Introduction

To forensic pathologists, wound age provides valuable information for the reconstruction of crime scenes and determination of the cause of death. As skin covers the outer surface of the body, it is the most vulnerable part, and dermal wound age is a critical issue in routine forensic autopsies. Inflammatory cell dynamics [1] and enzyme activities, such as ATPase, acid phosphatase, and alkaline phosphatase [10] are still applied to wound age estimation. Analyses of cytokine expressions in murine skin and immunohistochemical detection in human skin have also been reported, and their usefulness for dermal wound age estimation has been suggested [9, 16, 22, 31, 32].

Although cytokine mRNA can be detected by real-time polymerase chain reaction (PCR), and secreted cytokines can be quantified with the enzyme-linked immunosorbent assay, these methods have some limitations, for example, the need for a large sample volume. In addition, these techniques are laborious and time-consuming. Therefore, multiplexed examination of tissue proteins has been extremely difficult. These problems have been removed by the commercial availability of a bead-based immunoassay (Bio-Plex Suspension Array System, Bio-Rad, Hercules, CA). Although the ability to simultaneously quantify multiple signaling proteins would be beneficial to deciphering how they affect dermal wound healing, this technology has not been applied to the study of cytokine expression during dermal wound healing.

In this study, quantitative analyses of the time-dependent expressions of eight cytokines, including interleukin 2 (IL 2), interleukin 4 (IL 4), interleukin 6 (IL 6), interleukin 8 (IL 8), interleukin 10 (IL 10), granulocyte-macrophage colony-stimulating factor (GM-CSF), interferon γ (IFNγ), and tumor necrosis factor α (TNF α), in human dermal penetrating wounds, were performed to evaluate their utility for wound age estimation using bead-based immunoassay.

Materials and methods

Samples

We collected samples of human skin from skin penetrating injuries including incised wounds, stab wounds, laceration, and contusions, that could be dated chronologically from depositions taken at the scene of injuries. The postmortem interval was within 36 h. Skin samples 0.5 × 0.5 cm were taken parallel along the wound margin. The subcutis was then removed. Tissue samples from 104 dermal wounds taken from 29 males and 30 females were investigated. A total of 16 control samples from the left clavicular area taken from eight males and eight females were analyzed simultaneously (Table 1). The specimens were categorized into the following 10 groups according to wound age: 0.1–0.5 h (n = 17), 1–1.92 h (n = 15), 2–2.75 h (n = 13), 3–4.5 h (n = 14), 6–7 h (n = 11), 9 h (n = 8), 12–14.75 h (n = 10), 33–49 h (n = 5), 71–116.5 h (n = 6), 246–744 h (n = 5), and intact control tissue (n = 16).

Table 1 Samples analyzed in the present study
Full size table

Individuals with a known history of immunodeficiency or immunotherapy were excluded from the study.

Immunohistochemical studies

To identify cells infiltrated into the skin, immunostaining was performed using mouse anti-neutrophil elastase (Spring Bioscience, Fremont, CA) for neutrophils, mouse anti-CD68 (Zymed Laboratories, South San Francisco, CA) for macrophages, mouse anti-CD 3 (Nichirei, Tokyo, Japan) for T lymphocytes, mouse anti-L 26 (Nichirei, Tokyo, Japan) for B lymphocytes, and rabbit anti-vimentin (Abcam, Cambridge, UK) for fibroblasts as the primary antibodies.

Extraction of total protein

The skin samples were washed with Bio-Plex Cell Wash Buffer (Bio-Rad Laboratories), and pulverized mechanically in an oscillating mill with continuous addition of liquid nitrogen (Cryo-Press, Microtec, Funabashi, Japan). The resulting product was homogenized in a tube (MAZERA Z, EYALA, Tokyo, Japan) containing 500 μl of lysing solution (Bio-Plex Cell Lysis Kit, Bio-Rad Laboratories). Samples were placed in a 1.5-ml tube and frozen at −80°C. After thawing, the samples were sonicated on ice using Bioruptor (Cosmo Bio, Tokyo, Japan), and centrifuged at 6,000 rpm for 6 min at 4°C. The supernatant was filtered with cell strainer (BD, Franklin lakes, NJ), and transferred to a tube. The protein concentration was determined by the Bradford method [2] using a commercial assay kit (Bio-Rad Laboratories). The protein concentration of each sample was adjusted to 500 μg/ml with lysing solution.

Multiplex analysis

A Bio-Plex assay for eight cytokines (IL 2, IL 4, IL 6, IL 8, IL 10, GM-CSF, IFNγ, TNF α) was run following the manufacturer’s instruction (Bio-Rad Laboratories). Briefly, a standard curve was created from fourfold dilution series of premixed standards ranging from 3,200 to 0.2 pg/ml. The assay was performed in a 96-well filtration plate supplied with the Bio-Plex kit. Premixed beads coated with the target antibodies were added to each well, then washed twice with Bio-Plex Wash Buffer. Premixed standards and samples (50 μl) were then added to the wells, before shaking at 1,100 rpm for 30 s and incubation for 30 min together with shaking at 300 rpm at room temperature. Subsequently, wells were washed three times with Bio-Plex Wash Buffer, and 25 μl of the premixed detection antibodies was added, before shaking at 1,100 rpm for 30 s and at 300 rpm for 30 min at room temperature. Wells were again washed three times with Bio-Plex Wash Buffer, and 50 μl streptoavidin-phycoerythrin was added to the wells. After incubation for 10 min with shaking at 300 rpm, the wells were washed three times with Bio-Plex Wash Buffer, and the beads were resuspended in 125 μl Bio-Plex Assay Buffer. Using the Bio-Plex Suspension Array System, the factor was detected by the fluorescence of the beads, and simultaneously quantified by that of phycoerythrin. All samples were analyzed in duplicate.

Evaluation

The ratios of cytokine expression to total protein and cytokine concentration in dermal tissue were analyzed. Furthermore, differences of cytokine expressions between control and each injured tissue were statistically analyzed using the Dunnet test, and P values of 0.05 or less were considered statistically significant.

The research described in this report was conducted in accordance with the guidelines for forensic experimentation of Japanese Society of Legal Medicine.

Results

Immunohistochemical analysis within 5 mm from the edge

Neutrophils and macrophages were detected from 2 h post-injury, and strong infiltration was seen at 33–49 h. T and B lymphocytes also infiltrated simultaneously from 71 h. Although fibroblasts appeared in the intact skin, strong proliferation was shown from 246 h in injured skins. Thickening of the epidermis was seen from 71 h (Table 2).

Table 2 Immunohistochemical analysis of cellular dynamics during dermal wound healing within 5 mm from the edge
Full size table

Cytokine expressions per total protein

Time-dependent cytokine expressions per total protein are shown in Fig. 1. IL 10, GM-CSF, IFN γ, and TNF α increased from the early phase of dermal wound healing. TNF α showed a significant increase with very short survival times of less than 0.5 h, whereas rises of IL 10, GM-CSF, and IFN γ were seen until 1.92 h. These cytokines were active throughout the healing process, and peaks of TNF α at 9 h, IFN γ at 12–14.75 h, GM-CSF at 33–49 h, and IL 10 at 71 to 116.5 h were observed in the middle and late phases. IL 6 showed exclusive expressions in the middle phase, and peaked at 9 h. IL 2, IL 4, and IL 8 were mainly seen from the middle phase to the late phase. Peak expressions of IL 2 and IL 8 at 9 h, and IL 4 at 33–49 h were seen. On the whole, IL 6, IL 8, IFN γ, and TNF α were strongly expressed. Among the cytokines analyzed in the present study, IL 8 was the most strongly expressed.

Fig. 1
figure 1

Time course of cytokine expressions per total protein (cytokine/protein) and per tissue weight (cytokine/tissue) during dermal wound healing. The results represent the mean + SD per group.

Full size image

Cytokine expressions per tissue weight

Cytokine expressions per tissue weight were similar to those per total protein (Fig. 1). Increases of IL 10, GM-CSF, and IFN γ until 1.92 h were detected. Peaks of TNF α at 9 h, IFN γ at 12–14.75 h, and GM-CSF and IL 10 at 71 to 116.5 h were observed in the middle and late phases. IL 6 also peaked at 9 h. IL 2, IL 4, and IL 8 were mainly seen from the medium phase to the late phase, and peaked at 71–116.5 h. Overall, strong expressions of IL 6, IL 8, IFN γ, and TNF α were seen. In addition, the most abundant factor was IL 8.

Discussion

Dermal wound healing is a sequential biological process that involves the integration of chemotaxis of inflammatory cells, migration of keratinocytes, and remodeling of the scar tissue [26], and cellular infiltration has been considered to be one of the important tools for wound age estimation. Previous morphological analyses demonstrated that neutrophils and macrophages were detected during very early phase of dermal wound healing from 30 min and 3 h post-injury, respectively, whereas, lymphocytes were detected from 192 h [1]. In the present immnohistochemical analyses, neutrophils and macrophages simultaneously appeared from 2 h and strongly infiltrated at 33–49 h, whereas T and B lymphocytes infiltrated simultaneously from 71 h. These findings suggest few differences in behavior between nentrophils and macrophages, and remarkable changes of inflammatory cell dynamics were seen chiefly in the late phase. Neutrophils senescent within viable tissue are phogocytosed by macrophages [21], and the present study demonstrated that these phenomena would start from the early phase of dermal wound healing. Thus, it was considered that morphological analyses of cellular infiltration and proliferation were unsuitable for the markers of the early phase, in spite of many cases of a short survival time in forensic cases. On the other hand, reproducible time-dependent changes of cytokine expressions, which precede those of inflammatory cells, were observed in the present study. This indicates that quantitative analyses of these cytokines in tissue extracts can contribute to wound age estimation. Compared with cytokine expressions per tissue weight, significant changes of those per total protein were obvious, indicating that the latter would be useful as parameters. This dissociation between total protein and tissue weight may have resulted from postmortem water loss in organs and tissues. The potential parameters would be IL 10, GM-CSF, IFN γ, and TNF α in the early phase of dermal wound healing, IL 6 in the middle phase, and IL 2, IL 4, and IL 8 in the middle and late phases. The estimated ranges of wound age were rather wide in the immunohistochemical evaluations [24], whereas quantitative analyses of tissue extracts could indicate narrower intervals. Compared with the data obtained from animal experimental studies and previous histochemical analyses of human samples [11, 14, 15, 23], the early expression of IFN γ in the present study would appear significant. On the other hand, the expression patterns of IL 6, IL 8, IL 10, and TNF α resembled those reported in previous studies.

Cytokines can alter the behavior and properties of immune and other cells, and their combined effects are often more important than the function of one isolated component. Therefore, elucidating the interactive behavior of cytokines during dermal wound healing is essential for understanding dermal immune responses. Among the cytokines appearing in the early post-traumatic period, TNF α originates in keratinocytes [13], and GM-CSF in keratinocytes [3], endothelial cells [4], and fibroblasts [6]. Because TNF α promotes IL 10 production in macrophages [33], its expression would increase successively as shown in the present study. As these cytokines were seen throughout the posttraumatic period, they would play an important role in wound healing.

On the other hand, IFN γ was reported to have a chemotactic effect on neutrophils [17] and macrophages [20] during the early phase of wound healing. In the middle phase, IL 6 could be produced by keratinocytes [12] and fibroblasts [28] in response to TNF α, and has also been reported to express in endothelial cells, vascular smooth muscle cells [25], and macrophages [5].

Factors mainly detectable from the middle phase to the late phase were IL 2, IL 4, and IL 8. IL 8, which was the most plentiful cytokine in the present study, has been demonstrated to originate from keratinocytes [18], fibroblasts, endothelial cells [29], and neutrophils [30]. In addition, IL 8 proliferates keratinocytes and works as a potent chemokine for neutrophils [26] and lymphocytes [34]. IL 2 promotes T lymphocyte proliferation [19] and interacts with IFN γ on IL 8 mRNA production in keratinocytes [18]. On the other hand, IL 4 is thought to proliferate fibroblasts [7]. Significant expressions of IL 6, IL 8, IFN γ, and TNF α were suggestive of their having significant effects on dermal wound healing.

Quantitative analyses of cytokines at the wound site can be quite useful for wound age estimation, and multiplex analyses would be a promising method. Previous reports of cytokine interactions were performed with data collected by different methods and from different samples [8, 27], whereas the present multiplex data were gained from the same samples with a single method. Therefore, it is considered that more accurate interactions of cytokines could be demonstrated. Moreover, unlike previous dermal quantitative studies on cytokines mainly conducted by means of animal experiments, human penetrating wounds were analyzed directly in the present study. On the other hand, difference of age, sex, and type of wound, which might affect the cytokine pattern, could not been take into account in this study, because of relatively limited number of samples. Further investigations with more detailed classifications are required. Although the present study was oriented to practical forensic pathology, the data obtained would also be informative for various fields of medicine.