Published online before print March 27, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Department of Internal Medicine, The University of Texas Medical Branch, Galveston, Texas, USA; and
Shriners Hospitals for Children, Galveston, Texas, USA
1Correspondence: The University of Texas Medical Branch, Department of Internal Medicine, 301 University Boulevard, Galveston, TX 77555-0435, USA. E-mail: fsuzuki{at}utmb.edu
ABSTRACT
Using a mouse model of thermal injury, we studied why antimicrobial peptides are not produced at the burn-site tissues and how this defect contributes to the increased susceptibility to Pseudomonas aeruginosa burn-wound infection. Logarithmic growth of P. aeruginosa was demonstrated locally (at the burn site) and systemically (in circulation) in thermally injured mice exposed to 102 CFU/mouse of the pathogen beneath the burn wound. However, neither systemic nor local growth of the pathogen was observed in sham burn mice when they were infected intradermally with 106 CFU/mouse P. aeruginosa. Murine β-defensins (MBDs) were detected in the skin homogenates of sham burn mice. However, the amounts of MBDs were reduced greatly in the same tissue homogenates from thermally injured mice. Gr-1+CD11b+ cells, with an ability to suppress antimicrobial peptide production by skin keratinocytes, were isolated from tissues surrounding the burn areas, and these cells were not obtained from skin tissues of sham burn mice. After intradermal inoculation of Gr-1+CD11b+ cells, which were isolated from burn-site tissues, the production of antimicrobial peptides around the cell-inoculation site of sham burn mice decreased. Also, like thermally injured mice, these mice were shown to be susceptible to P. aeruginosa intradermal infection. These results indicate that sepsis stemming from P. aeruginosa burn-wound infection is accelerated by burn-induced Gr-1+CD11b+ cells with abilities to suppress antimicrobial peptide production by epidermal keratinocytes.
Key Words: thermal injury • β-defensin
INTRODUCTION
Patients sustaining major thermal injuries are especially prone to infection in their burn-wound area [1 ]. Pseudomonas aeruginosa is a common pathogen that causes burn wound infection. Without appropriate preventive treatment, the greater parts of burn-wound infection spread systemically and develop into sepsis [2 , 3 ]. Topical antibacterials, such as silver sulfadiazine, silver nitrate, and mafenide acetate, are useful for controlling the colonization and multiplication of microorganisms on the surface of burn wounds [4 ]. However, as a result of the burn-induced defects of host antibacterial defenses, the small amounts of bacteria, which escape from the above treatment, are enough to spread systemically [5 ].
The innate immune system is the first line of host defense against invaded bacteria [6 , 7 ]. The important roles of neutrophils and macrophages in antibacterial innate immunity have been described in many papers [8 9 10 11 12 ]. An additional effector mechanism in the innate immunity is the production of antimicrobial peptides, which have broad-range antimicrobial activities against Gram-positive and Gram-negative bacteria, fungi, and viruses [13 , 14 ]. Defensins and cathelicidins are two major groups of antimicrobial peptides. Most of antimicrobial peptides kill bacteria by adhering to and punching a hole in the fatty cell membranes of bacteria. Some of these peptides are synthesized naturally, and others are induced after exposure to microbes [13 , 14 ].
The importance of β-defensins and cathelicidins, which are distributed throughout the skin tissues on host resistance against various skin infections, has been reported [15 16 17 18 ]. Thus, cathelicidin-knockout mice were susceptible to group A streptococcus skin infection [15 ]. Human β-defensin-3 gene-transfected epidermis showed a highly protective shield against invasion of Staphylococcus aureus [19 ]. Also, synthesized β-defensin was shown to be protective of burn mice exposed to P. aeruginosa wound infection [20 , 21 ]. These data suggest that bacterial growth at the burn site and sepsis stemming from such local infection may be suppressed if sufficient amounts of antimicrobial peptides are produced at the local site of P. aeruginosa infection. However, the impaired production of antimicrobial peptides in tissues surrounding the burn area has been demonstrated [22 ]. In those studies, the mRNA expression of β-defensins was greatly impaired in full-thickness burn wounds of patients [22 ].
In this study using a mouse model of thermal injury, we examined why antimicrobial peptides are not produced at the burn-site tissues. In the results, Gr-1+CD11b+ cells, isolated from tissues surrounding the burn areas, were shown to be inhibitory on the production of antimicrobial peptides by epidermal keratinocytes. Gr-1+CD11b+ cells were not obtained from sham burn mice, which were resistant to locally infected P. aeruginosa. Intradermal inoculation of burn mouse-derived Gr-1+CD11b+ cells to sham burn mice resulted in the increased susceptibility of them to P. aeruginosa intradermal infection. Also, the decreased production of antimicrobial peptides in the Gr-1+CD11b+ cell-inoculation site of these mice was demonstrated. These results indicate that sepsis stemming from burn-wound infection of P. aeruginosa is up-regulated by Gr-1+CD11b+ cells, which are inhibitory on the production of antimicrobial peptides by epidermal keratinocytes.
MATERIALS AND METHODS
Mice
Seven- to 11-week-old male BALB/c mice purchased from The Jackson Laboratory (Bar Harbor, ME, USA) were used. The Institutional Animal Care and Use Committee (IACUC) of The University of Texas Medical Branch at Galveston (Galveston, TX, USA) approved all procedures for animal experiments performed in this study (IACUC Approval Number 04-04-019).
Reagents and media
Recombinant murine β-defensin 1 (MBD-1), MBD-2, MBD-3, MBD-4, and antibodies directed against these peptides were purchased from Alpha Diagnostic International (San Antonio, TX, USA). PE-conjugated anti-Gr-1 and FITC-conjugated anti-CD11b mAb (BD Biosciences, San Diego, CA, USA), FITC-conjugated anti-CD40 mAb (R&D Systems, Minneapolis, MN, USA), and FITC-conjugated anti-CD31 and anti-F4/80 mAb (BioLegend, San Diego, CA, USA) were used. Anti-F4/80 mAb was obtained from eBioscience (San Diego, CA, USA). DNase I was purchased from Sigma-Aldrich (St. Louis, MO, USA). RPMI-1640 medium supplemented with 10% FBS, 2 mM L-glutamine, and antibiotics (100 U/ml penicillin and 100 µg/ml streptomycin) was used as culture media for cells isolated from sites surrounding the burn area. Ca2+-free MEM (BioWhittaker, Walkersville, MD, USA), supplemented with 9% Ca2+-removed FBS, 4 ng/ml epithelial cell growth factor (Invitrogen, Carlsbad, CA, USA), and 0.05 mM CaCl2, was used as culture media for keratinocytes (keratinocyte media) [23
].
Thermal injury
Mice were anesthetized with pentobarbital (40 mg/kg, i.p.), and their back hair was shaved using an electric clipper. Thermal injury was induced by pressing a custom-made, insulated mold (with a 4x5-cm window) firmly against the shaved back of each mouse and subsequently immersing the area into 60°C water for 45 s. This procedure produced a third-degree burn on 25% total body surface area on a 27-g mouse [11
, 12
]. Immediately after thermal injury, physiologic saline (2 ml/mouse) was injected i.p. for fluid resuscitation. Sham burn mice (control mice) were not immersed into hot water, but they were anesthetized, and their back hair was shaved. Physiologic saline (2 ml/mouse, i.p.) was injected into these mice in the same manner as the burn group. All mice exposed to the burn injuries survived when they were not infected.
P. aeruginosa wound infection
The strain 180 of P. aeruginosa (American Type Culture Collection, Manassas, VA, USA) was used throughout the study. P. aeruginosa was grown in brain-heart infusion broth for 18 h at 37°C (shaking, 25 ml in a 250-ml Erlenmeyer flask). P. aeruginosa was harvested by centrifugation, washed twice, and diluted with physiological saline for appropriate concentrations. To induce sepsis stemming from burn-wound infection, P. aeruginosa at various doses was injected beneath the burn tissues 12 h after burn injury (0.2 ml/mouse, 0.5 ml insulin syringes with 27-gauge needles). The number of bacteria in the inoculum was confirmed by plating serial dilutions on brain-heart infusion agar and counting colonies 24 h after incubation at 37°C. The severities of infection were evaluated by their survival and growth of P. aeruginosa at burn-site tissues and in circulation. To determine the mortality rates, all mice were monitored every 12 h for 7 days after infection. To measure the quantity of bacteria in local burn tissues, mice were killed by cervical dislocation 6–24 h after the infection. Burn-wound tissues prepared as outlined below were cut and homogenized in 1 ml cold saline. Also, peripheral blood was collected from these mice by cardiac puncture 6–24 h after infection. Numbers of P. aeruginosa in the 1-g tissue (wet weight) or 1 ml blood were determined by serial dilution plate-count done in duplicate on brain-heart infusion agar.
Tissue surrounding the burn areas
Tissues surrounding the entire burn area were excised using a sterile biopsy punch (8 mm diameter, Sklar Instruments, West Chester, PA, USA). Four punch biopsies were randomly excised per mouse from tissues bordering 1 cm around the thermally injured area. The depth of the skin biopsy extended all the way to the skeletal muscles of the back such that all epidermal, dermal, and panniculus carnosus components were removed. Immediately following excision, the tissues were weighted and rinsed in cold, sterile PBS containing 1% antibiotics mixture (penicillin 100 U/ml, streptomycin sulicoricehate 100 µg/ml, and amphotericin B 0.25 µg/ml) for 2 min at 4°C. These tissues were used to measure antimicrobial peptides and as a source of inhibitor cells on the antimicrobial peptide production by skin keratinocytes. Also, tissues from mice infected with P. aeruginosa beneath their burn wound were used as specimens to measure bacterial growth at the infection sites.
Isolation of inhibitor cells
To separate dermal tissues from epidermis, tissues surrounding the burn areas obtained above were placed in PBS without Ca2+ and Mg2+ (–PBS) containing 0.25% trypsin/EDTA overnight at 4°C. Then, epidermal sheets and dermal tissues were separated using forceps. Dermal tissues obtained were cut into small pieces and digested by 0.025% DNase I in –PBS for 40 min at 37°C. The suspension was filtered through a sterile 60-µm steel mesh. The resulting cells were treated with magnetic beads coated with anti-Gr-1 mAb for 1 h at 4°C. Then, Gr-1+ cells were isolated using a magnet. Furthermore, Gr-1+ cells were labeled with FITC-conjugated anti-CD11b, anti-CD40, anti-CD31, anti-I-A/I-E, or anti-F4/80 mAb for 30 min at 4°C. Then, the expression of CD11b, CD31, CD40, I-A/I-E, and F4/80 by Gr-1+ cells was analyzed by a FACSCanto flow cytometer (Becton Dickinson, San Jose, CA, USA). To isolate Gr-1+CD11b+ cells, Gr-1+ cells were prepared from single-cell suspension by positive cell sorting using magnetic beads coated with anti-Gr-1 mAb. After selection, beads were detached using the detach-bead enzyme mixture (Dynal, Great Neck, NY, USA). Then, CD11b+ cells were additionally isolated from Gr-1+ cell preparation using CD11b magnetic particles (BD Biosciences).
Isolation of epidermal keratinocytes
The dorsal skin (6x5 cm) of sham burn mice was excised and incubated with –PBS containing 0.25% trypsin and 0.03% EDTA overnight at 4°C. The epithelium (epidermis and hair follicles) was separated from the underlying dermis with fine forceps. Then, epidermis obtained was placed in an Erlenmeyer flask containing Eagle’s minimal essential medium supplemented with 10% heat-inactivated FBS and 4 ng/ml epithelial cell growth factor (keratinocyte media) and mixed with a magnetic mixing bar at room temperature for 1 h [23
]. The freshly isolated epidermal cells were cultured in fibronectin-coated culture plates [23
]. Ten minutes after cultivation, the unattached cells were removed by a gentle wash of the cell layer and recultured in keratinocyte media. During the cultivation, one-half of the culture media was replaced with fresh keratinocyte media twice a week. Two weeks after primary cultivation, the cells were detached using a mixture of 0.25% trypsin and 0.03% EDTA and subcultured two times in keratinocyte media. The purity of keratinocytes was more than 92% when these cells were analyzed for their cell-surface expression of 
vβ6 integrin by flow cytometry [24
]. The cells that detached from the culture flasks were stored in liquid nitrogen by suspending 5 x 105 cells in 1 ml of the keratinocyte media containing 10% DMSO. Cells regrown from frozen stocks were used for the experiments throughout. Consistently, 3–4 µg/ml MBD-1 and 0.4 mg/ml MBD-4 were detected in culture fluids from 2 x 106 cells/ml of these cells 48 h after cultivation.
Measurement of Gr-1+CD11b+ suppressor activity on antimicrobial peptide production
Cells isolated from the burn-site area were functionally assayed based on their ability to suppress antimicrobial peptide production. Thus, epidermal keratinocytes (2x106 cells/ml) were cocultured with the putative suppressor cells (1x105–1x106 cells/ml). Forty-eight hours after cultivation, culture fluids were obtained by centrifugation (650 g, 10 min). Culture fluids obtained from the individual culture of keratinocytes and Gr-1+CD11b+ cell preparations under the same conditions served as controls. These culture fluids were assayed for antimicrobial peptides by ELISA.
Measurement of antimicrobial peptides
Tissues surrounding the burn areas were excised (as mentioned above) and homogenized on ice for 20 s using an Omni tissue homogenizer (Omni International, Marietta, GA, USA) in 1 ml PBS supplemented with 0.05% proteinase inhibitor mixture (Sigma-Aldrich). The homogenates were centrifuged (1000 g, 20 min), and supernatants of the homogenates were assayed for antimicrobial peptides. MBD-1, MBD-2, MBD-3, and MBD-4 in the assay samples were measured by ELISA according to the manufacturer’s protocols. Also, the amount of MBD-1, MBD-2, MBD-3, and MBD-4 in culture fluids of epidermal keratinocytes was determined by ELISA. The detection limit of MBD-1, MBD-2, MBD-3, and MBD-4 in the culture fluids or tissue homogenates was 5–21 ng/ml.
Also, the antimicrobial activity of skin homogenates obtained above was assayed, as described previously [25 ]. Thus, 100 CFU P. aeruginosa in 90 µl brain-heart infusion broth was added to 10 µl skin homogenate supernatants. One, 2, and 3 h after incubation at 37°C, media were harvested, and the numbers of bacteria were determined by a standard colony-counting method.
Statistical analysis
The results obtained were analyzed statistically using ANOVA test. Survival curves were analyzed using the Kaplan-Meier log rank test. Differences were considered significant at P < 0.05.
RESULTS
Susceptibility of thermally injured mice to P. aeruginosa wound infection
Mice with a third-degree burn over 25% of their total body surface area were prepared by immersing their shaved backs into 60°C water for 45 s [24
]. Twelve hours after burn injury, three groups of 12 mice were infected with 50, 103, and 104 CFU/mouse P. aeruginosa beneath their burn wound, respectively, and observed for their survival until 7 days after infection. Three groups of sham burn mice infected intradermally with 106, 107, or 108 CFU/mouse of the pathogen on their shaved backs were used as controls. In the results, P. aeruginosa at a number of 107 CFU/mouse was shown to be 1 LD50 in sham burn mice (Fig. 1A
), and all of the burn mice infected with 50 CFU/mouse of the pathogen died within 3 days of infection (Fig. 1B)
. As shown in Figure 2
, the growth of bacteria at infection-site tissues of burn mice exposed to 100 CFU/mouse P. aeruginosa was compared with that of sham burn mice infected intradermally with the same number of the pathogen. Blood specimens were also obtained from both groups of mice 1–48 h after infection. The numbers of bacteria in these samples were determined by a colony-counting method. P. aeruginosa was cleared quickly from the infection-site tissues of sham burn mice. However, the bacteria did grow progressively at the infection-site tissues of burn mice (Fig. 2A)
. Also, 6 x 105 CFU/ml bacteria was detected in the blood of burn mice 48 h after infection, when a significant number of pathogen were not detected in the blood of sham burn mice until 48 h after infection (Fig. 2B)
. These results indicate that as compared with sham burn mice, thermally injured mice are susceptible hosts for the local infections of P. aeruginosa.
 View larger version (14K): [in a new window] |
Figure 1. Susceptibility of thermally injured mice to P. aeruginosa wound infection. Mice, 12 h after burn injury (B), were infected with 50 CFU/mouse (12 mice,  ), 103 CFU/mouse (12 mice, •), or 104 CFU/mouse (12 mice,  ) P. aeruginosa beneath the burn wound. As controls (A), 106 CFU/mouse (12 mice, ), 107 CFU/mouse (12 mice, •), or 108 CFU/mouse (12 mice, ) P. aeruginosa was injected intradermally (shaved back) to sham burn mice.
|
 View larger version (15K): [in a new window] |
Figure 2. Numbers of P. aeruginosa at the infection-site tissues and in circulation of thermally injured mice. Mice, 12 h after burn injury (•), were infected with 100 CFU/mouse P. aeruginosa beneath the burn wound. As infection-site tissues (A), tissues bordering 1 cm around the thermally injured area were randomly excised using a biopsy punch from these mice 12–48 h after infection. Under the same schedule, blood specimens (B) were obtained from infected mice by cardiac puncture. As controls, infection-site tissues and blood specimens were obtained from sham burn mice infected intradermally with the same number of P. aeruginosa (). Numbers of P. aeruginosa in these specimens were determined by a standard colony-counting method. *, P < 0.05; **, P < 0.01; ***, P < 0.001, versus controls.
|
 View larger version (19K): [in a new window] |
Figure 3. Decreased antimicrobial activities in skin tissues surrounding the burn areas. Skin tissue specimens were excised from shaved backs of sham burn mice or sites surrounding the burn areas of mice 12 h (A) or 0.5–24 h (B) after thermal injury, as described in the text. These tissues were homogenized and centrifuged. (A) Supernatants obtained from the skin homogenates of sham burn mice () or thermally injured mice (•) were mixed with 100 CFU P. aeruginosa and incubated at 37°C. As a control (), the same number of bacteria was incubated with media. One to 3 h after incubation, the number of P. aeruginosa in these samples was counted by a standard colony-counting method. Data are displayed as the mean ± SEM (n=5 in each group). (B) Supernatants of the specimen homogenates were assayed for MBD-1 by ELISA. (C) Supernatants of the specimen homogenates were assayed for MBD-2, MBD-3, and MBD-4 by ELISA. Data are displayed as the mean ± SEM (n=5 in each group). *, P < 0.01; **, P < 0.001, versus skin specimens of sham burn mice.
|
 View larger version (36K): [in a new window] |
Figure 4. Impaired production of MBD-1 and MBD-3 by epidermal keratinocytes cocultured with Gr-1+CD11b+ cells from thermally injured mice. Epidermal keratinocytes (2x106 cells/ml) isolated from normal mouse skin were stimulated with (for MBD-3) or without (for MBD-1) 1 µg/ml LPS for 3 h and then cocultured with 5 x 105 cells/ml (A) or 1–10 x 105 cells/ml (B) Gr-1+CD11b+ cells isolated from tissues surrounding the burn area of mice 12 h after burn injury () or Gr-1+CD11b+ cells isolated from spleens of sham burn mice (•). Culture fluids harvested 48 h after cultivation were assayed for MBD-1 and MBD-3 by ELISA. Data displayed are representative of three experiments. *, P < 0.01; **, P < 0.001, versus controls.
|
 View larger version (18K): [in a new window] |
Figure 5. Flow cytometry analysis of Gr-1+CD11b+ cells isolated from tissues surrounding the burn area. Gr-1+CD11b+ cells (1x106 cells) from tissues surrounding the burn area of mice 12 h after burn injury and those from spleens of sham burn mice were stained with FITC-conjugated anti-CD31, anti-CD40, anti-I-A/I-E, or anti-F4/80 mAb (black lines) or isotype antibody (solid histograms) and analyzed by flow cytometry.
|
 View larger version (35K): [in a new window] |
Figure 6. Decreased MBD-1 production around the Gr-1+CD11b+ cell-inoculation site in sham burn mice, which were inoculated intradermally with 1 x 106 cells/mouse Gr-1+CD11b+ cells, which were isolated from burn-site tissues of mice 12 h after thermal injury. Twenty-four hours after cell inoculation, the tissues around the cell-inoculation site were harvested using a skin-biopsy punch and homogenized. As controls, MBD-1 production in tissues around intradermally injected MEM or 1 x 106 cells/mouse Gr-1+CD11b+ cells from spleens of sham burn mice was assayed in the same manner. Data displayed are representative of three experiments. *, P < 0.001, versus controls.
|
 View larger version (10K): [in a new window] |
Figure 7. Effect of Gr-1+CD11b+ cells on the resistance of sham burn mice to P. aeruginosa skin infection. Sham burn mice were inoculated intradermally with 1 x 106 cells/mouse Gr-1+CD11b+ cells, which were isolated from burn-site tissues of mice 1 h (12 mice, ), 6 h (12 mice,  ), or 12 h after thermal injury (16 mice, •). Twelve hours after cell inoculation, 103 CFU/mouse P. aeruginosa was infected at the Gr-1+CD11b+ cell-inoculation site in sham burn mice. As controls, sham burn mice injected intradermally with MEM (12 mice, ) or 1 x 106 cells/mouse Gr-1+CD11b+ cells from spleens of sham burn mice (12 mice,  ) were infected in the same manner. Data displayed are representative of three experiments. *, P < 0.001, versus controls.
|
 View larger version (22K): [in a new window] |
Figure 8. Mechanisms by which burn-associated Gr-1+CD11b+ cells inhibit MBD-1 production by keratinocytes. (A) Cytokine-producing profiles of Gr-1+CD11b+ cells (2x106 cells/ml), which were cultured for 48 h without any stimulation. Culture fluids harvested were assayed for IL-4, IL-10, IL-13, and CCL2 by ELISA. (B) Effector molecules for Gr-1+CD11b+ cell-mediated suppression of MBD-1 production by keratinocytes. In the presence or absence of anti-CCL2/anti-IL-10 mAb mixture (5 µg/ml each), epidermal keratinocytes (2x106 cells/ml, lower chamber) isolated from normal mouse skin were cultured with 1 x 106 cells/ml Gr-1+CD11b+ cells from tissues surrounding the burn area of mice 12 h after burn injury or Gr-1+CD11b+ cells isolated from spleens of sham burn mice. Culture fluids harvested 48 h after cultivation were assayed for MBD-1. *, P < 0.001, versus transwell cultures without mAb. (C) Effect of a mixture of recombinant CCL2 and IL-10 on MBD-1 production by keratinocytes. Epidermal keratinocytes (2x106 cells/ml) from normal mouse skin were cultured with a mixture of IL-10 and CCL2 (10 ng/ml each) for 48 h. Culture fluids harvested 48 h after cultivation were assayed for MBD-1. Data displayed are representative of three experiments. *, P < 0.001, versus controls.
|
In the present study, why antimicrobial peptides are not produced at the burn-site tissues and how this defect contributes to the increased susceptibility to P. aeruginosa burn wound infection were studied in a mouse model of thermal injury. A 50-CFU/mouse dose P. aeruginosa infected beneath the burn wound was shown to be lethal in burn mice, and 107 CFU/mouse of the intradermal pathogen was shown to be 1 LD50 in sham burn mice. Skin homogenates from sham burn mice contained antimicrobial peptides, and those from thermally injured mice did not. Gr-1+CD11b+ cells, which were obtained from tissues surrounding the burn area, were shown to be inhibitory on antimicrobial peptide production by skin keratinocytes. These cells were not isolated from the skin tissues of sham burn mice. Gr-1+CD11b– cells without an ability to inhibit antimicrobial peptide production were isolated from the skin tissues of sham burn mice. After the intradermal inoculation of these Gr-1+CD11b+ cells, the susceptibility of sham burn mice to skin P. aeruginosa infection was greatly increased. These results indicate that sepsis stemming from P. aeruginosa burn-wound infection is accelerated by Gr-1+CD11b+ cells appearing in response to burn injuries. However, as shown in Figure 7 , the resistance of sham burn mice to the pathogen did not completely disappear in sham burn mice inoculated with Gr-1+CD11b+ cells. This suggests that the susceptibility of burn mice to P. aeruginosa wound infection is additionally influenced by the other defects of host antibacterial resistance.
Antimicrobial peptides are natural peptides with potent, broad-spectrum, antibacterial activities [13
, 14
]. One family, defensins, is only 29–47 aa in length. Defensins act preferentially on microbes by permeabilizing microbial membranes rich in anionic phospholipids, with relative sparing of the host’s cholesterol- and neutral-phospholipid-rich cell membranes. Based on the pattern of their disulfide bonding, defensins are mainly classified into and β categories. - and β-defensins have broad-spectrum, antimicrobial activities against bacteria, fungi, and some enveloped viruses. Cathelicidins are also inhibitory against a wide range of bacteria, fungi, enveloped viruses, and protozoa [30
]. Cathelicidins (human, LL-37; mice, cathelin-related antimicrobial peptide) are structurally and evolutionally distinct from defensins in abundance and distribution.
Previous human studies described the presence of skin-localized immunodeficiency that allowed the entry of opportunistic microbes into the skin, and there were positive correlations between the decrease in antimicrobial peptide production by keratinocytes and the increase in surface infection [16 , 19 , 22 ]. Similar results were obtained in our studies using a mouse model of thermal injury. Thus, the same-sized skins were obtained from shaved backs of sham burn mice and tissues surrounding the burn area, and the activity of antimicrobial peptides in their skin homogenates was compared by the bioassay (antibacterial activity) and ELISA (MBD-1 and MBD-4). In the bioassay, antibacterial activities were demonstrated in the skin homogenates of sham burn mice, and the homogenates of skin tissues surrounding the burn area from mice 12 h after burn injury did not exhibit these activities. MBD-1 and MBD-4 were detected in the skin homogenates from sham burn mice. However, these antimicrobial peptides were not detected in the homogenates of skin tissues surrounding the burn area of mice. Subsequently, the accumulation of Gr-1+CD11b+ cells to tissues surrounding the burn area was demonstrated. When these cells were intradermally transferred to sham burn mice, decreased amounts of MBD-1 production were detected around the cell-inoculation site of these mice. In the in vitro studies, the production of MBD-1 and MBD-3 by epidermal keratinocytes was clearly inhibited by Gr-1+CD11b+ cells.
It has been described in recent papers that Gr-1+CD11b+ cells are demonstrated in mice 8–14 days after small scald burn injury [31 ], 1–3 days after surgical stress [27 ], and 5–84 days after peritonitis induced by cecal ligation and puncture (CLP) [32 ]. Gr-1+CD11b+ cells from surgically stressed mice and mice with CLP-induced sepsis expressed immature, cell-specific surface antigens (CD31, CD34) and F4/80 antigen. Gr-1+CD11b+ cells isolated from mice with scald burn injury that covered only 14% of their backs expressed macrophage progenitor markers (CD115, CD117) but not a monocyte/macrophage marker (F4/80 antigen). In our studies, Gr-1+CD11b+ cells with CD31 and F4/80 antigens were isolated from mice 12 h after severe burn injury (25% total body surface area burn). In our subsequent studies, two different Gr-1+CD11b+ IMC were demonstrated in spleens of the same severely burned mice. One was detected in mice 4–12 h after burn injury, and the other was demonstrated in mice 3–14 days after burn injury. Mice inoculated with either one of the IMC were shown to be susceptible to the infection. The suppressor cell activity of IMC appeared early after the injury was shown in a dual-chamber transwell, and cell-to-cell contact was required when IMC appearing late after the injury suppressed antimicrobial peptide production by keratinocytes. Early IMC did not express CD40 antigen, and late IMC expressed this surface antigen. The increased expression of CD40 has already been reported for Gr-1+CD11b+ cells isolated from surgically stressed mice [27 ]. All of these results suggest a possibility that Gr-1+CD11b+ cells described in this paper are different subsets from Gr-1+CD11b+ cells reported in earlier studies performed in mice with a smaller scald burn and mice with surgical stress. Further analytical studies for different sources of Gr-1+CD11b+ cells will be required.
Gr-1, CD11b, and F4/80 antigens have been shown to be expressed on the surface of IMC and monocytes [26 27 28 29 ]. IMC with immunosuppressive activities have already been demonstrated in the spleens of mice inoculated with tumor cells [33 34 35 ], exposed to protozoan infection [36 ], or subjected to graft-versus-host diseases [37 ]. Vascular endothelial growth factor (VEGF) and GM-CSF have been shown as inducers of IMC in tumor-bearing mice [33 , 38 , 39 ]. The increase in the plasma levels of these growth factors has been reported in patients and mice with thermal injuries [40 , 41 ]. In a clinical study of 36 thermally injured patients, circulating VEGF levels were immediately elevated on Day 1, reached their highest level on Day 14, and decreased to the normal range on Day 35 [40 ]. Therefore, these growth factors elevated in thermally injured hosts might participate in the elicitation of Gr-1+CD11b+ cells. On the other hand, various immunosuppressive activities in a certain population of monocytes have been reported [42 , 43 ]. These monocytes possessed the abilities to produce CCL17, CCL18, and CCL22. These are typical properties, which are displayed by alternatively activated monocytes [42 , 43 ]. Thus, Gr-1+CD11b+ cells with the ability to inhibit antimicrobial peptide production by epidermal keratinocytes may be composed of IMC and alternatively activated monocytes.
ACKNOWLEDGEMENTS
This work was supported by Shriners of North America grant 8610.
Received August 11, 2007; revised January 28, 2008; accepted February 25, 2008.
REFERENCES
/β-dependent, IFN-
secretion by bone marrow-derived macrophages controls an intracellular bacterial infection J. Immunol. 167,6453-6461
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
