SEARCH

SEARCH BY CITATION

Keywords:

  • Porcine islets;
  • preclinical studies;
  • revascularization;
  • suppress engraftment

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

The liver may not be an optimal site for islet transplantation due to obstacles by an instant blood-mediated inflammatory response (IBMIR), and low revascularization of transplanted islets. Therefore, intramuscular islet transplantation (IMIT) offers an attractive alternative, based on its simplicity, enabling easier access for noninvasive graft imaging and cell explantation. In this study, we explored the outcome of autologous IMIT in the minipig (n = 30). Using the intramuscular injection technique, we demonstrated by direct histological evidence the rapid revascularization of islets autotransplanted into the gracilius muscle. Islet survival assessment was performed using immunohistochemistry staining for insulin and glucagon up to a period of 6 months. Furthermore, we showed the crucial role of minimizing mechanical trauma to the myofibers and limiting exocrine contamination. Intramuscular islet graft function after transplantation was confirmed by documenting the acute insulin response to intravenous glucose in 5/11 pancreatectomized animals. Graft function after IMIT remained however significantly lower than the function measured in 12 out of 18 minipigs who received a similar islet volume in the liver through intraportal infusion. Collectively, these results demonstrated in a clinically relevant preclinical model, suggest IMIT as a promising alternative to intraportal infusion for the transplantation of β cells in certain medical situations.


Abbreviations
IBMIR

instant blood-mediated inflammatory response

IMIT

intramuscular islet transplantation

IT

islet transplantation

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

The field of β cell replacement therapies has progressed extensively over the last decades. It is well established that successful intraportal islet transplantation (IT) can restore endogenous β cell function to subjects with type 1 diabetes mellitus [1]. In fact, when the graft function is optimal, insulin independence can be consistently prolonged for up to 5 years in 50% of patients [2]. Several factors influence the outcome and performance of the graft upon implantation. For instance, preclinical studies have confirmed the significant differences in utilizing several sites for the implantation of islet grafts [3], but the most utilized clinical approach is embolization into the liver [1]. However, it has become evidently clear that the liver may not be the optimal environment as a recipient site for pancreatic islets, owing not only to immunologic [4], but also to anatomic [5, 6] and physiologic factors that may promote a decline in islet function [7]. Moreover, intrahepatic islet infusion is often associated with an immediate blood- mediated inflammatory reaction (IBMIR), thrombosis and hepatic tissue ischemia with elevated blood liver enzymes [8-10]. In addition, the cross-talk between activated coagulation and inflammatory mediators after implantation, dramatically affects islet cell survival and engraftment, resulting in β cell dysfunction or death, depicting primary nonfunction as a consequence of reduced functional islet mass [11]. This intrahepatic environment appears to potently impair the metabolic functions of transplanted islets [12]. Furthermore, the complications associated with graft recovery within the hepatic site, will further limit its potential applications in exploiting insulin-secreting cells obtained from alternative cell sources. These include xenogenic islets, immortalized β cell lines, embryonic stem cells, or adult progenitor cells, including β cell encapsulation.

Restoration of β cell function is a highly desirable goal for patients with unstable diabetes; therefore, the search for an alternative site that is safer for islet transplantation is imperative [3, 5].

In man, autotransplantation of minced tissue into striated muscle following blunt dissection has been successfully used in parathyroid surgery for several decades [13]. Initially demonstrated in rodents in the early 1980s [14], intramuscular islet transplantation (IMIT) has rarely been considered as a clinically feasible implantation site [15].

Furthermore, the optimal technique for intramuscular transplantation remains unclear. The recent report demonstrating a successful case of intramuscular islet autotransplantation after subtotal pancreatectomy [16] stimulated us to explore this site in a preclinical model. Among various potential sites, the muscle offers attractive prospects based on its surgical simplicity, easy access to the graft for noninvasive imaging techniques, as well as potential cell explantation [17].

In this study, using the minipig, we provide direct evidence of the survival, function and revascularization of autologous islets transplanted in the muscle. We also showed the necessity of minimizing trauma to the myofibers, and limiting exocrine contamination for opimizing islet survival. Finally we compared the function of autotransplanted islets following intramuscular transplantation to those obtained by standard intraportal infusion.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Animals

Nonsyngenic adult minipig (31 ± 2 kg; at least 2 years old; Denis’ breeding, Templeuve, France) were autotransplanted with various techniques. All animals, treated in accordance with French regulations, and our institutional ethical committee, were housed at our institution in the University Hospital Department for Experimental Research. A total of 30 minipigs received intramuscular islet auto transplantation for the purpose of this study. In 26 cases, initial surgery consisted of a left hemipancreatectomy. A total pancreatectomy was performed to explore graft function, initially in four animals, or secondarily, before graft explantation and sacrifice, in seven animals. Islet graft function was similarly assessed in 18 minipigs who received total pancreatectomy and intraportal islet infusion in the context of a previous study [18].

Pancreatectomy

All surgical procedures were performed under general anesthesia induced following administration of a 10 mg subcutaneous injection of nalbuphine (Nalbuphine®, Merck, Paris, France), an intramuscular injection of 0.3 mg/kg of midazolam (Hypnovel®; Roche, Neuilly-sur-Seine, France) and 5 mg/kg of ketamine (Ketamine Virbac®; Virbac, Carros, France). Tracheal intubation was performed subsequently with 0.2 mL/kg/h of intravenous propofol (Diprivan 1%®; AstraZeneca, Rueil-Malmaison, France). Twenty-four hours before surgery, two fentanyl patches with release rates of 25 and 50 μg/h (Durogesic; Janssen-Cilag, Issy-les-Moulineaux, France) were placed on the skin. In all cases, left pancreatectomy was performed with cautious dissection to maintain normal vascularization and to minimize warm ischemia until final excision of the gland. When the head of the pancreas was excised, the vascularization of the duodenum was carefully preserved as previously described [19]. The explanted pancreas was immediately cooled in a Hank's solution (Sigma, Saint-Quentin Fallavier, France) at 4°C. Minipigs with total pancreatectomy initially received daily injections of long acting porcine insulin, and adjusted to daily capillary glucose assessment. Minipigs with a residual pancreatic head did not require insulin injections to maintain normoglycemia.

Islet isolation

Within 30 min of cold ischemia, the pancreas was distended by infusion of 250 mg of porcine collagenase (Liberase PI®; Roche, Meylan, France) using an 18-G catheter (Optiva®; Ethicon, Pomezia, Italy) inserted in the pancreatic duct. Islets were isolated using state-of-the-art techniques as previously described [19]. Following Dithizone® staining (diphenylthiocarbazone; Sigma, Saint-Quentin Fallavier, France), islet mass was assessed by the same two investigators throughout the study and expressed in terms of the islet equivalent count per kilogram (IEQ/kg), the number of islets normalized to a diameter of 150 μm. Part of the islet preparation was purified using a discontinuous gradient of density (Histopaque® 1083; 1077; 1067 density; Sigma, Saint-Quentin Fallavier, France). The digested pancreatic tissue and the purified islets were resuspended in Hank's solution and supplemented with porcine serum albumin (Sigma, Saint-Quentin Fallavier, France) and transported at room temperature to the operating theatre.

Islet transplantation

Autologous intramuscular islet transplantation was performed in all cases in the gracilius muscle following exposition of the muscular aponevrosis through a skin incision. Each minipig received at least two and up to eight islet grafts in distinct intramuscular sites. The packed cell volume delivered in each site varied from 0.1 mL to 0.5 mL, depending on purity of the preparation. Two techniques of intramuscular islet implantation were compared. The muscular pouch technique was derived from the standard approach currently used for clinical parathyroid transplantation [13]. A 1 cm aponevrotomy was performed followed by blunt dissection of the myofibers to form a 1 ml pouch, in which islets could be spread out under direct vision. The aponevrosis was then hermetically closed with standard nonabsorbable sutures. Alternatively an injection technique was performed to minimize muscular trauma as recently suggested by Lund et al. in the rodent [20]. A 14-G catheter Optiva® was inserted in the muscle through the aponevrosis and slowly retracted while carefully injecting the islets to ensure their homogenous distribution along the 5 cm needle track. The implantation site was systematically marked with metal clips for subsequent identification at the time of explantation.

In the control group, hepatic islet transplantation was performed as previously described, by intraportal infusion through a catheter placed in a splenic branch at the time of pancreatectomy and exteriorized in the flank [19].

Graft survival

The presence of insulin positive cells was assessed by immunohistochemistry. The cell-transplanted sites were explanted at various times (3–180 days) after transplantation. Muscle biopsies were fixed (4% paraformaldehyde <18 h), embedded in paraffin and sectioned (7 μm width). One section every 280 μm in size was stained with Masson's trichrome, utilizing the light microscopic to determine cell localization. Subsequently, slides were prepared for immunostaining and immunofluorescence. After deparaffinization, and rehydratation, sections were subject to microwave (citrate buffer) or proteinase K (Dako France, Trappes, France) pretreatment for antigen recovery. Nonspecific binding was reduced with protein block (Dako France, Trappes, France). Sections were incubated with primary antibodies described in supplementary table 1, followed by biotinylated secondary antibodies (KPL Laboratory, les Ulis, France), and streptavidin-HRP antibody (45 min) followed by 3,3 DAB (Dako France, Trappes, France), or streptavidin Alexafluor 594 or 488 nm (1:400, Invitrogen–Molecular Probes; Eugene, OR, USA). Sections were counterstained with hematoxylin staining or Dapi. An Olympus light microscope was used to evaluate each tissue section at ×200 magnification. For each graft, at least three sections were selected with Masson's trichrome and examined with antiinsulin immunostaining to select the section that had the greatest number of positive stained cells. In each selected section, insulin positive area was calculated using Leica's QWIN imaging program (Leica Microsystems Imaging Solutions®, Cambridge, UK).

Graft revascularization

Islet revascularization was assessed by immunochemistry using anti-vWF immunostaining as previously described. When possible, vascular density was calculated using Leica's QWIN imaging program. The vascular density of an islet was defined as the immunostained vessel surface divided by the islet area.

Tissue hypoxia

Pimonidazole forms irreversible adducts with intracellular macromolecules under hypoxic conditions [21]. This technique identifies in vivo tissue with pO2 levels <10 mmHg [22]. One hour before tissue harvesting, animals (n = 3) were injected intravenously with 200 mg of pimonidazole (Hypoxyprobe-1 Omni Kit®; Burlington, MA, USA). Muscle samples were immediately fixed with 4% of paraformaldehyde and then processed in paraffin blocks. Immunostaining was performed with antipimonidazole primary antibody (Hypoxyprobe-1 Omni Kit®, polyclonal, 1/200) using the protocol described above. As a positive control, we used ischemic 3T3 mouse fibroblasts (gazed 1 h with 95% CO2) in the presence of 100 μg/mL of pimonidazole.

Graft function

Graft function was assessed after 1 month in 11 of the total pancreatectomized pigs prior to graft explantation at 4 weeks. A single-lumen radiopaque silicone catheter (Hickman®; Bard, Trappes, France) was initially placed in a jugular vein, and exteriorized on the neck to allow repeated blood samples. Acute insulin response was estimated prior to and after graft explantation, as previously described [13] during 10 min after intravenous injection of glucose (0.5 g per kg, Aguettant, Lyon, France). The first test was performed in pancreatectomized recipient immediately prior to graft explantation. A second test was performed 30 min after graft removal (negative control). The same test was also performed in 10 healthy control animals without any pancreatectomy (positive control). Insulinemia was evaluated by radioimmunoassay (Insulin-CT kit®; CIS BIO International, Gif-sur-Yvette, France). Overall glucose control by the graft was estimated in two among the four initially totally pancreatectomized pigs which did not received any exogenous insulin until graft assessment at 1 month.

Statistical analysis

All data were expressed as the mean ± sem. Mean values of continuous variables were compared using the Mann–Whitney U-test. The distributions of noncontinuous variables were compared using Fisher's exact test. Analyses were performed using Statview® (SAS, Cary, NC, USA), and p-values less than 0.05 were considered statistically significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Graft survival

IMIT was clinically uneventful and no mortality was observed. In two cases, a local abscess developed in one of the graft sites transplanted with unpurified islets. At the time of explantation, 3–180 days after the graft, 68 out of 69 graft sites as delineated by metal clips were successfully recovered (Figure 1C).

image

Figure 1. Intramuscular islet transplantation. Islets were transplanted in the gracilius muscle of adult minipigs in a muscular pouch formed by blunt dissection of the myofibers (panel A) or through intramuscular injection (panel B). Panel C shows a transplant site at the time of explantation at 3 to 180 days after transplantation, delineated by metal clips.

Download figure to PowerPoint

Some degree of fibrosis was present in all grafts when examined with trichrome staining (Figure 2A). In 43 out of 68 grafts, cells with a large cytoplasm were also observed within the fibrosis (white arrows), the beta cell phenotype of these cells was confirmed with immunostaining for antiinsulin antibodies (Figure 2B). Surviving islets were recovered in intramuscular grafts up to 180 days after transplantation. Antiglucagon antibody immunofluorescence also confirmed the presence of alpha cells (Figure 2D).

image

Figure 2. Islet survival. Following graft explantation, intramuscular islets were detected by Masson's trichrome staining (panel A, arrows) and the presence of beta cells was confirmed using antiinsulin immunostaining on consecutive slices (panel B). The integrity of intramuscular islets was confirmed with immunofluorescence with anti chromogranine A (panels C and D, green) and antiinsulin (red, panel C, red) or antiglugagon (panel D, red) antibodies on consecutive slices. The bar represents 100 μm.

Download figure to PowerPoint

Influence of transplantation technique

The overall number of surviving beta cells appeared highly variable, ranging from isolated cells to intact islets. To identify the factors that influenced graft survival, we quantified the islet cell surface (see Methods) in 29 intramuscular grafts removed 15 days after transplantation. As shown in Figure 3, no islets were present after dissection of the myofibers to form a muscular pouch (n = 7); however beta cells were present after the injection technique (p < 0.001). The mean islet surface of injected islets was higher in purified grafts (0.129 ± 0.035; n = 21) than in unpurified grafts (0.003 ± 0.002; n = 8) (p < 0.05). To examine the early outcome of islet survival, grafts were simultaneously performed in three animals with the two techniques and explanted after three days. The muscular pouch technique did not allow survival of transplanted islets, whereas surviving islets were consistently observed in injected grafts. The muscular pouch technique was thus abandoned for rest of the study.

image

Figure 3. Mean ± SEM surface of insulin positive cells within the graft (see methods) 15 days after the intramuscular transplantation of purified (n = 21) or unpurified islets (n = 8), using the pouch technique (n = 7) or the injection technique (n = 22).

Download figure to PowerPoint

Evidence of islet revascularization after intramuscular transplantation

To investigate the revascularization of purified islets injected in the muscle, we analyzed explanted grafts by anti-vWF immunostaining. Vascular density was calculated as described in the methods in six grafts removed 15 days after transplantation and in seven grafts removed 30 days after transplantation. As illustrated in Figure 4, intra-islet vascularization was already visible at 15 days, but appeared significantly higher at 30 days after transplantation (p < 0.01). Abundant vWF staining was associated with satisfactory islet oxygenation as illustrated by the absence of incorporation of injected pimonidazole in large and intact islets (Figure 5A,B,C). Conversely the lack of vWF staining in fragmented islet cell clusters was associated with pimonidazole staining that reflects a PO2 level inferior to 10 mmHg (Figure 5D,E,F).

image

Figure 4. Islet revascularization was assessed by immunochemistry using anti vWF immunostaining. The vascular density (immunostained vessel surface divided by the islet area) was estimated using an imaging program in 32 islets at day 15 (upper left panel) and 24 islets at day 30 (lower left panel). The bars represent 50 μm. Right panel shows box plots of islet vascular density at days 15 and 30.

Download figure to PowerPoint

image

Figure 5. Islet engraftment. Panel A shows a representative example of large intact islets immunostained for insulin, with abundant revascularization as depicted by immunostaining for von Willebrand factor (arrow heads, panel C), and no hypoxemic cells as depicted by immunostaining for pimonidazole (panel E). Panel B shows representative examples of small clusters of insulin positive cells (arrow heads), with little revascularization as depicted by immunostaining for von Willebrand factor (panel D), and hypoxemic cells as depicted by immunostaining for pimonidazole (panel F). The bars represent 50 μm.

Download figure to PowerPoint

Graft function

We then assessed islet graft function in totally pancreatectomized minipigs, 1 month after intramuscular islet transplantation. These animals received an overall volume of 45 853 ± 6981 150 μm islet equivalents (IEQ) with a purity of 72 ± 6% and a viability of 87 ± 1%, corresponding to 1040 ± 160 IEQ/kg. Significant islet graft function defined as mean acute insulin response to intravenous glucose (AIRglc) of 2 mU/L above baseline levels, could be documented in five animals (45%), with a maximum of 18.9 mU/L. In this last case, the persistence of insulin-secreting pancreatic tissue outside of the graft was excluded by repeating the test after graft explantation (Figure 6B). Islet graft function could be similarly documented in 12 out of 18 pancreatectomized minipigs (66%, NS vs. intramuscular islet transplantation) at 1 month after autologous intraportal transplantation of a similar islet volume (1007 ± 182 IEQ/kg, p = 0.37 vs. recipients of intramuscular islets). Overall, the mean AIRglc observed after intramuscular islet transplantation was significantly inferior to the level observed after intraportal islet transplantation (2.8 ± 1.7 mU/L vs. 6.2 + 1.3 mU/L, p = 0.02). As expected, AIRglc was substantially diminished in both groups when compared to the levels observed in healthy controls (35.9 ± 9.1 mUI/L, p < 0.01 vs. intramuscular and intraportal islet recipients). The two totally pancreatectomized IMIT recipients (1911 IEQ per kg and 1015 IEQ per kg) who received no exogenous insulin survived until graft assessment at 1 month. Their mean ± standard deviation fasting blood glucose was 231 ± 31 mg/dL (range 181–285 mg/dL) and 253 ± 35 mg/dL (range 176–304 mg/dL) without any exogenous insulin.

image

Figure 6. Islet graft function. Panel A shows mean ± SEM acute insulin response to intravenous glucose in healthy controls (n = 20, 36 ± 9 mU/L) or in pancreatectomized minipigs, 1 month after intrahepatic (n = 18, 6.2 + 1.3 mU/L) or intramuscular (n = 11, 2.8 ± 1.7 mU/L, p = 0.02 vs. intrahepatic) islet transplantation (1000 IEQ/kg). Panel B shows acute insulin response to intravenous glucose prior to graft explantation, and repeated after graft explantation for excluding the persistence of insulin-secreting pancreatic tissue outside of the graft.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

It is evidently clear that the liver in certain medical situations may not be an optimal site for islet transplantation, due to obstacles by an instant blood-mediated inflammatory response (IBMIR), and low revascularization of intraportally transplanted islets. Here we demonstrated, in a relevant preclinical model, direct evidence of the rapid revascularization, function and long-term survival of autologous islets transplanted in the muscle. We further showed in this minipig model, that islet survival could be significantly improved by limiting physical disruption of the muscle fibers with limited exocrine contamination.

The intramuscular site has been used for decades for autologous transplantation of parathyroid glands, another endocrine tissue [13]. Several rodent experiments have already suggested that it could also be used for islet cells, including two comprehensive rat studies that appeared during the completion of the present study [20, 23]. Interspecies discrepancies are however frequent in islet transplantation, and especially regarding implantation sites, where only few rodent studies were followed by successful clinical translation. The clinical relevance of the intramuscular site for islet cell transplantation has already been explored in large animals [15] or humans [16, 17, 24-26]. None of these studies provided sufficient understanding as to the influential factors required for a successful engraftment. The major clinical potential of IMIT and the need for more conclusive data prompted us to further scrutinize this new approach in a relevant preclinical model. For that purpose, we chose the minipig, a model previously used by our team and others for studying several nonimmunological aspects of islet isolation and transplantation prior to their successful clinical translation [19].

Firstly, we developed a reproducible, and simple technique of direct islet injection under the muscular aponevrosis. Skin incision was used here only to control under direct vision the intramuscular islet delivery and label the transplantation site to ensure subsequent graft recovery. When clinical translation is envisaged, IMIT could easily be performed percutaneously under local anesthesia with ultrasound guidance. Physical disruption of the muscle fibers, the technique routinely recommended for parathyroid transplantation, induced here a deleterious inflammatory reaction with massive monocyte infiltration. In order to clear necrotic cells, monocytes rapidly convert to macrophages that stimulate myogenesis and fibrosis [27], and likely prevented proper islet engraftment. As already reported after transplantation under the kidney capsule [28], this deleterious early inflammatory reaction appeared to be further favored by the presence of exocrine tissue. Like in the dog [15], the transplantation of unpurified islets even induced in the present study a local abscess in two animals. Noteworthy, Rafael et al. did not purify the islets in their clinical case, but placed them in culture for 2 days prior to transplantation [16]. Islet culture, which rapidly decreases the exocrine content of the preparation [29, 30], could represent an alternative to formal purification, especially in autologous transplantation when fewer islets are obtained from a pathological pancreas. Using an immunohistochemistry approach, we provided for the first time in a nonrodent model, direct evidence of the survival of intact islets expressing both insulin and glucagon up to 6 months after intramuscular transplantation. In the present study, islet survival seemed closely linked with revascularization and correct oxygenation. Revascularization is crucial for islet engraftment and function [31, 32]. Similarly, angiogenic activity can be observed during the first week after successful autotransplantation of parathyroid tissue [33, 34]. In the rodent, blood flow is restored in transplanted islets as early as 4–10 days after the graft [35]. In the present study, capillaries could be seen as early as 15 days in large engrafted porcine islets. Noteworthy vascular density of the graft or at least its reendothelialization depicted by the expression of von Willebrand factor, further increased with time as previously described by Lukinius et al. [36]. Noteworthy, some small beta cell clusters still expressed at 30 days high level of pimonidazole, a marker of hypoxia. Recent studies suggest that well vascularized islets can become hypoxic when blood glucose rises [37] or in special subpopulations of dormant islets [38]. However, the residual pancreatic head maintained normoglycemia, and the lack of immunostained capillaries suggests more a lack of revascularization than dormant islets. Using a bioengineered scaffold impregnated with proangiogenic factors approach, Witkowski et al. recently optimized islet vascularization engraftment [23]. By limiting islet loss in the early post transplantation period, this type of device should improve graft function.

In the last part of our study the function of intramuscular islet graft could be unequivocally demonstrated by stimulated insulin secretion after total pancreatectomy in half cases. IMIT seemed however less efficient than intraportal islet transplantation. One month after the transplantation the acute insulin response to glucose, an in vivo surrogate marker of graft function, was indeed 55% lower in recipients transplanted in the muscle compared with those transplanted in the liver with an identical islet mass (1000 IEQ/kg). After total pancreatectomy, IMIT allowed survival and partial glucose control in absence of exogenous insulin in sharp contrast with overt hyperglycemia and cachexia habitually observed within 3 weeks in the pig model [39-41].

An optimal site for islet implantation should (1) provide an appropriate environment including rich arterial supply, (2) permit normalization of blood glucose level, (3) require a mini invasive procedure and be also suitable for the monitoring and graft explantation [3]. Among the various potential alternatives to the classical intraportal infusion, IMIT offers several distinct advantages. The most obvious is the minimal invasiveness of the procedure, confirmed here by the absence of any severe adverse events when purified islets were transplanted. On a short term, IMIT also limits the direct contact of transplanted islets with blood, presumably responsible for the immediate loss of one third of islets after intraportal infusion [11]. In the longer term, intramuscular islets are revascularized through systemic arterial circulation, escaping the portal blood flow and its toxic xenobiotic content [42, 43]. Easily accessible, the intramuscular site would eventually greatly facilitate in vivo their functional monitoring and imaging of transplanted cells [17], as well as their eventual explantation if needed.

Few alternative sites, besides the liver, have allowed good functional results in man or relevant preclinical models [3]. Some interesting results were, however, recently reported in nonhuman primate after islet transplantation in the anterior chamber of the eye or in the omental pouch. In the first report, stimulated secretion was observed after 3 months in a baboon who received two allogenic islets transplantation in the anterior chamber of the eye. The animal developed however a cataract, questioning the safety of this procedure [44]. In the second report, elevated C peptide and decrease in exogenous insulin requirement was observed after islet transplantation in omental pouch in five diabetic monkeys [45]. As in our study, the authors conclude that the omental pouch is a potential extrahepatic transplant site, even if a greater islet mass may be required to achieve insulin independence compared to the liver. Like IMIT, these two sites provide an appropriate vascularization, and some function. However both the anterior chamber of the eye and omental pouch required a more complex procedure than IMIT, and do not offer an easy access to the graft monitoring and graft explantation remains questionable. In contrast, the brachioradialis muscle in man offers greater access to the graft for imaging, and functional testing to monitor even in the presence of residual orthotopic pancreatic tissue [17]. This site, which is routinely used for parathyroid autotransplantation, enables the possibility of easy graft explantation under local anesthesia, if needed [46].

Collectively, with recent reports of successful clinical translation of IMIT in Lille [17] and Uppsala [16], the present study further supports the exploration of striated muscle as an alternative to intraportal infusion for transplanting insulin-secreting cells. Our data demonstrated that minimizing myofiber trauma and exocrine contamination are essential to optimize the outcome of IMIT. Even if less efficient than standard intraportal infusion, the intramuscular site offers interesting perspectives in specific medical situations. Based on its simplicity, and minimal morbidity, intramuscular injection may already be the preferred choice to intraportal infusion, for islet autotransplantation following pancreatectomy, and in cases of pancreatic trauma [47]. Following a pancreatectomy for a potentially neoplastic tumor, intramuscular islet autotransplantation guarantees an easy access to the graft for noninvasive imaging or future extirpation [17]. Finally, in this context, intramuscular transplantation also appears as a good compromise for exploiting insulin-secreting cells obtained from alternative cell and less limited sources, such as stem cells or xenogenic cells, when they become available [48].

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

This study was supported by grants awarded by the 7th Framework Programme of the European Commission (Beta-Cell Therapy), European Genomic Institute for Diabetes (EGID, ANR-10-LABX-46), INSERM (Programme National de Recherche sur le Diabete), Conseil Regional Nord–Pas de Calais, University of Lille2, Fonds Européen de Developpement Economique et Regional, Agence de Biomédecine and OCS-01778-08-2005 from Oncosuisse. The authors are indebted to Drs. C. Bonner, L. Arnalsteen, B. Lefebvre, B. Vandewalle, E Moerman, B Lukowiak R. Ezzouaoui, S. Belaich, B. and I. Alunga for their support.

Disclosure

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References
  10. Supporting Information

Additional Supporting Information may be found in the online version of this article at the publisher's web site:

Table S1: Primary antibodies for immunohistochemistry and immunofluorescence.

Figure S1: Confirmation of specific positive endothelial immunostaining using anti-von Willebrand Factor antibodies on an intramuscular porcine vessel (A). With the same antibody, panel B shows a specific staining on vessels of a porcine pancreas without a specific staining of exocrine tissue or duct cells (arrow) (B).

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.