Scholarly article on topic 'Utilizing a highly responsive gene, yhjX, in E. coli based production of 1,4-butanediol'

Utilizing a highly responsive gene, yhjX, in E. coli based production of 1,4-butanediol Academic research paper on "Chemical engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Chemical Engineering Science
OECD Field of science
Keywords
{"1 / 4-Butanediol" / YhjX / Biosensor / "Host engineering" / "Metabolic engineering" / "Product toxicity"}

Abstract of research paper on Chemical engineering, author of scientific article — Heather L Szmidt-Middleton, Mario Ouellet, Paul D Adams, Jay D Keasling, Aindrila Mukhopadhyay

Abstract The role of yhjX, a predicted major facilitator superfamily protein, was examined in context of E. coli response to 1,4-butanediol (1,4-BDO). E. coli DH1 and MG1655, two commonly used metabolic engineering hosts, were both sensitive to the presence of 1,4-BDO in the growth medium, but to different extents. The strains also showed differences in the transcriptional response of the yhjX gene that was highly induced in response to 1,4-BDO. yhjX deletion improved growth of the E. coli strains in the control defined medium but did not significantly impact 1,4-BDO sensitivity. Overexpression of yhjX using a plasmid-borne copy and lactose-inducible promoter also did not result in an improvement in 1,4-BDO tolerance. However, the large differential expression of yhjX in response to this diol provided the foundation to develop a biosensor for the detection of 1,4-BDO using a fluorescent gene under the control of the yhjX promoter. A basic P yhjX :GFP biosensor in E. coli DH1 allows the detection of 4–7% 1,4-BDO in the extracellular medium and provides a tool for high throughput engineering for improving 1,4-BDO production strains.

Academic research paper on topic "Utilizing a highly responsive gene, yhjX, in E. coli based production of 1,4-butanediol"

~ CHEMICAL ENGINEERING SCIENCE

Contents lists available at ScienceDirect

Chemical Engineering Science

journal homepage: www.elsevier.com/locate/ces

Utilizing a highly responsive gene, yhjX, in E. coli based production of 1,4-butanediol$

Heather L Szmidt-Middleton a,b,c, Mario Ouelleta,b,c, Paul D Adams b,c, Jay D Keaslinga,c, Aindrila Mukhopadhyaya'b,c'*

a Fuel Synthesis Division, Joint Bioenergy Institute, 5885 Hollis St, Emeryville, CA, 94608, USA b Technology Division, Joint Bioenergy Institute, 5885 Hollis St, Emeryville, CA, 94608, USA

c Physical Biosciences Division, Lawrence Berkeley National Laboratory, 1 Cyclotron Road, MS 978, Berkeley, CA 94720, USA

CrossMark

HIGHLIGHTS

• Two different E. coli strains were found to be sensitive, but to different extents, to 1,4-BDO.

• The yhjX gene, encoding a transporter, was highly upregulated during 1,4-BDO exposure

• The 1,4-BDO related yhjX response also differed in the two E. coli strains.

• Neither yhjX gene deletion nor overexpression impacted 1,4-BDO tolerance in E. coli.

• We demonstrated the use of the PyhjX promoter as an E. coli biosensor for 1,4 BDO.

ARTICLE INFO ABSTRACT

The role of yhjX, a predicted major facilitator superfamily protein, was examined in context of E. coli response to 1,4-butanediol (1,4-BDO). E. coli DH1 and MG1655, two commonly used metabolic engineering hosts, were both sensitive to the presence of 1,4-BDO in the growth medium, but to different extents. The strains also showed differences in the transcriptional response of the yhjX gene that was highly induced in response to 1,4-BDO. yhjX deletion improved growth of the E. coli strains in the control defined medium but did not significantly impact 1,4-BDO sensitivity. Overexpression of yhjX using a plasmid-borne copy and lactose-inducible promoter also did not result in an improvement in 1,4-BDO tolerance. However, the large differential expression of yhjX in response to this diol provided the foundation to develop a biosensor for the detection of 1,4-BDO using a fluorescent gene under the control of the yhjX promoter. A basic PyhjX: GFP biosensor in E. coli DH1 allows the detection of 4-7% 1,4-BDO in the extracellular medium and provides a tool for high throughput engineering for improving 1,4-BDO production strains.

© 2013 The Authors. Published by Elsevier Ltd. All rights reserved.

Article history: Received 13 April 2013 Received in revised form 21 June 2013 Accepted 25 June 2013 Available online 3 July 2013

Keywords:

1,4-Butanediol

Biosensor Host engineering Metabolic engineering Product toxicity

1. Introduction

Green production of chemicals using microbial platforms has become an achievable target. While metabolic engineering continues to advance microbial production of high value commodities such as pharmaceuticals (e.g. taxol, Ajikumar et al., 2010; Chang and Keasling, 2006), nutraceuticals (e.g. tyrosine, Juminaga et al., 2012; lycopene, Das et al., 2007) and other high value commodities (Bomgardner, 2012), recently, bulk products such as solvents

☆This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-No Derivative Works License, which permits non-commercial use, distribution, and reproduction in any medium, provided the original author and source are credited.

n Corresponding author at: Lawrence Berkeley National Laboratory, Physical Biosciences Division, 1 Cyclotron Road, MS 978, Berkeley, CA 94720 USA. Tel./ fax: +1510 495 4252.

E-mail address: amukhopadhyay@lbl.gov (A. Mukhopadhyay).

(isobutanol, Atsumi et al., 2010), polymer precursors (styrene, McKenna and Nielsen, 2011) and biofuels (bisabolene, Peralta-Yahya et al., 2011; fatty acid ethyl esters, Steen et al., 2010; isopentenol, Chou and Keasling, 2012) are also being successfully targeted. Synthetic biology has begun to play an important role in key aspects of strain engineering including improved pathway regulation (Zhang et al., 2012a, 2012b), strategies and tools for chromosomal integration (Kuhlman and Cox, 2010), tolerance engineering (Dunlop et al., 2010) and pathway optimization (Moon et al., 2010), and is poised to become ubiquitous in many aspects of such efforts (Keasling, 2012). Strain development often requires an iterative process wherein numerous combinations of parts must be evaluated. Given the large number of parameters that require co-optimization, high-throughput strain generation is an essential aspect of strain engineering that must also be coupled with similarly paced measurements of strain performance. Synthetic biology can play a powerful role in the development of biosensors that aid in the development of high-throughput

0009-2509/$ - see front matter © 2013 The Authors. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016Zj.ces.2013.06.044

diagnostic assays to report on flux to intermediates (Pfleger et al., 2007) and final compounds (Dietrich et al., 2012).

Bulk chemicals, such as solvents or solvent-like compounds, are often toxic to microbes that may be used to produce them (Heipieper et al., 1994; Jarboe et al., 2010; Sangster, 1989). A complete description of 1,4-BDO production in Escherichia coli was recently published (Yim et al., 2011) and provides an excellent example of a toxic end product that was successfully engineered to obtain titers of almost 20 gL_ 1. However, this report, along with previous commentaries from the authors of this research (Burk, 2010), estimate a production level of about 100 g L"1 of 1,4-BDO (10% in water) required for the process to be commercially viable. Thus, host engineering to improve production by at least 5 fold is still required in the case of this compound. The toxicity of this compound to E. coli is hypothesized to be a key factor that restricts the upper limit in production levels (Burk, 2010). Recognizing 1, 4-BDO as an important bio-product, we conducted tolerance studies on this compound. We examined the role of a highly responsive gene, yhjX, encoding a putative major facilitator super-family 1 (MFS1) pump, in tolerance phenotypes and the applicability of the promoter of this locus for biosensor applications.

2. Materials and methods

2.1. Strain and plasmid construction

E. coli K-12 MG1655 and DH1 were used as the wild type strains for all experiments. The DH1 and MG1655 yhjX gene deletion strains were constructed using the yhjX-KO-Forward and yhjX-KO-Reverse primers and confirmed as described for E. coli BW25113 (Baba et al., 2006) (Supplementary Fig. S1). For the yhjX expression plasmid, the yhjX open reading frame was PCR amplified from E. coli BW25113 using the forward primer 5-gctcaggaattcatgacacctt-caaattatca-3' and the reverse primer 5'-gctcagggatccttaaagggagc-3', and cloned using restriction enzymes (BamHI and EcoRI) into a pBba5k vector containing a Piac promoter (Lee et al., 2011). Wild type and yhjX knockout strains were transformed with the Plac: yhjX plasmid for overexpression assays. An empty plasmid, with only the Plac promoter, transformed into the strains was used as the control in each case. For biosensor applications, a plasmid with the green fluorescent protein (GFP) driven by the yhjX gene promoter, PyhjX:GFP was obtained from an existing library (Zaslaver et al., 2006) and was transformed into E. coli DH1 and MG1655. Control strains consisted of the same construct with the yhjX promoter removed using restriction enzymes (BamHI and Xhol) and blunt end cloning.

2.2. Adapting cells

All experiments were performed with strains adapted to M9 minimal media (per liter: 200 ml 5 x M9 salts, 2 ml 1 M MgSO4, 50 ml 20% glucose, 20 ml 5% Casamino acids, 100 pl 0.5% Thiamine, 100 pl 1 M CaCl2) and made into single use stocks using methods described previously (Dunlop et al., 2011).

2.3. Culturing methods

For toxicity assays, adapted single use glycerol stocks were used to inoculate fresh M9 medium supplemented with 50 mg/ml of kanamycin when necessary. To determine the effect of 1,4-BDO (J.T. Baker, Avantor Performance Materials, PA, USA) on E. coli growth, a wide range of concentrations were tested first (data not shown) and then narrowed to a range that caused stress but not significant cell death. Growth assays to test the effect of different concentrations of 1,4-BDO were performed using 24-well

microtiter plates with low evaporation lids containing 800 ml of M9 medium and 3.75 ml of thawed single use stocks per well. Unless stated otherwise, all subsequent 1,4-BDO stress assays were conducted at 6% (v/v) 1,4-BDO.

2.4. Growth and fluorescence assays

For growth measurements, cultures were grown at 37 °C in triplicate and optical density (absorbance at 600 nm) was measured every 10 min using a plate reader (Synergy 4, BioTek, VT, USA). For combined growth and fluorescence measurements, cultures were grown at 37 °C in triplicate and optical density (absorbance at 595 nm) and fluorescence (Ex 470 nm/Ex 530 nm) were measured every 18 min using a plate reader (F200 Pro, Tecan, NC, USA).

2.5. Quantitative reverse transcriptase PCR (qPCR)

Overnight cultures of E. coli MG1655 and DH1 grown in M9 were diluted back into 6 separate cultures per strain at a dilution of 1:100 into fresh M9 medium. 1,4-BDO was added to 3 out of the 6 cultures per strain at mid-log phase (optical density at 600 nm [OD600]), 0.35-0.40/ ml. The strains were allowed to grow another hour then harvested, centrifuged (8000g, 4 °C, 5 min) and the pellet stored at - 80 °C. RNA was isolated using the Total RNA kit (Agilent, DE, USA) and any remaining genomic DNA contamination was removed using the Turbo DNA-free kit (Ambion, Life Sciences, NY, USA). cDNA was synthesized following the Superscript III Reverse Transcriptase protocol (Invitrogen, CA, USA). qPCR reactions were created using the PerfeCTa SYBR Green SuperMix ROX (Quanta Biosciences, MD, USA), cDNA template, the forward primer 5'-ccgctggctgacactcatcggtact-3' and the reverse primer 5'-caagccgaaagagaaagcgacctg-3' for yhjX and the forward primer 5'-gctgatgctggtgatgattg-3' and the reverse primer 5'-agtcg-cactttgccgtaatc-3' for the control gene hcaT. qPCR was performed using the StepOne Plus Real-Time PCR System (Applied Biosystems, CA, USA). Fold change was calculated using a published mathematical model for relative quantification in real-time PCR (Pfaffl, 2001). For methods and results from the microarray data see supplementary section.

2.6. Overexpression assay

The Isopropyl ß-D-1-thiogalactopyranoside (IPTG, Sigma Aldrich, MO, USA) inducible, Plac:yhjX expression plasmid was compared to an empty control plasmid in the E. coli strains using leaky (0 pM IPTG) or 5 pM IPTG induction. To determine the effect of yhjX expression on tolerance, 24-well plates were prepared as described for growth assays, with 0%, 6% and 10% 1,4-BDO.

2.7. GFP reporter assays for 1,4-butanediol

The PyhjX:GFP plasmid (Zaslaver et al., 2006) was transformed into E. coli MG1655, and DH1 and was assayed for fluorescence in response to varying levels of 1,4-BDO (0-12%).

3. Results

We examined the impact of 1,4-BDO on the growth of E. coli strains using an exogenous addition of 6% 1,4-BDO based on its published inhibitory levels (Burk, 2010). E. coli MG1655 was sensitive to 6% 1,4-BDO, showing about half maximal growth compared to the control without 1,4-BDO. However E. coli DH1 showed higher tolerance to 1,4-BDO compared to MG1655 (Fig. 1).

o o 0.4

£ 0.5

o o 0.4

DH1 0% DH1 6% DH1 10%

8 12 Hours

MG1655 0% MG1655 6% MG1655 10%

8 12 Hours

Fig. 1. Growth impact of 0%, 6% and 10% 1,4-BDO on E. coli strains DH1 (A), and MG1655 (B) measured as change in optical density over time.

Using quantitative RT-PCR we evaluated the differential expression of the most upregulated gene, yhjX, during 1,4-BDO exposure in E. coli DH1 in a preliminary gene expression experiment (Supplementary material). We confirmed the upregulation of yhjX upon exposure to 6% 1,4-BDO in both E. coli strains (Fig. 2). The yhjX transcript was upregulated to different levels the two strains in response to 1,4-BDO, 21 and 41 fold in DH1 and MG1655, respectively. Thus while MG1655 is more sensitive to 1,4-BDO, the transcriptional response of yhjX in DH1 appears to have greater dynamic range.

To examine if differences in the promoter or protein coding sequence of yhjX between DH1 and MG1655 could explain the different levels of upregulation, the yhjX coding sequence and 150 bp upstream of the start codon were amplified from each strain, cloned and sequenced. The sequences from the two strains were found to be identical to the sequences available on ecocyc. Thus, the same variant of the protein is being expressed to different levels using the same promoter sequences in response to 1,4-BDO in these strains. Downstream sequences and other regulatory mechanisms were not examined, but may also impact protein levels and the resulting function.

DH1 and MG1655 are two of the most commonly used E. coli strains for metabolic engineering, and yhjX was highly upregulated in the presence of 1,4-BDO in these strains. Thus, we tested the ability of the yhjX promoter to serve as a biosensor for 1,4-BDO. The promoter region of yhjX (PyhjX) used to control expression of the gene encoding GFP in a plasmid-borne system was introduced into the wild type E. coli strains. Our results indicate that the Pyhjx: GFP construct generates a fluorescent signal that increases with increases in the levels of 1,4-BDO (Fig. 3). E. coli DH1 produced the best response from this plasmid based system, and the optimal range of the GFP based reporter is from 4% to 7% 1,4-BDO (Fig. 4).

5 50 s

I- 30 _ c

O 10 o

MG1655

Fig. 2. Quantitative PCR measurement of the yhjX transcript in response to a one hour 6% 1,4-BDO exposure in E. coli strains DH1 and MG1655. Ratio shown is the level of yhjX cDNA for a given strain grown with or without 6% 1,4-BDO.

35000 30000 25000 20000 15000 10000 5000 0

1 PyhjX DH1 0% PyhjX DH1 6% DH1 Empty 0% 1 DH1 Empty 6%

8 12 Hours

20000 15000 10000 5000 0

PyhjX MG1655 0% PyhjX MG1655 6% MG1655 Empty 0% MG1655 Empty 6%

8 12 Hours

Fig. 3. Specific florescence in E. coli DH1 (A) and E. coli MG1655 (B) containing a Pyhjx: GFP construct using 1,4-BDO (0%, 6%) in the medium. Control vector in both strains is a Pyhjx only construct as described in the Section 2. Specific fluorescence (F/OD) is calculated as Fluorescence units/Abs at 595 nm.

Similar trends but slightly diminished reporter activity was observed in E. coli MG1655 (Fig. S2). A survey of other alcohols (e.g. 1,5-pentanediol, 1,3-propanediol) elicited variable fluorescent signals (Supplementary Fig. S3), suggesting that the PyhjX:GFP construct may be useful for the detection of other end metabolites also.

We also evaluated the impact of yhjX on 1,4-BDO sensitivity in E. coli. We constructed gene deletion strains in MG1655 and DH1. Deletion of yhjX resulted in an improved growth in control medium relative to the wild-type strains, especially for E. coli MG1655 and marginally also for E. coli DH1 (Fig. 5). However, the

O 15000

10000 5000 0

......I-........i

-0% £ c 04

-------1% o <£> Q 0.3

-----2% o

-------3%

-----4% 0.1

.......... 5% 0

O 15000

5000 0

y T 5%

'"/ ..-J'....... L -----6%

7% — 8% 9% 10% 12%

8 12 Hours

Fig. 4. Specific florescence in E. coli DH1 containing a PyhjX: GFP construct with increasing levels of 1,4-BDO (0-5%) (A), and (5-12%) (B). Equivalent survey of the PyhjX: GFP construct in E. coli MG1655 is provided in the supplementary section. Specific fluorescence (F/OD) is calculated as Fluorescence units/Abs at 595 nm.

response to 1,4-BDO remained similar to the wild type strains. Similarly, expression of the yhjX gene did not result in any improvement in 1,4-BDO tolerance (Fig. 6). Using IPTG levels greater than 5 pM during 1,4-BDO exposure resulted in additional reduction in growth (data not shown). Since the impact of yhjX overexpression may be better observed in a yhjX deleted strain, Plac:yhjX constructs were evaluated with and without 1,4-BDO exposure in both wild type and AyhjX strains and compared with the same strains harboring a control vector (Fig. 6). Similar to the wild type strains with a native copy of yhjX, the AyhjX strains also showed no improvement in 1,4-BDO tolerance using yhjX expression.

4. Discussion

For compounds such as 1,4-BDO, measuring the production of the final metabolite is a key step in metabolic engineering efforts. However, in the successful employment of a combinatorial approach, testing a large number of gene alterations and growth parameters, analytical measurements such as GC-MS, though precise, can be the rate-limiting step. Genes that have high and specific upregulation in response to the final product can be invaluable in the development of biosensors and are amenable for use in a high throughput manner. A recent example is that of a biosensor for butanol, developed using a heterologous n-butanol-sensing transcription factor-promoter pair in E. coli (Dietrich et al., 2012), that enables production measurements that are intrinsically

.....|.......|.......I H

DH1 WT 0% DH1 WT 6% DH1 WT 10% DH1 0% DH1 6% DH1 10%

8 12 Hours

MG1655 WT 0% MG1655 WT 6% MG1655 WT 10% MG1655 0% MG1655 6% MG1655 10%

12 Hours

Fig. 5. Comparison of the E. coli DH1 to E. coli DH1 AyhjX (A) and E. coli MG1655 to E. coli MG1655AyhjX (B) in the presence of 0%, 6% 10% 1,4-BDO in the medium.

high throughput. A review of the process to manufacture 1,4-BDO in E. coli states that an economically viable process requires 10% or more production (Burk, 2010; Yim et al., 2011), while the current maximum reported is about 2% 1,4-BDO (Yim et al., 2011). As such, considerable additional strain engineering remains in the efforts to improve production above this current 2%. A 1,4-BDO biosensor provides a useful tool to expedite such engineering. The growth conditions used for 1,4-BDO production itself would have little or no impact on the PyhjX:reporter performance, as the biosensor strain is likely to be used separate from the production strain. Under the growth conditions used in our study, the dynamic range of the generic GFP reporter is from 4% to 7% 1,4-BDO and is fairly specific to this diol relative to other components in the media. These limits are possibly defined by the toxicity of 1,4-BDO. While the toxicity may effectively induce genes only at concentrations of 4% or greater, the impact on cell growth beyond 7% places the upper limit on this detection. This range can be optimized using alternate reporter proteins that have greater sensitivity, reformulated media to reduce autofluorescence, different promoter strengths (such as by altering the RBS), and plasmid copy number. Such a biosensor can be used either within a production strain or via exogenous addition of extracted production cultures. A platform that uses exogenous detection allows both concentration and dilution of the production culture extract so as to further extend the 4-7% detection range. Further, an exogenous detection platform also eliminates the requirement of additional engineering of the production strain.

Since tolerance engineering for the development of robust hosts is a valuable goal we wanted to explore the potential role of yhjX in 1,4-BDO tolerance. yhjX is a putative major facilitator superfamily 1 transporter (Pao et al., 1998), suggested to be an oxalate-malate antiporter in most E. coli annotation databases

WT Empty 0% WT Empty 6% WT Empty 10% WT +yhjX 0% WT +yhjX 6% WT + yhjX 10%

- Empty 0% Empty 6% Empty 10%

- +yhjX 0% +yhjX 6%

- +yhjX 10%

8 12 Hours

Fig. 6. 1,4-BDO exposure in E. coli DH1 wild type or DH1 AyhjX strains containing either a plasmid with a Plac only (empty) or a Plac: yhjX (yhjX) construct and grown in the presence of 0%, 6% 10% 1,4-BDO in the medium.

(Keseler et al., 2011). Very recently, in E. coli MG1655, yhjX was found to be the sole target of a two-component system YpdA/YpdB that was highly and specifically induced by pyruvate (Fried et al., 2013). Ecogene.org (Zhou and Rudd, 2013) lists other studies where the yhjX transcript shows upregulation. These include the study of a solvent-tolerant E. coli strain (TK31, derived from JA300) in a cyclohexane:p-xylene mix (Shimizu et al., 2005), E. coli W3110 exposed to oxygen limitation and pH change (Hayes et al., 2006), E. coli W3110 exposed to benzoate (Kannan et al., 2008), and MG1655 treated with small toxic peptides (Fozo et al., 2008). A factor that makes it challenging to compare results from these various studies is that they were conducted in different E. coli strains, using different growth conditions and measured using different measurement platforms. However, the large number of conditions and compounds evaluated in other reports (Fried et al., 2013) indicates yhjX to be very specifically regulated in E. coli MG1655.

In the present study, in both DH1 and MG1655, the deletion of this gene resulted in an improved growth profile under control growth conditions but neither strain showed any significant change in 1,4-BDO tolerance. Expression of this gene, using a plasmid borne expression system, provided no additional tolerance in either the wild type or the AyhjX strains. Higher induction of the yhjX gene expression resulted in growth impairment in both E. coli strains. Membrane protein overexpression often results in a decrease in growth (Wagner et al., 2007), and in host engineering applications very low induction is typically ideal (Dunlop et al., 2011). It is possible that high levels of the YhjX transporter is only beneficial under specific conditions but is otherwise burdensome. Such a burden may explain the growth improvement observed in the gene deletion strains in control medium, but not during 1,4-BDO exposure. None of the conditions we tested could identify YhjX levels that would provide better 1,4-BDO tolerance relative to the control strains.

Genes that show such high and relatively specific upregulation in engineered strains have also been used to develop dynamic sensory systems. Recent examples of such use include promoters that respond to pathway intermediates and resulted in 7.5 fold improvement in production (Zhang et al., 2012a, 2012b). While in these cases, the dynamic sensory system was used to drive a

metabolic pathway; native regulation has also been used to express heterologous proteins in a robust, inducer free manner (Tsao et al., 2010). The PyhjX promoter may also serve as a promoter to regulate 1,4-BDO export or tolerance mechanisms, so as to provide optimal levels of the corresponding proteins, reduce conflict with promoters being used for other engineered parts, and to avoid the use of costly inducers such as IPTG.

Acknowledgments

We thank Marijke Frederix (JBEI, LBNL) for help with constructing the yhjX mutants in DH1 and MG1655. This work, conducted by the Joint BioEnergy Institute, was supported by the Office of Science, Office of Biological and Environmental Research, of the U. S. Department of Energy under Contract no. DE-AC02-05CH11231.

Appendix A. Supporting information

Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.ces.2013.06.044.

References

Ajikumar, P.K., Xiao, W.H., Tyo, K.E., Wang, Y., Simeon, F., Leonard, E., Mucha, O., Phon, T.H., Pfeifer, B., Stephanopoulos, G., 2010. Isoprenoid pathway optimization for Taxol precursor overproduction in Escherichia coli. Science 330, 70-74.

Atsumi, S., Wu, T.Y., Machado, I.M., Huang, W.C., Chen, P.Y., Pellegrini, M., Liao, J.C., 2010. Evolution, genomic analysis, and reconstruction of isobutanol tolerance in Escherichia coli. Mol. Syst. Biol. 6, 449.

Baba, T., Ara, T., Hasegawa, M., Takai, Y., Okumura, Y., Baba, M., Datsenko, K.A., Tomita, M., Wanner, B.L., Mori, H., 2006. Construction of Escherichia coli K-12 inframe, single-gene knockout mutants: the Keio collection. Mol. Syst. Biol. 2 (2006), 0008.

Bomgardner, M.M., 2012. The Sweet Smell Of Microbes. Chem. Eng. News 90, 25-29.

Burk, M.J., 2010. Sustainable production of industrial chemicals from sugars. Int. Sugar J. 112, 30-35.

Chang, M.C., Keasling, J.D., 2006. Production of isoprenoid pharmaceuticals by engineered microbes. Nat. Chem. Biol. 2, 674-681.

Chou, H.H., Keasling, J.D., 2012. Synthetic pathway for production of five-carbon alcohols from isopentenyl diphosphate. Appl. Environ. Microbiol. 78, 7849-7855.

Das, A., Yoon, S.H., Lee, S.H., Kim, J.Y., Oh, D.K., Kim, S.W., 2007. An update on microbial carotenoid production: application of recent metabolic engineering tools. Appl. Microbiol. Biotechnol. 77, 505-512.

Dietrich, J.A., Shis, D.L., Alikhani, A., Keasling, J.D., 2012. Transcription factor-based screens and synthetic selections for microbial small-molecule biosynthesis. ACS Synth. Biol.

Dunlop, M.J., Dossani, Z.Y., Szmidt, H.L., Chu, H.C., Lee, T.S., Keasling, J.D., Hadi, M.Z., Mukhopadhyay, A., 2011. Engineering microbial biofuel tolerance and export using efflux pumps. Mol. Syst. Biol. 7, 487.

Dunlop, M.J., Keasling, J.D., Mukhopadhyay, A., 2010. A model for improving microbial biofuel production using a synthetic feedback loop. Syst. Synth. Biol. 4, 95-104.

Fozo, E.M., Kawano, M., Fontaine, F., Kaya, Y., Mendieta, K.S., Jones, K.L., Ocampo, A., Rudd, K.E., Storz, G., 2008. Repression of small toxic protein synthesis by the Sib and OhsC small RNAs. Mol. Microbiol. 70, 1076-1093.

Fried, L., Behr, S., Jung, K., 2013. Identification of a target gene and activating stimulus for the YpdA/YpdB histidine kinase/response regulator system in Escherichia coli. J. Bacteriol. 195, 807-815.

Hayes, E.T., Wilks, J.C., Sanfilippo, P., Yohannes, E., Tate, D.P., Jones, B.D., Radmacher, M.D., BonDurant, S.S., Slonczewski, J.L., 2006. Oxygen limitation modulates pH regulation of catabolism and hydrogenases, multidrug transporters, and envelope composition in Escherichia coli K-12. BMC Microbiol. 6, 89.

Heipieper, H.J., Weber, F.J., Sikkema, J., Keweloh, H., Debont, J.A.M., 1994. Mechanisms of resistance of whole cells to toxic organic-solvents. Trends Biotechnol. 12, 409-415.

Jarboe, L.R., Zhang, X., Wang, X., Moore, J.C., Shanmugam, K.T., Ingram, L.O., 2010. Metabolic engineering for production of biorenewable fuels and chemicals: contributions of synthetic biology. J. Biomed. Biotechnol., 761042.

Juminaga, D., Baidoo, E.E., Redding-Johanson, A.M., Batth, T.S., Burd, H., Mukho-padhyay, A., Petzold, C.J., Keasling, J.D., 2012. Modular engineering of L-tyrosine production in Escherichia coli. Appl. Environ. Microbiol. 78, 89-98.

Kannan, G., Wilks, J.C., Fitzgerald, D.M., Jones, B.D., Bondurant, S.S., Slonczewski, J.L., 2008. Rapid acid treatment of Escherichia coli: transcriptomic response and recovery. BMC Microbiol. 8, 37.

Keasling, J.D., 2012. Synthetic biology and the development of tools for metabolic engineering. Metab. Eng. 14, 189-195.

Keseler, I.M., Collado-Vides, J., Santos-Zavaleta, A., Peralta-Gil, M., Gama-Castro, S., Muniz-Rascado, L., Bonavides-Martinez, C., Paley, S., Krummenacker, M., Altman, T., Kaipa, P., Spaulding, A., Pacheco, J., Latendresse, M., Fulcher, C., Sarker, M., Shearer, A.G., Mackie, A., Paulsen, I., Gunsalus, R.P., Karp, P.D., 2011. EcoCyc: a comprehensive database of Escherichia coli biology. Nucleic Acids Res. 39, D583-590.

Kuhlman, T.E., Cox, E.C., 2010. Site-specific chromosomal integration of large synthetic constructs. Nucleic Acids Res. 38, e92.

Lee, T.S., Krupa, R.A., Zhang, F., Hajimorad, M., Holtz, W.J., Prasad, N., Lee, S.K., Keasling, J.D., 2011. BglBrick vectors and datasheets: a synthetic biology platform for gene expression. J. Biol. Eng. 5, 12.

McKenna, R., Nielsen, D.R., 2011. Styrene biosynthesis from glucose by engineered E. coli. Metab. Eng. 13, 544-554.

Moon, T.S., Dueber, J.E., Shiue, E., Prather, K.L., 2010. Use of modular, synthetic scaffolds for improved production of glucaric acid in engineered E. coli. Metab. Eng. 12, 298-305.

Pao, S.S., Paulsen, I.T., Saier Jr., M.H., 1998. Major facilitator superfamily. Microbiol. Mol. Biol. Rev. 62, 1-34.

Peralta-Yahya, P.P., Ouellet, M., Chan, R., Mukhopadhyay, A., Keasling, J.D., Lee, T.S., 2011. Identification and microbial production of a terpene-based advanced biofuel. Nat. Commun. 2, 483.

Pfaffl, M.W., 2001. A new mathematical model for relative quantification in realtime RT-PCR. Nucleic Acids Res. 29, e45.

Pfleger, B.F., Pitera, D.J., Newman, J.D., Martin, V.J., Keasling, J.D., 2007. Microbial sensors for small molecules: development of a mevalonate biosensor. Metab. Eng. 9, 30-38.

Sangster, J., 1989. Octanol-Water partition-coefficients of simple organic-compounds. J. Phys. Chem. Ref. Data 18, 1111-1229.

Shimizu, K., Hayashi, S., Doukyu, N., Kobayashi, T., Honda, H., 2005. Time-course data analysis of gene expression profiles reveals purR regulon concerns in organic solvent tolerance in Escherichia coli. J. Biosci. Bioeng. 99, 72-74.

Steen, E.J., Kang, Y., Bokinsky, G., Hu, Z., Schirmer, A., McClure, A., Del Cardayre, S.B., Keasling, J.D., 2010. Microbial production of fatty-acid-derived fuels and chemicals from plant biomass. Nature 463, 559-562.

Tsao, C.Y., Hooshangi, S., Wu, H.C., Valdes, J.J., Bentley, W.E., 2010. Autonomous induction of recombinant proteins by minimally rewiring native quorum sensing regulon of E. coli. Metab. Eng. 12, 291-297.

Wagner, S., Baars, L., Ytterberg, A.J., Klussmeier, A., Wagner, C.S., Nord, O., Nygren, P. A., van Wijk, K.J., de Gier, J.W., 2007. Consequences of membrane protein overexpression in Escherichia coli. Mol. Cell. Proteomics 6, 1527-1550.

Yim, H., Haselbeck, R., Niu, W., Pujol-Baxley, C., Burgard, A., Boldt, J., Khandurina, J., Trawick, J.D., Osterhout, R.E., Stephen, R., Estadilla, J., Teisan, S., Schreyer, H.B., Andrae, S., Yang, T.H., Lee, S.Y., Burk, M.J., Van Dien, S., 2011. Metabolic engineering of Escherichia coli for direct production of 1,4-butanediol. Nat. Chem. Biol. 7, 445-452.

Zaslaver, A., Bren, A., Ronen, M., Itzkovitz, S., Kikoin, I., Shavit, S., Liebermeister, W., Surette, M.G., Alon, U., 2006. A comprehensive library of fluorescent transcrip-tional reporters for Escherichia coli. Nat. Methods 3, 623-628.

Zhang, F., Carothers, J.M., Keasling, J.D., 2012a. Design of a dynamic sensor-regulator system for production of chemicals and fuels derived from fatty acids. Nat. Biotechnol. 30, 354-359.

Zhang, F., Ouellet, M., Batth, T.S., Adams, P.D., Petzold, C.J., Mukhopadhyay, A., Keasling, J.D., 2012b. Enhancing fatty acid production by the expression of the regulatory transcription factor FadR. Metab. Eng. 14, 653-660.

Zhou, J., Rudd, K.E., 2013. EcoGene 3.0. Nucleic Acids Res. 41, D613-624.