Evaluation of continuous cell lines in antiviral studies with murine cytomegalovirus

Ganciclovir Treatment of Murine Cytomegalovirus Infection in Mice Immunosuppressed by Prior Infection with Friend Leukaemia Virus

Since many of the severe cytomegalovirus infections in humans occur in individuals immunosuppressed by the human immunodeficiency virus, we developed an analogous murine model for studying this disease. BALB/c mice infected with the Friend retrovirus complex (FV) were immunosuppressed (i.e., exhibited reduced spleen-cell mitogenic responses and natural killer cell activity) by 21 days after FV inoculation. Challenge with murine cytomegalovirus (MCMV) at that time led to mortality at doses generally non-lethal to normal mice. Superinfection of FV-infected mice with MCMV reduced spleen cell FV infectious centres and splenomegaly, and extended the lives of mice surviving the MCMV infection. Once-daily ganciclovir treatments of 12.5,25, and 50 mg kg−1 given to dually-infected mice for 5 days starting 24 h after MCMV inoculation resulted in 90–100% survival at 14 days (relative to MCMV inoculation) compared to 15% survival in the placebo group. By 70 days, survival in the drug-treated and placebo groups were 0–5%, these deaths being attributed to FV disease. Ganciclovir treatments reduced MCMV titres in spleen, liver, and kidney during treatment (day 4 of the infection), but lung and salivary gland titres rose between days 7 and 13 in surviving animals. Improved concanavalin A-induced mitogenic responses were noted on day 4 in mice treated with 25 and 50 mg kg−1. These results indicate that the FV/MCMV dual infection in mice can be used as a model for evaluating antiviral agents. Because of the complex nature of the interaction between FV and MCMV, the model may be more appropriate for advanced studies of well-defined viral inhibitors than for routine screening of potential new compounds.

Resolving the titer of murine cytomegalovirus by plaque assay using the M2-10B4 cell line and a low viscosity overlay

BACKGROUND

Murine cytomegalovirus (MCMV) is increasingly used as an infectious model to investigate host-pathogen interactions in mice. Detailed methods have been published for using primary murine embryonic fibroblasts (MEFs) for preparing stocks and determining viral titers of MCMV. For determining the titer of MCMV by plaque assay, these methods rely on a high viscosity media that restricts viral spreading through the supernatant of the culture, but is also usually too viscous to pipet. Moreover, MEFs must be repeatedly generated and can vary widely from batch-to-batch in purity, proliferation rates, and the development of senescence. In contrast, the M2-10B4 bone marrow stromal cell line (ATCC # CRL-1972), which is also permissive for MCMV, has been reported to produce high-titer stocks of MCMV and has the considerable advantages of growing rapidly and consistently. However, detailed methods using these cells have not been published.

METHODS

We modified existing protocols to use M2-10B4 cells for measuring MCMV titers by plaque assay.

RESULTS

We found that MCMV plaques could be easily resolved on monolayers of M2-10B4 cells. Moreover, plaques formed normally even when cultures of M2-10B4 cells were less than 50% confluent on the day of infection, as long as we also used a reduced viscosity overlay.

CONCLUSIONS

Overall, our protocol enabled us to use a consistent cell line to assess viral titers, rather than repeatedly producing primary MEFs. It also allowed us to start the assay with 4-fold fewer cells than would be required to generate a confluent monolayer, reducing the lead-time prior to the start of the assay. Finally, the reduced viscosity CMC could be handled by pipet and did not need to be pre-mixed with media, thus increasing its shelf-life and ease-of-use. We describe our results here, along with detailed protocols for the use of the M2-10B4 cell lines to determine the titer and grow stocks of MCMV.

Cytomegalovirus infection in burns: a review

Sepsis is responsible for significant morbidity and mortality in patients suffering from severe burn injuries. Burn patients are known to be immunocompromised, and it is generally accepted that the immunosuppressed patient may experience human cytomegalovirus (HCMV) infection and disease. Review of the very limited available literature identifies a seroconversion rate of between 18 and 22% for burn patients who were seronegative for HCMV prior to suffering their burn injury. Furthermore, approximately 50% of HCMV antibody-positive patients may reactivate. Blood products and allografted skin have clinically been identified as possible sources of HCMV transmission in burn patients. Experience in the treatment of infection or disease in burn patients is very scarce and limited to immunoglobulin therapy. Animal experiments have demonstrated that murine cytomegalovirus (MCMV)-seropositive skin grafts are able to infect immunodeficient mice as well as burned mice. Murine studies have also demonstrated that infection with MCMV enhances susceptibility to secondary bacterial infection and increases mortality in these animals. Burned mice challenged with MCMV have a significantly higher level of bacterial translocation to mesenteric lymph nodes than either control thermally injured mice without MCMV inoculation or non-burned mice injected with MCMV alone. In summary, it remains controversial whether HCMV infection per se alters outcome for the majority of burn patients. Subgroups of severely burned, seronegative patients may benefit from blood products and skin from seronegative donors. Antiviral strategies are not yet evaluated for the burn patient. Further investigations utilizing modern diagnostic techniques seem necessary.

Soluble IL-4 receptor improves the skin-graft-associated cytomegalovirus infection in thermally injured mice

The susceptibility of thermally injured mice (TI-mice) to murine cytomegalovirus (MCMV) infection is markedly influenced by burn-associated type 2 T cell responses, which are common with severe thermal injuries. In the present study, the effect of soluble IL-4 receptor (sIL-4R) on the skin-graft-associated MCMV infection was investigated. The marked growth of MCMV was demonstrated in the salivary glands of TI-mice grafted with MCMV seropositive [MCMV sero(+)] skin. However, the growth of MCMV was not demonstrated in the salivary glands of TI-mice grafted with MCMV sero(+) skin and treated with 50ng per mouse of sIL-4R. Compared with grafted normal mice, production of type 1 cytokines was markedly decreased when splenic T cells from TI-mice grafted with MCMV sero(+) skin were stimulated with anti-CD3 monoclonal antibody (mAb). The impaired type 1 cytokine production was recovered in cultures of splenic T cells from grafted TI-mice previously treated with sIL-4R. After grafting with MCMV sero(+) skin, the growth of MCMV was markedly inhibited in the salivary glands of severe combined immunodeficient (SCID) mice inoculated with T cells from TI-mice treated with sIL-4R. These results suggest that sIL-4R regulates the skin-graft-associated MCMV infection in TI-mice.

Susceptibility of thermally injured mice to cytomegalovirus infection

Thermally injured patients are very susceptible to infection with cytomegaloviruses. In this study a role of burn-associated type 2 T cell responses on the cytomegalovirus infection was examined in a mouse model of thermal injury. A predominance of type 2 T cell responses in splenic lymphocytes of thermally injured mice has been previously demonstrated. SCID mice inoculated with splenic T cells from thermally injured mice were susceptible to infection with a small amount (5 PFU/mouse) of murine cytomegalovirus (MCMV). Conversely, SCID mice inoculated with splenic T cells from normal mice were resistant to the same infection. High levels of IL-4 and IL-10, but not IFN-gamma and IL-2, were detected in sera of thermally injured mice (TI-mice) infected with MCMV when those were compared with sera of normal mice infected with MCMV. IL-4 and IL-10 (type 2 cytokines) were produced by splenic T cells from MCMV-infected TI-mice, when they were stimulated in vitro with anti-CD3 mAb. Type 1 cytokines (IFN-gamma and IL-2), however, were not produced by these T cells after the same stimulation. In contrast, splenic T cells from MCMV-infected normal mice produced type 1 cytokines by the stimulation with anti-CD3 mAb. These results suggest that the susceptibility of mice to MCMV infection is markedly influenced by burn-associated type 2 T cell responses.

Cadaveric skin allograft-associated cytomegalovirus transmission in a mouse model of thermal injury

As a routine procedure to provide temporary coverage for burn wounds, cadaveric skin allografts have been used in patients with massive thermal injuries. In this study, CMV infection associated with skin grafting was investigated. Graft-associated CMV transmission was shown in a mouse model of thermal injury. Skins from mice 100 days after a nonlethal dose of murine CMV (MCMV) infection contained MCMV DNA and mRNA, although the virus was not isolated from these murine skins. When these skins were grafted to burned mice, the marked growth of MCMV was demonstrated in salivary glands. No viral growth was shown in the salivary glands of unburned mice or CMV sero(+) mice after grafting with these skins. When severe combined immunodeficient beige (SCID-beige) mice were used as recipients for CMV sero(+) skins, all mice died within 30 days after the grafting. Only 1 PFU/mouse of MCMV was shown to be 1 LD(50) in SCID-beige mice, while a 50% mortality rate was shown in normal unburned mice infected with 5 x 10(5) PFU/mouse of MCMV. This indicates that a very small amount of CMV contained in skins is sufficient to induce CMV infection in immunocompromised hosts. On the other hand, human CMV (HCMV) DNA and mRNA were detected by PCR analysis in 55% (DNA) and 33% (mRNA) of cadaveric skins, although the isolation of HCMV from cadaveric skin homogenates was not achieved in tissue cultures. CMV sero(-) patients with severe burn injuries may have a high risk for CMV infection associated with allografts of cadaveric skins.

Propagation and titration of murine cytomegalovirus in a continuous bone marrow-derived stromal cell line (M2-10B4)

Murine cytomegalovirus (MCMV) can only be propagated effectively in mouse embryo fibroblast (MEF) cells. We demonstrate that MCMV replicates significantly better in M2-10B4 cells, a continuous line of murine bone marrow stromal cells. M2-10B4 cells were also comparable to MEF cells for detection of small amounts of MCMV reactivating from latently infected spleen explants. M2-10B4 cells will be very useful for studies of MCMV pathogenesis.

Selective cytotoxicity of ricin A chain immunotoxins towards murine cytomegalovirus-infected cells

Immunotoxins were constructed by linking immunoglobulins specific for murine cytomegalovirus (MCMV) to deglycosylated ricin A chain. Toxicities toward MCMV-infected and uninfected cells were determined by measuring the inhibition of protein synthesis following a 48-h exposure to immunotoxins commencing 24 h after infection. The 50% inhibitory concentrations ranged from 0.4 to 4 micrograms/ml for infected cells and from 22 to 120 micrograms/ml for uninfected cells. Selectivity indices ranged from 30 to 157. Control immunotoxins, which were constructed identically except that the immunoglobulin moiety had no specificity toward MCMV antigens, had 50% inhibitory concentrations of 50 and 100 micrograms/ml toward infected and uninfected cells, respectively.

Antiviral activity of 9-[[(ethoxyhydroxyphosphinyl)-methoxy]methoxy] guanine against cytomegalovirus and herpes simplex virus

An isosteric analog of 9-[2-(phosphonomethoxy)ethyl]guanine (PMEG), 9-[[(ethoxyhydroxyphosphinyl)methoxy]methoxy]guanine (SKI 1008), was evaluated for its in vitro antiviral activity against herpes simplex virus type 1 (HSV-1) and type 2 (HSV-2), murine cytomegalovirus (MCMV), and human cytomegalovirus (HCMV), and its in vivo antiviral efficacy against MCMV in mice. The in vitro anti-HSV activity of SKI 1008 was much lower than that of acyclovir, even though SKI 1008 showed similar antiviral activity against thymidine kinase positive (TK+) and thymidine kinase negative (TK-) strains. Like ganciclovir and PMEG, SKI 1008 selectively inhibited plaque formation of MCMV; the 50% effective concentration (EC50) and the 50% cytotoxic concentration (CC50) of SKI 1008, ganciclovir, and PMEG being 0.51 and 600, 1.65 and 461, and 0.06 and 12.1 micrograms/ml, respectively. The in vitro EC50 value of SKI 1008 against HCMV was comparable to that of ganciclovir (0.24 vs 0.16 microgram/ml) and was 12-fold higher than that of PMEG in a plaque reduction assay, but the therapeutic indices (the ratios of CC50 to EC50) of SKI 1008 and ganciclovir were higher than that of PMEG. The in vivo antiviral efficacy of SKI 1008 in MCMV-infected normal BALB/c and severe combined immunodeficiency (SCID) mice was lower than that of ganciclovir in terms of mortality and mean survival time.

Selective cytotoxicity towards cytomegalovirus-infected cells by immunotoxins consisting of gelonin linked to anti-cytomegalovirus antibody

An immunotoxin specific for cells infected with human cytomegalovirus (HCMV) was constructed by attaching the ribosome-inactivating enzyme, gelonin, through a disulfide linkage to polyclonal human immunoglobulin (IgG). In uninfected cells, there was no difference between [35S]methionine incorporation in untreated cultures and those treated with immunotoxin at 100 micrograms/ml. In HCMV-infected cells, there was a significant decrease in [35S]methionine incorporation in the immunotoxin-treated cultures, suggesting a selective cytotoxic effect on the virus-infected cells. An immunotoxin specific for murine cytomegalovirus (MCMV) was prepared by linking gelonin to polyclonal anti-MCMV IgG. Using this same parameter for assay of cytotoxicity, the anti-MCMV immunotoxin had a 50% cytotoxic concentration of 35 micrograms/ml in MCMV-infected cells and greater than 200 micrograms/ml in uninfected cells. MCMV yields measured at 7 days postinoculation were reduced by 2 log10 in cultures treated with immunotoxin at 20 micrograms/ml at 1 day postinoculation. These data suggest immunotoxins may have potential for eliminating CMV-infected cells from the host.

Nucleotide analogs related to acyclovir and ganciclovir are effective against murine cytomegalovirus infections in BALB/c and severe combined immunodeficient mice

Acyclovir phosphonate [9-(3-phosphono-propyloxymethyl)guanine; SR3722] and the S enantiomer (SR3772), R enantiomer (SR3773), and R,S enantiomeric mixture (SR3745A) of ganciclovir phosphonate (9-[((+/-)-1-hydroxymethyl-3-phosphono)propyloxymethyl]guanine) were evaluated for their antiviral activities against murine cytomegalovirus. In severe combined immunodeficient mice infected with murine cytomegalovirus, SR3773 and SR3745A (12.5, 25, and 50 mg/kg of body weight per day) were superior to ganciclovir in extending the mean time to death, whereas SR3722 and SR3772 was less potent than ganciclovir. In normal BALB/c mice, SR3773 and ganciclovir were approximately equally active in preventing death. SR3773 caused renal tubular damage when administered at 50 mg/kg/day for 15 days. These results suggest that SR3773 may have potential for use in the treatment of human cytomegalovirus infections, but it may also exhibit renal toxicity.

Treatment of murine cytomegalovirus infections in severe combined immunodeficient mice with ganciclovir, (S)-1-[3-hydroxy-2-(phosphonylmethoxy)propyl]cytosine, interferon, and bropirimine

Severe combined immunodeficient (SCID) mice were found to be highly susceptible to murine cytomegalovirus (MCMV) infection. Treatment of infected mice with ganciclovir (12.5, 25, and 50 mg/kg of body weight for 10 days) starting 24 h after virus challenge resulted in delays in death by 2 to 8 days, and no animals survived the infection. (S)-1-[3-Hydroxy-2-(phosphonylmethoxy)propyl]cytosine (HPMPC) was much more potent, with doses of 1, 3.2, and 10 mg/kg/day (for 10 days) increasing the mean survival time by 15 to 30 days. Twenty-day treatments with HPMPC starting 5 days after virus inoculation increased the mean survival time by 24 to 32 days, with once-weekly (50-mg/kg) treatments being equivalent to daily (10-mg/kg) treatments. Delays in the development of liver, lung, and spleen virus titers in ganciclovir- and HPMPC-treated groups correlated with extensions in the mean survival times relative to the survival times of the placebo controls. The two compounds were approximately equally toxic to uninfected BALB/c mice treated for 10 days, causing 80 to 100% mortality after a dose of 150 mg/kg and 0% mortality after a dose of 75 mg/kg. Thus, the relative therapeutic index of HPMPC was 50-fold greater than that of ganciclovir. Recombinant alpha interferon delta 4 alpha 1/alpha 2 (1 x 10(4) and 5 x 10(4) units per mouse per day) and bropirimine (100 and 300 mg/kg/day) provided no protection from the lethal MCMV infection. The severe combined immunodeficient mouse MCMV infection is an important new model that will permit chemotherapy regimens to be studied over several months.

4,6-dibenzamidopyrazolo[3,4-d]pyrimidine is a highly selective inhibitor of cytomegalovirus adsorption to cells

The heterocycle, 4,6-dibenzamidopyrazolo[3,4-d]pyrimidine (DBAPP), inhibited cytopathology induced by human, mouse, and vervet monkey cytomegaloviruses (CMV) in vitro at 0.2 to 0.5 microM, but did not inhibit cell replication at less than or equal to 30 microM. Herpes simplex viruses were unaffected by the inhibitor. The antiviral agent ganciclovir was effective against these CMVs at 3-10 microM in parallel assays. DBAPP and ganciclovir were synergistic inhibitors when used in combination. The heterocycle was only active if applied to cells before virus replication, indicating that it inhibited virus adsorption. Cells pre-treated 1 h with 30 microM DBAPP, then extensively rinsed, were resistant to infection by mouse CMV even 3 days after removal of the inhibitor. Human and monkey CMVs were able to infect cells and replicate within 24 h of drug removal. When virus and DBAPP were combined together then dialyzed to remove the compound, mouse CMV infectivity was decreased 1.7 logs, whereas human CMV and monkey CMV infectivity titers were relatively unaffected. Treatment of mice with DBAPP twice a day for 7 days starting 6 h after mouse CMV inoculation caused a moderate increase in number of survivors at 30 mg/kg. Cell to cell spread of the virus may account for poor efficacy of the compound when added after virus infection. DBAPP may serve as a tool to explore aspects of CMV adsorption or to characterize the cellular component of the CMV receptor.