DE10046069A1 - Synthesis of polynucleotide in microfluidics system, useful for preparing synthetic genes, comprises stepwise addition of sequemers to immobilized starter oligonucleotide - Google Patents

Abstract

Preparative synthesis of polynucleotides (I) using a microfluidics system (MS), is new. A starter oligonucleotide (ON) is either bound (directly or through a spacer) to the surface of the reaction space in MS, or is synthesized directly, with one end bound to the surface. ON corresponds to one end of (I) being synthesized. One or more rounds of hybridization of 'sequemers', each corresponding to a segment of (I) or its complement and having overlapping ends that allow hybridization, is performed. If required blunt ends, formed e.g. after a polymerase reaction, are converted to sticky ends suitable for hybridization. Where there are nicks between consecutive sequemers, ligation is performed, or if a segment of one or more nucleotides (nt) is missing (complementary to a segment of sequemer) this is completed by polymerase reaction, and ligation of the resulting nick if necessary. The process is repeated as necessary to assemble (I). Independent claims are also included for the following: (1) microfluidics element that includes bound polynucleotides produced by the new method, or their precursors; and (2) a kit or apparatus for the new method comprising appropriately coated microfluidics elements and optionally other components.

Description

Summary of the invention

The invention relates to a new process for the synthesis of polynucleotides using of microfluidic systems, microfluidic elements obtained with the process bound polynucleotides or precursors thereof, the use of micro fluidic element-bound precursors of the polynucleotides to be produced or the ent speaking microfluidic elements for the production of the polynucleotides and / or Kits or apparatus for the production of polynucleotides that are loaded accordingly Include microfluidic elements.

Background of the Invention

The term "microfluidics" denotes a promising technology that using miniaturized elements such as biochemical detection reactions in allowed very small scale. The miniaturization of biochemical reac that can be achieved in this way besides the economical use of high-quality or expensive reagents tien and materials further the special advantage of short reaction and detection times. Temperature-controlled processes can be set up very quickly and precisely controlled become. Through computers and suitable interfaces as well as micropumps, micro valves, Mi kroheizvorrichtungen, devices for creating fields, etc. can temperatures as well as liquid flows in the microfluidic microchannels and / or microcompar Timed reaction chambers can be precisely controlled. This allows the exact setting concentration in mixing processes. Various, for example opti cal detection methods can be used easily, especially if the micro fluidic elements are at least partially transparent.

The use of microfluidic systems has been used for many applications in the bio technology proposed, especially in the field of analytics, such as detection or the detection of nucleic acids, or also for syntheses, on the one hand in the combination Chemical chemistry (e.g. for 1,4-benzodiazepine derivatives, WO 96/15450), on the other hand to Am plication of existing DNA or RNA molecules with either temperature variation (e.g. WO 99/12016, DE 195 19 015) or under isothermal conditions (e.g. US 5,939,291, Use of denaturation by bases or use of optional paramagneti  electrostatically charged microparticles that attach the sample nucleic acids to the Bind surface and allow their separation in electrical fields). WO 99/36571 be writes the analytical use of "protector" nucleic acid sequences attached to fixed Substrates such as arrays or beads are bound, followed by hybridization with dissolved Most nucleic acids, then primers are attached to this and subsequent amplification and detection of the amplified nucleic acids formed in the analysis. WO 99/57310 describes microfluidic systems with a meandering structure for analysis. For example, Trennver drive for biomolecules (DNA, RNS) from Agilent under the brand DNA LabChip ™ , Aclara Biosciences Inc. offers plastic arrays under the name LabCard.

A still new area of using microfluidic elements relates to the implementation of biomolecular syntheses on a micro scale. Genetic engineering methods are becoming increasingly important in the field of biotechnology and the "life sciences". In particular, methods for generating recombinant genes and methods of "reverse genetics"("reversegenetics", in which the analysis or modification of the DNA is the starting point) are becoming increasingly important. The cloning and "in vitro" mutagenesis of gene constructs for applications in reverse genetics has so far tied up considerable time, human and financial resources. Here, simply making available, in particular longer DNA sequences, such as synthetic genes, would make enormous progress. In both academic and industrial research, gene synthesis methods using microfluidic elements would enable a direct transition from computer-aided DNA sequence analysis to function-oriented analysis and use of a wide variety of constructs - right up to use in various organisms in vivo. In addition to a sometimes drastic reduction in the time required, the focus of infrastructure and personnel on the actual core competencies for researchers could result in significant cost advantages. The free physical availability of digital genetic information (in the form of sequences) that can be obtained through the de novo synthesis of long DNA sequences with all possibilities for variation could remove existing limitations. It would enable new insights to be gained and new perspectives to be realized, by drawing conclusions from observing the flow made observable with their help and implementing genetic information. New solutions that are difficult or impossible to implement with conventional methods could be offered. In "green" (= plant) and "red" (animal-related) biotechnology and microbiology, there are a variety of specific applications, e.g. B .:

• - Improvement (qualitative and / or quantitative) of the expression of allogens and trans genes in various organisms based on codon optimization and / or Mo. dification of regulators (cis and trans elements), e.g. B. Expression of transgenes in plants, DNA vaccination or gene therapy;
• - Generation of suitable genes (or gene products that can be produced) for therapeuti cal applications, e.g. B. antisense molecule expression, ribozyme expression or Immunotoxin production;
• - Creation of artificial protein sequences e.g. B. from results of "molecular modeling" and genetic changes for protein structure and function studies of synthetic Designer proteins (e.g. addition or removal of protein domains or protein Leading sequences such as "leading signals", "homing signals", chimeric or fusion proteins from the combination of coding regions of different heterologous genes, etc.);
• - and / or new opportunities in the field of molecular evolution processes.

Against this background, the aim of the present invention is fully automated synthesis synthetic biomolecules such as oligonucleotides or polynucleotides, e.g. B. synthetic Ge ne etc.) by means of micro-compartmentalized microfluidic elements based on computer ge supported fully automatic synthesis machines. The elaborate in vitro con structure for the physical realization of suitable, biologically active nucleic acid moles cool, like RNA, DNA and gene variants, into production within the microcompartment microfluidic element. An essential task is the provision suitable methods and devices for the stated purposes.

General description of the invention

The invention relates to a method or a method for the preparative synthesis of poly nucleotides using microfluidic systems, in particular microfluidic elements, or the use of microfluidic systems, in particular microfluidic elements, for the synthesis of polynucleotides, comprising

• a) the use of a starter molecule oligonucleotide ("starter"), which on a Surface of a reaction space of a microfluidic system, preferably a Mi Krofluidikelements, especially a microchannel, directly or via a spacer  is bound, preferably via its 5'- or in particular its 3'-hydroxy, or directly with one end bound to the surface of the reaction space Siert, and which one of the ends, preferably the 3 'end of a strand of the polynucleotide sequence to be synthesized;
• b) one or more subsequent steps of hybridizing one or several sequemers, which sequence sections from the sequence of a syn correspond to the thetizing strand or its complementary sequence and their terminal sequences each overlap so that they hybridize with ei nem terminal section from the starter molecule or the respective preceding sequencer and the subsequent sequemer or in the case of the result a terminal blunt end, for example after polymerase reaction steps, after producing a sticky end, its hybridization with the above allow overlapping area of a subsequent sequencer, and
• c) in the case of nicks between such sequential ligation or to complete missing sections of 1 or more nucleotides that are complete a section of the opposite sequencer section tare, polymerase reactions and, if necessary, subsequent ligation Eliminate resulting nicks, repeating steps (ii) and (iii) so many times until the desired polynucleotide is obtained.

The invention also relates to microfluidic elements with those obtainable according to the method (non-covalently or preferably covalently) bound polynucleotides or precursors from producing the use of microfluidic element precursors lend polynucleotides, in particular the bound starter as defined above, or the appropriately charged (= bound with (non-covalently or preferably covalently) Polynucleotide precursors) microfluidic elements for the production of the polynucleo tides and kits or apparatus for the production of polynucleotides, accordingly loaded microfluidic elements and optionally further components.

The use (non-covalent or especially covalent) on a surface in one Microfluidic system of bound oligonucleotide starter molecules for the in-situ synthesis of Po In the prior art, lynucleotides are neither mentioned nor suggested. It simplifies however, the respective steps for the introduction of reagents (such as buffers, Enzymes or substrates), washing steps etc., by binding the resultant in each case  Fixes molecules, and enables, in particular without additional support materials, such as Beads or column materials that are conventionally used as carriers for nucleic acid synthesis be set to work what z. B. facilitates the handling of microfluidic systems (no clogging, no complicated loading procedures with carrier materials).

Detailed description of the preferred embodiments of the invention

The general terms used above and below preferably have the meanings given below, and instead of general terms above and below the more specific definitions can be used independently, which describes preferred embodiments of the invention:
Polynucleotides refers in particular to those nucleic acid derivatives which contain several nucleotides, in particular ribonucleotides (in RNA) or deoxyribonucleotides (in DNA), in 3 '- / 5' binding, for example complete genes, cistrons, ribozyme sequences, each with or without one or several suitable upstream and / or downstream sequences, such as promoter sequences, reporter sequences, sequences which contain restriction enzyme interfaces, or the like. Preferred are DNA or RNA with 10 or more, in particular special 20 to 100,000, preferably 20 to 10,000, especially 50 to 500, for. B. 80 to 90 nucleotides. In addition, as described below in the definition of sequemers, derivatives of nucleotides can also be present in the sequence. Such species include, for example, base modifications or modifications of the sugar / phosphate backbone. For example, they can increase resistance to nucleases (for example in the case of the manufacture of antisense molecules) and can therefore be sought. Polynucleotides according to the disclosure can in particular be double or single-stranded.

For example, in the 2-position of the ribose ring SH, SCH ₃ , F, N ₃ , CN, OCN, OCH ₃ , O (CH ₂ ) _z NH ₂ or O (CH ₂ ) instead of a hydrogen (DNS) or hydroxy (RNS) _z CH ₃ , in which z is 1 to 10, O (CH ₂ CH ₂ O) _v CH ₃ in which v is 0 to 12, in particular 1 to 3, CH ₂ CH (CH ₃ ) OCH ₃ , CH ₂ CH (OH ) CH ₂ OH, Cl, Br, CF ₃ , ONO ₂ , NO ₂ , NH ₂ , O-, S- or NH-C ₁ -C ₇ alkyl are present. Instead of the phosphodiester bond, a phosphorothioate, phosphorodithioate, sulfate, sulfonate, sulfonamide, sulfone, sulfite, sulfoxide, sulfide, formacetal, 3'-thioformacetal, 5'-thioether, hydroxylamine (with CH ₂ -NH-O-CH ₂ instead of Phosphodiester bond O - [(HO-) P (= O)] - O-CH ₂ ), methylene (methylimino) (with CH ₂ -N (CH ₃ ) -O-CH ₃ instead of the phosphodiester bond); Methylene oxy (methylimino) (with CH ₂ -ON (CH ₃ ) -CH ₂ instead of the phosphodiester bond), methylene - ((methylimino) -methylimino) (with CH ₂ -N (CH ₃ ) -N (CH ₃ ) - CH ₂ instead of the phosphodiester bond), car bonate, 5'-N-carbamate, amide (with CH ₂ - (C = O) -NH-CH ₂ instead of the phosphodiester bond, see he WO 92/20823), morpholino-carbamate (see Summerton, JE and Weiler, DD, US Patent No: 5,034,506) or in particular a peptide nucleic acid (PNA, see PE Nielsen, M. Egholm, RH Berg, O. Buchardt, Science 254, 1497 (1991)) or a " Locked Nucleic Acid "(LNA®, e.g. with bicyclo [2.2.1] ribunucleosides or xylosides, in which the oxygen of the 2'-OH group (or instead an S or an -NH-) and the C-4 'The ribose or xylose are connected to one another via a methylene group, see e.g. Pharmaceutical Manufacturing Int., 127-130 (1999) or Nucleosides Nucleotides 18, 1365-70 (1999) and the references mentioned therein; Exiqon AS, Vedbaek, Copenhagen ) are available (overview article on modifications: s see also Milligan et al., J. Med. Chem. 36 (14), 1923-37 (1993), and Uhlmann et al., Chemical Reviews 90 (4), 543-84 (1990)). Such modifications, in particular LNA sections, which have a particularly high affinity (usually 3 to 8 ° C. higher T _M , see below than in the case of unmodified nucleotides) in base pairings, are preferably found in the (in particular 3 ′ -) sections of Sequemers that are hybridized to the previous sequencer or starter (the advantage of this is that enzymatic reactions such as polymerase or ligation reactions are not hindered in the non-hybridized part). The invention therefore also relates in particular to the use of sequemers in the process according to the invention which (especially in this overlap region) contain modified nucleic acids, in particular LNA.

Suitable bases are the standard purine bases (adenine, guanine) or the standard pyrimidine bases (thymine, uracil and cytosine) or further derivatives, for example Xan thin, hypoxanthine, N-methyladenine, N-benzoyladenine, 2-methylthioadenine, 2-aminoadenine, 6 -Hydroxy purine, 2-amino-6-chloropurine, 2-amino-6-methylthiopurine, N-isobutyrylguanine, 5-fluorouracil, 5-chlorouracil, 5-bromouracil, dihydrouracil, 5-methylcytosine, 5-propynylthymine, 5-propynyluracil or 5 -Propinylcytosin. In addition, markers, in particular biotinylated ones (for indirect labeling methods) or by dyes, in particular special fluorescent dyes, such as fluorescein, rhodamine, hydroxycoumarin, benzofuran, Texas red, biman or ethidium / Tb ³⁺ , or luminescent markers, such as (iso ) -luminol derivatives, acridinium esters, Ru ²⁺ - (2,2'-bipyridyl) ₃ complexes, rhodamine or luminol, labeled nucleotides are used, which facilitates the detection of the resulting molecules and / or the synthesis of labeled molecules, such as probes, allows.

Further modifications can be found below in the section on chemical ligation. So far such modifications can hinder the ligation or polymerization reactions, the modifications can be found in the sequence in those places where they are the Do not interfere with ligation or polymerase reactions.

"Microfluidic systems" (= microfluidic systems) are preferably those systems which comprise one or more passages, chambers or lines which have at least one internal cross-sectional dimension, e.g. B. depth, width, length, diameter, etc., in microns dimension, in particular of 800 microns or less, especially in the range of 0.1 to 500 microns, z. B. between 5 and 500 microns. The microfluidic systems according to the invention preferably contain at least one microfluidic element with at least one channel on a μm scale (microchannel) or more thereof, which can also intersect and can be in a wide variety of shapes or geometries, for example linear, sawtooth-shaped, branched, meandering, spiral, circular or the like. Intersection points, if present, can be present, for example, as an intersection, T-junctions or the like. A microfluidic system for the purposes of the present invention preferably comprises a microfluidic element, ie a structure which contains one or more microchannels and microchambers, in particular a microchannel or a microchamber, in an integrated form, for example an array (chip), and in a monolith structure or is preferably constructed from two or more layers or parts, for example with a cover and a substrate layer, which defines the channels and chambers of the microfluidic element, for. B. As in Fig. 1. Preferably there is an essentially planar substrate layer which has at least one essentially flat surface with microfluidic channels and / or chambers, in particular a microchannel or a microchamber, on which a likewise essentially planar Top layer is placed.

A number of materials can be used for the top and substrate layers the. Preferably, since the devices are micro-engineered, the mate rialien selected so that they are compatible with known microfabrication techniques, z. B. photolithography, chemical wet etching, plasma or laser ablation, injection molding or other primary molding methods, pressing, embossing, CVD coating technology and the like chen. The substrate materials are also selected according to their suitability for the whole range of conditions to which the microfluidic systems are to be exposed finally extreme temperature, pH or salt concentration conditions or the An  application of electric or magnetic fields. In a preferred aspect of the inven The materials are materials that are also used in the semiconductor industry find application where microfabrication techniques are regularly used, in particular silicon-based, such as silicon, silicon dioxide, glass, quartz, silicone or polysilicon, or other semiconductor materials such as gallium arsenide. For example, the glass is FOTU RAN® from Mikroglas Technik AG, Mainz, Germany, if necessary with hydro xy-bearing silanes silanized, usable. In the case of semiconductor materials, it can often be appropriate, an insulating layer, for example made of silicon dioxide, over the material to layer, particularly where electrical fields are applied to the microfluidic element should find. Other preferred materials include a. Polymers, e.g. B. plastics, such as Po lymethyl methacrylate (PMMA), polycarbonate, polyester, polyamide, polytetrafluoroethylene (TEFLON®), polyvinyl chloride (PVC), polydimethylsiloxane (PDMS), polysulfone, polystyrene, Po lyolefins, such as polymethylpentene, polypropylene or polyethylene, polyvinylidine fluoride, acrylic nitrile-butadiene copolymer (ABS), block copolymers or mixtures of two or more of usable. Such polymeric substrate and cover layers are known, For example, the above-mentioned, microfabrication processes can be produced, or by means of Microtechnically produced mother molds using known molding processes ren, such as injection molding, embossing or stamping, or by polymerization of the monomer Precursor material within the form (original form). Such polymeric materials are coming trains because they can be easily manufactured, inexpensive and also disposable. However are Glass-based microfluidic elements are particularly preferred. All materials mentioned can surfaces or sections of the surface that have been specifically treated or coated (e.g. contained within the microchannels), for example by silanation (coating with a silane to which a compound molecule carrying a hydroxyl group is coupled is), etching of glass so that the surface has many hydroxyl groups, for example wise with Caro's acid, or otherwise introducing reactive groups (e.g. epoxy groups, activated ester groups, diene or dienophile groups or the like).

The microchannels and / or chambers of the microfluidic elements are introduced, in particular using the customary, in particular the microfabrication techniques mentioned above, into the upper layer of the substrate layer, as micro-scale pits or notches or the like. The cover layer ( 5 in FIG. 1) preferably has two mutually opposite, essentially parallel layers and can itself contain microchannels and / or chambers on the surface or simply cover existing microchannels or chambers in the substrate layer and themselves be completely planar, or it can also include one or more apertures, openings or connecting elements which penetrate to the outside.

Preferably, a surface of the cover layer and the side of the substrate layer, which comprises the microchannels and / or chambers, are connected to one another, for example by clamping, gluing or fusing, so that the cover layer completes and seals the channels or chambers upwards. To the outside leading supply and discharge elements for liquids, such as 3 and 4 in Fig. 1, and / or electrical conductors who led through the cover layer or laterally or below on the substrate layer.

In preferred embodiments of the invention, the microfluidic elements include optical detection window, which, for example, controls processes in micro channels or chambers allowed. Preferably the whole cover layer or the whole Substrate layer optically transparent, or in particular both. Optical detection window or the correspondingly transparent layers are in particular capable of UV, vision passable or infrared light. For example, are quartz or glass or transparent Polymer materials such as PMMA, polycarbonate suitable optically transparent materials.

The optically transparent areas are particularly suitable for monitoring and monitoring trolls. Suitable detection systems for this are, for example, those that use colorimetric, can detect fluorometric or radioactive signals or chemiluminescent signals, or further changes in refractive index or density, e.g. B. by means of photoacoustic Institutions. Examples of suitable detectors are spectrophotometers, photodiodes, Photomultipliers, microscopes, scintillation counters, cameras, cooled charge-coupled devi CE cameras (CCD), films and the like. Optical detectors are particularly preferred onssysteme; for example, fluorescence-based signals are activated using laser Fluorescence detection systems measured with lasers that have a suitable wavelength have to activate the fluorescent indicator within the system. The fluorescence is then measured, for example, using a photomultiplier tube. For the colorimetri detection are preferably used spectrophotometric detection systems det, which detect a light source directed at the sample and a measurement of the absorber or the optical permeability (see. The Photonics Design and Appli  cations Handbook, Books 1, 2, 3 and 4, annually by Laurin Publishing Co., Berkshire Com mon, Pittsfield, MA, USA with sources of optical components).

Non-optical detection systems can also be used, for example Tempe temperature sensors (useful for endothermic or exothermic reactions), conductivity, impe danz (e.g. by ISFETS), potentiometry (pH, ions) or amperometry (when using oxi dierbaren or reducible reagents, such as oxygen or organic oxidizable or reducible reagents). Examples of useful detection systems are immunological Systems, such as those that allow detection of double-stranded DNA, or especially those based on intercalating compounds that are fluorescent, Fluorescence quenching or photometric measurements measure the length syntheti enabled polynucleotide molecules.

The microfluidic elements can also include sections in which chromatographic Carrier materials, e.g. B. molecular sieves based on agarose, cellulose, polyacrylamide, Polymethacrylamide or polydimethacrylamide, ion exchange resins, separating materials for chromatography based on hydrophobic interactions, affinity resins, Gelaus final resins and the like, are present, for example, direct separations from Pro products of reagents.

The use of microfluidic systems requires appropriate control devices increasing the flows and residence times of solutions with buffers, reagents, enzymes etc., such as external pumping systems that can deliver liquids in a reproducible manner reproducible and accurately controlled quantities, and also the purity and desired if necessary, maintain sterility of solutions. The 205U Multichannel Cassette Pump from Watson Marley, Inc. (USA) is an example of such a pump. Alternatively, you can miniaturized mechanical pumps based on microelectromechanical systems (MEMS) can be used. These miniature pumps can be used internally or externally Microchannels and reaction chambers or the like. An example of such a pump pen are those of Shoji et al. in Electronics and Communications in Japan, Part 2, 70, 52-59 (1989), or diaphragm pumps, thermal pumps or the like. Other suitable fluids through microchannels include electro kinetic pumps - based on the principle of electroosmosis - e.g. B. as from Dasgupta et al. in anal. Chem. 66, 1792-1798 (1994), or electrophoretic metho  the z. B. the use of inert metal electrodes (z. B. of gold or platinum), the with external or internal circuits or electrical connections and suitable Control devices are in contact, require (see e.g. US 5,858,195).

The inflow and outflow can also be regulated by micro valves on the piezoelek trical bending principle, for example with piezo elements based on polysilicon or lithium niobate, based on ultrasonic sensors, based on tongue-shaped elements, that under the influence of an external magnetic field due to magnetorestrictive forces be bent, slide valve-controlled valves or fluidic elements where control small liquid movements big ones, and the like work. Derar term elements are e.g. B. described in US 5,837,200.

Electrical control devices are also required, for example for controlling Pumps, light elements and the like. An example can be found in WO 98/00707.

Microfluidic systems include, in particular, the micro elements mentioned, preferably channels and / or chambers and also detection systems (for example sensors, Detectors or the like) and optionally current-conducting coatings or Lei tion, microvalves and pumps as well as heating or cooling elements (except half or within the microfluidic elements) or other components, in particular how mentioned above. Control systems can also be used for direction and in Intensity of fluid movements, detection means for the control of the environment (at example temperature) of the microfluidic element, and units for detection or Control of data from the microfluidic element itself is present, such as voltage control elements, etc. These and other elements are preferably computer-controlled.

Starter molecule oligonucleotides can be directly on the surface (possibly via spacers) of each micro channel or the micro chamber are synthesized, starting from their 5'- or preferably the 3 'end. Then they can contain sequences that precede the egg Restriction cleavage sites to be synthesized occasionally, in particular those for type II restriction endonucleases. The separation of the finished syntheti The polynucleotide of interest can then be cleaved with the corresponding Restriction endonuclease (e.g. Alu I, Bam HI, Bgl I, Bst I, Eco RI, Eco RII, Fok I, Fnu Dl, Hae II, Hae III, Hind II, Hind III, Hpa I, Msp I, Not I, Pst I, Sac I, Sal I, Sau 3A, Sma I, Taq I,  Xho I, Xma I). Alternatively, the finished polynucleotide can be cleaved by Cleavable spacers occur (for example photolytically, or on a chemical one Ways, e.g. B. by acid or base treatment, reductive (z. B. reduction of bin end disulfide bridges) or thermally (for Retro-Diels-Alder reaction), or by Aufhe Practice of non-covalent bonds (e.g. addition of biotin to prevent biotin / Avi din-binding). In the case of in situ synthesis, for example (if necessary after Coating with a hydroxy-bearing silane) a spa bearing a hydroxyl group coupled molecule. Starting from this, the oligonucleotide synthesis is carried out according to übli Chen process, especially based on the CPG phosphoramidite process 3 'end of the starter molecule. In modified processes, the hydroxyl group of Ver binding molecule with a photolabile protecting group, such as pyrenylmethyl, hydroxystyryl (dialkyl) silyl, dimethoxybenzoinoxycarbonyl (DMBO), methylnitropiperonyloxycarbonyl (MeNPOC) or in particular 2-nitrophenylpropyloxycarbonyl (NPPOC), which after UV radiation (for example with a mercury vapor lamp, for example 100 W, or a laser, especially a XeF laser with a 351 nm wavelength) can be removed and thus the coupling of another desired phosphoramidite nomers enables. Starter molecules preferably have a length of 4 to 100 preferably from 6 to 80, for example from 6 to 50 nucleotides, unless otherwise stated give. However, they can also be longer.

Spacers (or linkers) are in particular those molecules that make it possible for the starter molecule to be synthesized at a sufficient distance from the surface of the Mi crocodile or the microchamber to reduce steric hindrance, and the furthermore preferably the cleavage of the available polynucleotide after completion Enable reaction. For example, there are photolabile linkers (less before if a photolabile protecting group such as NPPOC is used), acid or Ba Senlable linker, reductively cleavable linker (for example with disulfide bridges), or one thermal cleavage accessible linkers, e.g. B. cleavable in a Retro-Diels-Alder reaction. Examples of suitable spacers are polyethylene glycols, dicarboxylic acids (such as succinic acid right), polyamines and alkylenes, or cross-linkers such as N- (p-maleimidophenyl) isocyanate. Be linkers which are via groups prepared by the Diels-Alder reaction are particularly preferred the surface or the starter are bound, or contain such groups internally, and split back into diene and dienophile by thermal reaction (warming) can, so that the synthesized polynucleotide can be cleaved.  

"Bound" or "binding" (e.g. via a spacer element) does not mean covalent (e.g. about strong non-covalent bonds between the partners of bond pairs, such as between biotin and streptavidin or the like) or preferably covalently (via covalent bond). Bound means that with the usual washing and Reagent delivery steps do not dissociate from the surface (except in the case of deliberate detachment) of the microfluidic element, e.g. B. microchannel takes place. The connection polynucleotides and their end stages are preferably terminal.

Non-covalent binding refers to binding that occurs via binding pair interactions happen, one of the components of such a binding pair (direct or via spacers, covalently or non-covalently) to the surface of the microfluidic element, e.g. B. Microchannel, the other (directly or via spacer) is bound to the precursors of the synthesizing polynucleotides or the polynucleotides. Binding pairs can e.g. B. En zym / substrate, antigen / antibody, glutathione (derivative) / glutathione-S-transferase, lectin / lek tin binding protein or in particular biotin / avidin or biotin / streptavidin spacers also cause a non-covalent bond (e.g. binding via biotin on the surface area of the microchannel bound avidin or streptavidin or the like). Bin too Coupling pairs themselves can be spacers.

Directly synthesized means that the starter molecule is attached directly to the (optionally derivatized) surface (possibly via a spacer) of the microchannel or the Mi Krokammer is manufactured. Alternatively, a (preferably on the not on the Functional groups to be involved, such as amino or hydroxy groups Protected group protected) oligonucleotide as a starter molecule directly via its free 5'- or preferably 3'-hydroxy bound to the surface (if necessary via a spacer) to, for example, activated ester groups (such as 2,4-dinitrophenyloxycarbonyl), epoxy groups, dienes or dienophiles (e.g. in the case of allyl-derivatized nucleotides) or the like.

Including one end, especially 3 ', means that the corresponding starter molecule ent has neither only the terminal sequence of the polynucleotide to be produced (esp particularly usable with cleavable spacers) or optionally at the 5 'or preferably 3' end includes additional sequences, for example with restriction sites or others Sequences that are used, for example, for incorporation into vectors, such as plasmids, cosmids, Pha gemide, YACs or the like can be used.  

For the hybridization of overlapping complementary sections of sequeners to the starter molecule or with one another, stringent hybridization conditions are used above all. Stringent hybridization means that the hybridization is carried out under conditions in which the unspecific hybridization is essentially suppressed, so that essentially only the specific base pairings contribute to the formation of double strands. In general, more preferred highly stringent hybridization conditions are obtained if one works approximately 5 to 15 ° C below the "melting point" (T _m ) of the paired sequences at a defined pH and ionic strength. The T _m value is the temperature at which 50% of the target sequence binds to the complementary sequence for a given ionic strength and pH. Very strict conditions are chosen so that they are almost identical to the T _m for a specific pair (e.g. 0 to 5 ° C below the T _m value). The setting of suitable hybridization conditions is known to the person skilled in the art and can be carried out, for example, according to protocols such as in Sambrook et al., "Molecular Cloning, a Laboratory Manual", CSH Press, Cold Spring Harbor (1989) or in Hames and Higgins, "Nucleic Acid Hybridzation "A Practical Approach", IRL Press, Oxford 1985.

Blunt ends are ends of a polynucleotide (especially DNA) dop pelstranges, in which there is no unpaired nucleotide. Sticky ends ("sticky or knockout adhesive ends ") are double-stranded polynucleotide (especially DNA) molecules that have unpaired nucleotides at the free end. Especially for making one Sticky ends from blunt ends are suitable for cleavages with class II restriction enzymes or exonucleases (especially to generate free 3 'ends of lambda exonuclease or T7 exonuclease, New England Biolabs Inc., Beverly, MA, USA). Especially the Er Generation of free 3 'sticky ends is important for those made only with polymerases recursive procedure. See also DE 198 12 103. Sticky ends can also be obtained as a result that a single strand not bound to the surface of the microelement of a double strand is completely removed by dehybridization.

Sequemers are polynucleotide precursor molecules, the sequence segments from the sequence of a strand to be synthesized or its complementary sequence and whose terminal sequences each overlap so that they hybridize with a (Complementary) terminal section from the starter molecule or each  enable previous sequemer and subsequent sequemer, and are on synthetic route according to standard methods (then you can also use modified Nucleo tid structures include, as described above) or (for example as sections) from na naturally occurring sequences or sequence sections accessible. Have sequemers preferably a length from 4 to 1000, preferably from 6 to 100, in particular from 10 to 50 nucleotides unless otherwise stated. However, they can also last longer his. The mutually complementary overlap regions preferably each comprise because at least 3, preferably at least 4, for example 5 to 30 base pairs.

Nicks are simple breakpoints within oligo- or polynucleotides, i.e. that is, it is le digest only the backbone of the corresponding molecule without nucleosides missing len. Ligation reactions (ligation) allow the connection of the free ends of the back grats. Here, enzymatic and chemical ligation can be combined or only as desired one of these methods can be used. Suitable ligases or chemical ligations reactions are known to the person skilled in the art. Examples of chemical ligation are (i) metho for the ligation of the 3 'and 5' ends of unmodified oligo- or polynucleotides, like the cyanogen bromide method, either using a 3'-phosphate and a free 5'-hydroxy (see e.g. Dolinnaya, N.G., et al., Nucleic Acids Res. 19, 3067-72 and 3073-80 (1991); Shabarova, Z.A., et al., Nucleic Acids Res. 19, 4247-51 (1991); Dolin Oh well. N.G., et al., Nucl. Acids Res. 21: 5403-7 (1993); Kanaya, E., et al., Biochemistry 25, 7423-30 (1986); Luebke, K.J., et al., J. Am. Chem. Soc. 111: 8733-5 (1989); Rubin, E., et al., Nucl. Acids Res. 23: 3547-53 (1995); Dolinnaya, N.G., et al., Nucl. Acids Res. 16, 3721-38 (1988); or Fedorova, O.A., et al., Nucleosides Nucleotides 15 (6), 1137-47 (1996)) or a 5'-phosphate and a free 3'-hydroxy (James, K.D., et al., Chem. Biol. 4, 595-605 (1997)) or coupling methods using other coupling reagents, such as dicyclohexylcarbodiimide or in particular water-soluble Licher carbodiimides, such as ethyldiaminopropylcarbodiimide (see Harada et al., J. Mol. Evol. 38: 558-60 (1994); or (ii) chemical reactions to connect neighboring nuc leoside building blocks in which at least one of the building blocks is a modified nucleoside represents, for example by using oligonucleotide phosphoazides as reagents for chemical ligation (see Isagulyants, M.G., et al. Khim. Prir. Soedin (1987), the End ligation of homopyridine oligodeoxyribonucleotides (see e.g. Li, T., et al., Chem. Biol. 4, 209-214 (1997)), the ligation of oligodeoxyribonucleotides with a mercapto group in the 5'-position of the 2'-deoxyuridine via a disulfide bridge (see Ueno, Y., et al.,  Nucleosides Nucleotides 17 (1-3), 283-9 (1998)) or ligation of one with a 5'-iodo group modified oligonucleotide to a 3'-end modified as 3'-thiophosphate of the adj beard oligonucleotides (see e.g. Xu, Y.Z., et al., Tetrahedron Lett. 38, 5595-8 (1997)). Examples of enzymes for the enzymatic ligation are the DNA and the RNA ligase the bacteriophage T4 or the ligase from E. coli, for RNA ligation the RNA ligase from the bacteriophage T4. DNA molecules that lack a terminal phosphate be prepared for ligation by polynucleotide kinase from bacteriophage T4. Also thermostable ligases, e.g. B. taq ligase can be used, especially for cyclical ligation processes.

In contrast to Nicks, missing sections are characterized by a or several nucleosides of a sequence are missing. Polymerases enable the specific Filling these gaps (hereinafter "polymerase reaction"). As soon as there are only nicks left are closed by ligases. Chemical ligation (see above) is possible. Suitable polymerases and polymerase reactions are known to the person skilled in the art known. For the enzymatic polymerization, nucleoside triphosphates (which may also be used marked, e.g. B. by biotinyl residues or by fluorescent residues) as a sub strate used. There is an abundance of nucleic acid polymerases available, so probably from prokaryotic as well as from eukaryotic sources or viral origin. This are, for example, the DNA polymerase I from E. coli, in particular the Klenow fragment (without exonuclease activity), the bacteriophage T4 DNA polymerase, DNA polymerases α and β from mouse cells (A-9) or human cells (HeLa), and the DNA polymerase of Herpes simplex virus, or thermostable polymerases such as Taq, Pfu, Tth, Pwo or Pfu- Polymerase. The response to fill gaps can be, for example, under the conditions according to Bourguignon et al., J. Virol. 20, 290, (1976). For the production from DNA to RNA matrices can e.g. B. reverse transcriptase from retroviruses, such as from mouse or bird retroviruses. For the RNA production z. B. RNA polyme lawn from E. coli or bacteriophages, such as T7, SP6 or T3, or HeLa cell RNA polymers rase III, calf thymus RNA polymerase II or mouse cell RNA polymerase II (eukaryotic RNA polymerases) can be used. Mixtures of two or more polymerases are possible.  

Suitable conditions for ligation and polymerase reactions can be found e.g. B. in Sam brook et al., "Molecular Cloning - a Laboratory Manual", CSH Press, Cold Spring Harbor 1989, in Gassen et al., "Genetic Engineering Methods - A Collection of Instructions for the Molecular Biology Laboratory", Spektrum Akademischer Verlag, (ed.) Heidelberg 1999, FM Ausubel "Short Protocols in Molecular Biology", ^3rd ed, New York, Wiley. 1997; or in Asubel et al., "Current Protocols in Molecular Biology", vol. 1-3, Greene Publishing Associates and Wiley-Interscience, New York, 1987; as well as references mentioned therein.

"To include" means "to include at least". B. more in the process Steps may optionally be added to parts of devices such as microfluidic systems, more pieces.

Protecting groups are those groups which prevent the reaction of functional groups which are not intended to take part in a desired reaction and can be split off again at a suitable reaction stage without undesired side reactions. In games for such protecting groups, their introduction and their separation are known, for. , As described in Greene et al., Protective Groups in Organic Chemistry, ^2nd ed., John Wiley & Sons, otherwise mentioned New York (1991), or especially as in the present disclosure.

Further steps to be added in preferred embodiments are the spin-off of the polynucleotide obtained, which acts as a double strand with cleavage of the bond to Surface of the microchannel or the microchamber (for example by means of restrictions donucleases or a photo-cleavable or by base or acid or reducing agent tel, such as dithiothreitol, cleavable spacer group) or under the hybridization Conditions can be obtained by extracting the unbound single strand can, or by abolishing a non-covalent interaction (as between biotin and streptavidin). The product can be further cleaned, for example by gel electro phoresis (e.g. agarose gels), gel chromatography, HPLC, ion exchange chromatography or other standard procedures, or directly for the desired purpose ke are used, for example as a vaccine (possibly with adjuvants), or (possibly after Incorporation in vectors, such as plasmids) for the transformation of host cells or the like.

After the production of a polynucleotide, this can be amplified in a further step be, for example by polymerase chain reaction (PCR), ligase chain reaction  (LCR), Strand-Displacement Amplification (SDA), Nucleic Acid Sequence Based Amplifica tion (NASBA) or Branched DNA amplification (bDNA), or combinations thereof. RNA based amplification systems are for example the QB RNA replicase method and RNA transcription-based amplification systems (TMA). Other modifications are possible, for example adding sticky ends, restriction interfaces, etc.

Microfluidic elements with non-covalent or preferred available according to the process wise covalently bound polynucleotides (especially single strand or partially or completely double strand, the strand in the former case and one of the two in the second terminal strands are bound via its 5'- or in particular 3'-hydroxy) all those with on the surface of the microchannels or chambers, in particular one Microchannel or a microchamber, directly or via spacers non-covalently or preferentially as covalently bound via the 3'- or 5'-hydroxy starter molecules to the esp special [preferably already linked by polymerase and / or ligation treatment finished, optionally already converted into individual strands by strand separation Polynucleotides or moreover not only overlapping complementary by hybridization Sequences of the starter molecule and one or more sequences) or only partially already ligated and / or polymerase treated] at least partially double-stranded Starter / sequemer complexes are bound.

Also the use (not covalently or in particular covalently) on microfluidic elements bound precursors of the polynucleotides to be produced, in particular those bound Starter, as defined above, or the microflui charged accordingly with a starter dikelemente for the production of polynucleotides, the production in particular the Steps (i) to (iii) defined above and below is part of the invention.

Kits or apparatus for the production of polynucleotides that are loaded accordingly Microfluidic elements comprise, in particular comprise one or more microfluidic elements elements with (directly or via a spacer) on the surface of one or more microka channels or chambers not covalently or preferably covalently above their 5 'or preferred wise 3'-hydroxy bound starter molecules. There are also additional optional compos nenten, e.g. B. reagents and enzymes for ligation and / or polymerase reaction, which in ge suitable containers, such as vials or ampoules, are filled, for example in lyophi lized form or in suitable buffer solutions, and / or suitable reagents for the  Hybridization of sequemers. Other components, such as detectors, computers, interfaces, pumps, pipetting systems etc.

Preferably, the starters and / or especially the sequemers are micro synthetic method used. In particular, photolabile protecting groups are proposed preferably used as mentioned above for the in-situ synthesis of starter molecules. the Like methods in particular allow many different oligonucleotides on one Synthesize surface. So z. B. the 65536 possible 8-mer oligonucleotides, which leads to a gene synthesis from combinatorial monomers (starter molecule and Se quemere), s. u., are used by extremely flexible bio-array synthesis machines are synthesized on a solid phase. Such BioChip synthesis machines are today mostly using digital mirrors (digital mirror arrays), in particular manufactured by Texas Instruments. Synthesis machines of this type are used Synthesis of BioArrays for example from Skip Garner (Houston, Texas, Nimblegen (Ma dison, Wisconsin, USA) or FeBit (Mannheim, Germany). Visually controlled te solid-phase microsynthesis for the production of sequemers for combinatorial Gene synthesis meets all the requirements for the synthesis of oligonucleotide precursors molecules and the synthesis of long double-stranded polynucleotides on a microflui unite dikelement: The starter and the possible sequemers in particular manufactured at different defined locations of a microfluidic element, then by local exposure to light or heat cleave the oligonucleotides required and introduced to the starter or the sequemers already hybridized to the starter for hybridization and further reaction. Synthesis with (covalent bound, preferably via the 5'- or especially the 3'-hydroxy) combinatori oligonucleotides as starter molecules of pre-loaded microfluidic elements, especially arrays that originate from an oligosynthesis machine equipped with digital mirrors and their use within a gene synthesis machine that is capable of also to control the microfluidic elements in such a way that the polynucleo according to the invention tide synthesis can take place there, especially as just described. For example Such an array can be sent to a customer and there using the gene synthesis scheme machines for producing a polynucleotide, for example a gene, can be used.

Preferably only one polynucleotide mole is used in a microchannel or microchamber kül (a species) synthesized, but can be made using the above with  Synthesized machines also mirror several different molecules on ver different locations of one and the same microchamber or one and the same microchannel be synthesized.

A preferred embodiment of the invention relates to a combinatorial polynucleotide synthesis method which leads to the double-stranded polynucleotide, in particular DNA, or (after dehybridization) to single-stranded molecules. For this purpose, all possible sequence combinations for a given oligonucleotide length for the synthesis are prepared in a first step, preferably as described above, that is, for. B. for hexamers 4 ⁶ = 4096 or for octamers 4 ⁸ = 65536. The use of this combinatorial method for polynucleotide synthesis comprises (cf. FIG. 3):

• a) the use of a starter molecule oligonucleotide (“starter”) (for example 8 in FIG. 3) from the collection of molecules with the combinatorially produced sequence combinations, which on a surface of a reaction space of a microfluidic system, preferably a microfluidic element, in particular a micro channel, is bound directly or via a spacer, preferably via its 5'- or in particular its 3'-hydroxy, and which comprises one of the ends, preferably the 3 'end, of a strand of the polynucleotide sequence to be synthesized;
• b) then recursive steps of hybridization of one or more sequencers (e.g. n, n + 1,..., n + 5 in FIG. 3) from the collection of molecules with the combinatorial sequence combinations, which sequence segments correspond to the sequence of a strand to be synthesized or its complementary sequence and their terminal sequences each overlap in such a way that they enable hybridization with a terminal section from the starter molecule or the respectively preceding sequencer and the subsequent sequemer, and
• c) in the case of nicks between such sequencers, after at least two sequemers have been hybridized in the first step (ii) or in subsequent steps several partially overlapping sequemers or preferably one partially overlapping sequemer have been hybridized to the growing poly nucleotide to be produced, ligation (or , if necessary, to complete missing sections from 1 or more nucleotides, which are complementary to a section of the opposite sequencer section, polymerase reactions and then ligation, for example in the non-hybridizing part of the starter molecule oriented towards the binding surface) to remove the resulting render Nicks,
  repeating steps (ii) and (iii) as many times as necessary until the sequence of the desired polynucleotide is completed;
• d) and finally preferably release of the finished polynucleotide by Dehybridization (release of a single strand while the other is tied remains) or chemical cleavage from the surface or the spacer (e.g. pho chemically, by restriction enzymatic cleavage, acid or base action, reductive or thermal elimination (especially Retro-Diels-Alder- Reaction) and, if desired, cleaning, preferably as described above.

The use of shorter starters and Se is preferred for this combinatorial approach quemers, especially with 6 to 10, preferably 6 to 8 nucleotides, otherwise the number of the possible sequence variants and thus the administrative effort increase exponentially. For recursive, combinatorial procedures, ligation is the preferred response to the Link sequemere and the starter. The brevity of the single-strand precursor mo leküle usually does not allow efficient polymerization of the polynucleotide, especially DNA, with template-dependent polymerases starting from the 3 'end of one chain. On the other hand, the ligation has the disadvantage that it is relatively inefficient: the shorter the Sequemers are, the more ligation reactions are required. Here it may be necessary be to determine the optimal conditions (especially optimal nucleotide length).

A broader embodiment of the invention also relates to a method in which the syn thesis starting from a non-terminal, over a side group of the starter molecule (for example an amino group or an allyl group of a modified base) the surface of the microfluidic element-bound starter is bidirectional; but can NEN here because of the small overlap areas short n. , ., n + m forerunner at the spe specific sequencing problems due to low specificity.

Another embodiment of the invention relates to methods of sequence-individualized Synthesis (sequence-individualized means: starter molecule and sequemers are not out combinatorial molecules selected, but individually manufactured or isolated) - here can use non-recursive ("one tube") and recursive methods, or com combinations of these.  

The non-recursive method (see FIG. 2A) is based on the trust that the selectivity of the hybridization reaction of the sequeners and the starter is large enough that all the sequencers are found in the correct order and are completely ligated. In addition to simple ligation, cyclic ligation processes with thermostable ligases are also used. These methods work, but the occurrence of wrong products is possible, so the sequence has to be checked and if necessary the gene function has to be tested until the right product is found.

The invention thus preferably also relates to a process for the synthesis of Polynuc leotides, comprising

• a) the use of a starter molecule oligonucleotide ("starter") (z. B. 8 in Fig. 2A or 2B) with a specific sequence, which on a surface of a reaction space of a microfluidic system, preferably a microfluidic element, in particular a microchannel, directly or is bound via a spacer, preferably via its 5'- or in particular its 3'-hydroxy, and which comprises one of the ends, preferably the 3 'end, of a strand of the polynucleotide sequence to be synthesized;
• b) subsequent successive or simultaneous hybridization of one or more in each case (e.g. n, n + 1, ... n + 5 in FIG. 2A or n, n + 1, .. n + m in FIG. 2B ) which sequence sections correspond to the sequence of a strand to be synthesized or their complementary sequence and whose terminal sequences each overlap in such a way that they enable hybridization with a terminal section from the starter molecule or the respective preceding sequencer and the subsequent sequemer, and which together form the cover both complete double strands of the polynucleotide to be synthesized, so that only nicks are formed, or, if the ends of sequeners (including the starter) do not abut one another, also sections with unpaired nucleotides;
• c) and in the case of nicks between sequencers lined up in this way or for completing missing sections of 1 or more nucleotides which are complementary to a section of the opposite sequencer section, carries out polymerase reactions and then ligation to remove the resulting nicks;
  wherein steps (ii) and (iii), if not all of the sequeners are hybridized at once and then completed by ligase or ligase and polymerase reaction, are repeated as often as necessary until the sequence of the desired polynucleotide is completed;
• d) and finally preferably release of the finished polynucleotide and if necessary, cleaning, preferably as above (especially with the kombina toric polynucleotide synthesis method).

The recursive synthesis process is based on the combination of various advantageous properties of the variants presented above: Here, single-stranded, synthetic (or further obtained in another way) oligonucleotides (starters and sequemers) are used in independent steps, one after the other, in a suitable reaction as a template extend the growing polynucleotide (especially DNA) by hybridization and polymerase reaction ( Fig. 4). The use of polymerases with high activity, processivity and synthesis accuracy is a first, essential advantage over processes that work with ligases. The separation of the individual reaction steps of each addition reaction allows, unlike in "one tube" processes, largely to avoid incorrect hybridizations or incorrect sequence products. Since after the enzymatic synthesis reaction of a synthesis cycle there is a smooth end of the double-stranded molecule (in particular DNA), the provision of an unpaired 3'-ene to which the next n + t precursor molecule can be attached is necessary for this type of synthesis. There are several possible solutions here, for example the use of class II restriction endonucleases or exonucleases to generate sticky ends.

The invention relates in particular to such a recursive method with (preferably relatively longer) precursor molecules (cf. FIG. 4), comprising

• a) the use of a starter molecule oligonucleotide ("starter"), which on a Surface of a reaction space of a microfluidic system, preferably one Microfluidic elements, in particular a microchannel, directly or via a spa cer is bound, preferably via its 5'- or in particular its 3'-hydroxy, or directly with one end bound to the surface of the reaction space is synthesized, and which one of the ends, preferably the 3 'end, one Strand of the polynucleotide sequence to be synthesized;  
• b) a subsequent step of hybridizing a sequencer, which a Sequence section from the sequence of a strand of the poly to be synthesized corresponds to nucleotides and their terminal sequences in the case of the first hybridizers with a terminal section from the Starter molecule or in the case of further sequencing used sequentially mers of the same strand because of the polymerase mentioned below reaction blot ends and after production of sticky ends its hybridization with the overlapping area of a subsequent one Enable sequencers, and manufacture a sticky end (especially with a resulting free 3'-OH in the available single strand section); and
• c) to complete missing sections of 1 or more nucleotides, which are complementary to a section of the opposite sequence section, a polymerase reaction after each hybridization of a sequencer, and, if the covalent connection of the sequeners (also on the one not synthesized by the polymerase) Strand) is desired, if desired polymerase reactions to fill any gaps which may result (if the sequencers do not directly meet) and ligation or (if the sequencer ends meet) only ligation;
  wherein steps (ii) and (iii) are repeated as often as necessary until the sequence of the desired polynucleotide has been completed, and the covalent binding of the sequemers by ligation or polymerase reaction and ligation only after addition of more than one or all of the sequemers can be done;
• d) and finally preferably release of the finished molecule and ge if desired, cleaning, preferably as above (especially with the kombina toric polynucleotide synthesis method).

The "relatively longer precursor molecules", which are used here as sequemers, intend to preferably a length of 10 to 1000, in particular from 15 to 100, for example of 20 to 50 nucleotides. The starter can also be shorter and can in particular be 4 to 100 preferably 4 to 10, e.g. B. 6 to 8 nucleotides in length.

In a preferred embodiment, the invention also includes such syntheses which two or more methods, selected from combinatorial, sequence-individual ellen, recursive and non-recursive methods, especially the ones described above,  can be combined, which also includes processes in which intermediate stages isolated (e.g. cleaved and purified) and then further for the synthesis of polynucleotides be used. The combination of the production by means of a combi is particularly preferred Native synthesis for the preparation of longer precursor molecules, for example of 20 to 300, in particular 50 to 200 nucleotides in length, with subsequent use of the precursor molecules thus obtainable in the recursive method described above with polymerase reaction and ligation to longer polynucleotides.

A preferred embodiment of the invention also relates to the synthesis of a double strand polynucleotides according to one of the above methods, one strand of which is kova lent (especially via its 3 'end) directly or via a spacer to the surface a microchannel or microchamber is bound, the removal of the non-covalently bound strand by dehybridization, and the use of the resulting ko valent bound single strand for polymerase reaction to regenerate the complementary, non-covalently bound strand by polymerase reaction in the Mi Krofluidikelement and its release by dehybridization. This can be done any number of times repeated and so the free single strand can be produced in any number of copies.

Description of the drawings

Fig. 1: A schematic example of an inventive microfluidic (1), here embodied as an array of silicate glass; 2 = microchannel (meandering); 3 = supply line; 4 = derivative; 5 = top layer, here as a cover plate [shown here lifted off, connected to the lower part in the operating state (for example by gluing or clipping)]. Other elements such as brackets, pumps, controls, computers etc. are not shown.

Fig. 2A: Scheme for synthesis by the non-recursive ( "one tube") ligation techniques; For details see Examples 1 and 2. 6 = spacer; 7 = surface of the substrate (z. B. in the micro-channel ( 2 ) of FIG. 1); 8 = starter molecule (e.g. DNA). N, n + 1, n + 2, n + 3, n + 4, n + 5 = Seque mere. The sequemers and the starter molecule are shown here without ligation ("one tube").

2B, wherein the ligation and template-dependent polymerase reaction (.........) are combined the same ( "one tube"). Spacer not shown or absent in Fig. 2B.

Fig. 3: Recursive addition of sequemers: starter molecule, sequemers. Starter molecule and sequemers are obtained combinatorially. 6 = spacer; 7 = upper surface of the substrate (e.g. in the microchannel ( 2 ) according to FIG. 1); 8 = starter molecule (e.g. DNA); m = integer (number of oligonucleotides required to synthesize the target polynucleotide).

Fig. 4: Recursive method with (for example chemically synthesized) long precursor molecules. Starter molecule (spacer not shown here or absent); = Se quemere (after blunt ends have been eliminated, for example by means of exonuclease); ......... = template-dependent synthesized polynucleotide strand part; marks starting points of the polymerase reactions (preferably 3 'ends) after sticky end production. m, 7 , 8 as in FIG. 3.

Examples: The following examples serve to illustrate the invention without its Want to limit the scope.

example 1 Production of a DNA-90-mer (coding for the B chain of porcine insulin) (example for recursive method only with ligation) ( FIG. 2A)

On the surface of a microchannel with an approximately square cross-section, each with a 200 µm edge length perpendicular to the longitudinal axis of 5 cm in length, and with walls on a glass array made of silicate glass (a photo-structurable glass, e.g. FOTURAN® glass) according to FIG. 1 with a length of 5 cm and a width of 2 cm, the channel on the top being covered by a glass plate and the total reactive surface having an area of approximately 0.3 cm ² and approximately 10 pmol OH groups being available for binding, is using the phosphoamidite method and the 2- (2-nitrophenyl) propoxycarbonyloxy protecting group (see US 5,763,599) for the 5'-OH group of the nucleosides used, which after each coupling cycle by irradiation with a XeF laser at 351 nm (or a 100 W mercury vapor lamp) and a retro-Diels-Alder reaction which can be thermally cleaved at the 3 'end of a DNA 20-mer with a DNA which corresponds to the 3' end of the non-codogenic strand for corresponds to the porcine insulin B chain as a starter molecule (cf. Fig. 2) synthesized.

In a DNA synthesizer (e.g. ABI 380 B Automated Synthesizer, Applied Biosystems, Inc.), a DNA sequencer is synthesized independently of the phosphoramidite method, which contains the first 30 nucleic acids from the 5 'end of the coding strand for corresponds to the B chain of porcine insulin (see Sequemer n in Fig. 2). 20 µmol of this are hybridized to the starter molecule under stringent conditions and the excess is washed out with buffer.

Another DNA-30-mer (n + 1 sequencer) produced in the synthesizer is then which is the nucleotides 21 to 50 of the non-codogenic strand counted from 3 ′ for the Pork insulin B chain corresponds to the free end under stringent conditions of the n-sequencer hybridizes that all 10 base pairs are combined in the overlap area kidney.

The microchannel is then loaded with substrates and T4-DNA ligase (1 U / µl) (Gibco-BRL Life Technologies GmbH, Germany, Eggenstein) with T4-DNA ligase buffer and the starter molecule is ligated with the n + 1 sequencer , After the ligation is rinsed with buffer and a further DNA-30-mer (the n + 2-Sequemer, see Fig. 2), which was prepared in the DNA synthesizer and the sequence of nucleic acids 31 to 60 counted from the 5th '- End of the codogenic strand for the porcine insulin DNA corresponds, hybridized under stringent conditions, so that the overlapping regions of the n + 2 and n + 1 sequencers mate. Re-ligation joins the n- and n + 2-sequemers of the codogenic strand.

Another synthetic DNA-30-mer, which the nucleic acids 51 to 80 of the non- corresponds to the codogenic strand of the porcine insulin B chain gene (n + 3 sequencer) hybridizes with the overlapping portion of the ligated n + 2 sequencer and the n + 3- Sequemer with the n + 1 sequencer ligated analogously as above. Then again with one another DNA-30-mer obtained in the synthesizer, which contains nucleic acids 61 to 90 of the codogenic strand counted from the 5 'end, which corresponds to the B chain of the Pig insulin encoded (n + 4 sequencers), the overlapping part to the single-stranded end of the n + 3 sequencer hybridized and the n + 4 sequencer with the end of the n + 2 sequencer ligated. Finally, there is a DNA 10-mer which corresponds to the sequence of nucleotides 81 to 90, counted from the 3 'end, of the non-coding strand for the B chain from Schwei DNA corresponding to neinsulin (Sequemer n + 5) corresponds to the free single-strand segment of the sequencer n + 4 hybridized and then ligated to the n + 3 sequencer.  

The resulting double-stranded DNA, which encodes the B chain of insulin Sequence contains, is then split off by brief heating and can be used for different Purposes are used, for example for hybridization experiments or (for example after adding sticky ends) for incorporation into plasmids. About 10 pmol is obtained DNS (approx. 310 ng).

Example 2 Production of B-chain insulin DNA in a non-recursive way

To the starter molecule which is produced analogously to Example 1 and which is coupled to the surface of the microchannel, a hybridization intermediate product is initially produced by simultaneously adding a hybridization buffer which comprises all 6 sequencers (analogously n to n + 5 in FIG. 2) to which the individual sequemers, including the starter molecule, have not yet been ligated. The ligation is then carried out simultaneously for all league points as described in Example 1. The yield of correct molecules is somewhat lower here due to mismatches.

Example 3

Synthesis of the non-coding strand of the "green fluorescent protein" (GFP) from Aequoria victoria (Leptomedusae, Campanulidae) with chemically synthesized long precursor molecules using polymerase:
The coding counter strand of the non-coding strand to be synthesized for the fluorescent protein has the following sequence (cf. JP 1998234382):

Table 1 shows the synthetic (by normal chemical means or by combinatori cal synthesis using octamers) obtained oligonucleotides their use from the 3 'end first the sequence for the non-coding strand encode from nucleic acid 358 to 717 to the 3 'end of the complementary, shown above the strands (SEQ ID NO: 1):

Table 1

Oligonucleotides for the synthesis of the non-coding strand of the GFP gene

The individual procedure is as follows (short names of the oligonucleotides according to Tab. 1): On the surface of a microchannel as in Example 1, avidin (in a case tied with a binding silane). The 5 "-biotinylated oligonucleotide BIO1 (with a a pre-sequence containing a restriction site of 50 nucleic acids before the  coding sequence of the GFP gene complementary strand) is not covalently over the biotin residue is bound to the avidin (alternatively one is used with BIO1 except for the missing one Biotin identical oligonucleotide covalently to the substrate surface via a succinate Spacer bound - cleavage then takes place later using restriction endonuclease). The Oligonucleotides GFP-1b, GFP-2b, GFP-3b, GFP-4b and GFP-5b are at the 3 'end protects to prevent their prolongation in the polymerase reaction.

The 3'-protected oligonucleotide GFP-1b (at the 3 'end complementary to the last 18 Nucleic acids at the 3 'end of BIO1) are hybridized. A first polymerase reaction in Presence of nucleoside triphosphates and a Pfu / Taq polymerase mixture after extension of BIO1 to a partially double-stranded intermediate. Using T4 (Gen 6) -Exonuclease, the strand with the free 5 'end is partially removed, leaving the blunt End at the 3 'end of the remaining bound strand that at the 3' end of GFP-2b (last 20 nucleic acids at the 3 'end) end sequence complementary sequence as Sticky End is released. 3'-protected GFP is over this overlapping section 2b hybridized. This is followed by polymerase reaction and partial removal of the formed Strands to create a free 3 'sticky end as previously described. With that lies the again extended bound single strand. This now has the last one at the 3 'end ten 20 nucleic acids at the 3 'end of GFP-3b complementary nucleic acid sequence. to hybridization of this (3'-protected) section, subsequent polymerase reaction and partial removal of the double strand not bound to the microchannel (each like above) leads to the single strand which is bound again via biotin and is extended again. On whose The 3 'end is now the last 22 nucleic acids at the 3' end of GFP-4b complementary sequence. Hybridization of 3'-protected GFP-4b and subsequent Polymerase reaction and partial removal of the not directly bound to the microchannel which strands leads to the renewed single strand bound via biotin, the this time at the 3 'end of the last 26 nucleic acids at the 3' end of GFP-5b contains mental nucleic acids. Finally, 3'-protected GFP-5b is added to this strand hybridized in the overlap area. Another polymerase reaction leads to over biotin bound non-coding strand section of the GFP-DNA which contains the nucleic acids 358 to 717 of the coding strand is complementary, with a 5'-terminally elongated ene de of 50 nucleic acids encompassing the restriction site. This non-coding Section with the 5'-terminal extension and biotin at the 5'-end is now on the upper area of the microchannel via the biotin / avidin interaction.  

In parallel, the complementary coding section of the GFP- DNA from nucleic acid 1 to 401 (counted from its 3 'end) from the corresponding Oligonuc Leotide modules produced, which was bundled on a separate microfluidic element that is. This polynucleotide portion of the coding chain, which over its 5'-biotinyl end is bound to avidin on the surface of a microchannel, then by biotin over shot detached, isolated and in the microchannel obtained above with the non-coding Ab transferred and in the overlap area (358 to 401 of the coding strand) hybridizes this non-coding polynucleotide section. Polymerase reaction leads to complete non-coding strand with 3'-terminal and 5'-terminal one Re section and 5'-biotinyl residue. Code removal the strands (by endonuclease as described above or by dehybridization and washing) gives the desired single strand of DNA, which is still above the 5'-biotin is bound. This strand is released by treatment with an excess of biotin and cleaned. It can then be amplified or used as desired. Zwischenkontrol correct sequences of the intermediates are obtained by amplification (polymerase ket ten reaction) with suitable primers to the corresponding DNA sections (ex by sequencing).

Claims (11)

1. A method for preparative synthesis of polynucleotides using microfluidic systems, comprising

• a) the use of a starter molecule oligonucleotide, which is bound to a surface of a reaction space of a microfluidic system directly or via a spacer, or is directly synthesized with the binding of one end to the surface of the reaction space, and which is one of the ends of a strand of the includes synthesizing polynucleotide sequence;
• b) one or more subsequent steps of hybridization of one or more sequencers, which sequence segments correspond to the sequence of a strand to be synthesized or their complementary sequence and whose terminal sequences overlap in such a way that they hybridize with a terminal segment from the starter molecule or the respective preceding sequemer and the subsequent sequemer or, in the case of the result of a terminal blunt end, for example after polymerase reaction steps, after the manufacture of a sticky end, the hybridization thereof with the overlapping region of a subsequent sequencer, and
• c) in the case of nicks between sequencers lined up in this way or for the completion of missing sections from 1 or more nucleotides which are complementary to a section of the respective opposite section of the sequencer, polymerase reactions and, if necessary, subsequent ligation to remove the resulting nicks, the steps (ii) and (iii) repeated until the desired polynucleotide is obtained.

2. The method according to claim 1, characterized in that the microfluidic element is a micro-chamber or a micro-channel and the binding of the starter molecule via whose 5'- or in particular 3'-hydroxy takes place. 3. The method according to any one of the preceding claims 1 and 2, further comprising the Release of one of the two strands, especially the one not to the surface of the Microfluidic elements bound, or the double strand of the available Polynucleotide.   4. The method according to any one of claims 1 to 3, wherein two or more methods, selected from combinatorial, sequence-individual, recursive and non-recurs sive processes can be combined. 5. The method according to claim 4, wherein the production by means of combinatorial syn thesis for the preparation of longer precursor molecules, for example from 20 to 300, in particular from 50 to 200 nucleotides in length, with subsequent use of the precursor molecules thus obtainable in a recursive process using polymerase reaction and ligation to longer polynucleotides. 6. The method according to any one of the preceding claims 1 to 3, pronounced as a combinatorial polynucleotide synthesis method, wherein in a first step all possible sequence combinations for the synthesis are prepared, and which further comprises

• a) the use of a starter molecule oligonucleotide from the collection of molecules with the combinatorial sequence combinations, which is bound to a surface of a reaction space of a microfluidic system, preferably a microfluidic element, in particular a microchannel, directly or via a spacer, preferably via its 5 ' - or in particular its 3'-hydroxy, and which comprises one of the ends, preferably the 3 'end, of a strand of the polynucleotide sequence to be synthesized;
• b) then recursive steps of hybridization of one or more sequencers from the collection of molecules with the combinatorially produced sequence combinations, which sequence sections correspond to the sequence of a strand to be synthesized or its complementary sequence and whose terminal sequences overlap in such a way that they overlap the hybridization with enable a terminal section from the starter molecule or the respective preceding sequemer and the subsequent sequemer, and
• c) in the case of nicks between sequencers lined up in this way, after at least two sequemers have been hybridized in the first step (ii) or in subsequent steps a number of partially overlapping sequences or preferably one partially overlapping sequence have been hybridized to the growing polynucleotide to be produced, ligation (or , if necessary, to complete missing sections from 1 or more nucleotides which are complementary to a section of the opposite sequencer section, polymerase reactions and then ligation, for example in the non-hybridizing part of the starter molecule which is oriented towards the binding surface) in order to remove resulting nicks,
  repeating steps (ii) and (iii) as many times as necessary until the sequence of the desired polynucleotide is completed;
• d) and that, finally, the finished molecule is preferably released by dehybridization or chemical cleavage from the surface or the spacer and, if desired, cleaned.

7. The method according to any one of claims 1 to 3, pronounced as a sequence-individualized synthesis, comprising

• a) the use of a starter molecule oligonucleotide with a specific sequence, which is bound to a surface of a reaction space of a microfluidic system, preferably a microfluidic element, in particular a microchannel, directly or via a spacer, preferably via its 5 'or, in particular, its 3' Hydroxy, and which comprises one of the ends, preferably the 3 'end, of a strand of the polynucleotide sequence to be synthesized;
• b) subsequent sequential or simultaneous hybridization of in each case one or more sequencers, which sequence sections correspond to the sequence of a strand to be synthesized or its complementary sequence and whose end sequences each overlap in such a way that they hybridize with a terminal section from the starter molecule or the the preceding and subsequent sequemans are possible, and which together cover the two complete double strands of the polynucleotide to be synthesized, so that only nicks are formed, or, if the ends of sequeners (including the starter) do not meet, also sections with unpaired nucleotides; and
• c) in the case of nicks between sequencers lined up in this way or to complete missing sections from 1 or more nucleotides which are complementary to a section of the opposite sequencer section, carries out polymerase reactions and then ligation to eliminate the resulting render nicks;
  wherein steps (ii) and (iii), if not all of the sequeners are hybridized at once and then completed by ligase or ligase and polymerase reaction, are repeated as often as necessary until the sequence of the desired polynucleotide is completed;
• d) and preferably finally releases the finished molecule by dehybridization or chemical cleavage from the surface or the spacer and, if desired, cleans it.

8. The method according to any one of claims 1 to 3, characterized as a recursive method with relatively longer precursor molecules, comprising

• a) the use of a starter molecule oligonucleotide which is bound to a surface of a reaction space of a microfluidic system, preferably a microfluidic element, in particular a microchannel, directly or via a spacer, preferably via its 5'- or in particular its 3'-hydroxy, or is synthesized directly by binding one end to the surface of the reaction space and which comprises one of the ends, preferably the 3 'end, of a strand of the polynucleotide sequence to be synthesized;
• b) a subsequent step of hybridizing a sequencer, which corresponds to a sequence section from the sequence of a strand of the poly nucleotide to be synthesized and whose terminal sequences in the case of the first sequencer in each case hybridize with a terminal section from the starter molecule or in the case of sequentially used sequentially further seque mers of the same strand because of the blunt ends arising after the polymerase reaction steps mentioned below and after the production of sticky ends enable its hybridization with the overlapping region of a subsequent seque mers, and production of a sticky ends (in particular with a resultant free 3'-OH in the available single strand section); and
• c) to complete missing sections of 1 or more nucleotides that are complementary to a section of the opposite sequence section, each polymerase reaction after each hybridization of a sequencer, and, if the covalent connection of the sequeners is desired, polymerase reactions to fill in any resulting Gaps and ligation or just ligation;
  wherein steps (ii) and (iii) are repeated as often as necessary until the sequence of the desired polynucleotide is completed, and the covalent binding of the sequeners by ligation or polymerase reaction and ligation also take place only after the addition of more than one or all of the sequeners can;
• d) and preferably finally release and, if desired, cleaning of the finished molecule by dehybridization or chemical cleavage from the surface or the spacer.

9. The method according to any one of claims 1 to 5, further comprising the removal of the available non-covalently bound strand by dehybridization, and the Use of the resulting covalently bound single strand for polyme Race reaction to generate the complementary, non-covalently bound again NEN strand by means of polymerase reaction in the microfluidic element and its release tion by dehybridization, the release and synthesis again as often as desired can be fetched to produce the free single strand in any number of copies. 10. Microfluidic elements with according to one of the available in claims 1 to 7 those polynucleotide or precursors thereof. 11. Kits or apparatus for the production of polynucleotides according to one of the Claims 1 to 7, the correspondingly loaded microfluidic elements and gege optionally include other components.