PEDS Advance Access published online on January 5, 2008
Protein Engineering Design and Selection, doi:10.1093/protein/gzm085
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Short Communication |
Engineering Escherichia coli heat-resistance by synthetic gene amplification
1MRC Laboratory of Molecular Biology, Hills Road, Cambridge CB2 2QH, UK
3 To whom correspondence should be addressed. E-mail: chin{at}mrc-lmb.cam.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Organisms have evolved to exploit new environments by processes that involve both mutations and gene amplifications. Though in some cases amplified genes mutate to perform a different molecular function, in other cases altering gene copy number alone is sufficient to change organism function. Here we selected a library of genes, provided at high copy number, for their ability to confer survival on Escherichia coli cells at un-physiologically high temperatures. We find that a single gene (evgA), encoding a master transcriptional regulator, is overwhelmingly selected and allows survival upon heating to temperatures in excess of 50°C. While the detailed mechanisms of this resistance remained unclear, our results demonstrate the potential of copy number manipulation for the engineering of organisms
Keywords: copy number variation/synthetic biology/heat-resistance
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Gene amplification is the process by which a DNA segment is iterated to provide multiple copies of the DNA segment per cell. Though gene amplification was first observed in Escherichia coli over 40 years ago (Horiuchi et al., 1962
, 1963), the full extent of gene amplification, and the resulting copy number variation, in determining phenotypes across organisms is only just beginning to be appreciated. Gene amplification can contribute to tumorigenesis (Pollack et al., 1999), antibiotic resistance in microbes (Nilsson et al., 2006) and is causal to a range of neurological conditions (Niebuhr, 1978; Ledbetter et al., 1981; Hassold and Jacobs, 1984). Gene amplification can underlie adaptive evolution (Hastings et al., 2004; Kugelberg et al., 2006; Roth et al., 2006; Slack et al., 2006) and contribute to new gene function and genome evolution (Luscombe et al., 2004; Teichmann and Babu, 2004). Moreover, gene amplification contributes to the considerable gene copy number variation within the human population (Iafrate et al., 2004, Sebat et al., 2004) that may affect the variability of our phenotypic responses to a range of diseases, therapeutic agents and environmental cues (Gonzalez et al., 2005). Our emerging understanding of the role of copy number variation in determining organism function suggests that synthetic alteration of gene copy number, coupled with genetic selection may provide access to organisms with improved survival under extreme conditions.
Although gene copy number effects have been studied ever since plasmid-based genomic libraries became available in the mid 1970s (Clarke and Carbon, 1975), most studies have focussed on either resistance to antibiotic and cytotoxic agents (Chen and Bishai, 1998; Cotrim et al., 1999; Burger et al., 2000) or on limited variations of physiological conditions. For instance, a large body of previous work has explored cellular mechanisms for dealing with heat shock at elevated but non-lethal temperatures (Yura et al., 1993). In response to increase the temperature of E.coli growth from 37 to 42°C (a stressful but permissive temperature), 
H and E dependent transcription is activated (Yura et al., 1993; Lund, 2001). The molecular mechanisms by which many of the proteins encoded by heat inducible transcripts, including chaperones and operate have been extensively investigated (Arsene et al., 2000). In contrast, there has been a little work exploring the factors that might allow cells to survive at even higher temperatures, well beyond their normal physiological limits. Yeh et al. (1997) over-expressed a heterologous heat shock protein in E.coli and reported a modest effect on the thermal resistance (101- to 102-fold improvement in survival after 1 h at 47.5°C). Recently, Park et al. (2005) focussed on creating thermally resistant E.coli using a library of engineered zinc fingers aimed at repressing gene expression. A library of cells expressing zinc fingers, with a range of DNA binding specificities, was incubated for 2 h at 50°C, and zinc fingers able to confer a 102-fold increase in thermal resistance, after 2 h at 50°C, isolated. Lenski and co-workers have investigated the ability of E.coli strains propagated for many generations at 41.2°C to survive at 50°C, after pre-incubation at 41.2°C—which may allow for epigenetic change—they find that cells can survive up to 102-fold better after incubation for 4 h at 50°C (Riehle et al., 2003).
Here, we ask whether amplification of any gene, from a library of E.coli genes provided extra chromosomally at high copy number, can facilitate survival at high and otherwise lethal temperatures. We find that a single gene (evgA), encoding a master transcriptional regulator is overwhelmingly selected and allows the mesophilic gut bacterium E.coli to survive heating to temperatures in excess of 50°C. Our results demonstrate the potential of copy number manipulation for the engineering of organisms.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Construction of an E.coli gene library
We fragmented genomic DNA of E.coli W3110 in a Nebulizer DNA fragmentation device (Invitrogen), predominantly to about 300–1200 bp. We ligated oligonucleotide adapters to the fragments and ligated the resulting inserts into plasmid pW656 (as previously described, Christ and Winter (2006)). Plasmid pW656 is derived from pHEN1 (Hoogenboom et al., 1991) and carries an ampicillin resistance marker as well as the pUC origin of replication (300–500 copies per cell). After transformation into E.coli TG1 (Gibson, 1984), we obtained 5 x 106 transformants, from which we obtained plasmid DNA using a DNA miniprep kit (Qiagen). For selection, we transformed the plasmid DNA into E.coli TG1 yielding 1 x 108 independent transformants (and >99% library coverage by Poisson distribution).
Selection of amplified E.coli genes that confer thermal resistance
We grew cells in 2xTY broth supplemented with 100 µg/ml ampicillin at 37°C for 3 h starting from an initial OD600 of 0.1. We diluted cells 10-fold into 2xTY broth and heated them in an Eppendorf Mastercycler gradient PCR machine in a 96-well plate for 2 h (in a 100 µl volume). After the heat selection step, we plated cells on 2xTY agar plates supplemented with 100 µg/ml ampicillin and 2% glucose. We visually inspected the plates the next day after incubation at 37°C overnight. Selection temperatures in the first round were as follows: 40.7°C, 41.7°C, 43.5°C, 45.9°C, 48.7°C, 51.6°C, 54.4°C, 56.9°C, 58.9°C, 60.1°C and 60.3°C. Selection temperatures in second round were 47.5°C, 48.0°C, 48.6°C, 49.4°C, 50.3°C, 51.1°C, 51.9°C, 52.5°C and 52.9°C.
We extracted plasmid DNA after each round of selection using a DNA miniprep kit (Qiagen) and amplified the inserts by PCR using primers 5'-GCT TAT CAG ACC TTT ACA-3' and 5'-CCC TCA TAG TTA GCG TAA CGA-3'. We resolved the resulting products by agarose gel electrophoresis (PCR products exceed plasmid insert sizes by approximately 300 bp due to the amplification of flanking sequences). To analyze the heat resistance of individual clones, we transformed plasmid DNA into E.coli TG1 and plated on 2xTY agar plates supplemented with 100 µg/ml ampicillin and 2% glucose. We grew cultures from individual colonies in 2xTY broth supplemented with 100 µg/ml ampicillin and 4% glucose at 37°C. The next day, cells we the diluted cells and heated them in a gradient PCR machine (as described above).
We amplified fragments of the evgA gene by PCR from plasmid preparations of clone evgAf using primer pairs (I) 5'-GCT TAT CAG ACC TTT ACA-3', 5'-TGA CGA TAT CAG GCT TAA GTG TTT CCA C-3' and (II) 5'-ATG AGA TAT CGT CAT CAT TAA TGT CGA CAT CCC CGG AG-3', 5'-CCC TCA TAG TTA GCG TAA CGA-3'. We digested the amplified fragments with EcoRV and either XhoI (fragment I) or RsrII (fragment II) restriction endonuclease, ligated into XhoI/RsrII digested plasmid pW656 and transformed the resulting construct into E.coli TG1.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Selection of amplified E.coli genes conferring thermal resistance
To select for genes that, when provided at amplified copy number, confer thermal resistance (Fig. 1 schematic) we first transformed a library of high-copy-plasmid borne E.coli genomic fragments into E.coli cells. After growth in liquid media, the cells were heated for 2 h at temperatures ranging from 40°C to 60°C at 1–2.8°C intervals. Approximately 103 cells were obtained after heating at 48.7°C, while no colonies were obtained at higher temperatures in this first round of selection.
|
Plasmid DNA was isolated from cells surviving at 48.7°C and used to retransform fresh E.coli cells. The retransformed cells were used to inoculate liquid media and a second round of thermal selection carried out with heating at 47–53°C for 2 h (0.4–0.9°C steps were used). After the second round of selection, we obtained 31 colonies after heating at 50.3°C, whereas no colonies were obtained for the vector control. No colonies were obtained when heating the library at higher temperature while the next lower selection temperature (49.4°C) yielded
103 colonies. All 31 colonies from the 50.3°C selection step were sequenced of which 23 yielded analyzable DNA sequences. This revealed that only two clones harbouring a single gene, encoding the transcriptional activator EvgA (Utsumi et al., 1992), had been selected. Although both clones encode the complete EvgA open-reading frame, they have the opposite orientation in respect to the library plasmid (clone evgAf: W3110 2488937–2490052, 11 x observed; clone evgAb: W3110 2489885–2489013, 12 x observed). This strongly suggested that the two rounds of selection were sufficient to provide convergence in the library.
To examine the course of the selection for thermal resistance, we prepared DNA from the plasmid populations before selection and after each round of selection (Fig. 2A) and examined the distribution of clone size by PCR. Before any selection on the pool the plasmid inserts form a broad smear with no single clone dominant. After one round of selection (at 48.7°C), the distribution is similar to that before selection. However, after the second round of selection (at 49.4°C), two bands of 1300 and 1100 bp dominate. These bands correspond to the insert sizes of the EvgA encoding clones evgAf and evgAb. In order to analyze whether any additional clones had been selected we digested the template DNA with a restriction endonuclease (EcoRV) cleaving a 6 bp recognition site within the evgA gene. After PCR amplification of the digested preparation (Fig. 2A; R2 
evgA) a single band of about 1500 bp was observed (Fig. 2A; ydeO). We sequenced 16 clones after transformation of the preparation into E.coli TG1 of which (8/16) had an identical insert (W3110 1585694–1584416), encoding the transcriptional activator YdeO, which is known to be directly regulated by EvgA (Masuda and Church, 2003).
|
Synthetic evgA amplification confers thermal resistance
To confirm the effect of the selected clones on heat resistance, plasmid preparations were prepared and transformed into E.coli and survival was examined after heating. Considering the highest temperature under which we had observed colonies in the selection (50.3°C), we tested the next temperature step (51.1°C) to see if we would observe survival under this condition. This experiment confirmed that amplification of both evgA and ydeO significantly increased the thermal resistance (Fig. 2B). We also analyzed whether, in the case of the evgA gene, the observed effect was directly linked to the encoded transcription factor (rather than an effect mediated by DNA or RNA elements). For this purpose, we introduced a conservative aspartate to asparagine change into EvgA at position 52, a mutation, which had previously been shown to abolish EvgA activity in vivo (Utsumi et al., 1994). The experiment revealed that the D52 N mutation indeed completely abolished the heat resistant phenotype (Fig. 2B).
To further investigate the extent to which evgA amplification confers thermal resistance on E.coli, survival at a wider temperature range (48.3–54.1°C) was examined. For this purpose, clone evgAf and the background pW656 vector were transformed into E.coli and compared to wild-type bacteria. Outcomes of a typical heating experiment are shown in Figs 3 and 4. Cells transformed with the background vector or untransformed cells show a steep and comparable temperature dependent death (Fig. 4) indicating that neither plasmid propagation nor the presence of ampicillin for plasmid maintenance had a measurable effect on thermal resistance. In contrast, cells transformed with plasmid encoded EvgA show enhanced thermal resistance over a range of temperatures and the temperature dependent thermal resistance curve has a shallower slope that of wild-type cells (Fig. 4).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
EvgA is master transcriptional regulator, and part of a two-component signal transduction system in enterobacteria (Stock et al., 1989
; Perraud et al., 1998). It is activated by the sensor protein and histidine kinase EvgS on Asp 52 (Perraud et al., 1998) in response low pH, a physiological condition in enterobacteria encountered during their passage through the stomach. Previous work by Masuda and Church (2002, 2003) has shown that EvgA over-expression upregulates the transcription of 37 genes. Fifteen of these genes are upregulated directly by EvgA, whereas the remaining 22 are regulated indirectly through YdeO. EvgA, YdeO, and genes they regulate were shown to increase survival under acidic conditions (Masuda and Church, 2002, 2003). Since we find that, at elevated temperatures, evgA amplification is strongly selected and ydeO amplification is the next most strongly selected we suggest that the transcriptional network that confers thermal resistance overlaps with, and may be identical to, that turned on in response to acidic conditions. Since a conservative mutation (D52 N) of the phosphorylation site on plasmid encoded EvgA destroys thermal resistance, this would suggest that, in the case of evgA amplification, phosphorylation is still required for activation of downstream genes. Overlapping acid and thermal resistance pathway regulation may provide part of a mechanistic explanation for the observation that prior exposure of E.coli to acid stress confers increased subsequent thermal resistance (this has long been know by food engineers and complicates the pasteurization of grape juice) (Cheng et al., 2002).
The genes activated by EvgA include a number involved in periplasmic function—including periplasmic chaperones—as well as in membrane and permeability functions (Masuda and Church, 2002, 2003). This is in agreement with Lenski and co-workers experiments with long-term exposure to elevated temperatures that demonstrate the importance of extra-cytoplasmic stress response for prolonged incubation at elevated and lethal temperatures (Riehle et al., 2003).
Heating places thermal stress on every molecule in the cell, and therefore requires a response that is able to compensate for multiple potentially catastrophic defects in molecular function. Our results suggest that a complex molecular problem such as this can be effectively addressed by amplification of a single master transcriptional regulator. Thus, our experiments indicate that evgA amplification can confer significant thermal resistance on E.coli cells (102- to 104-fold increase in survival observed).
It will be interesting to examine the range of phenotypes that can be accessed by gene amplification, to investigate whether natural gene amplification hotspots for complex phenotypes correlate with the amplification of master regulators, and to quantitatively investigate how gene amplification alters the dynamics of gene regulation (Shen-Orr et al., 2002; Mangan et al., 2003) and the cell-to-cell variability in protein copy number (Rosenfeld et al., 2005).
|
Funding |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
This work was funded by the Medical Research Council.
|
Footnotes |
|---|
2 Present address: Garvan Institute of Medical Research, Victoria Road, Darlinghurst, Sydney, NSW 2010, Australia
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
D.C. was a Fellow of Trinity College (Cambridge University) throughout the course of this work. J.W.C. is an EMBO Young Investigator.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionFundingAcknowledgementsReferences |
|---|
Arsene F., Tomoyasu T., Bukau B. Int J Food Microbiol. (2000) 55:3–9.[CrossRef][Web of Science][Medline]
Burger H., Capello A., Schenk P.W., Stoter G., Brouwer J., Nooter K. Biochem. Biophys. Res. Commun. (2000) 269:767–774.[CrossRef][Web of Science][Medline]
Chen P., Bishai W.R. Infect. Immun. (1998) 66:5099–5106.
Cheng H.Y., Yang H.Y., Chou C.C. J. Food Prot. (2002) 65:260–265.[Web of Science][Medline]
Christ D., Winter G. J. Mol. Biol. (2006) 358:364–371.[CrossRef][Web of Science][Medline]
Clarke L., Carbon J. Proc. Natl. Acad. Sci. USA (1975) 72:4361–4365.
Cotrim P.C., Garrity L.K., Beverley S.M. J. Biol. Chem. (1999) 274:37723–37730.
Gibson T.J. Studies on the Epstein-Barr virus genome. (1984) Cambridge: University of Cambridge.
Gonzalez E., et al. Science (2005) 307:1434–1440.
Hassold T.J., Jacobs P.A. Annu. Rev. Genet. (1984) 18:69–97.[CrossRef][Web of Science][Medline]
Hastings P.J., Slack A., Petrosino J.F., Rosenberg S.M. PLoS Biol. (2004) 2:e399.[CrossRef][Medline]
Hoogenboom H.R., Griffiths A.D., Johnson K.S., Chiswell D.J., Hudson P., Winter G. Nucleic Acids Res. (1991) 19:4133–4137.
Horiuchi T., Horiuchi S., Novick A. Genetics (1963) 48:157–169.
Horiuchi T., Tomizawa J.I., Novick A. Biochim. Biophys. Acta (1962) 55:152–163.[Medline]
Iafrate A.J., Feuk L., Rivera M.N., Listewnik M.L., Donahoe P.K., Qi Y., Scherer S.W., Lee C. Nat. Genet. (2004) 36:949–951.[CrossRef][Web of Science][Medline]
Kugelberg E., Kofoid E., Reams A.B., Andersson D.I., Roth J.R. Proc. Natl. Acad. Sci. USA (2006) 103:17319–17324.
Ledbetter D.H., Riccardi V.M., Airhart S.D., Strobel R.J., Keenan B.S., Crawford J.D. N. Engl. J. Med. (1981) 304:325–329.[Web of Science][Medline]
Lund P.A. Adv. Microb. Physiol. (2001) 44:93–140.[Web of Science][Medline]
Luscombe N.M., Babu M.M., Yu H., Snyder M., Teichmann S.A., Gerstein M. Nature (2004) 431:308–312.[CrossRef][Medline]
Mangan S., Zaslaver A., Alon U. J. Mol. Biol. (2003) 334:197–204.[CrossRef][Web of Science][Medline]
Masuda N., Church G.M. J. Bacteriol. (2002) 184:6225–6234.
Masuda N., Church G.M. Mol. Microbiol. (2003) 48:699–712.[CrossRef][Web of Science][Medline]
Niebuhr E. Hum. Genet. (1978) 44:227–275.[CrossRef][Web of Science][Medline]
Nilsson A.I., Zorzet A., Kanth A., Dahlstrom S., Berg O.G., Andersson D.I. Proc. Natl. Acad. Sci. USA (2006) 103:6976–6981.
Park K.S., Jang Y.S., Lee H., Kim J.S. J. Bacteriol. (2005) 187:5496–5499.
Perraud A.L., Kimmel B., Weiss V., Gross R. Mol. Microbiol. (1998) 27:875–887.[CrossRef][Web of Science][Medline]
Pollack J.R., Perou C.M., Alizadeh A.A., Eisen M.B., Pergamenschikov A., Williams C.F., Jeffrey S.S., Botstein D., Brown P.O. Nat. Genet. (1999) 23:41–46.[Web of Science][Medline]
Riehle M.M., Bennett A.F., Lenski R.E., Long A.D. Physiol. Genomics (2003) 14:47–58.
Rosenfeld N., Young J.W., Alon U., Swain P.S., Elowitz M.B. Science (2005) 307:1962–1965.
Roth J.R., Kugelberg E., Reams A.B., Kofoid E., Andersson D.I. Annu. Rev. Microbiol. (2006) 60:477–501.[CrossRef][Web of Science][Medline]
Sebat J., et al. Science (2004) 305:525–528.
Shen-Orr S.S., Milo R., Mangan S., Alon U. Nat. Genet. (2002) 31:64–68.[CrossRef][Web of Science][Medline]
Slack A., Thornton P.C., Magner D.B., Rosenberg S.M., Hastings P.J. PLoS Genet. (2006) 2:e48.[CrossRef][Medline]
Stock J.B., Ninfa A.J., Stock A.M. Microbiol. Rev. (1989) 53:450–490.
Teichmann S.A., Babu M.M. Nat. Genet. (2004) 36:492–496.[CrossRef][Web of Science][Medline]
Utsumi R., Katayama S., Ikeda M., Igaki S., Nakagawa H., Miwa A., Taniguchi M., Noda M. Nucleic Acids Symp. Ser. (1992) 149–150.
Utsumi R., Katayama S., Taniguchi M., Horie T., Ikeda M., Igaki S., Nakagawa H., Miwa A., Tanabe H., Noda M. Gene (1994) 140:73–77.[CrossRef][Web of Science][Medline]
Yeh C.H., Chang P.F., Yeh K.W., Lin W.C., Chen Y.M., Lin C.Y. Proc. Natl. Acad. Sci. USA (1997) 94:10967–10972.
Yura T., Nagai H., Mori H. Annu. Rev. Microbiol. (1993) 47:321–350.[CrossRef][Web of Science][Medline]
Received August 28, 2007; revised October 22, 2007; accepted November 22, 2007.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||