Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 May;26(9-10):512-526.
doi: 10.1089/ten.TEA.2019.0319. Epub 2020 Jan 28.

Bioprinted Skin Recapitulates Normal Collagen Remodeling in Full-Thickness Wounds

Adam M Jorgensen  1 , Mathew Varkey  1 , Anastasiya Gorkun  1   2   3 , Cara Clouse  1 , Lei Xu  1 , Zishuai Chou  1 , Sean V Murphy  1 , Joseph Molnar  1   4 , Sang Jin Lee  1 , James J Yoo  1 , Shay Soker  1 , Anthony Atala  1
Affiliations

Bioprinted Skin Recapitulates Normal Collagen Remodeling in Full-Thickness Wounds

Adam M Jorgensen et al. Tissue Eng Part A. 2020 May.

Abstract

Over 1 million burn injuries are treated annually in the United States, and current tissue engineered skin fails to meet the need for full-thickness replacement. Bioprinting technology has allowed fabrication of full-thickness skin and has demonstrated the ability to close full-thickness wounds. However, analysis of collagen remodeling in wounds treated with bioprinted skin has not been reported. The purpose of this study is to demonstrate the utility of bioprinted skin for epidermal barrier formation and normal collagen remodeling in full-thickness wounds. Human keratinocytes, melanocytes, fibroblasts, dermal microvascular endothelial cells, follicle dermal papilla cells, and adipocytes were suspended in fibrinogen bioink and bioprinted to form a tri-layer skin structure. Bioprinted skin was implanted onto 2.5 × 2.5 cm full-thickness excisional wounds on athymic mice, compared with wounds treated with hydrogel only or untreated wounds. Total wound closure, epithelialization, and contraction were quantified, and skin samples were harvested at 21 days for histology. Picrosirius red staining was used to quantify collagen fiber orientation, length, and width. Immunohistochemical (IHC) staining was performed to confirm epidermal barrier formation, dermal maturation, vascularity, and human cell integration. All bioprinted skin treated wounds closed by day 21, compared with open control wounds. Wound closure in bioprinted skin treated wounds was primarily due to epithelialization. In contrast, control hydrogel and untreated groups had sparse wound coverage and incomplete closure driven primarily by contraction. Picrosirius red staining confirmed a normal basket weave collagen organization in bioprinted skin-treated wounds compared with parallel collagen fibers in hydrogel only and untreated wounds. IHC staining at day 21 demonstrated the presence of human cells in the regenerated dermis, the formation of a stratified epidermis, dermal maturation, and blood vessel formation in bioprinted skin, none of which was present in control hydrogel treated wounds. Bioprinted skin accelerated full-thickness wound closure by promoting epidermal barrier formation, without increasing contraction. This healing process is associated with human cells from the bioprinted skin laying down a healthy, basket-weave collagen network. The remodeled skin is phenotypically similar to human skin and composed of a composite of graft and infiltrating host cells. Impact statement We have demonstrated the ability of bioprinted skin to enhance closure of full-thickness wounds through epithelialization and normal collagen remodeling. To our knowledge, this article is the first to quantify collagen remodeling by bioprinted skin in full-thickness wounds. Our methods and results can be used to guide further investigation of collagen remodeling by tissue engineered skin products to improve ongoing and future bioprinting skin studies. Ultimately, our skin bioprinting technology could translate into a new treatment for full-thickness wounds in human patients with the ability to recapitulate normal collagen remodeling in full-thickness wounds.

Keywords: animal models; bioprinting; extracellular matrix; skin; tissue engineering; wound healing.

PubMed Disclaimer

Conflict of interest statement

No competing financial interests exist.

Figures

FIG. 1.
FIG. 1.
Bioprinted skin fabrication, surgical method, and experimental groups. Diagram of the bioprinted CAD/CAM design, printing process, skin layers, and in vitro histology and a visual description of the steps involved in the preparation, wound creation, treatment administration, protective bolster bandaging of animals, and the treatment groups in this study. (A) The three-layer biomimetic skin design consists of a hypodermal layer (preadipocytes), a dermal layer (dermal fibroblasts and microvascular endothelial cells), and epidermal layer (keratinocytes and melanocytes), each printed using a separate syringe. (B) CAD/CAM design of the square graft consisted of parallel deposited rods 3 cm in length with a 90° rotation with each layer. (C) Bioprinting is performed in a sterile environment. (D) H&E staining of bioprinted skin confirms high cellularity, while the fibrin hydrogel only control graft is acellular. (E) The final product measures about 3 × 3 cm and is then allowed to crosslink in fibrin for 1 h. (F) The anesthetized mice were surgically prepped with isopropyl alcohol and betadine swabs. (G) The area for the excisional wound was marked with one 2.5 × 2.5 cm square drawn with sterile marker on the dorsum. (H) For the creation of the defect, the skin wound was created by removing 2.5 × 2.5 cm of full-thickness skin in the central back along the thoracic and lumbar area. (I) Wounds were immediately treated either with bioprinted skin (I1) or a fibrin hydrogel vehicle control (I2); an additional group underwent no treatment (I3). (J) The wound area was covered with an Adaptic™ dressing and sutured in place at the four corners. (K) The wound was then covered with a Tegaderm™ bandage, and a folded piece of gauze was placed on top. (L) Surgical tape was then used to attach the bandage to the wound area, forming a bolster dressing. CAD/CAM, computer aided design and computer aided manufacturing; H&E, hematoxylin and eosin. Color images are available online.
FIG. 2.
FIG. 2.
Digital images of full-thickness wound closure over 21 days and H&E stained samples at day 21. (A) Digital photos were taken for each of the time points (days 0, 7, 14, 21) and compiled for each of the wounds. Gross morphology demonstrates accelerated wound closure time, and wounds were treated with bioprinted skin (images scaled to 3 × 3 cm; scale bar 5 mm). (B) Epithelialization in the bioprinted skin (top) was confirmed through H&E staining. The hydrogel only (middle) and untreated (bottom) wounds had only sparse coverage and incomplete epithelialization (stitched image at 20 × magnification; scale bar 200 μm; n = 4). Color images are available online.
FIG. 3.
FIG. 3.
Digital planimetry for total wound closure, contraction, and epithelialization. Individual plots for (A) total wound closure, (B) contraction, and (C) epithelialization over 21 days (****p < 0.0001; ***p < 0.001; NS, p > 0.05, n = 4). (D) The original wound area, outer healing wound border, and open wound area were measured for each treatment group at each time point using ImageJ. Total wound closure was calculated by subtracting the open wound area from the original wound area, divided by the original wound area [Total Wound Closure = (A − C)/A]. Contraction was calculated by subtracting the outer healing wound border from the original wound area, divided by the original wound area [Contraction = (A − B)/A]. Epithelialization was calculated by subtracting contraction from total wound closure (Epithelialization = Total Wound Closure − Contraction). The time to complete wound closure was quantified. (E) Combined wound healing analysis represents the total wound area at each time point divided into the percentage of wound area (red), percentage contraction (blue), and percentage epithelialization (green). With these analyses describing the healing wound over time, we can observe wound closure through competing parameters of contraction and epithelialization for each treatment. The best performing treatments were categorized as having the following: (1) small wound area, (2) least contraction, and (3) most epithelialization; the worst performing treatments had the opposing characteristics. Products are ranked from the best healing (left) to worst healing (right). Bioprinted skin had complete wound closure, greater epithelialization, and less contraction compared to hydrogel only and untreated wounds at day 21. Color images are available online.
FIG. 4.
FIG. 4.
Representative histological images for H&E, Masson's Trichrome, and Picrosirius Red. (A) Representative H&E stained histological images (Row 1; 20 × magnification; scale bar 50 μm; n = 4). Only the bioprinted skin group had developed an epidermis. Bioprinted skin-treated groups appear to have a dermis structure similar to healthy skin (thicker bundles, basket weave organization), while hydrogel only and untreated wounds had sparse or incomplete epidermis and a thick unorganized dermis. Representative Trichrome staining of tissues (Row 2; 20 × magnification; scale bar 50 μm; n = 4). Bioprinted skin showed the most similar staining intensity for keratin (red) and collagen (blue), compared to healthy skin. Hydrogel only treated wounds had some staining for keratin (red) and sparse dermal collagen staining (blue). Untreated wounds were highly disorganized, highly cellular (brown), and were not representative of healing skin. Representative Picrosirius Red staining of tissues viewed with polarizing light (Row 3; 20 × magnification; scale bar 50 μm; n = 4). Healthy skin is dominated by red mature collagen type I in a basket weave orientation. Bioprinted skin appeared most like healthy skin with this basket weave organized collagen fibers, while control hydrogel treated wounds had thick orange fibers organized in parallel, typical of scar formation. Untreated wounds appeared to show increased levels of immature green collagen type III fibers and were also organized more in parallel. (B) Summary table of histological findings. (C) MATLAB was used to quantify the relative percentage of each collagen color type by dividing the pixel count of each color by the total pixel count for each image, shown here as the average percentage for each treatment group (n = 4). Color images are available online.
FIG. 5.
FIG. 5.
Picrosirius Red analysis with CurveAlign and CT-FIRE to measure collagen fiber alignment, length, and width. (A) Representative picrosirius red stained histological images under polarized light (day 21, 40 × magnification; scale bar 20 μm; n = 4). (B) Collagen fiber alignment was measured using CurveAlign software and is represented as alignment, with zero being no alignment or more normal and 1 being complete alignment or more scar like (***p < 0.001; *p < 0.05; NS, p > 0.05; n > 10,000 fibers per group). Bioprinted skin had less aligned collagen fibers than the control hydrogel and wound only controls, but was more aligned than the human and mouse skin controls. (C) Collagen fiber length was measured using CT-FIRE software and is represented as length in pixels (****p < 0.0001; NS, p > 0.05; n > 10,000 fibers per group). Bioprinted skin has similar collagen fiber length to bioprinted skin and was significantly shorter than mouse skin, control hydrogel, and wound only treatment groups. (D) Collagen fiber width was measured using CT-FIRE software and is represented as width in pixels (****p < 0.0001; NS, p > 0.05; n > 10,000 fibers per group). Collagen in bioprinted skin was thinner than wound only samples, human skin, and mouse skin, but was significantly thicker than control hydrogel treated wounds. CT-FIRE, curvelet transform-fiber extraction. Color images are available online.
FIG. 6.
FIG. 6.
Representative scanning electron microscopy images. Scanning electron microscopy images of bioprinted skin, control hydrogel, and mouse skin (Row 1 scale bar 500 μm; row 2 scale bar 100 μm; n = 4). The bioprinted skin treated wounds displayed a unique separation in the dermis formed by the incorporated bioprinted skin in the wound area, with mouse epidermis healing over the top (E1–E3). Dermal collagen in the bioprinted skin appeared to be in a normal, basket-weave orientation (D1) similar to normal mouse skin (D3). Alternatively, the control hydrogel group had thick parallel collagen bundles in the dermis, typical of scar formation (D2). These results confirm our findings with picrosirius red staining and quantification in Figure 6.
FIG. 7.
FIG. 7.
Representative immunofluorescent images for Lamin A+C, Pan-Cytokeratin, Mel5, Adiponectin, and ZO-1. (A) Representative images of Pan-Cytokeratin (green) and Human Lamin A + C (red) immunofluorescent images (Column 1; 20 × magnification; scale bar 50 μm; n = 4). Bioprinted skin was the most like normal tissue, with a thick epidermis (arrows) and integrated human cells (circles, see also A1; 40 × magnification; scale bar 20 μm; n = 4), while the control hydrogel group only had sparse keratin staining in the epidermal region and no human cells. Representative images of Mel5 (green) and Pan-Cytokeratin (red) immunofluorescent images (column 2; 40 × magnification; scale bar 20 μm; n = 4). Bioprinted skin had some positive staining for Mel5 (circles) in both the superficial and deep dermis, while the control hydrogel group had little or no positive staining. Representative images of Adiponectin (green) and Vimentin (red) immunofluorescent images (column 3; 20 × magnification; scale bar 50 μm; n = 4). Bioprinted skin had intense staining for vimentin (arrows) in the dermis, but no positive adiponectin staining (circles), suggesting that preadipocytes in the bioprinted skin may have differentiated into fibroblasts. Control hydrogel treated wounds had no vimentin staining due to the high human specificity of the antibody, but there was some adiponectin staining deep in the dermis. Representative images of ZO-1 (green) and Pan-Cytokeratin (red) immunofluorescent images (column 4; 20 × magnification; scale bar 50 μm; n = 4). Bioprinted skin was the most like human skin, with ZO-1 (circles) staining highlighting vascular lumens throughout the dermis (see also A2; 40 × magnification; scale bar 20 μm; n = 4). The control hydrogel group also had positive staining, but the structures were less luminal in shape. (B) Summary table of histological findings. Color images are available online.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. American Burn Association. Burn incidence fact sheet. 2018. http://ameriburn.org/who-we-are/media/burn-incidence-fact-sheet (accessed November8, 2019)
    1. Peck M.D. Epidemiology of burns throughout the world. Part I: distribution and risk factors. Burns 37, 1087, 2011 - PubMed
    1. Bickers D.R., Lim H.W., Margolis D., et al. . The burden of skin diseases: 2004 a joint project of the American Academy of Dermatology Association and the Society for Investigative Dermatology. J Am Acad Dermatol 55, 490, 2006 - PubMed
    1. Kurd S.K., Hoffstad O.J., Bilker W.B., and Margolis D.J.. Evaluation of the use of prognostic information for the care of individuals with venous leg ulcers or diabetic neuropathic foot ulcers. Wound Repair Regen 17, 318, 2009 - PMC - PubMed
    1. Jeschke M.G., Shahrokhi S., Finnerty C.C., Branski L.K., and Dibildox M.. Wound coverage technologies in burn care: established techniques. J Burn Care Res 34, 612, 2013 - PMC - PubMed

Publication types

LinkOut - more resources