Mechanical characterization and microneedle puncture criteria for ex vivo bacterial biofilm infected porcine skin tissue

Introduction: There is an emerging interest in the use of microneedles (MNs) for the treatment of bacterial biofilm infections of the skin. Biofilms are composed of bacteria embedded in a hydrated three-dimensional matrix of extracellular polymer substance (EPS).[1] The EPS acts as a protective shell and is reported to physically compromise layers of the skin.[2] The benefits of MNs over widely used oral, topical, and intravenous antibiotics include minimizing the threat of antimicrobial resistance by locally and physically disrupting bacterial biofilms when present, while also limiting the pain the patient may encounter compared to use of hypodermic needles.[3] Although forces required for MN puncture of the stratum corneum (SC) are widely characterized for healthy human and porcine skin over a variety of body locations and demographics,[4,5] a thorough examination of the properties of biofilm infected skin and MN puncture criteria has not been reported. To address this gap, we focused on compression testing via flat-punch indentation and a single solid MN set up to explore whether biofilm-infected skin changes the mechanical properties of porcine skin and how biofilm formation might change needle penetration forces presented on skin. We specifically focused on Pseudomonas aeruginosa, a gram-negative bacterium that can infect open skin wounds and readily forms biofilms. To the best of our knowledge, this is the first-time mechanical characterization via indentation and MN insertion force criteria have been studied for bacterial biofilm infected skin.

Materials and

Methods: Ex vivo frozen Yucatan minipig dorsal skin was cut to a size of 6.35 cm × 2.54 cm. Following this, the hypodermis was removed and thawed overnight at 4°C. After 24 hours, samples were sterilized in 70% (v/v) ethanol and air dried in a biosafety cabinet. After a superficial slice through the SC, samples were placed on 0.6% (w/v) agar. P. aeruginosa PA01 (10 µL) at ~3.03×10^8 colony forming units (CFU)/mL was inoculated and spread evenly over the skin and incubated for 72 h. The control uninfected skin was prepared without bacteria. To quantify bacteria enumeration after incubation, six evenly distributed 8 mm biopsy punches were extracted from each sample to homogenize, dilute, plate on 1.5% (w/v) Pseudomonas isolation agar, and incubated until CFU enumeration (Fig. 1A). To visually detect P. aeruginosa, gram staining images were acquired from histology samples. Indentation and MN puncture studies were conducted using a flat indentation probe (3.25 mm diameter) or MN probe (tungsten, 30 μm tip radius) attached to a load cell (Fig. 2A). Samples were placed and clamped onto a metal block and stretched 15-30% of relaxed length.[6] Indentation tests were initiated after preload detection and lowered at a rate of 116 μm/s until a displacement of 10% of the sample thickness was achieved[7], eliminating substrate effects. Determining MN puncture force used the same set up without preload then lowered at a rate of 70 μm/s until 70-80% of sample height to identify a sudden force drop, indicating puncture.

Results, Conclusions, and Discussions: P. aeruginosa infection on porcine skin for 72 h was successful and equally distributed (Fig. 1A-B), where the infected skin had a bacteria density of ~10^8-10^9 CFU/mL. The gram stain histology images in Fig. 1C of uninfected epidermis exhibit a deep purple, due to the gentian violet solution used that stains the epidermal layer, while in the 72 h infected samples exhibited a physical and color change to the epidermal layer, where pink indicates the gram-negative P. aeruginosa. (Fig. 1D). In this study, we assumed that skin behaves as a linear elastic and isotropic material, where the elastic modulus (EM) via indentation into infected and non-infected skin can be solved for via Hertz contact theorem of a flat punch on an elastic half-space.[5,8] There was no significant difference in EM between 72 h infected (0.19 ± 0.07 MPa) and uninfected (0.18 ± 0.03 MPa) skin, while the depth of the indenter inside the skin displayed a significant difference between the groups, 0.35 ± 0.03 mm and 0.27 ± 0.02 mm, respectively (Fig. 2B-D). This finding may indicate that although the epidermal layer is compromised by the presence of the P. aeruginosa biofilm, leading to more displacement through the skin, the biofilm itself may be exhibiting a variable force like that of the unaltered SC on uninfected skin. Uninfected skin displayed reproducible MN punctures, while MN puncture through 72 h infected skin exhibited a wide variety of magnitudes with six replicates not exhibiting any puncture force (Fig. 2E). Reasoning for variable infection puncture results may include the inherent lack of homogeneity in skin, and formation of biofilm patches on the surface instead of a uniform structure. Future studies will include P. aeruginosa infection at 24 and 48 h incubation time, where the biofilm formation is expected to be minimal and have minor effect on skin compared to 72 h incubation. Findings from these studies can inform how MN used in treatment of biofilm infections may need to be fabricated to enhance their structural integrity and hence, efficacy, in treating bacterial biofilm skin infections.

Acknowledgements (Optional): David Silverberg - Department of Pathology & Laboratory Medicine