Facile methods for heterologous production of bacterial microcompartments in diverse host species

Bacterial microcompartments, present in nearly 20% of known bacterial species, are giant proteinaceous structures consisting of a protein shell that encloses a series of sequentially acting metabolic enzymes (Bobik et al., 2015). Their general purpose is to prevent small, volatile or toxic pathway intermediates from being released into the cytosol. Depending on context, bacterial microcompartments are abbreviated in the literature alternatively as MCPs or BMCs; for clarity here (and for keeping with Graf et al.), we use MCP to describe a microcompartment, and we use the widely accepted name BMC for the family of proteins that comprise the main components of MCP shells. A few types of MCPs have been characterized experimentally, beginning first with the carboxysome (which enhances CO2 fixation in cyanobacteria and some chemoautotrophs), and then the Pdu MCP (which allows for the metabolism of 1,2-propanediol without exposing the bacterial cytosol to the reactive intermediate propionaldehyde) (Fig. 1). More broadly, MCPs predicted to sequester diverse metabolic pathways have been identified by bioinformatic analyses and preliminary experimental studies (Abdul-Rahman et al., 2013; Jorda et al., 2013; Axen et al., 2014). Structural studies have played a key role in showing that MCPs are not merely inclusions in bacterial cells, but highly sophisticated molecular machines satisfying the concept of true organelles in bacteria (Kerfeld et al., 2005; Yeates et al., 2011). Revelations about the complexity and sophistication of MCPs have spurred much research into understanding more fully how they evolved, how they operate and how they might be engineered for unique applications. Within the broad scientific area of synthetic biology, diverse molecular systems are currently being pursued as potential frameworks for achieving spatial control over the various activities within cells (Lee et al., 2012). Among these, MCPs provide natural starting points for applications aimed at organizing metabolic processes (Kim and Tullman-Ercek, 2013; Lawrence et al., 2014). They are especially attractive for applications where limited molecular exchange between different subcellular regions is desirable, either to improve channelling of substrates through multiple enzymatic reactions, or to limit the escape of toxic metabolites or even toxic proteins (Yung et al., 2017). In addition, owing to their size – since thousands of protein subunits comprise a single shell – MCPs have internal volumes large enough to encapsulate hundreds of enzyme molecules. This capacity offers an advantage over current strategies for designing novel enclosures by protein engineering; those efforts have so far produced designed protein assemblies with internal volumes one or two orders of magnitude smaller than MCPs (Bale et al., 2016; Yeates et al., 2016). On the other hand, the size and complexity of MCPs present numerous experimental challenges, e.g. for robust production and assembly in heterologous systems. Several recent studies have advanced the goal of facile heterologous production of MCPs (Parsons et al., 2008; Bonacci et al., 2012; Sargent et al., 2013; Held et al., 2016). A common theme in many studies has been a reductionist approach to try to simplify or minimize the MCP shell by testing various subsets of shell proteins for their ability to assemble properly (Parsons et al., 2010; Choudhary et al., 2012; Lassila et al., 2014); some MCPs are comprised of seven or more different BMC shell proteins. The results have provided hints about the structural roles of individual shell proteins in MCP assembly, and have produced shells comprised of only a small number of distinct shell proteins, making them regular enough in shape to be structurally characterized (Jorda et al., 2016; Sutter et al., 2017). *For correspondence. E-mail yeates@mbi.ucla.edu; Tel. 310-2064866; Fax 310-206-3914. Microbial Biotechnology (2018) 11(1), 160–162 doi:10.1111/1751-7915.12869 Funding information NIH R01AI081146.

Bacterial microcompartments, present in nearly 20% of known bacterial species, are giant proteinaceous structures consisting of a protein shell that encloses a series of sequentially acting metabolic enzymes (Bobik et al., 2015). Their general purpose is to prevent small, volatile or toxic pathway intermediates from being released into the cytosol. Depending on context, bacterial microcompartments are abbreviated in the literature alternatively as MCPs or BMCs; for clarity here (and for keeping with Graf et al.), we use MCP to describe a microcompartment, and we use the widely accepted name BMC for the family of proteins that comprise the main components of MCP shells. A few types of MCPs have been characterized experimentally, beginning first with the carboxysome (which enhances CO 2 fixation in cyanobacteria and some chemoautotrophs), and then the Pdu MCP (which allows for the metabolism of 1,2-propanediol without exposing the bacterial cytosol to the reactive intermediate propionaldehyde) (Fig. 1). More broadly, MCPs predicted to sequester diverse metabolic pathways have been identified by bioinformatic analyses and preliminary experimental studies (Abdul-Rahman et al., 2013;Jorda et al., 2013;Axen et al., 2014). Structural studies have played a key role in showing that MCPs are not merely inclusions in bacterial cells, but highly sophisticated molecular machines satisfying the concept of true organelles in bacteria (Kerfeld et al., 2005;Yeates et al., 2011).
Revelations about the complexity and sophistication of MCPs have spurred much research into understanding more fully how they evolved, how they operate and how they might be engineered for unique applications.
Within the broad scientific area of synthetic biology, diverse molecular systems are currently being pursued as potential frameworks for achieving spatial control over the various activities within cells (Lee et al., 2012). Among these, MCPs provide natural starting points for applications aimed at organizing metabolic processes (Kim and Tullman-Ercek, 2013;Lawrence et al., 2014). They are especially attractive for applications where limited molecular exchange between different subcellular regions is desirable, either to improve channelling of substrates through multiple enzymatic reactions, or to limit the escape of toxic metabolites or even toxic proteins (Yung et al., 2017). In addition, owing to their size since thousands of protein subunits comprise a single shell -MCPs have internal volumes large enough to encapsulate hundreds of enzyme molecules. This capacity offers an advantage over current strategies for designing novel enclosures by protein engineering; those efforts have so far produced designed protein assemblies with internal volumes one or two orders of magnitude smaller than MCPs (Bale et al., 2016;Yeates et al., 2016). On the other hand, the size and complexity of MCPs present numerous experimental challenges, e.g. for robust production and assembly in heterologous systems.
Several recent studies have advanced the goal of facile heterologous production of MCPs (Parsons et al., 2008;Bonacci et al., 2012;Sargent et al., 2013;Held et al., 2016). A common theme in many studies has been a reductionist approach to try to simplify or minimize the MCP shell by testing various subsets of shell proteins for their ability to assemble properly (Parsons et al., 2010;Choudhary et al., 2012;Lassila et al., 2014); some MCPs are comprised of seven or more different BMC shell proteins. The results have provided hints about the structural roles of individual shell proteins in MCP assembly, and have produced shells comprised of only a small number of distinct shell proteins, making them regular enough in shape to be structurally characterized (Jorda et al., 2016;Sutter et al., 2017). This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. bs_bs_banner New work described in this issue analyses the challenges of heterologous MCP expression from a different angle. Rather than investigating what types of MCP constructse.g. what subsets of natural shell genesare compatible with assembly in a given target host (e.g. Escherichia coli), Graf et al. (2017) ask what hosts might be compatible with production of some MCP from its entire complement of genes. This is notable in view of the many genes required for native assembly and function of most MCP types (other than some carboxysomes, which are generally simpler). The authors focus on the Pdu MCP (Fig. 1) and the associated components of the cob/cbi cobalamin biosynthetic operon from Salmonella (cobalamin is required for 1,2-propanediol metabolism inside the Pdu MCP). Including enzymes, structural components and regulatory proteins, this amounts to 43 genes across a 38 kb segment of DNA.
To execute their heterologous expression studies, Graf et al. (2017) used an FRT (flippase-based) strategy for in vivo recombination to insert their 38 kb DNA segment into a plasmid with a broad host range across Gram negative bacteria. A plasmid harbouring all the pdu and cob/cbi genes allowed the investigators to rapidly assess the ability of diverse host bacterial species to produce native-like MCPs. MCPs could be isolated from 6 of the 10 species tested. The differences in outcomes provide opportunities for future studies to dissect the origins of the genetic interactions observed, including host factors and regulatory mechanisms essential to MCP function and assembly. In addition, the successful production of MCPs in several new bacterial hosts will enable new functional studies and promote new MCP applications in diverse contexts.

Conflict of interest
None declared.