Safety and pharmacodynamics of an engineered E. coli Nissle for the treatment of phenylketonuria: a first-in-human phase 1/2a study.

Phenylketonuria (PKU) is a rare disease caused by biallelic mutations in the PAH gene that result in an inability to convert phenylalanine (Phe) to tyrosine, elevated blood Phe levels and severe neurological complications if untreated. Most patients are unable to adhere to the protein-restricted diet, and thus do not achieve target blood Phe levels. We engineered a strain of E. coli Nissle 1917, designated SYNB1618, through insertion of the genes encoding phenylalanine ammonia lyase and L-amino acid deaminase into the genome, which allow for bacterial consumption of Phe within the gastrointestinal tract. SYNB1618 was studied in a phase 1/2a randomized, placebo-controlled, double-blind, multi-centre, in-patient study ( NCT03516487 ) in adult healthy volunteers (n = 56) and patients with PKU and blood Phe level ≥600 mmol l-1 (n = 14). Participants were randomized to receive a single dose of SYNB1618 or placebo (part 1) or up to three times per day for up to 7 days (part 2). The primary outcome of this study was safety and tolerability, and the secondary outcome was microbial kinetics. A D5-Phe tracer (15 mg kg-1) was used to study exploratory pharmacodynamic effects. SYNB1618 was safe and well tolerated with a maximum tolerated dose of 2 × 1011 colony-forming units. Adverse events were mostly gastrointestinal and of mild to moderate severity. All participants cleared the bacteria within 4 days of the last dose. Dose-responsive increases in strain-specific Phe metabolites in plasma (trans-cinnamic acid) and urine (hippuric acid) were observed, providing a proof of mechanism for the potential to use engineered bacteria in the treatment of rare metabolic disorders.

Characterization of an engineered live bacterial therapeutic for the treatment of phenylketonuria in a human gut-on-a-chip.

Engineered bacteria (synthetic biotics) represent a new class of therapeutics that leverage the tools of synthetic biology. Translational testing strategies are required to predict synthetic biotic function in the human body. Gut-on-a-chip microfluidics technology presents an opportunity to characterize strain function within a simulated human gastrointestinal tract. Here, we apply a human gut-chip model and a synthetic biotic designed for the treatment of phenylketonuria to demonstrate dose-dependent production of a strain-specific biomarker, to describe human tissue responses to the engineered strain, and to show reduced blood phenylalanine accumulation after administration of the engineered strain. Lastly, we show how in vitro gut-chip models can be used to construct mechanistic models of strain activity and recapitulate the behavior of the engineered strain in a non-human primate model. These data demonstrate that gut-chip models, together with mechanistic models, provide a framework to predict the function of candidate strains in vivo.