- Email: [email protected]
2′,5′-oligoadenylate synthetase from a lower invertebrate, the marine sponge Geodia cydonium, does not need dsRNA for its enzymatic activity
Biochimica et Biophysica Acta 1590 (2002) 140 – 149 www.bba-direct.com
2V,5V-oligoadenylate synthetase from a lower invertebrate, the marine sponge Geodia cydonium, does not need dsRNA for its enzymatic activity Annika Lopp a,*, Anne Kuusksalu a, To˜nu Reintamm a, Werner E.G. Mu¨ller b,1, Merike Kelve a a
Laboratory of Molecular Genetics, National Institute of Chemical Physics and Biophysics, Akadeemia tee 23, 12618 Tallinn, Estonia Institut fu¨r Physiologische Chemie, Abteilung fu¨r Angewandte Molekularbiologie, Universita¨t, Duesbergweg 6, D-55099 Mainz, Germany
b
Received 6 December 2001; received in revised form 21 March 2002; accepted 11 April 2002
Abstract Recently, the presence of 2V,5V-linked oligoadenylates and a high 2V,5V-oligoadenylate synthetase activity were discovered in a lower invertebrate, the marine sponge Geodia cydonium. It has been demonstrated that mammalian 2 – 5A synthetase isozymes require a dsRNA cofactor for their enzymatic activity. Our results show that, unlike mammalian 2 – 5A synthetases, the 2 – 5A synthetase from the sponge acts in a dsRNA-independent manner in vitro. A prolonged incubation of the G. cydonium extract with a high concentration of a micrococcal nuclease had no effect on the activity of the 2 – 5A synthetase. At the same time, the micrococcal nuclease was effective within 30 min in degrading dsRNA needed for the enzymatic activity in IFN-induced PC12 cells. These results indicate that the 2 – 5A synthetase from G. cydonium may be active per se or is activated by some other mechanism. The sponge enzyme is capable of synthesizing a series of 2 – 5A oligomers ranging from dimers to octamers. The accumulation of a dimer in the predominant proportion during the first stage of the reaction was observed, followed by a gradual increase in longer oligoadenylates. By its product profile and kinetics of formation, the sponge 2 – 5A synthetase behaves like a specific isoform of enzymes of the 2 – 5A synthetase family. D 2002 Elsevier Science B.V. All rights reserved. Keywords: 2 – 5A synthetase; Sponge; Geodia cydonium; Enzymatic activity; dsRNA
1. Introduction Higher vertebrates have evolved a number of defense mechanisms to control viral infection. One of the most extensively characterized mechanisms among them is an interferon (IFN)-stimulated 2V,5V-linked oligoadenylate (2– 5A) pathway or a 2– 5A system (reviewed in Refs. [1,2]). By now, it has been demonstrated that the 2 – 5A system functions as an antiviral pathway in vivo in animals [3]. Other possible functions such as growth control and the control of apoptosis have also been suggested for this enzyme system [4 –7]. The key enzyme of the 2 –5A pathway, (2V– 5V) oligoadenylate synthetase (2– 5A synthetase: EC 2.7.7.-) converts cellular ATP to a series of short 2V,5V-linked oligoadenylates (2 –5A) with a general formula pppA(2Vp5VA)n, n z 1 [8]. In *
Corresponding author. Tel.: +372-6-398-352; fax: +372-6-398-382. E-mail addresses: [email protected] (A. Lopp), [email protected] (W.E.G. Mu¨ller). 1 Fax: + 49-6131-395243.
human cells, three types of IFN-induced 2– 5A synthetases have been cloned, expressed and characterized corresponding to small (p40, p46), medium (p69, p71) and large (p100) isoforms [9]. The enzymatic activity of all these isoforms requires a cofactor, a double-stranded RNA [9,10]. Their enzymatic product, 2– 5A, functions as an allosteric activator of a latent endoribonuclease (RNase L), which degrades a single-stranded viral or cellular RNA [11,12]. 2 – 5A synthetases of the sizes similar to those of the synthetases found in humans have been identified in other vertebrates (mouse, rat, pig, woodchuck, chicken) [13,14]. The 2 –5A synthetase family also comprises homologous proteins without the 2 – 5A synthesizing activity; these members are called 2– 5A synthetase-like proteins [15,16]. Contrary to the wide occurrence of the 2– 5A synthetase in vertebrates, this enzyme has not been found in lower organisms such as bacteria and C. elegans [13,14,17]. In our earlier study, we discovered the presence of 2V,5Vlinked oligoadenylates and a high 2 –5A synthetase activity in a lower invertebrate, the marine sponge Geodia cydonium [18]. The 2V,5V phosphodiester bond in the oligoadenylates
0167-4889/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 8 8 9 ( 0 2 ) 0 0 2 0 7 - 0
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
was verified by the NMR analysis and the oligonucleotide composition was determined by using the MALDI-MS method [19]. We also succeeded in cloning the putative 2 –5A synthetase from the marine sponge G. cydonium [20]. Although we have demonstrated the presence of the 2 – 5A synthesizing activity in the sponge G. cydonium, the role of the 2 –5A synthesis in sponges has not been defined. Sponges are known to live in symbiosis with certain bacteria and unicellular eukaryotic organisms (reviewed in Ref. [21]). Besides, as sponges are filter-feeding metazoa, they are exposed to a huge amount of organic matter via the extraction of edible material from the water [22]. Accordingly, sponges have developed an effective chemical defense system as well as humoral and cellular defense mechanisms against toxic substances or organisms [23,24]. Whether the 2 – 5A synthetase in sponges is, similar to vertebrates, involved in defense reactions against viruses or microorganisms, is not known at present. It has been reported that the 2– 5A synthetase activity in the marine sponge G. cydonium can be modulated by some environmental stressors [25]. However, in order to understand the function of the 2 – 5A synthesis in sponges, the factors needed for the activation of the 2– 5A synthetase as well as catalytic properties of the enzyme will have to be established. While searching for the possibilities of a biochemical characterization of the sponge enzyme, we found that contrary to mammalian enzymes, the sponge 2 – 5A synthetase, which was immobilized to Hybond N + membrane, did not need poly(I)poly(C) as a dsRNA cofactor for its enzymatic activity in vitro. The nuclease treatments of the G. cydonium extracts that were effective in degrading the dsRNA needed for the enzymatic activity in mammalian cells did not influence the observed 2 – 5A synthetase activity. The enzyme from G. cydonium is capable of synthesizing 2 –5A oligomers from dimers to octamers with the accumulation of the dimer in a predominant proportion during the first stage of the reaction. The enzymatic studies demonstrate that in the absence of dsRNA, the sponge 2– 5A synthetase behaves like a specific isoform of enzymes of the 2 –5A synthetase family.
2. Materials and methods 2.1. Materials Adenosine 5V-triphosphate was purchased from Reanal (Budapest, Hungary). [14C]ATP (50 –60 mCi/mmol) was purchased from Amersham and purified by HPLC. Polyethyleneimine (PEI)-cellulose TLC-Ready-Plates were from Schleicher and Schuell (Germany). The membranes Hybond N + , Hybond N, Hybond C and PVDF were from Amersham International PLC (Buckinghamshire, UK). Poly(I)poly(C), creatine kinase (EC 2.7.3.2) and phosphocreatine were
141
obtained from Boehringer Mannheim (Germany). DNA (Type I, highly polymerized, from calf thymus) and micrococcal nuclease (EC 3.1.31.1; 107 U/mg protein) were obtained from Sigma Chemical Co. (St. Louis, MO, USA). The shrimp alkaline phosphatase (SAP) was from USB Corporation (Cleveland, OH, USA). 2.2. Sponge Specimens of the marine sponge G. cydonium (Porifera, Demospongiae, Geodiidae) were collected near Rovinj (Croatia) at depths between 15 and 35 m, cut to pieces and immediately frozen in liquid nitrogen. The frozen pieces were stored at 70 jC. 2.3. Cell extracts The frozen sponge tissue was mechanically broken in liquid nitrogen and during homogenization, a lysis buffer was added (1.5 –2 v/w), consisting of 10 mM HEPES, pH 7.6, 90 mM KCl, 1 mM Mg-acetate, 0.5% NP-40 and 2 mM 2-mercaptoethanol. The supernatant obtained after centrifugation (10,000 g; 10 min; 4 jC) was immediately frozen in liquid nitrogen and stored at 70 jC (crude extract). The extracts of g-IFN-induced PC12 cells were prepared as described earlier [26] and stored at 70 jC. Protein concentration in cell extracts was measured by a modified Bradford method [27]. 2.4. 2 –5A synthetase activity assay The 2 –5A synthetase activity in the extracts was determined after binding the enzyme to the poly(I)poly(C)membrane as described earlier [18], or simply after binding the enzyme to the positively charged nylon membrane Hybond N + . The crude sponge extract (20 Al, 200 –300 Ag of total protein) or IFN-induced PC12 cell extract (20 Al, 200 – 300 Ag of total protein) was added to a 0.16 cm2 membrane piece. The binding was performed in microtiter plate wells at room temperature by gently shaking for 30 min. The membrane was washed for 4 5 min with 200 Al of the reaction buffer (without ATP) and shortly dried on the filter paper. Then, 50 Al (10 Al for radioactive assays) of the reaction mixture (30 mM Tris – HCl, pH 8.0, 100 mM KCl, 5 mM MgCl2, containing 1 mM ATP as a substrate and 20,000 cpm of [14C]ATP-omitted in nonradioactive assays) was applied to each piece of the membrane. Using different batches of crude extracts, we obtained membranes with different 2 – 5A synthesizing activities (approximately 10 – 75% of ATP converted into its products per hour). Thus, each series of experiments was performed with the same batch of the extract. In activity assays ‘‘in solution,’’ we incubated the reaction mixture which contained 2 Al of the sponge extract (30 Ag of the total protein) in a final volume of 50 Al at 37 jC in the presence or absence of 100 Ag/ml of poly (I)poly(C).
142
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
Contrary to the crude G. cydonium extracts, the PC12 cell extracts contain a remarkable ATPase activity that competes with the 2 – 5A synthetase for the substrate. Therefore, to avoid the degradation of ATP through the phosphatase activity, the ATP regenerating system (0.5 mg/ ml creatine kinase, 30 mM phosphocreatine) was added to the reaction mixture in these experiments. After 0.5– 4 h, depending on the rate of conversion of ATP to its products, the reaction was terminated. For that, the reaction mixture was heated for 5 min at 95 jC and the products were analysed in their triphosphorylated form. Alternatively, to analyse the products in their ‘‘core’’ form the reaction mixture was treated with the SAP (0.04 U/Al, 1 –1.5 h at 37 jC) to remove the 5V phosphate groups followed by the inactivation of the enzyme for 15 min at 65 jC. 2.5. TLC separation of the products In radioactive assays, the reaction products were separated by TLC on PEI-cellulose [18,25]. As a mobile phase 0.4 M Tris –HCl (pH 8.6) containing 30 mM MgCl2 were used. The dried TLC plates were exposed to a CS-imaging screen and scanned by using the Bio-Rad GS-525 Molecular Imager System. The amounts of ATP and 2 –5A oligomers were quantified by the relative intensities of the corresponding spot areas on the autoradiograms. 2.6. HPLC analysis of the reaction products The reaction products were analysed on the HPLC apparatus (Waters) equipped with a C18 reverse-phase column (Supelcosilk LC-18, 300 4.6 mm, 5 Am). The 2V– 5V-linked triphosphorylated oligoadenylates were separated in the system which consisted of solvent A (50 mM NH4H2PO4, pH 7.0) and solvent B (MeOH/H2O, 1:1), by applying a 1 – 60% gradient of solvent B in solvent A for 30 min. The absorbance was measured at 254 nm. The dephosphorylated oligoadenylates (the ‘‘core’’ form of oligomers) were separated in the system which consisted of solvent A (0.1 M triethylammoniumacetate (TEAA), pH 7.0) and solvent B (40% acetonitrile/0.1 M TEAA, pH 7.0), by applying a 10 – 30% gradient of solvent B in solvent A for 30 min. The absorbance was measured at 265 nm. The retention times of ATP, adenosine and 2– 5A oligoadenylates, in either their triphosphorylated or ‘‘core’’ form were estimated by comparing them with those of authentic compounds and individual chemically synthesized 2 – 5A oligoadenylates. The structure of 2V,5V-linked oligoadenylates has been previously verified by enzymatic (ribonuclease U2 and alkaline phosphatase) treatments and by a biological assay in rabbit reticulocyte lysates [18] and also directly by NMR analysis and the MALDI-MS method [19]. The quantification of the products was performed by measuring the relative peak areas (Millenium32 version 3.05 software, Waters Corporation). In calculations, the molar absorption
coefficients of 1.54 10 4 , 2.58 10 4 , 3.60 10 4 , 4.16 104, and 5.0 104 were used for ATP, 2V,5V-linked A2, A3, A4, and A5, respectively [28]. 2.7. Nuclease treatments Micrococcal nuclease treatments were performed in ‘‘crude extracts’’ as well as on the membrane-bound material from the sponge extract. As a control, the IFN-induced PC12 cell extract was used as a dsRNA-dependent mammalian model system of the 2 –5A formation. 2.7.1. Treatment of the sponge material For the hydrolysis of nucleic acids in crude extracts, 95 Al of the crude extract were incubated at 37 jC in the presence of 10 mM Tris – HCl buffer, pH 8.6, 10 mM CaCl2 and 0.05 U/Al of the micrococcal nuclease (final volume 100 Al) for various periods of time (0 to 22 h). Then the extract was used to determine the Hybond N + -bound 2 – 5A synthetase activity or the 2 – 5A synthetase activity ‘‘in solution’’ as described above. To study the effect of the nuclease on the Hybond N + bound material, the membrane was subsequently washed and incubated with 100 Al of 100 mM glycine buffer (pH 8.6) containing 10 mM CaCl2, in the presence or absence of 0.05 U/Al of the micrococcal nuclease at room temperature for 30 min. After the washing step, the 2 –5A synthetase activity was determined. To follow the nucleic acid (endogenous or added) degradation in the extracts, the time-dependent increase in the absorbance at 260 nm was monitored. For that, the extracts were supplemented with 10 mM Tris –HCl, pH 8.6 and 10 mM CaCl2 in either the presence or absence of 0.05 U/Al of the micrococcal nuclease. 2.7.2. Experiments with PC12 cells To study the effect of the nuclease on dsRNA, the reaction mixture, which contained 1 mM ATP, 10 Ag/ml or 100 Ag/ml of poly(I)poly(C) and 2.5 mM CaCl2 in the reaction buffer, pH 8.0, was incubated for 30 min at 30 jC in the presence or absence of 0.05 U/Al of the micrococcal nuclease. The preincubated mixture was added to the Hybond N + -bound enzyme from the PC12 cell extracts in order to carry out the reaction. Alternatively, the poly(I)poly(C)-membrane was treated with the micrococcal nuclease as described and, after washing the membrane, the enzyme from the PC12 cell extract was bound to it and the 2 – 5A synthetase activity measured. In another series of experiments, the effect of the nuclease was studied after the formation of the 2 – 5A synthetase – dsRNA complex. The enzyme from PC12 cells was bound to the poly(I)poly(C)-membrane, the membrane was washed and treated with the micrococcal nuclease in the glycine buffer as described above. After washing the membrane, the 2 –5A synthetase activity assay was performed. Alternatively, the enzyme from PC12 cells was bound to the
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
143
membranes (Hybond C, Hybond N, PVDF) also gave positive results, suggesting that the protein binding was nonspecific. The crude extracts of G. cydonium may contain possible dsRNA activators of the 2 – 5A synthetase which bind to the Hybond N + membrane together with the enzyme and activate it. Therefore, in order to carry out the hydrolysis of nucleic acids, we undertook the treatment of the extracts with the micrococcal nuclease. 3.2. The nuclease treatment of sponge extracts
Fig. 1. Formation of the 2 – 5A reaction products by poly(I)poly(C)membrane-bound and Hybond N + membrane-bound material from G. cydonium extracts. The membrane-bound material was subjected to activity assay as described in Materials and methods. The products formed from 1 mM ATP during a 4-h synthesis were dephosphorylated and analyzed by HPLC. The peaks were identified on the basis of the retention times measured for authentic compounds and the products were quantified by relative peak areas. (A) Retention time: adenosine (peak 1)—6.9 min, (2 – 5) A2 (peak 2)—11.6 min, (2 – 5) A3 (peak 3)—15.5 min, (2 – 5) A4 (peak 4)—17.6 min, (2 – 5) A5 (peak 5)—18.9 min, (2 – 5) A6 (peak 6)—20.8 min. (B) Amounts of products (mean F SE; n = 3) by using the Hybond N + membrane (peaks 1,2 and 3) or poly(I)poly(C)membrane (peaks 1V, 2V and 3V).
Hybond N + -membrane followed by the sequential binding of the poly(I)poly(C), micrococcal nuclease treatment and the addition of the reaction mixture. Between all the steps, the membrane was washed. The reaction products were subjected to HPLC analysis.
3. Results
To follow the degradation of nucleic acids in the extracts, the micrococcal nuclease was added to the extract and an increase in the absorbance was registered at 260 nm. A sharp rise in A260 was observed during the first 30 min of incubation, followed by a subsequent slower increase (not shown). In the presence of the micrococcal nuclease the total increase in the absorbance was twice higher than in the absence of the nuclease, the latter, in its turn, was evidently caused by endogenous nucleases. These results confirm the occurrence of hydrolysis of nucleic acids in the extracts. To ensure complete hydrolysis of endogenous nucleic acids, the crude extracts of G. cydonium were incubated with the micrococcal nuclease for prolonged periods of time (up to overnight) followed by the 2 –5A synthetase activity assay. The results presented in Table 1 demonstrate that the pretreatment of the extract with a high concentration of the nuclease (see: Materials and methods) had no effect on the formation of 2V,5V-oligoadenylates in any of the preincubation time periods studied. The micrococcal nuclease also remained active in the preincubated extracts as proven by a characteristic increase in A260 of the added DNA (not shown). Moreover, the results show a high stability of the 2 –5A synthetase in the pretreatment conditions (Table 1).
3.1. 2 –5A synthetase activity test with sponge extracts using poly(I)poly(C)membrane-bound and Hybond N+-bound enzyme assays The poly(I)poly(C)membrane-bound enzyme activity test has been typically used to study the 2 –5A synthetase activity in mammalian cell extracts [29]. As shown in our earlier reports [18,19], the enzyme from G. cydonium was also capable of synthesizing 2V,5V-linked oligoadenylates when immobilized to the poly(I)poly(C)membrane. As a control of the binding specificity, we used a positively charged nylon membrane (Hybond N + ) instead of the poly(I)poly(C)membrane which was prepared on the basis of Hybond N + . Surprisingly, the use of the Hybond N + membrane also enabled the synthesis of 2 –5A oligomers. Analysis of the reaction products showed no difference between the profiles of the products (Fig. 1A) and the yields of 2– 5A oligomers (Fig. 1B), regardless of whether the poly(I)poly(C)membrane or Hybond N + membrane was used for the immobilization of the enzyme. The use of other
Table 1 The effect of the nuclease treatment of G. cydonium extract on the formation of 2V,5V-linked oligomers from ATP Preincubation time (h)
Micrococcal nuclease
0 + 1 + 3 + 22 +
Products (nmol/assay) p3A2
p3A3
p3A4
5.9 F 0.3a 5.6 F 0.5 6.3 F 0.4 6.3 F 0.3 6.0 F 0.5 6.4 F 0.6 5.6 F 0.4 6.0 F 0.4
1.3 F 0.09 1.2 F 0.10 1.4 F 0.08 1.6 F 0.07 1.5 F 0.10 1.8 F 0.10 1.1 F 0.06 1.3 F 0.08
0.22 F 0.02 0.25 F 0.02 0.28 F 0.02 0.34 F 0.01 0.33 F 0.03 0.46 F 0.03 0.19 F 0.01 0.25 F 0.02
The sponge extracts were incubated in the presence or absence of 0.05 U/Al of the micrococcal nuclease for indicated periods of time and the Hybond N + -bound material was subjected to activity assay as described in Materials and methods. The triphosphorylated products, formed from 1 mM ATP during a 2-h synthesis, were analyzed by the HPLC method. a The data represent mean F SE (n = 3).
144
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
Table 2 The effect of poly(I)poly(C) on the formation of 2V,5V-linked oligomers by the 2 – 5A synthetase from nuclease-treated G. cydonium extract Micrococcal nuclease
Poly(I)poly(C) (100 Ag/ml)
Products (nmol/assay) p3A2 a
+ + +
+
7.0 F 0.4 7.1 F 0.3 7.2 F 0.2 6.9 F 0.4
p3A3
p3A4
1.1 F 0.06 1.0 F 0.04 1.1 F 0.03 1.0 F 0.05
0.25 F 0.02 0.26 F 0.03 0.25 F 0.01 0.22 F 0.03
The sponge extracts were incubated overnight in the presence or absence of 0.05 U/Al of the micrococcal nuclease, thereafter the extract was used to carry out the synthesis of 2 – 5A oligomers ‘‘in solution’’ as described in Materials and methods. The triphosphorylated products, formed from 1 mM ATP during a 2-h synthesis, were analyzed by the HPLC method. a The data represent mean F SE (n = 3).
Thus, the degradation of nucleic acids in crude extracts did not influence the activity of the 2– 5A synthetase from the sponge G. cydonium. In order to rule out the fact that the binding to the Hybond N + membrane could cause the possible structural changes leading to the activation of the enzyme, the assay was carried out ‘‘in solution.’’ For that, the crude extracts were incubated overnight in the presence or absence of the micrococcal nuclease and then used for the determination of
Fig. 2. The effect of the micrococcal nuclease on the poly(I)poly(C)induced formation of 2 – 5A oligomers by the 2 – 5A synthetase from IFNinduced PC12 cells. The Hybond N + -bound enzyme from PC12 cell extracts was subjected to activity assay as described in Materials and methods. The HPLC profile of the products formed from 1 mM ATP during a 2-h synthesis in the absence of poly(I)poly(C) (A), in the presence of 10 and 100 Ag/ml of poly(I)poly(C), (B) and (C), respectively, or in the presence of 100 Ag/ml poly(I)poly(C) pretreated for 30 min with 0.05 U/Al of the micrococcal nuclease (D). Peaks 1, 2, 3 and 4 correspond to ATP, p3A2, p3A3 and p3A4, respectively.
2 – 5A synthetase activity either in the presence or absence of poly (I)poly(C). The results in Table 2 reveal that in all cases, the synthesis of 2 –5A oligomers remained the same. 3.3. Comparative studies in a dsRNA-dependent mammalian cell model system The effect of the micrococcal nuclease on the poly (I)poly(C)-activated formation of 2V,5V-linked oligoadenylates was studied in a mammalian cell model. For this purpose, the extracts of g-IFN-induced PC12 cells were used. When these extracts were applied to the Hybond N + membrane and the membrane-bound enzyme was used to carry out the 2– 5A synthesis, no products were formed (Fig. 2A). The presence of 10 Ag/ml of poly(I)poly(C) in the reaction mixture was sufficient to activate the Hybond N + -bound enzyme from PC12 cells (Fig. 2B). The formation of 2 –5A oligomers, mainly trimers and tetramers, was even more elevated in the presence of 100 Ag/ml of poly(I)poly(C) (Fig. 2C). On the contrary, no synthesis of 2– 5A oligomers was observed when the reaction mixture containing dsRNA was treated for 30 min with the micrococcal nuclease and
Fig. 3. The micrococcal nuclease treatment of the membrane-bound mammalian enzyme – poly(I)poly(C) complex and the membrane-bound material from G. cydonium extract. The 2 – 5A synthetase from IFNinduced PC12 cells or G. cydonium was bound to the Hybond N + membrane followed by the binding of poly(I)poly(C) (omitted in the case of sponge extracts), the nuclease treatment of the membrane and activity assay as described in Materials and methods. The HPLC profile of the products formed from 1 mM ATP during a 2-h synthesis by the mammalian (A, C) or during a 0.5-h synthesis by the sponge 2 – 5A synthetase (B, D) when the membrane-bound material was treated for 30 min in the absence (A, B) or presence (C, D) of 0.05 U/Al of the micrococcal nuclease. Peaks 1, 2, 3 and 4 correspond to ATP, p3A2, p3A3 and p3A4, respectively.
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
subsequently applied to the membrane-bound enzyme (Fig. 2D). Alternatively, the poly(I)poly(C)membrane was incubated in the presence of the micrococcal nuclease before binding the 2 –5A synthetase to it. Under these conditions, the nuclease treatment also abolished the synthesis of 2– 5A oligomers (not shown). These results demonstrate that 0.05 U/Al of the micrococcal nuclease was effective enough for degrading poly(I)poly(C) required for the enzyme activity in mammalian PC12 cells. Further, the effect of the nuclease on the synthesis of 2 – 5A was studied after binding the 2 – 5A synthetase to poly(I)poly(C), i.e. after the formation of the enzyme – dsRNA complex. For that, the enzyme from the PC12 cells was bound to the Hybond N + membrane followed by the binding of poly(I)poly(C). The micrococcal nuclease treatment of such a membrane abolished the synthesis of the products (Fig. 3A and C). The same result was obtained using the poly(I)poly(C)membrane-bound enzyme from PC12 cells. In contrast, when the Hybond N + -bound material from G. cydonium extracts was used in the assay, the nuclease treatment had no effect on the synthesis of 2 – 5A products (Fig. 3B and D).
145
3.4. A biochemical characterization of the 2 –5A oligomer formation by the 2 – 5A synthetase from G. cydonium On the basis of our finding that the formation of 2 –5A oligomers by the sponge 2 – 5A synthetase was poly (I)poly(C)-independent, all subsequent experiments were performed using the Hybond N + -bound enzyme without an additional dsRNA. The requirement for divalent cations, an optimal temperature and the pH value were determined to optimize the rate of formation of 2 –5A oligoadenylates. Fig. 4A demonstrates the yields of ATP oligomerization at different temperatures. It can be seen that the enzyme is active in quite a wide range of temperatures used (10 – 70 jC). The formation of 2– 5A oligomers was most effective in the temperature range of 25 –45 jC with an optimum at about 37 jC. Fig. 4B demonstrates the dependence of the formation of 2 –5A oligomers on pH of the assay buffer. It appears that for the maximal activity, basic pHs with an optimum at about 8 are needed. The quantities of 2 – 5A oligomers formed at different concentrations of MgCl2 are presented in Fig. 4C. The results show that the maximal rate of
Fig. 4. Biochemical characteristics of the formation of 2 – 5A oligomers by the Hybond N + -bound enzyme from G. cydonium extracts. (A) Temperaturedependence. After the incubation of 1 mM ATP+[14C]ATP with membrane-bound enzyme for 4 h, the reaction products were separated by TLC on PEIcellulose as described in Materials and methods. The quantification of the products on the autoradiograms was performed by the relative spot intensities. Each point represents a mean F SE (n = 4). (B) pH-dependence. After the incubation of 1 mM ATP with membrane-bound enzyme for 0.5 h at an indicated pH of the reaction mixture, the triphosphorylated products were analyzed by HPLC as described in Materials and methods. Each point represents a mean of the two independent experiments. (C) The effect of MgCl2. The products formed from 1 mM ATP at an increasing concentration of MgCl2 during a 4-h synthesis were dephosphorylated and analyzed by the HPLC method. The values represent means of the two independent experiments. (D) The time-course of formation of 2 – 5A oligomers. ATP (0.63 mM) was incubated with membrane-bound enzyme for indicated periods of time and the triphosphorylated products were analyzed by HPLC and quantified as described in Materials and methods.
146
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
formation of the products from 1 mM ATP was observed at 5 mM MgCl2, while no activity of the enzyme was detected in the absence of MgCl2 (Fig. 4C). In a separate series of experiments, we studied the ability of different divalent metal ions (Mn2 + , Zn2 + , Ca2 + ) to substitute Mg2 + in the oligomerization reaction of ATP. The results show that the potency of different divalent ions to activate the sponge synthetase decreased in the following order: Mg > Mn>Zn resulting under these conditions in the conversion of 96%, 85% and 38% of ATP, respectively (Fig. 5A – C). The products were not formed in the presence of CaCl2 (Fig. 5D). Mammalian 2– 5A synthetases have a characteristic pattern of 2– 5A oligomers synthesized by each protein [9,10]. Looking at the pattern of the oligomers formed by the sponge synthetase, we found a higher proportion of the 2 –5A dimer compared to that of the longer products (Fig. 5A). Fig. 4D shows the time-course of formation of 2V,5Vlinked di-, tri-, tetra-, and penta-adenylates from ATP. It demonstrates that the disappearance of ATP was quantitatively related to the appearance of oligonucleotides. During the first stage of the reaction, when up to 50% of ATP was used, the dinucleotide was prevalent but the ratio of dimer to longer oligomers decreased over time. Nevertheless, the molar ratio of dimer to total products was about 0.5 when approximately 90% of ATP was converted to its products (Fig. 4D).
Fig. 5. The effect of different divalent metal ions on the formation of products by the 2 – 5A synthetase from G. cydonium. The HPLC profile of the products formed from 1 mM ATP during a 3-h synthesis in the presence of 5 mM MgCl2 (A), 5 mM MnCl2 (B), 5 mM Zn-acetate (C) or 5 mM CaCl2 (D) in the reaction mixture, pH 7.6. Peaks 1, 2, 3 and 4 correspond to ATP, p3A2, p3A3 and p3A4, respectively. The group of peaks, designated as ‘‘5,’’ contains p3A5 and oligomers of longer chain length.
4. Discussion Mammalian 2– 5A synthetases were discovered as IFNinduced and dsRNA-activated enzymes (reviewed in Ref. [30]). Their dependence on dsRNA was proven by means of the enzymes purified from cell extracts [31 –33] and, later by means of recombinant proteins [34 –37]. In our earlier study, we discovered that the enzyme from the crude extract of a lower invertebrate, the marine sponge G. cydonium, also has a high 2 –5A synthetase activity when immobilized to the poly(I)poly(C)membrane [18]. Some years ago, we succeeded in cloning the putative 2 –5A synthetase from the sponge G. cydonium, which shows some, although limited, homology to small 2– 5A synthetase isoforms of mouse, chicken and human [20]. However, up to now, the expression of this cDNA clone in neither the bacterial nor insect cells has resulted in the active protein (Kuusksalu, A., unpublished data). It is not clear whether the polypeptide obtained is responsible for the observed 2 –5A synthetase activity in G. cydonium. Therefore, to further study the parameters needed for the 2 –5A synthetase activity in the sponge, we used the membrane-bound enzyme preparations from the crude extracts of G. cydonium in the present study. We found that the Hybond N + membrane as well as other membranes (Hybond C, Hybond N, PVDF) are capable of binding the 2 – 5A synthetase from the crude extracts of G. cydonium nonspecifically and this membrane can be used to carry out the synthesis of 2 – 5A without an additional dsRNA (Fig. 1). The 2 –5A synthetase from IFN-induced PC12 cells, used by us as a mammalian cell model, is also capable of nonspecific binding to Hybond N + , but no activity of the enzyme is observed in the absence of dsRNA. As expected, the 2– 5A synthesis can be induced in this case in a dose-dependent manner by adding poly(I)poly(C) as a dsRNA cofactor to the reaction mixture (Fig. 2A – C). Provided that the sponge 2 –5A synthetase behaves like mammalian enzymes, it is conceivable that the crude extract of G. cydonium contains dsRNA activators which bind together with the protein to the membrane and activate the 2 – 5A synthetase. To exclude this possibility, the sponge extracts were treated with the micrococcal nuclease, an enzyme used to hydrolyse single-stranded and doublestranded nucleic acids in crude cell-free extracts [38,39]. Our results in Table 1 reveal that the incubation of the extracts with the nuclease for prolonged periods of time (up to overnight) did not influence the activity of the sponge 2– 5A synthetase. On the contrary, the same concentration of the micrococcal nuclease was, within 30 min of incubation, effective in degrading poly(I)poly(C) required for the enzyme activity in PC12 cells (Fig. 2C and D). In another series of experiments with IFN-induced PC12 cells, we studied the effect of the nuclease on poly(I)poly(C) in the presence of the protein bound to it. We showed that the nuclease treatment of the 2 – 5A synthetase – poly(I)poly(C) complex also abolished the synthesis of 2 – 5A oligomers (Fig. 3A and C). Once again, the membrane-bound 2 –5A
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
synthetase from the sponge exhibited similar activity, irrespective of whether it had been subjected to the nuclease treatment (Fig. 3B and D). To ensure that binding of the sponge 2 –5A synthetase to the membrane would not bring about any structural modifications of the enzyme affecting its activation process, the formation of 2 – 5A oligomers was measured also ‘‘in solution.’’ The data in Table 2 show that the 2– 5A synthetase activity was the same independent of the micrococcal nuclease treatment and the presence or absence of poly (I)poly(C) in the reaction mixture. These data together with the abovementioned results suggest that the 2– 5A synthetase from G. cydonium, differently from mammalian 2– 5A synthetases, does not need dsRNA for its activity. The exact nature of interaction of an RNA activator with an RNA-binding domain in the proteins of the 2 – 5A synthetase family is not known [10,40]. Moreover, none of the proteins of this family isolated so far possesses a typical RNA-binding domain which is characteristic of many dsRNA-binding proteins (reviewed in Ref. [13]). The 2– 5A oligoadenylate synthetase, which has previously been believed to be activated only by dsRNA, can also be activated by RNA ligands with a little secondary structure [40]. Concerning natural activators of the enzyme, certain small viral RNAs have been shown to act as activators upon viral infection [41,42]. At the same time, little is known about cellular activators of 2– 5A synthetases in virus-noninfected cells. With regard to the latter, the activation of the 2 –5A synthetase by certain nuclear dsRNAs (HnRNA) has been proposed [43,44]. The only report that relates to the naturally occurring, non-dsRNA cellular activators of the enzyme demonstrates that high concentrations of fructose1,6-biphosphate can activate the 2– 5A synthetase [39]. There are several possibilities to explain our results. The corresponding sponge protein may be active per se or in the form of a hypothetical strong protein – nucleic acid complex which is resistant to the action of nucleases. Alternatively, it may be activated by a yet unknown component present in the sponge extract. Further experiments will be required to elucidate possible activators (if any) of the sponge 2– 5A synthetase. The optimization studies of the synthesis of 2 – 5A oligomers revealed that basic pH values with an optimum of 8.0, the presence of MgCl2 in excess with respect to ATP and the temperature optimum of 37 jC are necessary for the maximal rate of formation of products by the sponge 2– 5A synthetase (Fig. 4A – C). It has been established that mammalian 2 –5A synthetases belong to the class of nucleotidyltransferases which require divalent metal ions for their activity [45]. The presence of Mg2 + has been shown to be obligatory for the activity of mammalian 2 –5A synthetases [29]. Our results confirm that it is also valid for the sponge synthetase. Other divalent cations (Mn2 + , Zn2 + ), which may be chelated by carboxylate residues in the catalytic center of nucleotidyltransferases [46], were able to substitute Mg2 + with a
147
lower efficiency (Fig. 5). Several metal ions (copper, ion, zinc) have been shown to be strongly inhibitory to the small and medium human 2 – 5A synthetases [47]. The effects of divalent cations on the activity of the sponge 2– 5A synthetase deserve further investigation, since we observed a better formation of 2– 5A oligomers in the presence of zinc than in the presence of calcium. We have previously shown by the MALDI-MS analysis that the 2 – 5A synthetase from G. cydonium is capable of synthesizing oligoadenylates from dimers to octamers [19]. The kinetic studies of formation of different oligomers show that the 2– 5A dimer is the first product to accumulate in a predominant proportion during the first stage of the reaction. At a higher conversion percentage of ATP, the yields of the trimer and other longer oligomers gradually increase (Fig. 4D). However, the molar concentrations of oligomers higher than a pentamer remain low. The mammalian 2– 5A synthetases are characterized by different oligomer patterns and kinetic parameters of formation of 2 –5A oligomers [9,10]. The kinetics of formation of the products by the sponge synthetase (Fig. 4D) is similar to that observed by Justesen et al. [48] who carried out kinetic studies with the 2– 5A synthetase purified from a rabbit reticulocyte lysate. When 90% of ATP was converted to its products, we found the calculated molar ratio of dimer to total products to be about 0.5 vs. 0.41 calculated in experiments with the enzyme from rabbit reticulocytes [48]. The high ratio of dimer to total products is also typical of a large human isozyme p100 [49], whereas the medium isozyme p69 is quite different. Hovanessian et al. [50] have shown that at a maximum activity of each enzyme, the ratio of dimer to other oligoadenylates is 0.49 and 0.08 for the large human 2 – 5A synthetase isozyme p100 and medium isozyme p69, respectively. Taken together, our enzymatic studies demonstrate that the 2 – 5A synthetase from the sponge G. cydonium behaves like a specific isoform of enzymes of the 2 –5A synthetase family in the absence of dsRNA. The cellular abundance of specific 2 – 5A synthetase isoforms and their localization in different subcellular compartments may refer to the multiplicity of functions they exert in cells [51 – 53]. Although the general scheme for the functioning of the 2 –5A synthetase/RNase L system is well established, other mechanisms of actions of specific isozymes in mammalian cells cannot be ruled out. Lately, it has been demonstrated that besides its oligoadenylate synthesizing activity, the murine 2 – 5A synthetase isoform 9 – 2 functions as a proapoptotic protein of the Bcl-2 family in an RNase L-independent way [53]. Since 2V,5V-diadenylates have a low affinity to bind and thus activate the RNase L [49], it has been proposed that they have an independent antiviral activity or some other cellular effects [9,37,54]. Also, the potential role of the large human isozyme p100 in pre-RNA splicing through generating the 2V,5V-phoshodiester bond in the intermediate lariat structure has been proposed [55].
148
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
At present, we do not know the role of 2 –5A synthesis in the sponges. By now, the 2 – 5A synthetase as a single component of the whole mammalian 2– 5A system has been identified in sponges. Whether an interferon-like antimicrobial system exists in sponges is a matter of future studies. Thus, the specific need for 2– 5A products as extracellular or intracellular signalling molecules for sponges has to be established. The results of the present study give evidence of the dsRNA-independent synthesis of 2 – 5A oligomers. The preferential synthesis of dimeric 2 –5A molecules refers to the mode of action of the oligoadenylates formed which may not involve the RNase L pathway.
[13]
[14]
[15]
[16]
[17]
Acknowledgements This study was supported by the European Commission (Project Sponge) and the Estonian Science Foundation (grant No. 4221).
References [1] I.M. Kerr, The 2 – 5A system: a personal view, J. Interferon Res. 7 (1987) 505 – 510. [2] M.R. Player, P.F. Torrence, The 2 – 5A system: modulation of viral and cellular processes through acceleration of RNA degradation, Pharmacol. Ther. 78 (1998) 55 – 113. [3] A. Zhou, J. Paranjape, T.L. Brown, H. Nie, S. Naik, B. Dong, A. Chang, B. Trapp, R. Fairchild, C. Colmenares, R.H. Silverman, Interferon action and apoptosis are defective in mice devoid of 2V,5V-oligoadenylate-dependent RNase L, EMBO J. 16 (1997) 6355 – 6363. [4] G. Rysiecki, D.R. Gewert, B.R.G. Williams, Constitutive expression of a 2V,5V-oligoadenylate synthetase cDNA results in increased antiviral activity and growth suppression, J. Interferon Res. 4 (1989) 649 – 657. [5] M. Diaz-Guerra, C. Rivas, M. Esteban, Activation of the IFN-inducible enzyme RNase L causes apoptosis of animal cells, Virology 236 (1997) 354 – 363. [6] S. Salzberg, T. Hyman, H. Turm, Y. Kinar, Y. Schwartz, U. Nir, F. Lejbkowicz, E. Huberman, Ectopic expression of 2 – 5A synthetase in myeloid cells induces growth arrest and facilitates the appearance of a myeloid differentiation marker, Cancer Res. 57 (1997) 2732 – 2740. [7] J.C. Castelli, B.A. Hassel, A. Maran, P. Paranjape, J.A. Hewitt, X.-I. Li, Y.-T. Hsu, R.H. Silverman, R.J. Youle, The role of 2V– 5Voligoadenylate-activated ribonuclease L in apoptosis, Cell Death Differ. 5 (1998) 313 – 320. [8] I.M. Kerr, R.E. Brown, pppA2Vp5VA2Vp5VA: an inhibitor of protein synthesis synthesized with an enzyme fraction from interferon-treated cells, Proc. Natl. Acad. Sci. U. S. A. 75 (1978) 256 – 280. [9] D. Rebouillat, A.G. Hovanessian, The human 2V,5V-oligoadenylate synthetase family: interferon-induced proteins with unique enzymatic properties, J. Interferon Cytokine Res. 19 (1999) 295 – 308. [10] S.N. Sarkar, G.C. Sen, Production, purification and characterization of recombinant 2V,5V-oligoadenylate synthetases, METHODS: a companion to Methods in Enzymology 15 (1998) 233 – 242. [11] G. Floyd-Smith, E. Slattery, P. Lengyel, Interferon action: RNA cleavage pattern of a (2V– 5V) oligoadenylate-dependent endonuclease, Science 212 (1981) 1020 – 1032. [12] A. Zhou, B.A. Hassel, R.H. Silverman, Expression cloning of 2 – 5A-
[18]
[19]
[20]
[21] [22] [23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
dependent RNAase: a uniquely regulated mediator of interferon action, Cell 72 (1993) 753 – 765. J. Justesen, R. Hartmann, N.O. Kjeldgaard, Gene structure and function of the 2V,5V-oligoadenylate synthetase family, Cell. Mol. Life Sci. 57 (2000) 1593 – 1612. S. Kumar, C. Mitnik, G. Valente, G. Floyd-Smith, Expansion and molecular evolution of the interferon-induced 2V– 5V oligoadenylate synthetase gene family, Mol. Biol. Evol. 17 (2000) 738 – 750. R. Hartmann, H.S. Olsen, S. Widder, R. Joergensen, J. Justesen, p59OASL, a 2V– 5V oligoadenylate synthetase like protein: a novel human gene related to the 2V– 5V oligoadenylate synthetase family, Nucleic Acids Res. 26 (1998) 4121 – 4127. D. Rebouillat, I. Marie´, A.G. Hovanessian, Molecular cloning and characterization of two related and interferon-induced 56-kDa and 30-kDa proteins highly similar to 2V– 5V oligoadenylate synthetase, Eur. J. Biochem. 257 (1998) 319 – 330. P.J. Cayley, R.F. White, J.F. Antoniw, N.J. Walesby, I.M. Kerr, Distribution of the ppp(A2Vp)nA-binding protein and interferon-related enzymes in animals, plants, and lower organisms, Biochem. Biophys. Res. Commun. 108 (1982) 1243 – 1250. A. Kuusksalu, A. Pihlak, W.E.G. Mu¨ller, M. Kelve, The (2V– 5V) oligoadenylate synthetase is present in the lowest multicellular organisms, the marine sponges, Eur. J. Biochem. 232 (1995) 351 – 357. A. Kuusksalu, J. Subbi, T. Pehk, T. Reintamm, W.E.G. Mu¨ller, M. Kelve, Identification of the reaction products of (2V– 5V) oligoadenylate synthetase in the marine sponge, Eur. J. Biochem. 257 (1998) 420 – 426. M. Wiens, A. Kuusksalu, M. Kelve, W.E.G. Mu¨ller, Origin of the interferon-inducible (2V– 5V)oligoadenylate synthetases: cloning of the (2V– 5V)oligoadenylate synthetase from the marine sponge Geodia cydonium, FEBS Lett. 462 (1999) 12 – 18. T.L. Simpson, The Cell Biology of Sponges, Springer-Verlag, New York, 1984. S. Vogel, Current-induced flow through living sponges in nature, Proc. Natl. Acad. Sci. U. S. A. 74 (1977) 2069 – 2071. P. Proksch, Defensive role for secondary metabolites from marine sponges and sponge-feeding nudibranchs, Toxicon 32 (1994) 639 – 655. W.E.G. Mu¨ller, B. Blumbach, I.M. Mu¨ller, Evolution of the innate and adaptive immune systems: relationships between potential immune molecules in the lowest metazoan phylum (Porifera) and those in vertebrates, Transplantation 68 (1999) 1215 – 1227. H.C. Schro¨der, M. Wiens, A. Kuusksalu, M. Kelve, W.E.G. Mu¨ller, Modulation of 2V– 5Voligoadenylate synthetase by environmental stress in the marine sponge Geodia cydonium, Environ. Toxicol. Chem. 16 (1997) 1403 – 1409. A. Lopp, A. Kuusksalu, K. Samuel, M. Kelve, Expression and activity of 2 – 5A synthetase in the course of differentiation and apoptosis of PC12 cells, Cytokine 12 (2000) 737 – 741. M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976) 248 – 254. H. Cailla, C. Le Borgne De Kaouel, D. Roux, M. Delaage, J. Marti, Monoclonal antibodies to 5V-triphospho-(2V– 5V)adenyladenosine oligonucleotides, Proc. Natl. Acad. Sci. U. S. A. 79 (1982) 4742 – 4746. A.G. Hovanessian, R.E. Brown, E.M. Martin, W.K. Roberts, M. Kerr, I.M. Kerr, Enzymic synthesis, purification and fractionation of (2V – 5V)-oligoadenylic acid, Methods Enzymol. 79 (1981) 184 – 193. A.G. Hovanessian, Interferon-induced and double-stranded RNA-activated enzymes: a specific protein kinase and 2V,5V-oligoadenylate synthetases, J. Interferon Res. 11 (1991) 199 – 205. I.M. Kerr, R.E. Brown, A.G. Hovanessian, Nature of inhibitor of cellfree protein synthesis formed in response to interferon and doublestranded RNA, Nature 268 (1977) 540 – 542. M.A. Minks, D.K. West, S. Benvin, C. Baglioni, Structural requirements of double-stranded RNA for the activation of 2V,5V-oligo(A)po-
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
lymerase and protein kinase of interferon-treated HeLa cells, J. Biol. Chem. 254 (1979) 10180 – 10183. H. Samanta, J.P. Dougherty, P. Lengyel, Synthesis of (2V– 5V)(A)n from ATP. Characteristics of the reaction catalyzed by (2V– 5V)(A)n synthetase purified from mouse Ehrlich ascites tumor cells treated with interferon, J. Biol. Chem. 255 (1980) 9807 – 9813. Y. Mory, B. Vaks, J. Chebath, Production of two human 2V,5V-oligoadenylate synthetase enzymes in Escherichia coli, J. Interferon Res. 9 (1989) 295 – 304. S. Bandyopadhyay, A. Ghosh, S.N. Sarkar, G.C. Sen, Production and purification of recombinant 2V– 5V oligoadenylate synthetase and its mutants using the baculovirus system, Biochemistry 37 (1998) 3824 – 3830. S.N. Sarkar, S. Bandyopadhyay, A. Ghosh, G.C. Sen, Enzymatic characteristics of recombinant medium isozyme of 2V,5V-oligoadenylate synthetase, J. Biol. Chem. 274 (1999) 1848 – 1856. D. Rebouillat, A. Hovnanian, I. Marie´, A.G. Hovanessian, The 100kDa 2V,5V-oligoadenylate synthetase catalyzing preferentially the synthesis of dimeric pppA2Vp5VA molecules is composed of three homologous domains, J. Biol. Chem. 274 (1999) 1557 – 1565. H.R.B. Pelham, R.J. Jackson, An efficient mRNA-dependent translation system from reticulocyte lysates, Eur. J. Biochem. 67 (1976) 247 – 256. R.J. Suhadolnik, S.W. Li, R.W. Sobol, J.M. Varnum, 2V,5V A synthetase: allosteric activation by fructose 1,6-biphosphate, Biochem. Biophys. Res. Commun. 169 (1990) 1198 – 1203. R. Hartmann, P.L. Noerby, P.M. Martensen, P. Joergensen, M.C. James, C. Jacobsen, S.K. Moestrup, M.J. Clemens, J. Justesen, Activation of 2 – 5 oligoadenylate synthetase by single-stranded and double-stranded RNA aptamers, J. Biol. Chem. 273 (1998) 3236 – 3246. D.N. Sen Gupta, R.H. Silverman, Activation of interferon regulated dsRNA dependent enzymes by human immunodeficiency virus 1 leader RNA, Nucleic Acids Res. 17 (1989) 969 – 978. S.Y. Desai, R.C. Patel, G.C. Sen, P. Malhotra, G.D. Ghadge, B. Thimmappaya, Activation of interferon-inducible 2V– 5Voligoadenylate synthetase by adenoviral VAI RNA, J. Biol. Chem. 270 (1995) 3454 – 3461. T.W. Nilsen, P.A. Maroney, H.D. Robertson, C. Baglioni, Heterogenous nuclear RNA promotes synthesis of (2V– 5V)oligoadenylate and is cleaved by the (2V– 5V)oligoadenylate-activated endoribonuclease, Mol. Cell. Biol. 2 (1982) 154 – 160.
149
[44] H.R. Hubell, P.C. Sheetz, S.S. Iogal, I. Brodsky, K. Kariko, S.W. Li, R.J. Suhadolnik, R.W. Sobol, Heterogenous nuclear RNA from hairy cell leukemia patients activates 2V,5V-oligoadenylate synthetase, Anticancer Res. 11 (1991) 1927 – 1932. [45] D. Ferbus, J. Justesen, F. Besancßon, M.N. Thang, The 2V,5Voligoadenylate synthetase has a multifunctional 2V,5V nucleotidyl-transferase activity, Biochem. Biophys. Res. Commun. 100 (1981) 847 – 856. [46] G. Martin, W. Keller, Mutational analysis of mammalian poly(A) polymerase identifies a region for primer binding and a catalytic domain, homologous to the family X polymerases, and to other nucleotidyltransferases, EMBO J. 15 (1996) 2593 – 2603. [47] R. Hartmann, G. Walko, J. Justesen, Inhibition of 2V– 5Voligoadenylate synthetase by divalent metal ions, FEBS Lett. 507 (2001) 54 – 58. [48] J. Justesen, D. Ferbus, M.N. Thang, Elongation mechanism and substrate specificity of 2V,5V-oligoadenylate synthetase, Proc. Natl. Acad. Sci. U. S. A. 77 (1980) 4618 – 4622. [49] I. Marie´, J. Blanco, D. Rebouillat, A.G. Hovanessian, 69 kDa and 100 kDa isoforms of interferon-induced (2V– 5V)oligoadenylate synthetase exhibit differential catalytic parameters, Eur. J. Biochem. 248 (1997) 558 – 566. [50] A.G. Hovanessian, J. Svab, I. Marie´, N. Robert, S. Chamaret, A.G. Laurent, Characterization of 69- and 100-kDa forms of 2 – 5A synthetase from interferon-treated human cells, J. Biol. Chem. 263 (1988) 4945 – 4949. [51] J.C. Castelli, B.A. Hassel, K.A. Wood, X.-L. Li, K. Amemiya, M.C. Dalakas, P.F. Torrence, R.J. Youle, A study of the interferon antiviral mechanism: apoptosis activation by the 2 – 5A system, J. Exp. Med. 186 (1997) 967 – 972. [52] A. Ghosh, S.N. Sarkar, G.C. Sen, Cell growth regulatory and antiviral effects of the P69 isozyme of 2 – 5(A) synthetase, Virology 266 (2000) 319 – 328. [53] A. Ghosh, S.N. Sarkar, T.M. Rowe, G.C. Sen, A specific isozyme of 2V– 5Voligoadenylate synthetase is a dual function proapoptotic protein of the Bcl-2 family, J. Biol. Chem. 276 (2001) 25447 – 25455. [54] A. Tominaga, S. Saito, S. Kohno, K. Sakurai, Y. Hayakawa, R. Noyori, Antiviral effects of 2V,5Voligoadenylates (2 – 5As), and related compounds, Microbiol. Immunol. 34 (1990) 737 – 747. [55] J. Sperling, J. Chebath, H. Arad-Dann, D. Offen, P. Spann, R. Lehrer, D. Goldblatt, B. Jolles, R. Sperling, Possible involvement of (2V– 5V)oligoadenylate synthetase activity in pre-mRNA splicing, Proc. Natl. Acad. Sci. U. S. A. 88 (1991) 10377 – 10381.
2V,5V-oligoadenylate synthetase from a lower invertebrate, the marine sponge Geodia cydonium, does not need dsRNA for its enzymatic activity Annika Lopp a,*, Anne Kuusksalu a, To˜nu Reintamm a, Werner E.G. Mu¨ller b,1, Merike Kelve a a
Laboratory of Molecular Genetics, National Institute of Chemical Physics and Biophysics, Akadeemia tee 23, 12618 Tallinn, Estonia Institut fu¨r Physiologische Chemie, Abteilung fu¨r Angewandte Molekularbiologie, Universita¨t, Duesbergweg 6, D-55099 Mainz, Germany
b
Received 6 December 2001; received in revised form 21 March 2002; accepted 11 April 2002
Abstract Recently, the presence of 2V,5V-linked oligoadenylates and a high 2V,5V-oligoadenylate synthetase activity were discovered in a lower invertebrate, the marine sponge Geodia cydonium. It has been demonstrated that mammalian 2 – 5A synthetase isozymes require a dsRNA cofactor for their enzymatic activity. Our results show that, unlike mammalian 2 – 5A synthetases, the 2 – 5A synthetase from the sponge acts in a dsRNA-independent manner in vitro. A prolonged incubation of the G. cydonium extract with a high concentration of a micrococcal nuclease had no effect on the activity of the 2 – 5A synthetase. At the same time, the micrococcal nuclease was effective within 30 min in degrading dsRNA needed for the enzymatic activity in IFN-induced PC12 cells. These results indicate that the 2 – 5A synthetase from G. cydonium may be active per se or is activated by some other mechanism. The sponge enzyme is capable of synthesizing a series of 2 – 5A oligomers ranging from dimers to octamers. The accumulation of a dimer in the predominant proportion during the first stage of the reaction was observed, followed by a gradual increase in longer oligoadenylates. By its product profile and kinetics of formation, the sponge 2 – 5A synthetase behaves like a specific isoform of enzymes of the 2 – 5A synthetase family. D 2002 Elsevier Science B.V. All rights reserved. Keywords: 2 – 5A synthetase; Sponge; Geodia cydonium; Enzymatic activity; dsRNA
1. Introduction Higher vertebrates have evolved a number of defense mechanisms to control viral infection. One of the most extensively characterized mechanisms among them is an interferon (IFN)-stimulated 2V,5V-linked oligoadenylate (2– 5A) pathway or a 2– 5A system (reviewed in Refs. [1,2]). By now, it has been demonstrated that the 2 – 5A system functions as an antiviral pathway in vivo in animals [3]. Other possible functions such as growth control and the control of apoptosis have also been suggested for this enzyme system [4 –7]. The key enzyme of the 2 –5A pathway, (2V– 5V) oligoadenylate synthetase (2– 5A synthetase: EC 2.7.7.-) converts cellular ATP to a series of short 2V,5V-linked oligoadenylates (2 –5A) with a general formula pppA(2Vp5VA)n, n z 1 [8]. In *
Corresponding author. Tel.: +372-6-398-352; fax: +372-6-398-382. E-mail addresses: [email protected] (A. Lopp), [email protected] (W.E.G. Mu¨ller). 1 Fax: + 49-6131-395243.
human cells, three types of IFN-induced 2– 5A synthetases have been cloned, expressed and characterized corresponding to small (p40, p46), medium (p69, p71) and large (p100) isoforms [9]. The enzymatic activity of all these isoforms requires a cofactor, a double-stranded RNA [9,10]. Their enzymatic product, 2– 5A, functions as an allosteric activator of a latent endoribonuclease (RNase L), which degrades a single-stranded viral or cellular RNA [11,12]. 2 – 5A synthetases of the sizes similar to those of the synthetases found in humans have been identified in other vertebrates (mouse, rat, pig, woodchuck, chicken) [13,14]. The 2 –5A synthetase family also comprises homologous proteins without the 2 – 5A synthesizing activity; these members are called 2– 5A synthetase-like proteins [15,16]. Contrary to the wide occurrence of the 2– 5A synthetase in vertebrates, this enzyme has not been found in lower organisms such as bacteria and C. elegans [13,14,17]. In our earlier study, we discovered the presence of 2V,5Vlinked oligoadenylates and a high 2 –5A synthetase activity in a lower invertebrate, the marine sponge Geodia cydonium [18]. The 2V,5V phosphodiester bond in the oligoadenylates
0167-4889/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 7 - 4 8 8 9 ( 0 2 ) 0 0 2 0 7 - 0
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
was verified by the NMR analysis and the oligonucleotide composition was determined by using the MALDI-MS method [19]. We also succeeded in cloning the putative 2 –5A synthetase from the marine sponge G. cydonium [20]. Although we have demonstrated the presence of the 2 – 5A synthesizing activity in the sponge G. cydonium, the role of the 2 –5A synthesis in sponges has not been defined. Sponges are known to live in symbiosis with certain bacteria and unicellular eukaryotic organisms (reviewed in Ref. [21]). Besides, as sponges are filter-feeding metazoa, they are exposed to a huge amount of organic matter via the extraction of edible material from the water [22]. Accordingly, sponges have developed an effective chemical defense system as well as humoral and cellular defense mechanisms against toxic substances or organisms [23,24]. Whether the 2 – 5A synthetase in sponges is, similar to vertebrates, involved in defense reactions against viruses or microorganisms, is not known at present. It has been reported that the 2– 5A synthetase activity in the marine sponge G. cydonium can be modulated by some environmental stressors [25]. However, in order to understand the function of the 2 – 5A synthesis in sponges, the factors needed for the activation of the 2– 5A synthetase as well as catalytic properties of the enzyme will have to be established. While searching for the possibilities of a biochemical characterization of the sponge enzyme, we found that contrary to mammalian enzymes, the sponge 2 – 5A synthetase, which was immobilized to Hybond N + membrane, did not need poly(I)poly(C) as a dsRNA cofactor for its enzymatic activity in vitro. The nuclease treatments of the G. cydonium extracts that were effective in degrading the dsRNA needed for the enzymatic activity in mammalian cells did not influence the observed 2 – 5A synthetase activity. The enzyme from G. cydonium is capable of synthesizing 2 –5A oligomers from dimers to octamers with the accumulation of the dimer in a predominant proportion during the first stage of the reaction. The enzymatic studies demonstrate that in the absence of dsRNA, the sponge 2– 5A synthetase behaves like a specific isoform of enzymes of the 2 –5A synthetase family.
2. Materials and methods 2.1. Materials Adenosine 5V-triphosphate was purchased from Reanal (Budapest, Hungary). [14C]ATP (50 –60 mCi/mmol) was purchased from Amersham and purified by HPLC. Polyethyleneimine (PEI)-cellulose TLC-Ready-Plates were from Schleicher and Schuell (Germany). The membranes Hybond N + , Hybond N, Hybond C and PVDF were from Amersham International PLC (Buckinghamshire, UK). Poly(I)poly(C), creatine kinase (EC 2.7.3.2) and phosphocreatine were
141
obtained from Boehringer Mannheim (Germany). DNA (Type I, highly polymerized, from calf thymus) and micrococcal nuclease (EC 3.1.31.1; 107 U/mg protein) were obtained from Sigma Chemical Co. (St. Louis, MO, USA). The shrimp alkaline phosphatase (SAP) was from USB Corporation (Cleveland, OH, USA). 2.2. Sponge Specimens of the marine sponge G. cydonium (Porifera, Demospongiae, Geodiidae) were collected near Rovinj (Croatia) at depths between 15 and 35 m, cut to pieces and immediately frozen in liquid nitrogen. The frozen pieces were stored at 70 jC. 2.3. Cell extracts The frozen sponge tissue was mechanically broken in liquid nitrogen and during homogenization, a lysis buffer was added (1.5 –2 v/w), consisting of 10 mM HEPES, pH 7.6, 90 mM KCl, 1 mM Mg-acetate, 0.5% NP-40 and 2 mM 2-mercaptoethanol. The supernatant obtained after centrifugation (10,000 g; 10 min; 4 jC) was immediately frozen in liquid nitrogen and stored at 70 jC (crude extract). The extracts of g-IFN-induced PC12 cells were prepared as described earlier [26] and stored at 70 jC. Protein concentration in cell extracts was measured by a modified Bradford method [27]. 2.4. 2 –5A synthetase activity assay The 2 –5A synthetase activity in the extracts was determined after binding the enzyme to the poly(I)poly(C)membrane as described earlier [18], or simply after binding the enzyme to the positively charged nylon membrane Hybond N + . The crude sponge extract (20 Al, 200 –300 Ag of total protein) or IFN-induced PC12 cell extract (20 Al, 200 – 300 Ag of total protein) was added to a 0.16 cm2 membrane piece. The binding was performed in microtiter plate wells at room temperature by gently shaking for 30 min. The membrane was washed for 4 5 min with 200 Al of the reaction buffer (without ATP) and shortly dried on the filter paper. Then, 50 Al (10 Al for radioactive assays) of the reaction mixture (30 mM Tris – HCl, pH 8.0, 100 mM KCl, 5 mM MgCl2, containing 1 mM ATP as a substrate and 20,000 cpm of [14C]ATP-omitted in nonradioactive assays) was applied to each piece of the membrane. Using different batches of crude extracts, we obtained membranes with different 2 – 5A synthesizing activities (approximately 10 – 75% of ATP converted into its products per hour). Thus, each series of experiments was performed with the same batch of the extract. In activity assays ‘‘in solution,’’ we incubated the reaction mixture which contained 2 Al of the sponge extract (30 Ag of the total protein) in a final volume of 50 Al at 37 jC in the presence or absence of 100 Ag/ml of poly (I)poly(C).
142
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
Contrary to the crude G. cydonium extracts, the PC12 cell extracts contain a remarkable ATPase activity that competes with the 2 – 5A synthetase for the substrate. Therefore, to avoid the degradation of ATP through the phosphatase activity, the ATP regenerating system (0.5 mg/ ml creatine kinase, 30 mM phosphocreatine) was added to the reaction mixture in these experiments. After 0.5– 4 h, depending on the rate of conversion of ATP to its products, the reaction was terminated. For that, the reaction mixture was heated for 5 min at 95 jC and the products were analysed in their triphosphorylated form. Alternatively, to analyse the products in their ‘‘core’’ form the reaction mixture was treated with the SAP (0.04 U/Al, 1 –1.5 h at 37 jC) to remove the 5V phosphate groups followed by the inactivation of the enzyme for 15 min at 65 jC. 2.5. TLC separation of the products In radioactive assays, the reaction products were separated by TLC on PEI-cellulose [18,25]. As a mobile phase 0.4 M Tris –HCl (pH 8.6) containing 30 mM MgCl2 were used. The dried TLC plates were exposed to a CS-imaging screen and scanned by using the Bio-Rad GS-525 Molecular Imager System. The amounts of ATP and 2 –5A oligomers were quantified by the relative intensities of the corresponding spot areas on the autoradiograms. 2.6. HPLC analysis of the reaction products The reaction products were analysed on the HPLC apparatus (Waters) equipped with a C18 reverse-phase column (Supelcosilk LC-18, 300 4.6 mm, 5 Am). The 2V– 5V-linked triphosphorylated oligoadenylates were separated in the system which consisted of solvent A (50 mM NH4H2PO4, pH 7.0) and solvent B (MeOH/H2O, 1:1), by applying a 1 – 60% gradient of solvent B in solvent A for 30 min. The absorbance was measured at 254 nm. The dephosphorylated oligoadenylates (the ‘‘core’’ form of oligomers) were separated in the system which consisted of solvent A (0.1 M triethylammoniumacetate (TEAA), pH 7.0) and solvent B (40% acetonitrile/0.1 M TEAA, pH 7.0), by applying a 10 – 30% gradient of solvent B in solvent A for 30 min. The absorbance was measured at 265 nm. The retention times of ATP, adenosine and 2– 5A oligoadenylates, in either their triphosphorylated or ‘‘core’’ form were estimated by comparing them with those of authentic compounds and individual chemically synthesized 2 – 5A oligoadenylates. The structure of 2V,5V-linked oligoadenylates has been previously verified by enzymatic (ribonuclease U2 and alkaline phosphatase) treatments and by a biological assay in rabbit reticulocyte lysates [18] and also directly by NMR analysis and the MALDI-MS method [19]. The quantification of the products was performed by measuring the relative peak areas (Millenium32 version 3.05 software, Waters Corporation). In calculations, the molar absorption
coefficients of 1.54 10 4 , 2.58 10 4 , 3.60 10 4 , 4.16 104, and 5.0 104 were used for ATP, 2V,5V-linked A2, A3, A4, and A5, respectively [28]. 2.7. Nuclease treatments Micrococcal nuclease treatments were performed in ‘‘crude extracts’’ as well as on the membrane-bound material from the sponge extract. As a control, the IFN-induced PC12 cell extract was used as a dsRNA-dependent mammalian model system of the 2 –5A formation. 2.7.1. Treatment of the sponge material For the hydrolysis of nucleic acids in crude extracts, 95 Al of the crude extract were incubated at 37 jC in the presence of 10 mM Tris – HCl buffer, pH 8.6, 10 mM CaCl2 and 0.05 U/Al of the micrococcal nuclease (final volume 100 Al) for various periods of time (0 to 22 h). Then the extract was used to determine the Hybond N + -bound 2 – 5A synthetase activity or the 2 – 5A synthetase activity ‘‘in solution’’ as described above. To study the effect of the nuclease on the Hybond N + bound material, the membrane was subsequently washed and incubated with 100 Al of 100 mM glycine buffer (pH 8.6) containing 10 mM CaCl2, in the presence or absence of 0.05 U/Al of the micrococcal nuclease at room temperature for 30 min. After the washing step, the 2 –5A synthetase activity was determined. To follow the nucleic acid (endogenous or added) degradation in the extracts, the time-dependent increase in the absorbance at 260 nm was monitored. For that, the extracts were supplemented with 10 mM Tris –HCl, pH 8.6 and 10 mM CaCl2 in either the presence or absence of 0.05 U/Al of the micrococcal nuclease. 2.7.2. Experiments with PC12 cells To study the effect of the nuclease on dsRNA, the reaction mixture, which contained 1 mM ATP, 10 Ag/ml or 100 Ag/ml of poly(I)poly(C) and 2.5 mM CaCl2 in the reaction buffer, pH 8.0, was incubated for 30 min at 30 jC in the presence or absence of 0.05 U/Al of the micrococcal nuclease. The preincubated mixture was added to the Hybond N + -bound enzyme from the PC12 cell extracts in order to carry out the reaction. Alternatively, the poly(I)poly(C)-membrane was treated with the micrococcal nuclease as described and, after washing the membrane, the enzyme from the PC12 cell extract was bound to it and the 2 – 5A synthetase activity measured. In another series of experiments, the effect of the nuclease was studied after the formation of the 2 – 5A synthetase – dsRNA complex. The enzyme from PC12 cells was bound to the poly(I)poly(C)-membrane, the membrane was washed and treated with the micrococcal nuclease in the glycine buffer as described above. After washing the membrane, the 2 –5A synthetase activity assay was performed. Alternatively, the enzyme from PC12 cells was bound to the
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
143
membranes (Hybond C, Hybond N, PVDF) also gave positive results, suggesting that the protein binding was nonspecific. The crude extracts of G. cydonium may contain possible dsRNA activators of the 2 – 5A synthetase which bind to the Hybond N + membrane together with the enzyme and activate it. Therefore, in order to carry out the hydrolysis of nucleic acids, we undertook the treatment of the extracts with the micrococcal nuclease. 3.2. The nuclease treatment of sponge extracts
Fig. 1. Formation of the 2 – 5A reaction products by poly(I)poly(C)membrane-bound and Hybond N + membrane-bound material from G. cydonium extracts. The membrane-bound material was subjected to activity assay as described in Materials and methods. The products formed from 1 mM ATP during a 4-h synthesis were dephosphorylated and analyzed by HPLC. The peaks were identified on the basis of the retention times measured for authentic compounds and the products were quantified by relative peak areas. (A) Retention time: adenosine (peak 1)—6.9 min, (2 – 5) A2 (peak 2)—11.6 min, (2 – 5) A3 (peak 3)—15.5 min, (2 – 5) A4 (peak 4)—17.6 min, (2 – 5) A5 (peak 5)—18.9 min, (2 – 5) A6 (peak 6)—20.8 min. (B) Amounts of products (mean F SE; n = 3) by using the Hybond N + membrane (peaks 1,2 and 3) or poly(I)poly(C)membrane (peaks 1V, 2V and 3V).
Hybond N + -membrane followed by the sequential binding of the poly(I)poly(C), micrococcal nuclease treatment and the addition of the reaction mixture. Between all the steps, the membrane was washed. The reaction products were subjected to HPLC analysis.
3. Results
To follow the degradation of nucleic acids in the extracts, the micrococcal nuclease was added to the extract and an increase in the absorbance was registered at 260 nm. A sharp rise in A260 was observed during the first 30 min of incubation, followed by a subsequent slower increase (not shown). In the presence of the micrococcal nuclease the total increase in the absorbance was twice higher than in the absence of the nuclease, the latter, in its turn, was evidently caused by endogenous nucleases. These results confirm the occurrence of hydrolysis of nucleic acids in the extracts. To ensure complete hydrolysis of endogenous nucleic acids, the crude extracts of G. cydonium were incubated with the micrococcal nuclease for prolonged periods of time (up to overnight) followed by the 2 –5A synthetase activity assay. The results presented in Table 1 demonstrate that the pretreatment of the extract with a high concentration of the nuclease (see: Materials and methods) had no effect on the formation of 2V,5V-oligoadenylates in any of the preincubation time periods studied. The micrococcal nuclease also remained active in the preincubated extracts as proven by a characteristic increase in A260 of the added DNA (not shown). Moreover, the results show a high stability of the 2 –5A synthetase in the pretreatment conditions (Table 1).
3.1. 2 –5A synthetase activity test with sponge extracts using poly(I)poly(C)membrane-bound and Hybond N+-bound enzyme assays The poly(I)poly(C)membrane-bound enzyme activity test has been typically used to study the 2 –5A synthetase activity in mammalian cell extracts [29]. As shown in our earlier reports [18,19], the enzyme from G. cydonium was also capable of synthesizing 2V,5V-linked oligoadenylates when immobilized to the poly(I)poly(C)membrane. As a control of the binding specificity, we used a positively charged nylon membrane (Hybond N + ) instead of the poly(I)poly(C)membrane which was prepared on the basis of Hybond N + . Surprisingly, the use of the Hybond N + membrane also enabled the synthesis of 2 –5A oligomers. Analysis of the reaction products showed no difference between the profiles of the products (Fig. 1A) and the yields of 2– 5A oligomers (Fig. 1B), regardless of whether the poly(I)poly(C)membrane or Hybond N + membrane was used for the immobilization of the enzyme. The use of other
Table 1 The effect of the nuclease treatment of G. cydonium extract on the formation of 2V,5V-linked oligomers from ATP Preincubation time (h)
Micrococcal nuclease
0 + 1 + 3 + 22 +
Products (nmol/assay) p3A2
p3A3
p3A4
5.9 F 0.3a 5.6 F 0.5 6.3 F 0.4 6.3 F 0.3 6.0 F 0.5 6.4 F 0.6 5.6 F 0.4 6.0 F 0.4
1.3 F 0.09 1.2 F 0.10 1.4 F 0.08 1.6 F 0.07 1.5 F 0.10 1.8 F 0.10 1.1 F 0.06 1.3 F 0.08
0.22 F 0.02 0.25 F 0.02 0.28 F 0.02 0.34 F 0.01 0.33 F 0.03 0.46 F 0.03 0.19 F 0.01 0.25 F 0.02
The sponge extracts were incubated in the presence or absence of 0.05 U/Al of the micrococcal nuclease for indicated periods of time and the Hybond N + -bound material was subjected to activity assay as described in Materials and methods. The triphosphorylated products, formed from 1 mM ATP during a 2-h synthesis, were analyzed by the HPLC method. a The data represent mean F SE (n = 3).
144
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
Table 2 The effect of poly(I)poly(C) on the formation of 2V,5V-linked oligomers by the 2 – 5A synthetase from nuclease-treated G. cydonium extract Micrococcal nuclease
Poly(I)poly(C) (100 Ag/ml)
Products (nmol/assay) p3A2 a
+ + +
+
7.0 F 0.4 7.1 F 0.3 7.2 F 0.2 6.9 F 0.4
p3A3
p3A4
1.1 F 0.06 1.0 F 0.04 1.1 F 0.03 1.0 F 0.05
0.25 F 0.02 0.26 F 0.03 0.25 F 0.01 0.22 F 0.03
The sponge extracts were incubated overnight in the presence or absence of 0.05 U/Al of the micrococcal nuclease, thereafter the extract was used to carry out the synthesis of 2 – 5A oligomers ‘‘in solution’’ as described in Materials and methods. The triphosphorylated products, formed from 1 mM ATP during a 2-h synthesis, were analyzed by the HPLC method. a The data represent mean F SE (n = 3).
Thus, the degradation of nucleic acids in crude extracts did not influence the activity of the 2– 5A synthetase from the sponge G. cydonium. In order to rule out the fact that the binding to the Hybond N + membrane could cause the possible structural changes leading to the activation of the enzyme, the assay was carried out ‘‘in solution.’’ For that, the crude extracts were incubated overnight in the presence or absence of the micrococcal nuclease and then used for the determination of
Fig. 2. The effect of the micrococcal nuclease on the poly(I)poly(C)induced formation of 2 – 5A oligomers by the 2 – 5A synthetase from IFNinduced PC12 cells. The Hybond N + -bound enzyme from PC12 cell extracts was subjected to activity assay as described in Materials and methods. The HPLC profile of the products formed from 1 mM ATP during a 2-h synthesis in the absence of poly(I)poly(C) (A), in the presence of 10 and 100 Ag/ml of poly(I)poly(C), (B) and (C), respectively, or in the presence of 100 Ag/ml poly(I)poly(C) pretreated for 30 min with 0.05 U/Al of the micrococcal nuclease (D). Peaks 1, 2, 3 and 4 correspond to ATP, p3A2, p3A3 and p3A4, respectively.
2 – 5A synthetase activity either in the presence or absence of poly (I)poly(C). The results in Table 2 reveal that in all cases, the synthesis of 2 –5A oligomers remained the same. 3.3. Comparative studies in a dsRNA-dependent mammalian cell model system The effect of the micrococcal nuclease on the poly (I)poly(C)-activated formation of 2V,5V-linked oligoadenylates was studied in a mammalian cell model. For this purpose, the extracts of g-IFN-induced PC12 cells were used. When these extracts were applied to the Hybond N + membrane and the membrane-bound enzyme was used to carry out the 2– 5A synthesis, no products were formed (Fig. 2A). The presence of 10 Ag/ml of poly(I)poly(C) in the reaction mixture was sufficient to activate the Hybond N + -bound enzyme from PC12 cells (Fig. 2B). The formation of 2 –5A oligomers, mainly trimers and tetramers, was even more elevated in the presence of 100 Ag/ml of poly(I)poly(C) (Fig. 2C). On the contrary, no synthesis of 2– 5A oligomers was observed when the reaction mixture containing dsRNA was treated for 30 min with the micrococcal nuclease and
Fig. 3. The micrococcal nuclease treatment of the membrane-bound mammalian enzyme – poly(I)poly(C) complex and the membrane-bound material from G. cydonium extract. The 2 – 5A synthetase from IFNinduced PC12 cells or G. cydonium was bound to the Hybond N + membrane followed by the binding of poly(I)poly(C) (omitted in the case of sponge extracts), the nuclease treatment of the membrane and activity assay as described in Materials and methods. The HPLC profile of the products formed from 1 mM ATP during a 2-h synthesis by the mammalian (A, C) or during a 0.5-h synthesis by the sponge 2 – 5A synthetase (B, D) when the membrane-bound material was treated for 30 min in the absence (A, B) or presence (C, D) of 0.05 U/Al of the micrococcal nuclease. Peaks 1, 2, 3 and 4 correspond to ATP, p3A2, p3A3 and p3A4, respectively.
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
subsequently applied to the membrane-bound enzyme (Fig. 2D). Alternatively, the poly(I)poly(C)membrane was incubated in the presence of the micrococcal nuclease before binding the 2 –5A synthetase to it. Under these conditions, the nuclease treatment also abolished the synthesis of 2– 5A oligomers (not shown). These results demonstrate that 0.05 U/Al of the micrococcal nuclease was effective enough for degrading poly(I)poly(C) required for the enzyme activity in mammalian PC12 cells. Further, the effect of the nuclease on the synthesis of 2 – 5A was studied after binding the 2 – 5A synthetase to poly(I)poly(C), i.e. after the formation of the enzyme – dsRNA complex. For that, the enzyme from the PC12 cells was bound to the Hybond N + membrane followed by the binding of poly(I)poly(C). The micrococcal nuclease treatment of such a membrane abolished the synthesis of the products (Fig. 3A and C). The same result was obtained using the poly(I)poly(C)membrane-bound enzyme from PC12 cells. In contrast, when the Hybond N + -bound material from G. cydonium extracts was used in the assay, the nuclease treatment had no effect on the synthesis of 2 – 5A products (Fig. 3B and D).
145
3.4. A biochemical characterization of the 2 –5A oligomer formation by the 2 – 5A synthetase from G. cydonium On the basis of our finding that the formation of 2 –5A oligomers by the sponge 2 – 5A synthetase was poly (I)poly(C)-independent, all subsequent experiments were performed using the Hybond N + -bound enzyme without an additional dsRNA. The requirement for divalent cations, an optimal temperature and the pH value were determined to optimize the rate of formation of 2 –5A oligoadenylates. Fig. 4A demonstrates the yields of ATP oligomerization at different temperatures. It can be seen that the enzyme is active in quite a wide range of temperatures used (10 – 70 jC). The formation of 2– 5A oligomers was most effective in the temperature range of 25 –45 jC with an optimum at about 37 jC. Fig. 4B demonstrates the dependence of the formation of 2 –5A oligomers on pH of the assay buffer. It appears that for the maximal activity, basic pHs with an optimum at about 8 are needed. The quantities of 2 – 5A oligomers formed at different concentrations of MgCl2 are presented in Fig. 4C. The results show that the maximal rate of
Fig. 4. Biochemical characteristics of the formation of 2 – 5A oligomers by the Hybond N + -bound enzyme from G. cydonium extracts. (A) Temperaturedependence. After the incubation of 1 mM ATP+[14C]ATP with membrane-bound enzyme for 4 h, the reaction products were separated by TLC on PEIcellulose as described in Materials and methods. The quantification of the products on the autoradiograms was performed by the relative spot intensities. Each point represents a mean F SE (n = 4). (B) pH-dependence. After the incubation of 1 mM ATP with membrane-bound enzyme for 0.5 h at an indicated pH of the reaction mixture, the triphosphorylated products were analyzed by HPLC as described in Materials and methods. Each point represents a mean of the two independent experiments. (C) The effect of MgCl2. The products formed from 1 mM ATP at an increasing concentration of MgCl2 during a 4-h synthesis were dephosphorylated and analyzed by the HPLC method. The values represent means of the two independent experiments. (D) The time-course of formation of 2 – 5A oligomers. ATP (0.63 mM) was incubated with membrane-bound enzyme for indicated periods of time and the triphosphorylated products were analyzed by HPLC and quantified as described in Materials and methods.
146
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
formation of the products from 1 mM ATP was observed at 5 mM MgCl2, while no activity of the enzyme was detected in the absence of MgCl2 (Fig. 4C). In a separate series of experiments, we studied the ability of different divalent metal ions (Mn2 + , Zn2 + , Ca2 + ) to substitute Mg2 + in the oligomerization reaction of ATP. The results show that the potency of different divalent ions to activate the sponge synthetase decreased in the following order: Mg > Mn>Zn resulting under these conditions in the conversion of 96%, 85% and 38% of ATP, respectively (Fig. 5A – C). The products were not formed in the presence of CaCl2 (Fig. 5D). Mammalian 2– 5A synthetases have a characteristic pattern of 2– 5A oligomers synthesized by each protein [9,10]. Looking at the pattern of the oligomers formed by the sponge synthetase, we found a higher proportion of the 2 –5A dimer compared to that of the longer products (Fig. 5A). Fig. 4D shows the time-course of formation of 2V,5Vlinked di-, tri-, tetra-, and penta-adenylates from ATP. It demonstrates that the disappearance of ATP was quantitatively related to the appearance of oligonucleotides. During the first stage of the reaction, when up to 50% of ATP was used, the dinucleotide was prevalent but the ratio of dimer to longer oligomers decreased over time. Nevertheless, the molar ratio of dimer to total products was about 0.5 when approximately 90% of ATP was converted to its products (Fig. 4D).
Fig. 5. The effect of different divalent metal ions on the formation of products by the 2 – 5A synthetase from G. cydonium. The HPLC profile of the products formed from 1 mM ATP during a 3-h synthesis in the presence of 5 mM MgCl2 (A), 5 mM MnCl2 (B), 5 mM Zn-acetate (C) or 5 mM CaCl2 (D) in the reaction mixture, pH 7.6. Peaks 1, 2, 3 and 4 correspond to ATP, p3A2, p3A3 and p3A4, respectively. The group of peaks, designated as ‘‘5,’’ contains p3A5 and oligomers of longer chain length.
4. Discussion Mammalian 2– 5A synthetases were discovered as IFNinduced and dsRNA-activated enzymes (reviewed in Ref. [30]). Their dependence on dsRNA was proven by means of the enzymes purified from cell extracts [31 –33] and, later by means of recombinant proteins [34 –37]. In our earlier study, we discovered that the enzyme from the crude extract of a lower invertebrate, the marine sponge G. cydonium, also has a high 2 –5A synthetase activity when immobilized to the poly(I)poly(C)membrane [18]. Some years ago, we succeeded in cloning the putative 2 –5A synthetase from the sponge G. cydonium, which shows some, although limited, homology to small 2– 5A synthetase isoforms of mouse, chicken and human [20]. However, up to now, the expression of this cDNA clone in neither the bacterial nor insect cells has resulted in the active protein (Kuusksalu, A., unpublished data). It is not clear whether the polypeptide obtained is responsible for the observed 2 –5A synthetase activity in G. cydonium. Therefore, to further study the parameters needed for the 2 –5A synthetase activity in the sponge, we used the membrane-bound enzyme preparations from the crude extracts of G. cydonium in the present study. We found that the Hybond N + membrane as well as other membranes (Hybond C, Hybond N, PVDF) are capable of binding the 2 – 5A synthetase from the crude extracts of G. cydonium nonspecifically and this membrane can be used to carry out the synthesis of 2 – 5A without an additional dsRNA (Fig. 1). The 2 –5A synthetase from IFN-induced PC12 cells, used by us as a mammalian cell model, is also capable of nonspecific binding to Hybond N + , but no activity of the enzyme is observed in the absence of dsRNA. As expected, the 2– 5A synthesis can be induced in this case in a dose-dependent manner by adding poly(I)poly(C) as a dsRNA cofactor to the reaction mixture (Fig. 2A – C). Provided that the sponge 2 –5A synthetase behaves like mammalian enzymes, it is conceivable that the crude extract of G. cydonium contains dsRNA activators which bind together with the protein to the membrane and activate the 2 – 5A synthetase. To exclude this possibility, the sponge extracts were treated with the micrococcal nuclease, an enzyme used to hydrolyse single-stranded and doublestranded nucleic acids in crude cell-free extracts [38,39]. Our results in Table 1 reveal that the incubation of the extracts with the nuclease for prolonged periods of time (up to overnight) did not influence the activity of the sponge 2– 5A synthetase. On the contrary, the same concentration of the micrococcal nuclease was, within 30 min of incubation, effective in degrading poly(I)poly(C) required for the enzyme activity in PC12 cells (Fig. 2C and D). In another series of experiments with IFN-induced PC12 cells, we studied the effect of the nuclease on poly(I)poly(C) in the presence of the protein bound to it. We showed that the nuclease treatment of the 2 – 5A synthetase – poly(I)poly(C) complex also abolished the synthesis of 2 – 5A oligomers (Fig. 3A and C). Once again, the membrane-bound 2 –5A
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
synthetase from the sponge exhibited similar activity, irrespective of whether it had been subjected to the nuclease treatment (Fig. 3B and D). To ensure that binding of the sponge 2 –5A synthetase to the membrane would not bring about any structural modifications of the enzyme affecting its activation process, the formation of 2 – 5A oligomers was measured also ‘‘in solution.’’ The data in Table 2 show that the 2– 5A synthetase activity was the same independent of the micrococcal nuclease treatment and the presence or absence of poly (I)poly(C) in the reaction mixture. These data together with the abovementioned results suggest that the 2– 5A synthetase from G. cydonium, differently from mammalian 2– 5A synthetases, does not need dsRNA for its activity. The exact nature of interaction of an RNA activator with an RNA-binding domain in the proteins of the 2 – 5A synthetase family is not known [10,40]. Moreover, none of the proteins of this family isolated so far possesses a typical RNA-binding domain which is characteristic of many dsRNA-binding proteins (reviewed in Ref. [13]). The 2– 5A oligoadenylate synthetase, which has previously been believed to be activated only by dsRNA, can also be activated by RNA ligands with a little secondary structure [40]. Concerning natural activators of the enzyme, certain small viral RNAs have been shown to act as activators upon viral infection [41,42]. At the same time, little is known about cellular activators of 2– 5A synthetases in virus-noninfected cells. With regard to the latter, the activation of the 2 –5A synthetase by certain nuclear dsRNAs (HnRNA) has been proposed [43,44]. The only report that relates to the naturally occurring, non-dsRNA cellular activators of the enzyme demonstrates that high concentrations of fructose1,6-biphosphate can activate the 2– 5A synthetase [39]. There are several possibilities to explain our results. The corresponding sponge protein may be active per se or in the form of a hypothetical strong protein – nucleic acid complex which is resistant to the action of nucleases. Alternatively, it may be activated by a yet unknown component present in the sponge extract. Further experiments will be required to elucidate possible activators (if any) of the sponge 2– 5A synthetase. The optimization studies of the synthesis of 2 – 5A oligomers revealed that basic pH values with an optimum of 8.0, the presence of MgCl2 in excess with respect to ATP and the temperature optimum of 37 jC are necessary for the maximal rate of formation of products by the sponge 2– 5A synthetase (Fig. 4A – C). It has been established that mammalian 2 –5A synthetases belong to the class of nucleotidyltransferases which require divalent metal ions for their activity [45]. The presence of Mg2 + has been shown to be obligatory for the activity of mammalian 2 –5A synthetases [29]. Our results confirm that it is also valid for the sponge synthetase. Other divalent cations (Mn2 + , Zn2 + ), which may be chelated by carboxylate residues in the catalytic center of nucleotidyltransferases [46], were able to substitute Mg2 + with a
147
lower efficiency (Fig. 5). Several metal ions (copper, ion, zinc) have been shown to be strongly inhibitory to the small and medium human 2 – 5A synthetases [47]. The effects of divalent cations on the activity of the sponge 2– 5A synthetase deserve further investigation, since we observed a better formation of 2– 5A oligomers in the presence of zinc than in the presence of calcium. We have previously shown by the MALDI-MS analysis that the 2 – 5A synthetase from G. cydonium is capable of synthesizing oligoadenylates from dimers to octamers [19]. The kinetic studies of formation of different oligomers show that the 2– 5A dimer is the first product to accumulate in a predominant proportion during the first stage of the reaction. At a higher conversion percentage of ATP, the yields of the trimer and other longer oligomers gradually increase (Fig. 4D). However, the molar concentrations of oligomers higher than a pentamer remain low. The mammalian 2– 5A synthetases are characterized by different oligomer patterns and kinetic parameters of formation of 2 –5A oligomers [9,10]. The kinetics of formation of the products by the sponge synthetase (Fig. 4D) is similar to that observed by Justesen et al. [48] who carried out kinetic studies with the 2– 5A synthetase purified from a rabbit reticulocyte lysate. When 90% of ATP was converted to its products, we found the calculated molar ratio of dimer to total products to be about 0.5 vs. 0.41 calculated in experiments with the enzyme from rabbit reticulocytes [48]. The high ratio of dimer to total products is also typical of a large human isozyme p100 [49], whereas the medium isozyme p69 is quite different. Hovanessian et al. [50] have shown that at a maximum activity of each enzyme, the ratio of dimer to other oligoadenylates is 0.49 and 0.08 for the large human 2 – 5A synthetase isozyme p100 and medium isozyme p69, respectively. Taken together, our enzymatic studies demonstrate that the 2 – 5A synthetase from the sponge G. cydonium behaves like a specific isoform of enzymes of the 2 –5A synthetase family in the absence of dsRNA. The cellular abundance of specific 2 – 5A synthetase isoforms and their localization in different subcellular compartments may refer to the multiplicity of functions they exert in cells [51 – 53]. Although the general scheme for the functioning of the 2 –5A synthetase/RNase L system is well established, other mechanisms of actions of specific isozymes in mammalian cells cannot be ruled out. Lately, it has been demonstrated that besides its oligoadenylate synthesizing activity, the murine 2 – 5A synthetase isoform 9 – 2 functions as a proapoptotic protein of the Bcl-2 family in an RNase L-independent way [53]. Since 2V,5V-diadenylates have a low affinity to bind and thus activate the RNase L [49], it has been proposed that they have an independent antiviral activity or some other cellular effects [9,37,54]. Also, the potential role of the large human isozyme p100 in pre-RNA splicing through generating the 2V,5V-phoshodiester bond in the intermediate lariat structure has been proposed [55].
148
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
At present, we do not know the role of 2 –5A synthesis in the sponges. By now, the 2 – 5A synthetase as a single component of the whole mammalian 2– 5A system has been identified in sponges. Whether an interferon-like antimicrobial system exists in sponges is a matter of future studies. Thus, the specific need for 2– 5A products as extracellular or intracellular signalling molecules for sponges has to be established. The results of the present study give evidence of the dsRNA-independent synthesis of 2 – 5A oligomers. The preferential synthesis of dimeric 2 –5A molecules refers to the mode of action of the oligoadenylates formed which may not involve the RNase L pathway.
[13]
[14]
[15]
[16]
[17]
Acknowledgements This study was supported by the European Commission (Project Sponge) and the Estonian Science Foundation (grant No. 4221).
References [1] I.M. Kerr, The 2 – 5A system: a personal view, J. Interferon Res. 7 (1987) 505 – 510. [2] M.R. Player, P.F. Torrence, The 2 – 5A system: modulation of viral and cellular processes through acceleration of RNA degradation, Pharmacol. Ther. 78 (1998) 55 – 113. [3] A. Zhou, J. Paranjape, T.L. Brown, H. Nie, S. Naik, B. Dong, A. Chang, B. Trapp, R. Fairchild, C. Colmenares, R.H. Silverman, Interferon action and apoptosis are defective in mice devoid of 2V,5V-oligoadenylate-dependent RNase L, EMBO J. 16 (1997) 6355 – 6363. [4] G. Rysiecki, D.R. Gewert, B.R.G. Williams, Constitutive expression of a 2V,5V-oligoadenylate synthetase cDNA results in increased antiviral activity and growth suppression, J. Interferon Res. 4 (1989) 649 – 657. [5] M. Diaz-Guerra, C. Rivas, M. Esteban, Activation of the IFN-inducible enzyme RNase L causes apoptosis of animal cells, Virology 236 (1997) 354 – 363. [6] S. Salzberg, T. Hyman, H. Turm, Y. Kinar, Y. Schwartz, U. Nir, F. Lejbkowicz, E. Huberman, Ectopic expression of 2 – 5A synthetase in myeloid cells induces growth arrest and facilitates the appearance of a myeloid differentiation marker, Cancer Res. 57 (1997) 2732 – 2740. [7] J.C. Castelli, B.A. Hassel, A. Maran, P. Paranjape, J.A. Hewitt, X.-I. Li, Y.-T. Hsu, R.H. Silverman, R.J. Youle, The role of 2V– 5Voligoadenylate-activated ribonuclease L in apoptosis, Cell Death Differ. 5 (1998) 313 – 320. [8] I.M. Kerr, R.E. Brown, pppA2Vp5VA2Vp5VA: an inhibitor of protein synthesis synthesized with an enzyme fraction from interferon-treated cells, Proc. Natl. Acad. Sci. U. S. A. 75 (1978) 256 – 280. [9] D. Rebouillat, A.G. Hovanessian, The human 2V,5V-oligoadenylate synthetase family: interferon-induced proteins with unique enzymatic properties, J. Interferon Cytokine Res. 19 (1999) 295 – 308. [10] S.N. Sarkar, G.C. Sen, Production, purification and characterization of recombinant 2V,5V-oligoadenylate synthetases, METHODS: a companion to Methods in Enzymology 15 (1998) 233 – 242. [11] G. Floyd-Smith, E. Slattery, P. Lengyel, Interferon action: RNA cleavage pattern of a (2V– 5V) oligoadenylate-dependent endonuclease, Science 212 (1981) 1020 – 1032. [12] A. Zhou, B.A. Hassel, R.H. Silverman, Expression cloning of 2 – 5A-
[18]
[19]
[20]
[21] [22] [23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
dependent RNAase: a uniquely regulated mediator of interferon action, Cell 72 (1993) 753 – 765. J. Justesen, R. Hartmann, N.O. Kjeldgaard, Gene structure and function of the 2V,5V-oligoadenylate synthetase family, Cell. Mol. Life Sci. 57 (2000) 1593 – 1612. S. Kumar, C. Mitnik, G. Valente, G. Floyd-Smith, Expansion and molecular evolution of the interferon-induced 2V– 5V oligoadenylate synthetase gene family, Mol. Biol. Evol. 17 (2000) 738 – 750. R. Hartmann, H.S. Olsen, S. Widder, R. Joergensen, J. Justesen, p59OASL, a 2V– 5V oligoadenylate synthetase like protein: a novel human gene related to the 2V– 5V oligoadenylate synthetase family, Nucleic Acids Res. 26 (1998) 4121 – 4127. D. Rebouillat, I. Marie´, A.G. Hovanessian, Molecular cloning and characterization of two related and interferon-induced 56-kDa and 30-kDa proteins highly similar to 2V– 5V oligoadenylate synthetase, Eur. J. Biochem. 257 (1998) 319 – 330. P.J. Cayley, R.F. White, J.F. Antoniw, N.J. Walesby, I.M. Kerr, Distribution of the ppp(A2Vp)nA-binding protein and interferon-related enzymes in animals, plants, and lower organisms, Biochem. Biophys. Res. Commun. 108 (1982) 1243 – 1250. A. Kuusksalu, A. Pihlak, W.E.G. Mu¨ller, M. Kelve, The (2V– 5V) oligoadenylate synthetase is present in the lowest multicellular organisms, the marine sponges, Eur. J. Biochem. 232 (1995) 351 – 357. A. Kuusksalu, J. Subbi, T. Pehk, T. Reintamm, W.E.G. Mu¨ller, M. Kelve, Identification of the reaction products of (2V– 5V) oligoadenylate synthetase in the marine sponge, Eur. J. Biochem. 257 (1998) 420 – 426. M. Wiens, A. Kuusksalu, M. Kelve, W.E.G. Mu¨ller, Origin of the interferon-inducible (2V– 5V)oligoadenylate synthetases: cloning of the (2V– 5V)oligoadenylate synthetase from the marine sponge Geodia cydonium, FEBS Lett. 462 (1999) 12 – 18. T.L. Simpson, The Cell Biology of Sponges, Springer-Verlag, New York, 1984. S. Vogel, Current-induced flow through living sponges in nature, Proc. Natl. Acad. Sci. U. S. A. 74 (1977) 2069 – 2071. P. Proksch, Defensive role for secondary metabolites from marine sponges and sponge-feeding nudibranchs, Toxicon 32 (1994) 639 – 655. W.E.G. Mu¨ller, B. Blumbach, I.M. Mu¨ller, Evolution of the innate and adaptive immune systems: relationships between potential immune molecules in the lowest metazoan phylum (Porifera) and those in vertebrates, Transplantation 68 (1999) 1215 – 1227. H.C. Schro¨der, M. Wiens, A. Kuusksalu, M. Kelve, W.E.G. Mu¨ller, Modulation of 2V– 5Voligoadenylate synthetase by environmental stress in the marine sponge Geodia cydonium, Environ. Toxicol. Chem. 16 (1997) 1403 – 1409. A. Lopp, A. Kuusksalu, K. Samuel, M. Kelve, Expression and activity of 2 – 5A synthetase in the course of differentiation and apoptosis of PC12 cells, Cytokine 12 (2000) 737 – 741. M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976) 248 – 254. H. Cailla, C. Le Borgne De Kaouel, D. Roux, M. Delaage, J. Marti, Monoclonal antibodies to 5V-triphospho-(2V– 5V)adenyladenosine oligonucleotides, Proc. Natl. Acad. Sci. U. S. A. 79 (1982) 4742 – 4746. A.G. Hovanessian, R.E. Brown, E.M. Martin, W.K. Roberts, M. Kerr, I.M. Kerr, Enzymic synthesis, purification and fractionation of (2V – 5V)-oligoadenylic acid, Methods Enzymol. 79 (1981) 184 – 193. A.G. Hovanessian, Interferon-induced and double-stranded RNA-activated enzymes: a specific protein kinase and 2V,5V-oligoadenylate synthetases, J. Interferon Res. 11 (1991) 199 – 205. I.M. Kerr, R.E. Brown, A.G. Hovanessian, Nature of inhibitor of cellfree protein synthesis formed in response to interferon and doublestranded RNA, Nature 268 (1977) 540 – 542. M.A. Minks, D.K. West, S. Benvin, C. Baglioni, Structural requirements of double-stranded RNA for the activation of 2V,5V-oligo(A)po-
A. Lopp et al. / Biochimica et Biophysica Acta 1590 (2002) 140–149
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
lymerase and protein kinase of interferon-treated HeLa cells, J. Biol. Chem. 254 (1979) 10180 – 10183. H. Samanta, J.P. Dougherty, P. Lengyel, Synthesis of (2V– 5V)(A)n from ATP. Characteristics of the reaction catalyzed by (2V– 5V)(A)n synthetase purified from mouse Ehrlich ascites tumor cells treated with interferon, J. Biol. Chem. 255 (1980) 9807 – 9813. Y. Mory, B. Vaks, J. Chebath, Production of two human 2V,5V-oligoadenylate synthetase enzymes in Escherichia coli, J. Interferon Res. 9 (1989) 295 – 304. S. Bandyopadhyay, A. Ghosh, S.N. Sarkar, G.C. Sen, Production and purification of recombinant 2V– 5V oligoadenylate synthetase and its mutants using the baculovirus system, Biochemistry 37 (1998) 3824 – 3830. S.N. Sarkar, S. Bandyopadhyay, A. Ghosh, G.C. Sen, Enzymatic characteristics of recombinant medium isozyme of 2V,5V-oligoadenylate synthetase, J. Biol. Chem. 274 (1999) 1848 – 1856. D. Rebouillat, A. Hovnanian, I. Marie´, A.G. Hovanessian, The 100kDa 2V,5V-oligoadenylate synthetase catalyzing preferentially the synthesis of dimeric pppA2Vp5VA molecules is composed of three homologous domains, J. Biol. Chem. 274 (1999) 1557 – 1565. H.R.B. Pelham, R.J. Jackson, An efficient mRNA-dependent translation system from reticulocyte lysates, Eur. J. Biochem. 67 (1976) 247 – 256. R.J. Suhadolnik, S.W. Li, R.W. Sobol, J.M. Varnum, 2V,5V A synthetase: allosteric activation by fructose 1,6-biphosphate, Biochem. Biophys. Res. Commun. 169 (1990) 1198 – 1203. R. Hartmann, P.L. Noerby, P.M. Martensen, P. Joergensen, M.C. James, C. Jacobsen, S.K. Moestrup, M.J. Clemens, J. Justesen, Activation of 2 – 5 oligoadenylate synthetase by single-stranded and double-stranded RNA aptamers, J. Biol. Chem. 273 (1998) 3236 – 3246. D.N. Sen Gupta, R.H. Silverman, Activation of interferon regulated dsRNA dependent enzymes by human immunodeficiency virus 1 leader RNA, Nucleic Acids Res. 17 (1989) 969 – 978. S.Y. Desai, R.C. Patel, G.C. Sen, P. Malhotra, G.D. Ghadge, B. Thimmappaya, Activation of interferon-inducible 2V– 5Voligoadenylate synthetase by adenoviral VAI RNA, J. Biol. Chem. 270 (1995) 3454 – 3461. T.W. Nilsen, P.A. Maroney, H.D. Robertson, C. Baglioni, Heterogenous nuclear RNA promotes synthesis of (2V– 5V)oligoadenylate and is cleaved by the (2V– 5V)oligoadenylate-activated endoribonuclease, Mol. Cell. Biol. 2 (1982) 154 – 160.
149
[44] H.R. Hubell, P.C. Sheetz, S.S. Iogal, I. Brodsky, K. Kariko, S.W. Li, R.J. Suhadolnik, R.W. Sobol, Heterogenous nuclear RNA from hairy cell leukemia patients activates 2V,5V-oligoadenylate synthetase, Anticancer Res. 11 (1991) 1927 – 1932. [45] D. Ferbus, J. Justesen, F. Besancßon, M.N. Thang, The 2V,5Voligoadenylate synthetase has a multifunctional 2V,5V nucleotidyl-transferase activity, Biochem. Biophys. Res. Commun. 100 (1981) 847 – 856. [46] G. Martin, W. Keller, Mutational analysis of mammalian poly(A) polymerase identifies a region for primer binding and a catalytic domain, homologous to the family X polymerases, and to other nucleotidyltransferases, EMBO J. 15 (1996) 2593 – 2603. [47] R. Hartmann, G. Walko, J. Justesen, Inhibition of 2V– 5Voligoadenylate synthetase by divalent metal ions, FEBS Lett. 507 (2001) 54 – 58. [48] J. Justesen, D. Ferbus, M.N. Thang, Elongation mechanism and substrate specificity of 2V,5V-oligoadenylate synthetase, Proc. Natl. Acad. Sci. U. S. A. 77 (1980) 4618 – 4622. [49] I. Marie´, J. Blanco, D. Rebouillat, A.G. Hovanessian, 69 kDa and 100 kDa isoforms of interferon-induced (2V– 5V)oligoadenylate synthetase exhibit differential catalytic parameters, Eur. J. Biochem. 248 (1997) 558 – 566. [50] A.G. Hovanessian, J. Svab, I. Marie´, N. Robert, S. Chamaret, A.G. Laurent, Characterization of 69- and 100-kDa forms of 2 – 5A synthetase from interferon-treated human cells, J. Biol. Chem. 263 (1988) 4945 – 4949. [51] J.C. Castelli, B.A. Hassel, K.A. Wood, X.-L. Li, K. Amemiya, M.C. Dalakas, P.F. Torrence, R.J. Youle, A study of the interferon antiviral mechanism: apoptosis activation by the 2 – 5A system, J. Exp. Med. 186 (1997) 967 – 972. [52] A. Ghosh, S.N. Sarkar, G.C. Sen, Cell growth regulatory and antiviral effects of the P69 isozyme of 2 – 5(A) synthetase, Virology 266 (2000) 319 – 328. [53] A. Ghosh, S.N. Sarkar, T.M. Rowe, G.C. Sen, A specific isozyme of 2V– 5Voligoadenylate synthetase is a dual function proapoptotic protein of the Bcl-2 family, J. Biol. Chem. 276 (2001) 25447 – 25455. [54] A. Tominaga, S. Saito, S. Kohno, K. Sakurai, Y. Hayakawa, R. Noyori, Antiviral effects of 2V,5Voligoadenylates (2 – 5As), and related compounds, Microbiol. Immunol. 34 (1990) 737 – 747. [55] J. Sperling, J. Chebath, H. Arad-Dann, D. Offen, P. Spann, R. Lehrer, D. Goldblatt, B. Jolles, R. Sperling, Possible involvement of (2V– 5V)oligoadenylate synthetase activity in pre-mRNA splicing, Proc. Natl. Acad. Sci. U. S. A. 88 (1991) 10377 – 10381.