Honey Potential as Antibiofilm, Antiquorum Sensing and Dispersal Agent against Multispecies Bacterial Biofilm

bacterial bioﬁlm is quite complicated because of the mech-anisms underlying bacterial growth as well as resistance. Multispecies bioﬁlms have complementary metabolic strategies for obtaining nutrients and degrading host immune molecules 3 ） . Various strategies have been developed and implicated for the bioﬁlm control, however, natural and ef-Abstract: This study is first to test Pakistani honey bees, Apis dorsata and A. cerana honey samples as anti biofilm, anti quorum sensing (QS) and biofilm dispersal agents honey against multispecies biofilm of bacteria (obtained from obese patients). Briefly, five previously identified isolates Pseudomonas aeruginosa , Escherichia coli , Staphylococcus aureus , Morganella morganii and Klebsiella pneumoniae (MT448672-MT448676) were selected. Antibiogram study of all five isolates was tested against three antibiotics viz ., erythromycin (20 µg/mL), lincomycin (100 µg/mL) and rifampicin (100 µg/mL). In order to form multispecies biofilm, identified bacteria were grown in batch culture by mixing equal volumes (OD 590nm = 0.1) of 2, 3 and 5 bacterial isolates. In total 11 groups (g1-g11) were made. Crystal violet (CV) staining method was used to evaluate the antibiofilm potential and biofilm dispersal potential of both honey samples. QS inhibition in P. aeruginosa was measured following culture supernatant method. Antibiogram study showed significant ( p < 0.05) resistance by P. aeruginosa against tested antibiotics. E. coli , M. morganii and K. pneumoniae were significantly susceptible to erythromycin and S. aureus to lincomycin. Both honey samples at 2% and 5% concentrations showed significant ( p < 0.05) inhibition potential of multispecies biofilm by all test groups (g1-g11). Though A. dorsata honey significantly inhibited biofilm formation at 2 and 5% against all groups but 2% concentration was highly significant against g2-g4 groups. Regarding A. cerana honey, 2% concentration was significantly effective against g1, g4-g7 and g9-g11 groups. Both honey samples significantly inhibited QS at 2 and 5%. The 5% concentration of A. dorsata honey significantly dispersed biofilm by all groups compared to 2% which showed dispersal potential only by g2 and g3 groups. Accordingly, honey samples showed significant antibiofilm, anti-QS and biofilm dispersal potentials thus can be considered as good alternative to antibiotics.


Introduction
Bacterial biofilms are a major worldwide healthcare problem urinary tract infections and are associated with decreasing quality of life and significant patient morbidity 1 . Bacteria in biofilm mode protect themselves from antibiotics and the patient s immune system 2 . Treatment of ficient anti-biofilm forming agents are still being used widely. Bacterial cells in a biofilms can be up to a thousand times more resistant against antibiotics when compared to free living planktonic cells 4 . Pseudomonas aeruginosa is well known for its quorum sensing QS mechanism in developing biofilm and it regulates QS by secreting a light green pigment known as pyocyanin 5 . When the biofilm becomes mature, biofilm dispersal starts either actively by external harsh conditions due to antibacterial agents or passively by removing complete biofilm with sharp equipment 6 . Due to an increase resistance towards antibiotics, the need for natural antibiofilm agent like honey has generated renewed interest 7 . No studies have identified the exact antimicrobial mechanism of honey until now, although few have reported contributing factors such as high sugar contents 80 , minerals and vitamins 8 , its acidity pH 3.2-6.0 , production of hydrogen peroxide due to glucose oxidase; osmotic effects, antioxidant contents and stimulation of immunity neutrophils, that contribute towards antimicrobial properties of honey include 9 . Previously, Lu et al. 10 and Kim and Kang 11 , also reported antibiofilm activities of honey against P. aeruginosa and E. coli. Studies revealed that honey possesses strong anti-QS potentials at the concentration below its MIC values 12 14 . Moreover, a study by Park et al. 15 , revealed that honey is also effective in dispersing mature biofilm. With all this background, the current study has been designed to investigate the antibiofilm, anti QS and biofilm dispersal potential of two honey samples obtained from honey bees, A. dorsata and A. cerana against bacterial isolates obtained from obese patients.

Sample collection
A. dorsata and A. cerana honey samples were collected from the farmers of Sadiq Abad, Pakistan. The beekeepers raised the bees in hives under shelter close to fresh water source but away from noise. Initially artificial nectar was provided by dissolving equal parts granulated sugar to fill quart jars. Jars with top feeder lids are invert into the holes to nurture young colonies. The lids should be barely moist. The bees will drink what they need from the lids. After three weeks, bees are able to get nectar from agricultural crops, different floral sources and trees such as sider and bari, present in surrounding area. Both bees are indigenous so the honey is collected around the season. Honey samples were sterilized by spreading on agar plates and incubated for 24 hours at 37 . No bacterial growth proved the sterilization of honey samples. The test strains were obtained from Microbiology Lab, Department of Zoology, GC University, Lahore. Five test strains viz., Pseudomonas aeruginosa, Escherichia coli, Staphylococcus aureus, Morganella morganii and Klebsiella pneumoniae MT448672-MT448676 were used in current study.

Antibiogram study
Antibiotic susceptibility of the selected isolates was checked following disc diffusion assay 16 . Three antibiotics viz., erythromycin 20 µg/mL , lincomycin 100 µg/mL and rifampicin 100 µg/mL were used and dilutions were made 17 . Test strains were spread on agar plates and discs of 6 mm containing 20 µL volume of above mentioned antibiotics were placed. The plates were incubated at 37 for 24 hours. The zone of inhibition around the discs were measured in millimeter mm and expressed as mean SE. Criteria for resistance, susceptibility and intermediate resistance or susceptible were followed according to the Traub et al. 18 . Diameter of inhibition zones less than or equal to 13 mm resistant; diameter 14-21 mm intermediate susceptible; diameter greater than or equal to 22 mm susceptible were considered.

Formation of multispecies bio lm groups
Multispecies biofilms of the test strains were formed by growing isolates in batch cultures by mixing equal volumes OD 590nm 0.1 of 2, 3 and 5 bacterial isolates. Thus, total 2.4 Congo red agar assay CRA for bio lm detection Congo red assay medium was prepared following slight modifications of method by Liaqat et al. 19 . Congo red plates were streaked with test strains groups and incubated at 37 for 24 hours. Following 24 hours incubation, black colonies on plates indicated biofilm production.

Determination of inhibitory effects of A. dorsata and
A. cerana honey on bio lm formation Various six concentrations 2, 5, 8, 11, 14 and 17 v/v of honey samples were prepared and tested against all bacterial groups g1-g11 using slight modifications of CV staining method by Liaqat et al. 20,21 and the rest of the biofilm quantification was done following CV staining method. Bacterial inoculum of all test groups g1-g11 were prepared in nutrient broth and optical density OD 590 was adjusted to 0.1 approximately 1 10 8 CFU mL 1 . Then 50 µL of diluted culture was transferred to the test tubes containing 3 mL fresh LB-media along with 150 µL of diluted honey samples and incubated at 37 . After 72 hours media was discarded and test tubes were washed thrice with autoclaved distilled water, air dried and stained with 1 CV for 15-20 minutes. The stain was discarded again and test tubes were washed twice to remove the excess strain. The attached CV stain was dissolved in 33 glacial acetic acid and OD 590 was measured using UV-visible spectrophotometer. Experiment was run in triplicates. Erythromycin 20 µg/mL was used as positive control and experiment was run in triplicate.

Bio lm formation time kinetics of test strains
Following method by Liaqat et al. 22 , biofilm time kinetics study of test groups was performed at different time intervals of 24, 48, 72 and 96 hours at 37 . The CV staining method was used to quantify the adherent cells. Briefly, the overnight cultures of eleven test groups g1-g11 were prepared and the OD 590 was adjusted to 0.1 approximately 1 10 8 CFU mL 1 . The 50 µL of diluted culture was transferred to the test tubes containing 3 mL fresh LB-media and incubated for 24, 48, 72 and 96 hours at 37 . The experiment was run in triplicate.

Pyocyanin production time kinetics of P. aeruginosa
Culture supernatant method by Vinckx et al. 23 with slight modifications was followed for pyocyanin measurement after various intervals of time i.e., 24, 48, 72 and 96 hours. Briefly, the overnight cultures of P. aeruginosa were prepared and OD 590 was adjusted to 0.1 approximately 1 10 8 CFU mL 1 . The 50 µL of diluted culture was transferred to the test tubes containing 3 mL fresh LB-media and incubated for 24, 48, 72 and 96 hours at 37 . Bacterial cells were removed and pyocyanin was extracted and mixed with chloroform. For re-extraction of pyocyanin, 1 mL of acidified water 0.2 mol / HCl was used. OD 520 was measured. In order to obtain the concentration of pyocyanin as µg/mL, optical densities were multiplied by a factor 17.072 23 . The experiment was run in triplicate.

Antibio lm quanti cation of A. dorsata and A. cerana
honey samples Antibiofilm effects of both honey samples were measured following CV staining method 20,21 . The overnight bacterial cultures of all test groups g1-g11 were added into test tubes containing 3 mL fresh broth and 150 µL of honey sample 2 and 5 was used as test compound. Erythromycin 20 µg/mL , 1.4 and 3.5 fructose + sucrose solution equivalent to 70 of 2 and 5 honey were used as controls. Test tubes were incubated for 24 hours at 37 . After 24 hours, antibiofilm potential of honey was tested using CV staining method as described above determination of MIC . The experiment was run in triplicate.

Anti-QS activity of A. cerana and A. dorsata honey
samples Pyocyanin is blue color secondary metabolite and is produced by P. aeruginosa. Culture supernatant method was used for the pyocyanin measurement with slight modifications 23 . Briefly, the overnight cultures of P. aeruginosa was prepared and the OD 590 was adjusted to 0.1 approximately 1 10 8 CFU mL 1 . The 50 µL of diluted culture was transferred to the test tubes containing 3 mL fresh LBmedia along with 2 and 5 honey concentrations and erythromycin as positive control. Bacterial cells were removed, pyocyanin was extracted after centrifugation at 5,000 rpm and rest of the procedure was followed according to the method mentioned in section pyocyanin production time kinetics of P. aeruginosa .

Dispersal activity of A. cerana and A. dorsata honey samples
The method by Kavita et al. 24 with slight modifications was used to determine the biofilm dispersal potential of both honey samples against 11 multispecies biofilm groups g1-g11 . The multispecies bacterial biofilm was established up to 72 hours as mentioned in section biofilm formation time kinetics . 2 and 5 honey concentrations were added to treat biofilms for the next 24 hours. Test tubes were washed and 1 w/v crystal violet was used to stain the formed biofilm. Afterwards, 1 mL of 70 ethanol was added and using spectrophotometer, absorbance was measured at 590 nm. By using following formula, biofilm dispersal activity was measured:

Percent of biofilm dispersal
Absorbance of control absorbance of treated biofilm Absorbance of control 100

Statistical analysis
The data was expressed as means SE and were analyzed statistically using one-way ANOVA followed by Post Hoc Tukey test. Data with significant value *p 0.05, **p 0.01 and ***p 0.001 were considered statistically significant. All statistical analyses were performed using SPSS software and graphs were constructed using Microsoft Excel 2010 .

Antibiogram study
It was observed that P. aeruginosa was highly significantly p 0.01 resistant against all tested antibiotics lincomycin, erythromycin and rifampicin as no clear zones were observed. While other isolates, M. morganii, E. coli and K. pneumoniae were significantly p 0.05 more susceptible to erythromycin compared to other two antibiotics inhibition zones; ZI 30 mm, 23.6 mm and 25 mm, respectively . Among all isolates, S. aureus showed significantly highest susceptibility to lincomycin ZI 17 mm . Among three antibiotics, erythromycin was found to be highly effective against most of test isolates hence used as positive control in subsequent study Fig. 1S .

Congo red agar method CRA
Congo red agar assay results indicated strong biofilm ca- Fig. 1 Inhibitory effects of A. dorsata and A. cerana honey on biofilm formation of eleven multispecies biofilm test groups g1-g11 . a A. dorsata honey significantly p 0.05 inhibited the biofilm formation in all groups g1-g11 at 2 and 5 concentrations. Highly significant p 0.01 biofilm inhibition was seen at 2 and 5 concentrations against g2-g4 and g8-g11. b A. cerana significantly p 0.05 inhibited the biofilm formation of in all multispecies biofilm groups g1-g11 at 2 and 5 concentrations. The data was analyzed by one way ANOVA followed by Post Hoc Tukey test using SPSS software and graphs were constructed using excel 2010 . pability by all test groups g1-g11 as appearance of intense black colonies on all plates Fig. 2S .

Inhibitory effects of A. dorsata and A. cerana honey
on bio lm formation Among six tested concentrations 2, 5, 8, 11, 14 and 17 of both honey samples obtained from honey bees, A. dorsata and A. cerana, it was observed that A. dorsata honey significantly p 0.05 inhibited the biofilm formation in all groups g1-g11 at 2 and 5 concentrations. Highly significant p 0.01 biofilm inhibition was seen at 2 and 5 concentrations against g2-g4 and g8-g11. While, A. cerana significantly p 0.05 inhibited the biofilm formation of in all multispecies biofilm groups g1-g11 at 2 and 5 concentrations Fig. 1 .

Bio lm formation time kinetics
The multispecies biofilm formation time kinetics of eleven test groups g1-g11 was monitored using CV staining method for 24, 48, 72, 96 hours of incubation periods at 37 . All 11 multispecies biofilm test groups showed significantly p 0.05 high biofilm formation at 72 hours of incubation, after that there was a decline in the biofilm texture Fig. 3S .

Pyocyanin production time kinetics of P. aeruginosa
Time kinetics for pyocyanin production of P. aeruginosa was checked by incubating for 24, 48, 72, 96 hours at 37 . Significant p 0.05 production of pyocyanin by P. aeruginosa was observed after 96 hours incubation period Fig.  4S .

Antibio lm effect of A. dorsata and A. cerana honey
samples Antibiofilm study revealed that both honey samples possesses strong multispecies biofilm inhibition potentials compared to positive control erythromycin . A. dorsata honey showed significant p 0.05 biofilm inhibition at 2 against all tested groups of isolates. 2 and 5 concentrations showed almost same significant p 0.05 biofilm inhibition against g1 and g5-g11. A. cerana honey significantly p 0.05 inhibited the biofilm formation at 2 against all the groups g1-g11 compared to erythromycin positive control . Overall A. cerana honey was significantly p 0.05 effective as antibiofilm agent Fig. 2 .

Anti-QS activity of A. dorsata and A. cerana honey samples
Both honey samples showed strong anti-QS potentials against P. aeruginosa by inhibiting the production of pyocyanin. Both concentrations 2 and 5 of A. dorsata honey significantly p 0.05 inhibited the production of pyocyanin as compared to positive control erythromycin . A. cerana honey was significantly p 0.05 effective at 5 concentration as compared to 2 concentration and positive control erythromycin . Comparison of both honey samples showed that A. cerana honey was more significantly p 0.01 effective at high concentration 5 while A. dorsata honey showed significant p 0.05 results at both concentration 2 and 5 Fig. 3 .

Bio lm dispersal activity of A. dorsata and A. cerana
honey samples A. dorsata honey strongly dispersed the multispecies biofilm. A. dorsata honey significantly p 0.05 dispersed the multispecies biofilm of g2-g4 and g7-g11 at 5 concentration while there was no significant difference between 2 and 5 concentrations against g1, g5 and g6. A. cerana honey showed significant p 0.05 biofilm dispersal at 5 concentrations against g2-g5 and g7-g11. In g6, both honey concentrations 2 and 5 were significantly p 0.05 effective compared to positive control. Best biofilm dispersal was exhibited at 5 concentration in most multispecies biofilm groups either by A. dorsata or A. cerana honey samples, however 2 concentration was also effective compared to the erythromycin Fig. 4 .

Discussion
Nearly, 99 bacteria on the earth live in communities known as biofilms 20 22, 25 . Bacteria attach themselves to a surface through an extracellular polymeric substance EPS to form biofilm 26 . Bacteria have a well-developed mechanism of communication known as quorum sensing QS which helps in biofilm formation 22 e.g. pyocyanin production in P. aeruginosa 5 . Once a biofilm becomes mature, bacteria start to disperse from the biofilm 6 . Looking for a novel method to eradicate or inhibit biofilm growth has been an active area of research. This study used in vitro approach to assess antibiofilm, anti-QS and dispersal potential of two honey samples obtained bees i.e., A. cerana and A. dorsata against multispecies biofilm formed by five obese patient s isolates.
Five bacterial isolates i.e., P. aeruginosa, E. coli, S. aureus, M. morganii, and K. pneumoniae were used to check for antibiogram study. All isolates were obtained from fecal samples of obese patients and are in agreement with data obtained by authors, who reported the presence of S. aureus, P. aeruginosa and M. morganii in the fecal samples of the persons suffering from obesity related conditions 27,28 . Antibiotic susceptibility test was performed to check the antibiotic resistance pattern of identified isolates. Highly significant p 0.01 resistance to antibiotics erythromycin, lincomycin and rifampicin was noticed in case of P. aeruginosa. Corroborate with findings by Murray et al. 29 , who reported that P. aeruginosa was resistant towards most of the antibiotics due to its high pro-duction of EPS and strong biofilm forming ability. Likewise, E. coli, M. morganii and K. pneumoniae showed significant p 0.05 susceptibility towards erythromycin while S. aureus was significantly susceptible to lincomycin. The in-creased bacterial resistance observed in this study indicated over and miss-use of the antibiotics among visiting obese patients 30 .
Using congo red agar method, appearance of intense dorsata honey showed significant p 0.05 biofilm inhibition at 2 against all tested groups of isolates. 2 and 5 concentrations showed almost same significant p 0.05 biofilm inhibition against g1, g5-g11. b A. cerana honey significantly p 0.05 inhibited the biofilm formation at 2 against all the groups g1-g11 compared to erythromycin positive control . The data was analyzed by one way ANOVA followed by Post Hoc Tukey test using SPSS software and graph was constructed using excel 2010 .
black colonies by all eleven multispecies test groups on media plates indicated their strong biofilm forming potential. These findings are consistent with data obtained by Liaqat et al. 19 . Biofilm time kinetics study of all multispecies test groups was performed over a time span of 96 hours. Significant p 0.05 biofilm formation was seen after 72 hours of incubation. Decline in biofilm formation after 72 hours might be due to their growth in batch cultures where nutrient deprivation and waste accumulation lead to decreased biofilm growth and its dispersal 31 .
We next determined inhibitory effects of A. dorsata and A. cerana honey on biofilm formation of all multispecies  honey significantly p 0.05 dispersed the multispecies biofilm of g2-g4 and g7-g11 at 5 concentration while there was no significant difference between 2 and 5 concentrations against g1, g5 and g6. b A. cerana honey showed significant p 0.05 biofilm dispersal at 5 concentrations against g2-g5 and g7-g11. In g6, both honey concentrations 2 and 5 were significantly p 0.05 effective compared to positive control. The data was analyzed by one way ANOVA using SPSS software and graph was constructed using excel 2010 . biofilm groups g1-g11 . It was observed that both A. dorsata and A. cerana honey at 2 and 5 concentrations significantly p 0.05 inhibited the multispecies biofilm formation. Previously, McLoone et al. 32 determined the minimum inhibitory values of honey against S. aureus associated with skin wound related infection and found 2.5 as effective MIC. Similarly a study by Carnwath et al. 33 checked antimicrobial activities of 10 honey samples against S. aureus, E. coli and P. aeruginosa using serial dilution method to produce final honey concentrations MIC ranging from 2 to 16 . Results demonstrated that all honey samples inhibited the growth of all the tested microorganisms, with inhibitory concentrations ranging from 2 to 6 . Our next aim was to determine the antibiofilm effect of honey samples sources: A. dorsata and A. cerana . It was observed that both honey samples had significant p 0.05 biofilm inhibitory effects against multispecies biofilm at both concentrations 2 and 5 . Though, highly significant p 0.01 biofilm inhibition was observed at 2 concentration in both honey samples obtained from A. dorsata and A. cerana. This might be due to the presence of high amount of hydrogen peroxide whose production is associated with honey dilution. The greater the honey is diluted, greater would its production of hydrogen peroxide 34 and strongly it will reduce biofilm formation. However, our results are contradictory to findings by Sharahi et al. 35 , who reported that honey possessed more antibiofilm potentials at higher concentrations 25 .
Quorum sensing QS is important for biofilm maturation and stabilization in various pathogenic bacteria 20, 21 P. aeruginosa and S. aureus . Pseudomonas aeruginosa produces pyocyanin a signaling molecule to communicate and to initiate biofilm formation 36 . The maximum time required for pyocyanin production was between 24-48 hours incubation 5 . However, current study showed that P. aeruginosa significantly p 0.05 produced pyocyanin after 96 hours incubation. It was observed that 2 and 5 concentrations of A. dorsata and A. cerana honey possess significant p 0.05 anti-QS activities against P. aeruginosa, as noted by inhibition of pyocyanin production in this isolate. This could be explained by the fact that honey inhibits the genes las and rhl expression responsible for pyocyanin production in P. aeruginosa 37 . Likewise, pyocyanin inhibition might be due to inhibition of siderophore genes expression. Previously, Kronda et al. 37 also reported that manuka honey inhibits the siderophore genes expression in P. aeruginosa. Furthermore, high sugar content of honey also plays a major role in QS inhibition by making resistance in binding pyocyanin to its cytoplasmic receptors 38 .
Honey is also effective to disperse mature biofilm and led to release of planktonic cells 39 . Hence, our last aim was to determine the biofilm dispersal potential of two honey samples obtained from A. dorsata and A. cerana . It was observed that both A. dorsata and A. cerana honey samples were effective in dispersing multispecies biofilm. Both honey samples exhibited significant p 0.05 biofilm dispersal at 2 and 5 concentrations but 5 concentration was most effective as it caused highly significant p 0.01 multispecies biofilm dispersal of all test isolates groups. The biofilm dispersal effect of honey samples might be due to the presence of high osmotic pressure which induces the bacteria to degrade their EPS matrix, disperse and release planktonic cells. Similar results were reported by various other authors in past who described that higher concentration of honey are effective to disperse mature bacterial biofilms 13,19,40 .
Quality of honey is determined by its physiochemical parameters like color, pH, hydroxymethylfurfural HMF , electrical conductivity EC , moisture content, diastase index, Total antioxidant capacity and bioactive compounds determine the antioxidant and bioactive potential of the honey samples. Though, in Pakistan honey is consumed in many foods and herbal treatment but little information exists on physicochemical and antioxidant properties of honey from Asian bees, A. cerana and A. dorsata. For example, Kousar 42 reported 0.96 mS/cm EC values for A. dorsata honey. In ongoing study in our Lab, physicochemical and antioxidant potential of both honey samples have been studied proving A. cerana honey having higher antioxidant potential with higher phenolic, flavonoid content, ferric reducing power, and color intensity compared to A. dorsata honey.

Conclusion
In conclusion, the results of the current study showed that despites of the lot of number of antibiotics, natural and primitive compounds, honey possess great medical importance to treat various biofilm related infections. This study is first to test Pakistani honey bees, A. cerana and A. dorsata honey samples as anti biofilm, anti QS and biofilm dispersal agents against all multispecies antibiotic resistant bacterial groups g1-g11 . Taken together, the study findings support use of honey as effective remedy against biofilm related infections such as wound healing, stomach ulcer and biofilm related problems in medical instruments e.g. to eradicate biofilm from urinary tract and respiratory tract catheters. Ongoing study in lab will provide effective honey potential as adjuvants to antibiotics against multi drug resistant isolates.

Conflict of Interests
On behalf of all authors, the corresponding author states that there is no conflict of interest.

Supporting Information
This material is available free of charge via the Internet at doi: 10.5650/jos.ess21199