The Three-Dimensional Structure of Porcine Bladder... : Advances in Skin & Wound Care

• Updated on: 2024-04-09
• Read original article here

Author Information
Qinxin Liu, MD, is Research Fellow, Tissue Engineering and Wound Healing Laboratory, Division of Plastic Surgery, Brigham and Women’s Hospital, Harvard Medical School, and Trauma Surgeon, Department of Traumatic Surgery, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, China. Mengfan Wu, MD, PhD, is Postdoctoral Research Fellow, Tissue Engineering and Wound Healing Laboratory, Division of Plastic Surgery, Brigham and Women’s Hospital, Harvard Medical School and Plastic Surgeon, Department of Plastic Surgery, Peking University, Shenzhen Hospital, China. Mehran Karvar, MD, is Postdoctoral Research Fellow, Department of Surgery, Brigham and Women's Hospital. Shimpo Aoki, MD, PhD, is Postdoctoral Research Fellow, Brigham and Women’s Hospital, and Hand Surgeon, Minamitama Hospital, Japan. Yori Endo, MD, is Postdoctoral Research Fellow, Division of Plastic and Reconstructive Surgery, Department of Surgery, Brigham and Women's Hospital, Harvard Medical School. Ryoko Hamaguchi, MD, is Resident, Harvard General Brigham Plastic Surgery program. Chenhao Ma, MD, PhD, MSPH, is Postdoctoral Research Fellow, Division of Plastic Surgery, Brigham and Women's Hospital, Harvard Medical School, and Plastic Surgeon, Plastic Surgery Hospital, Chinese Academy of Medical Science. Dany Y. Matar is Undergraduate Student, Washington University, St Louis, Missouri. Dennis P. Orgill, MD, PhD, is Director and Plastic Surgeon, Wound Healing and Tissue Engineering Laboratory, Brigham and Women’s Hospital, and Professor of Surgery, Harvard Medical School. Adriana C. Panayi, MD, is Principal Investigator, Brigham and Women’s Hospital and Instructor, Harvard Medical School.
Acknowledgment: This work was funded by The Gillian Reny Stepping Strong Center for Trauma Innovation and ACell, Inc, to Brigham and Women’s Hospital.
The authors have disclosed no other financial relationships related to this article.
Submitted May 26, 2021; accepted in revised form July 30, 2021.
Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website ( www.ASWCjournal.com ).
Figure 1:
PORCINE BLADDER EXTRACELLULAR MATRIX (ECM) AND STUDY DESIGNA, Porcine bladder ECM. Three different three-dimensional structures of ECM were used: particulate urinary bladder matrix (pUBM) single-layer sheet UBM (fdUBM) and three-layer sheet UBM (lmUBM). A rectangular prism (w = 30 μm, l = 220 μm, h = 240 μm) from the pUBM was used to establish the surface area (SA) and volume (V) of the pUBM. The thickness of the fdUBM (T = 340 μm) and lmUBM (T = 100 μm) was measured. The area of a square cross-section of the fdUBM and lmUBM sheet necessary for a volume equal to that of the pUBM was 4.7 × 103 and 1.6 × 104 μm2, respectively. The surface-area-to-volume ratio (SAV) was 0.080 μm−1 for the pUBM, 0.01 μm−1 for the lmUBM, and 0.003 μm−1 for the fdUBM. Texture was accounted for by quantifying the distribution of pixel gray-scale values, measured as a standard deviation (SD), to calculate a correction factor (CF) for SA. The SAV, calculated as corrected SA over V was found to be 0.28 μm−1 for the pUBM, 0.02 μm−1 for the fdUBM, and 0.01 μm−1 for the lmUBM. B, Study design. Ninety db/db mice underwent a full-thickness dorsal skin excision (1 cm2). Forty-five db/db mice were treated with a topical particulate of porcine bladder ECM and covered with occlusive dressing (pUBM group) or covered with a low-stiffness (fdUBM + pUBM group) or high-stiffness (lmUBM + pUBM group) fenestrated sheet and occlusive dressing. Thirty mice were split into two groups and covered with a freeze-dried (fdUBM group) or laminated (lmUBM) sheet and occlusive dressing.
The wounds in the first group (n = 15) were covered with a 1.0 × 1.0-cm section of fdUBM (Cytal Wound Matrix 1-Layer scaffold; ACell, Inc). Similarly, the wounds in the second group (n = 15) were covered with a 1.0 × 1.0-cm section of lmUBM (Cytal Wound Matrix 3-Layer scaffold; ACell, Inc). Three groups (n = 45) were first treated with pUBM (MicroMatrix; ACell, Inc), as per the manufacturer’s guidelines. The wounds were then covered with either an occlusive dressing in the pUBM group (n = 15), a 1.0 × 1.0-cm section of Cytal Wound Matrix 1-Layer scaffold and occlusive dressing in the fdUBM + pUBM group (n = 15), or a 1.0 × 1.0-cm section of Cytal Wound Matrix 3-Layer scaffold and occlusive dressing in the lmUBM + pUBM group (n = 15). The final 15 mice were part of the db/db blank group; their wounds were covered with occlusive dressings.
The wounds remained covered with occlusive dressings throughout the experiment; dressings were only removed on inspection days to photograph the wounds. The entire wound, as well as 5 mm surrounding the wound, was collected at days 5, 10, 14, and 21 and split in half (see Table). Half of the wound tissue was fixed in 10% formalin followed by 70% ethanol for histologic and immunohistochemical analysis. The remaining tissue was cryopreserved using liquid nitrogen and stored at −80° C for biochemical analysis. The schedule for photography and sacrifice was based on previous publications (Table). 18,19
Wound Area Measurement
The wounds were photographed on days 5, 10, 14, and 21 using a Nikon D3100 SLR camera (Table). A ruler was included for scale. The photographs were imported into ImageJ software (version 1.52a; Media Cybernetics, Bethesda, Maryland), and the wound area was measured under double-blind conditions by selecting the wound border. Healing was assessed with the following formula:
Wound contraction
where n is day 5, 10, 14, or 21.
Histology
Tissues fixed for histology were embedded in paraffin, and 5-μm-thick cross-sections were cut, ensuring to include the wound bed and border. The sections were stained with standard stains (hematoxylin and eosin and Masson trichrome). For specialized staining, sections were deparaffinized, rehydrated, and probed with antibodies for Ki-67 (1:200, ab15580; Abcam, Cambridge, UK), CD45 (1:50, AF114; R&D Systems, Minneapolis, Minnesota), and anti-K17 (1:100, ab53707; Abcam) for 16 hours at 4° C. Secondary antibodies for anti-Ki-67, anti-CD45, and anti-K17 were goat anti-rabbit immunoglobulin G (IgG) H&L horseradish peroxidase (HRP) (1:2,000; ab205718 Abcam), goat IgG VisUCyte HRP polymer (VC003; Dilution R&D Systems), and goat anti-rabbit IgG H&L HRP (1:2,000, ab205718; Abcam), respectively. Finally, the samples were counterstained with hematoxylin. Double immunofluorescence was performed as previously described, 24 with primary anti-CD31 (1:400, ab7388; Abcam), anti-α-SMA (1:100, ab5694; Abcam), and secondary goat anti-rabbit (CD31, AS1111; α-SMA, AS1110; 1:400; Aspen Biotechnology, Hubei, China). Images were captured (Olympus model BX53, UCMAD3, T7; Olympus, Tokyo, Japan) and merged in ImageJ. Two assessors quantified the parameters under blinded conditions.
The investigators performed quantitative analyses using samples of the wound bed and borders on day 10 postsurgery, as previously described. 18,19 The hematoxylin and eosin- and Masson trichrome-stained sections were used to assess wound bed thickness and collagen deposition, respectively. Anti-CD45-, anti-Ki-67, and anti-K17-probed sections were used to assess leukocyte infiltration, cell proliferation, and keratinocyte migration, respectively. Anti-CD31-probed slides were used to calculate the microvessel density, and combined assessment of anti-CD31 and anti-SMA was used to establish the maturity of new blood vessels through the microvascular pericyte coverage index. 18,19
Scaffold Parameters Assessment
Scanning electron microscope photographs of the three scaffold structures were imported into ImageJ software and used to calculate the wound-exposed surface-area-to-volume ratio (SAV; Figure 1 A, Supplemental Methods, https://links.lww.com/NSW/A96 ).
Statistical Analysis
Statistically significant differences between the groups were assessed with a one-way analysis of variance. Data are expressed as mean ± SD, and statistical significance was set at P < .05. In multiple comparisons, a Bonferroni correction was used to adjust P values. Analysis was performed using GraphPad Prism version 8.00 for MacOS (GraphPad Software, La Jolla, California).
RESULTS
Wound Bed Thickness
The thickness of the wound bed in the blank mice (79 ± 55 μm) was significantly lower than in the fdUBM mice (200 ± 71 μm, P < .001), fdUBM + pUBM mice (360 ± 120 μm, P < .001), and lmUBM mice (260 ± 97 μm, P < .001). Wound bed thickness was similar between the blank and pUBM mice, suggesting that fdUBM enhances the effect of pUBM when used in combination ( Figures 2 A, B).
Figure 2:
HISTOLOGIC ANALYSISA, Wound bed on day 10. Representative sections of hematoxylin and eosin, Masson trichrome, Ki-67, and CD45-stained wounds of the pUBM, fdUBM, fdUBM + pUBM, lmUBM, lmUBM + pUBM, and blank groups on day 10. Scale bar = 100 μm. B, Wound bed thickness. The fdUBM, lmUBM, and fdUBM + pUBM groups had a significantly thicker wound bed than the blank group (P < .0001). C, Collagen deposition. The fdUBM and fdUBM + pUBM groups had significantly lower collagen deposition than the blank group (P < .0001). D, Cellular proliferation. Cellular proliferation was significantly higher in the fdUBM and fdUBM + pUBM groups than the blank, pUBM, lmUBM, and lmUBM + pUBM groups. E, Leukocyte infiltration. The pUBM group had significantly higher leukocyte infiltration than the blank, lmUBM, and lmUBM + pUBM groups. No difference in leukocyte infiltration was shown between the other groups.
Collagen Deposition
The lmUBM + pUBM (39% ± 7.6%) group had the highest collagen deposition, which was significantly higher than those of the blank (9.1% ± 4.9%; P < .0001), pUBM (19% ± 15%; P < .001), and fdUBM (13% ± 5.4%; P < .0001) groups. The fdUBM (13% ± 5.4%) group had significantly lower collagen deposition than the fdUBM + pUBM (27% ± 9%; P = .02) group. Likewise, the lmUBM (20% ± 6.7%) group had significantly lower collagen deposition than the lmUBM + pUBM (39% ± 7.6%; P = .001) group ( Figures 2 A, C).
Leukocyte Infiltration
Treatment with pUBM resulted in a significantly higher leukocyte infiltration (6% ± 1%) compared with the fdUBM + pUBM (3.7% ± 1.7%; P = .006), lmUBM + pUBM (3.7% ± 1%; P = .005), and blank (3.7% ± 1.4%; P = .006) groups. No difference was noted between the blank, fdUBM (4.4% ± 1.2%), and fdUBM + pUBM (4.6% ± 1.3%) groups ( Figures 2 A, D).
Cellular Proliferation
Cellular proliferation, as shown with anti-Ki-67 probing, was significantly higher in the fdUBM (37 ± 17 Ki-67+/high-power field) and fdUBM + pUBM (36 ± 13 Ki-67+/high-power field) groups than in the blank group (11 ± 5.8 Ki-67+/HPF; P < .0001 in both; Figures 2 A, E).
Keratinocyte Migration
The re-epithelialization rate was comparable among the six groups, indicating that the different UBM modalities did not impact keratinocyte migration across the wound surface. Specifically, the percentage of keratinocytes was 33% ± 11% in the blank group, 39% ± 10% in the pUBM group, 46% ± 20% in the fdUBM group, 47% ± 21% in the fdUBM + pUBM group, 35% ± 9.8% in the lmUBM group, and 41% ± 21% in the lmUBM + pUBM group. One sample in the fdUBM group and two samples in the lmUBM group achieved 100% re-epithelialization through the wound center ( Figures 3 A, B).
Figure 3:
KERATINOCYTE MIGRATIONA, Representative panel scans of K17-stained wounds on day 10. B, No differences were noted in the re-epithelialization rate between the six groups. C, Although no differences were noted between groups, two of the mice treated with fdUBM and one treated with lmUBM showed 100% re-epithelialization, as sectioned through the center of the wound.
Angiogenesis
Microvessel density was highest in the lmUBM group (43 ± 36 vessels/high-power field; Figures 5A, B) and was significantly higher than in the blank (12 ± 5.2 vessels/high-power field; P = .02) and pUBM (12 ± 3 vessels/HPF P = 0.01) groups, highlighting potential synergism ( Figures 4 A, B). No differences were noted in terms of microvessel maturity ( Figures 4 A, C).
Figure 4:
MICROVESSEL FORMATIONA, Immunofluorescent-stained sections of the wound bed, day 10. Anti-CD31, anti-α-SMA, and merged stained sections of the wound bed of the six groups on day 10. Scale bar = 100 μm. B, Microvessel density on day 10. Significant differences were noted in the density of microvessels with the fdUBM, fdUBM + pUBM, and lmUBM groups having higher densities than the blank group. C, Microvessel maturity on day 10. No differences were noted in the maturity of the vessels between the six groups.
Rate of Wound Healing
The perimeter of the wounds on days 5 and 10 postsurgery was significantly smaller in the fdUBM + pUBM and fdUBM groups. On day 5, the rate of wound closure of the fdUBM + pUBM-treated mice (48% ± 17% of the wound area on day 0) was higher than that of the pUBM (100% ± 21% of the wound area on day 0; P < .0001), indicating that the addition of the scaffold sheet improved the effect of the two modalities. The same result was found on day 10. The blank and pUBM groups healed the slowest and at a similar rate ( Figure 5 ).
Figure 5:
MOUSE WOUNDSThe particulate extracellular matrix increases the surface area of contact between the scaffold and the wound surface. Degradation of the sheet scaffold results in release of anti-inflammatory factors that may limit the inflammatory response, inducing an environment conducive to remodeling. Other factors released may act as chemoattractants that increase cellular infiltration and proliferation. Growth factor release from the wound bed may also be permissive of these functions.
DISCUSSION
The 3D structure and physical characteristics of an ECM scaffold (eg, surface area available for wound interaction) impact the scaffold’s ability to direct cell proliferation, infiltration, and angiogenesis, which, in turn, influences wound-healing processes, such as wound contraction and re-epithelialization. Although this consequence has previously been noted in other types of wound healing, such as cardiac regeneration, 3 the present study is the first to highlight that variations in 3D structure can directly affect the diabetic skin wound healing efficacy of scaffolds and the first to assess the use of porcine bladder ECM scaffold in a murine model.
Assessing decellularized porcine bladder ECM scaffold in three different physical forms highlighted differences in the wound-exposed surface area. The particulate form had a wound-exposed surface area that was seven times larger than that of the freeze-dried sheet and 14 times larger than that of the laminated sheet.
A possible synergism between the different forms may be hypothesized because the combined modality of freeze-dried sheet and particulate form resulted in faster wound closure than occurred each of the modalities was applied individually. This pattern was evident when analyzing wound bed thickness, which is classically considered to positively correlate with proper wound healing. The thickest wound bed was seen in the mice receiving the combined freeze-dried sheet and particulate modality. Although no significant difference was found in comparison with blank mice when the particulate was used on its own, combination treatment with the freeze-dried sheet increased the efficacy, suggesting synergistic effects. Treatment with the physically laminated sheet also resulted in a significantly thicker wound bed. Similarly, cellular proliferation was significantly increased when the wounds were treated with the freeze-dried sheet, whether on its own or in combination with the particulate. The particulate on its own, however, did not show similar efficacy. Likewise, significant increases in microvessel density were noted when the wounds were treated with the freeze-dried sheet, whether on its own or in combination with the particulate. Consistently, no such density increase was noted when the particulate was used alone. Despite the increase in vessel density, maturation of the microvessels was not compromised, with vessels reaching the same level of maturity. Notably, treatment with the particulate form on its own resulted in significantly elevated leukocyte infiltration. This effect was not evident when the particulate was used in combination with the freeze-dried sheet or the physically laminated sheet, suggesting that use of a sheet modified the inflammatory response induced by the particulate.
The aforementioned variations in wound healing may result because the particulate increases the surface area of contact between the wound interface and the sheet. This increased contact enhances the adequacy of the scaffold-wound bed interface and improves the biological effects on cellular signaling, including cell proliferation, infiltration, and angiogenesis. The freeze-dried sheet works with the particulate form to curtail the inflammatory response, limiting leukocyte infiltration to levels that are more permissive for optimal wound healing ( Figure 6 ). This finding is in agreement with prior research showing that noncross-linked ECM sheets can establish an anti-inflammatory environment that is permissive of constructive remodeling. 20 Further, the resorption process of the ECM scaffold is believed to enhance remodeling, because degradation products act as chemoattractants that recruit host cells. 20
Figure 6:
PROPOSED MECHANISM OF ACTION OF COMBINED SHEET AND PARTICULATE TREATMENTThe particulate extracellular matrix increases the surface area of contact between the scaffold and the wound surface. Degradation of the sheet scaffold results in release of anti-inflammatory factors that may limit the inflammatory response, inducing an environment conducive to remodeling. Other factors released may act as chemoattractants that increase cellular infiltration and proliferation. Growth factor release from the wound bed may also be permissive of these functions.
Taken together, these two mechanisms—increasing surface area contact and limiting the inflammatory response—cause the combination of sheet and particulate ECM to have different effects on wound healing, compared with the effects seen in each modality acting independently. This synergistic phenomenon was not replicated when using the physically laminated sheet with the particulate. Prior research has demonstrated that differentiation of stem cells is optimized when the mechanical properties of a scaffold are similar to that of the tissue being treated. 3,21,22 The physically laminated sheet, in addition to having the lowest wound-exposed surface area, was stiffer than both the freeze-dried sheet and the dorsal murine skin tissue; these differences may be the underlying reason for its limited combined effect.
Clinical Translatability
Porcine-derived scaffolds, owing to a close similarity to human skin, have been used for wound coverage since the 1960s. 2,23 Prior research has highlighted their clinical advantages, including easy handling, adherence to the wound surface, decreased pain, and inhibition of water loss from the wound surface. 24,25 In addition, no study to date has identified any allergic reactions to porcine-derived scaffolds. Given the results of this study, it can be hypothesized that application of these scaffolds also improves the perfusion of the wound and hence promotes wound healing.
Further, diabetic control, with optimization of glucose control, is clinically important for proper wound healing. To the authors’ knowledge, this study is the first to use a diabetic murine model to assess the use of UBM. Small animal models, such as mice, are highly predictive of human wound healing and offer an effective platform for wound therapy development and testing. This study offers evidence that UBM can be tested not only in small animals but also in animals that have been modified to exhibit pathologic healing.
For years, wound healing in the murine model was thought not to represent human healing accurately. Murine wounds were thought to contract, whereas human wounds re-epithelialize; hence, to control for contraction, researchers opted to stent the wounds. Chen et al, 26 however, showed that murine wound healing proceeds through both contraction and re-epithelialization, with the majority of contraction occurring after epithelial closure. In this study, quantitative assessment of healing was performed prior to epithelial closure (day 10 postsurgery). Based on this, as well as prior experience and publications, stenting is not a necessity because the effect of contraction prior to re-epithelialization is minor.
Clarifications, Limitations, and Future Research
Although the mice in this study were diabetic (db/db) and displayed impaired wound healing compared with wild-type mice (C57BL/6 J), 27 this model fails to replicate the pathophysiology of chronic diabetic wounds, such as ulcers. In a recent study, modification of the redox environment in db/db mice led to the development of chronic wounds. 19 To expand on the current work, future research could use a chronic wound model to assess the different scaffold combinations.
CONCLUSIONS
This study highlights that porcine bladder ECM is an effective skin wound healing modality in both sheet and particulate forms. Practitioners should consider the scaffold’s 3D structure when selecting for skin wound healing because variations in the physical structure can impact biological processes such as collagen deposition, cellular proliferation, and angiogenesis. Combinations of the various 3D structures can display synergistic effects. This study serves as a platform to guide future scaffold design and selection for different types of skin wounds and highlights the synergistic potential of combining various 3D structures of scaffolds in diabetic wound healing.
REFERENCES
1. Panayi AC, Orgill DP. Current use of biological scaffolds in plastic surgery. Plast Reconstr Surg 2019;143:209–20.