Efficacy of Honeydew Honey and Blossom Honey on Full-thickness Wound Healing in Mice

Introduction. The wound healing properties of honey, including blossom honey, are well known; however, the effects of honeydew honey during the wound healing process have not yet been investigated and thus remain unclear. Objective. This study compares the effects of honeydew honey with those of blossom honey. Materials and Methods. A total of 140 mice were divided into 2 control groups, which received either a hydrocolloid dressing (HCD; n = 22) or gauze (n = 22), and 4 experimental groups: honeydew honey (n = 23), Acacia honey (n = 23), Manuka honey (n = 22), and Japanese Pharmacopoeia honey (n = 28). Two circular full-thickness wounds were made and measured for 14 days. Each wound in the experimental groups was treated with 0.1 mL of honey and covered with gauze. Dressings in the control and experimental groups were changed daily. Results. The wounds in all of the honey groups and the HCD group were moist by day 14, while those in the gauze group were dry. The ratio of wound area to initial wound area and the number of inflammatory cells decreased during the inflammatory phase in all honey groups. However, the honey groups exhibited reepithelialization rates of < 40%, numerous neutrophils, weak wound contraction, and impaired collagen deposition in wounds after day 11. Conclusions. These results suggest honeydew honey and blossom honey both exert anti-inflammatory effects during the inflammatory phase. However, all of the honeys examined were less effective at promoting full-thickness wound healing than the controls. Further studies are warranted.

Honey is an ancient traditional wound healing agent that contains large amounts of sugar, water, and amino acids, thereby its topical application provides wounds with nutrients and increases the rate of granulation tissue growth.¹ Due to the high glucose concentration of honey, its high osmolarity draws intercellular fluid derived from the circulation, which decreases wound edema. This high osmolarity also is associated with its antibacterial activity that contributes to preventing bacterial growth.² The antibacterial effects of honey are generally attributed to hydrogen peroxide³ and antioxidant activity towards glucose involving the enzyme catalase.⁴ However, the antibacterial activity of Manuka honey is due to another factor.

Manuka honey is the most closely associated with wound healing, and it has been reported^5-7 to exhibit strong antibacterial activity and reduce sloughing/necrosis and wound inflammation in clinical settings. Molan⁸ reported that a component in Manuka honey other than hydrogen peroxide exhibits antibacterial activity, and Mavric et al⁹ subsequently demonstrated that the antibacterial activity of Manuka honey was due to methylglyoxal. Therefore, Manuka honey is currently the most commonly used honey for the treatment of wounds.

Previous studies^1,2,10 reported that honey is an effective promotor of wound healing, particularly chronic wounds such as infected wounds, burns, and ulcers. In experiments involving rats, Tualang honey induced greater reductions in the areas of full-thickness burn wounds than hydrofiber dressings.¹¹ In humans, honey has been found to promote reepithelialization and decrease inflammatory reactions in superficial burns¹² and also prevent the growth of Pseudomonas aeruginosa isolated from infected wounds due to its bactericidal activity.¹³ Furthermore, the topical application of Manuka honey to leg ulcers resulted in reductions in wound size, odor, and pain.⁵ In previous studies^14,15 by the authors herein involving mice, Indonesian honey, Manuka honey, and Japanese honey (Acacia honey, Buckwheat honey, and Chinese milk vetch honey) decreased inflammation and reduced the area of full-thickness wounds during the inflammatory phase. However, these findings on the positive effects of honey on full-thickness wound healing were mainly obtained using blossom honey; limited information is currently available on the effects of honeydew honey on wound healing.

Honeydew honey (100 g), produced from a sap eaten by insects, is composed of 16.3 g water, 80.5 g total sugars, 0.9 g minerals, 0.6 g amino acids, and 1.1 g acids with a pH of 5.2. Similarly, 100 g of blossom honey contains 17.2 g water, 79.7 g total sugars, 0.2 g minerals, 0.3 g amino acids, and 0.5 g acids (on average) with a pH of 3.9.¹⁶ Since the water and total sugar contents of honeydew honey and blossom honey are similar, the investigators hypothesized that honeydew honey exerts similar effects on wound healing as blossom honey. Mayer et al¹⁷ previously reported honeydew honey decreased the wound area and pain in leg ulcers. In addition, it has been found to produce hydrogen peroxide, similar to blossom honey, and the aqueous extract of Fir honeydew honey inhibited the tumor necrosis factor-α-induced production of matrix metalloproteinase-9 in keratinocytes; therefore, honeydew honey is effective in the treatment of infected or chronic wounds.¹⁸ However, the effects of honeydew honey during the wound healing process have not been investigated and, thus, remain unclear. In the present study, the authors focused on the effects of honeydew honey throughout the wound healing process and compared them with those of blossom honey and standard treatments. 

Animals 

One hundred forty 8-week-old male BALB/cCrSlc mice (Sankyo Lab Service Corporation, Inc, Toyama, Japan) with a mean weight of 21.5 g ± 1.3 g were used. They were caged individually in an air-conditioned room at 25.0°C ± 2.0°C with lights on from 08:45 to 20:45. Water and a commercial diet (CRF-1; Charles River Laboratories International, Inc, Kanagawa, Japan) were given freely. Experimental protocol and animal care were conducted in accordance with the Guidelines for the Committee on Animal Experimentation of Kanazawa University, Kanazawa, Japan (AP-153610).

Honey

Three types of blossom honey were used: Acacia honey (Robinia pseudoacacia; Yamada Bee Farm, Okayama, Japan), produced in Romania, which contained 1 × 10 of non-coliform bacteria; Manuka honey (Leptospermum scoparium; Yamada Bee Farm), produced in New Zealand; and Japanese Pharmacopoeia (JP) honey (Yamada Bee Farm). Honeydew honey (Yamada Bee Farm), produced in Bulgaria, also was used. No data were available on the bacterial contents of Manuka, JP, and honeydew honey. 

Injury induction and wound treatment

In accordance with a previous study,¹⁹ mice were anesthetized via inhalation anesthesia (1.75% isoflurane), and their dorsa were then shaved. Two circular (4-mm diameter) full-thickness wounds that included the panniculus muscle were made on both sides of the dorsum with a Kai sterile disposable biopsy punch (Kai Industries, Gifu, Japan). Mice were then divided into 6 groups: hydrocolloid dressing group (HCD; n = 22), gauze group (n = 22), Acacia honey group (n = 23), Manuka honey group (n = 22), JP honey group (n = 28), and honeydew honey group (n = 23).

In the experimental groups, 0.1 mL of honeydew honey, Acacia honey, Manuka honey, or JP honey was used to treat the wound based on the assigned group. All treatments were performed using clean implements disinfected with alcohol. The wounds to which honey was applied were covered with commercial gauze (Libatape, Kumamoto, Japan) in order to keep honey at the wound site and then wrapped twice with tape (Mesh pore tape; Nichiban, Tokyo, Japan) after application; gauze was changed and all wounds were treated with honey daily. 

In the control groups, a HCD (Tegaderm; 3M, Tokyo, Japan) or commercial gauze was used. The gauze was the same as that used in the experimental groups. The gauze treatment in the control groups was performed in order to confirm whether this gauze affects the wound healing process. In the control groups, the relevant dressing was applied and wrapped twice with tape. Dressings were changed daily for all control and experimental groups. 

Macroscopic examinations

The day of wounding was designated as day 0, and the process of wound healing was observed between days 0 and 14. Wounds were examined daily for signs of infection and necrotic tissue. Furthermore, the wound edges were traced on polypropylene sheets and daily photographs were taken. Traces on the sheets were captured with a scanner and transferred to a personal computer using a photo editing program (Adobe Photoshop Elements 7.0; Adobe System Inc, Tokyo, Japan). Wound area was calculated using an image analysis software (Scion Image Beta 4.02; Scion Corporation, Frederick, MD).

Tissue processing

Mice were euthanized with a massive intraperitoneal injection of pentobarbital sodium (0.05 mg/g weight) on day 3, 7, 11, or 14. The wounds and surrounding intact skin were harvested, stapled onto transparent plastic sheets to prevent the overcontraction of specimens, and fixed in 4% paraformaldehyde in 0.1 mol/L phosphate buffer (pH 7.4) for 15 hours. Specimens then were dehydrated in an alcohol series, cleaned in xylene, and embedded in paraffin to prepare 5-µm-thick serial sections. The resultant sections were subjected to hematoxylin and eosin (H&E), Azan, and immunohistological staining with antineutrophil antibody (Abcam Japan, Tokyo, Japan) to detect neutrophils, anti-mouse Mac-3 antibody (BD Pharmingen, Tokyo, Japan) to detect macrophages, and anti-α-smooth muscle actin (α-SMA) antibody prediluted (Abcam KK, Tokyo, Japan) to detect myofibroblasts. The procedure for unmasking the antigens was antigen-dependent, as detailed below. 

Immunohistochemical staining

After the relevant sections had been deparaffinized, antigen unmasking was accomplished by heating slides containing the sections in a water bath, followed by incubating them in sodium citrate buffer (10 mM sodium citrate and 0.05% Tween 20, pH 6.0) at about 100°C for 20 minutes. The slides used for anti-mouse Mac-3 antibody and anti-α-SMA staining were washed with phosphate-buffered saline (PBS), while those used for antineutrophil antibody staining were washed with 0.1% Tween in PBS. Slides were incubated with the antineutrophil antibody or Mac-3 antibody at a concentration of 1:100 in PBS or the anti-α-SMA antibody at 4°C overnight. Slides then were washed again with PBS or 0.1% Tween in PBS. In order to detect the primary antibodies, the slides used for anti-mouse Mac-3 antibody and antineutrophil antibody staining were incubated with polyclonal rabbit anti-rat immunoglobulin/horseradish peroxidase (HRP; Dako North America, Carpinteria, CA) at a concentration of 1:300 in 0.3% mouse serum (normal; Dako North America) in PBS at room temperature for 30 minutes, and the slides used for anti-α-SMA antibody staining were incubated with Dako Envision+system-HRP-labeled polymer anti-rabbit (ready to use; Dako North America) at room temperature for 30 minutes. Slides were washed again with PBS or 0.1% Tween in PBS and then incubated in Dako liquid DAB+ substrate chromogen system (a brown chromogen; Dako North America) for 5 minutes or until staining was detected at room temperature. Light hematoxylin counterstaining was performed for 1 minute to visualize cell nuclei. Slides were rinsed in distilled water, dehydrated, cleared, and mounted for analysis. Negative control slides were obtained by omitting each primary antibody.

Microscopic examinations

The investigators calculated the reepithelialization rate (%) as the length of the new epithelium/the distance between the wound edges and counted the numbers of macrophages, neutrophils, and myofibroblasts through a light microscope at X40 magnification. Then, they calculated the number of each cell type/mm² granulation tissue. The rates of collagen fibers in granulation tissue were calculated as the number of pixels of collagen fibers/number of pixels of granulation tissue using the photo editing program.

Statistical analysis

Data are expressed as the mean ± standard deviation and were analyzed using JMP 12 (SAS Institute, Cary, NC) (analysis of variance, multiple comparison Tukey-Kramer test). Differences were considered to be significant at P < .05. 

Macroscopic examination of wound healing

Wounds in the experimental groups were moist because honey remained on the wound surfaces the next day. The honey-coated gauze covering the wounds was easily removed. Scabs had not formed in any of the honey groups. However, moist necrotic tissue continued to cover the wounds on days 3 to 14 in all honey groups (Figure 1A). It was adhesive to the wound surface and was not removed from the wounds. On the other hand, the HCD absorbed large amounts of the exudate from the wounds in the inflammatory phase. The exudate did not spread from the HCD. Therefore, the wounds in the HCD group were kept moist. Necrotic tissue also appeared in the HCD group on days 5 to 7, and it was easily collected by removing the HCD. In the gauze group, wounds were covered with scabs until day 14 (Figure 1A); thus, they were kept dry. No signs of infection were observed in any of the groups. 

The wounds in the HCD group were completely covered with a new epithelium by day 11, while none of the wounds in any of the honey groups were covered by an epithelium on day 14 (Figure 1A). The ratio of wound area to initial wound area is shown in Figure 1B. No significant differences were observed in the wound area ratios between the honey groups during the study period. On days 2 to 8, the honey groups and gauze group exhibited significantly smaller wound area ratios than the HCD group (P < .0001). On days 9 to 11, no significant differences were noted in the wound area ratio between any of the groups. On days 12 and 13, the wound area ratios of the honey groups were significantly larger than that of the HCD group (P < .05). On day 14, the wound area ratios of the Acacia honey and JP honey groups were significantly larger than that of the HCD group (P = .0001; P = .0063, respectively).

Microscopic examination 

Reepithelialization. On day 3, new epithelium appeared at the wound edges in all groups (Figure 2A). No significant differences in the reepithelialization rates between the honeydew honey and blossom honey groups were noted. On day 3, the reepithelialization rates of the honey groups were significantly lower than those of the HCD and gauze groups (Acacia group vs. HCD group: P = .0002; other honey groups vs. HCD group: P < .0001; all honey groups vs. gauze group: P < .0001). The reepithelialization had been completed by day 11 in the HCD group and by day 14 in the gauze group. However, incomplete reepithelialization was observed in all honey groups on day 14 (Figures 1B, 2). All honey groups exhibited reepithelialization rates of < 40% on day 14 (honeydew honey: 33.7%; Acacia honey: 19.0%; Manuka honey: 38.2%; JP honey: 15.1%), which were significantly lower than those in the HCD and gauze groups (all honey groups vs. HCD group or gauze group: P < .001). 

Macrophages. All honey groups and the gauze group displayed significantly smaller numbers of macrophages than the HCD group on days 3 and 7 (P < .001 in both cases; Figure 3). In all groups, except the Manuka honey group, the number of macrophages decreased from days 3 to 7. There were no significant differences in the number of macrophages between the honeydew honey and blossom honey groups. 

Neutrophils. The number of neutrophils in the honeydew honey, Manuka honey, and JP honey groups was significantly smaller than that in the HCD group on day 3 (P = .0065; P = .0070; P = .0042, respectively). On day 11, numerous neutrophils were observed in all honey groups, and the number of neutrophils in the Acacia honey, Manuka honey, and JP honey groups was significantly larger than those in the HCD and gauze groups (P = .0379; P = .0007; P = .0228, respectively). On day 14, the number of neutrophils had slightly decreased in all honey groups; however, the number of neutrophils in the honeydew honey, Acacia honey, and Manuka honey groups was significantly higher than that in the HCD (P = .0104; P = .0006; P = .0462, respectively; Figure 4). There were no significant differences in the number of neutrophils between the honeydew honey and blossom honey groups. 

Myofibroblasts. On day 3, similar numbers of myofibroblasts started to appear in all groups (Figure 5A). There were no significant differences in the number of myofibroblasts between the honeydew honey and blossom honey groups during the study period. On day 7, the number of myofibroblasts had markedly increased in the HCD and gauze groups. The number of myofibroblasts in the honey groups was significantly smaller than those observed in the HCD and gauze groups on days 7 and 11 (P < .01; Figure 5). On day 14, the number of myofibroblasts in the JP honey and Manuka honey groups was significantly smaller than that in the HCD group (P = .0207; P = .0237, respectively). 

Collagen fibers. The rate of collagen fibers gradually increased from days 7 to 14 in all groups (Figure 6A). There were no significant differences between the honeydew honey and blossom honey groups. On day 11, the rate of collagen fibers was significantly lower in the Acacia honey, Manuka honey, and JP honey groups than in the HCD group (P = .0277; P = .0285; P = .0274, respectively). On day 14, the rate of collagen fibers was significantly lower in the Manuka honey and JP honey groups than in the HCD group (P = .0117; P = .0085, respectively; Figure 6). 

Dating back as far as ancient history, honey has been reported to exert positive effects on wound healing. Blossom honey has been used to treat chronic wounds (which are difficult to heal in clinical settings), resulting in wound healing or a reduction in pain.^8,20,21 Few studies have examined the effects of honeydew honey on wound healing and reported that it exerted anti-inflammatory effects.^17,18 In the present study, the authors investigated the effects of honeydew honey on full-thickness wounds throughout the wound healing process in mice. The honeydew honey and blossom honey groups showed decreased wound areas and neutrophil and macrophage numbers during the inflammatory phase. These results were consistent with the authors’ previous findings^14,15,22 and also support the notion that honeydew honey has anti-inflammatory effects on wound healing. However, contrary to the investigators’ expectations, the daily application of honeydew honey and blossom honey to full-thickness wounds appeared to be less effective for prolonged inflammation, reepithelialization, wound contraction, and collagen deposition. The reasons for impaired wound healing induced by the application of honey currently remain unclear; however, they propose the following suggestions. 

Diluted honey may be more effective than undiluted honey at promoting wound healing. Honey generates high amounts of hydrogen peroxide, which is one of its major antibacterial components.³ Hydrogen peroxide is produced via the glucose oxidase-mediated conversion of glucose in diluted honey.²¹ A previous study²³ used diluted honey to promote acute wound healing and found that it had positive effects. At a concentration of 10%, Acacia honey promoted epithelialization and wound contraction in rats with incision and burn wounds.²³ In addition, honey at a concentration of 1% stimulated monocytes to release inflammatory cytokines, including tumor necrosis factor-α, interleukin (IL)-1, and IL-6.^24,25 However, undiluted honey is often used for chronic wounds. Chronic wounds have large amounts of exudate and any applied honey is diluted by these exudates. As a result, hydrogen peroxide is produced in the wounds. In the mice in the present study, the amount of wound exudate was small; therefore, hydrogen peroxide levels were not considered to be high in the wounds. All honey groups exhibited delayed wound healing. 

Furthermore, the pH of the wound surface after the application of honey was low. Honey has a pH ranging from 3.5 to 4.5. The acidic pH of honey reduces the alkalinity of the wound surface pH, which ranges between 7.3 and 8.9 in nonhealing chronic wounds.²⁶ Manuka honey was previously shown²⁷ to be associated with significant reductions in wound pH and size after a 2-week treatment. Further, the wound surface pH of acute wounds, such as full-thickness wounds, ranges between 4.4 and 5.3 (unpublished data, October 2016). Lönnqvist et al²⁸ reported that epithelialization did not occur in environments with pH of < 5 in vivo. Honeydew honey has a pH of 5.2, while blossom honey has a pH of 3.9.¹¹ Therefore, the daily topical application of honey may reduce the surface pH of full-thickness wounds to pH < 5. In future experiments, the authors intend to investigate the relationship between wound surface pH and wound healing after honey treatments. 

The present results will contribute to further investigations on the effects of honey on wound healing. The mechanisms that promote wound healing by honey have not yet been elucidated. The active components of honey and their contribution to the wound healing process warrant further study.

In this study, the investigators have no information concerning active components or detailed ingredients of the honeys examined. This information is important to consider for the obtained results in the present and future studies. The investigators were not able to clarify the reason as to why the daily application of honey did not promote wound healing. Further studies with supporting evidence are needed in order to fully elucidate the effects of the topical application of honeydew honey and blossom honey on full-thickness wound healing.

Honeydew honey and blossom honey exerted similar effects on full-thickness wound healing in mice; they also decreased wound areas and the numbers of inflammatory cells during the inflammatory phase. However, the daily application of honey did not promote the wound healing process more than that of the control groups. 

* Takuto Sawazaki and Yukari Nakajima contributed equally to this research.

Affiliations: Department of Clinical Nursing, Graduate Course of Nursing Science, Division of Health Sciences, Graduate School of Medical Sciences, Kanazawa University, Kanazawa, Japan; Institute of Medical, Pharmaceutical and Health Sciences, Kanazawa University; and Division of Nursing, Faculty of Health Sciences, Kanazawa University

Correspondence: Toshio Nakatani, PhD, MD, Institute of Medical, Pharmaceutical and Health Sciences; Kanazawa University, 5-11-80 Kodatsuno, Kanazawa 920-0942, Japan;
nakatosi@staff.kanazawa-u.ac.jp

Disclosure: This study was supported by JSPS KAKENHI Grant Number JP25293430 and a grant from Yamada Bee Farm, Okayama, Japan. The authors disclose no financial or other conflicts of interest.