Identification of a DNA clone to phosphoenolpyruvate carboxykinase (GTP) from rat cytosol. Alterations in phosphoenolpyruvate carboxykinase RNA levels detectable by hybridization.

A recombinant plasmid containing a DNA segment complementary to phosphoenolpyruvate  carboxykinase (GTP) (EC 4.1.1.32) mRNA from rat kidney has been constructed and cloned. Using this probe, studies have been initiated to determine those factors which regulate phosphoenolpyruvate carboxykinase mRNA levels in rat liver and kidney. Enriched mRNA for phosphoenolpyruvate  carboxykinase was  prepared from poly(A)+ RNA isolated from the kidneys of acidotic rats injected with triamcinolone, by sedimentation through successive neutral and denaturing sucrose density gradients. Those RNA fractions in which phosphoenolpyruvate carboxykinase protein accounted for at least 15% of the total proteins ynthesized in a reticulocyte cell-free protein synthesizing system were combined and used to prepare double-stranded DNA with avian myeloblastosis virus reverse transcriptase. A poly(dC) tail was added to the double-stranded DNA with terminal transferase and the DNA was inserted into the Pst I site of a pBR322 tailed with poly(dG) and used to transform Escherichia coli HB101. A plasmid DNA containing a 250-base pair insert (pPCK1) was identified as a clone to phosphoenolpyruvate carboxykinase mRNA by hybridization-selected translation of rat kidney mRNA and immunoprecipitation of the translated products with antibody specific to the enzyme. Using pPCK1, the relative sequence abundance of phosphoenolpyruvate  carboxykinase mRNA from rat liver and kidney after various hormonal treatments were examined. The relative amount of hepatic phosphoenolpyruvate carboxykinase mRNA increases after starvation for 24 h or CAMP administration to normal fed rats and is also extremely high in the livers of diabetic rats. Conversely, the amount of phosphoenolpyruvate carboxykinase mRNA in the liv-

A recombinant plasmid containing a DNA segment complementary to phosphoenolpyruvate carboxykinase (GTP) (EC 4.1.1.32) mRNA from rat kidney has been constructed and cloned. Using this probe, studies have been initiated to determine those factors which regulate phosphoenolpyruvate carboxykinase mRNA levels in rat liver and kidney. Enriched mRNA for phosphoenolpyruvate carboxykinase was prepared from poly(A)+ RNA isolated from the kidneys of acidotic rats injected with triamcinolone, by sedimentation through successive neutral and denaturing sucrose density gradients. Those RNA fractions in which phosphoenolpyruvate carboxykinase protein accounted for at least 15% of the total proteins synthesized in a reticulocyte cell-free protein synthesizing system were combined and used to prepare double-stranded DNA with avian myeloblastosis virus reverse transcriptase. A poly(dC) tail was added to the double-stranded DNA with terminal transferase and the DNA was inserted into the Pst I site of a pBR322 tailed with poly(dG) and used to transform Escherichia coli HB101.
A plasmid DNA containing a 250-base pair insert (pPCK1) was identified as a clone to phosphoenolpyruvate carboxykinase mRNA by hybridization-selected translation of rat kidney mRNA and immunoprecipitation of the translated products with antibody specific to the enzyme. Using pPCK1, the relative sequence abundance of phosphoenolpyruvate carboxykinase mRNA from rat liver and kidney after various hormonal treatments were examined. The relative amount of hepatic phosphoenolpyruvate carboxykinase mRNA increases after starvation for 24 h or CAMP administration to normal fed rats and is also extremely high in the livers of diabetic rats. Conversely, the amount of phosphoenolpyruvate carboxykinase mRNA in the liv-*This work was supported in part by a grant from the Kroc Foundation and by Grants AM 25541 and CA 27414 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. This paper is dedicated to the memory of Dr. Merton F. Utter who discovered P-enolpyruvate carboxykinase and shared with us a lifelong interest in its regulation. ers of starved animals rapidly decreases after carbohydrate feeding or when insulin is injected into diabetic rats. The abundance of renal phosphoenolpyruvate carboxykinase mRNA is not altered in the diabetic animal but is increased by the combination of acidosis and triamcinolone injection. This pattern of hormonally induced alterations in the levels of phosphoenolpyruvate carboxykinase mRNA agrees well with previously reported changes in translatable enzyme mRNA.
The levels of translatable mRNA for cytosolic P-enolpyruvate carboxykinase' in rat liver and kidney are markedly altered by hormonal and dietary stimuli (1)(2)(3)(4)(5)(6). Insulin injection to a diabetic animal ( 5 ) or glucose administration to a starved rat (2) causes a rapid de-induction of P-enolpyruvate carboxykinase mRNA (tip = 45 min) as detected by translation of poly(A)' RNA in a cell-free protein synthesizing system. Also, translatable P-enolpyruvate carboxykinase mRNA was shown to be induced 8-fold in 90 min by the injection of BtzcAMP into carbohydrate-fed rats (6) or by the addition of the cyclic nucleotide to isolated hepatocytes incubated in vitro ( 5 ) .
In order to study the mechanisms responsible for these acute changes in enzyme mRNA, it is necessary to examine fluxes in the mRNA directly using a DNA probe to P-enolpyruvate carboxykinase. In the present communication, we report the construction and identification of a cloned, renal Penolpyruvate carboxykinase DNA. This probe, pPCK1, crosshybridizes with hepatic P-enolpyruvate carboxykinase mRNA. We have used it to demonstrate acute hormonally directed alterations in the relative levels of P-enolpyruvate carboxykinase mRNA in both rat liver and kidney.

EXPERIMENTAL PROCEDURES
Materials-Avian myeloblastosis virus reverse transcriptase was isolated as previously described (7). Nitrocellulose paper (BA185) was purchased from Schleicher and Schuell. [a-"2P]dCTP (specific activity, >600 Ci/mmol) was obtained from either New England Nuclear or Amersham/Searle and ["S]methionine was from Amersham/ Searle, Inc. Reticulocyte lysate translation kits were obtained from New England Nuclear and Eco RI and Pst I were from New England Biolabs. Guanidine isothiocynate was purchased from Fluka AG and triamcinolone acetonide (Kenalog 40) was a product of E. R. Squibb and Sons. Terminal transferase was purchased from P-L Biochemicals. Formamide, obtained from MCB Manufacturing Chemists, Inc., was deionized with AG 501 X 8 resin before use. Poly(A) (>300 bases) and rat liver tRNA, both from Sigma, were treated with proteinase K and phenol extracted before use (8). Oligo(dT)-cellulose and oligo(dT),r.,4 were purchased from Collaborative Research and deoxynucleotides were from Boehringer Mannheim Biochemicals. All other reagents were of analytical grade.
Construction of Recombinant Plasmids-A cDNA copy of the Penolpyruvate carboxykinase mRNA was prepared using avian myeloblastosis virus reverse transcriptase ((up) as described by Leis (7).
The plus DNA strand was synthesized in a similar reaction except that actinomycin D and oligo(dT),,.I, were omitted. The doublestranded DNA was then digested with nuclease SI and a poly(dC) sequence of approximately 20 bases was added to the 3'-ends using calf thymus terminal transferase by the procedure of Chang and Bollum (11). A poly(dG) (20 bases long) sequence was then added to 3'-OH ends of Pst I-treated pBR322 by a similar procedure. The plasmid and dsDNAs were then mixed and annealed to one another as described by Clarke and Carbon (12), except that annealing was continued for an additional 48 h a t room temperature.
Transformation of Escherichia coli HBlOl-The transformation of EK2 certified host E . coli HBlOl was performed in a P2 laboratory in compliance with the National Institutes of Health Guidelines for Research Involving Recombinant DNA Molecules. The procedure used was that described by Norgard et al. (13). Transformed colonies containing recombinant plasmids were selected for their resistance to tetracycline and sensitivity to ampicillin.
Identification of P-enolpyruvate Carboxykinase Clones-Transformed clones containing recombinant plasmids were grown on replica nitrocellulose filters, amplified on chloramphenicol-containing LB plates for 30 h, and treated as described by Grunstein and Hogness (14). Each colony was screened by hybridization to four different cDNA populations transcribed from mRNA which contained either high or low levels of translatable P-enolpyruvate carboxykinase mRNA. Those colonies which hybridized intensely to cDNAs prepared from mRNA extracted from kidneys of acidotic rats injected with triamcinolone and liver from diabetic animals, but which hybridized negligibly to cDNA made from spleen or mRNA from fetal rat liver, were selected and grown for further examination.
Plasmid DNA was isolated from clones selected by the above colony hybridization technique using the procedure described by Meagher et al. (15). These plasmid DNAs were rescreened by Northern blotting in order to select DNAs that hybridized to mRNA the size of P-enolpyruvate carboxykinase. The poly(A)' RNA isolated from liver or kidney was denatured (16) and separated by electrophoresis on an agarose gel containing 18% formaldehyde (17). After staining to determine the migration of the 18 S (2.3 kb) and 28 S (5.6 kb) RNA markers, the RNA was transferred to nitrocellulose in 20 X SSC essentially as described by Thomas (18). Plasmids which hybridized with an RNA species between 2.3 and 5.0 kb, the estimated size of P-enolpyruvate carboxykinase mRNA, were selected for further analysis.
Hybridization-selected Translation-Plasmid DNA from clones selected as outlined above, was linearized with Eco RI (19), denatured with alkali, bound to nitrocellulose Titers (20), and used to purify mRNA from a poly(A)' fraction from rat liver or kidney which contained high levels of translatable P-enolpyruvate carboxykinase mRNA. Poly(A)-containing RNA (40-80 pg) was incubated with a fiiter containing immobilized DNA for 6 h at 37 "C in 50% formamide, 25 mM Hepes, pH 7.0, 0.5 M NaCI, 1 mM EDTA, 0.58 SDS, 5 pg/ml of tRNA, 10 pg/ml of oligo(dA). The filters were washed in buffer without RNA at 37 "C followed by three successive washes in the same buffer without formamide (first wash), without NaCl (second wash), and without SDS (third wash). The bound mRNA was eluted by incubation for 2 min in 90% formamide, 25 mM Hepes, pH 7.0, 1 mM EDTA. The RNA was recovered by ethanol precipitation and translated in a rabbit reticulocyte protein synthesizing system.
Size of the Cloned P-enolpyruvate Carboxykinase DNA-Recombinant plasmid DNA was digested with Pst I using conditions described by New England Biolabs. The fragments were sized on a 5 8 polyacrylamide gel containing 90 m~ Tris-HC1, pH 7.5, 90 m~ boric acid, 4 mM EDTA (21). A Hue 111 digest of QXC174 was used for molecular weight markers.
Analysis of Sequence Abundance of P-enolpyruvate Carboxykinase mRNA by Northern Analysis-Northern analysis of RNA was performed as outlined above. Poly(A)' RNA was extracted (10) from the livers of rats that were: starved for 24 h; starved for 24 h then force-fed glucose (5 g/kg of body weight) for 2 h; starved for 24 h, force-fed glucose for 2 h, and then injected with B t s A M P (15 mg/kg of body weight) and theophylline (15 mg/kg of body weight) for 90 min; made diabetic by the injection of streptozotocin (70 mg/kg of body weight) for 4 days; made diabetic and then injected with insulin (5 units/kg of body weight) and force-fed glucose for 2 h. RNA was extracted from the kidneys of rats that were fed or made diabetic as above or made acidotic with NH4C1 gavage and then injected with triamcinolone (9). Twenty pg of each of the poly(A)+ mRNAs extracted from animals treated as described above were analyzed for the relative amount of P-enolpyruvate carboxykinase mRNA by Northern blotting (18). Plasmid '"P-DNA (lo8 cpm/pg) was prepared by nick translocation using E. coli DNA polymerase I (22). Hybridization was performed in 508 formamide, 5 X SSC, 0.54 SDS, 0.02% Ficoll, 0.02% polyvinylpyrrolidone, 10% dextran sulfate, 1.5 mg/ml of denatured salmon sperm DNA for 36 h a t 50 "C. The nitrocellulose papers were washed in 50% formamide, 5 X SSC, 0.02% Ficoll, 0.02% polyvinylpyrrolidone a t 37 "C and autoradiograms were prepared by the method of Swanstrom and Shank (23).

RESULTS AND DISCUSSION
Cloning of P-enolpyruvate Carboxykinase cDNA-P-enolpyruvate carboxykinase cDNA was synthesized from partially purified renal mRNA rather than from liver RNA for several reasons. The enzyme is immunochemically similar in both liver and kidney (24) and renal P-enolpyruvate carboxykinase mRNA can be easily induced to relatively high levels by acidosis and glucocorticoids (9). Also, kidney lacks albumin mRNA, a predominant hepatic mRNA, which is similar in size to the message for P-enolpyruvate carboxykinase. Renal poly (A)' was enriched about 15-fold for P-enolpyruvate carboxykinase mRNA by sedimentation in successive sucrose gradients. An SDS-polyacrylamide gel of the proteins translated in response to the added renal mRNA partially purified by this procedure is shown in Fig. 1. The RNA contained in fractions 7 and 8 were used for cDNA synthesis.
Prior to insertion into the plasmid pBR322, the newly synthesized cDNA was screened by Northern analysis to c o n f i that it hybridized to an mRNA of the correct size. Using this procedure, we noted that the cDNA hybridized with an RNA species migrating at 3.8 kb (data not shown). Previous studies using dimethyl sulfoxide sucrose gradients indicated that this was the approximate size of P-enolpyruvate carboxykinase mRNA (25).

Identification of pPCKl as a Clone to P-enolpyruuate
Carboxykinase-Identification of a plasmid carrying a Penolpyruvate carboxykinase insert (pPCK1) was performed using hybridization-selected translation (Fig. 2). Plasmid DNA was bound to nitrocellulose and hybridized with RNA extracted from the kidneys of acidotic rats induced with triamcinolone. The RNA hybridizing to the plasmid DNA was melted out and translated in a reticulocyte protein synthesizing system (Fig. 2). The predominant peptide translated in response to the selected RNA migrated with pure P-enolpyruvate carboxykinase (Fig. 2, lane 3 ) . This protein was positively identified as P-enolpyruvate carboxykinase by immunoprecipitation with antibody against the enzyme (Fig. 2, lane  4). Identical results were obtained if either renal or hepatic RNA was used as the source of the mRNA (data not shown). Our results thus indicate that the plasmid pPCKl contains  1 (left). Translation of partially purified mRNA from rat kidney. Renal mRNA isolated from acidotic rats which had been injected with triamcinolone was partially purified by sedimentation through two successive sucrose gradients as described under "Experimental Procedures." The RNA recovered from the second centrifugation (80% dimethyl sulfoxide, 5-20% sucrose gradient) was translated in a rabbit reticulocyte protein-synthesizing system and the proteins were separated by SDS-polyacrylamide gel electrophoresis. The arrom indicates the position of purified P-enolpyruvate carboxykinase (PEPCK) from rat kidney. Fractions 7 and 8 were used to synthesize cDNA as outlined under "Experimental Procedures. "   FIG. 2 (center). Identification of clone pPCKl carrying Penolpyruvate carboxykinase DNA by hybridization-selected translation. Poly(A)' RNA from the kidney of an acidotic rat injected with triamcinolone was hybridized to clone pPCKl which had been bound to nitrocellulose paper. The hybridized mRNA was released and translated in a cell-free protein-synthesizing system and sequences complementary to mRNA coding for full-length, immunoprecipitable enzyme. The size of the insert pPCKl was determined by polyacrylamide gel electrophoresis after Pst I digestion of the plasmid pPCKl (Fig. 3). The insert is approximately 250 bases in length as determined by comparison with known molecular weight markers generated by Hae I11 digestion of 6x174 DNA.

Hormonal Alterations in the Relative Amounts of P-enolpyruvate Carboxykinase mRNA in Rat Liver and Kidney-
A major unanswered question from previous studies with the enzyme is whether measurements of translatable RNA accurately reflect changes in total (hybridizable) P-enolpyruvate carboxykinase mRNA in vivo. The availability of a DNA probe specific for P-enolpyruvate carboxykinase now permits a direct measurement of the levels of P-enolpyruvate carboxykinase mRNA by a variety of hormonal conditions. Poly(A)-containing RNAs were isolated from a variety of dietary and hormonal conditions in which the translatable level of P-enolpyruvate carboxykinase mRNA had been measured (Table I). Each RNA was examined by Northern blotting using :'*P-labeled pPCKl as the probe specific for P-enolpy- ruvate carboxykinase mRNA. Northern analysis is a sensitive, semiquantitative technique which permits the estimation of relative sequence abundance of the enzyme mRNA under each condition selected. As shown in Fig. 4, pPCKl hybridizes to a RNA species at 23 S (3.8 kb) in both liver and kidney. No condition has been found where the size of the major hybridizing RNA species is altered. The degree of hybridization to the 23 S species mRNA is markedly altered by the hormonal stimuli (Fig. 4 ) in a pattern which agrees well with the levels of translatable mRNA for the enzyme presented in Table I. Starvation for 24 h causes an increase in the relative amount of P-enolpyruvate carboxykinase mRNA when compared to the fed animal. Feeding a starved animal glucose for 2 h lowers both translatable mRNA (Table I) and hybridizable mRNA (Fig. 4 ) to barely detectable levels. The levels of translatable mRNA are acutely raised by BkcAMP injection into starved, refed rats. As seen in Fig. 4, the levels of hybridizable mRNA for the enzyme were also rapidly increased by the CAMP injection. The highest level of hybridizable P-enolpyruvate carboxykinase mRNA was noted in diabetic rats. In these animals, the relative amount of enzyme mRNA could be rapidly lowered  (A)' RNA was prepared as described under "Experimental Procedures." Rats were starved for 24 h and when refed were given 5 mg/kg of glucose for 2 h by gavage. Those animals receiving BtZcAMP were fust starved, refed glucose for 2 h, and then injected with BtZcAMP (15 mg/kg of body weight) plus theophylline (15 mg/ kg of body weight). Diabetic rats received streptozotocin (70 mg/kg of body weight) by tail vein 4 days before killing. Diabetic animals receiving insulin were force-fed 5 mg/kg of glucose and injected with insulin (5 units/kg of body weight) for 2 h. Rats were made acidotic by NH,Cl gavage (10 mmol/kg of body weight) for 4.5 h. Triamcinolone (32 mg/kg of body weight) was injected intraperitoneally 90 min before the NHXI treatment. P-enolpyruvate carboxykinase was quantitated using a specific antibody either as referenced or by densimetric tracing of the immunoprecipitated protein after translation in the reticulocyte lysate system. Values are expressed as percentage of P-enolpyruvate carboxykinase mRNA relative to mRNA coding for total protein. See specific references for a more detailed treatment of the methods used. I U Liver Kidney  FIG. 4. Hormonally induced changes in the relative levels of P-enolpyruvate carboxykinase mRNA. The levels of P-enolpyruvate carboxykinase mRNA in the liver and kidney of animals treated as indicated were measured by Northern blotting using "*Plabeled pPCKl as a probe. The specific conditions for this experiment are given in detail under "Experimental Procedures." by the injection of insulin (Fig. 4).
The hormonal regulation of P-enolpyruvate carboxykinase mRNA was also examined in the kidney. No change in sequence abundance of this mRNA was found in kidneys ex-mboxykinase cDNA 10227 tracted from diabetic animals when compared to nondiabetic controls (data not shown). However, the relative amount of P-enolpyruvate carboxykinase mRNA was elevated in the kidneys of rats which had been made acidotic and injected with triamcinolone. Again, this is in agreement with the translational analysis of enzyme mRNA levels.
The mechanisms responsible for this turnover of P-enolpyruvate carboxykinase mRNA are not understood but the rapidity of the hormonal effects suggests that both synthesis and degradation play important roles in regulating the level of mRNA for this enzyme. Also, the apparent similarity between renal and hepatic forms of P-enolpyruvate carboxykinase noted in previous studies (24) is extended to the RNA level and c o n f i i e d with a DNA probe. This probe should allow us to examine the gene expression of P-enolpyruvate carboxykinase from two tissues which have different patterns of hormonal regulation.