Supporting Information
Supporting Information Design, construction and characterization of a set of biosensors for aromatic compounds Haoran Xue1║, Hailing Shi1║, Zhou Yu1║, Shuaixin He1║, Shiyu Liu1, Yuhang Hou1, Xingjie Pan1, Huan Wang1, Pu Zheng1, Can Cui1, Helena Viets1, Jing Liang1, Yihao Zhang1, Shuobing Chen1,2*, and Haoqian M. Zhang2,3* and Qi Ouyang2,3 1
Peking University Team for the International Genetically Engineered Machine Competition (iGEM), 2Center for Quantitative Biology, and 3Peking-Tsinghua Joint Center for Life Sciences, Peking University, Beijing, 100871, China. ║
These authors contribute equally to this work.
*Correspondence should [email protected]
be
addressed
1
to
[email protected]
and
I. Genetic and material sources of aromatics-‐responsive transcription factors NahR transcription activator is encoded by nahR gene that locates on NAH7 plasmid from the bacterium Pseudomonas putida. NAH7 plasmid encodes a regulator (NahR) and several enzymes (including NahF) that compose the nah operon. NahR responds to salicylate (SaA), an intermediate in naphthalene catabolic pathway, and activates the expression of the catabolic enzymes in nah operon (Fig. S1a). XylS transcription activator is encoded by xylS gene that locates on TOL plasmid pWW0 from the bacterium Pseudomonas putida. pWW0 plasmid encodes two regulators (XylS and XylR) and several enzymes that compose the xylene catabolic pathway. XylS responds to benzoate (BzO), an intermediate in toluene catabolic pathway, and activates the expression of the catabolic enzymes (Fig. S1b). HbpR transcription activator is encoded by hbpR gene from the bacterium Pseudomonas azelaica. HbpR is the master regulator of hbp operon responsible for 2-hydroxybiphenyl (2-HBP) degradation. HbpR responds to 2-HBP and activates the expression of the catabolic enzymes in hbp operon (Fig. S1c). DmpR transcription activator is encoded by dmpR gene from the bacterium Pseudomonas strain CF600. DmpR is the master regulator of dmp operon responsible for phenol (Phl) degradation. DmpR responds to Phl and activates the expression of the catabolic enzymes in dmp operon (Fig. S1d). The names and sequences of the promoters regulated by these four transcription factors are listed in Supplementary Table S1. The material sources of the coding sequences of these transcription factors used in this study are listed in Supplementary Table S2.
2
Supplementary Figure S1. Genetic sources of the aromatics-responsive transcription factors used in this study. (a) Genetic components of naphthalene catabolic pathway. NahR responds to salicylate and activates Pnah promoter, which drives the upstream metabolic pathway, and Psal promoter, which drives the downstream metabolic pathway. (b) Genetic components of benzoate catabolic pathway, which is a truncated version of xylene catabolic pathway. XylS responds to benzoate and activates Pm promoter. (c) Genetic components of 2-hydroxybiphenyl catabolic pathway. HbpR responds to biphenyl derivatives and activates PD and Pc promoters. (d) Genetic components of phenol catabolic pathway. DmpR responds to phenol derivatives and activates Po promoter.
3
Supplementary Table S1. Sequences of transcription factors and cognate promoters used in this study Transcription Factor
nahR
xylS
hbpR
dmpR
Cognate Promoter
Sequence of Cognate Promoter
Psal
TATTTATCAATATTGTTTGCTCCGTTATCGTT ATTAACAAGTCATCAATAAAGCCATCACGAGT ACCATAG
Pm
AATCCTTAGCTTTCGCTAAGGATGATTTTTGG ATTTCGCGGCCGCTACTAGAAAGGCCTACCCC TTAGGCTTTATGCAACTAGAGATT
Pc
TTTAATAAATTTATGAAATCGTGGTTGTGAGT TTTCATAATATGGTGAAGCTTGCCCGCCATGG CAAGTGCATTTCGGCAGCTGTTGGCGACCGCG GAAGGGGTTTACAGCCGTCTGGAGCTAGGCTT TCTGGCGCTCATTAAAATAAAAATCCTTATAA AACAGTATCCTAGCTTTTATGTCTGAGGCTGC TTAGTCAACCTGGCACGGTACTGGCTACGAGT CCCGC
Po
TAAGCATTTGCTCAAGCGGCCTTGGGCAATTG ATCAAATGCTTAAAAAGTCTGCGCAAGCGCGG CTTAATTTCGCTCGCTCCGATCATTCTAAAAA TTAGAAACACATTGAAAAACATTACCTTGAAG TCTGTTTTCAGACCTTGGCACAGCCGTTGCTT GATGTCCTGCG
4
Supplementary Table S2. Material sources of aromatics-responsive transcription factor genes Gene nahR xylS hbpR dmpR
Material Source of Coding Sequence Standard Biological Brick BBa_J61051 from Registry of Standard Biological Parts Gene synthesis Plasmid from Professor Jan Roelof van der Meer, Department of Fundamental Microbiology, University of Lausanne Plasmid from Professor Victoria Shingler, Department of Molecular Biology, Umeå University
5
II. Optimization of XylS and HbpR Biosensors As described in the main text, the primary genetic constructs for XylS and HbpR biosensors failed to exhibit desirable performance. The maximum induction folds for both of the biosensors were very low. This was largely due to exceedingly high sfGFP basal expressions when no inducer was added. We speculated that the high basal expression level might be attributed to two possible reasons: First, aromatics-responsive transcription factors might be excessively expressed, leading to activation of the TFs’ cognate promoters even without inducers. Second, translation strength of sfGFP reporter gene might be too strong, leading to high basal expression even when sfGFP's promoter is not activated. Based on these two speculations, we used a family of constitutive promoters from Registry of Standard Biological Parts to tune transcription factors' expression and used a family of ribosome binding sites (RBSs) to tune sfGFP expression. Constitutive promoter BBa_J23106 was chosen to drive the expression of hbpR gene in the primary construct. To tune the expression of hbpR gene, four constitutive promoters with increasing transcriptional strength, BBa_J23113, BBa_J23109, BBa_J23117 and BBa_J23114, were chosen. All four constitutive promoters are weaker than the original promoter BBa_J23106. Results of ON-OFF tests showed that promoter BBa_J23114 appeared to be the optimal choice (Fig. S2a). The performance of HbpR biosensor is not yet satisfying after the first round of optimization. So we conducted a second round of optimization. Three RBSs with increasing strength, BBa_B0032, BBa_B0031 and BBa_B034, were chosen from Registry of Standard Biological Parts and used to tune sfGFP expression. Results of dose-response curve tests showed that BBa_B0032 was the optimal choice (Fig. S2b). After two rounds of optimization, HbpR biosensor acquired a maximal induction response of over 50 fold. For XylS biosensor, BBa_J23106 was also chosen to drive the expression of xylS gene in the primary construct. As in the case of HbpR biosensor, four constitutive promoters with increasing transcriptional strength, BBa_J23113, BBa_J23109, BBa_J23114 and BBa_J23105, were chosen. Results also showed that BBa_J23114 was the optimal choice (Fig. S3). At this point, XylS biosensor already acquired a maximal induction response of over 100 fold, so the second round of optimization was not conducted. Sequences and relative strengths of the promoters and RBSs used in optimization of HbpR and XylS biosensor was listed in Supplementary Table S3 and S4.
6
Supplementary Figure S2. Optimization of HbpR biosensor. (a) Performance of HbpR biosensor variants with different constitutive promoters controlling hbpR gene. The HbpR biosensor with its hbpR gene controlled by J23114 promoter showed highest induction ration towards both 2-HBP and 2-ABP. (b) Performance of different versions of HbpR biosensor with different RBSs driving sfGFP expression. The constitutive promoter controlling hbpR gene was fixed as J23114 herein. HbpR biosensor with RBS B0032 driving the translation of sfgfp performed the best.
Supplementary Figure S3. Optimization of XylS biosensor. xylS gene was controlled by different constitutive promoters to create different versions of XylS biosensor. Biosensor using J23114 promoter showed highest induction ratio towards 7
BzO, 2-MeBzO, 3-MeBzO and 4-MeBzO. Supplementary Table S3. Sequences and relative promoter strengths of the constitutive promoters used in this study.
Categorized NO.a
Sequence
Relative Promoter Strength b
BBa_J23113
ctgatggctagctcagtcctagggattatgctagc
21
BBa_J23109
tttacagctagctcagtcctagggactgtgctagc
106
BBa_ J23117
ttgacagctagctcagtcctagggattgtgctagc
162
BBa_J23114
tttatggctagctcagtcctaggtacaatgctagc
256
BBa_J23105
tttacggctagctcagtcctaggtactatgctagc
623
BBa_J23106
tttacggctagctcagtcctaggtatagtgctagc
1185
a. The categorized number in Registry of Standard Biological Parts. http://partsregistry.org/Main_Page b. Relative promoter strength adopted from Registry of Standard Biological Parts. http://partsregistry.org/Main_Page
8
Supplementary Table S4. Sequences and relative strengths of the RBSs used in this study. Categorized NO. a
Sequence
Relative RBS Strength b
BBa_B0034
aaagaggagaaa
31121
BBa_B0031
tcacacaggaaacc
3588
BBa_B0032
tcacacaggaaag
2271
a. The categorized number in Registry of Standard Biological Parts. http://partsregistry.org/Main_Page b. The relative RBS strength was predicted using online RBS calculator v1.1: “Reverse Engineer RBS” (https://salis.psu.edu/software/reverse).
9
III. Detection Spectrum of Individual Biosensors & the Interferences on Sensing Capability. III.A. Detection Spectrums Chemical structures and full names of the aromatic compounds discussed below can be found in Supplementary Table S5. In this research, we selected 44 aromatic compounds to comprehensively represent the common aromatics present in the environment either naturally or due to industrial causes. To characterize the biosensors' response to these compounds, first we need to verify that the presence of these compounds exerts no significant toxic effect on the normal cellular processes of E. coli cells. To evaluate such possible toxicity, growth tests were conducted for each of the 44 aromatic compounds using Top10 E. coli strain (Fig. S4). Results showed that, apart from 4 aromatic compounds (5-ClSaD, 2,4,6-TClPhl, 2-HBP and 2-ABP), all the other 40 compounds exert no significant influence on E. coli cells' growth at 1000µM concentration. When the concentrations of the four severely toxic compounds were reduced to 100µM, their toxicity towards E. coli cells was no longer significant (Fig. S4). For each biosensor, ON-OFF test for each of the 44 aromatic compounds was then conducted to identify those compounds that can elicit significant responses. Those aromatic compounds that showed severe toxic effect at 1000µM concentration were added at 100µM concentration. All the other compounds were added at 1000µM concentration (Fig. S5). For NahR biosensor, 16 of the 44 compounds showed significant activation effects with induction folds higher than 20. They are listed as follows: SaA, 3-MeSaA, 4-MeSaA, 4-ClSaA, 5-ClSaA, AsPR, 2,4,6-TClPhl, 2-MeBzO, 3-MeBzO, 4-FBzO, 3-ClBzO, 4-ClBzO, 4-BrBzO, 3-IBzO, 3-MeOBzO and 3-HSaA. These compounds can be classified into salicylate and its derivatives, benzoate and its derivatives and a special compound 2,4,6-TClPhl. 2,4,6-TClPhl is a kind of polychlorinated phenol that is particularly hazardous to water environment and human health. Such a sensing capability of NahR biosensor has not been reported in previous researches. For XylS biosensor, 19 of the 44 compounds showed significant activation effects with induction folds higher than 20. They are listed as follows: BzO, 2-MeBzO, 3-MeBzO, 4-MeBzO, 2-FBzO, 4-FBzO, 2-ClBzO, 3-ClBzO, 4-ClBzO, 2-BrBzO, 4-BrBzO, 3-IBzO, 3-MeOBzO, SaA, 3-MeSaA, 4-ClSaA, 5-ClSaA, 3-MeBAD and 3-ClBAD. These compounds can be roughly classified into Benzoate and its derivatives and 10
Salicylate and its derivatives, with the exception of 3-MeBAD and 3-ClTOL. As illustrated in the main text, HbpR biosensor responds specifically to 2-HBP and 2-ABP. For DmpR biosensor, 4 of the 44 compounds showed significant activation effect with induction folds greater than 5. They are listed as follows: Phl, 2-MePhl, 2-ClPhl and Cat. III.B. Interferences Caused by Non-inducer Chemicals on Sensing Capability Previous studies reported that, for a given biosensor, when aromatic chemicals that are able or unable to induce significant response (typical inducers and non-typical inducers, respectively) co-exist, non-typical inducers might interfere with the response of the biosensor to other inducers, although the non-typical inducers themselves could hardly induce any significant response1 (Fig. 2a). For example, in Pseudomonas putida DOT-T1E, the TodS/TodT two-component system senses the presence of toluene to control expression of toluene metabolism pathway2. Despite the fact that toluene is a strong inducer for TodS (the histidine kinase), toluene derivatives with ortho-substitutions would reduce or totally abolish the in vivo responses of TodS through competing with toluene for binding to TodS3. Another example is much more similar to the cases of biosensors in this manuscript. In Escherichia coli K-12, the mhp operon encodes the catabolic pathways for the degradation of the aromatic compound 3-hydroxyphenylpropionate (3-HPP). The regulator MhpR is able to sense both 3-HPP, the substrate, and 3-(2,3-dihydroxyphenyl) propionate (DHPP), the catabolic product, but unable to sense henylpropionate (PP)4. However, PP is able to activate the MhpR regulator synergistically with the 3-HPP and DHPP; moreover, 3-HPP, DHPP, and PP bind independently to MhpR with similar affinities5. This is probably because that PP is the substrate of the metabolic pathway upstream of mhp pathway, and MhpR integrates the signals of 3-HPP, DHPP and PP, thus to rapidly modulate the transcription of mhp operon. These interferences revealed in previously studies strongly implied that some aromatic chemicals, for instance, the derivatives of the strong inducers, would probably influence the sensing capability of our biosensors, although they couldn’t induce any response by themselves alone. These interferences should be carefully assessed and taken into consideration when using our biosensors.
11
Supplementary Figure S4. The influence of aromatic compounds on the growth of E. coli. The final concentrations of aromatic compounds in the liquid culture were 1000 µM unless otherwise indicated. (a) Growth curves of E. coli Top10 strain when benzoate derivatives were added into the bacterial culture. (b) Growth curves of E. coli Top10 strain when salicylate derivatives were added into the bacterial culture. (c) Growth curves of E. coli Top10 strain when benzaldehyde derivatives, salicylaldehyde derivatives and benzyl alcohol derivatives were added into the bacterial culture. (d) Growth curves of E. coli Top10 strain when phenol derivatives, biphenyl derivatives and other common aromatic compounds were added into the bacterial culture. Note that when catechol was added into the culture, the OD600 value continued to rise after reaching stationary phase and the color of culture turned to brown (data not shown). This was probably caused by color reaction between catechol and substance in the culture. Subsequent ON-OFF tests using flow cytometer indicated that catechol caused no significant influence on the growth of E.coli.
12
Supplementary Figure S5. Detailed data for the detection spectrum of individual biosensors. The induction capability of 44 archetypical aromatic compounds to NahR (a), XylS (b), HbpR (c), and DmpR (d) biosensors were measured and compared to basal GFP fluorescence to calculate induction ratios. 5-ClSaD, 2,4,6-TClPhl, 2-HBP and 2-ABP were added at 100µM concentration due to severe toxicity at 1000µM concentration. All the other compounds were added at 1000µM concentration.
13
Supplementary Table S5. Acronyms, full names and structures of aromatic compounds used in this study.
Acronym
Full name
BzO
Benzoic acid
Structure
Responsive Biosensor(s)
COOH
XylS
COOH
2-MeBzO
2-Methyl-benzoic acid
XylS, NahR, NahF-R
XylS, NahR, NahF-R
COOH
3-MeBzO
3-Methyl-benzoic acid
COOH
4-MeBzO
2-FBzO
4-Methyl-benzoic acid
2-Fluoro-benzoic acid
XylS
COOH F
XylS
COOH
4-FBzO
F
2-ClBzO
XylS, NahR, NahF-R
4-Fluoro-benzoic acid
2-Chloro-benzoic acid
COOH Cl
XylS
COOH
3-ClBzO
3-Chloro-benzoic acid
XylS, NahR, NahF-R
Cl COOH
4-ClBzO
4-Chloro-benzoic acid
XylS
Cl
2-BrBzO
2-Bromo-benzoic acid
COOH Br
COOH
4-BrBzO
XylS, NahR, NahF-R
4-Bromo-benzoic acid
Br
COOH
3-IBzO
3-Iodo-benzoic acid I
XylS
XylS, NahR, NahF-R
COOH
3-MeOBzO
3-Methoxy-benzoic acid
OCH3
14
NahR, NahF-R
COOH
4-MeOBzO
4-Methoxy-benzoic acid
XylS
OCH3
COOH
4-HBzO
4-Hydroxy-benzoic acid
None
OH
SaA
2-Hydroxy-benzoic acid (Salicylic acid)
2-ABzO
2-Amino-benzoic acid
3-MeSaA
2-Hydroxy-3-methylbenzoic acid
4-MeSaA
4-ClSaA
5-ClSaA
3-HSaA
4-HSaA
COOH OH
COOH NH2
NahR, NahF-R
COOH OH
4-Chloro-2-hydroxybenzoic acid
COOH OH
COOH OH
2,3-Dihydroxybenzoic acid
OH
None
COOH OCOCH3
Aspirin Benzaldehyde
3-Methyl-benzaldehyde
F
2-Fluoro-benzaldehyde
Cl
2-Chloro-benzaldehyde
NahR, NahF-R None
CHO
2-ClBAD
NahR, NahF-R
COOH OH
2,4-Dihydroxybenzoic acid
CHO
2-FBAD
XylS, NahR, NahF-R
Cl
CHO
3-MeBAD
XylS, NahR, NahF-R
Cl
CHO
BAD
XylS, NahR, NahF-R
COOH OH
OH
AsPR
None
COOH OH
2-Hydroxy-4-methylbenzoic acid
5-Chloro-2-hydroxybenzoic acid
XylS, NahR, NahF-R
XylS None None
CHO
3-ClBAD
3-Chloro-benzaldehyde
XylS
Cl CHO
4-ClBAD
4-Chloro-benzaldehyde
None Cl
15
CHO
SaD
2-Hydroxy-benzaldehyde
5-ClSaD
5-Chloro-2-hydroxybenzaldehyde
OH
CHO
OH
Cl
CH2OH
BAL
NahF-R
Phenyl-methanol
NahF-R None
CH2OH
2-MeBAL
o-Tolyl-methanol
None
OH
Phl
Phenol
DmpR
OH
2-MePhl
2-Methyl-phenol
Cat
Benzene-1,2-diol (Catechol)
DmpR
OH OH
DmpR
OH
2-ClPhl
Cl
2-Chloro-phenol
DmpR
OH
2,4,6-TClPhl
2,4,6-Trichloro-phenol
Cl
Cl
NahR, NahF-R
Cl HO
2-HBP
Biphenyl-2-ol
2-ABP
Biphenyl-2-ylamine
TOL
Toluene
H2N
HbpR HbpR None
COOH
PAA
Phenyl-acetic acid
None
COOH
PPA
3-Phenyl-propionic acid
16
None
IV. Cytometry Raw Data of Dose-‐response Tests As we have mentioned in the main text, the four aromatics-responsive biosensors might function as novel inducible gene expression systems. However, to achieve satisfactory induction effect, we should verify that when these biosensors are responding to different concentrations of inducers, the sfGFP expressing population would be the homogeneous. In another word, when the histogram of sfGFP fluorescence of each test sample is plotted from raw flow cytometry data, the population distribution should have a single peak, rather than being bimodal or even multi-modal. To illustrate the point, we hereby display raw FACS data acquired in all dose-response tests conducted on our biosensors (Fig. 1c-f in main text). For clarity, the sfGFP fluorescence histograms of uninduced (0µM), half induced (10µM) and fully induced (100µM) states in each dose-response test were overlaid in one subplot. Results are shown in Supplementary Figure S5. No significant heterogeneity was observed in any of the tests.
17
a
b
e
i
COUNT
j
2-MeBzO 0 30 100
m
n
3-MeOBzO 0 30 100
q
100
f
5-ClSaA 0 10 100
Phl
101
102
103
r
0 30 100
c
3-MeSaA 0 30 100
AsPR 0 10 100
g
k
3-MeBzO 0 10 100
o
3-MeSaA 0 10 100
2-MePhl 0 30 100
104
s
4-MeSaA 0 1 100
3-IBzO 0 30 100
2-ClBzO 0 10 100
2-HBP 0 10 100
2-ClPhl 0 30 100
d
h
l
p
t
4-ClSaA 0 3 100
BzO
0 30 100
3-IBzO 0 10 100
2-ABP 0 30 100
Cat
0 30 100
FITC-A
Supplementary Figure S6. Representative sfGFP fluorescence histograms in dose-response tests. Each subplot represents an individual dose-response test. The histograms corresponding to uninduced (0µM), half induced (10µM) and fully induced (100µM) states are displayed in light gray, dark grey and black, respectively. a-g, Plots for NahR biosensor. h-n, Plots for XylS biosensor. o and p, Plots for HbpR biosensor. q-t, Plots for DmpR biosensor. The aromatic compounds used in each subplot and the inducer concentrations for half induced and fully induced state in each subplot are listed in Supplementary Table S6. Uninduced state means that no inducer was added. FITC-A denotes sfGFP fluorescence.
18
Supplementary Table S6. Information about the subplots of cytometry raw dataa.
Column1
Column2
Sensor Inducer
half
full
Row1
NahR
SaA
3 μM
100 μM NahR
Row2
NahR
5-‐ClSaA
10 μM 100 μM NahR
AsPR
Row3
XylS
2-‐MeBzO
30 μM 100 μM XylS
3-‐MeBzO 10 μM 100 μM
Row4
XylS
3-‐MeOBzO 30 μM 100 μM XylS
3-‐MeSaA 10 μM 100 μM
Row5
DmpR
Phl
2-‐MePhl
Sensor Inducer
30 μM 100 μM DmpR
half
full
3-‐MeSaA 30 μM 100 μM 10 μM 100 μM
30 μM 100 μM
Column3
Column4
Sensor Inducer
half
full
Row1
NahR
SaA
3 μM
100 μM NahR
Row2
NahR
5-‐ClSaA
10 μM 100 μM NahR
AsPR
Row3
XylS
2-‐MeBzO
30 μM 100 μM XylS
3-‐MeBzO 10 μM 100 μM
Row4
XylS
3-‐MeOBzO 30 μM 100 μM XylS
3-‐MeSaA 10 μM 100 μM
Row5
DmpR
Phl
2-‐MePhl
Sensor Inducer
30 μM 100 μM DmpR
a
half
full
3-‐MeSaA 30 μM 100 μM 10 μM 100 μM
30 μM 100 μM
Information is listed according to the positions of the subplots in Supplementary Figure S5.
19
V. Construction and Characterization of NahF-‐R Biosensor Enzyme NahF is a catabolic enzyme from naphthalene degradation pathway in bacterium Pseudomonas putida. It functions as a salicylaldehyde dehydrogenase to transform salicylaldehyde into salicylate using NAD+ as electron receptor. To construct NahF-R biosensor, nahF gene was constitutively expressed under promoter BBa_J23114 (Supplementary Table S3) and co-transformed with the plasmid containing nahR gene and Psal-sfgfp reporter. In addition to the experiments illustrated in the main text, ON-OFF tests were also conducted to determine the detection spectrum of NahF-R biosensor (Fig. S6a). Results indicate that SaD and 5-ClSaD can be degraded by enzyme NahF. Following this result, the dose-response of NahF-R biosensor to 5-ClSaD and 5-ClSaA were measured and compared to NahR biosensor's response to these compounds. Similar phenomena were observed as those demonstrated in the main text (Fig. S6b). It might be expected that since the amount of SaD and 5-ClSaD that was converted into SaA and 5-ClSaA is constrained by the thermodynamic equilibrium of the enzymatic reaction catalyzed by NahF, some efficiency loss will be associated with the conversion, making the NahF-R biosensor less sensitive to SaD and 5-ClSaD compared to SaA and 5-ClSaA. The reduction in sensitivity was indeed observed in the case of conversion from 5-ClSaD to 5-ClSaA (Fig. S7). However, in the case of conversion from SaD to SaA, almost no sensitivity loss was observed (Fig. 3). In fact, the concentration of SaA or 5-ClSaA that NahF-R biosensor responds to is not the total concentration in the liquid culture but rather the intracellular concentration. Taking conversion from SaD to SaA as an example, when the bacteria are added into the liquid culture, the SaD molecules in the medium readily diffuse into the cell. The NahF protein then converts SaD molecules into SaA molecules. The newly formed SaA then starts to diffuse out of the cell into the medium. Since the total volume of intracellular space is negligible compared to the total volume of extracellular medium in an exponential phase bacterial culture, the medium can be regarded as an infinitely large source of SaD molecules as well as an infinitely sink of SaA molecules. In such a condition, the intracellular concentration of SaA will eventually stop changing when the steady state is reached. And it is the steady state concentration of SaA that NahR protein responds to. This steady state, compared to the equilibrium state determined by the Gibbs free energy change of the enzymatic conversion from SaD to SaA, is co-determined by the kinetic properties of the enzymatic conversion process and the trans-membrane diffusion of SaD and SaA molecules and thus involves the interplay of many hidden parameters. In the light of such complexity, we accurately measured NahF-R biosensor's dose-response curves to SaD and 5-ClSaD (Fig. S7b-d). These response curves may serve as a convenient calibration for quantitative measurement of SaD and 5-ClSaD concentrations in practical samples. 20
图表标题
a
1.00
Normalized Induction Effect
0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10
NahF-R NahR
0.00
d
c
5-ClSaD NahR NahF-R
160
120
200
80
40
150
100
NahR
1
10
NahF-R
50
0
0 0
NahF-R 5-ClSaD 5-ClSaA
sfGFP Fluorescence (A. U.)
200
sfGFP Fluorescence (A. U.)
sfGFP Fluorescence (A. U.)
b
100
5-ClSaA NahR NahF-R
200
150
100
50
0 0
Conc. of Inducers (µM)
250
1
10
Conc. of Inducers (µM)
100
0
1
10
100
Conc. of Inducers (µM)
Supplementary Figure S7. Detailed characterization of NahF-R biosensor. (a) NahF-R biosensor's detection spectrum, compared with that of NahR biosensor. The black arrows highlight the compounds that can be detected by NahF-R but not by NahR, due to the fact that these compounds can be degraded by enzyme NahF into compounds that can be sensed by NahF biosensor. (b) NahR and NahF-R biosensors' dose-response curves to 5-ClSaD. (c) NahF-R biosensor's dose-response curves to 5-ClSaD and 5-ClSaA. (d) NahR and NahF-R biosensors' dose-response curves to 5-ClSaA.
21
VI. Test Protocols 1. Protocol for growth tests E. coli Top10 strain was transformed with plasmid Pc114-XylS from XylS biosensor and plasmid Pc-32-sfGFP from HbpR biosensor (see Supplementary Table S8) to mimic intracellular condition of normal biosensor strains while avoiding GFP expression when aromatic compounds were added. The transformed strain was plated onto Luria–Bertani (LB) agar plate and incubated overnight at 37 °C. Single colonies were picked and cultured overnight in fresh LB liquid medium in deep-well 96-well plates. The liquid cultures were then diluted into shallow-well 96-well plates by 100 fold by adding 2µL overnight culture into 200µL fresh LB medium. Forty-four aromatic compounds were then individually added to the diluted cultures to reach 1000 µM final concentration and the diluted cultures were then incubated for 14hrs at 30 °C to reach stationary phase. The OD600 values of cultures were measured every 20 minutes using a micro plate reader (Thermo Scientific Varioskan Flash). For those compounds that inhibit the growth of the transformed strain at 1000 µM concentration, their final concentrations used in the growth tests were reduced to 100µM and the tests were conducted again following the same procedure. 2. Protocol for ON-OFF tests E. coli Top10 strain was transformed with plasmids containing components of target biosensor, plated onto Luria–Bertani (LB) agar plate and incubated overnight at 37 °C. Single colonies were picked and cultured overnight in fresh LB liquid medium in deep-well 96-well plates. The liquid cultures were then diluted into shallow-well 96-well plates by 100 fold by adding 2µL overnight culture into 200µL fresh LB medium. For NahR, XylS and HbpR biosensor, 44 aromatic compounds were then individually added to the diluted cultures to reach 0, 100µM (for 5-ClSaD, 2,4,6-TClPhl, 2-HBP and 2-ABP) or 1000µM (for all the other 40 compounds) final concentration and the diluted cultures were then incubated for 12hrs at 30 °C to reach stationary phase. For DmpR biosensor, the diluted cultures were first incubated for 6hrs to reach late exponential/ early stationary phase. The cultures were then centrifuged at 4000rpm for 10 minutes. The supernatants were discarded and 200µL fresh LB medium was added to each well to re-suspend the pelleted culture. Then 44 aromatic compounds were then individually added to the diluted cultures to reach 0, 100µM (for 5-ClSaD, 2,4,6-TClPhl, 2-HBP and 2-ABP) or 1000µM (for all the other 40 compounds) final concentration and the cultures were further incubated for 4hrs. Then the cultures were diluted 200 fold by adding 1µL stationary phase culture into 200µL Phosphate Buffered Solution (PBS) containing 2mg/ml Kanamycin antibiotic to stop cell growth and protein synthesis. Then sfGFP fluorescence was measured using flow cytometer (BD LSRII). 22
3. Protocol for dose-response tests E. coli Top10 strain was transformed with plasmids containing components of target biosensor, plated onto Luria–Bertani (LB) agar plate and incubated overnight at 37 °C. Single colonies were picked and cultured overnight in fresh LB liquid medium in deep-well 96-well plates. The liquid cultures were then diluted into shallow-well 96-well plates by 100 fold by adding 2µL overnight culture into 200µL fresh LB medium. For NahR, XylS and HbpR biosensor, the target aromatic compound was then added to the diluted cultures to reach 0, 1, 3, 10, 30 and 100µM final concentrations (for some compounds, 0.1, 0.3 and 300µM final concentrations were also tested) and the diluted cultures were then incubated for 12hrs at 30 °C to reach stationary phase. For DmpR biosensor, the diluted cultures were first incubated for 6hrs to reach late exponential/ early stationary phase. The cultures were then centrifuged at 4000rpm for 10 minutes. The supernatants were discarded and 200µL fresh LB medium was added to each well to re-suspend the pelleted culture. Then the target aromatic compound was then added to the diluted cultures to reach 0, 1, 3, 10, 30 and 100µM final concentrations (for some compounds, 0.1, 0.3 and 300µM final concentrations were also tested) and the cultures were further incubated for 4hrs. Then the cultures were diluted 200 fold by adding 1µL stationary phase culture into 200µL Phosphate Buffered Solution (PBS) containing 2mg/ml Kanamycin antibiotic to stop cell growth and protein synthesis. Then sfGFP fluorescence was measured using flow cytometer (BD LSRII). 4. Protocol for interference tests E. coli Top10 strain was transformed with plasmids containing components of target biosensor, plated onto Luria–Bertani (LB) agar plate and incubated overnight at 37 °C. Single colonies were picked and cultured overnight in fresh LB liquid medium in deep-well 96-well plates. The liquid cultures were then diluted into shallow-well 96-well plates by 100 fold by adding 2µL overnight culture into 200µL fresh LB medium. For NahR, XylS and HbpR biosensor, the typical inducer was first added to the diluted cultures to reach 0, 1, 3, 10, 30 and 100µM final concentration. The interfering inducer was then added to the diluted culture to reach 0, 10, 30 and 100µM final concentration. The diluted cultures were then incubated for 12hrs at 30 °C to reach stationary phase. For DmpR biosensor, the diluted cultures were first incubated for 6hrs to reach late exponential/ early stationary phase. The cultures were then centrifuged at 4000rpm for 10 minutes. The supernatants were discarded and 200µL fresh LB medium was added to each well to re-suspend the pelleted culture. After this, the typical inducer was first added to the diluted cultures to reach 0, 1, 3, 10, 30 and 100µM final concentration. The interfering inducer was then added to the diluted culture to reach 0, 10, 30 and 100µM final concentration. The cultures were further incubated for 4hrs. 23
Then the cultures were diluted 200 fold by adding 1µL stationary phase culture into 200µL Phosphate Buffered Solution (PBS) containing 2mg/ml Kanamycin antibiotic to stop cell growth and protein synthesis. Then sfGFP fluorescence was measured using flow cytometer (BD LSRII).
24
VII. Parts, Plasmids and Strains Supplementary Table S7. Parts used in this study. Source & Description a BBa_B0015, double terminator used throughout the study BBa_J61051, terminator + coding sequence of NahR transcription factor + Psal promoter xylS BBa_K1031911, coding sequence of XylS transcription factor + terminator hbpR BBa_K1031300, coding sequence of HbpR transcription factor + terminator dmpR BBa_K1031211, constitutive promoter Pr + dmpR coding sequence BBa_J23113 constitutive promoter BBa_J23109 constitutive promoter BBa_J23117 constitutive promoter BBa_J23114 constitutive promoter BBa_J23115 constitutive promoter BBa_J23106 constitutive promoter sfGFP coding sequence of sfGFP Pm sequence of Pm promoter Pc sequence of Pc promoter Po sequence of Po promoter nahF coding sequence of nahF + terminator pSB1A3 High copy plasmid vector (100~300 copies per cell); replication origin: pUC19-derived pMB1; antibiotic resistance: ampicillin pSB4K5 Low copy plasmid vector (~5 copies per cell); replication origin: pSC101; antibiotic resistance: kanamycin pSB1C3 High copy plasmid vector (100~300 copies per cell); replication origin: pUC19-derived pMB1; antibiotic resistance: chloramphenicol. pV1398 pMMB66HE-based high copy cloning vector; antibiotic resistance: Ampicillin. BBa_B0034 RBS BBa_B0031 RBS BBa_B0032 RBS a. Biological parts whose names start with “BBa” or “pSB” are from the Registry of Standard Biological Parts, where the sequences and features of all parts can be obtained. Name Terminator nahR
25
Supplementary Tabel S8. Plasmids used in this study. Name Components of Insert a Backbone NahR-sfGFP BBa_J61051 + sfGFP + Terminator pSB1C3 Pc113-XylS BBa_J23113 + BBa_K1031911 + Terminator pSB4K5 Pc109-XylS BBa_J23109 + BBa_K1031911 + Terminator pSB4K5 Pc114-XylS BBa_J23114 + BBa_K1031911 + Terminator pSB4K5 Pc105-XylS BBa_J23105 + BBa_K1031911 + Terminator pSB4K5 Pc106-XylS BBa_J23106 + BBa_K1031911 + Terminator pSB4K5 Pc113-HbpR BBa_J23113 + BBa_K1031300 + Terminator pSB4K5 Pc109-HbpR BBa_J23109 + BBa_K1031300 + Terminator pSB4K5 Pc117-HbpR BBa_J23117 + BBa_K1031300 + Terminator pSB4K5 Pc114-HbpR BBa_J23114 + BBa_K1031300 + Terminator pSB4K5 Pc106-HbpR BBa_J23106 + BBa_K1031300 + Terminator pSB4K5 Pr-DmpR BBa_K1031211 + Terminator pV1398 Pm-eGFP Pm + BBa_B0034 + eGFP + Terminator pSB1A3 Pc-31-sfGFP Pc + BBa_B0031 + sfGFP + Terminator pSB1C3 Pc-32-sfGFP Pc + BBa_B0032 + sfGFP + Terminator pSB1C3 Pc-34-sfGFP Pc + BBa_B0034 + sfGFP + Terminator pSB1C3 Po-sfGFP Po + BBa_B0032 + sfGFP + Terminator pSB1C3 Pc114-NahF BBa_J23114 + BBa_B0034 + nahF pSB4K5 a. Listed are inserted components of the plasmids. The components can be traced back to the parts listed in Supplementary Table S7. Plasmid's backbone have been listed in another column. “+” means two biological parts were assembled using standard assembly.
26
Supplementary Table S9. Strains used in this study. Name Plasmid(s) transformed a NahR Biosensor Strain NahR-sfGFP one plasmid among Pc113-XylS, Pc113-XylS, XylS Biosensor Strain Pc113-XylS, Pc113-XylS and Pc106-XylS + Pm-eGFP one plasmid among Pc113-HbpR, Pc109-HbpR, Pc117-HbpR, Pc114-HbpR and Pc106-HbpR + one HbpR Biosensor Strain plasmid among Pc-31-sfGFP, Pc-32-sfGFP and Pc-34-sfGFP DmpR Biosensor Strain Pr-DmpR + Po-sfGFP NahF-R Biosensor Strain Pc114-NahF + NahR-sfGFP a. The plasmids listed can be traced back to the plasmids listed in Supplementary Table S8. "+" means that the plasmids listed are co-transformed into one strain.
27
VIII. Supplementary References 1
2
3
4
5
Cases, I. & de Lorenzo, V. Promoters in the environment: transcriptional regulation in its natural context. Nature reviews. Microbiology 3, 105-‐118, doi:10.1038/nrmicro1084 (2005). Lau, P. C. et al. A bacterial basic region leucine zipper histidine kinase regulating toluene degradation. Proceedings of the National Academy of Sciences of the United States of America 94, 1453-‐1458 (1997). Busch, A., Lacal, J., Martos, A., Ramos, J. L. & Krell, T. Bacterial sensor kinase TodS interacts with agonistic and antagonistic signals. Proceedings of the National Academy of Sciences of the United States of America 104, 13774-‐13779, doi:10.1073/pnas.0701547104 (2007). Diaz, E., Ferrandez, A., Prieto, M. A. & Garcia, J. L. Biodegradation of aromatic compounds by Escherichia coli. Microbiology and molecular biology reviews : MMBR 65, 523-‐569, table of contents, doi:10.1128/MMBR.65.4.523-‐569.2001 (2001). Manso, I. et al. 3-‐Hydroxyphenylpropionate and phenylpropionate are synergistic activators of the MhpR transcriptional regulator from Escherichia coli. The Journal of biological chemistry 284, 21218-‐21228, doi:10.1074/jbc.M109.008243 (2009).
28
Peking University Team for the International Genetically Engineered Machine Competition (iGEM), 2Center for Quantitative Biology, and 3Peking-Tsinghua Joint Center for Life Sciences, Peking University, Beijing, 100871, China. ║
These authors contribute equally to this work.
*Correspondence should [email protected]
be
addressed
1
to
[email protected]
and
I. Genetic and material sources of aromatics-‐responsive transcription factors NahR transcription activator is encoded by nahR gene that locates on NAH7 plasmid from the bacterium Pseudomonas putida. NAH7 plasmid encodes a regulator (NahR) and several enzymes (including NahF) that compose the nah operon. NahR responds to salicylate (SaA), an intermediate in naphthalene catabolic pathway, and activates the expression of the catabolic enzymes in nah operon (Fig. S1a). XylS transcription activator is encoded by xylS gene that locates on TOL plasmid pWW0 from the bacterium Pseudomonas putida. pWW0 plasmid encodes two regulators (XylS and XylR) and several enzymes that compose the xylene catabolic pathway. XylS responds to benzoate (BzO), an intermediate in toluene catabolic pathway, and activates the expression of the catabolic enzymes (Fig. S1b). HbpR transcription activator is encoded by hbpR gene from the bacterium Pseudomonas azelaica. HbpR is the master regulator of hbp operon responsible for 2-hydroxybiphenyl (2-HBP) degradation. HbpR responds to 2-HBP and activates the expression of the catabolic enzymes in hbp operon (Fig. S1c). DmpR transcription activator is encoded by dmpR gene from the bacterium Pseudomonas strain CF600. DmpR is the master regulator of dmp operon responsible for phenol (Phl) degradation. DmpR responds to Phl and activates the expression of the catabolic enzymes in dmp operon (Fig. S1d). The names and sequences of the promoters regulated by these four transcription factors are listed in Supplementary Table S1. The material sources of the coding sequences of these transcription factors used in this study are listed in Supplementary Table S2.
2
Supplementary Figure S1. Genetic sources of the aromatics-responsive transcription factors used in this study. (a) Genetic components of naphthalene catabolic pathway. NahR responds to salicylate and activates Pnah promoter, which drives the upstream metabolic pathway, and Psal promoter, which drives the downstream metabolic pathway. (b) Genetic components of benzoate catabolic pathway, which is a truncated version of xylene catabolic pathway. XylS responds to benzoate and activates Pm promoter. (c) Genetic components of 2-hydroxybiphenyl catabolic pathway. HbpR responds to biphenyl derivatives and activates PD and Pc promoters. (d) Genetic components of phenol catabolic pathway. DmpR responds to phenol derivatives and activates Po promoter.
3
Supplementary Table S1. Sequences of transcription factors and cognate promoters used in this study Transcription Factor
nahR
xylS
hbpR
dmpR
Cognate Promoter
Sequence of Cognate Promoter
Psal
TATTTATCAATATTGTTTGCTCCGTTATCGTT ATTAACAAGTCATCAATAAAGCCATCACGAGT ACCATAG
Pm
AATCCTTAGCTTTCGCTAAGGATGATTTTTGG ATTTCGCGGCCGCTACTAGAAAGGCCTACCCC TTAGGCTTTATGCAACTAGAGATT
Pc
TTTAATAAATTTATGAAATCGTGGTTGTGAGT TTTCATAATATGGTGAAGCTTGCCCGCCATGG CAAGTGCATTTCGGCAGCTGTTGGCGACCGCG GAAGGGGTTTACAGCCGTCTGGAGCTAGGCTT TCTGGCGCTCATTAAAATAAAAATCCTTATAA AACAGTATCCTAGCTTTTATGTCTGAGGCTGC TTAGTCAACCTGGCACGGTACTGGCTACGAGT CCCGC
Po
TAAGCATTTGCTCAAGCGGCCTTGGGCAATTG ATCAAATGCTTAAAAAGTCTGCGCAAGCGCGG CTTAATTTCGCTCGCTCCGATCATTCTAAAAA TTAGAAACACATTGAAAAACATTACCTTGAAG TCTGTTTTCAGACCTTGGCACAGCCGTTGCTT GATGTCCTGCG
4
Supplementary Table S2. Material sources of aromatics-responsive transcription factor genes Gene nahR xylS hbpR dmpR
Material Source of Coding Sequence Standard Biological Brick BBa_J61051 from Registry of Standard Biological Parts Gene synthesis Plasmid from Professor Jan Roelof van der Meer, Department of Fundamental Microbiology, University of Lausanne Plasmid from Professor Victoria Shingler, Department of Molecular Biology, Umeå University
5
II. Optimization of XylS and HbpR Biosensors As described in the main text, the primary genetic constructs for XylS and HbpR biosensors failed to exhibit desirable performance. The maximum induction folds for both of the biosensors were very low. This was largely due to exceedingly high sfGFP basal expressions when no inducer was added. We speculated that the high basal expression level might be attributed to two possible reasons: First, aromatics-responsive transcription factors might be excessively expressed, leading to activation of the TFs’ cognate promoters even without inducers. Second, translation strength of sfGFP reporter gene might be too strong, leading to high basal expression even when sfGFP's promoter is not activated. Based on these two speculations, we used a family of constitutive promoters from Registry of Standard Biological Parts to tune transcription factors' expression and used a family of ribosome binding sites (RBSs) to tune sfGFP expression. Constitutive promoter BBa_J23106 was chosen to drive the expression of hbpR gene in the primary construct. To tune the expression of hbpR gene, four constitutive promoters with increasing transcriptional strength, BBa_J23113, BBa_J23109, BBa_J23117 and BBa_J23114, were chosen. All four constitutive promoters are weaker than the original promoter BBa_J23106. Results of ON-OFF tests showed that promoter BBa_J23114 appeared to be the optimal choice (Fig. S2a). The performance of HbpR biosensor is not yet satisfying after the first round of optimization. So we conducted a second round of optimization. Three RBSs with increasing strength, BBa_B0032, BBa_B0031 and BBa_B034, were chosen from Registry of Standard Biological Parts and used to tune sfGFP expression. Results of dose-response curve tests showed that BBa_B0032 was the optimal choice (Fig. S2b). After two rounds of optimization, HbpR biosensor acquired a maximal induction response of over 50 fold. For XylS biosensor, BBa_J23106 was also chosen to drive the expression of xylS gene in the primary construct. As in the case of HbpR biosensor, four constitutive promoters with increasing transcriptional strength, BBa_J23113, BBa_J23109, BBa_J23114 and BBa_J23105, were chosen. Results also showed that BBa_J23114 was the optimal choice (Fig. S3). At this point, XylS biosensor already acquired a maximal induction response of over 100 fold, so the second round of optimization was not conducted. Sequences and relative strengths of the promoters and RBSs used in optimization of HbpR and XylS biosensor was listed in Supplementary Table S3 and S4.
6
Supplementary Figure S2. Optimization of HbpR biosensor. (a) Performance of HbpR biosensor variants with different constitutive promoters controlling hbpR gene. The HbpR biosensor with its hbpR gene controlled by J23114 promoter showed highest induction ration towards both 2-HBP and 2-ABP. (b) Performance of different versions of HbpR biosensor with different RBSs driving sfGFP expression. The constitutive promoter controlling hbpR gene was fixed as J23114 herein. HbpR biosensor with RBS B0032 driving the translation of sfgfp performed the best.
Supplementary Figure S3. Optimization of XylS biosensor. xylS gene was controlled by different constitutive promoters to create different versions of XylS biosensor. Biosensor using J23114 promoter showed highest induction ratio towards 7
BzO, 2-MeBzO, 3-MeBzO and 4-MeBzO. Supplementary Table S3. Sequences and relative promoter strengths of the constitutive promoters used in this study.
Categorized NO.a
Sequence
Relative Promoter Strength b
BBa_J23113
ctgatggctagctcagtcctagggattatgctagc
21
BBa_J23109
tttacagctagctcagtcctagggactgtgctagc
106
BBa_ J23117
ttgacagctagctcagtcctagggattgtgctagc
162
BBa_J23114
tttatggctagctcagtcctaggtacaatgctagc
256
BBa_J23105
tttacggctagctcagtcctaggtactatgctagc
623
BBa_J23106
tttacggctagctcagtcctaggtatagtgctagc
1185
a. The categorized number in Registry of Standard Biological Parts. http://partsregistry.org/Main_Page b. Relative promoter strength adopted from Registry of Standard Biological Parts. http://partsregistry.org/Main_Page
8
Supplementary Table S4. Sequences and relative strengths of the RBSs used in this study. Categorized NO. a
Sequence
Relative RBS Strength b
BBa_B0034
aaagaggagaaa
31121
BBa_B0031
tcacacaggaaacc
3588
BBa_B0032
tcacacaggaaag
2271
a. The categorized number in Registry of Standard Biological Parts. http://partsregistry.org/Main_Page b. The relative RBS strength was predicted using online RBS calculator v1.1: “Reverse Engineer RBS” (https://salis.psu.edu/software/reverse).
9
III. Detection Spectrum of Individual Biosensors & the Interferences on Sensing Capability. III.A. Detection Spectrums Chemical structures and full names of the aromatic compounds discussed below can be found in Supplementary Table S5. In this research, we selected 44 aromatic compounds to comprehensively represent the common aromatics present in the environment either naturally or due to industrial causes. To characterize the biosensors' response to these compounds, first we need to verify that the presence of these compounds exerts no significant toxic effect on the normal cellular processes of E. coli cells. To evaluate such possible toxicity, growth tests were conducted for each of the 44 aromatic compounds using Top10 E. coli strain (Fig. S4). Results showed that, apart from 4 aromatic compounds (5-ClSaD, 2,4,6-TClPhl, 2-HBP and 2-ABP), all the other 40 compounds exert no significant influence on E. coli cells' growth at 1000µM concentration. When the concentrations of the four severely toxic compounds were reduced to 100µM, their toxicity towards E. coli cells was no longer significant (Fig. S4). For each biosensor, ON-OFF test for each of the 44 aromatic compounds was then conducted to identify those compounds that can elicit significant responses. Those aromatic compounds that showed severe toxic effect at 1000µM concentration were added at 100µM concentration. All the other compounds were added at 1000µM concentration (Fig. S5). For NahR biosensor, 16 of the 44 compounds showed significant activation effects with induction folds higher than 20. They are listed as follows: SaA, 3-MeSaA, 4-MeSaA, 4-ClSaA, 5-ClSaA, AsPR, 2,4,6-TClPhl, 2-MeBzO, 3-MeBzO, 4-FBzO, 3-ClBzO, 4-ClBzO, 4-BrBzO, 3-IBzO, 3-MeOBzO and 3-HSaA. These compounds can be classified into salicylate and its derivatives, benzoate and its derivatives and a special compound 2,4,6-TClPhl. 2,4,6-TClPhl is a kind of polychlorinated phenol that is particularly hazardous to water environment and human health. Such a sensing capability of NahR biosensor has not been reported in previous researches. For XylS biosensor, 19 of the 44 compounds showed significant activation effects with induction folds higher than 20. They are listed as follows: BzO, 2-MeBzO, 3-MeBzO, 4-MeBzO, 2-FBzO, 4-FBzO, 2-ClBzO, 3-ClBzO, 4-ClBzO, 2-BrBzO, 4-BrBzO, 3-IBzO, 3-MeOBzO, SaA, 3-MeSaA, 4-ClSaA, 5-ClSaA, 3-MeBAD and 3-ClBAD. These compounds can be roughly classified into Benzoate and its derivatives and 10
Salicylate and its derivatives, with the exception of 3-MeBAD and 3-ClTOL. As illustrated in the main text, HbpR biosensor responds specifically to 2-HBP and 2-ABP. For DmpR biosensor, 4 of the 44 compounds showed significant activation effect with induction folds greater than 5. They are listed as follows: Phl, 2-MePhl, 2-ClPhl and Cat. III.B. Interferences Caused by Non-inducer Chemicals on Sensing Capability Previous studies reported that, for a given biosensor, when aromatic chemicals that are able or unable to induce significant response (typical inducers and non-typical inducers, respectively) co-exist, non-typical inducers might interfere with the response of the biosensor to other inducers, although the non-typical inducers themselves could hardly induce any significant response1 (Fig. 2a). For example, in Pseudomonas putida DOT-T1E, the TodS/TodT two-component system senses the presence of toluene to control expression of toluene metabolism pathway2. Despite the fact that toluene is a strong inducer for TodS (the histidine kinase), toluene derivatives with ortho-substitutions would reduce or totally abolish the in vivo responses of TodS through competing with toluene for binding to TodS3. Another example is much more similar to the cases of biosensors in this manuscript. In Escherichia coli K-12, the mhp operon encodes the catabolic pathways for the degradation of the aromatic compound 3-hydroxyphenylpropionate (3-HPP). The regulator MhpR is able to sense both 3-HPP, the substrate, and 3-(2,3-dihydroxyphenyl) propionate (DHPP), the catabolic product, but unable to sense henylpropionate (PP)4. However, PP is able to activate the MhpR regulator synergistically with the 3-HPP and DHPP; moreover, 3-HPP, DHPP, and PP bind independently to MhpR with similar affinities5. This is probably because that PP is the substrate of the metabolic pathway upstream of mhp pathway, and MhpR integrates the signals of 3-HPP, DHPP and PP, thus to rapidly modulate the transcription of mhp operon. These interferences revealed in previously studies strongly implied that some aromatic chemicals, for instance, the derivatives of the strong inducers, would probably influence the sensing capability of our biosensors, although they couldn’t induce any response by themselves alone. These interferences should be carefully assessed and taken into consideration when using our biosensors.
11
Supplementary Figure S4. The influence of aromatic compounds on the growth of E. coli. The final concentrations of aromatic compounds in the liquid culture were 1000 µM unless otherwise indicated. (a) Growth curves of E. coli Top10 strain when benzoate derivatives were added into the bacterial culture. (b) Growth curves of E. coli Top10 strain when salicylate derivatives were added into the bacterial culture. (c) Growth curves of E. coli Top10 strain when benzaldehyde derivatives, salicylaldehyde derivatives and benzyl alcohol derivatives were added into the bacterial culture. (d) Growth curves of E. coli Top10 strain when phenol derivatives, biphenyl derivatives and other common aromatic compounds were added into the bacterial culture. Note that when catechol was added into the culture, the OD600 value continued to rise after reaching stationary phase and the color of culture turned to brown (data not shown). This was probably caused by color reaction between catechol and substance in the culture. Subsequent ON-OFF tests using flow cytometer indicated that catechol caused no significant influence on the growth of E.coli.
12
Supplementary Figure S5. Detailed data for the detection spectrum of individual biosensors. The induction capability of 44 archetypical aromatic compounds to NahR (a), XylS (b), HbpR (c), and DmpR (d) biosensors were measured and compared to basal GFP fluorescence to calculate induction ratios. 5-ClSaD, 2,4,6-TClPhl, 2-HBP and 2-ABP were added at 100µM concentration due to severe toxicity at 1000µM concentration. All the other compounds were added at 1000µM concentration.
13
Supplementary Table S5. Acronyms, full names and structures of aromatic compounds used in this study.
Acronym
Full name
BzO
Benzoic acid
Structure
Responsive Biosensor(s)
COOH
XylS
COOH
2-MeBzO
2-Methyl-benzoic acid
XylS, NahR, NahF-R
XylS, NahR, NahF-R
COOH
3-MeBzO
3-Methyl-benzoic acid
COOH
4-MeBzO
2-FBzO
4-Methyl-benzoic acid
2-Fluoro-benzoic acid
XylS
COOH F
XylS
COOH
4-FBzO
F
2-ClBzO
XylS, NahR, NahF-R
4-Fluoro-benzoic acid
2-Chloro-benzoic acid
COOH Cl
XylS
COOH
3-ClBzO
3-Chloro-benzoic acid
XylS, NahR, NahF-R
Cl COOH
4-ClBzO
4-Chloro-benzoic acid
XylS
Cl
2-BrBzO
2-Bromo-benzoic acid
COOH Br
COOH
4-BrBzO
XylS, NahR, NahF-R
4-Bromo-benzoic acid
Br
COOH
3-IBzO
3-Iodo-benzoic acid I
XylS
XylS, NahR, NahF-R
COOH
3-MeOBzO
3-Methoxy-benzoic acid
OCH3
14
NahR, NahF-R
COOH
4-MeOBzO
4-Methoxy-benzoic acid
XylS
OCH3
COOH
4-HBzO
4-Hydroxy-benzoic acid
None
OH
SaA
2-Hydroxy-benzoic acid (Salicylic acid)
2-ABzO
2-Amino-benzoic acid
3-MeSaA
2-Hydroxy-3-methylbenzoic acid
4-MeSaA
4-ClSaA
5-ClSaA
3-HSaA
4-HSaA
COOH OH
COOH NH2
NahR, NahF-R
COOH OH
4-Chloro-2-hydroxybenzoic acid
COOH OH
COOH OH
2,3-Dihydroxybenzoic acid
OH
None
COOH OCOCH3
Aspirin Benzaldehyde
3-Methyl-benzaldehyde
F
2-Fluoro-benzaldehyde
Cl
2-Chloro-benzaldehyde
NahR, NahF-R None
CHO
2-ClBAD
NahR, NahF-R
COOH OH
2,4-Dihydroxybenzoic acid
CHO
2-FBAD
XylS, NahR, NahF-R
Cl
CHO
3-MeBAD
XylS, NahR, NahF-R
Cl
CHO
BAD
XylS, NahR, NahF-R
COOH OH
OH
AsPR
None
COOH OH
2-Hydroxy-4-methylbenzoic acid
5-Chloro-2-hydroxybenzoic acid
XylS, NahR, NahF-R
XylS None None
CHO
3-ClBAD
3-Chloro-benzaldehyde
XylS
Cl CHO
4-ClBAD
4-Chloro-benzaldehyde
None Cl
15
CHO
SaD
2-Hydroxy-benzaldehyde
5-ClSaD
5-Chloro-2-hydroxybenzaldehyde
OH
CHO
OH
Cl
CH2OH
BAL
NahF-R
Phenyl-methanol
NahF-R None
CH2OH
2-MeBAL
o-Tolyl-methanol
None
OH
Phl
Phenol
DmpR
OH
2-MePhl
2-Methyl-phenol
Cat
Benzene-1,2-diol (Catechol)
DmpR
OH OH
DmpR
OH
2-ClPhl
Cl
2-Chloro-phenol
DmpR
OH
2,4,6-TClPhl
2,4,6-Trichloro-phenol
Cl
Cl
NahR, NahF-R
Cl HO
2-HBP
Biphenyl-2-ol
2-ABP
Biphenyl-2-ylamine
TOL
Toluene
H2N
HbpR HbpR None
COOH
PAA
Phenyl-acetic acid
None
COOH
PPA
3-Phenyl-propionic acid
16
None
IV. Cytometry Raw Data of Dose-‐response Tests As we have mentioned in the main text, the four aromatics-responsive biosensors might function as novel inducible gene expression systems. However, to achieve satisfactory induction effect, we should verify that when these biosensors are responding to different concentrations of inducers, the sfGFP expressing population would be the homogeneous. In another word, when the histogram of sfGFP fluorescence of each test sample is plotted from raw flow cytometry data, the population distribution should have a single peak, rather than being bimodal or even multi-modal. To illustrate the point, we hereby display raw FACS data acquired in all dose-response tests conducted on our biosensors (Fig. 1c-f in main text). For clarity, the sfGFP fluorescence histograms of uninduced (0µM), half induced (10µM) and fully induced (100µM) states in each dose-response test were overlaid in one subplot. Results are shown in Supplementary Figure S5. No significant heterogeneity was observed in any of the tests.
17
a
b
e
i
COUNT
j
2-MeBzO 0 30 100
m
n
3-MeOBzO 0 30 100
q
100
f
5-ClSaA 0 10 100
Phl
101
102
103
r
0 30 100
c
3-MeSaA 0 30 100
AsPR 0 10 100
g
k
3-MeBzO 0 10 100
o
3-MeSaA 0 10 100
2-MePhl 0 30 100
104
s
4-MeSaA 0 1 100
3-IBzO 0 30 100
2-ClBzO 0 10 100
2-HBP 0 10 100
2-ClPhl 0 30 100
d
h
l
p
t
4-ClSaA 0 3 100
BzO
0 30 100
3-IBzO 0 10 100
2-ABP 0 30 100
Cat
0 30 100
FITC-A
Supplementary Figure S6. Representative sfGFP fluorescence histograms in dose-response tests. Each subplot represents an individual dose-response test. The histograms corresponding to uninduced (0µM), half induced (10µM) and fully induced (100µM) states are displayed in light gray, dark grey and black, respectively. a-g, Plots for NahR biosensor. h-n, Plots for XylS biosensor. o and p, Plots for HbpR biosensor. q-t, Plots for DmpR biosensor. The aromatic compounds used in each subplot and the inducer concentrations for half induced and fully induced state in each subplot are listed in Supplementary Table S6. Uninduced state means that no inducer was added. FITC-A denotes sfGFP fluorescence.
18
Supplementary Table S6. Information about the subplots of cytometry raw dataa.
Column1
Column2
Sensor Inducer
half
full
Row1
NahR
SaA
3 μM
100 μM NahR
Row2
NahR
5-‐ClSaA
10 μM 100 μM NahR
AsPR
Row3
XylS
2-‐MeBzO
30 μM 100 μM XylS
3-‐MeBzO 10 μM 100 μM
Row4
XylS
3-‐MeOBzO 30 μM 100 μM XylS
3-‐MeSaA 10 μM 100 μM
Row5
DmpR
Phl
2-‐MePhl
Sensor Inducer
30 μM 100 μM DmpR
half
full
3-‐MeSaA 30 μM 100 μM 10 μM 100 μM
30 μM 100 μM
Column3
Column4
Sensor Inducer
half
full
Row1
NahR
SaA
3 μM
100 μM NahR
Row2
NahR
5-‐ClSaA
10 μM 100 μM NahR
AsPR
Row3
XylS
2-‐MeBzO
30 μM 100 μM XylS
3-‐MeBzO 10 μM 100 μM
Row4
XylS
3-‐MeOBzO 30 μM 100 μM XylS
3-‐MeSaA 10 μM 100 μM
Row5
DmpR
Phl
2-‐MePhl
Sensor Inducer
30 μM 100 μM DmpR
a
half
full
3-‐MeSaA 30 μM 100 μM 10 μM 100 μM
30 μM 100 μM
Information is listed according to the positions of the subplots in Supplementary Figure S5.
19
V. Construction and Characterization of NahF-‐R Biosensor Enzyme NahF is a catabolic enzyme from naphthalene degradation pathway in bacterium Pseudomonas putida. It functions as a salicylaldehyde dehydrogenase to transform salicylaldehyde into salicylate using NAD+ as electron receptor. To construct NahF-R biosensor, nahF gene was constitutively expressed under promoter BBa_J23114 (Supplementary Table S3) and co-transformed with the plasmid containing nahR gene and Psal-sfgfp reporter. In addition to the experiments illustrated in the main text, ON-OFF tests were also conducted to determine the detection spectrum of NahF-R biosensor (Fig. S6a). Results indicate that SaD and 5-ClSaD can be degraded by enzyme NahF. Following this result, the dose-response of NahF-R biosensor to 5-ClSaD and 5-ClSaA were measured and compared to NahR biosensor's response to these compounds. Similar phenomena were observed as those demonstrated in the main text (Fig. S6b). It might be expected that since the amount of SaD and 5-ClSaD that was converted into SaA and 5-ClSaA is constrained by the thermodynamic equilibrium of the enzymatic reaction catalyzed by NahF, some efficiency loss will be associated with the conversion, making the NahF-R biosensor less sensitive to SaD and 5-ClSaD compared to SaA and 5-ClSaA. The reduction in sensitivity was indeed observed in the case of conversion from 5-ClSaD to 5-ClSaA (Fig. S7). However, in the case of conversion from SaD to SaA, almost no sensitivity loss was observed (Fig. 3). In fact, the concentration of SaA or 5-ClSaA that NahF-R biosensor responds to is not the total concentration in the liquid culture but rather the intracellular concentration. Taking conversion from SaD to SaA as an example, when the bacteria are added into the liquid culture, the SaD molecules in the medium readily diffuse into the cell. The NahF protein then converts SaD molecules into SaA molecules. The newly formed SaA then starts to diffuse out of the cell into the medium. Since the total volume of intracellular space is negligible compared to the total volume of extracellular medium in an exponential phase bacterial culture, the medium can be regarded as an infinitely large source of SaD molecules as well as an infinitely sink of SaA molecules. In such a condition, the intracellular concentration of SaA will eventually stop changing when the steady state is reached. And it is the steady state concentration of SaA that NahR protein responds to. This steady state, compared to the equilibrium state determined by the Gibbs free energy change of the enzymatic conversion from SaD to SaA, is co-determined by the kinetic properties of the enzymatic conversion process and the trans-membrane diffusion of SaD and SaA molecules and thus involves the interplay of many hidden parameters. In the light of such complexity, we accurately measured NahF-R biosensor's dose-response curves to SaD and 5-ClSaD (Fig. S7b-d). These response curves may serve as a convenient calibration for quantitative measurement of SaD and 5-ClSaD concentrations in practical samples. 20
图表标题
a
1.00
Normalized Induction Effect
0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10
NahF-R NahR
0.00
d
c
5-ClSaD NahR NahF-R
160
120
200
80
40
150
100
NahR
1
10
NahF-R
50
0
0 0
NahF-R 5-ClSaD 5-ClSaA
sfGFP Fluorescence (A. U.)
200
sfGFP Fluorescence (A. U.)
sfGFP Fluorescence (A. U.)
b
100
5-ClSaA NahR NahF-R
200
150
100
50
0 0
Conc. of Inducers (µM)
250
1
10
Conc. of Inducers (µM)
100
0
1
10
100
Conc. of Inducers (µM)
Supplementary Figure S7. Detailed characterization of NahF-R biosensor. (a) NahF-R biosensor's detection spectrum, compared with that of NahR biosensor. The black arrows highlight the compounds that can be detected by NahF-R but not by NahR, due to the fact that these compounds can be degraded by enzyme NahF into compounds that can be sensed by NahF biosensor. (b) NahR and NahF-R biosensors' dose-response curves to 5-ClSaD. (c) NahF-R biosensor's dose-response curves to 5-ClSaD and 5-ClSaA. (d) NahR and NahF-R biosensors' dose-response curves to 5-ClSaA.
21
VI. Test Protocols 1. Protocol for growth tests E. coli Top10 strain was transformed with plasmid Pc114-XylS from XylS biosensor and plasmid Pc-32-sfGFP from HbpR biosensor (see Supplementary Table S8) to mimic intracellular condition of normal biosensor strains while avoiding GFP expression when aromatic compounds were added. The transformed strain was plated onto Luria–Bertani (LB) agar plate and incubated overnight at 37 °C. Single colonies were picked and cultured overnight in fresh LB liquid medium in deep-well 96-well plates. The liquid cultures were then diluted into shallow-well 96-well plates by 100 fold by adding 2µL overnight culture into 200µL fresh LB medium. Forty-four aromatic compounds were then individually added to the diluted cultures to reach 1000 µM final concentration and the diluted cultures were then incubated for 14hrs at 30 °C to reach stationary phase. The OD600 values of cultures were measured every 20 minutes using a micro plate reader (Thermo Scientific Varioskan Flash). For those compounds that inhibit the growth of the transformed strain at 1000 µM concentration, their final concentrations used in the growth tests were reduced to 100µM and the tests were conducted again following the same procedure. 2. Protocol for ON-OFF tests E. coli Top10 strain was transformed with plasmids containing components of target biosensor, plated onto Luria–Bertani (LB) agar plate and incubated overnight at 37 °C. Single colonies were picked and cultured overnight in fresh LB liquid medium in deep-well 96-well plates. The liquid cultures were then diluted into shallow-well 96-well plates by 100 fold by adding 2µL overnight culture into 200µL fresh LB medium. For NahR, XylS and HbpR biosensor, 44 aromatic compounds were then individually added to the diluted cultures to reach 0, 100µM (for 5-ClSaD, 2,4,6-TClPhl, 2-HBP and 2-ABP) or 1000µM (for all the other 40 compounds) final concentration and the diluted cultures were then incubated for 12hrs at 30 °C to reach stationary phase. For DmpR biosensor, the diluted cultures were first incubated for 6hrs to reach late exponential/ early stationary phase. The cultures were then centrifuged at 4000rpm for 10 minutes. The supernatants were discarded and 200µL fresh LB medium was added to each well to re-suspend the pelleted culture. Then 44 aromatic compounds were then individually added to the diluted cultures to reach 0, 100µM (for 5-ClSaD, 2,4,6-TClPhl, 2-HBP and 2-ABP) or 1000µM (for all the other 40 compounds) final concentration and the cultures were further incubated for 4hrs. Then the cultures were diluted 200 fold by adding 1µL stationary phase culture into 200µL Phosphate Buffered Solution (PBS) containing 2mg/ml Kanamycin antibiotic to stop cell growth and protein synthesis. Then sfGFP fluorescence was measured using flow cytometer (BD LSRII). 22
3. Protocol for dose-response tests E. coli Top10 strain was transformed with plasmids containing components of target biosensor, plated onto Luria–Bertani (LB) agar plate and incubated overnight at 37 °C. Single colonies were picked and cultured overnight in fresh LB liquid medium in deep-well 96-well plates. The liquid cultures were then diluted into shallow-well 96-well plates by 100 fold by adding 2µL overnight culture into 200µL fresh LB medium. For NahR, XylS and HbpR biosensor, the target aromatic compound was then added to the diluted cultures to reach 0, 1, 3, 10, 30 and 100µM final concentrations (for some compounds, 0.1, 0.3 and 300µM final concentrations were also tested) and the diluted cultures were then incubated for 12hrs at 30 °C to reach stationary phase. For DmpR biosensor, the diluted cultures were first incubated for 6hrs to reach late exponential/ early stationary phase. The cultures were then centrifuged at 4000rpm for 10 minutes. The supernatants were discarded and 200µL fresh LB medium was added to each well to re-suspend the pelleted culture. Then the target aromatic compound was then added to the diluted cultures to reach 0, 1, 3, 10, 30 and 100µM final concentrations (for some compounds, 0.1, 0.3 and 300µM final concentrations were also tested) and the cultures were further incubated for 4hrs. Then the cultures were diluted 200 fold by adding 1µL stationary phase culture into 200µL Phosphate Buffered Solution (PBS) containing 2mg/ml Kanamycin antibiotic to stop cell growth and protein synthesis. Then sfGFP fluorescence was measured using flow cytometer (BD LSRII). 4. Protocol for interference tests E. coli Top10 strain was transformed with plasmids containing components of target biosensor, plated onto Luria–Bertani (LB) agar plate and incubated overnight at 37 °C. Single colonies were picked and cultured overnight in fresh LB liquid medium in deep-well 96-well plates. The liquid cultures were then diluted into shallow-well 96-well plates by 100 fold by adding 2µL overnight culture into 200µL fresh LB medium. For NahR, XylS and HbpR biosensor, the typical inducer was first added to the diluted cultures to reach 0, 1, 3, 10, 30 and 100µM final concentration. The interfering inducer was then added to the diluted culture to reach 0, 10, 30 and 100µM final concentration. The diluted cultures were then incubated for 12hrs at 30 °C to reach stationary phase. For DmpR biosensor, the diluted cultures were first incubated for 6hrs to reach late exponential/ early stationary phase. The cultures were then centrifuged at 4000rpm for 10 minutes. The supernatants were discarded and 200µL fresh LB medium was added to each well to re-suspend the pelleted culture. After this, the typical inducer was first added to the diluted cultures to reach 0, 1, 3, 10, 30 and 100µM final concentration. The interfering inducer was then added to the diluted culture to reach 0, 10, 30 and 100µM final concentration. The cultures were further incubated for 4hrs. 23
Then the cultures were diluted 200 fold by adding 1µL stationary phase culture into 200µL Phosphate Buffered Solution (PBS) containing 2mg/ml Kanamycin antibiotic to stop cell growth and protein synthesis. Then sfGFP fluorescence was measured using flow cytometer (BD LSRII).
24
VII. Parts, Plasmids and Strains Supplementary Table S7. Parts used in this study. Source & Description a BBa_B0015, double terminator used throughout the study BBa_J61051, terminator + coding sequence of NahR transcription factor + Psal promoter xylS BBa_K1031911, coding sequence of XylS transcription factor + terminator hbpR BBa_K1031300, coding sequence of HbpR transcription factor + terminator dmpR BBa_K1031211, constitutive promoter Pr + dmpR coding sequence BBa_J23113 constitutive promoter BBa_J23109 constitutive promoter BBa_J23117 constitutive promoter BBa_J23114 constitutive promoter BBa_J23115 constitutive promoter BBa_J23106 constitutive promoter sfGFP coding sequence of sfGFP Pm sequence of Pm promoter Pc sequence of Pc promoter Po sequence of Po promoter nahF coding sequence of nahF + terminator pSB1A3 High copy plasmid vector (100~300 copies per cell); replication origin: pUC19-derived pMB1; antibiotic resistance: ampicillin pSB4K5 Low copy plasmid vector (~5 copies per cell); replication origin: pSC101; antibiotic resistance: kanamycin pSB1C3 High copy plasmid vector (100~300 copies per cell); replication origin: pUC19-derived pMB1; antibiotic resistance: chloramphenicol. pV1398 pMMB66HE-based high copy cloning vector; antibiotic resistance: Ampicillin. BBa_B0034 RBS BBa_B0031 RBS BBa_B0032 RBS a. Biological parts whose names start with “BBa” or “pSB” are from the Registry of Standard Biological Parts, where the sequences and features of all parts can be obtained. Name Terminator nahR
25
Supplementary Tabel S8. Plasmids used in this study. Name Components of Insert a Backbone NahR-sfGFP BBa_J61051 + sfGFP + Terminator pSB1C3 Pc113-XylS BBa_J23113 + BBa_K1031911 + Terminator pSB4K5 Pc109-XylS BBa_J23109 + BBa_K1031911 + Terminator pSB4K5 Pc114-XylS BBa_J23114 + BBa_K1031911 + Terminator pSB4K5 Pc105-XylS BBa_J23105 + BBa_K1031911 + Terminator pSB4K5 Pc106-XylS BBa_J23106 + BBa_K1031911 + Terminator pSB4K5 Pc113-HbpR BBa_J23113 + BBa_K1031300 + Terminator pSB4K5 Pc109-HbpR BBa_J23109 + BBa_K1031300 + Terminator pSB4K5 Pc117-HbpR BBa_J23117 + BBa_K1031300 + Terminator pSB4K5 Pc114-HbpR BBa_J23114 + BBa_K1031300 + Terminator pSB4K5 Pc106-HbpR BBa_J23106 + BBa_K1031300 + Terminator pSB4K5 Pr-DmpR BBa_K1031211 + Terminator pV1398 Pm-eGFP Pm + BBa_B0034 + eGFP + Terminator pSB1A3 Pc-31-sfGFP Pc + BBa_B0031 + sfGFP + Terminator pSB1C3 Pc-32-sfGFP Pc + BBa_B0032 + sfGFP + Terminator pSB1C3 Pc-34-sfGFP Pc + BBa_B0034 + sfGFP + Terminator pSB1C3 Po-sfGFP Po + BBa_B0032 + sfGFP + Terminator pSB1C3 Pc114-NahF BBa_J23114 + BBa_B0034 + nahF pSB4K5 a. Listed are inserted components of the plasmids. The components can be traced back to the parts listed in Supplementary Table S7. Plasmid's backbone have been listed in another column. “+” means two biological parts were assembled using standard assembly.
26
Supplementary Table S9. Strains used in this study. Name Plasmid(s) transformed a NahR Biosensor Strain NahR-sfGFP one plasmid among Pc113-XylS, Pc113-XylS, XylS Biosensor Strain Pc113-XylS, Pc113-XylS and Pc106-XylS + Pm-eGFP one plasmid among Pc113-HbpR, Pc109-HbpR, Pc117-HbpR, Pc114-HbpR and Pc106-HbpR + one HbpR Biosensor Strain plasmid among Pc-31-sfGFP, Pc-32-sfGFP and Pc-34-sfGFP DmpR Biosensor Strain Pr-DmpR + Po-sfGFP NahF-R Biosensor Strain Pc114-NahF + NahR-sfGFP a. The plasmids listed can be traced back to the plasmids listed in Supplementary Table S8. "+" means that the plasmids listed are co-transformed into one strain.
27
VIII. Supplementary References 1
2
3
4
5
Cases, I. & de Lorenzo, V. Promoters in the environment: transcriptional regulation in its natural context. Nature reviews. Microbiology 3, 105-‐118, doi:10.1038/nrmicro1084 (2005). Lau, P. C. et al. A bacterial basic region leucine zipper histidine kinase regulating toluene degradation. Proceedings of the National Academy of Sciences of the United States of America 94, 1453-‐1458 (1997). Busch, A., Lacal, J., Martos, A., Ramos, J. L. & Krell, T. Bacterial sensor kinase TodS interacts with agonistic and antagonistic signals. Proceedings of the National Academy of Sciences of the United States of America 104, 13774-‐13779, doi:10.1073/pnas.0701547104 (2007). Diaz, E., Ferrandez, A., Prieto, M. A. & Garcia, J. L. Biodegradation of aromatic compounds by Escherichia coli. Microbiology and molecular biology reviews : MMBR 65, 523-‐569, table of contents, doi:10.1128/MMBR.65.4.523-‐569.2001 (2001). Manso, I. et al. 3-‐Hydroxyphenylpropionate and phenylpropionate are synergistic activators of the MhpR transcriptional regulator from Escherichia coli. The Journal of biological chemistry 284, 21218-‐21228, doi:10.1074/jbc.M109.008243 (2009).
28
