1-Galactosidase Fusion - Europe PMC

MOLECULAR AND CELLULAR BIOLOGY, Dec. 1985, p. 3436-3442 0270-7306/85/123436-07$02.00/0 Copyright C) 1985, American Society for Microbiology

Vol. 5, No. 12

Posttranscriptional Regulation and Assembly into Ribosomes of a Saccharomyces cerevisiae Ribosomal Protein1-Galactosidase Fusion LINDA GRITZ, NADJA ABOVICH, JOHN L. TEEM,t AND MICHAEL ROSBASH*

Department of Biology, Brandeis University, Waltham, Massachusetts 02254 Received 10 July 1985/Accepted 16 September 1985

To study the regulation of ribosomal protein genes, we constructed a 'lacZ fusion of the Saccharomyces cerevisiae RPSIA gene, containing the first 64 codons of RPSA. In a strain lacking an intact RPSIA gene (cells are viable due to the presence of an active RPSIB gene), f-galactosidase activity is 10-fold greater than in a strain containing RPS1A. RPSIA-lacZ mRNA levels are equal in the two strains, indicating that regulation is posttranscriptional. In the absence of the RP51A gene, the fusion protein is predominantly cytoplasniic and associated with polysomes, whereas in the presence of RPS1A, the fusion protein is predominantly nuclear, and none is associated with polysomes. Deletions were made in the RP51A-coding portion of the fusion gene. The most extensively deleted gene, containing only the first seven RP51A codons fused to lacZ, produced a high level of I-galactosidase activity in both the presence and the absence of the RP51A gene. In both cases, little or none of this shorter fusion protein was found associated with polysomes. Thus, a regulatory site (or sites) lies in the protein-coding region of RP51A. We suggest that posttranscriptional regulation of the rp5l fusion protein is related to assembly of the protein into ribosoines.

The coordinate production of ribosomal proteins (rproteins) poses an interesting problem in intracellular regulation. More than 70 proteins are coordinately synthesized with respect to each other, in relation to the synthesis of several species of rRNA, and in response to the varying demand for ribosomes under different environmental conditions. In Escherichia coli, coordinate synthesis is achieved, at least in part, by the arrangement of r-protein genes in polycistronic operons and regulation by autogenous control at the translational (9) and possibly transcriptional (18) levels. In eucaryotes, r-protein gene expression is probably more complex. Most r-protein genes are unlinked (6, 8, 10, 32), exist in multiple copies (6, 10, 20, 32), and contain introns (4, 10, 24). In addition, transcription, translation, and ribosome assembly occur in different compartments of the eucaryotic cell. Many Saccharomyces r-protein genes have been cloned (3, 10, 32) and are being used to study r-protein expression and regulation. When extra copies of the cloned teml gene encoding rpl are introduced into yeast on a multicopy plasmid, the steady-state level of mRNA increases three- to fourfold, but tcml protein synthesis is only slightly increased. Thus, dosage compensation of the tcml gene has been proposed to occur at the translational level (23). To facilitate the analysis of rpSl gene expression, the Saccharomyces cerevisiae RP51A gene has been fused to the E. coli lacZ gene. We have analyzed a number of integrated and episomal copies of RP51A-lacZ fusion genes in various strains designed to express different levels of rp5l. The data indicate that fusion protein levels are modulated by the level of expression of the RPSlA gene. We propose a regulatory mechanism involving ribosomal assembly.

MATERIALS AND METHODS Strains and media. The haploid S. cerevisiae DBY745 (ox adel-101 leu2-3 leu2-112 ura3-52) was constructed as described previously (5). PB12 is a derivative of DBY745 containing the LEU2 gene in place of the RPSIA gene (2). HR125-5D (a leu2-3 leu2-112 ura3-52 trpl his3 his4) was constructed as described previously (R. Jensen, Ph.D. dissertation, University of Oregon, Eugene, 1983). Yeast cells were grown in synthetic complete medium containing either 2% glucose or 2% galactose as described previously (27). Medium containing 5-bromo-4-chloro-3indolyl-,3-D-galactopyranoside was prepared as described previously (25). Plasmid and strain constructions. Plasmids pHZ18 (30) and pZA64 (previously designated pZA50; J. L. Teem, Ph.D. dissertation, Brandeis University, Waltham, Mass., 1983) were constructed as described previously. pZA64 is identical to pHZ18 except for the 5' end of the 'lacI-Z fusion gene; pZA64 contains the promoter and transcriptional and translational initiation sites of RP51A, whereas pHZ18 contains the GALIO UAS and CYCI transcriptional and translational initiation sites. YEp24 (5) and pLGSD5 (14) were as described previously. pPP-Z was a generous gift from H. Fried. pYE51B-1 is a derivative of pYE51B (2) containing the LEU2 gene. DNA cleavage, ligation, and transformation were performed as described previously (11, 12). Plasmids pHZ18 zA2p.m and pZA64A2p.m were constructed by partial EcoRI digestion and religation to remove the 2 ,im DNA portions of pHZ18 and pZA64. pHZ18MA2pm and pZA64A2V.Lm were cleaved with Sall which, upon transformation of DBY745, directed integration (22) of these plasmids at the homologous Sall site in the intron of the genomic copy of RP51A, creating strains HZint and ZAint, respectively (Fig. 1). The predicted structures of the RP51A locus in these strains were confirmed by Southern hybridization (data not shown). pZA7 was constructed by partial BAL 31 digestion of the

* Corresponding author. t Present address: Whitehead Institute, Cambridge, MA 02139.

3436

RIBOSOMAL

VOL. 5, 1985

PROTEIN-P-GALACTOSIDASE FUSIONS

3437

Im

pZA64

a

LP,/

;T/4

1. Remove 2F' DNA 2. Digest with Sal I 3. Transform DB745

GAL UAS

RP51A-lacZ

CYCI-RP51A

-

HZint:

Salt%

Sa I

Sal I

I ~~~~~~~~~~~~~~~~~~~~~

t ZAint I =~~~~~~~~~~~~ Sall *Sall

5'RP51A

RP5lA-locZ

5'RP51A

RP51A

:

FIG. 1. Construction of a galactose-inducible, glucose-repressible. CYCI-RP51A fusion gene. The 2p.m portion of pHZ18 was removed by partial EcokI digestion and religation to generate pHZ18M2p.m. Strain DBY745 was then transformed with pHZ18A2pLm cleaved in the RP5IA intron with Sall to direct integration of the plasmid at the homologous Sall site of the genomic RP51A. This integration event effected an exchange of 5' ends of the fusion and wild-type genes, resulting in an integrated RP5IA-lac(Z fusion gene with the genomic RP51A promoter adjacent to a CYCI-RP5JA fusion gene with a GAL-CYCl hybrid promoter. pZA64 was integrated in the same manner for comparison with , plasmid DNA; iizi, coding region. the galactose-inducible gene. --, chromosomal DNA;

RP5IA coding region of pZA64 as follows: pZA64 was cleaved at the RPSA-lacZ junction with BamnHI and then treated with BAL 31 nuclease for various lengths of time to remove 10 to 350 base pairs of the fusion gene. The remaining plasmid ends were made blunt with the large fragment of DNA polymerase, followed by cleavage in the lacZ DNA with SstI. To replace the lacIZ portion of the plasmid removed by BAL 31, the SmaI-SstI fragment of pZl (a plasmid used to construct pZA64; Teem, Ph.D. dissertation) containing this region of pZA64 was ligated to the BAL 31-treated DNA. Yeast cells transformed with these plasmids were identified as blue colonies on plates lacking uracil and containing 5-bromo-4-chloro-3-indolyl-,3-D-galactopyrano-side (7). The endpoints of the deletions were determined by DNA sequencing (26). pHZA7 was constructed by replacing the SalI-SstI fragment of pHZ18 with the SalI-SstI fragment of pZA7. cDNA synthesis. Yeast RNA was prepared as described previously (15). An oligonucleotide (5'TTCACCAGCG AGACGGGC3') complementary to the proximal portion of the 'lacI' fragment of pLG200 (13) was purchased from the Nucleic Acids Laboratory, University of Massachusetts Medical School, Worcester, Mass. cDNA was synthesized from 5 ,ug of total yeast RNA as described previously (30) except that dideoxyTTP was used instead of dTTP to limit the synthesis to a 30-base cDNA product, analogous to the procedure previously described (2).

I-Galactosidase

assay.

Enzyme

assays were

performed

as

described previously (19) except that the yeast cells were lysed by vortexing with glass beads. Immunoblot analysis. Exponentially growing cells were pelleted, suspended in sodium dodecyl sulfate sample buffer (17), lysed by vortexing with glass beads, and boiled for 3

min. An amount of the lysate (10 ,ul), corresponding to approximately 1 ml of the original culture, was subjected to electrophoresis in a 7.5% acrylamide sodium dodecyl sulfate gel (17), and the protein was transferred to nitrocellulose (31). The nitrocellulose filter was treated with rabbit antiserum raised against 3-galactosidase and then with 125[ labeled protein A (113 Ci/g; New England Nuclear Corp.) as described previously (29). Protein-antibody complexes were visualized by autoradiography. Genetic analysis. Yeast crosses and tetrad analysis were accomplished by standard techniques (21). Polysome preparation. Cells were grown in 50-ml cultures to an A6w of 0.6, then treated with 100 ,ug of cycloheximide per ml, quickly chilled, and harvested. The pellets were suspended in diethylpyrocarbonate-treated 50 mM Tris hydrochloride-100 mM NaCI-30 mM MgCl2-0.2% Triton X-100-1 mM 3-mercaptoethanol (pH 7.5) and broken by vortexing with glass beads at 4°C. The lysate, cleared of cell debris by centrifugation at 10,000 rpm for 10 min in an SS34 rotor, was loaded on a 10' to 30% sucrose gradient in the above buffer plus 0.48 M NaCl and centrifuged at 40,000 rpm for 60 min in an SW40 rotor at 0°C. Gradients were fractionated through an ISCO UA-5 absorbance monitor at a wavelength of 260 nm to identify the fractions containing ribosomal subunits, monosomes, and polysomes. Indirect immunofluorescence. Fixation and indirect immunofluorescence of yeast cells were performed as described previously (28) with monoclonal ,-galactosidase antibody 9E3, derived by T. Mason, and fluorescein isothiocyanateconjugated goat anti-mouse immunoglobulin G (Boehringer Mannheim Biochemicals, Indianapolis, Ind.). Cells were treated with 1 jg of 4,6-diamidino-2-phenylindole per ml to stain DNA.

3438

MOL. CELL. BIOL.

GRITZ ET AL. TABLE 1. Effect of RPSJ on ,B-galactosidase activitya

Activity of strain grown in carbon source: Glucose

Plasmid YEp24 pZA64 pPP-Z pZA7 pHZ18 pLGSD5 pHZA7 Integrant ZAint HZint

HZint(pYE51B-1) a

PB12(ARPS5A)

DBY745(RPSIA)

Fusion geneb

Plasmid or integrant

None RPSIA-lacZ (64) tcml-lacZ (372) RPSJA-lacZ (7) GAL-CYCI-RPSIA-IacZ (64) GAL-CYCI-lacZ (0) GAL-CYCI-RPSJA-lacZ (7)

0 30 51 303