ABSTRACT The NIE-115 mouse neuroblastoma cell line was used to investigate the role of the cholinergic receptor/ nitric oxide/cyclic GMP pathway in regulating neurite outgrowth. Here evidence is presented that both carbachol and 8-Bromo-cGMP inhibit neurite outgrowth. Nitric oxide donors, however, promote various forms of neurite outgrowth. DEANO and Spermine NONOate were shown to promote neurite outgrowth from lamellipodia, but had no effect on neurites that did not possess lamellipodia before administration of the donors. SNAP was shown to allow growth of the main shafts of neurites, but not that of lamellipodia or their extended neurites. This study suggests that nitric oxide promotes growth through a cGMP-independent pathway. During activation of cholinergic receptors, the stimulatory effects of nitric oxide appear to be masked by other pathways in the cholinergic cascade. INTRODUCTION In the cholinergic receptor/nitric oxide/cGMP pathway, the muscarinic cholinergic receptor, Ml, is coupled to the activation of nitric oxide synthase (NOS). NOS catalyzes the conversion of L-arginine into L-citrulline and nitric oxide gas Nitric oxide, in turn, has been shown to be coupled to the activation of guanylyl cyclase and the production of cyclic GMP [3,5,1 1), as well as several other, less well-characterized actions [21. Finally, cyclic GMP acts as a signal molecule for opening and closing membrane channels The present study investigates the role of nitric oxide in regulating neurite outgrowth in the NIE-115 mouse neuroblastoma cell line. Along with other receptors, NIE-115 cells express the Ml receptor. Previous studies conducted on the role of this pathway in neurite outgrowth regulation have been inconsistent. While it has been suggested that cGMP inhibits motility and causes neurite retraction in NIE-115 cells 111. other research have found that cGMP promotes neurite extension and motility in rat dorsal root ganglion (DRG) neurons [121. It has been shown that carbachol inhibits the growth- promoting effects of forskolin in human neuroblastoma NB-OKI cells [71, yet acetylcholinesterase promotes regeneration of neurites in cultured adult neurons of Aplysia [10]. The effects of nitric oxide on neurite outgrowth have also been controversial. While studies have shown that NO arrests growth of neuronal cells during differentiation 6,9 and inhibits extension of regenerating axons of rat DRG neurons [131, other research has indicated that nitric oxide promotes neurite extension and motility in rat DRG neurons [121 The present study investigates the role of nitric oxide on neurite outgrowth by examining three points along the MI receptor cascade: (1) at the level of the muscarinic receptor, using the cholinergic agonist carbachol, (2) at the level of nitric oxide, using NO donors DEANO, Spermine NONOate, and SNAP, and (3) at the level of cGMP using the membrane permeable analog, 8-Bromo-cGMP MATERIALS AND METHODS Materials S-nitroso-N-acetylpenicillamine (SNAP) and diethylamine nitric oxide, sodium salt (DEANO) were purchased from Research Biochemicals International (Natick, MA) Spermine NONOate was from Alexis (San Diego, CA, U.S.A.). Carbachol and 8-Bromo¬ cyclic GMP were from Calbiochem (San Diego, CA, U.S.A.). Cell Culture Mouse neuroblastoma clone NIE-115 cells (from UCSF Cell Culture Facility) were grown in Dulbecco's modified Eagle’s medium (DMEM) (Sigma, Saint Louis, MO U.S.A.) supplemented with 10% fetal bovine serum (HyClone, Logan, UT). Cells were incubated at 37°C in a humidified atmosphere containing 10% carbon dioxide. Cells were plated onto glass cover slips and placed in 35 mm plastic tissue culture dishes (Falcon, Becton Dickinson Labware, Lincoln Park, NJ). Cells grown to-60% confluency were then differentiated by treating with 2% dimethylsulfoxide (DMSO). All cells were cultured for at least 3 days before use and fed one day before use. Time-Lapse Video/OMDR Microscopy Immediately before each experiment, a glass chip was selected from one of the plastic culture dishes containing differentiated cells and placed in 10 mM HEPES buffer (phosphate buffered saline, PBS) at pH 7.4. The chip was then transferred to a well that fits into the stage of the microscope (Nikon Diaphot) and secured with a small amount of petroleum jelly applied to the edges of the chip. This well was then placed onto the stage and PBS was added immediately to prevent drying out of the cells. A field containing as many distinct cells as possible, of varied morphologies and with visible processes, was selected. Time-lapse video (Panasonic VHS) and optical disk (Panasonic OMDR) recordings were made simultaneously, at a rate of one frame every two minutes, using Nomarski D.I.C. optics and a Hammatsu C2400 SIT camera. A total of 69 frames was recorded for each treatment, 9 frames before and 60 frames after stimulation The images were acquired with a 20X bright field objective at low light intensities and at a controlled temperature of 30°C. A pipeline image processor (Megavision, Santa Barbara, CA) was used to digitize images, and its application programs were used for analysis of the entire time series. Select images were converted into bitmap files and transferred to Adobe Photoshop 3.0. Measurements and further analysis were done using Photoshop applications. Digitally processed images were printed from an HP LaserJet AMV printer. Controls It was found that the length of time the cells were in the well for video time- lapsing (-2.5 hours), coupled with the temperature of 30°C, increased the osmolarity of the bathing solution by more than 25%. To control for this effect, all experiments were done under constant perfusion (rate - 7.5 mL/hr) with the bathing solution. Because different cells have varied basal motility, an 18-minute control period of growth in PBS was recorded for each experiment prior to treatment, against which growth during the 18 minutes after treatment was compared. The control period was used to compare basal growth rates between experiments. Growth in PBS was considered as the control for all the treatments. Triplicate experiments using PBS were performed and recorded for the same total number of frames as for the treatment experiments. Neurite Outgrowth Assay After the initial 18-minute PBS control period, the test compound was added. Saline containing 100 uM DEANO was introduced into the chamber and allowed 20 minutes to react, after which constant perfusion with PBS was resumed. 1OOuM NO donor Spermine NONOate was applied using constant perfusion. Nitric oxide donor S- nitroso-N-acetylpenacillamine, at 10OuM in 1% DMSO, was also added using constant perfusion. The cyclic GMP analog, 8-Br-cGMP, was added at a concentration of ImM and allowed to react for 15 minutes before switching to constant perfusion at 20 uM. Cells treated with carbachol were first incubated with Fura-2 (SuM) for 30 minutes on a shaker table in the dark before experimentation. At addition of carbachol, the cells were viewed with a 20X Fluor objective under UV light and video taped (Panasonic VHS) to record calcium concentration increases in responding cells: the cells were recorded for approximately 5 seconds at 340 nm before stimulation (to obtain background fluorescence), and for one minute at 380 nm upon addition of carbachol. Carbachol was added at an initial concentration of 1 mM and allowed to react for 15 minutes before starting constant perfusion at 20 uM. Triplicate experiments were performed for each test compound. Cells were time-lapse recorded for an additional two hours from time of stimulation. In addition to studying the entire time series as movies and using Megavision program applications for image processing and individual frame analysis, images taken at times -18, 0, 10, 18, 30, 60, 90, and 120 minutes relative to time of treatment were analyzed and measured using Photoshop applications. RESULTS Cells grown in the PBS control exhibited a significant amount of growth, which manifested itself in a variety of ways (see figure 1). Initiation and extension of neurites from lamellipodia was observed. The number of neurites sprouting from each lamellipodium varied, but they were oftentimes delicate and finely branched. Frequently, instead of the neurites growing out from the ends of the lamellipodia, entire lamellipodia would develop into these highly-branched neurites, suggesting that the process is one involving the maturation of motile lamellipodia into neurites. Branches off the main shafts of processes lengthened in PBS, as well as the main shafts themselves, regardless of the original process length. Elongation and broadening of lamellipodia was also observed. DEANO, Spermine NONOate, and SNAP Inhibit SpecificTypes of Growth Cells grown in DEANO and Spermine NONOate exhibited the same types of growth (see figures 2 & 3). In either treatment, neurites are seen to extend and elongate from lamellipodia. As with cells grown in PBS, it is sometimes observed that entire lamellipodia develop into thin, distinct processes. Also analogous to the control, these neurites often appear delicate and finely branched. Both DEANO and Spermine NONOate, however, appear to inhibit growth in the main shafts of processes. Growth of neuronal cells treated with SNAP was different from that of cells treated with DEANO and Spermine NONOate (see figure 4). Unlike DEANO and Sperno, SNAP did not inhibit elongation of the main shafts of processes. As these processes extended, they did not branch and remained thick in diameter. It was observed, however, that SNAP did arrest the maturation and branching of lamellipodia. Carbachol and cGMP arrest growth The outgrowth exhibited by cells in PBS was almost entirely absent in carbachol- and cGMP-treated cells. Almost all neurite outgrowth was inhibited upon stimulation with carbachol (see figure 5). This result was observed both in cells that had a calcium concentration increase upon stimulation (as detected with fura-2 imaging), as well as in cells that did not. Likewise, almost all neurite outgrowth was arrested in cGMP-treated cells (see figure 6). Rapid Cell Death A few examples of rapid cell death (RCD) was observed in the experiments. This phenomenon is characterized by the speed with which the onset of dying becomes apparent and RCD takes over the fate of the cell. The initiation of RCD can be detected within two minutes - the time interval between frames. The observed RCD can be divided into two distinct categories: flat cell death and cluster cell death. Flat cell death (see figure 7) occurs to what the present study terms flat cells -flat, thin, roughly-circular cells with broad, thin lamellipodia spread out along the entire circumference. This type of RCD is characterized by the retraction of all processes and an implosion into a round, compact form. Flat cells that die in this manner are in contact with either the neurites or the soma of non-flat cells. The retraction that occurs upon death is often observed to pull at attached neurites and cell bodies. Four instances of flat cell death was observed: three in one SNAP experiment and one in a PBS control. The second type of death, cluster cell death, occurs to undifferentiated cells growing in a cloning group (see figure 8). Undifferentiated cells are defined here to be cells without processes. This type of death is characterized by a sudden bursting and bubbling, followed by continuous blebbing until the end of the time-lapse series In addition, it was found that one cluster cell death was often followed by other cluster cell deaths within the same group. Examples of cluster cell death were found in one SNAP, one Spermine NONOate, and one DEANO experiment. DISCUSSION Rapid cell death Due to the low frequency of rapid cell death, no conclusions can be drawn as to the trends that may govern its occurrence. It is possible that the two morphologically distinct types of RCD are governed by different mechanisms. Since it appears that only flat cells and undifferentiated cluster cells are subject to RCD, it is conceivable that those particular cell types possess or lack a certain quality, rendering them more susceptible. This quality may be correlated with the developmental stage, environment, treatment, and/or intercellular interactions That one cluster cell death often triggers other similar deaths in the same group suggests that there is intercellular communication between cells of the same cloning cluster. This communication may occur via gap junctions. It is also possible that the cell contents released by one cluster cell death triggers cluster cell deaths in susceptible cells in the surrounding area. Death of flat cells appears to be dependent on direct contact with the neurites and/or cell bodies of non-flat cells. This suggests that non-flat cells may cause flat cell death via a transmitted signal, and that this signal can travel through neurite and soma membranes. Concentration dependence of NO The three NO donors release nitric oxide for different lengths of time, at varied concentrations. DEANO, with a half life of 2.1 minutes, releases high concentrations of NO for a short period of time. DEANO will have released most of its nitric oxide after 15 minutes. Spermine NONOate, with a half life of 39 minutes, releases lower concentrations of nitric oxide for a longer period of time. The rate of release quickly plateaus, resulting in a steady, low rate of release. SNAP is also a long term donor, releasing even lower levels of nitric oxide for an even longer period of time Since different results were obtained using Sperno and SNAP, although both are slow-release molecules, this suggests that the effects of nitric oxide are concentration dependent. SNAP, consistently releasing lower levels of NO, inhibits the extension of thin, branching neurites from lamellipodia. In contrast, Sperno, releasing higher levels of nitric oxide, inhibits the elongation of the main shaft of processes. That cells grown in DEANO, the quick-release, high-level NO donor, and Sperno both exhibit the same changes in growth behaviors suggests that the duration of stimulus is not as crucial as the concentration of NO release. Another cholinergic pathway masks effects of nitric oxide Previous research has demonstrated that nitric oxide production is coupled with the calcium concentration increase that occurs upon stimulation of MI muscarinic receptors with carbachol [41. This study has found, however, that cells responding to carbachol with a calcium increase, and hence producing nitric oxide, exhibit neither the growth found in cells treated with nitric oxide donors nor that found in control conditions. This observation, coupled with the fact that carbachol arrested growth even in cells that did not show a calcium response, suggests that another cholinergic signal is involved in regulating neurite outgrowth, and that this signal masks the effects of nitric oxide Further evidence for this hypothesis lies in the effect of cGMP. It is widely accepted that nitric oxide activates guanylyl cyclase and consequently stimulates the production of cGMP (Moncada et al., 1991). However, the present study has shown that cells stimulated with 8-bromo-cGMP, a membrane permeable form of cGMP, demonstrate neither the growth found in cells treated with nitric oxide donors nor that seen in controls This suggests that nitric oxide regulates neurite outgrowth through a pathway other than CGMP. Previous research has shown that the NIE-115 cell line express other receptors in addition to the nitric oxide-cGMP coupled MI receptor. Yasuda et al have shown that NIE-115 cells express 15% MI and 65% M4. The M4 muscarinic receptor is coupled to the down regulation of cAMP, but not to a calcium response [14 Other research has demonstrated that intracellular cAMP promotes neurite outgrowth, expansion of lamellipodia, and motility [1. By this line of thought, stimulation of MA receptors by carbachol might arrest growth by causing a decrease in intracellular cAMP levels. It has also been shown that NIE-115 cells express nicotinic receptors [8]. Though not much is known about these receptors, it is generally accepted that they possess distinct functional properties. Their activation by carbachol may, in turn, play a role in neurite outgrowth regulation. CONCLUSIONS In studying the role of the cholinergic receptor/nitric oxide/cGMP pathway in regulating neurite outgrowth of the NIE-115 cell line, it was discovered that cells stimulated with different nitric oxide donors exhibited varied forms of growth. It was observed that cells treated with DEANO and Spermine NONOate expanded lamellipodia and demonstrated maturation of lamellipodia through neurite formation and extension. Growth of the main shafts of processes, however, was inhibited. Contrarily, it was observed that in SNAP, the main shafts of processes elongated, generally remained unbranched, and were thick in diameter. Growth of neurites from lamellipodia, however, was arrested. It was also observed that both carbachol and cGMP arrested neurite outgrowth. This study suggests that nitric oxide affects growth through a pathway other than the cGMP pathway. Evidence suggests that cholinergic stimulation activates additional pathways that mask the effects of nitric oxide, and that these other pathways play a role in inhibiting neurite outgrowth. LITERATURE CITED 1. Bolsover, S.R. et al. (1992). Intracellular cyclic AMP produces effects opposite to those of cyclic GMP and calcium on shape and motility of neuroblastoma cells. Cell Motility and the Cytoskeleton. 22:99-116. Förstermann, U. and Ishii, K (1996). Measurement of Cyclic GMP as an Indicator of Nitric Oxide Production. In Methods in Nitric Oxide Research (ed. M. Feelisch and J. S. Stamler), pp. 555-566. Chichester: John Wiley & Sons Ltd. 3. Hu, J. and El-Fakahany, E. (1993) Role of intercellular and intracellular communication by nitric oxide in coupling of muscarinic receptors to activation of guanylate cyclase in neuronal cells. J Neurochem. 61:578-585. Mathes, C. and Thompson, S. (1996). The Nitric Oxide/cGMP Pathway Couples Muscarinic Receptors to the Activation of Calcium Influx. Journal of Neuroscience. 16(5):1702-1709. Mckinney, M. et al (1990). Selective blockade of receptor-mediated cyclic GMP formation in NIE-115 neuroblastoma cells by an inhibitor of nitric oxide synthesis. Eur JPharmacol. 178:139-140. Munoz-Fernández, M.A. et al. (1994). Tumor necrosis factor- (TNF-a), Interferon-y and Interleukin-6 but not TNF-B induce differentiation of neuroblastoma cells: the role of nitric oxide. Journal of Neurochemistry. 62: 1330-1336. Nakagawa-Yagi, Y. et al. (1992). Suppressive effect of carbachol on forskolin- stimulated neurite outgrowth in human neuroblastoma NB-OKI cells. Biochemical and Biophysical Research Communications. 182(1):45-54. Oortgiesen, M. et al. (1989). Properties of Neuronal Type Acetylcholine Receptors in Voltage Clamped Mouse Neuroblastoma Cells. Neuroscience. 31(1):169-180. Peunova, N. and Enikolopov, G. (1995). Nitric oxide triggers a switch to growth arrest during differentiation of neuronal cells. Nature. 375:68-73 10. Srivatsan, M. and Peretz, B. (1997). Acetylcholinesterase promotes regeneration of neurites in cultured adult neurons of Aplysia. Neuroscience. 77(3):921-31. 11. Thompson, S.H. et al (1995). Calcium requirement for cGMP production during muscarinic activation of NIE-115 cells. Am J Physiol. 269:C979-201. 12. Tsukada, S. and Fukuda, J. (1996). Nitric oxide induced rapid neurite elongation and growthcone asymmetry of rat DRG neurons. Society for Neuroscience Abstracts. 22(1-3):739. 13. Wayne, D.B. and Skene, J.H.P. (1995). Endogenous nitric oxide modulates neurite growth in regenerating rat sensory neurons. Society for Neuroscience Abstracts. 21(1- 3):1292. 14. Yasuda, R.P. et al. (1993). Development of Antisera Selective for M4 and M5 Muscarinic Cholinergic Receptors: Distribution of M4 and M5 Receptors in Rat Brain. Molecular Pharmacology. 43(2):149-157. FIGURE LEGENDS Figure 1. Nomarski D.I.C. ÖMDR images showing neurite outgrowth in PBS. A: An image taken at start of PBS control. The cells possessed lamellipodia (a, b, c, d), as well as neurites without lamellipodia. B: After two hours of PBS treatment, the cells showed significant amount of growth. Neurites initiated and extended from lamellipodia (a, b in A), lamellipodia expanded (d in A), and processes elongated (e in A). Frequently, entire lamellipodia developed into fine, highly-branched neurites (c in A). Calibration bar in A, 20 um, and applies to A & B. Figure 2. DEANO (100 um) inhibits growth of main shafts of processes. A: An image taken at start of DEANO stimulation. B: Two hours after stimulation, lamellipodia had initiated and extended thin, branched neurites (a in A). The main shafts of processes, however, remained the same length (b in A). Calibration bar in A, 20 um, and applies to A & B. Figure 3. Spermine NONOate (100 um) inhibits growth of main shafts of processes. A: An image taken at start of Spermine NONOate stimulation. B: Two hours after stimulation, lamellipodia had initiated and extended neurites (a, b in A). The main shafts of processes, however, did not elongate (c in A). Calibration bar in A, 20 um, and applies to A & B. Figure 4. SNAP (100 um) inhibits the growth of neurites from lamellipodia. A: An image taken at start of SNAP stimulation. B: Two hours after stimulation, the main shafts of processes had elongated while remaining unbranched and thick in diameter (arrows in A). No neurite growth from lamellipodia is observed. Calibration bar in A, 20 um, and applies to A & B. Figure 5. Carbachol (1 mM during first 15 minutes of stimulation, 20 uM constant perfusion afterwards) arrests growth. A: An image taken at start of carbachol stimulation. B: Two hours after stimulation, no significant neurite outgrowth was observed. Calibration bar in A, 20 um, and applies to A & B. Figure 6. Cyclic GMP (1 mM during first 15 minutes of stimulation, 20 uM constant perfusion afterwards) arrests growth. A: An image taken at start of cGMP stimulation. B: Two hours after stimulation, no significant neurite outgrowth was observed. Calibration bar in A, 20 um, and applies to A & B. Figure 7. Flat cell death in PBS. A: An image taken at the start of PBS control: a and b point to neurites of neighboring non-flat cells, c points to the cell body of a flat cell. B: After 90 minutes of growth in PBS, neurites of neighboring non-flat cells (a, b in A) had grown to make contact with the lamellipodia ruffle of the flat cell. C: After two hours of growth in control conditions, the flat cell was dead. Note the characteristic compact, round shape of a dead flat cell. Calibration bar in A, 20 um, and applies to A-C. Figure 8. Cluster cell death in DEANO. A-B: Initiation of cluster cell death in cell 1 is evident within 2 minutes (images A and B were taken two minutes apart, at the start of DEANO stimulation). B-D: Progression of cluster cell death in cell 1 for the next 30 minutes, marked by bubbling and blebbing. E: Initiation of cluster cell death in cell 2, also marked by explosive bubbling. E-F: Progression of cluster cell death in both cells until the end of the two-hour time-lapse. Calibration bar in A, 20 um, and applies to A-F. FIGURES IVONON JNIIV