Extraction of mipomersen from liver using Biotage® Oligo SPE

Introduction

Accurate quantitation of therapeutic oligonucleotides in tissue is essential for pharmacokinetic, toxicology, and broader bioanalytical studies, particularly during clinical development and therapeutic monitoring. While plasma is often the primary matrix of choice, tissue analysis is indispensable for understanding distribution, metabolism, and potential toxicity. Among tissue matrices, liver is notably complex, with proteins, lipids, endogenous nucleic acids, and active nucleases that can bind, degrade, or interfere with oligonucleotide detection. These components contribute to significant matrix effects and reduced extraction efficiency if samples are processed without adequate clean-up.

The complexity of solid tissues further necessitates efficient homogenization and robust extraction workflows that differ markedly from traditional plasma or serum methods. As oligonucleotide therapeutics continue to gain popularity due to their high specificity for treating genetic diseases and disorders, optimized tissue sample preparation becomes increasingly critical, not only to remove endogenous compounds that contribute to matrix effects, but also to concentrate the analyte and ensure consistent recovery and reliable quantitation.

Mipomersen (brand name Kynamro®) serves as a representative model for evaluating extraction performance. Approved by the FDA in 2013 for the treatment of homozygous familial hypercholesterolemia, mipomersen is a second-generation antisense oligonucleotide that inhibits apolipoprotein B-100 synthesis, a key structural component of low-density lipoprotein (LDL), which contributes to high cholesterol levels. Chemically, mipomersen is a 20-mer single-stranded oligonucleotide with a phosphorothioate backbone, incorporating 2’-O-methoxyethyl (2’-OMe) modifications and 5-methylcytosine (5’-Me-dC) substitutions to enhance nuclease resistance, increase binding affinity, and improve overall tissue stability. These are key features of many clinically relevant antisense oligonucleotides.

This application note describes a streamlined workflow for extraction of mipomersen from lamb liver, combining bead-based homogenization with solid phase extraction (SPE) using Biotage® Oligo SPE plates via a novel weak anion exchange (WAX) sorbent chemistry.

The optimized workflow delivers high sensitivity and reliable quantitation across a dynamic range of 0.1-25 pmol/mg, with recovery ≥70 %, demonstrating its suitability for liver tissue analysis and its adaptability to other solid biological matrices.

Analytes

Figure 1. Molecular structure of mipomersen 

Name: Mipomersen

Molecular weight: 7177.0 g·mol⁻¹

Size (mers): 20

Linkers: PO₂S^3-

Modifications: 2’-OMe, 5’-Me-dC

Sequence: 5'-/52MOErG/*/i2MOErC/*/i2MOErC/* /i2MOErT/*/i2MOErC/*A* G*T*/iMe-dC/* T*G*/iMe-dC/* T*T*/iMe-dC/*/i2MOErG/*/i2MOErC/*/i2MOErA/* /i2MOErC/*/32MOErC/-3'

Key:
m = 2'-OMe ribose sugar
* = phosphorothioate linker bond
r = RNA bases
/iMe-dC/ = 5-methyl deoxycytidine
/i2MOErX/ = 2'-MOE ribose sugar

Sample preparation procedure

Format

Biotage® Oligo SPE 30 mg plate, part number: 654-0030-PX01

Processing

Samples were processed using Biotage® PRESSURE+ 96 positive pressure manifold, part number: PPM-96

Sample pre-treatment

All steps were performed using low-bind or polypropylene labware to minimize oligonucleotide adsorption. Lamb liver (50 mg) was pre-weighed directly into 2 mL Biotage® Lysera centrifuge tubes containing five 2.8 mm ceramic beads. Each tube received 10 µL of mipomersen (calibrator or QC at the desired concentration) and 220 µL nuclease-free water, resulting in an approximate 1:5 tissue-to-aqueous dilution. Samples were homogenized using a Biotage® Lysera for 1 min cycle at 5.5 m/s.

Following homogenization, 20 µL proteinase K (18 mg/mL) was added to the extracts, vortex-mixed, and digested for 1 h at 55 °C. After digestion, 250 µL of lysis buffer was added, vortex-mixed, and the tubes were shaken for an additional 5 minutes. Samples were then centrifuged at 13.300 rpm for 10 minutes. A 200 µL aliquot of the resulting supernatant was transferred into an Eppendorf tube and mixed with 500 µL of 50 mM ammonium acetate (pH 5.5).

This preparation resulted in a liver:lysis buffer:ammonium acetate ratio of 1:1:5, and the final mixture was used as the sample load in the SPE workflow.

Preliminary experiments on recovery were conducted using 50 mg lamb liver spiked with 25 µL mipomersen (10 pmol/µL).

Extraction procedure

A 700 µL aliquot of pre-treated liver was loaded onto the SPE plate and processed using a Biotage® PRESSURE+ 96 Positive Pressure Manifold at ~1 psi (fine control setting) for all steps. The plate-drying step was performed at ~20 psi. This increased pressure was applied after wash 3 and again following the elution step by gradually increasing the pressure using the fine control before switching to the coarse control to reach ~20 psi.

The full extraction protocol is demonstrated in Table 1.

Table 1. Mipomersen SPE extraction procedure 

Note: Ionic strength of wash/elution solvents is important. Guidelines for preparation are provided in the reagent preparation section.

Post-extraction

500 µL extracts were evaporated in 96-well collection plates using the TurboVap® 96 Dual with following parameters:

Gas Temp: 60 °C

Plate Temp: 60 °C

Gas Flow: 60 L/min

Plate Height: 48 mm

Time: 50 min

Extracts were reconstituted in 200 µL of mobile phase A:B (5:95). The plate was covered with a sealing mat and vortex-mixed for 25–30 min prior to LC–MS analysis.

Note: Reconstituted extracts were subjected to increased mixing times to ensure complete solubilization of the oligonucleotide, to prevent adsorption losses, and to produce a homogeneous solution for analysis.

The reconstitution solvent described above is optimized for hydrophilic interaction liquid chromatography (HILIC) separation as used in this application note. Extracts prepared using Biotage® Oligo SPE products are also compatible with the commonly used reversed-phase ion-pair (RP-IP) approach. The reconstitution solvent should therefore be adjusted to match the initial conditions of the chromatographic method used.

Analytical conditions

UHPLC conditions

• Instrument: Waters ACQUITY Premier UPLC System with FTN Sample Manager
• Column: ACQUITY Premier BEH Amide VanGuard^TM FIT UPLC 1.7 µm (50 x 2.1 mm), p/n 186009507 with a ACQUITY Premier BEH Amide VanGuard^TM cartridge 1.7 µm (5 x 2.1 mm), p/n 186009510
• Mobile Phase A: 30 mM ammonium acetate + 0.04% NH₄OH in 70:30 (v/v) water:acetonitrile, pH 8.8
  
• Mobile Phase B: 30 mM ammonium acetate + 0.04% NH₄OH in 30:70 (v/v) water:acetonitrile, pH 8.8
• Strong wash: Water: Acetonitrile (70:30, v/v)
• Purge/weak needle wash: water:acetonitrile (10:90, v/v)
• Seal wash: water:methanol (50:50, v/v)
• Flow rate: 0.5 mL/min
• Injection volume: 10 µL
• Column temperature: 40 °C
• Sample temperature: 10 °C
  

Table 2. UHPLC gradient conditions

Note: The LC flow state was diverted to waste initially and then set to MS at 0.8 mins followed by waste from 2.5 mins onwards.

MS/MS conditions

• Instrument: Waters Xevo TQ Absolute XR
• Ionization mode: ESI -
• Acquisition mode: MRM
• Desolvation Temperature: 600 °C
• Capillary Voltage: 2.00 kV
• Desolvation Gas Flow: 1000 L/hr
• Cone Gas Flow: 150 L/hr
• Source Temperature: 150 °C

Table 3. MRM Parameters

Note: MRM parameters for mipomersen were automatically set using the auto-dwell function. Although the parent-to-parent MRM approach is less selective, it provides an additional option for qualifier ions.

Results and discussion

Extraction recoveries

Recoveries of mipomersen from liver were consistently ≥70 % using the Biotage® PRESSURE+ 96, demonstrating reliable extraction performance, see Figure 2 for recoveries and corresponding matrix factors.

Figure 2. Recoveries and matrix factors using Biotage® Oligo SPE plate on the Biotage® PRESSURE+ 96 

Method development focused on optimizing tissue pre-treatment and extraction conditions to maximize oligonucleotide release and minimize losses prior to SPE. Spiking experiments at multiple workflow stages were used to assess the effects of sample dilution, homogenization, digestion, lysis, centrifugation, and SPE conditions on overall recovery.

When mipomersen was spiked into pre-treated liver supernatant immediately before SPE, recoveries were ≥90 %. In contrast, spiking into untreated tissue led to significant pre-extraction losses, see Figure 3, indicating incomplete analyte release from the matrix.

Homogenizing liver at 1:5 (w/v) ratio with water effectively eliminated the foaming observed when surfactant-containing lysis buffers were applied directly to tissue. Spiked liver samples were homogenized using Biotage® Lysera, followed by proteinase K digestion for 1h at 55 °C to promote protein breakdown and oligonucleotide release. Two pre-treatment strategies were evaluated. In the first, lysis buffer was added directly to the crude homogenate, followed by shaking, centrifugation, and transfer of 200 µL supernatant to a low-bind polypropylene tube prior to dilution with 500 µL equilibration buffer. In the second, 100 µL of crude homogenate was removed before lysis, mixed 1:1 with lysis buffer, shaken for 5 min, then diluted with 500 µL equilibration buffer and centrifuged. Performing centrifugation immediately after lysis buffer addition produced recoveries of ≥70 % and effectively removed insoluble material, see Figure 4. In contrast, delaying centrifugation until after equilibration buffer addition reduced recoveries around 25 %.

Proteinase K digestion further improved performance. With proteinase K, recoveries were ≥70 %, whereas omission of the enzyme reduced recoveries to around 40 % and increased variability, see Figure 5. Samples lacking proteinase K also required higher positive pressure during SPE processing. Increasing the proteinase K concentration two-fold or extending digestion to 2h did not produce additional gains in recovery or matrix factor, confirming that the standard digestion conditions were already sufficient for complete enzymatic activity.

SPE wash conditions also influenced matrix effects. Increasing the number of wash 3 cycles from a single wash to three washes improved matrix factors by approximately 0.2, indicating more effective removal of residual endogenous material without compromising analyte recovery, see Figure 6.

Figure 3. Spike into liver vs spike into pre-treated liver supernatant 

Figure 4. Effects of centrifugation during sample pre-treatment workflow  

Figure 5. Proteinase K digestion vs without proteinase K 

Figure 6. Effect of increasing the number of wash 3 cycles on matrix factors

Extract cleanliness

Extract cleanliness was assessed to confirm removal of endogenous matrix components. SDS-PAGE analysis comparing blank liver homogenate with Biotage® Oligo SPE processed extracts showed numerous high- and low-molecular-weight protein bands in the untreated liver, whereas both SPE-processed replicates exhibited no detectable protein bands.

A proteinase K treated liver sample confirmed effective enzymatic digestion of proteins. Together, these results demonstrate effective removal of proteinaceous material following SPE clean-up, see Figure 7.

Figure 7. SDS-PAGE comparison of blank liver homogenate (50 mg liver in 250 µL water) and Biotage® Oligo SPE processed liver extracts

Phospholipid profiling confirmed removal of phosphatidylcholine (PC) and lysophosphatidylcholine (LPC), two major ion-suppressing phospholipid classes in liver. SPE provided markedly cleaner extracts than protein precipitation, as shown in Figure 8.

Figure 8. Total ion chromatograms (TICs) showing phospholipid removal of PC (top chromatogram) and LPC (bottom chromatogram) from liver using Biotage® Oligo SPE compared to classic protein precipitation.

Finally, full scan MS analysis demonstrated complete removal of surfactant-derived signals in the SPE treated extracts. As shown in Figure 9, the lysis buffer control exhibited abundant surfactant-related signals between retention times 1.00 – 3.50 mins, whereas the SPE extract effectively eliminated this background, confirming improved clean-up and enhanced sample cleanliness. This ensures compatibility with downstream LC-MS/MS analysis and minimizes risk of ion suppression and contamination.

Figure 9. Full-scan total ion chromatograms (TICs) comparing a lysis buffer control (LHS) and a Biotage® Oligo SPE treated extract (RHS).

Linearity and LOQ

Calibration curves were constructed using liver spiked between 0.1-25 pmol/mg and extracted in duplicate. A 1/x weighted linear regression model was applied to generate the calibration and determine the correlation coefficient. No internal standards were used in this study, highlighting the reproducibility of uncorrected data and the robustness of the method. A representative calibration curve is shown in Figure 10. Good linearity was obtained over the concentration range investigated, with coefficient of determination (r²) greater than 0.99, see Tables 4 and 5. The limit of quantification (LOQ) was defined as the lowest concentration at which the analyte gave a signal to noise ratio of at least 10:1.

For precision and accuracy, three QC levels (low, medium and high) were analyzed in n=5. Accuracy was calculated as the mean relative error from the target concentration and precision was evaluated as the relative standard deviation (RSD %). Calibration standards and QCs were considered acceptable if they were within ±15 % of their nominal concentrations, except for the LOQ and lowest QC, which were allowed ±20 %. Precision and accuracy for mipomersen using the Biotage® PRESSURE+ 96 procedure are summarized in Table 6.

Figure 10. Calibration plot for mipomersen quantifier ion (1434.3 > 94.9)

Table 4. Calibrators and QC concentrations used to construct calibration curves

Table 5. Linearity and LOQ for mipomersen quantifier and qualifier ions

Table 6. QC precision and accuracy

Conclusion

An SPE method using Biotage® Oligo SPE 30 mg plate was optimized and developed for the extraction of mipomersen from lamb liver. The method demonstrated excellent quantitative performance, with linear calibration curves without the use of an internal standard from 0.1 – 25 pmol/mg using 1/x weighted regression and R² values ≥0.99, confirming linearity across the investigated range.

High and reproducible recoveries (≥70 %) were achieved, demonstrating the robustness of the SPE workflow. Matrix effects were well controlled, with matrix factors around 0.8 (including evaporative losses), indicating effective reduction of ion suppression. Excellent clean-up was demonstrated by SDS-PAGE analysis and LC-MS TICs, showing effective removal of liver proteins, phospholipids, and surfactants following SPE.

All data were generated using HILIC chromatographic separation. Although sensitivity differences between HILIC and RP-IP chromatography were not evaluated, the observed linearity, recoveries, and extract cleanliness demonstrate that this SPE procedure is suitable for quantitative analysis of oligonucleotides in lamb liver by LC-MS/MS.

Chemicals and reagents

• Mipomersen sodium standard was obtained from Insight Biotechnology (Wembley, UK)
• All solvents were HPLC grade from Rathburn Chemicals Ltd. (Scotland, UK)
• Nuclease-free water was purchased from Fisher Scientific (Loughborough, UK)
• Analytical grade reagents were from Sigma-Aldrich (Gillingham, UK)
• LC-grade water at 18 MΩ was obtained fresh daily from a Direct-Q 5 water purifier from Merck Life Science (Watford, UK)
• Lamb liver was obtained from the local butchers (Cardiff, UK)

Additional information

Mipomersen stock and working solution

Mipomersen stock was made up in nuclease-free water at a concentration of 1 mg/mL (1000 pmol/µL). An intermediate solution was then prepared by diluting the stock to 400 pmol/µL in 20:80 (v/v) water:acetonitrile. All working solutions were prepared from this intermediate solution using serial dilutions in 20:80 (v/v) water:acetonitrile. All mipomersen solutions were stored at -20 °C.

Reagent preparation

Equilibration buffer and wash 1: 50 mM ammonium acetate aq. (pH 5.5)

Dissolve 3.85 g ammonium acetate (MW = 77.08 g.mol^-1) in 900 mL water, adjust pH to 5.5 using glacial acetic acid. Bring volume to 1 L and store at room temperature for up to 1 week.

Wash 2: 50/50 (v/v) 50 mM ammonium acetate aq. (pH 5.5) / acetonitrile

Mix 200 mL 50 mM ammonium acetate aq. pH 5.5 with 200 mL acetonitrile and store at room temperature for up to 1 week.

Wash 3: 25/75 (v/v) 200 mM ammonium bicarbonate aq. (pH 8) / acetonitrile

Dissolve 1.58 g ammonium bicarbonate (MW = 79.06 g·mol^-1) in 80 mL water, sonicate for 1 min to aid solubility, adjust pH to 8 using 25 % ammonium hydroxide. Bring volume to 100 mL and add 300 mL acetonitrile. Store at room temperature for up to 1 week.

Elution buffer: 50/50 (v/v) 200 mM ammonium bicarbonate aq. pH 10 / acetonitrile

Dissolve 1.58 g ammonium bicarbonate (MW = 79.06 g·mol^-1) in 80 mL water, sonicate for 1 min to aid solubility, adjust pH to 10 using 25 % ammonium hydroxide. Bring volume to 100 mL and add 100 mL acetonitrile. Store at room temperature for up to 1 week.

Lysis buffer

The lysis buffer consisted of guanidine hydrochloride (6 M), TCEP hydrochloride (7 mM), urea (2 M), citric acid (4 mM), L-cysteine (8 mM), and 2 % (v/v) non-ionic surfactant (Triton X-100 or tergitol).

Dissolve 57.32 g guanidine hydrochloride (MW= 95.53 g·mol^-1) in approximately 40 mL water with sonication to aid solubility. Add 12.01 g urea (MW= 60.06 g·mol^-1) and sonicate until fully dissolved, then add 0.08 g citric acid (MW = 192.12 g·mol⁻¹), 0.20 g TCEP (MW = 286.65 g·mol⁻¹), and 0.097 g L-cysteine (MW = 121.16 g·mol⁻¹), sonicating briefly after each addition. Adjust the pH to 2.5 with glacial acetic acid. Bring the volume to 98 mL with water, add 2 mL surfactant (Triton X-100 or tergitol), and mix thoroughly. Store at 4 °C for up to 2 weeks.

Literature number:  AN1026 

Ordering information