Non-Invasive Liquid Biopsy

Frauke Henjes1, Lise Aagaard1, Darren Ellis1, Hannah Lindstroem1, Morten Oksvold2,3, Axl Neurauter1, Ketil W. Pedersen1. 1 Life Technologies AS, Ullerchausséen 52, 0379, Oslo, Norway; 2Department of Immunology, Institute for Cancer Research, Oslo University Hospital, Oslo, Norway; 3 Centre for Cancer Biomedicine, University of Oslo, Oslo, Norway

Introduction Invasive biopsies could often be replaced by simpler and safer non-invasive liquid biopsies using plasma or blood samples. A non-invasive liquid biopsy permits the analysis of multiple circulating biomarkers in one sample and facilitates the discovery of disease as well as treatment response monitoring in applications such as oncology. Here, we demonstrate that paramagnetic Dynabeads® provide a versatile and automation friendly tool for fast, sensitive, and high throughput isolation that can be tailored to isolate specific circulating biomarkers such as cells, exosomes, and cell-free nucleic acids. Circulating Tumor Cells (CTCs) Positive Isolation Negative Isolation For the positive isolation, Dynabeads® coupled with a monoclonal antibody targeting EpCAM CTCs can also be enriched by applying negative isolation. The advantage is that it is (epithelial cell adhesion molecule) are used to capture CTCs from whole blood samples. For independent of surface marker expression (e.g. EpCAM) on the CTCs. To deplete ® ® downstream applications, mRNA can be isolated using Dynabeads Oligo (dT)25 and white blood cells, Dynabeads CD45 are used. These beads deplete more than 95 % analyzed using gene expression analysis (Figure 1E) or sequencing. of the leukocytes from mononuclear cells (MNCs) (Figure 2), whole blood, and buffy coat. Addition of Dynabeads® CD15 could further increase cell depletion. Following A B C D negative isolation, downstream applications could include cell culturing and molecular analysis such as qPCR and sequencing.

A B Figure 2: Fluorescence histograms for CD45- Figure 1: A-D) Workflow for positive isolation of circulating tumors cells. E stained MNCs with inlaid ® A) Epithelial cells are captured from a complex matrix using Dynabeads scatter plots, measured Epithelial Enrich and lysed directly on-beads. B) After removal of beads, lysate before (A) and after (B) is mixed with Dynabeads® Oligo (dT) for mRNA isolation. C) Eluted mRNA is 25 depletion of CD45-positive reverse transcribed and analyzed using qRT-PCR. D) Expression of CK19 is cells using Dynabeads® measured to quantify epithelial cells. CD45. E) For proof-of-principle, 1 (blue), 3 (green), 10 (red), and 0 (black) SKBR3 breast carcinoma cells were spiked-into 1x106 MNCs and processed as described above. Mean Ct-values of biological replicates were taken, and the fold-changes of cDNA content were calculated in relation to 1 cell.

Exosomes A C Figure 3: Flow cytometric (FC) and Western immunoblot analysis of exosomes isolated with Dynabeads® coated with anti-human CD81 antibodies. Exosomes were immunoisolated from pre-enriched exosomes using magnetic beads coated with anti-human CD81. Exosomes were stained for B-cell surface antigens (CD20, HLA-DR) and exosomal antigens (CD63 CD81) followed by FC B analysis and presented as median fluorescence intensity (MFI). A) Scatterplot and fluorescence histogram of anti-CD81 magnetic beads in the absence or presence of exosomes labeled for CD81. B) CD81+ exosomes were stained for CD20 and HLA-DR in addition to the Figure 4: exosomal antigens CD81 and CD63. One of 3 representative experiments is A) Transmission Electron Microscopy (TEM) of pre-enriched shown. Sudhl4 exosomes labeled for CD63 2). CD63 was detected using C) CD81+ exosomes were lysed and processed for Western immunoblotting anti-CD63 antibodies followed by protein A gold (10 nm). analysis. Expression of the B-cell surface proteins (CD19, HLA-DR, CD45, B) TEM of exosomes isolated with Dynabeads® Human CD63 and CD20) and exosome markers (CD9, CD81, and CD63) were analyzed. Isolation/Detection (human cell culture media)1). Bar 500 nm. Ref. Oksvold et al., 2014, Clinical Therapeutics Vol. 36 (6).

Circulating Nucleic Acids In order to isolate sufficient amounts of cfDNA for downstream analysis, Figure 7: Four mL plasma samples from 10 individuals Figure 6: SILANE cfDNA Isolation from 4 mL Plasma or Serum (A-J) and pooled plasma derived from Cell-Free DNA typically 4 to 10 mL plasma or serum is necessary to be processed into BCT tubes (St1 and 12) processed manually and on an Lysis Bind Wash Air Dry Elute volumes below 100 µL. During processing, the work volumes can increase automated setting (Figure 5). The resulting eluates Sample Lysis Buffer Isopropanol 2x Wash Buffers ≤100 µL Elution over 3 fold to that of the input volume. Processing is therefore mostly done were then quantified by qPCR with fragment specific manually and is highly work and skill intensive, while automation presents probes. 6 A/B) Recovery rates of samples laced with 10 copies A A. 120 bp Fragment (%) A B B. 220 bp Fragment (%) A a serious challenge. Here we provide a solution using magnetic beads, 140 B 160 B of 120 bp (A) and 220 bp DNA (B) (~100%). C C ® TM 120 140 Dynabeads MyOne SILANE, in a highly sensitive, high-throughput C) Recovery rates of endogenous cfDNA of manual D D 120 100 E E automatable large volume sample processing. and automated sample processing in comparison to F 100 F 80 that of a non-automatable Competitor. Dynabeads® G 80 G 60 H H ® TM ® ® TM 60 Figure 5: Dynabeads MyOne SILANE Automation on Tecan Freedom Evo 150 I I MyOne SILANE yielded equally efficient recovery 40 40 Variomag Teleshake 95 manually and automated, albeit yielding slightly lower J J Sample processing plate (Inheco) Lysis/Binding Buffer 20 St1 20 St1 Liquid waste Custom magnet (48 deep-welled plate) Proteinase K St12 St12 Isopropanol recovery rates than that of Competitor (87% and 83% 0 0 Manual Tecan Manual Tecan ® Mean Mean Split Sample manually and on Tecan instrument, respectively, to A (4x 1 mL samples) C C. Total cfDNA Recovery (ng) D D. Total Protein Content (µg) A Competitor). B Final eluate 90 50 B 1.5 mL tubes Wash Buffer 2 (1st wash) C D) As a guideline for sample eluate purity, the total 80 45 C D 40 ® TM 70 D E 35 protein contents of Dynabeads MyOne SILANE 60 E F 30 F 50 Wash Buffer 2 Sample Pooling 4-to-2 manual and automated extractions and Competitor G (combine rows: 1>2, 3>4, 5>6, 7>8) 25 G 40 H Wash Buffer 1 20 H nd were compared. The lowest mean total protein content Wash Buffer 2 (2 wash) 30 I ® TM 15 I was achieved with Dynabeads MyOne SILANE 20 J 10 J St1 10 5 St 1 manual processing. Competitor gave medium protein St12 0 0 St Competitor Manual Tecan Mean Competitor Manual Tecan 12 LiHa disposable Tip disposal Start position: Final eluate 96- Variomag Thermoshake Dynabeads® Elution Buffer Sample Pooling 2-to-1 content but showed high sample to sample variation. TM (combine rows: 2>4, 6>8) Filtered tips Standard 13 mL tubes welled PCR plate (Inheco) MyOne SILANE

The Dynabeads® products referenced in this article are For Research Use Only. Not for use in diagnostic procedures. Thermo Fisher Scientific • 5791 Van Allen Way • Carlsbad, CA 92008 • lifetechnologies.com