Separation of Rat Epidermis and Dermis with Thermolysin to Detect Site-Specific Inflammatory mRNA and Protein

Evaluation of inflammatory mediators and neurotrophic factors from the skin can be limited due to the heterogeneity of cell types found in the inflamed dermis and epidermis^1,2,3. Several enzymes, chemical, thermal, or mechanical techniques involving separation of the two layers or for performing cell dissociation for evaluation have been reviewed recently⁴. Acid, alkali, neutral salt, and heat can divide the epidermis from dermis quickly, but cellular and extracellular swelling often occurs^5,6. Trypsin, pancreatin, elastase, keratinase, collagenase, pronase, dispase, and thermolysin are enzymes that have been used for epidermal-dermal separation^4,7. Trypsin and other broad scale proteolytic enzymes are active at 37–40 °C but must be monitored carefully to prevent dissociation of epidermal layers. Dispase cleaves the epidermis at the lamina densa, but requires 24 h for separation in the cold^4,8 or shorter timepoints at 37 °C^4,9. A limiting feature of all these techniques is the potential disruption of tissue morphology and loss of integrity of mRNA and protein.

To maintain the integrity of mRNA and protein, a skin separation method should be carried out in the cold for a short period of time. In evaluating skin separation techniques for inflammation studies, thermolysin is an effective enzyme to separate the epidermis from dermis at cold temperatures⁴. Thermolysin is active at 4 °C, cleaves epidermal hemidesmosomes from the lamina lucida, and separates the epidermis from dermis within 1–3 h^4,8,10. The goal of this report is to optimize the use of thermolysin for separation of inflamed rat epidermis from dermis to detect mRNA and protein levels for inflammatory mediators and neurotrophic factors. Several preliminary reports have been presented^{11,12,13,14,15}. The objective of this manuscript is to describe an optimal skin separation technique using thermolysin and demonstrate the detection of 1) markers of inflammation, 2) interleukin-6 (IL-6) mRNA, and 3) nerve growth factor (NGF) mRNA and protein in the epidermis of rats with carrageenan-induced inflammation (C-II)^16,17. A preliminary report using the complete Freund’s adjuvant model indicates that NGF mRNA and protein levels increase early during inflammation¹⁵. In mice, skin sensitization with the topical application of oxazolone causes an early rise in the IL-6 mRNA using in situ hybridization³⁶. Both IL-6 and NGF have been implicated in C-II^18,19, but there have been no reports describing mRNA or protein levels for IL-6 or NGF specifically from the epidermis during the acute stages of C-II.

The thermolysin technique is inexpensive and straightforward to perform. Furthermore, thermolysin separation of the epidermis from dermis allows for mRNA, western blot, and immunohistochemical analysis of inflammatory mediators and neurotrophic factors during the process of inflammation¹⁵. Investigators should be able to easily use this technique in both preclinical and clinical studies of skin inflammation.

This protocol follows the animal care guidelines of Oklahoma State University Center for Health Sciences IACUC (#2016-03).

1. Carrageenan-induced inflammation (C-II)

1. Anesthetize male and/or female Sprague Dawley rats (200–250 g; 8–9 weeks old) with isoflurane (or injectable anesthetic).
2. Check the depth of anesthesia by touching the cornea and lightly pinching the left hind paw. When the animal is appropriately anesthetized, no corneal or paw response will be observed.
3. Subcutaneously inject the right glabrous, hind paw with 100 µL of 1% (w/v) λ-carrageenan diluted in phosphate-buffered saline (PBS) and allow the rat to fully recover from anesthesia²⁰.
   1. Make sure that appropriate controls are used, such as naïve rats without isoflurane in this report. Preliminary studies indicate that naïve rats with or without isoflurane have the same basal expression of epidermal IL-6 and NGF.
      NOTE: Naïve rats are preferred controls for inflammation studies since subcutaneous saline or PBS cause a local inflammation^23,37.
4. At the 6-12 h time points, euthanize the rats with CO₂ and measure hind paw metatarsal thickness with calipers. Cut 1 mm x 2 mm pieces of glabrous hind paw skin with a sharp scalpel. If there is hair on the skin, shave the area before cutting 1 mm x 2 mm pieces of skin.
   NOTE: Make sure that the appropriate timepoints are chosen according to the specific studies.
5. Using microdissection forceps, transfer the skin into 1 mL of cold Dulbecco's Modified Eagle Medium (DMEM) in a microcentrifuge tube on ice and keep cold for 15–60 min.

2. Thermolysin separation of epidermis and dermis

1. Prepare and activate thermolysin.
   1. Prepare a solution of thermolysin, by adding 5 mg of Geobacillus stearothermophilus to 10 mL of PBS, at pH = 8 (concentration 500 µg/mL).
   2. Prepare a 1 M solution of calcium chloride (CaCl₂ anhydrous) by adding 1.11 g into 10 mL of distilled H₂O.
   3. To prevent autolysis of thermolysin, add 10 µL of calcium chloride to 10 mL of thermolysin solution. The calcium chloride final concentration will be 1 mM.
   4. Aliquot 1 mL of activated thermolysin into 10 wells of a 24 well cell culture plate on ice.
2. Use thermolysin enzyme digestion to separate the epidermis from dermis.
   1. Using microdissection forceps, transfer one skin sample into each well of activated thermolysin. Make sure not to immerse the skin in the thermolysin solution.
   2. Gently tap the skin on the side of the well to assist in releasing the skin sample from the forceps to float on the thermolysin solution.
   3. Float the skin into the thermolysin solution with the stratum corneum (outer epidermis) side up and dermis facing down. It is critical that the dermis faces down, or the effective separation will not take place.
      NOTE: The amount of time for thermolysin incubation must be determined empirically by the end-user. Glabrous, hind paw skin from Sprague Dawley rats (200–250 g; 8–9 weeks old) often requires 2.0–2.5 h for separation. Incubation time is expected to vary with species and age.
   4. After the appropriate incubation time in thermolysin, use microdissection forceps to transfer one skin sample into a well of a 6 well cell culture plate with 7–8 mL of cold (4 °C) DMEM. This allows more room for separation of the epidermis from the dermis.
   5. Immerse the skin into the DMEM.
   6. Gently brush the epidermis with the forceps around the perimeter of the skin until the near-translucent epidermis is observed at the borders. If this cannot be achieved, return the skin sample to the thermolysin solution for another 15–30 min.
   7. Once the epidermis noticeably separates from the dermis, then carefully hold both the epidermis and dermis with microdissection forceps and very slowly pull the epidermis from the dermis.
   8. Evaluate the translucence of the isolated epidermis and make sure it is optically consistent. See Figure 2 for an example of a 1 mm x 2 mm sample of rat epidermis. If there is a variation in the translucence, then proper separation has not occurred.
3. Inactivate thermolysin using ethylenediaminetetraacetic acid (EDTA) in the separated pieces of epidermis and dermis.
   CAUTION: The thermolysin that remains in the epidermis and dermis is still active and can damage the layers if not inactivated.
   1. Prepare a 0.5 M EDTA stock solution. To do so, slowly add 0.93 g EDTA into 5 mL of double-distilled water. Add sodium hydroxide to the solution until it clears. Ensure that the pH of the solution is ~8.0.
   2. Make a 5 mM EDTA solution in DMEM. Add 0.25 mL of 0.5 M EDTA stock solution to 25 mL of DMEM.
   3. Place the separated epidermis and dermis into the 5 mM EDTA/DMEM solution at 4 °C for 30 min to deactivate thermolysin’s activity.
4. Evaluate the epidermis with tinctorial histology^8,9,10.
   1. Fix a portion of the epidermis in a 10% neutral formalin, 4% paraformaldehyde, or 0.25% paraformaldehyde with 0.8% picric acid solution for 1 h at room temperature (RT) with agitation.
   2. Place the fixed epidermis in 10% sucrose in PBS for 1 h at RT with agitation.
   3. Freeze the epidermis in a tissue embedding matrix for sectioning. Cut 14 µm cross-sections using a cryostat and thaw-mount sections onto gelatin-coated glass microscope slides.
   4. Dry sections on a slide warmer and stain with a working solution of toluidine blue (TB; 10% TB in 1% sodium chloride) for 90 s. Appose coverslips with an aqueous mounting medium.
   5. Observe the epidermis with brightfield microscopy at 50x–250x.
      NOTE: If proper separation has occurred, the epidermis will be divided cleanly from the dermis and the five layers will be detected: stratum basale, stratum spinosum, stratum granulosum, stratum lucidum, and stratum corneum. An example of separated rat skin epidermis can be seen in Figure 3.

3. Protein extraction and western blot analysis

1. Perform western blotting on the separated tissue samples using previously published protocol²¹.
2. Homogenize the epidermis in 50 μL of lysis buffer (25mM Tris HCl, pH = 7.4, 150 mM NaCl, 1 mM EDTA, 5% glycerol, and 1% Triton X-100) containing a phosphatase and protease inhibitor cocktail.
3. Centrifuge samples at a max speed for 15 min at 4 °C and evaluate the supernatant for protein concentration using a protein assay kit.
4. Load equal concentrations of protein (30 µg) onto SDS gels, perform electrophoresis, and then transfer proteins to nitrocellulose or PVDF membranes.
5. Block membranes with 5% milk for 2 h and incubate overnight in primary antibody (mouse anti-NGF, E12, 1:1000).
6. Wash 3x with PBS with 0.3% tween for 10 min each and incubate with a labeled secondary antibody (e.g., alkaline phosphatase labeled rabbit anti-mouse IgG).
7. Use a scanning system to evaluate western blot signal (e.g., ECF substrate and an imaging platform).

4. Immunohistochemistry

1. Place tissue samples in a fixative for optimal immunoreactivity: 0.96% (w/v) picric acid and 0.2% (w/v) formaldehyde in 0.1 M sodium phosphate buffer, pH = 7.3^21,22,23 for 4 h at RT. Transfer to 10% sucrose in PBS overnight at 4 °C.
2. Perform standard immunohistochemistry on the tissue sections^21,22,23.
3. Embed the epidermis from animals into a single frozen block in embedding matrix and cut 10–30 µm sections on a cryostat. Mount the sections on gelatin-coated, glass microscope slides and dry at 37 °C for 2 h.
4. Wash sections for three, 10 min rinses in PBS and incubate for 24–96 h in primary antisera, [e.g., mouse anti-NGF (E12, 1:2000)] and rabbit anti-protein gene product 9.9 (PGP 9.5, 1:2000) diluted in PBS containing 0.3% (w/v) Triton X-100 (PBS-T) PBS-T with 0.5% bovine serum albumin (BSA) and 0.5% polyvinylpyrrolidone (PVP).
5. After primary antiserum incubation, rinse sections three times for 10 min in PBS and incubate 1 h at RT in Alexa Fluor 488 donkey anti-rabbit IgG (1:1000) and Alexa Fluor 555 donkey anti-mouse IgG (1:1000) diluted in PBS-T.
6. Rinse sections three times in PBS for 10 min and affix coverslips with non-fading mounting medium to retard fading of immunofluorescence.

5. RNA isolation and cDNA synthesis

1. Perform standard reverse transcriptase polymerase chain reaction (RT-PCR) on the skin samples²¹. Isolate total RNA using a phenol, guanidine isothiocyanate solution.
2. Carry out complementary DNA synthesis by Moloney murine leukemia virus reverse transcriptase.
3. Use the following primer sequences for NGF and IL-6 amplification:
   NGF (Sense) - GTGGACCCCAAACTGTTTAAGAAACGG
   NGF (Antisense) – GTGAGTCCTGTTGAAGGAGATTGTACCATG
   IL-6 (Sense) - GCAATTCTGATTGTATGAACAGCGATGATGC;
   IL-6 (Antisense) – GTAGAAACGGAACTCCAGAAGACCAGAG
4. Compare the levels of NGF and IL-6 mRNA to β-actin housekeeping gene:
   β-ACTIN (Sense) - TGCGTGACATTAAAGAGAAGCTGTGCTATG
   β-ACTIN (Antisense) – GAACCGCTCATTGCCGATAGTGATGA
5. Evaluate with qualitative RT-PCR using thermal cycler and quantitative real-time PCR (qRT-PCR) using a qRT-PCR system.

Carrageenan injection into the rat hind paw caused classic symptoms of inflammation such as redness and edema^16,17. The swelling of the hind paw was measured with mechanical calipers²⁰. A baseline value of the thickness of the paw was obtained for each rat before carrageenan treatment and measured again at 6 h and 12 h. Paw thickness was increased significantly compared to the baseline values (Figure 1).

Thermolysin incubation of the rat glabrous hind paw skin produced a sheet of epidermis. Brightfield microscopy was used to evaluate the effectiveness of thermolysin separation of the epidermis and dermis (Figure 2). The layers of the epidermis could be determined while focusing through the sheet at higher magnification. Toluidine blue staining of epidermal cross sections showed that the epidermis was separated from the dermis at the basement membrane (Figure 3). The epidermal rete ridges (epidermal pegs) along with the indentations from dermal papillae were intact. All keratinocyte cell layers were observed.

Western blotting of thermolysin-separated epidermis produced consistent results indicating stable protein levels during the technique at 4 °C. Very little NGF protein was detected in naïve rat epidermis, but NGF protein levels were upregulated (250%) after 6 h of C-II as compared to naïve animals (Figure 4). After 12 h of C-II, NGF levels were reduced compared to 6 h but remained elevated (55%) relative to controls. Immunohistochemistry for NGF in the separated epidermis provided reliable immunostaining and confirmed the results from western blots (Figure 5). NGF-immunoreactivity (ir) was not detected in naïve control epidermis, but at 6 h C-II, there was NGF-ir in most of the keratinocytes of the stratum granulosum and stratum lucidum. A few cells for the stratum spinosum were NGF-immunoreactive (IR) at 6 h C-II. At 12 h C-II, NGF-ir occurred in keratinocytes of the stratum granulosum and stratum lucidum with some cells in stratum granulosum intensely NGF-IR. At no timepoint was NGF-ir detected in stratum basale or stratum corneum. PGP9.5-IR intraepidermal, varicose nerve fibers were present in the separated epidermis from naïve and C-II rats (Figure 5).

Qualitative RT-PCR demonstrated that there was good quality mRNA from thermolysin-separated epidermis (Figure 6A, Figure 7A). Using actin as a housekeeping gene for quantitative real-time PCR, NGF mRNA expression in epidermis during C-II was significantly elevated (>3-fold) at 6 h compared to naïve rats (Figure 6B). At 12 h, NGF mRNA returned to baseline levels (Figure 6B). Using actin as a housekeeping gene for quantitative real-time PCR, IL-6 mRNA in epidermis during C-II was significantly elevated (>6-fold) at 6 h compared to naïve rats (Figure 7B). At 12 h, IL-6 mRNA levels dropped significantly from the 6 h amounts but remained elevated (2-fold) compared to naïve rats (Figure 7B).

Figure 1: Carrageenan injection produces paw edema.
A baseline value was obtained for each rat prior to carrageenan treatment. After 6 h and 12 h of treatment, paw thickness increased significantly compared to baseline values. The results are expressed as the SEM with six rats per treatment (*p < 0.05, **p < 0.01, ***p < 0.001; student’s t-test, unpaired, two-tail, was performed at each timepoint). Please click here to view a larger version of this figure.

Figure 2: Thermolysin produces a sheet of epidermis.
Brightfield microscopy revealed a translucent epidermal sheet approximately 1 mm x 2 mm in size. The layers of the epidermis could be determined while focusing through the sheet at higher magnification. Scale bar = 500 µm. Please click here to view a larger version of this figure.

Figure 3: Toluidine blue staining of the epidermis.
Brightfield microscopy determined that thermolysin caused an effective separation of the epidermis from dermis. Epidermal rete ridges (epidermal pegs; arrowheads) were observed along with indentations from dermal papillae (arrows). All keratinocyte cell layers were intact. SB: stratum basale, SS: stratum spinosum, SG: stratum granulosum, SL: stratum lucidum, SC: stratum corneum. Scale bar = 100 µm. Please click here to view a larger version of this figure.

Figure 4: NGF protein expression in epidermis during carrageenan-induced inflammation.
NGF levels were increased (250%) after 6 h of inflammation compared to naïve animals. After 12 h, the levels were reduced compared to 6 h but were elevated (55%) in contrast to naïve animals. The results are expressed as the SEM with three rats per group (*p < 0.05, **p < 0.01, ***p < 0.001; student’s t-test, unpaired, two-tail was performed at each timepoint) Please click here to view a larger version of this figure.

Figure 5: NGF and PGP9.5 immunoreactivity (ir) during C-II.
Columns A,B,C show NGF-ir, PGP9.5-ir, and DAPI nuclear staining, whereas columns A₁-C₁ show only NGF-ir. In all images, stratum corneum is towards the left and stratum basale towards the right. NGF-ir was not detected in naïve control epidermis (A, A₁). At 6 h C-II (B,B₁), NGF-ir was present in most keratinocytes of the stratum granulosum (short arrows) and stratum lucidum (long arrows). A few cells for the stratum spinosum were NGF-ir at 6 h C-II (large arrowheads). At 12 h C-II (C,C₁), NGF-ir occurred in keratinocytes of the stratum granulosum and stratum lucidum (long arrows) with some cells in stratum granulosum intensely NGF-ir (short arrows). At no timepoint was NGF-ir detected in stratum basale or stratum corneum. PGP9.5-ir intraepidermal nerve fibers were present in the separated epidermis from naïve and C-II rats (small arrowheads, A-C). SB: stratum basale, SS: stratum spinosum, SG: stratum granulosum, SL: stratum lucidum. Scale bar = 50 µm. Please click here to view a larger version of this figure.

Figure 6: NGF mRNA during carrageenan-induced inflammation.
NGF mRNA expression in thermolysin-separated epidermis during C-II was evaluated by qualitative PCR (A) and quantitative real-time PCR (B). Qualitative mRNA blots (A) for NGF and actin demonstrated that there was good quality mRNA that could be evaluated during inflammation. NGF mRNA expression in epidermis during C-II was evaluated by quantitative real time PCR using actin as a housekeeping gene (B). NGF mRNA was significantly elevated (>3-fold) after 6 h of C-II compared to naïve untreated rats, but levels returned to baseline at 12 h (B). Results are expressed as the SEM with three rats per group (*p < 0.05, **p < 0.01, ***p < 0.001; student’s t-test, unpaired, two-tail was performed at each timepoint). Please click here to view a larger version of this figure.

Figure 7: IL-6 mRNA during carrageenan-induced inflammation.
IL-6 mRNA expression in thermolysin-separated epidermis during C-II was evaluated by qualitative PCR (A) and quantitative real-time PCR (B). Qualitative mRNA blots (A) for IL-6 and actin showed there was good quality mRNA for evaluation during inflammation. IL-6 mRNA expression in epidermis during C-II was evaluated by quantitative real-time PCR using actin as a housekeeping gene (B). IL-6 mRNA was significantly elevated (>6-fold) after 6 h of C-II compared to naïve untreated rats (B). At 12 h, IL-6 mRNA levels were reduced significantly from 6 h levels but remained elevated (2-fold) compared to naïve rats (B). Results are expressed as the mean S.E.M. with two rats per group (*p < 0.05, **p < 0.01, ***p < 0.001, student’s t-test, unpaired, two-tail was performed at each timepoint). Please click here to view a larger version of this figure.

The study determined that the epidermis of rat hind paw glabrous skin was easily separated from dermis using thermolysin (0.5 mG/mL) in PBS with 1 mM calcium chloride at 4 °C for 2.5 h. Histological evaluation indicated that the epidermis was separated from the dermis at the basement membrane and that the epidermal rete ridges were intact. Thermolysin is an extracellular metalloendopeptidase produced by Gram-positive (Geo)Bacillus thermoproteolyticus²⁴. Its activity is stable at 4 °C but is functional over a wide range of temperatures^10,24,25. This enzyme has been used extensively for protein chemistry^24,25, but several groups have shown its application for skin separation into epidermal and/or dermal sheets^4,8,10,26,27. Walzer et al. were the first to report epidermal-dermal separation of human skin using thermolysin at 4 °C (250–500 μg/mL for 1 h)¹⁰. With light and electron microscopy, separation was determined to occur at the epidermal basement membrane between laminin and the bullous pemphigoid antigen site¹⁰.

Furthermore, hemidesmosomes, the attachments of basal keratinocytes to the basement membrane, were disrupted selectively^10,29. Rakhorst et al. compared thermolysin (4 °C, 500 µg/mL, overnight) to dispase for epidermal-dermal separation of rabbit buccal mucosa⁸. Thermolysin was incomplete in separating the mucosal epidermis from dermis signifying that differences may occur for species, incubation time, solution composition (no CaCl₂ to prevent thermolysin autolysis²⁴), source of thermolysin, and/or site-specific differences indicated from other studies^10,26,27. End users of the current protocol should make sure to use fresh thermolysin and always include calcium chloride but also should be aware of these potential limitations.

Although some investigators have used thermolysin at 37 °C^26,29, users of this protocol should be mindful to keep skin tissue at 4 °C to preserve the stability of protein and mRNA. We used DMEM at 4 °C as a solution for skin prior to and after thermolysin separation because of its usefulness in maintaining cells in culture²⁸, and it has been used previously for skin separation with thermolysin^{8,26,27,30,31}. However, Walzer et al. used sterile PBS supplemented with 200 µG/mL streptomycin, 200 U/mL penicillin, and 2.5 µg/mL fungizone¹⁰, whereas others have used different media (e.g., keratinocyte culture media free of epidermal growth factor) followed by PBS rinsing²⁹.

In the protocol, thermolysin separation was performed in 500 µg/mL thermolysin and 5 mM calcium chloride in PBS (pH = 8), similar to the original method¹⁰. DMEM has been used as a solution for thermolysin separation at 4 °C (overnight)⁸, and HEPES buffer has been used effectively with 500 μG/mL thermolysin solution at 37 °C for 2 h²⁹. However, we did not explore how culture medium or other buffers affects thermolysin’s activity for epidermal-dermal separation. Calcium chloride is an important addition to decrease autolysis of thermolysin^24,32 and deletion of this step may lead to incomplete cleavage of the epidermis from dermis⁸.

The size of the skin sample appears to influence the time needed for thermolysin incubation and the effectiveness of enzymatic cleavage of the epidermis from dermis. Investigators need to evaluate the appropriate sample size and incubation time for their own tissues. Floating the samples on the thermolysin solution with the epidermis facing upward is important for optimal enzyme effectiveness¹⁰. As noted earlier, the site of action for thermolysin is at the keratinocyte hemidesmosomes and basement membrane^4,10,29; therefore, thermolysin works inward from the edges of the skin. From our experience, the skin edges separate earlier than the middle of the sample, and it is important to allow enough time for complete enzymatic cleavage. When cleavage is complete, the epidermis should pull away easily from the dermis. If still attached, tugging on the layers may cause portions of dermis to come away with the epidermis.

A limitation of the thermolysin technique is the time required. Increasing the thermolysin concentration beyond 500 µG/mL does not decrease the time for separation and there is poor preservation of the epidermis at higher concentrations¹⁰. Epidermal-dermal separation methods have been reviewed recently⁴, and many methods take 30–60 min at 20–40 °C. Heat (50–60 °C) separation of skin occurs quickly (30 s to 10 min)⁴, but proteins and mRNA are known to degrade quickly at such high temperatures. Alternatively, sodium thiocyanate (2 N) at RT may be an acceptable rapid separation technique (5 min)^4,6, but protein and mRNA integrity have not been studied with this method⁶. The cold thermolysin method was chosen for the preservation of protein and mRNA, but there were no direct comparisons made between protein and mRNA integrity using other techniques.

In the present study, cold thermolysin digestion is demonstrated to be an effective method to separate the epidermis from the dermis for evaluation of mRNA and protein alterations during inflammation. During carrageenan-induced inflammation, NGF mRNA and protein levels and IL-6 mRNA levels were elevated at 6 h, returning close to baseline by 12 h. With immunohistochemistry, NGF immunoreactivity was increased in keratinocytes at 6 h and 12 h. An advantage of the thermolysin method is the ability to perform site-selective analysis. For example, the increased production of NGF and IL-6 in the current study is from keratinocytes, since dermal cells are excluded from the assays. This method allows for insight into the location and mediator types for sensitization of primary afferent terminals³³. In addition, this method allows for better understanding of the time course of neurotrophin production along with uptake and transport in primary afferents during inflammation^34,35.