|Home | About | Journals | Submit | Contact Us | Français|
Sordarin and its derivatives are antifungal compounds of potential clinical interest. Despite the highly conserved nature of the fungal and mammalian protein synthesis machineries, sordarin is a selective inhibitor of protein synthesis in fungal organisms. In cells sensitive to sordarin, its mode of action is through preventing the release of translation elongation factor 2 (eEF2) during the translocation step, thus blocking protein synthesis. To further investigate the cellular components required for the effects of sordarin in fungal cells, we have used the haploid deletion collection of Saccharomyces cerevisiae to systematically identify genes whose deletion confers sensitivity or resistance to the compound. Our results indicate that genes in a number of cellular pathways previously unknown to play a role in sordarin response are involved in its growth effects on fungal cells and reveal a specific requirement for the diphthamidation pathway of cells in causing eEF2 to be sensitive to the effects of sordarin on protein synthesis. Our results underscore the importance of the powerful genomic tools developed in yeast (Saccharomyces cerevisiae) to more comprehensively understanding the cellular mechanisms involved in the response to therapeutic agents.
Sordarin is a natural product originally isolated from Sordaria araneosa. A family of sordarin-related compounds has been described, all of them inhibiting cell growth by interfering with protein synthesis (6, 7). Even though there is a high degree of homology between the fungal and mammalian protein synthesis machineries, these inhibitors are highly specific for the fungal elongation step (24). Moreover, the sordarin derivatives display various levels of antifungal activity when tested against different species of fungi, with some of them being sensitive while others are resistant (15, 28).
Two proteins have been described as targets of sordarin and its derivatives, both involved in protein synthesis: translation elongation factor 2 (eEF2) (5, 7, 21) and the large ribosomal subunit protein rpP0 (13, 22). In the former target, characterization in yeast (Saccharomyces cerevisiae) of spontaneous resistant mutants along with site-directed mutagenesis has been used to identify residues 521, 523, and 524 as the most important amino acids in eEF2 involved in the binding of sordarin (41). Additional residues located at other sites in eEF2 have also been implicated in sordarin resistance (5, 21). Regarding RPP0, spontaneous resistant mutants have been isolated in the region from amino acid 137 to 144 (13, 22), and site-directed mutagenesis results also have implicated residues 119, 124, and 126 of rpP0 in sordarin resistance (39, 40). In any case, as sordarin directly binds to eEF2, this factor seems to be the primary target of the drug. However, since the ribosome strongly increases the affinity for the drug, the eEF2-ribosome complex must be considered as the functional target, where mutations in the stalk P0 protein must induce resistance through allosteric effects. These data suggest that the sordarin binding site seems to be quite susceptible to long-distance conformational changes.
Sordarin has been a valuable tool in analyzing some structural and functional aspects of the eukaryotic ribosome. Thus, it has been used in cryo-electron microscopy to visualize stalled ribosomes, associated with eEF2 (14, 42, 45). Moreover, it has facilitated the crystallization of eEF2 (19, 20), making it possible to dock the eEF2-sordarin X-ray structure in the molecular model of the yeast ribosome, revealing conformational changes within eEF2 in the ribosome which facilitate tRNA translocation (18, 42, 45).
Since mutants highly resistant to sordarin have been isolated only from strains with mutations in eEF2 and rpP0, these are probably the main targets involved in the inhibitory mechanism. However, the participation of additional proteins and pathways in the process cannot be ruled out. As an example, the remaining components of the ribosome stalk have been shown to be somehow involved in the sordarin mode of action, since in spontaneous mutants sordarin resistance is diminished by deletion of rpP1α and rpP2β (13). Due to the specificity of the sordarin mechanism of protein synthesis inhibition, any protein that affects this inhibition is expected to be involved, either directly or indirectly, in the translocation mechanism. Therefore, the search for these proteins could provide a way to identify new effectors of the translation process.
The availability of a nearly complete set of gene deletion mutants of yeast (49) enables genome-wide screenings of null mutants for a variety of phenotypes. Screening the yeast “disruptome” has proved to be a useful high-throughput functional genomic tool to identify genes involved in responsiveness to drugs (27, 44) as well as other phenotypes. Although mutations in genes encoding components involved in the translocation process are plausible candidates to render cells resistant or sensitive to sordarin, it is likely that other mechanisms participate in the response. Microbial cells can resist the action of any drug by three main mechanisms: (i) modification of the target, (ii) inactivation of the drug, and (iii) changes in the permeability of cells to the compound (31). Since it is highly improbable that deletion of a yeast gene can result in the inactivation of sordarin, we would expect to isolate mutants whose resistance to this compound is due to one of the other two mechanisms. Using this yeast deletion collection, we performed a chemical genomic screen to identify the spectrum of genes whose deletion alters the fitness profile in the presence of sordarin. This analysis revealed that a wide range of gene functions is required to protect cells against sordarin cytotoxicity. Most strikingly, our results reveal a role for diphthamide modification of eEF2 in sordarin-mediated toxicity.
The collection of nonessential haploid MATa yeast deletion strains derived from parent strain BY4741 (MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0) was obtained from Euroscarf (Frankfurt, Germany; http://web.uni-frankfurt.de/fb15/mikro/euroscarf/index.html) in a 96-well plate format. The deletion strains were generated by the Saccharomyces Genome Deletion Project (http://www-sequence.stanford.edu/group/yeast_deletion_project/deletions3.html) by replacing the target gene with a Kanr cassette, KanMX4, by means of a PCR-based gene disruption strategy (49).
Yeast strains were grown on standard rich medium (yeast-peptone-dextrose [YPD; 1% yeast extract, 2% Bacto peptone, 2% glucose]), synthetic minimal medium (SD; 0.17% yeast nitrogen base without amino acids, 0.5% ammonium sulfate, 2% glucose supplemented according to the requirements of the strains), and synthetic complete medium (SC; SD medium with 0.079% complete supplement mixture [ForMedium, Norwich, United Kingdom]). Glucose was replaced by 2% galactose and 1% raffinose when noted. All cultures were incubated at 28°C.
Petite mutants derived from parent strain BY4741 were isolated after growth in the presence of ethidium bromide (12). As expected, these cells were not viable when glycerol or ethanol were used as a nonfermentable carbon source.
Approximately 4,800 haploid deletion strains in the BY4741 background were screened for sensitivity/resistance to sordarin in SC liquid medium. Strains were pinned from 96-well frozen stock plates by using a stainless steel 96-pin replicator (Nalgene Nunc International) into 96-well plates containing 150 μl of liquid YPD medium supplemented with G418 (150 mg/liter; Gibco-BRL). The plates were incubated at 28°C for 3 days until they were confluent and then pin replicated onto liquid SC 96-well plates containing either no drug or sordarin derivative GM193663 (a gift from GlaxoSmithKine [15, 28]) at 2 ng/ml for sensitivity tests and 62.5 ng/ml for resistance. Plates were incubated at 28°C and growth was quantitatively scored every 24 h over a period of 5 days by optical density at 595 nm (OD595) readings using a microplate reader spectrophotometer (model 550; Bio-Rad Laboratories). The sordarin sensitivity/resistance was scored on the basis of the relative growth of each mutant against that of the wild type in the same plate and with that of the mutant in the control plates, where no inhibitor was added. Putative sordarin-sensitive and -resistant strains identified during the screening of the yeast knockout collection were further retested at least in duplicate under the same conditions described for the screening. Validation was performed at 62.5 ng/ml, 125 ng/ml, and 500 ng/ml for resistance phenotype testing and at 2 ng/ml for sensitivity testing.
Cultures grown in 96-well plates were resuspended using a 96-well plate shaker (Microtec; Infors AG), and four serial 10-fold dilutions in water were done. Five-microliter portions of the diluted cultures were spotted using a replica plater for a 96-well plate (Sigma-Aldrich) to OmnyTray plates (Nalgene Nunc International) containing SC solid medium. The plates were incubated at 28°C for 6 days and were imaged every 24 h.
To identify functional clusters of genes, statistical analysis of overrepresentation of functional categories affected by sordarin was performed by using Gene Ontology (GO) (2), functional annotations tools from FatiGo, FUNSPEC (http://funspec.med.utoronto.ca/) (36), the GO term finder at the Saccharomyces Genome Database (SGD) (http://www.yeastgenome.org/), and the MIPS functional catalogue (http://mips.gsf.de/genre/proj/yeast).
The gene classification was done subjectively, supported by SGD, the Yeast Proteome Database, and the Comprehensive Yeast Genome Database at MIPS as well as the literature. Microsoft Excel was used to further manipulate and analyze data.
Recombinant DNA techniques were carried out using standard molecular biology protocols (38). For the induced expression of diphtheria toxin fragment A (DTA), plasmid pYEUra3-Gal1-DTA was constructed. The primers DTA-F (5′-AGGATCCATATGGCGCTGATGATGTTGTTGA-3′) and DTA-R (5′-CAGTCGACTCGTCAGACACGATTTCCTG-3′) (restriction sites BamHI and SalI underlined) were used to amplify DTA (23) from plasmid pT7-7-DTA (a gift from the laboratory of R. J. Collier at Harvard Medical School). The amplified product was cloned in pGEM-T (Promega) vector and subsequently digested with BamHI and SalI and cloned in the expression plasmid pYEUra3 (Clontech) under the control of the GAL1 promoter to yield the galactose expression-induced plasmid pYEUra3-DTA. Yeast transformations were performed by a modification of the standard lithium acetate/polyethylene glycol method (11).
The dph3Δ mutant was constructed by use of a PCR-based gene deletion strategy using the KanMX6 marker, flanked by 40 bases of homology to the region upstream of the DPH3 open reading frame, including the AUG, and 40 bases of homology to the region downstream of it, including the stop codon. The PCR product was integrated by homologous recombination, replacing the wild-type gene with the Kanr marker by standard transformation of the parental strain BY4741. The deletion was confirmed by colony PCR of 10 geneticin-resistant transformant clones by use of primers DPH3-5′ (upstream of the deleted DPH3) and KanB (within the KanMX6 module).
Ribosomes and S-100 fractions were prepared from cells grown in YPD medium to an OD600 of 0.4 (37). The procedure and experimental conditions for polyphenylalanine synthesis directed by poly(U) are described in reference 35.
Exponentially (OD600 = 0.6) growing cells in YPD were collected, washed, and disrupted with sea sand in 1 volume of 80 mM KCl, 10 mM Tris-HCl, pH 7.4, 12.5 mM MgCl2, 5 mM 2-mercaptoethanol. S-100 extracts were prepared and sequentially precipitated with ammonium sulfate (37). The 30 to 70% fraction was resuspended in 50 mM Tris-HCl, pH 7.4, 5 mM 2-mercaptoethanol, and the protein concentration was determined by Bradford assay (3). The ADP ribosylation reaction was performed in a reaction mix containing, in 100 μl, 0.5 mg of yeast extract, 2 μg of DTA, and 2 μM [14C]NADH (9.18 GBq/mmol; Amersham Biosciences). The reaction was carried out for 40 min at 30°C and terminated by adding 1 ml of 10% trichloroacetic acid. Incorporation of ADP-ribose into eEF-2 was measured as radioactivity present in 10% trichloroacetic acid precipitates.
The sordarin in vitro binding assays were performed as described elsewhere (13). Cell-free S-50 extracts were prepared from cells grown in YPD medium to an OD600 of 1.5. To perform the assays, 0.5 mg of protein from an S-50 fraction and 0.1 μg of [3H]sordarin (36 kBq, 20 nmol; a gift from GlaxoSmithKline) were mixed in a final volume of 500 μl. After 30 min of incubation at room temperature, 400 μl from each sample were applied to a prepacked PD-10 Sephadex G-25 column (GE Healthcare) equilibrated with binding buffer (80 mM HEPES-KOH, pH 7.4, 10 mM MgOAc, 1 mM dithiothreitol). [3H]sordarin bound to macromolecules was measured by counting the radioactivity in the excluded fractions. The radioactivity counts obtained in the presence of 10 μg of cold sordarin were subtracted from all the data points. In addition, this radioactivity was corrected by the amount of eEF2 present in each sample (see “Electrophoretic methods” below).
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis was performed as described previously (38). Proteins transferred to a polyvinylidene difluoride membrane (Immobilon; Millipore) were treated either with polyclonal antibodies prepared against eEF2 or monoclonal antibody 3BH5 against the ribosomal protein P0 (48), and Western blots were developed with enhanced chemiluminescence. Densitometry of the films was processed using the software ImageJ 1.38x. The relative amount of eEF2 in each sample was normalized using the signal due to rpP0.
The availability of a genome-wide set of S. cerevisiae deletion mutants (10, 49) makes it possible to identify nearly all the nonessential genes that could alter the yeast sordarin sensitivity and, consequently, that might be related to its mode of action. In order to identify these genes, the complete set of isogenic nonessential haploid yeast deletion mutants for putative open reading frames larger than 300 nucleotides was screened for their ability to grow in the presence of different concentrations of the sordarin derivative GM193663.
The screening was initiated by testing sordarin for the ability to inhibit growth in the yeast strain BY4741. Based on the results of a growth curve analysis in SC medium with a broad range of sordarin concentrations (0 to 500 ng/ml) (Fig. (Fig.1),1), the primary screening for hypersensitive mutants was performed at a semi-inhibitory concentration of sordarin (2 ng/ml), while for resistant mutants a drug concentration (62.5 ng/ml) that severely inhibited growth of the wild type in SC medium was used. Nine mutants that were repeatedly unable to grow at 2 ng/ml of sordarin (see Table S1 in the supplemental material) and 104 mutants showing enhanced resistance to this compound (see Table S2 in the supplemental material) were identified. Out of these, 58 mutants were able to grow at very high concentrations of the drug (500 ng/ml), and therefore they were considered as superresistant.
The main cluster of the 104 genes (see Table S2 in the supplemental material), which when deleted induced resistance to 62.5 ng/ml of sordarin, is primarily involved in protein metabolism (39 out of 104 genes, or 37.5%), which is consistent with the known role of the drug as a protein synthesis inhibitor. According to GO at SGD, strains representing deletions of genes with functions in mitochondria (28 out of 104 genes, or 26.9%) and ribosomes (17 out of 104 genes, or 16.3%) were overrepresented among the sordarin-resistant set.
The relatively large proportion of resistant mutants that have mitochondrial defects raised the question as to whether this organelle as a whole is somehow related to sordarin function. To test this possibility, the response to sordarin of petite strains derived from S. cerevisiae BY4741 by ethidium bromide treatment was tested. However, no significant difference was detected in their sensitivity to the drug compared with that of the parental strain (see Fig. S1 in the supplemental material).
The most striking feature in the sordarin-resistant screen was the significant overrepresentation of strains deleted for genes directly involved in peptidyl-diphthamide metabolism. Diphthamidation is a specific posttranslational modification of eEF2 unique for this protein (47). The biosynthesis of diphthamide is accomplished by stepwise enzymatic additions to the His699 residue of eEF2, due to the action of five proteins, Dph1p to Dph5p, and a still-unidentified amidating enzyme (26, 29, 30). We identified mutants with deletions in four out of the corresponding genes (DPH1, DPH2, DPH4, and DPH5) in the sordarin-resistant set, with the dph1Δ, dph2Δ, and dph5Δ mutants being among the most resistant subset (see Table S2 in the supplemental material). The strain lacking DPH3 was not present in the deletion mutant collection, as this open reading frame is smaller than 300 nucleotides. Therefore, a strain with a knockout of this gene was generated as detailed in Materials and Methods and tested for sordarin resistance, together with the remaining dphΔ mutants, in a serial dilution spot test of growth. All the dphΔ mutants grew in the presence of 100 ng/ml of GM193663, a concentration that totally inhibited the growth of the parental strain (Fig. (Fig.2A).2A). No differences were observed between MATa and MATα yeast cells (data not shown). As the dph3Δ mutant seemed to grow slower than the remaining mutants on agar plates both in the absence and in the presence of sordarin, we performed a growth inhibition curve in liquid medium in order to have a more accurate estimation of the relative resistance level. The results shown in Fig. Fig.2B2B indicate that all dph1Δ to dph5Δ mutants have similar 50% inhibitory concentration values, in the range of 300 to 500 ng/ml of sordarin, suggesting a relationship between eEF2 diphthamidation and sordarin resistance.
We first wanted to confirm that, as expected, eEF2 is not diphthamidated in the selected dphΔ sordarin-resistant strains, testing their response to diphtheria toxin (26). We expressed in these strains the DTA under the control of the inducible GAL1 promoter in plasmid pYEUra-Gal1-DTA. As shown in Fig. Fig.3,3, the five dphΔ strains were resistant to DTA. Moreover, none of the dphΔ strains ADP ribosylate the factor when they were tested in a diphtheria toxin-dependent eEF2 ADP ribosylation assay from [14C]NADH (Table (Table11).
The preceding results indicated that yeast cells in which DPH1-5 genes are deleted express a nondiphthamidated eEF2 and, as a consequence, are highly resistant to diphtheria toxin and moderately resistant to sordarin. Additional evidence supporting that this higher sordarin resistance is related to protein synthesis was provided by in vitro inhibition of protein synthesis tests in extracts of the mutants. As shown in Fig. Fig.4,4, the dphΔ mutants show a higher 50% inhibitory concentration than the wild type, confirming that the absence of diphthamide is directly related to a higher resistance of the translation machinery to sordarin. Among the five mutants, the behavior of the dph3Δ mutant is slightly different, as its sensitivity in this test is halfway between that of the wild type and those of the remaining dphΔ mutants.
We wondered whether sordarin resistance linked to the absence of diphthamide in eEF2 is associated with effects on the binding of the drug to the factor. Spontaneous mutations in eEF2 that confer resistance to sordarin have been explained by the inability of sordarin to bind to the mutant eEF2 (7, 21). This can be observed in Fig. Fig.5A,5A, in the case of the fpr1-19 mutant, which carries an S523T mutation in eEF2 that abolishes drug binding, indicating that the factor is the only macromolecular target of sordarin. Nevertheless, spontaneous sordarin-resistant mutations have been observed for other genes such as RPP0, in which sordarin binding to the factor is not completely abolished (see results in Fig. Fig.5A5A for the fpr2-17 strain, which carries a double mutation, N139H A140W, in ribosomal protein P0), although this binding does not seem to affect the function of the factor during translocation (13). We then performed binding assays of radioactive sordarin to yeast extracts derived from the dphΔ mutants (Fig. (Fig.5B).5B). The bound radioactivity was normalized taking into account the amount of eEF2, the unique target of the drug, which was estimated by Western blotting using specific antibodies against this factor and also ribosomal protein P0 used as a loading control (Fig. (Fig.5C).5C). As shown in Fig. Fig.5B,5B, all dphΔ extracts can bind sordarin to 60 to 80% of the level seen for the wild type.
To gain insights into the sordarin mode of action and to reveal possible resistance mechanisms of yeast to this antifungal, we performed a chemical genomics screen of the haploid yeast deletion mutant collection.
Deletion of nine genes has been found to induce hypersensitivity to the drug. Most of them encode mitochondrial proteins (ATP1, ATP12, AFG3, SSQ1, and HEM14). Deletants for all these genes are unable to grow in a nonfermentable carbon source (43). Therefore, there seems to be a relationship between respiratory defects and sordarin sensitivity.
The number of identified sordarin-resistant strains is rather large (see Table S2 in the supplemental material). Some of the identified genes also encode mitochondrial proteins that can be related to processes that could affect the drug transport if this process is energy dependent. However, this does not seem to be the case, as the sensitivity to sordarin of petite strains is not different from that of the wild-type strain. Deletion of genes coding for proteins involved in cell wall organization might also inhibit drug transport. Unfortunately, it is difficult to relate most of the identified genes with the transport processes, and presently their relationship with sordarin resistance is not understood.
A number of the genes that we identified in our screen can be directly related to the sordarin target, namely, those encoding either ribosomal proteins or components of the diphthamide biosynthesis pathway. Seven ribosomal protein genes are found in this group, five of them encoding mitochondrial proteins. Since the prokaryote-type mitochondrial translation system is presumed insensitive to sordarin, it must be assumed that the detected effects are not directly related to their role as components of the sordarin target. Cytoplasmic ribosomal protein L12 is encoded by two gene copies, and deletion of the B copy results in about a 70% increase in the cell doubling time and a proportional reduction in the cellular content of rpL12 (4). Protein L12 binds closely to P0, forming the stalk base, and therefore the effect of its absence on sordarin interaction is not unexpected, as the absence of this protein in the stalk base may affect the interaction of eEF2 with the ribosome. While it is certainly surprising that only deletants in the RPL12B copy, and not in RPL12A, were found during the screening, this result provides an additional evidence confirming differences between the functional roles of ribosomal protein paralogs (25). Similarly to RPL12, ribosomal protein L1 is also encoded by two gene copies, and the disruption of the B copy results in a 100% doubling time increase (34). The effect of protein L1 on sordarin resistance is perhaps more surprising due to its location in the lateral protuberance at the opposite side of the stalk in the ribosomal 60S subunit (42). However, L1 with the aid of eEF3 is directly involved in the release of the deacylated tRNA from the ribosome (1, 46), and it cannot be ruled out that when the availability of L1 is reduced, the complete translocation process is slowed down, and an effect on the eEF2 conformation or the binding to the ribosome takes place, which affects the sordarin interaction. Alternatively, L1 absence may lead to more-efficient eEF2 release from the ribosome, and therefore cells are less sordarin sensitive.
We also identified genes of unknown function, and these deserve further analysis that can provide information on their possible function. Among these, YLR043w has properties that could relate the encoded protein with ribosomal activities. Thus, besides having an ATP-binding region and a chorismate mutase-like domain, this cytoplasmic protein contains an endoribonuclease L-PSP domain. A rat liver protein with this domain has been described as an inhibitor of protein synthesis due to its single-stranded RNA endoribonuclease activity (32). It is therefore tempting to speculate that in the absence of this protein the conformation of the eEF2-ribosome complex might have a diminished affinity for sordarin.
The most interesting conclusion of this study is, however, the clear relationship between EF2 diphthamidation and sordarin activity. The following different evidences are presented indicating that diphthamide is involved in the mechanism of sordarin inhibition: (i) the growth of dphΔ mutants in the presence of inhibitory concentrations of the drug, (ii) the lower sensitivity to sordarin of in vitro cell-free translation systems derived from these mutants, and (iii) the reduced sordarin binding to cell extracts derived from dphΔ mutants. Altogether, the results strongly suggest that the binding of sordarin to the complex between the 80S ribosome and eEF2 is significantly weaker when the factor is not diphthamidated, thus offering an explanation for the observed resistance of strains defective in the diphthamide biosynthesis pathway. Interestingly, while mutations of His699 of eEF2 show a higher sensitivity to translational inhibitors (33), the strains carrying deletions in genes DPH2 or DPH5 do not confer sensitivity to the translation inhibitors cycloheximide and hygromycin B (33). With the exception of the dph3Δ mutant (see below), we have preliminary results extending this observation to the remaining dphΔ mutants (data not shown), highlighting the specificity of the connection between sordarin sensitivity and the diphthamidation of eEF2.
It is interesting that not all the dphΔ mutants behave similarly with respect to sordarin activity, suggesting that the various intermediates in the synthesis can affect the eEF2 response differently. In this respect, dph3Δ mutants have especially interesting peculiarities. These are remarkable considering that Dph1 to Dph4 proteins are involved in the first step of diphthamide synthesis, that is, the transfer of the 3-amino-3-carboxypropyl group of S-adenosylmethionine to the imidazole C-2 of the precursor histidine residue (26). While the overall in vivo sordarin resistance levels are similar, the in vitro protein synthesis in dph3Δ extracts is more sensitive to the drug than are those for the remaining dphΔ mutants. In addition, the growth rate of mutants lacking DPH1, DPH2, DPH4, and DPH5 is largely indistinguishable from that of the wild type in the absence of sordarin, while dph3Δ mutants show a slow-growth phenotype. Moreover, there are reports indicating that dph3Δ mutants also show drug and temperature hypersensitivity not described for the remaining dphΔ mutants (9, 26) (our unpublished data). These differences may be due to the involvement of Dph3p in regulating the functions of the Elongator complex, the well-recognized target of zymocin (8, 16). Both Dph3p and the Elongator complex have been shown to be required for posttranscriptional modification of uridine in tRNAs at the wobble position of the anticodon (16, 17). The absence of these modifications in the dph3Δ mutant and not in the other dphΔ mutants could account for the observed differences in sordarin resistance.
The mechanism by which diphthamide modification affects sordarin inhibition is not obvious, since residue H699 and the sordarin binding site are far away from each other in the eEF2 crystal structure (19, 20). While the binding site for sordarin on eEF2 is a pocket located between domains III, IV, and V (19), diphthamide has been mapped in the tip of domain IV in eEF2, protruding into the solvent, with no flexible elements spanning the two sites (20). In addition, cryo-electron microscopy maps of 80S-bound ADP-ribosylated eEF2-GDPNP or eEF2-GDPNP-sordarin do not support a conformational change in the factor due to the binding of sordarin (45). Additionally, it must be mentioned that some sordarin-resistant mutations in eEF2 have been mapped far from the sordarin binding site. These have been reported in residues R180/V187 (domain I), A475 (domain II), F612 (domain IV), and G790 (domain V) (5, 21), thus demonstrating that regions of eEF2 that are located relatively far from the binding site can affect the sordarin inhibitory action. In all these cases, the ribosome is probably playing an important role. Taken together, our results indicate that, modulated by the ribosomal particle, the absence of diphthamide has a long-range effect in eEF2, such that it decreases the affinity of sordarin for the ribosome-bound eEF2.
In sum, this is the first study identifying the diphthamide biosynthetic pathway as an effector of sordarin resistance. Collectively, the mutants uncovered by this screen represent a valuable collection of candidate genes, whose further investigation in the future may help to define additional cellular pathways participating in the responses of cells to sordarin treatment. Genes whose deletions scored as sensitive mutants in our screen may represent possible targets for the generation of novel sordarin derivatives. The identification of a set of genes whose loss of function leads to enhanced sordarin resistance in yeast may also provide avenues to investigate the orthologs of these targets in pathogenic fungal species, leading also to new opportunities for antifungal therapies.
We thank M. C. Fernández Moyano for expert technical assistance. We also thank GlaxoSmithKline for providing yeast mutants and sordarin derivatives and R. J. Collier for the plasmid expressing DTA.
This work was supported by grants GEN2001-4707-C08-01, AGL2005-07245-C03-03 (to J.L.R.), BMC2003-03387 (to J.P.G.B.), and BFU2004-03079 (to M.R.) from the Ministerio de Educación y Ciencia, Spain and by an institutional grant to the Centro de Biología Molecular Severo Ochoa from the Fundación Ramón Areces. M.R.-M. and J.B. were recipients of fellowships from the Ministerio de Educación y Ciencia and the Junta de Castilla y León, respectively.
Published ahead of print on 19 February 2008.
†Supplemental material for this article may be found at http://aac.asm.org/.