Characterization of the Aplysia californica Cerebral Ganglion
F Cluster
STANISLAV S. RUBAKHIN, LINGJUN LI, TATIANA P. MOROZ, AND JONATHAN V. SWEEDLER
Department of Chemistry and Beckman Institute, University of Illinois, Urbana, Illinois 61801
the F cluster contains $50 small (,50 mm), whitish neurons
that have mostly silent or irregular spontaneous discharges, and
40 – 60% of studied F-cluster neurons respond to electrical
stimulation of cerebral nerves. The F clusters may be involved
in egg-laying behavior; Ferguson (Ferguson et al. 1989) found
that local electrical stimulation of F clusters induced a short
afterdischarge in bag cell clusters. After the cerebral ganglion
was backfilled from the right pleuroabdominal connective, a
group of four to six stained cells was visible in the F-cluster
region. Except for demonstration of angiotensinogen-, urotensin I- and urotensin II-like immunoreactivities of F-cluster
neurons (Gonzalez et al. 1992, 1995), little is known about the
biochemical properties of F-cluster cells.
We have performed comparative study of morphological,
electrophysiological, and chemical characteristics of the Fcluster neurons. We find that the cluster contains at least three
distinct neuronal subpopulations and that these neurons are
organized into a unique layered structure within the cluster.
Using mass spectrometry, we show that all three populations of
neurons have distinct and unique mass of peptides, and these
peptides are transported out of the ganglion to specific nerves.
The morphologically distinct neurons in the top layer of the F
cluster (CFT neurons) appear to be electrically coupled neuroendocrine cells that are similar to the light green cells of
Lymnaea stagnalis.
METHODS
INTRODUCTION
Animal and cell preparation
The cerebral ganglia of Aplysia californica play an essential
integrative role for receiving, transmitting, and processing information originating from the pedal, pleural, buccal ganglia,
and peripheral sensory sites such as rhinophores, anterior tentacles, mouth, and eyes. To date, a number of cerebral ganglion
neurons, neuronal groups, and pathways have been identified
and their functional properties described (Clatworthy and
Walters 1994; Kuenzi and Carew 1994; Rosen et al. 1983;
Strack and Jacklet 1993; Teyke et al. 1989; Xin et al. 1995; Xin
et al. 1996).
Although considerable progress has been made in determining the functional properties of several neurons and neuronal
groups, most cerebral neurons and neuronal clusters are poorly
characterized. Perhaps the least-characterized clusters on the
cerebral ganglia dorsal side are the medially located F clusters.
According to Jahan-Parwar (Jahan-Parwar and Fredman 1976),
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked ‘‘advertisement’’
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Forty-six cerebral ganglia were used to study the morphological
and electrophysiological properties of the cerebral F-cluster neurons.
A. californica (100 –200 g) were obtained from Aplysia Research
Facility (University of Miami, Miami, FL) and kept in an aquarium
containing continuously circulating, aerated and filtered artificial sea
water (ASW) at 14 –15°C until used. Animals were anesthetized by
injection of isotonic MgCl2 (;30 to ;50% of body wt) into the body
cavity. The head ganglia (cerebral, pedal, and pleural) were dissected
and placed in ASW containing (in mM) 460 NaCl, 10 KCl, 10 CaCl2,
22 MgCl2, 6 MgSO4, and 10 HEPES, pH 7.7 or in ASW-antibiotic
solution: ASW containing 100 units/ml penicillin G, 100 mg/ml streptomycin, and 100 mg/ml gentamicin, pH 7.7. In some experiments
low-Ca21 ASW was used, which contained (in mM) 460 NaCl, 10
KCl, 0.5 CaCl2, 32 MgCl2, 6 MgSO4, and 10 HEPES, pH 7.7.
Removing the nervous system sheath without protease treatment or
in hyperosmotic ASW resulted in damage of the top layer neurons in
the F cluster. Therefore the sheath was digested enzymatically by
incubating the ganglia in ASW-antibiotic solution containing 1% of
protease (Type IX: Bacterial; Sigma, St. Louis, MO) at 36°C for 1–3
h depending on animal size. Next, the ganglia were washed in fresh
ASW, and the dorsal side of the cerebral ganglia was desheathed.
0022-3077/99 $5.00 Copyright © 1999 The American Physiological Society
1251
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 17, 2017
Rubakhin, Stanislav S., Lingjun Li, Tatiana P. Moroz, and
Jonathan V. Sweedler. Characterization of the Aplysia californica
cerebral ganglion F cluster. J. Neurophysiol. 81: 1251–1260, 1999.
The cerebral ganglia neurons of Aplysia californica are involved in
the development and modulation of many behaviors. The medially
located F cluster has been characterized using morphological, electrophysiological and biochemical techniques and contains at least
three previously uncharacterized neuronal population. As the three
subtypes are located in three distinct layers, they are designated as top,
middle, and bottom layer F-cluster neurons (CFT, CFM, and CFB). The
CFT cells are large (92 6 25 mm), white, nonuniformly shaped, and
located partially in the sheath surrounding the ganglion. These neurons exhibit weak electrical coupling, the presence of synchronized
spontaneous changes in membrane potential, and a generalized inhibitory input upon electrical stimulation of the anterior tentacular (AT)
nerve. Similar to the CFT neurons, the CFM neurons (46 6 12 mm) are
mainly silent but do not show electrical coupling or synchronized
changes in membrane potential. Unlike the CFT neurons, the CFM
neurons exhibit weak action potential broadening during constant
current injection. Comparison of the peptide profiles of CFT, CFM,
and CFB (10 –30 mm) neurons using matrix-assisted laser desorption/
ionization time-of-flight mass spectrometry demonstrates distinct peptide molecular weights for each neuronal subtype with the masses of
these peptides not matching any previously characterized peptides
from A. californica. The mass spectra obtained from the AT nerve are
similar to the CFT neuron mass spectra, while upper labial nerve
contains many peptides observed in the CFM neurons located in
nongranular neuron region.
1252
S. S. RUBAKHIN, L. LI, T. P. MOROZ, AND J. V. SWEEDLER
Using 0.38-mm-diam tungsten wire (WPI, Sarasota, FL), the ganglia
were pinned dorsal side up to a silicone elastomer (Sylgard, Dow
Corning, Midland, MI) layer in a recording chamber containing 3– 4
ml of ASW-antibiotic media. The isolated ganglia preparation was
incubated in ASW-antibiotic solution at 14°C for $1 h before each
experiment.
Morphology
Electrophysiology
Small-diameter polyethylene suction electrodes, an A-M Systems
(Carlsborg, WA) differential AC amplifier (Model 1700) and an
isolated pulse stimulator (Model 2100) were used for extracellular
recording and stimulation of the anterior tentacular nerve (AT) (AT,
Jahan-Parwar and Fredman 1976; C2, Hoffmann 1939) nerve. Intracellular recording and stimulation were made using glass microelectrodes pulled from 1 mm borosilicate glass capillaries (WPI, Sarasota,
FL) and filled with 506.2 mM KCl and 5 mM HEPES solution (pH
7.6, 8 –15 MV). Signals were amplified with Axon Instruments (Foster City, CA) AxoClamp 2B amplifier, monitored and stored with a
Millennia Plus computer (Micron Electronics, Nampa, ID) using a
Digidata 1200A D/A - A/D converter (Axon Instruments, Foster City,
CA). The software packages pClamp 6 and AxoScope 1 (Axon
Instruments) were used for data acquisition and analysis. Records
were digitized at 2–10 kHz and filtered with low-pass Bessel filter at
10-kHz cutoff frequency. To reduce the digitization noise high-resolution records with signal amplitude #5 mV were digitally low-pass
filtered at 10 Hz if not otherwise stated. As currents .5 nA injected
through one recording microelectrode induced changes in potential
detected by a second microelectrode, we used a maximal stimulus of
3 nA to study synaptic connections between neurons. All experiments
were performed at room temperature (23–24°C).
Mass spectrometry
As matrix-assisted laser desorption/ionization time-of-flight mass
spectrometry (MALDI-TOF MS) is not tolerant of high concentrations of cations, the salts associated with cells and nerve tissue were
removed as previously described (Garden et al. 1996). Briefly, after
pinning down the cerebral ganglia, the physiological saline was replaced by an aqueous MALDI matrix solution, 10 mg/ml of 2,5-
FIG. 1. Schematic diagram of cerebral ganglia showing location of neuronal clusters, with shading demonstrating position of F-cluster top-layer (CFT)
and CFT-like neurons in the cerebral ganglia. C-B, cerebral-buccal connective;
C-PL, cerebral-pleural connective; C-P, cerebral-pedal connective; ULAB,
upper labial nerve; AT, anterior tentacular nerve; PT, posterior tentacular
nerve; O, optic nerve; LLAB, lower labial nerve. Nerve and cluster classification according to Jahan-Parwar and Fredman (1976).
dihydroxybenzoic acid (DHB; ICN Pharmaceuticals, Costa Mesa,
CA). Individual F cells were isolated based on their position, size, and
pigmentation. Tungsten needles were used to transfer each cell onto a
MALDI sample probe containing 0.5 mL DHB solution. Peripheral
nerve samples were prepared in a similar manner; a short section (,1
mm) of the interior of the nerve was isolated and placed on one of the
sample spots with DHB matrix solution. After drying at the room
temperature, samples either were inserted into a mass spectrometer for
immediate analysis or stored in the freezer for future analysis.
MALDI mass spectra were obtained using a Voyager Elite mass
spectrometer equipped with delayed ion extraction (PerSeptive Biosystems, Framingham, MA). A pulsed nitrogen laser (337 nm) was
used as the desorption/ionization source, and positive-ion mass spectra were acquired using reflectron mode with an accelerating voltage
at 19 –20 kV. Each unsmoothed mass spectrum is the average of
64 –128 laser pulses. The peaks are listed below as present in a
specific group of cells if 70% of the samples analyzed contain that
peak. Mass calibration was performed externally using either bovine
insulin (Sigma) and synthetic Aplysia a-bag cell peptide (BCP; American Peptide, Sunnyvale, CA) or a previously calibrated mass spectrum obtained from bag cells (Garden et al. 1998).
RESULTS
Morphology of the F cluster
As shown in Fig. 1, the two symmetrical F clusters of A.
californica are located in lateral part of the each cerebral
hemiganglion dorsal surface near the cerebral commissure and
contain ;70 white neurons with cell body size .50 mm and
nearly 100 neurons of ,50 mm in diameter. Cell bodies are
distributed in at least three layers (Fig. 2A). The top layer
contains bright white CFT neurons of 50 –100 mm in size that
mechanically contact and sometimes are embedded partially in
the nervous system sheath. Removal of the sheath without
enzyme pretreatment results in these cells being destroyed or
sticking to the sheath layer instead of the cerebral cluster. The
second layer is made up of smaller F-cluster (CFM) neurons
(30 – 60 mm). Two classes of the CFM neurons can be observed
depending on cytoplasm appearance; when viewed under ap-
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 17, 2017
To visualize the vascular system within the cerebral ganglia, Indigo
(Sigma) dispersed in ASW was injected by syringe in right pedal or
parapodial artery. Cross-sections of the cerebral ganglia were cut in
cryostat at 40 –70 mm thickness after dissection, fixation in 4%
paraformaldehyde, and freezing.
To study neuronal morphology, the cells were filled with 2%
Lucifer yellow (Sigma) or 15% lysine-fixable biotin dextran amine
(NeuroTrace BDA-10,000 Neuronal Tracer Kit, Molecular Probes,
Eugene, OR). The dyes were injected iontophoretically into neurons
using glass pipettes with resistance 1–2 MV for 1–2 h. The neurons
retained physiological properties (action potentials and electrical coupling) during and after injection. After 24 – 48 h incubation at 14°C in
ASW-antibiotic solution, the ganglia were fixed with 4% paraformaldehyde in 0.1 M PBS pH 7.5 for 2 h at 4°C. The avidin-horseradish
peroxidase solution (1.0 mg/ml avidin-horseradish peroxidase in PBS
containing 0.3% Triton-X100) was applied for 3 h after washing away
excess fixative with PBS. Rinsing the ganglia with PBS was followed
by addition of hydrogen peroxide to a final concentration of ;0.5%
for ;30 s and replacing it with 0.05% diaminobenzidine and ;0.5%
H2O2 solution for 3–5 min. Wash, dehydration in ascending concentrations of alcohols and clearing in methyl salicylate finalized preparation of the tissue for microscopic study. The cells were visualized
with fluorescence and transmission microscopy using an Olympus
IMT-2 inverted microscope (Lake Success, NY).
APLYSIA CEREBRAL GANGLION F CLUSTER
1253
FIG. 2. Distribution and morphology of F-cluster neurons. A: 40-mm-thick cross-section of the left cerebral
hemiganglia (magnification, 3160) through the F-cluster
region (dorsal side up). d, F-cluster layers: T, top layer; M,
middle layer; B, bottom layer. <, CFT and *, CFM granular
neurons. Black zone on photo is a vascular system compartment filled by Indigo. Blank areas are not observed
with thicker sections (50 – 60 mm). B: schematic representation of 2 CFT neuron somata locations and their collateral
projections through the neuropil to the AT nerves as determined by injection of biotin dextran amine (n 5 6) or
Lucifer yellow (n 5 7) with C being an expanded view of
the fine structure of CFT neuron terminals in the AT nerve.
s, putative region of signal molecule release (C).
somata are nonuniformly shaped. The axons travel throughout neuropil to the AT nerve (Fig. 2B) and have two morphologically distinct regions. Starting from the cell body
and ending near of the AT nerve base (Fig. 2C), the first
region is characterized by the absence of visible varicosities
and arborization. The second region lies mainly in the AT
nerve and displays multiple branching and varicosities,
which often are places of sharp changes in the direction of
the axon propagation. The diameter of the branches is most
commonly two to three times smaller than the diameter of
main axon. The axon forms many parallel thin branches at
Morphology of CFT neurons
To examine the structure of axon processes, the neurons
were filled electrophoretically with Lucifer yellow or biotin
dextran amine. The CFT neurons are generally unipolar and
have wide range of somata diameters from 50 to 150 mm
[92.6 6 24.8 (SD) mm, n 5 34, Table 1]. The CFT neuron
1.
neurons
TABLE
Electrophysiological characteristics of the CFT and CFM
Cell
CFT Neurons
Diameter, mm
92.6 6 24.8 (34)
Resting potential, mV
254.2 6 12.6 (34)
Input resistance, MV
26.5 6 9.4 (34)
Spike amplitude, mV*
74.7 6 11
(8)
Spike time of peak, ms
13.2 6 2.8 (8)
Spike half-width, ms
8.2 6 0.84 (8)
Max rise of spike, mV/ms
19.5 6 10
(8)
Max fall of spike, mV/ms
28.4 6 0.7 (8)
Spike time of fall slope, ms
18.1 6 4.6 (8)
CFM Neurons
46.1 6 11.7 (20)
235.7 6 10.7 (20)
55.5 6 30
(20)
78.6 6 14.5 (9)
4 6 0.83 (9)
5.4 6 1.5 (9)
51.7 6 27.8 (9)
218.2 6 5
(9)
6.5 6 2
(9)
Values are means 6 SD; n values in parentheses. *Spike amplitude was
measured from spike threshold to spike peak.
FIG. 3. Heterogeneity of electrophysiological parameters of F-cluster neurons. CFT neurons had significantly longer spike half-width and intensive
broadening of action potential (h, top) than the CFM located in granular neuron
region (3, bottom). Neurons were current clamped at 240 mV and stimulated
using 5-nA, 25-ms square pulses. Interval between pulses was 175 ms.
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 17, 2017
propriate lighting, some neurons contain granular material and
others clear cytoplasm. The first type is likely the granular
neurons described by Soinila and Mpitsos (1996). Generally
the granular neurons are located in the lateral part of the F
cluster opposite to the cerebral commissure. Interestingly, it
appears that some granular neurons extend out of the F cluster.
The clear CFM neurons occupy the portion of the middle layer
of the F-cluster located toward the cerebral commissure. Tiny
(10 –30 mm) neurons (CFB) form a third layer located just
above the neuropil. In some preparations, light-orange (80 –
120 mm) neurons are found in the caudal part of F cluster top
layer, whereas in most preparations, such cells are only observed in other regions of the cerebral ganglion. We further
characterize the CFT neurons and the CFM neurons of the F
cluster in the following text.
1254
S. S. RUBAKHIN, L. LI, T. P. MOROZ, AND J. V. SWEEDLER
FIG. 4. CFT neurons exhibited 3-phase spike adaptation to electrical stimulation. A: phase 1 contains fast
action potentials; the absence of action potentials and
membrane potential oscillations characterize phase 2,
whereas phase 3 exhibits broad action potentials (bottom). B: some preparations do not exhibit phase 2.
Action potentials were induced by brief 3-nA square
pulse current injection.
Fredman 1976; C1, Hoffmann 1939), with at least one artery
contacting the F-cluster region. Sections of all three cerebral
ganglia injected with Indigo reveal the presence of a blood
sinus in the granular cell region (Fig. 2A). Generally the AT
nerve has several major arteries and the ULAB a single artery.
Fast green injection into the artery approaching the left cerebral
ganglia induces a quick coloring of the left cerebral hemiganglion, the left AT, and ULAB nerves. No visible color changes
in the right cerebral hemiganglion, right AT, and right ULAB
nerves were observed for several minutes after the injection,
indicating some level of vascular system separation of the
FIG. 5. Synchronous spontaneous changes in
membrane potential of the CFT neurons. A: simultaneous registration of somatic membrane potential of
2 CFT neurons. B: recordings of somatic membrane
potential in CFT neuron (top) and A-cluster neuron
(bottom). C: action potentials appearing in 1 CFT
neuron (top) induced EPSP in target neuron (2,
bottom) but had little influence on membrane potentials (MP). D: depolarization of the neuron significantly (P 5 1.33 3 1029) increased spontaneous
changes in MP (SCMP) amplitude. Recording shows
the CFT neuron SCMP before and after steady depolarizing current injection. - - -, membrane potential.
Spikes are manually truncated in C and D.
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 17, 2017
the nerve surface (Fig. 2C) that may be the locations of
neurotransmitter and/or neuromodulator release into the
blood sinuses. It is difficult to make conclusions about the
placement of the axon in the nerve periphery because of the
protease treatment and potential mechanical disruptions due
to removal of the nerve sheath.
Indigo was injected into the right pedal or parapodial artery
to locate the vascular system within the cerebral ganglia and
cerebral nerves. A well-developed vascular system is observed
on both the dorsal and ventral sides of the cerebral ganglia, AT
and upper labial (ULAB) nerves (ULAB, Jahan-Parwar and
APLYSIA CEREBRAL GANGLION F CLUSTER
1255
cerebral hemiganglia. The location of the CFT neuron varicosities near these vascular structures may be important for the
rapid distribution of signal molecules released from the CFT
and granular neurons.
Morphology of CFM neurons
The CFM neurons located in granular neuron region are
unipolar and have smaller, round somata (46.1 6 11.7 mm, n 5
20) than CFT neurons. The neurons are not disrupted to any
great extent during the desheathing process due to their location in the second lower layer. The CFM axons travel toward
the space between the cerebral hemiganglia and there develop
branches and varicosities (not shown).
Electrophysiology
The CFT and CFM neurons are not only topologically and
morphologically distinct but have significantly different electrophysiological parameters. These properties are summarized
in Table 1, with aspects of action potential (AP) broadening
illustrated in Fig. 3. The CFT neurons become spontaneously
active generally after depolarization to 220 to 230 mV or
immediately after the cell membrane are penetrated by the
microelectrode. A three-phase adaptation of spike frequency to
continuous current injection (3 nA) is found in the CFT neurons
(Fig. 4). The first phase is characterized by fast, high-amplitude
action potentials. The second phase consists of increasing
potential oscillations finally resulting in spike generation,
which is the start of third phase including broad action potentials. Spike frequency in the first phase was higher than in the
third phase mainly because of broadened spikes. The CFM
neurons predominantly had only one or few fast action potentials at the start of continuous depolarization (3–5 nA).
The spontaneous changes occurred in somatic membrane
potential of the CFT neurons (Fig. 5). The spontaneous changes
in membrane potential (SCMP) had different levels of synchronization in randomly selected CFT neuron pairs. Some neuronal pairs had an almost one-to-one pattern of the SCMP (Fig.
5, A and C), whereas others had different patterns (not shown).
Measurements did not show similar SCMP in neurons from A
(Fig. 5B) and B clusters. Cell depolarization induced significant increase in amplitude of the SCMP (1.63 6 0.73 mV at
250 mV and 3.64 6 2.0 mV at 240 mV, P 5 1.33 3 1029)
and caused brief spike bursts with 10- to 120-s interburst
intervals (Fig. 5D).
To determine the cause of the SCMP, we examined whether
there are synaptic connections between CFT neurons; as shown
in Fig. 6, the CFT neurons are electrically coupled. The coupling coefficient varied from ,0.01 to 0.169 and was not
significantly different for neuronal pairs in the left and right F
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 17, 2017
FIG. 6. Electrical coupling of CFT neurons. A: locations of electrically coupled (gray
ovals) and not coupled cells (striped ovals) to
test neurons (filled ovals) in F and C clusters.
LF/RF, left/right F clusters; LC/RC, left/right
C-clusters, H cluster. B, top: simultaneous
recording from pair of the neighboring CFT
neurons from left F cluster. Action potentials
in presynaptic neuron elicited electrotonic
postsynaptic potentials in postsynaptic cell.
Coupling coefficient for this pair was 0.147.
B, bottom: registration of pair electrically
coupled right F-cluster CFT neuron (top) and
right C-cluster CFT-like neuron (bottom). Distance between their somata was ;250 mm
and coupling coefficient was 0.038.
1256
S. S. RUBAKHIN, L. LI, T. P. MOROZ, AND J. V. SWEEDLER
clusters: the left F cluster, 0.048 6 0.045 (n 5 49); the right F
cluster, 0.049 6 0.026 (n 5 38). Careful measurements of
interactions between the CFT neurons of opposite F clusters
revealed seldom and weak electrical coupling with a coupling
coefficient 0.012 6 0.0046 (n 5 3). The CFT neurons were
electrically coupled to top layer neurons of C cluster (Fig. 6, A
and B). It is, however, not clear how many neurons were
electrically coupled. In one CNS, for example, we located 24
coupled neuronal pairs using two test neurons (Fig. 6, A and B),
whereas in another CNS, only 6 coupled pairs were found
among different neurons. As expected, some neurons had detectable coupling to a specific neuron but not to each other. We
examined the effect of low-Ca21 ASW on the electrical junction between the CFT neurons by gradually lowering the calcium concentration, which did not abolish the electrical coupling (not shown). No significant difference in coupling
coefficient was found when two neurons current-clamped at
270 mV were injected with negative (21 nA; coupling coefficient 0.023 6 0.006, n 5 5) or positive current (11 nA;
coupling coefficient 0.027 6 0.003, n 5 5) pulses, indicating
that electrical coupling was not rectifying.
Presynaptic action potentials in pairs of the CFT neurons
with high coupling coefficients evoked EPSPs, but they were
of insufficient amplitude to elicit spike generation in postsynaptic neurons. Figure 7A demonstrates the time course of the
EPSPs (top) in a postsynaptic neuron after evoked action
potential generation in a presynaptic neuron (bottom). Only
neuronal pairs with a high coupling coefficient had a detectable
EPSP. The amplitude of the EPSPs increased during action
potential trains in the presynaptic neuron (Fig. 7B) that corresponded to spike broadening but not to spike amplitude
changes (Fig. 7C).
Electrical stimulation of the different regions of the AT
nerve was performed to determine the location of the CFT
Peptide profiling in F cluster by MALDI-TOF MS
MALDI mass spectrometric peptide profiling of F-cluster
cells isolated based on cell size and position reveals the presence of multiple peaks for each cell type. Essentially all peaks
in cellular MALDI-TOF mass spectra correspond to the protonated molecular weights of peptides. Figure 9 shows the
typical mass spectra of individual CFT, CFM and CFB neurons,
with consistent, characteristic peak patterns for each type. As
indicated in Table 2, a subset of putative peptides (mass peaks
at 1,182, 1,600, 1,526, 2,999, 3,735, 4,137, 4,620, 6,053, 6,110,
and 7,410) are present exclusively in CFT neurons. The characteristic masses of the peptides in small CFB neurons include
1,387, 1,433, 1,690, 3,389, and 3,775. The CFM neurons contained only one unique peptide 1247 and many molecules
observed in the CFT and CFB cells. In addition, two distinct
mass profiles are detected for these cells, indicating additional
heterogeneity in the CFM cell population. Individual cellular
mass spectra were obtained from 46 CFT cells, 40 CFM cells,
and 44 CFB cells from 10 animals.
The MALDI analysis of short sections of the ULAB (C1)
and AT nerve (C2) also revealed the presence of two reproducible and distinct spectra similar to those detected in the
F-cluster cells (20 samples from 4 animals). As shown in Figs.
FIG. 8. Response of the CFT neurons to electrical stimulation of the AT
nerve in normal ASW (left and right) and its disappearance in low Ca21 ASW
(middle). Spikes are manually truncated (top left record).
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 17, 2017
FIG. 7. Time course of electrotonic postsynaptic potentials. A: delay in
response of postsynaptic CFT neuron (top) to action potential generation in
presynaptic neuron (bottom). Inset: full record with the position in the record
(- - -). B: electrical coupling between 2 CFT neurons characterized by gradual
increasing amplitude of EPSPs (top) through action potential train in presynaptic neuron (bottom) elicited by current injection. C: changes in action
potential amplitude (■) and EPSP (E). Signals were filtered at 10 kHz, using
AxoClamp 2B internal filter. Presynaptic neuron was injected with current 3 nA.
terminals in the nerve. Stimulation of the AT nerve at the first
branching area resulted in either an initial low amplitude depolarization followed by prolonged hyperpolarization (24.1 6
1.4 mV, n 5 17) or just the hyperpolarization (Fig. 8). LowCa21 ASW eliminated the hyperpolarization, indicating the
involvement of chemical synapses in this effect. Surprisingly
stimulation of the posterior part of the left AT nerve produced
excitation in both the left and right F-cluster CFT neurons. This
effect indicated either strong electrical coupling between the
CFT neurons or the presence of processes from the opposite
F-cluster CFT neurons in the AT nerve that were not observed
during our morphological investigation. Jahan-Parwar and
Fredman (1976) demonstrated that the stimulation of the AT
nerve produced depolarization in ;60 –100% of the tested Fand C-cluster neurons, whereas hyperpolarization was detected
only in 7– 40% cases. As the sheath was removed without
enzymatic pretreatment, CFT neuron damage was possible. We
have not attempted to electrophysiologically characterize the
CFB neurons because of the small size of the cell bodies and
their location below the CFM neurons.
APLYSIA CEREBRAL GANGLION F CLUSTER
1257
FIG. 9. Matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectra from single F-cluster neurons: CFT (A), CFM
(B), and CFB (C).
DISCUSSION
Several authors have described complex structure of the A.
californica ganglia (Frazier et al. 1967; Kandel 1979; Vigna et
al. 1984). However, it has not been clear what is the functional
role for the layered organization and whether the neurons in
each layer are truly distinct. Using physiological, morphological, and biochemical approaches, we demonstrate that the
neurons located in the top, middle, and bottom layers of the
cerebral ganglion F cluster are distinct neuronal populations.
Important implications for the vertical distribution of cell bodies within the cluster include the distance from the vascular
system (or distance from the neuropil) and an increased number of neurons within a given area.
Several lines of evidence indicate the CFT neurons are a
homogeneous group of cells. They have similar morphology
and peptide profiles. Synchronized changes in somatic membrane potential of the CFT neurons are found in both F clusters.
The SCMP can be explained partially by electrical coupling
between the neurons. Stimulation of the AT nerve reveals a
putative external source of coordination of the CFT neurons as
the stimulation of the AT nerve induces prolonged hyperpolarization of the neurons. A possible functional role of the
hyperpolarization and synchronized membrane potential
changes is a shift of membrane potential below the potential of
calcium channel activation. Thus low-threshold calcium currents can evoke action potential formation as in the case of
GABA-induced hyperpolarization of a thalamocortical neurons
(Crunelli and Leresche 1991). Interestingly, Jahan-Parwar
(1972) found that application of the sperm sac content, food,
and the sexual gland compounds to A. californica oral veil
induced an increase the electrical activity in the AT nerve.
The morphology and physiology of the F-cluster cells suggests neuroendocrine roles for these cells. To confirm this, as
well as substantiate the multiple distinct populations of Fcluster neurons, individual neurons have been assayed using
MALDI-TOF MS. This technique has become a powerful
method for single cell peptide profiling (Garden et al. 1996,
1998; Jimenez et al. 1998; Li et al. 1994). MALDI-TOF MS
provides accurate mass (.0.01%) of the peptides present in
individual cells. As described in Fig. 9, MALDI reveals the
consistent presence of numerous peaks from 1,000 –10,000 Da.
We expect that most, if not all, of these peaks are from peptides
based on the extensive prior studies on MALDI of invertebrates (Garden et al. 1996, 1998; Jimenez et al. 1998; Li et al.
1994). As the masses of these peptides do not fit masses of
known Aplysia peptides, these represent previously uncharacterized peptides. As shown in Fig. 9, at least three different
patterns of peptide profiles are detected as a function of cell
size and location. This result is in a good agreement with the
morphological and electrophysiological data obtained from the
F cluster.
Several peptides, which have molecular mass of 1,455,
2,985, 4,575, and 9,148 Da, are present in most F-cluster
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 17, 2017
9A and 10A, the peak profiles seen in the AT nerve are similar
to the CFT neurons, whereas the ULAB nerve contains many
peptides observed in the CFM neurons (Figs. 9B and 10B).
Other tested nerves such as pleural-abdominal, cerebral-pleural, and cerebral-pedal do not show the presence of the Fcluster neuron peptides (Li et al. 1998)
1258
TABLE
S. S. RUBAKHIN, L. LI, T. P. MOROZ, AND J. V. SWEEDLER
2.
Putative neuropeptides in F cluster
CFB
CFM
CFT
1049
1148
1182
1247
1387
1433
1455
1526
1600
1690
1714
1764
1803
2041
2553
2569
2609
2810
2985
2999
3389
3735
3765
3775
4137
4215
4314
4373
4575
4620
5014
5022
6053
6110
7410
8429
8628
9149
11
11
—
—
11
11
11
—
—
11
—
1
—
—
1
—
1
—
11
—
1
—
1
1
—
11
—
11
11
—
—
—
—
—
—
1
1
1
11
11
—
11
—
—
11
—
—
—
11*
11
11*
1
1
—
1
1
11
—
—
—
1
—
—
11
11
11
11
—
1
1
—
—
—
1
1
1
—
—
11
—
—
—
11
11
11
—
11
—
—
11
—
1
—
1
1
11
—
1
—
—
11
—
11
—
11
11
1
1
1
1
11
—
11
11
n 5 44 for CFB, 40 for CFM, and 46 for CFT. The monoprotonated molecular
masses are listed as present in a specific group of cells if 70% of the samples
analyzed contain the peak. 11, molecular masses listed have relative intensity
more than 20% of the most intense peak; 1, molecular masses listed have
relative intensity ,20% of the most intense peak. *Only seen in granular cells.
FIG. 10. MALDI-TOF mass spectra from the
AT (A) and ULAB (B) nerves showing a large
number of similar peptide masses to the single-cell
spectra shown in Fig. 9. Buccalins S and G (BucS,
BucG) are neuropeptides described by Miller et al.
(1993).
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 17, 2017
Molecular Mass, Da
neurons. Possible reasons for the same peptides to be observed
in all cells include they are present in all neurons and, the most
likely, the peaks may be due to axons, terminals from some
neurons and pieces of neuropil being present in all layers/
samples. For example, we have frequently observed bag cell
peptides when examining both the left upper quadrant (LUQ)
cells and the R3–14 neurons in the abdominal ganglion, which
we have attributed to processes and terminals from the bag
cells (Garden et al. 1998). This potentially explains why we
observe many peaks in the CFM cells that also are observed
with either the CFT or the CFB neurons. The CFM cells are
located between the two other layers and so the possibility of
contamination from the other layers is higher. However, as
many peaks are only observed in a single cell type, the MALDI
data substantiates observations that the F-cluster neurons contain at least three distinct neuron types and suggests that all
types use peptides as extracellular signal molecules. Interestingly, on the basis of the peptide profile, the CFM neurons may
consist of two subclasses, potentially the so-called granular
neurons described by Soinila and Mpitsos (1996) and another
set of undescribed neurons. The same peptide profiles observed
in the CFT and CFM cells are observed in the cerebral AT and
ULAB connectives; this provides evidence that the molecular
masses detected in F-cluster neurons are likely a unique set of
putative peptide transmitters that are transported out of the
cerebral ganglion.
Although single-cell MALDI-TOF MS provides strong evidence for complex and different sets of peptides in each
neuronal subclass, the list of peptide masses presented in Fig.
9 and Table 2 may not be complete. Peptides with a mass ,900
Da are hard to identify or detect due to the presence of intense
lipid peaks in cellular samples. Thus peptides such as APGWamide, FMRFamide, or other low-molecular-weight peptides
were not detected in the observed mass range.
Given the large number of peptides observed, an important
question is are these truly unknown peptides? Although peptides of these masses have not been previously described, they
may closely resemble other known peptides with slight modifications. Several peptides have been suggested to be present in
the F cluster. For example, Gonzalez et al. (1992) reported
APLYSIA CEREBRAL GANGLION F CLUSTER
The assistance of and discussions with Drs. Leonid L. Moroz and Rhanor
Gillette are acknowledged gratefully as is the assistance of C. Sticha with mass
spectra acquisition.
This research was supported partially by National Institute of Neurological
Disorders and Stroke Grant NS-31609, National Science Foundation Grant
CHE-96 –22663, and the Sloan Foundation. A. californica were provided
partially by the NCRR National Resource for Aplysia at the University of
Miami under Division of Research Resources Grant RR-10294.
Address for reprint requests: J. V. Sweedler, 600 S. Mathews Ave., University of Illinois, Urbana, IL 61801.
Received 16 July 1998; accepted in final form 20 November 1998.
REFERENCES
BENJAMIN, P. R. AND ROSE, R. M. Electrotonic coupling and afterdischarges in
the Light Green Cells: a comparison with two other cerebral ganglia neu-
rosecretory cell types in the pond snail. Lymnaea stagnalis. Comp. Biochem.
Physiol. A Physiol. 77: 67–74, 1984.
CHIU, A. Y. AND STRUMWASSER, F. Two neuronal populations in the head
ganglia of Aplysia californica with egg-laying hormone-like immunoreactivity. Brain. Res. 294: 83–93, 1984.
CLATWORTHY, A. L. AND WALTERS, E. T. Comparative analysis of hyperexcitability and synaptic facilitation induced by nerve injury in two populations
of mechanosensory neurones of Aplysia californica. J. Exp. Biol. 190:
217–238, 1994.
CONLON, J. M., O’HARTE, F., SMITH D. D., BALMENT, R. J., AND HAZON, N.
Purification and characterization of urotensin II and parvalbumin from an
elasmobranch fish, Scyliorhinus canicula (common dogfish). Neuroendocrinology 55: 230 –235, 1992a.
CONLON, J. M., O’HARTE, F., SMITH, D. D., TONON, M. C., AND VAUDRY, H. Isolation
and primary structure of urotensin II from the brain of a tetrapod, the frog
Rana ridibunda. Biochem. Biophys. Res. Commun. 188: 578 –583, 1992b.
CRUNELLI, V. AND LERESCHE, N. A role for GABAB receptors in excitation and
inhibition of thalamocortical cells. Trends Neurosci. 14: 16 –12, 1991.
FERGUSON, G. P., TER MAAT, A., AND PINSKER, H. M. Egg laying in Aplysia.
II. Organization of central and peripheral pathways for initiating neurosecretory activity and behavioral patterns. J. Comp. Physiol. [A] 164: 849 –
857, 1989.
FRAZIER, W. T., KANDEL, E. R., KUPFERMANN, I., WAZIRI R., AND COGGESHALL R. E.
Morphological and functional properties of identified neurons in abdominal
ganglion of Aplysia californica. J. Neurophysiol. 30: 1288 –1351, 1967.
GARDEN, R. W., MOROZ, L. L., MOROZ, T. P., SHIPPY, S. A., AND SWEEDLER,
J. V. Excess salt removal with matrix rinsing: direct peptide profiling of
neurons from marine invertebrates using matrix-assisted laser desorption/
ionization time-of-flight mass spectrometry. J. Mass. Spectrom. 31: 1126 –
1130, 1996.
GARDEN, R. W., SHIPPY, S. A., LI, L., MOROZ, T. P., AND SWEEDLER, J. V.
Proteolytic processing of the Aplysia egg-laying hormone prohormone.
Proc. Natl. Acad. Sci. USA 95: 3972–3977, 1998.
GOMOT, A., GOMOT, L., MARCHAND, C. R., COLARD, C., AND BRIDE, J. Immunocytochemical localization of insulin-related peptide(s) in the central nervous system of the snail Helix aspersa Muller: involvement in growth
control. Cell Mol. Neurobiol. 12: 21–32, 1992.
GONZALEZ, G. C., MARTINEZ-PADRON, M., LEDERIS, K., AND LUKOWIAK, K.
Distribution and coexistence of urotensin I and urotensin II peptides in the
cerebral ganglia of Aplysia californica. Peptides 13: 695–703, 1992.
GONZALEZ, G. C., ROGER, I., GONZALEZ, E., MARTINEZ-PADRON, M., MOORE,
G. J., AND LUKOWIAK, K. Angiotensinogen-like epitopes are present in the
CNS of Aplysia californica and co-localize with urotensin I- and urotensin
II-like immunoreactivities in the cerebral ganglia. Neuroreport 6: 541–544,
1995.
HOFFMANN, H. In: Opisthobranchia. In Klassen und Ordnungen des Tierreichs. Mollusca. Gastropoda. Opisthobranchia, edited by H. G. Bronn.
Leipzig: Akademische Verlagsgesellschaft M.B.H., 1939, part 1, vol. 3, sect.
2, book 3.
JAHAN-PARWAR, B. Behavioral and electrophysiological studies on chemoreception in Aplysia. Am. Zoologist 12: 525–537, 1972.
JAHAN-PARWAR, B. AND FREDMAN, S. M. Cerebral ganglion of Aplysia: cellular
organization and origin of nerves. Comp. Biochem. Physiol. A Physiol. 54:
347–357, 1976.
JIMENEZ, C. R., LI, K. W., DREISEWERD, K., SPIJKER, S., KINGSTON, R.,
BATEMAN, R. H., BURLINGAME, A. L., SMIT, A. B., VAN MINNEN, J., AND
GERAERTS, W. P. Direct mass spectrometric peptide profiling and sequencing of single neurons reveals differential peptide patterns in a small neuronal
network. Biochemistry 37: 2070 –2076, 1998.
KANDEL, E. R. Behavioral Biology of Aplysia. San Francisco: W. H. Freeman,
1979.
KUENZI, F. M. AND CAREW, T. J. Head waving in Aplysia californica—
interganglionic pathways underlying the coordination and control of searching movements. J. Exp. Biol. 195: 75–90, 1994.
LI, K. W., HOEK, R. M., SMITH, F., JIMENEZ, C. R., van der Schors, R. C., van
Veelen, P. A., Chen, S., van der Greef J., Parish, D. C., and Benjamin, P. R.
Direct peptide profiling by mass spectrometry of single identified neurons
reveals complex neuropeptide-processing pattern. J. Biol. Chem. 269:
30288 –30292, 1994.
LI, L., MOROZ, T. P., GARDEN, R. W., FLOYD, P. D., WEISS, K. R., AND
SWEEDLER, J. V. Mass spectrometric survey of interganglionically transported peptides in Aplysia. Peptides 19: 1425–1433, 1998.
MILLER, M. W., BEUSHAUSEN, S., CROPPER, E. C., EISINGER, K., STAMM, S.,
VILIM, F. S., VITEK, A., ZAJC, A., KUPFERMANN, I., BROSIUS, J., AND WEISS,
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 17, 2017
urotensin I and urotensin II immunoreactivities in the F cluster.
The molecular weights of urotensin II varies from 1,364 to
1,526 in four different species (Conlon et al. 1992a,b; Pearson
et al. 1980; Waugh et al. 1995), based on MALDI spectra of
single F cells, we detected multiple peaks between 1,300 and
1,600, and one mass was 1,526. We expect that many of the
F-cluster peptides will be similar to peptides isolated from
related mollusks but with amino acid modifications producing
the unknown masses. In other work, Chiu and Strumwasser
(1984) demonstrated the presence of ELH immunoreactive
neurons near the F cluster. However, we do not detect molecular weights corresponding to ELH nor atrial gland peptides
[with the possible exception that the 6,053 peak is similar to
califin B molecular weight 6,072 (Rothman et al. 1986)].
However, the lack of all other peptides from the califin B gene
in the F cluster indicates this is unlikely a califin.
Because CFT neurons located in the cerebral ganglia clusters
have similar appearance, position, and physiological properties to
the A-type LGC-cluster neurons of L. stagnalis, we hypothesize
that the CFT neurons are homologous to the A-LGC cells of L.
stagnalis. Several lines of evidence support this hypothesis. The
A-LGC occupy the top position in the cluster (Smit et al. 1998),
and both the CFT and the A-LGC have neurohemal areas in the
nerves of cerebral ganglia innervating sensory areas of skin in
mollusks (Benjamin and Rose 1984; Smit et al. 1998). In addition,
both the CFT and the A-LGC are weakly electrically coupled, and
synchronized changes in membrane potential are reported. Last,
electrical stimulation of nerves containing terminals of the CFT
neurons or the A-LGC cells produced their long-term hyperpolarization. Both the A-LGC and the CFT neurons are peptidergic,
but, as expected from different species, the molecular weights of
the peptides do not match.
Interestingly, other mollusks such as Planorbarius corneus
and Helix aspersa have several clusters of light green neurons
with a position on the dorsal surface of cerebral ganglia which
is similar to the location of the F and C clusters of A. californica and the LGC clusters of L. stagnalis. Detailed study of the
LGC of L. stagnalis show their neuroendocrine function including synthesis and release of molluskan insulin-related peptides (Smit et al. 1998). A similar distribution of insulinimmunoreactive cells was found in mollusks P. corneus
(Sonetti et al. 1992) and H. aspersa (Gomot et al. 1992) that
supports the idea of a conserved location of neurosecretory
cells on cerebral ganglia dorsal surface of pulmonates and
opistobranchs. To confirm the homology, the peptides in the F
cluster are being isolated and characterized.
1259
1260
S. S. RUBAKHIN, L. LI, T. P. MOROZ, AND J. V. SWEEDLER
SONETTI, D., VAN HEUMEN W. R., AND ROUBOS, E. W. Light- and electronmicroscopic immunocytochemistry of a molluscan insulin-related peptide in
the central nervous system of Planorbarius corneus. Cell. Tissue Res. 267:
473– 481, 1992.
STRACK, S. AND JACKLET, J. W. Antiserum to an eye-specific protein identifies
photoreceptor and circadian pacemaker neuron projections in Aplysia.
J. Neurobiol. 24: 552–570, 1993.
TEYKE, T., WEISS, K. R., AND KUPFERMANN, I. A subpopulation of cerebral B
cluster neurones of Aplysia californica is involved in defensive head withdrawal but not appetitive head movements. J. Exp. Biol. 147: 1–20, 1989.
VIGNA, S. R., MORGAN, J. L., AND THOMAS, T. M. Localization and characterization of gastrin/cholecystokinin-like immunoreactivity in the central nervous system of Aplysia californica. J. Neurosci. 4: 1370 –1377, 1984.
WAUGH, D., YOUSON, J., MIMS, S. D., SOWER, S., AND CONLON, J. M. Urotensin
II from the river lamprey (Lampetra fluviatilis), the sea lamprey (Petromyzon marinus), and the paddlefish (Polyodon spathula). Gen. Comp. Endocrinol. 99: 323–332, 1995.
XIN, Y., WEISS, K. R., AND KUPFERMANN, I. Distribution in the central nervous
system of Aplysia of afferent fibers arising from cell bodies located in the
periphery. J. Comp. Neurol. 359: 627– 643, 1995.
XIN, Y., WEISS, K. R., AND KUPFERMANN, I. An identified interneuron contributes to aspects of six different behaviors in Aplysia. J. Neurosci. 16:
5266 –5279, 1996.
Downloaded from http://jn.physiology.org/ by 10.220.33.1 on June 17, 2017
K. R. The buccalin-related neuropeptides: isolation and characterization of
an Aplysia cDNA clone encoding a family of peptide cotransmitters. J. Neurosci. 13: 3346 –3357, 1993.
PEARSON, D., SHIVELY, J. E., CLARK, B. R., GESCHWIND I. I., BARKLEY, M.,
NISHIOKA, R. S., AND BERN, H. A. Urotensin II: a somatostatin-like peptide
in the caudal neurosecretory system of fishes. Proc. Natl. Acad. Sci. USA 77:
5021–5024, 1980.
ROSEN, S. C., KUPFERMANN, I., GOLDSTEIN, R. S., AND WEISS, K. R. Lesion of
a serotonergic modulatory neuron in Aplysia produces a specific defect in
feeding behavior. Brain. Res. 260: 151–155, 1983.
ROSEN, S. C., WEISS, K. R., COHEN, J. L., AND KUPFERMANN, I. Interganglionic
cerebral-buccal mechanoafferents of Aplysia: receptive fields and synaptic
connections to different classes of neurons involved in feeding behavior.
J. Neurophysiol. 48: 271–288, 1982.
ROTHMAN, B. S., HAWKE, D. H., BROWN, R. O., LEE, T. D., DEHGHAN, A. A.,
SHIVELY, J. E., AND MAYERI, E. Isolation and primary structure of the
califins, three biologically active egg-laying hormone-like peptides from the
atrial gland of Aplysia californica. J. Biol. Chem. 261: 1616 –1623, 1986.
SMIT, A. B., VAN KESTEREN, R. E., LI, K. W., VAN MINNEN, J., SPIJKER, S., VAN
HEERIKHUIZEN, H., AND GERAERTS, W. P. Towards understanding the role of
insulin in the brain: lessons from insulin-related signaling systems in the
invertebrate brain. Prog. Neurobiol. 54: 35–54, 1998.
SOINILA, S. AND MPITSOS, G. J. Putative neuroendocrine cells in the Aplysia
cerebral ganglion. Neurosci. Res. 25: 209 –216, 1996.