Combinatorial engineering of intergenic regions in operons tunes expression of multiple genes Brian F Pfleger 1 , Douglas J Pitera 1 , Christina D Smolke 1,4 , & Jay D Keasling 1–3 Many applications of synthetic biology require the balanced expression of multiple genes. Although operons facilitate coordinated expression of multiple genes in prokaryotes and eukaryotes, coordinating the many post-transcriptional processes that determine the relative levels of gene expression in operons by a priori design remains a challenge. We describe a method for tuning the expression of multiple genes within operons by generating libraries of tunable intergenic regions (TIGRs), recombining various post-transcriptional control elements and screening for the desired relative expression levels. TIGRs can vary the relative expression of two reporter genes over a 100-fold range and balance expression of three genes in an operon that encodes a heterologous mevalonate biosynthetic pathway, resulting in a sevenfold increase in mevalonate production. This technology should be useful for optimizing the expression of multiple genes in synthetic operons, both in prokaryotes and eukaryotes. The synthesis of natural or unnatural products in microorganisms usually involves the introduction of several genes encoding the enzymes of a metabolic pathway 1,2 . Often, in order to produce these molecules at commercially relevant levels, the genes must be expressed at appropriately balanced levels to avoid the accumulation of toxic intermediates or bottlenecks that result in growth inhibition or suboptimal yields. Similarly, manipulation of multisubunit proteins (for example, F 1 F 0 -ATPase, proteasomes and ion channels) usually requires coordinated expression of several genes to produce the subunits at the appropriate stoichiometries 3 . Development of tightly regulated, well-characterized molecular components for the construc- tion of optimized biological pathways is a central challenge in synthetic biology 4 . It is nearly impossible to predict the necessary strengths of the promoters and ribosome binding sites (RBSs) required to balance and coordinate the expression of multiple genes. Grouping multiple, related genes into operons, as is done naturally in prokaryotes 5 , is a convenient means for regulating several genes simultaneously without the need for multiple promoters. Internal ribosomal entry sequences (IRESs) from eukaryotic viruses and host stress response pathways perform a similar function and have been harnessed to create operons for heterologous expression of genes in eukaryotes 6–9 . With a single promoter controlling the transcription of several genes, relative expression of each open reading frame in the operon is controlled by altering post-transcriptional processes such as transcription termination 10,11 , mRNA stability 12,13 and translation initiation 14–16 . Sequences inserted into the intergenic regions of bacterial operons can direct the processing and segmental stability of a transcript containing multiple coding regions 17,18 . This type of directed mRNA processing results in differential production of the proteins encoded in the operon depending on the nature of the intergenic region between the coding regions. One of the major obstacles to designing and implementing this type of control is the difficulty in decoupling the many interrelated variables involved in post-transcriptional regulation 19 . We previously demonstrated that it is possible to differentially control the protein levels encoded by two or more genes in an operon using intergenic- region sequences 17,18 . Here, we simultaneously tune the expression of several genes within operons by generating and screening large libraries of TIGRs containing control elements that include mRNA secondary structures, RNase cleavage sites and RBS sequestering sequences. An operon reporter system (Fig. 1a) containing the genes encoding the red fluorescent protein DsRed (rfp EC ) 20,21 and the green fluorescent protein GFP (gfp UV ) facilitates high-throughput measure- ment of relative gene expression resulting from the TIGR libraries. A large library of TIGR sequences (410 4 ) was assembled combi- natorially from four sets of oligonucleotides (Supplementary Table 1 online) using PCR. Each oligonucleotide contained two 15-nt sequences that hybridized to a corresponding sequence in the neighboring oligonucleotide, such that a series of chimeric DNA molecules containing oligonucleotides from each of the four sets was created after several rounds of PCR (Fig. 1b). Between the hybridization sequences at either end of each oligonucleotide was a variable sequence that provided the diversity of features designed into the library. PCR amplification of this DNA pool with end-specific oligonucleotides enriched the population with full-length TIGRs containing a member from each set of oligonucleotides (Fig. 1b). Specific restriction sites incorporated into the amplification primers were used to clone the TIGR library between the two reporter genes. The TIGR pool that resulted from the assembly of the oligonucleo- tides was designed to contain three regions, two variable hairpin Received 25 April; accepted 21 May; published online 16 July 2006; doi:10.1038/nbt1226 1 Department of Chemical Engineering, University of California Berkeley, California 94720-1462, USA. 2 Department of Bioengineering, University of California Berkeley, California 94720-1762, USA. 3 Synthetic Biology Department, Physical Biosciences Division, Lawrence Berkeley National Laboratory, Berkeley, California 94720, USA. 4 Present address: Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125, USA. Correspondence should be addressed to J.D.K. (keasling@berkeley.edu). NATURE BIOTECHNOLOGY VOLUME 24 NUMBER 8 AUGUST 2006 1027 LETTERS © 2006 Nature Publishing Group http://www.nature.com/naturebiotechnology