Drugs Affecting Phospholipase C-Mediated Signal Transduction Block
the Olfactory Cyclic Nucleotide-Gated Current of Adult Zebrafish
LI MA AND WILLIAM C. MICHEL
Department of Physiology, University of Utah School of Medicine, Salt Lake City, Utah 84108
INTRODUCTION
Olfactory transduction is initiated when membrane-associated odorant receptors, located on the cilia of olfactory receptor neurons (ORN), bind an appropriate odorant. Results
obtained using a variety of experimental approaches suggest
that one of several distinct transduction cascades may be
activated upon ligand binding (Ache and Zhainazarov 1995;
Dionne and Dubin 1994). The most thoroughly studied
transduction pathways are GTP dependent and use either
cyclic nucleotides (Buck 1996; Zagotta and Siegelbaum
1996; Zufall et al. 1994) or inositol trisphosphate (IP3 )
(Bruch 1996; Honda et al. 1995; Restrepo et al. 1996) as
second messengers. A general strategy often used to associate an odor-evoked response with a specific transduction
cascade requires that one or more drugs specifically perturb
both odor and second-messenger–activated responses. A
problem with this approach is that the drugs affecting olfactory transduction are frequently not specific (Kleene 1994).
For example, the olfactory cyclic nucleotide-gated (CNG)
channel can be blocked by divalent cations (Frings et al.
1992; Zufall and Firestein 1993), calcium/calmodulin antagonists (W-7, trifluoperazine) (Kleene 1994), soluble gua-
nylate cyclase inhibitors [6-anilino-5,8-quinolinedione]
(Leinders-Zufall and Zufall 1995), and epithelial sodium
(amiloride) and calcium channel blockers (diltiazem)
(Frings et al. 1992). The olfactory IP3-activated current also
is blocked by divalent cations (Restrepo et al. 1992) and by
ruthenium red (Restrepo et al. 1990), a drug that interacts
with many calcium-binding proteins (Amann and Maggi
1991) and blocks voltage-gated calcium channels (Hamilton
and Lundy 1995). Because few studies have exhaustively
screened drug effects on all components of odor-activated
transduction pathways and because drug effects may vary
across species, it is essential to experimentally confirm that
drug effects described in one species account for perturbed
odor-evoked responses in other species.
The zebrafish is emerging as an increasingly popular
model for studies of the organization of the olfactory system ( Baier and Korsching 1994; Baier et al. 1994; Byrd
and Brunjes 1995; Friedrich and Korsching 1997; Weth
et al. 1996 ) and its development ( Barth et al. 1996; Byrd
et al. 1996; Hansen and Zeiske 1993; Vogt et al. 1997 ) .
Functional properties of the zebrafish olfactory system
have received less attention. Zebrafish respond behaviorally ( Algranati and Perlmutter 1981; Bloom and Perlmutter 1977; Dill 1974; Steele et al. 1990, 1991; Van Den
Hurk and Lambert 1983 ) and electrophysiologically ( Michel and Lubomudrov 1995 ) to most of the common water-soluble odorants detected by other fish species ( Hara
1992 ) . Most of the amino acid and bile salt odorants tested
appear to interact with at least partially independent odorant receptors ( Michel and Derbidge 1997 ) but the transduction cascades subsequently activated remain unknown.
Odorants stimulate phospholipase C (PLC) activity and
inositol trisphosphate production in the channel catfish
( Restrepo et al. 1990, 1993 ) and Atlantic salmon ( Lo et
al. 1994 ) . Components of the CNG pathway are present
in these fish species [ and the zebrafish ( Barth et al.
1996 ) ] , but at least in the catfish, odorants do not stimulate rapid cyclic nucleotide synthesis ( Restrepo et al.
1993 ) . Toward understanding the nature of transduction
cascade ( s ) present in zebrafish ORNs, we conducted two
experiments. In the first experiment, electro-olfactogram
( EOG ) methods were used to determine if drugs affecting
olfactory transduction in other vertebrates affect the amino
acid and bile salt-evoked responses of zebrafish. The two
drugs tested, ruthenium red and neomycin, are reported
to block IP3-gated channel ( Honda et al. 1995; Restrepo
et al. 1990 ) and PLC ( Slivka and Insel 1988 ) activity,
respectively. To explore the mechanistic basis of any drug
effects, a second series of experiments examined the ac-
0022-3077/98 $5.00 Copyright q 1998 The American Physiological Society
/ 9k26$$mr23 J746-7
02-06-98 08:24:40
neupa
LP-Neurophys
1183
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 14, 2017
Ma, Li and William C. Michel. Drugs affecting phospholipase
C-mediated signal transduction block the olfactory cyclic nucleotide-gated current of adult zebrafish. J. Neurophysiol. 79: 1183–
1192, 1998. Amino acid and bile salt odorants are detected by
zebrafish with relatively independent odorant receptors, but the
transduction cascade(s) subsequently activated by these odorants
remains unknown. Electro-olfactogram recording methods were
used to determine the effects of two drugs, reported to affect phospholipase C (PLC)/inositol tripohsphate (IP3 )-mediated olfactory
transduction in other vertebrate species, on amino acid and bile
salt-evoked responses. At the appropriate concentrations, either an
IP3-gated channel blocker, ruthenium red (0.01–0.1 mM), or a
PLC inhibitor, neomycin (50 mM), reduced amino-acid–evoked
responses to a significantly greater extent than bile salt-evoked
responses. Excised patch recording techniques were used to measure the affects of these drugs on second-messenger–activated currents. Ruthenium red and neomycin are both effective blockers of
the olfactory cyclic nucleotide-gated (CNG) current. Both drugs
blocked the CNG channel in a voltage-dependent and reversible
manner. No IP3-activated currents could be recorded. The differential effects of ruthenium red and neomycin on odor-evoked responses suggest the activation of multiple transduction cascades.
The nonspecific actions of these drugs on odor-activated transduction pathways and our inability to record an IP3-activated current
do not permit the conclusion that zebrafish, like other fish species,
use a PLC/IP3-mediated transduction cascade in the detection of
odorants.
1184
TABLE
L. MA AND W. C. MICHEL
1.
Composition of the intracellular and extracellular solutions
Solution
NaCl
Ringer
Ca2/ free
Low Cl0*
NaCl
High K/
137
137
8
150
—
KCl
2
2
2
—
150
CaCl2
MgCl2
1.8
1
1
—
—
1.6
—
—
HEPES
EGTA
5
—
10
10
5
5
—
—
5
5
Glucose
2
2
Glutamate
10
10
10
—
—
—
—
129
—
—
Solution contents are in millimolar. The osmolarity of all solutions was adjusted to 290–300 milliosmolar, the pH was adjusted to 7.4. Ca2/ free
solution (free Ca2/ Ç10 nM) was used in earlier studies (noted in text), HEPES, N-2-hydroxyethylpiperazine-N*-2-ethanesulfonic acid; EGTA, ethylene
glycol-bis(b-aminoethyl ether)-N,N,N*,N*-tetraacetic acid. * The calculated chloride reversal potential using the low Cl0 solution and Ca2/ free solution
was 060 mV.
tions these drugs on olfactory second-messenger – activated channel activity recorded from excised membrane
patches of dissociated olfactory receptor neurons.
aquaria equipped with under-gravel and charcoal filtration, and fed
a commercially available flake food diet (Tetramin).
Animal care and maintenance
Zebrafish (Danio rerio) were purchased from a commercial supplier (Steve Lambourne), housed in recirculating 10–20 gal
FIG .
1. Block of odor-evoked responses by ruthenium red is dose dependent. A: Mean { SE cysteine (Cys; 100 mM) and taurocholic acid (TCA;
1 mM) evoked responses recorded in the presence of 0.01 (n Å 4), 0.1
(n Å 13), 1 (n Å 4), and 10 (Cys, n Å 2; TCA, n Å 4) mM ruthenium
red. Data for each odor in the ruthenium red background are normalized
to their respective responses in normal artificial fresh water (AFW) recorded
before ruthenium red exposure. TCA-evoked responses are significantly
greater than the Cys-evoked response in the presence of 10 and 100 nM
ruthenium red [2-way analysis of variance (ANOVA), Fisher’s least significant difference (LSD) test, P õ 0.05]. B: ruthenium red (0.1 mM)
reduced the responses to several amino acids to a significantly greater extent
than the responses to the bile salts tested (1-way ANOVA, Fisher’s LSD
test P ° 0.05). In B, each odorant was tested on 3 fish. Average responses
to cysteine and TCA were 0.97 { 0.59 mV and 1.72 { 0.94 mV, respectively
(n Å 13). Arg, L-arginine; Cys, L-cysteine; Glu, L-glutamate; Met,
L-methionine; TCDCA, taurochenodeoxycholic acid; TCA, taurocholic
acid.
/ 9k26$$mr23 J746-7
EOG responses were obtained using methods as previously described (Michel and Lubomudrov 1995). Fish were immobilized
with an intramuscular injection of gallamine triethiodide (Flaxedil;
60 mg/g body wt) and secured to a silicone elastomer (Sylgard)
recording chamber. The olfactory organ immediately was provided
with a continuous flow of artificial freshwater (AFW; see Solutions
for composition), and the gills were irrigated with a flow of Ç5
ml/min of AFW containing the general anesthetic 3-aminobenzoic
acid ethyl ester (MS-222; 20 mg/l in AFW). The level of anesthesia was monitored by observation of reflexive movement of the
gills and eyes and increased if required. Anesthesia was not provided before immobilization to minimize the loss of afferent sensory activity associated with topical application of anesthetic
(Spath and Schweickert 1977). While under general anesthesia, the
small flap of epithelium covering the olfactory organ was removed
surgically to facilitate positioning of the recording electrode. The
recording electrode (3 M KCl in 1% agar, 5–10 mM tip diam) was
positioned between olfactory lamellae near the midline raphe in a
location that maximized the response to 100 mM cysteine (Cys).
An identical electrode positioned on the head served as the differential electrode and a silver/silver chloride ground electrode positioned beneath the fish in the AFW bath completed the electrical
circuit. Responses to olfactory stimuli were amplified (2,000–
5,000 times gain) and filtered (2 kHz) by a low-noise differential
DC amplifier and displayed on an oscilloscope. The amplified
signal also was digitized (100 Hz), stored, and displayed using
Axotape software (Axon Instruments). Each digital record contained Ç8 s of prestimulation and 32 s of poststimulation time.
Preparations were rejected if 100 mM Cys did not elicit a response
ú250 mV.
In situ odorant delivery
Methods for odorant delivery have been described previously
(Michel and Lubomudrov 1995). Briefly, the olfactory epithelium
was bathed by a carrier flow of AFW (or drugs prepared in AFW)
that was supplied by gravity-flow from polyethylene bottles, selected by a six-way valve and regulated to 3 ml/min with a flowmeter. Olfactory stimuli (50 ml) were introduced into the carrier
flow of AFW through a rotary loop injector (Rheodyne). The
bolus of odor reached a peak concentration of Ç84% of the loaded
concentration by 8 s after injection and decayed to the baseline
concentration after ca. 12 s (Michel and Lubomudrov 1995). The
concentrations reported have not been corrected for dilution. To
determine the effects of ruthenium red and neomycin on olfactory
sensitivity, odor-evoked responses measured in the presence of
these drugs were normalized to responses obtained in normal AFW
02-06-98 08:24:40
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 14, 2017
Electro-olfactogram methods
METHODS
PHARMACOLOGY OF ZEBRAFISH CNG CHANNEL
1185
frequently vesiculated, no attempt was made to determine the proportion of patches containing second-messenger–activated currents. The
magnitude of the second-messenger–activated current was measured
from the average of four replicate voltage ramps from 060 to /60
mV (duration Å 30 ms). The intracellular surface of the excised patch
was exposed to cAMP for 3–5 s before initiating the ramp protocol.
Leak currents, measured using an identical ramp protocol in the absence of second messenger, were subtracted from the second-messenger–activated current before calculating the slope conductance. The
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 14, 2017
FIG . 2. Phospholipase C inhibitor, neomycin (50 mM), reduces the responses to amino acid odorants to a significantly greater extent than the
responses to the bile salt odorants (1-way ANOVA, Fisher’s LSD test
P ° 0.05). Amino acid odorants were tested at a concentration of 100 mM,
and bile salt odorants were tested at a concentration of either 1 mM
(TCA) or 10 mM (all others). Data from 5 preparations are plotted as
means { SE. Gln, L-glutamine; GCA, glycocholic acid; His, L-histidine;
LCA, lithocholic acid; Lys, L-lysine.
before drug exposure. Odors tested in a drug background were
diluted with the appropriate background solution.
Dissociation of the olfactory epithelium
Fish were killed by decapitation and the olfactory rosettes were
dissected into ice-cold Ringer (see Table 1). After rinsing, the
rosettes were held in fresh, oxygenated Ringer at 47C until use. To
dissociate ORNs, a rosette was transferred into divalent-free Ringer
containing L-cysteine–activated (1.25 mg/ml) papain (0.25
mg/ml) for 15–30 min. The rosette was rinsed in divalent-free
Ringer, placed onto a concanavalin A-coated glass coverslip containing a single drop of divalent-free Ringer, and dissociated by
teasing with fine insect pins or by trituration through fire-polished
glass pipettes. The cells were allowed to settle and attach to the
coverslip for ¢10 min before use.
Excised patch recording techniques
Coverslips containing dissociated ORNs were placed in a recording chamber situated on the stage of a Nikon MM1 upright
microscope and viewed through a 140 water immersion lens with
an Optizoom image inverter/variable magnifier ( 10.8–2) and 110
eye pieces. Recordings were obtained with borosilicate glass patch
electrodes ( ú1 mm tip diam) filled with the appropriate pipette
solution and positioned on the soma or dendrite of an isolated ORN
under visual control. Signals were amplified with a commercial
patch clamp amplifier (Axopatch 200A), displayed on an analog
oscilloscope, and stored digitally using pClamp software (Axon
Instruments). Recordings were filtered at 1 kHz and digitized at
5 kHz. All reported voltages refer to the intracellular surface of
the membrane relative to the extracellular surface.
Membrane patches were excised from the soma or dendrite of
acutely dissociated ORNs in the inside-out configuration. The use of
low [Ca 2/ ] solutions and a brief air exposure often resulted in successful inside-out patches. The presence of one or many second-messenger–activated channels was confirmed by exposing the intracellular
face of the patch to either adenosine 3 *,5 *-cyclic monophosphate
(cAMP) or IP3 and noting either single- or multichannel activity. Only
excised patches estimated to have five or more second-messenger–
activated channels were used in this study. Since excised patches
/ 9k26$$mr23 J746-7
FIG . 3. Application of adenosine 3 *,5 *-cyclic monophosphate (cAMP) to
the intracellular surface of membrane patches excised from the dendrite or
soma of many zebrafish olfactory receptor neurons (ORNs) activates a dosedependent cyclic nucleotide-gated (CNG) current. A: CNG current in 1 insideout patch from the soma is activated by low micromolar concentrations of
cAMP and saturates at cAMP concentrations Ç10 mM. Patch was stimulated
with 0.4, 0.8, 1.6, 2, 3, 5, 10, 20, 30, and 40 mM cAMP). In this and all
subsequent patches, the CNG-specific current was calculated by subtracting
the ‘‘leak’’ currents (measured in the absence of cAMP; inset shows an example from another patch) from the total currents measured in the presence of
cAMP. B: plot of normalized slope conductances of the data shown in A were
fit with the Hill equation to yield a K1/2 for cAMP activation and Hill coefficient
for this patch of 2.4 mM and 2.1, respectively. Slope conductance for each
trace was normalized to the highest slope conductance. Data was obtained
from inside-out patches in symmetrical NaCl solutions.
02-06-98 08:24:40
neupa
LP-Neurophys
1186
L. MA AND W. C. MICHEL
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 14, 2017
FIG . 4. Pharmacology of the zebrafish olfactory CNG channel is similar to other vertebrate olfactory CNG channels. Two
calcium/calmodulin antagonists, N-(6-aminohexyl)-5-chloro-1-naphthalenesulphonamide (W7; A) and trifluoperazine (TFP;
B), blocked the CNG current in a dose-dependent fashion. A partial block was evident in the presence of low micromolar
concentrations of these drugs. Amiloride (C) and diltiazem (D) also blocked the CNG channel but required higher concentrations. Block by all of these drugs was completely reversible (not shown for W7 and TFP). Data plotted are representative
of data obtained from 3 to 5 different patches for each drug. CNG current was activated with 40 mM cAMP in all patches.
Data was obtained from inside-out patches in symmetrical NaCl solutions.
slope conductance was calculated from a linear regression of the
average of the leak-corrected current over the voltage range of 030
to /30 mV. Plots of the slope conductance versus second-messenger
concentration were fit with the Hill equation to calculate the K1/2 and
Hill coefficient for second messenger binding. Bi-ionic methods were
used to measure the potassium/sodium (K / /Na / ) permeability ratio
(Frings et al. 1992). Second messengers and drugs were applied to
excised patches using a flow microswitching device (Warner Instru-
/ 9k26$$mr23 J746-7
ments, Hamden, CT). Unless otherwise indicated, all data are reported
as means { SE.
Solutions
The compositions of all solutions are listed in Table 1. Stock
solutions of IP3 and cAMP were prepared in distilled water and
stored at 0207C. Stock solutions of ruthenium red, ( / )-cis-diltia-
02-06-98 08:24:40
neupa
LP-Neurophys
PHARMACOLOGY OF ZEBRAFISH CNG CHANNEL
zem, and amiloride were prepared in the NaCl bath solution and
stored at 47C. Stock solutions of N-(6-aminohexyl)-5-chloro-1naphthalenesulphonamide (W7) and trifluoperazine were prepared
in ethanol and stored at 47C. 6-anilino-5,8-quinolinedione
(LY-83583) was obtained from Calbiochem (San Diego, CA),
and a 10 mM stock solution, prepared in dimethyl sulfoxide, was
stored at 47C. All other chemicals were obtained from Sigma
Chemical (St. Louis, MO).
RESULTS
Second-messenger–activated currents in excised patch
recordings
Second-messenger–activated currents were identified by
application of cAMP or IP3 to the intracellular surface of
/ 9k26$$mr23 J746-7
membrane patches excised from the soma or dendrites of
isolated ORNs and recording currents during voltage clamp
protocols. A cAMP-activated current was recorded routinely,
but we failed to observe an IP3-activated current in excised
patches from ú40 zebrafish ORNs, including eight patches
that were sensitive to cAMP (data not shown). Our failure
to identify an IP3-activated current might have been due to
an inappropriate calcium concentration, the use of patches
generally excised from somatic membrane, or the absence
of these channels in zebrafish ORNs. The effects of ruthenium red and neomycin on the CNG current are considered
in the following sections.
General properties of zebrafish olfactory cng channel
Low micromolar concentrations of cAMP activated the
zebrafish olfactory CNG channel ( Fig. 3 ) . The average
K1 / 2 for cAMP was 3.0 { 0.6 mM ( n Å 10, range 0.94 –
6.6 mM ) . A Hill coefficient of 2.8 { 0.5 ( n Å 10 ) indicates
cooperativity of cAMP binding. At concentrations of
cAMP sufficient to elicit a saturating response, the average
slope conductance was 1.2 { 1.2 nS, ( n Å 102, range,
0.09 – 6.3 nS ) . The calculated K / / Na / permeability ratio
of 0.9 { 0.01 ( n Å 5 ) indicates that the zebrafish olfactory
CNG channel is a relatively nonspecific cation channel.
Shifting EC1 from 0 to 060 mV changed the reversal potential of the CNG current by only 1 mV ( n Å 3 patches;
data not shown ) .
Before determining the effects of ruthenium red and
neomycin on the zebrafish olfactory CNG current, the
basic pharmacology of the zebrafish CNG current was
compared with other vertebrate olfactory CNG currents.
Four of five drugs previously reported to block other vertebrate olfactory CNG channels reversibly blocked the
zebrafish olfactory CNG channel ( Fig. 4, Table 2 ) . At
1 – 10 mM, the calcium/ calmodulin antagonists W-7 and
trifluoperazine partially blocked the zebrafish olfactory
2. Effects of drugs reported to block other vertebrate
olfactory CNG conductances on the zebrafish olfactory CNG
conductance
TABLE
Percent Block (Most
Effective Dose)
Drug
Concentrations Tested,
mM
W7
Trifluoperazine
Diltiazim
Amiloride
LY-83583
1 (3), 10 (3), 100 (4)
1 (9), 10 (3), 100 (5)
20 (3), 40 (3), 1000 (5)
50 (3), 100 (3), 1000 (5)
10 (16), 80 (7)
060 mV
73.5
83
38.2
51.2
0.0
{
{
{
{
{
6.6
5.1
16.3
7.1
0.1
/60 mV
96.1
95.2
75.2
95.2
0.0
{
{
{
{
{
4.5
3.8
9.2
2.9
0.1
All drugs were applied with adenosine 3*,5*-cyclic monophosphate
(cAMP) to the intracellular membrane surface. The cAMP concentration
tested was always ¢10 mM. Percent Block is the average current measured
in the presence of a blocker at Vmem Å 060 and /60 mV expressed as a
proportion of the current elicited under control conditions. The percent
block reported is for the highest concentration tested. Extracellular (included
in the recording pipette) diltiazem (40 mM), amiloride (20 mM) and
6-anilino-5,8-quinolinedione (LY-83583; 20 mM) failed to block the cyclic
nucleotide-gated (CNG) current in 3 patches each. Parentheses enclose
number of patches. W7, N-(6-aminohexyl)-5-chloro-1-naphthalenesulphonamide.
02-06-98 08:24:40
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 14, 2017
Effects of ruthenium red and neomycin on odor-evoked
responses
Odor-evoked responses were affected significantly by
both of the drugs tested. Submicromolar ruthenium red significantly reduced the response to the amino acid L-cysteine
without affecting the response to the bile salt taurocholic
acid. Ten and 100 nM ruthenium red reduced the response
to 100 mM Cys by ca. 50 and 80%, respectively (Fig. 1A).
In contrast, responses elicited by 1 mM taurocholic acid
(TCA) were unaffected by 10 nM ruthenium red and reduced
by õ15% by 100 nM ruthenium red. When the ruthenium
red concentration was increased to either 1 or 10 mM, the
Cys-evoked response was eliminated and TCA-evoked responses were attenuated significantly. Cys-evoked responses
recovered slower than TCA-evoked responses regardless of
the ruthenium red concentrations tested (data not shown).
At 10 mM ruthenium red, the response to cysteine did not
recover within 30 min. The responses to three other amino
acids, thought to interact with odorant receptors independent
of the receptor(s) mediating the cysteine response (Michel
and Derbidge 1997), also were attenuated significantly by
100 nM ruthenium red (Fig. 1B).
Odorants stimulate the olfactory PLC activity of several
fish species (Boyle et al. 1987; Har Lo et al. 1993; Huque
and Bruch 1986; Restrepo et al. 1993). If PLC activation is
essential to odor transduction, then drugs blocking PLC activity also should block odor-evoked responses. The PLC
inhibitor, neomycin (50 mM) reduced amino acid-evoked
responses to a significantly [2-way analysis of variance
(ANOVA), Fisher’s least significant difference (LSD) test,
P õ 0.05] greater extent than bile salt-evoked responses
(Fig. 2). The responses elicited by the eight amino acids
tested were reduced by ú80–90%, whereas the responses
to the four bile salts tested were only reduced by õ20% to
Ç50%.
The selective reduction of amino acid-evoked, but not of
bile salt-evoked, responses by submicromolar ruthenium red
and by neomycin is consistent with the presence of a PLC/
IP3 transduction cascade activated by amino acids but not
by bile salts. However, the specific effects of ruthenium red
and neomycin on any potential odorant-activated transduction cascades in zebrafish are unknown. Consequently, in
the second phase of this investigation, we tested the effects
of these drugs on second-messenger–activated currents in
excised patch recordings.
1187
1188
L. MA AND W. C. MICHEL
CNG channel. At 100 mM, either calcium/ calmodulin antagonist effectively blocked the CNG current ( Fig. 4, A
and B ) . Amiloride and diltiazem partially blocked the
zebrafish olfactory CNG channel at millimolar concentrations ( Fig. 4, C and D ) but were ineffective at concentrations õ200 mM ( not shown ) . LY-83583 ( 10 mM ) blocked
the tiger salamander olfactory CNG channel ( LeindersZufall and Zufall 1995 ) but failed to block the zebrafish
olfactory CNG channel when concentrations ranging
from 10 to 80 mM were applied to either the intracellular
or extracellular surface of the membrane ( Table 2 ) . The
pharmacological data indicates that the zebrafish olfactory CNG channel has properties similar to other vertebrate CNG channels.
Effect of ruthenium red on the zebrafish olfactory CNG
current
To determine if ruthenium red affected the CNG current,
we coapplied 40 mM cAMP and one of several concentrations
of ruthenium red to the intracellular surface of the excised
membrane patch (Fig. 5, A, C, and D). The lowest concentra-
/ 9k26$$mr23 J746-7
tion of ruthenium red tested (100 nM) blocked Ç25% of the
outward component of the CNG current. At an intermediate
concentration of ruthenium red (1 mM), nearly 90% of the
outward component was blocked. At the highest ruthenium
red concentrations (10 and 100 mM), the block of the outward
component of the CNG current was generally complete. In
some patches, at the intermediate and higher ruthenium red
concentrations, a negative slope conductance at positive membrane potentials was observed, indicating the ruthenium red
block is voltage dependent. The inward component of the
CNG current also was blocked but only ruthenium red concentrations ¢10–100 mM (Fig. 5D). The block of the inward
component of the CNG current was also voltage dependent.
To determine if extracellular ruthenium red blocked the CNG
current, we included ruthenium red in the recording pipette
solution. Although the block developed slowly, extracellular
ruthenium red also affected the CNG current (Fig. 5B). Initially, only the outward component of the CNG current was
partially blocked by ruthenium red (the block was most notable at positive transmembrane potentials). After 30 min, both
inward and outward components of the CNG current were
blocked by ruthenium red.
02-06-98 08:24:40
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 14, 2017
FIG . 5. Intracellular
ruthenium
red
blocks the CNG channel at submicromolar
concentrations. A: ramp currents recorded in
the presence of cAMP and increasing concentrations of ruthenium red reveal that the
block is dose-dependent and affects both inward and outward components of the CNG
current. B: extracellular ruthenium red (10
mM ruthenium red in the recording pipette)
blocked the CNG channel in excised insideout patches. Block developed slowly; within
5 min the CNG current was blocked at more
positive transmembrane potentials (more
than /30 mV). In this patch, the outward
CNG current was blocked completely by 12
min. Block of the inward component developed more slowly but was essentially complete within 30 min. Different intracellular
ruthenium red concentrations are required to
block inward (C; n Å 4) and outward (D;
n Å 4) components of the CNG current. In
C and D, the inward and outward currents for
each patch were normalized to the cAMPactivated currents measured in the absence
of ruthenium red at 060 and /60 mV, respectively. CNG current was activated with
40 mM cAMP. Data was obtained from inside-out patches in symmetrical NaCl solutions.
PHARMACOLOGY OF ZEBRAFISH CNG CHANNEL
1189
To investigate the recovery from ruthenium red block,
ruthenium red was applied to the intracellular surface of an
excised patch and 40 mM cAMP was applied in ruthenium
red until a stable level of block was achieved. To start the
recovery period, ruthenium red was removed and cAMP was
tested alone. At each concentration of ruthenium red tested,
the magnitude of the CNG current was measured before, in
the presence of, and at several recovery intervals after the
removal of ruthenium red (Fig. 6). The data plotted are
representative of data obtained from at least three patches
at each concentration. In general, recovery from ruthenium
red block was relatively rapid and complete. Recoveries
from 100 nM and 1 mM ruthenium red were complete in
õ5 min. Recoveries from 10 and 100 mM ruthenium red
exposure were incomplete. Nearly 80% of the outward current recovered in õ5 min, but no further recovery was measured even after 30 min.
Effect of neomycin on the zebrafish olfactory CNG current
Application of neomycin to the intracellular surface of the
membrane partially blocked the olfactory CNG current. The
block was dose dependent (Fig. 7A). Neomycin (50–200
mM) blocked Ç50% of the outward current at /60 mV.
The effect of neomycin concentration on the average CNG
currents of three excised patches is illustrated in Fig. 7B.
Extracellular neomycin (included in the pipette solution,
n Å 3 patches) blocked only a small portion ( õ10%) of
the peak outward current (Fig. 7C). The most pronounced
neomycin block was observed when it was applied to both
sides of the membrane patch (Fig. 7D). In three patches
with 50 mM neomycin bathing the intracellular surface and
50–100 mM neomycin bathing the extracellular surface,
/ 9k26$$mr23 J746-7
nearly all of the outward component and Ç50% of the inward component of the CNG current was blocked.
DISCUSSION
Two significant findings are reported in this study. First,
drugs, previously reported to block phospholipase C activity
(neomycin) and IP3-gated channel activity (ruthenium red),
differentially affected odor-evoked responses of the zebrafish olfactory epithelium. Both drugs reduced the responses
evoked by amino acids to a significantly greater extent than
the responses evoked by bile salts. Submicromolar concentrations of ruthenium red (10–1,000 nM) selectively
blocked amino acid evoked responses, 10 mM ruthenium
red blocked all odor-evoked responses. Fifty micromolar
neomycin selectively blocked amino acid-evoked responses.
This neomycin concentration is 20- to 200-fold lower than
generally used to perturb PLC activity. Collectively, these
results are consistent with the interpretation that a PLC/IP3
transduction cascade mediates amino acid-evoked responses.
The second significant finding of this study was that ruthenium red and neomycin both blocked the zebrafish olfactory
CNG current. The later finding renders any conclusion about
the transduction pathway(s) used by amino acid and bile
salt odorants premature. To our knowledge, we are the first
to directly examine the actions of these drugs on the olfactory CNG channel.
Ruthenium red is an organometallic dye that reportedly
affects the function of many calcium-binding proteins
(Amann and Maggi 1991), blocks voltage-gated calcium
channel activity (Hamilton and Lundy 1995) and nociceptor
function (Dray et al. 1990), and associates with negatively
charged phospholipids (Voekler and Smejtek 1996). In ol-
02-06-98 08:24:40
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 14, 2017
FIG . 6. Time course of recovery from ruthenium red application was investigated at 4 ruthenium red concentrations (A–D). In all patches,
the ruthenium red block of the CNG current was
allowed to completely develop (plotted) before
returning the patch to cAMP alone and measuring recovery over time. Recovery was rapid and
complete at lower ruthenium red concentrations
( °1 mM). At higher ruthenium red concentrations ( ¢10 mM), the recovery was rapid and
largely complete, but a portion of the CNG current did not recover. Data from 4 different insideout patches are plotted in A–D. Each patch was
tested with the concentration of ruthenium red
indicated in the figure. CNG current was activated with 40 mM cAMP. Data was obtained
from inside-out patches in symmetrical NaCl solutions.
1190
L. MA AND W. C. MICHEL
faction, ruthenium red originally was used to block norleucine-evoked responses of the channel catfish olfactory epithelium as well as IP3-activated channel activity of enriched
catfish olfactory cilia membranes incorporated into lipid bilayers (Restrepo et al. 1990). Ruthenium red has similar
affects on the olfactory IP3-activated currents/channels of
other vertebrate (Honda et al. 1995; Okada et al. 1994; Restrepo et al. 1990, 1992) and invertebrate (Fadool and Ache
1992) species. The olfactory CNG channel must be added
to the list of cellular substrates affected by ruthenium red.
The block of the zebrafish olfactory CNG current by ruthenium red was dose dependent and effective at submicromolar
concentrations. One hundred-fold lower concentrations of
ruthenium red were required block the outward component
of the CNG current. Extracellular ruthenium red also blocked
the outward component of the CNG current to a greater
/ 9k26$$mr23 J746-7
extent than the inward component. Collectively, these findings suggest that the primary site of ruthenium red action is
on the intracellular surface. Ruthenium red association with
negatively charged phospholipid residues (Voekler and
Smejtek 1996) near the CNG channel presumably would
reduce the local monovalent cation concentration, resulting
in a block similar to that produced during intracellular acidification (Frings et al. 1992). If the adsorption of ruthenium
red to these membrane phospholipids is strong, ruthenium
red levels might be expected to persist even after ruthenium
red application is stopped. The slow and incomplete recovery
of both odor responsiveness and CNG currents after exposure to the higher ruthenium red concentrations might reflect
its adsorption to phospholipids.
Neomycin is an aminoglycoside antibiotic that reportedly
affects many cellular substrates. Neomycin blocks phospho-
02-06-98 08:24:40
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 14, 2017
FIG . 7. Phospholipase C (PLC) inhibitor neomycin blocked the CNG channel. A: intracellular neomycin blocked the
CNG current in a dose-dependent manner. Membrane patches became very unstable when exposed to concentrations of
neomycin ú200 mM. B: average { SE block of the outward component of the CNG current, measured in 3 excised patches,
is plotted as a function of intracellular neomycin concentration. C: extracellular neomycin (100 mM) weakly blocks the
outward component of the CNG current (neomycin included in the recording pipette). D: neomycin nearly completely
blocked the outward component of the CNG current when present on both sides of the membrane. In all patches, the CNG
current was activated with 40 mM cAMP. Data was obtained from inside-out patches in symmetrical NaCl solutions.
PHARMACOLOGY OF ZEBRAFISH CNG CHANNEL
/ 9k26$$mr23 J746-7
catfish (Miyamoto et al. 1992). The issue of whether multiple odor-activated transduction cascades are present in zebrafish has yet to be adequately resolved. If amino acid and
bile salt sensitivity is segregated among the ciliated and
microvillar receptor cells in zebrafish like pheromone and
amino acid sensitivity appears to be in goldfish (Zippel et
al. 1997), then differences in CNG channel microenvironment in these cell type might differentially affect the
blocking efficacy of these drugs. On the surface, however,
the selective reduction of amino acid-evoked responses, but
not of bile salt-evoked responses, by both ruthenium red (at
lower concentrations) and neomycin suggests the presence
of multiple transduction cascades. Because concentrations
of either drug significantly attenuating amino acid-evoked
responses had very little affect on the CNG current, it is
possible that an as-yet unidentified component involved in
the transduction of amino acid odorants by zebrafish is affected primarily by these drugs. For example, at low ruthenium red concentrations the block of amino acid-evoked
responses might be due to drug interactions with an anionic
binding site present on amino acid receptors but not on bile
salt receptors. At higher concentrations, ruthenium red might
eliminate all odor-evoked responses through its action on the
CNG current. Alternately, the block of amino acid-evoked
responses observed at lower drug concentrations might be
due to an action of these drugs on an independent transduction cascade, perhaps mediated by PLC and an IP3-gated
current. Our failure to identify an IP3-activated current may
have been due to any number of factors, including recording
from the soma rather than cilia or an inappropriate calcium
concentration. Because bile salt-evoked EOG responses and
CNG currents are both blocked 10 mM ruthenium red but
only slightly affected by 50 mM neomycin, it is perhaps
likely that bile salts activate a CNG transduction cascade.
We thank D. Derbidge, M. Pearson, and A. Reed for assistance with
electro-olfactogram recordings. We thank Dr. Mary Lucero for reviewing
an earlier version of the manuscript.
This work was supported by National Institute of Deafness and Other
Communications Disorders Grant DC-01418.
Address for reprint requests: W. C. Michel, Dept. of Physiology, University of Utah School of Medicine, 410 Chipeta Way, Salt Lake City, UT
84108.
Received 9 September 1997; accepted in final form 20 November 1997.
REFERENCES
ACHE, B. W. AND ZHAINAZAROV, A. Dual second-messenger pathways in
olfactory transduction. Curr. Opin. Neurobiol. 5: 461–466, 1995.
ALGRANATI, F. D. AND PERLMUTTER, A. Attraction of zebrafish, Brachydanio rerio, to isolated and partially purified chromatographic fractions.
Environ. Biol. Fish 6: 31–38, 1981.
AMANN, R. AND MAGGI, C. A. Ruthenium red as a capsaicin antagonist.
Life Sci. 49: 849–856, 1991.
BAIER, H. AND KORSCHING, S. Olfactory glomeruli in the zebrafish form
an invariant pattern and are identifiable across animals. J. Neurosci. 14:
219–230, 1994.
BAIER, H., ROTTER, S., AND KORSCHING, S. Connectional topography in the
zebrafish olfactory system: random positions but regular spacing of sensory neurons projecting to an individual glomerulus. Proc. Natl. Acad.
Sci. USA 91: 11646–11650, 1994.
BARTH, A. L., JUSTICE, N. J., AND NGAI, J. Asynchronous onset of odorant
receptor expression in the developing zebrafish olfactory system. Neuron
16: 23–34, 1996.
02-06-98 08:24:40
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 14, 2017
lipase C activity by binding to the enzyme’s substrate, phosphoinositol bisphosphate (PIP2 ) (Slivka and Insel 1988).
A variety of ion channel types, including skeletal muscle
L-type Ca 2/ channels, Ca 2/ -activated K / channels, and
mechanosensitive channels are blocked by neomycin and by
other aminoglycoside antibiotics (Haws et al. 1996; Winegar
et al. 1996). In isolated rat ORNs, 200 mM neomycin
blocked the [Ca 2/ ]i increase elicited by an IP3-activating
odor mixture, presumably by inhibiting PLC activity (Tareilus et al. 1995), but also appears to have partially blocked
the response to the cAMP-activating odor mixture. Our finding that neomycin blocks olfactory CNG current suggests
that the partial block of the response to the cAMP-activating
odor mixture previously noted may have been a result of a
direct action of neomycin upon the CNG channel. Although
higher concentrations of neomycin frequently are used to
perturb PLC function, we found that excised patches became
unstable at concentrations of neomycin ú200 mM. Because
neomycin only partially blocks the CNG current and is
known to have actions on other potential substrates involved
in olfactory transduction, particularly PLC activity, we did
not further characterize its actions on the CNG current.
Our findings that both ruthenium red and neomycin block
the olfactory CNG current are not likely to be due to properties unique to the zebrafish CNG channel. Biophysical and
pharmacological properties of the olfactory CNG channel of
zebrafish are similar, in most respects, to the olfactory CNG
channels of other vertebrates. In a low calcium environment
(ca. 10 nM), the K1 / 2 of 3.0 mM for cAMP activation of
zebrafish CNG channels is similar to K1 / 2 values reported
for the toad [3.4 mM (Nakamura and Gold 1987)], channel
catfish [2.5 mM (Bruch and Teeter 1990; Goulding et al.
1992)], frog [3.4–4 mM (Frings et al. 1992)], rat [2.5 mM
(Frings et al. 1992)], and the tiger salamander [20 mM
(Zufall et al. 1991)]. The Hill coefficient of 2.8 measured
for the zebrafish CNG channel is slightly larger than values
reported for other olfactory CNG channels (range, 1.4–2.1),
but is consistent with the general idea of cooperative cyclic
nucleotide binding to a heterotetrameric channel (Zagotta
and Siegelbaum 1996). The K / /Na / permeability ratio of
0.9 falls within the range of previously reported values
(Frings et al. 1992). Pharmacological properties of the zebrafish olfactory CNG channel are also similar to other olfactory CNG channels. Four of five drugs previously used to
block vertebrate olfactory CNG channels were effective
blockers of the zebrafish olfactory CNG channel. Low micromolar concentrations of the calcium/calmodulin antagonists W-7 and trifluoperazine blocked the zebrafish (current
report) and bullfrog (Kleene 1994) olfactory CNG channels.
As reported for other olfactory CNG channels (Zagotta
and Siegelbaum 1996), relatively high concentrations of
( / )-cis-diltiazem (1 mM) were required to partially block
the zebrafish olfactory CNG channels. Although shown to
be a relatively potent blocker of the olfactory CNG current
in the tiger salamander (Leinders-Zufall and Zufall 1995),
LY-83583 (10–80 mM) did not block the zebrafish olfactory
CNG channel from either the intracellular or extracellular
surface of the membrane.
Multiple odor-activated transduction cascades have been
confirmed in several species (Ache and Zhainazarov 1995;
Dionne and Dubin 1994) and are indicated in the channel
1191
1192
L. MA AND W. C. MICHEL
/ 9k26$$mr23 J746-7
MICHEL, W. C. AND LUBOMUDROV, L. M. Specificity and sensitivity of the
olfactory organ of the zebrafish, Danio rerio. J. Comp. Physiol. [A] 177:
191–199, 1995.
MIYAMOTO, T., RESTREPO, D., CRAGOE, E. J., JR ., AND TEETER, J. H. IP3and cAMP-induced responses in isolated olfactory receptor neurons from
the channel catfish. J. Membr. Biol. 127: 173–183, 1992.
NAKAMURA, T. AND GOLD, G. H. A cyclic nucleotide-gated conductance in
olfactory receptor cilia. Nature 325: 442–444, 1987.
OK ADA, Y., TEETER, J. H., AND RESTREPO, D. Inositol 1,4,5-trisphosphategated conductance in isolated rat olfactory neurons. J. Neurophysiol. 71:
595–602, 1994.
RESTREPO, D., BOEKHOFF, I., AND BREER, H. Rapid kinetic measurements
of second messenger formation in olfactory cilia from channel catfish.
Am. J. Physiol. 264 (Cell Physiol. 33): C906–C911, 1993.
RESTREPO, D., MIYAMOTO, T., BRYANT, B. P., AND TEETER, J. H. Odor
stimuli trigger influx of calcium into olfactory neurons of the channel
catfish. Science 249: 1166–1168, 1990.
RESTREPO, D., TEETER, J. H., HONDA, E., BOYLE, A. G., MARECEK, J. F.,
PRESTWICH, G. D., AND KALINOSKI, D.L. Evidence for an InsP3-gated
channel protein in isolated rat olfactory cilia. Am. J. Physiol. 263 (Cell
Physiol. 32): C667–C673, 1992.
RESTREPO, D., TEETER, J. H., AND SCHILD, D. Second messenger signaling
in olfactory transduction. J. Neurobiol. 30: 37–48, 1996.
SLIVK A, S. R. AND INSEL, P. A. Phorbol ester and neomycin dissociate
bradykinin receptor-mediated arachidonic acid release and polyphosphoinositide hydrolysis in Madin-Darby canine kidney cells. J. Biol. Chem.
263: 14640–14647, 1988.
SPATH, M. AND SCHWEICKERT, W. The effect of metacaine (MS-222) on
the activity of the efferent and afferent nerves in the teleost lateral-line
system. Naunyn-Schmiedeberg’s Arch. Pharmacol. 297: 9–16, 1977.
STEELE, C. W., OWENS, D. W., AND SCARFE, A. D. Attraction of zebrafish
Brachydanio-rerio to alanine and its suppression by copper. J. Fish Biol.
36: 341–352, 1990.
STEELE, C. W., SCARFE, A. D., AND OWENS, D. W. Effects of group size
on the responsiveness of zebrafish Brachydanio-rerio Hamilton Buchanan
to alanine a chemical attractant. J. Fish Biol. 38: 553–564, 1991.
TAREILUS, E., NOÉ, J., AND BREER, H. Calcium signals in olfactory neurons.
Biochim. Biophys. Acta Mol. Cell Res. 1269: 129–138, 1995.
VAN DEN HURK, R. AND LAMBERT, J.G.D. Ovarian steroid glucuronides
function as sex pheromones for male zebrafish, Brachydanio rerio. Can.
J. Zool. 61: 2381–2387, 1983.
VOEKLER, D. AND SMEJTEK, P. Adsorption of ruthenium red to phosholipid
membranes. Biophys. J. 70: 818–830, 1996.
VOGT, R. G., LINDSAY, S. M., BYRD, C. A., AND SUN, M. Spatial patterns
of olfactory neurons expressing specific odor receptor genes in 48-hourold embryos of zebrafish Danio rerio. J. Exp. Biol. 200: 433–443, 1997.
WETH, F., NADLER, W., AND KORSCHING, S. Nested expression domains
for odorant receptors in zebrafish olfactory epithelium. Proc. Natl. Acad.
Sci. USA 93: 13321–13326, 1996.
WINEGAR, B. D., HAWS, C. M., AND LANSMAN, J. B. Subconductance block
of single mechanosensitive ion channels in skeletal muscle fibers by
aminoglycoside antibiotics. J. Gen. Physiol. 107: 433–443, 1996.
ZAGOTTA, W. N. AND SIEGELBAUM, S. A. Structure and function of cyclic
nucleotide-gated channels. Annu. Rev. Neurosci. 19: 235–263, 1996.
ZIPPEL, H. P., SORENSEN, P. W., AND HANSEN, A. High correlation between
microvillous olfactory receptor cell abundance and sensitivity to pheromones in olfactory nerve-sectioned goldfish. J. Comp. Physiol. [A] 180:
39–52, 1997.
ZUFALL, F. AND FIRESTEIN, S. Divalent cations block the cyclic nucleotidegated channel of olfactory receptor neurons. J. Neurophysiol. 69: 1758–
1768, 1993.
ZUFALL, F., FIRESTEIN, S., AND SHEPHERD, G. M. Analysis of single cyclic
nucleotide-gated channels in olfactory receptor cells. J. Neurosci. 11:
3573–3580, 1991.
ZUFALL, F., FIRESTEIN, S., AND SHEPHERD, G. M. Cyclic nucleotide-gated
ion channels and sensory transduction in olfactory receptor neurons.
Annu. Rev. Biophys. Biomol. Struct. 23: 577–607, 1994.
02-06-98 08:24:40
neupa
LP-Neurophys
Downloaded from http://jn.physiology.org/ by 10.220.33.6 on June 14, 2017
BLOOM, H. D. AND PERLMUTTER, A. A sexual aggregating pheromone system in the zebrafish, Brachydanio rerio (Hamilton-Buchanan). J. Exp.
Zool. 199: 215–226, 1977.
BOYLE, A. G., PARK, Y. S., HUQUE, T., AND BRUCH, R. C. Properties of
phospholipase C in isolated olfactory cilia from the channel catfish (Ictalurus punctatus). Comp. Biochem. Physiol. B Biochem. Mol. Biol. 88:
767–775, 1987.
BRUCH, R. C. Phosphoinositide second messengers in olfaction. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 113: 451–459, 1996.
BRUCH, R. C. AND TEETER, J. H. Cyclic AMP links amino acid chemoreceptors to ion channels in olfactory cilia. Chem. Senses 15: 419–430, 1990.
BUCK, L. B. Information coding in the vertebrate olfactory system. Annu.
Rev. Neurosci. 19: 517–544, 1996.
BYRD, C. A. AND BRUNJES, P. C. Organization of the olfactory system in
the adult zebrafish: histological, immunohistochemical, and quantitative
analysis. J. Comp. Neurol. 358: 247–259, 1995.
BYRD, C. A., JONES, J. T., QUATTRO, J. M., ROGERS, M. E., BRUNJES, P. C.,
AND VOGT, R. G. Ontogeny of odorant receptor gene expression in zebrafish, Danio rerio. J. Neurobiol. 29: 445–458, 1996.
DILL, L. M. The escape response of the zebra danio ( Brachydanio rerio).
II. The effect of experience. Anim. Behav. 22: 723–730, 1974.
DIONNE, V. E. AND DUBIN, A. E. Transduction diversity in olfaction. J.
Exp. Biol. 194: 1–21, 1994.
DRAY, A., FORBES, C. A., AND BURGESS, G. M. Ruthenium red blocks the
capsaicin-induced increase in intracellular calcium and activation of
membrane currents in sensory neurones as well as the activation of peripheral nociceptors in vitro. Neurosci. Lett. 110: 52–59, 1990.
FADOOL, D. A. AND ACHE, B. W. Plasma membrane inositol 1,4,5-trisphosphate-activated channels mediate signal transduction in lobster olfactory
receptor neurons. Neuron 9: 907–918, 1992.
FRIEDRICH, R. W. AND KORSCHING, S. I. Combinatorial and chemotopic
odorant coding in the zebrafish olfactory bulb visualized by optical imaging. Neuron 18: 737–752, 1997.
FRINGS, S., LYNCH, J. W., AND LINDEMANN, B. Properties of cyclic nucleotide-gated channels mediating olfactory transduction. Activation, selectivity, and blockage. J. Gen. Physiol. 100: 45–67, 1992.
GOULDING, E. H., NGAI, J., KRAMER, R. H., COLICOS, S., AXEL, R.,
SIEGELBAUM, S. A., AND CHESS, A. Molecular cloning and single-channel
properties of the cyclic nucleotide-gated channel from catfish olfactory
neurons. Neuron 8: 45–58, 1992.
HAMILTON, M. G. AND LUNDY, P. M. Effect of ruthenium red on voltagesensitive Ca2/ channels. J. Pharmacol. Exp. Ther. 273: 940–947, 1995.
HANSEN, A. AND ZEISKE, E. Development of the olfactory organ in the
zebrafish, Brachydanio rerio. J. Comp. Neurol. 333: 289–300, 1993.
HAR LO, Y., BRADLEY, T. M., AND RHOADS, D. E. Stimulation of Ca 2/ regulated olfactory phospholipase C by amino acids. Biochemistry 32:
12358–12362, 1993.
HARA, T. J. Fish Chemoreception. Amsterdam: Elsevier, 1992.
HAWS, C. M., WINEGAR, B. D., AND LANSMAN, J. B. Block of single L-type
Ca 2/ channels in skeletal muscle fibers by aminoglycoside antibiotics.
J. Gen. Physiol. 107: 421–432, 1996.
HONDA, E., TEETER, J. H., AND RESTREPO, D. InsP3-gated ion channels in
rat olfactory cilia membrane. Brain Res. 703: 79–85, 1995.
HUQUE, T. AND BRUCH, R. C. Odorant-and guanine nucleotide-stimulated
phosphoinositide turnover in olfactory cilia. Biochem. Biophys. Res. Commun. 137: 36–42, 1986.
KLEENE, S. J. Inhibition of olfactory cyclic nucleotide-activated current by
calmodulin antagonists. Br. J. Pharmacol. 111: 469–472, 1994.
LEINDERS-ZUFALL, T. AND ZUFALL, F. Block of cyclic nucleotide-gated
channels in salamander olfactory receptor neurons by the guanylyl cyclase
inhibitor LY83583. J. Neurophysiol. 74: 2759–2762, 1995.
LO, Y. H., BELLIS, S. L., CHENG, L.-J., PANG, J., BRADLEY, T. M., AND
RHOADS, D. E. Signal transduction for taurocholic acid in the olfactory
system of Atlantic salmon. Chem. Senses 19: 371–380, 1994.
MICHEL, W. C. AND DERBIDGE, D. S. Functional characterization of amino
acid and bile salt receptors in the olfactory system of the zebrafish, Danio
rerio. Brain Res. 764: 179–187, 1997.