- Email: [email protected]
An improved method for culturing cerebellar Purkinje cells with differentiated dendrites under a mixed monolayer setting
Brain Research Protocols 3 Ž1998. 192–198
Protocol
An improved method for culturing cerebellar Purkinje cells with differentiated dendrites under a mixed monolayer setting Shigeki Furuya ) , Asami Makino, Yoshio Hirabayashi Laboratory for Cellular Glycobiology, Frontier Research Program, The Institute of Physical and Chemical Research (RIKEN), 2-1 Hirosawa, Wako, Saitama 351-01, Japan Accepted 29 July 1998
Abstract We report here a novel cell culture protocol which facilitates in vitro survival and dendritic differentiation of cerebellar Purkinje cells in a monolayer, mixed culture setting. We found that the type of culture medium is a critical factor for the maintenance of these cells. Purkinje cells present in the single cell suspension of embryonic rat cerebellum were best maintained in a medium based on Dulbecco’s modified Eagle’s medium ŽDMEM.rF-12 without the addition of known neurotrophic factors. These cells maintained in DMEMrF-12based media displayed an approximately 2.5–3.5-fold increase in survival compared with cells maintained in the widely used Basal Medium Eagle’s ŽBME.-based serum-free culture medium with the same supplements. This novel protocol permits not only enhanced survival but also accelerated, improved dendritic differentiation of these cells. Purkinje cells developed highly branched spiny dendrites by 14–16 days in vitro, which matches the time course of the dendritic growth of these cells in vivo. The Purkinje cells expressed metabotropic glutamate receptor 1a in the cell bodies and branched dendrites, and the intradendritic calcium concentration increased when trans-ACPD, a selective agonist of this receptor, was applied. This novel protocol allows the development of functional branched dendrites and therefore is useful for electrophysiological and ion-imaging studies on dendrites of Purkinje cells grown in vitro. q 1998 Elsevier Science B.V. All rights reserved. Themes: Development and regeneration Topics: Cell differentiation and migration Keywords: Cerebellar Purkinje cell; Cell culture; Dendritic differentiation; Glutamate receptor; Survival
1. Type of research Ø Growing morphologically and functionally mature rat cerebellar Purkinje cells under a dispersed, mixed monolayer culture setting.
Ø Attachment of cerebellar neurons to culture plates: 2.0– 5.0 h. Ø Phenotypic and functional maturation of Purkinje cells: 2–3 weeks.
3. Materials 2. Time required 3.1. Animals Ø Dissecting fetuses and preparing single cell suspension of cerebellum: 1.5–2.0 h.
Ø Pregnant Wistar rats Žgestation day 20–21.. 3.2. Special equipment
) Corresponding author. [email protected]
Fax:
q 81-48-467-9614;
E-mail:
Ø Multiwell cell culture plates Ž12-well plates, Falcon 3043, Becton Dickinson Labware, NJ. or low fluores-
1385-299Xr98r$ - see front matter q 1998 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 5 - 2 9 9 X Ž 9 8 . 0 0 0 4 0 - 3
S. Furuya et al.r Brain Research Protocols 3 (1998) 192–198
Ø Ø Ø Ø Ø Ø Ø Ø Ø Ø Ø
cent plastic coverslips Ž13.5 mm ø, Celldesk LF, MS92132, Sumilon, Sumitomo Bakelite, Tokyo. coated with poly-L-ornithine. The coverslips are used for examination of cultures for immunofluorescent staining. Cell culture dishes Ž10 cm ø: Falcon 3003, 15 cm ø: Falcon 3025.. Membrane filter unit Ž0.22 mm. Že.g., Millex-GV, Millipore.. Microdissection forceps ŽTaxal No. 5, Fontax.. Dissection microscope. Forceps Žlarge, small.. Scissors Žlarge, small.. Disposable conical tubes ŽFalcon 2170, 2195 or equivalent.. Phase-contrast microscope. Ice bath Spatulas Žsmall.. Thermo-controlled water bath Ž33–358C..
3.3. Chemicals and reagents All solutions except for ether and ethanol should be sterilized by membrane filtration Ž0.22 mm filter unit, e.g., Sterivexe-GV, Millipore.. Ø Poly-L-ornithine stock solution Ž10 mgrml.. Ø Phosphate-buffered saline ŽpH 7.2. Ø Dissection medium: CaqqrMgqq-free Hank’s BSS ŽGibco BRL Life Technologies, a21250. containing 10 mgrml gentamicin. Ø DNase solution: 0.05% DNase Žbovine pancreas; Boehringer Mannheim. plus 12 mM MgSO4 in Caqqfree Hank’s BSS. Store at y208C. Ø Trypsin solution: 1.0% Žwrv. trypsin in Caqq, Mgqq free Hank’s BSS. Store at y208C. Ø Seeding solution: DMEMrF-12 ŽGibco BRL Life Technologies, a12400. containing 10% fetal calf serum ŽFCS, heat inactivated.. Ø Culture medium Ž serum-free DM EM rF-12 . : DMEMrF-12 ŽGibco BRL Life Technologies, a12400. containing putrescine Ž100 mM., sodium selenite Ž30 nM., glutamine Žfinal 3.9 mM. and gentamicin Ž10 mgrml.. Ø N3 supplement Ž=200 stock solution.: Progesterone Ž4 mM., bovine insulin Ž2 mgrml. ŽI-1882, Sigma. and transferrin Ž20 mgrml. ŽT-7786, Sigma.. Store aliquots Ž400–500 ml. at y808C. Ø Tri-iodothyronine ŽT3.: stock concentration: 0.5 mgrml. Store aliquots Ž50 ml. at y808C. Ø Bovine serum albumin: stock concentration: 50 mgrml. Store aliquots Ž200–500 ml. at y808C. Ø Cytosine arabinoside ŽAraC.: stock concentration: 1 mM. Store aliquots Ž200 ml. at y208C. Ø Ether to anaesthetize pregnant rats. Ø Ethanol Ž80%..
193
4. Detailed procedure 4.1. Poly-L-ornithine-coated plates or coÕerslips 4.1.1. Preparation of poly-L-ornithine solution A. Dissolve 100 mg of poly-L-ornithine ŽP-4638, Sigma. in 10 ml of Milli Q water ŽG 18 M V .. B. Sterilize using a membrane filter unit Ž0.22 mm, e.g., MILLEX-GV.. C. Dispense the solution into aliquots Ž400–500 mlrtube. and store at y208C. 4.1.2. Coating of plates or coÕerslips A. Dilute the poly-L-ornithine stock solution to concentrations of 400–500 mgrml and 10 mgrml with filtersterilized 18 M V water. B. Plate 400–500 mgrml of poly-L-ornithine solution to each well of 12 well plates Ž100 mlr22.2 mm ø well, 80 mlr13.5 mm ø plastic cover slip. and 10 mgrml of poly-L-ornithine solution to culture dishes Ž1 mlr10 cm ø: Falcon 3003.. Incubate plates and dishes in a CO 2 Ž5%. incubator overnight. C. Wash wells with filter-sterilized 18 M V water Ž3–4 times, 1 mlr22.2 mm ø well.. D. Dry dishes in a clean bench. 4.2. Preparation of cerebellar neurons 4.2.1. Dissociation of cerebellum A. Anesthetize the pregnant rat Žgestation day 20–21. with ether vapor, sterilize the abdomen with 80% ethanol solution, open the peritoneal cavity and excise the uterine horns. B. Transfer the two uterine horns to a 15 cm cell culture dish containing PBS Ž70–80 ml., remove embryos using scissors and forceps, and transfer them to a second cell culture dish Ž15 cm ø. containing dissection medium. C. Remove whole brains from the embryos and transfer them to ice-cold dissection medium Ž35 ml. in a Falcon 2170 tube. D. Pour the ice-cold CaqqrMgqq-free Hank’s BSS containing the brains onto a third cell culture dish Ž10 cm ø., and dissect cerebella from whole brains using microdissection forceps ŽTaxal No. 5. under the dissection microscope w16x. Note that the meninges should be removed completely. After the dissection of cerebella, pick up each cerebellum using forceps. Do not chop or cut the cerebella into small pieces. E. Then, transfer the cerebella to ice-cold dissection medium Ž12 ml. in a Falcon 2170 tube. Wash cerebella by centrifugation at 1000 rpm for 1–2 min. F. Wash the cerebella again in 10 ml of fresh dissection medium. G. Add 2 ml of 0.1% trypsin diluted with dissection medium to the tube and incubate the tube at 33–358C for 10 min in a thermo-controlled water bath.
194
S. Furuya et al.r Brain Research Protocols 3 (1998) 192–198
H. After incubation, add 10 ml of dissection medium to the tube and centrifuge it at 1000 rpm for 3 min. Remove supernatant carefully using an aspirator. I. Repeat step H. J. Add 1 ml of DNase solution to the tube and dissociate the trypsin-treated cerebella by gentle trituration with a Pasteur pipette. After large aggregates are dissociated, triturate them further with a Pasteur pipette equipped with a sterilized pipette tip. After the complete dissociation of aggregates of the cerebella, add 10 ml of dissection medium to the tube and then centrifuge it at 1200 rpm for 5 min. K. Suspend the cell pellet in 1–1.5 ml of seeding medium by trituration with a Pasteur pipette. To remove non-neuronal cells such as fibroblasts and glial cells, seed the single cell suspension onto a 10 mgrml poly-Lornithine-coated culture dish Ž10 cm ø, Falcon 3003., and incubate it at 20 min in a CO 2 incubator Ž378C, 5%CO 2 .. L. After incubation, recover the single cell suspension, count the number of cells and dilute the single cell suspension to a density of 5 = 10 6 cellsrml. Dilution of the cell density to below 5 = 10 6 cellsrml will cause a decrease in the viability of both Purkinje cells and granule cells. 4.2.2. Seeding and feeding the cultures A. Seed the cell suspension onto culture dishes Ž40 mlr22.2 mm ø well or 13.5 mm ø plastic cover slip.. B. Incubate cultures for at least 3 h in a CO 2 incubator Ž378C, 5% CO 2 . C. Add 0.4 ml of serum-free DMEMrF-12 containing N3 supplement Ž=1. and T3 Ž0.5 ngrml. to each well. This results in a medium containing approximately 1%
FCS. Compared to other completely serum-free media, improved growth of Purkinje cells is obtained under this condition Žsee Results.. When the serum-free condition is to be established, wash cultures in serum-free DrF and then add serum-free DrF containing N3 supplement Ž=1., T3 Ž0.5 ngrml. and bovine serum albumin Ž100 mgrml.. D. Replace half of the medium with fresh serum-free DMEMrF-12 containing the supplements and AraC Ž4 mM. at 9 and 16 DIV. 5. Results 5.1. The surÕiÕal of Purkinje cells in different culture media The maintenance of cerebellar Purkinje cells under dissociated culture conditions has been carried out to study molecular mechanisms underlying the development and physiology of Purkinje cells. In these studies, a serum-free medium based on Basal Medium Eagle’s ŽBME. w1,4,13,14,17,18,20x or serum-containing media based on Minimum Essential Medium ŽMEM. w3,8,11,15x has been used. While rat cerebellar Purkinje cells in trypsin-dissociated cerebella could be maintained using pre-existing protocols, we developed a novel culture protocol which allows Purkinje cells exhibiting improved survival and highly differentiated dendrites to be obtained w5–7x. We determined the number of Purkinje cells following maintenance in different culture media ŽFig. 1.. When cerebellar neuronal cultures were maintained in different culture media containing Fischer’s supplements including
Fig. 1. Cell survival of cerebellar Purkinje cells and other neurons in different culture media. ŽA. Purkinje cells. Cerebellar neuronal cultures Žstarting cell number: 2 = 10 5 cellsrwell. were maintained for 14 days in a BME, MEM or DMEMrF-12-based medium. At 14 DIV, cultures were fixed with 4% Žwt.rvol.. paraformaldehyde in 100 mM NaPi buffer, and stained for calbindin D-28K, a Purkinje cell-specific marker. After taking photographs of randomly selected fields, the mean number of calbindin D-28K-positive cells per mm2 was counted Žmean " S.E.M., n s 71–102.. Values were obtained from two independent experiments. ŽB. MTT assay. Cerebellar neurons Žstarting cell number: 2 = 10 5 cellsrwell. were maintained as in ŽA., and the viability of whole cerebellar neurons was assessed by MTT assay at 14 DIV. After a 4 h incubation with 0.55 mgrml of MTT at 378C, cultures were washed in PBS, and the blue formazan produced was solubilized in 1 ml of pure dimethyl sulfoxide. The optical density of the formed blue formazan was measured at 595 nm. Data are the means" S.E.M. of two independent experiments, each performed on triplicate wells. When not visible, the sizes of the error bars are less than those of the symbols.
S. Furuya et al.r Brain Research Protocols 3 (1998) 192–198
195
Fig. 2. Phenotypic growth of Purkinje cells maintained in different culture media. Primary dissociated cerebellar neuronal cultures were prepared and maintained in different culture media as described in the text. Purkinje cells were visualized by immunocytochemical staining for calbindin D-28K ŽA–G. or metabotropic glutamate receptor 1a ŽH. as described w5,7x. Antibodies for calbindin D-28K and metabotropic glutamate receptor 1a were obtained from SIGMA ŽMO. and CHEMICON ŽCA., respectively. Cultures were stained at DIV 14 ŽA–F, H. or DIV 28 ŽG.. Culture media used were serum-free BME ŽA, B., serum-free MEM ŽC., serum-free DMEMrF-12 ŽD., and DMEMrF-12 containing 1% FCS ŽE–H.. All media contained Fischer’s supplements. Bar in ŽA. indicates 100 mm for A, C, D, and E, and in B it indicates 50 mm for B, F, G and H.
196
S. Furuya et al.r Brain Research Protocols 3 (1998) 192–198
insulin, transferrin, progesterone, putrescine, selenium and T3, we observed that the in vitro growth of Purkinje cells was influenced by the culture medium itself. Primary cerebellar neuronal cultures were prepared from fetal rats at 21 days of gestation ŽE21. and maintained in serum-free BME, MEM, or DMEMrF-12-based culture medium. At 14 days in vitro ŽDIV., Purkinje cells were visualized with immunocytostaining for calbindin D-28K, a Purkinje cellspecific marker, and the number of these cells were counted. Among these serum-free culture media, the highest ratio of Purkinje cell survival was obtained using a DMEMrF12-based culture medium. The mean number of these cells increased up to 2.5-fold in cultures maintained in serumfree DMEMrF-12 medium compared with cultures in a serum-free BME-based medium ŽFig. 1A.. Furthermore, we observed that a more significant improvement in the survival was obtained by the addition of 1% fetal calf serum ŽFCS. together with 0.1 mgrml BSA instead of 1 mgrml BSA in the DMEMrF-12-based medium ŽFig. 1A.. Under this condition, the number of Purkinje cells increased 3.5-fold compared with that in a serum-free BME-based medium. The addition of 10% FCS to the medium did not support the survival of these cells Ždata not shown.. In the culture, other cerebellar neurons, such as granule cells and interneurons were present in addition to Purkinje cells. The major neuronal population was granule cells and this neuronal type consitituted ) 90% of
microtubule associated protein 2 ŽMAP2.-positive cells Ždata not shown.. Although Purkinje cells constituted only 2.5% of MAP2-positive cerebellar neurons, the percentage of Purkinje cells maintained using the present protocol was higher than that obtained using a previously described protocol w18x. Upon comparison of the viability of granule cells maintained in different medium by the tetrazolium salt Ž3-Ž4,5-dimethylthiazol-2-yl.-2,5-diphenyl tetrazolium bromide, MTT. reduction method w9x, DMEMrF-12-based media gave the highest ratio in the survival of these neurons among the culture media tested ŽFig. 1B.. Immunocytostaining of cerebellar neurons for MAP2 protein also demonstrated increased survival of granule neurons in DMEMrF-12-based media Žnot shown.. 5.2. Differentiation of functional dendrites In addition to promotion of the survival, the DMEMrF12-based medium containing 1% FCS promoted the phenotypic maturation, in particular dendritogenesis, of the neurons ŽFig. 2.. Compared to the cells grown in a serum-free BME-based medium ŽFig. 2A and B., Purkinje cells grown in the DMEMrF-12-based media developed more complicated, branched dendritic arbors by 14 days in vitro ŽFig. 2D–F.. The growth of dendrites continued for at least 30 days in vitro, and then the cells developed highly ramified dendritic branches as found in those cells grown in vivo ŽFig. 2G.. Purkinje cells failed to develop branched den-
Fig. 3. Pseudo-color images of changes in intracellular wCa2q x induced by trans-ACPD in a cultured Purkinje cell. The fluorescent calcium indicator Calcium Greene-I Ž5 mM in 1.5 M KCl, Molecular Probes, OR. was injected into the soma of a Purkinje cell grown in DMEMrF-12 containing 1% FCS for 15 DIV by pressure microinjection ŽTransjector 5246, Eppendorf.. After injection, cells were merged in artificial cerebrospinal fluid containing tetrodotoxin Ž2 mM. and perfused at 1 mlrmin. A laser scanning confocal system ŽMRC-600, Bio-Rad. equipped with an inverted was used to measure changes in fluorescence intensity following application of trans-ACPD Ž100 mM.. The cell was illuminated with an excitation wavelength of 488 nm, and fluorescent images were obtained using the time courserratiometric software module ŽBio-Rad. through a 515 nm filter. Changes in the fluorescent intensity of Calcium Greene-I were analyzed using the COMOS program ŽBio-Rad. and expressed by maximum pixel values between 1 Žred. and 256 Žwhite. with pseudo-color images ŽNIH Image, ver. 1.61.. ŽA.: before application of trans-ACPD, ŽB.: 1 min after application of trans-ACPD. Bar indicates 50 mm.
S. Furuya et al.r Brain Research Protocols 3 (1998) 192–198
drites when grown in a serum-free MEM-based medium ŽFig. 2C.. Purkinje cells maintained in our DMEMrF-12-based medium expressed metabotropic glutamate receptor 1a ŽmGluR1a . ŽFig. 2H. and ionotropic glutamate receptors 2 and 3 Ždata not shown.. The distribution of the labeling for mGluR1a was patchy and found in spine-like protrusions on dendrites. We next investigated whether these branched dendrites were functional in terms of response to a glutaminergic agonist. Extracellular application of Ž".-1aminocyclopentane-trans-1, 3-dicarboxylic acid Ž transACPD., a selective agonist for group I and group II metabotropic glutamate receptors, has been shown to elicit an increase in the intracellular concentration of Caqq in dendrites of Purkinje cells in acutely prepared slices w10x. As shown in Fig. 3, application of 100 mM trans-ACPD caused a large increase in Calcium Green-1 fluorescence intensity in the distal dendritic branches Žcompare Fig. 3A with B.. These observations indicate that distal dendrites of the cultured Purkinje cells also contain a functional signal transduction pathway which is associated with mGluR1a. Furthermore, cultured Purkinje cells maintained using this protocol show ionotropic glutamate receptor responsiveness similar to that observed in Purkinje cells in acutely prepared thin slices ŽHartell and Furuya, manuscript in preparation..
6. Discussion In this report, we have described a novel protocol that allows the maintenance of rat cerebellar Purkinje cells exhibiting a high survival ratio and well-differentiated dendrites under the condition of a mixed, dissociated culture. This protocol is based on a previously used protocol described by Fischer that allows the maintenance of mouse Purkinje cells in the presence of other cerebellar neurons such as granule cells w4,18x, and can be applied in the case of mouse Purkinje cells as well. Although the survival of Purkinje cells was improved using this protocol, we observed a gradual decrease in the number of immature Purkinje cells during the first week of culture. By 14 days in vitro, the mean number of Purkinje cells was decreased to 50–60% of the level of that at 1 day in vitro Ždata not shown.. A similar decrease in the number of Purkinje cells in the early culture period was reported in the case of mouse cerebellum under a similar monolayer condition w1x. Compared to Purkinje cells, the death of granule cells was more extensive and about 60% of granule cells die during the early culture period under low-Kq culture conditions ŽFuruya, Makino, and Hirabayashi, unpublished observation.. This might reflect the naturally occurring apoptotic death of cerebellar granule cells in vivo w19x. At present, although we do not know why immature Purkinje cells decrease gradually in number in a monolayer culture setting, one of the causes of this early gradual death might be
197
insufficient trophic support due to the absence or lower number of astroglial cells under the present conditions. Consequently, this may lead to the death of approximately 50% of Purkinje cells. 6.1. Troubleshooting One problem frequently encountered was poor cell survival and dendritic development. As reported in previous works w1,18x, we also observed that reduced cell density caused the poor survival of both Purkinje and granule cells. To avoid problems caused by reduced cell density, we maintain the mean cell density of 2.5 = 10 3 cellsrmm2 when seeding cells. Under this condition, we routinely obtain functional Purkinje cells, which exhibit spontaneous synaptic currents and glutamate receptor responsiveness. Detailed characterization of the electrophysiological properties of these cells obtained using this protocol will be described elsewhere. 6.2. AlternatiÕe and support procedures While cerebellar granule cells have been reported to play an indispensable role in the phenotypic maturation of Purkinje cell dendrites under dissociated culture conditions w1x, a granule cell-free protocol for Purkinje cell culture was reported by Brorson et al. w2x. This method is an adaptation of a method for the co-culture of hippocampal neurons with astroglial cells, and allowed the long-term survival of Purkinje cells under low-cell density conditions as a result of the trophic support of glial cells in the absence of granule cells or the addition of trophic factors. Purkinje cells grown using this method also develop, with few exceptions, physiological characteristics similar to those observed for Purkinje cells grown in vivo. Similar granule cell-free protocol was developed for chick embryonic Purkinje cells w12x. 7. Quick procedure A. Remove cerebella from embryonic rat brains on gestational day 20–21. B. Incubate cerebellar neurons in 0.1% trypsin solution. C. Dissociate cerebella by gentle trituration. D. Wash and suspend cells Ž5 = 10 6 cellsrml. in DMEMrF-12 containing 10% FCS. E. Plate cell suspension onto poly-L-ornithine-coated culture dishes. F. After 3–5 h, add serum-free DMEMrF-12 that contain supplements. G. Replace half of the medium with the fresh medium at 9 and 16 DIV. 8. Essential literature references Refs. w1,2,5,7,18x.
198
S. Furuya et al.r Brain Research Protocols 3 (1998) 192–198
Acknowledgements This work was supported by the Frontier Research Program ŽFRP. of RIKEN, and the Grants-in-Aid for Encouragement of Young Scientist ŽNo. 07780710, 08780757, 09780733 to S.F.. from the Ministry of Education, Science and Culture of Japan. We are grateful to Prof. Y. Nagai ŽFRP, RIKEN and Mitsubishi Kagaku Institute of Life Science. for encouragement of our work.
w9x
w10x
w11x
w12x
References w1x C.A. Baptista, M.E. Hatten, R. Blazeski, C.A. Mason, Cell–cell interactions influence survival and differentiation of purified Purkinje cells in vitro, Neuron 12 Ž1994. 243–260. w2x J.R. Brorson, D. Bleakman, S.J. Gibbons, R.J. Miller, The properties of intracellular calcium stores in cultured rat cerebellar neurons, J. Neurosci. 11 Ž1991. 4024–4043. w3x S. Cohen-Cory, C.F. Dreyfus, I.B. Black, NGF and excitatory neurotransmitters regulate survival and morphogenesis of cultured cerebellar Purkinje cells, J. Neurosci. 11 Ž1991. 462–471. w4x G. Fischer, Cultivation of mouse cerebellar cells in serum free, hormonally defined media: survival of neurons, Neurosci. Lett. 28 Ž1982. 325–329. w5x S. Furuya, J. Mitoma, A. Makino, Y. Hirabayashi, Ceramide and its interconvertable metabolite sphingosine function as indispensable lipid factors involved in survival and dendritic differentiation of cerebellar Purkinje cells, J. Neurochem. 71 Ž1998. 366–377. w6x S. Furuya, T. Hashikawa, Y. Hirabayashi, Restricted occurrence of an unusual ganglioside GD1a in rat brain and its possible involvement in dendritic growth of cerebellar Purkinje neurons, J. Neurosci. Res. 44 Ž1996. 73–83. w7x S. Furuya, K. Ono, Y. Hirabayashi, Sphingolipid biosynthesis is necessary for dendritic growth and survival of cerebellar Purkinje cells in culture, J. Neurochem. 65 Ž1995. 1551–1561. w8x D.L. Gruol, C.L. Franklin, Morphological and physiological differ-
w13x
w14x
w15x
w16x
w17x
w18x
w19x
w20x
entiation of Purkinje neurons in cultures of rat cerebellum, J. Neurosci. 7 Ž1987. 1271–1293. M.B. Hansen, S.E. Nielsen, K. Berg, Re-examination and further development of a precise and rapid dye method for measuring cell growthrcell kill, J. Immunol. Methods 119 Ž1989. 203–210. N.A. Hartell, Induction of cerebellar long-term depression requires activation of glutamate metabotropic receptors, NeuroReport 5 Ž1994. 913–916. P.E. Hockberger, H.Y. Tseng, J.A. Connor, Fura-2 measurements of cultured rat Purkinje neurons show dendritic localization of Ca2q influx, J. Neurosci. 9 Ž1989. 2272–2284. P.L. Jeffrey, J. Meaney, O. Tolhurst, R.P. Weinberger, Epigenetic factors controlling the development of avian Purkinje neurons, J. Neurosci. Methods 67 Ž1996. 163–175. L. Larkfors, R.M. Lindsay, R.F. Alderson, Characterization of the ¨ responses of Purkinje cells to neurotrophin treatment, J. Neurochem. 66 Ž1996. 1362–1373. D.J. Linden, M.H. Dickinson, M. Smeyne, J.A. Connor, A long-term depression of AMPA currents in cultured cerebellar Purkinje neurons, Neuron 7 Ž1991. 81–89. H.T.J. Mount, C.F. Dreyfus, I.B. Black, Neurotrophin-3 selectively increases cultured Purkinje cell survival, NeuroReport 5 Ž1994. 2497–2500. B. Rogister, G. Moonen, Microexplant cultures of the cerebellum, in: S. Fedoroff, A. Richardson ŽEds.., Protocols for Neuronal Cell Culture, 2nd edn., Humana Press, Totowa, NJ, 1997, pp. 13–23. K. Schilling, M.H. Dickinson, J.A. Connor, J.I. Morgan, Electrical activity in cerebellar cultures determines Purkinje cell dendritic growth patterns, Neuron 7 Ž1991. 891–902. A. Weber, M. Schachner, Maintenance of immunologically identified Purkinje cells from mouse cerebellum in monolayer culture, Brain Res. 311 Ž1984. 119–130. K.A. Wood, B. Dipasquale, R.J. Youle, In situ labeling of granule cells for apoptosis-associated DNA fragmentation reveals different mechanisms of cell loss in developing cerebellum, Neuron 11 Ž1993. 621–632. M. Yuzaki, D. Forrest, L.M. Verselis, S.C. Sun, T. Curran, J.A. Connor, Functional NMDA receptors are transiently active and support the survival of Purkinje cells in culture, J. Neurosci. 16 Ž1996. 4651–4661.
Protocol
An improved method for culturing cerebellar Purkinje cells with differentiated dendrites under a mixed monolayer setting Shigeki Furuya ) , Asami Makino, Yoshio Hirabayashi Laboratory for Cellular Glycobiology, Frontier Research Program, The Institute of Physical and Chemical Research (RIKEN), 2-1 Hirosawa, Wako, Saitama 351-01, Japan Accepted 29 July 1998
Abstract We report here a novel cell culture protocol which facilitates in vitro survival and dendritic differentiation of cerebellar Purkinje cells in a monolayer, mixed culture setting. We found that the type of culture medium is a critical factor for the maintenance of these cells. Purkinje cells present in the single cell suspension of embryonic rat cerebellum were best maintained in a medium based on Dulbecco’s modified Eagle’s medium ŽDMEM.rF-12 without the addition of known neurotrophic factors. These cells maintained in DMEMrF-12based media displayed an approximately 2.5–3.5-fold increase in survival compared with cells maintained in the widely used Basal Medium Eagle’s ŽBME.-based serum-free culture medium with the same supplements. This novel protocol permits not only enhanced survival but also accelerated, improved dendritic differentiation of these cells. Purkinje cells developed highly branched spiny dendrites by 14–16 days in vitro, which matches the time course of the dendritic growth of these cells in vivo. The Purkinje cells expressed metabotropic glutamate receptor 1a in the cell bodies and branched dendrites, and the intradendritic calcium concentration increased when trans-ACPD, a selective agonist of this receptor, was applied. This novel protocol allows the development of functional branched dendrites and therefore is useful for electrophysiological and ion-imaging studies on dendrites of Purkinje cells grown in vitro. q 1998 Elsevier Science B.V. All rights reserved. Themes: Development and regeneration Topics: Cell differentiation and migration Keywords: Cerebellar Purkinje cell; Cell culture; Dendritic differentiation; Glutamate receptor; Survival
1. Type of research Ø Growing morphologically and functionally mature rat cerebellar Purkinje cells under a dispersed, mixed monolayer culture setting.
Ø Attachment of cerebellar neurons to culture plates: 2.0– 5.0 h. Ø Phenotypic and functional maturation of Purkinje cells: 2–3 weeks.
3. Materials 2. Time required 3.1. Animals Ø Dissecting fetuses and preparing single cell suspension of cerebellum: 1.5–2.0 h.
Ø Pregnant Wistar rats Žgestation day 20–21.. 3.2. Special equipment
) Corresponding author. [email protected]
Fax:
q 81-48-467-9614;
E-mail:
Ø Multiwell cell culture plates Ž12-well plates, Falcon 3043, Becton Dickinson Labware, NJ. or low fluores-
1385-299Xr98r$ - see front matter q 1998 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 5 - 2 9 9 X Ž 9 8 . 0 0 0 4 0 - 3
S. Furuya et al.r Brain Research Protocols 3 (1998) 192–198
Ø Ø Ø Ø Ø Ø Ø Ø Ø Ø Ø
cent plastic coverslips Ž13.5 mm ø, Celldesk LF, MS92132, Sumilon, Sumitomo Bakelite, Tokyo. coated with poly-L-ornithine. The coverslips are used for examination of cultures for immunofluorescent staining. Cell culture dishes Ž10 cm ø: Falcon 3003, 15 cm ø: Falcon 3025.. Membrane filter unit Ž0.22 mm. Že.g., Millex-GV, Millipore.. Microdissection forceps ŽTaxal No. 5, Fontax.. Dissection microscope. Forceps Žlarge, small.. Scissors Žlarge, small.. Disposable conical tubes ŽFalcon 2170, 2195 or equivalent.. Phase-contrast microscope. Ice bath Spatulas Žsmall.. Thermo-controlled water bath Ž33–358C..
3.3. Chemicals and reagents All solutions except for ether and ethanol should be sterilized by membrane filtration Ž0.22 mm filter unit, e.g., Sterivexe-GV, Millipore.. Ø Poly-L-ornithine stock solution Ž10 mgrml.. Ø Phosphate-buffered saline ŽpH 7.2. Ø Dissection medium: CaqqrMgqq-free Hank’s BSS ŽGibco BRL Life Technologies, a21250. containing 10 mgrml gentamicin. Ø DNase solution: 0.05% DNase Žbovine pancreas; Boehringer Mannheim. plus 12 mM MgSO4 in Caqqfree Hank’s BSS. Store at y208C. Ø Trypsin solution: 1.0% Žwrv. trypsin in Caqq, Mgqq free Hank’s BSS. Store at y208C. Ø Seeding solution: DMEMrF-12 ŽGibco BRL Life Technologies, a12400. containing 10% fetal calf serum ŽFCS, heat inactivated.. Ø Culture medium Ž serum-free DM EM rF-12 . : DMEMrF-12 ŽGibco BRL Life Technologies, a12400. containing putrescine Ž100 mM., sodium selenite Ž30 nM., glutamine Žfinal 3.9 mM. and gentamicin Ž10 mgrml.. Ø N3 supplement Ž=200 stock solution.: Progesterone Ž4 mM., bovine insulin Ž2 mgrml. ŽI-1882, Sigma. and transferrin Ž20 mgrml. ŽT-7786, Sigma.. Store aliquots Ž400–500 ml. at y808C. Ø Tri-iodothyronine ŽT3.: stock concentration: 0.5 mgrml. Store aliquots Ž50 ml. at y808C. Ø Bovine serum albumin: stock concentration: 50 mgrml. Store aliquots Ž200–500 ml. at y808C. Ø Cytosine arabinoside ŽAraC.: stock concentration: 1 mM. Store aliquots Ž200 ml. at y208C. Ø Ether to anaesthetize pregnant rats. Ø Ethanol Ž80%..
193
4. Detailed procedure 4.1. Poly-L-ornithine-coated plates or coÕerslips 4.1.1. Preparation of poly-L-ornithine solution A. Dissolve 100 mg of poly-L-ornithine ŽP-4638, Sigma. in 10 ml of Milli Q water ŽG 18 M V .. B. Sterilize using a membrane filter unit Ž0.22 mm, e.g., MILLEX-GV.. C. Dispense the solution into aliquots Ž400–500 mlrtube. and store at y208C. 4.1.2. Coating of plates or coÕerslips A. Dilute the poly-L-ornithine stock solution to concentrations of 400–500 mgrml and 10 mgrml with filtersterilized 18 M V water. B. Plate 400–500 mgrml of poly-L-ornithine solution to each well of 12 well plates Ž100 mlr22.2 mm ø well, 80 mlr13.5 mm ø plastic cover slip. and 10 mgrml of poly-L-ornithine solution to culture dishes Ž1 mlr10 cm ø: Falcon 3003.. Incubate plates and dishes in a CO 2 Ž5%. incubator overnight. C. Wash wells with filter-sterilized 18 M V water Ž3–4 times, 1 mlr22.2 mm ø well.. D. Dry dishes in a clean bench. 4.2. Preparation of cerebellar neurons 4.2.1. Dissociation of cerebellum A. Anesthetize the pregnant rat Žgestation day 20–21. with ether vapor, sterilize the abdomen with 80% ethanol solution, open the peritoneal cavity and excise the uterine horns. B. Transfer the two uterine horns to a 15 cm cell culture dish containing PBS Ž70–80 ml., remove embryos using scissors and forceps, and transfer them to a second cell culture dish Ž15 cm ø. containing dissection medium. C. Remove whole brains from the embryos and transfer them to ice-cold dissection medium Ž35 ml. in a Falcon 2170 tube. D. Pour the ice-cold CaqqrMgqq-free Hank’s BSS containing the brains onto a third cell culture dish Ž10 cm ø., and dissect cerebella from whole brains using microdissection forceps ŽTaxal No. 5. under the dissection microscope w16x. Note that the meninges should be removed completely. After the dissection of cerebella, pick up each cerebellum using forceps. Do not chop or cut the cerebella into small pieces. E. Then, transfer the cerebella to ice-cold dissection medium Ž12 ml. in a Falcon 2170 tube. Wash cerebella by centrifugation at 1000 rpm for 1–2 min. F. Wash the cerebella again in 10 ml of fresh dissection medium. G. Add 2 ml of 0.1% trypsin diluted with dissection medium to the tube and incubate the tube at 33–358C for 10 min in a thermo-controlled water bath.
194
S. Furuya et al.r Brain Research Protocols 3 (1998) 192–198
H. After incubation, add 10 ml of dissection medium to the tube and centrifuge it at 1000 rpm for 3 min. Remove supernatant carefully using an aspirator. I. Repeat step H. J. Add 1 ml of DNase solution to the tube and dissociate the trypsin-treated cerebella by gentle trituration with a Pasteur pipette. After large aggregates are dissociated, triturate them further with a Pasteur pipette equipped with a sterilized pipette tip. After the complete dissociation of aggregates of the cerebella, add 10 ml of dissection medium to the tube and then centrifuge it at 1200 rpm for 5 min. K. Suspend the cell pellet in 1–1.5 ml of seeding medium by trituration with a Pasteur pipette. To remove non-neuronal cells such as fibroblasts and glial cells, seed the single cell suspension onto a 10 mgrml poly-Lornithine-coated culture dish Ž10 cm ø, Falcon 3003., and incubate it at 20 min in a CO 2 incubator Ž378C, 5%CO 2 .. L. After incubation, recover the single cell suspension, count the number of cells and dilute the single cell suspension to a density of 5 = 10 6 cellsrml. Dilution of the cell density to below 5 = 10 6 cellsrml will cause a decrease in the viability of both Purkinje cells and granule cells. 4.2.2. Seeding and feeding the cultures A. Seed the cell suspension onto culture dishes Ž40 mlr22.2 mm ø well or 13.5 mm ø plastic cover slip.. B. Incubate cultures for at least 3 h in a CO 2 incubator Ž378C, 5% CO 2 . C. Add 0.4 ml of serum-free DMEMrF-12 containing N3 supplement Ž=1. and T3 Ž0.5 ngrml. to each well. This results in a medium containing approximately 1%
FCS. Compared to other completely serum-free media, improved growth of Purkinje cells is obtained under this condition Žsee Results.. When the serum-free condition is to be established, wash cultures in serum-free DrF and then add serum-free DrF containing N3 supplement Ž=1., T3 Ž0.5 ngrml. and bovine serum albumin Ž100 mgrml.. D. Replace half of the medium with fresh serum-free DMEMrF-12 containing the supplements and AraC Ž4 mM. at 9 and 16 DIV. 5. Results 5.1. The surÕiÕal of Purkinje cells in different culture media The maintenance of cerebellar Purkinje cells under dissociated culture conditions has been carried out to study molecular mechanisms underlying the development and physiology of Purkinje cells. In these studies, a serum-free medium based on Basal Medium Eagle’s ŽBME. w1,4,13,14,17,18,20x or serum-containing media based on Minimum Essential Medium ŽMEM. w3,8,11,15x has been used. While rat cerebellar Purkinje cells in trypsin-dissociated cerebella could be maintained using pre-existing protocols, we developed a novel culture protocol which allows Purkinje cells exhibiting improved survival and highly differentiated dendrites to be obtained w5–7x. We determined the number of Purkinje cells following maintenance in different culture media ŽFig. 1.. When cerebellar neuronal cultures were maintained in different culture media containing Fischer’s supplements including
Fig. 1. Cell survival of cerebellar Purkinje cells and other neurons in different culture media. ŽA. Purkinje cells. Cerebellar neuronal cultures Žstarting cell number: 2 = 10 5 cellsrwell. were maintained for 14 days in a BME, MEM or DMEMrF-12-based medium. At 14 DIV, cultures were fixed with 4% Žwt.rvol.. paraformaldehyde in 100 mM NaPi buffer, and stained for calbindin D-28K, a Purkinje cell-specific marker. After taking photographs of randomly selected fields, the mean number of calbindin D-28K-positive cells per mm2 was counted Žmean " S.E.M., n s 71–102.. Values were obtained from two independent experiments. ŽB. MTT assay. Cerebellar neurons Žstarting cell number: 2 = 10 5 cellsrwell. were maintained as in ŽA., and the viability of whole cerebellar neurons was assessed by MTT assay at 14 DIV. After a 4 h incubation with 0.55 mgrml of MTT at 378C, cultures were washed in PBS, and the blue formazan produced was solubilized in 1 ml of pure dimethyl sulfoxide. The optical density of the formed blue formazan was measured at 595 nm. Data are the means" S.E.M. of two independent experiments, each performed on triplicate wells. When not visible, the sizes of the error bars are less than those of the symbols.
S. Furuya et al.r Brain Research Protocols 3 (1998) 192–198
195
Fig. 2. Phenotypic growth of Purkinje cells maintained in different culture media. Primary dissociated cerebellar neuronal cultures were prepared and maintained in different culture media as described in the text. Purkinje cells were visualized by immunocytochemical staining for calbindin D-28K ŽA–G. or metabotropic glutamate receptor 1a ŽH. as described w5,7x. Antibodies for calbindin D-28K and metabotropic glutamate receptor 1a were obtained from SIGMA ŽMO. and CHEMICON ŽCA., respectively. Cultures were stained at DIV 14 ŽA–F, H. or DIV 28 ŽG.. Culture media used were serum-free BME ŽA, B., serum-free MEM ŽC., serum-free DMEMrF-12 ŽD., and DMEMrF-12 containing 1% FCS ŽE–H.. All media contained Fischer’s supplements. Bar in ŽA. indicates 100 mm for A, C, D, and E, and in B it indicates 50 mm for B, F, G and H.
196
S. Furuya et al.r Brain Research Protocols 3 (1998) 192–198
insulin, transferrin, progesterone, putrescine, selenium and T3, we observed that the in vitro growth of Purkinje cells was influenced by the culture medium itself. Primary cerebellar neuronal cultures were prepared from fetal rats at 21 days of gestation ŽE21. and maintained in serum-free BME, MEM, or DMEMrF-12-based culture medium. At 14 days in vitro ŽDIV., Purkinje cells were visualized with immunocytostaining for calbindin D-28K, a Purkinje cellspecific marker, and the number of these cells were counted. Among these serum-free culture media, the highest ratio of Purkinje cell survival was obtained using a DMEMrF12-based culture medium. The mean number of these cells increased up to 2.5-fold in cultures maintained in serumfree DMEMrF-12 medium compared with cultures in a serum-free BME-based medium ŽFig. 1A.. Furthermore, we observed that a more significant improvement in the survival was obtained by the addition of 1% fetal calf serum ŽFCS. together with 0.1 mgrml BSA instead of 1 mgrml BSA in the DMEMrF-12-based medium ŽFig. 1A.. Under this condition, the number of Purkinje cells increased 3.5-fold compared with that in a serum-free BME-based medium. The addition of 10% FCS to the medium did not support the survival of these cells Ždata not shown.. In the culture, other cerebellar neurons, such as granule cells and interneurons were present in addition to Purkinje cells. The major neuronal population was granule cells and this neuronal type consitituted ) 90% of
microtubule associated protein 2 ŽMAP2.-positive cells Ždata not shown.. Although Purkinje cells constituted only 2.5% of MAP2-positive cerebellar neurons, the percentage of Purkinje cells maintained using the present protocol was higher than that obtained using a previously described protocol w18x. Upon comparison of the viability of granule cells maintained in different medium by the tetrazolium salt Ž3-Ž4,5-dimethylthiazol-2-yl.-2,5-diphenyl tetrazolium bromide, MTT. reduction method w9x, DMEMrF-12-based media gave the highest ratio in the survival of these neurons among the culture media tested ŽFig. 1B.. Immunocytostaining of cerebellar neurons for MAP2 protein also demonstrated increased survival of granule neurons in DMEMrF-12-based media Žnot shown.. 5.2. Differentiation of functional dendrites In addition to promotion of the survival, the DMEMrF12-based medium containing 1% FCS promoted the phenotypic maturation, in particular dendritogenesis, of the neurons ŽFig. 2.. Compared to the cells grown in a serum-free BME-based medium ŽFig. 2A and B., Purkinje cells grown in the DMEMrF-12-based media developed more complicated, branched dendritic arbors by 14 days in vitro ŽFig. 2D–F.. The growth of dendrites continued for at least 30 days in vitro, and then the cells developed highly ramified dendritic branches as found in those cells grown in vivo ŽFig. 2G.. Purkinje cells failed to develop branched den-
Fig. 3. Pseudo-color images of changes in intracellular wCa2q x induced by trans-ACPD in a cultured Purkinje cell. The fluorescent calcium indicator Calcium Greene-I Ž5 mM in 1.5 M KCl, Molecular Probes, OR. was injected into the soma of a Purkinje cell grown in DMEMrF-12 containing 1% FCS for 15 DIV by pressure microinjection ŽTransjector 5246, Eppendorf.. After injection, cells were merged in artificial cerebrospinal fluid containing tetrodotoxin Ž2 mM. and perfused at 1 mlrmin. A laser scanning confocal system ŽMRC-600, Bio-Rad. equipped with an inverted was used to measure changes in fluorescence intensity following application of trans-ACPD Ž100 mM.. The cell was illuminated with an excitation wavelength of 488 nm, and fluorescent images were obtained using the time courserratiometric software module ŽBio-Rad. through a 515 nm filter. Changes in the fluorescent intensity of Calcium Greene-I were analyzed using the COMOS program ŽBio-Rad. and expressed by maximum pixel values between 1 Žred. and 256 Žwhite. with pseudo-color images ŽNIH Image, ver. 1.61.. ŽA.: before application of trans-ACPD, ŽB.: 1 min after application of trans-ACPD. Bar indicates 50 mm.
S. Furuya et al.r Brain Research Protocols 3 (1998) 192–198
drites when grown in a serum-free MEM-based medium ŽFig. 2C.. Purkinje cells maintained in our DMEMrF-12-based medium expressed metabotropic glutamate receptor 1a ŽmGluR1a . ŽFig. 2H. and ionotropic glutamate receptors 2 and 3 Ždata not shown.. The distribution of the labeling for mGluR1a was patchy and found in spine-like protrusions on dendrites. We next investigated whether these branched dendrites were functional in terms of response to a glutaminergic agonist. Extracellular application of Ž".-1aminocyclopentane-trans-1, 3-dicarboxylic acid Ž transACPD., a selective agonist for group I and group II metabotropic glutamate receptors, has been shown to elicit an increase in the intracellular concentration of Caqq in dendrites of Purkinje cells in acutely prepared slices w10x. As shown in Fig. 3, application of 100 mM trans-ACPD caused a large increase in Calcium Green-1 fluorescence intensity in the distal dendritic branches Žcompare Fig. 3A with B.. These observations indicate that distal dendrites of the cultured Purkinje cells also contain a functional signal transduction pathway which is associated with mGluR1a. Furthermore, cultured Purkinje cells maintained using this protocol show ionotropic glutamate receptor responsiveness similar to that observed in Purkinje cells in acutely prepared thin slices ŽHartell and Furuya, manuscript in preparation..
6. Discussion In this report, we have described a novel protocol that allows the maintenance of rat cerebellar Purkinje cells exhibiting a high survival ratio and well-differentiated dendrites under the condition of a mixed, dissociated culture. This protocol is based on a previously used protocol described by Fischer that allows the maintenance of mouse Purkinje cells in the presence of other cerebellar neurons such as granule cells w4,18x, and can be applied in the case of mouse Purkinje cells as well. Although the survival of Purkinje cells was improved using this protocol, we observed a gradual decrease in the number of immature Purkinje cells during the first week of culture. By 14 days in vitro, the mean number of Purkinje cells was decreased to 50–60% of the level of that at 1 day in vitro Ždata not shown.. A similar decrease in the number of Purkinje cells in the early culture period was reported in the case of mouse cerebellum under a similar monolayer condition w1x. Compared to Purkinje cells, the death of granule cells was more extensive and about 60% of granule cells die during the early culture period under low-Kq culture conditions ŽFuruya, Makino, and Hirabayashi, unpublished observation.. This might reflect the naturally occurring apoptotic death of cerebellar granule cells in vivo w19x. At present, although we do not know why immature Purkinje cells decrease gradually in number in a monolayer culture setting, one of the causes of this early gradual death might be
197
insufficient trophic support due to the absence or lower number of astroglial cells under the present conditions. Consequently, this may lead to the death of approximately 50% of Purkinje cells. 6.1. Troubleshooting One problem frequently encountered was poor cell survival and dendritic development. As reported in previous works w1,18x, we also observed that reduced cell density caused the poor survival of both Purkinje and granule cells. To avoid problems caused by reduced cell density, we maintain the mean cell density of 2.5 = 10 3 cellsrmm2 when seeding cells. Under this condition, we routinely obtain functional Purkinje cells, which exhibit spontaneous synaptic currents and glutamate receptor responsiveness. Detailed characterization of the electrophysiological properties of these cells obtained using this protocol will be described elsewhere. 6.2. AlternatiÕe and support procedures While cerebellar granule cells have been reported to play an indispensable role in the phenotypic maturation of Purkinje cell dendrites under dissociated culture conditions w1x, a granule cell-free protocol for Purkinje cell culture was reported by Brorson et al. w2x. This method is an adaptation of a method for the co-culture of hippocampal neurons with astroglial cells, and allowed the long-term survival of Purkinje cells under low-cell density conditions as a result of the trophic support of glial cells in the absence of granule cells or the addition of trophic factors. Purkinje cells grown using this method also develop, with few exceptions, physiological characteristics similar to those observed for Purkinje cells grown in vivo. Similar granule cell-free protocol was developed for chick embryonic Purkinje cells w12x. 7. Quick procedure A. Remove cerebella from embryonic rat brains on gestational day 20–21. B. Incubate cerebellar neurons in 0.1% trypsin solution. C. Dissociate cerebella by gentle trituration. D. Wash and suspend cells Ž5 = 10 6 cellsrml. in DMEMrF-12 containing 10% FCS. E. Plate cell suspension onto poly-L-ornithine-coated culture dishes. F. After 3–5 h, add serum-free DMEMrF-12 that contain supplements. G. Replace half of the medium with the fresh medium at 9 and 16 DIV. 8. Essential literature references Refs. w1,2,5,7,18x.
198
S. Furuya et al.r Brain Research Protocols 3 (1998) 192–198
Acknowledgements This work was supported by the Frontier Research Program ŽFRP. of RIKEN, and the Grants-in-Aid for Encouragement of Young Scientist ŽNo. 07780710, 08780757, 09780733 to S.F.. from the Ministry of Education, Science and Culture of Japan. We are grateful to Prof. Y. Nagai ŽFRP, RIKEN and Mitsubishi Kagaku Institute of Life Science. for encouragement of our work.
w9x
w10x
w11x
w12x
References w1x C.A. Baptista, M.E. Hatten, R. Blazeski, C.A. Mason, Cell–cell interactions influence survival and differentiation of purified Purkinje cells in vitro, Neuron 12 Ž1994. 243–260. w2x J.R. Brorson, D. Bleakman, S.J. Gibbons, R.J. Miller, The properties of intracellular calcium stores in cultured rat cerebellar neurons, J. Neurosci. 11 Ž1991. 4024–4043. w3x S. Cohen-Cory, C.F. Dreyfus, I.B. Black, NGF and excitatory neurotransmitters regulate survival and morphogenesis of cultured cerebellar Purkinje cells, J. Neurosci. 11 Ž1991. 462–471. w4x G. Fischer, Cultivation of mouse cerebellar cells in serum free, hormonally defined media: survival of neurons, Neurosci. Lett. 28 Ž1982. 325–329. w5x S. Furuya, J. Mitoma, A. Makino, Y. Hirabayashi, Ceramide and its interconvertable metabolite sphingosine function as indispensable lipid factors involved in survival and dendritic differentiation of cerebellar Purkinje cells, J. Neurochem. 71 Ž1998. 366–377. w6x S. Furuya, T. Hashikawa, Y. Hirabayashi, Restricted occurrence of an unusual ganglioside GD1a in rat brain and its possible involvement in dendritic growth of cerebellar Purkinje neurons, J. Neurosci. Res. 44 Ž1996. 73–83. w7x S. Furuya, K. Ono, Y. Hirabayashi, Sphingolipid biosynthesis is necessary for dendritic growth and survival of cerebellar Purkinje cells in culture, J. Neurochem. 65 Ž1995. 1551–1561. w8x D.L. Gruol, C.L. Franklin, Morphological and physiological differ-
w13x
w14x
w15x
w16x
w17x
w18x
w19x
w20x
entiation of Purkinje neurons in cultures of rat cerebellum, J. Neurosci. 7 Ž1987. 1271–1293. M.B. Hansen, S.E. Nielsen, K. Berg, Re-examination and further development of a precise and rapid dye method for measuring cell growthrcell kill, J. Immunol. Methods 119 Ž1989. 203–210. N.A. Hartell, Induction of cerebellar long-term depression requires activation of glutamate metabotropic receptors, NeuroReport 5 Ž1994. 913–916. P.E. Hockberger, H.Y. Tseng, J.A. Connor, Fura-2 measurements of cultured rat Purkinje neurons show dendritic localization of Ca2q influx, J. Neurosci. 9 Ž1989. 2272–2284. P.L. Jeffrey, J. Meaney, O. Tolhurst, R.P. Weinberger, Epigenetic factors controlling the development of avian Purkinje neurons, J. Neurosci. Methods 67 Ž1996. 163–175. L. Larkfors, R.M. Lindsay, R.F. Alderson, Characterization of the ¨ responses of Purkinje cells to neurotrophin treatment, J. Neurochem. 66 Ž1996. 1362–1373. D.J. Linden, M.H. Dickinson, M. Smeyne, J.A. Connor, A long-term depression of AMPA currents in cultured cerebellar Purkinje neurons, Neuron 7 Ž1991. 81–89. H.T.J. Mount, C.F. Dreyfus, I.B. Black, Neurotrophin-3 selectively increases cultured Purkinje cell survival, NeuroReport 5 Ž1994. 2497–2500. B. Rogister, G. Moonen, Microexplant cultures of the cerebellum, in: S. Fedoroff, A. Richardson ŽEds.., Protocols for Neuronal Cell Culture, 2nd edn., Humana Press, Totowa, NJ, 1997, pp. 13–23. K. Schilling, M.H. Dickinson, J.A. Connor, J.I. Morgan, Electrical activity in cerebellar cultures determines Purkinje cell dendritic growth patterns, Neuron 7 Ž1991. 891–902. A. Weber, M. Schachner, Maintenance of immunologically identified Purkinje cells from mouse cerebellum in monolayer culture, Brain Res. 311 Ž1984. 119–130. K.A. Wood, B. Dipasquale, R.J. Youle, In situ labeling of granule cells for apoptosis-associated DNA fragmentation reveals different mechanisms of cell loss in developing cerebellum, Neuron 11 Ž1993. 621–632. M. Yuzaki, D. Forrest, L.M. Verselis, S.C. Sun, T. Curran, J.A. Connor, Functional NMDA receptors are transiently active and support the survival of Purkinje cells in culture, J. Neurosci. 16 Ž1996. 4651–4661.