Antibacterial effect of Czech and Mānuka honey on selected mastitis pathogens

The work is focused on the antibacterial effect of four types of blossom honey, one honeydew and two Mānuka honey (MAN100+ and MAN400+) on selected pathogenic microorganisms isolated from cow’s milk (Staphylococcus aureus 51 and S. aureus 428), sheep’s milk (S. aureus 627), and from the Czech Collection of Microorganisms (Streptococcus uberis CCM 4617, Streptococcus agalactiae CCM 6187, Enterococcus faecalis CCM 4224 and Escherichia coli CCM 4787). The concentrations of honey samples were 20% and 30%. The obtained results showed a 100% inhibitory effect of MAN400+ on all tested bacterial strains even at a concentration of 20% and also a comparable inhibitory effect of Mānuka honey with Czech honeydew. The results indicate that honey had an inhibitory effect against the tested bacterial species which may cause mastitis. Blossom honey, honeydew, bacterial infection, mammary gland, inhibitory effect Bacterial infection of the mammary gland (mastitis) is a frequent and costly disease not only in dairy cows, but in small ruminants such as sheep and goats, as well (Menzies and Ramanoon 2001; Leitner et al. 2004; Vyletělová et al. 2011; Manga and Vyletělová 2013; Gelasakis et al. 2015; Kvapilík et al. 2015). In general, mastitis is caused by pathogenic bacteria, especially Staphylococcus aureus, Streptococcus uberis, coagulase-negative staphylococci (CNS), Corynebacterium bovis, Streptococcus agalactiae, Escherichia coli and others (Hanuš et al. 1992; Hanuš et al. 2004; Vyletělová-Klimešová et al. 2014; Bogdanovičová et al. 2016). Due to the increasing number of antibiotic-resistant strains in both human and veterinary medicine, research is focused on the use of alternative antibiotic-free treatments. Among the resistant bacteria causing serious diseases are mainly methicillin-resistant S. aureus, vancomycin-resistant Enterococcus spp., multi-resistant Mycobacterium tuberculosis or carbapenem-resistant bacteria of the genus Enterobacteriaceae (Rattan et al. 1998; Lukášová and Šustáčková 2003; Vanderhaeghen et al. 2010; Gupta et al. 2011). One possibility of preventing mammalian infection is to use the antimicrobial effect of animal products containing plant substances, such as honey. Cooper et al. (1999) tested the sensitivity of 58 S. aureus strains isolated from infected wounds to Mānuka and pasture honey and found no significant differences among the isolates in sensitivity to honey. Al i et al. (2005) also confirmed the inhibitory effect of honey, declared as fennel, on gram-positive and gram-negative mastitis pathogens (S. aureus, Klebsiella pneumoniae, Pseudomonas aeruginosa, S. agalactiae, Proteus spp.). The use of honey in practice is mainly spread in New Zealand, where Mānuka honey is exclusively used for the treatment of mastitis infections (Allen and Molan 1997). However, the question remains whether any local honey could be as effective as Mānuka honey. This work was therefore focused on verifying and comparing the antimicrobial effect of different types of honey originating from northern Moravia and New Zealand on selected bacterial pathogens that can cause mastitis. ACTA VET. BRNO 2018, 87: 387–393; https://doi.org/10.2754/avb201887040387 Address for correspondence: Marcela Klimešová Dairy Research Institute, Ltd. Ke Dvoru 12a, 160 00 Prague, Czech Republic Phone: +420 608 260 395 E-mail: marcela.vyletelova@seznam.cz http://actavet.vfu.cz/ Materials and Methods The origin of bacterial strains and honey samples The tested strains were isolated from cow’s milk suspected of mastitis (S. aureus 51 and S. aureus 428) and from sheep’s milk (S. aureus 627). Other strains originated from the Czech Collection of Microorganisms in Brno (CCM; S. uberis CCM 4617, S. agalactiae CCM 6187, Enterococcus faecalis CCM 4224 and E. coli CCM 4787). Seven kinds of honey (five from small sellers and two commercial) were tested: blossom honey K [the main content (38%) of pollen from the Pyrus/Prunus-T family], BK (with 62% of pollen from the Cruciferae/ Brassicaceae family), S (with 82% of pollen from Cruciferae family) and MM (with 75% content of Cruciferae family pollen), honeydew M (with 74% content of Myosotis sylvatica pollen) and commercial Mānuka honey (MAN100+ and MAN400+, Watson & Son, New Zealand). Chemical compositions and more detailed pollen spectrum of the honey samples as measured by the Intertek Food Services, Bremen, Germany, are given in Tables 1, 2, and 3. The testing procedure for the inhibitory effect of honey Blood agar (without the addition of blood) was prepared to contain honey at the concentrations of 20% and 30%. Honey was homogenized in sterile distilled water and added to the agar base after sterilization and cooling to 45±1 °C. Afterwards, the dilution plate method was performed (ČSN EN ISO 7218, 2008). The tested bacteria were suspended in sterile distilled water and the relevant dilution at a volume of 0.1 ml was inoculated onto the surface of agar with honey and incubated at 37 ± 1 °C for 24–48 h. The total count of colonies was determined in cfu·ml-1 and the results were expressed as a percentage of inhibited bacteria. For this purpose, blood agar without the addition of honey or sugar was used as bacterial growth control. Another control modelling the osmotic pressure of honey was blood agar containing 20% and 30% of honey sugars: fructose (38.2%), sucrose (1.3%), maltose (7.3%) and glucose (31.3%) according to Belitz and Grosch (1992). Statistical analysis Paired t-test was used to compare variables between MAN400+ as a reference honey and others species of honey on the mastitis set of pathogens (MS Excel – Microsoft, Redmond, Washington, USA). P values < 0.05 were considered significant.


The origin of bacterial strains and honey samples
The tested strains were isolated from cow's milk suspected of mastitis (S. aureus 51 and S. aureus 428) and from sheep's milk (S. aureus 627).Other strains originated from the Czech Collection of Microorganisms in Brno (CCM; S. uberis CCM 4617, S. agalactiae CCM 6187, Enterococcus faecalis CCM 4224 and E. coli CCM 4787).
Seven kinds of honey (five from small sellers and two commercial) were tested: blossom honey K [the main content (38%) of pollen from the Pyrus/Prunus-T family], BK (with 62% of pollen from the Cruciferae/ Brassicaceae family), S (with 82% of pollen from Cruciferae family) and MM (with 75% content of Cruciferae family pollen), honeydew M (with 74% content of Myosotis sylvatica pollen) and commercial Mānuka honey (MAN100+ and MAN400+,Watson & Son,New Zealand).Chemical compositions and more detailed pollen spectrum of the honey samples as measured by the Intertek Food Services, Bremen, Germany, are given in Tables 1, 2, and 3.The testing procedure for the inhibitory effect of honey Blood agar (without the addition of blood) was prepared to contain honey at the concentrations of 20% and 30%.Honey was homogenized in sterile distilled water and added to the agar base after sterilization and cooling to 45±1 °C.Afterwards, the dilution plate method was performed (ČSN EN ISO 7218, 2008).The tested bacteria were suspended in sterile distilled water and the relevant dilution at a volume of 0.1 ml was inoculated onto the surface of agar with honey and incubated at 37 ± 1 °C for 24-48 h.The total count of colonies was determined in cfu•ml -1 and the results were expressed as a percentage of inhibited bacteria.For this purpose, blood agar without the addition of honey or sugar was used as bacterial growth control.Another control modelling the osmotic pressure of honey was blood agar containing 20% and 30% of honey sugars: fructose (38.2%), sucrose (1.3%), maltose (7.3%) and glucose (31.3%) according to Belitz and Grosch (1992).

Statistical analysis
Paired t-test was used to compare variables between MAN400+ as a reference honey and others species of honey on the mastitis set of pathogens (MS Excel -Microsoft, Redmond, Washington, USA).P values < 0.05 were considered significant.

Results
The results regarding the inhibitory effect of honey samples are summarized in Table 4.For the 20% concentration of honey, the most significant inhibition (100%) of all the tested strains was found for honey MAN400+.Besides that, honey samples M, MM and MAN100+ significantly (60-100%) inhibited S. aureus and Streptococcus spp.strains while honey K was effective (89-94%) only against S. uberis and S. agalactiae.Among the tested samples, Mānuka honey MAN400+ showed the highest inhibitory effect at both concentrations of 20% and 30%.The effectivity of other samples was in the sequence: honeydew M and honey MAN100+ as comparable with a very good effect, followed by honey MM, then honey BK and K, with honey S being the weakest.
No significant difference was found between MAN400+ and M at both concentrations, between MAN400+ and MM at a 20% concentration, and MAN400+ and BK and MAN100+ at a 30% concentration (Table 5).
As a whole, the microorganis-ms most sensitive to honey were bacteria of the genus Streptococcus at both concentrations of honey.Furthermore, the highest inhibitory effect at the concentration of 20% was found for Staphylococcus strains, followed by E. faecalis and E. coli.At the higher concentration of 30%, the order of sensitivity was as follows: E. coli, E. faecalis, and Staphylococcus spp.

Discussion
In general, both Mānuka honey samples showed the highest inhibitory effect at both concentrations of 20% and 30% which was most likely caused by the presence 389 peroxide, which can be reduced in vivo by the catalase activity in tissues or blood.The effect of Mānuka honey originates partly from hydrogen peroxide and partly from the phytochemical component MGO, so this type of honey may be more effective in vivo.In their later work, Cooper et al. (2002) tested Mānuka and pasture honey against eighteen strains of methicillin-resistant S. aureus, seven strains of vancomycin-sensitive enterococci isolated from infected wounds and twenty strains of vancomycin-resistant enterococci isolated from hospital environmental surfaces using an agar incorporation technique to determine minimum inhibitory concentration (MIC).For all of the strains tested, the MIC values for Mānuka and pasture honey were below 10% (v/v).However, the concentration of artificial honey required to achieve equivalent inhibition in vitro was at least three times higher.Mousa et al. (2012) tested different concentrations of honey (undiluted honey, 10%, 30%, 50% and 70% v/v) against S. aureus and Streptococcus pyogenes by the disc diffusion method in vitro and concluded that these honey samples were comparable with standard antibiotics as ampicillin, penicillin G, amoxicillin, gentamicin, tobramycin, erythromycin and chloramphenicol.
In this study, comparable results with MAN 100+ were achieved by honeydew M. Its high activity can be explained by a higher content of mineral substances, which is typical for honeydews.Likewise, high inhibitory effect was found for MM honey containing the highest content of mineral substances among the tested blossom honey samples.This characteristic is manifested by higher conductivity (0.32 mS/cm).Honey B and BK showed similar effects at a 20% concentration; however, at a 30% concentration, honey BK was more effective.This difference can be explained by the pollen composition and contents of hydroxymethylfurfural and diastase.The weakest inhibitory effect was shown in honey S, although its composition was very similar to honey BK.It differed only in a higher content of pollen grains of the family Crucifereae and a lower content of family Rosaceae.
The antibacterial influence of honey samples was confirmed on both gram-positive and gram-negative bacteria in our study.Wilkinson and Cavanagh (2005) tested the antibacterial effect of thirteen honey samples (including three commercial ones) on gramnegative bacteria E. coli and P. aeruginosa, achieving results comparable with our findings on E. coli.The antibacterial activity of honey was tested using standard diffusion methods with honey samples at four  1.90 ns concentrations (10%, 5%, 2.5% w/v).All of the tested honey types had an inhibitory effect on the growth of E. coli and P. aeruginosa, one of the honey samples still acted against E. coli and three showed anti-P.aeruginosa activity at 2.5%.This study showed that several honey types, besides the commercial antibacterial honey, can inhibit E. coli and P. aeruginosa and may have a therapeutic potential in the case of Gram-negative bacteria.
The use of honey as a wound dressing is well known in both traditional and modern medicine.Many studies have reported the effectivity of honey in the removal of bacterial infections in ulcers and abscesses and indicate that it may be suitable for the intramammary treatment of mastitis.Allen and Molan (1997) tried out the sensitivity of bacteria (Actinomyces pyogenes, K. pneumoniae, Nocardia asteroides, S. aureus, S. agalactiae, Streptococcus dysgalactiae, S. uberis) that usually cause mastitis in dairy cows to antibacterial honey activity.The growth of all seven species tested was completely inhibited by a typical honey (with antibacterial activity attributed to its content of hydrogen peroxide) at a concentration of 10% (v/v) in agar plates.Moreover, two species were inhibited even by 5% honey.
The obtained results indicate that honey could be a suitable raw material for the production of natural sustainable products applicable as alternative therapy in the case of mastitis or during the cow's drying period, as well as for protection against the spreading of pathogenic and resistant microorganisms in dairy production.This study presents preliminary results and a more detailed study should be conducted in the near future.
d = arithmetic mean; SD = standard deviation; t = paired t-test; ns = non-significant; P = probability value of

Table 4 .
The results on the inhibitory effect of honey.

Table 5 .
Comparison of efficacy between honey types and MAN400+.