Suppressor Analysis of Fimbrin (Sac6p) Overexpression in Yeast

Corresponding author: Alison E. M. Adams, Department of Molecular and Cellular Biology, Life Sciences South, University of Arizona, Tucson, AZ 85721., adamsa{at}u.arizona.edu (E-mail)

Communicating editor: F. WINSTON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Yeast fimbrin (Sac6p) is an actin filament-bundling protein that is lethal when overexpressed. To identify the basis for this lethality, we sought mutations that can suppress it. A total of 1326 suppressor mutations were isolated and analyzed. As the vast majority of mutations were expected to simply decrease the expression of Sac6p to tolerable levels, a rapid screen was devised to eliminate these mutations. A total of 1324 mutations were found to suppress by reducing levels of Sac6p in the cell. The remaining 2 mutations were both found to be in the actin gene and to make the novel changes G48V (act1-20) and K50E (act1-21). These mutations suppress the defect in cytoskeletal organization and cell morphology seen in ACT1 cells that overexpress SAC6. These findings indicate that the lethal phenotype caused by Sac6p overexpression is mediated through interaction with actin. Moreover, the altered residues lie in the region of actin previously implicated in the binding of Sac6p, and they result in a reduced affinity of actin for Sac6p. These results indicate that the two mutations most likely suppress by reducing the affinity of actin for Sac6p in vivo. This study suggests it should be possible to use this type of suppressor analysis to identify other pairs of physically interacting proteins and suggests that it may be possible to identify sites where such proteins interact with each other.

OVEREXPRESSION of actin and many actin-associated proteins is lethal to the cell, suggesting that the correct stoichiometry of cytoskeletal components is critical. For example, in yeast, actin (encoded by ACT1), Abp1p, and fimbrin (encoded by SAC6) are lethal under at least some conditions when overexpressed (DRUBIN et al. 1988 ; MAGDOLEN et al. 1993 ; this study). However, little is known about why overexpression of these genes is lethal.

In this study, we set out to address the reason for the lethality when the actin filament-bundling protein Sac6p is overexpressed. To better understand why SAC6 overexpression is lethal and to identify proteins that may interact with, or compensate for, an excess of Sac6p, a genetic screen was devised to isolate genes that, when mutated, can suppress the growth defect caused by SAC6 overexpression. Overexpression of SAC6 from the GAL1 promoter inhibits cell growth, so suppressors could be isolated on galactose plates that selected for the pGAL1-SAC6 plasmid. However, as there were numerous ways in which the levels of Sac6p in the cell could be reduced to tolerable levels, it was essential to have a rapid way of eliminating such mutations from the analysis. We therefore devised a screen that simply involved plate tests. Using this screen, we were able to rapidly eliminate all but 2 of 1326 suppressors isolated. The remaining two mutations were both found to be in the ACT1 gene, indicating that the lethality resulting from overexpression of Sac6p occurs via the specific interaction of this protein with actin. These mutations make novel changes in actin (G48V and K50E) and, thus, further increase the collection of mutant actin alleles that are available for analysis of actin cytoskeleletal function in yeast. In addition, these two mutations change residues in a region of actin previously implicated in the binding of Sac6p, suggesting that they suppress the overexpression lethality by reducing the affinity of actin for Sac6p. Biochemical analysis of the interaction between the mutant actins and Sac6p confirms that both act1-20 and act1-21 result in a decreased affinity of actin for Sac6p, providing a molecular explanation for our in vivo findings. These studies suggest that it may be possible to map unknown sites of interactions of proteins through analysis of mutations that suppress lethality (or some other phenotype) caused by overexpression of an interacting protein.

These studies are important in showing clearly that overexpression phenotypes, which are often considered suspect (largely because of the possibility that they result from nonspecific interactions) can arise from a highly specific and relevant interaction. Moreover, they demonstrate that such phenotypes, which are often viewed as a poor starting point for genetic analysis of interacting proteins, can lead to the identification of normal binding partners through suppressor analysis.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Yeast strains, plasmids, and media:
The yeast strains used in this study are listed in Table 1. Plasmids used in this study are listed in Table 2. Media for yeast growth and sporulation and the methods for mating, sporulation, and tetrad dissection are described in SHERMAN et al. 1974 . Rich medium containing glucose is referred to as YEPD; rich medium containing galactose is referred to as YEP-Gal. Galactose plates contained 2% galactose. Osmotic sensitivity was determined by growth of cells on rich medium (YEPD or YEP-Gal) containing 900 mM NaCl. Growth on plates was determined by placing cell cultures onto plates using a 32-prong inoculator.

Construction of plasmids containing SAC6 under the control of the GAL1 promoter:
A pGAL1-SAC6 overexpression plasmid pAAB157 was constructed in two steps. The first step placed a portion of the 5' end of the SAC6 gene behind a GAL1 promoter. The vector pAAB121, a centromere-containing plasmid carrying the yeast URA3 gene and GAL1-10 promoter (Table 2), was prepared by cutting with BamHI (two sites in the polylinker, adjacent to the GAL1 promoter), filling the ends, and then cutting with XbaI (in the polylinker, distal to the GAL1 promoter). A fragment containing a blunt-ended EcoRI site, followed by the 5' end of the SAC6 gene (from 25 bp 5' of the start codon) to the XbaI site within the coding sequence, was obtained from a cDNA construct described previously (ADAMS et al. 1991 ). This fragment was ligated into the blunt-ended BamHI and XbaI sites, respectively, of pAAB121 (above). The blunt-ended BamHI sites of pAAB121 and the EcoRI site of the fragment were destroyed in the cloning process. The second step involved fusing the remaining 3' portion of SAC6 from the XbaI site, resulting in the generation of plasmid pAAB157 containing the entire SAC6 coding region. To this end, the pAAB121-derived construct described above was cut with SphI (distal to the XbaI site in the polylinker), the ends were made blunt, and the linearized plasmid was cut with XbaI. The 3' end of SAC6 (from a wild-type genomic clone described in ADAMS et al. 1991 ), from the XbaI site in the coding sequence to a blunt-ended BamHI site (engineered, overlapping with the NcoI site 1245 bp 3' of the SAC6 stop codon), was inserted into the XbaI and blunt-ended SphI sites of the plasmid described above to generate pAAB157 (the BamHI site was regenerated while the SphI site was destroyed in the process). The cloning junctions at the 5' end of the gene and the XbaI site were sequenced. The first part of the construct was derived from a cDNA clone, and, thus does not contain the SAC6 intron (ADAMS et al. 1991 ).

The entire pGAL1-SAC6 fragment from pAAB157, flanked by EcoRI (5' of the GAL1-10 promoter) and the regenerated BamHI site described above, was inserted into the EcoRI-BamHI sites of pAAB107 (LEU2 CEN6) and pAAB149 (LYS2 CEN) to create pAAB328 and pAAB329, respectively.

Isolation of suppressors:
Spontaneous suppressors of the growth inhibition caused by SAC6 overexpression were isolated as follows. A SAC6 strain of mating-type a or (IGY182 or IGY181, respectively) carrying a SAC6 overexpression plasmid pAAB329 (pGAL1-SAC6 LYS2 CEN) and reporter plasmid pAAB416 (pGAL1-LacZ URA3 CEN) was plated onto synthetic-glucose or synthetic-raffinose media lacking uracil and lysine (to select for maintenance of both plasmids) at ~100–200 cells per plate and grown for 4–10 days. Unless otherwise indicated, cells were grown at 30°. Individual colonies were resuspended in sterile water and spread onto synthetic-galactose plates lacking uracil and lysine, such that one colony from the glucose or raffinose plate was put on one entire galactose plate. Cells were grown on galactose plates for 3–7 days. Isolates were streaked for single colonies on synthetic-galactose medium lacking uracil and lysine. To ensure the isolation of independent suppressors, only one colony from each galactose plate was usually selected; in those cases where more than one colony per plate was chosen, only one isolate from any one plate was subsequently retained unless it was clear that additional isolates were genotypically or phenotypically distinct. A total of 1326 colonies from 1173 plates were analyzed. For the isolation of most suppressors, 60-mm plates were used instead of the more standard 100-mm plates because they required much less media and still usually yielded at least one colony per plate.

Genetic analysis of suppressors:
Suppressors containing plasmid pAAB329 [pGAL1-SAC6 LYS2] were analyzed to determine whether the mutations were plasmid linked or genomic and dominant or recessive. As we expected the majority of mutants to be trivial (see RESULTS), a rapid approach was used to eliminate them (see Figure 1). Thus, revertants were crossed to a wild-type strain of opposite mating type (AAY1947 or IGY184) that contained plasmid pAAB328 [pGAL1-SAC6 LEU2], and diploids were selected on synthetic-glucose plates lacking lysine and leucine. The resulting strains, containing two pGAL1-SAC6 plasmids, were tested for growth on plates containing galactose and lacking leucine or lysine. As described in RESULTS, for most revertants, it was possible to determine whether the suppressor mutation was plasmid linked or genomic, and recessive or dominant, by simply scoring growth on these plates. However, this analysis was critically dependent on the ability of the various suppressor strains to mate. Some 51 of the 1326 suppressors were unable to mate on plates; of these, 6 could be mated in liquid cultures, and the remaining 45 were analyzed by transformation rather than by mating. In these cases, the cells were transformed with fresh [pGAL1-SAC6 LEU2] plasmid and tested for growth on synthetic galactose plates lacking leucine. Growth indicated that suppression was not caused by a plasmid-linked mutation, and these isolates were analyzed further.

View larger version (37K): In this window In a new window Download PPT slide   Figure 1. Identification of suppressors of the lethality resulting from Sac6p overexpression. Each straight line inside the cells represents a haploid genome, and each circle represents a particular plasmid whose selectable marker is shown. For simplicity, the pGAL1-lacZ plasmid has been omitted from each cell. X represents either a genomic or plasmid-linked mutation. Suppressor strains carrying a plasmid containing pGAL1-SAC6 and LYS2 are selected on galactose-containing plates lacking lysine. These strains are crossed to wild-type cells carrying a LEU2 pGAL1-SAC6 plasmid. Diploids are selected on glucose-containing plates lacking lysine and leucine, and are then tested for growth on galactose-containing plates lacking lysine or leucine. Some 126 recessive genomic suppressor mutations result in no growth on either plate; ~1200 recessive, plasmid-linked suppressor mutations give growth on plates lacking lysine and no growth on plates lacking leucine.

Sequence analysis of the act1-20 and act1-21 mutations:
As described in RESULTS, preliminary evidence indicated the two suppressor mutations were in ACT1. The presence of an act1 mutation in each case was confirmed by isolating genomic DNA from each mutant, and by amplifying and cycle sequencing their ACT1 genes, as described previously (HONTS et al. 1994 ).

Generation of strains containing act1-20::HIS3 or act1-21::HIS3:
To confirm that the two act1 mutations identified were sufficient for suppression, strains carrying mutant actin genes with just these mutations were generated by a method similar to that described by WERTMAN et al. 1992 . Thus, genomic DNA was isolated from strains IGY186 (act1-20) and IGY173 (act1-21). The mutant actin genes were isolated via PCR with exo⁺ pfu DNA polymerase, using the primer set AAO119 (5' ATTTTTCACGCTTACTGC 3') and AAO120 (5' ACATAAACATACGCGCAC 3'). In each case, this generated a 1522-bp fragment, from 56 bp 5' of the coding sequence to 1466 bp within the coding sequence of actin. The fragments were cut with XhoI and BsmI (which cut within the coding region at nucleotides 128 and 1300, respectively) and were cloned into pAAB508 (Table 2) cut with XhoI and BsmI in the coding region of ACT1. This manipulation simply replaced the wild-type actin sequence, from XhoI to BsmI, with the same stretch from the mutants. The presence of each mutation was verified through complete DNA sequencing of one strand. Plasmids were then transformed into an adenine methylase-deficient bacterial strain, GM48. Plasmids isolated from this strain were cut with BclI (just 3' of the mutant actin genes), and in each case, a 1.8-kb BamHI fragment containing HIS3 from pAAB4 (Table 2) was inserted. These plasmids were cut with EcoRI (flanking the act1::HIS3 fragment) and transformed into the ACT1⁺/ACT1⁺ diploid strain IGY123. His⁺ transformants were selected and sporulated, and tetrads were dissected. His⁺ segregants were isolated, and the presence of just the act1-20 or act1-21 mutation in a haploid segregant from each diploid was verified by isolating genomic DNA, amplifying the coding region of the mutant actin genes by PCR (using oligonucleotides AAO119 and AAO120), and then sequencing from the XhoI to BsmI site.

X-Gal overlays:
X-Gal overlays were conducted by the method used in Roy Parker's lab. Briefly, a 1% solution of agarose/H₂O was prepared and mixed 1:1 with 1 M NaPO₄, pH 7.0. One milliliter of 10% SDS was added per 100-ml final volume. Two-hundred microliters of 2% X-Gal in DMF was added per 10 ml. Strains containing the pGAL1-LacZ URA3 and pGAL1-SAC6 plasmids were streaked onto galactose plates lacking uracil and lysine, selecting for maintenance of both plasmids, and grown for 3–4 days at 30°. Molten X-Gal mix was poured directly on top of the patches of cells on the plate. Color development was assayed after incubation for 1–2 hr at 37°.

Purification of actin and Sac6p, and actin filament-binding and cross-linking assays:
Actin was purified from wild-type and mutant cells, as described previously (HONTS et al. 1994 ), using strains IGY181 (ACT1), IGY185 (act1-20), and IGY173 (act1-21); see Table 1. Sac6p was purified from a strain overexpressing Sac6p, as described previously (SANDROCK et al. 1997 ). Actin filament-binding and cross-linking assays were performed as described previously (HONTS et al. 1994 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Genetic screen to identify suppressors of the growth defect caused by overexpression of Sac6p:
To identify suppressors of the lethality caused by Sac6p overexpression, we devised a screen that would allow us to rapidly identify mutants of interest. Cells of the SAC6 strain IGY181 (mating type ) or IGY182 (mating-type a) carrying a plasmid containing SAC6 under the control of the GAL1 promoter were plated on galactose (conditions that lead to overexpression of SAC6), and spontaneous suppressors that could grow under these conditions were isolated. A total of 1326 suppressor strains (most of which were independent isolates, as described in MATERIALS AND METHODS) were obtained and characterized further.

Identification of recessive genomic suppressor mutations:
We expected to obtain several classes of suppressors, including (i) plasmid-linked mutations that no longer result in overproduction of wild-type Sac6p, through mutation of either the GAL1 promoter or the SAC6 gene; (ii) lys2::HIS3 gene convertants, in which the lys2::HIS3 genomic locus is replaced by LYS2 (thus enabling cells to grow without lysine in the absence of the pGAL1-SAC6 LYS2 plasmid); (iii) trans-acting genomic mutations that decrease levels of expression from the GAL1 promoter; and (iv) genomic suppressor mutations of interest. The first three classes of suppressors were not of interest, but were expected to be the most common. It was, therefore, necessary to eliminate them rapidly. To this end, the following approach was taken (see Figure 1). Suppressor strains carrying a [pGAL1-SAC6 LYS2] plasmid were crossed to a SAC6 strain of opposite mating type containing a fresh [pGAL1-SAC6-LEU2] plasmid. Diploids were selected on synthetic glucose plates lacking lysine and leucine. The resulting diploids, carrying two different SAC6 overexpression plasmids, were purified and tested for their ability to grow on galactose plates lacking lysine or leucine (see Figure 1 and Table 3). As outlined below, from these tests, we were able to distinguish between genomic and plasmid-linked mutations and between dominant and recessive mutations:

Plasmid-linked mutations: If the suppressor was caused by a recessive mutation in the [pGAL1-SAC6 LYS2] plasmid, then the resulting diploid should be able to grow on galactose plates lacking lysine, but not lacking leucine (Figure 1). As shown in Table 3, the vast majority of mutants were of this type. If the suppressor was caused by a dominant mutation in the [pGAL1-SAC6 LYS2] plasmid, i.e., a mutation that can suppress even in the presence of a fresh plasmid, the resultant diploid should be able to grow on galactose plates lacking either lysine or leucine. However, growth on the latter would be dependent on the presence of the [pGAL1-SAC6 LYS2] plasmid. None of the revertants fell into this class (Table 3).

lys2::HIS3 gene convertants: If the suppressor was caused by replacement of the lys2::HIS3 locus by the LYS2 gene from the plasmid, then diploids resulting from a cross to wild-type carrying the [pGAL1-SAC6 LEU2] plasmid would be able to grow on synthetic galactose lacking lysine but not lacking leucine and, therefore, would appear similar to strains carrying recessive, plasmid-linked mutations (see above and Table 3). As neither category of mutation was of interest to us, the distinction was unimportant, but could be made by analysis of such strains on synthetic galactose plates lacking histidine (see Table 3).

Genomic suppressor mutations: The class of suppressor mutations that we wished to isolate in this screen were those that were genomic. Such mutations could be identified in the crosses described above. Thus, if a suppressor mutation were genomic and recessive, it would be masked by the wild-type copy in the resulting diploid strain. The diploid would then be unable to grow on synthetic galactose plates lacking either lysine or leucine (Figure 1 and Table 3). Some 126 suppressor mutations were of this class (Table 3). If the suppressor mutation were genomic and dominant, however, the diploid should be able to suppress the lethality caused by either the [pGAL1-SAC6 LYS2] or the [pGAL1-SAC6 LEU2] plasmid and, thus, should grow on synthetic galactose plates lacking either lysine or leucine. In contrast to the dominant plasmid mutations (above), however, such suppressors would not be dependent on the presence of the LYS2-containing plasmid. No suppressors of this type were obtained.

Identification of genomic suppressor mutations that act by reducing the levels of expression from the GAL1 promoter:
Among the genomic suppressor mutations, we expected to obtain mutations in the galactose regulatory genes which could result in decreased expression of SAC6 to tolerable levels. To rapidly identify such mutations, a reporter plasmid, pAAB416, containing pGAL1-LacZ URA3, was included in the initial strain (IGY181 or IGY182). This plasmid allowed assessment of the relative levels of induction from the GAL1 promoter. Expression was measured using an X-Gal overlay assay, as described in MATERIALS AND METHODS. Suppressors were evaluated on the basis of the relative level of blue color development (resulting from levels of ß-galactosidase activity). Only two mutants developed blue color as dark as the controls. The remaining mutants presumably grow in the presence of galactose because of reduced levels of expression of SAC6 from the GAL1 promoter. Indeed, consistent with this expectation, of 23 mutants that had reduced levels of expression and that were analyzed, 20 carried mutations in either SPT5 or GAL11, two genes known to be involved in transcriptional regulation (SUZUKI et al. 1988 ; SWANSON et al. 1991 ). The genes defective in the remaining three mutants of this group were not identified and were not studied further.

Identification of act1 mutant alleles:
The two mutants that were dark blue on X-Gal indicator plates were analyzed further. Preliminary phenotypic analysis suggested these mutations might be in the ACT1 gene. Thus, in the presence of the pGAL1-SAC6 plasmid, both mutants were found to have temperature-sensitive and osmotic-sensitive phenotypes reminiscent of many act1 mutant strains (BOTSTEIN et al. 1997 ): the cells grew more poorly at 37° than at 23° on either YEPD or YEP-Gal plates, and they failed to grow on media containing 900 mM salt. Therefore, we tested whether the two suppressors isolated in this study were in the ACT1 gene. In each case, we obtained preliminary evidence from complementation and/or linkage analyses that the mutations were in ACT1 (not shown). Therefore, we isolated genomic DNA from each mutant (strains IGY186 and IGY173) and amplified and cycle sequenced their ACT1 genes as described previously (HONTS et al. 1994 ). In each case, the entire coding region was sequenced on one strand, and the segment containing the mutation was sequenced on both strands. For one (hereafter referred to as act1-20), a single nucleotide change in the ACT1 coding region was found which resulted in a Gly-to-Val change at amino acid 48. For the other (hereafter referred to as act1-21), a single change was found which resulted in a Lys-to-Glu change at residue 50. Interestingly, an actin mutation was isolated in a screen for endocytosis mutants (MUNN et al. 1995 ). This allele, designated end7-1, alters the same residue as act1-20, except end7-1 results in a glycine to aspartic acid substitution (MUNN et al. 1995 ).

Demonstration that act1-20 or act1-21 is sufficient for suppression of the lethality caused by overexpression of SAC6:
We confirmed that these mutations were sufficient to cause the suppression phenotype by replacing the wild-type ACT1 gene with mutant actin genes (linked to HIS3) harboring either act1-20 or act1-21 and showing that, in each case, the resulting cells were viable when SAC6 was overexpressed (Figure 2). To this end, ACT1/ACT1 diploid strains were first transformed with act1::HIS3 (see MATERIALS AND METHODS). Heterozygous act1-20::HIS3/ACT1 or act1-21::HIS3/ACT1 diploids were then transformed with a URA3-containing plasmid carrying pGAL1-SAC6 (pAAB364). Transformants were selected on synthetic dextrose plates lacking uracil, cells were sporulated, and tetrads dissected. In the case of act1-20, 14 out of 18 tetrads dissected on YEPD yielded 4 viable spores, and His⁺:His^- segregated 2:2. Some 24 of 56 segregants were Ura⁺ and, therefore, carried the pGAL1-SAC6 plasmid. Of these, 12 were viable on galactose and were His⁺ (act1-20), whereas 12 were inviable on galactose and were His^- (ACT1). These results indicate that the act1-20 mutation is necessary and sufficient to suppress the lethality caused by SAC6 overexpression.

View larger version (86K): In this window In a new window Download PPT slide   Figure 2. Suppression of Sac6p overexpression lethality by act1-20 or act1-21 mutant alleles. Growth of ACT1 (A and B), act1-20 (C and D), or act1-21 (E and F) cells containing a pGAL1-SAC6 plasmid (pAAB364; A, C, and E) or control plasmid (pAAB121; B, D, and F) on -Ura dextrose or -Ura galactose after 3 days at 30°. Strains shown are (A) IGY172 (ACT1 strain containing pAAB364), (B) IGY169 (ACT1 strain containing pAAB121), (C) IGY171 (act1-20 strain containing pAAB364), (D) IGY170 (act1-20 strain containing pAAB121), and (E) IGY105 (act1-21 strain containing pAAB364). act1-21 cells containing the vector pAAB121 and, therefore, expressing normal levels of Sac6p are inviable (F). act1-21 cells containing the pGAL1-SAC6 plasmid are viable even on glucose, most likely because of low (but nevertheless elevated) levels of Sac6p expression from the GAL1 promoter on glucose.

In the case of act1-21, genetic analysis revealed that act1-21 not only suppresses the lethality caused by overexpression of SAC6, but that it also makes the cells dependent on overexpression of SAC6 for viability (see below and Figure 2). Heterozygous act1-21::HIS3/ACT1 diploids transformed with the URA3-containing plasmid carrying SAC6 (pAAB364) were selected on synthetic dextrose media lacking uracil, cells were sporulated, and tetrads were dissected on galactose. (Tetrads were dissected on galactose-containing plates because act1-21 cells grew best when they carried the pGAL1-SAC6 plasmid and were grown on galactose; see below.) Two or fewer viable spores were obtained per tetrad. A total of 32 viable segregants were obtained. Of these, 12 were Ura⁺ and, therefore, carried the pGAL1-SAC6 plasmid (the Ura^- segregants are discussed below). All 12 of these Ura⁺ segregants were His⁺ (act1-21), confirming that overexpression of SAC6 is lethal in ACT1 cells, and showing that the act1-21 mutation is necessary and sufficient for suppression of the lethality caused by overexpression of SAC6.

Overexpression of SAC6 suppresses the lethality caused by the act1-21 mutation:
Analysis of the Ura^- segregants from the act1-21::HIS3/ACT1 diploid above revealed that 20 out of 20 were His^- (ACT1). As none were His⁺ (act1-21), act1-21 cells are viable only in the presence of the pGAL1-SAC6 plasmid. act1-21 cells carrying the pGAL1-SAC6 plasmid are viable on either glucose or galactose, indicating that there must be enough expression of SAC6 from the GAL1 promoter, even on glucose, to suppress the lethality caused by act1-21. Indeed, the act1-21 mutant originally had to go through a period of growth under noninducible conditions before its identification on galactose (see MATERIALS AND METHODS); any mutants that were dependent on higher levels of expression of SAC6 would not have been isolated in this scheme. Growth on glucose, however, is much weaker than on galactose, suggesting that higher levels of SAC6 are optimal (Figure 2).

Additional evidence confirmed that overexpression of SAC6 suppresses the lethality caused by the act1-21 mutation. Thus, IGY174 (act1-21) mutant cells containing the [LYS2 pGAL1-SAC6] plasmid are unable to lose the plasmid unless they have first been transformed with (i) a [LEU2 pGAL1-SAC6] plasmid (pAAB328), or (ii) a high-copy-number (2µ) plasmid carrying URA3 SAC6 (pAAB144), which allows the cells to grow on medium containing glucose. (In the presence of the LEU2 or URA3 control plasmid [pAAB107 or pAAB132, respectively], the cells are still not able to lose the [LYS2 pGAL1-SAC6] plasmid.)

act1-20 and act1-21 suppress the defects in actin organization seen in cells overexpressing SAC6:
To gain insight into the underlying basis of the suppression by the mutant actin alleles, cells were stained with rhodamine-labeled phalloidin, and the intracellular distribution of actin in wild-type or mutant cells overexpressing SAC6 was examined by fluorescence microscopy (Figure 3). As reported previously (ADAMS and PRINGLE 1984 ; KILMARTIN and ADAMS 1984 ), wild-type cells expressing normal levels of Sac6p have a polarized distribution of actin, with patches concentrated in regions of growth and cables extending through the cytoplasm in a direction roughly parallel to the axis of polarized growth (Figure 3A). Wild-type cells overexpressing SAC6, however, have an abnormal distribution of actin. In particular, cells show delocalized actin patches and a loss of actin cables (Figure 3B). [If ACT1 cells overexpressing SAC6 are stained with anti-actin antibodies instead of phalloidin, ~27% of the cells are seen to have densely stained actin cables or bars. Similar bars have been seen previously in cells stained with antiactin antibodies, rather than phalloidin, but the nature of these bars is unknown (HAARER et al. 1990 ). In the present case, in which SAC6 is overexpressed, these bars likely reflect an increase in actin filament cross-linking activity.] In addition, wild-type cells overexpressing SAC6 are often larger and rounder than normal. act1-20 or act1-21 mutant cells overexpressing SAC6, however, show a distribution of actin and cell size and morphology very similar to wild-type cells expressing normal levels of Sac6p (Figure 3D and Figure F). These results indicate that the act1 suppressor mutations restore function to cells overexpressing SAC6 by directly affecting the way in which Sac6p organizes actin.

View larger version (108K): In this window In a new window Download PPT slide   Figure 3. Actin organization in ACT1 wild-type (A and B), act1-20 mutant (C and D), or act1-21 mutant (F) cells expressing normal levels of SAC6 (A and C) or overexpressing SAC6 under the control of the inducible GAL1 promoter (B, D, and F). Cells carrying the SAC6-containing plasmid (pAAB364, B, D, and F) or vector control (pAAB121, A and C) were grown on minimal medium lacking uracil (selective conditions) and containing 2% glucose (repressing conditions) to midlog, and were then shifted to galactose (to induce expression from the GAL1 promoter) for ~16 hr. Cells were then fixed and stained with rhodamine-phalloidin, as described previously (CHOWDHURY et al. 1992 ), and examined under a Zeiss Axioskop using a 100x objective. All prints are at the same magnification. Strains are as described in Figure 2. act1-21 cells containing the vector pAAB121 (i.e., expressing normal levels of Sac6p) are inviable, so E does not contain any cells.

Actin mutations that suppress the overexpression of Sac6p map to a defined region of the actin monomer and reduce the affinity of actin for Sac6p in vitro:
The two new act1 alleles lie in the region of actin that was previously implicated in the binding of Sac6p (Figure 4). Indeed, act1-21 changes one of the residues altered in act1-125 (K50A, D51A, WERTMAN et al. 1992 ), which has been shown to encode actin with a lower affinity for Sac6p than wild-type actin (HONTS et al. 1994 ). This finding suggests that the mechanism of suppression of the lethality caused by overexpression of Sac6p may be simply by reducing the affinity of Sac6p for actin. To test this idea, we purified actin from ACT1 wild-type and act1-20 and act1-21 mutant cells, and we asked whether the mutant actins are defective in (i) binding to wild-type Sac6p and (ii) forming bundles of actin filaments in the presence of Sac6p. Wild-type or mutant actins were mixed with Sac6p under polymerization conditions, and filaments of actin and bound Sac6p were sedimented by high-speed centrifugation. Under these conditions, nonpolymerized actin and unbound Sac6p remain in the supernatant. As reported previously and shown in Figure 5A, most wild-type actin is found in the high-speed pellet with Sac6p, indicating that most actin forms filaments to which Sac6p is bound. In the case of Act1-20p and Act1-21p, most mutant actin is similarly found in the high-speed pellet, but in each case (and most dramatically in the case of Act1-21p), a significant fraction of Sac6p remains in the supernatant (Figure 5B and Figure C). These findings indicate that both Act1-20p and Act1-21p are able to form filaments, but the mutant filaments are defective in binding to Sac6p.

View larger version (97K): In this window In a new window Download PPT slide   Figure 4. Location in a model of rabbit muscle G-actin of residues altered by two novel actin alleles. A model of the rabbit muscle actin molecule was displayed from the atomic coordinates of the rabbit muscle actin, as determined by KABSCH et al. 1990 . Location of residues previously identified as important for Sac6p interaction with actin are indicated by gray shading (HONTS et al. 1994 ). Residues altered by the suppressor mutations are indicated by black shading.

View larger version (23K): In this window In a new window Download PPT slide   Figure 5. Actin-binding and cross-linking assays for wild-type and mutant actins with wild-type Sac6p. SDS-PAGE of supernatants (S) or pellets (P) obtained by high-speed (HS) or low-speed (LS) centrifugation of wild-type or mutant actins polymerized in the presence (+) or absence (-) of Sac6p, as described in MATERIALS AND METHODS. The positions of yeast actin (A) and Sac6p (S) bands in the gels are indicated. The proteins were resolved in a 10% SDS-polyacrylamide gel and stained with Coomassie blue R-250. The data are representative of multiple experiments.

As Sac6p is an actin filament-bundling protein, we tested whether defects in binding of Sac6p to mutant actin result in defects in bundle formation. To this end, mixtures of actin and Sac6p were subjected to low-speed centrifugation—conditions under which cross-linked actin filaments sediment but individual filaments do not. As shown in Figure 5A, in the case of wild-type, most actin and bound Sac6p are found in the pellet, indicating the formation of bundles of actin filaments. In the case of Act1-20 and Act1-21 mutant actins, however, a significant fraction of actin and Sac6p is found in the supernatant, indicating a defect in actin filament bundle formation. As expected, (i) in the absence of actin, most Sac6p is found in the supernatant after either high- or low-speed centrifugation (Figure 5D), and (ii) in the absence of Sac6p (Figure 5, A–C), most wild-type or mutant actin is found in the high-speed pellet (indicating the presence of F-actin) and in the low-speed supernatant (indicating the absence of bundles).

Together, these results show that Act1-20 and Act1-21 mutant actins are defective in binding to Sac6p, and that they have a defect in Sac6p-induced actin filament bundle formation. Therefore, we conclude that these mutations suppress the Sac6p overexpression lethality by reducing the affinity of actin for Sac6p. Moreover, as the lethal phenotype of Sac6p is mediated through an interaction with actin, it is highly specific.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In yeast, numerous proteins are lethal when overexpressed, and the overexpression phenotype can frequently provide insight into the functions of the overexpressed protein (PHIZICKY and FIELDS 1995 ). The use of the overexpression phenotype as a starting point for genetic analysis is, therefore, likely to lead to the identification of proteins that interact with the overexpressed protein; indeed, there are several cases where this has proven to be so. For example, several studies with yeast have identified high-copy suppressors of the overexpression phenotype (e.g., MEEKS-WAGNER and HARTWELL 1986 ; WEINSTEIN and SOLOMON 1990 ; MAGDOLEN et al. 1993 ; ARCHER et al. 1995 ; PHIZICKY and FIELDS 1995 ; ALVAREZ et al. 1998 ; SMITH et al. 1998 ). Rarely, however, have genomic mutations that suppress the overexpression phenotype been sought [although there are examples, e.g., suppressors of the phenotype caused by overexpression of the GAL4-VP16 fusion protein (BERGER et al. 1992 ) and suppressors of the phenotype caused by overexpression of the Schizosaccharomyces pombe cdc7 protein (FANKHAUSER and SIMANIS 1994 )]. Indeed, in recent reviews of genetic methods used to study the yeast cytoskeleton (BOTSTEIN et al. 1997 ) or other aspects of biology (PHIZICKY and FIELDS 1995 ), this particular brand of suppressor analysis was not even mentioned. The reason for this may lie in the suspicion that overexpressed proteins might interfere with cellular functions through nonspecific mechanisms or through altered specificity and, therefore, may identify proteins that do not interact with the protein of interest under normal conditions. However, dosage suppressors that ameliorate a phenotype caused by mutation have been identified in numerous studies with yeast (e.g., FRANCISCO et al. 1994 ; KIRKPATRICK and SOLOMON 1994 ; VALLEN et al. 1994 ; MACHIN et al. 1995 ; PRENDERGAST et al. 1995 ; GLERUM et al. 1996 ; LAPINSKAS et al. 1996 ; SAPPERSTEIN et al. 1996 ; HERMANN et al. 1997 ; HOVLAND et al. 1997 ; HUANG and ELLEDGE 1997 ; KAGAMI et al. 1997 ; SCHMIDT et al. 1997 ; WEBB et al. 1997 ; DESRIVIERES et al. 1998 ; GRANDIN et al. 1998 ; ZHANG et al. 1998 ). As such suppressors often identify proteins of interest, it should be similarly possible to identify mutations that suppress an overexpression phenotype. In this study, we used suppressor analysis to examine the nature of the lethality caused by overexpression of the actin filament-bundling protein fimbrin, which is encoded by the SAC6 gene (ADAMS et al. 1991 ). We found a very low frequency of mutations of interest, but we showed that those that were obtained were highly specific and were novel mutations in the actin gene.

Utility of the suppressor screen devised in this study to identify physically interacting proteins:
We have described a suppressor screen that should be useful in identifing suppressors of other genes that are lethal when overexpressed. The suppressor screen devised in this study is likely to be of general use in identifying interacting proteins, as (i) it led to the identification of a protein known to physically interact with Sac6p, and (ii) the screen was effective in identifying just mutations of interest (all the trivial mutations were rapidly eliminated early in the analysis). It is striking that in a screen of the entire genome, the only mutations identified as suppressors of the lethality caused by Sac6p overexpression were those that changed residues in a protein known to interact with Sac6p in vivo. However, it is probable that screens such as the one described here will not always identify just physically interacting proteins, as there are multiple ways in which the activity of proteins can potentially be downregulated.

Utility of the approach in mapping sites of interaction of proteins:
The observation that the actin mutations act1-20 and act1-21 suppress Sac6p overexpression is explained by the finding that the mutant actins have a decreased affinity for Sac6p. Both mutations are within the DNase I-binding loop in subdomain 2 of actin (see Figure 4), and they may effect local or more global changes. As they alter residues important for Sac6p binding, however, it is likely that these mutations further define the site of interaction of actin with Sac6p and fimbrins in general. This finding suggests that it may be possible to use this approach to map where, on a particular protein, an overexpressed protein binds. For example, if overexpression of a protein causes a lethal (or any other) phenotype, it may be possible to map where that protein binds to another protein through analysis of mutations that suppress the overexpression phenotype. Moreover, it may be possible to map sites of interaction of individual domains if overexpression of just that domain has a phenotype.

act1-20 and act1-21 are novel actin alleles:
Previous studies have identified a variety of mutant actin alleles (e.g., SHORTLE et al. 1984 ; ADAMS and BOTSTEIN 1989 ; DUNN and SHORTLE 1990 ; WERTMAN et al. 1992 ; MUNN et al. 1995 ). These alleles have provided a collection that has been invaluable in the analysis of both the interactions and functions of actin (BOTSTEIN et al. 1997 ). In this study, we describe two additional alleles that further increase this collection. One of these alleles, act1-20, changes the same residue altered by the end7-1 mutation. As end7-1 results in a defect in endocytosis (MUNN et al. 1995 ), and sac6 null mutant cells are defective in endocytosis (KUBLER and RIEZMAN 1993 ), it is likely that end7-1 is defective in endocytosis because of a defect in binding to Sac6p.

The other allele, act1-21, is particularly interesting, as it has the novel phenotype of being dependent on overexpression of an actin-binding protein for viability. The underlying biochemical basis for this dependence is unknown, but it will likely prove interesting and may shed light on cytoskeletal dynamics. Interestingly act1-125 (K50A, D51A; WERTMAN et al. 1992 ), which changes the same residue as act1-21 (K50E), is also suppressed by overexpression of Sac6p (HONTS et al. 1994 ), but in the case of act1-125, the cells are viable at 23° without extra Sac6p, and it is only the growth defect at 37° that is suppressed.

Overexpression of Sac6p leads to a highly specific phenotype that is more severe than that of the sac6 null mutation:
We previously found that sac6 null mutant cells are viable at room temperature but inviable at 37° (ADAMS et al. 1991 ). In contrast, we found that overexpression of Sac6p is lethal, even at 23°. This finding indicates that overexpression of Sac6p is more toxic to the cells than is a complete absence of this protein. There are several possible explanations for this observation, including that excess Sac6p (i) interferes nonspecifically with cellular function; (ii) causes excess bundling of actin, which is more deleterious to the cells than is a deficiency of bundling; (iii) titrates out some factor that is essential for cell growth; and (iv) competes with an essential actin-binding protein(s) for sites of interaction on actin. However, the finding that actin mutations suppress the overexpression lethality indicates that lethality is not simply through nonspecific interaction, but rather, through a specific association of Sac6p with actin. It is possible that excess Sac6p hyperstabilizes actin, making it less dynamic, and consistent with this idea is the observation that mutations of SAC6 (e.g., sac6-19) that result in an increased affinity of Sac6p for actin are also more deleterious to the cell than is the sac6 null mutation (SANDROCK et al. 1997 ). However, it is equally likely that either mutant Sac6-19 protein or excess Sac6p outcompetes some other actin-binding protein whose interaction with actin is essential for cell viability. Alternatively, Sac6p may be titrating out some factor other than actin, but if so, binding of Sac6p to actin must be required for that factor to bind to Sac6p.

	FOOTNOTES

¹ Present address: Ventana Genetics, Salt Lake City, UT 84108.
² Present address: Department of Microbiology and Molecular Genetics, University of Vermont, Burlington, VT 05405.

	ACKNOWLEDGMENTS

We thank Roy Parker, Ted Weinert, and members of our lab for helpful discussions during the course of this work; and Johanna O'Dell and Ted Weinert for comments on the manuscript. We are grateful to Skip Vaught and his staff in the DNA Sequencing Service at the University of Arizona for assistance with DNA sequencing and to Roy Parker for plasmids. This work was supported by a grant from the National Institutes of Health (GM45288), a predoctoral National Research Service Award (CA09213) to T.M.S., and an NSF Predoctoral Minority Fellowship to K.A.T.

Manuscript received August 18, 1998; Accepted for publication December 18, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ADAMS, A. E. M. and D. BOTSTEIN, 1989  Dominant suppressors of yeast actin mutations that are reciprocally suppressed. Genetics 121:675-683[Abstract/Free Full Text].

ADAMS, A. E. M. and J. R. PRINGLE, 1984  Relationship of actin and tubulin distribution to bud growth in wild-type and morphogenetic-mutant Saccharomyces cerevisiae.. J. Cell Biol. 98:934-945[Abstract/Free Full Text].

ADAMS, A. E. M., D. BOTSTEIN, and D. G. DRUBIN, 1991  Requirement of yeast fimbrin for actin organization and morphogenesis in vivo.. Nature 354:404-408[Medline].

ALVAREZ, P., A. SMITH, J. FLEMING, and F. SOLOMON, 1998  Modulation of tubulin polypeptide ratios by the yeast protein Pac10p. Genetics 149:857-864[Abstract/Free Full Text].

ARCHER, J. E., L. R. VEGA, and F. SOLOMON, 1995  Rbl2p, a yeast protein that binds to beta-tubulin and participates in microtubule function in vivo. Cell 82:425-434[Medline].

BERGER, S. L., B. PINA, N. SILVERMAN, G. A. MARCUS, and J. AGAPITE et al., 1992  Genetic isolation of ADA2: a potential transcriptional adaptor required for function of certain acidic activation domains. Cell 70:251-265[Medline].

BOTSTEIN, D., D. AMBERG, J. MULHOLLAND, T. HUFFAKER, A. ADAMS, et al., 1997 The yeast cytoskeleton, pp. 1–90 in The Molecular and Cellular Biology of the Yeast Saccharomyces: Cell Cycle and Cell Biology, edited by J. R. PRINGLE, J. R. BROACH and E. W. JONES. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

CHOWDHURY, S., K. W. SMITH, and M. C. GUSTIN, 1992  Osmotic stress and the yeast cytoskeleton: phenotype-specific suppression of an actin mutation. J. Cell Biol. 118:561-571[Abstract/Free Full Text].

DESRIVIERES, S., F. T. COOKE, P. J. PARKER, and M. N. HALL, 1998  MSS4, a phosphatidyl-4-phosphate 5-kinase required for organization of the actin cytoskeleton in Saccharomyces cerevisiae. J. Biol. Chem. 273:15787-15793[Abstract/Free Full Text].

DRUBIN, D. G., K. G. MILLER, and D. BOTSTEIN, 1988  Yeast actin-binding proteins: evidence for a role in morphogenesis. J. Cell Biol. 107:2551-2561[Abstract/Free Full Text].

DUNN, T. M. and D. SHORTLE, 1990  Null alleles of SAC7 suppress temperature-sensitive actin mutations in Saccharomyces cerevisiae.. Mol. Cell. Biol. 10:2308-2314[Abstract/Free Full Text].

FANKHAUSER, C. and V. SIMANIS, 1994  The cdc7 protein kinase is a dosage dependent regulator of septum formation in fission yeast. EMBO J. 13:3011-3019[Medline].

FRANCISCO, L., W. WANG, and C. S. CHAN, 1994  Type 1 protein phosphatase acts in opposition to Ipl1 protein kinase in regulating yeast chromosome segregation. Mol. Cell. Biol. 14:4731-4740[Abstract/Free Full Text].

GLERUM, D. M., A. SHTANKO, and A. TZAGOLOFF, 1996  SCO1 and SCO2 act as high copy suppressors of a mitochondrial copper recruitment defect in Saccharomyces cerevisiae.. J. Biol. Chem 271:20531-20535[Abstract/Free Full Text].

GRANDIN, N., A. DE ALMEIDA, and M. CHARBONNEAU, 1998  The Cdc14 phosphatase is functionally associated with the Dbf2 protein kinase in Saccharomyces cerevisiae.. Mol. Gen. Genet. 258:104-116[Medline].

HAARER, B. K., S. H. LILLIE, A. E. M. ADAMS, V. MAGDOLEN, and W. BANDLOW et al., 1990  Purification of profilin from Saccharomyces cerevisiae and analysis of profilin-deficient cells. J. Cell Biol. 110:105-114[Abstract/Free Full Text].

HERMANN, G. J., E. J. KING, and J. M. SHAW, 1997  The yeast gene, MDM20, is necessary for mitochondrial inheritance and organization of the actin cytoskeleton. J. Cell Biol. 137:141-153[Abstract/Free Full Text].

HONTS, J. E., T. S. SANDROCK, S. M. BROWER, J. L. O'DELL, and A. E. M. ADAMS, 1994  Actin mutations that show suppression with fimbrin mutations identify a likely fimbrin-binding site on actin. J. Cell Biol. 126:413-422[Abstract/Free Full Text].

HOVLAND, P. G., M. TECKLENBERG, and R. A. SCLAFANI, 1997  Overexpression of the protein kinase Pak1 suppresses yeast DNA polymerase mutants. Mol. Gen. Genet. 256:45-53[Medline].

HUANG, M. and S. J. ELLEDGE, 1997  Identification of RNR4, encoding a second essential small subunit of ribonucleotide reductase in Saccharomyces cerevisiae.. Mol. Cell. Biol. 17:6105-6113[Abstract].

KABSCH, W., H. G. MANNHERZ, D. SUCK, E. F. PAI, and K. C. HOLMES, 1990  Atomic structure of the actin:DNase I complex. Nature 347:37-44[Medline].

KAGAMI, M., A. TOH-E, and Y. MATSUI, 1997  SRO9, a multicopy suppressor of the bud growth defect in the Saccharomyces cerevisiae rho3-deficient cells, shows strong genetic interactions with tropomyosin genes, suggesting its role in organization of the actin cytoskeleton. Genetics 147:1003-1016[Abstract].

KILMARTIN, J. V. and A. E. M. ADAMS, 1984  Structural rearrangements of tubulin and actin during the cell cycle of the yeast Saccharomyces. J. Cell Biol. 98:922-933[Abstract/Free Full Text].

KIRKPATRICK, D. and F. SOLOMON, 1994  Overexpression of yeast homologs of the mammalian checkpoint gene RCC1 suppresses the class of alpha-tubulin mutations that arrest with excess microtubules. Genetics 137:381-392[Abstract].

KUBLER, E. and H. RIEZMAN, 1993  Actin and fimbrin are required for the internalization step of endocytosis in yeast. EMBO J. 12:2855-2862[Medline].

LAPINSKAS, P. J., S. J. LIN, and V. C. CULOTTA, 1996  The role of the Saccharomyces cerevisiae CCC1 gene in the homeostasis of manganese ions. Mol. Microbiol. 21:519-528[Medline].

MA, H., S. KUNES, P. J. SCHATZ, and D. BOTSTEIN, 1987  Plasmid construction by homologous recombination in yeast. Gene 58:201-216[Medline].

MACHIN, N. A., J. M. LEE, and G. BARNES, 1995  Microtubule stability in budding yeast: characterization and dosage suppression of a benomyl-dependent tubulin mutant. Mol. Biol. Cell 6:1241-1259[Abstract].

MAGDOLEN, V., D. G. DRUBIN, G. MAGES, and W. BANDLOW, 1993  High levels of profilin suppress the lethality caused by overproduction of actin in yeast cells. FEBS Lett. 316:41-47[Medline].

MEEKS-WAGNER, D. and L. H. HARTWELL, 1986  Normal stoichiometry of histone dimer sets is necessary for high fidelity of mitotic chromosome transmission. Cell 44:43-52[Medline].

MUNN, A. L., B. J. STEVENSON, M. I. GELI, and H. RIEZMAN, 1995  end5, end6, and end7: mutations that cause actin delocalization and block the internalization step of endocytosis in Saccharomyces cerevisiae.. Mol. Biol. Cell 6:1721-1742[Abstract].

PHIZICKY, E. M. and S. FIELDS, 1995  Protein-protein interactions: methods for detection and analysis. Microbiol. Rev. 59:94-123[Abstract/Free Full Text].

PRENDERGAST, J. A., C. PTAK, T. G. ARNOSON, and M. J. ELLISON, 1995  Increased ubiquitin expression suppresses the cell cycle defect associated with the yeast ubiquitin conjugating enzyme, CDC34 (UBC3). Evidence for a noncovalent interaction between CDC34 and ubiquitin. J. Biol. Chem. 270:9347-9352[Abstract/Free Full Text].

SANDROCK, T. M., J. L. O'DELL, and A. E. M. ADAMS, 1997  Allele-specific suppression by formation of new protein-protein interactions in yeast. Genetics 147:1635-1642[Abstract].

SAPPERSTEIN, S. K., V. V. LUPASHIN, H. D. SCHMITT, and M. G. WATERS, 1996  Assembly of the ER to Golgi SNARE complex requires Uso1p. J. Cell Biol. 132:755-767[Abstract/Free Full Text].

SHERMAN, F., G. FINK and C. LAWRENCE, 1974 Methods in Yeast Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SCHMIDT, A., M. BICKLE, T. BECK, and M. N. HALL, 1997  The yeast phosphatidylinositol kinase homolog TOR2 activates RHO1 and RHO2 via the exchange factor ROM2. Cell 88:531-542[Medline].

SHORTLE, D., P. NOVICK, and D. BOTSTEIN, 1984  Construction and genetic characterization of temperature-sensitive mutant alleles of the yeast actin gene. Proc. Natl. Acad. Sci. USA 81:4889-4893[Abstract/Free Full Text].

SIKORSKI, R. S. and P. HIETER, 1989  A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae.. Genetics 122:19-27[Abstract/Free Full Text].

SMITH, J. S., C. B. BRACHMANN, L. PILLUS, and J. D. BOEKE, 1998  Distribution of a limited Sir2 protein pool regulates the strength of yeast rDNA silencing and is modulated by Sir4p. Genetics 149:1205-1219[Abstract/Free Full Text].

SUZUKI, Y., Y. NOGI, A. ABE, and T. FUKASAWA, 1988  GAL11 protein, an auxiliary transcription factor for genes encoding galactose-metabolizing enzymes in Saccharomyces cerevisiae.. Mol. Cell. Biol. 8:4991-4999[Abstract/Free Full Text].

SWANSON, M. S., E. A. MALONE, and F. WINSTON, 1991  SPT5, an essential gene important for normal transcription in Saccharomyces cerevisiae, encodes an acidic nuclear protein with a carboxy-terminal repeat. Mol. Cell. Biol. 11:3009-3019[Abstract/Free Full Text].

TOENJES, K. A., 1998 Functional analysis of yeast fimbrin. Ph.D. Thesis, University of Arizona, Tucson, AZ.

VALLEN, E. A., W. HO, M. WINEY, and M. D. ROSE, 1994  Genetic interactions between CDC31 and KAR1, two genes required for duplication of the microtubule organizing center in Saccharomyces cerevisiae.. Genetics 137:407[Abstract].

WEBB, G. C., M. HOEDT, L. J. POOLE, and E. W. JONES, 1997  Genetic interactions between a pep7 mutation and the PEP12 and VPS45 genes: evidence for a novel SNARE component in transport between the Saccharomyves cerevisiae golgi complex and endosome. Genetics 147:467-478[Abstract].

WEINSTEIN, B. and F. SOLOMON, 1990  Phenotypic consequences of tubulin overproduction in Saccharomyces cerevisiae: differences between alpha-tubulin and beta-tubulin. Mol. Cell. Biol. 10:5295-5304[Abstract/Free Full Text].

WERTMAN, K. F., D. G. DRUBIN, and D. BOTSTEIN, 1992  Systematic mutational analysis of the yeast ACT1 gene. Genetics 132:337-350[Abstract].

ZHANG, C.-J., M. M. CAVENAGH, and R. A. KAHN, 1998  A family of Arf effectors defined as suppressors of the loss of Arf function in the yeast Saccharomyces cerevisiae.. J. Biol. Chem. 273:19792-19796[Abstract/Free Full Text].

This article has been cited by other articles:

				S. R. Gross and T. G. Kinzy Improper Organization of the Actin Cytoskeleton Affects Protein Synthesis at Initiation Mol. Cell. Biol., March 1, 2007; 27(5): 1974 - 1989. [Abstract] [Full Text] [PDF]

				J. B. Moseley and B. L. Goode The Yeast Actin Cytoskeleton: from Cellular Function to Biochemical Mechanism Microbiol. Mol. Biol. Rev., September 1, 2006; 70(3): 605 - 645. [Abstract] [Full Text] [PDF]

				M. Ralser, U. Nonhoff, M. Albrecht, T. Lengauer, E. E. Wanker, H. Lehrach, and S. Krobitsch Ataxin-2 and huntingtin interact with endophilin-A complexes to function in plastin-associated pathways Hum. Mol. Genet., October 1, 2005; 14(19): 2893 - 2909. [Abstract] [Full Text] [PDF]

				W. P. Voth, J. D. Richards, J. M. Shaw, and D. J. Stillman Yeast vectors for integration at the HO locus Nucleic Acids Res., June 15, 2001; 29(12): e59 - e59. [Abstract] [Full Text] [PDF]

				R. Munshi, K. A. Kandl, A. Carr-Schmid, J. L. Whitacre, A. E. M. Adams, and T. G. Kinzy Overexpression of Translation Elongation Factor 1A Affects the Organization and Function of the Actin Cytoskeleton in Yeast Genetics, April 1, 2001; 157(4): 1425 - 1436. [Abstract] [Full Text]

J. L. Whitacre, D. A. Davis, K. A. Toenjes, S. M. Brower, and A. E. M. Adams Generation of an Isogenic Collection of Yeast Actin Mutants and Identification of Three Interrelated Phenotypes Genetics, February 1, 2001; 157(2): 533 - 543. [Abstract] [Full