The research report should be a detailed narrative explaining the function, biological processes, and localization of the gene product. Citations should be given for all claims.
You should prioritize authoritative reviews and primary scientific literature when conducting research. You can supplement
this with annotations you find in gene/protein databases, but these can be outdated or inaccurate.
We are specifically interested in the primary function of the gene - for enzymes, what reaction is catalyzed, and what is the substrate specificity? For transporters, what is the substrate? For structural proteins or adapters, what is the broader structural role? For signaling molecules, what is the role in the pathway.
We are interested in where in or outside the cell the gene product carries out its function.
We are also interested in the signaling or biochemical pathways in which the gene functions. We are less interested in broad pleiotropic effects, except where these elucidate the precise role.
Include evidence where possible. We are interested in both experimental evidence as well as inference from structure, evolution, or bioinformatic analysis. Precise studies should be prioritized over high-throughput, where available.
Identity is verified. The requested protein is the S. cerevisiae S288c gene product Swe1, encoded by SWE1/YJL187C and represented by UniProt P32944. It is the budding-yeast homolog of Wee1 and a protein kinase whose established primary role is to inhibit the cyclin-dependent kinase Cdc28/Cdk1. This identity agrees with the supplied protein-kinase, kinase-like, ATP-binding, and Ser/Thr-kinase-domain annotations. Papers concerning MoSwe1 from Magnaporthe oryzae, Swe1 proteins from Candida, or Wee1 proteins in other organisms were excluded from the functional annotation.
Swe1’s principal catalytic output is phosphorylation of Tyr19 in Cdc28, particularly Cdc28 assembled with mitotic B-type cyclins such as Clb2. This lowers mitotic CDK activity, postponing nuclear division until a bud and an appropriately organized cortical cytoskeleton are available. Swe1 can additionally bind and inhibit Clb2–Cdc28 in a partly phosphorylation-independent manner. Its activity is controlled spatially: Swe1 accumulates in the nucleus during S phase, subsequently appears on the daughter side of the septin-organized bud neck, and is then hyperphosphorylated, ubiquitylated, and degraded. Thus, Swe1 is best annotated as the central effector of a morphogenesis-surveillance circuit, not merely as a generic mitotic inhibitor. (asano2005concertedmechanismof pages 1-1, shulewitz1999hsl7localizesto pages 13-14, raspelli2013theproteinkinase pages 21-26)
The literature consistently identifies budding-yeast Swe1 as the S. cerevisiae Wee1-family kinase that negatively regulates Cdc28. No conflicting identity was found for SWE1/YJL187C. The supplied InterPro assignments—protein-kinase domain, kinase-like fold, ATP-binding region, and catalytic kinase signature—are fully consistent with this experimentally established activity. Although UniProt describes the protein as belonging broadly to the Ser/Thr protein-kinase superfamily, its defining physiological reaction is phosphorylation of a tyrosine, Cdc28 Tyr19. Such catalytic specificity is characteristic of Wee1-family kinases and is not contradicted by classification based on the overall eukaryotic protein-kinase fold. (asano2005concertedmechanismof pages 1-1, asano2005concertedmechanismof pages 1-3)
| Functional claim | Molecular/spatial mechanism | Strongest evidence type | Representative source and date/DOI | Confidence / caveat |
|---|---|---|---|---|
| Identity verified: SWE1 is YJL187C/P32944, the S. cerevisiae Wee1-family mitosis-inhibitory protein kinase | Its protein-kinase domain is consistent with phosphorylation-dependent inhibition of the major CDK Cdc28; non-S. cerevisiae Swe1 orthologs are excluded | Sequence/family annotation supported by organism-specific biochemical and genetic literature | Asano et al., published online 26 May 2005, 10.1038/sj.emboj.7600683 (asano2005concertedmechanismof pages 1-1, asano2005concertedmechanismof pages 1-3) | High; identity, organism, locus, and Wee1-family function are concordant |
| Primary catalytic output: Swe1 phosphorylates Cdc28/Cdk1 Tyr19 | ATP-dependent transfer of phosphate to Tyr19 in the Cdc28 ATP-binding region inhibits predominantly B-type cyclin–Cdc28 activity and restrains mitotic entry | Biochemical kinase assays plus SWE1/MIH1 dosage genetics and nonphosphorylatable CDC28-Y19 mutant analysis | Sia et al., November 1996, 10.1091/mbc.7.11.1657; Asano et al., May–June 2005, 10.1038/sj.emboj.7600683 (asano2005concertedmechanismof pages 1-1, shulewitz1999hsl7localizesto pages 13-14) | High; Cdc28 Tyr19 is the best-established physiologically central substrate |
| Cyclin-complex specificity favors mitotic Clb–Cdc28, especially Clb2–Cdc28 | Cyclin identity and Clb2 surfaces facilitate Swe1 recognition; G1 Cln–Cdc28 is comparatively resistant, allowing Start while mitotic CDK remains checkpoint-sensitive | Comparative cyclin–CDK biochemistry, cyclin mutants, interaction mapping, and cell-cycle assays | Asano et al., 2005, 10.1038/sj.emboj.7600683 (asano2005concertedmechanismof pages 3-3, asano2005concertedmechanismof pages 8-9) | High for preferential mitotic-complex regulation; exact quantitative specificity depends on assay context |
| Swe1 also inhibits Clb2–Cdc28 by direct binding, partly independently of Tyr19 phosphorylation | Stable Swe1 association can suppress Clb2–Cdc28 activity even when phosphotyrosine-dependent inhibition is impaired | Co-complex/kinase assays and genetics using phosphorylation-resistant Cdc28 | Mechanistic synthesis in Asano et al., 2005, 10.1038/sj.emboj.7600683 (asano2005concertedmechanismof pages 1-1) | Moderate–high; experimentally supported, but Tyr19 phosphorylation is the canonical and dominant annotation |
| Swe1 localization is dynamic: S-phase nucleus → cytoplasm/daughter-side bud neck → degradation | Nuclear Swe1 encounters Cdc28; subsequent recruitment to the daughter side of the septin collar places it in a spatially organized inactivation module | Fluorescence localization, synchronized-cell time courses, and localization-mutant analysis | Longtine et al., June 2000, 10.1128/MCB.20.11.4049-4061.2000; Lee et al., September 2005, 10.4161/cc.4.10.2049 (raspelli2013theproteinkinase pages 21-26, lee2005monitoringthecell pages 5-5) | High for nuclear and daughter-neck localization; exact sequence of trafficking and phosphorylation events remains partly model-based |
| Septins–Hsl1–Hsl7 create the bud-neck platform that promotes Swe1 downregulation | Proper septin assembly activates/localizes Hsl1; Hsl7 acts as an adaptor recruiting Swe1. Budding, actin, or septin defects disrupt this module, stabilizing active Swe1 | Localization microscopy, deletion genetics, phospho-state analysis, and epistasis | Shulewitz et al., October 1999, 10.1128/MCB.19.10.7123; McMillan et al., October 1999, 10.1128/MCB.19.10.6929 (shulewitz1999hsl7localizesto pages 13-14, mcmillan1999themorphogenesischeckpoint pages 10-11) | High; Hsl1/Hsl7 are central spatial regulators, although related neck kinases provide partial redundancy |
| Clb2–Cdc28 and Cdc5 sequentially phosphorylate Swe1 to promote its inactivation and turnover | Clb2–Cdc28 primes Swe1, enhancing interaction with the Cdc5 polo-box domain; bud-neck Cdc5 then hyperphosphorylates Swe1, facilitating ubiquitylation/proteasomal degradation | In vitro phosphorylation, polo-box binding, kinase/localization mutants, overexpression, and degradation time courses | Asano et al., 26 May 2005, 10.1038/sj.emboj.7600683 (asano2005concertedmechanismof pages 1-3, asano2005concertedmechanismof pages 3-3, asano2005concertedmechanismof pages 8-9) | High for sequential phosphorylation and Cdc5-dependent turnover; the responsible ubiquitin machinery may vary with cell-cycle context |
| Mih1 antagonizes Swe1 | Mih1, the budding-yeast Cdc25-family phosphatase, removes inhibitory phosphate from Cdc28 Tyr19, restoring mitotic CDK activity | Genetic dosage/epistasis and Cdc28 phosphotyrosine measurements | Sia et al., November 1996, 10.1091/mbc.7.11.1657; Shulewitz et al., October 1999, 10.1128/MCB.19.10.7123 (shulewitz1999hsl7localizesto pages 13-14) | High; Swe1–Mih1 opposition is the core Cdc28-Y19 control module |
| Swe1 is the principal morphogenesis-checkpoint effector | When bud formation or cytoskeletal organization fails, stabilized Swe1 lowers mitotic Clb–Cdc28 activity, delays nuclear division, and prolongs polarized growth; successful budding permits Swe1 destruction and isotropic bud growth | Bud-defective and actin-perturbed mutants, SWE1 deletion/overexpression, CDC28-Y19 mutants, and cell-morphology assays | Sia et al., 1996, 10.1091/mbc.7.11.1657; McMillan et al., 1999, 10.1128/MCB.19.10.6929 (mcmillan1999themorphogenesischeckpoint pages 10-11, galli2017yeasthaspinkinase pages 20-26) | High; the checkpoint delays rather than invariably arrests mitosis, and additional pathways can contribute under stress |
| Representative quantitative observations support a highly phosphorylated, spatially controlled switch | Approximately 35–40 Swe1 phosphorylation sites have been detected in vivo. Under nocodazole arrest, Cdc5 bud-neck localization was about 42% in wild type, 12% in hsl1Δ, and 23% in hsl7Δ; a localization-competent EGFP-Cdc5 construct reached about 98% in wild type | Phosphosite mapping and quantitative fluorescence microscopy | Asano et al., 2005, 10.1038/sj.emboj.7600683 (raspelli2013theproteinkinase pages 21-26, asano2005concertedmechanismof pages 3-3) | Moderate–high; values are experiment- and construct-specific and should not be treated as universal physiological constants |
| 2023–2024 target-specific mechanistic advances are scarce | Recent S. cerevisiae studies generally use the established Swe1–Cdc28 morphogenesis-checkpoint framework rather than redefining Swe1 catalysis, specificity, or localization | Literature search and comparison with foundational primary studies | The current mechanistic model remains anchored in the 1993–2017 primary literature summarized above (asano2005concertedmechanismof pages 1-1, lee2005monitoringthecell pages 5-5) | High as a literature-scope conclusion; recent work on similarly named Swe1 proteins in other fungi cannot be transferred directly to P32944 |
Table: This table links the principal functional annotations of S. cerevisiae Swe1/P32944 to their molecular mechanisms and strongest supporting evidence. It also separates well-established conclusions from model-dependent details and notes the limited target-specific advances reported in 2023–2024.
The primary reaction can be represented as:
ATP + Cdc28–Tyr19 → ADP + Cdc28–phospho-Tyr19
Tyr19 lies in the ATP-binding region of Cdc28/Cdk1. Its phosphorylation reduces the activity of mitotic Clb–Cdc28 complexes and thereby delays entry into mitosis. The opposing Cdc25-family phosphatase Mih1 removes this inhibitory phosphate, so the physiological state of Cdc28 is determined by the balance between Swe1 and Mih1. Genetic dosage experiments showed that nuclear-division timing in unbudded cells is especially sensitive to SWE1 and MIH1, supporting a causal kinase–phosphatase switch rather than a simple correlation. Sia, Herald & Lew, November 1996, DOI 10.1091/mbc.7.11.1657. (shulewitz1999hsl7localizesto pages 13-14, galli2017yeasthaspinkinase pages 20-26)
The best-established physiological substrate is Cdc28/Cdk1, with inhibitory phosphorylation focused on Tyr19. Swe1 preferentially controls Cdc28 associated with mitotic B-type cyclins, particularly Clb2–Cdc28, rather than uniformly inhibiting every Cdc28 complex. This selectivity allows G1/S cell-cycle functions to proceed while making mitotic activation conditional on successful morphogenesis. Mitotic-cyclin depletion strongly reduces Swe1 phosphorylation and turnover, whereas loss of S-phase Clb5/Clb6 causes only an approximately 10-minute delay, attributable largely to delayed S-phase progression. Clb2 also has a specific role in Cdc5 recruitment to the neck and in the feedback that eliminates Swe1. (asano2005concertedmechanismof pages 3-3, asano2005concertedmechanismof pages 8-9)
Swe1 regulation of Clb2–Cdc28 is not limited to phosphotransfer. Direct association can inhibit the complex even when Tyr19-dependent inhibition is impaired, establishing a secondary, phosphorylation-independent mechanism. Nevertheless, Tyr19 phosphorylation remains the canonical and best-supported basis for functional annotation. Asano et al., published online 26 May 2005, DOI 10.1038/sj.emboj.7600683. (asano2005concertedmechanismof pages 1-1)
Budding yeast must coordinate nuclear division with formation of a daughter compartment. When bud emergence, actin organization, or septin assembly is defective, Swe1 remains active and inhibits mitotic Clb–Cdc28. The resulting delay provides time to correct morphogenesis before chromosome segregation. SWE1 is required for the normal delay in nuclear division in cells unable to form a bud, whereas budded cells are much less dependent on Cdc28 Tyr19 phosphorylation for ordinary cell-cycle timing. This distinction is why the pathway is termed the morphogenesis checkpoint or, more cautiously, morphogenesis surveillance. Sia et al., November 1996. (galli2017yeasthaspinkinase pages 20-26)
Reduced Clb–Cdc28 activity also postpones the transition from apical or polarized bud growth to isotropic growth. Consequently, stabilized or excessive Swe1 produces elongated buds or cells. Conversely, eliminating Swe1 can permit nuclear division despite defective budding or actin organization, risking division without an adequate daughter compartment and formation of abnormal or binucleate cells. Thus, the elongated morphology is mechanistically informative: it reflects prolonged polarized growth caused by delayed mitotic CDK activation, rather than a direct structural role for Swe1 in the cell wall or cytoskeleton. (raspelli2013theproteinkinase pages 21-26, shulewitz1999hsl7localizesto pages 13-14, galli2017yeasthaspinkinase pages 20-26)
Swe1 is an intracellular, dynamically localized signaling enzyme rather than a membrane, secreted, or structural protein.
Accordingly, Swe1’s catalytic action on Cdc28 can occur in the nucleus and likely other intracellular compartments, whereas a major portion of its own downregulation occurs at the septin-organized bud neck. The exact trafficking sequence and compartment in which every phosphorylation occurs remain partly model-dependent, but nuclear accumulation and asymmetric daughter-neck localization are supported experimentally. (raspelli2013theproteinkinase pages 21-26, lee2005monitoringthecell pages 5-5)
The septin collar acts as a signaling scaffold. Correct bud and septin assembly permits the Nim1-related kinase Hsl1 to localize and become active at the neck. Hsl7 serves as an adaptor that recruits Swe1 to this platform. Loss of HSL1 or HSL7 disrupts efficient Swe1 neck localization, reduces its normal phosphorylation, stabilizes Swe1, increases Cdc28 Tyr19 phosphorylation, and produces a transient G2/M delay with elongated buds. Partial redundancy from related neck-associated kinases helps explain why these deletions delay rather than permanently arrest the cycle. Shulewitz, Inouye & Thorner, October 1999, DOI 10.1128/MCB.19.10.7123; McMillan et al., October 1999, DOI 10.1128/MCB.19.10.6929. (shulewitz1999hsl7localizesto pages 13-14, mcmillan1999themorphogenesischeckpoint pages 10-11)
As mitotic cyclin activity rises, Clb2–Cdc28 phosphorylates Swe1. This priming promotes binding of Swe1 to the polo-box domain of the Polo-like kinase Cdc5. Hsl1–Hsl7 and the bud-neck scaffold facilitate Cdc5 recruitment, after which Cdc5 further phosphorylates Swe1. Hyperphosphorylation promotes Swe1 downregulation and degradation, creating a feedback circuit in which rising mitotic CDK activity helps eliminate its inhibitor. Asano et al., 26 May 2005. (asano2005concertedmechanismof pages 1-3, asano2005concertedmechanismof pages 3-3, asano2005concertedmechanismof pages 8-9)
This mechanism explains how morphogenesis and cell-cycle progression are integrated: septin organization supplies spatial permission, while increasing Clb2–Cdc28 and Cdc5 activities provide temporal permission. Swe1 is therefore regulated by a coincidence-detection system rather than a single linear on/off reaction. (asano2005concertedmechanismof pages 1-3, lee2005monitoringthecell pages 5-5)
Hyperphosphorylated Swe1 is ubiquitylated and degraded by the proteasome. SCF-associated mechanisms, including reported Met30 involvement, and cell-cycle-dependent APC-related turnover have both been implicated. However, the relative contribution of individual ubiquitin ligases varies by cell-cycle context and has historically been less settled than the upstream Hsl1/Hsl7–Cdc28–Cdc5 phosphorylation pathway. It is therefore safest to annotate phosphorylation-coupled ubiquitin/proteasome turnover as established while treating a single universal E3 assignment as less certain. McMillan et al., October 2002, DOI 10.1091/mbc.e02-05-0283. (mcmillan2002determinantsofswe1p pages 16-16, asano2005concertedmechanismof pages 8-9, raspelli2013theproteinkinase pages 21-26)
Swe1 also participates in responses to interrupted DNA synthesis and genotoxic stress, where inhibition of mitotic CDK can reinforce delayed chromosome segregation. In budding yeast this branch is partly redundant with Mec1/Rad53-dependent controls, explaining why Swe1 may contribute substantially without always being strictly required. Its well-established meiotic role similarly uses inhibitory Cdc28 phosphorylation to help maintain checkpoint-dependent prophase delay. These contexts reuse the same biochemical activity—control of Cdc28 Tyr19—rather than establishing a different primary enzymatic function.
Reports also connect Swe1 to spindle behavior and age-dependent spindle-pole-body segregation. Such findings broaden the physiological consequences of the Swe1–Cdc28 module, but they should remain secondary annotations: the direct and most reproducible molecular function is still Cdc28 inhibition, and some spindle phenotypes may arise downstream from changed CDK timing rather than from Swe1 acting as a spindle structural component.
Several useful measurements illustrate the architecture of the pathway:
A targeted search found little 2023–2024 primary work that materially revises the biochemical annotation of S. cerevisiae Swe1/P32944. Recent budding-yeast studies continue to invoke the established Swe1–Cdc28 morphogenesis-checkpoint framework, while the decisive mechanistic evidence remains concentrated in the foundational 1993–2017 literature. This is not evidence that the annotation is obsolete; rather, it reflects a mature mechanism whose core claims—Cdc28 Tyr19 phosphorylation, mitotic-cyclin preference, septin/Hsl1/Hsl7-dependent spatial regulation, Cdc5-assisted turnover, and Mih1 antagonism—are strongly established.
Recent papers on proteins named Swe1 in pathogenic fungi were not used to assign new substrates or pathways to P32944. For example, organism-specific claims involving cell-wall-integrity MAPKs or autophagy proteins in other fungi cannot be transferred to S. cerevisiae Swe1 without direct evidence.
Recommended primary annotation:
Swe1 is the S. cerevisiae Wee1-family mitosis-inhibitory protein kinase that phosphorylates Cdc28/Cdk1 on Tyr19, preferentially regulating mitotic Clb–Cdc28 complexes. Through this reaction—and a secondary direct-binding inhibitory mechanism—it delays mitotic entry and the transition from polarized to isotropic bud growth when bud formation or cytoskeletal organization is defective.
Recommended process annotations: morphogenesis checkpoint; negative regulation of mitotic entry; inhibitory phosphorylation of Cdc28; coordination of bud growth with nuclear division; phosphorylation-coupled protein degradation; and secondary roles in replication-stress and meiotic checkpoint responses.
Recommended localization annotation: nucleus during S phase; cytoplasm; and daughter side of the septin-associated mother–bud neck, where Hsl1/Hsl7- and Cdc5-dependent regulation promotes Swe1 inactivation and degradation.
Evidence confidence: high for identity, Cdc28 Tyr19 phosphorylation, morphogenesis-checkpoint function, Mih1 antagonism, and bud-neck regulation; moderate for the universal identity of the responsible ubiquitin ligase and for assigning downstream spindle phenotypes as direct rather than CDK-mediated effects.
References
(asano2005concertedmechanismof pages 1-1): Satoshi Asano, Jung-Eun Park, Krisada Sakchaisri, Li-Rong Yu, Sukgil Song, Porntip Supavilai, Timothy D Veenstra, and Kyung S Lee. Concerted mechanism of swe1/wee1 regulation by multiple kinases in budding yeast. The EMBO Journal, 24:2194-2204, Jun 2005. URL: https://doi.org/10.1038/sj.emboj.7600683, doi:10.1038/sj.emboj.7600683. This article has 153 citations.
(shulewitz1999hsl7localizesto pages 13-14): Mark J. Shulewitz, Carla J. Inouye, and Jeremy Thorner. Hsl7 localizes to a septin ring and serves as an adapter in a regulatory pathway that relieves tyrosine phosphorylation of cdc28 protein kinase insaccharomyces cerevisiae. Molecular and Cellular Biology, 19:7123-7137, Oct 1999. URL: https://doi.org/10.1128/mcb.19.10.7123, doi:10.1128/mcb.19.10.7123. This article has 240 citations and is from a domain leading peer-reviewed journal.
(raspelli2013theproteinkinase pages 21-26): E Raspelli. The protein kinase swe1: new players in its regulatory pathway and analysis of its involvement in mitotic spindle dynamics. Unknown journal, 2013.
(asano2005concertedmechanismof pages 1-3): Satoshi Asano, Jung-Eun Park, Krisada Sakchaisri, Li-Rong Yu, Sukgil Song, Porntip Supavilai, Timothy D Veenstra, and Kyung S Lee. Concerted mechanism of swe1/wee1 regulation by multiple kinases in budding yeast. The EMBO Journal, 24:2194-2204, Jun 2005. URL: https://doi.org/10.1038/sj.emboj.7600683, doi:10.1038/sj.emboj.7600683. This article has 153 citations.
(asano2005concertedmechanismof pages 3-3): Satoshi Asano, Jung-Eun Park, Krisada Sakchaisri, Li-Rong Yu, Sukgil Song, Porntip Supavilai, Timothy D Veenstra, and Kyung S Lee. Concerted mechanism of swe1/wee1 regulation by multiple kinases in budding yeast. The EMBO Journal, 24:2194-2204, Jun 2005. URL: https://doi.org/10.1038/sj.emboj.7600683, doi:10.1038/sj.emboj.7600683. This article has 153 citations.
(asano2005concertedmechanismof pages 8-9): Satoshi Asano, Jung-Eun Park, Krisada Sakchaisri, Li-Rong Yu, Sukgil Song, Porntip Supavilai, Timothy D Veenstra, and Kyung S Lee. Concerted mechanism of swe1/wee1 regulation by multiple kinases in budding yeast. The EMBO Journal, 24:2194-2204, Jun 2005. URL: https://doi.org/10.1038/sj.emboj.7600683, doi:10.1038/sj.emboj.7600683. This article has 153 citations.
(lee2005monitoringthecell pages 5-5): Kyung S. Lee, Satoshi Asano, Jung-Eun Park, Krisada Sakchaisri, and Raymond L. Erikson. Monitoring the cell cycle by multi-kinase-dependent regulation of swe1/wee1 in budding yeast. Cell Cycle, 4:1346-1349, Sep 2005. URL: https://doi.org/10.4161/cc.4.10.2049, doi:10.4161/cc.4.10.2049. This article has 32 citations and is from a peer-reviewed journal.
(mcmillan1999themorphogenesischeckpoint pages 10-11): John N. McMillan, Mark S. Longtine, Rey A. L. Sia, Chandra L. Theesfeld, Elaine S. G. Bardes, John R. Pringle, and Daniel J. Lew. The morphogenesis checkpoint in saccharomyces cerevisiae: cell cycle control of swe1p degradation by hsl1p and hsl7p. Molecular and Cellular Biology, 19:6929-6939, Oct 1999. URL: https://doi.org/10.1128/mcb.19.10.6929, doi:10.1128/mcb.19.10.6929. This article has 231 citations and is from a domain leading peer-reviewed journal.
(galli2017yeasthaspinkinase pages 20-26): MARTINA GALLI. Yeast haspin kinase regulates mitotic cell cycle events: from g2/m transition to polarisome dispersion. Text, Dec 2017. URL: https://doi.org/10.13130/galli-martina_phd2017-12-01, doi:10.13130/galli-martina_phd2017-12-01. This article has 0 citations and is from a peer-reviewed journal.
(mcmillan2002determinantsofswe1p pages 16-16): John N. McMillan, Chandra L. Theesfeld, Jacob C. Harrison, Elaine S. G. Bardes, and Daniel J. Lew. Determinants of swe1p degradation in saccharomyces cerevisiae. Molecular biology of the cell, 13 10:3560-75, Oct 2002. URL: https://doi.org/10.1091/mbc.e02-05-0283, doi:10.1091/mbc.e02-05-0283. This article has 109 citations and is from a domain leading peer-reviewed journal.