5
Apart from the choice of analytical technique for oligonucleotide analysis, other considerations include the type of analysis required (i.e. impurity profiling, analysis of aggregate or assay determination), and whether the biopolymer is being analysed in its single or double stranded state, or both. Generally, all of these analytical criteria can be met with one or more of the following chromatographic techniques, ion pair reverse phase HPLC (IP-RP-HPLC) [34], anion exchange HPLC (AX-HPLC) [32] and size exclusion HPLC (SEC) [35]. Capillary gel electrophoresis (CGE) [33] has been included in this review because of its complementary and orthogonal nature to the more traditional separation techniques.
Analysis of Single-Stranded Oligonucleotides
The preferred methods for determining the purity of single stranded oligonucleotides are IP-RP-HPLC [34] and AX-HPLC [32]. Both techniques have the resolving power to separate deletion and extension sequences from the full-length oligonucleotides and for adduct impurities. Separations based on IP-RP-HPLC rely on the formation of transient ion pairs between the oligonucleotide and the ion pair reagents (as well as the more recognised interactions between the ion pair reagent and the stationary phase), resulting in increased and improved separation [36]. Commonly used ion pairing reagents include triethylammonium acetate (TEAA) [37], hexylammonium acetate (HAA) [38] and combination of 1,1,1,3,3,3- hexafluor-2- propanol (HFIP) and triethylammonia (TEA) [38]. In Figure 2, a 21-mer oligonucleotide has been analysed using IP-RP-HPLC before and after column purification [39]. In the crude sample, the increased hydrophobicity of the
longmers and adduct impurities results in increased retention. Conversely, the shorter mers are more polar eluting before the main peak. Apart from the choice of ion pair reagent, other factors such as particle size and the use of non-porous stationary phases (such as monoliths) can enhance the oligonucleotide separation via improved mass transfer and enhanced peak capacity in IP-RP-HPLC [40] and UPLC [41]. In Figure 3, a capillary monolith column eluted with TEAA was used to baseline resolve a mixture of homologous oligothymidylic oligonucleotides (12 to 18-mers) with very high efficiencies within five minutes [42].
AX-HPLC offers a complementary seperation alternative to IP-RP-HPLC and resolves oligonucleotides (and related impurities) primarily based on their charge differences [32]; however, hydrophilic and hydrophobic interactions may also play a role in the separation mechanism [43]. High resolution oligonucleotide separations can be achieved by optimising key parameters such as the ion exchange sorbent (e.g. diethylaminoethyl (DEAE) bonded onto a polystyrene matrix [44]), pH, organic solvents and counterions. Figure 4 shows an AX-HPLC separation of impurities present in a phosphorothioated 19-mer antisense oligonucleotide [45]. The resolved impurities include oxidised phosphodiester (P=O) groups (impurities A1, A2 and A3) and shortmers, where n-1 (impurity A4) and n-2 (impurity A7). In CGE, the gel-filled capillary resolves oligonucleotides via a ‘sieving’ mechanism primarily according to size. To overcome the bias associated with electrokinetic injection, internal standards are often included in the analysis. This is exemplified in Figure 5 which shows a CGE electropherogram of a fully phosphorothioated 20-mer resolved into its deletion sequence impurities in the presence of a 23-mer internal standard [46].
Figures 4. AX-HPLC Analysis of a Phosphorothioated 19-mer oligonucleotide (FT19), showing chromatograms showing resolution of the full length sequence from P=O impurities (A1, A2 and A3 in (a) to (c)) and and n-x deletion sequence impurities (A7 and A4 in (d) to (e)). See reference [45] for more details.
Figure 5. CGE of Phosphorothioated 20-mer oligonucleotide showing resolution of the full length sequence from n-1, n-2, n-3 deletion sequence impurities with a 23-mer internal standard (T23). See reference [46] for more details.
resolutions are improved as oligonucleotide length is reduced and on-column base pairing or secondary structure is eliminated from the analysis. Consequently, chromatographic separations of the individual single strands are more discriminating and have the ability to resolve closely related impurities formed during manufacture or on stability, resulting in a more complete measure of purity and greater understanding of the impurity profile. Analysis of the individual strands can be performed either prior to annealing the single strands or via the use of denaturing techniques which dissociate the duplex during the analysis, allowing for on-column separation of the individual sense and antisense strands.
The latter approach can give incomplete resolution between the two strands or their related impurities and consequently, analysis of the pre-annealed single strands is preferred. However, a denaturing impurities method would still be required to ensure that no additional degradation has occurred during the subsequent manufacturing processes (such as, annealing, desalting, lyophilisation) and during storage of the API. Figure 6 shows the separation of sense and antisense siRNA strands from their related impurities [49].
IP-RP-HPLC [34] and AX-HPLC [32] are capable of being run in both non-denaturing and denaturing conditions. In the latter approach, factors such as increased column temperature, increased organic modifier, extremes of pH, and low ionic strength in the mobile phase are used to dissociate or ‘melt’ the duplex into individual strands and eliminate any secondary structures. In contrast, opposing chromatographic conditions are used to create non-denaturing methodology. In particular, the use of a high ionic strength mobile phase is critical to maintaining the duplex structure and other secondary structure conformations such as aggregated impurities.
This approach can be very useful if the analytical goal is to separate a duplex oligonucleotide from aggregate impurities and excess unhybridised single stranded impurities. Figure 7 shows a SEC chromatogram of a 21-mer SiRNA duplex containing both aggregate and residual single stranded impurities [49]. Whilst non- denaturing SEC can be used to determine levels of residual single strand impurities, it suffers from poor specificity impacting on sensitivity. We have demonstrated [49] that residual single stranded impurities (≤1% w/w) co-elute under the tailing section of the duplex peak via SEC and are not detected [49]. However, non-denaturing SAX can provide excellent specificity for the detection and quantification of residual single strand impurities.
Figure 3. IP-RP HPLC Separation of oligonucleotides d(pT)12 to d(pT)18 using a monolith column. See reference [42] for more details.
Analysis of Double-Stranded Oligonucleotides
Of the various classes of oligonucleotide therapeutics, double stranded oligonucleotides such as siRNA (short interfering RNA) offer some of the more significant analytical challenges from a quality control perspective. The individual antisense and sense strands are non-covalently hybridised to form an alpha-helical duplex which is the active pharmaceutical ingredient (API). Non-denaturing chromatographic techniques, aimed at maintaining the native duplex structure of the oligonucleotide during analysis, are routinely used. Non-hybridised single strands may be present as impurities in the API. In addition, aggregated oligonucleotide impurities (larger than the targeted duplex conformation) can be formed during manufacture or during storage of the API [47]. The presence of these impurities is of concern from both a potency and safety perspective [48]. Whilst chromatographic analysis of the intact duplex is obviously necessary, analysis of the individual sense and antisense strands is equally important. Chromatographic
Figure 6. Denaturing RP-IP-UPLC Analysis of a 21-mer siRNA double stranded oligonucleotide showing separation of sense and antisense strands from related impurities. See reference [49] for more details.
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52
