2. MATERIALS & METHODS
2.1. Study site
This study was performed at the Heukcheon stream of the Namhangang river in Yangpyoeng-gun, Kyeongki-do, Korea in March–April 2017. The experiment site was a small pond (30 m wide × 10 m long; maximum and mean depth of 1.5 m and 0.8 m, respectively) connected to the Heukcheon stream. The stream bed was comprised of silt and mud. The site had eight species of freshwater fishes, but only one bitterling species,Rhodeus pseudosericeus , and one mussel species, Unio douglasiae sinuolatus . R. pseudosericeus is considered an endangered species by the Ministry of the Environment of Korea, thus we received permission from the Ministry of Environment of Korea (Permission number, 2017-22) to perform this study.
2.2. Induction of R. pseudosericeus spawning
To study the development of the minute tubercles at each larval developmental stage and the position and migration of larvae in mussels, the induction of R. pseudosericeus spawning on mussels was performed simultaneously in large quantities. Mussels were collected on 5 March 2017, before the spawning period of R. pseudosericeus . A total of 150 mussels (shell length 35–70 mm) were collected using a kick net (mesh size 3 × 3 mm) and placed in the small pond where the spawning experiments would be conducted. The captured mussels were then placed on fine sand inside a plastic box (60 cm length × 60 cm width × 20 cm height) through which water could pass but not R. pseudosericeus individuals. The sealed box was placed in another pond next to the pond where the experiment would be performed. The spawning induction experiment was prepared on 27 March 2017, at night. The mussels were then divided into three boxes with 50 mussels each, and the boxes were places at 3-m intervals. On the morning of 29 March 2017, 36 hours later, the plastic box containing the mussels was removed to complete the spawning induction experiment. Water from the pond was collected in three plastic boxes (100 cm length × 100 cm width × 60 cm height), which were transferred to a laboratory with an oxygen generator.
2.3. Mussel rearing in the aquarium
In the laboratory, an experimental aquarium (60 cm width × 60 cm length × 60 m height) was prepared, and sand was evenly spread (10 cm height) on the bottom of the aquarium. For spawning induction, the three groups of 50 mussels were separately placed in three glass tanks. Oxygen was supplied so that dissolved oxygen (DO) was maintained above 7 mg/L, and aquarium heaters were used to keep water temperature around 20 ± 1°C. A natural 13:11 h light:dark photoperiod was used. The mussels were maintained in the experimental aquarium and fed daily with a liveChlorella sp. suspension derived from an indoor aquarium.
2.4. Observation of R. pseudosericeus larval development stage and position inside the mussels
After 1 day in the tanks, three mussels per day for 30 days were checked for R. pseudosericeus larvae. The presence of larvae on the four gills (left or right, outer or inner) of the U. d.sinuolatus mussels was checked by using a mussel-opening device that enabled mussels to be opened to approximately 1 cm. Mussels with spawn had their adductor muscle cut and were examined for the position, number, and developmental stage of larvae. Mussels without spawned were housed in different tanks.
To evaluate the changes in larval position in the mussels, the gills were divided into nine parts (Figures 1, 2); from the gill demibranch to its point of contact with the suprabranchial cavity, the gill was divided into lower part (L), middle part (M), and upper part (U); it was also divided into three parts in the other direction, 3 being the farthest from the outlet, followed by 2 and 1. Moreover, the larvae’s position was accurately recorded and photographed (Canon, Mark II, Japan) by measuring the transverse length of the siphon of the mussel and the longitudinal length from the suprabranchial cavity to the gill demibranch. The developmental stages of the R. pseudosericeuslarvae were determined under a stereoscopic microscope (Nikon, SMZ-10, Japan) using the AxioVision LE program (version 4.5, Carl Zeiss, Germany), following Kim et al. (2006).
2.5. Observation of the larvae’s minute tubercles
The development trend of the minute tubercles at each larval developmental stage was determined using scanning electron microscopy (SEM), and the height of the minute tubercles was measured. For the SEM analysis, three specimens at each stage of larval development were fixed for 24 h under carcodylate-buffered 2.5% glutaraldehyde, dehydrated in an ethanol graded series, and dried to a critical point with liquid CO2. The dried samples were sputter-coated with gold and then examined under SEM (Supra40VP, Carl Zeiss, Germany). For photographic documentation and assessment of the minute tubercles, a Carl Zeiss vision camera (LE REL. 4.4, Carl Zeiss, Germany) was used during SEM.
To facilitate the description of the distributional patterns, the surface of the larvae was divided into three regions (Figure 3) following Kim et al. (2008): (1) Anterior yolk sac projection covering eyes and head (hereafter referred to as EHR), (2) surface of wing-like projection composed of a pair of dorsal and one ventral yolk sac (hereafter referred to as WLP), and (3) posterior regions of yolk sac and most parts of the body including caudal fin-fold region (hereafter referred to as PR). During the 30 days of experiment, no dead mussels were found, and the minute tubercle heights were measured at each larval developmental stage. Thirty minute tubercles per region were measured from three regions per larva removed from the tank.
2.6. R. pseudosericeus utilization of host mussel
Host use of R. pseudosericeus was determined by recording the position of larvae within the four gill demibranchs and the number and frequency of larvae. To compare the size of mussels with and without larvae, mussel shell lengths were measured to the nearest 0.01 mm.
2.7. Statistical analyses
Statistical analyses were conducted using SYSTAT softwater (Systat version 18.0, SPSS Inc.,, Chicago, IL, USA). A two sample t-test was performed to compare the size of mussels with and without larvae. The Kruskal-Wallis H test was used to test the difference in number and frequency of larvae among different gill parts related to mussel size. Statistical significance was considered when P < 0.05.