PDF
ePub
Citation
Print
INTRODUCTION
There is an extreme worldwide lack of transplantable organs, and many patients are on waiting lists. According to 2013 data from the Korean Network for Organ Sharing (KONOS), there were 13,689 and 4,263 patients waiting for kidney and liver transplantations, respectively.1 These numbers are increasing by approximately 20% annually, but the numbers of donors remain flat. In 2013, just 1,351 of the 4,263 patients received liver transplantation, but less than one-third were deceased donor transplants.1
This demand and supply mismatch of organs is a persistent worldwide problem,234 and although many studies have been carried out to identify viable alternatives to deceased donor organs, these projects remain in the early stages.5 Part of this effort is a liver support system that is largely categorized into two systems. The first is an artificial liver support system that mainly focuses on detoxification also known as the Molecular Absorbent Recirculating System (MARS™), and the second is a bio-artificial liver support system that detoxifies and has synthetic and regulatory functions. Unfortunately, these systems are extremely expensive, and problems with the membrane and biocompatibility make it unlikely that these solutions will be clinically viable.678
Hepatocyte transplantation is another alternative. The donor's hepatocytes are integrated into the recipient's liver following their insertion into the portovenous system and subsequent translocation to the hepatic sinusoids.91011 This approach has been reported to be clinically effective in congenital metabolic diseases. For example, serum bilirubin levels are reportedly decreased by 25-50% after hepatocyte transplantation in patients with Crigler-Najjar syndrome.1213141516 However, the effect decreased with time, and patients eventually required whole liver transplantation due to hepatocyte graft failure after 4-48 months.
Regarding functional cell differentiation, there have been reports of endothelial cells producing dopamine,17 pancreatic cells secreting insulin,18 and cells functioning as hepatocytes.19 In addition, cells differentiated from autologous cells that function as hepatocytes have been transplanted into human recipients,20 but the lack of angiogenesis associated with these functional differentiated cells remains problematic.
Vasculogenesis and angiogenesis are unique processes. The former is the formation of primary capillary blood vessels from mesodermal cells that first differentiate into angioblasts and then endothelial cells.21 Angiogenesis involves the sprouting of new vessels from existing vessels.21 This is a complex process that involves basement membrane degradation and proliferation, migration, and endothelial cell tube formation. Angiogenesis is critical for the development and wound healing of normal tissue, as well as ensuring adequate oxygen and nutrient supplies to transplanted tissues.22
Angiogenesis is stimulated by hypoxia-inducible factors (HIFs) and the expression of many other signals including vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), and angiopoietin2 (Ang2).23 VEGF is a signaling peptide that facilitates angiogenesis,24 and immunohistochemical labeling of this protein has been used as a histological assessment tool.252627 It increases the permeability of endothelial cells and facilitates the creation of new vessels.28 When overexpressed, it can facilitate the development of cancer, atherosclerosis, rheumatic arthritis, diabetic retinopathy, and so on.2930
Other proteins such as CD31, CD34, CD105, and von Willebrand factor (vWF) also play important roles in angiogenesis. CD34 is a 110-kDa transmembrane glycoprotein expressed in white blood cells, endothelial cells, and stem cells. It can also be found around the splenic marginal zone and in dendritic cells in vessels, nerves, sweat glands, and hair pouches.3132 The function of CD34 is not precisely known, but it is used to identify gastrointestinal stromal tumors and solitary fibrous tumors.3334 It is also used to assess the degree of angiogenesis.27313233343536373839 CD105 (endoglin) is a 180-kDa transmembrane glycoprotein, which is part of the transforming growth factor (TGF)-β receptor complex. It controls angiogenesis by controlling cellular proliferation, differentiation, and migration.40 Unlike the VEGF or CD34, CD105 is weakly expressed in normal tissue and blood vessels but is strongly expressed in tumors.404142
Immunohistochemical labeling for the proteins mentioned above is helpful for assessing the degree of angiogenesis in different model systems.3135373839434445 In the present study, we collected adipose tissue-derived stem cells from adult patients with liver disease, placed them in artificial scaffolds, and implanted these scaffolds in severe combined immunocompromised (SCID) mice to determine whether they induced angiogenesis.
MATERIALS AND METHODS
Experimental model
We employed SCID mice as an experimental model to study angiogenesis in transplanted bio-artificial liver models (Fig. 1). Four scaffolds were transplanted considering possible errors and effectiveness. Adipose tissue-derived stem cells from human patients were implanted into the artificial scaffolds made of type I collagen, and the scaffolds were implanted into the stomach surface of SCID mice. Tissues attached to and near the scaffolds were analyzed 30 days after implantation, which is reportedly a sufficient period of time for adequate angiogenesis.4647 The tissue was processed into frozen sections, and immunohistochemistry labeling was performed to examine the expression of different angiogenesis markers.
Artificial scaffold production
Artificial scaffolds were comprised of adipose tissue-derived stem cells and type I collagen. Adult adipose cells were obtained from abdominal subcutaneous fat acquired during liver tumor resection from consenting patients, and the type I collagen was from pig skin. Harvest of tissues was performed under anesthesia according to procedures approved by the Institutional Review Board (IRB, 4-2013-0757) of Yonsei University Hospital, Korea. Type I collagen is the most abundant collagen in the human body, and it is a constituent of the extracellular matrix of internal organs such as the liver and spleen. Previous studies have shown that angiogenesis occurs through foreign body reaction;46 therefore, type I collagen was considered the perfect material for artificial scaffolds in this experimental model. The model for the artificial scaffold was supplied by a Nanotechnology Laboratory at Yonsei University. An illustration of the artificial scaffold model is shown in Fig. 2.
Artificial scaffold implantation into SCID mice
The 12-week old male SCID mice (Charles River Laboratory, MA, USA) weighing ~20 g were used for the experiment. The study was carried out after receiving approval from the laboratory animal rule committee (animal experiment protocol number 2013-0346). Artificial scaffold implantation proceeded as follows. Anesthesia was induced by placing the animal in an inhalation chambers (RC2-rodent circuit controller, VetEquip, CA, USA) at inhalation level 4, and anesthesia was maintained at level 1.5-2 by placing an inhalation anesthetic tube near the animal's mouth. Under anesthesia, hair around the abdomen was removed, and the incision site at the left lateral decubitus position was disinfected with betadine. After making an incision in the subcostal area of the abdominal wall with surgical scissors, forceps were used to expose the stomach. Silicon tubes were used to scratch the stomach's surface, and the scaffolds were placed near the scratch. Vicryl 5-0 was used to fix the scaffolds by the fundus and pylorus of the stomach. The fixed artificial scaffolds were pressed into the abdominal cavity using cotton swabs, and the abdominal wall was sutured layer by layer using Vicryl 4-0. The anesthetic level was lowered during suturing to ensure a fast recovery of consciousness. The suture site was closely checked for 2-3 days after surgery.
Tissue processing
The specimens including the artificial scaffolds were removed 30 days after implantation to search for evidence of angiogenesis. They were immediately frozen after the mice were sacrificed. 12-µm thick tissue slices were processed from frozen blocks of tissue. The slides were stained with hematoxylin and eosin (H&E) to observe the general characteristics of the artificial scaffolds. To block endogenous peroxidase, the slides were immersed in 0.3% H2O2 solution for 5-10 minutes and then washed for 5 minutes in distilled water. Next, they were stained with hematoxylin for 5 minutes, washed with distilled water three times, washed with phosphate-buffered saline (PBS), immersed in ammonia for 1 minute, washed with distilled water 3 times, and immersed in eosin for 5 minutes. After 2 minutes in 80% ethanol, 2 minutes in 100% ethanol, and 2 minutes in xylene, the slides were cover-slipped with Permount solution and observed under a light microscope (BX51, Olympus, Tokyo, Japan).
Known angiogenesis biomarkers (VEGF, CD34, and CD105) were immunohistochemically assessed. The frozen sections containing the artificial scaffolds and surrounding tissue were fixed for 5-10 minutes. To block endogenous peroxidase, the frozen sections were immersed in 0.3% hydrogen peroxide solution for 5-10 minutes and washed with distilled water for 5 minutes. Afterward, the blocks were placed a moist box and incubated at 4℃ for a day with one of three primary antibodies directed against CD34, CD105, or VEGF (Table 1). Next, the sections were washed three times with PBS for 5 minutes each. Based on the visualization system instructions, the sections were incubated with secondary antibodies including anti-rabbit IgG (HRP) (sc-2749, Santa Cruz, CA, USA) and anti goat IgG (HRP) (sc-2741, Santa Cruz, CA, USA) at room temperature for 2 hours and then washed with PBS four times for 5 minutes each. The 1 : 250 diluted streptoavidin (DAKO, Glostrup, Denmark) reaction was carried out for more than 30 minutes, and then the blocks were washed with phosphate buffer saline (PBS) three times for 5 minutes each. Afterwards, the sections were chemically stained with the brown water-soluble dye diaminobenzidine (Vectorlabs, CA , USA) After washing with distilled water, the sections were incubated with hematoxylin for counterstaining. After a final wash in flowing distilled water, the slides were immersed for 10 minutes in PBS. For dehydration and mounting, the blocks were immersed in 95% alcohol for 2 minutes followed by xylene for 2 minutes. The slides were then cover-slipped with Permount solution and examined under a microscope.
RESULT
Gross findings
The gross findings in mice that were sacrificed and dissected 30 days after artificial scaffold implantation are shown in Fig. 3. The artificial scaffolds were covered with the omentum and adhered to the fundus and pylorus suture areas, as well as to the surrounding organs.
H&E staining
We observed the H&E-stained artificial scaffolds under low magnification to assess their overall appearance. Fig. 4 shows collagen surrounding the scaffold that contained adipose tissue-derived stem cells. The collagen membrane was disrupted where the scaffold was sutured to fix it to the pylorus and fundus, and this membrane linked the artificial scaffold and the stomach.
We next observed the H&E-stained artificial scaffolds under high magnification to better assess its characteristics (Figs. 5, 6). The basement membrane surrounding the outer side of the artificial scaffold contained eosin-stained collagen without nuclei (Fig. 5), and we also observed hematoxylin-stained adipose tissue-derived stem cells within the artificial scaffold (Fig. 6).
Immunohistochemistry with VEGF, CD34, CD105
VEGF-labeled slides showed brown staining starting from the membrane of the artificial scaffold (Fig. 7A, B), but there was no evidence of endothelial cells or vascular structure at ×200 magnification. Although collagen was stained brown by diaminobenzidine, the inside of the cells were not positive for VEGF (Fig. 7E, F).
The inside and outside of the artificial membrane were assessed under high magnification to view cell shapes and differences in VEGF immunohistochemistry. Mouse omentum cells were observed outside the scaffold, and there was no positive staining for VEGF (Fig. 7E). The inside of the artificial scaffold contained hematoxylin-stained adipose tissue-derived stem cells and diaminobenzidine-stained collagen but was negative for VEGF (Fig. 7F).
CD34-labeled slides also showed progressive brown stain starting from the membrane of the artificial scaffold (Fig. 8A, B), but again there was no sign of endothelial cells or vascular structures at ×200 magnification, and the inside of cells were negative for CD34 expression (Fig. 8D, E). Both the inside and outside of the artificial scaffold were checked under high magnification in order to see the cell shape and difference in CD34 immunohistochemistry staining. Mouse omentum cells were observed outside the scaffold, and there was no positive staining for CD34 (Fig. 8E). The inside of the artificial scaffold contained hematoxylin-stained adipose tissue-derived stem cells and diaminobenzidine-stained collagen but was negative for CD34 (Fig. 8F).
The CD105 immunohistochemistry stained slide also showed progressive brown staining starting from the membrane of the artificial scaffold (Fig. 9A, B). There were no endothelial cells or vascular structures at ×200 magnification, and the inside of the scaffold contained only stained collagen and was negative for CD105. Mouse omentum cells were observed outside the scaffold, and there was no positive staining for CD105 (Fig. 9E). On the other hand, the inside of the artificial scaffold contained hematoxylin-stained adipose tissue-derived stem cells and diaminobenzidine-stained collagen but was negative for CD105 (Fig. 9F). Microscopic observation failed to reveal any endothelial cells or vascular structures within the artificial scaffolds. Although diaminobenzidine-stained brown collagen was visible inside the scaffolds, the three angiogenesis markers were not expressed.
DISCUSSION
In the past 20 years, most tissue engineering projects have employed thin structures such as skin or vesical arteries.484950 Manufacturing thick tissues such as muscle, liver, and kidney on scaffolds was not considered possible due to insufficient diffusion of oxygen and nutrients.51 To overcome this problem, the pig liver experimental model was developed by removing most of the cells and only leaving the vessels with mechanical perfusion. Still, microfilaments and microtubules were damaged in the process.52
In the present study, we designed an experimental model under the assumption that vessels from the omentum will grow into the artificial scaffolds implanted onto the vessel-rich mouse stomach. In general, active liver regeneration occurs for 7 days after liver resection in mice, and most show regeneration within a month.53 Previous studies reported that 30 days is a sufficient period of time for angiogenesis to occur. Therefore, we carried performed immunohistochemical labeling 1 month after artificial scaffold implantation. We expected to observe angiogenesis in the artificial scaffolds, but light microscopy of H&E-stained and antibody-labeled sections failed to reveal any evidence of angiogenesis.
There were diaminobenzidine-stained brown regions indicating that there could be transport of materials involved in angiogenesis from outside the artificial scaffold to inside. Specifically, the suture sites where the artificial scaffolds were fixed onto the mice appeared to contain connecting tracts. However, if there was angiogenesis from outside the artificial scaffolds, there would be evidence of endothelial cells or tubular structures inside the scaffolds, and these changes were not observed. In addition, only collagen was stained by diaminobenzidine; there was no evidence of angiogenesis marker expression.
There are several possible reasons why we were unable to find evidences of angiogenesis markers. First, previous studies used 12 weeks old C57bl/6 male mice, whereas we used SCID mice to prevent a reaction to the implanted adipose-derived stem cells. While our animal model choice might have helped maintain the adipose-derived stem cells within the artificial scaffolds, it also could have been an obstacle preventing foreign body reaction and angiogenesis. Secondly, it is possible that the type I collagen scaffolds were not appropriate for inducing angiogenesis. But as van Amerongen et al. reported, angiogenesis occurred through foreign body reaction 7 days after the implantation of type I collagen into mice.46 Thirdly, our time points may have been too short, even though previous studies have shown that angiogenesis starts as early as 7 days after implantation and is clearly observed within 14 and 30 days.4647 For a more appropriate comparison, it would have been better to have a comparison group of scaffold removal at 90 or 180 days.
If our second presumption is correct, it would indicate that artificial scaffolds made with type I collagen do not elicit sufficient tissue reaction. Although there might be adhesion to surrounding tissue, the implanted cells are maintained without angiogenesis, and this suggests that type I collagen artificial scaffolds are an acceptable biomaterial for transplantation. Additional experiments on normal mice are needed to provide additional support for this theory.
In conclusion, there was no evidence of angiogenesis in the artificial scaffolds comprised of type I collagen 30 days after implantation. Similarly, we did not observe endothelial cells or angiogenesis markers. Although the results were negative, they do indicate that such scaffolds could be useful as a new biomaterial. Scaffolds made with type I collagen induced minimal tissue reactions and could maintain implanted cells. Additional research is needed to assess their utility in transplantation.
XML Download