A novel approach for the endothelialization of xenogeneic decellularized vascular tissues by human cells utilizing surface modification and dynamic culture

概要

www.nature.com/scientificreports

OPEN

A novel approach
for the endothelialization
of xenogeneic decellularized
vascular tissues by human cells
utilizing surface modification
and dynamic culture
Wen‑Jin Ho 1, Mako Kobayashi 2,6, Kozue Murata 1,3,4, Yoshihide Hashimoto 2, Kenji Izumi 5,
Tsuyoshi Kimura 2, Hideo Kanemitsu 1,7, Kazuhiro Yamazaki 1, Tadashi Ikeda 1,
Kenji Minatoya 1, Akio Kishida 2 & Hidetoshi Masumoto 1,3*
Decellularized xenogeneic vascular grafts can be used in revascularization surgeries. We have
developed decellularization methods using high hydrostatic pressure (HHP), which preserves
the extracellular structure. Here, we attempted ex vivo endothelialization of HHP-decellularized
xenogeneic tissues using human endothelial cells (ECs) to prevent clot formation against human
blood. Slices of porcine aortic endothelium were decellularized using HHP and coated with gelatin.
Human umbilical vein ECs were directly seeded and cultured under dynamic flow or static conditions
for 14 days. Dynamic flow cultures tend to demonstrate higher cell coverage. We then coated the
tissues with the E8 fragment of human laminin-411 (hL411), which has high affinity for ECs, and
found that Dynamic/hL411showed high area coverage, almost reaching 100% (Dynamic/Gelatin vs
Dynamic/hL411; 58.7 ± 11.4 vs 97.5 ± 1.9%, P = 0.0017). Immunostaining revealed sufficient endothelial
cell coverage as a single cell layer in Dynamic/hL411. A clot formation assay using human whole
blood showed low clot formation in Dynamic/hL411, almost similar to that in the negative control,
polytetrafluoroethylene. Surface modification of HHP-decellularized xenogeneic endothelial tissues
combined with dynamic culture achieved sufficient ex vivo endothelialization along with prevention of
clot formation, indicating their potential for clinical use as vascular grafts in the future.
Cardiovascular disease remains a major cause of death ­globally1,2. Several standardized options are available for
treating ischemic heart and peripheral arterial diseases, such as vascular graft i­ mplantation3, stent i­ mplantation4,
and drug i­ ntervention5. Among these, vascular graft implantation using autologous grafts provides high treatment efficacy, better quality of life, and ­prognosis6. Even though autologous vascular grafts are the most suitable
for revascularization surgery targeting small arteries, graft preparation is not always ideal because of the poor
quality of autologous grafts, possibly associated with comorbidity or previous surgery in the p
­ atient7, or shortage
of preparation time owing to hemodynamic instability during surgery, especially in emergent cases. To address
this dilemma, an advanced bioengineering graft platform needs to be established.

1

Department of Cardiovascular Surgery, Graduate School of Medicine, Kyoto University, 54 Shogoin
Kawahara‑cho, Sakyo‑ku, Kyoto 606‑8507, Japan. 2Department of Material‑Based Medical Engineering, Institute
of Biomaterials and Bioengineering, Tokyo Medical and Dental University, Tokyo, Japan. 3Clinical Translational
Research Program, RIKEN Center for Biosystems Dynamics Research, Kobe, Japan. 4Institute for Advancement
of Clinical and Translational Science, Kyoto University Hospital, Kyoto, Japan. 5Tokai Hit., Co, Ltd., Fujinomiya,
Japan. 6Present address: Department of Materials Processing, Graduate School of Engineering, Tohoku University,
Sendai, Japan. 7Present address: Department of Cardiovascular Surgery, Kitano Hospital, Osaka, Japan. *email:
masumoto@kuhp.kyoto-u.ac.jp
Scientific Reports |

(2022) 12:22294

| https://doi.org/10.1038/s41598-022-26792-w

1
Vol.:(0123456789)

www.nature.com/scientificreports/

Figure 1.  Cell viability assay. (a) A custom-made bioreactor system for dynamic cell culture. Upper: schematic
of the entire system. Lower: the perfusion culture chamber. (b) Representative fluorescence images of cultured
tissues. “D2” indicates “Two days of culture”. (c) Area coverage. (d) Signal intensity. HUVECs, human umbilical
venous endothelial cells; HHP, high hydrostatic pressure; hL411, human laminin-411. Scale bars: 3 cm in (a),
5 mm in (b) (lower magnification) and 1 mm in (b) (higher magnification; rightmost column). *P < 0.05,
**P < 0.01, ***P < 0.001 (vs Dynamic/hL411). †P < 0.05, ‡P < 0.01 (vs day 2).
As a new paradigm to prepare vascular grafts as surrogate autologous grafts, decellularization of xenogeneic
vascular tissue has recently been introduced, wherein immunological issues mediated by xenogeneic cells are
­mitigated8,9. Two major approaches are used for decellularization: chemical reagent-based10 and physical manipulation-based11 methods. To maintain the extracellular matrices (ECMs) after decellularization with a sufficient
efficiency of decellularization and microorganism removal, we developed an original physical decellularization
method called a high hydrostatic pressure (HHP) ­method12–14. We recently confirmed that implantation of
xenogeneic HHP-decellularized vascular tissue in a large animal model could confer endothelial cell migration
and maintenance of endothelium in the implanted vascular t­ issue15. However, thromboembolism occurred after
implantation when the inner surface of the vascular grafts was damaged by dry conditions during surgery, possibly owing to the absence of an intact endothelial cell layer before implantation; this encouraged us to address
this problem for the broader application of this technology.
Ex vivo recellularization is an approach used to reconstruct intact tissue structures, and can possibly help
prevent acute thromboembolism after ­implantation16. To develop humanized vascular grafts from HHP-decellularized xenogeneic vascular tissues with biological vascular function, recellularization using human vascular
cells might be an optimal approach. In particular, endothelialization of decellularized vascular tissues before
in vivo implantation could be crucial to confer endothelial function to the graft, which might help prevent acute
clot formation after implantation.
In the present study, we hypothesized and validated a novel method for the endothelialization of HHPdecellularized xenogeneic vascular tissues with human endothelial cells utilizing tissue surface modification
by coating the tissue surface with an ECM that possesses high affinity with human endothelial cells, along with
a dynamic flow culture method. We also confirmed that the optimized endothelialization protocol sufficiently
prevented acute clot formation against human blood, and further validated this in an ex vivo test.

Results

The combination of coating with the E8 fragment of human laminin‑411 and dynamic flow cul‑
ture enables sufficient endothelial cell coverage on the decellularized vascular tissue.  Dynamic

flow culture has been reported to improve the recellularization efficiency of decellularized pericardial ­tissues17,
possibly owing to the reproduction of the physical conditions in the living body. We prepared a custom-made
bioreactor that could confer dynamic flow of the culture medium and incubated gelatin-coated slices of porcine
aorta endothelium seeded with human umbilical venous endothelial cells (HUVECs) under dynamic flow conditions (Fig. 1a). We found sufficient coverage of HUVECs on decellularized tissue in earlier incubation periods
regardless of the dynamic condition (area coverage at day7: Static/Gelatin vs Dynamic/Gelatin; 76.0 ± 4.5 vs
Scientific Reports |
Vol:.(1234567890)

(2022) 12:22294 |

https://doi.org/10.1038/s41598-022-26792-w

2

www.nature.com/scientificreports/

Figure 2.  Histological evaluation of recellularized tissues. (a) Representative hematoxylin and eosin (H&E)
staining for HUVECs (arrow). (b) Representative H&E staining for HUVECs in each culture condition.
Arrows indicate HUVECs. (c) Representative immunohistochemical staining for CD31-positive HUVECs
(dotted line). (d) Representative immunohistochemical staining for CD31-positive HUVECs in each culture
condition. Arrows indicate HUVECs. (e) Representative immunohistochemical staining for VE-cadherinpositive HUVECs (dotted line). (f) Representative immunohistochemical staining for CD31-positive HUVECs
in each culture condition. Arrows indicate HUVECs. DAPI, 4’,6-diamidino-2-phenylindole; VEcad, Vascular
endothelial-cadherin. Scale bars: 10 µm in (a), (c) and (e); 100 µm in (b), (d) and (f).
88.0 ± 6.2%, P = 0.9881), but coverage started to decline after 7 days of incubation in the static incubation groups
[Static/Gelatin: day2 vs day4 vs day7 vs day14; 97.8 ± 1.6 vs 90.7 ± 3.4 vs 76.0 ± 4.5 vs 42.8 ± 11.3%, P = 0.4342
(day2 vs day7), < 0.0001 (day2 vs day14), 0.0163 (day7 vs day14)]. Although the area coverage was 97.3 ± 1.4%
at 2 days after incubation, coverage significantly declined at 14 days compared to that at 2 days even in dynamic
culture [Dynamic/Gelatin: day2 vs day4 vs day7 vs day14; 97.3 ± 1.4 vs 91.3 ± 5.9 vs 88.0 ± 6.2 vs 58.7 ± 11.4%,
P = 0.9992 (day2 vs day7), 0.0019 (day2 vs day14), 0.0631 (day7 vs day14)] (Fig. 1b,c).
Laminin-411 has a high affinity for human endothelial cells. To further improve the efficiency of endothelialization, we coated decellularized tissues with the recombinant E8 fragment of human laminin-411 (hL411)18,19.
The combination of dynamic flow culture and coating with hL411 significantly improved the efficiency of coverage at 14 days after incubation, reaching almost 100% (area coverage at day14: Dynamic/Gelatin vs Dynamic/
hL411; 58.7 ± 11.4 vs 97.5 ± 1.9%, P = 0.0017 (Fig. 1b,c). ...

この論文で使われている画像