Fig. 4: Treatment with CAP facilitates bacterial clearance in a 3D skin model of staphylococcal wound infection.

a Schematic representation of experimental timeline. Bioengineered human skin was matured, wounded and inoculated with S. aureus before treatment with CAP or helium. The experimental timeline is not drawn to scale. Created with BioRender.com. b Macroscopic and microscopic images (H&E and MGG) of bioengineered skin. Enlarged view of the boxed region is shown in the inset of H&E-stained micrograph. Scale bar: macroscopic image 1 cm; H&E-stained (large) 100 µm; inset 20 µm; MGG-stained section 20 µm. c Depth-resolved optical section taken 337 µm deep (z-step: 0.5 µm) into bioengineered skin to image live (calcein, green) and dead (ethidium homodimer, red) cells. Scale bar 50 µm. d Epidermal keratinocytes (keratin 14) and dermal-epidermal junction (involucrin and laminin 5) in bioengineered skin. e Immunohistochemical staining (keratin 1, PCNA, vimentin) of bioengineered skin. Scale bar in (d) and (e) is 20 µm. f Bioengineered human skin inoculated with S. aureus. Gram stain (upper panel) and immunohistochemical staining using anti-S. aureus antibodies (lower panel). Scale bar 20 µm. g Violin plot with mean denoted in red. Bioengineered wounds were analyzed for the number of living bacteria (24 h. p. i). CFU average in NT group was normalized to 1 and compared to other groups. n = 3 for NT, n = 7 for control (helium) and n = 10 for CAP. Absolute CFU values and exact p are reported in Supplementary Data 4. *p < 0.05 versus NT. One-way ANOVA non-parametric Kruskal–Wallis with Dunn’s multiple comparison post hoc test. Results are expressed as mean ± s.e.m. of three independent experiments. SC stratum corneum, E epidermis, D dermis, h. p. i. hours post inoculation, H&E hematoxylin and eosin, MGG May Grunwald-Giemsa, NT no treatment, CAP cold atmospheric plasma.