INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

Ubiquitin-mediated protein degradation is the major nonlysosomal proteolytic pathway. Ubiquitin, a 76-amino acid polypeptide that is highly conserved in eukaryotes, marks proteins for degradation by the 26 S proteasome (1). This system is involved in a wide variety of cellular processes including DNA repair, cell cycle progression, signal transduction, and antigen presentation (2). Ubiquitin-mediated protein degradation is known to account for the turnover of cyclins, cyclin-dependent kinase inhibitors, p53, c-Jun, c-Fos, and other oncoproteins (reviewed in Ref. 3).

Although the 26 S proteasome is able to degrade proteins involved in diverse processes, the mechanism is highly specific. Protein degradation is usually brought about by the attachment of polyubiquitin moieties to a lysine residue on the target protein. This acts as a sorting signal targeting conjugated proteins to the 26 S proteasome, which then degrades the polyubiquitinated protein into small peptides (reviewed in Ref. 2 and 4).

The 26 S proteasome consists of two multiprotein subcomplexes, the 20 S core and the 19 S cap. The barrel-shaped 20 S enzymatic core consists of four stacked rings each consisting of seven polypeptides (5). The 19 S regulatory cap complex comprises at least 20 different subunits, which fall into two classes, ATPases and non-ATPases (6). In the presence of ATP, the 19 S cap binds to each end of the 20 S core and confers ATP dependence and specificity for ubiquitinated substrates on the 26 S complex (7). The budding yeast SUG1 gene (8) and the fission yeast mts2⁺ gene (9) are known to encode ATPase components of the 19 S cap, while the fission yeast mts3⁺ and mts4⁺ genes encode non-ATPase subunits (10, 11). At the restrictive temperature the phenotypes of the temperature-sensitive (ts)¹ mts2-1 and mts3-1 strains are transient cell cycle arrest at metaphase, indicating that the metaphase to anaphase transition is blocked (9, 10). This is similar to ts sug1-1 mutants that are unable to segregate their DNA and arrest after replication with an anucleate bud at the restrictive temperature (13).

Previously, mutations in the mts2⁺ (9), mts3⁺ (10) and mts4⁺ (11) genes were isolated in a screen for fission yeast mutants that are both resistant to the microtubule destabilizing drug methyl 2-benzimidazolecarbamate (MBC^R) and temperature sensitive (ts) for growth. In this communication we describe the cloning and characterization of a fourth mutant identified in the same screen, mts5-1. This mutation is rescued by a cDNA clone containing the pad1⁺ gene. pad1⁺ was originally isolated by its ability to confer staurosporine resistance on wild type cells when present on a multicopy plasmid (12). Further analysis showed that the pad1⁺ gene isolated in this screen was truncated, lacking 29 amino acids from the carboxyl-terminal end. Pad1 was originally described as a positive regulator of the Schizosaccharomyces pombe transcription factor Pap1 since overexpression of the truncated pad1⁺ gene caused an increase in Pap1-dependent transcription (12). Recently cloning of the human homologue of Pad1, Poh1, was reported (14). Poh1 has been shown to encode a subunit of the human 26 S proteasome. In this communication we demonstrate that Pad1 is a subunit of the 26 S proteasome in fission yeast and that the mts5-1/pad1-1 strain is defective in the degradation of ubiquitin conjugates. Furthermore we describe genetic interactions between the mts5⁺/pad1⁺ gene and two other genes encoding non-ATPase subunits of the 19 S cap, mts3⁺ and mts4⁺. In addition we have isolated the mouse Pad1 homologue which can rescue the ts phenotype of mts5-1/pad1-1, illustrating the highly conserved nature of the 26 S proteasome between eukaryotes.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Yeast Strains-- The S. pombe strains described in this report were all derived from the 972h and 975h⁺ wild type heterothallic strains. mts2-1(mts2leu1-32h) (9), mts3-1(mts3leu1-32h) (10), mts4-1(mts4leu1-32h) (11), mts5-1(mts5leu1-32h) (this study), the pad1⁺HA strain was made by adding a NotI site to the 3' end of pad1 just before the stop codon using nmt-84 5'-oligo CCTGGCATATCATCAATT and 3'-oligo AAGGCCTAGCGGCCGCAGAAGGCAACGGAATCAAGCATACAAGC. The resulting polymerase chain reaction product was cut with XhoI/SmaI and cloned into pREP3X. A 111-base pair 3xHA NotI fragment cut from pBSK-3xHA was cloned into the NotI site of pREP3Xpad1⁺ to create plasmid pREPpad1HA which has a HA-tagged pad1⁺ gene under the control of the nmt1 promoter. This construct was shown to complement the ts phenotype of pad1-1. Finally a 909-base pair EcoRI/ScaI fragment was excised from pREPpad1HA (the whole 5' end of the pad1⁺ tagged cDNA) and cloned into pUC19ura4SphI(pUC19 carrying ura4⁺ at the SphI site). The construct was linearized using the remaining SmaI site and transformed into SP813(leu1.32ade6-210ura4D-18) and ura4⁺ transformants selected. This created strain pad1⁺HA(leu1.32ade6-210ura4D-18pad1⁺::HA) which has a HA-tagged pad1⁺ gene under the control of the endogenous promoter.

Media and Genetic Methods-- Media and general method for handling S. pombe are as described elsewhere (15). Plates containing MBC consisted of YPD with the MBC added after autoclaving.

Plasmids-- The S. pombe library used in this work was a wild type cDNA library cloned in to the S. pombe expression vector pREP1 (a gift from Chris Norbury). The plasmid pREP1 contains the S. pombe thiamine-repressible nmt1⁺ promoter and was used for routine cloning in S. pombe (16, 17). The integrating vector pJK210 (18) was used for integrating cloned S. pombe cDNA.

Fluorescence Microscopy-- For immunofluorescence microscopy cells were grown overnight in YPD (0.5% yeast extract, 0.5% peptone, and 2% glucose) at 25 °C then shifted to 35 °C for subsequent samplings. Immediately before fixation, sorbitol was added to the sample to give a final concentration of 1.2 M. The culture was then fixed in 38% paraformaldehyde at 25 °C, or 35 °C for ts mutants, for 30 min. The cells were then prepared as described previously (19). The anti-tubulin antibody TAT1 was used to stain microtubules (20), and 4',6-diamidoino-2-phenylindole was added at 1 µg/ml in Vectashield (Vector H-1000) directly to cells on the slide. Stained cells were observed on a Zeiss Axioplan microscope. Images were processed using IPLab software and Digital Smart capture.

SDS-PAGE and Immunoblotting-- SDS-PAGE was performed using standard protocols. Separated proteins were stained with Coomassie Brilliant Blue or transferred to nitrocellulose (ECL Hybond, Amersham Pharmacia Biotech). Western blot analysis was carried out as detailed elsewhere (9). Anti-ubiquitin (DAKO Z0658) and anti-actin antibodies (Amersham N350) were used at 1:2000.

Flow Cytometry-- Flow cytometry was carried out using a Becton Dickinson FACScan using the method described previously (10, 21).

Co-purification of Pad1 with the 26 S Proteasome-- S. pombe cell extracts were prepared from the pad1⁺::HA strain as described previously (22). 5 liters of cells were grown in yeast extract media to an A₅₉₅ of approximately 1.0. Cell extracts were fractionated by DEAE-cellulose chromatography (Whatman) and by glycerol gradient centrifugation. Specific peptidase activity, as measured by the cleavage of the fluorogenic peptide succinimidyl-Leu-Leu-Val-Tyr-7-amino-4-methylcoumarin, was determined in the presence of ATP for the fractions collected from glycerol gradient centrifugation, as described previously (22). Each lane in the blot contains 7.5 µg of protein. The antisera used in each blot are anti-HA (Boehringer 1583816), anti-Mts4 (11), and antisera raised against the 20 S subcomplex of the 26 S proteasome (22).

Immunoprecipitations-- S. pombe extracts were prepared as described previously (14) from the S. pombe strain containing the HA-tagged version of the pad1⁺ gene, in the following buffer: 10 mM Tris, pH 7.5, 50 mM NaCl, 10% glycerol, 0.5 mM phenylmethylsulfonyl fluoride. 2-mg extracts were incubated with either 4 µl of anti-Mts4 antisera or 4 µl of preimmune sera in a total volume of 1 ml for 2 h at 4 °C. 100 µl of a 50% slurry of protein A-Sepharose (Amersham Pharmacia Biotech) was added, and the extract was incubated for a further 2 h at 4 °C. After five washes in the extract buffer, the proteins bound to the antibodies were released by boiling in SDS loading buffer.

Isolation of Mouse Pad1-- A Stratagene Uni-ZAP mouse testis cDNA library was screened using standard protocols with 50 ng of full-length S. pombe pad1⁺ DNA randomly primed with [³²P]dCTP. Six single colony positive plaques were isolated. One clone was sequenced on both strands using vector oligos and was found to be homologous to S. pombe pad1⁺.

DNA Sequencing-- Automated fluorescent sequencing was performed using an Applied Biosystems automated sequencing machine 373A, according to the manufacturer's instructions. Double-stranded DNA was labeled using an ABI Dydeoxyterminator cycle sequencing kit (ABI 401113). Data were analyzed using Applied Biosystems 373A software.

Genomic DNA Extraction from S. pombe-- Cells were grown to an A₅₉₅ of 1.0 and resuspended in 0.2 volume extraction buffer (100 mM Tris-HCl, pH 8.0, 2% Triton, 1% SDS). RNase was added to a final concentration of 5 µg/ml, and cells were incubated at 37 °C for 1 h. 2 volumes of phenol and 0.5 volumes of acid-washed glass beads were added, and cells were vortexed for 2 min. Cell debris was spun down, and the upper phase was cleaned using a standard phenol chloroform extraction and ethanol precipitation protocol.

Assaying Ubiquitin-¹²⁵I-Lysozyme Conjugate Degradation of 26 S Proteasomes-- The assay was performed essentially as described previously (22) on 26 S preparations prepared from wild type cells and pad1-1 cells grown at the permissive temperature of 25 °C. Aliquots of glycerol gradient fractions 4 and 5, containing the peak in 26 S proteolytic activity, were incubated at 37 °C in the presence of ubiquitin conjugates with or without ATP in a final volume of 150 µl. Rates of conjugate degradation are expressed as percentage degradation per 60 min (37 °C) in the presence of 0.2 mg of protein per sample (22).

Preparation of Protein Extracts from Fission Yeast-- Cells were harvested and resuspended in 100 µl of cell lysis buffer (10 mM Tris, pH 7.8, 10% glycerol, 50 mM NaCl, 0.1% Triton, protease inhibitor Complete Tablet (Bohringer 1697498), 0.5 mM phenylmethylsulfonyl fluoride). Acid-washed glass beads (0.5 mm, Sigma G-9268) were added to just below the meniscus of the liquid. Cells were broken by vigorous vortexing for 30 s followed by 30 s on ice, for 15 min. Cells were pelleted by centrifugation at high speed for 5 min at 4 °C. The supernatant was clarified by further centrifugation for 15 min at 4 °C. Supernatants were transferred to a clean Eppendorf tube and stored at 20 °C.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

mts5-1 Is Rescued by a cDNA Clone Containing the pad1⁺ Gene-- A S. pombe cDNA library was screened to isolate clones capable of rescuing the mts5-1 ts phenotype. One clone was identified and found to contain the pad1⁺ gene and was shown to be the authentic gene by homologous integration (Fig. 1). From now on we will refer to mts5-1 as pad1-1. The mutation in the pad1-1 allele was identified by sequencing. Genomic DNA was isolated from pad1-1 and the pad1 ts allele was amplified by polymerase chain reaction. The mutation was found to be a G to A transition at position 160. This changes the corresponding codon from GGT to AGT, substituting serine for glycine (Fig. 1).

View larger version (97K): [in this window] [in a new window]   Fig. 1.   Comparison of the protein sequence of S. pombe Pad1 (PAD1) with mouse Pad1 (MAD1) (Y13071), S. cerevisiae Mpr1 (MPR1) (28), human JAB1 (26), and human Poh1 (POH1) (U86782) (14). Dotted lines represent gaps inserted to achieve maximum alignment. Identical residues are shown in reverse font. Note that Pad1, mPad1, Mpr1, and Poh1 contain more identical amino acids than JAB1. The star at position 160 denotes the position of the glycine to serine substitution in pad1-1. The position within the protein sequence is shown on the left- and right-hand sides. The mouse pad1+ gene was isolated from a cDNA library using the fission yeast pad1+ gene as a probe.

pad1-1 Arrests with the Same Phenotype as the 26 S Proteasome Mutants mts2-1 and mts3-1-- In pad1-1 16% of cells arrest transiently at metaphase with a short mitotic spindle and condensed DNA after 4 h of growth at the restrictive temperature, consistent with a metaphase arrest (Figs. 2 and 3). With further incubation at the restrictive temperature the number of cells at metaphase decreases, while the number of cells with a septated phenotype, elongated septate cell with decondensed DNA in one daughter cell only, increases (Fig. 3). After 24 h of incubation at the restrictive temperature, no metaphase cells are present and 95% of the cells display the septated phenotype. FACS analysis demonstrates that cells arrest with a G₂ content of DNA (Fig. 4). This is similar to the 26 S proteasome mutant mts2-1 isolated in the same screen. This mutant arrests transiently at metaphase after 4 h of growth at the restrictive temperature with a G₂ content of DNA and forms septated cells with DNA in one daughter cell after extended incubation at the restrictive temperature. These observations coupled with the MBC^R phenotype, indicated that Pad1 might be a subunit of the 26 S proteasome.

View larger version (125K): [in this window] [in a new window]   Fig. 2.   Immunofluorescence microscopy of the pad1-1 mutant at the restrictive temperature. A field of cells is shown. Cells arrest transiently at metaphase after 4 h of growth at the restrictive temperature with a short mitotic spindle and condensed DNA. Cells were fixed with paraformaldehyde after 4 h of growth at 35 °C. The fixed cells were stained with the anti-tubulin antibody TAT1 (20) shown in red, and the DNA specific stain 4',6-diamidoino-2-phenylindole shown in blue.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Percentage of pad1-1 cells showing a metaphase arrest and septated phenotype. A pad1-1 culture was grown at 25 °C to mid log phase and shifted to 35 °C. A, samples were taken over a course and fixed for fluorescence microscopy. The number of cells in metaphase, short mitotic spindle, and condensed DNA was counted (black circles) along with the number of cells with a septated phenotype (as shown in B) (black squares). In a wild type culture grown under identical conditions there were 0.5-1% metaphase cells at each time point, and no cells with the septated phenotype were observed.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   FACS analysis of the pad1-1 mutant. Cells arrest with a G2 content of DNA consistent with metaphase arrest. Flow cytometry was carried out on an exponentially growing wild type diploid culture (A), an exponentially growing haploid wild type culture (B), and a nitrogen-starved wild type haploid culture (C) to determine the position of the 4n, 2n, and 1n peaks, respectively. Flow cytometry was performed on asynchronous cultures of pad1-1, mts2-1, and a wild type cells shifted from the permissive to the restrictive temperature. The culture was sampled at the times shown, and the DNA content was determined by flow cytometry on a Becton Dickinson FACScan. pad1-1 and mts2-1 arrest with a 2n content of DNA. The graphs broaden with increasing time at the restrictive temperature due to an increase in aberrant elongated cells.

Pad1 Co-purifies with Known Subunits of the 26 S Proteasome-- To determine whether Pad1 is a subunit of the 26 S proteasome in fission yeast, the 26 S proteasome was purified from wild type S. pombe cells by anion exchange chromatography and glycerol gradient centrifugation as described previously for other fission yeast 26 S proteasome subunits (22). Using anti-HA antibodies against a HA-tagged pad1⁺, Pad1 was found to co-purify with the fractions containing the highest levels of 26 S proteasome activity. Furthermore, these are the fractions where known subunits of the 26 S proteasome are found (22) (Fig. 5A).

View larger version (19K): [in this window] [in a new window]   Fig. 5.   A, co-purification of Pad1 with the fission yeast 26 S proteasome complex. Specific peptidase activity, as measured by the cleavage of the fluorogenic peptide succinimidyl-Leu-Leu-Val-Tyr-7-amino-4-methylcoumarin, was determined in the presence of ATP for the fractions collected from the glycerol gradient centrifugation. This was performed using pooled fractions from DEAE chromatography that also contained this activity. Western blots of the same fractions, containing 7.5 µg of protein/lane, were probed with either anti-Mts4, anti-20 S, or anti-HA (against Pad1) antisera. The various antisera used in each blot is indicated, and molecular mass markers are shown in kilodaltons. B, co-immunoprecipitation of Mts4 and Pad1. Western blot of immunoprecipitation using preimmune sera (lane 1) and anti-Mts4 antisera. The immunoprecipitated fractions were separated on a 10% SDS-PAGE gel, transferred to nitrocellulose, and probed with either anti-Mts4 antisera (top panel) or anti-HA monoclonal antisera (bottom panel). Molecular mass markers are shown in kilodaltons. C, an increase in high molecular weight protein conjugates is seen in the temperature-sensitive pad1-1 and mts2-1 26 S proteasome mutants. Western analysis was performed on extracts made from pad1-1, mts2-1, and wild type cells at 25 and 35 °C (4 and 8 h) with an antibodies against ubiquitin (DAKO Z0658) and actin (Amersham N350), to act as a loading control. An increase in high molecular weight protein conjugates is seen in pad1-1 and mts2-1 when incubated at the restrictive temperature, 35 °C. A wild type culture grown under identical conditions contains no high molecular weight ubiquitinated proteins at 25 °C or 35 °C. pad1-1 and mts2-1 have low levels of ubiquitinated conjugates even at 25 °C indicating that the 26 S proteasome is not fully functional in these mutants even at the permissive temperature.

The pad1-1 Mutant Shows an Increase in High Molecular Weight Ubiquitinated Proteins at the Restrictive Temperature-- An increase in high molecular weight ubiquitin conjugates has previously been reported in mts2-1 after incubation at the restrictive temperature (9). A similar assay was undertaken to ascertain whether pad1-1 showed a similar increase, which would suggest it also had a proteolysis defect. Wild type, mts2-1 (9), and pad1-1 strains were grown overnight at 25 °C to mid log phase, and then shifted to the restrictive temperature of 35 °C. Cells were analyzed at 0, 4, and 8 h after the temperature shift, and proteins were isolated. Equal amounts of protein were run on an SDS-PAGE gel and electroblotted. Blots were probed with either anti-ubiquitin antibodies or anti-actin antibodies to act as a loading control (Fig. 5C). High molecular weight ubiquitinated protein conjugates are present in pad1-1 and mts2-1 at the permissive temperature, and with incubation at the restrictive temperature of 35 °C these levels increase significantly. A wild type culture grown under identical conditions has no detectable high molecular weight ubiquitinated conjugates at 25 or 35 °C. An increase in high molecular weight ubiquitinated proteins is consistent with a defect in 26 S proteasome mediated protein degradation (22). The low levels of high molecular weight ubiquitinated proteins observed at 25 °C indicate that the proteasome is defective even at this temperature.

Ubiquitin Conjugate Degradation Is Impaired in pad1-1-- Analysis of the degradation of ubiquitin conjugates provides a specific test for 26 S proteasome function (22). After growing cells at the permissive temperature ubiquitin-¹²⁵I-lysozyme conjugate degradation was measured in glycerol gradient fractions possessing the highest activities of the 26 S proteasome. The data are summarized in Table I. The results demonstrate that reduced levels of conjugate degradation are seen in the pad1-1 strain compared with wild type. Residual degradation rates of the pad1-1 26 S proteasomes can be explained by nonspecific cleavage of free lysozyme (22). These results are consistent with pad1-1 having a defect in the degradation of ubiquitin conjugates.

Overexpression of Mts3 and Mts4 Will Rescue pad1-1 at an Intermediate Temperature-- The pad1-1 strain will grow at the permissive temperature of 25 °C, but not at the restrictive temperature, 35 °C, or at an intermediate temperature of 32 °C. Plasmids expressing the non-ATPase 19 S cap subunits Mts3 (10) and Mts4 (11) under the control of the thiamine repressible nmt1 promoter were transformed into the pad1-1 strain and streaked onto selective medium with and without thiamine at 25, 30, 32, and 35 °C to test for full or partial rescue of the ts phenotype of pad1-1. Overexpression of Mts3 and Mts4 at 32 °C but not 35 °C will rescue the temperature sensitive phenotype of pad1-1 (Fig. 6). Overexpression of the ATPase 19 S cap subunit Mts2 (9) was unable to rescue the temperature sensitivity of pad1-1 at 32 or 35 °C (data not shown). In addition, pad1-1 was found to be synthetically lethal with mts3-1. Asci containing four viable, three viable, and two viable spores segregated in a ratio of 4, 18, and 1, respectively. In the tetrads containing three viable spores the missing one was deduced to be the pad-1mts3-1 double mutant. The same conclusion was drawn from the tetrads containing the two viable spores, which both gave rise to wild type colonies. The double mutant spores were examined microscopically and found to divide once or twice to give a maximum of four cells. An explanation for the partial rescue and the synthetic lethality results is that Pad1 interacts directly with Mts3 in the 19 S regulatory complex. The overexpression data suggests that Pad1 may also interact with Mts4 directly although the mts4-1 and pad1-1 alleles are not synthetically lethal.

View larger version (51K): [in this window] [in a new window]   Fig. 6.   Overexpression of cDNAs encoding the 19 S cap subunits Mts3 and Mts4 will rescue the temperature sensitive phenotype of pad1-1 at an intermediate temperature. pad1-1 was transformed with pREP1, mts3+ (pmts3), mts4+ (pmts4), and pad1+ (ppad1) cDNAs under the control of the thiamine repressible promoter in pREP1. Pad1-1 cells containing these plasmids were streaked to single colonies at the permissive (25 °C), restrictive (35 °C), and two intermediate (30 and 32 °C) temperatures.

A Mouse Pad1 Homologue Rescues the pad1-1 ts Phenotype-- pad1⁺ is essential for cell viability (12). Proteins with highly similar sequences exist in Saccharomyces cerevisiae and humans (Table II). We also isolated a mouse cDNA from a library using the S. pombe pad1⁺ gene as a probe. The mouse protein shares 68.1% identity with the S. pombe Pad1 protein over its whole length. In addition, the mouse pad1⁺ gene, when expressed from a S. pombe expression vector, can rescue the temperature sensitivity of pad1-1 (Fig. 7). This demonstrates that the mouse gene is the functional homologue of S. pombe pad1⁺ gene.

View larger version (49K): [in this window] [in a new window]   Fig. 7.   The cDNA encoding the mouse homologue of fission yeast Pad1 can rescue the pad1-1 strain at 35 °C. pad1-1 was transformed with the mouse pad1+ cDNA in pREP1 (mpad1+), S. pombe pad1+ cDNA in pREP1 (ppad1+), and pREP1 alone. pad1-1 cells containing these plasmids were then streaked to single colonies at the permissive (25 °C) and restrictive (35 °C) temperatures.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

In this study we provide biochemical and genetic evidence that the pad1⁺ gene encodes a subunit of the 19 S regulatory complex of the 26 S proteasome. First, we have shown that the Pad1 protein co-purifies with peak levels of 26 S proteasome activity after anion exchange chromatography followed by glycerol gradient centrifugation. Furthermore, the fractions containing the highest 26 S proteasome activity and Pad1 protein levels were also the fractions in which the Mts4 protein, subunit 2 of the 19 S complex of S. pombe (11) was present. Second, the Pad1 protein was present when the 19 S complex was immunoprecipitated from yeast crude extracts using an antibody against the Mts4 protein (11). Third, we show that polyubiquitin conjugates build up in the pad1-1 mutant strain, as has been shown for other 26 S proteasome mutants (9), and that the pad1-1 mutant strain is defective in ubiquitin-lysozyme conjugate degradation. Fourth, 26 S proteasome function in S. pombe is essential for cell growth, and loss of activity results in a characteristic phenotype, arrest at the metaphase stage of mitosis (9-11). The phenotype of the pad1-1 mutant strain at the restrictive temperature is arrest at the metaphase stage of mitosis consistent with a mutation in a 26 S proteasome subunit. Finally, we have shown a genetic interaction between the pad1-1 strain and the mts3-1 and mts4-1 19 S mutant strains. Taken together the data presented provides compelling evidence that the pad1⁺ gene encodes a subunit of the 19 S complex. The isolation of the human pad1⁺ homologue, Poh1, and its identification as a subunit of the human 26 S proteasome confirms our results (14).

The 19 S complex is thought to have at least three biochemical functions; the recognition of polyubiquitinated substrates (23); an isopeptidase activity, which cleaves ubiquitin from the substrate for it to be recycled (24); and an anti-chaperone activity to unfold the substrate protein and present it to the 20 S catalytic complex (25). We have assayed purified Pad1 protein expressed in Escherichia coli for polyubiquitin binding and isopeptidase activity, and preliminary results show no evidence for either activity.²

The original isolation of pad1⁺ was as a truncated gene on a multicopy plasmid which conferred staurosporine resistance on wild type cells (12). This can be explained by our identification of Pad1 as a 26 S proteasome subunit. A truncated pad1⁺ gene could produce an abnormal protein that might be expected to disrupt the function of the 26 S proteasome leading to staurosporine resistance, possibly due to decreased proteolysis of a protein required for staurosporine resistance. Such a mechanism could also explain the MBC^R observed in the mts2-1, mts3-1, mts4-1, and pad1-1 proteasome mutant strains and why the original genetic screen appears to have enriched for proteasome mutants. Pad1 was originally described as a positive regulator of an AP-1 transcription factor Pap1 since overexpression of Pad1 caused an increase in the levels of Pap1. However, in this communication we have described the identification of Pad1 as a novel subunit of the 26 S proteasome in fission yeast. We postulate that decreased degradation of Pap1 in proteasome mutants results in elevated transcription of genes required for resistance to MBC and staurosporine. Experiments to test this hypothesis are in progress.

We also show that pad1-1 interacts genetically with both mts3-1 and mts4-1. This evidence is based on two observations. First, overexpression of either mts3⁺ or mts4⁺ cDNAs will partially rescue pad1-1 lethality at an intermediate temperature. Second, mts3-1 and pad1-1 are synthetically lethal. Since the pad1⁺, mts3⁺, and mts4⁺ genes are not sequence homologues, and therefore presumably do not encode proteins of similar biochemical activities, we hypothesise that the partial rescue of the pad1-1 mutant could be due to stabilization of the heat labile Pad1 protein in the pad1-1 strain. This raises the possibility that the Pad1 protein and the Mts3 and Mts4 proteins interact directly in the 19S regulatory complex.

Previously, a human gene JAB1 which has limited homology to Pad1 over the N-terminal half of the protein (Fig. 1), was isolated (26). As the JAB1 gene could not rescue a pad1 deletion strain, however, it was not a functional homologue (26). The functional Pad1 mouse homologue that we have isolated and the recently described Poh1 human homologue both have greater sequence identity with Pad1 than JAB1 has (Table II) and both are functional homologues of pad1 mutants. Consistent with these observations JAB1 has recently been reported to be a subunit of a high molecular weight complex called the signalosome, distinct from the 26S proteasome and implicated in signal transduction (27). Although JAB1 cannot rescue a pad1 deletion strain, overexpression of JAB1 from a high copy number vector in wild type S. pombe cells was as efficient as pad1⁺ overexpression in conferring drug resistance (26). In addition, overexpression of both JAB1 and S. pombe Pad1 also causes c-Jun-dependent AP-1 transcriptional activity in mammalian cells, leading to the conclusion that Pad1 and JAB1 are co-activators of AP-1 transcription factors (26). In this study however, we have shown that Pad1 is a subunit of the 26 S proteasome in fission yeast, and therefore we offer an alternative mechanism for the above results. Overexpression of JAB1, like overexpression of the truncated Pad1 protein, somehow disrupts 26 S proteasome function and results in elevated levels of AP-1/Pap1, thus leading to drug resistance.