J. Exp. Biol. (1966),+4, 131-148 With 13 text-figures Printed in Great Britain ASPECTS OF THE NEUROPHYSIOLOGY OF BUCCINUM UNDATUM L. (GASTROPODA) I. CENTRAL RESPONSES TO STIMULATION OF THE OSPHRADIUM BY D. F. BAILEY* AND M. S. LAVERACK Gatty Marine Laboratory, University of St Andrews, Scotland (Received 23 June 1965) INTRODUCTION The osphradium of molluscs is an organ located in the mantle cavity either on or near the ctenidium (Fretter & Graham, 1962). The function of this organ has been the subject of some speculation but it is now generally accepted to be a sense organ. This conclusion is based upon its constancy of position in the line of the inhalant current (Yonge, 1947) and the histological studies of such workers as Bernard (1890) and Stork (1935). The extensive research of Bernard showed that, besides ciliated and mucous cells, the epithelium of the 'organe de Spengel' (osphradium) in a wide variety of molluscs contains large numbers of neuro-epithelial cells. These cells are linked by multipolar nerve cells which synapse with each other at various levels. Axons of the multipolar cells then converge to form tracts and small nerves which usually run to the branchial or osphradial ganglia; the former ganglion is not present in all cases. In the cases investigated by these and other workers there does not appear to be a specific 'sensory' region of the osphradium; rather, the whole organ surface bears nerve endings. In Buccinum, however, Dakin (1912) states that 'the osphradium... attains a degree of complexity which is probably never exceeded in the Mollusca', and he describes three specialized regions of the osphradial leaflets. These are sensory, glandular and ciliated, the first being the most extensive and occupying 'the greater part of the free lateral surface of the leaflets'. Dakin also saw 'free nerve-endings' in the osphradial epithelium and considered these to be ' without doubt the important sensory structure in the organ'. Interesting observations on the layout of the innerva- tion of the osphradium have been made by means of the electron microscope (Ander- son, 1963). The sensitivity of the osphradium to chemical stimuli has been suggested many times. Copeland (1918) excised the osphradia of Busycon, and by comparison of the food-finding ability of the experimental and unoperated control animals decided that the olfactory organ(s) were located in the osphradium. Henschel (1932) and Brock (1936) also came to the same conclusion for Nassa and Buccinum respectively. More recently Wolper (1950), working on Pahidina, Brown & Noble (i960) on Bullia and Michelson (i960) on Australorbis have all favoured the chemoreceptive theory on behavioural evidence. • Present address: Pharmacological Research Department, Pfizer Ltd., Sandwich, Kent. 9-2
18
Embed
ASPECTS OF THE NEUROPHYSIOLOGY OF BUCCINUM UNDATUM … · These solutions were prepared by adding dilute hydrochloric acid or sodium hydro-xide to sea water and testing the pH of
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
J. Exp. Biol. (1966),+4, 131-148With 13 text-figuresPrinted in Great Britain
ASPECTS OF THE NEUROPHYSIOLOGY OFBUCCINUM UNDATUM L. (GASTROPODA)
I. CENTRAL RESPONSES TO STIMULATION OF THE OSPHRADIUM
BY D. F. BAILEY* AND M. S. LAVERACK
Gatty Marine Laboratory, University of St Andrews, Scotland
(Received 23 June 1965)
INTRODUCTION
The osphradium of molluscs is an organ located in the mantle cavity either on ornear the ctenidium (Fretter & Graham, 1962). The function of this organ has been thesubject of some speculation but it is now generally accepted to be a sense organ. Thisconclusion is based upon its constancy of position in the line of the inhalant current(Yonge, 1947) and the histological studies of such workers as Bernard (1890) andStork (1935). The extensive research of Bernard showed that, besides ciliated andmucous cells, the epithelium of the 'organe de Spengel' (osphradium) in a widevariety of molluscs contains large numbers of neuro-epithelial cells. These cells arelinked by multipolar nerve cells which synapse with each other at various levels.Axons of the multipolar cells then converge to form tracts and small nerves whichusually run to the branchial or osphradial ganglia; the former ganglion is not presentin all cases.
In the cases investigated by these and other workers there does not appear to be aspecific 'sensory' region of the osphradium; rather, the whole organ surface bearsnerve endings. In Buccinum, however, Dakin (1912) states that 'the osphradium...attains a degree of complexity which is probably never exceeded in the Mollusca',and he describes three specialized regions of the osphradial leaflets. These are sensory,glandular and ciliated, the first being the most extensive and occupying 'the greaterpart of the free lateral surface of the leaflets'. Dakin also saw 'free nerve-endings'in the osphradial epithelium and considered these to be ' without doubt the importantsensory structure in the organ'. Interesting observations on the layout of the innerva-tion of the osphradium have been made by means of the electron microscope (Ander-son, 1963).
The sensitivity of the osphradium to chemical stimuli has been suggested manytimes. Copeland (1918) excised the osphradia of Busycon, and by comparison of thefood-finding ability of the experimental and unoperated control animals decided thatthe olfactory organ(s) were located in the osphradium. Henschel (1932) and Brock(1936) also came to the same conclusion for Nassa and Buccinum respectively. Morerecently Wolper (1950), working on Pahidina, Brown & Noble (i960) on Bullia andMichelson (i960) on Australorbis have all favoured the chemoreceptive theory onbehavioural evidence.
• Present address: Pharmacological Research Department, Pfizer Ltd., Sandwich, Kent.9-2
132 D. F. BAILEY AND M. S. LAVERACK
Hulbert & Yonge (1937) proposed a different hypothesis since they noted that manyherbivorous and plankton-feeding gastropods also have well-developed osphradia.They postulated therefore that the osphradium acted as a sensory organ 'concernedwith the estimation of the amount of sediment carried into the mantle cavity by thewater currents created by the lateral cilia on the gills'. This view was reiterated byYonge (1947) after an extensive survey of the palHal organs of the Gastropoda. Kohn(1961) reports that Yonge has now revised this opinion somewhat and considersthat a chemosensory role for the osphradium may have evolved secondarily from theoriginal particle-detecting mechanism.
The only report of an attempt to record directly the activity of the osphradial nervecells has been the unsuccessful one by Kohn and Tateda (Kohn, 1961). The presentpaper describes neurophysiological experiments that throw further light on thefunction of the osphradium.
MATERIALS AND METHODS
The basic dissection to expose the nervous system was carried out as described inLaverack & Bailey (1963). Some attempts were made to record nervous activity in theosphradial nerves by means of platinum-wire electrodes. These were placed beneaththe nerve and then raised to lift the nerve into air for recording purposes. This tech-nique was superseded by the use of metal-filled glass microelectrodes. These wereplaced in the supra-intestinal ganglion after the ganglion had been isolated from theremainder of the CNS. The nerves to the osphradium and ctenidium which arise fromthe ganglion were left intact. The entire preparation was then removed to a specialexperimental dish (Fig. 1). The peripheral organ-bearing region of the preparationwas pinned so that the body wall rested against a partition separating the upper andlower chambers of the dish. The ganglion was then placed in a small depression in thefloor of the upper chamber in such a way that the nerves lay along a slot leading to thetop of the partition. The preparation was then covered in cool sea water, which wasused throughout as a physiological saline. All connective tissue was then carefullydissected from the ganglion, and the ctenidium was removed from the peripheralmantle area to avoid possible complicating factors due to ctendial sensory receptors.Using this procedure it was possible to keep the preparation still and, by rotating theganglion, to expose any part of its surface to the vertically mounted indium micro-electrode. Without these precautions the motility of the fragments of body wall wastoo great to allow stable penetration of the ganglion.
Immediately prior to the insertion of the electrode the exposed surface of the ganglionwas desheathed by tearing the sheath with two pairs of fine forceps, and the lowerchamber was drained of sea water via a tube inserted in the floor of the dish. Stimu-lating solutions could then be applied to the osphradium by pipette without furtherdilution, and later removed without physically disturbing the preparation. By main-taining a pool of sea water in the upper chamber a gentle creeping flow of saline alongthe nerves into the lower chamber was achieved. This obviated any chance of directstimulation of the axons by diffusion of stimulants away from the peripheral area.
The indium-filled glass microelectrodes used were prepared largely in the mannerdescribed by Gesteland, Howland, Lettvin & Pitts (1959). Tip diameters were 3-5 jiand the electrode resistance was less than 100 D. The electrodes were mounted in a
Neurophysiology of Buccinum undatum L. I 133
hydraulic advance system for insertion into the nervous tissue. The electrical activityof the nerve cells was displayed and recorded by conventional means.
Intracellular glass microelectrodes were also used in a number of experiments. Thesewere prepared from borosilicate tubing and filled with 2-9 M-KC1. An 'Amatnick'd.c. preamplifier (Bioelectric Instruments Inc.) and a high-gain d.c. oscilloscope(Telequipment) were used for display purposes.
Fig. 1. Experimental dish used when recording central nervous activity in response to stimula-tion of the osphradium, the lower figure showing preparation and recording electrode in situ.Bw., Body wall; Ga., supra-intestinal ganglion; I.EL, indifferent electrode; Lo., lower chamberof dish; Ot., osphradium; R.EL, recording electrode; Up., upper chamber of dish; W., wax.
It was early found that a sea-water extract of Mytilus was an effective stimulus andthis was subsequently used as a test stimulus for all experiments. The extract wasobtained by crushing 8-12 g. of whole Mytilus in 50 ml. of sea water and allowing theresultant mixture to stand for a least 30 min. The turbid supernatant solution was thendrawn off as required. Chemical stimulation of the preparation was accomplished bysolutions of known synthetic chemicals dissolved in filtered sea water. Stock solutionsof the various chemicals (B.D.H. ' Analar' reagents) were made up at io - 1 or io~2 Mconcentrations and stored in a refrigerator. These solutions were replaced at frequentintervals. Dilutions of the stock solutions were made up as necessary, and the pH wasadjusted to 7-5 immediately prior to use.
The chemical test solutions were applied to the osphradium by means of glassdropping pipettes of 1-2 ml. capacity. In all cases the solutions were spread as evenlyas possible over the whole osphradium. The chemicals used as stimuli in the course ofthe experiment were as follows: adipic acid, /?-alanine, /-aspartic acid, betaine hydro-
Other methods used to indicate the variety of effective stimuli on the osphradiumwere as follows:
(1) Touch—using a paint-brush.(2) Carborundum particles of the following sizes suspended in sea water: less than
20 /i, 20-200 fi, and greater than 200 fi in diameter. The graded particle sizes wereobtained by sieving carborundum powders through 20 fi and 200 /i-mesh sieves.
(3) Sea water adjusted to pH values of 10-05, 9"4> 8-5, 7-4, 6-8, 4-9, 2-3 and 1-4.These solutions were prepared by adding dilute hydrochloric acid or sodium hydro-xide to sea water and testing the pH of small samples with a Pye 'Dynacap' pH meter.
(4) Sea water, adjusted to saline concentrations equivalent to 200, 160, 133, 115,75, 50 and 25% sea water, and also distilled water, were used as test solutions. Thesaline solutions were prepared from known volumes of sea water by evaporation or bydilution with distilled water as necessary.
(5) Sea water to which 86-28 g./l. of mannitol had been added, producing a solutionwith an osmotic pressure approximately equivalent to 150% sea water.
The onset and duration of application of stimuli were indicated by means of amanually operated switch circuit producing an interruption of a time-marker pulsefed into the lower beam amplifier. In the ca9e of all stimuli applied in sea water theactive application of the stimulus was marked on the records and the preparation wasleft for some seconds, or, if a response was observed, until the response ceased. Thestimulating solution was then washed away with filtered sea water. The washing wascarried out in the same way as the stimulation and, besides restoring the preparationto normal, served as a useful control in cases where the osphradium proved sensitiveto mechanical distortion as well as to the chemical stimulus.
RESULTS
Efferent activity in the ospkradial nerves
In the majority of preparations spontaneous activity was seen in the axons of theosphradial nerves (see Fig. 2 A), using platinum wire electrodes. In each preparationattempts were made to record nervous responses to applications of the followingstimuli to the osphradium: (1) touch, (2) flowing sea water, (3) particles of carborundumin sea water, (4) chemicals in sea water.
All attempts to record afferent activity in the nerves in response to the above stimulifailed (see Bailey & Laverack, 1963). The spontaneous activity described above andshown in Fig. 2 A was found to be entirely of an efferent nature, since cutting theosphradial nerves proximal to the recording electrodes abolished the activity in allcases, whilst cutting the nerves peripheral to the electrodes usually left the activityundiminished.
In some cases the efferent activity in the intact nerves should be seen to increase or,less frequently, activity could be invoked in a 'silent' nerve when certain stimuli wereapplied to the osphradium. In all cases the stimuli which elicited such responses
Neurophysiology o/Buccinum undatum L. I 135
were of an extremely acidic nature (e.g. unbuffered io~2 M betaine hydrochloride insea water) and the response lasted for several minutes. A response of this type is shownin Fig. 2B and C. The experiment was repeatable if the stimulating solution waswashed off the osphradium within 10 sec., but prolonged exposure to the stimulus ledto refractoriness of the preparation.
Behavioural experiments upon whole animals showed that such acidic materialsinduced strong retraction responses when they were introduced into sea water nearthe animal. This suggests that the efferent activity observed in the osphradial nervesin vitro was either of a protective nature or, more probably, was produced as a resultof the arrival of injury discharges in the afferent sensory neurones at the ganglionicsynapses.
Fig. 2. Efferent activity recorded from the osphradial nerves. A, Background activity; B andC, ' acid' response recorded from a previously ' silent' nerve, the traces are continuous. Time-marks indicate i sec., upper for A only and lower for B and C.
The origins of the reflex efferent activity described above appear to reside in thesupra-intestinal ganglion, since the removal of the remainder of the CNS failed toabolish, or even to cause a visible reduction in, the observed response (Bailey, inpress).
The experiments utilizing platinum wire electrodes therefore failed to show anysensory activity in the osphradial nerves. Afferent fibres must be present in the nerves,however, since reflex efferent activity could be observed in response to noxiousstimuli when all other afferent routes were severed. These results show that theosphradial nerves are of the ' mixed' type and suggest that the afferent pathways areaxons of small diameter. Electron microscope studies have shown that the largestaxons in these nerves are only 0-5-1 -o /i in diameter, and the smallest range down to0-2 ji in diameter (Bailey, 1964).
In the hope of obtaining records of the afferent activity from these small fibresattempts were made to split the osphradial nerves into smaller bundles for furtherexperiments with platinum wire electrodes. This technique failed because of thefragility of the nerve fibres after the integrity of the nerve sheath was interrupted.
136 D. F. BAILEY AND M. S. LAVERACK
Spontaneous activity of neurones in the supra-intestinal ganglion
Penetrations of the supra-intestinal ganglion with indium electrodes establishedthat active neurones were common in the isolated ganglion. In the majority of casesthe background discharge was of a relatively regular frequency (see Fig. 3A). Suchactivity was frequently modifiable in response to stimulation of the osphradium, aswill be shown below.
In a small number of preparations on-going activity of a different type was observed.This took the form of periodic bursts of spikes from small groups of neurones in theabsence of overt stimulation. This type of result is shown in Fig. 3B. The similarity ofthe patterns of activity and their regularity of occurrence suggest that some pacemakerneurones may be involved. In two of the preparations exhibiting such activity it waspossible to modify the bursts by chemically or mechanically stimulating the osphradium(see Fig. 4), but in others the periodicity and pattern remained unchanged.
Fig. 3. On-going activity of neurons in the supra-intestinal ganglion. A, Regular dischargefrom several neurones. Time-marker frequency io/sec. B, periodic bursts of spikes fromseveral neurones. Time-marker frequency, 1 /sec.
t \Fig. 4. Effects of mechanical stimulation of the osphradium. Periodicity is interrupted, avolley of potentials occurs, and then a quiet period which lasts for up to 30 sec. after the endof the stimulus. Arrows indicate onset and end of stimulation. Time-marker frequency, 1 /sec.
Central nervous responses in the supra-intestinal ganglion
These observations were made in order to obtain information concerning the sen-sory modalities of the osphradial receptors.
(1) Responses to particles in suspension in sea water
In order to test the theory of osphradial function proposed by Hulbert & Yonge(1937) and Yonge (1947) sea water containing suspended carborundum particles ofknown size was used as a stimulus. The smallest particles in the first range ( < 20 ft)would remain in suspension for many minutes but the largest settled out after 5-30 sec.Similarly the smallest particles of the middle range (20-200 ft) remained in suspension
Neurophysiology of Buccinum undatum L. I 137
for some few seconds whilst the larger particles settled rapidly. The particles of thelast range (> 200 fi) settled almost immediately, the largest particles being of a similarsize to large sand grains.
Stimulation of the osphradium with any of the above failed to elicit a central responseon any occasion. The smallest and middle-range particles are of a size which couldeasily be swept into suspension by wave action, whilst the largest particles shouldrepresent a considerably supramaximal stimulus in view of their rapid settling pro-perties. It is therefore possible to rule out sensitivity to sediments in suspension as asensory modality of the receptors of the osphradium of Buccinum, unless there be veryfew neurones involved which possibly may not have been monitored.
(2) Responses to variations in pH
Sea water to various known pH values between 1-4 and 10-05 were used as stimulantson a number of occasions. Central responses to such stimuli were only elicited by thosesolutions having pH values of less than 4-0 or greater than 9-0. The normal pH of seawater is in the region of 7-8 and the buffering effects of dissolved bicarbonate are wellknown. Local pH variations, e.g. near decomposing organic material in still water,are possible, but it seems unlikely that such high and low values as those necessary toelicit responses experimentally would be reached. The responses obtained in theseexperiments closely resembled those illustrated in Fig. 2B and C, for the 'acid'responses in the osphradial nerves and repeated or prolonged stimulation with suchsolutions again caused an irreversible diminution or loss of the central responses.
(3) Responses to ionic and/or osmotic concentration
To test the osphradium for sensitivity to ionic or osmotic variations in the surround-ing medium sea water was concentrated or diluted to produce a range of solutions from200% sea water to distilled water. Central responses to the applications of suchsolutions to the osphradium were limited to the extreme conditions of the upper andlower limits. Responses were only seen in cases where 200% sea water or dilutionsof 50 % or less were used. Both these values can be considered as unlikely in the normalenvironment, though the latter might be attained in estuarine regions. Generally,however, Buccinum lives below E.L.W.M. and would not be expected to encounter suchvariations. The osphradial receptors do not appear to be sensitive to an increase in theosmotic concentration alone since a stimulus solution of mannitol in sea water osmo-tically equivalent to 150% sea water also failed to elicit any central response.
(4) Responses to tactile and chemical stimuli
(i) Responses to touch and 'extracts'
The only stimuli which consistently elicited central responses in the supra-intestinalganglion when they were applied to the osphradium were tactile and chemical. Theformer had to be of sufficient magnitude to move and distort the osphradium and in-volved forces greater than those exerted by the suspended particles which failed toinvoke any central responses. The effective chemical stimuli were sea-water extractsof the mussel, Mytilus edulis, similar extracts of the visceral hump of Buccinum undatumand neutral sea-water solutions of some synthetic chemicals.
D. F. BAILEY AND M. S. LAVERACK
The 'Mytilus extract' was tested on active specimens of Buccinum in tanks andproved effective in provoking the proboscis extension response which is typical of theobserved feeding behaviour of this species. This mussel extract was therefore usedas a test stimulus in all the following electrophysiological experiments so that abehaviourally potent stimulus might be compared with the effects of various syntheticchemicals.
c
1 1 «
r i
1 1 1
Fig. 5. Responses of a central neurone in the supra-intestinal ganglion. A, Response totouching the osphradium; B, control application of sea water to the osphradium illustratingthe absence of response to water flow; C, prolonged response to 'Mytilus extract', the tracesbeing continuous. Time-marker frequency, io/sec. The interruptions in the time-markerindicate the duration of the stimuli in A and B. In C the same feature at the beginning of thetrace represents the active application of the stimulus, whilst the gaps at the end of the recordindicate washing the osphradium with sea water.
Typical records of the nervous responses to the tactile and 'extract' stimuli de-scribed above are shown in Fig. 5. In this case only one unit is visible, a situation notencountered in the majority of the recordings made with indium microelectrodes.The most striking difference between the two responses is their duration. Fig. 6 showstime/frequency graphs of the responses shown in Fig. 5 with which many of thefollowing results may be compared. The response to touch was short and was notprolonged beyond the active application of the stimulus. The 'Mytilus response' onthe other hand had a long time-course, but adapted slowly even when the extractremained in contact with the osphradium. Both these responses can be compared witha control application of sea water which elicited no response and shows that simplewater flow over the osphradium is not a natural stimulus to osphradial receptors.Jets of water of sufficient force to distort the osphradium could, however, elicit a briefresponse, but by exercising some care during' washing' these responses were abolished.
Nervous activity similar to the 'Mytilus response' could also be recorded fromcentral neurones following stimulation of the osphradium by sea-water extracts of thevisceral hump of both male and female Buccinum (see Fig. 7H). No difference between
Neurophysiology of Buccinum undatum L. I 139
the responses to extracts from male and female was discernible, suggesting that theosphradium is not differentially sensitive to any male or female sex-attractants whichmight conceivably be produced in this region.
The origin of the afferent activity leading to the observed central responses to touchand to the various extracts appears to differ. In several experiments the osphradiumwas carefully removed from the mantle after the various stimuli had been applied andthe results recorded. The stimuli were then reapplied and whilst the response to extractsand synthetic chemicals was abolished, the response to touch was generally unchanged.It would appear therefore that the majority of the mechanoreceptors are situated in themantle musculature underlying the osphradium. Because of their position it is likelythat they comprise the same types of receptor as have already been described (Laverack& Bailey, 1963).
10
8
£4 6
req
fa4
2
0
!-
-
A
- •;"' j
fil l '- 5 '!
[I
- !
-1st,
-B
-
-
-
ft1\1
HAAA/AM AT^ Av v W V V - ^ V V N ̂ / \
10 0 10 20 30
Time (sec.)
50 60 70
Fig. 6. Graph* of the activity shown in Fig. 5. The frequency is calculated by the method ofPringle & Wil»on (1952). A, Response to touching the osphradium; B, response to 'Mytilusextract'. St., Application of the stimulus.
(ii) Responses to synthetic chemicals
Neurones in the supra-intestinal ganglion were found to respond after osphradialstimulation by neutral sea-water solutions of the following chemicals: io"3 M/-glutamic acid, io~3 M /-aspartic acid, io"3 M adipic acid, io~a M glutaric acid, io~2 Msuccinic acid, io~2 M malonic acid, io~a M betaine hydrochloride, io~2 M trimethyl-amine oxide hydrochloride. On one occasion only, weak responses were also obtainedusing io~3 M /-glutamine, and a saturated solution of glutathione.
Each type of response, excepting the last two, was confirmed several times in eachpreparation in which it occurred and proved to be regularly repeatable in differentpreparations. Single-unit preparations showing these responses were not obtained inthose experiments in which indium microelectrodes were used. Because of this poordiscrimination, the large number of units responding, and the inherent variation ofthe size of the impulses from single units, it was found impossible to plot time/frequency graphs of the responses of individual neurones to the above stimuli. The
140 D. F. BAILEY AND M. S. LAVERACK
graphs of the nervous activity which are included below are therefore indications ofthe total nervous activity involved in each response. The threshold concentrations ofthe various stimuli were also difficult to determine accurately because of the lengthyexperimental procedure and the condition of recording centrally rather than at theprimary receptor. In each preparation at least thirty successive stimuli were applied
30
20
10
1 = 0
20
10
St.
Hi
£
St.
- 5 0 10 20Time (sec.)
I30
30
20
10
0
30
20
10
0
-
J•i r
I
•-V
K\St.
V
St.
(ii)
n i
w.w.1 . • • 1
B
i i i
- 5 0 10 20Time (sec)
30
Fig. 7. (i) Graphs of multi-unit responses from supra-intestinal ganglion neurones elicited by' Mytilus extract' and glutamic acid. Total number of impulses for each second are plottedagainst time and the stimuli (St.) and washing (W.) are indicated by the solid bars at thebeginning and end of the response. A, response to ' Mytilus extract'; B, response to glutamicacid ( IO- 'M.) . (ii) Effects of extracts of the visceral hump of Buccmum placed upon the osphra-dium of a male animal: A, response to male 'visceral hump extract'; B, response to female' visceral hump extract'.
and this, together with the time taken up by washing solutions away from the osphra-dium and allowing time for recovery after each stimulus, led to a total experimentaltime of at least 2 hr. During this time the electrode had to remain in the same positionwithin the ganglion. Such prolonged conditions of stability were seldom achieved onaccount of the movements of the preparation which arose for a variety of reasons. Ontwo occasions, however, virtually complete sets of data were obtained and on thebasis of these, and the confirmation obtained from many other more fragmentaryresults, it is possible to give a reasonably precise account of the various excitatoryresponses of the supra-intestinal ganglion cells.
The glutamic acid response. Fig. j(i) shows a central response to io~8 M /-glutamicacid placed on the osphradium, compared with the 'Mytilus extract' response fromthe same preparation. Both the pattern and the time-course of the response are similarin both cases. The glutamic acid response thus closely mimics the action of 'Mytilusextract' and would appear to be its synthetic equivalent for all intents and purposes.The threshold for the central response to glutamic acid is approximately 5 x io"6 M.
Neurophysiology of Buccinum undatum L. I 141
The aspartic acid response. This type of response is shown in Fig. 8, where a io~3 Msolution was used. The duration of the response is similar to that for 'Mytilus extract'but there is a more gradual initial increase in frequency. The threshold for the responsewould appear to be lower than that for glutamic acid, however, as a io"6 M solutionproduced a just noticeable increase in activity, whilst a io~7 M solution did not.
30 1-
20
S.•asD.o
10
10 4020 30
Time (sec.)
Fig. 8. Graph of the 'aspartic acid response' plotted as in Fig. 7.
50
30
|_ 20n
4)
'a" 10
°Q_- 5
30 r
8 20 -
0 10 20 30
Time (sec.)
0 10 20
Time (sec.)
30
30
H 20
i- 5 0 10 20
Time (sec.)
30
20
10
30
- 5 0 10 20 30
Time (sec.)
Fig. 9. Effects of addition of (A) adipic acid, (B) glutaric acid, (C) succinic acid,and (D) malonic acid, plotted as in Fig. 7.
142 D. F. BAILEY AND M. S. LAVERACK
The adipic acid response. The response as shown in Fig. 9 A was elicited by a io"8 Msolution. The maximal frequency of the response is lower and attained more slowlythan in the response to 'Mytilus extract'. The response is also of a noticeably shorterduration and the threshold, approximately 5 x io"4 M, is higher than for glutamicacid.
The ghitaric acid response. Glutaric acid stimulation is followed by a short volley ofspikes (see Fig. 9B). The threshold concentration was of the order of io~2 M.
The succimc acid response. The characteristic curve of activity shows low maximalfrequency, a slow rate of increase and a short duration compared with the equivalent'Mytilus' response (see Fig. 9C). Threshold concentration was greater than io~3 M.
Fig. 10. Response of a single neurone in the supra-intestinal ganglion to betaine (IO~'M)applied to the osphradium. A-D are continuous traces. The bar in A represents the activeapplication of the stimulus. Time-mark in D represents i sec.
The malonic acid response. A io~2 M solution of malonic acid was an effectivestimulus in the case shown in Fig. 9D, but was ineffective in three other experiments.
The betaine response. Central nervous responses to io"3 M betaine hydrochloride insea water were observed in four experiments, but the same solutions proved in-effective in an equal number of cases where central responses to other chemicalstimuli had been seen. When present, the response was generally similar to that for'Mytilus extract' or glutamic acid, though the peak frequency was lower and theduration somewhat shorter. Fig. 10 shows one example of a betaine response wherea single unit was recorded.
The trimethylamine oxide response. The central responses to io~2 M TMO were veryweak in several cases but in others produced a result like that of betaine. The responsewas more consistent in its occurrence than that for betaine and only on one occasionin eight trials was there no visible response in a preparation which responded to'Mytilus extract' and glutamic acid. The TMO threshold concentration was betweenio~2 and io~3 M.
The l-glutamine and glutathione responses. A solution of io~s M /-glutamine provedan effective stimulus on only one occasion, in spite of frequent trials in the course ofexperiments upon preparations which responded to other chemical stimuli.
Neurophysiology of Buccinum undatum L. I 143
Glutathione was used only once as a stimulus since it was almost insoluble in coldsea water, though a saturated solution did elicit a weak central response on thisoccasion.
Microelectrode studies
In experiments utilizing indium microelectrodes considerable numbers of unitswere monitored in any particular investigation. Further studies were made usingintracellular saline-filled microelectrodes in order to ascertain the response of indi-vidual central neurones to osphradial stimulation.
•
Fig. 11. This neurone, initially inactive, responded to 'Mytilus extract' but not to tactilestimulation of the osphradium. Resting potential 35 mV; vertical bar represents 20 MV. Thedepression of the lower beam indicates the active application of the stimulus. Time-markerfrequency is i/sec.
L
Fig. 12. Intracellular recording from supra-intestinal ganglion cell whose initial activity waginhibited by both tactile (A) and chemical (B,' Mytilus extract') stimulation of the osphradium.The two parts of the trace in B are continuous. Resting potential, 50 mV. The vertical barrepresents 20 mV and the horizontal bar 1 sec. The lower beam depressions represent the dura-tion of the stimulus in A, and both stimulus (first) and washing (second and third) in B.
Particular central neurones often showed differing responses to peripheral stimula-tion by different modalities of stimulus. The same stimuli could also produce differentresults in other central neurones. Figures 11,12 and 13 are representative of the resultsobtained. Figure 11 is a record from an inactive cell that underwent depolarizationleading to propagated spikes sometime after the osphradium was flooded with 'Myti-lus extract'. This same cell was unresponsive to mechanical stimulation of the mantleregion.
Different central responses are illustrated in Fig. 12, in which a previously dis-charging cell (basal frequency about 2/sec.) was inhibited by both mechanical andchemical stimulation of the osphradium. The two types of peripheral input alsoapparently converge on a third type of central c«fl whose responses are shown in
D. F. BAILEY AND M. S. LAVERACK
Fig. 13. In this case the background spike discharge of about 3/sec. was inhibited bjmechanical but stimulated by chemical stimulation of the osphradium.
A tabular representation of the various types of central neurone which can exist inrespect of mechanical and chemical stimulation is shown in Table 1. This indicatesthe variety of effects that can occur at a central neurone in response to peripheral
Fig. 13. Recording from an active supra-intestinal ganglion cell. A, Inhibition caused by tactilestimulation of the osphradium; B, shows the excitation caused by 'Mytilus extract' applied tothe same organ. Resting potential, 30 mV. Lower beam depressions, as in Fig. 10; verticalbar represents 10 mV; time-marker frequency is i/sec.
Table 1. Combinations of possible input to active and inactive central neurones
( + , Events actually demonstrated; —, events possible but not so far recorded.)
A. Active neuroneTactile input
Combination ... 1Excitation +No effectInhibition
Chemosensory inputExcitation +No effectInhibition
8
B. Inactive neuroneTactile input
CombinationExcitationNo effect
Chemosensory inputExcitationNo effect
1
+
stimulation by these two types of stimulation. It also shows that of the possible com-binations over hah0 have been demonstrated. If the central unit be actively dischargingthen peripheral input may excite it further, have no effect or inhibit the prevalentactivity. Of these possibilities it has been shown that both tactile and chemical stimulimay inhibit (singly or in combination); that both may excite; or, in the case of chemo-sensory input (Table 1 A), may have no effect. In the case of a 'silent' neurone thereare four possibilities and of these three have been seen (Table 1B).
DISCUSSION
Investigation of central nervous activity in the supra-intestinal ganglion of Buccinumhas demonstrated conclusively the chemoreceptive function of the osphradium inthis species and the absence of any central activity caused by mechanoreceptors of asufficient sensitivity for the monitoring of inhaled sediments.
Neurophysiology of Bucciiium undatum L. I 145
Central nervous responses to stimulation of the osphradium
All attempts to record afferent nervous impulses in the osphradial nerves failed.The experiments did, however, show that efferent bursts of activity could be initiatedby noxious stimuli. The prolonged nature of the efferent bursts and their irreversibleloss with prolonged stimulation suggest that this activity was provoked by injurydischarges in the osphradial receptors which were henceforward refractory. Theresults indicate that there are interneurones or motor neurones passing from the CNSto the osphradium via the osphradial ganglion and thus this organ should also beconsidered as an effector of some kind. The function of the nervous activity directedtoward the osphradium is by no means clear, but observations of the organ duringmechanical stimulation showed that local withdrawal responses took place. Noxiousstimuli also caused contractions in the mantle region and it is probable that theseresulted from the observed reflex efferent activity in the osphradial nerves and con-stitute protective responses.
The results of the experiments carried out on central responses to varied stimulationof the osphradium represent the first successful electrophysiological investigation ofosphradial function in relation to chemoreception in gastropods. The results have thedisadvantage that they are not the responses of the receptors or their afferent axonsbut the information which has been obtained is, nevertheless, useful in filling theconsiderable gap in our knowledge of this subject.
The chemosensory abilities of gastropods, especially the marine forms, have recentlybeen reviewed by Kohn (1961) and behavioural responses to sea water and inorganicions, predators and distant food are well documented. Mating and homing are furtheraspects in which chemoreception may have some part to play. Apparently the onlyprevious attempt to obtain electrophysiological data on chemoreception in gastropodswould appear to be that of Kohn and Tateda (see Kohn, 1961), though other un-successful attempts may not have been published (Duncan, personal communication).
The experimental data indicate an osphradial sensitivity to gross mechanicaldistortion, extremes of pH and ionic/osmotic conditions and to various syntheticorganic chemicals and animal extracts. Mechanical stimulation of the type mentionedabove is not mimicked by stimulation with particles. Thus it seems unlikely that sedi-ments swept into the mantle cavity via the siphon are effective in stimulating theosphradium. The copious mucous secretions of the hypobranchial gland also serve toremove particles from the contents of the mantle cavity.
The osphradium also appears insensitive to variations in pH over a wide range ofvalues above and below that of normal sea water. The observed responses to extremevalues of pH are most probably caused only by injury discharges in the osphradialreceptors.
Small variations in the overall ionic and osmotic concentrations of the environmenthave not been followed by any recordable central activity. There is a possibility thatthe animal can detect areas where the sea water has been greatly diluted, as repeatablecentral nervous responses were elicited experimentally using such stimuli. Whethersuch activity is meaningful or merely represents injury discharges, analogous to thosecaused by extremes of pH, is difficult to determine with certainty. Some marinemolluscs are, however, tolerant to fresh water to a remarkable degree and fresh-water
10 Exp. Biol. 44, 1
146 D. F. BAILEY AND M. S. LAVERACK
turgor has been used as an aid to dissection in some cases (Krijgsman & Brown, i960).This would suggest that diluted sea water or fresh water are not very harmful to suchanimals. It has been noted in the present experiments that preparations from Bucdnumappear to suffer no ill effects from exposure to distilled water. This tolerance andthe fact that the experimentally induced nervous activity could be repeated withoutdiminution of the response suggests that such responses were not caused by injurydischarges in the receptors. Thus they may have some significance to the animal andtherefore be a means for the detection of areas of low salinity. Behavioural ex-periments using choice-chamber techniques might throw some light on this aspectsince the overt responses of the gnimai to such factors could then be studied.
The most illuminating results obtained in the course of this work are those showingcentral activity concomitant with osphradial stimulation by 'mussel extract' and thesimilar effects of synthetic chemicals, notable amino- and dicarboxylic acids. DamagedMytihis edulis is a readily acceptable food for Bucdnum and its attractiveness to thepredator over some distance is easily demonstrated. The osphradium is certainly in-volved in this response (Brock, 1936), though chemoreceptors elsewhere on the body,e.g. rhinophores and general body wall, may also be involved. The osphradium, con-cealed within the mantle cavity, is not involved in contact chemoreception because ofits position and must therefore be considered as a seat of olfactory sense.
Results showing that the 'natural responses' of central neurones could be evoked bysolutions of certain synthetic chemicals are revealing in two respects. First, theyindicate the possible nature of the effective constituent of the animal-extract stimuli,though this can only be confirmed by fractionation of the latter and subsequent testingon both the whole animal and the osphradial preparation. Secondly, the results suggesta relationship between chemical structure and effective stimulation of the osphradialchemoreceptors.
The most effective chemical stimuli were glutamic and aspartic acids. Both werecapable of eliciting responses at low concentrations, which is compatible with the theoryof olfaction. It also is worth noting in respect of the sensitivity of Bucdnum to ' Mytihisextract' that Mytihis muscle has been shown to contain both glutamic and asparticacids in free form (Potts, 1958). The occurrence of these chemicals in the fresh ex-tract is thus virtually certain. Responses to amino acids have been noted in a numberof marine animals, e.g. Cardnides (Case & Gwilliam, 1961, 1963; Case, Gwilliam &Hanson, i960), Nerds (Case & Gwilliam, 1936), Limuhis (Barber, 1956, 1963),Nassarius (Henschel, 1932) and Panulirus (Laverack, 1964). Amino acid chemo-receptors may consequently be a common feature amongst marine invertebrates.
The relationship between chemical structure and effective stimulation
It is clear that glutamic and aspartic acids are the synthetic compounds whichmimic most closely the substantial, and perhaps more natural, responses to 'Mytihisextract'. The dicarboxylic acids on the other hand are rather less effective and higherthresholds are demonstrated. The majority of the dicarboxylic acids tested werenevertheless found to be effective stimuli. Adipic and glutaric acids were those whichwere most stimulatory and also those which are structurally most similar to glutamicand aspartic acids. Figure 14 illustrates the structural formulae of those chemicalsfound to be adequate stimuli for the osphradial chemoreceptors. It may therefore be
Neurophysiobgy of Buccinum undatum L. I 147
postulated that the requirement for stimulation of the osphradial chemoreceptors is astraight chain compound of four to six carbon atoms, bearing terminal carboxylgroups. The smaller dicarboxylic acids and the long chain of glutathione are verymuch less effective as stimuli.
The observations involving betaine, TMO and the solitary response obtained withglutamine are not compatible with the hypothesis stated above. It is, however, possiblethat more than one type of chemoreceptor is present in the osphradial epithelium and,whilst those responding to the dicarboxylic amino-acids predominate, there are otherswith different properties (see Laverack, 1963). The high thresholds for the three com-pounds, which are similar to the less effective short-chain dicarboxylic acids, indicatea lack of sensitivity in the latter receptors suggesting a non-specific site of action.
Fig. 14. Structural chemical formulae of the effective and partially effective syntheticchemical stimuli employed during the experiments upon osphradial sensitivity.
Central neurone variability
The reactions of central neurones of the supra-intestinal ganglion to varied stimula-tion of the osphradial region are diverse and complex. Table I shows that eightdifferent types of response have been demonstrated and others might well be found ina more comprehensive study. Considerable integration must occur in the supra-intestinal ganglion. The activity of the peripheral receptors may have both excitatoryand inhibitory effects on central neurones. However, without a far greater number ofresults of this topic, it is impossible to show whether there is any relationship betweenany specific type of peripheral receptor and a particular type of central reaction.
SUMMARY
1. Recordings have been made from various central neurones of Buccinum unda-tum, the common whelk, particularly in response to peripheral stimulation of theosphradium.
2. These central responses show that the osphradium is sensitive to the applicationof a restricted range of chemical stimuli, but insensitive to mechanical stimulationwith suspended particulate matter.
3. The most pronounced central activity was seen in response to the addition of asea-water extract of Mytilus edulis. This animal is one normally preyed upon byBuccinum. The synthetic chemicals most closely mimicking this effect were glutamic
148 D. F. BAILEY AND M. S. LAVERACK
and aspartic acids, which were effective at low-threshold concentrations. Otherchemicals of 4-6 carbon-atom chains with terminal carboxyl groups were effectivestimuli and it therefore appears that the dicarboxylic nature of the effective aminoacids plays a major part in the stimulatory process.
4. Intracellular recordings indicate that the peripheral input from osphradialchemoreceptors and mantle mechanoreceptors may be either excitatory, inhibitory orwithout effect on central neurones.
This work was carried out whilst one of the authors (D.F.B.) held a D.S.I.R.Studentship. Part of the equipment was obtained on a Royal Society Grant-in-Aidto M.S.L.
REFERENCES
ANDERSON, E. (1963). Cellular and subcellular organization of the osphradium in Busycon. Proc.16th Int. Cong. Zool. 3, 280.
BAILBY, D. F. (1064). Ph.D. Thesis, University of St Andrews.BAILEY, D. F. & LAVERACK, M. S. (1963). Central responses to chemical stimulation of a gastropod
osphradium. Nature, Land., 300, 1122—3.BARBER, S. B. (1956). Chemoreception and proprioception in Limulus. J. Exp. Zool. 131, 51^73.BARBER, S. B. (1963). Properties of Limulus chemoreceptors. Proc. 16th Int. Cong. Zool. 3, 76.BERNARD, F. (1890). Recherches sur les organes palleaux des gasKropodes prosobranches. Ann. Set.
nat. Zool. Ser. 7, 9, 88-404.BROCK, F. (1936). Suche, Aufnahme und enzymatische Spaltung der Nahrung durch die Wellhom-
schnecke, Buccinum undatum L. Zoologica, Stuttgart, 34 (92), 1—136.BROWN, A. C. & NOBLE, R. G. (i960). Functions of the osphradium in Bullia (Gastropoda). Nature,
Lond., 188, 1045.CASE, J. & GWILLIAM, G. F. (1961). Amino acid sensitivity of the dactyl chemoreceptors of Cardnides
maenas. Biol. Bull, iai, 449-55.CASE, J. & GWILLIAM, G. F. (1963). Amino acid detection by marine invertebrates. Proc. 16th Int.
Cong. Zool. 3, 75.CASE, J., GWILLIAM, G. F. & HANSON, F. (i960). Dactyl chemoreceptors of brachyurans. Biol. Bull.
119, 308.COPELAND, M. (1918). The olfactory reactions and organs of the marine snails Alectrion obsolete (Say)
and Busycon canaliculatum. Linn. J. exp. Zool. 35, 177-227.DAKIN, W. J. (1912). Buccinum (the Whelk). L. M. B. C. Mem. 20.FRETTER, V. & GRAHAM, A. (1962). British Prosobranch Molluscs. Ray Soc. Monograph, 144.GESTELAND, R. C , HOWLAND, B., LETTVTN, J. Y. & PITTS, W. H. (1959). Comments on microelectrodes.
Proc. Imt. Radio Engri, N. Y., 47, 1856-62.HENSCHEL, J. (1932). Untersuchengen Uber den chemischer Sinn von Naua reticulata. Win. Meeres-
untersuch. 31, 131-59.HULBERT, G. C. E. B. & YONGE, C. M. (1937). A possible function of the osphradium in the Gastro-
poda. Nature, Lond., 139, 840.KOHN, A_ J. (1961). Chemoreception in gastropod molluscs. Amer. Zool. 1, 291—308.KRIJGSMAN, B. J. & BROWN, A. C. (i960). Water rigor as an aid when operating on marine Gastropoda.
Nature, Lond., 187, 69.LAVBRACK, M. S. (1963). AspectB of chemoreception in Crustacea. Comp. Biochem. Physiol. 8, 141-51.LAVERACK, M. S. (1964). The antennular sense organs of Panultrus argut. Comp. Biochem. Physiol. 13,
301-21.
LAVERACK, M. S. & BAILEY, D. F. (1963). Movement receptors in Buccinum undatum. Comp. Biochem.Physiol. 8, 289-98.
MiCHELSON, E. H. (i960). Chemoreception in the snail Australorbitglabratus. Amer.J.trop.Med.Hyg.9, 480-7.
POTTS, W. T. W. (1958). The inorganic and amino-acid composition of some lamellibranch muscles.J. Exp. Biol. 35, 749-64-
PRINOLE, J. W. S. & WILSON, V. J. (1952). The response of a sense organ to a harmonic stimulus.J. Exp. Biol. 39, 220-34.
STORK, H. A. (1935). BeitrSge zur Histologie und Morphologie des Osphradiums. Arch, nterl. Zool. 1,71-99.
WOLPER, C. (1950). Das Osphradium der Paludina vwipara. Z. vergl. Physiol. 33, 272-86.YONGE, C. M. (1047). The pallia! organs in the aspidobranch Gastropoda and their evolution throughout