Genetically engineered Sertoli cells are able to survive ... - Nature

Gene Therapy (2004) 11, 694–700 & 2004 Nature Publishing Group All rights reserved 0969-7128/04 $25.00 www.nature.com/gt

RESEARCH ARTICLE

Genetically engineered Sertoli cells are able to survive allogeneic transplantation JM Dufour1, R Hemendinger2, CR Halberstadt2, P Gores2, DF Emerich3, GS Korbutt1 and RV Rajotte1,4 1

Department of Surgery, Surgical-Medical Research Institute, University of Alberta, Edmonton, Canada T6G 2N8; 2Department of General Surgery and The Transplant Center, Carolinas Medical Center, Charlotte, NC 28232, USA; 3Sertoli Technologies, Inc., Cranston RI, 02921, USA; and 4Department of Medicine, University of Alberta, Edmonton, Canada T6G 2N8

The immunoprotective nature of the testis has led to numerous investigations for its ability to protect cellular grafts. Sertoli cells (SCs) are at least partially responsible for this immunoprotective environment and survive allogeneic and xenogeneic transplantation. The ability of SCs to survive transplantation leads to the possibility that they could be engineered to deliver therapeutic proteins. As a model to test this hypothesis, we examined the ability of SCs that produce green fluorescent protein (GFP) to survive transplantation and continue expressing GFP. SCs were isolated from transgenic mice engineered to express GFP and transplanted as aggregates under the kidney capsule of severe combined immunodeficient (SCID) and Balb/c mice. Using this paradigm, it was possible to compare the survival of

transgenic SCs directly in both immunodeficient and immunocompetent recipients. Fluorescence microscopy of the kidney capsule and immunohistochemistry of the grafts for GFP and GATA-4 revealed the presence of GFP-expressing SCs under the kidney capsule of SCID and Balb/c mice at both 30 and 60 days post-transplantation. In contrast, islets transplanted to Balb/c mice were rejected. Thus, SCs survive transplantation and continue to express GFP raising the possibility that SCs can be engineered using transgenic technology to produce proteins, such as insulin, factor VIII, or dopamine for the treatment of diabetes, hemophilia or Parkinson’s disease, respectively. Gene Therapy (2004) 11, 694–700. doi:10.1038/sj.gt.3302218 Published online 15 January 2004

Keywords: Sertoli cell; GFP; transplantation

Introduction The testis is considered an immunoprivileged site and has been investigated for its ability to protect cellular grafts from the host’s immune system.1–4 This immunoprotective environment is due in part to the secretion of immunosuppressive molecules by Sertoli cells (SCs)5–7 that allow ectopically transplanted SCs to survive as allografts8–10 and concordant (rat-to-mouse)11 and discordant (pig-to-rat)12,13 xenografts in rodents. In addition to the ability of transplanted SCs to protect themselves in immunologically foreign environments, SCs were shown to protect islets and neuronal cells from immunemediated rejection.9,10,14–19 The ability of SCs to protect themselves from immunemediated destruction, as well as their ability to produce a variety of proteins, has novel and potentially far-reaching implications for gene therapy. Gene therapy is an emerging field with the goal of replacing defective and deficient proteins to restore function in disease states. Traditionally, these therapies have involved the use of viral vectors or other methods to produce the expression (often transient) of the gene of interest, raising questions of safety and efficacy.20–22 Cell-based gene therapy has been proposed as one means of overcoming these issues Correspondence: RV Rajotte, Surgical-Medical Research Institute, 1074 Dentistry/Pharmacy Centre, University of Alberta, Edmonton, Alberta, Canada T6G 2N8 Received 10 June 2003; accepted 8 October 2003

but has itself been hampered by the lack of an abundant, safe and immunologically acceptable source of the tissue.21 Theoretically, transgenic animals can be designed to produce specific therapeutic proteins overcoming the issues of transient expression and safety associated with viral vectors, but this is hampered by immune rejection of the allogeneic or xenogeneic tissue. The use of transgenic SCs could overcome this issue by providing long-term gene expression in an immunoprivileged environment allowing the efficacious use of allogeneic or even xenogeneic cells. While the use of SCs to deliver therapeutic proteins has potential applications across numerous clinical situations, the greatest amount of data collected to date have been focused on diabetes and Parkinson’s disease (PD).9,10,14–19 It is not our goal to review these data here; however, it seems intuitively obvious that the consistent ability to graft SCs in animal models of diabetes and PD successfully provides a strong foundation for further exploration of transgenic SCs in these same preclinical systems. For instance, islet transplantation for the treatment of type I diabetes was recently demonstrated to be a viable option.23 However, in order to treat the majority of type I diabetic patients, two major problems must be overcome: elimination of the requirement for chronic immunosuppression needed to prevent human islet allograft destruction and an unlimited supply of insulin-producing tissue. The use of SCs engineered to produce insulin following appropriate stimulus would overcome both issues mentioned above. Similarly, this approach could be utilized for the

Transplantation of GFP Sertoli cells JM Dufour et al

treatment of PD since cotransplantation of SCs improved the engraftment of human neuronal cells when transplanted in hemi-Parkinsonian rat striatum.19 In this study, we hypothesized that if SCs were engineered using transgenic technology to produce therapeutic proteins such as insulin or dopamine for the treatment of diseases such as diabetes or PD, then we could eliminate the need for immunosuppressive drugs and cotransplantation of cells. The results would enable us to overcome the major hurdles of gene therapy. In order to test this hypothesis, we used SCs isolated from transgenic mice engineered to produce green fluorescent protein (GFP) as a model and examined whether they would be able to survive allogeneic rejection and continue to express GFP.

Results Characterization of SCs from TgN(GFPU)5Nagy mice To verify the presence of GFP in SCs isolated from TgN(GFPU)5Nagy transgenic mice, testicular sections from TgN(GFPU)5Nagy and wild-type mice were immunostained for GFP (Figure 1a and b). GFP-positive SCs were present in the TgN(GFPU)5Nagy but not in the wild-type testicular sections. To further verify the expression of GFP by SCs, seminiferous tubules and SCs were isolated from TgN(GFPU)5Nagy mouse testes and examined under blue-light fluorescence. An aliquot containing seminiferous tubules was collected during the isolation procedure after collagenase digestion. These tubules contained many GFP-expressing cells (Figure 1c).

At the time of transplantation, SCs that had been cultured for 2 days to form aggregates were analyzed and also contained GFP-expressing cells (Figure 1d and e). Since the seminiferous tubules and aggregates contain other cells including germ cells and peritubular myoid cells, isolated cells were also cultured on chamber slides as a monolayer. SCs cultured in a monolayer can be identified by their characteristic morphology. These SCs expressed GFP (Figure 1f). Nonetheless, it should be noted that not all of the cells from the TgN(GFPU)5Nagy mouse testes expressed GFP (Figure 1d). SCs isolated from Balb/c mice were not fluorescent (data not shown). Prior to transplantation, the composition of the transplanted cells was determined by assessing the proportion of immunoreactive GFP- or GATA-4-positive SCs and smooth muscle a-actin-positive peritubular myoid cells.24 The cellular preparation was shown to contain 54.771.9% SCs and 2.870.6% myoid cells. Of the SCs, 66.173.3% were also GFP positive. The remaining cells were almost exclusively germ cells as determined by hematoxylin and eosin stain. Since germ cells are unable to survive at the higher temperature of the abdominal cavity, we did not expect them to survive after transplantation underneath the kidney capsule.25

695

Transplantation of TgN(GFPU)5Nagy SC aggregates to severe combined immunodeficient (SCID) mice To assess whether SCs that produce GFP would be able to survive transplantation and continue to express the transgene, SC aggregates expressing GFP were transplanted under the left kidney capsule of

Figure 1 Characterization of SCs from TgN(GFPU)5Nagy mice. Immunohistochemistry was performed on testicular sections from TgN(GFPU)5Nagy (a) and wild-type (b) mice (n ¼ 3) using a mouse monoclonal anti-GFP antibody followed by Cy3-conjugated goat anti-mouse secondary antibody. In addition, seminiferous tubules and SCs were isolated from TgN(GFPU)5Nagy mouse testes and examined under blue-light fluorescence for GFP expression. Seminiferous tubules (c), SCs that were cultured for 2 days to form aggregates prior to transplantation (d, e) and SCs cultured on chamber slides as a monolayer (f) (n ¼ 6 for each). Arrowhead, SC; arrow, SC aggregate. Bar in e, 100 mm for c–e; bar in f, 50 mm for a, b and f. Gene Therapy


696

immunocompromised SCID mice. Nephrectomies were performed after 30 and 60 days and kidneys were analyzed immediately with blue-light illumination for the presence of GFP-positive cells under the kidney capsule. Many brightly fluorescent GFP-positive cells were detected in each of the grafts (Figure 2a and b). Negative controls consisting of the kidney from a SCID mouse did not fluoresce (Figure 2c). To be certain that the GFP-positive cells detected in the grafts were SCs, sections from grafts 30 and 60 days posttransplant were double immunostained for GFP and GATA-4 (a marker for SCs).24 GFP-positive SCs were present in 100% of the grafts at both 30 and 60 days posttransplant (Figure 3). However, not all of the SCs expressed GFP. This is consistent with the variable expression of GFP in the SCs from the TgN(GFPU)5Nagy mouse testes. The proportion of GFP-positive cells in the grafts was 46.674.4% GFP-positive SCs, 52.774.4%

Figure 3 GFP- and GATA-4-positive SCs in SCID mouse grafts. Sections from grafts 30 (a–d) and 60 (e–h) days post-transplant (n ¼ 3/time point) were double immunostained for GFP (red, b, d, f and h) and GATA-4 (brown, a, c, e and g). GATA-4 is used as a marker for SCs.24 Sections were counterstained with hematoxylin. Arrow, SC. Bar in f, 50 mm, for a, b, e and f; bar in h, 25 mm for c, d, g and h.

GFP-negative SCs and 0.870.5% GFP-positive cells that were a cell type other than SCs.

Transplantation of TgN(GFPU)5Nagy mouse islets to Balb/c mice Prior to testing the survival of GFP-expressing SCs as allografts, it was first necessary to confirm that nonimmunoprivileged cells from the genetically undefined TgN(GFPU)5Nagy mice would be immunologically rejected by H-2d Balb/c mice. Islets isolated from TgN(GFPU)5Nagy mice were transplanted to the renal subcapsular space of diabetic Balb/c mice, and all animals rejected these islets within 17 days, mean graft survival 14.7571.5 days7s.e. (Figure 4). Figure 2 Transplantation of TgN(GFPU)5Nagy SC aggregates to SCID mice. Grafts were removed at 30 (a) and 60 (b) days post-transplant (n ¼ 3/ time point) and kidneys analyzed immediately with blue-light illumination for the presence of GFP-positive cells under the kidney capsule. Photomicrograph of the kidney from a SCID mouse as a negative control (c, n ¼ 3). Bar in c, 100 mm for a–c. Gene Therapy

Transplantation of TgN(GFPU)5Nagy SC aggregates to Balb/c mice To determine whether GFP-expressing SCs can survive as allografts and continue to produce the transgene,


697

Figure 4 Function of TgN(GFPU)5Nagy mouse islets transplanted to diabetic Balb/c mice (n ¼ 4). Nonfasting blood glucose levels are shown from the time of transplantation (day 0) to the time of graft rejection.

SC aggregates expressing GFP were transplanted under the left kidney capsule of Balb/c mice. Nephrectomies were performed after 30 and 60 days and kidneys were analyzed immediately under blue-light illumination for the presence of GFP-positive cells under the kidney capsule. GFP-positive cells were found in all the grafts at both 30 and 60 days post-transplant (Figure 5a–d). In fact, some of the grafts (four of 10) contained fluorescent tubules (Figure 5b and d). The kidney from a Balb/c mouse was used as a negative control and did not fluoresce (Figure 5e). To confirm that the GFP-positive cells detected in the grafts were SCs, sections from grafts 30 and 60 days posttransplant were immunostained for GFP and/or GATA4. When tissue sections were immunostained for GFP alone, positive cells were detected in six of the 10 grafts, while seven of the 10 grafts were positive when immunostained for GATA-4 alone. When sections were double stained for GFP and GATA-4, GFP-positive SCs were present in 50% of the grafts (Figure 6). Only a few infiltrating lymphocytes were present in the grafts. These data support the hypothesis that genetically engineered SCs can survive as allografts and express genes of interest.

Discussion The use of transgenic tissue in gene therapy offers the advantages of long-term, stable expression of proteins without the safety concerns associated with viral vectors. However, issues of immune rejection prevent the widespread use of this tissue for transplantation. The ability of transgenic SCs to establish a locally immunoprivileged environment in transplant recipients could overcome these obstacles. The current study provides ‘proof-ofprinciple’ data of this concept by demonstrating that SCs obtained from transgenic mice engineered to produce GFP survived transplantation and continued to express GFP in both immunocompromised SCID mice and immunocompetent Balb/c mice for at least 60 days (ie

Figure 5 Transplantation of TgN(GFPU)5Nagy SC aggregates to Balb/c mice. Grafts were removed 30 (a, b) and 60 (c, d) days post-transplant (n ¼ 5/time point) and kidneys were analyzed immediately under bluelight illumination for the presence of GFP-positive cells under the kidney capsule. As a negative control, the kidney from a Balb/c mouse was analyzed for autofluorescence (e, n ¼ 5).

the longest time point examined). Together these data suggest that SCs can be easily altered to express foreign proteins and that this expression does not inhibit the natural ability of SCs to protect themselves against allogeneic rejection. In the current study, transgenic SCs exhibited robust GFP fluorescence in the native testis, as well as in vitro following enzymatic digestion from the testis and most importantly following transplantation into two different recipient animals. These data indicate that the expression of GFP in these animals is nontoxic to normal SC development and that the expression of the transgene can be maintained following conventional manipulation used for isolation and transplantation. The expression of GFP did not interfere with the SC’s ability to protect itself from immune rejection as demonstrated by the fact that GFP-expressing SCs survived long-term in allogeneic recipients, while pancreatic islets were rapidly rejected. These data are all the more impressive considering the suggestion from previous work that the GFP protein itself is immunogenic.26 When GFP-positive SCs were transplanted into SCID mice, 100% of the grafts survived as determined by both fluorescence microscopy and immunohistochemistry for GATA-4 (a marker for SCs). In contrast, the survival of GFP SCs in Balb/c mice was more variable with surviving SCs detected in 50–70% of the allografts. While the underlying reason for the variability is currently unknown, several factors likely contribute including the Gene Therapy


698

cells. Most notable are the recent reports identifying the negative issues using adenoviral vectors as a means of inserting genes into cells.22,27 Other vectors have had similar unacceptable safety profiles,21,22 not to mention that the current vectors are limited as to the size of the gene constructs and hence, the types of proteins that can be generated. On the other hand, the creation of stably transfected cells has been achieved by microinjections into the embryos.28,29 The development of transgenic animals has been well established for the study of a variety of diseases and for the production of human proteins for therapeutic purposes.30–34 SCs have the capacity to engraft in various environments and produce a variety of proteins that are immunoprotective. In addition, these cells are naturally very prolific protein producers. Hence, the use of stably transfected SCs isolated from transgenic animals such as a pig could become an ‘off-the-shelf’ source of cells that not only protect themselves from immunedestruction but also produce therapeutic proteins. In conclusion, this study demonstrates the ability of SCs isolated from transgenic animals to engraft and express a transgene in an animal model. These data have major implications for the field of gene therapy by providing a means of long-term delivery of crucial proteins needed to reverse disease states without immunosuppressive drugs and with minimal safety issues. Future studies will focus on the expression of therapeutically relevant protein expression and secretion by SCs to reverse a particular disease state.

Materials and methods

Figure 6 TgN(GFPU)5Nagy mouse SCs survive and express GFP in Balb/c mice. Sections from grafts 30 (a–d) and 60 (e–h) days posttransplant (n ¼ 5/time point) were double immunostained for GFP (red, b,d,f and h) and GATA-4 (brown, a, c, e and g). Sections were counterstained with hematoxylin. Arrow, SC. Bar in e, 50 mm, for a, b, e and f; bar in g, 25 mm for c, d, g and h.

need to optimize the isolation, culture and composition of SC grafts differentially across syngeneic–allogeneic– xenogeneic barriers. In addition to these variables, future studies should also include a more detailed time-course analysis and large enough sample sizes to provide insight into the true variability associated with each approach. These studies must also consider the potential contribution of the therapeutic protein secreted to the process of immunological protection. These reservations withstanding, the present data provide the first successful reported example of a discordant cell-based transplant approach to gene therapy that does not require significant chemical or physical manipulation of the recipient’s immune system. Gene therapy has tremendous untapped medical potential, but has been mired in several practical and ethical problems. The major issue with the most common forms of gene therapy is identifying a vector that can safely, reproducibly and efficiently transfect nondividing Gene Therapy

Animals One heterozygous male TgN(GFPU)5Nagy mouse and one wild-type strain female mouse were purchased from the Jackson Laboratory (Bar Harbor, ME, USA) as a breeding pair. TgN(GFPU)5Nagy mice created by germline transmission of a 129 ES line (H-2b haplotype) were maintained by crossing with ICR (outbred) mice.35 Male TgN(GFPU)5Nagy offspring were identified by the analysis of tail biopsies using a Zeiss fluorescent microscope and used as SC donors. Male SCID (Taconic Farms) and Balb/c (H-2d haplotype; University of Alberta, Health and Laboratory Animal Services) mice, aged 6–8 weeks, were used as recipients. SC isolation and characterization SCs were isolated using a technique similar to that previously described for rat SCs.9. Briefly, testicles from adult male TgN(GFPU)5Nagy transgenic mice were surgically removed and placed in a 50 ml conical tube containing cold (41C) Hank’s balanced salt solution (HBSS) supplemented with 0.25% (w/v) fraction V bovine serum albumin (BSA; Sigma Chemical Co., St Louis, MO, USA). The testes were cut into 1-mm fragments with scissors, digested for 6 min at 371C with collagenase Type V (2.5 mg/ml; Sigma) and then washed three times with HBSS. The tissue was resuspended in calcium-free medium supplemented with 1 mM EGTA and further digested with trypsin (25 mg/ml; Boehringer Mannheim, Laval, Canada) and DNase (4 mg/ml, Boehringer) for 10 min at 371C. The digest was passed


through a 500-mm nylon mesh, washed with HBSS and cultured in nontreated Petri dishes (10 cm diameter) containing Ham’s F10 media supplemented with 10 mmol/l D-glucose, 2 mmol/l L-glutamine, 50 mmol/l isobutylmethylxanthine, 0.5% BSA, 10 mmol/l nicotinamide, 100 U/ml penicillin, 100 mg/ml streptomycin and 10% fetal calf serum. Cells were incubated for 48 h at 371C to allow the formation of SC aggregates (100– 300 mm diameter) before transplantation under the kidney capsule.

SC transplantation Prior to transplantation, the purity and number of SCs was determined. Specifically, we assessed the number of GFP- or GATA-4-positive SCs and smooth muscle aactin-positive peritubular myoid cells in a representative aliquot after dissociation of the cell aggregates using techniques described previously.24 In each preparation a minimum of 500 single cells were counted. To determine the number of SCs isolated, three representative aliquots of the cell suspension were measured for total cellular DNA content using a Hoefer DyNa Quant 200 fluorometric assay (Amersham Pharmacia Biotech, San Francisco, CA, USA). Aliquots were washed with citrate buffer (150 mmol/l NaCl, 15 mmol/l citrate, 3 mmol/l EDTA, pH 7.4), resuspended in TNE buffer (10 mM Tris, 0.2 mM NaCl, 1 mM EDTA, pH 7.4) and sonicated. Aliquots of 10 ml were assayed in triplicate by diluting them in 2 ml of assay solution (0.1 mg/ml Hoechst 33258 in 1 TNE) and measuring fluorescence (365 nm excitation/460 nm emission). A sixpoint (0–500 ng/ml) DNA standard curve was generated using calf thymus DNA. For transplantation, aliquots consisting of 12.6877.95 million SCs as aggregates (6.6 pg DNA/cell) were aspirated into polyethylene tubing (PE-50), pelleted by centrifugation and gently placed under the left renal subcapsular space of Halothane-anesthetized SCID (n ¼ 6) and Balb/c (n ¼ 10) mice.24 Diabetic mice Balb/c mice were rendered diabetic by intraperitoneal injection of streptozotocin (Streptozotocin, Sigma, St Louis, MO, USA) at a dose of 275 mg/kg body weight. Diabetes was confirmed by the presence of hyperglycemia and only those mice with nonfasting blood glucose levels exceeding 20 mM were used as recipients. Islets were isolated from TgN(GFPU)5Nagy mice by collagenase (Sigma, Type V) digestion, dextran gradient purification and subsequently cultured 24 h prior to transplantation of 500 TgN(GFPU)5Nagy mouse islets underneath the renal capsule of diabetic Balb/c mice.9 After transplantation, nonfasting blood glucose levels were measured twice a week and graft rejection was defined as two successive blood glucose levels 414 mM. Graft characterization Nephrectomies were performed at 30 and 60 days posttransplant and the graft-bearing kidneys were examined immediately using blue-light fluorescent microscopy for GFP-expressing cells underneath the kidney capsule. Kidneys were then immersed in Z-fix and embedded in paraffin. After deparaffinization, and rehydration, sections were heated for 15 min in 0.01 M sodium citrate buffer (pH 6.0), in a microwave at full power.9,24 Sections

were incubated with 10% hydrogen peroxide to quench endogenous peroxidases, blocked with 1% BSA and single stained for GATA-4, or GFP or double stained for GATA-4 and GFP. For GATA-4, sections were incubated with goat polyclonal anti-GATA-4 (1:50; Santa Cruz Biotechnology, Santa Cruz, CA, USA) primary antibody for 30 min, followed by biotinylated horse anti-goat secondary antibody for 20 min and diaminobenzamide as the chromagen using an ABC kit (Vector Laboratories, Burlingame, CA, USA). For GFP, sections were incubated with mouse monoclonal anti-GFP (1:100; Chemicon International, Inc., Temecula, CA, USA) primary antibody for 30 min, followed by Cy3-conjugated goat antimouse secondary antibody (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA) for 20 min. GFP-positive cells were only detected in the grafts. No GFP-positive cells were present in the kidney.

699

Acknowledgements This work was funded by the Alberta Heritage Foundation for Medical Research (AHFMR), Canadian Diabetes Association (CDA), Canadian Institutes of Health Research, a collaborative grant from the Juvenile Diabetes Research Foundation International (JDRF) and Sertoli Technologies, Inc., Edmonton Civic Employees’ Charitable Assistance Fund and MacLachlan Fund. GSK is a senior scholar of the AHFMR and recipient of a scholarship from the CDA and a Career Development Award from the JDRF. JMD is a recipient of a Postdoctoral Fellowship from the JDRF and the Izaak Walton Killiam Memorial. We acknowledge the technical assistance of Daniel Bruch, Deb Dixon, Lynnette Elder, Crystal Harris and Monique Tourand.

References 1 Barker CF, Billingham RE. Immunologically privileged sites. Adv Immunol 1977; 25: 1–54. 2 Gores PF et al. Long-term survival of intratesticular porcine islets in non-immunosuppressed beagles. Transplantation 2003; 75: 613–618. 3 Selawry HP. Islet transplantation to immunoprivileged sites. In: Lanza RP, Chick WL (eds) Pancreatic Islet Transplantation Vol. II Immunomodulation of Pancreatic Islets. R.G. Landes Co.: Austin, TX, 1994, pp 75–86. 4 Streinlein JW. Unraveling immune privilege. Science 1995; 270: 1158–1159. 5 Cameron DF, Whittington K, Schultz RE, Selawry HP. Successful islet/abdominal testis transplantation does not require Leydig cells. Transplantation 1990; 50: 649–653. 6 Selawry HP, Whittington KB. Prolonged intratesticular islet allograft survival is not dependent on local stereoidogenesis. Horm Metab Res 1988; 20: 562–565. 7 Whitmore III WF, Karsh L, Gittes RF. The role of germinal epithelium and spermatogenesis in the privileged survival of intratesticular grafts. J Urol 1985; 134: 782–786. 8 Bellgrau D et al. A role for CD95 ligand in preventing graft rejection. Nature 1995; 377: 630–632. 9 Korbutt GS, Elliott JF, Rajotte RV. Cotransplantation of allogeneic islets with allogeneic testicular cell aggregates allows long-term graft survival without systemic immunosuppression. Diabetes 1997; 46: 317–322. 10 Selawry HP, Cameron DF. Sertoli cell-enriched fractions in successful islet cell transplantation. Cell Transplant 1993; 2: 123–129. Gene Therapy


700 11 Yang H, Wright Jr JR. Co-encapsulation of Sertoli enriched testicular cell fractions further prolongs fish-to-mouse islet xenograft survival. Transplantation 1999; 67: 815–820. 12 Dufour JM et al. Long-term survival of neonatal porcine Sertoli cells in non-immunosuppressed rats. Xenotransplantation 2003; 10: 577–586. 13 Saporta S, Cameron DF, Borlongan CV, Sanberg PR. Survival of rat and porcine Sertoli cell transplants in the rat striatum without cyclosporine-A immunosuppression. Exp Neurol 1997; 146: 299–304. 14 Dufour JM, Rajotte RV, Kin T, Korbutt GS. Immunoprotection of rat islet xenografts by co-transplantation with Sertoli cells and a single injection of antilymphocyte serum. Transplantation 2003; 75: 1594–1596. 15 Kin T, Rajotte RV, Dufour JM, Korbutt GS. Development of an immunoprivileged site to prolong allograft survival. Cell Transplant 2002; 11: 547–552. 16 Korbutt GS et al. Testicular Sertoli cells exert both protective and destructive effects on syngeneic islet grafts in non-obese diabetic mice. Diabetologia 2000; 43: 474–480. 17 Sanberg PR, Borlongan CV, Saporta S, Cameron DF. Testisderived Sertoli cells survive and provide localized immunoprotection for xenografts in rat brain. Nat Biotech 1996; 14: 1692–1695. 18 Suarez-Pinzon W et al. Testicular Sertoli cells protect islet b-cells from autoimmune destruction in NOD mice by a transforming growth factor-b1-dependent mechanism. Diabetes 2000; 49: 1810–1818. 19 Willing AE et al. Sertoli cells decrease microglial response and increase engraftment of human hNT neurons in the hemiparkinsonian rat striatum. Brain Res Bull 1999; 48: 441–444. 20 Kay MA et al. In vivo hepatic gene therapy: complete albeit transient correction of factor IX deficiency in hemophilia B dogs. Proc Natl Acad Sci USA 1994; 91: 2353–2357. 21 Romano G, Micheli P, Pacilio C, Giordano A. Latest developments in gene transfer technology: achievements, perspectives, and controversies over therapeutic applications. Stem Cells 2000; 18: 19–39.

Gene Therapy

22 Thomas CE, Ehrhardt A, Kay MA. Progress and problems with the use of viral vectors for gene therapy. Nat Rev Gentet 2003; 4: 346–358. 23 Shapiro AMJ et al. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regime. N Engl J Med 2000; 343: 230–238. 24 Dufour JM, Rajotte RV, Korbutt GS. Development of an in vivo model to study testicular morphogenesis. J Androl 2002; 23: 635–644. 25 Chakraborty J. Conditions adversely affecting Sertoli cells. In: Russell LD, Griswold MD (eds) The Sertoli Cell. Cache River Press: Clearwater, FL, 1993, pp 578–579. 26 Stripecke R et al. Immune response to green fluorescent protein: implications for gene therapy. Gene Therapy 1999; 6: 1305–1312. 27 Marshall E. Gene therapy on trial. Science 2000; 288: 951–957. 28 Giraldo P, Montoliu L. Size matters: use of YACs, BACs and PACs in transgenic animals. Transgenic Res 2001; 10: 83–103. 29 Papaioannou VE. The coming of age of the transgenic era. Int J Dev Biol 1998; 42: 841–846. 30 Houdebine LM. The production of pharmaceutical proteins from the milk of transgenic animals. Reprod Nutr Dev 1995; 35: 609–617. 31 Illmensee K. Biotechnology in reproductive medicine. Differentiation 2002; 69: 167–173. 32 Lai L et al. Transgenic pig expressing the enhanced green fluorescent protein produced by nuclear transfer using colchicine-treated fibroblasts as donor cells. Mol Reprod Dev 2002; 62: 300–306. 33 Levy JH, Weisinger A, Ziomek CA, Echelard Y. Recombinant antithrombin: production and role in cardiovascular disorder. Semin Thromb Hemost 2001; 27: 405–416. 34 Pollock DP et al. Transgenic milk as a method for the production recombinant antibodies. J Immunol Methods 1999; 231: 147–157. 35 Hadjantonakis AK et al. Generating green fluorescent mice by germline transmission of green fluorescent ES cells. Mech Dev 1998; 76: 79–90.