262
BrainResearch, 583 (1992) 262-270 © 1992Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00
BRES 17898
Biochemical characterization of independent olfactory receptor sites for 5'-AMP and taurine in the spiny lobster Kirby S. Olson a
a,
Henry G. Trapido-Rosenthal
Departmentof Biology, Georgia State University, Atlanta, GA (USA) and (USA)
b
b
and Charles D. Derby
a
The Whitney Laboratory, University of Florida, St. Augustine, FL
(Accepted 18 February 1992)
Key words: Olfaction; Chemoreception; Transduction; Receptor binding; Lobster; Crustacean; AMP; Taurine
To understand the initial events in chemosensory transduction (i.e. binding of odorants to olfactory receptor sites), we have utilized a radioligand-receptor binding assay on a tissue preparation that is enriched in dendritic membrane from olfactory receptor cells in the spiny lobster Panulirus argus. Radioligands used were tritiated adenosine 5'-monophosphate (AMP) and taurine, which are two of the most excitatory compounds for spiny lobsters. Our results indicate the existence of two independent types of binding sites-a taurine binding site and an AMP binding site. For both the taurine and AMP binding sites, odorant binding is rapid, reversible, and saturable. K D values for the taurine and AMP binding sites are 2.3 and 2.0 J.LM, respectively, and B max values are 159 and 3.2 fmolj J.Lg protein, respectively. The fact that the specificity, affinity, and independence of these two binding sites as defined in these biochemical studies are in agreement with those from electrophysiological studies suggests that these binding sites are olfactory receptor sites involved in sensory transduction. .
INTRODUCfION
The olfactory system of the Florida spiny lobster Panulirus argus has proven to be a useful experimental model system for understanding mechanisms of olfactory function. Peripheral and central olfactory elements have been studied in this species using various techniques, including extracellular and intracellular electrophysiology, histology and anatomical tract tracing, electron microscopy, immunocytochemistry, biochemistry, and behavioral conditioning. Studies using these techniques have led to an understanding of important olfactory phenomena including the effect of peri receptor events, such as odorant degradation and uptake, on odorant detection15.16.39.55, and involvement of second messengers and ion channels in sensory transduction in olfactory receptor cel!s29,45,46. Prior work has also investigated the mechanisms of neural coding of the quality and intensity of single odorant compounds and odorant mixtures by the peripheral and central olfactory systems20,25,28,35-37, mechanisms
of mixture interactions and their effects on neural coding of and behavioral responses to odorant mixtures 2,3.5.17,18.25- 28,32,46, and functional organization of the chemosensory processing centers in the CNS 1,49,52. An important event in chemosensory transduction which has not been directly studied in the spiny lobster is the initial event in transduction-the binding of odorant molecules to the membrane-bound receptor molecules on the olfactory receptor cells. Our knowledge of the binding properties of these receptor sites in spiny lobsters comes solely from indirect studies using electrophysiological and behavioral techniques. These indirect studies indicate that the olfactory system of P. argus is most sensitive to select amino acids, amines, adenine nucleotides, quaternary ammonium compounds, and ammonium, with response thresholds to single compounds typically being in the micromolar to nanomolar range 4,12,18,22.23,26,32. Receptor cells with narrowly tuned excitatory response spectra have been found which are best excited by either AMP, ATP, betaine, cysteine, glutamate, ammonium, or tau-
Correspondence: K. Olson, Department of Biology, Georgia State University, P.O. Box 4010, Atlanta, GA 30302-4010, USA. Fax: (I) (404) 651-2509.
263 rine 14,20,26,34 . The high degree of narrow tuning of the cells is revealed by the difference in the stimulus concentrations that produce a half-maximal response for the two most excitatory compounds for a given cell, which is often several log units apart. Cluster analysis of the cells based on their excitatory response spectra to single compounds demonstrates several distinct classes of cells20 ,26. Some responses are inhibitory or suppressive (i.e. result in a reduction in the spontaneous rate of spiking activity or a hyperpolarization from the resting membrane potential, respectively). Each cell can produce both excitatory or inhibitory responses, though not necessarily the same response to the same compounds as other cells in the population 19,21,26,44,46. Thus, these indirect studies of odorant-receptor binding based on responses to single compounds suggest that the spiny lobster's olfactory system contains numerous types of relatively independent receptor sites that mediate excitation or inhibition, and that these receptor sites are heterogeneously distributed across the olfactory cells of the system. The use of chemical mixtures in studies of the spiking responses of olfactory receptor cells of the spiny lobster reveals a potential for greater complexity in the association of odorant molecules and receptor sites than indicated by studies using single compounds53. Mixture interactions, especially mixture suppression, are common in this system2 ,3,13.21,25- 27. For example, the spiking responses of olfactory receptor cells to certain binary mixtures are often not predictable from the responses to the individual components of the mixtures26 ,27,38,41. Determining the extent to which inhibition of ligand-receptor binding contributes to mixture suppression requires direct measurements of the interaction of odorant molecules with receptor sites. Receptor binding assays using radioligands have been developed for several chemoreceptor systems, including chemoreception by unicellular organisms", larval gastropods58, hydra", olfaction in fishlO,1l ,3 1,48,51 and mammals30,50 , and gustation in fish8,42 . No such radioligand-receptor binding techniques have been applied to crustacean chemoreception, and only rarely to other arthropods 60. In this paper, we describe a radioligand-receptor binding assay using a tissue fraction that is enriched in dendritic membrane from olfactory receptor cells in the spiny lobster. We use this assay to characterize two classes of binding sites: one for adenosine 5/-monophosphate (AMP) and one for taurine. AMP and taurine were selected as stimuli because they are strong excitants of receptor cell activity 20,21,24,26,34 and behav-
ior 18 ,32,33, and electrophysiological evidence suggests that AMP and taurine probably operate through two
independent receptor sites20 ,26. The biochemical data presented in this paper demonstrate the existence of independent binding sites for AMP and for taurine, both of which have rapid, reversible, and saturable binding characteristics. Correlation of these biochemical results with existing behavioral and electrophysiological data suggests that taurine and AMP are binding to separate olfactory receptor sites.
MATERIALS AND METHODS Materials. [3H1Adenosine 5'-monophosphate (specific activity = 29.0 Ciyrnmol), [3Hltaurine (specific activity = 25.6 Ciymrnol), and [3Hladenosine (specific activity = 31.0 Ciymmol) were purchased from DuPont-New England Nuclear. All other chemicals were from Sigma Chemical Co. ANa +-free, sucrose-containing Tris buffer was prepared with the following composition: 50 mM KCl, 10 mM Tris base, 320 mM sucrose, 12.9 mM CaCl z·2H zO, 23.1 mM MgClz·6 HzO, and 25.6 mM MgS04·7 HzO, with pH adjusted to 7,8 using HC!. The pH and ion concentrations used in this buffer approximate those found in seawater to which receptor sites are normally exposed, and are identical to those of the artificial seawater used in electrophysiological studies ZO. Animals. Spiny lobsters i Panulirus argus) were collected from commercial fish houses in the Florida Keys. In experiments using frozen tissue, lateral filaments of the antennules (olfactory organ) were removed from live animals, placed in tubes filled with Tris buffer, and frozen in dry ice until they were transported to a - 80°C freezer, where they were stored until used. In experiments using fresh antennules, the lobsters were transported to the laboratory where they were held in aquaria with running seawater, and were fed a mixed diet of shrimp, squid, and fish. Lateral antennular filaments were excised from these animals as needed and used immediately. Tissue preparation. Collection of tissue enriched in dendritic membrane from receptor cells is relatively simple, due to the organization of receptor cells within antennules. The paired lateral filaments of the antennules of spiny lobsters contain ca. 600,000 chemoreceptor cells that innervate the aesthetasc sensilla'". Dendrites of these receptor cells are the only neuronal tissue within the sensilla, since the somata and axons of the receptor cells lie in the lumen of the antennule below the sensilla. Distal processes of supporting cells (glia) are also located in the sensilla40. Thus, by collecting only the aesthetasc sensilla, it is possible to harvest a tissue fraction highly enriched in olfactory receptor sites but free of non-dendritic regions of receptor cells, synapses, central nervous tissue, muscle, and most other tissue. The aesthetasc sensilla were manually removed from the antennule by scraping with a scalpel, and subsequently maintained at 4°C in Tris buffer. To extract membrane from the sensilla, the tissue was homogenized manually using a glass-glass homogenizer, and then sonicated (five times for 5 s each on setting 35 of a Fisher Sonic Dismembrator Model 300 with a 4 mm titanium tip). The tissue was centrifuged at 6,000 x g for 10 min, resulting in a pellet (PI) containing sensillar cuticle and a supernatant (S'l) containing the dendritic membrane. The Sl was removed, and the PI was resuspended in Tris buffer and centrifuged again at 6,000x g for 10 min. The resulting supernatant was removed and combined with the previous Sl. The Sl was diluted with 3 ml of Tris buffer and centrifuged at 150,000X g for 30 min. The supernatant was then removed, the pellet was resuspended in 3 ml of fresh Tris buffer, and then recentrifuged at 150,000X g for an additional 30 min to provide an additional washing of the pellet. The resulting P2 pellet was then resuspended in Tris buffer to form the P2 fraction, a suspension of tissue enriched in dendritic membrane from olfactory receptor cells, which was used in subsequent binding assays. Binding assay. Binding activity of the olfactory tissue was assayed by incubating aliquots of the P2 fraction and radiolabel in Tris buffer at 4°C. The protein concentration of the tissue in each incubation tube
264 was approximately 25 ~g/tube in a final incubation volume of 50 ~I (equivalent to approximately 3 antennulesy'tube). The concentration of radiolabel was 1 ~M, unless specified otherwise . Incubations were 60 min in duration, except where otherwise indicated. (This time was determined from association kinetics experiments to be sufficient for saturation of the binding sites with labeled ligand.) Following incubation, bound radioligand was separated from free radioligand by rapid filtration under vacuum through 0.45 ~m pore-size cellulose acetate filters (Type HAWP, Millipore) pre-soaked in 0.3% polyethylenimine in buffer, using a Hoefer vacuum filtration manifold. Filters were washed with two 5-ml vols. of cold Tris buffer until dry, and were then dissolved by placing them in scintillation vials containing 1 ml of ethylene glycol monomethylether for 30 min prior to addition of scintillation fluid. Radioactivity was measured in Ecolite( +) (lCN Biomedical Inc.) using a Beckman model LS5801 liquid scintillation counter, with a counting efficiency of 43-50%. Total binding and non-specific binding were characterized by incubation of tissue and radiolabeled ligand in the absence or presence of 1 mM concentration of the unlabeled ligand. Specific binding was defined as the difference between total binding and non-specific binding. Binding is expressed in terms of fmol of labeled compound bound per ~g of membrane protein . Protein concentrations were determined according to Bradford ? using bovine serum albumin as the standard. The binding data were statistically analyzed using Graphpad Software (INPLOT). Association and dissociation experiments. Association experiments were performed to determine the time of incubation necessary to attain steady-state binding. P2 fractions were incubated with 1 ~M radioligand for 0, 5, 10, 15, 30, 45, 60 and 75 min, with (non-specific binding) or without (total binding) 1 mM unlabeled ligand. Dissociation experiments were performed to examine reversibility of binding. P2 fractions were incubated with 1 ~M radioligand for 60 min, at which time 1 mM unlabeled ligand was added. Incubation then continued for 0, 1, 2, 3, 4, 5, 10, and 30 min, at which time the samples were filtered. Saturation experiments. Binding was measured for the P2 fraction incubated in increasing concentrations of ligand. Ligand concentrations from 10 nM to 'Il.l mM were tested by diluting 1 ~M radioligand with sufficient concentrations of either buffer (for concentrations less than 1 ~M) or unlabeled ligand (for concentrations greater than 1 ~M). The amount of binding at each ligand concentration was calculated by adjusting the specific activity of the radioligand to account for the added unlabeled ligand. Inhibition experiments. To determine if AMP and taurine bind to independent binding sites, the P2 fraction was incubated in 1 ~M radioligand alone or in 1 ~M radioligand plus increasing concentrations (10 nM to 1 mM) of unlabeled AMP or taurine. Specific binding in the absence of unlabeled ligand was defined as 100%, and binding in the presence of unlabeled ligand was expressed as a percentage of this value. When the unlabeled ligand was the same compound as the labeled ligand, the binding at 1 mM of unlabeled ligand was defined as the non-specific binding. These experiments were designed to distinguish independence of the two binding sites and to .determine the IC so values (the concentration of unlabelled compound that inhibits 50% of the maximum specific binding), which are used to compare the degree of inhibition by different compounds . Adenosine production assay. For determinations of dephosphorylation of AMP to adenosine, frozen tissue was prepared as a P2 fraction. Aliquots of this fraction equivalent to 3 antennules each were incubated in 500 1£1 of buffer with 1 ~M [3H)AMP, 1 mM adenosine, and various concentrations of unlabeled AMP, a,l3-methyleneadenosine 5'-diphosphate (AMPCP), or /:I-glycerophosphate (I3GP). After 1 h of incubation at 4°C, samples of incubation media were removed, and unlabeled AMP, ADP, and ATP were all added to a concentration of 1 mM. The samples were then stored at - 80°C until they were analyzed for [3Hladenosine by anion-exchange HPLC and liquid scintillation spectrophotometry", Specific enzymatic activity was determined by subtracting from the above values (which represent total enzymatic activity) the values for non-specific dephosphorylation of [3H)AMP that occurred in incubations containing 1 mM unlabeled AMP.
RESULTS
Preliminary experiments Significant specific binding for [3H}AMP and for [3Hltaurine was found in the P2 tissue fraction. Specific binding was 70-80% of total binding for taurine, and 50-70% of total binding for AMP. However, no specific binding was found in the S2 and PI tissue fractions from fresh antennules (n = 6 from 2 experiments run in triplicate; P> 0.05, r-test). Both the S2 and P2 were also tested for their ability to convert AMP to adenosine via 5'-ectonucleotidase or alkaline phosphatase. The S2 fraction converts 8 pmolyp,g protein while the P2 converts 16 pmol /u g protein; so, unlike specific binding, enzyme activity is not restricted to the P2. Binding was similar for P2 fraction prepared from fresh antennules and for P2 fraction prepared from antennules previously frozen at -80D e in Tris buffer. Values for specific binding of [3H]taurine in these two conditions were 7.56 ± 1.07 vs. 7.00 ± 1.36 fmol /g.g protein, respectively (mean ± S.E.M., n = 9 from 3 experiments run in triplicate; P> 0.05, r-test), Thus, for convenience, frozen antennules were used to prepare P2 in all subsequent experiments.
Association and dissociation experiments Association curves for specific binding of [3H]taurine and [3H]AMP are shown in Figs. 1 and 2, respectively. Dissociation curves are shown in Figs. 3 and 4. Specific binding values (in fmol/p,g protein) and curves fitted to a one-site binding model are shown for each analysis. The results show that specific binding of [3H]taurine and [3H]AMP is rapid and reversible, which is characteristic of the binding of odorants to receptor sites.
[3H]-TAURINE ASSOCIATION CURVE 0 Z
10
::J
a
CD
>-
Z
..J ..J
H
u
I-
-c H
LL.
0
0::
H
CL
U W
~
CL
VI
---
H
a
Z
::J
..:
6
"'-
W
0::
8
w
VI
W
..J
::E LL. ~
lt
t
10
20
30 40 50 TIME ( MI N )
60
70
80
Fig. 1. Association of [3Hltaurine to P2 olfactory tissue fraction. Values are means ± S.E.M. for one experiment run in triplicate . Solid curve represents the one-site model that most closely fits the data.
265 [3H]-AMP ASSOCIATION CURVE 0
z
[3H]-AMP DISSOCIATION CURVE
100
3.0
0
z
:::>
:::>
0
CD
0
en z
75
>-
....J ....J
>-
--
--
--
W D-
w
(/)
D-
25
'" I
0
cr
0 0
2.0 1.5
(!)
"'....... (/)
w
1.0
'" '"
0.5
D-
o
CD
>-
z
>--< W
....J ....J
I-
u
cr
b c..:>
(!)
--
0'"
i=; UJ
0-
(/l
UJ Z H
::::> '" w 0
'"
1.5
BrainResearch, 583 (1992) 262-270 © 1992Elsevier Science Publishers B.V. All rights reserved 0006-8993/92/$05.00
BRES 17898
Biochemical characterization of independent olfactory receptor sites for 5'-AMP and taurine in the spiny lobster Kirby S. Olson a
a,
Henry G. Trapido-Rosenthal
Departmentof Biology, Georgia State University, Atlanta, GA (USA) and (USA)
b
b
and Charles D. Derby
a
The Whitney Laboratory, University of Florida, St. Augustine, FL
(Accepted 18 February 1992)
Key words: Olfaction; Chemoreception; Transduction; Receptor binding; Lobster; Crustacean; AMP; Taurine
To understand the initial events in chemosensory transduction (i.e. binding of odorants to olfactory receptor sites), we have utilized a radioligand-receptor binding assay on a tissue preparation that is enriched in dendritic membrane from olfactory receptor cells in the spiny lobster Panulirus argus. Radioligands used were tritiated adenosine 5'-monophosphate (AMP) and taurine, which are two of the most excitatory compounds for spiny lobsters. Our results indicate the existence of two independent types of binding sites-a taurine binding site and an AMP binding site. For both the taurine and AMP binding sites, odorant binding is rapid, reversible, and saturable. K D values for the taurine and AMP binding sites are 2.3 and 2.0 J.LM, respectively, and B max values are 159 and 3.2 fmolj J.Lg protein, respectively. The fact that the specificity, affinity, and independence of these two binding sites as defined in these biochemical studies are in agreement with those from electrophysiological studies suggests that these binding sites are olfactory receptor sites involved in sensory transduction. .
INTRODUCfION
The olfactory system of the Florida spiny lobster Panulirus argus has proven to be a useful experimental model system for understanding mechanisms of olfactory function. Peripheral and central olfactory elements have been studied in this species using various techniques, including extracellular and intracellular electrophysiology, histology and anatomical tract tracing, electron microscopy, immunocytochemistry, biochemistry, and behavioral conditioning. Studies using these techniques have led to an understanding of important olfactory phenomena including the effect of peri receptor events, such as odorant degradation and uptake, on odorant detection15.16.39.55, and involvement of second messengers and ion channels in sensory transduction in olfactory receptor cel!s29,45,46. Prior work has also investigated the mechanisms of neural coding of the quality and intensity of single odorant compounds and odorant mixtures by the peripheral and central olfactory systems20,25,28,35-37, mechanisms
of mixture interactions and their effects on neural coding of and behavioral responses to odorant mixtures 2,3.5.17,18.25- 28,32,46, and functional organization of the chemosensory processing centers in the CNS 1,49,52. An important event in chemosensory transduction which has not been directly studied in the spiny lobster is the initial event in transduction-the binding of odorant molecules to the membrane-bound receptor molecules on the olfactory receptor cells. Our knowledge of the binding properties of these receptor sites in spiny lobsters comes solely from indirect studies using electrophysiological and behavioral techniques. These indirect studies indicate that the olfactory system of P. argus is most sensitive to select amino acids, amines, adenine nucleotides, quaternary ammonium compounds, and ammonium, with response thresholds to single compounds typically being in the micromolar to nanomolar range 4,12,18,22.23,26,32. Receptor cells with narrowly tuned excitatory response spectra have been found which are best excited by either AMP, ATP, betaine, cysteine, glutamate, ammonium, or tau-
Correspondence: K. Olson, Department of Biology, Georgia State University, P.O. Box 4010, Atlanta, GA 30302-4010, USA. Fax: (I) (404) 651-2509.
263 rine 14,20,26,34 . The high degree of narrow tuning of the cells is revealed by the difference in the stimulus concentrations that produce a half-maximal response for the two most excitatory compounds for a given cell, which is often several log units apart. Cluster analysis of the cells based on their excitatory response spectra to single compounds demonstrates several distinct classes of cells20 ,26. Some responses are inhibitory or suppressive (i.e. result in a reduction in the spontaneous rate of spiking activity or a hyperpolarization from the resting membrane potential, respectively). Each cell can produce both excitatory or inhibitory responses, though not necessarily the same response to the same compounds as other cells in the population 19,21,26,44,46. Thus, these indirect studies of odorant-receptor binding based on responses to single compounds suggest that the spiny lobster's olfactory system contains numerous types of relatively independent receptor sites that mediate excitation or inhibition, and that these receptor sites are heterogeneously distributed across the olfactory cells of the system. The use of chemical mixtures in studies of the spiking responses of olfactory receptor cells of the spiny lobster reveals a potential for greater complexity in the association of odorant molecules and receptor sites than indicated by studies using single compounds53. Mixture interactions, especially mixture suppression, are common in this system2 ,3,13.21,25- 27. For example, the spiking responses of olfactory receptor cells to certain binary mixtures are often not predictable from the responses to the individual components of the mixtures26 ,27,38,41. Determining the extent to which inhibition of ligand-receptor binding contributes to mixture suppression requires direct measurements of the interaction of odorant molecules with receptor sites. Receptor binding assays using radioligands have been developed for several chemoreceptor systems, including chemoreception by unicellular organisms", larval gastropods58, hydra", olfaction in fishlO,1l ,3 1,48,51 and mammals30,50 , and gustation in fish8,42 . No such radioligand-receptor binding techniques have been applied to crustacean chemoreception, and only rarely to other arthropods 60. In this paper, we describe a radioligand-receptor binding assay using a tissue fraction that is enriched in dendritic membrane from olfactory receptor cells in the spiny lobster. We use this assay to characterize two classes of binding sites: one for adenosine 5/-monophosphate (AMP) and one for taurine. AMP and taurine were selected as stimuli because they are strong excitants of receptor cell activity 20,21,24,26,34 and behav-
ior 18 ,32,33, and electrophysiological evidence suggests that AMP and taurine probably operate through two
independent receptor sites20 ,26. The biochemical data presented in this paper demonstrate the existence of independent binding sites for AMP and for taurine, both of which have rapid, reversible, and saturable binding characteristics. Correlation of these biochemical results with existing behavioral and electrophysiological data suggests that taurine and AMP are binding to separate olfactory receptor sites.
MATERIALS AND METHODS Materials. [3H1Adenosine 5'-monophosphate (specific activity = 29.0 Ciyrnmol), [3Hltaurine (specific activity = 25.6 Ciymrnol), and [3Hladenosine (specific activity = 31.0 Ciymmol) were purchased from DuPont-New England Nuclear. All other chemicals were from Sigma Chemical Co. ANa +-free, sucrose-containing Tris buffer was prepared with the following composition: 50 mM KCl, 10 mM Tris base, 320 mM sucrose, 12.9 mM CaCl z·2H zO, 23.1 mM MgClz·6 HzO, and 25.6 mM MgS04·7 HzO, with pH adjusted to 7,8 using HC!. The pH and ion concentrations used in this buffer approximate those found in seawater to which receptor sites are normally exposed, and are identical to those of the artificial seawater used in electrophysiological studies ZO. Animals. Spiny lobsters i Panulirus argus) were collected from commercial fish houses in the Florida Keys. In experiments using frozen tissue, lateral filaments of the antennules (olfactory organ) were removed from live animals, placed in tubes filled with Tris buffer, and frozen in dry ice until they were transported to a - 80°C freezer, where they were stored until used. In experiments using fresh antennules, the lobsters were transported to the laboratory where they were held in aquaria with running seawater, and were fed a mixed diet of shrimp, squid, and fish. Lateral antennular filaments were excised from these animals as needed and used immediately. Tissue preparation. Collection of tissue enriched in dendritic membrane from receptor cells is relatively simple, due to the organization of receptor cells within antennules. The paired lateral filaments of the antennules of spiny lobsters contain ca. 600,000 chemoreceptor cells that innervate the aesthetasc sensilla'". Dendrites of these receptor cells are the only neuronal tissue within the sensilla, since the somata and axons of the receptor cells lie in the lumen of the antennule below the sensilla. Distal processes of supporting cells (glia) are also located in the sensilla40. Thus, by collecting only the aesthetasc sensilla, it is possible to harvest a tissue fraction highly enriched in olfactory receptor sites but free of non-dendritic regions of receptor cells, synapses, central nervous tissue, muscle, and most other tissue. The aesthetasc sensilla were manually removed from the antennule by scraping with a scalpel, and subsequently maintained at 4°C in Tris buffer. To extract membrane from the sensilla, the tissue was homogenized manually using a glass-glass homogenizer, and then sonicated (five times for 5 s each on setting 35 of a Fisher Sonic Dismembrator Model 300 with a 4 mm titanium tip). The tissue was centrifuged at 6,000 x g for 10 min, resulting in a pellet (PI) containing sensillar cuticle and a supernatant (S'l) containing the dendritic membrane. The Sl was removed, and the PI was resuspended in Tris buffer and centrifuged again at 6,000x g for 10 min. The resulting supernatant was removed and combined with the previous Sl. The Sl was diluted with 3 ml of Tris buffer and centrifuged at 150,000X g for 30 min. The supernatant was then removed, the pellet was resuspended in 3 ml of fresh Tris buffer, and then recentrifuged at 150,000X g for an additional 30 min to provide an additional washing of the pellet. The resulting P2 pellet was then resuspended in Tris buffer to form the P2 fraction, a suspension of tissue enriched in dendritic membrane from olfactory receptor cells, which was used in subsequent binding assays. Binding assay. Binding activity of the olfactory tissue was assayed by incubating aliquots of the P2 fraction and radiolabel in Tris buffer at 4°C. The protein concentration of the tissue in each incubation tube
264 was approximately 25 ~g/tube in a final incubation volume of 50 ~I (equivalent to approximately 3 antennulesy'tube). The concentration of radiolabel was 1 ~M, unless specified otherwise . Incubations were 60 min in duration, except where otherwise indicated. (This time was determined from association kinetics experiments to be sufficient for saturation of the binding sites with labeled ligand.) Following incubation, bound radioligand was separated from free radioligand by rapid filtration under vacuum through 0.45 ~m pore-size cellulose acetate filters (Type HAWP, Millipore) pre-soaked in 0.3% polyethylenimine in buffer, using a Hoefer vacuum filtration manifold. Filters were washed with two 5-ml vols. of cold Tris buffer until dry, and were then dissolved by placing them in scintillation vials containing 1 ml of ethylene glycol monomethylether for 30 min prior to addition of scintillation fluid. Radioactivity was measured in Ecolite( +) (lCN Biomedical Inc.) using a Beckman model LS5801 liquid scintillation counter, with a counting efficiency of 43-50%. Total binding and non-specific binding were characterized by incubation of tissue and radiolabeled ligand in the absence or presence of 1 mM concentration of the unlabeled ligand. Specific binding was defined as the difference between total binding and non-specific binding. Binding is expressed in terms of fmol of labeled compound bound per ~g of membrane protein . Protein concentrations were determined according to Bradford ? using bovine serum albumin as the standard. The binding data were statistically analyzed using Graphpad Software (INPLOT). Association and dissociation experiments. Association experiments were performed to determine the time of incubation necessary to attain steady-state binding. P2 fractions were incubated with 1 ~M radioligand for 0, 5, 10, 15, 30, 45, 60 and 75 min, with (non-specific binding) or without (total binding) 1 mM unlabeled ligand. Dissociation experiments were performed to examine reversibility of binding. P2 fractions were incubated with 1 ~M radioligand for 60 min, at which time 1 mM unlabeled ligand was added. Incubation then continued for 0, 1, 2, 3, 4, 5, 10, and 30 min, at which time the samples were filtered. Saturation experiments. Binding was measured for the P2 fraction incubated in increasing concentrations of ligand. Ligand concentrations from 10 nM to 'Il.l mM were tested by diluting 1 ~M radioligand with sufficient concentrations of either buffer (for concentrations less than 1 ~M) or unlabeled ligand (for concentrations greater than 1 ~M). The amount of binding at each ligand concentration was calculated by adjusting the specific activity of the radioligand to account for the added unlabeled ligand. Inhibition experiments. To determine if AMP and taurine bind to independent binding sites, the P2 fraction was incubated in 1 ~M radioligand alone or in 1 ~M radioligand plus increasing concentrations (10 nM to 1 mM) of unlabeled AMP or taurine. Specific binding in the absence of unlabeled ligand was defined as 100%, and binding in the presence of unlabeled ligand was expressed as a percentage of this value. When the unlabeled ligand was the same compound as the labeled ligand, the binding at 1 mM of unlabeled ligand was defined as the non-specific binding. These experiments were designed to distinguish independence of the two binding sites and to .determine the IC so values (the concentration of unlabelled compound that inhibits 50% of the maximum specific binding), which are used to compare the degree of inhibition by different compounds . Adenosine production assay. For determinations of dephosphorylation of AMP to adenosine, frozen tissue was prepared as a P2 fraction. Aliquots of this fraction equivalent to 3 antennules each were incubated in 500 1£1 of buffer with 1 ~M [3H)AMP, 1 mM adenosine, and various concentrations of unlabeled AMP, a,l3-methyleneadenosine 5'-diphosphate (AMPCP), or /:I-glycerophosphate (I3GP). After 1 h of incubation at 4°C, samples of incubation media were removed, and unlabeled AMP, ADP, and ATP were all added to a concentration of 1 mM. The samples were then stored at - 80°C until they were analyzed for [3Hladenosine by anion-exchange HPLC and liquid scintillation spectrophotometry", Specific enzymatic activity was determined by subtracting from the above values (which represent total enzymatic activity) the values for non-specific dephosphorylation of [3H)AMP that occurred in incubations containing 1 mM unlabeled AMP.
RESULTS
Preliminary experiments Significant specific binding for [3H}AMP and for [3Hltaurine was found in the P2 tissue fraction. Specific binding was 70-80% of total binding for taurine, and 50-70% of total binding for AMP. However, no specific binding was found in the S2 and PI tissue fractions from fresh antennules (n = 6 from 2 experiments run in triplicate; P> 0.05, r-test). Both the S2 and P2 were also tested for their ability to convert AMP to adenosine via 5'-ectonucleotidase or alkaline phosphatase. The S2 fraction converts 8 pmolyp,g protein while the P2 converts 16 pmol /u g protein; so, unlike specific binding, enzyme activity is not restricted to the P2. Binding was similar for P2 fraction prepared from fresh antennules and for P2 fraction prepared from antennules previously frozen at -80D e in Tris buffer. Values for specific binding of [3H]taurine in these two conditions were 7.56 ± 1.07 vs. 7.00 ± 1.36 fmol /g.g protein, respectively (mean ± S.E.M., n = 9 from 3 experiments run in triplicate; P> 0.05, r-test), Thus, for convenience, frozen antennules were used to prepare P2 in all subsequent experiments.
Association and dissociation experiments Association curves for specific binding of [3H]taurine and [3H]AMP are shown in Figs. 1 and 2, respectively. Dissociation curves are shown in Figs. 3 and 4. Specific binding values (in fmol/p,g protein) and curves fitted to a one-site binding model are shown for each analysis. The results show that specific binding of [3H]taurine and [3H]AMP is rapid and reversible, which is characteristic of the binding of odorants to receptor sites.
[3H]-TAURINE ASSOCIATION CURVE 0 Z
10
::J
a
CD
>-
Z
..J ..J
H
u
I-
-c H
LL.
0
0::
H
CL
U W
~
CL
VI
---
H
a
Z
::J
..:
6
"'-
W
0::
8
w
VI
W
..J
::E LL. ~
lt
t
10
20
30 40 50 TIME ( MI N )
60
70
80
Fig. 1. Association of [3Hltaurine to P2 olfactory tissue fraction. Values are means ± S.E.M. for one experiment run in triplicate . Solid curve represents the one-site model that most closely fits the data.
265 [3H]-AMP ASSOCIATION CURVE 0
z
[3H]-AMP DISSOCIATION CURVE
100
3.0
0
z
:::>
:::>
0
CD
0
en z
75
>-
....J ....J
>-
--
--
--
W D-
w
(/)
D-
25
'" I
0
cr
0 0
2.0 1.5
(!)
"'....... (/)
w
1.0
'" '"
0.5
D-
o
CD
>-
z
>--< W
....J ....J
I-
u
cr
b c..:>
(!)
--
0'"
i=; UJ
0-
(/l
UJ Z H
::::> '" w 0
'"
1.5
