To improve our knowledge of cellular and molecular neotissue formation, a murine model of the TEVG was recently developed. The grafts were implanted as infrarenal vena cava interposition grafts in C57BL/6 mice. This model achieves similar results to those achieved in our clinical investigation, but over a far shortened time-course.
Biodegradable scaffolds seeded with bone marrow mononuclear cells (BMCs) are often used for reconstructive surgery to treat congenital cardiac anomalies. The long-term clinical results showed excellent patency rates, however, with significant incidence of stenosis. To investigate the cellular and molecular mechanisms of vascular neotissue formation and prevent stenosis development in tissue engineered vascular grafts (TEVGs), we developed a mouse model of the graft with approximately 1 mm internal diameter. First, the TEVGs were assembled from biodegradable tubular scaffolds fabricated from a polyglycolic acid nonwoven felt mesh coated with ε-caprolactone and L-lactide copolymer. The scaffolds were then placed in a lyophilizer, vacuumed for 24 hr, and stored in a desiccator until cell seeding. Second, bone marrow was collected from donor mice and mononuclear cells were isolated by density gradient centrifugation. Third, approximately one million cells were seeded on a scaffold and incubated O/N. Finally, the seeded scaffolds were then implanted as infrarenal vena cava interposition grafts in C57BL/6 mice. The implanted grafts demonstrated excellent patency (>90%) without evidence of thromboembolic complications or aneurysmal formation. This murine model will aid us in understanding and quantifying the cellular and molecular mechanisms of neotissue formation in the TEVG.
Congenital heart defects are serious conditions that affect nearly 8% of live births in the United States. Approximately 25% of those infants with congenital heart defects or 2.4 per 1,000 live birth, require invasive treatment in the first year of their life1. The most effective treatment for congenital heart disease is reconstructive surgery. Unfortunately, complications arising from the use of currently available vascular conduits are the most significant cause of postoperative morbidity and mortality.
To address this problem, we developed the first tissue engineered vascular grafts (TEVGs) for clinical use2. TEVGs were constructed from biodegradable polyester tubes seeded with autologous bone marrow derived-mononuclear cells (BM-MNCs) and implanted as venous conduits for congenital heart surgery. The results showed excellent patency rates at 1-3 years of follow-up, but with significant incidence of stenosis3,4. It was clear that a better understanding of vascular neotissue formation and the mechanism underlying the development of TEVG stenosis was needed. In order to better understand the development of TEVGs and the mechanism of stenosis development, an ovine model was created5,6. In this model, the TEVGs successfully transformed into living vessels and were similar in both morphology and function of native veins. This use of a large animal model was a good first step in providing important pre-clinical information that aided clinical usage of TEVGs. However, full understanding of the cellular and molecular mechanisms of vascular neotissue formation in TEVGs using large animal models is limited due to limitations in molecular characterization of vascular cell phenotypes due to lack of species specific molecular tools. To overcome these shortcomings, a murine model of TEVGs was developed by reason of the rapid advancement in mouse genetics and their extensive molecular characterization with the added advantage of a shortened time scale.
The murine IVC interposition model faithfully recapitulated the process of neovessel formation that occurs in large animals and humans, but over a much shorter time course6-9. Here, a detailed protocol for small-scale graft manufacturing using biodegradable scaffolds, BM-MNC harvesting and isolation, BM-MNC seeding on scaffold, and graft implantation in a murine model were described.
NOTE: All animal procedures were approved by the Nationwide Children's Hospital Institutional Animal Care and Use Committee.
1. Graft Manufacturing
2. Bone Marrow Mononuclear Cell Harvesting and Isolation
3. Cell Seeding
4. Graft Implantation
A schematic of TEVG implantation is shown in Figure 1. Bone marrow was harvested from a donor mouse and mono nuclear cells were isolated using density centrifugation and then seeded onto a biodegradable scaffold. The seeded scaffolds were incubated O/N and implanted to a recipient mouse as an inferior vena cava interposition graft.
Figure 2 illustrates the scanning electron microscopy of the PGA-P(CL/LA) scaffold. The internal diameter was approximately 1 mm and the wall thickness was approximately 0.17 mm. Total porosity was 78.5% and mean pore sizes were 45.4 ± 17.6 µm.
The gross image of a TEVG interposed into the IVC of C57BL/6 mice is shown in Figure 3. Right after implantation (A) and 2 weeks after implantation (B). (C) shows the gross image of a native IVC in comparison. It is evident that graft was filled with neotissue, however, it still kept the shape of the original graft. Our previous result showed that total graft degradation takes up to 12 weeks10.
The ultrasound B-mode and color Doppler image of a 2 weeks implanted TEVG Figures 4A-B to show the patency of the implanted grafts. A native IVC image was added for comparison (Figure 4, C). Cell seeding improved patency of the graft significantly, patencies at 2 weeks for unseeded and seeded grafts are 65% and 91%, respectively (p<0.05, Fisher's exact test).
Figure 5 shows the cell infiltration and extracellular matrix deposition in TEVG after 2 weeks of implantation. Explanted grafts stained with H&E staining showed neotissue formation throughout the graft and remainders of scaffold material (A). Harts and Masson's trichrome stain showed deposition of Elastin in the intimal layer (B) and collagen in the medial layer (C). Endothelial cell lining was observed throughout the intimal layer (D) and confluent smooth muscle cells were present throughout the TEVG, especially below the EC lining (E) as shown by CD31 and αSMA staining, respectively. By the second week after implantation, macrophage infiltration was evident, as indicated with F4/80 staining (F).
Figure 1. Schematic of TEVG implantation. Bone marrow mono nuclear cells were harvested from a donor mouse, seeded onto a biodegradable scaffold, and then implanted as an IVC interposition graft to a recipient mouse.
Figure 2. Scanning electron microscopy images of a scaffold. The average outer diameter was 1.45 mm, luminal diameter was 1.1 mm, and length was 3 mm.
Figure 3. Implanted TEVG after 2 weeks and native IVC. A) Immediately after implantation, B) 2 weeks after implantation, and C) native IVC for comparison.
Figure 4. Ultrasound images 2 weeks and native IVC. A) B-Mode and B) color Doppler image of TEVG after 2 weeks. C) Native IVC for comparison.
Figure 5. Representative images of the 2 weeks post operative TEVG. A) H&E (inner lumen), B) Harts (pink = elastin), C) Massion's trichrome (blue = collagen), D) CD31 (brown = endothelial cell), E) αSMA (brown = smooth muscle cell), and F) F4/80 (brown = macrophage).
The mouse model of TEVG is a valuable tool to study cellular and molecular mechanisms of neotissue formation and the development of stenosis. The seeded BM-MNC was shown in both histological and SEM images of the seeded cells on the graft11. Cell seeding efficiency was also shown using a DNA assay7. Using this model system we showed that cell seeding reduces the incidence of the development of TEVG stenosis, which was the primary mode of failure in our human clinical trial3. The seeded cells rapidly disappeared from the TEVG, suggesting that they exert their effect via a paracrine mechanism8,12. It was also shown that neovessel formation is an immune-mediated regenerative process, and that the neovessel arises from ingrowth of the vascular cells (EC and SMC) from the neighboring vessel wall along the luminal surface of the scaffold. Moreover, normal macrophage infiltration is essential for vascular neotissue formation, but when this process is excessive it leads to the development of TEVG stenosis7,8. These results show the relevance of this mouse model to human disease. The microsurgical techniques can be adapted to other vascular applications including arterial grafts13 and artery to venous fistula.
We operate over 1,000 IVC interposition graft implantations each year and the mortality rate is less than 1%. For a successful graft implantation, there are several important points to mention.
First, injectable anesthesia is preferred for this operation because it is necessary to turn the mouse around several times during anastomosis. The effect of anesthesia usually lasts around an hour and the entire surgery time for a skilled micro surgeon is approximately 30-45 min, which provides enough time to complete the surgery. If the animal wakes up during the surgery, 0.05-0.1 ml of an anesthesia cocktail can be injected intramuscularly.
Second, most of the small aortic branches needed to be ligated to prevent excess blood loss while separating the IVC and the aorta from the abdominal area. Also keep in mind not to injure the IVC when it is separated from the aorta.
Third, when the graft was sutured, the first two sutures on both sides are the most important steps of the entire process. When they are not placed carefully, it is difficult to separate the front and back layers of the IVC. If the layers are not clearly identified, there is a higher chance to accidentally suture them together. While suturing the IVC to the graft, it is important not to over stretch the IVC, which can cause tearing. Also make sure not to cut off too much IVC to replace the same length of the graft. Usually approximately 1 to 2 mm of IVC was removed to replace a 3 to 5 mm graft due to the longitudinal tension of the IVC. If the same length of the IVC is removed, it makes anastomosis extremely challenging.
Fourth, there are significant strain to strain variations with respect to TEVG formation and stenosis development. For example, unseeded C57BL/6 mice (wild type) showed a much higher rate of stenosis than SCID/bg mice (immunodeficient mice)8,12. In addition, from our experience, SCID/bg mice are much easier to operate in comparison to C57BL/6 mice due to its stiffer IVC wall14. Keep that in mind when operating on a new strain of mice.
The graft manufacturing, BM-MNC harvesting, and cell seeding processes are straight forward, there are only a couple of processes to be cognizant of. First, for the graft manufacturing, it is very important not to 'bunch' the felt while push it using the 18 G blunt needle. We have tried different methods to push the felt down to the distal end of the pipet tips. Pushing the graft with a blunt needle showed the best outcome, by far. The inner diameter of the blunt needle is slightly bigger than the outer diameter of the needle that the felt is wrapped around. Therefore, the smaller needle can snugly slide into the bigger blunt needle and the felt can subsequently be pushed down with equal amounts of force all around the circumference of the felt, which prevents bunching of the felt. We previously observed that bunched grafts showed lower patency at 2 weeks. Also, it is critical to remove all bubbles because bubbles will cause holes on the scaffold surface and blood will leak through the hole after implantation. Second, when the bones are separated for BM-MNC harvesting, make sure to clean the bone as much as possible so they do not include too much muscle or hair. If necessary, it is recommended to use a tissue strainer before the density centrifugation process.
This model of TEVG worked well on venous circulation, which is a low pressure and high flow system. However, in arterial circulation, a high pressure and high flow system, we observed high rates of rupture with PGA grafts due to its fast degradation characteristic. Which means the graft loses its structural strength before the neotissue gains enough strength to resist the arterial pressure. We have used polylactic acid (PLA) grafts previously, but showed high prevalence of aneurysm formation. Also it takes more than 2 years to completely degrade PLA in vivo, which is more than the life span of mice. For a future application in arterial circulation graft manufacturing techniques, for example, an electrospun graft is currently under development.
The authors have nothing to disclose.
This work was supported, in part, by a grant from the NIH (RO1 HL098228) to CKB.
Name | Company | Catalog Number | Comments |
polyglycolic acid (PGA) felt | Biomedical Structures | Custome ordered | |
ɛ-caprolactone and L- lactide copolymer P(LA/CL) |
Gunze Inc. | Custome ordered | |
Pipet tip, 0.1-10 μl | Fisher Sientific | 02-707-456 | |
Lyophilizer | Labconco | 7070020 | |
RPMI medium 1604 | gibco | 11875-093 | |
Petri dish | BD | 353003 | |
24 well plate | Corning | 3526 | |
15cc tube | BD | 352096 | |
Ficoll | Sigma | 10831-100ml | Also called 'Histopaque' |
DPBS | gibco | 14190-144 | |
Littauer Bone Cutter 4.5" Straight | Roboz | RS-8480 | For BM harvesting |
Forceps 4.5" | Roboz | RS-8120 | For BM harvesting |
Scissors 4.5" | Roboz | RS-5912 | For BM harvesting |
Microscope | Leica | M80 | |
C57BL/6J (H-2b), Female | Jackson Laboratories | 664 | 8-12 weeks |
Ketamine Hydrochloride Injection | Hospira Inc. | NDC 0409-2053 | |
Xylazine Sterile Solution | Akorn Inc. | NADA# 139-236 | |
ketoprofen | Fort Dodge Animal Health | NDC 0856-4396-01 | |
Ibuprofen | PrecisionDose | NDC 68094-494-59 | |
Heparin Sodium | Sagent Pharmaceticals | NDC 25021-400 | |
Saline solution (Sterile 0.9% Sodium Chloride) | Hospira Inc. | NDC 0409-0138-22 | |
0.9% Sodium Chloride Injection | Hospira Inc. | NDC 0409-4888-10 | |
Petrolatum Ophthalmic Ointment | Dechra Veterinary Products | NDC 17033-211-38 | |
Iodine Prep Pads | Triad Disposables, Inc. | NDC 50730-3201-1 | |
Alcohol Prep Pads | McKesson Corp. | NDC 68599-5805-1 | |
Cotton tipped applicators | Fisher Sientific | 23-400-118 | |
Fine Scissor | FST | 14028-10 | |
Micro-Adson Forcep | FST | 11018-12 | |
Clamp Applying Forcep | FST | 00072-14 | |
S&T Vascular Clamp | FST | 00396-01 | |
Spring Scissors | FST | 15008-08 | |
Colibri Retractors | FST | 17000-04 | |
Dumont #5 Forcep | FST | 11251-20 | |
Dumont #7 – Fine Forceps | FST | 11274-20 | |
Dumont #5/45 Forceps | FST | 11251-35 | |
Tish Needle Holder/Forceps | Micrins | MI1540 | |
Black Polyamide Monofilament Suture, 10-0 | AROSurgical Instruments Corporation | TI638402 | For sutureing the graft |
Black Polyamide Monofilament Suture, 6-0 | AROSurgical Instruments | SN-1956 | For musculature and skin closure |
Non-Woven Songes | McKesson Corp. | 94442000 | |
Absorbable hemostat | Ethicon | 1961 | |
1 ml Syringe | BD | 309659 | |
3 ml Syringe | BD | 309657 | |
10 ml Syringe | BD | 309604 | |
18G 1 1/2 in, Needle | BD | 305190 | |
25G 1 in., Needle | BD | 305125 | |
30G 1 in., Needle | BD | 305106 | |
Warm Water Recircultor | Gaymar | TP-700 | |
Warming Pad | Gaymar | TP-22G | |
Trimmer | Wahl | 9854-500 |