Current Microbiology

, Volume 59, Issue 4, pp 469–474

Interference of Cranberry Constituents in Cell–Cell Signaling System of Vibrio harveyi


  • Mark Feldman
    • Institute of Dental Sciences, Faculty of Dental MedicineHebrew University-Hadassah
    • Department of Prosthodontics, Faculty of Dental MedicineHebrew University-Hadassah
  • Ervin I. Weiss
    • Department of Prosthodontics, Faculty of Dental MedicineHebrew University-Hadassah
  • Itzhak Ofek
    • Department of Human Microbiology, Sackler Faculty of MedicineTel Aviv University
    • Institute of Dental Sciences, Faculty of Dental MedicineHebrew University-Hadassah

DOI: 10.1007/s00284-009-9462-3

Cite this article as:
Feldman, M., Weiss, E.I., Ofek, I. et al. Curr Microbiol (2009) 59: 469. doi:10.1007/s00284-009-9462-3


Cranberry juice has long been recognized in folk medicine as a therapeutic agent, mainly in urinary track infections. It acts as an antibiofilm agent against various pathogens. Quorum sensing is process where bacteria communicate with each other via signal molecules known as autoinducers. This process is strongly involved in various bacterial pathological and physiological pathways. Various strains of Vibrio harveyi bacteria were incubated with different concentrations of nondialyzable material of cranberry (NDM) with or without addition of exogenous autoinducer. Bioluminescence regulated by the autoinducers was measured in GENios reader. Effect of NDM alone or NDM supplemented with autoinducer on quorum sensing was determined as change in bioluminescence in each treated sample compared to appropriate control in every strain. Using model of V. harveyi, we found an inhibitory effect of cranberry constituents on bacterial signaling system. This effect was reversible, since exogenous autoinducer was able to recover bioluminescence which was decreased by NDM. We hypothesized that cranberry NDM interacts with V. harveyi quorum sensing by competition with autoinducer for binding to autoinducer sensor.


Bacterial cell–cell communication (quorum sensing) is associated with bacterial density and regulates various bacterial properties such as biofilm formation, virulence, genetic competence, and others. This process is mediated by signal molecules known as autoinducers (AIs). The bacterium Vibrio harveyi has been proposed as a model for investigating quorum sensing related to AI-1, AI-2 [3]. V. harveyi produces AIs that regulate its bioluminescence: species-specific acylated homoserine lactone (AHL) termed as AI-1, furanons based AI termed as AI-2 and the recently described CAI-1 (V. cholerae AI) [13].

Cranberry constituents have been shown to affect pathogenesis and physiological pathways of various types of bacteria. Antiadhesion activity of cranberry is manifested by decreasing length of P-fimbriae of E. coli [18] as well as by affecting adhesion of H. pylori [8, 25]. Studies demonstrated the inhibitory effect of cranberry juice and its constituents on mutans streptococci extracellular enzymes lead to impaired biofilm formation [10, 15]. Additional studies indicated that the antibiofilm properties of NDM is also mediated by inhibition of the activity, protein secretion, and gene expression of extracellular enzymes fructosyltransferase (FTF) [11] as well as enzymatic activity of glucosyltransferases (GTF) of the cariogenic mutans streptococci [19]. Both, GTF and FTF are crucial factors in dental biofilm formation and maturation. Moreover, NDM has the potential to inhibit proteinases of various bacteria associated in pathogenesis of periodontal diseases, as Porphyromonas gingivalis, Tannerella forsythia, and Treponema denticola [6]. By inhibition of periopathogenes proteinases, cranberry polyphenols may prevent destructive process occurring in periodontitis [24]. Cranberry extract has been tested also in variety of disorders as an antiviral [23] and an anticancer [14] agent.

In this study, we tested the hypothesis that the antibiofilm effect of cranberry NDM is also associated with influence on bacterial quorum sensing.

Materials and Methods

Preparation of NDM

NDM was prepared as described previously [21, 22]. In brief, concentrated juice from the American cranberry Vaccinium macrocarpon was obtained from Ocean Spray, Inc. (Lakeville-Middleboro, MA, USA). The juice was exhaustively dialyzed for 4 days at 4°C against distilled water in 15,000 MW cutoff dialysis bags and lyophilized. The high molecular weight, nondialyzable material, designated NDM, exhibits tannin-like properties, is highly soluble in water, devoid of proteins, carbohydrates, and fatty acids, and contains 56.6% carbon and 4.14% hydrogen [22]. Using the same source of Cranberry extract from Ocean Spray, Inc. and similar isolation method as we did, it was found by Labrecque et al. [16] that this fraction of NDM contains 0.35% anthocyanins (0.055% cyanidin-3-galactoside, 0.003% cyanidin-3-glucoside, 0.069% cyanidin-3-arabinoside, 0.116% peonidin-3-galactoside, 0.016% peonidin-3-glucoside, and 0.086% peonidin-3-arabinoside) and 65.1% proanthocyanidins as determined by colorimetric assays. Because such colorimetric methods are specific for phenolic compounds rather than for a specific phenol containing compound such as PAC, it was suggested that at least two-thirds of NDM is composed of a unique molecular form or segments of phenolic compounds that are distinct from oligomeric PAC (e.g., tannins) and that exhibits potent antiadhesion activity. Further studies are required to resolve such a complex hypothetical structure.

Bacterial Strains

Vibrio harveyi mutant strains, defected in their sensor 1 or 2 or both and wild type (Table 1) (kindly provided by B. Bassler, Princeton University), were used in order to examine the effect of NDM on quorum sensing.
Table 1

Bacterial strains used in this study

V. harveyi strain

Relevant genotype

Relevant phenotype


Wild type

Wild type


(BB7) luxP::Tn5

Sensor-1+, sensor-2; AI-1+, AI-2+


(BB7) luxN::Tn5

Sensor-1, sensor-2+; AI-1+, AI-2+

JAF 375

(BB 960) luxN::Cmr

Constitutive luminescence sensor-1, sensor-2; AI-1+, AI-2+

At the tested concentrations of NDM bacterial growth and viability was not significantly affected (data not shown).

Effect of NDM on Quorum Sensing

In order to evaluate the effect of NDM on the AI cascade, overnight cultures of V. harveyi mutants, BB886 [4] and BB170 [3] lacking the sensor for AI-2 (S2) or sensor for AI-1 (S1), respectively, as well as wild type (BB120) [5] were grown for 16 h to cell density 1.5–2.5 × 109 CFU/ml as previously described [2], diluted 1:10,000 in fresh AB medium [3] and further incubated with NDM at concentrations of 3.2 and 16 μg/ml. Bioluminescence was recorded continuously and measured every 30 min in GENios reader (TECAN, Austria) at 30°C for duration of 24 h. The cranberry compound at above concentrations supplemented with AB medium had no effect on bioluminescence.

The luminescence emitted during growth of bacterial culture was measured with the GENios reader and the intensity of light emission was calculated relatively to the turbidity of the culture (OD595). The data obtained between 3.5 and 8 h of bacterial incubation was calculated as the area under the curve (AUC) of each sample of all strains and presented as percentage of bioluminescence compared with control of untreated sample (100%).

Introduction of Exogenous AIs to NDM-Treated Bacteria

Further experiment was designed to investigate the mode of NDM’s action on quorum sensing. The above strains of V. harveyi were grown as described above. After reaching cell density of 1.5–2.5 × 109 CFU/ml, the inoculum was diluted as described before and bacteria with absence or presence of above doses of NDM were incubated with 10% (v/v) and 30% (v/v) of spent medium containing AI-1 isolated from V. harveyi MM30 (AI-1+, AI-2) or AI-2 isolated from V. harveyi BB152 (AI-1, AI-2+) prepared as described previously [3].

Spent medium of nonproducing AI mutant MM77 (AI-0) was used as base level luminescence control at each dose of NDM. Bioluminescence was recorded as described above and data are presented as percentage of bioluminescence in each tested sample compared to control without NDM but with the addition of appropriate exogenous AI. NDM at the tested concentrations, supplemented with either AI-1 or AI-2 containing medium, had no effect on bioluminescence.

Little change in bioluminescence was recorded in the first 2–3 h of incubation and after 8 h of incubation the bioluminescence of all samples reached the plato level. Therefore, the period of incubation between 3.5 and 8 was chosen for calculation of NDM’s effect on quorum sensing since most of the changes in bioluminescence were observed during this period.

Interaction of NDM with Signal Molecules

In this experiment, we checked the ability of NDM to bind to AI molecule, thus forming nondialyzable NDM-AI complexes, which should decrease the amount of free AI. Briefly, AI-1/2 containing medium at 10% (v/v) and 30% (v/v) were preincubated without NDM as control or with NDM = 16 μg/ml in 3,500 MWCO dialysis tubing (Pierce, Rockford, IL, USA) during 4 h at 30°C. The chosen cutoff dialysis tube allows small molecules of AI-1 (158 Da) or AI-2 (187 Da) to pass thru, while high molecular weight NDM (14,000 Da) or complex NDM-AI are unable to penetrate thru the dialysis tubing. The dialyzed liquids of each sample, collected at time points: 0, 1, 2, 3, 4 h of dialysis were incubated with V. harveyi S2 or S1 mutant strains. Bioluminescence was recorded similarly to the method described above and data are presented as percentage of bioluminescence in each sample treated with dialyzed mixture of AI-1/AI-2 with NDM and compared to the samples treated with AI-1/AI-2 alone.


Results in Fig. 1a, b demonstrate the effect of NDM on bioluminescence of V. harveyi wild type (BB 120). NDM concentration of 3.2 μg/ml had minor influence on quorum sensing of the wild type. Increasing NDM’s dose to 16 μg/ml caused a more pronounced reduction of bacterial signal, up to 57% compare to control. Exposure of the bacteria to 30% (v/v) of exogenous AI-1 reversed that inhibitory influence of NDM (Fig. 1a). In contrast, the introduction of exogenous AI-2 caused no significant alteration to the inhibition effect of quorum sensing produced by NDM (Fig. 1b).
Fig. 1

Effect of 3.2 and 16 μg/ml of NDM on V. harveyi BB120 (wild type) bioluminescence with the addition of exogenous AI-1 (a) or AI-2 (b) containing medium. Samples were compared to controls of bacteria not exposed to NDM (100%). The data are expressed as the mean ± standard deviation of duplicates from three separate experiments. Values marked with an asterisk are significantly different from that of the nontreated control (P < 0.05, Student’s t test)

As shown in Fig. 2a, b, V. harveyi mutant strains defected in their quorum sensing sensors, S1 or S2, were much more sensitive to the influence of NDM cranberry constituents than the wild type, resulting in 74% inhibition in S2 mutant (Fig. 2a) and 84% inhibition in S1 mutant (Fig. 2b) at concentration of NDM = 16 μg/ml.
Fig. 2

Effect of 3.2 and 16 μg/ml of NDM on V. harveyi BB886 (sensor-1+, sensor-2; AI-1+, AI-2+) bioluminescence with the addition of exogenous AI-1 containing medium (a). Effect of 3.2 and 16 μg/ml of NDM on V. harveyi BB170 (sensor-1, sensor-2+; AI-1+, AI-2+) bioluminescence with the addition of exogenous AI-2 containing medium (b). Samples were compared to the appropriate controls of untreated with NDM bacteria (100%). The data are expressed as the mean ± standard deviation of duplicates from three separate experiments. Values marked with an asterisk are significantly different from that of the nontreated control (P < 0.05, Student’s t test)

As was recorded above with the wild type, also in the mutant strains, exogenous AI-1 demonstrated pronounced dose-dependent restoring effect on the bioluminescence repressed by NDM. The strain BB886 (S2) incubated with the NDM cranberry compound at 16 μg/ml and exogenous AI-1 at 10% (v/v) and 30% (v/v) recovered its bioluminescence by 42% and 68%, respectively, compared to the same bacteria exposed to the above dose of NDM alone (Fig. 2a). In contrast, this restoring influence was minor (9%) in the mutant strain BB170 (S1) when it was exposed to 30% (v/v) of exogenous AI-2 in the presence of NDM (Fig. 2b).

Preincubation of AI-1/AI-2 with NDM showed no significant difference in bioluminescence as compared to the effect of AI-1/AI-2 pre incubated alone (Fig. 3).
Fig. 3

Effect of AIs preincubated with NDM during 4 h of dialysis on bioluminescence of V. harveyi mutant strains: Effect of 10% (a) and 30% (b) of AI-1 preincubated with NDM on bioluminescence of BB886 mutant (sensor-1+, sensor-2; AI-1+, AI-2+); Effect of 10% (c) and 30% (d) of AI-2 preincubated with NDM on bioluminescence of BB170 mutant (sensor-1, sensor-2+; AI-1+, AI-2+)


Bacterial pathogenesis can be modified by several means, the most effective being a bactericidal approach. However, this approach has numerous undesirable side effects [12]. Inhibition of bacterial quorum sensing systems, rather than bactericidal or bacteriostatic strategies, may find application in many fields, such as medicine, agriculture, and food technology. This approach is beneficiary because it does not impose selective pressure for the development of resistance strains as with antibiotics, because quorum sensing is not directly involved in processes essential for growth of bacteria [26]. Several studies showed that natural agents also may inhibit quorum sensing of different bacteria at concentrations below their MIC [1, 7, 9, 20].

In this study, using model of V. harveyi strains, we explored the effect of cranberry NDM on quorum sensing as one of the potential antibiofilm modes of action of this compound.

Our observations demonstrate that NDM decreases bioluminescence during Log growth phase (3.5–8 h) of bacteria, probably due to interaction with bacterial cell surface receptors. It seems that cranberry NDM binds to AIs receptors in higher efficiency than either auto-secreted endogenous AI-1 or AI-2. In addition to its inhibitory effect, the beginning stage of bioluminescence induction of bacteria exposed to NDM at highest concentration of 16 μg/ml was delayed by 0.5 h in wild type and by 1.5 h in mutant strains compare to untreated control (data not shown). Once being attached to a quorum sensing sensor, the cranberry compound interferes in the autoinducer–receptor binding process, which is manifested in reduction of bioluminescence. Indeed, proanthocyanidins (PAC)—a major component of NDM may act as an extracellular mediator, interacting with bacterial surface proteins [17].

In this study, we also observed the difference in inhibition level between wild-type and bacterial strains defected in their quorum sensing sensors. It may be explained by higher ability of NDM to block one receptor than two receptors in the same time. Therefore, it seems that wild type carrying both AI sensors is more tolerant to the inhibitory activity of NDM than mutant strains which are lacking one of the sensors. Bioluminescence level among all samples (treated and nontreated with NDM) was similar after additional bacterial incubation (8 h). It may be explained by increasing amount of secreted endogenous AI into the medium due to bacterial population growth resulting in the competition of the AI with the NDM cranberry constituents for receptor binding sites. The fact that addition of exogenous AI reduced the inhibitory effect of NDM is another evidence for the reversible mode of inhibition of cranberry compound. Interestingly, exogenous AI-1 was more pronounced in recovery the bioluminescence inhibition by cranberry constituents than exogenous AI-2, which may be explained by a less affinity of the NDM constituents to the sensor of AI-2, demonstrating a diverse affinity mode of the NDM to the different sensors sites of the quorum sensing cascade.

It was recorded that NDM did not affect quorum sensing of JAF 375 mutant lacking both receptors and producing constitutive bioluminescence (data not shown). These findings revealed that mode of action of NDM is mainly on extracellular organelles as the sensors, and not on the intracellular cascade of the quorum sensing. In addition, we determined that cranberry constituents did not chemically interact with the signal molecules, since AI-1/AI-2 preincubated either with or without NDM during dialysis showed similar activity on V. harveyi.

Summarizing the above observations, we hypothesize that one of the proposed means of action of NDM cranberry constituents is on cell–cell communication. This mechanism indicates another mode of the antibiofilm action of cranberry constituents supporting its known therapeutic benefits in medicine.


This study is part of the Ph.D. thesis of M. Feldman. The authors thank the Israeli cranberry consortium for their valuable advices and support.

Copyright information

© Springer Science+Business Media, LLC 2009