JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M603107200 on August 21, 2006

J. Biol. Chem., Vol. 281, Issue 42, 31616-31626, October 20, 2006
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
281/42/31616    most recent
M603107200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Reinke, A.
Right arrow Articles by Powers, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Reinke, A.
Right arrow Articles by Powers, T.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Caffeine Targets TOR Complex I and Provides Evidence for a Regulatory Link between the FRB and Kinase Domains of Tor1p*

Aaron Reinke, Jenny C.-Y. Chen, Sofia Aronova, and Ted Powers1

From the Section of Molecular and Cellular Biology, College of Biological Sciences, University of California, Davis, California 95616

Received for publication, March 31, 2006 , and in revised form, August 4, 2006.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The target of rapamycin (TOR) kinase is an important regulator of growth in eukaryotic cells. In budding yeast, Tor1p and Tor2p function as part of two distinct protein complexes, TORC1 and TORC2, where TORC1 is specifically inhibited by the antibiotic rapamycin. Significant insight into TORC1 function has been obtained using rapamycin as a specific small molecule inhibitor of TOR activity. Here we show that caffeine acts as a distinct and novel small molecule inhibitor of TORC1: (i) deleting components specific to TORC1 but not TORC2 renders cells hypersensitive to caffeine; (ii) rapamycin and caffeine display remarkably similar effects on global gene expression; and (iii) mutations were isolated in Tor1p, a component specific to TORC1, that confers significant caffeine resistance both in vivo and in vitro. Strongest resistance requires two simultaneous mutations in TOR1, the first at either one of two highly conserved positions within the FRB (rapamycin binding) domain and a second at a highly conserved position within the ATP binding pocket of the kinase domain. Biochemical and genetic analyses of these mutant forms of Tor1p support a model wherein functional interactions between the FRB and kinase domains, as well as between the FRB domain and the TORC1 component Kog1p, regulate TOR activity as well as contribute to the mechanism of caffeine resistance.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Rapamycin is an immunosuppressive and anti-proliferative antibiotic that targets the TOR2 kinase, a member of the phosphatidylinositol 3-kinase-like kinase family of protein kinases (1). TOR is a large (~280 kDa) protein and assembles into distinct membrane-associated protein complexes, termed TOR complex 1 (TORC1) and TORC2, in yeast as well as in higher eukaryotes (1). In yeast, TORC1 contains Tor1p or Tor2p as well as several additional proteins, including Kog1p, Lst8p, and Tco89p (2, 3). TORC1 mediates a multitude of rapamycin-sensitive activities related to cell growth, including control of translation, gene expression, and protein trafficking and stability (1, 2). TORC2 contains Tor2p as well as Lst8p, Avo1p-Avo3p, and Bit61p (2-5), and its activity is required for polarized cell growth and cytoskeletal organization (1, 2). TORC2 does not interact with nor is inhibited by rapamycin (1, 2). In higher eukaryotes, TORC1 consists of mammalian TOR (mTOR), mLST8/GbetaL (the ortholog of Lst8p), and Raptor (the ortholog of Kog1p), whereas TORC2 consists of mTOR, mLST8/GbetaL, and mAVO3/Rictor (the ortholog of Avo3p) (1, 6-9).

Much of what we know about TORC1 has come from the use of rapamycin as a specific small molecule inhibitor of TOR activity (1). Rapamycin, in conjunction with the highly conserved prolyl-isomerase FKBP, binds to the FRB domain of TOR located immediately N-terminal to its kinase domain (1, 10). A single amino acid change at a highly conserved serine residue within the FRB domain is sufficient to prevent binding of the rapamycin-FKBP complex to TOR and confer rapamycin resistance both in vivo and in vitro (1, 10-16). How rapamycin affects TORC1 activity remains unclear, although a number of in vitro studies have shown that the rapamycin-FKBP complex impairs TOR kinase activity (12, 17-21). Alternatively, it has been proposed that rapamycin prevents binding of a regulatory partner, for example Raptor, to mTOR that is essential for its activity (1, 7, 22-24). Interestingly, rapamycin-resistant mutant forms of mTOR have been shown to possess reduced kinase activity, prompting the suggestion that the FRB domain may somehow itself be involved in substrate recognition (19).

In addition to rapamycin, a number of other pharmacological agents have been shown to affect TOR, particularly in mammalian cells, including members of the methylxanthine family of compounds such as caffeine (25-27). Caffeine affects a diverse array of cellular processes related to cell growth, DNA metabolism, and cell cycle progression, most likely by acting as a low affinity ATP analog (28, 29). Both caffeine and the related compound theophylline have been shown to inhibit phosphorylation of mTOR-dependent substrates both in vitro as well as in vivo (25-27). Caffeine has not been used widely as a tool for probing mTOR function, however, potentially due to the pleiotropic behavior of this compound as well as the fact that it interacts with mTOR with relatively low affinity, i.e. in the submillimolar range (25-27).

In Saccharomyces cerevisiae, increased sensitivity to caffeine has been correlated with defects in the "cell integrity pathway," whereby cell well and/or plasma membrane stability is monitored in response to osmotic or thermal stress (30, 31). Several signaling pathways have been implicated in cell integrity maintenance, however, the relevant targets for caffeine have not yet been identified (30, 31). In this regard, we as well as others have reported that mutation of distinct components of TORC1 lead to defects in cellular integrity, including increased sensitivity to caffeine (3, 32). Independently, a large scale screen of yeast deletion mutants for altered drug sensitivities also identified TORC1 mutants as displaying increased caffeine sensitivity (33). Here we have pursued these observations and present evidence that TORC1 is indeed a significant target for caffeine in yeast. As part of these studies, we have identified mutations within the FRB and kinase domains of Tor1p that together confer significant levels of caffeine resistance. Characterization of these mutants provides independent evidence that the FRB domain is an important regulator of TOR kinase function.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Strains, Media, and General Methods—Strains and plasmids used in this study are listed in Tables 1 and 2, respectively. Cells were cultured in YPD (2% yeast extract, 1% peptone, and 2% dextrose) or synthetic complete dextrose (SCD) medium (0.8% yeast nitrogen base without amino acids, pH 5.5, 2% dextrose) supplemented with amino acids as described (34). Rapamycin (Sigma) was dissolved in Me2SO, and caffeine (Sigma) was dissolved in H2O; both compounds were added to liquid cultures and agar plates at the final concentrations indicated in the text and figure legends. Yeast transformations were performed using a lithium acetate procedure (35).


View this table:
[in this window]
[in a new window]
  		TABLE 1
Strains of S. cerevisiae used for this study

 


View this table:
[in this window]
[in a new window]
  		TABLE 2
Plasmids used this study

 
Construction of Yeast Strains—All gene deletions strains were constructed by replacing an entire open reading frame with a selectable maker following transformation of a linear fragment of DNA constructed by PCR. Specifically, we used His3MX6 as a selectable marker from plasmid pFA6a-His3MX6 (36) and used forward and reverse primers that contained 50 bp corresponding to the 5'- or 3'-ends of the target open reading frame, followed by 20 bp corresponding to the 5'- or 3'-ends of the marker gene. All disruptions were confirmed by PCR using different primers that could detect specific integration events for each linear DNA fragment. Kog1p, Tco89p, and Lst8p were each tagged at their C termini with multiple copies of the Myc epitope using a PCR-based gene-tagging method as described previously (4).

Plasmid Construction—Plasmid pPL132 was constructed in multiple steps. The starting plasmid was pYDF23, which carries the TOR1-1 allele under control of its native promoter (16). We first introduced sequences corresponding to three copies of the HA epitope (HA3) at the N terminus of the coding region of TOR1-1. This was accomplished by PCR amplification using genomic DNA from strain PLY298 (3) that expresses HA3-TOR1 to generate a linear fragment of DNA that encodes the promoter region of TOR1 followed by the N terminus of TOR1 fused to HA3. This fragment was used in a co-transformation along with gapped pYDF23 to generate an intact HA3-tagged TOR1-1 gene. Next, the FRB region of this plasmid was removed by digestion with BglII and NdeI followed by transformation of strain JK9-3da to recover the wild-type FRB sequences by gap repair. The success of each step was verified by restriction digestion, PCR, and DNA sequence analysis. The functionality of pPL132 was demonstrated by its ability to rescue cell integrity defects of strain PLY254 carrying a tor1 deletion in the W303a background (3).

Mutagenesis of pPL132—Hydroxylamine mutagenesis of pPL132 was carried out essentially as described (37). Briefly, 10 µg of plasmid was incubated at 75 °C in 1 M hydroxylamine for ~15 min, followed by transformation of W303a cells. Cells were allowed to recover for 6 h in SCD minus leucine media and were then plated on SCD minus leucine agar plates containing 9 mM caffeine and then incubated for several days at 30 °C. Control transformations onto plates lacking caffeine demonstrated that ~3 x 105 cells were transformed using this procedure. Plasmids were recovered from several colonies that appeared on caffeine-containing plates and were used to transform fresh W303a cells, followed by re-plating on 9 mM caffeine plates. Three plasmids that passed this criteria where subjected to a combination of Gap-repair and fragment exchange reactions, where in each case the caffeine-resistant phenotype was localized to an NcoI-BsiWI fragment that spanned the FRB and kinase domains of TOR1. This region was sequenced, and all three plasmids were found to possess nucleotide changes predicted to create an A1957V mutation in TOR1.

Error-prone PCR mutagenesis of pPL132 was carried out essentially as described (38) with several modifications. A 1583-bp fragment spanning the NcoI and BsiWI sites in TOR1 was amplified using pPL132 DNA as template. Mutagenic PCR conditions used 5 mM MnCl2, 2 mM each dCTP, dGTP, and dTTP, and 1 mM dATP. A 1383-bp fragment was excised from pPL132 by digestion with NcoI and BsiWI, and the resulting linearized plasmid vector was mixed with the PCR product in a 1:30 ratio and used to co-transform W303a cells. Following recovery in SCD minus leucine media, cells were plated directly onto SCD minus leucine agar plates containing 9 mM caffeine. Control transformations onto plates lacking caffeine demonstrated that ~3.5 x 105 cells were transformed using this procedure. Following several days of growth at 30 °C, plasmids were rescued from colonies that appeared on caffeine-containing plates, re-tested by transformation of fresh W303a cells, and then ultimately sequenced. Site-directed mutagenesis was carried out using a QuikChange XL II kit (Stratagene) and pPL132 DNA as a template, following instructions provided by the manufacturer.

Gene Expression Studies—Northern blot analysis was performed as described (39). A 10-ml culture was grown to mid log phase (A600 = 0.5) and harvested by centrifugation, and total RNA was extracted by performing a hot phenol method. 20 µg of total RNA was loaded onto a 1.5% agarose, 6.9% formaldehyde gel and run in 1x E buffer (20 mM MOPS, pH 7.0, 5 mM sodium acetate, 0.5 mM EDTA). The RNA was transferred overnight onto a Duralon-UV membrane (Stratagene, La Jolla, CA) and probed overnight at 65 °C in Church hybridization buffer (0.5 M NaPO4, pH 7.2, 7% SDS, 1 mM EDTA). Membranes were washed in 2x sodium saline citrate (1 x SSC = 0.15 M NaCl and 0.015 M sodium citrate) buffer, exposed to a phosphorimaging screen, and analyzed using a Storm 860 imaging system and software provided by the manufacturer (Amersham Biosciences). DNA templates for the Northern probes were made as described previously (39). Relative levels of specified mRNAs were normalized to actin (ACT1) mRNA.

For gene expression analysis using DNA microarrays, 250-ml cultures were harvested for each sample, poly(A) mRNA was isolated, and fluorescently labeled cDNAs were prepared essentially as described (40). Arrays were scanned using a GenePix 4000a microarray scanner (Axon Instruments, Inc.) and analyzed using GenePix Pro 3.0 provided by the manufacturer. Data were uploaded to an AMAD microarray data base using a Linux operating system, and analyzed using software Cluster 3.0 and Java Treeview (all publicly available at derisilab.ucsf.edu/microarray/index.html). All microarray data have been deposited in the NCBI Gene Expression Omnibus (GEO, www.ncbi.nlm.nih.gov/geo) and are available through GEO Series accession numbers GSE4584 [NCBI GEO] and GSE4586 [NCBI GEO] .

Immune Complex Kinase Assay and Co-immunoprecipitations—The protocol for kinase assays using PHAS-I as a substrate was developed based on previous reports (5, 19, 21) with several modifications based on our immunoprecipitation conditions for TORC1 (3). Two liters of each strain was grown in SCD minus leucine media to mid log phase (0.5 A600/ml) at 30 °C. Cells were washed with sterile H2O, pelleted again, and resuspended in yeast extract buffer (YEB, 50 mM HEPES-KOH, pH 7.1, 100 mM beta-glycerol phosphate, 50 mM NaF, 5 mM EGTA, 5 mM EDTA, 10% glycerol, 0.25% Tween 20, and 150 mM KCl). Cells were again pelleted, resuspended in 2 ml of YEB containing protease inhibitors (mixture tablet, Roche Applied Science), 2 mM dithiothreitol, and 2 mM phenylmethylsulfonyl fluoride and frozen dropwise by transfer pipette into liquid nitrogen. The cell pellet was then transferred to a pre-chilled mortar and pestle containing liquid nitrogen and ground into a fine powder (~150 strokes followed by the addition of liquid nitrogen, repeated three times). After the liquid nitrogen boiled off, the powder that remained was transferred to 1.5-ml microcentrifuge tubes and thawed. All of the following steps were performed at 4 °C unless stated otherwise. Extracts were then centrifuged twice at 20,000 x g for 20 min, transferring the supernatant to a new tube each time. 500 µl of each extract was then transferred into new tubes and pre-cleared by binding to 25 µl of protein A-coupled beads (Amersham Biosciences) for 45 min. 5 µl of affinity-purified anti-HA polyclonal antibody (3) and 25 µl of fresh protein A-coupled beads were incubated together at room temperature for 1 h and were added to each tube of cleared extract. Extract-bead mixtures were then incubated for 3 h at 4 °C with end-over-end rotation. Following this incubation, beads were washed five times with 1 ml of YEB. To each tube of beads, 56 µl of kinase buffer (YEB containing protease inhibitors (mixture tablet, Roche Applied Science)), 2 mM dithiothreitol, 4 mM MnCl2, and 1.5 µg of PHAS-I (Stratagene) substrate was added. Reactions were started by adding 4 µl of ATP mix (1.5 mM ATP, 2.5 µCi/µl[{gamma}-32P]ATP, 3000 Ci/mmol, Amersham Biosciences). Where indicated, caffeine was added at final concentrations listed in the text and figure legends prior to addition of ATP. Samples were incubated at 30 °C for 15 min, and reactions were then stopped by addition of 15 µl of 4x SDS-PAGE sample buffer. Samples were analyzed by SDS-PAGE on a 7.5%/12.5% discontinuous gel. The top of the gel was used for immunoblotting of Tor1p, and the bottom was stained with Coomassie Blue, exposed to a phosphorimaging screen, and analyzed using a Storm 860 imaging system and software provided by the manufacturer (Amersham Biosciences). Control experiments demonstrated that no background labeling of PHAS-I occurred in extracts prepared from tor1{Delta} cells that carried an empty plasmid vector (i.e. in the absence of HA3-TOR1), demonstrating the specificity of this system. Coimmunoprecipitation experiments were carried out as described previously (3, 4).


Figure 1
View larger version (61K):
[in this window]
[in a new window]
  		FIGURE 1.
TORC1 is a target for caffeine in yeast. A, wild-type (WT), tor1{Delta}, and tco89{Delta} cells derived from W303a and JK9-3da strain backgrounds were grown to mid log phase (A600 = 0.5), washed with sterile water, serially diluted, and plated onto a YPD agar plate or YPD agar plates that contained caffeine at the concentrations indicated and incubated at 30 °C for 3 days. Also shown are cells plated onto a YPD agar plate and incubated at 37 °C for ~2 days. B, wild-type and heterozygous diploid mutants, where each mutant was deleted for one copy of an indicated TORC1 and/or TORC2 component, all in the W303a/{alpha} background, were grown as described in A, plated onto YPD agar plates that contained the indicated concentration of caffeine, and incubated at 30 °C for 3 days. C, comparison of effects of rapamycin and caffeine on global gene expression, where untreated W303a cells were compared with cells treated with an indicated concentration of caffeine or rapamycin. Shown are results of hierarchical cluster analysis of genes whose expression is affected 4-fold or greater (log scale = 10) in at least one array. D, a subset of data from C showing DAL cluster genes, RTG target genes, and NDP genes (40-43) whose expression is increased following treatment with rapamycin or caffeine. E, response of three different strain backgrounds (W303, S288c, and {sum}2000) to rapamycin versus caffeine. Shown are results of hierarchical cluster analysis of genes whose expression is affected 4-fold or greater (log scale = 10) in at least three arrays, where untreated cells were compared with cells treated with 9 mM caffeine (denoted as "C") or 200 ng/ml rapamycin (denoted as "R"). In C and D, cells were grown in YPD to mid log phase (A600 = 0.5) and treated with rapamycin or caffeine, as indicated, for 30 min. All data for C and D are available under GEO accession number GSE4584.

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
TORC1 Is a Target for Caffeine in Yeast—We reported recently that cells singly deleted for genes encoding two non-essential TORC1-specific components, Tor1p or Tco89p, display a number of cell integrity-related phenotypes, including temperature sensitive (ts) growth, hypersensitivity to the cell wall binding agent calcofluor white, and hypersensitivity to caffeine (3) (Fig. 1A). Both the ts and calcofluor white-sensitive phenotypes were restricted to the W303a strain background that contains a mutant allele of the SSD1 gene, termed ssd1-d, which is known to sensitize mutants for cell integrity defects (3) (Fig. 1A). By contrast, here we have found that hypersensitivity to caffeine was observed for tor1{Delta} and tco89{Delta} mutants in both the W303a and JK9-3da strain backgrounds, where the latter possesses a wild-type SSD1 gene, suggesting caffeine has a more direct influence on TORC1 (Fig. 1A).

Consistent with this conclusion, a recent high throughput analysis of yeast heterozygous diploid mutants for altered drug sensitivities also identified mutants in three TORC1 genes, TOR1, KOG1, and TCO89, as being caffeine-hypersensitive (33) (note that this study refers to TCO89 using the obsolete name SHD7). Importantly, this study was conducted using strains derived from the S288c strain background, which carries a functional SSD1 allele (33).

We tested the generality of these findings by constructing diploid strains that contained single gene disruptions of individual components of TORC1 and/or TORC2 and then examined these mutants for sensitivity to caffeine. Consistent with previous findings (33), deletion of components specific to TORC1 (e.g. TOR1, KOG1, and TCO89) resulted in the greatest sensitivity to caffeine (Fig. 1B). In addition, we found that deletion of components that are shared between TORC1 and TORC2 (e.g. TOR2 and LST8) also displayed hypersensitivity but at a higher concentration of caffeine (Fig. 1B). By contrast, deletion of components specific to TORC2 (e.g. AVO1 and AVO3) showed no increased sensitivity to caffeine, compared with wild-type cells (Fig. 1B and data not shown).

Treating yeast cells with rapamycin results in widespread and characteristic changes in gene expression (1, 40-43). We reasoned that, if caffeine targets TORC1, then treating cells with this drug might elicit effects similar to rapamycin with respect to these transcriptional readouts. Accordingly, we examined global transcriptional changes when wild-type W303 cells were grown in rich media (YPD) and treated with either rapamycin or caffeine (Fig. 1C). Indeed, remarkable overlap was observed with respect to transcriptional changes caused by treatment with either drug (Fig. 1C). Importantly, these similarities included not only repression of r-protein gene expression but also included increased expression of distinct nitrogen-regulated gene clusters, for example RTG target, DAL, and nitrogen discrimination pathway genes, that together are characteristic of the cellular response to rapamycin (Fig. 1D) (40-43). A survey of three different wild-type strain backgrounds confirmed the overall similarity of the transcriptional response to treatment with rapamycin and caffeine (Fig. 1E). A number of strain-specific differences were also observed for each drug, indicating that genetic factors in addition to TORC1 contribute to the precise response of cells to rapamycin as well as caffeine (Fig. 1E). This observation is consistent with the finding that different wild-type strain backgrounds display unique basal sensitivities to both rapamycin as well as caffeine (3) (Fig. 1A).

Isolation of Caffeine-resistant Alleles of TOR1—We reasoned that, if TORC1 was a target for caffeine, then it might be possible to isolate mutants in TOR1 that afforded increased resistance to the drug. In a preliminary experiment, we transformed wild-type W303a cells with a hydroxylamine-treated plasmid that expressed an epitope-tagged TOR1 gene (HA3-TOR1) and grew the resulting transformed cells on agar plates containing 9 mM caffeine, a concentration that was completely inhibitory for growth for this strain background (see "Materials and Methods"). Three independently derived, plasmid-dependent caffeine-resistant colonies were obtained, and subsequent analyses revealed that each TOR1 gene possessed a mutation that was predicted to create an alanine to valine substitution at position 1957 within the FRB domain of Tor1p. Site-directed mutagenesis was used to confirm that this A1957V single amino acid substitution was necessary as well as sufficient for growth of cells in the presence of 9 mM caffeine (Table 3).


View this table:
[in this window]
[in a new window]
  		TABLE 3
Caffeine-resistant phenotypes of TOR1 mutants

W303a (WT) and PLY254 (tor1{Delta}) were transformed with the indicated plasmids and grown on SCD minus leucine plates that contained the indicated concentrations of caffeine. Following 2-4 days of growth, colony sizes were scored as follows: +++, strong growth; ++, moderate growth; +/–, weak growth; 0, no growth.

 
Based on these results, we mutagenized the FRB as well as adjacent kinase domains using error-prone PCR in an effort to generate additional caffeine-resistant TOR1 alleles (see "Materials and Methods"). Six independent TOR1 mutants were isolated that allowed W303a cells to grow on agar plates containing up to 12 mM caffeine, two of which are shown in Fig. 2A. None of these mutants conferred increased resistance to a number of other drugs tested, including doxycycline, chlorpromazine, camptothecin, cycloheximide, and hygromycin B, suggesting they were specific for resistance to caffeine.3 Sequence analysis revealed that, although each mutant contained a number of different predicted amino acid substitutions, all shared similar substitutions at two specific positions: (i) a change at position isoleucine 1954 within the FRB domain to either valine (two isolates) or threonine (four isolates); (ii) a change at position tryptophan 2174 within the kinase domain to either arginine (five isolates) or cysteine (one isolate) (Fig. 2B). Site-directed mutagenesis was used to demonstrate that a combination of two mutations, I1954V and W2176R, was both necessary as well as sufficient for growth in the presence of 12 mM caffeine (Table 3). Each caffeine-resistant allele conferred similar levels of resistance in both wild-type as well as tor1{Delta} cells, where in the latter strain the sole source of TOR1 was plasmid-encoded (Table 3; see below). We note that no mutants were isolated that afforded growth of W303a wild-type or tor1{Delta} cells on agar plates containing a concentration of caffeine higher than 12 mM, suggesting we had encountered one or more unknown constraints by this approach.


Figure 2
View larger version (50K):
[in this window]
[in a new window]
  		FIGURE 2.
Identification and characterization of caffeine-resistant TOR1 mutants. A, WT (W303a) cells were transformed with a control plasmid (pRS315) or plasmids that expressed wild-type HA3-TOR1 (pPL132) or either one of two caffeine-resistant TOR1 mutants, HA3-TOR1-1a1 or HA3-TOR1-1a7 (pPL164 and pPL165, respectively). Cells were grown in SCD media lacking leucine at 30 °C to mid log phase, serial dilutions were made, and cells were plated onto SCD minus leucine agar plates that contained the indicated concentrations of caffeine. Photographs of plates were taken following ~3 days of growth. B, alignments of segments of the FRB and kinase domains of selected TOR homologues showing the location of amino acids identified in this study where mutations result in caffeine resistance. Also shown is position Ser-1972 within the FRB domain, which is the site of mutations that confer rapamycin resistance, as well as position Leu-2174 within the kinase domain, which corresponds to the "gatekeeper" position as defined by Shokat and co-workers (45). Numbering of amino acids is according to Tor1p. C, Northern blot analysis of tor1{Delta} (PLY254) cells transformed with plasmids expressing either HA3-TOR1 (pPL132) or HA3-TOR1 (I1954V/W2176R) (pPL158). Cells were grown to mid log phase in SCD minus leucine media at 30 °C and treated with rapamycin or caffeine at the indicated concentrations for 30 min. Cells were harvested, and total RNA was isolated and used for Northern blot analysis, probing for the specified mRNAs. D, effects of caffeine on global gene expression in tor1{Delta} (PLY254) cells expressing either HA3-TOR1 or HA3-TOR1 (I1954V/W2176R). Shown are results of hierarchical cluster analysis, based on genes whose expression was affected 4-fold or greater (log scale = 10) in cells expressing HA3-TOR1 treated with 6 and 9 mM caffeine. Data for D are available under GEO accession number GSE4586.

 
Interestingly, we observed that the I1954V and W2176R double mutant enabled tor1{Delta} cells derived from the JK9-3da background to grow on concentrations of caffeine up to 20 mM (data not shown). This finding is consistent with our above conclusion that genetic factors in addition to TORC1 contribute to the detailed response of cells to caffeine; however, as also stated above, these genetic factors are likely to be distinct from the SSD1 gene.

We compared the effects of caffeine on gene expression in cells expressing the HA3-TOR1 (I1954V/W2176R) double mutant versus wild-type HA3-TOR1. Unless stated otherwise, for this and all remaining experiments, we used a tor1{Delta} strain so that the sole source of TOR1 would be plasmid-encoded. Northern blot analysis was used to examine expression of two representative target genes, CIT2 and RPL30, which are negatively and positively regulated by TORC1, respectively (39, 42) (Fig. 2C). We observed a significant difference in the response of these genes to caffeine within the two strains, where a higher concentration of drug was required to elicit a comparable response in cells expressing TOR1 (I1954V/W2176R) (Fig. 2C). Microarray analysis of global gene expression patterns confirmed that, for essentially all genes affected by caffeine, a significantly higher dose of drug was also required to show a comparable change in expression in cells expressing mutant versus wild-type TOR1 (Fig. 2D). By contrast, no change in response to rapamycin was observed for these strains with respect to CIT2 or RPL30 expression (Fig. 2C, compare lanes 2 and 9), consistent with our finding that the TOR1 (I1954V/W2176R) mutant does not confer rapamycin resistance (see below).

Functional Dissection of Tor1p FRB and Kinase Domain Mutations—Wesought to explore further the molecular basis for the caffeine resistant phenotypes observed for both the TOR1 (I1954V/W2176R) double as well as TOR1 (A1957V) single mutants described above. Toward this end, we constructed versions of TOR1 that expressed the I1954V FRB domain and W2176R kinase domain mutations singly, and we constructed a double mutant that combined the A1957V FRB and W2176R kinase domain mutations, and then examined the phenotypes of these mutants under a variety of conditions (Fig. 3, A and B). We observed that both the I1954V and A1957V FRB domain single mutants afforded a comparable level of caffeine resistance, where cells expressing these alleles grew at 9 mM but not 12 mM caffeine (Fig. 3A). Interestingly, the W2176R kinase domain mutation in combination with the A1957V FRB mutation also allowed growth at 12 mM caffeine, demonstrating that this kinase domain mutation could be combined with either of the two FRB domain mutations to confer increased caffeine resistance (Fig. 3B). By contrast, the W2176R kinase domain on its own conferred no caffeine resistance and, moreover, displayed both temperature-sensitive growth as well as rapamycin hypersensitivity, compared with wild-type HA3-TOR1, suggesting that this mutation impaired Tor1p function (Fig. 3A). Interestingly, the I1954V/W2176R double mutant was not temperature-sensitive, but it did display rapamycin hypersensitivity (Fig. 3A). These differences in phenotypes are unlikely to be explained by differences in protein stability, because Western blot analysis of cell extracts demonstrated that each mutant produced comparable steady-state levels of Tor1p (Fig. 3C).


Figure 3
View larger version (60K):
[in this window]
[in a new window]
  		FIGURE 3.
Molecular basis for caffeine resistance by identified TOR1 mutants. A and B, tor1{Delta} (PLY254) cells were transformed with a control plasmid (pRS315) or plasmids that expressed wild-type HA3-TOR1 (pPL132) or mutant HA3-TOR1 alleles as indicated. Cells were grown in SCD minus leucine media to mid log phase at 30 °C, serially diluted, and plated out on SCD minus leucine agar plates that contained the indicated concentrations of caffeine or rapamycin. All plates were incubated at 30 °C, except for the indicated plate that was incubated at 37 °C. Photographs of plates were taken following ~3 days of growth. C, Western blot analysis of HA3-Tor1p expressed in strains used in A. Cells were grown in SCD minus leucine media to mid log phase at 30 °C, harvested, and processed for Western blot analysis as described (3). D, testing the ability of different TOR1 alleles to suppress the synthetic lethality of tor1{Delta} tco89{Delta} cells. Strain PLY365 was transformed with a control plasmid (pRS315) or plasmids that expressed TOR2 (pJK4), wild-type HA3-TOR1 (pPL132), or mutant HA3-TOR1 alleles as indicated. Equivalent amounts of cells were then patched onto SCD minus leucine agar plates that either lacked or contained 1 mM 5-fluororotic acid to counter select for the TOR2 URA3 plasmid (pNB100) present in PLY365. In E: Upper panels, results of immune complex kinase assays where Tor1p kinase activity is measured from tor1{Delta} (PLY254) cells that expressed the indicated alleles of TOR1. The blot is a representative experiment showing the amount of 32P incorporated into PHAS-I, and the graph shows the average of three independent experiments, where error bars denote the standard deviation of the mean. Lower panels, Western blot analysis of extracts depicts levels of Tor1p present in each kinase assay.

 
We tested further the functionality of the FRB and kinase domain mutants by examining their ability to suppress the synthetic lethality of a tor1{Delta} tco89{Delta} double mutant (3). As reported previously (3), a wild-type copy of TOR2 carried on a LEU2 plasmid allows tor1{Delta} tco89{Delta} cells to drop a TOR2 URA3 plasmid, as judged by the appearance of colonies following counter selection against the URA3 gene on agar plates containing 5-fluororotic acid (Fig. 3D). Similarly, we observed that the HA3-TOR1 LEU2 (pPL132) plasmid also allowed for production of 5-fluororotic acid-resistant colonies with equal efficiency as the TOR2 LEU2 plasmid (Fig. 3D). By contrast, neither the W2176R kinase domain single mutant nor the I1954V/W2176R double mutant allowed production of 5-fluororotic acid-resistant colonies, consistent with their impaired phenotypes described above. Remarkably, for both single FRB mutants, I1954V and A1957V, reproducibly more 5-fluororotic acid-resistant colonies were obtained in this assay, suggesting these alleles allowed tor1{Delta} tco89{Delta} cells to drop the TOR2 URA3 plasmid with greater efficiency than wild-type TOR1 or TOR2 (Fig. 3D).


Figure 4
View larger version (23K):
[in this window]
[in a new window]
  		FIGURE 4.
An activating FRB domain mutation influences interactions between Tor1p and Kog1p. A, immunoprecipitation (IP) of HA3-T or1p (left panels) and co-immunoprecipitation of Kog1p-Myc (right panels) from extracts prepared from strain PLY671 expressing the indicated plasmids. B, quantification of data, where levels of Kog1p-Myc present in unbound (U) and bound (B) fractions are expressed as a fraction of the total amount of Kog1p-Myc present in each reaction. Similar results were observed in two independent sets of experiments.

 
To gain insight into the molecular basis for these observed phenotypic differences, we examined the kinase activity of wild-type as well as several mutant HA3-Tor1p proteins using an immune complex kinase assay that employed mammalian PHAS-I protein as a model substrate (5, 19, 21; see "Materials and Methods"). Western blot analysis confirmed that an equivalent amount of Tor1p protein was used in each assay (Fig. 3E, bottom panels). We observed that the W2176R kinase domain mutant had severely diminished activity, in comparison to wild-type Tor1p (Fig. 3E, upper panels, compare lanes 1 and 4). By contrast, the I1954V/W2176R double mutant displayed an intermediate activity, where it was improved relative to the kinase domain mutation alone yet possessed ~30% of the activity of wild-type (Fig. 3E, upper panels, compare lanes 1, 4, and 5). Remarkably, both of the single FRB domain mutations displayed a significant (~2- to 3-fold) increase in kinase activity compared with wild-type (Fig. 3E, upper panels, compare lane 1 with lanes 2 and 3). Thus, in general there was remarkable agreement in terms of the in vivo phenotypes observed for these mutants and their in vitro kinase activity.

To extend these results further, we asked whether the behavior of the FRB and/or kinase domain mutations correlated with differences in the integrity of TORC1 by performing co-immunoprecipitation experiments. For this, we introduced plasmids that expressed wild-type HA3-TOR1, the single I1954V FRB domain mutation, the single W2176R kinase domain mutation, or an untagged control vector, into strains expressing Mycepitope-tagged versions of Kog1p, Lst8p, or Tco89p. Extracts were prepared from these strains and used for immunoprecipitations using anti-HA antibody directed against HA3-Tor1p, followed by Western blotting using anti-Myc antibody to assess the degree of co-precipitation of the different TORC1 components. We performed these experiments in strains expressing endogenous Tor1p, reasoning that competition for assembly with wild-type protein might reveal subtle differences.

Indeed, we observed that the I1954V FRB mutant afforded a subtle but reproducible increase in the amount of co-precipitating Kog1p-Myc, compared with wild-type HA3-Tor1p (Fig. 4, A and B). By contrast, no significant differences were observed between wild-type and the FRB domain mutant with respect to interactions with Lst8p-Myc or Tco89p-Myc (data not shown). Similarly, no significant differences were observed for interactions between the W2176R kinase domain mutant and any of the TORC1 partners (Fig. 4, A and B, and data not shown). Thus, we conclude that the overall stability of TORC1 is not significantly impacted by either of these Tor1p mutations; rather, the activating I1954V FRB domain mutation results in only subtle modulation of interactions between Tor1p and Kog1p.

Rapamycin-resistant TOR Alleles Display Several Impaired Phenotypes—We sought to determine whether caffeine resistance was in general associated with mutations within the FRB domain by examining the behavior of the S1972R rapamycinresistant TOR1-1 allele. Toward this end, we transformed tor1{Delta} cells with plasmids that expressed TOR1, TOR1-1, or a control vector. We observed that, although the TOR1-1 plasmid conferred an expected level of rapamycin resistance, this mutant was hypersensitive to caffeine, relative to wild-type TOR1 (Fig. 5A). Similar results were also obtained using rapamycin-resistant strains in the JK9-3da background that expressed chromosomal TOR1-1 or TOR2-1 mutant genes (Fig. 5B). Two other observations indicated the TOR1-1 allele encodes a Tor1p protein with impaired activity. First, we observed that this allele was unable to suppress the synthetic lethality of the tor1{Delta} tco89{Delta} double mutant, in contrast to wild-type TOR1 (Fig. 3D). Second, Tor1-1p displayed reduced in vitro kinase activity relative to wild-type Tor1p in the immune complex kinase assay (Fig. 3E). Together these results demonstrate that adjacent mutations within the FRB domain lead to different phenotypes and suggest that this domain modulates Tor1p activity in distinct ways. This conclusion is also consistent with our finding that the caffeine-resistant FRB domain mutations do not on their own confer increased rapamycin resistance (Fig. 3A and data not shown).


Figure 5
View larger version (33K):
[in this window]
[in a new window]
  		FIGURE 5.
Rapamycin-resistant alleles of TOR are hypersensitive to caffeine. A, tor1{Delta} (PLY254) cells were transformed with a control plasmid (pRS315) or plasmids that expressed wild-type TOR1 (pPL132) or TOR1-1 (pYDF23) and streaked out onto SCD minus leucine agar plates, which contained the indicated concentrations of caffeine or rapamycin. B, wild-type (JK9-3da), TOR1-1 (3H11-1c), and TOR2-1 (JH12-17b) cells were streaked out onto YPD agar plates containing the indicated concentrations of caffeine or rapamycin. For A and B, plates were photographed following incubation at 30 °C for ~3 days. Note that different concentrations of caffeine were used in A and B due to differences in basal sensitivities of W303a versus JK9-3da.

 


Figure 6
View larger version (33K):
[in this window]
[in a new window]
  		FIGURE 6.
Caffeine-resistant mutations in Tor1p affect interactions with caffeine in vitro. Immune complex kinase assay shows activities of wild-type HA3-Tor1p and HA3-Tor1p (I1954V/W2176R) in the presence of the indicated concentrations of caffeine. Blots show representative samples of the amount of 32P incorporated into PHAS-I, and the graph depicts the average of three independent experiments, where error bars denote ±S.D. Data were fit using linear regression to calculate an IC50 by caffeine for each kinase (26). Note that the exposure of the two blots has been adjusted so that the intensities of samples without caffeine appear equivalent for both WT HA3-Tor1p and HA3-Tor1p (I1954V/W2176R), to compensate for the reduced activity of the mutant, and has no effect on calculation of IC50 values.

 
Tor1p (I1954V/W2176R) Double Mutant Confers Caffeine Resistance in Vitro—To begin to test for physical interactions between Tor1p and caffeine, we used the immune complex kinase assay described above to determine whether caffeine inhibited Tor1p kinase activity, and, if so, whether this inhibition was influenced by the I1954V and W2176R mutations. We observed that caffeine inhibited the kinase activity of HA3-Tor1p with an IC50 of 0.28 mM (Fig. 6, left panels), in very good agreement with the reported IC50 of ~0.4 mM caffeine for inhibition of mTOR (26). By contrast, the IC50 for inhibition of HA3-Tor1p (I1954V/W2176R) was increased ~17-fold to 4.9 mM (Fig. 6, right panels). Based upon these results we conclude that together the I1954V and W2176R mutations in Tor1p result in impaired interactions with caffeine in vitro that is likely to contribute, at least in part, to caffeine-resistant growth of cells expressing this mutant in vivo.

FRB and Kinase Domain Mutations Likely Act Exclusively in cis—Both TORC1 as well as TORC2 have been proposed to function as dimers (1, 5). In addition, genetic evidence from Drosophila indicates that TOR in higher eukaryotes may also function as a multimer (44). Given these findings, we were interested in determining whether the I1954V FRB/W2176R kinase domain mutations function strictly within the same TOR1 gene product to confer high levels of caffeine resistance or whether they might also function in trans. Toward this end, we introduced the I1954V and W2176R mutations into cells on separate TOR1 CEN/ARS plasmids and asked whether these cells were able to grow on agar plates containing 12 mM caffeine. Caffeine-resistant colonies were indeed obtained, however, sequence analysis of plasmids retrieved from these cells revealed that in every case growth at 12 mM caffeine correlated with a recombination event between the plasmids such that both mutations were expressed in the context of a single TOR1 gene (data not shown). Based on this observation, we conclude that these mutations are likely to act strictly within the context of a single Tor1p protein to confer highest levels of caffeine resistance.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
We have demonstrated that TORC1-dependent signaling represents an important target for caffeine in yeast. Our analysis of caffeine-resistant TOR1 alleles has been particularly informative in that it demonstrates the close correspondence between the effects of rapamycin and caffeine on global gene expression may be directly attributable to inhibition of TORC1. Of potentially greater importance, however, is our finding that caffeine-resistant mutations within the FRB domain result in increased Tor1p function, as defined both by a number of in vivo phenotypes as well as by their in vitro kinase activity. Together with our finding that the rapamycin-resistant FRB domain mutation instead causes impaired Tor1p function, as has been shown previously for mTOR (19), these results underscore the importance of the FRB domain as an important modulator of TOR activity.


Figure 7
View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 7.
Working model for mechanism of caffeine resistance for Tor1p mutants. Model shows possible interactions between FRB and kinase domain mutations that together contribute to highest levels of caffeine resistance, which may also involve interactions with Kog1p. For clarity, only the FRB and kinase domains of Tor1p are shown. See text for details.

 
We can begin to account for the caffeine resistance conferred by the TOR1 I1954V/W2176R as well as A1957V/W2176R double mutants by the working model presented in Fig. 7. According to this model, the W2176R kinase domain mutation impairs interactions with both caffeine as well as ATP, resulting in a Tor1p with severely impaired kinase activity. This suggestion is consistent with the proximity of Trp-2176 to Leu-2174, where the latter amino acid corresponds to the so-called "gatekeeper" identified recently by Shokat and co-workers (45) as likely being located within the ATP binding pocket and involved in nucleotide recognition within the phosphatidylinositol 3-kinase and PIKK protein kinase families. When the kinase domain mutation is combined with either one of the kinase-activating FRB mutations, however, the resulting double mutants now possess sufficient kinase activity to both support adequate Tor1p function as well as display significant caffeine resistance. By themselves, the FRB domain mutants are able to provide limited caffeine resistance, presumably due to their increased kinase activity. Consistent with this interpretation is the fact that the S1972R rapamycin-resistant TOR1-1 allele possesses diminished kinase activity and is hypersensitive to caffeine.

Our results also begin to address the molecular basis by which caffeine-resistant mutations within the FRB domain influence Tor1p kinase function with the finding that the I1954V mutation results in increased association between Tor1p and its partner Kog1p. This result is consistent with previous findings for mTOR that the FRB domain is an important regulatory point for interactions with Raptor and, moreover, that Raptor is an important regulator of mTORC1 activity (6-8). The relationship between the FRB domain of Tor1p and Raptor/Kog1p is likely to be complicated, however, because available evidence suggests Raptor is unlikely to interact directly with the FRB domain of mTOR (7). We point out that our findings do not rule out the possibility that direct interactions between the FRB and kinase domains may additionally account the mutant phenotypes described here. The development of a more sensitive in vitro reconstituted system will be important for addressing these questions, as well as for exploring the exact mode of inhibition of TORC1 by caffeine.

It is likely that caffeine will turn out to interact with other components in yeast as well, including Tor2p within the context of TORC2. For example, an allele of TOR2 has been described that is hypersensitive to caffeine and that impairs receptor-mediated endocytosis by what appears to be a TORC2-specific pathway (46). Similarly, temperature-sensitive mutations have been identified in the TORC2-specific component AVO3 that also display caffeine hypersensitivity (47).3 These findings are consistent with the fact that TORC2 is also strongly linked to cell integrity maintenance (5, 30). It is therefore unclear why we did not observe increased sensitivity to caffeine in AVO1/avo1{Delta} or AVO3/avo3{Delta} heterozygous mutants compared with wild type (Fig. 1A). One possibility is that TORC2 is less sensitive to changes in the dosage of these TORC2-specific genes compared with components of TORC1, at least with respect to this phenotype. A prediction of these findings is that it should be possible to isolate caffeine-resistant alleles of TOR2. Indeed, in a preliminary study we have obtained TOR2 mutants that confer limited caffeine resistance.3 Interestingly, these mutations appear to map to a distinct region of the kinase domain of TOR2. Whether these alleles affect TORC1 and/or TORC2 readouts remains to be determined.

In addition to mTOR, caffeine has been shown to inhibit the in vitro kinase activity of a number of other mammalian PIKKs, including ATM, ATR, and DNA-PK (26, 48, 49). Inhibition of ATM and ATR by caffeine has been correlated with the ability of the drug to suppress DNA damage-induced cell cycle checkpoints (26). This correlation has been questioned recently, however, following observations that several substrates of ATM and/or ATR are either not affected or instead become hyperphosphorylated following caffeine treatment in vivo (28, 29). Thus, the target for caffeine with respect to DNA damage-induced cell cycle arrest apparently remains to be identified (29). By contrast, there is a better correlation between the in vitro and in vivo effects of caffeine on mTOR-dependent substrates (19, 25). Together with our results presented here, we believe that inhibition by caffeine is likely to represent a common feature of TOR signaling. Here it is relevant that the TOR1 mutations identified in this study correspond to highly conserved amino acids within TOR proteins from across the phylogenetic spectrum (Fig. 2B). Thus, it should be possible to identify and/or construct caffeine-resistant TOR alleles in other organisms as well. Such an approach may prove helpful in distinguishing the effects of caffeine on TOR-dependent versus TOR-independent readouts in other systems.


   FOOTNOTES
 
* This work was supported by National Sciences Foundation Grant MCB-1031221 and by American Cancer Society Research Scholar Grant RSG-04-075-01-TBE (to T. P.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

1 To whom correspondence should be addressed. Tel.: 530-754-5052; Fax: 530-752-3085; E-mail: tpowers{at}ucdavis.edu.

2 The abbreviations used are: TOR, target of rapamycin; mTOR, mammalian TOR; SCD, synthetic complete dextrose; HA, hemagglutinin; MOPS, 4-morpholinepropanesulfonic acid; GEO, NCBI Gene Expression Omnibus; PHASI, phosphorylated heat and acid stable protein regulated by insulin. Back

3 A. Reinke, J. C.-Y. Chen, S. Aronova, and T. Powers, unpublished results. Back


   ACKNOWLEDGMENTS
 
We thank S. Fairclough, C. Iverson, and K. P. Wedaman for their technical assistance in constructing plasmid pPL132 and thank C. Iverson and C. Cortez for helping to establish conditions for efficient PCR mutagenesis. We thank B. Bonner for his help with co-immunoprecipitation experiments and thank members of the Powers laboratory and J. Nunnari for discussions.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Wullschleger, S., Loewith, R., and Hall, M. N. (2006) Cell 124, 471-484[CrossRef][Medline] [Order article via Infotrieve]
  2. Loewith, R., Jacinto, E., Wullschleger, S., Lorberg, A., Crespo, J. L., Bonenfant, D., Oppliger, W., Jenoe, P., and Hall, M. N. (2002) Mol. Cell 10, 457-468[CrossRef][Medline] [Order article via Infotrieve]
  3. Reinke, A., Anderson, S., McCaffery, J. M., Yates, J., 3rd, Aronova, S., Chu, S., Fairclough, S., Iverson, C., Wedaman, K. P., and Powers, T. (2004) J. Biol. Chem. 279, 14752-14762[Abstract/Free Full Text]
  4. Wedaman, K. P., Reinke, A., Anderson, S., Yates, J. I., McCaffery, J. M., and Powers, T. (2003) Mol. Biol. Cell 14, 1204-1220[Abstract/Free Full Text]
  5. Wullschleger, S., Loewith, R., Oppliger, W., and Hall, M. N. (2005) J. Biol. Chem. 280, 30697-30704[Abstract/Free Full Text]
  6. Hara, K., Maruki, Y., Long, X., Yoshino, K., Oshiro, N., Hidayat, S., Tokunaga, C., Avruch, J., and Yonezawa, K. (2002) Cell 110, 177-189[CrossRef][Medline] [Order article via Infotrieve]
  7. Kim, D.-H., Sarbassov, D. D., Ali, S. M., King, J. E., Latek, R. R., Erdjument-Bromage, H., Tempst, P., and Sabatini, D. M. (2002) Cell 110, 163-175[CrossRef][Medline] [Order article via Infotrieve]
  8. Kim, D.-H., Sarbassov, D. D., Ali, S. M., Latek, R. R., Guntur, K. V. P., Erdjument-Bromage, H., Tempst, P., and Sabatini, D. M. (2003) Mol. Cell 11, 895-904[CrossRef][Medline] [Order article via Infotrieve]
  9. Sarbassov, D. D., Ali, S. M., Kim, D. H., Guertin, D. A., Latek, R. R., Erdjument-Bromage, H., Tempst, P., and Sabatini, D. M. (2004) Curr. Biol. 14, 1296-1302[CrossRef][Medline] [Order article via Infotrieve]
  10. Schmelzle, T., and Hall, M. N. (2000) Cell 103, 253-262[CrossRef][Medline] [Order article via Infotrieve]
  11. Chen, J., Zheng, X. F., Brown, E. J., and Schreiber, S. L. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 4947-4951[Abstract/Free Full Text]
  12. Brown, E. J., Beal, P. A., Keith, C. T., Chen, J., Shin, T. B., and Schreiber, S. L. (1995) Nature 377, 441-446[CrossRef][Medline] [Order article via Infotrieve]
  13. Heitman, J., Movva, N. R., and Hall, M. N. (1991) Science 253, 905-909[Abstract/Free Full Text]
  14. Lorenz, M. C., and Heitman, J. (1995) J. Biol. Chem. 270, 27531-27537[Abstract/Free Full Text]
  15. Stan, R., McLaughlin, M. M., Cafferkey, R., Johnson, R. K., Rosenberg, M., and Livi, G. P. (1994) J. Biol. Chem. 269, 32027-32030[Abstract/Free Full Text]
  16. Zheng, X.-F., Fiorentino, D., Chen, J., Crabtree, G. R., and Schreiber, S. L. (1995) Cell 82, 121-130[CrossRef][Medline] [Order article via Infotrieve]
  17. Brunn, G. J., Hudson, C. C., Sekulic, A., Williams, J. M., Hosoi, H., Houghton, P. J., Lawrence, J. C., Jr., and Abraham, R. T. (1997) Science 277, 99-101[Abstract/Free Full Text]
  18. Scott, P. H., Brunn, G. J., Kohn, A. D., Roth, R. A., and Lawrence, J. C., Jr. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 7772-7777[Abstract/Free Full Text]
  19. McMahon, L. P., Choi, K. M., Lin, T. A., Abraham, R. T., and Lawrence, J. C., Jr. (2002) Mol. Cell. Biol. 22, 7428-7438[Abstract/Free Full Text]
  20. Crespo, J. L., and Hall, M. N. (2003) Microbiol. Mol. Biol. Rev. 66, 579-591
  21. Alarcon, C. M., Heitman, J., and Cardenas, M. E. (1999) Mol. Biol. Cell 10, 2531-2546[Abstract/Free Full Text]
  22. Schalm, S. S., Fingar, D. C., Sabatini, D. M., and Blenis, J. (2003) Curr. Biol. 13, 797-806[CrossRef][Medline] [Order article via Infotrieve]
  23. Yonezawa, K., Tokunaga, C., Oshiro, N., and Yoshino, K. (2004) Biochem. Biophys. Res. Commun. 313, 437-441[CrossRef][Medline] [Order article via Infotrieve]
  24. Nojima, H., Tokunaga, C., Eguchi, S., Oshiro, N., Hidayat, S., Yoshino, K., Hara, K., Tanaka, N., Avruch, J., and Yonezawa, K. (2003) J. Biol. Chem. 278, 15461-15464[Abstract/Free Full Text]
  25. Scott, P. H., and Lawrence, J. C., Jr. (1998) J. Biol. Chem. 273, 34496-34501[Abstract/Free Full Text]
  26. Sarkaria, J. N., Busby, E. C., Tibbetts, R. S., Roos, P., Taya, Y., Karnitz, L. M., and Abraham, R. T. (1999) Cancer Res. 59, 4375-4382[Abstract/Free Full Text]
  27. McMahon, L. P., Yue, W., Santen, R. J., and Lawrence, J. C., Jr. (2005) Mol. Endocrinol. 19, 175-183[Abstract/Free Full Text]
  28. Kaufmann, W. K., Heffernan, T. P., Beaulieu, L. M., Doherty, S., Frank, A. R., Zhou, Y., Bryant, M. F., Zhou, T., Luche, D. D., Nikolaishvili-Feinberg, N., Simpson, D. A., and Cordeiro-Stone, M. (2003) Mutat. Res. 532, 85-102[Medline] [Order article via Infotrieve]
  29. Cortez, D. (2003) J. Biol. Chem. 278, 37139-37145[Abstract/Free Full Text]
  30. Levin, D. E. (2005) Microbiol. Mol. Biol. Rev. 69, 262-291[Abstract/Free Full Text]
  31. Martin, H., Rodriguez-Pachon, J. M., Ruiz, C., Nombela, C., and Molina, M. (2000) J. Biol. Chem. 275, 1511-1519[Abstract/Free Full Text]
  32. Torres, J., Di Como, C. J., Herrero, E., and Angeles de la Torre-Ruiz, M. (2002) J. Biol. Chem. 277, 43495-43502[Abstract/Free Full Text]
  33. Lum, P. Y., Armour, C. D., Stepaniants, S. B., Cavet, G., Wolf, M. K., Butler, J. S., Hinshaw, J. C., Garnier, P., Prestwich, G. D., Leonardson, A., Garrett-Engele, P., Rush, C. M., Bard, M., Schimmack, G., Phillips, J. W., Roberts, C. J., and Shoemaker, D. D. (2004) Cell 116, 121-137[CrossRef][Medline] [Order article via Infotrieve]
  34. Sherman, F. (1991) Methods Enzymol. 194, 3-21[CrossRef][Medline] [Order article via Infotrieve]
  35. Gietz, R. D., and Woods, R. A. (2002) Methods Enzymol. 350, 87-96[CrossRef][Medline] [Order article via Infotrieve]
  36. Longtine, M. S., McKenzie III, A., Demarini, D. J., Shah, N. G., Wach, A., Brachat, A., Philippsen, P., and Pringle, J. R. (1998) Yeast 14, 953-961[CrossRef][Medline] [Order article via Infotrieve]
  37. Sikorski, R. S., and Boeke, J. D. (1991) Methods Enzymol. 194, 302-318[Medline] [Order article via Infotrieve]
  38. Collins, C. A., and Guthrie, C. (1999) Genes Dev. 13, 1970-1982[Abstract/Free Full Text]
  39. Powers, T., and Walter, P. (1999) Mol. Biol. Cell 10, 987-1000[Abstract/Free Full Text]
  40. Chen, J. C., and Powers, T. (2006) Curr. Genet. 49, 1-13[Medline] [Order article via Infotrieve]
  41. Cardenas, M. E., Cutler, N. S., Lorenz, M. C., Di Como, C. J., and Heitman, J. (1999) Genes Dev. 13, 3271-3279[Abstract/Free Full Text]
  42. Komeili, A., Wedaman, K. P., O'Shea, E. O., and Powers, T. (2000) J. Cell Biol. 151, 863-878[Abstract/Free Full Text]
  43. Shamji, A. F., Kuruvilla, F. G., and Schreiber, S. L. (2000) Curr. Biol. 10, 1574-1581[CrossRef][Medline] [Order article via Infotrieve]
  44. Zhang, Y., Billington, C. J., Jr., Pan, D., and Neufeld, T. P. (2006) Genetics 172, 355-362[Abstract/Free Full Text]
  45. Alaimo, P. J., Knight, Z. A., and Shokat, K. M. (2005) Bioorg. Med. Chem. 13, 2825-2836[CrossRef][Medline] [Order article via Infotrieve]
  46. deHart, A. K. A., Schnell, J. D., Allen, D. A., Tsai, J.-Y., and Hicke, L. (2003) Mol. Biol. Cell 14, 4676-4684[Abstract/Free Full Text]
  47. Ho, H. L., Shiau, Y. S., and Chen, M. Y. (2005) Curr. Genet. 47, 273-288[CrossRef][Medline] [Order article via Infotrieve]
  48. Saiardi, A., Resnick, A. C., Snowman, A. M., Wendland, B., and Snyder, S. H. (2005) Proc. Natl. Acad. Sci. U. S. A. 102, 1911-1914