- Email: [email protected]
Electrophysiological demonstration of independent olfactory receptor types and associated neuronal responses in the trout olfactory bulb
Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
Electrophysiological demonstration of independent olfactory receptor types and associated neuronal responses in the trout olfactory bulb ´ ´ Labergea,*, Toshiaki J. Haraa,b Frederic a
Department of Zoology, University of Manitoba, Winnipeg, Manitoba, Canada R3T 2N2 Freshwater Institute, Fisheries and Oceans Canada, Winnipeg, Manitoba, Canada R3T 2N6
b
Received 25 April 2003; received in revised form 28 October 2003; accepted 29 October 2003
Abstract The present study attempts to highlight the principles by which peripheral olfactory information of across- and withinclass odorant signals is transformed into bulbar neuron responses. For this purpose, we performed electro-olfactogram cross-adaptation and mixture experiments as well as single unit recording of olfactory bulb neurons using amino acid, bile acid and F-prostaglandin stimulants in brown and rainbow trout. The results show that amino acids, a bile acid and a F-prostaglandin activate independent receptor types. However, within the class of amino acids, different receptor types are only partially independent. Neurons responsive to bile acid and amino acids were segregated to the mid-dorsal and latero-posterior olfactory bulb, respectively. Of the 43 responsive olfactory bulb neurons studied in brown trout, 41 showed specificity for one odorant class. Olfactory bulb neurons gained responsiveness to new amino acids with increasing stimulant concentration. We conclude that different odorant classes activate specific neurons located in different regions of the trout olfactory bulb, and that information distinguishing related amino acids can be represented in a limited number of bulbar neurons with distinct response profiles under the conditions investigated. 䊚 2003 Elsevier Inc. All rights reserved. Keywords: Olfaction; Fish; Salmonidae; Olfactory bulb; Amino acid; Extracellular recording; Electro-olfactogram; Receptor; Mixture; Cross-adaptation
1. Introduction In vertebrates, the detection of odorants by the olfactory system is accomplished by receptors found on the dendritic part of sensory neurons residing in large areas of the olfactory epithelium. Research shows that primary olfactory sensory neurons converge on specific areas of the olfactory bulb glomerular layer according to the receptor type they express (Ressler et al., 1994; Vassar et al., 1994; Mombaerts et al., 1996; Wang et al., 1998). This organizational feature is not restricted *Corresponding author. Tel.: q1-49-4221-9160224; fax: q 1-49-4221-9160179. E-mail address: [email protected] (F. Laberge).
to vertebrates as demonstrated in Drosophila (Gao et al., 2000; Vosshall et al., 2000). Therefore, a spatial pattern of activity observed in the olfactory bulb upon stimulation with an odorant could represent which receptor type(s) is activated. In fish, electroencephalographic (EEG) recording of neuronal activity induced by stimulation with different odorant classes demonstrated a topography of olfactory projections to the olfactory ¨ bulb in salmonids (Thommesen, 1978; Doving et al., 1980; Hara and Zhang, 1996, 1998). A chemotopy of odor representations in the olfactory bulb has recently been demonstrated in channel catfish by using both EEG and single unit recording (Nikonov and Caprio, 2001). Responses to
1095-6433/04/$ - see front matter 䊚 2003 Elsevier Inc. All rights reserved. doi:10.1016/S1095-6433(03)00345-3
398
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
mixtures of amino acids, bile acids and nucleotides were found in different regions of the catfish olfactory bulb. These results show that the selectivity of single olfactory bulb neurons underlies the specific foci of EEG activity obtained with different odorant classes in teleosts. Imaging of primary olfactory neuron synaptic terminals in the olfactory bulb has been used successfully to localize afferents activated by different odorants in zebrafish and the mouse (Friedrich and Korsching, 1997, 1998; Fuss and Korsching, 2001; Wachowiak and Cohen, 2001; Fried et al., 2002). Different odorants show specific bulbar afferent activation patterns even if some overlap of response exists in the case of related odorants. The correspondence between these afferent maps and the activation of secondary olfactory neurons is not well studied. The fidelity of the afferent’s image will be maintained in second-order responses on the condition that intrabulbar processing is absent or minimal. Research in zebrafish shows that despite specific bulbar activation patterns for single amino acids demonstrated by imaging studies (Friedrich and Korsching, 1997, 1998; Fuss and Korsching, 2001), individual mitral cell responses are not specific to a single amino acid (Friedrich and Laurent, 2001), suggestive of a combinatorial code for amino acid quality. Friedrich and Laurent (2001) further suggested that temporal patterning of mitral cell population responses during the course of a stimulus increase the distinctiveness of amino acid representations in zebrafish. Most single olfactory bulb neurons also respond to several amino acids in other fish species (MacLoed, 1976; Bodznick, 1978; Meredith, 1981; Kang and Caprio, 1995). Recent results show that most olfactory bulb neurons are not even specific to an odorant class in goldfish (Hanson and Sorensen, 2001; Masterman et al., 2001). In the present study, single olfactory bulb neuron responses were recorded to study how they correlate with the established topography of EEG responses to odorants of different classes in salmonid fish. Additionally, neuron response characteristics within a zone of the olfactory bulb responsive to amino acids were investigated to examine if the relative independence of amino acid receptor mechanisms found in the olfactory epithelium could be maintained in the response profiles of olfactory bulb neurons.
2. Materials and methods 2.1. Fish maintenance Brown trout (Salmo trutta), 1–3 years old, and rainbow trout (Oncorhynchus mykiss) of the Langley strain (1–2 years old) were originally obtained from the Whiteshell Provincial Fish Hatchery, Manitoba, Canada. Additional rainbow trout of the Tagwerker strain, 2–4 years old, were obtained from the Rockwood Aquaculture Research Centre, Freshwater Institute. They were held in laboratory tanks at the Freshwater Institute with constant flowing aerated, dechlorinated Winnipeg city water (10.5–11.5 8C). Lighting conditions were 12 h on–12 h off with dusk and dawn simulation accomplished by low intensity light bulbs on 30 min before and after the 12 h illumination period. The fish were fed to satiation twice a week with commercial trout pellets. All experiments complied with the Canadian Council on Animal Care guidelines. 2.2. Electrophysiological recordings Fish were tranquilized by exposure to water containing MS-222 (0.5 gyl), fully anesthetized by intraperitoneal injection of amobarbital (30 mgykg body mass), and immobilized by intramuscular injection of Flaxedil (gallamine triethiodide; 3–5 mgykg body mass) before being secured on a holder in a flow-through trough. The gills were continuously perfused (0.4 lymin) with dechlorinated water. For electro-olfactogram (EOG) recording, skin and cartilage covering the right side naris were removed and the exposed naris was continuously perfused with dechlorinated water. Exposed parts of the fish were covered with wet tissue and kept moist throughout the experiment. EOGs were recorded as previously described (Evans and Hara, 1985; Sveinsson and Hara, 2000). The recording electrode position above the olfactory epithelium was adjusted so that a maximal response to a standard stimulant, 10y5 M L-serine, was achieved. Electrical signals were amplified with a DC preamplifier (Type 7P1, Grass Instruments, Quincy, MA) and recorded on a polygraph (Model 79, Grass Instruments). For single unit recording, the roof of the skull was opened, and cartilage and mesenchymal tissues removed to expose the dorsal brain from the olfactory nerves to the telencephalon. Tungsten
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
microelectrodes (impedances5 MV, World Precision Instruments, Sarasota, FL) were used. A reference electrode was placed on the dorsal skin of the snout. Electrical signals were amplified with an AC amplifier (model P511, Grass Instruments; 1y2 amplitude lower than 30 Hz and higher than 3 kHz) and fed to a slopeyheight window discriminator (Frederick Haer and Co., Bowdoinham, ME) to isolate single units. A typical recording comprised several units firing well above noise level. In order to be confident that the output of the window discriminator represented the activity of a single unit, the position of the recording electrode was adjusted until an easily isolable unit was obtained. Unit activity was displayed on an oscilloscope and spike magnitude and shape were carefully compared over time to verify that the signal of interest was from a single unit. The outputs were then recorded on a polygraph (model 79, Grass Instruments). Electrode surface location and depth were noted for every recording. A constant monitoring of neural activity was also maintained with an audio output to a speaker. 2.3. Chemical stimulation To perfuse the olfactory epithelium and deliver chemical stimuli without flow interruption, the method of Sveinsson and Hara (2000) was used. A minimal recovery period of 2 min between each 10 s stimulation was allowed. Stock solutions of test stimulants were prepared with distilled water. Stock solutions were stored at 4 8C and aliquots (10 or 100 ml) were diluted with 10 ml of dechlorinated water immediately before testing. The stimulants used were several L-amino acids (alanine, Ala; arginine, Arg; aspartic acid, Asp; cysteine, Cys; glutamine, Gln; glutamic acid, Glu; histidine, His; lysine, Lys; methionine, Met; serine, Ser; tryptophan, Trp; tyrosine, Tyr), the bile acid taurocholic acid (TCA) and prostaglandin F2a (PGF2a). PGF2a was only used with brown trout because the rainbow trout olfactory system is insensitive to this chemical (Kitamura et al., 1994; Hara and Zhang, 1998; Laberge and Hara, 2003a). The chemicals were purchased from Sigma Chemical Co (St. Louis, MO) with the exception of PGF2a, which was from Cayman Chemical (Ann Arbor, MI). Cross-adaptation experiments were performed to determine whether multiple receptor types are involved in the detection of amino acids in trout.
399
These experiments were done by replacing the constant flow of dechlorinated water with an adapting amino acid solution. The adapting solution (brown trout: 10y5 M Arg, Cys, Glu or Ser; rainbow trout: 10y3 M Arg or Cys) was perfused over the olfactory organ until the response declined and stabilized at the tonic level. Then, various test stimuli (listed above) dissolved in the adapting stimulus solution were applied at 2 min intervals (Sveinsson and Hara, 1990). Each stimulant was tested before, during and after adaptation. The adapted response to a stimulant was divided by the averaged responses before and after adaptation to express the cross-adaptation results in percentage of unadapted responses. PGF2a and TCA (10y7 M) were used in these experiments to represent additional odorant classes and compare the level of independence between across- and within-class receptor types. Binary and trinary mixture experiments were also performed with low concentrations of amino acids to supplement the cross-adaptation experiments. They consisted of EOG tests of two equipotent amino acid solutions, a mixture containing the two stimulants in the same concentration as in the individual tests, and the tests of twice the concentration of the mixture’s components as previously done in catfish (Caprio et al., 1989). The concentrations used were 10y6 M for Ser and Arg, 10y5 or 2=10y5 M for Glu, and between 10y8 and 10y7 M for Cys. Variations in concentration show that some adjustments were needed for the solutions to have equivalent EOG responses in each fish. Each fish was tested at least three times with each solution to allow the measurement of potentially small EOG amplitude differences. In the case of the trinary mixture, three equipotent solutions were used along with tests of three times of the mixture component’s concentration. The method used in Caprio et al. (1989) was followed for calculation of the mixture discrimination index (MDI) and independent component index (ICI). MDI was measured by dividing the response to the mixture by the larger response to twice the concentration of one of its components. ICI was measured by dividing the response to the mixture by the sum of the responses to its components presented individually. Theoretically, if the stimulants in the mixture activate the same olfactory receptive mechanism, the MDI equals 1. Mixture suppression would be indicated by a MDI significantly less than 1, whereas mixture enhancement
400
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
odorant classes and a water control. The amino acids were tested at 10y5 M while the bile acid and prostaglandin were tested at 10y7 M. The amino acids Cys, Arg, Ser and Glu (only tested in rainbow trout) were selected to represent stimulants within an odorant class. These amino acids were chosen to represent odorants that activate similar (Cys and Ser) or independent but overlapping receptive mechanisms (Cys and Arg) according to our EOG results. The determination of neuron response thresholds was only attempted in rainbow trout. 2.4. Statistics
Fig. 1. EOG concentration–response relationships in brown (A) and rainbow (B) trout for selected amino acids. Response magnitude is expressed as a percentage of the response to 10y5 M L-serine. Stimulants: L-cysteine (solid squares), L-serine (open squares), L-arginine (solid circles), and L-glutamic acid (open circles). Each point represents mean"S.D. The sample size is 4 in brown and 5 in rainbow trout (Tagwerker and Langley strains combined).
would be indicated by a MDI significantly greater than 1. Research on amino acid olfaction done in catfish showed no evidence of mixture suppression and concluded that the mechanism for mixture enhancement is the simultaneous activation of multiple types of receptor binding sites by the different components of the mixture (Caprio et al., 1989; Kang and Caprio, 1991). Thus, we expected to observe a mixture enhancement with amino acid pairs that activate different receptive mechanisms and an even larger mixture enhancement, as reflected by a large MDI, with the trinary mixture of amino acids if this mixture activated more receptive mechanisms than a binary mixture. An ICI equal to 1 would mean that the components of a mixture are independent. The lower the ICI value is, the more related the mixture components are. Responses of single olfactory bulb neurons were recorded to representative stimulants from three
A one-way ANOVA followed by a Dunnett post hoc test upon significance (P-0.05) was performed on the percentage of unadapted responses obtained in the cross-adaptation experiments. A significant response in the Dunnett test between the stimulation with the chemical used as adaptant and another stimulant was taken as evidence of the activation of distinct receptor sites. A twotailed paired t-test comparing the EOG response magnitudes of a binary mixture and twice the concentration of its most potent component was performed to determine if they differed. The statistical threshold was set at (P-0.05) for the ttests. The same test was applied to the mixture and three times the concentration of its most potent component for the trinary mixture. Even though most single neuron response types could be determined visually, statistical analysis proved useful when dealing with small responses. ANOVAs were calculated on single trials. For this purpose, a neuron activity record was divided into 30-s prestimulus, 10-s stimulus, and 30-s post-stimulus periods. The periods were divided into 3-s (prestimulus and post-stimulus) or 2.5-s (stimulus) time bins and spikes were counted in each time bin. The 2.5-s time bin counts were multiplied by a factor of 1.2 to make them equivalent to the 3-s time bins. Upon a significant ANOVA result (P0.05), Tukey post hoc tests were done to determine which period(s) was responsible for the significant increase or decrease in spike frequency. 3. Results 3.1. EOG responses The EOG concentration–response relationships for the four amino acids used in brown trout are
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
401
Fig. 2. EOG responses of brown trout to selected chemicals during adaptation to 10y5 M L-cysteine (A); 10y5 M L-serine (B); 10y5 M arginine (C) and 10y5 M L-glutamic acid (D). Sample size is 5 in A, B, C and 3 in D. Asterisks indicate a statistically significant difference when compared to the test of the adapting stimulus. Vertical bars represent mean"S.D.
shown in Fig. 1A. Note that Cys is very potent and that the response threshold is approximately 10y8 M for all amino acids used. These curves helped us to determine the equipotent amino acid concentrations for the mixture experiments in brown trout. The same curves for amino acids used in EOG experiments in rainbow trout are shown in Fig. 1B. No obvious differences could be observed between the EOG responses of Tagwerker and Langley rainbow trout (data not shown). The results from brown trout cross-adaptation experiments are shown in Fig. 2. Adaptation with Cys (Fig. 2A), the most potent amino acid, suppressed significantly responses to Cys and Ser, but not to PGF2a, TCA, Arg and Glu. Adaptation with Ser, Arg and Glu (Fig. 2B–D) only suppressed significantly the responses to the adaptant amino acid. The responses to the chemical classes of prostaglandins and bile acids did not differ significantly before and during adaptation to amino acids. Fig. 3 shows the results of rainbow trout cross-adaptation experiments using high concentrations (10y3 M) of Cys and Arg. Only TCA had a
significant response under Cys adaptation while the responses to Arg and Lys, although not significant, were not suppressed totally. The small response to Glu under Cys adaptation was the result of a response in only one fish and is not significant. Responses to the amino acids Trp, Tyr, Ala, Ser, Asp, Gly, Gln, His and Met were suppressed totally by Cys adaptation, while adaptation with Arg totally suppressed only responses to Arg itself, Lys and Gly. Statistical analysis showed that all the binary mixtures of the amino acids used in brown trout had greater responses than expected if the concentration of the most potent of the mixture’s component was doubled, except in the case of the Glu-Cys mixture (Ps0.07) (Table 1). The calculated values of MDI were all larger than 1, supporting the existence of mixture enhancement for the selected amino acids. Also shown on Table 1 are the values of ICI for these mixtures. The trinary mixture of Cys, Ser and Arg also showed a statistically significant mixture enhancement, along with a large MDI and a small ICI (Table 1). The binary mixture experiments conducted in rain-
402
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
Fig. 3. EOG responses of rainbow trout to selected amino acids tested at 10y5 M and TCA at 10y7 M during adaptation to 10y3 M y3 L-cysteine (A) and 10 M L-arginine (B). Sample size is indicated above the bars. Asterisks indicate a statistically significant difference when compared to the test of the adapting stimulus. Vertical bars represent mean"S.D.
bow trout yielded results comparable to those obtained in brown trout with the exceptions that the Ser-Cys mixture enhancement was not significant despite a MDI greater than 1 and the mixture of Glu and Cys produced a significant enhancement (Table 2). The amino acid pair Arg-Lys was additionally tested in rainbow trout after the crossadaptation results revealed the similarity between these two stimulants. The MDI value of 1 and a small ICI value were unique to this amino acid pair in the present study. 3.2. Single neuron recording in the olfactory bulb Of the total of 84 neurons recorded in eight brown trout, 35 neurons responded specifically to
amino acids, 6 neurons responded specifically to the bile acid TCA, 2 neurons responded to all the stimulants tested and 41 neurons did not respond to any of the stimulants tested (Table 3). Neurons responding specifically to PGF2a could not be found. A few additional neurons responded to the water control, presumably because of mechanical disruption of the flow to the naris, and are excluded from the present analysis. The response thresholds of rainbow trout olfactory bulb neurons to TCA and the amino acids Cys, Arg, Ser and Glu were determined successfully in four neurons. The response type did not change with increasing stimulant concentration. These results are listed in Table 4 along with the specificity of five additional rainbow trout neurons tested at 10y5 M only.
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408 Table 1 Results of the mixture experiments in brown trout
403
Table 2 Results of the mixture experiments in rainbow trout
Mixture (n)
MDIa
ICIb
Statistical significancec
Mixture (n)
MDIa
ICIb
Statistical significancec
Glu-Arg (4) Glu-Cys (4) Glu-Ser (4) Arg-Ser (6) Arg-Cys (6) Ser-Cys (6) Arg-Ser-Cys (5)
1.203 1.141 1.298 1.277 1.228 1.239 1.343
0.772 0.806 0.949 0.903 0.805 0.850 0.691
Yes No Yes Yes Yes Yes Yes
Glu-Arg (5) Glu-Cys (5) Glu-Ser (5) Arg-Ser (5) Arg-Cys (5) Arg-Lys (4) Ser-Cys (5)
1.336 1.220 1.297 1.266 1.224 1.011 1.180
0.856 0.887 0.815 0.849 0.894 0.646 0.792
Yes Yes Yes Yes Yes No No
a MDI, mixture discrimination index. In binary mixtures, MDIsABy(2A or 2B). In the trinary mixture, MDIsABCy (3A, 3B or 3C). b ICI, independent component index. In binary mixtures, ICIsABy(AqB). In the trinary mixture, ICIsABCy(AqBq C). c A paired t-test was performed to determine if the amplitude of the mixture response differed significantly from the response to twice (AB/2A or 2B) or thrice (ABC/3A, 3B or 3C) the concentration of its most stimulatory component.
3.3. Distribution of the amino acid and bile acid responsive neurons in brown trout For the purpose of the topographical analysis, the olfactory bulb was divided into dorsal and ventral halves at 1000 mm deep. Morphological observation showed that the maximal depth of the olfactory bulb was approximately 2000 mm. Further, these halves were divided into a rostral zone, and two medial and lateral zones at the middle and posterior levels of the bulb for a total of ten bulbar regions. Fig. 4 illustrates the general regions where olfactory bulb neurons responsive to amino acids and the bile acid TCA were found. Bile acid
a MDI, mixture discrimination index. MDIsABy(2A or 2B). b ICI, independent component index. ICIsABy(AqB). c A paired t-test was performed to determine if the amplitude of the mixture response differed significantly from the response to twice the concentration of its most stimulatory component (AB/2A or 2B).
responsive neurons were found along the dorsal midline of the mid-section of the olfactory bulb. Amino acid responsive neurons were found mainly along the dorsoventral extent of the latero-posterior bulb. Some neurons responsive to amino acids were also found in the ventral part of the medioposterior olfactory bulb and, occasionally in the middle of the bulb. Also shown in Fig. 4 are some examples of neuron responses recorded from the olfactory bulb. The proportion of neurons of the posterior olfactory bulb responding to the selected stimuli used in this study was greater in the lateral part compared to the medial part (79% of 33 neurons, lateral; 41% of 22 neurons, medial). Distribution of the electrode surface positions of single neuron recordings could not highlight differences related to cell response specificity with-
Table 3 Response specificity of recorded single neurons in brown trout Response type Cys(q) Ser(q) Cys(q) Cys(q) Ser(q) TCA(q) Arg(q) Cys(q) Ser(y) Cys(q) Ser(y) Cys(y) Ser(q) Cys(y) Ser(y) Ser(q) Arg(y) Cys(q) Ser(q) Cys(y) Ser(y) No response
Number of neurons Arg(q)
Arg(q) Arg(y)
Arg(q) TCA(q) PGF(q) Arg(y) TCA(y) PGF(y)
(q) and (y) indicates increasedydecreased firing.
13 7 6 6 2 2 1 1 1 1 1 1 1 41
404
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
Table 4 Response specificity of recorded single neurons in rainbow trout Response type Neurons with response thresholdsa Cys(q) 10y6 Ser(q) 10y5 Arg(q) 10y5 Cys(q) 10y6 Ser(q) 10y6 Arg(q) 10y5 Glu(q) 10y5 Arg(q) 10y5 Lys(q) 10y5b Arg(y) 10y7 Cys(q) 10y5 Ser(q) 5=10y5 Neurons tested at 10y5 M only Cys(q) Ser(q) Glu(q) Cys(q) Ser(q) Cys(q) Arg(q) Ser(y) Glu(y) Arg(y) (q) and (y) indicates increasedydecreased firing. a The concentrations tested were between 10y8 and 5=10y5 M. b L-lysine was only tested in this neuron.
in the amino acid responsive zones (not shown). Therefore, neurons with different response specificities were divided into five groups and graphed according to their recording depth in order to see if different layers of the olfactory bulb could be responsible for most responses of one given type. The five groups are, in the order presented in Fig. 5: Cys(q)-Ser(q)-Arg(q), all neurons showing inhibitory responses, Cys(q), Cys(q)-Ser(q) and neurons of the posterior bulb showing no response. Bile acid neurons were excluded from this analysis because they were clearly found only in the dorsal olfactory bulb. Fig. 5 shows that no clear relationship exists between recording depth distribution and neuron response profiles in the brown trout olfactory bulb. 4. Discussion 4.1. Evidence for multiple amino acid receptor types in trout Previous electrophysiological studies done in fish highlighted the presence of several partially independent receptor types involved in the detection of amino acids (Caprio, 1982; Hara, 1982; Ohno et al., 1984; Caprio et al., 1989; Sveinsson and Hara, 1990) and bile acids (Zhang and Hara, 1994; Li and Sorensen, 1997; Michel and Derbidge, 1997) in the olfactory epithelium. Our crossadaptation and mixture results strongly suggest that multiple receptor types are involved in amino acid detection in both brown and rainbow trout
without major differences between the two species. The extent to which one odorant’s response is suppressed by adaptation of the olfactory organ with a second odorant could represent a measure of independence of the receptor sites binding the two odorants. From Fig. 2A, for example, Ser and Cys would stimulate the same or similar receptor sites, while Arg would stimulate a great number of receptor sites independent of those for Cys or Ser. PGF2a or TCA would activate a receptor site group independent of all the receptors involved in amino acid detection. A large MDI value in the trinary mixture experiment with Cys, Arg and Ser compared to the smaller values of MDI obtained in binary mixtures using the same chemicals strongly supports the idea that each amino acid used in the brown trout study activates an independent receptor type. This finding suggests that different neutral amino acid receptor types exist in trout in addition to the basic and acidic types as was previously demonstrated in catfish by the use of complex amino acid mixtures (Kang and Caprio, 1991). Kang and Caprio (1991) also proposed that more than one common receptor type is involved in the detection of basic amino acids in catfish. However, the MDI value very close to 1 obtained with the binary mixture of Arg and Lys in rainbow trout provides no evidence for more than one basic amino acid receptor type in trout. 4.2. Neuronal responses in the olfactory bulb The specificity of single olfactory bulb neurons was investigated in the brown trout by testing with three different odor classes (amino acid, bile acid, prostaglandin) as well as testing three different amino acids having similar (Cys-Ser) or different (Cys-Arg) binding characteristics as shown by the EOG data. Several neurons specifically responsive to TCA or amino acids were found. They were distributed in different bulb regions resembling those found in a previous EEG investigation of salmonid bulbar responses (Hara and Zhang, 1998). The TCA responsive region is centered in the mid-section of the dorsal olfactory bulb and the amino acid responsive region in the lateral part of the posterior bulb. The amino acid and bile acid zones are not totally exclusive since a few amino acid responsive neurons were found in the TCA region and vice versa. Overall, the bulk of neurons responsive to each odorant class are segregated into different bulbar regions. The regions respon-
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
405
Fig. 4. Darkened regions show the general distribution of bile acid (A) and amino acid (B) responsive units in the brown trout olfactory bulb. The drawings represent sections of the left olfactory bulb. The olfactory nerve is to the right and the telencephalon to the left. The activity of five representative single neurons is also shown around the bulb drawings. The neuron response type is specified in a box above the traces and the bar under the traces represents the 10 s stimulus duration. The stimuli tested for each neuron are (from top to bottom) L-cysteine, L-arginine, L-serine, TCA and prostaglandin F2a.
sive to amino acids and bile acids in brown trout differ from their equivalent regions in catfish and zebrafish. In catfish and zebrafish, the amino acid responsive region is found more rostrally than in salmonids, and the presence of bile acid responses in the ventral olfactory bulb is restricted to catfish (Friedrich and Korsching, 1998; Nikonov and Caprio, 2001). Most amino acid responsive neurons showed increased firing when stimulated with 10y5 M of all three amino acids used or Cys alone, the most potent amino acid. However, complex response patterns comprising inhibition and responses restricted to Arg or Ser alone were observed. The demonstration of bulbar responses restricted to Arg or Ser confirms the mixture results that receptor types for basic (Arg) and more than one shortchain neutral (Ser and Cys) amino acids exist in
the brown trout olfactory epithelium. As an example, the observed excitatory neuron responding to Ser only cannot be explained by intra-bulbar processing because Ser has EOG characteristics close to Cys, but is less potent than Cys. Primary input to the bulb has to be the source of the specific Ser response. Not enough amino acids were tested to make the claim that these bulbar neurons are specific for an amino acid or a group of closely related amino acids, but the selectivity revealed in a limited number of neurons seems responsible for the uniqueness of related amino acid representations by olfactory bulb neurons. The determination of amino acid response thresholds in rainbow trout bulbar neurons clearly shows that olfactory bulb neurons can gain responses to new amino acids with increasing stimulant concentration. This finding could result
406
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
Fig. 5. Depth distribution of some brown trout single neurons according to their response types. Groups are: (1) Cys(q)Ser(q)-Arg(q) neurons; (2) all neurons showing inhibitory responses; (3) Cys(q) neurons; (4) Cys(q)-Ser(q) neurons; (5) neurons of the posterior bulb showing no response.
from an amino acid receptor loss of specificity with increasing stimulant concentration or the expression of more than one receptor type in each sensory neuron (Medler et al., 1998). Alternatively, the input of olfactory receptor neurons bearing different amino acid receptor types could converge on single olfactory bulb neurons because fish mitral cells innervate multiple glomeruli (Allison, 1953). The idea that higher concentration of an amino acid activates more receptor types is supported by the observation that binary mixtures of high amino acid concentrations do not produce a clear mixture enhancement in brown trout (F. Laberge, personal observation). A gradual loss of specificity in the activity patterns of olfactory bulb afferents stimulated with increasing amino acid concentrations in zebrafish also supports the above idea (Fuss and Korsching, 2001). The number of effective odorants to activate a mouse olfactory receptor expressed in a heterologous system was also shown to change with odorant concentration (Kajiya et al., 2001). These findings imply that the identity of an amino acid olfactory signal could be lost or change with increasing stimulant concentration, an issue that deserves further investigation. Unfortunately, it is not known if the high amino acid concentrations (10y5 M and higher) used in single neuron recording studies, including this one, enable us to study biologically relevant olfactory responses. Indeed, it is not known if fish can be conditioned to discriminate high concentrations of
amino acids (Zippel et al., 1993 showed that goldfish could not), with the possible exception of conditioned cardiac responses obtained in channel catfish (Little, 1981). Thus, future olfactory discrimination conditioning experiments in fish will have to take into account odor concentrations and variations in concentration. The existence of nonspecific bulbar neurons in goldfish points to species differences in olfactory coding strategies used at the bulb level (Hanson and Sorensen, 2001; Masterman et al., 2001). Thus, the finding of Friedrich and Laurent (2001) that stimulus-specific temporal patterns of firing could be used in amino acid signal optimization by zebrafish mitral cells could derive from the absence of selective bulbar neurons like those observed in brown trout. Surprisingly, neurons specific for PGF could not be found in brown trout despite robust EOG responses to this chemical class (Hara and Zhang, 1998; Laberge and Hara, 2003a). The telencephalon and preoptic area were also unsuccessfully probed for PGF responses (F. Laberge, unpublished observation). This stands in marked contrast to our recent finding of a PGF-specific neuron population in the lake whitefish olfactory system, a related salmonid fish (Laberge and Hara, 2003b). The lack of PGF response in the brown trout bulb could be explained by the existence of a small group of responsive neurons that would have been missed in the present investigation. Alternatively, the responsiveness of the PGF second-order neurons could be seasonal. The fish used here were kept on a 12 h light–12 h dark photoperiod and did not reproduce in captivity. Another possibility is that the responsiveness of PGF neurons could be activated by intra-specific interactions. More research is needed in brown trout because of the recent demonstration of the potential for PGF2a to be a reproductive pheromone acting on both sexes of that species (Laberge and Hara, 2003a; Moore et al., 2002). In conclusion, stimulants from different odorant classes stimulate specific neurons located in different regions of the trout olfactory bulb. Multiple olfactory receptor types are involved in amino acid detection in trout. Stimulant concentration could be an important determinant of the number of receptor types and olfactory bulb neurons activated by amino acids. The existence of bulbar neurons with selective response profiles to one amino acid even at high stimulant concentration, as revealed in this study, suggests that these neurons could
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
potentially be used by the trout olfactory system to encode amino acid quality. Acknowledgments This study was supported by Natural Sciences and Engineering Research Council of Canada grant OGP 0007576. F. Laberge was supported by Fonds pour la Formation de Chercheurs et l’Aide a` la ´ Recherche (Quebec) and University of Manitoba Graduate Fellowship. Brown trout were kindly provided by the West Hawk Lake Hatchery, Department of Natural Resources, Manitoba. Four anonymous reviewers provided helpful comments on the manuscript. References Allison, A.C., 1953. The morphology of the olfactory system in the vertebrates. Biol. Rev. 28, 195–244. Bodznick, D., 1978. Characterization of olfactory bulb units of sockeye salmon with behaviorally relevant stimuli. J. Comp. Physiol. A 127, 147–155. Caprio, J., 1982. High sensitivity and specificity of olfactory and gustatory receptors of catfish to amino acids. In: Hara, T.J. (Ed.), Chemoreception in Fishes. Elsevier, Amsterdam, pp. 109–134. Caprio, J., Dudek, J., Robinson II, J.J., 1989. Electro-olfactogram and multiunit olfactory receptor responses to binary and trinary mixtures of amino acids in the channel catfish, Ictalurus punctatus. J. Gen. Physiol. 93, 245–262. ¨ Doving, K.B., Selset, R., Thommesen, G., 1980. Olfactory sensitivity to bile acids in salmonid fishes. Acta Physiol. Scand. 108, 123–131. Evans, R.E., Hara, T.J., 1985. The characteristics of the electroolfactogram (EOG): its loss and recovery following olfactory nerve section in rainbow trout (Salmo gairdneri). Brain Res. 330, 65–75. Fried, H.U., Fuss, S.H., Korsching, S.I., 2002. Selective imaging of presynaptic activity in the mouse olfactory bulb shows concentration and structure dependence of odor responses in identified glomeruli. Proc. Nat. Acad. Sci. USA 99, 3222–3227. Friedrich, R.W., Korsching, S.I., 1997. Combinatorial and chemotopic odorant coding in the zebrafish olfactory bulb visualized by optical imaging. Neuron 18, 737–752. Friedrich, R.W., Korsching, S.I., 1998. Chemotopic, combinatorial, and noncombinatorial odorant representations in the olfactory bulb revealed using a voltage-sensitive axon tracer. J. Neurosci. 18, 9977–9988. Friedrich, R.W., Laurent, G., 2001. Dynamic optimization of odor representations by slow temporal patterning of mitral cell activity. Science 291, 889–894. Fuss, S.H., Korsching, S.I., 2001. Odorant feature detection: activity mapping of structure response relationships in the zebrafish olfactory bulb. J. Neurosci. 21, 8396–8407. Gao, Q., Yuan, B., Chess, A., 2000. Convergent projections of Drosophila olfactory neurons to specific glomeruli in the antennal lobe. Nat. Neurosci. 3, 780–785.
407
Hanson, L.R., Sorensen, P.W., 2001. Single-unit recording demonstrates that pheromones are discriminated by a combination of labeled-line and population coding in the goldfish olfactory bulb. Association for Chemoreception Sciences Meeting XXIII (April 25–29), Sarasota, FL, USA. Hara, T.J., 1982. Structure–activity relationships of amino acids as olfactory stimuli. In: Hara, T.J. (Ed.), Chemoreception in Fishes. Elsevier, Amsterdam, pp. 135–157. Hara, T.J., Zhang, C., 1996. Spatial projections to the olfactory bulb of functionally distinct and randomly distributed primary neurons in salmonid fishes. Neurosci. Res. 26, 65–74. Hara, T.J., Zhang, C., 1998. Topographic bulbar projections and dual neural pathways of the primary olfactory neurons in salmonid fishes. Neuroscience 82, 301–313. Kajiya, K., Inaki, K., Tanaka, M., Haga, T., Kataoka, H., Touhara, K., 2001. Molecular bases of odor discrimination: reconstitution of olfactory receptors that recognize overlapping sets of odorants. J. Neurosci. 21, 6018–6025. Kang, J., Caprio, J., 1991. Electro-olfactogram and multiunit olfactory receptor responses to complex mixtures of amino acids in the channel catfish, Ictalurus punctatus. J. Gen. Physiol. 98, 699–721. Kang, J., Caprio, J., 1995. Electrophysiological responses of single olfactory bulb neurons to amino acids in the channel catfish, Ictalurus punctatus. J. Neurophysiol. 74, 1421–1434. Kitamura, S., Ogata, H., Takashima, F., 1994. Olfactory responses of several species of teleost to F-prostaglandins. Comp. Biochem. Physiol., Part A 107, 463–467. Laberge, F., Hara, T.J., 2003. Behavioural and electrophysiological responses to F-prostaglandins, putative spawning pheromones, in three salmonid fishes. J. Fish Biol. 62, 206–221. Laberge, F., Hara, T.J., 2003. Non-oscillatory discharges of an F-prostaglandin responsive neuron population in the olfactory bulb-telencephalon transition area in lake whitefish. Neuroscience 116, 1089–1095. Li, W., Sorensen, P.W., 1997. Highly independent olfactory receptor sites for naturally occurring bile acids in the sea lamprey, Petromyzon marinus. J. Comp. Physiol. A 180, 429–438. Little, E.E., 1981. Conditioned cardiac response to the olfactory stimuli of amino acids in the channel catfish, Ictalurus punctatus. Physiol. Behav. 27, 691–697. MacLoed, N.K., 1976. Spontaneous activity of single neurons in the olfactory bulb of the rainbow trout (Salmo gairdneri) and its modulation by olfactory stimulation with amino acids. Exp. Brain Res. 25, 267–278. Masterman, R., Hanson, L.R., Sorensen, P.W., 2001. The goldfish olfactory bulb encodes pheromone information using at least two spatial maps. Association for Chemoreception Sciences Meeting XXIII (April 25–29), Sarasota, FL, USA. Medler, K.F., Tran, H.N., Parker, J.M., Caprio, J., Bruch, R.C., 1998. Metabotropic glutamate receptor expression in olfactory receptor neurons from the channel catfish, Ictalurus punctatus. J. Neurobiol. 35, 94–104. Meredith, M., 1981. The analysis of response similarity in single neurons of the goldfish olfactory bulb using aminoacids as odor stimuli. Chem. Senses 6, 277–293. Michel, W.C., Derbidge, D.S., 1997. Evidence of distinct amino acid and bile salt receptors in the olfactory system of the zebrafish, Danio rerio. Brain Res. 764, 179–187.
408
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
Mombaerts, P., Wang, F., Dulac, C., et al., 1996. Visualizing an olfactory sensory map. Cell 87, 675–686. ´ K.H., Lower, N., Kindahl, H., 2002. The Moore, A., Olsen, role of F-series prostaglandins as reproductive priming pheromones in the brown trout (Salmo trutta). J. Fish Biol. 60, 613–624. Nikonov, A.A., Caprio, J., 2001. Electrophysiological evidence for a chemotopy of biologically relevant odors in the olfactory bulb of the channel catfish. J. Neurophysiol. 86, 1869–1876. Ohno, T., Yoshii, K., Kurihara, K., 1984. Multiple receptor types for amino acids in the carp olfactory cells revealed by quantitative cross-adaptation method. Brain Res. 310, 13–21. Ressler, K.J., Sullivan, S.L., Buck, L.B., 1994. Information coding in the olfactory system: evidence for a stereotyped and highly organized epitope map in the olfactory bulb. Cell 79, 1245–1255. Sveinsson, T., Hara, T.J., 1990. Multiple olfactory receptors for amino acids in Arctic char (Salvelinus alpinus) evidenced by cross-adaptation experiments. Comp. Biochem. Physiol., Part A 97, 289–293. Sveinsson, T., Hara, T.J., 2000. Olfactory sensitivity and specificity of Arctic char, Salvelinus alpinus, to a putative
male pheromone, prostaglandin F2a. Physiol. Behav. 69, 301–307. Thommesen, G., 1978. The spatial distribution of odour induced potentials in the olfactory bulb of char and trout (Salmonidae). Acta Physiol. Scand. 102, 205–217. Vassar, R., Chao, S.K., Sitcheran, R., Nunez, J.M., Vosshall, L.B., Axel, R., 1994. Topographic organization of sensory projections to the olfactory bulb. Cell 79, 981–991. Vosshall, L.B., Wong, A.M., Axel, R., 2000. An olfactory sensory map in the fly brain. Cell 102, 147–159. Wachowiak, M., Cohen, L.B., 2001. Representation of odorants by receptor neuron input to the mouse olfactory bulb. Neuron 32, 723–735. Wang, F., Nemes, A., Mendelsohn, M., Axel, R., 1998. Odorant receptors govern the formation of a precise topographic map. Cell 93, 47–60. Zhang, C., Hara, T.J., 1994. Multiplicity of salmonid olfactory receptors for bile acids as evidenced by cross-adaptation and ligand binding assay. Chem. Senses 19, 579. Zippel, H.P., Voigt, R., Knaust, M., Luan, Y., 1993. Spontaneous behaviour, training and discrimination training in goldfish using chemosensory stimuli. J. Comp. Physiol. A 172, 81–90.
Electrophysiological demonstration of independent olfactory receptor types and associated neuronal responses in the trout olfactory bulb ´ ´ Labergea,*, Toshiaki J. Haraa,b Frederic a
Department of Zoology, University of Manitoba, Winnipeg, Manitoba, Canada R3T 2N2 Freshwater Institute, Fisheries and Oceans Canada, Winnipeg, Manitoba, Canada R3T 2N6
b
Received 25 April 2003; received in revised form 28 October 2003; accepted 29 October 2003
Abstract The present study attempts to highlight the principles by which peripheral olfactory information of across- and withinclass odorant signals is transformed into bulbar neuron responses. For this purpose, we performed electro-olfactogram cross-adaptation and mixture experiments as well as single unit recording of olfactory bulb neurons using amino acid, bile acid and F-prostaglandin stimulants in brown and rainbow trout. The results show that amino acids, a bile acid and a F-prostaglandin activate independent receptor types. However, within the class of amino acids, different receptor types are only partially independent. Neurons responsive to bile acid and amino acids were segregated to the mid-dorsal and latero-posterior olfactory bulb, respectively. Of the 43 responsive olfactory bulb neurons studied in brown trout, 41 showed specificity for one odorant class. Olfactory bulb neurons gained responsiveness to new amino acids with increasing stimulant concentration. We conclude that different odorant classes activate specific neurons located in different regions of the trout olfactory bulb, and that information distinguishing related amino acids can be represented in a limited number of bulbar neurons with distinct response profiles under the conditions investigated. 䊚 2003 Elsevier Inc. All rights reserved. Keywords: Olfaction; Fish; Salmonidae; Olfactory bulb; Amino acid; Extracellular recording; Electro-olfactogram; Receptor; Mixture; Cross-adaptation
1. Introduction In vertebrates, the detection of odorants by the olfactory system is accomplished by receptors found on the dendritic part of sensory neurons residing in large areas of the olfactory epithelium. Research shows that primary olfactory sensory neurons converge on specific areas of the olfactory bulb glomerular layer according to the receptor type they express (Ressler et al., 1994; Vassar et al., 1994; Mombaerts et al., 1996; Wang et al., 1998). This organizational feature is not restricted *Corresponding author. Tel.: q1-49-4221-9160224; fax: q 1-49-4221-9160179. E-mail address: [email protected] (F. Laberge).
to vertebrates as demonstrated in Drosophila (Gao et al., 2000; Vosshall et al., 2000). Therefore, a spatial pattern of activity observed in the olfactory bulb upon stimulation with an odorant could represent which receptor type(s) is activated. In fish, electroencephalographic (EEG) recording of neuronal activity induced by stimulation with different odorant classes demonstrated a topography of olfactory projections to the olfactory ¨ bulb in salmonids (Thommesen, 1978; Doving et al., 1980; Hara and Zhang, 1996, 1998). A chemotopy of odor representations in the olfactory bulb has recently been demonstrated in channel catfish by using both EEG and single unit recording (Nikonov and Caprio, 2001). Responses to
1095-6433/04/$ - see front matter 䊚 2003 Elsevier Inc. All rights reserved. doi:10.1016/S1095-6433(03)00345-3
398
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
mixtures of amino acids, bile acids and nucleotides were found in different regions of the catfish olfactory bulb. These results show that the selectivity of single olfactory bulb neurons underlies the specific foci of EEG activity obtained with different odorant classes in teleosts. Imaging of primary olfactory neuron synaptic terminals in the olfactory bulb has been used successfully to localize afferents activated by different odorants in zebrafish and the mouse (Friedrich and Korsching, 1997, 1998; Fuss and Korsching, 2001; Wachowiak and Cohen, 2001; Fried et al., 2002). Different odorants show specific bulbar afferent activation patterns even if some overlap of response exists in the case of related odorants. The correspondence between these afferent maps and the activation of secondary olfactory neurons is not well studied. The fidelity of the afferent’s image will be maintained in second-order responses on the condition that intrabulbar processing is absent or minimal. Research in zebrafish shows that despite specific bulbar activation patterns for single amino acids demonstrated by imaging studies (Friedrich and Korsching, 1997, 1998; Fuss and Korsching, 2001), individual mitral cell responses are not specific to a single amino acid (Friedrich and Laurent, 2001), suggestive of a combinatorial code for amino acid quality. Friedrich and Laurent (2001) further suggested that temporal patterning of mitral cell population responses during the course of a stimulus increase the distinctiveness of amino acid representations in zebrafish. Most single olfactory bulb neurons also respond to several amino acids in other fish species (MacLoed, 1976; Bodznick, 1978; Meredith, 1981; Kang and Caprio, 1995). Recent results show that most olfactory bulb neurons are not even specific to an odorant class in goldfish (Hanson and Sorensen, 2001; Masterman et al., 2001). In the present study, single olfactory bulb neuron responses were recorded to study how they correlate with the established topography of EEG responses to odorants of different classes in salmonid fish. Additionally, neuron response characteristics within a zone of the olfactory bulb responsive to amino acids were investigated to examine if the relative independence of amino acid receptor mechanisms found in the olfactory epithelium could be maintained in the response profiles of olfactory bulb neurons.
2. Materials and methods 2.1. Fish maintenance Brown trout (Salmo trutta), 1–3 years old, and rainbow trout (Oncorhynchus mykiss) of the Langley strain (1–2 years old) were originally obtained from the Whiteshell Provincial Fish Hatchery, Manitoba, Canada. Additional rainbow trout of the Tagwerker strain, 2–4 years old, were obtained from the Rockwood Aquaculture Research Centre, Freshwater Institute. They were held in laboratory tanks at the Freshwater Institute with constant flowing aerated, dechlorinated Winnipeg city water (10.5–11.5 8C). Lighting conditions were 12 h on–12 h off with dusk and dawn simulation accomplished by low intensity light bulbs on 30 min before and after the 12 h illumination period. The fish were fed to satiation twice a week with commercial trout pellets. All experiments complied with the Canadian Council on Animal Care guidelines. 2.2. Electrophysiological recordings Fish were tranquilized by exposure to water containing MS-222 (0.5 gyl), fully anesthetized by intraperitoneal injection of amobarbital (30 mgykg body mass), and immobilized by intramuscular injection of Flaxedil (gallamine triethiodide; 3–5 mgykg body mass) before being secured on a holder in a flow-through trough. The gills were continuously perfused (0.4 lymin) with dechlorinated water. For electro-olfactogram (EOG) recording, skin and cartilage covering the right side naris were removed and the exposed naris was continuously perfused with dechlorinated water. Exposed parts of the fish were covered with wet tissue and kept moist throughout the experiment. EOGs were recorded as previously described (Evans and Hara, 1985; Sveinsson and Hara, 2000). The recording electrode position above the olfactory epithelium was adjusted so that a maximal response to a standard stimulant, 10y5 M L-serine, was achieved. Electrical signals were amplified with a DC preamplifier (Type 7P1, Grass Instruments, Quincy, MA) and recorded on a polygraph (Model 79, Grass Instruments). For single unit recording, the roof of the skull was opened, and cartilage and mesenchymal tissues removed to expose the dorsal brain from the olfactory nerves to the telencephalon. Tungsten
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
microelectrodes (impedances5 MV, World Precision Instruments, Sarasota, FL) were used. A reference electrode was placed on the dorsal skin of the snout. Electrical signals were amplified with an AC amplifier (model P511, Grass Instruments; 1y2 amplitude lower than 30 Hz and higher than 3 kHz) and fed to a slopeyheight window discriminator (Frederick Haer and Co., Bowdoinham, ME) to isolate single units. A typical recording comprised several units firing well above noise level. In order to be confident that the output of the window discriminator represented the activity of a single unit, the position of the recording electrode was adjusted until an easily isolable unit was obtained. Unit activity was displayed on an oscilloscope and spike magnitude and shape were carefully compared over time to verify that the signal of interest was from a single unit. The outputs were then recorded on a polygraph (model 79, Grass Instruments). Electrode surface location and depth were noted for every recording. A constant monitoring of neural activity was also maintained with an audio output to a speaker. 2.3. Chemical stimulation To perfuse the olfactory epithelium and deliver chemical stimuli without flow interruption, the method of Sveinsson and Hara (2000) was used. A minimal recovery period of 2 min between each 10 s stimulation was allowed. Stock solutions of test stimulants were prepared with distilled water. Stock solutions were stored at 4 8C and aliquots (10 or 100 ml) were diluted with 10 ml of dechlorinated water immediately before testing. The stimulants used were several L-amino acids (alanine, Ala; arginine, Arg; aspartic acid, Asp; cysteine, Cys; glutamine, Gln; glutamic acid, Glu; histidine, His; lysine, Lys; methionine, Met; serine, Ser; tryptophan, Trp; tyrosine, Tyr), the bile acid taurocholic acid (TCA) and prostaglandin F2a (PGF2a). PGF2a was only used with brown trout because the rainbow trout olfactory system is insensitive to this chemical (Kitamura et al., 1994; Hara and Zhang, 1998; Laberge and Hara, 2003a). The chemicals were purchased from Sigma Chemical Co (St. Louis, MO) with the exception of PGF2a, which was from Cayman Chemical (Ann Arbor, MI). Cross-adaptation experiments were performed to determine whether multiple receptor types are involved in the detection of amino acids in trout.
399
These experiments were done by replacing the constant flow of dechlorinated water with an adapting amino acid solution. The adapting solution (brown trout: 10y5 M Arg, Cys, Glu or Ser; rainbow trout: 10y3 M Arg or Cys) was perfused over the olfactory organ until the response declined and stabilized at the tonic level. Then, various test stimuli (listed above) dissolved in the adapting stimulus solution were applied at 2 min intervals (Sveinsson and Hara, 1990). Each stimulant was tested before, during and after adaptation. The adapted response to a stimulant was divided by the averaged responses before and after adaptation to express the cross-adaptation results in percentage of unadapted responses. PGF2a and TCA (10y7 M) were used in these experiments to represent additional odorant classes and compare the level of independence between across- and within-class receptor types. Binary and trinary mixture experiments were also performed with low concentrations of amino acids to supplement the cross-adaptation experiments. They consisted of EOG tests of two equipotent amino acid solutions, a mixture containing the two stimulants in the same concentration as in the individual tests, and the tests of twice the concentration of the mixture’s components as previously done in catfish (Caprio et al., 1989). The concentrations used were 10y6 M for Ser and Arg, 10y5 or 2=10y5 M for Glu, and between 10y8 and 10y7 M for Cys. Variations in concentration show that some adjustments were needed for the solutions to have equivalent EOG responses in each fish. Each fish was tested at least three times with each solution to allow the measurement of potentially small EOG amplitude differences. In the case of the trinary mixture, three equipotent solutions were used along with tests of three times of the mixture component’s concentration. The method used in Caprio et al. (1989) was followed for calculation of the mixture discrimination index (MDI) and independent component index (ICI). MDI was measured by dividing the response to the mixture by the larger response to twice the concentration of one of its components. ICI was measured by dividing the response to the mixture by the sum of the responses to its components presented individually. Theoretically, if the stimulants in the mixture activate the same olfactory receptive mechanism, the MDI equals 1. Mixture suppression would be indicated by a MDI significantly less than 1, whereas mixture enhancement
400
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
odorant classes and a water control. The amino acids were tested at 10y5 M while the bile acid and prostaglandin were tested at 10y7 M. The amino acids Cys, Arg, Ser and Glu (only tested in rainbow trout) were selected to represent stimulants within an odorant class. These amino acids were chosen to represent odorants that activate similar (Cys and Ser) or independent but overlapping receptive mechanisms (Cys and Arg) according to our EOG results. The determination of neuron response thresholds was only attempted in rainbow trout. 2.4. Statistics
Fig. 1. EOG concentration–response relationships in brown (A) and rainbow (B) trout for selected amino acids. Response magnitude is expressed as a percentage of the response to 10y5 M L-serine. Stimulants: L-cysteine (solid squares), L-serine (open squares), L-arginine (solid circles), and L-glutamic acid (open circles). Each point represents mean"S.D. The sample size is 4 in brown and 5 in rainbow trout (Tagwerker and Langley strains combined).
would be indicated by a MDI significantly greater than 1. Research on amino acid olfaction done in catfish showed no evidence of mixture suppression and concluded that the mechanism for mixture enhancement is the simultaneous activation of multiple types of receptor binding sites by the different components of the mixture (Caprio et al., 1989; Kang and Caprio, 1991). Thus, we expected to observe a mixture enhancement with amino acid pairs that activate different receptive mechanisms and an even larger mixture enhancement, as reflected by a large MDI, with the trinary mixture of amino acids if this mixture activated more receptive mechanisms than a binary mixture. An ICI equal to 1 would mean that the components of a mixture are independent. The lower the ICI value is, the more related the mixture components are. Responses of single olfactory bulb neurons were recorded to representative stimulants from three
A one-way ANOVA followed by a Dunnett post hoc test upon significance (P-0.05) was performed on the percentage of unadapted responses obtained in the cross-adaptation experiments. A significant response in the Dunnett test between the stimulation with the chemical used as adaptant and another stimulant was taken as evidence of the activation of distinct receptor sites. A twotailed paired t-test comparing the EOG response magnitudes of a binary mixture and twice the concentration of its most potent component was performed to determine if they differed. The statistical threshold was set at (P-0.05) for the ttests. The same test was applied to the mixture and three times the concentration of its most potent component for the trinary mixture. Even though most single neuron response types could be determined visually, statistical analysis proved useful when dealing with small responses. ANOVAs were calculated on single trials. For this purpose, a neuron activity record was divided into 30-s prestimulus, 10-s stimulus, and 30-s post-stimulus periods. The periods were divided into 3-s (prestimulus and post-stimulus) or 2.5-s (stimulus) time bins and spikes were counted in each time bin. The 2.5-s time bin counts were multiplied by a factor of 1.2 to make them equivalent to the 3-s time bins. Upon a significant ANOVA result (P0.05), Tukey post hoc tests were done to determine which period(s) was responsible for the significant increase or decrease in spike frequency. 3. Results 3.1. EOG responses The EOG concentration–response relationships for the four amino acids used in brown trout are
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
401
Fig. 2. EOG responses of brown trout to selected chemicals during adaptation to 10y5 M L-cysteine (A); 10y5 M L-serine (B); 10y5 M arginine (C) and 10y5 M L-glutamic acid (D). Sample size is 5 in A, B, C and 3 in D. Asterisks indicate a statistically significant difference when compared to the test of the adapting stimulus. Vertical bars represent mean"S.D.
shown in Fig. 1A. Note that Cys is very potent and that the response threshold is approximately 10y8 M for all amino acids used. These curves helped us to determine the equipotent amino acid concentrations for the mixture experiments in brown trout. The same curves for amino acids used in EOG experiments in rainbow trout are shown in Fig. 1B. No obvious differences could be observed between the EOG responses of Tagwerker and Langley rainbow trout (data not shown). The results from brown trout cross-adaptation experiments are shown in Fig. 2. Adaptation with Cys (Fig. 2A), the most potent amino acid, suppressed significantly responses to Cys and Ser, but not to PGF2a, TCA, Arg and Glu. Adaptation with Ser, Arg and Glu (Fig. 2B–D) only suppressed significantly the responses to the adaptant amino acid. The responses to the chemical classes of prostaglandins and bile acids did not differ significantly before and during adaptation to amino acids. Fig. 3 shows the results of rainbow trout cross-adaptation experiments using high concentrations (10y3 M) of Cys and Arg. Only TCA had a
significant response under Cys adaptation while the responses to Arg and Lys, although not significant, were not suppressed totally. The small response to Glu under Cys adaptation was the result of a response in only one fish and is not significant. Responses to the amino acids Trp, Tyr, Ala, Ser, Asp, Gly, Gln, His and Met were suppressed totally by Cys adaptation, while adaptation with Arg totally suppressed only responses to Arg itself, Lys and Gly. Statistical analysis showed that all the binary mixtures of the amino acids used in brown trout had greater responses than expected if the concentration of the most potent of the mixture’s component was doubled, except in the case of the Glu-Cys mixture (Ps0.07) (Table 1). The calculated values of MDI were all larger than 1, supporting the existence of mixture enhancement for the selected amino acids. Also shown on Table 1 are the values of ICI for these mixtures. The trinary mixture of Cys, Ser and Arg also showed a statistically significant mixture enhancement, along with a large MDI and a small ICI (Table 1). The binary mixture experiments conducted in rain-
402
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
Fig. 3. EOG responses of rainbow trout to selected amino acids tested at 10y5 M and TCA at 10y7 M during adaptation to 10y3 M y3 L-cysteine (A) and 10 M L-arginine (B). Sample size is indicated above the bars. Asterisks indicate a statistically significant difference when compared to the test of the adapting stimulus. Vertical bars represent mean"S.D.
bow trout yielded results comparable to those obtained in brown trout with the exceptions that the Ser-Cys mixture enhancement was not significant despite a MDI greater than 1 and the mixture of Glu and Cys produced a significant enhancement (Table 2). The amino acid pair Arg-Lys was additionally tested in rainbow trout after the crossadaptation results revealed the similarity between these two stimulants. The MDI value of 1 and a small ICI value were unique to this amino acid pair in the present study. 3.2. Single neuron recording in the olfactory bulb Of the total of 84 neurons recorded in eight brown trout, 35 neurons responded specifically to
amino acids, 6 neurons responded specifically to the bile acid TCA, 2 neurons responded to all the stimulants tested and 41 neurons did not respond to any of the stimulants tested (Table 3). Neurons responding specifically to PGF2a could not be found. A few additional neurons responded to the water control, presumably because of mechanical disruption of the flow to the naris, and are excluded from the present analysis. The response thresholds of rainbow trout olfactory bulb neurons to TCA and the amino acids Cys, Arg, Ser and Glu were determined successfully in four neurons. The response type did not change with increasing stimulant concentration. These results are listed in Table 4 along with the specificity of five additional rainbow trout neurons tested at 10y5 M only.
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408 Table 1 Results of the mixture experiments in brown trout
403
Table 2 Results of the mixture experiments in rainbow trout
Mixture (n)
MDIa
ICIb
Statistical significancec
Mixture (n)
MDIa
ICIb
Statistical significancec
Glu-Arg (4) Glu-Cys (4) Glu-Ser (4) Arg-Ser (6) Arg-Cys (6) Ser-Cys (6) Arg-Ser-Cys (5)
1.203 1.141 1.298 1.277 1.228 1.239 1.343
0.772 0.806 0.949 0.903 0.805 0.850 0.691
Yes No Yes Yes Yes Yes Yes
Glu-Arg (5) Glu-Cys (5) Glu-Ser (5) Arg-Ser (5) Arg-Cys (5) Arg-Lys (4) Ser-Cys (5)
1.336 1.220 1.297 1.266 1.224 1.011 1.180
0.856 0.887 0.815 0.849 0.894 0.646 0.792
Yes Yes Yes Yes Yes No No
a MDI, mixture discrimination index. In binary mixtures, MDIsABy(2A or 2B). In the trinary mixture, MDIsABCy (3A, 3B or 3C). b ICI, independent component index. In binary mixtures, ICIsABy(AqB). In the trinary mixture, ICIsABCy(AqBq C). c A paired t-test was performed to determine if the amplitude of the mixture response differed significantly from the response to twice (AB/2A or 2B) or thrice (ABC/3A, 3B or 3C) the concentration of its most stimulatory component.
3.3. Distribution of the amino acid and bile acid responsive neurons in brown trout For the purpose of the topographical analysis, the olfactory bulb was divided into dorsal and ventral halves at 1000 mm deep. Morphological observation showed that the maximal depth of the olfactory bulb was approximately 2000 mm. Further, these halves were divided into a rostral zone, and two medial and lateral zones at the middle and posterior levels of the bulb for a total of ten bulbar regions. Fig. 4 illustrates the general regions where olfactory bulb neurons responsive to amino acids and the bile acid TCA were found. Bile acid
a MDI, mixture discrimination index. MDIsABy(2A or 2B). b ICI, independent component index. ICIsABy(AqB). c A paired t-test was performed to determine if the amplitude of the mixture response differed significantly from the response to twice the concentration of its most stimulatory component (AB/2A or 2B).
responsive neurons were found along the dorsal midline of the mid-section of the olfactory bulb. Amino acid responsive neurons were found mainly along the dorsoventral extent of the latero-posterior bulb. Some neurons responsive to amino acids were also found in the ventral part of the medioposterior olfactory bulb and, occasionally in the middle of the bulb. Also shown in Fig. 4 are some examples of neuron responses recorded from the olfactory bulb. The proportion of neurons of the posterior olfactory bulb responding to the selected stimuli used in this study was greater in the lateral part compared to the medial part (79% of 33 neurons, lateral; 41% of 22 neurons, medial). Distribution of the electrode surface positions of single neuron recordings could not highlight differences related to cell response specificity with-
Table 3 Response specificity of recorded single neurons in brown trout Response type Cys(q) Ser(q) Cys(q) Cys(q) Ser(q) TCA(q) Arg(q) Cys(q) Ser(y) Cys(q) Ser(y) Cys(y) Ser(q) Cys(y) Ser(y) Ser(q) Arg(y) Cys(q) Ser(q) Cys(y) Ser(y) No response
Number of neurons Arg(q)
Arg(q) Arg(y)
Arg(q) TCA(q) PGF(q) Arg(y) TCA(y) PGF(y)
(q) and (y) indicates increasedydecreased firing.
13 7 6 6 2 2 1 1 1 1 1 1 1 41
404
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
Table 4 Response specificity of recorded single neurons in rainbow trout Response type Neurons with response thresholdsa Cys(q) 10y6 Ser(q) 10y5 Arg(q) 10y5 Cys(q) 10y6 Ser(q) 10y6 Arg(q) 10y5 Glu(q) 10y5 Arg(q) 10y5 Lys(q) 10y5b Arg(y) 10y7 Cys(q) 10y5 Ser(q) 5=10y5 Neurons tested at 10y5 M only Cys(q) Ser(q) Glu(q) Cys(q) Ser(q) Cys(q) Arg(q) Ser(y) Glu(y) Arg(y) (q) and (y) indicates increasedydecreased firing. a The concentrations tested were between 10y8 and 5=10y5 M. b L-lysine was only tested in this neuron.
in the amino acid responsive zones (not shown). Therefore, neurons with different response specificities were divided into five groups and graphed according to their recording depth in order to see if different layers of the olfactory bulb could be responsible for most responses of one given type. The five groups are, in the order presented in Fig. 5: Cys(q)-Ser(q)-Arg(q), all neurons showing inhibitory responses, Cys(q), Cys(q)-Ser(q) and neurons of the posterior bulb showing no response. Bile acid neurons were excluded from this analysis because they were clearly found only in the dorsal olfactory bulb. Fig. 5 shows that no clear relationship exists between recording depth distribution and neuron response profiles in the brown trout olfactory bulb. 4. Discussion 4.1. Evidence for multiple amino acid receptor types in trout Previous electrophysiological studies done in fish highlighted the presence of several partially independent receptor types involved in the detection of amino acids (Caprio, 1982; Hara, 1982; Ohno et al., 1984; Caprio et al., 1989; Sveinsson and Hara, 1990) and bile acids (Zhang and Hara, 1994; Li and Sorensen, 1997; Michel and Derbidge, 1997) in the olfactory epithelium. Our crossadaptation and mixture results strongly suggest that multiple receptor types are involved in amino acid detection in both brown and rainbow trout
without major differences between the two species. The extent to which one odorant’s response is suppressed by adaptation of the olfactory organ with a second odorant could represent a measure of independence of the receptor sites binding the two odorants. From Fig. 2A, for example, Ser and Cys would stimulate the same or similar receptor sites, while Arg would stimulate a great number of receptor sites independent of those for Cys or Ser. PGF2a or TCA would activate a receptor site group independent of all the receptors involved in amino acid detection. A large MDI value in the trinary mixture experiment with Cys, Arg and Ser compared to the smaller values of MDI obtained in binary mixtures using the same chemicals strongly supports the idea that each amino acid used in the brown trout study activates an independent receptor type. This finding suggests that different neutral amino acid receptor types exist in trout in addition to the basic and acidic types as was previously demonstrated in catfish by the use of complex amino acid mixtures (Kang and Caprio, 1991). Kang and Caprio (1991) also proposed that more than one common receptor type is involved in the detection of basic amino acids in catfish. However, the MDI value very close to 1 obtained with the binary mixture of Arg and Lys in rainbow trout provides no evidence for more than one basic amino acid receptor type in trout. 4.2. Neuronal responses in the olfactory bulb The specificity of single olfactory bulb neurons was investigated in the brown trout by testing with three different odor classes (amino acid, bile acid, prostaglandin) as well as testing three different amino acids having similar (Cys-Ser) or different (Cys-Arg) binding characteristics as shown by the EOG data. Several neurons specifically responsive to TCA or amino acids were found. They were distributed in different bulb regions resembling those found in a previous EEG investigation of salmonid bulbar responses (Hara and Zhang, 1998). The TCA responsive region is centered in the mid-section of the dorsal olfactory bulb and the amino acid responsive region in the lateral part of the posterior bulb. The amino acid and bile acid zones are not totally exclusive since a few amino acid responsive neurons were found in the TCA region and vice versa. Overall, the bulk of neurons responsive to each odorant class are segregated into different bulbar regions. The regions respon-
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
405
Fig. 4. Darkened regions show the general distribution of bile acid (A) and amino acid (B) responsive units in the brown trout olfactory bulb. The drawings represent sections of the left olfactory bulb. The olfactory nerve is to the right and the telencephalon to the left. The activity of five representative single neurons is also shown around the bulb drawings. The neuron response type is specified in a box above the traces and the bar under the traces represents the 10 s stimulus duration. The stimuli tested for each neuron are (from top to bottom) L-cysteine, L-arginine, L-serine, TCA and prostaglandin F2a.
sive to amino acids and bile acids in brown trout differ from their equivalent regions in catfish and zebrafish. In catfish and zebrafish, the amino acid responsive region is found more rostrally than in salmonids, and the presence of bile acid responses in the ventral olfactory bulb is restricted to catfish (Friedrich and Korsching, 1998; Nikonov and Caprio, 2001). Most amino acid responsive neurons showed increased firing when stimulated with 10y5 M of all three amino acids used or Cys alone, the most potent amino acid. However, complex response patterns comprising inhibition and responses restricted to Arg or Ser alone were observed. The demonstration of bulbar responses restricted to Arg or Ser confirms the mixture results that receptor types for basic (Arg) and more than one shortchain neutral (Ser and Cys) amino acids exist in
the brown trout olfactory epithelium. As an example, the observed excitatory neuron responding to Ser only cannot be explained by intra-bulbar processing because Ser has EOG characteristics close to Cys, but is less potent than Cys. Primary input to the bulb has to be the source of the specific Ser response. Not enough amino acids were tested to make the claim that these bulbar neurons are specific for an amino acid or a group of closely related amino acids, but the selectivity revealed in a limited number of neurons seems responsible for the uniqueness of related amino acid representations by olfactory bulb neurons. The determination of amino acid response thresholds in rainbow trout bulbar neurons clearly shows that olfactory bulb neurons can gain responses to new amino acids with increasing stimulant concentration. This finding could result
406
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
Fig. 5. Depth distribution of some brown trout single neurons according to their response types. Groups are: (1) Cys(q)Ser(q)-Arg(q) neurons; (2) all neurons showing inhibitory responses; (3) Cys(q) neurons; (4) Cys(q)-Ser(q) neurons; (5) neurons of the posterior bulb showing no response.
from an amino acid receptor loss of specificity with increasing stimulant concentration or the expression of more than one receptor type in each sensory neuron (Medler et al., 1998). Alternatively, the input of olfactory receptor neurons bearing different amino acid receptor types could converge on single olfactory bulb neurons because fish mitral cells innervate multiple glomeruli (Allison, 1953). The idea that higher concentration of an amino acid activates more receptor types is supported by the observation that binary mixtures of high amino acid concentrations do not produce a clear mixture enhancement in brown trout (F. Laberge, personal observation). A gradual loss of specificity in the activity patterns of olfactory bulb afferents stimulated with increasing amino acid concentrations in zebrafish also supports the above idea (Fuss and Korsching, 2001). The number of effective odorants to activate a mouse olfactory receptor expressed in a heterologous system was also shown to change with odorant concentration (Kajiya et al., 2001). These findings imply that the identity of an amino acid olfactory signal could be lost or change with increasing stimulant concentration, an issue that deserves further investigation. Unfortunately, it is not known if the high amino acid concentrations (10y5 M and higher) used in single neuron recording studies, including this one, enable us to study biologically relevant olfactory responses. Indeed, it is not known if fish can be conditioned to discriminate high concentrations of
amino acids (Zippel et al., 1993 showed that goldfish could not), with the possible exception of conditioned cardiac responses obtained in channel catfish (Little, 1981). Thus, future olfactory discrimination conditioning experiments in fish will have to take into account odor concentrations and variations in concentration. The existence of nonspecific bulbar neurons in goldfish points to species differences in olfactory coding strategies used at the bulb level (Hanson and Sorensen, 2001; Masterman et al., 2001). Thus, the finding of Friedrich and Laurent (2001) that stimulus-specific temporal patterns of firing could be used in amino acid signal optimization by zebrafish mitral cells could derive from the absence of selective bulbar neurons like those observed in brown trout. Surprisingly, neurons specific for PGF could not be found in brown trout despite robust EOG responses to this chemical class (Hara and Zhang, 1998; Laberge and Hara, 2003a). The telencephalon and preoptic area were also unsuccessfully probed for PGF responses (F. Laberge, unpublished observation). This stands in marked contrast to our recent finding of a PGF-specific neuron population in the lake whitefish olfactory system, a related salmonid fish (Laberge and Hara, 2003b). The lack of PGF response in the brown trout bulb could be explained by the existence of a small group of responsive neurons that would have been missed in the present investigation. Alternatively, the responsiveness of the PGF second-order neurons could be seasonal. The fish used here were kept on a 12 h light–12 h dark photoperiod and did not reproduce in captivity. Another possibility is that the responsiveness of PGF neurons could be activated by intra-specific interactions. More research is needed in brown trout because of the recent demonstration of the potential for PGF2a to be a reproductive pheromone acting on both sexes of that species (Laberge and Hara, 2003a; Moore et al., 2002). In conclusion, stimulants from different odorant classes stimulate specific neurons located in different regions of the trout olfactory bulb. Multiple olfactory receptor types are involved in amino acid detection in trout. Stimulant concentration could be an important determinant of the number of receptor types and olfactory bulb neurons activated by amino acids. The existence of bulbar neurons with selective response profiles to one amino acid even at high stimulant concentration, as revealed in this study, suggests that these neurons could
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
potentially be used by the trout olfactory system to encode amino acid quality. Acknowledgments This study was supported by Natural Sciences and Engineering Research Council of Canada grant OGP 0007576. F. Laberge was supported by Fonds pour la Formation de Chercheurs et l’Aide a` la ´ Recherche (Quebec) and University of Manitoba Graduate Fellowship. Brown trout were kindly provided by the West Hawk Lake Hatchery, Department of Natural Resources, Manitoba. Four anonymous reviewers provided helpful comments on the manuscript. References Allison, A.C., 1953. The morphology of the olfactory system in the vertebrates. Biol. Rev. 28, 195–244. Bodznick, D., 1978. Characterization of olfactory bulb units of sockeye salmon with behaviorally relevant stimuli. J. Comp. Physiol. A 127, 147–155. Caprio, J., 1982. High sensitivity and specificity of olfactory and gustatory receptors of catfish to amino acids. In: Hara, T.J. (Ed.), Chemoreception in Fishes. Elsevier, Amsterdam, pp. 109–134. Caprio, J., Dudek, J., Robinson II, J.J., 1989. Electro-olfactogram and multiunit olfactory receptor responses to binary and trinary mixtures of amino acids in the channel catfish, Ictalurus punctatus. J. Gen. Physiol. 93, 245–262. ¨ Doving, K.B., Selset, R., Thommesen, G., 1980. Olfactory sensitivity to bile acids in salmonid fishes. Acta Physiol. Scand. 108, 123–131. Evans, R.E., Hara, T.J., 1985. The characteristics of the electroolfactogram (EOG): its loss and recovery following olfactory nerve section in rainbow trout (Salmo gairdneri). Brain Res. 330, 65–75. Fried, H.U., Fuss, S.H., Korsching, S.I., 2002. Selective imaging of presynaptic activity in the mouse olfactory bulb shows concentration and structure dependence of odor responses in identified glomeruli. Proc. Nat. Acad. Sci. USA 99, 3222–3227. Friedrich, R.W., Korsching, S.I., 1997. Combinatorial and chemotopic odorant coding in the zebrafish olfactory bulb visualized by optical imaging. Neuron 18, 737–752. Friedrich, R.W., Korsching, S.I., 1998. Chemotopic, combinatorial, and noncombinatorial odorant representations in the olfactory bulb revealed using a voltage-sensitive axon tracer. J. Neurosci. 18, 9977–9988. Friedrich, R.W., Laurent, G., 2001. Dynamic optimization of odor representations by slow temporal patterning of mitral cell activity. Science 291, 889–894. Fuss, S.H., Korsching, S.I., 2001. Odorant feature detection: activity mapping of structure response relationships in the zebrafish olfactory bulb. J. Neurosci. 21, 8396–8407. Gao, Q., Yuan, B., Chess, A., 2000. Convergent projections of Drosophila olfactory neurons to specific glomeruli in the antennal lobe. Nat. Neurosci. 3, 780–785.
407
Hanson, L.R., Sorensen, P.W., 2001. Single-unit recording demonstrates that pheromones are discriminated by a combination of labeled-line and population coding in the goldfish olfactory bulb. Association for Chemoreception Sciences Meeting XXIII (April 25–29), Sarasota, FL, USA. Hara, T.J., 1982. Structure–activity relationships of amino acids as olfactory stimuli. In: Hara, T.J. (Ed.), Chemoreception in Fishes. Elsevier, Amsterdam, pp. 135–157. Hara, T.J., Zhang, C., 1996. Spatial projections to the olfactory bulb of functionally distinct and randomly distributed primary neurons in salmonid fishes. Neurosci. Res. 26, 65–74. Hara, T.J., Zhang, C., 1998. Topographic bulbar projections and dual neural pathways of the primary olfactory neurons in salmonid fishes. Neuroscience 82, 301–313. Kajiya, K., Inaki, K., Tanaka, M., Haga, T., Kataoka, H., Touhara, K., 2001. Molecular bases of odor discrimination: reconstitution of olfactory receptors that recognize overlapping sets of odorants. J. Neurosci. 21, 6018–6025. Kang, J., Caprio, J., 1991. Electro-olfactogram and multiunit olfactory receptor responses to complex mixtures of amino acids in the channel catfish, Ictalurus punctatus. J. Gen. Physiol. 98, 699–721. Kang, J., Caprio, J., 1995. Electrophysiological responses of single olfactory bulb neurons to amino acids in the channel catfish, Ictalurus punctatus. J. Neurophysiol. 74, 1421–1434. Kitamura, S., Ogata, H., Takashima, F., 1994. Olfactory responses of several species of teleost to F-prostaglandins. Comp. Biochem. Physiol., Part A 107, 463–467. Laberge, F., Hara, T.J., 2003. Behavioural and electrophysiological responses to F-prostaglandins, putative spawning pheromones, in three salmonid fishes. J. Fish Biol. 62, 206–221. Laberge, F., Hara, T.J., 2003. Non-oscillatory discharges of an F-prostaglandin responsive neuron population in the olfactory bulb-telencephalon transition area in lake whitefish. Neuroscience 116, 1089–1095. Li, W., Sorensen, P.W., 1997. Highly independent olfactory receptor sites for naturally occurring bile acids in the sea lamprey, Petromyzon marinus. J. Comp. Physiol. A 180, 429–438. Little, E.E., 1981. Conditioned cardiac response to the olfactory stimuli of amino acids in the channel catfish, Ictalurus punctatus. Physiol. Behav. 27, 691–697. MacLoed, N.K., 1976. Spontaneous activity of single neurons in the olfactory bulb of the rainbow trout (Salmo gairdneri) and its modulation by olfactory stimulation with amino acids. Exp. Brain Res. 25, 267–278. Masterman, R., Hanson, L.R., Sorensen, P.W., 2001. The goldfish olfactory bulb encodes pheromone information using at least two spatial maps. Association for Chemoreception Sciences Meeting XXIII (April 25–29), Sarasota, FL, USA. Medler, K.F., Tran, H.N., Parker, J.M., Caprio, J., Bruch, R.C., 1998. Metabotropic glutamate receptor expression in olfactory receptor neurons from the channel catfish, Ictalurus punctatus. J. Neurobiol. 35, 94–104. Meredith, M., 1981. The analysis of response similarity in single neurons of the goldfish olfactory bulb using aminoacids as odor stimuli. Chem. Senses 6, 277–293. Michel, W.C., Derbidge, D.S., 1997. Evidence of distinct amino acid and bile salt receptors in the olfactory system of the zebrafish, Danio rerio. Brain Res. 764, 179–187.
408
F. Laberge, T.J. Hara / Comparative Biochemistry and Physiology Part A 137 (2004) 397–408
Mombaerts, P., Wang, F., Dulac, C., et al., 1996. Visualizing an olfactory sensory map. Cell 87, 675–686. ´ K.H., Lower, N., Kindahl, H., 2002. The Moore, A., Olsen, role of F-series prostaglandins as reproductive priming pheromones in the brown trout (Salmo trutta). J. Fish Biol. 60, 613–624. Nikonov, A.A., Caprio, J., 2001. Electrophysiological evidence for a chemotopy of biologically relevant odors in the olfactory bulb of the channel catfish. J. Neurophysiol. 86, 1869–1876. Ohno, T., Yoshii, K., Kurihara, K., 1984. Multiple receptor types for amino acids in the carp olfactory cells revealed by quantitative cross-adaptation method. Brain Res. 310, 13–21. Ressler, K.J., Sullivan, S.L., Buck, L.B., 1994. Information coding in the olfactory system: evidence for a stereotyped and highly organized epitope map in the olfactory bulb. Cell 79, 1245–1255. Sveinsson, T., Hara, T.J., 1990. Multiple olfactory receptors for amino acids in Arctic char (Salvelinus alpinus) evidenced by cross-adaptation experiments. Comp. Biochem. Physiol., Part A 97, 289–293. Sveinsson, T., Hara, T.J., 2000. Olfactory sensitivity and specificity of Arctic char, Salvelinus alpinus, to a putative
male pheromone, prostaglandin F2a. Physiol. Behav. 69, 301–307. Thommesen, G., 1978. The spatial distribution of odour induced potentials in the olfactory bulb of char and trout (Salmonidae). Acta Physiol. Scand. 102, 205–217. Vassar, R., Chao, S.K., Sitcheran, R., Nunez, J.M., Vosshall, L.B., Axel, R., 1994. Topographic organization of sensory projections to the olfactory bulb. Cell 79, 981–991. Vosshall, L.B., Wong, A.M., Axel, R., 2000. An olfactory sensory map in the fly brain. Cell 102, 147–159. Wachowiak, M., Cohen, L.B., 2001. Representation of odorants by receptor neuron input to the mouse olfactory bulb. Neuron 32, 723–735. Wang, F., Nemes, A., Mendelsohn, M., Axel, R., 1998. Odorant receptors govern the formation of a precise topographic map. Cell 93, 47–60. Zhang, C., Hara, T.J., 1994. Multiplicity of salmonid olfactory receptors for bile acids as evidenced by cross-adaptation and ligand binding assay. Chem. Senses 19, 579. Zippel, H.P., Voigt, R., Knaust, M., Luan, Y., 1993. Spontaneous behaviour, training and discrimination training in goldfish using chemosensory stimuli. J. Comp. Physiol. A 172, 81–90.