Development of safe and efficient retroviral vectors for ... - Nature

2 downloads 0 Views 275KB Size Report
Keywords: retrovirus; gene therapy; Gaucher disease. Gaucher disease is caused by a deficiency in the enzyme version. Both constructs are either packaged by ...
Gene Therapy (1997) 4, 1393–1400  1997 Stockton Press All rights reserved 0969-7128/97 $12.00

BRIEF COMMUNICATION

Development of safe and efficient retroviral vectors for Gaucher disease M Havenga1, R Fisher2, P Hoogerbrugge 2,3, B Roberts4, D Valerio1,2 and HHG van Es1,2 1

Gene Therapy Section, Department of Molecular Cell Biology, Medical Faculty, Leiden University; 2IntroGene BV, Leiden; Department of Pediatrics, Sophia Children’s Hospital, Rotterdam, The Netherlands; and 4Genzyme Corporation, Framingham, MA, USA 3

We have generated amphotropic and Gibbon ape leukemia (GaLV) viruses carrying either a full-length (IG-GC2) or a shortened glucocerebrosidase cDNA (IG-GC4). For all recombinant retroviruses, a single infection was sufficient to augment glucocerebrosidase activity in unselected Gaucher type I and type II fibroblasts to levels which can be considered therapeutic. Transfer efficiency of the glucocerebrosidase cDNA into normal human and Gaucher type I CD34+ cells, using supernatant transduction, ranged from 4 to 50% as established on vector-positive CFU-GM. In

these experiments, GaLV and amphotropic virus were equally efficient in transducing early human progenitors. Importantly, mixing amphotropic and GaLV pseudotyped retroviruses resulted in significantly higher transduction efficiencies as compared with single infections, up to 70% vector-positive CFU-GM. Glucocerebrosidase activity, measured in the progeny of human CD34+ cells, increased up to 460% compared with mock-infected CD34+ cells. Upon transduction of Gaucher CD34+ bone marrow cells the glucocerebrosidase deficiency was reversed.

Keywords: retrovirus; gene therapy; Gaucher disease

Gaucher disease is caused by a deficiency in the enzyme b-glucocerebrosidase (hGC, EC 3.2.1.45). Gene therapy offers an attractive treatment for Gaucher disease since the clinical phenotype manifests itself predominantly in cells derived from the pluripotent hemopoietic stem cell (PHSC). Transfer of a correct copy of the glucerebrosidase cDNA into PHSCs should result in hGC expressing hemopoietic lineages, including macrophages. With recombinant retroviruses, up to 50–60% of human primitive hemopoietic progenitor cells (CFUs) can be transduced ex vivo using clinically accepted supernatant transduction protocols.1 However, with a few exceptions,2,3 the presence of transduced genes in the hemopoietic system of large animals has been disappointing, with only 0.01–5% transduced cells detected in the long term.4,5 Factors limiting ex vivo retroviral transduction of PHSCs are now known to include virus binding,6 ex vivo cycling and maintenance of PHSCs,7 and amphotropic receptor expression on PHSCs.8 To increase retroviral infection into PHSCs, a prerequisite for treatment of Gaucher disease by stem cell gene therapy, alternative transduction methods, culture conditions and retroviral vectors are being explored (see Ref. 9 for a detailed coverage of alternative retroviral transduction methods). To study alternative PHSC transduction methods we generated four different recombinant retroviral producer cell lines. The recombinant viruses carry either a full-length version of the human glucocerebrosidase cDNA or a 39-truncated

Correspondence: HHG van Es, IntroGene BV, PO Box 2048, 2301CA Leiden, The Netherlands Received 8 May 1997; accepted 17 July 1997

version. Both constructs are either packaged by the amphotropic virus producer cell line, PA317,10 or the Gibbon ape leukemia virus producer cell line, PG13,11 to determine which retroviral receptor, GLVR-2 12 or GLVR1,13 is more abundantly expressed on human PHSCs. This article describes the generation and characterization of the four retrovirus producer cell lines and the employment of these recombinant viruses for transduction of primary human cells, including fibroblasts and CD34+ cells isolated from bone marrow and mobilized peripheral blood. In the 39 LTR of the retroviral constructs the enhancer in U3 of the wild-type Moloney murine leukemia virus (MoMLV) is replaced by a mutant (PyF101) polyoma enhancer (DMo + PyF101). This backbone was chosen since it is known that the presence of DMo + PyF101 provides sustained expression in primate hemopoietic bone marrow cells4 and since this hybrid LTR is known to abolish leukemogenicity of MoMLV.14,15 The 59 LTR is derived from Moloney murine sarcoma virus (MoMSV) including part of the RNA packaging signal (for a detailed description of the IG-GC vectors described here see Ref. 16). The remaining 39 part of the packaging signal and other 39 sequences upstream of the ATG of the inserted transgene are derived from MoMLV. In the latter part of the vector the ATGgag has been mutated to a TAG and a cryptic splice acceptor site is present. After one round of replication the 59 LTR driving expression of the transgene will contain the U3 and R region derived from the 39 LTR carrying the mutant polyoma enhancer PyF101. Therefore, in both IG-GC2 and IGGC4, transcription of the hGC cDNA is under the control of DMo + PyF101-LTR. IG-GC2 contains a full-length human placental glucocerebrosidase cDNA, whereas IG-

Gene therapy for Gaucher disease M Havenga et al

1394

GC4 contains a 160 bp deletion in the 39 untranslated region of the hGC cDNA. As expected, the difference in length of the hGC cDNAs is sufficient to discriminate between the constructs (Figure 1). Virus producing cell clones were generated by cotransfecting ecotropic packaging cell lines with IG-GC2 or IGGC4 and a neomycin resistance plasmid. Ecotropic virus was used to infect the amphotropic retroviral packaging cell line, PA31710 and the Gibbon ape leukemia virus (GaLV) packaging cell line, PG13.11 Both an increase in hGC activity and Western analysis demonstrated the presence of the hGC protein (data not shown). From each producer pool, more than a hundred cell clones were screened for production of hGC transferring virus.17 Clones were isolated and coded PA2 (PA317/IG-GC2), PA4 (PA317/IG-GC4), PG2 (PG13/IG-GC2) and PG4 (PG13/IG-GC4). Master cell banks of the retrovirus producers were tested for sterility, absence of mycoplasm, absence of replication competent retrovirus (RCR) using a S+/L− foci test (,1 RCR/ml) after amplification on Mus dunni cells and absence of adventitious virus. Moreover, the producer cells were subjected to DNA fingerprinting and provirus-specific identity testing using PCR and Southern analysis. In addition, clinical lots of virus supernatant were tested for endotoxin levels and general safety. Both the master cell banks and clinical lots passed all safety tests. Because it is known that the PyF101 enhancer is less active in fibroblasts compared with the wild-type enhancer, 18 we first verified whether the recombinant

producer cell clones contain the intact mutant PyF101 enhancer element. Genomic DNA was extracted from the producer clones using standard techniques.18 The isolated DNA was digested with either NheI or with SstI. With these two enzymes the presence of the mutant PyF101 enhancer can be investigated in both the 59 and 39 LTR (Figure 2). When blotted and probed with the PyF101 fragment, hybridization signals corresponding to the expected proviral DNA fragments of 3.4 (IG-GC2) or 3.2 kb (IG-GC4) were detected, confirming that the PyF101 enhancer is present in all four producer cell lines in both the 59 and 39 LTR (Figure 2). Additional hybridizing fragments represent flanking genomic DNA sequences. To determine a functional virus titer from the PA2, PA4, PG2 and PG4 virus producer cell lines, genomic DNA of infected Gaucher type II fibroblasts was extracted, digested with EcoRI, blotted, and probed with an hGC fragment. Quantification of signal intensities between the endogenous DNA fragment (20 kb) and the proviral DNA fragment (3.4 kb for IG-GC2 or 3.2 kb for IG-GC4) resulted in a virus titer of 3 × 105, 1 × 105, 2 × 105, and 8 × 104 for PA2, PA4, PG2 and PG4, respectively (Figure 3). In addition, the virus titer derived from the PA2 producer cell clone was also determined on NIH/3T3 cells using a different approach19 and found to be 6 × 105 functional virus particles per milliliter (data not shown). Gaucher type I and II primary fibroblasts were infected with PA2, PA4, PG2 or PG4 derived virus supernatant.

Figure 1 (a) Schematic presentation of the proviral structure of IG-GC2 and IG-GC4. Shown are the splice donor (SD), splice acceptor (SA), primer binding sites (P− and P+) and the packaging signal (C). Details on vector construction are published elsewhere.16 The arrows denote the approximate location of oligonucleotides MH3 and GCo4 which are used to discriminate between IG-GC2 and IG-GC4 in a PCR reaction. (b) Ethidium bromide stained gel of PCR results with oligonucleotides MH3 and GCo4 obtained by mixing plasmid IG-GC2 and IG-GC4 in various concentrations (0–100%). The DNA isolation procedure for PCR analysis has been described.4 PCR analysis on transduced myeloid cells derived from CD34+ bone marrow cells or individual CFU-GM-derived colonies was performed directly using this suspension. Oligonucleotides GCo4 (59-CAGCCCATGTTCTACCAC-39) and MHo3 (59-GGATCCCTAGGCTTTTGC-39) detect the presence of both IG-GC2 and IG-GC4 provirus.

Gene therapy for Gaucher disease M Havenga et al

1395

Figure 2 Southern analysis to assay for the presence of the polyoma enhancer in both the 59 and 39 LTR. Genomic DNA of the retroviral producer cell lines PA2, PA4, PG2 and PG4 as well as from the parental cell lines PA317 (PA) and PG13 (PG) was isolated. After digestion with either NheI (N) or SstI (S) the genomic DNA was blotted and probed with the complete PyF101 enhancer fragment. For the IG-GC2 construct, present in PA2 and PG2, the 32P-labeled PyF101 enhancer fragment specifically recognizes a 3.4 kb proviral DNA fragment. For the IG-GC4 construct, present in PA4 and PG4, a 3.2 kb proviral DNA fragment is recognized.

Figure 3 Southern analysis on genomic DNA isolated from infected Gaucher type II fibroblasts. Genomic DNA (10 mg) of noninfected (−) and PA2, PA4, PG2, or PG4 infected Gaucher type II fibroblasts was digested with NheI which cuts in both LTRs thereby liberating the entire proviral sequence. Quantification of signal intensity ratios between the endogenous hGC fragment (± 20 kb) and the proviral fragment (3.4 kb and 3.2 kb) using a phosphor imager, revealed a ratio of 0.0 (−), 0.3 (PA4), 0.8 (PA2), 0.4 (PG2) and 0.2 (PG4).

After 48 h their hGC activity was determined using standard techniques (Figure 4).17,20 Based on these results we concluded that a single infection with either PA2 or PG2 virus supernatant (n = 12) augmented hGC activity levels in both the Gaucher type I and type II fibroblasts to levels comparable with normal human fibroblasts. After transduction with either PA4 or PG4 virus supernatant

(n = 12), correction levels of approximately 50–70% of normal were achieved. Human GC expression was further confirmed using Western analysis (Figure 5). Therefore, all four producer cell lines generate retrovirus particles that transfer an active hGC gene to primary human cells and are capable of stably correcting the Gaucher defect. After informed consent, bone marrow cells were harvested from healthy donors (n = 7), from a patient undergoing allogeneic bone marrow transplantation for NHL (n = 1), and from a patient with Gaucher disease, who underwent surgery for implantation of a porth-acath catheter (n = 1). Peripheral blood stem cells were harvested after mobilization with chemotherapy followed by G-CSF treatment in patients with NHL (n = 3). Low density mononuclear cells (,1.064 g/cm3 ) were obtained by Ficoll gradient separation.21 Human CD34+ cells were isolated using either immunoabsorption columns (Cellpro, Bothell, WA, USA) or magnetic antibodies (Miltenyi, Bergisch Gladbach, Germany). Mononuclear cells or purified CD34 + cells were infected daily for 4 days in the presence of 4 mg/ml protamine sulphate (Pharmacia BV, Woerden, The Netherlands). To stimulate hemopoietic proliferation we only added 50 ng/ml IL-3 (Sandoz, Basel, Switzerland).7 In all experiments, a mock infection was included using a retrovirus carrying the multidrug resistance gene (MDR). The virus-backbone of the MDR construct is identical to the IG-GC series and the virus titer was approximately 105 particles per milliliter as established by vincristine resistance on A2780 cells (unpublished results). Oligonucleotides GCo4 (59-CAGCCCATGTTCTACCAC-39) and MHo3 (59GGATCCCTAGGCTTTTGC-39) were used for PCR analysis on a suspension of mature myeloid cells derived from transduced CD34+ cells and CFU-GM and prepared as described previously.4 In Table 1 the results from 11 independent transduc-

Gene therapy for Gaucher disease M Havenga et al

1396

Figure 4 Specific hGC activity measured as nanomoles/h/mg20 protein in bulk-infected, unselected Gaucher type I (a) or type II (b) primary fibroblasts (n = 12). (a) Black bar (N): hGC activity measured in normal human fibroblasts (2000 ± 390). White bar: residual hGC activity in Gaucher type I fibroblasts (475 ± 15). Grey bars: hGC activity in Gaucher type I fibroblasts infected with virus derived from PA2 (2050 ± 150), PA4 (1350 ± 250), PG2 (2350 ± 400) and PG4 (970 ± 75). (b) Black bar (N): hGC activity measured in normal human fibroblasts (2000 ± 390). White bar: residual hGC activity in Gaucher type II fibroblasts (300 ± 15). Grey bars: hGC activity in Gaucher type II fibroblasts infected with virus derived from PA2 (2950 ± 200), PA4 (1500 ± 120), PG2 (3800 ± 300) and PG4 (1400 ± 200).

Figure 5 Western analysis31 on 20 mg of total protein from (a) retroviral producer cell lines and (b) infected and unselected Gaucher type II fibroblasts. (a) Protein lysates prepared from retroviral producer cell lines PA2, PA4, PG2 and PG4. Parental cell lines PA317, and PG13 are denoted as PA and PG, respectively. One unit of Cerezyme was used as a positive control (+). (b) Protein lysates prepared from Gaucher type II fibroblasts infected once with PA2, PA4, PG2 and PG4-derived virus. Protein lysate of non-infected Gaucher type II fibroblasts is denoted as GtII.

tion experiments on human CD34+ cells are summarized. From these results several conclusions can be drawn. First, all four recombinant viruses are able to transduce human CD34+ cells efficiently, since the presence of the provirus could be confirmed by PCR in both the mature myeloid progeny derived from CD34+ cells and after flow cytometric sorting of cells positive for the CD34+ antigen directly after transduction (data not shown). Second, there is a large difference in retroviral infectability of CD34+ cells derived from different donors, irrespective of the virus tropism used, ie GaLV versus amphotropic. Third, a high increase in hGC activity correlates with high percentages of vector-positive CFUs. Fourth, in con-

trast to other groups who reported on increased transduction efficiency using a GaLV pseudotyped virus,22,23 we were unable to detect a significant difference between GaLV and amphotropic pseudotyped viruses in transducing human CD34+ cells. Moreover, the results obtained show that increased overall transduction can be achieved when GaLV and amphotropic virus are added simultaneously to CD34 + cells (n = 3), which may result in an improved therapeutic window for Gaucher patients. Since each virus uses a different receptor, competition between the two pseudotyped viruses was not expected and did not occur. 12,13 Whether GaLV or amphotropic virus infects PHSCs more efficiently is

Gene therapy for Gaucher disease M Havenga et al

1397 Table 1 Summary of results obtained from transduction experiments on human hemopoietic cells using a standard, 4 day transduction protocol Medical status

Source

Sample

Virus

MOIa

Relative GC activityb

PCRc

PCR+ CFU-GMd

Total %

IG-GC2 IG-GC4 IG-GC2 + 4 Normal Non-Hodgkin Non-Hodgkin Normal Normal Normal Non-Hodgkin Normal

BM PB PB BM BM BM PB BM

CD34 + CD34 + Post-Ficoll Post-Ficoll Post-Ficoll CD34+ CD34+ CD34+

PA2 PA2 PA2 PA2 PA2 PA2 PA2 PA2

0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3

12 × 1.0 × 1.3 × 1.0 × 2.1 × 1.0 × 1.0 × 1.0 ×

+ + + + + + + +

ND ND ND ND ND ND ND ND

— — — — — — — —

— — — — — — — —

— — — — — — — —

Normal

BM

CD34+

PA2 PA4 PG2 PG2 + PA4

1.2 0.4 0.8 0.4 + 0.2

4.6 × 2.4 × 2.6 × 4.0 ×

+ + + +

11/20 — 8/20 6/20

— 6/20 — 4/20

— — — 4/20

55 30 40 70

Normal

BM

CD34+

PA2 PA4 PG2 PG4 PA2 + PG4 PG2 + PA4

1.2 0.4 0.8 0.3 0.6 + 0.2 0.4 + 0.2

1.2 × 1.0 × 1.2 × 1.2 × 1.4 × 1.4 ×

+ + + + + +

2/24 — 3/24 — 5/24 2/24

— 1/24 — 2/24 4/24 3/24

— — — — 2/24 1/24

8 4 12 8 44 24

Non-Hodgkin

BM

CD34+

PA2 PA4 PG2 PG4

1.2 0.4 0.8 0.3

1.0 × 1.4 × 1.3 × 1.3 ×

+ + + +

1/24 — 1/24 —

— 6/24 — 3/24

— — — —

4 24 4 12

a

MOI, multiplicity of infection as determined on Gaucher type II fibroblasts. hGC activity measured in the PA2, PA4, PG2, or PG4 transduced cells divided by the hGC activity measured in mock-infected cells. c Provirus-specific PCR on mature myeloid progeny to determine the presence of the proviral structure (+, present). d The number of vector-positive CFU-GM divided by the total number of CFU-GM analyzed by PCR is used to determine the percentage of transduction efficiency. CFU-GM and myeloid progenitors derived from human CD34+ cells were obtained as has been described elsewhere.30 ND, not determined. b

unknown at present but long-term in vivo transduction studies in baboons and rhesus monkeys indicate that GaLV is not superior to amphotropic virus in transducing PHSCs.24 The overall transduction results shown here are similar to those reported in other studies in which normal bone marrow or mobilized peripheral blood was transduced with a hGC retroviral vector.25 Three of our four recombinant retroviruses were also tested for their ability to correct b-glucocerebrosidase deficiency in cells derived from CD34+ cells of a Gaucher type I patient. Correction of the Gaucher phenotype was clearly achieved in the mature myeloid progeny when compared with the averaged hGC activity measured in the progeny of eight different CD34 + samples derived from normal donors (Figure 6). PCR and subsequent Southern analysis on the transduced myeloid progeny of the Gaucher CD34+ cells confirmed the presence of the IG-GC2 and IG-GC4 provirus. The CFU-GM-based retroviral transduction efficiency in the Gaucher-derived CD34+ cells, was up to 52% (Figure 6). Although the number of vector-positive CFU-GM between amphotropic and GaLV was similar, a comparison between Southern hybridization signals suggested that more copies of the IG-GC2 construct are present per c.f.u. when PG2 virus was used. This difference in copy number could explain the difference in hGC activity measured in the mature myeloid progeny. Studies by others using a hGC retrovirus and supernatant transduction of Gaucher CD34+

cells have shown increases in hGC enzyme levels to a level approximately three-fold higher than normal hGC levels.1,19 Based on transduction efficiency and levels of expression, we conclude that the recombinant hGC viruses we have generated are as efficient in infecting human CD34 + hemopoietic cells as other retroviral vectors for Gaucher disease including MFG-GC and LGC. A comparative study between MFG-GC and LGC-derived recombinant viruses indicated that MFG-transduced cells contained approximately five-fold higher levels of spliced RNA per vector copy compared with the LN vectors, suggesting that MFG retroviral vectors provide higher levels of expression in target cells.26 In the IG-GC2 and IG-GC4 vectors, the splice donor and splice acceptor are both derived from the LN vector. One major difference between the LN vector and IG-GC is the presence of the mutant PyF101 polyoma enhancer in the virus LTRs which might explain the levels of expression measured in differentiated CD34+ cells. To determine whether we could increase retroviral transduction using centrifugation, NIH/3T3 cells were infected for 2.5 h at 1100 g (n = 5). A typical example of results obtained with diluted PA2 virus is given in Figure 7. From these results it can be concluded that without centrifugation a 50-fold dilution in virus titer results in a complete reduction in hGC activity from 375 ± 22 arbitrary hGC units to 0 units. In contrast, using centrifugation, a 50-fold dilution in virus titer results in

Gene therapy for Gaucher disease M Havenga et al

1398

Figure 6 (a) Specific hGC activity measured as nanomoles/h/mg protein in the mature myeloid progeny derived from Gaucher type I CD34+ cells (values are the mean of two independent infections). Black bar (N): averaged hGC activity measured in the mature myeloid progeny from normal human individuals (10 500 ± 3500, n = 8). White bar (Mock): hGC activity measured in the mature myeloid progeny after transduction with a retroviral vector carrying a MDR cDNA. Grey bars represent hGC activity in PA2, PA4 and PG2 transduced cells. (b) PCR and Southern analysis on mock-transduced and PA2, PG2 or PA4 transduced myeloid progeny derived from Gaucher type I CD34+ cells (each in duplicate) to confirm the presence of retroviral sequence. The expected length using oligonucleotides MH3 and GCo4 is a 590 bp (IG-GC2) or 430 bp (IGGC4) fragment. (c) The percentage of CFU-GM found positive for the IGGC provirus.

a less than two-fold reduction in hGC activity (from 355 ± 64 arbitrary hGC units to 205 ± 14). Therefore, centrifugation for 2.5 h at 1100 g increases the retroviral transduction efficiency of diluted virus preparations on NIH/3T3 cells. The next step was to demonstrate increased transduction of human CD34+ cells using the centrifugation protocol. For this purpose human CD34 + cells were seeded in virus supernatant supplemented with protamine-HCl and IL-3 and subsequently centrifuged for 2.5 h at 1100 g either once or four times for 4 days. Analysis of transduction efficiency into human CD34+ cells is as described previously. A comparison of the total number of c.f.u.s obtained by seeding equal numbers of CD34+ cells after transduction, with or without centrifugation, indicated no increased cell death due to the 1100 g centrifugation force (data not shown). The results obtained on the number of vector-positive CFU-GM and increased hGC

activity are summarized in Table 2. These results indicate that four times centrifugation of our undiluted virus preparations on to CD34+ cells did not result in increased numbers of vector-positive CFU-GM. Evidently the virus concentration was already sufficient for efficient CD34+ cell transduction. This is further witnessed by the finding that a single exposure to undiluted retrovirus preparations, with or without centrifugation, is almost as efficient as four exposures during 4 days. It is generally accepted that centrifugation does not act on the virus to target cell ratio but that it promotes virus–cell contact which increases the chance of successful retroviral uptake in target cells.27 Since we did not observe increases in retroviral infection of NIH/3T3 cells (data not shown) and human CD34+ cells using undiluted virus preparations, we conclude that virus binding is not a limiting factor in our recombinant virus preparations. In summary, we have developed a panel of retroviral vectors carrying the hGC cDNA driven by an LTR containing the mutant polyoma enhancer PyF101 which is known to give sustained expression in human hemopoietic cells. No replication competent retrovirus could be detected in any of the four master cell banks nor was there any RCR detectable in 5% of 10 liter lots of retrovirus produced by two of the cell lines, namely PA2 (PA317 derived) and PG4 (PG13 derived). This is in line with previous observations using PA317 and a construct similar to IGGC2 or IGGC4 but containing the human MDR1 cDNA instead of the hGC cDNA (JJB Boesen, unpublished observations). The recombinant viruses are able to infect human hemopoietic CD34+ cells efficiently and are capable of correcting the Gaucher phenotype in both primary fibroblasts and Gaucher-derived CD34+ cells. Importantly, we have shown that mixing GaLV and amphotropic virus results in an overall increased transduction as measured by the number of vector-positive CFU-GM, which predicts that multitropism infection might improve the therapeutic window for treatment of Gaucher disease by gene therapy. Furthermore, there is no competition for infection, which is in line with the observations that GaLV and amphotropic virus use two related but different retroviral receptors. We believe, therefore, that our recombinant PA2, PA4 and PG2, and PG4 viruses are ideally suited for gene marking studies to compare the effect of virus tropism within one stem cell graft. Transduction of one stem cell graft with two viruses is preferred to exclude differences in transduction efficiency due to differences in the composition of a retroviral supernatant lot. Components which might negatively influence transduction of human CD34+ cells are physical parameters such as pH, nutrient concentrations and concentrations of certain (murine) growth factors known to act on human cells, such as TGF-b.28,29 Upon reinfusion of the transduced stem cell graft, PCR analysis on blood samples will determine which of the two viruses is present more abundantly and hence, which retroviral receptor, GLVR-1 or GLVR-2, is expressed more abundantly on PHSCs. These recombinant viruses can also be used to determine which ex vivo PHSC culture conditions or transduction protocols are superior. In such studies a stem cell graft will have to be split after which each graft will be transduced with one virus only. Finally these retroviruses provide essential ingredients for pilot trials in Gaucher patients.

Gene therapy for Gaucher disease M Havenga et al

1399

Figure 7 Centrifugation-facilitated retroviral transduction of NIH/3T3 cells. (a) 200 ml of diluted PA2 virus supernatant was added to 2000 NIH/3T3 seeded 24 h before infection in 96-well plates. (b) Identical as described for (a), except that the 96-well plate was subsequently centrifuged for 2.5 h at 1100 g and then returned to a 10% CO2 incubator at 37°C. Forty-eight hours after infection, hGC activity of both (a) and (b) was measured as described.17,20

Table 2 Summary of results obtained from transduction experiments on human hemopoietic cells using various transduction protocols Medical status

Non-Hodgkin

Normal

Source

BM

BM

Sample

CD34 +

CD34+

Virus

MOIa

Transduction protocol

Relative GC activityb

PCR+ CFU-GMc IG-GC2 (%)

IG-GC4 (%)

PA2 PG4

1.2 0.3

4 × centrifugation 2.5 h/1100 g

1.0 × 4.1 ×

5/24 (20) —

— 3/24 (12)

PA2 PG4

1.2 0.3

Standard, 4 day transduction

1.0 × 1.0 ×

1/24 (4) —

— 3/24 (12)

PA2 PA4 PG2

1.2 0.3 0.8

1 × centrifugation 2.5 h/1100 g

2.0 × 1.9 × 2.1 ×

6/20 (30) — 9/20 (45)

— 5/20 (25) —

PA2

1.2

Standard, 1 day transduction

ND

5/20 (25)



PA2 PA4 PG2

1.2 0.3 0.8

4 × centrifugation 2.5 h/1100 g

1.4 × 1.0 × 2.2 ×

5/20 (25) — 10/20 (50)

— 6/20 (30) —

PA2 PA4 PG2

1.2 0.3 0.8

Standard, 4 day transduction

2.5 × 2.0 × 2.3 ×

11/20 (55) — 8/20 (40)

— 6/20 (30) —

a

MOI, multiplicity of infection as determined on Gaucher type II fibroblasts. hGC activity measured in the PA2, PA4, PG2, or PG4 transduced cells divided by the hGC activity measured in mock-infected cells. c The number of vector-positive CFU-GM divided by the total number of CFU-GM analyzed by PCR is used to determine the percentage of transduction efficiency. ND, not determined. b

Acknowledgements We acknowledge the Department of Pediatrics of the University Hospital in Leiden and Department of Haematology of the Sophia Childrens Hospital in Rotterdam for bone marrow from normal donors and a Gaucher type I patient, respectively. We also wish to thank the Department of Pediatrics of the University Hospital in Rotterdam for bone marrow of NHL patients and Dr V van Beusechem for critically reading the manuscript. This

research was supported by ‘Praeventiefonds’ grant 0028359 and Genzyme Corporation.

References 1 Xu L et al. Correction of the enzyme deficiency in hematopoietic cells of Gaucher patients using a clinically acceptable retroviral supernatant transduction protocol. Exp Hematol 1994; 22: 223– 230.

Gene therapy for Gaucher disease M Havenga et al

1400

2 Brenner MK et al. Gene marking to determine whether autologous marrow infusion restores long-term haemopoiesis in cancer patients. Lancet 1993; 342: 1134–1137. 3 Bienzle D et al. Gene transfer into hematopoietic stem cells: longterm maintenance of in vitro activated progenitors without marrow ablation. Proc Natl Acad Sci USA 1994; 91: 350–354. 4 van Beusechem VW et al. Long-term expression of human adenosine deaminase in rhesus monkeys transplanted with retrovirus-infected bone marrow cells. Proc Natl Acad Sci USA 1992; 89: 7640–7644. 5 Dunbar CE et al. Improved retroviral gene transfer into murine and rhesus peripheral blood or bone marrow repopulating cells primed in vivo with stem cell factor and granulocyte colonystimulating factor. Proc Natl Acad Sci USA 1996; 93: 11871–11876. 6 Chuck AS, Palsson BO. Membrane adsorption characteristics determine the kinetics of flow-through transductions. Biotechnol Bioengineer 1996; 51: 260–270. 7 Knaan Shanzer S, Valerio D, van Beusechem VW. Cell cycle state, response to hemopoietic growth factors and retroviral vector-mediated transduction of human hemopoietic stem cells. Gene Therapy 1996; 3: 323–333. 8 van Es HHG et al. Expression studies of the amphotropic receptor GLVR2 in mammalian cells and tissues including human CD34+ cells. Cold Spring Harbor Gene Therapy Meeting, 1996, Abstr. 309. 9 Havenga MJE et al. Retroviral stem cell gene therapy. Stem Cells 1997; 15: 162–179. 10 Miller AD, Buttimore C. Redesign of retrovirus packaging cell lines to avoid recombination leading to helper virus production. Mol Cell Biol 1986; 6: 2895–2902. 11 Miller AD et al. Construction and properties of retrovirus packaging cells based on Gibbon ape leukemia virus. J Virol 1991; 65: 2220–2224. 12 van Zeijl M et al. A human amphotropic retrovirus receptor is a second member of the Gibbon ape leukemia virus receptor family. Proc Natl Acad Sci USA 1994; 91: 1168–1172. 13 O’Hara B et al. Characterization of a human gene conferring sensitivity to infection by Gibbon ape leukemia virus. Cell Growth Differ 1990; 1: 119–127. 14 Danos O, Mulligan RC. Safe and efficient generation of recombinant retroviruses with amphotropic and ecotropic host ranges. Proc Natl Acad Sci USA 1988; 85: 6460–6464. 15 Brightman BK, Farmer C, Fan H. Escape from in vivo restriction of Moloney mink cell focus inducing viruses driven by the Mo + PyF101 long terminal repeat (LTR) by LTR alterations. J Virol 1993; 67: 7140–7148. 16 Vogels R et al. Improved recombinant retroviral vectors for gene therapy. Patent publication, 1996; Wo96/35798. 17 Havenga MJE et al. Improved screening method for recombinant retrovirus producing clones. Biotechniques 1996; 21: 1004–1007.

18 Stewart TA, Pattengale PK, Leder P. Spontaneous mammary adenocarcinomas in transgenic mice that carry and express MTV/myc fusion genes. Cell 1984; 38: 627–637. 19 Bahnson AB et al. Transduction of CD34 + enriched cord blood and Gaucher bone marrow cells by a retroviral vector carrying the glucocerebrosidase gene. Gene Therapy 1994; 1: 176–184. 20 Aerts JMFG et al. The occurrence of two immunologically distinguishable b-glucocerebrosidases in human spleen. Eur J Biochem 1985; 150: 565–574. 21 Dicke KA, Tridente G, van Bekkum DW. The selective elimination of immunologically competent cells from bone marrow and lymphocyte cell mixtures: in vitro test for detection of immunocompetent cells in fractionated mouse spleen cell suspensions and primate bone marrow suspensions. Transplantation 1969; 8: 422–434. 22 von Kalle C et al. Increased gene transfer into human hematopoietic progenitor cells by extended in vitro exposure to a pseudotyped retroviral vector. Blood 1994; 84: 2890–2897. 23 Bunnell BA et al. High-efficiency retroviral-mediated gene transfer into human and nonhuman primate peripheral blood lymphocytes. Proc Natl Acad Sci USA 1995; 92: 7739–7743. 24 Kiem HP et al. Gene transfer into baboon hematopoietic repopulating cells: comparison between amphotropic and a gibbon ape leukemia virus (GALV) pseudotyped vectors. Blood 1996; 88: 1708. 25 Nimgaonkar MT et al. Transduction of mobilized peripheral blood CD34+ cells with the glucocerebrosidase cDNA. Gene Therapy 1994; 1: 201–207. 26 Krall WJ et al. Increased levels of spliced RNA account for augmented expression from the MFG retroviral vector in hematopoietic cells. Gene Therapy 1996; 3: 37–48. 27 Hodgkin PD et al. Murine cytomegalovirus binds reversibly to mouse embryo fibroblasts: implications for quantitation and explanation of centrifugal enhancement. J Virol Meth 1988; 22: 215–230. 28 Einerhand MP et al. Factors affecting the transduction of pluripotent hematopoietic stem cells: long-term expression of a human adenosine deaminase gene in mice. Blood 1993; 81: 254–263. 29 Hatzfeld J et al. Purification and release from quiescence of umbilical cord blood early progenitors reveal their potential to engraft adults. Blood Cells 1994; 20: 430–434. 30 Fruehauf S et al. Frequency analysis of multidrug resistance-1 gene transfer into human primitive hematopoietic progenitor cells using the cobblestone area-forming cell assay and detection of vector-mediated P-glycoprotein expression by Rhodamine123. Hum Gene Ther 1996; 7: 1219–1231. 31 van Es HHG et al. Expression of the plasmodial pfmdr1 gene in mammalian cells is associated with increased susceptability to chloroquine. Mol Cell Biol 1994; 14: 2419–2428.