Background

The bacterial cytosol microenvironment is home to thousands of diverse reactions and is tightly regulated in a reducing redox state. This poses challenges in recombinant bacterial protein overexpression as the reducing environment can prevent the correct formation of critical disulphide bonds in proteins that require them for structure and/or function (Kadokura and Beckwith, 2010). As such, these proteins are often expressed by E. coli into insoluble inclusion bodies (Berkmen, 2012) that renders them unusable. In favorable cases, proteins from inclusion bodies can be refolded in vitro after solubilizing in strong denaturants (Qin et al., 2015). However, this is typically a lengthy and reagent consuming process and the yield of the correctly refolded protein is variable.

One alternative to conventional recombinant bacterial overexpression is cell-free synthesis, where proteins can be expressed using the bacterial transcription and protein production machinery, but without the confines of a cell (Apponyi et al., 2008; Ozawa and Loh, 2014; Gao et al., 2019; Dondapati, 2020). Cell-free protocols are particularly useful in the expression of cysteine-rich proteins, as the redox environment can be controlled by simply changing the ratios of reduced and oxidized glutathione (GSH:GSSG) and varying the concentration of reducing agents such as dithiothreitol (DTT). After optimization, the yield of protein with correctly formed disulphide bonds suitable for downstream experiments (e.g., binding studies, structural and biophysical characterisation) can be greatly increased.

Beyond proteins with disulphide bonds, cell-free protein synthesis systems can hold additional advantages over traditional recombinant protein production such as the ability to readily incorporate unnatural amino acids (Gao et al., 2019; Dondapati et al., 2020). In addition, proteins that are too toxic for a host cell can be expressed in a cell-free environment as there is no requirement to maintain cell viability and this has allowed high throughput production and screening for membrane proteins, enzymes, and other therapeutics (Khambhati et al., 2019).

Our optimized cell-free expression protocol allows direct control of the redox conditions in the reaction mixture and demonstrates a distinct advantage over recombinant expression for a class of proteins known as the fungal hydrophobins (Siddiquee et al., 2020). This class of proteins is characterized by four disulphide bonds and the correct bonding pattern is required for hydrophobins to self-assemble into amphipathic layers at hydrophilic:hydrophobic interfaces (Sunde et al., 2008; Ball et al., 2020). These unique assemblies have suggested a range of biotechnological applications from drug delivery systems to coating implants and the engineering of surfaces to increase biocompatibility (Wosten and Scholtmeijer, 2015; Berger and Sallada, 2019). While many hydrophobins can be overexpressed in E. coli to a very high yield, the protein is almost exclusively found misfolded within inclusion bodies and cannot be used for downstream applications. Therefore, a long process of in vitro refolding and multi-step purification is typically required to separate the correctly folded protein from the partially folded and misfolded species (Kwan et al., 2006; Pille et al., 2015; Pham et al., 2016). To overcome this problem, we have developed a cell-free expression system for hydrophobin production based on the published S30 E. coli lysate (Apponyi et al., 2008).

The S30 lysate is one of the most used extracts in cell free synthesis of proteins and contains the cellular transcriptional and translational machinery. A Pubmed search with “S30 cell free” as search terms gives >150 articles. While the S30 lysate can be purchased commercially (e.g., from Promega), it can be produced readily in laboratories that perform recombinant protein expression and purification at a fraction of the cost of commercially available cell-free expression kits. This cell-free reaction involves incubating the lysate and other key cofactors with a gene of interest cloned downstream of a constitutive promoter in a dialysis set up. The dialysis set up allows continuous exchange of lower molecular reagents (e.g., amino acids, rNTPs, ATP) and offers higher protein yields per unit of components that are the more expensive or time-consuming to prepare (e.g., transcription and translation machinery in the S30 extract as well as plasmid DNA).

To aid purification, we have cloned the gene into the vector pET-MCSIII that has a hexa-histidine affinity tag but lacks the lac operator (Neylon et al., 2000). The protocol presented herein can be easily amended to enable the expression of other proteins with disulphide bonds. The apparatus used in this protocol can be easily prepared using disposable plastic tubes commonly used in most scientific laboratories and can be scaled up as required. The efficiency and low cost of this set up allow multiple proteins (e.g., a protein and its mutants) to be produced in parallel cell-free reactions using one preparation of dialysis buffer. Isotopically labeled proteins can also be easily produced by substituting in labeled amino acids in the reaction mixture. In some cases, this has allowed Nuclear Magnetic Resonance (NMR) and mass spectroscopy experiments to be carried out on the protein of interest within one day of setting up the reaction and without the need for further purification.