Thraustochytrids are non-photosynthetic marine/estuarine stramenopile protists that are frequently observed and/or isolated from marine and estuarine waters, sediments, algal and plant materials both as saprotrophs and parasites. Their bio-volume in coastal waters is estimated at 43% of that of the bacterioplankton. The ubiquitousness and ability to use a wide variety of organic substrates (including bacterivory) argue for their ecological importance as decomposers . In addition, due to their high production of PUFAs (polyunsaturated fatty acids) such as docosahexaenoic acid and docosapentaenoic acid , they are considered very important as food resources for higher organisms in marine systems . Because of these distinctive features of the thraustochytrids, their ecological significance in the coastal ecosystems has been studied. Kimura et al. biogeographically demonstrated the abundance of thraustochytrids was closely related with the density of POM (particulate organic matter). And Bongiorni and Dini show the abundance and composition of thraustochytrids change with habitats and seasons in Mediterranean coastal areas. However, the effective techniques that can separately estimate the abundance of each genus or systematic groups of thraustochytrids, still have not been established. In spite of their ecological significance, therefore, relatively little is known about their ecological influence and impact in situ. On the other hand, viruses are highly abundant in marine environments and are recognized as important pathogens in controlling bacterial and algal biomass nutrient cycling [, and in maintaining the bio-diversity of bacteria and microalgae . To date, more than thirty algal viruses have been isolated and characterized to different levels of resolution; and particularly, the relationships between algal blooms and viruses have been intensively investigated . The viral infection is considered to affect the dynamics of algal blooms both quantitatively (biomass) and qualitatively (clonal composition). Heterosigma akashiwo-HaV(Heterosigma akashiwo virus) and Heterocapsa circularisquama-HcRNAV (Heterocapsa circularisquama RNA virus) are well known host-virus systems . In both cases, the host and their virus dynamics were tightly linked each other . Considering the fact that viruses can’t reproduce without their specific host, fluctuations in abundance of certain virus may reflect the host dynamics. Therefore, studies on viruses that infect thraustochytrids lead up to novel information about their host. Previously, we reported two distinct viruses infecting thraustochytrids: AuRNAV (Aurantiochytrium RNA virus: reported as SssRNAV) and SmDNAV (Sicyoidochytrium minutum DNA virus) . AuRNAV is a single-strand RNA virus infecting Aurantiochytrium sp. (formerly Schizochytrium sp., see Yokoyama and Honda, ; and SmDNAV is a double-strand DNA virus infecting S. minutum. The two hosts are taxonomically distant within the family Thraustochytriaceae. Here we describe the seasonal change in abundance of viruses infecting the thraustochytrids in Hiroshima Bay, Japan and discuss the ecology of thraustochytrids from the viewpoint of the host-virus relationships. The temporal changes in the field data collected during 2004 are shown in Significant decreases in practical salinity units (psu) were observed three times; in May, Jun, and Aug . The diatoms in the genera Chaetoceros and Skeletonema dominated in the water column almost throughout the survey period except during a Heterosigma akashiwo bloom where the highest cell density was 9.7 × 104 cells ml-1 on Jun 11 . Six phylogenetically distant thraustochytrid clones were used to isolate and enumerate their viruses throughout the survey; and consequently, only two types of virus respectively infecting Aurantiochytrium sp. NIBH N1-27 (here regarded as AuRNAV) and Sicyoidochytrium minutum NBRC 102975 (SmDNAV-like viruses) were detected and successfully isolated . AuRNAV showed a drastic increase in abundance (1.8 × 103 infectious units ml-1) at ten days after the H. akashiwo bloom termination. Then, they rapidly decreased to below the detection limit at 3.0 infectious units ml-1 after 23 Jul .Whereas, the viruses infecting SmDNAV-like viruses did not show any changes in abundance during the survey period which fluctuated at <2.5 × 101 infectious units ml-1 . Considering the fact that viruses can’t reproduce without their specific host, fluctuations in abundance of certain virus may reflect the host dynamics. In addition, our preliminary inoculation experiments showed that Aurantiochytrium sp. NIBH N1-27 attached to H. akashiwo cells and propagated more efficiently than other thraustochytrid strains tested (data not shown). Therefore, we thought that the spike peak of thraustochytrids existed after the H. akashiwo bloom but before the peak of AuRNAV. To confirm the hypothesis, we also estimated the abundance of thraustochytrids during the field survey in 2005. In the present field survey, we detected two distinct types of thraustochytrid virus; they were AuRNAV and SmDNAV-like. Considering both virus types were isolated from a variety of coastal environments in Japan, at least two distinct thraustochytrid-virus combinations may be widely distributed and functioning in universal eukaryotic decomposing systems.
However, no viral agents causing lysis of the other four tested thraustochytrid clones were detected throughout the present survey. Of course, this does not deny the possibility of the existence of other thraustochytrid-infecting viruses. In this study, we used only six host strains to examine virus abundance; as a result, only the viruses which caused lysis of tested hosts were isolated. It may be possible to isolate a wider variety of viruses by using more thraustochytrid strains as hosts. As well, it should be noted that host-virus combinations not accompanied with drastic cell lysis may have been overlooked in this study. The fluctuation patterns in abundance of the two virus types were different from each other. AuRNAV showed a remarkable increase in abundance following the H. akashiwo bloom in 2004 . Since AuRNAV does not infect H. akashiwo , the increase is considered reflecting the drastic increase and viral lysis of Aurantiochytrium sp. NIBH N1-27-type thraustochytrids, which occurred following the H. akashiwo bloom. Actually, we succeeded in detecting an increase in thraustochytrid abundance after the peak of the H. akashiwo bloom also in 2005, which was accompanied with the temporal increase of AuRNAV . Whereas, no statistically significant relationship was found between the abundance of thraustochytrids and AuRNAV from the Pearson’s correlation coefficient analysis (data not shown). A possible explanation for the ecological events which was observed in 2004 is reasonable; i.e., H. akashiwo rapidly multiplied due to the enough amount of nutrient supply originated from land water , and it caused drastic increase and dominance of Aurantiochytrium sp. NIBH N1-27-type thraustochytrids (utilizing H. akashiwo cells). The spike peak of AuRNAV detected in 2004 was considered as the result of virus infection to the dominant thraustochytrids. On the other hand, in 2005, land water supply was less than 2004 before the H. akashiwo bloom. Then, the bloom scale was not as large as in 2004 . Although the similar events should have occurred also in 2005, each event may have occurred at lower level, thus, the event sequence was not so obvious in 2005 compared to 2004. Another possibility is difference in the species composition of thraustochytrids. Untapped organic matter remained in water column and/or released organic matter derived from virally lysed NIBH N1-27-type thraustochytrid cells were considered to be substrates for multiplication of the other types of thraustochytrid. At the peak of thraustochytrids detected in 2005, Aurantiochytrium sp. NIBH N1-27-type may not have been as dominant as in 2004. Although we checked the total thuraustochytrids abundance to grasp the tendency of their dynamics, the resolution quality was too low to verify the hypothesis. Techniques for separately estimating the abundance of each genus or systematic groups of thraustochytrids is essential.