Holistic engineering of cell-free systems through proteome-reprogramming synthetic circuits

Synthetic biology has focused on engineering genetic modules that operate orthogonally from the host cells. A synthetic biological module, however, can be designed to reprogram the host proteome, which in turn enhances the function of the synthetic module. Here, we apply this holistic synthetic biology concept to the engineering of cell-free systems by exploiting the crosstalk between metabolic networks in cells, leading to a protein environment more favorable for protein synthesis. Specifically, we show that local modules expressing translation machinery can reprogram the bacterial proteome, changing the expression levels of more than 700 proteins. The resultant feedback generates a cell-free system that can synthesize fluorescent reporters, protein nanocages, and the gene-editing nuclease Cas9, with up to 5-fold higher expression level than classical cell-free systems. Our work demonstrates a holistic approach that integrates synthetic and systems biology concepts to achieve outcomes not possible by only local, orthogonal circuits.

To measure the protein expression of our multi-strain cell lysates and controls, we designed a vector with the gene encoding deGFP under the control of the T7 promoter. Different versions of this plasmid were built and tested to find the vector that yielded the highest deGFP expression in our system (Methods, Section M1). For the quantification of deGFP produced by our E. coli celllysates, we constructed a calibration curve using purified eGFP (BioVision, Inc) as a standard ( Supplementary Fig. 6, Methods, Section M4).

Supplementary Note 2. Optimization of cell lysate preparation and reaction buffer
The whole-cell lysate preparation protocol was optimized by varying pellet to sonication buffer ratio, energy input for the sonication of the cellular pellet, and runoff duration. Different growth media were tested following the same whole-cell lysate preparation protocol, and the media that showed an advantage over the others was incorporated into the protocol in all subsequent whole cell lysate preparations ( Supplementary Fig. 2). The conditions for CFPS were optimized by varying template DNA concentration, expression time, temperature, and agitation speed during incubation ( Supplementary Fig. 1). The optimized protocol for obtaining our multi-strain cell lysates and for the assembly of the CFPS reactions is detailed in Methods, Section M2.

Supplementary Note 3. SDS-PAGE analysis of CFPS reactions supplemented with purified translation machinery
To confirm that the translation machinery protein concentrations supplemented are comparable to those present in BL-18SWCE and BL-7SWCE, we analyzed and compared the reactions in Fig. 2B&C with control reactions through SDS-PAGE ( Supplementary Fig. 7). The change in band intensities shows the staged increase in translation machinery proteins in the reactions supplemented with the purified mixtures. During gel analysis, it is noticeable that we can closely match and surpass the concentrations of the overexpressed proteins in BL-18SWCE and BL-7SWCE. There is a prominent difference between the intensities of the intentionally overexpressed proteins across the samples, as seen in Fig. 3C indicated in black. The intended enrichment of CFP and EF-Tu/EF-Ts is clearly observed in the BL-CFPWCE and BL-S1WCE, respectively. All over-expressed proteins in BL-7SWCE were enriched by more than 2-fold with some as high as 12-fold, except for RF-1 and RF-2, which showed no significant difference. This result was anticipated because the strain expressing those proteins comprise ~1% of the total inoculum. There is also an 82% and 67% increase in the intensity of EF-Tu and EF-Ts, respectively, when comparing BL-S1WCE to BL-7SWCE, as would be expected by the dilution of Strain-1 in BL-7SWCE. Table 1   Supplementary Table 1

. Addition of maltose to cell-free reactions
Maltose was added at various concentrations to stimulate inorganic phosphate utilization to maltose metabolism, as described previously 2 . Two different types of extracts were prepared to test this modification, one BL-7S consortium grown with maltose supplemented in the 2xYTP growth media (Black) and one without (Grey). The addition of maltose to the growth media was intended to increase the concentration of maltose phosphorylase, an enzyme critical for the metabolism of maltose. CFPS reactions were assembled with these lysates and variable amounts of maltose. The extract cultured with standard conditions showed a decrease in expression with increasing maltose. The standard deviation also increased with maltose. The extract grown with maltose showed a 36% increase in expression with 15 mM maltose. Extracts grown with maltose produced over 2-fold more protein than the extract without in the 15 mM maltose condition. Data are presented as mean values & each error bar represents normalized s.d. (n=3 independent experiments). Standard two-tail t-test. Source data are provided as a Source Data file. Supplementary Fig. 4. Comparison of deGFP expression between CFPS reactions assembled using S12 supplement and the optimized reaction buffer.
The expression of deGFP using conventional S12 buffer 1 and an optimized reaction buffer was assessed in reactions assembled using BL-7SWCE. Reactions assembled using the optimized buffer showed ~2-fold more deGFP expression compared to reactions assembled using conventional S12 We built a standard curve for deGFP quantification by diluting pure eGFP (BioVision, Inc) using the dilution buffer specified in Methods, Section M4. Independent serial dilutions were made in triplicate, and the fluorescence intensity of each dilution was measured using NanoQuant plate (Tecan) and an m1000Pro Infinite plate reader. Excitation and emission wavelength used to measure the fluorescence of eGFP were 488 and 507nm, respectively. The individual values obtained from each dilution were plotted, and the linear equation from the fitted trendline was used to quantify deGFP expression through this study. Each error bar represents s.d. of the three independent serial dilutions. Plasmid maps show the plasmids used throughout this study for CFPS of deGFP. Source data are provided as a Source Data file.