Nucleic acid therapeutics applications notebook Table of Contents

Introduction...... 3 Analysis of Nucleic Acids...... 5 Analysis of Phosphorothioate Oligonucleotides...... 6 Analysis of ss- and dsRNA...... 8 Anion-Exchange ESI/MS of Oligonucleotides—Analysis of Impurities in the Target Oligonucleotide Product ...... 12 Anion-Exchange ESI/MS of Oligonucleotides—Analysis of Oxidation Products of Purified Biotinylated Oligonucleotides...... 15 Analysis of 2´, 5´-Linkage Isomers in RNA and DNA, and Elucidation of Aberrant Linkage Position...... 17 Analysis of Nucleoside Mono-, Di-, and Triphosphates...... 22 Method Development Using Anion-Exchange Chromatography for Nucleic Acids...... 23 Application Notes...... 29 High-Resolution Analysis and Purification of Oligonucleotides with the DNAPac PA100 Column...... 30 Product Focus: Systems ...... 36 Biocompatible LC Systems...... 37 Orbitrap MS Instruments...... 38 Product Focus: Consumables ...... 39 Nucleic Acid Columns ...... 40 Column Selection Guide for Oligonucleotide Separations ...... 41 Product Focus: Software ...... 42 Chromeleon 7 Software...... 43 References...... 44 Select Peer-Reviewed Publications...... 45 Introduction

THERAPEUTIC NUCLEIC ACIDS MARKETPLACE There are various models in which the active pharma- Over the last 15–20 years, development of nucleic acid ceutical ingredient (API) is mainly RNA. The first, short in- therapeutics has exploded from a modest but dedicated terfering RNA (siRNA) employs specific short (19–23 base) effort by specialty research and development organizations RNA sequences that are incorporated into a cellular complex into a widespread enterprise fully embraced by all major RNA-induced silencing complex (RISC) that functions to pharmaceutical firms. This explosion is supported by degrade RNA complementary to the siRNA. This results in the discovery of Ribonuclease H (RNase H) function for diminution of a very specific subset of cytosolic mRNA that antisense oligonucleotides, and the more recent realization is targeted by the administered sequence, severely limiting that cellular mechanisms for control of gene expression and the expression of the target gene product. transcriptome regulation can be manipulated to produce Immunostimulatory RNA (isRNA) is a mode that therapeutics for ailments that were previously considered employs sequence specific motifs, some of which are very unresponsive to drugs. Recently, options for therapeutics simple (e.g., CpG where the C is not methylated). Introduc- with exquisite selectivity also provide the opportunity for tion of the specific RNA sequence motif induces a cellular new therapeutic modes and intellectual properties. For immune system directing degradation of foreign nucleic example, Macugen successfully passed clinical trials and acids. This is mediated by the cellular toll-like receptors release protocols to become the first approved aptamer (TLRs), TLR7, 8, and 9. therapeutic. This success encourages further efforts and miRNA is another RNA model that employs a series opportunities for therapeutic nucleic acid developers. of cellular complexes to degrade mRNA in a sequence- specific fashion. This is almost certainly the pathway that THERAPEUTIC NUCLEIC ACID MODELS is conscripted by the siRNA approach, but represents a Several models of nucleic acid therapeutics have been mechanism whereby the cell controls expression and turn- described, including antisense oligodeoxynucleotides; over of specific mRNA. Many miRNA families described ribonucleic acid (RNA) interference; immunostimulatory regulate different functions in plants and animals, including nucleic acids; microRNA (miRNA) (along with other neogenesis and malignant transformations. The sources of noncoding RNA modes); and aptamers. the endogenous miRNA in humans include highly conserved Antisense oligonucleotides are nucleic acid therapeu- noncoding regions on chromosomes (initially thought to be tics that are injected into a person for distribution to target junk DNA) and sequences inserted in specific genes. There organs by various means. From there, these therapeutics are other noncoding RNA (ncRNA) models, with important recruit RNase H to degrade the resulting deoxyribonucleic ncRNA motifs being reported regularly. These may arise acid (DNA)/RNA hybrids, thus minimizing translation of the from endogenous variations on the miRNA process de- target messenger RNA (mRNA). scribed above.

3 Introduction Aptamers are synthetic nucleic acids selected to inter- 3) anion exchange-HPLC-electrospray ionization mass act with specific structures—such as cell surface receptors, spectrometry (ESI-MS) of derivatized oligonucleotides; proteins, and other chemical signatures—in a highly specific 4) preparative and analytical separations of nucleic acid manner. They are typically modified with protecting groups diastereoisomers; 5) coupling of anion-exchange HPLC to (2´-O-methyl phosphorothioate linkages, DNA bases, ESI-MS using automated desalting with a Titanium-based sometimes inverted 3´-3´ linkages, and conjugation to LC system; 6) separation and analysis of RNA contain- polyethylene glycol). Most therapeutic targets for aptamers ing aberrant linkage isomers (not resolved by single-stage have been extracellular, so they operate in the circulatory ESI-MS alone, including identification of the position of the system where these protecting groups prolong their circula- aberrant linkages); 7) analysis of nucleoside mono-, di-, and tory half-life. triphosphates; and 8) useful information for new method development. BIOSEPARATION SOLUTIONS FOR NUCLEIC ACID THERAPEUTICS THERMO SCIENTIFIC AND DIONEX INTEGRATED SYSTEMS Thermo Scientific™ offers solutions for the purification Dionex™ Products are now a part of the Thermo Scien- and analysis of essentially all classes of oligonucleotides and tific brand, creating exciting new possibilities for scientific their modifications. Purification efforts are supported by both analysis. Now, leading capabilities in LC, ion chroma- high-capacity Thermo Scientific DNASwift™ SAX-1S Mono- tography (IC), and sample preparation are together in one lith Columns (5 × 150 mm) and DNAPac™ PA-100 Semi-Pre- portfolio with those in MS. Combining Dionex’s innovation parative Columns (9 × 250 mm and 22 × 250 mm). These are in chromatography with Thermo Scientific’s leadership housed in bioinert column bodies to prevent column fouling position in MS, a new range of powerful and simplified by corrosion of stainless steel frits and tubing, and typically workflow solutions is now possible. employ either PEEK™ or Titanium high-performance liquid These Thermo Scientific integrated solutions can expand chromatography (HPLC) systems. The solutions described your capabilities and provide tools for new possibilities: in this notebook include: 1) analysis of phosphorothioate • IC and MS oligonucleotides; 2) analysis of single-stranded (ss) and • LC and MS double-stranded (ds) RNA for therapeutic applications; • Sample preparation and MS

4 Introduction Analysis of nucleic acids Nucleic acid therapeutics applications notebook Analysis of Phosphorothioate Oligonucleotides

Phosphorothioate (PS) linkages are introduced into another by only one (All PS vs. 1 PO and 1PO vs. 2PO) or oligonucleotides to restrict the susceptibility of therapeutic two (All PS vs. 2PO) atoms. oligonucleotides (ONs) to degradation by plasma and The broad peaks suggest that some of the 16,384 dia- tissue nucleases. A very large fraction of therapeutic ONs stereoisomers may be resolved from one another. In some harbor one or more PS linkages, so their properties must be cases, only one or two PS linkages are employed, so the thoroughly understood for effective method development. diastereoisomers from these ONs may be resolved. Anion-exchange chromatography on DNAPac columns allows resolution of ONs that differ in some cases by only one atom. One example of this includes analysis of antisense O B O B1 1 Column: DNAPac PA100 (4 × 250 mm) ONs. These typically employ phosphorothioate linkages. Eluents: A. 20 mM Tris pH 8 (pH 8) Phosphorothioates are ONs in which one of the oxygen 24 mM NaOH (pH 12.4) H B. A + 0.375 M NaClO O O H O O 4 bound only to the phosphorous atom is replaced with a Gradient: 15–86% B in 20 min (pH 8) P P 15–88% B in 20 min (pH 12.4) sulfur atom. The introduction of the sulfur atom results in a S O O O (both nonlinear, curve 4) Flow Rate: 1.5 mL/min chiral center at the phosphorus atom, so each linkage harbors O B O B 2 2 Temperature: 30 °C two configurations and forms diastereoisomers. An ON with Inj. Volume: 10 µL 14 Detection: UV, 260 nm 15 bases can harbor 14 linkages and thus, 2 (16,384) pos- H H Phosphorothioate Phosphodiester Sample: dT15, Reversed-phase purified sible diastereoisomers. This results in fairly broad peaks. In Linkage Linkage the example presented here (Figure 1), dT15 phosphorothio- All PS 1 PO ate, previously purified by reversed-phase chromatography, 2 PO was analyzed at pH 8 and 12.4 using NaClO4 eluent. At pH 8, the components elute with broad peaks and are poorly resolved. The major peak represents the full length, fully substituted phosphorothioate. The broad plateau eluting prior to the major peak represents n-x and incompletely thioated ONs. At pH 12.4, the peaks eluting earlier than the major component are much better resolved, and represent incompletely phosphorothioated components which lack the sulfur atom at one or more linkages. These elute as partially resolved peaks between 7 and 15.5 min. Thus, the peaks 13078 labeled here as All PS, 1 PO, and 2 PO differ from one Figure 1. Resolution of phosphorothioate oligonucleotides with incomplete thiolation.

6 Analysis of Phosphorothioate Oligonucleotides In Figure 2, purification of phosphorothioate diastereo- Column: DNASwift SAX-1S (5 × 150 mm) Temperature: 30 °C isomers is shown. A 21-base sequence—the sense strand Eluents: A. 40 mM Tris.Cl, pH 7 Inj. Volume: 2 µL, 6 mg/mL B. A + 1.25 M NaCl Detection: UV, 260 nm of an eGFP RNAi sample as both RNA and as DNA—is Gradient: 24–48% B in 16.7 min Sample: eGFP Sense, 6 mg/mL Flow Rate: 1.5 mL/min separated on a DNASwift SAX-1S column. This is a hybrid (Sequence and PS monolith, where anion-exchange nanobeads, with chemistry positions as shown) eGFP* (sense strand): 5´ AGC UGAs CCC UGA AGsU UCA UdCdT 3´ similar to that of the DNAPac PA200 column, are attached to * __ ___ indicates position of phosphorothioate linkage s the monolith surface. Four diastereoisomers 950 In this example, two phosphorothioate (PS) linkages are 5´____ s ____ s ____3´ 5´____ s ______3´ inserted at linkages 6 and 14 in the sequence. When sepa- s A C rated on the DNASwift (or DNAPac) columns, the four DNA 5´______s ____ 3´ B 5´____ s ______3´ D components are resolved into three peaks, and the four RNA s s components are all resolved. RNA, Sense, 2 PS (4 isomers) mA Since the DNASwift column is designed with capacity 260 B, C similar to porous-bead based anion exchangers (but with bet- A ter peak shape), this separation can be scaled up for lab-scale D purifications. DNA, Sense, 2 PS (4 isomers) Here, all four components from the RNA sample were –50 isolated, desalted, and subjected to ESI-MS, where each 010 20 Minutes component produced the mass of the full-length ON with 27299 both PS linkages. Since this sequence does not self-associate, this observation demonstrates that the diastereoisomers are Figure 2. Separation of isobaric diastereoisomers on the DNASwift SAX-1S column. resolved.

7 Analysis of Phosphorothioate Oligonucleotides Analysis of ss- and dsRNA

The discovery that short dsRNA species can effect either strand is considered an impurity, and so it must be exquisite control of the expression and turnover of RNA in characterized. animals, plants, and protists led to an explosive increase in RNA and DNA differ by only one atom per nucleotide. the development of potential RNA therapeutics. The applica- This difference allows them to adopt slightly different solu- tions presented here underline the crucial role the DNAPac tion conformations, and thus slightly different interactions plays in RNA therapeutic analysis and meeting the needs of with the stationary phases. Figure 4 demonstrates that DNA developers preparing pharmaceuticals in this rapidly growing tends to elute significantly earlier than RNA at pH 7. RNA field. might be expected to elute later at a very high pH (> 12), This new class of pharmaceuticals consists of two where the 2´ hydroxyl will begin to ionize, but not at pH 7 as complementary ssRNAs of 19–21 bases that are annealed shown. Hence, the solution conformation influences the ON to form a single dsRNA. In practice, excess ssRNA of interaction with the stationary phase. RNA degrades at pH values as low as 8, hence, pH values above 8 have been considered as a poor choice for Column: DNASwift SAX-1S (5 × 150 mm) Temperature: 30 °C the separation of RNA ONs. However, pH-induced RNA Eluents: A. 40 mM Tris, pH 7 Inj. Volume: 2, 4 µL, 6 mg/mL B. A + 1.25 M NaCl Detection: UV, 260 nm degradation is a slow process, and does not usually occur in Gradient: 24–48% B in 16.7 min Sample: 21 mer Sense RNA Flow Rate: 1.5 mL/min the time frame for ON analysis on column. To evaluate the eGFP (sense strand): 5´ AGC UGA CCC UGA AGU UCA UdCdT 3´ RNA degradation by pH during chromatography, an RNA sample was run at pH 11 using different flow rates to change the oligoribonucleotide residence time on the column. In the example presented here (Figure 3), the number of peaks and relative peak area for both RNA and DNA remained essentially unchanged for residence times from 7 to 21 min, indicating a lack of on-column degradation for mA 260 that time period even at this very high pH, representing a thousand-fold increase in hydroxide concentration over In RNA form pH 8. Further, control of extensive hydrogen bonds, such as those formed in G-tetrad ladders of considerable length, may require pH values up to 12.4 or temperatures > 95 °C. On the In DNA form lower pH end, depurination becomes more likely as the pH falls to ≤ 6. Since depurination will lead to strand scission, 01020 Minutes pH values ≤ 6.5 are not typically recommended for ON AEC. 27300

Figure 3. Relative retention of RNA and DNA on a DNASwift mono- lith column.

8 Analysis of ss- and dsRNA By modifying chromatographic temperature, conditions Column: DNAPac PA200 (4 × 250 mm) Eluents: A. 4 mM diisopropylamine to pH 11 with NaOH to better resolve each component may be obtained. As the B. A + 0.33 M NaClO4 Gradient: 30–70% B in 3.2 column volumes temperature is increased, the elution positions of the two Flow Rate: 1.0 mL/min (top), 0.67 mL/min (middle), 0.33 mL/min (bottom) ssRNAs and the dsRNA will increase. This optimizes the Temperature: 30 °C Inj. Volume: 5 µL resolution of all three components for the new RNA pharma- Detection: UV, 260 nm ceuticals, allowing ssRNA impurity characterization. Sample: eGFP antisense RNA (sequence as shown) Sequence: 5’AUG AAC UUC AGG GUC AGC UdTdG 3’ 20 Flow: 1.00 mL/min 59.6 95.3 95.3 1.25 M NaCl: Column: DNAPac PA200 (2 × 250 mm) Temperature: 30 °C Eluents: A. 20 mM Tris pH 7 Inj. Volume: 10 µL

43.2 % B. A + 1.25 M NaCl Detection: UV, 260 nm 1.89 1.25 1.25 0.93 0.63 Gradient: 26–56 % B in 15 min Sample: Sense and antisense strands of –2 Flow Rate: 300 µL/min 4.5 6.0 7.5 eGFP ssRNA, ssDNA, and duplexes 20 Flow: 0.67 mL/min 60.3 Antisense sequence: 5’-AUGAACUUCAGGGUCAGCUdTdG-3’ 95.3 95.3 Sense sequence: 5’-AGCUGACCCUGAAGUUCAUdCdT-3’ 1.25 M NaCl: mA

260

44.3 % 1,230 eGFP antisense eGFP sense 1.88 1.26 0.93 0.69 RNA RNA –2 6.15 8.00 10.00 12.15

20 Flow: 0.33 mL/min 60.3

95. 2 1.25 M NaCl:

46.9 % mA260 1.96 dsRNA

1.22 0.90 0.70 0.70 –2 15.0 19.0 23.5 Minutes eGFP antisense RNA 27301 eGFP sense RNA Figure 4. On-column stability: effect of high pH on eGFP antisense Duplex eGFP RNA RNA. Peak labels indicate relative area (area %) at 7, 10.5, and (excess antisense strand) 20.5 min residence time; pH 11. –20 0510 15 Minutes 27302

DNAPac and DNASwift columns are useful for Figure 5. Resolution of duplex, ssRNA at 30 °C on the DNAPac the analysis of both ss- and dsRNA. The use of elevated PA200 column: 325–750 mM NaCl in 17.2’, 300 µL/min, pH 7. temperatures to control hydrogen bonding, within or between ONs, is a widely used technique. Pellicular anion-exchange chromatography of dsRNA and DNA/RNA hybrids can also Figure 6 shows that the retention of each RNA compo- benefit from careful selection of temperature during chroma- nent has increased, thereby exchanging the relative elution tography. As shown in Figure 5, two complementary RNA order of the ds- and antisense ssRNA strands. strands, and the duplex formed after annealing them together, were individually analyzed on a DNAPac PA200 column at pH 7 and 30 °C. Under these conditions, the two ssRNA were well resolved from one another, and both were resolved from the duplex. In this example, a small molar excess of the antisense strand was present, and the excess appeared as a small peak eluting at the position of the antisense ssRNA. This is typical of many efforts to prepare dsRNA; this sepa- ration technique demonstrates an easy titration method of the two ssRNA components to molar equivalence.

9 Analysis of ss- and dsRNA At 60 °C, it is seen that the retention of each RNA com- Column: DNAPac PA200 (2 × 250 mm) Temperature: 40 °C Eluents: A. 20 mM Tris pH 7 Inj. Volume: 10 µL ponent has increased further, causing the elution order of the B. A + 1.25 M NaCl Detection: UV, 260 nm two ssRNAs (sense and antisense) to reverse. Gradient: 26–56% B in 15 min Sample: Sense and antisense strands of Flow Rate: 300 µL/min eGFP ssRNA, ssDNA, and duplexes

Antisense sequence: 5’-AUGAACUUCAGGGUCAGCUdTdG-3’ Sense sequence: 5’-AGCUGACCCUGAAGUUCAUdCdT-3’ Column: DNAPac PA200 (2 × 250 mm) Temperature: 60 °C Eluents: A. 20 mM Tris pH 7 Inj. Volume: 10 µL 1,230 eGFP antisense eGFP sense B. A + 1.25 M NaCl Detection: UV, 260 nm RNA Gradient: 26–56% B in 15 min RNA Sample: Sense and antisense strands of Flow Rate: 300 µL/min eGFP ssRNA, dsRNA ssDNA, and duplexes Antisense sequence: 5’-AUGAACUUCAGGGUCAGCUdTdG-3’ Sense sequence: 5’-AGCUGACCCUGAAGUUCAUdCdT-3’ 630 eGFP antisense mA260 RNA eGFP antisense RNA eGFP sense RNA dsRNA eGFP sense RNA Duplex eGFP RNA (excess antisense strand)

–20 mA260 0510 15 Minutes eGFP antisense RNA 27303

Figure 6. Resolution of duplex, ssRNA at 40 °C on the DNAPac eGFP sense RNA PA200 column: 325–750 mM NaCl in 17.2’, 300 µL/min, pH 7 Duplex eGFP RNA (excess antisense strand) –20 0 7.5 15.0 As seen in Figure 7, at 50 °C, the retention of each RNA Minutes 27305 component has increased, causing the two ssRNAs (sense and antisense) to elute closer together. Figure 8. Resolution of duplex, ssRNA at 60 °C on the DNAPac PA200 column: 325–750 mM NaCl in 17.2’, 300 µL/min pH 7.

Column: DNAPac PA200 (2 × 250 mm) Temperature: 50 °C At 70 °C (Figure 9), the retention of each RNA compo- Eluents: A. 20 mM Tris pH 7 Inj. Volume: 10 µL B. A + 1.25 M NaCl Detection: UV, 260 nm nent has again increased, causing the two ssRNAs (sense and Gradient: 26–56% B in 15 min Sample: Sense and antisense strands of antisense) to elute more and further apart. Flow Rate: 300 µL/min eGFP ssRNA, ssDNA, and duplexes : 1,230 Antisense sequence 5’-AUGAACUUCAGGGUCAGCUdTdG-3’ Sense sequence: 5’-AGCUGACCCUGAAGUUCAUdCdT-3’

eGFP eGFP antisense RNA sense RNA

dsRNA

mA260 eGFP antisense RNA

eGFP sense RNA

Duplex eGFP RNA (excess antisense strand)

–20 0612 18 Minutes 27304

Figure 7. Resolution of duplex, ssRNA at 50 °C on the DNAPac PA200 column: 325–750 mM NaCl in 17.2’, 300 µL/min, pH 7

10 Analysis of ss- and dsRNA Column: DNAPac PA200 (2 × 250 mm) Temperature: 70 °C Column: DNAPac PA200 (2 × 250 mm) Temperature: 80 °C Eluents: A. 20 mM Tris pH 7 Inj. Volume: 10 µL Eluents: A. 20 mM Tris pH 7 Inj. Volume: 10 µL B. A + 1.25 M NaCl Detection: UV, 260 nm B. A + 1.25 M NaCl Detection: UV, 260 nm Gradient: 26–56% B in 15 min Gradient: 26–56% B in 15 min Sample: Sense and antisense strands of Sample: Sense and antisense strands of Flow Rate: 300 µL/min Flow Rate: 300 µL/min eGFP ssRNA, eGFP ssRNA, ssDNA, and duplexes ssDNA, and duplexes Antisense sequence: 5’-AUGAACUUCAGGGUCAGCUdTdG-3’ Antisense sequence: 5’-AUGAACUUCAGGGUCAGCUdTdG-3’ Sense sequence: 5’-AGCUGACCCUGAAGUUCAUdCdT-3’ Sense sequence: 5’-AGCUGACCCUGAAGUUCAUdCdT-3’ 1,230 eGFP antisense 1,230 RNA eGFP antisense eGFP sense RNA RNA eGFP sense dsRNA RNA

mA260 mA260 eGFP antisense RNA eGFP antisense RNA dsRNA (melting)

eGFP sense RNA eGFP sense RNA Duplex eGFP RNA Duplex eGFP RNA (excess antisense strand) (excess antisense strand) –20 –20 0612 18 0 7.5 15.0 Minutes Minutes 27306 27307

Figure 9. Resolution of duplex, ssRNA at 70 °C on the DNAPac Figure 10. Resolution of duplex, ssRNA at 80 °C on the DNAPac PA200 column: 325–750 mM NaCl in 17.2’, 300 µL/min, pH 7. PA200 column: 325–750 mM NaCl in 17.2’, 300 µL/min, pH 7.

At 80 °C (Figure 10), the retention of each RNA a difference in the peak areas for the sense and antisense component has increased, and the duplex has begun to melt strands in the duplex preparation due to partial melting of the (causing both ssRNAs–sense and antisense–to appear in duplex. This peak area difference allows a direct assessment the duplex chromatogram). Therefore, sense and antisense of the relative molar inequivalence of the sense and antisense strands are less well resolved at 80° C than at 70 °C. strands. Using the molar extinction coefficients and the peak This series of temperature assays (30–80 °C) reveals areas of each strand, the user can calculate the amount (in the ability to optimize the separation of both ssRNAs from this case, of sense strand) required to prepare a perfect 1:1 each other and from the duplex. The 80 °C trace also reveals molar ratio for duplex formation.

11 Analysis of ss- and dsRNA Anion-Exchange ESI/MS of Oligonucleotides— Analysis of Impurities in theTarget Oligonucleotide Product

Mass spectrometry capabilities have been applied to WPS-3000 TBFC AE-separated ONs after manual desalting on C18 cartridges Sample Loop 1 From Pump Quarternary by ESI and MALDI-TOF approaches. These techniques Syringe 2 6 Valve IV Pump are valuable for the identification of ON impurities. ESI 53 Bridge 4 was shown to provide superior results compared to simple Tube Sampling 1 Needle 2 6 OligoTrap DNASwift MALDI-TOF techniques, especially for longer ONs. A FV CSV Waste 3 5 or Acclaim or DNAPac high-throughput method for automated desalting of IP- 4 Fraction PA-II Valve Z RPLC-separated ONs was developed by Novatia LLC. Wash Liquid Waste Reservoir Syringe Here, this method is adapted to automatically desalt anion- Y X UV Detector exchange-separated ONs. The automated system employs Well Plate or Sample Tray From Wash Detector Port the Thermo Scientific Dionex UltiMate 3000 Titanium LC Carousel system equipped with the WPS-3000 TBFC fraction-collect- Waste ing autosampler as shown in Figure 11. With this system, 27308 ONs are applied to DNAPac or DNASwift pellicular AE Figure 11. Automated purification and desalting using a well-plate sampler with injection, fraction collection, and column selection columns, and eluted using a salt gradient. Peaks detected by valves. absorbance are collected to unique positions in the autosam- pler, and the collected fractions automatically desalted on an Acclaim® reversed-phase guard column using an ion-pair reagent and methanol as eluents. The desalted sample may be collected and submitted for ESI-MS at a later time, or im- mediately directed to the ESI-MS inlet.

12 Anion-Exchange ESI/MS of Oligonucleotides—Analysis of Impurities in the Target Oligonucleotide Product In the example presented here (Figure 12), an oligode- The starting flow rate of an ion-pair eluent system (as oxynucleotide (ODN) was purified on a DNAPac PA200 shown in Figure 13) was 0.7 mL/min. The blue trace indi- column using NaCl as eluent and at pH 12, to examine the cates the output of a high-sensitivity conductivity detector, purity of a DNAPac-purified ON. and shows elution of the salt from the sample in the first 20 s. At 0.7 min, the eluent switches from 1.5% to 40%

CH3OH in the ion-pair eluent, and the flow is reduced to 0.3 mL/minute for possible introduction into the MS. The Column: DNAPac PA200 (2 × 250 mm) Flow Rate: 300 µL/min Eluents: A. Deionized water Temperature: 30 °C ON elutes between 1.7 and 2.0 min, and the initial conditions B. 1.25 M NaCl Inj. Volume: 20 µL, 37 µg/mL C. 0.1 M NaOH Detection: UV, 260 nm are restored to re-equilibrate the cartridge at 2.3 min. Gradient: 16–68% B in 20 min Sample: 21mer DNA, sequence indicated When the UltiMate WPS-3000 TBFC was used, the Isocratic 10% C Fractions: 1; fraction 70 sample was collected back into the autosampler instead of Dx-96: 5´ ATT gTA ggT TCT CTA ACG CTg 3´ being directed into a mass spectrometer; thus, a flow rate 300 reduction during ODN elution (for MS compatibility) was 17.72 optional.

Column: Acclaim PA II (4.6 × 50 mm) Flow Rate: 700 µL/min for 0.8 min Eluents: A. 0.375% DIEA, 300 µL/min for 1.6 min 0.75% HFIP, 10 µM EDTA 700 µL/min for 1.6 min B. 40 % CH OH in eluent A Temperature: 30 °C mA 3 260 Gradient: 1.5% B for 0.7 min, Inj. Volume: 20 µL 100% B for 1.0 min Detection: Conductivity for salt elution 1.5% B for 2.3 min UV, 260 nm for oligonucleotide Sample: DNAPac-Purified Dx 131 0.09 1.85 Oligonucleotide Elution of salt plug 100%

–10 (70-70) Fr. 12 15 18 21 Minutes 27309 40% CH3OH: 1.5% 1.5% Elute Figure 12. Purification of a 21-base oligonucleotide, MW = 6427. Switch eluent, oligonucleotide 80–195 mM NaClO in 15 min, pH 12. Salt reduce flow 4 removal for mass spec. Re-equilibrate cartridge

Absorbance, mA260 As shown in Figure 13, the purified ODN sample was Conductivity, µS desalted on an Acclaim 4.6 × 50 mm guard column before Flow: 700 µL/min Flow: 700 µL/min submission to ESI-MS. Flow: 300 µL/min

0 1.5 3.0 Minutes 27310

Figure 13. Oligonucleotide desalting on an Acclaim PA-II column;

IP-RPLC retention, oligonucleotide elution with CH3OH.

13 Anion-Exchange ESI/MS of Oligonucleotides—Analysis of Impurities in the Target Oligonucleotide Product The ProMass deconvolution peak report indicates the 4 presence of three nucleic acid components and a sodium ad- 12.0E duct. This purified ODN’s two contaminants include an n-C Dx96 Deconvoluted ESI-MS 6427.3 and an n-A–both at < 5% of the intensity of the full-length 10.0E4 ODN. Thus, the DNAPac-purified ODN harbored small amounts of two different n-1 impurities, which were identi- 8.0E4 fied by AXLC-ESI/MS.

6.0E4 Intensity

4.0E4

2.0E4 6114.7 6138.0 6450.2

0.0E4 6000 6100 6200 6300 6400 6500 6600

Dx96: 5´ A TTg TAg gTT CTC TAA CGC Tg 3´ Mass Peak List Sorted by Intensity: Mass Intensity Delta Mass Relative% %Total Presumed Identity 6427.3 9.1E+4 0.0 100.00 90.3 Target, 6427.2, A1-G21 6114.7 4.3E+3 –313.4 4.44 4.0 T2-G21, or n-A 6450.2 2.0E+3 22.6 3.10 2.8 n+Na+ (adduct ) 6138.0 2.0E+0 –289.6 2.10 1.9 n -C 6436.5 1.1E+3 8.5 1.10 1.0 27311

Figure 14. The (deconvoluted) spectra of the sample in Figure 13. The ESI-MS shows impurities in Dx96 primary peak.

14 Anion-Exchange ESI/MS of Oligonucleotides—Analysis of Impurities in the Target Oligonucleotide Product Anion-Exchange ESI/MS of Oligonucleotides— Analysis of Oxidation Products of Purified Biotinylated Oligonucleotides

To further demonstrate the separation, desalting, and Column: DNAPac PA200 (2 × 250) mm) Inj. Volume: 5 µL ESI-MS process, a biotinylated oligodeoxynucleotide was Eluents: A. 10 mM NaOH (pH 12) Detection: UV, 260 nm selected with the sequence shown. This ODN was purified B. 1.25 M NaCl in eluent A Sample: Derivatized oligonucleotide, Gradient: 16–68% B in 20 min 0.2 mg/mL on a 2 × 250 mm DNAPac PA200 column using a NaCl Flow Rate: 300 µL/min Fractions: # 83, collected with WPS BTFC autosampler using gradient at 300 µL/min at pH 12. Under these conditions, the peak collection full-length sample appears to resolve from its degradation Biotin: 5´ CTG CTT GTA GGA TCT TTA AAG ACG A 3´ 300 products and the ON failure sequences (Figure 15). 17.44 85.0 O Biotin HN NH 68.0

H2N

mA260 S O 1.25 M NaCl: 16% 16.0

0 Fr. (83) Fr.

01020 Minutes 27312

Figure 15. DNAPac purification and desalting of Dx-131 (5´-biotin).

15 Anion-Exchange ESI/MS of Oligonucleotides—Analysis of Oxidation Products of Purified Biotinylated Oligonucleotides After desalting, the sample is submitted to ESI-MS analysis, and the raw and deconvoluted spectra are shown in Figure 16. As seen, the biotinylated ODN appears to be free of ON- derived contaminants, and only a small amount of sodium adduction is observed. However, an impurity is observed having a mass of ~16 amu larger than that expected for the biotinylated 25 mer. The deconvolution peak report generated by ProMass software highlights the unexpected mass, and proposes an addition of oxygen to the purified ODN, indicating an oxidation. Since biotin oxidation is possible, this oxygen is proposed to be due to the oxidation of the biotin moiety, possibly from separation at pH 12. Hence, this system shows effective purification of a modified ODN, its subsequent and automated desalting, and successful ESI-MS analysis revealing a possible oxidation.

Biotin oxidation?

Mass Peak List Sorted by Intensity, Dx130: Biotin-CTG CTT GTA GGA TCT TTA AAG ACG A Mass(Da) ± Std. Dev. Intensity Score Delta-Mass %Relative %Total Presumed Identity 8102.1 0.2 4.94E+004 16.10 0.0 100.00 66.45 Target Mass: 8101.5, /5Bio/1-A26 8117.0 0.2 2.03E+004 14.25 14.9 41.07 27.29 8101.5 Target + 15.5 (*O Oxidation) 8139.0 0.2 3.46E+003 10.50 36.9 6.99 4.65 8101.5 Target + 37.5 (*O Oxidation, *Na Adduct) 27313

Figure 16. ESI-MS of desalted Dx-131 (5´-biotin) with raw and de- convoluted spectra, and MS Peak report.

16 Anion-Exchange ESI/MS of Oligonucleotides—Analysis of Oxidation Products of Purified Biotinylated Oligonucleotides Analysis of 2´, 5´-Linkage Isomers in RNA and DNA, and Elucidation of Aberrant Linkage Position

The use of RNA as a therapeutic tool has been demon- Column: DNASwift SAX-1S (5 × 150 mm) Probe Aberrant linkage position(s) strated for both ds- (RNA interference and regulatory Eluents: A. 40 mM Tris.Cl, pH 7 Dio-1 none B. A + 1.25 M NaCl Dio-2 1 microRNA) and ss- (such as aptamer) forms. This has stimu- Gradient: 26–42% B in 16.7 min Dio-3 1, 2 lated great interest and led to numerous efforts to prepare Flow Rate: 1.5 mL/min Dio-4 1, 3 Temperature: 30 °C Dio-5 10 oligoribonucleotides (ORNs) for therapeutic applications. Inj. Volume: 3 µL Dio-6 10, 11 Detection: UV, 260 nm Dio-7 10, 12 Unlike DNA, RNA harbors a 2´ hydroxyl moiety; under Dio-8 5 Sample: eGFP antisense, 1.5 mg/mL conditions used for synthetic nucleic acid release and depro- Dio-9 15 Dio-10 20 tection, this can result in a phosphoryl-migration, producing Dio-11 18, 20 2´, 5´-linkages. These aberrant linkages are known to inhibit Dio-12 19, 20 14.49 certain human nucleases and polymerases, and so agencies 13.19 responsible for regulating RNA therapeutic development Dio-9 13.17 require demonstration of their presence and linkage position. Dio-1 13.08 In addition, some commercial preparations of RNAi Dio-11 12.95 libraries employ RNA and DNA with 2´, 5´-linkages, so Dio-10 methods to confirm their presence are also needed. The Dio-4 12.75 12.63 Dio-12 presence of 2´, 5´-linkages does not change the ionic charac- mA260 Dio-8 12.42 ter, hydrophobicity, or mass of the RNA or DNA. Thus, com- 12.36 Dio-3 11.35 monly employed analyses were not considered adequate by 11.09 Dio-2 the regulatory agencies. However, ONs with 2´, 5´-linkages Dio-5 9.97 have been shown to influence ON solution conformation; so, Dio-7 to the extent that the presence of these linkages impact this Dio-6 conformation, they may also influence their interaction with 0612 18 Minutes stationary phases. eGFP Sequence: 5´-A U G A A C U U C A G G G U C A G C U U G-3´ Here, a DNASwift SAX-1S high-capacity monolith Position: 123 5 10 11 12 15 18 19 20 column is used to to resolve and purify 12 ORN samples with 27314 zero to two 2´, 5´-linkages (Figure 17). Using NaCl as the Figure 17. DNASwift resolution of RNA linkage isomers. 4.5 µg eluent, resolution of a single-sequence RNA with injected, 325–575 mM NaCl in 10 CV. 2´, 5´-linkages to different positions within the sequence, was

17 Analysis of 2´, 5´-Linkage Isomers in RNA and DNA, and Elucidation of Aberrant Linkage Position Column: DNAPac PA200 (2 × 250 mm) Flow Rate: 300 µL/min Column: DNAPac PA200 (4 × 250 mm) Probe Aberrant linkage position(s) Eluents: A. 40 mM Tris.Cl, pH 7 Temperature: 60 °C Eluents: A. 40 mM Tris.Cl, pH 8 Dio-1 none B. A + 1.25 M NaCl Inj. Volume: 25 µL B. A + 0.33 M NaClO4 Dio-2 1 Gradient: 28–52% B in 16.7 min Detection: UV, 260 nm Gradient: 3–36% B in 20 min Dio-3 1, 2 Curve 4 Sample: eGFP antisense RNA, Flow Rate: 1.0 mL/min Dio-4 1, 3 50 µg/mL Temperature: 30 °C Dio-5 10 Inj. Volume: 3 µL Dio-6 10, 11 Asterisk (*) indicates position of 2´,5´-linkage Detection: UV, 260 nm Dio-7 10, 12 Dio-8 5 12.85 Sample: PDAse-II Digests of eGFP Dio-9 15 AUG AAC UUC AGG GUC* AGC UUG NMPs Digestion Products Dio-10 20 12.78 Dio-11 18, 20 AUG AAC UUC AGG GUC AGC UUG (normal) 395 1.44 Dio-12 19, 20

12.67 1.53 1.60 AUG AA*C UUC AGG GUC AGC UUG Dio-1 12.49 AUG AAC UUC AGG GUC AGC* UU*G 1.65 Dio-10 12.38

Dio-12 A*U*G AAC UUC AGG GUC AGC UUG 2.47 Dio-2

12.18 2.93 Dio-11

AUG AAC UUC A*GG GUC AGC UUG mA 4.05 260 3.36

Dio-8 11.66 AUG AAC UUC A*G*G GUC AGC UUG 4.25

Dio-6 11.54 4.47 Dio-5

AUG AAC UUC A*GG* GUC AGC UUG 4.16 4.83 5.55

Dio-9 0510 15 5.36 5.98 Dio-4 Minutes 27315 5.68 6.34 Dio-3 1.53 1.42 1.60 Dio-1:L0 12.84 Figure 18. DNAPac PA200 column purity check of RNA, 21 mers. –5 0 6.0 13.5 Minutes

eGFP Sequence: 5´-A U G A A C U U C A G G G U C A G C U U G-3´ remarkably similar to that obtained on the DNAPac PA200 Position: 12 35 10 11 12 15 18 19 20 column, revealing that several of these linkage isomers are 27316 resolved from one another and also from the ORN lacking the Figure 19. Elution of PDAse-II digestion products. PDAse-II fails to aberrant linkages. cleave at 2´, 5´-linkages leaving partially digested products. While partial resolution of the ORN harboring the aber- rant linkages at one position from those harboring them at position is indicative on the length of the digested fragment. different positions is useful, it does not constitute a definitive This allows a definitive demonstration of the presence of analysis for the presence of these aberrant linkages. these linkages. However, it does not allow elucidation of the However, these linkages are resistant to cleavage by phos- position of these aberrant linkages. phodiesterase-II [Calf-Spleen exonuclease, PDAse-II (E.C.: In order to establish the position of the aberrant 2´, 3.1.16.1)], so digestion with PDAse-II will produce digestion 5´-linkages in ORNs, the following test was repeated. First, products reflective of their position within the sequence. the authors used the DNASwift hybrid monolith column to As seen in Figure 19, DNAPac PA200 column purify ORNs with and without aberrant linkages. Then, their chromatography of PDAse-II-treated, mixed-base RNA 21 purity was verified and they were digested with PDase-II. mers containing one or two aberrant 2´-5´ linkages at differ- The digestion products were then purified and desalted using ent positions in the RNA, results in specific, differentially the automated protocol on an Acclaim PA-II column, and the retained degradation products. The chromatographic elution desalted digests were examined by ESI-MS.

18 Analysis of 2´, 5´-Linkage Isomers in RNA and DNA, and Elucidation of Aberrant Linkage Position DNASwift SAX-1S Purification A. On-Line Desalting Conductivity Column: DNASwift SAX-1S Temperature: 30 °C Trace (5 × 150 mm) Inj. Volume: 85 µL, 1.5 mg/mL (NaCl Eluents: A. 40 mM Tris, pH 7 Detection: UV, 280 nm Washout) B. A + 1.25 M NaCl 1,200 Sample: eGFP antisense 21 mer Gradient: 26–42% B in 16.7 min Purified Flow Rate: 1.5 mL/min Dio-11 Elution A. Dio-11

Flow: mA260 1000 µL/min 1000 µL/min F18 F20 F21 0 F19 0 Minutes 21 300 µL/min

DNAPac PA200 Purity Assessments 100% Column: DNAPac PA200 (2 × 250 mm) Inj. Volume: B. 0.63 µL, 800 µg/mL Eluents: A. 20 mM Tris, pH 7 C. 25 µL, 50 µg/mL Fr - 1 UV Trace B. A + 1.25M NaCl Detection: UV, 260 nm Eluent 4: Gradient: 28–52% B in 12.0 min, curve 4 1.5% Sample: B. Machine-grade Dio-11 Flow Rate: 300 µL/min C. Purified desalted Dio-11 012 Temperature: 60 °C Minutes (eGFP antisense). 1.0 E6 B. ESI-MS of B. Dio-11 C. Dio-11 6712.7 Machine-grade RNA Fraction 20 Purified Desalted Dio-11 Full-length purity: 78% Full-length purity: 97% 8.0 E5

6.0 E5 0 Minutes 16 0 Minutes 16

27317 Intensity 4.0 E5

Figure 20. RNA purification using the DNASwift SAX-1S column 6757.6

2.0 E5 6776.8 The DNASwift purification of one of the ORNs is shown in Figure 20. 0.0 E0 6110 6288 6466 6644 6822 7000 In Figure 21, the desalted, full-length ORN is desalted Mass (d) (top panel) and examined by ESI-MS (bottom panel) to 27318 verify purity. Only the full-length mass and two sodium Figure 21. RNA desalting using the Acclaim PA-II guard column. adduct masses are observed.

19 Analysis of 2´, 5´-Linkage Isomers in RNA and DNA, and Elucidation of Aberrant Linkage Position This sample, amongst several others harboring aberrant from ORNs known to harbor aberrant linkages at specific linkages, was then treated with PDase-II positions in the sequence (Figure 22). The PDase-II digestion products were subsequently These were then desalted using the Acclaim PA2 column purified on a 4× 250 mm DNAPac PA200 column with with reversed-phase eluents instead of ion-pair eluents, and the desalted fragments were submitted to ESI-MS analyses. NaClO4 eluent at pH 7, and several fragments were collected The desalting protocol and an example mass raw spectrum (not deconvoluted) reveal a molecular mass of

Column: DNAPac PA200 (4 × 250 mm) Inj. Volume: 65 µL 1327.2 amu (Figure 23). This is consistent with a base Eluents: A. Deionized Water Detection: UV, 260 nm composition of A U G . Given the ORN sequence, the only B. A + 0.33M NaClO 2 1 1 4 Samples: PDAse-II Digests of eGFP C. 0.2 M Tris.MSA, pH 7 position in which these bases are contiguous comprises the 750 µg/mL Gradient: 0–38% B in 25 min 10% C isocratic Probe Aberrant linkage position(s) four bases at the 5´ end of the sequence. This indicates the Flow Rate: 0.8 mL/min Dio-3 1, 2 aberrant linkage to be at position 1 or at 1 and other positions Temperature: 30 °C Dio-5 10 Dio-6 10, 11 in the first four bases. RNA Fragments from Injection 3´ NMPs Phosphodiesterase digestion Spike

Column: Acclaim PA-II (4.6 × 50 mm) Flow Rate: 500 µL/min Eluents: A. 20 mM Ammonium formate, pH 6 Temperature: 30 °C

B. A + 40% CH3OH Inj. Volume: 25 µL Gradient: 1 % B, hold for 24 s Detection: UV, 260 nm Step to 100% B, hold for 27 s Fr-4 Samples: Purified PDAse-II Step to 1% B, hold for 129 s digests of eGFP RNA (Dio-3) 41 µg/mL Dio-6 mA 260 Dio-3 fragment desalting using Acclaim PA-II 900 Fr-2 Fr-3 Dio-5

Fr-1 Salt (Conductivity trace) mA Dio-3 260 RNA Fragment (Absorbance trace at 260nm) 0 Minutes 12 27319

Figure 22. DNAPac purification of PDAse-II digests of eGFP anti- 0 sense RNA. Collected fraction Gradient trace –400 0123 Minutes

ESI-MS of purified desalted digestion fragment

Fragment = A2U1G1

27320

Figure 23. Oligonucleotide fragment desalting.

20 Analysis of 2´, 5´-Linkage Isomers in RNA and DNA, and Elucidation of Aberrant Linkage Position Table 1 summarizes the data for the samples purified of three to four bases were revealed. Further, cases of two on the DNASwift column, digested with PDase-II; and aberrant linkages separated by a normal linkage include Dio- products purified on the DNAPac PA200, desalted on the 7 and Dio-11 (aberrant linkage at the 3´ end). In each case, Acclaim PAII column, and evaluated by ESI-MS. The table the fragments represent the latest-eluting component and aligns the ESI-MS mass assignments with the sequence and comprised 4-base RNAs. Dio-11 occurs at the 3´-end, so this positions of the 21-base RNA at each aberrant linkage. As fragment terminates with a 3´-hydroxyl group instead of a summarized here, fragments from two to four bases were phosphate. The fragment base compositions allows sequence observed, each with base compositions consistent with the inference from the (known common) parent RNA sequence, 2´-linked base, and one (or two) bases to the 3´ side of the thereby confirming that assignment of the positions of the 2´, aberrant linkage. In cases where a single aberrant linkage 5´-linkages in RNA samples can be deduced from the ESI- was inserted, fragments were either two or three bases long. MS-derived base composition and the sequence (unless the In cases where two aberrant linkages were present, fragments sequence harbors frequent repeats).

Name Fragment Number Sequence and Position of Digest Fragment Fragment ID Mass

Dio1: none 5′-AUG AAC UUC AGG GUC AGC UUG -3′ None (only NMPs) – Dio3: Fr 1 5′-A*U*G AAC UUC AGG GUC AGC UUG -3′ ApUpGpAp, 1327.2 Dio5: Fr 2 5′-AUG AAC UUC A*GG GUC AGC UUG -3′ ApGp 691.2 Dio5: Fr 3 5′-AUG AAC UUC A*GG GUC AGC UUG -3′ ApGpGp 1037.2 Dio6: Fr 4 5′-AUG AAC UUC A*G*G GUC AGC UUG -3′ ApGpGp 1037.2 Dio7: Fr 5 5′-AUG AAC UUC A*GG* GUC AGC UUG -3′ ApGpGpGp 1382.2 Dio8: Fr 6 5′-AUG AA*C UUC AGG GUC AGC UUG -3′ ApCpUp 958.1 Dio9: Fr 7 5′-AUG AAC UUC AGG GUC* AGC UUG -3′ CpAp 652.1 Dio11: Fr 8 5′-AUG AAC UUC AGG GUC AGC* UU*G -3′ CpUpUpG-OH 1200.2 Table 1. Identification of phosphodiesterase-II RNA fragments (and resulting mass) based on known sequence. Elution on DNAPac PA200 col- umn. eGFP antisense strand: 5′-AUG AAC UUC AGG GUC AGC UUG-3′. Modifications:Underlined residues have 2’-5’ linkages. Emboldened bases indicate identified fragments.

21 Analysis of 2´, 5´-Linkage Isomers in RNA and DNA, and Elucidation of Aberrant Linkage Position Analysis of Nucleoside Mono-, Di-, and Triphosphates

The analysis of the nucleoside triphosphates in PCR Column: DNAPac PA200 (4 × 250 mm) Elution Order: reaction mixtures allows verification that the dNTPs (or Eluents: A. Deionized water 1. dCMP 13. dCTP NTPs for RNA amplification protocols) have not degraded. B. 0.1 M NaOH 2. CMP 14. CTP D. 0.25 M NaCl 3. dAMP 15. dATP Analysis of amplification cocktails after amplification can Gradient: 12–60% B in 29 min, 4. AMP 16. dTDP Isocratic 10% B 5. dCDP 17. ATP reveal which component may have limited the amplification, Flow Rate: 1.4 mL/min 6. CDP 18. UDP and also reveal possible product inhibition. Here, a linear Temperature: 23–27 °C, as indicated 7. dTMP 19. dGDP Inj. Volume: 10 µL, 37 µg/mL 8. dADP 20. GDP gradient is used to elute all 24 dNXPs and NXPs found in Detection: UV, 260 nm 9. UMP 21. dTTP 10. ADP Sample: Nucleoside mono-, di-, and 22. UTP RNA and DNA amplification mixes at 23–27 °C. 11. dGMP triphosphates as DNA and RNA 23. dGTP The optimum temperatures for RNA and DNA compo- 12. GMP 24. GTP

195 nents with this gradient were observed to be different: dAMP 27° C

CMP 23 °C for DNA components, and 27 °C for RNA compo- AMP dTMP

nents. Minor changes to tubing length, mixer design, and

dCMP CDP

P

ADP dATP dADP dADP

AT column oven may contribute to somewhat different results

GDP

dGMP dGMP

dGTP dCDP dCTP UMP GTP CTP dGDP dTDP UDP dTTP for different chromatography systems. UTP

mA260 25° C dAMP

CMP AMP

dADP 23° C

dTMP

CDP dCMP UDP dATP

P

ADP

AT GDP

dGMP

dCDP dGTP

dCTP UMP GTP dGDP CTP dTTP dTDP UTP GMP –2 0102030 Minutes 27321

Figure 24. Analysis of nucleoside phosphates: effect of temperature using the DNAPac PA200 column.

22 Analysis of Nucleoside Mono-, Di- and Triphosphates Method Development Using Anion-Exchange Chromatography for Nucleic Acids

This section demonstrates method development for allows users with sequence or composition information to nucleic acid separations through nucleic acid interactions predict eluent pH values likely to produce resolution between with anion-exchange stationary phases. ONs eluting near one another at neutral pH. The development of ONs to be used for pharmaceutical While the effects of pH are qualitatively similar in these applications require critical analyses. This analyis is sup- different eluents, perchlorate-based systems elute ONs at ported by Dionex anion-exchange chromatography columns much lower concentrations, and use lower gradient slopes that were designed to employ ON chemistry for selectivity (5 mM/mL vs. 15 mM/mL for NaCl) without significant optimization. loss of resolution. Perchlorate is also a chaotropic salt, so it In Figure 25, a linear salt gradient was used to elute a influences hydration of both the ON and stationary phase. mixed-base 25 mer sequence. The top panel is with NaCl This tends to minimize the effect of hydrophobic interactions eluent and the bottom panel uses NaClO4 as the eluent salt. between the nucleobases and the phase. Therefore, the use of In both panels, increasing eluent pH produces a pronounced NaClO4 tends to encourage retention primarily on net charge. increase in retention of the ON. This is due to the ioniza- Similarly, reducing the pH will minimize the charge on the tion of the tautomeric oxygen on G and T bases (T shown as nucleobases. Thus, reducing the pH with NaClO4 eluent will inset). This ionization confers a formal increase in negative tend to encourage retention bases primarily on ON length. charge for each T and G as the pH increases between 9 and For DNAPac and DNASwift columns, pH represents a more 11 (this does not occur if the ON contains only C and A). powerful influence for control of retention, while salt form Since different ONs often contain different proportions of G exerts a smaller influence. and T, their pH-dependent retention increases will differ. This

23 Method Development using Anion-Exchange Chromatography for Nucleic Acids Column: DNAPac PA200 (4 × 250 mm) Column: DNAPac PA200 (4 × 250 mm) Flow Rate: 1.2 mL/min (both) O O– Eluents: A. 40 mM buffer, pH as indicated Eluents: A. 40 mM buffer, pH as indicated Temperature: 25 °C B. 1.25 M NaCl in eluent A CH CH (± 20% CH CN) Inj. Volume: 12 µL, 40 µg/mL HN 3 N 3 3 or 0.33M NaClO4 in eluent A B. 1.25 M NaCl in eluent A Detection: UV, 260 nm Gradient: Top: 26–72% B in 32 min O N HO N Gradient: Top: 26–61% B in 24 min Sample: Dx 79, sequence as indicated Bottom: 21–56% in 19 min R R Flow Rate: 1.2 mL/min (both) pH 7 pH 11 25 mer sequence: CTG AAT GTA GGT TCT CTA ACG CTG A Temperature: 25 °C Inj. Volume: 14 µL, 40 µg/mL Flow: 1.20 mL/min 20% CH CN 13.80 Detection: UV, 260 nm 3 pH 12 9.93 14.35 pH 11 Sample: Dx 79, Sequence as indicated 6.02 mA260 25 mer sequence: CTG AAT GTA GGT TCT CTA ACG CTG A 5.42 pH 10 pH 9 4.61 pH 8 A. Flow: 1.20 mL/min, 15 mM/mL NaCl pH 6.5 pH 12 0 Minutes 20 pH 11 mA260 pH 10 Flow: 1.20 mL/min 22.74 No CH CN pH 9 3 22.57 pH 12 pH 8 15.04 pH 11 mA 7.41 01020 260 pH 10 Minutes pH 9 6.88 pH 8 6.15 pH 6.5 B. Flow: 1.20 mL/min, 5 mM/mL NaClO 4 0 10 20 30 Minutes pH 12 27323 mA260 pH 11 pH 10 pH 9 Figure 26. Influence of solvent on DNAPac PA200 retention: NaCl pH 8 eluents. 01020 Minutes 27322 Oligonucleotides may be modified with chemical deriva- Figure 25. Effect of eluent salts on pH-induced retention using the DNAPac PA200 column and showing the effects of pH and salt form tives for use as capture probes or as specific detection on single-stranded nucleic acid retention. beacons. These are often very hydrophobic, and can introduce additional hydrophobic interactions that influence chromatographic peak shape. These need to be well Some ONs used for critical applications harbor more controlled for critical analyses. hydrophobic characteristics. These require a more specific To demonstrate the utility of salt form for control of approach for critical analyses. Like NaClO , the addition of 4 peak shape, a 25 nt sample without and with one or more solvent to the eluent will tend to reduce interactions between common derivatives was analyzed on a DNAPac PA200 the nucleobases and the stationary phase. column with NaCl and NaClO eluents (Figure 27). In Figure 26, pH-dependent retention in NaCl eluent on 4 As seen in the top panel of Figure 27, NaCl eluent were the PA200 column is compared in the presence and absence used without solvent to separate these very hydrophobic of 20% CH CN. With this solvent, pH-dependent retention 3 samples. The peak width values (at half-height [PW½], in persists, but overall retention is significantly reduced. The minutes) reveal PW½ values of 0.275 ±0.07 min. However, effect of solvent also impacts pH-dependent retention, in that the same samples separated with NaClO -based eluents pro- ON elution time increases slightly when the pH shifts from 4 duced PW½ values of 0.064 ± 0.007 min. In this case, use of 11 to 12 without solvent, but decreases with the programmed NaClO eluent reduced the peak width parameter to < 25% pH shift in 20% solvent. This effect is not observed when 4 of that observed using NaCl eluents. perchlorate eluents are used without solvent. While use of such hydrophobic derivatives is not re- An important use of high pH eluents is control of quired for many ON assays, and often hydrophobic deriva- Watson-Crick or other hydrogen-bonding interactions. For tives have no effect on ON peak shape, Figure 27 illustrates most ONs, these bonds are overcome at pH values above a simple method to control peak shape when hydrophobic 10.5 or 11, and eventually GC or poly-G hydrogen bonds are derivatives must be used. theoretically denatured at pH 12.4. Hence, use of high pH is also a useful method for controlling unwanted inter- and intrastrand interactions during chromatography.

24 Method Development using Anion-Exchange Chromatography for Nucleic Acids Panel A Panel B Panel A Panel B Column: DNAPac PA200 (4 × 250 mm) Column: DNAPac PA200 (4 × 250 mm) Column: DNAPac PA200 (4 × 250 mm) Column: DNAPac PA200 (4 × 250 mm) Eluents: A. 20 mM Tris, pH 8 Eluents: A. 20 mM Tris, pH 8 Eluents: A. 20 mM Tris, pH 8, no CH3CN Eluents: A. 20 mM Tris, pH 8, 20% CH3CN B. A + 1.25 M NaCl B. A + 0.33 M NaClO 4 B. A + 1.25 M NaCl, no CH3CN B. A + 1.25 M NaCl, 20% CH3CN Gradient: 27–59% B in 18.5 min Gradient: 24.2–48.5% B in 13.3 min Gradient: 27–59% B in 18.5 min Gradient: 26–56% B in 17.4 min Flow Rate: 1.2 mL/min Flow Rate: 1.2 mL/min Flow Rate: 1.2 mL/min Flow Rate: 1.2 mL/min Temperature: 35 °C Temperature: 35 °C Temperature: 35 °C Temperature: 35 °C Inj. Volume:: 3–5 µL, ~50 µg/mL Inj. Volume: 5 µL, 50 µg/mL Inj. Volume: 3–5 µL, ~50 µg/mL Inj. Volume: 5 µL, 50 µg/mL Detection: UV, 260 nm Detection: UV, 260 nm Detection: UV, 260 nm Detection: UV, 260 nm Sample: 25 mer ± derivative Sample: 25 mer ± derivative Sample: 25 mer ± derivative Sample: 25 mer ± derivative

Peak labels indicate peak width (half height, min) 0.356 Peak labels indicate peak width (half height, min) A 0.346 A. 20% CH CN 3 0.073 Dx131 (78+5' TET) 0.055 0.068 Dx133 (78+5' TET,+3' TAMRA) 0.057

mA 0.059 260 0.240 0.233 0.200 Dx132 (78+5’Flr) Dx131 (78+5' TET) Dx130 (78+5'Biotin) Dx133 (78+5' TET,+3' TAMRA) Dx78 Dx132 (78+5’ Flr) Dx130 (78+5' Biotin) 0816 Minutes Dx78 mA260 0.069 0.073 B. No Solvent 0.356 B 0.062 Dx78-5'TET,+3'TAMRA Dx131 (78+5' TET) 0.346 0.057 Dx133 (78+5' TET,+3' TAMRA) mA Dx78-5'TET 260 0.200 0.060 Dx132 (78+5’Flr) 0.233 Dx78-5’Flr 0.240 Dx130 (78+5'Biotin) Dx78-5'Biotin Dx78 Dx78

0510 15 08 16 Minutes Minutes 27324 27325 Figure 27. Effect of salt form on hydrophobic DNA peak width: NaCl Figure 28. Effect of solvent derivatized on peak width. vs. NaClO4.

As in the previous example, ONs modified with very The peak width values (at half-height, [PW½] in minutes) hydrophobic chemical derivatives for use as capture probes reveal PW½ values of 0.275 ±0.07 min. However, the same or as specific detection beacons introduce additional samples separated with 20% CH3CN produced PW½ values hydrophobic interactions that influence chromatographic of 0.062 ± 0.008 min. In this case, use of solvent eluent peak shape. As a second example of peak shape control, a reduced the peak width parameter to < 25% of that observed 25 nt sample without and with one or more common deriva- using NaCl eluents. tives was analyzed on a DNAPac PA200 column with While use of such hydrophobic derivatives is not re- NaCl eluents with and without 20% acetonitrile as solvent quired for many ON assays, and often hydrophobic deriva- (Figure 28). tives have no effect on ON peak shape, Figure 28 illustrates In the top panel of this example, eluents without sol- a second method to control peak shape when hydrophobic vent were used to separate these very hydrophobic samples. derivatives must be used.

25 Method Development using Anion-Exchange Chromatography for Nucleic Acids Column: DNASwift SAX-1S (5 × 150 mm) Order of Elution: Column: DNASwift SAX-1S (5 × 150 mm) Temperature: As indicated Eluents: A. 40 mM Tris, pH 7 Cal610 Eluents: A. 30 mM Tris to pH 8 with MSA Inj. Volume: 16 µL B. 1.25 M NaCl in eluent A Parent sequence (unlabeled) B. A + 1.25 M NaCl Detection: UV, 260 nm Gradient: 16–80% B in 15 min Biotin Gradient: 8–64% B in 15 min Sample: AR25 (20 mer) 0.5 mg/mL Flow Rate: 1.5 mL/min Texas Red Flow Rate: 1.77 mL/min (Top) (Sequence as shown) Inj. Volume: 8 µL FAM Detection: UV, 260 nm FAM + Iowa Black Sequence: 5´-GGG ATG CAG ATC ACT TTC CG-3’ Joe Sample: Derivatized oligonucleotides, TET 0.4 mg/mL 173 µL

HEX 169 µL 159 µL Dx-78 Sequence: 5´CTG CTT GTA GGA TCT TTA AAG ACG T3´ 164 µL 162 µL 30

mA260 13.32 Dx-78-HEX 10.35 11.66 30 °C Dx-78-TET 10.11 10.52 40 °C 50 °C

Dx-78-JOE 9.79 60 °C 70 °C Dx-78-FAM + Iowa Black quencher 9.21 mA 260 0612 18 Dx-78-Flr (FAM Mixed isomers) Minutes 27327 7.70 8.05 Dx-78-Texas Red Figure 30. Effect of temperature on DNASwift PW½ and resolution.

Dx-78-Biotin Dx-78- Unlabeled parent sequence 7.40 In Figure 30, a 20-base ON was examined on a DNAS- wift SAX-1S hybrid monolith column at different tempera- Dx-78-CAL610 7.11 tures. Adequate separation is observed at 30 °C, but increas- 0 0816 Minutes ing the temperature increases retention and reduces peak 27326 width (improving resolution), especially of the components Figure 29. DNASwift oligonucleotide probes: fluorophore series. eluting near the full-length AR25 20 mer. This illustrates the advantages of high-temperature ON chromatography. Another advantage is that operation above the melting tem- Like the DNAPac columns, the DNASwift hybrid perature of the ON eliminates possible intra- and interstrand monolith column also resolves ONs harboring hydropho- hydrogen bonding. bic probe structures from one another, and from the parent Since use of low pH, NaClO4, and solvent all tend to sequence lacking the derivatives. In Figure 29, 3 µg samples reduce the influence of nucleobases on retention, combining were separated using a steep (survey) gradient on NaCl at these three conditions should tend to promote elution primar- pH 7. Each of the ONs bearing the derivatives are resolved ily in order of ON length, and this is a common goal during from one another and all, except the ONs with Cal610 or analysis. biotin, are resolved from the parent sequence without modi- fications. As demonstrated, when using DNAPac columns, a less steep gradient will better resolve the ON derivatives, and addition of a small amount of solvent, use of NaClO4, or even use of higher temperature with a column oven, will improve peak shape.

26 Method Development using Anion-Exchange Chromatography for Nucleic Acids To demonstrate this effect, 21 ONs with alterations at Since the increase in pH increases ON retention, and 5′ and 3′ ends, and containing 21–25 bases were separated since the retention increases depend on the number of T and on a DNAPac PA200 column at pH 6.5, using NaClO4 G bases. Changes in pH can also influence retention order. In eluents with 20% CH3CN. The earliest- and latest-eluting the cases where two closely related ONs must be resolved, components for each ON length were overlaid together on for example n and n-1 sequences, changes in pH can allow this set of chromatograms. The components of each length control of selectivity. are at least partially resolved from the components of every In Figure 32, the effect of pH on selectivity with two adjacent length, even though they all have nearly identi- 23-base ODNs (sequences provided) in NaCl, and NaClO4 cal sequence. This demonstrates that ONs with related base eluent systems is demonstrated. Both ONs exhibit dramati- sequences are resolved from one another primarily on the cally increased retention with increasing pH. This is due to number of bases under these conditions. the ionization of the tautomeric oxygen on each G and T as the pH increases (T shown). These two ONs also exhibit pH- dependent elution order reversals in each salt. With NaCl, the

Column: DNAPac PA200 (4 × 250 mm) Temperature: 30 °C elution order reverses between pH 9 and 10. With NaClO4, Eluents: A. 20 mM phosphate, pH 6.5, Inj. Volume: 5 µL the reversal is between pH 9 and 11 (coelution at pH 10). with 20% CH3CN Detection: UV, 260 nm B. 0.33 M NaClO in eluent A 4 Sample: Oligonucleotides of The extent of the retention difference between the ONs is Gradient: 21–43% B in 12 min related sequence, 40 µg/mL Flow Rate: 1.2 mL/min more dramatic in NaCl than in NaClO4. This derives from Elution Order: By length 21 the chaotropic nature of NaClO , that suppresses hydropho- 22 4 Oligonucleotide length: bic interactions, contributing to the separation in NaCl. 23

Dx98 24 Dx96

Dx94 25 Dx90 Dx91 Dx88 Oligonucleotide 23 mer sequences: Dx87 1. (Dx88): GA TTG TAG GTT CTC TAA CGC TGA Dx80 mA Dx86 260 Dx84 2. (Dx89): TGA TTG TAG GTT CTC TAA CGC TG

A. Flow: 1.20 mL/min, 15mM NaCl/mL gradient

2: pH 11 1: pH 11 mA260 2: pH 10 1: pH 10 2: pH 9 1: pH 9 2: pH 8 1: pH 8 34567 Minutes 27328 B. Flow: 1.20 mL/min, 5 mM NaClO4/mL gradient Figure 31. Oligonucleotide elution based primarily on length using 2: pH 11 the DNAPac PA200 column. Excerpted from Analytical Biochemis- 1: pH 11 try 2005, 338, 39–47. 2: pH 10 Ionization of Thymine – mA260 1: pH 10 O O CH CH 2: pH 9 HN 3 N 3 1: pH 9 O N HO N As mentioned earlier, often, unwanted hydrogen-bond- 2: pH 8 R R ing within- or between-oligonucleotide strands can interfere 1: pH 8 pH 7 pH 11 0 10 20 30 with chromatographic separations regardless of the separa- Minutes tion mode. With anion-exchange chromatography, high pH 27329 or high temperature can be used. In this case where the pH Figure 32. Effect of salt form on pH-dependent selectivity. must remain low, unwanted hydrogen bonding interactions can be controlled either by increased temperature, or by the addition of chaotropic agents, such as 6 M urea or 25% formamide. These agents increase the eluent viscosity, and thus the operating pressures, so their use requires some care to ensure that rapid increases in flow (and thus very high pressure) do not damage the instrument.

27 Method Development using Anion-Exchange Chromatography for Nucleic Acids IP-RPLC does elute ONs based on their size, but does sodium salts are used which are known to influence pH mea- not support facile control of selectivity, and does not resolve surements; also, relatively high salt concentrations are used, many ONs having identical mass, such as RNA linkage usually between 50 mM and 1 M. To examine the effect of isomers, or diastereoisomeric ONs, even when coupled to sodium ion on pH measurement, the pH measurements were single-stage MS. repeated using salt concentrations of 100, 400, and 800 mM. Because ON retention is profoundly influenced by pH It was found that the pH vs. eluent 3 to eluent 4 proportion is between 6 and 12.4, the authors have developed a method essentially identical under each condition. Figure 33 illus- to quickly and reliably deliver eluents at pH values in that trates a standard curve prepared with NaCl range. While it is possible to employ eluent selection valves This figure also provides the Excel-derived sixth-order to deliver premade eluents at various pH values, these are polynomial equation, which is then employed to calculate cumbersome to execute as each eluent must be separately the proportions of eluents 2 and 3 to deliver a desired eluent prepared and multiple valve systems must be employed. pH. To verify that the desired pH is correctly delivered, we The method developed is a programmable system with employ a calibrated, in-line pH electrode. Using this system quaternary eluent proportioning. This allows use of one pair one can program a desired pH with eluents 2 and 3, and of eluents for pH control, and the other pair of eluents to use eluents 1 and 4 to prepare simple or complex multistep provide a salt gradient. Usually, the gradient eluents are 1) gradients. deionized water, and 4) a salt solution (e.g., 1.25 M NaCl or

0.33 M NaClO4). Eluent 2 and 3 are used for pH program- ming. Eluent 2 is 0.2 M NaOH, and eluent 3 is TAD buffer, a Sixth order polynomial equation, x = desired pH combination of Tris-base, 2-amino-2-methyl-1-propanol, and % [E2] = 0.0317x6 - 1.8247x5 + 43.524x4 - 550.61x3 + 3894.1x2 - 14584x + 22572 % [E3] = 20 - % [E2] di-isopropylamine, each at 0.2 M, and adjusted with meth- ane sulfonic acid (MSA) to pH 7.2. These buffers harbor pK pH vs. Eluent Proportion [E2] (Duplicate Assays) 20 values ranging from pH 8 to 11. While HCl is typically used 18 to adjust pH for eluents, it was found that HCl-adjustment 16 of this buffer set tends to promote microbial growth, while 14 12 MSA tends to inhibit it. Proportioning between eluents 2 %E2 and 3 allows delivery of a target pH. Generally, a fixed 10 8 fraction of the total delivered eluent (e.g., 20%) is used to 6 generate the desired pH, but this system also allows pH 4 gradient formation. 2 An advantage of this system is that the salt form can 0 678910 11 12 13 be quickly changed by preparing a single new eluent (e.g., Resulting pH 0.33 M LiClO ). To determine the proportion of each of 4 Calculation of Eluent Proportions the pH-generating eluents, the authors prepared a standard Desired % E2 % E3 Desired % E2 % E3 curve from the pH delivered by each proportion (0–20%) of pH [NaOH] [Buffer] pH [NaOH] [Buffer] 7.00 0.1 19.9 9.75 11.4 8.6 the two eluents. The pH is monitored with a calibrated pH 7.25 0.8 19.2 10.00 11.9 8.1 7.50 2.2 17.8 10.25 12.5 7.5 detector placed in the eluent line after the column and UV 7.75 3.8 16.2 10.50 13.0 7.0 detector. Since three buffers are used, the relationship of 8.00 5.3 14.7 10.75 13.5 6.5 8.25 6.7 13.3 11.00 13.9 6.1 proportion to pH is not linear. Each buffer contributes a dis- 8.50 7.9 12.1 11.25 14.3 5.7 8.75 8.8 11.2 11.50 14.8 5.2 tinct pH curve, and these are designed to overlap. Our pH vs. 9.00 9.6 10.4 11.75 15.7 4.3 relative proportion data is converted into the standard curve 9.25 10.2 9.8 12.00 17.3 2.7 9.50 10.8 9.2 using a sixth-order polynomial regression using Microsoft 27330 Excel, and the data with the regression line printed to use for day-to-day experiments. In most of the authors' work, Figure 33. Dial-a-pH standard curve: pH vs. eluent proportion using 0.2 M NaOH vs. 0.2 M TAD. TAD = 0.2 M Tris, 0.2 M, 2-amino- 2-methyl-1-propanol, 0.2 M di-isopropylamine, pH 7.2.

28 Method Development using Anion-Exchange Chromatography for Nucleic Acids Application notes Nucleic acid therapeutics applications notebook Application Note 100

High-Resolution Analysis and Purification of Oligonucleotides with the DNAPac™ PA-100 Column

INTRODUCTION REAGENTS AND STANDARDS The DNAPac™ PA-100 anion-exchange column is Deionized water, 17.8 MΩ-cm resistance or better specifically engineered to provide unit-base resolution of Sodium hydroxide solution, 50% w/w, low carbonate synthetic oligonucleotides to 60 bases and beyond. This Tris(hydroxymethyl)aminomethane (molecular column is packed with a pellicular anion-exchange resin biology grade) in which quaternary amine-functionalized microbeads are Sodium chloride bound to a 13-µm diameter nonporous polymeric substrate Sodium perchlorate monohydrate (see Figure 34). The rapid mass transport characteristics of Disodium ethylenediaminetetraacetate (EDTA), this resin result in higher resolution oligonucleotide dihydrate (molecular biology grade) separations than are possible with traditional macroporous pd(A) (Pharmacia Biotech Inc., Piscataway, resins or reversed-phase columns. Additionally, the 12–18, DNAPac PA-100 can be operated routinely under strongly New Jersey, USA)

denaturing conditions, including high temperature or high pd(A)40–60, (Pharmacia Biotech Inc.)

pH. Anion-exchange separations under denaturing conditions pd(G)12–18, (Pharmacia Biotech Inc.) are particularly useful for the resolution of oligonucleo- -20 sequencing primer, crude 17-mer, tides with high guanine (G) content or with regions of (Genosys Biotechnologies, Inc. Woodlands, complementary sequence. Texas, USA) In this application note, methods for the separation of various synthetic oligonucleotides on the 4-mm CONDITIONS diameter DNAPac PA-100 analytical column are presented. Columns: DNAPac PA-100 analytical, 4 × 250 mm Included are pH 8.0 and pH 12.4 gradient programs. Also DNAPac PA-100 guard, 4 × 50 mm presented are strategies for the scale up of analytical separations to semipreparative purifications using a DNAPac PA-100 semipreparative, 22 x 250 mm 22-mm diameter column. Eluent A: Deionized water EQUIPMENT Eluent B: 0.20 M NaOH Dionex DX 500 HPLC system consisting of: Eluent C: 0.25 M Tris-Cl, pH 8.0 GP40 Gradient Pump Eluent D: Option 1: 1.0 M NaCl AD20 UV/Visible Absorbance Detector Option 2: 2.0 M NaCl

LC20 Chromatography Enclosure Option 3: 0.375 M NaClO4 Dionex PeakNet Chromatography Workstation Flow Rate: 1.5 mL/min, except where noted Inj. Volume: 200 µL Detection: UV, 260 nm

30 High-Resolution Analysis and Purification of Oligonucleotides with the DNAPac PA100Application Column Note 100 1 PREPARATION OF SOLUTIONS AND REAGENTS

Eluent A: Deionized H2O Vacuum degas 1 L of deionized water. 100 nm diameter MicroBeads

Eluent B: 0.20 M NaOH µm Dilute 10.4 mL of low carbonate 50% w/w sodium 6.5 hydroxide to 1 L with degassed, deionized water. Pellets of NaOH are coated with a layer of sodium carbonate and are not suitable for eluent preparation.

Eluent C: 0.25 M Tris-Cl, pH 8.0 Dissolve 30.28 g of tris(hydroxymethyl)amino- methane in 800 mL of deionized water. Adjust the 11710 pH to 8.0 by the addition of approximately 10 mL Figure 34. Pellicular structure of the DNAPac PA-100 resin. of concentrated HCl. Add deionized water to a final volume of 1 L and vacuum degas the solution. SAMPLE PREPARATION

pd(A)12–18, pd(A)40–60, and pd(G)12–18 Eluent D, Option 1: 1.0 M NaCl Each lyophilized DNA homopolymer sample Dissolve 58.45 g of sodium chloride in 800 mL of was dissolved in TE buffer to produce a 1-mg/mL deionized water. Add deionized water to a final volume of solution. Before injection, DNA was diluted to the 1 L and vacuum degas the solution. appropriate concentration with deionized water. The “p” in the chemical symbol indicates that the oligomers are phosphorylated. Eluent D, Option 2: 2.0 M NaCl Dissolve 116.9 g of sodium chloride in 800 mL of deionized water. Add deionized water to a final volume of -20 Sequencing Primer 1 L and vacuum degas the solution. The lyophilized, unpurified synthetic oligonucleotide was suspended in TE buffer. To remove particulate matter, the sample was centrifuged at 12,000 x g for 5 minutes and Eluent D, Option 3: 0.375 M NaClO 4 the supernatant was transferred to a fresh tube. The DNA Dissolve 52.67 g of sodium perchlorate monohydrate concentration was determined by measuring the absorbance in 800 mL of deionized water. Add deionized water to a at 260 nm and applying the manufacturer’s suggested final volume of 1 L and vacuum degas the solution. conversion factor of 1.0 A260 = 30 µg/mL. The oligonucleo- tide was diluted with deionized water to the indicated 0.5 M EDTA, pH 8.0 concentration before injection. Add 93.1 g of disodium ethylenediaminetetraacetate dihydrate to 400 mL of deionized water. While stirring, RESULTS AND DISCUSSION add NaOH (approximately 10 g of pellets) until the EDTA High-Resolution Separations at pH 8.0 is completely dissolved and the pH is stable at 8.0. Add The high resolving power of the DNAPac PA-100 deionized water to a final volume of 500 mL and autoclave can be demonstrated by the analysis of synthetic DNA to sterilize. homopolymers. Typical results for the analysis of 1 µg

of pd(A)12–18 are shown in Figure 35. A gradient of 100 TE Buffer, pH 8.0 to 450 mM NaCl in the presence of 25 mM Tris-Cl 10 mM Tris-Cl, pH 8.0 pH 8.0 was used to elute the pd(A) oligonucleotides. 1 mM EDTA, pH 8.0

31 High-Resolution Analysis and Purification of Oligonucleotides with the DNAPac PA100 Column Time (min) %A %C %D Curve Time (min) %A %C %D Curve 0.00 80 10 10 5 0.00 85 10 5 5 20.00 45 10 45 4 Peaks: 20.00 64 10 26 4 Peaks: 0.018 6 1. pd(A) 0.018 6 1. pd(A) 5 12 5 12 2. pd(A) 3 2. pd(A) 3 13 13 3. pd(A)14 4 3. pd(A)14 4 4. pd(A)15 4. pd(A)15 5. pd(A) 1 7 5. pd(A) 1 7 16 16 AU AU 2 2 6. pd(A)17 6. pd(A)17 7. pd(A)18 7. pd(A)18

0.00 0.00

12 14 16 18 20 10 12 14 16 18 20 Minutes Minutes 10952 10953

Figure 35. Separation of 1 µg of pd(A)12–18 at pH 8.0 with a Figure 36. Separation of 1 µg of pd(A)12–18 at pH 8.0 with a 19 100 to 450 mM NaCl gradient. Small peaks represent dephos- to 98 mM NaClO4 gradient. Slightly different selectivity relative phorylated impurites (see text). Eluent D, Option 1 was used: to NaCl elution is observed (see Figure 35). Eluent D, Option 3

1.0 M NaCl. was used: 0.375 M NaClO4.

The sample was injected at a relatively low NaCl concentration of 100 mM to ensure efficient binding Time (min) %A %C %D Curve 0.00 80 10 10 5 of the oligonucleotides to the column. The pd(A) 2.00 49 10 41 5 homopolymer mixture was resolved into seven main 25.00 39 10 51 3 0.004 peaks, which represent the seven phosphorylated oligonucleotides described by the manufacturer. Also

detected were seven minor peaks, which coeluted with AU

the d(A)12–18 oligonucleotides produced by dephosphoryla-

tion of the pd(A)12–18 sample with alkaline phosphatase (data not shown). -0.001

DNA can be eluted from the DNAPac PA-100 510152025 by a variety of anions other than chloride, including Minutes 10954 1 2 1 acetate, bromide, and perchlorate. The use of NaClO4 1 Figure 37. Separation of 1.5 g of pd(A) at pH 8.0 with a as an eluent has been thoroughly investigated. Perchlo- µ 40-60 shallow 410 to 510 mM NaCl gradient. The sample was injected rate is particularly useful for the analysis of phospho- at 100 mM NaCl to ensure efficient binding of the DNA to the rothioate DNA, in which sulfur has been substituted column. Eluent D, Option 1 was used: 1.0 M NaCl. for non-bridging oxygen on the phosphate backbone. Since this sulfur substitution results in an increased oligonucleotides are longer and possess a stronger negative affinity of DNA for the DNAPac resin, conventional salt charge than the pd(A)12–18 species, higher NaCl concentra- gradients such as 0–2 M sodium chloride often will not tions are needed for elution. After sample injection at 1,2 elute phosphorothioates. For the pd(A)12–18 sample, 100 mM of NaCl, the NaCl concentration was stepped

gradient elution with NaClO4 rather than NaCl produced to 410 mM over 2 minutes. A shallow (410 to 510 mM) similar chromatographic results (see Figure 36). Relative NaCl gradient over 23 minutes was used for elution of the

to NaCl, much lower concentrations of NaClO4 are DNA. This gradient strategy couples the high resolving needed for DNA elution. power of shallow gradients with short run times. A somewhat different NaCl gradient was used for

analysis of a pd(A)40-60 oligonucleotide mixture (see Figure 37). In addition to the 21 expected peaks, at least one additional minor peak was detected. Since these

32 High-Resolution Analysis and Purification of Oligonucleotides with the DNAPac PA100 Column High pH Separations Time (min) %A %C %D Curve Self-complementary sequences or poly-G stretches 0.00 80 10 10 5 can result in intra- and intermolecular associations of 20.00 0 10 90 3 0.008 oligonucleotides. These associations can prevent efficient oligonucleotide separation under non-denaturing conditions.

The DNAPac PA-100 can be operated at high temperature AU (to 90 °C) or at high pH (to pH 12.4) for the separation of samples that contain such sequences. As an example, the 0.000 attempted separation of pd(G)12–18 at room temperature with a 100 to 900 mM NaCl gradient at pH 8.0 is shown in Figure 38. The poly-G sequence composition causes the 02468 10 12 14 16 18 20 Minutes sample to elute as one broad, diffused band. At pH 12.4, 10955 hydrogen bonding is abolished. As a consequence, the Figure 38. Attempted separation of 3 µg of pd(G)12–18 with a seven primary pdG homopolymers are resolved at pH 12.4, NaCl gradient at pH 8.0. Self-complementary sequences or as are a number of minor contaminating species (see Fig- poly-G regions in DNA can prevent oligonucleotide separation under non-denaturing conditions. Eluent D, Option 1 was used: ure 39). Relative to pH 8.0, an additional negative charge is 1.0 M NaCl. present on the bases G and T at pH 12.4. This extra charge provides additional separation selectivity because it causes oligonucleotides with high levels of G and T to elute later Time (min) %A %B %D Curve Peaks: 1. pd(G)12 than sequences low in G and T. 0.00 82.5 12.5 5 5 2. pd(G)13 2.00 62.5 12.5 25 5 3. pd(G)14 0.010 20.00 42.5 12.5 45 4 1 4. pd(G) 3 5 6 7 15 Analytical to Semipreparative Scale Up 4 5. pd(G) 2 16 DNAPac PA-100 columns are available in diameters 6. pd(G)17 7. pd(G) of 9 mm and 22 mm for semipreparative applications. AU 18 Analytical separations on the 4-mm diameter column can be scaled directly to the larger columns, so that methods development for semipreparative applications 0.000 can be performed rapidly on the analytical column with 02468101214161820 small amounts of sample. For direct scale up of analytical Minutes gradient chromatography, a constant column volume 10956 scaling rule should be observed: The number of column- Figure 39. Resolution of pd(G)12–18 homopolymers at pH 12.4 volumes of eluent delivered over the duration of the with a 500 to 900 mM NaCl gradient. The sample was injected at 100 mM of NaCl. At high pH, hydrogen bonding between gradient should remain constant. To demonstrate scale poly-dG sequences is eliminated. Eluent D, Option 2 was used: up of a method from the 4-mm diameter to the 22-mm 2.0 M NaCl. diameter column, an analytical separation was developed for the “-20 primer,” a crude synthetic 17-mer with the The analytical method was modified in the following three sequence 5'-GTAAAACGACGGCCAGT-3'. A 7.5 to ways to accomplish the transfer: 124 mM NaClO gradient over 20 minutes at pH 8.0 4 1) The flow rate of the analytical method was decreased was used for separation of the primer from the various from 1.5 mL/min to 0.33 mL/min. A 10.0-mL/min failure sequences in the sample. Figure 40a shows an flow rate on the 22-mm diameter column is equivalent analytical separation of 1.0 µg of this sample on the to a 0.33-mL/min flow rate on the 4-mm column 4-mm diameter column. because flow rate scales with the cross-sectional area The objective of the scale up exercise was the of the column: transfer of this analytical method from the 4-mm diameter column at a flow rate of 1.5 mL/min to the 22-mm diameter (2 mm)2 10.0 mL/min × π = 0.33 mL/min{1} column at 10.0 mL/min. π(11 mm)2

33 High-Resolution Analysis and Purification of Oligonucleotides with the DNAPac PA100 Column Time (min) %A %C %D Curve Time (min) %A %C %D Curve 0.00 88 10 2 5 0.00 88 10 2 5 20.00 57 10 33 4 90.00 57 10 33 4 Full length 0.03 0.03 Full length

AU AU

0.00 0.00

024681012141618 20 010203040506070 80 90 Minutes Minutes 10957 10958

Figure 40a . Separation of the -20 primer from synthesis Figure 40b . Separation of 1 µg of the -20 primer with a failure sequences. A 7.5 to 124 mM NaClO4 gradient was used 90-minute NaClO4 gradient delivered at 0.33 mL/min. Note for analysis of the 1-µg sample. Eluent D, Option 3 was used: the similarity between this separartion and the 20-minute 0.375 M NaClO . 4 separation shown in Figure 40a. Eluent D, Option 3 was used:

0.375 M NaClO4. 2) The duration of the analytical method was increased to 90 minutes. The constant column volume scaling rule dictates that the 20-minute gradient at 1.5 mL/min is Time (min) %A %C %D Curve equivalent to a 90.9-min gradient at 0.33 mL/min: 0.00 88 10 2 5 90.00 57 10 33 4 0.03 Full length (1.5 mL/min) 20 min × = 90.9 min {2} (0.33 mL/min)

AU Separation of 1 µg of DNA on the 4-mm diameter column with a 90-minute gradient at a flow rate of 0.33 mL/min is shown in Figure 7b. Nearly identical 0.00 chromatograms were obtained in Figures 7a and 7b, 0 10 20 30 40 50 60 70 80 90 which demonstrates the scaling fidelity of separations Minutes 10959 on the DNAPac PA-100. 3) The final step in the scale up was to transfer this Figure 40c. Scale up of the -20 primer analytical separation 90-minute separation from the 4-mm diameter to the to the 22-mm diameter DNAPac column. A 30-µg sample was analyzed at a flow rate of 10.0 mL/min. Eluent D, Option 3 was 22-mm diameter column. For this transfer, the grad- used: 0.375 M NaClO4. ient used for Figure 40b was delivered to the 22-mm diameter column at a flow rate of 10.0 mL/min. Figure 40c shows the separation of 30 µg of DNA by this overloading the analytical capacity of the DNAPac method. The chromatography scales in a highly column. The full-length oligonucleotide can elute as a predictable manner, so that analytical methods can very broad peak under overload conditions. However, a be transferred easily to preparative applications. highly pure full-length oligonucleotide typically is present throughout this peak. Figure 41a shows an example of Oligonucleotide Purification overload purification of 150 µg of the -20 primer on the Injections up to approximately 10 µg of crude 4-mm diameter analytical column. The separation was

synthetic DNA yield sharp, symmetric peaks on the performed with a 90-minute NaClO4 gradient at a flow rate analytical 4-mm diameter column. On the 9-mm and of 0.33 mL/min, exactly as in the analytical separation in 22-mm diameter columns, upper load limits for analytical Figure 7b. The target oligonucleotide appears as an intense chromatography are approximately 50 µg and 300 µg, 3-minute peak beginning at about 61 minutes. Three 1- respectively. Purification of larger samples is possible by minute fractions were collected as shown in Figure 41b and

34 High-Resolution Analysis and Purification of Oligonucleotides with the DNAPac PA100 Column reanalyzed (see inset for fraction #2). The overall purity of the oligonucleotide collected from 61 to 64 minutes was 1.2 > 98%. For another example of overload purification, a 1-mg purification of crude primer DNA to > 97% purity on the 4-mm diameter DNAPac has been demonstrated in reference 1. AU

CONCLUSION

By using a variety of eluent salts, highly efficient 0.0 separations of synthetic DNA are possible on the DNAPac 0102030405060708090 PA-100. The DNAPac pellicular anion-exchange resin Minutes provides higher resolution separations of single-stranded 10960 DNA than macroporous resins or reversed-phase columns. Figure 41a . Purification of 150 µg of the -20 primer on the Analytical and semipreparative chromatography addition- 4-mm diameter column. The full-length primer elutes as a broad peak under these overload conditions. Chromatographic condi- ally can be performed under denaturing conditions for the tions are identical to those used for Figure 40b. separation of difficult samples. High-resolution analytical chromatography can be transferred directly to the 9-mm and 22-mm diameter columns for semipreparative

applications. Fraction 2 0.04 REFERENCES AU 1. Thayer, J.R.; McCormick, R.M.; Avdalovic, N. Methods in Enzymology, 1996, 271, 147–174. 0.00 2. Bergot, B.J.; Egan, W. J. Chromatogr. 1992, 02468101214161820 Minutes 599, 35.

1.2 F1

F2

F3

AU

0.0

54 56 58 60 62 64 66 68 70 Minutes 10961

Figure 41b . Expanded view of full-length primer peak from Figure 41a. Fractions are indicated. Purity of pooled fractions was >98%. Inset: rechromatography of fraction 2.

35 High-Resolution Analysis and Purification of Oligonucleotides with the DNAPac PA100 Column Product focus: systems Nucleic acid therapeutics applications notebook Biocompatible LC Systems

Exceptional results and speed, resolution, versatility with Benefits of our Bio-LC systems include: Bio-LC separations • Superior chromatographic performance Thermo Fisher Scientific provides solutions for • Industry-leading range of biocompatible pumps biopharmaceutical LC analysis. Thermo Scientific Dionex • Titanium and PEEK flow-path for full biocompatibility UltiMate™ 3000 Titanium Systems provide solutions for • Dual-gradient pump for true parallel, tandem, or biochromatographic demands from micro to analytical multidimensional chromatography range. System components are perfectly matched to meet • High-precision autoinjections from 0.1 to 250 μL the requirements for high-performance analysis as well as (default) with ultralow carryover purification. The wide range of solvent options allows easy • Sample fractionation and reinjection with the WPS- implementation of different gradient profiles, essential for 3000TBFC Thermostatted Biocompatible Pulled-Loop method development. Additionally, these systems provide Well Plate Autosampler with Integrated Fraction ™ superior ease of use and are compatible with all Thermo Collection and Thermo Scientific Dionex Chromeleon Scientific mass spectrometers, including our hybrid and Chromatography Data System (CDS) software Orbitrap™ instruments. The UltiMate 3000 Titanium System ensures full biocompatibility—critical to integrity of proteins during separation—while delivering high day-to-day reproducibility and robust operation, even under harsh salt and pH conditions.

37 Product Focus: Systems Orbitrap MS Instruments

Single-point control and automation for improved ease of Some of the benefits of our Orbitrap MS instruments use in LC/MS and IC/MS include: Thermo Fisher Scientific provides advanced integrated • High-confidence, high-resolution/accurate mass IC/MS and LC/MS solutions which combine ease of use with (HR/AM) intact mass analysis modest price and space requirements. UltiMate 3000 System • Resolving power >240,000 full width at half maximum Wellness technology and automatic MS calibration allow (FWHM) on our most advanced systems continuous operation with minimal maintenance. • Spectral multiplexing for enhanced duty cycle The Thermo Scientific Dionex IC and Reagent-Free™ • Self-cleaning ion sources for low-maintenance IC (RFIC™) systems automatically remove mobile phase operation ions for effort-free transition to MS detection. The Orbitrap • Chromeleon CDS software for single-point method family makes up some of the most powerful MS instruments configuration, instrument control, and data management on the market to date. With resolving powers and MSn • Compatible with all existing IC and LC methods capabilities, these systems offer broad screening possibilities Our Orbitrap MS instruments are not only ideal for and confidence in identification and quantitation analysis. interfacing with our chromatography systems, they also provide confidence with HR/AM detection to deliver excellent performance and tremendous versatility.

38 Product Focus: Systems Product focus: consumables Nucleic acid therapeutics applications notebook Nucleic Acid Columns

Analysis of oligonucleotide purity, screening, and • Capable of n-1, n+1 resolution for oligonucleotides purification • Compatible with solvent, high pH, and high Thermo Fisher Scientific provides a variety of nucleic temperatures acid columns for oligonucleotide purity analysis, fast • Provides easy scale up screening, and purification. Both the DNAPac PA100 and Additional column technologies within the DNAPac PA200 Semi-Preparative Columns are nucleic acid, anion- family include the DNASwift SAX-1S Monolith anion- exchange, polymer-based columns. The DNAPac columns exchange column, which provides exceptionally high purity provide excellent separation of oligonucleotides—including and yields of oligonucleotides. This column combines full length from n-1, n+1 and other failure sequences— DNAPac and monolith technology to provide exceptionally and support screening of synthetic oligonucleotides for high resolution and capacity for oligonucleotide purification, production yield and failure sequences. The DNAPac PA200 making it the ideal column for therapeutic and diagnostic column offers improved efficiency and enhanced stability research. All of these nucleic acid columns provide high under alkaline conditions. resolution of full length from n-1, n+1, and other failure Benefits of our DNAPac column line include: sequences that may not be possible with other columns on • High-resolution separation of oligonucleotides and the market. nucleic acids

40 Product Focus: Consumables Columns for Carbohydrate Separations Particle Size Pore Nominal Surface % USP Phase Phase Particle Type (µm) Size (Å) Area (m2/g) Carbon Endcapping Code Code Page

ssor ies HyperREZ XP Carbohydrate H+ spherical, polymer 8.0 – – – – L17 690 4-152 ColumnCarbohydrate Pb2+ Selectionspherical, polymer Guide8.0 –for – Oligonucleotide– – L34 691 4-152

Acc e Carbohydrate Ca2+ spherical, polymer 8.0 – – – – L19 692 4-152 Carbohydrate Na+ spherical, polymer 10.0 – – – – – 693 4-152

nd Separations Organic Acid spherical, polymer 8.0 – – – – L17 696 4-152 Sugar Alcohol spherica l, polymer 8.0 – – – – L19 697 4-152 ns a lu m Columns for Oligonucleotide Separations LC Co il ity re ssure ue nts Column Target Applications Base Ma trix Ma terial Substrate Crosslinking Latex Crosslinking Capacity Recommended El Recommended Flow Rate Solvent Compatib Maximum Backp pH Range DNAPac High resolution 13µm diameter 55% 5% 40µeq Hydroxide 1.5mL/min 0–100% 4000psi 2–12.5 PA100 separations of nonporous substrate (28MPa) single and double agglomerated with stranded DNA or alkyl quaternary RNA ammonium oligonucleotides functionalized latex 100nm MicroBeads DNAPac High resolution 8µm diameter 55% 5% 40µeq Hydroxide, 1.2mL/min 0–100% 4000psi 2–12.5 PA200 separations of nonporous substrate acetate/ (28MPa) single and double agglomerated with hydroxide stranded DNA or alkyl quaternary RNA ammonium oligonucleotides functionalized latex 130nm MicroBeads

DNASwift High resolution Monolith; N/A N/A 50mg, NaClO4 0.5–2.5mL Most 1500psi 6.0–12.4 separations for polymethacrylate of a 20 mer and NaCl Common substrate agglomerated oligonucleotide Organic oligonucleotides with quaternary amine Solvents functionalized latex 201 3

2012 - d Consum ab les mns a n hy C olu og rap Chr oma t

41 Product Focus: Consumables

4-120 Product focus: software Nucleic acid therapeutics applications notebook Chromeleon 7 Chromatography Data System Software

The fastest way to get from samples to results • Accelerate analyses and learn more from your data Discover Chromeleon CDS software version 7, the through dynamic, interactive displays. chromatography software that streamlines your path from • Deliver customized reports using the built-in Excel- samples to results. Get rich, intelligent functionality and compatible speadsheet. outstanding usability at the same time with Chromeleon CDS Chromeleon CDS software version 7 is a forward- ™ software version 7—the Simply Intelligent chromatography looking solution to your long-term chromatography data software. needs. It is developed using the most modern software tools • Enjoy a modern, intuitive user interface designed around and technologies, and innovative features will continue to be the principle of operational simplicity. added for many years to come. ™ • Streamline laboratory processes and eliminate errors The Cobra integration wizard uses an advanced with eWorkflows, which enable anyone to perform a mathematical algorithm to define peaks. This ensures that complete analysis perfectly with just a few clicks. noise and shifting baselines are no longer a challenge in • Access your instruments, data, and eWorkflows instantly difficult chromatograms. When peaks are not fully resolved, ™ in the Chromeleon Console. the SmartPeaks integration assistant visually displays integration options. Once a treatment is selected, the • Locate and collate results quickly and easily using appropriate parameters are automatically included in the powerful built-in database query features. processing method. • Interpret multiple chromatograms at a glance using Chromeleon CDS software version 7 ensures data MiniPlots. integrity and reliability with a suite of compliance tools. • Find everything you need to view, analyze, and report Compliance tools provide sophisticated user management, data in the Chromatography Studio. protected database structures, and a detailed interactive audit trail and versioning system.

43 Product Focus: Software References Nucleic acid therapeutics applications notebook Select Peer-Reviewed Publications

Sproat, B.; Colonna, F.; Mullah, B.; Tsou, D.; Andrus, A.; Chen, S-H.; Qian, M.; Brennan, J.M.; Gallo, J.M. Determi- Hampel, A.; Vinayak, R. An Efficient Method for the nation of Antisense Phosphorothioate Oligonucleotides Isolation and Purification of Oligoribonucleotides. and Catabolites in Biological Fluids and Tissue Extracts Nucleosides and Nucleotides 1995, 14, 255–273. using Anion-Exchange High-Performance Liquid Chro- Bourque, A.J.; Cohen, A.S. Quantitative Analysis of Phos- matography and Capillary Gel Electrophoresis. phorothioate Oligonucleotides in Biological J. Chromatogr., B 1997, 692, 43–51. Fluids using Direct Injection Fast Anion-Exchange Chen, J-K.; Schultz, R.G.; Lloyd, D.H.; Gryaznov, S. Chromatography and Capillary Gel Electrophoresis. Synthesis of Oligodeoxyribonucleotide N3´-P5´ Phos- J. Chromatogr., B 1994, 662, 343–349. phoramidates. Nucleic Acids Res. 1995, 23, 2661–2668. Thayer, J.R.; McCormick, R.M.; Avdalovic, N. High Resolu- Morgan, M.T.; Bennet, M.T.; Drohat, AC. Excision of 5-Ha- tion Nucleic Acid Separations by High Performance Liq- logenated Uracils by Human Thymine DNA Glycosyl- uid Chromatography. In Methods in Enzymol., Karger, ase. J. Biol. Chem. 2007, 282, 27578–27586. B, Hancock, B., Eds.; Academic Press: New York, 1996, Jurczyk, S.C.; Horlacher, J.; Devined, K.G.; Benner, S.A.; 271, 147–174. Battersby, T.R. Synthesis and Characterization of Grant, G.P.G.; Popova, A.; Qin, P.Z. Diastereomer Character- Oligonucleotides containing 2´-Deoxyxanthosine using izations of Nitroxide-Labeled Nucleic Acids. Biochem. Phosphoramidite Chemistry. Helv. Chim. Acta 2000, 83, & Biophys. Res. Commun. 2008, 371, 451–455. 1517–1524. Xu, Q.; Musier-Forsyth, K.; Hammer, R.P.; Barany, G. Use Wincott, F.; DiRenzo, A.; Shaffer, C.; Grimm, S.; Tracz, T.; of 1,2,4-Dithiazolidine-3,5-dione (DtsNH) and 3-Eth- Workman, C.; Sweedler, D.; Gonzalez, C.; Scaringe, S.; oxy-1,2,4-dithiazoline-5-one (EDITH) for Synthesis of Usman, N. Synthesis, Deprotection, Analysis and Puri- Phosphorothioate-Containing Oligodeoxyribonucleo- fication of RNA and Ribozymes.Nucl. Acids Res. 1995, tides. Nucl. Acids Res. 1996, 24, 1602–1607. 23, 2677–2684. Wang, X.D.; Gou, P.R. Polymerase-Endonuclease Amplifica- Grünewald, C., Kwon, T.; Piton, N.; Förster, U.; Wachtveitl, tion Reaction for Large-Scale Enzymatic Production on J.; Engels, J.W. RNA as Scaffold for Pyrene Excited Antisense Oligonucleotides. Nature Proceedings 2009 Complexes. Bioorg. & Med. Chem. 2008, 16, 19–26. (hdl:10101/npre.2009.3711.1 posted 2 Sept 2009). Tsou, D.; Hampel, A.; Andrus, A.; Vinayak, R. Large Scale Fearon, K.L.; Stults, J.T.; Bergot, B.J.; Christensen, L.M.; Synthesis of Oligoribonucleotides on a High-Loaded Raible, A.M. Investigation of the ‘n-1’ Impurity in Phos- Polystyrene (HLP) Support. Nucleosides. & Nucleotides phorothioate Oligodeoxynucleotides Synthesized by the 1995, 14, 1481–1492. Solid-Phase β-Cyanoethyl Phosphoramidite Method Vinayak, R.; Andrus, A.; Sinha, N.D.; Hampel, A. Assay using Stepwise Sulfurization. Nucl. Acids Res. 1995, 23, of Ribozyme-Substrate Cleavage by Anion-Exchange 2754–2761. High-Performance, Liquid Chromatography. Anal. Bio- Koziolkiewicz, M., Owczarek, A.; Domañski, K.; Nowak, chem. 1995, 232, 204–209. M.; Guga, P.; Stec. W.J. Stereochemistry of Cleavage of Internucleotide Bonds by Serratia marcescens Endo- nuclease. Bioorg. Med. Chem. 2001, 9, 2403–2409.

45 Select Peer-Reviewed Publications Sproat, B.S.; Rupp, T.; Menhardt, N.; Keane, D.; Beijer, B. Soutschek, J.; Akinc, A.; Bramlage, B.; Charisse, K.; Con- Fast and Simple Purification of Chemically Modified stien, R.; Donoghue, M.; Elbashir, S.; Geick, A.; Had- Hammerhead Ribozymes using a Lipophilic Capture wider, P.; Harborth,J.; John, M.; Kesavan, V.; Lavine, G.; Tag. Nucl. Acids Res. 1999, 27, 1950–55. Pandey, R.K.; Racie, T.; Rajeev, K.G.; Rohl, I.; Toud- Murray, J.B.; Dunham, C.M.; Scott, W.G. A pH-Dependent jarska, I.; Wang, G.; Wuschko, S.; Bumcrot, D.; Kote- Conformational Change, rather than the Chemical liansky, V.; Limmer, S.; Manoharan, M.; Vornlocher, Step, Appears to be Rate-Limiting in the Hammerhead H-P. Therapeutic Silencing of an Endogenous Gene by Ribozyme Cleavage Reaction. J. Mol. Biol. 2002, 315, Systemic Administration of Modified siRNAs.Nature 121–130. 2004, 432, 173–178. Earnshaw, D.J.; Masquida, B.; Müller, S.; Sigurdsson, S.Th.; Pham, J.W.; Radhakrishnan, I.; Sontheimer, E.J. Thermody- Eckstein, F.; Westhof, E.; Gait, M.J. Inter-Domain namic and Structural Characterization of 2´-Nitrogen- Cross-Linking and Molecular Modeling of the Hairpin Modified RNA Duplexes. Nucl. Acids Res. 2004, 32, Ribozyme. J. Mol. Biol. 1997, 274, 197–212. 3446–3455. Micura, R.; Pils, W.; Höbartner, C.; Grubmayr, K.; Ebert, Frank-Kamentetsy, M.; Grefhorst, A.; Anderson, N.N.; Ra- M-O.; Jaun, B. Methylation of the Nucleobases in RNA cie, T.N.; Bramlage, B.; Akinc, A.; Butler, D.; Charisse, Oligonucleotides Mediates Duplex-Hairpin Conversion. K.; Dorkin, R.; Fan, Y.; Gamba-Vitalo, C.; Hadwiger, P.; Nucl. Acids Res. 2001, 29, 3997–4005. Jayaraman, M.; John, M.; Jayaprakash, K.N.; Maier, M.; Massey, A.P.; Sigurdsson, S.Th. Chemical Syntheses of In- Nechev, L.; Rajeev, K.G.; Read, T.; Rohl, I.; Soutschek, hibitory Substrates of the RNA-RNA Ligation Reaction J.; Tan, P.; Wong, J.; Wang, G.; Zimmermann, T.; de Catalyzed by the Hairpin Ribozyme. Nucl. Acids Res. Fougerolles, A.; Vornlocher, H-P.; Langer, R.; Anderson, 2004, 32, 2017–22. D.G.; Manoharan, M.; Koteliansky, V.; Horton, J.D.; Prater, C.E.; Saleh, A.D.; Wear, M.P.; Miller, P.S. Chimeric Fitzgerald, K. Therapeutic RNAi Targeting PCSK9 RNAse H-Competent Oligonucleotides Directed to the Acutely Lowers Plasma Cholesterol in Rodents and HIV-1 Rev Response Element. Bioorg. Med. Chem. LDL Cholesterol in Nonhuman Primates. Proceedings, 2007, 15, 5386–5395. Natl. Acad. Sci. USA 2008, 105, 11915–11920. Cohen, C.; Forzan, M.; Sproat, B.; Pantophlet, R.; Mc- Li, F.; Pallan, P.S.; Maier, M.A.; Rajeev, K.G.; Mathieu, S.L.; Gowan, I.; Burton, D.; James, W. An Aptamer that Kreutz, C.; Fan, Y.; Sanghvi, J.; Micura, R.; Rozners, R.; Neutralizes R5 Strains of HIV-1 Binds to Core Residues Manoharan, M.; Egli, M. Crystal Structure, Stability and of gp120 in the CCR5 Binding Site. Virology 2008, 381, in vitro RNAi Activity of Oligoribonucleotides Contain- 46–54. ing the Ribo-difluorotoluyl Nucleotide: Insights into Sub- Mackman, R.L.; Zhang, L.; Prasad, V.; Boojamra, C.G.; strate Requirements by the Human RISC Ago2 Enzyme. Douglas, J.; Grant, D.; Hui, H.; Kim, C.U.; Laflamme, Nucl. Acids Res. 2007, 35, 6424–6438. G.; Parish, J.; Stoycheva, A.D.; Swaminathan, S.; Mikat, V.; Heckel, A. Light-Dependent RNA Interfer- Wang, K.; Cihlar, T. Synthesis, Anti-HIV Activity, and ence with Nucleobase-Caged siRNAs. RNA 2007, 13, Resistance Profile of Thymidine Phospohonomethoxy 2341–2347. Nucleosides and their bis-isopropyloxymethylcarbonyl Rhodes, A.; Deakin, A.; Spaull, J.; Coomber, B.; Aitken, A.; (bisPOC) Prodrugs. Bioorg. Med. Chem. 2007, 15, Life, P.; Rees, S. The Generation and Characterization of 5519–5528. Antagonist RNA Aptamers to Human Oncostatin M. Nair, T.M.; Myszka, D.G.; Davis, D.R. Surface Plasmon J. Biol. Chem. 2000, 275, 28555–28561. Resonance Kinetic Studies of the HIV TAR RNA Floege, J.; Ostendorf, T.; Janssen, U.; Burg, M.; Radeke, Kissing Hairpin Complex and its Stabilization by H.H.; Vargeese, C.; Gill, S.C.; Green, L.S.; Janjic, N. 2-Thiouridine Modification.Nucl. Acids Res. 2000, 28, Novel Approach to Specific Growth Factor Inhibition in 1935–1940. Vivo: Antagonism of Platelet-Derived Growth Factor in Thayer, J.R.; Rao, S.; Puri, N.; Burnett, C.A.; Young, M. Glomerulonephritis by Aptamers. American Journal of Identification of Aberrant 2´-5´ RNA Linkage Isomers Pathology 1999, 154, 169–179. by Pellicular Anion-Exchange Chromatography. Anal. Bioch. 2007, 361, 132–139.

46 Select Peer-Reviewed Publications Puffer, B.; Moroder, H.; Aigner, M.; Micura, R. 2’-Meth- Johansson, M.K.; Cook, R.M.; Xu, J.; Raymond, K.N. Time ylseleno-Modified Oligoribonucleotides for X-ray Crys- Gating Improves Sensitivity in Energy Transfer Assays tallography Synthesized by the ACE RNA Solid-Phase with Terbium Chelate/Dark Quencher Oligonucleotide Approach. Nucl. Acids Res. 2008, 36, 970–983. Probes. J. Am. Chem. Soc. 2004, 126, 16451–16455. Murray, J.B.; Szöke, H.; Szöke, A.; Scott, W.G. Capture Ouyang, X.; Shestopalov, I.A.; Sinha, S.; Zheng, G.; Pitt, and Visualization of a Catalytic RNA Enzyme-Product C.L.W.; Li, W-H.; Olson, A.J.; Chen, J.K. Versatile Syn- Complex using Crystal Lattice Trapping and X-ray Ho- thesis and Rational Design of Caged Morpholinos. lographic Reconstruction. Molec. Cell 2000, 5, 279–287. J. Am. Chem. Soc. 2009, 131, 13255–13269. Pallan, P.S.; Wilds, C.H.; Wawrzak, Z.; Krishnamurthy, Semenyuk, A.; Ahnfelt, M.; Estmer, C.; Yong-Hao, X.; R.; Eschenmoser, A.; Egli, M. Why does TNA Cross- Földesi, A.; Kao, Y-S.; Chen, H-H.; Kao, W-C.; Peck, Pair more Strongly with RNA than with DNA? An K.; Kwiatkowski, M. Cartridge-Based High-Throughput Answer from X-ray Analysis. Chem. Int. Ed. 2003, 42, Purification of Oligonucleotides for Reliable Oligo- 5893–5895. nucleotide Arrays. Analytical Biochemistry 2006, 356, Yang, H.; Lam, S.L. Effect of 1-Methyladenine on the Ther- 132–141. modynamic Stabilities of Double-Helical DNA Struc- Junker, H-D.; Hoehn, S.T.; Bunt, R.C.; Marathius, V.; Chen, tures. FEBS Letters 2009, 583, 1548–1553. J.; Turner, C.J.; Stubbe, J. Synthesis, Characterization Fearon, K.L.; Hirschbein, B.L.; Nelson, J.S.; Foy, M.F.; and Solution Structure of Tethered Oligonucleotides Nguyen, M.Q.; Okruszek, A.; McCurdy, S.N.; Frediani, Containing an Internal 3´-Phosphoglycolate, 5´-Phos- J.E.; DeDionisio, L.A.; Raible, A.M.; Cagle, E.N.; Boyd, phate Gapped Lesion. Nucl. Acids Res. 2006, 30, V. An Improved Synthesis of Oligodeoxynucleotide 5497–5508. N3´-P5´ Phosphoramidates and their Chimera using Gill, S.; O’Neill, R.; Lewis, R.J.; Connolly, B.A. Interaction Hindered Phosphoramidite Monomers and a Novel of the Family-B DNA Polymerase from the Archaeon Handle for Reverse Phase Purification.Nucl. Acids Res. Pyrococcus Furiosus with Deaminated Bases. 1998, 26, 3813–3824. J. Mol. Biol. 2007, 372, 855–863. Tucker, C.E.; Chen, L-S.; Judkins, M.B.; Farmer, J.A.; Gill, Ye, Y.; Munk, B.H.; Muller, J.G.; Cogbill, A.; Burrows, S.C.; Drolet, D.W. Detection and Plasma Pharmacoki- C.J.; Schlegel, B. Mechanistic Aspects of the Formation netics of an Anti-Vascular Endothelial Growth Factor of Guanidinohydantoin from Spiroiminodihydantoin Oligonucleotide-Aptamer (NX1838) in Rhesus Mon- under Acidic Conditions. Chem. Res. Toxicol. 2009, 22, keys. J. Chromatogr., B 1999, 732, 203–12. 526–535. Lee, B.M., Xu, J.; Clarkson, B.K.; Martinez-Yamout, M.A.; Misiaszek, R.; Uvaydov, Y.; Crean, C.; Geacintov, N.E.; Dyson, H.J.; Case, D.A.; Gottesfeld, J.M.; Wright, P.E. Shafirovich, V. Combination Reactions of Superoxide Induced Fit and “Lock and Key” Recognition of 5S with 8-oxo-7,8-Dihydroguanine Radicals in DNA. RNA by Zinc Fingers of Transcription Factor IIIA. J. Biol Chem. 2005, 280, 6293–6300. J. Mol. Biol. 2006, 357, 275–291. Boojamra, C.G.; Parrish, J.P.; Sperandio, D.; Gao, Y.; Petra- Eon-Duval, A.; Burke, G. Purification of Pharmaceutical- kovsky, O.V.; Lee, S.K.; Markevitch, D.Y.; Vela, J.E.; Grade Plasmid DNA by Anion-Exchange Chromatog- Laflamme, G.; Chen, J.M.; Ray, A.S.; Barron, A.C.; raphy in and RNAse-Free Process. J. Chromatogr., B. Sparacino, M.L.; Desai, M.C.; Kim, C.U.;Cihlar T.; 2004, 804, 327–335. Mackman, R.L. Design Synthesis and Anti-HIC activ- Harsch, A.; Marzill, L.A.; Bunt, R.C.; Stubbe, J.; Vouros, ity of 4´-Modified Carbocyclic Nucleoside Phosphonate P. Accurate and Rapid Modeling of Iron-Bleomycin- Reverse Transcriptase Inhibitors. Bioorg. Med. Chem. Induced DNA Damage using Tethered Duplex Oligo- 2009, 17, 1739–1746. nucleotides and Electrospray Ionization Ion Trap Mass Conn, G.L.; Brown, T.; Leonard, G.A. The Crytal Spectrometric Analysis. Nucl. Acids Res. 2000, 28, Structure of the RNA/DNA Hybrid 1978–1995. r(GAAGAGAAGC.d(GCTTCTCTTC) Biczo, R.; Hirsh, D.J. Structure and Dynamics of a DNA- shows Significant Differences to that found in Solution. Based Model System for the Study of Electron Spin-Spin Nucl. Acids Res. 1999, 27, 555–561. Interactions. J. Inorg. Biochem. 2009, 103, 362–372.

47 Select Peer-Reviewed Publications Nordin, B.E.; Schimmel, P. Plasticity of Recognition of the Thayer, J.R.;Yansheng, W.; Hansen, E.; Angelino, MD.; 3´-end of Mischarged tRNA by class-I Aminoacyl-tRNA Rao, S. Separation of Oligonucleotide Phosphorothioate Synthetases. J. Biol Chem. 2002, 277, 20510–20517. Diastereoisomers by Anion-Exchange Chromatography. Fonvielle, M.; Chemama, M.; Villet, R.; Lecerf, M.; Bouhss, Analytical Chemistry 2010 (in preparation). A.; Valéry, J-M.; Ethève-Quelquejeu, M.; Arthur, M. Li, H.; Miller, P.S.; Seidman, M.M. Selectivity and Affinity Aminoacyl-tRNA Recognition by the FemXWv Trans- of DNA Triplex Forming Oligonucleotides Containing ferase for Bacterial Cell Wall Synthesis. Nucl. Acids Res. the Nucleoside Analogues 2′-O-methyl-5-(3-amino- 2009, 37, 1589–1601. 1-propynyl)uridine and 2′-O-Methyl-5-propynyluridine. Garcia-Garcia, C.; Draper, D.E. Electrostatic Interactions in a Org. Biomol. Chem. 2008, 6, 4212–4217. Peptide-RNA Complex. J. Mol. Biol. 2003, 331, 75–88. Yang, C.J.; Wang, L.; Wu, Y.; Kim, Y.; Medley, C.D.; Lin, Perry, K.; Hwang, Y.; Bushman, F.D.; Van Duyne, G.D. H.; Tan, W. Synthesis and Investigation of Deoxyribo- Structural Basis for Specificity in the Poxvirus Topoi- nucleic acid/Locked Nucleic Acid Chimeric Molecular somerase. Molec. Cell 2006, 213, 343–354. Beacons Nucl. Acids Res. 2007, 35, 4030–4041. Lochmann, D.; Weyermann, J.; Georgens, C.; Prassl, R.; Kim, H-J.; Leal, N.A.; Benner, S.A. 2´-Deoxy-1-methylp- Zimmer, A. Albumin-Protamine-Oligonucleotide seudocytidine, a Stable Analog of 2´-deoxy-5-methyliso- Nanoparticles as a New Antisense Delivery System. Part cytidine. Bioorganic & Medicinal Chemistry 2009, 17, 1: Physicochemical Characterization. Euro. J. Pharma- 3728–3732. ceut. and Biopharmaceut. 2005, 59, 419–429. Siegel, R.W.; Bellon, L.; Beigelman, L.; Kao, C.C. Moieties Allen T.D.; Wick, K.L.; Matthews, K.S. Identification of in an RNA Promoter Specifically Recognized by a Viral Amino Acids in Lac Repressor Protein Cross-Linked to RNA-Dependent RNA Polymerase. Proc. Nat’l. Acad. Operator DNA Specifically Substituted with Bromode- Sci. U.S. 1998, 95, 11613–11618. oxyuridine. J. Biol Chem. 1991, 266, 6113–6119. Thayer, J.R.; Puri, N.; Burnett, C.; Hail, M.E.; Rao, S. Iden- Matthew-Fenn, R.S.; Das, R.; Silverman, J.A.; Walker, P.A.; tification of RNA Linkage Isomers by Anion-Exchange Harbury, P.A.B. A Molecular Ruler for Measuring Quan- Purification with ESI-MS of Automatically Desalted titative Distance Distributions. PLoS ONE 2008, 3(10), Phosphodiesterase-II Digests. Analytical Biochemistry e3229. Doi10.1371/journal.pone.0003229. 2009 (in press). Thayer, J.R.; Flook, K.J.; Woodruff, A.; Rao, S.; Pohl, C.A. Thayer, J.R.; Rao, S.; Puri, N. Detection of Aberrant 2´-5´ New Monolith Technology for Automated Anion-Ex- Linkages in RNA by Anion Exchange. In Current Pro- change Purification of Nucleic Acids.J. Chromatogr., B tocols in Nucleic Acid Chemistry 2008. Ed. S. Beaucage, 2010 (submitted). 10.13.1-11. Wiley Interscience, John Wiley and Sons. Thayer, J.R.; Barreto, V.; Rao, S.; Pohl, C. Control of Oligo- Thayer, J.R.; Flook, K.J.; Woodruff, A.; Rao, S.; Pohl, C.A. nucleotide Retention on a pH-Stabilized Strong Anion New Monolith Technology for Automated Anion- Exchange Column. Anal. Biochem. 2005, 338, 39–47. Exchange Purification of Nucleic Acids.J. Chromatogr., B 2010, 878, 933–941.

www.thermoscientific.com/dionex ©2012 Thermo Fisher Scientific Inc. All rights reserved. ISO is a trademark of the International Standards Organization. PEEK is a trademark of Victrex, PLC. All other trademarks are the property of Thermo Fisher Scientific Inc. and its subsidiaries. This information is presented as an example of the capabilities of Thermo Fisher Scientific Inc. products. It is not intended to encourage use of these products in any manners that might infringe the intellectual property rights Thermo Scientific Dionex products are of others. Specifications, terms and pricing are subject to change. Not all products are available in all countries. Please designed, developed, and manufactured under an ISO 9001 Quality System. consult your local sales representative for details.

Australia +61 3 9757 4486 Denmark +45 70 23 62 60 Japan +81 6 6885 1213 Switzerland +41 62 205 9966 Austria +43 1 333 50 34 0 France +33 1 60 92 48 00 Korea +82 2 3420 8600 Taiwan +886 2 8751 6655 Belgium +32 53 73 42 41 Germany +49 6126 991 0 Netherlands +31 76 579 55 55 UK/Ireland +44 1442 233555 Brazil +55 11 3731 5140 India +91 22 2764 2735 Singapore +65 6289 1190 USA and Canada +847 295 7500 China +852 2428 3282 Italy +39 02 51 62 1267 Sweden +46 8 473 3380

AN70321_E 12/12S