Neuron-Specific Modulation by Serotonin of Regenerative Outgrowth and lntracellular Calcium within the CNS of Helisoma trivolvis

Michelle Murrain,*+ A. Don Murphy,t Linda R. Mills, and Stanley B. Kater

Department of Anatomy and Neurobiology and Program in Neuronal Growth, Colorado State University, Fort Collins, Colorado, USA and +Department of Biological Sciences, University of Illinois at Chicago, Chicago, Illinois, USA

SUMMARY

W e have investigated the cell-specific effect of serotonin (5-HT) on regenerating neurons within the adult central nervous system of the pond snail, Helisoma trivolvis. In culture, 5-HT arrests outgrowth of buccal neurons B19 but not neurons B5 (Haydon, McCobb, and Kater, 1984). After axotomy, neurons within the Helisoma nervous system typically exhibit profuse regenerative outgrowth. This study, on neurons within the CNS, shows that 5-HT selectively inhibits the outgrowth of specific identified neurons, and also causes significant elevations in intracellular calcium concentrations as

measured by the calcium indicator dye, Fura-2. The out- growth of neurons B19 and Cl was selectively inhibited when ganglia were incubated in 5 X 10-5M 5-HT. The outgrowth of buccal neurons B5, however, was not af- fected. Moreover, 5-HT caused significant transient ele- vations of calcium concentrations in neurons B19 over 30 minutes, but neurons B5 did not show any increases in calcium concentrations with the addition of 5-HT. These results suggest that the effect of 5-HT upon outgrowth of regenerating neurons may be due to an increase in the intracellular calcium concentration.

INTRODUCTION

The formation, maintenance, and plasticity of nervous systems reflect the actions of a variety of factors. Some factors, such as neurotransmitters and growth factors, can promote or inhibit the growth of neuronal processes. The state of the ner- vous system may well reflect a dynamic balance between growth-promoting and growth-inhibiting stimuli.

Neurotransmitters have been implicated in the inhibition of neurite outgrowth in culture in many systems, both invertebrate (e.g., Haydon et al., 1984) and vertebrate (e.g., Mattson, DOU, and

Received January 10, 1990; accepted February 22, 1990. Journal of Neurohiology, Vol. 2 1, No. 4, pp. 6 1 1-6 18 (1990) 0 1990 John Wiley & Sons, Tnc. CCC 0022-3034/90/0406 1 1-08$04.00

* To whom correspondence should be addressed. t Present address: School of Natural Sciences, Hampshire

College, Amherst, MA 01002. USA.

Kater, 1988). More specifically, the neurotrans- mitter, serotonin, has been shown in Helisoma cell culture to inhibit dramatically neurite outgrowth of a specific identified neuron, B19, and to have no effect on another neuron, B5 (Haydon et al., 1984). Neurotransmitters and growth factors are known to block the effect of growth-inhibiting transmitters also (McCobb, Cohan, Conner, and Kater, 1988; Mattson, Murrain, Guthrie, and Kater, 1989).

The nervous system of the pond snail, Heli- soma, provides an opportunity to examine the plasticity of morphology at the level of identified neurons, and to illuminate the role of a specific neurotransmitter in modulating regeneration of specific neurons in the adult animal.

Thus far, neurotransmitters that retard out- growth have had the effect of increasing the level of intracellular calcium within those neurons ( Kater, Mattson, Cohan, and Conner, 1988). It has been possible to make precise analyses of both growth

611

612 Murrain et al.

cone behavior and intracellular calcium largely by employing individual neurons within isolated cell culture systems.

The present study represents a first step toward examining the role of neurotransmitters in adult neuroplasticity within the context of a more nor- mal environment than that afforded by cell cul- ture. These experiments make use of the ability of individual identified neurons to be studied within the buccal ganglia of Ifelisornu when placed in organ culture. Neurons in organ-cultured ganglia (Murphy, Barker. Loring, and Kater, 1985) dis- play regenerative capabilities remarkably similar to those observed in situ. Thus, it is possible to ask directly about the role of neurotransmitters in re- generative neuronal outgrowth in the context of the more normal environment of the nerve trunks. This report also makes use of newly developed technology for monitoring intracellular calcium concentration within the buccal ganglion. Our re- sults demonstrate that 1 ) 5-HT causes a neuron- selective inhibition of regenerative outgrowth; re- generation of B 19 but not B5 is inhibited by 5-HT; and 2) 5-HT produces a neuron-selective increase in intracellular Caz+; Ca2+ increases in B19 but not in B5 with addition of 5-HT. Taken together, these results demonstrate that the growth inhibi- tory effect of 5-HT, previously reported in cell cul- ture, also has a basis of action in the context of more normal environments provided by buccal ganglia in organ culture.

MATERIALS AND METHODS

Animal Maintenance and Culturing Procedures

The pond snail, Heluoma trivolvis was used for these experiments, from an ongoing inbred colony. The ani- mals were dissected as described previously (Kater and Kaneko, 1972). The circumesophageal ring of ganglia and the buccal ganglia were removed from animals leaving the cerebrobuccal connectives (CBCs) intact. The axons of the identified neurons Cl , B19, and R5 were severcd by crushing the CBCs, the heterobuccal and ventrobuccal nerves, and the esophageal trunks, re- spectively. There were two groups of cultured ganglia: one group in defined, modified Liebowitz medium as described previously (Wong. Hadley, Kater, and Iiauser, 1981), and one in the same medium with 5 X Msolution of serotonin. These ganglia were cul- tured for 2 days. At this time, the cerebral and buccal ganglia were cut away from the circumesophageal ring of ganglia and prepared for dye filling.

Measurement of Outgrowth

Individual neurons were filled with Lucifer Yellow (LY) as described previously ( Hadley, Wong, Kater, Barker, and Bulloch, 1982). Briefly, electrodes were filled with 4% LY, and the cells were filled by injecting 5 nA of hyperpolarizing current for 5 to 7 min.

Because the outgrowth of these neurons was often very complex with many small processes and varicosi- ties, and extremely difficult to quantify with standard morphometric techniques, we used a semiquantitative measurement technique. Based on extensive previous observations of the growth of buccal neurons B5, B19, and cerebral neuron Cl following axotomy (Bulloch, Kater, and Murphy, 1980; Murphy and Kater, 1980; Wong et al., 1981; Bulloch and Kater. 1982; Hadley et al., 1982; Hadley and Kater, 1983). an arbitrary subjec- tive scale was established, from 0.0 for no outgrowth, to 5.0 for maximal outgrowth. Outgrowth observed in- cluded sprouting within the ganglion, as well as neurite extension into the nerves. The extent of these types of outgrowth were the criteria for the subjective scales. Ex- perimental (treated with serotonin) and control (cul- tured without serotonin) neurons were scored blindly. Level 3.0 represents expected normal levels of out- growth after 2 days in culture. The significance of sero- tonin’s effect on outgrowth was confirmed using the Student’s t test.

Imaging of lntracellular Calcium in Buccal Neurons

The buccal ganglia were pinned out, dorsal-side up, and specific neurons were impaled with microelectrodes containing 10 m M of the calcium indicator dye, Fura-2 ( pentapotassium salt) in distilled water. The resistance of the electrodes was 100-200 MQ. The Fura-2 was ion- tophoresed into the cells with a 6-9 nA hyperpolarizing current, delivered in 200 ms pulses (at 1 Hz) to prevent blockage of the electrodes.

The procedure used to visualize intracellular calcium was very similar to that of Cohan, Conner, and Kater ( 1987). Only buccal ganglia were suitable for this tech- nique; cerebral ganglia had sheaths too thick to see neurons clearly within it. Once cells in the buccal gan- glia were filled with Fura-2, the ganglia were placed in a dish with a glass bottom, with a Sylgard (Dow Corning) block above it. Two mL of normal Iielisoma saline (Hadley et al., 1982) was placed in the dish. The cells were visualized on a Zeiss ICM microscope, to which an RCA S.I.T. camera is attached. The output from the camera was fed to a Quantex (Santa Clara, CA) QX7 Image Processing System, with an IBM AT computer base. The image was converted to a digital image, and averaged for 540 ms ( 16 frames). The fluorescent source was put through a computer-controlled filter wheel, which rotated between 350 and 380 nm interference filters. The emission was filtered with a 495 nm long- pass emission filter. Images were taken with each inter-

Inhibition qfRegeneration Within CNS 613

ference filter, and a ratio (R) was obtained. The calcium concentration was determined by the equation (Gryn- kiewicz, Poenie, and Tsien, 1985):

We were able to image the calcium concentrations in situ by using a computer thresholding opcration that subtracted fluorescent scatter from the surrounding tis- sue from the image. This involved making all pixels in the image at a given intensity and below black; this did not affect the intensity or color conversion of pixels above that intensity.

RESULTS

Inhibition of Regeneration with Serotonin

Upon axotomy, neurons within the CNS of Heli- soma send out new processes (Murphy and Kater, 1980). We found that serotonin (5-HT) could de- crease selectively the extent of outgrowth in spe- cific identified neurons, buccal neurons B 19, and cerebral neurons C 1. Other neurons, specifically buccal neurons B5, were unaffected.

Figure 1 shows the outgrowth of buccal neurons B5 and cerebral neurons C1 with 5-HT and con- trol media. The arrows indicate the location of axotomy, which varied slightly from animal to an- imal. In control cultures (no 5-HT), there was ex- tensive outgrowth from the site of axotomy with many small processes and complex varicosities. After 2 days in culture postaxotomy in the pres- ence of serotonin, there was little or no outgrowth in neurons C 1. The average value for outgrowth of C I was 3.5 f 1 .0 ( n = 1 1 ) in control media, com- pared to 0.7 k 0.6 ( n - 1 1 ) in media containing serotonin. Similar effects were seen in B 19. where regeneration was inhibited significantly in the presence of 5-HT also (Fig. 2) . Control B19s had an average outgrowth value of 3.3 -t 0.9 ( n = 15) compared to 0.7 k 0.6 ( n = 18) for B19s in 5-HT media. This inhibitory effect was selective: neurons B5 showed robust outgrowth following axotomy even in the presence of 5-HT. Figure 3 shows a graph comparing the subjective outgrowth measurements in all three neurons, C1, B19, and B5. There were significant differences in the out- growth of both B19s and CIS between those cul- tured in 5-HT and in normal media ( p < 0.00 l Student’s t test), and no significant difference in the outgrowth of B5s in normal and 5-HT media.

CONTROL 5-HT

Figure 1 otomy, in control and 5 X white arrows indicate the point of axotomy.

Photographs of B5s and Cls 2 days after ax- M 5-HT media. The

lntracellular Calcium Concentrations in Neurons B19 and 85

Previous work in culture demonstrated that 5-HT causes the arrest of neurite outgrowth in B 19s and an increase in intracellular Ca2+. In order to de- termine whether changes in intracellular calcium concentration might play a role in the regulation of regenerate morphology described above, we filled individual buccal neurons B19 and B5 with Fura-2.

The intracellular calcium concentration ob- tained in this way in cell bodies was 147.6 k 32.6 n M ( n = 5 ) for normal B5s and 155.5 * 15.9 nil4 ( n = 22) for normal B19s. The resting Ca2’ levels in neurons within the nervous system generally vaned by about 35 n M in different parts of indi- vidual cells, and no one area was significantly dif- ferent from another (ANOVA, p = 0.65, six cells). For these experiments, measurements were taken primarily from the cell bodies. because they were the most consistently measurable part of the cells.

614 Murrain et al.

B19 I I CONTROL

I I

f 5-HT I

I Figure 2 Tracings of two regenerating buccal neurons 19. The neuron on the top was incubated in control medium, and the neuron underneath was incubated in medium con- taining 5 X lo-* M 5-HT. The dotted lines indicate points of axotomy.

Addition of 5-HT caused a significant increase in intracellular calcium concentrations in B19. Figure 4 shows a graph of the time course of the serotonin response in one B19. After addition of 2 X M 5-HT the concentration of Ca2+ in-

creased in this cell from the initial level by 1 17% as measured in the cell body, from 134 to 291 nM. After 20 min, the calcium level began to decrease. The average peak calcium level after addition of serotonin was significantly higher than the rest ( p

4 'i Outgrowth in Neurons B19, B5 and C l

0 Control El 5-HT

* pc.001 -r

B5 B19 c1

Figure 3 Summary of outgrowth measurements in neurons B 19, B5, and C1. The y axis is in the subjective growth units (see text). Averages for neurons B19: Control: 3.33 f 0.9 ( n = 15); Experimental: ( 5 X lo-' M 5-HT media). 0.67 k 0.64 ( n = 18). Averages for neurons C1: Control: 3.5 f I . 1 ( n = 1 1 ); Experimental: 0.68 k 0.6 ( n = 1 1 ) . Averages for neurons B5: Control: 2.89 f 1.06 ( n = 13); Experimental: 2.5 k 0.65 ( n = 14). NeuronsB19 andCl show significant inhibition of outgrowth as compared to control neurons (in both cases p < 0.001. Student's t test).

Inhibition of Regeneration Within CNS 615

Response in Normal B19 to 5-HT

3s0 1

50 1 " I - , ' , . ,

0 10 20 30 Time

Figure 4 An example of the time course of the 5-HT response in B 19s. This graph represents one cell, out of 11 such cells.

< 0.005 paired T , n = 1 l ) , and the intracellular calcium levels in B19s reached peak at 8.5 k 1.5 min. However, by 30 min, despite the continued presence of 5-HT, the average calcium level de- creased close to rest (not significantly higher than rest levels, n = 1 1 ).

The addition of 10 pcM cobalt could block the response to serotonin completely ( n = 4). An ex- ample cell is in Figure 5 . Moreover, neurons in ganglia bathed in 0 m M Ca2' (without EGTA)

Cobalt can block 5-HT response in B19s

300

. A E, 200 0 4

8

0 2 4 6 8 10 12 14 16 Time post 5-HT (min)

Figure 5 An example of a serotonin response in a B19 in normal media (filled squares) and 5-HT response with 10 pM cobalt applied (open squares). Clearly, in the presence of cobalt, there is no rise in calcium levels with the addition of 2 X lo-' 5-HT. Similar results were seen in 4 out of 4 cells.

saline showed no increase in intracellular calcium concentrations when serotonin is added. These re- sults strongly suggest that the rise in intracellular calcium in response to 5-HT is triggered by the influx of calcium.

Paralleling the effects of serotonin on neurite outgrowth, the increase in intracellular Ca2+ was cell-specific: B5 exposed to 5-HT did not show a similar increase. Figure 6 shows a comparison of the peak response in calcium concentrations to 5-HT in B5s and B19s. B5s often showed a small decrease in calcium with 5-HT (four out of six cells), or did not respond at all (two out of six cells), but never showed an increase in calcium concentrations.

After 2 days of incubation in 5-HT medium, there was no significant elevation of intracellular calcium in B5 or B 19 as compared to ganglia in- cubated in control media. We found that the level of calcium in B19s was 143 k 33.2 ( n = 8) in controls and 122 f 8.9 ( n = 5 ) within cells in ganglia incubated in 5-HT. B5s showed concen- trations of 145.9 * 23.2 ( n = 6 ) in controls and 125 ? 20 (y1 = 3 ) in 5-HT (Fig. 7 ) . After 2 days of incubation in 5-HT, the calcium Concentrations were not significantly different from rest levels.

DISCUSSION

Factors that can modulate outgrowth of neurons in the nervous system can have a profound influ-

Peak Calcium Responses to 5-HT

250 1 200

0 B5 B 19

Figure 6 A comparison of peak responses in B5s ( n = 6) and B19s ( n = 11). The y axis is the percent of resting calcium levels. The line represents rest levels (100%). In B19s the average time to reach peak intra- ccllular calcium is 8.5 * 1.5 min.

616 Murrain et al.

2 day incubation in normal media or 5-HT 200 1

ll- I- 150

E c v

.- 100 0 --. 6 4

3 50 --.

.i F - 0

B19s B5s B 19s B5s Control 5-HT

Figure 7 Calcium levels in regenerate neurons after 2 days of incubation either in control media, or 5 X I 0-5 M serotonin. There are no significant differences between the groups.

ence on the morphology and connectivity of these neurons, and thus affect how the nervous system as a whole develops and operates. Increasing evi- dence shows that properties associated with partic- ular classes of neurons either may enhance or in- hibit neuritic growth from other neurons. For in- stance, dendrite formation by the Medial Giant Interneuron of grasshopper embryos was en- hanced in regions of ncuropile-containing sensory axon terminals (Shankland, Bentley, and Good- man, 1982). On the other hand, striatal neurons inhibit the growth of neurites from mesencephalic dopaminergic neurons from mouse embryos when grown in co-culture ( Denis-Dohini, Glow inski, and Prochiantz, 1983 ) , Neurotransmitters have been implicated in regulation of neurite outgrowth also.

In this study, we have found that the neuro- transmitter serotonin causes a cell-specific inhibi- tion of regenerative outgrowth in identified neurons within the adult central nervous system of Helisoma. Further. we have shown that there is a cell-specific increase in the intracellular calcium concentrations in buccal neurons exposed to 5-HT. These data suggest that the effect of 5-HT upon regeneration is due to an increase in the in- tracellular calcium concentration.

In earlier cell culture experiments (Haydon et al., I Y84), the neurotransmitter serotonin was chosen as the test agent because it is known that there is a defined serotonergic input to the area of the nervous system containing neurons B19 and B5. The identified neurons Cl release 5-HT into the buccal ganglia, consequently excite neuron

B19, and activate the motor program for feeding behavior (Granzow and Kater, 1977). Addition of serotonin to the media bathing the buccal ganglion will evoke the feeding motor program also. Clearly the activity of neuron C 1 is involved in the control of feeding behavior of the mature animal, but while involved in the execution of this normal in- tegrated behavior, does the continued release of 5-HT tonically suppress the outgrowth of neurites from ncuron B 1 Y? Recent experiments suggest that during embryogenesis, release of 5-HT from the terminals of C1 could be the signal that halts extension of neuron B 19’s dendrites: adult snails deprived of 5-HT during embryogenesis show aberrant morphology and connectivity of neurons B19 (Goldberg and Kater, 1989).

The growth-inhibiting effects of neurotransmit- ters seem to be mediated through a calcium-de- pendent pathway (Kater et al., 1988). Specifically, in B 19s in culture, 5-HT evokes a sustained rise in intracellular calcium (Cohan et al.. 1987). We have shown in this study that 5-HT can inhibit the outgrowth of B 19s within ganglia and nerves, and elevate intracellular calcium under these condi- tions, thus this inhibition is likely to be via a cal- cium-dependent pathway also. The increase in in- tracellular calcium in neurons B 19 in the presence of serotonin could be due to influx from ligand- gated calcium channels, from voltage-sensitive cal- cium channels, or both. Serotonin is known to cause a depolarization leading to an increased spike frequency in neurons R19. The increase in intracellular-free calcium could result also from a release of calcium from intracellular stores. That

Inhibition of Regeneration Within CNS 61 7

cobalt and zero calcium saline can block the effect entirely suggests that the increase in intracellular Ca2+ is triggered by its influx.

In contrast to neuron B19. the outgrowth of neuron B5 was unaffected by incubation with 5-HT. Neuron B5 also showed no increases in in- tracellular calcium with addition of 5-HT, rein- forcing the hypothesis that an increased level of calcium is the likely mechanism for the inhibition of outgrowth in B 19s.

The measurements made in this study were pri- marily measurements from the cell bodies of these neurons. It was often possible to measure intracel- lular calcium in axons and large dendrites of the neurons in the ganglion. However, measurement of the intracellular calcium levels at the small and highly branched growing tips of regenerating neurons was often difficult. Consequently, cell bodies were selected as the most reliable site to take calcium measurements. Although we recog- nize that increases in intracellular calcium in the cell body may not mirror changes elsewhere in the cell, in a few cases we were able to measure cal- cium in the growing tips, and intracellular calcium did indeed increase in a similar fashion to calcium in the cell bodies. Furthermore, experiments in cell culture with the same Fura-2-dye-filling tech- nique indicate that increases in intracellular cal- cium with addition of serotonin occur both in the cell body and in the growth cones (Murrain, un- published).

The results reported here suggest that calcium is the critical factor in inhibiting outgrowth of B19s; the window of time that calcium is significantly higher than at rest (5-20 min) may well be suffi- cient to inhibit the outgrowth of these neurons. The finding that the rise of intracellular calcium in cells within the nervous system is transient con- trasts with the results found in culture: intracellu- lar calcium remains elevated near peak levels as long as 2 hours after the introduction of 5-HT (Murrain, unpublished).

An additional factor recently proposed by Alkon and Rasmussen ( 1988) is that the cycling of calcium, that is, influx and efflux (or sequestra- tion) might be critical, rather than absolute, levels of intracellular calcium. It is possible then, within the nervous system of Hclisoma, that serotonin causes a sustained influx of calcium, which is compensated by buffering and extrusion mecha- nisms within 30 min after addition of 5-HT. In that case we would not see a sustained increase of calcium by our measurement techniques. Thus, it remains possible either that a transient risc in in- tracellular calcium concentration could cause a

prolonged inhibition of outgrowth, or that a sus- tained. but partially masked, influx of calcium is necessary for the inhibition to occur.

We would like to thank Barbara Bertram, Denny Giddings, and Ping Dou for technical assistance. Bar- bara Hayes for software assistance, and Drs. Roy Ritz- mann and Joffre Mercier for critically reviewing this manuscript.

REFERENCES

ALKON, D. L., and RASMUSSEN, H. (1988). A Spatial- temporal model of cell activation. Science 239:998- 1004.

BUI,I.OCH, A. G. M., and K4TER, S. B. ( 1982). Neurite outgrowth and selection of new electrical connections by adult Helisoma neurons. .I. Neurtiphysiol. 48:569- 583.

BULLOCH. A. G. M.. U r m . S. B., and MURPHY, A. D. (1980). Connectivity changes in an isolated mollus- can ganglion during In Vivn culture. .J. Neurobiol. 11:53 1-546.

COHAN, C. S., CONNER. J. A., and KATER, S. B. ( 1987). Electrically and chemically mediated increases in in- tracellular calcium in neuronal growth cones. J . Neurosci. 7:3 5 88-3.599.

DENIS-DOHINI, GLOWINSKI, S. J., and PROCHIANTZ, A. ( 1983). Spccific influencc of striatal target neurons on the In Vitro outgrowth of mesencephalic dopa- minergic neurites: a morphological quantitative study. J. hTeurosci. 32292-2299.

GOLDBERG, J. I., and KATER, S. B. (1989). Expression and function of the neurotransmitter serotonin dur- ing development of- the Helisomu nervous system. Dev. Biol. 131:483-495.

GRANZOW, B., and KATER, S. B. ( 1977) . Identified higher order neurons controlling the feeding motor program of Hdisoma. Xeurosci. 2: 1049- 1063.

GRYNKIEWICZ, POENIE, G. M . , and TSIEN, R. Y. (1985). A new generation of Ca indicators with greatly improved ff uorescence properties. J. Bid. C’hem. 260:3440-3450.

HADLEY, R. D., WONG. R. G., KATER, S. B., BARKER, D. L., and BULLOCH, A. G. M. ( 1982). Formation of novel central and peripheral connections between molluscan central neurons in organ cultured ganglia. J. Neurobiol. 13:2 17-230.

HADLEY, R. D., and KATEK, S. B. ( 1983). Competence to form electrical connections in restricted to growing neurites in the snail Helisoma. J . Neurosci.

HAYDON, P. G., MCCOBB~ D. P., and KATER, S. B. ( 1984). Serotonin selectively inhibits growth cone motility and synaptogenesis of specific identified neurons. Science 226:561-564.

KATER. S. B., MATTSON, M. P., COHAN, C. S., and CONNER. J. A. ( 1988). Calcium regulation ofthe neu-

3(5):924-932.

618 Murrain et al.

ronal growth cone. Trends in Nwrosci. 11 (7):3 15- 321.

KATER, S. B., and KAN~KO. C. R. S. ( 1972). An endog- enously bursting neuron in the gastropod mollusc, Helisoma trivolvis: Characterization of activity In Vivo. J. Cornp. Physiol. 79:l-14.

MATTSON, M. P., Dou, P., and KATER, S. B. (1988). Outgrowth-regulating actions of glutamate in isolated hippocampal neurons. J. Neurosci. 8:2087-2 100.

MATTSON, M. P.; MIJRRAIN, M., GUTHRIE. P. €3.. and KATER, S. B. (1989). Fibroblast growth factor and glutamate: Opposing roles in the generation and de- generation of hippocampal neuroarchitecture. J . Neurosci. 9f 11):3728-3740.

MCCOBB, D. P., COHAN. C. S., CONNER, J. A., and KATEK, S. B. (1988). Interactive effects of serotonin

and acetylcholine on neurite elongation. Neuron

MURPHY, A. D., and LITER, S. B. (1980). Sprouting and functional regeneration of an identified neuron in Helisoma. Brain Re.7. 186:25 1-272.

MURPHY, A. D., BARKER, D. L., LORING, J. F., and KATER, S. B. (1985). Sprouting and functional re- generation of an identified serotonergic neuron fol- lowing axotomy. J. Ncurobiol. 16: 137-1 5 1.

SHANKLAND, M., BENTLEY, D., and GOODMAN, C. S. ( 1982). Afferent innervation shapes the dendritic branching pattern of the medial giant interneuron in grasshopper embryos raised in culture. Dev. B i d .

WOYG, R. G., HADI,EY, R. P., KATER, S. B., and HAIJSER, G. C. ( 198 1 ). Neurite outgrowth in mollus- can organ and cell cultures: the role of conditioning factor(s). J . Neurosci. 1:1008-1021.

1:377-385.

92:507-520.