/. Biochem. 86, 1301-1311 (1979)
Characterization of Polyriboadenylate Polymerase from Tetrahymena pyriformis Hisao UEYAMA 1 Department of Biochemistry, Kyoto Prefectural University of Medicine, Kamigyo-ku, Kyoto, Kyoto 602 Received for publication, May 17, 1979
Poly(A) polymerase [polyadenylate nucleotidyltransferase, EC 2.7.7.19] was extracted from Tetrahymena pyriformis. The enzyme was demonstrated to be present in three forms by column chromatography on DEAE-cellulose, and they were termed poly(A) polymerase la, Ib, and II in order of increasing affinity to the column. The properties of enzymes la and Ib were similar except that la utilizes poly(A) as a primer rather efficiently. Enzyme II differed from enzymes la and Ib not only in elution profile on DEAE-cellulose column chromatography but also in pH and temperature preferences, molecular weight, requirement for divalent cations, sensitivity to salts at high ionic strength, optimal primer concentration, and subcellular localization. The molecular weights of enzymes la and Ib measured by gel filtration were both 43,000, and that of enzyme II was 95,000. All three enzymes required Mn ! + for maximal activity; Mg t + could replace Mn 1+ in the reaction of enzyme II, but only partially. In the presence of 0.1 M ammonium sulfate the activities of enzymes la and Ib were both completely inhibited, whereas enzyme II still showed 42% of its original activity. These findings suggest that there are two distinct types of poly(A) polymerase in Tetrahymena pyriformis.
PoIy(A) polymerase is thought to be an enzyme that post-transcriptionally synthesizes poly(A) chains at the 3'-termini of heterogeneous nuclear RNA (hnRNA) and mRNA of eukaryotic cells (13). In the cases of N.I.H.-Swiss mouse embryos (4) and mouse lymphoma cells (5), only one form of poly(A) polymerase was extracted, which probably carries out the above function in the cells. On the other hand, the existence of more than one poly(A) polymeiase in a single cell type has been reported (2, 3, 6-10), but the significance of 1
Present address: Department of Medical Biochemistry, Shiga University of Medical Science, Seta, Ohtsu, Shiga 520-21. Vol. 86, No. 5, 1979
1301
such multiple forms of poly(A) polymerase is still unknown (11). Poly(A) polymerase has not previously been isolated from Tetrahymena pyriformis, which has been used for several studies on nucleic acid metabolism, including work on DNA polymerases (12, 13) and RNA polymerases (14, 15), although poly(A) chains of 80-150 nucleotides (16) or 4S (17) were found to be attached to its hnRNA and mRNA. In order to determine how poly(A) polymerase is involved in the synthesis and turnover of mRNA in Tetrahymena pyriformis, it is necessary to extract and characterize this enzyme. The present paper reports the possible occurrence of two distinct types of poly(A) polymerase in this organism.
1302
H. UEYAMA
MATERIALS AND METHODS Materials—a-Amanitin, actinomycin D, rifampicin, and cordycepin (3'-deoxyadenosine) 5'triphosphate were obtained from Sigma. Poly(A), poly(C), and poly(U) were from P-L Biochemicals, poly(G) from Sigma, poly(l) from Miles Laboratories and tRNA from Boehringer Mannheim. tRNA having a 3'-phosphate end was prepared with sodium periodate and aniline (75). Tetrahymena RNA and rat liver rRNA were extracted from ribosomal pellets by phenol-sodium dodecyl sarcosinate (SDSa) treatment {19). It was observed by sucrose density gradient centrifugation that rat liver RNA has two main peaks of 18S and 28S, whereas Tetrahymena RNA has a broad distribution with one peak smaller than 18S rRNA. Phosphoenolpyruvate and pyruvate kinase were obtained from Boehringer Mannheim. [2-'H]ATP, [U- 14 C]GTP, [U- 14 C]UTP, [U- 14 C]CTP, [8- s H]poly(A), [8-"C]ADP, and [a-"P]ATP were from the Radiochemical Centre. Hexylene glycol was purchased from Eastman. Preparation of the Crude Extract—Tetrahymena pyriformis (the amicronucleate strain) was cultured aerobically with 2 % peptone (Mikuni, Japan) and 0.5% yeast extract (Difco). After incubation at 28°C for 36 h (cell density 1.5-3 x 107ml), cells were collected by centrifugation at 1,300 xg for 10 min, and washed with buffer A (50 rriM Tris-HCl, pH 7.6, containing 2 IHM 2mercaptoethanol). The pellet was resuspended in buffer A and stored below —20°C. All purification procedures after this step were carried out at 0-4°C. The thawed cells were sonicated for 1 min with an Insonator (Kubota, Japan) at 1.8 A (140 W). The sonicated cell suspensions were centrifuged at 35,000 x g for 20 min. The resulting supernatant was centrifuged at 122,000xg for 120 min. The supernatant was used as the enzyme preparation and the pellet was used for extraction of RNA. Assay of Poly(A) Polymerase—The standard assay mixture contained 50 mM Tris-HCl buffer (pH 8.5), 5 mM 2-mercaptoethanol, 1 mM MnCl,, 0.5 mg of Tetrahymena RNA, 10-100 fi\ of enzyme, and 0.5 mM [ 8 H]ATP (2 ftCi/fimol), in a volume of 250 [t\. After incubation for 30 min at 28°C, 1 ml of ice-cold 5 % trichloroacetic acid (TCA)-5 mM
sodium pyrophosphate (NaPPi) was added to the reaction mixture and the acid-insoluble precipitate was collected on a glass membrane filter (GF/C, 2.4 cm, Whatman). After washing twice with 10 ml of 5% TCA-5mM NaPPi, the filters were put into glass vials to which 10 ml of PCS (Amersham/Searle) was added. The radioactivity was measured in a Packard scintillation spectrometer (model 3255). Protein was determined by the biuret method, Lowry's method or WarburgChristian's method (20) depending on its concentration and the presence of ammonium sulfate in the enzyme preparation. One unit of the enzyme was defined as the amount of enzyme which catalyzes the incorporation of 1 nmol of AMP into acid-insoluble precipitate in 30 min. Gel Filtration—A column (1.5x90 cm) of Ultrogel (ACA44, LKB) previously equilibrated with 50 mM Tris-HCl buffer (pH 7.6) was loaded with 1.0 ml of sample and developed with the same buffer, collecting 3 ml fractions. The molecular weight of each poly(A) polymerase was estimated using the following markers: myoglobin (Mr = 17,000), soybean trypsin inhibitor (Mr=21,5OO), ovalbumin (Mr=45,0OO), and bovine serum albumin (Mr=68,000). When gel filtration was used at the last step in the purification of poly(A) polymerases, 2-4 ml of each enzyme sample was loaded on the column and 3 ml fractions were collected. Analysis of the Reaction Product—The reaction conditions were as described above except that the specific activity of [ J H]ATP used was 20 ftC\/ftmo\. The glass membrane filter ( G F / Q used to collect acid-insoluble precipitate was soaked in 0.5 ml of 0.3 N KOH and incubated for 18 h at 37°C. The hydrolysate was neutralized with 1 N perchloric acid and centrifuged. Twenty fi\ of the supernatant was developed by ascending paper chromatography with the following solvent: w-butanol/acetone/acetic acid/5 % NH 4 OH/H,O (35 : 25 : 15 : 15 : 10, by vol.). After development for 10 h the radioactivities of spots detected by ultraviolet absorption were determined. Isolation of the Nuclei—The cells collected by centrifugation (1,300 x g, 5 min) were washed once with buffer D (0.05 mM PIPES-Na, pH 7.5, containing 0.5 M hexylene glycol and 1 mM CaCl,) and then resuspended in the same buffer. The cell homogenate obtained with a Potter-Elvehjem /. Biochem.
POLY(A) POLYMERASE OF Tetrahymena
1303
homogenizer was centrifuged at 1,500 xg for 5 min. The pellet was washed once with buffer D and then suspended in buffer D containing 1.6 M sucrose. The suspension was layered on an equal volume of buffer D containing 2 . 1 M sucrose in SW27 rotor tubes and centrifuged at 55,000Xg for 60 min. The nuclear pellet obtained was washed once with buffer D and used to investigate the localization of the enzymes. DNA was determined by the method of Burton (21) and RNA by means of the orcinol reaction (22). Assay of Poly (A)-Degrading Enzyme—The assay mixture was as follows: 50 ITIM Tris-HCl (pH 8.5), 1 rain MnCl,, 5 mM 2-mercaptoethanol, 0.5 mM [»H]poly(A) (0.1 ftCi/ftmol nucleotides), enzyme, and 0.5 mM ATP, in a volume of 250 //I. After incubation of the reaction mixture at 28°C for 30 min, 100 //I of 5 mg/ml RNA, followed rapidly by 650 ft\ of 7.7 % TCA, was added and the mixture was centrifuged. The radioactivity of 0.5 ml aliquots of the acid-soluble supernatant was measured in the liquid scintillation spectrometer.
RESULTS Purification of Poly (A) Polymerase—The crude extract (740 ml) from 450 g of Tetrahymena cells was subjected to DEAE-cellulose column chromatography (4 X 30 cm). A typical elution profile is shown in Fig. 1. The enzyme activity was found in two fractions, i.e., the washing with the starting buffer (buffer A) and the fraction eluted with ammonium sulfate. The enzyme which was not adsorbed on the column was termed poly(A) polymerase I and the enzyme eluted from the column with 0.25 M ammonium sulfate was termed poly(A) polymerase II. These enzymes were further purified as follows. The pH of the enzyme I preparation (1,030 ml) was adjusted to 8.5 from 7.6 with 50 mM Tris2 mM 2-mercaptoethanol. The solution (2,060 ml) was concentrated to 530 ml with a hollow fiber apparatus (DC 2, Amicon) and applied to a column of DEAE-cellulose (1.5x30 cm) previously equilibrated with buffer B (50 mM Tris-HCl, pH 8.5,
'o x E
« o
w O
a. o o _c 0. 5
10
20
30 Fraction
40
SO
Fig. 1. 1st DEAE-cellulose column chromatography of po'y(A) polymerases from Tetrahymena pyriformis. The crude extract obtained from 20 g of Tetrahymena cells as described in " MATERIALS AND METHODS" was applied to a column of DEAE-cellulose (1.0x30cm) previously equilibrated with buffer A (50 mM Tris-HCl, pH 7.6, containing 2 mM 2-mercaptoethanol). The column was washed with buffer A, then eluted stepwise with ammonium sulfate, 0.05 M and 0.25 M. PoIy(A) polymerase activity was assayed under standard assay conditions as described in "MATERIALS AND METHODS." • , Protein concentration; O, po'y(A) polymerase activity; , ammonium sulfate concentration. Vol. 86, No. 5, 1979
1304
H. UEYAMA same buffer, the column was eluted with a linear gradient from 0 to 0.3 M NaCl. The fractions containing enzyme activity (fractions 27 through 34) were combined, then concentrated to 6.2 ml with a Mini-Module hollow fiber apparatus (Asahikasei, Japan), and applied to a column of Ultrogel. The eluted fractions showing enzyme activity were rechromatographed on the Ultrogel column after concentration to 2.2 ml with the Mini-Module. Poly(A) polymerase la was collected in fractions 28 and 29 (Fig. 2), and the molecular weight of enzyme la was estimated to be 43,000. The yield was 25 % and the degree of purification was 104-fold (Table I). The final enzyme preparation had no poly(A)-degrading activity. Enzyme Ib (170 ml) was concentrated with the Mini-Module to 26 ml and dialyzed against 5 liters of 10 mM Na acetate buffer (pH5.5) containing 2mM 2mercaptoethanol. The dialyzed solution was
o
5
05
AMP Inco rporated
(dpmx 10 )
containing 2 mM 2-mercaptoethanol). Most of the enzyme activity (la), which utilized poly(A) as a primer much more effectively than Tetrahymena RNA, was eluted by washing with buffer B and a small portion of the original activity (Ib), which utilized the RNA in preference to poly(A) as a primer, was adsorbed on the column and then eluted with buffer A. Enzyme la fraction (690 ml) was brought to 0.45 saturation with solid ammonium sulfate and the mixture was centrifuged at 15,000 xg for 15 min. The supernatant (750 ml) was then brought to 0.55 saturation with ammonium sulfate and centrifuged. The pellet was dissolved in 25 ml of buffer C (10 mM MES-Na, pH 6.0, containing 2 mM 2-mercaptoethanol) and dialyzed against 5 liters of the buffer. The dialyzed enzyme solution (28 ml) was applied to a column of CM-cellulose (1.5x30 cm) previously equilibrated with buffer C and, after washing with the
I
5 LOG (Mr)
20
30
40
50
Fraction Fig. 2. Gel nitration chromatography on Ultrogel of poly(A) polymerases from Tetrahymena pyriformis. Each poly(A) polymerase solution was applied to a column of Ultrogel (1.5x90 cm) equilibrated with 50 mM Tris-HCl buffer (pH 7.6), collecting 3 ml fractions. Three independent elution profiles are superimposed in the figure to compare the elution positions of the poly(A) polymerases. The void volume (KJ is indicated by an arrow. The poly(A) polymerase activity was assayed under the standard assay conditions as described in " MATERIALS AND METHODS." D, Poly(A) polymerase la; • , poly(A) polymerase Ib; • , poly(A) polymerase II. The inset shows the determination of the molecular weight of each poly(A) polymerase. Myoglobin ( • ) ; soybean trypsin inhibitor ( • ) ; ovalbumin (O); bovine serum albumin ( • ) ; poly(A) polymerase la and Ib ( A ) ; poly(A) polymerase II (A). J. Biochem.
POLY(A) POLYMERASE OF Tetrahymena
1305
applied to a column of CM-cellulose (1.5 x 30 cm) 30 cm) previously equilibrated with buffer A. The equilibrated with the same buffer. The column column was washed with buffer A containing was washed with the starting buffer and then 0.05 M ammonium sulfate, then eluted with a linear eluted with a linear gradient of NaCl (0-0.3 M). , gradient from 0.05 to 0.15 M ammonium sulfate. The fractions containing enzyme activity were The fractions containing enzyme activity (fractions combined, concentrated to 3.9 ml with the Mini23 through 45) were combined, concentrated to Module and applied to a column of Ultrogel. 3.8 ml with the Mini-Module and applied to a The column chromatography on CM-cellulose was column of Ultrogel. Poly(A) polymerase II was effective in removing poly(A)-degrading activity, eluted in fractions nearer to the void volume than but the increase in the specific activity of enzyme polymerases la and Ib (Fig. 2). The molecular Ib was small (Table 1). The molecular weight of weight of this enzyme was estimated to be 95,000. enzyme Ib was estimated to be 43,000 by gel The recovery was 27 % and the degree of purificafiltration. tion was 315-fold. The final enzyme preparation was found to be free from poly(A)-degrading Enzyme II (400 ml) was first fractionated by activity. The above purification procedures are ammonium sulfate precipitation. The active ensummarized in Table I. zyme protein precipitated with ammonium sulfate between 0.33 and 0.45 saturation was dissolved in Properties of Poly (A) Polymerases—All three 10 ml of buffer A and dialyzed against 2 liters of polymerases required an RNA primer and showed the same buffer. The dialyzed enzyme solution no requirement for a DNA template (Table II). (12.4 ml) was applied to a column of Sepharose Among primers tested, RNA obtained from the 6B (1.8x40 cm) equilibrated with buffer A. The ribosomal pellet of Tetrahymena pyriformis was active fractions (17-21) were combined and then most effective for poly(A) polymerase Ib, and applied to a column of DEAE-cellulose (1.5 x Tetrahymena RNA, tRNA, or rRNA of rat liver TABLE I. Purification of poly (A) polymerases from Tetrahymena pyriformis. Enzyme fraction Crude extract6 1st DEAE-cellulose
II
Specific activity1 (units/mg)
Total enzyme1 (units)
12,300
1.10
13,500
5,880
0.818
4,810
Yield (%)
Hnd DEAE-cellulose 45-55% (NH4),SO, CM-cellulose 1st Ultrogel Hnd Ultrogel
2,930 633 106 30.7 6.62
1.03 2.75 12.9 39.4 114
3,020 1,740 1,370 1,210 756
100 58 45 40 25
Ilnd DEAE-cellulose Ib CM-cellulose Ultrogel
566 50.8 3.24
1.86 2.13 22.0
1,050 108 71.4
100 10 7
1st DEAE-cellulose 33-45% (NHJ.SO, Sepharose 6B Ilnd DEAE-cellulose Ultrogel
859 250
7,800 6,220 4,900 4,540 2,140
100 80 63 58 27
la I
Total protein (mg)
101 40.7 6.18
9.09 24.9 48.5 112 346
»~The primer used was RNA obtained from the ribosomal pellet of Tetrahymtna. 450 g of Tetrahymena cells. Vol. 86, No. 5, 1979
b
The starting material was
H. UEYAMA
1306
TABLE IT. Primer requirements of poly (A) polymcrases from Tetrahymena pyriformis. The reaction mixture contained 50 mM Tris-HCl buffer (pH 8.5), 5 mM 2-mercaptoethanol, 1 mM MnCl t , 0.5 unit of enzyme, 0.5 DIM [*H]ATP and various primers, in a volume of 250 pi\. Just before termination of the reactions after incubation at 28°C for 30 min, 0.05 ml of 5 mg/m] RNA was added to the reaction mixtures as a carrier. The concentration used was 1 mg/ml for all primers except Tetrahymena RNA and rat 4iver rRNA (2 mg/m] for the former and 1.5 mg/ml for the latter). The denatured DNA was prepared by heating the native DNA at 99°C for 10 min, then cooling it rapidly. The activated DNA was prepared by DNase treatment. Poly(A) polymerase Primers Tetrahymena RNA —primer Poly(A) Poly(Q • PoIy(U) PoIy(G) Poly© Rat liver rRNA tRNA (3'-CCA-0H) tRNA (3'-CC-P) tRNA (3'-CC-OH) Calf thymus native DNA Calf thymus denatured DNA Calf thymus activated DNA
la
Ib
n
100
100
100
0
0
0
324
1
10
1 14 8 10 80 95 2 48 12 10
0 0 4 0 21 52 18 66 1 3
4 0 11 3 82 152 1 351 5 2
7
4
5
was effective for enzyme II. Only poly(A) polymerase l a could utilize poly(A) effectively as a primer. A free 3'-terminal hydroxy group must be essential for the enzyme reactions, because t R N A having a 3'-phosphate end (tRNA-CCP) was far less active than normal tRNA, while tRNA having a 3'-CC-OH end, prepared from the tRNAC C P by phosphatase treatment, was as active as normal t R N A (Table II). The effects of R N A concentration on the poly(A) polymerase reaction were different among polymerases la, Ib, and II. Polymerase Ib required about the same concentration of R N A as in the standard assay conditions (2 mg/m]) for maximal activity, while polymerases l a and II showed
maximal activities at one-half (1 mg/ml) and onequarter (0.5 mg/ml) of this R N A concentration, respectively. Poly(A) polymerases la and Ib differed from polymerase II in p H preference. Enzymes la and Ib had pH optima at 8.5 but that of II was at 9.0. With regard to the optimal temperature of the reaction, la and Ib were most active at 28°C, which is also optimal for the growth of Tetrahymena, while II was most active at 33°C. Divalent cations were absolutely required for poly(A) polymerase activity, and Mn ! + was more effective than Mg I + . The optimal concentrations of M n ' + and Mg I + for enzyme II were 1-2 mM and 2-4 mM, respectively, but enzymes la and Ib could be activated only by Mn 2 + , as shown in Fig. 3. Mg I + , evenat higher concentrations than indicated in Fig. 3 (up to 20 mM), could not replace Mn 1 + in the reactions of enzymes la and Ib. Polymerase II differed from polymerases I in the effect of salts at high ionic strength. As shown in Fig. 4, in the presence of 0.1 M ammonium sulfate the activities of poly(A) polymerases la and Ib were completely depressed but enzyme II still showed 42 % of its original activity. Similar results were obtained with NaCl and KC1. All three poly(A) polymerases incorporated ATP most efficiently into acid-insoluble precipitate. The incorporation of other labeled ribonucleoside triphosphates was less than 17% of that of A T P ; this specificity for ATP is thought to be an important characteristic of a poIy(A) polymerase {11). None of the three polymerases was affected by DNA-dependent R N A polymerase inhibitors such as ar-amanitin, actinomycin D, and rifampicin. All three poly(A) polymerases were inhibited by other ribonucleoside triphosphates at a concentration equimolar with substrate ATP (0.5 mM). Reversal of the inhibiting effect of the ribonucleotides could not be observed on further addition of Mn 1 + , so these ribonucleotides seem to inhibit poly(A) polymerase activities, but not by chelating M n l + . The addition of 0.5 mM A D P did not affect polymerase n , suggesting that the substrate is not A D P but ATP. However, the activities of poly(A) polymerase l a and Ib were inhibited about 50% by 0.5 mM ADP. When [ " Q A D P at 0.5 mM was used as a substrate in the reaction of enzyme la or Ib, incorporation of the radioactivity into acid-insoluble precipitate was found to be less than
J. Biochem.
1307
POLY(A) POLYMERASE OF Tetrahymena
100
-O
O
1
2
3
4
Divalent Cations (mM)
Fig. 3. Divalent cation requirements of poly(A) polymerases from Tetrahymena pyriformis. The reaction mixture contained 50 mM Tris-HCI buffer (pH 8.5), 5 mM 2-mercaptoethanol, 0.5 mg of RNA obtained from the ribosomal pellet of Tetrahymena, 100 p\ of enzyme, 0.5 mM PH]ATP, and divalent cations at various concentrations, in a volume of 250 ftl. Maximal activity of each enzyme was obtained with 1 mM MnClt, amounting to 1,408 disintegrations per min (dpm), 6,021 dpm and 5,845 dpm with enzymes la, Ib, and II, respectively. Poly(A) polymerase la, —D— with MnCl,, - - • - - with MgCl,; poly(A) polymerase Ib, — • — with MnCl., - - • - - with MgCl,; poly(A) polymerase II, — • — with MnCl,, - - • -- with MgCl,.
10
20 30 Ammonium Sulfate (mM)
100
Fig. 4. Effect of ammonium sulfate on poly(A) polymerases from Tetrahymena pyriformis. The control assay mixture contained 50 mM Tris-HCI buffer (pH 8.5), 5 mM 2-mercaptoethanol, 1 mM MnCl,, 0.5 mg of RNA obtained from the ribosomal pellet of Tetrahymena, 100 fi\ of enzyme, and 0.5 mM PHJATP, in a volume of 250 /il. AH enzymes used were ammonium sulfate-free after gel filtration on Ultrogel. The control enzyme activities were 1,550 disintegrations per min (dpm), 3,577 dpm, and 2,666 dpm with enzymes la, Ib, and II, respectively. D, Poly(A) polymerase la; • , poly(A) polymerase Ib; • , poly(A) polymerase II.
11 % of that of ATP, and when 2 mM pohsphoenolpyruvate and 10 units of pyruvate kinase were added as an ATP-generating system, the incorporation was greatly enhanced. These results suggest Vol. 86, No. 5, 1979
that the true substrate for the polymerizing reaction of enzyme la or Ib is not ADP but ATP. Cordycepin 5'-triphosphate, which is probably the active form of cordycepin that inhibits poly(A)
1308
H. UEYAMA
synthesis in vivo, inhibited all three enzyme activities to similar extents (40-60%) at a concentration equimolar with substrate ATP (0.5 ITTM). Products of Poly(A) Polymerase Reactions— The reaction products were hydrolyzed and the radioactivities of 2'(3')-AMP and adenosine were compared. As shown in Table in, the reaction product of each poly(A) polymerase under the conditions of the standard assay was oligo(A) consisting of less than 10 AMP residues. Natural poly(A) found on nuclear and cytoplasmic mRNAs in Tetrahymena pyriformis was reported to be 80-150 nucleotides in length (75), and there is a major difference in length between the natural poly(A) and the reaction products. However, the chain length of poly(A) polymerizing reaction products in vitro has been reported to be dependent on primer concentration (3, 9, 23) and time of
TABLE m . Average chain length of the reaction products of each poly (A) polymerase from Tetrahymena pyriformis. The reaction mixture was as described in "MATERIALS AND METHODS." The primer, other than poly(A), was RNA obtained from the ribosomal pellet of Tetrahymena pyriformis. After incubation of the reaction mixture at 28°C for 30 min, the reaction was terminated with 1 ml of ice-cold 5% TCA containing 5 mM NaPPi. The acid-insoluble precipitate collected on a glass membrane filter (GF/C) was hydrolyzed with 0.3 N KOH at 37°C for 18 h. Twenty ft\ of the supernatant obtained by centrifugation of the hydrolysate neutralized with 1 N HC1O4 was developed by ascending paper chromatography. The recovery of radioactivity was 85-90% in all cases. Poly(A) polymerase
Primer concentration (mg/ml)
Average chain length
0.5
2.0
1.0 2.0 poly(A) 1.0
1.7 1.4 2.9
Ib
0.5 1.0 2.0
4.6 3.9 3.2
U
0.25 0.5 1.0
30.0 24.2 16.2
2.0
8.8
la
incubation (3, 23). Decreasing the concentration of primer RNA resulted in a progressive elongation of the average chain length of the product of enzyme II, but had little effect on that of enzyme Ib (Table HI). The product of enzyme la was also oligo(A) even with a lower concentration of RNA or poly(A) as a primer. The covalent linkage of the reaction product to RNA primer was examined by sucrose density gradient centrifugation and nearest-neighbor analysis. In the former experiment, 28S rRNA purified from rat liver was used as a primer. After incubation of the reaction mixture containing 28S rRNA for 30 min, the reaction was terminated with ice-cold 90% ethanol containing 2% potassium acetate and 0.01% SDSa. The RNA extracted from the precipitate by phenol-SDSa treatment after centrifugation at 3,000 xg for 15 min was dissolved in 0.05 M Na acetate buffer (pH 5.1) and layered on a 10-30% linear sucrose gradient {19) in a Spinco SW 27 rotor tube. Sucrose density gradient centrifugation was carried out at 65,000 x g for 18 h at 8°C. The centrifuged sample was fractionated with an Auto Densi-Flow (Searle) into 31 fractions. Radioactivity was found in 28S rRNA. To confirm that the reaction product is covalently linked to 3'-termini of the primer, tRNA having a 3'-CC-OH end and [a-38P]ATP were used in the latter experiment. After incubation of the reaction mixture at 28°C for 30 min, the reaction was terminated with 5% TCA-5 mM NaPPi. The acid-insoluble precipitate collected on GF/C was soaked in 0.5 ml of 1 N KOH and incubated for 16 h at 37°C. The neutralized' hydrolysate with 1 N perchloric acid was centrifuged and 20 fi\ of the supernatant was subjected to high-pressure liquid chromatography (/iBondapak/NH,, Waters Associates Inc.). After development with 0.01 M NH4H,PO« (pH 3.0), the radioactivity of each 2'(3')-mononucleotide was measured in a liquid scintillation spectrometer. It was found that most (82-96%) of the radioactivity (total 2,660-52,100 cpm) was recovered in 2'QO-CMP and 2'(3>AMP, although the ratios of radioactivities of these two nucleotides (AMP to CMP) were different among the three enzymes: 12.4, 0.14, and 0.21 for enzymes la, Ib, and n , respectively. The above results suggest that the reaction products are covalently linked to 3'termini of the RNA primer. J. Eiochem.
POLY(A) POLYMERASE OF Tetrahymena
Nuclear Localization of Poly (A) Polymerase II—It was reported that nuclear poly(A) polymerase in calf thymus (2, 3, 6) and HeLa cells (10) can be activated by either MnI+ or Mgl+, and that • cytoplasmic poly(A) polymerase can be activated only by MnI+. In order to study whether enzymes I and II also differ in subcellular localization, nuclei of Tetrahymena were isolated. The hexylene glycol method used was preferred to both aqueous and nonaqueous techniques since the procedure is simple and leakage of soluble nuclear components is prevented (24). The nuclei were purified over 14.7-fold from the whole cell, based on the ratio of DNA to protein and of RNA to DNA. The nuclear pellet obtained from 26 g of Tetrahymena cells was suspended in 9 ml of buffer B, sonicated at 1.8 A for 1 min, and then centrifuged at 105,000 xgr for 1 h. The poly(A) polymerase activity of the resulting supernatant was resistant to ammonium sulfate at 0.1 M, suggesting that it may be due to poly(A) polymerase II. The supernatant was applied to a column of DEAE-cellulose (0.3x30 cm) previously equilibrated with buffer B and then the column was eluted with a linear gradient from 0 to 0.3 M ammonium sulfate. Poly(A) polymerase activity was found only in fractions 16 through 20, showing that only enzyme II was present in the nuclei. DISCUSSION Poly(A) polymerase of Tetrahymena pyriformis was separated in three forms by column chromatography on DEAE-cellulose. The three enzymes were further purified and classified into two types (la, Ib, II) on the basis of their properties. These enzymes required RNA as a primer, not a DNA template. They can be easily distinguished from RNA polymerase type II because of their insensitivity to a-amanitin. It is also clear that they differ from RNA polymerase types I and III in sensitivity to salts at high ionic strength, requirement for divalent cations and molecular weight. The high sensitivity of poly(A) polymerases I to ammonium sulfate is not compatible with the report that cr-amanitin-insensitive RNA polymerases of Tetrahymena are not affected by the salt up to 50 mM (75). It was reported that RNA polymerases of Tetrahymena can be activated by M n " and Mg1+ to similar extents (15), while Vol. 86, No. 5, 1979
1309
these three poly(A) polymerases are activated by Mn1+ more effectively than by MgI+. The molecular weights of eukaryotic RNA polymerases (25) are far larger than those of the poly(A) polymerases. There is a possibility that polynucleotide phosphorylase and tRNA nucleotidyltransferase are involved in the poly(A) polymerase-detecting system when the incorporation of a labeled nucleotide into acid-insoluble precipitate is measured. The former enzyme is known not to be present in eukaryotic cells (11), but some reports on eukaryotic polynucleotide phosphorylase have appeared (26, 27). However, the properties of the three enzymes are not consistent with those of polynucleotide phosphorylase, which requires not ATP but ribonucleoside diphosphates and can be activated only by MgI+ (26). It was reported that tRNA nucleotidyltransferase utilizes tRNA as the best primer but can also use rRNA to some extent (28), and it is possible that some of the poly(A) polymerases are identical with this enzyme. However, poly(A) polymerase II seems to be different from tRNA nucleotidyltransferase, because the reaction products of poly(A) polymerase II were poly(A) of 9-30 nucleotides in length, which is not consistent with the observation that only a single AMP residue is added to the primer by tRNA nucleotidyltransferase (28). Besides, poly(A) polymerase II is present in nuclei, whereas tRNA nucleotidyltransferase has been reported to exist not in nuclei but in the cytosol and mitochondria (29). As the reaction products of poly(A) polymerase la or Ib were oligo(A) with more than 2 residues, these enzymes are not identical with tRNA nucleotidyltransferase, either. It has been reported (30) that tRNA nucleotidyltransferase of Tetrahymena pyriformis incorporates CTP to the same extent as ATP, that Mg1+ is twice as effective as Mn1+ on this enzyme activity and that tRNA is utilized as a primer much more efficiently than rRNA (14 times) by this enzyme. On the contrary, poly(A) polymerases la and Ib incorporate not CTP but only ATP (the incorporation of CTP by enzymes la and Ib was 4% and 0% of that of ATP, respectively), Mg1+ was not effective on the poly(A) polymerases la and Ib reactions, and tRNA was utilized only 1.2-2.5 times more efficiently than rRNA as a primer by enzyme la or Ib. Thus, it seems likely that all three enzymes
1310
arc poIy(A) polymerase or polyadenylate nucleotidyltransferase [EC 2.7.7.19]. The next problem is whether poIy(A) polymerases la and Ib are identical or not. The properties of polymerase II were quite dissimilar to those of polymerases la and Ib, and it is clear that enzyme II is different from enzymes la and Ib. If the enzyme Ib preparation contains poly(A)degrading activity and conversely the enzyme la preparation does not, the difference that enzyme la utilizes poly(A) as a primer much better than enzyme Ib might be accounted for, but the final enzyme preparations of la and Ib, both free of poly(A)-degrading activity, still showed a difference in the utilization of poly(A) as a primer. The properties of the two enzymes were similar, except for the primer specificity, and therefore they are classified as type I poly(A) polymerase in this report, though further studies are clearly necessary. The significance of the occurrence of two types of poly(A) polymerase in Tetrahymena pyriformis cannot be easily deduced from the properties of the enzymes obtained. It is possible that enzyme la or Ib is mitochondrial in origin, but no poly(A) polymerase activity was detected in the extracts of mitochondria isolated by two different methods (31, 32). Therefore, poly(A) polymerase la and Ib probably exist in the extramitochondrial cytoplasm. In yeast, it is considered that one of the poly(A) polymerases found in the nuclei may add a short oligo(A) sequence to mRNA for initiation of the poly(A) tracts of the mRNA and that the other enzyme, preferring poly(A) itself, could elongate the short oligo(A) tract (7). However, the properties of the poly(A) polymerases purified from Tetrahymena pyriformis are not consistent with such a sequential mechanism. Since poly(A) was inactive as a primer for enzyme Ib, this enzyme may be involved in the initiation process of poly(A) synthesis. However, cordycepin 5'-triphosphate, which has been reported to inhibit the initiation of poIy(A) synthesis at far lower concentrations than those of the substrate ATP (33, 34), affected the activity of enzyme Ib by only 56% at a concentration equimolar with ATP. The absence of this enzyme in the nuclei also seems inconsistent with the above hypothesis. Recently, it was confirmed that poly(A) synthesis in HeLa cells and mouse sarcoma ascites
H. UEYAMA cells involves two steps: de novo synthesis and terminal addition (33, 35). Poly(A) polymerase la may be involved in the terminal turnover of poly(A) tracts in the cytoplasm, because poIy(A) is an effective primer for this enzyme. On the other hand, poly(A) polymerase II probably functions in the de novo synthesis and terminal addition of poly(A) in the nuclei. This hypothesis is compatible with the report that the nucleus is the site of de novo synthesis of poly(A), and that 3'-terminal addition of poly(A) occurs both in the nuclei and the cytoplasm (35). The author wishes to express his gratitude to Dr. T. Matsuura and Prof. K. Ueda, Shiga University of Medical Science, for their valuable suggestions and discussions during the course of this work. The author also wishes to thank Prof. A. Iwashima, Kyoto Prefectural University of Medicine, for his helpful advice in the preparation of this manuscript. REFERENCES 1. Darnell, J.E., Jelinek, W.R., & Molloy, G.R. (1973) Science 181, 1215-1221 2. Winters, M.A. & Edmonds, M. (1973) /. Biol. Chem. 248, 4756-4762 3. Winters, M.A. & Edmonds, M. (1973) J. Biol. Chem. 248, 4763^1768 4. Hadidi, A. & Sethi, V.S. (1976) Biochim. Biophys. Ada 425, 95-109 5. Mtiller, W.E.G., Schroder, H.C., Arendes, J., Steffen, R., Zahn, R.K., & Dose, K. (1977) Eur. J. Biochcm, 76, 531-540 6. Tsiapalis, CM., Dorson, J.W., &Bollum, F.J. (1975) /. Biol. Chem. 250, 4486-4496 7. Haff, L.A. & Keller, E.B. (1975) J. Biol. Chem. 250, 1838-1846 8. Pellicer, A., Salas, J., & Salas, M.L. (1978) Biochim. Biophys. Ada 519, 149-162 9. Niessing, J. (1975) Eur. J. Biochem. 59, 127-135 10. Nevins, J.R. & Joklik, W.K. (1977) /. Biol. Chem. 252, 6939-6947 11. Edmonds, M. & Winters, M.A. (1976) in Progress in Nucleic Acid Research and Molecular Biology (Cohn, W.E., ed.) Vol. 17, pp. 149-179, Academic Press, New York 12. Westergaard, O. & Lindberg, B. (1972) Eur. J. Biochem. 28, 422-431 13. Crerar, M. & Pearlman, R.E. (1974) / . Biol. Chem. 249, 3123-3131
14. Lee, Y.C. & Byfield, J.E. (1970) Biochemistry 9, 3947-3959 J. Biochem.
POLY(A) POLYMERASE OF Tetrahymena 15. Higashinakagawa, T., Tashiro, F., & Mita, T. (1975) J. Biochem. TJ, 783-793 16. Rodriguez-Pousada, C. & Hayes, D.H. (1976) Eur. J. Biochem. 71, 117-124 17. Ron, A., Horovitz, O., & Sarov, I. (1976) / . Mol. Evol. 8, 137-142 18. Fraenkel-Conrat, H. & Steinschneider, A. (1968) in Methods in Enzymology (Grossman, L. & Moldave, K., eds.) Vol. 12B, pp. 243-246, Academic Press, New York 19. Muramatsu, M. (1973) in Methods in Cell Biology (Prescott, D.M., ed.) Vol. 7, pp. 23-51, Academic Press, New York 20. Layne, E. (1957) in Methods in Enzymology (Colowick, S.P. & Kaplan, N.O., eds.) Vol. 3, pp. 447454, Academic Press, New York 21. Burton, K. (1968) in Methods in Enzymology (Grossman, L. & Moldave, K., eds.) Vol. 12A, pp. 163-166, Academic Press, New York 22. Schneider, W.C. (1957) in Methods in Enzymology (Colowick, S.P. & Kaplan, N.O., eds.) Vol. 3, pp. 680-684, Academic Press, New York 23. Twu, J.S. & Bretthauer, R.K. (1971) Biochemistry 10,1576-1582 24. Wray, W., Conn, P.M., & Wray, V.P. (1977) in Methods in Cell Biology (Stein, G., Stein, J., & Kleinsmith, L.J., eds.) Vol. 16, pp. 69-86, Academic Press, New York
Vol. 86, No. 5, 1979
1311 25. Roeder, R.G. (1976) in RNA Polymerase (Losick, R. & Chamberlin, M., eds.) pp. 285-329, Cold Spring Harbor Laboratory 26. Grunberg-Manago, M. (1963) in Progress in Nucleic Acid Research (Davidson, J.N. & Cohn, W.E., eds.) pp. 93-133, Academic Press, New York 27. See, Y.P. & Fitt, P.S. (1970) Biochem. J. 119, 517-524 28. Deutscher, M.P. (1973) / . Biol. Chem. 248, 31163121 29. Mukerji, S.K. & Deutscher, M.P. (1972) / . Biol. Chem. 247, 481^88 30. Fitt, P.S. (1966) / . Protozool. 13, 507-509 31. Kobayashi, S. (1965) / . Biochem. 58, 444-457 32. Flavell, R.A. & Jones, I.G. (1970) Biochem. J. 116, 811-817 33. Brawerman, G. (1976) in Progress in Nucleic Acid Research and Molecular Biology (Cohn, W.E., ed.) Vol. 17, pp. 117-148, Academic Press, New York 34. Rose, K.M., Roe, F.J., & Jacob, S.T. (1977) Biochim. Biophys. Ada 478, 180-191 35. Sawicki, S.G., Jelinek, W., & Darnell, J.E. (1977) / . Mol. Biol. 113, 219-235
Characterization of Polyriboadenylate Polymerase from Tetrahymena pyriformis Hisao UEYAMA 1 Department of Biochemistry, Kyoto Prefectural University of Medicine, Kamigyo-ku, Kyoto, Kyoto 602 Received for publication, May 17, 1979
Poly(A) polymerase [polyadenylate nucleotidyltransferase, EC 2.7.7.19] was extracted from Tetrahymena pyriformis. The enzyme was demonstrated to be present in three forms by column chromatography on DEAE-cellulose, and they were termed poly(A) polymerase la, Ib, and II in order of increasing affinity to the column. The properties of enzymes la and Ib were similar except that la utilizes poly(A) as a primer rather efficiently. Enzyme II differed from enzymes la and Ib not only in elution profile on DEAE-cellulose column chromatography but also in pH and temperature preferences, molecular weight, requirement for divalent cations, sensitivity to salts at high ionic strength, optimal primer concentration, and subcellular localization. The molecular weights of enzymes la and Ib measured by gel filtration were both 43,000, and that of enzyme II was 95,000. All three enzymes required Mn ! + for maximal activity; Mg t + could replace Mn 1+ in the reaction of enzyme II, but only partially. In the presence of 0.1 M ammonium sulfate the activities of enzymes la and Ib were both completely inhibited, whereas enzyme II still showed 42% of its original activity. These findings suggest that there are two distinct types of poly(A) polymerase in Tetrahymena pyriformis.
PoIy(A) polymerase is thought to be an enzyme that post-transcriptionally synthesizes poly(A) chains at the 3'-termini of heterogeneous nuclear RNA (hnRNA) and mRNA of eukaryotic cells (13). In the cases of N.I.H.-Swiss mouse embryos (4) and mouse lymphoma cells (5), only one form of poly(A) polymerase was extracted, which probably carries out the above function in the cells. On the other hand, the existence of more than one poly(A) polymeiase in a single cell type has been reported (2, 3, 6-10), but the significance of 1
Present address: Department of Medical Biochemistry, Shiga University of Medical Science, Seta, Ohtsu, Shiga 520-21. Vol. 86, No. 5, 1979
1301
such multiple forms of poly(A) polymerase is still unknown (11). Poly(A) polymerase has not previously been isolated from Tetrahymena pyriformis, which has been used for several studies on nucleic acid metabolism, including work on DNA polymerases (12, 13) and RNA polymerases (14, 15), although poly(A) chains of 80-150 nucleotides (16) or 4S (17) were found to be attached to its hnRNA and mRNA. In order to determine how poly(A) polymerase is involved in the synthesis and turnover of mRNA in Tetrahymena pyriformis, it is necessary to extract and characterize this enzyme. The present paper reports the possible occurrence of two distinct types of poly(A) polymerase in this organism.
1302
H. UEYAMA
MATERIALS AND METHODS Materials—a-Amanitin, actinomycin D, rifampicin, and cordycepin (3'-deoxyadenosine) 5'triphosphate were obtained from Sigma. Poly(A), poly(C), and poly(U) were from P-L Biochemicals, poly(G) from Sigma, poly(l) from Miles Laboratories and tRNA from Boehringer Mannheim. tRNA having a 3'-phosphate end was prepared with sodium periodate and aniline (75). Tetrahymena RNA and rat liver rRNA were extracted from ribosomal pellets by phenol-sodium dodecyl sarcosinate (SDSa) treatment {19). It was observed by sucrose density gradient centrifugation that rat liver RNA has two main peaks of 18S and 28S, whereas Tetrahymena RNA has a broad distribution with one peak smaller than 18S rRNA. Phosphoenolpyruvate and pyruvate kinase were obtained from Boehringer Mannheim. [2-'H]ATP, [U- 14 C]GTP, [U- 14 C]UTP, [U- 14 C]CTP, [8- s H]poly(A), [8-"C]ADP, and [a-"P]ATP were from the Radiochemical Centre. Hexylene glycol was purchased from Eastman. Preparation of the Crude Extract—Tetrahymena pyriformis (the amicronucleate strain) was cultured aerobically with 2 % peptone (Mikuni, Japan) and 0.5% yeast extract (Difco). After incubation at 28°C for 36 h (cell density 1.5-3 x 107ml), cells were collected by centrifugation at 1,300 xg for 10 min, and washed with buffer A (50 rriM Tris-HCl, pH 7.6, containing 2 IHM 2mercaptoethanol). The pellet was resuspended in buffer A and stored below —20°C. All purification procedures after this step were carried out at 0-4°C. The thawed cells were sonicated for 1 min with an Insonator (Kubota, Japan) at 1.8 A (140 W). The sonicated cell suspensions were centrifuged at 35,000 x g for 20 min. The resulting supernatant was centrifuged at 122,000xg for 120 min. The supernatant was used as the enzyme preparation and the pellet was used for extraction of RNA. Assay of Poly(A) Polymerase—The standard assay mixture contained 50 mM Tris-HCl buffer (pH 8.5), 5 mM 2-mercaptoethanol, 1 mM MnCl,, 0.5 mg of Tetrahymena RNA, 10-100 fi\ of enzyme, and 0.5 mM [ 8 H]ATP (2 ftCi/fimol), in a volume of 250 [t\. After incubation for 30 min at 28°C, 1 ml of ice-cold 5 % trichloroacetic acid (TCA)-5 mM
sodium pyrophosphate (NaPPi) was added to the reaction mixture and the acid-insoluble precipitate was collected on a glass membrane filter (GF/C, 2.4 cm, Whatman). After washing twice with 10 ml of 5% TCA-5mM NaPPi, the filters were put into glass vials to which 10 ml of PCS (Amersham/Searle) was added. The radioactivity was measured in a Packard scintillation spectrometer (model 3255). Protein was determined by the biuret method, Lowry's method or WarburgChristian's method (20) depending on its concentration and the presence of ammonium sulfate in the enzyme preparation. One unit of the enzyme was defined as the amount of enzyme which catalyzes the incorporation of 1 nmol of AMP into acid-insoluble precipitate in 30 min. Gel Filtration—A column (1.5x90 cm) of Ultrogel (ACA44, LKB) previously equilibrated with 50 mM Tris-HCl buffer (pH 7.6) was loaded with 1.0 ml of sample and developed with the same buffer, collecting 3 ml fractions. The molecular weight of each poly(A) polymerase was estimated using the following markers: myoglobin (Mr = 17,000), soybean trypsin inhibitor (Mr=21,5OO), ovalbumin (Mr=45,0OO), and bovine serum albumin (Mr=68,000). When gel filtration was used at the last step in the purification of poly(A) polymerases, 2-4 ml of each enzyme sample was loaded on the column and 3 ml fractions were collected. Analysis of the Reaction Product—The reaction conditions were as described above except that the specific activity of [ J H]ATP used was 20 ftC\/ftmo\. The glass membrane filter ( G F / Q used to collect acid-insoluble precipitate was soaked in 0.5 ml of 0.3 N KOH and incubated for 18 h at 37°C. The hydrolysate was neutralized with 1 N perchloric acid and centrifuged. Twenty fi\ of the supernatant was developed by ascending paper chromatography with the following solvent: w-butanol/acetone/acetic acid/5 % NH 4 OH/H,O (35 : 25 : 15 : 15 : 10, by vol.). After development for 10 h the radioactivities of spots detected by ultraviolet absorption were determined. Isolation of the Nuclei—The cells collected by centrifugation (1,300 x g, 5 min) were washed once with buffer D (0.05 mM PIPES-Na, pH 7.5, containing 0.5 M hexylene glycol and 1 mM CaCl,) and then resuspended in the same buffer. The cell homogenate obtained with a Potter-Elvehjem /. Biochem.
POLY(A) POLYMERASE OF Tetrahymena
1303
homogenizer was centrifuged at 1,500 xg for 5 min. The pellet was washed once with buffer D and then suspended in buffer D containing 1.6 M sucrose. The suspension was layered on an equal volume of buffer D containing 2 . 1 M sucrose in SW27 rotor tubes and centrifuged at 55,000Xg for 60 min. The nuclear pellet obtained was washed once with buffer D and used to investigate the localization of the enzymes. DNA was determined by the method of Burton (21) and RNA by means of the orcinol reaction (22). Assay of Poly (A)-Degrading Enzyme—The assay mixture was as follows: 50 ITIM Tris-HCl (pH 8.5), 1 rain MnCl,, 5 mM 2-mercaptoethanol, 0.5 mM [»H]poly(A) (0.1 ftCi/ftmol nucleotides), enzyme, and 0.5 mM ATP, in a volume of 250 //I. After incubation of the reaction mixture at 28°C for 30 min, 100 //I of 5 mg/ml RNA, followed rapidly by 650 ft\ of 7.7 % TCA, was added and the mixture was centrifuged. The radioactivity of 0.5 ml aliquots of the acid-soluble supernatant was measured in the liquid scintillation spectrometer.
RESULTS Purification of Poly (A) Polymerase—The crude extract (740 ml) from 450 g of Tetrahymena cells was subjected to DEAE-cellulose column chromatography (4 X 30 cm). A typical elution profile is shown in Fig. 1. The enzyme activity was found in two fractions, i.e., the washing with the starting buffer (buffer A) and the fraction eluted with ammonium sulfate. The enzyme which was not adsorbed on the column was termed poly(A) polymerase I and the enzyme eluted from the column with 0.25 M ammonium sulfate was termed poly(A) polymerase II. These enzymes were further purified as follows. The pH of the enzyme I preparation (1,030 ml) was adjusted to 8.5 from 7.6 with 50 mM Tris2 mM 2-mercaptoethanol. The solution (2,060 ml) was concentrated to 530 ml with a hollow fiber apparatus (DC 2, Amicon) and applied to a column of DEAE-cellulose (1.5x30 cm) previously equilibrated with buffer B (50 mM Tris-HCl, pH 8.5,
'o x E
« o
w O
a. o o _c 0. 5
10
20
30 Fraction
40
SO
Fig. 1. 1st DEAE-cellulose column chromatography of po'y(A) polymerases from Tetrahymena pyriformis. The crude extract obtained from 20 g of Tetrahymena cells as described in " MATERIALS AND METHODS" was applied to a column of DEAE-cellulose (1.0x30cm) previously equilibrated with buffer A (50 mM Tris-HCl, pH 7.6, containing 2 mM 2-mercaptoethanol). The column was washed with buffer A, then eluted stepwise with ammonium sulfate, 0.05 M and 0.25 M. PoIy(A) polymerase activity was assayed under standard assay conditions as described in "MATERIALS AND METHODS." • , Protein concentration; O, po'y(A) polymerase activity; , ammonium sulfate concentration. Vol. 86, No. 5, 1979
1304
H. UEYAMA same buffer, the column was eluted with a linear gradient from 0 to 0.3 M NaCl. The fractions containing enzyme activity (fractions 27 through 34) were combined, then concentrated to 6.2 ml with a Mini-Module hollow fiber apparatus (Asahikasei, Japan), and applied to a column of Ultrogel. The eluted fractions showing enzyme activity were rechromatographed on the Ultrogel column after concentration to 2.2 ml with the Mini-Module. Poly(A) polymerase la was collected in fractions 28 and 29 (Fig. 2), and the molecular weight of enzyme la was estimated to be 43,000. The yield was 25 % and the degree of purification was 104-fold (Table I). The final enzyme preparation had no poly(A)-degrading activity. Enzyme Ib (170 ml) was concentrated with the Mini-Module to 26 ml and dialyzed against 5 liters of 10 mM Na acetate buffer (pH5.5) containing 2mM 2mercaptoethanol. The dialyzed solution was
o
5
05
AMP Inco rporated
(dpmx 10 )
containing 2 mM 2-mercaptoethanol). Most of the enzyme activity (la), which utilized poly(A) as a primer much more effectively than Tetrahymena RNA, was eluted by washing with buffer B and a small portion of the original activity (Ib), which utilized the RNA in preference to poly(A) as a primer, was adsorbed on the column and then eluted with buffer A. Enzyme la fraction (690 ml) was brought to 0.45 saturation with solid ammonium sulfate and the mixture was centrifuged at 15,000 xg for 15 min. The supernatant (750 ml) was then brought to 0.55 saturation with ammonium sulfate and centrifuged. The pellet was dissolved in 25 ml of buffer C (10 mM MES-Na, pH 6.0, containing 2 mM 2-mercaptoethanol) and dialyzed against 5 liters of the buffer. The dialyzed enzyme solution (28 ml) was applied to a column of CM-cellulose (1.5x30 cm) previously equilibrated with buffer C and, after washing with the
I
5 LOG (Mr)
20
30
40
50
Fraction Fig. 2. Gel nitration chromatography on Ultrogel of poly(A) polymerases from Tetrahymena pyriformis. Each poly(A) polymerase solution was applied to a column of Ultrogel (1.5x90 cm) equilibrated with 50 mM Tris-HCl buffer (pH 7.6), collecting 3 ml fractions. Three independent elution profiles are superimposed in the figure to compare the elution positions of the poly(A) polymerases. The void volume (KJ is indicated by an arrow. The poly(A) polymerase activity was assayed under the standard assay conditions as described in " MATERIALS AND METHODS." D, Poly(A) polymerase la; • , poly(A) polymerase Ib; • , poly(A) polymerase II. The inset shows the determination of the molecular weight of each poly(A) polymerase. Myoglobin ( • ) ; soybean trypsin inhibitor ( • ) ; ovalbumin (O); bovine serum albumin ( • ) ; poly(A) polymerase la and Ib ( A ) ; poly(A) polymerase II (A). J. Biochem.
POLY(A) POLYMERASE OF Tetrahymena
1305
applied to a column of CM-cellulose (1.5 x 30 cm) 30 cm) previously equilibrated with buffer A. The equilibrated with the same buffer. The column column was washed with buffer A containing was washed with the starting buffer and then 0.05 M ammonium sulfate, then eluted with a linear eluted with a linear gradient of NaCl (0-0.3 M). , gradient from 0.05 to 0.15 M ammonium sulfate. The fractions containing enzyme activity were The fractions containing enzyme activity (fractions combined, concentrated to 3.9 ml with the Mini23 through 45) were combined, concentrated to Module and applied to a column of Ultrogel. 3.8 ml with the Mini-Module and applied to a The column chromatography on CM-cellulose was column of Ultrogel. Poly(A) polymerase II was effective in removing poly(A)-degrading activity, eluted in fractions nearer to the void volume than but the increase in the specific activity of enzyme polymerases la and Ib (Fig. 2). The molecular Ib was small (Table 1). The molecular weight of weight of this enzyme was estimated to be 95,000. enzyme Ib was estimated to be 43,000 by gel The recovery was 27 % and the degree of purificafiltration. tion was 315-fold. The final enzyme preparation was found to be free from poly(A)-degrading Enzyme II (400 ml) was first fractionated by activity. The above purification procedures are ammonium sulfate precipitation. The active ensummarized in Table I. zyme protein precipitated with ammonium sulfate between 0.33 and 0.45 saturation was dissolved in Properties of Poly (A) Polymerases—All three 10 ml of buffer A and dialyzed against 2 liters of polymerases required an RNA primer and showed the same buffer. The dialyzed enzyme solution no requirement for a DNA template (Table II). (12.4 ml) was applied to a column of Sepharose Among primers tested, RNA obtained from the 6B (1.8x40 cm) equilibrated with buffer A. The ribosomal pellet of Tetrahymena pyriformis was active fractions (17-21) were combined and then most effective for poly(A) polymerase Ib, and applied to a column of DEAE-cellulose (1.5 x Tetrahymena RNA, tRNA, or rRNA of rat liver TABLE I. Purification of poly (A) polymerases from Tetrahymena pyriformis. Enzyme fraction Crude extract6 1st DEAE-cellulose
II
Specific activity1 (units/mg)
Total enzyme1 (units)
12,300
1.10
13,500
5,880
0.818
4,810
Yield (%)
Hnd DEAE-cellulose 45-55% (NH4),SO, CM-cellulose 1st Ultrogel Hnd Ultrogel
2,930 633 106 30.7 6.62
1.03 2.75 12.9 39.4 114
3,020 1,740 1,370 1,210 756
100 58 45 40 25
Ilnd DEAE-cellulose Ib CM-cellulose Ultrogel
566 50.8 3.24
1.86 2.13 22.0
1,050 108 71.4
100 10 7
1st DEAE-cellulose 33-45% (NHJ.SO, Sepharose 6B Ilnd DEAE-cellulose Ultrogel
859 250
7,800 6,220 4,900 4,540 2,140
100 80 63 58 27
la I
Total protein (mg)
101 40.7 6.18
9.09 24.9 48.5 112 346
»~The primer used was RNA obtained from the ribosomal pellet of Tetrahymtna. 450 g of Tetrahymena cells. Vol. 86, No. 5, 1979
b
The starting material was
H. UEYAMA
1306
TABLE IT. Primer requirements of poly (A) polymcrases from Tetrahymena pyriformis. The reaction mixture contained 50 mM Tris-HCl buffer (pH 8.5), 5 mM 2-mercaptoethanol, 1 mM MnCl t , 0.5 unit of enzyme, 0.5 DIM [*H]ATP and various primers, in a volume of 250 pi\. Just before termination of the reactions after incubation at 28°C for 30 min, 0.05 ml of 5 mg/m] RNA was added to the reaction mixtures as a carrier. The concentration used was 1 mg/ml for all primers except Tetrahymena RNA and rat 4iver rRNA (2 mg/m] for the former and 1.5 mg/ml for the latter). The denatured DNA was prepared by heating the native DNA at 99°C for 10 min, then cooling it rapidly. The activated DNA was prepared by DNase treatment. Poly(A) polymerase Primers Tetrahymena RNA —primer Poly(A) Poly(Q • PoIy(U) PoIy(G) Poly© Rat liver rRNA tRNA (3'-CCA-0H) tRNA (3'-CC-P) tRNA (3'-CC-OH) Calf thymus native DNA Calf thymus denatured DNA Calf thymus activated DNA
la
Ib
n
100
100
100
0
0
0
324
1
10
1 14 8 10 80 95 2 48 12 10
0 0 4 0 21 52 18 66 1 3
4 0 11 3 82 152 1 351 5 2
7
4
5
was effective for enzyme II. Only poly(A) polymerase l a could utilize poly(A) effectively as a primer. A free 3'-terminal hydroxy group must be essential for the enzyme reactions, because t R N A having a 3'-phosphate end (tRNA-CCP) was far less active than normal tRNA, while tRNA having a 3'-CC-OH end, prepared from the tRNAC C P by phosphatase treatment, was as active as normal t R N A (Table II). The effects of R N A concentration on the poly(A) polymerase reaction were different among polymerases la, Ib, and II. Polymerase Ib required about the same concentration of R N A as in the standard assay conditions (2 mg/m]) for maximal activity, while polymerases l a and II showed
maximal activities at one-half (1 mg/ml) and onequarter (0.5 mg/ml) of this R N A concentration, respectively. Poly(A) polymerases la and Ib differed from polymerase II in p H preference. Enzymes la and Ib had pH optima at 8.5 but that of II was at 9.0. With regard to the optimal temperature of the reaction, la and Ib were most active at 28°C, which is also optimal for the growth of Tetrahymena, while II was most active at 33°C. Divalent cations were absolutely required for poly(A) polymerase activity, and Mn ! + was more effective than Mg I + . The optimal concentrations of M n ' + and Mg I + for enzyme II were 1-2 mM and 2-4 mM, respectively, but enzymes la and Ib could be activated only by Mn 2 + , as shown in Fig. 3. Mg I + , evenat higher concentrations than indicated in Fig. 3 (up to 20 mM), could not replace Mn 1 + in the reactions of enzymes la and Ib. Polymerase II differed from polymerases I in the effect of salts at high ionic strength. As shown in Fig. 4, in the presence of 0.1 M ammonium sulfate the activities of poly(A) polymerases la and Ib were completely depressed but enzyme II still showed 42 % of its original activity. Similar results were obtained with NaCl and KC1. All three poly(A) polymerases incorporated ATP most efficiently into acid-insoluble precipitate. The incorporation of other labeled ribonucleoside triphosphates was less than 17% of that of A T P ; this specificity for ATP is thought to be an important characteristic of a poIy(A) polymerase {11). None of the three polymerases was affected by DNA-dependent R N A polymerase inhibitors such as ar-amanitin, actinomycin D, and rifampicin. All three poly(A) polymerases were inhibited by other ribonucleoside triphosphates at a concentration equimolar with substrate ATP (0.5 mM). Reversal of the inhibiting effect of the ribonucleotides could not be observed on further addition of Mn 1 + , so these ribonucleotides seem to inhibit poly(A) polymerase activities, but not by chelating M n l + . The addition of 0.5 mM A D P did not affect polymerase n , suggesting that the substrate is not A D P but ATP. However, the activities of poly(A) polymerase l a and Ib were inhibited about 50% by 0.5 mM ADP. When [ " Q A D P at 0.5 mM was used as a substrate in the reaction of enzyme la or Ib, incorporation of the radioactivity into acid-insoluble precipitate was found to be less than
J. Biochem.
1307
POLY(A) POLYMERASE OF Tetrahymena
100
-O
O
1
2
3
4
Divalent Cations (mM)
Fig. 3. Divalent cation requirements of poly(A) polymerases from Tetrahymena pyriformis. The reaction mixture contained 50 mM Tris-HCI buffer (pH 8.5), 5 mM 2-mercaptoethanol, 0.5 mg of RNA obtained from the ribosomal pellet of Tetrahymena, 100 p\ of enzyme, 0.5 mM PH]ATP, and divalent cations at various concentrations, in a volume of 250 ftl. Maximal activity of each enzyme was obtained with 1 mM MnClt, amounting to 1,408 disintegrations per min (dpm), 6,021 dpm and 5,845 dpm with enzymes la, Ib, and II, respectively. Poly(A) polymerase la, —D— with MnCl,, - - • - - with MgCl,; poly(A) polymerase Ib, — • — with MnCl., - - • - - with MgCl,; poly(A) polymerase II, — • — with MnCl,, - - • -- with MgCl,.
10
20 30 Ammonium Sulfate (mM)
100
Fig. 4. Effect of ammonium sulfate on poly(A) polymerases from Tetrahymena pyriformis. The control assay mixture contained 50 mM Tris-HCI buffer (pH 8.5), 5 mM 2-mercaptoethanol, 1 mM MnCl,, 0.5 mg of RNA obtained from the ribosomal pellet of Tetrahymena, 100 fi\ of enzyme, and 0.5 mM PHJATP, in a volume of 250 /il. AH enzymes used were ammonium sulfate-free after gel filtration on Ultrogel. The control enzyme activities were 1,550 disintegrations per min (dpm), 3,577 dpm, and 2,666 dpm with enzymes la, Ib, and II, respectively. D, Poly(A) polymerase la; • , poly(A) polymerase Ib; • , poly(A) polymerase II.
11 % of that of ATP, and when 2 mM pohsphoenolpyruvate and 10 units of pyruvate kinase were added as an ATP-generating system, the incorporation was greatly enhanced. These results suggest Vol. 86, No. 5, 1979
that the true substrate for the polymerizing reaction of enzyme la or Ib is not ADP but ATP. Cordycepin 5'-triphosphate, which is probably the active form of cordycepin that inhibits poly(A)
1308
H. UEYAMA
synthesis in vivo, inhibited all three enzyme activities to similar extents (40-60%) at a concentration equimolar with substrate ATP (0.5 ITTM). Products of Poly(A) Polymerase Reactions— The reaction products were hydrolyzed and the radioactivities of 2'(3')-AMP and adenosine were compared. As shown in Table in, the reaction product of each poly(A) polymerase under the conditions of the standard assay was oligo(A) consisting of less than 10 AMP residues. Natural poly(A) found on nuclear and cytoplasmic mRNAs in Tetrahymena pyriformis was reported to be 80-150 nucleotides in length (75), and there is a major difference in length between the natural poly(A) and the reaction products. However, the chain length of poly(A) polymerizing reaction products in vitro has been reported to be dependent on primer concentration (3, 9, 23) and time of
TABLE m . Average chain length of the reaction products of each poly (A) polymerase from Tetrahymena pyriformis. The reaction mixture was as described in "MATERIALS AND METHODS." The primer, other than poly(A), was RNA obtained from the ribosomal pellet of Tetrahymena pyriformis. After incubation of the reaction mixture at 28°C for 30 min, the reaction was terminated with 1 ml of ice-cold 5% TCA containing 5 mM NaPPi. The acid-insoluble precipitate collected on a glass membrane filter (GF/C) was hydrolyzed with 0.3 N KOH at 37°C for 18 h. Twenty ft\ of the supernatant obtained by centrifugation of the hydrolysate neutralized with 1 N HC1O4 was developed by ascending paper chromatography. The recovery of radioactivity was 85-90% in all cases. Poly(A) polymerase
Primer concentration (mg/ml)
Average chain length
0.5
2.0
1.0 2.0 poly(A) 1.0
1.7 1.4 2.9
Ib
0.5 1.0 2.0
4.6 3.9 3.2
U
0.25 0.5 1.0
30.0 24.2 16.2
2.0
8.8
la
incubation (3, 23). Decreasing the concentration of primer RNA resulted in a progressive elongation of the average chain length of the product of enzyme II, but had little effect on that of enzyme Ib (Table HI). The product of enzyme la was also oligo(A) even with a lower concentration of RNA or poly(A) as a primer. The covalent linkage of the reaction product to RNA primer was examined by sucrose density gradient centrifugation and nearest-neighbor analysis. In the former experiment, 28S rRNA purified from rat liver was used as a primer. After incubation of the reaction mixture containing 28S rRNA for 30 min, the reaction was terminated with ice-cold 90% ethanol containing 2% potassium acetate and 0.01% SDSa. The RNA extracted from the precipitate by phenol-SDSa treatment after centrifugation at 3,000 xg for 15 min was dissolved in 0.05 M Na acetate buffer (pH 5.1) and layered on a 10-30% linear sucrose gradient {19) in a Spinco SW 27 rotor tube. Sucrose density gradient centrifugation was carried out at 65,000 x g for 18 h at 8°C. The centrifuged sample was fractionated with an Auto Densi-Flow (Searle) into 31 fractions. Radioactivity was found in 28S rRNA. To confirm that the reaction product is covalently linked to 3'-termini of the primer, tRNA having a 3'-CC-OH end and [a-38P]ATP were used in the latter experiment. After incubation of the reaction mixture at 28°C for 30 min, the reaction was terminated with 5% TCA-5 mM NaPPi. The acid-insoluble precipitate collected on GF/C was soaked in 0.5 ml of 1 N KOH and incubated for 16 h at 37°C. The neutralized' hydrolysate with 1 N perchloric acid was centrifuged and 20 fi\ of the supernatant was subjected to high-pressure liquid chromatography (/iBondapak/NH,, Waters Associates Inc.). After development with 0.01 M NH4H,PO« (pH 3.0), the radioactivity of each 2'(3')-mononucleotide was measured in a liquid scintillation spectrometer. It was found that most (82-96%) of the radioactivity (total 2,660-52,100 cpm) was recovered in 2'QO-CMP and 2'(3>AMP, although the ratios of radioactivities of these two nucleotides (AMP to CMP) were different among the three enzymes: 12.4, 0.14, and 0.21 for enzymes la, Ib, and n , respectively. The above results suggest that the reaction products are covalently linked to 3'termini of the RNA primer. J. Eiochem.
POLY(A) POLYMERASE OF Tetrahymena
Nuclear Localization of Poly (A) Polymerase II—It was reported that nuclear poly(A) polymerase in calf thymus (2, 3, 6) and HeLa cells (10) can be activated by either MnI+ or Mgl+, and that • cytoplasmic poly(A) polymerase can be activated only by MnI+. In order to study whether enzymes I and II also differ in subcellular localization, nuclei of Tetrahymena were isolated. The hexylene glycol method used was preferred to both aqueous and nonaqueous techniques since the procedure is simple and leakage of soluble nuclear components is prevented (24). The nuclei were purified over 14.7-fold from the whole cell, based on the ratio of DNA to protein and of RNA to DNA. The nuclear pellet obtained from 26 g of Tetrahymena cells was suspended in 9 ml of buffer B, sonicated at 1.8 A for 1 min, and then centrifuged at 105,000 xgr for 1 h. The poly(A) polymerase activity of the resulting supernatant was resistant to ammonium sulfate at 0.1 M, suggesting that it may be due to poly(A) polymerase II. The supernatant was applied to a column of DEAE-cellulose (0.3x30 cm) previously equilibrated with buffer B and then the column was eluted with a linear gradient from 0 to 0.3 M ammonium sulfate. Poly(A) polymerase activity was found only in fractions 16 through 20, showing that only enzyme II was present in the nuclei. DISCUSSION Poly(A) polymerase of Tetrahymena pyriformis was separated in three forms by column chromatography on DEAE-cellulose. The three enzymes were further purified and classified into two types (la, Ib, II) on the basis of their properties. These enzymes required RNA as a primer, not a DNA template. They can be easily distinguished from RNA polymerase type II because of their insensitivity to a-amanitin. It is also clear that they differ from RNA polymerase types I and III in sensitivity to salts at high ionic strength, requirement for divalent cations and molecular weight. The high sensitivity of poly(A) polymerases I to ammonium sulfate is not compatible with the report that cr-amanitin-insensitive RNA polymerases of Tetrahymena are not affected by the salt up to 50 mM (75). It was reported that RNA polymerases of Tetrahymena can be activated by M n " and Mg1+ to similar extents (15), while Vol. 86, No. 5, 1979
1309
these three poly(A) polymerases are activated by Mn1+ more effectively than by MgI+. The molecular weights of eukaryotic RNA polymerases (25) are far larger than those of the poly(A) polymerases. There is a possibility that polynucleotide phosphorylase and tRNA nucleotidyltransferase are involved in the poly(A) polymerase-detecting system when the incorporation of a labeled nucleotide into acid-insoluble precipitate is measured. The former enzyme is known not to be present in eukaryotic cells (11), but some reports on eukaryotic polynucleotide phosphorylase have appeared (26, 27). However, the properties of the three enzymes are not consistent with those of polynucleotide phosphorylase, which requires not ATP but ribonucleoside diphosphates and can be activated only by MgI+ (26). It was reported that tRNA nucleotidyltransferase utilizes tRNA as the best primer but can also use rRNA to some extent (28), and it is possible that some of the poly(A) polymerases are identical with this enzyme. However, poly(A) polymerase II seems to be different from tRNA nucleotidyltransferase, because the reaction products of poly(A) polymerase II were poly(A) of 9-30 nucleotides in length, which is not consistent with the observation that only a single AMP residue is added to the primer by tRNA nucleotidyltransferase (28). Besides, poly(A) polymerase II is present in nuclei, whereas tRNA nucleotidyltransferase has been reported to exist not in nuclei but in the cytosol and mitochondria (29). As the reaction products of poly(A) polymerase la or Ib were oligo(A) with more than 2 residues, these enzymes are not identical with tRNA nucleotidyltransferase, either. It has been reported (30) that tRNA nucleotidyltransferase of Tetrahymena pyriformis incorporates CTP to the same extent as ATP, that Mg1+ is twice as effective as Mn1+ on this enzyme activity and that tRNA is utilized as a primer much more efficiently than rRNA (14 times) by this enzyme. On the contrary, poly(A) polymerases la and Ib incorporate not CTP but only ATP (the incorporation of CTP by enzymes la and Ib was 4% and 0% of that of ATP, respectively), Mg1+ was not effective on the poly(A) polymerases la and Ib reactions, and tRNA was utilized only 1.2-2.5 times more efficiently than rRNA as a primer by enzyme la or Ib. Thus, it seems likely that all three enzymes
1310
arc poIy(A) polymerase or polyadenylate nucleotidyltransferase [EC 2.7.7.19]. The next problem is whether poIy(A) polymerases la and Ib are identical or not. The properties of polymerase II were quite dissimilar to those of polymerases la and Ib, and it is clear that enzyme II is different from enzymes la and Ib. If the enzyme Ib preparation contains poly(A)degrading activity and conversely the enzyme la preparation does not, the difference that enzyme la utilizes poly(A) as a primer much better than enzyme Ib might be accounted for, but the final enzyme preparations of la and Ib, both free of poly(A)-degrading activity, still showed a difference in the utilization of poly(A) as a primer. The properties of the two enzymes were similar, except for the primer specificity, and therefore they are classified as type I poly(A) polymerase in this report, though further studies are clearly necessary. The significance of the occurrence of two types of poly(A) polymerase in Tetrahymena pyriformis cannot be easily deduced from the properties of the enzymes obtained. It is possible that enzyme la or Ib is mitochondrial in origin, but no poly(A) polymerase activity was detected in the extracts of mitochondria isolated by two different methods (31, 32). Therefore, poly(A) polymerase la and Ib probably exist in the extramitochondrial cytoplasm. In yeast, it is considered that one of the poly(A) polymerases found in the nuclei may add a short oligo(A) sequence to mRNA for initiation of the poly(A) tracts of the mRNA and that the other enzyme, preferring poly(A) itself, could elongate the short oligo(A) tract (7). However, the properties of the poly(A) polymerases purified from Tetrahymena pyriformis are not consistent with such a sequential mechanism. Since poly(A) was inactive as a primer for enzyme Ib, this enzyme may be involved in the initiation process of poly(A) synthesis. However, cordycepin 5'-triphosphate, which has been reported to inhibit the initiation of poIy(A) synthesis at far lower concentrations than those of the substrate ATP (33, 34), affected the activity of enzyme Ib by only 56% at a concentration equimolar with ATP. The absence of this enzyme in the nuclei also seems inconsistent with the above hypothesis. Recently, it was confirmed that poly(A) synthesis in HeLa cells and mouse sarcoma ascites
H. UEYAMA cells involves two steps: de novo synthesis and terminal addition (33, 35). Poly(A) polymerase la may be involved in the terminal turnover of poly(A) tracts in the cytoplasm, because poIy(A) is an effective primer for this enzyme. On the other hand, poly(A) polymerase II probably functions in the de novo synthesis and terminal addition of poly(A) in the nuclei. This hypothesis is compatible with the report that the nucleus is the site of de novo synthesis of poly(A), and that 3'-terminal addition of poly(A) occurs both in the nuclei and the cytoplasm (35). The author wishes to express his gratitude to Dr. T. Matsuura and Prof. K. Ueda, Shiga University of Medical Science, for their valuable suggestions and discussions during the course of this work. The author also wishes to thank Prof. A. Iwashima, Kyoto Prefectural University of Medicine, for his helpful advice in the preparation of this manuscript. REFERENCES 1. Darnell, J.E., Jelinek, W.R., & Molloy, G.R. (1973) Science 181, 1215-1221 2. Winters, M.A. & Edmonds, M. (1973) /. Biol. Chem. 248, 4756-4762 3. Winters, M.A. & Edmonds, M. (1973) J. Biol. Chem. 248, 4763^1768 4. Hadidi, A. & Sethi, V.S. (1976) Biochim. Biophys. Ada 425, 95-109 5. Mtiller, W.E.G., Schroder, H.C., Arendes, J., Steffen, R., Zahn, R.K., & Dose, K. (1977) Eur. J. Biochcm, 76, 531-540 6. Tsiapalis, CM., Dorson, J.W., &Bollum, F.J. (1975) /. Biol. Chem. 250, 4486-4496 7. Haff, L.A. & Keller, E.B. (1975) J. Biol. Chem. 250, 1838-1846 8. Pellicer, A., Salas, J., & Salas, M.L. (1978) Biochim. Biophys. Ada 519, 149-162 9. Niessing, J. (1975) Eur. J. Biochem. 59, 127-135 10. Nevins, J.R. & Joklik, W.K. (1977) /. Biol. Chem. 252, 6939-6947 11. Edmonds, M. & Winters, M.A. (1976) in Progress in Nucleic Acid Research and Molecular Biology (Cohn, W.E., ed.) Vol. 17, pp. 149-179, Academic Press, New York 12. Westergaard, O. & Lindberg, B. (1972) Eur. J. Biochem. 28, 422-431 13. Crerar, M. & Pearlman, R.E. (1974) / . Biol. Chem. 249, 3123-3131
14. Lee, Y.C. & Byfield, J.E. (1970) Biochemistry 9, 3947-3959 J. Biochem.
POLY(A) POLYMERASE OF Tetrahymena 15. Higashinakagawa, T., Tashiro, F., & Mita, T. (1975) J. Biochem. TJ, 783-793 16. Rodriguez-Pousada, C. & Hayes, D.H. (1976) Eur. J. Biochem. 71, 117-124 17. Ron, A., Horovitz, O., & Sarov, I. (1976) / . Mol. Evol. 8, 137-142 18. Fraenkel-Conrat, H. & Steinschneider, A. (1968) in Methods in Enzymology (Grossman, L. & Moldave, K., eds.) Vol. 12B, pp. 243-246, Academic Press, New York 19. Muramatsu, M. (1973) in Methods in Cell Biology (Prescott, D.M., ed.) Vol. 7, pp. 23-51, Academic Press, New York 20. Layne, E. (1957) in Methods in Enzymology (Colowick, S.P. & Kaplan, N.O., eds.) Vol. 3, pp. 447454, Academic Press, New York 21. Burton, K. (1968) in Methods in Enzymology (Grossman, L. & Moldave, K., eds.) Vol. 12A, pp. 163-166, Academic Press, New York 22. Schneider, W.C. (1957) in Methods in Enzymology (Colowick, S.P. & Kaplan, N.O., eds.) Vol. 3, pp. 680-684, Academic Press, New York 23. Twu, J.S. & Bretthauer, R.K. (1971) Biochemistry 10,1576-1582 24. Wray, W., Conn, P.M., & Wray, V.P. (1977) in Methods in Cell Biology (Stein, G., Stein, J., & Kleinsmith, L.J., eds.) Vol. 16, pp. 69-86, Academic Press, New York
Vol. 86, No. 5, 1979
1311 25. Roeder, R.G. (1976) in RNA Polymerase (Losick, R. & Chamberlin, M., eds.) pp. 285-329, Cold Spring Harbor Laboratory 26. Grunberg-Manago, M. (1963) in Progress in Nucleic Acid Research (Davidson, J.N. & Cohn, W.E., eds.) pp. 93-133, Academic Press, New York 27. See, Y.P. & Fitt, P.S. (1970) Biochem. J. 119, 517-524 28. Deutscher, M.P. (1973) / . Biol. Chem. 248, 31163121 29. Mukerji, S.K. & Deutscher, M.P. (1972) / . Biol. Chem. 247, 481^88 30. Fitt, P.S. (1966) / . Protozool. 13, 507-509 31. Kobayashi, S. (1965) / . Biochem. 58, 444-457 32. Flavell, R.A. & Jones, I.G. (1970) Biochem. J. 116, 811-817 33. Brawerman, G. (1976) in Progress in Nucleic Acid Research and Molecular Biology (Cohn, W.E., ed.) Vol. 17, pp. 117-148, Academic Press, New York 34. Rose, K.M., Roe, F.J., & Jacob, S.T. (1977) Biochim. Biophys. Ada 478, 180-191 35. Sawicki, S.G., Jelinek, W., & Darnell, J.E. (1977) / . Mol. Biol. 113, 219-235
