Removal of dye-labeled dideoxy terminators from DNA sequencing reactions

[0002] This application claims benefit of U.S. Provisional Application Nos. 60/158,188, filed Oct. 6, 1999 and 60/164,050, filed Nov. 8, 1999. Both of these applications are incorporated herein by reference for all purposes.

[0003] 1. Field of the Invention

[0004] This invention pertains to the field of the purification of polynucleotides from mixtures that contain fluorescent dye-labeled molecules. The polynucleotides can include, for example, reaction products from nucleic acid sequencing reactions.

[0005] 2. Background

[0006] The demands of the Human Genome Project and the commercial implications of polymorphism and gene discovery have driven the development of significant improvements in DNA sequencing technology. Contemporary approaches to DNA sequencing have imposed stringent demands on reliability and throughput for DNA sequencers. Recent reports have demonstrated the extraordinary potential of capillary electrophoresis (CE) for DNA sequencing given the inherent speed, resolving power and ease of automation associated with this method as compared to slab gel electrophoretic methods (Carrilho et al., Anal. Chem.1996, 68, 3305-3313; Tan and Yeung, Anal. Chem.1997, 69, 664-674; Swerdlow et al., Anal. Chem.1997, 69, 848-855).

[0007] Relative to cross-linked gel capillary electrophoretic columns, the recent development of replaceable polymer solutions to achieve size separation of single-stranded DNA fragments has increased the lifetime of the columns and eliminated the requirements of gel pouring and casting (Ruiz-Martinez et al., Anal. Chem.1993, 65, 2851-2858). Additionally, improvements in the composition of the separation matrix have led to sequencing over 1000 bases per run (Carrilho et al., Anal. Chem.1996, 68, 3305-3313). Automated capillary electrophoresis systems for DNA sequencing have been introduced commercially by three major scientific instrument manufacturers (Beckman Coulter CEQ™ 2000 DNA Analysis System; Amersham Pharmacia MegaBACE 1000 DNA Sequencing System; and PE Biosystems ABI Prism 3700 DNA Analyzer).

[0008] Realizing the potential of this new generation of automated DNA sequencers is proving difficult, however, as problems in read length and accuracy remain, primarily due to the limitations associated with the methods currently available for purifying the products of cycle sequencing reactions. Indeed, the critical importance of sample preparation for the successful implementation of capillary electrophoresis has not been sufficiently emphasized.

[0009] In contrast to slab gel electrophoresis, primer extension products are introduced into the capillary column using electrokinetic injection, which provides focusing of the single-stranded DNA fragments at the head of the column (Swerdlow et al., Proc. Natl. Acad. Sci. U.S.A.1988, 85, 9660-966). However, electrokinetic injection is biased toward high electrophoretic mobility ions, such as chloride, deoxynucleotides and dideoxynucleotides, which, if present in the sequencing reaction solution, negatively affect the focusing of single-stranded DNA fragments. Consequently, to increase the amount of DNA injected into the capillary column, and to improve the focusing of the injected DNA, an effective removal of these small ionic species is required.

[0010] A further group of especially problematic high electrophoretic mobility ions are the dye-labeled fluorescent dideoxynucleotide terminators, and in particular the recently commercialized terminators having two fluorescent moieties configured as energy transfer pairs (ABI PRISM BigDye™ Terminators from PE Biosystems and DYEnamic ET™ Terminators from Amersham Pharmacia). These reagents, and the hydrolysis products derived therefrom, have been found to be particularly troublesome with respect to quantitative removal from primer extension reactions, resulting in the presence of fluorescent artifacts (routinely described as “dye blobs”) that negatively impact the automated analysis of sequencing data (Rosenblum, B.; Lee, L.; Spurgeon, S.; Khan, S.; Menchen, S.; Heiner, C. and Chen, S., Nucleic Acids Research,1997, 25, 4500-4504). Dye-labeled sequencing primers, when employed as an alternative to dye-labeled terminators, afford similar problems (Jingyue, J.; Ruan, C.; Fuller, C.; Glazer, A. and Mathies, R., Proc. Natl. Acad. Sci. USA,1995, 92, 4347-4351. Jingyue, J.; Kheterpal, I.; Scherer, J.; Ruan, C.; Fuller, C.; Glazer, A. and Mathies, R., Anal. Biochem.,1995, 231, 131-140. Jingyue, J.; Glazer, A. and Mathies, R., Nucleic Acids Research,1996, 24, 1144-1148. Lee, L.; Spurgeon, S.; Heiner, C.; Benson, S.; Rosenblum, B.; Menchen, S.; Graham, R.; Constantinescu, A.; Upadhya, K. and Cassel, J., Nucleic Acids Research,1997, 25, 2816-2822).

[0011] Fluorescent energy transfer dye-labeled dideoxynucleotide triphosphate terminators suitable for use in DNA sequencing are described in U.S. Pat. No. 5,800,996. Fluorescent energy transfer dye-labeled primers suitable for use in DNA sequencing are described in U.S. Pat. Nos. 5,688,648, 5,707,804 and 5,728,528.

[0012] During the course of cycle-sequencing reactions, deoxynucleotide triphosphates (dNTPs) and dye-labeled dideoxynucleotide triphosphates (ddNTPs) undergo hydrolysis of the phosphate ester bonds during the denaturation step that proceeds each amplification cycle when the temperature is elevated to from 95° C. to 99° C. This results in the generation of dye-labeled artifacts including dideoxynucleotide diphosphates (ddNDPs), dideoxynucleotide monophosphates (ddNMPs) and dideoxynucleosides. The ddNTPs derived from the pyrimidine bases, dideoxythymidine (ddTTP) and dideoxycytidine (ddCTP), are particularly labile in this regard. The dye-labeled ddNTPs, ddNDPs and ddNMPs elute from capillary electrophoretic columns prior to the dye-labeled primer extension products, and consequently do not directly interfere with the interpretation of the sequencing data. However, the intensity of the signals associated with the dye-labeled ddNTPs, ddNDPs and ddNMPs may exceed those of the primer extension products by many orders of magnitude. This discontinuity in signal intensity has proven problematic with respect to automated base-calling software, resulting, in a worse-case scenario, in the software interpreting the fluorescent artifact peaks as primer extension products and the primer extension products as baseline noise. A further significant complication is associated with the presence of dye-labeled dideoxynucleosides, in that they are found to co-elute from both capillary electrophoretic columns and slab gels with the primer extension products. Not only is the intensity of the artifact signal disproportionately high as compared to the signals associated with the primer extension products, but the width of the peak is sufficiently broad to obscure the analysis of from 5 to 20 bases.

[0013] The sample preparation scheme now routinely employed for both slab gel electrophoresis and CE consists of desalting DNA sequencing samples by ethanol or isopropanol precipitation, followed by reconstitution of the DNA fragments and template in a mixture of formamide-0.5 M EDTA (49: 1) prior to loading or injection (Figeys et al., 1996, 744, 325-331; Sambrook, J.; Fritsch, E. F.; Maniatis, T. Molecular Cloning: A Laboratory Manual;Cold Spring Harbor Laboratory Press: Cold Spring Harbor, N.Y., 1989; section 9.49). Although widely utilized, this method has been found to exhibit variable reproducibility in terms of DNA recovery, to provide marginal performance with respect to the quantitative removal of dye-labeled artifacts, and is not easily automated (Tan, H.; Yeung, E. S. Anal. Chem.1997, 69, 664-674, and Hilderman, D.; Muller, D. Biotechniques1997, 22, 878-879).

[0014] High electrophoretic mobility ionic species in DNA sequencing samples are not the only contaminants that cause degradation in sequencing read length. Template DNA also has been shown to interfere with the analysis of primer extension products in both thin slab gels (Tong et al., Biotechniques1994, 16, 684-693), and capillary columns (Swerdlow et al., Electrophoresis1996, 17, 475-483). Upon injection of the sequencing reaction solution, a current drop and significant deterioration in the resolving power of the capillary column is observed when template DNA is present in the sample (Salas-Solano et al., Anal. Chem.1998, 70, 1528-1535). However, at present, template DNA removal is seldom considered an essential aspect of sample preparation for DNA sequencing by capillary electrophoresis.

[0015] Only two approaches to sample preparation that address the need for template removal have been proposed thus far. In the first approach, which is described in U.S. Pat. No. 5,484,701, a biotinylated primer enables the capture and purification of primer extension products on streptavidin magnetic particles. After extensive washing of the primer extension products immobilized on the streptavidin magnetic particles to remove the sequencing reaction constituents including template DNA, release of the primer extension products is effected by heating the streptavidin magnetic particles to from about 90° C. to 100° C. in a formamide solution.

[0016] Although this approach has considerable utility in conjunction with slab gel electrophoresis (in which formamide is often added to sequencing samples to facilitate denaturation of duplex DNA and to increase the viscosity of the sample to facilitate slab gel loading), it has recently been shown to be problematic when utilized in conjunction with capillary electrophoresis. At least three distinct problems (exclusive of cost) have been identified as being associated with this approach. First, the formamide solution utilized to effect release of immobilized primer extension products is incompatible with electrokinetic injection, owing to the high ionic strength of the solution due to the presence of high electrophoretic mobility ions (most notably 10 mM EDTA or 30-140 mM sodium acetate in 95% formamide). In the absence of salt in the formamide solution, the efficiency of release of biotinylated primer extension products has been shown to be significantly reduced from >95% to <40% (Tong and Smith, Anal. Chem.1992, 64, 2672-2677). The effective ionic strength of the release solution has been shown to be still further increased by decomposition of 95% formamide that occurs when the solution is heated and results in release of ammonia. Second, samples recovered from streptavidin magnetic particles are found to be contaminated with protein derived from streptavidin. Release of immobilized primer extension product results from the denaturation of the streptavidin that is covalently linked to the magnetic particle. Streptavidin is a multi-subunit protein with a high isoelectric point. Denaturation of immobilized streptavidin is always accompanied by the concomitant release of those protein subunits that are not covalently linked to the magnetic particles. This contaminating protein acts in a manner somewhat analogous to template DNA, as a consequence of its anionic character and high molecular weight. Finally, dye-labeled fluorescent dideoxynucleotide terminators, and hydrolysis products derived therefrom, have been found to bind nonspecifically to streptavidin magnetic particles, and to be released into the formamide solution upon denaturation of streptavidin. Thus, the nonspecifically bound terminators can accompany the “purified” primer extension product and adversely impacting the subsequent analysis.

[0017] The second approach to purification of primer extension product that includes template DNA removal utilizes a multi-step methodology involving: (1) Ultrafiltration to remove template DNA; (2) Vacuum concentration to reduce sample volume; (3) Size exclusion chromatography (two sequential gel filtration columns) to reduce the ionic strength and remove dye-labeled artifacts; and (4) Vacuum concentration to reduce sample volume prior to analysis (Ruiz-Martinez et al., Anal. Chem.1998, 70, 1516-1527, and Salas-Solano et al., Anal. Chem.1998, 70, 1528-1535). Although this approach affords excellent samples for CE analysis, it is generally complex, costly, time consuming and unsuitable for automation in a high throughput environment. In fact, as compared to the throughput potential of multi-column capillary electrophoresis DNA sequencers, the aforementioned methodology would constitute the rate-limiting step in a sequencing laboratory.

[0018] The methods currently employed to purify primer extension products for analysis by capillary and slab gel electrophoresis have been summarized above. Although ethanol or isopropanol precipitation are known to reduce the level of both high electrophoretic mobility ions and dye-labeled artifacts present in the sample, the residual level of dye-labeled artifacts remains problematic. Size exclusion chromatography in spin columns can remove both high electrophoretic mobility ions and dye-labeled artifacts, but is the least automation compatible for high throughput environments and the most expensive. Additionally, the use of spin columns results in considerable sample dilution, often introducing the necessity for a precipitation or evaporation step to concentrate the sample prior to analysis. The biotin/streptavidin methodology should, in principle, quantitatively remove dye-labeled artifacts. However, dye-labeled artifacts bind nonspecifically to immobilized streptavidin and are released along with the primer extension products when the sample is heated in formamide. Another disadvantage of the biotin/streptavidin methods is that short molecules are preferentially adsorbed to the matrix, resulting in a decrease in signal for longer molecules.

[0019] Thus, none of the methods currently available provide for the quantitative removal of all of the potentially contaminating constituents associated with DNA sequencing reactions. Consequently, a method is needed to circumvent this considerable limitation if the extraordinary potential of capillary electrophoresis for DNA sequencing is to be realized in the not too distant future. The present invention fulfills this and other needs.

[0020] The present invention provides methods for removing fluorescent or other dye-labeled molecules from a mixture that includes one or more polymers, such as polynucleotides, and the dye-labeled molecules. The methods involve contacting the mixture with a plurality of particles that are composed of a cross-linked three-dimensional hydrophilic polymeric matrix that has trapped within it a porous adsorptive hydrophobic material. The fluorescent dye-labeled molecules pass through the hydrophilic matrix and become preferentially adsorbed onto the porous hydrophobic materials. Subsequent removal of the particles from the mixture thus also removes the adsorbed fluorescent dye-labeled molecules.

[0021] In some embodiments, the invention provides methods for the purification of fluorescent primer extension products derived from polymerase chain reactions (PCR) and cycle sequencing reactions (CSR). The purified products are substantially free of dye-labeled primers and terminators, including dye-labeled energy transfer primers and terminators (ddNTPs), as well as their hydrolysis products, which include dye-labeled ddNDPs, ddNMPs and dideoxynucleosides. The methods afford samples that are free of fluorescent artifacts and thus are suitable for DNA sequencing by either capillary or slab gel electrophoresis.

[0022] The methods of the invention involve, in some embodiments, extending a primer by means of a template-directed primer extension reaction in the presence of either dye-labeled primers or dye-labeled terminators;

[0023] contacting the constituents of the primer extension reaction with a plurality of particles that are composed of a cross-linked three-dimensional hydrophilic polymeric matrix in which porous adsorptive hydrophobic materials are entrapped, so as to effect the preferential absorption to the particles of the dye-labeled primers or dye-labeled terminators, as well as dye-labeled artifacts derived therefrom; and

[0024] separating physically the particles from the liquid phase that contains the constituents of the primer extension reaction.

[0025] Some embodiments of the methods further involve purifying the liquid phase that contains the remaining constituents of the primer extension reaction, if necessary, and analyzing the liquid phase that contains the remaining constituents of the primer extension reaction by capillary or slab gel electrophoresis.

[0026] The invention also provides methods for preparing dye-labeled polynucleotides that are substantially free of unincorporated dye-labeled reactants. These methods involve:

[0027] (a) annealing a primer to a template and contacting the annealed primer with a polymerase in a reaction mixture that comprises the dye-labeled reactant, thereby extending the primer to form a plurality of dye-labeled polynucleotides;

[0028] (b) contacting the reaction mixture with a plurality of particles that are composed of porous hydrophobic materials that are retained within a hydrophilic polymeric matrix, so as to effect the preferential absorption of the dye-labeled reactant, and dye-labeled artifacts derived therefrom; and

[0029] (c) separating the particles of (b) from the reaction mixture that contains the dye-labeled polynucleotides.

[0030] Other aspects of the invention include, for example, the use of blocking reagents to precondition the magnetic particles so as to minimize the nonspecific binding of primer extension products.

[0031] The file of this patent contains at least one drawing executed in color. Copies of this patent with color drawing(s) will be provided by the Patent and Trademark Office upon request and payment of the necessary fee.

[0032] FIG. 1 summarizes the cycle sequencing methodology from which the invention can be used to purify the primer extension products. A sequencing ladder is generated by repetition of several cycles in which a primer is first annealed to template DNA that provides a hybrid suitable for subsequent extension of the primer by the action of a thermal stable DNA polymerase in the presence of deoxynucleotide triphosphates. Each of the primer extension products is eventually terminated by incorporation of a dye-labeled dideoxynucleotide triphosphate terminator.

[0033] FIG. 2 illustrates the cycle sequencing methodology while emphasizing that a fluorescent dye-labeled dideoxynucleotide triphosphate terminator can be substituted for an unlabeled terminator, thereby generating a sequencing ladder suitable for detection in an automated DNA sequencer having fluorescence detection capabilities.

[0034] FIG. 3 illustrates the chemical structures of the dye-labeled BigDye™ Terminators (PE Applied Biosystems, Foster City Calif.) derived from the pyrimidine bases thymine and cytosine. For dideoxythymidine (ddT), R is OH; for dideoxythymidine monophosphate (ddTMP), R is HPO₄⁻, for dideoxythymidine diphosphate (ddTDP), R is HP₂O₇²⁻; and for dideoxythymidine triphosphate (ddTTP), R is HP₃O₁₀³⁻. For dideoxycytidine (ddC), R is OH; for dideoxycytidine monophosphate (ddCMP), R is HPO₄⁻, for dideoxycytidine diphosphate (ddCDP), R is HP₂O₇²⁻; and for dideoxycytidine triphosphate (ddCTP), R is HP₃O₁₀³⁻.

[0035] FIG. 4 illustrates the chemical structures of the dye-labeled BigDye™ Terminators (PE Applied Biosystems, Foster City Calif.) derived from the purine bases adenine and guanine. For dideoxyadenosine (ddA), R is OH; for dideoxyadenosine monophosphate (ddAMP), R is HPO₄⁻, for dideoxyadenosine diphosphate (ddADP), R is HP₂O₇²⁻; and for dideoxyadenosine triphosphate (ddATP), R is HP₃O₁₀³⁻. For dideoxyguanosine (ddG), R is OH; for dideoxyguanosine monophosphate (ddGMP), R is HPO₄⁻, for dideoxyguanosine diphosphate (ddGDP), R is HP₂O₇²⁻; and for dideoxyguanosine triphosphate (ddGTP), R is HP₃O₁₀³⁻.

[0036] FIG. 5 illustrates the raw data recorded on an Perkin Elmer ABI PRISM™ 310 Genetic Analyzer. The primer extension products were purified by isopropanol precipitation as recommended by the instrument manufacturer. Note the presence of dye-labeled artifacts in the front on the capillary electrophoretic separation (ddNTPs, ddNDPs and ddNMPs) as well as the ddT artifact near the center of the electrophoretic separation. Primer extension products terminated with ddA are depicted in green; primer extension products terminated in ddT are depicted in red; primer extension products terminated in ddC are depicted in blue; and primer extension products terminated with ddG are depicted in black.

[0037] FIG. 6 shows the processed DNA sequencing data that corresponds to the raw data shown in FIG. 5. Note that the region between 200 and 214 bases has been assigned by the software the sequence GCTTT TTTTT TTTNA (SEQ ID NO:1), despite the fact that visual inspection of the processed data clearly indicates that the sequence is probably GCTTT CCAGT CGGAA (SEQ ID NO:2). This anomaly results from the presence of the dye-labeled ddT artifact in the capillary electrophoretic separation. Primer extension products terminated with ddA are depicted in green; primer extension products terminated in ddT are depicted in red; primer extension products terminated in ddC are depicted in blue; and primer extension products terminated with ddG are depicted in black. The software assigns an “N” to primer extension products that cannot be identified.

[0038] FIG. 7 illustrates the raw data recorded on an Perkin Elmer ABI PRISM™ 310 Genetic Analyzer. The primer extension products were purified by the method described herein by adding 20 μL of product to the magnetic particles pelleted from 100 μL of a 10% (v/v) suspension. The contents of each well were mixed by drawing the suspension into a pipette tip and then displacing the suspension 10 times. To each of the supernatants removed from the magnetic particles was added 20 μL of formamide containing 10% (w/v) sorbitol. The resulting solution was mixed briefly by vortexing and then analyzed. Note that after purification by the method described herein, only the ddTTP and ddTCP artifacts are recorded at magnitudes exceeding those of the primer extension products. Primer extension products terminated with ddA are depicted in green; primer extension products terminated in ddT are depicted in red; primer extension products terminated in ddC are depicted in blue; and primer extension products terminated with ddG are depicted in black.

[0039] FIG. 8 shows the processed DNA sequencing data that corresponds to the raw data shown in FIG. 7. Although some residual ddT artifact can been seen by careful inspection of the baseline in the region between 190 and 200 bases, the magnitude of the artifact signal is some low relative to the signals associated with the primer extension products that the analysis software was unimpeded. Primer extension products terminated with ddA are depicted in green; primer extension products terminated in ddT are depicted in red; primer extension products terminated in ddC are depicted in blue; and primer extension products terminated with ddG are depicted in black. The software assigns an “N” to primer extension products that can not be identified.

[0040] FIG. 9 illustrates the raw data recorded on an Perkin Elmer ABI PRISM™ 310 Genetic Analyzer. The primer extension products were generated from a cycle sequencing reaction comprised of 2 μL of pUC19 template DNA (125 ng/μL), 4 μL of primer (M13/pUC primer-48 24mer), 6 μL of HPLC grade water, and 8 μL of Terminator Ready Reaction Mix (PE Applied Biosystems) diluted 1:1 with halfBD™ Dye Terminator Sequencing Reagent (Genpak, Ltd., Stony Brook N.Y.). The primer extension products (20 μL) were added to the magnetic particles retained from 100 μL of a 10% (w/v) suspension in each well of a multiwell plate. The contents of each well are mixed by drawing the suspension into a pipette tip and then displacing the suspension 10 times. The multiwell plate was placed on a magnet to pellet the particles and the supernatants transferred to sequencer sample tubes. To each of the tubes was added 20 μL of formamide containing 10% (w/v) sorbitol. The samples are mixed by vortexing briefly and analyzed. Note that after purification by the method described above, none of the dye-labeled artifacts are recorded at magnitudes exceeding those of the primer extension products. Primer extension products terminated with ddA are depicted in green; primer extension products terminated in ddT are depicted in red; primer extension products terminated in ddC are depicted in blue; and primer extension products terminated with ddG are depicted in black.

[0041] FIG. 10 shows the processed DNA sequencing data that corresponds to the raw data shown in FIG. 9. Careful inspection of the baseline reveals that dye-labeled ddT artifact cannot be detected. Primer extension products terminated with ddA are depicted in green; primer extension products terminated in ddT are depicted in red; primer extension products terminated in ddC are depicted in blue; and primer extension products terminated with ddG are depicted in black. The software assigns an “N” to primer extension products that can not be identified.

[0042] FIG. 11 shows a time course of the removal of fluorescent dNTPs by MagaCharc™ beads. The relative amount of fluorescence was determined over time between zero and 120 seconds.

[0043] FIG. 12 shows a response curve for increasing amounts of MagaCharc™ beads added to a mixture of fluorescent dNTPs. The beads were incubated with the dNTPs for one minute and removed. Relative fluorescent units remaining were then determined.

[0044] The present invention provides methods and kits for removing unincorporated fluorescent dye-labeled molecules from a mixture that includes a plurality of fluorescent dye-labeled polynucleotides and the unincorporated fluorescent dye-labeled molecules. The methods of the invention provide an efficient and effective means by which one can remove unincorporated fluorescent dyes from the products of, for example, DNA sequencing reactions. The removal of the dyes from the reaction products greatly reduces or eliminates artifacts that can result from the presence of unincorporated fluorescent dyes in samples that are loaded onto gels or capillary electrophoresis columns, for example.

[0045] Moreover, the methods are simple to perform and highly efficient, and are amenable to high-throughput, automated sequencing. Purification of dye-labeled polynucleotides using the methods of the invention is particularly well suited to automation. One need not carry out a centrifugation step or a vacuum filtration step, both of which are difficult to adapt to an automated robotic sequencing system. Another significant advantage of the methods and kits of the invention is that the use of modified (e.g., biotinylated) primers is not required.

[0046] The methods involve contacting a mixture that contains a plurality of fluorescently labeled polynucleotides, as well as fluorescent dye-labeled molecules that are not incorporated into the polynucleotide, with a plurality of particles that are composed of a cross-linked hydrophilic polymeric matrix in which are entrapped crystalline or particulate porous hydrophobic materials to which the fluorescent dye-labeled molecules can adsorb. The unincorporated fluorescent dye-labeled molecules pass rapidly through the hydrophilic matrix and become adsorbed onto the hydrophobic materials, while the fluorescently-labeled molecules that are incorporated into polynucleotides pass through the hydrophilic matrix much more slowly due to their molecular size. The particles are then removed from the mixture, thus also removing the adsorbed unincorporated fluorescent dye-labeled molecules.

[0047] The methods of the present invention are useful for the purification of primer extension products that are free of dye-labeled artifacts that result from the presence of dye-labeled ddNTPs and the hydrolysis products derived therefrom. The products are obtained in a form that is optimal for automated DNA sequencing by slab gel or particularly capillary electrophoresis, and for other analytical methods.

[0048] Primer Extension Reactions

[0049] The purification methods of the invention are useful for purifying a wide variety of products that are obtained by polymerase-mediated, template-directed extension of oligonucleotide primers. These reactions are often used in the characterization of nucleic acids, including DNA and RNA. The purification methods can be used, for example, to purify the products of polymerase chain reaction, ligase chain reaction, and other amplification methods that employ primer extension and/or ligation. These and other protocols that involve primer extension are known to those of skill in the art. Examples of these techniques are found in Berger and Kimmel, Guide to Molecular Cloning Techniques, Methods in Enzymology152 Academic Press, Inc., San Diego, Calif. (Berger); Sambrook et al. (1989) Molecular Cloning—A Laboratory Manual(2nd ed.) Vol. 1-3, Cold Spring Harbor Laboratory, Cold Spring Harbor Press, N.Y., (Sambrook et al.); Current Protocols in Molecular Biology,F. M. Ausubel et al., eds., Current Protocols, a joint venture between Greene Publishing Associates, Inc. and John Wiley & Sons, Inc., (1994 Supplement) (Ausubel); Cashion et al., U.S. Pat. No. 5,017,478; and Carr, European Patent No. 0,246,864. Examples of techniques sufficient to direct persons of skill through in vitro amplification methods are found in Berger, Sambrook, and Ausubel, as well as Mullis et al., (1987) U.S. Pat. No. 4,683,202; PCR Protocols A Guide to Methods and Applications (Innis et al. eds) Academic Press Inc. San Diego, Calif. (1990) (Innis); Arnheim & Levinson (Oct. 1, 1990) C&EN 36-47; The Journal Of NIH Research(1991) 3: 81-94; (Kwoh et al. (1989) Proc. Nat'l. Acad. Sci. USA86: 1173; Guatelli et al. (1990) Proc. Nat'l. Acad. Sci. USA87, 1874; Lomell et al. (1989) J. Clin. Chem.,35: 1826; Landegren et al., (1988) Science,241: 1077-1080; Van Brunt (1990) Biotechnology,8: 291-294; Wu and Wallace, (1989) Gene,4: 560; and Barringer et al. (1990) Gene,89:117.

[0050] The purification methods of the invention are particularly useful where a very clean primer extension product preparation is required. DNA sequencing, in particular where capillary electrophoresis is used, provides an illustrative example of an analytical method for which the methods of the invention can solve major drawbacks that have prevented capillary electrophoresis-mediated DNA sequencing from reaching its full potential.

[0051] In these methods, a cycle sequencing reaction is carried out as summarized in FIG. 1. In a typical embodiment, a primer is allowed to hybridize to the template DNA at a suitable annealing temperature, which is typically between about 50° C. and about 55° C., in preparation for primer extension. The polymerase enzyme, deoxynucleotide triphosphates (dNTPs), dideoxynucleotide terminators (ddNTPs) and other necessary reactants are added to the annealed template-primer complex. The temperature is then raised to an formed between the extended primers and the template DNA are denatured, e.g., by heating to a temperature of from about 95° C. to about 99° C., or other suitable method, thereby effecting release of the terminated primer extension products and liberating the template DNA prior to initiating a second cycle of primer extension. Routinely, this cycle is repeated from about 10 to 50 times.

[0052] In presently preferred embodiments, the primer extension products produced in the aforementioned cycle contain dye-labeled dideoxynucleotide terminators, as illustrated in FIG. 2. The chemical structures of the dye-labeled dideoxynucleotide terminators contained within the ABI PRISM® BigDye™ Terminator Cycle Sequencing Ready Reaction Kit are appropriate temperature for the particular polymerase, which is generally between about 60° C. and about 70° C. for a thermostable polymerase or between about room temperature and about 37° C. for a non-thermostable polymerase, to facilitate template-directed primer extension. Finally, the hybrids illustrated in FIGS. 3 and 4. The dideoxynucleotide triphosphates derived from the pyrimidine bases (ddCTP and ddTTP) are illustrated in FIG. 3, and the dideoxynucleotide triphosphates derived from the purine bases (ddATP and ddGTP) are illustrated in FIG. 4. These reagents represent the state of the art and are each comprised of two dyes configured as energy transfer pairs. In each of the reagents the fluorescein moiety acts as an energy transfer donor and the rhodamine moiety as a energy transfer acceptor. The efficiency of energy transfer approaches 100%. The brightness of these reagents is 2- to 3-fold greater than the previous generation of dideoxynucleotide terminators that were derived from the dichlororhodamines. However, the BigDye™ terminators are particularly hydrophobic and exhibit limited solubility at neutral and low pH. Consequently, BigDye™ terminators, and the hydrolysis products derived therefrom, have proven particularly difficult to remove from primer extension products. Hydrolysis of BigDye™ terminators, which results in progressive dephosphorylation of the nucleotide triphosphate, and is thought to occur during the denaturation stage of the cycle sequencing reaction when the temperature to elevated to from 95° C. to 99° C., produces the corresponding ddNDPs, ddNMPs and dideoxynucleosides. The ddNTPs initially present in the reaction are less hydrophobic than the ddNDPs which result from their hydrolysis, which are less hydrophobic than the ddNMPs which result from their hydrolysis, which are less hydrophobic than the dideoxynucleosides which result from their hydrolysis. The ddNTPs derived from the pyrimidine bases are more susceptible to hydrolytic dephosphorylation than the ddNTPs derived from the purine bases. The mechanistic basis for this differentiation has not been explained.

[0053] Purification of Primer Extension Products

[0054] The purification of primer extension products free of dye-labeled dideoxynucleotide triphosphates (ddNTPs), and hydrolysis products derived therefrom (ddNDPs, ddNMPs and dideoxynucleosides), can be achieved by contacting the primer extension reaction constituents, after thermal cycling, for a precise period of from about 10 seconds to 5 minutes, more preferably between about 10 seconds and about 2 minutes, with particles that are composed of a cross-linked hydrophilic polymeric matrix. Entrapped within the cross-linked hydrophilic matrix are porous hydrophobic materials.

[0055] A. Particles

[0056] The adsorptive porous hydrophobic materials are preferably of a composition that selectively retains from aqueous solution of moderate to high ionic strength, through attractive forces, low molecular weight organic molecules said to exhibit hydrophobic properties owing to the presence of predominately aromatic and aliphatic hydrocarbon substituents, and exhibiting limited solubility in aqueous solution. The hydrophobic materials are often (but not necessarily) crystalline or particulate, and are preferably comprised of either activated charcoal, porous hydrophobic polymers, or either alkyldimethylsilane- or aryldimethylsilane-coated silica and glass particles. Examples of hydrophobic polymers include divinylbenzene, latex, polystyrene and polymethylmethacrylate. Examples of alkyldimethylsilane and aryldimethylsilane treated silica and glass particles include those prepared from butyldimethylchlorosilane (C4), octyldimethylchlorosilane (C8), octadecyldimethylsilane (C18) and phenyldimethylsilane. In presently preferred embodiments, the mean diameter of the porous hydrophobic materials is approximately two orders of magnitude smaller than the diameter of the particle. Particles that include a hydrophobic material are described in, for example, U.S. Pat. No. 5,790,964.

[0057] The porous hydrophobic materials are entrapped within a cross-linked, three-dimensional hydrophilic polymeric matrix. This matrix is highly permeable so that low and relatively low molecular weight molecules (i.e., those having a molecular weight of less than about 5,000 daltons) can pass through the matrix while larger molecules cannot. The porosity of the hydrophilic matrix results from varying the percentage of cross-linking associated with the matrix. Matrices are preferably prepared from either acrylamide, agarose or dextran by cross-linking with appropriate reagents so as to entrap the porous hydrophobic materials while simultaneously controlling the porosity of the hydrophilic matrix. In presently preferred embodiments, the hydrophilic matrices are polyacrylamide, cross-linked at between about 2.5-15%, and most preferably have about 5% cross-linking. The particles are usually three-dimensional, of an irregular shape, and typically have mean diameters of from about 5 to 40 microns.

[0058] In a preferred embodiment, the particles also include a magnetizable constituent to allow the particles to be separated from a mixture by placing the particles in proximity to a magnet. A variety of different magnetizable constituents are suitable for use in the particles. These include, for example, ferric oxide, nickel oxide, barium ferrite, and ferrous oxide. Suitable magnetic particles can be prepared by, for example, mixing equivalent amounts of iron oxide and hydrophobic particles (e.g., charcoal) with acrylamide/bis-acrylamide. These components are vigorously mixed to form a “cake,” which is then passed through a coffee grinder or equivalent. The ground material is then passed through a ball mill to obtain particles that are preferably about 5-20 μm in diameter.

[0059] MagaCharc™ AA particles (Cortex Biochem, San Leandro Calif.) are an example of a commercially available particle that is suitable for use in the methods of the invention. MagaCharc™ particles are prepared by cross-linking of a 1:1 mixture (w/w) of charcoal (Norit SX-Ultra) and iron oxide (Fe₃O₄) within a polymeric matrix that is prepared from 80% (w/w) polyacrylamide cross-linked with 5% N,N-methylene-bis-acrylamide and containing 15% (w/w) acrylic acid. The presence of the iron oxide distributed throughout the particle, preferably uniformly, renders the particle magnetizable and thereby facilitates the removal of supernatants from the particles by placing the particles in proximity to a magnet.

[0060] Nonspecific absorption associated with a hydrophilic matrix can be reduced by including an acrylic acid or methacrylic acid monomer during the polymerization of the matrix. This ensures that the matrix exhibits a residual anionic charge which is mutually repulsive, from an electrostatic standpoint, with respect to the polynucleotides that are present in the primer extension reaction.

[0061] Another way to reduce nonspecific is to precondition the particles with one of several reagents known to minimize the nonspecific absorption of nucleic acids on blotting membranes including bovine serum albumin (BSA), Denhardt's reagent, heparin sulfate, linear polyacrylamide (LPA), nonfat dried milk, polyvinylpyrrolidone (PVP), salmon sperm DNA and Tween 20. For example, particles are preferably batch preconditioned with heparin sulfate.

[0062] B. Purification Methods

[0063] To effect removal of the unincorporated dye molecules from the desired labeled polynucleotides, the particles are added to the mixture that contains the polynucleotides and the dye molecules (e.g., a primer extension reaction or cycle sequencing reaction). The hydrophobic ddNTPs, ddNDPs, ddNMP and dideoxynucleosides selectively permeate the hydrophilic coating of the particles, owing to their relatively low molecular weight as compared to the primer extension products, and are absorbed by the hydrophobic cores. After an appropriate incubation period, and often mixing, the particles are removed from the mixture. The unincorporated dye molecules are removed along with the particles, thus leaving the desired labeled polynucleotides substantially free of the dye molecules.

[0064] Relatively short primer extension products can also be absorbed. However, the loss of short primer extension products can be rendered unimportant by the judicious choice of an appropriate primer. In general terms, the efficiency of dye-labeled terminator removal has been found to be related to the hydrophobicity of each reagent. Consequently, ddNMPs are more efficiently removed than ddNDPs, which are more efficiently removed than ddNTPs.

[0065] The particles are generally added to the reaction mixtures as an aqueous suspension. The reaction mixtures and the particle suspensions should be allowed to come to room temperature before adding the particles to the reaction mixtures. For samples that have been stored at 4° C., this usually takes approximately 15 minutes. Generally, between about 1 μg and 10 μg (inclusive) of particles is added per 1 μl of reaction mixtures. More preferably, between 3 μg and 7 μg (inclusive) of particles is added per 1 μl of reaction mixture, and still more preferably the amount of particles added is between 4 μg and 6 μg per μl of reaction mixture. For example, about 100 μg to about 1 μg, more preferably about 400 μg to about 600 μg, of particles are typically added to about 100 μl of reaction mixture. The supernatant is removed after pelleting the particles with a magnet.

[0066] The particles are incubated in the reaction mixture for a precise period of time, generally between about ten seconds to about 5 minutes. More preferably, the incubation time is about 30 seconds to about 2 minutes. During the incubation, it is preferable to mix the reaction mixtures thoroughly. Mixing can be by any method known to those of skill in the art, including pipetting up and down, and the like. Vortexing is particularly suitable. One example of a suitable commercially available vortexer is the Vortex Genie 2 Model G-560 (Fisher, Pittsburgh, Pa., Catalog No. 12-812). For use with microtiter plates, the Vortex Genie 2 96-well plate head is suitable (Fisher, Catalog No. 12-812C).

[0067] It is often desirable to optimize the amount of particles and the time of incubation for a particular dye-labeled reactant, amount of dye reagent in each reaction, reaction volume, dilution buffer, and/or particular sequencing instrument. For example, DNA sequencers such as the Applied Biosystems Prism 3700® DNA Analyzer, which use laminar flow heads are more sensitive to purification than are other sequencers. Similarly, the use of different dilutions of commercially available DNA sequencing kit reagents can affect the optimal conditions for purification.

[0068] Relatively quick optimization of the purification conditions can be achieved by setting up reactions in a microtiter plate. Varying amounts of particles are then added to different wells, and incubation is conducted for a particular time. To optimize the time of incubation, reactions in different wells can be incubated with the particles for different times. After removal of the particles from the reaction mixtures, the samples are then run on the sequencer, and the conditions that provide the best results are then employed routinely for that sequencer and dye reagent.

[0069] After incubation of the samples with the particles, the particles are removed. Removal can be by any suitable method known to one of skill in the art. In preferred embodiments, magnetic particles are employed and removal is accomplished by placing the sample containers in contact with a magnet. Suitable magnets include, for example, neodynium magnets that exhibit greater than 12 KGs (Kgauss) reminance. Magnets that have rare earth magnetic elements such as the Prolinx RapXtract™ Magnetic Separator are suitable.

[0070] After separation of the particles from the samples, the supernatant can then be removed for use. For example, the samples can be placed on a loading plate which contains loading buffer for use in a sequencer.

[0071] The purification methods of the invention substantially remove the unincorporated dye-labeled reactants. By “substantially remove” is meant that at least about 60% of the unincorporated dye-labeled reactants are removed from a mixture. More preferably, at least about 75% of the unincorporated dye-labeled reactants are removed, still more preferably at least about 90%, and most preferentially substantially all of the unincorporated dye-labeled reactants are removed.

[0072] C. Purification Kits

[0073] The invention also provides kits for carrying out the purification methods of the invention. For example, the invention provides kits that include one or more microtiter plates which have aliquots of the particles in some or all of their wells. Such kits allow ready incorporation of the purification methods of the invention into an automated sequencing operation. The kits can also include written instructions for the use of the particles to remove unincorporated dye-labeled reactants from polynucleotides.

[0074] In presently preferred embodiments, the wells of the microtiter plates contain an amount of the particles that has been determined to provide optimal results for a given dye-labeled reactant, for a given dilution of the reactant, and/or for a particular DNA sequencer. For example, the invention provides kits in which the microtiter plate wells contain an amount of particles that is optimized for a PE Biosystems ABI PRISM® 373, 377, 310, or 3700 sequencer using a specific dilution of PE Biosystems BigDye® Ready Reaction Mix. For example, one kit would be optimized for use with a PRISM® 373 using a 4:20 (total μl BigDye® Ready Reaction Mix: total μl reaction volume) dilution. Table 1 shows the amounts of particles per well in kits for the indicated specific sequencers and dilutions of BigDye® Ready Reaction Mix.

[0075] The following examples are offered to illustrate, but not to limit the present invention.

[0076] A region of pUC19 plasmid DNA was sequenced by DNA Cycle Sequencing using modifications to the ABI PRISM® BigDye™ Terminator Cycle Sequencing Ready Reaction Kit (PE Applied Biosystems, Foster City Calif.). For each sample, the following ingredients were dispensed into MicroAmp™ Reaction tubes with caps (PE Applied Biosystems): 2 μL of pUC19 template DNA (125 ng/μL), 4 μL of primer (1 pmol/μL, M13/pUC-48 24mer), 6 μL of HPLC grade water, and 8 μL of Terminator Ready Reaction Mix (PE Applied Biosystems). Capped tubes were placed in a GeneAmp PCR System 9700 Thermal Cycler (PE Applied Biosystems) preheated to 96° C. The reactions were denatured at 96° C. for five minutes and extended by 30 thermal cycles at 96° C. for 20 seconds, 50° C. for 20 seconds and 60° C. for 4 min.

[0077] After thermal cycling, the reactions were purified away from dye-labeled terminators and other reaction constituents by isopropanol precipitation. Twenty μL of deionized water and 60 μL of isopropanol was added to each of the MicroAmp™ tubes (60% isopropanol in final volume). The tubes were sealed and the contents mixed by inverting several times. Allowed tubes to stand at room temperature for 15 min. Collected recipitate by centrifugation at 2000×g for 45 min. Discarded supernatant and centrifuged at 700×g for 1 min. Dissolved precipitate in 20 μL of formamide containing 10% (w/v) sorbitol and analyzed sample on an ABI PRISM™ 310 Genetic Analyzer (PE Applied Biosystems) by capillary electrophoresis. The raw data recorded is illustrated in FIG. 5. Note the presence of dye-labeled artifacts in the front on the capillary electrophoretic separation (ddNTPs, ddNDPs and ddNMPs) as well as the ddT artifact near the center of the electrophoretic separation. The processed DNA sequencing data corresponding to the raw data is illustrated in FIG. 6. Note that the region between 200 and 214 bases has been assigned by the analysis software the DNA sequence GCTTT TTTTT TTTNA (SEQ ID NO:1), despite the fact that visual inspection of the processed data clearly indicates that the DNA sequence is probably GCTTT CCAGT CGGAA (SEQ ID NO:2). This anomaly results from the presence of the dye-labeled ddT artifact in the capillary electrophoretic separation.

[0078] A region of pUC19 plasmid DNA was sequenced by DNA Cycle Sequencing using modifications to the ABI PRISM® BigDye™ Terminator Cycle Sequencing Ready Reaction Kit (PE Applied Biosystems, Foster City Calif.). For each sample, the following ingredients were dispensed into MicroAmp™ Reaction tubes with caps (PE Applied Biosystems): 2 1 μL of pUC19 template DNA (125 ng/μL), 4 μL of primer (1 pmol/μL, M13/pUC-48 24mer), 6 μL of HPLC grade water, and 8 μL of Terminator Ready Reaction Mix (PE Applied Biosystems). Capped tubes were placed in a GeneAmp PCR System 9700 Thermal Cycler (PE Applied Biosystems), preheated to 96° C. The reactions were denatured at 96° C. for five minutes and extended by 30 thermal cycles at 96° C. for 20 seconds, 50° C. for 20 seconds and 60° C. for 4 min.

[0079] After thermal cycling, the reactions were purified away from dye-labeled terminators by magnetic particle absorption. MagaCharc™ particles (Cortex Biochem, San Leandro Calif.) were batch preconditioned to minimize nonspecific absorption of primer extension products. The supernatant was removed from 5 mL of a 10% (v/v) suspension of MagaCharc™ particles and replaced with 10×Denhardt's reagent (Denhardt, D. T., Biochem. Biophys. Res. Commun.1966, 23, 641). The suspension was placed on a rotary mixer for 1 hour at room temperature and then washed 3 times with 5 mL aliquots of HPLC grade water and then finally suspended in 5 mL of HPLC grade water.

[0080] To each well of a multiwell plate was added 100 μL of a 10% (v/v) suspension of MagaCharc™ particles. The multiwell plate was placed on a magnet to pellet the particles and the supernatant removed. To each well was then added 20 μL of cycle sequencing reaction product and the contents of each well were mixed by drawing the particle suspension into a pipette tip and then displacing the suspension 10 times. The multiwell plate was then placed on a magnet to pellet the particles and the supernatants transferred to sequencer sample tubes. To each of the tubes was added 20 μL of formamide containing 10% (w/v) sorbitol. The samples were mixed by vortexing briefly and analyzed on an ABI PRISM™ 310 Genetic Analyzer (PE Applied Biosystems) by capillary electrophoresis.

[0081] The raw data recorded is illustrated in FIG. 7. Note that after purification by the method described above, only the ddTTP and ddTCP artifacts are recorded at magnitudes exceeding those of the primer extension products. The processed DNA sequencing data corresponding to the raw data is illustrated in FIG. 8. Although some residual ddT artifact can been seen by careful inspection of the baseline in the region between 190 and 200 bases, the magnitude of the artifact signal is so low relative to the signals associated with the primer extension products that the analysis software assigned the correct DNA sequence even in the presence of trace amounts of the artifact.

[0082] A region of pUC19 plasmid DNA was sequenced by DNA Cycle Sequencing using modifications to the ABI PRISM® BigDye™ Terminator Cycle Sequencing Ready Reaction Kit (PE Applied Biosystems, Foster City Calif.). For each sample, the following ingredients were dispensed into MicroAmp™ Reaction tubes with caps (PE Applied Biosystems): 2 μL of pUC19 template DNA (125 ng/μL); 4 μL of primer (1 pmol/μL, M13/pUC-48 24mer); 6 μL of HPLC grade water; and 8 μL of Terminator Ready Reaction Mix (PE Applied Biosystems) that had been diluted (1:1) with halfBD™ Dye Terminator Sequencing Reagent (Genpak Ltd., Stony Brook N.Y.). Capped tubes were placed in a GeneAmp PCR System 9700 Thermal Cycler (PE Applied Biosystems), preheated to 96° C. The reactions were denatured at 96° C. for five minutes and extended by 30 thermal cycles at 96° C. for 20 seconds, 50° C. for 20 seconds and 60° C. for 4 min.

[0083] After the thermal cycling, the reactions were purified away from dye-labeled terminators by magnetic particle absorption. MagaCharc™ particles (Cortex Biochem, San Leandro Calif.) were batch preconditioned to minimize nonspecific absorption of primer extension products as described in Example 2. To each well of a multiwell plate was added 100 μL of a 10% suspension of MagaCharc™ particles. The multiwell plate was placed on a magnet to pellet the particles and the supernatant removed. To each well was then added 20 μL of cycle sequencing reaction product and the contents of each well were mixed by drawing the particle suspension into a pipette tip and then displacing the suspension 10 times. The multiwell plate was then again placed on a magnet to pellet the particles and the supernatants transferred to sequencer sample tubes.

[0084] To each of the tubes was added 20 μL of formamide containing 10% (w/v) sorbitol. The samples were mixed by vortexing briefly and analyzed on an ABI PRISM™ 310 Genetic Analyzer (PE Applied Biosystems) by capillary electrophoresis. The raw data recorded is illustrated in FIG. 9. Note that after purification by the method described above, none of the dye-labeled artifacts are recorded at magnitudes exceeding those of the primer extension products. The processed DNA sequencing data corresponding to the raw data is illustrated in FIG. 10. Careful inspection of the baseline reveals that dye-labeled ddT artifact cannot be detected.

[0085] This Example shows a time course and a titration curve for the removal of unincorporated fluorescent dye-labeled deoxynucleotides using MagaCharc™ particles Cortex Biochem, San Leandro Calif.). 8 μl of Big Dye in a total volume of 20 μl (mock reaction containing no template) was heated to 96° C. for 30 min.

[0086] The time course experiment was conducted by adding the amounts of MagaCharc™ particles shown in FIG. 11 to the mixture of dye-labeled ndeoxynucleotides. The samples were mixed by vortexing and incubated for the indicated time periods. The relative fluorescent units remaining in the mixture for each of the time points is shown in FIG. 11. These results show that over 70% of the fluorescently labeled molecules are removed after a 60 second incubation when 240 μg or more of the magnetic particles are used. Further incubation time did not significantly increase the amount removed.

[0087] To investigate the effect of increasing amounts of magnetic particles on removal of the fluorescent dyes, mixtures of dye-labeled deoxynucleotides were heated as described above, brought to room temperature, and the indicated amounts of MagaCharc™ particles were added. The samples were vortexed for 1 minute, and the magnetic particles removed. The amount of relative fluorescent units remaining in the liquid phase was then determined.

[0088] The results of this experiment are shown in FIG. 12. The maximum effectiveness was achieved at about 200 μg of MagaCharc™ particles.

[0089] This Example describes examples of reaction conditions that are suitable for use with the purification methods and kits of the invention. In the following reaction mixtures, all template and primer DNA is preferably resuspended in deionized water. These reaction conditions are well-suited for use of pUC and pGem templates.

[0090] 4:20 Reactions (4 μBigDye® Ready Reaction Mix (PE Biosystems) in 20 μl reaction)

[0091] Add reagents in the following order:

[0092] Gently tap tube to mix

[0093] Briefly centrifuge to collect sample at the bottom of the tube

[0094] For best results, prepare sequencing reactions first by preparing a master mix of BigDye® Ready Reaction Mix with PE Biosystems 5×Sequencing Buffer (e.g., a mix containing 200 μl of PE Biosystems 5×Sequencing Buffer, 400 μl of BigDye® Ready Reaction Mix and 800 μl of deionized water will be sufficient for 96 samples); and next by aliquoting 14 μl of master mix to tubes containing each primer/template mixture.

[0095] 4:10 Reactions (4 μl BigDye® Ready Reaction Mix (PE Biosystems) in 10 μl reaction)

[0096] Add reagents in the following order:

[0097] Gently tap to mix

[0098] Briefly centrifuge to collect sample at the bottom of the tube

[0099] It is understood that the examples and embodiments described herein are for illustrative purposes only and that various modifications or changes in light thereof will be suggested to persons skilled in the art and are to be included within the spirit and purview of this application and scope of the appended claims. All publications, patents, and patent applications cited herein are hereby incorporated by reference for all purposes.