Characterization of an ADP-ribosylation Factor-like 1 Protein inSaccharomyces cerevisiae *

ADP-ribosylation factors (ARFs) are highly conserved ∼20-kDa guanine nucleotide-binding proteins that enhance the ADP-ribosyltransferase activity of cholera toxin and are believed to participate in vesicular transport in both exocytic and endocytic pathways. Several ARF-like proteins (ARLs) have been cloned fromDrosophila, rat, and human; however, the biological functions of ARLs are unknown. We have identified a yeast gene (ARL1) encoding a protein that is structurally related (>60% identical) to human, rat, and Drosophila ARL1. Biochemical analyses of purified recombinant yeast ARL1 (yARL1) protein revealed properties similar to those ARF and ARL1 proteins, including the ability to bind and hydrolyze GTP. Like other ARLs, recombinant yARL1 protein did not stimulate cholera toxin-catalyzed auto-ADP-ribosylation. yARL1 was not recognized by antibodies against mammalian ARLs or yeast ARFs. Anti-yARL1 antibodies did not cross-react with yeast ARFs, but did react with human ARLs. On subcellular fractionation, yARL1, similar to yARF1, was localized to the soluble fraction. The amino terminus of yARL1, like that of ARF, was myristoylated. Unlike Drosophila Arl1, yeastARL1 was not essential for cell viability. Like rat ARL1, yARL1 might be associated in part with the Golgi complex. However, yARL1 was not required for endoplasmic reticulum-to-Golgi protein transport, and it may offer an opportunity to define an ARL function in another kind of vesicular trafficking, such as the regulated secretory pathway.


MATERIALS AND METHODS
Yeast Culture-Yeast culture media were prepared as described by Sherman et al. (44). YPD and YPGal contained 1% Bacto-yeast extract, 2% Bacto-peptone, and 2% glucose or 2% galactose, respectively; SD contained 0.7% Difco yeast nitrogen base (without amino acids) and 2% glucose. Nutrients essential for auxotrophic strains were supplied at specified concentrations (44). Sporulation, growth, and mating were carried out as described (45). Yeast cells were transformed by the lithium acetate method (46). Plasmids were constructed by standard protocols (47). Yeast strains are listed in Table I.
Polymerase Chain Reaction-Unless otherwise specified, the protocol used for PCR amplification was 35 cycles of 1 min at 95°C, 1 min at 52°C, and 1 min at 72°C, followed by extension at 72°C for 10 min, in a solution containing 50 mM KCl, 10 mM Tris-Cl (pH 8.3), 1.5 mM MgCl 2 , 0.01% gelatin, 20 mM each dNTP, 0.1% Tween, 25 pmol of each amplification primer, and 2.5 units of Taq polymerase (total volume of 100 ml). Samples of reaction mixtures were subjected to electrophoresis o a 1.5% agarose gel. All PCR products were purified, subcloned, and sequenced by the dideoxy chain termination method (48). DNA sequences were analyzed, and multiple protein alignments were prepared using a GeneWorks software package (IntelliGenetics, Inc./Betagen, Mountain View, CA).
Isolation of Yeast ARL1 cDNA-Yeast ARL1 was isolated initially by PCRs from a yeast gt11 cDNA library. Five PCRs were used to obtain cDNA segments and to assemble a composite sequence of the full-length coding region. Primers 1222 (5Ј-TTGACACCAGACCAACTGGTAATG-3Ј), 1222B (5Ј-ACCGGCGCTCAGCTGGAATT-3Ј), 1218 (5Ј-GGTGGC-GACGACTCCTGGAGCCCG-3Ј), and 1218A (5Ј-CGTCAGTATCGGCG-GAATTC-3Ј) are complementary to sequences immediately upstream or downstream of the EcoRI insertion site of vector gt11. Degenerate nucleotides (ARL1-R1 and ARL1-R2) correspond to part of the consensus sequences VWDLGGQD and TSIRPYWR in human, rat, and Drosophila ARL1. The open reading frame was completed using the rapid amplification of 5Ј-cDNA ends procedure (49). A yeast gt11 cDNA library (CLONTECH) served as template in the one site-specific PCR used to capture 3Ј-and 5Ј-ends as described previously (10,11).
Yeast ARL1 Gene Disruption-Yeast ARL1 DNA generated by PCR was subcloned into the pGEM-7Zf plasmid, resulting in pGyL1. The yeast URA3 gene was inserted at the single EcoNI site in the yeast ARL1 gene as follows (see Fig. 4A). The 3.8-kb DNA fragment containing the yeast URA3 gene and two hisG repeat sequences was excised from the plasmid pNKY51 (50) by digestion with BglII and BamHI; the 5Ј-overhangs were filled in with Klenow fragment. Plasmid pGyL1, containing the yeast ARL1 gene, was linearized at the internal EcoNI site; the overhang ends were filled in with Klenow fragment, and the cDNA was ligated to the 3.8-kb hisG-URA3-hisG fragment, resulting in pGyL1U.
Gene disruption mutants were constructed by a one-step gene replacement method (51). Briefly, the ϳ4.8-kb DNA fragment excised from pGyL1U by digestion with XhoI and BamHI was used to transform various Ura Ϫ strains (see Table I), and uracil prototrophs were selected. DNA blot analysis of the URA ϩ cells confirmed that the yeast ARL1 gene contained an additional 3.8 kb, corresponding to the hisG-URA3-hisG gene. Elimination of URA3 and one hisG repeat was carried out as described previously (11). Double deletions of yeast ARL1 and yeast ARF3 were performed in the yeast arl1 mutants (arl1::hisG, ura3) and arf3 mutants (arf3::hisG, ura3) that are listed in Table I.

Expression and Purification of Recombinant Proteins-
The open reading frame of yeast ARL1 was obtained by PCR using primers that incorporated unique NdeI and BamHI sites at the initiating methionine and 6 base pairs 3Ј of the stop codon, respectively. For the His-tagged yARL1 fusion protein, a DNA fragment containing the yeast ARL1 coding region was generated by amplifying yeast genomic DNA with sequence-specific primers. The PCR product was purified and annealed to the expression vector pET15b (Novagen), yielding pET15byL1. For the nonfusion protein, PCR products were digested with NdeI and BamHI, purified, and annealed to the expression vector pT7 (52), yielding pT7yARL1. BL21(DE3) cells containing expression plasmids were grown to a density of A 600 ϭ 1.0, at which time the inducer, isopropyl-1-thio-␤-D-galactopyranoside, was added to a final concentration of 1 mM. After 3 h, cells were harvested by centrifugation, washed once in 20 mM Tris (pH 7.4) and 1 mM EDTA, and stored at Ϫ80°C until used. For large-scale production of recombinant proteins, 5 ml of overnight culture were used to inoculate 1 liter of LB broth containing ampicillin (100 g/ml), followed by shaking at 37°C. When A 600 reached 0.6 -0.8, protein production was induced with 0.5 mM isopropyl-1-thio-␤-D-galactopyranoside for 3 h, and bacteria were collected by centrifugation and stored at Ϫ20°C. Cell pellets were suspended in 10 ml of phosphatebuffered saline (pH 7.4) containing lysozyme (0.5 mg/ml) and disrupted by sonification. The lysate was centrifuged after the addition of Triton X-100 to 1%, and His-tagged fusion protein was isolated on Ni 2ϩnitrilotriacetic acid resin (QIAGEN Inc., Chatsworth, CA) by standard methods. Purity was assessed by SDS-PAGE and staining with Coomassie Blue. Protein was quantified by Coomassie Blue or silver stain assays (Bio-Rad).
Polyclonal Antibody Production-The recombinant N-terminal Histagged fusion proteins (yARL1, yARF1, yARF3, yARP1, hARL1, hARL2, hARL3, mouse ARL4, hARL5, and carboxypeptidase Y) were synthesized in Escherichia coli using the pET15b expression plasmid, isolated on Ni 2ϩ -nitrilotriacetic acid resin, and further purified by SDS-PAGE. Denatured purified proteins from the SDS-PAGE gel were used as antigens to raise polyclonal antibodies in rabbits essentially as described (53).
Western Blot Analysis-Yeast cell concentration was assessed by absorbance at 600 nm; cells were suspended in radioimmune precipitation assay buffer (50 mM Tris-HCl (pH 8.0), 0.1% SDS, 0.5% deoxycholic acid, 150 mM NaCl, and 1% Nonidet P-40) to a final A 600 of 30. Whole cell extracts were then prepared by vortexing with glass beads for 2 min at 4°C and clarified by a brief centrifugation. After SDS-PAGE, proteins were transferred electrophoretically to Immobilon-P membranes (Millipore Corp.). Incubation with antibodies was carried out in phosphate-buffered saline (pH 7.4) containing 0.1% Tween 20 and 5% dried skim milk at room temperature for 60 min. The monoclonal anti-HA antibody (HA.11, Berkeley Antibody Co., Richmond, CA) and horseradish peroxidase-conjugated goat anti-mouse IgG ϩ IgM (H ϩ L) were each diluted 1:1000. Bound antibodies were detected with the ECL system (Amersham Corp.) according to the manufacturer's instructions. Primary and secondary antibodies and luminol substrate were removed from the blot using the blot-stripping protocol (Amersham Corp.).
To synthesize 3 H-labeled myristoylated yARL1 in yeast, [ 3 H]myristic acid was added to yeast culture as described by Simon and Aderem (56). Briefly, yeast cells were grown to a density of 2 ϫ 10 7 cells/ml at 25°C in a rotary air shaker, harvested by centrifugation, and suspended at a density of 2 ϫ 10 8 cells/ml. Cerulenin (10 mg/ml in dimethyl sulfoxide) was added to a final concentration of 2.5 g/ml. After 10 min, [ 3 H]myristic acid was added to a final concentration of 30 Ci/ml, and cells were incubated for 1 h at 25°C in a rotary shaker. Cells were collected and washed with 10 mM NaN 3 in double distilled H 2 O. Glass beads (300 ml) and 300 ml of lysis buffer (50 mM Tris-Cl (pH 7.5), 1% SDS, 1 mM EDTA, and 1 mM phenylmethylsulfonyl fluoride) were added, and the mixture was agitated vigorously for 90 s at room temperature before immersion in a boiling water bath for 6 min. Immunoprecipitation, electrophoresis, and autoradiography were done essentially as described (57) using 10 l of anti-yARL1 or anti-yARF1 serum prepared in our laboratory.
Protein Processing in the Golgi Apparatus-Yeast cells were grown at 24°C overnight in selective minimal medium containing 200 mM (NH 4 ) 2 SO 4 to an A 600 of 0.5. After 1 h at 37°C, cells were transferred to sulfate-free selective minimal medium (final A 600 ϭ 5) and incubated for 5 min at 37°C before the addition of 30 Ci of Pro-mix L-[ 35 S]-label (blend of [ 35 S]methionine and [ 35 S]cysteine, 14.3 mCi/ml) per A 600 unit. After 5 min, labeling was terminated by the addition of 1% (v/v) chase solution (0.3% (w/v) cysteine, 0.4% (w/v) methionine, and 100 mM (NH 4 ) 2 SO 4 ). Samples were removed at the indicated times thereafter and added to equal volumes of ice-cold 20 mM NaN 3 in double distilled H 2 O. Cells were collected, and cell lysates were prepared as described above. Immunoprecipitation, electrophoresis, and autoradiography were done essentially as described (57) using 10 l of anti-carboxypeptidase Y serum prepared in our laboratory.
Nucleotide Binding and Hydrolysis-Binding of GTP␥S to purified recombinant yARL1 was determined by a filter trapping method (58). Unless otherwise specified, 1 g of His-tagged yARL1 fusion protein was incubated at 30°C in 20 mM HEPES (pH 7.5), 100 mM NaCl, 1 mM dithiothreitol, 1 mM EDTA, 0.5 mM MgCl 2 , 20 g/ml bovine serum albumin, and 10 M [ 35 S]GTP␥S (Ͼ1000 Ci/mmol; Amersham Corp.) without or with 3 mM sonified DL-␣-dimyristoylphosphatidylcholine and 2.5 mM (0.1%) sodium cholate in a final volume of 50 l. Duplicate or triplicate samples were transferred to 2 ml of ice-cold 20 mM Tris-Cl (pH 7.4), 100 mM NaCl, 10 mM MgCl 2 , and 1 mM dithiothreitol before rapid filtration on 0.45-m HA filters (Millipore Corp.). The amount of nucleotide bound to the fusion protein was quantified by scintillation counting. Data were fitted to a first-order rate equation.
GTP hydrolysis was determined by binding [␣-32 P]GTP to 5.0 M recombinant yARL1 protein as described by Randazzo and Kahn (59), followed by its dilution (1:9) into 25 mM HEPES (pH 7.4), 100 mM NaCl, 2.5 mM MgCl 2 , 0.1% Triton X-100, 1 mM dithiothreitol, and 1 mM GTP with bovine brain phosphoinositides (1 mg/ml) and incubation at 30°C. Every 5 min, samples were transferred to 2 ml of ice-cold 20 mM Tris-Cl (pH 7.4), 100 mM NaCl, 10 mM MgCl 2 , and 1 mM dithiothreitol. The conversion of GTP to GDP was determined by thin-layer chromatography as described (59). A blank without protein was used to determine background, which was subtracted from each experimental value.
Subcellular Fractionation by Velocity Sedimentation on Sucrose Density Gradients-Cells were harvested by centrifugation from cultures (50 ml) grown in YPD medium to midexponential phase (A 600 ϭ 1). Cells (ϳ0.5 g) were washed by repeated suspension in ice-cold H 2 O and centrifugation, incubated with lyticase to form spheroplasts, suspended in 0.2 ml of ice-cold lysis buffer (10 mM Tris-HCl (pH 7.4), 1 mM EDTA, and 12.5% sucrose) containing protease inhibitors (1 mg/ml aprotinin, 1 mg/ml leupeptin, 1 mg/ml pepstatin, 1 mM benzamidine, and 1 mM phenylmethylsulfonyl fluoride), and disrupted on ice with 20 strokes in a Dounce homogenizer. The cell lysate was centrifuged (1500 ϫ g) for 5 min to remove unbroken cells and large cellular debris. For gradient fractionation of cell organelles, 0.8 ml of the clarified supernatant were loaded on top of a manually generated five-step sucrose gradient (0.8 ml each of 60, 50, 40, 30, and 20% sucrose in lysis buffer) and then subjected to centrifugation at 37,000 rpm (ϳ170,000 ϫ g) for 3 h at 4°C in a Beckman SW 55 rotor. Ten fractions were collected manually from the bottom. Samples (12 l) of fractions were boiled for 5 min in SDS-PAGE sample buffer containing 5% SDS, subjected to SDS-PAGE, and analyzed by immunoblotting.
Indirect Immunofluorescence-Cells were grown in 5 ml of minimal selective medium with 2% glucose to a cell density of 1-2 ϫ 10 7 cells/ml and prepared for indirect immunofluorescence as described (60) with the following modifications. To each culture was added 0.6 ml of 37% formaldehyde for fixation, and the cultures were gently shaken at 30°C for 60 min. Cells were collected by centrifugation (2500 ϫ g, 5 min), washed once in 5 ml of solution P (1.2 M sorbitol and 0.1 M potassium phosphate (pH 6.5)), suspended in 1 ml of solution P, transferred to a microcentrifuge tube, and washed again. The cell pellet was suspended in 1 ml of solution P and incubated at 30°C for 30 min with 5 l of lyticase (10,000 units/ml, in solution P) and 1 l of ␤-mercaptoethanol. The cells were collected by centrifugation (3000 ϫ g, 5 min), washed in solution P, and suspended in 100 -200 l of solution P. Samples (20 l) of cells were placed in each well of a multiwell slide that had been coated with 0.1% polylysine. Following aspiration of excess cells, the slide was immersed in methanol (Ϫ20°C) for 6 min and then transferred to acetone (Ϫ20°C) for 30 s. Cells were washed once in solution P and incubated for 1 h with antibody blocking buffer (100 mM Tris-HCl (pH 9.0), 150 mM NaCl, 5% nonfat milk, and 0.1% Tween 20), followed by overnight incubation with the primary antibody in antibody blocking buffer. After primary and secondary antibody incubations, cells were washed extensively with a buffer containing 100 mM Tris-HCl (pH 9.0) and 150 mM NaCl and with a buffer containing 100 mM Tris-HCl (pH 9.5), 100 mM NaCl, and 50 mM MgCl 2 . Affinity-purified rabbit anti-yARL1, mouse monoclonal anti-HA (12CA5), and fluorescein isothiocyanate-conjugated secondary antibodies (Cappel) were diluted 1:1000 for use. Fluorescence microscopy was performed with a Nikon Microphot SA microscope. Cells were viewed at a magnification of ϫ1000. Exposure times for immunofluorescence photographs were 15 and 30 s.

RESULTS AND DISCUSSION
Analyses of Protein Sequences of Yeast ARL1-The yeast cDNA homologue of human and rat ARL1 (yARL1) was identified as a product of PCR amplification using degenerate probes derived from conserved sequences in members of the ARL family (see "Materials and Methods"). The full-length open reading frame of yARL1 (549 bases) encodes a protein of 183 amino acids. Alignment of the deduced amino acid sequence of yARL1 and other related ARFs and ARLs revealed that yARL1 is more identical (61%) to mammalian ARL1 than to ARFs (48 -56%) or other ARLs (35-45%) ( Table II). The entire S. cerevisiae genome sequence data base was searched for mammalian ARL sequences; yARL1 was the only ARL homologue identified. Like other ARL1 proteins, yARL1 has a glycine at position 2, the site of N-myristoylation in ARF proteins. In addition, yARL1 lacks cysteine residues near the carboxyl terminus, which are sites of isoprenylation in non-ARF members of the Ras superfamily. Alignment of the yARL1 protein with other ARL and ARF proteins is shown in Fig. 1 These are residues and sequences that are thought to be involved in binding of the guanine nucleotide and magnesium ion or in protein-protein interactions and are very highly conserved in the ARF family.
Specific Immunoreactivity of Antibody against yARL1-Recombinant proteins were purified, and antibodies were prepared as described under "Materials and Methods." Purified His-tagged fusion ARLs (yARL1, hARL1, hARL3, and hARL5), yARP1, 2 , and ARFs (yARF1, yARF2, and yARF3) (ϳ30 -40 ng) were subjected to SDS-PAGE under reducing conditions, followed by transfer of proteins to polyvinylidene difluoride membranes. Recombinant proteins were visualized by silver staining ( Fig. 2A). At a dilution of 1:5000, the polyclonal antibody specific for yARL1 did not cross-react with yARF1, yARF2, yARF3, or yARP1 (Fig. 2B). It reacted, however, with human ARLs (hARL5 Ͼ hARL1 Ͼ hARL3). In addition, polyclonal antibodies against yARF1, yARF2, yARF3, and yARP1 failed to react with yARL1 on Western analysis (data not shown). Thus, yARL1 was immunologically distinguishable from yeast ARFs, and the anti-hARL1 antibody proved to be a sensitive and specific probe for yARL1 protein. Immunoblotting with this antiserum allowed detection of yARL1 at nanogram levels (ϳ1-2 ng), whereas no signal was detected with recombinant yARF1 and yARF2 (up to 100 ng) or yARF3 (up to ϳ40 ng).
DNA and RNA Blot Analyses-Yeast ARL1 is a single copy gene and is located on chromosome II (Yeast Genome Data Bank). Total RNA from mid-log growth yeast cultures in either glucose-or galactose-containing medium was subjected to electrophoresis, transferred to GeneScreen Plus, hybridized with the yARL1 cDNA probe, and, after stripping, with a yeast ␤-tubulin probe. The ϳ0.75-kb yARL1 RNA, similar to that of yARF1, was not repressed by growth in glucose (Fig. 3).
Yeast ARL1 Is Not an Essential Gene-To learn more about the function of yARL1, the gene and flanking regions were isolated from genomic DNA using PCR, and a construct was created in which its open reading frame was disrupted by the URA3 marker gene (Fig. 4A). A DNA fragment containing the ARL1::hisG-URA3-hisG sequence was used to transform ura3/ ura3 diploid yeast (SEY6210.5) (Table I) (51). Ura ϩ transformants were isolated and used to confirm the correct replacement of one of the two genomic copies of yARL1. DNA blot analysis confirmed that one copy of the yeast ARL1 gene contained the 2.1-kb (EcoRI/BamHI-digested) or 2.3-kb (EcoRIdigested) fragment, whereas disrupted yeast ARL1 contained 5.9-and 6.1-kb bands, corresponding to the insertion of the 3.8-kb (hisG-URA3-hisG) fragment into 2.1-and 2.3-kb fragments (Fig. 4B). The verified heterozygous diploids were then subjected to sporulation and tetrad dissection. On germination at 30°C, most diploid cells gave rise to four viable spores. As each of the strains (haploid) containing the arl1 disruption (hisG-URA3-hisG) was viable, yeast ARL1 is not an essential gene. Ura ϩ spores, but not Ura Ϫ , were also confirmed the correct replacement of yARL1 (Fig. 4B).
Three ARF genes are present in S. cerevisiae, and two of them (ARF1 and ARF2) are believed to participate in vesicular trafficking in the Golgi system. The double deletion of yeast  ARF1 and ARF2 is lethal, but each of the five known human ARF proteins can restore vegetative growth to this double deletion mutant (61)(62)(63). Cells with double deletions of yeast ARL1 and ARF3 or ARL1 and ARF1 were also viable (data not shown). Proper disruption of the specific genes was confirmed by PCR on genomic DNA prepared from colonies of the mutants. This result confirmed that yeast ARL1, unlike Drosophila Arl1 (38), is not essential for cell viability, and its deletion was not being complemented by yeast ARF1 or ARF3.
To assess whether yeast arl1 can affect growth phenotype, the growth rates of wild-type, arl1 mutant, and overexpressed yARL1 strains were determined. We constructed a recombinant yARL1 clone (yARL1-HA) with a 9-amino acid influenza virus HA epitope (64) fused to its C terminus, placed the HAtagged allele (yARL1-HA) under the control of the ADH1 promoter, and expressed it in wild-type and arl1 mutant yeast. The arl1 mutants and yeast overexpressing yARL1 had 15-20 and ϳ40% lower specific growth rates, respectively, than wildtype cells in glucose synthetic medium (data not shown). It is conceivable that overexpressed yARL1, similar to yARF1, can interfere with ARF (or ARL)-mediated vesicular transport (65). Yeast lysate protein (ϳ20 g) was subjected to SDS-PAGE and stained with Coomassie Blue (Fig. 5A). Expression of yARL1 was confirmed by using antibody against yARL1 (Fig. 5B). However, Western analyses with either anti-yARL1 or anti-HA antibodies detected two proteins of ϳ21 and ϳ22 kDa when yARL1 was overexpressed in wild-type or arl1 mutant strains (Fig. 5, B and C). Overexpression of yARF3 in the same system did not yield a doublet pattern. We currently do not have a good explanation for this observation. It is necessary to investigate further possible post-translational modifications of yARL1. Expression of yARL1 protein, like that of yARF1, was not carbon FIG. 2. Specific immunoreactivity of antibody against yARL1. ϳ20 -40 ng of purified recombinant His-tagged ARFs and ARLs were subjected to SDS-PAGE on a 15% gel. yARP1 (temporarily named Lpe21p in the Yeast Genome Project) is the yeast homologue of mammalian ARP1 (77). A, silver-stained gel; B, proteins transferred to nitrocellulose and reacted with anti-hARL1 antibody, followed by detection using the ECL system. Positions of protein standards indicated on the left.  lanes 1 and 2; Ura Ϫ , lanes 3 and 4), a heterozygous disrupted diploid cell SEY6210.5 ( lanes 5 and 6), and a wild-type SEY6210.5 (lanes 7 and 8). Isolated DNA was digested with EcoRI (E) or BamHI/EcoRI (B/E), separated by electrophoresis on a 1% (w/v) agarose gel using Tris borate buffer, transferred to GeneScreen Plus, and hybridized with a 32 P-labeled 584-bp fragment of yARL1. source-dependent (data not shown).
Function of yARL1 Differs from That of yARF-To evaluate the role of yARL1 in protein transport from the ER to the Golgi apparatus or within the Golgi apparatus, we measured the glycosylation and proteolytic processing of carboxypeptidase Y, an enzyme that traverses this secretory pathway en route to the vacuole. Pulse-chase labeling with 35 S-labeled cysteine and methionine of wild-type, arf1, and arl1 mutant cells was followed by immunoprecipitation of carboxypeptidase Y. The coreglycosylated P1 form of the carboxypeptidase Y proenzyme in the ER is converted to the P2 form by further glycosylation in the Golgi apparatus and finally is proteolytically processed in the vacuole to the mature form. Similar to the wild-type cells, the arl1 mutant converted carboxypeptidase Y from the ER to Golgi apparatus and vacuole forms (Fig. 6). The arf1 mutant, however, accumulated core-glycosylated carboxypeptidase Y in the P1 form, as expected (57). Therefore, yARL1 has a biological function clearly different from that of yARF1 and is not required for ER-to-Golgi protein transport.
Amino Terminus of yARL1 Is Myristoylated-ARF proteins are myristoylated in vivo, and the cellular functions of ARF involve its reversible association with specific intracellular membranes in myristoylation-dependent and guanine nucleotide-regulated reactions (66,67). Binding of the activating nucleotide (GTP) results in a conformational change in the protein with increased affinity for phospholipids and membranes (68,69). The inactive GDP-bound form of ARF is soluble (69,70). Although ARL3 has been reported not to be N-myristoylated (41), we tested whether ARL1 can be myristoylated. When yARL1 or hARL1 was coexpressed in bacteria with yeast Nmyristoyltransferase, each was myristoylated (Fig. 7). Myristoylated yARF1 served as a positive control. His-tagged yARL1, with the amino-terminal sequence MGSSHHHHHH-, was myristoylated poorly. To determine whether native yARL1 is myristoylated, wild-type yeast expressing HA-tagged yARL1 and wild-type and arl1 mutant cells were grown in medium with [ 3 H]myristic acid. Yeast cell lysates were immunoprecipitated with anti-yARL1 or anti-yARF1 antibodies. Both yARL1 and yARF1 were identified as myristoylated proteins (Fig. 7). Thus, the biological function of ARL1, similar to that of ARF, might be influenced by myristoylation. Subcellular Localization of yARL1-As the cellular localization of yARL1 could provide hints regarding its function, a cell lysate was subjected to velocity sedimentation on sucrose density gradient centrifugation. The distribution of yARL1, yARF1, and Emp47p, a Golgi marker protein (60), among the fractions was determined by Western blot analysis (Fig. 8).
Most of the yARL1, apparently in the soluble cytoplasmic form, was at the top of the gradient. Although yARF1 is known to function in the Golgi apparatus, Ͼ90% of it was in the soluble cytoplasmic fraction. Fig. 8 also shows that yARL1 comigrated with yARF1.
Since yARL1 and hence possibly yARF1 dissociate from membranes upon cell lysis, we sought to investigate the cellular localization of yARL1 by indirect immunofluorescence. Using affinity-purified anti-yARL1 antibodies, a punctate staining was observed (Fig. 9). This is reminiscent of what has been seen by others with antibodies directed against Golgi-and transport vesicle-associated proteins, such as Ypt1p (71,72), Kex2p (73), Sed5p (74), and Ypt31p (75). As shown in Fig. 9, a control experiment with antibody 12CA5 directed against the HA epitope of HA-Pmr1, which is assumed to localize primarily to the Golgi compartment (76), resulted in a picture similar to that obtained with anti-yARL1 antibodies. In both cases, neither perinuclear staining typical for ER localization nor staining of the vacuole was evident. From the combined results of subcellular fractionation and indirect immunofluorescence, we FIG. 6. Immunoprecipitation of labeled carboxypeptidase Y. Wild-type (ARF1/ARL1), arl1 mutant, and arf1 mutant cells were grown, radiolabeled for 5 min, and further treated as described under "Materials and Methods." P1 is the core-glycosylated carboxypeptidase Y found in the ER; P2 is the outer chain-glycosylated form in the Golgi apparatus; and M is the mature form that results from proteolytic processing in the vacuole. The chase times are indicated. The lysate of spheroplasts of wild-type yeast was fractionated by sucrose gradient centrifugation. Samples of fractions were subjected to SDS-PAGE, and proteins were transferred to polyvinylidene difluoride filters; yARL1, yARF1, and a Golgi marker (Emp47p) were identified with specific antibodies and detected using the ECL system by exposure to Hyperfilm-MP. Positions of protein standards are indicated on the left. Gradient fractions are numbered from the bottom. Lane W indicates the yeast total lysate. assume that at least part of yARL1 is associated with Golgi membranes, which was recently reported for ARL1 in normal rat kidney cells (43).
Biochemical Properties of Recombinant yARL1 Protein-To determine whether the yeast ARL1 gene product has ARF activity, a recombinant protein was constructed, expressed, and purified from E. coli. The His-tagged yARL1 fusion protein did not stimulate auto-ADP-ribosylation of cholera toxin A1 protein in the presence of 100 M GTP␥S, a nonhydrolyzable GTP analogue, and detergent (SDS) (data not shown). It is known that myristoylation is not absolutely required for the activation of cholera toxin A by rat ARF1.
To assess the binding of GTP␥S, a slowly hydrolyzed analogue of GTP, to yARL1, the nitrocellulose filter trapping method was used. Binding had achieved steady state in 60 min at 30°C (Fig. 10, inset). GTP␥S binding to yARL1, like that to ARF, Drosophila ARL1, and rat ARL1, was influenced by added phospholipids and was concentration-dependent ( Fig.  10) (38,43). In contrast, binding of GTP␥S to hARL2 and hARL3 was affected very little by added lipid or detergent (39,41). Maximal rates of GTP hydrolysis by hARL2 and hARL3 were 0.0074 and 0.005 min Ϫ1 , respectively (39,41). Recombinant yARL1 had a rate of ϳ0.02 min Ϫ1 , similar in magnitude to that of Drosophila ARL1 (0.05 min Ϫ1 ) (38).
ARL1 was first identified in D. melanogaster as an essential gene encoding a structural relative of the ARF family that is 55% identical to mammalian ARF1 (38). PCRs using highly degenerate primers led to the identification of cDNA fragments encoding at least five human ARL proteins and several rat and mouse homologues (38 -43). The ARL proteins are at least 40% identical to the ARF proteins, but little is known about their cellular function. We report here the identification and characterization of yeast ARL1 with the functional characteristics of an ARL1 protein that is expressed widely in Drosophila, human, and rat tissues. yARL1, at its normal level of expression, clearly cannot replace yARF1 and yARF2, as their double deletion is lethal (61). The specific antibody described should prove useful in defining the cellular role(s) of ARL1 and in distinguishing among the many structurally related members of the ARF family. There are still many gaps in our understanding of the roles of ARL proteins in membrane budding and fusion processes. The newly identified yARL1, which is apparently associated with the Golgi apparatus, is not essential for ER-to-Golgi protein transport. Thus, it offers an opportunity to define its function in another kind of membrane vesicular transport, such as the regulated secretory pathway. FIG. 9. Indirect immunofluorescence to localize yARL1. Wildtype cells (YPH250) transformed with pl161 coding for the HA epitope of HA-Pmr1 and wild-type and arl1 mutant cells were fixed with formaldehyde. Spheroplasts were prepared and treated with affinity-purified polyclonal anti-yARL1 antibody (arl1 mutant and ARL1 panels) or monoclonal 12CA5 antibody against the HA epitope of the Golgi marker HA-Pmr1 (HA-Pmr1p panel). Phase images of the cells are on the right, and indirect immunofluorescence images on the left.