Wound healing after burn is a complex pathophysiological process, including inflammation, cell proliferation, wound contraction and collagen metabolism and encompass complex interaction between cells, and with the extracellular matrix and cytokines (1). The main wound-repairing cells are the fibroblasts (1), which undergo a series of phenotypic changes in post-traumatic period, leading to activation, proliferation, migration, synthesis and secretion of extracellular matrix, including collagen, ultimately resulting in scar formation and wound remodeling (2). However, the regulatory mechanisms responsible for the succinate changes during wound repair are poorly understood. Burn-related hypertrophic scar formation impacts patients' quality of life (3). Therefore, accelerating wound healing, preventing and/or reducing hypertrophic scar formation is one of the main objectives of burn research.
The rate of healing of burn wounds is influenced by the physical, nutritional status, systemic factors and certain extraneous factors (3). Burn injury-mediated destruction of the skin barrier normally induces microbial invasion, in turn leading to the development of systemic infection and septic shock by the release of endotoxins. Endotoxins can cause destruction of cellular components and extracellular matrix through its decomposition, slow wound healing and even cause wounds to deepen and expand (4).
In this study, we investigated (a) the influence of lipopolysaccharide (LPS) on the biological characteristics of the normal skin fibroblast proliferation and collagen synthesis to explore the possible role of LPS in early wound healing and (b) whether prolonged LPS treatment renders hypertrophic scar tissue phenotype and behavior to normal skin fibroblasts.
Ethics Review and Informed Consent
The study protocol was reviewed and approved by the Ethics Committee of The First Hospital Affiliated to the Chinese PLA General Hospital (No. 20110016). All enrolled patients were required to provide signed informed consent to participate in the study.
Fibroblast Cell Culture from Skin Tissue Explants and LPS Treatment
Fibroblasts were isolated from hypertrophic scar and site-matched normal skin tissue explants of the same subject from 20 patients with hypertrophic scar in proliferative stage and cultured in vitro as described previously (5, 6). Passage 3 cells were used for subsequent experiments. Normal skin and hypertrophic scar tissue fibroblasts were treated with increasing concentrations of LPS (Sigma, St. Louis, MO) (0.005–1 μg/mL) for indicated time points.
Fibroblast Proliferation Activity
Cell proliferation was evaluated by 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide (MTT) reduction assays (BIOLEAF Biotechnology, China) as per the manufacturer's instructions using logarithmic growth phase cells (1 × 104 cells/mL) plated in 96-well plates and treated with LPS up to 9 days.
Fibroblasts were treated with LPS for 6 days before being harvested by trypsinization (to avoid losing floating cells) and stained with propidium iodide (20 μg/mL) (BIOLEAF Biotechnology, China). Flow cytometry was performed by using a FACSCalibur (BD Biosciences, San Jose, CA), equipped with the filter set for quadruple fluorescence signals. At least 50,000 positive events were measured and data were analyzed using CellQuest Pro software (BD Biosciences).
Excised scar tissue and normal skin explants were fixed with paraformaldehyde (4%, 18 h, 4 °C) and postfixed (70% ethanol, 16 h) before dehydration and paraffin embedding. Paraffin sections were stained with hematoxylin/eosin as described earlier (7).
Immunohistochemistry was performed according to standard protocols. Briefly, fibroblasts were blocked and incubated with primary antibody (α-Type I Collagen: 1:50; α-Type III Collagen: 1:50; α-Vimentin: 1:25; α-proliferating cell nuclear antigen [α-PCNA]: 1:25; α-smooth muscle actin (α-SMA): 1:25) (Zhongshan Golden Bridge Biotechnology, Beijing, China) for 2 h at 4 °C. The sections were washed with PBS, before being incubated with biotinylated goat antimouse IgG secondary antibody for 25 min at 25 °C. The stain was developed with diaminobenzidine tetrahydrochloride chromogen and observed under Leica DM 2000 microscope (Olympus, Japan).
Determination of Collagen Synthesis
Pepsin-resistant, extracellular collagen synthesis by confluent fibroblast monolayers was quantified by [3H]-proline incorporation into collagens. Fibroblasts were treated with LPS for 6 days before addition of [3H]-proline (0.5 μCi/well) for 24 h. Data are expressed either as cpm of [3H]-proline incorporation in 103 cells or were analyzed as described previously (8).
Transmission Electron Microscopy
Cultured cells were collected by centrifugation and fixed using 3% glutaraldehyde. After dehydration, they were treated with acetone in 1% osmium tetroxide and then embedded using the conventional EPON812 embedding matrix of ultrathin sections, double staining with uranyl acetate and citrate aluminum before being observed and analyzed using the JEM-200EX transmission electron microscope. Nucleocytoplasmic ratio was determined as described earlier (9). Briefly, contours of cells and nuclei were drawn on each image to construct two binary mask-images—one for the entire cell and the other for just the nuclei. Boolean operation, XOR, was applied to the aforementioned image to obtain the binary mask image of just the cytoplasm. The cytoplasmic and nucleic binary mask images were used to numerically compute the NcArea and CitArea vectors, respectively, for nuclear and cytoplasmic areas. Nucleocytoplasmic ratio (NcRatio) was determined by using the following formula: NcRatio = (NcArea/CitArea); the ratios were calculated for 10 independent cells and were represented as mean ± standard deviation (SD).
All data were analyzed using the SPSS (13.0 version) statistical package for data description, the measurement data of the t-test and analysis of variance. Data are represented as mean ± SD. P-value of <0.05 was considered as significant difference.
Establishment of Fibroblast Explant Culture and Effect of LPS on its Growth and Proliferation
Normal skin (Fig. 1A, top) and hypertrophic scar tissue specimens (Fig. 1A, bottom) were used to obtain the primary explant cultures (Figs. 1B and 1C). Immunohistochemistry with α-type I collagen (Fig. 1D), type III collagen (Fig. 1E) and vimentin (Fig. 1F) confirmed the cultures as bona fide fibroblasts. Fibroblast culture was treated with increasing concentration of LPS for 9 days. Beyond day 3, LPS induced a dose-dependent increase in cell proliferation at concentrations of 0.005–0.10 μg/mL (P < 0.05) (Fig. 2A). However, although LPS of 0.50 μg/mL showed an overall induction of proliferation compared to the untreated cells, it was comparatively lowered than at lower concentrations (P > 0.05). At a concentration of 1.00 μg/mL, LPS induced a significant inhibition of cell proliferation (P > 0.01) (Fig. 2A). The effect of LPS on cell proliferation was corroborated and mimicked by its effect on growth rate (Fig. 2B), with LPS stimulating and inhibiting growth at concentrations of <0.5 and >0.5 μg/mL, respectively (P < 0.05 for growth stimulation and P < 0.01 for growth inhibition compared to control untreated fibroblasts).
Effect of LPS on Cell-Cycle Progression of Fibroblast
We next wanted to investigate if the observed effects after LPS treatment were owing to their impact on cell cycle. After treatments, fibroblasts were stained with propidium iodide and DNA content was analyzed by flow cytometry (Table 1). At low concentrations (0.005–0.100 μg/mL), LPS induced transition from G1 to S phase in a dose-dependent fashion, suggesting active cell proliferation (P < 0.01) (Figs. 3A–3E). At 0.5 μg/mL of LPS, S-phase progression was still significantly higher than the control group (P < 0.01) (Fig. 3F), but LPS of 1.0 μg/mL inhibited progression into S phase (P < 0.01) (Fig. 3G). Taken together (Figs. 1, 2, 3, 1–3), these results suggest that LPS promote or inhibit cell proliferation in a dose-dependent fashion by impacting transition through the G0/G1 into S phase.
Table 1. Cell count percent in different cell-cycle phases after the treatment of human normal skin fibroblast with different concentrations of LPS
98.35 ± 2.34
0.43 ± 0.08
1.23 ± 0.09
91.23 ± 1.02
0.06 ± 0.02
8.71 ± 0.98
86.37 ± 0.57
1.13 ± 0.05
12.51 ± 1.10
88.79 ± 1.96
1.53 ± 0.23
11.78 ± 0.93
89.1 ± 1.23
0.71 ± 0.09
10.2 ± 0.94
90.2 ± 0.09
0.46 ± 0.11
9.34 ± 0.98
93.52 ± 1.99
0.25 ± 0.14
6.23 ± 0.92
Effect of LPS on Collagen Synthesis by the Cultured Fibroblasts
Compared with the control group, LPS of 0.005–0.1 μg/mL resulted in dose-dependent increase in 3H-proline incorporation (P < 0.05) (Fig. 4). LPS of 0.5 μg/mL induced collagen synthesis higher than in the untreated cells, but were comparatively lower than in the low LPS treatment groups. Treatment with LPS of 1.0 μg/mL resulted in significant attenuation of collagen synthesis as is evident by the lowering of 3H-proline incorporation (P < 0.05) (Fig. 4).
Control untreated fibroblasts, normal skin fibroblasts treated with LPS of 0.1 μg/mL and hypertrophic scar tissue ultrastructural morphology were compared by transmission electron microscopy (TEM) (Fig. 5). Treatment with LPS of 0.1 μg/mLS resulted in diversified morphological changes in the ultrastructure of fibroblasts after different generation of passages (generations 4–10, Figs. 5A–5D), resulting in large cells with high nuclear cytoplasm ratio (0.47 ± 0.21 in normal skin fibroblasts, 0.62 ± 0.13 in generation 4, 0.73 ± 0.02 in generation 6, 0.85 ± 0.12 in generation 8, 0.89 ± 0.01 in generation 10 and 0.87 ± 0.09 in hypertrophic scar tissue; P < 0.05 in each case when compared to normal skin fibroblasts; P > 0.05 when generation 8 and hypertrophic scar tissue are compared). In addition, the photomicrographs obtained distinctly showed that treatment with LPS of 0.1 μg/mL resulted in irregular nuclear membrane, and visible cytoplasm of the rough endoplasmic reticulum, microfilaments and microtubules. Interestingly, the ultrastructure of the HTS fibroblasts mimicked the ultrastructure of LPS-treated normal skin fibroblast cells (Fig. 5E). HE staining confirmed similar nucleocytoplasmic ratio and fiber cell morphology in scar tissue and normal skin fibroblasts after eight generations of passage and treatment with LPS of 0.10 μg/mL (Fig. 5F). Immunohistochemical analyses revealed similar expression of PCNA in the nucleus and type-1 procollagen and α-SMA in the cytoplasm of normal skin fibroblasts treated with LPS and hypertrophic scar tissue (Table 2). In fact, collagen synthesis rate as determined by 3H-proline incorporation was similar in the scar tissue (2,133 ± 276 cpm) and LPS-treated fibroblasts (2,235 ± 342 cpm) and was significantly higher than the untreated fibroblasts (772 ± 79 cpm) (P < 0.01 in each case is compared to the untreated group).
Table 2. Phenotypic changes comparison among hypertrophic scar tissue or explant fibroblasts (generated from normal skin) treated with LPS of 0.1 μg/mL for more than eight generations
Proliferating cell nuclear antigen
Type I collagen
Number of “+” denotes the relative expression levels of the corresponding marker; P > 0.05 for α-SMA for the two groups.
Normal skin fibroblast + LPS
Hypertrophic scar tissue
Gram-negative bacteria are the common microorganisms found in burn wound infection and release endotoxin to the wound surface (10, 11). LPS is the main component of endotoxin, and its biological activity, cytotoxicity and immunological activity dictate, to a large extent, the process of wound healing. Our study thus assumes significance in understanding how natural endotoxin released at the wound microenvironment dictates wound healing.
Our results of LPS-induced stimulation of fibroblast proliferation corroborate to earlier findings by other groups (12); however; we observed similar effects at one-tenth concentration of those used in these studies. Conversely, we found growth inhibition at concentrations of 1 μg/mL in comparison to these studies, showing cell proliferation even at 2 μg/mL. This apparent discrepancy can be possibly explained by the biological potency of LPS from different sources. LPS can promote cellular proliferation either through facilitating intracellular calcium signaling (13) or through activating downstream signaling cascade by stimulating expression of c-myc, c-fos and hsp70 (14), which directly impacts on cell-cycle progression. Alternatively, LPS can induce cell proliferation by activating cytokine secretion which would stimulate downstream prodivision signaling cascades. On the flip side, at high concentrations LPS inhibits cell proliferation by possible inhibition of DNA synthesis (15) and hence causing an arrest in the G0–G1 phase. It is worth noting that MTT assay is highly dependent on the metabolic state of a certain cell type and given that the metabolic status of the normal fibroblasts and scar derived (myo-) fibroblasts differs, and the assay does not provide accurate quantitation of cell proliferation. Moreover, LPS might influence the metabolic status of the cells. However, we also determined cell growth rate (Fig. 2B) and the results corroborate the findings of Fig. 2A.
Given that proline is an important and stable constituent of collagen, and is rarely present in other tissues, 3H-proline incorporation and pepsin digestion method is a reliable indicator of collagen content in the cells. We found that LPS stimulates collagen synthesis at low concentrations and has a diametrically opposite effect at higher concentrations.
Hypertrophic scar results from fibroblastic hyperplasia and is characterized by excessive accumulation of extracellular matrix (6, 16, 17), which is a major difference in comparison to normal skin fibroblasts. In fact, gene expression analysis has shown distinct gene signature expressions in hypertrophic scar tissues (18, 19).
In vitro studies have shown that fibroblasts after LPS challenge can consume glucose-3H-thymidine deoxyribose, in turn enhancing the synthesis of hyaluronic acid intake, suggesting a direct role of LPS in tissue repair and scar formation (16, 17). Hence, we decided to test if normal skin fibroblasts stimulated by LPS and hypertrophic scar fibroblasts passaged to maintain their biological characteristics have similar structural morphology. Our TEM analyses combined with our observation of similar expression levels of PCNA, α-SMA and type I procollagen synthesis shows that the ultrastructure of fibroblasts post-LPS treatment and their own proliferative scar tissue fibroblasts are similar although the underlying mechanisms are still unknown.
It has been suggested that TGF-β, IFN-γ closely regulate hypertrophic scar formation by stimulating fibroblast proliferation, differentiation and promoting the deposition of extracellular matrix (20–22). Moreover, Toll-like receptors (TLRs), especially TLR4 has been indicated to be a major lipid-A recognition protein in the LPS receptor complex (23) and might mediate the signaling cascade pertaining to the inflammatory response involved in hypertrophic scar formation. In fact, TLR4 has been shown to be overexpressed at both the transcript and the protein expression levels in hypetrophic scar tissue obtained from burn wounds in comparison to normal dermal fibroblast explant cultures (24). Hence, it would be interesting to ascertain if the secretion of different cytokines into the wound microenvironment dictates hypertrophic scar formation and whether TLR4 expression levels can provide mechanistic answers to our observations reported in this study. In summary, our study suggests that LPS can be conducive or detrimental to wound healing based on the stoichiometry and can ultimately lead to scar tissue formation. Future studies will focus on identifying the changes in cytokine secretion and gene expression pattern post-LPS treatment, which in turn will allow successful transition to any potential therapeutic benefit.