Stem Cells, Vol. 14, No. 2, 232-238,
March 1996
© 1996 AlphaMed Press
Long-Term Expression of Human Growth Hormone (hGH) in Mice Containing Allogeneic Yolk Sac Cell Derived Neovascular Implants Expressing hGH
Yanzhang Wei,
Jinhua Li,
Thomas E. Wagner
Edison Biotechnology Institute, Molecular and Cellular Biology Program and Department of Biological Science, Ohio University, Athens, Ohio, USA
Key Words. Yolk sac cell • Human growth hormone • Gene delivery • Gene therapy • Allogeneic mice • Rat-anti-mouse B7.2 antibody
Dr. Thomas E. Wagner, Edison Biotechnology Institute of Ohio University, Athens, OH 45701, USA.
 |
Abstract
|
|---|
We have established a systemic gene delivery animal model system by using cultured murine embryonic yolk sac cells, which can be easily genetically modified in vitro and participate in angiogenesis in vivo when basement membrane proteins (Matrigel) are provided in syngeneic mice. In the present study, we successfully applied this system to allogeneic mice. In order to suppress donor cell-specific immune responses, the costimulatory signal transduction pathway of T cell activation was blocked by treating the recipient allogeneic C57BL/6 mice with rat-antimouse B7.2 antibody. As a result of this suppression, human growth hormone, the therapeutic gene product, could be detected for over 340 days, while it could only be detected in mice treated with rat-IgG2a, the isotype control of anti-B7.2, for fewer than 50 days. This is the first ex vivo gene delivery system that can express a therapeutic gene product, long-term, in an allogeneic host.
|
Introduction
|
|---|
Different primary cells, including skin fibroblast cells [1, 2], keratinocytes [3–5], epithelial cells [6], endothelial cells [7] and myoblast cells [8–10], have been intensively tested as target cells for gene therapy in ex vivo systems. However, problems with these primary cell-based gene therapies exist in access of cellular material, culture in vitro, transfection and reimplantation. Recently, we have developed a new ex vivo gene delivery animal model system by using cultured murine yolk sac cells. The therapeutic gene product in our system, human growth hormone (hGH), was expressed for more than seven months in syngeneic mice [11]. But practical utilization of this system in human gene therapy would require the use of allogeneic donor yolk sac cells. Although these cultured murine yolk sac cells do not express major histocompatibility complex (MHC) in vitro, our attempts to express hGH, long-term, in allogeneic mice failed. This failure suggests that an immune response by the host to the endothelial decedents of the embryonic yolk sac cellular graft occurred in allogeneic recipients of this cell-based gene therapy system.
One approach to prevent the rejection of donor cells is to make the host immune system unresponsive by treatment with panimmunosuppressive drugs, such as cyclosporine A or monoclonal antibodies to CD3. However, these drugs must be taken frequently during the life of the individual, depress the immune system, and often result in increased infections and cancer. Recently, it has been found that the binding of T cell surface CD28 or CTLA4 by the ligands, B7s (B7.1 and B7.2), during T cell receptor engagement is critical for proper T cell signaling [12–15]. When the interaction of CD28 with its ligand is blocked, antigen-specific T cells are inappropriately induced into a state of antigen-specific T cell anergy [16]. In one study, treatment of mice with CTLA4 Ig, a recombinant ligand of B7s, dramatically prolonged the persistence of xenogeneic human pancreatic islet grafts [17].
In this study, in order to make the host mice specifically unresponsive to allogeneic yolk sac cells, antibodies against B7.2, one of the CD28 ligands that are expressed on antigen-presenting cells, were used to block the CD28-B7 costimulatory signal transduction pathway. After this treatment, the costimulatory pathway of T cell signaling was successfully blocked and, consequently, the therapeutic gene product can be detected in the allogeneic host for over 340 days.
|
Materials and Methods
|
|---|
Cells and Animals
Cells used in this study were long-term cultured murine yolk sac cells, YS4, and cells genetically modified from YS4 with cytomegalovirus immediate early promoter/human growth hormone-fusing gene, GH-YS4. The genetic modification and routine maintenance of these cells have been described previously [11]. Mice used in this study were C57BL/6 mice purchased from Jackson Laboratory (Bar Harbor, ME) at 8 to 10 weeks of age.
Antibodies
All of the following antibodies were purchased from Pharmigen (San Diego, CA): purified rat IgG2a, fluorescein isothiocyanate (FITC)-rat IgG2a, FITC-CD3 (
), FITC-hamster IgG, FITC-L3T4 (CD4), FITC-Ly-2 (CD8a) and FITC-anti-B7.2. For the blocking experiment, purified rat antimouse B7.2 antibody was prepared in-house. Briefly, the supernatant of the hybridoma cell culture, GL-1 (HB-253, American Type Culture Collection [ATCC]), was harvested and precipitated with saturated ammonium sulfate. The centrifuged pellet was dissolved in phosphate buffered saline (PBS) solution and dialyzed for 48 h at 4° C against a >20 time volume of PBS with four solution exchanges. The dialyzed antibody-containing solution was then purified by running it through an IgG-specific protein A column (Avidchrome-Protein A kit, Sigma; St. Louis, MO). The concentration of this antibody solution was determined by spectrophotometer at A280 by comparison to purified rat antimouse B7.2 (Pharmigen).
Antibody Administration and Implantation of GH-YS4 Cells
Considering the date that GH-YS4 cells were implanted as day 0, at day -1, 0, 1, 3, 5 and 7, eight C57BL/6 mice were each i.p. injected with 50 µg of purified rat antimouse B7.2 antibody in 0.2 ml of PBS, and four C57BL/6 mice were each i.p. injected with 50 µg of purified rat IgG as controls. At day 0, all of the above 12 mice and other four untreated C57BL/6 mice were each s.c. implanted with 5 x 106 GH-YS4 cells in 0.5 ml of Matrigel as described previously [11].
Fluorescence-Activated Cell Sorting (FACS) Analysis
In order to know whether the administration of rat-antimouse B7.2 antibody blocks B7.2 antigen expression, a 50 µl blood sample was collected respectively from a normal C57BL/6 mouse, a C57BL/6 mouse implanted with GH-YS4 cells for seven days and injected with purified isotype control rat IgG2a six times, and a C57BL/6 mouse implanted with GH-YS4 cells for seven days and injected with rat-antimouse B7.2 antibody six times. These blood samples were washed with 1 ml of PBS twice and divided into two parts, one stained with FITC-conjugated rat-antimouse B7.2 antibody and the other stained with FITC-conjugated rat IgG2a as control. In order to determine whether the antibody administration influences the T cell population, a 200 µl blood sample was collected from one C57BL/6 mouse implanted with GH-YS4 cells and injected with purified rat-IgG2a for two months, and one C57BL/6 mouse implanted with GH-YS4 cells and injected with rat-antimouse B7.2 antibody for two months, respectively. Each sample was washed with PBS twice, divided into four parts, and stained with FITC-labeled rat IgG2a as an isotype control (the isotype control for the CD3 antibody was FITC-labeled hamster IgG), FITC-labeled rat-antimouse CD3 antibody, FITC-labeled rat-antimouse CD4 antibody and FITC-labeled rat-antimouse CD8a antibody, respectively. After staining, all of the red blood cells in the stained blood samples were lysed with 1x lysing solution (Becton Dickinson; San Jose, CA) for 10 min in the dark and then washed with staining buffer (0.1 g sodium azide and 0.1 g bovine serum albumin in 100 ml of PBS) twice. The samples were then analyzed using a Becton Dickinson FACsort (Mansfield, MA) equipped with a Lysis II program (Becton Dickinson; San Jose, CA).
Radioimmunoassay (RIA)
A 100 µl serum sample was made from each experimental mouse at different scheduled times. A Tandem-R HGH solid phase II-site RIA kit (Hybritech, Inc.; San Diego, CA) was used in this study to measure the hGH levels in the serum samples according to the procedures provided by the kit. The radioactivity was counted using a 
-counter (5500, Backman; Novi, MI). The concentration of hGH was calculated according to a standard curve made from the standard hGH samples provided with the kit.
|
Results
|
|---|
Anti-B7.2 Antibody Administration Blocked B7.2 Antigen Expression
The total leukocytes of normal C57BL/6 mice express small amounts of the B7.2 antigen (Fig. 1A
). GH-YS4 cell implantation stimulates more B7.2 antigen expression (Fig. 1B), suggesting that GH-YS4 cells can stimulate an immune response in C57BL/6 mice. However, as Figure 1C shows, after anti-B7.2 antibody i.p. injection, no B7.2 antigen could be detected by FACS on whole leukocytes from the GH-YS4-implanted mice expressing hGH (see below).
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 1. Blockage of B7.2 antigen expression. At the end of the antibody administration (seven days after the GH-YS4 cell implantation), B7.2 antigen expression on total leukocytes of normal C57BL/6 mouse (A); C57BL/6 mouse implanted with GH-YS4 cells (B); and C57BL/6 mouse implanted with GH-YS4 cells and treated with rat-antimouse B7.2 antibody (C) was analyzed by FACS. The isotype controls were FITC-rat IgG2a.
|
|
Anti-B7.2 Antibody Injection Decreases T Lymphocyte Proliferation
In order to know whether the B7.2 antigen blockage influences T lymphocyte production, a number of different subtypes of T lymphocytes in non-B7.2 antigen-blocked, GH-YS4 cell-implanted mice (Fig. 2A, 2C, 2E) or B7.2 antigen-blocked, GH-YS4 cell-implanted mice (Fig. 2B, 2D, 2F) were analyzed by FACS. The results indicate that the numbers of all of the CD3+ cells (Fig. 2A, 2B), CD4+ cells (Fig. 2C, 2D) and CD8a+ cells (Fig. 2E, 2F) among total leukocytes decreased two months after the antimouse B7.2 antibody treatment.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 2. Analysis of T cells from rat IgG2a-treated or rat-antimouse B7.2 antibody-treated C57BL/6 mice. Two months after GH-YS4 cell implantation, 200 µl blood samples were collected from a rat IgG2a-treated mouse that had stopped expressing hGH (A, C and E) or from a rat-antimouse B7.2-treated mouse that was still expressing hGH (B, D and F). Different subtypes of T cells, CD3+ T cells (A and B), CD4+ T cells (C and D), and CD8a+ T cells (E and F) were analyzed by FACS.
|
|
Sustained Expression of hGH in Antimouse B7.2 Antibody-Treated Allogeneic Mice
Although our systemic gene delivery system can express hGH in syngeneic mice for more than seven months [11], when applied to allogeneic mice, it can only express hGH for fewer than 50 days (Fig. 3A). However, after rat-antimouse B7.2 antibody treatment, the expression of hGH in allogeneic mice was markedly prolonged (Fig. 3B). Five out of eight C57BL/6 mice implanted with GH-YS4 cells and treated with rat-antimouse B7.2 antibody expressed hGH in their sera for more than 340 days. This prolonged hGH expression was due to the treatment of rat-antimouse B7.2 antibody because the isotype (rat IgG) treatment had no effect on the hGH expression (Fig. 3C) compared to the nontreated C57BL/6 mice (Fig. 3A).
View larger version (20K):
[in this window]
[in a new window]
|
Fig. 3. hGH levels in GH-YS4 cells implanted C57BL/6 mice. A) Normal C57BL/6 mice without antibody treatment; B) C57BL/6 mice treated with rat-antimouse B7.2 antibody; C) C57BL/6 mice treated with rat-IgG2a. After GH-YS4 cell implantation at different scheduled dates, the hGH levels of both rat IgG2a-treated and rat-antimouse B7.2 antibody-treated mice were measured by RIA.
|
|
|
Discussion
|
|---|
Murine embryonic yolk sac cells have the potential to differentiate into both hematopoietic cells (hematopoiesis) and endothelial cells (vasculogenesis) [11, 18]. We have taken advantage of the vasculogenesis of murine embryonic yolk sac cells by using them as an ex vivo systemic gene delivery system for gene therapy [11]. There are at least two principal advantages of this system for gene therapy. Upon introduction into the host animal together with Matrigel, the genetically modified yolk sac cells (GH-YS4) take part in new blood vessel formation in basement membrane protein (Matrigel) implants. Our previous work demonstrated that the new blood vessels formed in the Matrigel-GH-YS4 cell implant and the GH-YS4 cells participated in the new blood vessel formation [11]. Therefore, the cellular platform expressing the therapeutic gene product is the endothelial lining of an aspect of the circulatory system of the recipient of this cellular implant providing highly efficient delivery of the gene product systemically throughout the host. Also, since the yolk sac cells used in our system are early embryonic cells which do not initially express MHC class I or II molecules [19], the induction of allogeneic host tolerance to this yolk sac cell transplantation may be facilitated. If such tolerance induction proved possible, a universal yolk sac cell-based gene delivery system would have significant advantages over other ex vivo systems currently used which require a syngeneic cellular platform derived from the host. Therefore, in this study, we tried to apply this system to allogeneic recipient mice.
Even after culture on Matrigel-coated flasks for 24 h, during which most of the GH-YS4 cells formed organized linear structures, they still do not express MHC I or MHC II [unpublished data] surface markers. This might be the reason why Balb/c mouse-derived GH-YS4 cells can survive in allogeneic C57BL/6 mice for up to 50 days. However, the GH-YS4 cells only persisted in C57BL/6 mice for fewer than 50 days (Fig. 3A
), suggesting that the implantation of GH-YS4 cells did stimulate an immune response in C57BL/6 mice. Therefore, we attempted to suppress this immune response. One approach to prevent rejection of donor cells is to make the host immune system unresponsive by treatment with panimmunosuppressive drugs, such as cyclosporine A or monoclonal antibodies to CD3. However, the use of such drugs must be continued to assure maintenance of the allograft, and continued panimmunosuppression is always accompanied by increased susceptibility to infection and oncogenesis. Therefore, a more selective induction of graft tolerance by induced T cell anergy was attempted.
Recent immunological studies demonstrated that in order to be functionally activated, T helper cells (CD4+) in both humans and mice must be signaled not only by the T cell receptor for antigen/MHC pathway but also the CD28(CTLA-4)/B7 pathway, the so-called costimulatory pathway [20]. Blockage of this costimulatory pathway results in T cell anergy [21–23]. In this study, we blocked the costimulatory pathway by using rat-antimouse B7.2 antibody treatment to induce anergy in T cells specific to donor yolk sac-derived endothelial cells. Our results indicated that without stimulation, C57BL/6 leukocytes express a small amount of B7.2 antigen (Fig. 1A), while after implantation of GH-YS4 cells, the B7.2 antigen expression increased (Fig. 1B). After administration with rat-antimouse B7.2 antibody, no functional B7.2 antigen could be detected (Fig. 1C). These results are similar to those obtained by Linsley et al. [24]. Our results also indicate that the blockage of this costimulatory pathway affects T cell proliferation in vivo. The number of CD3 T cells, CD4 T cells and CD8a T cells significantly decreased after antibody treatment. In vitro studies by Hathocock et al. showed that anti-B7.2 antibody (GL-1) treatment dramatically decreased the T cell proliferative rate [25]. Our previous work in syngeneic mouse experiments demonstrated that hGH expression in mice did not affect T cell proliferation [unpublished data].
Most importantly, the administration of rat-antimouse B7.2 antibody markedly prolonged the expression of the therapeutic gene product, hGH, in allogeneic C57BL/6 mice (Fig. 3B). Five of eight experimental mice treated with rat-antimouse B7.2 antibody and implanted with GH-YS4 cells expressed hGH at the level of 1 ng/ml for over 340 days, while the control mice either untreated or treated with rat IgG isotype and implanted with GH-YS4 cells expressed hGH for fewer than 50 days (Fig. 3A, 3C). The short-term expression of hGH in the other three experimental mice (Fig. 3B) may be due to the short time of antibody administration (six times in nine days). By comparison, in a xenogeneic pancreatic islet implantation experiment, the recipient mice were treated with anti-B7 antibody with the dose we used in our study, but for 14 days with daily injections [17]. The drop of hGH from 50 ng/ml during the days immediately after implantation to about 1 ng/ml at 100 days after implantation can be explained as follows: at the early time after implantation, all of the implanted cells secrete hGH. As time goes on, cells inside the gel implant apparently die because of a shortage of nutrients and only the cells at the surface of the gel implant remain alive, successfully differentiate and form blood vessels. This has been confirmed by the observation that, after excision of the implant, neovascularization can only be observed on the surface of the gel implant.
To date, the longest therapeutic gene expression in an animal model has been achieved by Heartlein et al. [26] and is over 500 days. However, this was achieved in severe combined immunodeficiency mice by kidney capsule injection with genetically modified rabbit fibroblasts. The present study is the first gene therapy model to express a therapeutic gene product in normal allogeneic mice for over 340 days.
|
Acknowledgments
|
|---|
This research was funded by the Ohio Department of Development's Thomas Edison program, a grant from the National Institute of Allergy and Infectious Disease (5-RO1-AI33280) and by a research contract sponsored by Progenitor, Inc.
|
Footnotes
|
|---|
Provisionally accepted December 21, 1995.
|
References
|
|---|
-
St. Louis D, Verma IM. An alternative approach to somatic cell gene therapy. Proc Natl Acad Sci USA 1988;85:3150–3154.[Abstract/Free Full Text]
-
Palmer TD, Rosman GJ, Osborne WRA et al. Genetically modified skin fibroblasts persist long after transplantation but gradually inactivate introduced genes. Proc Natl Acad Sci USA 1991;88:1330–1334.[Abstract/Free Full Text]
-
Fenjves ES, Gordon DA, Pershing LK et al. Systemic distribution of apolipoprotein E secreted by grafts of epidermal keratinocytes: implication for epidermal function and gene therapy. Proc Natl Acad Sci USA 1989;86:8803–8807.[Abstract/Free Full Text]
-
Morgan JR, Barrandon Y, Green H et al. Expression of an exogenous growth hormone gene by transplantable human epidermal cells. Science 1987;237:1476–1479.[Abstract/Free Full Text]
-
Gerrard AJ, Hudson DL, Brownlee GG et al. Towards gene therapy for haemophilia B using primary human keratinocytes. Nat Genet 1993;3:180–183.[Medline]
-
Smith GH, Gallahan D, Zwiebel JA et al. Long-term in vivo expression of genes introduced by retrovirus-mediated transfer into mammary epithelial cells. J Virol 1991;65:6365–6370.[Abstract/Free Full Text]
-
Zwiebel JA, Freeman SM, Kantoff PW et al. High level recombinant gene expression in rabbit endothelial cells transduced by retroviral vectors. Science 1989;243:220–243.[Abstract/Free Full Text]
-
Dai Y, Roman M, Naviaux RK et al. Gene therapy via primary myoblasts: long-term expression of factor IX protein following transplantation in vivo. Proc Natl Acad Sci USA 1992;89:10892–10895.[Abstract/Free Full Text]
-
Lynch CM, Clowes MM, Osborne WRA et al. Long-term expression of human adenosine deaminase in vascular smooth muscle cells of rats: a model for gene therapy. Proc Natl Acad Sci USA 1992;89:1138–1142.[Abstract/Free Full Text]
-
Yao SN, Kurachi K. Expression of human factor IX in mice after injection of genetically modified myoblast. Proc Natl Acad Sci USA 1992;89:3357–3361.[Abstract/Free Full Text]
-
Wei YZ, Quertermous T, Wagner TE. Directed endothelial differentiation of cultured embryonic yolk sac cells in vivo provides a novel cell-based system for gene therapy. STEM CELLS 1995;13:541–547.[Abstract]
-
Jenkins MK, Taylor PS, Norton SD et al. CD28 delivers a costimulatory signal involved in antigen-specific IL-2 production by human T cells. J Immunol 1991;147:2461–2466.[Abstract/Free Full Text]
-
June CH, Ledbetter JA, Linsley PS et al. Role of the CD28 receptor in T-cell activation. Immunol Today 1990;11:211–216.[Medline]
-
Reiser H, Freeman GJ, Razi-Wolf Z et al. Murine B7 antigen provides an efficient costimulatory signal for activation of murine T lymphocytes via the T-cell receptor/CD3 complex. Proc Natl Acad Sci USA 1992;89:271–275.[Abstract/Free Full Text]
-
Damle NK, Klussman K, Linsley PS et al. Differential costimulatory effects of adhesion molecules B7, ICAM-1, LFA-3, and VCAM-1 on resting and antigen-primed CD4+ T lymphocytes. J Immunol 1992;148:1985–1992.[Abstract]
-
Harding FA, McArthur JG, Gross JA et al. CD28-mediated signalling co-stimulates murine T cells and prevents induction of anergy in T-cell clones. Nature 1992;356:607–609.[Medline]
-
Lenschow DJ, Zeng Y, Thistlethwaite JR et al. Long-term survival of xenogeneic pancreatic islet grafts induced by CTLA4Ig. Science 1992;257:789–792.[Abstract/Free Full Text]
-
Palis J, McGrath KE, Kingsley PD. Initiation of hematopoiesis and vasculogenesis in murine yolk sac explants. Blood 1995;86:156–163.[Abstract/Free Full Text]
-
Barut BA, Antczak M, Winnard A et al. Isolation and characterization of endothelial cell lines from murine yolk sac. Blood 1994;84(suppl 1):74.[Abstract/Free Full Text]
-
Abbas AK, Lichtman AH, Pober JS. Cellular and Molecular Immunology, 2nd ed. Philadelphia, PA: W.B. Saunders Company, 1994:163-164.
-
Freeman GJ, Borriello F, Hodes RJ et al. Uncovering of functional alternative CTLA-4 counter-receptor in B7-deficient mice. Science 1993;262:907–909.[Abstract/Free Full Text]
-
Freeman GJ, Gribben JG, Boussiotis VA et al. Cloning of B7-2: a CTLA-4 counter-receptor that costimulates human T cell proliferation. Science 1993;262:909–911.[Abstract/Free Full Text]
-
Kay MA, Holterman AX, Meuse L et al. Long-term hepatic adenovirus-mediated gene expression in mice following CTLA4Ig administration. Nat Genet 1995;11:191–197.[Medline]
-
Linsley PS, Wallace PM, Johnson J et al. Immunosuppression in vivo by a soluble form of the CTLA-4 T cell activation molecule. Science 1992;257:792–795.[Abstract/Free Full Text]
-
Hathcock KS, Laszlo G, Dickler HB et al. Identification of an alternative CTLA-4 ligand costimulatory for T cell activation. Science 1993;262:905–907.[Abstract/Free Full Text]
-
Heartlein MW, Roman VA, Jiang JL et al. Long-term production and delivery of human growth hormone in vivo. Proc Natl Acad Sci USA 1994;91:10967–10971.[Abstract/Free Full Text]
Received November 9, 1995;
accepted for publication January 8, 1996.
Top
Abstract
Introduction
