Optimization of biological networks is often limited by wet lab labor and cost, and the lack of convenient computational tools. Here, we describe METIS, a versatile active machine learning workflow with a simple online interface for the data-driven optimization of biological targets with minimal experiments. We demonstrate our workflow for various applications, including cell-free transcription and translation, genetic circuits, and a 27-variable synthetic CO2-fixation cycle (CETCH cycle), improving these systems between one and two orders of magnitude. For the CETCH cycle, we explore 1025 conditions with only 1,000 experiments to yield the most efficient CO2-fixation cascade described to date. Beyond optimization, our workflow also quantifies the relative importance of individual factors to the performance of a system identifying unknown interactions and bottlenecks. Overall, our workflow opens the way for convenient optimization and prototyping of genetic and metabolic networks with customizable adjustments according to user experience, experimental setup, and laboratory facilities.

Enzymatic carbon dioxide fixation is one of the most important metabolic reactions as it allows the capture of inorganic carbon from the atmosphere and its conversion into organic biomass. However, due to the often unfavorable thermodynamics and the difficulties associated with the utilization of CO2, a gaseous substrate that is found in comparatively low concentrations in the atmosphere, such reactions remain challenging for biotechnological applications. Nature has tackled these problems by evolution of dedicated CO2-fixing enzymes, i.e., carboxylases, and embedding them in complex metabolic pathways. Biotechnology employs such carboxylating and decarboxylating enzymes for the carboxylation of aromatic and aliphatic substrates either by embedding them into more complex reaction cascades or by shifting the reaction equilibrium via reaction engineering. This review aims to provide an overview of natural CO2-fixing enzymes and their mechanistic similarities. We also discuss biocatalytic applications of carboxylases and decarboxylases for the synthesis of valuable products and provide a separate summary of strategies to improve the efficiency of such processes. We briefly summarize natural CO2 fixation pathways, provide a roadmap for the design and implementation of artificial carbon fixation pathways, and highlight examples of biocatalytic cascades involving carboxylases. Additionally, we suggest that biochemical utilization of reduced CO2 derivates, such as formate or methanol, represents a suitable alternative to direct use of CO2 and provide several examples. Our discussion closes with a techno-economic perspective on enzymatic CO2 fixation and its potential to reduce CO2 emissions.

One of the biggest challenges to realize a circular carbon economy is the synthesis of complex carbon compounds from one-carbon (C1) building blocks. Since the natural solution space of C1−C1 condensations is limited to highly complex enzymes, the development of more simple and robust biocatalysts may facilitate the engineering of C1 assimilation routes. Thiamine diphosphatedependent enzymes harbor great potential for this task, due to their ability to create C−C bonds. Here, we employed structure-guided iterative saturation mutagenesis to convert oxalyl-CoA decarboxylase (OXC) from Methylobacterium extorquens into a glycolyl-CoA synthase (GCS) that allows for the direct condensation of the two C1 units formyl-CoA and formaldehyde. A quadruple variant MeOXC4 showed a 100 000-fold switch between OXC and GCS activities, a 200-fold increase in the GCS activity compared to the wild type, and formaldehyde affinity that is comparable to natural formaldehyde-converting enzymes. Notably, MeOCX4 outcompetes all other natural and engineered enzymes for C1−C1 condensations by more than 40-fold in catalytic efficiency and is highly soluble in Escherichia coli. In addition to the increased GCS activity, MeOXC4 showed up to 300-fold higher activity than the wild type toward a broad range of carbonyl acceptor substrates. When applied in vivo, MeOXC4 enables the production of glycolate from formaldehyde, overcoming the current bottleneck of C1− C1 condensation in whole-cell bioconversions and paving the way toward synthetic C1 assimilation routes in vivo.

Formate can be envisioned at the core of a carbon-neutral bioeconomy, where it is produced from CO2 by (electro-)chemical means and converted into value-added products by enzymatic cascades or engineered microbes. A key step in expanding synthetic formate assimilation is its thermodynamically challenging reduction to formaldehyde. Here, we develop a two-enzyme route in which formate is activated to formyl phosphate and subsequently reduced to formaldehyde. Exploiting the promiscuity of acetate kinase and N-acetyl-γ-glutamyl phosphate reductase, we demonstrate this phosphate (Pi)-based route in vitro and in vivo. We further engineer a formyl phosphate reductase variant with improved formyl phosphate conversion in vivo by suppressing cross-talk with native metabolism and interface the Pi route with a recently developed formaldehyde assimilation pathway to enable C2 compound formation from formate as the sole carbon source in Escherichia coli. The Pi route therefore offers a potent tool in expanding the landscape of synthetic formate assimilation.

Initiation of RNA synthesis by the hepatitis C virus (HCV) RNA-dependent RNA polymerase (RdRp) NS5B has been extensively studied in vitro and in cellulo. Intracellular replication is thought to rely exclusively on terminal de novo initiation, as it conserves all genetic information of the genome. In vitro, however, additional modes of initiation have been observed. In this study, we aimed to clarify whether the intracellular environment allows for internal initiation of RNA replication by the HCV replicase. We used a dual luciferase replicon harboring a terminal and an internal copy of the viral genomic 5= untranslated region, which was anticipated to support noncanonical initiation. Indeed, a shorter RNA species was detected by Northern blotting with low frequency, depending on the length and sequence composition upstream of the internal initiation site. By introducing mutations at either site, we furthermore established that internal and terminal initiation shared identical sequence requirements. Importantly, lethal point mutations at the terminal site resulted exclusively in truncated replicons. In contrast, the same mutations at the internal site abrogated internal initiation, suggesting a competitive selection of initiation sites, rather than recombination or templateswitching events. In conclusion, our data indicate that the HCV replicase is capable of internal initiation in its natural environment, although functional replication likely requires only terminal initiation. Since many other positive-strand RNA viruses generate subgenomic messenger RNAs during their replication cycle, we surmise that their capability for internal initiation is a common and conserved feature of viral RdRps. IMPORTANCE Many aspects of viral RNA replication of hepatitis C virus (HCV) are still poorly understood. The process of RNA synthesis is driven by the RNAdependent RNA polymerase (RdRp) NS5B. Most mechanistic studies on NS5B so far were performed with in vitro systems using isolated recombinant polymerase. In this study, we present a replicon model, which allows the intracellular assessment of noncanonical modes of initiation by the full HCV replicase. Our results add to the understanding of the biochemical processes underlying initiation of RNA synthesis by NS5B by the discovery of internal initiation in cellulo. Moreover, they validate observations made in vitro, showing that the viral polymerase acts very similarly in isolation and in complex with other viral and host proteins. Finally, these observations provide clues about the evolution of RdRps of positive-strand RNA viruses, which might contain the intrinsic ability to initiate internally.