Introduction
Appendicularians are pelagic tunicates that form an important component of marine planktonic communities worldwide; they are notable for their construction of elaborate mucus structures (or ‘houses’), which they use to collect small food particles. Paffenhöfer (Reference Paffenhöfer1973) first reported the successful cultivation of a neritic appendicularian species, Oikopleura dioica Fol, 1872, using a unique apparatus comprising a culture vessel mounted on a motor-driven frame, which was rotated in a circular orbit to gently agitate the water and suspend animals and their food in the water (Paffenhöfer, Reference Paffenhöfer1970). Subsequently, several additional methods and devices have been developed and applied to the cultivation of appendicularians. These include a rotating paddle (Fenaux and Gorsky, Reference Fenaux and Gorsky1985), water injection with a rotating paddle (Galt, Reference Galt and Strathmann1987), a temperature-gradient plankton kreisel (Gorsky et al., Reference Gorsky, Fenaux and Palazzoli1986), a diffusion tube (Patry et al., Reference Patry, Bubel, Hansen and Knowles2020), and improved rotating culture vessel (Sato et al., Reference Sato, Yu, Tanaka and Ishimaru1999). Because the rotating paddle device has a simple structure and is easy to construct, it has become indispensable in maintaining O. dioica cultures in the laboratory (Chioda et al., Reference Chioda, Eskeland and Thompson2002; Nishida, Reference Nishida2008; Bouquet et al., Reference Bouquet, Spriet, Troedsson, Otterå, Chourrout and Thompson2009; Martí-Solans et al., Reference Martí-Solans, Ferrández-Roldán, Godoy-Marín, Badia-Ramentol, Torres-Aguila, Rodríguez-Marí, Bouquet, Chourrout, Thompson, Albalat and Cañestro2015; Masunaga et al., Reference Masunaga, Liu, Tan, Scott and Luscombe2020). Consequently, significantly more ecological, physiological, and genetic information has been accumulated for this species of the Oikopleuridae family compared to other families of Appendicularia (Kowalevskiidae and Fritillariidae).
Fritillarid species are common in appendicularian assemblages and sometimes occur in high densities (Tokioka, Reference Tokioka1955; Shiga, Reference Shiga1985; Presta et al., Reference Presta, Hoffmeyer and Capitanio2015), however, laboratory investigations of their life cycles are limited. Paffenhöfer (Reference Paffenhöfer, Persoone and Jaspers1976) determined the generation time of Fritillaria borealis Lohmann, 1896 by conducting rearing experiments with his unique cultivation apparatus. Sato et al. (Reference Sato, Yu, Tanaka and Ishimaru1999) modified Paffenhöfer's instrumentation and cultivated Fritillaria formica digitata Lohmann in Lohmann & Bückmann, 1926 (Sato, Reference Sato2000) and Appendicularia sicula Fol, 1874 (Sato et al., Reference Sato, Yu, Tanaka and Ishimaru1999) through several life cycles to examine their generation time. Recently, Henriet et al. (Reference Henriet, Aasjord and Chourrout2022) reported a morphogenetic study of Fritillaria borealis typica Lohmann, 1896 cultured in the laboratory for several months using the rotating paddle method. Although these studies revealed some new information on the ecology and physiology of fritillarids, information on their biology remains limited.
In considering the cultivation of zooplankton species with highly fragile mucoid structures, the author speculated that any major disturbance or thrusting movement of the water by a rotating paddle blade or water injection at the inlet openings should be limited as much as possible because such hydraulic pressure could distort or twist mucilaginous constructions. The methods of Paffenhöfer (Reference Paffenhöfer1970) and Sato et al. (Reference Sato, Yu, Tanaka and Ishimaru1999), in which culture contents are circulated due to shearing between the water and the inside wall of the culture vessel, are suitably gentle and limit the destruction of the fragile structures. However, the former requires a large space and specific installation sites because the culture vessel and mounting flame move in a circle. This device may also be top-heavy at sites such as shipboard laboratories. Although Sato et al. (Reference Sato, Yu, Tanaka and Ishimaru1999) later mitigated these disadvantages by having the culture vessel rotate on a fixed axis in the same position, much labour and time are required to construct this apparatus, and due to the placement of the vessel on a rotating shaft, it is not suitable for larger culture vessels, as the weight can lead to breakage of the device.
Therefore, the author tried to improve on the available cultivation devices, using a novel method that circulates the water smoothly and gently without any oscillation and strong hydraulic pressure. At the time of this study, a species of the family Fritillariidae (Fritillaria sp.) was collected at a site near our laboratory and reared using the improved device. This study reports this new cultivation device as well as the growth, house morphology, and house renewal rate of this species.
Materials and methods
The device used in this study consisted of two units: a rotating polypropylene cylinder driven by a motor mounted on a metal frame pedestal, and a polycarbonate bucket that formed the culture vessel (Figure 1). The cylinder (12 cm in diameter × 13 cm in height) comprised a lidded 2 L polypropylene bottle, the bottom of which was removed using a heavy-duty cutter knife (Figure 1A). This was fastened directly onto the axis of a speed-controllable motor with a gear head (US206-401 and 2GN90 K, Oriental Motor Co. Ltd., Tokyo, Japan) (Figure 1B, C). A metal frame pedestal was made using a store-brought angle bar and a metal plate for assembling a steel rack or a metal gadget (Figure 1B). A bucket containing 9 L of seawater was set at an angle of 20° to facilitate lateral and vertical water movement. The cylinder-motor unit was attached to the top of the bucket, such that the cylinder was partially submerged in the seawater (Figure 1D), and rotated at 3–5 rpm.
Figure 1. Cultivation device: (A) cylinder and bottle cap. A pinion gear to hold the motor axis was glued onto the bottle cap with epoxy adhesive. Small holes were drilled around the pinion gear so that epoxy adhesive could extend to both sides of the cap. Four large holes in the lid form an air vent during experiments; (B) disassembled metal pedestal. The leg plates were cut to fit the edge of the bucket; (C) assembled cylinder-motor unit; (D) experimental setup with an angled polycarbonate bucket containing seawater. bc, bottle cap; pg, pinion gear; av, air vent; rc, rotating cylinder; mbp, main body of pedestal; m, motor; mc, motor clamp; lp, leg plate; gh, gear head; p, pedestal; sc, speed controller; pb, polycarbonate bucket.
The animals used in the experiment were collected from surface waters in Sagami Bay, off Misaki Marine Biological Station, School of Science, the University of Tokyo, Miura, Kanagawa Prefecture, Japan (35°16’N; 139°60’E). Surface water was sampled by submerging a 20 L polyethylene tank. In the laboratory, this was gently poured into a beaker illuminated from behind. The animals were sorted and transferred into the polycarbonate bucket filled with ambient seawater filtered through a 30-μm mesh using a wide-bore pipette. The transferred individuals were then maintained with setting the cylinder-motor unit to expand their houses and resume feeding. Individuals were transferred into fresh mesh-filtered ambient seawater every 0.5–2 days, depending on the density and size of the animals. The number of transferred individuals varied from dozens of premature individuals up to 200 individuals with trunk lengths of 200–300 μm.
At maturity, the oocyte-filled posterior region of the trunk could be seen with the naked eye. About 10 of these individuals were transferred to a 50 ml flask and agitated to release gametes and fertilize eggs. The contents of the flask were poured into the bucket of the cultivation device, and cultivation experiments were then conducted under a 12 L:12 D light cycle at three different temperatures (20, 23, and 26°C) with the cultivation device placed in an incubator. Every 24 h after fertilization, 5–11 individuals were sampled and fixed in a 5% buffered formaldehyde seawater solution, and their trunk lengths (the distance between the upper lip of the mouth and the posterior edge of the trunk) and tail lengths were immediately measured under a microscope. Some specimens were captured with a digital camera (Panasonic DMC-GH2) fitted on the microscope to facilitate morphological observation of the animal's body. The growth rate was calculated by examining the change in body weight during the cultivation period. The trunk length was converted into the ash-free dry weight using a conversion equation obtained for Fritillaria haplostoma Fol, 1872 (Hopcroft et al., Reference Hopcroft, Roff and Bouman1998). The trunk lengths of some mature individuals could not be measured because their gonads had ruptured and released eggs at the sampling time. Therefore, the relationship between tail length and trunk length was determined to obtain the trunk length of the ruptured animals. The instantaneous growth rate (g) was calculated as follows:
where Wm and W1 are the mean ash-free dry weights of individuals on the final cultivation day and one day after fertilization, respectively, and t is the generation time in days.
A single individual was also maintained in the culture vessel at 23°C for 12 h, and the discarded houses were collected every 1–2 h and counted to determine the house renewal rate. For morphological observations of the house, cultured alga (Tetraselmis sp.) was added to the water. The algae adhered to the outer membrane of the house and thus diminished house transparency and facilitated the observation of its shape and structure. The houses occupied by animals in the culture vessel were recorded using a digital camera (Canon Power Shot G16) in close-up and video mode. Discarded houses were also transferred into a Petri dish and photographed with the digital camera.
Results
The cultivated Fritillaria individuals (Figure 2) resembled Fritillaria formica tuberculata Lohmann in Lohmann & Bückmann, 1926 or F. haplostoma, considering the strongly bent compact trunk and dorsal oikoplastic layer located upstream of the stomach (Figure 2B) for the former, or the narrow tail musculature (Figure 2C) and large upper lip (Figure 2F) for the latter. However, they did not have the protruding lips with lateral lobes of the former or elongated cylindrical testis of the latter (Bückmann and Kapp, Reference Bückmann and Kapp1975; Fenaux, Reference Fenaux1993). Therefore, these individuals are provisionally regarded here as Fritillaria sp.
Figure 2. Fritillaria sp.: (A) right-side view of the whole animal; (B) left-side view of the trunk with the tail detached; (C) proximal portion of the tail showing narrow musculature. The tail fin was partially torn; (D) distal part of the tail showing the paired amphichordal cells on each side of the notochord tip; (E) dorsal view of the posterior trunk; (F) dorsal view of the mouth. hr, house rudiment; st, stomach; ov, ovary; te, testis; nt, notochord; mu, musculature; ac, amphichordal cell; ul, upper lip.
Fritillaria sp. produced an extraordinary mucilaginous and fragile house that was easily distorted or broken, even by moderate water turbulence or gentle prodding. However, the rotating cylinder of the novel cultivation device never damaged or distorted their houses, and the animals were maintained in good condition through more than ten filial generations.
The animals matured and released their gametes 4–5 days, 3–4 days, and 3 days after fertilization at 20, 23, and 26°C, respectively. The mean tail length reached 2137 ± 177 μm (± SD, n = 7) on the final day at 23°C (Figure 3) and was not statistically different from those of 11 individuals at 20°C and eight individuals at 26°C (Kolmogorov–Smirnov test for consistency with a normal distribution, 0.16 < D < 0.19, 0.40 < P < 0.81; Bartlett test for the equality of the error variances, χ2 = 2.44, P = 0.30; one-way ANOVA, F 2,23 = 3.42, P = 0.64). The relationships between tail length (Ta, μm) and trunk length (Tr, μm) were plotted at each temperature and found to overlap (Figure 4), and thus all data were combined to fit one regression equation in a form according to Shiga (Reference Shiga1976):
Figure 3. Changes in trunk length (open symbols) and tail length (solid symbols) of Fritillaria sp. over time after fertilization at 20°C (squares and solid lines), 23°C (triangles and dashed lines), and 26°C (circles and dotted lines). Trunk lengths include calculated values from tail length (see text and Figure 4) after mature individuals were first observed (indicated by arrows). Each point represents mean ± standard deviation.
Figure 4. Relationships between tail length (Ta) and trunk length (Tr) of Fritillaria sp. at 20°C (blue squares, n = 33), 23°C (green triangles, n = 22), and 26°C (red circles, n = 13). For regression, all data were combined and fitted to one power function, as shown in the graph. The line denotes the regression line for all data.
By applying this equation, the mean trunk length was estimated as 1026 ± 85 μm on the final day at 23°C. Instantaneous growth rates were calculated as 1.11, 1.34, and 1.72 at 20, 23, and 26°C, respectively, representing a 3.0–5.6-fold daily increase in body weight.
The house was typically barrel-like or cylindrical (Figure 5); however, some had a conical shape (Figure 5E), even when occupied. Houses of large individuals had longitudinal lengths of 7–10 mm. The house enveloped the entire animal body and food-concentrating filter and featured a funnel-like dent on one side, leading to a single water inlet just behind the occupying animal (Figure 5C, F). The elastic food-concentrating filter was alternatingly inflated with tail beating and deflated with tail arrest. The outer membrane of the house surrounding the animal and food-concentrating filter did not respond to the tail's movement (Supplementary video S1). The animals did not exhibit vigorous swimming with the house during feeding cycles, as was reported for Fritillaria pellucida (Busch, 1851) (Bone et al., Reference Bone, Gorski and Pulsford1979) and F. borealis (Flood, Reference Flood2003). The house renewal rate was estimated as 28 ± 2.8 houses d−1 at 23°C (n = 6).
Figure 5. House of Fritillaria sp.: (A) video frame of an occupied house with the animal undulating its tail, left-side view; (B) video frame of an occupied house with the animal resting its tail, right-side view; (C) video frame of an occupied house, view from the water-inlet side; (D, E) photographs of houses just after being abandoned; (F) schematic of an occupied house with the animal inflating food-concentrating filter. tr, trunk; ta, tail; ifcf, inflated food-concentrating filter; sfcf, shrunken food-concentrating filter; fd, funnel-like dent; wi, water inlet.
Discussion
Tokioka (Reference Tokioka1956) compared the taxonomic characters of ten forms of Fritillaria and proposed that they should be placed into sibling species forming a Fritillaria haplostoma-complex. Fritillaria sp. resembles F. haplostoma, especially in the trunk shape of small specimens from the lagoon waters of the Palao Islands (Tokioka, Reference Tokioka1955). The testis of Fritillaria sp. was roundish (Figure 2B, E), differing from the elongated cylindrical testis of F. haplostoma, although that of large individuals of the latter caught in the oceanic waters of the Palao Islands by Tokioka (Reference Tokioka1955) showed roundish objects resembling the testis in the posterior part of the trunk. The taxonomy of a specific group of the genus Fritillaria is unclear; thus, further studies, including taxonomical scrutiny of Fritillaria sp., F. haplostoma, and related species, are needed. Considering these facts, the laboratory cultivation of various forms of fritillarids will be beneficial to taxonomical differentiation by providing specific ecological and physiological information and materials for the genetic analysis of each species.
Cultivation device
The cultivation device with the rotating cylinder worked effectively during the rearing of this appendicularian species as it did not damage or distort the extremely delicate houses or induce animals to discard their houses. Because the culture vessel itself remained stationary, the device could be set up in an incubator or shipboard laboratory and the design could be applied to larger culture vessels, as needed. The large culture vessels will help to reduce the wall effect and contact with the water–air surface, to maintain a larger number of animals or species with a larger body or house, and to prevent rapid depletion of food with an adequate animal concentration in the vessel. We also cultivated other planktonic animals (R. Sato, pers. obs.). For example, the fritillarid species F. pellucida was cultivated through three filial generations, the doliolid Dolioletta gegenbauri (Uljanin, 1884) was cultivated through two full life cycles with alternation of asexual and sexual generations, and a thecosome pteropod species (probably genus Creseis) was maintained for one week up to spawning, exhibiting normal floating in the water by spreading its feeding net. These results indicate that this device could be effective for cultivating various zooplankton.
The rotating paddle method performed effectively in the cultivation of F. borealis typica (Henriet et al., Reference Henriet, Aasjord and Chourrout2022). This species produces a compact house relative to the animal's body size and swims with the deflated house after pumping water into the food-concentrating filter during its feeding cycle (Flood, Reference Flood2003). This indicates that the house of this form may be durable against any thrusting movement of the water caused by the rotating blade, resulting in the successful culture of the species with this device. It is not clear whether the continuous cultivation of Fritillaria sp. with its extremely fragile mucoid house is possible using other techniques such as the rotating paddle method or those mentioned in the introduction. However, the rotating cylinder device used in this study is perfectly suitable for cultivating this fragile fritillarid continuously. The abundance of different cultivation methods and devices will benefit the development of aquatic biology since the successful cultivation of a particular species is crucial to understanding the function of that species in the ecosystem, as this is difficult to observe directly in the natural environment (Paffenhöfer and Harris, Reference Paffenhöfer and Harris1979).
Growth and house of Fritillaria sp
The generation time of Fritillaria sp. was only several days, which is consistent with those of other appendicularian species such as O. dioica (Fenaux, Reference Fenaux1976; Paffenhöfer, Reference Paffenhöfer, Persoone and Jaspers1976) or some oceanic warm-water forms (Sato et al., Reference Sato, Yu, Tanaka and Ishimaru1999). A short generation time of 6–7 days was reported even for the cold-water species F. borealis typica at a low temperature of 12°C (Henriet et al., Reference Henriet, Aasjord and Chourrout2022). The instantaneous growth rates calculated here were also within the range of those obtained for fritillarid species incubated in microcosms at 28°C in situ (Hopcroft et al., Reference Hopcroft, Roff and Bouman1998). These appendicularian growth rates are higher than those of copepods (Huntley and Lopez, Reference Huntley and Lopez1992). Due to the short life cycle and high growth rate, appendicularian somatic productivity can sometimes be comparable with or higher than that of dominant crustacean zooplankton (Nakamura, Reference Nakamura1998; Tomita et al., Reference Tomita, Ikeda and Shiga1999). Therefore, Fritillaria sp. could play the role of a significant secondary producer, along with other appendicularian species.
According to Flood (Reference Flood2003), the compact house of F. borealis does not fully enclose the animal. By contrast, the Fritillaria sp. house reached lengths nearly ten times longer than the animal's trunk, and its outer membrane completely covered the animal. These morphological differences between their houses may reflect each species' behaviour and functional response. Contractility of the entire compact house of F. borealis without a non-deformable outer membrane may enable the animal to swim with the deflated house during its feeding cycle. Contrarily, Fritillaria sp. did not exhibit vigorous swimming with the house during feeding cycles as the particularly fragile large house could be damaged by such behaviour. Images of the house of Fritillaria rex Hopcroft and Robison, Reference Hopcroft and Robison2005, a large mesopelagic species, were recorded using a remotely operated vehicle and show the structure enveloping the animal and the food-concentrating filter, expressed as ‘a diffuse outer sphere’ (Hopcroft and Robison, Reference Hopcroft and Robison2005). This structure seems to correspond to the outer membrane of the Fritillaria sp. house. The house of F. borealis does not have the diffuse outer structure around the animal and the food-concentrating filter (Flood, Reference Flood2003). Thus, such a structure may be characteristic of the house of a distinct group of fritillarid forms. A single water inlet observed in the house of Fritillaria sp., which is different from the house of oikopleurid species with paired openings, was also described in F. borealis (Flood, Reference Flood2003), but it is not obvious in the in situ images of F. rex (Hopcroft and Robison, Reference Hopcroft and Robison2005). This conformity suggests that the single water inlet may be one of the characteristics of the Fritillaria house. However, we need to examine other fritillarid species since Flood (Reference Flood2003) proposed that the general house morphology of F. borealis belonging to the subgenus Eurycercus should not be applied to the entire genus, including the other subgenus Acrocercus, which is characterized by a distal part of the tail ending in a simple process which is not notched (Fenaux, Reference Fenaux1993; Shiga, Reference Shiga, Chihara and Murano1997) like that of Fritillaria sp. (Figure 2D).
The house renewal rate of Fritillaria sp. is one of the highest reported (Flood and Deibel, Reference Flood, Deibel and Bone1998; Sato et al., Reference Sato, Tanaka and Ishimaru2003). Sato et al. (Reference Sato, Tanaka and Ishimaru2003) suggested that those species producing houses with lower carbon content have higher house renewal rates. This indicates that the house of Fritillaria sp. may contain a lower amount of carbon, which may account for the severe fragility of the house. The larger size of the house relative to the animal could also contribute to its extreme fragility due to the reduced density of the materials per unit volume of the house. Despite the possible low carbon content of the house, this form may significantly contribute to the cycling and transport of organic matter in the marine environment as an active producer of macroscopic aggregates owing to the high house renewal rate.
In general, many fritillarid appendicularians have smaller bodies than the oikopleurids, except for large mesopelagic species (Hopcroft and Robison, Reference Hopcroft and Robison2005). The small size of the fritillarids may mean that they have a lower contribution to and impact on marine ecosystems when compared with oikopleurids. Additionally, the former has received less research attention, and there is a shortage of ecological and physiological information relating to them. The results of this study emphasize that the fritillarid appendicularians may be a more important producer and transporter of organic matter in marine ecosystems than what is currently proposed due to their high growth and house production rates.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0025315423000541.
Data
Raw data are available from the author upon reasonable request.
Acknowledgements
I thank the staff of Misaki Marine Biological Station, School of Science, the University of Tokyo, for their co-operation. I am grateful to Dr K. Hidaka, Japan Fisheries Research and Education Agency, for critically reading the manuscript.
Author's contributions
The author performed the entirety of this research.
Financial support
This work was supported by the Research Institute of Marine Invertebrates.
Competing interest
None.
