A Skin Chamber To Investigate Wound Infection and Healing in the Mouse

The development of a simple, convenient, and reliable polypropylene screw-capped skin chamber, which can be inserted into mice, is described. All implanted chambers of normal immuno-competent mice (n = 10), or immuno-suppressed mice (n = 10) remained in-situ for 15 days. Wound infection was established by a clinical isolate of Pseudomonas aeruginosa in immuno-competent mice (n = 10) 1 day after chamber implantation and chambers remained in-situ for 10 days. Similar infections of wounds among mice immuno-suppressed with cyclophosphamide resulted in the mouse becoming moribund due to systemic invasion by the bacterium. The authors conclude that this mouse skin chamber will be of potential value for studying wound healing during the inflammatory and early proliferative phases, and the influence of infection and treatments on these processes in immuno-suppressed and immuno-competent mice.

The skin is the body’s primary protective barrier and serves a significant role in body fluid homeostasis, thermoregulation, and host defense. Once the integrity of the skin has been breached by traumatic injury, the healing process commences and involves both removal and repair of the damaged tissue. Augmented innate and specific immune reactivity against potential infections occurs.¹ Healing is a progressive process of inflammation, proliferation, and re-modeling that involves an array of cells and cell signals. This process can be adversely affected by the extent of the injury, the immune status of the patient, and the presence of bacterial or other infection.² Much remains to be learned regarding the early phases of healing and its influence by infectious agents. Several animal models of severe skin trauma were previously developed for this purpose that used sterile chambers either attached to, or inserted through the epidermis.^3,4 These systems maintain sterility at the trauma site or permit the establishment of monoculture infections and allow for sampling of exudate and tissue, as well as application of therapy. Effective epidermal chamber applications have been previously described in pigs³ and rats,⁴ but to our knowledge, has not yet been attempted in the mouse. This is most likely due to the difficulties resulting from its small size and thinness, and therefore, more fragile nature of the mouse epidermis. While pig skin more closely resembles human skin, rodents have been used extensively to elucidate the genes involved in wound healing and as models of infection, wounding, and immunity.^5–10 The availability of genetically defined strains plus the abundance of biomarkers and reagents for mouse experimentation adds to the value of mice as an in-vivo research tool to study wound healing and infection.

A device was developed that is suitable for use in mice with normal immune function or in mice that have been treated with an immunosuppressant, to mimic a reduction in immune function much like what is observed following a severe burn. The following describes a simple and reliable full-thickness murine skin wound chamber and will discuss potential uses in studies of healing and infection.

Chamber
The chamber has an overall height of 10 mm, an outer diameter (OD) of 18 mm at its base, weighs 1.8 g with a capacity of 400 µL ex-vivo, and can be autoclaved at 121˚C for 15 minutes and re-used. The body of the chamber is made from a 1.8-mL capacity polypropylene capped tube (Scientific Specialties Inc, Hanover, Md) and has 4 parts (Figure 1). A) Cap with a polypropylene threaded lid, 13 mm outer diameter (OD) and 6-mm high with silicon o-ring. B) Stainless steel circlip with 12 mm internal diameter (ID). C) O-ring with a 0.5-mm thick stainless steel flat circular ring with an ID of 13 mm and an OD of 18 mm. D) Body consists of an open-ended polypropylene cylinder 8 mm long with an ID of 8 mm and an OD of 13 mm. It has a thread at one end to accommodate the lid and a moulded flange 18 mm in diameter at the opposite end in 4 equal sized sections. Figure 2 shows a cross sectional diagram of the device in situ.

Mice
All animal experiments were conducted with permission from the Animal Ethics Committee of Murdoch University, Western Australia. Six- to 8-week-old female Swiss mice weighing approximately 22 g to 30 g were obtained from the Animal Resource Centre, Murdoch, Western Australia. Mice were housed 5 per cage in the small animal holding facility at Murdoch University, Western Australia. Animals were subjected to a 12-hour light cycle with food and water ad libitum.

Bacteria
The Microbiology Department of Royal Perth Hospital, Perth Western Australia, donated a Pseudomonas aeruginosa clinical isolate of human origin. All incubations were at 37˚C (98.6˚F). The bacterium was cultured overnight on nutrient agar ([NA], Difco, England). A single typical colony was sub-cultured into 10 mL of Triptone soy broth ([TSB], Oxoid Ltd, Hampshire, UK) and incubated with shaking overnight. This culture (1 mL) was used to inoculate 9 mL of TSB, which was incubated for 5.5 hours (with shaking) to consistently produce 109 colony-forming units (cfu) per milliliter. Preparations for inoculation of the skin chamber were diluted in phosphate buffered saline (PBS) to deliver approximately 1000 cfu per mouse. All cfu counts were determined by standard serial dilutions on NA plates.

Chamber insertion
Mice were allowed to acclimatize in the holding facility for at least 24 hours prior to surgery. Mice were injected with 60-µg Nembutal™/gram intra-perotoneally (IP). Once anesthetized, hair (an area of approximately 2 cm²) on the animals’ left thorax immediately behind the shoulder was removed using electric clippers and was swabbed with iodine/povidone solution. A horizontal incision approximately 10 mm in length was made through the epidermis down to the underlying connective tissue. The layers of skin were separated from the underlying muscle fascia using a scalpel before the flange of the chamber was inserted beneath the skin and seated on top of the muscle fascia. The skin layer was repositioned over the chamber flange and the chamber was secured in place using 4 sutures (4/0 using a 19-mm needle; Dafilon®, B Braun, Tuttlingen, Germany) through the skin layer and around each flange of the chamber. Next, the o-ring was placed on top of the skin effectively clamping the skin between the o-ring and the flange. The circlip was then fitted using circlip pliers to retain the o-ring and protect the sutures. The positioning of the chamber just behind the shoulder (Figure 3) largely reduced the possibility of interference by the mouse. A 100-µL aliquot of PBS was added to the chamber at the completion of the implantation surgery. Postoperative analgesia was provided by a single dose of 0.3 µg of buprenorphine (Buprenex®, Reckitt Benckiser, Slough, UK) injected subcutaneously, and 500 mg/L of soluble paracetamol (Panadol®, GlaxoSmithKline, Brentford, UK), which was added to the drinking water thereafter for 7 days post surgery.

Cyclophosphamide treatment
The method of Carmeño-Vivas et al⁷, which uses cyclophosphamide as the immunosuppressant for experiments involving immunosuppression of mice was followed. Each mouse was given 200 mg/kg of cyclophosphamide IP.

Routine chamber sampling
Mice were anesthetized by controlled delivery of Rhodia Halothane (Rhodia, Merial, Australia) to a perspex chamber through a Fluotec Mark 2 Vaporiser (Cyprane Ltd, Yorkshire, UK). Once under deep sedation, a pair of modified surgical forceps was used to grasp the body of the chamber below the cap to prevent a rotational force being applied to the chamber, which might dislodge the skin/chamber interface seal. With the cap removed, a sterile micropipette was used to transfer exudate from the chamber. Generally, it was found that a volume of 50 µL–100 µL could be withdrawn daily for the first 2 to 3 days after implantation. Subsequent sampling usually required the addition of sterile saline to augment the exudate volume.

Inoculation of the chamber with bacteria
The chamber was inoculated with bacteria on the day following chamber implantation. Bacterial suspensions were diluted in sterile PBS and 100-µL volumes were aliquotted into the chamber using a micropipette. Negative controls were given PBS.

Chamber persistence in situ
Mice recovered well from the implantation surgery and within 1 to 2 hours showed no adverse signs or reactions and resumed feeding as normal. Shortly after recovery, the mice frequently attempted to reach the chamber with their hind feet, but within 24 hours attempts to dislodge it generally subsided, and thereafter it was well tolerated even after inoculating the chamber with bacteria.

The persistence of chambers in situ was assessed using groups of mice, which received different treatments: normal, immunocompetent mice (immunpos), cyclophosphamide treated mice (immunneg), normal mice infected (immunpos/bactpos), and cyclophosphamide mice infected (immunneg/bactpos). Chambers in immunpos and immunneg mice consistently remained in place for 15 days. Sporadic loss of chambers in both groups generally occurred after 15 days with 60% or 70% of chambers still in place at day 22 (Figure 4).

Mouse infection studies
With immunpos/ bactpos mice, chambers remained in place for 9 days; the mice showed no clinical signs of illness. Despite obvious inflammation and pus formation in the chamber, no viable bacteria were cultured from biopsies taken from the spleen and liver of these mice. In contrast, immunneg/bactpos mice became moribund and were euthanized or died 3 days after bacteria were inoculated into the chamber (Figure 5) and biopsies from the liver and spleen recovered viable P aeruginosa (data not shown).

Mice are generally the animals of choice for research purposes because of the large number of genetically defined strains, and there are many biomarkers and immune reagents commercially available. In the past, mice have been used to study wound infection and other purposes by direct subcutaneous bacterial injection in conjunction with traumatizing the skin by burning or scalding. Apart from the ethical concern of this procedure, difficulties arise with maintaining sterility and sampling from traumatized sites.

This device is easily implantable into an anesthetized mouse and the entire procedure takes approximately 10 to 15 minutes. Hemorrhage was rarely observed and the intact skin around the chamber, which is clamped between the o-ring and the flange, effectively sealed the chamber in place. This allows for solutions to be deposited into the chamber immediately after implantation, without leakage around the body of the chamber. This sealing also prevents unintentional microbial contamination.

All of the inserted chambers stayed in situ for 15 days in normal, noninfected, immuno-competent mice in the absence of bacterial inoculation. In contrast, mice that had their chambers inoculated with P aeruginosa remained in place for a shorter period (up to 9 days), presumably reflecting the impact of infection and inflammation on the normal healing process.

Whilst the chamber is a foreign body not normally present in a wound progressing to sepsis, its impact in relation to an infection is assessable by including appropriate sterile controls. The reason for the failure of the chambers to remain in situ after the 9- to 15-day period has not been systematically investigated at this stage, but it is likely that pressure necrosis of the epidermis overlying the chamber flange may have contributed. Notwithstanding, the chamber is a suitable device to investigate processes relating to the early- and mid-phases of normal wound healing—the inflammatory and proliferative stages. During this time, the device can be used as a convenient tool to investigate and influence the healing process within the chamber through the introduction of infectious and therapeutic agents.

From the School of Veterinary & Biomedical Science, Division of Health Science, Murdoch University, Murdoch, Western Australia; AB Health Pty Ltd, Perth, Western Australia Disclosure: AusIndustry and AB Health Pty Ltd provided funding for this study.

The authors thank AusIndustry and AB Health Pty Ltd for the funding to undertake this project. We also thank Professor Fiona Wood for helpful preliminary discussions, and Mr. Ernest Etherington from the workshop of the Division of Science and Engineering at Murdoch University (Murdoch, Western Australia) for his technical assistance with the design and manufacturing of the chamber. 

Address correspondence to:
Peter M. Geerlings, Bsc,(Hons)
Murdoch University
South Street Murdoch 6150 Western Australia
E-mail: p.geerlings@murdoch.edu.au
Phone: (08) 9360 2684