All chemicals and reagents are listed and described below in alphabetical order (Table 1). For all substances the purity is at least p.a. or higher. Purity of LNA-modified oligonucleotides containing a phosphorothioate backbone was determined by EXIQON with HPLC-ESI-MS analysis (Table 2).

In this study murine, as well as porcine EDTA-plasma was used. As an internal standard 1.5 μL of a 200 pmol/μL antimiR oligonucleotide Scr, or 132 solution were added to 28.5 μL of plasma to achieve a final concentration of 10 pmol/μL. For Proteinase K digestion 30 μL of plasma containing antimiR oligonucleotides were mixed with 40 μL proteinase K solution and 330 μL Proteinase K digestion buffer. The mixture was incubated at 55°C with 450 rpm on an Eppendorf Thermomixer comfort for 16 h overnight. Subsequently, antimiR oligonucleotides were precipitated by adding 100 μg of glycogen and 2 volumes ethanol followed by incubation for 4 h at −20°C. Then the samples were centrifuged at 21100 x g for 30 min and 4°C with a HERAEUS FRESCO 21 centrifuge. Supernatant was removed and pellets were dissolved in 30 μL ammonium citrate dibasic (AHC) solution at RT.

For MALDI-mass spectrometry (MALDI-MS) analysis a 5800 MALDI-TOF/TOF mass spectrometer (Sciex) was used. A special metal polish paste provided by Sciex successfully removed even the slightest impurities on the MALDI target plate so that a homogenous crystallization behavior of the analyte/matrix mix and thus a higher reproducibility of the analytical results was ensured. 5 μL of an antisense oligonucleotide (ASO) solution were mixed thoroughly with 5 μL of THAP matrix-solution in a 0.5 mL tube by vortexing with subsequent spotting of 0.5 μL of this solution per spot onto an MALDI target plate. This added a dilution factor of four on spot in terms of molarity. The Acquisition method was set up to the following parameters: Fixed Laser Intensity: 6100; Mass Range (Da): 4.000 to 6.000; Focus Mass: 5.000; Delay Times (ns): 600; X2 Deflector: 0.055; Shots/sub-spectrum: 200; Total Shots/Spectrum: 3000; Stage Velocity (μm/s): 600; Bin Size (ns): 1.0; Vertical Offset (% full scale): −0.5; Detector Voltage Multiplier: 0.67; Laser Pulse Rate (Hz): 400. The other parameters were set to standard. For analysis the TOF/TOF^TM Series Explorer^TM Software Version 4.1.0 (build 12), Oracle Database Schema Version 4.0.0, Data Version 4.0.4 was used.

For quantification of oligonucleotides, peak areas were utilized. For each analysis, three technical replicates per sample were generated and the values were averaged with subsequent determination of standard deviation (SD) as well as the coefficient of variation (CV).

The limit of detection (LOD) and lower limit of quantification (LLOQ) of antimiR oligonucleotides in mice as well as pig plasma were determined by setting up several dilutions of antimiR132. In addition, antimiRScr was added to each sample at a final concentration of 10 pmol/μL as an internal standard. Afterward 5 μL of antimiR oligonucleotide solution were mixed thoroughly with 5 μL of THAP matrix-solution (Table 1) in a 0.5 mL tube by vortexing with subsequent spotting of 0.5 μL of this solution per spot on an MALDI target plate. MS analysis was performed as described above. Ratios obtained from antimiR132 and antimiRScr were inserted into the calibration curve equation (Figure 3), which was then resolved to x to obtain the amount of substance. Taking into account all dilution factors, the concentration of the substance was then calculated.

All experimental procedures involving mice were performed in accordance with the local and national regulations and approved by local animal welfare bodies of the Hannover Medical School, Germany or University of Kaposvar, Hungary. Male C57BL/6N Crl mice were administered once at the beginning of the experiment with 20 mg/kg antimiR132 intravenous (i.v.) and sacrificed after various time points to remove plasma and organs, namely heart and liver. The respective substance and dose combination was applied intravenously by administration into the lateral tail veins. For blood collection, the mouse was placed under anesthesia using isoflurane. The study plan is depicted in Table 3.

Additionally, heart and liver were taken 24 h, 72 h, and 1 weeks post application of antimiR132.

The tissue and plasma samples for this assay development study were derived from an animal study and used in all described experiments in different quantities and specification.

The protein concentration of mouse plasma was determined by Bradford protein assay. Plasma or tissue extracts were diluted with Millipore water if necessary and 5 μL were mixed with 200 μL of Bradford reagent (Table 1) and 800 μL of Millipore water with subsequent incubation for 5 min at RT. Afterward absorbance levels were determined at λ = 595 nm in triplicates. Protein concentration was then calculated for the obtained absorbance levels based on the calibration curve done with bovine serum albumin (Supplementary Figure 1).

Protein samples (50 – 100 μg) were mixed with 0.25 volumes of Laemmli buffer (Table 1) and incubated for 10 min at 95°C. If necessary, proteins were alkylated by addition of 1 μL of a 40% acrylamide solution and incubation at RT for 30 min. Proteins were applied to a 12% Acryl-/Bisacrylamid SDS-gel (Table 1) and separation was performed with a SDS-running buffer (Table 1) and a 5% stacking gel (Table 1) for 15 min at 100 V and 1h at 180 V. The Page Ruler^TM Prestained Protein Ladder (Table 1) was used. After separation gels were washed 3 times with Millipore water for 10 min with subsequent treatment with a fixation solution (Table 1). Staining was performed with 20 mL PageBlue Protein Staining Solution (Table 1). Destaining was done 3 times with 200 mL Millipore water each for 15 min.

Six matrices were prepared: 9-aminoacridine (AA), α-cyano-4-hydroxycinnamic acid (CHCA), 2,5-dihydroxybenzoic acid (DHB), 6-Aza-2-thiothymine (ATT), 3-hydroxypicolinic acid, 2,4,6-Trihydroxyacetophenone (THAP), and 2,5-diaminobenzophenone (DABP) (Table 1). In addition, a final concentration of 5 mg/mL L-Fucose was added to every matrix solution. Every matrix was then mixed with 30 pmol/μL antimiR oligonucleotide 24 and spotted on a MALDI target plate and analyzed. Analysis was performed in linear negative- as well as in linear positive detection mode for all matrix/antimiR oligonucleotide solutions. Reflector detection mode and linear positive detection mode did not yield any evaluable results. Only HPA and THAP could show evaluable results in linear detection mode. Best signal intensities were obtained using THAP in linear negative detection mode (Supplementary Figure 2). These results indicated best ionization efficiency for THAP as ionization matrix and thus it was decided to use THAP for further analysis.

Full extraction of an analyte from a biological sample is crucial for quantification. Since phosphorothioate modified oligonucleotides were reported to strongly bind to plasma proteins as well as cell surface proteins, a proteinase K digestion was used to digest those proteins and thus releasing oligonucleotides and decreasing the background noise originated from complex matrix components (Beltinger et al., 1995; Geary et al., 2001). The workflow consisted of two main steps: a) protein digestion at 55°C using proteinase K and b) glycogen-facilitated precipitation of the oligonucleotide. Subsequently, oligonucleotides were measured by MALDI-MS (Figure 1). Effectiveness of protein removal is shown in Supplementary Figures 3, 4.

For extraction of antimiR oligonucleotides from complex matrices four steps were necessary. With being the most important step, Proteinase K digestion was followed by precipitation of antimiR oligonucleotides with subsequent resuspension and MS analysis.

For assay validation purposes, the LOD, LLOQ and area of quantification (AOQ) values for detection of antimiR132 were determined. Validation of the isolation protocol was performed by adding antimiR132 at various concentrations (10, 5, 2, 1, 0.5, 0.25, and 0.1 pmol) together with 10 pmol antimiRScr as internal standard to murine plasma from untreated mice and measured with the developed assay. Analysis was performed with 10 replicates for determination of the LOD being at 0.25 pmol and three replicates for determination of the LLOQ and the AOQ. The LLOQ was determined to be at 0.25 pmol. AOQ was determined to be linear in the range from 0.25 up to 25 pmol (Figures 2, 3).

For the quantification of antimiR132 in murine plasma, 20 mg/kg antimiR132 was delivered i.v. to 21 mice. At seven timepoints (3 min; 30 min; 1 h; 9 h, 1 days; 3 days; 1 week) blood samples were taken from which 100 – 200 μL plasma were obtained. AntimiRScr was added as an internal standard in a final concentration of 10 pmol/μL. By increasing the NaCl concentration from 100 mM to 250 mM in the Proteinase K buffer, a higher efficiency of the enzymatic reaction was achieved. The quantification of antimiR oligonucleotides was carried out as described previously. The plasma concentrations determined for each time were averaged over all three biological replicates with the standard deviation and the coefficient of variation (CV) being determined (Figure 4).

Using the MALDI-MS analysis, a plausible kinetic could be determined. The coefficient of variation (CV) varied from 2.4 to 18.7% and the peak concentration was calculated to 2.124 pmol/μL plasma. The plasma concentration decreased rapidly and plasma half-life of antimiR132 could be determined to be approximately 2 h. However, at least 9 h post treatment no antimiR132 was detected (Figure 5).

To evaluate the pharmacokinetic properties of antimiR132, further blood plasma samples from pig taken 15 and 30 min; 1, 6, 12, and 24 h after application were examined for the presence of antimiR132 and subsequently quantified (Figure 5).

Using MALDI-MS, a plausible kinetic could be determined. The peak concentration could be calculated at 2.124 pmol/μL. As expected, the plasma concentration decreased rapidly with the plasma half-life of antimiR132 being after approximately 2 h with no longer detection after 9 h. The concentration fell below the LOD and LLOQ after 12 h (Figure 4).