|
|
||||||||
1 Department of Veterinary Science, The University of Melbourne, Victoria 3010, Australia
2 Division of Infection and Immunity, The Walter and Eliza Hall Institute of Medical Research, The Royal Melbourne Hospital, Victoria 3050, Australia
Correspondence
Glenn F. Browning
glenfb{at}unimelb.edu.au
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
An experimental model for CAV-induced disease is required for the assessment of attenuation. Such a model does not initially need to represent the full spectrum of pathology observed in the field, but rather must enable assessment of attenuation under conditions that produce most severe pathology. We aimed to develop a model based on yolk sac inoculation of 7 day embryos and post-mortem at embryonic day 20. This parallels the vertical transmission of CAV to embryos in naïve breeder flocks infected at the point of lay, which has been recognized as resulting in the most severe naturally occurring form of disease induced by CAV. The model needed to consistently reproduce severe CAV-specific lesions. A further objective was to derive a standard grading system for lesion severity to apply to the model. Viruses demonstrating attenuation in this stringent model may have potential as vaccine candidates.
A full genome clone of virus CAU269/7 has been produced in the plasmid vector pGEM 4T (Brown et al., 2000
). Previous studies have established equivalent infectivity in cell culture of virus recovered by transfection of the cloned genome and the parental virus. In this study, we aimed to compare the infectivity and pathogenicity of the parental and regenerated viruses in the embryonic model. In previous work, we produced mutated viruses from the genomic clone of CAU269/7 virus, and some CAVs containing specific mutations in VP2 have been shown to grow to moderate titres in cell culture (Peters et al., 2005
, 2006
). The studies reported here investigated the ability of these viruses to cause disease in embryos.
| METHODS |
|---|
|
|
|---|
Viral stocks for inoculation were prepared from 400 ml MDCC-MSB1 cells infected using an adaptation of a previously described method (Todd et al., 1990
). Briefly, 400 ml of culture was sonicated at low frequency in an ice bath, SDS was added to 0.5 % and the lysate was incubated for 30 min at 37 °C. Cellular debris was removed by pelleting at 10 000 g for 30 min. Virus was then purified by ultracentrifugation at 80 000 g for 3 h at 15 °C. Viral pellets were washed in RPMI medium and pelleted again at 80 000 g for 3 h.
Mutant viruses were titrated in 96-well trays (Nunc) in 200 µl vols containing 5x105 MDCC-MSB1 cells ml1 using an adaptation of a previously described method (Goryo et al., 1987
). CAV grows slowly in culture and thus the end point of titration can only be established after approximately seven passages. Duplicate 10-fold dilution series of viral stocks were prepared, ranging from a final dilution of 0.05 through to 0.5x1010, and inoculated onto cells. At intervals of 48 h cells were passaged into fresh medium at a dilution of 1 in 4. Each well was scored for evidence of cytopathogenic effect (CPE), and cultures were serially passaged until no difference was detected between successive passages in the lowest dilution at which CPE was observed. The result was confirmed using an indirect immunofluorescence assay (IFA) on the end-point dilution with the anti-VP3 monoclonal antibody mAbVP3 (TropBio, James Cook University, Townsville, Qld, Australia). Viral stocks were resuspended at a final titre of 104.5 TCID50 ml1.
Infection model in embryonated eggs.
An infection model was developed in embryonated eggs in order to assess infectivity and the in vivo phenotype of wild-type and mutant CAV. The model was initially used to verify equivalence between virus regenerated from pCAU269/7 DNA transfection and CAU269/7 virus. Yolk sac inoculation of ten 7 day embryos (E7) with parental virus, virus regenerated from pCAU269/7 and clarified MDCC-MSB1 lysate was repeated on at least two separate occasions.
Chicken embryos were inoculated with viruses Mut C86R, Mut R101G, Mut H103Y, Mut R129G, Mut Q131P, Mut R/K/K150/151/152G/A/A, Mut D/E161/162G/G, Mut L163P, Mut D169G or Mut E186G, wild-type CAU269/7 virus, RPMI 1640 medium supplemented with 10 % fetal calf serum, or MDCC-MSB1 lysate (for n-values see Table 2
) (Peters et al., 2005
, 2006
). Experimental treatments were repeated at least twice on separate occasions. The yolk sacs of specific-pathogen-free 7 day embryos (E7) (SPAFAS Australia) were inoculated with 0.5 ml of inoculum containing 104 TCID50 of virus. Embryos were monitored by candling to detect death before day E21. There was no significant difference in the rate of pre-hatch embryonic mortality between uninfected control embryos and infected embryos, and embryonic mortality was in all cases less than 10 % of embryos incubated.
|
Assessment of viral infectivity.
The establishment of viral infection was assessed by CAV-specific PCR of liver homogenates and by IFA on bone marrow squash preparations to detect VP3. Cellular DNA was prepared from liver homogenates by proteinase K digestion and SDS lysis, and phenol/chloroform extraction. Cellular DNA was resuspended in 50 µl 0.1 M Tris/HCl (pH 8.0), 1 mM EDTA (pH 8.0). CAV DNA was amplified by PCR using the upstream primer CAV2, 5'-GCGGAGCCGCGCAGGGGCAA-3', and the downstream primer CAV10, 5'-TGCTCACGTATGTCAGGTTC-3'. A 100 µl reaction mixture was prepared containing 300 µM each of dATP, dCTP, dGTP and dTTP (Promega), 2 mM MgCl2 (Promega), 0.2 µM of each primer, 10 µl 10xTaq DNA polymerase buffer (Promega), 2 U Taq DNA polymerase (Promega) and 2 µl template DNA. The PCR reaction was incubated at 96 °C for 2 min, followed by 40 cycles at 96 °C for 4 s, 63 °C for 60 s and 72 °C for 60 s, with a final incubation at 72 °C for 5 min. The PCR products were analysed by agarose (1 %) gel electrophoresis and infection verified on the basis of detection of a 677 bp band.
IFAs were performed on squash preparations of bone marrow to detect cells expressing VP3. Cells were fixed in ice-cold methanol (90 %) for 5 min and blocked for 1 h with a solution of 5 % BSA in PBS containing 0.1 % Tween 20 (PBST) at 37 °C in a humidified chamber. The samples were then incubated with mAbVP3 diluted 1/200 in 0.1 % BSA/PBST for 1 h at 37 °C in a humidified chamber. Samples were then incubated with sheep anti-mouse antibody conjugated to fluorescein isothiocyanate (Dako) diluted 1/100 in 0.1 % BSA/PBST for 1 h. The preparations were washed between all incubations with PBS containing 0.1 % BSA and 0.5 % Tween 20.
Lesion scores.
A grading system was established to allow consistent classification of lesion severity associated with CAV infection. Gross lesions within the thymus, bone marrow and spleen were all scored on a scale of 1 to 4. Gross lesions within the bursa and haemorrhages were scored on a scale of 1 to 3. From this a cumulative lesion score was derived for the overall severity of pathology. Those lesions most central to the pathology of CAV such as thymic and bone marrow lesions were double weighted in the cumulative lesion score method to reflect their central role in CAV infection. This allowed us to include both minor lesions and significant lesions in our assessment. This method is supported by the findings in positive control-infected birds where all birds will have severe thymic and bone marrow lesions, but not all birds have the minor lesions. The weighting of parameters according to their significance in determining an overall score for a trait is a common method in pathology studies and in statistical models in general, and is used in an attempt to arrive at broad definitions of disease that are not biased by over-representation of minor pathology.
In all cases, a score of 1 indicated no pathology. Table 1
outlines the scoring system used for gross pathology. Cumulative scores between 20 and 26 indicated severe CAV lesions, scores between 15 and 19 indicated moderate lesions, and mild CAV lesions scored between 7 and 14. Median lesion scores in each group of infected birds were compared using the MannWhitney test. Between trial variation in data was analysed using an ANOVA test and was found to be not significant.
|
| RESULTS |
|---|
|
|
|---|
Gross pathology after wild-type virus infection
Severe CAV-specific pathology was present in all birds treated with parental or regenerated CAU269/7 virus in five different experimental trials. Petechial haemorrhages of mild to moderate extent were found in the superficial and deep fascial and subcutaneous tissues. The spleen was reduced in size by between 50 and 80 % and, on sectioning, the parenchyma appeared diffusely pale and there were focal subcapsular haemorrhages. In all lobes of the thymic chains there was a severe loss of parenchyma and either lobar haemorrhages or a subcapsular, gelatinous, red fluid in the lobes, consistent with acute cytolysis. A moderate to severe myelophthisis was evident in the medulla of longitudinally sectioned femurs, and the contents appeared either pale and fatty or red and liquefied, consistent with acute lysis. Paleness of the marrow and loss of cells evident as holes in the marrow is very obvious macroscopically in infected animals. In a minority of birds the Bursa of Fabricius was grossly reduced in size, the walls appeared thinner than normal, and the capsule and parenchymal folds had an appearance of increased convolutions and folding. No additional pathology was found in any organ systems beyond the CAV-specific lesions.
Infection with viruses mutated in VP2
Attenuation of CAV-specific lesions was seen in all embryos infected with VP2 mutated viruses (Table 2
). Viruses could be categorized according to their degree of attenuation into two broad categories: highly attenuated viruses (Mut C86R, Mut R101G, Mut H103Y, Mut R129G, Mut Q131P, Mut R/K/K150/151/152G/A/A, Mut D/E161/162G/G and Mut E186G) and moderately attenuated viruses (Mut L163P and Mut D169G) (Peters et al., 2005
, 2006
). Cumulative lesion scores in embryos infected with any of the highly attenuated viruses were low (medians between 9 and 12) and were intermediate (median 15) for both the moderately attenuated viruses (Table 2
). Median cumulative lesion scores in embryos infected with mutant viruses were significantly lower than in embryos infected with wild-type CAU269/7.
Mean body weights of embryos infected with mutant viruses were not significantly different from that of uninfected embryos, but were significantly greater than the mean body weight of embryos infected with wild-type virus (Table 3
), indicating that the VP2 mutations attenuated the growth suppression induced by wild-type virus infection.
|
For embryos infected with all VP2 mutated viruses, except Mut D169G, median thymic, splenic and bone marrow lesion scores were significantly lower than those of embryos infected with wild-type virus (Table 2
). Median thymic and splenic lesion scores were intermediate in embryos infected with Mut D169G. Either median haemorrhage scores were low or there were no haemorrhages detected in embryos infected with mutant viruses. Median haemorrhage scores in all embryos infected with VP2 mutants were significantly lower than those in embryos infected with wild-type virus, with the exceptions of embryos infected with Mut R/K/K150/151/152G/A/A and Mut D169G.
Median bursal lesion scores were similar in embryos infected with either wild-type or VP2 mutant viruses, and the incidence of grossly apparent bursal lesions was sporadic (Table 2
). However, there was a significant reduction in mean bursa : body weight ratio in embryos infected with CAU269/7 relative to uninfected embryos (Table 3
). In embryos infected with Mut C86R, Mut R101G, Mut R129G, Mut Q131P, Mut L163P, Mut D169G and Mut E186G the mean bursa : body weight ratio was significantly greater than in embryos infected with wild-type virus.
There were no differences detected between the mean PCVs of embryos infected with wild-type virus, VP2 mutant viruses or uninfected embryos. The normal haematocrit cannot be attributed to an absence of CAV infection in the bone marrow as VP3-positive cells were identified by immunofluorescence in bone marrow smears taken from the femoral medullary cavity of infected embryos. It is more likely that normal haematocrits were observed because the rate of loss of infected precursor cells due to CAV infection did not exceed the rate of their maturation and release into the circulation. A reduction in haematocrit is a common finding in most reports of birds infected with CAV at 1 day of age. A haematocrit less than 27 %, and typically between 9 and 23 %, is found at 714 days post-infection for chicks infected at 1 day of age (normal 3237.5 %) (Connor et al., 1991
; Imai et al., 1999
; von Bulow & Fuchs, 1986
). Reports of confirmed outbreaks of CAV in five flocks in New Zealand described the presence of typical CAV-induced lesions in lymphoid tissues, but no changes were observed in the haematocrit (Stanislawek & Howell, 1994
). These reports indicate that anaemia is not always found in field CAV infections. The anaemia reported in birds infected at 1 day of age is a transient feature of the pathology due to the rapid maturation of erythroblastoid cells in the avian bone marrow. Birds infected at 1 day of age develop a peracute severe anaemia at 7 days post-infection, and the haematocrit rapidly returns to normal at 14 days. A normal haematocrit may have been observed in this study due to the 14 day time interval between infection and sampling. Haematocrit data has not been published previously for CAV infection in an embryonic model. The blood was watery on venepuncture due to primary coagulopathy arising from thrombocytopaenia rather than due to low PCV. This distinction in the pathogenesis has been described in previous publications (Engstrom et al., 1988
; Goryo et al., 1989
; McNulty, 1991
; McNulty et al., 1990
; Taniguchi et al., 1982
; Yuasa et al., 1983
).
Mean thymus : body weight and mean spleen : body weight ratios of embryos infected with all VP2 mutant viruses were significantly greater than those of embryos infected with wild-type virus, and were not significantly different from the ratios in uninfected embryos (Table 3
).
Gastrointestinal lesions were seen in 12 of 18 embryos infected with Mut R/K/K150/151/152G/A/A in two independent experiments, and were not seen in embryos infected with any other virus. Of the 12 embryos with gastrointestinal lesions, three had lesions in the crop, one in the proventriculus, one in the ventriculus, one in the duodenum, seven in the jejunum and two in the ileum. The majority of embryos infected with Mut R/K/K150/151/152G/A/A had mild CAV lesions, and only two had lesions graded as moderate (Table 2
). Gross lesions in the crop, proventriculus, ventriculus and intestines consisted of thinning of the gastrointestinal wall, hyperaemia and serosal haemorrhages. Lesions were also found in the liver of six of the embryos, with a generalized pattern of irregular nodularity and purplish-red discoloration. In some there were subcapsular hepatic haemorrhages and focal or marginal infarcts in the hepatic lobes. In one embryo haemorrhage and irregular roughening was seen on the synovial surfaces of both femorotibial and tibiotarsal joints.
Viral infection was detected by CAV-specific PCR on liver homogenates from embryos infected with all mutants and wild-type CAU269/7 (data not shown). PCR product was not amplified from uninfected embryos. Whilst it is possible that DNA from virus inoculated into the yolk sac at day E7 may be detected in the liver at day E21 simply by passive translocation, it is unlikely as the inoculating dose is 104 TCID50 and after 14 days it is therefore unlikely this would be detected in the liver without amplification by virus replication. Together with the IFA results, the PCR results confirm the establishment of virus infection. Bone marrow cells positive for VP3 expression by IFA were found in the femoral medulla of all infected embryos.
| DISCUSSION |
|---|
|
|
|---|
Severe CAV-specific lesions were consistently and uniformly produced in all embryos infected with wild-type virus. Cumulative lesion scores, mean body weight and mean thymus : body weight ratio were found to be representative of data obtained for all other parameters measured. Significant growth depression, measured as a reduction in mean body weight, was seen in embryos infected with wild-type CAV. Infection with wild-type CAV resulted in a significant reduction in thymic and splenic weights, and consistently produced severe lesions in the thymus, spleen and bone marrow, and haemorrhages. Gross bursal pathology was only seen sporadically, but there was a reduction in the mean bursa : body weight ratio in embryos infected with CAV, indicating that bursal effects, even though grossly inapparent, were a significant component of the disease. The bursa is not considered to be a target organ for CAV infection, but bursal lesions secondary to T-lymphocyte depletion in the thymus have been reported previously (Rosenberger & Cloud, 1989
).
In this study, the PCV was normal in all embryos infected with wild-type and VP2 mutant viruses, even though VP3-positive cells were found in bone marrow smears. Thus, the rate of loss of infected precursor cells due to CAV infection did not exceed the rate of their maturation and release into the circulation. Anaemia is a common finding in most reports of birds infected with CAV at 1 day of age, with the PCV reduced between 7 and 14 days after infection (Connor et al., 1991
; Imai et al., 1999
; von Bulow & Fuchs, 1986
). However, there are reports of outbreaks of disease due to CAV in flocks with typical lesions in lymphoid tissues, but no signs of anaemia (Stanislawek & Howell, 1994
). As anaemia in birds infected at 1 day of age is transient, it is possible that the lack of anaemia in infected birds in our study was due to the 14 day time interval between infection and sampling. PCV data have not been published previously for CAV infection of embryos.
CAVs mutated in VP2 were infectious in an embryonic infection model and highly attenuated with respect to growth depression and CAV-specific pathology. As for infection with wild-type virus, cumulative lesion scores and body weight were highly predictive of the pathological scoring and organ weights for all lymphoid tissues. Based on the degree of attenuation, the mutant viruses cluster into two groups: Mut C86R, Mut R101G, Mut H103Y, Mut R129G, Mut Q131P, Mut R/K/K150/151/152G/A/A, Mut D/E161/162G/G and Mut E186G were highly attenuated, and Mut L163P and Mut D169G were moderately attenuated. Attenuation was found consistently for lesions in the thymus, spleen, bone marrow, thymic and splenic weights, and for haemorrhage. There was no growth depression associated with infection by the group of highly attenuated mutant viruses and a moderate reduction in mean body weight was only found for Mut L163P. As these viruses are infectious in vivo and concurrently highly attenuated, they are of interest as potential vaccine candidates.
Atypical gastrointestinal lesions were observed in embryos infected with virus Mut R/K/K150/151/152G/A/A in two independent experiments. These lesions were not seen in embryos infected with other viruses or in uninfected embryos and were therefore a specific effect of this virus. Typical CAV-specific pathology was mild in all embryos infected with Mut R/K/K150/151/152G/A/A. However, severe haemorrhages were seen in the subcutaneous tissues and fascial sheets. The full thickness necrosis of the intestinal wall and serosal haemorrhages are not likely to be a result of severe anaemia and subsequent ischaemic injury to the gastrointestinal tract, as the PCV was normal and a more generalized pattern of injury would be expected with anoxia, with involvement of the myocardium and renal tissue rather than the intestine alone.
A change in viral tropism associated with mutagenesis is perhaps a more plausible explanation for the specific distribution of lesions in the gastrointestinal tract in embryos infected with Mut R/K/K150/151/152G/A/A. The virus contains three mutations that neutralize basic residues in a region of VP2 predicted to be basic amphipathic
-helix. The mutations are also located in the genome at the junction of ORF2 and ORF3 in a segment of Alu repeat sequence. The mechanism by which translation of VP1 from the polycistronic mRNA is initiated is unknown; however, the Alu repeat region could play a role in this process. As a change in tropism is more likely to involve an alteration in structural viral proteins such as VP1, the unique phenotype of this mutation, not seen with other mutations that similarly neutralize the basic charge in VP2, may be due to an influence on VP1 gene translation rather than due to substitutions in VP2.
This study has established that VP2 is an effective target for attenuation through site-directed mutagenesis and is the first study to demonstrate attenuation of viral effects in vivo through modification of VP2. This is in contrast to previous studies examining virus constructs mutated in VP3, which were not replication competent (Danen-Van Oorschot et al., 1997
, 1999
, 2000
). The mechanism by which VP2 mutagenesis attenuates virulence remains unknown. It is probable that VP2 has multiple functions and that directed mutagenesis that maintains infectivity may maintain functions necessary to replication, whilst modifying functions associated with virulence.
| ACKNOWLEDGEMENTS |
|---|
| REFERENCES |
|---|
|
|
|---|
Büchen-Osmond, C. (2006). ICTVdB Management (2006). 00.016.0.02. Gyrovirus. In ICTVdB The Universal Virus Database, version 4. Columbia University, New York, USA.
Connor, T. J., McNeilly, F., Firth, G. A. & McNulty, M. S. (1991). Biological characterisation of Australian isolates of chicken anaemia agent. Aust Vet J 68, 199201.[Medline]
Danen-Van Oorschot, A. A., Fischer, D. F., Grimbergen, J. M., Klein, B., Zhuang, S., Falkenburg, J. H., Backendorf, C., Quax, P. H., Van der Eb, A. J. & Noteborn, M. H. (1997). Apoptin induces apoptosis in human transformed and malignant cells but not in normal cells. Proc Natl Acad Sci U S A 94, 58435847.
Danen-Van Oorschot, A. A., van der Eb, A. J. & Noteborn, M. H. (1999). BCL-2 stimulates apoptin-induced apoptosis. Adv Exp Med Biol 457, 245249.[Medline]
Danen-Van Oorschot, A. A., van Der Eb, A. J. & Noteborn, M. H. (2000). The chicken anemia virus-derived protein apoptin requires activation of caspases for induction of apoptosis in human tumor cells. J Virol 74, 70727078.
Engstrom, B., Fossum, O. & Luthman, M. (1988). Blue wing disease of chickens: experimental infection with a Swedish isolate of chicken anaemia agent and avian reovirus. Avian Pathol 17, 3350.[Medline]
Farkas, T., Tanaka, A., Kai, K. & Kanoe, M. (1996). Cloning and sequencing of the genome of chicken anaemia virus (CAV) TK-5803 strain and comparison with other CAV strains. J Vet Med Sci 58, 681684.[Medline]
Goryo, M., Suwa, T., Matsumoto, S., Umemura, T. & Itakura, C. (1987). Serial propagation and purification of chicken anaemia agent in MDCC-MSB1 cell line. Avian Pathol 16, 149163.[Medline]
Goryo, M., Suwa, T., Umemura, T., Itakura, C. & Yamashiro, S. (1989). Histopathology of chicks inoculated with chicken anaemia agent (MSB1TK5803 strain). Avian Pathol 18, 7389.[Medline]
Imai, K., Mase, M., Tsukamoto, K., Hihara, H. & Yuasa, N. (1999). Persistent infection with chicken anaemia virus and some effects of highly virulent infectious bursal disease virus infection on its persistency. Res Vet Sci 67, 233238.[CrossRef][Medline]
Lamichhane, C. M., Snyder, D. B., Goodwin, M. A., Mengel, S. A., Brown, J. & Dickson, T. G. (1991). Pathogenicity of CL-1 chicken anemia agent. Avian Dis 35, 515522.[CrossRef][Medline]
McNulty, M. S. (1991). Chicken anaemia agent: a review. Avian Pathol 20, 187203.[Medline]
McNulty, M. S., Connor, T. J., McNeilly, F. & Spackman, D. (1989). Chicken anemia agent in the United States: isolation of the virus and detection of antibody in broiler breeder flocks. Avian Dis 33, 691694.[CrossRef][Medline]
McNulty, M. S., Mackie, D. P., Pollock, D. A., McNair, J., Todd, D., Mawhinney, K. A., Connor, T. J. & McNeilly, F. (1990). Production and preliminary characterisation of monoclonal antibodies to chicken anemia agent. Avian Dis 34, 352358.[CrossRef][Medline]
Peters, M. A., Jackson, D. C., Crabb, B. S. & Browning, G. F. (2005). Mutation of chicken anemia virus VP2 differentially affects serine/threonine and tyrosine protein phosphatase activities. J Gen Virol 86, 623630.
Peters, M. A., Crabb, B. S., Washington, E. A. & Browning, G. F. (2006). Site-directed mutagenesis of the VP2 gene of chicken anemia virus affects virus replication, cytopathology and host-cell MHC class I expression. J Gen Virol 87, 823831.
Pringle, C. R. (1999). Virus taxonomy at the XIth International Congress of Virology, Sydney, Australia, 1999. Arch Virol 144, 20652070.[CrossRef][Medline]
Rosenberger, J. K. & Cloud, S. S. (1989). The effects of age, route of exposure, and coinfection with infectious bursal disease virus on the pathogenicity and transmissibility of chicken anemia agent (CAA). Avian Dis 33, 753759.[CrossRef][Medline]
Stanislawek, W. L. & Howell, J. (1994). Isolation of chicken anemia virus from broiler chickens in New Zealand. N Z Vet J 42, 5862.[Medline]
Taniguchi, T., Yuasa, N., Maeda, M. & Horiuchi, T. (1982). Haematological changes in dead and moribund chicks induced by chicken anemia agent. Natl Inst Anim Health Q (Tokyo) 22, 6169.[Medline]
Todd, D., Creelan, J. L., Mackie, D. P., Rixon, F. & McNulty, M. S. (1990). Purification and biochemical characterization of chicken anaemia agent. J Gen Virol 71, 819823.
von Bulow, V. & Fuchs, B. (1986). Properties of the causative agent of avian infectious anaemia (chicken anaemia agent, CAA) in vitro. Dtsch Tierarztl Wochenschr 93, 911.[Medline]
Yuasa, N., Taniguchi, T., Imada, T. & Hihara, H. (1983). Distribution of chicken anemia agent (CAA) and detection of neutralizing antibody in chicks experimentally inoculated with CAA. Natl Inst Anim Health Q (Tokyo) 23, 7881.[Medline]
Received 5 February 2007;
accepted 23 March 2007.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
| J MED MICROBIOL | ALL SGM JOURNALS | |