Toward selective breeding of a hermaphroditic oyster Ostrea chilensis: roles of nutrition and temperature in improving fecundity and synchrony of gamete release.
KEY WORDS: selective breeding, Ostrea chilensis, gonadosomic index, conditioning, reproduction
Selective breeding in Ostrea chilensis (New Zealand flat oyster) is hampered by the inability to apply routine methods such as strip-spawning and external larval rearing to achieve known crosses. Such methods are commonly used in the breeding of nonlarviparous oyster species such as the Pacific oyster Crassostrea gigas. To maximize the potential for a commercial selective breeding program with O. chilensis, this study aimed to elucidate some particularly challenging features of its reproduction: low female fecundity, simultaneous hermaphroditism, internal fertilization, and the inability to rear larvae externally prior to the early trochophore stage of development. A better understanding of reproductive physiology in this species is needed to simplify hatchery rearing and facilitate breeding programs.
Techniques for stimulating oogenesis and synchronizing gamete release are important in hatcheries to reduce brood stock maintenance costs and obtain larval cohorts. Lack of synchrony in gamete development can lead to overrepresentation of fertile individuals during mass matings, reducing the effective population size and thus genetic diversity, which has been shown to be a very real concern in the breeding of other flat oyster species such as Ostrea edulis (Bierne et al. 1998, Naciri-Graven et al. 2000, Launey et al. 2001, Lallias et al. 2010). The role of known environmental factors such as feed and temperature on gonadogenesis is of interest to commercial hatcheries, especially if it is possible to increase reproductive success through higher female sex ratios/ fecundity and by improving synchrony of gamete release.
The species Ostrea chilensis is native to both New Zealand and Chile. It is closely related to other commonly cultured flat oyster species such as Ostrea edulis and Ostrea angasi, but despite similarities with these species, it has a unique and perhaps more challenging reproductive physiology for hatchery production. In natural habitats, O. chilensis populations have unusually low fertility, with one study reporting only 10%-12% of individuals undergoing gametogenesis and even fewer brooding during the austral summer in New Zealand's Foveaux Strait (Cranfield & Michael 1989). In warmer waters, other studies report higher levels of brooding (Brown et al. 2010), but such levels are still below normal hatchery reproductive rates for other commercially produced flat oyster species, and exponentially lower than for oysters such as Crassostrea gigas. In addition to only few oysters being reproductive at any time, O. chilensis also produce far fewer eggs than any other commonly cultured flat oysters (~7,000-120,000 eggs per brood) (Jeffs & Creese 1996). Consequently, hatcheries are required to hold significant numbers of brood stock (costly), but must also simultaneously be vigilant to ensure desired larval yields while avoiding buildup of relatedness during mass matings.
The establishment of known parental pairs for selective breeding in this species is currently hampered by the inability to use the strip-spawning and external larval rearing methods routinely used in other nonbrooding species such as the Pacific oyster. By contrast, Ostrea chilensis is a larviparous protandric hermaphrodite that first achieves sexual maturity as a male, but subsequently the majority of individuals produce both male and female gametes, a feature commonly observed in other oviparous ostreids (Buroker et al. 1983). Although a few O. chilensis are single-sex, most contain both female and male gonad, and release gametes of both sexes simultaneously during a breeding cycle (Jeffs et al. 1996), previous studies of O. chilensis suggest that there is disproportionate investment in male gametogenesis, with most reproductive effort in this species being devoted to sperm production (Jeffs et al. 1996, Jeffs et al. 1997, Jeffs 1998). Sperm are released as spermatozeugmata continuously through the reproductive season (Jeffs 1999), a factor which opens up the potential for self-fertilization, or for sperm from different paternal lines to fertilize mature oocytes of the same female, thus compromising production of full-sibling families. Fertilization is internal and larvae brood in the mantle cavity (Chaparro et al. 1993). Larvae are liberated at the pediveliger stage (between 15 and 50 days, depending on water temperature) each with a well-developed foot and eyespot, settling on a substrate within 24-48 h postrelease (Cranfield & Michael 1989). No hatchery methods for external fertilization and larval rearing in this species have been demonstrated, nor are the authors aware of any ongoing breeding programs, despite commercial demand.
The efficiency of single-pair mating in selective breeding programs requires determination of reproductive sex, yet sexually mature specimens of Ostrea chilertsis are phenotypically identical and most are simultaneous hermaphrodites containing both male and female gonad. Hence there is no easy way to determine gamete contributions in parentage, as the type of gonad smears often used with gonochoric species are inappropriate for assessing sex in this species. Male and female gametes appear in the same follicle, even though different sex gametes may be at different stages of maturity. Progress has been made for rapid in vivo sex determination techniques using protein recognition in species such as Crassostrea gigas (Li et al. 2010), or noninvasive analysis with magnetic resonance imaging (MRI) (Davenel et al. 2006), as well as nuclear magnetic resonance fast field-cycling relaxometry (Davenel et al. 2010). Such techniques are expensive, however, and require complex equipment not commonly available in hatcheries. Thus, there remains no inexpensive way to differentiate sex or determine reproductive stage in O. chilensis without sacrificing specimens for histology.
The Ostrea chilensis species exhibit a cyclically skewed sex ratio and asynchronous gonadal development, both factors which decrease the effective breeding population size (Nb) (Beaumont 2010). Hence, a selective breeding program in O. chilensis requires either (1) large genetic diversity within a brood stock cohort (a costly route that requires feeding and maintaining large numbers of brood stock with potentially low reproductive success rates), or (2) an increase in the female sex ratio of the population. In a protandric species for which sperm release is continuous, the availability of sperm is not a limiting factor. At the same time, strategies to increase female gametogenesis are crucial to increasing breeding population size. In addition, enhanced synchrony in the timing of gamete release would permit more efficient fertilization and the subsequent rearing of tightly spaced larval cohorts.
The role of temperature on the gametogenic cycle of Ostrea chilensis in both natural and laboratory conditions has been well described (Walne 1963, Westerkov 1980, Buroker et al. 1983, Jeffs et al. 1997). Studies have established temperature thresholds for gonad development and spawning, as well as gamete and larval quality under varying feed regimes and have examined fecundity and sex ratios in O. chilensis from wild populations. Variable environmental conditions in natural populations or short-term studies in laboratory tanks make it difficult to extrapolate findings from these studies to the type of long-term control over gonadogenesis and gametogenesis that is possible in hatcheries. Although several papers already address selective breeding in this species (Toro & Newkirk 1991, Jeffs 1999), none of these studies have examined directly how control over environmental factors in a hatchery setting--such as control over the combined effects of both temperature and feed--can affect sex ratios and gametogenic synchrony in the breeding of O. chilensis. Such knowledge is important to hatcheries to determine how to best optimize reproductive cycles and timing of production for this species. This paper demonstrates that it is possible to influence brood stock sex ratios, to increase developmental synchrony, and to improve fecundity by manipulating simple, controllable exogenous factors during the breeding cycle. Such baseline information is crucial for establishing selective breeding programs for this protandric, hermaphroditic, and larviparous species.
A total of 1,570 brood stock were transferred from an oyster farm in the Marlborough Sounds (41[degrees] 14' S 174[degrees] 13' E) to the Cawthron Institute in Nelson, New Zealand on May 31, 2011. Here, oysters were divided into two treatments: one (n = 785) was placed in a high feed environment and held in an outdoor nursery where water temperature fluctuated with ambient air temperature ("nursery"); the other group (also n = 785) was maintained in strictly temperature- and feed-controlled conditions inside the hatchery ("conditioned"). Oysters in each of the treatments were held for 6 mo until December 13, 2011. At 3-wk intervals, 40-65 additional oysters were shipped from the same aquaculture site in the Marlborough Sounds, and samples of 40 individuals from the farm, as well as nursery and conditioned treatments, were examined at regular intervals for reproductive status using histological analysis.
Treatment I: Ambient Controlled Conditioning = "Nursery"
"Nursery" brood stock were held in flow-through systems, with seawater pumped from shallow outdoor ponds where natural phytoplankton blooms are sustained by the addition of nutrients. As photosynthetic rates increased in spring and summer, diurnal spikes in pH and temperature occurred within the nursery-feed ponds. The pH levels ranged from 7.6 to 9.1 (mean 8.3); normal diurnal air temperatures varied from 6[degrees]C to 21 [degrees]C. Temperature slow increased during spring, reaching a mean of 18[degrees]C on November 5, 2011. In this treatment, the oysters were exposed to broader fluctuations in temperature, pH, ammonia, nitrite, salinity, and dissolved oxygen than oysters held on the farm or in the controlled conditioning (treatment 2). Overall, however, this treatment provided both high levels of feed as well as a more rapid rise of conditioning temperature than farm conditions.
Treatment 2: Indoor Hatchery Brood Stock Tanks = "Conditioned"
Oysters (n = 785) in the conditioned treatment were held in flow-through brood stock tanks using temperature-controlled filtered seawater inside the hatchery. Oysters were fed an ad libitum diet of Isochrysis galhana and Chaetoceros calcitrans. Water temperatures were maintained at 10[degrees]C (which was initial temperature at translocation from the farm) for the first 2 mo, and then raised to 18[degrees]C over 2-wk period ending August 20, 2011. They were thereafter maintained at 17-18[degrees]C (above the spawning threshold in this species). As temperatures in the comparable nursery treatment did not reach 18[degrees]C consistently until November 5, 2011, the rapid and consistent increase of water temperature in the controlled conditioning provided a comparative with the nursery treatment. The peak temperature of conditioning in both nursery and conditioned treatments occurred several months before natural peak temperatures (e.g., temperature on farms in the region).
Reference Population = "Farmed"
New oysters (n = 800) from the same farm location as the original samples were shipped to the Cawthron Institute on each of the sampling dates for comparison with oysters sampled from the nursery and conditioned treatments.
Approximately 40 brood stock from each of the treatments were sampled every 3 wk over the 6-mo period. For each collection date (16 total), tissue samples for histology were taken randomly from 20 oysters per treatment. Both mature (50-75 mm) and juvenile (20-35 mm) oysters were sampled at each occasion. Because of ecomorphological variability, exact age could not be inferred from size. Farmed oysters in the juvenile category, however, were selected after the first year of production, whereas larger oysters had been held on the farm for more than 18 mo.
There are several means of assessing gametogenesis in bivalve molluscs, but in a hermaphroditic species such as Ostrea chilensis, histology is preferable over other methods because different sections of gonad can vary considerably in their sex and stage of development (Hollis 1962). Oysters were weighed (wet weight of visceral mass), measured, photographed, and the presence or absence of larvae or eggs in the mantle cavity was recorded prior to histological preparation. A visual gonad condition assessment was made, and brooding individuals were recorded, including stage of brooding in eggs (white sic, grey sic, black sic). For each sampling occasion, samples (n = 20) were fixed and 5-mm paraffin sections were prepared using standard hematoxylin and eosin (H&E) methods. Slides were examined by light microscopy to assess relative abundance of male and female products and stages of gametogenesis according to the gonad classification scheme in Jeffs et al. (1998). Sections were taken as much as possible so that the follicles in the inner portion of a gonad, which lie at an angle to the genital ducts into which they discharge their content, allowed for a transverse cut across the gonoducts to capture release of spermatozoa or oocytes.
In addition to staining with H&E, several other methods were evaluated for assessing gonadal development. For instance, a subset of random slides (n = 44), were prepared using Eosin-Alcian Blue-Neutral Red (trichrome), as well as Alcian Blue-Alcian Yellow, periodic acid-Schiff, and periodic acid-Schiff diastase. Results from these staining experiments suggest that the best technique for visualizing gonad in this species is a trichrome stain of Eosin-Alcian Blue-Neutral Red that makes it possible to discern reproductive cell maturity, particularly in sperm where Alcian Blue stained sperm precursor cells bright blue, spermatids red with Neutral Red, and egg material pink with Eosin. The intensity of Neutral Red stain associated with mature sperm indicates that the Eosin-Alcian blue-Neutral Red technique is useful for establishing the sperm from spermatids, which is possible, but more challenging with standard H&E.
Magnetic Resonance Imaging
It is not possible to determine sex without sacrificing Ostrea chilensis brood stock for histology because the sexes do not differ phenotypically and gonad smears are not reliable in a hermaphroditic species. Noninvasive imaging techniques like MRI and fast field-cycling nuclear magnetic resonance relaxometry have recently shown promise for examining reproductive state (Pouvreau et al. 2006, Flahauw et al. 2012, Smith & Reddy 2012). In related studies, they have also been useful in distinguishing sex of gonad in Crassostrea gigas (Davenel et al. 2006). The use of MRI is not previously reported in O. chilensis, and thus this method was evaluated on a sample of 10 brooding and nonbrooding individuals. Specimens were immersed in sea water to expel air bubbles, banded and scanned using a Siemens Avento imager operating at 1.5T (60 MHz) equipped with a wrist probe. Over the 6-mo time frame of this study, MRI assessment for a large number of samples was not feasible from a cost perspective. The technique was nevertheless evaluated for future effectiveness in assessing presence or absence of eggs or larvae in the mantle cavity, and compared with histological techniques for assessing gonad development stage and sex.
The effects of treatment were analyzed using one-way analysis of variance (ANOVA) and Fisher least significant difference post hoc test to examine significant differences between means of a female gonad condition index (GCI) (maturity and oocyte fraction). The female GCI, derived from histological examination by developmental classification, was compared for accuracy with proportions of brooders to nonbrooders in each treatment and was found to be related. One-way ANOVAs were also used to detect significant differences in sex of gonad between treatments.
Size and Weight Differences among Treatments
Mean shell size between farmed oysters and treatments did not differ at the [alpha] = 0.05 level of significance between the three treatments (ANOVA [F.sub.(1,5)] = 2.41, P = 0.104). Tissue wet-weight differed over time between the farmed and treatment oysters, but in relation primarily to the differential timing of brooding and subsequent decrease of condition rather than an overall indication of fitness. As wet-weight and condition are imprecise indicators of reproductive state, no results were based on wet weight and subsequent discussions in this study concentrate entirely on data derived from histological analysis of gonad.
From all samples, 96.1 % of the oysters examined histologically had visible signs of gonadal activity (eggs/sperm or gamete precursors). Overall, 3.9% of oysters were reproductively unproductive, with no significant difference between treatments at P > 0.05. In all but 0.1% of these individuals, specific pathogens were observed, including Bonamia exitiosa (2.1%) and Bucephalus longicornutus (1.7%), both of which degrade reproductive capacity (Hine & Jones 1994). Oysters infected with either of these pathogens, where gonad was largely unproductive, were excluded from further statistical analysis as the pathogens were clearly identifiable as the cause of reproductive failure, and would have skewed results. Apicomplexan infection was also observed (prevalence 49%), whereas Microsporidium rapuae and Rickettsiae were noted at low prevalence and intensity of infection, but with no discernable impact on reproductive effort.
Because Ostrea chilensis are facultative hermaphrodites, the female and male gonads are evident within the same gonad section, and often the same follicle (Fig. 1). Although MRI techniques clearly allowed for differentiation of brooding and nonbrooding individuals (Fig. 2), these techniques did not provide suitable visualization of sex of gonad or gonad ripeness as compared with histology sections.
Up to 25 mm, Ostrea chilensis were predominantly male (96% of specimens had >90% male gonad), regardless of environmental factors such as temperature or feeding regimen. Only 8% showed development of female germ cells and. in rare cases, eggs. Staining with Eosin-Alcian Blue-Neutral Red clearly distinguished mature sperm from precursors, and allowed us to ascertain definitively that mature sperm were being produced even during the winter period in oysters smaller than 25 mm.
Timing of Spawning
Oocytes were abundant in most oysters from all treatments by mid-October, at which time the size and appearance of the germinal vesicle, and the presence of oocytes were clear indications of advanced vitellogenesis. Although male gonad exhibited different phases of development within follicles of the same individual, female gametes were generally all in similar stages of development.
Early stages of female development were frequently observed in the same section of gonad containing late stage male products, thus suggesting that within an individual, male gametes are available before female products mature. Female gonad containing follicles with ripe oocytes were observed simultaneous to male gonad in a range of developmental stages, thus also supporting prior conclusions that males trickle spawn continuously. The release of sperm was observed even after brooding commenced. Total gonad coverage, however, including male follicles, became less abundant during brooding, suggesting that continued release of sperm was derived from existing precursors, but that no new development occurs during this period. In many cases, sperm were observed in early brooding individuals within the gonoducts, which is not physiologically impossible during brooding, as sperm may be carried on the exhalent stream without the adductor muscle contractions observed with egg release. Resorption of female gonad occurred shortly after brooding commenced and there was no evidence of new germinal cell development in late-stage brooding females. In all late-stage brooders, atresia was observed within the female gonad, a finding which supports Jeffs' (1998) conclusion that Ostrea chilensis are unlikely to initiate a second brooding cycle during the same spawning season.
Seasonal Reproductive Pattern
Sperm development was continuous throughout the winter period, but was markedly reduced, with male gonad production increasing rapidly again during spring. Oogenesis in the majority of specimens appeared to be suspended between May and September. Primary and secondary oocytes appeared to be inactive, rather than undergoing active development or atresia. A range of oocyte stages were present in all treatments by mid-September 2011, but brooding had not yet commenced (Fig. 3). By late-October, stages of development varied by treatment, and the first brooding individuals were recorded in treatment 1 (nursery).
At the final (December 13th) sampling, many oysters in both treatments were spent, with evidence of a drop in reproductive activity: follicle walls did not show active germinal layers for future gonad recuperation, cell membranes were ruptured, apparent oocyte wastes were dispersed in the lumen with significant lysosome activity associated with phagocytosis.
Most oysters exhibited simultaneous hermaphroditism, but did not exhibit synchronous male and female development. Subsequent to the protandric juvenile period, when most gonad is male, few (1%) oysters >25 mm across all treatments displayed exclusively male gonad or all female gonad. These sexually mature individuals may have been in transition between phases, or may have simply been expressions of a singular sex, based on genetic or environmental influences. As all individuals were sacrificed for histology, it is impossible to determine if some of the population are exclusively of the same sex throughout their adult lifetime, or whether some individuals of this species are sequential rather than simultaneous hermaphrodites such that the observed histology of single-sex gonad was merely a transitional phase. Differences in 100% female individuals between treatments existed, but were significant only between farmed and the two experimental treatments (ANOVA [F.sub.(2,21)] = 6.34, P = 0. 018). The amount of female gonad was remarkably consistent between the nursery and conditioned treatments (Fig. 4).
Proportion of male gonad, as measured by percentage of total gonad by weighted radius/density remained high even during early stages of brooding. Female gonad production (by weighted radius/density within GCI) was overall more than 53% of effort in the farmed oysters, 59% in the nursery treatment, and 61% in the conditioned treatment during the peak reproductive period from September 2 to December 13.
Not only did hermaphroditic individuals have nonsynchronous male and female gametogenesis, but there also appeared to be differential resorption of male and female gametic products. Resorption appeared to be a selective process, wherein developing spermatocytes and spermatids continued to be released into the lumen, and in many cases, large quantities of spermatozeugmata accumulated within gonoducts even after phagocytes were observed invading empty follicular spaces previously occupied by oocytes.
Synchrony of Oocyte Maturation and Brooding
With an ad libitum diet of Isochrysis galhana and Chaetoceros calcitrans, and a steady conditioning temperature of 18[degrees]C, the "conditioned" treatment exhibited tighter synchrony of oocyte maturation than those held in the hatchery nursery ponds, where feed was ad libitum but water temperature varied with ambient air temperature. The pattern in gonad maturation observed from the histology was confirmed by peaks in brooding, where conditioned oysters peaked several weeks prior to the nursery treatment (Fig. 3). Hatchery conditioned oysters, where temperature was much more tightly controlled than in the nursery, also exhibited a tighter curve in synchrony of brooding, with earlier initial brooding as well as earlier peak that tapered off quickly. In both nursery and conditioned treatments, brooding occurred over a much shorter period than the farmed group (Fig. 5). The difference in spread of the means in brooding between nursery (19.78 [m.sup.2] 2.1) and conditioned (24.25 [m.sup.2] 3.0) groups was ~3 wk, a really pronounced difference relative to much later reproductive activity in the farmed oysters. In the farmed group, water temperatures rose at a much slower and variable rate, with the oysters exhibiting correspondingly delayed gonad development. A much lower percentage of farmed oysters brooded overall (59%) during the 6-mo experimental period. Earlier elevated conditioning temperatures in the nursery and controlled treatments clearly accelerated the brooding cycle, as evident when comparing brooding peaks of the treatments relative to the farmed controls. From the farmed site, ~79% were found to be brooding larvae during the 6-mo experimental period, as compared with the much higher percentage brooding in the nursery and conditioned treatments. It is unclear, however, whether the total number of brooders was significantly different between the treatments and the farmed oysters, as sampling was terminated on December 13. In the hatchery (treatment 1), brooding had already peaked by the November 18th sampling date when the nursery (treatment 2) exhibited its maximum numbers of brooding oysters, and brooding appeared to be declining in both treatments when sampling was terminated. Of the samples taken from the farm in the Marlborough Sounds, the fact that fewer oysters had brooded by the final sampling occasion may not indicate lower overall reproductive activity, as many farmed oysters still possessed ripe gonad, thus indicating the potential for further spawning after the final date of the experiments. In natural populations, peak spat collection generally occurs in early January, a date that is also predicted by the curve in Figure 3.
The design of a selective breeding program requires not only identification of traits important for commercial production (e.g., growth and morphological characteristics, disease resistance), but also selection of a brood stock population that contains adequate genetic variability in these desired traits (Gaffney & Scott 1984, Boudry et al. 2004). Suboptimal fertilization rates, potentially leading to inbreeding depression, arise not only from small breeding populations, but from disparities in sex ratio and asynchronous gametogenesis (Gaffney 2006). Controlling these factors increases success rates, while decreasing costs of a breeding program. The total number of reproductive females (which, it can be assumed, is related to total productive female gonad proportion) within a brood stock population determines the potential effective population breeding size ([N.sub.b]). In a hermaphrodite species such as Ostrea chilensis, increasing the proportion of female gonad will in theory decrease the required population of brood stock oysters, thus reducing overall conditioning costs.
Research increasingly elucidates the role of genetic factors in controlling the steroids and neuropeptides that regulate reproductive pathways (Yusa 2007, Hedrick & Hedgecock 2010, Dheilly et al. 2012), thus it may be possible in the future to condition brood stock to be predominantly female, or to induce gametic maturation in oysters "on demand" through genetic or hormonal manipulation. Until this is possible, knowledge of how environmental factors such as temperature and feed affect the gametogenic cycle is important for hatcheries, particularly if better reproductive rates and synchrony of gamete release can be achieved.
Sex Ratios and Fecundity
The easiest way to create genetic diversity within the population is through controlled in vitro fertilization and external rearing of known parental crosses. As described, however, Ostrea chilensis are simultaneous hermaphrodites, wherein at any given sampling occasion, only a very small number of oysters are solely of one sex. Establishing desired crosses is more challenging when gametes of both sexes are released simultaneously, and there are few all-male or all-female specimens within a brood stock population. It is particularly difficult to breed known crosses if differentiating male and female specimens is impossible without sacrificing them for histological examination. In O. chilensis, there is thus no easy way to track parentage except through molecular marker-based pedigrees, which is too costly for most commercial hatcheries. Single-pair mating is possible (e.g., see experiments with Ostrea edulis by Newkirk & Haley 1983), but nonetheless remains inefficient as a large-scale breeding technique.
Mass mating remains the commercially viable option for breeding, thus necessitating creative strategies for tracing parentage and maximizing breeding population size. The proportion of female gonad within a population is one measure of reproductive potential in a larviparous species. Both Jeffs (1998) and Cranfield and Allen (1977) have indicated that male gonad forms the largest proportion of adult natural reproductive effort during the peak reproductive season, with 70%-90% of gonad releasing male gametes and only 10%-12% producing female gametes. By contrast, this study found that in all samples, including farmed oysters, the amount of female gonad activity was much greater than previously reported in natural populations. There are several possible explanations for this observation. Bottom-sampled oysters from natural populations that were not suspended in the water column may have experienced lower food availability and higher parasite loads than long-line farmed or hatchery-held oysters such as in this study. Density of bottom-sampled oysters is also likely to be lower, thus also potentially indicating density dependence in gonad development and by inference, also fertilization success rates. Probably, there is a relationship between feed availability and vitellogenesis. In a closely related species, Ostrea edulis, the role of feed and thus lipid availability is a key factor in female reproductive effort (Ruiz et al. 1992). On the one hand, it was not possible in the research design to differentiate the effect of temperature from feed on gonad development in the two treatments. In addition, because the feed in the natural, farmed environment was not measured, no comment can be offered on this comparative. On the other hand, it was discovered that no significant differences were observed in total female gonad proportions between the nursery and hatchery populations, even though synchrony of gametogenesis was earlier and slightly tighter in the hatchery (treatment 1) cohort. Nevertheless, in these experiments, a feed threshold may have been reached in both "nursery" and "conditioned" oysters such that even if excess feed was ingested, it did not lead to redirection of somatic growth toward reproductive effort.
The consistency in stages of female gonad development that were observed within an individual are easily explained by the fact fertilizable eggs must be discharged within a short period to ensure continuity in the developmental stage of brooded larvae. Sperm are also continuously produced in Ostrea chilensis, with different stages of development of spermatids even within the same individual, thus ensuring sperm are readily available throughout the breeding season. Despite this, strip-spawning of eggs, although theoretically possible, is not feasible without the possible introduction of self-sperm to the egg pool. External fertilization also requires external larval rearing of progeny for which a method has not been demonstrated in this species. An alternative method of isolating spawning females is therefore desirable for tracing parentage. Isolation with controlled pipette-based fertilization, however, does not seem to work as in a small subtrial of oysters maintained in isolation, unfertilized eggs were observed to undergo resorption in the mantle cavity when nonself-sperm were not present. It would be difficult to estimate timing of fertilization. Eggs likely became ripe, but without nonself-sperm cues to trigger egg release, deteriorate and are resorbed. Such findings suggest that self-fertilization potentially does not occur, but also suggests that isolation is not a solution for maintaining genetic lines unless a ready supply of nonself-sperm can be provided over time to induce oocyte development and release. Based on the fact that most specimens clearly exhibited trickle spawning of sperm even after eggs were being released, the possibility of self-fertilization cannot be definitely ruled out, especially because it is an area that requires further investigation using genetic methods. In this study, the research team also observed fertilized eggs in follicle ducts above the gill plates, thus implying that nonself-sperm are either passing through the ostia (not previously believed to be possible) or potentially that self-fertilization is occurring. There is no possibility, based on the location of these observations of fertilized embryos, that they were artifacts of the histological sampling methods or cross-contamination.
Exposure to nearby individuals of a specific sex is important for initiating gametogenesis in a number of bivalve species (Korpelainen 1990). Galtsoff (1940) demonstrated that the existence of chemical cues in Crassostrea virginica sperm influenced spawning in females, but found egg release to be independent of the concentration of sperm. In Ostrea chilensis, similar chemical cues appear to be implicated in synchrony of egg release. A disproportionate number of sperm were observed in the mantle cavities of ripe females, whereas sperm were in several cases observed swarming on the gill plates simultaneous to large egg masses passing through the gonoducts. Thus it is reasonable to assume that uptake of spermatozeugmata within the mantle cavity of O. chilensis may induce spawning through chemical cues. If so, the frequency and intensity of fertilization could well be a function of the spatial density in a breeding population. Despite the findings, it is not possible to conclude definitively about the influence of feed and temperature on female gonad development, as these factors may be only part of the picture. Indeed, lower reproductive rates in wild populations may be partly a function of lower densities of oysters (e.g., in benthic conditions or on farms), as opposed to the high densities of oysters in nursery/hatchery conditions.
Skewed sex ratios among hermaphrodite species generally result from factors that vary spatially within a patchy environment (e.g., food supply, parasitic stress, sex, and proximity of nearby oysters) (Buroker 1985). Advocates of such theories suggest that protandric species will invariably undergo sex reversal resulting in a predominance of femaleness among larger size classes because a male's ability to fertilize is independent of size, but available energy reserves in larger individuals will permit egg production (Buroker 1985, O Foighil & Taylor 2000). Bias in the overall sex ratio of a population increases reproductive success in low-density populations (Morton 1991). Generally, however, this bias is toward females rather than males, as energetic investments in oocyte production are generally greater than those for spermatozoa, particularly in a brooding larviparous species where reproductive investment is much greater than in oviparous females. The Ostrea chilensis are thus unusual, as unlike Ostrea edulis or Ostrea angasi, they appear to devote considerable effort to male gametogenesis and continuously trickle-spawn sperm even after brooding has commenced. Because free-swimming spermatozeugmata are relatively short lived, the continuous release of sperm through trickle-spawning ensures that sperm are always available whenever there is the potential for eggs to be fertilized (Foighil 1989).
The influence of environmental factors such as temperature, photoperiod, lunar cycles and tidal activity, salinity, and spatial density on gamete production is well explored under natural conditions, but research on how to control these factors in hatcheries is still limited (c.f. Jeffs & Creese 1996). The ability to manipulate easily controllable and known environmental variables such as feed and temperature is of particular interest to commercial hatcheries to control physiological development, especially if it is possible to tighten the synchrony of spawning for larval cohorts, and to increase reproductive success through higher female sex ratios and fecundity.
In temperate-zone bivalves such as Crassostrea gigas, variations in temperature appear to be the principal environmental factor affecting reproductive maturity phases, with the availability of food during preconditioning and conditioning an important secondary factor (Chavez-Villalba et al. 2003). Enriquez-Diaz et al. (2009) found that gametogenesis in C. gigas of the same origin, but cultured in different coastal locales, responded faster to increased seawater temperature, but that the intensity of the response was influenced by food availability. Similar results have been found in the European flat oyster Ostrea edulis, wherein spawning begins earlier in well-fed oysters, provided that a minimum temperature threshold is reached (Ruiz et al. 1992). Previous work in O. edulis indicates that preconditioning, and hence vitellogensis, is the period in which feed availability is likely to be most influential on fecundity (Joyce 2013). Earlier studies in Ostrea chilensis report mixed results in experiments on feed and temperature, but do not dismiss the interrelationship between these factors completely (Chaparro 1990, Toro & Morande 1998). The length of experimental period, and thus timing of feed delivery--whether during preconditioning, conditioning, or throughout the entire gametic cycle--will likely have an important influence on results from any of these types of studies.
Synchrony of Gametogenesis
Synchrony of oogenesis has a number of advantages for hatchery efficiency, leading to a larger breeding population size and more closely spaced cohorts of larvae for rearing purposes. Greater synchrony in egg release makes mass mating more efficient by increasing [N.sub.b], thus allowing for greater genetic diversity, whereas requiring shorter periods of maintenance for brood stock. Oysters in the hatchery or nursery conditions spawned earlier and more synchronously than oysters conditioned under less-controlled natural ("farmed") conditions, as reflected both in greater productivity (by oocyte size and percentage) and higher brooding rates in a shorter period.
In central New Zealand, mean winter seawater temperatures do not drop below 8[degrees]C and thus early-stage gametic products are present throughout the year. The research team observed that although lower water temperatures were correlated primarily with spermatogenesis, maturing oocytes were present throughout the winter study period and that the mean date for sperm maturation occurred before the mean date of oocyte maturation. Data on mean water temperature and spawning in natural populations are readily available from prior studies of oysters in Tasman Bay (Brown et al. 2010). Temperature thresholds are already used for routine conditioning of this species in commercial hatchery production. Temperature is known to be the controlling factor in gametogenesis, which begins as water temperatures approach 15[degrees]C (in the Tasman Sea, this occurs usually during November), and is followed by a peak in brooding generally observed around mid-December or early January (Jeffs & Creese 1996). This peak was brought forward by several weeks in the conditioning and nursery treatments as a result of artificially raising water temperatures earlier in the season (final maximum conditioning temperature was reached on October 8, 2011 in the conditioned treatment, and the nursery shortly thereafter on November 3, 2011).
Much higher numbers of individuals were brooding on the farm (suspended in the water column) and in the two treatments than previously reported for natural populations (Fig. 3). The significant difference between total brooding recorded for farmed oysters and conditioning treatments, however, should be interpreted with caution, as surface water temperature at the farm in the outer Marlborough Sounds finally reached 18[degrees]C at the end of the sampling period on December 13, 2011, whereas in the two hatchery treatments (nursery and controlled), temperatures had been above 18[degrees]C for almost 2 mo.
Rate of development and size of larvae are strongly influenced by temperature. Although the team did not calculate brooding times for individual oysters, the higher temperatures on average in the hatchery treatments may have led to a shortened brooding period. Brooding periods averaged 3 wk (as estimated based on overall difference in the peak of means between oocyte maturity and total brooders) rather than the 4-5 wk previously reported in colder Chilean waters. Larvae were also significantly smaller in the treatments than those reported in the Chilean studies where veligers are >450 [micro]m, whereas average size of larvae at release in these trials were ~230 [micro]m. As spawning events could not be readily recorded (only presence of brooding), the time from oocyte maturity to larval release is a rough estimate based on total brooding individuals overall, and thus it is impossible determine the exact effects of higher temperatures in correlation to shortened brood times. Nevertheless, results strongly indicate that percentage of female gonad, timing of gamete release, size of larvae, and length of brooding period are all temperature dependent.
A prerequisite for an efficient selective breeding program is to establish, as the research team did in this study, continuity in handling of brood stock with stringent control of temperature to optimize management of environmental influences that affect synchrony during the reproductive cycle. Data from this study suggest that environmental factors such as temperature, and most likely also feed, strongly influence femaleness during gonad development, and subsequently, the proportion of brooding individuals (density dependence was not tested, but should be considered in further research). Tighter control over temperature, and a more rapid increase in spring temperature, increases overall egg production with earlier and more synchronous egg release. Tighter synchrony has a number of advantages in hatcheries as it results in greater effective breeding population sizes, while reducing costs through improved efficiency in batch production, both factors that are valuable in hatchery production and will aid in developing a selective breeding program for this species.
Beaumont, A. 2010. Biotechnology and genetics in fisheries and aquaculture. Chichester, UK: Wiley-Blackwell. 216 pp.
Bierne, N., S. Launey, Y. Naciri-Graven & F. Bonhomme. 1998. Early effect of inbreeding as revealed by microsatellite analyses on Ostrea edulis larvae. Genetics 148:1893-1906.
Boudry, P., L. Degremont, N. Taris, H. McCombie, P. Haffray & B. Ernande. 2004. Genetic variability and selective breeding for traits of aquacultural interest in the Pacific oyster (Crassostrea gigas). Bull. Aquacult. Assoc. Can. 104:12-18.
Brown, S., S. Handley, K. Michael & D. Schiel. 2010. Annual pattern of brooding and settlement in a population of the flat oyster Ostrea chilensis from central New Zealand. N. Z. J. Mar. Freshw. Res. 44:217-227.
Buroker, N., P. Chanley, H. Cranfield & P. Dinamani. 1983. Systematic status of two oyster populations of the genus Tiostrea from New Zealand and Chile. Mar. Biol 77:191-200.
Buroker, N. 1985. Evolutionary patterns in the family ostreidae: larviparity vs. Oviparity. J. Exp. Mar. Biol. Ecol. 90:233-247.
Chaparro, O. 1990. Effect of temperature and feeding on conditioning of Ostrea chilensis Philippi, 1845. Aquacult. Fish. Manage. 21:399-406.
Chaparro, O.. R. Thompson & J. Ward. 1993. In vivo observations of larval brooding in the Chilean oyster, Ostrea chilensis Philippi, 1845. Biol. Bull. 185:365-372.
Chavez-Villalba, J.. J. Cochard, M. Le Pennec, J. Barret, M. Enriquez-Diaz & C. Caceres-Martinez. 2003. Effects of temperature and feeding regimes on gametogenesis and larval production in the oyster Crassostrea gigas. J. Shellfish Res. 22:721-731.
Cranfield. H. & R. Allen. 1977. Fertility and larval production in an unexploited population of oysters Ostrea lutaria Hutton, from Foveaux Strait. N. Z. J. Mar. Freshw. Res. 11:239-253.
Cranfield, H. & K. Michael. 1989. Larvae of the incubatory oyster Tiostrea chilensis (Bivalvia: Ostreidae) in the plankton of central and southern New Zealand. N. Z. J. Mar. Freshw. Res. 23:51-60.
Davenel, A., R. Gonzalez, M. Suquet, S. Quellec & R. Robert. 2010. Individual monitoring of gonad development in the European flat oyster Ostrea edulis by in vivo magnetic resonance imaging. Aquaculture 307:165-169.
Davenel, A., S. Quellec & S. Pouvreau. 2006. Noninvasive characterization of gonad maturation and determination of the sex of Pacific oysters by MRI. Magnetic Resonance Imaging 24:1103-1110.
Dheilly, N., C. Lelong, A. Huvet, K. Kellner, M.-P. Dubos, G. Riviere, P. Boudry & P. Favrel. 2012. Gametogenesis in the Pacific oyster Crassostrea gigas: a microarrays-based analysis identifies sex and stage specific genes. PLoS One 7:e36353.
Enriquez-Diaz, M., S. Pouvreau, J. Chavez-Villalba & M. Le Pennec. 2009. Gametogenesis, reproductive investment, and spawning behavior of the Pacific oyster Crassostrea gigas: evidence of an environment-dependent strategy. Aquacult. Int. 17:491-506.
Flahauw, E., S. Quellec, A. Davenel, L. Degremont, S. Lapegue & P. Hatt. 2012. Gonad volume assessment in the oyster Crassostrea gigas: comparison between a histological method and a magnetic resonance imaging (MRI) method. Aquaculture 370-371:84-89.
Gaffney, P. M. 2006. The role of genetics in shellfish restoration. Aquatic Living Res. 19:277-282.
Gaffney, P. & T. Scott. 1984. Genetic heterozygosity and production traits in natural and hatchery populations of bivalves. Aquaculture 42:289-302.
Galtsoff, P. 1940. Physiology of reproduction of Ostrea virginica: stimulation of spawning in the male oyster. Biol. Bull. 78:117-135.
Hedrick, P. & D. Hedgecock. 2010. Sex determination: genetic models for oysters. J. Hered. 101:602-611.
Hine, P. & J. Jones. 1994. Bonamia and other aquatic parasites of importance to New Zealand. N.Z. J. Zool. 21:49-56.
Hollis, P. J. 1962. Studies on the New Zealand mud-oyster Ostrea lutaria Hutton 1873. Wellington, New Zealand: Zoology Department, Victoria University. 167 pp.
Jeffs, A. 1999. The potential for developing controlled breeding in the Chilean oyster. Aquacult. Int. 7:189-199.
Jeffs, A. G. 1998. Gametogenic cycle of the Chilean oyster, Tiostrea chilensis (Philippi, 1845), in North-Eastern New Zealand. Invertebr. Reprod. Dev. 34:109-116.
Jeffs, A. & R. Creese. 1996. Overview and bibliography of research on the Chilean oyster Tiostrea chilensis (Philippi, 1845) from New Zealand waters. J. Shellfish Res. 15:305-311.
Jeffs, A., R. Creese & S. Hooker. 1996. Annual pattern of brooding in populations of Chilean oysters, Tiostrea chilensis (Philippi, 1845) from Northern New Zealand. J. Shellfish Res. 15:617-622.
Jeffs, A., R. Creese & S. Hooker. 1997. The potential for Chilean oysters, Tiostrea chilensis (Philippi, 1845), from two populations in northern New Zealand as a source of larvae for aquaculture. Aquacult. Res. 28:433-441.
Joyce, A. 2013. Neuroendocrine control of sexual differentiation and gamete maturation in oysters. J. Shellfish Res. 32:213-221.
Korpelainen, H. 1990. Sex ratios and conditions required for environmental sex determination in animals. Biol. Rev. 65:147-184.
Lallias, D., P. Boudry, S. Lapegue, J. W. King & A. R. Beaumont. 2010. Strategies for the retention of high genetic variability in European flat oyster (Ostrea edulis) restoration programmes. Conserv. Genet. 11:1899-1910.
Launey, S., M. Barre. A. Gerard & Y. Naciri-Graven. 2001. Population bottleneck and effective size in bonamia ostreae-resistant populations of Ostrea edulis as inferred by microsatellite markers. Genet. Res. 78:259-270.
Li, Y., G. Siddiqui & G. Wikfors. 2010. Non-lethal determination of sex and reproductive condition of Eastern oysters Crassostrea virginica gmelin using protein profiles of hemolymph by protein chip and SELDI-TOF-MS technology. Aquaculture 309:258-264.
Morton, B. 1991. Do the bivalvia demonstrate environment-specific sexual strategies? J. Zool. 223:131-142.
Naciri-Graven. Y., S. Launey, N. Lebayon, A. Gerard & J. P. Baud. 2000. Influence of parentage upon growth in Ostrea edulis: evidence for inbreeding depression. Genet. Res. 76:159-168.
Newkirk, G. & L. Haley. 1983. Selection for growth rate in the European oyster, Ostrea edulis: response of second generation groups. Aquaculture 33:149-155.
O Foighil. D. 1989. Role of spermatozeugmata in the spawning ecology of the brooding oyster Ostrea edulis. Gamete Res. 24:219-228.
O Foighil. D. & D. J. Taylor. 2000. Evolution of parental care and ovulation behavior in oysters. Mol. Phylogenet. Evol. 15:301-313.
Pouvreau, S., M. Rambeau. J. C. Cochard & R. Robert. 2006. Investigation of marine bivalve morphology by in vivo MR imaging: first anatomical results of a promising technique. Aquaculture 259:415-423.
Ruiz, C., D. Martinez, G. Mosquera, M. Abad & J. L. Sanchez. 1992. Seasonal variations in condition, reproductive activity and biochemical composition of the flat oyster, Ostrea edulis, from San Cibran (Galicia, Spain). Mar. Biol. 112:67-74.
Smith, P. T. & N. Reddy. 2012. Application of magnetic resonance imaging (MRI) to study the anatomy and reproductive condition of live Sydney rock oyster, Saccostrea glomerata (Gould). Aquaculture 334-337:191-198.
Toro, J. & P. Morande. 1998. Effects of food ration and temperature on length of brooding period, larval development and size of pediveligers released in the Chilean oyster Ostrea chilensis. J. World Aquacult. Soc. 29:267-270.
Toro, J. E. & G. F. Newkirk. 1991. Response to artificial selection and realized heritability estimate for shell height in the Chilean oyster Ostrea chilensis. Aquat. Living Resour. 4:101-108.
Walne, P. 1963. Breeding of the Chilean oyster (Ostrea chilensis Philippi) in the laboratory. Nature 197:676.
Westerkov, K. 1980. Aspects of the biology of the dredge oyster Ostrea lutaria Hutton, 1873. Dunedin, New Zealand: Department of Marine Science, University of Otago. 192 pp.
Yusa, Y. 2007. Causes of variation in sex ratio and modes of sex determination in the Mollusca: an overview. Am. Malacol. Bull. 23:89-98.
A. JOYCE, (1) * S. WEBB, (2) H. MUSSELY, (2) K. HEASMAN, (2) A. ELLIOT (3) AND N. KING (2)
(1) Department of Biological and Environmental Sciences, University of Gothenburg, Sven Loven Centre for Marine Sciences, Stromstad, Sweden; (2) Cawthron Institute, 98 Halifax Street East, Nelson 7010, New Zealand; (3) Kono Seafood, Wakatii House Montgomery Square, Nelson, New Zealand
* Corresponding author. E-mail: firstname.lastname@example.org
|Printer friendly Cite/link Email Feedback|
|Author:||Joyce, A.; Webb, S.; Mussely, H.; Heasman, K.; Elliot, A.; King, N.|
|Publication:||Journal of Shellfish Research|
|Date:||Dec 1, 2015|
|Previous Article:||Exploring restoration methods for the Olympia oyster Ostrea lurida carpenter, 1864: effects of shell bed thickness and shell deployment methods on...|
|Next Article:||Development of a DNA microarray-based identification system for commercially important Korean oyster species.|