In vitro and in vivo characterization of hNT neuron neurotransmitter phenotypes

Brain Research Bulletin, Vol. 53, No. 3, pp. 263–268, 2000 Copyright © 2000 Elsevier Science Inc. Printed in the USA. All rights reserved 0361-9230/00/$–see front matter

PII S0361-9230(00)00329-4

In vitro and in vivo characterization of hNT neuron neurotransmitter phenotypes Samuel Saporta,1,2,4* Alison E. Willing,1,4 Lucy O. Colina,2 Tanja Zigova,2,4 Melissa Milliken,2 Marcel M. Daadi2,4,5 and Paul R. Sanberg2,3,4 Departments of 1Anatomy; 2Neurosurgery; 3Psychiatry, Pharmacology and Psychology; 4Neuroscience Program, University of South Florida College of Medicine, Tampa, FL, USA; and 5Layton BioScience, Inc., Sunnyvale, CA, USA [Received 19 May 2000; Accepted 1 June 2000] ABSTRACT: The hNT neuron exhibits many characteristics of neuroepithelial precursor cells, making them an excellent model to study neuronal plasticity in vitro and in vivo. These cells express a number of neurotransmitters in vitro, including dopamine, ␥-aminobutyric acid and acetylcholine. However, there have been few reports of the neurotransmitters that hNT neurons express in vivo. The present study examined whether hNT neurons express the same neurotransmitters in vivo as they do in vitro. First, the expression of tyrosine hydroxylase (TH), glutamic acid decarboxylase (GAD), choline acetyltransferase (ChAT) and the human specific nuclear marker NuMA by hNT neurons was confirmed. Nineteen normal animals were then transplanted with 80,000 hNT neurons aimed at the striatum, hippocampus or cerebral cortex. Five additional animals received injections of medium. All animals received daily intraperitoneal injections of cyclosporine (10 mg/kg) and survived 30 days. Sections through the transplants were examined for NuMA-positive hNT neurons, and for the presence of the three neurotransmitter markers: TH, GAD and ChAT. The hNT neurons were found in the striatum and cortex. Of the hNT neurons found within the rat striatum, 33% were ChAT-positive. In the cortex, only 4% of the neurons expressed ChAT. No GADpositive hNT neurons were detected at either site. No NuMApositive neurons were found in the hippocampus. The implanted hNT neurons did not induce activation of astrocytes as determined by immunocytochemistry for glial fibrillary acidic protein (GFAP). Moreover, no hNT neuron was found to express GFAP in vivo. Together, these data suggest that the hNT neurons engraft in the new host tissue, maintain their neuronal identity and may be guided in differentiation according to local environmental cues. © 2000 Elsevier Science Inc.

hNT neuron. The hNT neurons are derived from an embryonal human teratocarcinoma cell line (NT2N) following treatment with retinoic acid [1,31]. These non-mitotic cells display many characteristics of undifferentiated neuronal precursors and express the neuroepithelial marker, nestin, as well as fetal tau protein [17,23, 30,33]. Morphologically, hNT neurons kept in long-term tissue culture take on the appearance of mature neurons with welldeveloped neurites [30]. In vivo, the hNT neurons acquire markers of mature neurons such as highly phosphorylated neurofilament proteins and adult CNS tau protein [17,24,30]. However, hNT neurons develop more slowly than expected when implanted within the CNS [18,24]. For example, hNT neurons transplanted into the hippocampus and subcortical white matter were reported to have good outgrowth of axons and simple dendrites [33], while hNT neurons transplanted into the neocortex of nude mice had little neurite outgrowth [18]. Similarly, hNT neurons transplanted into the striatum of rats with ischemic stroke had little neurite outgrowth [32]. These results suggest that the maturation and phenotype of hNT neurons may be guided by the microenvironment of the area in which they are transplanted [4,17,32,33]. It is known that hNT neurons express a number of neurotransmitters in cell culture, including acetylcholine, ␥-aminobutyric acid (GABA) and dopamine [11,12,16,37,38]. However, there have been few reports of the neurotransmitter phenotype acquired by the hNT neurons in vivo [15]. In this report we confirm that cultured hNT neurons express tyrosine hydroxylase (TH), glutamic acid decarboxylase (GAD), choline acetyltransferase (ChAT) and the human specific nuclear marker NuMA and report that 30-dayold transplants of hNT neurons differentially express neurotransmitters, depending on where they are transplanted within the CNS.

KEY WORDS: Transplant, Neuronal development, Differentiation, Neurotransmitter phenotype, Local environmental cues.

MATERIALS AND METHODS

INTRODUCTION Cell Culture

Neural transplantation has become an established experimental procedure for evaluating efficacy of a variety of cells for potential therapeutic use [2,7,19], and as a tool for studying neuronal plasticity and development within the central nervous system (CNS) [17,33]. A recent neuronal cell line of interest in brain repair is the

The hNT neurons (Layton BioScience Inc., Sunnyvale, CA, USA) were thawed at 37°C until just before the last ice crystals were gone. The cells were gently transferred to a 15-ml centrifuge tube filled with 10 ml of Dulbecco’s modified Eagle’s medium

* Address for correspondence: Samuel Saporta, Ph.D., Department of Anatomy and the Neuroscience Program, University of South Florida College of Medicine, 12901 Bruce B. Downs Boulevard, Tampa, FL 33612, USA. Fax: ⫹1-(813)-974-3078; E-mail: [email protected]

263

264 (DMEM), 10% fetal bovine serum (FBS) and 0.1% Gentamicin (Sigma Chemical Co., St. Louis, MO, USA), centrifuged at 700 rpm for 7 min and resuspended in 1 ml of the DMEM/FBS media. Viability and cell number were assessed using the trypan blue dye exclusion method. The hNT neurons were then plated at a density of 106 cells/ml on eight-well culture slides (Nunc, Naperville, IN, USA). After 24 h, the medium was changed to DMEM:F12, 50 ␮g/ml Gentamicin, and 0.1% insulin-transferrin-selenium. Cell Preparation for Transplantation Vials of hNT were thawed rapidly in a 37°C water bath with occasional swirling and then transferred to a 15-ml tube containing DMEM:F12. The cells were centrifuged at 1,000 rpm for 7 min and the supernatant aspirated. The cells were resuspended in 1 ml of HBSS ⫹ 15 mM HEPES and a 1-␮l sample of the cells was removed to determine cell viability. Cell concentration was adjusted to 80,000 cells/␮l in preparation for implantation. Cells were cultured for 5 days prior to examination. Transplantation Surgery Nineteen male Sprague–Dawley rats with an average weight of 250 g (Zivic-Miller, Zelienople, PA, USA) were anesthetized with Equithesin (3.5 ml/kg), and positioned in a stereotaxic frame (Kopf, Tujunga, CA, USA). The skin over the cranium was incised, bregma visualized, and a hole drilled through the skull over the transplantation site. Eighty thousand hNT neurons contained in a 1-␮l vol. of medium were transplanted in the striatum (n ⫽ 6), hippocampus (n ⫽ 6) or parietal cortex (n ⫽ 6). Five additional animals received a 1-␮l injection of medium in the striatum (n ⫽ 2), hippocampus (n ⫽ 1) or cortex (n ⫽ 2). The coordinates used were: AP: ⫹1, ML: ⫺3.5, DV: ⫺4.9 for the striatum; AP: ⫺1.0, ML: ⫺4.5, DV: ⫺2.0 for the cortex and AP: ⫺4.2, ML: ⫺2.5, DV: ⫺3.0 for the hippocampus. All animals received daily intraperitoneal injections of cyclosporine (10 mg/kg) beginning 2 days prior to surgery and continuing throughout the duration of the study. Animals were allowed to survive for 30 days, at which time they were deeply anesthetized and perfused transcardially with saline, followed by 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.2). Brains were removed, placed in 20% sucrose in 0.1 M phosphate buffer (pH 7.2) overnight and 30-␮m frozen sections through the transplant collected in a 1-in-5 series. Care was taken to ensure that animals were adequately anesthetized during surgery and comfortable during the recovery period, according to University of South Florida IACUC guidelines and the Principles of Laboratory Animal Care (National Institutes of Health publication No. 86-23, 1985) throughout this study. Immunocytochemistry Cell cultures. Cultures were fixed with 4% paraformaldehyde for 20 min, thoroughly washed in 0.1 M phosphate-buffered saline (PBS) and then incubated in 10% normal serum of the host secondary antibody with 0.03% Triton X-100 in PBS for 1 h. Subsequently, the slides were incubated for 24 h in the same solution containing primary antibodies against TH (1:4,000; mouse monoclonal; Diasorin, Stillwater, MN, USA), GAD (1: 1,600; rabbit polyclonal; Chemicon, Temecula, CA, USA), human nuclear matrix antibody (NuMA; 1:500; mouse monoclonal; Calbiochem, La Jolla, CA, USA) or ChAT (1:500; goat polyclonal; Chemicon). After several washes with PBS, the slides were incubated in the appropriate biotinylated secondary antibody (1:300; Vector, Burlingame, CA, USA). The antibody complex was amplified using an avidin-biotin kit (ABC-Elite kit; Vector) and visualized with 3,3⬘-diaminobenzidine (DAB; ImmunoPure Metal

SAPORTA ET AL. Enhanced DAB; Pierce, Rockford, IL, USA). Finally, slides were rinsed in 0.1 M PBS and coverslipped with 95% glycerol. Deletion of primary antibodies resulted in the absence of neuronal labeling. Tissue sections. For brain tissue, sections were immunostained with TH, GAD or ChAT, as described above. In order to verify the neurotransmitter phenotype of implanted hNT neurons, a second series of sections was reacted with one of the primary antibodies, above, rinsed with 0.1 M phosphate buffer and double labeled with NuMA overnight. Binding of NuMA was detected with VIP (Vector) as the chromogen. A third set of sections from each animal was used only for NuMA immunohistochemistry, and was used to assess the number of surviving hNT neurons within the transplant. Unbiased stereology using the optical dissector method [34,35] was used to calculate the number of NuMA-positive hNT neurons in the striatum. Selected sections were reacted for glial fibrillary acidic protein (GFAP) (1:300; mouse monoclonal; Harlan, Indianapolis, IN, USA) and visualized with DAB, or double labeled for NuMA and GFAP (1:100; rabbit polyclonal; Biogenex, San Ramon, CA, USA). Primary antibodies used for double labeling were visualized with goat anti-mouse antiserum conjugated to a fluorescein-5-isothiocyanate-equivalent Alexa dye (1:200; Molecular Probes) and goat anti-rabbit antiserum conjugated to a rhodamine equivalent Alexa dye (1:800; Molecular Probes, Eugene, OR, USA), respectively. RESULTS After 5 days in vitro, the hNT neurons displayed typical neuronal morphologies with refractile cell bodies under phase contrast illumination and expressed the human specific cell marker NuMA. Immunocytochemistry with antisera to TH, GAD and ChAT detected immunoreactive cells (Fig. 1), demonstrating that cultured hNT neurons are dopaminergic, GABAergic or cholinergic. To determine whether or not these neurotransmitter phenotypes found in vitro were maintained in vivo, the hNT neurons were implanted into three different brain regions: striatum, cortex and hippocampus. After 30 days post-transplant survival time, NuMApositive hNT neurons were found in the striatum and cortex in close proximity to the graft site (Fig. 2). No NuMA-positive hNT neurons were found in the hippocampus. Transplants of hNT neurons within the striatum, tended to follow the path of the implantation needle, but were confined to the striatum. These striatial hNT neurons varied in size from 10 –18 ␮m and exhibited short neurites. More NuMA-positive cells were found within the corpus callosum than in the cortex in cortical transplants, suggesting that the implantation needle was inserted deeper than originally planned, or that hNT neurons preferentially survive in the corpus callosum as compared to neocortex. Within the corpus callosum the hNT neurons were bipolar and oriented along the trajectory of callosal fibers (data not shown). Cells that were NuMA-positive were also found on the alveus of the hippocampus. Since there was some question regarding the number of hNT neurons placed into the cortex and hippocampus, only striatal transplants were used to quantitate the number of surviving hNT neurons. Cortical hNT transplants were used for neurotransmitter immunohistochemical analysis only. The number of surviving hNT neurons in the striatum was calculated from transplants in six animals (Table 1). The mean number of striatal hNT neurons in animals was 9,722 ⫾ 204. The number of surviving hNT neurons in the striatum was approximately 12% of the total number of hNT cells transplanted. The survival of hNT neurons in striatum in this study is very similar to that reported previously [32]. Approximately 33% of NuMA-positive neurons within striatal grafts were ChAT-positive (Fig. 3), and all the ChAT-positive

IN VITRO AND IN VIVO HNT NEURON PHENOTYPES

265

FIG. 1. Multiple neurotransmitter phenotypes of hNT neurons in vitro. (A) Phase contrast photomicrograph showing the morphology of the hNT neurons after 5 days in vitro. (B) Photomicrograph of hNT neurons labeled with human nuclear matrix antibody (NuMA). This marker specifically labels cells of human origin and was subsequently used to identify transplanted hNT neurons in vivo (see Fig. 2). (C), (D) and (E) Examples of hNT neurons expressing tyrosine hydroxylase (TH), choline acetyltransferase (ChAT) and glutamic acid decarboxylase (GAD), respectively. Scale bar [shown in (E)]: 30 ␮m in (B), and 45 ␮m in (A), (C), (D), (E).

neurons within the graft were NuMA-immunoreactive. Interestingly, no GAD- or TH-positive neurons were found within the striatal grafts. In cortical grafts, 4% of the hNT neurons coexpressed ChAT and NuMA (Table 1). These grafts were not immunoreactive for GAD or TH. No NuMA-positive cells were seen in the hippocampus, though Numa-positive hNT neurons were found on the alveus of the hippocampus. None of these NuMA-positive cells were TH-, GAD- or ChAT-positive. There was no prominent astroglial activation detected within the transplant (Fig. 2). Interestingly GFAP labeling of the graft showed the presence of large, radial-orientated, glial fibers at the host-transplant interface (Fig. 2) that appear to be extended from host astrocytes. Double labeling with GFAP and NuMA confirmed that these GFAP-positive fibers were derived from host astrocytes and that no NuMA-positive hNT neurons also expressed GFAP. DISCUSSION The present study investigated the neuronal phenotype of the hNT neurons in vivo by implanting them within the hippocampus, necortex and striatum. After a 4-week post-transplantation survival time, the implanted hNT neurons survived, integrated and showed healthy cell bodies as identified by a specific marker to human derived cells, NuMA. The grafted hNT neurons did not induce a pronounced astroglial response and did not, themselves, express an

astrocytic phenotype. Interestingly, the implanted hNT neurons expressed different neurotransmitter markers, according to the area of brain in which they were placed. Within the striatum and cortex, 33% and 4% of the hNT neurons expressed ChAT, respectively. In both regions, the implanted hNT neurons were GABA- and THnegative. However, no NuMA-positive neurons could be detected in the hippocampus. The striatum and cerebral cortex have intrinsic cholinergic neurons [36], though fewer are found in the cortex than in the striatum. The hippocampus receives extensive cholinergic input, but does not have intrinsic cholinergic neurons [6,28,29]. Repressive signals are known to operate during differentiation particularly in early embryogenesis [8,18,20]. The difference between striatum and cortex in the percentage of hNT neurons expressing ChAT is puzzling and may be a result of regional differences between these two structures. It was expected that some GAD-positive hNT neurons would be found at all transplant sites. GABAergic neurons are prevalent throughout the striatum, hippocampus and cortex [14,22,25,28,29]. Hartley et al. [13] recently demonstrated that hNT neurons are capable of forming functional GABAergic synapses after 5 weeks in culture. The failure to find GAD immunoreactivity in the present study cannot be attributed to failure of the immunohistochemical reaction because the GAD antibody used in this study binds to both

266

SAPORTA ET AL.

FIG. 2. Transplanted hNT neurons within the rat striatum. (A) Photomicrograph through an hNT transplant in the striatum showing the distribution of human nuclear matrix antibody (NuMA)-positive cells. (B) Higher power photomicrograph of the graft in (A) showing NuMA-positive nuclei of hNT neurons. (C) and (D) Low and higher power photomicrographs of GFAP in an hNT transplant and at the host-transplant interface, respectively. There is no evidence of astrogliosis surrounding the transplant. (E) Extensions of glial fibrillary acidic protein (GFAP)-positive fibers from host astrocytes (red) extend through the NuMA-positive (green) hNT graft. Scale bar [shown in (E)]: 30 ␮m for (B), (D) and (E), and 360 ␮m for (A) and (C).

TABLE 1 GAD65 and GAD67 epitopes, and numerous well-marked GADpositive neurons were invariably seen in host tissue. Similarly, numerous well-marked TH-positive neurons were seen in the host hypothalamus and substantia nigra when TH immunohistochemistry was performed. hNT neurons are considered to be committed neuronal progenitor cells [16,24,31,33]. Therefore, TH and GAD expression in neuronally committed progenitor cells may be related to a common maturational pathway under the influence of specific growth factors. For example, the expression of TH can be induced in striatal GABAergic cells with acidic fibroblast growth factor and dopamine [21]. In addition, nearly 95% of TH-induced

NUMBER OF hNT NEURONS AFTER TRANSPLANTATION WITHIN THE STRIATUM AND CORTEX AND THE NUMBER OF THOSE NEURONS EXPRESSING CHOLINE ACETYLTRANSFERASE (ChAT) Transplant Site

Striatum Cortex

NuMA-IR

NuMA-IR and ChAT-IR

9672 ⫾ 223 7983 ⫾ 299

3170 ⫾ 91 308 ⫾ 73

Values are mean ⫾ SEM of transplants in six animals. NuMA-IR, NuMA-immunoreactive neurons; ChAT-IR, choline acetyltransferase-immunoreactive neurons.

IN VITRO AND IN VIVO HNT NEURON PHENOTYPES

267

FIG. 3. Transplanted hNT neurons express choline acetyltransferase (ChAT) within the rat striatum. Photomicrograph through the hNT transplant in the striatum. Numerous small hNT neurons express ChAT throughout the transplant. Arrow and arrowhead show examples of ChAT-positive hNT neurons shown at high magnification in (B) and (C). Large, mature endogenous striatal cholinergic neurons are seen in host striatum around the transplant. Scale bar [shown in (A)]: 15 ␮m for (B), (C) and 105 ␮m for (A).

stem cell-derived neuronal progeny coexpress GABA [5]. Thus, the TH/GABA differentiation pathway may be dormant or repressed under these specific in vivo conditions. Alternatively, transplanted hNT neurons may express other neurotransmitter phenotypes or require different post-transplant survival times that were not tested in this study. Delayed maturation may be a common feature of cells transplanted into the nervous system. In vitro, the hNT neurons take on a multipolar neuronal appearance within 1 week of plating and can elaborate functional synapses within 5 weeks when plated on a bed of astrocytes [13]. In vivo and in concordance with previous studies using human derived cells, the hNT neurons require at least 4 weeks to acquire neuritic processes and perhaps longer to elaborate specific neurotransmitter phenotype. Transplanted hNT neurons produce adult forms of neurofilament proteins 2– 4 months after transplantation [17]. Extensive elaboration of neurites is not seen in hNT neurons until 6 months in vivo [17,33]. Additionally, other types of transplanted cells have been shown to undergo delayed maturation in vivo. For example, fetal GABAergic neurons transplanted in the thalamus [27] and dopaminergic neurons transplanted into striatum [26] had delayed maturation times in vivo following transplantation as compared to their normal ontogenic development or maturation in vitro. Thus, longer transplant survival times may be necessary before the hNT neurons express specific neurotransmitter phenotypes. Interestingly, recent studies in our laboratory suggest that the initial period of retinoic acid treatment of the NT2N/D1 precursor cells may affect the neurotransmitter fate of hNT neurons (Zigova, T., Willing, A., Daadi, M., Saporta, S., Sanberg. P., unpublished results). We are currently

investigating the neurotransmitter phenotype fate of cultured hNT neurons subjected to different lengths of retinoic acid treatment. The present results demonstrate that hNT neurons maintain their neuronal identity after transplantation in adult rat brain and appear to differentiate according to instructive signals from the local environment. These characteristics make the hNT neurons an ideal source of neurons for reconstructive cell transplantation therapy of injured or diseased nervous system. ACKNOWLEDGEMENTS

The authors wish to thank Italo Masiello for expert assistance with digital imaging. A portion of these results was presented at the annual meeting of the American Society for Neural Transplantation and Repair, 1999. This study was supported, in part, by Layton BioScience, Inc. Drs. Saporta and Willing are consultants, and Dr. Sanberg is a consultant and Board member, for Layton BioScience, Inc.

REFERENCES 1. Andrews, P. W. Retinoic acid induces neuronal differentiation of a cloned human embryonal carcinoma cell line in vitro. Dev. Biol. 103:285–293; 1984. 2. Armstrong, R. J. E.; Svendsen, C. N. Neural stem cells: From cell biology to cell replacement. Cell Transplant. 9:139 –152; 2000. 3. Borlongan, C. V.; Tajima, Y.; Trojanowski, J. Q.; Lee, V. M.-Y.; Sanberg, P. R. Transplantation of cryopreserved human embryonal carcinoma-derived neurons (NT2N cells) promotes functional recovery in ischemic rats. Exp. Neurol. 149:310 –321; 1998. 4. Daadi, M. M.; Weiss, S. Generation of tyrosine hydroxylase-producing

268

5.

6. 7. 8. 9.

10. 11. 12.

13. 14. 15. 16. 17.

18. 19.

20.

21.

SAPORTA ET AL. neurons from precursors of the embryonic and adult forebrain. J. Neurosci. 19:4484 – 4497; 1999. Cassel, J. C.; Jackisch, R.; Duschek, M.; Hornsperger, J. M.; Richards, M. H.; Kelche, C.; Hertting, G.; Will, B. Long term effects of septohippocampal lesions and intrahippocampal grafts on acetylcholine concentration, muscarinic stimulated formation of inositol phospholipids and electrically evoked release of neurotransmitters in the rat hippocampus. Exp. Brain Res. 83:633– 642; 1991. Du, X.; Iacovitti, L. Multiple signaling pathways direct the initiation of tyrosine hydroxylase gene expression in cultured brain neurons. Mol. Brain Res. 50:1– 8; 1997. Edge, A. S. B.; Gasse, M. E.; Dinsmore, J. Xenographic cell therapy: Current progress and future development of porcine cell transplantation. Cell Transplant. 7:213–225; 1998. Faus, I.; Hsu, H. J.; Fuchs, E. Oct-6: A regulator of keratinocyte gene expression in stratified squamous epithelia. Mol. Cell Biol. 14:3263– 3275; 1994. Freeman, T. B.; Olanow, C. W.; Hauser, R. A.; Nauert, G. M.; Smith, D. A.; Borlongan, C. V.; Sanberg, P. R.; Snow, B.; Vingerhoets, F. J. G.; Calne, D.; Gauger, L. L. Bilateral fetal nigral transplantation into the post-commissural putamen as a treatment for Parkinson’s disease: Six month follow-up. Ann. Neurol. 39:379 –388; 1995. Guegan, C.; Ceballos-Picot, I.; Nicole, A.; Kato, H.; Onteniente, B.; Sola, B. Recruitment of several neuroprotective pathways after permanent focal ischemia in mice. Exp. Neurol. 154:371–380; 1998. Guillemain, I.; Patey, G.; Alonso, G.; Privat, A.; Chaudieu, I. Phenotypic and morphological characterization of the differentiated NTera2 cells. Soc. Neurosci. Abstr. 24:1527; 1998. Hardy, M.; Younkin, D.; Tang, C. M.; Pleasure, J.; Shi, Q. Y.; Williams, M.; Pleasure, D. Expression of non-NMDA glutamate receptor channel genes by clonal human neurons. J. Neurochem. 63: 482– 489; 1994. Hartley, R. S.; Margulis, M.; Fishman, P. S.; Lee, V. M.; Tang, C. M. Functional synapses are formed between human NTera2 (NT2N, hNT) neurons grown on astrocytes. J. Comp. Neurol. 407:1–10; 1999. Houser, C. R.; Esclapez, M. Localization of mRNAs encoding two forms of glutamic acid decarboxylase in the rat hippocampal formation. Hippocampus 4:530 –545; 1994. Hurlbert, M. S.; Gianani, R. I.; Hutt, C.; Freed, C. R.; Kaddis, F. G. Neural transplantation of hNT neurons for Huntington’s disease. Cell Transplant. 8:143–151; 1999. Iacovitti, L.; Stull, N. D. Expression of tyrosine hydroxylase in newly differentiated neurons from a human cell line (hNT). Neuroreport 14:1471–1474; 1997. Kleppner, S. R.; Robinson, K. A.; Trojanowski, J. Q.; Lee, V. M.-Y. Transplanted human neurons derived from a teratocarcinoma cell line (NTera-2) mature, integrate, and survive for over 1 year in the nude mouse brain. J. Comp. Neurol. 357:618 – 632; 1995. Liem, K. F. Jr.; Tremml, G.; Roelink, H.; Jessell, T. M. Dorsal differentiation of neural plate cells induced by BMP-mediated signals from epidermal ectoderm. Cell 82:969 –979; 1995. Lindvall, O.; Brundin, P.; Widner, H.; Rehncrona, S.; Gustavii, B.; Frackowiak, R.; Leemders, K. L.; Sawie, G.; Rothewell, J. C.; Marsden, C. D. Grafts of fetal dopamine neurons survive and improve motor function in Parkinson’s disease. Science 247:574 –577; 1990. Martinez, S.; Crossley, P. H.; Cobos, I.; Rubenstein, J. L.; Martin, G. R. FGF8 induces formation of an ectopic isthmic organizer and isthmocerebellar development via a repressive effect on Otx2 expression. Development 126:1189 –1200; 1999. Max, S. R.; Bossio, A.; Iacovitti, L. Co-expression of tyrosine hydroxylase and glutamic acid decarboxylase in dopamine differentiation factor-treated striatal neurons in culture. Dev. Brain Res. 91:140 –142; 1996.

22. McCormick, D. A. Neurotransmitter actions in the thalamus and cerebral cortex. J. Clin. Neurophysiol. 9:212–223; 1992. 23. Miyazono, M.; Lee, V. M.-Y.; Trojanowski, J. Q. Proliferation, cell death, and neuronal differentiation in transplanted human embryonal carcinoma (NTera-2) cells depend on the graft site in nude and severe combined immunodeficient mice. Lab. Invest. 73:273–283; 1995. 24. Miyazono, M.; Nowell, P. C.; Finan, J. L.; Lee, V. M.; Trojanowski, J. Q. Long-term integration and neuronal differentiation of human embryonal carcinoma cells (NTera-2) transplanted into the caudoputamen of nude mice. J. Comp. Neurol. 376:603– 613; 1996. 25. Mugnaini, E.; Oertel, W. H. An atlas of the distribution of GABAergic neurons and terminals in the rat. In: Bjo¨rklund, A.; Ho¨kfelt, T., eds. Handbook of chemical neuoranatomy, Vol. 4. Amsterdam: Elsevier; 1985:436 – 608. 26. Naimi, S. N.; Jeny, R.; Hantraye, P.; Peschanski, M.; Riche, D. Ontogeny of human striatal DARPP-32 neurons in fetuses and following xenografting to the adult rat brain. Exp. Neurol. 137:15–25; 1996. 27. Nothias, F.; Salin, P.; Peshanski, M.; Chesselet, M.-F. Glutamic acid decarboxylase gene expression in thalamic reticular neurons transplanted as a cell suspension in the adult thalamus. Mol. Brain Res. 20:245–253; 1993. 28. Ottersen, O. P.; Storm-Mathisen, J. Glutamate- and GABA-containing neurons in the mouse and rat brain, as demonstrated with a new immunocytochemical technique. J. Comp. Neurol. 229:374 –392; 1984. 29. Ottersen, O. P.; Storm-Mathisen, J.; Bramham, C.; Torp, R.; Laake, J.; Gundersen, V. A quantitative electron microscopic immunocytochemical study of the distribution and synaptic handling of glutamate in rat hippocampus. Prog. Brain Res. 83:99 –114; 1990. 30. Pleasure, S. J.; Lee, V. M.-Y. NTera 2 cells: A human cell line which displays characteristics expected of a human committed neuronal progenitor cell. J. Neurosci. Res. 35:585– 602; 1993. 31. Pleasure, S. J.; Page, C. D.; Lee, V. M.-Y. Pure, postmitotic polarized human neurons derived from Ntera2 cells provide a system for expressing exogenous proteins in terminally differentiated neurons. J. Neurosci. 12:1802–1815; 1992. 32. Saporta, S.; Borlongan, C. V.; Sanberg, P. R. Neural transplantation of human neuroteratocarcinoma (hNT) neurons into ischemic rats. A quantitative dose-response analysis of cell survival and behavioral recovery. Neuroscience 91:519 –525; 1999. 33. Trojanowski, J. Q.; Mantione, J. R.; Lee, J. H.; Seid, D. P.; You, T.; Inge, L. J.; Lee, V. M.-Y. Neurons derived from a human teratocarcinoma cell line establish molecular and structural polarity following transplantation into the rodent brain. Exp. Neurol. 122:283–294; 1993. 34. West, M. J. New stereological methods for counting neurons. Neurobiol. Aging 14:275–285; 1993. 35. West, M. J.; Gundersen, H. J. G. Unbiased stereological estimation of the number of neurons in the human hippocampus. J. Comp. Neurol. 29:1–22; 1990. 36. Woolf, N. J.; Butcher, L. L. Cholinergic neurons in the caudateputamen complex proper are intrinsically organized: A combined Evans blue and acetylcholinesterase analysis, Brain Res. Bull. 5:487– 507; 1981. 37. Zeller, M.; Strauss, W. L. Retinoic acid induces cholinergic differentiation of NTera 2 human embryonal carcinoma cells. Int. J. Dev. Neurosci. 13:437– 445; 1995. 38. Zigova, T.; Willing, A. E.; Borlongan, C. V.; Saporta, S.; Snable, G. L.; Sanberg, P. R. Lithium chloride induces the expression of tyrosine hydroxylase in hNT neurons. Exp. Neurol. 157:251–258; 1999.