Tumor Necrosis Factor Alpha Enhances the Adenoviral Transduction of CD34+ Hematopoietic Progenitor Cells
http://www.100md.com
《干细胞学杂志》
a Institute for Transfusion Medicine, Charité—Universit?tsmedizin Berlin, Germany;
b Memorial Sloan-Kettering Cancer Center, New York, New York, USA
Key Words. TNF- ? CD34+ cells ? Adenovirus ? Transduction ? Capping
Anja Moldenhauer, M.D., Institute for Transfusion Medicine, Charité—Universit?tsmedizin Berlin, Augustenburger Platz 1, 13351 Berlin, Germany. Telephone: 49-160-1090837; Fax: 49-450 553988; e-mail: anja.moldenhauer@charite.de
ABSTRACT
The purpose of this study was to improve the transduction efficiency of adenoviral vectors (Ad) in human CD34+ hematopoietic progenitor cells. CD34+ cells from cord blood or mobilized peripheral blood were incubated with tumor necrosis factor-alpha (TNF-). After removal of free TNF-, the cells were infected with an Ad encoding green fluorescent protein (GFP). One day later, viable cells were counted and analyzed for GFP and CD34 by flow cytometry. To visualize vectoral trafficking, CD34+ cells were incubated with fluorophore-conjugated Ad. Plating efficiencies of hematopoietic progenitors before and after transduction were evaluated by methylcellulose assays. Pretreatment with TNF- increased the transduction efficiency more than twofold (39.2% versus 15.5%) in a dose-dependent manner and strongly improved the survival of GFP-positive CD34+ cells. Time course experiments showed that TNF- incubation times as short as 10 minutes were still effective. Neutralizing antibodies to TNF receptor II and RGD peptides diminished the TNF--dependent increase in transduction efficiency. No TNF--dependent increase in adenoviral receptors (coxsackie-adenovirus receptor, v?3-integrin) occurred. Analysis of viral binding demonstrated a significantly higher incidence of local concentrations of Ad along the cell surface (caps) in virus-positive cells of the TNF--treated group. Plating efficiency, especially the formation of granulocyte-macrophage colony forming units, was enhanced by TNF- pretreatment. We conclude that brief incubation with TNF- before addition of the Ad significantly increased the Ad transduction efficiency in CD34+ cells, and improved post-transduction survival of progenitors of the granulocyte-macrophage lineage. This finding correlates with increased Ad capping at the cell surface and suggests an alteration of Ad trafficking.
INTRODUCTION
Ex vivo adenoviral transduction of CD34+ progenitor cells has been an objective in the field of gene therapy in the past two decades . Technical obstacles are a low transduction efficiency, which can only be overcome by a tremendous increase in viral load (103 multiplicity of infection ), leading to a significant increase in toxicity and a diminished cloning efficiency .
Tumor necrosis factor-alpha (TNF-) is known for its proliferation and apoptosis-inducing capacities . Adenovirally infected cells have been demonstrated to escape TNF--induced apoptosis by expressing inhibitory glycoproteins of the viral expression cassette . At least three open reading frames within the early region 3, including 14.7K (also called receptor internalization and degradation ?), the complex of 10.4K and 10.5K, and the adenoviral glycoprotein E1B 19K, block TNF-dependent apoptosis .
In this study, we demonstrate that preincubation of CD34+ cells with TNF- increases the transduction efficiency, cell survival, and plating efficiency of hematopoietic progenitors.
MATERIALS AND METHODS
Adenoviral Vectors
In the present study we used AdGFP, an E1–E3– replication-deficient, serotype 5 adenoviral vector (Ad) that contains the green fluorescent protein (GFP) reporter gene driven by a cytomegalovirus early-intermediate promoter enhancer in the E1 position . Viral particle concentrations were determined by the absorbance at 260 nm and the extinction coefficient for Ad of 9.09 x 10–12 ml/particles and centimeters . Stocks were titrated on 293 cells (American Type Culture Collection; Rockville, MD; http://www.atcc.org) by plaque assays, and titers were expressed as infectious particles of plaque-forming units (pfu)/ml. The MOI was calculated as pfu/target cell. Carbocyanine dye Cy3 (Amersham; Arlington Heights, IL; http://www.amershambiosciences.com) covalently conjugated to the capsid of Ad was used for trafficking experiments .
Preparation and TNF- Treatment of CD34+ Cells
CD34+ cells (0.5–1 x 105) were isolated from cord blood or mobilized peripheral blood by immunomagnetic bead separation (Miltenyi Biotech; Auburn, CA; http://www.miltenyibiotec.com) according to the manufacturer’s instructions. The purity of isolated CD34+ cells was >95% as analyzed by flow cytometry. Before starting the experiments, the thawed cells were washed with fetal bovine serum (FBS) and cultured overnight in complete culture medium (Iscove’s modified Dulbecco’s medium supplemented with 20% FBS, 2 mM L-glutamine, 50 μg/ml gentamicin, 20 mM mercaptoethanol). A total of 105 CD34+ cells were incubated with TNF- (specific activity 1 x 108 U/mg; Genentech; San Francisco, CA; http://www.gene.com) at 0.01, 0.1, and 1 μg/ml for 2 hours or were kept in complete medium only. For time course experiments, the cells were incubated with 1 μg/ml TNF-. Aliquots of 105 cells for viral transduction were removed after 10, 30, 60, 120, and 240 minutes. Cells were manually counted in a hemocytometer, and their viability was determined by trypan blue exclusion.
Adenoviral Infection
After removal of TNF- by washing the samples twice with FBS, the cells were infected with an AdGFP at an MOI of 500 for 12 hours in serum-free medium (X-Vivo 15; BioWhittaker, Inc.; Walkersville, MD; http://www.cambrex.com). Excess virus was removed and the cells were resuspended in complete culture medium plus 20 ng/ml c-kit ligand (stem cell factor; Amgen; Thousand Oaks, CA; http://www.amgen.com). Twenty-four hours later, viable cells were counted and analyzed for GFP and CD34 expression by flow cytometry.
For TNF- inhibition studies, CD34+ cells were mixed with either monoclonal anti-TNF- receptor I or II antibodies (anti-TNFR1 and 2, R&D Systems; Minneapolis, MN; http://www.rndsystems.com) for 1 hour at 37°C before incubation with TNF- (100 ng/ml). Antibody concentrations ranged from 0.1 μg/ml to 100 μg/ml. In some experiments, the cells were incubated with both anti-TNFR 1 and 2 antibodies. Fc receptor antibodies were used as negative controls. To antagonize adenoviral entry, TNF--stimulated CD34+ cells were incubated with either fiber (1.8 μg/ml), 100 nM RGD (GRGDSP), or a control peptide (GRGESP) (all GIBCO Life Technology; Gaithersburg, MD; http://www.lifetech.com) for 1 hour before addition of AdGFP. Percent inhibition was calculated as the ratio of the number of TNF--treated, transduced CD34+GFP+ cells with anti-TNFR or blocking substance compared with the number of CD34+GFP+ cells without neutralizing antibodies.
Flow Cytometry
To determine the number of transduced progenitors, cells were stained with phycoerythrin (PE)-conjugated, monoclonal anti-human CD34 (Immunotech; Marseilles, France) after transduction with AdGFP. Dead cells were excluded by propidium iodide. Apoptosis was assessed using annexin-V-PE (Pharmingen; San Diego, CA; http://www.bdbiosciences.com/pharmingen) as described previously .
To stain for coxsackie-adenovirus receptor (CAR), the monoclonal anti-CAR antibody RmcB , kindly provided by Phil Leopold, as the supernatant from a mouse hybridoma cell line, was dialyzed with phosphate-buffered saline (PBS) six times through a Centriprep YM-10 filter (Millipore; Bedford, MA; http://www.millipore.com) in order to remove protein residues. Concentrated anti-CAR was biotinylated according to the manufacturer’s instructions (EZ-Link Sulfo-NHS-LC Biotinylation Kit; Pierce; Rockford, IL; http://www.piercenet.com). Briefly, 2 mg of anti-CAR in 1 ml PBS were mixed with a 20-fold molar excess of biotin dissolved in distilled water. To measure CAR expression, the CD34+ cells were stained with anti-CAR/biotin first, then stained with avidin-fluorescein isothiocyanate (FITC; Immunotech) and anti-CD34-PE. A549 cells (ATCC) served as the CAR-positive controls. v?3 expression was assessed using the monoclonal FITCconjugated anti-human CD51/CD61 antibody (Pharmingen).
Clonogenic Assay
Pre- and post-transduction hematopoietic colony formation capacity was evaluated in methylcellulose cultures stimulated with interleukin-3, erythropoietin, and c-kit-ligand, similar to the method of MacKenzie et al. . Briefly, 1 x 103 cells/ml were plated in 30% fetal calf serum/Iscove’s modified Dulbecco’s medium and 0.8% methylcellulose (Dow Chemicals; Saddlebrook, NJ; http://www.dow.com), supplemented with 20 ng/ml c-kit-ligand, 50 ng/ml interleukin-3, and 6 U/ml erythropoietin (Amgen). Cultures were scored at 14 days for granulocyte-macrophage colony-forming units (CFU-GM), mixed colony-forming units (CFU-MIX), and BFU-E. In two experiments, fluorescence-activated cell sorted, GFP+ CD34+ cells were analyzed for cobblestone area formation on the murine stroma cell line MS-5 as previously described .
Fluorescence Microscopy
To visualize vectoral trafficking, 2 x 104 CD34+ cells were resuspended in 30 μl X-Vivo 15 and incubated for 30 minutes at 37°C with Cy3-conjugated Ad (1011 particles/ml, MOI 5,000). Excess virus was removed by washing the samples with PBS three times. The cells were placed on glass slides, cytospun (500 g, 3 minutes), and counterstained with 0.1 mg/ml Hoechst in mounting media (Permount; Fisher Scientific; Hanover Park, IL; http://www.fisherscientific.com). Ten images per slide were taken with a Microphot SA microscope (Nikon; Garden City, NY; http://www.nikon-image.com/eng) equipped with a 100 W mercury arc and a cooled charge-coupled device camera (Princeton Instruments; Trenton, NJ; http://www.prinsci.com), and recorded by the MetMorph program (Universal Imaging; West Chester, PA; http://www.image1.com) .
RESULTS AND DISCUSSION
Prior Incubation with TNF- Increased the Transduction Efficiency in CD34+ Cells
The use of high adenovector MOI (>1,000), which lead to a high transduction efficiency, is usually limited by vector-induced cytotoxicity . In this study, prior incubation with TNF- increased the adenoviral transduction efficiency significantly in a dose-dependent manner (Fig. 1). The effects were observed in both fresh and frozen purified CD34+ cells independent of the cell source. The highest GFP positivity (on average 39.2%) was obtained with 1 μg/ml TNF-. Initial saturation was observed at 0.1 μg/ml. In the absence of TNF-, similar transduction efficiencies have been achieved by increasing the viral load (MOI >1,000) , but this was generally associated with a higher cytotoxicity. At the MOI used here, cell viability without TNF- ranged between 50%-72%, while at the highest TNF- concentration (1 μg/ml), 61.9%-97.8% of the cells were viable as judged by trypan blue exclusion. When the cells were pretreated with TNF-, not only the percentage of GFP+ CD34+ cells, but also the number of vital cells rose significantly. This resulted in a sixfold increase of GFP+ CD34+ cells (24.4 ± 4.1 versus 4.2 ± 0.8 x 103), which was still present on the sixth day post-transduction (Fig. 2A, 2B). The same TNF--associated effect was also present at lower MOIs of 10 and 100, while the maximum transduction effciency was achieved at an MOI of 500. CD34+ cells isolated from both cord blood and leukapheresis products showed a comparable increase in transduction following TNF--treatment. Time course experiments demonstrated that the highest number of CD34+GFP+ cells (8.4 x 104) occurs after 2-hours of incubation with TNF-. However, a significant increase in the percentage of cells transduced was observed already after 10 minutes of TNF- exposure followed by adenoviral infection and subsequent culture for 24–48 hours (51.5% versus 23.6%). Annexin V staining excluded the possibility of TNF--dependent increased apoptosis (Table 1), while in the absence of TNF-, more cells became necrotic after adenoviral infection. This confirms a direct cytotoxic effect of the adenoviral vector.
Figure 1. TNF- concentration-dependent increase in GFP-expressing CD34+ cells 24 hours after adenoviral infection. Cord blood-derived CD34+ cells were incubated with TNF- at various concentrations (0; 0.01; 0.1; 1 μg/ml, indicated on top of each graph) prior to adenoviral infection (MOI 500). At the highest concentration of TNF-, twice as many cells expressed GFP as in the absence of TNF-. The numbers in each box represent the frequency of positive cells.
Figure 2. Cell count, transduction efficiency, and cell viability. Twenty-four hours after adenoviral transduction post-TNF- treatment (A), not only the percentage of GFP+ CD34+ cells (black), but also the number of viable cells had doubled (63.8 ± 7.5 versus 29.9 ± 3.5 x 103, white), leading to a sixfold increase of CD34+GFP+ cells. Five days after transduction (B), the frequency of GFP expression in TNF--treated CD34+ cells was still higher, but the cell counts and viabilities (gray) were not significantly different. Average results of six or seven independent experiments ± standard error (four cord blood, three leukapheresis products). *p < 0.05 compared with no TNF- treatment.
Table 1. TNF- incubation time, percentage of GFP+ cells and Annexin-positive cells, and absolute number of GFP+ cells
We investigated whether blocking its signaling pathway could inhibit the supportive effect of TNF-. Usually, using a 10-fold higher concentration of anti-TNFR than TNF- inhibits cellular effects of TNF- . Blocking studies demonstrated a significant (p = 0.04), almost 40% inhibition of transduction when TNF- receptor II (TNFR2) was neutralized (Fig. 3). In contrast, neutralization of TNF- receptor I (TNFR1) had no effect on adenoviral transduction efficiency. Simultaneous incubation with both anti-TNF receptors had no cumulative inhibitory effect confirming that the observed mechanism is partially triggered through interaction with TNF receptor II. The reason why anti-TNFR2 did not completely block the TNF-dependent increase in the transduction efficiency remains unclear in our study. However, it is possible that the anti-TNFR2-TNFR complex may induce the uptake of adenoviral particles by an unidentified mechanism, thereby preventing a complete inhibition of the TNF--dependent increase of the adenoviral transduction efficiency. This TNFR neutralization study is also consistent with previous publications showing that TNF receptor I is responsible for all in vitro growth-related effects on more committed human bone marrow progenitor cells , while TNF receptor II is directly involved in signaling growth inhibition in the most primitive human and murine stem cells .
Figure 3. Inhibition of TNF- by neutralizing antibodies. CD34+ cells were mixed with either anti-TNFRI or II antibodies for 1 hour at 37°C before incubation with TNF- (100 ng/ml). Soluble anti-TNFR2 (black) antibody reduced Ad transduction efficiency in CD34+ cells to about 60% at the highest antibody concentration (100 μg/ml). Blockage of TNFR1 (white) had no effect. The ordinate represents the ratio of the number of transduced cells compared to the number of TNF--treated transduced cells in the absence of antibodies. The figures represent the average results of two independent experiments (CD34+ cells derived from one cord blood and one leukapheresis product). *p < 0.05 compared to anti-Fc.
Adenoviral Receptors
Other approaches to increase the Ad transduction efficiency are based on the upregulation of Ad receptors . Therefore, we analyzed the amount of CAR and v?3-integrins present on the surface of TNF--treated CD34+ cells. While CAR is responsible for the initial attachment of the adenovirus on the cell surface , integrins are involved in the internalization of the virus . Flow cytometry showed a slight but measurable upregulation of CAR in TNF--treated CD34+ cells in a dose-dependent manner. A significant upregulation was seen in cord blood-derived CD34+ cells that were subsequently cultured in serum-free medium for 12 hours to simulate the conditions of Ad infection (Fig. 4A, 4B). However, since adenoviral internalization usually occurs within the first 30 minutes of infection , this upregulation can hardly be responsible for the improved transduction efficiency. No increase in v?3-integrin expression was observed on the cell surface. Although the number of cell surface receptors involved in Ad cell attachment and internalization, e.g., CAR and v?3-integrins , did not increase after TNF- treatment, we attempted to block Ad entry using the receptor antagonists, fiber, and RGD peptide. Fiber did not influence GFP positivity, but the integrin-binding motif RGD reduced the number of GFP+ cells by 41% (Fig. 5). This reflected not only a decrease in the absolute cell count but also a reduced percentage of GFP+ cells (28.9% versus 40.5%). These findings suggest that the TNF--related increase in transduction efficiency is not caused by an increase in the amount of CAR or v?3 integrins on the cell surface, but rather by the amount of other integrins, such as v?5, which also play a role in adenoviral infection and are blocked by RGD .
Figure 4. CAR expression. CAR expression in leukapheresis-derived (A) and (B) (1–5; 8) and cord blood-derived CD34+ cells (C) (1; 9–11) relative to TNF- concentration. While no upregulation of CAR was observed in CD34+ cells from leukapheresis products, a significant increase was observed in cord-blood derived CD34+ cells treated with 1 μg/ml TNF-, which were subsequently cultured in serum-free medium for 12 hours. Twelve hours represents the incubation period with the adenoviral vector. The figures reflect the representative results from one of three independent experiments. 1) IgG; 2 and 9) no TNF; 3) TNF- 0.01 μg/ml; 4) TNF- 0.1 μg/ml; 5 and 10) TNF- 1 μg/ml; 8 and 11) TNF- 1 μg/ml, after 12 hours in serum-free medium.
Figure 5. Inhibition of Ad transduction by adenoviral receptors. TNF--stimulated cord and peripheral blood-derived CD34+ cells were incubated with either fiber (1.8 μg/ml), 100 nM RGD (GRGDSP), or a control peptide (GRGESP) for 1 hour before addition of AdGFP. While fiber had no effect, RGD peptide reduced the number of TNF--treated CD34+GFP+ cells by more than 40%. The control peptide (Control) did not affect the number of transduced cells. *p < 0.05 compared to the control.
Viral Attachment in Fluorescence Microscopy
Since no increase in Ad receptors, which could explain the increased transduction efficiency, was detected, we studied the direct interactions of the cells with cytochrome-labeled adenovirus by fluorescence microscopy (Fig. 6). Viral caps (the concentration of viruses at one site on the cell surface) appeared in CD34+ cells pretreated with TNF- (Fig. 6C). Some of these caps persisted after the cells had been cultured in serum-free medium for another 12 hours (Fig. 6D). Since the numbers of virus-positive cells were similar with and without TNF- pretreatment (Table 2), enhanced viral uptake can be attributed to the improved kinetics of adenoviral trafficking into CD34+ cells rather than to an increased percentage of CD34+ cells being attached by Ad vectors. That was reflected in the number of viral caps, which was significantly higher in the TNF--treated, virus-positive CD34+ cells than in the corresponding non-TNF--treated cells (18.1% ± 5% versus 8% ± 2.4%; p < 0.05; Table 2). In the context of phagocytosis, capping has been observed as a phenomenon that occurs shortly before internalization of materials concentrating on the cell surface . Pretreatment of CD34+ cells with TNF- could affect the Ad internalization pathway by triggering phosphatidylinositol-3-OH kinase activation, a signaling molecule involved in adenovirus internalization , thereby inducing the formation of caps. Therefore, the internalization of Ad vectors by TNF--treated CD34+ cells would appear to be a form of phagocytosis.
Figure 6. Adenoviral adherence to the cell surface of cord blood-derived CD34+ cells (size bar: 1 μm). In the non-TNF--treated CD34+ cells, the red-fluorescing Ad vectors dispersed along the cell surface and attached to the cells at various sites. After incubation with TNF-, the Ad vectors concentrated on one side of the cells, giving them a cap-like appearance (arrows). Hoechst was used to stain the nuclei of viable cells (blue). The figures demonstrate the representative results of four experiments (two cord blood, two leukapheresis products). A) Virus-free negative control; B) non-TNF--treated CD34+ cells; C) CD34+ cells incubated with TNF- (1 μg/ml, 2 hours); D) CD34+ cells incubated with TNF- (1 μg/ml, 2 hours) and subsequently cultured in serum-free medium for 12 hours.
Table 2. Adenoviral cap formation in CD34+ cells
TNF- Improved Plating Efficiency of Transduced CD34+ Cells
Previous studies demonstrated a mainly TNFR1-related bifunctional effect of TNF-. On the one hand, high concentrations of TNF- have a direct cytotoxic effect on more committed hematopoietic progenitors when the cells are exposed to TNF- for 24 hours or longer . Since TNF- is known to induce apoptosis , we were concerned that the observed effect could be related to defects in the cellular membrane that might lead to uncontrolled viral entry through membrane leakage. However, the main difference to our experimental approach is the short TNF--incubation period, which did not increase the apoptosis rate. In the absence of a viral vector, short-term incubation with TNF- did not increase the percentage of annexin+ propidium iodide– cells as determined after 3 days in culture (no TNF-: 4.6%; preincubation with 1 μg/ml TNF- for 2 hours: 3.5%). Similar observations have been described previously .
On the other hand, TNF- supports interleukin-3 and GM-CSF-dependent proliferation of hematopoietic progenitor cells at low concentrations , favoring the GM lineage . Therefore, incubation of CD34+ cells with TNF- before adenoviral infection increased the number and plating efficiency of the cells (Table 3). Notably, the number of CFU-GM increased in a TNF- dose-dependent manner, which was observed in both fresh and frozen CD34+ cells independent of their origin. Although the initial pretreatment values were not achieved (6.3 ± 2.7 x 103), TNF- helped the cells to overcome, in part, the adenoviral toxicity, and an almost fivefold expansion of CFU-GM was observed at the highest TNF- concentration (4.9 ± 1.4 versus 1.0 ± 0.51 x 103). An increase of CFU-GM formation may suggest that this technique could drive lineage-specific commitment or that the less primitive CD34+ cells are being transduced, both of which are unwanted. However, since agranulocytosis after myeoloablative therapy is one of the major risk factors for septic complications in bone marrow and stem cell transplantation , this might also represent a beneficial side effect.
Table 3. Plating efficiency of CD34+ cells before and after transduction.
Long-term culture-initiating cells from GFP+, fluorescence-sorted CD34 cells from leukapheresis products were investigated. The number of cobblestone areas observed after 2 weeks, which represents the precommitted precursors of the white blood cell lineage, was higher in the TNF--pretreated population (initial value 44.5 ± 17.1; no TNF- 42.5 ± 13 versus TNF- 51.3 ± 15.9), though this difference was not significant (p > 0.05). Thus, TNF- incubation not only raised the absolute cell number in a short time, but also improved long-term cell survival of adenovirally transduced CD34+ cells.
No enhancement of the transduction efficiency in TNF--treated CD34+ cells was achieved when we used the MIGR1 vector , a retroviral construct that drives gene expression off the murine stem cell virus promoter. Nathwany et al. recently demonstrated TNF--mediated doubling of the transduction efficiency of adeno-associated viral gene transfer in CD34+ cells . Their results, as well as ours, indicate that the TNF--induced activation of the CMV promoter, when present in retroviral and adenoviral vectors, led to a higher expression of the transduced genes.
CONCLUSION
TNF- not only increased the Ad transduction efficiency but also decreased the Ad-related cytotoxicity. The most likely explanations for these phenomena are the concentration of viruses on one side of the cell before entry (capping), TNFR2-dependent upregulation of integrins, and, possibly, activation of the CMV promoter. Since the toxicity of intravascular Ad vector administration has been confirmed, these findings suggest a new approach in the ex vivo gene manipulation of human hematopoietic progenitor cells. Further studies on nonobese diabetic severe-combined-immunodeficient mice are necessary to assess the number of transduced repopulating cells in vivo.
ACKNOWLEDGMENT
We would like to thank P. Leopold and R.G. Crystal for providing adenoviral vectors for the transduction and trafficking studies, and J.C. Mulloy for performing the retroviral infections. We are further indebted to R.G. Crystal for helpful discussions and P. Loeser for critical proofreading. A.M. was sponsored by a postdoctoral fellowship from the German Academic Exchange Service. This study was financed in part by NCI Grant P01 CA 59350, NIH Grants HL 61401 and HL 66952, NIH Cancer Center Support Grant CA 08748 (M.A.S.M.), and the BMBF grant 0311591 (A.M.).
REFERENCES
Neering SJ, Hardy SF, Minamoto D et al. Transduction of primitive human hematopoietic cells with recombinant adenovirus vectors. Blood 1996;88:1147–1155.
Frey BM, Hackett NR, Bergelson JM et al. High-efficiency gene transfer into ex vivo expanded human hematopoietic progenitors and precursor cells by adenovirus vectors. Blood 1998;91:2781–2792.
MacKenzie KL, Hackett NR, Crystal RG et al. Adenoviral vector-mediated gene transfer to primitive human hematopoietic progenitor cells: assessment of transduction and toxicity in long-term culture. Blood 2000;96:100–108.
Santiago-Schwarz F, Divaris N, Kay C et al. Mechanisms of tumor necrosis factor-granulocyte-macrophage colony-stimulating factor-induced dendritic cell development. Blood 1993;82:3019–3028.
Shisler J, Yang C, Walter B et al. The adenovirus E3-10.4K/14.5K complex mediates loss of cell surface Fas (CD95) and resistance to Fas-induced apoptosis. J Virol 1997;71:8299–8306.
Gooding LR, Aquino L, Duerksen-Hughes PJ et al. The E1B 19,000-molecular-weight protein of group C adenoviruses prevents tumor necrosis factor cytolysis of human cells but not of mouse cells. J Virol 1991;65:3083–3094.
Mittereder N, March KL, Trapnell BC. Evaluation of the concentration and bioactivity of adenovirus vectors for gene therapy. J Virol 1996;70:7498–7509.
Leopold PL, Ferris B, Grinberg I et al. Fluorescent virions: dynamic tracking of the pathway of adenoviral gene transfer vectors in living cells. Hum Gene Ther 1998;9:367–378.
Moldenhauer A, Nociari MM, Dias S et al. Optimized culture conditions for the generation of dendritic cells from peripheral blood monocytes. Vox Sang 2003;84:228–236.
Hsu KH, Lonberg-Holm K, Alstein B et al. A monoclonal antibody specific for the cellular receptor for the group B coxsackieviruses. J Virol 1988;62:1647–1652.
Vincent T, Harvey BG, Hogan SM et al. Rapid assessment of adenovirus serum neutralizing antibody titer based on quantitative, morphometric evaluation of capsid binding and intracellular trafficking: population analysis of adenovirus capsid association with cells is predictive of adenovirus infectivity. J Virol 2001;75:1516–1521.
Ferrajoli A, Talpaz M, Kurzrock R et al. Analysis of the effects of tumor necrosis factor inhibitors on human hematopoiesis. STEM CELLS 1993;11:112–119.
Rusten LS, Jacobsen SE. Tumor necrosis factor (TNF)-alpha directly inhibits human erythropoiesis in vitro: role of p55 and p75 TNF receptors. Blood 1995;85:989–996.
Jacobsen SE, Jacobsen FW, Fahlman C et al. TNF-alpha, the great imitator: role of p55 and p75 TNF receptors in hematopoiesis. STEM CELLS 1994;12(suppl 1):111–128.
Huang S, Endo RI, Nemerow GR. Upregulation of integrins alpha v beta 3 and alpha v beta 5 on human monocytes and T lymphocytes facilitates adenovirus-mediated gene delivery. J Virol 1995;69:2257–2263.
Kitazono M, Rao VK, Robey R et al. Histone deacetylase inhibitor FR901228 enhances adenovirus infection of hematopoietic cells. Blood 2002;99:2248–2251.
Shayakhmetov DM, Papayannopoulou T, Stamatoyannopoulos G et al. Efficient gene transfer into human CD34(+) cells by a retargeted adenovirus vector. J Virol 2000;74:2567–2583.
Rebel VI, Hartnett S, Denham J et al. Maturation and lineage-specific expression of the coxsackie and adenovirus receptor in hematopoietic cells. STEM CELLS 2000;18:176–182.
Magnusson MK, Hong SS, Boulanger P et al. Genetic retargeting of adenovirus: novel strategy employing "deknobbing" of the fiber. J Virol 2001;75:7280–7289.
Serrander L, Skarman P, Rasmussen B et al. Selective inhibition of IgG-mediated phagocytosis in gelsolin-deficient murine neutrophils. J Immunol 2000;165:2451–2457.
Li E, Brown SL, Von Seggern DJ et al. Signaling antibodies complexed with adenovirus circumvent CAR and integrin interactions and improve gene delivery. Gene Ther 2000;7:1593–1599.
Rusten LS, Jacobsen FW, Lesslauer W et al. Bifunctional effects of tumor necrosis factor alpha (TNF alpha) on the growth of mature and primitive human hematopoietic progenitor cells: involvement of p55 and p75 TNF receptors. Blood 1994;83:3152–3159.
Broxmeyer HE, Williams DE, Lu L et al. The suppressive influences of human tumor necrosis factors on bone marrow hematopoietic progenitor cells from normal donors and patients with leukemia: synergism of tumor necrosis factor and interferon-gamma. J Immunol 1986;136:4487–4495.
Akahane K, Hosoi T, Urabe A et al. Effects of recombinant human tumor necrosis factor (rhTNF) on normal human and mouse hemopoietic progenitor cells. Int J Cell Cloning 1987;5:16–26.
Murase T, Hotta T, Saito H et al. Effect of recombinant human tumor necrosis factor on the colony growth of human leukemia progenitor cells and normal hematopoietic progenitor cells. Blood 1987;69:467–472.
Caux C, Saeland S, Favre C et al. Tumor necrosis factor-alpha strongly potentiates interleukin-3 and granulocyte-macrophage colony-stimulating factor-induced proliferation of human CD34+ hematopoietic progenitor cells. Blood 1990;75:2292–2298.
K?rbling M, Anderlini P. Peripheral blood stem cell versus bone marrow allotransplantation: does the source of hematopoietic stem cells matter? Blood 2001;98:2900–2908.
Mulloy JC, Cammenga J, MacKenzie KL et al. The AML1-ETO fusion protein promotes the expansion of human hematopoietic stem cells. Blood 2002;99:15–23.
Nathwani AC, Hanawa H, Vandergriff J et al. Efficient gene transfer into human cord blood CD34+ cells and the CD34+CD38– subset using highly purified recombinant adeno-associated viral vector preparations that are free of helper virus and wild-type AAV. Gene Ther 2000;7:183–195.(A. Moldenhauera, J-H Shie)
b Memorial Sloan-Kettering Cancer Center, New York, New York, USA
Key Words. TNF- ? CD34+ cells ? Adenovirus ? Transduction ? Capping
Anja Moldenhauer, M.D., Institute for Transfusion Medicine, Charité—Universit?tsmedizin Berlin, Augustenburger Platz 1, 13351 Berlin, Germany. Telephone: 49-160-1090837; Fax: 49-450 553988; e-mail: anja.moldenhauer@charite.de
ABSTRACT
The purpose of this study was to improve the transduction efficiency of adenoviral vectors (Ad) in human CD34+ hematopoietic progenitor cells. CD34+ cells from cord blood or mobilized peripheral blood were incubated with tumor necrosis factor-alpha (TNF-). After removal of free TNF-, the cells were infected with an Ad encoding green fluorescent protein (GFP). One day later, viable cells were counted and analyzed for GFP and CD34 by flow cytometry. To visualize vectoral trafficking, CD34+ cells were incubated with fluorophore-conjugated Ad. Plating efficiencies of hematopoietic progenitors before and after transduction were evaluated by methylcellulose assays. Pretreatment with TNF- increased the transduction efficiency more than twofold (39.2% versus 15.5%) in a dose-dependent manner and strongly improved the survival of GFP-positive CD34+ cells. Time course experiments showed that TNF- incubation times as short as 10 minutes were still effective. Neutralizing antibodies to TNF receptor II and RGD peptides diminished the TNF--dependent increase in transduction efficiency. No TNF--dependent increase in adenoviral receptors (coxsackie-adenovirus receptor, v?3-integrin) occurred. Analysis of viral binding demonstrated a significantly higher incidence of local concentrations of Ad along the cell surface (caps) in virus-positive cells of the TNF--treated group. Plating efficiency, especially the formation of granulocyte-macrophage colony forming units, was enhanced by TNF- pretreatment. We conclude that brief incubation with TNF- before addition of the Ad significantly increased the Ad transduction efficiency in CD34+ cells, and improved post-transduction survival of progenitors of the granulocyte-macrophage lineage. This finding correlates with increased Ad capping at the cell surface and suggests an alteration of Ad trafficking.
INTRODUCTION
Ex vivo adenoviral transduction of CD34+ progenitor cells has been an objective in the field of gene therapy in the past two decades . Technical obstacles are a low transduction efficiency, which can only be overcome by a tremendous increase in viral load (103 multiplicity of infection ), leading to a significant increase in toxicity and a diminished cloning efficiency .
Tumor necrosis factor-alpha (TNF-) is known for its proliferation and apoptosis-inducing capacities . Adenovirally infected cells have been demonstrated to escape TNF--induced apoptosis by expressing inhibitory glycoproteins of the viral expression cassette . At least three open reading frames within the early region 3, including 14.7K (also called receptor internalization and degradation ?), the complex of 10.4K and 10.5K, and the adenoviral glycoprotein E1B 19K, block TNF-dependent apoptosis .
In this study, we demonstrate that preincubation of CD34+ cells with TNF- increases the transduction efficiency, cell survival, and plating efficiency of hematopoietic progenitors.
MATERIALS AND METHODS
Adenoviral Vectors
In the present study we used AdGFP, an E1–E3– replication-deficient, serotype 5 adenoviral vector (Ad) that contains the green fluorescent protein (GFP) reporter gene driven by a cytomegalovirus early-intermediate promoter enhancer in the E1 position . Viral particle concentrations were determined by the absorbance at 260 nm and the extinction coefficient for Ad of 9.09 x 10–12 ml/particles and centimeters . Stocks were titrated on 293 cells (American Type Culture Collection; Rockville, MD; http://www.atcc.org) by plaque assays, and titers were expressed as infectious particles of plaque-forming units (pfu)/ml. The MOI was calculated as pfu/target cell. Carbocyanine dye Cy3 (Amersham; Arlington Heights, IL; http://www.amershambiosciences.com) covalently conjugated to the capsid of Ad was used for trafficking experiments .
Preparation and TNF- Treatment of CD34+ Cells
CD34+ cells (0.5–1 x 105) were isolated from cord blood or mobilized peripheral blood by immunomagnetic bead separation (Miltenyi Biotech; Auburn, CA; http://www.miltenyibiotec.com) according to the manufacturer’s instructions. The purity of isolated CD34+ cells was >95% as analyzed by flow cytometry. Before starting the experiments, the thawed cells were washed with fetal bovine serum (FBS) and cultured overnight in complete culture medium (Iscove’s modified Dulbecco’s medium supplemented with 20% FBS, 2 mM L-glutamine, 50 μg/ml gentamicin, 20 mM mercaptoethanol). A total of 105 CD34+ cells were incubated with TNF- (specific activity 1 x 108 U/mg; Genentech; San Francisco, CA; http://www.gene.com) at 0.01, 0.1, and 1 μg/ml for 2 hours or were kept in complete medium only. For time course experiments, the cells were incubated with 1 μg/ml TNF-. Aliquots of 105 cells for viral transduction were removed after 10, 30, 60, 120, and 240 minutes. Cells were manually counted in a hemocytometer, and their viability was determined by trypan blue exclusion.
Adenoviral Infection
After removal of TNF- by washing the samples twice with FBS, the cells were infected with an AdGFP at an MOI of 500 for 12 hours in serum-free medium (X-Vivo 15; BioWhittaker, Inc.; Walkersville, MD; http://www.cambrex.com). Excess virus was removed and the cells were resuspended in complete culture medium plus 20 ng/ml c-kit ligand (stem cell factor; Amgen; Thousand Oaks, CA; http://www.amgen.com). Twenty-four hours later, viable cells were counted and analyzed for GFP and CD34 expression by flow cytometry.
For TNF- inhibition studies, CD34+ cells were mixed with either monoclonal anti-TNF- receptor I or II antibodies (anti-TNFR1 and 2, R&D Systems; Minneapolis, MN; http://www.rndsystems.com) for 1 hour at 37°C before incubation with TNF- (100 ng/ml). Antibody concentrations ranged from 0.1 μg/ml to 100 μg/ml. In some experiments, the cells were incubated with both anti-TNFR 1 and 2 antibodies. Fc receptor antibodies were used as negative controls. To antagonize adenoviral entry, TNF--stimulated CD34+ cells were incubated with either fiber (1.8 μg/ml), 100 nM RGD (GRGDSP), or a control peptide (GRGESP) (all GIBCO Life Technology; Gaithersburg, MD; http://www.lifetech.com) for 1 hour before addition of AdGFP. Percent inhibition was calculated as the ratio of the number of TNF--treated, transduced CD34+GFP+ cells with anti-TNFR or blocking substance compared with the number of CD34+GFP+ cells without neutralizing antibodies.
Flow Cytometry
To determine the number of transduced progenitors, cells were stained with phycoerythrin (PE)-conjugated, monoclonal anti-human CD34 (Immunotech; Marseilles, France) after transduction with AdGFP. Dead cells were excluded by propidium iodide. Apoptosis was assessed using annexin-V-PE (Pharmingen; San Diego, CA; http://www.bdbiosciences.com/pharmingen) as described previously .
To stain for coxsackie-adenovirus receptor (CAR), the monoclonal anti-CAR antibody RmcB , kindly provided by Phil Leopold, as the supernatant from a mouse hybridoma cell line, was dialyzed with phosphate-buffered saline (PBS) six times through a Centriprep YM-10 filter (Millipore; Bedford, MA; http://www.millipore.com) in order to remove protein residues. Concentrated anti-CAR was biotinylated according to the manufacturer’s instructions (EZ-Link Sulfo-NHS-LC Biotinylation Kit; Pierce; Rockford, IL; http://www.piercenet.com). Briefly, 2 mg of anti-CAR in 1 ml PBS were mixed with a 20-fold molar excess of biotin dissolved in distilled water. To measure CAR expression, the CD34+ cells were stained with anti-CAR/biotin first, then stained with avidin-fluorescein isothiocyanate (FITC; Immunotech) and anti-CD34-PE. A549 cells (ATCC) served as the CAR-positive controls. v?3 expression was assessed using the monoclonal FITCconjugated anti-human CD51/CD61 antibody (Pharmingen).
Clonogenic Assay
Pre- and post-transduction hematopoietic colony formation capacity was evaluated in methylcellulose cultures stimulated with interleukin-3, erythropoietin, and c-kit-ligand, similar to the method of MacKenzie et al. . Briefly, 1 x 103 cells/ml were plated in 30% fetal calf serum/Iscove’s modified Dulbecco’s medium and 0.8% methylcellulose (Dow Chemicals; Saddlebrook, NJ; http://www.dow.com), supplemented with 20 ng/ml c-kit-ligand, 50 ng/ml interleukin-3, and 6 U/ml erythropoietin (Amgen). Cultures were scored at 14 days for granulocyte-macrophage colony-forming units (CFU-GM), mixed colony-forming units (CFU-MIX), and BFU-E. In two experiments, fluorescence-activated cell sorted, GFP+ CD34+ cells were analyzed for cobblestone area formation on the murine stroma cell line MS-5 as previously described .
Fluorescence Microscopy
To visualize vectoral trafficking, 2 x 104 CD34+ cells were resuspended in 30 μl X-Vivo 15 and incubated for 30 minutes at 37°C with Cy3-conjugated Ad (1011 particles/ml, MOI 5,000). Excess virus was removed by washing the samples with PBS three times. The cells were placed on glass slides, cytospun (500 g, 3 minutes), and counterstained with 0.1 mg/ml Hoechst in mounting media (Permount; Fisher Scientific; Hanover Park, IL; http://www.fisherscientific.com). Ten images per slide were taken with a Microphot SA microscope (Nikon; Garden City, NY; http://www.nikon-image.com/eng) equipped with a 100 W mercury arc and a cooled charge-coupled device camera (Princeton Instruments; Trenton, NJ; http://www.prinsci.com), and recorded by the MetMorph program (Universal Imaging; West Chester, PA; http://www.image1.com) .
RESULTS AND DISCUSSION
Prior Incubation with TNF- Increased the Transduction Efficiency in CD34+ Cells
The use of high adenovector MOI (>1,000), which lead to a high transduction efficiency, is usually limited by vector-induced cytotoxicity . In this study, prior incubation with TNF- increased the adenoviral transduction efficiency significantly in a dose-dependent manner (Fig. 1). The effects were observed in both fresh and frozen purified CD34+ cells independent of the cell source. The highest GFP positivity (on average 39.2%) was obtained with 1 μg/ml TNF-. Initial saturation was observed at 0.1 μg/ml. In the absence of TNF-, similar transduction efficiencies have been achieved by increasing the viral load (MOI >1,000) , but this was generally associated with a higher cytotoxicity. At the MOI used here, cell viability without TNF- ranged between 50%-72%, while at the highest TNF- concentration (1 μg/ml), 61.9%-97.8% of the cells were viable as judged by trypan blue exclusion. When the cells were pretreated with TNF-, not only the percentage of GFP+ CD34+ cells, but also the number of vital cells rose significantly. This resulted in a sixfold increase of GFP+ CD34+ cells (24.4 ± 4.1 versus 4.2 ± 0.8 x 103), which was still present on the sixth day post-transduction (Fig. 2A, 2B). The same TNF--associated effect was also present at lower MOIs of 10 and 100, while the maximum transduction effciency was achieved at an MOI of 500. CD34+ cells isolated from both cord blood and leukapheresis products showed a comparable increase in transduction following TNF--treatment. Time course experiments demonstrated that the highest number of CD34+GFP+ cells (8.4 x 104) occurs after 2-hours of incubation with TNF-. However, a significant increase in the percentage of cells transduced was observed already after 10 minutes of TNF- exposure followed by adenoviral infection and subsequent culture for 24–48 hours (51.5% versus 23.6%). Annexin V staining excluded the possibility of TNF--dependent increased apoptosis (Table 1), while in the absence of TNF-, more cells became necrotic after adenoviral infection. This confirms a direct cytotoxic effect of the adenoviral vector.
Figure 1. TNF- concentration-dependent increase in GFP-expressing CD34+ cells 24 hours after adenoviral infection. Cord blood-derived CD34+ cells were incubated with TNF- at various concentrations (0; 0.01; 0.1; 1 μg/ml, indicated on top of each graph) prior to adenoviral infection (MOI 500). At the highest concentration of TNF-, twice as many cells expressed GFP as in the absence of TNF-. The numbers in each box represent the frequency of positive cells.
Figure 2. Cell count, transduction efficiency, and cell viability. Twenty-four hours after adenoviral transduction post-TNF- treatment (A), not only the percentage of GFP+ CD34+ cells (black), but also the number of viable cells had doubled (63.8 ± 7.5 versus 29.9 ± 3.5 x 103, white), leading to a sixfold increase of CD34+GFP+ cells. Five days after transduction (B), the frequency of GFP expression in TNF--treated CD34+ cells was still higher, but the cell counts and viabilities (gray) were not significantly different. Average results of six or seven independent experiments ± standard error (four cord blood, three leukapheresis products). *p < 0.05 compared with no TNF- treatment.
Table 1. TNF- incubation time, percentage of GFP+ cells and Annexin-positive cells, and absolute number of GFP+ cells
We investigated whether blocking its signaling pathway could inhibit the supportive effect of TNF-. Usually, using a 10-fold higher concentration of anti-TNFR than TNF- inhibits cellular effects of TNF- . Blocking studies demonstrated a significant (p = 0.04), almost 40% inhibition of transduction when TNF- receptor II (TNFR2) was neutralized (Fig. 3). In contrast, neutralization of TNF- receptor I (TNFR1) had no effect on adenoviral transduction efficiency. Simultaneous incubation with both anti-TNF receptors had no cumulative inhibitory effect confirming that the observed mechanism is partially triggered through interaction with TNF receptor II. The reason why anti-TNFR2 did not completely block the TNF-dependent increase in the transduction efficiency remains unclear in our study. However, it is possible that the anti-TNFR2-TNFR complex may induce the uptake of adenoviral particles by an unidentified mechanism, thereby preventing a complete inhibition of the TNF--dependent increase of the adenoviral transduction efficiency. This TNFR neutralization study is also consistent with previous publications showing that TNF receptor I is responsible for all in vitro growth-related effects on more committed human bone marrow progenitor cells , while TNF receptor II is directly involved in signaling growth inhibition in the most primitive human and murine stem cells .
Figure 3. Inhibition of TNF- by neutralizing antibodies. CD34+ cells were mixed with either anti-TNFRI or II antibodies for 1 hour at 37°C before incubation with TNF- (100 ng/ml). Soluble anti-TNFR2 (black) antibody reduced Ad transduction efficiency in CD34+ cells to about 60% at the highest antibody concentration (100 μg/ml). Blockage of TNFR1 (white) had no effect. The ordinate represents the ratio of the number of transduced cells compared to the number of TNF--treated transduced cells in the absence of antibodies. The figures represent the average results of two independent experiments (CD34+ cells derived from one cord blood and one leukapheresis product). *p < 0.05 compared to anti-Fc.
Adenoviral Receptors
Other approaches to increase the Ad transduction efficiency are based on the upregulation of Ad receptors . Therefore, we analyzed the amount of CAR and v?3-integrins present on the surface of TNF--treated CD34+ cells. While CAR is responsible for the initial attachment of the adenovirus on the cell surface , integrins are involved in the internalization of the virus . Flow cytometry showed a slight but measurable upregulation of CAR in TNF--treated CD34+ cells in a dose-dependent manner. A significant upregulation was seen in cord blood-derived CD34+ cells that were subsequently cultured in serum-free medium for 12 hours to simulate the conditions of Ad infection (Fig. 4A, 4B). However, since adenoviral internalization usually occurs within the first 30 minutes of infection , this upregulation can hardly be responsible for the improved transduction efficiency. No increase in v?3-integrin expression was observed on the cell surface. Although the number of cell surface receptors involved in Ad cell attachment and internalization, e.g., CAR and v?3-integrins , did not increase after TNF- treatment, we attempted to block Ad entry using the receptor antagonists, fiber, and RGD peptide. Fiber did not influence GFP positivity, but the integrin-binding motif RGD reduced the number of GFP+ cells by 41% (Fig. 5). This reflected not only a decrease in the absolute cell count but also a reduced percentage of GFP+ cells (28.9% versus 40.5%). These findings suggest that the TNF--related increase in transduction efficiency is not caused by an increase in the amount of CAR or v?3 integrins on the cell surface, but rather by the amount of other integrins, such as v?5, which also play a role in adenoviral infection and are blocked by RGD .
Figure 4. CAR expression. CAR expression in leukapheresis-derived (A) and (B) (1–5; 8) and cord blood-derived CD34+ cells (C) (1; 9–11) relative to TNF- concentration. While no upregulation of CAR was observed in CD34+ cells from leukapheresis products, a significant increase was observed in cord-blood derived CD34+ cells treated with 1 μg/ml TNF-, which were subsequently cultured in serum-free medium for 12 hours. Twelve hours represents the incubation period with the adenoviral vector. The figures reflect the representative results from one of three independent experiments. 1) IgG; 2 and 9) no TNF; 3) TNF- 0.01 μg/ml; 4) TNF- 0.1 μg/ml; 5 and 10) TNF- 1 μg/ml; 8 and 11) TNF- 1 μg/ml, after 12 hours in serum-free medium.
Figure 5. Inhibition of Ad transduction by adenoviral receptors. TNF--stimulated cord and peripheral blood-derived CD34+ cells were incubated with either fiber (1.8 μg/ml), 100 nM RGD (GRGDSP), or a control peptide (GRGESP) for 1 hour before addition of AdGFP. While fiber had no effect, RGD peptide reduced the number of TNF--treated CD34+GFP+ cells by more than 40%. The control peptide (Control) did not affect the number of transduced cells. *p < 0.05 compared to the control.
Viral Attachment in Fluorescence Microscopy
Since no increase in Ad receptors, which could explain the increased transduction efficiency, was detected, we studied the direct interactions of the cells with cytochrome-labeled adenovirus by fluorescence microscopy (Fig. 6). Viral caps (the concentration of viruses at one site on the cell surface) appeared in CD34+ cells pretreated with TNF- (Fig. 6C). Some of these caps persisted after the cells had been cultured in serum-free medium for another 12 hours (Fig. 6D). Since the numbers of virus-positive cells were similar with and without TNF- pretreatment (Table 2), enhanced viral uptake can be attributed to the improved kinetics of adenoviral trafficking into CD34+ cells rather than to an increased percentage of CD34+ cells being attached by Ad vectors. That was reflected in the number of viral caps, which was significantly higher in the TNF--treated, virus-positive CD34+ cells than in the corresponding non-TNF--treated cells (18.1% ± 5% versus 8% ± 2.4%; p < 0.05; Table 2). In the context of phagocytosis, capping has been observed as a phenomenon that occurs shortly before internalization of materials concentrating on the cell surface . Pretreatment of CD34+ cells with TNF- could affect the Ad internalization pathway by triggering phosphatidylinositol-3-OH kinase activation, a signaling molecule involved in adenovirus internalization , thereby inducing the formation of caps. Therefore, the internalization of Ad vectors by TNF--treated CD34+ cells would appear to be a form of phagocytosis.
Figure 6. Adenoviral adherence to the cell surface of cord blood-derived CD34+ cells (size bar: 1 μm). In the non-TNF--treated CD34+ cells, the red-fluorescing Ad vectors dispersed along the cell surface and attached to the cells at various sites. After incubation with TNF-, the Ad vectors concentrated on one side of the cells, giving them a cap-like appearance (arrows). Hoechst was used to stain the nuclei of viable cells (blue). The figures demonstrate the representative results of four experiments (two cord blood, two leukapheresis products). A) Virus-free negative control; B) non-TNF--treated CD34+ cells; C) CD34+ cells incubated with TNF- (1 μg/ml, 2 hours); D) CD34+ cells incubated with TNF- (1 μg/ml, 2 hours) and subsequently cultured in serum-free medium for 12 hours.
Table 2. Adenoviral cap formation in CD34+ cells
TNF- Improved Plating Efficiency of Transduced CD34+ Cells
Previous studies demonstrated a mainly TNFR1-related bifunctional effect of TNF-. On the one hand, high concentrations of TNF- have a direct cytotoxic effect on more committed hematopoietic progenitors when the cells are exposed to TNF- for 24 hours or longer . Since TNF- is known to induce apoptosis , we were concerned that the observed effect could be related to defects in the cellular membrane that might lead to uncontrolled viral entry through membrane leakage. However, the main difference to our experimental approach is the short TNF--incubation period, which did not increase the apoptosis rate. In the absence of a viral vector, short-term incubation with TNF- did not increase the percentage of annexin+ propidium iodide– cells as determined after 3 days in culture (no TNF-: 4.6%; preincubation with 1 μg/ml TNF- for 2 hours: 3.5%). Similar observations have been described previously .
On the other hand, TNF- supports interleukin-3 and GM-CSF-dependent proliferation of hematopoietic progenitor cells at low concentrations , favoring the GM lineage . Therefore, incubation of CD34+ cells with TNF- before adenoviral infection increased the number and plating efficiency of the cells (Table 3). Notably, the number of CFU-GM increased in a TNF- dose-dependent manner, which was observed in both fresh and frozen CD34+ cells independent of their origin. Although the initial pretreatment values were not achieved (6.3 ± 2.7 x 103), TNF- helped the cells to overcome, in part, the adenoviral toxicity, and an almost fivefold expansion of CFU-GM was observed at the highest TNF- concentration (4.9 ± 1.4 versus 1.0 ± 0.51 x 103). An increase of CFU-GM formation may suggest that this technique could drive lineage-specific commitment or that the less primitive CD34+ cells are being transduced, both of which are unwanted. However, since agranulocytosis after myeoloablative therapy is one of the major risk factors for septic complications in bone marrow and stem cell transplantation , this might also represent a beneficial side effect.
Table 3. Plating efficiency of CD34+ cells before and after transduction.
Long-term culture-initiating cells from GFP+, fluorescence-sorted CD34 cells from leukapheresis products were investigated. The number of cobblestone areas observed after 2 weeks, which represents the precommitted precursors of the white blood cell lineage, was higher in the TNF--pretreated population (initial value 44.5 ± 17.1; no TNF- 42.5 ± 13 versus TNF- 51.3 ± 15.9), though this difference was not significant (p > 0.05). Thus, TNF- incubation not only raised the absolute cell number in a short time, but also improved long-term cell survival of adenovirally transduced CD34+ cells.
No enhancement of the transduction efficiency in TNF--treated CD34+ cells was achieved when we used the MIGR1 vector , a retroviral construct that drives gene expression off the murine stem cell virus promoter. Nathwany et al. recently demonstrated TNF--mediated doubling of the transduction efficiency of adeno-associated viral gene transfer in CD34+ cells . Their results, as well as ours, indicate that the TNF--induced activation of the CMV promoter, when present in retroviral and adenoviral vectors, led to a higher expression of the transduced genes.
CONCLUSION
TNF- not only increased the Ad transduction efficiency but also decreased the Ad-related cytotoxicity. The most likely explanations for these phenomena are the concentration of viruses on one side of the cell before entry (capping), TNFR2-dependent upregulation of integrins, and, possibly, activation of the CMV promoter. Since the toxicity of intravascular Ad vector administration has been confirmed, these findings suggest a new approach in the ex vivo gene manipulation of human hematopoietic progenitor cells. Further studies on nonobese diabetic severe-combined-immunodeficient mice are necessary to assess the number of transduced repopulating cells in vivo.
ACKNOWLEDGMENT
We would like to thank P. Leopold and R.G. Crystal for providing adenoviral vectors for the transduction and trafficking studies, and J.C. Mulloy for performing the retroviral infections. We are further indebted to R.G. Crystal for helpful discussions and P. Loeser for critical proofreading. A.M. was sponsored by a postdoctoral fellowship from the German Academic Exchange Service. This study was financed in part by NCI Grant P01 CA 59350, NIH Grants HL 61401 and HL 66952, NIH Cancer Center Support Grant CA 08748 (M.A.S.M.), and the BMBF grant 0311591 (A.M.).
REFERENCES
Neering SJ, Hardy SF, Minamoto D et al. Transduction of primitive human hematopoietic cells with recombinant adenovirus vectors. Blood 1996;88:1147–1155.
Frey BM, Hackett NR, Bergelson JM et al. High-efficiency gene transfer into ex vivo expanded human hematopoietic progenitors and precursor cells by adenovirus vectors. Blood 1998;91:2781–2792.
MacKenzie KL, Hackett NR, Crystal RG et al. Adenoviral vector-mediated gene transfer to primitive human hematopoietic progenitor cells: assessment of transduction and toxicity in long-term culture. Blood 2000;96:100–108.
Santiago-Schwarz F, Divaris N, Kay C et al. Mechanisms of tumor necrosis factor-granulocyte-macrophage colony-stimulating factor-induced dendritic cell development. Blood 1993;82:3019–3028.
Shisler J, Yang C, Walter B et al. The adenovirus E3-10.4K/14.5K complex mediates loss of cell surface Fas (CD95) and resistance to Fas-induced apoptosis. J Virol 1997;71:8299–8306.
Gooding LR, Aquino L, Duerksen-Hughes PJ et al. The E1B 19,000-molecular-weight protein of group C adenoviruses prevents tumor necrosis factor cytolysis of human cells but not of mouse cells. J Virol 1991;65:3083–3094.
Mittereder N, March KL, Trapnell BC. Evaluation of the concentration and bioactivity of adenovirus vectors for gene therapy. J Virol 1996;70:7498–7509.
Leopold PL, Ferris B, Grinberg I et al. Fluorescent virions: dynamic tracking of the pathway of adenoviral gene transfer vectors in living cells. Hum Gene Ther 1998;9:367–378.
Moldenhauer A, Nociari MM, Dias S et al. Optimized culture conditions for the generation of dendritic cells from peripheral blood monocytes. Vox Sang 2003;84:228–236.
Hsu KH, Lonberg-Holm K, Alstein B et al. A monoclonal antibody specific for the cellular receptor for the group B coxsackieviruses. J Virol 1988;62:1647–1652.
Vincent T, Harvey BG, Hogan SM et al. Rapid assessment of adenovirus serum neutralizing antibody titer based on quantitative, morphometric evaluation of capsid binding and intracellular trafficking: population analysis of adenovirus capsid association with cells is predictive of adenovirus infectivity. J Virol 2001;75:1516–1521.
Ferrajoli A, Talpaz M, Kurzrock R et al. Analysis of the effects of tumor necrosis factor inhibitors on human hematopoiesis. STEM CELLS 1993;11:112–119.
Rusten LS, Jacobsen SE. Tumor necrosis factor (TNF)-alpha directly inhibits human erythropoiesis in vitro: role of p55 and p75 TNF receptors. Blood 1995;85:989–996.
Jacobsen SE, Jacobsen FW, Fahlman C et al. TNF-alpha, the great imitator: role of p55 and p75 TNF receptors in hematopoiesis. STEM CELLS 1994;12(suppl 1):111–128.
Huang S, Endo RI, Nemerow GR. Upregulation of integrins alpha v beta 3 and alpha v beta 5 on human monocytes and T lymphocytes facilitates adenovirus-mediated gene delivery. J Virol 1995;69:2257–2263.
Kitazono M, Rao VK, Robey R et al. Histone deacetylase inhibitor FR901228 enhances adenovirus infection of hematopoietic cells. Blood 2002;99:2248–2251.
Shayakhmetov DM, Papayannopoulou T, Stamatoyannopoulos G et al. Efficient gene transfer into human CD34(+) cells by a retargeted adenovirus vector. J Virol 2000;74:2567–2583.
Rebel VI, Hartnett S, Denham J et al. Maturation and lineage-specific expression of the coxsackie and adenovirus receptor in hematopoietic cells. STEM CELLS 2000;18:176–182.
Magnusson MK, Hong SS, Boulanger P et al. Genetic retargeting of adenovirus: novel strategy employing "deknobbing" of the fiber. J Virol 2001;75:7280–7289.
Serrander L, Skarman P, Rasmussen B et al. Selective inhibition of IgG-mediated phagocytosis in gelsolin-deficient murine neutrophils. J Immunol 2000;165:2451–2457.
Li E, Brown SL, Von Seggern DJ et al. Signaling antibodies complexed with adenovirus circumvent CAR and integrin interactions and improve gene delivery. Gene Ther 2000;7:1593–1599.
Rusten LS, Jacobsen FW, Lesslauer W et al. Bifunctional effects of tumor necrosis factor alpha (TNF alpha) on the growth of mature and primitive human hematopoietic progenitor cells: involvement of p55 and p75 TNF receptors. Blood 1994;83:3152–3159.
Broxmeyer HE, Williams DE, Lu L et al. The suppressive influences of human tumor necrosis factors on bone marrow hematopoietic progenitor cells from normal donors and patients with leukemia: synergism of tumor necrosis factor and interferon-gamma. J Immunol 1986;136:4487–4495.
Akahane K, Hosoi T, Urabe A et al. Effects of recombinant human tumor necrosis factor (rhTNF) on normal human and mouse hemopoietic progenitor cells. Int J Cell Cloning 1987;5:16–26.
Murase T, Hotta T, Saito H et al. Effect of recombinant human tumor necrosis factor on the colony growth of human leukemia progenitor cells and normal hematopoietic progenitor cells. Blood 1987;69:467–472.
Caux C, Saeland S, Favre C et al. Tumor necrosis factor-alpha strongly potentiates interleukin-3 and granulocyte-macrophage colony-stimulating factor-induced proliferation of human CD34+ hematopoietic progenitor cells. Blood 1990;75:2292–2298.
K?rbling M, Anderlini P. Peripheral blood stem cell versus bone marrow allotransplantation: does the source of hematopoietic stem cells matter? Blood 2001;98:2900–2908.
Mulloy JC, Cammenga J, MacKenzie KL et al. The AML1-ETO fusion protein promotes the expansion of human hematopoietic stem cells. Blood 2002;99:15–23.
Nathwani AC, Hanawa H, Vandergriff J et al. Efficient gene transfer into human cord blood CD34+ cells and the CD34+CD38– subset using highly purified recombinant adeno-associated viral vector preparations that are free of helper virus and wild-type AAV. Gene Ther 2000;7:183–195.(A. Moldenhauera, J-H Shie)