Stem Cells, Vol. 17, No. 2, 117-120,
March 1999
© 1999 AlphaMed Press
Recombinant Adeno-Associated Virus-Based Vectors Provide Short-Term Rather Than Long-Term Transduction of Primitive Hematopoietic Stem Cells
Ronald van Osa,
Hava Avrahamb,
Naheed Banub,
Peter M. Maucha,
Jennifer Whaterc,
Yangming Yangc,
Bin Duc
a Joint Center for Radiation Therapy, Department of Radiation Oncology, Harvard Medical School, Boston, MA;
b Division of Experimental Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA;
c Department of Infectious Diseases, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA
Key Words. Primitive stem cells • rAAV • Gene therapy • Transduction • Cobblestone-area-forming cells • Long-term repopulation
Dr. Ronald van Os, Department of Hematology, Leiden University Medical Center, Building 1, C2-R, PO Box 9600, 2300 RC Leiden, The Netherlands.
 |
Abstract
|
|---|
Bone marrow stem cells collected from B6-Gpi-1a mice pretreated with 5-fluorouracil were incubated for 2 h at 37°C in the presence of the recombinant adenovirus-associated virus-based vector (rAAV) SSV9. As measured in vitro immediately following transduction, SSV9 was found to be effective in transducing the primitive cobblestone-area-forming cell (CAFC)-35 subset (60% transduction efficiency). However, this did not predict long-term expression as the presence of the transgene could not be detected six months after transplantation of 1-2 x 106 transduced bone marrow stem cells into lethally irradiated recipients. CAFC analysis of bone marrow cells and Southern blot analysis of bone marrow and spleen cells were negative, and polymerase chain reaction analysis showed less than 0.1% transduction in bone marrow cells. Therefore, based on our study we conclude that rAAV transiently transduces hematopoietic stem cells but fails to integrate into the genome, leading to the loss of the reporter gene within the first six months after transplantation in vivo.
|
Introduction
|
|---|
The ability to successfully transfer genes into different cell types using viral vectors has created new prospects for the treatment of a variety of genetic diseases [1]. Of the many genetic diseases affecting humans, those which involve derivatives of hematopoietic stem cells appear to be amenable targets for gene therapy [2]. In the case of disorders of the hematopoietic system, primitive stem cells within the hematopoietic hierarchy need to be successfully transduced with the gene of interest to ensure a continuous supply of healthy cells. In various blood diseases, the genes causing deficient function of mature blood cell types have now been identified [3], allowing for their use in gene replacement therapy in autologous stem cells.
Over the last decade, it has been proposed that adeno-associated virus (AAV)-based vectors may have a potential advantage over the currently used retroviral vectors due to their nonpathogenicity, physical stability, and ability to also infect nondividing cells. Therefore, we decided to use a transduction protocol shown to be efficient at transducing nondividing neural cells [4] for the transduction of hematopoietic stem cells. Bone marrow cells transduced according to this protocol were tested for transduction efficiency immediately following gene transfer as well as six months after transplantation into lethally irradiated recipients. Our results indicate important differences among the various methods for measuring transduction efficiencies and show that transplantation in vivo is required to accurately determine the transduction efficiency of functional long-term repopulating (LTR) stem cells.
|
Materials and Methods
|
|---|
Congenic C57BL/6J-Gpi-1a/Gpi-1a (B6-Gpi-1a) mice (Jackson Laboratories; Bar Harbor, ME) were used as donors and C57BL/6J (B6-Gpi-1b) mice (Jackson Laboratories) as recipients. Bone marrow was collected three days after 5-fluorouracil (5-FU) administration by crushing tibiae and femora to prepare a single-cell suspension in sterile Hank's balanced salt solution (HBSS). In experiment 1, nucleated cells were counted in trypan blue on a hemocytometer, and then 10,000,000 viable cells were incubated on a layer of stromal cells (FBMD-1) for 2 h at 37°C. In experiment 2, 10,000,000 cells were incubated for 2 h in a waterbath at 37°C. Following transduction, cells were either used in the cobblestone-area forming cell (CAFC) assay or injected into the tail veins of the recipient mice. Recipient mice received 2 x 106 (experiment 1) or 106 (experiment 2) bone marrow cells 4-6 h after 9.5 Gy total-body irradiation.
The SSV9 vector used in this study was designed and constructed as described [4]. Following CsCl purification, the vector titer was calculated by determining the proportion of lacZ-positive (blue) cells using a standard staining procedure [4]. In this case, 107 functional transduction units per ml were found. In both experiments, bone marrow cells were incubated with 450 µl total volume of medium alone or 450 µl medium with vector. Thus, approximately 4.5 x 106 functional transduction units and 1 x 107 bone marrow cells were incubated for 2 h at 37°C, with (experiment 1) or without (experiment 2) a stromal support layer of FBMD-1 cells. Therefore, in our experiments, the multiplicity of infection (MOI) was 0.45. Cell viability after transduction was more than 90%, as determined by trypan blue exclusion.
In vitro determination of hematopoietic stem and progenitor cell frequencies was performed by limiting dilution analysis of CAFC in microcultures, according to the method described previously [5, 6]. Cobblestone areas (CA) were tested for the presence of the transgene by adding 5 µl X-gal (Sigma; St. Louis, MO) per well to the medium and incubating the plates at 37°C for 3-4 h. Subsequently, the CA were individually analyzed for ß-galactosidase activity. We calculated two CAFC frequencies, one without X-gal staining (total CAFC) and one where only lacZ-positive CAFC colonies were used (negative CAFC were excluded) to calculate the frequency of positive CAFC. Only homogeneously stained CAFC colonies were considered positive for lacZ. The frequency of positive CAFC divided by the total CAFC is a measure of the transduction efficiency.
Transduced bone marrow cells (Gpi-1a) were transplanted into lethally irradiated recipients (Gpi-1b) to measure the transduction efficiency of the LTR stem cells and the persistence of gene expression. At six months following transplantation of transduced bone marrow, recipients were sacrificed for analysis of donor chimerism and expression of lacZ in bone marrow and spleen as described [4, 7].
|
Results
|
|---|
In experiment 1, transduction was performed on a stromal support layer containing FBMD-1 cells. Transduction efficiency was determined by staining CA with X-gal to visualize the enzymatic activity of ß-galactosidase. Early-appearing CAFC-7 showed a considerable transduction efficiency of about 35%. Of the late-appearing CAFC-35, 60% were positive for lacZ (Table 1). In experiment 2, 57% of CAFC-7 were lacZ-positive, as were 38% of CAFC-35. This means that all subsets could be efficiently transduced with SSV9-lacZ and that lacZ activity could be detected in vitro up to five weeks following transduction.
Six months after transplantation of transduced cells into lethally irradiated recipients, bone marrow cells were analyzed for the presence of the lacZ gene. No significant ß-galactosidase activity could be detected in stem cells assayed at this time. Both early- and late-appearing CAFC showed 0% transduction efficiency (Table 2). At six months following transplantation of the transduced cells, the bone marrow and spleens of all mice were analyzed for the presence of the vector. Southern blot analysis was performed to reveal whether the gene was integrated into the genome. However, no vector could be detected in the bone marrow of bone marrow transplantation recipients (data not shown). Polymerase chain reaction with vector-specific primers showed the presence of the transgene at very low levels (0.1% of the cells). Thus, the vectors were not integrated into the genome of the long-term repopulating stem cells. An alternative explanation might be that lacZ causes a growth disadvantage for transduced cells, as shown for endothelial cells and hematopoietic cells transduced with retroviral lacZ vectors [8, 9]. This may also lead to the loss of the transgene after transplantation.
|
Discussion
|
|---|
The results presented in this paper indicate that a previously efficient transduction protocol for rat neurons was not successful at transducing LTR stem cells in the bone marrow. Failure to integrate seems to be the major obstacle for gene therapy using adenovirus-associated vectors in non-dividing hematopoietic stem cells. We also observed a discrepancy between measuring transduction efficiency in vitro immediately after transduction and in vivo following transplantation of transduced cells into lethally irradiated mice. This difference may be explained by the maintenance of the vector in episomes. Since late-appearing CAFC seemed to be transduced efficiently with the SSV9 vector when tested immediately following transduction, the loss of lacZ activity in vivo strongly suggests that the vector failed to integrate. However, it could be possible that the ß-galactosidase enzyme was produced for a limited time.
The CAFC assay was designed as an in vitro alternative for assays of hematopoietic stem cells in vivo. In the murine system, it was shown that early-appearing CAFC represent short, transient, repopulating stem cells equivalent to colony-forming units-spleen whereas late-appearing CAFC (day 28 and later) are representative of LTR stem cells [5]. Recently, it was shown that upon transduction with retroviral vectors containing the gene for the enhanced green fluorescent protein, the transduction efficiency of in vivo LTR cells could be accurately predicted by the frequency of CAFC day 35 [10]. Similarly, retroviral transduction of human CAFC week 6 correlated well with transduction of human NOD/SCID repopulating stem cells [11]. The lack of predictability of the CAFC assay in our studies thus seems to be related to the use of AAV-based vectors that can function in episomes for the duration of the CAFC assay (five weeks), but when no integration occurs, the transgene is eventually lost. Therefore, we can conclude that with AAV-based vectors, transduction of late-appearing CAFC representing LTR cells is not indicative for persistence of the transgene in vivo. An alternative explanation might be that the site of integration does not favor expression, but our negative Southern blots of various hematopoietic tissues indicates lack of integration rather than failure of expression. A similar loss of expression of the transgene over time might be expected when cells are grown in stroma-supported long-term bone marrow cultures (such as in the CAFC assay) beyond five weeks.
AAV vectors have been proposed as an alternative to retroviral vectors for achieving efficient transduction of non-dividing cells [12]. However, the efficiency of integration of such vectors in hematopoietic cells is low despite the prolonged episomal persistence of rAAV [13]. Our data also suggest a transient integration or prolonged episomal expression of the transgene. Presence of the transgene in stem cells immediately after transduction reflects efficient entry of rAAV particles, although expression declines with cell passage when the vector is not successfully integrated into the genome. Since rAAV integration was found to be dependent on the MOI, it may also be possible that, in our experiments, the MOI was too low to ensure genomic integration [14]. Indeed, using a higher MOI, one other study has shown successful in vivo long-term expression of the transgene when rAAV was used to transduce hematopoietic stem cells [15].
In conclusion, a better understanding of vector biology is needed to design and package vectors best suited for transduction of hematopoietic stem cells. This knowledge may then lead to higher levels of integration into quiescent hematopoietic stem cells, an outcome that is necessary for effective gene therapy applications. In addition, improvements in gene transfer should be carefully evaluated to ensure proper genomic integration and long-term in vivo persistence of the transgene.
|
Acknowledgments
|
|---|
The authors wish to thank Dr. Jerome E. Groopman, Division of Experimental Medicine, Beth Israel Deaconess Medical Center, Boston, for helpful discussions and Janet Delahanty for editing the manuscript. We also appreciated the technical assistance of John M.K. Mislow.
This research was supported by NIH Grants RO1-CA10941-26, R01-HL51456, R01-HL55445, and NIH/NIAID 5PO1-HL4351007
|
References
|
|---|
-
Anderson WF. Gene therapy. Sci Am 1995;273:124-128.[Medline]
-
Walsh CE, Liu JM, Miller JL et al. Gene therapy for human hemoglobinopathies. Proc Soc Exp Biol Med 1993;204:289-300.[Abstract]
-
Karlsson S. Treatment of genetic defects in hematopoietic cell function by gene transfer. Blood 1991;78:2481-2492.[Free Full Text]
-
Du B, Wu P, Boldt-Houle D et al. Efficient transduction of human neurons with an adenovirus-associated virus vector. Gene Ther 1996;3:254-261.[Medline]
-
Ploemacher R. Cobblestone Area Forming Cell (CAFC) Assay. Culture of Hemopoietic Cells. New York, NY: Wiley-Liss, Inc., 1994:1-21.
-
van Os R, Dawes D, Mislow J et al. Host conditioning with 5-fluorouracil and kit-ligand to provide for long-term bone marrow engraftment. Blood 1997;89:2376-2383.[Abstract/Free Full Text]
-
Ploemacher R, van Os R, van Beurden C et al. Murine hematopoietic stem cells with long-term engraftment and marrow repopulating ability are less radiosensitive to gamma radiation than are spleen colony forming cells. Int J Radiat Biol 1992;61:489-499.[Medline]
-
Baer RP, Whitehill TE, Sarkar R et al. Retroviral-mediated transduction of endothelial cells with the lac Z gene impairs cellular proliferation in vitro and graft endothelialization in vivo. J Vasc Surg 1996;24:892-899.[Medline]
-
Clapp DW, Freie B, Srour E et al. Myeloproliferative sarcoma virus directed expression of beta-galactosidase following retroviral transduction of murine hematopoietic cells. Exp Hematol 1995;23:630-638.[Medline]
-
Zijlmans JMJM, Bierhuizen MFA, Jumelet E et al. Gene transfer in hematopoietic stem cells with in vivo long-term repopulating ability is accurately assessed in vitro with green fluorescent protein and the cobblestone area forming cell assay. Exp Hematol 1998;26:763a.
-
Van Hennik PB, Verstegen MMA, Bierhuizen MFA et al. Highly efficient transduction of the green fluorescent protein gene in human umbilical cord blood stem cells capable of cobblestone formation in long-term cultures and multilineage engraftment of immunodeficient mice. Blood 1998;92:4013-4022.[Abstract/Free Full Text]
-
Muzyczka N. Use of adeno-associated virus as a general transduction vector for mammalian cells. Curr Top Microbiol Immunol 1992;158:97-129.[Medline]
-
Goodman S, Xiao X, Donahue RE et al. Recombinant adeno-associated virus-mediated gene transfer into hematopoietic progenitor cells. Blood 1994;84:1492-1500.[Abstract/Free Full Text]
-
Nienhuis AW, Bertran J, Hargrove P et al. Gene transfer into hematopoietic cells. STEM CELLS 1997;15(suppl 1):123-134.
-
Ponnazhagan S, Yoder MC, Srivastava A. Adeno-associated virus type 2-mediated transduction of murine hematopoietic cells with long-term repopulating ability and sustained expression of a human globin gene in vivo. J Virol 1997;71:3098-3104.[Abstract]
accepted for publication January 18, 1999.
Top
Abstract
Introduction
