A Prevascularized Sinus Tract on the Liver Surface for Islet Transplantation : Transplantation

Journal Logo

Original Basic Science

A Prevascularized Sinus Tract on the Liver Surface for Islet Transplantation

Li, Feng MD1; Lv, Yi MD1; Li, Xiaohang MD1; Yang, Zhaoming MD1,2; Guo, Tingwei MD1; Zhang, Jialin MD, PhD1

Author Information
Transplantation 107(1):p 117-128, January 2023. | DOI: 10.1097/TP.0000000000004236
  • Free
  • SDC
  • Infographic

Abstract

INTRODUCTION

Since the Edmonton protocol was proposed in 2000,1 >1500 people have received islet transplants worldwide.2 Most clinical islet transplants are performed through the portal vein,3 but this approach inevitably causes inherent defects, such as the instant blood-mediated inflammatory reaction (IBMIR),4 portal embolism,5 liver necrosis,6 and inability to retrieve the graft. It has been reported that up to 70% of grafts were lost in the early stage after transplantation, and islets from 2 to 4 donors are usually needed clinically to meet the needs of 1 recipient.7,8 The lack of suitable transplantation sites and approaches has become a bottleneck, restricting the development of islet transplantation.2

Finding an ideal transplantation site that conforms to physiology and is conducive to the survival of islet grafts is 1 of the 3 key issues (β-cell source, transplantation site, and immune protection) that need to be resolved in the field of islet transplantation.9 Alternative islet transplantation sites that have been reported so far include subrenal capsule,10 omentum,11 subcutaneous,12,13 anterior eye chamber, etc.14 These models played important roles in elucidating the favorable microenvironment for the survival of islet grafts, but until now, they have been rarely used in clinical practice.

In recent studies, high hopes are placed on the liver surface, and islets were transplanted on the liver surface by means of protein gel,15 cell sheet,16,17 or membrane structure.18,19 However, almost all of these methods require large islet equivalents, reflecting that the local blood supply and microenvironment of the unmodified liver surface were not conducive to the survival of islet grafts. It is worth considering whether the local microenvironment on the liver surface can be further optimized by some engineering methods before islet transplantation.

In this study, for the first time, a prevascularized sinus tract (PST) on the liver surface was constructed for islet transplantation in a mouse model by temporarily embedding a 4× silk thread between the liver surface and the attached decellularized human amniotic membrane (DAM), and ~200 syngeneic islets (the marginal dose of intrahepatic islet transplantation) were transplanted into the PST to observe the therapeutic effect of these islets on the recipient diabetic mice.

MATERIALS AND METHODS

Animals

Specific pathogen-free grade, male C57bl/6J or Balb/c mice, weighing 20–25 g, were obtained from the Laboratory Animal Resource Center of Liaoning Province. All experimental protocols were conducted in accordance with the National Institutes of Health Guide for the Use and Care of Laboratory Animals (NIH Publications No. 8023, revised 1978) and approved by the Ethical Committee of First Hospital of China Medical University ([2020] No. 276).

Preparation of the DAM

All human amniotic membranes were obtained from the Obstetrics Department of the First Hospital of China Medical University, and the entire research protocol was approved and supervised by the Ethics Committee of the First Hospital of China Medical University [(2020) No. 276]. The selection criteria for participants and the written informed consent signing process were the same as described previously.20

The DAM was prepared with some modifications based on the method described previously.21,22 Briefly, the obtained fresh human amniotic membrane (Figure 1A) was repeatedly washed with 4 °C precooled phosphate buffer saline (PBS; Hyclone, Logan, UT) to remove clots, and then the human amniotic membrane was divided into small pieces and frozen in a solution consisting of a mixture of equal volumes of Dulbecco’s Modified Eagle Medium (Hyclone) and glycerol (Solarbio, Beijing, China) at −80 °C. Next, it was resuscitated before use (Figure 1B) and wiped with a cotton swab dipped in 0.5 M sodium hydroxide solution (Xinxing, Liaoning, China) to remove amniotic membrane epithelial cells. Finally, the DAM was repeatedly washed with PBS and placed on ice for later use (Figure 1C).

F1
FIGURE 1.:
Preparation and morphological observation of the DAM. A, The obtained fresh human amniotic membrane. B, Human amniotic membrane after cryopreservation and resuscitation. C, DAM. D, HE staining of the human amniotic membrane. E, HE staining of the DAM. F, Masson’s staining of the DAM. DAM, decellularized human amniotic membrane; HE, hematoxylin-eosin. Scale bars, 50 μm.

Preparation of the 4× Silk Thread

Two 4-0 silk threads (Ethicon, USA) were stacked together, and then the head and tail ends of the silk threads were tied into a 4× silk thread (approximately 1 mm in diameter). The prepared 4× silk thread was trimmed into small sections about 1-cm long and subjected to high-temperature steam sterilization for later use (Figure 2A). For allogeneic islet transplantation, to increase the space of the PST to accommodate 500 islets, the 4-0 silk thread was replaced with a 3-0 silk thread, and accordingly, the diameter of the resulting 4× silk thread was increased to about 1.2 mm.

F2
FIGURE 2.:
Creation of the PST on the liver surface. A, The 4× silk thread of about 1 cm (diameter of about 1 mm). B, The 4× silk thread was laid on the curved surface of the left lobe of the liver. C, A DAM was covered on the 4× silk thread with its edge glued to the liver surface. D, Two weeks after the 4× silk thread was embedded; the PST was created by the removal of the 4× silk thread after reopening the abdominal cavity. The arrow indicates the entrance of the PST. DAM, decellularized human amniotic membrane; PST, prevascularized sinus tract.

Construction of the PST on the Liver Surface

Two weeks before islet transplantation, the recipient C57bl/6J mice were anesthetized with isoflurane (RWD, Shenzhen, China) inhalation via a small animal anesthetic machine (MIDMARK, USA). The mice were fixed, and the left liver lobe was exposed through an incision of about 2.5 cm in the midline of the abdomen. The 4× silk thread prepared above was laid on the curved surface of the left lobe of the liver (Figure 2B), and then, a DAM was covered on the 4× silk thread. Subsequently, the edge of the DAM (approximately 4-mm wide area) was glued to the liver surface by Fibrin Sealant (RAAS, Shanghai, China); thus, the 4× silk thread was embedded between the DAM and the liver surface (Figure 2C) (caution: the fibrin sealant should not be in contact with the embedded 4× silk thread). Finally, the abdominal wall was sutured precisely with 6-0 silk thread. All mice involved in surgery were injected subcutaneously with tramadol hydrochloride (30 mg/kg) (Tianlong, Liaoning, China) and cefazolin sodium (90 mg/kg) (Lukang, Shandong, China) in the first 3 d after surgery.23 At the time of transplantation, the PST, allowing islet transplantation, was created by the removal of the embedded 4× silk thread after reopening the abdominal cavity (Figure 2D).

As a control group, similar to the construction method of the PST, a DAM was attached to the surface of the left lobe of the liver, but no silk thread was embedded (the DAM group). The corresponding schematic protocol is available in Figure 3A.

F3
FIGURE 3.:
Histological analysis of the PST on mouse liver surface. A, Diagram of the experimental protocol. HE staining of local tissues in each group: the DAM group (B1) and the PST group (B2). Masson’s staining of local tissues in each group: the DAM group (C1) and the PST group (C2). CD68(+) immunohistochemical staining of local tissues in each group: the DAM group (D1) and the PST group (D2). CD31(+) immunohistochemical staining of local tissues in each group: the DAM group (E1) and the PST group (E2). Compared with the DAM group, the percentage of CD68(+) macrophages (F) and CD31 (+) vascular endothelial cells (G) in the local tissues of the PST group increased significantly (**P < 0.01 and ***P < 0.001 compared with the DAM group), and there was a correlation (H) between the 2 (r2 = 0.7674; P < 0.0001). DAM, decellularized human amniotic membrane; HE, hematoxylin-eosin; PST, prevascularized sinus tract. Scale bars, 100 μm.

Western Blotting

Western blotting was used to detect the expression of proangiogenic-related proteins in local tissues of the PST (the corresponding schematic protocol is available in Figure 4A). The local tissues (n = 4, in each group) from the sham operation control group (the Con group), the DAM group, and the PST group were procured and lysed as previously described.24 Samples containing 40 μg of total protein were electrophoresed and then transferred to 0.22 μm polyvinylidene fluoride membranes (Millipore, USA). After blocking with 5% nonfat milk for 2 h (except for vascular endothelial growth factor A [VEGFA]: 3% nonfat milk for 1 h), the membranes were probed with primary antibodies against hypoxia inducible factor-1α (HIF-1α; ab1, Abcam, UK), Wnt3a (WL02179, Wanleibio, Liaoning, China), VEGFA (ab1316, Abcam, UK), matrix metalloprotein 9 (MMP9; ab38898, Abcam, UK), β-actin (66009-1-Ig, Proteintech, China), and α-Tubulin (ab7291, Abcam, UK) at 4 °C overnight. After incubating the corresponding secondary antibody for 1 h at room temperature, proteins were visualized by an enhanced chemiluminescence kit (Proteintech, Wuhan, China). The intensity of bands was measured using the Image Lab 5.0 software (Bio-Rad, USA).

F4
FIGURE 4.:
Proangiogenesis proteins were upregulated in local tissues of the PST on the liver surface. A, Diagram of the experimental protocol. Protein levels of HIF-1α (B), Wnt3a (C), VEGFA (D), and MMP9 (E) were assessed by Western blot. Quantification of protein levels showed the enhanced expression of HIF-1α (F), Wnt3a (G), VEGFA (H), and MMP9 (I). Results are shown as means ± SD of 4 independent experiments. One-way ANOVA was performed for the multiple-group followed by the Tukey’s test. HIF-1α: *P < 0.05 compared with the Con group, **P < 0.01 compared with the Con group, and ††P < 0.01 compared with the DAM group. Wnt3a: n.s. compared with the Con group, *P < 0.05 compared with the Con group, and **P < 0.01 compared with the DAM group. VEGFA: ‡‡P < 0.01 compared with the Con group, **P < 0.01 compared with the Con group, and ††P < 0.01 compared with the DAM group. MMP9: n.s. compared with the Con group, **P < 0.01 compared with the Con group, and ††P < 0.01 compared with the DAM group. ANOVA, analysis of variance; CON, control; DAM, decellularized human amniotic membrane; HIF-1α, hypoxia inducible factor-1α; MMP9, matrix metalloprotein 9; n.s., not significant; PST, prevascularized sinus tract; VEGFA, vascular endothelial growth factor A; WB, Western blot.

Mouse Islet Isolation

Donor mouse islets were isolated on the basis of our previous experience.25 Briefly, the common bile duct was punctured and 1–2 mL of collagenase V solution (1 mg/mL) (Sigma-Aldrich, China) was injected to inflate the pancreas. Subsequently, the pancreas was digested in a 37 °C water bath for 10–20 min. Afterward, islets were purified by Histopaque-1077 (TBD, Tianjin, China) density gradient centrifugation. Finally, about 200 islets were handpicked (for allogeneic transplantation, about 500 islets from Balb/c mice were collected) and placed in precooled medium 199 (Gibico, USA) at 4 °C for later use. The quality control was the same as before: purity and vitality were both >90%.23

Establishment of Diabetes Model

One week before transplantation, the recipient mice were induced to become diabetic by intraperitoneal (i.p.) injection of Streptozotocin (Sigma-Aldrich, China) at 180 mg/kg, as previously done.23 Mice with a nonfasting blood glucose of 20–30 mmol/L were included in the follow-up study.

Marginal Dose Syngeneic Islet Transplantation

Recipient mice were randomly divided into 6 groups (n = 6, the corresponding schematic protocol is available in Figure 5A):naive group (normal mice, no special treatment), 1-stage decellularized human amniotic pouch (DHAP1) group (diabetic mice, ~200 islets were transplanted into DHAP1), 2-stage decellularized human amniotic pouch (DHAP2) group (diabetic mice, ~200 islets were transplanted into DHAP2), PST group (diabetic mice, ~200 islets were transplanted into the PST on the liver surface), kidney capsule (KC) group (diabetic mice, ~200 islets were transplanted under the KC), and Sham group (diabetic mice, sham operation).

F5
FIGURE 5.:
Technical essentials of islet transplantation in each group of mice. A, Diagram of the experimental protocols. B, Islet transplantation in the DHAP1 group. The edge of a DAM was glued to the surface of the left lobe of the liver, and ~200 islets (*) were transplanted in a pouch-like structure between the liver surface and the DAM, and then the entrance of the pouch (arrow) was sealed with the Fibrin Sealant. C, Islet transplantation in the DHAP2 group. After the edge of the DAM was glued to the liver surface for 14 d, the DAM was partially lifted, and ~200 islets (*) were transplanted into the pouch-like structure between the liver surface and the DAM, and then the entrance of the pouch (arrow) was sealed with the Fibrin Sealant. D, Islet transplantation in the PST group. After implantation of the 4× silk thread between the DAM and the liver surface for 14 d, the embedded 4× silk thread was pulled out to form the PST on the liver surface, and ~200 islets (*) were subsequently transplanted into the PST. The entrance of the PST (arrow) was sutured with an 11-0 silk thread. E, Islet transplantation in the KC group. Around two hundred islets (*) were transplanted under the upper polar capsule by inserting a silicone tube from the lower pole of the left kidney, and then electrocoagulation was performed to close the lower pole capsule entrance (arrow). DAM, decellularized human amniotic membrane; DHAP1, the 1-stage decellularized human amniotic pouch; DHAP2, the 2-stage decellularized human amniotic pouch; IPGTT, intraperitoneal glucose tolerance test; KC, kidney capsule; PST, prevascularized sinus tract; STZ, streptozotocin.

Procedures in the PST group (Figure 5D): 2 wk after the silk thread was embedded, the recipient mice were anesthetized again. The left lobe of the liver was exposed by carefully reopening the original abdominal incision. After the local DAM was torn, one end of the embedded 4× silk thread was exposed, and then the entire 4× silk thread was slowly pulled out. Next, the previously prepared islets were centrifuged into a pellet and delivered within the PST using a silicone tube connected to the microsyringe (Gaoge, Shanghai, China). Immediately afterward, the DAM at the entrance of the PST was closed with 11-0 suture to prevent leakage of islet suspension.

Procedures in the DHAP1 group (Figure 5B): on the day of islet transplantation, the recipient mice were anesthetized. Similar to the construction method of the PST, the entire edge of the DAM was pasted on the surface of the left liver lobe (except for a small opening reserved at the lower right edge of the left liver lobe). Subsequently, a silicone tube with a diameter of 1 mm was temporarily inserted between the liver surface and the DAM. At the time of transplantation, the temporarily implanted silicone tube was pulled out to form the DHAP1, and then the islet suspension was delivered within the pouch. Immediately afterward, the entrance of the pouch was closed with fibrin sealant (RAAS, Shanghai, China).

Procedures in the DHAP2 group (Figure 5C): 2 wk before transplantation, a DAM was attached to the surface of the left lobe of the liver as previously described (caution: no silk thread was embedded). On the day of islet transplantation, the recipient mice were anesthetized, and the left lobe of the liver was exposed. Similar to the construction method of the DHAP1, the local potential gap between the DAM and the liver surface was carefully separated to temporarily implant a silicone tube as before. At the time of transplantation, the temporarily implanted silicone tube was pulled out to form the DHAP2, and then the islet suspension was delivered within the pouch. Immediately afterward, the entrance of the pouch was closed with fibrin sealant (RAAS, Shanghai, China).

Procedures in the KC group (Figure 5E): as a positive control group, islets were transplanted under the KC according to the previous method.10

Evaluation of Syngeneic Islet Graft Metabolic Function

The nonfasting blood glucose of the recipient mice was monitored using a blood glucosemeter (ACCU-CHEK, Roche, USA). For the first 40 d after transplantation, the nonfasting blood glucose was measured every 2 d. The reversal of diabetes was defined as 2 consecutive readings <11.1 mmol/L and maintained until the islet grafts were removed.

On the 40th day after transplantation, an i.p. glucose tolerance test was performed to further evaluate the function of the islet grafts as previously described.25 Briefly, after fasting overnight, mice in each group were intraperitoneally injected with 50% glucose solution (Kelun, Sichuan, China) at 2-g/kg body weight, and then, blood glucose was measured at 0, 15, 30, 60, 90, and 120 min after the injection.

Long-term Syngeneic Islet Graft Function

For the protection of animal ethics and welfare, each group of mice receiving islet transplantation measured nonfasting blood glucose once a month from the 40th day to the 128th day after transplantation.

On the 128th day after transplantation, these mice were anesthetized again. The pedicle of the left lobe of the liver was ligated to completely remove the islet graft in the DHAP1, DHAP2, and PST groups as previously described.25 As a positive control group, islet grafts in the KC group were totally removed on the 128th day after transplantation by resection of the left kidney. Subsequently, the mice in each group continued to be fed for 1 wk, and their nonfasting blood glucose was monitored.

Allogeneic Islet Transplantation Into the PST on the Liver Surface

Approximately 500 islets from Balb/c mice were transplanted into the PST on the liver surface of recipient C57bl/6J mice, in the presence or absence of rapamycin therapy (1 mg/kg daily, i.p., from day 0 to day 7 posttransplant, dissolved in 0.2% carboxylethyl cellulose, 0.25% polysorbate-80 in 0.9% NaCl, all Sigma-Aldrich, China). Nonfasting blood glucose was measured daily for the first 20 d after transplantation, and then changed to weekly blood glucose measurement. Once the blood glucose value was found to be >13.6 mmol/L, the blood glucose was remeasured the next day. Rejection was defined as 2 consecutive readings >13.6 mmol/L followed by excision of the graft for histological observation.

Histological Assessment

The samples were fixed with 4% paraformaldehyde (Biosharp, Anhui, China), dehydrated with gradient ethanol (Shengtai, Tianjin, China), permeabilized with xylene (Shengtai, Tianjin, China), embedded in paraffin, and cut into slices.

For the hematoxylin-eosin (HE) staining, the slices were stained with a hematoxylin and eosin staining kit (Solarbio, Beijing, China). For the Masson’s staining, the sections were stained using a Masson’s staining solution kit (Servicebio, Wuhan, China).

For immunohistochemical staining, the sections were subjected to antigen retrieval and endogenous peroxidase activity quenching, and then incubated with anti-CD31 (ab28364, Abcam, UK), anti-CD68 (GB11067, Servicebio, Wuhan, China), or anti-insulin (ab7842, Abcam, UK) primary antibodies at 4 °C overnight. Subsequently, the sections were incubated with corresponding secondary antibodies and detected using a diaminobenzidine chromogenic kit (MXB, Fuzhou, China).

A fluorescence microscope (ECLIPSE 80i, Nikon, Japan) was used for image acquisition, and Image J software was used for semiquantitative analysis of the image.26

Statistical Analysis

The statistical analysis was performed using GraphPad Prism 8.0 software. The measurement data were expressed as mean ± SD. The differences between the groups were compared by Student’s unpaired t-test with 2-tailed P or 1-way analysis of variance followed by Tukey’s multiple comparisons test. P < 0.05 was considered statistically significant.

RESULTS

Preparation and Morphological Observation of the DAM

After repeated washing with PBS, the obtained blood-stained fresh human amniotic membrane (Figure 1A) was transformed into a white semipermeable membrane. After cryopreservation and resuscitation, the naked eye shape of amniotic membrane did not change significantly (Figure 1B). HE staining showed that the frozen amniotic membrane had a complete shape and clear layers (Figure 1D). A single layer of amniotic epithelial cells can be seen in the epithelial layer, and a few fibroblasts can be seen in the fibroblast layer. After rubbing with sodium hydroxide solution (Figure 1C), amniotic membrane epithelial cells and internal fibroblasts disappeared, but the collagen-rich basal layer, dense layer, and sponge layer remained (Figure 1E and F).

Morphological Observation of the PST on the Liver Surface

After the edges were glued with Fibrin Sealant, a DAM could cover the 4× silk thread and stick to the liver surface (Figure 2C). On the 14th day after operation, there was no obvious displacement among the DAM, embedded 4× silk thread, and the liver surface. By carefully pulling out the implanted 4× silk thread, the PST on the liver surface can be formed (Figure 2D, the arrow shows the entrance of the PST).

The cross-section of the PST stained with HE (Figure 3B2) shows that the sinus cavity (*) of the PST was located between the DAM and the liver surface. The sinus wall on the side adjacent to the liver surface was composed of a large number of new granulation tissues, containing a large number of inflammatory cells (mainly neutrophils and macrophages [Figure 3D2]) and new capillaries (Figure 3E2). Masson’s staining shows that the DAM and the granulation tissue of the sinus wall were rich in collagen (Figure 3C2). Compared with the DAM group, the proportion of CD68(+) macrophages (Figure 3D1, D2, and F; P < 0.01) and CD31(+) vascular endothelial cells (Figure 3E1, E2, and G; P < 0.001) in the local tissues of the PST group increased significantly, and there was a correlation between the 2 (Figure 3H; Y = 1.107 × X + 0.4585; r2 = 0.7674; P < 0.0001).

Proangiogenesis Proteins Were Upregulated in Local Tissues of the PST on the Liver Surface

Compared with the DAM group, the expression of HIF-1α, Wnt3a, VEGFA, and MMP9 protein in the local tissues of the PST group was significantly upregulated (Figure 4; P < 0.01).

Nonfasting Blood Glucose Measurements After Syngeneic Transplantation

After islet transplantation, the blood glucose of the mice in the DHAP1, DHAP2, PST, and KC groups decreased gradually. However, only the blood glucose of the PST group (19.3 ± 3.3 d) and KC group (17.9 ± 3.4 d) reversed to the normal range, whereas the Sham group did not see a significant drop in blood glucose during the entire observation period (Figure 6A). There was no statistical difference in diabetes reversal rate between PST group and KC group (P > 0.05; Figure 6C).

F6
FIGURE 6.:
Regulation of nonfasting blood glucose by syngeneic islet grafts within the PST. A, Nonfasting blood glucose monitoring of recipient mice and control mice in the first 40 d after transplantation. Data points represent blood glucose mean ± SD. The dotted straight line represents 11.1 mmol/L. B, The area under the blood glucose curve of mice in each group, expressed as mmol/L/38 d. The n.s.1 compared with the DHAP2 group, ††††P < 0.0001 compared with other groups. The n.s. 2 compared with the KC group, ****P < 0.0001 compared with other groups; ‡‡‡‡P < 0.0001 compared with other groups except the DHAP1 group; §§§§P < 0.0001 compared with other groups except the PST group (1-way ANOVA followed by the Tukey’s test). C, The percentage of recipients achieving euglycemia. P = 0.0005 for the post hoc comparison of the PST and DHAP1 (or DHAP2) group. P = 0.3860 for the post hoc comparison of the PST and KC group (Log-rank [Mantel-Cox] test). ANOVA, analysis of variance; AUC, area under the curve; DHAP1, the 1-stage decellularized human amniotic pouch; DHAP2, the 2-stage decellularized human amniotic pouch; IPGTT, intraperitoneal glucose tolerance test; KC, kidney capsule; n.s., not significant; PST, prevascularized sinus tract; STZ, streptozotocin.

The area under the nonfasting glucose curve in the PST group (448.5 ± 32.4 mmol/L/38 d) was comparable to that in the KC group (443.9 ± 37.1 mmol/L/38 d; P > 0.05) but significantly lower than that in the DHAP1 group (783.8 ± 88.0 mmol/L/38 d; P < 0.0001) and DHAP2 group (761.7 ± 44.4 mmol/L/38 d; P < 0.0001) (Figure 6B).

Intraperitoneal Glucose Tolerance Test for Syngeneic Islet Grafts

As shown in Figure 7B, the area under the glucose tolerance curve value in the PST group (1879.8 ± 126.8 mmol/L/120 min) was similar to that in the KC group (1802.0 ± 141.6 mmol/L/120 min; P > 0.05), but significantly lower than that in the DHAP1 group (2997.5 ± 174.9 mmol/L/120 min; P < 0.0001) and DHAP2 group (2909.8 ± 147.4 mmol/L/120 min; P < 0.0001), indicating that the islet grafts in the PST group had excellent ability to regulate blood glucose.

F7
FIGURE 7.:
Intraperitoneal glucose tolerance test of syngeneic islet grafts in each group, 40 d posttransplantation. A, Blood glucose curve post dextrose bolus. Data points represent blood glucose mean ± SD. B, Areas under the blood glucose curve post dextrose bolus, expressed as mmol/L per 120 min. **P < 0.01 compared with the Sham group, the n.s.1 compared with the DHAP2 group, ††††P < 0.0001 compared with other groups; ***P < 0.001 compared with the Sham group, ‡‡‡‡P < 0.0001 compared with other groups except the DHAP1 and Sham group; The n.s.2 compared with the KC group, ****P < 0.0001 compared with other groups; §§§§P < 0.0001 compared with other groups except the PST group (1-way ANOVA followed by the Tukey’s test). ANOVA, analysis of variance; DHAP1, the 1-stage decellularized human amniotic pouch; DHAP2, the 2-stage decellularized human amniotic pouch; KC, kidney capsule; n.s., not significant; PST, prevascularized sinus tract.

Long-term Islet Graft Retrieval After Syngeneic Transplantation

On the 128th day after transplantation, mice in the DHAP1, DHAP2, PST, and KC groups were operated again to completely remove the left liver lobe (left kidney) where the islet grafts were located. The nonfasting blood glucose levels of mice in each group rose to near the pretransplant level within 1 wk after surgery (Figure 8).

F8
FIGURE 8.:
Long-term function of syngeneic islet grafts transplanted into the PST on the liver surface. Similar to the KC group, the PST recipients maintained normoglycemia >100 d until the islet grafts were removed (arrow), from which point the blood glucose levels returned to near the pretransplant level in a week. In contrast, during the entire observation period, the nonfasting blood glucose of the mice in the DHAP1 and DHAP2 groups did not return to normal blood glucose levels, and only a certain degree of decline was observed. The data points represent blood glucose mean ± SD. The dotted straight line represents 11.1 mmol/L. DHAP1, the 1-stage decellularized human amniotic pouch; DHAP2, the 2-stage decellularized human amniotic pouch; KC, kidney capsule; PST, prevascularized sinus tract.

Histopathological Staining of the Syngeneic Islet Grafts

On the 128th day after transplantation, the islet grafts of each group of mice were resected for histopathological observation. As shown in Figure 9A3–D3, the islet grafts in the PST group were oval in cross section, occupying the position where the previous PST was located, that is, one side was surrounded by surrounding liver parenchyma, and the other side was covered by the DAM. Furthermore, the islet grafts in the PST group had abundant local blood supply (Figure 9D3), uniformly expressed insulin in the cytoplasm (Figure 9C3), no excessive fibrosis (Figure 9B3), and no obvious local inflammatory cell infiltration (Figure 9A3). However, in the DHAP1 group (Figure 9A1–D1) and DHAP2 group (Figure 9A2–D2), only a small number of islet grafts can be seen between the liver surface and the DAM. As a positive control group, the corresponding pictures of islet grafts in the KC group were shown in Figure 9A4–D4.

F9
FIGURE 9.:
Histological morphology of syngeneic islets transplanted long-term into the PST on the liver surface. HE staining of islet grafts in each group: the DHAP1 group (A1), the DHAP2 group (A2), the PST group (A3), and the KC group (A4). Masson’s staining of islet grafts in each group: the DHAP1 group (B1), the DHAP2 group (B2), the PST group (B3), and the KC group (B4). Insulin(+) immunohistochemical staining of islet grafts in each group: the DHAP1 group (C1), the DHAP2 group (C2), the PST group (C3), and the KC group (C4). CD31(+) immunohistochemical staining of islet grafts in each group: the DHAP1 group (D1), the DHAP2 group (D2), the PST group (D3), and the KC group (D4). DHAP1, the 1-stage decellularized human amniotic pouch; DHAP2, the 2-stage decellularized human amniotic pouch; HE, hematoxylin-eosin; KC, kidney capsule; PST, prevascularized sinus tract. Scale bars, 100 μm.

Efficacy of Allogeneic Transplantation Into the PST on the Liver Surface

To determine whether the PST on the liver surface is efficacious across the major histocompatibility complex–completely mismatched alloimmune barrier, islets from Balb/c mice were transplanted into the PST on the liver surface of C57bl/6J mice, in the presence or absence of rapamycin therapy. As shown in Figure 10C1, in the absence of immunosuppressive agents, islet grafts survived within the PST 10 d after transplantation. However, rejection occurred within 16 d (median survival was 12.5 d). Rapamycin therapy led to prolonged allograft survival (median survival was 55 d; P < 0.01) and even partial allograft survived >90 d (Figure 10A). In some mice treated with rapamycin, islet graft morphology was preserved (Figure 10B4) and showed clear insulin immunohistochemical staining (Figure 10C4), whereas in islet grafts that had been immunorejected, islet structure was destroyed (Figure 10B2 and B3) with a little or no insulin staining (Figure 10C2 and C3).

F10
FIGURE 10.:
Allogeneic islet transplantation into the PST on the liver surface. A, Impact of an allogeneic barrier upon diabetes reversal using the PST on the liver surface. P = 0.0035 for the post hoc comparison of the Control and Rapa group (Log-rank [Mantel-Cox] test). HE staining of islet grafts in each group: Control group, 10 d posttransplant (B1); Control group, 13 d posttransplant (B2); Rapa group, 42 d posttransplant (B3); Rapa group, 91 d posttransplant (B4). Insulin(+) immunohistochemical staining of islet grafts in each group: Control group, 10 d posttransplant (C1); Control group, 13 d posttransplant (C2); Rapa group, 42 d posttransplant (C3); Rapa group, 91 d posttransplant (C4). HE, hematoxylin-eosin; PST, prevascularized sinus tract. Scale bars, 100 μm.

DISCUSSION

At present, most clinical islet transplants are performed via the portal vein, but the inherent defects of the transplant site such as IBMIR cause up to 70% of the islet grafts to be lost.7 Usually, islets from 2 to 4 donors are needed to meet the needs of 1 recipient.2,27 The optimized alternative transplantation site is expected to significantly improve the efficiency of islet transplantation, allowing limited organ resources to benefit more patients.

In this study, a PST on the liver surface was constructed for islet transplantation in a mouse model by temporarily embedding a 4× silk thread between the liver surface and the attached DAM, for the first time, as a novel promising clinical islet transplantation site.

Our results show that the local blood supply of the PST was abundant, and the inner wall of the PST was rich in collagen, which mimics the microenvironment of the islets in the pancreas to a certain extent.28,29 At the same time, proangiogenic proteins including VEGFA and HIF-1α were significantly upregulated in the local tissues of the PST, which will help revascularization of islet grafts.30-32 Therefore, it is reasonable to speculate that the PST is a site beneficial to the survival of islet grafts.

Transplantation of a marginal dose of syngeneic islets into the PST routinely reversed the hyperglycemia of the recipient mice and maintained normoglycemia for >100 d until the graft was removed. The islet grafts within the PST achieved significantly better results to the islet grafts in the nonprevascularized control group (the DHAP1 and DHAP2 group) and comparable results to the islet grafts in the KC group (the gold-standard approach in mice but not feasible for humans because of anatomical differences33) with respect to glycemic control and glucose tolerance. These results indicated that islet grafts in the PST demonstrated excellent glycemic regulation capability. Besides, although not a fully immune-privileged site, PST on the liver surface can still support allogeneic islet grafts survival in the presence of immunosuppression.

The islet grafts transplanted into the PST may exhibit the following advantages. Compared with subcutaneous islet transplantation, the insulin secreted by the islet graft within the PST can be more physiologically delivered into the liver to regulate blood glucose. Compared with islet transplantation via portal vein, it can avoid defects such as IBMIR and portal hypertension and reverse the recipient’s hyperglycemia with a smaller transplant equivalent. The islet graft on the liver surface facilitates monitoring of its function and preserves the possibility of removing part of the liver to retrieve the graft when necessary, especially when transplanting stem cell–derived beta cells in the future.34,35 Compared with the reported methods of islet transplantation on the liver surface,15-18 the PST on the liver surface was preintegrated with the rich blood supply network in the liver, avoided the damage of islet digestion into single cells, and for the first time achieved a marginal dose islet transplantation to reverse the diabetes of the recipient mice. Compared with the hepatic sinus tract in our previous research,25,36 the PST on the liver surface retains the advantages of the hepatic sinus tract and can completely avoid the risk of damaging the intrahepatic vasculature. More importantly, it is no longer difficult to expand the receiving space of the PST by adjusting the amount and shape of the embedded biomaterials, which helps to meet the needs of islet volume in clinical practice in the future. It is comforting that although the construction of PST requires secondary operations, these procedures are not difficult for an experienced laparoscopic surgeon. The above-mentioned characteristics of PST have laid a solid foundation for its future clinical application prospects. Furthermore, the construction method of the PST provides a novel engineering strategy for cell transplantation on the surface of other solid organs.

To the best of our knowledge, this study is the first in the world to utilize the temporary foreign body reaction (FBR) to construct a PST on the surface of a solid organ. Studies have shown that the initiation of FBR is closely related to the provisional matrix formed on the material surface because of local tissue damage caused by foreign body implantation.37 However, unlike subcutaneous or hepatic parenchyma, laying the silk thread on the liver surface may not cause local tissue damage, so it is an interesting question whether the desired FBR can still be triggered. Additionally, it is also worth exploring whether neovascularization can penetrate the liver capsule to preconnect PST with the rich network of hepatic sinusoid. At the technical level, this study innovatively selected a self-made silk thread with good flexibility to achieve a close fit with the irregular curved liver surface. Furthermore, this study used fibrin sealant for the first time to firmly adhere DAM to the liver surface, which provided a solid guarantee for the construction of PST on the liver surface.

On the other hand, this study applies DAM to the construction of PST for the first time in the world. Amniotic membrane is an easily obtained natural biomaterial, and its good biocompatibility and safety have been fully clinically recognized in the fields of ophthalmology and burns for years. In contrast to complex artificial scaffolds used in other prevascularization strategies,38,39 the long-term implantation of amniotic membrane in vivo does not need to worry about excessive fibrosis. Studies have shown that amniotic membrane has biological functions such as promoting angiogenesis, preventing adhesion, and anti-inflammatory.40 Our previous study also confirmed that the amniotic membrane extract contains multiple growth factors, including hepatocyte growth factor, epidermal growth factor, etc, which have protective effects on islet survival and function.20 Therefore, compared with other prevascularization strategies, islet transplantation into PST on the liver surface may be independent of the additional administration of expensive growth factors39,41,42 or other accessory cells (such as autologous adipose-derived stem cells or fibroblasts).38,43 In this study, the flexibility of the amniotic membrane ensured a tight fit to the liver surface. Furthermore, in contrast to other cell sheet/membrane structures, the strength of the amniotic membrane prevents the pull-out operation of the silk thread from causing the collapse of the PST. More importantly, compared with reported keratinocyte sheets,19 the antiadhesion feature of the amniotic membrane avoids the dense adhesion of intra-abdominal organs in the surgical area, making it possible to completely resect the graft if necessary. Additionally, amniotic membrane is rich in collagen and other extracellular matrix components, which may mimic the tissue basement membrane to a certain extent, thereby facilitating the growth of granulation tissue and the survival of islet grafts.

Nevertheless, there are some limitations to the present study. The effects of different materials of implanted foreign bodies on the function of islet grafts have been reported,12 but only silk was studied in this study. The optimal strategy for constructing PST on the liver surface remains to be further elucidated in subsequent studies. Additionally, because of technical barriers, this study could not directly compare the engraftment of islet grafts in each group of mice in the early posttransplantation period, which could only be reflected by indirect indicators such as nonfasting blood glucose and glucose tolerance.

In conclusion, by attaching a DAM to the surface of mouse liver and temporarily embedding a 4× silk thread, a PST lined with granulation tissue can be formed. The PST on the liver surface is a favorable site for islet transplantation, and the islet grafts within the PST can effectively reverse the blood glucose of the recipient diabetic mice.

ACKNOWLEDGMENTS

This work was inspired by the research of Professor A.M. James Shapiro from University of Alberta12,44 and the study of Professor Wanxing Cui from MedStar Georgetown University Hospital.18 The authors thank Professor Tao Meng from the Obstetrics Department of the First Hospital of China Medical University for her technical assistance for this study.

REFERENCES

1. Shapiro AM, Lakey JR, Ryan EA, et al. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Engl J Med. 2000;343:230–238.
2. Shapiro AM, Pokrywczynska M, Ricordi C. Clinical pancreatic islet transplantation. Nat Rev Endocrinol. 2017;13:268–277.
3. Pepper AR, Korbutt GS. Conditioning the liver into a favorable niche for pancreatic islet engraftment. Am J Transplant. 2021;21:2927–2928.
4. Bennet W, Groth CG, Larsson R, et al. Isolated human islets trigger an instant blood mediated inflammatory reaction: implications for intraportal islet transplantation as a treatment for patients with type 1 diabetes. Ups J Med Sci. 2000;105:125–133.
5. Sakata N, Hayes P, Tan A, et al. MRI assessment of ischemic liver after intraportal islet transplantation. Transplantation. 2009;87:825–830.
6. Sakata N, Obenaus A, Chan N, et al. Factors affecting islet graft embolization in the liver of diabetic mice. Islets. 2009;1:26–33.
7. Yin D, Ding JW, Shen J, et al. Liver ischemia contributes to early islet failure following intraportal transplantation: benefits of liver ischemic-preconditioning. Am J Transplant. 2006;6:60–68.
8. Vivot K, Benahmed MA, Seyfritz E, et al. A metabolomic approach (1H HRMAS NMR Spectroscopy) supported by histology to study early post-transplantation responses in islet-transplanted livers. Int J Biol Sci. 2016;12:1168–1180.
9. Vantyghem MC, de Koning EJP, Pattou F, et al. Advances in β-cell replacement therapy for the treatment of type 1 diabetes. Lancet. 2019;394:1274–1285.
10. Szot GL, Koudria P, Bluestone JA. Transplantation of pancreatic islets into the kidney capsule of diabetic mice. J Vis Exp. 2007:e404.
11. Berman DM, Molano RD, Fotino C, et al. Bioengineering the endocrine pancreas: intraomental islet transplantation within a biologic resorbable scaffold. Diabetes. 2016;65:1350–1361.
12. Pepper AR, Gala-Lopez B, Pawlick R, et al. A prevascularized subcutaneous device-less site for islet and cellular transplantation. Nat Biotechnol. 2015;33:518–523.
13. Patikova A, Vojtiskova A, Fabryova E, et al. The optimal maturation of subcutaneous pouch can improve pancreatic islets engraftment in rat model. Transplantation. 2022;106:531–542.
14. Mojibian M, Harder B, Hurlburt A, et al. Implanted islets in the anterior chamber of the eye are prone to autoimmune attack in a mouse model of diabetes. Diabetologia. 2013;56:2213–2221.
15. Brubaker CE, Kissler H, Wang LJ, et al. Biological performance of mussel-inspired adhesive in extrahepatic islet transplantation. Biomaterials. 2010;31:420–427.
16. Fujita I, Utoh R, Yamamoto M, et al. The liver surface as a favorable site for islet cell sheet transplantation in type 1 diabetes model mice. Regen Ther. 2018;8:65–72.
17. Lee YN, Yi HJ, Kim YH, et al. Evaluation of multi-layered pancreatic islets and adipose-derived stem cell sheets transplanted on various sites for diabetes treatment. Cells. 2020;9:1999.
18. Cui W, Khan KM, Ma X, et al. Human amniotic epithelial cells and human amniotic membrane as a vehicle for islet cell transplantation. Transplant Proc. 2020;52:982–986.
19. Inagaki A, Imura T, Nakamura Y, et al. The liver surface is an attractive transplant site for pancreatic islet transplantation. J Clin Med. 2021;10:724.
20. Yang Z, Li X, Zhang C, et al. Amniotic membrane extract protects islets from serum-deprivation induced impairments and improves islet transplantation outcome. Front Endocrinol (Lausanne). 2020;11:587450.
21. Saghizadeh M, Winkler MA, Kramerov AA, et al. A simple alkaline method for decellularizing human amniotic membrane for cell culture. PLoS One. 2013;8:e79632.
22. Gholipourmalekabadi M, Mozafari M, Salehi M, et al. Development of a cost-effective and simple protocol for decellularization and preservation of human amniotic membrane as a soft tissue replacement and delivery system for bone marrow stromal cells. Adv Healthc Mater. 2015;4:918–926.
23. Li F, Lv Y, Li X, et al. Comparative study of two different islet transplantation sites in mice: hepatic sinus tract vs splenic parenchyma. Cell Transplant. 2020;29:963689720943576.
24. Wang D, Chen F, Fang B, et al. MiR-128-3p alleviates spinal cord ischemia/reperfusion injury associated neuroinflammation and cellular apoptosis via SP1 suppression in rat. Front Neurosci. 2020;14:609613.
25. Li F, Jiao A, Li X, et al. Survival and metabolic function of syngeneic mouse islet grafts transplanted into the hepatic sinus tract. Transplantation. 2018;102:1850–1856.
26. Espes D, Lau J, Quach M, et al. Rapid restoration of vascularity and oxygenation in mouse and human islets transplanted to omentum may contribute to their superior function compared to intraportally transplanted islets. Am J Transplant. 2016;16:3246–3254.
27. Bellin MD, Dunn TB. Transplant strategies for type 1 diabetes: whole pancreas, islet and porcine beta cell therapies. Diabetologia. 2020;63:2049–2056.
28. Aamodt KI, Powers AC. Signals in the pancreatic islet microenvironment influence β-cell proliferation. Diabetes Obes Metab. 2017;19(Suppl 1):124–136.
29. Bowers DT, Song W, Wang LH, et al. Engineering the vasculature for islet transplantation. Acta Biomater. 2019;95:131–151.
30. Chen H, Jia P, Kang H, et al. Upregulating Hif-1α by hydrogel nanofibrous scaffolds for rapidly recruiting angiogenesis relative cells in diabetic wound. Adv Healthc Mater. 2016;5:907–918.
31. Chittezhath M, Gunaseelan D, Zheng X, et al. Islet macrophages are associated with islet vascular remodeling and compensatory hyperinsulinemia during diabetes. Am J Physiol Endocrinol Metab. 2019;317:E1108–E1120.
32. Staels W, Heremans Y, Heimberg H, et al. VEGF-A and blood vessels: a beta cell perspective. Diabetologia. 2019;62:1961–1968.
33. Stokes RA, Simond DM, Burns H, et al. Transplantation sites for porcine islets. Diabetologia. 2017;60:1972–1976.
34. Korbutt GS, Pepper AR. Vascularized stem cell-derived β-cell spheroids: a “single step” in the right direction for the treatment of type 1 diabetes. Transplantation. 2022;106:12–13.
35. Krentz NAJ, Shea LD, Huising MO, et al. Restoring normal islet mass and function in type 1 diabetes through regenerative medicine and tissue engineering. Lancet Diabetes Endocrinol. 2021;9:708–724.
36. Tomei AA. Engineering confined and prevascularized sites for islet transplantation. Transplantation. 2018;102:1793–1794.
37. Kharbikar BN, Chendke GS, Desai TA. Modulating the foreign body response of implants for diabetes treatment. Adv Drug Deliv Rev. 2021;174:87–113.
38. Skrzypek K, Nibbelink MG, Karbaat LP, et al. An important step towards a prevascularized islet macroencapsulation device-effect of micropatterned membranes on development of endothelial cell network. J Mater Sci Mater Med. 2018;29:91.
39. Scheiner KC, Coulter F, Maas-Bakker RF, et al. Vascular endothelial growth factor-releasing microspheres based on poly(ε-caprolactone-PEG-ε-caprolactone)-b-Poly(L-lactide) multiblock copolymers incorporated in a three-dimensional printed poly(dimethylsiloxane) cell macroencapsulation device. J Pharm Sci. 2020;109:863–870.
40. Leal-Marin S, Kern T, Hofmann N, et al. Human amniotic membrane: a review on tissue engineering, application, and storage. J Biomed Mater Res B Appl Biomater. 2021;109:1198–1215.
41. Uematsu SS, Inagaki A, Nakamura Y, et al. The optimization of the prevascularization procedures for improving subcutaneous islet engraftment. Transplantation. 2018;102:387–395.
42. Sakurai T, Satake A, Nagata N, et al. The development of new immunoisolatory devices possessing the ability to induce neovascularization. Cell Transplant. 2003;12:527–535.
43. Fumimoto Y, Matsuyama A, Komoda H, et al. Creation of a rich subcutaneous vascular network with implanted adipose tissue-derived stromal cells and adipose tissue enhances subcutaneous grafting of islets in diabetic mice. Tissue Eng Part C Methods. 2009;15:437–444.
44. Pepper AR, Pawlick R, Bruni A, et al. Harnessing the foreign body reaction in marginal mass device-less subcutaneous islet transplantation in mice. Transplantation. 2016;100:1474–1479.

Supplemental Digital Content

Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.