|Home | About | Journals | Submit | Contact Us | Français|
Embolic occlusion of the portal vein due to islet transplantation is one of the major reasons for reduced survival of transplanted islets. In this study, we examined the location of islets as well as the correlation between islet and portal vein size after intraportal islet transplantation and evaluated liver and islet pathology.
The liver was divided into peripheral and central sites. Islet and liver apoptosis/necrosis were significantly higher at peripheral sites. In regions without liver apoptosis or necrosis, portal vein diameter was significantly larger and embolic ratios were significantly lower.
BALB/c mice were intraportally transplanted with 800 islets and the liver was examined at postoperative day (POD) 0 (n = 7), POD 2 (n = 4) and POD 28 (n = 3). Liver specimens were stained for hematoxylin and eosin (necrosis), insulin and TUNEL (apoptosis). We evaluated distance from liver surface to islets, islet and portal vein diameter, embolic ratio (islet diameter/portal vein diameter), apoptosis/necrosis of islets and apoptosis/necrosis of the liver tissue surrounding the islet.
Transplanted islets and liver tissue exhibited more injury at peripheral sites, in part, due to smaller diameters of portal venules that result in more frequent emboli following islet transplantation.
Though islet transplantation is considered a useful treatment for type 1 diabetes mellitus (T1DM), there are a number of issues that need to be overcome prior to routine clinical application. One such issue is loss of transplanted islets due to a variety of reasons, including isolation procedures, immunosuppression toxicity, instant blood mediated inflammatory reaction, inflammation and rejection.1,2 Over 50% of transplanted islets are estimated to fail engraftment.1,3,4 Recently, it was reported that embolization of the portal vein after islet transplantation is likely to be one of the major components for impairment of engraftment.4 The underlying mechanism(s) is thought to be that emboli occur due to islet transplantation within the portal vein, inducing ischemia within the surrounding liver tissue and inflammatory cytokines that ultimately lead to islet apoptosis, necrosis and graft loss.4–6
We have previously assessed the location of ischemia after islet transplantation using magnetic resonance imaging (MRI).6–8 MRI findings revealed that: (1) islets were dispersed throughout the whole liver in our and other MRI studies6,9,10 and (2) ischemia and necrosis due to islet transplantation early after transplantation was observed primarily at the peripheral surface of the liver.4,6 These reports suggest that ischemic/necrotic changes appear to be more pronounced in liver. We now report transplanted islet location within the liver in addition to portal vein diameter and how these factors relate to islet and liver tissue apoptosis and necrosis. Our results may guide future clinical intra-portal islet transplantation and improve animal studies of the underlying mechanisms involved in transplanted islet loss.
Blood glucose and serum insulin levels improved in all mice after islet transplantation. Three mice that survived to the final time point (POD 28), achieved euglycemia and serum insulin levels were improved in comparison with pretransplantation (Fig. 2). In other words, islet transplantation was effective in all transplanted diabetic mice.
The median of the distance from liver periphery to the islets (indicated as “distance to islet”) (Fig. 1C) at POD 0 was 484.6 µm (n = 284 islets). Therefore, we divided the liver into peripheral and central sites by demarcating a line at 484.6 µm from the liver surface. Islet distribution (peripheral vs. central) was not significantly different at any of the time points, but POD 28 trended towards significance (p = 0.08) (Table 1). Comparison between peripheral and central sites revealed that islet apoptosis and liver apoptosis/necrosis were significantly increased in peripheral sites (Fig. 3A–C) and islet necrosis tended to be enhanced at the peripheral site (Fig. 3D).
One of the reasons for differences between peripheral and central sites could be due to portal vein and islet diameters. Figure 3 showed the correlation among distance to islet, islet diameter, portal vein diameter and embolic ratio. Distance to islet was significantly correlated with islet diameter, portal vein diameter and embolic ratio (Fig. 4A–C). Portal vein diameter significantly correlated with islet diameter (Fig. 4D) and embolic ratio (Fig. 4E). In addition to the known fact that portal vein diameter is smaller at peripheral sites and larger at central sites, these findings suggest that: (1) most islets appeared to be lodged in portal venules by approximately the same size (total portal vein obstruction (100%) was detected in 44.3% of islets and 85.2% of islets caused over 90% of portal vein obstruction) and (2) embolic ratio was higher at the peripheral and lower at the central sites which is likely why total venous obstruction tended to occur primarily at the peripheral sites.
A significant amount of liver tissue surrounding transplanted islets was apoptotic (93.1%: 107/115 islets). Evaluating a number of parameters demonstrated that distance to islet and islet diameter decreased liver apoptosis (Fig. 5A and B). Portal vein diameter was significantly smaller (Fig. 5C) and embolic ratio was significantly higher (Fig. 5D) when apoptosis was present. Similar to liver apoptosis, liver necrosis was observed at shorter distances to transplanted islets (Fig. 6A), smaller portal vein diameter (Fig. 6C) and higher embolic ratio (Fig. 6D). There was no significant difference in islet diameters (Fig. 6B).
While there was a good correlation between liver injury (apoptosis/necrosis) and various factors, we found no significant correlation for islet injury. There was only a significant correlation between distance to islet and islet apoptosis (R2 = 0.05, p < 0.01). However, apoptotic islets were significantly increased at POD 2 (68.7 ± 3.8%, n = 115 islets) in comparison with islets at POD 0 (6.4 ± 1.0%, p < 0.0001, n = 284). We did not observe any apoptotic cells at POD 28.
Currently the liver is the most successful organ used for clinical islet transplantation but many islets fail to engraft.1,3,4 Thus, a significant number of transplanted islets is required (over 10,000 IEQ) in order to improve the diabetic condition.3,11 To overcome loss of islet engraftment, many clinical and experimental trials have been performed.11–14 Recently, portal vein obstruction due to islet transplantation itself has been considered one of the primary reasons for islet loss.4,5 The mechanism of graft loss suggests that islets become lodged in the portal vein, block blood flow and subsequently induce local ischemia of adjacent liver tissue. The ischemic liver tissue induces apoptosis and necrosis further damaging the transplanted islets.5 We and other groups have examined transplanted islets using MRI6,9,10 and we could clearly identify that the transplanted islets were dispersed throughout the whole liver. In spite of these MRI findings, liver necrosis at the early stages of transplantation appeared to be limited to peripheral sites.6 Based on these findings, we hypothesized that peripheral liver sites suffer from ischemia due to transplanted islets as 100% of emboli occur in this region. To validate our hypothesis, we examined the degree and the distribution of embolization in the liver histologically.
Firstly, we examined the distribution of islets at POD 0 and found that they were dispersed throughout the whole the liver similar to our MRI findings (preliminary data). There were no variances between peripheral and central regions of the liver (Table 1). It is assumed that most of the islets are located in portal venules of approximately the same diameter since islet diameter significantly correlated with portal vein diameter (Fig. 4D) and most of the embolic ratios were over 90%. In other words, islet location appeared to be decided by portal vein diameter. Ischemic changes of the liver at POD 2 were detected in all recipients (data not shown) and over 90% of the liver tissue surrounding islets had apoptosis and necrosis though this was more severe at peripheral sites (Fig. 3A and B). Portal vein diameter was smaller and embolic ratios were also higher at peripheral sites (Fig. 4B and C). Therefore, islets could easily cause portal vein emboli leading to local tissue ischemia that then induces apoptosis and necrosis (Fig. 5C and D; Fig. 6C and D).
In one of the four mice sacrificed at POD 2 and one of the three mice sacrificed at POD 28 there was an immediate decrease of blood glucose level after transplantation. Then a rapid increase in blood glucose level right before liver recovery was detected in two of the four animals sacrificed at POD 2. The rapid decrease and increase of blood glucose level can be caused by the destruction of transplanted islets. Destruction of islets causes a temporary increase in serum insulin and results in a rapid decrease of blood glucose. Our preliminary data revealed that serum insulin was significantly increased at 15 minutes after islet transplantation and remained at an elevated level for the first few days due to islet hypoxia, unspecific inflammation or instant blood-mediated inflammatory reaction (data not shown). After the insulin spike subsides, blood glucose levels increase again. We believe the rapid decrease and increase in blood glucose level is influenced by the degree of islet destruction.
While liver injury was significantly reflected by these factors, correlation between islet damage and the various factors (distance to islet, islet diameter, portal vein diameter and embolic ratio) were not significant (data not shown). Islet obstruction in the portal vein may be one of the reasons for islet damage but other factors may also play a role. These include local inflammation and instant blood mediated reaction that can influence islet survival.
To prevent islet emboli, it may be useful to use smaller islets for transplantation.15 Larger diameter portal vein and lower embolic ratio are the main factors that could prevent liver damage (Fig. 5C and D; Fig. 6C and D) and so a decreased embolic ratio could be achieved by using smaller islets. There are many other merits to using smaller islets (<100 µm): (1) smaller islets function better as measured by insulin release,16 (2) smaller islets tend to have better survival from decreased hypoxia,16 (3) smaller islet have higher vascular density15,17 and (4) smaller islets have better survival after cryopreservation.17,18 The evidence shows that using smaller islets can contribute to improving the outcome of islet transplantation. In other words, islet size can be one of the critical factors in deciding the success or failure of islet transplantation. Recently, Cavallari et al. described a method for managing the size of isolated islets.19 They showed that dispersed single islet cells gathered and became cell clusters called pseudoislets with a one week culture. The size of pseudoislets was approximately 100 µm and with small variances in size. With this technique, larger islets can be used effectively as smaller pseudoislets for transplantation.
In summary, we found that the distribution of transplanted islets in the liver appears to change over time. Transplanted islets and the liver tissue in peripheral sites were more damaged as portal vein diameter was smaller in this region and islet could readily embolize following transplantation. An important step towards clinical improvements of transplanted islets is to understand the post-transplant condition of the liver after islet transplantation.
BALB/c female mice weighing 22–27 g (Charles River Laboratories. Inc., Boston, MA) were used as both donors and recipients. The mice were housed under pathogen-free conditions with a 12-hour light cycle and free access to food and water.All animal care and treatment procedures were in accordance with institutional regulations and the Institutional Animal Care Use.
Streptozotocin (STZ, 200 mg/kg/mouse, Sigma-Aldrich, St. Louis, MO) was injected intraperitoneally. Blood glucose levels were measured by Accuchek Advantage glucose monitors (Roche, Indianapolis, IN) and diabetes was diagnosed when the blood glucose levels were greater than 250 mg/dL.
Murine islets were isolated by collagenase (collagenase V, Sigma-Aldrich) digestion, separation by Ficoll (Sigma-Aldrich) discontinuous gradients and purification. The isolated islets were cultured in M199 with 11 mmol/L and 10% fetal bovine serum at 37°C in 5% CO2/95% air overnight.20–22 Following culture, we performed syngeneic islet intraportal transplantation into diabetic mice.6 Fourteen mice were transplanted with 800 islet equivalents (IEQ = 150 µm) per recipient. IEQ’s were calculated by measuring islet size via a microscope. Liver recoveries were performed at three time points; (1) day of transplantation (POD 0: within 1 hour after transplantation, n = 7), (2) early after transplantation POD 2 (n = 4) and (3) a late time point, POD 28 (n = 3). Tissue was placed into 10% formalin and immersed for 24 hours.
Transplanted mice at POD 2 (n = 4) and POD 28 (n = 3) had blood glucose and serum insulin measurements performed (serum insulin measurement was performed on three of the four at POD 2 and all mice at POD 28). Blood glucose levels were measured by Accu-chek Advantage glucose monitors while serum insulin was measured with a rat/mouse insulin enzyme-linked immnosorbent assay kit (Linco, MO).
Harvested livers were placed into 10% formalin immediately after exicision and immersed for 24 hours. Liver specimens were then embedded in paraffin and sections were cut at 5 µm. Sections were stained for: (1) Hematoxylin and eosin (H andE) for liver necrosis. (2) Insulin immunohistochemistry to identify islets. Primary antibodies were guinea pig anti-insulin antibody (Dako, Carpunteria, CA) diluted 1:100. After incubating with biotinylated secondary IgG antibody (Vector Laboratories, Burlingame, CA, and Santa Cruz Inc.), the specimens were colored with AEC+ (Dako) and counterstained with hematoxylin.6,7 (3) Apoptosis was detected by the TdT-mediated dUTP-biotin nick end labeling (TUNEL) method using an in situ apoptosis detection kit (Promega, Madison, WI, USA). Sections were treated with proteinase K (Dako) and incubated with TdT enzyme for 60 min at 37°C. After washing in PBS, the sections were further incubated with streptavidin horseradish peroxidase (HRP) solution and visualized with peroxidase substrate solution containing 3,3’-diamin-obenzidine DAB.6
Islets and surrounding liver tissue were assessed for necrosis (defined as destruction of cell structure with granulation, H&E: Fig. 1A and C) and apoptosis (TUNEL: Fig. 1B and D). The same islet was used for Figure 1C–E. Apoptosis of transplanted islets was expressed as the percentage of TUNEL positive islets relative to total islet cells [(TUNEL positive cells)/(Total islet cells) × 100 (%)].6 Necrosis of islets and apoptosis/necrosis of surrounding liver tissue were scored either as absent (zero) or present (one).6 We measured, (1) the distance from the periphery of the liver to the islet mass (i.e., distance to islet), (2) the transplanted islet and (3) the portal vein diameter at the level of the islet (Fig. 1F) using image analysis software (Image J ver. 1.40, NIH). The ratio of islet diameter divided by portal vein diameter [(islet size)/(portal vein diameter) × 100 (%)], was abbreviated as the “embolic ratio.”
Transplanted islets at POD 0 were divided into two groups by the median of the distance to islet (peripheral and central sites) and we compared the two groups in islet distribution at each time point. Assessment of the distribution at each time point was performed by chi square test. We also compared both groups for liver and islet apoptosis and necrosis at POD 2. The statistical assessment was performed using a Student t test. To evaluate the correlation among the four factors (distance to islet, islet diameter, portal vein diameter and embolic ratio) that might influence islet/liver damage, we performed regression analysis. The degree of correlation was decided by the correlation coefficient (R value): 0–0.2 (0–0.04 as R2 value) was defined as no correlation, 0.2–0.4 (0.04–0.16 as R2 value) as a weak correlation and 0.4–1.0 (0.16–1.0 as R2 value) as a strong correlation. Finally, we compared islet/liver apoptosis/necrosis with islet diameter, portal vein diameter and embolic ratio. Statistical analysis was Student t test. Data were expressed as mean ± standard error of the mean (SEM). All the differences were considered significant at p < 0.05.
This work was supported by NIH/NIDDK Grant #1R01-DK077541 (EH) and a research fellowship of the Uehara Memorial Foundation (NS). We are very appreciative of the microsurgical technical support of John Chrisler and Loma Linda University Microsurgery Laboratory, and the kind help in specimen processing by John Hough.