Comparison and Evaluation of 2,4-Dinitrouorobenzene Induced Allergic Contact Dermatitis Model in Mice

2,4-Dinitrouorobenzene (DNFB) induced allergic contact dermatitis model is widely used in the research of inammation, immune response and pruritus. However, there are several modeling cycles of DNFB model in many literatures, which brings some confusion for choice in the experiments. Here, we established two-weeks and ve-weeks models of DNFB and aimed to compare and evaluate them. Methods Two-week and ve-week DNFB models were established in male SPF C57BL/6 mice. Results The results showed that the scratching bouts of the ACD mice in two groups both signicantly increased. Following the modeling process, the skin of the neck and back of the modeling site of the ACD mice in two-weeks group gradually appeared erythema, edema, ulceration, scabbing, scaling and dryness. The skin surface features of ACD mice in ve-weeks group obviously showed erythema, edema, ulceration, scabbing, scaling, re-scabbing, re-scaling, harden and dryness.The whole modeling process are both accompanied by severe scratching reaction after the skin repeated exposure to DNFB. Moreover, the results of hematoxylin-eosin (HE) staining showed that the two modeling methods both had thickened epidermis (p < 0.001) and hyperkeratosis. The toluidine blue (TB) staining revealed obvious inammatory cell inltration (p < 0.0001) in two groups. Our results and inammation.


Introduction
Allergic contact dermatitis (ACD) is known as a delayed hypersensitivity in ammatory reaction [1,2], that occurs when the skin or mucous membrane is exposed to certain external irritants or allergens [3,4]. This reaction is mediated by a hapten-speci c T cell-dependent reaction. After the sensitized body is attacked by antigen, it is mainly characterized by the in ltration of in ammatory cells and the release of in ammatory factors. The clinical symptoms are mainly allergic contact reactions such as erythema, erosion, scab, edema and itching in the sensitized skin [5][6][7]. The T cells in the delayed hypersensitivity reaction play a key role in transplant rejection, graft-versus-host disease, autoimmunity and tumor immunity [8]. Therefore, studying the effect of drugs on the induction and regulation of ACD is very important for nding new drugs and elucidating the mechanism of action.
A good experimental animal disease model is helpful for people to understand the occurrence and development laws of human diseases more conveniently and effectively, and to study preventive measures. The allergic contact dermatitis model induced by 2,4-Dinitro uorobenzene (DNFB) is one of the most commonly used animal models in the research of in ammation or chronic pruritus [9][10][11][12][13][14]. A good animal model must have the characteristics of accurate replication and similarity with clinical pathogenesis, which plays a key role in the research of disease pathogenesis and e cacy. DNFB is a hapten that activates skin keratinocytes and produces cytokines after the rst contact with the body [15].
Cytokines combine with antigen-presenting cells to transfer to local lymph nodes, activate antigenspeci c T cells, and make them produce memory T cells [16]. This process is called the sensitization stage [3]. After 4-7 days of sensitization, apply the hapten DNFB to the skin of the ear or the neck and back to supplement and activate memory T cells. Cytokines are produced by helper T cells 1 (Th1 cells) and helper T cells 2 (Th2 cells) to cause local delayed allergic reactions [4,14,17,18]. This reaction generally occurs about 18 hours after exposure to the antigen and reaches a peak in 24-48 hours. There are literatures using the ACD two-week model to study the role of Toll-like receptors [19] and monoclonal antibody NaV1.7 channels in chronic pruritus [10,[20][21][22], while studying the development of chronic pruritus in sensory neurons requires the use of ACD in the BRAF signaling pathway for ve weeks Model [13]. In order to clarify a more suitable animal model during the actual operation, this study compared and evaluated the two modeling methods of DNFB-induced contact allergic reaction in mice.

Animals
Male SPF C57BL/6 mice, weighing approximately 20-25 g, are provided by Changzhou Cavens Laboratory Animal Co., Ltd., free to eat and water during the feeding and experiment period. The ambient temperature is 20-25°C, and the relative humidity is 40 %-60%. All experimental protocols described in this study were followed the protocol outlined in the Guide for the Care and Use of Laboratory Animals of Anhui Medical University Animal Care Committee, which comply with the National Institutes of Health Guidelines for the Care and Use of laboratory Animals in ARRIVE guidelines. Two weeks ACD model established by DNFB ACD model was induced by repeated exposure mice skin to DNFB-acetone solution [4,[22][23][24]. DNFB was dissolved in a mixture of acetone:olive oil (4:1) [15]. C57BL/6 male mice were randomly divided into normal saline control, solvent control (acetone: olive oil 4:1) and DNFB model, with 5 mice in each group. After a week of adaptive feeding, the rostral area of the neck and the abdomen of all mice were shaved and the shaving area were about 2cm×2cm. Two days later after shaving, 0.5% DNFB-acetone solution 50µL was dripped on the shaving area of the abdominal skin, and evenly applied for initial sensitization [25]. Repeat the operation at the next day for intensive sensitization. After 5 days, 50µL of 0.15% DNFB was dripped into the shaved part of the neck and back of mice in DNFB model, and evenly smeared. The saline control mice and the solvent control mice were dripped with 50µL saline and 50µL acetone olive oil (4: 1) respectively. And the challenge with 50µL 0.15% DNFB-acetone solution was painting on rst, third, fth and seventh day [9,10]. The speci c timeline and sequence could be seen in Figure 1A. The video of itching behavior of mice was recorded for 60 minutes after each application 24 hours.

Five weeks ACD model established by DNFB
The intervention measures are similar as the two-week ACD model. C57BL/6 male mice were randomly divided into saline control, solvent control (acetone: olive oil 4:1) [17,26] and DNFB model, with 5 mice in each group. Mice were shaved before the experiment. 0.15% DNFB 100µL were dripped on the shaving area of mouse abdomen, and evenly smeared for sensitization. After sensitization for one week, 0.15% DNFB 50µL was applied to the shaving area of the neck and back of model mice twice a week for four weeks, nine times in total [14,27]. The control mice were coated with saline and acetone olive oil (4:1) respectively. The speci c treatment and intervention measures were shown in Figure 1B. After each treatment for 24 hours, itch behavior video was recorded continuously for 60 minutes.

Scratching behavior test
Behavioral tests were videotaped using SONY digital video camcorders [13,28,29]. On the test day, Mice were placed in a plastic arena (10×10×13 cm) and given at least 10 minutes to get accustomed to recording conditions prior to recordings. Then mice were recorded for 60 minutes for spontaneous scratches. Videos were played back on a computer for assessments by observers [28][29][30][31]. Hind limb scratching behavior towards the painted area was observed for 60 minutes. A scratch is de ned as a bout of scratching that occurs after the mouse lifts its hind paw to the moment the hind paw is returned to the ground or mouth [32,33].

Skin morphology studies and histological observation
The skin which was challenged were xed by 4% paraformaldehyde, and then the para n sections were made to be stained with hematoxylin-eosin (HE) solutions and toluidine blue (TB) staining [13]. And the staining images were visualized under a microscope, and images were taken. Finally, Image J software was used to measure the relative thickness of the epidermis from HE staining images and count the number of mast cells from TB staining images [13,31].

Statistical analysis
Results are expressed as mean ± SEM of all independent experiments. Statistical analysis was conducted using GraphPad Prism 8.0 statistical software. For multiple-group comparisons, a one-way variance analysis (ANOVA) with Bonferroni's multiple comparison test were used. When p < 0.05, the difference is signi cant.

Results
Scratching behavioral changes in two-week and ve-week ACD mice model induced by DNFB The experimental results were shown in Figure 2. The number of scratches in the DNFB two-week ACD model increased signi cantly and showed an upward trend. The number of scratching in the saline control group and the vehicle control group did not increase signi cantly (Figure 2A). In the ve-week ACD model group of DNFB, the scratches increased obviously. And the scratching times of hind limbs of ACD model mice in saline control group and solvent control group had no obvious increasing trend ( Figure   2B). At the end of the two-week model building cycle, the number of scratches in the model group was signi cantly increased compared with the two-week model's normal saline control group and solvent control group, and the difference was statistically signi cant ( Figure 2C, **** p < 0.0001). At the end of the ve-week model construction period, the number of scratches in the model group was signi cantly higher than that in the saline control group and the solvent control group and was statistically signi cant ( Figure 2D, **** p < 0.0001).
Skin gross morphology changes in two-week and ve-week ACD mice model induced by DNFB The experimental results are shown in Figure 3. In the two modeling methods, the skin surface of the neck and back of the mice in the saline control group and the vehicle control group has no obvious change, and the mice had better mobility. With the progress of the modeling process, the skin of the two-week model mice was ulcerated on the fth day after applying DNFB on the neck and back. On the seventh day after applying DNFB on the neck and back, the skin in the ulceration area became crusted and hardened.
At the end of modeling of two weeks, the skin was scabbed ( Figure 3C). On the eighth day after applying DNFB on the neck and back of ve-week model mice, the skin at the neck and back modeling site ulcerated. On the eleventh day, the skin had scab before administration. In the fteenth day of administration, the scab of the skin had faded, and the skin under the scab was exposed and was dry. Before administration on the eighteenth day, the skin ulceration at the modeling site was aggravated.
Before administration on the twenty-second day, the skin began to scab, and the skin around the administration area began to turn up. Before the administration on the twenty-fth day, the turned skin around the administration area of the mice began to fade and the skin ulcerated. At the end of the ve- week model, the scab on the neck and back fell off and exposed the skin ( Figure 3F).
Skin histopathological changes in two-week and ve-week ACD mice model induced by DNFB As shown in Figure 4, HE staining results showed that the skin epidermis thickness of mice in the twoweek model saline control group and solvent control group had no obvious thickening and the edge was clear ( Figure 4A, 4B). Compared with saline control group and solvent control group, the model group mice obviously thickened the skin epidermis at the modeling site ( Figure 4C), and the difference was signi cant (****p < 0.0001, Figure 4G). The skin at the modeling site was hyperkeratosis, spongy degeneration and dermal capillary dilatation. There was no obvious thickening of skin epidermis thickness at the modeling site of saline control group and solvent control group in the ve-week model ( Figure 4D, 4E). Compared with normal saline control group and solvent control group, the skin epidermis in the model group is obviously thickened (Figure 4F), and the difference is statistically signi cant (**** p < 0.0001, Figure 4G), dermal capillaries dilate, skin hyperkeratosis and spongy degeneration.

In ammatory mast cell changes in two-week and ve-week ACD mice model induced by DNFB
The experimental results are shown in Figure 5. Compared with the mice in the control group and the vehicle control group, the skin of the two-week ACD model group has obvious in ltration of mast cells ( Figure 5C), and the difference is statistically signi cant ( Figure 5G, **** p < 0.0001). Compared with the mice in the saline control group and the vehicle control group, the skin of the ve-week ACD model group showed obvious in ltration of mast cells ( Figure 5F), and the difference of mast cell count was signi cant ( Figure 5G, **** p < 0.0001).

Discussion
ACD is a chronic skin disease with recurrence, pruritus and in ammation [34]. It is seriously affecting human health and quality of life [35][36][37]. ACD is caused by delayed allergy, and patients are often accompanied by itching, so itching is the most prominent clinical symptom of the disease [38]. Studies have shown that local sensitization by exogenous allergens on the basis of the destruction of the skin barrier is an important basis for inducing ACD [7], and the immune in ammatory response is mainly based on the secretion of IL-4 by Th2 cells [39,40].
ACD is a common in ammatory skin disease caused by the skin or skin mucous membrane contacting with external allergens. It usually needs to be connected by proteins in the epidermis to form a new antigenic determinant, causing immune system activation [41,42]. Primary irritant contact dermatitis is a strong irritation of the contact material to the skin, and dermatitis occurs after contact. The form, scope and severity of dermatitis are different due to the different contact modes and individual reactions. In severe cases, the erythema is swollen and there are many papules and blisters. When the in ammation is severe, the blisters may rupture, erosion, exudate crusting and peel off. The ACD model established by DNFB is one of the most commonly used models [4,10]. DNFB is a hapten of small molecular substance, which is easy to combine with skin tissue protein without signi cantly changing the internal structure of tissue protein. The DNFB-induced ACD model, as a model of immunity and itch, plays an important role in the research of T cell in ammation-related and itch [3].
In this study, the two-week model [9,10] and the ve-week model of ACD [13,14] in mice were established by using the methods of DNFB abdominal sensitization and neck back excitation. The modeling methods of ACD are compared in order to provide su cient experimental and theoretical basis for the selection of appropriate animal models for research in related elds. According to the experimental data, the number of scratches in the ACD model mice induced by the two methods was signi cantly increased (p < 0.0001), which indicated that there were signi cant differences in itch behavior: the itch behavior results of ACD two-week model and ACD ve-week model mice increased signi cantly and showed an upward trend, while the itch behavior results of normal saline control group and solvent control group had no obvious upward trend. In addition, the HE staining results of the skin where the mice were modeled by the two modeling methods showed that the epidermis was thickened (p < 0.0001), hyperkeratosis, and telangiectasia. The results of toluidine blue staining showed the in ltration of mast cells in the model group obviously (p < 0.0001). Therefore, the results of this study show that the two methods of DNFBinduced ACD can successfully produce in ammation in the skin of the mouse model, and there are signi cant differences in itch behavior. However, in this experiment, the two modeling methods caused different changes in the appearance of the skin of the mouse model. The two-week model modeling cycle was relatively short. After DNFB was administered on the back and neck, the skin at the administration site of mice festered and hardened. The shape, scope and degree of dermatitis in the neck and back of the ve-week model mice were more serious than those of the two-week model. After the administration of DNFB, the skin was ulcerated, crusted, scabed, re-crusted, and then dropped. Therefore, we believe that in the experimental research, we can choose the appropriate modeling time in combination with our research eld and research direction. For the study of pharmacokinetics, the ve-week model may be more suitable, and there is enough time to observe the effects of drugs.  Figure 1 Protocol for two-week and ve-week ACD model induction in mice. Two-week ACD model building. DNFB was dissolved in a mixture of acetone: olive oil (4:1). Mice were sensitized with 50 µL 0.5% DNFB solution by topical application of shaved abdomen skin. Repeat the operation on the next day. 5 days later, mice were challenged with 50µL 0.15% DNFB solution by painting the nape of neck, then on day 1, 3, 5, and 7.
Mice were painted with 100 µL 0.15% DNFB acetone solution onto the clipped abdominal skin for sensitization. Then 50 µL 0.15% DNFB challenge was repeated twice a week for 4 weeks, 9 times in total, from 7 days after the sensitization. The video of itching behavior was recorded for 60 minutes after 24 hours of each challenged on the nape of neck (n = 5).   Comparison of the skin histopathological changes and epidermal thickness variation in two ACD models. in the ve-week ACD model and has analytical signi cance, **** p < 0.0001. All data are expressed as the mean ± SEM (n=5). **** p < 0.0001 compared with control groups.