JBC PeproTech; Our Business is Cytokines!

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


     

Originally published In Press as doi:10.1074/jbc.M003923200 on July 7, 2000

J. Biol. Chem., Vol. 275, Issue 44, 34293-34305, November 3, 2000
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
275/44/34293    most recent
M003923200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nemoto, Y.
Right arrow Articles by Bankaitis, V. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nemoto, Y.
Right arrow Articles by Bankaitis, V. A.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Functional Characterization of a Mammalian Sac1 and Mutants Exhibiting Substrate-specific Defects in Phosphoinositide Phosphatase Activity*

Yasuo NemotoDagger §, Brian G. Kearns, Markus R. Wenk§||, Hong Chen§, Kensaku MoriDagger , James G. Alb Jr., Pietro De Camilli§**, and Vytas A. BankaitisDaggerDagger

From the Dagger  Brain Science Institute, The Institute of Physical and Chemical Research, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan, the § Howard Hughes Medical Institute and Department of Cell Biology, Yale University School of Medicine, New Haven, Connecticut 06510, and the  Department of Cell Biology, University of Alabama at Birmingham, Birmingham, Alabama 35294-0005

Received for publication, May 8, 2000, and in revised form, July 6, 2000


    ABSTRACT
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

The Saccharomyces cerevisiae SAC1 gene was identified via independent analyses of mutations that modulate yeast actin function and alleviate the essential requirement for phosphatidylinositol transfer protein (Sec14p) activity in Golgi secretory function. The SAC1 gene product (Sac1p) is an integral membrane protein of the endoplasmic reticulum and the Golgi complex. Sac1p shares primary sequence homology with a subfamily of cytosolic/peripheral membrane phosphoinositide phosphatases, the synaptojanins, and these Sac1 domains define novel phosphoinositide phosphatase modules. We now report the characterization of a rat counterpart of Sac1p. Rat Sac1 is a ubiquitously expressed 65-kDa integral membrane protein of the endoplasmic reticulum that is found at particularly high levels in cerebellar Purkinje cells. Like Sac1p, rat Sac1 exhibits intrinsic phosphoinositide phosphatase activity directed toward phosphatidylinositol 3-phosphate, phosphatidylinositol 4-phosphate, and phosphatidylinositol 3,5-bisphosphate substrates, and we identify mutant rat sac1 alleles that evoke substrate-specific defects in this enzymatic activity. Finally, rat Sac1 expression in Delta sac1 yeast strains complements a wide phenotypes associated with Sac1p insufficiency. Biochemical and in vivo data indicate that rat Sac1 phosphatidylinositol-4-phosphate phosphatase activity, but not its phosphatidylinositol-3-phosphate or phosphatidylinositol-3,5-bisphosphate phosphatase activities, is essential for the heterologous complementation of Sac1p defects in vivo. Thus, yeast Sac1p and rat Sac1 are integral membrane lipid phosphatases that play evolutionary conserved roles in eukaryotic cell physiology.


    INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Phosphoinositides regulate diverse cellular processes such as cell proliferation, signal transduction, organization of the cytoskeleton, and membrane trafficking. They do so by regulating the intracellular localization and biological properties of a variety of proteins involved in these processes. Such effector proteins recognize specific phosphoinositides via lipid binding modules such as C2 domains, pleckstrin homology domains, FYVE domains, and SH21 domains (reviewed in Refs. 1-3).

The levels of specific phosphoinositide pools are regulated by intricate networks of enzymes and proteins that govern the rates of phosphoinositide synthesis, transport, and degradation (4-7). The physiological importance of proper regulation of phosphoinositide levels is underscored by abnormalities resulting from mutations in enzymes that effect synthesis, degradation, and subcellular localization of these molecules. For example, mutations in genes encoding an eye-specific CDP-diacylglycerol synthase (8), diacylglycerol kinase (rdgA) (9), and membrane-bound phosphatidylinositol transfer protein (rdgB) (10, 11) cause retinal degeneration in Drosophila. In mice, the vibrator mutation leads to a lethal neurodegenerative disease which is the result of reduced levels of phosphatidylinositol transfer protein alpha  (PITPalpha ) (12, 13). The human disease oculocerebrorenal syndrome of Lowe arises from mutations in the OCRL-1 gene which encodes an inositol polyphosphate-5-phosphatase (14). Given these findings, the characterization of proteins implicated in the metabolic turnover of phoshoinositides in mammalian cells has broad implications in biology and medicine.

Saccharomyces cerevisiae SAC1 was originally isolated as a gene whose dysfunction suppresses specific mutations in the actin structural gene ACT1 (15). SAC1 was independently identified in a genetic screen for mutations that bypass the normally essential requirement for Sec14p, the major yeast phosphatidylinositol/phosphatidylcholine transfer protein, in protein transport from the Golgi complex to the cell surface (16-18). In addition to these "bypass Sec14p" effects, other phenotypes also result from Sac1p insufficiencies in yeast. These sac1 phenotypes include: acquisition of an inositol auxotrophy without obvious defects in de novo synthesis of inositol or phosphatidylinositol (19), retardation of cell growth and disorganization of the actin cytoskeleton (15), disorganized chitin deposition at cold temperatures (15, 16), genetic interactions with several sec mutations (16), deficiencies in ATP uptake and preprotein traslocation from the cytosol into the lumen of the endoplasmic (20, 21), acquisition of multiple drug sensitivities (22), and enhanced rates of diacylglycerol-driven phosphatidylcholine biosynthesis via the CDP-choline pathway (23).

Further interest in Sac1p has been fueled by the recent demonstration that regions of considerable homology to Sac1p (Sac1 domains) are present in a subfamily of inositol polyphosphate-5-phosphatases, the synaptojanins. Synaptojanin 1, the first identified member of these enzymes, plays a critical role in regulation of a phosphoinositide pool involved in actin function and the recycling of synaptic vesicles in nerve terminals (24, 25). Furthermore, it has been found that yeast Sac1p and the Sac1 domains of yeast Inp52p (Sjl2p), Inp53p (Sjl3p), and human synaptojanin 1 have enzymatic activity. These proteins are phosphoinositide phosphatases that employ phosphatidylinositol 3-phosphate (PI(3)P), PI(4)P, and PI(3,5)P2, but not PI(4,5)P2, as substrates (26, 27). While yeast Sac1p has been analyzed in considerable detail, Sac1 proteins from other species have yet to be characterized. Herein we report the biochemical and functional properties of rat Sac1, the first Sac1 homologue to be defined in multicellular organisms.

    EXPERIMENTAL PROCEDURES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Isolation of Rat SacI cDNA-- Sac1 domains from S. cerevisiae Sac1p, Inp51p, Inp52p, Inp53p, and Fig4p and rat synaptojanin 1 were aligned via the PILEUP algorithm and adjusted by visual inspection. The degenerate oligonucleotides 5'-G(G/C)(G/A/T/C)AA(T/C)TT(T/C)(G/A)(T/ A)(G/A/T/C)GA(G/A)AC(G/A/T/C)GA-3' and 5'-G(G/C)(G/A/T/C)AA(T /C)TT(T/C)(G/A)(T/A)(G/A/T/C)GA(G/A)CA(G/A/T/C)GA-3' were derived from the highly conserved motifs, (A/G)NF(V/N)ETE and NCLDCLD, respectively. These oligonucleotides were used as primers for reverse transcriptase-polymerase chain reaction (PCR) from rat brain total RNA essentially as described (28). The PCR products were subcloned into pCRII vector (Invitrogen) and sequenced. Sequence analysis identified a clone with the 480-base pair insert encoding a part of a protein highly homologous to budding yeast Sac1p. This DNA fragment was radiolabeled with [alpha -32P]dCTP (Amersham Pharmacia Biotech) and used as a probe to screen a rat brain lambda  ZAPII cDNA library (Stratagene) following standard techniques (29). Nucleotide sequencing was performed by the chain termination method (30) by using double stranded plasmid DNA as template and the Sequenase version 2.0 sequencing kit (Amersham Pharmacia Biotech).

Northern Blot Analysis-- A Northern blot filter with 2 µg of poly(A)+ RNA from various adult rat tissues (rat multiple tissue Northern blot, CLONTECH) was probed for rat Sac1 by using the PCR-generated entire coding region. Rat glyceraldehyde-3-phosphate dehydrogenase cDNA (31) was used as control for equal loading of poly(A)+ RNA (data not shown).

Production of Anti-rat Sac1 Antibodies-- To generate antibodies against rat Sac1, the coding sequence for the amino-terminal 54 amino acid residues was amplified by PCR and subcloned into pGEX4T-1 vector (Amersham Pharmacia Biotech). The GST fusion protein was purified by glutathione-Sepharose 4B column chromatography (Amersham Pharmacia Biotech) according to the manufacturer's protocols and used to immunize rabbits. The antibodies were affinity purified using the fusion protein immobilized on polyvinylidene difluoride membrane following standard procedures (32).

Rat Liver Subcellular Fractionation-- Homogenization and subcellular fractionation of rat liver was performed as described by Fleischer and Kervina (33). Rat liver microsomes were isolated and fractionated following the discontinuous sucrose gradient centrifugation procedure described by Palade and co-workers (34). Materials from the interface between 0.25, 0.86 M and 0.86, 1.14 M sucrose were designated as Golgi light and Golgi heavy fractions, respectively. Fractions from 1.14, 1.18, and 1.24 M sucrose layers were defined as carrier vesicle fraction 1 and 2 (CV1 and CV2), and the endoplasmic reticulum (ER), respectively. Alkali sodium carbonate extraction of microsomes was carried out as described previously (35).

Mammalian Expression Plasmid Construction-- The sequence encoding a 10-amino acid Myc epitope, EEQKLISEDL (36), was inserted next to the initiation codon of rat Sac1 cDNA by a PCR-mediated procedure as described previously (28). The NH2-terminal Myc-tagged cDNA was subcloned into an eukaryotic expression vector pcDNA3 (Invitrogen). The plasmid for expression of NH2-terminal Myc-tagged Sac1 homology domain of rat Sac1 (amino acids 1-520) was constructed essentially in the same way. To express green fluorescent protein (GFP) fused with the COOH terminus of rat Sac1 (amino acids 521-587), the region was amplified by PCR and subcloned into pEGFP C1 (CLONTECH).

Expression and Purification of Recombinant Rat Sac1 Protein-- The region coding for the Sac1 homology domain of rat Sac1 (amino acids 1-520) was amplified by PCR and subcloned into a modified pFastBac1 baculovirus transfer vector (Life Technologies, Inc.) downstream of the coding sequence of glutathione S-transferase (GST) (48). A recombinant virus was produced and amplified in Sf9 insect cells by the Bac-to-Bac baculovirus expression system (Life Technologies, Inc.) following the manufacturer's instructions. A high titer solution of the recombinant baculovirus was used to infect Sf9 cells at 106 cells/ml. The cells were harvested after 72 h of incubation. The GST-Sac1 domain of rat Sac1 was purified from the lysed cells by glutathione-Sepharose 4B column chromatography.

Yeast Media and Genetic Techniques-- Minimal defined and complex media (YPD, supplemented with glucose to 2%, v/v) have been described (37). Lithium acetate yeast transformation (38) and gene disruption (39) methods were employed as described. Yeast strains are listed in Table I.

                              
View this table:
[in this window]
[in a new window]
 
Table I
Yeast strains

Site-directed Mutagenesis-- Site-directed mutagenesis of rat Sac1 was performed using the Quick Change site-directed mutagenesis kit (Stratagene). Two primers were generated for each mutation, each containing the desired mutation and designed to anneal to the same sequence on opposite strands of the plasmid. Mutagenic primers used were as follows: for D391N, 5'-CGCAGCAACTGTATGAATTGTCTAGACAG-3' and 5'-CTGTCTAGACAATTCATACAGTTGCTGCG-3'; for A442V, 5'-CCTGGG CCGATAATGTTAATGCTTGTGCC-3' and 5'-GGCACAAGCATTAACATTATCGGCCCAGG-3'; for R480H, 5'-GGCTTCAACTCATTATTACACTACTACAAGAACAAC-3' and 5'-GTTGTT CTTGTAGTAGTGTAATAATGAGTTGAAGCC-3'. Correctly mutagenized clones were confirmed by sequence analysis of the entire rat Sac1 cDNA and subcloned into a centromeric vector for constitutive PGK promoter-driven expression in the appropriate yeast strains (40).

Immunoblot Analysis of Yeast Lysates-- Yeast strains were grown overnight in selective minimal medium at 25 °C to an A600 of 1.0. Spheroplasts were prepared as described (41, 42), washed in 1.2 M sorbitol, 10 mM Tris-HCl, pH 7.4, 0.5 mM phenylmethylsulfonyl fluoride, and resuspended in lysis solution (0.3 M sorbitol, 10 mM Tris-HCl, pH 7.4, 0.5 mM phenylmethylsulfonyl fluoride). Cells were incubated on ice for 30 min with periodic trituration every 5 min. Lysates were clarified by centrifugation for 5 min at 2,000 × g. Alkaline extraction of the yeast cell lysates was described previously (19). Thirty µg of protein for each sample was resolved by SDS-PAGE, and immunoblots were developed using the enhanced chemiluminescence system (Amersham Pharmacia Biotech).

Invertase Assays-- Yeast strains were grown to mid-logarithmic growth phase in standard YPD medium at 25 °C, washed, and then shifted to low glucose (0.1%) YPD (to induce synthesis of secretory invertase) and 37 °C (to impose Sec14p deficiency in sec14-1ts strains). After 2 h at 37 °C, 1-ml samples were removed from each culture, poisoned with sodium azide at a final concentration of 10 mM on ice. Cells were processed as described in detail elsewhere for invertase assays, and secretion indices calculated as described (43).

Bulk Phospholipid Analysis and Quantification by 32P Radiolabeling-- In experiments where bulk phospholipid species were identified and quantified in yeast, cells were grown to mid-logarithmic phase in inositol containing minimal medium and radiolabeled with [32P]orthophosphate for 20 min at 25 °C. Lipids were extracted and resolved by two-dimensional paper chromatography using previously described solvents (23, 44, 45). Extracts derived from equal numbers of cells were loaded on each chromatogram, and specific 32P-radiolabeled phospholipids were identified by autoradiography, excised, and quantified by liquid scintillation counting, or identified and quantified by phosphorimaging.

Inositol Radiolabeling and Phosphoinositide Analysis-- Approximately 1 × 105 yeast cells were inoculated into 1-ml of SD medium supplemented with appropriate amino acids and containing 10-20 µCi of myo-[3H]inositol (American Radiolabel Co., St. Louis, MO). Cells were grown to steady state at 30 °C, harvested by centrifugation, and washed twice with 3-ml of ice-cold water. Phosphoinositides were extracted by addition of 100 µl of 0.5 N HCl, followed by 400 µl of chloroform/methanol (1:2, v/v) and vortexing vigorously with glass beads for 2 min. For improvement of recovery, 200 µg of crude brain phosphoinositides (Sigma) were added as lipid carrier. Aqueous and organic phases were then generated by adding 300 µl each of chloroform and 1 M KCl, and then separated by brief centrifugation. The aqueous phase was re-extracted with 400 µl of chloroform, and the organic phases were pooled and dried under a stream of nitrogen. Lipids were deacylated with methylamine as described (46). Samples were re-dissolved in water, separated by high performance liquid chromatography (HPLC) on a Partisphere SAX ion exchange column (4.6 × 125 mm; Whatman, Clifton, NJ), and eluted at a flow rate of 1 ml/min using the following gradient profile: buffer B: 0-10 min, 0% B; 10-55 min, 0-35% B, 55-70 min, 35-100% B; buffer B: 1.4 M (NH4)2HPO4, pH 3.7. Radioactivity was measured using an on-line liquid scintillation counter (Packard Instrument Co.).

Miscellaneous Procedures-- Polyclonal rabbit antibodies against protein-disulfide isomerase and calnexin were purchased from StressGen Biotechnologies (Victoria, Canada). Rabbit polyclonal antibody against type I InsP3 receptor was described by Takei et al. (47). Mouse monoclonal antibody against GM130 was purchased from Transduction Laboratories (Lexington, KY), while mouse anti-Myc monoclonal antibody (clone 9E10; American Type Culture Collection) was used for detection of Myc epitope. Protein concentration was determined using the BCA assay (Pierce Biochemicals). Transfection of cells, immunostaining procedures, and other techniques and materials used are described elsewhere (23, 47, 48).

    RESULTS
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Identification of a Rat Sac1 Homologue-- S. cerevisiae Sac1p represents a prototypical member of a growing family of Sac1-domain containing proteins. Sac1 homology domains have been identified in four other open reading frames in the budding yeast genome, and in mammalian inositol 5-phosphatases typified by synaptojanin (24, 49, 50, 51). Proteins of this family display the Sac1 homology domains at their amino termini, and these Sac1 homology domains typically share 24-32% amino acid identity with budding yeast Sac1p. In an attempt to identify a mammalian counterpart of Sac1p, we designed a set of oligonucleotide primers based on the conserved motifs in these Sac1 domains and isolated a partial cDNA fragment by reverse transcriptase-PCR from rat brain. Subsequent screening of a rat brain library identified a cDNA exhibiting an open reading frame of 1,764 nucleotides that encodes a protein of 587 amino acids (Mr = 67,035). This protein, which we designate rat Sac1, shares 35% amino acid identity to S. cerevisiae Sac1p (Fig. 1A).


View larger version (122K):
[in this window]
[in a new window]
  		Fig. 1.   Primary structure of rat Sac1. A, amino acid sequence comparison of rat Sac1 and proteins homologous to Sac1p. Amino acid sequences of rat Sac1 and S. cerevisiae Sac1p are aligned with those of putative Sac1p homologues by the Clustal method (66). Amino acid residues are numbered at the left. The conserved CX5R(T/S) motif (26, 53) commences at rat Sac1 residue 389 and is highlighted by asterisks at top. The sequence data for rat Sac1 are available from GenBank/EMBL/DDBJ under accession number AF251186. The amino acid sequences of putative human, Caenorhabditis elegans, Drosophila melanogaster, Schizosaccharomyces pombe, and Arabidopsis thaliana Sac1 are derived from AB020658, Z81072, AE003735, AL022599, and AF049236, respectively. B, Kyte-Doolittle hydrophilicity profile of rat Sac1 (67). Rat Sac1 has a putative transmembrane segment near its COOH terminus (amino acids 521 to 543). Bar corresponding to amino acid residues is at top, and hydrophobic regions lie below the 0 line.

Data base searches reveal the presence, in diverse species, of proteins that are closely related to rat and budding yeast Sac1p. The widespread existence of such putative homologues suggests an evolutionary conservation of Sac1p function in cell physiology (Fig. 1A). In support of this notion, all three amino acid residues presently known to be important for in vivo Sac1p function (49, 52), and a CX5R(T/S) active site motif found in other metal independent protein and inositol polyphosphate phosphatases (26, 53), are strongly conserved among all the putative Sac1 homologues from mammals, insects, nematodes, fungi, and plants. A distinguishing feature of yeast Sac1p with regard to other phosphoinositidases is that it represents an integral membrane protein with a single putative transmembrane domain that resides near the COOH terminus of the protein (16, 19). A similar situation applies to rat Sac1. Kyte-Doolittle hydropathy analysis identifies a stretch of 23 hydrophobic amino acid residues (Phe521 to Gly543) that likely constitutes a transmembrane segment near the rat Sac1 COOH terminus (Fig. 1B). The presence of a putative COOH-terminal transmembrane segment is a structural feature shared by the other Sac1 homologues depicted in Fig. 1A.

Rat Sac1 Is Widely Expressed and Localizes to the Endoplasmic Reticulum-- Northern blot analyses of adult rat tissues demonstrate that rat Sac1 mRNA is expressed as a 4-kilobase transcript in all tissues tested (Fig. 2A). The 4-kilobase transcript is expressed most strongly in brain, spleen, liver, and kidney, and is expressed at lower levels in the other tissues examined. Immunoblot analyses were performed with these same rat tissues to determine expression of rat Sac1 at levels. These immunoblotting experiments employed a polyclonal antibody raised against the NH2 terminus of rat Sac1 (see "Experimental Procedures"). This antibody detected a single 65-kDa protein in all rat tissues examined (Fig. 2B), in excellent agreement with the calculated Mr of rat Sac1. The relative abundance of this protein band correlated very closely with the rat Sac1 mRNA levels detected in these same tissues (Fig. 2, A and B). These data demonstrate that rat Sac1 is expressed in a wide spectrum of adult rat tissues. We have also identified a murine Sac1 that exhibits homology to rat Sac1. This murine Sac1 is also expressed in all adult mouse tissues tested, and is expressed in murine embryonic stem cells as well (not shown). Thus, mammalian Sac1 appears to be ubiquitously expressed in both adult and embryonic tissues.


View larger version (36K):
[in this window]
[in a new window]
  		Fig. 2.   Rat Sac1 is ubiquitously expressed. A, tissue distribution of rat Sac1 mRNA. Rat Sac1 mRNA was detected by Northern blotting with the probe spanning the entire coding region. RNA size markers are indicated in kilobase at the left. B, Western blot analysis of tissue distribution of rat Sac1. Total protein extracts (50 µg each) from various rat tissues were subjected to SDS-PAGE and rat Sac1 was detected by immunoblotting with anti-rat Sac1 antibody. Molecular weight markers are indicated in kDa at the left. In both panels: lane 1, heart; lane 2, brain; lane 3, spleen; lane 4, lung; lane 5, liver; lane 6, skeletal muscle; lane 7, kidney; lane 8, testis.

Yeast Sac1p localizes to the Golgi and endoplasmic reticulum (19). To characterize the intracellular localization of rat Sac1 in mammalian cells, we first probed rat liver subcellular fractions for the rat Sac1 antigen by immunoblotting. In agreement with the presence of a putative transmembrane domain, the rat Sac1 antigen fractionated with particulate fractions. Microsomes containing Golgi and endoplasmic reticulum membranes were particularly enriched in rat Sac1 immunoreactivity (Fig. 3A). To better define the compartment in which rat Sac1 resides, rat liver microsomes were further fractionated by equilibrium sucrose density gradient centrifugation. Analyses of the fractions revealed a substantial co-enrichment of rat Sac1 with the endoplasmic reticulum marker protein calnexin (Fig. 3B). No significant enrichment of rat Sac1 was recorded in fractions enriched in GM130, a Golgi marker protein (Fig. 3B). These data suggest that rat Sac1 localizes predominantly to the ER.


View larger version (41K):
[in this window]
[in a new window]
  		Fig. 3.   Cofractionation of rat Sac1 with calnexin. A, immunoblot analyses of rat liver subcellular fractions. Rat liver homogenates were fractionated by differential fractionation and discontinuous sucrose gradient centrifugation following the procedure of Fleischer and Kervina (33). The low-speed P1 fraction was further resolved into nuclear membrane (NC), plasma membrane (PM), and mitochondrial (MT) fractions. Each fraction (50 µg of protein) was separated by SDS-PAGE and analyzed by immunoblotting with anti-rat Sac1 antibody, anti-GM130 antibody, or anti-calnexin antibody as indicated. B, crude rat liver microsomes (TM) were fractionated into the Golgi light (GL), Golgi heavy (GH), carrier vesicle 1 and 2 (CV1 and CV2), and ER fractions by sucrose step gradient centrifugation method of Jin et al. (34). Fractions were analyzed by immunoblotting as described above in A.

To gain an independent assessment of rat Sac1 localization, we performed an indirect immunofluorescence study. While we generally failed to detect sufficient signal to assess the subcellular localization of endogenous protein, we found particularly high levels of rat Sac1 in cerebellar Purkinje cells. Both Purkinje cell bodies and the dendritic processes in the molecular layer were intensely stained with anti-rat Sac1 antibodies (Fig. 4A). Furthermore, their axons in the granule cell layer, the white matter, and the deep cerebellar nuclei were also stained, albeit to a lesser degree (Fig. 4A; data not shown). As shown in Fig. 4B, the distribution of rat Sac1 in Purkinje cells closely resembled that of type I InsP3 receptor, a resident protein of the ER membrane (47). To better assess the precise distribution of rat Sac1, we sought to employ mammalian cells that are more amenable to fine morphological resolution. We transfected Chinese hamster ovary (CHO) cells with a vector driving expression of rat Sac1 tagged with the Myc epitope. Indirect immunofluorescence methods, using the anti-Myc monoclonal 9E10 antibody as primary antibody, were subsequently employed to visualize the distribution of the Myc-tagged rat Sac1 in these transfected cells. These experiments demonstrated that rat Sac1 localized to reticular structures, and double-label immunofluorescence experiments demonstrated a colocalization of rat Sac1 with the ER marker protein-disulfide isomerase (Fig. 5, A-D). The putative rat Sac1 transmembrane domain was required for the observed ER localization. The Myc-tagged protein which lacks the putative COOH-terminal transmembrane domain exhibited a diffuse cytosolic staining in CHO cells (Fig. 5, E and F). By contrast, a chimeric protein consisting of jellyfish GFP fused to the rat Sac1 COOH-terminal domain exhibited a reticular distribution consistent with ER localization (Fig. 6). Taken together, the data indicate that rat Sac1 is localized predominantly to ER membranes, and that the rat Sac1 COOH-terminal transmembrane domain is both necessary and sufficient to direct localization to this compartment.


View larger version (100K):
[in this window]
[in a new window]
  		Fig. 4.   Localization of rat Sac1 immunoreactivity in the Purkinje cell layer of the cerebellum. Sections of paraformaldehyde-fixed rat cerebellum were immunostained either with rabbit polyclonal antibody against rat Sac1 (A) or with rabbit polyclonal antibody against InsP3 receptor (B) and visualized with Cy3-conjugated goat anti-rabbit antibody for immunofluorescence analysis. The scale bar corresponds to 20 µm.


View larger version (129K):
[in this window]
[in a new window]
  		Fig. 5.   Localization of recombinant rat Sac1 to the endoplasmic reticulum in CHO cells. CHO cells were transfected with the Myc-tagged construct for full-length rat Sac1 (A-D) or its Sac1 domain (amino acids 1-520; lacking the most COOH-terminal portion) (E and F). The cells were co-incubated with rabbit polyclonal antibody against protein-disulfide isomerase (A, C, and E) and mouse monoclonal anti-Myc antibody (9E10) (B, D, and F) and stained for double-label immunofluorescence analysis. Recombinant rat Sac1 distributes to a reticular structure characteristic of the endoplasmic reticulum. The scale bar corresponds to 15 µm in A-D, and 21 µm in E and F.


View larger version (152K):
[in this window]
[in a new window]
  		Fig. 6.   The COOH-terminal portion of rat Sac1 confers localization to the endoplasmic reticulum. Images of CHO cells transfected with the construct driving expression of a chimeric protein, where GFP is fused to the most COOH-terminal portion of rat Sac1 (amino acids 521-587), were captured by fluorescence microscopy. Visualization of GFP fluorescence marked the intracellular distribution of the COOH-terminal domain of rat Sac1 (A and C). For visualization of the endoplasmic reticulum, these same cells were co-stained with a rabbit polyclonal antibody directed against a resident of the endoplasmic reticulum lumen, protein-disulfide isomerase (B and D). The scale bar corresponds to 10 µm in panels A and B, and 21 µm in panels C and D.

Rat Sac1 Is an Integral Membrane Protein with Its NH2 Terminus Disposed to the Cytoplasm-- The hydropathy profile of the deduced rat Sac1 primary sequence suggests that rat Sac1 is an integral membrane protein anchored to membranes via the COOH-terminal transmembrane domain (Fig. 1B). We also expected the NH2-terminal Sac1 domain to be oriented toward the cytoplasm. To test these predictions, rat liver microsomes were stripped with alkaline sodium carbonate to remove all but integral membrane proteins (35), and the membrane and soluble fractions were probed for rat Sac1 antigen by immunoblotting. The data clearly show that rat Sac1 is resistant to extraction from membranes with alkaline carbonate (Fig. 7A). By contrast, protein-disulfide isomerase, a soluble resident protein of the ER lumen, was efficiently released from microsomal membranes by the same extraction procedure. On the basis of its resistance to alkaline extraction, we conclude that rat Sac1 is an integral membrane protein.


View larger version (32K):
[in this window]
[in a new window]
  		Fig. 7.   Rat Sac1 is an integral membrane protein with a cytoplasmic NH2 terminus. Rat liver microsomes were isolated and resuspended in 0.1 M Na2CO3, pH 11.5, and the alkali-inextractable fraction was obtained by centrifugation at 70,000 rpm for 30 min in a Beckman TL-100 ultracentrifuge (A). Alternatively rat liver microsomes were incubated with the indicated concentration of proteinase K either in the absence or presence of Triton X-100 for 10 min at room temperature. The reaction was terminated by addition of phenylmethylsulfonyl fluoride to a final concentration of 2 mM and the mixture was precipitated by trichloroacetic acid (B). The samples were analyzed by SDS-PAGE and immunoblotting with antibody against the amino terminus of rat Sac1 or with antibody against protein-disulfide isomerase (PDI).

To determine how rat Sac1 is oriented within microsomal membranes, we employed protease digestion as a probe for rat Sac1 topology. In these experiments, intact rat liver microsomes were challenged with exogenous proteinase K, and the shaved microsomes were subsequently probed for the rat Sac1 antigen with antibody raised against the rat Sac1 NH2-terminal domain (see "Experimental Procedures"). As clearly shown in Fig. 7B, rat Sac1 was efficiently degraded by low concentrations of proteinase K (1 µg/ml), regardless of whether microsomes were left intact, or their integrity was disrupted by solubilization in Triton X-100 buffer. Microsomal integrity was not compromised in detergent-free incubations as judged by the resistance of the lumenal protein protein-disulfide isomerase to proteinase K digestion. Protein-disulfide isomerase is not inherently resistant to proteinase K because Triton X-100 permeabilization of microsomes rendered protein-disulfide isomerase sensitive to proteinase K challenge (Fig. 7B). These topology mapping data indicate that rat Sac1 assumes an orientation with its NH2 terminus disposed toward the cytoplasm.

Expression of Rat Sac1 and the Mutant Forms in Yeast-- We next explored the functional similarity between rat Sac1 and yeast Sac1p. To study this issue in detail, it was necessary to develop a functional assay for rat Sac1. To this end, rat Sac1 was expressed in yeast strains rendered free of endogenous Sac1p by deletion of the nonessential SAC1 structural gene. We constructed a centromere-based yeast shuttle vector where the rat Sac1 cDNA was placed under the control of the yeast phosphoglycerate kinase (PGK) promoter. This powerful promoter drives constitutive expression of genes, and the rat Sac1 expression plasmid was designated YCp(rSAC1).

To extend these functional analyses we also wished to analyze mutant forms of rat Sac1 that were defective in phosphoinositide phosphatase activity. We had previously characterized several sac1 missense alleles that produce fully stable mutant forms of yeast Sac1p that are nonetheless nonfunctional in vivo (19, 49, 52). These alleles (sac1-8, sac1-10, and sac1-22), and the cognate amino acid substitutions, are depicted in Fig. 8A. As these amino acid residues are conserved in rat Sac1, we introduced the corresponding point mutations into rat SAC1 cDNA by site-directed mutagenesis. Plasmids for expression of these D391N, A442V, and R480H mutant forms of rat Sac1 (corresponding to the yeast sac1-8, -10, and -22 proteins, respectively; Fig. 8A) were also transformed into Delta sac1 yeast strains for analysis. Immunoblotting experiments demonstrate that wild-type and the mutant forms of rat Sac1 are expressed in Delta sac1 yeast strains as stable proteins that accumulate to levels comparable to those observed for wild-type rat Sac1 (Fig. 8B). These immunoreactive species represent heterologous rat Sac1 forms on the basis that these species were detected in strains harboring the appropriate YCp(rSAC1) or YCp(rsac1) plasmids, but not in strains carrying a YCp(URA3) control plasmid.


View larger version (34K):
[in this window]
[in a new window]
  		Fig. 8.   Expression of wild-type and mutant forms of rat Sac1 in yeast. A, alignments of sequences from yeast proteins with Sac1 domains (Inp51p(Sjl1p), Inp52p(Sjl2p), Inp53p(Sjl3p), and Fig4p), and rat synaptojanin and rat Sac1, that correspond to functionally critical sequence motifs of yeast Sac1p. These critical Sac1p motifs are defined by three Sac1p missense substitutions that individually preserve protein stability, yet compromise function with regard to control of inositol lipid metabolism, inositol auxotrophy, growth at cold temperatures, complementation of bypass Sec14p phenotypes, and actin cytoskeleton function (19, 49). The involved residues are highlighted in Sac1p and, when conserved, in these Sac1 domains. The missense substitutions are identified by the amino acid substitutions and by the corresponding allelic designations (at top). Corresponding substitutions generated in rat Sac1 are identified at bottom. Residues conserved in all of the indicated Sac1 domains are highlighted by boxes. The position of the first residue of each motif in the primary sequence is given. B, cell-free extracts were prepared from YCp(rSAC1) yeast strains expressing rat Sac1 or each of the three mutant forms. The extracts with equivalent protein amounts were analyzed by SDS-PAGE and immunoblotting with anti-rat Sac1 serum. Missense proteins are labeled under the rsac1 umbrella. All of these proteins accumulated to steady-state levels comparable to wild type rat Sac1. C, rat Sac1 is expressed as an integral membrane protein in yeast. A yeast strain deleted for the SAC1 gene (CTY244) was transformed with a control vector or with a centromere-based rat SAC1 expression plasmid, YCp(rSAC1), and was grown overnight to a low cell density (A600 = 0.8-1.0). The cells were then subjected to osmotic lysis. The resultant lysate was subjected to two rounds of centrifugation (see "Experimental Procedures"). The whole cell (WC), 100,000 × g supernatant (S100), and 100,000 × g pellet (P100) fractions were resolved by SDS-PAGE and probed by immunoblotting with anti-rat Sac1 antibody. As a negative control, lysates prepared from the Delta sac1,YCp(URA3) strain were also probed. For alkaline extraction experiments, osmotic lysate was prepared and half of the sample was adjusted to 0.1 M Na2CO3, pH 11.5. After a 1-h incubation on ice, the samples were centrifuged at 100,000 × g and the resulting S100 and P100 fractions were probed by immunoblotting for rat Sac1 and Kes1p, a peripheral membrane protein control.

Fractionation experiments were employed to determine whether the normal integral membrane disposition of rat Sac1 is preserved upon expression of this protein in yeast. Rat Sac1 was quantitatively recovered in the membrane pellet (P100) generated by centrifugation of whole yeast cell lysate at 100,000 × g (Fig. 8C). Kes1p, a peripheral membrane-associated protein (54), distributes both to soluble (S100) and P100 fractions. Moreover, while alkaline carbonate treatment efficiently strips Kes1p from membranes, rat Sac1 is fully resistant to extraction from the P100 fraction (Fig. 8C). On the basis of these data, we conclude that YCp(rSAC1) and the cognate YCp(rsac1) plasmids drive robust expression of rat Sac1 and its mutant forms in yeast. The data also indicate that the heterologous rat Sac1 adopts an integral membrane disposition in yeast cells. The functional properties of rat Sac1 and its mutant forms were then subjected to more detailed analysis in the heterologous yeast system.

Rat Sac1 Is a Phosphoinositide Phosphatase with a Substrate Specificity Similar to That of Yeast Sac1p-- Yeast Sac1p is a phosphoinositide phosphatase whose inactivation results in a marked accumulation of specific phosphoinositides in vivo (23, 26, 27, 55). To investigate whether rat Sac1 has similar properties, we first addressed this question in an in vivo context by comparing the phosphoinositide profiles of isogenic wild-type, Delta sac1, and Delta sac1,YCp(rSAC1) yeast strains. [3H]Inositol radiolabeling experiments confirmed previous demonstrations that Delta sac1 mutations effect a about 2.5-fold increase in PI(3)P, an 8-10-fold increase in PI(4)P, and approximately a 10-fold increase in PI(3,5)P2 levels compared with those found in wild-type yeast strains (Fig. 9, A and B). Although PI(4,5)P2 levels do not deviate significantly from those in wild-type yeast, a small (but reproducible) reduction in PI levels was recorded in Delta sac1 strains. Expression of rat Sac1 in Delta sac1 yeast restored PI(3)P, PI(4)P, and PI(3,5)P2 levels to normal. These data indicate that rat Sac1 expression fully corrects the pleiotropic perturbations in phosphatidylinositol metabolism associated with loss of Sac1p function in yeast, and strongly suggest that rat Sac1 and yeast Sac1 share considerable biochemical similarity with regard to their respective phosphoinositide phosphatase activities.


View larger version (29K):
[in this window]
[in a new window]
  		Fig. 9.   Phosphoinositide levels in sac1 mutant strains. Cells were labeled to steady-state with [3H]inositol, and phosphoinositides were extracted and separated by HPLC after deacylation. Typical chromatographic tracings are shown in A. Data are expressed as radioactivity in each fraction (cpmi) normalized to the total radioactivity of each run (cpmtot). B, levels of phosphoinositides from a wild-type SAC1 strain, the sac1 deletion mutant strains transformed with empty YCp(URA3) vector, or YCp-borne rSAC1 or the mutants encoding rat Sac1D391N, rat Sac1A442V, rat Sac1R480H, respectively. Data are expressed as mean ± S.D. of at least four independent experiments each.

To test whether rat Sac1 has intrinsic phosphoinositide phosphatase activity directed against the relevant phosphoinositide species, we expressed the Sac1 homology domain of rat Sac1 (amino acids 1-520; lacking the transmembrane segment) in insect cells. The intrinsic phosphatase activity of purified recombinant protein was then measured under conditions where labeled yeast phosphoinositides were presented as in vitro substrates (see "Experimental Procedures"). As shown in Fig. 10, the Sac1 domain of rat Sac1 exhibits intrinsic phosphoinositide phosphatase activity when PI(3)P, PI(4)P, or PI(3,5)P2 are employed as substrates in vitro. By contrast, PI(4,5)P2 is not a suitable substrate for the rat Sac1 enzyme under the experimental conditions employed. These results formally prove that rat Sac1 displays phosphoinositide phosphatase activity against PI(3)P, PI(4)P, and PI(3,5)P2 substrates, and that the Sac1 homology domain is the catalytic entity. When coupled with the in vivo data described above, these results also demonstrate that the enzymatic properties of rat Sac1 are substantially similar to those of yeast Sac1p.


View larger version (22K):
[in this window]
[in a new window]
  		Fig. 10.   Phosphoinositide phosphatase activity of recombinant Sac1 domain from rat Sac1 in vitro. Phosphoinositide substrates were prepared from total cellular extracts of Delta sac1 strains labeled to steady state with [3H]inositol. Radiolabeled substrates were incubated with purified GST-Sac1 domain of rat Sac1 or a control protein GST-Grb2 expressed in insect cells for 1 h at 37 °C. Reaction products were extracted, deacylated, and resolved by HPLC. A, typical chromatograms of reactions with either the control protein (ctrl) or GST-Sac1 domain of rat Sac1 (rSAC1). The inset shows the chromatography of PI(3,5)P2 and PI(4,5)P2 on an expanded scale. B, levels of reactants of three independent experiments (mean ± S.D.).

Identification of Rat Sac1 Mutations That Elicit Substrate-specific Enzymatic Defects-- To characterize the enzymatic activities of rat Sac1 point mutant forms, we analyzed the phosphoinositide profiles of Delta sac1 yeast strains expressing rat Sac1D391N, rat Sac1A442V, or rat Sac1R480H as well. As in the case of the yeast Sac1pD337N and Sac1p A387V proteins (the products of the sac1-8 and sac1-10 alleles, respectively) expression of neither rat Sac1D391N nor rat Sac1A442V had any significant effect of reducing the massive accumulation of PI(4)P observed in Delta sac1 strains (Fig. 9B). Interestingly, expression of these two mutant rat Sac1 forms did significantly reduce accumulation of PI(3)P, and particularly PI(3,5)P2, in the Delta sac1 strain (Fig. 9B). These data suggest that rat Sac1D391N and rat Sac1A442V retain considerable activity against these phosphoinositide species, i.e. the D391N and A442V missense substitutions effect a more pronounced decrease in the phosphatidylinositol-4-phosphatase activity than in the phosphatidylinositol-3- and 5-phosphatase activities. Thus, the rat Sac1D391N and rat Sac1A442V proteins exhibit significant positional specificity in their respective phosphoinositide phosphatase defects. By contrast, rat Sac1R480H expression restores essentially normal phosphoinositide profiles in Delta sac1 yeast mutants (Fig. 9B). This difference is in agreement with the unique phenotype of the corresponding yeast mutant form (Sac1pR425H, i.e. the product of the sac1-22 allele) in comparison to the sac1-8 and sac1-10 alleles (Ref. 49; see below). Furthermore, the yeast Sac1pR425H is capable of effecting a nearly normal regulation of inositol phospholipid metabolism when cells are grown in the absence of exogenous inositol (23, 49, 55). Rat Sac1R480H, however, is a fully functional phosphoinositide phosphatase, even under inositol-replete growth conditions, while Sac1pR425H is not. These results indicate that this particular mutant rat Sac1 behaves somewhat differently from its Sac1pR425H counterpart in yeast.

Rat Sac1 Expression Complements Delta sac1-associated Phenotypes in Yeast-- Our demonstration that rat Sac1 and yeast Sac1p share similar biochemical activities prompted us to determine whether rat Sac1 can functionally substitute for yeast Sac1p activity in vivo. To this end, we determined whether rat Sac1 expression could complement Delta sac1-associated phenotypes in yeast. Sac1p deficiency in yeast exerts pleiotropic effects (15, 16, 19, 20, 22, 23, 49). Hallmark biological phenotypes include: (i) a restoration of cell viability and Golgi secretory function to Sec14p-deficient yeast strains (i.e. a bypass Sec14p phenotype), (ii) an imposition of an inositol auxotrophy to yeast strains that are fully capable of synthesizing inositol de novo, (iii) a synthetic lethality of sac1 alleles with mutations that inactivate tryptophan biosynthesis, and (iv) a marked defect in the ability of cells to grow at 13 °C (cold sensitivity for growth).

We first analyzed the ability of rat Sac1 and its mutant forms to complement the bypass Sec14p phenotype of Delta sac1 strains. Yeast strains expressing a thermolabile Sec14p cannot grow at the restrictive temperature of 37 °C because of their inability to transport secretory proteins from the yeast Golgi complex. Inactivation of yeast Sac1p (i.e. in Delta sac1 strains) restores both growth and secretory competence to sec14ts strains at 37 °C (16, 17, 41). Expression of rat Sac1 re-imposes temperature-sensitive growth to Delta sac1,sec14ts strains (Fig. 11, A and B) and, as expected, this growth defect at 37 °C is fully complemented by incorporation of a wild-type SEC14 gene into this strain (data not shown).


View larger version (55K):
[in this window]
[in a new window]
  		Fig. 11.   Rat Sac1 functionally substitutes for yeast Sac1p. A, rat Sac1 expression restores temperature sensitivity to Delta sac1,sec14-1ts strains at 37 °C. Yeast strains were streaked for individual colonies on uracil-free selective minimal plates and placed at either 26 or 37 °C. Growth was scored after 3 days. To assess the functional state of defined missense rat Sac1 mutants, yeast strain CTY243 (Delta sac1,sec14-1ts) was individually transformed with centromeric plasmids that drive expression of wild-type rSAC1, mutant rSAC1s encoding rat Sac1D391N, rat Sac1A442V, and rat Sac1R480H, respectively, and an empty URA3 vector. As an additional control, Delta sac1 strains carrying a wild-type copy of SEC14 were also plated, as indicated, to demonstrate that there was no inherent temperature sensitivity in strains transformed with the rsac1 expression plasmids. Plates were scored for growth after 3 days at the indicated conditions. Expression of either the rat Sac1D391N or rat Sac1A442V mutant (designated N391 and V442, respectively) is permissive for suppression of sec14-1ts by Delta sac1. Expression of rat Sac1R480H (designated H480) effects a functional substitution for Sac1p function as evidenced by lack of growth of this strain at 37 °C. B, invertase secretion in Delta sac1 yeast mutants carrying YCp(rSAC1). For secretion of invertase under Sec14p proficient and deficient conditions, respectively, wild-type and sec14-1ts strains were also employed as indicated. The Delta sac1,sec14-1ts strains transformed with YCp(rSAC1) show greater than a 2-fold reduction in invertase secretory ability compared with the parent strain transformed with the control YCp URA3 vector. Actual secretion index values obtained for each strain from at least three experiments are given above each corresponding bar. C, Delta sac1 strains harboring the indicated expression plasmids were either plated onto inositol-replete or inpatient defined minimal media as indicated, and incubated at 26 °C for 5 days. Both rat Sac1D391N and rat Sac1A442V scored as inactive in this complementation assay, as judged by the inability of their expression to restore inositol prototrophy to Delta sac1 strains. Expression of either rat Sac1p or rat Sac1R480H complemented the Delta sac1-associated inositol auxotrophy in yeast. Designations of rat Sac1 mutants are described in the legend to Fig. 8A. D, Delta sac1 strains harboring the indicated expression plasmids were streaked for isolation onto YPD agar as indicated, and incubated at 26 °C for 3 days or 13 °C for 10 days, as indicated. Expression of rat Sac1 complemented the Delta sac1-associated cold-sensitivity for growth.

The phenotypic complementation by rat Sac1 of bypass Sec14p phenotypes associated with Delta sac1 alleles in yeast extends to re-imposition of secretory defects to Delta sac1,sec14ts mutants. Invertase secretion efficiency provides a measure of yeast secretory competence. This efficiency is quantified by an invertase secretion index that relates the percentage of secreted invertase relative to the total amount of invertase produced by the cells (41, 43, 56). Wild-type yeast cells secrete invertase rapidly and efficiently. Thus, virtually all of the total invertase activity is localized to the cell surface after induction of invertase synthesis by a 2-h shift of cells to 37 °C in low glucose medium in wild-type cells (secretion index = 0.92 ± 0.05; Fig. 11B). Under these same conditions, sec14ts mutants localize only a minor fraction of the total invertase activity to the cell surface (secretion index = 0.14 ± 0.03) because the temperature shift to 37 °C induces Sec14p deficiency in these mutants. The bulk of the secretory invertase is trapped within the yeast Golgi complex as a consequence of defects in Sec14p-dependent protein transport from this organelle (41, 57). The secretion index of Delta sac1,sec14ts strains at 37 °C (0.80 ± 0.03; Fig. 11B) is similar to that of wild-type strains, and this improvement in invertase secretion quantifies the efficiency with which sac1 alleles suppress Sec14p secretory defects (16). Expression of rat Sac1 in the Delta sac1,sec14ts strain markedly reduces the secretion index from 0.80 ± 0.03 to 0.30 ± 0.08; a value approaching that measured for sec14ts mutants (Fig. 11B). These results demonstrate a substantial re-imposition of the sec14ts secretory block in Delta sac1,sec14ts mutants expressing rat Sac1.

In other studies, we have uncoupled discrete yeast Sac1p activities by comparing the phenotypes of yeast carrying sac1-8, sac1-10, and sac1-22 missense alleles to the phenotypes of yeast strains harboring Delta sac1 null mutations (see Refs. 21 and 49; data not shown). The pleiotropic nature of Delta sac1-associated phenotypes, when coupled with the likelihood that yeast Sac1p is a multifunctional protein (see "Discussion"), raises the possibility that rat Sac1p expression might effect a selective rescue of sac1 phenotypes. Counter to this possibility, we find that complementation of Delta sac1-associated phenotypes by rat Sac1 is not restricted to bypass Sec14p. Rat Sac1 expression also fully complements the inositol auxotrophy (Fig. 11C) and cold-sensitivity for growth (Fig. 11D) phenotypes that characterize Delta sac1 yeast strains.

Rat Sac1p Expression Corrects Aberrant Phosphatidylcholine Metabolism in Delta sac1 Strains-- While derangement of phosphoinositide metabolism is a signature of sac1 yeast mutants, such mutants exhibit other striking abnormalities in phospholipid metabolism. One such example involves the dramatic effects on phosphatidylcholine (PC) metabolism that are evoked by Sac1p deficiency; i.e. sac1 mutations effect a specific 3-fold increase in the rate of metabolic flux through the CDP-choline pathway for PC biosynthesis (23). To determine whether rat Sac1 expression corrects this metabolic defect in Delta sac1 strains as well, we measured rates of PC biosynthesis in wild-type and Delta sac1 yeast, and compared these values with those recorded in an isogenic Delta sac1,YCp(rSAC1) strain. The appropriate yeast strains were cultured to early logarithmic growth phase in inositol-containing medium at 25 °C and pulse-radiolabeled for 20 min with [32P]orthophosphate. Bulk phospholipids were extracted from each culture, resolved by paper chromatography, and quantified. Although [32P] does not specifically radiolabel the PC pool that is synthesized via the CDP-choline pathway (PC generated via methylation of phosphatidylethanolamine is also radiolabeled by this regimen), the short period of radiolabeling preferentially monitors CDP-choline pathway activity (23, 45). As reported previously (23), the Delta sac1 mutant incorporates [32P] into PC at rates that are 3.5-fold greater than those measured for the isogenic wild-type strain (Fig. 12). Expression of rat Sac1 in the Delta sac1 strain reduces the rate of [32P] incorporation into PC to what is essentially a wild-type value. These data indicate rat Sac1 expression not only restores proper phosphoinositide metabolism to Delta sac1 mutants, but that it also fully corrects the elevated rate of metabolic flux through the CDP-choline pathway that accompanies Sac1p deficiency in yeast.


View larger version (19K):
[in this window]
[in a new window]
  		Fig. 12.   Phosphatidylcholine synthesis in wild-type and sac1 mutant yeast strains. Yeast strains with the indicated genotypes (at bottom) were grown to mid-logarithmic growth phase in medium containing inositol (0.1 mM) and choline (1 mM). Cell cultures were then pulse-radiolabled with [32P]orthophosphate (10 µCi/ml) for 30 min at 25 °C. Phospholipids were extracted and resolved by two-dimensional paper chromatography. Radiolabeled PC was identified and quantified by phosphorimaging. PC values are given in arbitrary phosphorimager units. Total incorporation of [32P] into chloroform-soluble counts for all strains were between 10,000 and 12,000 cpm per A600 of cells (23). Data are expressed as mean ± S.D. from at least three independent experiments. Strains employed were: CTY182 (SEC14,SAC1), CTY244 (CTY182Delta sac1) transformed with YCp(URA3), and CTY244 transformed with YCp(rSAC1).

Rat Sac1 Function in Yeast Requires PI(4)P Phosphatase Activity-- Analyses of the rat Sac1 point mutants indicate that the D391N and A442V substitutions ablate the ability of rat Sac1p to complement the bypass Sec14p phenotype of Delta sac1 strains (Fig. 11A). This failure to functionally complement Delta sac1 mutations correlates with the profound decrease in the PI(4)P phosphatase activity that characterizes these mutant rat Sac1 forms (Fig. 9B). In addition, expression of either rat Sac1D391N or rat Sac1A442V fails to complement the inositol auxotrophy of Delta sac1 strains (Fig. 11C) and the signature growth defects of Delta sac1 strains at 13 °C (not shown). Interestingly, rat Sac1R480H retains the ability to substitute for Sac1p in each of these biological assays for Sac1p function in yeast (Fig. 11, A and C; data not shown), in agreement with the biochemical data that the R480H missense substitution in rat Sac1 spares phosphoinositide phosphatase activity (Fig. 9B). It is also worth emphasizing that, in yeast Sac1p, the R425H substitution (i.e. the sac1-22 allele) represents an unusual case. While the sac1-22 allele conforms to all other known sac1 alleles with respect to its association with cold-sensitive growth in yeast, it is unique in that it is the only sac1 mutation that does not impose an obligate inositol auxotrophy to the host yeast strain. The bypass Sec14p phenotype and the dramatic accumulation of phosphoinositide in sac1-22 yeast strains are nonetheless inositol-dependent phenotypes since these phenotypes are not manifested when sac1-22 mutants are cultured in inositol-free growth medium (23, 49).

    DISCUSSION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES

Rat Sac1 Is a Functional Homologue of Yeast Sac1p-- Detailed characterization of the physiological and biochemical aspects of Sac1p function in yeast reveals that Sac1p modulates a complex regulatory network of interactions that integrate inositol lipid metabolism with the metabolism of other lipids, with the dynamics and organization of the actin cytoskeleton, and with secretory pathway function. As such, Sac1p occupies a central niche in the regulation of lipid-meditated signaling in yeast. Yeast Sac1p is somewhat unusual among phosphoinositide metabolizing enzymes in that it is an integral membrane protein (19). In addition, the functional importance of yeast Sac1p in regulating lipid signaling processes raises the question of whether higher eukaryotes similarly employ integral membrane Sac1 homologues as central regulators of lipid signaling pathways. Our results indicate that proteins with considerable homology to yeast Sac1p are indeed conserved throughout the eukaryotic kingdom, and the first identification and characterization of a higher eukaryotic Sac1 is reported herein.

The first line of evidence that rat Sac1 is an authentic yeast Sac1p homologue is that both proteins are very similar in size, and that these polypeptides share substantial primary sequence identity. More importantly, rat Sac1 resembles yeast Sac1p in that it exhibits a candidate COOH-terminal transmembrane domain, and fractionates as an integral membrane protein in both mammalian and yeast cells. Subcellular fractionation experiments and indirect immunofluorescence data independently demonstrate that rat Sac1 localizes to the mammalian endoplasmic reticulum, and the COOH-terminal transmembrane domain is necessary and sufficient to direct rat Sac1 to this compartment.

The second line of evidence that rat Sac1 and yeast Sac1p are genuine homologues stems from our demonstration that these proteins share conservation in their enzymatic activities. The rat Sac1 cytoplasmic domain exhibits intrinsic phosphoinositide phosphatase activity, and the mammalian enzyme exhibits a substrate specificity that is essentially indistinguishable from that of yeast Sac1p. Specifically, rat Sac1 acts on PI(3)P, PI(4)P, and PI(3,5)P2, but not PI(4,5)P2, in vivo and in vitro (Figs. 9 and 10). As this rather unique substrate specificity is conserved in the yeast and mammalian Sac1 proteins, we presume this substrate specificity is an important feature of Sac1 protein function in vivo. In this regard, we note that both S. cerevisiae Sac1p and rat Sac1 elaborate considerable phosphatase activity against PI(3,5)P2, in addition to their potent PI(4)P phosphatase activities (26, 27; Figs. 7 and 8). PI(3,5)P2 is detected in yeast, plant, and animal cells, and intracellular levels of this phosphoinositide rise dramatically when cells are subjected to osmotic stress (58). It is tempting to speculate that Sac1p modulates the interplay between PI(3,5)P2 and the actin cytoskeleton rearrangements that accompany yeast cell responses to osmotic stress. Such a regulatory Sac1 function may help attenuate phosphoinositide signaling-mediated actin rearrangements once cells have successfully adapted to osmotic challenge.

Finally, yeast Sac1p and rat Sac1 share functional similarity in vivo. This conclusion derives from our finding that point mutations which compromise yeast Sac1p phosphoinositide phosphatase activity evoke qualitatively similar effects when placed in the context of the rat protein. Mutant rat Sac1 proteins harboring missense substitutions that correspond to those present in yeast the sac1-8 and sac1-10 alleles are defective in phosphoinositide phosphatase enzymatic activity (Fig. 9). These biochemical defects render rat Sac1 incapable of fulfilling yeast Sac1p functions in vivo, as judged by the failure of rat sac1-8 and rat sac1-10 alleles to complement the bypass Sec14p, inositol auxotrophy, and "cold sensitivity for growth" phenotypes associated with sac1 strains. By contrast, the rat sac1 allele that corresponds to the yeast sac1-22 mutation spares the phosphatase activity and does not compromise the ability of rat Sac1 to faithfully execute Sac1p functions in yeast. Accordingly, the yeast sac1-22 gene product behaves differently from the yeast sac1-8 and sac1-10 gene products, and appears to be a strongly defective phosphoinositide phosphatase only when cells are cultured in the presence of exogenous inositol (23, 49). The basis for this property remains to be investigated.

The close correlation between the phosphoinositide phosphatase activities and the biological functions of rat Sac1, as assessed by the capacity to complement sac1 phenotypes in budding yeast, strongly suggests this enzyme activity is obligatorily required for the biological function of yeast Sac1p. In particular, the PI(4)P phosphatase activity is emphasized. In vivo experiments indicate that the rat sac1-8 and rat sac1-10 proteins (rat Sac1D391N and rat Sac1A442V, respectively), while nonfunctional in yeast and ablated for PI(4)P phosphatase activity, nonetheless retain significant activity against PI(3)P and PI(3,5)P2 (Fig. 9). These data highlight two issues: (i) that these missense substitutions elicit a rather specific inactivation of the PI(4)P phosphatase activity without dramatically compromising the ability of Sac1 protein to recognize and hydrolyze phosphoinositides at the 3- and 5-positions of the inositol ring, and (ii) that elevated PI(4)P is itself somehow associated with the sac1 phenotypes emphasized in this study. This latter issue is an important one because, while the highly elevated levels of PI(4)P that accumulate in sac1 strains have previously been proposed to form the basis of sac1-associated phenotypes in yeast (particularly the bypass Sec14p phenotype; Refs. 23, 26, 55, and 59), the formal possibility that the accompanying accumulation of either PI(3)P and/or PI(3, 5)P2 contributes to (or actually determines) these phenotypes has not been excluded. So far, the data reported herein strongly reinforce the notion that PI(4)P accumulation is indeed a major biochemical defect that underlies many sac1 phenotypes. Whether the various phenotypic effects are directly related to PI(4)P accumulation is unclear. Elevation of PI(4)P in sac1 yeast evokes multiple secondary derangements in the metabolism of lipids that do not contain inositol head groups, and PI(4)P accumulation is insufficient to evoke bypass Sec14p in sac1 mutants (23).

While the phosphoinositide phosphatase activities of rat Sac1 and yeast Sac1p no doubt constitute major aspects of their physiological functions, other evidence suggests that Sac1 phosphoinositide phosphatase activity may not fully account for the biological functions of individual Sac1 domains. We have expressed in yeast a chimeric protein where the Sac1 domain of rat synaptojanin 1 is fused to the transmembrane and carboxyl-terminal domain of rat Sac1. This chimera fails to rescue sac1-associated phenotypes (not shown). This is an unanticipated result because the Sac1 domain of synaptojanin 1 has a phosphoinositide phosphatase activity with the same substrate specificity as yeast Sac1p and rat Sac1 (26, 27).

Phosphoinositide Metabolism and the Endoplasmic Reticulum Functions-- Our demonstration that rat Sac1 localizes to the endoplasmic reticulum, with a topology that exposes the catalytic domain of this protein to the cytoplasm, raises the question as to which phosphoinositide kinases produce the phosphoinositide pools that serve as substrates for rat Sac1. There is indirect evidence to suggest that Sac1p acts on a pool of PI(4)P produced by Pik1p, one of the two PI 4-kinases in yeast (23, 59). It is unlikely that Sac1p action is restricted to this pool because a pik1-101ts mutation (60) elicits only modest effects on PI(4)P accumulation in sac1 mutant yeast.2 In mammals, PI 4-kinase alpha , which is structurally similar to the yeast Stt4p PI 4-kinase, is localized to the endoplasmic reticulum while PI 4-kinase beta , which resembles yeast Pik1p, is localized to the Golgi complex (61). Based on the localization of rat Sac1, we speculate that the PI(4)P pool generated by PI 4-kinase alpha  likely provides the major source of physiological substrate for rat Sac1 action.

What physiological functions of the endoplasmic reticulum might be subject to Sac1-mediated regulation? Based on reconstitution experiments, Sac1p has been shown to be a stimulatory component for ATP uptake into the endoplasmic reticulum in yeast (20). Clearly, such an activity must exist in mammalian endoplasmic reticulum membranes as well. Thus, mammalian Sac1 proteins may recapitulate the function of yeast Sac1p in facilitating chaperone-mediated protein folding in the lumen of the endoplasmic reticulum (20, 21). Present data, however, suggest that yeast Sac1p is not itself an ATP transporter. Rather, it appears to be a regulatory factor required for optimal transporter activity (21). Thus, yeast Sac1p appears to be a multifunctional protein, and the possibility that rat Sac1 also stimulates ATP import into the mammalian endoplasmic reticulum in a manner that is independent of its phosphoinositide phosphatase activities must now be investigated.

From the perspective of phosphoinositide metabolism, a pool of PI(4)P generated in a Pik1p-dependent manner is somehow required for Golgi secretory function in yeast (59, 60), and PI(4)P may also be involved in stimulating protein transport from the endoplasmic reticulum in mammalian cells. In this regard, Cleves et al. (16) demonstrated that the sac1-6 allele exhibits negative genetic interactions with mutations in six SEC genes, all of which are implicated in protein transport between the endoplasmic reticulum and the Golgi complex. The finding that PI 4-kinase beta  drives synthesis of a PI(4,5)P2 pool required for maintaining the structural integrity of the mammalian Golgi complex (62) suggests a mechanism for how Sac1-mediated regulation of phosphoinositide pools might contribute to the activity of trafficking pathways from the endoplasmic reticulum to the Golgi complex. Sac1-mediated phosphoinositide degradation may also play a more direct role in the maintenance of endop