- Email: [email protected]
Optimisation of herpes simplex virus-based vectors for delivery to human peripheral blood mononuclear cells
Journal of Immunological Methods 270 (2002) 235 – 246 www.elsevier.com/locate/jim
Recombinant Technology
Optimisation of herpes simplex virus-based vectors for delivery to human peripheral blood mononuclear cells Konstantina Papageorgiou a, David A. Isenberg b, David S. Latchman a,* a
Institute of Child Health, University College London, 30 Guilford Street, London, WC1N 1EH, UK b Centre for Rheumatology, Department of Medicine, University College London, London, UK Received 3 February 2002; received in revised form 2 May 2002; accepted 16 July 2002
Abstract Peripheral blood mononuclear cells (PBMCs) represent a significant target for gene delivery both for therapeutic and experimental purposes. Thus far however, it has proved difficult to develop vectors capable of high efficient gene delivery to unstimulated PBMCs. We have tested a range of different vectors derived from herpes simplex virus (HSV) which differ in their degree of disablement in terms of their gene delivery efficiency to unstimulated human PBMCs and ability to deliver a reporter gene. None of the viruses had any significant toxic effect in PBMCs. However, optimal gene delivery to unstimulated PBMCs was obtained with a semidisabled virus lacking functional genes encoding ICP34.5 and Vmw65 which was more efficient than either nondisabled or more extremely disabled viruses. Expression of green fluorescent protein (GFP) with this virus was observed in up to 50% of PBMCs 1 day after infection, and reporter gene expression was detectable by Western blotting and immunofluorescence at undiminished levels at the longest time points tested, up to 5 days after infection. This optimised HSV vector may thus represent an effective tool for gene delivery to unstimulated PBMCs in culture. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Peripheral blood mononuclear cells; Virus vector; Gene delivery; Herpes simplex virus
1. Introduction A very large number of laboratories have attempted to develop viral or nonviral vectors for the delivery of genes to human hematopoietic stem cells in the hope of manipulating such cells ex vivo and then returning them to patients suffering from a wide range of immunological and hematological diseases (for recent reviews see Heim and Dunbar, 2000; Williams and * Corresponding author. Tel.: +44-20-7-905-2189; fax: +44-207-242-8437. E-mail address: [email protected] (D.S. Latchman).
Smith, 2000). However, much less attention has been paid to the development of methods for delivering genes to human peripheral blood mononuclear cells obtained directly from adult individuals. Nonetheless, such efficient gene delivery to these cells could be of both therapeutic and scientific importance. Thus, for example, such transduction could be used to deliver the gene encoding the defective enzyme to the peripheral blood cells of patients with mucopolysaccharide diseases (see, for example, Pan et al., 1999; Stroncek et al., 1999). Similarly, in our previous work, we have demonstrated that cytokine IL-6 induces enhanced expression of the 90 kDa heat shock protein (HSP90)
0022-1759/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 2 2 - 1 7 5 9 ( 0 2 ) 0 0 2 9 9 - 5
236
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
in peripheral blood mononuclear cells (Stephanou et al., 1997, 1998). As IL-6 activates both the NF-IL-6/ NF-IL-6h transcription factors and the unrelated STAT-3 transcription factor (Nakajima et al., 1993; Wegenka et al., 1993), it is necessary to introduce dominant negative forms of each of these factors into peripheral blood cells in order to dissect the mechanisms by which IL-6 stimulates the HSP90 promoter. For these reasons, some previous studies have investigated the means of delivering genes to peripheral blood mononuclear cells. Thus, it has been reported that such gene delivery can be achieved either by using naked DNA (Jensen et al., 2000), retroviral viral vectors based on Moloney murine leukemia virus (Dardalhon et al., 2000), lentiviral vectors (Li et al., 2000; Chinnasamy et al., 2000), or adeno-associated viral vectors (Dunbar, 1999). Nonetheless, each of these approaches has considerable disadvantages. Thus, since retroviral vectors based on Moloney murine leukemia virus will only infect dividing cells, it is necessary to stimulate the peripheral blood cells in some way in order to obtain efficient delivery (Dardalhon et al., 2000). Similarly, lentiviral vectors do not appear to achieve high efficiency gene delivery unless HIV-1 accessory proteins are also provided (Chinnasamy et al., 2000). Clearly, this would not be appropriate in a clinical situation and would unnecessarily complicate experimental procedures. Moreover, in a comparison between a lentiviral vector and an adeno-associated virus vector, it was concluded that while the adenoassociated virus vector gave a higher initial efficiency of gene delivery, this fell off rapidly with the time compared to the retroviral vector (Dunbar, 1999). Interestingly, efficient gene delivery to human T cells has been reported with a vector based on the herpes virus, Herpes Saimiri (Hiller et al., 2000). However, since this virus transforms T cells to a continuously proliferating phenotype, it is evidently not suitable for clinical use in humans or for experimental procedures where it is necessary to maintain the normal phenotype of the target cell. Our laboratory has extensively investigated the use of vectors based on another member of the herpes virus family, herpes simplex virus type 1 (HSV-1), for gene delivery to different cell types (for review, see Latchman, 2000, 2001). Although this virus naturally infects neuronal cells establishing life-long asympto-
matic latent infections, it can infect a variety of different cell types. Thus, for example, we have shown that the virus can deliver reporter genes effectively to cardiac cells both in culture and in the intact heart (Coffin et al., 1996a), and can also deliver to human dendritic cells and hematopoietic cells with high efficiency (Coffin et al., 1998). Moreover, this virus has a number of advantages as a gene delivery vector, notably the ability to be grown efficiently in culture, and since it is a very large DNA virus with a genome of over 150 kilobases, the ability to accept several different inserts encoding different genes which may need to be delivered in parallel. A key issue in the development of vectors derived from HSV, both for gene therapy and for testing gene functions, is the need to disable the virus so that it cannot carry out a lytic infection while retaining the ability to prepare viral stocks in culture and deliver genes effectively to a specific cell type. In our previous studies, we have demonstrated that the degree of disablement required for optimal gene delivery differs depending on the cell type being targeted. Thus, for example, in the initial stages of vector preparation, we constructed a virus which lacks the viral gene encoding the ICP34.5 protein and is therefore nonneurovirulent upon injection into the brain being unable to replicate in a variety of nondividing cells including neurones (Coffin et al., 1996b). This virus continues to replicate on dividing cells in culture, thereby allowing stocks of the vector to be prepared. Moreover, this virus produces optimal gene delivery to the peripheral nervous system following injection into the mouse footpad (Palmer et al., 2000). However, following direct injection into brain, this virus produces only low efficiency gene delivery (Coffin et al., 1996b). For higher efficiency gene delivery to the central nervous system, it is necessary to remove the viral gene encoding the immediate early protein ICP27 (Howard et al., 1998). This results in a virus that is unable to replicate lytically in any cell type, and must be grown in culture on cells artificially expressing ICP27. Hence, an entirely replication-deficient virus is required for the efficient gene delivery to the central nervous system, whereas a partially incompetent virus produces optimal delivery to the peripheral nervous system. Similarly, virus lacking ICP34.5 and ICP27 can be further disabled,
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
for example, by the removal of the gene encoding another immediate early protein ICP4. Although this protein must similarly be provided within cultured cells for replication to occur, infection of cells with this highly disabled virus lacking ICP34.5, ICP4, and ICP27 results in the production of essentially noviral protein and represents an optimally disabled virus for gene therapy applications in the central nervous system (Thomas et al., 1999b; Lilley et al., 2001). In this report, we have therefore compared viral vectors lacking ICP34.5, and ICP27 or ICP34.5, ICP27, and ICP4 in terms of their ability to deliver genes in a nontoxic and efficient manner to human peripheral blood mononuclear cells in order to determine whether HSV can be used as a vector for these cells and the optimal level of disablement required for a safe and efficient vector.
237
2.3. Infection Cells of 106 were taken and spun down in 15-ml falcon tubes. Most of the media was removed by pipetting and virus at Multiplicities of Infection (MOI) of 10 was added. The tubes were flicked and left at 37 jC, 5% CO2 for 1 h. The cells were then resuspended in 1 ml of full grown RPMI (10% Foetal Calf Serum, 1% Penicillin Streptomycin) and transferred to one well of a 24-well plate. The cells were then left overnight and assessed on the three following days. 2.4. Infection assessment 2.4.1. GFP expression The cells were observed under an inverted fluorescent microscope and scanned for cells expressing GFP. Total gene delivery rate was calculated as the percentage of green cells from the whole population.
2. Methods and materials
Blood of 20– 50 ml was taken from normal volunteers and centrifuged at 1500 rpm (450 g), 4 jC for 10 min. The serum was then removed and stored for further tests. The cells were then diluted 1:3 with RPMI serum-free media (GIBCO BRL, Scotland) and layered over 15 ml of lymphoprep (Pharmacia, UK) in a 50-ml falcon tube. The samples were then spun at 1500 rpm (450 g) for 45 min, at 4 jC with zero brake rate. The white blood cell layer was then collected with a sterile plastic Pasteur pipette and diluted 1:5 with RPMI serum-free media. The cells were spun at 1500 rpm (450 g) for 10 min, at 4 jC. The pellet was then resuspended in 5 ml of serum-free RPMI. The cell concentration was measured by counting with Trypan blue.
2.4.2. X-gal staining A sample of 100 Al was taken from a resuspended culture of infected PBMCs. These were then centrifuged in a 15-ml falcon tube at 1500 rpm (450 g) at 4 jC for 10 min. The supernatant was removed and the cells were fixed in 100 Al of 1 PBS/0.1% glutaraldehyde. The cells were incubated at room temperature for 10 min and spun again at 1500 rpm at 4 jC for 10 min. The fix was removed and the cells resuspended in 200 Al of X-gal solution (150 Ag/ml 4Chloro, 5-bromo, 3-indolyl-b-galactoside in DMSO, 1 PBS, 10 mM sodium phosphate, 1 mM MgCl2, 3.3 mM K4Fe(CN)6, 3.3 mM K3Fe(CN)6). The cells were plated out in a 24-well plate, and incubated overnight at 37 jC. The cells were then observed under the light microscope and the percentage of cells expressing the lacZ reporter gene (blue cells) was determined.
2.2. Cell counting
2.5. Cell culture
When cells are resuspended in Trypan blue, dead cells do not exclude the dye and therefore turn blue. From a homogeneous mix of RPMI and PBMCs, 10 Al were taken and added to 10 Al of a 0.04% Trypan blue solution. The later mix of 10 Al was loaded on a hemocytometer counting chamber.
BHK cells were maintained using standard tissue culture procedures. Virus strain 1764 was propagated on BHK cells, while virus strains 17 + 27- and 176427-4- were grown on a BHK-derived cell line described previously which expresses ICP4 and ICP27 (Thomas et al., 1999b; Lilley et al., 2001)
2.1. Isolation of PBMCs
238
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
(27/12/M:4 cell line). This cell line complements for the genes that have been deleted and there is no sequence overlap between the viral genes present in the cell line and the deleted viral genome. Complementing cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% Fetal Calf Serum, 1% Penicillin Streptomycin and in the case of 27/12/M:4 with 800 Ag/ml of G418 and 750 Ag/ml of Zeocin. Hexamethylene bisacetamide (HMBA) was included in the growth media. 2.6. Viral culture Ten 175 cm2 flasks of 100% confluent cells were split by using a versene/10% Trypsin (GIBCO) solution and resuspended in DMEM FGM. The cells were then transferred to ten 850 cm2 roller bottles (100 ml/bottle), and incubated at 37 jC, 5% CO2 for 2 days, or until 70 –80% confluent. The cells were then infected with 106 pfu of virus/bottle, and grown in fresh FGM for another 2 –3 days at 32 jC, or at least until they are ready to harvest (media becomes orange and cells become detached from vessel surfaces). The cells were harvested and frozen at 80 jC. The cells were then thawed and spun at 3500 rpm (2500 g) at 4 jC for 30 min to remove the cell debris. The supernatant was then passed through a 0.45-Am filter and spun again at 12,000 rpm (28,000 g) at 4 jC for 2 h in a JA12 rotor. The pellet was resuspended in its own medium or a small volume (approximately 200 Al) of serum-free DMEM, and sonicated to homogeneity. The virus stock was then titred and stored at 70 jC. 2.7. Viral titre Appropriate cells for lytic growth of each virus were grown on a 6-well plate, at 37 jC, 5% CO2, until 70– 80% confluent. Successive 1:10 dilutions of the virus starting from 10 2 to 10 10 were prepared in 500 Al of Serum Free DMEM. These were added to the cells. The cells were then incubated at 37 jC, 5% CO2, for 1 h. After the incubation, 2 ml of 1:2 of 1.6% carboxymethyl cellulose full grown medium DMEM was added to each well. For noncomplementing cell lines (BHK), 3 mM hexamethylene bisacetamide (HMBA) were added to the media.
2.8. Western blotting The cell pellets were freeze-thawed five times in liquid nitrogen. The samples were then resuspended with 20 Al of 0.1 M Tris (pH 8). They were then spun at 14,000 rpm (13,000 g) at 4 jC for 30 min. The supernatant was collected and used on a 10% acrylamide gel. Alternatively, the cells were directly lysed in loading buffer (2.3% sodium dodecyl sulphate (SDS), 0.0625 M Tris, 10% glycerol, 5% h-mercaptoethanol and bromophenol blue) at a 1:1 ratio. The samples were subjected to SDS-polyacrylamide gel electrophoresis (PAGE). The resolving gel consisted of 10% acrylamide (in 1.5 M Tris, 0.4% SDS, pH 8.8), and the stacking/loading gel consisted of 5% acrylamide (in 0.5 M Tris, 0.4% SDS, pH 6.8). The samples were denatured at 95 jC for 5 min in loading buffer, and then immediately placed on ice and loaded on the gel. The gels were run in running buffer (0.129 M glycine, 0.025 M Tris, 0.1% SDS, pH 8.3) at 30 mA, for approximately 4 h. They were then transferred onto a Hybond-C membrane (Amersham, UK), in blotting buffer (0.192 M glycine, 0.025 M Tris, 20% methanol pH 8) at 200 mA overnight at 4 jC. The membrane was then blocked in 4% Marvel/0.1% PBS Tween for 2 h, on a shaker. Then, appropriate antibody was added at 1:1000 dilution in 4% Marvel/0.1% PBS Tween, and placed on a shaker for 1 h or more to intensify the signal. The membrane was then briefly washed twice in 0.1% PBS Tween, and incubated on the shaker for another 1 h with the secondary antibody conjugated with horseradish peroxidase (HRP) (1:1000 dilution in 4% Marvel/0.1% PBS Tween). The membrane was passed through a series of 5 5-min washes in 0.1% PBS Tween and then incubated in an enhanced chemiluminescent (ECL) (Amersham) solution for approximately 1 min. The membrane was then exposed to X-OMAT film from Kodak, for 1 min. Further washes or exposure periods took place if necessary. 2.9. Virus nomenclature All viruses were derived from HSV-1 strain 17syn+ (Brown et al., 1973). Viruses 2, 5, and 6 have the 1814 mutation in the gene encoding VP16 (VMW65) and the genes ICP34.5 and ORF P completely deleted
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
(between nt. 124945 and 125723) (Coffin et al., 1996a,b; Ace et al., 1989). Viruses 3 and 4 also have a deletion of nt. 113273 to 116869 which contain the genes UL54, 55 and 56 (Howard et al., 1998). UL54 encodes for the essential IE protein ICP27, while UL55 and 56 are nonessential genes. Viruses 3 and 4 were also deleted for the endogenous LATP2 regions (nt. 118768 and 120470) in order to prevent recombination instability that was found to occur after insertion of LATP2-containing expression cassettes outside of the LAT region (Palmer et al., 2000).
3. Results and discussion To assess the ability of HSV to deliver a reporter gene to peripheral blood mononuclear cells (PBMCs) and the optimal degree of viral disablement, we utilised four viruses differing in their degree of disablement but each having the Cytomegalovirus (CMV) immediate early gene promoter driving the expression of a green fluorescent protein (GFP) reporter gene (see Fig. 1 and Palmer et al., 2000; Thomas et al., 1999b; Lilley et al., 2001, for further details of these viruses). Virus 1 contains the CMV-GFP construct inserted into the nonessential UL43 gene which has no effect on the functioning of the virus. It is therefore an essentially wild type virus. Virus 2 contains the ICP34.5 mutation as well as a mutation in the gene encoding Vmw65, which transactivates the viral immediate early genes with the CMV-GFP gene inserted into the nonessential LAT P2 site. It is, however, still able to replicate on dividing cells but not on nondividing cells, and is therefore a partially disabled virus. Virus 3 contains, as well as the ICP34.5 mutation, a mutation in the gene encoding ICP27 which renders it nonreplication competent on any cell type. Finally, the most disabled virus, virus 4, also contains mutations in both copies of the genes encoding ICP4 providing the maximal level of disablement by inactivating two essential immediate early genes. In initial experiments, we infected PBMCs with each of these viruses at a multiplicity of infection (MOI) of 10 plaque forming units per cell. The effect on the cell viability of such viral infection was measured on the basis of the ability of live cells to
239
exclude Trypan blue. As illustrated in Fig. 2, no significant decrease was observed in any of the viral infected cells at 1, 2, or 3 days after infection. A small decrease was observed with virus 3 but was not significant. Most interestingly, infection, even with wild type virus (virus 1), demonstrated no significant decrease in viability. This is in accordance with previous findings suggesting that B cells, T cells, and monocytes do not support productive infection with HSV (Tenney and Morahan, 1987; Albers et al., 1989; Kemp et al., 1990). Having established that none of the viruses caused significant death of PBMCs, we investigated their ability to deliver the GFP reporter gene to PBMCs by measuring the number of cells in the culture which were positive for GFP fluorescence at intervals after infection. The results of these experiments (Fig. 3) revealed significant differences between the various viruses in their efficiency of gene delivery. Thus, viruses 3 and 4, which contained mutations in the immediate early gene encoding ICP27, either alone (virus 3) or in combination with a mutation in ICP4 (virus 4), showed relatively low efficiencies of gene delivery with 10% of the PBMCs showing GFP fluorescence even at 1 day after infection. Similarly, the essentially wild type virus (virus 1) showed a very low efficiency of gene delivery. The greatest efficiency of gene delivery with over 40% of the cells showing GFP fluorescence at day 1 after infection was achieved with the semidisabled virus (virus 2), in which the genes encoding ICP34.5 and the virion transactivator Vmw65 had been inactivated but still contained functional copies of all the genes required for replication in all cell types. Hence, a partially disabled virus lacking functional ICP34.5 or Vmw65 appears to provide efficient gene delivery to PBMCs with nearly half of the cells expressing the reporter gene at 1 day after infection. The high efficiency of this virus compared to either less or more disabled viruses is likely to reflect the balance between inhibitory effects on nonviral transgene expression, which may occur in nondisabled viruses and the impaired efficiency of gene delivery with highly disabled viruses. This balance is different in different cell types such as neurones of the central versus peripheral nervous systems (Coffin et al., 1996b; Lilley et al., 2001), necessitating the need to determine the optimal vector for each cell type.
240
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
241
Fig. 2. Viability of PBMCs infected with the viruses shown in Fig. 1 as determined by the ability of live cells to exclude Trypan blue at intervals after infection. Values indicate the mean of at least nine experiments with each virus whose standard deviation is shown by the bars.
To confirm that this partially disabled virus will infect all cell types in PBMC, PBMCs were infected with virus. FACS was then used to separate monocytes from lymphocytes on the basis of cell size, and B cells were distinguished from T lymphocytes on the basis of their reactivity with antibodies to CD19 and CD3. Successful infection (as assayed by GFP positivity) was observed in all cell types with the greatest efficiency being observed in monocytes. Hence, our virus can successfully deliver a reporter gene to T lymphocytes, B lymphocytes, and monocytes. Although virus 2 produced high efficiency gene delivery as indicated in Fig. 4, the number of cells showing GFP fluorescence declines with time, reducing to approximately 20% of the cells after 2 days of infection and 10% of cells after 3 days. As we did not observe significant cell death following infection with
virus 2 (see Fig. 2), this loss of GFP positivity does not appear to be due to the death of infected cells during the culture period. To investigate whether the decline in GFP positivity was due to silencing of the CMV promoter which drives GFP in all the viruses tested so far, we used two further viruses (see Fig. 1) which both contained mutations in the genes encoding ICP34.5 and Vmw65 as in virus 3, but in which the GFP gene was driven either by the murine sarcoma virus promoter (virus 5) or the Moloney murine leukemia virus promoter (virus 6). As expected from our previous experiments, both these viruses have no significant effect on the viability of infected PBMCs (data not shown). When the percentage of GFP positive cells was determined at intervals following infection with these viruses, virus 5 produced a similar number of
Fig. 1. Details of the viruses used in this study. An indicates a gene which has been functionally inactivated or deleted. The site of insertion of each virus is shown. Note that deletion of the LATP2 element in the viruses does not affect their replication ability but is essential to prevent recombination in stability due to the presence of elements (LAP1 and LATP2) designed to give longer term expression in the inserted construct (Thomas et al., 1999a). For further details of these viruses, see Coffin et al. (1996a,b), Howard et al. (1998), Lilley et al. (2001), and Palmer et al. (2000).
242
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
Fig. 3. Percentage of GFP positive PBMCs at daily intervals following infection with each of the viruses 1 – 4 shown in Fig. 1. Values indicate the mean of at least nine experiments with each virus whose standard deviation is shown by the bars.
GFP positive cells at day 1 after infection to that observed with virus 3 (Fig. 4). In contrast, the percentage of GFP positive cells observed with virus 6 was somewhat lower, suggesting the MMLV promoter is somewhat weaker than the MSV or CMV promoters in PBMCs. However, in both cases, a decline in the number of cells showing GFP fluorescence was observed at days 2 and 3, indicating that changing the viral promoter used to drive GFP expression does not result in longer persistence of GFP fluorescence in the infected cells. To investigate the time course of GFP expression further, we utilised Western blotting with an antibody to GFP. Cells were infected with virus 5 and harvested at various times after infection. As illustrated in Fig. 5, GFP expression in the infected cells was not observed after 4 h of infection but was observed at both the 8and 24-h time points. Since the virus enters the infected cell within 1 h after exposure, this indicates that the GFP being detected in the infected cells is not the result of the pseudotransfer of GFP made in the cells on which the virus was previously grown and
then packaged into the viral particle. Rather, it represents new GFP protein derived from the GFP gene in the virus and synthesised in the PBMCs providing further evidence that our vectors can express a reporter gene in the cells. To investigate the persistence of GFP expression, we also carried out Western blotting on extracts prepared from cells infected with virus 5 at 1, 2, and 3 days after infection. Surprisingly in this experiment (Fig. 6), we detected high levels of GFP at all time points after infection with no noticeable decline in GFP expression over the 3-day period despite the decline in the number of cells showing GFP fluorescence (compare Figs. 4 and 6). This suggests that GFP is either stable in the infected cell or continues to be made so that protein levels remain similar, but that GFP fluorescence is quenched in the PBMCs over the 3-day time period so that considerably less cells appear positive for GFP even though GFP protein is still present. To confirm this, we stained PBMCs infected with virus 5 using an antibody to GFP and compared the
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
243
Fig. 4. Percentage of PBMCs which were GFP positive at daily intervals following infection with viruses 5 and 6 illustrated in Fig. 1. Values indicate the mean of five experiments with each virus whose standard deviation is shown by the bars.
percentage of cells staining with that observed by GFP fluorescence. As illustrated in Fig. 7, the percentage of cells which reacted with the anti-GFP antibody was higher at every time point tested than the number of cells showing GFP fluorescence. Moreover, there was very little decrease in the number of antibody positive cells with time for at least 5 days after infection.
Hence, GFP fluorescence underestimates the number of infected cells in PBMC culture. In addition, it is clear that GFP expression is relatively stable with time in infected PBMCs as assayed by immunofluorescence and Western blotting, and that the decline in GFP fluorescence is due to quenching of the signal with time.
Fig. 5. Western blotting with an antibody to GFP using cells prepared at the indicated times after infection with virus 5 or from uninfected cells. The results of Western blotting for GFP and for the control actin protein are illustrated.
244
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
Fig. 6. Western blotting of cells prepared at the indicated periods after infection with virus 5 and probed with antibody to h-galactosidase, GFP or actin.
Interestingly, since virus 5 contains a h-galactosidase gene under the control of the Rous sarcoma virus promoter (Fig. 1), we were also able to measure the
levels of a second reporter gene in these experiments. As illustrated in Fig. 6, Western blotting with an antibody to h-galactosidase similarly revealed high
Fig. 7. Percentage of PBMCs showing either GFP fluorescence or positive staining with antibody to GFP at the indicated time after infection with virus 5.
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
levels of this protein being made in the infected cells, and showed that it too was clearly detectable at high level at 3 days after infection. Hence, two reporter proteins persist in the infected PBMCs for at least 3– 5 days following infection with the optimum HSV vector. Unfortunately, we were unable to determine the percentage of h-galactosidase positive cells in a culture of PBMCs using the well-known X-Gal stain for h-galactosidase activity. Although this method readily allowed us to detect h-galactosidase positive cells in cultures of neuronal and other cell types infected with our vectors, it did not result in any positive staining cells in the PBMC cultures, suggesting that h-galactosidase activity is not detectable by this means in infected PBMCs even though the protein is detectable by Western blotting (data not shown). These experiments therefore indicate that two distinct reporter genes can be expressed in PBMCs for at least 3 days in culture using an HSV vector which has been disabled by inactivation of the genes encoding ICP34.5 and Vmw65. Moreover, this can be achieved without significantly affecting the viability of the cells which appear to be resistant to cell death even following infection with an essentially wild type virus. These findings therefore indicate that the optimum HSV vector which we have identified can be used to express specific genes in PBMCs in order to assess their effect in short-term assays. Moreover, by investigating the time course of expression of a particular delivered gene using both Western blotting and immunofluorescence assays, it may be possible to carry out longer term assays in PBMCs where protein production/persistence is observed for longer periods following infection. Similarly, although further experiments will be required to test whether our optimal vector affects the functioning of PBMCs in general or of individual cell types, in the population, it is already clear that the cells survive infection even with wild type virus, with no significant loss of viability being observed. Hence, HSV represents a useful addition to the limited repertoire of virus vectors capable of delivering genes to cultured human PBMCs. This will allow, for example, the delivery of genes encoding dominant negative forms of specific transcription factors in order to elucidate the mechanisms involved in the induction of HSP90 by IL-6 which occurs in all PBMC cell types (Dhillon et al., 1993; Stephanou et
245
al., 1997, 1998). Similarly, it may ultimately be possible to specifically deliver genes to one cell type within PBMCs, following further optimisation of our vectors for a specific PBMC cell type.
Acknowledgements K.P. is supported by a Research Studentship from the Arthritis Research Campaign.
References Ace, C.I., McKee, T.A., Ryan, M., Cameron, J.M., Preston, C.M., 1989. Construction and characterization of a herpes simplex virus type I mutant unable to transduce immediate-early gene expression. Journal of Virology 63, 2260. Albers, I., Kirchner, H., Domke-Opitz, I., 1989. Resistance of human blood monocytes to infection with herpes simplex virus. Virology 169, 466. Brown, S.M., Ritchie, D.A., Subak-Sharpe, J.H., 1973. Genetic studies with herpes simplex virus type-1. Journal of General Virology 18, 329. Chinnasamy, D., Chinnasamy, N., Enriquez, M.J., Otsu, M., Morgan, R.A., Candotti, F., 2000. Lentiviral-mediated gene transfer into human lymphocytes: role of HIV-1 accessory proteins. Blood 96, 1309. Coffin, R.S., Howard, M.K., Cumming, D.V.E., Dollery, C.M., McEwan, J., Yellon, D.M., Marber, M.S., Maclean, A.R., Brown, S.M., Latchman, D.S., 1996a. Gene delivery to cardiac cells in vitro and in vivo using herpes simplex virus vectors. Gene Therapy 3, 560. Coffin, R.S., Maclean, A.R., Latchman, D.S., Brown, S.M., 1996b. Gene delivery to the central and peripheral nervous systems of mice using HSV1 ICP34.5 deletion mutant vectors. Gene Therapy 3, 886. Coffin, R.S., Thomas, S.K., Thomas, N.S.B., Lilley, C.E., Pizzey, A.R., Griffiths, C.H., Gibb, B.J., Wagstaff, M.J.D., Inges, S.J., Binks, M.H., Chain, B.M., Thrasher, A.J., Rutault, K., Latchman, D.S., 1998. Pure populations of transduced primary human cells can be produced using GFP expressing herpes virus vectors and flow cytometry. Gene Therapy 5, 718. Dardalhon, V., Jaleco, S., Rebouissou, C., Ferrand, C., Skander, N., Swainson, L., Tiberghien, P., Spits, H., Noraz, N., Taylor, N., 2000. Highly efficient gene transfer in naive human T cells with a murine leukemia virus-based vector. Blood 96, 885. Dhillon, V.B., McCallum, S., Norton, P.M., Twomey, B.M., Erkeller-Yuksel, F., Lydyard, P., Isenberg, D.A., Latchman, D.S., 1993. Differential heat shock protein over-expression and its clinical relevance in systemic lupus erythematosus. Annals of the Rheumatic Diseases 52, 436. Dunbar, C.E., 1999. In vivo marking of rhesus monkey lymphocytes by adeno-associated viral vectors: direct comparison with retroviral vectors. Blood 94, 2263.
246
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
Heim, D.A., Dunbar, C.E., 2000. Hematopoietic stem cell gene therapy: towards clinically significant gene transfer efficiency. Immunological Reviews 178, 29. Hiller, C., Wittmann, S., Slavin, S., Fickenscher, H., 2000. Functional long-term thymidine kinase suicide gene expression in human T cells using a herpesvirus saimiri vector. Gene Therapy 7, 664. Howard, M.K., Kershaw, T., Gibb, B., Storey, W., Maclean, A.R., Zeng, B.-W., Jenner, P., Brown, S.M., Woolf, C.J., Anderson, P.N., Coffin, R.S., Latchman, D.S., 1998. High efficiency gene transfer to the central nervous system of rodents and primates using herpes virus vectors lacking functional ICP27 and ICP34.5. Gene Therapy 5, 1137. Jensen, M.C., Clarke, P., Tan, G., Wright, C., Chung-Chang, W., Clark, T.N., Zhang, F., Slovak, M.L., Wu, A.M., Forman, S.J., Raubitschek, A., 2000. Human T lymphocyte genetic modification with naked DNA. Molecular Therapy 1, 49. Kemp, L.M., Estridge, J.K., Brennan, A., Katz, D., Latchman, D.S., 1990. Mononuclear phagocytes and HSV-1 infection: increased permissivity in differentiated U937 cells is mediated by posttranscriptional regulation of viral immediate early gene expression. Journal of Leukocyte Biology 47, 483. Latchman, D.S., 2000. Gene therapy using herpes simplex virus based vectors. Histology and Histopathology 15, 1253. Latchman, D.S., 2001. Gene delivery and gene therapy with herpes simplex-virus based vectors. Gene 264, 1. Li, Y., Starr, S.E., Lisziewicz, J., Ho, W.Z., 2000. Inhibition of HIV1 replication in chronically infected cell lines and peripheral blood mononuclear cells by retrovirus-mediated antitat gene transfer. Gene Therapy 7, 321. Lilley, C.E., Groutsi, F., Han, Z.Q., Palmer, J.A., Anderson, P.N., Latchman, D.S., Coffin, R.S., 2001. Multiply disabled herpes simplex virus vectors optimised for the widespread and long term delivery of genes to the brain and spinal cord. Journal of Virology 75, 4343. Nakajima, T., Kinoshita, S., Sasagawa, T., Sasaki, K., Naruto, M., Kishimoto, T., Akira, S., 1993. Phosphorylation at threonine235 by a ras-dependent mitogen activated protein kinase cascade is essential for transcription factor NF-IL6. Proceedings of the National Academy of Sciences of the United States of America 90, 2207. Palmer, J., Branston, R., Lilley, C.E., Robinson, M., Groutsi, F., Smith, J., Latchman, D.S., Coffin, R.S., 2000. Development and optimisation of herpes simplex virus vectors for multiple long
term gene delivery to the peripheral nervous system. Journal of Virology 74, 5604. Pan, D., Shankar, R., Stroncek, D.F., Whitley, C.B., 1999. Combined ultrafiltration – transduction in a hollow-fibre bioreactor facilitates retrovirus-mediated gene transfer into peripheral blood lymphocytes from patients with mucopolysaccharidosis type II. Human Gene Therapy 10, 2799. Stephanou, A.S., Amin, V., Isenberg, D.A., Akira, S., Kishimoto, T., Latchman, D.S., 1997. Interleukin-6 activates heat shock protein 90h gene expression. Biochemical Journal 321, 103. Stephanou, A.S., Isenberg, D.A., Akira, S., Kishimoto, T., Latchman, D.S., 1998. The NF-IL6 and STAT-3 signalling pathways cooperate to mediate the activation of the HSP90h gene by IL-6 but have opposite effects on its inducibility by heat shock. Biochemical Journal 330, 189. Stroncek, D.F., Hubel, A., Shankar, R.A., Burger, S.R., Pan, D., McCullough, J., Whitley, C.B., 1999. Retroviral transduction and expansion of peripheral blood lymphocytes for the treatment of mucopolysaccharidosis type II, Hunter’s syndrome. Transfusion 39, 343. Tenney, D.J., Morahan, P.S., 1987. Effects of differentiation of human macrophage-like U937 cells on intrinsic resistance to herpes simplex virus type 1. Journal of Immunology 139, 3076. Thomas, S.K., Gough, G., Latchman, D.S., Coffin, R.S., 1999a. The herpes simplex virus latency associated transcript encodes a protein which greatly enhances virus growth, can compensate for deficiencies in immediate-early gene expression and is likely to function during the reactivation from virus latency. Journal of Virology 73, 6618. Thomas, S.K., Lilley, C.E., Latchman, D.S., Coffin, R.S., 1999b. Equine herpes virus 1 gene 12 can substitute for Vmw65 in the growth of herpes simplex virus 1 (HSV1) allowing the generation of optimised cell lines for the propagation of vectors with multiple immediate-early gene defects. Journal of Virology 73, 7399. Wegenka, U.M., Buschmann, J., Lutticken, C., Heinrich, P.C., Horn, F., 1993. Acute-phase response factor, a nuclear factor binding to acute-phase response elements, is rapidly activated by interleukin-6 at the posttranslational level. Molecular and Cellular Biology 13, 276. Williams, D.A., Smith, F.O., 2000. Progress in the use of gene transfer methods to treat genetic blood diseases. Human Gene Therapy 11, 2059.
Recombinant Technology
Optimisation of herpes simplex virus-based vectors for delivery to human peripheral blood mononuclear cells Konstantina Papageorgiou a, David A. Isenberg b, David S. Latchman a,* a
Institute of Child Health, University College London, 30 Guilford Street, London, WC1N 1EH, UK b Centre for Rheumatology, Department of Medicine, University College London, London, UK Received 3 February 2002; received in revised form 2 May 2002; accepted 16 July 2002
Abstract Peripheral blood mononuclear cells (PBMCs) represent a significant target for gene delivery both for therapeutic and experimental purposes. Thus far however, it has proved difficult to develop vectors capable of high efficient gene delivery to unstimulated PBMCs. We have tested a range of different vectors derived from herpes simplex virus (HSV) which differ in their degree of disablement in terms of their gene delivery efficiency to unstimulated human PBMCs and ability to deliver a reporter gene. None of the viruses had any significant toxic effect in PBMCs. However, optimal gene delivery to unstimulated PBMCs was obtained with a semidisabled virus lacking functional genes encoding ICP34.5 and Vmw65 which was more efficient than either nondisabled or more extremely disabled viruses. Expression of green fluorescent protein (GFP) with this virus was observed in up to 50% of PBMCs 1 day after infection, and reporter gene expression was detectable by Western blotting and immunofluorescence at undiminished levels at the longest time points tested, up to 5 days after infection. This optimised HSV vector may thus represent an effective tool for gene delivery to unstimulated PBMCs in culture. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Peripheral blood mononuclear cells; Virus vector; Gene delivery; Herpes simplex virus
1. Introduction A very large number of laboratories have attempted to develop viral or nonviral vectors for the delivery of genes to human hematopoietic stem cells in the hope of manipulating such cells ex vivo and then returning them to patients suffering from a wide range of immunological and hematological diseases (for recent reviews see Heim and Dunbar, 2000; Williams and * Corresponding author. Tel.: +44-20-7-905-2189; fax: +44-207-242-8437. E-mail address: [email protected] (D.S. Latchman).
Smith, 2000). However, much less attention has been paid to the development of methods for delivering genes to human peripheral blood mononuclear cells obtained directly from adult individuals. Nonetheless, such efficient gene delivery to these cells could be of both therapeutic and scientific importance. Thus, for example, such transduction could be used to deliver the gene encoding the defective enzyme to the peripheral blood cells of patients with mucopolysaccharide diseases (see, for example, Pan et al., 1999; Stroncek et al., 1999). Similarly, in our previous work, we have demonstrated that cytokine IL-6 induces enhanced expression of the 90 kDa heat shock protein (HSP90)
0022-1759/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 2 2 - 1 7 5 9 ( 0 2 ) 0 0 2 9 9 - 5
236
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
in peripheral blood mononuclear cells (Stephanou et al., 1997, 1998). As IL-6 activates both the NF-IL-6/ NF-IL-6h transcription factors and the unrelated STAT-3 transcription factor (Nakajima et al., 1993; Wegenka et al., 1993), it is necessary to introduce dominant negative forms of each of these factors into peripheral blood cells in order to dissect the mechanisms by which IL-6 stimulates the HSP90 promoter. For these reasons, some previous studies have investigated the means of delivering genes to peripheral blood mononuclear cells. Thus, it has been reported that such gene delivery can be achieved either by using naked DNA (Jensen et al., 2000), retroviral viral vectors based on Moloney murine leukemia virus (Dardalhon et al., 2000), lentiviral vectors (Li et al., 2000; Chinnasamy et al., 2000), or adeno-associated viral vectors (Dunbar, 1999). Nonetheless, each of these approaches has considerable disadvantages. Thus, since retroviral vectors based on Moloney murine leukemia virus will only infect dividing cells, it is necessary to stimulate the peripheral blood cells in some way in order to obtain efficient delivery (Dardalhon et al., 2000). Similarly, lentiviral vectors do not appear to achieve high efficiency gene delivery unless HIV-1 accessory proteins are also provided (Chinnasamy et al., 2000). Clearly, this would not be appropriate in a clinical situation and would unnecessarily complicate experimental procedures. Moreover, in a comparison between a lentiviral vector and an adeno-associated virus vector, it was concluded that while the adenoassociated virus vector gave a higher initial efficiency of gene delivery, this fell off rapidly with the time compared to the retroviral vector (Dunbar, 1999). Interestingly, efficient gene delivery to human T cells has been reported with a vector based on the herpes virus, Herpes Saimiri (Hiller et al., 2000). However, since this virus transforms T cells to a continuously proliferating phenotype, it is evidently not suitable for clinical use in humans or for experimental procedures where it is necessary to maintain the normal phenotype of the target cell. Our laboratory has extensively investigated the use of vectors based on another member of the herpes virus family, herpes simplex virus type 1 (HSV-1), for gene delivery to different cell types (for review, see Latchman, 2000, 2001). Although this virus naturally infects neuronal cells establishing life-long asympto-
matic latent infections, it can infect a variety of different cell types. Thus, for example, we have shown that the virus can deliver reporter genes effectively to cardiac cells both in culture and in the intact heart (Coffin et al., 1996a), and can also deliver to human dendritic cells and hematopoietic cells with high efficiency (Coffin et al., 1998). Moreover, this virus has a number of advantages as a gene delivery vector, notably the ability to be grown efficiently in culture, and since it is a very large DNA virus with a genome of over 150 kilobases, the ability to accept several different inserts encoding different genes which may need to be delivered in parallel. A key issue in the development of vectors derived from HSV, both for gene therapy and for testing gene functions, is the need to disable the virus so that it cannot carry out a lytic infection while retaining the ability to prepare viral stocks in culture and deliver genes effectively to a specific cell type. In our previous studies, we have demonstrated that the degree of disablement required for optimal gene delivery differs depending on the cell type being targeted. Thus, for example, in the initial stages of vector preparation, we constructed a virus which lacks the viral gene encoding the ICP34.5 protein and is therefore nonneurovirulent upon injection into the brain being unable to replicate in a variety of nondividing cells including neurones (Coffin et al., 1996b). This virus continues to replicate on dividing cells in culture, thereby allowing stocks of the vector to be prepared. Moreover, this virus produces optimal gene delivery to the peripheral nervous system following injection into the mouse footpad (Palmer et al., 2000). However, following direct injection into brain, this virus produces only low efficiency gene delivery (Coffin et al., 1996b). For higher efficiency gene delivery to the central nervous system, it is necessary to remove the viral gene encoding the immediate early protein ICP27 (Howard et al., 1998). This results in a virus that is unable to replicate lytically in any cell type, and must be grown in culture on cells artificially expressing ICP27. Hence, an entirely replication-deficient virus is required for the efficient gene delivery to the central nervous system, whereas a partially incompetent virus produces optimal delivery to the peripheral nervous system. Similarly, virus lacking ICP34.5 and ICP27 can be further disabled,
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
for example, by the removal of the gene encoding another immediate early protein ICP4. Although this protein must similarly be provided within cultured cells for replication to occur, infection of cells with this highly disabled virus lacking ICP34.5, ICP4, and ICP27 results in the production of essentially noviral protein and represents an optimally disabled virus for gene therapy applications in the central nervous system (Thomas et al., 1999b; Lilley et al., 2001). In this report, we have therefore compared viral vectors lacking ICP34.5, and ICP27 or ICP34.5, ICP27, and ICP4 in terms of their ability to deliver genes in a nontoxic and efficient manner to human peripheral blood mononuclear cells in order to determine whether HSV can be used as a vector for these cells and the optimal level of disablement required for a safe and efficient vector.
237
2.3. Infection Cells of 106 were taken and spun down in 15-ml falcon tubes. Most of the media was removed by pipetting and virus at Multiplicities of Infection (MOI) of 10 was added. The tubes were flicked and left at 37 jC, 5% CO2 for 1 h. The cells were then resuspended in 1 ml of full grown RPMI (10% Foetal Calf Serum, 1% Penicillin Streptomycin) and transferred to one well of a 24-well plate. The cells were then left overnight and assessed on the three following days. 2.4. Infection assessment 2.4.1. GFP expression The cells were observed under an inverted fluorescent microscope and scanned for cells expressing GFP. Total gene delivery rate was calculated as the percentage of green cells from the whole population.
2. Methods and materials
Blood of 20– 50 ml was taken from normal volunteers and centrifuged at 1500 rpm (450 g), 4 jC for 10 min. The serum was then removed and stored for further tests. The cells were then diluted 1:3 with RPMI serum-free media (GIBCO BRL, Scotland) and layered over 15 ml of lymphoprep (Pharmacia, UK) in a 50-ml falcon tube. The samples were then spun at 1500 rpm (450 g) for 45 min, at 4 jC with zero brake rate. The white blood cell layer was then collected with a sterile plastic Pasteur pipette and diluted 1:5 with RPMI serum-free media. The cells were spun at 1500 rpm (450 g) for 10 min, at 4 jC. The pellet was then resuspended in 5 ml of serum-free RPMI. The cell concentration was measured by counting with Trypan blue.
2.4.2. X-gal staining A sample of 100 Al was taken from a resuspended culture of infected PBMCs. These were then centrifuged in a 15-ml falcon tube at 1500 rpm (450 g) at 4 jC for 10 min. The supernatant was removed and the cells were fixed in 100 Al of 1 PBS/0.1% glutaraldehyde. The cells were incubated at room temperature for 10 min and spun again at 1500 rpm at 4 jC for 10 min. The fix was removed and the cells resuspended in 200 Al of X-gal solution (150 Ag/ml 4Chloro, 5-bromo, 3-indolyl-b-galactoside in DMSO, 1 PBS, 10 mM sodium phosphate, 1 mM MgCl2, 3.3 mM K4Fe(CN)6, 3.3 mM K3Fe(CN)6). The cells were plated out in a 24-well plate, and incubated overnight at 37 jC. The cells were then observed under the light microscope and the percentage of cells expressing the lacZ reporter gene (blue cells) was determined.
2.2. Cell counting
2.5. Cell culture
When cells are resuspended in Trypan blue, dead cells do not exclude the dye and therefore turn blue. From a homogeneous mix of RPMI and PBMCs, 10 Al were taken and added to 10 Al of a 0.04% Trypan blue solution. The later mix of 10 Al was loaded on a hemocytometer counting chamber.
BHK cells were maintained using standard tissue culture procedures. Virus strain 1764 was propagated on BHK cells, while virus strains 17 + 27- and 176427-4- were grown on a BHK-derived cell line described previously which expresses ICP4 and ICP27 (Thomas et al., 1999b; Lilley et al., 2001)
2.1. Isolation of PBMCs
238
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
(27/12/M:4 cell line). This cell line complements for the genes that have been deleted and there is no sequence overlap between the viral genes present in the cell line and the deleted viral genome. Complementing cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% Fetal Calf Serum, 1% Penicillin Streptomycin and in the case of 27/12/M:4 with 800 Ag/ml of G418 and 750 Ag/ml of Zeocin. Hexamethylene bisacetamide (HMBA) was included in the growth media. 2.6. Viral culture Ten 175 cm2 flasks of 100% confluent cells were split by using a versene/10% Trypsin (GIBCO) solution and resuspended in DMEM FGM. The cells were then transferred to ten 850 cm2 roller bottles (100 ml/bottle), and incubated at 37 jC, 5% CO2 for 2 days, or until 70 –80% confluent. The cells were then infected with 106 pfu of virus/bottle, and grown in fresh FGM for another 2 –3 days at 32 jC, or at least until they are ready to harvest (media becomes orange and cells become detached from vessel surfaces). The cells were harvested and frozen at 80 jC. The cells were then thawed and spun at 3500 rpm (2500 g) at 4 jC for 30 min to remove the cell debris. The supernatant was then passed through a 0.45-Am filter and spun again at 12,000 rpm (28,000 g) at 4 jC for 2 h in a JA12 rotor. The pellet was resuspended in its own medium or a small volume (approximately 200 Al) of serum-free DMEM, and sonicated to homogeneity. The virus stock was then titred and stored at 70 jC. 2.7. Viral titre Appropriate cells for lytic growth of each virus were grown on a 6-well plate, at 37 jC, 5% CO2, until 70– 80% confluent. Successive 1:10 dilutions of the virus starting from 10 2 to 10 10 were prepared in 500 Al of Serum Free DMEM. These were added to the cells. The cells were then incubated at 37 jC, 5% CO2, for 1 h. After the incubation, 2 ml of 1:2 of 1.6% carboxymethyl cellulose full grown medium DMEM was added to each well. For noncomplementing cell lines (BHK), 3 mM hexamethylene bisacetamide (HMBA) were added to the media.
2.8. Western blotting The cell pellets were freeze-thawed five times in liquid nitrogen. The samples were then resuspended with 20 Al of 0.1 M Tris (pH 8). They were then spun at 14,000 rpm (13,000 g) at 4 jC for 30 min. The supernatant was collected and used on a 10% acrylamide gel. Alternatively, the cells were directly lysed in loading buffer (2.3% sodium dodecyl sulphate (SDS), 0.0625 M Tris, 10% glycerol, 5% h-mercaptoethanol and bromophenol blue) at a 1:1 ratio. The samples were subjected to SDS-polyacrylamide gel electrophoresis (PAGE). The resolving gel consisted of 10% acrylamide (in 1.5 M Tris, 0.4% SDS, pH 8.8), and the stacking/loading gel consisted of 5% acrylamide (in 0.5 M Tris, 0.4% SDS, pH 6.8). The samples were denatured at 95 jC for 5 min in loading buffer, and then immediately placed on ice and loaded on the gel. The gels were run in running buffer (0.129 M glycine, 0.025 M Tris, 0.1% SDS, pH 8.3) at 30 mA, for approximately 4 h. They were then transferred onto a Hybond-C membrane (Amersham, UK), in blotting buffer (0.192 M glycine, 0.025 M Tris, 20% methanol pH 8) at 200 mA overnight at 4 jC. The membrane was then blocked in 4% Marvel/0.1% PBS Tween for 2 h, on a shaker. Then, appropriate antibody was added at 1:1000 dilution in 4% Marvel/0.1% PBS Tween, and placed on a shaker for 1 h or more to intensify the signal. The membrane was then briefly washed twice in 0.1% PBS Tween, and incubated on the shaker for another 1 h with the secondary antibody conjugated with horseradish peroxidase (HRP) (1:1000 dilution in 4% Marvel/0.1% PBS Tween). The membrane was passed through a series of 5 5-min washes in 0.1% PBS Tween and then incubated in an enhanced chemiluminescent (ECL) (Amersham) solution for approximately 1 min. The membrane was then exposed to X-OMAT film from Kodak, for 1 min. Further washes or exposure periods took place if necessary. 2.9. Virus nomenclature All viruses were derived from HSV-1 strain 17syn+ (Brown et al., 1973). Viruses 2, 5, and 6 have the 1814 mutation in the gene encoding VP16 (VMW65) and the genes ICP34.5 and ORF P completely deleted
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
(between nt. 124945 and 125723) (Coffin et al., 1996a,b; Ace et al., 1989). Viruses 3 and 4 also have a deletion of nt. 113273 to 116869 which contain the genes UL54, 55 and 56 (Howard et al., 1998). UL54 encodes for the essential IE protein ICP27, while UL55 and 56 are nonessential genes. Viruses 3 and 4 were also deleted for the endogenous LATP2 regions (nt. 118768 and 120470) in order to prevent recombination instability that was found to occur after insertion of LATP2-containing expression cassettes outside of the LAT region (Palmer et al., 2000).
3. Results and discussion To assess the ability of HSV to deliver a reporter gene to peripheral blood mononuclear cells (PBMCs) and the optimal degree of viral disablement, we utilised four viruses differing in their degree of disablement but each having the Cytomegalovirus (CMV) immediate early gene promoter driving the expression of a green fluorescent protein (GFP) reporter gene (see Fig. 1 and Palmer et al., 2000; Thomas et al., 1999b; Lilley et al., 2001, for further details of these viruses). Virus 1 contains the CMV-GFP construct inserted into the nonessential UL43 gene which has no effect on the functioning of the virus. It is therefore an essentially wild type virus. Virus 2 contains the ICP34.5 mutation as well as a mutation in the gene encoding Vmw65, which transactivates the viral immediate early genes with the CMV-GFP gene inserted into the nonessential LAT P2 site. It is, however, still able to replicate on dividing cells but not on nondividing cells, and is therefore a partially disabled virus. Virus 3 contains, as well as the ICP34.5 mutation, a mutation in the gene encoding ICP27 which renders it nonreplication competent on any cell type. Finally, the most disabled virus, virus 4, also contains mutations in both copies of the genes encoding ICP4 providing the maximal level of disablement by inactivating two essential immediate early genes. In initial experiments, we infected PBMCs with each of these viruses at a multiplicity of infection (MOI) of 10 plaque forming units per cell. The effect on the cell viability of such viral infection was measured on the basis of the ability of live cells to
239
exclude Trypan blue. As illustrated in Fig. 2, no significant decrease was observed in any of the viral infected cells at 1, 2, or 3 days after infection. A small decrease was observed with virus 3 but was not significant. Most interestingly, infection, even with wild type virus (virus 1), demonstrated no significant decrease in viability. This is in accordance with previous findings suggesting that B cells, T cells, and monocytes do not support productive infection with HSV (Tenney and Morahan, 1987; Albers et al., 1989; Kemp et al., 1990). Having established that none of the viruses caused significant death of PBMCs, we investigated their ability to deliver the GFP reporter gene to PBMCs by measuring the number of cells in the culture which were positive for GFP fluorescence at intervals after infection. The results of these experiments (Fig. 3) revealed significant differences between the various viruses in their efficiency of gene delivery. Thus, viruses 3 and 4, which contained mutations in the immediate early gene encoding ICP27, either alone (virus 3) or in combination with a mutation in ICP4 (virus 4), showed relatively low efficiencies of gene delivery with 10% of the PBMCs showing GFP fluorescence even at 1 day after infection. Similarly, the essentially wild type virus (virus 1) showed a very low efficiency of gene delivery. The greatest efficiency of gene delivery with over 40% of the cells showing GFP fluorescence at day 1 after infection was achieved with the semidisabled virus (virus 2), in which the genes encoding ICP34.5 and the virion transactivator Vmw65 had been inactivated but still contained functional copies of all the genes required for replication in all cell types. Hence, a partially disabled virus lacking functional ICP34.5 or Vmw65 appears to provide efficient gene delivery to PBMCs with nearly half of the cells expressing the reporter gene at 1 day after infection. The high efficiency of this virus compared to either less or more disabled viruses is likely to reflect the balance between inhibitory effects on nonviral transgene expression, which may occur in nondisabled viruses and the impaired efficiency of gene delivery with highly disabled viruses. This balance is different in different cell types such as neurones of the central versus peripheral nervous systems (Coffin et al., 1996b; Lilley et al., 2001), necessitating the need to determine the optimal vector for each cell type.
240
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
241
Fig. 2. Viability of PBMCs infected with the viruses shown in Fig. 1 as determined by the ability of live cells to exclude Trypan blue at intervals after infection. Values indicate the mean of at least nine experiments with each virus whose standard deviation is shown by the bars.
To confirm that this partially disabled virus will infect all cell types in PBMC, PBMCs were infected with virus. FACS was then used to separate monocytes from lymphocytes on the basis of cell size, and B cells were distinguished from T lymphocytes on the basis of their reactivity with antibodies to CD19 and CD3. Successful infection (as assayed by GFP positivity) was observed in all cell types with the greatest efficiency being observed in monocytes. Hence, our virus can successfully deliver a reporter gene to T lymphocytes, B lymphocytes, and monocytes. Although virus 2 produced high efficiency gene delivery as indicated in Fig. 4, the number of cells showing GFP fluorescence declines with time, reducing to approximately 20% of the cells after 2 days of infection and 10% of cells after 3 days. As we did not observe significant cell death following infection with
virus 2 (see Fig. 2), this loss of GFP positivity does not appear to be due to the death of infected cells during the culture period. To investigate whether the decline in GFP positivity was due to silencing of the CMV promoter which drives GFP in all the viruses tested so far, we used two further viruses (see Fig. 1) which both contained mutations in the genes encoding ICP34.5 and Vmw65 as in virus 3, but in which the GFP gene was driven either by the murine sarcoma virus promoter (virus 5) or the Moloney murine leukemia virus promoter (virus 6). As expected from our previous experiments, both these viruses have no significant effect on the viability of infected PBMCs (data not shown). When the percentage of GFP positive cells was determined at intervals following infection with these viruses, virus 5 produced a similar number of
Fig. 1. Details of the viruses used in this study. An indicates a gene which has been functionally inactivated or deleted. The site of insertion of each virus is shown. Note that deletion of the LATP2 element in the viruses does not affect their replication ability but is essential to prevent recombination in stability due to the presence of elements (LAP1 and LATP2) designed to give longer term expression in the inserted construct (Thomas et al., 1999a). For further details of these viruses, see Coffin et al. (1996a,b), Howard et al. (1998), Lilley et al. (2001), and Palmer et al. (2000).
242
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
Fig. 3. Percentage of GFP positive PBMCs at daily intervals following infection with each of the viruses 1 – 4 shown in Fig. 1. Values indicate the mean of at least nine experiments with each virus whose standard deviation is shown by the bars.
GFP positive cells at day 1 after infection to that observed with virus 3 (Fig. 4). In contrast, the percentage of GFP positive cells observed with virus 6 was somewhat lower, suggesting the MMLV promoter is somewhat weaker than the MSV or CMV promoters in PBMCs. However, in both cases, a decline in the number of cells showing GFP fluorescence was observed at days 2 and 3, indicating that changing the viral promoter used to drive GFP expression does not result in longer persistence of GFP fluorescence in the infected cells. To investigate the time course of GFP expression further, we utilised Western blotting with an antibody to GFP. Cells were infected with virus 5 and harvested at various times after infection. As illustrated in Fig. 5, GFP expression in the infected cells was not observed after 4 h of infection but was observed at both the 8and 24-h time points. Since the virus enters the infected cell within 1 h after exposure, this indicates that the GFP being detected in the infected cells is not the result of the pseudotransfer of GFP made in the cells on which the virus was previously grown and
then packaged into the viral particle. Rather, it represents new GFP protein derived from the GFP gene in the virus and synthesised in the PBMCs providing further evidence that our vectors can express a reporter gene in the cells. To investigate the persistence of GFP expression, we also carried out Western blotting on extracts prepared from cells infected with virus 5 at 1, 2, and 3 days after infection. Surprisingly in this experiment (Fig. 6), we detected high levels of GFP at all time points after infection with no noticeable decline in GFP expression over the 3-day period despite the decline in the number of cells showing GFP fluorescence (compare Figs. 4 and 6). This suggests that GFP is either stable in the infected cell or continues to be made so that protein levels remain similar, but that GFP fluorescence is quenched in the PBMCs over the 3-day time period so that considerably less cells appear positive for GFP even though GFP protein is still present. To confirm this, we stained PBMCs infected with virus 5 using an antibody to GFP and compared the
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
243
Fig. 4. Percentage of PBMCs which were GFP positive at daily intervals following infection with viruses 5 and 6 illustrated in Fig. 1. Values indicate the mean of five experiments with each virus whose standard deviation is shown by the bars.
percentage of cells staining with that observed by GFP fluorescence. As illustrated in Fig. 7, the percentage of cells which reacted with the anti-GFP antibody was higher at every time point tested than the number of cells showing GFP fluorescence. Moreover, there was very little decrease in the number of antibody positive cells with time for at least 5 days after infection.
Hence, GFP fluorescence underestimates the number of infected cells in PBMC culture. In addition, it is clear that GFP expression is relatively stable with time in infected PBMCs as assayed by immunofluorescence and Western blotting, and that the decline in GFP fluorescence is due to quenching of the signal with time.
Fig. 5. Western blotting with an antibody to GFP using cells prepared at the indicated times after infection with virus 5 or from uninfected cells. The results of Western blotting for GFP and for the control actin protein are illustrated.
244
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
Fig. 6. Western blotting of cells prepared at the indicated periods after infection with virus 5 and probed with antibody to h-galactosidase, GFP or actin.
Interestingly, since virus 5 contains a h-galactosidase gene under the control of the Rous sarcoma virus promoter (Fig. 1), we were also able to measure the
levels of a second reporter gene in these experiments. As illustrated in Fig. 6, Western blotting with an antibody to h-galactosidase similarly revealed high
Fig. 7. Percentage of PBMCs showing either GFP fluorescence or positive staining with antibody to GFP at the indicated time after infection with virus 5.
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
levels of this protein being made in the infected cells, and showed that it too was clearly detectable at high level at 3 days after infection. Hence, two reporter proteins persist in the infected PBMCs for at least 3– 5 days following infection with the optimum HSV vector. Unfortunately, we were unable to determine the percentage of h-galactosidase positive cells in a culture of PBMCs using the well-known X-Gal stain for h-galactosidase activity. Although this method readily allowed us to detect h-galactosidase positive cells in cultures of neuronal and other cell types infected with our vectors, it did not result in any positive staining cells in the PBMC cultures, suggesting that h-galactosidase activity is not detectable by this means in infected PBMCs even though the protein is detectable by Western blotting (data not shown). These experiments therefore indicate that two distinct reporter genes can be expressed in PBMCs for at least 3 days in culture using an HSV vector which has been disabled by inactivation of the genes encoding ICP34.5 and Vmw65. Moreover, this can be achieved without significantly affecting the viability of the cells which appear to be resistant to cell death even following infection with an essentially wild type virus. These findings therefore indicate that the optimum HSV vector which we have identified can be used to express specific genes in PBMCs in order to assess their effect in short-term assays. Moreover, by investigating the time course of expression of a particular delivered gene using both Western blotting and immunofluorescence assays, it may be possible to carry out longer term assays in PBMCs where protein production/persistence is observed for longer periods following infection. Similarly, although further experiments will be required to test whether our optimal vector affects the functioning of PBMCs in general or of individual cell types, in the population, it is already clear that the cells survive infection even with wild type virus, with no significant loss of viability being observed. Hence, HSV represents a useful addition to the limited repertoire of virus vectors capable of delivering genes to cultured human PBMCs. This will allow, for example, the delivery of genes encoding dominant negative forms of specific transcription factors in order to elucidate the mechanisms involved in the induction of HSP90 by IL-6 which occurs in all PBMC cell types (Dhillon et al., 1993; Stephanou et
245
al., 1997, 1998). Similarly, it may ultimately be possible to specifically deliver genes to one cell type within PBMCs, following further optimisation of our vectors for a specific PBMC cell type.
Acknowledgements K.P. is supported by a Research Studentship from the Arthritis Research Campaign.
References Ace, C.I., McKee, T.A., Ryan, M., Cameron, J.M., Preston, C.M., 1989. Construction and characterization of a herpes simplex virus type I mutant unable to transduce immediate-early gene expression. Journal of Virology 63, 2260. Albers, I., Kirchner, H., Domke-Opitz, I., 1989. Resistance of human blood monocytes to infection with herpes simplex virus. Virology 169, 466. Brown, S.M., Ritchie, D.A., Subak-Sharpe, J.H., 1973. Genetic studies with herpes simplex virus type-1. Journal of General Virology 18, 329. Chinnasamy, D., Chinnasamy, N., Enriquez, M.J., Otsu, M., Morgan, R.A., Candotti, F., 2000. Lentiviral-mediated gene transfer into human lymphocytes: role of HIV-1 accessory proteins. Blood 96, 1309. Coffin, R.S., Howard, M.K., Cumming, D.V.E., Dollery, C.M., McEwan, J., Yellon, D.M., Marber, M.S., Maclean, A.R., Brown, S.M., Latchman, D.S., 1996a. Gene delivery to cardiac cells in vitro and in vivo using herpes simplex virus vectors. Gene Therapy 3, 560. Coffin, R.S., Maclean, A.R., Latchman, D.S., Brown, S.M., 1996b. Gene delivery to the central and peripheral nervous systems of mice using HSV1 ICP34.5 deletion mutant vectors. Gene Therapy 3, 886. Coffin, R.S., Thomas, S.K., Thomas, N.S.B., Lilley, C.E., Pizzey, A.R., Griffiths, C.H., Gibb, B.J., Wagstaff, M.J.D., Inges, S.J., Binks, M.H., Chain, B.M., Thrasher, A.J., Rutault, K., Latchman, D.S., 1998. Pure populations of transduced primary human cells can be produced using GFP expressing herpes virus vectors and flow cytometry. Gene Therapy 5, 718. Dardalhon, V., Jaleco, S., Rebouissou, C., Ferrand, C., Skander, N., Swainson, L., Tiberghien, P., Spits, H., Noraz, N., Taylor, N., 2000. Highly efficient gene transfer in naive human T cells with a murine leukemia virus-based vector. Blood 96, 885. Dhillon, V.B., McCallum, S., Norton, P.M., Twomey, B.M., Erkeller-Yuksel, F., Lydyard, P., Isenberg, D.A., Latchman, D.S., 1993. Differential heat shock protein over-expression and its clinical relevance in systemic lupus erythematosus. Annals of the Rheumatic Diseases 52, 436. Dunbar, C.E., 1999. In vivo marking of rhesus monkey lymphocytes by adeno-associated viral vectors: direct comparison with retroviral vectors. Blood 94, 2263.
246
K. Papageorgiou et al. / Journal of Immunological Methods 270 (2002) 235–246
Heim, D.A., Dunbar, C.E., 2000. Hematopoietic stem cell gene therapy: towards clinically significant gene transfer efficiency. Immunological Reviews 178, 29. Hiller, C., Wittmann, S., Slavin, S., Fickenscher, H., 2000. Functional long-term thymidine kinase suicide gene expression in human T cells using a herpesvirus saimiri vector. Gene Therapy 7, 664. Howard, M.K., Kershaw, T., Gibb, B., Storey, W., Maclean, A.R., Zeng, B.-W., Jenner, P., Brown, S.M., Woolf, C.J., Anderson, P.N., Coffin, R.S., Latchman, D.S., 1998. High efficiency gene transfer to the central nervous system of rodents and primates using herpes virus vectors lacking functional ICP27 and ICP34.5. Gene Therapy 5, 1137. Jensen, M.C., Clarke, P., Tan, G., Wright, C., Chung-Chang, W., Clark, T.N., Zhang, F., Slovak, M.L., Wu, A.M., Forman, S.J., Raubitschek, A., 2000. Human T lymphocyte genetic modification with naked DNA. Molecular Therapy 1, 49. Kemp, L.M., Estridge, J.K., Brennan, A., Katz, D., Latchman, D.S., 1990. Mononuclear phagocytes and HSV-1 infection: increased permissivity in differentiated U937 cells is mediated by posttranscriptional regulation of viral immediate early gene expression. Journal of Leukocyte Biology 47, 483. Latchman, D.S., 2000. Gene therapy using herpes simplex virus based vectors. Histology and Histopathology 15, 1253. Latchman, D.S., 2001. Gene delivery and gene therapy with herpes simplex-virus based vectors. Gene 264, 1. Li, Y., Starr, S.E., Lisziewicz, J., Ho, W.Z., 2000. Inhibition of HIV1 replication in chronically infected cell lines and peripheral blood mononuclear cells by retrovirus-mediated antitat gene transfer. Gene Therapy 7, 321. Lilley, C.E., Groutsi, F., Han, Z.Q., Palmer, J.A., Anderson, P.N., Latchman, D.S., Coffin, R.S., 2001. Multiply disabled herpes simplex virus vectors optimised for the widespread and long term delivery of genes to the brain and spinal cord. Journal of Virology 75, 4343. Nakajima, T., Kinoshita, S., Sasagawa, T., Sasaki, K., Naruto, M., Kishimoto, T., Akira, S., 1993. Phosphorylation at threonine235 by a ras-dependent mitogen activated protein kinase cascade is essential for transcription factor NF-IL6. Proceedings of the National Academy of Sciences of the United States of America 90, 2207. Palmer, J., Branston, R., Lilley, C.E., Robinson, M., Groutsi, F., Smith, J., Latchman, D.S., Coffin, R.S., 2000. Development and optimisation of herpes simplex virus vectors for multiple long
term gene delivery to the peripheral nervous system. Journal of Virology 74, 5604. Pan, D., Shankar, R., Stroncek, D.F., Whitley, C.B., 1999. Combined ultrafiltration – transduction in a hollow-fibre bioreactor facilitates retrovirus-mediated gene transfer into peripheral blood lymphocytes from patients with mucopolysaccharidosis type II. Human Gene Therapy 10, 2799. Stephanou, A.S., Amin, V., Isenberg, D.A., Akira, S., Kishimoto, T., Latchman, D.S., 1997. Interleukin-6 activates heat shock protein 90h gene expression. Biochemical Journal 321, 103. Stephanou, A.S., Isenberg, D.A., Akira, S., Kishimoto, T., Latchman, D.S., 1998. The NF-IL6 and STAT-3 signalling pathways cooperate to mediate the activation of the HSP90h gene by IL-6 but have opposite effects on its inducibility by heat shock. Biochemical Journal 330, 189. Stroncek, D.F., Hubel, A., Shankar, R.A., Burger, S.R., Pan, D., McCullough, J., Whitley, C.B., 1999. Retroviral transduction and expansion of peripheral blood lymphocytes for the treatment of mucopolysaccharidosis type II, Hunter’s syndrome. Transfusion 39, 343. Tenney, D.J., Morahan, P.S., 1987. Effects of differentiation of human macrophage-like U937 cells on intrinsic resistance to herpes simplex virus type 1. Journal of Immunology 139, 3076. Thomas, S.K., Gough, G., Latchman, D.S., Coffin, R.S., 1999a. The herpes simplex virus latency associated transcript encodes a protein which greatly enhances virus growth, can compensate for deficiencies in immediate-early gene expression and is likely to function during the reactivation from virus latency. Journal of Virology 73, 6618. Thomas, S.K., Lilley, C.E., Latchman, D.S., Coffin, R.S., 1999b. Equine herpes virus 1 gene 12 can substitute for Vmw65 in the growth of herpes simplex virus 1 (HSV1) allowing the generation of optimised cell lines for the propagation of vectors with multiple immediate-early gene defects. Journal of Virology 73, 7399. Wegenka, U.M., Buschmann, J., Lutticken, C., Heinrich, P.C., Horn, F., 1993. Acute-phase response factor, a nuclear factor binding to acute-phase response elements, is rapidly activated by interleukin-6 at the posttranslational level. Molecular and Cellular Biology 13, 276. Williams, D.A., Smith, F.O., 2000. Progress in the use of gene transfer methods to treat genetic blood diseases. Human Gene Therapy 11, 2059.