A Novel High-Throughput Ex Vivo Ovine Skin Wound Model for Testing Emerging Antibiotics

Summary

The protocol describes a step-by-step method to set up an ex vivo ovine wounded skin model infected with Staphylococcus aureus. This high-throughput model better simulates infections in vivo compared with conventional microbiology techniques and presents researchers with a physiologically relevant platform to test the efficacy of emerging antimicrobials.

Abstract

The development of antimicrobials is an expensive process with increasingly low success rates, which makes further investment in antimicrobial discovery research less attractive. Antimicrobial drug discovery and subsequent commercialization can be made more lucrative if a fail-fast-and-fail-cheap approach can be implemented within the lead optimization stages where researchers have greater control over drug design and formulation. In this article, the setup of an ex vivo ovine wounded skin model infected with Staphylococcus aureus is described, which is simple, cost-effective, high throughput, and reproducible. The bacterial physiology in the model mimics that during infection as bacterial proliferation is dependent on the pathogen's ability to damage the tissue. The establishment of wound infection is verified by an increase in viable bacterial counts compared to the inoculum. This model can be used as a platform to test the efficacy of emerging antimicrobials in the lead optimization stage. It can be contended that the availability of this model will provide researchers developing antimicrobials with a fail-fast-and-fail-cheap model, which will help increase success rates in subsequent animal trials. The model will also facilitate the reduction and refinement of animal use for research and ultimately enable faster and more cost-effective translation of novel antimicrobials for skin and soft tissue infections to the clinic.

Introduction

Skin infections are an important global issue, with large economic costs to healthcare providers around the world. The development of multidrug resistance and biofilm formation by pathogens plays a key role in the prevalence of non-healing wounds^1,2,3,4. As a result of this, skin and soft tissue infections are one of the more common reasons for extended hospitalization and subsequent readmission⁵. Delays in wound healing are costly for both the patient and healthcare providers, with some estimates suggesting around 6.5 million patients are affected annually in the US. In the UK, skin infections and associated complications result in approximately 75,000 deaths annually^2,4,6.

Staphylococcus aureus (S. aureus) is a formidable wound pathogen frequently isolated from patient wounds^2,7. The rate of emergence of multidrug resistance increased drastically in the 2000s. During this time, around 60% of acute bacterial skin and skin structure infections were culture positive for methicillin-resistant S. aureus¹. The increasing number of multidrug-resistant strains among Staphylococci, and indeed other pathogens, within the last 2 decades indicates an urgent need for the rapid development of antibiotics with new modes of action that can overcome resistance.

However, since the early 2000s, antibiotic discovery programs have been dominated by longer developmental times and low success rates, with only 17% of novel antibiotics entering clinical trials in the US achieving market approval⁸. This suggests a disparity between results from in vitro testing of emerging antibiotics and their clinical outcomes. It can be contended that this disparity is largely due to differences in bacterial physiology during infections in vivo and during conventional microbiological methods when testing the efficacy of antibiotics in the in vitro preclinical stages. Therefore, novel laboratory methods that are more representative of bacterial physiology during infection are needed to improve the success rates in antibiotic discovery programs.

Current methods for studying skin infections include studies in live animals (e.g., mice), ex vivo skin models (e.g., porcine), and 3D tissue-engineered skin models (e.g., human)^9,10,11,12. Studies in live animals are strictly regulated and have relatively low throughput. In animal models, wounding and infection cause significant distress to the animals and raise ethical concerns. Human skin models, ex vivo or tissue-engineered, require ethical approval, compliance with local and global legislation (the Human Tissue Act, the Declaration of Helsinki), and there is difficulty in acquiring tissues, with some requests taking years to fulfil^13,14. Both model types are labor intensive and require significant expertise to ensure experimental success. Some current ex vivo skin infection models require pre-inoculated discs and additives for the wound bed to enable infection; although these models are incredibly useful, there are limitations in the infection process as additives limit the utilization of the wound bed as a nutrient source^10,15,16,17. The model described in this study uses no additives to the wound bed, which ensures that the pathology of infection and viable cell counts are a result of direct utilization of the wound bed as the only nutrient source.

Given the need for new laboratory methods, a novel high-throughput ex vivo ovine model of skin infections for use in evaluating the efficacy of emerging antibiotics has been developed. Skin infection studies face many challenges-high costs, ethical concerns, and models that do not show a full picture^20,21. Ex vivo models and 3D explant models allow for better visualization of the disease process and the impact treatments can have from a more clinically relevant model. Here, the setup of a novel ovine skin model is described, which is simple, reproducible, and clinically relevant and has high throughput. Ovine skin was chosen as sheep are one of the large mammals commonly used to model responses to infections in vivo. Moreover, they are readily available from abattoirs, ensuring a steady supply of skin for research, and their carcasses are not scalded, ensuring good tissue quality. This study used S. aureus as the exemplar pathogen; however, the model works well with other microorganisms.

Protocol

Lambs' heads from the R.B Elliott and Son Abattoir were used as the source of skin samples in this project. All lambs were slaughtered for consumption as food. Instead of discarding the heads, these were repurposed for research. Ethics approval was not required as the tissue was sourced from waste discarded from abattoirs.

1. Sterilization

1. Disinfect forceps prior to collection of the heads by taking clean forceps and performing dry heat sterilization in an oven at 200 °C for 1 h. Autoclave all glassware at 121 °C for 15 min before use.
2. Carry out all described work within a microbiology class 2 cabinet. Prepare all reagents as per the manufacturer's instructions.

2. Sample collection

1. In the abattoir, Swaledale lambs were slaughtered by stunning using electricity or a captive-bolt pistol and exsanguinated. Collect lamb heads no more than 4 h post-slaughter.

3. Preparation of the heads

1. Disinfect the forehead section of the lamb by pouring approximately 100 mL of 200 ppm chlorine dioxide solution onto the sample area. Shave the forehead section of the head using electric clippers and wash the area with 200 mL of 200 ppm chlorine dioxide solution.
2. Wipe the area with ethanol and blue roll and cover the sample area with hair removal cream for 35 min. Gently scrape off the hair removal cream using a scraping tool and assess the sample area. If a significant amount of hair remains, repeat the hair removal process.
3. Use a further 200 mL of chlorine dioxide solution to rinse the area, and then rinse with ethanol and wipe with a blue roll.
4. Using a sterile 8 mm biopsy punch, cut out 8 mm skin samples from the prepared area. Remove the samples using sterile forceps and a 15-blade scalpel, ensuring that all cutaneous fat is removed.
5. Place the samples in a sterile 0.5 L jar filled with sterile phosphate-buffered saline (PBS), and then transfer them to sterile 50 mL tubes with 50 mL of 200 ppm chlorine dioxide solution, invert twice, and leave to sterilize for 30 min.
6. Remove the samples from the chlorine dioxide solution and wash them by placing them in a 50 mL tube filled with 40 mL of sterile PBS. Once washed, place each individual skin sample in a separate well of a 24-well plate.
7. Add 350 µL of pre-warmed medium while maintaining the sample at the air-liquid interface. The composition of the media is as follows: MK media (Medium 199 with Hanks' salts, L-glutamate, and 1.75 mg/mL sodium bicarbonate) and Ham's F12 in a 1:1 ratio, with added FBS (10% v/v), EGF (10 ng/mL), insulin (5 µg/mL), penicillin-streptomycin (100 U/mL), and amphotericin B (2.5 µg/mL).
8. Seal the 24-well plates with a gas permeable plate seal and incubate at 37 °C in a humidified 5% CO₂ tissue incubator for up to 24 h.

4. Maintenance of skin samples

1. After incubation, remove the culture medium and rinse the samples in 500 µL of sterile PBS. Add antibiotic-free media to each sample and incubate at 37 °C in a humidified 5% CO₂ tissue incubator for 24 h to remove residual antibiotics in the sample.
2. If turbidity or fungal infection develop in the antibiotic-free media after 24 h, then discard the sample.

5. Preparation of the inoculum

1. Prepare a 50 mL tube with 10 mL of sterile tryptic soy broth. Take a fresh agar plate of S. aureus and use a swab to transfer several colonies into the broth. Incubate for 18 h at 37 °C at 150 rpm.
2. Centrifuge at 4,000 x g for 3 min. Remove the supernatant and resuspend the cell pellet in 10 mL of sterile PBS. Repeat twice to ensure adequate washing of the cells.
3. Adjust the inoculum to 0.6 OD₆₀₀ in sterile PBS. Confirm the inoculum load by undertaking a manual viable plate count.

6. Infection of skin samples

1. Prepare a fresh 24-well plate with 400 µL of pre-warmed antibiotic-free media and add in the 24-well inserts using sterile forceps.
2. Remove the media from the skin samples, wash with 500 µL of sterile PBS, and remove the wash. Use sterile forceps to gently hold the sample to the bottom of the well.
3. Use a 4 mm punch biopsy to make a central wound flap, piercing through to a rough depth of 1-2 mm. Then, use a 15-blade scalpel and sterile toothed allis tissue forceps to remove the top layer of the wound flap. Variability in the wound dimensions may affect the outcome of the infection and the endpoint colony forming units (CFU).
4. Once all the samples have been wounded, transfer them to the 24-well inserts using sterile forceps. Pipette 15 µL of the bacterial inoculum into the wound bed. Then, incubate for 24 h at 37 °C in a humidified 5% CO₂ tissue incubator.
5. If longer incubation periods are necessary, remove the media and replace it with fresh media every 24 h and incubate in the same conditions.

7. Determination of the bacterial load

1. Remove the media from the bottom of the wells. With sterile forceps, transfer each sample into a separate 50 mL tube filled with 1 mL of sterile PBS.
2. Use a fine-tipped homogenizer to homogenize the surface of the sample. Take care to ensure that the wound bed is in direct contact with the tip of the homogenizer.
3. Homogenize each sample for 35 s on medium/high. The homogenizer detaches the bacteria from the surface of the wound bed to allow for enumeration of the bacterial load.
4. Once all the samples are processed, vortex each sample, in turn, prior to pipetting. This is to ensure the bacterial homogenate is mixed.
5. Pipette 20 µL of vortexed homogenate into the corresponding well of a 96-well plate containing 180 µL of sterile PBS.
6. Serially dilute each sample homogenate to 1 x 10⁻⁷ and pipette 10 µL of the diluted homogenate onto a tryptic soy agar plate in triplicate.
7. Incubate the agar plate for 18 h at 37 °C. Then, count the number of colonies to determine the CFU for each sample.

Results

The identification of a route to sterilize the skin before setting up the wound infection model was challenging. The challenge lay in sterilizing the skin without damaging the different skin layers, which may then go on to have unintended consequences in the outcome of infection. To identify an appropriate sterilization regime, different treatments were tried for varying lengths of time, as outlined in Table 1. Contamination was recorded as the development of turbidity after 48 h in the MK medium used to...

Discussion

The development of antimicrobials is an important but expensive venture that is estimated to cost around $1 billion and take around 15 years to complete. Over 90% of antimicrobial drug discovery and preclinical studies of antimicrobial drug efficacy are carried out by academic researchers and small to medium companies with typically less than 50 employees²². These teams are very financially constrained, which makes the failure of lead molecules in later stages of translational research calamitous....

Materials

Name	Company	Catalog Number	Comments
24 Well Companion Plate	SLS	353504
4 mm Biopsy Punch	Williams Medical	D7484
50 ml centrifuge tubes	Fisher Scientific	10788561
8 mm Biopsy Punch	Williams Medical	D7488
Amphotericin B solution, sterile	Sigma	A2942
Colour Pro Style Cordless Hair Clipper	Wahl	9639-2117X	Hair Clippers
Dual Oven Incubator	SLS	OVe1020	Sterilising oven
Epidermal growth factor	SLS	E5036-200UG
Ethanol	Honeywell	458600-2.5L
F12 HAM	Sigma	N4888
Foetal bovine serum	Labtech International	CA-115/500
Forceps	Fisher Scientific	15307805
Hair Removal Cream	Veet	Not applicable
Heracell VIOS 160i	Thermo Scientific	15373212	Tissue culture incubator
Heraeus Megafuge 16R	VWR	521-2242	Centrifuge
Homogenizer 220, Handheld	Fisher Scientific	15575809
Homogenizer 220, plastic blending cones	Fisher Scientific	15585819
Insert Individual 24 well 0.4um membrane	VWR International	353095
Insulin, recombinant Human	SLS	91077C-1G
Medium 199 (MK media)	Sigma	M0393
Microplate, cell culture Costar 96 well	Fisher Scientific	10687551
Multitron	Infors	Not applicable	Bacterial incubator
PBS tablets	Sigma	P4417-100TAB
Penicillin-Streptomycin	SLS	P0781
Plate seals	Fisher Scientific	ESI-B-100
Safe 2020	Fisher Scientific	1284804	Class II microbiology safety cabinet
Scalpel blade number 15	Fisher Scientific	O305
Scalpel Swann Morton	Fisher Scientific	11849002
Sodium bicarbonate	Sigma	S5761-1KG
Toothed Allis Tissue Forceps	Rocialle	RSPU500-322
Tryptic Soy Agar	Merck Life Science UK Limited	14432-500G-F
Tryptic Soy Broth	Merck Life Science UK Limited	41298-500G-F
Vimoba Tablets	Quip Labs	VMTAB75BX