SEARCH

SEARCH BY CITATION

Keywords:

  • Human leukocyte antigen;
  • insulin-independence;
  • islets transplantation;
  • morphology;
  • optical projection tomography;
  • vascularization

Abstract

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

Long-term insulin independence after islets of Langerhans transplantation is rarely achieved. The aims of this study were to identify the histological and immunological features of islets transplanted in a type 1 diabetic patient who died of a cerebral hemorrhage after >13 years insulin independence. Islets were pooled from two donors with respectively one and five HLA mismatches. Insulin-positive islets were found throughout the right and left liver, and absent in the pancreas. Two- and three-dimensional analysis showed that islets lost their initial rounded and compact morphology, had a mean diameter of 136 μm and were constituted of an unfolded epithelial band of 39.1 μm. Leukocyte phenotyping showed no evidence of a tolerogenic environment in the islet-containing portal spaces. Finally, HLA typing of microdissected islets showed HLA from the best matched donor in all 23 microdissection samples, compared to 1/23 for the least matched donor. This case report demonstrates that allogeneic islets can survive over 13 years while maintaining insulin independence. Allogeneic islets had unique morphologic features and implanted in the liver regardless of their size. Finally, our results suggest that, in this case, rejection had been prevalent over autoimmunity, although this hypothesis warrants further investigation.

Abbreviations
HLA

human leukocytes antigen

IEQ

islet equivalents

OPT

optical projection tomography

Since the introduction of the Edmonton protocol, insulin independence is achieved in more than 70% of patients 1 year after clinical islet transplantation [1]. However, long-term insulin independence over the 10-year mark has remained exceptional due to several reasons including instant blood-mediated inflammatory reaction, acute and chronic rejection, recurrence of autoimmunity, drug toxicity, widespread amyloid deposition and low levels of β-cell proliferation [2, 3].

Case Report

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

Herein, we report the case of a 36-year-old woman with a 27-year history of type 1 diabetes who received intraportally an islet-after-kidney allograft in June 1996 and died of a cerebral hemorrhage in October 2009 after >13 years of insulin-independence. Prior to transplant, metabolic control (HbA1c at 11.2%) was poor while injecting a mean insulin dose of 16 IU/day, her BMI was 22.1 kg/m2, and she suffered from severe hypoglycemia unawareness [3]. She received a total of 8800 islet equivalents (IEQ) per kilogram body weight pooled from two deceased donors with respectively five (106 700 IEQ) and one (421 300 IEQ) HLA mismatches (Supporting Table S1). Except for cold ischemia time and intensive care unit (ICU) stay, both donors had similar characteristics (Supporting Table S2). Two months after transplantation, she became insulin free for the rest of her life with excellent metabolic control (Supporting Figure S1). Panel-reactive antibody of the patient was 0% prior to transplantation, and she never developed de novo anti-HLA antibodies over 13 years of follow-up. The immunosuppressive protocol consisted of cyclosporine, mycophenolate mofetil and low-dose steroids [3].

Material and Methods

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

Liver and pancreas autopsies

Autopsy and studies on liver and pancreas specimens were performed after obtaining oral informed consent from the patient's family, as per our standard institutional rules.

Immunofluorescence and morphologic quantifications

Peripheral and central sections of the right and left liver were embedded in paraffin. Slides were stained for CD34, insulin and glucagon as previously described [9] and scanned with the Mirax scan microscope (Zeiss, Feldbach, Germany). Quantifications of insulin (in blue) and glucagon (in red) positive cells were performed in digital images using the offline MetaMorph imaging software for microscopy (Universal Imaging, West Chester, PA, USA). The software was automatically programmed to quantify blue and red fluorescence within the areas of interest. The total area of all islet structures measured in the liver was 285 714 μm2, among which 254 129 μm2 was insulin positive (89%) and 31 584 μm2 glucagon positive (11%). In the pancreas, the area of 32 islets was measured, among which 169 392 μm2 was glucagon positive (99.97%) and 58.3 μm2 was insulin positive (0.03%). The β-cell:α-cell ratio was calculated as the insulin-positive area/glucagon-positive area for each organ.

Optical projection tomography (OPT)

Human samples (embedded in paraffin) were stained with primary guinea pig anti-insulin (A0564, Dako) and rabbit anti-glucagon (AB932, Millipore) antibodies, followed by secondary anti-guinea pig Alexa 488 (A11073, Invitrogen) and anti-rabbit Alexa Fluor 594 (A11012, Invitrogen) antibodies. OPT scanning was carried out using the Bioptonics 3001 OPT M scanner (Bioptonics) with exciter D560/40x and emitter E610Ipv2 filter (Chroma), or exciter D480/30 and emitter HQ535/50 filter (Chroma) when visualizing Alexa 594 and 488, respectively (4,5). Tomographic reconstructions were generated using the NRecon V1.6.1.0 (Skyscan, Belgium) software and reconstructed images were further assessed using Bioptonics viewer V2.0 as previously described (4). Movies and images were finally constructed using Image J 1.43 u software. Volumetric and diameter calculations were done on the Imaris software (Bitplane Inc., South Windsor, CT, USA).

Insulin staining, microdissection and nested HLADRB1 PCR

Five to nine consecutive frozen sections were mounted on glass slides with a PEN-membrane (Leica Microsystem). Sections were fixed with methanol for 2 min, permeabilized with 0.5% Triton X-100 for 5 min and successively stained with antiguinea-pig insulin (1:250, invitrogen) and Alexa 488 antiguinea-pig for 1 hour respectively. The slides were air-dried at room temperature and prepared for microdissection on a Leica LMD6500 (Leica Microsystem). Microdissections were performed in the left (n = 13) and right (n = 10) parts of the liver. Microdissected tissues were recovered in PCR micro-tubes containing 20 μL of 50 mM NH4SO4 200 mM Tris-HCL ph 8.8 buffer and heated at 95°C for 5 min before. Thereafter, a first PCR was performed with 10 μL of each microdissected sample using the generic DRBP1/DRBP2 primers [18]. Products of the first PCR were diluted 1:200 and subjected to a second PCR step with group-specific primers. Primers were chosen according to HLA incompatibilities between donor 1, 2 and recipient respectively DRB1*10, DRB1*12 and DRB1*03 [4]. Product sizes were expected to be 203 (DR10), 163 (DR12) and 211(DR3) base pairs. Amplification profiles and reagents were identical as previously described [19]. PCRs amplifications were detected by agarose gel electrophoresis.

Results and Discussion

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

Post mortem histological analysis of the liver and the pancreas revealed the presence of insulin-positive islets throughout the left and right liver lobes (Figure 1A). Islets were also found in the pancreatic parenchyma, but did not stain for insulin. Rare isolated β cells were found scattered in the pancreas (Figure 1B), corresponding to what has been described in long-standing type 1 diabetic patients [6]. The overall β-cell:α-cell ratio was 89:11 in the liver (left/right, central/peripheral), compared to 1:100 in the pancreas (Figure 1C). The large number of native α cells in the pancreas may have caused preferential turnover of β cells in the liver. In the liver, β cells had a core location, whereas α cells had a mantle location (Figure 1a), δ- and PP-positive cells were only rarely detected in the liver in comparison to native pancreatic islets (data not shown) suggesting that glucagon, somatostatin and PP were mainly produced in the pancreas. Taken together, these data discard the notion that insulin-independence was achieved as a result of native β-cell regeneration and argue against a potential beneficial effect of chronic immunosuppression on endogenous β-cell proliferation [6].

image

Figure 1. Morphological analysis of the transplanted islets. Representative histological sections of islets present in the liver (A) and the pancreas (B) after triple-staining for insulin (blue), glucagon (red) and CD34 (green). Independent islet structures (referred to in the text as «epithelial bands») are numbered 1–2-3–4-5. Scale bar: 100 μm. (C) The proportion of insulin (I) or glucagon (G) positive cells was calculated as the area of I or G divided by (I+G) and expressed as percentage (%). The results represent 163 independent islet structures in the right and left liver (peripheral and central part) and 32 islets in the pancreas. The β-cell:α-cell ratio was constantly similar in the liver (respectively 89/11 (right peripheral), 90/10 (right central), 90/10 (left peripheral) and 87/13 (left central). (D) The mean diameter (μm) of each independent islet structure (insulin and/or glucagon positive) was assessed by immunofluorescence in the right lobe (central part n = 35, peripheral part n = 50) and in the left lobe (central part n = 37, peripheral part n = 41). (E) Representative blocks of liver tissue containing transplanted islets and healthy pancreas tissue are depicted. The blocks were stained for insulin (green) and glucagon (red), and analyzed using the optical projection tomography technology. (F) Volumetric quantification of insulin and/or glucagon positive islet structures based on optical projection tomography in the liver and healthy pancreas. Results show size distribution according to diameter categories of the islets structures in the liver (squares, n = 250), healthy pancreas (circles, n = 369) compared to the islets size distribution measured prior to transplantation of donor 1 (triangles) and 2 (inverted triangles). Error bars represent the mean and SEM.

Download figure to PowerPoint

The advantage of large versus small size in long-term islet survival after liver implantation is a matter of debate. Two-dimensional analysis of liver sections revealed that the mean diameter of islet sections was 39.1 μm (Figure 1D). Recently, we have shown that human pancreatic islets are constituted of one compact epithelial band measuring on average 42 μm [7]. Thus, it was unclear whether the observed diameter of 39.1 μm represented the true size of the transplanted islets or only the epithelial band of larger islets that had unfolded during implantation or over the years. Three-dimensional analysis using consecutive serial sections confirmed that the islets had lost their initial rounded and compact morphology and that islets were mainly constituted of one unfolded epithelial band of 39.1 μm (Supporting Figure S2). When the volume and actual dimensions of the transplanted islets were measured by optical projection tomography (Figure 1E, Supporting Videos S1 and S2), the corresponding mean diameter size was estimated to be 136 μm. Interestingly, the size distribution of the transplanted islets analyzed in the liver specimen and of the islets isolated from the better-matched donor 2 prior to transplantation was identical (Figure 1F, Supporting Table S1). If one considers that islets have a predefined size, these results suggest that (i) islets stemming from donor 2 preferentially survived in the liver, and (ii) that islets implanted in the liver regardless of their size, which argues against recent data suggesting that smaller islets have an engraftment advantage in human islet transplantation [8, 9]. Of interest, the Nordic group has published similar results on a patient deceased 4.5 years after a first islet infusion [10].

Vascularization of transplanted islets either depends on intraislet endothelial cells originating from the donor or from ingrowing host cells [11]. Using the endothelial cell marker CD34, histological analysis of the transplanted islets revealed only occasional vascular channels (7%) penetrating the core of the epithelial bands (referred as central vasculature), whereas peripheral endothelial cells (external to the epithelial band) were present in 94% of the transplanted islets. In 6% of the islets, CD34 staining was not detected (Figure 1A, Supporting Table S3). This finding is in sharp contrast to the native pancreas of the patient, where 78% of the islets exhibited central vasculature (Figure 1B, Supporting Table S4). Thus, the peripheral location of the endothelial cells suggests that the intrinsic vasculature of the islets in the liver did not persist over the years and was replaced by a peripheral neovasculature originating from the recipient.

To assess whether long-term islet survival was associated with an immuno-protective microenvironment, CD4, CD8, Foxp3, IL-17, CD20 and CD68 positive cells were stained by immunochemistry. The respective numbers of each subset were compared in portal spaces with or without islets following normalization to the surface of the portal space after exclusion of the vasculature and islets (Supporting Figure S3). No significant difference, except for a trend for increased CD68+ cells (p = 0.07), was detected in the number of leukocytes surrounding the islets (Supporting Figure S4). In conclusion, no evidence of a tolerogenic environment was found in the portal spaces containing the islets.

Donor–recipient HLA matching positively influences long-term graft survival in terms of allogeneic rejection [12, 13], although it has been suggested to be able to trigger recurrence of autoimmunity [14, 15]. Based on the results of pancreas transplantation, recurrence of autoimmunity seems to be increased in nonimmunosuppressed patients [16]. Nevertheless, using an appropriate immunosuppressive regimen, the majority of HLA-matched transplanted pancreases achieve long-term insulin independence [17]. Since no such data are as yet available concerning islet transplantation, we investigated whether the islets present in the liver 13 years after transplantation came from donor 1 (HLA match: 1/5) or donor 2 (HLA match: 5/6) using in situ microdissection (Figure 2A) and a two-stage nested PCR [18-20]. HLA from both donors was detected in 1 sample whereas all 23 micro-dissected samples contained HLA from donor 2 (Figure 2B) demonstrating that islets from donor 2 mainly mediated insulin independence. Although, the complexity of the microdissection technique renders difficult the analysis of larger numbers of islets, these observations suggest that, in this case, the beneficial effect of HLA-matching on control of alloimmunity was more important than the potential detrimental effect on triggering of autoimmunity. Of note, the patient tested positive for anti-GAD65 antibodies and negative for anti-IA2. She had negative PRA reactivity before transplantation and never developed anti-HLA antibodies over 13 years of follow-up. Thus, we cannot exclude that islets from donor 1 suffered more from longer cold ischemia time and ICU stay prior to transplantation (Supporting Table S2).

image

Figure 2. Immunological analysis of the transplanted islets. (A) Five to nine serial cryosections of the liver were prepared on a PEN membrane and stained for insulin (green). The slides were air-dried at room temperature and prepared for microdissection on a Leica LMD6500. Representative islets in a portal space before and after microdissection. Scale bar: 100 μm. (B) Islets stemming from the left (MD 1–13) and right liver (MD 14–23), respectively, were stained for insulin and microdissected (see also supporting figure S4). Control microdissections were performed in portal spaces without islets. A two-stage nested PCR based on HLA-DRB1 incompatibilities between donor 1 (left, 10), 2 (right, 12) and recipient, respectively HLA-DRB1*10, -DRB1*12 and -DRB1*03, was performed. HLA-DRB1*10, DRB1*12 from donors 1 and 2 respectively was amplified in one microdissection, whereas all other microdissections were positive for HLA-DRB1*12 from donor 2 only. DNA of both donors 1 and 2, which had been stored at the time of transplantation, was used as positive control. Islets-free portal spaces were microdissected and amplified for donor 1, 2 and recipient (middle, 3) as further controls. CN: negative control based on two-stage nested PCR of a microdissection-free sample. MD: microdissection, D1: donor 1 HLADRB1*10+, D2: donor 2 HLADRB1*12+. Recipient HLADRB1*03+.

Download figure to PowerPoint

Conclusion

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

Overall, this case demonstrates intrahepatic survival of allogeneic islets of Langerhans for more than 13 years with excellent metabolic function. Insulin immunofluorescence staining demonstrated the absence of insulin-positive cells within the native pancreas’ islets and the presence of insulin-positive islets in the liver with significant reshaping. Insulin independence was mainly mediated by islets stemming from the better (5/6) HLA-matched donor without evidence of a tolerogenic cellular environment in the islet-containing portal spaces. We have to acknowledge that it remains a single case, and that, because of technical issues, only a limited number of islets could be analyzed. Thus, further investigations are warranted to confirm the hypothesis of HLA-matching benefit. However, understanding the unique features that contributed to long-term insulin independence in this case and in the next cases likely to be reported in the near future may help turn into a rule what has been as yet an exception.

Acknowledgments

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

This work was supported by grants from the Swiss National Research Foundation (# 3200B-120376 to D.B. and T.B., SCORE grant #3232230-126233 to C.T.), the Juvenile Diabetes Research Foundation (# 31-2008-416 to T.B.) and the E. & L. Schmidheiny Foundation (to Y.D.M.).

We thank Gisella Puga-Yung and Dela Golshayan for critical advice and help in the performance of the experiments, and David Matthey-Doret, Lisa Perez, Solange Masson, Corinne Sinigaglia for their excellent technical assistance.

Disclosure

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

Supporting Information

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

Disclaimer: Supplementary materials have been peer-reviewed but not copyedited.

FilenameFormatSizeDescription
ajt12138-sup-0001-videoS1.avi2492KSupporting Material and Methods: Three-dimensional analysis.
ajt12138-sup-0002-videoS2.avi7092KSupporting Material and Methods: Immunohistochemistry and leukocyte quantification.
ajt12138-sup-0003-FigureS1.doc6076K

Figure S1: Long-term insulin independence after allogeneic islet transplantation.

Figure S2: Three-dimensional analysis of islets implanted in the liver.

Figure S3: Leukocyte subsets in portal space with or without islets.

Figure S4: Immunologic analysis of the transplanted islets.

Table S1: HLA status of patient and islets donors.

Table S2: Donor characteristics and islets isolation outcomes.

Table S3: Islets transplanted in the liver have a peripheral vasculature.

Table S4: Islets in the patient's pancreas have both central and peripheral vasculature.

Video S1: Isosurface reconstruction of the patient liver.

Video S2: Isosurface reconstruction of the human pancreas.

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.