|
|
||||||||
1 School of Marine Sciences, Darling Marine Center, University of Maine, Walpole, Maine 04573
2 Marine Sciences Center, University of New England, Biddeford, Maine 04005
* To whom correspondence should be addressed. E-mail: sherij{at}maine.edu
| Abstract |
|---|
|
|
|---|
16 to 26 h), are able to fertilize eggs at extremely low external sperm concentrations (ca. 101 sperm ml1), and have a longevity that varies with temperature. It is possible that such prolonged sperm longevity may be achieved by reductions in motility, reactivation of quiescent sperm by chemical cues, or intermittent swimming. | Introduction |
|---|
|
|
|---|
In contrast to egg-broadcasters, many free-spawners that retain eggs for internal fertilization (i.e., egg-brooders) may use filter-feeding or suspension-feeding mechanisms to concentrate dilute sperm from the water throughout a prolonged period of egg viability (Yund, 2000; Pemberton et al., 2003b). If females can overcome direct dilution effects and obtain sperm from distant males, adaptations that increase sperm longevity (e.g., inactivity, intermittent swimming, and activation by chemical cues) may be important in fertilization in the field. Sperm longevity is thought to enhance fertilization in some internally fertilizing ascidians (Bishop, 1998; Pemberton et al., 2003b) and bryozoans (Manríquez et al., 2001).
Because many marine invertebrates spawn under a wide range of temperatures (Andronikov, 1975), thermal effects may have important consequences for gamete longevity. A few studies have examined the effects of temperature on sperm viability (Andronikov, 1975; Greenwood and Bennett, 1981; Manríquez et al., 2001), but effects on longevity have received little attention. Temperature has a direct effect on all metabolic processes (Hochachka and Somero, 1984); hence higher temperatures might be expected to reduce longevity (Greenwood and Bennett, 1981). However, increased temperature also decreases water viscosity (Dorsey, 1968; Jumars et al., 1993), which can profoundly affect the energy requirements of small organisms (such as sperm) that swim in a low Reynolds number environment (Denny, 1993; Podolsky and Emlet, 1993; Fuiman and Batty, 1997; Hunt von Herbing, 2002). If sperm are subject to both direct metabolic effects and viscosity effects, the overall thermal effects on longevity and fertilization may be complex.
In this study, we use the colonial ascidian Botryllus schlosseri (Pallas, 1766) as a model system to experimentally investigate the effects of time, temperature, and mixing on the viability of naturally released sperm. Physical disturbance created by mixing might alter longevity patterns by inducing or suppressing swimming or by physically damaging gametes (Mead and Denny, 1995; Denny et al., 2002). Botryllus schlosseri has been the subject of several laboratory and field fertilization studies (e.g., Grosberg, 1991; Yund and McCartney, 1994; Yund, 1995, 1998; Stewart-Savage et al., 2001). Released sperm exit the colony through the exhalent siphon and rapidly disperse into the surrounding seawater (Milkman, 1967). Sperm are then acquired by another colony, and fertilization takes place internally. Botryllus schlosseri is a geographically widespread species (Grosberg, 1988), so colonies experience a wide range of temperature regimes during the reproductive season (e.g., Yund and Stires, 2002).
| Materials and Methods |
|---|
|
|
|---|
Colonies of B. schlosseri were collected from the Damariscotta River, Maine. Animals were grown on glass microscope slides (7.6 x 5.1 cm) in the flowing seawater system at the University of Maines Darling Marine Center and monitored. When colonies with at least 20 eggs were approaching takeover (late stage 5 through early stage 6; Milkman, 1967), they were isolated (at least 12 h before siphon opening) in 50 ml of unfiltered sperm-free seawater (aged at least 7 days in the dark at 15 °C) at 15 °C and fed phytoplankton (Isochrysis or Tetraselmis sp.) at densities of approximately 105 cells ml1. Water and food were changed daily. Colonies were determined to have completed siphon opening when phytoplankton were present in the digestive tract, indicating that eggs had been ovulated, and these colonies were subsequently treated as virgin females for the experiment.
Sperm collection and aging
For each experimental trial, naturally spawned sperm were obtained by isolating 24 colonies in 200 ml of sperm-free seawater (see above) for 4 h during peak sperm release (Stewart-Savage and Yund, 1997). A 15-ml sample of the undiluted solution was collected to determine the sperm concentration. Sperm were then quickly diluted 100-fold: 160 ml of mixed sperm suspension were added to 16. 1 of aged seawater (15 °C) and mixed. We immediately collected a sample and used it to obtain an initial fertilization level. The remaining suspension was then aliquotted into four glass containers, each containing 3.8 1. The aliquots were then aged under experimental conditions, consisting of all combinations of two temperatures (15 °C and 22 °C) and two mixture treatments (mixed and unmixed). Mixed suspensions were constantly stirred with a magnetic mixer. The 15 °C and 22 °C temperatures were selected to represent the range under which spawning normally occurs in the Damariscotta River estuary (Yund and Stires, 2002). Each container was sampled at 8, 24, 48, and 72 h. We sampled two areas of the unmixed container. One sample was removed from the top, and then a second sample was carefully pipetted off the bottom while minimizing disturbance to the solution. By contrast, the sample from the mixed solution integrated different heights within the container. This sampling scheme yielded three experimental treatments: top and bottom of the unmixed solution, and the mixed solution. Differences between the top and bottom treatments might reveal aspects of sperm behavior (e.g., buoyant or upward-swimming sperm vs. immotile sinking or downward-swimming sperm), while the mixed treatment tests the effects of physical disturbance and aeration on sperm viability.
Sperm counts were determined by concentrating a fresh aliquot (13.5 ml) of the undiluted sperm suspension by two orders of magnitude through centrifugation. Polyvinyl alcohol (Sigma-Aldrich; 1.5 ml of a 1 g ml1 solution) was added to the sperm solution before centrifugation to prevent sperm from adhering to container surfaces. The concentrated suspension (150 µl) obtained by centrifugation was fixed with 25% gluteraldehyde and subsequently counted on a hemacytometer (three samples, two counts each). Raw hemacytometer counts were multiplied by a dilution factor to obtain the concentration of the aging and fertilization solution. Because of the low concentration of B. schlosseri sperm, all 25 squares of the hemacytometer had to be scored to obtain non-zero counts. Consequently, our values are probably accurate only to the nearest order of magnitude.
Fertilization assays
All fertilization assays were conducted in 50-ml chambers with virgin female colonies. At each time point, 40 ml of each experimental solution was transferred into the chambers along with 5 ml (105 cells ml1) of algae. The fertilization assays were then incubated at the relevant experimental temperature for 4 h. At the end of the incubation, colonies were rinsed with aged seawater to terminate the collection of sperm and were returned to isolation. All colonies were maintained at a constant temperature to standardize development times. After 1620 h, an incision was made in each zooid, and the unfertilized eggs and developing embryos were counted with the aid of a stereomicroscope. Results are expressed as percent fertilization. A total of 14 trials were conducted over two reproductive seasons. Sperm-free controls were not included, because colonies were isolated in sperm-free seawater well in advance of siphon opening. In addition, stage of embryo development was assessed as a control for the timing of fertilization, to ensure that fertilizations were the product of the sperm we added. Some of the embryos in six female colonies were too far along in development (i.e., >24 h), and so must have been the product of fertilization by contaminating sperm prior to our experiment. Data from these individuals were excluded from the analysis.
Statistical analysis
All percent fertilization data were arcsine transformed to meet normality assumptions. A few fertilization assays were omitted during the experiment, due either to the death of experimental animals or to a shortage of virgin females at appropriate time points. Consequently, fertilization data were analyzed with a software package that accommodates an unbalanced design (Proc GLM ver. 6.07, SAS Institute, Cary, North Carolina). A repeated-measures ANOVA with trial nested within treatment and temperature was conducted. The effects of temperature and treatment were tested against the nested term, and all effects involving time were tested against the residual error. Type III sums of squares are reported for all sources of variation. Significance was determined at the 5% level. An additional multi-way ANOVA assessed whether egg number differed among times, temperatures, and treatments to test whether effects in the main analysis could have been an artifact of variation in egg number.
Sperm half-life
Two sperm half-life estimates were generated by treating time as a continuous variable. For each temperature, a logarithmic regression of mean percent fertilization (not transformed), averaging across all treatments, was conducted with time as the dependent variable. Sperm half-lives were calculated from these equations by solving for the time at which fertilization dropped to half of the initial (0 h) value. Estimates obtained from our data were compared to previously published half-lives of other free-spawning marine invertebrates. We used data previously reviewed by Manríquez et al. (2001), but included published values for other egg-brooding taxa.
| Results |
|---|
|
|
|---|
|
|
Sperm half-lives
Logarithmic regressions of fertilization with time were highly significant for both temperatures (15 °C, r2 = 0.967, P < 0.01; 22 °C, r2 = 0.790, P < 0.05). Sperm half-lives (the time at which fertilization drops to 50% of its initial level) were estimated at both temperatures, using the following logarithmic regression equations: 15 °C, % fertilization = 29.369 log (time) + 72.291; 22 °C, % fertilization = 31.375 log (time) + 68.456. The resulting half-lives were 26.3 h at 15 °C and 16.1 h at 22 °C (Fig. 2).
|
|
| Discussion |
|---|
|
|
|---|
Williams and Bentley (2002) reported similar sperm longevity in a free-spawner (the polychaete Arenicola), but only at a sperm concentration four orders of magnitude higher than the 101 sperm ml1 reported here. High longevity of concentrated sperm is predicted by the respiratory dilution effect (Chia and Bickell, 1983), but the combination of high longevity and very low concentration has not previously been reported. In addition, Arenicola is an egg-broadcasting species that retains its eggs in a burrow and pumps sperm-laden water past them (Williams et al., 1997). Consequently, it appears to function much like an egg-brooder (Williams and Bentley, 2002). Prolonged motility has been observed in two species of deep-sea echinothuriid sea urchins, Araeosoma fenestratum and Sperosoma antillense (Young, 1994); but sperm concentrations were not reported and fertilization was not assayed. The presence of lipid stores attached to the mitochondria of these sperm may enable the sperm to swim for prolonged times (Eckelbarger et al., 1989; Young, 1994). However, the role of lipid stores in swimming has been questioned (Eckelbarger, 1994).
The only free-spawners known to have comparable sperm longevity at low concentration (Fig. 3) are another brooding ascidian, Diplosoma listeranum (Bishop, 1998), with an estimated sperm half-life of 8 h at 101 sperm ml1, and a brooding bryozoan, Cellaporella hyalina, with an estimated sperm half-life of 1 h at 101102 sperm ml1 (Manríquez et al., 2001). This pattern suggests that sperm longevity may be consistently higher in egg-brooding invertebrates than in egg-broadcasters. However, this relationship may break down when more taxa are studied. Some brooders lack any apparent mechanism to efficiently collect dilute sperm from distant males via a filter-feeding or suspension-feeding mechanism (e.g., brooding corals).
Although the other two brooding taxa have dilute sperm longevities greater than those of the broadcasting species, the longevities of D. listeranum and C. hyalina are nevertheless somewhat lower than in B. schlosseri. In contrast to these two species, B. schlosseri lacks the ability to store sperm (Stewart-Savage et al., 2001). D. listeranum and C. hyalina can both store sperm for several months (Bishop and Ryland, 1991; Bishop and Sommerfeldt, 1996) and weeks (Manríquez, 1999), respectively. In the absence of a mechanism to store sperm, extended sperm longevity in B. schlosseri may represent an alternative adaptation to further enhance fertilization.
The high longevity of sperm in active filter-feeding or suspension-feeding brooders raises the question of how very dilute sperm can remain viable for such an extended period of time. A sperms life span is based on its consumption of energy reserves (usually phospholipids; Harumi et al., 1990), which is a function of the amount of energy consumed for motility. If B. schlosseri sperm are taken up passively through filter-feeding mechanisms, sperm may need to swim only short distances (i.e., within the maternal zooid) to reach an egg. Consequently, longevity would be prolonged if sperm were relatively inactive while in the water column, and then activated within the maternal zooid. The activation of dilute, inactive sperm by egg exudates has been reported in some broadcast-spawning solitary ascidians (Miller, 1974; Bolton and Havenhand, 1996; Jantzen et al., 2001) and in abalone (Riffell et al., 2002). Similarly, packets of bryozoan sperm alter their flagellar waveforms when released, thus probably saving energy and enhancing longevity, but increase the generation of waveforms to enter the maternal zooids (Temkin and Bortolami, 2004). B. schlosseri sperm might also conserve energy through intermittent swimming after release, as suggested for another colonial ascidian (Bishop, 1998). The continued presence of sperm in our top sample (Fig. 1A) after many hours supports this interpretation, but might also have been due to convection currents within the sperm-aging containers. We have never observed much activity in B. schlosseri sperm obtained from testes macerates, but it is possible that sperm are activated prior to natural release. Mixing had no effect on longevity (Table 1), so physical disturbance apparently does not activate sperm.
Temperature effects
This study suggests that temperature has a small but significant influence on longevity. The estimated half-life for sperm aged at 15 °C was 10 h longer than for those kept at 22 °C. B. schlosseri is found as far north as Newfoundland and south to North Carolina, and so experiences a wide range of temperature regimes (Pollock, 1998). Within the Damariscotta River estuary, water temperatures during spawning season can be as high as 22 °C at landward sites and as low as 13 °C in seaward sites (Yund and Stires, 2002).
Temperature directly controls metabolic activity in all cells, which influences energy consumption and survival (Hochachka and Somero, 1984). Due to temperature compensation, changes in sperm viability are expected to be minimal in eurythermal species that routinely spawn over a wide range of temperatures (Andronikov, 1975; Manríquez et al., 2001). However, actual fertilization levels have been shown to decrease with increased temperature, due to mechanical damage or exhaustion of energy reserves (in a temperate sea urchin; Greenwood and Bennett, 1981). But any strictly metabolic approach to understanding temperature effects on sperm may be overly simplistic. Recent work on other small organisms has indicated that temperature effects may overestimate direct metabolic expenditures (Podolsky and Emlet, 1993; Fuiman and Batty, 1997; Hunt von Herbing, 2002). Water viscosity decreases with increasing temperature (Dorsey, 1968; Vogel, 1984; Denny, 1993; Jumars et al., 1993), thus altering the energetic requirements of swimming. For example, Podolsky and Emlet (1993) demonstrated that an increase in temperature from 12 °C to 22 °C increased swimming speed in sand dollar larvae, and that about 40% of the speed increase and 67% of the Q10 could be attributed to the effect of reduced viscosity. This viscosity effect should be even more pronounced in smaller organisms (like sperm) that inhabit lower Reynolds number environments (Vogel, 1984), and most evident at lower temperatures where changes in viscosity are most pronounced (Jumars et al., 1993). However, viscosity effects will only be relevant during times when sperm are actually swimming, or if viscosity stimulates swimming behavior.
It is also possible that temperature did not have a direct effect on sperm longevity per se, but rather an indirect effect via fertilization. The fertilization assays were conducted at the same temperatures used for aging the sperm suspensions. The clearance rates of some filter-feeders increase with temperature (Riisgard and Manríquez, 1997; Lisbjerg and Peterson, 2001; Turker et al., 2003), but the ramifications for sperm capture in brooders have not yet been explored. Consequently, we cannot exclude the possibility that temperature affected the rate at which sperm were removed from the suspensions, as well as the viability of sperm.
Potential evolutionary implications
Because there is a trade-off between sperm velocity and longevity, it has been suggested that fast sperm are advantageous under conditions of sperm competition, and long-lived sperm are advantageous under conditions of sperm limitation (Levitan, 1993, 2000). Although B. schlosseri possesses extremely long-lived sperm, this species does not experience sperm limitation in nature (fertilization levels in nature are generally >85%; Phillippi et al., 2004), and male-phase colonies in experimental populations compete for access to eggs (Yund and McCartney, 1994; Yund, 1995, 1998). Similarly, other brooding free-spawners with high sperm longevity are not thought to be sperm-limited, and they exhibit reproductive traits (e.g., sperm storage, female choice) that are typically associated with sperm competition (Bishop, 1996, 1998; Bishop et al., 2000; Pemberton et al., 2003a). Hence the trade-off between velocity and longevity may not apply to brooders, but only to egg-broadcasting free-spawners.
Overall, Botryllus schlosseri appears to promote fertilization through the longevity of water-born sperm and through the ability to concentrate dilute sperm from the water column. The substantial longevity of dilute sperm reported here potentially allows viable sperm to disperse a great distance from a source population. In the field, most eggs are fertilized by sperm from nearby sources (Grosberg, 1987; Yund, 1995, 1998), but eggs in isolated colonies can be fertilized from distances of 40 m or more (Yund and McCartney, 1994). Given the often high density of B. schlosseri colonies in nature (>1000 colonies m2 at some sites; Grosberg, 1982) and the fact that colonies continuously dribble sperm over a period of 45 days (Stewart-Savage and Yund, 1997), more sperm are probably released than are utilized locally. This surplus is likely to be advected away by currents and could contribute greatly to gene flow between populations if sperm remain viable during their journey. Sperm dispersal could be of particular importance to gene flow in an organism like B. schlosseri that has a very short-lived larval stage (Grosberg, 1987; Yund, 1995).
| Acknowledgments |
|---|
| Footnotes |
|---|
| Literature Cited |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. A. Pechenik, J. S. Pearse, and P.-Y. Qian Effects of Salinity on Spawning and Early Development of the Tube-Building Polychaete Hydroides elegans in Hong Kong: Not Just the Sperm's Fault? Biol. Bull., April 1, 2007; 212(2): 151 - 160. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. D. D. Bishop and A. J. Pemberton The third way: spermcast mating in sessile marine invertebrates Integr. Comp. Biol., August 1, 2006; 46(4): 398 - 406. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Phillippi, E. Hamann, and P. O. Yund Fertilization in an Egg-Brooding Colonial Ascidian Does Not Vary With Population Density Biol. Bull., June 1, 2004; 206(3): 152 - 160. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |