- Email: [email protected]
Photoinactivation of neurones in the pond snail, Lymnaea stagnalis: estimation of a safety factor
Brain Research, 524 (1990) 149-152 Elsevier
149
BRES 24199
Photoinactivation of neurones in the pond snail, estimation of a safety factor
Lymnaea stagnalis:
C.J.H. Elliott 1'2 a n d H . - U . K l e i n d i e n s t 2 1Department of Biology, Universityof York, Heslington, York (U.K.) and 2Abteilung Huber, Max Planck lnstitutfiir Verhaltensphysiologie, Seewiesen (E R. G.) (Accepted 10 April 1990)
Key words: Photoinactivation; Lyrnnaea; Neuron; Lucifer yellow; Mollusc
Neurones were irradiated with blue laser light (440 nm). The intensity of light for reliable cell killing (0.5 MW.m-2) was much greater than that used to kill arthropod neurones. In wild snails, there was no difference in the intensity to kill Lucifer yellow-filled neurones and unfilled neurones, probably because of the red pigments in the cell bodies. In laboratory-reared snails, which have much less pigmentation, only the filled cells were killed. The technique of photoinactivation of neurones filled with the fluorescent dye Lucifer yellow was introduced to test models of pattern generation in the crustacean stomatogastric ganglion 8 and has also proved useful with insect preparations 9. Recently there has been rapid progress in understanding the CNS of gastropod molluscs, including the feeding system of Lyrnnaea stagnalis 2. A powerful test of the model of this feeding system would be inactivation of the identified pre-motor interneurones. As a first step, we report here experiments in which Lymnaea motoneurones were photoinactivated. We show that the intensity of blue light required to kill the bright red and orange neurones of Lymnaea is much higher than that used in arthropods, and, in wild-caught snails, is at the level at which direct tissue damage begins. However, damage to unstained cells can be eliminated by using laboratory-reared Lymnaea stagnalis in which the pigmentation is always much weaker. In the lab snails, photoinactivation can be safely achieved by irradiation of filled cells with 0.5 MW.m -2 blue laser light for 2 min with no danger to unstained cells. Experiments were performed on identified motoneurones in the buccal ganglia of Lymnaea stagnalis kept in a standard snail water and fed daily with lettuce. 'Wild caught' snails were obtained from Blades Biological (Kent), while 'laboratory-reared' snails were raised in York with lettuce as the sole diet. The CNS, accompanied by only a short length of oesophagus, was dissected out and pinned down in a sylgard dish containing standard Lymnaea saline 4. The ganglia were treated with 0.1% protease (Sigma XIV) for 2-5 min and the neuronal
somata impaled with glass microelectrodes (20-40 MO). Neurones were identified by location or pattern of synaptic inputs 2'4. To stain cells, Lucifer yellow C H (used as a 3% solution in 1 M lithium chloride or in distilled water) or 5(6) carboxyfluorescein (5-CF; Eastman Kodak ref 9953, saturated solution at p H 7 in distilled water) were used. In control experiments 1 M lithium chloride or 4 M potassium acetate were used. A Krypton (Coherent Innova 90K; wavelength 415 nm) or Helium Cadmium (439 Omnichrome with a 20x beam expander; wavelength 442 nm) Laser was used to supply blue light. A prism or front silvered mirror was used to direct the beam onto the preparation. In a few experiments this optical system was used to give a spot 1 mm in diameter. In most experiments, a lens of focal length 250 mm was inserted into the beam focussing it to a 100 ~m spot. The ganglia were observed with a stereo microscope, fitted with a G G 475 (Schott) barrier filter. When cells from wild-caught Lymnaea stagnalis were filled with Lucifer yellow and illuminated with 0.5 MW.m -2, the cell body shone brightly, but the neuropilar processes were never visible. If the beam was directed onto the nerves containing the axon of the stained cells, then the axon could be clearly seen. It would appear that the red to orange pigment in the neuronal somata blocks either the laser beam or the yellow fluorescence. Steady illumination at 0.5 MW.m -2 depolarises and then kills the motoneurones (Figs. 1 and 2A). During killing, first the membrane resistance and then the membrane potential decline to zero, a high frequency barrage of action
Correspondence: C.J.H. Elliott, Department of Biology, University of York, Heslington, York, YO1 5DD, U.K. 0006-8993/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
150 potentials is recorded and all synaptic inputs cease 1"3"8. Fig. 2A summarises the results of 2 min illumination. The proportion of the filled neurones depolarised and killed increases with beam intensity with a threshold just below 0.1 MW.m -2. At low intensity, (0.1 MW.m-2), some cells apparently remain unaffected for up to 7 min, but are always eventually depolarised and then killed. At maximum intensity (1 MW.m -2) 50% of the motoneurones are killed in less than 2 min, the fastest taking just 10 s. Filling the cells for longer periods of time did not reduce the time needed to kill the 4 cluster motoneurones. In control experiments unfilled cells of wild-caught snails Lymnaea stagnalis were also irradiated. These were found to be damaged by the level of illumination required to kill filled cells. The sequence of events is similar to that observed in filled cells but generally slower (Fig. 1). Fig. 2A summarises the damage produced by 2 min irradia-
tion. Note that with a beam intensity of 0.5 MW.m -z (the minimum to kill filled cells) 50% of unstained wild cells are damaged sufficiently to be noticeably depolarised. In 9 experiments, the time taken to kill stained and unstained pairs of 4 cluster cells was compared. Only in 5 out of these 9 experiments was the filled motoneurone killed faster than its contralateral, unfilled homologue. In three preparations the sensitivity was equal, while in the remaining two the unfilled cell showed signs of damage first. After irradiation, the neurones are discoloured: the tissue appeares white and opaque and the boundaries of the cell bodies are no longer visible. In wild snails, the bleaching occurs throughout the illuminated spot, irrespective of whether any of the neurones have been filled.
A: wild snails 100
beam on filled
beam i n t e n s i t y MW m -2
0.1
0.5.
1.0
~ .
.
cell
50-
~
25-
control elo© trode[ r._~aiced
--
+
--
+
--
+
dye (lucifer yellow)
B: lab snails 10o
[C~3 OK depol killed
,~ =
Q) tO
E 25
bellm on control coil
% 5
none rain
tO0
mV
40 mV 1
L
minute
Fig. 1. Photoinactivation of Lymnaea neurones. Simultaneous recordings from two 4 cluster cells (4 CL) one filled, the other, in the contralateral ganglion, unfilled. Illumination of the filled cell with 0.1 MW.m -2 (bar) causes depolarisation and a barrage of spikes. At the end the electrode was raised, showing the resting potential to be less than 5 mV. Five min irradiation was needed before any change was apparent in the unstained cell and it was killed after 15 min.
5-CF dye
LY
beam intensity 0.5 MW m -2
Fig. 2. The efficiency of photoinactivation killing of buccal neurones in Lymnaea stagnalis. A: in wild-caught snails: the influence of Lucifer yellow (42 cells). Cells were filled with Lucifer yellow (3% in 1 M lithium chloride) by passing 500 ms pulses o f - 2 . 5 n A at 1 Hz for 10 min. Lithium chloride electrodes were used in control cells. The damage to the cell was assessed after 2 rain irradiation. At a beam intensity of 0.1 MW.m 2, depolarisation is seen in only 25% of filled cells; unstained cells are not affected. At 0.5 MW.m -2, all filled cells are damaged and about 10% killed, while 50% of unstained cells are still unaffected. At 1 MW.m -z, approximately 50% of cells are killed irrespective of staining. B: in laboratoryreared Lymnaea stagnalis, Lucifer yellow (LY; 3% in distilled water) or 5-CF (saturated solution in distilled water) was injected into 4 cluster buccal motoneurones by passing a steady -1 n A current. Control cells were penetrated with potassium acetate electrodes. Damage to the cells was assessed after 2 min irradiation. Control cells were unaffected, 50% of 5-CF-injected cells were depolarised and 50% undamaged, while all LY cells were killed. Compare the results of this treatment with the wild-caught snails (center bars of A).
151
on beam
A O"
off
0 +
> v
E
-25'
snail snail
2] 1
-50' c-75
O
(3.
B
-100
6
5'o 16o
E ¢,-
+
010~E v
Q) 0 t-
5-
2 ~ o
-50
6
5'0
time (s)
1O0 150
Fig. 3. Irradiation of control cells (0.5 MW.m -2, 2 min) does not affect either (A) the membrane potential or (B) the membrane resistance. The membrane resistance was measured by passing a triangular waveform current (+ 2.5 nA, period 4 s) continuously through one electrode and measuring the voltage changes with a second electrode.
Tissue damage began at roughly the same time as cell depolarisation, and is accompanied by increasing autofluorescence. The most likely cause of damage in the unfilled cells is absorbtion of the blue light by the reddish-orange pigment of the Lymnaea neurones. Since the pigmentation of neurones in laboratory-reared snails is much less, they would be expected to absorb less light and so would be damaged less easily. This is indeed the case: Fig. 2B shows that none of the eight unfilled motoneurones tested in laboratory reared snails were affected by 2 min irradiation with 0.5 MW.m -2. Fig. 3 shows detailed analysis of two experiments, indicating that there was no change in membrane resistance or resting potential within the 2 min test period. With prolonged illumination, a change in membrane potential was first recorded after 10 min, when discolouration of the cells began. However, with the Lucifer yellow-filled cells, the same beam intensity (0.5 MW.m -2) is sufficient to kill all within 2 min (Fig. 2B). It took longer to kill cells if less dye was injected. Filling the smaller motoneurones (e.g. 7 cells) gave quicker fills for the same period of dye injection. These results show that, with the laboratory-reared snails, there is a good safety factor - - at least 8 min - between the irradiation required to kill filled and unfilled cells. By contrast with the wild snails, the laboratory reared snails have less pigment, the axons and large diameter
dendritic branches of filled cells were visible through the outer layer of cell bodies (unlike the wild snails). In Helisoma and the crustacean stomatogastric ganglia, neurones have been killed by the fluorescent dye 5-CF, which has a higher quantum yield than Lucifer yellow3'5'7. However, in Lymnaea stagnalis 5-CF does not kill cells nor does Lucifer yellow. Although the cell bodies and axons of 5-CF-filled cells glowed strongly, only half (3 out of 6) of the 5-CF cells were depolarised within the 2 min test period and the others were not affected. It would seem that injection of Lucifer yellow gives more reliable killing; perhaps the 5-CF can cross the cell membrane. One possible cause of damage is an increase in temperature. The laser beam was set to 1 MW.m -2 and a thermistor placed in the bath. The thermistor output was monitored with a Technotherm 9500 digital meter. Focussing the beam onto the element increased the temperature from 21 to 34 °C. The same temperature change was measured with the thermistor hidden within the buccal ganglion. If the thermistor is raised above the surface with a small Luficer yellow droplet on the end, then the beam produced a greater increase in temperature (up to 90 °C). These experiments show that the photoinactivation of neurones in the CNS of Lymnaea follows a pattern similar to that reported from arthropods8'9: the neurone is depolarised, generating a high frequency barrage of spikes and synaptic potentials disappear (Fig. 1). Importantly, however, successful killing of the snail neurones requires a much greater light intensity than arthropod neurones. In the same apparatus, 2 min illumination at only 5 kW.m -2 kills auditory neurones of the cricket Gryllus campestris 9 - - this is one hundredth of the energy required in Lymnaea. In Acheta, a helium-cadmium laser (10 MW.m -2) killed parts of neurones in 3 s6. One possible explanation of the slower killing of the snail neurones is that the 4 cells have large (up to 100/~m diameter) cell bodies, and this means that the concentration of Lucifer yellow might be lower than that in arthropods. It would then not be surprising that, in the laboratory reared snails, the 7 cells, which have much smaller (20/~m) cell bodies, are killed more quickly. The second important result is that in Lymnaea successful inactivation requires laboratory reared snails. In wild-caught snails, there is little difference between the susceptibility of stained and unstained neurones to photodamage (Fig. 2), and the same result was found when the time to kill pairs of stained and unstained 4 cluster cells was measured. The results with the less pigmented laboratory reared snails show that the endogenous red pigment is an important absorber of blue light. With the reduced pigment in lab-reared snails, killing of cells filled with Lucifer yellow is regularly achieved in 2
152 min irradiation, but d a m a g e to unfilled cells is only seen after 10 min (Fig. 3). O u r results show that at these beam intensities, the t e m p e r a t u r e o f the bath is raised above 30 °C and that the same t e m p e r a t u r e increase also occurs inside the ganglia. If the intracellular t e m p e r a t u r e is quickly raised, it seems likely that this could contribute to cell death, even
We would like to thank Prof. E Huber for his help and encouragement, Mr. A. Willis for rearing snails and the SERC (U.K.) for financial support.
1 Arshavsky, Yu.I., Orlovsky, G.N. and Panchin, Yu.V., Control of locomotion in marine mollusc - - Clione limacina. V. Photoinactivation of efferent neurons, Exp. Brain Res., 59 (1985) 203-205. 2 Benjamin, P.R. and Elliott, C.J.H., Snail feeding oscillator: the central pattern generator and its control by modulatory interneurons. In J.W. Jacklet (Ed.), Neuronal and Cellular Oscillators, Dekker, New York, 1989. 3 Cohan, C.S., Hadley, R.D. and Kater, S.B. 'Zap axotomy': localized fluorescent excitation of single dye-filled neurons induces growth by selective axotomy, Brain Research, 270 (1983) 93-101. 4 Ellion, C.J.H. and Benjamin, P.R., Interactions of the pattern generating interneurons controlling feeding in Lymnaea stagnalis,
J. Neurophysiol., 54 (1985) 1396-1411. 5 Harris-Warrick, R.M. and Flamm, R.E., Multiple mechanisms of bursting in a conditional bursting neuron, J. Neurosci., 7 (1987) 2113-2128. 6 Jacobs, G.A. and Miller, J.P., Functional properties of individual neuronal branches isolated in situ by laser photoinactivation, Science, 228 (1985) 344-346. 7 Kater, S.B. and Hadley, R.B., Video monitoring of neuronal plasticity, Trends Neurosci., 5 (1985)80-82. 8 Miller, J.P. and Selverston, A.I., Rapid killing of single neurons by irradiation of injected dye, Science, 206 (1979) 702-704. 9 Selverston, A.I., Kleindienst, H.-U. and Huber, E, Synaptic connectivity between cricket auditory interneurons as studied by selective photoinactivation, J. Neurosci., 5 (1985) 1283-1292.
without the production of toxic products from the p h o t o d e c o m p o s i t i o n of the red pigments or dyes.
149
BRES 24199
Photoinactivation of neurones in the pond snail, estimation of a safety factor
Lymnaea stagnalis:
C.J.H. Elliott 1'2 a n d H . - U . K l e i n d i e n s t 2 1Department of Biology, Universityof York, Heslington, York (U.K.) and 2Abteilung Huber, Max Planck lnstitutfiir Verhaltensphysiologie, Seewiesen (E R. G.) (Accepted 10 April 1990)
Key words: Photoinactivation; Lyrnnaea; Neuron; Lucifer yellow; Mollusc
Neurones were irradiated with blue laser light (440 nm). The intensity of light for reliable cell killing (0.5 MW.m-2) was much greater than that used to kill arthropod neurones. In wild snails, there was no difference in the intensity to kill Lucifer yellow-filled neurones and unfilled neurones, probably because of the red pigments in the cell bodies. In laboratory-reared snails, which have much less pigmentation, only the filled cells were killed. The technique of photoinactivation of neurones filled with the fluorescent dye Lucifer yellow was introduced to test models of pattern generation in the crustacean stomatogastric ganglion 8 and has also proved useful with insect preparations 9. Recently there has been rapid progress in understanding the CNS of gastropod molluscs, including the feeding system of Lyrnnaea stagnalis 2. A powerful test of the model of this feeding system would be inactivation of the identified pre-motor interneurones. As a first step, we report here experiments in which Lymnaea motoneurones were photoinactivated. We show that the intensity of blue light required to kill the bright red and orange neurones of Lymnaea is much higher than that used in arthropods, and, in wild-caught snails, is at the level at which direct tissue damage begins. However, damage to unstained cells can be eliminated by using laboratory-reared Lymnaea stagnalis in which the pigmentation is always much weaker. In the lab snails, photoinactivation can be safely achieved by irradiation of filled cells with 0.5 MW.m -2 blue laser light for 2 min with no danger to unstained cells. Experiments were performed on identified motoneurones in the buccal ganglia of Lymnaea stagnalis kept in a standard snail water and fed daily with lettuce. 'Wild caught' snails were obtained from Blades Biological (Kent), while 'laboratory-reared' snails were raised in York with lettuce as the sole diet. The CNS, accompanied by only a short length of oesophagus, was dissected out and pinned down in a sylgard dish containing standard Lymnaea saline 4. The ganglia were treated with 0.1% protease (Sigma XIV) for 2-5 min and the neuronal
somata impaled with glass microelectrodes (20-40 MO). Neurones were identified by location or pattern of synaptic inputs 2'4. To stain cells, Lucifer yellow C H (used as a 3% solution in 1 M lithium chloride or in distilled water) or 5(6) carboxyfluorescein (5-CF; Eastman Kodak ref 9953, saturated solution at p H 7 in distilled water) were used. In control experiments 1 M lithium chloride or 4 M potassium acetate were used. A Krypton (Coherent Innova 90K; wavelength 415 nm) or Helium Cadmium (439 Omnichrome with a 20x beam expander; wavelength 442 nm) Laser was used to supply blue light. A prism or front silvered mirror was used to direct the beam onto the preparation. In a few experiments this optical system was used to give a spot 1 mm in diameter. In most experiments, a lens of focal length 250 mm was inserted into the beam focussing it to a 100 ~m spot. The ganglia were observed with a stereo microscope, fitted with a G G 475 (Schott) barrier filter. When cells from wild-caught Lymnaea stagnalis were filled with Lucifer yellow and illuminated with 0.5 MW.m -2, the cell body shone brightly, but the neuropilar processes were never visible. If the beam was directed onto the nerves containing the axon of the stained cells, then the axon could be clearly seen. It would appear that the red to orange pigment in the neuronal somata blocks either the laser beam or the yellow fluorescence. Steady illumination at 0.5 MW.m -2 depolarises and then kills the motoneurones (Figs. 1 and 2A). During killing, first the membrane resistance and then the membrane potential decline to zero, a high frequency barrage of action
Correspondence: C.J.H. Elliott, Department of Biology, University of York, Heslington, York, YO1 5DD, U.K. 0006-8993/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
150 potentials is recorded and all synaptic inputs cease 1"3"8. Fig. 2A summarises the results of 2 min illumination. The proportion of the filled neurones depolarised and killed increases with beam intensity with a threshold just below 0.1 MW.m -2. At low intensity, (0.1 MW.m-2), some cells apparently remain unaffected for up to 7 min, but are always eventually depolarised and then killed. At maximum intensity (1 MW.m -2) 50% of the motoneurones are killed in less than 2 min, the fastest taking just 10 s. Filling the cells for longer periods of time did not reduce the time needed to kill the 4 cluster motoneurones. In control experiments unfilled cells of wild-caught snails Lymnaea stagnalis were also irradiated. These were found to be damaged by the level of illumination required to kill filled cells. The sequence of events is similar to that observed in filled cells but generally slower (Fig. 1). Fig. 2A summarises the damage produced by 2 min irradia-
tion. Note that with a beam intensity of 0.5 MW.m -z (the minimum to kill filled cells) 50% of unstained wild cells are damaged sufficiently to be noticeably depolarised. In 9 experiments, the time taken to kill stained and unstained pairs of 4 cluster cells was compared. Only in 5 out of these 9 experiments was the filled motoneurone killed faster than its contralateral, unfilled homologue. In three preparations the sensitivity was equal, while in the remaining two the unfilled cell showed signs of damage first. After irradiation, the neurones are discoloured: the tissue appeares white and opaque and the boundaries of the cell bodies are no longer visible. In wild snails, the bleaching occurs throughout the illuminated spot, irrespective of whether any of the neurones have been filled.
A: wild snails 100
beam on filled
beam i n t e n s i t y MW m -2
0.1
0.5.
1.0
~ .
.
cell
50-
~
25-
control elo© trode[ r._~aiced
--
+
--
+
--
+
dye (lucifer yellow)
B: lab snails 10o
[C~3 OK depol killed
,~ =
Q) tO
E 25
bellm on control coil
% 5
none rain
tO0
mV
40 mV 1
L
minute
Fig. 1. Photoinactivation of Lymnaea neurones. Simultaneous recordings from two 4 cluster cells (4 CL) one filled, the other, in the contralateral ganglion, unfilled. Illumination of the filled cell with 0.1 MW.m -2 (bar) causes depolarisation and a barrage of spikes. At the end the electrode was raised, showing the resting potential to be less than 5 mV. Five min irradiation was needed before any change was apparent in the unstained cell and it was killed after 15 min.
5-CF dye
LY
beam intensity 0.5 MW m -2
Fig. 2. The efficiency of photoinactivation killing of buccal neurones in Lymnaea stagnalis. A: in wild-caught snails: the influence of Lucifer yellow (42 cells). Cells were filled with Lucifer yellow (3% in 1 M lithium chloride) by passing 500 ms pulses o f - 2 . 5 n A at 1 Hz for 10 min. Lithium chloride electrodes were used in control cells. The damage to the cell was assessed after 2 rain irradiation. At a beam intensity of 0.1 MW.m 2, depolarisation is seen in only 25% of filled cells; unstained cells are not affected. At 0.5 MW.m -2, all filled cells are damaged and about 10% killed, while 50% of unstained cells are still unaffected. At 1 MW.m -z, approximately 50% of cells are killed irrespective of staining. B: in laboratoryreared Lymnaea stagnalis, Lucifer yellow (LY; 3% in distilled water) or 5-CF (saturated solution in distilled water) was injected into 4 cluster buccal motoneurones by passing a steady -1 n A current. Control cells were penetrated with potassium acetate electrodes. Damage to the cells was assessed after 2 min irradiation. Control cells were unaffected, 50% of 5-CF-injected cells were depolarised and 50% undamaged, while all LY cells were killed. Compare the results of this treatment with the wild-caught snails (center bars of A).
151
on beam
A O"
off
0 +
> v
E
-25'
snail snail
2] 1
-50' c-75
O
(3.
B
-100
6
5'o 16o
E ¢,-
+
010~E v
Q) 0 t-
5-
2 ~ o
-50
6
5'0
time (s)
1O0 150
Fig. 3. Irradiation of control cells (0.5 MW.m -2, 2 min) does not affect either (A) the membrane potential or (B) the membrane resistance. The membrane resistance was measured by passing a triangular waveform current (+ 2.5 nA, period 4 s) continuously through one electrode and measuring the voltage changes with a second electrode.
Tissue damage began at roughly the same time as cell depolarisation, and is accompanied by increasing autofluorescence. The most likely cause of damage in the unfilled cells is absorbtion of the blue light by the reddish-orange pigment of the Lymnaea neurones. Since the pigmentation of neurones in laboratory-reared snails is much less, they would be expected to absorb less light and so would be damaged less easily. This is indeed the case: Fig. 2B shows that none of the eight unfilled motoneurones tested in laboratory reared snails were affected by 2 min irradiation with 0.5 MW.m -2. Fig. 3 shows detailed analysis of two experiments, indicating that there was no change in membrane resistance or resting potential within the 2 min test period. With prolonged illumination, a change in membrane potential was first recorded after 10 min, when discolouration of the cells began. However, with the Lucifer yellow-filled cells, the same beam intensity (0.5 MW.m -2) is sufficient to kill all within 2 min (Fig. 2B). It took longer to kill cells if less dye was injected. Filling the smaller motoneurones (e.g. 7 cells) gave quicker fills for the same period of dye injection. These results show that, with the laboratory-reared snails, there is a good safety factor - - at least 8 min - between the irradiation required to kill filled and unfilled cells. By contrast with the wild snails, the laboratory reared snails have less pigment, the axons and large diameter
dendritic branches of filled cells were visible through the outer layer of cell bodies (unlike the wild snails). In Helisoma and the crustacean stomatogastric ganglia, neurones have been killed by the fluorescent dye 5-CF, which has a higher quantum yield than Lucifer yellow3'5'7. However, in Lymnaea stagnalis 5-CF does not kill cells nor does Lucifer yellow. Although the cell bodies and axons of 5-CF-filled cells glowed strongly, only half (3 out of 6) of the 5-CF cells were depolarised within the 2 min test period and the others were not affected. It would seem that injection of Lucifer yellow gives more reliable killing; perhaps the 5-CF can cross the cell membrane. One possible cause of damage is an increase in temperature. The laser beam was set to 1 MW.m -2 and a thermistor placed in the bath. The thermistor output was monitored with a Technotherm 9500 digital meter. Focussing the beam onto the element increased the temperature from 21 to 34 °C. The same temperature change was measured with the thermistor hidden within the buccal ganglion. If the thermistor is raised above the surface with a small Luficer yellow droplet on the end, then the beam produced a greater increase in temperature (up to 90 °C). These experiments show that the photoinactivation of neurones in the CNS of Lymnaea follows a pattern similar to that reported from arthropods8'9: the neurone is depolarised, generating a high frequency barrage of spikes and synaptic potentials disappear (Fig. 1). Importantly, however, successful killing of the snail neurones requires a much greater light intensity than arthropod neurones. In the same apparatus, 2 min illumination at only 5 kW.m -2 kills auditory neurones of the cricket Gryllus campestris 9 - - this is one hundredth of the energy required in Lymnaea. In Acheta, a helium-cadmium laser (10 MW.m -2) killed parts of neurones in 3 s6. One possible explanation of the slower killing of the snail neurones is that the 4 cells have large (up to 100/~m diameter) cell bodies, and this means that the concentration of Lucifer yellow might be lower than that in arthropods. It would then not be surprising that, in the laboratory reared snails, the 7 cells, which have much smaller (20/~m) cell bodies, are killed more quickly. The second important result is that in Lymnaea successful inactivation requires laboratory reared snails. In wild-caught snails, there is little difference between the susceptibility of stained and unstained neurones to photodamage (Fig. 2), and the same result was found when the time to kill pairs of stained and unstained 4 cluster cells was measured. The results with the less pigmented laboratory reared snails show that the endogenous red pigment is an important absorber of blue light. With the reduced pigment in lab-reared snails, killing of cells filled with Lucifer yellow is regularly achieved in 2
152 min irradiation, but d a m a g e to unfilled cells is only seen after 10 min (Fig. 3). O u r results show that at these beam intensities, the t e m p e r a t u r e o f the bath is raised above 30 °C and that the same t e m p e r a t u r e increase also occurs inside the ganglia. If the intracellular t e m p e r a t u r e is quickly raised, it seems likely that this could contribute to cell death, even
We would like to thank Prof. E Huber for his help and encouragement, Mr. A. Willis for rearing snails and the SERC (U.K.) for financial support.
1 Arshavsky, Yu.I., Orlovsky, G.N. and Panchin, Yu.V., Control of locomotion in marine mollusc - - Clione limacina. V. Photoinactivation of efferent neurons, Exp. Brain Res., 59 (1985) 203-205. 2 Benjamin, P.R. and Elliott, C.J.H., Snail feeding oscillator: the central pattern generator and its control by modulatory interneurons. In J.W. Jacklet (Ed.), Neuronal and Cellular Oscillators, Dekker, New York, 1989. 3 Cohan, C.S., Hadley, R.D. and Kater, S.B. 'Zap axotomy': localized fluorescent excitation of single dye-filled neurons induces growth by selective axotomy, Brain Research, 270 (1983) 93-101. 4 Ellion, C.J.H. and Benjamin, P.R., Interactions of the pattern generating interneurons controlling feeding in Lymnaea stagnalis,
J. Neurophysiol., 54 (1985) 1396-1411. 5 Harris-Warrick, R.M. and Flamm, R.E., Multiple mechanisms of bursting in a conditional bursting neuron, J. Neurosci., 7 (1987) 2113-2128. 6 Jacobs, G.A. and Miller, J.P., Functional properties of individual neuronal branches isolated in situ by laser photoinactivation, Science, 228 (1985) 344-346. 7 Kater, S.B. and Hadley, R.B., Video monitoring of neuronal plasticity, Trends Neurosci., 5 (1985)80-82. 8 Miller, J.P. and Selverston, A.I., Rapid killing of single neurons by irradiation of injected dye, Science, 206 (1979) 702-704. 9 Selverston, A.I., Kleindienst, H.-U. and Huber, E, Synaptic connectivity between cricket auditory interneurons as studied by selective photoinactivation, J. Neurosci., 5 (1985) 1283-1292.
without the production of toxic products from the p h o t o d e c o m p o s i t i o n of the red pigments or dyes.