The crustacean stomatogastric ganglion, consisting of about 30 identifiable neurones, produces several distinct patterned outputs governing rhythmic movements of the stomach. This system has become a model for addressing cellular properties and synaptic interactions underlying pattern generation in neural networks (see Selverston and Moulins, 1987). Studies of the system have been limited by the need to use acutely isolated intact or semi-intact ganglion preparations. Also, interesting membrane responses and synaptic interactions occur in the neuropile, but recording from sites other than the somata has been difficult using an intact preparation. Primary culture of dissociated neurones from the ganglion may yield new insights on regional membrane properties and longterm (days) responses to neuromodulators of individual neurones that have regenerated neuritic processes.

This paper presents our culture methods and some observations of outgrowth from neurones that have been dissociated from the stomatogastric ganglia of two evolutionarily diverse species, the Maine lobster Homarus americanus (Fig. 1) and a tropical semi-terrestrial crab Cardisoma carnifex (Fig. 2). The only other study on stomatogastric neurones in culture of which we are aware used a freshwater crayfish (Pacifastacus leniusculus), in which neurones were plated in media supplemented with serum (Krenz and Fischer, 1988).

Fig. 1.

Photomicrographs of outgrowth from isolated Homarus americanus stomato-gastric neurones. (A) Low-power phase contrast image of soma with residual axon and lateral processes, 2 days in culture. Arrowhead marks end of residual axon. (B) Higher-power image of the growth cone from A taken at the same time as A. (C) Same growth cone on day 3. Note directed growth extending left from the veil. (D) Composite photographs pieced together from two focal planes (soma and growth cone) to give an example of neuronal surface membrane contour. Note the large veil. Day 2 in culture. (B–D) Hoffman interference modulation contrast microscopy. All cells in defined medium. Calibration bars, 50 μm.

Fig. 1.

Photomicrographs of outgrowth from isolated Homarus americanus stomato-gastric neurones. (A) Low-power phase contrast image of soma with residual axon and lateral processes, 2 days in culture. Arrowhead marks end of residual axon. (B) Higher-power image of the growth cone from A taken at the same time as A. (C) Same growth cone on day 3. Note directed growth extending left from the veil. (D) Composite photographs pieced together from two focal planes (soma and growth cone) to give an example of neuronal surface membrane contour. Note the large veil. Day 2 in culture. (B–D) Hoffman interference modulation contrast microscopy. All cells in defined medium. Calibration bars, 50 μm.

Fig. 2.

Hoffmann modulation contrast images of directed outgrowth from a Cardisoma carnifex stomatogastric neurone at different culture ages (as indicated) in static defined medium. At day 1, filopodia are seen at the peripheral margin of the small veil. By day 2, there has been extensive directed outgrowth from day 1 filopodia. Further extension of several processes occurred by day 3. Note the fibrillar appearance of the stabilized major neurite. Scale bar, 50 μm.

Fig. 2.

Hoffmann modulation contrast images of directed outgrowth from a Cardisoma carnifex stomatogastric neurone at different culture ages (as indicated) in static defined medium. At day 1, filopodia are seen at the peripheral margin of the small veil. By day 2, there has been extensive directed outgrowth from day 1 filopodia. Further extension of several processes occurred by day 3. Note the fibrillar appearance of the stabilized major neurite. Scale bar, 50 μm.

The culturing procedure is based on that developed for crustacean X-organ neurones (Graf et al. 1988; Cooke et al. 1989) with a number of modifications that optimize conditions for culturing the larger and more delicate stomatogastric neurones. Ganglia were removed from adult animals with 0.5 cm lengths of connective, and submerged for 5 min in an antibiotic crab or lobster saline containing 100 units ml−1 penicillin-G, 75 units ml−1 streptomycin (both from frozen powder) and 0.25 μmol ml−1 fungizone (all from Gibco), pH7.6. The dissection and plating were performed in a laminar horizontal flow hood (Baker) using a Wild M5 dissecting microscope equipped with fibre optic illumination. Dissected ganglia were transferred to 35 mm culture dishes containing an enzymatic saline: 2 mg ml−1 collagenase/dispase (Boehringer Mannheim) and 0.4 mg ml−1 elastase I (Sigma) in crab or lobster saline (pH 7.6) sterilized by filtration (0.22 μm pore size). The covered dishes were agitated (22 cycles min−1) on a shaker for l h at room temperature. Somata with neurites were further separated from their surrounding ganglionic tissues by freeing the loosened connective tissue using a fine jet of enzymatic saline blown from a capillary tube drawn down to an inner tip diameter of 10–20 μm. The dangling somata, held to the ganglion by their neurites, were rinsed of enzyme with two changes of filtered antibiotic saline. Unidentified somata were then individually aspirated using ethanol-sterilized capillary tubes, pulled and fire-polished to yield a smooth tip with an inside diameter of roughly 100 μm. The neurones were plated onto 35 mm culture dishes (for substrata, see below) containing 2ml of defined medium; the fluid depth was approximately 2 mm. The defined medium was prepared by mixing sterile Leibowitz-15 (L-15, Gibco) with an equal volume of buffered doublestrength crab or lobster saline (pH 7.9, 20 mmol l−1 Hepes) containing nutrient and antibiotic constituents. At final volume, these were: 120 mmol l−1 D-glucose, 2mmol l−1 L-glutamine (from frozen stock) and 0.l mg ml−1 gentamicin. Without adjustments, the final medium pH was usually near the optimal value, 7.7. Medium was usually made just prior to use. The substrata used included uncoated Primaria dishes (Falcon 3801) and Falcon 3001 dishes coated with sterile 0.l mg ml−1 poly-L-lysine (Sigma). After neurone dispersal, the dishes were covered and not moved for at least 2 h to aid neuronal adhesion to the substratum. They were then transferred to a sterile, humidified portable incubator chamber (Billups Rothanberg), and placed in the dark at room temperature (20–21 °C).

Outgrowth generally occurred from sites where membrane was in direct contact with the substratum. If only the soma contacted the substratum, with primary neurite suspended in the medium or severed during dissociation, it sometimes adhered via sparse, thin, membraneous protrusions that clung to the dish after 1 day in culture (not shown). Extensive outgrowth occurred if the severed end of the primary neurite adhered to the substratum. Two forms of outgrowth predominated: a large, thin, veiling (lamellipodial) type (Fig. ID), and a directed branching type (Figs 1A, 2). These forms were not mutually exclusive (Fig. 1B,C) and, when both types were present, the neurone’s growth was in transition, usually from veiling to branching. Maximum veil size was typically reached by day 2 and was maintained for a week in static culture. The breadth reached by the veil roughly equalled the soma diameter. The margins of veils were skirted with growth cones that appeared phase dark under phase contrast optics (not shown).

Neurones displaying the more directed (linear) growth morphology (Fig. 2) developed neuritic extensions that represented the stabilization of processes following outgrowth from one or a few motile growth cones. The neuritic processes arose from the periphery of a veil within a day of plating (compare day 1 and day 2 of Fig. 2). Well-stabilized regions of the major neurites often appeared to have filamentous structures in parallel array (Fig. 2, day 2).

The neurones having well-directed outgrowth continued to show process elongation in static culture up to day 7, at which time the cultures were used for other purposes. Typical neurite extension rates may be seen from the development of the leading growth margins of the Cardisoma neurone shown in Fig. 2: from day 1 to day 2 the primary neurite elongated 165 μm, and from day 2 to day 3 the new growth traversed 143 μm. These growth rates varied from cell to cell with a range of 80–170 μm day−1.

The stomatogastric somata and primary neurites in situ appear from ultrastructural observations to be entirely covered by an unbroken glial sheath (King, 1976), which may explain why it was necessary to change the enzymatic regime from that used for the X-organ neurones. These appear to separate more easily from surrounding glia or other material. Successful outgrowth appears to depend largely on whether conditions are provided that favour neuronal adherence and direct contact with the substratum.

Since these neurones grow vigorously in a simple medium that is completely defined, exploration of conditions and factors that modify cell outgrowth and electrical responses should be facilitated.

We thank Professor D. K. Hartline for supplying animals for culture experiments, encouragement and suggestions; Dr B. R. Jones for suggestions on the manuscript; and Ms B. A. Tomiyasu for demonstrating the dissection. This research was supported by National Institutes of Health grant NS15453 and National Science Foundation grant BNS84–04459 to IC, National Institutes of Health grants to the University of Hawaii (Research Centers in Minority Institutions G12 RR03061, and Biomedical Research Support), and by the University of Hawaii Foundation.

Cooke
,
I.
,
Graf
,
R.
,
Grau
,
S.
,
Haylett
,
B.
,
Meyers
,
D.
and
Ruben
,
P.
(
1989
).
Crustacean peptidergic neurons in culture show immediate outgrowth in simple medium
.
Proc. natn. Acad. Sci. U.S.A
.
86
,
402
406
.
Graf
,
R. A.
,
Grau
,
S.
,
Haylett
,
B.
and
Cooke
,
I. M.
(
1988
).
Crustacean peptidergic neurons in primary culture show immediate outgrowth in defined media
.
Neurosci. Abstr
.
14
,
869
.
King
,
D.
(
1976
).
Organization of crustacean neuropil. II. Distribution of synaptic contacts on identified motor neurons in lobster stomatogastric ganglion
.
J. Neurocytol
.
5
,
239
266
.
Krenz
,
W. D.
and
Fischer
,
P.
(
1988
).
Ionic conductances in crustacean stomatogastric neurons in primary culture
.
Neurosci. Abstr
.
14
,
295
.
Selverston
,
A. I.
and
Moulins
,
M.
(
1987
).
The Crustacean Stomatogastric System
.
New York
:
Springer-Verlag
.