BioFactura (United States)

Abstract We introduce a method to reduce the cost of synthesizing proteins and other biological sequences designed by a generative model by as much as a trillion-fold. In particular, we make our generative models manufacturing-aware, such that model-designed sequences can be efficiently synthesized in the real world with extreme parallelism. We demonstrate by training and synthesizing samples from generative models of antibodies, T cell antigens and DNA polymerases. For example, we train a manufacturing-aware generative model on 300 million observed human antibodies and synthesize∼1017generated designs from the model, achieving a sample quality comparable to a state-of-the-art protein language model, at a cost of 103dollars. Using previous methods, synthesis of a library of the same accuracy and size would cost roughly a quadrillion (1015) dollars.

As part of the Reproducibility Project: Cancer Biology, we published Registered Reports that described how we intended to replicate selected experiments from 29 high-impact preclinical cancer biology papers published between 2010 and 2012. Replication experiments were completed and Replication Studies reporting the results were submitted for 18 papers, of which 17 were accepted and published by eLife with the rejected paper posted as a preprint. Here, we report the status and outcomes obtained for the remaining 11 papers. Four papers initiated experimental work but were stopped without any experimental outcomes. Two papers resulted in incomplete outcomes due to unanticipated challenges when conducting the experiments. For the remaining five papers only some of the experiments were completed with the other experiments incomplete due to mundane technical or unanticipated methodological challenges. The experiments from these papers, along with the other experiments attempted as part of the Reproducibility Project: Cancer Biology, provides evidence about the challenges of repeating preclinical cancer biology experiments and the replicability of the completed experiments.

Concerns regarding outbreaks of human monkeypox or the potential reintroduction of smallpox into an immunological naïve population have prompted the development of animal models and countermeasures. Here we present a marmoset model of monkeypox and smallpox disease utilizing a relevant poxvirus via a natural exposure route. We found that 1000 plaque forming units (PFU) of Monkeypox virus was sufficient to recapitulate smallpox disease, to include an incubation period of approximately 13 days, followed by the onset of rash, and death between 15 and 17 days. Temporally accurate manifestation of viremia and oral shedding were also features. The number of lesions ranged from no lesions to 299, the most reported in a marmoset exposed to a poxvirus. To both evaluate the efficacy of our antibodies and the applicability of the model system, marmosets were prophylactically treated with two monoclonal antibodies, c7D11 and c8A. Of three marmosets, two were completely free of disease and a single marmoset died 8 days after the mock (n = 1) or PBS control(s) (n = 2). Evaluation of the serum levels of the three animals provided a possible explanation to the animal succumbing to disease. Interestingly, more females had lesions (and a greater number of lesions) and lower viral burden (viremia and oral shedding) than males in our studies, suggesting a possible gender effect.

The precise product quality and lower cost of goods demands of the growing biosimilars industry are driving biomanufacturing innovation. Biosimilar cell lines that produce complex glycoproteins such as monoclonal antibodies must be both highly productive and express a product with critical quality attributes closely matching those of the innovator reference. In this work, a biomanufacturing platform is described that harnesses the commercially-established NS0 host cell in new ways to create stable, highly productive cell lines with characteristics meeting the current demands. A cholesterol metabolic selection marker and implementation strategy that can be generically applied are shown to yield high expressing cell lines as well as eliminate the need for cholesterol addition, which has been a significant barrier in both stainless steel reactors as well as in single-use plastic systems. Additionally, for the first time, a multiplex selection strategy was implemented that served to increase NS0 cell line specific productivity >10-fold and volumetric yields >6-fold. The best overall performing cell line had a Qp of 28.5 picograms per cell per day and was rapidly adapted to a lean production medium. Yields in l-glutamine fed-batch shaker cultures exceeded 500mg/L. An initial screening of four feeding strategies resulted in a final 13-day yield of over 1.4g/L in small shaker culture. Overall, this work shows both the strategy to develop biosimilar cell lines and the commercial potential of a novel expression system highly suited for the manufacture of biosimilars of reference biologics currently produced in murine cells. © 2018 American Institute of Chemical Engineers Biotechnol. Prog., 34:455-462, 2018.

SYM‐1219, a novel oral granule formulation of secnidazole, is under development as single‐dose treatment for bacterial vaginosis. This 4‐way, randomized, crossover study evaluated the effects of SYM‐1219 on electrocardiographic (ECG) parameters in 52 healthy subjects. Subjects were administered single doses of SYM‐1219, 2 g (proposed therapeutic dose), 6 g (supratherapeutic dose), placebo, and moxifloxacin (positive control). Serial digital 12‐lead ECGs were recorded pre‐ and postdose; blood samples were taken to determine plasma secnidazole concentrations. A high‐precision QT technique measured ECGs. The primary end point was change from baseline QTcF (∆QTcF); data were analyzed with the objective of excluding QT effects >10 milliseconds at postdosing time points and with exposure‐response analysis. Safety and tolerability were assessed. Single doses of 2 g and 6 g SYM‐1219 did not have a clinically relevant effect on the QTcF interval; an effect >10 milliseconds could be excluded at all postdosing time points. A shallow slope of the exposure‐response relationship was seen (0.058 millisecond per μg/mL; 90%CI 0.042, 0.073); in this model, the effect on QTc can be predicted to be <10 milliseconds up to a secnidazole plasma concentration of ∼125 μg/mL, approximately 3.4‐fold higher than anticipated peak therapeutic plasma levels. The moxifloxacin QT response demonstrated assay sensitivity. The most frequently reported treatment‐emergent adverse events with SYM‐1219 were headache, dizziness, and nausea. This thorough QT study demonstrated that SYM‐1219 in doses and plasma concentrations up to 3‐fold above therapeutically relevant levels does not have a clinically concerning effect on ECG parameters, including the QT interval.

Many recent advances in bioprocessing have been enabled by developments in miniaturization and microfluidics. A continuing challenge, however, is integrating multiple unit operations that require distinct spatial boundaries, especially with included labile biological components. We have suggested "biofabrication" as a means for organizing cells and biomolecules in complex configurations while preserving function of individual components. Polysaccharide films of chitosan and alginate that are assembled on-chip by electrodeposition are "smart" configurable interfaces that mediate communication between the biological systems and microfabricated devices. Here, we demonstrate the scalable performance of a production address, where incubated cells secrete antibodies, and a capture address, where secreted antibody is retained with specificity and subsequently assayed. The antibody exchange from one electro-address to another exemplifies integrated in-film bioprocessing, facilitated by the integrated biofabrication techniques used. This in-film approach enables complex processes without need for microfluidics and valving. Finally, we have shown scalability by reducing electrode sizes to a 1 mm scale without compromising film biofabrication or bioprocessing performance. The in situ reversible deposition of viable cells, productivity characterization, and capture of secreted antibodies could find use in bioprocessing applications such as clonal selection, run-to-run monitoring, initial scale-up, and areas including drug screening and biopsy analysis.

BackgroundThere is a significant requirement for the development and acquisition of reagents that will facilitate effective diagnosis, treatment, and prevention of Lassa fever. In this regard, recombinant Lassa virus (LASV) proteins may serve as valuable tools in diverse antiviral applications. Bacterial-based systems were engineered for expression and purification of recombinant LASV nucleoprotein (NP), glycoprotein 1 (GP1), and glycoprotein 2 (GP2).ResultsFull-length NP and the ectodomains of GP1 and GP2 were generated as maltose-binding protein (MBP) fusions in the Rosetta strains of Escherichia coli (E. coli) using pMAL-c2x vectors. Average fusion protein yields per liter of culture for MBP-NP, MBP-GP1, and MBP-GP2 were 10 mg, 9 mg, and 9 mg, respectively. Each protein was captured from cell lysates using amylose resin, cleaved with Factor Xa, and purified using size-exclusion chromatography (SEC). Fermentation cultures resulted in average yields per liter of 1.6 mg, 1.5 mg, and 0.7 mg of purified NP, GP1 and GP2, respectively. LASV-specific antibodies in human convalescent sera specifically detected each of the purified recombinant LASV proteins, highlighting their utility in diagnostic applications. In addition, mouse hyperimmune ascitic fluids (MHAF) against a panel of Old and New World arenaviruses demonstrated selective cross reactivity with LASV proteins in Western blot and enzyme-linked immunosorbent assay (ELISA).ConclusionThese results demonstrate the potential for developing broadly reactive immunological assays that employ all three arenaviral proteins individually and in combination.