Previous Article | Next Article ![]()
Applied and Environmental Microbiology, May 2003, p. 2580-2586, Vol. 69, No. 5
0099-2240/03/$08.00+0 DOI: 10.1128/AEM.69.5.2580-2586.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
National Research Institute of Fisheries and Environment of Inland Sea,1 and Japan Society for the Promotion of Science, Ohno, Saeki, Hiroshima 739-0452,2 SDS Biotech K.K., Tsukuba, Ibaraki 300-2646, Japan3
Received 5 August 2002/ Accepted 12 February 2003
|
|
|---|
|
|
|---|
The first interesting aspect of the H. circularisquama-HcV system is the ecological relationship. The abundance of virus-like particles in the sea was estimated to be 105 to 109 ml-1, which was much higher than had been estimated before the 1990s (1, 2, 28, 43), and evidence showing the ecological importance of algal viruses has gradually accumulated (21, 33, 35). In the previous study on a large double-stranded DNA algal virus, HaV (22), and its host, Heterosigma akashiwo, which causes dense blooms in coastal environments (8, 10), it was shown that HaV has a considerable impact on the dynamics of blooms in the natural environment (20, 21, 35). Thus, the interrelationship between HcV and H. circularisquama is of interest from the viewpoint of bloom dynamics, especially the degradation of H. circularisquama blooms. Considering that many dinoflagellate species cause red tides, the host-virus system is undoubtedly useful material with which to examine viral impact on dinoflagellate blooms.
The second aspect of interest is related to fisheries and environmental remediation research. Because H. circularisquama has caused heavy commercial damage to the aquaculture industry of bivalves such as short-necked clams, blue mussels, Pacific oysters, and pearl oysters in the western part of Japan (17-19), detailed investigations of the biology of H. circularisquama have been conducted. In the process, it was shown that H. circularisquama has several characteristics distinct from the other representative red tide-causing microalgae, such as members of the genus Chattonella and H. akashiwo: it can tolerate high temperature and salinity (44), it forms a temporary cyst that is tolerant to the cataclysm of ambient conditions (11, 40) or bacterial attack (26), and it can also attack other phytoflagellates through direct cell contact (39, 40). As HcV is a natural infective agent of H. circularisquama, it seemed meaningful to assess the possibility of its use as a microbiological agent for controlling blooms.
On the basis of these backgrounds, the objective of the present study was to examine the growth characteristics and intraspecies host specificity of HcV by elucidating the interaction between HcV and H. circularisquama through laboratory experiments.
|
|
|---|
Viruses.
Ten clonal HcV strains used in the present study were free from bacterial contamination: five strains (HcV 01 to 05) were isolated from the surface water of Wakinoura Fishing Port in Fukuoka Prefecture, Japan, on 12 August 1999, and the others (HcV 06 to 10) were from the surface water of Fukura Bay, Hyogo Prefecture, Japan, on 19 August 1999. As a typical lytic virus strain, HcV 03 was principally used in the present experiments. The virus stock was inoculated into a fresh culture of H. circularisquama HU9433-P and incubated under the conditions given above, and the newly obtained viral suspension made cell-free by centrifugation (2,000 rpm for 10 min) was used as an inoculum in each experiment. Viral abundance was estimated by the extinction dilution method in our experiments (22, 32), and the most probable number was calculated (27). Thus, although virus abundance was measured immediately after inoculation in each experiment, the multiplicity of infection (MOI) in each inoculation was calculated 10 to 14 days postinoculation.
Virus sensitivity and growth conditions of host cultures.
Preliminary experiments were designed to define the difference in sensitivity of H. circularisquama to HcV infection in relation to the growth phase of the host culture. Samples (3 ml) of H. circularisquama HU9433-P culture in the late log phase and stationary phase were inoculated with HcV 03 at an MOI of 29 and 4.9 infectious units cell-1, respectively, which were sufficiently high to make most of the cells in the cultures simultaneously exposed to viral attack. These assays were carried out under the conditions given above. In parallel, the growth of H. circularisquama HU9433-P without viral inoculation was also monitored as a control. Algal growth was determined with a Turner Designs fluorometer (model 10-005R) equipped with a 436-nm excitation filter and >650-nm emission filter. The fluorescent unit indicates the relative biomass of the host alga. Each assay was run in triplicate.
Furthermore, a semicontinuous culture experiment was designed to verify whether the physiological conditions were related to the virus sensitivity of the host cells. H. circularisquama HU9433-P was inoculated into eight series of flasks and incubated for 3 days under the conditions described above, and 0, 33, 50, or 67% of the culture was replaced with fresh SWM3 every 24 h. After 6 days of semicontinuous dilution, a fresh HcV suspension with or without heat-treatment (100°C, 5 min) was added to each host culture to give an MOI of 0.49 infectious units cell-1 in all eight flasks. Thereafter, the flasks were incubated without semicontinuous dilution. Throughout the experiments, the abundance of host cells was monitored by direct counting under an optical microscope.
Effect of temperature on algicidal activity of HcV.
One hundred microliters of a vigorously growing culture of H. circularisquama HU9433-P was inoculated into 3 ml of fresh SWM3 and transferred to four different temperatures (15, 20, 25, and 30°C). After 5 days of acclimation at each temperature, when they were in the exponential growth phase, an aliquot of the new virus suspension was inoculated to give an initial MOI of 73, 29, 24, and 21 infectious units cell-1 at 15, 20, 25, and 30°C, respectively. Light conditions were as given above, and host growth was monitored by the use of a Turner Designs fluorometer (model 10-005R). All experiments were run in triplicate. In parallel, the growth of H. circularisquama HU9433-P without viral inoculation was also monitored as a control.
One-step growth experiment.
In order to estimate the latent period and the burst size of HcV, one-step growth experiments were designed. In the experiments at 20°C and 25°C, algal host cultures were inoculated with HcV 03 at an initial MOI of 64 and 198 infectious units cell-1, respectively. Light conditions were as defined above. The abundances of host cells and total infectious centers (free viruses and infected cells) were monitored periodically by microscopic direct counting and the extinction dilution method, respectively (22, 32). On the basis of the changes in algal cell abundance and the viral abundance, the burst size and latent period were calculated.
During the experiments, aliquots of the algal culture at 20°C were periodically prepared for transmission electron microscopy by a previously reported method (9, 36). Thin sections were stained with uranyl acetate and lead citrate and observed under a JEOL JEM-1010 transmission electron microscope.
Intraspecies host specificity.
The 53 H. circularisquama strains were independently inoculated with each of the 10 HcV strains. First, 0.6 ml of exponentially growing host culture was inoculated with 0.2 ml of a fresh virus suspension diluted 24 times with SWM3 after the surviving cells had been excluded by centrifugation (7,000 rpm for 5 min) and incubated as described above. Lysis of the host algal culture was regarded as being caused by viral infection on the basis of visible characteristics (formation of a pale greenish pellet). Host-virus combinations with indistinct results were reexamined. The resultant data sets were converted to a Euclidean distance matrix and analyzed by unweighted pair-group method analysis of clustering in PHYLIP (Phylogeny Inference Package, version 3.5 [6]). A bootstrap analysis (100 replicates) was used to test the robustness and stability of the branching.
|
|
|---|
![]() View larger version (26K): [in a new window] |
FIG. 1. Algicidal effects of HcV03 on growth of H. circularisquama HU9433-P at 20°C. HcV strain 03 was inoculated at the exponential growth phase (on the third day) at an MOI of 29 () or stationary phase (on the 20th day) at an MOI of 4.9 ( ). As a control experiment, host growth without viral inoculation is also indicated ( ). Bars indicate the standard deviation (n = 3).
|
In the semicontinuous culture experiment, the intermittent replacement of the host algal culture with fresh culture medium should have the following effects: (i) reducing host cell concentration, (ii) reducing the concentration of waste products excreted from host cells, and (iii) supplying the essential growth components in SWM3. On the basis of the data shown in Fig. 2, the relative growth of H. circularisquama was estimated by calculating (cell density 5 days after virus inoculation)/(cell density at virus inoculation). When the replacement percentage of the culture was high, the growth activity was high in the control cultures but the host cells were highly sensitive to viral infection, and vice versa (Fig. 3).
![]() View larger version (40K): [in a new window] |
FIG. 2. Changes in abundance of H. circularisquama HU9433-P in semicontinuous culture experiments. Either 0% (A), 33% (B), 50% (C), or 67% (D) of the cultures was replaced with fresh SWM3 for 6 days before viral inoculation. Arrows indicate the time of viral inoculation (). As a control, the abundance of the host without viral inoculation is also indicated ( ). Bars indicate the standard deviation (n = 3).
|
![]() View larger version (25K): [in a new window] |
FIG. 3. Relative growth of H. circularisquama HU9433-P in semicontinuous culture experiments (see Fig. 2) estimated by calculating (cell density 5 days after virus inoculation)/(cell density at virus inoculation). Solid and open bars indicate experiments with and without viral inoculation, respectively.
|
Our speculation is that the physiological condition of H. circularisquama cells was presumably diverse rather than identical (flat) even in a clonal batch culture, and it was closely related with their ability to change into temporary cysts that were more resistant to viral attack. From the exponential growth phase through the stationary phase, it is likely that the proportion of host cells more changeable into temporary cysts increased to cause the low sensitivity to viral infection.
Effect of temperature on algicidal activity of HcV.
The algicidal effect of HcV 03 occurred over a wide range of temperatures, 15 to 30°C, and algal lysis was remarkable at 25 to 30°C (Fig. 4). A negative effect of temperature on the algicidal activities of viruses was not as clear as the interaction between HaV and H. akashiwo (23).
![]() View larger version (30K): [in a new window] |
FIG. 4. Algicidal effects of HcV 03 on growth of H. circularisquama HU9433-P at 15°C, 20°C, 25°C, and 30°C. Host growth with () and without ( ) viral inoculation is shown in each graph. Bars indicate the standard deviation (n = 3).
|
One-step growth experiment.
The one-step growth experiments revealed the growth parameters of HcV. To calculate the burst size, the abundance of hosts and viruses of 48 h to 64 h and 32 h to 64 h was determined in the experiments at 20°C and 25°C, respectively. At 20°C, the latent period and burst size were estimated at 56 h and 1,800 infectious particles cell-1, respectively (Fig. 5A). At 25°C, virus propagation was faster, and the burst size was higher; the latent period was 40 h, and the burst size was 2,440 infectious particles cell-1 (Fig. 5B). These data agree with the idea that the more vigorously growing host cells at 25°C are preferable for viral growth because of their higher biosynthesis activity (44). The growth parameters calculated through the one-step growth experiments were similar to those previously estimated by transmission electron microscopy, 48 to 72 h and 1,300 infectious particles cell-1 (36). The burst size of HcV was comparable to that of Chrysochromulina ercina virus (CeV) (30), and the latent period was somewhat longer than those of the other microalgal viruses reported to date (5, 7, 14, 25, 31). Of course, it should be noted that these parameters are affected by the physiological condition of the host cells (3). Even though the burst size of HcV was relatively high among those of the large double-stranded DNA algal viruses, it was smaller than those of the small algal viruses such as Heterosigma akashivo nuclear inclusion virus (HaNIV) (15) and HcSV (Y. Tomaru, K. Nagasaki, K. Tarutani, and M. Yamaguchi, Abstr. 3rd International Algal Virus Workshop, abstr. O-11, 2002).
![]() View larger version (30K): [in a new window] |
FIG. 5. Changes in abundance of H. circularisquama HU9433-P () and HcV 03 ( ) in one-step growth experiments at 20°C (A) and 25°C (B); the initial MOI was 64 and 198 infectious units cell-1, respectively.
|
![]() View larger version (167K): [in a new window] |
FIG. 6. Transmission electron micrographs of thin sections of H. circularisquama HU9433-P at 0 h (A), 8 h (B), 16 h (C), 24 h (D), and 32 h (E) postinfection with HcV 03. (F) Thin section of a surviving cell at 48 h postinfection, Bars: 500 nm (A to D); 2 µm (E and F).
|
Intraspecies host specificity.
Viral lysis occurred in 502 of the 530 (53 host strains and 10 virus strains) combinations between virus strains and host strains tested (502 of 530 = 94.7%). Among the 53 H. circularisquama strains examined for viral sensitivity, 46 were lysed by all 10 HcV strains. In contrast, HB9, HO4, HcAG-1, HcAG-2, HcAG-3, HcAG-4, and HcAG-5 were resistant to some of them, but there was no host strain that showed complete resistance to all 10 HcV strains tested. It was most notable that the five strains from Ago Bay (HcAG-1 to HcAG-5) showed relatively high resistance to viral infection and that the sensitivities to HcV strains 01, 04, 07, and 09 were complementary to those of HcV strains 02, 03, 05, 08, and 10, i.e., the HcAG strains that were sensitive to the former group were resistant to the latter, and vice versa.
Although no obvious relationship between host specificity and the locality of virus strains was found with respect to these results, it was notable that the virus sensitivity spectrum of H. circularisquama HB9 reflected the locality of viruses; it was sensitive to HcV strains 01 to 05 but resistant to HcV strains 06 to 10. The occurrence of resistant combinations between HcV and H. circularisquama was as low as 5.3% (28 of 530), which was about one-fifth of that observed between HaV and H. akashiwo (
28%) reported previously (24). This was also certified by analyzing the algicidal activity spectra of the 10 HcV strains by means of unweighted pair-group method analysis, which also proved the low diversity among them (Fig. 7).
![]() View larger version (9K): [in a new window] |
FIG. 7. Dendrogram showing levels of relatedness among the 10 HcV clones based on the algicidal spectra against 53 H. circularisquama strains by means of unweighted pair-group method analysis. Neighbor-joining analysis gave a similar tree (data not shown). The bootstrap values from 100 resamplings (>50%) are shown above each branch in italics. The scale bar beneath the tree represents a Euclidean distance. Note that the levels of relatedness among the 10 virus strains are extremely high, although they appear divided into two clusters in the tree.
|
Future view.
Although viruses have recently been considered an important component in aquatic ecosystems (33, 38, 43), it has not yet been sufficiently clarified how HcV regulates H. circularisquama populations and how it affects the disintegration of its blooms. Considering that the HcV strains were isolated from natural seawater where H. circularisquama dominated, it is probable that there is a close interaction between HcV and H. circularisquama in the natural environment. Based on the present study, HcV was shown to be highly effective in infecting vigorously growing host cells at a wide range of temperatures and to have a comparatively high growth activity and a relatively wide host strain range for H. circularisquama. This fundamental information on HcV will be helpful in understanding the ecology of the host-virus system in future studies and also in measuring the possibility of its use as a tool for controlling H. circularisquama blooms.
We are grateful to Takuji Uchida (Hokkaido National Fisheries Research Institute) and Ichiro Imai (Kyoto University), who kindly provided the algal cultures tested in the present study. Thanks are also extended to Kensho Nishida and Yoko Shirai (FEIS) for technical cooperation.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»