MB Final Project

Small Interfering RNA Analysis and Purification by Liquid Chromatography

MB Final Project
Chemical Information Retrieval Course
Drexel University, Department of Chemistry, Philadelphia, PA


Ribonucleic acid interference (RNAi) is a recently discovered natural process for regulation of gene expression. First discovered in plants and then in worms and mammals, the pathway uses short interfering RNA (siRNA) to silence gene expression. This discovery led to a surge in interest in using siRNA for biomedical research and drug development. Different biomedical and pharmaceutical companies are now evaluating the use of siRNA as therapeutics for treatment of different diseases such as cancer, macular degeneration and viral infections. As a result, good analysis and purification techniques are needed when developing these drugs to show safety and efficacy. In this summary, different analytical techniques are reviewed for analysis and purification of siRNA molecules.


Ribonucleic acid is a biologically important molecule that consists of a long chain of oligonucleotide units. Each nucleotide contains a ribose sugar, a nitrogenous base, and a phosphate group (Figure 1)[25]. There are four bases in RNA: adenine, guanine, cytosine, and uracil (Figure 2)[25].
Figure 1. Nucleotide unit consisting of a ribose sugar, a base, and a phosphate group.

Figure 2. The four bases in RNA are: Adenine (A), Guanine (G), Cytosine (C), and Uracil (U).

Adenine and guanine are purines, cytosine and uracil are pyrimidines. The phosphate groups are attached to the 3' position of one ribose and the 5' position of the next ribose and creating the backbone of the RNA strand. The bases can form hydrogen bonds and are referred to as base pair (bp). In the Watson-Crick base pairing system for RNA, adenine forms a base pair with uracil and guanine forms a base pair with cytosine (Figure 3)[25].

Figure 3. RNA Structure: adenine forms a base pair with uracil (A-U) and guanine with cytosine (G-C).

In RNA, a base pair between guanine and uracil can also occur. The hydrogen bonding between the base pairs is critical in the double stranded RNA (dsRNA) stability. The guanine and cytosine form three intermolecular hydrogen bonds, where as adenine and uracil base pair forms two intermolecular hydrogen bonds (Figure 4). Therefore, double stranded RNA (dsRNA) or deoxyribonucleic acid (DNA) with high G-C content tends to be more stable. DNA or dsRNA are relatively stable at room temperature but the two nucleotide strands can separate, or 'melt', at higher temperatures. The melting point is dependent on the length of the molecules, the extent of any mispairing, and the G-C content.

Figure 4. The G-C pair forms three intermolecular hydrogen bonds; the A-U pair forms two intermolecular hydrogen bonds (chemdraw).

RNA is very similar to DNA-the hereditary material in humans and almost all other organisms, but differs in a few important structural details. In the cell, RNA is usually single stranded and has much shorter chain of nucleotides while DNA is usually double stranded or double helix structure with very long chain of nucleotides. Also, RNA nucleotides contain ribose sugar while DNA contains deoxyribose (lack of hydroxyl group on the 2' position on the ribose sugar molecule). Lastly, RNA has the base uracil rather than thymine that is present in DNA (Figure 5)[25].

Figure 5. RNA vs. DNA

The first reports of HPLC analysis and purification of biopolymers, such as peptides, deoxyribonucleic acid (DNA) and ribonucleic acid (RNA) fragments were in the 1980s[1]. The chromatographic behaviors for small molecules compared to biomolecules was very different and most researchers believed that the retention mechanism for the two were fundamentally different. Some researchers suggested that the macromolecules or biopolymers precipitated at the column inlet until they were desorbed by a gradient of the mobile phase and eluted without any further interaction with the stationary phase. Snyder et al. however successfully applied separation theory for small molecules to macromolecules or biomolecules[2]. They showed that the separation of macromolecules differs from small molecules in degree, but not in the kind of separation. Most recently, with the RNAi discovery, analysis and purification of siRNAs have been reported.

Types of RNA

RNA is central to protein synthesis in the cell. There are different types of RNA such as messenger RNA (mRNA), transfer RNA (tRNA), ribosomal RNA (rRNA), dsRNA, and small interfering RNA (siRNA). mRNA carries information from DNA to the ribosome which is the site of protein synthesis in the cell. tRNA is a small RNA chain (~80 nucleotides) that transfers a specific amino acid to a growing polypeptide chain at the ribosome site of protein synthesis. rRNA combines with a protein to from ribosome in the cytoplast. dsRNA is RNA with two complimentary strands, similar to DNA, and can form the genetic material of some viruses referred to as double stranded RNA viruses. siRNA is a class of small dsRNA molecules typically about 19-25 nucleotides in length that play different roles in biology.

Importance and Discovery of siRNA

More recently it was discovered that siRNA can be used in the gene silencing pathway in the ribonucleic acid interference (RNAi) mechanism for regulation of gene expression[3]. In this mechanism, siRNA degrades mRNA , thereby preventing protein translation. siRNAs were first discovered by David Baulcombe's group at the Sainsbury laboratory in Norwich, England, as part of a study of post-transcriptional gene silencing in plants[4]. Shortly after, in 2001 Thomas Tuschl, Keith Rosenberg and colleagues showed that synthetic siRNAs were able to induce RNAi in mammalian cells[5]. In 2002, Ming Jiang and Jo Milner showed that siRNA could inhibit viral infections in human cells[6]. During the same year, Science named RNAi as "Technology of the Year". Morris et al. showed that siRNA could be used for gene silencing in human cells[7]. In 2006, Drs. Andrew Z. Fire from Stanford University School of Medicine and Craig C. Mello from University of Massachusetts Medical School won the Nobel Prize in medicine or physiology for discovering the RNAi mechanism.

This significant discovery and the use of siRNA for gene silencing in animals and human cells led to a surge in interest in using siRNA for biomedical and drug development research. Different biomedical and pharmaceutical companies started to explore the use of siRNA as therapeutics for the treatment of different diseases such as cancer, macular degeneration and viral infections. In 2005, Acuity Pharmaceutical was the first to enter Phase I clinical trials with an siRNA drug for age-related macular degeneration[8]. In 2006, Alnylam Pharmaceuticals reported first data showing siRNA drug delivery in primates. During the same time, Merck purchased siRNA Therapeutics and initiated an extensive study of siRNAs for drug development and manufacturing[9].

siRNA Structure and Preparation

siRNA is a small double stranded RNA (usually 21 nucleotide length) with 2 nucleotide overhangs on each 3' end (Figure 6)[26].

Figure 6. Schematic representation of siRNA molecule (19-nucleotide duplex)

Each strand has a 5' phosphate group and a 3' hydroxyl group. siRNAs for therapeutic use are prepared synthetically via stepwise synthesis. Double stranded siRNA are prepared by hybridization or annealing of two complimentary single-stranded RNA counterparts. The synthetic process is very efficient, with yields greater than 99% per coupling step. However, each strand can introduce its own set of impurities with each synthetic step and carried over to the final siRNA molecule. These impurities include mismatched sequences and noncomplementary single stranded sequences. The presence of these impurities in a therapeutic mixture can lead to unwanted and nontargeted gene silencing. In addition, the presence of nonhybridized single-strand RNA can lead to decrease in therapeutic potency. Therefore, a major challenge in developing siRNA therapeutics is assurance of purity to minimize off-target gene silencing. As a result, siRNA analysis and purification is very critical in developing safe and efficacious therapeutics.

Analysis of siRNA

There have been many different methods reported for the analysis and purification of oligonucleotides, including polyacrylamide gel electrophoresis (PAGE), capillary gel electrophoresis (CGE), anion-exchange high performance liquid chromatography (AX HPLC)[10-12], and ion-pair reversed phase liquid chromatography (IP RPLC)[13]. Oligonucleotide purification by PAGE method can lead to high purity product, but method is generally very long and the yield is limited. Anion exchange HPLC separates oligonucleotides according to the number of charged groups and the separation selectivity usually decreases with oligonucleotide length. Furthermore, in anion-exchange HPLC the mass transfer of oligonucleotides in the stationary phase is very slow[14,15]. This can be improved by using non-porous or monolithic stationary phases, but the limited mass capacity of these phases limits their usefulness for preparative scale applications. Ion-pair RPLC has been used most extensively for the analysis and separation of oligonucleotides. Another significant advantage in using IP RPLC (with volatile eluent components) is the ability to directly couple to mass spectrometry (MS) detection, where as anion-exchange RPLC methods are incompatible with MS. Soft ionization techniques in MS such as electrospray ionization (ESI)[16], or matrix assisted laser desorption ionization (MALDI) can be used with IP RPLC for the analysis of oligonucleotides. MS can provide very useful information regarding the sequence characterization and identification.

Martin Gilar et al. reported the evaluation of IP RPLC for oligonucleotide analysis and studied the retention mechanism with this separation technique[17]. They found that mass transfer of the oligonucleotides in the stationary phase was the major factor contributing to peak broadening on porous C18 stationary phases. In order to overcome this problem, they used small particle size (2.5 µm) columns with increased column temperature and relatively slow flow rates. Smaller particle sizes (2.5 µm vs. 5 µm) shorten the diffusion path and improve the separation of large slowly diffusing molecules such as oligonucleotides. Van Deemter curves for 50 x 4.6 mm columns packed with 5, 3.5 and 2.5 µm particle size columns showed that the mass transfer in the stationary phase (or the C term of Van Deemter equation) had a major impact on the oligonucleotide separation. Also, the separation performance with the smaller particle size column (2.5 µm) was significantly better than that of columns packed with 5 µm particles (Figure 7)[17].

Figure 7. Impact of sorbent particle size on the separation of a 2-30mer oligodeoxythymidine ladder. The separation was performed using a XTerra MS C18 30 x 4.6 mm column packed with 2.5 µm (A), 3.5 µM (B), and 5 µm sorbent (C). Reference: Martin Gilar et al. Journal of Chromatography A, 958 (2002):167-182.

The impact of the flow rate was also evaluated and it was found that slower flow rates (0.5 mL/min compared to 2 mL/min) afforded improved separations. Finally, the increased column temperature resulted in faster mass transfer in the stationary phase and improved the separation. However, more specialized columns such as Xterra hybrid-silica columns that are stable with higher pH mobile phase and temperature need to be used. In addition, higher temperatures can only be used in single stranded siRNA analysis, because double stranded RNAs have melting temperatures in the range of 20 °C and at higher column temperatures they would degrade on the column to their corresponding complementary single strands.

An ion-pairing buffer consisting of triethylammonium acetate (TEAA) along with a shallow linear gradient of organic modifier is typically used as a mobile phase for the analysis of siRNAs (Figure 8)[24].

Figure 8. UHPLC separation of siRNA duplex from excess of upper chain ssRNA (A) and excess of single-stranded lower chain (B). (C) Separation of siRNA duplex prepared by annealing nearly stoichiometric ratio of complementary RNA oligonucleotides. Column: Waters OST BEH C18, 2.1 x 50 mm, 1.7 µm. Mobile phase A: 100 mM TEAA (pH 7.0); mobile phase B: 20% Acetonitrile in mobile phase A. Gradient: from 35% B (7% Acetonitrile) to 85% B (13% Acetonitrile) in 10 min, 0.2 mL/min, 20 °C. The eluent was monitored at 260 nm.

The proper choice of mobile phase has a great impact on the siRNA separation. Some earlier studies have shown that non-ion-pair buffers such as ammonium acetate and acetonitrile have been used for the separation of oligonucleotides[18]. The results showed that the separation performance was acceptable up to ~10-mer, but sharply declined for longer oligonucleotides. In this case, the RPLC separation was solely based on oligonucleotide hydrophobicity. Since the different nucleobases in RNA have different hydrophobicities, the separation of these heterooligonucleotides can be very challenging. In addition, it has also been shown that smaller length oligonucleotides can be separated more easily than longer lengths. This is due to smaller difference in charge to length for longer chain oligonucleotides (100mer and 101mer have 1 % difference in charge/length) compared to smaller lengths (10mer and 11mer have 10% difference in charge/length)[19]. The decrease in separation selectivity with the increase in oligonucleotide length has also been described for ion-pairing RPLC[20,21] and capillary gel electrophoresis techniques[22].

With the introduction of ion-pairing buffers, such as TEAA, in addition to hydrophobicity of the nucleobases, charge-charge interaction is introduced in the separation mechanism that allows for better separation of the oligonucleotides including the ones with longer length. Also, the studies showed that the composition of the different bases also has an effect on the separation. The separation of guanine (G) has shown to be problematic for HPLC analysis due to the strong inter- and intramolecular interactions of the G-rich RNA sequences[23]. Finally, shallow linear gradients of organic modifiers are used because studies have shown that there is a sharp change of the retention factor in oligonucleotides (or large biopolymers compared to small molecules) with a small change in mobile phase strength. The changes in acetonitrile content have a greater effect on the retention of longer oligonucleotides than on shorter ones. For example, Martin Gilar et al. have reported that retention factor k for 15-mer oligonucleotide decreases from 100, 13.5, to 3.2 with a very small change of mobile phase composition from 8, 9, to 10% acetonitrile, respectively.

Purification of siRNA

Sean M. McCarthy et al. have most recently demonstrated the use of IP-RPLC for the analysis and purification of siRNA[24]. Their studies were conducted using ultra high performance liquid chromatography (UHPLC) instrument. Small scale purification of single strand RNA was performed using an analytical size column and TEAA mobile phase. As the column was overloaded, typical peak broadening was observed with some retention shift of impurities eluting before the main peak. "Heart-cutting" strategy in the preparative separation was used (or main peak was collected from its apex to about 30% of the peak height) to obtain good purity of the target oligonucleotide without much loss in product recovery. In this case, 1-2 mL single fractions were collected with about 50-70% recovery and >95% purity (Figure 9)[24}.
Figure 9. RP IP chromatographic purification of a crude synthetic RNA. The column loading was varied from 1.4 to 140 nmol load. Fractions were collected for each purification from the peak apex to 30% of the peak height. The eluent was monitored at 260 nm. Column: XBridge OST BEH C18, 4.6 x 50 mm, 2.5 µm. Mobile phase A: 100 mM TEAA (pH 7.0); mobile phase B: 20% Acetonitrile in mobile phase A. Gradient: from 30%B (6% Acetonitrile) to 52.5%B (10.5% Acetonitrile) in 30 min, 1 mL/min, 60°C.

Besides single-stranded RNA analysis, Sean M. McCarthy et al. also performed double-stranded RNA analysis and purification. However, several different points need to be considered when analyzing dsRNAs. The primary concern is the stability of the dsRNA under IP-RPLC conditions. Most dsRNA have melting temperatures in the range of 45 and 65 °C. Therefore some on-column melting of dsRNA can occur during analysis and purification, especially at higher column temperatures. The on-column duplex melting is typically observed as a dramatic peak broadening. If the column temperature reaches or exceeds the duplex melting point then the duplex rapidly melts on injection and on column and elutes as the two corresponding single strands. As a result, duplex RNA can be analyzed using lower column temperatures and 20 °C was used in this research study. Purification of the duplex then is possible if there is an efficient separation of the target duplex from truncated species. In this study, full length duplex RNA was collected by using heart-cutting strategy that provided duplex target material with purity greater then 98%.

In addition, siRNA purification using on-column annealing was also successfully demonstrated. In this approach, the first strand is injected at initial gradient conditions, followed by injection of the complementary strand. The gradient elution conditions are immediately started afterward. The results showed formation of duplex RNA on-column that could then be purified similarly as the single strands. However, the results also showed the on-column annealing approach was less efficient where increased amounts of non-annealed single strands were also observed.


Chromatography was discovered over 100 years ago and it plays a major role in analysis and purification in various research areas. Many types of chromatography were developed over the years that are used for analysis of compounds from small molecules to large biomolecules. With the more recent discovery of RNAi mechanism, siRNA analysis and purification has become very important. Different research groups have reported the use of few separation techniques, such as capillary electrophoresis, anion exchange and ion-pair liquid chromatography for the analysis and purification of siRNA molecules. In this summary, the ion-pair liquid chromatography technique was in more detail reviewed for the efficient analysis and preparative purification of siRNA molecules.


[1] Regnier F.E. HPLC of proteins, peptides, and polynucleotides. Anal. Chem. 55 (1983), p. 1298A.
[2] Snyder L.R., Stadalius M.A., Quarry M.A. Gradient elution in reversed-phase HPLC separation of macromolecules. Anal. Chem. 55 (1983), p. 1412A.
[3] RNA interference, Wikipedia.
[4] Hamilton A, Baulcombe D. A Species of Small Antisense RNA in Posttrasncriptional Gene Silencing in Plants. Science 286, 5441 (1999), p. 950-952.
DOI: 10.1126/science.286.5441.950
[5] Elbashir S., Harborth J., Lendeckel W., Yalcin A., Weber K., Tuschl T. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411, 6836 (2001), p. 494-498.
[6] Jiang M., Milner J. Selective silencing of viral gene expression in HPV-positive human cervical carcinoma cells treated with siRNA, a primer of RNA interference. Oncogene 21 (2002), p. 6041-6048.
[7] Morris K.V., Chan S.W., Jacobsen D.E., Looney D.J. Small Interfering RNA-Induced Transcriptional Gene Silencing in Human Cells. Science 305, 5688 (2004), p. 1289-92.
[8] http://www.medicalnewstoday.com/articles/44387.php
[9] Jarvis, Lisa M. Developing drugs based on RNA interference hinges on finding better ways to safely and potently get molecules into cells. C&EN September 7, 2009, vol. 87, issue 36, p.18-27.
[10] Bourque A.J., Cohen A.S. Quantitative analysis of phosphothioate oligonucleotides in biological fluids using direct injection fast anion-exchange chromatography and capillary gel electrophoresis. J. Chromatogr B 662 (1994), p. 343-349.
[11] Srivatsa G.S., Klopchin P., Batt M., Feldman M., Carlson R.H., Cole D.L. Selectivity of anion exchange chromatography and capillary gel electrophoresis for the analysis of phosphorothioate oligonucleotides. J. Pharm. Biomed. Anal. 16 (1997), p. 619-630.
[12] Fountain K.J., Gilar M., Gebler J.C. Analysis of native and chemically modified oligonucleotides by tandem ion-pair reversed-phase high-pressure liquid chromatography/electrospray ionization mass spectrometry. Rapid Commun. Mass Spectrom. 17 (2003), p. 646-653.
[13] Gilar M. Analysis and Purification of Synthetic Oligonucleotides by Reversed-Phase High Performance Liquid Chromatography with Photodiode Array and Mass Spectrometry Detection. Anal. Biochem. 298 (2001), p. 196-206.
[14] Heeter G.A., Liapis A.I. Model discrimination and estimation of the intraparticle mass transfer parameters for the adsorption of bovine serum albumin onto porous adsorbent particles by the use of experimental frontal analysis data. J. Chromatogr. A 776 (1997), p. 3-13.
[15] Miyabe K., Guichon G. Kinetic study of the mass transfer of bovine serum albumin in anion-exchange chromatography. J. Chromatogr. A 866 (2000), p. 147-171.
[16] Huber C.G., Krajete A. Analysis of Nucleic Acids by Capillary Ion-pair Reversed-phase HPLC Coupled to Negative-ion Electrospray Ionization Mass Spectrometry. Anal. Chem. 71 (1999), p. 3730-3739.
[17] Gilar M., Fountain K.J., Budman Y., Neue U.D., Yardley K.R., Rainville P.D., Russell II R.J., Gebler J.C. Ion-pair reversed-phase high-performance liquid chromatography analysis of oligonucleotides: Retention prediction. J. Chromatogr. A 958 (2002), p. 167-182.
[18] Eckstein H., Schott H. The Use of Multidimensional Chromatography for the isolation of Synthetic Oligodeoxyribonucleotides on a Preparative Scale. Chromatographia 19 (1984), p. 236-239.
[19] Baba Y., Fukuda M., Yoza N. Computer-Assisted Retention Prediction System for Oligonucleotides in Gradient Anion-Exchange Chromatography. J Chromatogr. A 458 (1988), p. 385-394.
[20] Haupt W., Pingoud A. Comparison of several high-performance liquid chromatography techniques for the separation of oligodeoxynucleotides according to their chain lengths. J. Chromatogr. A 260 (1983), p. 419-427.
[21] Huber C.G., Oefner P.J., Bonn G.K. High-Resolution Liquid Chromatography of Oligonucleotides on Nonporous Alkylated Styrene-Divinylbenzene Copolymers. Anal. Biochem. 212 (1993), p. 351-358.
[22] Kleparnik K., Foret F., Berka J., Goetzinger W., Miller A.W., Karger B.L. The use of elevated column temperature to extend DNA sequencing read lengths in capillary electrophoresis with replaceable polymer matrices. Electrophoresis 17 (1996), p. 1860-1866.
[23] Oefner P.J. Allelic discrimination by denaturing high-perforamnce liquid chromatography. J. Chromatogr. B 739 (2000), p. 345-355.
[24] McCarthy S.M., Gilar M., Gebler J. Reversed-phase ion-pair liquid chromatography analysis and purification of small interfering RNA. Anal. Biochem. 90 (2009), p. 181-188.
[25]RNA Structure, Google search http://www.google.com/images?hl=en&expIds=17259,27744,27868,27936&sugexp=ldymls&xhr=t&q=rna+structure&cp=5&qe=cm5hIHN0&qesig=JYTMiGyZU5gwM6TYYkNohg&pkc=AFgZ2tkV29cMAC5FZ-LaIfyH8cYdbXnioeC6_Y6DZrrKj1t2Vvz5tmSNAfH4wmp663N10ubVAs7tG04lIjXlvPlIi9CU7PurWg&safe=off&um=1&ie=UTF-8&source=univ&ei=vrn5TN-lPIW8lQevzfT_Bg&sa=X&oi=image_result_group&ct=title&resnum=1&sqi=2&ved=0CCsQsAQwAA&biw=1258&bih=596
[26] siRNA Structure http://www.gene-quantification.de/si-rna.html#def