您的当前位置:首页正文

利用脱氧核酶调控基因表达的DNA计算模型研究

2022-12-08 来源:易榕旅网
第3l卷第12期 计 算 机 学 报 Vol_31 No.12 2008年12月 CHINESE JOURNAL OF COMPUTERS Dec.2008 利用脱氧核酶调控基因表达的DNA计算模型研究 张凤月” 宋海刚 温俊 海” 张 强。 刘 波 (北京理工大学生命科学与技术学院生物医学与工程系 北京 100081) (中国科学技术部基础研究管理中心 北京 100862) 3 (大连大学辽宁省智能信息处理重点实验室 辽宁大连 116622) 4 (中国检验检疫科学研究院北京100029) 摘要 用于逻辑调控基因表达分子自动机的研究是DNA计算的重要研究领域.文中将脱氧核酶技术应用于 DNA计算研究当中,利用脱氧核酶的特性,特别是可以作为反义药物的特点,作为构建分子自动机的主要材料,设 计了调控基因H—ras表达的DNA计算模型,而且模型也可适用于其它过表达基因的调控.结合DNA计算具备的 高度并行性和智能性的优点,该模型为DNA计算在基因表达调控方面的应用做了进一步探索. 关键词DNA计算;脱氧核酶;基因表达调控 中图法分类号TP301 DNA Computing Model with DNAzyme to Control Gene Expression ZHANG Feng—Yue” SONG Hai—Gang WEN Jun—Hai ZHANG Qiang。’ LIU Bo ”(Department of Biomedical Engineering,School of Life Science and Technology,Beijing Institute of Technology,Beijing 100081) (Fundamental Research Management Center of Ministry of Science and Technology of China,Beijing 100862) 。 (Liaoning Key Laboratory of Intelligent Information Processing,Dalian University,Dalian,Liaoning 116622) ”(Chinese Academy of Inspection and Quarantine.Beijing 100029) Abstract A molecular computer for logical control of gene expression is an essential research field of DNA computing.In this paper,with its characteristic especially being antisense medica— tion,DNAzyme technique is used in construction of DNA computing model as basis materia1. Furthermore,a DNA computing model is designed to control H—ras gene expression,and this model also has potential to control other over expressed genes.Benefited from the massive paral— lelism and intelligence feature of DNA computing,this model explores further application of DNA computing to contro1 of gene expression. Keywords DNA computing;DNAzyme;control of gene expression l IntrOductiOn ready long—standing.but it was not until 1 994 when Adleman successfully solved a 7 vertexes of Scientific researcher attempted to use the or— the Hamihon Road Problem with DNA molecules, ganism’S biological macromolecule(DNA,RNA, this aspect’S research only then obtained a major proteins,etc.)to improve the electronic computer breakthroughL .Afterward,the DNA computing or construct a new type of computer vision was al— has made the outstanding contributions on solving 收稿日期:2oo8一O4—20;最终修改稿收到日期:2008—11-02.本课题得到国家自然科学基金(60403001,30740036,30870573)、辽宁省智能信 息处理重点实验开放课题(2005—1)以及新世纪优秀人才支持计划资助.张凤月,男,1973年生,博士,讲师,主要研究方向为生物计算和 生物信息学.宋海刚,男,1975年生,博士,助理研究员,主要研究方向为系统工程、计算机应用.温俊海,男,1969年生,博士,副教授,主要 研究方向为图像处理与重建.张强,男,1971年生,博士,教授,主要研究领域为DNA计算、神经网络、计算机动画.刘 波,女,1973年 生,博士,副研究员,主要研究方向为DNA计算和植物保护.E-mail:liubobj@126.corn. 2098 计 算 机 学 报 2008越 hard computational problems such as NP problem, which is difficult to solve by electronic computer currently.The representative of the study include: Lipton proposed a solution of Satisfiability problem (SAT)by using a simple contact network in a tube in 1995[ ]:0uyang imitated Adleman’s method to put forward a solution for the Maximal Clique Problem in 1997/。];Eric tried to solve SAT by a cellular automata model as well as by the plasmid suggested in reference[43;Qihuang Liu,Wiscon— sin—Madison University,did the surface computing and solved a 4 variables SAT problem in 2000,this experiment indicated the possibility of constructing DNA computer on the surfaceE ];Braich solved a 20 variables of the 3-SAT problem based Acrydite M separation technology in 2002[ .In addition,for the general model of DN A computing,researchers take the Sticker Systems[ ]and the Splicing System modelE as representative.and many related re— search results has been reported. In 200 1,the programmable semiautomatic tube DNA computer based on the biological macro— molecule,which proposed by Benenson from Israe— li Weitzman Academy of Science caused the DNA computation research to obtain unDrecedented pro- gress[ .In 2002,a DNA computer was developed for the analysis of gene expression by Suyama. This computer was mainly composed of the molec— ular computation module and the diagnostic part. The former selected the correct result responded through the biochemistry reactions。the 1atter car- ried on analysis to the obtained result[ .After proposed the programmable semiautomatic tube DNA computer and unified automata theory and life science,Benenson put forward a molecular au— tomaton for the logical regulation gene expression to open up a new research direction in DNA corn— putation in 2004E .which related test simulation has been carried on 72 vitro[ .This automaton is mainly composed of three programmable modules, respectively were the computation module,the in— put module and the output module.The elementa— ry function of this automaton is to diagnose the specific mRNA density automatically in tube ac— cording to the hypothesis logical rule,and to pro— duce the suppressor for gene expression corre— sponding the diagnosis result(or release drugs). Thereby it can be used as gene therapy,i.e.anti— sense therapyE13].And the essentia1 of this compu— tation module is using endonuclease Fok I as the hardware molecule,and after operation and transi- tion。this automaton have two conditions of“yes’’ and“no”.corresponds in the biological“positive diagnosis’’and“negative diagnosis”separately. Afterward,the related Turning machine,the bio一 1ogieal Turning machine and the molecular automa— ton model also obtained further elaborated[ .In 2006,Georg designed a logic operation model without enzyme,which using short otigonucleoti— des as input and output and implement Boolean logic operation function(“AND”“0R”and “NO’’)[ 引.Carrying on the multistage logic opera— tions on the chip surface.a DNA logic computing model was proposed with more complex struc— ture ̄ 引,in addition,according to the Boolean logic operation,the use of oligonucleotide—based distri— bution so as to produce output fluorescent mole— cule,also caused using DNA to carry out the logic operation examination easierc17].Harvard Univer— sity system biology’s scholar further designed a DNA logic operation model with RNAi technology in 2007,this mode1 was confirmed that it could be applied in artificial cultivation of renal cell[ .With the use of molecular release’s configurationalIy an— tropy,to design specific nucleotide sequences to catalyze the release of the output molecular can amplify the logic operation signal .In 2008, Peng Yin designed a variety of complex structures like linear,branched,circuit and autonomous lOCO— motion by the use of DNA strand self—assembling and disassembling,which has established more solid foundation for DNA computing[zo]. In this paper,a DNA computing model is pro— posed by using DNAzyme technology of molecular biology,after integrated molecular automata for logic control of gene expression with cancer re— search related to polygenes.This study, DNAzyme—based molecular automata for logic reg— utation of gene expression,is expected to improve its application in tumor treatment,gene regulation and so on.Furthermore,this research provides a potential means for research of regulation of gene expression and establishes a basis for its applica— tion in living organisms. 2 DNAzyme—Based DNA Computing Model for Regulation Gene Expression 2・1 DNAzyme DNAzyme(Deoxyribozyme,DNA),is a cata— lytic DNA molecules,also known as DNA enzyme or enzyme—DNA,and is a new member of biologi— cal catalysts together with protein enzymes and ri— bozyme.The main enzyme activity are ligase,met— al chelating and phosphatase activity. 12期 张凤月等:利用脱氧核酶调控基因表达的DNA计算模型研究 2099 It has been proved that DNAzyme 1 0—23 can snip mRNA in the AU point,thereby,in theory only through adj usting the nueleotide components of its substrate’s recognition site,any of the mR— NA start code can be cleaved,which implies a mas— ter key for the control protein expressionL j.So DNAzyme has potential applications in many areas such as gene therapy,disease diagnosis and analysis tools of molecular biology[ 一 .As a spe— cia1 kind of DNA molecules,the specia1 features of DNAzyme make it suitable for DNA computing, especially as a active molecular for the regulation of gene expresston; and it IS posstble oi usmg DNAzyme to design logical operation due to the in— dependence of the its domains.The DNAzyme 10—23 molecular model is shown in Fig.1,and its DNAzyme sequence is indicated in Fig.2. GC TACAA TA DNAxyme c cG G 5 一 一3 3 一 —5 RNA G U Fig.1 10—23 DNAzyme molecular model Gc I A AA T A C C DNAxy1TIe G G A G J— ACC AG UUAC ¨3,一UI-A……¨I UUCCGUC AU一一0; H—ras rllRNA I Fig.2 DNAzyme sequence of the proposed DNA computing model and its active site in H—ras mRNA molecules 2.2 Selection of the Target Gene With the in—depth research in the lire sciences, in particular the completion of human genome pro— iect(HGP)and the startup of post HGP,lots of disease genes have been revealed and some diseases will be cured through gene therapy.It is now un— ambiguous that the tumor is a polygenic disease, due to the inactivation of tumor suppressor genes and the activation of cancer genes,as well as the disorder of cell cycle regulation gene on the level of gene expression.For example.when p1 6(tumor suppressor genes)and CyclinD1(cancer gene)gene bind CDK4(eyclin—dependent kinase),the former inhibits cell growth and the latter promotes cell growth.That missing p1 6 or arising CyclinD1 ex— pression will cause uncontrolled cell growth and in— duce cel1 cancer finally[ .In human tumors,a class of important genes,ras gene family(including three functional genes that H—ras,K—ras and N—ras),has to be taken seriously,the current re— searches have shown that malignant tumors are as— sociated with disorder of pl 6,CyclinD1 and H—ras gene expression in gastric cancer’colorectal canc— er,and other common gastrointestinal cancer. Therefore:the rules to determine these cancer cells can be summed up as:p1 6 CyclinD1十H~ras 十( 1ower expression;十over expression). The human H—ras gene was chosen as con— trolled target to construct a DNA computing model for regulation of gene expression.Human H—ras is positioned in chromosome 1 1,its family encode highly similar proteins of which the molecular weight is 2100 Dalton and called p21 protein.The D2 1,located in the inner membrane,plays an im— portant role in adenosine receptor cyclase signaling pathway with the GTP activity.Searched through bioinformatics methods,the entire length of hu— man H—ras gene sequence is 41026 bp.An EST se— quenee,which is a gene transcription fragment of the mRNA,is selected as the target for design of the DNA computing model,and the corresponding cDNA sequences is following: 1 ccggcctngg ncccggccct ggncccgggg gcagtcgcgc ctgtgaacgg tggggcagga gaccctgtag aggaccccg 8 l ggccgcaccg tggaggagcg atgacggaat ataagctggt ggtg— gtgggc gccggcggtg tgggcaagag tgcgctgacc 1 6 1 atc— cagctga tccagaacca ttttgtggac gaatacgacc ccactataga gggttnctac cggaagcagg tgg・ 2.3 Molecular Automaton ModeI The molecular automata mode1 shown in Fig.3 includes three modules.The hardware module is a restrictive endonuclease which can recognize and snip a specific sequence of double—stranded DNA and then form a sticky or blunt ends;The software module contains the restrictions enzyme site,the interva1 sequence and a double—stranded DNA with sticky ends. The input and output module are DNA mole— cules,and encoded DNA sequences are taken as state set and symbo1 set.The state set include state of SO and S1,as well as the symbol set is“a” and“b”:the transit rules are described.If the current state iS SO and the symbol read is“b”, then the unit,which is composed of the hardware module and the software module.will change its state to S1 and the symbol to“a”.and move one position to the right as wel1.The output state iS SO and the symbol is“t”(terminator)after 21OO Software molecule 计 算 机 学 报 2008年 Output molecule Fig.3 The theoretical model of molecular automata operations. the snip site of DNAzyme iS the AU sequence of between 115 to 120 nucleotide.’ 2.4.2 Regulate rules of mode1 2.4 Model Design Based on the research about the theoretical model of molecular automata。A DNA computing model is designed to regulate H—ras gene expres— The DNA computing model recognizes the lev~ el condition of gene H—ras expression,and releases inhibiting molecules if the gene is over expression, or maintains the original state if it iS norma1. 2.4.3 Composition and function of DNA computing mode1 The designed DNA computing model includes sion after the target gene is selected and the related DNAzyme iS constructed. 2.4.1 Basic principles of model The model encoding has been checked by re— lated bioinformatics software in order to ensure the specificity between the designed DNA computing three parts of which are computing module,diag— nostic module and output module.The respective models to the regulated target.The specified re— gion,from 181 to 210 nucleotide of the target gene,serves as a template for model design,and @ structure and their relation are described in Fig.4, and encode of these sequences are 1isted in Table 1. Fig.4 The structure of DNA computing model based DNAzyme 10—23 for regulation H—ras gene expression and the computing processes Table 1 DNA Computing Model Components and Encode 12期 张风月等:利用脱氧核酶调控基因表达的DNA计算模型研究 21O1 The purpose of computing module is for the detection of target mRNA.It is a ssDNA strand and the encode is complementary to the region from 181 to 200 nucleotide of H—ras cDNA. The function of the diagnostic module is tran— sits state according to the proceeded results of the computing module;the sequences of the diagnostic module are composed of diagnostic strand and its complementary,separation stand and restriction enzyme strand.The diagnostic strand forms doub— le-stranded structure with its complementary.The diagnostic strand and the computing module share the same encoding strategy.The structure of the separate stand is double—stranded and its encode is as same as that of the region of 200 to 210 nucleo— tide of H—ras cDNA.Located in the middle of the diagnostic and the separate stand,the restriction enzyme sequence is only single—stranded and forms a circle structure. An endonuclease sequence and a DNAzyme se— quence constitute the output module.The endonu— clease sequence is complement to the endonuctease sequence in the diagnostic module.As a suppressor, the structure and activate site of DNAzyme is indica— ted in Fig.3,of which the strand of CGAATA TAAGGCGGT recognize and complete to the sequenee of 1 1 5 to 1 20 nucleotide of target mRNA. The stand of GGCTAGCTACAACGA is the coral site because it can snip AU site specifically. 2.5 Biological Computing Processes of the Model The first step,a certain amount of computing module(equivalent to the number of mRNA ex— pressed by gene H—ras)was inserted into the target cell’s cytoplasm。in there is the transcript inclu— ding the mRNA of gene H—ras.According to the principle of Watson—Crick,the computing module is complementary to the region of 1 81 to 200 nucle— otide of mRNA.Then (1)If the amount of computing module三三=the amount of transcription mRNA of gene H—ras,all the region of 181 to 200 nucleotide of these mRNA will be double—stranded structure(20bp),which is shown in Fig.4④; (2)If the amount of computing module<the amount of transcripted mRNA of gene H—ras,the region of l 8 1 to 200 nucleotide of those redundant mRNA will remain the origina1 single—strand structure. The second step,a certain amount of the diag— nostic module enter the cel1.and because the diag— nostic strand of this module is the same as the computing module’s,then: (1)If the first step produces the results of (1),the diagnostic module will remain the original structure indicated by Fig.4( ; (2)If the first step produces the results of (2),the diagnostic module wil1 form complemen— tary structure with these redundant mRNA in the region of 181 to 200 nucleotide(25bp),which is shown in Fig.4@; The third step,if the second step produces the results of(2),the complementary sequence(1 2 bases)will be single strand due to its molecular force less than double—stranded force.So the ex— posed single—stranded restriction enzyme sequence form a completed double—stranded site with the complementary sequence of the output module, and the diagnostic module changes its state indica— ted by Fig.4@. The fourth step,the DNAzyme is released by adding restriction enzyme Bcl I,which snip the output module in site. Furthermore,the DNAzyme cleavers mRNA molecules in AU site af— ter it is attached to the region of l15 to 120 nucleo— tide。so the over expressed gene,H—ras,is proper— ly regulated,and the operation is indicated in Fig.4④. After these four steps operations,the DNA computing based DNAzyme fulfills the designed purpose and maintains the normal function of cells through diagnose target gene and degradation ex— cessive mRNA when gene is over expressed. 3 Conclusion and Prospect Gene expression and regulation of organisms are the essential function,any out of control in this function wil1 no doubt have adverse effects,such as has been shown that cancer is the outcome of a polygenes interaction,of which oncogene activa— tion and over expression play an important role.In this paper, a DNA computing model with DNAzyme to regulate gene expression of H—ras gene is constructed in theory.This model has a good versatility in the study of Gene expression and regulation,especially over expressed genes, since the output molecules is DNAzyme.For ex— ample.only taken different encode,the model will adapt to regulate over expressed CyclinD1 gene. Despite that the current DNA computing mod— els for regulation of gene expression are not very complicated as well as the target genes are less,it is enough for scientists to create biosensors and drug transport systems for specific human cells.In addition to the treatment of cancer,these resear— 21O2 计 算 机 学 报 2008年 ches have good prospects in functiona1 genomics, Dekker Encyclopedia of Nanoscience and Nanotechnology. New York:Marcel Dekker,Inc.,2004:2043-2056 l B,Ben—Dor U,Adar R et a1.An autono- [123 Benenson Y,G|mous molecular computer for logical control of gene expres— sion.Nature,2004,429:423—429 sense.Nature,2004,427: [13] Anne C.Automata make anti351—352 especially to identify complex signaling pathway in a comprehensive relations of different genes.From that point of view that the non-linear trait is mainly characteristic of life system itself.DNA computa— tion with its high parallelism,intelligent and affin— ity to organisms will be a potential research tool for complex phenomenon of life research。further ro E。Benenson Y.Bringing DNA computers to life: [143 ShapiTapping the computing power of biological molecules gives study will have a major significance both in the field of information science and life science. References [13 Adleman L M.Molecular computation of solution to combi— natorial problems.Science,1994,266:1021—1024 [2]Lipton Richard J.DNA solution of hard computational prob— lems.Science,1995,268:542—545 [33 Ouyang Qi.DNA solution of the maximal clique prohlem. Science,1997,278:446—449 [43 Head T,Lommerse P H M,Spaink H P.Computing with DNA by operating on plasmids.BioSystems,2000,57:87— 93 [53 Liu Qing—Hua,Wang Li—Man,Anthony G F,Anne E C et a1.DNA computing on surfaces.Nature,2000,403:175一 】78 [63 Braieh R S,Chelyapov N,Johnson C et a1.Solution of a 20一 Variable 3-SAT problem on a DNA computer.Science, 2002,296:499-502 [73 Karl L,Paun G,Rozenberg et a1.DNA computing,sticker systems,and universality.Aeta Informatica,1998,35:401— 420 [8]Paun G,Rozenherg G,Salomaa A.DNA Computing:New Computing Paradigms.Berlin:Springer-Varlag,1 998 [9]Benenson Y,Paz—Elizur T,Adar R et a1.Programmable and autonomous computing machine made of biomolecules.Na— ture,2001,414:430—434 [1o]Sakakibara Y.Intelligent DNA chips:Logical operation of gene expression profiles on DNA computers.Geneme Infor— matics,2002,11:33—42 [1 13 Benenson Y,Shapiro E.Molecular computing machines// Schwarz James A,Contescu Cristian I,Putyera Karol eds. ZHANG Feng—Yue,born in 1973, Ph.D.,lecturer.His research interests include biological computing and bioin— formatics. SONG Hal-Gang,born in 1975,Ph.DI,assistant re— searcher.His research interests include systems engineering rise to tiny machines that can speak directly to living cells. Scientific American,2006,5:45-51 [15] Georg S,David S,David Y Zh et a1.Enzyme—free nucleic acid lc}gic circuits.Science,2006,314:1585一l588 [16] Brian M F.Scott L C.Ghadiri M.Modular multi—level cir— cults from immobilized DNA-based logic gates.Journal of the American Chemical Society,2007,129:l4875-14879 [17] Wataru Y,Yohei Yokobayashi.Photonic boolean logic gates based on DNA aptamers.Chemica1 Communications。2007 (2):195—197 [183 Keller R,Leonidas B,Rohan M et al。A universal RNAi- based logic evaluator that operates in mammalian cells.Na— ture Biotechnoloy,2007,25:795—801 [193 David YZh,Andrew J T,Bernard Y et a1.Engineering en— tropy-driven reactions and networks catalyzed by DNA.Sci— ence,2007,318:1121—1125 [203 Peng Yin,Harry M T Choi,Colby R Calvert.Programming biomolecular self-assembly pathways.Nature,2008,451, 318—322 [213 Santoro S W,Joyce G F.Mechanism and utility of an RN Ar cleaving DNA enzyme.Biochemistry,1998,37:13330— 13342 [22] Achenbach J C,Chiuman W,Cruz R P,Li Y.DNAzymes: From creation in vitro to application in vivo.Current Phar— maceutical Bj0techn010gy,Z004,5:321—336 [233 Cairns M J,Hopkins T M。Witherington C,Sun L Q.The influence of arm length asymmetry and base substitution on the ac-tivity of the 10—23 DNA enzyme.Antisense and Nucle— ic Acid Drug Development,2000,10:323—332 [243 Azechi H,Nishida N,Fukuda Y et a1.Disruption of the pl 6/Cyclin D1/Retinoblastoma protein pathway in the major~ ity of human hepatocellular carcinomas.Oncology,2001, 60:346—354 and computer applications. WEN Jun-Hai,born in 1969,Ph.D.,associate profes— sor.His research interests include image processing and im— age reconstructiOn. ZHANG Qiang,born in 1971,Ph.D.,professor.His research interests include DNA computing,neural networks and computer animation. LIU Bo,born in 1973,Ph.D.,associate researcher. Her research interests include DNA computing and plant protection. 

因篇幅问题不能全部显示,请点此查看更多更全内容