the production of decarboxylated AdoMet responsive to the demands of the ..... A clone (pSD-1.35) with an insert size of approximately. 1350 bases was chosen ...
Vol.,261, No. 25, Issue of September 5, pp. 11697-11703,1986 Printed in U.S.A.
THEJOURNAL OP BIOLOGICAL CHEMISTRY 0 1986 hy The American Swiety of Biological Chemists, Inc,
Isolation of a cDNA Clone Encoding S-Adenosylmethionine Decarboxylase EXPRESSION OF THE GENEIN
MITOGEN-ACTIVATED LYMPHOCYTES* (Received for publication, November 19,1985)
Michael Mach, Michael W. White, Michael Neubauer, Jay L. Degen, and David R. Morris From the Department of Biochemistry, SJ-70,University of Washington, Seattle, Washington 98195
S-Adenosylmethionine decarboxylase was purified ine moiety for polyamine biosynthesis (for reviews see Wilfrom bovine liver and digested with endopeptidase liams-Ashman and Pegg, 1981; Tabor and Tabor, 1984; Pegg, Lys-C; the resulting peptides were chromatographi1984). Mammalian AdoMet decarboxylase has a subunit mocally separated. Peptides containing either methionine lecular weight of approximately 32,000 and associates to dior tryptophan weresubjected to sequence analysis. An mers and higher oligomers. The enzymes from both prokaroligonucleotide mixture of 48 sequences, which was 17 yotic and eukaryotic sources contain covalently linked pyrunucleotides in length, was synthesized based on one of voyl residues which are essential for enzymatic activity. these peptide sequences. This syntheticoligonucleotide The metabolic step catalyzed by AdoMet decarboxylase mixture waslabeled and used to screena bovine cDNA joins the pathways of AdoMet metabolism and polyamine library in phage Xgtll. A clone was identified which contained a 1350-nucleotide insert. This insert con- biosynthesis and is regulated. The activity of AdoMet decartained nucleotide sequences coding for amino acid se- boxylase is positively controlled by putrescine, thus making quences of two of the peptidesthat were analyzed, thusthe production of decarboxylated AdoMet responsive to the proving that this cDNA clone codes for S-adenosyl- demands of the polyamine biosynthetic pathway. The cellular level of AdoMet decarboxylase is regulated by a variety of methionine decarboxylase. A subcloned fragment from thecoding region of the stimuli and also by spermidine levels (reviewedin Pegg, 1984). cDNA was used as a probe to analyze the expression of During both liver regeneration and activation of lymphocytes this gene in mitogen-activated lymphocytes. Northern by mitogens, the elevation of enzyme activity has been shown blots revealed two message species of 2.4 and 3.6 kil- to parallel an increase in antigenically reactive protein (Pegg, obases in length. Both mRNAs were coordinately ex- 1979; Seyfried et al., 1982). The short biological half-life of pressed and were present in polysomes. The levels of AdoMet decarboxylase results in rapid changes in the cellular these mRNAs increased -4-fold by 9 h after activation levels of the enzyme and, in several instances, changes in the of the cells. The magnitude of the increase in these half-life generate altered tissue levels of the enzyme (reviewed messages is to be compared with an 8- to 10-fold in- in Pegg, 1984). crease in the rate of synthesis of the protein. The In order to furtherexamine the regulation of this important apparent increase in translational efficiency of this enzyme, we have isolated a cDNA clone coding for bovine message upon lymphocyte activation wasconfirmed by analyzing polysomes from these cells. In resting lym- AdoMet decarboxylase. This cDNA probe has been used to phocytes, the average size of polysomes containing examine the level and efficiency of translation of AdoMet mRNA coding for S-adenosylmethionine decarboxyl- decarboxylase mRNA during activation of bovine lymphoase was 1.4 ribosomes per mRNA, and this value in- cytes by mitogenic lectin. creased to2.7 in stimulated cells. Thus, it appears that EXPERIMENTALPROCEDURES the increase in translational efficiency of this mRNA Isolation of AdoMet Decarboxylase: Purification and Analysis of arises from an elevated rate of translational initiation, leading tomore ribosomes per polysome encoding this Peptide Fragments-AdoMet decarboxylase was purified from calf particular message. This is not a general effect on the liver as described previously (Seyfried et al., 1982). Fifteen nmol of expression of all proteins, since there is no change in the native enzyme were carboxymethylated with iodoacetic acid the translational efficiency of cytoplasmic actin upon (Crestfield et al., 1963). The carboxymethylated protein was digested with 36 pg of endoproteinase Lys-C (Boehringer Mannheim) for 4 h activation of lymphocytes.
AdoMet’ decarboxylase serves an essential role in the biosynthesis of spermidine and spermine by catalyzing the formation of decarboxylated AdoMet, the donor of the propylam-
* This work was partially supported by Research Grant CA39053 from the National Institutes of Health, the National Science Foundation Research Grant PCM8318133,and by Fellowship 22-02592-01 from the American Cancer Society. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solelyto indicate this fact. The abbreviations used are: AdoMet, S-adenosylmethionine; ConA, concanavalin A; SDS, sodium dodecyl sulfate.
a t 37 “C, and the resulting peptides were separated by reverse phase high performance liquid chromatography with a Varian 5000 liquid chromatograph ona pBondapak Cls column (Waters Associates). Chromatography was done at pH 2.0 using 0.1% trifluoroacetic acid as theaqueous phase and eluting with an increasing concentration of acetonitrile (Mahoney and Hermodson, 1980). An empirically determined amount of trifluoroacetic acid, usually 0.07-0.08%, was added to the acetonitrile in order to obtain a flat absorbance base line throughout the gradient. A second buffer system for further analysis and purification of peptides consisted of an aqueous phase of 5 mM NH4HCO8, adjusted to pH 6.5 with formic acid, and a gradient of acetonitrile was used for elution (Karplus, 1984). The flow rate was 2.0 ml in both systems. The absorbance was monitored a t 200 nm, and fractions corresponding to absorbance peaks were collected manually. Amino acid compositions were determined on a Dionex amino acid analyzer (Model D500) after hydrolysis in 5.7 N HCl at 100 “C for
11697
11698
Isolation of a cDNA Clone Encoding AdoMet Decarboxylase
16-24 h. Automated sequence analysis was performed with a Beckman sequencer (Model 890C) in the presence of Polybrene (Pierce), using the method of Edman and Berg (1967) and the quadrol program of Brauer et al. (1975). Manual sequencing was performed according to Tarr (1982) as modified by Karplus (1984). Methionine-containing peptides were detected by amino acid analysis and tryptophan-containing peptides by measuring fluorescence at 348 nm (excitation at 287 nm). Construction and Screening of Bovine cDNA Library-Poly(A)+ RNA was isolated from 10-h activated bovine lymhocyte cultures as described (Degen et al., 1983). Double-stranded cDNA was synthesized, and molecules larger than 600 base pairs were isolated (Degen et al., 1983, Stein et al., 1978). The Xgtll vector (Young and Davis, 1983)was cleavedwith EcoRI and treatedwith calf alkaline phosphatase to reduce intramolecular religation. The size-fractionated cDNA was then ligated into the phage arms using EcoRI linkers and packaged in vitro (Enquist and Sternberg, 1979). A library was obtained from 150 ng of double-stranded cDNA, which contained 4.7 X 10' independent recombinants and 20% wild type phage. The library was screened by the insitu plaque hybridization technique of Benton and Davis (1977) as modified by Woo (1979). All hybridizations with the oligonucleotide mixture (obtained from Creative Biomolecules, San Francisco, CA) were carried out in 6 X ssc (1 X s s c , 15 mM sodium citrate, pH 7.0, containing 0.15 M NaCl) at 37 "C. The filters were then washed three times at 37 "C in a solution of 2 X SSC and 0.1% SDS for 2 h each and at42 "C in 2 X SSC, 0.1% SDS for 15 min. For labeling the oligonucleotide mixture, 150 ng was incubated for 45 min at 37 "C with 200 pCi [m3'P]ATP (specificactivity 3000 Ci/mmol), 20 units of T4 polynucleotide kinase (Bethesda Research Laboratories) in50 mM Tris-HC1, pH 7.4,lO mM MgC12,5 mM dithiothreitol, 0.1 mM spermidine trihydrochloride, and 0.1 mM EDTA. The labeled oligonucleotide mixture was separated from unincorporated ATP on a 0.8 X 25-cm column of Sephadex G50 superfine in 10 mM Tris-HC1, pH 8.0, 0.1 M NaCl, and 1 mM EDTA. Typically, 1.0-1.5 X lo9 cpm was incorporated per pgof oligonucleotide mixture using this protocol. Initial Sequence Characterization of cDNA Inserts-Bacteriophage DNA was isolated by a rapid small scale isolation procedure (Maniatis et al., 1982), starting with 25 Petri plates. Inserts were cut out with EcoRI, isolated by electrophoresis in gels of low melting agarose, and used as template for the sequencing method of Sanger et al. (1977). Insert DNA (0.5 pg) and 0.1 pg of oligonucleotide, in 25 pl of 10 mM Tris-HC1, pH 8.0, and 5 mM MgClz, weresealed into a glass capillary tube and incubated at 100 "C for 3 min. The tube was plunged into ice water and 1 unit of Escherichia coli DNA polymerase I (Klenow fragment) and 20pCiof [cY-~'P]~CTP were added. From this point on, the Amersham sequencing protocol for the dideoxy method was followed. Subcloning of Fragments of the 1350-Base Insert-Cutting the 1350-base cDNA with restriction endonuclease Sau3A generated three fragments. These fragments were subcloned into the EcoRI/ BamHI site of M13mplO and mpll. Thesubclone that hybridized to 32P-labeled oligonucleotide mixture was sequenced by the dideoxy method as described in the Amersham cloning and sequencing manual. Quantitative Assay of AdoMet DecarboxylasemRNA Sequences by Solution Hybridization-AdoMet decarboxylase mRNA sequences in preparations of total RNA were measured by hybridization to 35Slabeled single-stranded cDNA. Preparation of single-stranded hybridization probes specific for AdoMet decarboxylase mRNA was exactly as described by Durnam and Palmiter (1983). Hybridization mixtures contained 0.6 M NaC1, 4 mM EDTA, 10 mM Tris-HC1, pH 7.5, 7-20 pg of total RNA, and about 1000 cpm of =S-labeled cDNA in a final volume of 30 p1. The mixtures were covered with paraffin oil, centrifuged briefly, and incubated at 68 "C for approximately 16 h. In each experimentastandard curve was included, consisting of known amounts of single-stranded phage DNA. Followinghybridization, the samples were treated with nuclease S1 toremove unhybridized cDNA (Durnam and Palmiter, 1983). Northern Blot Analysis-Total RNA was purified from bovine lymphocytes as previously described (Chirgwin et al., 1979).The RNA was electrophoresed on 1.2% agarose gels containing 6.5% formaldehyde (v/v), 20 mM sodium phosphate, pH 7.2, and 50 mM sodium citrate and blotted onto nitrocellulose using 20 X SSC. The nitrocellulose blot was baked for 2 h a t 80 "C and thenprehyhridized at 42 "C in 6 x SSC containing 2 X Denhardt's solution, 50% formamide, 20 mM sodium phosphate, pH 7.2, and yeast tRNA (0.2 mg/ml). Denatured AdoMet decarboxylase cDNA insert, which had previously
been labeled with [w3'P]dATP by nick translation (Durnam and Palmiter, 19831, was injected into the prehybridization solution and the blot allowed to hybridize 36 h at 44 "C. The blot was then washed three times in 2 X SSC containing 0.1% SDS at room temperature for 20 min per wash, followed by twowashes in 0.1 X SSC containing 0.1% SDS a t 50°C for 30 min per wash. After washing, the blot was air-dried and exposed to x-ray film at -80 "C using a DuPont Cronex Lightning Plus intensifying screen. Preparation of Polysomal RNA-All steps were done on ice or at 4 "C using autoclaved glassware and buffers. Cells (1 x 10') were collected by centrifugation a t 300 X g for 15 min, washed in 20 ml of phosphate-buffered saline (0.14 M NaCl, 5 mM KC1, 8 mM Na2P0,, and 1.5 mM KHzP04,pH 7.31, and collected again by centrifugation a t 300 X g for 15 min. The cells were then resuspended in 0.75 ml of low salt buffer (LSB, 20 mM Tris-HC1, pH 7.4, containing 10 mM NaCl and 3 mM MgC1,) and allowed to swell for 3 min. As rapidly as possible, 0.25 ml of a detergent-containing buffer (LSB containing 1.2% (v/v) Triton N-101 and 0.2 M sucrose) was added tothe suspension, and the mixture was homogenized with eight strokes in a Dounce homogenizer with a tight-fitting pestle. The lysate was centrifuged at 27,000 X g for 30 s, and the supernatant was poured into a tube with 0.1 ml of LSB containing heparin (10 mg/ml); the contents were mixed by inversion. A solution of 4 M NaCl was then added to give a final concentration of0.15 M, and the lysate (-1.1 ml) was applied to a 0.5-1.5 M linear sucrose gradient (11ml, in LSB) prepared in an autoclaved SW 40.1 polyallomer tube (Beckman) and centrifuged at 36,000 rpm for 110 min at 4 "C in this rotor. After centrifugation, the gradients were fractionated into 12 1-ml fractions using an ISCO Density Gradient Fractionator equipped with an absorbance monitor (see absorbance profiles of unstimulated and stimulated cells, Fig. 8). To each fraction, SDS and proteinase K (0.5%and 0.1 mg/ml, final concentrations, respectively) were added, and thefractions were incubated for 30 min a t 37 "C. Fractions were then combined into six pools as indicated and extracted with an equal volume of phenol/chloroform (1:l).This was followed by twoextractions with equal volumes of chloroform, and theRNA wasprecipitated with 2.5 volumes of ethanol at -20 "C overnight. After collection by centrifugation, the RNA was resuspended in water and analysis was performed by Northern blots as described above. RESULTS
Partial Amino Acid Sequence of AdoMet Decarboxylaseand Selection of an Oligonucleotide Probe-AdoMet decarboxylase was purified from the livers of calves treated with methylglyoxal bis(guany1hydrazone)in order to elevate the levels of the enzyme (Seyfried et al., 1982). On SDS-polyacrylamide gel electrophoresis, the purified enzyme showeda single staining band corresponding to a molecule weight of 32,000, the reported subunit molecular weight of the bovine, mouse, and rat enzymes (Seyfried et al., 1982; Pegg, 1977; Sakai et al., 1979). No sequence data could be obtained from the amino terminus of the intact protein, suggesting the presence of a blocked amino terminus. The position of the pyruvoyl residue, which is present in AdoMet decarboxylase from mammalian cells (Pegg, 1977), has not been identified in the polypeptide chain, but in a bacterial pyruvoyl enzyme, Lactobacillus histidine decarboxylase, the prosthetic group was localized to the amino terminus (Riley and Snell, 1970). Thus, our data are consistent with a similar location of the pyruvoyl residue in mammalian AdoMet decarboxylase. Approximately 15 nmol of purified AdoMet decarboxylase was carboxymethylated, digested with endopeptidase Lys-C, and subjected to high performance liquid chromatography. Eighteen major peaks were observed, which is close to the number expected based on 19 lysine residues per molecule (determined by amino acid analysis, data not shown). Fractions corresponding to absorbance peaks in the first chromatography step were subjected to analytical chromatography in a second solvent system (see "Experimental Procedures"), and those peptide mixtures which resolved into two or more peaks were further purified in this second system. Peptides givingonly one peak in the second solvent system were
Isolation of a cDNA Clone Encoding AdoMet Decarboxylase subjected directly to Edman degradation. The purity of all peptides sequenced was confirmed by unambiguous chromatograms of the phenylthiohydantoin derivatives after Edman degradation. Due to theirunique codon assignments, peptides containing the amino acids methionine and tryptophanwere of particular importance for designing oligonucleotide probes. After purification, all peptides were analyzed for methionine by amino acid analysis andtryptophan by fluorescence at 348 nm. Chromatograms of five peptides, together with partial amino acid sequence analysis, are shown in Fig. 1. Due to limiting amounts of peptide material available after purification, only peptide number 116 could be sequenced entirely to the carboxyl terminus. In peptide number 118,the phenylthiohydantoin derivative of the second amino acid could not be identified, suggesting a modified amino acid in the position. A portion of peptide 31 was identified for which there were relatively few potential coding sequences. The possible codon assignments for positions 5 to 10 are given in Fig. 2. In constructing a synthetic oligonucleotide mixture, the third nucleotide of the last isoleucine codon was excluded. This gave a 17-nucleotide long sequence with 48 coding possibilities. This oligonucleotide mixture was purchased commercially and used to screen a cDNA library. Identification of a cDNA Clone Coding for AdoMet Decarboxylase-Double-stranded cDNA fragments of greater than 600 base pairs in length were prepared from bovine lymphocyte poly(A)+ RNA and cloned into the EcoRI site of bacteriophage Xgtll using EcoRI linkers (Young and Davis, 1983). The cDNA library was screened by filter hybridization using the synthetic oligonucleotide mixture as probe after labeling with 32Pusing polynucleotide kinase. Of the 4 x lo5 recombinants screened, 40 hybridized to the oligonucleotide mixture. Thehybrids were stable for 15 min at 45 “C in2 X SSC, conditions considered to be stringent enough to require a perfect match for formation of a stable hybrid (Wallace et al., 1979). A clone (pSD-1.35) with an insertsize of approximately 1350 bases was chosen for further characterization. For preliminary confirmation of the identity of clone pSD-1.35, partial sequence was determined from the site of binding of the synthetic oligonucleotide probe. The inserted DNA was cut out with EcoRI and purified by agarose gel electrophoresis. The insert was heat-denatured and sequence analysis was performed by the dideoxy method (Sanger et al., 1977), using the syntheticoligonucleotide mixture as primer and dCTP as
11699
the labeled nucleotide (see “Experimental Procedures”). The preliminary sequence of this clone was found to be consistent with amino acids 1 to 3 in peptide 31 (Ser-Asp-Gly). Restriction endonuclease analysis of clone pSD-1.35 (see Fig. 3) showed the presence of several endonuclease sites. The restriction endonuclease Sau3A cut theinsert into fragments of 550, 600, and 180 base pairs. The 180-base pair fragment was mapped to the 5‘-end of the cDNA clone and was found t o hybridize to the synthetic oligonucleotide mixture. This fragment was subcloned into M13mplO which was cleaved with EcoRI and BamHI and then sequenced entirely by the dideoxy method (Sanger et al., 1977). The results are presented in Fig. 4. The site complementary to the synthetic oligonucleotide probe is located from base 93 through 109. The base composition of the 17-nucleotide sequence is 29% G + C , indicating that the hybridization conditions of the initial screening were quite stringent indeed. The nucleotide sequence from base 81 through 170 codes precisely for the amino acid sequence of peptide 31. In addition, nucleotides 1 through 32 code for the sequence of amino acids 7 through 16, and part of amino acid 6, of peptide 297. These identities between amino acid sequences in the protein and sequences translated from the nucleotide sequence of the cDNA clearly establish the clone as coding for bovine AdoMet decarboxylase. The nucleotide sequence continues into the vector to the 5’-side of the sequence given in Fig. 4. This, togetherwith the fact that the sequence of peptide 297 was truncated in the clone, suggested that the original 1350-base clone lacked a portion of the 5’-end of the protein sequence of the messenger RNA and is therefore an incomplete cDNA copy. The Level of AdoMet Decarboxylase mRNA after Mitogenic Activation of Lymphocytes-Total RNA was extracted from bovine lymphocytes at various times after mitogenic activation by ConA. Northern blots were prepared after agarose gel electrophoresis in the presence of formaldehyde, and hybridization was performed using AdoMet decarboxylase cDNA probe (Fig. 5). Two hybridizing bands were observed with sizes relative to prokaryotic and eukaryotic rRNA standards, of 2.4 and 3.6 kilobase pairs. These mRNA species are found also on Northern blotsprepared from cytoplasmic, rather than total, RNA (data notshown), and therelative proportion of these two hybridizing species did not change during the course of mitogenesis (Fig. 5 ) or with varying stringency of hybridization (data not shown). Densitometric scanning of these Northern blots showed that during the first 24 h after T
70.06
0.8 97:DVTESG IRDLIP GSVI
FIG. 1. Elution profilesof peptides in high performance liquid chromatography system 11. Following initial separation of the peptides in system I, the major peaks were rechromatographed at neutral pH in system I1 (sse “Experimental Procedures”).Left panel, preparative chromatography for purification of peptides 116, 118, and 297, respectively. Right panel, analytical chromatography of peptides 22 and 31.
22:LLEVW FSRQ
31:SDGMIKTIHI TPEPEFSWS FETNLSQTSY
-ip \
0.4
0.02
0.2
\ OJ 0
20
0
2040
J
C
I
20
b
r
20
4
11700
Isolation of a cDNA Clone Encoding AdoMet Decarboxylase
recting for these increases in total cellular RNA, the cellular levels of AdoMet decarboxylase mRNA increased 3.3- and 5.5-fold at 12 and 24 h in the experiment of Fig. 6. Table I T h rGTlryp T y r T h r Ile His lie Thr summarizes the levels of AdoMet decarboxylase mRNA per cell derived from four independent RNA preparations. From these data, the average increases are 3.9-fold at 12 h and 5.4A A GGN ACN UUAG AC $G UN ACE AUC ACN fold at 24 h. Fig. 7 presents amore detailed time course of the u u relative cellular level of AdoMet decarboxylase mRNA and FIG. 2. Partial amino acid sequence of peptide 31 from bovine AdoMet decarboxylase and possible nucleotide coding compares it with the rate of synthesis of the enzyme protein sequences. Bold letters highlight the RNA sequence used for synthe- as determined previously (Seyfried et al., 1982). As can be sis of the deoxyoligonucleotidemixture. N indicates that any one of seen, there is approximately a 2-fold discrepancy between the the four bases could be used. relative change in therate of enzyme synthesisand the relative level of its mRNA, this suggests an element of translational control which is manifested as theobserved increase Ec So Hd Nc Hd Er Ec in efficiency of translation of AdoMet decarboxylase mRNA. Distribution of AdoMet Decarboxylase mRNAinPolysomes-Polysomes from lysates containing identical numbers of unstimulated and 8-h ConA-activated cells were fraction0 BO 400 550 650 1350 ated on sucrose gradients (see Fig. 8). RNA was purified from FIG. 3. Restriction endonuclease map for the insert region the gradient fractions, separated on formaldehyde/agarose of the recombinant DNA clone pSD-1.35. Hatched area of the insert indicates the region of this clone which was sequenced by the gels, and transferred to nitrocellulose. The polysome Northdideoxynucleotide methods. Restriction endonuclease abbreviations: ern blots were then probed with nick-translated AdoMet Ec, EcoRI; Sa, Sau3A; Hd, HindIII;Nc, NcoI; Er, EcoRV. decarboxylase cDNA insert. Because the level of AdoMet decarboxylase mRNA increased about 3-fold by 8 hafter ConA addition to thecells, the level of AdoMet decarboxylase activation, the autoradiograph of the 8-hpolysome Northern mRNA relative to total RNA increased approximately 3- to blot was underexposed -%fold to closely match the hybridi4-fold. A more quantitative estimateof AdoMet decarboxylase zation signal on the blot from unstimulated cells (see Fig. 9). mRNAlevelwas obtained by solution hybridization to a This was done to facilitate comparison of mRNA distributions single-stranded DNA probe complementary to the coding across the polysome gradients. Autoradiographs, matched in region of mRNA. A typical experiment is given in Fig. 6. this way, were quantitated by densitometric scanning, and the Varying amounts of total cellular RNA were added to a data are expressed as the percentage of total hybridization constant amount of probe, under conditions where hybridi- (Fig. 9). AdoMet decarboxylase mRNA from resting lymphozation to a single-stranded DNA standard was linear with cytes was translated on monosomes (75%, gradient pool 2) input (see “Experimental Procedures”). The amount of S1 and polysomes containing 2-5 ribosomes (25%, gradient pool nuclease-resistant radioactivity was determined, and thelevel 3). In activated lymphocytes, the amount of AdoMet decarof AdoMet decarboxylase mRNA relative to total cellular boxylase mRNA on monosomes (58%) was decreased, with a RNA was calculated from the slopes of plots such as those small increase in the proportion of mRNA on polysomes given in Fig. 6. Inthis particular experiment, a 2.4-fold containing 2-5 ribosomes (30%) and greater than a 10-fold increase in the level of the specific mRNA relative to total increase on polysomes containing more than 6 ribosomes RNAwas observed both at 12 and 24 h after mitogenic (12%, gradient pool 4). This apparent shift inAdoMet decaractivation. In mitogen-activated bovine lymphocytes, the total boxylase mRNA onto larger polysomes does not seem to be a RNA per cell was shown to increase 1.3- and 2.2-fold at 12 general effect on ConA stimulation on all message species, and 24 h, respectively (Fillingame and Morris, 1973). Cor- since the size of polysomes containing actin mRNA has been G
T
Y
W
T
I
Ile
Arq
Asp
Leu
H
Ile
T
I
G Pl yr o
Ser
Val
Ile
Asp
Ala
Tht
:,let
G A A T T C G T G A C C T G A T A C C A G G T T C T G T C A T T G A T G C C A C A A T G 20
1
Ser
T yGrC l yyPsrAoP s nh e
40 Met
Asn
Gly
Net
Lys
Ser
Asp
Gly
T T C A A T C C T T G T G G G T A T T C A A T G A A T G G G A T G A A A T C G G A T G G A 60
80
l’hr T hT rrTpy r
Ile
His
I l eP G r oPl urToh r
Glu
S ePrh e
Tyr
Asp
‘T}t
120 Val
S e t T hGrP l uh e
Asn T hGrlSneLre u
Set
Asr;
G T T A G C T T T G A A A C A A A C T T A A G T C A G A C C T C C T A T G A T G A C C T G A T 140
160
FIG. 4. Nucleotide sequence of pSD-0.18. The nucleotide sequences corresponding to amino acid sequences of peptides 31 and 297 from AdoMet decarboxylase are indicated by an ouerbar.
1bo
Isolation of a cDNA Clone Encoding AdoMet Decarboxylase TABLE I
Time after Con A
0 0.5 I
11701
2 4
- 3.6 Kb - 2.3 Kb 3.4 3.7 3.9 5.7 FIG. 5. Northern gel analysis ofAdoMet 5.4 decarboxylasc mRNA species. Equal amounts of total RNA isolated from unstimulated cells and cells activated for 0.5, 1, 2, and 4 h were electrophoresed in a 1.2% agarose/formaldehyde gel and then blotted onto nitrocellulose. The 180-base pair insert from clone pSD-0.18 (see Fig. 3) was labeled with [cY-~'P]~ATP by nick translation, denatured, and used as a hybridization probe. The mRNA nucleotide length was estimated from rRNA standards which were co-electrophoresed. Kb, kilobase.
Influence of lymphocyte activation on the level of AdoMet decarboxylase mRNA Total RNA was isolated from lymphocytes at the indicated times after activation with ConA. 35S-Labeledsingle-stranded cDNA probe was prepared from the 180-base insert of clone pSD-0.18. This probe (about 1000 cpm input) was hybridized to various amounts of total RNA (between 7.5 and 20 pg) and hybridization was detected by resistance to nuclease SI as described in Fig. 6. Four independent RNA DreDarations wereanalvzed as indicated. Time after ConA
I"
I1
Relative RNA level 111 IV IVb
Average
1.0 1.0 0 1.0 1.0 1.0 1.0 4.4 4.2 10-12 3.9 6.3 24 4.3 5.3 'Independently prepared RNA samples are designated by roman numerals I-IV. * Data inthis column were obtained by hybridization of sample IV in buffer containing 20% formamide.
2501
8 2001
+422iLI
6 501
00
2
4
6
8
101214
FIG. 6. Quantitative analysis of the content of AdoMet decarboxylase mRNA sequences in total RNA at various times after cell activation. 35S-Labeledsingle-stranded DNA was prepared from the insert of clone pSD-0.18. The probe was hybridized to the indicated amounts of total RNA isolated from unstimulated cells (A) and cells activated for 12 (0)and 24 (0)h. Thedata represent counts above a background with no RNA added to the hybridization. The inset shows results using complementary singlestranded (ss) DNA of pSD-0.18 as a standard.
shown to be identical in resting and activated lymphocytes (Degen et al., 1983). DISCUSSION
A cDNA clone was isolated which hybridized efficiently to an oligonucleotide mixture that was synthesized based on partial amino acid sequence data for AdoMet decarboxylase. Preliminary confirmation of the identity of the clone was complicated by the fact that AdoMet decarboxylase mRNA was very inefficiently translated, if a t all, in reticulocyte lysates using conditions previously described for translation of actin mRNA (Degen et al., 1983). Attempts to identify the clone by hybridization of lymphocyte poly(A)+RNA, followed by elution, translation, and immunoprecipitation, were unsuccessful. It was therefore necessary to devise the procedure described for preliminary identification, which used the oligonucleotide mixture as primer for the dideoxy sequencing
I
6
12 18 Time ofter ConA- hours
24
FIG. 7. Comparison of the rate of AdoMet decarboxylase synthesis and the level of its mRNA. The relative level of AdoMet decarboxylase mRNA (0)was determined as described in the legends to Fig. 6 and Table I. The rate of synthesis of the protein (0)was taken from Seyfried et al. (1982). Both sets of data were normalized to thevalues in resting cells.
method. Preliminary analysis by this procedure was consistent with the amino acid sequence, and thus more extensive characterization of the clone was justified. This approach shouldbe generally applicable in cases where the known amino acid sequence extends over the 3'-end of the site of oligonucleotide hybridization. Rigorous identification of the cDNA clone as one coding for AdoMet decarboxylase was provided by DNA sequence analysis of the 180-base pair EcoRIISau3A fragment which contained the sequence of the synthetic oligonucleotide probe. The fragment encompassed a complete open reading frame and contained a nucleotide sequence corresponding to all 30 amino acids of peptide 31. In addition, this cDNA fragment also contained a nucleotide sequence coding for the 10 carboxyl-terminal amino acids of peptide 297, which was interrupted by an EcoRI site at the 5'-end of the coding sequence. Thus, although we have not been able to obtain hybrid elution-translation data for this cDNA clone, it seems quite clear that it indeed codes for AdoMet decarboxylase. This clone was used to monitor the level of AdoMet decarboxylase mRNA during lymphocyte mitogenesis. The rate of synthesis of AdoMet decarboxylase increased disproportionately to the general increase in translational
Isolation of a cDNA Clone Encoding AdoMet Decarboxylase A. Unstimulated cells
I
B. 8 hr. activated cells
Unstimulatedcells
\ 8hr. activated
cells O
l
2
3
4
5
6
O
i
2
3
4
5
6
Pooled fractions
J
I
2
I
I
4
I
I
6
I
I
8
1
I
L
IO
Fraction number FIG. 8. Absorbance profiles at 254 nm of polysomes from unstimulated and 8-h ConA-activated cells. Lysates corresponding to equal numbers of cells were loaded onto sucrose gradients and, after centrifugation, the gradients were fractionated into 12 1-ml fractions (see “Experimental Procedures”).
activity in mitogen-activated bovine lymphocytes (Degen et al., 1983); AdoMet decarboxylase synthesis increased 10-fold by 9 h after activation (Seyfried et al., 1982), as compared with a %fold increase in the overall rate of protein synthesis at the same time (Degen et al., 1983). The level of AdoMet decarboxylase mRNA was determined by solution hybridization and confirmed by quantitation of Northern blots. As early as 3 h after activation, we observed a doubling in the cellular level of sequences coding for AdoMet decarboxylase and by 10 h the increase was 4-fold. These results reveal a highly significant 2.0- to 2.5-fold discrepancy in the rate of synthesis of AdoMet decarboxylase compared with the level of its mRNA. This is in clear distinctionto the situation with actin inthese cells, where the rateof synthesis of the protein and the cellular level of the mRNA increased inparallel (Degen et al., 1983). Thus, thereappears to be an increase in the efficiency of translation of AdoMet decarboxylase mRNA during the course of mitogenic activation, although that of actin mRNA remains constant. The average size of polysomes containing AdoMet decarboxylase mRNA was calculated from the data of Fig. 9 to be 1.4 ribosomes permRNA in resting lymphocytes. This is considerably smaller than the predicted polysome size of 1011 ribosomes for a protein of M , 32,000; this prediction is based on actin which is 1.5 times larger and is foundon polysomes of 15 ribosomes in these gradients, consistent with
FIG. 9. Distribution of AdoMet decarboxylase mRNA in polysomes isolated from unstimulated ( A )and 8-h activated cells ( B ) .Lysates from 1x IO9cells were fractionated on sucrose gradients (12 1-ml fractions, see Fig. 8 for absorbance profiles). Fractions were pooled as described below, and RNA was prepared from aliquots of each gradient pool. The RNA was electrophoresed on 1.2% agarose/ formaldehyde gels, blotted, and hybridized as described in Fig. 5. The amount of hybridization was quantitated from autoradiographs by densitometric scanning and expressed as the percent of the total hybridizable material across the gradient. Polysome size was estimated from Fig. 8. A , unstimulated cells: ribosomal subunits, pool 1 (fractions 1and 2); monosomes, pool 2 (fractions 3 and 4); polysomes containing 2-4 ribosomes, pool 3 (fractions 5 and 6); polysomes containing greater than 4 ribosomes, pool 4 (fractions 7 and 8); large polysomes, pools 5 and 6 (fractions 9 and 1 0 , l l and 12). B, fractions from activated cells were pooled identically: subunits, pool 1; monosomes, pool 2; polysomes containing 2-5 ribosomes, pool 3; polysomes containing greater than 5 ribosomes, pool 4 (an average polysome size of 8 ribosomes was used for this fraction in calculating the overall ribosomes per mRNA, see“Discussion”);large polysomes, pools5 and 6.
our previous report (Degen et al., 1983). Thus, inresting lymphocytes, AdoMet decarboxylase mRNA is quite inefficiently translated compared to actin. In 8-h ConA-activated lymphocytes, the average size of polysomes containing AdoMet decarboxylase mRNA increased to 2.7 ribosomes per mRNA. This %fold increase in average polysome size easily accounts for the discrepancy between the elevation in rate of AdoMet decarboxylase synthesis and theincrease in the level of its mRNA after stimulation with ConA. These datastrongly suggest that theincreased translational efficiency of AdoMet decarboxylase mRNA is due to increased rate of translational initiation, leading to more ribosomes per polysome encoding this message. Recently, it has been reported by Kahana and Nathans (1985) that the cDNA nucleotide sequence of another polyamine biosynthetic enzyme, ornithine decarboxylase from mouse, contains an unusual 5’untranslated region similar to theyeast protein GCN-4 (Thireos et al., 1984; Hinnebusch, 1984). This yeast message has been demonstrated to be under translational control in uiuo, and hence it was suggested that ornithinedecarboxylase may also be translationally controlled in uiuo. The present demonstration thatAdoMet decarboxylase is translationally reg-
Isolation of a cDNA Clone Encoding AdoMet Decarboxylase ulated in vivo suggests the existence of a common element of translational regulation for these two important enzymes of polyamine biosynthesis. Acknowledgments-We would like to express our gratitude to C. Seyfried for help in purification of AdoMet decarboxylase, to Drs. A. Karplus and K. A. Walsh for invaluable aid in amino acid sequence analysis, and to C. Degnin for help in subcloning AdoMet decarboxylase cDNA and in culturing lymphocytes.
REFERENCES Benton, W. D., and Davis, R. W. (1977) Science 196,180-182 Brauer, A.W., Margolis, M.N., and Haber, E. (1975) Biochemistry 14,3029-3035 Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J., and Rutter, W. J. (1979) Biochemistly 1 8 , 5294-5299 Crestfield, A. M., Moore, S., and Stein, W. H. (1963) J . Biol. Chem. 238.622-627 Degen, J. L., Neubauer, M. G., Degen, S. J. F., Seyfreid, C. E., and Morris, D. R. (1963) J. Biol. Chem. 258, 12153-12162 Durnarn, D., and Palrniter, R.D. (1983) Anal. Biochem. 1 3 1 , 385393 Edrnan, P., and Berg, G. (1967) Eur. J. Biochem. 1,80-91 Enquist, L., and Sternberg, N. (1979) Methods Enzymoi.68,281-298 Fillingame, R. H., and Morris, D. R. (1973) Biochemistry 12, 44794487 Hinnebusch, A. G . (1984) Proc. Natl. Acad. Sci. U. S. A . 8 1 , 64426446 Kahana, C., and Nathans, D. (1985) Proc. Natl. Acad. Sci. U. S. A . 8 2 , 1673-1677
11703
Karplus, A. (1974) Ph.D. Thesis, University of Washington, Seattle Mahoney, W. C., and Hermodson, M. A. (1980) J. Biol. Chem. 2 5 5 , 11199-11203 Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) Molecular Cloning, A Laboratory Manuul, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY Pegg, A. E. (1977) FEBS Lett. 8 4 , 33-36 Pegg, A. E. (1979) J. B i d . Chem. 254,3249-3253 Pegg, A. E. (1984) Cell Biochem. Funct. 2, 11-15 Riley, W. D., and Snell, E. E. (1970) Biochemistry 9 , 1485-1491 Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl. Acad. Sci. U. S. A . 74,5463-5467 Sakai, T., Hori, C., Kano, K., and Oka, T. (1979) Biochemistry 1 8 , 5541-5548 Seyfried, C . E., Oleinick, 0. E., Degen, J. L., Resing, K., and Morris, D. R. (1982) Biochim. Biophys. Acta 7 1 6 , 169-177 Stein, J. P., Catterall, J. F., Woo,S. L. C., Means, A. R., and O’Malley, B. W. (1978) Biochemistry 17,5763-5772 Tabor, C. W., and Tabor, H. (1984) Adu. Enzymol. 66,251-282 Tarr, G . (1982) in Methods in Protein Sequence Analysis (Elzinga, M.,ed) pp. 223-232, Humana Press, Clifton, NJ Thireos, G., Driscoll-Penn, M., and Greer, H. (1984) Proc. Natl. Acad. Sci. U. S. A . 81,5096-5100 Wallace, R. B., Shaffer, J., Murphy, R. F., Bonner, J., Hirose, T., and Itakura, K. (1979) Nucleic Acid Res. 6 , 3543-3557 Williams-Ashman, H. G., and Pegg, A. E. (1981) in Polyamines in Biology and Medicine (Morris, D. R., and Marton, L. J., eds) pp. 343, Marcel Dekker, New York Woo, S. L. C. (1979) Methods Enzymol. 6 8 , 389-395 Young, R A., and Davis, R. W. (1983) Proc. Natl. Acad. Sci. U. S. A. 80,1194-1198
