Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
HYPOTHESIS State Key Laboratory of Protein and Plant Gene Research, School of Life Sciences, Peking University, China Please cite this article as: Xinmiao Fu. Potential protein-encoded synthesis of DNA and RNA. Hypothesis 2014, 12(1): e6, doi:10.5779/hypothesis.v12i1.366 *Correspondence: [email protected] 1/5 Received: 2014/02/27; Accepted: 2014/05/12; Posted online: 2014/09/12 © 2014 Xinmiao Fu. This is an Open Access article distributed by Hypothesis under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Potential protein-encoded synthesis of DNA and RNA Xinmiao Fu* mono- and/or dinucleotides are assumed presence of antibiotics such as neomy- HYPOTHESIS Synthesis of DNA encoded to be arranged along the characteristic cin . In addition, regarding the origin of by TALE proteins TALE proteins are secret- organisms9. amino acids of TALE and PUF, and then the central dogma, it is widely believed assembled as oligonucleotides by ligase that the codon-amino acid stereochemi- cifically recognize host DNA sequences ed by bacterial plant pathogens and spe- On the other hand, the code of DNA or condensation agents. This hypothesis cal pairing occurred during early evo- via their modular DNA-binding domain of suggests a new protein-based strategy for lution and played a crucial role in the synthesizing DNA and RNA molecules. current ribosome-mediated translation INTRODUCTION The central dogma process, which does not involve the direct 3,4 ful DNA targeting tool utilized in different recognition by RVDs of TALE repeats also suggests the possibility that the tandem repeats9. Each repeat comprises sequence information of the RVDs in a around 34 conserved amino acids and TALE protein can be transferred residue targets a specific base pair by using re- by residue to DNA. Although the minimal interaction between codons and corre- peat variable diresidues (RVDs) at posinumber of repeats in an artificial TALE 10,11 sponding amino acids5,6,7,8. tions 12 and 13 (Figure 1B). Structural 1 Francis Crick in 1958 and later formuto significantly activate gene expression lated in 19702, is based on extensive In this article, I suggest the possibility that studies have revealed that multiple TALE in cells is reported to be 6.510, it is con- of molecular biology, as proposed by ABSTRACT As stated by the central dogma repeats form a superhelical structure to ceivable that mono- and dinucleotide protein can be artificially transferred resi- track along the sense strand of the DNA due by residue directly to DNA and RNA, duplex within its internal layer and that the duplexes can be arranged specifically respectively, based on transcription ac- 12th residue in each repeat stabilizes the tivator-like effectors (TALE) and Pumilio/ RVD loop while the 13th residue makes (Figure 1B) based on the following report- of molecular biology, in living organisms genetic and biochemical studies on liv- the sequence information encoded by the sequence information of DNA is trans- ing organisms and is fundamental to life ferred to RNA and then to protein, but sciences. It follows that the sequence such information cannot be transferred information of DNA can be self-replicated back from protein to nucleic acids. In this or transferred residue by residue to RNA article, it is proposed that the sequence in- and then to protein (Figure 1A). However, fem-3 mRNA-binding factors (PUF). This formation encoded by protein can be artificially transferred back to DNA and RNA, respectively, based on transcription activator-like effectors (TALE) and Pumilio/fem-3 mRNA-binding factors (PUF). Specifically, Illustration by Eloïse Kremer along the RVD loops of TALE repeats ed observations. First, the minimal num- a nucleotide-specific contact12. Since ber of repeats found in naturally occurring the DNA specificity can be designed by the sequence information encoded by hypothesis, if proved experimentally, sugTALE proteins is as low as 1.513, indicating customizable assembly of TALE repeats, protein cannot be transferred back to gests a new strategy for the synthesis of that a mononucleotide base pair alone TALEN (a TALE protein fused with a nucle- is likely to be sufficient to interact with nucleic acids. In some special cases, short DNA/RNA molecules. ase protein) is being deployed as a power- a TALE repeat. Second, each base pair DNA can also serve as a template in directing in vitro protein synthesis in the from the DNA duplex12 was reported to HYPOTHESIS Vol.12, No.1 | 2014 | hypothesisjournal.com HYPOTHESIS Potential protein-encoded synthesis of DNA and RNA 2/5 Fu DNA independently interact with the RVD loop contains an RNA-binding domain that of one TALE repeat, also strengthening the comprises multiple PUF repeats and also possibility of an interaction between a forms a superhelical structure to bind the TALE repeat and the mononucleotide RNA molecule within its inner concave base pair. Third, mono- or dinucleotide surface18. Three-dimensional structural base pairs, although unable to exist studies19 revealed that each PUF repeat, stably in solution , were reported to though not making direct interactions 14 translation RNA protein (PUF) TALE repeats Characteristic residues NG NN HD NI Mono- or dideoxynucleotide base pairs T G C A A T G C A T G C A C G T T A C G T A C G NI NG NN HD NI NG NN HD (ligation or condensation) Oligodeoxynucleotide duplex be stably formed in the hydrophobic with either the phosphate backbone or cavities cages15,16. the 2' hydroxyl groups of RNA, recog Fourth, the apparent dissociation con- nizes a single RNA base through its stant between optimized TALE pro- three conserved amino acids, two of of self-assembled teins and target DNA was reported to which make hydrogen bonds or Van der be as low as 0.16 nM17. Lastly, TALE Waals interactions with the edge of an repeats were reported to be flexible in RNA base and a third residue that stacks conformation to cope with the B-form with the same base and/or the preceding conformation of the DNA duplex12. base. The code of RNA-binding speci- Together, these observations strongly T G C A A T G C A T G C A C G T T A C G T A C G PUF repeats suggest that mono- or dinucleotide base pairs can be arranged along the corresponding characteristic amino acids of TALE proteins through sequence-specific ficity by PUF repeats has been decoded and artificially evolved20,21 (also illustrated in Figure 1C), by which modular PUF repeats capable of selectively binding specific RNA sequences can be created. Characteristic residues CQ NQ SE SR CQ NQ SE SR NQ NQ SE SR interactions as described above. As The bioinformatics analysis results de- Mono- or diribonucleotides A U G C A U G C A U GC such, these independent but arranged scribed here indicate that, although a (ligation or condensation) base pairs would be easily ligated or majority contain 8 PUF repeats, many A U G C A U G C A U G C condensed as an oligodeoxynucleotide of the naturally occurring PUF proteins duplex (as illustrated in Figure 1B). In this only carry 2 or 3 PUF repeats (Table 1). In Oligodeoxynucleotides regard, the sequence information of mul- particular, the PUF repeats in some PUF tiple RVDs discontinuously encoded in proteins are not sequentially connected the TALE protein is transferred to DNA. but discontinuously present as discrete synthesis of RNA by RNA ligase or condensation agents from mono- or diribonu- Synthesis of RNA encoded by PUF pro- units of one, two or three repeats (as from protein (e.g., TALE or PUF) to DNA or RNA is illustrated (highlighted in red cleotides that are arranged in a sequence-specific manner along the characteristic teins Similarly, the transfer of sequence or blue). B Schematic illustration of the synthesis of a DNA duplex by DNA ligase residues of multiple PUF repeats. The amino acid code of RNA base specificity information from protein to RNA can or condensation agents from mono- or dideoxyribonucleotide base pairs that was adopted from earlier studies potentially be achieved using PUF, which Figure 1 | Illustration of the modified central dogma from potential are arranged in a sequence-specific manner along the characteristic residues of synthesis of DNA and RNA encoded by protein. A Sequence information TALE repeats in a designed TALE protein. The amino acid code of DNA sequence transfer (from DNA to RNA to protein) takes place in nature and is summarized by specificity was adopted from Bogdanove et al.9. C Schematic illustration of the the central dogma of molecular biology (black). Sequence information transfer . 20,21 HYPOTHESIS exemplified in Figure 1). These observations indicate that the minimal number of PUF repeats for their stable interaction Vol.12, No.1 | 2014 | hypothesisjournal.com HYPOTHESIS Potential protein-encoded synthesis of DNA and RNA 3/5 Fu Repeat No. 2 3 4 5 6 7 8 9 Total Protein No.a 7 38 94 91 167 91 532 4 1024 Ratio (%) 0.7 3.7 9.2 8.9 16.3 8.9 52 0.4 100 Table 1 | Varying number of PUF repeats in naturally occurring PUF proteins. a A total of 1024 PUF proteins were found in SMART database (http://smart.embl.de/; date of 2013/6/5). The number of PUF repeats in each protein was counted and a summary of PUF proteins with different repeats is presented here. PUF = Pumilio/fem-3 mRNA-binding factors; SMART = Simple Modular Architecture Research Tool. dideoxyribonucleotide base pairs due to and the mono- and diribonucleotides at blocks. Specifically, if the building block space hindrance, and thus would not be a concentration of approximately 1 mM. is mononucleoside 3',5'-biphosphate, suitable to catalyze the ligation. An alternative approach to overcome this difficulty is to use chemical agents, such as cyanogen bromide23, cyanamide24 and imidazole25, which have been reported to be able to condense mono- and dideoxyribonucleotide base pairs into oligonucleotide duplexes. Since the base If polynucleotide phosphorylase is used for ligation , a specific diribonucleoside 26,27 (with a free 3'-terminal hydroxyl group) corresponding to the last two PUF mono- and diribonucleotides into oligori- added and incubated with the PUF pro- agents such as cyanogen bromide31 and tein. Polynucleotide phosphorylase can montmorillonite32,33. To this end, mixed even 1. It follows that the mono- and/or of the TALE protein to each of these RVDs of the TALE protein, condensation diribonucleotides arranged along the deoxynucleotides can be evaluated by these agents should be carried out ef- concave surface of a designed PUF pro- by commercial systems based on iso- ficiently due to the entropy reduction. tein could be assembled as an oligoribo- thermal titration calorimetry or surface PUF-encoded synthesis of RNA To syn- oligoribonucleotides by the enzyme. nucleotide (as illustrated in Figure 1C). As plasma resonance. According to the thesize specific single-stranded RNA, the such, the oligoribonucleotide sequence early reports12,22 on the interaction be- gene encoding multiple PUF repeats can be designed and assembled according the preliminary conditions used here are to the principles and methods described protein. as follows: TALE at a concentration of previously20,21, and the PUF protein can approximately 2 μM and the mono- and be expressed and purified similarly as re- dideoxyribonucleotides at a concentra- ported previously19,21. The binding affinity tion of approximately 1 mM. of the PUF protein to various types of multiple TALE repeats plus the N- and One approach to assembling these mono- and diribonucleotides can be eval- C-terminal signals of TALE can be mono- and dideoxyribonucleotide base uated by commercial systems based on designed de- pairs is to use DNA ligase plus ATP and isothermal titration calorimetry or those scribed previously (as reviewed by optimal ligation buffer. However, given based on surface plasma resonance. Bogdanove et al.9) and the protein can that the DNA duplex has been reported The results of this kind of experiment will be expressed and purified as described to be absorbed in the inner surface of determine the approach and conditions recently . Mixed 5'-phosphate mono- the superhelical structure of the TALE that are suitable for the subsequent liga- (4 types) and dideoxyribonucleotides protein12, DNA ligase bearing a molecu- tion and condensation. The preliminary (16 types) are then incubated with the lar weight of 43 kDa may not be able to conditions used here are as follows: PUF access those partially buried mono- and at a concentration of approximately 2 μM EXPERIMENTAL DESIGN TALE-encoded synthesis of DNA To synthesize specific DNA duplexes, the gene encoding 12,22 and assembled as purified TALE protein for a certain length The third approach to assemble these bonucleotides is to use condensation pairs are aligned along the characteristic tween TALE and short DNA duplexes, is not required. nucleoside 3',5'-diphosphates can be of time. In particular, the binding affinity is solely determined by the characteristic it is dinucleoside pyrophosphate, ATP repeats of the PUF protein and mixed with RNA bases may be as low as 2, or amino acids of PUF repeats in the PUF ATP should be included for ligation; if be added to the mixture to initiate the 5'-biphosphate mono- and diribonucleo- addition of mononucleoside 3',5'-diphos- sides are incubated with the PUF protein phates to the diribonucleoside, which will for a certain length of time before add- be further elongated, base by base, to ing these condensation agents and other crucial chemicals (e.g., metal ions). A substitute for polynucleotide phos- SIGNIFICANCE AND CONCLUSION In phorylase is RNA ligase28,29,30, which is summary, it is proposed that sequence- known to catalyze the formation of an specific DNA and RNA molecules can internucleotide phosphodiester bond be- possibly be synthesized according to tween an oligonucleotide acceptor with the characteristic amino acid sequences a 3'-terminal hydroxyl (minimal acceptor of designed TALE and PUF proteins, re- being trinucleotides) and an oligonucle- spectively. In terms of thermodynamics, otide donor molecule with a 5'-terminal TALE and PUF proteins arrange or fix the phosphate (minimal donor being trinucle- free mono- or dinucleotides, leading to oside diphosphate, dinucleoside pyro- a reduction in the entropy of the nucleo- phosphate and mononucleoside 3',5'-bi- tides and thus facilitating the subsequent phosphates). For this purpose, a specific ligation or condensation. This hypothesis triribonucleoside (with a free 3'-terminal is clearly experimentally testable. If hydroxyl group) corresponding to the proved, it is of interest to further exam- last three PUF repeats of the PUF protein ine whether Trp RNA-binding attenuation serves as the starting primer, and dinu- proteins34 and pentatricopeptide repeat cleoside pyrophosphate or mononucleo- proteins35, both interacting with RNA in side 3',5'-biphosphates serve as building a sequence-specific manner, can serve HYPOTHESIS Vol.12, No.1 | 2014 | hypothesisjournal.com HYPOTHESIS Potential protein-encoded synthesis of DNA and RNA 4/5 Fu a similar purpose as PUF. More impor- It was proposed recently8 that the third 2 Crick F. Central dogma of molecular biology. 8 Yarus M, Widmann JJ, Knight R. RNA-amino http://dx.doi.org/10.1007/978-1-4612-5190-3 tantly, proving these possibilities by characteristic residue of the PUF re- Nature. 1970;227(5258):561-3. acid binding: a stereochemical era for the genetic PMCid:PMC557514 rational design of experiments suggests peat, which stacks on and sandwiches http://dx.doi.org/10.1038/227561a0 code. J Mol Evol. 2009;69(5):406-29. 15Sawada T, Yoshizawa M, Sato S, Fujita M. a new protein-based strategy for syn- successive bound RNA bases19, may PMid:4913914 http://dx.doi.org/10.1007/s00239-009-9270-1 Minimal nucleotide duplex formation in water thesizing DNA/RNA molecules, which have played a role during the evolution 3 McCarthy BJ, Holland JJ. Denatured DNA as a PMid:19795157 through enclathration in self-assembled hosts. Nat is of interest in life science research and of such pairing. Therefore, it is of inter- direct template for in vitro protein synthesis. Proc 9 Bogdanove AJ, Voytas DF. TAL effectors: cus- Chem. 2009;1(1):53-6. biotechnology. est to further investigate the evolutionary Natl Acad Sci U S A. 1965;54(3):880-6. tomizable proteins for DNA targeting. Science. http://dx.doi.org/10.1038/nchem.100 significance of the other two character- http://dx.doi.org/10.1073/pnas.54.3.880 2011;333(6051):1843-6. PMid:21378801 istic residues of the PUF repeat that PMid:8900275 http://dx.doi.org/10.1126/science.1204094 16Sawada T, Fujita M. A single Watson-Crick make hydrogen bonds or Van der Waals 4 Hulen C, Legault-Demare J. In vitro synthesis of PMid:21960622 G x C base pair in water: aqueous hydrogen interactions with RNA.H large peptide molecules using glucosylated single- 10Boch J, Scholze H, Schornack S, Landgraf bonds in hydrophobic cavities. J Am Chem Soc. certain limitations (e.g., the binding re- ACKNOWLEDGEMENT This work was stranded bacteriophage T4D DNA template. Nucleic A, Hahn S, Kay S, et al. Breaking the code of DNA 2010;132(20):7194-201. peats in naturally occurring TALE and supported by research grants from the Acids Res. 1975;2(11):2037-48. binding specificity of TAL-type III effectors. Science. http://dx.doi.org/10.1021/ja101718c http://dx.doi.org/10.1093/nar/2.11.2037 2009;326(5959):1509-12. PMid:20429562 PMid:1052527 PMCid:PMC343570 http://dx.doi.org/10.1126/science.1178811 17Meckler JF, Bhakta MS, Kim MS, Ovadia R, 5 Woese CR, Dugre DH, Saxinger WC, Dugre SA. PMid:19933107 Habrian CH, Zykovich A, et al. Quantitative analysis The molecular basis for the genetic code. Proc Natl 11Moscou MJ, Bogdanove AJ. A simple cipher of TALE-DNA interactions suggests polarity effects. Acad Sci U S A. 1966;55(4):966-74. governs DNA recognition by TAL effectors. Science. Nucleic Acids Res. 2013;41(7):4118-28. http://dx.doi.org/10.1073/pnas.55.4.966 2009;326(5959):1501. http://dx.doi.org/10.1093/nar/gkt085 PMid:5219702 PMCid:PMC224258 http://dx.doi.org/10.1126/science.1178817 PMid:23408851 PMCid:PMC3627578 6 Woese CR, Dugre DH, Dugre SA, Kondo M, PMid:19933106 18Wang X, Zamore PD, Hall TM. Crystal struc- Saxinger WC. On the fundamental nature and evo- 12Deng D, Yan C, Pan X, Mahfouz M, Wang J, ture of a Pumilio homology domain. Mol Cell. lution of the genetic code. Cold Spring Harb Symp Zhu JK, et al. Structural basis for sequence-spe- 2001;7(4):855-65. Quant Biol. 1966;31:723-36. cific recognition of DNA by TAL effectors. Science. http://dx.doi.org/10.1016/S1097-2765(01)00229-5 http://dx.doi.org/10.1101/SQB.1966.031.01.093 2012;335(6069):720-3. 19Wang X, McLachlan J, Zamore PD, Hall TM. PMid:5237212 http://dx.doi.org/10.1126/science.1215670 Modular recognition of RNA by a human pumilio- 7 Hlevnjak M, Polyansky AA, Zagrovic B. PMid:22223738 PMCid:PMC3586824 homology domain. Cell. 2002;110(4):501-12. Sequence signatures of direct complementarity be- 13Boch J, Bonas U. Xanthomonas AvrBs3 family- http://dx.doi.org/10.1016/S0092-8674(02)00873-5 tween mRNAs and cognate proteins on multiple lev- type III effectors: discovery and function. Annu Rev 20Filipovska A, Razif MF, Nygard KK, Rackham O. els. Nucleic Acids Res. 2012;40(18):8874-82. Phytopathol. 2010;48:419-36. A universal code for RNA recognition by PUF pro- http://dx.doi.org/10.1093/nar/gks679 http://dx.doi.org/10.1146/annurev-phyto-080508-081936 teins. Nat Chem Biol. 2011;7(7):425-7. PMid:22844092 PMCid:PMC3467073 PMid:19400638 http://dx.doi.org/10.1038/nchembio.577 14Saenger W. Principles of Nucleic Acid Structure. PMid:21572425 The potential sequence information transferring from TALE and PUF to DNA and RNA, respectively, however, would not occur in nature, presumably due to PUF proteins were found to be less than National Natural Science Foundation of and 10 (Table 1), respectively). In China (No. 31100559 and No. 31270804 other words, it is unlikely that in nature to X.F.) and the National Basic Research genetic information is stored in proteins Program of China (973 Program) (No. like TALE and PUF, which is then trans- 2012CB917300 to X.F.). 30 13 ferred into nucleic acids. In this regard, the hypothesis does not challenge the central dogma. ABOUT THE AUTHOR Dr. Fu is a protein scientist focusing on the mechanism of protein biogenesis, folding, assembly Nevertheless, the specific amino acid- and degradation, as well as protein evo- nucleotide interaction between TALE lution. His long-term goal is to elucidate and DNA, or between PUF and RNA, how the genetic information encoded may provide new insights into the ori- by DNA is accurately transferred to the gin of the central dogma. For instance, three-dimensional structure information regarding the origin of the ribosome- of proteins, which is usually facilitated mediated translation from mRNA to by molecular chaperones and folding protein, a direct codon-amino acid catalysts in cells. stereochemical pairing was suggested to occur during evolution5,6, which later evolved to the current translation form that does not involve such direct pairing. REFERENCES 1 Crick FH. On protein synthesis. Symp Soc Exp Biol. 1958;12:138-63. PMid:13580867 New York, Berlin: Springer; 1984. HYPOTHESIS Vol.12, No.1 | 2014 | hypothesisjournal.com HYPOTHESIS Potential protein-encoded synthesis of DNA and RNA 5/5 Fu 21Dong S, Wang Y, Cassidy-Amstutz C, Lu G, 27Slomovic S, Portnoy V, Yehudai-Resheff S, 33Ferris JP, Ertem G. Oligomerization of ribo- Bigler R, Jezyk MR, et al. Specific and modular Bronshtein E, Schuster G. Polynucleotide phosphory- nucleotides on montmorillonite: reaction of the binding code for cytosine recognition in Pumilio/ lase and the archaeal exosome as poly(A)-polymer- 5'-phosphorimidazolide of adenosine. Science. FBF (PUF) RNA-binding domains. J Biol Chem. ases. Biochim Biophys Acta. 2008;1779(4):247-55. 1992;257(5075):1387-9. 2011;286(30):26732-42. http://dx.doi.org/10.1016/j.bbagrm.2007.12.004 http://dx.doi.org/10.1126/science.1529338 http://dx.doi.org/10.1074/jbc.M111.244889 PMid:18177749 34Antson AA, Dodson EJ, Dodson G, Greaves RB, PMid:21653694 PMCid:PMC3144504 28England TE, Gumport RI, Uhlenbeck OC. Chen X, Gollnick P. Structure of the trp RNA-binding 22Yin P, Deng D, Yan C, Pan X, Xi JJ, Yan N, et al. Dinucleoside pyrophosphate are substrates for attenuation protein, TRAP, bound to RNA. Nature. Specific DNA-RNA hybrid recognition by TAL effec- T4-induced RNA ligase. Proc Natl Acad Sci U S A. 1999;401(6750):235-42. tors. Cell Rep. 2012;2(4):707-13. 1977;74(11):4839-42. http://dx.doi.org/10.1038/45730 http://dx.doi.org/10.1016/j.celrep.2012.09.001 http://dx.doi.org/10.1073/pnas.74.11.4839 PMid:10499579 PMid:23022487 PMid:200936 PMCid:PMC432051 35Schmitz-Linneweber C, Small I. Pentatricopeptide 23Sokolova NI, Ashirbekova DT, Dolinnaya NG, 29England TE, Uhlenbeck OC. Enzymatic oligo- repeat proteins: a socket set for organelle gene ex- Shabarova ZA. Chemical reactions within DNA du- ribonucleotide synthesis with T4 RNA ligase. pression. Trends Plant Sci. 2008;13(12):663-70. plexes. Cyanogen bromide as an effective oligo- Biochemistry. 1978;17(11):2069-76. http://dx.doi.org/10.1016/j.tplants.2008.10.001 deoxyribonucleotide coupling agent. FEBS Lett. http://dx.doi.org/10.1021/bi00604a008 PMid:19004664 1988;232(1):153-5. 30Kikuchi Y, Hishinuma F, Sakaguchi K. Addition http://dx.doi.org/10.1016/0014-5793(88)80406-X of mononucleotides to oligoribonucleotide accep- 24Ibanez JD, Kimball AP, Oro J. Possible prebiotic tors with T4 RNA ligase. Proc Natl Acad Sci U S A. condensation of mononucleotides by cyanamide. 1978;75(3):1270-3. Science. 1971;173(3995):444-6. http://dx.doi.org/10.1073/pnas.75.3.1270 http://dx.doi.org/10.1126/science.173.3995.444 PMid:274717 PMCid:PMC411452 PMid:17770449 31Kanaya E, Yanagawa H. Template-directed polym- 25Ibanez JD, Kimball AP, Oro J. Condensation erization of oligoadenylates using cyanogen bromide. of mononucleotides by imidazole. J Mol Evol. Biochemistry. 1986;25(23):7423-30. 1971;1(1):112-4. http://dx.doi.org/10.1021/bi00371a026 http://dx.doi.org/10.1007/BF01659398 32Kawamura K, Ferris JP. Kinetic and mechanistic PMid:5173650 analysis of dinucleotide and oligonucleotide forma- 26Grunberg-Manago M, Oritz PJ, Ochoa S. tion from the 5'-phosphorimidazolide of adenosine Enzymatic synthesis of nucleic acidlike polynucleo- on Na(+)-montmorillonite. J Am Chem Soc. 1994; tides. Science. 1955;122(3176):907-10. 116(17):7564-72. http://dx.doi.org/10.1126/science.122.3176.907 http://dx.doi.org/10.1021/ja00096a013 HYPOTHESIS Vol.12, No.1 | 2014 | hypothesisjournal.com