Synthetic Interactions of the Post-Golgi sec Mutations of Saccharomyces cerevisiae

Corresponding author: Peter Novick, Department of Cell Biology, Yale University School of Medicine, P.O. Box 208002, New Haven, CT 06520-8002., peter.novick{at}yale.edu (E-mail)

Communicating editor: D. BOTSTEIN

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In the budding yeast Saccharomyces cerevisiae, synthetic lethality has been extensively used both to characterize interactions between genes previously identified as likely to be involved in similar processes as well as to uncover new interactions. We have performed a large study of the synthetic lethal interactions of the post-Golgi sec mutations. Included in this study are the interactions of the post-Golgi sec mutations with each other, with mutations affecting earlier stages of the secretory pathway, with selected mutations affecting the actin cytoskeleton, and with selected cell division cycle (cdc) mutations affecting processes thought to be important for or involving secretion, such as polarity establishment and cytokinesis. Synthetic negative interactions of the post-Golgi sec mutations appear (as predicted) to be largely stage specific, although there are some notable exceptions. The significance of these results is discussed in the context of both secretory pathway function and the utility of synthetic lethality studies and their interpretation.

THE term "synthetic lethality" was first used by DOBZHANSKY 1946 to describe the phenomenon where alleles of different genes are separately viable, but inviable when combined in a double mutant. Such interactions can be interpreted in several ways, depending upon the characteristics of the interacting alleles (reviewed in GUARENTE 1993 ). For example, if both mutations are null, the interpretation is that the genes are required in parallel pathways with a common essential function, loss of which is lethal. This type of interaction is frequently the result of genetic redundancy. In cases where single null mutations are lethal, mutations with partial function must be used to evaluate synthetic phenotypes. Synthetic lethality in such cases is then usually interpreted as indicating that both genes function in the same essential pathway. Such synergy could result from successive reductions in flow through the pathway at different discrete stages such that the lower throughput is the product of reduced efficiency of the two steps. More frequently, synthetic lethality is interpreted as resulting from genes affecting the same stage of a pathway, as when mutations weaken the interactions of subunits of a complex, although a synthetic lethal interaction does not necessarily imply a physical interaction.

In the budding yeast Saccharomyces cerevisiae, synthetic lethality was first noted in studies of the allele specificity of suppressors of actin mutations (NOVICK et al. 1989 ). It has subsequently been used extensively as a means of assessing whether gene products are involved in similar cellular processes, and it has been exploited in screens for new mutations affecting similar biological processes (BENDER and PRINGLE 1991 ). In the secretory pathway in yeast (KAISER et al. 1997 ), synthetic lethality has been used extensively both to characterize interactions between genes previously identified as likely to be involved in similar processes as well as to uncover new interactions. This pathway begins with the cotranslational translocation into the endoplasmic reticulum (ER) of proteins destined for export. Proteins begin to be modified in the ER and are subsequently transported to the Golgi complex via small vesicle carriers. In the Golgi, proteins are further modified, and then sorted from proteins that will reside in the vacuole, and transported via post-Golgi secretory vesicles that fuse with the plasma membrane. A combination of phenotypic and genetic analyses has divided this pathway into several stages: the formation of vesicles from the ER, the fusion of these vesicles with the Golgi, intra-Golgi transport, and the fusion of post-Golgi vesicles with the plasma membrane.

We present here a study of the synthetic lethal interactions and synthetic negative genetic interactions of one class of yeast secretory mutations, the post-Golgi sec mutations. These are temperature-sensitive alleles of the SEC1, SEC2, SEC3, SEC4, SEC5, SEC6, SEC8, SEC9, SEC10, and SEC15 genes (NOVICK et al. 1981 ). Their gene products include several members of a large complex required for vesicle tethering at the plasma membrane, a rab family GTPase and its accessory proteins, and SNAREs and their regulators. This study comprises the interactions of the post-Golgi sec mutations with each other, with mutations affecting earlier stages of the secretory pathway, with selected mutations affecting the actin cytoskeleton, and with selected cell division cycle (cdc) mutations affecting processes thought to be important for or involving secretion, such as polarity establishment and cytokinesis. A number of these interactions have been analyzed in earlier publications (SALMINEN and NOVICK 1987 ; NAIR et al. 1990 ; BOWSER et al. 1992 ; POTENZA et al. 1992 ; MOYA et al. 1993 ; GOVINDAN et al. 1995 ); here we extend, and in some cases reexamine, this analysis.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The S. cerevisiae strains used in this study are listed in Table 1. For each post-Golgi sec mutation, at least one temperature-sensitive (ts) allele was used (sec2-41, sec4-8, sec5-24, sec6-4, sec8-9, sec9-4, sec10-2, and sec15-1). In the cases of sec1, two alleles were used (sec1-1 and sec1-13), and for sec3, three alleles were used (sec3-2, sec3-4, and sec3-5). Standard procedures were used for yeast growth, mating, and genetics (GUTHRIE and FINK 1991 ). Matings were usually performed using complementation of auxotrophic markers to select for diploids. Where auxotrophic selection was not available, complementation of ts growth was used. Yeast were generally grown prior to sporulation on YPD plates; however, strains that did not sporulate efficiently on YPD were grown on presporulation plates. All strains were sporulated at 25° on sporulation medium containing 1% potassium acetate, 0.5% glucose, 0.2% raffinose, and 0.1% yeast extract. Growth phenotypes were assessed following replica plating of colonies that germinated at 25° onto YPD plates at 25°, 30°, 34°, and 37°. Where alleles were tagged with auxotrophic markers, testing for the presence of the marker was also by replica plating. In most cases, the presence of double mutants was inferred from the growth phenotypes of the other colonies in the tetrads. A minimum of 9 informative tetrads (those containing double mutants) out of at least 12 tetrads dissected were obtained for each cross. This was considered to be sufficient for cases in which no interaction was observed. In most cases at least 12 or more informative tetrads were obtained, particularly in instances where results were contrary to expectations. Interactions were considered to be significant if double mutants were dead or their growth dramatically reduced at 25°, or if the restrictive temperature was reduced by at least 3°.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Interaction of the post-Golgi secretory mutations with each other:
We first looked at the interactions of the post-Golgi sec mutations with each other, dividing them into three functional categories. Several of the post-Golgi sec gene products, those encoded by SEC3, SEC5, SEC6, SEC8, SEC10, and SEC15, are components of a large complex that is peripherally associated with and involved in vesicle tethering at the plasma membrane (BOWSER et al. 1992 ; TERBUSH and NOVICK 1995 ; TERBUSH et al. 1996 ; FINGER et al. 1998 ; GUO et al. 1999 ). When the post-Golgi sec mutations are crossed with mutations in the components of this tethering complex, we find that most combinations are synthetically lethal or lower the restrictive temperature, regardless of whether both components are complex members (Fig 1). The sole exceptions to this are that neither sec1-1 nor sec3-2 interacts with sec6-4, although the more severe alleles sec1-13, sec3-4, and sec3-5 are all synthetically lethal with sec6-4 (Fig 1).

View larger version (31K): In this window In a new window Download PPT slide   Figure 1. Crosses of post-Golgi sec mutants to mutants in members of the tethering complex. White boxes indicate no interaction, black boxes indicate synthetic lethality at 25°, and gray boxes indicate that the restrictive temperature of the double mutant was at least 3° lower than that of either single mutant. AL, the mutations are allelic; ND, not determined. Data for some crosses were previously published: sec2-41 x sec3-2, sec5-24, sec6-4, and sec10-2 (NAIR et al. 1990 ); sec4-8 x sec3-2, sec5-24, sec10-2, and sec15-1 (SALMINEN and NOVICK 1987 ); sec6-4 x sec1-1 and sec3-2 (POTENZA et al. 1992 ); sec8-9 x sec1-1, sec3-2, sec6-24, and sec10-2 (BOWSER et al. 1992 ); sec15-1 x sec3-2, sec5-24, and sec10-2 (SALMINEN and NOVICK 1987 ); sec2-41 x sec15-1 (SALMINEN and NOVICK 1987 ; NAIR et al. 1990 ); sec2-41 x sec8-9 (NAIR et al. 1990 ; BOWSER et al. 1992 ); sec6-4 x sec8-9 (BOWSER et al. 1992 ; POTENZA et al. 1992 ); sec8-9 x sec15-1 (SALMINEN and NOVICK 1987 ; BOWSER et al. 1992 ).

SNAREs and their regulators form the second subcategory of post-Golgi sec mutations. We have also included here sec17-1 and sec18-1, which are mutations defective for the yeast NSF and -SNAP homologs, respectively (WILSON et al. 1989 ; GRIFF et al. 1992 ). Although the observed secretory block for these two mutations is in ER-to-Golgi transport (NOVICK et al. 1981 ), we include them here as they are also required for post-Golgi secretion and physically interact with yeast post-Golgi SNAREs (BRENNWALD et al. 1994 ; GRAHAM and EMR 1991 ). The other members of this category are the two alleles of the gene encoding the SNARE interacting protein Sec1p (CARR et al. 1999 ), sec1-1 and sec1-13, and an allele of the SNAP-25 homolog Sec9p, sec9-4 (BRENNWALD et al. 1994 ). We find that sec1 and sec9 mutations display negative synthetic interactions with each other and with the other post-Golgi sec mutations (Fig 2). No interactions were observed in crosses with sec17-1 and sec18-1 (Fig 2).

View larger version (56K): In this window In a new window Download PPT slide   Figure 2. Crosses of post-Golgi sec mutants to mutants in SNAREs and their regulators. White boxes indicate no interaction, black boxes indicate synthetic lethality at 25°, and gray boxes indicate that the restrictive temperature of the double mutant was at least 3° lower than that of either single mutant. AL, the mutations are allelic; ND, not determined. Data for some crosses were previously published: sec2-41 x sec1-1, sec9-4, sec17-1, and sec18-1 (NAIR 1990, no. 27); sec4-8 x sec17-1 and sec18-1 (SALMINEN and NOVICK 1987 ); sec6-4 x sec18-1 (POTENZA et al. 1992 ); sec8-9 x sec1-1 and sec9-4 (BOWSER et al. 1992 ).

The last grouping of post-Golgi sec mutations are those affecting the rab-family GTPase, Sec4p (SALMINEN and NOVICK 1987 ), and its accessory proteins. These include the guanine nucleotide exchange factor (GEF), Sec2p (WALCH-SOLIMENA et al. 1997 ), the GDP dissociation inhibitor Gdi1p (allelic with SEC19; GARRETT et al. 1994 ), and the guanine-nucleotide releasing protein, Dss4p (MOYA et al. 1993 ). We find that sec4-8 and sec2-41 interact with all of the other post-Golgi sec mutations (Fig 3). In contrast, the interactions of sec19-1 and dss4 are more specific. There is no negative synthetic interaction between either sec9-4 or sec15-1 and sec19-1. dss4 interacts only with sec2-41, sec3-2, sec3-5, sec4-8, sec9-4, and sec15-1, but not with sec1-1, sec1-13, sec3-4, sec5-24, sec6-4, and sec10-2 (Fig 3).

View larger version (74K): In this window In a new window Download PPT slide   Figure 3. Crosses of post-Golgi sec mutants to mutants of sec4 and its accessories. White boxes indicate no interaction, black boxes indicate synthetic lethality at 25°, and gray boxes indicate that the restrictive temperature of the double mutant was at least 3° lower than that of either single mutant. AL, the mutations are allelic. Data for some crosses were previously published: sec2-41 x sec1-1, sec3-2, sec5-24, sec6-4, sec9-4, sec10-2, and sec19-1 (NAIR et al. 1990 ); sec4-8 x sec3-2, sec5-24, sec8-9, sec10-2, sec15-1, and sec19-1 (SALMINEN and NOVICK 1987 ); sec19-1 x sec6-4 (POTENZA et al. 1992 ); sec19-1 x sec8-9 (BOWSER et al. 1992 ); sec19-1 x sec15-1 (SALMINEN and NOVICK 1987 ); dss4 x sec1-1, sec2-41, sec3-2, sec4-8, sec5-24, sec6-4, sec8-9, sec9-4, sec10-2, and sec15-1 (MOYA et al. 1993 ).

Interaction of post-Golgi sec mutations with those affecting earlier stages of the secretory pathway:
We next crossed the post-Golgi sec mutants to those affecting ER-to-Golgi and intra-Golgi transport (Fig 4). With one striking exception we see no strong interaction of any of the post-Golgi sec mutations with the sec mutations affecting either ER-to-Golgi transport (sec12-4, sec13-1, sec16-2, sec20-1, sec21-1, sec22-3, and sec23-1), or with those affecting intra-Golgi transport (sec7-1 and sec14-3). This notable exception is the lowering of the temperature at which lethality occurs from 37° to 30°, compared to both single mutants, in the sec14-3 sec15-1 double mutant. SEC14 encodes a phosphatidylinositol/phosphatidylcholine transfer protein (BANKAITIS et al. 1990 ).

View larger version (31K): In this window In a new window Download PPT slide   Figure 4. Crosses of post-Golgi sec mutants to early secretory pathway mutants. White boxes indicate no interaction and the gray box indicates that the restrictive temperature of the double mutant was at least 3° lower than that of either single mutant. Synthetic lethality was not observed in any of these crosses. ND, not determined. Data for some crosses were previously published: sec2-41 x sec7-1, sec12-1, sec13-1, sec14-3, sec20-1, sec21-1, sec22-3, and sec23-1 (NAIR et al. 1990 ); sec4-8 x sec7-1, sec12-1, sec13-1, sec14-3, sec16-2, sec20-1, sec21-1, sec22-3, and sec23-1 (SALMINEN and NOVICK 1987 ); sec6-4 x sec7-1, sec12-1, sec13-1, sec14-3, sec16-2, sec20-1, sec21-1, and sec23-1 (POTENZA et al. 1992 ); sec8-9 x sec7-1, sec12-1, sec13-1, sec16-2, sec20-1, sec21-1, sec22-3, and sec23-1 (BOWSER et al. 1992 ); sec15-1 x sec22-3 (SALMINEN and NOVICK 1987 ).

The YPT1 gene product has been implicated in transport from the ER to the Golgi and in intra-Golgi transport (BACON et al. 1989 ; JEDD et al. 1995 ). We crossed four mutant alleles of YPT1 to the post-Golgi sec mutants. Three of these alleles are ts (ypt1-1, ypt1-3, and ypt1^I121,V161; SEGEV and BOTSTEIN 1987 ; SCHMITT et al. 1988 ; WUESTEHUBE et al. 1996 ), and one of them (ypt1-1) is quite slow growing (SEGEV and BOTSTEIN 1987 ). The remaining allele, ypt1-2, has no growth phenotype, but is synthetically lethal in combination with some ER-to-Golgi sec mutations and is defective in an in vitro ER to Golgi transport assay (BACON et al. 1989 ). We find that the post-Golgi sec mutations display negative genetic interactions with all of these ypt1 mutations, although not all combinations are deleterious or lethal (Fig 5).

View larger version (59K): In this window In a new window Download PPT slide   Figure 5. Crosses of post-Golgi sec mutants to ypt1 alleles. ypt1ts refers to ypt1I121,V161 (SCHMITT et al. 1988 ). White boxes indicate no interaction, black boxes indicate synthetic lethality at 25°, and gray boxes indicate that the restrictive temperature of the double mutant was at least 3° lower than that of either single mutant. ND, not determined.

Interaction of post-Golgi sec mutations with cytoskeletal and cdc mutations:
The actin cytoskeleton has been implicated in post-Golgi secretion (NOVICK and BOTSTEIN 1985 ), most particularly in the requirement for a functional actin cytoskeleton to properly localize several of the post-Golgi SEC gene products [Sec4p (AYSCOUGH et al. 1997 ; WALCH-SOLIMENA et al. 1997 ) and Sec8p (AYSCOUGH et al. 1997 ; FINGER et al. 1998 )] and for correctly polarized secretion (NOVICK and BOTSTEIN 1985 ). We find that none of the post-Golgi sec mutations interacts with the act1-1 ts actin mutant allele (Fig 6). Another cytoskeletal protein, the type V myosin Myo2p, has also been associated with post-Golgi secretion (JOHNSTON et al. 1991 ; GOVINDAN et al. 1995 ), and previous studies have identified a subset of post-Golgi sec mutations (sec 4-8, sec5-24, sec8-9, sec9-4, sec10-2, and sec15-1) that are synthetically lethal with myo2-66 (GOVINDAN et al. 1995 ). We have now documented an additional synthetic lethal interaction of myo2-66 with the severe sec3-4 mutation (Fig 6).

View larger version (45K): In this window In a new window Download PPT slide   Figure 6. Crosses of post-Golgi sec mutants to cytoskeletal and cdc mutants. White boxes indicate no interaction, black boxes indicate synthetic lethality at 25°, and gray boxes indicate that the restrictive temperature of the double mutant was at least 3° lower than that of either single mutant. ND, not determined. Data for some crosses were previously published: act1-1 x sec1-1, sec2-41, sec4-8, sec6-4, sec9-4, sec10-2, and sec15-1 (SALMINEN and NOVICK 1987 ); myo2-66 x sec1-1, sec2-41, sec3-2, sec4-8, sec5-24, sec6-4, sec8-9, sec9-4, sec10-2, and sec15-1 (GOVINDAN et al. 1995 ).

We were also curious as to the possibility of genetic interaction of the post-Golgi sec mutations with septins, as both sets of gene products are implicated in cytokinesis (LONGTINE et al. 1996 ; FINGER and NOVICK 1997 ; FINGER et al. 1998 ; ROTH et al. 1998 ), and recent reports suggest that septins may be localized to synaptic vesicles (BEITES et al. 1999 ) as well as to the plasma membrane and that they coassociate with members of the mammalian sec6/8 complex (HSU et al. 1998 ) and with the t-SNARE syntaxin-1 (BEITES et al. 1999 ). The only interactions we see with cdc12-6, a tight ts allele of an essential septin (HAARER and PRINGLE 1987 ; LONGTINE et al. 1996 ), are synthetic lethality with sec9-4 and a weak interaction with sec2-41 at 30° (Fig 6).

CDC42 and CDC24 encode, respectively, a rho-family GTPase and its GEF, which are important for establishment of yeast cell polarity (SLOAT and PRINGLE 1978 ; ADAMS et al. 1990 ; JOHNSON and PRINGLE 1990 ; ZHENG et al. 1994 ). We have also crossed all of the post-Golgi sec mutants to the ts cdc24-4 and cdc42-1 mutants. We find no interaction of cdc24-4 with any of the post-Golgi sec genes. In contrast, several of the post-Golgi sec genes interact with cdc42-1. Strong interactions are seen with sec5-24, sec8-9, sec10-2, and sec15-1, all components of the plasma membrane-associated tethering complex (Fig 6). Weaker interactions are seen with sec3-2, sec4-8, and sec9-4 (Fig 6).

The final group of crosses we performed with the post-Golgi sec mutants are those with the ts cdc28-1 mutant. CDC28 encodes the yeast cyclin-dependent kinase, and its association with different cyclins triggers changes in the site of secretion over the course of the cell cycle (LEW and REED 1993 ). A strong interaction between cdc28-1 and sec3-2, where the temperature at which lethality occurred for the double mutant was lowered from 37° in each of the single mutants to 34°, with significantly impaired growth at 30°, was the only interaction detected (Fig 6).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Synthetic lethality, a strong genetic interaction whereby two viable mutations are lethal when combined in a double mutant, is usually interpreted as indicating that the gene products in question function at the same stage of a biological pathway or in parallel pathways. We have completed a large study of the synthetic lethal interactions of post-Golgi sec mutations in combination with each other, with other secretory pathway mutations, with selected mutations affecting the actin cytoskeleton, and with selected cdc mutations affecting processes thought to be important for or involving secretion, such as polarity establishment and cytokinesis. As this is the largest study of this kind of which we are aware, the results are informative for interpreting synthetic lethality in general, in addition to the specific implications of these results with regard to post-Golgi secretion.

Two models have been invoked to explain observed synthetic lethal interactions of partially functional mutations (GUARENTE 1993 ), such as the ts post-Golgi sec mutations. The first model, the "pipeline model," proposes that reduced flux at multiple stages of a pathway limits the flow through the pathway below levels required for viability. Alternatively, synthetic lethality is interpreted as resulting from genes affecting the same stage of a pathway, as when mutations weaken the interactions of subunits of a complex, although a physical interaction is not necessarily implied.

There is ample evidence, including the data presented here, that the pipeline model does not hold true for the secretory pathway, as mutations affecting ER-to-Golgi and intra-Golgi transport are not, in general, synthetically lethal with the post-Golgi sec mutations. In fact, it has previously been shown that even within the ER-to-Golgi transport stage, interactions occur among those mutations affected in vesicle budding from the ER, or among mutations affecting fusion of ER-derived vesicles with the Golgi, but not between the two classes of mutations (KAISER and SCHEKMAN 1990 ). The interactions of the post-Golgi sec mutations with each other in almost every possible pairwise combination also indicate that synthetic lethal interactions are characteristic of mutations affecting the same stage of a biological pathway. Furthermore, these extensive interactions suggest that post-Golgi secretion requires the concerted action of all of the post-Golgi SEC gene products, and that any division of post-Golgi secretion into substages may be an artificial construct, rather than a reflection of discrete targeting, docking, and fusion steps. Although many of the interactions found are between components of complexes (e.g., between sec8 and all other components of the tethering complex), in many cases interactions were seen between genes encoding proteins not thought to be binding partners (e.g., sec4 and ypt1). In some cases, mutations affecting proteins that are known to physically interact did not genetically interact. For example, sec17-1 and sec18-1 did not interact with mutations affecting the t-SNARE Sec9p or the SNARE-interacting protein Sec1p, even though Sec17p and Sec18p act to disassemble the SNARE complex that contains Sec9p (BRENNWALD et al. 1994 ; GROTE and NOVICK 1999 ) and that is bound by Sec1p (CARR et al. 1999 ). Since sec17-1 and sec18-1 block many stages of membrane traffic (GRAHAM and EMR 1991 ), it is possible that post-Golgi traffic is affected less severely than other stages at intermediate temperatures. Alternatively, the lack of interactions may indicate that the different mutations affect different aspects of SNARE complex function. While sec18-1 is known to block SNARE complex disassembly, sec1-1, sec1-13, and sec9-4 may affect complex assembly or membrane fusion.

A frequently stated caution with regard to interpretation of these types of negative interactions is that the combination of two sickly mutations may result in a more severe phenotype that is additive, rather than synergistic (BOTSTEIN et al. 1997 ). We do not find that this is the case in our studies. For example, sec3-4, sec3-5, and sec1-13 are all quite slow growing at 25° and also display lethality in combination with the severe ypt1-1 mutation. However, synthetic lethal interactions are also seen with weaker alleles of sec1, sec3, and ypt1, suggesting that these interactions are, in fact, specific. Furthermore, none of the severe post-Golgi mutations displays any synthetic phenotype with ER-to-Golgi sec mutations, or with the sickly act1-1 mutation. If the phenotypes were simply additive, we would expect to see lethality in all cases. As we do not, we see no reason to interpret any of these interactions as nonspecifically additive.

In the case of the interactions of the post-Golgi sec mutations with alleles of ypt1, we propose, using the stage-specificity model for synthetic lethality, that these interactions are reflective of a role for Ypt1p in post-Golgi secretion, in addition to its previously documented roles in ER-to-Golgi and intra-Golgi transport (BACON et al. 1989 ; JEDD et al. 1995 ). It has been shown that the post-Golgi sec mutants sec1-1 and sec6-4 accumulate multiple classes of vesicles, including those that are immunoreactive with antibodies against Sec4p and those that are immunoreactive with antibodies against Ypt1p (MULHOLLAND et al. 1997 ). sec4-8 mutants also accumulate vesicles that are immunoreactive with antibodies against Ypt1p (MULHOLLAND et al. 1997 ). These results support the suggestion that the genetic interactions between post-Golgi sec mutations and ypt1 mutations are indicative of the cooperation of their gene products in vivo to effect exocytosis. We have previously reported that high-copy SEC3 lowers the restrictive temperature for ypt1-3 and ypt1^I121,V161 (FINGER and NOVICK 1997 ). As sec3-4 and sec3-5 mutants accumulate membranes from earlier stages of the secretory pathway, in addition to post-Golgi secretory vesicles, and are partially blocked in early stages of transport (FINGER and NOVICK 1997 ), the genetic interactions of post-Golgi sec genes with ypt1 mutant alleles may reflect a possible role for Sec3p in these earlier stages of transport.

The only other strong interaction seen with a mutation in an earlier stage of the secretory pathway is that between sec14-3, defective in a phosphotidyl inositol/phosphotidyl choline (PI/PC) transfer protein (BANKAITIS et al. 1990 ), and sec15-1. Sec15p is found (at least when overexpressed) on post-Golgi vesicles and appears to be an effector for the rab protein Sec4p (GUO et al. 1999 ). If Sec15p binds to vesicles via interaction with PI lipids, this could explain that genetic interaction. In mammalian cells an effector for Rab5, EEA1, also binds to phosphatidylinositol trisphosphate (SIMONSEN et al. 1998 ). This could point to a general requirement for PI lipids in rab effector activity. Alternatively, as both act1 and sec14-1 mutations are suppressed by sac1 (CLEVES et al. 1989 ), this interaction could also reflect the requirement for actin in yeast exocytosis.

Although actin mutants accumulate post-Golgi secretory vesicles and are partially defective in exocytosis of invertase (NOVICK and BOTSTEIN 1985 ), the role of actin in secretion has been elusive. None of the post-Golgi sec mutations interact with the ts actin mutation, act1-1 (this study and SALMINEN and NOVICK 1987 ), but such interactions may certainly be allele-specific. Mutations in three actin-associated proteins, Myo2p (GOVINDAN et al. 1995 ), Pfy1p (profilin; HAARER et al. 1996 ; FINGER and NOVICK 1997 ), and Tpm1p/Tpm2p (tropomyosin; LIU and BRETSCHER 1992 ), do genetically interact with the post-Golgi sec mutations, and it is difficult to imagine that such interactions would be reflective of roles for these three proteins that are completely independent of their well-characterized roles in the actin cytoskeleton. A ts profilin mutant, pfy1-111, did not accumulate post-Golgi vesicles upon shift to the restrictive temperature, suggesting that its role in secretion may be indirect (FINGER and NOVICK 1997 ). Tropomyosin-containing actin cables were recently demonstrated to be required for the Myo2p-dependent polarized delivery of post-Golgi secretory vesicles (PRUYNE et al. 1998 ), and actin is also required for polarized distribution of Sec4p (AYSCOUGH et al. 1997 ; WALCH-SOLIMENA et al. 1997 ) and Sec8p (AYSCOUGH et al. 1997 ; FINGER et al. 1998 ). It is possible, although unlikely given the phenotype of the mutant, that the actin allele that was used in this study is preferentially defective in other actin-requiring processes. An interpretation of the lack of genetic interaction between actin and the post-Golgi sec mutants that fits the available data is that the stage of the secretory pathway at which actin functions is upstream of that defined by the post-Golgi sec mutants.

Finally, the crosses with cell cycle (cdc) mutants divide the secretory mutants into several classes, in contrast to the other categories where most or all sec mutations displayed similar interactions. With the septin mutation cdc12-6, we see interactions only with sec9-4, the yeast SNAP-25 homolog (BRENNWALD et al. 1994 ), and with sec2-41, defective in the GEF for the rab-family GTPase Sec4p (WALCH-SOLIMENA et al. 1997 ). The mammalian septins CDCrel-1 and Nedd5 are both able to bind to the t-SNARE syntaxin-1 (BEITES et al. 1999 ), suggesting that the genetic interaction between cdc12-6 and sec9-4 could potentially reflect physical interactions of yeast septins and t-SNAREs. A significant fraction of the mammalian septin CDCrel-1 is localized to synaptic vesicles (BEITES et al. 1999 ); however, there is currently no evidence for yeast septins localizing to post-Golgi secretory vesicles, where Sec2p is found (WALCH-SOLIMENA et al. 1997 ). This interaction could also indicate a role for Sec2p outside of its known function in the secretory pathway. Interactions are, somewhat surprisingly, not seen with members of the plasma membrane-associated vesicle tethering complex, although the corresponding mammalian sec6/8 complex copurifies with septins (HSU et al. 1998 ). These results are consistent, however, with the finding that SEC gene products do not appear to colocalize with septins at cytokinesis and that at least one of them, Sec3p, can localize to the division site independent of septin function (FINGER et al. 1998 ). These results imply that in yeast the functions of this complex and the septins are separable.

The other cdc mutations used in this study are cdc28-1, affecting a cyclin-dependent kinase that regulates changes in the sites of secretion during the cell cycle (LEW and REED 1993 ), and cdc24-4 and cdc42-1, affecting the GEF for a rho-family GTPase involved in cell polarity establishment (including establishment of the polarity of the actin cytoskeleton; ZHENG et al. 1994 ) and the rho-family GTPase (JOHNSON and PRINGLE 1990 ), respectively. We find that cdc28-1 interacts only with sec3-2, a mild allele of SEC3. This is consistent with results that localization of Sec3p, which is important for establishing sites of polarized secretion in yeast, is defective only in cdc28-1 mutants among many mutants tested (FINGER et al. 1998 ). Interestingly, Sec3p contains one or more consensus sites for Cdc28p-dependent phosphorylation (FINGER and NOVICK 1998 ). None of the post-Golgi sec mutations interact with cdc24-4, but several interact with cdc42-1, which is formally downstream of cdc24. The strongest interactions are seen between cdc42-1 and sec5-24, sec8-9, sec10-2, and sec15-1, all components of the plasma membrane tethering complex. Weaker interactions are seen with sec3-2, sec4-8, and sec9-4. The interactions with these cdc mutations are consistent with the polarity of the secretory pathway being established and/or regulated at different stages of the hierarchy for establishment of the general polarity of the yeast cell. Other possible interpretations of these data include Cdc42p functioning in secretion beyond a role in the initial polarization of the secretory pathway, or the tethering complex working with Cdc42p to perform other, nonsecretory functions.

Although the yeast genome is now completely sequenced, no functions have yet been ascribed to a considerable fraction of the genes (GOFFEAU et al. 1996 ). Synthetic lethality is one of many tools that can be used to understand the relationships between genes of both known and unknown functions. Our results indicate that synthetic lethality studies, while of considerable utility, must be interpreted with some caution. A failure to observe synthetic effects does not establish that the gene products do not interact. Furthermore, the observation of synthetic interactions does not by itself reveal the underlying relationship between the gene products. However, in combination with other more direct approaches, synthetic lethality studies can provide important insights into gene function.

	FOOTNOTES

¹ Present address: Laboratory of Molecular Biology, University of Wisconsin, 1525 Linden Dr., Madison, WI 53706.

	ACKNOWLEDGMENTS

We thank Mary Travers for expert technical assistance. These studies were supported by a National Institutes of Health training grant and by a Miles Scholar Award to F.P.F., and by National Institutes of Health grant GM-35370 to P.N.

Manuscript received March 21, 2000; Accepted for publication July 3, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ADAMS, A. E. M., D. I. JOHNSON, R. M. LONGNECKER, B. F. SLOAT, and J. R. PRINGLE, 1990  CDC42 and CDC43, two additional genes involved in budding and the establishment of cell polarity in the yeast Saccharomyces cerevisiae.. J. Cell Biol. 111:131-142[Abstract/Free Full Text].

AYSCOUGH, K. R., J. STRYKER, N. POKALA, M. SANDERS, and P. CREWS et al., 1997  High rates of actin filament turnover in budding yeast and roles for actin in establishment and maintenance of cell polarity revealed using the actin inhibitor latrunculin-A. J. Cell Biol. 137:399-416[Abstract/Free Full Text].

BACON, R. A., A. SALMINEN, H. RUOHOLA, P. NOVICK, and S. FERRO-NOVICK, 1989  The GTP-binding protein Ypt1 is required for transport in vitro: the Golgi apparatus is defective in ypt1 mutants. J. Cell Biol. 109:1015-1022[Abstract/Free Full Text].

BANKAITIS, V. A., J. R. AITKEN, A. E. CLEVES, and W. DOWHAN, 1990  An essential role for a phospholipid transfer protein in yeast Golgi function. Nature 347:561-562[Medline].

BEITES, C., H. XIE, R. BOWSER, and R. TRIMBLE, 1999  The septin CDCrel-1 binds syntaxin and inhibits exocytosis. Nat. Neurosci. 2:434-439[Medline].

BENDER, A. and J. PRINGLE, 1991  Use of a screen for synthetic lethal and multicopy suppressee mutants to identify two new genes involved in morphogenesis in Saccharomyces cerevisiae. Mol. Cell. Biol. 11:1295-1305[Abstract/Free Full Text].

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, edited by J. R. PRINGLE, J. R. BROACH and E. W. JONES. Cold Spring Harbor Laboratory Press, Plainview, NY.

BOWSER, R., H. MULLER, B. GOVINDAN, and P. NOVICK, 1992  Sec8p and Sec15p are components of a plasma membrane-associated 19.5S particle that may function downstream of Sec4p to control exocytosis. J. Cell Biol. 118:1041-1056[Abstract/Free Full Text].

BRENNWALD, P., B. KEARNS, K. CHAMPION, S. KERÄNEN, and V. BANKAITIS et al., 1994  Sec9 is a SNAP-25-like component of a yeast SNARE complex that may be the effector of Sec4 function in exocytosis. Cell 79:245-258[Medline].

CARR, C. M., E. GROTE, M. MUNSON, F. M. HUGHSON, and P. J. NOVICK, 1999  Sec1p binds to SNARE complexes and concentrates at sites of secretion. J. Cell Biol. 146:334-344.

CLEVES, A. E., P. J. NOVICK, and V. A. BANKAITAS, 1989  Mutations in the SAC1 gene suppress defects in yeast Golgi and yeast actin function. J. Cell Biol. 109:2939-2950[Abstract/Free Full Text].

DOBZHANSKY, T., 1946  Genetics of natural populations. XIII. Recombination and variability in populations of Drosophila pseudoobscura. Genetics 31:269-290[Free Full Text].

FINGER, F. P. and P. NOVICK, 1997  Sec3p is involved in secretion and morphogenesis in Saccharomyces cerevisiae.. Mol. Biol. Cell 8:647-662[Abstract].

FINGER, F. P. and P. NOVICK, 1998  Spatial regulation of exocytosis: lessons from yeast. J. Cell Biol. 142:609-612[Free Full Text].

FINGER, F. P., T. E. HUGHES, and P. NOVICK, 1998  Sec3p is a spatial landmark for polarized secretion in budding yeast. Cell 92:559-571[Medline].

GARRETT, M. D., J. E. ZAHNER, C. M. CHENEY, and P. J. NOVICK, 1994  GDI1 encodes a GDP dissociation inhibitor that plays an essential role in the yeast secretory pathway. EMBO J. 13:1718-1728[Medline].

GOFFEAU, A., B. G. BARRELL, H. BUSSEY, R. W. DAVIS, and B. DUJON et al., 1996  Life with 6000 genes. Science 274:563-567.

GOVINDAN, B., R. BOWSER, and P. NOVICK, 1995  The role of Myo2, a yeast class V myosin, in vesicular transport. J. Cell Biol. 128:1055-1068[Abstract/Free Full Text].

GRAHAM, T. R. and S. D. EMR, 1991  Compartmental organization of Golgi-specific protein modification and vacuolar protein sorting events defined in a yeast sec18 (NSF) mutant. J. Cell Biol. 114:207-218[Abstract/Free Full Text].

GRIFF, I. C., R. SCHEKMAN, J. E. ROTHMAN, and C. A. KAISER, 1992  The yeast SEC17 gene product is functionally equivalent to mammalian alpha-SNAP protein. J. Biol. Chem. 267:12106-12115[Abstract/Free Full Text].

GROTE, E. and P. J. NOVICK, 1999  Promiscuity in rab-SNARE interactions. Mol. Biol. Cell 10:4149-4161[Abstract/Free Full Text].

GUARENTE, L., 1993  Synthetic enhancement in gene interaction: a genetic tool come of age. Trends Genet. 9:362-366[Medline].

GUO, W., D. ROTH, C. WALCH-SOLIMENA, and P. NOVICK, 1999  The exocyst is an effector for Sec4p, targeting secretory vesicles to sites of exocytosis. EMBO J. 18:1071-1080[Medline].

GUTHRIE, C. and G. R. FINK, 1991  Guide to yeast genetics and molecular biology. Methods Enzymol. 194:3-37[Medline].

HAARER, B. K. and J. R. PRINGLE, 1987  Immunofluorescence localization of the Saccharomyces cerevisiae CDC12 gene product to the vicinity of the 10-nm filaments in the mother-bud neck. Mol. Cell. Biol. 7:3678-3687[Abstract/Free Full Text].

HAARER, B. K., A. CORBETT, Y. KWEON, A. S. PETZOLD, and P. SILVER et al., 1996  SEC3 mutations are synthetically lethal with profilin mutations and cause defects in diploid-specific bud-site selection. Genetics 144:495-510[Abstract].

HSU, S., C. HAZUKA, R. ROTH, D. FOLETTI, and J. HEUSER et al., 1998  Subunit composition, protein interactions, and structures of the mammalian brain sec6/8 complex and septin filaments. Neuron 20:1111-1122[Medline].

JEDD, G., C. RICHARDSON, R. LITT, and N. SEGEV, 1995  The Ypt1 GTPase is essential for the first two steps of the yeast secretory pathway. J. Cell Biol. 131:583-590[Abstract/Free Full Text].

JOHNSON, D. I. and J. R. PRINGLE, 1990  Molecular characterization of CDC42, a Saccharomyces cerevisiae gene involved in the development of cell polarity. J. Cell Biol. 111:143-152[Abstract/Free Full Text].

JOHNSTON, G. C., J. A. PRENDERGAST, and R. A. SINGER, 1991  The Saccharomyces cerevisiae MYO2 gene encodes an essential myosin for vectorial transport of vesicles. J. Cell Biol. 113:539-551[Abstract/Free Full Text].

KAISER, C. A. and R. SCHEKMAN, 1990  Distinct sets of SEC genes govern transport vesicle formation and fusion early in the secretory pathway. Cell 61:723-733[Medline].

KAISER, C. A., R. E. GIMENO and D. SHAYWITZ, 1997 Protein secretion, membrane biogenesis, and endocytosis, pp. 91–227 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, Plainview, NY.

LEW, D. J. and S. I. REED, 1993  Morphogenesis in the yeast cell cycle: regulation by Cdc28 and cyclins. J. Cell Biol. 120:1305-1320[Abstract/Free Full Text].

LIU, H. and A. BRETSCHER, 1992  Characterization of TPM1 disrupted yeast cells indicates an involvement of tropomyosin in directed vesicular transport. J. Cell Biol. 118:285-299[Abstract/Free Full Text].

LONGTINE, M. S., D. J. DEMARINI, M. L. VALENCIK, O. S. AL-AWAR, and H. FARES et al., 1996  The septins: roles in cytokinesis and other processes. Curr. Opin. Cell Biol. 8:106-119[Medline].

MOYA, M., D. ROBERTS, and P. NOVICK, 1993  Dss4-1 is a dominant suppressor of sec4-8 that encodes a nucleotide exchange protein that aids Sec4p function. Nature 361:460-463[Medline].

MULHOLLAND, J., A. WESP, H. RIEZMAN, and D. BOTSTEIN, 1997  Yeast actin cytoskeleton mutants accumulate a new class of Golgi-derived secretory vesicle. Mol. Biol. Cell 8:1481-1499[Abstract].

NAIR, J., H. MULLER, M. PETERSON, and P. NOVICK, 1990  Sec2 protein contains a coiled-coil domain essential for vesicular transport and a dispensable carboxy terminal domain. J. Cell Biol. 110:1897-1909[Abstract/Free Full Text].

NOVICK, P. and D. BOTSTEIN, 1985  Phenotypic analysis of temperature-sensitive yeast actin mutants. Cell 40:405-416[Medline].

NOVICK, P., S. FERRO, and R. SCHEKMAN, 1981  Order of events in the yeast secretory pathway. Cell 25:461-469[Medline].

NOVICK, P., B. C. OSMOND, and D. BOTSTEIN, 1989  Suppressors of yeast actin mutants. Genetics 121:659-671[Abstract/Free Full Text].

POTENZA, M., R. BOWSER, H. MULLER, and P. NOVICK, 1992  SEC6 encodes an 85 kD soluble protein required for exocytosis in yeast. Yeast 8:549-558[Medline].

PRUYNE, D. W., D. H. SCHOTT, and A. BRETSCHER, 1998  Tropomyosin-containing actin cables direct the Myo2p-dependent polarized delivery of secretory vesicles in budding yeast. J. Cell Biol. 143:1931-1945[Abstract/Free Full Text].

ROTH, D., W. GUO, and P. NOVICK, 1998  Dominant negative alleles of SEC10 reveal distinct domains involved in secretion and morphogenesis in yeast. Mol. Biol. Cell 9:1725-1739[Abstract/Free Full Text].

SALMINEN, A. and P. J. NOVICK, 1987  A ras-like protein is required for a post-Golgi event in yeast secretion. Cell 49:527-538[Medline].

SCHMITT, H. D., M. PUZICHA, and D. GALLWITZ, 1988  Study of a temperature sensitive mutant of the ras-related YPT1 gene product in yeast suggests a role in the regulation of intracellular calcium. Cell 53:635-647[Medline].

SEGEV, N. and D. BOTSTEIN, 1987  The ras-like yeast YPT1 gene is itself essential for growth, sporulation, and starvation response. Mol. Cell. Biol. 7:2367-2377[Abstract/Free Full Text].

SIMONSEN, A., R. LIPPE, S. CHRISTOFORIDIS, J. GAULLIER, and A. BRECH et al., 1998  EEA1 links PI(3)K function to Rab5 regulation of endosome fusion. Nature 394:494-498[Medline].

SLOAT, B. F. and J. R. PRINGLE, 1978  A mutant of yeast defective in cellular morphogenesis. Science 200:1171-1173[Abstract/Free Full Text].

TERBUSH, D. R. and P. NOVICK, 1995  Sec6, Sec8, and Sec15 are components of a multisubunit complex which localizes to small bud tips in Saccharomyces cerevisiae.. J. Cell Biol. 130:299-312[Abstract/Free Full Text].

TERBUSH, D. R., T. MAURICE, D. ROTH, and P. NOVICK, 1996  The exocyst is a multiprotein complex required for exocytosis in Saccharomyces cerevisiae.. EMBO J. 15:6483-6494[Medline].

WALCH-SOLIMENA, C., R. N. COLLINS, and P. J. NOVICK, 1997  Sec2p mediates nucleotide exchange on Sec4p and is involved in polarized delivery of post-Golgi secretory vesicles. J. Cell Biol. 137:1495-1509[Abstract/Free Full Text].

WILSON, D. W., C. A. WILCOX, G. C. FLYNN, E. CHEN, and W.-J. KUANG et al., 1989  A fusion protein required for vesicle-mediated transport in both mammalian cells and yeast. Nature 339:355-359[Medline].

WUESTEHUBE, L. J., R. DUDEN, A. EUN, S. HAMAMOTO, and P. KORN et al., 1996  New mutants of Saccharomyces cerevisiae affected in the transport of proteins from the endoplasmic reticulum to the Golgi complex. Genetics 142:393-406[Abstract].

ZHENG, Y., R. CERIONE, and A. BENDER, 1994  Control of the yeast bud-site assembly GTPase Cdc42. J. Biol. Chem. 269:2369-2372[Abstract/Free Full Text].

This article has been cited by other articles:

				J. A. Songer and M. Munson Sec6p Anchors the Assembled Exocyst Complex at Sites of Secretion Mol. Biol. Cell, February 1, 2009; 20(3): 973 - 982. [Abstract] [Full Text] [PDF]

				C. Kurischko, V. K. Kuravi, N. Wannissorn, P. A. Nazarov, M. Husain, C. Zhang, K. M. Shokat, J. M. McCaffery, and F. C. Luca The Yeast LATS/Ndr Kinase Cbk1 Regulates Growth via Golgi-dependent Glycosylation and Secretion Mol. Biol. Cell, December 1, 2008; 19(12): 5559 - 5578. [Abstract] [Full Text] [PDF]

				Y. Fang, R. Sugiura, Y. Ma, T. Yada-Matsushima, H. Umeno, and T. Kuno Cation Diffusion Facilitator Cis4 Is Implicated in Golgi Membrane Trafficking via Regulating Zinc Homeostasis in Fission Yeast Mol. Biol. Cell, April 1, 2008; 19(4): 1295 - 1303. [Abstract] [Full Text] [PDF]

				Y. Ma, T. Kuno, A. Kita, T. Nabata, S. Uno, and R. Sugiura Genetic Evidence for Phospholipid-Mediated Regulation of the Rab GDP-Dissociation Inhibitor in Fission Yeast Genetics, November 1, 2006; 174(3): 1259 - 1271. [Abstract] [Full Text] [PDF]

				E. Milgrom, R. W. West Jr., C. Gao, and W.-C. W. Shen TFIID and Spt-Ada-Gcn5-Acetyltransferase Functions Probed by Genome-wide Synthetic Genetic Array Analysis Using a Saccharomyces cerevisiae taf9-ts Allele Genetics, November 1, 2005; 171(3): 959 - 973. [Abstract] [Full Text] [PDF]

				L. F. Ciufo, J. W. Barclay, R. D. Burgoyne, and A. Morgan Munc18-1 Regulates Early and Late Stages of Exocytosis via Syntaxin-independent Protein Interactions Mol. Biol. Cell, February 1, 2005; 16(2): 470 - 482. [Abstract] [Full Text] [PDF]

				C. Z. Chen, M. Calero, C. J. DeRegis, M. Heidtman, C. Barlowe, and R. N. Collins Genetic Analysis of Yeast Yip1p Function Reveals a Requirement for Golgi-Localized Rab Proteins and Rab-Guanine Nucleotide Dissociation Inhibitor Genetics, December 1, 2004; 168(4): 1827 - 1841. [Abstract] [Full Text] [PDF]

				O. C. Rodriguez and R. E. Cheney Human myosin-Vc is a novel class V myosin expressed in epithelial cells J. Cell Sci., January 3, 2002; 115(5): 991 - 1004. [Abstract] [Full Text] [PDF]

				X. Zhang, E. Bi, P. Novick, L. Du, K. G. Kozminski, J. H. Lipschutz, and W. Guo Cdc42 Interacts with the Exocyst and Regulates Polarized Secretion J. Biol. Chem., December 7, 2001; 276(50): 46745 - 46750. [Abstract] [Full Text] [PDF]

				J. E. Adamo, J. J. Moskow, A. S. Gladfelter, D. Viterbo, D. J. Lew, and P. J. Brennwald Yeast Cdc42 functions at a late step in exocytosis, specifically during polarized growth of the emerging bud J. Cell Biol., November 12, 2001; 155(4): 581 - 592. [Abstract] [Full Text] [PDF]