Virus resistance in the toxic bloom-forming dinoflagellate Heterocapsa circularisquama to single-stranded RNA virus infection


*E-mail; Tel. (+81) 829 55 3529; Fax (+81) 829 54 1216.


HcRNAV is the only known cultured dinoflagellate-infecting RNA virus. Lysis of its host dinoflagellate Heterocapsa circularisquama caused by HcRNAV is followed by apparent cell regrowth. Here we investigate the mechanism supporting the survival phenomenon. The proportion of normal cells with intact nucleus decreased to ∼8% by 3 days post infection, and then, increased to > 90% at 15 days post infection. There were abnormal cells lacking an intact nucleus, and this was followed by propagation of virus-resistant survivor cells. The proportion of HcRNAV-resistant cells in three different subcultures and temporal fluctuations were compared: a clonal H. circularisquama culture without virus inoculation (virus-sensitive, VS), a surviving isolate from the HcRNAV-inoculated Culture-VS incubated in autoclaved medium (virus-resistant, VR) and a portion of Culture-VR incubated with HcRNAV (VR incubated with virus, VR + V). The proportion of HcRNAV-resistant cells in Culture-VS was 0% and in Culture-VR + V was > 94% during the experiment; and Culture-VR fluctuated from 4% to 71%. Hence, the virus resistance was assumed to be reversible. Using Northern hybridization, viral genome accumulation was not detected in Culture-VR + V cells either inoculated with HcRNAV or transfected with HcRNAV-genome; thus, intracellular viral RNA replication was assumed to be interrupted in the virus-resistant cells.


Since the discovery of the high abundance of viruses in natural seawater, aquatic viruses have been intensively studied (Bergh et al., 1989; Suttle, 2007). Some reports show that viral infection is one of the factors affecting phytoplankton dynamics preventing algal host populations from reaching high levels (Brussaard, 2004). All of the eukaryotic microalgal viruses in laboratory culture are lytic (Brussaard, 2004). However, some microalgal hosts and their viruses were found to be coexisting in natural waters during all seasons, e.g. the cosmopolitan prasinophyte Micromonas pusilla and its viruses (Zingone et al., 1999). Laboratory experiments also showed the phytoplankton population recovered after virus-induced lysis in several microalgal host–virus systems allowing both virus and alga to coexist, e.g. Emiliania huxleyi and EhV, and M. pusilla and its DNA viruses (Thyrhaug et al., 2003; Zingone et al., 2006). Thyrhaug and colleagues (2003) suggested the viral host recognition molecules released during cell lysis caused by algal DNA viruses protected algal host cell surfaces from further infections and this enabled such host–virus coexistence. However, the mechanism supporting the resistance of microalgae against RNA virus infection has not been determined.

Heterocapsa circularisquama is a bloom-forming small thecate dinoflagellate that kills bivalves (Horiguchi, 1995), and is one of the noxious marine microorganisms causing harmful algal blooms. The H. circularisquama blooms are accompanied by a specific increase in the abundance of a positive-strand RNA virus (HcRNAV) lytic to H. circularisquama (Tomaru et al., 2007). Based on field surveys in western Japan (Nagasaki et al., 2004; Tomaru and Nagasaki, 2004), HcRNAV infection is now considered to be one of the significant factors affecting the population dynamics of H. circularisquama bloom termination. However, because there is a host–virus co-occurrence observed almost every year (Tomaru et al., 2007), we speculate that there is a host–virus system for preventing from completely killing the alga (total collapse). Indeed, HcRNAV did not cause complete lysis of the suitable host cultures and the recovery of H. circularisquama population was always observed in laboratory experiments (Tomaru et al., 2004). Understanding this phenomenon is essential for showing the ecological dynamics of H. circularisquama and determining the mechanism that supports the continuous host–virus coexistence in nature.

Here we focused on the recovery process of H. circularisquama cultures following HcRNAV infection and examined the virus sensitivity (and/or resistance) of isolates subcloned from originally clonal H. circularisquama cultures with or without HcRNAV infection to determine the virus-resistance mechanism of the survivor host cells using RNA transfection and Northern hybridization techniques. This is the first report investigating the recovery of algal host populations after RNA virus infection. The data provide clues for the molecular mechanism enabling the host–virus coexistence of host–virus populations in natural environments.


Growth experiment

The H. circularisquama strain HU9433-P cells inoculated with HcRNAV34 showed a decrease in cell number within the initial 3 days post infection (dpi), and then the alga regrew despite the high virus titer of between 106 and 107 infectious units ml−1 (Fig. 1A). Using fluorescence microscopy (FLM), two cell types having distinctive features were identified during the process of cell lysis due to viral infection: normal cells equipped with an intact nucleus in which chromosomes were condensed, and the other was abnormal cells retaining an intact cell shape but lacking an intact nucleus where a dim green fluorescence diffused intracellularly (Fig. 2). Using transmission electron microscopy (TEM), both normal and abnormal cells were observed. Normal H. circularisquama cells had cytoplasmic structures typical of dinoflagellates with no symptoms of viral infection. In contrast, aggregations of virus-like particles and disruptions of nucleus, chromosomes spread out in the cytoplasms, were observed in the abnormal cells with diagnostic HcRNAV infections (Tomaru et al., 2004) (Fig. 3). Considering the results of TEM observations, the diffused green fluorescence observed in the abnormal cells shown using FLM, indicating that SYBR-Gold stainable nucleic acids, including both RNA and DNA, may be the increase of viral genomic RNA and/or degradation of host nucleus DNA due to viral infection.

Figure 1.

A. Temporal changes in cell number of Heterocapsa circularisquama in a growth experiment with addition of HcRNAV suspension (closed squares) or autoclaved virus suspension (open squares) and viral titer (open circle: infectious units ml−1). HcRNAV34 was inoculated into a mid-exponential phase culture of H. circularisquama HU9433-P (arrow) at an moi of 74 on day 4.
B. Temporal changes in the percentage of host cells harbouring intact nucleus in the total surviving cells (n > 100) in the virus-inoculated culture of experiment (A) as measured using FLM with SYBR-Gold staining from day 4 to the end of this experiment.
C. Temporal changes in the percentage of cells harbouring HcRNAV-like particles in the total surviving cells (n > 50) in the virus inoculated culture of experiment (A) measured using TEM from day 4 to day 15. ND indicates no data.

Figure 2.

Optical (A and B) and fluorescent micrographs (C and D) of Heterocapsa circularisquama HU9433-P before (A and C) and after inoculation of HcRNAV34 (B and D) at 1 dpi. Fluorescence staining was performed using SYBR-Gold. Note the nucleus of the uninfected cell is intact (A) and infected cell is degraded (B). Bars indicate 10 μm. IN: intact nucleus stained with SYBR-Gold, Ch: chloroplast, SN: SYBR-Gold stainable nucleic acids.

Figure 3.

Transmission electron micrographs of Heterocapsa circularisquama HU9433-P infected with HcRNAV34.
A. Thin section of a cell at 1 dpi harbouring a degraded nucleus (DN) where accumulation of HcRNAV-like particles (arrows) are observed in the cytoplasm.
B. a higher magnification of the unordered aggregation of HcRNAV-like particles. Bars in (A) and (B) indicate 2 μm and 200 nm, respectively.

The normal cell numbers estimated by FLM count rapidly decreased over 3 dpi (the minimum proportion was ∼8%), and then gradually increased with the growth of normal host cells at the end of the experiment (Fig. 1B). Changes in the proportion of normal cells measured by TEM showed a similar trend to that estimated by FLM, although the fluctuation dynamics was less significant (Fig. 1C). These results suggest that a portion of the host cells in HcRNAV-inoculated culture escaped viral infection and grew in spite of the high virus abundance.

Temporal changes in virus sensitivity

To examine temporal changes in virus sensitivity of the host cells cultured using various conditions, we conducted the growth experiments for the host strain H. circularisquama HU9433-P (original culture, Culture-O) (Fig. 4A) (see Experimental procedures). The H. circularisquama strain HU9433-P cells inoculated with HcRNAV34 at day 4 (Culture-O + V) showed a slight decrease in cell number during the initial 2 dpi (day 6), and then regrew (Fig. 4A). Next, we examined the proportions of virus-resistant cells in Culture-O and Culture-O + V respectively. The proportion of virus-resistant cells isolated from Culture-O + V at day 8 (4 dpi) was 97% (data not shown), whereas in contrast, the Culture-O at day 0 was 0% (data not shown). We then prepared three different subcultures (Culture-VS, -VR and -VR + V) originating from an axenic clonal strain of H. circularisquama HU9433-P: Culture-VS, a virus-sensitive clonal isolate from Culture-O, which was cultured in SWM3 medium; Culture-VR, a virus-resistant clonal isolate from Culture-O + V, which was incubated in SWM3 medium; and Culture-VR + V, the same virus-resistant clone with that of Culture-VR but incubated in SWM3 medium with added HcRNAV (see Fig. 4 and Experimental procedures). They were used for the following experiments.

Figure 4.

Temporal changes in virus resistance of three different Heterocapsa circularisquama subcultures.
A. Temporal changes in cell number of H. circularisquama in a growth experiment with addition of HcRNAV suspension (moi = 23.3) (open squares) or autoclaved virus suspension (closed squares) at day 4 (indicated with an arrow). One hundred H. circularisquama cells were isolated from the original culture (Culture-O) at day 0 and from virus inoculated culture (Culture-O + V) at day 8 (= 4 dpi), and then the virus sensitivity of these isolates was examined (see Experimental procedures).
B. Proportion of HcRNAV-resistant cells in the three subcultures of H. circularisquama Culture-VS, Culture-VR and Culture-VR + V respectively. Culture-VS is the virus-sensitive clone isolated from Culture-O. Culture-VS and -VR + V are the same virus-resistant clone isolated from Culture-O + V cultured using a virus-free medium and a virus-added medium respectively. On each sampling day, 100 cells were isolated from each subculture and the successfully established subclonal cultures were used for the virus sensitivity tests (see Experimental procedures). *Number of resistant subclonal cultures/total subclonal cultures.

To estimate the proportion of HcRNAV-resistant cells of three H. circularisquama subcultures (Culture-VS, -VR and -VR + V), the virus sensitivity of these host cultures was tested on days 51, 101 and 255 post initiation of the experiment (Fig. 4B). Throughout the experiment, all of the subclonal cultures (100%) recloned from Culture-VS showed cell lysis due to viral inoculation, while most of those (94–99%) from Culture-VR + V showed resistance to HcRNAV34 (Fig. 4B), whereas the frequency of cells resistant to HcRNAV34 in Culture-VR fluctuated between 4% and 71% (Fig. 4B). This suggests that the survivor cells that escape viral lysis sustain resistance to HcRNAV during continuous exposure to viral infection pressure (Culture-VR + V), while the virus-resistant cell proportion sporadically decreased when released from viral infection pressure (Culture-VR). Hence, the sensitivity (i.e. susceptible or resistant) of H. circularisquama against HcRNAV is assumed to be ‘reversible’ (= not rigid). Here, the susceptible or resistant cell ratio was measured as a parameter showing the overall culture sensitivity.

Susceptibility to HcV

HcV (H. circularisquama virus) is another virus specifically infectious to H. circularisquama, and is large (∼0.2 μm), icosahedral and harbours a double-stranded DNA (dsDNA) genome (Tarutani et al., 2001). Culture-VS was sensitive to both HcV10 and HcRNAV34. Although Cultures-VR and -VR + V showed resistance to HcRNAV34, they were sensitive to HcV10 (data not shown). This suggests that the mechanism that supported the resistance of H. circularisquama cells to HcRNAV differed from the HcV infection.

Intracellular replication of the HcRNAV genome

To determine if the cells in Culture-VR + V permitted to replicate viral RNA due to virion inoculations, accumulation of the complementary negative-strand RNA of HcRNAV genome was examined by using Northern-blot analysis. Accumulation of the viral genome was observed in the cells of Culture-VS inoculated with HcRNAV34, whereas it was undetectable in the cells of Culture-VR + V with the virus inoculations (Fig. 5). This indicates that replication of the viral genome did not occur in Culture-VR + V cells even when they were exposed to intact virions of HcRNAV34.

Figure 5.

Accumulation of positive- and negative-strand RNA of HcRNAV in Heterocapsa circularisquama cells inoculated with HcRNAV34 virions. Total RNA was extracted from H. circularisquama cells at 0 and 24 hours post infection (hpi), separated by gel electrophoresis and blotted onto membranes. The membranes were then probed with strand-specific DIG-labelled RNA probes. The genomic RNA of HcRNAV34 was used as a positive control. Total RNA stained with SYBR Gold is shown. (1) HcRNAV34 viral RNA; (2) Culture-VR + V without virus inoculation; (3) Culture-VS without virus inoculation; (4) virus-inoculated Culture-VR + V at 24 hpi; (5) SWM3-inoculated Culture-VR + V at 24 hpi (negative control); (6) virus-inoculated Culture-VS at 24 hpi; (7) SWM3-inoculated Culture-VS at 24 h (negative control).

To determine whether the resistance exhibited by the Culture-VR + V cells was due to intracellular permissiveness for viral replication, direct transfection of viral RNA into H. circularisquama cells was performed. In Culture-VS cells that show susceptibility to HcRNAV34, the physically introduced viral genomic RNA successfully replicated from 24 through 48 h post inoculation (hpi) (Fig. 6); in contrast, no accumulation of the negative-strand RNA was detected in Culture-VR + V cells transfected with HcRNAV34 genomic RNA at 48 hpi (Fig. 6), suggesting the suppression of intracellular viral replication. The suppression may be due to the lack of replication-associated protein activity (such as RNA-dependent RNA polymerase) through the negative-strand RNA replication process.

Figure 6.

Accumulation of negative-strand RNA of HcRNAV in Heterocapsa circularisquama Culture-VS (1) and Culture-VR + V (4) cells transfected with HcRNAV34 genomic RNA. Total RNA was extracted from H. circularisquama cells at 0 (immediately after inoculation), 24 and 48 h post infection, and blotted onto membranes. The membranes were then probed with a negative strand-specific DIG-labelled RNA probe. Heterocapsa circularisquama cells in Culture-VR + V bombarded by gold particles with no viral genomic RNA (2) and the cells inoculated with gold particles and viral genomic RNA (3) were run in parallel as negative controls.


Recovery and resistance mechanisms

Using the transfection technique with Northern hybridization method for detecting accumulation of viral negative-strand RNA, no clear evidence of intracellular viral RNA replication in the resistant host cell culture (Culture-VR + V) was detected. The intracellular suppression of the viral genome replication would thus primarily contribute to the regrowth of the host culture after viral lysis.

Regrowth of eukaryotic microalgae following viral lysis due to their suitable dsDNA viruses were reported in several host–virus combinations: Phaeocystis pouchetii (Prymnesiophyceae) and PpV, Pyramimonas orientalis (Prasinophyceae) and PoV, Chrysochromulina ericina (Prymnesiophyceae) and CeV, E. huxleyi (Prymnesiophyceae) and EhV, and M. pusilla (Prasinophyceae) and MpV (Thyrhaug et al., 2003; Zingone et al., 2006). Thyrhaug and colleagues (2003) suggested that excess viral molecules released during cell lysis may compete with complete virus particles for receptor sites on the host cell surface, thus reducing infection, which contribute to stable coexistence of microalgal hosts and viruses. According to Zingone and colleagues (2006), it is difficult to explain the resistance mechanism of M. pusilla against MpV based on inhibition alone. There is the possibility M. pusilla cells could mutate to become virus-resistant at the cell surface, this resistance was sustained at least for several years in culture without virus addition (Waters and Chan, 1982; Zingone et al., 2006) and no positive result supporting the possibilities of latent or lysogenic infections for M. pusilla was obtained (Zingone et al., 2006). Although there were a number of previous studies describing the phenomena of host regrowth following viral lysis and virus resistance of microalgae, the mechanisms supporting the host response to viral infection have not been enough determined. Recent study only demonstrated that dsDNA virus (EhV) susceptibility of E. huxleyi is determined by ploidy levels, the haploid phase is virus resistance but the diploid phase is not (Frada et al., 2008). There are few reports concerning the host algal response against RNA virus infections, and the ploidy levels of H. circularisquama seem not to be related to HcRNAV resistance (data not shown). In future studies, involvement of inhibitors and change in cell surface should be assessed that may support resistance of H. circularisquama against HcRNAV.

Reversibility of viral susceptibility

Here we focused on the survivor cells of H. circularisquama that had escaped from HcRNAV infection. Throughout the experiment, only a low percentage (1–6%) of subclones from Culture-VR + V (the resistant subclone cultured in a virus-added medium) showed sensitivity to HcRNAV34. Cells isolated from Culture-VR + V were incubated in virus-free SWM3 for about a month prior to the sensitivity test; our prediction is that sensitive host cells became dominant in some of the subclones isolated from Culture-VR + V during the pre-incubation period. Considering the temporal fluctuation in the proportion of resistant (and sensitive) cells in Culture-VR, this prediction is most likely. The susceptibility to viral infection is presumably reversible, i.e. not determined by irreversible mutation. The pre-incubation period (c. 1 month; see Experimental procedures) may be long enough for the host cells to change from ‘virus-resistant’ to ‘virus-sensitive’.

The proportion of resistant cells in Culture-VR temporarily increased on day 101. The proportion of the virus-resistant cells gradually decrease under a virus-free condition; however, it may fluctuate for a considerable period. In natural environments, HcRNAV is suspended in a water column and preserved in sediments for several months after its occurrence (Tomaru et al., 2007). To sustain a H. circularisquama population after its bloom disintegrations, the fluctuation of the virus-resistant cell proportion may be a merit for reducing a possibility of accidental virus infections.

HcRNAV inoculation to a vigorously growing H. circularisquama culture even with a high multiplicity of infection (moi) did not cause complete lysis; a small proportion (∼8%) of host cells kept their shape and growth ability intact (Fig. 1). Therefore, what is the mechanism that supports the survivor cells' escape from viral infection? There may be two possible explanations: (i) only a very small proportion of the host cells in Culture-VS already express virus resistance or (ii) indeed perhaps all of the cells in Culture-VS were once infected by HcRNAV; then a defence mechanism against viral infection was switched on in a small proportion of them. Because no subclone isolated from Culture-VS showed resistance to HcRNAV (Fig. 4B), the proportion of survivable cells (expressing virus resistance) would be significantly low. Assuming the growth rates of the virus-resistant survivor cells and the uninfected sensitive cells were equal to each other and the proportion of the former cells at 3 dpi was 8% (based on the experiment in Fig. 1), the frequency for surviving cells in the original culture is estimated to be 0.3–0.5% using regression analysis (data not shown).

In typical bacteriophage–host relationships, resistant host cells have physiological disadvantages, e.g. low growth rate, compared with its susceptible cells (Lenski and Levin, 1985). In contrast, we could not find physiological differences between the susceptible H. circularisquama cells and the resistant cells; i.e. their growth rates were not significantly different (data not shown). The HcRNAV resistance systems of H. circularisquama may not require a large physiological cost.

Ecological implication

Heterocapsa circularisquama blooms occur almost every year in a number of semi-enclosed bays in Japan, and in most cases their occurrences are accompanied by specific increases in infectious viruses (Tomaru et al., 2004). Virus infection is one of the most important factors affecting the dynamics of H. circularisquama blooms (Nagasaki et al., 2004), while long-term coexistence of H. circularisquama and HcRNAV was also observed in natural environments (Tomaru et al., 2007). Considering that the response of H. circularisquama to HcRNAV is reversible to some extent as observed here, host–virus coexistence in natural environments may be enabled due to the mutual control between the host–virus populations; i.e. the number of natural cells expressing virus resistance can decrease when the HcRNAV infection pressure is lowered, and vice versa. For instance, they coexisted in Ago Bay in 2002 for more than 2 months where their concentrations ranged 100−102 algal cells ml−1 and 100−103 virus infectious units ml−1, and in Imari Bay in 2002 they coexisted for almost 2 months where their concentrations ranged < 100−101 algal cells ml−1 and < 100−103 virus infectious units ml−1 (Tomaru et al., 2007). These observations support the idea that a mutual controlling system may prevent the host population from reaching high abundance in natural environments.

To understand the viral impact on H. circularisquama populations in nature, however, would be difficult because the HcRNAV-resistant host cells are lysed by HcV. Coexistence of HcV-like and HcRNAV-like viruses in natural blooms of H. circularisquama was observed occasionally (Tomaru and Nagasaki, 2004). A recent study shows a diverse response of H. circularisquama strains to HcRNAV, i.e. sensitive, delayed lysis and resistant (Mizumoto et al., 2008), and diverse infection specificity of HcRNAV to H. circularisquama (K. Nagasaki, unpubl. data).


We show that the virus resistance of H. circularisquama cells to HcRNAV infection is a reversible feature and may be related to an intracellular suppression mechanism barring viral genome replication. This may be significant for both host and virus population dynamics and survival in natural environments. In future studies, determination of the suppression mechanism and factors driving the change in the sensitivity will be required for further understanding the host–virus system, and further ecological and physiological studies are necessary to understand the relationship they mutually maintain in situ.

Experimental procedures

Algal host and viruses

The origins of the clonal virus isolate (HcRNAV34) and the clonal algal host isolate (H. circularisquama HU9433-P) used in this study were previously reported (Tomaru et al., 2004). Heterocapsa circularisquama HU9433-P has not been exposed to any infectious viruses since its isolation in 1994 (Tomaru et al., 2004). Cell cultures were grown in modified SWM3 medium (Itoh and Imai, 1987) enriched with 2 nm Na2SeO3 and incubated under a 12 h light, 12 h dark cycle, light (130–150 μmol photons m−2 s−1) provided by cool white fluorescent illumination at 20°C. The virus stock was inoculated into a fresh culture of H. circularisquama and incubated until host cell lysis. The lysate was filtered through a 0.2 μm polycarbonate membrane filter (Whatman, Middlesex, UK) to remove host cell debris. A large dsDNA virus strain infecting H. circularisquama (HcV10) was also used. The lysate of HcV-inoculated host culture (incubation was the same as mentioned above) was centrifuged at 4900 g, 4°C for 5 min and the supernatants were then examined as described below. Titration for the viruses was conducted using the extinction dilution method as described previously (Suttle, 1993; Tomaru et al., 2004).

Growth experiment

An exponentially growing culture of H. circularisquama HU9433-P (1000 ml) was inoculated with HcRNAV34 at an moi of 74. A H. circularisquama HU9433-P culture inoculated with an autoclaved viral suspension served as the control. Incubation conditions were as described above. A 60 ml aliquot of cell suspension was sampled from each culture at 0, 1, 3, 5, 7, 9, 11 and 14 dpi. The samples were used for monitoring the host cell number and the virus titer, and used for observation by FLM and TEM. The host cell number was estimated using optical microscopy, and enumeration of the lytic agents was conducted using the extinction dilution method (Suttle, 1993; Tomaru et al., 2004). A 50 ml aliquot of the cell suspension was used for TEM observation. The host cells were harvested by centrifugation at 860 g, 4°C for 10 min and fixed with 1% glutaraldehyde in SWM3 for > 4 h at 4°C. Preparation for TEM was performed as previously described (Tomaru et al., 2004). The proportion of host cells harbouring the intracellular virus-like particles was estimated by counting more than 50 cell sections. One-millilitre aliquot of cell suspension was used for FLM. The host cells were fixed with 1% glutaraldehyde for > 4 h at 4°C, and stained with SYBR-Gold (Molecular Probes) at a final concentration of 1 × 10−4 dilution of the commercial stock. The stained samples were filtered onto a 3 μm polycarbonate membrane filters (Nuclepore), then the filters were mounted on a glass slide with a drop of low-fluorescence immersion oil and covered with another drop of immersion oil and a cover slip. The slides were viewed at a magnification of 1000× with an Olympus BX50 epifluorescence microscope where at least 100 cells per sample were observed. The cell images at 0 and 1 dpi were photographed using a Zeiss 13 Axiocam HRc Camera with an epifluorescence microscope Zeiss Axiovert 200 M and FITC filter set (Carl Zeiss).

Temporal change in virus sensitivity

One hundred cells of H. circularisquama HU9433-P in logarithmic growth phase of the original axenic culture (Culture-O in Fig. 4A) were independently isolated (day 0) and incubated under the condition mentioned above, and then 89 isolates were successfully established and subcultured from them. After 28 days of incubation, virus sensitivity of these isolates was examined as mentioned below, and then one virus-sensitive subculture was selected (Culture-VS). Culture-VS was defined at day 42 and used in the following experiments. This virus-sensitive subculture was maintained in the culture medium SWM3 under the condition mentioned above. A 50 μl of the cell suspension was transferred to 5 ml of the fresh culture medium once a month.

At day 4, HcRNAV suspension was inoculated into a vigorously growing cell culture (1.29 × 104 cells ml−1) at an moi of 23 (Culture-O + V) (Fig. 4A). Host cultures inoculated with an autoclaved virus suspension served as a control. One hundred survivor cells were independently isolated into SWM3 medium from the virus-inoculated culture at day 8 (i.e. at 4 dpi) (Culture-O + V, Fig. 4A), and each of them was washed three times with the sterilized medium using a micropipetting method before the isolations to wash out viruses around the cells; as a result, 80 isolates were successfully cultured. Then, the virus sensitivity test was conducted for them at day 28. One among the 80 subcultures that showed resistance to HcRNAV34 was selected and used for further experiments. It was continuously incubated in SWM3 without (Culture-VR) or with an addition of a fresh suspension of HcRNAV34 (v/v = 10%) (Culture-VR + V) under the temperature and light conditions mentioned above. The titer of the HcRNAV34 suspension added to the medium was > 106 infectious units ml−1. Cultures-VR and -VR + V were established and defined at day 42. Thus, after 42 days from the start of the experiments, we prepared three distinctive subcultures (Culture-VS, -VR and -VR + V) originating from the initial H. circularisquama HU9433-P culture (Culture-O). Each was maintained in its conditioned medium, SWM3 (Culture-VS and -VR) and HcRNAV-added SWM3 (Culture-VR + V). A 50 μl of the cell suspension was transferred to 5 ml of the fresh culture medium once a month.

On day 51, 100 cells were isolated using a microcapillary from each subculture; consequently, 76, 83 and 92 clones from Culture-VS, -VR and -VR + V, respectively, were subcultured. As well, respectively, 89, 93 and 94 clones on day 101, and 88, 96 and 91 clones on day 255 were subcultured. The virus sensitivity tests for the subclonal cultures at days 51, 101 and 255 were conducted at days 83, 128 and 282 respectively.

The virus sensitivity tests for the clones from each culture were conducted as follows: a HcRNAV suspension was inoculated (v/v = 3%) independently to the exponentially growing subclonal cultures of H. circularisquama, and they were incubated under the conditions above. The occurrence of algal lysis was monitored by optical microscopy. Host cultures inoculated with SWM3 served as controls. Algal lysis was scored when an aggregation of lysed cells was observed on the bottom of the culture vessels. Algal cultures that were not lysed after 14 dpi were scored as resistant host isolates for HcRNAV34.

Sensitivity to HcV and HcRNAV

Exponentially growing cultures of H. circularisquama (Culture-VS, -VR and -VR + V) were inoculated with HcRNAV34 or HcV10 (v/v = 5%). Cultures inoculated with an autoclaved viral suspension served as controls. The initial moi for Culture-VS versus HcV10, Culture-VS versus HcRNAV34, Culture-VR versus HcV10, Culture-VR versus HcRNAV34, Culture-VR + V versus HcV10 and Culture-VR + V versus HcRNAV34 were 0.4, 6.6, 0.3, 5.2, 0.2 and 4.8 respectively. Incubation conditions were as described above. Algal growth was monitored using a Turner Designs fluorometer (model 10-005R).

Viral genome replication in H. circularisquama cells

To determine whether viral genomic RNA can replicate in virus-resistant H. circularisquama cells, the accumulation of viral positive-strand RNA and the complementary negative-strand RNA specifically synthesized during the replication process was examined using Northern blot analysis with a positive strand-specific RNA probe and the negative strand-specific RNA probe respectively (Mizumoto et al., 2007). Culture-VS and -VR + V were used for the following experiments.

An exponentially growing culture of each subculture (10 ml) was inoculated with virus lysate at an moi of 79 and incubated as mentioned above. Host cultures inoculated with SWM3 served as controls. An aliquot of cell suspension (1.5 ml) was taken from the culture before inoculation and at 24 hpi. Host cells were collected by centrifugation at 3000 g, 4°C for 3 min, and the pellets were stored at −80°C until RNA extraction. Northern blot analysis was performed as described previously (Mizumoto et al., 2007).

Direct transfection experiment

Viral RNA transfection into the cells of Culture-VS and -VR + V was performed as follows: exponentially growing H. circularisquama cells (c. 2.5 × 106 cells) were collected for particle bombardment on no. 3 quantitative filter paper (47 mm Advantec, Tokyo, Japan) by gravity filtration. Gold particles (0.6 μm; Bio-Rad Laboratories) were coated with the RNA extracted from HcRNAV34. Then, the filter paper was set approximately 6 cm below the microcarrier launch assembly and bombarded with the viral RNA-coated gold particles at a rupture pressure of 1350 lb/in2 in a vacuum of 28.5 in. Hg using the Bio-Rad Biolistic PDS-1000/He Particle Delivery System (Bio-Rad Laboratories, Hercules, CA) as described by Mizumoto and colleagues (2007). After bombardment, the filter paper was placed in fresh SWM3 medium (50 ml) and gently shaken to release the transfected cells. The culture was incubated as described above. A 1.5 ml aliquot was sampled from the cell suspension at 0, 24 and 48 h post transfection. Cells were pelleted by centrifugation (3000 g for 3 min) and stored at −80°C until RNA extraction. Northern blot analysis was performed as previously described (Mizumoto et al., 2007).


This study was partially supported by the Industrial Technology Research Grant Program in 2000-2004 from the New Energy and Industrial Technology Development Organization of Japan (NEDO) and the Ministry of Agriculture, Forestry and Fisheries, Japan. Thanks are due to I. Imai (Hokkaido University, Japan) who kindly provided H. circularisquama.