Characterization of Na⁺ and Ca²⁺ Currents in Bag Cells of Sexually Immature Aplysia Californica

Sexual maturation in A. californica is defined by the onset of copulation and egg-laying behavior. Yet the reproductive system appears fully formed several months prior to sexual maturity when the animals enter stage 13 of development (stage 13; Cash and Carew, 1989). The appearance of the bag cell clusters on the abdominal ganglion is an important developmental feature of early stage 13, occurring at approximately 120 days post-metamorphosis. At this stage, the animals can be considered sexually immature adults.

The bag cells store hormones that cause egg laying when released. Both an 11 g sexually immature A. californica and a 200 g mature A. californica 4–6 months after reaching stage 13 are equipped to reproduce. Yet reproductive behavior and egg laying are observed only in A. californica weighing more than 150 g. The physiological features that distinguish 11 g from 200 g adult A. californica have been documented, leading to several hypotheses for the absence of reproductive behavior in A. californica weighing less than 150 g. Two possibilities are that sexually immature, early stage 13 animals do not synthesize bag cell hormones or that they do not release them. McAllister et al. (1983) reported that, in early stage 13 A. californica, the first 10 cells of the newly forming abdominal ganglion contain messenger RNA for egg-laying hormone. Nick et al. (1996b) reported that bag cell clusters from immature A. californica (11–20 g) can be electrically stimulated to secrete egg-laying hormone in concentrations sufficient to induce egg-laying behavior in mature adults. These studies suggest that the pertinent hormones and the cellular machinery for secretion are present in immature A. californica. Nevertheless, an ability to release egg-laying hormones naturally may be lacking in sexually immature adults.

Hormone secretion from mature bag cells is caused by trains of action potentials called the afterdischarge (AD; Kupfermann and Kandel, 1970). Occasionally, a large electrical stimulus can provoke an AD in an excised bag cell cluster from stage 13 animals that are far from sexual maturity (11 g; Nick et al. 1996a). This is very rare, and the resulting AD is short-lived. The inability of young A. californica to sustain an AD coincided with high K⁺ current densities and low Ca²⁺ current densities in immature A. californica compared with their sexually mature counterparts capable of AD.

The AD is caused in part by ion movement through channels that carry Na⁺ and Ca²⁺ into the cell during an action potential. Characterization of the inward voltage-gated currents for Na⁺ and Ca²⁺ in bag cells from sexually immature A. californica may reveal kinetic and pharmacological differences that provide insights into the regulation of excitability during sexual maturation. In a previous study (Fieber, 1995), Na⁺ and Ca²⁺ currents of bag cells from A. californica of known sexual maturity were studied in short-term tissue culture. In the present study, bag cell Na⁺ and Ca²⁺ currents from sexually immature A. californica were studied under exactly the same conditions.

Aplysia californica were obtained from the University of Miami Aplysia Facility where they were raised from eggs laid by wild-caught brood stock and maintained in community tanks at 15 °C. All experiments were performed on cells from sexually immature animals. Sexual immaturity was defined by an absence of sexual behaviors such as copulation or egg laying. Subsequent observations on animals from the same cohorts as those used in experiments indicated that these cohorts did not engage in mating behavior for approximately 2 months after electrophysiological experiments.

Experiments to assess whole-cell ionic currents were performed with the whole-cell variation of the patch-clamp technique (Hamill et al. 1981) on cells maintained in short-term tissue culture at 19–21 °C. Abdominal ganglia with intact bag cell clusters from A. californica of known ages were removed from animals anesthetized for 30 min in a 1:1 mixture of isotonic MgCl₂ and sea water and were digested in an enzyme solution in artificial sea water (ASW) for 24 h at room temperature. The enzyme solution consisted of 12.5 mg of neutral protease (Boehringer Mannheim, Indianapolis, IN, USA), 5 mg of hyaluronidase and 1 mg of collagenase type XI (both from Sigma, St Louis, MO, USA) in 5 ml of ASW. The ASW consisted of (in mmol l⁻¹): 417 NaCl, 55 MgCl₂, 10 CaCl₂, 10 KCl, 10 Hepes/KOH, pH 7.6. After digestion, the connective tissue sheath was removed from the clusters and the bag cells were dispersed onto 35 mm culture dishes in ASW, plus 100 i.u. ml⁻¹ penicillin, 100 μg ml⁻¹ streptomycin and 0.25 μg ml⁻¹ fungizone, and maintained at room temperature (approximately 19–21 °C) until use in experiments 4–48 h later. Electrophysiological recordings were made with an Axopatch 200A patch-clamp amplifier, using the PClamp 6 programs (Axon Instruments, Foster City, CA, USA) for data acquisition and storage on a 100 MHz IBM-compatible computer. The intracellular solution consisted of (in mmol l⁻¹): 450 CsCl, 2.9 CaCl₂, 2.5 MgCl₂, 10 EGTA, 5 Na₂ATP, 300 μmol l⁻¹ GTP and 40 Hepes/CsOH, pH 7.4. Pipette resistances were less than 1 MΩ. ASW was used for recording Na⁺ and K⁺ currents. Ca²⁺ currents were not recorded during Na⁺ current recordings because of their relatively greater latency and small amplitude in 10 mmol l⁻¹ Ca²⁺. To study Ba²⁺ currents through Ca²⁺ channels, the solution consisted of Ba²⁺-Cs⁺-TEA⁺ containing (in mmol l⁻¹): 460 tetraethylammonium chloride (TEACl), 10.4 CsCl, 11 BaCl₂, 55 MgCl₂ and 10 Hepes/CsOH, pH 7.6. The cells were bathed in different solutions during experiments by applying solutions from 1 μl pipettes attached to gravity-dispensed reservoirs. All voltage-activated currents were elicited from a holding voltage by test depolarizations with a duration of 25–500 ms separated by a pause of 4–10 s at the holding voltage.

Calculated voltage errors were usually smaller than 3 mV following approximately 80 % series resistance compensation. These errors were not corrected in the data presented. The PClamp 6 P/4 leak subtraction protocol was used in some of the data presented. Whole-cell capacitance was estimated from the capacitance setting on the patch-clamp amplifier.

All reagents were obtained from Sigma Chemical Co. unless otherwise noted. Tetrodotoxin (TTX; Research Biochemicals Inc.; Natick, MA, USA) was diluted and divided into small samples, then dried under N₂ and stored at −20 °C until it was dissolved in ASW at the time of experiments. Nifedipine and ω-conotoxin GVIA (ω-conotoxin) were obtained lyophilized from Calbiochem (San Diego, CA, USA). Nifedipine was diluted daily from stock solutions of 5 mmol l⁻¹ in ethanol stored at −4 ° C in foil-wrapped vials, and was applied in the dark. Solutions of ω-conotoxin were made up at the time of use and were discarded after 24 h. ω-Conotoxin was applied to cells in a solution containing low levels of Ca²⁺ consisting of ASW solution lacking CaCl₂. After control recordings had been collected, each drug was bath-applied from a 1 μl pipette. Wash-out of drug effects using extracellular solution was attempted after studying the effects of the drug.

The Na⁺ equilibrium potential was calculated from the Nernst equation using Na⁺ activity values from Robinson and Stokes (1968). Data are presented as mean ± S.D. in the stated number of cells, unless otherwise noted.

Statistical comparisons were made with the Datadesk program (Data Description, Inc., Ithaca, NY, USA).

Na⁺ and Ca²⁺ currents are described from 54 bag cells from 24 A. californica from nine different egg masses. The animals studied ranged in age from 4.25 to 7 months and weighed 12–125 g. Cultured bag cells from immature animals were 10–30 μm in diameter. Input resistances were 1.19±1.7 GΩ (N=23). Less than 50 % of cells studied in whole-cell voltage-clamp experiments had inward Na⁺ currents, Ca²⁺ currents or both, although outward currents were commonly recorded. The number of bag cells per culture with Na⁺ and Ca²⁺ currents did not vary widely with the length of time cells were in culture.

Fig. 1 shows Na⁺ currents in immature bag cells. Na⁺ currents activated at test depolarizations less negative than −30 mV elicited from a holding potential (V_h) of −70 mV and peaked within 2 ms during each depolarizing test pulse (Fig. 1A). Na⁺ current amplitude peaked at test depolarizations of 10–20 mV, as shown in the current–voltage relationship (I–V) of Fig. 1B. Although variable in amplitude, the Na⁺ current densities of immature bag cells were approximately one-third of those of Na⁺ currents in adult bag cells (approximately150 pA pF⁻¹ at activation peak; Fieber, 1995). The average zero current potential was close to the Na⁺ equilibrium potential of 84 mV.

Like mature Na⁺ currents, immature Na⁺ currents were susceptible to block by TTX (Fig. 2). In six bag cells from three different animals, Na⁺ currents were reversibly blocked by 82±5 % by bath-applied 50 nmol l⁻¹ TTX.

During an AD, mature bag cells undergo a change in resting potential to −30 mV (Fisher and Kaczmarek, 1990). The sensitivity of bag cell whole-cell Na⁺ current to depolarized V_h was determined to assess the ability of Na⁺ currents from immature bag cells to fire repetitively at −30 mV. Sensitivity to V_h was evaluated in three ways (Fig. 3). Families of Na⁺ currents were recorded from different V_h values, Na⁺ currents were recorded at a single test potential from different V_h values held for 1 s prior to the test pulse, and two repetitive depolarizations to a single test potential, with varied interval, were elicited from a V_h of −70 mV and −30 mV.

Changes in the duration of the depolarized V_h did not affect the amplitude of immature Na⁺ currents. The I–V curves in Fig. 3A from an individual cell show that, when families of Na⁺ currents were elicited from each V_h maintained for a total of approximately 1 min except during the depolarizations, the current amplitudes were maximal at a V_h of −70 and −50 mV, and were 54±7 % of maximum at a V_h of −30 mV (N=3). For comparison, a plot of the average Na⁺ current amplitude as a function of V_h held for 1 s prior to a single test pulse is shown in Fig. 3B. The data were fitted to a Boltzman distribution with the voltage at which the current was half-inactivated (V_1/2) was calculated as −27 mV. This value of V_1/2 is similar to the degree of current inactivation observed when depolarized V_h was maintained for 1 min (Fig. 3A). The V_1/2 of Fig. 3B was not significantly different from the average mature adult V_1/2 for Na⁺ current of −21 mV (Fieber, 1995; two-way analysis of variance, ANOVA). Fig. 3C shows recovery of Na⁺ currents from inactivation in a double-pulse protocol. Time constants were 1.9 ms at V_h=−70 mV, which is not significantly different from the adult time constant of 2.9 ms (Fieber, 1995; ANOVA), and 9.5 ms at V_h=−30 mV, which is significantly different from the adult value of 6.8 ms (two-way ANOVA, P<0.01). Thus, Na⁺ currents of immature bag cells were slower to recover from inactivation at V_h values relevant to the AD.

Ca²⁺ currents, using Ba²⁺ as the charge carrier, were studied by eliciting test potentials from V_h=−70 mV. Ca²⁺ currents were short-lived in whole-cell recordings, beginning to run down within approximately 2 min of whole-cell dialysis and disappearing within 10 min of the whole-cell current recording configuration being established. Variations on the normal recording protocol were made in an attempt to inhibit Ca²⁺ current run-down. Maintaining the cells at a very negative V_h of −120 mV for 1 min before any test potentials to elicit Ca²⁺ currents did not inhibit run-down beginning within 2 min of whole-cell dialysis (N=3). Ca²⁺ currents also disappeared with time in cell-attached recordings of single Ca²⁺ channel currents in which the cytosol remained intact (data not shown).

Ca²⁺ currents were smaller in amplitude than Na⁺ currents from the same cells. Ca²⁺ current densities in immature bag cell were approximately 100 pA pF⁻¹ at the activation peak, which is less than the Ca²⁺ current densities of mature bag cells (Fieber, 1995). Immature Ca²⁺ currents, like mature Ca²⁺ currents, activated at test potentials less negative than approximately −40 mV and peaked at approximately 20 mV (Fig. 4A,B). Immature Ca²⁺ currents were blocked by the Ca²⁺ channel blocker Cd²⁺ (100 μmol l⁻¹; not shown; Bean, 1989).

The pharmacological sensitivity of Ca²⁺ currents, with examples shown in Fig. 5, was similar to that of L-type currents in mature bag cells. Blocking using bath-applied nifedipine (10 μmol l⁻¹) was variable, ranging from 0 to 100 % block with an average block of 61±10 % in 11 bag cells (Fig. 5A). Block by ω-conotoxin (10 μmol l⁻¹) was ⩽5% in six cells, with an average block of 2.3±5 % (Fig. 5B).

Fig. 6 shows the susceptibility of peak Ca²⁺ current amplitude to depolarized V_h. Like immature Na⁺ currents, immature Ca²⁺ current amplitudes were not affected by changing the duration for which the depolarized V_h was maintained. The I–V curves recorded at different V_h values maintained for approximately 1 min (Fig. 6A) show that peak current declined at a V_h of −30 mV. The plot of different V_h values maintained for 1 s before each test pulse versus whole-cell Ca²⁺ current amplitude is shown in Fig. 6B. These data are fitted by a Boltzman distribution with a V_1/2 of −23 mV; the mature adult average V_1/2 was −34 mV (Fieber, 1995). These values are not significantly different (two-way ANOVA). A depolarized V_1/2 is characteristic of L-type Ca²⁺ currents.

Immature bag cells may be incapable of action potential trains because their channels are regulated in an inactivated state. To test for the presence of inactivated Ca²⁺ channels capable of being opened by a depolarizing prepulse (Artalejo et al. 1991a,b), experiments were performed on nine cells with Ca²⁺ currents in which 5 ms prepulses to 120 mV were given immediately prior to each of 20 successive test pulses to elicit Ca²⁺ currents. For comparison, five cells with Ca²⁺ currents were subjected to test pulses without the 120 mV prepulse. Test pulses were to 20 mV from a V_h of −90 mV. All test pulses were separated by 10 s at −90 mV. Of the nine cells in which the 120 mV prepulse protocol was used, the initial Ca²⁺ current amplitude increased by 20–60 % during successive test pulses in five cells, was maintained in three cells, and decreased in one cell. Of the five cells with Ca²⁺ currents not receiving the 120 mV prepulse, the initial Ca²⁺ current amplitude was maintained in three cells, increased by 20 % in one cell and decreased in one cell. In the eight cells in which the 120 mV prepulse preserved or enhanced Ca²⁺ current amplitudes during the 20 test pulses, however, the Ca²⁺ currents were irretrievably less than 50 % of the starting amplitude within 12 min of whole-cell dialysis.

The slower recovery of immature bag cell Na⁺ currents from inactivation and the labile nature of the immature bag cell Ca²⁺ currents compared with mature A. californica bag cells suggest that immature currents might not be sustained at the action potential frequencies observed during an AD. To test this possibility, Na⁺ and Ca²⁺ currents were elicited for 10 and 20 ms, respectively, at 4 Hz, an action potential frequency typical of the initial stage of AD (Dudek and Kossatz, 1982). V_h in these experiments was −70 mV or −30 mV, values that are typical of the initial, Na⁺-dependent phase (−70 mV) and the later, Ca²⁺-dependent phase (−30 mV) of the AD, respectively (Fisher and Kaczmarek, 1990). The results are shown in Fig. 7 for Na⁺ currents and in Fig. 8 for Ca²⁺ currents. 4 Hz Ca²⁺ current experiments were executed within 2 min of the start of whole-cell dialysis, a period during which little Ca²⁺ current run-down occurred in other whole-cell Ca²⁺ current experiments.

Na⁺ current amplitudes were maintained or increased during 1 min of 4 Hz test pulses in three cells at a V_h of −70 mV (Fig. 7A). Na⁺ currents were also maintained or increased at a V_h of −30 mV (Fig. 7B) in three out of four experiments. In contrast, Ca²⁺ current run-down during the 1 min protocol was common, occurring in seven out of 10 cells at a V_h of −70 mV (Fig. 8A) and in 12 out of 12 cells at a V_h of −30 mV (Fig. 8B).

Activation of voltage-gated Na⁺, Ca²⁺ and K⁺ currents by an initial electrical or chemical stimulus causes 30 min of repetitive action potentials in A. californica bag cells, known as the afterdischarge (AD). This AD causes hormone release from bag cells in sexually mature animals (Kupfermann and Kandel, 1970). In contrast, immature bag cells from animals weighing less than 11 g cannot sustain a full AD when they are exogenously stimulated (Nick et al. 1996a), and thus they do not exhibit AD-dependent hormone release. In the latter study, the ability to sustain an AD correlated with increasing animal mass between 11 and 50 g. The development of the capability of sexually immature A. californica to sustain an AD coincided with a decrease in the current density of three outward K⁺ currents and an increase in the current density of two Ca²⁺ currents and the A-type K⁺ current. The present study suggests several reasons additional to the observations of these authors why the bag cell clusters from a percentage of immature A. californica failed to sustain an AD. These additional reasons are a low percentage of cells with Na⁺ and Ca²⁺ currents, a low density of Na⁺ currents, and the time- and voltage-dependent disappearance of Ca²⁺ currents in whole-cell recordings.

The Na⁺ and Ca²⁺ currents of mature A. californica bag cells have been well characterized (DeReimer et al. 1985b; Fieber, 1995; Nerbonne and Gurney, 1987; Strong et al. 1987). Ionic currents in bag cells of sexually immature A. californica studied in short-term tissue culture had many similarities to those present in bag cells from sexually mature A. californica under the same conditions. The activation range, time course, pharmacology and susceptibility to depolarized V_h of immature Na⁺ and Ca²⁺ currents are all nearly identical to these characteristics in mature bag cell currents. The similarities are sufficiently numerous to suggest that the Na⁺ and Ca²⁺ channels are the same molecular entities in immature and mature A. californica. The failure of immature bag cells to sustain an AD may be due in part to the relatively lower abundance of functional Na⁺ and Ca²⁺ channels in immature A. californica and the lability of immature bag cell Ca²⁺ currents.

Only 50 % of cultured bag cells from immature A. californica ranging from 4.25 to 7 months in age at 12–125 g had measurable inward currents, in contrast to 90 % of bag cells from mature animals cultured under the same conditions (Fieber, 1995). Nick et al. (1996a) found that immature bag cells from animals incapable of an AD at body masses less than 11 g responded to TEA⁺ or to electrical stimulation with long depolarizations that sometimes summated to an action potential. Nick et al. (1996a) did not report the percentage of immature cells per A. californica cohort that responded with a depolarization, indicating at least the presence of Ca²⁺ currents and K⁺ currents, or the percentage that had whole-cell Ca²⁺ currents. The generally lower Ca²⁺ current density of immature bag cells that probably contributed to the failure to sustain an AD in some immature animals was also observed by these authors. New observations made here, such as a similar, low density of Na⁺ currents and differences in the number of immature bag cells possessing functional Na⁺ and Ca²⁺ currents, might further explain the inhibition of the AD in a percentage of immature animals.

Nick et al. (1996a) observed that, as the mass of the animal increased above 11 g, an increasing proportion of bag cell clusters were capable of AD and that excised bag cell clusters from A. californica over 50 g were capable of producing an AD. In the present study, the proportion of dissociated bag cells with measurable Na⁺ and Ca²⁺ currents in this exact size range was uniformly low. There are several explanations for why an AD might occur in some 11–50 g A. californica despite the scarcity of Na⁺ and Ca²⁺ currents observed in cultured bag cells from animals in this size range. First, there may be abundant bag cells in clusters from 11–50 g animals with Na⁺ and Ca²⁺ channels, but these cells rarely survive in culture. This explanation would require a large difference in the survivorship of bag cells from immature and mature A. californica in culture, which is not likely since bag cell cultures of mature (Fieber, 1995) and immature A. californica yield approximately the same number of cells after enzymatic treatment. Second, bag cells from 11–50 g animals may lose functional channels in culture. This explanation is also unlikely because approximately 50 % of the bag cells from immature animals dissociated only 4 h before experimentation lack functional Na⁺ and Ca²⁺ channels, and this number does not vary widely with the time that cells are in culture, as would be expected if channels were being resynthesized, degraded or downregulated in large numbers. Third, low Na⁺ and Ca²⁺ channel densities in cultured bag cells from 11–50 g A. californica may be representative of channel numbers in vivo, but these bag cell clusters may require very few electrically excitable cells to sustain at least a short AD. The high input resistances of immature compared with mature bag cells (Nick et al. 1996a; present study) suggest that relatively fewer Na⁺ and Ca²⁺ channels would be required in immature cells than in mature cells for the generation of an action potential. However, the minimum number of action-potential-capable cells in a cluster required to sustain an AD is unknown.

The 4 Hz test pulse experiments suggest that Na⁺ currents do not inactivate with 4 Hz activation but that Ca²⁺ currents do inactivate. These experiments were conducted under conditions in which AD-induced second messengers known to increase Na⁺ and Ca²⁺ currents and to sustain the AD in mature A. californica (Kaczmarek and Strumwasser, 1984; DeReimer et al. 1985b; Fieber, 1995) are absent. The availability of Na⁺ currents to be sustained during 4 Hz stimulation suggests that bag cells of immature A. californica are probably capable of repetitive firing during the initial, 1 min Na⁺-dependent phase of the AD (Kaczmarek et al. 1982) in spite of their greater tendency to inactivate in a standard double-pulse protocol than mature Na⁺ currents. In contrast, Ca²⁺ currents in bag cells of a percentage of immature A. californica may not be available during the subsequent, Ca²⁺-dependent phase of the AD. Protein kinase C levels increase during the AD to enhance Ca²⁺ current amplitudes. It is not known whether the manoeuvres Nick et al. (1996a) used to initiate depolarizations in immature bag cells (TEA⁺, 0.1 nA current injection) increase protein kinase C levels. These authors demonstrated that protein-kinase-C-sensitive Ca²⁺ currents are not present in immature A. californica under 11 g, suggesting that protein kinase C does not play the same role in maintenance of immature Ca²⁺ currents as it does in mature Ca²⁺ currents. It is possible that protein kinase C may prevent the Ca²⁺ current run-down observed in immature bag cells.

The run-down of Ca²⁺ currents in immature bag cells indicates that these currents may be susceptible to wash-out, use-dependent block or cumulative inactivation. Experiments to determine the contribution of whole-cell dialysis versus use-dependent inhibition to Ca²⁺ current run-down were inconclusive. Short-term (4 min) run-down was prevented in some cells by using depolarizing prepulses to avoid use-dependent block or cumulative inactivation, but this procedure did not prevent eventual Ca²⁺ current run-down during the course of longer experiments. Holding the cells at a negative V_h for 1 min before test potentials to elicit currents did not retard subsequent Ca²⁺ current run-down. This is an additional indication that run-down was not solely activity-dependent. However, the failure to avoid run-down in cell-attached single-channel experiments indicates that whole-cell dialysis is not the sole cause of run-down. Since Ca²⁺ current run-down was not observed in mature bag cells in vitro (Fieber, 1995), it is unlikely that the current carrier, Ba²⁺, or the constituents of the intracellular solution are overtly damaging to bag cells. The possibility remains that immature bag cells in vivo have labile Ca²⁺ currents during repetitive activation of this current. Although Ca²⁺ current run-down and cumulative inactivation are observations that support the idea of resting modulation to prevent the AD, both occurred during studies of isolated Ca²⁺ currents in the absence of possible modulating effects of depolarizations that activate K⁺ currents. Action potentials themselves are stimulatory in mature bag cells.

Modulation of Na⁺ and Ca²⁺ currents by cyclic AMP and protein kinase C changes the height, duration and frequency of action potentials during the AD in sexually mature A. californica (Kaczmarek et al. 1978, 1980; Kaczmarek and Strumwasser, 1981, 1984; DeReimer et al. 1985a,b; Strong et al. 1987), sustaining and strengthening it. The scarcity of Na⁺ and Ca²⁺ currents in whole-cell recordings from immature A. californica and the run-down of Ca²⁺ currents, together with the up-regulation of K⁺ currents in immature A. californica observed by Nick et al. (1996a), suggest that bag cell currents are also regulated in sexually immature adults, but with a different result from that in mature adults. Ion channel modulation in immature bag cells may retard the occurrence of action potentials rather than causing changes in action potential amplitude and frequency, as occurs in mature adults. Such regulation would be present in closed or resting channels of immature A. californica and regulation would retard the opening of Na⁺ and Ca²⁺ channels in response to a voltage stimulus that would otherwise open them, and would ensure that outward K⁺ flow during depolarizations inhibits action potentials and the premature release of egg-laying hormones. This form of regulation could be effected by phosphorylation of ion channels which, in the case of bag cell Ca²⁺ channels, is known to be associated with changes in their functional expression during the AD (Fisher and Kaczmarek, 1990; DeReimer et al. 1994).

The trigger for a change from the resting state may be a hormonal or electrical signal, transforming functionally inexcitable cells into cells that are able to contribute to repetitive action potential firing within the bag cell cluster.

Supported by NSF IBN-9508705 and NIH RR-10294. The NCRR supports investigator-initiated research grants for the study of various non-mammalian systems. Please contact Dr Jill Carrington at gillc@ep.ncrr.nih.gov or (301) 435-0776 for more information.