Vol. 130, No. 2
JOURNAL OF BACTERIOLOGY, May 1977, p. 869-876 Copyright C 1977 American Society for Microbiology
Printed in U.S.A.
Isolation and Characterization of a Polynucleotide Phosphorylase from Bacillus amyloliquefaciens ROBERT J. ERICKSON* AND JOSEPHINE C. GROSCH Molecular Biology Department, Miles Laboratories, Inc., Elkhart, Indiana 46514
Received for publication 8 December 1976
Bacillus amyloliquefaciens BaM-2 produces large amounts of extracellular and the synthesis of these proteins appears to be dependent upon abnormal ribonucleic acid metabolism. A polynucleotide phosphorylase (nucleoside diphosphate:polynucleotide nucleotidyl transferase) was identified, purified, and characterized from this strain. The purification scheme involved cell disruption, phase partitioning, differential (NH4)2SO4 solubilities, agarose gel filtration, and diethylaminoethyl-Sephadex chromatography. The purified enzyme demonstrated the reactions characteristic of polynucleotide phosphorylase: polymerization, phosphorolysis, and inorganic phosphate exchange with the phosphate of a nucleotide diphosphate. The enzyme was apparently primer independent and required a divalent cation. The reactions for the synthesis of the homopolyribonucleotides, (A),, and (G),, were optimized with respect to pH and divalent cation concentration. The enzyme is sensitive to inhibition by phosphate ion and heparin and is partially inhibited by rifamycin SV and synthetic polynucleotides. enzymes,
3-
The enzyme polynucleotide phosphorylase (PNPase) has been isolated from a variety of microorganisms (7,21) and catalyzes the biochemical reaction summarized in the following reaction scheme: NDP + (NMP),, f (NMP), +1 + Pi, where NDP is nucleoside 5'-diphosphate, NMP is nucleoside monophosphate, and Pi is inorganic pyrophosphate. The forward reaction results in the polymerization of ribonucleoside5'-diphosphates and has been important in the supply of synthetic polyribonucleic acids for biological research. Some forms of PNPase require that n be greater than a certain critical value (primer-dependent PNPase), whereas other forms will initiate synthesis from the component mononucleotides (10). Primer-dependent syntheses have been utilized to modify naturally occurring ribonucleic acid (RNA) by addition of a defined sequence to the 3'-termini of the molecule (17). The reverse reaction results in the processive degradation of polynucleotide chains in the 3' to 5' direction (5,11), and in this capacity the enzyme has been utilized to remove (adenylic acid),, [(A)j] segments from eucaryotic messenger RNA (mRNA) molecules (20,23). The apparent ubiquity of the enzyme in microorganisms suggests an important role in cell physiology; however, an unequivocal demonstration of its function in cell metabolism is absent from the literature. Two divergent, contradictory views have been presented to explain
the role of PNPase. One view suggests that PNPase participates in RNA metabolism by increasing the turnover rate of ribonucleotides through the processive degradative pathway (15,18). The opposing viewpoint suggests that the polymerization reaction is the physiologically important function and that the enzyme may modify the 3'-hydroxyl end of mRNA molecules so as to increase the half-life of the message (16). The recent demonstration of (A),, segments on the 3'-termini of some mRNA molecules in Escherichia coli (12) may add support to this latter viewpoint and may indicate that only certain classes of mRNA's are modified. Bacillus amyloliquefaciens produces large quantities of extracellular proteins and the mRNA molecules for these products have an apparently long half-life (4, 6). We have initiated a study of RNA metabolism in B. amyloliquefaciens and, in this report, describe the purification and characterization of one of two PNPase activities found in cell extracts. MATERIALS AND METHODS Bacterial cells. B. amyloliquefaciens BaM-2 is asporogenic and was isolated by E. W. Boyer. Large quantities of the strain were grown and concentrated for us by G. Mercer of the Marschall Research Laboratory, Miles Laboratories. Chemicals. Isotopically labeled and unlabeled synthetic polynucleotides are products of Miles Research Products Division. Tritiated NDPs and inorganic 32p were purchased from Schwarz/Mann, and 869
870
J. BACTERIOL.
ERICKSON & GROSCH
heparin, actinomycin D, rifamycin SV, N-ethylmaleimide, and dithiothreitol (DTT) were obtained from Calbiochem. The nonionic detergent Nonidet P-40 (NP-40) was supplied by Shell Chemical Co. All reagents used in the gel electrophoresis were products of Canalco. Phenylmethylsulfonyl fluoride (PMSF) was a product of Sigma, polyethylene glycol 6000 (PEG) was from Matheson, and (NH4)2SO4 (enzyme grade) was from Schwarz/Mann. Bio-Gel A 1.5m (200 to 400 mesh) was purchased from Bio-Rad, and the dextran T-500 and diethylaminoethyl (DEAE)-Sephadex A-50 were from Pharmacia. Standard polymerization assay. A standard reaction was carried out in a volume of 50 ul at 370 C in a mixture that contained the following: adenosine 5'diphosphate (ADP), 6.8 nmol; ethylenediaminetetraacetic acid (EDTA), 68 nmol, NH4Cl, 1.0 ,imol; MgCl2, 0.45 A mol; DTT, 45 pmol; NP-40, 0.03%; dium cacodylate (pH 5.0), 1.1 ,umol; and [3H]ADP was added to a specific activity of 20 cpm/pmol of ADP in our systems of quantitation. After 20 min of incubation, the reaction was stopped by the addition of a mixture of trichloroacetic acid (100%)-saturated Na4P207-saturated NaH2PO4 in a volume ratio of 1:1:1 at 4°C. The insoluble product was collected and washed on nitrocellulose filters. The filters were dried, placed in a toluene-based scintillation fluid, and assayed in a liquid scintillation spectrometer. Standard phosphorolysis assay. The phosphorolytic activity of the preparations was assessed in 50 Al of reaction mixture containing: P042, 90 nmol; MgCl2, 0.45 ,Lmol; EDTA, 68 nmol; NH4Cl, 1.0 ,umol; DTT, 45 pmol; tris-(hydroxymethyl)aminomethane (Tris)-hydrochloride (pH 7.5), 2.5 ,umol; NP-40, 0.03%; and [3H](A)0, 100 ng (specific activity of 87 cpm/ng). The reaction was incubated at 37°C for 20 min and assayed for remaining acid-insoluble polynucleotide as described for the polymerase assay. Assay for inorganic [32P]ADP exchange reaction. The exchange reaction between the S-phosphate of ADP and inorganic [32p] was carried out as previously described (16). E. coli PNPase was obtained from J. Colbourn of Miles Research Products Department and used as a positive control. Enzyme purification. (i) Cell disruption. Fifty grams of cells frozen in 8% dimethyl sulfoxide were suspended in 200 ml of buffer I (0.01 M Tris-hydrochloride, pH 8.4, 1 mM EDTA, pH 7, 0.01 M MgCl2, 0.3 mM DTT). The cell suspension was centrifuged at 12,000 x g for 10 min. The cell pellet was resuspended in 125 ml of buffer I plus 20 ml of 25-,tm glass powder (Heat Systems). The cell suspension was disrupted for 10 min with a sonifier (Bronson model S110) at 10 A. The temperature during cell breakage was held at 8°C or below by alternating the sample from an ice to ethylene glycol-dry ice bath. Forty milligrams of PMSF freshly dissolved in 4 ml of 95% ethanol was added to the cell extract after min 1 of breakage. The extract was centrifuged at 12,000 x g for 1.5 h. The supernatant fraction (143 ml) was designated fraction I and retained. (ii) Phase partition. Fraction I was partitioned between phases of dextran and PEG by the Shorenstein and Losick (19) modification of the Babinet (2) procedure. Thirty-five milligrams of PMSF and 0.3 so-
ml of 0.1 M DTT were added per 100 ml of fraction I. Fraction I (143 ml) was treated with 45.9 ml of 30% (wt/wt) PEG and 16.5 ml of 20% (wt/wt) dextran, both dissolved in water. After stirring for 30 min, the mixture was centrifuged at 14,000 x g for 10 min. Two phases were obtained; the upper phase, containing the PEG, was discarded. To the dextran phase (26 ml), 62 ml of buffer I, 9 mg of PMSF, 0.075 ml of 0.1 M DTT, 25.4 ml of 30% PEG, and 13.2 g of NaCl were added. The mixture was stirred for 30 min and centrifuged as before. The PEG phase was discarded. To the dextran phase (22 ml), 52.6 ml of buffer I, 8 mg of PMSF, 0.07 ml of 0.1 M DTT, 21.6 ml of 30% PEG, and 22.7 g of NaCl were added. The mixture was stirred for 45 min and centrifuged as before. The dextran phase was discarded. The PEG phase (80 ml) containing the PNPase activity was dialyzed for 2 h against two changes of 4 volumes of buffer I plus 0.05 M NaCl and 5% (vol/vol) glycerol. (iii) (NH4)2SO4 fractionation. After dialysis, (NH4)2SO4 (16.3 g/100 ml) was added, and 1 N NaOH was used to maintain the pH at 7.6. The mixture was stirred for 30 min and centrifuged 5 min at 10,000 x g. Two phases were obtained; the upper phase, containing the PEG, was discarded. To the lower phase, (NH4)2SO4 (7 g/100 ml) was added; the mixture was stirred for 30 min and centrifuged 45 min at 23,500 x g, and the precipitate was discarded. The 30 to 52% (NH4)2SO4 precipitate was made by adding (10 g/100 ml) (NH4)2SO4 to the supernatant fraction. The mixture was stirred and centrifuged as described before, and the precipitate was dissolved in 4 ml of buffer II (0.01 M Trishydrochloride, pH 8, 1 mM EDTA, pH 7, 0.01 M MgCl2, 0.3 mM DTT, 10% [vol/vol] glycerol, and 0.1 M NaCl) and designated AS 52. (iv) Agarose gel filtration. AS 52 was applied to a Pharmacia K26/40 column of Bio-Gel A-1.5-m agarose and eluted with buffer II plus 0.1 M NaCl at a flow rate of 5 ml/h. Fractions of 2.2 ml were collected, and 5 ,ul was assayed for enzyme activity. The effluent in tubes 29 to 34 (PNPase I) and tubes 40 to 42 (PNPase II) showed activity (See Fig. 1). These two activities were pooled separately and dialyzed overnight against storage buffer (0.01 M Tris, pH 8.4, 0.01 M MgCl2, 0.1 mM EDTA, pH 7,0.03 mM DTT, 25 mM NH4Cl, 50% [vol/vol] glycerol). (v) DEAE-Sephadex chromatography. PNPase I activity from the agarose gel filtration step was dialyzed into buffer II containing 0.05 M NaCl. This material was applied to a column of DEAE-Sephadex (Pharmacia K9/15) at a flow rate of 10 ml/h. The material was fractionated by stepwise elution with buffer II, which contained 0.05 M, 0.1 M, 0.3 M, and 0.5 M NaCl. Fractions 50 to 56 eluted with 0.5 M NaCl contained the PNPase I activity. This material was pooled and dialyzed into storage buffer. (vi) Polyacrylamide gel electrophoresis. Sodium dodecyl sulfate (SDS)-gel electrophoresis was performed as described by Weber and Osborn (22) on gels (8 cm long) containing 7.5% acrylamide and 0.1% SDS. Five to twenty microliters from each fraction containing PNPase activity were diluted 1:1 into a gel buffer containing 0.01 M sodium phosphate, pH 7.2, 3% SDS, 4 M urea, and 0.002 M DTT
VOL. 130, 1977
=~ ~ ~.
B. AMYLOLIQUEFACIENS POLYNUCLEOTIDE PHOSPHORYLASE
and heated for 2 min in a bath of boiling water. The samples were then dialyzed overnight at room temperature against gel buffer. To the dialysate was added glycerol to a concentration of 10% (vol/vol) and 0.003% bromophenol blue. One hundred fifty microliters of sample was applied to each gel. Samples were stacked at 1 mA per tube for 2 h and subjected to electrophoresis at 15 mA per gel for 3.5 h. Gels were stained with a 0.25% (wt/vol) solution of Coomassie brilliant blue R in methanol-acetic acid-water (5:1:5, vol/vol/vol) for 2 h. The gels were electrophoretically destained in 7.5% acetic acid. The molecular weight of PNPase I was determined by SDS-gel electrophoresis as described previously (22).
RESULTS Separation of two PNPase activities by aga-
gel filtration. When the material from final phase partition was fractionated by differential solubility in 32 to 52% (NH4)2S04 and the resulting precipitate (AS 52) was dissolved in buffer and fractionated by agarose gel filtration, two peaks of PNPase activity were resolved (Fig. 1). The major peak (PNPase I) was eluted first and was followed by a second orthophosphate-sensitive small peak of activity (PNPase II). The fractions demonstrating PNPase I and PNPase II activities were pooled separately, and the degree of purity was assessed by SDS-polyacrylamide gel electrophoresis. The PNPase I fraction showed significantly fewer bands than PNPase II (Fig. 2A) and also rose
B
A
871
C
Ii
FIG. 2. SDS-polyacrylamide gel electrophoresis of samples of PNPase activity. (A) Pooled agarose fractions of PNPase I activity, (B) pooled agarose fractions of PNPase II activity, (C) PNPase I activity eluted from DEAE-Sephadex column.
demonstrated the reactions of a true PNPase. The relationship between the two activities is, at present, not clear, and PNPase II will be the subject of future investigations. The remaining analyses pertain to the further purification and characterization of PNPase I. DEAE-Sephadex chromatography. The pooled PNPase I material described above was fractionated by stepwise elution on DEAESephadex columns. The active material was eluted from the column in the presence of 0.5 M NaCl (Fig. 3). Gel electrophoresis in the presence of SDS and urea produced a single band of protein (Fig. 2C) that migrated at a rate characteristic of a protein of molecular weight 7.4 x 104 (Fig. 4). PNPase I is composed of a single j100 polypeptide chain or oligomers thereof. The specific activity of this material was calculated 40 to be 170 nmol of AMP incorporated per mg of so protein per min. Effect of pH and divalent cations on polymerization activity. Maximum polymerization 60 activity for the incorporation of ADP into acidinsoluble material was observed in cacodylate buffer at pH 5.0 (Fig. 5A), and pH dependence appeared as a relatively sharp peak. The syn~40 so0I thesis of (guanylic acid)n [(G),,] from GDP, on ~~~~~~60 the other hand, produced a broad pH range j20 with maximal polymerization at pH 7.0 in Trislt40~ hydrochloride buffer (Fig. 5B). The concentra2 0 1a tion of the Tris-hydrochloride buffer (i.e., 1.0 to 10 mM) did not alter the extent of polymerization, whereas the cacodylate buffer was found to be inhibitory at a concentration above 2.0 mM. Fracffon Number The polymerization reaction is completely deFIG. 1. Agarose gel filtration on AS 52 fraction of BaM-2 cell extracts obtained as described in Materi- pendent upon the presence of divalent cations. als and Methods. Symbols: Solid line, percent of Both the syntheses of (A),, and (G),, exhibit this absorbance at 260 nm, (0) micromoles of GMP incor- dependence: (A),, synthesis is optimal in 10 mM MgCl2, and (G). synthesis is optimal at 5 mM porated per 50-,pl sample in 20 min. la
872
J. BACTERIOL.
ERICKSON & GROSCH
=E
0 020
0
0
'
I0.3M
15'
0
0.1 M
S
I
'I
0.75 K 10
¶5
I
I
I
~I'
I
a
.0 4c
= B
0 a
a
.0
1.0
I
a
0
10
A
I
0.50 "
0
0
CP a a-
5
_S 00CM
nlI
20
10
0
_
X
_
~~~~~~~~~ ~ ~
S OC00
v
LX
30
40
Il
0.25
2
|I
I
50
In
v
60
Fraction Number FIG. 3. DEAE-Sephadex chromatography of PNPase L. Symbols: Solid line, percentage of absorbance at 260 nm; (0) nanomoles of AMP incorporated per 50-sl sample in 20 min; numbers above bars signify molarity of NaCl used for elution at designated points in fractionation. 9-
8~ b
7-
6_
PNPeSa I
%b
BSA
O'
5-~
IgO (H chain)
6, 0
ovalbumin
4-
x 4AC
3-
a
19 (L chain)
o "
\
I
O
I
I
I
.2
.4
.8
trypsin
I
.8
1.0
Mobility
FIG. 4. Estimation of molecular weight of PNPase
I by SDS-polyacrylamide gel electrophoresis.
MgCl2. At higher concentrations, Mg2+ beinhibitory in both systems (Fig. 6). Manganese can serve as the cation in both comes
syntheses, but the final amount of polymerization is only 0.5 of that observed in the presence of Mg2+. Inorganic [32P]ADP exchange reaction. In the presence of NDP and phosphate, PNPase can exchange the 8-phosphate of the nucleotide with orthophosphate (7, 21). This reaction is a distinguishing property of PNPase. Table 1 shows that PNPase I of B. amyloliquefaciens carries out this reaction as efficiently as the well-characterized PNPase of E. coli. PNPase I activity as a function of time of incubation. The reaction stoichiometry of PNPase indicates that as the polymerization reaction proceeds, the accumulation of orthophosphate will tend to drive the reaction in the opposite direction. The polymerization reaction studied as a function of time demonstrates that this is the case for both (A),, and (G), syntheses (Fig. 7). In the case of (A). synthesis, the forward reaction continues for 30 min at a point when approximately 45% of the added ADP has been rendered acid insoluble. At this point, the phosphorolysis reaction becomes dominant and polymer degradation becomes evident. In the case of (G), synthesis, only about 5% of the added NDP is polymerized before the reaction is inhibited. The cause of this pronounced inhibition is not known. In addition, there is only a slight amount of (G),, degradation during the time course of the experiment. Phosphorolysis reaction of PNPase I. The phosphorolysis reaction evident in the preceeding experiments can be studied by following the
VOL. 130, 1977
B. AMYLOLIQUEFACIENS POLYNUCLEOTIDE PHOSPHORYLASE
873
Q.5
~0
0 0
0
0
0.4
c
a.
1.0
~50.3 0
4t
02.2 z
a 0.5 0
E
.!0.1
c
0
E IC
0
I
0
0 .~0.4
h.06
40
[MgCI2I
mM
80
FIG. 6. Dependence of polymerization reaction on divalent cation concentration. PNPase I was added to standard polymerization reaction at a concentration of 0.5 pg per 50-pl assay. Symbols: (O) (A)n synthesis, (0) (G),, synthesis.
0
0
20
10
0.
-C IL
a 0.2
TABLE 1. Exchange reaction between inorganic phosphate and the /-phosphate of ADP catalyzed by PNPase I and E. coli PNPase 9.0
8.0
7.0 p
6.0
Sample descriptiona 5.0
H
FIG. 5. Dependence of polymerization reaction on pH. With the exception ofpH, the standard polymerization assay was used in the presence of 1.0 pg of enzyme. (A) Response for (A),, synthesis; (B) synthesis of (G),,. Symbols: (@) Tris-hydrochloride buffer, (0) sodium cacodylate buffer.
degradation of [3H](A),, in the absence of the NDPs. The stoichiometry of the reaction indicates that the degradative pathway would require PO42-, and this is shown to be the case in Fig. 8A. The purified PNPase I shows no degradation of [3H](A),L after 30 min of incubation in the absence of orthophosphate, indicating no contaminating ribonuclease activity. The addition of as little as 2.0 nmol Of P042- initiates phosphorolysis, and the degradation is complete in the presence of 20 nmol. The resistant isotope remaining in the presence of high levels of orthophosphate may be related to the resistant fraction described previously (11), which was postulated to be residual oligomeric material that cannot bind to the enzyme and is, therefore, not degraded. The dependence of phosphorolysis upon pH is shown in Fig. 8B, and the data suggest an
cpm retained on fil-
ter
PNPase I from BaM-2 89,000 PNPase from E. coli 67,000 400 No protein a Concentrations of the enzymes as PNPase I, 10 ,ug per assay, and E. coli, 2.8 U per assay, as obtained from Miles Research Products.
explanation for the pH optimum of the polymerization reaction of ADP. The polymer is rapidly degraded at neutral or alkaline pH, but the phosphorolysis reaction is almost completely inhibited in cacodylate buffer at pH 5.0 to 4.5. Inhibitory studies on PNPase I. The effects of various inhibitors of PNPase I are summarized in Fig. 9. As expected, inorganic PO42markedly inhibits the reaction at levels above 5.0 nmol per assay. Rifamycin SV, a potent inhibitor of B. subtilis deoxyribonucleic acid (DNA)-dependent RNA polymerase, demonstrates partial inhibition of the reaction, but only at relatively high concentrations. Unexpectedly, heparin proved to be a very potent inhibitor of PNPase I. It is also of interest to note that both (A),, and (G),, inhibited the synthesis of (A),, by PNPase I. The reaction proved relatively insensitive to N-ethylmaleimide and high concentrations of KCI.
874
J. BACTERIOL.
ERICKSON & GROSCH
DISCUSSION The presence of PNPase in extracts of B. subtilis has been previously described (3, 21) but was not studied in detail. PNPase I of B. amyloliquefaciens, as described in the present investigation, is an apparently primer-independent form of the enzyme that can be purified to a single protein band upon SDS-polyacrylamide gel electrophoresis. The primer inde22.5 CL
2 0
~0 1.5 z
a 1.0
o
0
E c
0.5
L
0
30
0
60
120
Minutes Incubation
FIG. 7. Polymerization as a function of incubation time. Enzyme was added at a level of 1.0 pg per assay. Symbols: (a) (A)n synthesis, (0) (G)n synthesis.
pendence of our preparation must be qualified, since we have not rigorously proved the absence of nucleic acid contamination, and very small amounts of nucleic acid tightly bound to the enzyme could escape detection by conventional means. The immediate initiation of (A), and (G)" synthesis at the maximal rate might be indicative of the existence of such a contaminant (7). The enzyme possesses many of the properties and characteristics of the well-studied PNPase of Micrococcus luteus (5, 7, 10, 11, 21). PNPase I can carry out the polymerization, phosphorolysis, and inorganic [32P]ADP exchange reactions that are characteristic of this class of enzyme. The divalent cation requirement, as well as the inhibition by high concentration of the cations, is comparable to that described for the other enzymes (7, 21). The major difference in the reaction conditions utilized in studies with B. amyloliquefaciens PNPase I is the relatively low pH optimum. Although the pH optimum is dependent upon reaction conditions, the value is usually reported to be between a pH of 8 and 9 for other enzymes. In addition, the molecular weight value of 7.4 x 104 for PNPase I is significantly smaller than the 2.5 x 105 value reported for the M. luteus enzyme (10). The existence of two separable PNPase activities is not unique to B. amyloliquefaciens. The PNPase of M. luteus is purified as a primerindependent form which, upon limited tryptic hydrolysis, yields a slightly smaller primerdependent form of the enzyme (10). B. amyloliquefaciens produces large quantities of extracellular proteases, and these enzymes may act
A
100
B
so 01
3)
c
-C
JOURNAL OF BACTERIOLOGY, May 1977, p. 869-876 Copyright C 1977 American Society for Microbiology
Printed in U.S.A.
Isolation and Characterization of a Polynucleotide Phosphorylase from Bacillus amyloliquefaciens ROBERT J. ERICKSON* AND JOSEPHINE C. GROSCH Molecular Biology Department, Miles Laboratories, Inc., Elkhart, Indiana 46514
Received for publication 8 December 1976
Bacillus amyloliquefaciens BaM-2 produces large amounts of extracellular and the synthesis of these proteins appears to be dependent upon abnormal ribonucleic acid metabolism. A polynucleotide phosphorylase (nucleoside diphosphate:polynucleotide nucleotidyl transferase) was identified, purified, and characterized from this strain. The purification scheme involved cell disruption, phase partitioning, differential (NH4)2SO4 solubilities, agarose gel filtration, and diethylaminoethyl-Sephadex chromatography. The purified enzyme demonstrated the reactions characteristic of polynucleotide phosphorylase: polymerization, phosphorolysis, and inorganic phosphate exchange with the phosphate of a nucleotide diphosphate. The enzyme was apparently primer independent and required a divalent cation. The reactions for the synthesis of the homopolyribonucleotides, (A),, and (G),, were optimized with respect to pH and divalent cation concentration. The enzyme is sensitive to inhibition by phosphate ion and heparin and is partially inhibited by rifamycin SV and synthetic polynucleotides. enzymes,
3-
The enzyme polynucleotide phosphorylase (PNPase) has been isolated from a variety of microorganisms (7,21) and catalyzes the biochemical reaction summarized in the following reaction scheme: NDP + (NMP),, f (NMP), +1 + Pi, where NDP is nucleoside 5'-diphosphate, NMP is nucleoside monophosphate, and Pi is inorganic pyrophosphate. The forward reaction results in the polymerization of ribonucleoside5'-diphosphates and has been important in the supply of synthetic polyribonucleic acids for biological research. Some forms of PNPase require that n be greater than a certain critical value (primer-dependent PNPase), whereas other forms will initiate synthesis from the component mononucleotides (10). Primer-dependent syntheses have been utilized to modify naturally occurring ribonucleic acid (RNA) by addition of a defined sequence to the 3'-termini of the molecule (17). The reverse reaction results in the processive degradation of polynucleotide chains in the 3' to 5' direction (5,11), and in this capacity the enzyme has been utilized to remove (adenylic acid),, [(A)j] segments from eucaryotic messenger RNA (mRNA) molecules (20,23). The apparent ubiquity of the enzyme in microorganisms suggests an important role in cell physiology; however, an unequivocal demonstration of its function in cell metabolism is absent from the literature. Two divergent, contradictory views have been presented to explain
the role of PNPase. One view suggests that PNPase participates in RNA metabolism by increasing the turnover rate of ribonucleotides through the processive degradative pathway (15,18). The opposing viewpoint suggests that the polymerization reaction is the physiologically important function and that the enzyme may modify the 3'-hydroxyl end of mRNA molecules so as to increase the half-life of the message (16). The recent demonstration of (A),, segments on the 3'-termini of some mRNA molecules in Escherichia coli (12) may add support to this latter viewpoint and may indicate that only certain classes of mRNA's are modified. Bacillus amyloliquefaciens produces large quantities of extracellular proteins and the mRNA molecules for these products have an apparently long half-life (4, 6). We have initiated a study of RNA metabolism in B. amyloliquefaciens and, in this report, describe the purification and characterization of one of two PNPase activities found in cell extracts. MATERIALS AND METHODS Bacterial cells. B. amyloliquefaciens BaM-2 is asporogenic and was isolated by E. W. Boyer. Large quantities of the strain were grown and concentrated for us by G. Mercer of the Marschall Research Laboratory, Miles Laboratories. Chemicals. Isotopically labeled and unlabeled synthetic polynucleotides are products of Miles Research Products Division. Tritiated NDPs and inorganic 32p were purchased from Schwarz/Mann, and 869
870
J. BACTERIOL.
ERICKSON & GROSCH
heparin, actinomycin D, rifamycin SV, N-ethylmaleimide, and dithiothreitol (DTT) were obtained from Calbiochem. The nonionic detergent Nonidet P-40 (NP-40) was supplied by Shell Chemical Co. All reagents used in the gel electrophoresis were products of Canalco. Phenylmethylsulfonyl fluoride (PMSF) was a product of Sigma, polyethylene glycol 6000 (PEG) was from Matheson, and (NH4)2SO4 (enzyme grade) was from Schwarz/Mann. Bio-Gel A 1.5m (200 to 400 mesh) was purchased from Bio-Rad, and the dextran T-500 and diethylaminoethyl (DEAE)-Sephadex A-50 were from Pharmacia. Standard polymerization assay. A standard reaction was carried out in a volume of 50 ul at 370 C in a mixture that contained the following: adenosine 5'diphosphate (ADP), 6.8 nmol; ethylenediaminetetraacetic acid (EDTA), 68 nmol, NH4Cl, 1.0 ,imol; MgCl2, 0.45 A mol; DTT, 45 pmol; NP-40, 0.03%; dium cacodylate (pH 5.0), 1.1 ,umol; and [3H]ADP was added to a specific activity of 20 cpm/pmol of ADP in our systems of quantitation. After 20 min of incubation, the reaction was stopped by the addition of a mixture of trichloroacetic acid (100%)-saturated Na4P207-saturated NaH2PO4 in a volume ratio of 1:1:1 at 4°C. The insoluble product was collected and washed on nitrocellulose filters. The filters were dried, placed in a toluene-based scintillation fluid, and assayed in a liquid scintillation spectrometer. Standard phosphorolysis assay. The phosphorolytic activity of the preparations was assessed in 50 Al of reaction mixture containing: P042, 90 nmol; MgCl2, 0.45 ,Lmol; EDTA, 68 nmol; NH4Cl, 1.0 ,umol; DTT, 45 pmol; tris-(hydroxymethyl)aminomethane (Tris)-hydrochloride (pH 7.5), 2.5 ,umol; NP-40, 0.03%; and [3H](A)0, 100 ng (specific activity of 87 cpm/ng). The reaction was incubated at 37°C for 20 min and assayed for remaining acid-insoluble polynucleotide as described for the polymerase assay. Assay for inorganic [32P]ADP exchange reaction. The exchange reaction between the S-phosphate of ADP and inorganic [32p] was carried out as previously described (16). E. coli PNPase was obtained from J. Colbourn of Miles Research Products Department and used as a positive control. Enzyme purification. (i) Cell disruption. Fifty grams of cells frozen in 8% dimethyl sulfoxide were suspended in 200 ml of buffer I (0.01 M Tris-hydrochloride, pH 8.4, 1 mM EDTA, pH 7, 0.01 M MgCl2, 0.3 mM DTT). The cell suspension was centrifuged at 12,000 x g for 10 min. The cell pellet was resuspended in 125 ml of buffer I plus 20 ml of 25-,tm glass powder (Heat Systems). The cell suspension was disrupted for 10 min with a sonifier (Bronson model S110) at 10 A. The temperature during cell breakage was held at 8°C or below by alternating the sample from an ice to ethylene glycol-dry ice bath. Forty milligrams of PMSF freshly dissolved in 4 ml of 95% ethanol was added to the cell extract after min 1 of breakage. The extract was centrifuged at 12,000 x g for 1.5 h. The supernatant fraction (143 ml) was designated fraction I and retained. (ii) Phase partition. Fraction I was partitioned between phases of dextran and PEG by the Shorenstein and Losick (19) modification of the Babinet (2) procedure. Thirty-five milligrams of PMSF and 0.3 so-
ml of 0.1 M DTT were added per 100 ml of fraction I. Fraction I (143 ml) was treated with 45.9 ml of 30% (wt/wt) PEG and 16.5 ml of 20% (wt/wt) dextran, both dissolved in water. After stirring for 30 min, the mixture was centrifuged at 14,000 x g for 10 min. Two phases were obtained; the upper phase, containing the PEG, was discarded. To the dextran phase (26 ml), 62 ml of buffer I, 9 mg of PMSF, 0.075 ml of 0.1 M DTT, 25.4 ml of 30% PEG, and 13.2 g of NaCl were added. The mixture was stirred for 30 min and centrifuged as before. The PEG phase was discarded. To the dextran phase (22 ml), 52.6 ml of buffer I, 8 mg of PMSF, 0.07 ml of 0.1 M DTT, 21.6 ml of 30% PEG, and 22.7 g of NaCl were added. The mixture was stirred for 45 min and centrifuged as before. The dextran phase was discarded. The PEG phase (80 ml) containing the PNPase activity was dialyzed for 2 h against two changes of 4 volumes of buffer I plus 0.05 M NaCl and 5% (vol/vol) glycerol. (iii) (NH4)2SO4 fractionation. After dialysis, (NH4)2SO4 (16.3 g/100 ml) was added, and 1 N NaOH was used to maintain the pH at 7.6. The mixture was stirred for 30 min and centrifuged 5 min at 10,000 x g. Two phases were obtained; the upper phase, containing the PEG, was discarded. To the lower phase, (NH4)2SO4 (7 g/100 ml) was added; the mixture was stirred for 30 min and centrifuged 45 min at 23,500 x g, and the precipitate was discarded. The 30 to 52% (NH4)2SO4 precipitate was made by adding (10 g/100 ml) (NH4)2SO4 to the supernatant fraction. The mixture was stirred and centrifuged as described before, and the precipitate was dissolved in 4 ml of buffer II (0.01 M Trishydrochloride, pH 8, 1 mM EDTA, pH 7, 0.01 M MgCl2, 0.3 mM DTT, 10% [vol/vol] glycerol, and 0.1 M NaCl) and designated AS 52. (iv) Agarose gel filtration. AS 52 was applied to a Pharmacia K26/40 column of Bio-Gel A-1.5-m agarose and eluted with buffer II plus 0.1 M NaCl at a flow rate of 5 ml/h. Fractions of 2.2 ml were collected, and 5 ,ul was assayed for enzyme activity. The effluent in tubes 29 to 34 (PNPase I) and tubes 40 to 42 (PNPase II) showed activity (See Fig. 1). These two activities were pooled separately and dialyzed overnight against storage buffer (0.01 M Tris, pH 8.4, 0.01 M MgCl2, 0.1 mM EDTA, pH 7,0.03 mM DTT, 25 mM NH4Cl, 50% [vol/vol] glycerol). (v) DEAE-Sephadex chromatography. PNPase I activity from the agarose gel filtration step was dialyzed into buffer II containing 0.05 M NaCl. This material was applied to a column of DEAE-Sephadex (Pharmacia K9/15) at a flow rate of 10 ml/h. The material was fractionated by stepwise elution with buffer II, which contained 0.05 M, 0.1 M, 0.3 M, and 0.5 M NaCl. Fractions 50 to 56 eluted with 0.5 M NaCl contained the PNPase I activity. This material was pooled and dialyzed into storage buffer. (vi) Polyacrylamide gel electrophoresis. Sodium dodecyl sulfate (SDS)-gel electrophoresis was performed as described by Weber and Osborn (22) on gels (8 cm long) containing 7.5% acrylamide and 0.1% SDS. Five to twenty microliters from each fraction containing PNPase activity were diluted 1:1 into a gel buffer containing 0.01 M sodium phosphate, pH 7.2, 3% SDS, 4 M urea, and 0.002 M DTT
VOL. 130, 1977
=~ ~ ~.
B. AMYLOLIQUEFACIENS POLYNUCLEOTIDE PHOSPHORYLASE
and heated for 2 min in a bath of boiling water. The samples were then dialyzed overnight at room temperature against gel buffer. To the dialysate was added glycerol to a concentration of 10% (vol/vol) and 0.003% bromophenol blue. One hundred fifty microliters of sample was applied to each gel. Samples were stacked at 1 mA per tube for 2 h and subjected to electrophoresis at 15 mA per gel for 3.5 h. Gels were stained with a 0.25% (wt/vol) solution of Coomassie brilliant blue R in methanol-acetic acid-water (5:1:5, vol/vol/vol) for 2 h. The gels were electrophoretically destained in 7.5% acetic acid. The molecular weight of PNPase I was determined by SDS-gel electrophoresis as described previously (22).
RESULTS Separation of two PNPase activities by aga-
gel filtration. When the material from final phase partition was fractionated by differential solubility in 32 to 52% (NH4)2S04 and the resulting precipitate (AS 52) was dissolved in buffer and fractionated by agarose gel filtration, two peaks of PNPase activity were resolved (Fig. 1). The major peak (PNPase I) was eluted first and was followed by a second orthophosphate-sensitive small peak of activity (PNPase II). The fractions demonstrating PNPase I and PNPase II activities were pooled separately, and the degree of purity was assessed by SDS-polyacrylamide gel electrophoresis. The PNPase I fraction showed significantly fewer bands than PNPase II (Fig. 2A) and also rose
B
A
871
C
Ii
FIG. 2. SDS-polyacrylamide gel electrophoresis of samples of PNPase activity. (A) Pooled agarose fractions of PNPase I activity, (B) pooled agarose fractions of PNPase II activity, (C) PNPase I activity eluted from DEAE-Sephadex column.
demonstrated the reactions of a true PNPase. The relationship between the two activities is, at present, not clear, and PNPase II will be the subject of future investigations. The remaining analyses pertain to the further purification and characterization of PNPase I. DEAE-Sephadex chromatography. The pooled PNPase I material described above was fractionated by stepwise elution on DEAESephadex columns. The active material was eluted from the column in the presence of 0.5 M NaCl (Fig. 3). Gel electrophoresis in the presence of SDS and urea produced a single band of protein (Fig. 2C) that migrated at a rate characteristic of a protein of molecular weight 7.4 x 104 (Fig. 4). PNPase I is composed of a single j100 polypeptide chain or oligomers thereof. The specific activity of this material was calculated 40 to be 170 nmol of AMP incorporated per mg of so protein per min. Effect of pH and divalent cations on polymerization activity. Maximum polymerization 60 activity for the incorporation of ADP into acidinsoluble material was observed in cacodylate buffer at pH 5.0 (Fig. 5A), and pH dependence appeared as a relatively sharp peak. The syn~40 so0I thesis of (guanylic acid)n [(G),,] from GDP, on ~~~~~~60 the other hand, produced a broad pH range j20 with maximal polymerization at pH 7.0 in Trislt40~ hydrochloride buffer (Fig. 5B). The concentra2 0 1a tion of the Tris-hydrochloride buffer (i.e., 1.0 to 10 mM) did not alter the extent of polymerization, whereas the cacodylate buffer was found to be inhibitory at a concentration above 2.0 mM. Fracffon Number The polymerization reaction is completely deFIG. 1. Agarose gel filtration on AS 52 fraction of BaM-2 cell extracts obtained as described in Materi- pendent upon the presence of divalent cations. als and Methods. Symbols: Solid line, percent of Both the syntheses of (A),, and (G),, exhibit this absorbance at 260 nm, (0) micromoles of GMP incor- dependence: (A),, synthesis is optimal in 10 mM MgCl2, and (G). synthesis is optimal at 5 mM porated per 50-,pl sample in 20 min. la
872
J. BACTERIOL.
ERICKSON & GROSCH
=E
0 020
0
0
'
I0.3M
15'
0
0.1 M
S
I
'I
0.75 K 10
¶5
I
I
I
~I'
I
a
.0 4c
= B
0 a
a
.0
1.0
I
a
0
10
A
I
0.50 "
0
0
CP a a-
5
_S 00CM
nlI
20
10
0
_
X
_
~~~~~~~~~ ~ ~
S OC00
v
LX
30
40
Il
0.25
2
|I
I
50
In
v
60
Fraction Number FIG. 3. DEAE-Sephadex chromatography of PNPase L. Symbols: Solid line, percentage of absorbance at 260 nm; (0) nanomoles of AMP incorporated per 50-sl sample in 20 min; numbers above bars signify molarity of NaCl used for elution at designated points in fractionation. 9-
8~ b
7-
6_
PNPeSa I
%b
BSA
O'
5-~
IgO (H chain)
6, 0
ovalbumin
4-
x 4AC
3-
a
19 (L chain)
o "
\
I
O
I
I
I
.2
.4
.8
trypsin
I
.8
1.0
Mobility
FIG. 4. Estimation of molecular weight of PNPase
I by SDS-polyacrylamide gel electrophoresis.
MgCl2. At higher concentrations, Mg2+ beinhibitory in both systems (Fig. 6). Manganese can serve as the cation in both comes
syntheses, but the final amount of polymerization is only 0.5 of that observed in the presence of Mg2+. Inorganic [32P]ADP exchange reaction. In the presence of NDP and phosphate, PNPase can exchange the 8-phosphate of the nucleotide with orthophosphate (7, 21). This reaction is a distinguishing property of PNPase. Table 1 shows that PNPase I of B. amyloliquefaciens carries out this reaction as efficiently as the well-characterized PNPase of E. coli. PNPase I activity as a function of time of incubation. The reaction stoichiometry of PNPase indicates that as the polymerization reaction proceeds, the accumulation of orthophosphate will tend to drive the reaction in the opposite direction. The polymerization reaction studied as a function of time demonstrates that this is the case for both (A),, and (G), syntheses (Fig. 7). In the case of (A). synthesis, the forward reaction continues for 30 min at a point when approximately 45% of the added ADP has been rendered acid insoluble. At this point, the phosphorolysis reaction becomes dominant and polymer degradation becomes evident. In the case of (G), synthesis, only about 5% of the added NDP is polymerized before the reaction is inhibited. The cause of this pronounced inhibition is not known. In addition, there is only a slight amount of (G),, degradation during the time course of the experiment. Phosphorolysis reaction of PNPase I. The phosphorolysis reaction evident in the preceeding experiments can be studied by following the
VOL. 130, 1977
B. AMYLOLIQUEFACIENS POLYNUCLEOTIDE PHOSPHORYLASE
873
Q.5
~0
0 0
0
0
0.4
c
a.
1.0
~50.3 0
4t
02.2 z
a 0.5 0
E
.!0.1
c
0
E IC
0
I
0
0 .~0.4
h.06
40
[MgCI2I
mM
80
FIG. 6. Dependence of polymerization reaction on divalent cation concentration. PNPase I was added to standard polymerization reaction at a concentration of 0.5 pg per 50-pl assay. Symbols: (O) (A)n synthesis, (0) (G),, synthesis.
0
0
20
10
0.
-C IL
a 0.2
TABLE 1. Exchange reaction between inorganic phosphate and the /-phosphate of ADP catalyzed by PNPase I and E. coli PNPase 9.0
8.0
7.0 p
6.0
Sample descriptiona 5.0
H
FIG. 5. Dependence of polymerization reaction on pH. With the exception ofpH, the standard polymerization assay was used in the presence of 1.0 pg of enzyme. (A) Response for (A),, synthesis; (B) synthesis of (G),,. Symbols: (@) Tris-hydrochloride buffer, (0) sodium cacodylate buffer.
degradation of [3H](A),, in the absence of the NDPs. The stoichiometry of the reaction indicates that the degradative pathway would require PO42-, and this is shown to be the case in Fig. 8A. The purified PNPase I shows no degradation of [3H](A),L after 30 min of incubation in the absence of orthophosphate, indicating no contaminating ribonuclease activity. The addition of as little as 2.0 nmol Of P042- initiates phosphorolysis, and the degradation is complete in the presence of 20 nmol. The resistant isotope remaining in the presence of high levels of orthophosphate may be related to the resistant fraction described previously (11), which was postulated to be residual oligomeric material that cannot bind to the enzyme and is, therefore, not degraded. The dependence of phosphorolysis upon pH is shown in Fig. 8B, and the data suggest an
cpm retained on fil-
ter
PNPase I from BaM-2 89,000 PNPase from E. coli 67,000 400 No protein a Concentrations of the enzymes as PNPase I, 10 ,ug per assay, and E. coli, 2.8 U per assay, as obtained from Miles Research Products.
explanation for the pH optimum of the polymerization reaction of ADP. The polymer is rapidly degraded at neutral or alkaline pH, but the phosphorolysis reaction is almost completely inhibited in cacodylate buffer at pH 5.0 to 4.5. Inhibitory studies on PNPase I. The effects of various inhibitors of PNPase I are summarized in Fig. 9. As expected, inorganic PO42markedly inhibits the reaction at levels above 5.0 nmol per assay. Rifamycin SV, a potent inhibitor of B. subtilis deoxyribonucleic acid (DNA)-dependent RNA polymerase, demonstrates partial inhibition of the reaction, but only at relatively high concentrations. Unexpectedly, heparin proved to be a very potent inhibitor of PNPase I. It is also of interest to note that both (A),, and (G),, inhibited the synthesis of (A),, by PNPase I. The reaction proved relatively insensitive to N-ethylmaleimide and high concentrations of KCI.
874
J. BACTERIOL.
ERICKSON & GROSCH
DISCUSSION The presence of PNPase in extracts of B. subtilis has been previously described (3, 21) but was not studied in detail. PNPase I of B. amyloliquefaciens, as described in the present investigation, is an apparently primer-independent form of the enzyme that can be purified to a single protein band upon SDS-polyacrylamide gel electrophoresis. The primer inde22.5 CL
2 0
~0 1.5 z
a 1.0
o
0
E c
0.5
L
0
30
0
60
120
Minutes Incubation
FIG. 7. Polymerization as a function of incubation time. Enzyme was added at a level of 1.0 pg per assay. Symbols: (a) (A)n synthesis, (0) (G)n synthesis.
pendence of our preparation must be qualified, since we have not rigorously proved the absence of nucleic acid contamination, and very small amounts of nucleic acid tightly bound to the enzyme could escape detection by conventional means. The immediate initiation of (A), and (G)" synthesis at the maximal rate might be indicative of the existence of such a contaminant (7). The enzyme possesses many of the properties and characteristics of the well-studied PNPase of Micrococcus luteus (5, 7, 10, 11, 21). PNPase I can carry out the polymerization, phosphorolysis, and inorganic [32P]ADP exchange reactions that are characteristic of this class of enzyme. The divalent cation requirement, as well as the inhibition by high concentration of the cations, is comparable to that described for the other enzymes (7, 21). The major difference in the reaction conditions utilized in studies with B. amyloliquefaciens PNPase I is the relatively low pH optimum. Although the pH optimum is dependent upon reaction conditions, the value is usually reported to be between a pH of 8 and 9 for other enzymes. In addition, the molecular weight value of 7.4 x 104 for PNPase I is significantly smaller than the 2.5 x 105 value reported for the M. luteus enzyme (10). The existence of two separable PNPase activities is not unique to B. amyloliquefaciens. The PNPase of M. luteus is purified as a primerindependent form which, upon limited tryptic hydrolysis, yields a slightly smaller primerdependent form of the enzyme (10). B. amyloliquefaciens produces large quantities of extracellular proteases, and these enzymes may act
A
100
B
so 01
3)
c
-C
