Rupika Madhavan
Abstract
Sinorhizobium meliloti is a species of quorum-sensing bacteria that has a symbiotic relationship with its plant host, Medicago sativa. Quorum sensing allows bacteria to com- municate by excreting and uptaking chemical signals, called autoinducers, and change the behavior of the entire population once a critical concentration of autoinducer is reached. S. meliloti is predicted to have at least 2 quorum sensing networks (QSNs): the hypoth- esized Mel system and the known Sin system. The Sin QSN has a synthase (SinI) which produces several long-chain AHLs with a wide range of carbon-chain length (C12 to C18). Though it has been studied for a while, not much is known about the specifics of the Sin QSN. In this paper, we test the effects of several AHLs (C4-C18) on various strains of S. meliloti to see if more can be understood about the complex behavioral network in these small bacteria.
1 Introduction
Quorum sensing (QS) is a process through which bacteria communicate with each other [1]. It is facilitated by three main parts: (1) autoinducers, which are the chemical signals used by the system for effective communca- tion; (2) synthases (which produce the chem- ical signals); and (3) receptors (which bind to the autoinducers). In ram negative bacteria, like Sinorhizobium meliloti, N -acyl homoser- ine lactons (or AHLs) are the preferred au- toinducer. The bacteria ”learn” about and respond to environmental changes by pre- ducing, emitting, binding to, and responding to these autoinducers.This system allows for population-wide behavioral changes, which are essential for survival. Uncovering the mechanisms used by different quorum sens- ing networks (QSNs) will allow us to better manipulate the behavior of harmful or bene-
ficial bacteria around us.
One of the best understood (QSNs) is that of the bacteria Vibrio fischeri, the biolumi- nescent bacteria that live in a Hawaiian squid [1]. It has a fairly simple system; there are only two proteins involved (See Figure 1). LuxI (the synthase) produces the AHL 3-oxo- C6, which leaves the cell. In a population of bacteria, these AHLs increase in concen- tration and absorbed by the cells. Once in the cell, an autoinducer attaches to the re- ceptor, LuxR. The LuxR-AHL complex then causes the transcription of the luxICDABE operon, which encodes for luciferase, the bi- oluminescent enzyme. In addition, luxI is also encoded within this operon, further pro- ducing AHLs, which, in turn, bind to LuxR. This positive feedback loop accelerates the process, and causes the whole population to produce light. This bacterial circuit is sim-
1
Figure 1: The well-known QSN of V. fischeri. LuxI is a synthase that produces the autoin- ducer, 3-oxo-C6. It is released from the cell, where more AHLs have been excreted. It is then taken in by a cell and attaches to LuxR, the receptor protein. These two to- gether bind to the ”lux-box” (an operon with several genes including luxI and the gene for luciferase) and promote the production of lu- ciferase, the bioluminescent protein. In addi- tion, luxI is also promoted, and thus creates a positive feedback loop that accelerates the quorum sensing process.
ple, but provides a nice insight into the way in which bacteria work. The goal of a scien- tist studying a quorum sensing network is to discover the circuit through which his or her bacteria communicates.
The bacteria discussed in this paper are Sinorhizobium meliloti, gram-negative bacte- ria who fix nitrogen for their host, the alfalfa plant Medicago sativa [2]. The system is more complicated than the V. fischeri. It is sus- pected that there are two QSNs at work in S. meliloti : the Sin system and the hypothe- sized Mel system [3]. As the Mel system has not been located, the Sin system has been
at the center of QS research on S. meliloti. The Sin system is essential for symbiosis with the host plant; strains without a functional synthase cannot successfully invade the host plant’s roots [4]. In addition, Medicago sativa produces AHL-like signals that can interfere with the activities of the Sin QSN [5].
Three proteins work within the framework of the Sin system: SinI, SinR, and ExpR [6, 3]. Sin I is a LuxI homolog; i.e., SinI is the synthase which produces the autoin- ducers used by the system. The position of sinR with respect to sinI on the genome (di- rectly upstream; see Figure 2) would imply that SinR is a LuxR homolog [7]; however, it has been shown that ExpR is the protein that actually binds to the AHLs produced by SinI [4]. Though ExpR is an LuxR solo (expR is not near sinI on the bacterial genome) [8], it does have a binding site is located in the intergenic space between sinI and sinR [7].
Several AHLs have been shown to be used by the rhizobia. Of these, those with carbon chains of lengths between 12 and 18 have been shown to be synthesized by the Sin system; though there is evidence of other AHLs be- ing produced by the bacteria [6]. This wide range of carbon-chain length is rather pecu- liar; these lengths correspond directly to the ability of the AHL to diffuse. Though much is known about S. meliloti, a lot is still left to discover, such as what the purpose of sinR is, whether SinR interacts with Sin AHLs, and whether all the AHLs predicted to be pro- duced by SinI are actually used by the Sin QSN. Once this is known, the bigger purpose of this experiment may be carried out: to dis- cover the diffusion patterns of AHLs in the bacteria.
In this paper, we try to seek the answers to these questions. Our experiment testing the purpose of SinR is running and will pro- vide answers within the next week, but we have shown that of the many AHLs produced by SinI, only 3-oxo-C14, C16:1, and 3-oxo-C16:1
2
Figure 2: The arrangement of the quroum- sensing-related genes in the S. meliloti chro- mosome. It has been predicted that the ExpR binding site is in the intergenic region be- tween sinR and sinI. The lux-box -like gene is predicted to be the binding site for SinR.
change the response behavior of the sinI mu- tant strain when grown in Ty medium.
2 Methods and Materials
The fluorescence and optical density curves were produced by using a BioTek machine. Gen5 software was used to program the ma- chine to take measurements of the optical density and fluorescence every 10 minutes for 48 hours. Polystyrene 48-well plates and 96- well plates were primarily used in these ex- periments.
Five strains of bacteria were studied in these series of experiments: 8530 (wild type), 1021 (expR mutant), MG32 (sinI mu- tant), MG75 (sinI, expR double mutant), and MG170 (sinR mutant). Each of these strains contained a psinI-gfp plasmid−i.e., when the sinI promoter was activated (psinI ), GFP would fluoresce green when blue light was shone on it. Glycerol stock of these strains were kept in a -80 C freezer.
Ice from the glycerol stock cultures was scraped onto a pipette tip and dropped into 5 mL of Ty medium and grown overnight (around 14 hours) in a 30 C incubator. These new cultures were used to streak plates of 1.5% agar (made with antibiotics strepto- mycin at 400 µg/mL and spectinomycin at 50 µg/mL) and then incubated at 30 C un- til enough colonies were formed, but before to many colonies started to merge. These agar plates were used for several experiments,
and then retired when colonies from a single clone were no longer able to be picked. Clonal colonies were picked with a pipette tip and dropped into 5 mL of Ty and grown overnight (around 9 hours) until the optical density was between 0.2 and 0.4. These colonies were di- luted 100-fold, returning them to an OD be- tween 0.002 and 0.004, so the fluorescence due to expression of the sinI promoter could be seen at all stages of growth.
The autoinducers were treated in four ways:
1. Originally, 400 µL of diluted bacteria were added to each well. The AHLs (stored in ethyl acetate) were added di- rectly to the bacteria in the well plates after the bacteria were placed in the wells; however, as ethyl acetate kills bac- teria and the effects from this were no- ticed, it was decided that a new method needed to be used.
2. Phosphate buffer dilutions were used next: The AHLs in ethyl acetate were evaporated in polycarbonate tubes, leav- ing just the AHLs. Phosphate buffer was then added to the appropriate concen- tration and mixed throroughly. Since AHLs do not mix well with phosphate buffer, the concentration of the AHLs was doubted.
3. Further experiments were conducted by placing the AHLs directly into the well plate, and then evaporating. Since the ethyl acetate ate up the polystyrene, 150-200 µL of deionized water were added to the bottom of each well, followed by pipetting 4 µL of the autoinducer-ethyl acetate mixture on top of the water. The wells were placed in a 60 C oven and left to evaporate un- til all the liquid was gone from the wells. As the ethyl acetate is lighter and evapo- rates more quickly, the water was a pro-
3
Strain Relevant Characteristics Source Rm8530 1021 expR+ SmR Pellock et al., 2002 Rm1021 expR102 ::ISRm2011-1 expR, SmR Galibert et al., 2001 MG32 8530 δsinI, SmR Gao et al., 2005 MG75 1021 δsinI, SmR Gao et al., 2005 MG170 8530 δsinR, SmR Gao, et. al, 2012
Table 1: The various names and characteristics of the strains studied in these experiments.
tective barrier between the ethyl acetate and the water. The plate was allowed to cool to room temperature, and the 100- fold dilution of colonies mixed with an- tibiotics was added to the wells. How- ever, some wells were still affected nega- tively by the ethyl acetate.
4. Final experiments were conducted by di- luting the AHLs with ethanol. Ethanol does not eat away polystyrene, and so did not have a negative affect on the well plates. The AHLs were serially diluted in ethanol, and 4 µL of the diluted solution was added to the appropriate wells. The well plate was left under the hood until all the ethanol was evaporated. Ty and the bacteria were then added to the evap- orated wells, and mixed with a pipette.
After adding both autoinducers and bacte- ria to each of the wells, the well plates were placed in the BioTek machine and data was collected for 48 hours at 30 C.
Two experiments were conducted testing the effects of phosphate on psinI activity. The first experiment used a 100 mM phos- phate buffer and Ty medium mixture. The second experiment used a 2 mM phosphate buffer and Ty medium mixture. To add phos- phate to the Ty medium, a 1 M solution of phosphate buffer at a pH of 7 was diluted using Ty medium to the appropriate concen- trations. The various strains were then grown in this mixture.
Furthermore, to test the purpose of SinR, the psinI-gfp plasmid and a plasmid contain-
ing sinR were transformed into E. coli cells via electroporation. The results of these ex- periments have not yet been delivered, so fur- ther detail will be provided when the results come in.
3 Results
In our experiments, we see that only MG32 (sinI mutant) has any response to a varia- tion in the concentration of AHL (Figure 3); it is the only one of the strains not producing AHL (it lacks sinI ) but also has expR (the re- ceptor). The psinI activity in MG32 is signif- icantly lower than in MG75, as shown in Fig- ure 7. However, with the right AHLs, psinI activity increases in MG32, with 3-oxo-C16
raising psinI activity to a level greater than the MG75. The C16:1 and 3-oxo-C14 raise the psinI activity, but not enough to overcome the psinI activity of MG75. As Figure 3 and Figure 5 show, of all the AHLs believed to be produced by the Sin system (C12-C18), only 3-oxo-C16, C16:1, and 3-oxo-C14 showed any increase in psinI activity in the MG32 strain.
Looking at Figure 3 and Figure 4, we see that none of the AHLs have any effect on psinI expression in MG75, the only other strain that does not produce its own AHLs. Furthermore, MG75 (sinI, expR double mu- tant) and 1021 (expR mutant) have fluores- cence curves that increase continuously de- spite the leveling-off of the optical density.
The most significant impact on psinI ex- pression aside from removing sinI itself was
4
0.5
1
1.5
2
2.5
3
3.5
C4
C6
3oC6
C8
C12
C14
3oC14
C16
C16:1
3oC16:1
C18
No AHL
Figure 3: sinI promoter activity of various strains of S.meliloti with all AHLs. 8530 is the wild-type strain with psinI-gfp; 1021 is the expR mutant with psinI-gfp; MG32 is the sinI mutant with psinI-gfp; MG75 is the sinI, expR double mutant with psinI-gfp; MG170 is the sinR mutant with psinI-gfp. It is important to note that data from the 8530 is not reliable; as it forms EPS, a colony forms at the bottom of the wells, making the fluorescence hard to read. Fluorescence curves of the 8530 (not shown) show the eccentric, and hard to map behavior of the 8530.
5
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8 Optical Density of 1021 and MG75 with short−chain AHLs
Time (h)
e n s it y (
A U
1
1.5
2
2.5
3
3.5
5 Fluorescence of 1021 and MG75 with short−chain AHLs
Time (h)
F lu
o re
c o u n ts
)
−expR −expR/−sinI
Figure 6: sinI promoter activity of 1021 (expR mutant) and MG75 (sinI, expR double mutant) with various AHLs (C12, 3-oxo-C14, 3-oxo-C16, C16:1, and C18). The two have identical fluorescence curves, in spite of the fact that MG75 does not have a functional sinI.
0 10 20 30 40 50 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Time (h)
O p
1
1.5
2
2.5
3
3.5
4
4.5
Time (h)
F lu
o re
MG32
Figure 7: sinI promoter activity of MG32 (sinI mutant) and MG75 (sinI, expR double mutant) with various AHLs (C12, 3-oxo-C14, 3-oxo-C16, C16:1, and C18). the psinI activity of MG75 is significantly greater than that of MG32, unless an autoinducer binds to ExpR
6
0.5
1
1.5
2
2.5
3
3.5
100 µM
10 µM
1 µM
100 nM
No AHL
Figure 4: sinI promoter activity of MG75 (sinI, expR double mutant) with several long- chain AHLs, with AHLs prepared by way of method 4. MG75 does not respond to any of the AHLs.
No AHL C12 3oC14 3oC16 C16:1 C18 0
0.5
1
1.5
2
2.5
3
3.5
4
c o u n ts
)
100 µM
10 µM
1 µM
100 nM
No AHL
Figure 5: sinI promoter activity of MG32 (sinI mutant) with several long-chain AHLs, with AHLs prepared by way of method 3. The concentration of AHL needed for a re- sponse is extremely high (100 µM!)
C14 3oC14 C16 C16:1 3oC16 No AHL 0
0.5
1
1.5
2
2.5
3
3.5
1.5 µM
150 nM
15 nM
1.5 nM
Figure 9: sinI promoter activity of MG32 (sinI mutant) with long-chain AHLs, with AHLs prepared by way of method 4. The concentration of AHL needed for a response is lower (1.5 µM).
removing sinR, with around a 3-fold decrease in fluorescence. Disrupting expR reduced the fluorescence by a factor of two. Once expR was removed, disrupting sinI from the expR mutant had little to no effect; the expR mu- tant and expR, sinI double mutant had the same fluorescence curves (Figure 6). As this result was surprising, PCR was done to con- firm that these two strains were in fact differ- ent; the results showed that they were.
As the experimental technique improved, the concentration of AHL required to induce a response decreased significantly (compare Figure 5 and Figure 9). The final experiments show a response to concentrations around 150 nM.
When the phosphate concentration was in- creased to 2 mM, no effect was detected; how- ever, when the phosphate concentration was increased to 100 mM, a significant change in the behavior of the bacteria was detected. The fluorescence curves had an interesting pattern (Figure 8).
7
0.1
0.2
0.3
0.4
0.5
0.6
Time (h)
O p
2
3
4
5
6
7
8
Time (h)
F lu
o re
No AHL
Figure 8: sinI promoter activity of MG32 (sinI mutant) with various AHLs (C6, 3-oxo-C6, C8, C12, C14, 3-oxo-C14, C16, 3-oxo-C16, C16:1, and C18)
4 Discussion
The main interest of our physics group in S. meliloti has to do with the wide range of AHLs it produces. As this group previously did an experiment looking at the QS diffu- sion patterns of Vibrio fischeri, we thought it would be interesting to see how an organ- ism interacted with a wide range of autoin- ducers. Our hypothesis was that the short- chain AHLs would diffuse more quickly and over a larger area than the long-chain AHLs, and thereby control bacterial population be- havior on a grand scale, while the long-chain AHLs would control behavior at a more lo- cal range. However, in the process of learn- ing more about this species, we have discov- ered that there is much left to understand about the interesting quorum sensing net- work of these bacteria before diffusion exper- iments can be conducted.
Marketon et al. (2002) identified six AHLs being produced by the Sin QSN: C8-HL, C12- HL, 3-oxo-C14, 3-oxoC16 : 1, C16:1, and C18
[6]. They also discovered AHLs with shorter chains: 3-oxo-C6 and C6; however, these AHLs were unaffected by a disruption in the sinI gene, implying that the Sin QSN does not produce these AHLs. Teplitski, et al. (2003) found a response from C14, 3-oxo-C14, C16, 3-oxo-C16 and C16:1, but did not find a response from C8, C12, or C18 [9]. We only found a response in 3-oxo-C14, 3-oxo-C16 and C16:1. It should be noted that the environ- ment of the rhizobia has a profound effect on the way that these bacteria interact [10]; the Teplitski group found a response from C14-HL and C16-HL in different growth media than the one used in this experiment (Ty).
We are unsure of why S. meliloti would produce so many AHLs without using all of them. What is clear, however, is that the primary AHLs seem to be 3-oxo-C14, 3-oxo- C16:1, and C16:1 in that order. This is fur- ther supported by an experiment conducted by Gao et. al (2007), in which a gene en- coding for AiiA (which inactivates AHLs) was inserted into Rm1021 [11]. The exper-
8
Figure 10: SinI produces several AHLs; of these, only 3-oxo-C14, C16:1, and 3-oxo-C16:1 bind to ExpR, which represses sinR activity, but promotes sinI activity when bound to AHL. SinR constantly upregulates sinI, but is repressed by ExpR and promoted by PhoB. A positive feedback loop with expR allows for the autoregulation of expR.
9
iment showed that without AiiA, 3-oxo-C16:1
was shown to be the most abundant QS sig- nal, followed by C16:1 and C16 (in minimal medium), and finally 3-oxo-C14. With the AiiA insertion, however, C16:1 was the only AHL detected, with its signal only reduced by 90%. At these levels, this paper also showed that even at these low concentrations of AHL, the bacteria still swarmed, indicating that this AHL allows for quorum sensing in even the direst circumstances.
Not much is known about SinR; it is not very soluble. McIntosh et al. (2009) showed that eliminating sinR reduces sinI expres- sion to background levels, while the overex- pression of sinR increases sinI expression 8- fold, implying that SinR is the primary pro- moter of sinI [10]. Looking at the fluores- cence of the sinR mutant in our experiments, we see the same thing; the fluorescence drops down to the same levels as the sinI mutant. However, looking at Figure 3, and consider- ing the fact that sinI and sinR transcribe in the same direction, there is the possibil- ity that disrupting sinR also disrupts sinI. As Figure 3 and Figure 5show, eliminating expR increases psinI activity two-fold (com- pare the sinI mutant with the expR mutant and sinI /expR double mutant). Since SinR seems to be the biggest promoter of sinI [10], this implies that ExpR represses sinR expres- sion somehow, an idea suggested by McIntosh et al. (2009).
Comparing MG32 with the other strains in Figure 5, we see that while disrupting sinI reduces psinI activity to background levels, activity is regained with the addition of AHL. This implies that ExpR regulates itself. Our experiments show that without expR, we see no response in psinI activity due to increased AHL concentrations. Furthermore, looking at Figure 6, we can see that once expR has been disrupted, removing sinI has no effect on the activity of psinI ; i.e., ExpR is neces- sary for AHL-dependent quorum sensing.
We see that phosphate has some clear effect on the system. McIntosh et al. (2009) showed that in phosphate-limiting conditions, sinR is significantly induced [10]. This dependence of sinR on phosphate levels is mediated through the Pho regulon. One study found that there are around 96 Pho boxes in the S. meliloti genome [12]. Another study located one Pho box upstream of sinR [13].
Putting all of this together, we can organize a new circuit diagram for S. meliloti (Fig- ure 4). The next step for this experiment (once we know which autoinducers create a response in psinI activity and what sinR is doing) is to calculate the diffusion curves for the quorum sensing response. As this paper is being written, we are conducting an exper- iment that will test if SinR interacts with any of the Sin AHLs; we have transformed a plas- mid encoding sinR with the psinI-gfp gene and one with expR and psinI-gfp into E. coli. Using the BioTek machine, we will add AHL and see if the fluorescence increases with any of these AHLs. This will decouple sinR and expR from the QS circuit, and we will be able to see which AHLs, if any, are the ligands for SinR and see if ExpR can bind to AHL with- out SinR. If our E. coli experiments show that SinR does not interact with AHL, we will be able to confirm our suspicions that ExpR, and only ExpR, binds to the Sin AHLs.
5 Conclusion
Out of the five strains used in this experi- ment, only the psinI activity of MG32 re- sponds to an increase in autoinducer con- centration. The only autoinducers that in- duce this increased response are the 3-oxo- C14, C16:1, and 3-oxo-C16:1. ExpR appears to repress sinR expression. SinR has the biggest effect on psinI activity. These three autoin- ducers will be used to perform diffusion ex- periments once more data is collected.
10
6 Acknowledgements
The research conducted in this paper was funded by an NSF grant: NSF DMR- 1156737.
7 References
References
[1] C. M. Waters and B. L. Bassler, “Quo- rum sensing: cell-to-cell communication in bacteria.,” Annu Rev Cell Dev Biol, vol. 21, pp. 319–346, 2005.
[2] N. Gurich and J. E. Gonzalez, “Role of quorum sensing in sinorhizobium meliloti-alfalfa symbiosis,” Journal of Bacteriology, vol. 191, pp. 4372– 4382, Jul 2009. LR: 20100927; GR: 1R01GM069925/GM/NIGMS NIH HHS/United States; JID: 2985120R; 0 (Bacterial Proteins); OID: NLM: PMC2698488; 2009/04/24 [aheadof- print]; ppublish.
[3] M. M. Marketon and J. E. Gonzalez, “Identification of two quorum-sensing systems in sinorhizobium meliloti,” Journal of Bacteriology, vol. 184, pp. 3466–3475, Jul 2002. LR: 20100914; GENBANK/AF456804; JID: 2985120R; 0 (Bacterial Proteins); 0 (Chaperonin 60); 0 (Escherichia coli Proteins); 0 (Membrane Proteins); 0 (TraR pro- tein, Agrobacterium tumefaciens); 104042-79-7 (TraM protein, bacte- rial); 1192-20-7 (homoserine lactone); 147757-14-0 (FlaD protein, Bacteria); 148591-53-1 (TrbI protein, E coli); 96-48-0 (4-Butyrolactone); OID: NLM: PMC135146; ppublish.
[4] M. M. Marketon, S. A. Glenn, A. Eber- hard, and J. E. Gonzalez, “Quorum sens-
ing controls exopolysaccharide produc- tion in sinorhizobium meliloti,” Jour- nal of Bacteriology, vol. 185, pp. 325– 331, Jan 2003. LR: 20091118; JID: 2985120R; 0 (Bacterial Proteins); 0 (Culture Media); 0 (LuxI protein, Bac- teria); 0 (Polysaccharides, Bacterial); 0 (SinI protein, Bacillus subtilis); 0 (Tran- scription Factors); 1192-20-7 (homoser- ine lactone); 147757-14-0 (FlaD protein, Bacteria); 73667-50-2 (succinoglycan); 96-48-0 (4-Butyrolactone); OID: NLM: PMC141839; ppublish.
[5] N. D. Keshavan, P. K. Chowdhary, D. C. Haines, and J. E. Gonzalez, “L- canavanine made by medicago sativa in- terferes with quorum sensing in sinorhi- zobium meliloti,” Journal of Bacteriol- ogy, vol. 187, pp. 8427–8436, Dec 2005. LR: 20100921; JID: 2985120R; 0 (In- doles); 0 (Plant Extracts); 0 (Polysac- charides, Bacterial); 543-38-4 (Canava- nine); 548-54-9 (violacein); OID: NLM: PMC1317012; ppublish.
[6] M. M. Marketon, M. R. Gronquist, A. Eberhard, and J. E. Gonzalez, “Characterization of the sinorhizobium meliloti sinr/sini locus and the pro- duction of novel n-acyl homoserine lac- tones,” Journal of Bacteriology, vol. 184, pp. 5686–5695, Oct 2002. LR: 20091118; JID: 2985120R; 0 (Bacterial Proteins); 0 (Culture Media); 0 (SinI protein, Bacil- lus subtilis); 1192-20-7 (homoserine lac- tone); 147757-14-0 (FlaD protein, Bac- teria); 96-48-0 (4-Butyrolactone); OID: NLM: PMC139616; ppublish.
[7] F. W. Bartels, M. McIntosh, A. Fuhrmann, C. Metzendorf, P. Plat- tner, N. Sewald, D. Anselmetti, R. Ros, and A. Becker, “Effector-stimulated single molecule protein-dna interac- tions of a quorum-sensing system in
11
sinorhizobium meliloti,” Biophysical journal, vol. 92, pp. 4391–4400, Jun 15 2007. LR: 20091118; JID: 0370626; 0 (DNA, Bacterial); 1192-20-7 (homoser- ine lactone); 96-48-0 (4-Butyrolactone); OID: NLM: PMC1877767; 2007/03/23 [aheadofprint]; ppublish.
[8] S. Subramoni and V. Venturi, “Luxr-family ’solos’: bachelor sen- sors/regulators of signalling molecules.,” Microbiology, vol. 155, pp. 1377–1385, May 2009.
[9] M. Teplitski, A. Eberhard, M. R. Gronquist, M. Gao, J. B. Robinson, and W. D. Bauer, “Chemical identi- fication of n-acyl homoserine lactone quorum-sensing signals produced by sinorhizobium meliloti strains in defined medium,” Archives of Microbiology, vol. 180, pp. 494–497, Dec 2003. LR: 20111117; GR: GM 08500/GM/NIGMS NIH HHS/United States; GR: GM 53830/GM/NIGMS NIH HHS/United States; JID: 0410427; 0 (Culture Media); 1192-20-7 (homoserine lac- tone); 96-48-0 (4-Butyrolactone); 2003/04/29 [received]; 2003/09/12 [revised]; 2003/10/02 [accepted]; 2003/10/31 [aheadofprint]; ppublish.
[10] M. McIntosh, S. Meyer, and A. Becker, “Novel sinorhizobium meliloti quorum sensing positive and negative regulatory feedback mechanisms respond to phos- phate availability,” Molecular microbiol- ogy, vol. 74, pp. 1238–1256, Dec 2009. JID: 8712028; 0 (Acyl-Butyrolactones); 0 (Bacterial Proteins); 0 (Phosphates); 104220-36-2 (PhoB protein, Bacteria); 147757-14-0 (FlaD protein, Bacteria); 2009/11/02 [aheadofprint]; ppublish.
[11] M. Gao, H. Chen, A. Eberhard, M. R. Gronquist, J. B. Robinson, M. Con-
nolly, M. Teplitski, B. G. Rolfe, and W. D. Bauer, “Effects of aiia- mediated quorum quenching in sinorhi- zobium meliloti on quorum-sensing sig- nals, proteome patterns, and sym- biotic interactions,” Molecular plant- microbe interactions : MPMI, vol. 20, pp. 843–856, Jul 2007. LR: 20071203; GR: GM042893-14/GM/NIGMS NIH HHS/United States; JID: 9107902; 0 (Bacterial Proteins); 0 (Proteome); 1192-20-7 (homoserine lactone); 96-48- 0 (4-Butyrolactone); EC 3.1.1.- (Car- boxylic Ester Hydrolases); EC 3.1.1.- (N- acyl homoserine lactonase); ppublish.
[12] Z.-C. Yuan, R. Zaheer, R. Morton, and T. M. Finan, “Genome prediction of phob regulated promoters in sinorhizo- bium meliloti and twelve proteobacte- ria.,” Nucleic Acids Res, vol. 34, no. 9, pp. 2686–2697, 2006.
[13] E. Krol and A. Becker, “Global tran- scriptional analysis of the phosphate starvation response in sinorhizobium meliloti strains 1021 and 2011.,” Mol Genet Genomics, vol. 272, pp. 1–17, Aug 2004.
12