Using Biological Properties of Viruses to Develop Experimental Models for Testing Antiviral Compounds

Il’ya Vinogradov, Galina Kochneva, Elena Ryabchikova
State Research Center of Virology and Biotechnology “Vector”
Koltsovo, Novosibirsk region, RUSSIA 630559

          The development of effective chemical antiviral compounds is one of the most important directions of modern science. There are very few examples of chemical antiviral preparations used in medical practice. One of these preparations is remantadine, which affects the operation of endosomes: endosomes are responsible for uncoating of influenza virus and release of its gnome into cytoplasm [5,7]. The invention of new antiviral compounds is laborious and expensive work, including sequential testing of antiviral activity in cell cultures and in animal models.
          While development of cell culture models for testing of an antiviral usually goes smoothly, elaboration of appropriate animal model is a complicated task. Not every viral disease can be adequately modeled in animals. Sometimes a virus has only one natural host, such as smallpox virus. The routine approach is to use animals that are susceptible to a virus and develop fatal disease. Such models permit the evaluation of antiviral activity using changes in mortality. However, mortality rate is integral parameter, representing the influence of the antiviral compound on a disease in all aspects of the infection. The goal of our study was to examine which mechanisms may be responsible for the fatal outcome in virus infection and which parameters of the infection can be use to evaluate the antiviral activity.

MATERIALS AND METHODS:
          Virus. Strain EP-2 was isolated from sick elephant in zoo in Germany, and preliminarily identified as elephant pox [1]. Sequence analysis showed that the agent belonged to the cowpox virus species [6]. We used uncloned virus, passage three times in chick Charon-allantoic membrane. Virus activity was evaluated in TV-1 cell culture by biotitration.
          Animals and infection. Outbred mice ICR weighing (group I) 7-9 g (166 animals) and (group II) 10-14 g (97 animals) were infected i.p. by doses of 104 to 107 PFU. Control animals for both groups received the same volume of physiological saline.
          Virological studies. Fragments of abdominal wall having size of 1 cm2 and peritoneal washings were sampled aseptically from 4-6 mice infected by 106 PFU at 24 and 48 h postinfection. Virus titers were evaluated by PFU technique in both fragments and washings.
          Microscopy. Mice were sacrificed by cervical decomposition at various times postinfection. Blood smears were prepared and cells were counted. Liver, spleen, mesentery and abdominal wall were dissected from 3 mice at each time point. Samples were fixed in 4% paraformaldehyde in Eagle MEM. For electron microscopy fixed samples were postfixed in 1% osmium oxide, processed by routine method and embedded in epon-araldit mixture. Samples for light microscopy and immunohistochemistry were routinely processed and embedded in paraffin wax. Paraffin sections were stained by hematoxyline-eosine. Viral antigens were detected using human serum (titer 1:10,000) in dilution 1:200. Antigen-antibody complexes were visualized using polyspecific biotynilated antibodies conjugated with immunoperoxidase (VIP-kit, Vectastain, Vector Lab., USA). More details of the experiments may be found in [8].

RESULTS AND DISCUSSION:
          The results obtained showed that development of disease and its outcome in mice infected with EP-2 strain of cowpox virus depended on animal age. Younger mice (group I) showed fatal disease after infection with 107 and 106 PFU (all mice died in 4-6 days). Mortality of mice infected by 105 PFU reached to 40% during 14 days postinfection. Visual signs of disease (conjunctivitis, tousled wool, depressed activity) were noted on day 3 postinfection.
          Mice of group II remained visually healthy after infection with the same doses during 14 days; none of 85 mice was died. It should be noted that mice of both groups did not develop secondary skin lesions (pocks) during 14 days of observation.
          Reproduction of EP-2 strain of cowpox virus was more active in abdominal wall and peritoneal washings of younger mice, group I (Table 1).
          Macroscopic examination of infected young mice did not find visible pathological changes in the visceral organs and mucoses, except in the spleen in some mice. In those cases, the spleen was pale and looked bloodless. All young mice demonstrated opaque bloody exudate in peritoneal cavity; evidence for peritonitis. No exudate was detected in mice of group II and of control group. Visceral organs of group II mice were visually normal. Blood counts remained significantly unchanged during whole period of the observation (14 days) in both groups.
          Light microscopy revealed large hemorrhages, fibrin deposits and necroses infiltrated by neutrophils in mesentery and peritoneal wall of young mice from day 2 postinfection. Small subcapsular focal necroses, hemostasis, erythrocyte depletion of red pulp, foci of lymphocytes having picnotic nucleus were observed in spleen of young mice infected with 106 and 107 PFU from day 3 postinfection. Ultrastructure studies showed apoptosis of lymphocytes in these foci, which was most prominent in mice infected with 106 PFU on days 4 and 5 postinfection. The pattern of damage to spleen, mesentery and abdominal wall was the same in mice infected with all doses, but the level of lesions decreased with decrease of a dose. No pathological changes were detected in liver of all mice during whole period of observation.
          Pathological changes in the mesentery and abdominal wall of more adult mice (group II) were incomparably lower than in young mice. Necroses of some mesothelium cells surrounded by leukocytes were observed.
          Immunohistochemical studies revealed rare infected cells in splenic parenchyma, and in cells located at the periphery of the organ in mice infected by 107 and 106 PFU of EP-2 strain on days 4-5 postinfection. No relation between apoptotic lymphocytes and location of the infected cells was registered. Bright positive staining was observed in cells of splenic capsule and mesentery (Fig. 1). Immunohistochemistry was unable to detect infected cells in mice liver. Liver of a mouse infected with ectromelia virus served as positive control to detect orthopoxviral antigens, and showed extensive specific staining of hepatocytes and macrophages.
          Electron microscopic examination of abdominal wall and mesentery of young mice found replication of EP-2 strain in mesothelium cells and fibroblasts of mice after 48 h of infection with 106 and 107PFU. Number of infected cells grew during later period, and signs of virus replication were observed also in endothelial, adventitial, fat, nerve envelope, myosatellites, smooth in striated muscle cells of mesentery and abdominal wall (Fig. 2, 3). In mice of group II replication of the virus was much poorer, only infected fibroblasts were observed in mesentery and abdominal wall. In spleen of young mice, replication of EP-2 strain was observed in fibroblasts, smooth muscle, endothelial and undifferentiated cells on day 4-6 postinfection with 105, 106 and 107 PFU. It should be noted that replication of EP-2 strain was not detected in macrophages in all the studied organs, in contrast to ectromelia virus, which is known to replicate in macrophages actively [4]. Electron microscopy did not detect replication of EP-2 strain in a liver of all examined mice.
          Ultrastructural parameters of reproduction of EP-2 strain of cowpox virus were identical to those described for other orthopoxviruses. However, the virus demonstrated some features: different types of A-type inclusion bodies, unusual tubular structures. The most interesting feature was production of very few extracellular enveloped virions, which are responsible for dissemination of orthopoxviral infection [2,3]. Details of ultrastructural characteristics of replication of EP-2 strain of cowpox virus are published in [8].
          Replication of EP-2 strain of cowpox virus was registered in mesentery and abdominal wall of all mice of group II infected with 107 PFU, but number of infected cells was incomparably lower, and their set was reduced to those in fibroblasts. No other cellular types were involved in infection. Many infected cells were destroyed. Many neutrophils, monocytes, macrophages and large lymphocytes were located around the infected cells, and the extent of infiltration was markedly larger than in young mice. Thus, EP-2 strain replicated in mice of group II, but the level of replication was far less than in young mice.
          The results of this study showed ability of EP-2 strain of cowpox virus to infect various cellular types in mice, including such highly differentiated as striated and smooth muscle, and fat cells. At the same time, the zone of virus replication was clearly limited by area of application of viral inoculum (peritoneal cavity and adjoining tissues). Pathological changes also were noted within the same borders. Evidently, massive replication of the virus in tissues bordering the peritoneal cavity caused inflammation and destruction of these tissues, which visually resembled peritonitis. The effect was dose- and age-dependent, and so we suppose that outcome of the infection mostly was related to organism resistance and its physiological condition.
          From our analysis of the data, we conclude that generally the infection in fatally sick mice is related to two main factors: production of very few numbers of extracellular enveloped particles and the inability of EP-2 strain to replicate in macrophages. Thus, the fatal disease in mice caused by EP-2 strain of cowpox virus represents a disease depending firstly on characteristics of the infected animal. This calls into doubt application of such an experimental model for testing of antiviral action of a compound.
          However, the virus readily replicated in various mouse cells, and this provides the opportunity and possibility to test antiviral preparation, using a non-fatal model, i.e., criteria other than mortality. Biotitration of tissues supporting the virus replication may serve as one of these criteria. Other criteria may be changes in the number of infected cells in a tissue as detected by immunoperoxidase or immunofluorescence techniques. Using electron microscopy we are able to provide details on the virus replication in target cells, and thereby to confirm primary data obtained in vitro about the replication stage of a virus altered by antiviral preparations. Usage of animals that do not develop fatal infection, but which can support replication of a virus, allows us to test a preparation in vivo, against background of blood and all biologically active substances. Advantage of this approach include a possibility to confirm primary data about antiviral activity obtained in vitro without using animal models, which exactly mimic the human disease and which very often are expensive and sometimes simply do not exist.

TABLES & FIGURES:

          Table 1. Concentration of strain EP-2 of cowpox virus in peritoneal washings and abdominal wall of mice

KEY WORDS:

          cowpox virus, mice, age-dependent infection, microscopy

REFERENCES:

          1. Baxby D. Variability in the characteristics of pocks produced on the chick chorioallantois by white pock mutants of cowpox and other poxviruses. J. Hyg. Camb.1969, V. 67, 637-647.
          2. Blasco R., Moss B. Role of cell-associated enveloped vaccinia virus in cell-to-cell spread. J. Virol. 1992, V. 66, 4170-4179.
          3. Buller R.M.L., Palumbo G.J. Poxvirus pathogenesis. Microbiol. Rev. 1991, V. 55, 80-122.
          4. Kochneva G.V., Urmanov I.H., Ryabchikova E.I., Streltsov V.V., Serpintsky O.I. Fine mechanisms of ectromelia virus thymidine kinase-negative mutants avirulence. Virus Res. 1994, V. 34, 49-61.
          5. Kornilaeva G.V., Molibog E.V., Bukrinskaia A.G. pH-dependent hemolytic and neuraminidase activity of remantadine-sensitive and -resistant variants of influenza virus. Vopr Virus. 1983, V. 28, 30-32.
          6. Meyer H., Neubauer H., Pfeffer M. Amplification of “Variola virus-specific” sequences in German Cowpox Virus Isolates. J. Vet. Med. 2002, V. 49, 17-19.
          7. Molotkovskii Iu.G., Manevich E.M., Shvets V.I., Smirnova M.M., Krylova V.N. Membrane structure of remantadine-sensitive and remantadine-resistant influenza virus studied with fluorescent phospholipid probes. Vopr Virusol. 1983, V. 28, 182-187.
          8. Vinogradov I.V., Kochneva G.V., Shchelkunov S.N., Riabchikova E.I. An experimental infection caused by the EP-2 strain of cowpox virus in mice of different ages. Vopr Virusol. 2003, V. 48, 34-38.