Propagation of Rat Parvovirus in Thymic Lymphoma Cell Line C58(NT)D and Subsequent Appearance of a Resistant Cell Clone after Lytic Infection

Abstract

Rat parvovirus (RPV) is nonpathogenic in rats but causes persistent lymphocytotropic infection. We found that RPV was propagated in rat thymic lymphoma cell line C58(NT)D and induced apoptosis. Interestingly, a resistant subclone, C58(NT)D/R, from surviving cells after lytic infection had differentiated phenotypic modifications, such as increased cell adherence, resistance to apoptosis, and suppressed tumorigenicity.

Recent molecular studies on parvoviral pathogenicity suggest that the viral nonstructural (NS) protein in which coded genes are highly homologous among parvoviruses correlates with cytotoxicity (2, 8, 17). The productive and cytotoxic activity of the NS protein is modulated by cellular factors that may vary with the host cell type, particularly in oncogene-transformed cells (2, 13). We have shown that a transcriptional coactivator, CREB binding protein, is required for NS1-mediated viral and cellular transcription in parvovirus-infected cells, resulting in cell proliferation and differentiation to achieve its lytic cycle (16).

Newly recognized rodent parvoviruses, also called orphan parvoviruses, are widespread among laboratory mice and rats (21, 23, 24), and viruses isolated have been classified as mouse parvovirus and rat parvovirus (RPV) (1, 5, 6). Although mouse parvovirus grows well in a murine T-cell clone (L3), causing a cytopathic effect (CPE) (10), effective in vitro RPV propagation has not been established. We found that RPV involved lytic infection in the rat thymic lymphoma cell line C58(NT)D and developed in vitro propagation for RPV. Interestingly, virus-resistant cell clones isolated from subcultures of surviving cells acquired differentiated phenotypes such as reduced tumorigenicity and sensitivity to apoptosis.

Viral propagation in cell lines.

The RPV strain was isolated from rats infected spontaneously at our facility (23) and passaged three times in specific-pathogen-free newborn Sprague-Dawley rats. The virus stock was prepared from infected spleens at 7 days postinoculation (p.i.) We initially inoculated a supernatant of 5% infected spleen homogenate into cell lines and primary cell cultures of rats and hamsters to find cells permitting virus propagation. C6 (rat glioma), BRL-3A (Buffalo rat liver), RBL-2H3 (rat basophilic leukemia), BHK-21 (Syrian hamster kidney), and Y3-Ag1.2.3 (rat myeloma) cells were obtained from the Riken Cell Bank, Tsukuba, Japan, and C58(NT)D (rat thymic lymphoma) cells were purchased from the American Type Culture Collection, Manassas, Va. Infected cells were observed for appearance of the viral CPE and examined for presence of the viral antigen by immunofluorescent antibody (IFA) and hemagglutination (HA) ability assays. The isolated virus was propagated only in C58(NT)D cells, not in other cells tested (Table 1). Parvoviral DNA was also detected only in infected C58(NT)D cells (data not shown). In contrast, prototype RV-13 (7) was propagated, with titers in C58(NT)D cells lower than those in other cells. Propagation of the isolate in C58(NT)D cells and a difference in cell tropism was thus clearly demonstrated between the isolate and prototype RV-13 viruses.

TABLE 1.

Propagation of the RPV isolate and prototype RV (RV-13) in several cell lines^a

Cell line	Origin	Propagation of cell line as shown by indicated assay
		RPV isolate		RV (RV-13)
		IFAb	HA titer	IFA	HA titer
C58(NT)D	Rat thymic lymphoma	++	>256	+	16
Y3-Ag1.2.3	Rat myeloma	−		−
C6	Rat glioma	−		++	>256
RBL-2H3	Rat basophilic leukemia	−		++	>256
BRL-3A	Buffalo rat liver	−		++	>256
RE	Rat embryonic fibroblast	−		++	>256
BHK-21	Hamster kidney	−		++	>256

^aThe isolated strain of RPV and prototype strain of RV were inoculated into each cell line, and viral propagation was assayed by IFA and HA. When these tests were negative, inoculated cells were passaged three times and confirmed to be consistently negative. 

^bViral antigen was demonstrated in more than 30% of the inoculated cells (++), <10% (+), or no cells (−). 

To identify the propagated virus in C58(NT)D cells, we conducted HA ability, HA inhibition, and IFA assays. The propagated virus agglutinated mouse and rat erythrocytes at HA titers of 2⁷ to 2⁹ at 4°C, but not guinea pig erythrocytes. Prototype RV, in contrast, agglutinated erythrocytes of the rat and guinea pig but not those of the mouse (data not shown). The propagated virus was serologically reacted with antisera prepared from experimentally infected rats with RV or RPV by IFA, but only reacted to anti-RPV serum by HA inhibition (data not shown). Thus, the propagated virus was identical to the original isolate strain of RPV(23). In the present study, the isolate strain, designated RPV/UT, was propagated only in the thymic lymphoma cell line, i.e. C58(NT)D cells (3), and involved persistent infection, confirming the lymphotropism of this virus in in vitro experiments.

Viral cytotoxicity and apoptosis induction to C58(NT)D cells.

To preliminarily assess viral cytotoxicity and apoptosis induction, C58(NT)D cells were infected with the RPV/UT virus at a theoretical multiplicity of infection of 2, and cell viability was measured by trypan blue staining. Infected C58(NT)D cells arrested cell growth for 2 days p.i., involved CPE, and reduced the cell number to less than 10% of that of mock-infected cells at 4 days p.i. (Fig. 1A, C, and D). Viral infectivity of the infected cell culture reached 10^5.5 50% tissue culture infective doses/100 μ1 at 2 days p.i. (Fig. 1B). Oligonucleosomal DNA ladders were clearly visible in infected C58(NT)D cells at 3 days p.i. (Fig. 1E). To confirm that apoptosis was induced in RPV-infected C58(NT)D cells, we examined the expression pattern of the Bcl-2 gene, which regulates cell survival against apoptosis in lymphocyte development and selection. Significant Bcl-2 downregulation was demonstrated in infected C58(NT)D cells at 3 days p.i. by Western blot analysis (Fig. 1F). These findings indicate that RPV/UT induces apoptosis in infected C58(NT)D cells.

FIG. 1.

Cytotoxicity and apoptosis induction of RPV in C58(NT)D cells. (A) Kinetics of cell survival in the RPV-infected (●) and the mock-infected (■) C58(NT)D cells. (B) Infectivity titers of the infected C58(NT)D cells (multiplicity of infection, 2). CPE in the infected cells (C) and the mock-infected control (D) at 4 days p.i. (E) DNA fragmentation in the RPV-infected cells (∗) and mock-infected cells (unmarked lanes). M, marker. (F) Western blot analysis showing Bcl-2 and β-actin as a control in the infected and the mock-infected cells.

Certain autonomous parvoviruses were assumed to induce apoptosis in infected cells (4, 11, 12, 15, 18, 26). Our previous study (15) demonstrated caspase-3-dependent apoptosis in H-1 virus-infected C6 cells. It was also reported that H-1 virus-induced apoptosis in U937 cells was mediated by downregulation of c-Myc oncoprotein overexpression (18). In the present study, we showed that RPV/UT mediates downregulation of Bcl-2 expression and induces apoptosis in infected C58(NT)D cells. The Bcl-2 family acts as an upstream checkpoint of caspase 3 and mitochondrial dysfunction in the apoptosis pathway and has specific roles in T-cell development and selection (14, 25). Our data suggest that RPV, like H-1 parvovirus, induces apoptosis mediated by the activation of the caspase 3 signaling pathway.

Isolation of virus-resistant cell clone C58(NT)D/R.

The CPE of the virus led to massive cell death, with surviving cells numbering less than 10% of mock-infected cells. Surviving cells proliferated continuously, accompanied by phenotypic modification such as increased cell adherence through subsequent serial passages at 3- to 4-day intervals. Changes in cell proliferation, viral infectivity of cultured fluids, and the viral antigen ratio during serial passages are summarized in Fig. 2A to F. Proliferation of surviving cells was reduced for comparison with mock-infected cells and recovered at the fifth passage. Cells positive to the viral antigen by IFA assay progressively decreased at the 4th (Fig. 2C), 8th (Fig. 2D), and 10th (Fig. 2E) passages, and finally disappeared at the 12th (Fig. 2F) and subsequent passages. The cell adherence of all surviving cells was enhanced more than that of parental C58(NT)D cells (Fig. 2G and H). The appearance of resistant subclones was repeatedly confirmed, and 10 resistant subclones were derived from surviving cells isolated by limiting dilution. All resistant subclones had the same morphological properties and resistance to RPV infection (data not shown). Neither the viral antigen nor viral DNA was detected in them by IFA assay or PCR, indicating that surviving cells completely eliminated the virus. One of the subclones, C58(NT)D/R, was observed on the cell surface structure by scanning electron microscopy (SEM). Samples were fixed with 2% glutaraldehyde and 1% osmium tetraoxide, dehydrated, and critical-point dried with carbon dioxide. They were coated with platinum and examined using SEM (JSM-6320F microscope; JEOL, Tokyo, Japan). Numerous elongated microvilli covering the surface of C58(NT)D/R cells were also observed (Fig. 2I and J).

FIG. 2.

Persistent infection of RPV in C58(NT)D cells and an appearance of resistant cells. (A) Cell growth ratio of the RPV-infected C58(NT)D cells (●) compared to uninfected cells (■) at serial cell passages. (B) Infective titers of cell fluids at the indicated cell passages. Cells with a positive reaction to the viral antigen by IFA assay were progressively decreased at the 4th (C), 8th (D), and 10th (E) passages, and finally disappeared at the 12th passage (F) or later. The cells at the 12th passage (H) showed enhanced cell adherence compared to the parental C58(NT)D cells (G). One of the resistant cell clones, C58(NT)D/R (J), was observed by SEM and compared to the C58(NT)D cells (I). Bar = 1 μm.

Phenotypic modification on C58(NT)D/R.

We studied the susceptibility to apoptosis of irradiated C58(NT)D and C58(NT)D/R cells using a cell death detection kit enzyme-linked, immunosorbent assay (Boehringer, Mannheim, Germany). X-ray irradiation at 6 or 8 Gy was done at 150 V and 20 mA, using a device (model MBR-1520R from Hitachi Medical, Japan). Interestingly, C58(NT)D/R cells involved definitive resistance to apoptosis induced by X-ray irradiation, while parental C58(NT)D cells showed enhanced DNA fragmentation, indicating apoptosis at 24 h postirradiation (Fig. 3).

FIG. 3.

Resistance of C58(NT)D/R cells to irradiation-induced apoptosis. C58(NT)D/R cells and parental C58(NT)D cells were irradiated with 6 Gy (○ and ●) or 8 Gy (▵ and ▴) of X rays, and DNA fragmentation was assayed as an indicator of apoptotic change by using the cell death detection kit enzyme-linked immunosorbent assay (Boeringer).

Since parvoviruses inhibit tumorigenesis mediated with oncoviruses and chemical carcinogens (19, 20), we compared tumorigenicity between C58(NT)D and C58(NT)D/R cells. Eight female nude mice (5 weeks old, BALB/c^−nu/nu) purchased from CLEA Japan (Tokyo, Japan) in each group were inoculated subcutaneously with 5 × 10⁵ C58(NT)D and C58(NT)D/R cells, respectively. Mice were housed in an isolator controlled at 23 ± 2°C and 55% ± 10 % relative humidity, and given free access to autoclaved food (NMF; Oriental Yeast, Tokyo, Japan) and water. They were inspected daily, and maximum and minimum diameters of tumor mass were measured over 28 days. Seven of the eight injected with C58(NT)D cells developed tumors, averaging 68.8 mm² in size within 28 days p.i. In contrast, only two of the eight developed tumors, averaging 23.4 mm², in the C58(NT)D/R-injected group within 28 days p.i. (Fig. 4), indicating that the resistant C58(NT)D/R subclone possesses more suppressed tumorigenicity than the original C58(NT)D clone.

FIG. 4.

Tumorigenicity of C58(NT)D and C58(NT)D/R cells in nude mice. (A) Tumor incidence in nude mice inoculated with C58(NT)D (thick line) and C58(NT)D/R (thin line) is indicated (n = 8). (B) Tumor size was calculated by averaging the diameters of the tumor areas in the tumor-bearing nude mice inoculated with C58(NT)D and C58(NT)D/R and is indicated as mean ± standard error (error bars). Photographs of examples of tumors in C58(NT)D (C)- and C58(NT)D/R (D)-inoculated mice are shown.

Resistant subclones and their phenotypic modification were reported to occur in the K562 human leukemia cell line (22) and the U937 human promonocytic cell line (9) infected with the H-1 virus. A cell clone, KS, resistant to lytic infection with the H-1 virus, shows a suppressed malignant phenotype and expressed wild-type p53, undetectable in parental K562 cells (22). Resistant cell clone RU from H-1 virus-infected U937 cells significantly decreased tumorigenicity and the accumulation of c-Myc and exhibits monocytic differentiation in cell surface antigens and nonspecific esterase activity (9). Our findings with C58(NT)D/R, such as enhanced cell adherence, microvillus increase and elongation on the cell surface, resistance to apoptosis, and suppressed tumorigenicity, also suggest that the resistant clone altered the differentiated or activated state of lymphoma cells. The appearance of resistant clones from RPV-infected C58(NT)D cells is considered to involve the same mechanism acting on H-1 virus-infected K562 and U937 cells.

In conclusion, we clarified that RPV was propagated in the rat thymic lymphoma cell line C58(NT)D and induced apoptosis and isolated subclones resistant to RPV infection involving differentiated phenotypes. This resistant subclone is expected to contribute much to the understanding of apoptotic and anticancer mechanisms mediated by parvovirus infection.