High Vascular Density and Oxygenation of Pancreatic Islets Transplanted in Clusters into Striated Muscle

Abstract

Pancreatic islet transplantation is presently almost exclusively performed using the intraportal route for transplantation into the liver. However, islets at this site are poorly revascularized and, when also considering the poor long-term results of clinical islet transplantation, there has in recent years emerged an increased interest to evaluate alternative sites for islet transplantation. Striated muscle is easily accessible and has for decades been used for autotransplantation of parathyroid glands. Moreover, it is almost the only tissue in the adult where physiological angiogenesis occurs. The present study tested the hypothesis that striated muscle would provide good conditions for revascularization and oxygenation of transplanted islets. Because we previously have observed similar revascularization of islets implanted to the renal subcapsular site and intraportally into the liver, islets grafted to the kidney were for simplicity besides native islets used for comparison. Islets grafted into muscle were found to have three times more blood vessels than corresponding islets at the renal subcapsular site at 2 month follow-up, but still less vascular numbers than native islets. The oxygen tension in 2-month-old intramuscular islet grafts was sixfold higher than in corresponding renal subcapsular grafts, and 70% of that in native islets. However, the oxygenation of surrounding muscle was only 50% of that in renal cortex, and connective tissue constituted a larger proportion of the intramuscular than the renal subcapsular grafts, suggesting exaggerated early islet cell death at the former site. We conclude that the intramuscular site provides excellent conditions for vascular engraftment, but that interventions to improve early islet survival likely are needed before clinical application. Such could include bioengineered matrices that not only spatially disperse the islet, but also could provide local supply of oxygen carriers, growth and survival factors, strategies that are much more easily applied at the intramuscular than the intrahepatic site.

Introduction

During the last decade transplantation of pancreatic islets intraportally into the liver has become a treatment for selected patients with type 1 diabetes worldwide (9). However, several of the islets die early on due to islet blood interactions (18), innate immunity (8), and hypoxic events (10). Moreover, in contrast to whole-organ transplantation, there seems to be a progressive decline in function of such islet transplants, and very few patients remain insulin independent at 5 years posttransplantation (25). Because the histocompatibility barrier, the underlying autoimmune disease, and the immunosuppressive agents used are the same for whole-pancreas transplants, it is most likely that issues related to the isolation of islets and/or adaptation of the implanted islets to their new microenvironment play an important role in this context. A difference between the two modes of transplantation is that whole-pancreas transplants have an intact endogenous vascular system, which is directly anastomosed to that of the recipient. The islets, on the other hand, become disconnected from their vascular supply when they are isolated by collagenase digestion prior to transplantation.

Pancreatic islets normally have a very dense capillary network. Although revascularization is rapidly initiated following transplantation, similarly low numbers of blood vessels tend to form within mouse and human islets transplanted intraportally into the liver or beneath the renal capsule (13,16). This results in an impaired oxygenation of islets beneath the renal capsule, which seems even to be aggravated in intrahepatically transplanted islets (7,21). Experimentally, several other sites than the liver have been tested with variable success. Within the Nordic Network for Clinical Islet Transplantation particular interest has been focused on the intramuscular site, bearing in mind the feasibility of this site for autotransplantation of parathyroid glands and its easy access. In one patient in Stockholm, islets were due to severe hereditary chronic pancreatitis autotransplanted into muscle with high and stable c-peptide production for at least 2.5 years as a result (23). Not much is known with regard to engraftment and subsequent oxygenation of islets at this site. The present study tested the hypothesis that intramuscularly transplanted islets will become better engrafted and oxygenated than islets implanted to the renal subcapsular site.

Materials and Methods

Animals

The experiments were performed on inbred, male Wistar-Furth rats weighing ~325 g and purchased from B&K Universal (Sollentuna, Sweden). The animals had free access to tap water and standard rat chow (R3, Ewos, Södertälje, Sweden) throughout the study. All experiments were approved by the animal ethics committee for Uppsala University.

Islet Isolation, Culture, and Transplantation

Pancreatic islets were prepared by collagenase (Boehringer-Mannheim, Mannheim, Germany) digestion, as described elsewhere (1). Groups of ~150 islets were cultured free-floating for 4–7 days in RPMI-1640 medium supplemented with 10% (v/v) calf serum (Sigma-Aldrich, St. Louis, MO) (1), and the medium was changed every second day. At transplantation, ~250 islets were packed in a braking pipette and implanted beneath the renal capsule on the dorsal side of the left kidney, or packed and injected through a butterfly needle (25 gauge) superficially into the left iliopsoas muscle, in pentobarbital-anesthetized (60 mg/kg, IP; Apoteket, Umeå, Sweden) syngeneic rats.

Oxygen Tension Measurements

Two months posttransplantation, the animals were anesthetized with an IP injection of thiobutabarbital (120 mg/kg; Inactin^®, Research Biochemicals International, Natick, MA), placed on an operating table maintained at 37°C, and tracheostomized. Oxygen tension was measured in the native and transplanted islets with modified Clark-type microelectrodes (Unisense, Aarhus, Denmark), essentially as previously described (4,24). Islet grafts were exposed by a skin incision, and the tissue surface embedded in surroundings of cotton wool soaked in Ringer solution, and covered with mineral oil. In control animals (not transplanted), the pancreas was exposed by an abdominal midline incision and immobilized over a cylindric plastic block attached to the operating table and then superfused with mineral oil (Kebo Grave). In the transplanted islets and surrounding tissue ≥ 10 measurements of oxygen tension were performed in each animal. In the pancreas of control animals, measurements were performed in three to five superficial pancreatic islets and surrounding exocrine parenchyma. Multiple measurements were usually performed within the same islet; the mean was calculated to obtain the oxygen tension value for one islet. The mean of all measurements in each tissue and animal was calculated and considered to be one experiment. During the electrode measurements, blood pressure, body temperature, and tissue temperature (CT D85, Ellab, Copenhagen, Denmark) were continuously recorded with a MacLab Instrument (AD Instruments, Hastings, UK) connected to a Power Macintosh 6100 computer. At the end of the oxygen tension measurements, a blood sample was collected for analysis of hematocrit and blood gases in order to ascertain that animals fulfilled inclusion criteria (pH >7.30, pO₂ >10 kPa, pCO₂ <6.8, and hematocrit >40). During measurements the mean arterial blood pressure was kept within the range of 100–130 mmHg in all animals, and nonfasting blood glucose concentrations were ~5 mmol/L (test reagent strips; Freestyle lite).

Vascular Density

After electrode measurements, the graft-bearing organs or the pancreas of control animals were retrieved, fixed in formalin, and paraffin embedded. Sections (5 μm) of the islet grafts or control pancreata were prepared, and stained with an insulin antibody or the lectin Bandeiraea (Griffonia) simplicifolia (BS-1) to visualize β-cells and the islet blood vessels, respectively (16). In each animal, ≥ 10 tissue sections stained with BS-1 from the islet transplants or the control pancreas were randomly chosen and evaluated for islet vascular density. Connective tissue surrounded individual islets in the renal subcapsular grafts. The numbers of blood vessels in the endocrine and connective tissues were here counted separately. The respective fractions of islets and connective tissue in the islet grafts were determined by a direct point counting technique (29). The area of the investigated tissue was determined using a computerized system for morphometry (Scion Image, Scion, MD, USA). Vascular density [i.e., the number of blood vessels found per measured islet or graft area (mm²)] was then calculated.

Statistical Analysis

All values are given as mean ± SEM. When only two groups were compared was Student's unpaired t-test used. Multiple comparisons between data were performed by using ANOVA (Statview; Abacus Concepts, Berkeley, CA) and the Bonferroni post hoc test. For all comparisons, p < 0.05 was considered to be statistically significant.

Results

Vascular Density in Islets

Native pancreatic islets of nontransplanted control animals had a vascular density of ~700 capillaries/mm², whereas the vascular density in islets transplanted to the kidney or into muscle was much lower at the 2 month follow-up (Figs. 1 and 2). Nevertheless, islets grafted into muscle had three times more blood vessels than corresponding islets at the renal subcapsular site. The vascular densities of the connective tissue surrounding individual islets in all grafts were higher than in transplanted islets per se, and similar between renal subcapsular and intramuscular islet grafts (stroma vascular density; range 2000–4000 blood vessels/mm² in both renal subcapsular and intramuscular grafts). Connective tissue constituted a larger proportion of the intramuscular than the renal subcapsular grafts (22.0 ± 3.8% vs. 7.4 ± 0.8%, p < 0.05).

Figure 1.

Insulin and endothelium staining (Bandeiraea simplicifolia; red) of native rat islets (A, B), and 2-month-old renal subcapsular (C, D) or intramuscular islet grafts (D, E). Scale bars: 50 μm (A, B), and 100 μm (C–F).

Figure 2.

Vascular density in native rat islets, and 2-month-old renal subcapsular or intramuscular islet grafts. All values are given as mean ± SEM for 4–7 experiments. ^∗p < 0.05 when compared to native islets and ^#p < 0.05 when compared to renal subcapsular islet grafts.

Islet Oxygen Tension

The oxygen tension in native islets was ~40 mmHg, and lower in both renal subcapsular and intramuscular islet grafts (Fig. 3). However, the oxygen tension was sixfold higher in the intramuscular than the renal subcapsular islet grafts at 2 months posttransplantation. The implantation organs differed in their oxygenation with approximately doubled oxygen tension in the renal cortex compared to muscle (32.7 ± 3.5% vs. 15.4 ± 3.1%, p < 0.05).

Figure 3.

Oxygen tension in native rat islets, and 2-month-old renal subcapsular or intramuscular islet grafts. All values are given as mean ± SEM for 4–7 experiments. ^∗p < 0.05 when compared to native islets and ^#p < 0.05 when compared to renal subcapsular islet grafts.

Discussion

Clinical islet transplantations have been almost exclusively performed through the intraportal route. However, despite advances in immunosuppressive regimens at least two donor pancreases are needed to reverse type 1 diabetes, which is far more than the alleged 10–20% of the total islet volume suggested to be enough to maintain normoglycemia in humans (26). Indeed, the functional capacity of the transplanted islets has been shown to only correspond to about 20% of that found in a nondiabetic person (25). The concept of pancreatic islet transplantation is also severely hampered by the gradually declining graft function (25,27). In order to improve the result of clinical islet transplantation, new strategies are clearly needed (28).

We have previously observed that pancreatic islets transplanted to the liver or beneath the renal capsule are poorly revascularized, and that this is associated with impaired blood perfusion, low oxygenation, perturbed metabolism, and dedifferentiation of β-cells in the transplanted islets (5,7,14,17). Interestingly, in the adult, angiogenesis is normally induced only in ovaries (during ovarian cycling), in the placenta (during placental development), and in muscles (during exercise in response to hypoxia) (22). Thus, a clear difference exists between striated muscle and other referred implantation sites. In recent experiments in mice using intravital microscopy (Christoffersson et al., unpublished observation), we have observed that single islets implanted to cremaster muscle rapidly restore a functional islet vascular network, which seems to have a similar B-A-D cell perfusion order as previously described to be predominant for native islets (3,19).

In the present work, we implanted clusters of islets in order to more resemble the clinical situation. Although the vascular network was not fully restored when compared to native islets, much more intraislet blood vessels were found in the intramuscularly transplanted islets than in islets implanted beneath the renal capsule. Moreover, the oxygenation of the 2-month-old intramuscularly transplanted islets was sixfold higher than of the renal subcapsular islet grafts and 70% of that of native islets. The discrepancy between oxygenation and revascularization may reflect that the correlation between vascular numbers and tissue oxygenation is not directly linear, and that a smaller increase in capillary density may cause a much better improvement in oxygen tension. We have earlier observed such effects when using different interventions to improve islet graft revascularization [e.g. (11,20)]. Higher blood perfusion of individual blood vessels in the intramuscular islet grafts could, if present, also contribute to a better oxygen tension [cf. (12)], as well as possibly diffusion of oxygen from the surroundings when considering the highly vascularized rich stroma compartment of the grafts.

For measurements of oxygen tension in islet grafts, we used our previously described technique (4). All electrodes inserted into tissues may to some extent affect the tissue, although the small diameter (tip diameter 2–4 μm) of the Clark microelectrodes used minimizes tissue distortion. The recordings can be expected to mainly report extracellular interstitial oxygen tension values, because any injured cells rupture and form part of the extracellular compartment. No profound effects on tissue physiology and oxygen consumption seem, however, to occur, because stable long-term measurements of oxygen tension are achieved, and even the small oscillations of metabolism in pancreatic islets synchronized with oscillations in insulin release can readily be measured (2). We have also earlier repeatedly performed blood flow measurements before and after multiple islet oxygen tension measurements in islet grafts and found no alterations in the microcirculation of the tissue by the electrode recordings (4,6).

Intramuscularly transplanted islets were interspersed by more connective tissue than islets transplanted to the kidney, which is likely to reflect increased early cell death with concomitant fibrosis. Immediately after transplantation islets are supplied by oxygen solely by diffusion from the surrounding tissues. Hypoxic death could be expected to be more pronounced at the intramuscular site before islet graft revascularization commences, when considering the much lower oxygen tension in muscle than renal cortex. Increased graft fibrosis and a need for higher numbers of islets to cure diabetic recipients at the intramuscular site have also been described recently in rat when implanting islets in clusters, but decreased fibrosis and improved results were observed by dispersing the islets and attempting more of a “pearls on a string” fashion for islet transplantation (15). Because graft function at this site mainly seems to be restricted by early islet cell death, we chose in the present work to use normoglycemic rats as recipients for our studies of vascular engraftment. Interestingly, means to improve early islet survival by better implantation techniques (e.g., in a “pearls on a string” fashion) by bioengineered matrices to spatially disperse the islets and for local supply of oxygen carriers, growth and survival factors, could easily be applied at the intramuscular in contrast to the intrahepatic site. When considering the much better vascular engraftment of islets at the intramuscular than the renal subcapsular or previously investigated hepatic site (7,13,16), muscle as implantation site could provide advantages for long-term function in clinical islet transplantation when combined with interventions to support early survival.