The polychaetes are the most speciose and morphologically diverse class of the Phylum Annelida, with over 5000 identified species. They occupy every part of the marine ecosystem, but are especially abundant in the littoral zone. Spionidan polychaetes inhabit these near-shore regions, mostly living at the sediment-water interface, and numbering several thousand per square meter in many cases (Zajac, 1991). Spionidan and other interstitial polychaetes are responsible for recycling much of the organic matter of the littoral zone. As such they are of interest ecologically as a major component of the meiofaunal community (Levin, 1984). Littoral polychaetes have also been used as pollution-indicator species (Reish and Gerlinger, 1997). Larval form and ecology vary widely within the polychaetes, making them potentially excellent models for study of larval ecology and its evolution (Eckman, 1996; Young, 1994). Because of the morphological diversity within polychaetes, built on a simple segmented annelid body plan, they are also an ideal group for comparative studies of morphogenesis (Berrill, 1961; Irvine, 1998; Irvine et al., 1999).
Many spionidan polychaetes are particularly easy to obtain because of their beach and estuarine habitats and their abundance. Two common types are Polydora, which builds tubes in sand or on various hard substrates, and Chaetopterus, which builds U-shaped mucus tubes beneath sand or mud flats. Both types feed by extracting organic material from their surroundings. Polydora does this by sweeping nearby water and substrate with its long anterior palps, and Chaetopterus by catching material in a mucus net from water pumped through its tube. We have developed methods for culturing each of these animals in the laboratory. These methods may be applied generally to any polychaete with pelagic larvae.
While our interest in these species was for comparative studies of developmental gene expression, the methods outlined could be applied to rearing of any larval stocks needed for a variety of research projects. Examples include studies in larval ecology, such as settlement patterns (Toonen and Pawlik, 1994), or population dynamics (Eckman, 1996). Other uses could be for environmental studies such as pollutant genotoxicity assays (Jha et al., 1996) or studies of pollution indicator species (Rice and Simon, 1980; Yokohama, 1990). Additional applications might be found in mariculture (Nel et al., 1996; Ruck and Cook, 1998). None of what we describe is technically sophisticated, but we hope that by documenting our experiences in detail we can save other workers from "re-inventing the wheel," and allow them to devote more time to improving the methodology and using it to accomplish their scientific goals.
Chaetopterus has been a favorite organism for the study of early development because of the easy availability of large numbers of gametes that can be fertilized in vitro (Eckberg and Hill, 1996). This worm is of interest because of its unusual morphology during later development compared with other polychaetes (Fig. 1). We also worked on a related animal, without the morphological specializations of Chaetopterus, for comparison in embryological studies. Polydora cornuta (Fig. 2), from the related family Spionidae, was chosen because its developmental patterns are more typical of polychaetes (Blake, 1969; Irvine et al., 1999), and it is readily available on the New England coast.
The genus Chaetopterus is a species complex sometimes mistakenly described as monotypic (Petersen, 1984a, b). The species we have used is that described by Enders (1909) as Chaetopterus variopedatus, which is different from the type species C. variopedatus Renier - the type specimen of which has, unfortunately, been lost (M. E. Petersen, Copenhagen Museum, pers. comm.). Eckberg and Hill (1996) call a similar worm C. pergamentaceous. In the body of this paper we will refer to the animal described as Chaetopterus until a complete revision of the genus provides a more definitive name (M. E. Petersen, unpubl. data).
P. cornuta (formerly known as P. ligni) is a member of the polydorid species complex (Blake, 1969). It is easily obtained on the estuarine beaches of New England due to its location in shallow subtidal sand flats accessible to wading. The techniques for its culture described here would be applicable to any interstitial polychaete, including other spionids or interstitial members of other families such as Capitella.
Methods for obtaining and fertilizing gametes have been described by Henry (1986) and Eckberg and Hill (1996). Gravid adult worms may be obtained from animal suppliers: Marine Biological Laboratory, Woods Hole, Massachusetts; and Cape Fear Biological Supply, Beaufort, North Carolina (Cape Fear will sex worms and send equal numbers of males and females). Adults from North Carolina are gravid from April or May through October. The worms can be kept in room temperature aerated aquaria in 32 ppt artificial seawater and adults sexed according to the color of their gametogenic parapodia (white for males, orange for females). Adult male and female worms are best kept in separate tanks to prevent premature spawning. The adults were kept inside their tubes at all times, and access to the worm was achieved by making a small longitudinal cut in the tube. Longitudinal cuts are repaired quickly by the worm. Transverse or chevron-shaped cuts cause irreparable damage to the tube, decreasing the ability of the adult to feed and survive.
Oocytes are obtained by cutting off a single parapodium from a gravid female and emptying the contents (several thousand eggs) into a 150-µm screen immersed in 0.2-µm filtered artificial seawater (FASW) at 32 ppt salinity in a 35-mm petri dish. Oocytes are washed through this screen; retained mucous and parapodial fragments are discarded. The sieved oocytes are washed three more times by transferring them into 100-mm petri dishes of fresh FASW and then incubated in FASW for about 40 min before fertilization to allow for germinal vesicle breakdown. Sperm is prepared by cutting off the parapodium of a ripe male without allowing the contents to mix with seawater. The "dry sperm" is extracted with a Pasteur pipette and diluted 1:1000 with FASW. This sperm suspension is examined through a stereo microscope and, when the sperm were seen to be motile, 1 to 3 drops added to the eggs in a 100-mm petri dish. Knowing the optimal amount of sperm suspension to use to obtain complete fertilization without excessive polyspermy is a matter of experience, because the concentration of dry sperm varies between animals and over time in the same animal.
Swimming ciliated blastulae develop by 12 h at room temperature (21° -23° C). These blastulae may then be transferred to the pelagic culture apparatus described in Section III below.
Like many interstitial polychaetes, P. cornuta is fertilized internally and broods its embryos inside the mother's tube. This makes obtaining early embryonic stages difficult, because gametes are not easily harvested for in vitro fertilization. However, we have developed a "simulated beach" for harboring adult worms that breed in the laboratory; early swimming larvae may then be extracted from the culture water for rearing in a pelagic culture vessel.
P. cornuta is common in shallow subtidal regions of estuaries close to the entrance of a freshwater source. We have harvested P. cornuta at the east side of Eel Pond, Woods Hole, Massachusetts, and at the head of Waquoit Bay, Falmouth, Massachusetts, near the headquarters building of the Waquoit Bay National Estuarine Reserve. The worms construct vertical tubes in sand, which are barely visible to the naked eye. Populations can be quite dense, numbering 14,000 per m2 in midsummer in an estuary in southeastern Connecticut (Zajac, 1991). P. cornuta lives within the upper 2-3 cm of sediment. Population peaks for the New England coast occur in late spring and early summer (Zajac, 1991), making this the best time to collect the adults.
We collected the adults using a "polychaete scoop" of our own construction (Fig. 3). This is made from a modified galvanized sheet metal dryer vent (available at any good building supply store) fastened to a garden hoe, as a handle. The scoop is dragged along the sediment catching the top polychaete-containing layer. The vent has a hole covered with galvanized wire mesh, which in turn is covered with fiberglass window screening adhered with silicone caulk. This hole allows water to pass through the scoop while the mesh catches any polychaetes that swim out of the sediment. The full scoop of sediment is then sieved through two screens, an upper one with an 8-mm mesh, and a lower one constructed of fiberglass window screening (approximately 2 mm mesh size). The upper sieve catches larger stones and macrobenthic organisms, which are then discarded. P. cornuta and other meiofauna are retained on the lower sieve; these are emptied into a bucket with some of the sieved sediment. This operation is repeated numerous times, resulting in a bucket of concentrated meiofauna containing the desired species. The sieving operation must be performed gently, with the lower sieve submerged in seawater so that the interstitial organisms are not ground up in the sand or against the mesh. Additional sediment is sieved only through the coarse mesh, to remove rocks and larger meiofauna. It is retained for future use in the "simulated beach."
Once a sufficient quantity of sediment has been sieved, the concentrated meiofauna must be taken back to the laboratory to separate the desired species from the other interstitial species present. This is important to avoid the isolation of larvae of different species, which might be difficult to distinguish from those of the study species. In addition, it is necessary to remove meiofauna that might feed on the study species or that might outcompete the study species for food. We transfer a small amount of sediment to a shallow petri dish for examination under a low-power stereo microscope. The worms hide in their sand-mucus tubes, but they can be captured by agitating the tube with a wide-bore Pasteur pipette until the worm swims out and then sucking it into the pipette for transfer to another dish. The worms are checked to make sure they are the correct species before being added to the culture vessel. (Refer to Appendix 1 for identification of P. cornuta.)
We transport unsorted worms by car from Woods Hole, Massachusetts, to Chicago, Illinois by layering the sieved sediment containing the worms into shallow, covered plastic containers (the same as used for the adult worm culture) along with some extra sediment and seawater, allowing for air space above the water. We stack the containers and surround them with foam insulation to prevent overheating. After the two-day trip the worms are sorted and placed into the simulated beach culture described below.
The "simulated beach" is made by depositing about 3 cm of sediment from the collection site (sieved through 8-mm screen as described above) into a wide shallow vessel, deep enough to allow for 6-8 cm of water on top. (Refer to Fig. 4) The sediment should first be frozen to kill any fauna remaining and then rinsed several times with fresh water to remove some of the dead organic material, which could produce unwanted products of decomposition. After deposition of the sediment, natural or artificial seawater (ASW) is added. The seawater should be changed at least twice to dilute out any remaining fresh water and to remove some of the silty organic debris, which will wash out of the newly placed sediment. Washed autoclaved sand may be layered on top if the sediment is too silty. Once the substrate has settled, the sorted worms can be scattered over the surface. They will colonize the sediment and construct new tubes within a few hours. We have cultured 225 worms in one 30 cm culture vessel with good results (i.e., 0.25 worms per cm2).
Two operations are necessary to maintain the culture long enough to produce larvae. First, adult animals must be fed. For the 30 cm culture vessel we use 4 ml of concentrated liquid invertebrate food (Roti Rich, Aquaculture Supply) diluted in some ASW and broadcast evenly over the sediment once per week. Over-feeding should be avoided to prevent excessive bacterial growth due to decomposing food matter.
Second, the culture water must be changed regularly to prevent waste products, bacteria, and unwanted volunteer organisms from building up, and to retrieve any larval offspring that may have been released into the seawater. We change the water by siphoning the old water into a 35-µm sieve (Fig. 5a) so that any pelagic larvae will be caught for separate culture, as described in Section III below. If any other meiofauna are captured in the sieve they may be removed before the desired polychaete larvae are transferred to the pelagic culture apparatus (Fig. 5b), by gently backwashing the sieve with ASW from a squeeze bottle.
Two changes of water per week should be sufficient to maintain healthy cultures. The inlet of the siphon tube may be passed back and forth through the culture vessel to ensure that as many larvae as possible are caught. Fresh seawater is added through a side inlet so that the disturbance of the sediments is minimized. If natural seawater is used it may be necessary to filter out suspended fauna, such as small crustaceans, that might invade the colony.
In our experience, pelagic larvae begin to appear in the culture water around 40 d after beginning the culture. Using these methods we have collected as many as 1000 larvae from two culture vessels in one water change. However, the number of larvae retrieved varies over time, with a periodicity of around 30 d, probably due to the breeding and life history characteristics of the worms (Fig. 6). In our hands, culture life has been 3-4 months, or roughly 2 months of larval production. After this time the adult population declines and larval collections drop off.
The simplest methods for rearing pelagic marine larvae involve closed cultures, i.e., the culture medium is not continuously changed as in a flowing seawater system (Dean and Mazurkiewicz, 1975; Strathmann, 1987). Closed cultures are the only practical alternative for inland laboratories without access to natural seawater. Successful closed culture systems only require a means of moving the culture water to keep the larvae suspended and a means of periodically changing the water to reduce the concentration of waste products. The devices we constructed for these purposes are easily and inexpensively made from readily available materials. (Refer to Fig. 5a, 5b, for illustrations, Appendix 2 for details of construction, and Appendix 3 for parts suppliers.)
As culture water we used artificial seawater (ASW; Instant Ocean) mixed in 50-gallon batches, using tap water, and aerated constantly. Several clean clam shells (Mercenaria sp.) were added to the tank because of reports that unknown factors in the shells improve marine animal culture in the laboratory (M. LaBarbera, University of Chicago, pers. comm.). We filtered the ASW through a 5-µm bottle top filter (Nalgene filter holder; Polysciences filters). Failure to filter the water resulted in contaminating crustaceans and algae in our cultures. Four liters of ASW were used in each vessel. One-half of the water was changed twice weekly by pouring the water into the sieve shown in Figure 5a, which was filled with water above the level of the mesh so that the larvae are not forced against the screen or allowed to dry out. The mesh size should be based on the size of the larvae to be maintained – we used 50 µm for young larvae, and 100 µm later in development. Once the water has run through the sieve, the concentrated larvae are gently poured back into the culture vessel and the sieve backwashed with a squeeze bottle of ASW to release any larvae caught on it.
For feeding of pelagic larvae we had the best results with unicell algae cultured in the laboratory. Attempts to use commercially available aquarium food, such as the liquid invertebrate food used for adult Polydora above, or dried rotifers, resulted in excessive bacterial growth in the colonies and poor survival of the larvae. Several types of algae were used, but Dunaliella salina, a green alga, and Isochrysis galbana, a golden-brown alga, gave adequate nutrition and were easy to maintain in culture.
The starting cultures were obtained from Carolina Biological. Autoclaved half-liter wide-mouth glass bottles with cotton stoppers were used as culture vessels to avoid bacterial or fungal contamination. We supplemented 0.2-µm filtered artificial seawater (FASW) with F/2 medium at standard concentration [Aquaculture Supply, (Strathmann, 1987)] as the culture medium. The bottles were placed in sunlight and swirled daily to aerate. For feeding, 1 ml of each of the two algal species were added per l of culture twice per week, after changing the culture water.
The most important variable for the success of cultures appears to be the density of larvae. Chaetopterus cultures were attempted with initial densities ranging from 0.67 to 5.5 larvae per milliliter (Fig. 7). Cultures at the low end of this range had the lowest mortality, and subsequent cultures were grown at 1 larva per ml or less. Since Polydora larvae were never available in the large numbers possible with Chaetoperus, these cultures were always maintained at 0.1 larva per ml or less.
Under our culture conditions, Chaetoperus growth in body length is roughly linear for the first 28 days, but as expected, the size variance for larvae of the same age increases over time at a given culture density (Fig. 8). P. cornuta larval growth was measured in terms of setiger number, which increased by approximately one setiger per day (Fig. 9). All of our cultures were maintained on a normal light/dark cycle, but it should be noted that Dean and Mazurkiewicz (1975) have reported faster growth for P. cornuta kept in constant darkness.
Both Chaetoperus and P. cornuta larvae underwent metamorphosis spontaneously in the pelagic culture vessels upon reaching a competent developmental stage. For Chaetopterus this is stage L6, which requires approximately 60 d from fertilization (Irvine et al., 1999). P. cornuta required approximately 20 d in culture to reach competence at 17 setigers. It was unnecessary to add substrate to induce metamorphosis. Faster or more reliable metamorphosis might be achieved by transferring late-stage larvae to a dish with substrate (Dean and Mazurkiewicz, 1975), and adding excess potassium (Yool et al., 1986) or neurotransmitter analogs (Morse, 1985) to the culture medium.
Live larvae may be narcotized by adding 0.1% propylene phenoxytol (also known as propylene glycol phenol ether, Dow Chemical) in FASW 1:1 to the specimens in seawater (0.05% final concentration). This anesthetized later-stage specimens sufficiently for photography, especially in combination with immobilization, by compressing the larvae with a cover slip. Early-stage larvae were relaxed in 4% ethanol in FASW because propylene phenoxytol was not effective at these stages.
The authors would like to thank Jonathan J. Henry, Mary E. Petersen, James A. Blake, and Michael LaBarbera for scientific and technical advice. Thanks also to Susan E. Hill for critical reading of an earlier version of this manuscript. We acknowledge the assistance of Waquoit Bay National Estuarine Research Reserve for allowing us to collect Polydora from their beach. S.Q.I. was supported by National Institutes of Health training grant T32HD07136-20, National Science Foundation dissertation improvement grant 9623453, and a Hind's Fund grant from the University of Chicago. M.Q.M. was supported by National Science Foundation grant 9315653.
Appendix 1: Identification of P. cornuta adults
Appendix 2: Details for construction of the pelagic culture apparatus
Appendix 3: List of suppliers
Figure 1: Chaetopterus larval stages (metatrochophore and metamorphosed juvenile; photomicrographs)
Figure 2: Polydora cornuta larval stages (nectochaete larva and metamorphosed juvenile; photomicrographs)
Figure 3: Interstitial polychaete scoop (diagram)
Figure 4: Interstitial polychaete culture vessel (diagram)
Figure 5: Pelagic culture apparatus and larvae sieve (diagrams)
5a. Section through the sieve used for capturing larvae and changing culture water.
5b. Axonometric drawing of the pelagic culture vessel described in Appendix 2.
Figure 6: Polydora cornuta larval collection history
Figure 7: Chaetopterus survivorship as a function of initial culture density
Figure 8: Chaetopterus larval growth as a function of time
Figure 9: Polydora cornuta larval growth as a function of time
Figure 10: Diagnostic characters for the identification of Polydora cornuta adults (diagrams).