A semi-synthetic organism with an expanded genetic alphabet
A semi-synthetic organism with an expandedgenetic alphabet Denis A. Malyshev1, Kirandeep Dhami1, Thomas Lavergne1, Tingjian Chen1, Nan Dai2, Jeremy M. Foster2, Ivan R. Corre & Floyd E. Romesberg1 Organisms are defined by the information encoded in their genomes, suggest that decomposition is mediated by phosphatases. As no degra- and since the origin of life this information has been encoded using a dation was observed upon incubation in spent media, decomposition two-base-pair genetic alphabet (A–T and G–C). In vitro, the alphabet seems to occur within the periplasm. No increase in stability was observed has been expanded to include several unnatural base pairs (UBPs)1–3.in cultures of single-gene-deletion mutants of E. coli BW25113 lacking We have developed a class of UBPs formed between nucleotides bear- a specific periplasmic phosphatase19 (as identified by the presence of a ing hydrophobic nucleobases, exemplified by the pair formed between Sec-type amino-terminal leader sequence), including phoA, ushA, appA, d5SICS and dNaM (d5SICS–dNaM), which is efficiently PCR-amplified1 aphA, yjjX, surE, yfbR, yjjG, yfaO, mutT, nagD, yggV, yrfG or ymfB, and transcribed4,5 in vitro, and whose unique mechanism of repli- suggesting that decomposition results from the activity of multiple phos- cation has been characterized6,7. However, expansion of an organ- phatases. However, the extracellular stability of [a-32P]-dATP was sig- ism's genetic alphabet presents new and unprecedented challenges: nificantly greater when 50 mM potassium phosphate (KPi) was added to the unnatural nucleoside triphosphates must be available inside the the growth medium (Extended Data Fig. 3). Thus, we measured [a-32P]- cell; endogenous polymerases must be able to use the unnatural tri- dATP uptake from media containing 50 mM KPi after induction of the phosphates to faithfully replicate DNA containing the UBP within transporter with isopropyl-b-D-thiogalactoside (IPTG) (Extended Data the complex cellular milieu; and finally, the UBP must be stable in Fig. 4). Although induction with 1 mM IPTG resulted in slower growth, the presence of pathways that maintain the integrity of DNA. Here consistent with the previously reported toxicity of NTTs17, it also resulted we show that an exogenously expressed algal nucleotide triphosphate in maximal [a-32P]-dATP uptake. Thus, after addition of 1 mM IPTG, transporter efficiently imports the triphosphates of both d5SICS and we analysed the extracellular and intracellular stability of [a-32P]-dATP dNaM (d5SICSTP and dNaMTP) into Escherichia coli, and that the as a function of time (Extended Data Fig. 5). Cells expressing PtNTT2 endogenous replication machinery uses them to accurately replicate were found to have the highest levels of intracellular [a-32P]-dATP, and a plasmid containing d5SICS–dNaM. Neither the presence of the although both extra- and intracellular dephosphorylation was still observed, unnatural triphosphates nor the replication of the UBP introduces a the ratio of triphosphate to dephosphorylation products inside the cell notable growth burden. Lastly, we find that the UBP is not efficiently remained roughly constant, indicating that the extracellular concen- excised by DNA repair pathways. Thus, the resulting bacterium is the trations and PtNTT2-mediated influx are sufficient to compensate for first organism to propagate stably an expanded genetic alphabet.
To make the unnatural triphosphates available inside the cell, we previ- Likewise, we found that the addition of KPi increased the extracel- ously suggested using passive diffusion of the free nucleosides into the lular stability of d5SICSTP and dNaMTP (Extended Data Fig. 2), and cytoplasm followed by their conversion to the corresponding tripho-sphate via the nucleoside salvage pathway8. Although we have shown that analogues of d5SICS and dNaM are phosphorylated by the nucle- oside kinase from Drosophila melanogaster8, monophosphate kinases are more specific9, and in E. coli we found that overexpression of the endogenous nucleoside diphosphate kinase results in poor growth. As an alternative, we focused on the nucleotide triphosphate transport- ers (NTTs) of obligate intracellular bacteria and algal plastids10–14. We expressed eight different NTTs in E. coli C41(DE3)15–17 and measured the uptake of [a-32P]-dATP as a surrogate for the unnatural triphosphates (Extended Data Fig. 1). We confirmed that [a-32P]-dATP is efficiently transported into cells by the NTTs from Phaeodactylum tricornutum (PtNTT2)18 and Thalassiosira pseudonana (TpNTT2)18. Although NTTs from Protochlamydia amoebophila (PamNTT2 and PamNTT5)15 also import [a-32P]-dATP, PtNTT2 showed the most activity, and both it and TpNTT2 are known to have broad specificity18, making them the most promising NTTs for further characterization.
Transport via an NTT requires that the unnatural triphosphates are sufficiently stable in culture media; however, preliminary characteriza- Figure 1 Nucleoside triphosphate stability and import. a, Chemical tion of d5SICSTP and dNaMTP indicated that decomposition occurs structure of the d5SICS–dNaM UBP compared to the natural dG–dC base pair.
b, Composition analysis of d5SICS and dNaM in the media (top) and in the presence of actively growing E. coli (Extended Data Fig. 2). Similar cytoplasmic (bottom) fractions of cells expressing PtNTT2 after 30 min behaviour was observed with [a-32P]-dATP, and the dephosphorylation incubation; dA shown for comparison. 3P, 2P, 1P and 0P correspond to products detected by thin-layer chromatography (TLC) for [a-32P]-dATP, triphosphate, diphosphate, monophosphate and nucleoside, respectively; [3P] or by high-performance liquid chromatography (HPLC) and matrix- is the intracellular concentration of triphosphate. Error bars represent s.d.
assisted laser desorption/ionization (MALDI) for d5SICSTP and dNaMTP, of the mean, n 5 3.
1Department of Chemistry, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037, USA. 2New England Biolabs, 240 County Road, Ipswich, Massachusetts 01938, USA.
2014 Macmillan Publishers Limited. All rights reserved
when a stationary phase culture was diluted 100-fold into fresh media, for our purposes. Thirty minutes after their addition to the media, neither the half-lives of both unnatural triphosphates (initial concentrations of of the unnatural triphosphates was detected in cells expressing TpNTT2; 0.25 mM) were found to be approximately 9 h, which seemed sufficient in contrast, 90 mM of d5SICSTP and 30 mM of dNaMTP were found in the cytoplasm of cells expressing PtNTT2 (Fig. 1b). Although intracel- lular decomposition was still apparent, the intracellular concentrations of intact triphosphate are significantly above the sub-micromolar KM lac operator lac promoter values of the unnatural triphosphates for DNA polymerases20, setting TK-1 Okazaki processing site Primase recognition sequence the stage for replication of the UBP in a living bacterial cell.
The replication of DNA containing d5SICS-dNaM has been validated in vitro with different polymerases, primarily family A polymerases, such as the Klenow fragment of E. coli DNA polymerase I (pol I)20,21. As the majority of the E. coli genome is replicated by pol III, we engineered a lacI promoter lac operator lac prom plasmid to focus replication of the UBP to pol I. Plasmid pINF (the infor- mation plasmid) was constructed from pUC19 using solid-phase DNA lac operator synthesis and circular-extension PCR to replace the dA–dT pair at posi- tion 505 with dNaM paired opposite an analogue of d5SICS (dTPT322) (Fig. 2a, b). This positions the UBP 362 bp downstream of the ColE1 origin of replication where leading-strand replication is mediated by pol I23, and within the TK-1 Okazaki processing site24, where lagging- strand synthesis is also expected to be mediated by pol I. Synthetic pINF was constructed using the d5SICS analogue because it should be efficiently replaced by d5SICS if replication occurs in vivo, making it possible to differentiate in vivo replicated pINF from synthetic pINF.
To determine whether E. coli can use the imported unnatural triphos- phates to stably propagate pINF, C41(DE3) cells were first transformed with a pCDF-1b plasmid encoding PtNTT2 (hereafter referred to as pACS, for accessory plasmid, Fig. 2a) and grown in media containing 0.25 mM of both unnatural triphosphates, 50 mM KPi and 1 mM IPTG to induce transporter production. Cells were then transformed with pINF, and after a 1-h recovery period, cultures were diluted tenfold with the same media supplemented with ampicillin, and growth was monitored via culture turbidity (Extended Data Table 1). As controls, cells were also transformed with pUC19, or grown without either IPTG or without the unnatural triphosphates. Again, growth was significantly slower in the presence of IPTG, but the addition of d5SICSTP and dNaMTP resulted in only a slight further decrease in growth in the absence of pINF, and interest- ingly, it eliminated a growth lag in the presence of pINF (Fig. 2c), suggesting that the unnatural triphosphates are not toxic and are required E. coli pACS for the efficient replication of pINF.
To demonstrate the replication of pINF, we recovered the plasmid from cells after 15 h of growth. The introduction of the UBP resulted in Figure 2 Intracellular UBP replication. a, Structure of pACS and pINF.
dX and dY correspond to dNaM and a d5SICS analogue22 that facilitated plasmid construction (see Methods). cloDF, origin of replication; Sm, streptomycin resistance gene; AmpR, ampicillin resistance gene; ori, ColE1 origin of replication; lacZa, b-galactosidase fragment gene. b, Overview of pINF construction. A DNA fragment containing the unnatural nucleotide was synthesized via solid-phase DNA synthesis and then used to assemblesynthetic pINF via circular-extension PCR29. X, dNaM; Y9, dTPT3 (an analogueof d5SICS22); y, d5SICS (see text). Colour indicates regions of homology. The doubly nicked product was used directly to transform E. coli harbouring pACS.
c, The addition of d5SICSTP and dNaMTP eliminates a growth lag of cells harbouring pINF. EP, electroporation. Error bars represent s.d. of the mean,n 5 3. d, LC-MS/MS total ion chromatogram of global nucleoside content inpINF and pUC19 recorded in dynamic multiple reaction monitoring (DMRM) mode. pINF and pUC19 (control) were propagated in E. coli in the presence or absence of unnatural triphosphates, and with or without PtNTT2 induction.
The inset shows a 100-fold expansion of the mass-count axis in the d5SICS region. e, Biotinylation only occurs in the presence of the UBP, the unnatural triphosphates and transporter induction. After growth, pINF was recovered, and a 194-nucleotide region containing the site of UBP incorporation (nucleotides 437–630) was amplified and biotinylated. B, biotin; SA, streptavidin. The natural pUC19 control plasmid was prepared identically to pINF. A 50-bp DNA ladder is shown to the left. f, Sequencing analysis demonstrates retention of the UBP. An abrupt termination in the Sangersequencing reaction indicates the presence of UBP incorporation (site indicated Acquisition time (min) with arrow).
2014 Macmillan Publishers Limited. All rights reserved
a small (approximately twofold) reduction in the copy number of pINF, as gauged by its ratio to pACS (Extended Data Table 1); we determinedthat the plasmid was amplified 2 3 107-fold during growth (approxi-mately 24 doublings) based on the amount of recovered plasmid and the transformation efficiency. To determine the level of UBP retention,the recovered plasmid was digested, dephosphorylated to single nucle-osides, and analysed by liquid chromatography-tandem mass spectrom- etry (LC-MS/MS)25. Although the detection and quantification of dNaM were precluded by its poor fragmentation efficiency and low production counts over background, signal for d5SICS was clearly observable (Fig. 2d). External calibration curves were constructed using the unnat-ural nucleoside and validated by determining its ratio to dA in synthetic oligonucleotides (Extended Data Table 2). Using the resulting calibra- tion curve, we determined the ratio of dA to d5SICS in recovered pINFwas 1,106 to 1, which when compared to the expected ratio of 1,325to 1, suggests the presence of approximately one UBP per plasmid. No d5SICS was detected in control experiments in which the transporter was not induced, or when the unnatural triphosphates were not addedto the media, or when pUC19 was used instead of pINF (Fig. 2d, inset), Figure 3 Intracellular stability of the UBP. E. coli C41(DE3) pACS was demonstrating that its presence results from the replication of the UBP transformed with pINF and grown after a single dose of d5SICSTP and and not from misinsertion of the unnatural triphosphates opposite a natural dNaMTP was provided in the media. UBP retention in recovered pINF, OD600,and relative amount of d5SICSTP and dNaMTP in the media (100% 5 0.25 nucleotide. Importantly, as the synthetic pINF contained an analogue mM), were determined as a function of time. Error bars represent s.d. of the of d5SICS, and d5SICS was only provided as a triphosphate added to mean, n 5 3.
the media, its presence in pINF confirms in vivo replication.
To independently confirm and quantify the retention of the UBP in with replication (Extended Data Fig. 7b, c, d), but even then, retention the recovered plasmid, the relevant region was amplified by PCR in the of the UBP remained at approximately 45% and 15%, at days 3 and 6 presence of d5SICSTP and a biotinylated dNaMTP analogue4 (Fig. 2e).
respectively. Moreover, when d5SICS-dNaM was lost, it was replaced Analysis by streptavidin gel shift showed that 67% of the amplified DNA by dA–dT, which is consistent with the mutational spectrum of DNA contained biotin. No shift was observed in control experiments where pol I20. Finally, the shape of the retention versus time curve mirrors that the transporter was not induced, or when unnatural triphosphates were of the growth versus time curve. Taken together, these data suggest that not added, or when pUC19 was used instead of pINF, demonstrating in the absence of unnatural triphosphates, the UBP is eventually lost that the shift results from the presence of the UBP. Based on a calibra- by replication-mediated mispairing, and not from the activity of DNA tion curve constructed from the shifts observed with the amplification repair pathways.
products of controlled mixtures of DNA containing dNaM or its fully We have demonstrated that PtNTT2 efficiently imports d5SICSTP natural counterpart (Methods and Extended Data Fig. 6), the observed and dNaMTP into E. coli and that an endogenous polymerase, possibly gel shift corresponds to a UBP retention of 86%. Similarly, when the ampli- pol I, efficiently uses the unnatural triphosphates to replicate DNA con- fication product obtained with d5SICSTP and dNaMTP was analysed taining the UBP within the cellular environment with reasonable efficiency by Sanger sequencing in the absence of the unnatural triphoshates1,26,27, and fidelity. Moreover, the UBP appears stable during both exponen- the sequencing chromatogram showed complete termination at the posi- tial and stationary phase growth despite the presence of all DNA repair tion of UBP incorporation, which with an estimated lower limit of read- mechanisms. Remarkably, although expression of PtNTT2 results in a through detection of 5%, suggests a level of UBP retention in excess of somewhat reduced growth rate, neither the unnatural triphosphates nor 95% (Fig. 2f). In contrast, amplification products obtained from pINF replication of the UBP results in significant further reduction in growth.
recovered from cultures grown without PtNTT2 induction, without added The resulting bacterium is the first organism that stably harbours DNA unnatural triphosphates, or obtained from pUC19 propagated under containing three base pairs. In the future, this organism, or a variant with identical conditions, showed no termination. Overall, the data unam- the UBP incorporated at other episomal or chromosomal loci, should biguously demonstrate that DNA containing the UBP was replicated provide a synthetic biology platform to orthogonally re-engineer cells, in vivo and allow us to estimate that replication occurred with fidelity with applications ranging from site-specific labelling of nucleic acids (retention per doubling) of at least 99.4% (24 doublings; 86% retention; in living cells to the construction of orthogonal transcription networks 0.99424 5 0.86). This fidelity corresponds to an error rate of approxi- and eventually the production and evolution of proteins with multiple, mately 1023, which is comparable to the intrinsic error rate of some poly- different unnatural amino acids.
merases with natural DNA28.
The high retention of the UBP over a 15-h period of growth (approx- imately 24 doublings) strongly suggests that it is not efficiently excised To prepare electrocompetent C41(DE3) pACS cells, freshly transformed E. coli by DNA repair pathways. To test further this hypothesis and to exam- C41(DE3) pACS was grown overnight in 2 3 YT medium (1.6% tryptone, 1% yeast ine retention during prolonged stationary phase growth, we repeated extract, 0.5% NaCl) supplemented with streptomycin and KPi. After 100-fold dilu- the experiments, but monitored UBP retention, cell growth and unnat- tion into the same medium and outgrowth at 37 uC to OD600 5 0.20, IPTG was added ural triphosphate decomposition for up to 6 days without providing any to induce expression of PtNTT2. After 40 min, cultures were rapidly cooled, washed additional unnatural triphosphates (Fig. 3 and Extended Data Fig. 7). At with sterile water and resuspended in 10% glycerol. An aliquot of electrocompetent 15 and 19 h of growth, the cultures reached an optical density at 600 nm cells was mixed with pINF and electroporated. Pre-warmed 2 3 YT medium con- taining streptomycin, IPTG and KPi was added, and an aliquot was diluted 3.3-fold 600) of approximately 0.9 and 1.2, respectively, and both d5SICSTP and dNaMTP decomposed to 17–20% and 10–16% of their initial 0.25-mM in the same media supplemented with 0.25 mM each of dNaMTP and d5SICSTP.
The resulting mixture was allowed to recover at 37 uC with shaking. After recovery, concentrations (Extended Data Fig. 7a). In agreement with the experiments cultures were centrifuged. Spent media was analysed for nucleotide composition by described above, retention of the UBP after 15 h was 97 6 5% and .95%, HPLC (Extended Data Fig. 7a); cells were resuspended in fresh medium containing as determined by gel shift and sequencing, respectively, and after 19 h it streptomycin, ampicillin, IPTG, KPi and 0.25 mM each of dNaMTP and d5SICSTP, was 91 6 3% and .95%. As the cultures entered stationary phase and the and grown with shaking. At defined time points, OD600 was determined and ali- triphosphates decomposed completely, plasmid loss began to compete quots were removed and centrifuged. Spent media were analysed for nucleotide 2014 Macmillan Publishers Limited. All rights reserved
composition, and pINF was recovered by spin column purification. UBP retention 17. Haferkamp, I. & Linka, N. Functional expression and characterisation of membrane was characterized by LC-MS/MS, PCR amplification and gel electrophoresis, or transport proteins. Plant Biol. 14, 675–690 (2012).
sequencing, as described in the Methods.
18. Ast, M. et al. Diatom plastids depend on nucleotide import from the cytosol. Proc.
Natl Acad. Sci. USA 106, 3621–3626 (2009).
Online Content Any additional Methods, Extended Data display items and Source 19. Baba, T. et al. Construction of Escherichia coli K-12 in-frame, single-gene knockout Data are available in the ; references unique to these mutants: the Keio collection. Mol. Syst. Biol. 2, 2006.0008 (2006).
sections appear only in the online paper.
20. Lavergne, T., Malyshev, D. A. & Romesberg, F. E. Major groove substituents and polymerase recognition of a class of predominantly hydrophobic unnatural basepairs. Chemistry 18, 1231–1239 (2012).
Received 27 November 2013; accepted 8 April 2014.
21. Seo, Y. J., Hwang, G. T., Ordoukhanian, P. & Romesberg, F. E. Optimization of an Published online 7 May 2014.
unnatural base pair toward natural-like replication. J. Am. Chem. Soc. 131,3246–3252 (2009).
Malyshev, D. A. et al. Efficient and sequence-independent replication of DNA 22. Li, L. et al. Natural-like replication of an unnatural base pair for the expansion of the containing a third base pair establishes a functional six-letter genetic alphabet.
genetic alphabet and biotechnology applications. J. Am. Chem. Soc. 136, 826–829 Proc. Natl Acad. Sci. USA 109, 12005–12010 (2012).
Yang, Z., Chen, F., Alvarado, J. B. & Benner, S. A. Amplification, mutation, and 23. Tomizawa, J. & Selzer, G. Initiation of DNA synthesis in Escherichia coli. Annu. Rev.
sequencing of a six-letter synthetic genetic system. J. Am. Chem. Soc. 133, Biochem. 48, 999–1034 (1979).
24. Allen, J. M. et al. Roles of DNA polymerase I in leading and lagging-strand Yamashige, R. et al. Highly specific unnatural base pair systems as a third base pair replication defined by a high-resolution mutation footprint of ColE1 plasmid for PCR amplification. Nucleic Acids Res. 40, 2793–2806 (2012).
replication. Nucleic Acids Res. 39, 7020–7033 (2011).
Seo, Y. J., Malyshev, D. A., Lavergne, T., Ordoukhanian, P. & Romesberg, F. E.
25. Hashimoto, H. et al. Structure of a Naegleria Tet-like dioxygenase in complex with Site-specific labeling of DNA and RNA using an efficiently replicated and 5-methylcytosine DNA. Nature 506, 391–395 (2013).
transcribed class of unnatural base pairs. J. Am. Chem. Soc. 133, 19878–19888 26. Malyshev, D. A., Seo, Y. J., Ordoukhanian, P. & Romesberg, F. E. PCR with an expanded genetic alphabet. J. Am. Chem. Soc. 131, 14620–14621 (2009).
Seo, Y. J., Matsuda, S. & Romesberg, F. E. Transcription of an expanded geneticalphabet. J. Am. Chem. Soc. 131, 5046–5047 (2009).
27. Hirao, I. et al. An unnatural hydrophobic base pair system: site-specific Betz, K. et al. Structural insights into DNA replication without hydrogen bonds.
incorporation of nucleotide analogs into DNA and RNA. Nature Methods 3, J. Am. Chem. Soc. 135, 18637–18643 (2013).
729–735 (2006).
Betz, K. et al. KlenTaq polymerase replicates unnatural base pairs by inducing a 28. Goodman, M. F. Error-prone repair DNA polymerases in prokaryotes and Watson–Crick geometry. Nature Chem. Biol. 8, 612–614 (2012).
eukaryotes. Annu. Rev. Biochem. 71, 17–50 (2002).
Wu, Y., Fa, M., Tae, E. L., Schultz, P. G. & Romesberg, F. E. Enzymatic 29. Quan, J. & Tian, J. Circular polymerase extension cloning for high-throughput phosphorylation of unnatural nucleosides. J. Am. Chem. Soc. 124, 14626–14630 cloning of complex and combinatorial DNA libraries. Nature Protocols 6, 242–251 Yan, H. & Tsai, M. D. Nucleoside monophosphate kinases: structure, mechanism, Supplementary Information is available in the .
and substrate specificity. Adv. Enzymol. 73, 103–134 (1999).
10. Winkler, H. H. & Neuhaus, H. E. Non-mitochondrial ATP transport. Trends Biochem.
Acknowledgements We thank I. Haferkamp and J. Audia for kindly providing the NTT Sci. 24, 64–68 (1999).
plasmids and helpful discussions, and P. Ordoukhanian for providing access to the 11. Amiri, H., Karlberg, O. & Andersson, S. G. Deep origin of plastid/parasite ATP/ADP Center for Protein and Nucleic Acid Research at TSRI. This work was supported by the translocases. J. Mol. Evol. 56, 137–150 (2003).
US National Institutes of Health (NIH) (GM 060005).
12. Hatch, T. P., Al-Hossainy, E. & Silverman, J. A. Adenine nucleotide and lysine transport in Chlamydia psittaci. J. Bacteriol. 150, 662–670 (1982).
Author Contributions D.A.M., K.D., T.C. and F.E.R. designed the experiments. D.A.M., 13. Winkler, H. H. Rickettsial permeability: an ADP-ATP transport system. J. Biol. Chem.
K.D. and T.L. performed the experiments. N.D., J.M.F. and I.R.C.J. performed LC-MS/MS 251, 389–396 (1976).
analysis. D.A.M., K.D. and F.E.R. analysed data and D.A.M. and F.E.R. wrote the 14. Horn, M. & Wagner, M. Bacterial endosymbionts of free-living amoebae. J. Eukaryot.
manuscript with assistance from the other authors.
Microbiol. 51, 509–514 (2004).
15. Haferkamp, I. et al. Tapping the nucleotide pool of the host: novel nucleotide carrier Author Information Reprints and permissions information is available at proteins of Protochlamydia amoebophila. Mol. Microbiol. 60, 1534–1545 (2006).
The authors declare competing financial interests: details 16. Miroux, B. & Walker, J. E. Over-production of proteins in Escherichia coli: mutant are available in the Readers are welcome to comment on hosts that allow synthesis of some membrane proteins and globular proteins at the . Correspondence and requests for materials should be high levels. J. Mol. Biol. 260, 289–298 (1996).
addressed to F.E.R. 2014 Macmillan Publishers Limited. All rights reserved
incubated with the substrate with shaking at 37 uC for 30 min and then pelleted Materials. 2 3 YT, 2 3 YT agar, IPTG, ampicillin and streptomycin were obtained (8,000 r.c.f. (relative centrifugal force) for 5 min, 4 uC). An aliquot of the media from Fisher Scientific. Ampicillin and streptomycin were used at 100 mg ml21 and fraction (40 ml) was mixed with acetonitrile (80 ml) to precipitate proteins31, and 50 mg ml21, respectively. All pET-16b constructs containing the nucleotide trans- then incubated at 22 uC for 30 min. Samples were either analysed immediately by porters were kindly provided by I. Haferkamp (Technische Universita¨t Kaisers- HPLC or stored at 280 uC until analysis. Analysis began with centrifugation (12,000 lautern, Germany) with the exception of pET16b-RpNTT2, which along with the r.c.f. for 10 min at 22 uC), then the pellet was discarded, and the supernatant was C41(DE3) E. coli strain, was provided by J. P. Audia (University of South Alabama, reduced to approximately 20 ml by SpeedVac, resuspended in buffer A (see below) USA). Plasmids pUC19 and pCDF-1b were obtained from Thermo Scientific and to a final volume of 50 ml, and analysed by HPLC (see below).
EMD Millipore, respectively. Plasmids were purified using the PureLink Quick Preparation of cytoplasmic fraction for nucleoside triphosphate analysis. To Plasmid DNA Miniprep Kit (Life Technologies). OneTaq, DeepVent, Q5 Hot Start analyse the intracellular desphosphorylation of the unnatural nucleoside triphos- High-Fidelity DNA Polymerases, and all restriction endonucleases were obtained phate, cell pellets were subjected to 3 3 100 ml washes of ice-cold KPi (50 mM).
from New England Biolabs. In general, PCR reactions were divided into multiple Pellets were then resuspended in 250 ml of ice cold KPi (50 mM) and lysed with 250 ml aliquots with one followed in real time using 0.5 3 Sybr Green I (Life Technologies); of lysis buffer L7 of the PureLink Quick Plasmid DNA Miniprep Kit (200 mM following PCR, the aliquots were recombined, purified by spin column (DNA Clean NaOH, 1% w/v SDS), after which the resulting solution was incubated at 22 uC for and Concentrator-5; Zymo Research, Irvine, California, USA) with elution in 20 ml 5 min. Precipitation buffer N4 (350 ml, 3.1 M potassium acetate, pH 5.5) was added, of water, then separated by agarose gel electrophoresis, followed by band excision and the sample was mixed to homogeneity. Following centrifugation (.12,000 r.c.f.
and recovery (Zymoclean Gel DNA Recovery Kit), eluting with 20 ml of water unless for 10 min, at 22 uC) the supernatant containing the unnatural nucleotides was stated otherwise. Polyacrylamide gels were stained with 1 3 Sybr Gold (Life Tech- applied to a Hypersep C18 solid phase extraction column (Thermo Scientific) pre- nologies) for 30 min, agarose gels were cast with 1 3 Sybr Gold. All gels were visu- washed with acetonitrile (1 ml) and buffer A (1 ml, see HPLC protocol for buffer alized using a Molecular Imager Gel Doc XR1 equipped with 520DF30 filter (Bio-Rad) composition). The column was then washed with buffer A and nucleotides were and quantified with Quantity One software (Bio-Rad). The sequences of all DNA eluted with 1 ml of 50% acetonitrile:50% triethylammonium bicarbonate (TEAB) oligonucleotides used in this study are provided in Supplementary Information.
0.1 M (pH 7.5). The eluent was reduced to approximately 50 ml in a SpeedVac and Natural oligonucleotides were purchased from IDT (San Diego, California, USA).
its volume was adjusted to 100 ml with buffer A before HPLC analysis.
The concentration of dsDNA was measured by fluorescent dye binding (Quant-iT HPLC protocol and nucleoside triphosphate quantification. Samples were applied dsDNA HS Assay kit, Life Technologies) unless stated otherwise. The concentra- to a Phenomenex Jupiter LC column (3 mm C18 300 A ˚ , 250 3 4.6 mm) and subjected tion of ssDNA was determined by UV absorption at 260 nm using a NanoDrop to a linear gradient of 0–40% B over 40 min at a flow rate of 1 ml min21. Buffer A: 1000 (Thermo Scientific). [a-32P]-dATP (25 mCi) was purchased from PerkinElmer 95% 0.1 M TEAB, pH 7.5; 5% acetonitrile. Buffer B: 20% 0.1 M TEAB, pH 7.5; 80% (Shelton, Connecticut, USA). Polyethyleneimine cellulose pre-coated Bakerflex TLC acetonitrile. Absorption was monitored at 230, 273, 288, 326 and 365 nm.
plates (0.5 mm) were purchased from VWR. dNaM phosphoramidite, dNaM and Each injection series included two extra control samples containing 5 nmol of d5SICS nucleosides were obtained from Berry & Associates Inc. (Dexter, Michigan, dNaMTP or d5SICSTP. The areas under the peaks that corresponded to tripho- USA). Free nucleosides of dNaM and d5SICS (Berry & Associates) were converted to sphate, diphosphate, monophosphate and free nucleoside (confirmed by MALDI- the corresponding triphosphates under Ludwig conditions30. After purification by anion TOF) were integrated for both the control and the unknown samples (described exchange chromatography (DEAE Sephadex A-25) followed by reverse phase (C18) above). After peak integration, the ratio of the unknown peak to the control peak HPLC and elution through a Dowex 50WX2-sodium column, both triphosphates adjusted for the loss from the extraction step (62% and 70% loss for dNaM and were lyophilized and kept at 220 uC until use. The d5SICSTP analogue dTPT3TP22 d5SICS, respectively, Extended Data Table 3), provided a measure of the amount and the biotinylated dNaMTP analogue dmmo2SSBIOTP4 were made as reported of each of the moieties in the sample. To determine the relative concentrations of previously. MALDI-TOF mass spectrometry (Applied Biosystems Voyager DE-PRO unnatural nucleotide inside the cell, the amount of imported unnatural nucleotide System 6008) was performed at the TSRI Center for Protein and Nucleic Acid Research.
(dXTP, mmol) was then divided by the volume of cells, which was calculated as the Construction of NTT expression plasmids. The PtNTT2 gene was amplified from product of the volume of a single E. coli cell (1 mm3 based on a reported average value32; plasmid pET-16b-PtNTT2 using primers PtNTT2-forward and PtNTT2-reverse; that is, 1 3 1029 ml per cell) and the number of cells in each culture (OD600 of 1.0 the TpNTT2 gene was amplified from plasmid pET-16b-TpNTT2 using primers equal to 1 3 109 cells per ml (ref. 32)). The RpNTT2 sample was used as a negative TpNTT2-forward and TpNTT2-reverse. A linear fragment of pCDF-1b was gener- control and its signal was subtracted to account for incomplete washing of nucle- ated using primers pCDF-1b-forward and pCDF-1b-reverse. All fragments were otide species from the media.
purified as described in Materials. The pCDF-1b fragment (100 ng, 4.4 3 10214 mol) dATP uptake. To analyse the intracellular desphosphorylation of dATP, after induc- and either the PtNTT2 (78 ng, 4.4 3 10214 mol) or TpNTT2 (85 ng, 4.4 3 10214 mol) tion of the transporter, the uptake reaction was initiated by the addition of dATP fragment were then assembled together using restriction-free circular polymerase (spiked with [a-32P]-dATP) to a final concentration of 0.25 mM, followed by incu- extension cloning29 in 1 3 OneTaq reaction buffer, MgSO bation at 37 uC with shaking for 30 min. The culture was then centrifuged (8,000 4 adjusted to 3.0 mM, 0.2 mM of dNTP, and 0.02 U ml21 of OneTaq DNA under the following thermal r.c.f. for 5 min at 22 uC). Supernatant was analysed by TLC. Cell pellets were washed cycling conditions: initial denaturation (96 uC, 1 min); 10 cycles of denaturation three times with ice-cold KPi (50 mM, 100 ml) to remove excess radioactive sub- (96 uC, 30 s), touchdown annealing (54 uC to 49.5 uC for 30 s (20.5 uC per cycle)), strate, lysed with NaOH (0.2 M, 100 ml) and centrifuged (10,000 r.c.f. for 5 min at extension of 68 uC for 5 min, and final extension (68 uC, 5 min). Upon completion, 22 uC) to remove cell debris; supernatant was analysed by TLC.
the samples were purified and used for heat-shock transformation of E. coli XL10.
TLC analysis. Samples (1 ml) were applied on a 0.5 mm polyethyleneimine cellu- Individual colonies were selected on lysogeny broth (LB)-agar containing strep- lose TLC plate and developed with sodium formate pH 3.0 (0.5 M, 30 s; 2.5 M, 2.5 min; tomycin, and assayed by colony PCR with primers PtNTT2-forward/reverse or 4.0 M, 40 min). Plates were dried using a heat gun and quantified by phosphorimaging TpNTT2-forward/reverse. The presence of the NTT genes was confirmed by sequenc- (Storm Imager, Molecular Dynamics) and Quantity One software.
ing and double digestion with ApaI/EcoO109I restriction endonucleases with the Optimization of nucleotide extraction from cells for HPLC injection. To min- following expected pattern: pCDF-1b-PtNTT2 (2,546/2,605 bp), pCDF-1b-TpNTT2 imize the effect of the lysis and triphosphate extraction protocols on the decom- (2,717/2,605 bp), pCDF-1b (1,016/2,605 bp). The complete nucleotide sequence position of nucleoside triphosphate within the cell, the extraction procedure was of the pCDF-1b-PtNTT2 plasmid (pACS) is provided in Supplementary Information.
optimized for the highest recovery with the lowest extent of decomposition (Extended Growth conditions to quantify nucleoside triphosphate uptake. E. coli C41(DE3)16 Data Table 3). To test different extraction methods, cells were grown as described freshly transformed with pCDF-1b-PtNTT2 was grown in 2 3 YT with streptomy- above, washed, and then 5 nmol of either dNaMTP or d5SICSTP was added to the cin overnight, then diluted (1:100) into fresh 2 3 YT medium (1 ml of culture per pellets, which were then subjected to different extraction protocols including boil- uptake with [a-32P]-dATP; 2 ml of culture per uptake with d5SICSTP or dNaMTP) ing water, hot ethanol, cold methanol, freeze and thaw, lysozyme, glass beads, NaOH, supplemented with 50 mM potassium phosphate (KPi) and streptomycin. A nega- trichloroacetic acid (TCA) with Freon, and perchloric acid (PCA) with KOH33.
tive control with the inactive transporter pET-16b-RpNTT2, was treated identically The recovery and composition of the control was quantified by HPLC as described except ampicillin was used instead of streptomycin. Cells were grown to an OD600 above to determine the most effective procedure. Method 3—that is, cell lysis with of approximately 0.6 and the NTT expression was induced by the addition of IPTG NaOH (Extended Data Table 3)—was found to be most effective and reprodu- (1 mM). The culture was allowed to grow for another hour (final OD600 approxi- cible, thus we further optimized it by resuspension of the pellets in ice-cold KPi mately 1.2) and then assayed directly for uptake as described below using a method (50 mM, 250 ml) before addition of NaOH to decrease dephosphorylation after cell adapted from a previous paper15.
lysis (Method 4). Cell pellets were then processed as described above. See above for Preparation of media fraction for unnatural nucleoside triphosphate analysis.
the final extraction protocol.
The experiment was initiated by the addition of either dNaMTP or d5SICSTP Preparation of the unnatural insert for pINF construction. The TK-1-dNaM (10 mM each) directly to the media to a final concentration of 0.25 mM. Cells were oligonucleotide containing dNaM was prepared using solid-phase DNA synthesis 2014 Macmillan Publishers Limited. All rights reserved
with ultra-mild DNA synthesis phosphoramidites on CPG ultramild supports (10 ml) was spread onto 2 3 YT-agar containing streptomycin with 10- and 50-fold (1 mmol, Glen Research, Sterling, Virginia, USA) and an ABI Expedite 8905 syn- dilutions for the determination of viable colony forming units after overnight growth thesizer. After the synthesis, the DMT-ON oligonucleotide was cleaved from the at 37 uC to calculate the number of the transformed pINF molecules (see the section solid support, deprotected and purified by Glen-Pak cartridge according to the on calculation of the plasmid amplification). Transformation, recovery and growth manufacturer's recommendation (Glen Research), and then subjected to 8 M urea were carried out identically for the natural control plasmid. In addition, a negative 8% PAGE. The gel was visualized by ultraviolet shadowing, the band corresponding control was run and treated identically to pINF transformation except that it was to the 75-mer was excised, and the DNA was recovered by crush and soak extraction, not subjected to electroporation (Extended Data Fig. 7b). No growth in the untrans- filtration (0.45 mm), and final desalting over Sephadex G-25 (NAP-25 Columns, GE formed negative control samples was observed even after 6 days. No PCR amp- Healthcare). The concentration of the single stranded oligonucleotide was deter- lification of the negative control was detected, which confirms that unamplified mined by ultraviolet absorption at 260 nm assuming that the extinction coefficient of pINF plasmid is not carried through cell growth and later detected erroneously as dNaM at 260 nm is equal to that of dA. TK-1-dNaM (4 ng) was next amplified by the propagated plasmid.
PCR under the following conditions: 1 3 OneTaq reaction buffer, MgSO4 adjusted Analysis of pINF replication in E. coli. After recovery, the cells were centrifuged to 3.0 mM, 0.2 mM of dNTP, 0.1 mM of dNaMTP, 0.1 mM of the d5SICSTP ana- (4,000 r.c.f. for 5 min, 4 uC), and spent media (0.15 ml) was removed and analysed logue dTPT3TP, 1 mM of each of the primers pUC19-fusion-forward and pUC19- for nucleotide composition by HPLC (Extended Data Fig. 7a). The cells were resus- fusion-reverse, and 0.02 U ml21 of OneTaq DNA Polymerase (in a total of 4 3 50 ml pended in fresh 2 3 YT media (1.5 ml, streptomycin, ampicillin, 1 mM IPTG, 50 mM reactions) under the following thermal cycling conditions: initial denaturation (96 uC, KPi, 0.25 mM dNaMTP, 0.25 mM d5SICSTP) and grown overnight at 37 uC while 1 min) followed by 12 cycles of denaturation (96 uC, 10 s), annealing (60 uC, 15 s), shaking (250 r.p.m.), resulting in tenfold dilution compared to recovery media or and extension (68 uC, 2 min). An identical PCR without the unnatural triphosphates 33.3-fold dilution compared to the originally transformed cells. Aliquots (100 ml) was run to obtain fully natural insert under identical conditions for the construction were taken after 15, 19, 24, 32, 43, 53, 77 and 146 h, OD600 was determined, and the of the natural control plasmid. Reactions were subjected to spin column purification cells were centrifuged (8,000 r.c.f. for 5 min, 4 uC). Spent media were analysed for and then the desired PCR product (122 bp) was purified by a 4% agarose gel.
nucleotide composition by HPLC (Extended Data Fig. 7a), and the pINF and pACS pUC19 linearization for pINF construction. pUC19 (20 ng) was amplified by plasmid mixtures were recovered and linearized with NdeI restriction endonuclease; PCR under the following conditions: 1 3 Q5 reaction buffer, MgSO4 adjusted to pINF plasmid was purified by 1% agarose gel electrophoresis (Extended Data Fig. 7b) 3.0 mM, 0.2 mM of dNTP, 1 mM of each primers pUC19-lin-forward and pUC19- and analysed by LC-MS/MS. The retention of the UBP on the pINF plasmid was lin-reverse, and 0.02 U ml21 of Q5 Hot Start High-Fidelity DNA Polymerase (in a quantified by biotin gel shift mobility assay and sequencing as described below.
total of 4 3 50 ml reactions with one reaction containing 0.5 3 Sybr Green I) under Mass spectrometry of pINF. Linearized pINF was digested to nucleosides by treat- the following thermal cycling conditions: initial denaturation (98 uC, 30 s); 20 cycles ment with a mixture of nuclease P1 (Sigma-Aldrich), shrimp alkaline phosphatase of denaturation (98 uC, 10 s), annealing (60 uC, 15 s), and extension (72 uC, 2 min); (NEB), and DNase I (NEB), overnight at 37 uC, following a previously reported and final extension (72 uC, 5 min). The desired PCR product (2,611 bp) was purified protocol25. LC-MS/MS analysis was performed in duplicate by injecting 15 ng of by a 2% agarose gel.
digested DNA on an Agilent 1290 UHPLC equipped with a G4212A diode array PCR assembly of pINF and the natural control plasmid. A linear fragment was detector and a 6490A Triple Quadrupole Mass Detector operating in the positive amplified from pUC19 using primers pUC19-lin-forward and pUC19-lin-reverse.
electrospray ionization mode (1ESI). UHPLC was carried out using a Waters XSelect The resulting product (800 ng, 4.6 3 10213 mol) was combined with either the nat- HSS T3 XP column (2.1 3 100 mm, 2.5 mm) with the gradient mobile phase con- ural or unnatural insert (see above) (56 ng, 7.0 3 10213 mol) and assembled by cir- sisting of methanol and 10 mM aqueous ammonium formate (pH 4.4). MS data cular overlap extension PCR under the following conditions: 1 3 OneTaq reaction acquisition was performed in Dynamic Multiple Reaction Monitoring (DMRM) buffer, MgSO4 adjusted to 3.0 mM, 0.2 mM of dNTP, 0.1 mM of dNaMTP, 0.1 mM mode. Each nucleoside was identified in the extracted chromatogram associated of the d5SICSTP analogue dTPT3TP, and 0.02 U ml21 of OneTaq DNA Polymer- with its specific MS/MS transition: dA at m/z 252R136, d5SICS at m/z 292R176, ase (in a total of 4 3 50 ml reactions with one reaction containing 0.5 3 Sybr Green I) and dNaM at m/z 275R171. External calibration curves with known amounts of using the following thermal cycling conditions: initial denaturation (96 uC, 1 min); the natural and unnatural nucleosides were used to calculate the ratios of indivi- 12 cycles of denaturation (96 uC, 30 s), annealing (62 uC, 1 min), and extension (68 uC, dual nucleosides within the samples analysed. LC-MS/MS quantification was vali- 5 min); final extension (68 uC, 5 min); and slow cooling (68 uC to 10 uC at a rate of dated using synthetic oligonucleotides1 containing unnatural d5SICS and dNaM 20.1 uC s21). The PCR product was analysed by restriction digestion on 1% aga- (Extended Data Table 2).
rose and used directly for E. coli transformation. The d5SICS analogue dTPT322 DNA biotinylation by PCR to measure fidelity by gel shift mobility assay.
pairs with dNaM, and dTPT3TP was used in place of d5SICSTP as DNA contain- Purified mixtures of pINF and pACS plasmids (1 ng) from growth experiments ing dTPT3–dNaM is better PCR amplified than DNA containing d5SICS–dNaM, were amplified by PCR under the following conditions: 1 3 OneTaq reaction buf- and this allowed for differentiation of synthetic and in vivo replicated pINF, as well fer, MgSO4 adjusted to 3.0 mM, 0.3 mM of dNTP, 0.1 mM of the biotinylated as facilitated the construction of high-quality pINF (UBP content .99%).
dNaMTP analogue dMMO2SSBIOTP, 0.1 mM of d5SICSTP, 1 mM of each of the Preparation of electrocompetent cells for pINF replication in E. coli. C41(DE3) primers pUC19-seq-forward and pUC19-seq-reverse, 0.02 U ml21 of OneTaq DNA cells were transformed by heat shock34 with 200 ng of pACS plasmid, and the trans- Polymerase, and 0.0025 U ml21 of DeepVent DNA Polymerase in a total volume of formants were selected overnight on 2 3 YT-agar supplemented with streptomy- 25 ml in an CFX Connect Real-Time PCR Detection System (Bio-Rad) under the cin. A single clone of freshly transformed C41(DE3) pACS was grown overnight in following thermal cycling conditions: initial denaturation (96 uC, 1 min); 10 cycles 2 3 YT medium (3 ml) supplemented with streptomycin and KPi (50 mM). After of denaturation (96 uC, 30 s), annealing (64 uC, 30 s), and extension (68 uC, 4 min).
100-fold dilution into the same fresh 2 3 YT media (300 ml), the cells were grown PCR products were purified, and the resulting biotinylated DNA duplexes (5 ml, at 37 uC until they reached an OD600 of 0.20 at which time IPTG was added to a 25–50 ng) were mixed with streptavidin (1 ml, 1 mg ml21, Promega) in phosphate final concentration of 1 mM to induce the expression of PtNTT2. Cells were grown buffer (50 mM sodium phosphate, pH 7.5, 150 mM NaCl, 1 mM EDTA), incubated for another 40 min and then growth was stopped by rapid cooling in ice water with for 30 min at 37 uC, mixed with 5 3 non-denaturing loading buffer (Qiagen), and intensive shaking. After centrifugation in a prechilled centrifuge (2,400 r.c.f. for 10 min, loaded onto 6% non-denaturing PAGE. After running at 110 V for 30 min, the gel 4 uC), the spent media was removed, and the cells were prepared for electropora- was visualized and quantified. The resulting fragment (194 bp) with primer regions tion by washing with ice-cold sterile water (3 3 150 ml). After washing, the cells underlined and the unnatural nucleotide in bold (X represents dNaM or its bioti- were resuspended in ice-cold 10% glycerol (1.5 ml) and split into 50-ml aliquots.
nylated analogue dMMO2SSBIO) is 59-GCAGGCATGCAAGCTTGGCGTAATC Although we found that dry ice yielded better results than liquid nitrogen for freezing cells to store for later use, freshly prepared cells were used for all reported experi- ments as they provided higher transformation efficiency of pINF and higher rep- lication fidelity of the UBP.
Electroporation and recovery for pINF replication in E. coli. The aliquot of cells Streptavidin shift calibration for gel shift mobility assay. We have already reported was mixed with 2 ml of plasmid (400 ng), transferred to 0.2 cm gap electroporation a calibration between streptavidin shift and the fraction of sequences with UBP in cuvette and electroporated using a Bio-Rad Gene Pulser according to the manu- the population (see Supplementary Fig. 8 of ref. 1). However, we found that spiking facturer's recommendations (voltage 25 kV, capacitor 2.5 mF, resistor 200 V, time the PCR reaction with DeepVent improves the fidelity with which DNA contain- constant 4.8 ms). Pre-warmed 2 3 YT media (0.95 ml, streptomycin, 1 mM IPTG, ing d5SICS-dMMO2SSBIO is amplified, and thus we repeated the calibration with 50 mM KPi) was added, and after mixing, 45 ml was removed and combined with added DeepVent. To quantify the net retention of the UBP, nine defined mixtures 105 ml of the same media (3.33-fold dilution) supplemented with 0.25 mM of dNaMTP of the TK-1-dNaM template and its fully natural counterpart were prepared (Extended and d5SICSTP. The resulting mixture was allowed to recover for 1 h at 37 uC with Data Fig. 6a), subjected to biotinylation by PCR and analysed by mobility-shift assay shaking (210 revolutions per min (r.p.m.)). The original transformation media on 6% non-denaturing PAGE as described above. For calibration, the mixtures 2014 Macmillan Publishers Limited. All rights reserved
TK-1-dNaM template and its fully natural counterpart with a known ratio of USA). Products were eluted off the beads with deionized water and sequenced unnatural and natural templates (0.04 ng) were amplified under the same condi- directly on a 3730 DNA Analyzer (Applied Biosystems). Sequencing traces were tions over nine cycles of PCR with pUC19-fusion primers and analysed identically collected using Applied Biosystems Data Collection software v3.0 and analysed to samples from the growth experiment (see the section on DNA biotinylation by with the Applied Biosystems Sequencing Analysis v5.2 software.
PCR). Each experiment was run in triplicate (a representative gel assay is shown in Analysis of Sanger sequencing traces. Sanger sequencing traces were analysed as Extended Data Fig. 6b), and the streptavidin shift (SAS, %) was plotted as function described previously1,26 to determine the retention of the unnatural base pair. In brief, of the UBP content (UBP, %). The data was then fit to a linear equation, SAS 5 the presence of an unnatural nucleotide leads to a sharp termination of the sequencing 0.77 3 UBP 1 2.0 (R2 5 0.999), where UBP corresponds to the retention of the profile, whereas mutation to a natural nucleotide results in ‘read-through'. The extent UBP (%) in the analysed samples after cellular replication and was calculated from of this read-through after normalization is inversely correlated with the retention the SAS shift using the equation above.
of the unnatural base pair. Raw sequencing traces were analysed by first adjusting Calculation of plasmid amplification. The cells were plated on 2 3 YT-agar the start and stop points for the Sequencing Analysis software (Applied Biosystems) containing ampicillin and streptavidin directly after transformation with pINF, and then determining the average signal intensity individually for each channel (A, and the colonies were counted after overnight growth at 37 uC. Assuming each cell C, G and T) for peaks within the defined points. This was done separately for the is only transformed with one molecule of plasmid, colony counts correspond to the parts of the sequencing trace before (section L) and after (section R) the unnatural original amount of plasmid that was taken up by the cells. After overnight growth, nucleotide. The R/L ratio after normalization (R/L)norm for sequencing decay and the plasmids were purified from a specific volume of the cell culture and quantified.
read-through in the control unamplified sample (R/L 5 0.55(R/L)norm 1 7.2, see As purified plasmid DNA represents a mixture of the pINF and pACS plasmids, ref. 26 for details) corresponds to the percentage of the natural sequences in the digestion restriction analysis with NdeI exonuclease was performed to linearize pool. Therefore, an overall retention (F) of the incorporation of the unnatural base both plasmids, followed by 1% agarose gel electrophoresis (Extended Data Fig. 7b).
pair during PCR is equal to 1 – (R/L)norm. As significant read-through (over 20%) An example of calculations for the 19-h time point with one of three triplicates is was observed in the direction of the pUC19-seq2-forward primer even with the provided in Supplementary Information.
control plasmid (synthetic pINF); sequencing of only the opposite direction (pUC19- Fragment generation for Sanger sequencing to measure fidelity. Purified mix- seq-reverse) was used to gauge fidelity. Raw sequencing traces are shown in Fig. 2f tures of pINF and pACS plasmids (1 ng) after the overnight growth were amplified and provided as Supplementary Data.
by PCR under the following conditions: 1 3 OneTaq reaction buffer, MgSO4 adjustedto 3.0 mM, 0.2 mM of dNTP, 0.1 mM of dNaMTP, 0.1 mM of the d5SICSTP ana- Ludwig, J. & Eckstein, F. Rapid and efficient synthesis of nucleoside5 logue dTPT3TP, 1 mM of each of the primers pUC19-seq2-forward and pUC19- using 2-chloro-4H-1,3,2-benzodioxaphosphorin-4-one. J. Org. Chem. 54, seq-reverse (see below), and 0.02 U ml21 of OneTaq DNA Polymerase in a total 631–635 (1989).
volume of 25 ml under the following thermal cycling conditions: initial denatura- Alpert, A. & Shukla, A. Precipitation of Large, High-Abundance Proteins from tion (96 uC, 1 min); and 10 cycles of denaturation (96 uC, 30 s), annealing (64 uC, 30 s), Serum with Organic Solvents in ABRF 2003: Translating Biology using Proteomics and extension (68 uC, 2 min). Products were purified by spin column, quantified to and Functional Genomics Poster no. P111-W measure DNA concentration and then sequenced as described below. The sequenced Kubitschek, H. E. & Friske, J. A. Determination of bacterial cell volume with the fragment (304 bp) with primer regions underlined and the unnatural nucleotide in Coulter Counter. J. Bacteriol. 168, 1466–1467 (1986).
bold (X, dNaM) is 59-GCTGCAAGGCGATTAAGTTGGGTAACGCC AGGGT Yanes, O., Tautenhahn, R., Patti, G. J. & Siuzdak, G. Expanding coverage of the metabolome for global metabolite profiling. Anal. Chem. 83, 2152–2161 (2011).
Seidman, C. E., Struhl, K., Sheen, J. & Jessen, T. Introduction of plasmid DNA into cells. Curr. Prot. Mol. Biol. Chapter 1, Unit 1.8 (2001).
Knab, S., Mushak, T. M., Schmitz-Esser, S., Horn, M. & Haferkamp, I. Nucleotide parasitism by Simkania negevensis (Chlamydiae). J. Bacteriol. 193, 225–235(2011).
Audia, J. P. & Winkler, H. H. Study of the five Rickettsia prowazekii proteins Sanger sequencing. The cycle sequencing reactions (10 ml) were performed on a annotated as ATP/ADP translocases (Tlc): Only Tlc1 transports ATP/ADP, while 9800 Fast Thermal Cycler (Applied Biosystems) with the Cycle Sequencing Mix (0.5 ml) Tlc4 and Tlc5 transport other ribonucleotides. J. Bacteriol. 188, 6261–6268 of the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems) con- taining 1 ng template and 6 pmol of sequencing primer pUC19-seq-reverse under Hofer, A., Ekanem, J. T. & Thelander, L. Allosteric regulation of Trypanosoma bruceiribonucleotide reductase studied in vitro and in vivo. J. Biol. Chem. 273, the following thermal cycling conditions: initial denaturation (98 uC, 1 min); and 25 cycles of denaturation (96 uC, 10 s), annealing (60 uC, 15 s), and extension (68 uC, Reijenga, J. C., Wes, J. H. & van Dongen, C. A. M. Comparison of methanol 2.5 min). Upon completion, the residual dye terminators were removed from the and perchloric acid extraction procedures for analysis of nucleotides reaction with Agencourt CleanSEQ (Beckman-Coulter, Danvers, Massachusetts, by isotachophoresis. J. Chromatogr. B Biomed. Sci. Appl. 374, 162–169 (1986).
2014 Macmillan Publishers Limited. All rights reserved
Extended Data Figure 1 Natural triphosphate uptake by NTTs. a, Survey of a measurable uptake of dATP although this activity was not reported before.
reported substrate specificity (KM, mM) of the NTTs assayed in this study.
This can possibly be explained by the fact that substrate specificity was only b, PtNTT2 is significantly more active in the uptake of [a-32P]-dATP compared characterized using competition experiments, and assay sensitivity might to other nucleotide transporters. Raw (left) and processed (right) data are not have been adequate to detect this activity15. References 35, 36 are cited in shown. Relative radioactivity corresponds to the total number of counts this figure.
produced by each sample. Interestingly, both PamNTT2 and PamNTT5 exhibit 2014 Macmillan Publishers Limited. All rights reserved
Extended Data Figure 2 Degradation of unnatural triphosphates in down the dephosphorylation of both unnatural triphosphates.
growth media. Unnatural triphosphates (3P) of dNaM and d5SICS are a, Representative HPLC traces (for the region between ,20 and 24 min).
degraded to diphosphates (2P), monophosphates (1P) and nucleosides (0P) in dNaM and d5SICS nucleosides are eluted at approximately 40 min and not the growing bacterial culture. Potassium phosphate (KPi) significantly slows shown. b, Composition profiles.
2014 Macmillan Publishers Limited. All rights reserved
Extended Data Figure 3 Effect of potassium phosphate on dATP uptakeand stability in growth media. a, KPi inhibits the uptake of [a-32P]-dATP atconcentrations above 100 mM. Raw (left) and processed (right) data are shown.
The NTT from Rickettsia prowazekii (RpNTT2) does not mediate the uptake ofany of the dNTPs and was used as a negative control: its background signalwas subtracted from those of PtNTT2 (black bars) and TpNTT2 (white bars).
Relative radioactivity corresponds to the total number of counts produced byeach sample. b, KPi (50 mM) significantly stabilizes [a-32P]-dATP in themedia. Triphosphate stability in the media is not significantly affectedby the nature of the NTT expressed. 3P, 2P and 1P correspond to triphosphate,diphosphate and monophosphate states, respectively. Error bars represents.d. of the mean, n 5 3.
2014 Macmillan Publishers Limited. All rights reserved
Extended Data Figure 4 dATP uptake and growth of cells expressing total number of counts produced by each sample. b, A stationary phase culture PtNTT2 as a function of inducer (IPTG) concentration. Growth curves and of C41(DE3) pACS cells was diluted 100-fold into fresh 2 3 YT media [a-32P]-dATP uptake by bacterial cells transformed with pCDF-1b-PtNTT2 containing 50 mM KPi, streptomycin, and IPTG at the indicated (pACS) plasmid as a function of IPTG concentration. a, Total uptake of concentrations and were grown at 37uC. Error bars represent s.d. of the radioactive substrate (left) and total intracellular triphosphate content (right) mean, n 5 3.
are shown at two different time points. Relative radioactivity corresponds to the 2014 Macmillan Publishers Limited. All rights reserved
Extended Data Figure 5 Stability and uptake of dATP in the presence of respectively. 3P, 2P and 1P correspond to nucleoside triphosphate, diphosphate 50 mM KPi and 1 mM IPTG. Composition of [a-32P]-dATP in the media and monophosphate, respectively. M refers to a mixture of all three compounds (left) and cytoplasmic fraction (right) as a function of time. TLC images and that was used as a TLC standard. The position labelled ‘Start' corresponds to their quantifications are shown at the bottom and the top of each of the panels, the position of sample spotting on the TLC plate.
2014 Macmillan Publishers Limited. All rights reserved
Extended Data Figure 6 Calibration of the streptavidin shift (SAS). a, TheSAS is plotted as a function of the fraction of template containing the UBP.
Error bars represent s.d. of the mean, n 5 3. b, Representative data.
SA, streptavidin.
2014 Macmillan Publishers Limited. All rights reserved
Extended Data Figure 7 Decomposition of unnatural triphosphates, pINF data are shown in three lanes with the untransformed control shown in the quantification, and retention of the UBP with extended cell growth.
fourth, rightmost lane (see Methods). c, Number of pINF doublings as a a, Dephosphorylation of the unnatural nucleoside triphosphate. 3P, 2P, 1P and function of time. The decrease starting at approximately 50 h is due to the loss 0P correspond to triphosphate, diphosphate, monophosphate and nucleoside of the pINF plasmid that also results in increased error. See the section on pINF states, respectively. The composition at the end of the 1 h recovery is shown at replication in E. coli in the Methods for details. d, UBP retention (%) as a the right. b, Restriction analysis of pINF and pACS plasmids purified from function of growth as determined by gel shift (data shown in Fig. 3) and Sanger E. coli, linearized with NdeI restriction endonuclease and separated on a 1% sequencing (sequencing traces are available as Supplementary Data). In a, c agarose gel (assembled from independent gel images). Molar ratios of pINF/ and d, error shown is the s.d. of mean, n 5 3.
pACS plasmids are shown at the top of each lane. For each time point, triplicate 2014 Macmillan Publishers Limited. All rights reserved
Extended Data Table 1 OD600 of E. coli cultures and relative copy number of plasmid (pINF or control pUC19) as determined by itsmolar ratio to pACS after 19 h of growth X, NaM; Y, 5SICS.
2014 Macmillan Publishers Limited. All rights reserved
Extended Data Table 2 Relative quantification by LC-MS/MS using synthetic oligonucleotides containing d5SICS and dNaM * dA/d5SICS and dA/dNaM ratios were calculated assuming that randomized nucleotides (N) around the unnatural base are distributed equally.
2014 Macmillan Publishers Limited. All rights reserved
Extended Data Table 3 Summary of the most successful extraction methods * Recovery of all nucleotides (3P, 2P, 1P and nucleoside).
{ Calculated as a ratio of 3P composition (%) before and after the extraction.
References 37, 38 are cited in this figure. Details of methods 3 and 4 can be found online 2014 Macmillan Publishers Limited. All rights reserved
Source: http://apinano.org.br/documentos/A-semi-synthetic-organism-with-an-expanded.pdf
Principles and processes in biotechnology.pmd
Ever since the days of Rene Descartes, the French philosopher,mathematician and biologist of seventeenth century, all human Chapter 11 knowledge especially natural sciences were directed to develop Biotechnology : Principles and technologies which add to the creature comforts of human lives, as also value to human life. The whole approach to Chapter 12
Microsoft word - primary research-1.docx
Primary research Establishment and characterization of two primary breast cancer cell lines from young Indian breast cancer patients: mutation analysis Santhi Latha, Chintamani , and Sunita Saxena Author Affiliations For all author emails, please log on. Cancer Cell International 2014, 14:14 doi:10.1186/1475-2867-14-14