Herpesvirus saimiri-based vector biodistribution using noninvasive optical imaging

Abstract

Herpesvirus saimiri (HVS) is capable of infecting a range of human cell types with high efficiency and the viral genome persists as high copy number, circular, nonintegrated episomes which segregate to progeny upon cell division. This allows the HVS-based vector to stably transduce a dividing cell population and provide sustained transgene expression for an extended period of time both in vitro and in vivo. Here we assess the dissemination of HVS-based vectors in vivo following intravenous and intraperitoneal administration. Bioluminescence imaging of an HVS-based vector expressing luciferase demonstrates that the virus can infect and establish a persistent latent infection in a variety of mouse tissues. Moreover, the long-term in vivo maintenance of the HVS genome as a nonintegrated circular episome provided sustained expression of luciferase over a 10-week period. A particularly high level of transgene expression in the liver and the ability of HVS to infect and persist in hepatic stellate cells suggest that HVS-based vectors may have potential for the treatment of inherited and acquired liver diseases.

Introduction

Noninvasive imaging of gene expression in vivo has emerged as a powerful tool for gene therapy applications during the last few years.1, 2 Several diagnostic techniques have been developed for clinical applications, specifically magnetic resonance imaging, optical imaging and radionuclide imaging techniques such as positron emission tomography.3, 4, 5, 6, 7, 8 Many of these technologies have been developed for imaging in laboratory animals as the ability to monitor the location, magnitude and kinetics of gene expression without the need of sacrificing the subject has obvious benefits. Fluorescence and bioluminescence reporter genes, such as the green fluorescent protein (GFP) and luciferase, are particularly attractive as optical signatures due to their low cost in living animal studies.9, 10 Bioluminescence imaging utilizes an energy-dependent reaction catalysed by a range of luciferases on substrates which emit photons that can be detected using a low-light cooled charged coupled device (CCD) or photon-counting cameras.9, 10 Both firefly luciferase (which catalyses D-luciferin to produce oxyluciferin in the presence of oxygen and cofactors emitting light with a peak wavelength at 562 nm) and renilla luciferase (which catalyses the oxidation of coelenterazine producing light at 482 nm) have been used as reporter genes in living animals.11, 12, 13 The major advantage of using these reporters is the minimal background fluorescence produced in animal studies, as luciferase is an insect photoprotein and not present in mammalian organisms. Bioluminescence imaging has been utilized to evaluate a number of nonviral and viral vector systems in small animal models.14, 15, 16, 17 These experiments have validated the use of noninvasive bioluminescence imaging of vector-based gene delivery as well as imaging of metastasis and cell trafficking.3, 4, 5, 6, 7

Herpesvirus saimiri (HVS) is the prototype gamma-2 herpesvirus, or rhadinovirus.18, 19 It persistently infects its natural host, the squirrel monkey, without causing any obvious disease.19 Although HVS infection of other species of New World primates can result in lymphoproliferative diseases, this can be completely eliminated by deletion of the transforming genes, STP and Tip.20 The vectors utilized herein possess these deletions and as such are incapable of transforming any cell type. HVS has several properties that make it amenable to development as a gene delivery vector. HVS offers the potential to incorporate large amounts of heterologous DNA and infects a broad range of human cell lines.21, 22, 23, 24 Upon infection, the viral genome can persist by virtue of episomal maintenance and stably transfer heterologous gene expression.21, 22, 23, 24, 25, 26 The long-term presence of the HVS episome is believed not to affect cell growth, as the in vitro growth rates of uninfected human carcinoma cells and cells persistently infected with HVS-GFP showed no significant difference in the doubling time of the cell populations. Moreover, studies in a wide range of primate and human cells have previous shown no effect of the deleted STP viruses on cell transformation.27 This suggests that the maintenance of the HVS episome does not disrupt the growth machinery of human cells in vitro.28

To further assess HVS as a gene delivery vector, the in vivo properties of HVS-based vectors have been appraised. Initially, ex vivo tumour xenograft experiments were performed in nude mice comparing uninfected or HVS-infected human carcinoma cell lines.28, 29 Tumour xenografts were allowed to grow over a period of 3 months and animals were then analysed for spread of the virus, persistence of the episome and expression of the GFP transgene. Results demonstrated that the HVS-based vector remained latent in the xenograft without spreading to other organs. Moreover, the long-term maintenance of the HVS genome, as a nonintegrated circular episome, provided efficient sustained expression of a heterologous transgene.28, 29 In addition, we have demonstrated that HVS can efficiently infect solid tumour xenografts derived from a variety of human carcinoma cells via direct intratumoral injections. Upon infection of both the tumour xenografts and spheroid cultures, HVS-based vectors can establish a persistent episomal infection within the tumour xenograft allowing expression of a heterologous transgene.30 Interestingly, HVS was also shown to infect and establish a latent episomal infection within certain mouse tissues, specifically the liver and spleen.30

In this study, we have examined the delivery profiles of the HVS vector in vivo after systemic administration. Our data demonstrate that a HVS-based vector can infect and establish a persistent episomal infection in a range of mouse tissues. Moreover, the HVS episome provided sustained expression of the heterogous transgene, particularly in the liver. Further analysis shows that the HVS-based vector can also infect and persist in hepatic stellate cells and therefore may have potential for use in the gene therapy of chronic liver disease.

Results

Production and analysis of HVS-Luc

A recombinant HVS-based vector expressing the luciferase reporter gene, HVS-Luc, was engineered using the HVS bacterial artificial chromosome (HVS-BAC)31 in a two-step process (Figure 1a). Initially, a luciferase expression cassette containing the luciferase gene under the control of the SV40 promoter was inserted into a shuttle vector, p4421. In the second stage an I-Ppo-I fragment containing the luciferase cassette and the kanamycin resistance gene was excised from p4421-Luc and directly cloned into the HVS-BAC, previously linearized with I-Ppo-I. Recombinant constructs were screened by double antibiotic selection to create the HVS-BAC-expressing luciferase, HVS-Luc. Restriction digestion and pulse field gel electrophoresis analysis of HVS-Luc clones produced band sizes that were consistent with those expected from the published sequence.18 The HVS-Luc BACs differed only in the orientation of the luciferase cassette. Figure 1b shows pulse field gel electrophoresis analysis of wild-type HVS in comparison with the recombinant HVS-Luc, demonstrating correct insertion of the Luc cassette into the HVS-BAC.

Figure 1
figure1

Production and analysis of a recombinant HVS-Luc-based vector. (a) A luciferase expression cassette containing the luciferase gene under the control of the SV40 promoter alongside a kanamycin resistence cassette were inserted into the HVS-BAC at the unique I-Ppo-I restriction, to generate HVS-Luc. (b) To confirm that the heterologous DNA was inserted into the correct position within the HVS-BAC, a restriction digest on wild-type HVS and HVS-Luc was performed using I-Ppo-I and Age I restriction enzymes. (c) OMK and Miapaca cells lytically and latently infected with HVS-BAC and HVS-Luc were analysed for luciferase expression. Error bars indicate the variations between three replicate assays, each performed in duplicate.

To confirm that the recombinant HVS-Luc virus grows to levels comparable to wild-type virus, growth time courses were performed. OMK cells were infected with wild-type HVS and HVS-Luc at both high (5 × 104 PFU/well) and low (103 PFU/well) MOI in the wells of a six-well dish. The supernatant was harvested on days 2 to 7 post-infection and the virus titre measured. Similar maximum virus titres were observed for wild-type HVS and HVS-Luc (data not shown). In addition, to verify the expression of luciferase in HVS-Luc-infected cells, reporter-based assays were performed. Initially, the fully permissive OMK cells were infected with wild-type HVS or HVS-Luc, cell extracts were harvested after 48 h and assessed for luciferase activity (Figure 1c). Moreover, to assess whether HVS-Luc could express luciferase during a persistent latent infection, a stably transduced MiaPaca cell line was produced as described previously.28 MiaPaCa cells were infected with either wild-type HVS or HVS-Luc and cultured in the presence of hygromycin. After 2 weeks, only those cells which had been successfully transduced remained viable, with 100% exhibiting the RFP phenotype when analysed by fluorescence microscopy (data not shown). To assess luciferase expression in these stably transduced cells, cell extracts were harvested after a further 4 weeks in culture and assays for luciferase activity (Figure 1c). Results demonstrate that the recombinant HVS-Luc virus is able to express luciferase in both lytic and latently infected cell lines. Interestingly, comparison of the SV40 promoter activity suggests a higher level of reporter gene expression during the latent state. At present, we are unable to determine whether this is due to the SV40 promoter being downregulated during a production infection or is more active in the latently infected MiaPaCa cell line.

HVS predominately infects the liver following intravenous (i.v.) and intraperitoneal (i.p.) administration to mice

Bioluminescence was used to assess the distribution of HVS in mice. A dose of 1 × 106 PFU of HVS was injected either i.v. or i.p. into each mouse. After 3 h, an image was taken to confirm the presence of the virus and the activity of the luciferase gene (data not shown). Successive images were taken at weeks 1, 4 and 10 post-HVS injection. As expected, the principal source of light was the liver, independent of the route of administration of the HVS (Figure 2a). During the 10-weeks duration of the experiment, the activity of the luciferase gene did not change significantly (Figure 2b), which suggested that a persistent infection was established and that luciferase gene expression was stable.

Figure 2
figure2

HVS-Luc can infect mouse tissues. (a) BALB/c nu/nu mice were injected with a dose of 106 PFU HVS either i.v. or i.p. At 1, 4 and 10 weeks post-injection, the animals were injected i.p. with luciferin and imaged using the IVIS system. (b) Levels of luciferase were measured at 1, 4 and 10 weeks post-injection. An ROI was drawn and the amount of light quantified using the maximum value of photons/second/cm2/steradian. The figure shows the mean value and the standard derivation. (c) The presence of HVS DNA was identified by the PCR amplification of ORF73 from DNA extracted from tissue samples of mice at 10 weeks post-infection by i.v. and i.p. routes. GAPDH DNA was detected in all samples, indicating that sufficient amounts of DNA had been extracted from each tissue.

To verify the presence of the HVS genome in mouse tissues, PCR amplification was utilized using DNA isolated from mouse tissue samples at 1, 4 and 10 weeks post-infection. Specific amplification of the open reading frame (ORF) 73, a gene latently expressed in human carcinoma cell lines,25 was performed. In addition, as a control, GAPDH DNA was detected in all samples, indicating that sufficient amounts of DNA had been extracted from each tissue. The HVS ORF73 and mGAPDH were PCR amplified with the primers and conditions reported previously.28 Viral DNA was readily detected from a variety of mouse organs, specifically the liver, spleen, kidney, heart and lung tissues, via i.v. injection (Figure 2c). Similar results were observed for mice injected via an i.p. route (data not shown). These results demonstrate that HVS can infect a variety of mouse tissues via systemic routes and the presence of HVS episomes can still be detected 10 weeks post-injection.

HVS-Luc can establish a latent persistent infection in mouse tissues

We have previously demonstrated that HVS can persist in human tumour xenografts via a direct injection.30 To determine whether HVS can establish a persistent latent infection in mouse tissue via systemic injections, RT-PCR amplification was performed using RNA isolated from mouse tissue samples at 1, 4 and 10 weeks post-infection. As a control, GAPDH expression was detected in all samples. The establishment of a persistent infection was indicated by the expression of the latently expressed ORF73 transcript.25 Results demonstrate that ORF73 expression was observed in the liver, spleen, kidney, heart and lung tissues via i.v. injection (Figure 3). Similar results were observed for mice injected via an i.p. route (data not shown). These results demonstrate that HVS can infect and establish a persistent latent infection in a variety of mouse tissues via systemic routes.

Figure 3
figure3

The presence of HVS latent and lytic gene expression was identified by RT-PCR amplification of ORFs73, 57 and 47 from RNA extracted from each tissue sample of mice infected with HVS-Luc via the i.v. route. GAPDH RNA was detected in all samples, indicating that sufficient amounts of RNA had been extracted from each tissue.

RT-PCR analysis also identified low levels of ORF57 expression, an early lytically expressed gene,32 in mouse tissues (Figure 3). This suggests that there may be low levels of lytic gene expression within the persistently infected mouse tissues. We have previously observed a very low level of spontaneous replication in stably transduced lung carcinoma cell lines and human tumour xenografts.25, 28, 29 However, RT-PCR analysis to identify late lytic gene expression, ORF47, glycoprotein L, failed to identify any transcripts (Figure 3). Therefore, this low level of lytic gene expression does not result in late gene expression and therefore infectious virus production, and seems not to affect the long-term persistence of the HVS-Luc vector in mouse tissues.

HVS-Luc is maintained as a circular nonintegrated episome in mouse tissues

We have previously shown that the HVS genome can persist as a circular episome in human carcinoma cell lines and human tumour xenografts.25, 28, 29 To verify the existence of HVS genomes in a circular, nonintegrated episomal form in mouse tissues after systemic injection, viral DNA conformation was assessed using the Gardella gel technique. To obtain the required sensitivity to detect HVS episomes in vivo, we utilized a PCR modification of the Gardella gel technique, which has previously been used to detect latent herpesvirus episomes in vivo.33, 34, 35 In contrast to analysing the Gardella gel by Southern blotting, each lane of the gel is cut into horizontal slices after electrophoresis. Each slice was then analysed for HVS-Luc genomes by PCR. A derivative of the A549 cell line stably transduced with HVS-GFP,25 (which contains HVS as a circular nonintegrated episome) was also analysed as a suitable control. Gardella analysis was performed on mouse tissues after 4 and 10 weeks post-infection. Results show a band corresponding to episomal HVS DNA present in the liver via i.v. and i.p. injection routes at 10 weeks post-infection (Figure 4). Similar results were observed at 4 weeks and also for the spleen, kidney, heart and lung tissues (data not shown). This result suggests that HVS-Luc can establish a persistent episomal infection, where the HVS genome remains as a circular nonintegrated episome which is not incorporated into the host cell genome, within mouse tissues via systemic injection, and this is independent of a productive infection. The Gardella gel analysis suggests that HVS-Luc establishes a predominantly latent state within mouse tissues as it has previously been shown that latently infected cells contain genomes that are covalently, closed circular in conformation.33, 34, 35 In contrast, although many forms of the genome are present in productively replicating cells, Gardella analysis of such cells reveals that genomes are mainly in their linear conformation. This is demonstrated in the reactivated (TPA-induced) A549 cell line (Figure 4). Therefore, we believe that little, if any, lytic replication occurs in mouse tissues due to the lack of linear genomes identified by the modified Gardella technique.

Figure 4
figure4

HVS is maintained as a circular episome in vivo. Gardella gel and PCR analysis of latently infected A549 cells and lytically induced A549 cells and liver cells harvested 10 weeks post-infection via the i.v. and i.p. routes. Gardella gels were cut into horizontal slices after electrophoresis and melted at 65°C, and 5 μl was analysed directly by PCR for HVS genomes (ORF73).

HVS-Luc provides sustained transgene expression in mouse tissues

The noninvasive optical imaging suggests that HVS-Luc episomes can provide sustained luciferase expression in certain mouse tissues after systemic administration. To further quantify the luciferase expression in these tissues, reporter-based assays were performed. Luciferase expression was measured from mouse tissues harvested at 1, 4 and 10 weeks post-infection. Results demonstrate that luciferase expression was observed in the liver, spleen, kidney, heart and lung tissues upon i.v. injection (Figure 5). In particular, high levels of luciferase expression were observed in the liver samples. Interestingly, the levels of luciferase in each tissue remain constant throughout the 10-week period. Similar results, but slightly lower levels of expression, were observed for mice injected via an i.p. route. These data suggest that the establishment of a persistent latent episomal infection by HVS-Luc allows sustained expression of a transgene in a variety of mouse tissues, particularly high levels in the liver. Furthermore, little, if any, silencing of the heterologous SV40 promoter contained in the HVS episome was observed in these in vivo experiments during this period.

Figure 5
figure5

HVS-Luc provided sustained expression in mouse tissues. Tissues harvested from mice at week 10 post-infection were frozen in liquid nitrogen and pulverized into a fine powder. Samples were then assayed for luciferase expression. Error bars indicate the variations between three replicate assays, each performed in duplicate.

HVS-directed gene expression in hepatocytes and hepatic stellate cells

The particularly high level of luciferase expression in the liver provided by HVS-Luc led us to analyse the liver in more detail. Histological analysis of HVS-Luc-infected livers was carried out to assess the architectural damage and graded for the presence of inflammatory infiltrate. Results showed no gross signs of hepatocyte damage or increased inflammatory infiltrate, indicating that neither the viral vector nor the transgene is hepatotoxic (Figure 6). In addition, to confirm the presence of the HVS latent episome and luciferase expression throughout the liver sections, immunohistochemistry was performed using antibodies directed to ORF73 and luciferase, respectively. Immunohistochemical analysis of ORF73 and luciferase expression revealed widespread infection of hepatocytes that persisted up to at least 10 weeks post-infection (Figure 7). These data provide evidence that the HVS vector can deliver gene expression to hepatocytes; however, it was also of interest to determine if the vector may also be of use for hepatic myofibroblasts (HM). HM are key wound-healing and profibrogenic cells of the liver that can be generated in vitro by culturing freshly isolated hepatic stellate cells for several days on plastic in serum-containing media.36 Rat HM produced in this way were shown to be infected with high efficiency by an HVS-GFP vector. As shown in Figure 8a, an MOI of 0.4 was associated with a high percentage (46.3±2.19%) of GFP-positive HM, indicating that HVS vectors may be used for delivery of gene expression to this hepatic cell type.

Figure 6
figure6

Liver histology. Photomicrographs are of haematoxylin and eosin-stained frozen liver sections from HVS-Luc-infected mice at 4 and 10 weeks post-infection. Photomicrographs are at × 200 magnification and HV denotes hepatic vein.

Figure 7
figure7

Immunohistochemical stain for ORF73 and Luciferase. Photomicrographs are of negative control and ORF73 (a) and luciferase (b) stained frozen liver sections at × 200 magnification. Liver sections are from HVS-Luc-infected mice at 4 and 10 weeks post-infection, black arrows denote ORF73 and luciferase-positive hepatocytes.

Figure 8
figure8

(a) HVS-GFP-infected rat hepatic myofibroblasts. Photomicrographs at × 100 magnification of activated rat HM infected with HVS-GFP (MOI's of 0.1, 0.2 and 0.4 PFU/cell) 24 h post-infection with infection efficiencies of 46.3±2.19, 36.4±3.04 and 24.5±3.69%, respectively. HVS-GFP-expressing cells were visualized using UV fluorescence microscopy under a FITC filter. (b) The presence of HVS latent expression was identified by RT-PCR amplification of ORF from RNA extracted from infected rat hepatic myofibroblasts with HVS-Luc. Actin RNA was detected in all samples indicating that sufficient amounts of RNA had been extracted from each tissue.

To determine whether HVS can establish a persistent latent infection in rat HM cells, RT-PCR amplification was performed using RNA isolated at 3, 7 and 10 days post-infection (Figure 8b). As a control, actin expression was detected in all samples. Results demonstrate that ORF73 expression was observed in the rat HM cells at all time points, demonstrating that HVS can infect and establish a persistent latent infection in this in vitro model of hepatic myofibroblasts.

Discussion

We have previously shown that direct injection of HVS into human tumour xenografts results in a high efficiency of cell infection that can provide long-term transgene expression.30 Furthermore, we were able to detect only limited viral spread from the injection site to only the liver and spleen, suggesting that this virus may undergo very low levels of lytic replication in vivo. Little is known about the dissemination of HVS in vivo following i.v. or i.p. administration; so, to address this lack of knowledge, we constructed an HVS-based vector expressing luciferase that would allow the use of a more sensitive detection method than the previously used GFP-expressing virus.30 Through luciferase imaging and PCR analysis, we show that HVS displays wide cell tropism and provides long-term transgene expression in mice following i.v. or i.p. administration of high-titre virus.

Our previous studies had utilized GFP imaging and PCR screening of extracted tissue to determine viral dissemination in mice given an intratumoral administration of HVS.30 The construction of an HVS virus expressing luciferase allows a higher degree of sensitivity in whole-body imaging and therefore the HVS-Luc virus provides an ideal tool to study the establishment of persistent infection and viral replication in vivo. Following both i.v. and i.p. administration of HVS, luciferase bioluminescence was consistently detected in the liver. Photon flux measurements from the region of interest revealed consistent luciferase expression over the experimental period, suggesting sustained expression up to 10 weeks. Bioluminescence imaging with luciferase has the increased sensitivity of viral detection in vivo. Analysis suggests that bioluminescence imaging sensitivity is around 102–103 luciferase-expressing cells, depending on the system;37 therefore it requires a low amount of signal to achieve detection. However, the bioluminescence imaging systems cannot differentiate between sources of light that are in close anatomic proximity. Since the light is scattered due to the refractive index at cell membranes and organelles, the spatial resolution of bioluminescence imaging is approximately 2–3 mm.38 This is low in comparison to imaging modalities that use more penetrating radiation such as positron emission tomography, single positron emission computed tomography and computed tomography.37 Furthermore, the imaging data are in a two-dimensional format; so, the light detected is a consequence of all light emitted in a given plane, and further method development is needed to produce three-dimensional imaging/cross-sectional imaging. In addition, luciferase expression is under the control of the SV40 promoter which may have cell-type dependency. Therefore, to support the imaging data, we harvested the major organs from the imaged mice and performed PCR analysis to check for the presence of viral DNA. PCR analysis revealed that the majority of luciferase activity observed in the abdomen is accounted for by viral infection of the liver, whereas lower levels of infection occur in the spleen and kidney. Although viral DNA could be detected in the heart, bioluminescence imaging of mice did not reveal luciferase expression in this region, and this may be accounted for by an insufficient number of cells being infected and expressing luciferase. It is generally accepted that 106 cells need to express luciferase before detection can be achieved,38 in addition to which the heart is shielded by the ribcage which may reduce detectable light compared to the lower abdomen. Heparan sulphate has been identified as the cell surface receptor for the HVS homologue human herpesvirus-8,39 and it is likely that HVS shares the same receptor. Therefore, the infection of different tissues within the mouse following i.v. or i.p. injection of HVS is likely to be due, at least in part, to the ubiquitous expression of this cell surface molecule.

We next wanted to determine whether the viral DNA detected in mouse tissues supported viral replication. Therefore, we performed RT-PCR analysis for the HVS-associated lytic genes ORF57 and ORF47 and the latently expressed gene ORF73 as a control. We were able to detect low levels of lytic gene expression in the tissues where we had previously detected viral DNA by PCR analysis. This finding is not surprising since we have previously observed low levels of lytic replication in stably transduced cells in vitro and in vivo.25, 28, 29 However, no late gene expression was observed, suggesting that the small amount of lytic gene expression was insufficient to produce infectious virions. Moreover, the highly sensitive, modified Gardella technique could not detect any linear viral DNA from extracted tissues which would be indicative of lytic replication occurring. Also, the absence of virally induced cytopathic effect in the liver sections supports the absence of lytic replication occurring in vivo. However, it is still not possible fully to rule out the occurrence of lytic replication and, since the HVS-Luc virus maintains the required replication machinery, there is still a pressing need for production of a replication-disabled HVS vector.

One of the attractive features of using HVS as a gene therapy vector is the ability of its genome to be maintained as a circular episome that does not integrate into host DNA, thereby eliminating the risk of insertional mutagenesis. Following i.p. and i.v. injection of HVS-Luc, we detected viral DNA predominantly in the liver, but also in the lung, kidney, heart and spleen, and we were able to confirm that the DNA detected in these tissues was maintained as a nonintegrated circular episome. Also, the HVS episomes were able to drive transgene expression in mouse tissues, and luciferase activity was detectable in all tissues where viral DNA was detected. Levels of luciferase activity are consistently higher following i.v. injection. The levels of luciferase activity were highest in the liver and spleen, the two tissues which had the highest levels of HVS DNA by PCR analysis and bioluminescence imaging. Thus, the amount of viral DNA present, which can be linked to the degree of viral infection, determines the levels of gene expression in vivo.

The high level of infection and persistence of HVS in the liver suggests that it may be a useful vector for the treatment of inherited and acquired liver disease.40 Moreover, the infection and persistence within the spleen could provide a useful gene therapy vector for the genetic modification of lymphocytes. However, future analysis is now required to determine which types of cells are infected for each specific organ. To analyse the potential of HVS as a vector for liver diseases, we were interested to determine whether HVS could be used to infect hepatic myofibroblasts (HM). HM cells are key wound-healing and profibrogenic cells of the liver that are generated as a consequence of hepatic injury and inflammation.36 There is growing interest in the development of gene delivery systems for HM with a view to attenuating their profibrogenic phenotype in chronic liver disease.41, 42 Since HM are classically a feature of the injured liver, it was not possible to assess the potential for HVS-Luc to transduce them in the in vivo infection study. Therefore, we used an in vitro model of HM cells, hepatic stellate cells, and determined high levels of transduction by HVS-GFP and long-term gene expression. Although adenoviral and baculovirus vectors have been shown to infect hepatic stellate cells,41, 42, 43 this is the first report we are aware of to show a herpesvirus-based vector infecting hepatic stellate cells in vitro.

We have reported here the first study examining the dissemination of HVS in mice using noninvasive imaging techniques. It is apparent that HVS possesses many characteristics that make it a promising gene delivery vector: HVS accommodates large amounts of heterologous DNA, can infect a wide range of human cell types and avoids problems with chromosomal integration through maintenance of the genome as a high copy number episome. Furthermore, HVS can infect solid tumour masses and, following i.v. and i.p. injection, can achieve a high level of infection in the liver without evidence of cytopathic effect. With a greater understanding of the cellular receptors of HVS, it may be possible to modify HVS and increase its targeting to the liver, while reducing its ability to infect other tissues. It is now necessary to produce an HVS vector that is incapable of lytic replication to deliver a therapeutic transgene as a proof-of-principle experiment that HVS can indeed be used as a therapeutic gene delivery vector.

Materials and methods

Virus preparation and cell culture

In order to produce HVS-Luc, an initial shuttle vector, p4421-Luc, was constructed. A luciferase expression cassette, containing the luciferase gene under the control of the SV40 promoter plus SV40 polyadenylation signals and SV40 enhancer, was excised from pGL3-Control (Promega), as a SalI–SmaI fragment and subcloned into p4421, previously digested with SalI and EcoRV, generating p4421-Luc. A larger cassette was then excised from p4421-Luc containing the luciferase expression cassette and the kanamycin resistance gene with I-Ppo-I. This I-Ppo-I fragment was then directed subcloned into the linearized HVS-BAC,31 previously digested with I-Ppo-I. Recombinant BACs were selected using kanamycin and chloramphenicol antibiotic resistance screening. DNA of the recombinant BACs was purified on a Maxiprep column (Qiagen) using the manufacturer's low-copy plasmid protocol.

To generate recombinant HVS-Luc virus, the HVS-Luc DNA was transfected into the fully permissive Owl Monkey Kidney (OMK) cells maintained in Dulbecco's modified Eagle medium (DMEM) (Invitrogen) supplemented with 10% foetal calf serum (FCS), using an integrin-targeting peptide combined with lipofectin (Invitrogen) as described previously.44 To purify large virus stocks of the recombinant viruses, HVS-Luc and HVS-GFP,31 roller bottle cultures of OMK cells were infected with virus at a low moi (0.1 PFU/cell) and incubated at 37°C for 4–5 days. Virus was pelleted, resuspended in Tris buffer (100 mM Tris-HCl (pH 8.0), 50 mM NaCl and 10 mM EDTA), layered onto a sucrose gradient (20–40%) and centrifuged at 19 000 rpm for 30 min. Purified infectious virus was harvested from the central fraction, pelleted and resuspended in PBS.

HVS-Luc stably transduced MiaPaCa cells were produced as described previously.28 Briefly, 106 cells were infected with HVS-Luc at a multiplicity of infection of 1 and cultured in the presence of 300 μg/ml of hygromycin (Invitrogen). The HVS-GFP A549 stably transduced cell line has been described previously.25

Noninvasive imaging

BALB/c nu/nu mice were obtained from Harlan (Oxfordshire, UK) and kept in a germ-free environment with irradiated food and acidified water ad libitum. Experiments were conducted after appropriate ethics approval and licensing was obtained in accordance with the Guidance on the Operation of the Animals (Scientific Procedures) Act 1986 (House of Commons, 1990).

Bioluminescence in vivo was conducted using a cryogenically cooled IVIS system (Xenogen Corp., Alameda, CA, USA) connected to a computer running the Living Image 2.20 software (Xenogen Corp.). Mice were injected with a dose of 106 PFU HVS either i.v. or i.p. Before imaging, the animals were injected IP with 100 μl of a 150 mg/ml PBS solution with luciferin (Xenogen Corp.). After that, mice were anaesthetized in a plastic chamber with isofluorane and transferred to the IVIS system. A grey scale body surface image was collected under dim illumination, followed by acquisition and overlay of the pseudocolour image corresponding to the light detected by the camera during an exposition time of 5 min. A region of interest (ROI) was drawn and the amount of light quantified using the maximum value of photons/s/cm2/steradian inside the ROI.13

PCR and RT-PCR analysis

The brain, liver, lung, kidney, heart, spleen and colon were removed, snap frozen in liquid nitrogen and stored at −80°C prior to RNA, DNA and protein analysis. Tissue samples were homogenized in 10 volumes of Trizol (Invitrogen), and DNA or RNA was then extracted as described by the manufacturer. The HVS ORF73, ORF57 and mGAPDH were amplified as described previously.28 The reaction, 30 cycles (1 min, 92°C, 1 min 58°C, 1 min 72°C) was performed with 4 U of Taq Polymerase (Promega). RT-PCR analysis was performed using the Superscript™ Preamplification system (Life Technologies).

Gardella gel electrophoresis and PCR analysis

Tissue was cut into pieces of approximately 1–2 mm3, incubated with a solution of 150 U/ml collagenase/dispase (Sigma) in medium containing 10% FCS for 4 h at 37°C and then dissociated by passing through a 23-gauge needle. Episomal DNA molecules were detected using the Gardella technique with a PCR modification, as described previously.33, 34, 35 Horizontal gels were prepared in two steps. Initially, a 0.75% agarose gel in Tris-borate-EDTA buffer was poured. Once solidified, 5 cm of the gel was removed and replaced with 0.8% agarose containing 2% sodium dodecyl sulphate and 1 mg/ml of self-digested pronase (Sigma). Tissue and cell pellets were resuspended in sample buffer (15% Ficoll, 0.01% bromophenol blue) and electrophoresed at 4°C for 2 h at 40 V and then 18 h at 160 V. Each lane of the gel was then cut into horizontal slices after electrophoresis and melted at 65°C, and 5 μl was analysed directly by PCR for HVS genomes. A single round of PCR was performed to amplify ORF73 as described previously.

Luciferase reporter assay

HVS-Luc-infected cells (1 × 106) were harvested in 200 μl of passive lysis buffer (Promega). Quantitation of relative light units was determined using the dual luciferase Stop & Glo reagent using the manufacturer's directions (Promega) and a Berthold luminometer (EG & G Berthold) with a dual injector system. All assays were performed in triplicate, and each experiment was repeated a minimum of three times.

Mouse tissues were harvested at week 1, 4 and 10 post-injection and frozen in liquid nitrogen. To assess luciferase expression in these tissues, a modified assay was utilized as described previously.45 The frozen tissues (200 mg) were then pulverized into fine powder by hand grinding with a dry-ice chilled porcelain pestle and mortar, and the powder stored at −70°C. Frozen tissues were thawed and 500 μl of passive lysis buffer (Promega) was added to each sample. Samples were vortexed for 15 s, frozen and thawed three times, using alternating liquid nitrogen and 37°C water baths, and centrifuged for 3 min at 10 000 g. Quantitation of relative light units was determined using 50 μl of supernatant as described above. All assays were performed in triplicate, and each experiment was repeated a minimum of three times.

Liver pathology and hepatic stellate cell culture

Liver sections were cut at 7 μm and air dried. Sections were then stained with haematoxylin and eosin using standard protocols as described previously.46 Sections were assessed for architectural (zonal) damage including hepatocyte ballooning or changes in hepatocyte morphology and secondly graded for the presence of an inflammatory infiltrate.

HSC were isolated from normal livers of 350 g adult male Sprague–Dawley rats by sequential perfusion with collagenase and pronase, followed by discontinuous density centrifugation in 11.5% Optiprep (Life Technologies, UK). HSC were cultured on plastic in DMEM, supplemented with 100 U/ml penicillin, 100 μg/ml streptomycin, 2 mM L-glutamine, 16% FCS and maintained at 37°C at an atmosphere of 5% CO2. Activated HSC were generated by continuous culture of freshly isolated cells on plastic for 7 days. To infect with HVS-GFP, activated HSC were washed three times in serum-free DMEM and subsequently incubated with HVS-GFP (0.1, 0.2, 0.4 MOI) for 4 h at 37°C at an atmosphere of 5% CO2. Cells were then washed three times in DMEM containing 16% serum and incubated for a further 20 h. HVS-GFP-expressing cells were visualized using UV fluorescence microscopy under a FITC filter.

Immunohistochemistry

Frozen mouse liver cryosections were cut (10 μm), air-dried, then fixed in dry acetone for 15 min. Endogenous peroxidase activity was blocked by 0.06% hydrogen peroxide in methanol pretreatment for 15 min, then further blocked using the Avidin/Biotin blocking kit (Vector Laboratories, UK). The polyclonal rabbit anti-ORF73 primary antibody,47 was diluted 1:50 and incubated for 1.5 h at room temperature. The goat antiluciferase antibody (Promega) was used at 1:100 dilution, binding overnight at 4°C. Secondary and anti-IgG HRP-conjugated tertiary antibodies were incubated for 20 min (Vector Laboratories, UK). ORF73 expression was visualized by diaminobenzidine (DAB) staining. Slides were counterstained with Mayer's haematoxylin for 30 s, cleared in methanol and xylene and then mounted in DPX.

References

  1. 1

    Bogdanov Jr A, Weissleder R . The development of in vivo imaging systems to study gene expression. Trends Biotechnol 1998; 16: 5–10.

  2. 2

    Nichol C, Kim EE . Molecular imaging and gene therapy. J Nucl Med 2001; 42: 1368–1374.

  3. 3

    Min JJ, Gambhir SS . Gene therapy progress and prospects: noninvasive imaging of gene therapy in living subjects. Gene Therapy 2004; 11: 115–125.

  4. 4

    Rudin M, Weissleder R . Molecular imaging in drug discovery and development. Nat Rev Drug Discov 2003; 2: 123–131.

  5. 5

    Weissleder R . Scaling down imaging: molecular mapping of cancer in mice. Nat Rev Cancer 2002; 2: 11–18.

  6. 6

    Weissleder R, Ntziachristos V . Shedding light onto live molecular targets. Nat Med 2003; 9: 123–128.

  7. 7

    Gambhir SS . Molecular imaging of cancer with positron emission tomography. Nat Rev Cancer 2002; 2: 683–693.

  8. 8

    Groot-Wassink T et al. Quantitative imaging of Na/I symporter transgene expression using positron emission tomography in the living animal. Mol Ther 2004; 9: 436–442.

  9. 9

    Contag PR, Olomu IN, Stevenson DK, Contag CH . Bioluminescent indicators in living mammals. Nat Med 1998; 4: 245–247.

  10. 10

    Contag CH et al. Visualizing gene expression in living mammals using a bioluminescent reporter. Photochem Photobiol 1997; 66: 523–531.

  11. 11

    Contag CH, Jenkins D, Contag PR, Negrin RS . Use of reporter genes for optical measurements of neoplastic disease in vivo. Neoplasia 2000; 2: 41–52.

  12. 12

    Bhaumik S, Gambhir SS . Optical imaging of Renilla luciferase reporter gene expression in living mice. Proc Natl Acad Sci USA 2002; 99: 377–382.

  13. 13

    Wu JC, Sundaresan G, Iyer M, Gambhir SS . Noninvasive optical imaging of firefly luciferase reporter gene expression in skeletal muscles of living mice. Mol Ther 2001; 4: 297–306.

  14. 14

    Iyer M, Berenji M, Templeton NS, Gambhir SS . Noninvasive imaging of cationic lipid-mediated delivery of optical and PET reporter genes in living mice. Mol Ther 2002; 6: 555–562.

  15. 15

    Hildebrandt IJ, Iyer M, Wagner E, Gambhir SS . Optical imaging of transferrin targeted PEI/DNA complexes in living subjects. Gene Therapy 2003; 10: 758–764.

  16. 16

    Adams JY et al. Visualization of advanced human prostate cancer lesions in living mice by a targeted gene transfer vector and optical imaging. Nat Med 2002; 8: 891–897.

  17. 17

    De A, Lewis XZ, Gambhir SS . Noninvasive imaging of lentiviral-mediated reporter gene expression in living mice. Mol Ther 2003; 7: 681–691.

  18. 18

    Albrecht J et al. Primary structure of the herpesvirus saimiri genome. J Virol 1992; 66: 5047–5058.

  19. 19

    Fickenscher H, Fleckenstein B . Herpesvirus saimiri. Philos Trans R Soc 2001; 356: 545–567.

  20. 20

    Duboise SM et al. STP and Tip are essential for herpesvirus saimiri oncogenicity. J Virol 1998; 72: 1308–1313.

  21. 21

    Grassmann R, Fleckenstein B . Selectable recombinant herpesvirus saimiri is capable of persisting in a human T-cell line. J Virol 1989; 63: 1818–1821.

  22. 22

    Simmer B et al. Persistence of selectable herpesvirus saimiri in various human haematopoietic and epithelial cell lines. J Gen Virol 1991; 72: 1953–1958.

  23. 23

    Stevenson AJ et al. Assessment of herpesvirus saimiri as a potential human gene therapy vector. J Med Virol 1999; 57: 269–277.

  24. 24

    Stevenson AJ et al. Herpesvirus saimiri-based gene delivery vectors maintain heterologous expression throughout mouse embryonic stem cell differentiation in vitro. Gene Therapy 2000; 7: 464–471.

  25. 25

    Hall KT et al. Analysis of gene expression in a human cell line stably transduced with herpesvirus saimiri. J Virol 2000; 74: 7331–7337.

  26. 26

    Calderwood MA, White RE, Whitehouse A . Development of herpesvirus-based episomally maintained gene delivery vectors. Expert Opin Biol Ther 2004; 4: 493–505.

  27. 27

    Desrosiers RC, Daniel PS, Waldron LM, Letvin NL . Nononcogenic deletion mutants of herpesvirus saimiri are defective for in vitro immortalization. J Virol 1986; 57: 701–705.

  28. 28

    Smith PG, Coletta PL, Markham AF, Whitehouse A . In vivo episomal maintenance of a herpesvirus saimiri-based gene delivery vector. Gene Therapy 2001; 8: 1762–1769.

  29. 29

    Giles MS et al. The herpesvirus saimiri ORF73 regulatory region provides long-term transgene expression in human carcinoma cell lines. Cancer Gene Ther 2003; 10: 49–56.

  30. 30

    Smith PG et al. Efficient infection and persistence of a herpesvirus saimiri-based gene delivery vector into human tumour xenografts and multicellular spheroid cultures. Cancer Gene Ther 2005; 12: 248–256.

  31. 31

    White RE, Calderwood MA, Whitehouse A . Generation and precise modification of a herpesvirus saimiri bacterial artificial chromosome demonstrates that the terminal repeats are required for both virus production and episomal persistence. J Gen Virol 2003; 84: 3393–3403.

  32. 32

    Goodwin DJ et al. The open reading frame 57 gene product of herpesvirus saimiri shuttles between the nucleus and cytoplasm and is involved in viral RNA nuclear export. J Virol 1999; 73: 10519–17337.

  33. 33

    Decker LL, Klaman LD, Thorley-Lawson DA . Detection of the latent form of Epstein–Barr virus DNA in the peripheral blood of healthy individuals. J Virol 1996; 70: 3286–3289.

  34. 34

    Stewart JP et al. Lung epithelial cells are a major site of murine gammaherpesvirus persistence. J Exp Med 1998; 187: 1941–1951.

  35. 35

    Decker LL, Babcock GJ, Thorley-Lawson DA . Detection and discrimination of latent and replicative herpesvirus infection at the single cell level in vivo. Methods Mol Biol 2001; 174: 111–116.

  36. 36

    Friedman SL . Molecular regulation of hepatic fibrosis, an integrated cellular response to tissue injury. J Biol Chem 2000; 275: 2247–2250.

  37. 37

    Edinger M et al. Advancing animal models of neoplasia through in vivo bioluminescence imaging. Eur J Cancer 2002; 38: 2128–2136.

  38. 38

    Luker GD et al. Noninvasive bioluminescence imaging of herpes simplex virus type 1 infection and therapy in living mice. J Virol 2002; 76: 12149–12161.

  39. 39

    Akula SM, Wang FZ, Vieira J, Chandran B . Human herpesvirus 8 interaction with target cells involves heparan sulfate. Virology 2001; 282: 245–255.

  40. 40

    Prieto J et al. The promise of gene therapy in gastrointestinal and liver diseases. Gut 2003; 52 (Suppl 2): ii49–ii54.

  41. 41

    Janoschek N, van de Leur E, Gressner AM, Weiskirchen R . Induction of cell death in activated hepatic stellate cells by targeted gene expression of the thymidine kinase/ganciclovir system. Biochem Biophys Res Commun 2004; 316: 1107–1115.

  42. 42

    Gao R et al. High efficiency gene transfer into cultured primary rat and human hepatic stellate cells using baculovirus vectors. Liver 2002; 22: 15–22.

  43. 43

    Weiskirchen R et al. Comparative evaluation of gene delivery devices in primary cultures of rat hepatic stellate cells and rat myofibroblasts. BMC Cell Biol 2000; 1: 4.

  44. 44

    Hart SL et al. Lipid-mediated enhancement of transfection by a nonviral integrin-targeting vector. Hum Gene Ther 1998; 9: 575–585.

  45. 45

    Manthorpe M et al. Gene therapy by intramuscular injection of plasmid DNA: studies on firefly luciferase gene expression in mice. Hum Gene Ther 1993; 4: 419–431.

  46. 46

    Wright MC et al. Gliotoxin stimulates the apoptosis of human and rat hepatic stellate cells and enhances the resolution of liver fibrosis in rats. Gastroenterology 2001; 121: 685–698.

  47. 47

    Hall KT et al. Characterization of the herpesvirus saimiri ORF73 gene product. J Gen Virol 2000; 81: 2653–2658.

Download references

Acknowledgements

This work was supported in part by grants to AW from the Association of International Cancer Research, Yorkshire Cancer Research, Candlelighter's Trust and the Royal Society, to NL from Cancer Research UK, and to DAM from the Wellcome Trust (050443/Z and 068524/Z/02/Z) and the Medical Research Council (COG component grant 69900279).

Author information

Correspondence to A Whitehouse.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Smith, P., Oakley, F., Fernandez, M. et al. Herpesvirus saimiri-based vector biodistribution using noninvasive optical imaging. Gene Ther 12, 1465–1476 (2005). https://doi.org/10.1038/sj.gt.3302543

Download citation

Keywords

  • herpesvirus
  • vector
  • live imaging
  • biodistribution

Further reading