Nano and Micro Technologies
Stephanie Otten, Bachelor of Science
Scientist
Merck
Cambridge, Massachusetts, United States
Stephen Kasper
Associate Principal Scientist
Merck, United States
Brian Squadroni
Associate Principal Scientist
Merck, United States
Yi Kuang
Senior Scientist
Merck, United States
Bonnie Howell
Vice President of Research Science
Merck, United States
Andrea Peier
Senior Director
Merck, United States
Xi Ai
Principal Scientist
Merck, United States
Jason Cassaday
Senior Principal Scientist
Merck, United States
Cyclic peptides are a compelling modality to target notoriously difficult-to-drug intracellular protein-protein interactions (PPIs), due to their size and ability to bind flat PPI interfaces with exceptional affinity and selectivity. However, cyclic peptides typically lack cell permeability, which hampers the ability to functionally characterize them in cell-based assays within drug discovery screening efforts. Advances in microfluidics have enabled permeabilization of the cytoplasmic membrane by physical cell deformation (i.e., mechanoporation), resulting in intracellular delivery of impermeable macromolecules in vector- and electrophoretic-free approaches. Nevertheless, the throughput of these approaches is limited by having to pre-mix cells and payloads, a manually intensive process. Therefore, we designed a system that would integrate a microfluidic mechanoporation chip with inline microplate-based dispensing to increase throughput to a scale compatible with downstream cell-based assays.
To characterize the delivery efficiency of low molecular weight macromolecules and subsequent membrane resealing, we used a fluorescent tracer, AlexaFluor488-dextran (3 kDa). Next, we built a system, DµVS (dispensing-microfluidic vortex shedding), integrating a microfluidic vortex shedding (µVS) chip with inline microplate-based dispensing, syncing an electronic pressure regulator, flow rate sensor, on/off dispense tip valve, and an x-y motion control platform in a software-driven feedback loop. To validate the DµVS system, we leveraged two orthologous cell-based assays and a set of peptides directed at MDM2, a negative regulator of the tumor suppressor p53. We characterized the cytosolic delivery of an impermeable peptide using a click chemistry- and NanoBRET-based cell permeability assay in 96-well format. Furthermore, we used a p53 reporter cell assay to identify functional activity of otherwise cell-inactive MDM2-binding peptides in 96- and 384-well format.
When cells were pre-mixed with AF-488-dextran, µVS processing substantially increased the number of dextran-positive cells (94.6 ± 5.2% and 85.0 ± 15.1%, in HeLa and HCT116, respectively). Both cell lines rapidly resealed and when fitted to a one-phase decay model, HeLa displayed a resealing t1/2 of 2.8 min and HCT116 resealed with t1/2 of 1.1 min. This resealing time established a brief window (< 1 min) in which cells can be processed (without a payload) by DµVS and quickly shuttled to microplates containing various payloads (i.e., peptides) at different concentrations. The DµVS system accurately (< 10 %CV) filled 96-well and 384-well plates with mechanoporated cells in approximately 2.5 and 5.5 min, respectively. Using a high-throughput cell-based permeability assay downstream of DµVS, we observed increased intracellular levels of cyclic peptides directed at p53/MDM2, including an impermeable anionic peptide, MP-950. In the, MP-950 showed a clear dose response and low-micromolar potency (EC50 = 2.4 µM), compared to displaying no functional activity in unprocessed cells. Furthermore, we tested a broader set of 15 peptides in the DmVS-p53 reporter assay and observed that the peptides generally fall into three groups: (i) cell-active peptides whose activities are unchanged by DμVS processing, (ii) peptides that show enhanced activity in DμVS-processed cells, and (iii) peptides that show no activity in either format. Peptides that showed enhanced activity with DµVS processing were confirmed via dose response studies.
In conclusion, to improve on the existing throughput of microfluidic mechanoporation methodologies, we integrated a μVS chip with inline microplate-based dispensing, in a system we call DμVS. We designed and constructed the DμVS system to rapidly shuttle permeabilized cells to standard microplates containing various peptides at different concentrations. We then validated the DμVS method using p53/MDM2 peptides in both a click-chemistry-based permeability assay, as well as a functional p53 reporter assay. This combination of vector-free ICD, microfluidics, and high throughput-amenable cell-based assays, could have a broad impact in drug discovery, enabling the development of novel modalities against challenging intracellular targets.