Structure-based engineering of Tor complexes reveals that two types of yeast TORC1 produce distinct phenotypes.

Certain proteins assemble into diverse complex states, each having a distinct and unique function in the cell. Target of rapamycin (Tor) complex 1 (TORC1) plays a central role in signalling pathways that allow cells to respond to the environment, including nutritional status signalling. TORC1 is widely recognised for its association with various diseases. The budding yeast Saccharomyces cerevisiae has two types of TORC1, Tor1-containing TORC1 and Tor2-containing TORC1, which comprise different constituent proteins but are considered to have the same function. Here, we computationally modelled the relevant complex structures and then, based on the structures, rationally engineered a Tor2 mutant that could form Tor complex 2 (TORC2) but not TORC1, resulting in a redesign of the complex states. Functional analysis of the Tor2 mutant revealed that the two types of TORC1 induce different phenotypes, with changes observed in rapamycin, caffeine and pH dependencies of cell growth, as well as in replicative and chronological lifespan. These findings uncovered by a general approach with huge potential - model structure-based engineering - are expected to provide further insights into various fields such as molecular evolution and lifespan.

Overexpression of polyphosphate polymerases and deletion of polyphosphate phosphatases shorten the replicative lifespan in yeast.

We previously found that overexpression of phosphate starvation-responsive genes by disrupting PHO80 led to a shortened replicative lifespan in yeast. To identify lifespan-related genes, we screened upregulated genes in the pho80Δ mutant and focused on the VTC genes, which encode the vacuolar polyphosphate (polyP) polymerase complex. VTC1/VTC2/VTC4 deletion restored the lifespan and intracellular polyP levels in pho80Δ. In the wild type, overexpression of VTC5 or a combination of the other VTCs caused high polyP accumulation and shortened lifespan. Similar phenotypes were caused by the deletion of polyP phosphatase genes-vacuolar PPN1 and cytosolic PPX1. The polyP-accumulating strains exhibited stress sensitivities. Thus, we demonstrated that polyP metabolic enzymes participate in replicative lifespan, and extreme polyP accumulation shortens the lifespan.

Secreted acid phosphatases maintain replicative lifespan via inositol polyphosphate metabolism in budding yeast.

Secreted acid phosphatases (APases) dephosphorylate extracellular organic phosphate compounds to supply inorganic phosphate (Pi) to maintain cellular functions. Here, we show that APases are necessary to maintain a normal replicative lifespan in Saccharomyces cerevisiae. Deletion of all four APase genes shortened the lifespan in yeast strains on synthetic media (irrespective of the concentrations of Pi in the media), but it did not affect the intracellular ortho- and polyphosphate levels. Deletion of inositol-pentakisphosphate 2-kinase (IPK1), which encodes inositol-pentakisphosphate 2-kinase, restored the lifespan in APase-null mutants, and IPK1 overexpression shortened the lifespan in wild-type strains. Overexpression of inositol hexakisphosphate (IP6 ) and heptakisphosphate kinases, KCS1 and VIP1, recovered the lifespan in APase-null mutants. Thus, yeast APases modulate the replicative lifespan, probably through dephosphorylation of intracellular IP6 .

Cyclin-dependent kinase Pho85p and its cyclins are involved in replicative lifespan through multiple pathways in yeast.

Lifespan is determined by genetic factors and influenced by environmental factors. Here, we find that the phosphate signal transduction (PHO) pathway is involved in the determination of replicative lifespan in budding yeast. Extracellular phosphate does not affect the lifespan. However, deletion of PHO80 (cyclin) and PHO85 (cyclin-dependent kinase) genes, that is, negative regulators of the PHO pathway, shortens the lifespan, which is restored by further deletion of PHO4 (transcriptional activator). Four of the other nine Pho85p cyclin genes are also required to maintain normal lifespan. The short-lived mutants show a metabolic profile that is similar to strains with normal lifespan. Thus, Pho85p kinase genetically determines replicative lifespan in combination with relevant cyclins. Our findings uncover novel cellular signals in longevity regulation.

Proline metabolism regulates replicative lifespan in the yeast Saccharomyces cerevisiae.

In many plants and microorganisms, intracellular proline has a protective role against various stresses, including heat-shock, oxidation and osmolarity. Environmental stresses induce cellular senescence in a variety of eukaryotes. Here we showed that intracellular proline regulates the replicative lifespan in the budding yeast Saccharomyces cerevisiae. Deletion of the proline oxidase gene PUT1 and expression of the γ-glutamate kinase mutant gene PRO1-I150T that is less sensitive to feedback inhibition accumulated proline and extended the replicative lifespan of yeast cells. Inversely, disruption of the proline biosynthetic genes PRO1, PRO2, and CAR2 decreased stationary proline level and shortened the lifespan of yeast cells. Quadruple disruption of the proline transporter genes unexpectedly did not change intracellular proline levels and replicative lifespan. Overexpression of the stress-responsive transcription activator gene MSN2 reduced intracellular proline levels by inducing the expression of PUT1, resulting in a short lifespan. Thus, the intracellular proline levels at stationary phase was positively correlated with the replicative lifespan. Furthermore, multivariate analysis of amino acids in yeast mutants deficient in proline metabolism showed characteristic metabolic profiles coincident with longevity: acidic and basic amino acids and branched-chain amino acids positively contributed to the replicative lifespan. These results allude to proline metabolism having a physiological role in maintaining the lifespan of yeast cells.

Parallel mapping with site-directed hydroxyl radicals and micrococcal nuclease reveals structural features of positioned nucleosomes in vivo.

Micrococcal nuclease (MNase) has been widely used for analyses of nucleosome locations in many organisms. However, due to its sequence preference, the interpretations of the positions and occupancies of nucleosomes using MNase have remained controversial. Next-generation sequencing (NGS) has also been utilized for analyses of MNase-digests, but some technical biases are commonly present in the NGS experiments. Here, we established a gel-based method to map nucleosome positions in Saccharomyces cerevisiae, using isolated nuclei as the substrate for the histone H4 S47C-site-directed chemical cleavage in parallel with MNase digestion. The parallel mapping allowed us to compare the chemically and enzymatically cleaved sites by indirect end-labeling and primer extension mapping, and thus we could determine the nucleosome positions and the sizes of the nucleosome-free regions (or nucleosome-depleted regions) more accurately, as compared to nucleosome mapping by MNase alone. The analysis also revealed that the structural features of the nucleosomes flanked by the nucleosome-free region were different from those within regularly arrayed nucleosomes, showing that the structures and dynamics of individual nucleosomes strongly depend on their locations. Moreover, we demonstrated that the parallel mapping results were generally consistent with the previous genome-wide chemical mapping and MNase-Seq results. Thus, the gel-based parallel mapping will be useful for the analysis of a specific locus under various conditions.

The forkhead-like transcription factor (Fhl1p) maintains yeast replicative lifespan by regulating ribonucleotide reductase 1 (RNR1) gene transcription.

In eukaryotes, numerous genetic factors contribute to the lifespan including metabolic enzymes, signal transducers, and transcription factors. As previously reported, the forkhead-like transcription factor (FHL1) gene was required for yeast replicative lifespan and cell proliferation. To determine how Fhl1p regulates the lifespan, we performed a DNA microarray analysis of a heterozygous diploid strain deleted for FHL1. We discovered numerous Fhl1p-target genes, which were then screened for lifespan-regulating activity. We identified the ribonucleotide reductase (RNR) 1 gene (RNR1) as a regulator of replicative lifespan. RNR1 encodes a large subunit of the RNR complex, which consists of two large (Rnr1p/Rnr3p) and two small (Rnr2p/Rnr4p) subunits. Heterozygous deletion of FHL1 reduced transcription of RNR1 and RNR3, but not RNR2 and RNR4. Chromatin immunoprecipitation showed that Fhl1p binds to the promoter regions of RNR1 and RNR3. Cells harboring an RNR1 deletion or an rnr1-C428A mutation, which abolishes RNR catalytic activity, exhibited a short lifespan. In contrast, cells with a deletion of the other RNR genes had a normal lifespan. Overexpression of RNR1, but not RNR3, restored the lifespan of the heterozygous FHL1 mutant to the wild-type (WT) level. The Δfhl1/FHL1 mutant conferred a decrease in dNTP levels and an increase in hydroxyurea (HU) sensitivity. These findings reveal that Fhl1p regulates RNR1 gene transcription to maintain dNTP levels, thus modulating longevity by protection against replication stress.

Ole1, fatty acid desaturase, is required for Atg9 delivery and isolation membrane expansion during autophagy in Saccharomyces cerevisiae.

Macroautophagy, a major degradation pathway of cytoplasmic components, is carried out through formation of a double-membrane structure, the autophagosome. Although the involvement of specific lipid species in the formation process remains largely obscure, we recently showed that mono-unsaturated fatty acids (MUFA) generated by stearoyl-CoA desaturase 1 (SCD1) are required for autophagosome formation in mammalian cells. To obtain further insight into the role of MUFA in autophagy, in this study we analyzed the autophagic phenotypes of the yeast mutant of OLE1, an orthologue of SCD1. Δole1 cells were defective in nitrogen starvation-induced autophagy, and the Cvt pathway, when oleic acid was not supplied. Defects in elongation of the isolation membrane led to a defect in autophagosome formation. In the absence of Ole1, the transmembrane protein Atg9 was not able to reach the pre-autophagosomal structure (PAS), the site of autophagosome formation. Thus, autophagosome formation requires Ole1 during the delivery of Atg9 to the PAS/autophagosome from its cellular reservoir.

Random sample consensus combined with partial least squares regression (RANSAC-PLS) for microbial metabolomics data mining and phenotype improvement.

In recent years, the advent of high-throughput omics technology has made possible a new class of strain engineering approaches, based on identification of possible gene targets for phenotype improvement from omic-level comparison of different strains or growth conditions. Metabolomics, with its focus on the omic level closest to the phenotype, lends itself naturally to this semi-rational methodology. When a quantitative phenotype such as growth rate under stress is considered, regression modeling using multivariate techniques such as partial least squares (PLS) is often used to identify metabolites correlated with the target phenotype. However, linear modeling techniques such as PLS require a consistent metabolite-phenotype trend across the samples, which may not be the case when outliers or multiple conflicting trends are present in the data. To address this, we proposed a data-mining strategy that utilizes random sample consensus (RANSAC) to select subsets of samples with consistent trends for construction of better regression models. By applying a combination of RANSAC and PLS (RANSAC-PLS) to a dataset from a previous study (gas chromatography/mass spectrometry metabolomics data and 1-butanol tolerance of 19 yeast mutant strains), new metabolites were indicated to be correlated with tolerance within certain subsets of the samples. The relevance of these metabolites to 1-butanol tolerance were then validated from single-deletion strains of corresponding metabolic genes. The results showed that RANSAC-PLS is a promising strategy to identify unique metabolites that provide additional hints for phenotype improvement, which could not be detected by traditional PLS modeling using the entire dataset.

IREN, a novel EF-hand motif-containing nuclease, functions in the degradation of nuclear DNA during the hypersensitive response cell death in rice.

The hypersensitive response (HR), a type of programmed cell death that is accompanied by DNA degradation and loss of plasma membrane integrity, is a common feature of plant immune responses. We previously reported that transcription of IREN which encodes a novel EF-hand containing plant nuclease is controlled by OsNAC4, a key positive regulator of HR cell death. Transient overexpression of IREN in rice protoplasts also led to rapid DNA fragmentation, while suppression of IREN using RNA interference showed remarkable decrease of DNA fragmentation during HR cell death. Maximum DNA degradation associated with the recombinant IREN was observed in the presence of Ca(2+) and Mg(2+) or Ca(2+) and Mn(2+). Interestingly, DNA degradation mediated by the recombinant IREN was completely abolished by Zn(2+), even when Ca(2+), Mg(2+), or Mn(2+) were present in the reaction buffer. These data indicate that IREN functions in the degradation of nuclear DNA during HR cell death.

Metabolomic approach for improving ethanol stress tolerance in Saccharomyces cerevisiae.

The budding yeast Saccharomyces cerevisiae is widely used for brewing and ethanol production. The ethanol sensitivity of yeast cells is still a serious problem during ethanol fermentation, and a variety of genetic approaches (e.g., random mutant screening under selective pressure of ethanol) have been developed to improve ethanol tolerance. In this study, we developed a strategy for improving ethanol tolerance of yeast cells based on metabolomics as a high-resolution quantitative phenotypic analysis. We performed gas chromatography-mass spectrometry analysis to identify and quantify 36 compounds on 14 mutant strains including knockout strains for transcription factor and metabolic enzyme genes. A strong relation between metabolome of these mutants and their ethanol tolerance was observed. Data mining of the metabolomic analysis showed that several compounds (such as trehalose, valine, inositol and proline) contributed highly to ethanol tolerance. Our approach successfully detected well-known ethanol stress related metabolites such as trehalose and proline thus, to further prove our strategy, we focused on valine and inositol as the most promising target metabolites in our study. Our results show that simultaneous deletion of LEU4 and LEU9 (leading to accumulation of valine) or INM1 and INM2 (leading to reduction of inositol) significantly enhanced ethanol tolerance. This study shows the potential of the metabolomic approach to identify target genes for strain improvement of S. cerevisiae with higher ethanol tolerance.

Yeast Cyc8p and Tup1p proteins function as coactivators for transcription of Stp1/2p-dependent amino acid transporter genes.

The yeast Cyc8p-Tup1p complex is known to serve primarily as a transcriptional corepressor in a variety of biological processes. However, less is known about its function as a coactivator. Herein, we found tryptophan transporter genes, TAT1 and TAT2, that, when overexpressed, suppressed the slow growth of Δcyc8. We observed that the addition of tryptophan to Δcyc8 cultures partially restored cell growth, and the deletion of CYC8 and TUP1 reduced transcriptional levels of TAT1 and TAT2. Tup1p bound to the promoter region of TAT1 and TAT2 genes that were dependent on STP1 and STP2 (encoding DNA-binding activator proteins) for expression. Similarly, transcription of the other Stp1/2p-dependent amino acid transporter (AAT) genes also required CYC8 and TUP1 gene functions. These data indicate that Cyc8p-Tup1p plays a role as a transcriptional coactivator for AAT genes via Stp1/2p activators and that lowering intracellular tryptophan by CYC8 deletion causes slow growth.

A metabolomics-based strategy for identification of gene targets for phenotype improvement and its application to 1-butanol tolerance in Saccharomyces cerevisiae.

BACKGROUND: Traditional approaches to phenotype improvement include rational selection of genes for modification, and probability-driven processes such as laboratory evolution or random mutagenesis. A promising middle-ground approach is semi-rational engineering, where genetic modification targets are inferred from system-wide comparison of strains. Here, we have applied a metabolomics-based, semi-rational strategy of phenotype improvement to 1-butanol tolerance in Saccharomyces cerevisiae. RESULTS: Nineteen yeast single-deletion mutant strains with varying growth rates under 1-butanol stress were subjected to non-targeted metabolome analysis by GC/MS, and a regression model was constructed using metabolite peak intensities as predictors and stress growth rates as the response. From this model, metabolites positively and negatively correlated with growth rate were identified including threonine and citric acid. Based on the assumption that these metabolites were linked to 1-butanol tolerance, new deletion strains accumulating higher threonine or lower citric acid were selected and subjected to tolerance measurement and metabolome analysis. The new strains exhibiting the predicted changes in metabolite levels also displayed significantly higher growth rate under stress over the control strain, thus validating the link between these metabolites and 1-butanol tolerance. CONCLUSIONS: A strategy for semi-rational phenotype improvement using metabolomics was proposed and applied to the 1-butanol tolerance of S. cerevisiae. Metabolites correlated with growth rate under 1-butanol stress were identified, and new mutant strains showing higher growth rate under stress could be selected based on these metabolites. The results demonstrate the potential of metabolomics in semi-rational strain engineering.

Transcription factor genes essential for cell proliferation and replicative lifespan in budding yeast.

Many of the lifespan-related genes have been identified in eukaryotes ranging from the yeast to human. However, there is limited information available on the longevity genes that are essential for cell proliferation. Here, we investigated whether the essential genes encoding DNA-binding transcription factors modulated the replicative lifespan of Saccharomyces cerevisiae. Heterozygous diploid knockout strains for FHL1, RAP1, REB1, and MCM1 genes showed significantly short lifespan. (1)H-nuclear magnetic resonance analysis indicated a characteristic metabolic profile in the Δfhl1/FHL1 mutant. These results strongly suggest that FHL1 regulates the transcription of lifespan related metabolic genes. Thus, heterozygous knockout strains could be the potential materials for discovering further novel lifespan genes.

Changes in transcription and metabolism during the early stage of replicative cellular senescence in budding yeast.

Age-related damage accumulates and a variety of biological activities and functions deteriorate in senescent cells. However, little is known about when cellular aging behaviors begin and what cellular aging processes change. Previous research demonstrated age-related mRNA changes in budding yeast by the 18th to 20th generation, which is the average replicative lifespan of yeast (i.e. about half of the population is dead by this time point). Here, we performed transcriptional and metabolic profiling for yeast at early stages of senescence (4th, 7th, and 11th generation), that is, for populations in which most cells are still alive. Transcriptional profiles showed up- and down-regulation for ∼20% of the genes profiled after the first four generations, few further changes by the 7th generation, and an additional 12% of the genes were up- and down-regulated after 11 generations. Pathway analysis revealed that these 11th generation cells had accumulated transcripts coding for enzymes involved in sugar metabolism, the TCA cycle, and amino acid degradation and showed decreased levels of mRNAs coding for enzymes involved in amino acid biosynthetic pathways. These observations were consistent with the metabolomic profiles of aging cells: an accumulation of pyruvic acid and TCA cycle intermediates and depletion of most amino acids, especially branched-chain amino acids. Stationary phase-induced genes were highly expressed after 11 generations even though the growth medium contained adequate levels of nutrients, indicating deterioration of the nutrient sensing and/or signaling pathways by the 11th generation. These changes are presumably early indications of replicative senescence.

Metabolic profiling of retrograde pathway transcription factors rtg1 and rtg3 knockout yeast.

Rtg1 and Rtg3 are two basic helix-loop-helix (bHLH) transcription factors found in yeast Saccharomyces cerevisiae that are involved in the regulation of the mitochondrial retrograde (RTG) pathway. Under RTG response, anaplerotic synthesis of citrate is activated, consequently maintaining the supply of important precursors necessary for amino acid and nucleotide synthesis. Although the roles of Rtg1 and Rtg3 in TCA and glyoxylate cycles have been extensively reported, the investigation of other metabolic pathways has been lacking. Characteristic dimer formation in bHLH proteins, which allows for combinatorial gene expression, and the link between RTG and other regulatory pathways suggest more complex metabolic signaling involved in Rtg1/Rtg3 regulation. In this study, using a metabolomics approach, we examined metabolic alteration following RTG1 and RTG3 deletion. We found that apart from TCA and glyoxylate cycles, which have been previously reported, polyamine biosynthesis and other amino acid metabolism were significantly altered in RTG-deficient strains. We revealed that metabolic alterations occurred at various metabolic sites and that these changes relate to different growth phases, but the difference can be detected even at the mid-exponential phase, when mitochondrial function is repressed. Moreover, the effect of metabolic rearrangements can be seen through the chronological lifespan (CLS) measurement, where we confirmed the role of the RTG pathway in extending the yeast lifespan. Through a comprehensive metabolic profiling, we were able to explore metabolic phenotypes previously unidentified by other means and illustrate the possible correlations of Rtg1 and Rtg3 in different pathways.

Crystal structure of the N-terminal domain of the yeast general corepressor Tup1p and its functional implications.

The yeast Cyc8p-Tup1p protein complex is a general transcriptional corepressor of genes involved in many different physiological processes. Herein, we present the crystal structure of the Tup1p N-terminal domain (residues 1-92), essential for Tup1p self-assembly and interaction with Cyc8p. This domain tetramerizes to form a novel antiparallel four-helix bundle. Coiled coil interactions near the helical ends hold each dimer together, whereas interdimeric association involves only two sets of two residues located toward the chain centers. A mutagenesis study confirmed that the nonpolar residues responsible for the association of the protomers as dimers are also required for transcriptional repression. An additional structural study demonstrated that the domain containing an Leu(62) → Arg mutation that had been shown not to bind Cyc8p exhibits an altered structure, distinct from the wild type. This altered structure explains why the mutant cannot bind Cyc8p. The data presented herein highlight the importance of the architecture of the Tup1p N-terminal domain for self-association.

Construction and analysis of EST libraries of the trans-polyisoprene producing plant, Eucommia ulmoides Oliver.

Eucommia ulmoides Oliver is one of a few woody plants capable of producing abundant quantities of trans-polyisoprene rubber in their leaves, barks, and seed coats. One cDNA library each was constructed from its outer stem tissue and inner stem tissue. They comprised a total of 27,752 expressed sequence tags (ESTs) representing 10,520 unigenes made up of 4,302 contigs and 6,218 singletons. Homologues of genes coding for rubber particle membrane proteins that participate in the synthesis of high-molecular poly-isoprene in latex were isolated, as well as those encoding known major latex proteins (MLPs). MLPs extensively shared ESTs, indicating their abundant expression during trans-polyisoprene rubber biosynthesis. The six mevalonate pathway genes which are implicated in the synthesis of isopentenyl diphosphate (IPP), a starting material of poly-isoprene biosynthesis, were isolated, and their role in IPP biosynthesis was confirmed by functional complementation of suitable yeast mutants. Genes encoding five full-length trans-isoprenyl diphosphate synthases were also isolated, and two among those synthesized farnesyl diphosphate from IPP and dimethylallyl diphosphate, an assumed intermediate of rubber biosynthesis. This study should provide a valuable resource for further studies of rubber synthesis in E. ulmoides.

GABA metabolism pathway genes, UGA1 and GAD1, regulate replicative lifespan in Saccharomyces cerevisiae.

Many of the genes involved in aging have been identified in organisms ranging from yeast to human. Our previous study showed that deletion of the UGA3 gene-which encodes a zinc-finger transcription factor necessary for γ-aminobutyric acid (GABA)-dependent induction of the UGA1 (GABA aminotransferase), UGA2 (succinate semialdehyde dehydrogenase), and UGA4 (GABA permease) genes-extends replicative lifespan in the budding yeast Saccharomyces cerevisiae. Here, we found that deletion of UGA1 lengthened the lifespan, as did deletion of UGA3; in contrast, strains with UGA2 or UGA4 deletions exhibited no lifespan extension. The Δuga1 strain cannot deaminate GABA to succinate semialdehyde. Deletion of GAD1, which encodes the glutamate decarboxylase that converts glutamate into GABA, also increased lifespan. Therefore, two genes in the GABA metabolism pathway, UGA1 and GAD1, were identified as aging genes. Unexpectedly, intracellular GABA levels in mutant cells (except for Δuga2 cells) did not differ from those in wild-type cells. Addition of GABA to culture media, which induces transcription of the UGA structural genes, had no effect on replicative lifespan of wild-type cells. Multivariate analysis of (1)H nuclear magnetic resonance spectra for the whole-cell metabolite levels demonstrated a separation between long-lived and normal-lived strains. Gas chromatography-mass spectrometry analysis of identified metabolites showed that levels of tricarboxylic acid cycle intermediates positively correlated with lifespan extension. These results strongly suggest reduced activity of the GABA-metabolizing enzymes extends lifespan by shifting carbon metabolism toward respiration, as calorie restriction does.

A novel human dynactin-associated protein, dynAP, promotes activation of Akt, and ergosterol-related compounds induce dynAP-dependent apoptosis of human cancer cells.

There are several human genes that may encode proteins whose functions remain unknown. To find clues to their functions, we used the mutant yeast defective in Mad2, a component of the spindle checkpoint complex. Phenotypes that were provoked by the expression of a human C18orf26 protein in the mutant yeast encouraged further characterization of this protein in human cells. This protein was designated dynAP (dynactin-associated protein) because of its interaction with dynactin subunits that comprised a microtubule-based motor protein complex. The dynAP is a transmembrane protein localizing to Golgi apparatus and plasma membrane in a microtubule-dependent manner. This protein was expressed in half of human cancer cell lines but barely in normal human fibroblasts tested. The SV40-transformed fibroblasts expressed dynAP. Importantly, the expression of dynAP activated Akt (also known as protein kinase B) by promoting Ser47³ phosphorylation required for the full activation, whereas knockdown of dynAP abolished this activation. The ergosterol-related compounds identified by the yeast cell-based high-throughput screen abrogated activation of Akt and induced apoptosis in a dynAP-dependent manner. We propose a possible advantage of dynAP expression in cancer cells; the survival of cancer cells that express dynAP is supported by dynAP-induced activation of Akt, sustaining high rates of proliferation. The inactivation of dynAP by the selected compounds nullifies this advantage, and thereby, the apoptotic machinery is allowed to operate. Taken together, dynAP can be a new target for cancer therapy, and the selected chemicals are useful for developing a new class of anticancer drugs.