Scale up biologics manufacturing

Description of proposed project

Manufacturing many therapeutic, precision fermentation and industrial proteins remains a bottleneck: half of biologics fail to scale at the manufacturing stage, and even “successful” proteins often require months to years of strain and process optimization. In developing nations important medicines like insulin, erythropoietin (EPO), and recombinant Hep B vaccines face supply issues due to manufacturing cost and under-developed cold chains. With many drugs coming off patent, cheap manufacturing of biosimilars would broaden access to these important medicines. Yeast has long been used as a production platform for biologics and other proteins due to their cheaper media requirements and lack of viral contamination risk, but relatively poor titers and non-human post-translational modifications has led them to be disfavored over the more costly mammalian CHO cells.

The ultimate goal of the company is to discover and evolve genes which enable cheap, efficient, facile manufacturing of important proteins; changing the biomanufacturing landscape by enabling production of previously impossible proteins, and enabling facile, cheap production of current proteins. I am trying to start a company utilizing a continuous directed evolution platform I developed (OrthoRep Assisted Continuous Library Evolution, ORACLE) which continuously mutagenizes large gene libraries in yeast, allowing for unmatched search of sequence space when placed under evolutionary pressure. I have already shown that unexpected or unknown genes evolve to solve imposed evolutionary constraints in a way that is, to the best of my knowledge, unprecedented. To apply this approach to protein manufacturing, I need to build systems which couple secretion of a protein of interest to cell growth or fluorescence.

By iterating selection in pooled cultures and switching between orthogonal screening methods (growth-based, FACS, droplet sorting), ORACLE should uncover non-obvious chaperones, trafficking factors, and detoxifiers that increase titers of both tractable and previously unproducible targets.

To start, we will:

1. Build and validate general selection tools

2. Run pilot ORACLE campaigns on a representative set of 8–12 proteins, to isolate evolved genes that boost secretion proxies by more than 2x.

3. Characterize and port top hits for mechanism (folding vs. trafficking), combine effectors in polycistronic cassettes, and test portability in CHO and Pichia pastoris, aiming for ≥25% of hits active in all systems.

4. Contract a 1–5 L bench-scale fermentation at a CRO to validate that evolved strains recapitulate shake-flask titer gains in bioreactors, measuring purity by ELISA/HPLC and glycosylation fidelity by lectin assays.

The initial goal is to establish a rapid, generalizable discovery engine for secretion-enhancing genes, shortening the strain development timeline from ~90 days to ~30 days, lowering costs, and unlocking production of complex and post-translationally modified proteins currently out of reach.

Given preliminary, still unpublished data generated in my final months of grad school I have a high degree of confidence this system will be productive. To achieve our long term goals of transforming biomanufacturing, this project will require more investment but a relatively small up front amount will allow me to get started, delivering proof of concept and building momentum for further efforts.

Why are you qualified to work on this?

I, Alexander Pisera, am the founder of Synlibris, Inc (the C-corp I founded prior to operations for grant purposes). I completed my PhD at UC Irvine as of Dec 2024 and am the developer of the core platform, ORACLE. During the final two years of my PhD I conceived, built, and validated the platforms’ ability to produce evolved genes inaccessible to other means. I am the sole first author on the paper describing ORACLE, and two other papers on OrthoRep / evolution related work.

I am the first named inventor on the patent covering the high efficiency P1 integration system, and a named inventor on the patent applying the mating based P1 transfer to gene editor evolution. Upon securing funding I will be working on the company full time, developing the platform for application to the problem of protein expression. I have over a decade of bench experience, realistic expectations of timelines enabled by ORACLE, and a demonstrated ability to deliver well-characterized functional systems on rapid timelines.

I have hands on experience in yeast, e.coli, and mammalian settings and have optimized complex protocols in all, with a wealth of experience for molecular biology assays, next-gen sequencing, and advanced techniques such as multicolor flow cytometry, emulsion PCR, lentiviral production, GPCR signaling assays, and primary cell handling. Prior to graduate school I spent three years in the biotech industry and developed a network there, as well as maintaining connections with undergraduate professors at Johns Hopkins University. During graduate school, I built a network of connections as a PD Soros Fellow. My company has just gone through the Blueprint program at The Engine, and is in Harvard Innovation Labs.

I am the sole founder am screening co-founders.

What would you do if not funded?

I’ve applied to NSF, NIH SBIR, and am working on trying to find other grants.

It's no secret the biotech VC fundraising environment is bleak for early stage work so I am not optimistic on traditional venture means, but starting to put more effort in there.

Being pretty fresh out of grad school (did flip a condo during it) I only have 6 months of financial runway before I have to move on. Frankly, if I don't do it no one will; it's a complicated system to work with.

Supporting documents

ORACLE paper: https://www.biorxiv.org/content/10.1101/2025.03.22.644768v2

Recent pubs: https://www.nature.com/articles/s41559-025-02792-7#Abs1

https://pubs.acs.org/doi/abs/10.1021/acssynbio.4c00786