Rapid On-Cell Selection of High-Performance Human Antibodies

Abstract

Phage display is a critical tool for developing antibodies. However, existing approaches require many time-consuming rounds of biopanning and screening of potential candidates due to a high rate of failure during validation. Herein, we present a rapid on-cell phage display platform which recapitulates the complex in vivo binding environment to produce high-performance human antibodies in a short amount of time. Selection is performed in a highly stringent heterogeneous mixture of cells to quickly remove nonspecific binders. A microfluidic platform then separates antigen-presenting cells with high throughput and specificity. An unsupervised machine learning algorithm analyzes sequences of phage from all pools to identify the structural trends that contribute to affinity and proposes ideal candidates for validation. In a proof-of-concept screen against human Frizzled-7, a key ligand in the Wnt signaling pathway, antibodies with picomolar affinity were discovered in two rounds of selection that outperformed current gold-standard reagents. This approach, termed μCellect, is low cost, high throughput, and compatible with a wide variety of cell types, enabling widespread adoption for antibody development.

M13 phage: a versatile building block for a highly specific analysis platform

“…C Schematic illustration of the μCellect methodology. Copyright 2022 American Chemical Society [54] the collision-based force effect with the two-dimensional separation-based concentration effect to increase the biopanning stringency, phages with high-affinity can be screened within only 6 h (Fig. 3B) [53].…”

“…The biopanning stringency can be tuned by changing the ratio of cell types. With the assistance of machine learning and next-generation sequencing (NGS), binders with picomolar affinity can be identified with only two rounds of biopanning [54].…”

Application of microfluidic technology in antibody screening

“…Philpott et al presented a biopanning platform with a throughput of 6 × 10 7 cells per h (μCellect), and high-affinity antibodies against human Frizzled-7 were discovered in two rounds of selection. [81] Similarly, microfluidic panning has been optimized to identify high-affinity multiple-target phage display peptides in one round without requiring any bacterial culture. [82] The entire screening completed in only 4 h, resulting in less time and labor consumption.…”

Unsupervised machine learning leads to an abiotic picomolar peptide ligand

“…Recent examples largely include genetically-encoded discovery methods, where round-to-round enrichment serves as the label in supervised learning analysis. [36][37][38] In contrast to genetically-encoded platforms, AS-MS has unparalleled use of abiotic chemical libraries, meaning its datasets could enable the development of an ML approach to broadly connect chemical space. AS-MS has historically been a screening tool; 39 but has been advanced with target-focused libraries, 16,40 and onward to de novo discovery with peptidomimetics against multiple biomolecular targets with large libraries (>10 8 members).…”