Cell-free TXTL synthesis of infectious bacteriophage T4 in a single test tube reaction

Cell-free TXTL synthesis of infectious bacteriophage T4 in a single test tube reaction The bottom-up construction of biological entities from genetic information provides a broad range of opportunities to better understand fundamental processes within living cells, as well as holding great promise for the development of novel bio- medical applications. Cell-free transcription–translation (TXTL) systems have become suitable platforms to tackle such topics because they recapitulate the process of gene expression. TXTL systems have advanced to where the in vitro con- struction of viable, complex, self-assembling deoxyribonucleic acid-programmed biological entities is now possible. Previously, we demonstrated the cell-free synthesis of three bacteriophages from their genomes: MS2, UX174, T7. In this work, we present the complete synthesis of the phage T4 from its 169-kbp genome in one-pot TXTL reactions. This achieve- ment, for one of the largest coliphages, demonstrates the integration of complex gene regulation, metabolism and self- assembly, and brings the bottom-up synthesis of biological systems to a new level. Key words: cell-free transcription/translation; bacteriophages; self-assembly; cell-free synthetic biology convenient technology for information-based constructive biol- 1. Introduction ogy to address questions at both the basic and applied level (1, 7). The construction of complex biochemical systems in test tube The development of an all-Escherichia coli TXTL toolbox, for exam- reactions has become, in recent years, a thriving research area. ple, has opened new perspectives for the synthesis of biochemi- Fostered by novel capabilities, deoxyribonucleic acid (DNA) cal systems through the execution of natural or synthetic gene assembly in particular, the bottom-up synthesis of genetically networks (8). The in vitro metabolism of this system, composed of programmed biological systems in vitro has arisen as a means to an ATP regeneration system and a carbon source, fuels TXTL, quantitatively dissect the molecular mechanisms found in yielding up to 2 mg/ml of protein synthesis in batch mode and up living cells (1, 2). Such an approach offers, at the same time, to 6 mg/ml in semi-continuous mode (9, 10), allowing the simulta- neous expression of many genes. The transcription, based on the fresh perspectives for engineering active systems readily applicable to biotechnologies, biomanufacturing and medicine endogenous E. coli RNA polymerase and housekeeping sigma fac- (3–6). Cell-free transcription–translation (TXTL) systems have tor 70 (11), acts like a versatile operating system building on a emerged as amenable platforms for such realizations because vast repertoire of regulatory elements, as opposed to traditional they recapitulate the process of gene expression in vitro. TXTL platforms limited to a few bacteriophage RNA polymerases Whereas such systems were initially employed almost exclu- and promoters. This added complexity permits the execution of sively to produce protein outside living organisms, over the past complex gene regulation pathways. This experimental platform 15 years cell-free TXTL has been transformed into a highly has proven useful to prototype single genetic parts in isolation as Submitted: 27 November 2017; Received (in revised form): 2 January 2018; Accepted: 4 January 2018 V The Author(s) 2018. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/synbio/article-abstract/3/1/ysy002/4821891 by Ed 'DeepDyve' Gillespie user on 16 March 2018 2| Synthetic Biology, 2018, Vol. 3, No. 1 T7 T7 T4 T4 60 nm d ds sDNA DNA, , 40 kb 40 kbp p dsDNA dsDNA, , 170 kbp 170 kbp 57 g 57 ge en ne es s 289 g 289 ge enes nes φ φX174 X174 190 nm ssDNA ssDNA, , 5.6 kb 5.6 kb 32 nm 11 g 11 ge en ne es s Unique Unique T4 T4 tr traits: aits: • • in intr tron on s sp pliced liced g genes enes MS2 MS2 • spe • spec cific ific σ σ f fa ac ct to or rs s • 8 tRNAs • 8 tRNAs mRN mRNA A, , 3.6 kb 3.6 kb 27 nm 4 4 g gen enes es Figure 1. Overview of phages synthesized in TXTL. The synthesis of MS2, UX174 and T7 has been described previously (10, 19). Phage schematics have been drawn pro- portionally to size. The genomes shown in red are drawn proportionally to length. The table summarizes the characteristics of each phage and some of the optimum TXTL synthesis parameters for each of them. The number of structural genes is the number of gene products making the phage. well as circuit motifs (2, 12–14). The scalability of this system assembled phages are necessary, including scaffolding and cat- allows the completion of gene circuits from the femtoliter scale alytic proteins, and chaperone-like proteins that assist polypep- in cell-sized liposomes or microfluidics chips (15–18), to milliliters tide chain folding and protein assembly (20). The four virion in test tube reactions. parts (T4 head, tail, long tail fibers and whiskers) are assembled We recently demonstrated that this TXTL toolbox can proc- through four independent, linear pathways that converge to ess larger DNA programs than elementary circuit motifs. The form a mature infectious phage (23). The completed T4 virion is bacteriophages MS2, UX174 and T7 are completely synthesized 190 nm tall and 65 nm wide at its maximum. The complexity of after a few hours by just adding viral genomes to a one-pot gene regulation, metabolism, structure, and self-assembly TXTL mixture (10, 19)(Figure 1). Surprisingly, the TXTL system found in T4 surpasses by far the ones found in the three other was able to synthesize infectious T7 phages in vitro, concur- phages described above. T4 gene regulation includes overlap- rently with DNA replication, revealing unexpected potentials of ping genes, internal translation initiation, eukaryote-like cell-free protein synthesis. Together with MS2 and UX174, coli- introns spliced genes, translational bypassing, RNA processing, phages turned out to be a measure of the capabilities of TXTL to specific sigma- and anti-sigma factors, and its own DNA replica- effectively achieve gene expression and complex self- tion machinery (21). It is the degree of complexity found in T4 assemblies into functioning wholes. In this work, we demon- that positions it between the simplicity of a parasitic existence strate that the TXTL toolbox can be challenged with the phage of a virus and the elaborate architecture of a free-living cell. T4, one of the largest enterobacteriophages that infects E. coli. Remarkably, infectious T4 phages are produced in one-pot 2. Materials and methods batch mode TXTL reactions using the same procedure used for 2.1 Reagents the three other phages, by just adding the T4 genome to a standard TXTL reaction. All chemicals were reagent-grade and purchased from Sigma The coliphage T4 is among the largest and most complex Aldrich unless otherwise specified. viruses developed into a model system for the study of funda- mental molecular events in biological systems, rivaling in size 2.2 Phage T4 stock preparation and complexity the eukaryotic herpesviruses. The T4 genome is comprised of 168 903 bp, encoding 289 genes (20). Whereas Phages were produced using a single-plaque, multi-cycle tech- only 62 genes have been deemed essential to form viable T4 nique previously well described in Chen et al. (24). Briefly, E. coli phages in vivo, many others support a wide array of functions B cells were grown in Luria-Bertani (LB) media in an overnight including nucleoside metabolism, suppression of the host culture. This culture was diluted 50 into fresh LB media and a genome, encoding 8 tRNAs, and the timing of gene expression single 4–5 h wild-type T4 plaque grown on an LB agar plate with over the course of infection (21). A T4 phage particle is com- E. coli B cells was cored with a flame-sterilized Pasteur pipette posed of more than 1500 proteins from about 50 different genes. and blown into the liquid culture. The culture was then allowed It has been determined that 24 proteins are involved in the head to incubate for 2 h and then monitored until the onset of super- assembly, 22 in the tail assembly, and 6 in formation of the tail infection. Superinfection was tested for by adding CHCl3 to a fibers (22). More than 10 other genes not present in the small volume of the infected cells and monitoring for lysis after Downloaded from https://academic.oup.com/synbio/article-abstract/3/1/ysy002/4821891 by Ed 'DeepDyve' Gillespie user on 16 March 2018 M. Rustad et al. | 3 5 min. The cells were allowed to incubate for an additional 2 h an energy buffer and 20 canonical amino acids solution (26). The before deoxyribonucleases (DNase) I (Sigma Aldrich) was added energy buffer is composed of 50 mM Hepes pH 8, 1.5 mM ATP and to a final concentration of 5 lg/ml. The solution was shaken at GTP, 0.9 mM CTP and UTP, 0.2 mg/ml tRNA, 0.26 mM coenzyme A, 37 C for an additional 5 min before the volume was then centri- 0.33 mM NAD, 0.75 mM cAMP, 0.068 mM folinic acid, 1 mM sper- fuged at 10 000g for 10 min at 4 C to pellet the infected cells. midine, 30 mM 3-PGA, either 10–15 mM maltose or 20–40 mM The supernatant was discarded and the pellet resuspended in maltodextrin. A typical cell-free reaction is composed of 33% (v/ 10 ml cold 1 TM [50 mM Tris (pH 7.8), 10 mM MgCl ] buffer con- v) of E. coli crude extract. The other 66% of the reaction volume is taining 5 lg/ml DNase I. The concentrated cells were lysed with composed of the energy mixture, the amino acids and plasmids. the addition of 200 ml of CHCl3 and heavy vortexing. The solu- The amino acid concentration was adjusted between 1.5 mM and tion was centrifuged at 10 000g for 10 min at 4 C to remove cel- 3 mM of each of the 20 amino acids. Mg-glutamate, K-glutamate lular lysis products. This stock was stored at 4 C over CHCl3 and PEG8000 (polyethelyne glycol, average molecular weight of until a fraction was removed for further purification. The 8000) concentrations were adjusted based on the reaction test removed volume was purified by 5–45% (wt/vol) sucrose gra- made, see text (typically 60 mM K-glutamate, 5 mM Mg- dient and centrifuged at 35 000 rpm for 20 min in a swinging glutamate and 2% PEG8000 for P70a-deGFP, a reference plasmid bucket rotor in an ultracentrifuge system. The bands were for the TXTL system (10). Cell-free reactions are carried out in a removed, re-diluted by 3 in 1 TM and re-centrifuged at volume of 5 mlto 20 mlat 29–30 C. The controls included two 25 000 rpm for 1 h to pellet the phage. Each pellet was resus- assays based on rifampicin, an inhibitor of the core RNA poly- pended using 0.2 ml 1 TM buffer at 4 C overnight. merase, and DNase I. Rifampicin was used at 100lg/ml (122 mM) and DNase I at 1 mg/ml in TXTL reactions. 2.3 T4 genome extraction 2.5 Plaque assay Concentrated T4 stock was diluted 10 with Millipore H2O (Milli-Q Advantage A10) before adding an equal volume of satu- The bacteriophage was counted using the standard plaque rated Tris: Phenol: Chloroform (ThermoFisher Scientific) solu- assay using the E. coli strain P301. The cells were grown in LB tion to disrupt structural phage proteins. The layered solution broth at 37 C, following a two-stage cascading culture protocol. was gently shaken by hand for 2 min until an emulsion was Incubate a 5 ml LB pre-culture overnight to saturation; then, formed. This was immediately followed by 5 min of centrifuga- dilute the saturated pre-culture 50: 1 in 50 ml LB and grow host tion at 4 C at 13 300 rpm. Due to the large size of the T4 genome, cells 3.5 h to mid-log phase. Centrifuge 50 ml culture for 10 min all following DNA handling steps were performed using wide- at 5000g and resuspend in 5 ml LB. The plates were prepared bore pipette tips to avoid shear damage. The aqueous phase as follows: each sample was added to a solution of 2.6 ml of was slowly pipetted off while avoiding disruption of the protein 0.6% liquid LB-agar solution (45 C) and 0.025 ml of cell culture boundary. The pipetted volume was again mixed with an equal dispensed on a 1.5% solid LB-agar plate. Plates were incubated volume of Tris: Phenol: Chloroform and the process was at 37 C for 7 h. The error bars shown in the plots are standard repeated for a total of three phenol extractions. A final chloro- deviations of multiple repeats. form back extraction was performed to remove phenol from the genomic solution followed by 5 min of centrifugation at 4 C and 2.6 Electron microscopy 13 000 rpm before removing the aqueous phase to a fresh To visualize phages originating from the TXTL system, reactions Eppendorf tube. This volume was then estimated and 2.8 vol- were plated as above at concentrations high enough to com- umes of 95% ethanol (EtOH) (ThermoFisher Scientific) were pletely lyse all cells on the plate. Phages were recovered from added for a total EtOH concentration of 70%. This was gently the plates by scraping the top agar from the plates into 2.5 ml mixed and allowed to sit overnight at 20 C to precipitate DNA 1 TM buffer. This extract was homogenized by vortexing and in solution. This solution was then centrifuged for 30 min at clarified by high speed centrifugation. The supernatant was 13 300 rpm at 4 C to pellet the T4 DNA. The supernatant was then filtered through a 0.45 mm sterile filter, and phages were discarded and 500 ll of 70% EtOH was added. The pellet was pelleted in a Ti-50.2 ultracentrifuge rotor at 30 000 rpm, 4 C for washed by flicking the tube and re-centrifuged at 13 000 rpm at 90 min (Beckman Coulter). Pelleted phage were resuspended 4 C, discarding the ethanol after. This step was repeated again overnight at 4 Cin 1 TM buffer. Phages were adsorbed to and for 2 total washes. The final pellet was allowed to air dry for imaged on carbon-coated formvar grids stained with 2% w/v 20 min at room temperature before resuspension of the DNA uranyl acetate in a Tecnai Spirit TEM (FEI). pellet in 100 ll of Millipore water. The final genomic concentra- tion was determined using a NanoDrop One at 260 nm. Following genomic extraction, a fraction of T4 DNA was plated 3. Results and discussion on agar with and without an E. coli B carpet to ensure that no The goal of the first experiment was to plot the kinetics of phage plaques formed, indicating all phage were disrupted by the synthesis to determine the reaction incubation time and the process or that there was no bacterial contamination, potential yield of the system. Samples from a cell-free reaction respectively. set at the optimum biochemical settings (1 nM genome, 5 mM Mg, 60 mM K, 3% PEG8000) were plated at 1 h interval. No detect- 2.4 TXTL system and T4 TXTL reactions able phages were synthesized during the first 2 h. Most of the Preparation of the all-E. coli TXTL system (myTXTL, Arbor phages were synthesized between 2 and 4 h of incubation Biosciences) used in this work was described previously in sev- (Figure 2). Phage yield increases slightly after 4 h to plateau eral articles (11, 25). Transcription and translation are performed between 6 and 10 h. Although slightly slower than T7 (19), the by the endogenous molecular components provided by an E. coli T4 synthesis kinetics in TXTL presents the same trends as T7, cytoplasmic extract, without addition of exogenous purified with a burst of phages that lasts a few hours. Subsequently, all TXTL molecular components. TXTL reactions are composed of the TXTL reactions were incubated for 10–12 h. Downloaded from https://academic.oup.com/synbio/article-abstract/3/1/ysy002/4821891 by Ed 'DeepDyve' Gillespie user on 16 March 2018 4| Synthetic Biology, 2018, Vol. 3, No. 1 Our next experiment consisted in varying the genome con- Potassium glutamate (K-glu) and magnesium glutamate centration in TXTL reaction. The goal was to probe the production (Mg-glu) are essential ions for transcription and translation capacity limit of the synthesis reaction, using the optimal reac- reactions and therefore highly important to be set at precise tions settings described above. Genome concentrations above the concentrations for T4 TXTL synthesis. Across a respective range saturating limit are extraneous and genome concentrations of 0–100 mM K-glu and 4–7 mM Mg-glu, optimum concentrations for K-glu were found to be between 40 and 80 mM while there below optimum do not fully exploit the assembly pathways and was no statistical difference between concentrations tested chemical energy present in the reaction. We observed that at using Mg-glu (Figure 4). Outside of the 4–7 mM Mg-glu concen- 1 nM and above the reaction was saturated (Figure 3). Throughout tration range phage synthesis was notably smaller. The opti- the course of this work, T4 phage synthesis reached 10 PFU/ml mized concentrations that were used in reactions were 60 mM several times. It corresponds to about 1.66 10 phage produced and 5 mM K-glu and Mg-glu, respectively. per genome added to the reaction. We did not observe increased It has been well established theoretically and experimentally T4 synthesis when dNTPs were added to the cell-free reaction, that molecular crowding is a fundamental aspect of macromo- suggesting that DNA replication does not occur. However, one lecular assembly, especially in biological systems (27–29). limitation to address quantitatively this question was the high Molecular crowding favors self-assembly by an entropy-driven viscosity of the reaction due to the size of the T4 genome. In TXTL, some of the components show correlations to process. In a solution containing large concentrations of macro- phage yield when their concentrations are varied, such as ion molecules, such as a cytoplasm or an extract-based cell-free expression system, a net gain of entropy is created by decreas- concentration and molecular crowders. We optimized the con- ing the excluded volume when macromolecules self-assemble. centrations of magnesium, potassium, and PEG for T4 synthesis. Figure 2. Kinetics of T4 TXTL synthesis. A concentration of 1 nM T4 genome and Figure 3. T4 TXTL synthesis as a function of genome concentration. Cell-free 3% PEG was used for this experiment. No phage is produced during the first 2 h reactions with no genome added had no plaques. Standard TXTL reactions were of incubation. T4 synthesis reaches plateau within just a few hours, typically carried out with 3% PEG. Reactions were incubated for 12 h. The optimum phage 7 9 between 10 and 10 . synthesis is observed at 1 nM genome. Figure 4. Magnesium and potassium settings for T4 TXTL synthesis. (A) Optimum phage synthesis is observed for a magnesium concentration of 4–7 mM. Phage yields are smaller outside this range. (B) 40–80 mM potassium is optimum for T4 TXTL synthesis. Downloaded from https://academic.oup.com/synbio/article-abstract/3/1/ysy002/4821891 by Ed 'DeepDyve' Gillespie user on 16 March 2018 M. Rustad et al. | 5 It has been directly observed and modeled for the assembly of entire phages are expressed. By just increasing the PEG8000 viruses such as HIV in vitro (30–32). Molecular crowding also concentration to 3–4%, we could increase TXTL T7 synthesis by alters the rates and the equilibrium constants of biochemical a few orders of magnitude. We varied the concentration of reactions by increasing the effective concentrations of macro- PEG8000 (1% ¼ 1.25 mM) in a TXTL reaction at optimal settings molecules, promoting in particular the association of macromo- for T4 synthesis (1 nM T4 genome, 5 mM Mg-glu and 60 mM lecules (33, 34). Consequently, molecular crowding in TXTL K-glu). The largest effect from any optimized parameter was the reactions is an essential parameter to vary, especially when PEG concentration, showing a >100 000-fold PFU/ml increase self-assembled systems are expressed. We already demon- when increasing from 0.5% to 1.5% v/v (Figure 5). strated the dramatic effect of PEG8000, a typical crowder for To eliminate any source of artifacts and contaminations, we in vitro reactions, on the synthesis of the phage T7 in TXTL (10). performed eight types of controls during all stages of experi- In the case of T7, we showed that the standard molecular crow- mentation throughout the course of the work (Figure 6A), from ders settings in TXTL, 1.5–2% PEG8000, are not optimal when genome extractions to cell-free reactions. Each of the T4 genome preparations used in this study was plated on LB agar multiple times to detect contaminating host cells (control 1). None were detected in the 30 repeats made. A similar control, done in parallel with control 1, was made by mixing the T4 genome with the E. coli host B (control 2) to test random T4 genome uptake by host cells. No plaque was formed in the 30 repeats. Control 1 and 2 covered all the T4 stocks prepared for this work. There were no observed plaques when the T4 genome was added to a cell-free reaction and incubated for <1 min (control 3) in the 10 performed repeats. When rifampicin (100 lg/ml or 122 mM), an inhibitor of the E. coli core RNA poly- merase already tested in TXTL (10), was added to a cell-free reaction at time 0 with 1 nM T4 genome, no plaque was observed (control 4). When DNase I (1 mg/ml) was added to a cell-free reaction at time 0 with 1 nM T4 genome, no plaque was observed (Control 5). However, when DNase I was added to the cell-free reaction after 12 h of incubation and incubated for 30 min at 37 C, an average of 10 plaque forming unit per millili- ter (PFU/ml) was measured (control 6). In each of the final Figure 5. T4 TXTL synthesis as a function of PEG8000 concentration. Standard repeats, the equivalent of seven cell-free reactions of 12 ml (no TXTL reactions were carried out with optimum settings: 5 mM magnesium, 60 mM potassium, 1 nM genome. Maximum TXTL T4 synthesis occurs around DNA added, 84 ml total) was plated at 0 (Control 7) and 12 h 3% PEG. The phage yields varies by orders of magnitude, as observed previously (Control 8) incubation. No colonies were observed when plated for the other phages UX174 and T7 (10). showing that our cell-free expression system does not contain Figure 6. Experimental controls. (A) Table summarizing the controls and repeats. These controls were carried out during the course of experimentation. (B) Example of a set of controls including a positive (labeled as þ, TXTL reaction containing 1 nM T4 genome) and controls 6, 5 and 4. (C) Example of control plates (negative on the left with a lawn of cells and positive on the right) for T4 TXTL synthesis. (D) Electron microscopy image of T4 phages from plaques produced by a TXTL reaction containing 1 nM of the T4 genome. Downloaded from https://academic.oup.com/synbio/article-abstract/3/1/ysy002/4821891 by Ed 'DeepDyve' Gillespie user on 16 March 2018 6| Synthetic Biology, 2018, Vol. 3, No. 1 intact E. coli cells, as discussed previously (10). An example of prototyping genetic networks with cell-free transcription- the controls (Figure 6B) was performed as follows: a cell-free translation reactions. Methods, 86, 60–72. reaction containing 1 nM T4 genome was prepared and then 3. Bundy,B.C., Franciszkowicz,M.J. and Swartz,J.R. (2008) split into equal volume reactions to which was added (i) water Escherichia coli-based cell-free synthesis of virus-like particles. (positive control), (ii) rifampicin at time 0 (100 lg/ml or 122 mM), Biotechnol. Bioeng., 100, 28–37. (iii) DNase I at time 0 (1 mg/ml) and (iv) DNase I added after 12 h 4. Carlson,E.D., Gan,R., Hodgman,C.E. and Jewett,M.C. (2011) of incubation. The plaque assay for T4 was carried out according Cell-free protein synthesis: applications come of age. to standard procedures (Figure 6C). We concluded that the syn- Biotechnol. Adv., 30, 1185–1194. thesis of the phage T4 measured on plaques does not come 5. Dudley,Q.M., Karim,A.S. and Jewett,M.C. (2014) Cell-free met- from stock contamination. Despite multiple attempts, it was abolic engineering: biomanufacturing beyond the cell. not possible to directly visualize the phage T4 from TXTL reac- Biotechnol. J., 10, 69–82. tions by electron microscopy (EM) because of the high viscosity 6. Pardee,K., Slomovic,S., Nguyen,P.Q., Lee,J.W., Donghia,N., of the solutions and the relatively low number of phages syn- Burrill,D., Ferrante,T., McSorley,F.R., Furuta,Y., Vernet,A. thesized (10 PFU/ml at best). As an additional verification step, et al. (2016) Portable, on-demand biomolecular manufactur- however, we visualized the phage T4 by EM from plaques pro- ing. Cell, 167, 248–259.e212. duced from a TXTL reaction (Figure 6D), thus confirming cell- 7. Perez,J.G., Stark,J.C. and Jewett,M.C. (2016) Cell-free synthetic free synthesis of the phage T4. biology: engineering beyond the cell. Cold Spring Harb. Perspect. Biol., 8, a023853. 8. Shin,J. and Noireaux,V. (2012) An E. coli cell-free expression 4. Summary and conclusions toolbox: application to synthetic gene circuits and artificial The objective of this letter is to report the synthesis of the largest cells. ACS Synth. Biol., 1, 29–41. biological entity so far achieved in a TXTL reaction. By demon- 9. Caschera,F. and (2014) Synthesis of 2.3 mg/ml of protein strating that TXTL systems can process a DNA program larger with an all Escherichia coli cell-free transcription-translation than 100 kbp encoding for about 100 essential genes, we bring the system. Biochimie, 99, 162–168. capabilities of the cell-free technology to a new level. We believe 10. Garamella,J., Marshall,R., Rustad,M. and Noireaux,V. (2016) that the total synthesis of a phage of this size and complexity is a The all E. coli TX-TL toolbox 2.0: a platform for cell-free syn- significant milestone towards the TXTL execution of a minimal thetic biology. ACS Synth. Biol., 5, 344–355. cell genome for self-reproduction of a living entity. It has been 11. Shin, J. and Noireaux, V. (2010). Efficient cell-free expression estimated by different approaches and groups that the minimal with the endogenous E. coli RNA polymerase and sigma fac- set of genes required by a free-living cell is between 300 and 500 tor 70. J. Biol. Eng., 4, 8. genes (35–38). The complete synthesis of T4 in an open environ- 12. Niederholtmeyer,H., Sun,Z.Z., Hori,Y., Yeung,E., Verpoorte,A., ment like TXTL offers unique possibilities to interrogate quantita- Murray,R.M. and Maerkl,S.J. (2015) Rapid cell-free forward engi- tively the links between gene expression, self-assembly and neering of novel genetic ring oscillators. Elife,4, e09771. metabolism, which is left for a future study. Our future work will 13. Sun,Z.Z., Yeung,E., Hayes,C.A., Noireaux,V. and Murray,R.M. also consist of elucidating a few surprising results. For example, (2014) Linear DNA for rapid prototyping of synthetic biologi- the literature suggests that the phage T4 initiates its self- cal circuits in an Escherichia coli based TX-TL cell-free system. assembly at the inner membrane of E. coli (20). It is unlikely to ACS Synth. Biol., 3, 387–397. happen in TXTL because addition of liposomes prepared from E. 14. Takahashi,M.K., Chappell,J., , Sun,Z.Z., Kim,J., coli membranes to TXTL reactions did not increase phage synthe- Singhal,V., Spring,K.J., Al-Khabouri,S., Fall,C.P., Noireaux,V. sis (not shown). This observation suggests that there are other et al. (2015a) Rapidly characterizing the fast dynamics of RNA thermodynamic modes that allow efficient self-assembly of com- genetic circuitry with cell-free transcription-translation plex biological systems in in vitro reactions lacking natural tem- (TX-TL) systems. ACS Synth. Biol., 4, 503–515. plates used for self-organization. The demonstration that MS2, 15. Caschera,F. and Noireaux,V. (2016) Compartmentalization of UX174, T7 and now T4 can be entirely produced in TXTL reactions an all-E. coli cell-free expression system for the construction also paves the way toward phage engineering for biomedical of a minimal cell. Artif. Life, 22, 185–195. applications. We expect that the all-E. coli TXTL platform can be 16. Karzbrun,E., Tayar,A.M., Noireaux,V. and Bar-Ziv,R.H. (2014) employed to create synthetic phages, by recombineering for Programmable on-chip DNA compartments as artificial cells. example, useful for phage therapy or other applications (39). Science, 345, 829–832. 17. Majumder,S., Garamella,J., Wang,Y.L., DeNies,M., Noireaux,V. and Liu,A.P. (2017) Cell-sized mechanosensitive and biosens- Funding ing compartment programmed with DNA. Chem. Commun., The Office of Naval Research [N00014-13-1-0074]; the Human 53, 7349–7352. Frontier Science Program [RGP0037/2015]. Noireaux laboratory 18. Tayar, A., Karzbrun, E., Noireaux, V. and Bar-Ziv, R. (2015) receives research funds from Arbor Biosciences, a distributor of Propagating gene expression fronts in a one-dimensional myTXTL cell-free protein expression kit. coupled system of artificial cells. Nat. Phys., 19. Shin,J., Jardine,P. and (2012) Genome replication, syn- Conflict of interest statement. None declared. thesis, and assembly of the bacteriophage T7 in a single cell-free reaction. ACS Synth. Biol., 1, 408–413. References 20. Yap,M.L. and Rossmann,M.G. (2014) Structure and function of bacteriophage T4. Future Microbiol., 9, 1319–1327. 1. Hockenberry,A.J. and Jewett,M.C. (2012) Synthetic in vitro cir- cuits. Curr. Opin. Chem. Biol., 16, 253–259. 21. Miller,E.S., Kutter,E., Mosig,G., Arisaka,F., Kunisawa,T. and 2. Takahashi,M.K., Hayes,C.A., Chappell,J., Sun,Z.Z., Murray,R.M., Ruger,W. (2003) Bacteriophage T4 genome. Microbiol. Mol. Biol. Noireaux,V. and Lucks,J.B. (2015b) Characterizing and Rev., 67, 86–156, table of contents. Downloaded from https://academic.oup.com/synbio/article-abstract/3/1/ysy002/4821891 by Ed 'DeepDyve' Gillespie user on 16 March 2018 M. Rustad et al. | 7 31. Rivas,G., Fernandez,J.A. and Minton,A.P. (2001) Direct obser- 22. Mesyanzhinov,V.V., Leiman,P.G., Kostyuchenko,V.A., Kurochkina,L.P., Miroshnikov,K.A., Sykilinda,N.N. and vation of the enhancement of noncooperative protein Shneider,M.M. (2004) Molecular architecture of bacteriophage self-assembly by macromolecular crowding: indefinite linear T4. Biochemistry (Mosc), 69, 1190–1202. self-association of bacterial cell division protein FtsZ. Proc. 23. Wood,W.B., Edgar,R.S., King,J., Lielausis,I. and Henninger,M. Natl. Acad. Sci. USA, 98, 3150–3155. (1968) Bacteriophage assembly. Fed. Proc., 27, 1160–1166. 32. Smith,G.R., Xie,L., Lee,B. and Schwartz,R. (2014) Applying 24. Chen,X., Mathews,C.K., Wheeler,L.J., Maley,G., Maley,F. and molecular crowding models to simulations of virus capsid Coombs,D.H. (1995) An immunoblot assay reveals that assembly in vitro. Biophys. J., 106, 310–320. bacteriophage T4 thymidylate synthase and dihydrofolate 33. Minton,A.P. (1983) The effect of volume occupancy upon the reductase are not virion proteins. J. Virol., 69, 2119–2125. thermodynamic activity of proteins: some biochemical con- 25. Sun,Z.Z., Hayes,C.A., Shin,J., Caschera,F., Murray,R.M. and sequences. Mol. Cell. Biochem., 55, 119–140. Noireaux,V. (2013) Protocols for implementing an Escherichia 34. (1997) Influence of excluded volume upon macromolec- 0 0 coli based TX-TL cell-free expression system for synthetic ular structure and associations in crowded media. Curr. biology. J. Vis. Exp., e50762. Opin. Biotechnol., 8, 65–69. 26. Caschera,F. and Noireaux,V. (2015) Preparation of amino acid 35. Gil,R., Sabater-Munoz,B., Latorre,A., Silva,F.J. and Moya,A. mixtures for cell-free expression systems. Biotechniques, 58, (2002) Extreme genome reduction in Buchnera spp.: toward 40–43. the minimal genome needed for symbiotic life. Proc. Natl. 27. Minton,A.P. (2000) Implications of macromolecular Acad. Sci. USA, 99, 4454–4458. crowding for protein assembly. Curr. Opin. Struct. Biol., 10, 36. , Silva,F.J., Pereto,J. and Moya,A. (2004) Determination of 34–39. the core of a minimal bacterial gene set. Microbiol. Mol. Biol. 28. (2006) How can biochemical reactions within cells differ Rev., 68, 518–537. from those in test tubes? J. Cell Sci., 119, 2863–2869. 37. Glass,J.I., Assad-Garcia,N., Alperovich,N., Yooseph,S., 29. Zimmerman,S.B. and Minton,A.P. (1993) Macromolecular Lewis,M.R., Maruf,M., Hutchison,C.A. 3rd, Smith,H.O. and crowding: biochemical, biophysical, and physiological conse- Venter,J.C. (2006) Essential genes of a minimal bacterium. quences. Annu. Rev. Biophys. Biomol. Struct., 22, 27–65. Proc. Natl. Acad. Sci. USA, 103, 425–430. 30. del Alamo,M., Rivas,G. and Mateu,M.G. (2005) Effect of macro- 38. Zhang,L.Y., Chang,S.H. and Wang,J. (2010) How to make a mini- molecular crowding agents on human immunodeficiency mal genome for synthetic minimal cell. Protein Cell, 1, 427–434. virus type 1 capsid protein assembly in vitro. J. Virol., 79, 39. Reardon,S. (2014) Phage therapy gets revitalized. Nature, 510, 14271–14281. 15–16. Downloaded from https://academic.oup.com/synbio/article-abstract/3/1/ysy002/4821891 by Ed 'DeepDyve' Gillespie user on 16 March 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Synthetic Biology Oxford University Press

Cell-free TXTL synthesis of infectious bacteriophage T4 in a single test tube reaction

Free
7 pages

Loading next page...
 
/lp/ou_press/cell-free-txtl-synthesis-of-infectious-bacteriophage-t4-in-a-single-QUjVJ23Mf0
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press.
eISSN
2397-7000
D.O.I.
10.1093/synbio/ysy002
Publisher site
See Article on Publisher Site

Abstract

The bottom-up construction of biological entities from genetic information provides a broad range of opportunities to better understand fundamental processes within living cells, as well as holding great promise for the development of novel bio- medical applications. Cell-free transcription–translation (TXTL) systems have become suitable platforms to tackle such topics because they recapitulate the process of gene expression. TXTL systems have advanced to where the in vitro con- struction of viable, complex, self-assembling deoxyribonucleic acid-programmed biological entities is now possible. Previously, we demonstrated the cell-free synthesis of three bacteriophages from their genomes: MS2, UX174, T7. In this work, we present the complete synthesis of the phage T4 from its 169-kbp genome in one-pot TXTL reactions. This achieve- ment, for one of the largest coliphages, demonstrates the integration of complex gene regulation, metabolism and self- assembly, and brings the bottom-up synthesis of biological systems to a new level. Key words: cell-free transcription/translation; bacteriophages; self-assembly; cell-free synthetic biology convenient technology for information-based constructive biol- 1. Introduction ogy to address questions at both the basic and applied level (1, 7). The construction of complex biochemical systems in test tube The development of an all-Escherichia coli TXTL toolbox, for exam- reactions has become, in recent years, a thriving research area. ple, has opened new perspectives for the synthesis of biochemi- Fostered by novel capabilities, deoxyribonucleic acid (DNA) cal systems through the execution of natural or synthetic gene assembly in particular, the bottom-up synthesis of genetically networks (8). The in vitro metabolism of this system, composed of programmed biological systems in vitro has arisen as a means to an ATP regeneration system and a carbon source, fuels TXTL, quantitatively dissect the molecular mechanisms found in yielding up to 2 mg/ml of protein synthesis in batch mode and up living cells (1, 2). Such an approach offers, at the same time, to 6 mg/ml in semi-continuous mode (9, 10), allowing the simulta- neous expression of many genes. The transcription, based on the fresh perspectives for engineering active systems readily applicable to biotechnologies, biomanufacturing and medicine endogenous E. coli RNA polymerase and housekeeping sigma fac- (3–6). Cell-free transcription–translation (TXTL) systems have tor 70 (11), acts like a versatile operating system building on a emerged as amenable platforms for such realizations because vast repertoire of regulatory elements, as opposed to traditional they recapitulate the process of gene expression in vitro. TXTL platforms limited to a few bacteriophage RNA polymerases Whereas such systems were initially employed almost exclu- and promoters. This added complexity permits the execution of sively to produce protein outside living organisms, over the past complex gene regulation pathways. This experimental platform 15 years cell-free TXTL has been transformed into a highly has proven useful to prototype single genetic parts in isolation as Submitted: 27 November 2017; Received (in revised form): 2 January 2018; Accepted: 4 January 2018 V The Author(s) 2018. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/synbio/article-abstract/3/1/ysy002/4821891 by Ed 'DeepDyve' Gillespie user on 16 March 2018 2| Synthetic Biology, 2018, Vol. 3, No. 1 T7 T7 T4 T4 60 nm d ds sDNA DNA, , 40 kb 40 kbp p dsDNA dsDNA, , 170 kbp 170 kbp 57 g 57 ge en ne es s 289 g 289 ge enes nes φ φX174 X174 190 nm ssDNA ssDNA, , 5.6 kb 5.6 kb 32 nm 11 g 11 ge en ne es s Unique Unique T4 T4 tr traits: aits: • • in intr tron on s sp pliced liced g genes enes MS2 MS2 • spe • spec cific ific σ σ f fa ac ct to or rs s • 8 tRNAs • 8 tRNAs mRN mRNA A, , 3.6 kb 3.6 kb 27 nm 4 4 g gen enes es Figure 1. Overview of phages synthesized in TXTL. The synthesis of MS2, UX174 and T7 has been described previously (10, 19). Phage schematics have been drawn pro- portionally to size. The genomes shown in red are drawn proportionally to length. The table summarizes the characteristics of each phage and some of the optimum TXTL synthesis parameters for each of them. The number of structural genes is the number of gene products making the phage. well as circuit motifs (2, 12–14). The scalability of this system assembled phages are necessary, including scaffolding and cat- allows the completion of gene circuits from the femtoliter scale alytic proteins, and chaperone-like proteins that assist polypep- in cell-sized liposomes or microfluidics chips (15–18), to milliliters tide chain folding and protein assembly (20). The four virion in test tube reactions. parts (T4 head, tail, long tail fibers and whiskers) are assembled We recently demonstrated that this TXTL toolbox can proc- through four independent, linear pathways that converge to ess larger DNA programs than elementary circuit motifs. The form a mature infectious phage (23). The completed T4 virion is bacteriophages MS2, UX174 and T7 are completely synthesized 190 nm tall and 65 nm wide at its maximum. The complexity of after a few hours by just adding viral genomes to a one-pot gene regulation, metabolism, structure, and self-assembly TXTL mixture (10, 19)(Figure 1). Surprisingly, the TXTL system found in T4 surpasses by far the ones found in the three other was able to synthesize infectious T7 phages in vitro, concur- phages described above. T4 gene regulation includes overlap- rently with DNA replication, revealing unexpected potentials of ping genes, internal translation initiation, eukaryote-like cell-free protein synthesis. Together with MS2 and UX174, coli- introns spliced genes, translational bypassing, RNA processing, phages turned out to be a measure of the capabilities of TXTL to specific sigma- and anti-sigma factors, and its own DNA replica- effectively achieve gene expression and complex self- tion machinery (21). It is the degree of complexity found in T4 assemblies into functioning wholes. In this work, we demon- that positions it between the simplicity of a parasitic existence strate that the TXTL toolbox can be challenged with the phage of a virus and the elaborate architecture of a free-living cell. T4, one of the largest enterobacteriophages that infects E. coli. Remarkably, infectious T4 phages are produced in one-pot 2. Materials and methods batch mode TXTL reactions using the same procedure used for 2.1 Reagents the three other phages, by just adding the T4 genome to a standard TXTL reaction. All chemicals were reagent-grade and purchased from Sigma The coliphage T4 is among the largest and most complex Aldrich unless otherwise specified. viruses developed into a model system for the study of funda- mental molecular events in biological systems, rivaling in size 2.2 Phage T4 stock preparation and complexity the eukaryotic herpesviruses. The T4 genome is comprised of 168 903 bp, encoding 289 genes (20). Whereas Phages were produced using a single-plaque, multi-cycle tech- only 62 genes have been deemed essential to form viable T4 nique previously well described in Chen et al. (24). Briefly, E. coli phages in vivo, many others support a wide array of functions B cells were grown in Luria-Bertani (LB) media in an overnight including nucleoside metabolism, suppression of the host culture. This culture was diluted 50 into fresh LB media and a genome, encoding 8 tRNAs, and the timing of gene expression single 4–5 h wild-type T4 plaque grown on an LB agar plate with over the course of infection (21). A T4 phage particle is com- E. coli B cells was cored with a flame-sterilized Pasteur pipette posed of more than 1500 proteins from about 50 different genes. and blown into the liquid culture. The culture was then allowed It has been determined that 24 proteins are involved in the head to incubate for 2 h and then monitored until the onset of super- assembly, 22 in the tail assembly, and 6 in formation of the tail infection. Superinfection was tested for by adding CHCl3 to a fibers (22). More than 10 other genes not present in the small volume of the infected cells and monitoring for lysis after Downloaded from https://academic.oup.com/synbio/article-abstract/3/1/ysy002/4821891 by Ed 'DeepDyve' Gillespie user on 16 March 2018 M. Rustad et al. | 3 5 min. The cells were allowed to incubate for an additional 2 h an energy buffer and 20 canonical amino acids solution (26). The before deoxyribonucleases (DNase) I (Sigma Aldrich) was added energy buffer is composed of 50 mM Hepes pH 8, 1.5 mM ATP and to a final concentration of 5 lg/ml. The solution was shaken at GTP, 0.9 mM CTP and UTP, 0.2 mg/ml tRNA, 0.26 mM coenzyme A, 37 C for an additional 5 min before the volume was then centri- 0.33 mM NAD, 0.75 mM cAMP, 0.068 mM folinic acid, 1 mM sper- fuged at 10 000g for 10 min at 4 C to pellet the infected cells. midine, 30 mM 3-PGA, either 10–15 mM maltose or 20–40 mM The supernatant was discarded and the pellet resuspended in maltodextrin. A typical cell-free reaction is composed of 33% (v/ 10 ml cold 1 TM [50 mM Tris (pH 7.8), 10 mM MgCl ] buffer con- v) of E. coli crude extract. The other 66% of the reaction volume is taining 5 lg/ml DNase I. The concentrated cells were lysed with composed of the energy mixture, the amino acids and plasmids. the addition of 200 ml of CHCl3 and heavy vortexing. The solu- The amino acid concentration was adjusted between 1.5 mM and tion was centrifuged at 10 000g for 10 min at 4 C to remove cel- 3 mM of each of the 20 amino acids. Mg-glutamate, K-glutamate lular lysis products. This stock was stored at 4 C over CHCl3 and PEG8000 (polyethelyne glycol, average molecular weight of until a fraction was removed for further purification. The 8000) concentrations were adjusted based on the reaction test removed volume was purified by 5–45% (wt/vol) sucrose gra- made, see text (typically 60 mM K-glutamate, 5 mM Mg- dient and centrifuged at 35 000 rpm for 20 min in a swinging glutamate and 2% PEG8000 for P70a-deGFP, a reference plasmid bucket rotor in an ultracentrifuge system. The bands were for the TXTL system (10). Cell-free reactions are carried out in a removed, re-diluted by 3 in 1 TM and re-centrifuged at volume of 5 mlto 20 mlat 29–30 C. The controls included two 25 000 rpm for 1 h to pellet the phage. Each pellet was resus- assays based on rifampicin, an inhibitor of the core RNA poly- pended using 0.2 ml 1 TM buffer at 4 C overnight. merase, and DNase I. Rifampicin was used at 100lg/ml (122 mM) and DNase I at 1 mg/ml in TXTL reactions. 2.3 T4 genome extraction 2.5 Plaque assay Concentrated T4 stock was diluted 10 with Millipore H2O (Milli-Q Advantage A10) before adding an equal volume of satu- The bacteriophage was counted using the standard plaque rated Tris: Phenol: Chloroform (ThermoFisher Scientific) solu- assay using the E. coli strain P301. The cells were grown in LB tion to disrupt structural phage proteins. The layered solution broth at 37 C, following a two-stage cascading culture protocol. was gently shaken by hand for 2 min until an emulsion was Incubate a 5 ml LB pre-culture overnight to saturation; then, formed. This was immediately followed by 5 min of centrifuga- dilute the saturated pre-culture 50: 1 in 50 ml LB and grow host tion at 4 C at 13 300 rpm. Due to the large size of the T4 genome, cells 3.5 h to mid-log phase. Centrifuge 50 ml culture for 10 min all following DNA handling steps were performed using wide- at 5000g and resuspend in 5 ml LB. The plates were prepared bore pipette tips to avoid shear damage. The aqueous phase as follows: each sample was added to a solution of 2.6 ml of was slowly pipetted off while avoiding disruption of the protein 0.6% liquid LB-agar solution (45 C) and 0.025 ml of cell culture boundary. The pipetted volume was again mixed with an equal dispensed on a 1.5% solid LB-agar plate. Plates were incubated volume of Tris: Phenol: Chloroform and the process was at 37 C for 7 h. The error bars shown in the plots are standard repeated for a total of three phenol extractions. A final chloro- deviations of multiple repeats. form back extraction was performed to remove phenol from the genomic solution followed by 5 min of centrifugation at 4 C and 2.6 Electron microscopy 13 000 rpm before removing the aqueous phase to a fresh To visualize phages originating from the TXTL system, reactions Eppendorf tube. This volume was then estimated and 2.8 vol- were plated as above at concentrations high enough to com- umes of 95% ethanol (EtOH) (ThermoFisher Scientific) were pletely lyse all cells on the plate. Phages were recovered from added for a total EtOH concentration of 70%. This was gently the plates by scraping the top agar from the plates into 2.5 ml mixed and allowed to sit overnight at 20 C to precipitate DNA 1 TM buffer. This extract was homogenized by vortexing and in solution. This solution was then centrifuged for 30 min at clarified by high speed centrifugation. The supernatant was 13 300 rpm at 4 C to pellet the T4 DNA. The supernatant was then filtered through a 0.45 mm sterile filter, and phages were discarded and 500 ll of 70% EtOH was added. The pellet was pelleted in a Ti-50.2 ultracentrifuge rotor at 30 000 rpm, 4 C for washed by flicking the tube and re-centrifuged at 13 000 rpm at 90 min (Beckman Coulter). Pelleted phage were resuspended 4 C, discarding the ethanol after. This step was repeated again overnight at 4 Cin 1 TM buffer. Phages were adsorbed to and for 2 total washes. The final pellet was allowed to air dry for imaged on carbon-coated formvar grids stained with 2% w/v 20 min at room temperature before resuspension of the DNA uranyl acetate in a Tecnai Spirit TEM (FEI). pellet in 100 ll of Millipore water. The final genomic concentra- tion was determined using a NanoDrop One at 260 nm. Following genomic extraction, a fraction of T4 DNA was plated 3. Results and discussion on agar with and without an E. coli B carpet to ensure that no The goal of the first experiment was to plot the kinetics of phage plaques formed, indicating all phage were disrupted by the synthesis to determine the reaction incubation time and the process or that there was no bacterial contamination, potential yield of the system. Samples from a cell-free reaction respectively. set at the optimum biochemical settings (1 nM genome, 5 mM Mg, 60 mM K, 3% PEG8000) were plated at 1 h interval. No detect- 2.4 TXTL system and T4 TXTL reactions able phages were synthesized during the first 2 h. Most of the Preparation of the all-E. coli TXTL system (myTXTL, Arbor phages were synthesized between 2 and 4 h of incubation Biosciences) used in this work was described previously in sev- (Figure 2). Phage yield increases slightly after 4 h to plateau eral articles (11, 25). Transcription and translation are performed between 6 and 10 h. Although slightly slower than T7 (19), the by the endogenous molecular components provided by an E. coli T4 synthesis kinetics in TXTL presents the same trends as T7, cytoplasmic extract, without addition of exogenous purified with a burst of phages that lasts a few hours. Subsequently, all TXTL molecular components. TXTL reactions are composed of the TXTL reactions were incubated for 10–12 h. Downloaded from https://academic.oup.com/synbio/article-abstract/3/1/ysy002/4821891 by Ed 'DeepDyve' Gillespie user on 16 March 2018 4| Synthetic Biology, 2018, Vol. 3, No. 1 Our next experiment consisted in varying the genome con- Potassium glutamate (K-glu) and magnesium glutamate centration in TXTL reaction. The goal was to probe the production (Mg-glu) are essential ions for transcription and translation capacity limit of the synthesis reaction, using the optimal reac- reactions and therefore highly important to be set at precise tions settings described above. Genome concentrations above the concentrations for T4 TXTL synthesis. Across a respective range saturating limit are extraneous and genome concentrations of 0–100 mM K-glu and 4–7 mM Mg-glu, optimum concentrations for K-glu were found to be between 40 and 80 mM while there below optimum do not fully exploit the assembly pathways and was no statistical difference between concentrations tested chemical energy present in the reaction. We observed that at using Mg-glu (Figure 4). Outside of the 4–7 mM Mg-glu concen- 1 nM and above the reaction was saturated (Figure 3). Throughout tration range phage synthesis was notably smaller. The opti- the course of this work, T4 phage synthesis reached 10 PFU/ml mized concentrations that were used in reactions were 60 mM several times. It corresponds to about 1.66 10 phage produced and 5 mM K-glu and Mg-glu, respectively. per genome added to the reaction. We did not observe increased It has been well established theoretically and experimentally T4 synthesis when dNTPs were added to the cell-free reaction, that molecular crowding is a fundamental aspect of macromo- suggesting that DNA replication does not occur. However, one lecular assembly, especially in biological systems (27–29). limitation to address quantitatively this question was the high Molecular crowding favors self-assembly by an entropy-driven viscosity of the reaction due to the size of the T4 genome. In TXTL, some of the components show correlations to process. In a solution containing large concentrations of macro- phage yield when their concentrations are varied, such as ion molecules, such as a cytoplasm or an extract-based cell-free expression system, a net gain of entropy is created by decreas- concentration and molecular crowders. We optimized the con- ing the excluded volume when macromolecules self-assemble. centrations of magnesium, potassium, and PEG for T4 synthesis. Figure 2. Kinetics of T4 TXTL synthesis. A concentration of 1 nM T4 genome and Figure 3. T4 TXTL synthesis as a function of genome concentration. Cell-free 3% PEG was used for this experiment. No phage is produced during the first 2 h reactions with no genome added had no plaques. Standard TXTL reactions were of incubation. T4 synthesis reaches plateau within just a few hours, typically carried out with 3% PEG. Reactions were incubated for 12 h. The optimum phage 7 9 between 10 and 10 . synthesis is observed at 1 nM genome. Figure 4. Magnesium and potassium settings for T4 TXTL synthesis. (A) Optimum phage synthesis is observed for a magnesium concentration of 4–7 mM. Phage yields are smaller outside this range. (B) 40–80 mM potassium is optimum for T4 TXTL synthesis. Downloaded from https://academic.oup.com/synbio/article-abstract/3/1/ysy002/4821891 by Ed 'DeepDyve' Gillespie user on 16 March 2018 M. Rustad et al. | 5 It has been directly observed and modeled for the assembly of entire phages are expressed. By just increasing the PEG8000 viruses such as HIV in vitro (30–32). Molecular crowding also concentration to 3–4%, we could increase TXTL T7 synthesis by alters the rates and the equilibrium constants of biochemical a few orders of magnitude. We varied the concentration of reactions by increasing the effective concentrations of macro- PEG8000 (1% ¼ 1.25 mM) in a TXTL reaction at optimal settings molecules, promoting in particular the association of macromo- for T4 synthesis (1 nM T4 genome, 5 mM Mg-glu and 60 mM lecules (33, 34). Consequently, molecular crowding in TXTL K-glu). The largest effect from any optimized parameter was the reactions is an essential parameter to vary, especially when PEG concentration, showing a >100 000-fold PFU/ml increase self-assembled systems are expressed. We already demon- when increasing from 0.5% to 1.5% v/v (Figure 5). strated the dramatic effect of PEG8000, a typical crowder for To eliminate any source of artifacts and contaminations, we in vitro reactions, on the synthesis of the phage T7 in TXTL (10). performed eight types of controls during all stages of experi- In the case of T7, we showed that the standard molecular crow- mentation throughout the course of the work (Figure 6A), from ders settings in TXTL, 1.5–2% PEG8000, are not optimal when genome extractions to cell-free reactions. Each of the T4 genome preparations used in this study was plated on LB agar multiple times to detect contaminating host cells (control 1). None were detected in the 30 repeats made. A similar control, done in parallel with control 1, was made by mixing the T4 genome with the E. coli host B (control 2) to test random T4 genome uptake by host cells. No plaque was formed in the 30 repeats. Control 1 and 2 covered all the T4 stocks prepared for this work. There were no observed plaques when the T4 genome was added to a cell-free reaction and incubated for <1 min (control 3) in the 10 performed repeats. When rifampicin (100 lg/ml or 122 mM), an inhibitor of the E. coli core RNA poly- merase already tested in TXTL (10), was added to a cell-free reaction at time 0 with 1 nM T4 genome, no plaque was observed (control 4). When DNase I (1 mg/ml) was added to a cell-free reaction at time 0 with 1 nM T4 genome, no plaque was observed (Control 5). However, when DNase I was added to the cell-free reaction after 12 h of incubation and incubated for 30 min at 37 C, an average of 10 plaque forming unit per millili- ter (PFU/ml) was measured (control 6). In each of the final Figure 5. T4 TXTL synthesis as a function of PEG8000 concentration. Standard repeats, the equivalent of seven cell-free reactions of 12 ml (no TXTL reactions were carried out with optimum settings: 5 mM magnesium, 60 mM potassium, 1 nM genome. Maximum TXTL T4 synthesis occurs around DNA added, 84 ml total) was plated at 0 (Control 7) and 12 h 3% PEG. The phage yields varies by orders of magnitude, as observed previously (Control 8) incubation. No colonies were observed when plated for the other phages UX174 and T7 (10). showing that our cell-free expression system does not contain Figure 6. Experimental controls. (A) Table summarizing the controls and repeats. These controls were carried out during the course of experimentation. (B) Example of a set of controls including a positive (labeled as þ, TXTL reaction containing 1 nM T4 genome) and controls 6, 5 and 4. (C) Example of control plates (negative on the left with a lawn of cells and positive on the right) for T4 TXTL synthesis. (D) Electron microscopy image of T4 phages from plaques produced by a TXTL reaction containing 1 nM of the T4 genome. Downloaded from https://academic.oup.com/synbio/article-abstract/3/1/ysy002/4821891 by Ed 'DeepDyve' Gillespie user on 16 March 2018 6| Synthetic Biology, 2018, Vol. 3, No. 1 intact E. coli cells, as discussed previously (10). An example of prototyping genetic networks with cell-free transcription- the controls (Figure 6B) was performed as follows: a cell-free translation reactions. Methods, 86, 60–72. reaction containing 1 nM T4 genome was prepared and then 3. Bundy,B.C., Franciszkowicz,M.J. and Swartz,J.R. (2008) split into equal volume reactions to which was added (i) water Escherichia coli-based cell-free synthesis of virus-like particles. (positive control), (ii) rifampicin at time 0 (100 lg/ml or 122 mM), Biotechnol. Bioeng., 100, 28–37. (iii) DNase I at time 0 (1 mg/ml) and (iv) DNase I added after 12 h 4. Carlson,E.D., Gan,R., Hodgman,C.E. and Jewett,M.C. (2011) of incubation. The plaque assay for T4 was carried out according Cell-free protein synthesis: applications come of age. to standard procedures (Figure 6C). We concluded that the syn- Biotechnol. Adv., 30, 1185–1194. thesis of the phage T4 measured on plaques does not come 5. Dudley,Q.M., Karim,A.S. and Jewett,M.C. (2014) Cell-free met- from stock contamination. Despite multiple attempts, it was abolic engineering: biomanufacturing beyond the cell. not possible to directly visualize the phage T4 from TXTL reac- Biotechnol. J., 10, 69–82. tions by electron microscopy (EM) because of the high viscosity 6. Pardee,K., Slomovic,S., Nguyen,P.Q., Lee,J.W., Donghia,N., of the solutions and the relatively low number of phages syn- Burrill,D., Ferrante,T., McSorley,F.R., Furuta,Y., Vernet,A. thesized (10 PFU/ml at best). As an additional verification step, et al. (2016) Portable, on-demand biomolecular manufactur- however, we visualized the phage T4 by EM from plaques pro- ing. Cell, 167, 248–259.e212. duced from a TXTL reaction (Figure 6D), thus confirming cell- 7. Perez,J.G., Stark,J.C. and Jewett,M.C. (2016) Cell-free synthetic free synthesis of the phage T4. biology: engineering beyond the cell. Cold Spring Harb. Perspect. Biol., 8, a023853. 8. Shin,J. and Noireaux,V. (2012) An E. coli cell-free expression 4. Summary and conclusions toolbox: application to synthetic gene circuits and artificial The objective of this letter is to report the synthesis of the largest cells. ACS Synth. Biol., 1, 29–41. biological entity so far achieved in a TXTL reaction. By demon- 9. Caschera,F. and (2014) Synthesis of 2.3 mg/ml of protein strating that TXTL systems can process a DNA program larger with an all Escherichia coli cell-free transcription-translation than 100 kbp encoding for about 100 essential genes, we bring the system. Biochimie, 99, 162–168. capabilities of the cell-free technology to a new level. We believe 10. Garamella,J., Marshall,R., Rustad,M. and Noireaux,V. (2016) that the total synthesis of a phage of this size and complexity is a The all E. coli TX-TL toolbox 2.0: a platform for cell-free syn- significant milestone towards the TXTL execution of a minimal thetic biology. ACS Synth. Biol., 5, 344–355. cell genome for self-reproduction of a living entity. It has been 11. Shin, J. and Noireaux, V. (2010). Efficient cell-free expression estimated by different approaches and groups that the minimal with the endogenous E. coli RNA polymerase and sigma fac- set of genes required by a free-living cell is between 300 and 500 tor 70. J. Biol. Eng., 4, 8. genes (35–38). The complete synthesis of T4 in an open environ- 12. Niederholtmeyer,H., Sun,Z.Z., Hori,Y., Yeung,E., Verpoorte,A., ment like TXTL offers unique possibilities to interrogate quantita- Murray,R.M. and Maerkl,S.J. (2015) Rapid cell-free forward engi- tively the links between gene expression, self-assembly and neering of novel genetic ring oscillators. Elife,4, e09771. metabolism, which is left for a future study. Our future work will 13. Sun,Z.Z., Yeung,E., Hayes,C.A., Noireaux,V. and Murray,R.M. also consist of elucidating a few surprising results. For example, (2014) Linear DNA for rapid prototyping of synthetic biologi- the literature suggests that the phage T4 initiates its self- cal circuits in an Escherichia coli based TX-TL cell-free system. assembly at the inner membrane of E. coli (20). It is unlikely to ACS Synth. Biol., 3, 387–397. happen in TXTL because addition of liposomes prepared from E. 14. Takahashi,M.K., Chappell,J., , Sun,Z.Z., Kim,J., coli membranes to TXTL reactions did not increase phage synthe- Singhal,V., Spring,K.J., Al-Khabouri,S., Fall,C.P., Noireaux,V. sis (not shown). This observation suggests that there are other et al. (2015a) Rapidly characterizing the fast dynamics of RNA thermodynamic modes that allow efficient self-assembly of com- genetic circuitry with cell-free transcription-translation plex biological systems in in vitro reactions lacking natural tem- (TX-TL) systems. ACS Synth. Biol., 4, 503–515. plates used for self-organization. The demonstration that MS2, 15. Caschera,F. and Noireaux,V. (2016) Compartmentalization of UX174, T7 and now T4 can be entirely produced in TXTL reactions an all-E. coli cell-free expression system for the construction also paves the way toward phage engineering for biomedical of a minimal cell. Artif. Life, 22, 185–195. applications. We expect that the all-E. coli TXTL platform can be 16. Karzbrun,E., Tayar,A.M., Noireaux,V. and Bar-Ziv,R.H. (2014) employed to create synthetic phages, by recombineering for Programmable on-chip DNA compartments as artificial cells. example, useful for phage therapy or other applications (39). Science, 345, 829–832. 17. Majumder,S., Garamella,J., Wang,Y.L., DeNies,M., Noireaux,V. and Liu,A.P. (2017) Cell-sized mechanosensitive and biosens- Funding ing compartment programmed with DNA. Chem. Commun., The Office of Naval Research [N00014-13-1-0074]; the Human 53, 7349–7352. Frontier Science Program [RGP0037/2015]. Noireaux laboratory 18. Tayar, A., Karzbrun, E., Noireaux, V. and Bar-Ziv, R. (2015) receives research funds from Arbor Biosciences, a distributor of Propagating gene expression fronts in a one-dimensional myTXTL cell-free protein expression kit. coupled system of artificial cells. Nat. Phys., 19. Shin,J., Jardine,P. and (2012) Genome replication, syn- Conflict of interest statement. None declared. thesis, and assembly of the bacteriophage T7 in a single cell-free reaction. ACS Synth. Biol., 1, 408–413. References 20. Yap,M.L. and Rossmann,M.G. (2014) Structure and function of bacteriophage T4. Future Microbiol., 9, 1319–1327. 1. Hockenberry,A.J. and Jewett,M.C. (2012) Synthetic in vitro cir- cuits. Curr. Opin. Chem. Biol., 16, 253–259. 21. Miller,E.S., Kutter,E., Mosig,G., Arisaka,F., Kunisawa,T. and 2. Takahashi,M.K., Hayes,C.A., Chappell,J., Sun,Z.Z., Murray,R.M., Ruger,W. (2003) Bacteriophage T4 genome. Microbiol. Mol. Biol. Noireaux,V. and Lucks,J.B. (2015b) Characterizing and Rev., 67, 86–156, table of contents. Downloaded from https://academic.oup.com/synbio/article-abstract/3/1/ysy002/4821891 by Ed 'DeepDyve' Gillespie user on 16 March 2018 M. Rustad et al. | 7 31. Rivas,G., Fernandez,J.A. and Minton,A.P. (2001) Direct obser- 22. Mesyanzhinov,V.V., Leiman,P.G., Kostyuchenko,V.A., Kurochkina,L.P., Miroshnikov,K.A., Sykilinda,N.N. and vation of the enhancement of noncooperative protein Shneider,M.M. (2004) Molecular architecture of bacteriophage self-assembly by macromolecular crowding: indefinite linear T4. Biochemistry (Mosc), 69, 1190–1202. self-association of bacterial cell division protein FtsZ. Proc. 23. Wood,W.B., Edgar,R.S., King,J., Lielausis,I. and Henninger,M. Natl. Acad. Sci. USA, 98, 3150–3155. (1968) Bacteriophage assembly. Fed. Proc., 27, 1160–1166. 32. Smith,G.R., Xie,L., Lee,B. and Schwartz,R. (2014) Applying 24. Chen,X., Mathews,C.K., Wheeler,L.J., Maley,G., Maley,F. and molecular crowding models to simulations of virus capsid Coombs,D.H. (1995) An immunoblot assay reveals that assembly in vitro. Biophys. J., 106, 310–320. bacteriophage T4 thymidylate synthase and dihydrofolate 33. Minton,A.P. (1983) The effect of volume occupancy upon the reductase are not virion proteins. J. Virol., 69, 2119–2125. thermodynamic activity of proteins: some biochemical con- 25. Sun,Z.Z., Hayes,C.A., Shin,J., Caschera,F., Murray,R.M. and sequences. Mol. Cell. Biochem., 55, 119–140. Noireaux,V. (2013) Protocols for implementing an Escherichia 34. (1997) Influence of excluded volume upon macromolec- 0 0 coli based TX-TL cell-free expression system for synthetic ular structure and associations in crowded media. Curr. biology. J. Vis. Exp., e50762. Opin. Biotechnol., 8, 65–69. 26. Caschera,F. and Noireaux,V. (2015) Preparation of amino acid 35. Gil,R., Sabater-Munoz,B., Latorre,A., Silva,F.J. and Moya,A. mixtures for cell-free expression systems. Biotechniques, 58, (2002) Extreme genome reduction in Buchnera spp.: toward 40–43. the minimal genome needed for symbiotic life. Proc. Natl. 27. Minton,A.P. (2000) Implications of macromolecular Acad. Sci. USA, 99, 4454–4458. crowding for protein assembly. Curr. Opin. Struct. Biol., 10, 36. , Silva,F.J., Pereto,J. and Moya,A. (2004) Determination of 34–39. the core of a minimal bacterial gene set. Microbiol. Mol. Biol. 28. (2006) How can biochemical reactions within cells differ Rev., 68, 518–537. from those in test tubes? J. Cell Sci., 119, 2863–2869. 37. Glass,J.I., Assad-Garcia,N., Alperovich,N., Yooseph,S., 29. Zimmerman,S.B. and Minton,A.P. (1993) Macromolecular Lewis,M.R., Maruf,M., Hutchison,C.A. 3rd, Smith,H.O. and crowding: biochemical, biophysical, and physiological conse- Venter,J.C. (2006) Essential genes of a minimal bacterium. quences. Annu. Rev. Biophys. Biomol. Struct., 22, 27–65. Proc. Natl. Acad. Sci. USA, 103, 425–430. 30. del Alamo,M., Rivas,G. and Mateu,M.G. (2005) Effect of macro- 38. Zhang,L.Y., Chang,S.H. and Wang,J. (2010) How to make a mini- molecular crowding agents on human immunodeficiency mal genome for synthetic minimal cell. Protein Cell, 1, 427–434. virus type 1 capsid protein assembly in vitro. J. Virol., 79, 39. Reardon,S. (2014) Phage therapy gets revitalized. Nature, 510, 14271–14281. 15–16. Downloaded from https://academic.oup.com/synbio/article-abstract/3/1/ysy002/4821891 by Ed 'DeepDyve' Gillespie user on 16 March 2018

Journal

Synthetic BiologyOxford University Press

Published: Jan 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off