- Email: [email protected]
Comparison of initiating abilities of primers of different length in polymerization reactions catalyzed by DNA polymerases from thermoacidophilic archaebacteria
Biochimica et Biophysica Acta, 1008 (1989) 102-107 Elsevier
102
BBA 91946
Comparison of initiating abilities of primers of different length in polymerization reactions catalyzed by DNA polymerases front thermoacidophilic archaebacteria Ivan Sh. Bukhrashvili 1, David Z. C h i n c h a l a d z e 1, O l g a I. L a v r i k 2, Asja S. Levina 2, G e o r g e A. N e v i n s k y 2 a n d D a v i d A. Prangishvili 1 t Institute of Molecular Biology and Biological Physics, Georgian SSR Academy of Sciences, Tbilisi and 2 Novosibirsk Institute of Bioorganic Chemistry, Siberian Branch of the USSR Academy of Sciences, Novosibirsk (U.S.S.R.)
(Received 9 January 1989)
Key words: Pro,~ein-DNA interaction; DNA polymerization; DNA polymerase; Primer; (Thermoacidophilic archaebacteria)
Optimal conditions for polymerization reaction catalyzed on poly(dA) and poly(dT) templates by DNA polymerases from thermoacidophilic archaebacteria - DNA polymerase A from Sulfolobus acidocaldarius and DNA polymerase B from Thermoplasma acidophilum - have been established. Values of K m and Vmax (60. C) for a set of primers d(pA)n and d(pT)n have been estimated. Minimal primers for both enzymes are dNMP. Lengthening of primers by each mononucleotide increases their affinity about 2.16-fold. Linear dependence of log K m and of |Og Vmax on the number of mononucleotide links in primers (n) has breaking point at n--10. The value of Vma~ is about 20% of that for decanucleotide. The affinity of the primer d(pA)gp(rib* ) with a deoxyribosylurea residue at the 3'-end does not differ essentially from that of d(pA)9. Substitution of the 3'-terminal nucleotide of a complementary primer for a noncomplementary nucleotide, e.g., substitution of 3'-terminal A for C in d(pA)10 in the reaction catalyzed on poly(dT), decreases the affinity of a primer by one order of magnitude.
Introduction
Archaebacteria constitute a group of microorganisms - most living in extreme habitats - which has been
classified as a third primary kingdom besides those of true bacteria (eubacteria) and eukaryotes [1]. In many respects, cellular components of archaebacteria show unique features, whereas in others they resemble eubacteria or eukaryotes [2]. Recently, a number of reports have appeared describing DNA polymerases of archaebacteria. It seems that, similar to eubacteria and eukaryotic cells, archaebacteria contain several DNA polymerases differing from each other by physicochemical and biochemical criteria. Tentatively, the enzymes can be grouped into three types. One type of archaebacterial DNA polymerase is reminiscent of eukaryotic DNA polymerase a in being a multipolypeptide complex and sensitive to aphidicolin and N-ethylmaleimide; such enzymes were purified from Halobacterium halobium [3], MethanococCorrespondence: D.A. Prangishvili, Institute of Molecular Biology and Biological Physics, Georgian Academy of Sciences, L. Gotua Street 14, 380060 Tbilisi, U.S.S.R.
cus vannielii [4], Suifolobus acidocaldarius [5] and Sulfolobus solfataricus [6]. The second type of enzyme, with M r < 50000 - insensitive to aphidicolin and N-ethylmaleimide and inhibited by dideoxynucleotide triphosphates - resembles to some extent DNA polymerase fl of eukaryotic cells; enzymes of this type were purified from H. halobium [3], Thermoplasma acidophilure [7], and S. acidocaldarius [5,8]. Archaebacterial DNA polymerases of the third type were purified from S. acidocaldarius [5,9] and Methanobacterium thermoautotrophicum [10]; they show certain similarities to eubacterial DNA polymerase I: exonuclease activities have been found associated with the enzymes, a polypeptide of Mr > 100000 constitutes a tingle subunit of the enzymes. All three types of enzyme have been found in S. acidocaldarius and designated DNA polymerases C, B and A corresponding to aforesaid succession; immunochemical ~nalysis showed that low-molecularweight DNA polymerase B is not a product of proteolytic degradation of DNA polymerases A and C [5]. Further studies are required to establish the relationships of archaebacterial DNA polymerases to their eubacterial and eukaryotic counterparts. A detailed investigation of archaebacterial DNA polymerases will
0167-4781/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
103 doubtless help to elucidate the evolution of replication systems. Here we present some characteristics of the interaction of primers with archaebacterial DNA polymerases - DNA polymerase A from S. acidocaldarius (PA) and DNA polymerase B from T. acidophilum (PB). An analogous study has been recently carried out with Escherichia coil DNA polymerase I and DNA polymerase a from human placenta [11-16]. Both archaebacterial enzymes, polymerases A and B, being isolated from thermophilic species, had maximal activity at 60 ° C and revealed not only the kingdom-specific characteristics of DNA polymerases, but also the effect of elevated temperatures on the interactions of the enzymes with primers. Materials and Methods
Enzymes DNA polymerase A from S. acidocaldarius and DNA polymerase B from T. acidophilum were isolated as described in Refs. 5, 7. One unit of enzyme activity was defined as the amount of protein catalyzing the incorporation into activated DNA of 1 nmol of [3H]dNMP in 60 min at 60 ° C. Substrates Calf thymus DNA was obtained from Serva, poly(dA)- poly(dT) (1 : 1) from Boehringer-Mannheim, poly(dA), poly(dT), dATP and dTTP from Niktibav, U.S.S.R. [3H]dATP, [3H]dTTP, (spec. act., 0.75-2.5 Ci/mmoi) and [all]dAMP (20 Ci/mmol) (Izotop, U.S.S.R.) were purified by thin-layer chromatography on Kieselgel-60 1'254 (Merk) in a dioxane/ammonium hydroxide/water (6 : 4:1) system and were homogeneous as indicated by ion-exchange and reversed-phase chromatography conducted as in Ref. 11. Poly(dT) consisted of 100-700 nucleotides, and poly(dA) of 70-150 nucleotides. Template concentration (base concentration) was estimated using the following molar absorption coefficients for mononucleotides in the compositk,n of polynucleotide chains - 8.43- 103 (268 nm) and 9.1.103 (257 nm). The synthesis and characterization of oligoadenylates and oligothymidilates used, as well as of d(pA)9pC and d(pA)9p(rib* ), a decanucleotide with the terminal base substituted for the urea residue, were as described earlier [11]. All oligonucleotides were homogeneous as revealed by reversed-phase and ion-exchange chromatography. We used the following molar absorption coefficient to determine the oligonucleotide concentration: for d(pT)7, 60.5. 103; d(PT) s, 69- 103; d(pT)~ 0, 87-103 (267 rim) [11,12]; dAMP, 15.3-103; d(pA)2, 24.5- 103; d(pA)4, 43.5. 103; d(pA)6, 62. 103; d(pA), 71- 103; d(pA) 8, 80- 103; d(pA)~0, 98.7- 103; d(pA)l 2, 117.2- 103, d(pA)16,154.5.103 [13]; d(pA)9pC, 98.5- 103; d(pA)9p(rib*), 89.5.103 (260 nm) [14].
DNA polymerase assay The polymerization reaction was carried out at 60 ° C. The assay mixture for polymerases A (50-100 /~l) contained 20 mM Tfis-HCl (pH 7.2)/0.2 mM MnCl 2/75 mM KCI/0.2 mM [3H]dATk or 0.5 mM [3H]dTTP, poly(dA)-poly(dT) 2 A260/ml or poly(dT)-d(pA),, or poly(dA) - d(pT),,; concentration of primers d(pA),, and d(pT),, varied over a wide range and that of templates (base concentration) was 0.18 mM for poly(dT) and 0.44 mM for poly(dA). The assay mixture for polymerases B (50-100 /zl) contained 20 mM Tris-HCl (pH 7.2), 0.2 mM MnCl 2, 75 mM KCI, 0.1 mM [3HldATP or 0.2 mM [3H]dTTP, 0.35 A26o/ml poly(dA)-poly(dT) or 0.12 mM (base concentration) poly(dT) or 0.15 mM (base concentration) poly(dA) and complementary primers d(pA),, or d(pT),,, respectively. The polymerization reaction was initiated by addition of DNA polymerase. The initial rates of polymerization increased linearly with an increase of enzyme concentration to 150 units/ml of polymerase A or B in the case of poly(dT) and to 500 units/ml of these enzymes in the case of poly(dA). In the standard assay mixture 20-80 units/ml of DNA polymerase were present. As demonstrated earlier, estimation of acid-preciipitable material formed does not allow accurate detection of dNMP incorporation into primers shorter than 6 nucleotides; the efficiency of precipitation by trichloracetic acid of oligomers containing 9, 8, 7, 6, or 5 nucleotide units is about 95, 89, 82, 65 and 3~ of that of decanucleotide, respectively [13,15]. Due to this, [3H]dNMP incorporation was detected by the usual procedure only in the case of primers longer than 6 nucleotides: acid-insoluble material was collected by filtration on Whatman GF/C glass filters (diameter 2 cm) and washed (eight times) by 5 ml of cold 5~ trichloracetic acid containing 0.2 M KH2PO4; actual incorporation of [3H]dNMP was estimated using the above coefficients of precipitation efficiency. If the primer used was shorter than 6 nucleotides the following procedure was followed. The polymerization reaction was stopped by addition of assay mixture aliquots (10-50/zl) to 0.1 ml of 0.1 M EDTA. The mixture was applied to sandwich filters (diameter 2 cm) consisting of two Whatman G F / C glass filters with a layer of Lichrosor RP-18 (Merck) between them, and the filters were washed (15 times) by 5 ml of 3-3.5~ aqueous methanol, containing 20 mM Tris-HCi (pH 7.2). Oligomers longer than two nucleotides remained adsorbed on filters. In the case of d(pN)5 t,nd d(pN)lo primers, [3H]dNMP incorporation was tested by both procedures. The feasibility of using dNMP as a primer for polymerases A and B was studied using [3H]dAMP and unlabeled dATP. Aliquots of standard assay (50 /~l)
104
were applied to a column (2 x 60 mm) with the ion-exchanger Aminosilochrome AC-300 (Novosibirsk University, U.S.S.R.) equilibrated with water. An excess of [3H]dAMP was eluted (100 p l / m i n ) by 3 mM KH2PO4 (pH 7.5), and oligoadenylates longer than two monomers by I m KH2PO4. The eluates were applied to filters, the filters were dried and radioactivity was counted.
lo
t_
E
Kinetic constants Km and Vma,, values were determined by constructing direct linear plots as described in Ref. 17. An error was estimated according to the following equation:
I 0.5
! 1.0
1.5
2.0
i 2.5
Poly ( d A ) . poly(dT ) c o n c e n t r a t ion( A26o1 ml ) n
Fig. 1. Plots of the initial rates of polymerization reactions catalyzed by polymerase A (1) ( x ) and polymerase B (2) (o) against the concentration cf a poly(dA)-poly(dT) template-primer complex (substrate: [3H]dATP).
X i -- ~
zl(~) =
.lOO$
.
j-I
where d is an error; terms x~ are values for Km and Vmax determined using an intersection point of two lines; n, the number of intersection points. The values of Vma~ are in cpm/min and are reduced to the common specific activity of [3H]dNTP. The values of Vma~ for d(pT)m and d(pA)m were taken to be 100~ and Vma~ values for other primers - d(pA)n and d(pT)n - were expressed as fractions thereof, respectively. The error of K m and Vma~ was in the range of 10-40~. Results
Maximal rates of polymerization reaction catalyzed on poly(dA) • poly(dT) by both archaebacterial enzymes studied, polymerases A and B, were found to occur at 60 o C, pH 7.2, in the presence of 0.2 mM MnCI2 and 75 mM KCI. These conditions were used in all experiments conducted. Fig. 1 shows the dependence of initial rates of the polymerization reaction on the concentration of the poly(dA), poly(dT) template-primer complex. For
polymerases A and B the picture is different. The polymerization catalyzed by polymerase A is not inhibited by a concentration of the template-primer complex 8-t~mes in excess of the K m value (0.25 A2eo/ml), while concentrations of the complex even slightly exceeding (1.5-times) the K m value (0.2 A2~o/ml) significantly decrease the reaction rate in the case of polymerase B. The inhibitory effect of dNTPs is also more pronounced in the case of PB. As an example, Fig. 2 shows the dependence of the initial rates of polymerization catalyzed by polymerases A and B on dATP concentration. The affinity of dATP for polymerase A ( K m -- 10 ItM) is about 14-times that for polymerase B (K m = 140 aM). To study the kinetic parameters of oligonucleotide primers in DNA polymerase reactions it was necessary to determine the concentrations of all other substrates that saturated without an inhibitory effect. Table I summarizes the data obtained and indicates the concentrations of dATP, dTrP, poly(dA) and poly(dT) used in further studies. The minimal primer for polymerases A and B on poly(dT) template proved to be dAMP. A direct linear
TABLE I K~, for templates and dNTP in polymerization reactions catalyzed by polymerases A and B, and maximal concentrations of these substrates not having an inhibitory effect on reactions (Cmax) Substtate
Poly(dA) a Poly(dT) b dATP b dTTP a
Polymerase A
Polymerase B
Km
Cmax
concentration in standard assay
Km
Cmax
concentration in standard assay
0.5 A26o/ml 0.16 A2eo/ml 0.14 mM 0.15 mM
_ c _ c 0.2 mM 0.5 mM
4 A26o/lld 1.5 A26o/ml 0.2 mM 0.5 mM
0.2 A26o/rld 0.05 A2eo/ml 10 ltM 25/tM
_ c _ c 0.1 mM 0.2 mM
3 A26o/ml 1 A2~/ml 0.1 mM 0.2 mM
i The primer was d(pT)]o. b The primer was d(pA)l0. c Reaction is not inhibited by the substrate in concentrations up to 10-20 Km. The error of K m estimation was within 20-40~.
105 :TABLE II
o 100t
1
K m for d(pA), and d(pT), primers in polymeri=ation reactions catalyzed by polymerases A and B on poly(dT) and polycdA) templates, respectively.
e
The error of K m estimation was within 10-40~.
~ 6o
Primers
~ 4o
E
I
!
1
2
[3H]dATP concentration(Ml(xl0 4) Fig. 2. Plots of the initial rats of polymerization reactions catalyzed by po,lymerase A (1) ( × ) and polymerase B (2) (e) against [3H]dATP concentration. t--
d(pA) d(pA) 2 d(pA)4 d(pA) 6 d(pA) 7 d(pA) s d(pA)gp(rib* ) d(pA)gpC d(pA)lo d(pA)!2 d(pA) 16
polymerase A
polymerase B
240 120 19 2.0 1.5 0.6 0.44 !.4 0.2 0.43 1.8
160 80 14 3.2 1.6 0.94 0.33 1.6 0.15 0.21 1.4
d(pT)~ d(pT) s d(pT) lo
o
co
~
3c
-6g
2c
,
5
g m (/gM)
4
3 2 1 [3H] dAM P concentrotlon( M )( xl0 4)
Fig. 3. Direct linear plot, constructed according to Eisenthal and Cornish-Bowden [16], of the initial rates of [ 3H]dAMP-primer elongation by polymerase B against primer concentration.
p|ot of dAMP elongation by polymerase A against dAMP concentration, constructed according to Eisenthal and Cornish-Bowden [17] is presented in Fig. 3. ¢-
£
0.3 0.15 0.05
0.33 0.11 0.035
According to the graphical procedure of Eisenthal and Comish-Bowden [16], we estimated the values of K m and Vma, f~r set of primers d(pA). (n = 1, 2, 4, 6, 7. 8, 9, 10, 12, 16) and d(pT). (n = 7, 8, 10) as well as for d(pA)opC and d(pA)oP(rib* ) in polymerization reactions catalyzed by polymerase A or B on templates complementary to these primers, poly(dT) and poly(dA), respectively. In each case, a direct linear plot was constructed. As an example, Fig. 4 presents such a plot of the initial elongation rates of d(pA)6 primer by polymerase A against primer concentration. The K m values obtained are presented in Table II. Fig. 5 shows a plot of log Km against the length (n) of oligoadeny|ate p,-~ers, it is linear in the intervals 1 < n < 10 and 10 < n < 16, having the breaking point at n = 10. The lengthening of a primer by one mononucleotide unit (n - 1-10) in the case of both enzymes results in about 2.16-fold increase of oligoadenylate primer affinity, corresponding to Gibbs' energy change,
u~
3
•
"-g
//~2
~E
f
e
6
"7 5 lO
5
[cl(pA)6]concentrat ,or' ( M ) ( x l O 6 ) Fig. 4. Direct linear plot, constructed according to Eisenthal and Comish-Bowden [16], of the initial rates of d(pA)6 primer elongation by PA against primer concentration.
4 3
/ I
4
13
12
16
2'0
Fig. 5. Plots of log Km against primer length, n: @, polymerase A; x, polymerase B.
106 R T . In K m. The dependence of AG O for d(pA) upon the x 2.0
Y '
1.5
1.0
4
8
I'~
16
20 n
Fig. 6. Plots of log Vmax against primer length, n: o, polymerase A; ®, polymerase B.
AG ffi -0.51 kcal/mol. The similar increase in affinity
is observed for d(pT), primers (Table II). A comparison of maximal rates o f d(pA)m and d(pT)lo elongation shows that the v~lue of Vma,, for the latter primer is about 1~ of that of the former one. The plot of log Vmax against n, similarly to that of log K m, is linear, with an inflection point at n - 10 (Fig. 6). The Vm~,, value for dAMP constitutes about 20~ of the value for d(pA)l o. Discussion
The melting point of the complexes of dNTP with the complementary polynucleotide is less than 0 °C [13]. However, it has been shown that dNMP and NMP are minimal primers for DNA polymerase a from human placenta and E. coil DNA polymerase I, the enzymes having maximal activity at 30-40 °C [13-16]. This may be due to the fact that the stability of template-primer duplex bound to the enzyme is determined mainly not by complementary interactions but by the interaction with the enzyme of a template and the 3'-terminal nucleotide of a primer. The phosphate group of the latter forms with the enzyme an Me2+-dependent electrostatic contact ( A G - - 1 . 2 kcal/mol) and its 3'-OH group, as well as the oxygen atom of ~ group form with the enzyme hydrog,~n bonds ( A G - - 5 . 6 kcal/mol); all other nucleotides of a primer interact only with a template, forming Watson-Crick hydrogen bonds [11,13,15,16]. The increase of optimal reaction temperatures up to 60°C in the case of DNA polymerases from thermophilic archaebacteria could have necessitated additional contacts between a primer and the enzyme, or an elongation of a minimal primer. However, as has been demonstrated (Fig. 3, Table II), dAMP is a minimal primer also for thermophilic archaebacterial DNA polymerases, and in the case of these enzymes, the K m v a l u e s (0.2 / t M ) for such a primer (estimated at 60 o C) exceed only 2-5-times those for mesophilic eubacterial or eukaryotic DNA polymerases (estimated at 30 ° C). Assuming - according to Refs. 12-14 - that K m for primers are close to Kd, Gibbs' energy values, G °, of the complex formation were calculated as equal to
number of nucleotide units (n) is linear up to n - 10. Each primer unit enhances the affinity by a factor of 2.16 (AG O= -0.51 kcal/mol). The result may be described by a common equation: AG = -6.54-0.51(n - 1 ) (kcal/mol), where -6.54 kcai/mol is Gibbs' energy of dAMP. The data of Table II show that the oligonucleotide d(pA)9p(rib* ) with a urea residue in place of the terminal base, has nearly the same affinities for polymerases A and B as d(pA)9; this suggests that the contribution of the 3'-terminal base to the interaction of a primer with an enzyme-template complex does not differ from that of the other bases of the primer. The data available indicate a specific strong interaction of the 3'-terminal nucleotide residue of a primer with an enzyme-template complex ( A G = - 6 . 0 kcal/mol); interactions of the other nucleotide units of a primer are much weaker (2-I03-times, AG ffi -0.51 kcal/mol). It is reasonable to suggest that the latter interactions are determined by base-pairing between primer units and the template and that archaebacterial DNA polymerases studied interact with only the 3'-terminal nucleotide of a primer, as in the case of DNA polymerase from human placenta and E. coil DNA polymerase I [12-14]. With respect to the mode of interaction with primers, polymerases A and B also bear some other resemblances with eubacterial and eukaryotic DNA polymerases. The plots of log K m and log Vma,, against n are linear with the breaking point at n - lP in case of all enzymes. This may mean that only 10 mononucleotide units of a primer are present in the range of the protein globule of the DNA polymerase. Interactions between template and primer chains outside the enzyme molecule can change the character of the dependence of K m and Vmax o n n [14,15]. In the case of DNA polymerase a from human placenta, it has been shown that substitution of the 3'-terminal base of a primer for a noncomplementary one results in a significant decrease of primer affinity: the value of K m for d(pT)10pC is 39-times that for d(pT) n. As seen from the data of Table II, values of K~ for d(pA)9pC are 7- and ll-times those for d(pA)lo in the case of polymerase A and polymerase B, respectively. The maximal rates of dAMP primer elongation catalyzed by polymerases A and B constitute about 20~ of that of the d(pA)lo primer, and lengthening of a primer by each nucleotide (n ffi 1-10) increases the value of Vma~ 1.2-times: V~ffi 20~V10(1.2) "-1. A similar dependence is observed for DNA polymerase a from human placenta: V~ffi 7%V1o(1.36)"-1 [13,15]. Table II shows that the affinity of d(pT), primers for polymerases A and B is higher than that of d(pA), primers of the same length. The same was observed for eukaryotic DNA polymerase a [13]. It must be noted
107
that in the case of archaebactefial enzymes the maximal rate of d(pA)l o elongation is about 50-100-times higher than that of d(pT)~0 primer. The results obtained demonstrate that the mode of interaction with primers of thennophilic archaebacter/al DlqA polymerases does not differ much from that of mesophilic eubacterial or eukaryotic DNA polymerases. References 1 Woese, C.R., Magrum, L.J. and Fox, G.E. (1978) J. Mol. Evol. 11, 245-252. 2 Woese, C.R. and Olsen, (3. (1986) System. AppL Microbioi. 7, 161-177. 3 Kohiyarna, M., Nakayama, M. and Ben Mahrez, K. (1986) System. Appl. Microbiol. 7, 79-82. 4 Zabel, H.-P., Fischer, H., Holler, E. and Winter, J. (1985) System. Appl. Microbiol. 6, 111-118. 5 Prangishvili, D.A. (1986) Mol. Biol. USSR 20, 477-488. 6 Rossi, M., Rella, R., Pensa, M., Battolucci, S., De Rosa, M., Gambacorta, A., Raia, C.A. and DelrAversano Orabona, N. (1986) System. Appl. Microbiol. 7, 337-341.
7 Chinchaladze, D.Z., PrangJshvili, D.A., Kachabava, L.A. and Zaafishvili, M.M. (1985) Mol. Biol. USSR 19, 1466-1475. g Forterre, P., Nadal, M., Elie, C., Mirambeau, G., Jaxel, C. and Duguet, M. (1986) System. App|. Microbiol. 7, 67-71. 9 I¢~mczak, L.J., Grummt, F. and Burger, K.J (1985) Nucleic Acids Res. 13, 5269-5282. 10 Klimczak, L.J., Grummt, F. and Burger, K.J. (1986) Biochemistry 25, 4850-4855. 11 Levina, A.S., Nevinsky, G.A. and Lawik, O.l. (1985) Bioorg. Chem. USSR, 11, 358-362. 12 Nevinsky, G.A., Frolova, E.I., Podust, V.N., Levina, A.S., Lavrik, O.I. and Khalabuda, O.V. (1986) Bioorg. Chem. USSR, 12, 357-368. 13 Venjaminova, A.G., Levina, A.S., Nevinsky, G.A. and Podust, V.N. (1987) Mol. Biol. USSR 21, 1378-1385. 14 Nevinsky, G.A., Levina, A.S., Frolova, E.I. and Podust, V.N. (1987) Mol. Biol. USSR 21, 1193-1200. 15 Nevinsky, G.A., Frolova, E.I., Levina, A.S., Podust, V.N. and Lebedev, A.V. (1987) Bioorg. Chem. USSR 13, 45-57. 16 Knorr¢, D.G., Lavrik, O.I. and Nevinsky, G.A. (1988) Biochimie 70, 655-661. 17 Eisenthai, R. and Cornish-i~iowdcn, ,,~. [l~J4j bi~;i~m..L IJ.,, 715-720.
102
BBA 91946
Comparison of initiating abilities of primers of different length in polymerization reactions catalyzed by DNA polymerases front thermoacidophilic archaebacteria Ivan Sh. Bukhrashvili 1, David Z. C h i n c h a l a d z e 1, O l g a I. L a v r i k 2, Asja S. Levina 2, G e o r g e A. N e v i n s k y 2 a n d D a v i d A. Prangishvili 1 t Institute of Molecular Biology and Biological Physics, Georgian SSR Academy of Sciences, Tbilisi and 2 Novosibirsk Institute of Bioorganic Chemistry, Siberian Branch of the USSR Academy of Sciences, Novosibirsk (U.S.S.R.)
(Received 9 January 1989)
Key words: Pro,~ein-DNA interaction; DNA polymerization; DNA polymerase; Primer; (Thermoacidophilic archaebacteria)
Optimal conditions for polymerization reaction catalyzed on poly(dA) and poly(dT) templates by DNA polymerases from thermoacidophilic archaebacteria - DNA polymerase A from Sulfolobus acidocaldarius and DNA polymerase B from Thermoplasma acidophilum - have been established. Values of K m and Vmax (60. C) for a set of primers d(pA)n and d(pT)n have been estimated. Minimal primers for both enzymes are dNMP. Lengthening of primers by each mononucleotide increases their affinity about 2.16-fold. Linear dependence of log K m and of |Og Vmax on the number of mononucleotide links in primers (n) has breaking point at n--10. The value of Vma~ is about 20% of that for decanucleotide. The affinity of the primer d(pA)gp(rib* ) with a deoxyribosylurea residue at the 3'-end does not differ essentially from that of d(pA)9. Substitution of the 3'-terminal nucleotide of a complementary primer for a noncomplementary nucleotide, e.g., substitution of 3'-terminal A for C in d(pA)10 in the reaction catalyzed on poly(dT), decreases the affinity of a primer by one order of magnitude.
Introduction
Archaebacteria constitute a group of microorganisms - most living in extreme habitats - which has been
classified as a third primary kingdom besides those of true bacteria (eubacteria) and eukaryotes [1]. In many respects, cellular components of archaebacteria show unique features, whereas in others they resemble eubacteria or eukaryotes [2]. Recently, a number of reports have appeared describing DNA polymerases of archaebacteria. It seems that, similar to eubacteria and eukaryotic cells, archaebacteria contain several DNA polymerases differing from each other by physicochemical and biochemical criteria. Tentatively, the enzymes can be grouped into three types. One type of archaebacterial DNA polymerase is reminiscent of eukaryotic DNA polymerase a in being a multipolypeptide complex and sensitive to aphidicolin and N-ethylmaleimide; such enzymes were purified from Halobacterium halobium [3], MethanococCorrespondence: D.A. Prangishvili, Institute of Molecular Biology and Biological Physics, Georgian Academy of Sciences, L. Gotua Street 14, 380060 Tbilisi, U.S.S.R.
cus vannielii [4], Suifolobus acidocaldarius [5] and Sulfolobus solfataricus [6]. The second type of enzyme, with M r < 50000 - insensitive to aphidicolin and N-ethylmaleimide and inhibited by dideoxynucleotide triphosphates - resembles to some extent DNA polymerase fl of eukaryotic cells; enzymes of this type were purified from H. halobium [3], Thermoplasma acidophilure [7], and S. acidocaldarius [5,8]. Archaebacterial DNA polymerases of the third type were purified from S. acidocaldarius [5,9] and Methanobacterium thermoautotrophicum [10]; they show certain similarities to eubacterial DNA polymerase I: exonuclease activities have been found associated with the enzymes, a polypeptide of Mr > 100000 constitutes a tingle subunit of the enzymes. All three types of enzyme have been found in S. acidocaldarius and designated DNA polymerases C, B and A corresponding to aforesaid succession; immunochemical ~nalysis showed that low-molecularweight DNA polymerase B is not a product of proteolytic degradation of DNA polymerases A and C [5]. Further studies are required to establish the relationships of archaebacterial DNA polymerases to their eubacterial and eukaryotic counterparts. A detailed investigation of archaebacterial DNA polymerases will
0167-4781/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)
103 doubtless help to elucidate the evolution of replication systems. Here we present some characteristics of the interaction of primers with archaebacterial DNA polymerases - DNA polymerase A from S. acidocaldarius (PA) and DNA polymerase B from T. acidophilum (PB). An analogous study has been recently carried out with Escherichia coil DNA polymerase I and DNA polymerase a from human placenta [11-16]. Both archaebacterial enzymes, polymerases A and B, being isolated from thermophilic species, had maximal activity at 60 ° C and revealed not only the kingdom-specific characteristics of DNA polymerases, but also the effect of elevated temperatures on the interactions of the enzymes with primers. Materials and Methods
Enzymes DNA polymerase A from S. acidocaldarius and DNA polymerase B from T. acidophilum were isolated as described in Refs. 5, 7. One unit of enzyme activity was defined as the amount of protein catalyzing the incorporation into activated DNA of 1 nmol of [3H]dNMP in 60 min at 60 ° C. Substrates Calf thymus DNA was obtained from Serva, poly(dA)- poly(dT) (1 : 1) from Boehringer-Mannheim, poly(dA), poly(dT), dATP and dTTP from Niktibav, U.S.S.R. [3H]dATP, [3H]dTTP, (spec. act., 0.75-2.5 Ci/mmoi) and [all]dAMP (20 Ci/mmol) (Izotop, U.S.S.R.) were purified by thin-layer chromatography on Kieselgel-60 1'254 (Merk) in a dioxane/ammonium hydroxide/water (6 : 4:1) system and were homogeneous as indicated by ion-exchange and reversed-phase chromatography conducted as in Ref. 11. Poly(dT) consisted of 100-700 nucleotides, and poly(dA) of 70-150 nucleotides. Template concentration (base concentration) was estimated using the following molar absorption coefficients for mononucleotides in the compositk,n of polynucleotide chains - 8.43- 103 (268 nm) and 9.1.103 (257 nm). The synthesis and characterization of oligoadenylates and oligothymidilates used, as well as of d(pA)9pC and d(pA)9p(rib* ), a decanucleotide with the terminal base substituted for the urea residue, were as described earlier [11]. All oligonucleotides were homogeneous as revealed by reversed-phase and ion-exchange chromatography. We used the following molar absorption coefficient to determine the oligonucleotide concentration: for d(pT)7, 60.5. 103; d(PT) s, 69- 103; d(pT)~ 0, 87-103 (267 rim) [11,12]; dAMP, 15.3-103; d(pA)2, 24.5- 103; d(pA)4, 43.5. 103; d(pA)6, 62. 103; d(pA), 71- 103; d(pA) 8, 80- 103; d(pA)~0, 98.7- 103; d(pA)l 2, 117.2- 103, d(pA)16,154.5.103 [13]; d(pA)9pC, 98.5- 103; d(pA)9p(rib*), 89.5.103 (260 nm) [14].
DNA polymerase assay The polymerization reaction was carried out at 60 ° C. The assay mixture for polymerases A (50-100 /~l) contained 20 mM Tfis-HCl (pH 7.2)/0.2 mM MnCl 2/75 mM KCI/0.2 mM [3H]dATk or 0.5 mM [3H]dTTP, poly(dA)-poly(dT) 2 A260/ml or poly(dT)-d(pA),, or poly(dA) - d(pT),,; concentration of primers d(pA),, and d(pT),, varied over a wide range and that of templates (base concentration) was 0.18 mM for poly(dT) and 0.44 mM for poly(dA). The assay mixture for polymerases B (50-100 /zl) contained 20 mM Tris-HCl (pH 7.2), 0.2 mM MnCl 2, 75 mM KCI, 0.1 mM [3HldATP or 0.2 mM [3H]dTTP, 0.35 A26o/ml poly(dA)-poly(dT) or 0.12 mM (base concentration) poly(dT) or 0.15 mM (base concentration) poly(dA) and complementary primers d(pA),, or d(pT),,, respectively. The polymerization reaction was initiated by addition of DNA polymerase. The initial rates of polymerization increased linearly with an increase of enzyme concentration to 150 units/ml of polymerase A or B in the case of poly(dT) and to 500 units/ml of these enzymes in the case of poly(dA). In the standard assay mixture 20-80 units/ml of DNA polymerase were present. As demonstrated earlier, estimation of acid-preciipitable material formed does not allow accurate detection of dNMP incorporation into primers shorter than 6 nucleotides; the efficiency of precipitation by trichloracetic acid of oligomers containing 9, 8, 7, 6, or 5 nucleotide units is about 95, 89, 82, 65 and 3~ of that of decanucleotide, respectively [13,15]. Due to this, [3H]dNMP incorporation was detected by the usual procedure only in the case of primers longer than 6 nucleotides: acid-insoluble material was collected by filtration on Whatman GF/C glass filters (diameter 2 cm) and washed (eight times) by 5 ml of cold 5~ trichloracetic acid containing 0.2 M KH2PO4; actual incorporation of [3H]dNMP was estimated using the above coefficients of precipitation efficiency. If the primer used was shorter than 6 nucleotides the following procedure was followed. The polymerization reaction was stopped by addition of assay mixture aliquots (10-50/zl) to 0.1 ml of 0.1 M EDTA. The mixture was applied to sandwich filters (diameter 2 cm) consisting of two Whatman G F / C glass filters with a layer of Lichrosor RP-18 (Merck) between them, and the filters were washed (15 times) by 5 ml of 3-3.5~ aqueous methanol, containing 20 mM Tris-HCi (pH 7.2). Oligomers longer than two nucleotides remained adsorbed on filters. In the case of d(pN)5 t,nd d(pN)lo primers, [3H]dNMP incorporation was tested by both procedures. The feasibility of using dNMP as a primer for polymerases A and B was studied using [3H]dAMP and unlabeled dATP. Aliquots of standard assay (50 /~l)
104
were applied to a column (2 x 60 mm) with the ion-exchanger Aminosilochrome AC-300 (Novosibirsk University, U.S.S.R.) equilibrated with water. An excess of [3H]dAMP was eluted (100 p l / m i n ) by 3 mM KH2PO4 (pH 7.5), and oligoadenylates longer than two monomers by I m KH2PO4. The eluates were applied to filters, the filters were dried and radioactivity was counted.
lo
t_
E
Kinetic constants Km and Vma,, values were determined by constructing direct linear plots as described in Ref. 17. An error was estimated according to the following equation:
I 0.5
! 1.0
1.5
2.0
i 2.5
Poly ( d A ) . poly(dT ) c o n c e n t r a t ion( A26o1 ml ) n
Fig. 1. Plots of the initial rates of polymerization reactions catalyzed by polymerase A (1) ( x ) and polymerase B (2) (o) against the concentration cf a poly(dA)-poly(dT) template-primer complex (substrate: [3H]dATP).
X i -- ~
zl(~) =
.lOO$
.
j-I
where d is an error; terms x~ are values for Km and Vmax determined using an intersection point of two lines; n, the number of intersection points. The values of Vma~ are in cpm/min and are reduced to the common specific activity of [3H]dNTP. The values of Vma~ for d(pT)m and d(pA)m were taken to be 100~ and Vma~ values for other primers - d(pA)n and d(pT)n - were expressed as fractions thereof, respectively. The error of K m and Vma~ was in the range of 10-40~. Results
Maximal rates of polymerization reaction catalyzed on poly(dA) • poly(dT) by both archaebacterial enzymes studied, polymerases A and B, were found to occur at 60 o C, pH 7.2, in the presence of 0.2 mM MnCI2 and 75 mM KCI. These conditions were used in all experiments conducted. Fig. 1 shows the dependence of initial rates of the polymerization reaction on the concentration of the poly(dA), poly(dT) template-primer complex. For
polymerases A and B the picture is different. The polymerization catalyzed by polymerase A is not inhibited by a concentration of the template-primer complex 8-t~mes in excess of the K m value (0.25 A2eo/ml), while concentrations of the complex even slightly exceeding (1.5-times) the K m value (0.2 A2~o/ml) significantly decrease the reaction rate in the case of polymerase B. The inhibitory effect of dNTPs is also more pronounced in the case of PB. As an example, Fig. 2 shows the dependence of the initial rates of polymerization catalyzed by polymerases A and B on dATP concentration. The affinity of dATP for polymerase A ( K m -- 10 ItM) is about 14-times that for polymerase B (K m = 140 aM). To study the kinetic parameters of oligonucleotide primers in DNA polymerase reactions it was necessary to determine the concentrations of all other substrates that saturated without an inhibitory effect. Table I summarizes the data obtained and indicates the concentrations of dATP, dTrP, poly(dA) and poly(dT) used in further studies. The minimal primer for polymerases A and B on poly(dT) template proved to be dAMP. A direct linear
TABLE I K~, for templates and dNTP in polymerization reactions catalyzed by polymerases A and B, and maximal concentrations of these substrates not having an inhibitory effect on reactions (Cmax) Substtate
Poly(dA) a Poly(dT) b dATP b dTTP a
Polymerase A
Polymerase B
Km
Cmax
concentration in standard assay
Km
Cmax
concentration in standard assay
0.5 A26o/ml 0.16 A2eo/ml 0.14 mM 0.15 mM
_ c _ c 0.2 mM 0.5 mM
4 A26o/lld 1.5 A26o/ml 0.2 mM 0.5 mM
0.2 A26o/rld 0.05 A2eo/ml 10 ltM 25/tM
_ c _ c 0.1 mM 0.2 mM
3 A26o/ml 1 A2~/ml 0.1 mM 0.2 mM
i The primer was d(pT)]o. b The primer was d(pA)l0. c Reaction is not inhibited by the substrate in concentrations up to 10-20 Km. The error of K m estimation was within 20-40~.
105 :TABLE II
o 100t
1
K m for d(pA), and d(pT), primers in polymeri=ation reactions catalyzed by polymerases A and B on poly(dT) and polycdA) templates, respectively.
e
The error of K m estimation was within 10-40~.
~ 6o
Primers
~ 4o
E
I
!
1
2
[3H]dATP concentration(Ml(xl0 4) Fig. 2. Plots of the initial rats of polymerization reactions catalyzed by po,lymerase A (1) ( × ) and polymerase B (2) (e) against [3H]dATP concentration. t--
d(pA) d(pA) 2 d(pA)4 d(pA) 6 d(pA) 7 d(pA) s d(pA)gp(rib* ) d(pA)gpC d(pA)lo d(pA)!2 d(pA) 16
polymerase A
polymerase B
240 120 19 2.0 1.5 0.6 0.44 !.4 0.2 0.43 1.8
160 80 14 3.2 1.6 0.94 0.33 1.6 0.15 0.21 1.4
d(pT)~ d(pT) s d(pT) lo
o
co
~
3c
-6g
2c
,
5
g m (/gM)
4
3 2 1 [3H] dAM P concentrotlon( M )( xl0 4)
Fig. 3. Direct linear plot, constructed according to Eisenthal and Cornish-Bowden [16], of the initial rates of [ 3H]dAMP-primer elongation by polymerase B against primer concentration.
p|ot of dAMP elongation by polymerase A against dAMP concentration, constructed according to Eisenthal and Cornish-Bowden [17] is presented in Fig. 3. ¢-
£
0.3 0.15 0.05
0.33 0.11 0.035
According to the graphical procedure of Eisenthal and Comish-Bowden [16], we estimated the values of K m and Vma, f~r set of primers d(pA). (n = 1, 2, 4, 6, 7. 8, 9, 10, 12, 16) and d(pT). (n = 7, 8, 10) as well as for d(pA)opC and d(pA)oP(rib* ) in polymerization reactions catalyzed by polymerase A or B on templates complementary to these primers, poly(dT) and poly(dA), respectively. In each case, a direct linear plot was constructed. As an example, Fig. 4 presents such a plot of the initial elongation rates of d(pA)6 primer by polymerase A against primer concentration. The K m values obtained are presented in Table II. Fig. 5 shows a plot of log Km against the length (n) of oligoadeny|ate p,-~ers, it is linear in the intervals 1 < n < 10 and 10 < n < 16, having the breaking point at n = 10. The lengthening of a primer by one mononucleotide unit (n - 1-10) in the case of both enzymes results in about 2.16-fold increase of oligoadenylate primer affinity, corresponding to Gibbs' energy change,
u~
3
•
"-g
//~2
~E
f
e
6
"7 5 lO
5
[cl(pA)6]concentrat ,or' ( M ) ( x l O 6 ) Fig. 4. Direct linear plot, constructed according to Eisenthal and Comish-Bowden [16], of the initial rates of d(pA)6 primer elongation by PA against primer concentration.
4 3
/ I
4
13
12
16
2'0
Fig. 5. Plots of log Km against primer length, n: @, polymerase A; x, polymerase B.
106 R T . In K m. The dependence of AG O for d(pA) upon the x 2.0
Y '
1.5
1.0
4
8
I'~
16
20 n
Fig. 6. Plots of log Vmax against primer length, n: o, polymerase A; ®, polymerase B.
AG ffi -0.51 kcal/mol. The similar increase in affinity
is observed for d(pT), primers (Table II). A comparison of maximal rates o f d(pA)m and d(pT)lo elongation shows that the v~lue of Vma,, for the latter primer is about 1~ of that of the former one. The plot of log Vmax against n, similarly to that of log K m, is linear, with an inflection point at n - 10 (Fig. 6). The Vm~,, value for dAMP constitutes about 20~ of the value for d(pA)l o. Discussion
The melting point of the complexes of dNTP with the complementary polynucleotide is less than 0 °C [13]. However, it has been shown that dNMP and NMP are minimal primers for DNA polymerase a from human placenta and E. coil DNA polymerase I, the enzymes having maximal activity at 30-40 °C [13-16]. This may be due to the fact that the stability of template-primer duplex bound to the enzyme is determined mainly not by complementary interactions but by the interaction with the enzyme of a template and the 3'-terminal nucleotide of a primer. The phosphate group of the latter forms with the enzyme an Me2+-dependent electrostatic contact ( A G - - 1 . 2 kcal/mol) and its 3'-OH group, as well as the oxygen atom of ~ group form with the enzyme hydrog,~n bonds ( A G - - 5 . 6 kcal/mol); all other nucleotides of a primer interact only with a template, forming Watson-Crick hydrogen bonds [11,13,15,16]. The increase of optimal reaction temperatures up to 60°C in the case of DNA polymerases from thermophilic archaebacteria could have necessitated additional contacts between a primer and the enzyme, or an elongation of a minimal primer. However, as has been demonstrated (Fig. 3, Table II), dAMP is a minimal primer also for thermophilic archaebacterial DNA polymerases, and in the case of these enzymes, the K m v a l u e s (0.2 / t M ) for such a primer (estimated at 60 o C) exceed only 2-5-times those for mesophilic eubacterial or eukaryotic DNA polymerases (estimated at 30 ° C). Assuming - according to Refs. 12-14 - that K m for primers are close to Kd, Gibbs' energy values, G °, of the complex formation were calculated as equal to
number of nucleotide units (n) is linear up to n - 10. Each primer unit enhances the affinity by a factor of 2.16 (AG O= -0.51 kcal/mol). The result may be described by a common equation: AG = -6.54-0.51(n - 1 ) (kcal/mol), where -6.54 kcai/mol is Gibbs' energy of dAMP. The data of Table II show that the oligonucleotide d(pA)9p(rib* ) with a urea residue in place of the terminal base, has nearly the same affinities for polymerases A and B as d(pA)9; this suggests that the contribution of the 3'-terminal base to the interaction of a primer with an enzyme-template complex does not differ from that of the other bases of the primer. The data available indicate a specific strong interaction of the 3'-terminal nucleotide residue of a primer with an enzyme-template complex ( A G = - 6 . 0 kcal/mol); interactions of the other nucleotide units of a primer are much weaker (2-I03-times, AG ffi -0.51 kcal/mol). It is reasonable to suggest that the latter interactions are determined by base-pairing between primer units and the template and that archaebacterial DNA polymerases studied interact with only the 3'-terminal nucleotide of a primer, as in the case of DNA polymerase from human placenta and E. coil DNA polymerase I [12-14]. With respect to the mode of interaction with primers, polymerases A and B also bear some other resemblances with eubacterial and eukaryotic DNA polymerases. The plots of log K m and log Vma,, against n are linear with the breaking point at n - lP in case of all enzymes. This may mean that only 10 mononucleotide units of a primer are present in the range of the protein globule of the DNA polymerase. Interactions between template and primer chains outside the enzyme molecule can change the character of the dependence of K m and Vmax o n n [14,15]. In the case of DNA polymerase a from human placenta, it has been shown that substitution of the 3'-terminal base of a primer for a noncomplementary one results in a significant decrease of primer affinity: the value of K m for d(pT)10pC is 39-times that for d(pT) n. As seen from the data of Table II, values of K~ for d(pA)9pC are 7- and ll-times those for d(pA)lo in the case of polymerase A and polymerase B, respectively. The maximal rates of dAMP primer elongation catalyzed by polymerases A and B constitute about 20~ of that of the d(pA)lo primer, and lengthening of a primer by each nucleotide (n ffi 1-10) increases the value of Vma~ 1.2-times: V~ffi 20~V10(1.2) "-1. A similar dependence is observed for DNA polymerase a from human placenta: V~ffi 7%V1o(1.36)"-1 [13,15]. Table II shows that the affinity of d(pT), primers for polymerases A and B is higher than that of d(pA), primers of the same length. The same was observed for eukaryotic DNA polymerase a [13]. It must be noted
107
that in the case of archaebactefial enzymes the maximal rate of d(pA)l o elongation is about 50-100-times higher than that of d(pT)~0 primer. The results obtained demonstrate that the mode of interaction with primers of thennophilic archaebacter/al DlqA polymerases does not differ much from that of mesophilic eubacterial or eukaryotic DNA polymerases. References 1 Woese, C.R., Magrum, L.J. and Fox, G.E. (1978) J. Mol. Evol. 11, 245-252. 2 Woese, C.R. and Olsen, (3. (1986) System. AppL Microbioi. 7, 161-177. 3 Kohiyarna, M., Nakayama, M. and Ben Mahrez, K. (1986) System. Appl. Microbiol. 7, 79-82. 4 Zabel, H.-P., Fischer, H., Holler, E. and Winter, J. (1985) System. Appl. Microbiol. 6, 111-118. 5 Prangishvili, D.A. (1986) Mol. Biol. USSR 20, 477-488. 6 Rossi, M., Rella, R., Pensa, M., Battolucci, S., De Rosa, M., Gambacorta, A., Raia, C.A. and DelrAversano Orabona, N. (1986) System. Appl. Microbiol. 7, 337-341.
7 Chinchaladze, D.Z., PrangJshvili, D.A., Kachabava, L.A. and Zaafishvili, M.M. (1985) Mol. Biol. USSR 19, 1466-1475. g Forterre, P., Nadal, M., Elie, C., Mirambeau, G., Jaxel, C. and Duguet, M. (1986) System. App|. Microbiol. 7, 67-71. 9 I¢~mczak, L.J., Grummt, F. and Burger, K.J (1985) Nucleic Acids Res. 13, 5269-5282. 10 Klimczak, L.J., Grummt, F. and Burger, K.J. (1986) Biochemistry 25, 4850-4855. 11 Levina, A.S., Nevinsky, G.A. and Lawik, O.l. (1985) Bioorg. Chem. USSR, 11, 358-362. 12 Nevinsky, G.A., Frolova, E.I., Podust, V.N., Levina, A.S., Lavrik, O.I. and Khalabuda, O.V. (1986) Bioorg. Chem. USSR, 12, 357-368. 13 Venjaminova, A.G., Levina, A.S., Nevinsky, G.A. and Podust, V.N. (1987) Mol. Biol. USSR 21, 1378-1385. 14 Nevinsky, G.A., Levina, A.S., Frolova, E.I. and Podust, V.N. (1987) Mol. Biol. USSR 21, 1193-1200. 15 Nevinsky, G.A., Frolova, E.I., Levina, A.S., Podust, V.N. and Lebedev, A.V. (1987) Bioorg. Chem. USSR 13, 45-57. 16 Knorr¢, D.G., Lavrik, O.I. and Nevinsky, G.A. (1988) Biochimie 70, 655-661. 17 Eisenthai, R. and Cornish-i~iowdcn, ,,~. [l~J4j bi~;i~m..L IJ.,, 715-720.