B1 Lymphocytes and Myeloid Dendritic Cells in Lymp
http://www.100md.com
病菌学杂志 2005年第6期
Program of Infection and Cancer, Abteilung F010 and INSERM U375
Program of Functional and Structural Genomic Research, Deutsches Krebsforschungszentrum
Institute of Immunology, University of Heidelberg, Heidelberg, Germany
The Stephan Angeloff Institute of Microbiology, Bulgarian Academy of Sciences, Sofia, Bulgaria
Institut de Recherche contre les Cancers de l'Appareil Digestif (IRCAD) INSERM U375, Strasbourg, France
ABSTRACT
Due to their oncolytic properties and apathogenicity, autonomous parvoviruses have attracted significant interest as possible anticancer agents. Recent preclinical studies provided evidence of the therapeutic potential of minute virus of mice prototype strain (MVMp) and its recombinant derivatives. In a murine model of hemangiosarcoma, positive therapeutic outcome correlated with high intratumoral expression of MVMp-encoded genes in tumors and lymphoid organs, especially in tumor-draining lymph nodes. The source and relevance of this extratumoral expression, which came as a surprise because of the known fibrotropism of MVMp, remained unclear. In the present study, we investigated (i) whether the observed expression pattern occurs in different tumor models, (ii) which cell population is targeted by the virus, and (iii) the immunological consequences of this infection. Significant MVMp gene expression was detected in lymphoid tissues from infected tumor-free as well as melanoma-, lymphoma-, and hemangiosarcoma-bearing mice. This expression was especially marked in lymph nodes draining virus-injected tumors. Fluorescent in situ hybridization analysis, multicolor fluorescence-activated cell sorting, and quantitative reverse transcription-PCR revealed that MVMp was expressed in rare subpopulations of CD11b (Mac1)-positive cells displaying CD11c+ (myeloid dendritic cells [MDC]) or CD45B (B220+ [B1 lymphocytes]) markers. Apart from the late deletion of cytotoxic memory cells (CD8+ CD44+ CD62L–), this infection did not lead to significant alteration of the immunological profile of cells populating lymphoid organs. However, subtle changes were detected in the production of specific proinflammatory cytokines in lymph nodes from virus-treated animals. Considering the role of B1 lymphocytes and MDC in cancer and immunological surveillance, the specific ability of these cell types to sustain parvovirus-driven gene expression may be exploited in gene therapy protocols.
INTRODUCTION
Over the last few years, naturally occurring or genetically manipulated oncolytic parvoviruses attracted significant attention as means to treat cancer. Autonomous parvoviruses are small (18 to 26 nm) nonenveloped nucleus-replicating viruses. Their capsid contains a single-stranded DNA genome of about 5,000 nucleotides. The prototype strain of minute virus of mice, MVMp, is the prototype strain of autonomous parvoviruses and one of the best-studied members to date. MVMp and the closely related parvovirus H-1 have emerged as the most suitable therapeutic agents, due to their oncotropism, infectiousness for tumor cells of both human and rodent origin, and good safety profile (9, 22, 27, 28). The mechanism of parvovirus-induced tumor suppression is not fully understood yet. Besides previously described direct oncolytic effects, the recent preclinical experimental data reporting successful treatment of hemangiosarcoma by MVMp parvovirus and its recombinant IP-10-expressing derivative revealed the role of the immune system in the strong therapeutic effect of MVMp (15). Using T-cell-deficient mice (C57BL/6 nu+/nu+) as recipients for polyoma virus middle T antigen (PymT)-transformed endothelial cell implants (termed H5V cells), we showed that animals inoculated with MVMp-infected cells did develop tumors more slowly than the ones receiving mock-treated H5V cells. However, the inhibition was of short duration: all mice in both groups eventually developed tumors and metastases and succumbed to the disease. This contrasted with the long-term antitumor protection given by MVMp in immunocompetent animals under similar conditions. Nevertheless, MVMp alone was not successful in the treatment of preestablished tumors, which required reinforcement of the virus intrinsic activity by IP-10. In this system, MVMp was thus able by itself to delay tumor development, probably due to its oncotoxic activity, yet some interaction of the virus (and harbored transgene) with the host immune system appeared to be crucial for the eventual cure of the disease. In the course of this work, MVMp expression was found to preferentially occur not only in the tumor, as expected from the oncotropic character of the virus, but also in tumor-draining lymph nodes (DLN). This came as a surprise considering the fact that the prototype and the immunosupressive strain of MVM (MVMi) are shown to be restrictive for growth in each other's host cell: e.g., T lymphocytes and fibroblasts, respectively (32). MVMi and MVMp have the same genetic organization and share 96% of their sequences (29), with the target cell specificity of the respective viruses being defined by distinct amino acid substitutions within the capsid proteins (2). However, competition experiments have shown that both viruses recognize the same receptor on each cell type, and penetration and uncoating seem not to be the restrictive steps (30). The detection of significant MVMp-driven gene expression in tumor-draining lymph nodes suggested that this barrier may not apply to all lymphoid cells and that the viral cycle may progress at least to the level of transcription in a subset of these cells. With the view of using MVMp and related parvoviruses in the fight against cancer, the interaction of these agents with lymphoid cells represents a crucial issue regarding both safety and possible immunomodulating effects that may interfere with tumor formation. This prompted us to further investigate this question in the present study by determining the cellular origin of MVMp expression in lymphoid tissues. The data obtained confirm that draining lymph nodes are a major site of MVMp expression following virus injection into various tumor grafts in recipient mice and trace this expression back to distinct subtypes of lymphoid cells.
MATERIALS AND METHODS
Cells. The mouse cell lines H5V (13) and B78/H1 (a weakly immunogenic amelanotic subclone of the murine B16 melanoma cell line) (4) were kindly provided by S. Sozzani (IRF Negri, Milan, Italy). The A9 (fibrosarcoma) cell line was obtained from the American Type Culture Collection (Manassas, Va.). All cell lines were maintained in Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 2 mM L-glutamine, 100-μg/ml penicillin, 100-U/ml streptomycin, and 10% (H5V and B78) or 5% (A9) fetal bovine serum (FBS).
Lymphocyte single-cell suspensions were prepared from spleen and lymph nodes taken from C57BL/6 mice using standard techniques. After resuspension in RPMI-1640 containing 10% FBS, nonessential amino acids, and 5 μM 2-mercaptoethanol, 5 x 106 cells/well were plated into 24-well plates and exposed to various stimuli: lipopolysaccharide (LPS; 5 μg/ml), concanavalin A (ConA; 5 μg/ml), or a combination of phorbol myristate acetate (PMA; 0.5 ng/ml) with ionomycin (0.5 μg/ml) (all purchased from Sigma (Munich, Germany). Three hours later, cells were inoculated with MVMp and harvested at different times postinfection, depending on further analysis.
Viruses. MVMp was produced and titrated by plaque assay in A9 indicator cells, as described previously (31). Virus titers are expressed in PFU per milliliter. MVMp DNA replication and protein expression in infected cell cultures were assessed by Southern and Western blotting, respectively. For Western blotting, MVMp-infected monolayers were harvested, washed, resuspended in phosphate-buffered saline (PBS), and lysed by three freeze/thaw cycles 2 or 48 h postinfection (hpi). Ten micrograms of total protein was fractionated by sodium dodecyl sulfate-8% polyacrylamide gel electrophoresis, blotted onto nitrocellulose membranes (Amersham Pharmacia Biothech Europe, Freiburg, Germany), and probed with the NS1-specific antiserum (Sp7) or VP1 and VP2-specific rabbit monoclonal antibodies (21). Modified Hirt's extraction of parvoviral DNA and subsequent Southern blotting analysis of viral replicative forms were performed as described previously (8).
Murine leukemia retrovirus (MuLV clone LP-BM5mix) was received as a gift from H. C. Morse III (National Institutes of Health, Bethesda, Md.) (25) and used to induce mouse AIDS (MAIDS) as described below.
Mice and tumor models. For all tumor models, 6- to 8-week-old female C57BL/6 mice were purchased from CRL (Sulzfeld, Germany). Hemangiosarcoma and melanoma tumors were induced by subcutaneous (s.c.) implantation of 5 x 106 H5V or 2 x 105 B78 cells in the right flanks of syngeneic C57BL/6 mice, as described previously (4). Lymphoproliferative disorder was induced by intraperitoneal (i.p.) injection of 100 μl of MuLV according to established protocols (14, 18, 25). Starting 2 weeks postinfection, MuLV-injected B6 mice display progressive immunodeficiency and lymphoproliferation ending in development of lymphomas, a syndrome called MAIDS.
MVMp was applied per animal carrying an established tumor at 5 (H5V) or 14 (B78) days postimplantation or 3 weeks postinjection of LP-BM5mix. Parvoviral infections were performed via intratumoral (i.t.) or extratumoral (i.p.) route. Since MAIDS mice do not display early visible subcutaneous tumors, the i.t. route was imitated by injection of MVMp into palpable enlarged lymph node. For an expression analysis, mice were given three daily injections of 109 PFU/day and sacrificed 24 h later. For long-term immunologic studies, mice were either exposed to the 3-week-long parvoviral treatment or injected with MVMp-infected tumor cells and sacrificed once moribund (15).
Mice were kept in laminar flow safety cabinets (BDK Luft- und Reinraumtechnik, Sonnenbühl, Germany). Procedures for animal handling, treatment, and care were performed in compliance with European guidelines.
Analysis of parvovirus expression by RT-PCR. Total RNA was extracted from isolated tissues (kept at –80°C) or cultured cells by using Trizol reagent (Life Technologies) according to the manufacturer's instructions. DNase I-treated RNA was converted to cDNA and analyzed according to reverse transcription-PCR (RT-PCR) protocols detailed previously (14). The following PCR primers were used: MVMp, 5'-ACGCTCACCATTCACGACACCGAAA-3' and 5'-ATCATAGGCCTCGTCGTGCTCTTTG-3' (415-bp-long PCR product); PymT, 5'-TTCTGAGCAACCCGACCTAT-3' and CTTCTTAGGTGGCGTTGCAT-3 (439 bp), with probe 5'-GCCACTCCTATCCCCCAACCC-3'; and ?-actin, 5'-ATGTTTGAGACCTTCAACAC-3' and 5'-ACGTCACACTTCATGATCC-3' (489 bp). PCR products were analyzed by 2% gel electrophoresis and visualized, depending on the level of expression, either by direct staining of the gels with ethidium bromide or SybrGreen II (Roche Diagnostics, Mannheim, Germany) or by blotting on Hybond N+ membranes and hybridization with fluorescein-labeled DNA probe. Enhanced chemiluminescence (ECL) detection was performed according to the manufacturer's instructions (Amersham Pharmacia Biothech Europe, Freiburg, Germany). The intensity of PCR signals was measured, and levels of expression were normalized with ?-actin as a reference for matching the cDNA input for PCR (14). In some experiments, viral transcripts were quantified by using LightCycler-based real-time quantitative RT-PCR (QRT-PCR) technology (Roche Diagnostics) (16).
FISH. Fluorescent in situ hybridization (FISH) of DNA was performed as previously described (3). Briefly, lymphocytes were pelleted on slides (Cytospin 2; Shandon, United Kingdom), air dried, and incubated for 12 h at 60°C. After RNase A and pepsin treatment, the slides were fixed with formaldehyde, treated with increasing concentrations of ethanol, and denatured in 70% formamide. An NS1 sequence-specific probe was nick labeled with biotin-14-dCTP (Invitrogen) and hybridized with the DNA on the slides overnight at 37°C. After washing and blocking with 10% bovine serum albumin, slides were exposed to streptavidin-fluorescein isothiocyanate (FITC) complex. The nucleus of the cells was visualized by DAPI (4',6'-diamidino-2-phenylindole) staining. The images were captured with a Hamamatsu digital camera using Openlab 2.2.5 software (Improvision, Ltd.).
Analysis of cytokine expression. Cytokine expression was evaluated at the RNA level by QRT-PCR (15). Murine IP-10, gamma interferon (IFN-), interleukin-10 (IL-10), and IL-12 p40-specific primers were obtained from Search-LC (Heidelberg, Germany). Protein levels were measured with previously established enzyme-linked immunosorbent assays (ELISAs) (14).
FACS analysis and cell purification. To determine the immunological status of infected mice, three-color staining of cells derived from spleen and lymph nodes was performed and analyzed by fluorescence-activated cell sorting (FACS) with the FACS DIVA (Becton Dickinson, Heidelberg, Germany). FITC-, phycoerythrin-, or biotin-labeled antibodies to T-cell receptor /? (TCR/?), CD4, CD8, ThB, CD45 (B220), Dx5 (NK 1.1), CD11b (Mac1), CD11c, CD32, CD43, CD44, CD62L, CD69, IFN-, immunoglobulin M (IgM), and IgG were from Becton Dickinson. Avidin-allophycocyanin complex was obtained from Molecular Probes Europe BV (Leiden, Netherlands). Isolation of cell populations from lymphoid organs was achieved by two- or three-color FACS of accordingly stained single-cell suspensions. Usually, 2 x 107 to 4 x 107 cells underwent the sorting procedure and purified populations were collected by centrifugation, reanalyzed for purity, suspended in lysis buffer, and processed for automatic mRNA isolation with a MagNA Pure mRNA preparation kit I (Roche), followed by QRT-PCR analysis of viral transcripts (16).
RESULTS
Tissue distribution pattern of MVMp transcripts in different tumor models. Several mouse tumor models (B78 melanoma, MAIDS-associated lymphoma, and H5V hemangiosarcoma), as well as tumor-free B6 mice, were compared in terms of parvovirus expression. Once tumors became palpable (14 days for melanoma, 4 days for hemangiosarcoma, and 3 weeks for MAIDS), animals received three daily injections of 109 PFU of MVMp via the extratumoral (i.p.) or intratumoral (i.t.) route. One day after injection, brain, thymus, lungs, kidney, intestine, liver, spleen, draining or distant lymph nodes, and tumor (when applicable) were isolated and examined by RT-PCR for the presence of viral transcripts. While the majority of organs displayed no or sporadic viral expression (data not shown), the spleen, liver, lymph nodes, and tumor represented the sites of the most prominent and consistent MVMp expression (Fig. 1A). The levels of MVMp transcripts in these lymphoid tissues were comparable after extratumoral (i.p.) MVMp inoculation and similar to the ones detected in tumor-free mice (Fig. 1A, upper panel). In contrast, the intratumoral injections of MVMp led to preferential accumulation of viral transcripts in the lymph nodes draining the inoculated tumor (DLN), which was particularly striking with hemangiosarcoma and melanoma models (Fig. 1A, lower panel). Interestingly, despite the strong lymphoproliferative reaction occurring in MAIDS, the level of lymphoid viral transcription in this model was not preferentially increased (Fig. 1A). Viral transcripts declined to nondetectable levels within 2 weeks postinoculation, although traces of viral DNA were measurable by PCR up to 1 month later (data not shown).
These data suggested that MVMp was expressed either in tumor cells metastasized to DLN or in resident lymphoid cells. Since H5V tumor cells can be easily identified with PymT as a marker, the pattern of PymT expression was evaluated in MVMp-expressing organs of hemangiosarcoma-bearing mice. As seen in Fig. 1B, tumors showed strong accumulation of PymT, whereas this marker was barely detectable in spleens and not detectable in DLN even after PCR-ECL amplification of the signal. The lack of extratumoral PymT transcripts expression in DLN that carry the highest number of viral transcripts led us to conclude that tumor cells were not the source of viral expression in these organs. A similar conclusion would apply for the melanoma model, knowing that B78/H1 melanoma cells are not metastasizing upon subcutaneous implantation. It therefore appeared likely that the extratumoral expression of MVMp was supported by some of the lymph node cells themselves.
Parvovirus infection of isolated splenocytes. To confirm the ability of some cells located in lymphoid organs of the immune system (termed immunocytes in the text below) to express MVMp, in vitro experiments were carried out, in which this parvovirus was inoculated to naive or activated single-cell suspensions derived from the spleens or lymph nodes of B6 mice (multiplicity of infection [MOI] = 10 PFU/cell). Viral infection was monitored at the DNA (Southern blot), RNA (RT-PCR), protein (Western blot), and fluorescent microscopy levels. Besides wild-type MVMp, the MVMp green fluorescent protein-expressing recombinant (MVMp/GFP) was used, allowing visualization of virus-driven gene expression in individual cells. A9 fibroblastic cells, which are highly permissive for the MVMp virus, served as a positive control. As shown in Fig. 2, at least some cells derived from lymphoid organs could be targets for replication and expression of MVMp. Viral DNA replicative forms were synthesized, peaking at 48 hpi and rapidly decaying afterwards (Fig. 2A). There was thus no evidence of infection persisting with time in these cultures. Furthermore, a significant production of virus/vector transcripts (Fig. 2B) and proteins (Fig. 2C) was observed in activated immunocytes after infection with wild-type or recombinant MVMp as well as with the GFP-encoding recombinant (Fig. 2B). It is noteworthy that, even under activation conditions, only a very small fraction (less than 5%) of these cells was able to support MVMp-driven gene expression, as was apparent from the analysis of MVMp/GFP-infected cultures by fluorescent microscopy (data not shown). It should also be stated that MVMp expression could be detected in nonstimulated immunocytes, yet to a much lower level than in cultures stimulated with ConA, LPS, or PMA-ionomycin (data not shown). Interestingly, while mRNA expression was not dependent on the method of stimulation, viral proteins and replicative forms were more abundant with the stimulatory regimens indicated in Fig. 2. Notably, neither production of progeny infectious particles—shown by plaque assay—nor significant cytotoxic effect—more than 10% of growth reduction by MTT (3-[4,5-dimethylthiazol-2-yl)]-2,5-diphenyltetrazolium bromide) assay—could be detected upon MVMp infection (data not shown). Given the small proportion of susceptible immunocytes in these cultures, it cannot be ruled out, however, that a productive and/or lytic infection took place in a few cells but got masked by the overgrowth of nonpermissive cells. As a whole, the above results were reproducible with cells derived from lymph nodes or spleen. Altogether, these observations substantiated the possibility that a minor fraction of nontumor cells residing in lymphoid organs may support MVMp infection, at least up to a certain extent.
Identification of lymphoid cell subpopulations targeted by MVMp in vivo. With the objective of tracing MVMp expression in lymphoid organs back to distinct cells, lymphoid organs were collected from infected mice and used to prepare single-cell suspensions. The fraction of these cells that were amplifying viral DNA was determined by fluorescent in situ hybridizations of cytospined cells with an NS1 DNA-specific probe. The cellular nuclei were colored blue by DAPI staining, while the viral DNA was visualized through the green FITC labeling of the hybridization probe. Results obtained with DLN from H5V-bearing mice are shown in Fig. 3A. The proportion of cells replicating the viral DNA was estimated at about 1% of the whole DAPI-stained population. This is in agreement with above in vitro experiments showing that only a small population of cultured immunocytes was competent for MVMp/GFP-mediated gene transduction. Altogether the data provide evidence to suggest that MVMp replication and expression are restricted to a discrete fraction of these cells. In order to further characterize this fraction, distinct subpopulations of immunocytes were isolated by FACS sorting and individually tested by QRT-PCR for the expression of viral transcripts. A stepwise fractionation procedure was carried out. In a first step, a 95 to 98% pure population of T lymphocytes (TCR/?+ as well as separately tested CD4+ and CD8+ cells), B lymphocytes (B220+/ThB+), and Mac1+ cells were isolated from lymph nodes or spleen. As summarized in Fig. 3B (left panel), the Mac1+-enriched cell population was a major source of virus expression in DLN of H5V tumor-bearing mice (2,500 transcripts/μg of input mRNA), while viral transcripts were nondetectable in the two other cell types. Since the Mac1 (CD11b) molecule is present on macrophages, NK cells, and subsets of B and dendritic cells, the Mac1-positive population was further sorted with additional markers. These experiments showed no viral expression in NK cells but strong accumulation of transcripts in Mac1+ B220+ (1,200 copies/μg of RNA) and Mac1+ CD11c+ (500 copies/μg of RNA)-positive cells derived from analyzed DLN (Fig. 3B, right panel) of hemangiosarcoma-bearing mice. Sorting and QRT-PCR analysis of spleens obtained from MVMp-infected H5V mice and lymphoid organs derived from na?ve mice confirmed the observed pattern (data not shown). The constellation of markers defines two rather rare populations residing in peripheral lymphoid organs: myeloid dendritic cells (Mac1+ and CD11c+) and a subset of B lymphocytes, termed B1 (Mac1+ and B220+) (Fig. 3C). Interestingly, compared with na?ve animals, DLN from H5V tumor-bearing mice showed a threefold enrichment with these two cell types (Fig. 3C), whereas the absence of progressive accumulation of B1 lymphocytes and destruction of dendritic cells were previously reported in lymph nodes of MAIDS-affected mice (18, 23). This difference between both tumor models is likely to account, at least in part, for the much greater MVMp expression detected in DLN from H5V versus MAIDS-carrying animals (Fig. 1B). The predilection of MVMp expression for Mac1+ B220+ and Mac1+ CD11c+ cells defines a new and unexpected component of MVMp tropism. This property may be exploited to design vector derivatives for the selective transduction of this narrow subset of lymphoid cells in the context of gene (immuno) therapy protocols.
Impact of MVMp on the immune status of infected mice. Since two very distinct cell populations participating in diverse immune reactions were targets for MVMp in lymphoid organs, we investigated whether infected mice showed alteration of their immune status. To this end, spleens and lymph nodes from infected or mock-treated mice were collected and analyzed by FACS for the distribution of immunocyte populations (CD4+ and CD8+ T cells, B220+ B cells, Dx5+ NK cells, and Mac1+ macrophages) and their activation status: CD69 (early activation), CD44 and CD62L (activation/memory), IFN- (activated cytotoxic T lymphocytes [CTL]), and surface Igs (activated B cells). Furthermore the expression of various cytokines was monitored by QRT-PCR and ELISA (Fig. 4). The FACS analysis of lymphoid organs failed to reveal significant changes in the distribution and activation of immunocyte populations over the course of time following exposure of either naive or tumor-bearing mice to MVMp (data not shown). The only difference was in the late reduction of memory CTL (CD8+ CD44+ CD62L–) in half of the animals studied, starting from 4 weeks after infection (Fig. 4A). This phenomenon was observed in mice exposed to the live virus or to the virus-infected cells. As mentioned above, virus expression was not detectable anymore at this time point. Since T cells did not seem to belong to the population able to support viral expression in lymphoid tissues of infected mice, the memory CTL depletion was likely to occur through an indirect mechanism. We further determined whether MVMp infection of animals affected the lymphoid expression of cytokines known to be produced (IP-10, IL-10, and IL-12 p40) or not (IFN-) by the target cells (34). As shown in Fig. 4B, a striking up-regulation of IP-10 expression was observed in lymphoid tissues derived from infected B6 mice, while no main differences were detected for the other cytokines. In vitro, MVMp itself did not induce IFN- expression in lymphocyte cultures either but significantly augmented LPS-induced IFN- production (Fig. 4C). Interestingly, no such costimulation was observed in ConA and PMA-ionomycin-primed cultures. Since MVMp target populations (B1 and MDC) are known to produce IP-10, but not IFN-, these changes in cytokine expression might be due to direct and indirect viral impact. It therefore appeared that MVMp infection, although affecting a minor fraction of immunocytes, may still have a milieu-dependent impact on the general immune status of the host.
DISCUSSION
Despite growing interest in the use of oncolytic parvovirus in cancer therapy, there is little information about the in vivo distribution of MVMp and the possible impact of therapeutic doses of this virus on the immune status of the host (22). The present study shows that the previously observed extratumoral expression of MVMp in DLN of hemangiosarcoma-bearing mice is a tumor model-independent feature of MVMp and can be traced back to two distinct immunocyte types: i.e., myeloid dendritic cells and B1 lymphocytes. As discussed previously, MVMp is considered to be strictly fibrotropic, in contrast with its close allotropic relative MVMi, which productively infects lymphocyte cells (32). The allotropic determinants were assigned to distinct amino acid substitutions within the capsid proteins, mapping at positions 317 and 321 (2). To rule out that the MVM-driven gene expression detected in lymphoid tissues was caused by contaminating MVMi virions, our virus stock was verified by sequencing and bioassay. Sequence analysis (performed by the QIAGEN Sequencing Services, Hilden, Germany) failed to reveal the above-mentioned MVMi-specific VP changes. The bioassay, involving detection of the cytopathogenic effect (CPE) in the MVMi-permissive T-cell lymphoma EL-4 (32), showed no signs of CPE after three passages of the cultured EL-4 cells inoculated with our MVMp stock at MOIs ranging from 0.1 to 100 PFU/ml. These assays confirm that tropism for MDC and B1 cells is a characteristic of MVMp. It was reported that under restrictive cell culture conditions, the MVM life cycle gets blocked at an early step, which is limiting for the onset of virus DNA transcription and amplification (12, 32). MVMp may have a similar barrier in most lymphoid cells (but MDC and B1 lymphocytes) in vivo, since these two cell types were the only hematopoietic cell populations in which MVMp replication and expression could be detected. In agreement with the in vivo data, cultures of activated immunocytes were unable to produce MVMp progeny viruses and did not show very significant growth alterations or cytopathic effects after infection, as expected from the overgrowth of the majority of nonpermissive cells. Further investigations are therefore required to determine whether MVMp DNA replication and expression in MDC and B1 lymphocytes result in a productive and/or lytic infection.
Brownstein et al. (5) have previously reported that upon infection in vivo, MVMp could be found in cells located in mesenteric lymph nodes of mice. The nuclear pattern of viral imprints suggested that these cells were sites of virus replication rather than sequestration. Furthermore, MVMp targets were described as lymphoid by their morphological appearance. On the basis of these observations and our data, we first speculated that the enhanced MVMp-driven gene expression detected in LN of tumor-bearing mice was due to the cancer-associated proliferation of lymphocytes (MAIDS and hemangiosarcoma) or precursor hematopoietic cells (normoblasts in melanoma-bearing mice) in these organs (15, 25; unpublished observation). However, FACS experiments did not support this hypothesis. Expanded T and conventional B (B2) cells isolated from lymphoid organs of tumor-bearing MVMp-infected mice were negative for parvovirus transcription. Moreover, MAIDS mice, which have the most dramatic increase of B2 cells, displayed viral transcript levels comparable to those in tumor-free B6 mice. This phenomenon was rather surprising and could indicate that either the hypothetical viral target population was not expanding or the activation/transformation status of MAIDS lymphocytes was not sufficient to support viral expression (e.g., this ability of lymphocytes would depend on the way in which activation and proliferation are induced). Puzzling at first, this result could be presently explained by the discovered affinity of MVMp for MDC and B1 lymphocytes in lymphoid tissues and the previously reported decline of dendritic cells and nonexpansion of B1 lymphocytes in contrast to the vast proliferation of T and B2 cells in MAIDS-infected animals (18, 23).
A possible modulation of B1 cells by parvoviruses and recombinant derivatives is certainly worth considering, given the increasing evidence pointing to the intercalation of immune and anticancer reactions (11). Production of autoreactive antibodies in hemolytic anemia and development of systemic lupus erythematosus in mice were ascribed to B1 cell activity (17, 26). Besides the possible role in mediating antitumor reactions, B1 cells may themselves be targets for malignant transformation and are thought to participate in the pathogenesis of leukemic disorders (6). The proneness of B1 cells to transformation was linked to the fact that they represent the only self-renewing lymphocyte population and constitutively express activated STAT3, a pro-oncogenic member of the STAT family (20). Probably, changes caused by MVMp infection may account for some of the previously observed antileukemic effects of parvoviruses (33). On the other hand, the interaction of parvoviruses with dendritic cells is not without precedent. Dendritic cells were recently reported to be able to take up and process porcine parvovirus-like particles, resulting in viral antigen presentation (24). It was not determined, however, whether this processing concerned input viruses or de novo-synthesized viral products. The presence of MVMp transcripts in MDC may be indicative of the capacity of these cells for sustaining the virus uptake and replication steps required for the onset of MVMp gene transcription. Clinical consequences of MDC infection by MVMp are not clear yet. However, they may be far-reaching, assuming the role of these cells in immunologic surveillance. To this end, recent data describing impact of deregulated expression of certain molecules located on the surface of MDC for the anticancer immune response are of special interest. For example, B7-H1-mediated suppression of dendritic cell function in tissues and DLN of ovarian cancer patients was shown to contribute to the inhibition of immune responses and disease progression. The up-regulation of B7-H1 on MDCs in the tumor microenvironment down-regulated T-cell immunity, and the blockade of B7-H1 enhanced MDC-mediated T-cell activation and was accompanied by down-regulation of T-cell IL-10 and up-regulation of IL-2 and IFN- (7). T cells conditioned with the B7-H1-blocked MDCs had a more potent ability to inhibit autologous human ovarian carcinoma growth in nonobese diabetic-severe combined-immunodeficient (NOD-SCID) mice. It is possible that such an alteration of the surface B7-H1 expression by MVMp-infected MDC would optimize an ongoing antitumor reaction in parvovirus-exposed tumor-bearing mice and at least partially explain the T-cell-dependent character of MVMp-induced oncosuppression.
Infection of MDC and B1 cells by MVMp was not accompanied by any detectable toxicity in host animals, according to clinical observations and histopathological examinations. In addition, no cytopathic effects were observed in immunocyte cultures infected in vitro with MVMp even at high MOIs (100 PFU/cell), as determined by lactate dehydrogenase and MTT assays. Considering that B1 and dendritic cells represent a very minor fraction of the total immunocyte population, a toxic effect of MVMp on these cells may still have been overlooked. No dramatic phenotypic changes were detected in the immune status of affected animals, according to FACS and cytokine profile analysis. Yet, some subtle alterations were induced as a result of the infection of mice with MVMp, as exemplified by a marked increase in the chemokine IP-10. Since IP-10 may be produced by MVMp target cells (34), the induction of these chemokines may be a direct consequence of the infection of these cells with MVMp. On the other hand, IP-10 is a chemoattractant and activator of lymphocytes (10). Hence the direct impact of MVMp on B1 and MDCs may be accompanied by indirect effects on other immune cells, as mediated by IP-10 and other chemokines produced by the former cells upon infection. In this regard, it is worth noting that in vitro infection with MVMp induces LPS-activated immunocyte cultures to release IFN-. IFN- is a marker of activated CTL and NK cells (i.e., cells which are not direct targets for MVMp gene expression). Hence, the induction of this cytokine is likely to occur through an indirect process. Knowing that IP-10 is endowed with both antitumoral and antiangiogenic properties (1, 10) and that IFN- is an antitumor molecule (19), the MVMp-mediated direct or indirect induction of these and related cytokines represents a candidate mechanism contributing to the antineoplastic activity of parvoviruses. This possibility is supported by our recent work showing that recombinant parvoviruses supplemented with distinct cytokine transgenes are indeed endowed with an enhanced oncosuppressive capacity, compared with parental viruses (27). Further work is, however, required to demonstrate the role and assess the relative contribution of immunomodulating cytokine induction in the overall ability of parvoviruses to inhibit oncogenesis. Expression of recombinant cytokines transduced by parvovirus vectors was shown to be accompanied by infiltration of tumors with immune cells (35). This effect was assumed to be responsible for the enhanced oncosuppressive activity of these vectors and to be due to their oncotropism leading to intratumoral transgene expression. Although this is likely to be true in part, the present study suggests that transgene expression in lymphoid tissues may play a significant and possibly even more important role in the potentiation of the anticancer activity of chemokine-transducing parvovirus vectors. In combination with the possibility discussed above of MVMp improving the T-cell priming ability of MDC through modulation of surface molecules (B7-H1), these observations may indicate that the infection of MDC with transgene (IP-10)-expressing MVMp recombinants could result in the more efficient attraction of tumor-specific lymphocytes to the de-blocked dendritic cells within DLN and therefore represent a new approach for cancer immunotherapy.
In summary, this study revealed a novel facet of MVMp-host interactions: namely, the selective ability of two rare lymphoid subpopulations, MDC and B1 lymphocytes, to sustain parvovirus-driven gene expression in tumor-bearing animals. Since autonomous parvoviruses are contemplated for use as oncolytic nonpathogenic therapeutic treatments against cancer, the present data indicate that, besides tumor cells, MDC and B1 lymphocytes need to be further investigated for the outcome of parvovirus infection. Intensive analysis of molecular changes caused by MVMp in target populations should lead to better understanding of the immune component of parvovirus-induced oncosuppression, guide the choice of transgenes to be inserted into parvovirus vectors in order to enhance their anticancer potency, and improve the risk-versus-benefit assessment of wild-type and recombinant parvoviruses with respect to specific tumor types.
ACKNOWLEDGMENTS
This work was supported by the Biotechnology program of the European Union. Zahari Raykov is recipient of a DKFZ doctoral fellowship.
REFERENCES
Angiolillo, A. L., C. Sgadari, and G. Tosato. 1996. A role for the interferon-inducible protein 10 in inhibition of angiogenesis by interleukin-12. Ann. N. Y. Acad. Sci. 795:158-167.
Ball-Goodrich, L. J., and P. Tattersall. 1992. Two amino acid substitutions within the capsid are coordinately required for acquisition of fibrotropism by the lymphotropic strain of minute virus of mice. J. Virol. 66:3415-3423.
Barker, P. E., L. Savelyeva, and M. Schwab. 1993. Translocation junctions cluster at the distal short arm of chromosome 1 (1p36.1-2) in human neuroblastoma cells. Oncogene 8:3353-3358.
Bottazzi, B., S. Walter, D. Govoni, F. Colotta, and A. Mantovani. 1992. Monocyte chemotactic cytokine gene transfer modulates macrophage infiltration, growth, and susceptibility to IL-2 therapy of a murine melanoma. J. Immunol. 148:1280-1285.
Brownstein, D. G., A. L. Smith, E. A. Johnson, D. J. Pintel, L. K. Naeger, and P. Tattersall. 1992. The pathogenesis of infection with minute virus of mice depends on expression of the small nonstructural protein NS2 and on the genotype of the allotropic determinants VP1 and VP2. J. Virol. 66:3118-3124.
Chiorazzi, N., and M. Ferrarini. 2003. B cell chronic lymphocytic leukemia: lessons learned from studies of the B cell antigen receptor. Annu. Rev. Immunol. 21:841-894.
Curiel, T. J., S. Wei, H. Dong, X. Alvarez, P. Cheng, P. Mottram, R. Krzysiek, K. L. Knutson, B. Daniel, M. C. Zimmermann, O. David, M. Burow, A. Gordon, N. Dhurandhar, L. Myers, R. Berggren, A. Hemminki, R. D. Alvarez, D. Emilie, D. T. Curiel, L. Chen, and W. Zou. 2003. Blockade of B7-H1 improves myeloid dendritic cell-mediated antitumor immunity. Nat. Med. 9:562-567.
Dupont, F., L. Tenenbaum, L.-P. Guo, P. Spegelaere, M. Zeicher, and J. Rommelaere. 1994. Use of an autonomous parvovirus vector for selective transfer of a foreign gene into transformed human cells of different tissue origins and its expression therein. J. Virol. 68:1397-1406.
Faisst, S., D. Guittard, A. Benner, J. Y. Cesbron, J. R. Schlehofer, J. Rommelaere, and T. Dupressoir. 1998. Dose-dependent regression of HeLa cell-derived tumours in SCID mice after parvovirus H-1 infection. Int. J. Cancer 75:584-589.
Farber, J. M. 1997. Mig and IP-10: CXC chemokines that target lymphocytes. J. Leukoc. Biol. 61:246-257.
Finn, O. J. 2003. Cancer vaccines: between the idea and the reality. Nat. Rev. Immunol. 3:630-641.
Gardiner, E. M., and P. Tattersall. 1988. Evidence that developmentally regulated control of gene expression by a parvoviral allotropic determinant is particle mediated. J. Virol. 62:1713-1722.
Garlanda, C., C. Parravicini, M. Sironi, M. De Rossi, R. Wainstok de Calmanovici, F. Carozzi, F. Bussolino, F. Colotta, A. Mantovani, and A. Vecchi. 1994. Progressive growth in immunodeficient mice and host cell recruitment by mouse endothelial cells transformed by polyoma middle-sized T antigen: implications for the pathogenesis of opportunistic vascular tumors. Proc. Natl. Acad. Sci. USA 91:7291-7295.
Giese, N. A., R. T. Gazzinelli, J. K. Actor, R. A. Morawetz, M. Sarzotti, and H. C. Morse III. 1996. Retrovirus-elicited interleukin-12 and tumour necrosis factor-alpha as inducers of interferon-gamma-mediated pathology in mouse AIDS. Immunology 87:467-474.
Giese, N. A., Z. Raykov, L. DeMartino, A. Vecchi, S. Sozzani, C. Dinsart, J. J. Cornelis, and J. Rommelaere. 2002. Suppression of metastatic hemangiosarcoma by a parvovirus MVMp vector transducing the IP-10 chemokine into immunocompetent mice. Cancer Gene Ther. 9:432-442.
Giese, T. 2001. Development of quantitative RT-PCR tests for the expression of cytokine genes on the Light Cycler, p. 251-261. In S. Meuer, C. Wittwer, and K. Nakagawara (ed.), Rapid cycle real time PCR: methods and applications. Springer, Berlin, Germany.
Hirose, S., M. Wakiya, Y. Kawano-Nishi, J. Yi, R. Sanokawa, S. Taki, T. Shimamura, T. Kishimoto, H. Tsurui, and H. Nishimura. 1993. Somatic diversification and affinity maturation of IgM and IgG anti-DNA antibodies in murine lupus. Eur. J. Immunol. 23:2813-2820.
Iciek, L. A., T. J. Waldschmidt, M. M. Griffiths, and K. H. Brooks. 1994. B-1 cells in systemic autoimmune responses: IgM+, Fc epsilon Rdull B cells are lost during chronic graft-versus-host disease but not in murine AIDS or collagen-induced arthritis. Immunol. Investig. 23:293-311.
Ikeda, H., L. J. Old, and R. D. Schreiber. 2002. The roles of IFN gamma in protection against tumor development and cancer immunoediting. Cytokine Growth Factor Rev. 13:95-109.
Karras, J. G., R. A. McKay, T. Lu, J. Pych, D. A. Frank, T. L. Rothstein, and B. P. Monia. 2000. STAT3 regulates the growth and immunoglobulin production of BCL(1) B cell lymphoma through control of cell cycle progression. Cell Immunol. 202:124-135.
Kestler, J., B. Neeb, S. Struyf, J. Van Damme, S. F. Cotmore, A. D'Abramo, P. Tattersall, J. Rommelaere, C. Dinsart, and J. J. Cornelis. 1999. cis requirements for the efficient production of recombinant DNA vectors based on autonomous parvoviruses. Hum. Gene Ther. 10:1619-1632.
Kimsey, P. B., H. D. Engers, B. Hirt, and C. V. Jongeneel. 1986. Pathogenicity of fibroblast- and lymphocyte-specific variants of minute virus of mice. J. Virol. 59:8-13.
Masuda, A., G. F. Burton, A. K. Szakal, and J. G. Tew. 1995. Loss of follicular dendritic cells in murine-acquired immunodeficiency syndrome. Lab. Investig. 73:511-520.
Moron, G., P. Rueda, I. Casal, and C. Leclerc. 2002. CD8alpha– CD11b+ dendritic cells present exogenous virus-like particles to CD8+ T cells and subsequently express CD8 alpha and CD205 molecules. J. Exp. Med. 195:1233-1245.
Morse, H. C., III, R.A. Morawetz, N. A. Giese, S. K. Chattopadhyay, Y. Tang, T. N. Fredrickson, and J. W. Hartley. 1997. Immunologic havoc induced by the 60 kD Gag protein of the MAIDS defective virus. Leukemia 11:167-169.
Okamoto, M., M. Murakami, A. Shimizu, S. Ozaki, T. Tsubata, S. Kumagai, and T. Honjo. 1992. A transgenic model of autoimmune hemolytic anemia. J. Exp. Med. 175:71-79.
Olijslagers, S., A. Y. Dege, C. Dinsart, M. Voorhoeve, J. Rommelaere, M. H. Noteborn, and J. J. Cornelis. 2001. Potentiation of a recombinant oncolytic parvovirus by expression of apoptin. Cancer Gene Ther. 8:958-965.
Russell, S. J., A. Brandenburger, C. L. Flemming, M. K. L. Collins, and J. Rommelaere. 1992. Transformation-dependent expression of interleukin genes delivered by a recombinant parvovirus. J. Virol. 66:2821-2828.
Sahli, R., G. K. McMaster, and B. Hirt. 1985. DNA sequence comparison between two tissue-specific variants of the autonomous parvovirus, minute virus of mice. Nucleic Acids Res. 13:3617-3633.
Spalholz, B. A., and P. Tattersall. 1983. Interaction of minute virus of mice with differentiated cells: strain-dependent target cell specificity is mediated by intracellular factors. J. Virol. 46:937-943.
Tattersall, P. 1972. Replication of the parvovirus MVM. I. Dependence of virus multiplication and plaque formation on cell growth. J. Virol. 10:586-590.
Tattersall, P., and J. Bratton. 1983. Reciprocal productive and restrictive virus-cell interactions of immunosuppressive and prototype strains of minute virus of mice. J. Virol. 46:944-955.
Telerman, A., M. Tuynder, T. Dupressoir, B. Robaye, F. Sigaux, E. Shaulian, M. Oren, J. Rommelaere, and R. Amson. 1993. A model for tumor suppression using H-1 parvovirus. Proc. Natl. Acad. Sci. USA 90:8702-8706.
Weighardt, H., G. Jusek, J. Mages, R. Lang, K. Hoebe, B. Beutler, and B. Holzmann. 2004. Identification of a TLR4- and TRIF-dependent activation program of dendritic cells. Eur. J. Immunol. 34:558-564.
Wetzel, K., P. Menten, G. Opdenakker, J. Van Damme, H. J. Grone, N. Giese, A. Vecchi, S. Sozzani, J. J. Cornelis, J. Rommelaere, and C. Dinsart. 2001. Transduction of human MCP-3 by a parvoviral vector induces leukocyte infiltration and reduces growth of human cervical carcinoma cell xenografts. J. Gene Med. 3:326-337.(Zahari Raykov, Larissa Sa)
Program of Functional and Structural Genomic Research, Deutsches Krebsforschungszentrum
Institute of Immunology, University of Heidelberg, Heidelberg, Germany
The Stephan Angeloff Institute of Microbiology, Bulgarian Academy of Sciences, Sofia, Bulgaria
Institut de Recherche contre les Cancers de l'Appareil Digestif (IRCAD) INSERM U375, Strasbourg, France
ABSTRACT
Due to their oncolytic properties and apathogenicity, autonomous parvoviruses have attracted significant interest as possible anticancer agents. Recent preclinical studies provided evidence of the therapeutic potential of minute virus of mice prototype strain (MVMp) and its recombinant derivatives. In a murine model of hemangiosarcoma, positive therapeutic outcome correlated with high intratumoral expression of MVMp-encoded genes in tumors and lymphoid organs, especially in tumor-draining lymph nodes. The source and relevance of this extratumoral expression, which came as a surprise because of the known fibrotropism of MVMp, remained unclear. In the present study, we investigated (i) whether the observed expression pattern occurs in different tumor models, (ii) which cell population is targeted by the virus, and (iii) the immunological consequences of this infection. Significant MVMp gene expression was detected in lymphoid tissues from infected tumor-free as well as melanoma-, lymphoma-, and hemangiosarcoma-bearing mice. This expression was especially marked in lymph nodes draining virus-injected tumors. Fluorescent in situ hybridization analysis, multicolor fluorescence-activated cell sorting, and quantitative reverse transcription-PCR revealed that MVMp was expressed in rare subpopulations of CD11b (Mac1)-positive cells displaying CD11c+ (myeloid dendritic cells [MDC]) or CD45B (B220+ [B1 lymphocytes]) markers. Apart from the late deletion of cytotoxic memory cells (CD8+ CD44+ CD62L–), this infection did not lead to significant alteration of the immunological profile of cells populating lymphoid organs. However, subtle changes were detected in the production of specific proinflammatory cytokines in lymph nodes from virus-treated animals. Considering the role of B1 lymphocytes and MDC in cancer and immunological surveillance, the specific ability of these cell types to sustain parvovirus-driven gene expression may be exploited in gene therapy protocols.
INTRODUCTION
Over the last few years, naturally occurring or genetically manipulated oncolytic parvoviruses attracted significant attention as means to treat cancer. Autonomous parvoviruses are small (18 to 26 nm) nonenveloped nucleus-replicating viruses. Their capsid contains a single-stranded DNA genome of about 5,000 nucleotides. The prototype strain of minute virus of mice, MVMp, is the prototype strain of autonomous parvoviruses and one of the best-studied members to date. MVMp and the closely related parvovirus H-1 have emerged as the most suitable therapeutic agents, due to their oncotropism, infectiousness for tumor cells of both human and rodent origin, and good safety profile (9, 22, 27, 28). The mechanism of parvovirus-induced tumor suppression is not fully understood yet. Besides previously described direct oncolytic effects, the recent preclinical experimental data reporting successful treatment of hemangiosarcoma by MVMp parvovirus and its recombinant IP-10-expressing derivative revealed the role of the immune system in the strong therapeutic effect of MVMp (15). Using T-cell-deficient mice (C57BL/6 nu+/nu+) as recipients for polyoma virus middle T antigen (PymT)-transformed endothelial cell implants (termed H5V cells), we showed that animals inoculated with MVMp-infected cells did develop tumors more slowly than the ones receiving mock-treated H5V cells. However, the inhibition was of short duration: all mice in both groups eventually developed tumors and metastases and succumbed to the disease. This contrasted with the long-term antitumor protection given by MVMp in immunocompetent animals under similar conditions. Nevertheless, MVMp alone was not successful in the treatment of preestablished tumors, which required reinforcement of the virus intrinsic activity by IP-10. In this system, MVMp was thus able by itself to delay tumor development, probably due to its oncotoxic activity, yet some interaction of the virus (and harbored transgene) with the host immune system appeared to be crucial for the eventual cure of the disease. In the course of this work, MVMp expression was found to preferentially occur not only in the tumor, as expected from the oncotropic character of the virus, but also in tumor-draining lymph nodes (DLN). This came as a surprise considering the fact that the prototype and the immunosupressive strain of MVM (MVMi) are shown to be restrictive for growth in each other's host cell: e.g., T lymphocytes and fibroblasts, respectively (32). MVMi and MVMp have the same genetic organization and share 96% of their sequences (29), with the target cell specificity of the respective viruses being defined by distinct amino acid substitutions within the capsid proteins (2). However, competition experiments have shown that both viruses recognize the same receptor on each cell type, and penetration and uncoating seem not to be the restrictive steps (30). The detection of significant MVMp-driven gene expression in tumor-draining lymph nodes suggested that this barrier may not apply to all lymphoid cells and that the viral cycle may progress at least to the level of transcription in a subset of these cells. With the view of using MVMp and related parvoviruses in the fight against cancer, the interaction of these agents with lymphoid cells represents a crucial issue regarding both safety and possible immunomodulating effects that may interfere with tumor formation. This prompted us to further investigate this question in the present study by determining the cellular origin of MVMp expression in lymphoid tissues. The data obtained confirm that draining lymph nodes are a major site of MVMp expression following virus injection into various tumor grafts in recipient mice and trace this expression back to distinct subtypes of lymphoid cells.
MATERIALS AND METHODS
Cells. The mouse cell lines H5V (13) and B78/H1 (a weakly immunogenic amelanotic subclone of the murine B16 melanoma cell line) (4) were kindly provided by S. Sozzani (IRF Negri, Milan, Italy). The A9 (fibrosarcoma) cell line was obtained from the American Type Culture Collection (Manassas, Va.). All cell lines were maintained in Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 2 mM L-glutamine, 100-μg/ml penicillin, 100-U/ml streptomycin, and 10% (H5V and B78) or 5% (A9) fetal bovine serum (FBS).
Lymphocyte single-cell suspensions were prepared from spleen and lymph nodes taken from C57BL/6 mice using standard techniques. After resuspension in RPMI-1640 containing 10% FBS, nonessential amino acids, and 5 μM 2-mercaptoethanol, 5 x 106 cells/well were plated into 24-well plates and exposed to various stimuli: lipopolysaccharide (LPS; 5 μg/ml), concanavalin A (ConA; 5 μg/ml), or a combination of phorbol myristate acetate (PMA; 0.5 ng/ml) with ionomycin (0.5 μg/ml) (all purchased from Sigma (Munich, Germany). Three hours later, cells were inoculated with MVMp and harvested at different times postinfection, depending on further analysis.
Viruses. MVMp was produced and titrated by plaque assay in A9 indicator cells, as described previously (31). Virus titers are expressed in PFU per milliliter. MVMp DNA replication and protein expression in infected cell cultures were assessed by Southern and Western blotting, respectively. For Western blotting, MVMp-infected monolayers were harvested, washed, resuspended in phosphate-buffered saline (PBS), and lysed by three freeze/thaw cycles 2 or 48 h postinfection (hpi). Ten micrograms of total protein was fractionated by sodium dodecyl sulfate-8% polyacrylamide gel electrophoresis, blotted onto nitrocellulose membranes (Amersham Pharmacia Biothech Europe, Freiburg, Germany), and probed with the NS1-specific antiserum (Sp7) or VP1 and VP2-specific rabbit monoclonal antibodies (21). Modified Hirt's extraction of parvoviral DNA and subsequent Southern blotting analysis of viral replicative forms were performed as described previously (8).
Murine leukemia retrovirus (MuLV clone LP-BM5mix) was received as a gift from H. C. Morse III (National Institutes of Health, Bethesda, Md.) (25) and used to induce mouse AIDS (MAIDS) as described below.
Mice and tumor models. For all tumor models, 6- to 8-week-old female C57BL/6 mice were purchased from CRL (Sulzfeld, Germany). Hemangiosarcoma and melanoma tumors were induced by subcutaneous (s.c.) implantation of 5 x 106 H5V or 2 x 105 B78 cells in the right flanks of syngeneic C57BL/6 mice, as described previously (4). Lymphoproliferative disorder was induced by intraperitoneal (i.p.) injection of 100 μl of MuLV according to established protocols (14, 18, 25). Starting 2 weeks postinfection, MuLV-injected B6 mice display progressive immunodeficiency and lymphoproliferation ending in development of lymphomas, a syndrome called MAIDS.
MVMp was applied per animal carrying an established tumor at 5 (H5V) or 14 (B78) days postimplantation or 3 weeks postinjection of LP-BM5mix. Parvoviral infections were performed via intratumoral (i.t.) or extratumoral (i.p.) route. Since MAIDS mice do not display early visible subcutaneous tumors, the i.t. route was imitated by injection of MVMp into palpable enlarged lymph node. For an expression analysis, mice were given three daily injections of 109 PFU/day and sacrificed 24 h later. For long-term immunologic studies, mice were either exposed to the 3-week-long parvoviral treatment or injected with MVMp-infected tumor cells and sacrificed once moribund (15).
Mice were kept in laminar flow safety cabinets (BDK Luft- und Reinraumtechnik, Sonnenbühl, Germany). Procedures for animal handling, treatment, and care were performed in compliance with European guidelines.
Analysis of parvovirus expression by RT-PCR. Total RNA was extracted from isolated tissues (kept at –80°C) or cultured cells by using Trizol reagent (Life Technologies) according to the manufacturer's instructions. DNase I-treated RNA was converted to cDNA and analyzed according to reverse transcription-PCR (RT-PCR) protocols detailed previously (14). The following PCR primers were used: MVMp, 5'-ACGCTCACCATTCACGACACCGAAA-3' and 5'-ATCATAGGCCTCGTCGTGCTCTTTG-3' (415-bp-long PCR product); PymT, 5'-TTCTGAGCAACCCGACCTAT-3' and CTTCTTAGGTGGCGTTGCAT-3 (439 bp), with probe 5'-GCCACTCCTATCCCCCAACCC-3'; and ?-actin, 5'-ATGTTTGAGACCTTCAACAC-3' and 5'-ACGTCACACTTCATGATCC-3' (489 bp). PCR products were analyzed by 2% gel electrophoresis and visualized, depending on the level of expression, either by direct staining of the gels with ethidium bromide or SybrGreen II (Roche Diagnostics, Mannheim, Germany) or by blotting on Hybond N+ membranes and hybridization with fluorescein-labeled DNA probe. Enhanced chemiluminescence (ECL) detection was performed according to the manufacturer's instructions (Amersham Pharmacia Biothech Europe, Freiburg, Germany). The intensity of PCR signals was measured, and levels of expression were normalized with ?-actin as a reference for matching the cDNA input for PCR (14). In some experiments, viral transcripts were quantified by using LightCycler-based real-time quantitative RT-PCR (QRT-PCR) technology (Roche Diagnostics) (16).
FISH. Fluorescent in situ hybridization (FISH) of DNA was performed as previously described (3). Briefly, lymphocytes were pelleted on slides (Cytospin 2; Shandon, United Kingdom), air dried, and incubated for 12 h at 60°C. After RNase A and pepsin treatment, the slides were fixed with formaldehyde, treated with increasing concentrations of ethanol, and denatured in 70% formamide. An NS1 sequence-specific probe was nick labeled with biotin-14-dCTP (Invitrogen) and hybridized with the DNA on the slides overnight at 37°C. After washing and blocking with 10% bovine serum albumin, slides were exposed to streptavidin-fluorescein isothiocyanate (FITC) complex. The nucleus of the cells was visualized by DAPI (4',6'-diamidino-2-phenylindole) staining. The images were captured with a Hamamatsu digital camera using Openlab 2.2.5 software (Improvision, Ltd.).
Analysis of cytokine expression. Cytokine expression was evaluated at the RNA level by QRT-PCR (15). Murine IP-10, gamma interferon (IFN-), interleukin-10 (IL-10), and IL-12 p40-specific primers were obtained from Search-LC (Heidelberg, Germany). Protein levels were measured with previously established enzyme-linked immunosorbent assays (ELISAs) (14).
FACS analysis and cell purification. To determine the immunological status of infected mice, three-color staining of cells derived from spleen and lymph nodes was performed and analyzed by fluorescence-activated cell sorting (FACS) with the FACS DIVA (Becton Dickinson, Heidelberg, Germany). FITC-, phycoerythrin-, or biotin-labeled antibodies to T-cell receptor /? (TCR/?), CD4, CD8, ThB, CD45 (B220), Dx5 (NK 1.1), CD11b (Mac1), CD11c, CD32, CD43, CD44, CD62L, CD69, IFN-, immunoglobulin M (IgM), and IgG were from Becton Dickinson. Avidin-allophycocyanin complex was obtained from Molecular Probes Europe BV (Leiden, Netherlands). Isolation of cell populations from lymphoid organs was achieved by two- or three-color FACS of accordingly stained single-cell suspensions. Usually, 2 x 107 to 4 x 107 cells underwent the sorting procedure and purified populations were collected by centrifugation, reanalyzed for purity, suspended in lysis buffer, and processed for automatic mRNA isolation with a MagNA Pure mRNA preparation kit I (Roche), followed by QRT-PCR analysis of viral transcripts (16).
RESULTS
Tissue distribution pattern of MVMp transcripts in different tumor models. Several mouse tumor models (B78 melanoma, MAIDS-associated lymphoma, and H5V hemangiosarcoma), as well as tumor-free B6 mice, were compared in terms of parvovirus expression. Once tumors became palpable (14 days for melanoma, 4 days for hemangiosarcoma, and 3 weeks for MAIDS), animals received three daily injections of 109 PFU of MVMp via the extratumoral (i.p.) or intratumoral (i.t.) route. One day after injection, brain, thymus, lungs, kidney, intestine, liver, spleen, draining or distant lymph nodes, and tumor (when applicable) were isolated and examined by RT-PCR for the presence of viral transcripts. While the majority of organs displayed no or sporadic viral expression (data not shown), the spleen, liver, lymph nodes, and tumor represented the sites of the most prominent and consistent MVMp expression (Fig. 1A). The levels of MVMp transcripts in these lymphoid tissues were comparable after extratumoral (i.p.) MVMp inoculation and similar to the ones detected in tumor-free mice (Fig. 1A, upper panel). In contrast, the intratumoral injections of MVMp led to preferential accumulation of viral transcripts in the lymph nodes draining the inoculated tumor (DLN), which was particularly striking with hemangiosarcoma and melanoma models (Fig. 1A, lower panel). Interestingly, despite the strong lymphoproliferative reaction occurring in MAIDS, the level of lymphoid viral transcription in this model was not preferentially increased (Fig. 1A). Viral transcripts declined to nondetectable levels within 2 weeks postinoculation, although traces of viral DNA were measurable by PCR up to 1 month later (data not shown).
These data suggested that MVMp was expressed either in tumor cells metastasized to DLN or in resident lymphoid cells. Since H5V tumor cells can be easily identified with PymT as a marker, the pattern of PymT expression was evaluated in MVMp-expressing organs of hemangiosarcoma-bearing mice. As seen in Fig. 1B, tumors showed strong accumulation of PymT, whereas this marker was barely detectable in spleens and not detectable in DLN even after PCR-ECL amplification of the signal. The lack of extratumoral PymT transcripts expression in DLN that carry the highest number of viral transcripts led us to conclude that tumor cells were not the source of viral expression in these organs. A similar conclusion would apply for the melanoma model, knowing that B78/H1 melanoma cells are not metastasizing upon subcutaneous implantation. It therefore appeared likely that the extratumoral expression of MVMp was supported by some of the lymph node cells themselves.
Parvovirus infection of isolated splenocytes. To confirm the ability of some cells located in lymphoid organs of the immune system (termed immunocytes in the text below) to express MVMp, in vitro experiments were carried out, in which this parvovirus was inoculated to naive or activated single-cell suspensions derived from the spleens or lymph nodes of B6 mice (multiplicity of infection [MOI] = 10 PFU/cell). Viral infection was monitored at the DNA (Southern blot), RNA (RT-PCR), protein (Western blot), and fluorescent microscopy levels. Besides wild-type MVMp, the MVMp green fluorescent protein-expressing recombinant (MVMp/GFP) was used, allowing visualization of virus-driven gene expression in individual cells. A9 fibroblastic cells, which are highly permissive for the MVMp virus, served as a positive control. As shown in Fig. 2, at least some cells derived from lymphoid organs could be targets for replication and expression of MVMp. Viral DNA replicative forms were synthesized, peaking at 48 hpi and rapidly decaying afterwards (Fig. 2A). There was thus no evidence of infection persisting with time in these cultures. Furthermore, a significant production of virus/vector transcripts (Fig. 2B) and proteins (Fig. 2C) was observed in activated immunocytes after infection with wild-type or recombinant MVMp as well as with the GFP-encoding recombinant (Fig. 2B). It is noteworthy that, even under activation conditions, only a very small fraction (less than 5%) of these cells was able to support MVMp-driven gene expression, as was apparent from the analysis of MVMp/GFP-infected cultures by fluorescent microscopy (data not shown). It should also be stated that MVMp expression could be detected in nonstimulated immunocytes, yet to a much lower level than in cultures stimulated with ConA, LPS, or PMA-ionomycin (data not shown). Interestingly, while mRNA expression was not dependent on the method of stimulation, viral proteins and replicative forms were more abundant with the stimulatory regimens indicated in Fig. 2. Notably, neither production of progeny infectious particles—shown by plaque assay—nor significant cytotoxic effect—more than 10% of growth reduction by MTT (3-[4,5-dimethylthiazol-2-yl)]-2,5-diphenyltetrazolium bromide) assay—could be detected upon MVMp infection (data not shown). Given the small proportion of susceptible immunocytes in these cultures, it cannot be ruled out, however, that a productive and/or lytic infection took place in a few cells but got masked by the overgrowth of nonpermissive cells. As a whole, the above results were reproducible with cells derived from lymph nodes or spleen. Altogether, these observations substantiated the possibility that a minor fraction of nontumor cells residing in lymphoid organs may support MVMp infection, at least up to a certain extent.
Identification of lymphoid cell subpopulations targeted by MVMp in vivo. With the objective of tracing MVMp expression in lymphoid organs back to distinct cells, lymphoid organs were collected from infected mice and used to prepare single-cell suspensions. The fraction of these cells that were amplifying viral DNA was determined by fluorescent in situ hybridizations of cytospined cells with an NS1 DNA-specific probe. The cellular nuclei were colored blue by DAPI staining, while the viral DNA was visualized through the green FITC labeling of the hybridization probe. Results obtained with DLN from H5V-bearing mice are shown in Fig. 3A. The proportion of cells replicating the viral DNA was estimated at about 1% of the whole DAPI-stained population. This is in agreement with above in vitro experiments showing that only a small population of cultured immunocytes was competent for MVMp/GFP-mediated gene transduction. Altogether the data provide evidence to suggest that MVMp replication and expression are restricted to a discrete fraction of these cells. In order to further characterize this fraction, distinct subpopulations of immunocytes were isolated by FACS sorting and individually tested by QRT-PCR for the expression of viral transcripts. A stepwise fractionation procedure was carried out. In a first step, a 95 to 98% pure population of T lymphocytes (TCR/?+ as well as separately tested CD4+ and CD8+ cells), B lymphocytes (B220+/ThB+), and Mac1+ cells were isolated from lymph nodes or spleen. As summarized in Fig. 3B (left panel), the Mac1+-enriched cell population was a major source of virus expression in DLN of H5V tumor-bearing mice (2,500 transcripts/μg of input mRNA), while viral transcripts were nondetectable in the two other cell types. Since the Mac1 (CD11b) molecule is present on macrophages, NK cells, and subsets of B and dendritic cells, the Mac1-positive population was further sorted with additional markers. These experiments showed no viral expression in NK cells but strong accumulation of transcripts in Mac1+ B220+ (1,200 copies/μg of RNA) and Mac1+ CD11c+ (500 copies/μg of RNA)-positive cells derived from analyzed DLN (Fig. 3B, right panel) of hemangiosarcoma-bearing mice. Sorting and QRT-PCR analysis of spleens obtained from MVMp-infected H5V mice and lymphoid organs derived from na?ve mice confirmed the observed pattern (data not shown). The constellation of markers defines two rather rare populations residing in peripheral lymphoid organs: myeloid dendritic cells (Mac1+ and CD11c+) and a subset of B lymphocytes, termed B1 (Mac1+ and B220+) (Fig. 3C). Interestingly, compared with na?ve animals, DLN from H5V tumor-bearing mice showed a threefold enrichment with these two cell types (Fig. 3C), whereas the absence of progressive accumulation of B1 lymphocytes and destruction of dendritic cells were previously reported in lymph nodes of MAIDS-affected mice (18, 23). This difference between both tumor models is likely to account, at least in part, for the much greater MVMp expression detected in DLN from H5V versus MAIDS-carrying animals (Fig. 1B). The predilection of MVMp expression for Mac1+ B220+ and Mac1+ CD11c+ cells defines a new and unexpected component of MVMp tropism. This property may be exploited to design vector derivatives for the selective transduction of this narrow subset of lymphoid cells in the context of gene (immuno) therapy protocols.
Impact of MVMp on the immune status of infected mice. Since two very distinct cell populations participating in diverse immune reactions were targets for MVMp in lymphoid organs, we investigated whether infected mice showed alteration of their immune status. To this end, spleens and lymph nodes from infected or mock-treated mice were collected and analyzed by FACS for the distribution of immunocyte populations (CD4+ and CD8+ T cells, B220+ B cells, Dx5+ NK cells, and Mac1+ macrophages) and their activation status: CD69 (early activation), CD44 and CD62L (activation/memory), IFN- (activated cytotoxic T lymphocytes [CTL]), and surface Igs (activated B cells). Furthermore the expression of various cytokines was monitored by QRT-PCR and ELISA (Fig. 4). The FACS analysis of lymphoid organs failed to reveal significant changes in the distribution and activation of immunocyte populations over the course of time following exposure of either naive or tumor-bearing mice to MVMp (data not shown). The only difference was in the late reduction of memory CTL (CD8+ CD44+ CD62L–) in half of the animals studied, starting from 4 weeks after infection (Fig. 4A). This phenomenon was observed in mice exposed to the live virus or to the virus-infected cells. As mentioned above, virus expression was not detectable anymore at this time point. Since T cells did not seem to belong to the population able to support viral expression in lymphoid tissues of infected mice, the memory CTL depletion was likely to occur through an indirect mechanism. We further determined whether MVMp infection of animals affected the lymphoid expression of cytokines known to be produced (IP-10, IL-10, and IL-12 p40) or not (IFN-) by the target cells (34). As shown in Fig. 4B, a striking up-regulation of IP-10 expression was observed in lymphoid tissues derived from infected B6 mice, while no main differences were detected for the other cytokines. In vitro, MVMp itself did not induce IFN- expression in lymphocyte cultures either but significantly augmented LPS-induced IFN- production (Fig. 4C). Interestingly, no such costimulation was observed in ConA and PMA-ionomycin-primed cultures. Since MVMp target populations (B1 and MDC) are known to produce IP-10, but not IFN-, these changes in cytokine expression might be due to direct and indirect viral impact. It therefore appeared that MVMp infection, although affecting a minor fraction of immunocytes, may still have a milieu-dependent impact on the general immune status of the host.
DISCUSSION
Despite growing interest in the use of oncolytic parvovirus in cancer therapy, there is little information about the in vivo distribution of MVMp and the possible impact of therapeutic doses of this virus on the immune status of the host (22). The present study shows that the previously observed extratumoral expression of MVMp in DLN of hemangiosarcoma-bearing mice is a tumor model-independent feature of MVMp and can be traced back to two distinct immunocyte types: i.e., myeloid dendritic cells and B1 lymphocytes. As discussed previously, MVMp is considered to be strictly fibrotropic, in contrast with its close allotropic relative MVMi, which productively infects lymphocyte cells (32). The allotropic determinants were assigned to distinct amino acid substitutions within the capsid proteins, mapping at positions 317 and 321 (2). To rule out that the MVM-driven gene expression detected in lymphoid tissues was caused by contaminating MVMi virions, our virus stock was verified by sequencing and bioassay. Sequence analysis (performed by the QIAGEN Sequencing Services, Hilden, Germany) failed to reveal the above-mentioned MVMi-specific VP changes. The bioassay, involving detection of the cytopathogenic effect (CPE) in the MVMi-permissive T-cell lymphoma EL-4 (32), showed no signs of CPE after three passages of the cultured EL-4 cells inoculated with our MVMp stock at MOIs ranging from 0.1 to 100 PFU/ml. These assays confirm that tropism for MDC and B1 cells is a characteristic of MVMp. It was reported that under restrictive cell culture conditions, the MVM life cycle gets blocked at an early step, which is limiting for the onset of virus DNA transcription and amplification (12, 32). MVMp may have a similar barrier in most lymphoid cells (but MDC and B1 lymphocytes) in vivo, since these two cell types were the only hematopoietic cell populations in which MVMp replication and expression could be detected. In agreement with the in vivo data, cultures of activated immunocytes were unable to produce MVMp progeny viruses and did not show very significant growth alterations or cytopathic effects after infection, as expected from the overgrowth of the majority of nonpermissive cells. Further investigations are therefore required to determine whether MVMp DNA replication and expression in MDC and B1 lymphocytes result in a productive and/or lytic infection.
Brownstein et al. (5) have previously reported that upon infection in vivo, MVMp could be found in cells located in mesenteric lymph nodes of mice. The nuclear pattern of viral imprints suggested that these cells were sites of virus replication rather than sequestration. Furthermore, MVMp targets were described as lymphoid by their morphological appearance. On the basis of these observations and our data, we first speculated that the enhanced MVMp-driven gene expression detected in LN of tumor-bearing mice was due to the cancer-associated proliferation of lymphocytes (MAIDS and hemangiosarcoma) or precursor hematopoietic cells (normoblasts in melanoma-bearing mice) in these organs (15, 25; unpublished observation). However, FACS experiments did not support this hypothesis. Expanded T and conventional B (B2) cells isolated from lymphoid organs of tumor-bearing MVMp-infected mice were negative for parvovirus transcription. Moreover, MAIDS mice, which have the most dramatic increase of B2 cells, displayed viral transcript levels comparable to those in tumor-free B6 mice. This phenomenon was rather surprising and could indicate that either the hypothetical viral target population was not expanding or the activation/transformation status of MAIDS lymphocytes was not sufficient to support viral expression (e.g., this ability of lymphocytes would depend on the way in which activation and proliferation are induced). Puzzling at first, this result could be presently explained by the discovered affinity of MVMp for MDC and B1 lymphocytes in lymphoid tissues and the previously reported decline of dendritic cells and nonexpansion of B1 lymphocytes in contrast to the vast proliferation of T and B2 cells in MAIDS-infected animals (18, 23).
A possible modulation of B1 cells by parvoviruses and recombinant derivatives is certainly worth considering, given the increasing evidence pointing to the intercalation of immune and anticancer reactions (11). Production of autoreactive antibodies in hemolytic anemia and development of systemic lupus erythematosus in mice were ascribed to B1 cell activity (17, 26). Besides the possible role in mediating antitumor reactions, B1 cells may themselves be targets for malignant transformation and are thought to participate in the pathogenesis of leukemic disorders (6). The proneness of B1 cells to transformation was linked to the fact that they represent the only self-renewing lymphocyte population and constitutively express activated STAT3, a pro-oncogenic member of the STAT family (20). Probably, changes caused by MVMp infection may account for some of the previously observed antileukemic effects of parvoviruses (33). On the other hand, the interaction of parvoviruses with dendritic cells is not without precedent. Dendritic cells were recently reported to be able to take up and process porcine parvovirus-like particles, resulting in viral antigen presentation (24). It was not determined, however, whether this processing concerned input viruses or de novo-synthesized viral products. The presence of MVMp transcripts in MDC may be indicative of the capacity of these cells for sustaining the virus uptake and replication steps required for the onset of MVMp gene transcription. Clinical consequences of MDC infection by MVMp are not clear yet. However, they may be far-reaching, assuming the role of these cells in immunologic surveillance. To this end, recent data describing impact of deregulated expression of certain molecules located on the surface of MDC for the anticancer immune response are of special interest. For example, B7-H1-mediated suppression of dendritic cell function in tissues and DLN of ovarian cancer patients was shown to contribute to the inhibition of immune responses and disease progression. The up-regulation of B7-H1 on MDCs in the tumor microenvironment down-regulated T-cell immunity, and the blockade of B7-H1 enhanced MDC-mediated T-cell activation and was accompanied by down-regulation of T-cell IL-10 and up-regulation of IL-2 and IFN- (7). T cells conditioned with the B7-H1-blocked MDCs had a more potent ability to inhibit autologous human ovarian carcinoma growth in nonobese diabetic-severe combined-immunodeficient (NOD-SCID) mice. It is possible that such an alteration of the surface B7-H1 expression by MVMp-infected MDC would optimize an ongoing antitumor reaction in parvovirus-exposed tumor-bearing mice and at least partially explain the T-cell-dependent character of MVMp-induced oncosuppression.
Infection of MDC and B1 cells by MVMp was not accompanied by any detectable toxicity in host animals, according to clinical observations and histopathological examinations. In addition, no cytopathic effects were observed in immunocyte cultures infected in vitro with MVMp even at high MOIs (100 PFU/cell), as determined by lactate dehydrogenase and MTT assays. Considering that B1 and dendritic cells represent a very minor fraction of the total immunocyte population, a toxic effect of MVMp on these cells may still have been overlooked. No dramatic phenotypic changes were detected in the immune status of affected animals, according to FACS and cytokine profile analysis. Yet, some subtle alterations were induced as a result of the infection of mice with MVMp, as exemplified by a marked increase in the chemokine IP-10. Since IP-10 may be produced by MVMp target cells (34), the induction of these chemokines may be a direct consequence of the infection of these cells with MVMp. On the other hand, IP-10 is a chemoattractant and activator of lymphocytes (10). Hence the direct impact of MVMp on B1 and MDCs may be accompanied by indirect effects on other immune cells, as mediated by IP-10 and other chemokines produced by the former cells upon infection. In this regard, it is worth noting that in vitro infection with MVMp induces LPS-activated immunocyte cultures to release IFN-. IFN- is a marker of activated CTL and NK cells (i.e., cells which are not direct targets for MVMp gene expression). Hence, the induction of this cytokine is likely to occur through an indirect process. Knowing that IP-10 is endowed with both antitumoral and antiangiogenic properties (1, 10) and that IFN- is an antitumor molecule (19), the MVMp-mediated direct or indirect induction of these and related cytokines represents a candidate mechanism contributing to the antineoplastic activity of parvoviruses. This possibility is supported by our recent work showing that recombinant parvoviruses supplemented with distinct cytokine transgenes are indeed endowed with an enhanced oncosuppressive capacity, compared with parental viruses (27). Further work is, however, required to demonstrate the role and assess the relative contribution of immunomodulating cytokine induction in the overall ability of parvoviruses to inhibit oncogenesis. Expression of recombinant cytokines transduced by parvovirus vectors was shown to be accompanied by infiltration of tumors with immune cells (35). This effect was assumed to be responsible for the enhanced oncosuppressive activity of these vectors and to be due to their oncotropism leading to intratumoral transgene expression. Although this is likely to be true in part, the present study suggests that transgene expression in lymphoid tissues may play a significant and possibly even more important role in the potentiation of the anticancer activity of chemokine-transducing parvovirus vectors. In combination with the possibility discussed above of MVMp improving the T-cell priming ability of MDC through modulation of surface molecules (B7-H1), these observations may indicate that the infection of MDC with transgene (IP-10)-expressing MVMp recombinants could result in the more efficient attraction of tumor-specific lymphocytes to the de-blocked dendritic cells within DLN and therefore represent a new approach for cancer immunotherapy.
In summary, this study revealed a novel facet of MVMp-host interactions: namely, the selective ability of two rare lymphoid subpopulations, MDC and B1 lymphocytes, to sustain parvovirus-driven gene expression in tumor-bearing animals. Since autonomous parvoviruses are contemplated for use as oncolytic nonpathogenic therapeutic treatments against cancer, the present data indicate that, besides tumor cells, MDC and B1 lymphocytes need to be further investigated for the outcome of parvovirus infection. Intensive analysis of molecular changes caused by MVMp in target populations should lead to better understanding of the immune component of parvovirus-induced oncosuppression, guide the choice of transgenes to be inserted into parvovirus vectors in order to enhance their anticancer potency, and improve the risk-versus-benefit assessment of wild-type and recombinant parvoviruses with respect to specific tumor types.
ACKNOWLEDGMENTS
This work was supported by the Biotechnology program of the European Union. Zahari Raykov is recipient of a DKFZ doctoral fellowship.
REFERENCES
Angiolillo, A. L., C. Sgadari, and G. Tosato. 1996. A role for the interferon-inducible protein 10 in inhibition of angiogenesis by interleukin-12. Ann. N. Y. Acad. Sci. 795:158-167.
Ball-Goodrich, L. J., and P. Tattersall. 1992. Two amino acid substitutions within the capsid are coordinately required for acquisition of fibrotropism by the lymphotropic strain of minute virus of mice. J. Virol. 66:3415-3423.
Barker, P. E., L. Savelyeva, and M. Schwab. 1993. Translocation junctions cluster at the distal short arm of chromosome 1 (1p36.1-2) in human neuroblastoma cells. Oncogene 8:3353-3358.
Bottazzi, B., S. Walter, D. Govoni, F. Colotta, and A. Mantovani. 1992. Monocyte chemotactic cytokine gene transfer modulates macrophage infiltration, growth, and susceptibility to IL-2 therapy of a murine melanoma. J. Immunol. 148:1280-1285.
Brownstein, D. G., A. L. Smith, E. A. Johnson, D. J. Pintel, L. K. Naeger, and P. Tattersall. 1992. The pathogenesis of infection with minute virus of mice depends on expression of the small nonstructural protein NS2 and on the genotype of the allotropic determinants VP1 and VP2. J. Virol. 66:3118-3124.
Chiorazzi, N., and M. Ferrarini. 2003. B cell chronic lymphocytic leukemia: lessons learned from studies of the B cell antigen receptor. Annu. Rev. Immunol. 21:841-894.
Curiel, T. J., S. Wei, H. Dong, X. Alvarez, P. Cheng, P. Mottram, R. Krzysiek, K. L. Knutson, B. Daniel, M. C. Zimmermann, O. David, M. Burow, A. Gordon, N. Dhurandhar, L. Myers, R. Berggren, A. Hemminki, R. D. Alvarez, D. Emilie, D. T. Curiel, L. Chen, and W. Zou. 2003. Blockade of B7-H1 improves myeloid dendritic cell-mediated antitumor immunity. Nat. Med. 9:562-567.
Dupont, F., L. Tenenbaum, L.-P. Guo, P. Spegelaere, M. Zeicher, and J. Rommelaere. 1994. Use of an autonomous parvovirus vector for selective transfer of a foreign gene into transformed human cells of different tissue origins and its expression therein. J. Virol. 68:1397-1406.
Faisst, S., D. Guittard, A. Benner, J. Y. Cesbron, J. R. Schlehofer, J. Rommelaere, and T. Dupressoir. 1998. Dose-dependent regression of HeLa cell-derived tumours in SCID mice after parvovirus H-1 infection. Int. J. Cancer 75:584-589.
Farber, J. M. 1997. Mig and IP-10: CXC chemokines that target lymphocytes. J. Leukoc. Biol. 61:246-257.
Finn, O. J. 2003. Cancer vaccines: between the idea and the reality. Nat. Rev. Immunol. 3:630-641.
Gardiner, E. M., and P. Tattersall. 1988. Evidence that developmentally regulated control of gene expression by a parvoviral allotropic determinant is particle mediated. J. Virol. 62:1713-1722.
Garlanda, C., C. Parravicini, M. Sironi, M. De Rossi, R. Wainstok de Calmanovici, F. Carozzi, F. Bussolino, F. Colotta, A. Mantovani, and A. Vecchi. 1994. Progressive growth in immunodeficient mice and host cell recruitment by mouse endothelial cells transformed by polyoma middle-sized T antigen: implications for the pathogenesis of opportunistic vascular tumors. Proc. Natl. Acad. Sci. USA 91:7291-7295.
Giese, N. A., R. T. Gazzinelli, J. K. Actor, R. A. Morawetz, M. Sarzotti, and H. C. Morse III. 1996. Retrovirus-elicited interleukin-12 and tumour necrosis factor-alpha as inducers of interferon-gamma-mediated pathology in mouse AIDS. Immunology 87:467-474.
Giese, N. A., Z. Raykov, L. DeMartino, A. Vecchi, S. Sozzani, C. Dinsart, J. J. Cornelis, and J. Rommelaere. 2002. Suppression of metastatic hemangiosarcoma by a parvovirus MVMp vector transducing the IP-10 chemokine into immunocompetent mice. Cancer Gene Ther. 9:432-442.
Giese, T. 2001. Development of quantitative RT-PCR tests for the expression of cytokine genes on the Light Cycler, p. 251-261. In S. Meuer, C. Wittwer, and K. Nakagawara (ed.), Rapid cycle real time PCR: methods and applications. Springer, Berlin, Germany.
Hirose, S., M. Wakiya, Y. Kawano-Nishi, J. Yi, R. Sanokawa, S. Taki, T. Shimamura, T. Kishimoto, H. Tsurui, and H. Nishimura. 1993. Somatic diversification and affinity maturation of IgM and IgG anti-DNA antibodies in murine lupus. Eur. J. Immunol. 23:2813-2820.
Iciek, L. A., T. J. Waldschmidt, M. M. Griffiths, and K. H. Brooks. 1994. B-1 cells in systemic autoimmune responses: IgM+, Fc epsilon Rdull B cells are lost during chronic graft-versus-host disease but not in murine AIDS or collagen-induced arthritis. Immunol. Investig. 23:293-311.
Ikeda, H., L. J. Old, and R. D. Schreiber. 2002. The roles of IFN gamma in protection against tumor development and cancer immunoediting. Cytokine Growth Factor Rev. 13:95-109.
Karras, J. G., R. A. McKay, T. Lu, J. Pych, D. A. Frank, T. L. Rothstein, and B. P. Monia. 2000. STAT3 regulates the growth and immunoglobulin production of BCL(1) B cell lymphoma through control of cell cycle progression. Cell Immunol. 202:124-135.
Kestler, J., B. Neeb, S. Struyf, J. Van Damme, S. F. Cotmore, A. D'Abramo, P. Tattersall, J. Rommelaere, C. Dinsart, and J. J. Cornelis. 1999. cis requirements for the efficient production of recombinant DNA vectors based on autonomous parvoviruses. Hum. Gene Ther. 10:1619-1632.
Kimsey, P. B., H. D. Engers, B. Hirt, and C. V. Jongeneel. 1986. Pathogenicity of fibroblast- and lymphocyte-specific variants of minute virus of mice. J. Virol. 59:8-13.
Masuda, A., G. F. Burton, A. K. Szakal, and J. G. Tew. 1995. Loss of follicular dendritic cells in murine-acquired immunodeficiency syndrome. Lab. Investig. 73:511-520.
Moron, G., P. Rueda, I. Casal, and C. Leclerc. 2002. CD8alpha– CD11b+ dendritic cells present exogenous virus-like particles to CD8+ T cells and subsequently express CD8 alpha and CD205 molecules. J. Exp. Med. 195:1233-1245.
Morse, H. C., III, R.A. Morawetz, N. A. Giese, S. K. Chattopadhyay, Y. Tang, T. N. Fredrickson, and J. W. Hartley. 1997. Immunologic havoc induced by the 60 kD Gag protein of the MAIDS defective virus. Leukemia 11:167-169.
Okamoto, M., M. Murakami, A. Shimizu, S. Ozaki, T. Tsubata, S. Kumagai, and T. Honjo. 1992. A transgenic model of autoimmune hemolytic anemia. J. Exp. Med. 175:71-79.
Olijslagers, S., A. Y. Dege, C. Dinsart, M. Voorhoeve, J. Rommelaere, M. H. Noteborn, and J. J. Cornelis. 2001. Potentiation of a recombinant oncolytic parvovirus by expression of apoptin. Cancer Gene Ther. 8:958-965.
Russell, S. J., A. Brandenburger, C. L. Flemming, M. K. L. Collins, and J. Rommelaere. 1992. Transformation-dependent expression of interleukin genes delivered by a recombinant parvovirus. J. Virol. 66:2821-2828.
Sahli, R., G. K. McMaster, and B. Hirt. 1985. DNA sequence comparison between two tissue-specific variants of the autonomous parvovirus, minute virus of mice. Nucleic Acids Res. 13:3617-3633.
Spalholz, B. A., and P. Tattersall. 1983. Interaction of minute virus of mice with differentiated cells: strain-dependent target cell specificity is mediated by intracellular factors. J. Virol. 46:937-943.
Tattersall, P. 1972. Replication of the parvovirus MVM. I. Dependence of virus multiplication and plaque formation on cell growth. J. Virol. 10:586-590.
Tattersall, P., and J. Bratton. 1983. Reciprocal productive and restrictive virus-cell interactions of immunosuppressive and prototype strains of minute virus of mice. J. Virol. 46:944-955.
Telerman, A., M. Tuynder, T. Dupressoir, B. Robaye, F. Sigaux, E. Shaulian, M. Oren, J. Rommelaere, and R. Amson. 1993. A model for tumor suppression using H-1 parvovirus. Proc. Natl. Acad. Sci. USA 90:8702-8706.
Weighardt, H., G. Jusek, J. Mages, R. Lang, K. Hoebe, B. Beutler, and B. Holzmann. 2004. Identification of a TLR4- and TRIF-dependent activation program of dendritic cells. Eur. J. Immunol. 34:558-564.
Wetzel, K., P. Menten, G. Opdenakker, J. Van Damme, H. J. Grone, N. Giese, A. Vecchi, S. Sozzani, J. J. Cornelis, J. Rommelaere, and C. Dinsart. 2001. Transduction of human MCP-3 by a parvoviral vector induces leukocyte infiltration and reduces growth of human cervical carcinoma cell xenografts. J. Gene Med. 3:326-337.(Zahari Raykov, Larissa Sa)