Bacterial Quorum Sensing, Cooperativity, and Anticipation of Stationary-Phase Stress

Bacterial quorum sensing, cooperativity, and anticipation of stationary-phase stress

Eunhye Gooa,1, Charlotte D. Majerczykb,1, Jae Hyung Ana,1, Josephine R. Chandlerb, Young-Su Seoc, Hyeonheui Hama, Jae Yun Lima, Hongsup Kima, Bongsoo Leea, Moon Sun Janga, E. Peter Greenbergb,2, and Ingyu Hwanga,2

aDepartment of Agricultural Biotechnology, Seoul National University, Seoul 151-921, Korea; bDepartment of Microbiology, University of Washington, Seattle, WA 98195; and cDepartment of Microbiology, Pusan National University, Busan 609-735, Korea

Contributed by E. Peter Greenberg, October 18, 2012 (sent for review September 10, 2012) Acyl-homoserine lactone–mediated quorum sensing (QS) regulates system. In B. glumae, TofI-R regulates toxoflavin production, diverse activities in many species of Proteobacteria. QS-controlled motility, protein secretion, and QsmR, an isocitrate lyase regulator- genes commonly code for production of secreted or excreted public type transcriptional regulator (10–12). Less is known about gene goods. The acyl-homoserine lactones are synthesized by members control by C8-HSL in B. thailandensis via BtaI-R and B. pseu- of the LuxI signal synthase family and are detected by cognate domallei via BpsI-R (13–15). In the present study, we found that members of the LuxR family of transcriptional regulators. QS QS induces cellular enzymes for production of oxalate (HOOC- affords a means of population density-dependent gene regulation. COOH), which is excreted into the culture medium and likely Control of public goods via QS provides a fitness benefit. Another shared by other members of the population, and as such is a potential role for QS is to anticipate overcrowding. As population public good. The oxalate protects stationary-phase cells from self- density increases and stationary phase approaches, QS might in- intoxication and killing as a result of ammonia production. duce functions important for existence in stationary phase. Here we provide evidence that in three related species of the genus Results Burkholderia QS allows individuals to anticipate and survive sta- QS Is Essential for Stationary-Phase Survival of Burkholderia. We tionary-phase stress. Survival requires QS-dependent activation of assessed the role of QS during stationary phase in nutrient-rich cellular enzymes required for production of excreted oxalate, which LB broth using WT and C8-HSL-QS signal synthesis mutants of serves to counteract ammonia-mediated alkaline toxicity during B. glumae, B. pseudomallei, and B. thailandensis. Exponential MICROBIOLOGY stationary phase. Our findings provide an example of QS serving growth of each QS mutant was comparable to that of its isogenic as a means to anticipate stationary phase or life at the carrying WT strain; however, whereas the WT strain survived long peri- capacity of a population by activating the expression of cytoplasmic ods in stationary phase, the QS mutants showed massive and enzymes, altering cellular metabolism, and producing a shared re- rapid population crashes commencing shortly after the onset of source or public good, oxalate. stationary phase (Fig. 1A). Mutant population crashes were averted by adding C8-HSL to the growth media (Fig. 1A), in- Burkholderia carrying capacity | glumae | pseudomallei | thailandensis | dicating that QS is involved in stationary-phase survival in these cell death three bacterial species.

cyl-homoserine lactone (AHL)-mediated quorum sensing Massive Population Crashes Are Related to Medium Alkalization in A(QS) regulates diverse activities, including bioluminescence, Stationary Phase. To determine the cause of the observed pop- biofilm formation, motility, and virulence factor formation, in ulation crashes, we monitored the pH of the culture fluids many Proteobacteria (1–3). AHLs are synthesized most typically during growth. The pH of both the WT and QS mutants of B. by members of the LuxI family signal synthases and detected by glumae, B. pseudomallei,andB. thailandensis was approximately members of the LuxR family of transcriptional regulators (1–3). 7 early in growth and rose to between 7.5 and 8 as cultures A large body of work has characterized the molecular mecha- entered stationary phase (Fig. 1B). During continued station- nisms of bacterial QS; demonstrating the population-wide ben- ary-phase growth, the pH of WT cultures fell to 7 or lower, efits that drive QS-mediated cooperative behavior has proven whereas the pH of the QS mutant cultures rose above 8. The pH difficult, however. Cooperative activities benefit individuals within increase in the QS mutants was correlated with a drop in cell agroup(4,5). viability (Fig. 1). QS-controlled genes commonly code for the production of To test whether cell death was caused by base toxicity, we extracellular public goods that can be shared by all members of added 100 mM Hepes (pH 7) to the growth medium. We found the group regardless of which members produce them. These that addition of this strong buffer to the growth medium spared extracellular products are often important for nutrient acquisi- the QS mutants from death in stationary phase (Fig. S1). We tion, interspecies competition, or virulence (6–8). In Pseudomo- adjusted the pH of the culture medium with sodium hydroxide nas aeruginosa, QS control of secreted proteases provides fitness and monitored cell growth to determine the sensitivity of B. glu- benefits, because the proteases are produced only when they can mae and B. thailandensis to alkaline conditions. We found that the be used efficiently (9). Other potential roles of QS in bacteria have been proposed, including the hypothesis that QS enables bacteria to anticipate population carrying capacity in a given Author contributions: C.D.M., E.P.G., and I.H. designed research; E.G., C.D.M., J.H.A., J.R.C., Y.-S.S., H.H., H.K., B.L., and M.S.J. performed research; E.G., C.D.M., J.H.A., J.R.C., J.Y.L., E.P.G., environment. Anticipation of stationary phase might allow indi- and I.H. analyzed data; and E.G., C.D.M., J.H.A., J.R.C., E.P.G., and I.H. wrote the paper. viduals to modify their physiology in preparation for survival at The authors declare no conflict of interest. population carrying capacity. Data deposition: The RNAseq data reported in this paper have been deposited in the Here we address the question of whether QS is involved in Gene Expression Omnibus (GEO) database, www.ncbi.nlm.nih.gov/geo (accession no. anticipation of stationary-phase stress in three closely related GSE36485). bacteria: the rice pathogen Burkholderia glumae, the opportunistic 1E.G., C.D.M., and J.H.A. contributed equally to this work. human pathogen Burkholderia pseudomallei, and the nonpatho- 2To whom correspondence may be addressed E-mail: [email protected] or [email protected]. genic saprophyte Burkholderia thailandensis. Each species contains This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. a conserved N-octanoyl homoserine lactone (C8-HSL) signaling 1073/pnas.1218092109/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1218092109 PNAS Early Edition | 1of6 Downloaded by guest on October 1, 2021 A B. glumae B. pseudomallei B. thailandensis 12

WT 4 WT Δ btaI1 tofI::Ω WT Δ bpsI1 qsmR-135::ISlacZ- 2 qsmR::Ω PrhaBo-Tp/FRT Δ bpsI1 + 4 μM tofI::Ω + 1 μM C8-HSL Δ btaI1 + 1 μM C8-HSL

Population density (Log CFU/ml) C8-HSL 0 6 12 24 28 0612 24 32 48 0 6 12 24 28 48

B 10

pH 7

WT 6 WT Δ btaI1 tofI::Ω WT qsmR-135::ISlacZ- 5 qsmR::Ω Δ bpsI1 PrhaBo-Tp/FRT tofI::Ω + 1 μM C8-HSL Δ bpsI1 + 4 μM C8-HSL Δ btaI1 + 1 μM C8-HSL 4 0 612 2428 0612 24 32 48 0 6 12 24 28 48 Time (hours)

Fig. 1. Cell viability and culture medium pH of B. glumae, B. pseudomallei,andB. thailandensis grown in LB broth at 37 °C with shaking. At the indicated times, small volumes were removed and used to determine cell numbers as cfu by plate-counting methods (A), or cells were removed by centrifugation and pH was measured with a pH electrode and meter (B). Open blue circles indicate WT strains of three Burkholderia spp.; open orange triangles indicate C8-HSL synthase mutants BGS2 (B. glumae BGR1 tofI::Ω), CM135 (B. pseudomallei 1026b ΔpurM ΔbspI1), and JBT101 (B. thailandensis E264 ΔbtaI1); open green squares indicate QsmR mutants BGS9 (B. glumae BGR1 qsmR::Ω) and BT09539 (B. thailandensis E264 qsmR-135::ISlacZ-PrhaBo-Tp/FRT); and ﬁlled red triangles indicate C8-HSL synthase mutants grown in media containing 1 μMor4μM C8-HSL. Error bars represent the error ranges of experiments performed in triplicate.

WT bacteria survived at pH 8, but not at pH 9 (Fig. S2). These a two-gene operon and are responsible for oxalate synthesis (18) data support the view that culture medium alkalization is the (Fig. 4 A and B). Oxalate is an acidic molecule found in multiple cause of QS mutant cell death. domains of life. Its physiological role in bacteria has been largely It is well documented that alkalization in other bacteria occurs uncharacterized, and it has been predicted to exist primarily as due to ammonia production as a result of amino acid catabolism a metabolic end product. We hypothesized that the acidity of in complex media, such as LB broth (16, 17). Thus, we monitored oxalate might counter base intoxication in these species. Con- ammonia production in the WT and C8-HSL-QS signal synthesis sequently, we measured oxalate levels in WT and QS mutant mutants of B. glumae, B. pseudomallei,andB. thailandensis grown cultures of B. glumae, B. pseudomallei, and B. thailandensis. The in LB broth. We found that ammonia was produced continuously WT of each species produced oxalate in amounts presumably by both WT and QS mutants (Fig. 2A); however, the ammonia sufﬁcient to neutralize the accumulated ammonia, whereas the production of QS mutants seemed to decrease at 12 h after in- QS mutants produced very little oxalate (Fig. 2B). This indicates cubation in LB, but not after incubation in buffered LB (Figs. 2A that QS is required for production of a neutralizing agent, oxa- and 3A). The decrease in ammonia production corresponded to late, in WT bacteria to avoid alkaline toxicity. Alkaline conditions the growth defect of the mutants (Fig. 1A). Thus, the QS-de- were not required for oxalate synthesis, however (Fig. 3B). pendent shift in metabolism that limits alkalization is not a shift Because ObcA and ObcB are required for the production of away from ammonia production. Instead, we posit that QS con- oxalate in B. glumae (18, 19), we hypothesized that obcA and trols metabolism such that pH changes are constrained. obcB mutants also would be subject to ammonia-induced base toxicity in stationary phase. As expected, both obcA and obcB QS-Dependent Oxalate Production Counteracts Base Toxicity. To mutants showed massive population crashes corresponding to identify the QS-controlled process or processes involved in coun- increased pH in stationary phase (Fig. 4C). In B. thailandensis tering ammonia-induced alkalization, we performed an RNAseq and B. pseudomallei, a single Obc homolog, Obc1, is predicted transcriptome comparison of WT B. glumae and its QS mutant to be responsible for oxalate biosynthesis (20). The N termi- strain, and drew on published B. glumae proteomics data (12). nus of Obc1 shares 54% amino acid identity with full-length The QS-activated genes identiﬁed included those encoding the ObcA, whereas the C terminus of Obc1 has no similarity with enzymes ObcA and ObcB (Table S1). obcA and obcB compose ObcA or ObcB (Fig. S3). We measured oxalate levels in the

2of6 | www.pnas.org/cgi/doi/10.1073/pnas.1218092109 Goo et al. Downloaded by guest on October 1, 2021 A B. glumae B. pseudomallei B. thailandensis 50 120 70 WT WT tofI::Ω Δ btaI1 100 60 40 qsmR::Ω qsmR-135::ISlacZ- tofI::Ω + 1 μM C8-HSL 50 PrhaBo-Tp/FRT 80 Δ btaI1 + 1 μM C8-HSL 30 40 60 20 30 40 Ammonia (mM) WT 20 10 Δ bpsI1 20 Δ bpsI1 + 4 μM C8-HSL 10

0 6212 428 0612 24 32 48 0 6212 4

B 20 60 35 WT WT tofI::Ω 50 30 Δ btaI1 15 qsmR::Ω qsmR-135::ISlacZ- tofI::Ω + 1 μM C8-HSL 25 PrhaBo-Tp/FRT 40 Δ btaI1 + 1 μM C8-HSL 20 10 30 15

Oxalate (mM) 20 WT 10 MICROBIOLOGY 5 Δ bpsI1 10 Δ bpsI1 + 4 μM C8-HSL 5

0 6212 428 0612 24 32 48 016224 Time (hours)

Fig. 2. Ammonia and oxalate production by B. glumae, B. pseudomallei, and B. thailandensis. Cultures were grown as described in Fig. 1, and concentrations of ammonia (A) or oxalate (B) were measured as described in Materials and Methods. Open circles indicate the WT strains; open triangles, C8-HSL synthase mutants; open squares, QsmR mutants; and ﬁlled triangles, C8-HSL synthase mutants grown in media containing 1 μMor4μM C8-HSL. Error bars represent the error ranges of experiments performed in triplicate.

WT and an obc1 mutant of B. thailandensis and found that the stationary-phase death, as do QS mutants. Thus, we tested cell latter did not produce oxalate, confirming that Obc1 is respon- viability of B. glumae and B. thailandensis qsmR mutants grown sible for oxalate biosynthesis in B. thailandensis (Fig. 4D). Con- in LB broth, and found a stationary-phase survival defect in sistent with the hypothesis that oxalate is required for stationary- these mutants (Fig. 1). The qsmR mutant population crashes phase survival, the obc1 mutant of B. thailandensis exhibited were correlated with rising pH and a lack of oxalate production alkalization and a massive population crash in stationary phase (Figs. 1 and 2). Therefore, we believe that C8-HSL QS controls (Fig. 4E). a battery of genes, including qsmR, and that it is the qsmR product that directly activates genes for oxalate synthesis. Regulation of obc Genes by QsmR. Because our results show that oxalate production in B. glumae, B. pseudomallei,andB. thai- Discussion landensis requires a functional C8-HSL QS system, we asked There has been recent debate about the selective forces that led whether the obc operon was directly or indirectly activated by to the evolution of AHL QS. One idea is that QS control of QS. Insight came from our transcriptomics analysis of B. glu- public goods provides an advantage to cooperating individuals mae, which showed that both a functional C8-HSL QS system (21–24). We believe we have uncovered another particularly and the QsmR regulator are required for activation of obcA and interesting role for AHL QS in B. glumae, B. pseudomallei, and obcB (Table S1). These data indicate that the obc genes belong B. thailandensis. The enormous population crashes of QS mu- to the QsmR regulon, which is activated by QS. To further tants in stationary phase represent a case in which QS serves to confirm this hypothesis, we measured expression levels of obcA anticipate crowding before the cells experience loss of viability. and obcB from a chromosomal obcAB-gusA transcriptional fu- Our findings suggest that bacteria use QS to determine the pop- sion in B. glumae. We found that QsmR activates obcA and ulation density to anticipate carrying capacity to avoid population obcB expression and interacts directly with the obcA-B pro- collapse. Specifically, the three species of Burkholderia that we moter (Fig. 5). studied sense increasing population densities by QS and alter their metabolism; thus, the group is protected from high con- B. glumae and B. thailandensis QsmR Mutants Behave Like QS Mutants. centrations of ammonia, a toxic end product of energy metabo- Given our finding of similar low levels of obcAB expression lism. The metabolic rewiring induced by QS leads to excretion of in the B. glumae qsmR and QS mutants, we reasoned that oxalate, a public good that provides protection from base toxicity qsmR mutants might fail to accumulate oxalate and experience to the group.

Goo et al. PNAS Early Edition | 3of6 Downloaded by guest on October 1, 2021 A The fact that oxalate is produced at neutral pH suggests that 40 anticipatory production of oxalate by QS-dependent mechanisms is neither dependent on alkaline pH nor operated by a simple pH-dependent regulatory mechanism. Where might 30 B. glumae, B. pseudomallei,andB. thailandensis encounter alkaline pH in natural habitats? It is plausible that B. glumae experiences high pH in the plant apoplast owing to active defense responses to infection in plants. Apoplasts contain sugars 20 and organic acids rather than amino acids (26); however, under WT apoplastic conditions, plants alkalize the infection site as a tofI::Ω defense mechanism (27). In this scenario, it is conceivable that Ammonia (mM) 10 qsmR::Ω QS-dependent oxalate production can contribute to the fitness of B. glumae by preventing plant-mediated alkalization in plant apoplasts during infection. Oxalate production also has been linked to pathogenic B. glumae isolates. A survey of 200 isolates from Japan found that all 180 virulent species produced oxalate, 03326 912 428 2whereas the 20 avirulent isolates did not (19). Avoiding alkalization is also important because AHLs are degraded above pH B 25 8.0 (28, 29); thus, anticipatory production of oxalate in a QS- dependent manner might be one way of protecting AHL signals 20 in alkaline conditions. B. pseudomallei might experience amino WT acid-rich environments in mammals during infection, and all of tofI::Ω these species exist in soil, where pH can vary. Much of the literature on biological oxalate synthesis fo- 15 qsmR::Ω cuses on the formation of oxalate crystals that cause kidney stones in humans and crystal formation in plants (30–32). 10 Oxalate production by certain plant pathogenic fungi is important for virulence (33, 34). Oxalate traditionally has been Oxalate (mM) considered a metabolic end product, but in the Burkholderia 5 species that we studied, oxalate clearly serves to protect against base toxicity. This finding is consistent with a previous report indicating that growth inhibition of B. pseudomallei is due to am- monium toxicity after spontaneous loss of oxalate production (35). 0 326 912 43228 In B. glumae, both ObcA and ObcB synthesize oxalate from Time (hours) acetyl-CoA and oxaloacetate derived from the tricarboxylic acid cycle (18, 19). In bacteria, oxalate synthesis more commonly Fig. 3. Ammonia and oxalate levels in culture fluid of the WT, tofI mutant, occurs through a separate pathway involving glyoxylate as an and qsmR mutant of B. glumae grown in LB supplemented with 100 mM intermediate (36, 37); however, we found no evidence of this Hepes. Open circles, the WT strains; open triangles, C8-HSL synthase mu- pathway in any of the three Burkholderia species that we stud- tants; open squares, QsmR mutants. Error bars represent the error ranges of experiments performed in triplicate. ied. We show that Obc1 is responsible for oxalate synthesis in B. thailandensis. It is very likely that the B. pseudomallei Obc1 is also responsible for oxalate biosynthesis, given that it shares 95% When bacteria use amino acids as a carbon source, deamination amino acid identity with the Obc1 protein of B. thailandensis.In results in ammonia release, which can result in alkalization (16, support of conserved functionality among the Obc enzymes, the 17). For many bacterial species, ammonia accumulation is not B. mallei obc1 gene can complement a B. glumae obcA mutation a direct cause of population crashes in stationary phase, because (20). Therefore, oxalate synthesis in B. glumae, B. thailandensis, pH homeostasis mechanisms allow cells to survive the alkaline and B. pseudomallei seems to share a biochemical pathway, in conditions. However, we found that alkaline pH is toxic to WT that oxalate is made in the branched tricarboxylic acid cycle in strains of B. glumae and B. thailandensis, and that these species a QS-dependent manner. and B. pseudomallei use C8-HSL QS to limit environmental al- The results of the present study show that QS can function to kalization. QS induces a change in metabolism such that cells allow anticipation of overcrowding and promote bacterial survival excrete copious amounts of oxalate. In a sense, oxalate is an ideal at maximum population carrying capacity, and provide insight into organic acid for neutralizing ammonia because of its dianionic how QS bacteria have evolved to control both public and private nature. Therefore, QS control of cellular metabolism serves an goods. It can be imagined that QS evolved in a common ancestor anticipatory function and involves the production of an excreted of B. glumae, B. thailandensis, and B. pseudomallei to anticipate public good, oxalate. the onset of stationary phase and prepare cells to survive in this It has been reported that QS controls carbon flux in some crowded environment. Survival involves induction of cellular bacteria, which results in the maintenance of neutral pH during functions that are required for production of oxalate, an excreted culture (6, 25). We believe, however, that QS control of oxalate public good. This is one solution to a problem encountered by bacteria that use amino acids for energy―an end product of amino in these Burkholderia species represents a unique mechanism acid metabolism is ammonia, and ammonia production leads to of anticipatory pH control. Notably, QS-controlled oxalate increasing pH in a local environment. accumulation begins in late exponential phase, long before the toxic effects of ammonia-induced alkalinization lead to re- Materials and Methods duced cell viability. Moreover, oxalate likely is not an excreted Bacterial Strains and Growth Conditions. The bacterial strains and plasmids used product imported later and used for carbon and energy, given are listed in Table S2. Strains of B. glumae and B. thailandensis were grown in LB that it cannot serve as a carbon source for B. thailandensis or broth (0.1% tryptone, 0.5% yeast extract, and 0.5% NaCl; USB Products) or in LB B. glumae. broth buffered with 100 mM Hepes (pH 7.0) at 37 °C. Strains of B. pseudomallei

4of6 | www.pnas.org/cgi/doi/10.1073/pnas.1218092109 Goo et al. Downloaded by guest on October 1, 2021 A bglu_2g 18760 18770 18780 18790 18880 obcB obcA

95 91 29 83 13 5 Tn3-gusA fusions + + -- -+Oxalate production (H) EN NN N (E) 1 kb pOBC1 // // BC25 12 10

10 9 20 WT obcA::Tn3-gusA83 8 8 15

obcB::Tn3-gusA29 pH 6 7

Oxalate (mM) 10 4 6 pHCFU obcA::Tn3-gusA83 5 5 2 obcB::Tn3-gusA29

Population density (Log CFU/ml) 4 0 0 12 24 12 24 28 36 48 DE30 12 10 MICROBIOLOGY

25 WT 10 9 obc1-117::ISlacZ hah-Tc 20 8 8 pH 15 6 7 Oxalate (mM) 10 4 6

pHCFU 5 5 2 obc1-117::ISlacZ hah-Tc

Population density (Log CFU/ml) 4 0 12 24 0 12 24 28 48 Time (hours) Time (hours)

Fig. 4. Massive population crashes owing to alkaline toxicity in obcAB and obc1 B. glumae and B. thailandensis mutants. (A) Organization of obcA and obcB genes in B. glumae. Vertical bars indicate the positions and orientations of the Tn3-gusA insertions. Oxalate production is indicated above the restriction map. N, NotI; E, EcoRI; H, HindIII. (B–E) Oxalate concentration (B and D) and cell numbers (blue) and pH (red) (C and E) in cultures grown as described in Figs. 1 and 2. Strains included the WT B. glumae BGR1 and B. thailandensis E264 (blue circles), BOBA83 (B. glumae BGR1 obcA::Tn3-gusA83) and BT00401 (B. thailandensis E264 obc1-117::ISlacZ hah-Tc) (red triangles), and BOBB29 (B. glumae BGR1 obcB::Tn3-gusA29) (orange squares). Error bars represent the error ranges of experiments performed in triplicate.

Bp82 and CM135 were grown in LB broth with 1.2 mM adenine sulfate and using pobc-F (5′-GAACGGCCCTCTCTC TATGG-3′) and pobcR1 (5′-ACAT- 0.005% thiamine hydrochloride and on unsupplemented LB agar. C8-HSL was TCGGCGACTTATTTCCC-3′) primers. The PCR product was labeled with supplemented to a final concentration of 1 μMor4μM at the beginning of each biotin using a Pierce Lightshift Chemiluminescent Electrophoretic Mo- growth experiment, as indicated. bility Shift Assay Kit, in accordance with the manufacturer’s instructions. For nonspecific competitor DNA, the 242-bp upstream region of katE Ammonia and Oxalate Measurements. A gas-sensing ammonia ion-selective was amplified as described previously (11). Purified QsmR-His (200 nM) electrode (Thermo Scientific) was used to detect the gas phase of am- was incubated with 3 nM of biotin-labeled DNA in binding buffer [50 mM monia after addition of the alkaline reagent to the culture medium. Tris-HCl (pH 8.0), 150 mM NaCl, and 0.1 mM EDTA] for 15 min at 28 °C. Oxalate was measured using an oxalate assay kit (Libios) in accordance Unlabeled target DNA (30 nM) and a nonspecific unlabeled competitor with the manufacturer’s instructions (38). In brief, oxalate was converted DNA (3 nM) were added to the binding reaction. The reaction mixtures to carbon dioxide and hydrogen peroxide by oxalate oxidase, and hy- were separated on a nondenaturing 4% (vol/vol) polyacrylamide gel and drogen peroxide was measured by reaction with 3-(dimethylamino) transferred to nitrocellulose membranes, followed by detection with benzoic acid to form a blue-colored compound and catalyzed by per- streptavidin-HRP chemiluminescence (Pierce), in accordance with the man- oxidase, and the absorbance at 590 nm was measured. ufacturer’s instructions.

Electrophoretic Mobility Shift Assays. QsmR-His was puriﬁed as described ACKNOWLEDGMENTS. This work was supported by the Creative Re- previously (11). The 216-bp upstream region of obcAB was PCR-ampliﬁed search Initiatives Program of the National Research Foundation of Korea

Goo et al. PNAS Early Edition | 5of6 Downloaded by guest on October 1, 2021 A 900 B - + + + 0.2 μM QsmR-His 800 obcA::Tn3-gusA83 + + + + 3 nM labeled target promoter region - - + - 30 nM unlabeled target promoter region 700 - - - + 3 nM unlabeled katE 600 obcB::Tn3-gusA29

U/min/CFU) 500 -10 labeled DNA + QsmR-His 400

300

200

GUS activity (10 100 Free DNA 0 WT tofI:: qsmR:: tofI:: WT tofI:: qsmR:: tofI:: + C8-HSL + C8-HSL

Fig. 5. Regulation of obcAB genes by QsmR in B. glumae.(A) Expression of a chromosomal obcAB-gusA transcriptional fusion in B. glumae requires TofI and QsmR. Where indicated, synthetic C8-HSL was added to a ﬁnal concentration of 1 μM. (B) Electrophoretic mobility shift assay indicating binding of QsmR-His to the obc promoter region. This assay was performed using 0.2 μM QsmR-His, 3 nM labeled target promoter DNA, 30 nM unlabeled target promoter DNA, and 3 nM unlabeled katE promoter DNA. Error bars represent the error ranges of experiments performed in triplicate.

(Grant 2010-0018280, to I.H.) and National Institute of Allergy and In- ter of Excellence for Biodefense and Emerging Infectious Diseases (to fectious Disease Award U54AI057141 to the Northwest Regional Cen- E.P.G.).

1. Fuqua C, Greenberg EP (2002) Listening in on bacteria: Acyl-homoserine lactone 21. Diggle SP, Gardner A, West SA, Griffin AS (2007) Evolutionary theory of bacterial signalling. Nat Rev Mol Cell Biol 3(9):685–695. quorum sensing: When is a signal not a signal? Philos Trans R Soc Lond B Biol Sci 362 2. Parsek MR, Greenberg EP (2005) Sociomicrobiology: The connections between quo- (1483):1241–1249. rum sensing and biofilms. Trends Microbiol 13(1):27–33. 22. Williams P, Winzer K, Chan WC, Cámara M (2007) Look who’s talking: Communication 3. Ng W-L, Bassler BL (2009) Bacterial quorum-sensing network architectures. Annu Rev and quorum sensing in the bacterial world. Philos Trans R Soc Lond B Biol Sci 362 – Genet 43:197 222. (1483):1119–1134. 4. Brown SP, Johnstone RA (2001) Cooperation in the dark: Signalling and collective 23. Keller L, Surette MG (2006) Communication in bacteria: An ecological and evolu- action in quorum-sensing bacteria. Proc Biol Sci 268(1470):961–965. tionary perspective. Nat Rev Microbiol 4(4):249–258. 5. Diggle SP, Griffin AS, Campbell GS, West SA (2007) Cooperation and conflict in 24. West SA, Griffin AS, Gardner A, Diggle SP (2006) Social evolution theory for micro- quorum-sensing bacterial populations. Nature 450(7168):411–414. organisms. Nat Rev Microbiol 4(8):597–607. 6. Studer SV, Mandel MJ, Ruby EG (2008) AinS quorum sensing regulates the Vibrio fi- 25. Van Houdt R, Aertsen A, Michiels CW (2007) Quorum-sensing-dependent switch to scheri acetate switch. J Bacteriol 190(17):5915–5923. butanediol fermentation prevents lethal medium acidification in Aeromonas hydro- 7. Chandler JR, Heilmann S, Mittler JE, Greenberg EP (2012) Acyl-homoserine lactone- – dependent eavesdropping promotes competition in a laboratory co-culture model. phila AH-1N. Res Microbiol 158(4):379 385. ISME J 6:1336–1344. 26. Grignon C, Sentenac H (1991) pH and ionic conditions in the apoplast. Annu Rev Plant – 8. Smith RS, Iglewski BH (2003) P. aeruginosa quorum-sensing systems and virulence. Physiol Plant Mol Biol 42:103 128. Curr Opin Microbiol 6(1):56–60. 27. Nachin L, Barras F (2000) External pH: An environmental signal that helps to rationalize 9. Darch SE, West SA, Winzer K, Diggle SP (2012) Density-dependent fitness benefits in pel gene duplication in Erwinia chrysanthemi. Mol Plant Microbe Interact 13(8):882–886. quorum-sensing bacterial populations. Proc Natl Acad Sci USA 109(21):8259–8263. 28. Byers JT, Lucas C, Salmond GPC, Welch M (2002) Nonenzymatic turnover of an Erwinia 10. Kim J, et al. (2004) Quorum sensing and the LysR-type transcriptional activator ToxR carotovora quorum-sensing signaling molecule. J Bacteriol 184(4):1163–1171. regulate toxoflavin biosynthesis and transport in Burkholderia glumae. Mol Microbiol 29. Yates EA, et al. (2002) N-acylhomoserine lactones undergo lactonolysis in a pH-, 54(4):921–934. temperature-, and acyl chain length-dependent manner during growth of Yersinia 11. Kim J, et al. (2007) Regulation of polar flagellum genes is mediated by quorum pseudotuberculosis and Pseudomonas aeruginosa. Infect Immun 70(10):5635–5646. sensing and FlhDC in Burkholderia glumae. Mol Microbiol 64(1):165–179. 30. Coe FL, Evan A, Worcester E (2005) Kidney stone disease. J Clin Invest 115(10): 12. Goo E, Kang Y, Kim H, Hwang I (2010) Proteomic analysis of quorum sensing-de- 2598–2608. pendent proteins in Burkholderia glumae. J Proteome Res 9(6):3184–3199. 31. Baker PR, Cramer SD, Kennedy M, Assimos DG, Holmes RP (2004) Glycolate and 13. Chandler JR, et al. (2009) Mutational analysis of Burkholderia thailandensis quorum glyoxylate metabolism in HepG2 cells. Am J Physiol Cell Physiol 287(5):C1359–C1365. sensing and self-aggregation. J Bacteriol 191(19):5901–5909. 32. Green MA, Fry SC (2005) Vitamin C degradation in plant cells via enzymatic hydrolysis 14. Ulrich RL, Hines HB, Parthasarathy N, Jeddeloh JA (2004) Mutational analysis and of 4-O-oxalyl-L-threonate. Nature 433(7021):83–87. biochemical characterization of the Burkholderia thailandensis DW503 quorum- 33. Williams B, Kabbage M, Kim HJ, Britt R, Dickman MB (2011) Tipping the balance: – sensing network. J Bacteriol 186(13):4350 4360. Sclerotinia sclerotiorum secreted oxalic acid suppresses host defenses by manipulating 15. Ulrich RL, et al. (2004) Role of quorum sensing in the pathogenicity of Burkholderia the host redox environment. PLoS Pathog 7(6):e1002107. pseudomallei. J Med Microbiol 53(Pt 11):1053–1064. 34. Chen C, Sun Q, Narayanan B, Nuss DL, Herzberg O (2010) Structure of oxalacetate 16. Sezonov G, Joseleau-Petit D, D’Ari R (2007) Escherichia coli physiology in Luria-Bertani acetylhydrolase, a virulence factor of the chestnut blight fungus. J Biol Chem 285(34): broth. J Bacteriol 189(23):8746–8749. 26685–26696. 17. McFall E, Newman EB (1996) Amino acids as carbon sources. Escherichia coli and 35. Rogul M, Carr SR (1972) Variable ammonia production among smooth and rough Salmonella: Cellular and Molecular Biology, eds Neidhardt FC, et al. (American Society for Microbiology, Washington, DC), 2nd Ed, pp 358–379. strains of Pseudomonas pseudomallei: Resemblance to bacteriocin production. J – 18. Nakata PA, He C (2010) Oxalic acid biosynthesis is encoded by an operon in Bur- Bacteriol 112:327 380. kholderia glumae. FEMS Microbiol Lett 304(2):177–182. 36. Kasai T, Suzuki I, Asai T (1963) Glyoxylate oxidation in Acetobacter with reference to 19. Li HQ, Matsuda I, Fujise Y, Ichiyama A (1999) Short-chain acyl-CoA-dependent production the formation of oxalic acid. J Gen Appl Microbiol 9:49–58. of oxalate from oxaloacetate by Burkholderia glumae, a plant pathogen which causes 37. Singh R, et al. (2009) An ATP and oxalate generating variant tricarboxylic acid cycle grain rot and seedling rot of rice via the oxalate production. JBiochem126(1):243–253. counters aluminum toxicity in Pseudomonas fluorescens. PLoS ONE 4(10):e7344. 20. Nakata PA (2011) The oxalic acid biosynthetic activity of Burkholderia mallei is en- 38. Laker MF, Hofmann AF, Meeuse BJ (1980) Spectrophotometric determination of coded by a single locus. Microbiol Res 166(7):531–538. urinary oxalate with oxalate oxidase prepared from moss. Clin Chem 26(7):827–830.

6of6 | www.pnas.org/cgi/doi/10.1073/pnas.1218092109 Goo et al. Downloaded by guest on October 1, 2021