Karrikins from plant smoke modulate bacterial quorum sensing

Article abstract of DOI:10.1039/c3cc47501h

The discovery that plant smoke contains germination stimuli has led to the identification of a new class of signaling molecules named karrikins. Here we report a potential second role for these molecules: in various bacterial species-A. tumefaciens, P. aeruginosa and V. harveyi-they modulate bacterial quorum-sensing (QS), with very different outcomes. the Partner Organisations 2014.

Full text of DOI:10.1039/c3cc47501h

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Karrikins from plant smoke modulate bacterial
quorum sensing†
Cite this: Chem. Commun., 2014,
50, 5322
Aviad Mandabi, Hadas Ganin, Pnina Krief, Josep Rayo and Michael M. Meijler*
Received 30th September 2013,
Accepted 7th November 2013
DOI: 10.1039/c3cc47501h
www.rsc.org/chemcomm
The discovery that plant smoke contains germination stimuli has led to
the identification of a new class of signaling molecules named karrikins.
Here we report a potential second role for these molecules: in various
bacterial species – A. tumefaciens, P. aeruginosa and V. harveyi – they
modulate bacterial quorum-sensing (QS), with very different outcomes.
Nature has evolved many different mechanisms to modulate cross-talk
between different organisms (e.g. animals, plants, bacteria, fungi) and
most of these are based on secretion and recognition of small signaling
molecules.1–3 It was discovered recently that naturally occurring bute-
nolides, derived from the smoke of burnt plant material, stimulate seed
germination in a wide range of plant species. One specific family of
compounds, the karrikins, was identified and characterized as active
compounds that promote this intriguing phenomenon.4–7 While cur-
rent studies focus on the mechanism and mode of action of karrikins
in plants, we decided to explore whether these compounds are able to
affect bacterial group behavior especially in QS systems. QS describes
Fig. 1 Karrikins and structural similarities with known QS molecules and
inhibitors. P. aeruginosa AIs: N-(3-oxododecanoyl)-L-homoserine lactone
(3-oxo-C12-HSL), N-butyryl-L-homoserine lactone (C4-HSL). A. tumefaciens
AI: N-(3-oxooctanoyl)-L-homoserine lactone (3-oxo-C8-HSL). P. syringae
AI: N-(3-oxohexanoyl)-L-homoserine lactone (3-oxo-C6-HSL). V. harveyi
AIs: AI-1, N-(3-hydroxybutanoyl)-L-homoserine lactone (3-OH-C4-HSL);
AI-2, (S)-4,5-dihydroxy-2,3-pentanedione (DPD).
the mechanism used by a population of microorganisms to act as a Another hypothesis – not mutually exclusive with the first – is that
single multicellular organism in a cell-density dependent manner karrikins are used by the plants to manipulate certain bacteria to
through secretion and sensing of small diffusible molecules, enabling their advantage in a post-fire environment, to provide them nutrition,
intercellular communication leading to synchronized gene expression.8 protection from other pathogens, while repressing virulence factor
Given the known ubiquitous interactions between plants and bacteria production and pathogenicity. It is known that plants recognize
and the structural similarities between karrikins and certain QS certain QS molecules, such as the primary P. aeruginosa AI, 3-oxo-
molecules (e.g. short chain AHLs and AI-2, Fig. 1) as well as QS C12-HSL, which influence gene expression in plants and trees.13,16
inhibitors (e.g. patulin, Fig. 1),9–15 we chose to examine potential
By studying the effects of these molecules on bacteria we
interactions between karrikins and three different bacterial species. hope to gain a better understanding of the role of karrikins in
Interestingly, the lactone moiety of karrikins resembles one class of nature and more general principles of chemical guidance of
signaling molecules, of Gram-negative bacteria, autoinducers-1 (AI-1) coexistence and warfare.
and the pyran moiety resembles a second class of QS molecules,
autoinducers-2 (AI-2) (Fig. 1).
Here, we report the effects of two members of the karrikin family,
KAR1 and KAR2, on QS systems of different types of bacteria: (i) a
We hypothesize that in a post-fire environment, if a plant can plant pathogen, Agrobacterium tumefaciens;2,3,14,16 (ii) an opportunistic
sense these molecules to their advantage (more space, less competi- pathogen, Pseudomonas aeruginosa, that is able to infect both humans
tion) to promote germination it is reasonable to assume that bacteria and plants;12,13 (iii) V. harveyi a luminescent marine bacterium and
have developed means to sense this opportunity to proliferate. opportunistic pathogen of marine animals.
The synthesis of KAR1 and KAR2 was performed following
procedures described by Goddard-Borger et al. with minor modifica-
Department of Chemistry and The National Institute for Biotechnology in the Negev,
tions (Scheme S1, ESI†).17 We first examined the effects of the
Ben-Gurion University of the Negev, Beer-Sheva, Israel. E-mail: meijler@bgu.ac.il
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c3cc47501h karrikins on P. aeruginosa wild-type strain PAO1-lux, containing the
5322 | Chem. Commun., 2014, 50, 5322--5325
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
luminescent reporter gene luxCDABE cloned downstream of lasI.18
We also examined a potential agonistic effect, using the P. aeruginosa
PAO-JP2 (PAO1-lux lasIÀ, rhlIÀ) strain,18 which is unable to produce
N-(3-oxododecanoyl)-L-homoserine lactone (3-oxo-C12-HSL) and N-
butanoyl-L-homoserine lactone (C4-HSL). No effects were observed
for both compounds in the two reporter assays, and we conclude
that KAR1/KAR2 does not interact with the las system in P. aerugi-
nosa, which is the higher in hierarchy between the two primary QS
systems in P. aeruginosa. We then investigated the rhl system of
P. aeruginosa, using the luminescent reporter strain PAO-JP2 (pKD-rhlA)
(PAO-JP2 with the rhlA promoter fused upstream of the luxCDABE
operon).18 Interestingly, KAR1/KAR2 showed concentration depen-
Fig. 3 Activation of QS by different concentrations of KAR1/2, in the absence
of 3-oxo-C8-AHL in A. tumefaciens A136 pCF218 pMV26 after 15 hours.
dent inhibition of the rhl system in the presence of 10 mM C4-HSL for an opportunity to colonize new plants, though an argument
(Fig. 2a). In order to examine the physiological relevance of rhl against this hypothesis would be that the time scale of new plant
inhibition in wild-type P. aeruginosa, pyocyanin assays were con- growth is not compatible with such a signaling scenario – although
ducted, as the production of this important virulence factor is one may argue that forest fire smoke can linger for days. Still, the
controlled in part by RhlR. Addition of KAR1/KAR2 to P. aeruginosa maximal QS activation in this strain was roughly 50-fold less for the
strain PAO1 (wild-type) reduced production of pyocyanin (Fig. S1, karrikins compared to the natural autoinducer 3-oxo-C8-HSL,
ESI†). Interestingly, when we tested the effects of the karrikins on the prompting us to examine whether karrikins might also serve as QS
quorum sensing system of the related plant pathogen P. syringae inhibitors through partial agonism. Employing the same experi-
(Fig. S2, ESI†),19 which employs 3-oxo-C6-HSL as its autoinducer, we mental conditions used for the agonist assay, but with added synthetic
observed neither agonist nor antagonist effects – suggesting that the 3-oxo-C8-HSL (400 pM), we observed no modulatory effects for the
response to KAR1/2 is highly specific, and the QS inhibitory effects in karrikins. These results suggest that the affinity of 3-oxo-C8-HSL to the
P. aeruginosa appear to be based on direct competition with C4-HSL primary QS receptor TraR is much higher compared with KAR1/KAR2.
for binding to RhlR (Fig. 2a).
Still, KAR1/KAR2 may very well interact with an unknown protein that
We then examined whether the plant pathogen A. tumefaciens can regulate the activation of QS in A. tumefaciens.
would be affected by karikkins. We used a luminescent reporter
Next, we examined the effects of KAR1/2 on the AI-2 signaling
strain, A. tumefaciens A136 pCF218 pMV26 (lacking Ti plasmid, TraR pathway, as given the putative importance of AI-2 based interspecies
response regulator, traIÀ and TraI promoter fused to luxCDABE).20 signaling.3,21–24 We used modified strains of V. harveyi, BB170
Addition of KAR1/2 at different concentrations in the absence of (luxNÀ, AI-1 receptor),23 and MM32 (luxNÀ, luxSÀ),21,24 which
N-(3-oxooctanoyl)-L-homoserine lactone (3-oxo-C8-HSL) resulted in respond only to the presence of AI-2. While potential effects of plant
activation of the QS response cascade. Both karikkins served as smoke on marine bacteria would be far fetched at best, and
agonists in a concentration dependent manner (Fig. 3). These results structural similarities between karrikins and AI-2 are not strong,
suggest that KAR1/2 may be sensed by A. tumefaciens as early alerts we did observe a clear synergistic effect in a concentration dependent
manner (Fig. 4a) for KAR1/2 on BB170. In addition, we added KAR1/
2 to V. harveyi strain MM32 in the absence of exogenous AI-2. No
response was observed, suggesting that KAR1/2 does not act as
agonist in these bacteria. However, upon addition of synthetic AI-2
(133 nM) the same synergistic effect was observed as in BB170
(Fig. 4b), suggesting that KAR1/KAR2 is affecting the AI-2 signaling
pathway in some manner, for instance through interaction with the
AI-2 receptor, LuxP. Alternatively, KAR1/KAR2 may be recognized by
an unknown receptor, which can affect AI-2 induced gene expres-
sion, the biosynthesis of AI-2 or both.
Finally, we examined the effects of karrikins in vivo and studied
their physiological relevance with regard to bacterial pathogenesis in
plants. Blackwell and coworkers showed that QS inhibitors effectively
inhibited virulence of Pectobacterium carotovera in bean and potato rot
models.25 We conducted two experiments, based on P. aeruginosa plant
infection models described by Rahme and coworkers.26 We focused on
the activity of KAR1 in P. aeruginosa, as the QS inhibitory or agonist
effects in the other bacteria were less pronounced, and we tested the
effect of KAR1 on the infection of Arabidopsis thaliana plants and lettuce
midriffs. While KAR1 did not prevent infection of living plants, we did
Fig. 2 (a) Inhibition of QS by KAR1/KAR2, in P. aeruginosa PAO-JP2 (pKD-
rhlA) in the presence of 10 mM C4-HSL, after 6 hours. At 200 mM and
observe a significant reduction in loss of plant leaves from 24 h
beyond KAR1 caused some growth inhibition. (b) Inhibition of pyocyanin
production in P. aeruginosa strain PAO1 by KAR1 and KAR2.
to 48 h post-infection (Fig. 5), in the presence of 100 mM KAR1
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 5322--5325 | 5323

View Article Online
ChemComm
Communication
answer, especially since the origin of karrikins is still unknown.
However, here we provide evidence of the ability of two molecules of
the karrikin family, KAR1/2, to affect QS in three different bacteria,
two of which (P. aeruginosa, A. tumefaciens) are known to interact
with plants/trees in nature. While in P. aeruginosa a clear QS
antagonist effect was observed on the rhl system, which controls
the expression of pyocyanin, in A. tumefaciens we only measured a
mild agonist effect. Although the marine bacterium V. harveyi is not
likely to encounter karrikins in nature, the results were interesting
given the proposed interspecies signaling role ascribed to AI-2.
The ability of KAR1/2 to activate or inhibit QS pathways might
reveal a new type of interkingdom communication. Further studies
on the activity and mechanisms of action of these molecules are
needed in order to answer more fundamental questions. Although
we determined that KAR1/2 most likely interferes with P. aeruginosa
QS through inhibition of the rhl system, direct identification of
proteins that bind KAR1/2 will give us further insight from a
mechanistic point of view. The presence, localization and identity
of KAR1/2 receptor in bacteria are currently under investigation.
We would like to thank Eilon Shani, Mark Estelle and Pieter
Dorrestein (UCSD) for their advice and support, and E. P. Greenberg,
B. L. Bassler, M. G. Surette, P. A. Sokol and S. E. Lindow for
generously providing bacterial strains. This work was supported
by the European Research Council (Starting Grant 240356, MMM).
Fig. 4 (a) Activation of QS by synergism by different concentrations of
KAR1/2 in V. harveyi BB170. (b) Activation of QS by synergism by different
concentrations of KAR1/2 in the presence of exogenous synthetic 133 nM
AI-2 in V. harveyi MM32. Above 200 mM KAR1 was toxic.
Notes and references
1 H. Chung, S. J. Pamp, J. A. Hill, N. K. Surana, S. M. Edelman,
E. B. Troy, N. C. Reading, E. J. Villablanca, S. Wang, J. R. Mora,
Y. Umesaki, D. Mathis, C. Benoist, D. A. Relman and D. L. Kasper,
Cell, 2012, 149, 1578.
2 J. E. Gonzalez and N. D. Keshavan, Microbiol. Mol. Biol. Rev., 2006,
70, 859.
3 M. B. Miller and B. L. Bassler, Annu. Rev. Microbiol., 2001, 55, 165.
4 S. D. S. Chiwocha, K. W. Dixon, G. R. Flematti, E. L. Ghisalberti,
D. J. Merritt, D. C. Nelson, J. A. M. Riseborough, S. M. Smith and
J. C. Stevens, Plant Sci., 2009, 177, 252.
5 K. W. Dixon, D. J. Merritt, G. R. Flematti and E. L. Ghisalberti, Acta
Hortic., 2009, 813, 155.
6 G. R. Flematti, E. L. Ghisalberti, K. W. Dixon and R. D. Trengove,
Science, 2004, 305, 977.
7 D. C. Nelson, J. A. Riseborough, G. R. Flematti, J. Stevens, E. L.
Ghisalberti, K. W. Dixon and S. M. Smith, Plant Physiol., 2009, 149, 863.
8 W. C. Fuqua, S. C. Winans and E. P. Greenberg, J. Bacteriol., 1994,
176, 269.
9 A. C. Hayward, Annu. Rev. Phytopathol., 1991, 29, 65.
10 J. Loh, D. P. Lohar, B. Andersen and G. Stacey, J. Bacteriol., 2002,
184, 1759.
Fig. 5 Effect of KAR1 on infection of Arabidopsis thaliana Colombia by
P. aeruginosa wild type strain PAO1; upper panel, left: with 100 mM KAR1, after
24 h, and right: after 48 h; lower panel, left: without KAR1, after 24 h, and right:
after 48 h. The bar graphs reflect the ratio of average leaf area between 24 and
48 h. Plant leaf surface areas were calculated using ImageJ 1.47t.
11 J. T. Loh, J. P. Yuen-Tsai, M. G. Stacey, D. Lohar, A. Welborn and
G. Stacey, Mol. Microbiol., 2001, 42, 37.
12 S. T. Schenk, E. Stein, K. H. Kogel and A. Schikora, Plant Signaling
Behav., 2012, 7, 178.
13 E. K. Shiner, K. P. Rumbaugh and S. C. Williams, FEMS Microbiol.
Rev., 2005, 29, 935.
14 S. B. Von Bodman, W. D. Bauer and D. L. Coplin, Annu. Rev.
Phytopathol., 2003, 41, 455.
15 T. B. Rasmussen, M. E. Skindersoe, T. Bjarnsholt, R. K. Phipps, K. B.
Christensen, P. O. Jensen, J. B. Andersen, B. Koch, T. O. Larsen, M.
Hentzer, L. Eberl, N. Hoiby and M. Givskov, Microbiology, 2005, 151, 1325.
16 J. Zhu, P. M. Oger, B. Schrammeijer, P. J. Hooykaas, S. K. Farrand
and S. C. Winans, J. Bacteriol., 2000, 182, 3885–3895.
(36 Æ 11% vs. 65 Æ 8%, p o 0.03), indicating that the karrikin
mediated reduction in P. aeruginosa virulence slows down plant loss.27
This hypothesis was further strengthened by a slightly reduced induc-
tion of soft rot in lettuce midriffs upon infection with P. aeruginosa in
the presence of increasing amounts of KAR1 (Fig. S2, ESI†).
In this study we focused on the potential role of the karrikin
family in the modulation of QS (agonist, antagonist or synergism) in
different bacterial species. Whether the presence of karrikins, in the
post-forest-fire environment, simultaneously influences plant seed 17 E. D. Goddard-Borger, E. L. Ghisalberti and R. V. Stick, Eur. J. Org.
Chem., 2007, 3925.
18 K. Duan and M. G. Surette, J. Bacteriol., 2007, 189, 4827.
19 G. Dulla and S. E. Lindow, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 3082.
germination and bacterial group behavior (e.g. suppression of
virulence factors, nutrient production, antibiotic production) for
the benefit of habitat rehabilitation is still a major question to 20 S. P. Bernier, A. L. Beeston and P. A. Sokol, BMC Biotechnol., 2008, 8, 59.
5324 | Chem. Commun., 2014, 50, 5322--5325
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
21 B. L. Bassler, E. P. Greenberg and A. M. Stevens, J. Bacteriol., 1997, 24 K. Winzer, K. R. Hardie, N. Burgess, N. Doherty, D. Kirke,
179, 4043.
M. T. Holden, R. Linforth, K. A. Cornell, A. J. Taylor, P. J. Hill and
P. Williams, Microbiology, 2002, 148, 909.
22 S. T. Miller, K. B. Xavier, S. R. Campagna, M. E. Taga,
M. F. Semmelhack, B. L. Bassler and F. M. Hughson, Mol. Cell, 25 A. G. Palmer, E. Streng and H. E. Blackwell, ACS Chem. Biol., 2011, 6, 1348.
2004, 15, 677. 26 M. Starkey and L. G. Rahme, Nat. Protocols, 2009, 4, 117.
23 S. Schauder, K. Shokat, M. G. Surette and B. L. Bassler, Mol. 27 Administration of KAR1 alone – without bacteria – does not affect
Microbiol., 2001, 41, 463.
plant leaf surface area significantly (see Fig. S3, ESI†).
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 5322--5325 | 5325