The Challenge
- Seeking antibodies with similar function as a clinical benchmark Ab
- Also find broadly reactive antibodies to 4 related but different protein targets
- Aiming to shorten downstream timelines screening antibody hits
The Single Cell Solution
- Implement a competition assay in-line to discovery with a functional benchmark antibody
- Screen all antibodies in parallel with the competition assay
Highlights
- Over 2,000 antibody hits discovered from 100M donor human PBMCs
- Competition to clinical benchmark antibody determined for all hits in parallel
Problem: Taking on the Challenge
You may have a benchmark, or even a clinical-stage antibody, with a well-characterized functional profile. The next objective is to identify antibodies that compete with this benchmark for antigen engagement to quickly bin them into possible functional antibodies and likely non-functional.
In this case study, human peripheral blood cells (PBMCs) from donors recently vaccinated with the seasonal influenza vaccine were comprehensively profiled using AbTheneum screening. The campaign focused on identifying antibodies that were cross-reactive to a highly diverse panel of hemagglutinin (HA) proteins and that competed with a clinical-stage, flu-neutralizing antibody, MEDI8852.
Solution: Finding a Better Way
AbTheneum combines multiple assays during each antibody campaign.
After obtaining activated memory B cells from human PBMCs using a commercial kit, cells were deposited onto a picoliter device. 3 capture slides were generated to screen all antibodies for binding to 4 HA proteins and competition with MEDI8852. The 4 HA proteins from a different viral subtype were selected based on their vastly divergent sequences:
- A/California/7/2009 HA,
- A/Vietnam/ 1194/2004 HA,
- A/California/7/2004,
- A/Shanghai/ 1/2013 HA,
which are named in this work H1N1, H3N2, H5N1, and H7N9, respectively.
Over 2,000 anti-HA antibody hits were discovered from 100M donor PBMCs, with 15 mAbs cross-reactive to all 4 HAs. AbTheneum screened for cross-reactivity to all 4 HA proteins and MEDI8852 competition (Figure 1).
Figure 1. Screening plan and small cropped region from the same area on all 3 slides across 8 screening conditions. Slides were scanned on a fluorescent slide scanner and aligned across all slides and images.
Just want the highlights?
All sequences are captured in AbTheneum and sequenced by Next Generation Sequencing (NGS). 5 unique mAbs were expressed (Figure 2) with diverse binding profiles. The cross-reactivity screening results against the 4 HA proteins was 100% confirmed by ELISA. The competition assay against MEDI8852 antibody also
confirmed. Ab3-009 (Fig. 2, purple) shows cross-reactivity against all 4 HA proteins in the study and competes with MEDI8852, meeting target goals.
Figure 2. Five antibodies with diverse binding profiles were expressed and tested by ELISA against all HA proteins and competition with MEDI8852. Small image cutout shows AbTheneum screening result compared to curves generated by ELISA. Negative MEDI8852 screening result means antibody blocks/competes with MEDI8852.
Impact: What We Achieved
This case study demonstrates AbTheneum enables rapid filtering from thousands of antibodies to find mAbs with rare properties. It also offers the advantage of layering multiple screens, delivering layers of data on top of the natively paired full-length sequences from human cells. The ELISA data confirmed the screening data captured by AbTheneum, showcasing the predictive power of the cross-reactivity and competition assays.
Thinking about a similar study?
