WO2003054202A1

WO2003054202A1 - Generation of human neural crest stem cell line and its utilization in human transplantation

Info

Publication number: WO2003054202A1
Application number: PCT/US2001/013354
Authority: WO
Inventors: Seung U. Kim; Evan Y. Snyder; John H. Wolfe
Original assignee: University Of British Columbia; The Children's Medical Center Corporation; University Of Pennsylvania
Priority date: 2000-05-05
Filing date: 2001-04-25
Publication date: 2003-07-03
Also published as: AU2001257262A1

Abstract

The present invention provides human neural crest stem cells which are pluripotent, self-renewing, and can be maintained as a stable cell line via retrovirus-mediated vmyc gene transfer. These human neural crest stem cells and their daughter progeny cells are uncommitted and undifferentiated cells which can subsequently be induced to a range of alternative recognized types of differentiated and functional cells including neurons, Schwann cells, adrenal chromaffin cells, and skeletal muscle cells. These human neural crest stem cells and their progeny cells are suitable for implantation in-vivo for cell therapy treatment of human neurological disorders and diseases.

Description

GENERATION OF HUMAN NEURAL CREST STEM CELL LINE AND ITS UTILIZATION IN HUMAN TRANSPLANTATION

PROVISIONAL PATENT APPLICATION

A Provisional Patent Application enabling and descriptive of the present invention was filed with the U.S. Patent and Trademark Office as Application No. 60/155,011 on September 21, 1999.

RESEARCH SUPPORT

Work for this invention was supported by grants from the Canadian Myelin Research Initiative.

FIELD OF THE INVENTION

The present invention is concerned with the development of differentiated cells and tissues deriving from neural crest cells; and is particularly directed to the isolation, expansion, and maintenance as stable clones of human neural crest stem cells and their direct daughter progeny cells.

BACKGROUND OF THE INVENTION

The neural crest of the vertebrate embryo is the main source of the cells of the peripheral nervous system (PNS) and autonomic nervous system (ANS). The neural crest is a transitory embryonic structure arising from the lateral ridges of the neural primordium when they join mediodorsally during the closure of the neural tube. This structure has one striking developmental feature. Its cells have the capacity to undergo migration at precise periods of development along apparently definite pathways, and to settle finally in particular locations where they differentiate into a variety of cell types. At the trunk level, the PNS arises entirely from the neural crest, while in the head, ectodermal placodes (that is, thickenings of the superficial ectoderm which differentiate at variable distances from the main neural primordium, the neural plate) also participate in the formation of the sensory ganglia of certain cranial nerves.

Besides the PNS, the neural crest gives rise to a wide variety of structures. All the melanocytes of the body except those of the retinal pigmented epithelium and certain endocrine and paracrine cells (the adrenal medulla, the calcitonin-producing cells, the type I cells of the carotid bone) are of neural crest origin. Moreover, the cephalic neural crest yields mesenchymal cells that differentiate into the brain meninges, the entire facial and visceral arch skeleton and dermis, the musculo- connective wall of the large arterial trunks arising from the aortic arches, and the connective tissues of the buccal and pharyngeal region, including that of the salivary, thyroid, parathyroid, and thymus glands. Neural crest development thus involves the establishment of different cell lineages and a patterning of crest cell migration and diversification is established during embryogenesis.

In the generation of the cells of the vertebrate nervous system, and in the control of their phenotypic choices, the two best understood experimental systems are arguably the rat sympathoadrenal lineage that yields neurons and endocrine cells of the rat autonomic nervous system; and the avian lineage of neural crest cells in- vivo. Each experimental system has provided some useful information and insight. The sympathoadrenal lineage is derived from the neural crest and experimentally gives rise primarily to chromaffin cells of the adrenal gland and sympathetic neurons. These two cell types can be distinguished at a number of levels. Chromaffin cells are small cells, without significant processes, and they secrete primarily the catecholamine epinephrine into the circulation. Sympathetic neurons are much larger cells, with dendritic and axonal processes that receive and send synaptic connections, respectively. These neurons secrete primarily norepinephrine, a catecholamine compound. While both cell types store their neurotransmitters and neuropeptides in vesicles, the chromaffin vesicles are about 150-350 run in diameter, while neuronal synaptic vesicles are only 50 run in diameter. A third, apparently minor, population of cells in this lineage, known as small intensely fluorescent (SIF) cells, is somewhat intermediate in character between neurons and chromaffin cells and has vesicles of 75-120 nm in diameter. In addition to these chemical and morphological characteristics, chromaffin cells and sympathetic neurons can be distinguished by numerous molecular markers, such as genes encoding specific cytoskeletal, vesicle and surface proteins; and neurotransmitter-synthesizing enzymes such as phenyl N-methyl transferase (which is chromaffin cell specific). These markers have proven useful both in the identification of intermediate stages of differentiation and in the identification and isolation of a sympathoadrenal progenitor cell.

The step at which a multipotential neural crest cell becomes committed to the sympathoadrenal lineage is not well understood. Catecholamine-producing cells that resemble neurons or chromaffin cells have been observed in cultures of neural crest cells but these possible sympathoadrenal precursors have not as yet been isolated for further study. Committed sympathoadrenal progenitors that are at least bipotential (giving rise to either chromaffin cells or sympathetic neurons) have, however, been isolated from embryonic adrenal medullae as well as embryonic and neonatal rat sympathetic ganglia. Embryonic progenitors can be isolated by fluorescence- activated cell sorting (FACS), using several surface membrane antigens. In cell culture, these progenitors can be induced to become neurons or chromaffin cells that express cell type-specific antigens or genes, as well as the appropriate morphology and ultra-structure.

For more detailed information, empirical evidence, and reviews of the sympathoadrenal lineage system, the reader is directed to the following list of relevant and representative publications: Patterson et al.. Cell 62: 1035-1038 (1990); Stemple, D.L. and D . Anderson. Cell 71: 973-985 (1992); Anderson, D.J., Ann. Rev. Neurosci. 16: 129-158 (1993); Le Douarin, N.M., Science 231 : 1515-1522 (1986); Kim et al.. Neurosci. Res. 22: 50-59 (1989); Lo et al.. Dev. BioL 145: 139-153 (1991); lessen et al.. Neuron 12: 509-527 (1994); Anderson,

D.L., Curr. Opin. Neurobiol. 3: 8-13 (1993); and the references cited within each of these publications. The avian embryo has also provided considerable information via in-vivo and in-vitro experimental studies of the patterning of neural crest derivatives. Much of the present knowledge of the developmental potential and fate of neural crest cells comes from research studies in avian systems. Fate maps have been established in aves and provide evidence that several different crest cell derivatives may originate from the same position along the neural tube. Schwann cells, melanocytes and sensory and sympathetic neurons can all derive from the truncal region of the avian neural tube. On the other hand, some derivatives were also found to originate from specific regions of the neural crest, e.g., enteric ganglia from the vagal and sacral regions. These studies revealed that the developmental potential of the neural crest population at a given location along the neural tube is greater than its developmental fate. This also suggests that a new environment encountered by the migrating neural crest cells influences their developmental fate.

Moreover, single-cell lineage analysis in-vivo, as well as clonal analysis in- vitro, have reportedly shown that early avian neural crest cells are multipotential during or shortly after their detachment and migration from the neural tube. In avian systems, certain clones derived from single neural crest cells in culture were found to contain both catecholaminergic and pigmented cells; and that avian neural crest cells from the cephalic region could generate clones which gave rise to highly heterogeneous progeny when grown on growth-arrested fibroblast feeder cell layers. An in-vivo demonstration of the multipotency of early neural crest cells has been conducted in chickens. Individual neural crest cells, prior to their migration from the neural tube, were injected with a fluorescent dye. After 48 hours, the clonal progeny of injected cells were found to reside in many or all of the locations to which neural crest cells migrate, including sensory and sympathetic ganglia, peripheral motor nerves and the skin. Phenotypic analysis of the labelled cells revealed that at least some neural crest cells are multipotent in-vivo.

Furthermore, following migration from the neural tube, these early multipotent neural crest cells become segregated into different sublineages, which generate restricted subsets of differentiated derivatives. The mechamsms whereby neural crest cells become restricted to the various sublineages continue to be poorly understood. The fate of avian neural crest derivatives is known to be controlled in some way by the embryonic location in which their precursors come to reside. Yet the mechanism of specification for neural crest cells derivatives remains unknown to date. In culture studies described above, investigators reported that clones derived from primary neural crest cells exhibited a mixture of phenotypes. Some clones contained only one differentiated cell type whereas other clones contained many or all of the assayable crest phenotypes.

In short, the respective contributions of the avian "in embryo" and in-vitro approaches to understanding of the neural crest cell differentiating potentialities are quite limited. It is pointed out that the search for survival and proliferation factors acting locally on neural crest derivatives when they are wandering and/or settling in various embryonic locations constitutes the recurring challenge for further understanding their complex patterning and the highly diversified variety of their phenotypes.

For more detailed information, empirical data, and reviews of the avian lineage analyses for neural crest cells, the reader is directed to the following list of representative publications: Le Douarin et al., Dev. Biol. 159: 24-49 (1993); N.M. Le Douarin, Nature 286: 663-669 (1980); N.M. Le Douarin, The Neural Crest. Cambridge University Press, Cambridge, U.K., 1982; Sieber-Blum et al.. Dev. Biol. 80: 96-106 (1980); Baroffio et aL, Proc. Natl. Acad. Sci. USA 85: 5325-5329 (1988); Bronner-Fraser et al.. Neuron 3: 755-766 (1989); Cohen et al.. Dev. Biol. 46: 262-280 (1975); Dupin et al.. Proc. Natl. Acad. Sci. USA 87: 1119-1123 (1990); Duff et al.. Dev. Biol. 147: 451-459 (1991); Bronner-Frasier et al.. Nature 355: 161-164 (1988).

Another approach and line of research inquiry in this technical area has been the search to seek out, isolate and portray a "stem cell" for neurons and glia derived from the mammalian neural crest. Most illustrative of this approach are the scientific publications and issued U.S. patents of D.J. Anderson and his colleagues. In the main, these investigators have identified mammalian neural crest cells - chiefly rat and mouse cells - using antibodies to cell surface antigens; and have used sub-cloning to examine the developmental potential of these cells and their clonal progeny. Three main findings emerge from their analyses. First, some single mammalian neural crest cells are multipotent. Second, some multipotent neural crest cells generate multipotent progeny, indicating that they are capable of self- renewal and therefore are stem cells. Third, some multipotent neural crest cells can also generate some clonal progeny that form only neurons or glia, implying that the stem cells may eventually produce committed neuroblasts and glioblasts. Their experiments collectively suggest that in-vivo neural crest cells may maintain their multipotency as they migrate and proliferate; and that initial lineage decisions may occur within developing ganglia via the generation of committed blast cells. Their data also demonstrate the existence and clonal propagation of a mammalian stem cell for neurons and glia via experimental manipulation of such cells and their environment.

Some of the relevant scientific publications of Anderson and his colleagues include: D.J. Anderson, Neuron 3: 1-12 (1989); D.J. Anderson, Annu. Rev. Neurosci. 16: 129-158 (1993); Stemple, D.L. and D.J. Anderson, CeU 71: 973-985 (1992); Anderson et al.. J. Neurosci. 11: 3507-3519 (1991); Anderson, D.L. and R. Axel, Cell 42: 649-662 (1985); Anderson, D.L. and R. Axel, Cdl 47: 1079-1090 (1986); Shah et al.. Cell 85: 331-343 (1996); Lo et al. Dev. Biol. 145: 139-153 (1991).

The reader is also directed to the following issued U.S. Patents for further information: U.S. Patent Nos. 6,001,654; 5,928,947; 5,849,553; 5,693,482; 5,824,489; 5,654,183; and 5,589,376. The texts of all these issued U.S. Patents, individually and collectively, are expressly incorporated by reference herein.

It will be noted and appreciated, however, that despite this large body of reported research inquiries, printed scientific publications, and issued patents, no one has yet isolated, or expanded, or characterized a human neural crest stem cell as such; and no person has to date succeeded in creating stable cultures or clones of human neural crest stem cells which proliferate and can be maintained indefinitely under in-vitro conditions; and no person, in so far as is presently known, has demonstrated that a human neural crest stem cell implanted in-vivo can develop into several alternative and different kinds of differentiated cells in-situ, including neurons, Schwann cells, adrenal chromaffin cells, and/or skeletal muscle cells. Clearly, were such an innovation created, it would be generally seen and acknowledged as a major advance and unforeseen development in this technical field.

SUMMARY OF THE INVENTION

The present invention has multiple aspects. A first major aspect provides a primordial human neural crest stem cell and its descendant progeny cells which are suitable for implantation in-vivo into a living host subject, said primordial human neural crest stem cell comprising: a pluripotent and self-renewing neural crest stem cell of human origin which (i) carries native human genomic DNA which has not been genetically modified by human intervention means;

(ii) remains uncommitted and undifferentiated while passaged in- vitro using as a mitotic cell line;

(iii) is implantable in-vivo as an uncommitted cell; (iv) optionally migrates in-vivo after implantation from the implantation site to another anatomic site for in-vivo integration within the living host subject; (v) integrates in-situ after implantation into the body of the living host subject at a local anatomic site; and

(vi) differentiates in-situ after integration into at least one recognized type of differentiated cell of neural crest origin.

A second major aspect provides a genetically modified human neural crest stem cell and its descendant progeny cells maintained as a stable cell line in-vitro and suitable for on-demand implantation in-vivo into a living host subject, said genetically modified human neural crest stem cell comprising: a primordial neural crest stem cell of human origin which (i) remains uncommitted and undifferentiated while passaged in- vitro as a mitotic, self-renewing cell line;

(ii) is implantable in-vivo as an uncommitted cell; (iii) optionally migrates in-vivo after implantation from the implantation site to another anatomic site for integration within the body of the living host subject;

(iv) integrates in-situ after implantation into the body of the living host subject at a local anatomic site; and

(v) differentiates in-situ after integration into a recognized type of differentiated cell of neural crest origin; and human genomic DNA which has been genetically modified to include a viral vector carrying at least one DNA segment comprised of an exogenous gene coding for a specific protein product.

BRIEF DESCRIPTION OF THE FIGURES

The present invention may be more completely and easily understood when taken in conjunction with the accompanying drawing, in which:

Figs. 1A-1C illustrate the isolation, expansion and maintenance as a cultured cell line of human neural crest stem cells;

Figs. 2A-2D are photographs illustrating attributes of human neural crest stem cells; Figs. 3A-3F are photographs illustrating HNC10 human neural crest stem cells after differentiating into alternative and different functional cells;

Fig. 4 is a photograph showing the reverse transcriptase-polymerase chain reaction analysis of genes expressed by the HNC10 human neural crest stem cell line; Figs. 5A-5C are graphs illustrating the differentiation of HNC10 human neural crest stem cells into alternative types of differentiated cells after stimulation by various neurotropic factors;

Figs. 6A-6H illustrate the in-vivo changes caused by lacZ-expressing HNC10 human neural crest stem cells which were implanted into newborn mouse brains using an intra- ventricular implantation technique and determined readily by their blue color reaction; and Figs. 7A-7D illustrate the results of HNC10 human neural crest stem cells which were implanted into myelin mutant shaker rat brains four weeks after implantation in-vivo.

DETAILED DESCRIPTION OF THE INVENTION

The present invention constitutes the isolation, expansion, and maintenance of human neural crest stem cells and their direct descendent progeny cells as a pure cloned culture; and the use of such cloned human neural crest cells as therapeutic implants when placed at a prechosen anatomic site in a living human or animal host. The generation and in-vitro culture of purified clones constituted of human neural crest stem cells and their direct daughter progeny cells is unique in this field. These stem cells and their direct progeny are medically suitable and biologically compatible for use as cellular implants in humans - particularly those afflicted with debilitating neurological disorders and diseases.

Among the many benefits and major advantages provided by the present invention are the following:

1. When human neural crest stem cells are implanted into pathological myelin- deficient areas of the CNS, their differential potential is strongly influenced by the environmental signals at the site of implantation. Thus, the implanted cells are useful for treating a variety of diseases characterized by profuse white matter degeneration - that would benefit by the replacement of myelin-forming cells. When neural crest stem cells are implanted into different areas of the developing nervous system, they generate progeny that become myelin-forming cells in that area.

2. Cell therapy has become a most promising strategy for the treatment of many human diseases including neurological disorders. The objective of cell therapy is to replace lost cells and restore the function of damaged cells and tissues. Transplantation of renewable, homogenous and well-characterized human neural crest stem cells into the damaged target tissue or organ will replace lost cells and should restore damaged function. These stem cells and their direct progeny are ideal cells to serve as donor cells for cell therapy in various neurological diseases including motor and sensory neuropathy, multiple sclerosis, Parkinson disease, Huntington disease, amyotrophic lateral sclerosis (ALS), spinal cord injury, stroke, Duchenne muscular dystrophy and pain control.

3. The stable immortalized human neural crest stem cell line described here can be expanded readily and provide a renewable and homogeneous population of glial, neuronal and muscle cells. These cells are most valuable for future research studies of fundamental questions in developmental neurobiology, cell and gene therapies, and development of new therapeutic drugs .

A considerable number of conventionally known abbreviations and minimal designations are used by practitioners in this art as jargon. To aid the reader in understanding the detailed information and description presented hereinafter, a summary of such terms is given below.

Abbreviation Proper title/name or Designation and nomenclature

A2B5 A clone of monoclonal antibody specifically react to

GT1 ganglioside. p75NGFR Low affinity nerve growth factor receptor. FORSE-1 A cell surface molecule specific in earliest stage of neural tube formation and a marker for neural stem cells.

PSA-NCAM Polysialic acid - neural cell adhesion molecule, a cell type marker for neural progenitor cells.

NG2 A monoclonal antibody recognizes glycoprotein; similar to PSA-NCAM.

NF-L; -M; -H Neurofilament triplet proteins; low (L), medium (M) and high (H) molecular weight protein.

01; 04 Clones of monoclonal antibodies specifically react to surface antigens of oligodendrocytes.

GFAP Glial fibrillary acidic protein. trk A/B/C Cellular receptors specific for neurotrophins, NGF,

BDNF, NT-3 and NT-4/5.

MBP Myelin basic protein. B7-2 Co-stimulatory factor specific for antigen presenting cells (APC) and a marker for microglia.

PO Protein zero, a myelin protein specifically found in the

PNS, and a marker for Schwann cells.

S100 Protein found in Schwann cells. FGF-2 Fibroblast growth factor-2, also called basic fibroblast growth factor.

NGF Nerve growth factor. BDGF Brain derived growth factor. Abbreviation Proper title/name or Designation and nomenclature

GGF Neuregulin/glial growth factor.

NT-3 Neurotrophin-3.

PDGF BB Platelet derived growth factor BB isoform.

TGF-βl Transforming growth factor-βl.

BMP2 Bone morphogenic protein-2.

CNPase Cyclicnucleotide phosphodiesterase, a cell-type specific marker for oligodendrocytes.

TH Tyrosine hydroxylase.

DRG Dorsal root ganglia.

NCSC Neural crest stem cell(s).

PNS Peripheral nervous system.

ANS Autonomic nervous system.

CNS Central nervous system.

The disclosure and detailed description of the present invention is presented as an incongruous approach and contrasting position to the conventional opinions and ordinary expectations of practitioners working in this technical field. The present invention often contravenes and stands opposite to conventional views and positions; and provides many striking examples of differences and distinctions of cell qualities and cell attributes not previously recognized or appreciated. For these reasons, among others, it is deemed both valuable and useful to provide the reader first with the novel neural crest cell paradigm and organized system of neural crest cellular development which underlies and supports the mode and manner by which the newly cloned human stem cells yield functional and completely differentiated cells.

I. The Neural Crest Cell Model And System Of Cellular Development The substantive value and real significance of the present invention can only be properly recognized and truly appreciated in the context of the model and system of cellular development that these unique cells evidence and embody. Much confusion, misleading views, ambiguity and inconsistency has been reported and unfortunately is the overall result to date of the different investigative attempts to elucidate and specify the stages of developmental and the various differentiated outcomes originating with and from the primordial neural crest stem cell. As evidenced and illustrated by the experiments and empirical data described herein, the present invention - for the first time - is able to present an organized system of developmental stages and different cell lineages and forms which originate with a single human neural crest stem cell and continue until a fully differentiated, phenotype cell is yielded. The model system of neural crest cellular origins and development is given by Flow Scheme A below.

As shown in Flow Scheme A, the original embryonic source of all these cells is the neural crest stem cell. Such stem cells are: (i) uncommitted and undifferentiated cells; (ii) pluripotent cells having an unlimited proliferation capacity; and (iii) are able to self-renew and self-maintain their existence when replicating by producing two daughter progeny cells, one of which becomes a self- renewed stem cell and the other becoming a direct and true descendent cell, now designated a "pre-progenitor cell". The daughter self-renewed stem cell is identical to and indistinguishable from its parent cell. However, the daughter pre-progenitor cell is markedly different from the parent stem cell. In comparison to its parent, the pre-progenitor cell has a large, but limited proliferation capacity. This direct descendent daughter cell is itself a multipotent cell which is and remains uncommitted and undifferentiated as such. However, the rate of proliferation for the daughter pre-progenitor cells is much greater than its stem cell parent. Stem cells are not found in abundance in any tissue, embryonic or adult. The pre-progenitor cell, however, proliferates rapidly during its limited number of reproductive cycles. Also the progeny of the single pre-progenitor cell becomes altered and thus these limited number of daughter cells are now termed "primary progenitor cells".

The dominant characteristic property and attribute of the "primary progenitor cells" is their ability to become committed to a single cell lineage and line of development. It is therefore at this third stage of neural crest cell descendancy that the multipotency aspect and capacity of the ancestor cells becomes lost forever; and that the primary progenitor cell is the stage which becomes influenced by external stimuli and chemical signals in the local environment such that an irreversible commitment is made to one category of cell lineage and cell embodiment type. As shown by Flow Scheme A, the model system shows that not less than four (4) separate and individual cell pathway lineages exist. Each lineage provides for its own family tree and pedigree; and each pathway provides at least one outcome and typically several different possible formats of differentiation. The primary progenitor cell is also able to reproduce itself, presumably both before and after commitment to a separate cell lineage; but its proliferation capacity is believed to be markedly restricted and limited in comparison to its ancestors. Once cell lineage commitment occurs in the primary progenitor cell, the development of the cells continues within carefully controlled and pre-selected pathway limits. Also, cell differentiation as such occurs only after a prior commitment to a set cell lineage has been made. At these latter stages of development, the particular form and phenotypic properties of the cell are decided; and a completely differentiated cell subsequently emerges as the full and final embodiment in the progression of events from early, to middle, to late stages of cell differentiation - as these have been classically identified and reviewed in the scientific literature. The nature of signals and molecules which determine cell lineage or cell fate are currently virtually unknown.

The human neural crest stem cells and their progeny cells - the pre- progenitor cells and the primary progenitor cells - in this model system of cellular development are the cells constituting and comprising the present invention. As disclosed and described in full hereinafter, these cells are unique in their pluripotent properties; are uncommitted and undifferentiated cells; and are demonstrably able to be implanted in-vivo and yet provide a range of fully differentiated cells which differ in function, in morphology, and in phenotypic cell properties.

II. Isolation, Expansion, And Immortalization Of Human Neural Crest Stem Cells

Isolation

Dissociated cell cultures were established from human embryonic dorsal root ganglia (DRG) of 15 week gestation by trypsin treatment as described previously by Kim et al.. L Neurosci. Res. 22: 50-59 (1989). DRG cultures grown for 1-2 weeks were consisted of small (10 μm in diameter) or larger (15 μm in diameter) nerve cells in singles or clusters, more numerous spindle shaped (15-20 μm in length) Schwann cells, flat polyclonal fibroblasts and a small number of neural crest stem cells [see Fig. 1A]. DRG cultures with 80% confluency were subjected to retro virally mediated transduction of vmyc oncogene and subsequent cloning. Retrovirus-Mediated Gene Transfer

Two different xenotropic, replication-incompetent retroviral vectors were used to infect human NCSCs. The retroviral vector used for transducing NCSCs with vmyc oncogene was an amphotropic replication-incompetent retroviral vector encoding vmyc (transcribed from the LTR plus neo transcribed from an internal SV40 early promoter) which permitted the propagation of human NCSC clones by genetic means. And also confirmation of the monoclonal origin of all progeny. This amphotropic vector was generated using the ecotropic retroviral vector encoding vmyc to infect the GP+envAM12 amphotropic packaging lines [Snyder et aL , Ceil 68: 1-20 (1992); Ryder et_aL , Neurobiol. 21 : 356-375 (1990);

Markowitz et al.. Virology 167: 400-406 (1988)] . The second retroviral vector encoding lacZ transcribed from the viral long terminal repeat (LTR) plus neo transcribed from an internal SV40 early promoter similar to the BAG vector [Snyder et al.. Cell 68: 1-20 (1992)]. This vector provided a stable, histochemically- and immuno-detectable genetic marker for transplantation experiments. Successful infectants were selected and expanded. Supernatants from these new producer cells contained replication-incompetent retroviral particles bearing an amphotropic envelope at a titer of 4xl0⁵ CFUs which efficiently infected the human neural cells as indicated by G418-resistance. No helper or replication-competent recombinant viral particles were produced.

Clones of Immortalized Human Neural Crest Stem Cells

After 7-14 days of G418 selection, ten G418-resistant clones, HNC10/A2, Bl, B3, C2, D3, D4, E2, F5, FI, and G6 were isolated and expanded. The cloned NCSCs were tripolar or multipolar in morphology with 10-15 μm in size [see Fig. IB]. HNC10/C2, one of these expanded clones, was subjected to elaborate study. In the proliferative growth condition, the clone HNC10/C2 exhibited a doubling time of about 24 hr [see Fig. 1C].

Characteristics of the HNC10 Clone of Immortalized Cells

The HNC10/C2 cell line grows in culture in serum-free, chemically-defined medium as single cells or clusters that can be subcultured and passaged weekly. The cells are typically grown on coverslips and can be processed for immunocytochemical staining to reveal cell type specific markers.

A summary of the cell-type specific markers and histochemical characteristics for this stable clone of human neural crest stem cells is given by Table 1 below. A more complete description of the experiments and empirical data supporting the information provided by Table 1 is presented hereinafter. The content and formulation of the serum-free, chemically-defined medium in which these cells were grown and maintained is given by Table 1A.

Table 1: Immortalized Hi iman Neural Crest Stem Cell Characteristics

Presence (H-) Meaning/Value

Marker or of Marker Substance (Type) Absence (- Indicator Detected human mitochondria'⁰ (+) human cell origin cell organelle pan-myc oncoprotein²⁾ (+) contains vmvc cellular oncoprotein nestin⁽³⁾ (+) NCSC cytoskeletal protein p75NGFR⁽⁴⁾ (+) NCSC surface receptor antigen vimentin⁽⁵⁾ (+) NCSC cytoskeletal protein

A2B5⁽⁶⁾ (+) NCSC surface antigen

Musashi⁽⁷⁾ (+) neural progenitor transcription cells factor

PORSE-l⁽⁸⁾ (+) neural progenitor surface antigen cells

PSA-NCAM⁽⁹⁾ (+) neural progenitor surface antigen cells

NG2 proteoglycan⁽¹⁰⁾ (+) progenitor cells/ surface antigen glial precursor cells

NF-L^(,,) (+) early stage neuron specific neuron marker

NF-M^(,,) (+) middle stage neuron specific neuron marker

NF-H^(,,) (+) differentiated neuron specific neuron marker β tubulin^(I1) (-) differentiated neuron specific neuron marker isotype III peripherin^(1I) (-) mature peripheral PNS specific neuron nervous system neuron marker trkA^(l,) (-) neuron neurotrophin (NGF) receptor protein

OiOO (-) oligodendrocyte specific oligodendrocyte marker

04^(1I) (-) oligodendrocyte specific oligodendrocyte marker

GFAP^(II) (-) astrocyte cell specific astrocyte marker Table 1: Immortalized Human Neural Crest Stem Cell Characteristics (continued)

Presence (+) Meaning/Value

Marker or of Marker Substance (Type) Absence (-) Indicator Detected

MBP^(,,) oligodendrocyte specific oligodendrite marker

B7-2^(,,) microglia specific marker PO™ Schwann cell specific Schwann cell marker trkB/trkC^(I,) special classes of neurotrophin receptor neurons protein

S100^(,,,) glial cell mature Schwann cell marker chromogranin^(IV> adrenal chromaffin cell chromaffin cell specific marker desmin^(V) skeletal muscle cytoskeletal protein cell myosin^(V) skeletal muscle cytoskeletal protein cell

Table 1 References:

1. Flax et aL, Nature Biotech 16: 1033-1039 (1998). 2. Spotts, R. and B. Hann, MpL Cell BioL 10: 3952-3956 (1990). 3. Lendahl et al.. Cell 60: 585-595 (1990); Stemple, D.L. and D.J. Anderson, Cell 71: 973-

985 (1992); Friedman et al.. J. Comp. Neurol. 295: 43-51 (1990); Hockfield et al.. L

Neurosci. 5: 3310-3328 (1985); Reynolds et al.. Science 255: 1707-1710 (1992).

4 Chao, M.V., Neuron 9: 583-593 (1992); Morrison et_aL, Cell 96: 737-749 (1999). 5 Houle, J. and S. Fedoroff. Dev. Brain Res. 9: 189-195 (1983). 6 Eisenbarth et_ah, Proc. Nat. Acad. Sci. USA 46: 4913-4917 (1979). 7 Sakakibara, S. and H. Okano, J. Neurosci. 17: 8300-8312 (1997).

Tole, S. and P.H. Patterson, J. Neurosci. 15: 970-980 (1995).

9 Grinspan, J.B. and B. Franceschino, Neurosci. Res. 41: 540-551 (1995).

10, Stallcup, W.B. and L. Beaslev. J. Neurosci. 7: 2737-2744 (1987).

II. See Table 2 references.

Ill See Table 3 references.

IV See Table 4 references.

V. See Table 5 references. Table 1A: UBC1 Serum-Free Chemically-Defined Medium

[mg/liter] [mg/liter] albumin 1000 - 4000 aluminum chloride. 6H₂0 0.001 - 0.01

D-biotin 0.015 - 0.7 ammonium metavanadate 0.0006 - 0.0012 ethanolamine 10 - 50 barium chloride 0.001 - 0.003 galactose 800 - 1000 cobalt chloride. 6H₂0 0.001 - 0.003 linoleic acid¹ 0.04 - 0.12 chromic potassium sulfate 0.001 - 0.003 oleic acid¹ 0.04 - 0.12 cupric sulfate. 5H₂0 0.0012 - 0.005 putrescine. 2H₂0 0.08 - 0.32 ferrous sulfate. 7H₂0 0.42 - 0.84 pyruvate² 55 - 220 germanium dioxide 0.005 - 0.01 retinal acetate¹ 0.01 - 0.05 lithium chloride 0.01 - 0.02 vitamin B 12 0.34 - 1.36 molybdic acid. 2H₂0 0.0001 - 0.0002 ascorbic acid 20 - 100 nickel nitrate. 6H₂0 0.0001 - 0.0002 catalase 20 - 100 rubidium chloride 0.00001 - 0.00002 glutathone, reduced 0.05 - 2.0 silver chloride 0.0000044 superoxide dismutase 20 - 100 sodium selenite 0.03 - 0.3 tocopherol acetate 0.025 - 0.5 stannous chloride 0.0001 - 0.0003

HEPES 2900 - 6600 titanium oxide 0.001 - 0.003 zinc sulfate. 7H₂0 0.43 - 1.29 insulin/human³ 5 - 10 transferrin/human 5 - 10 hydrocortisone 0.2 - 0.4 progesterone 0.013 - 0.026 triiodofhyronine 0.02 - 0.06

1. Water-soluble forms of linoleic/oleic acids and retinal acetate should be used.

2. Pyruvate should be omitted when DMEM is used as a basal medium.

3. Insulin/ 10 me ; should be dissolved in 1 ml of 0.1 N Hcl and then add 1 ml H,0.

As summarized by Table 1 above, the immortalized clone of HNC10 cells were immunoreaction-positive with an antibody against human mitochondria and also for pan-myc oncoprotein indicating that HNC10 cells are uniquely of human origin and also contain a copy or multiple copies of vmyc-encoding retrovirus (as determined relative to positive and negative controls run in parallel). Almost all HNC10 cells tested were immunoreaction-positive for nestin and p75NGFR indicating that HNC10 cell line is indeed constituted of neural crest stem cells. Other cell-type markers for neural crest stem cells, vimentin and A2B5, were also demonstrated in HNC10 cells. In addition, the HNC10 cells were also immunoreaction-positive for Musashi, a transcription factor determining differentiation of neural progenitor cells; FORSE-1, a surface antigen marker for neural progenitor cells; PSA-NCAM, a surface antigen marker for neural progenitors and also for oligodendrocyte-type 2 astrocyte (O-2A) pre-progenitor cells; and NG2 proteoglycan, a surface antigen marker for neural progenitor cells. These data, therefore, have established for the first time that the HNC10 cell lines (immortalized via retrovirus-mediated vmyc transfer and initially derived from embryonic DRG) are truly and properly the self-renewing and multipotent human neural crest stem cells. Never before, insofar as is presently known, has a human neural crest stem cell been isolated, expanded and maintained as a stable cell line; and, moreover, no immortalized neural crest stem cell clone has been developed and characterized in depth to demonstrate and prove its human origins and stem cell properties.

II. Lineage Commitment Of Human Neural Crest Stem Cells Human neural crest cells and their progeny cells can commit to at least four different cell lineages when cultured in-vitro using media containing 5 % fetal bovine serum and 5 % horse serum. A variety of different immunological and histochemical markers have been identified in the earlier published scientific literature which offers reliable test assay [surface antigen phenotypes] procedures by which to separate and distinguish different cell types among the possible choices of cell lineage development. The experiments and empirical data unequivocally demonstrating a commitment to a specific cell lineage and the undifferentiated nature of such committed cells are described in detail hereinafter. However, the sum and substance of these differences is given in summary form by Tables 2-5 respectively below.

Table 2: Neuronal Cell Lineage Characteristics

Absence (-) Meaning or

Marker or Value of Substance

(Type) Presence (+) of Indicator Detected

NF-L⁽¹⁾ (+) early stage neuron neurofilament protein- specific neuron marker

NF-M (+) middle stage neuron neurofilament protein- specific neuron marker

NF-H (+) mature neuron neurofilament protein- specific neuron marker

β tubulin⁽²⁾ (+) mature neuron specific neuron marker isotype III peripherin⁽³⁾ (+) mature peripheral specific neuron marker nervous system neuron trkA⁽⁴⁾ ⁽+⁾ sensory and receptor protein specific sympathetic neurons for NGF trkB/trkC⁽⁵⁾ (-) special classes of receptor protein specific neurons for BDNF (trkB) or NT-3 (trkC)

Table 2 References:

1. Julien et al., Biochem. Biophvs. Acta. 909: 10-20 (1987); Lee et al., EMBO 1 7: 1947-1955 (1988); Myers et al., EMBO J. 6: 1617-1626 (1987).

2. Ferreira, A. and A. Caceres, Neurosci. Res. 32: 516-529 (1992).

3. Parysek, L.M. and R.D. Goldman, Neurosci. 8: 555-563 (1988); Gorham et al.. Dev. Brain Res. 57: 235-248 (1990).

4. Chao, M.V. , Neuron 9: 583-587 (1992).

5. Obermeyer et al., L BioL Chem. 268: 22963-67 (1993). Table 3: Glial Cell Lineage Characteristics

Absence (-) Meaning or Marker or Value of Substance (Type) Presence (+) Indicator Detected glial fibrillary⁽¹⁾ (+) astrocyte and cellular protein of acidic protein Schwann cell astrocyte and Schwann cells (GFAP)

S100⁽²⁾ (+) Schwann cells Schwann cell specific marker

Ol^{3) Θ oligodendrocyte oligodendrocyte specific marker

O4⁽⁴⁾ (-) oligodendrocyte oligodendrocyte specific marker

MBP⁽⁵⁾ (-) oligodendrocyte oligodendrocyte specific marker p₀(6) (+) Schwann cell Schwann cell specific marker

B7-2⁽⁷⁾ (-) microglia microglia specific marker

Table 3 References

1. Jessen et al.. Neuron 12: 509-527 (1994).

2. Jessen et al.. Development 109: 91-103 (1990).

3. Sommer. L. and M. Schachner. Dev. Biol. 83: 311-327 (1981).

4. Schachner et al., Dev. Biol. 83: 328-338 (1981).

5. Sternberger et al.. Proc. Nat. Acad. Sci. USA 75: 2521-2524 (1978).

6. Jessen et al.. Neuron 12: 509-527 (1994).

7. Satoh et al.. Brain Res. 704: 92-96 (1995). Table 4: Lineage Characteristics of Adrenal Chromaffin Cells

Absence (-) Meaning or

Marker or Value of Substance (Type) Presence (-) Indicator Detected

chromogranin⁽¹⁾ (+) adrenal chromaffin cell surface cell marker tyrosine hydroxylase⁽²⁾ (+) sympathetic neurons specific of the autonomic enzyme nervous system

PNMT⁽³⁾ adrenal chromaffin cell marker cell

Table 4 References

1. Wilson, B.S. and R.V. Lloyd, Arm L Pathol. 115: 458-468 (1984).

2. Pickel et aL, Proc. Nat. Acad. Sci. USA 72: 659-663 (1975).

3. Connett, R. and N. Kirschner, BioL Chem. 245: 329-334 (1970).

Table 5: Lineage Characteristics of Skeletal Muscle Cells

Absence (+) Meaning or

Marker or Value of Substance

(Type) Presence (-) Indicator Detected

morphologically flat, (+) myoblasts physical elongated, multinucleated appearance cells⁽¹⁾ of cells desmin⁽²⁾ (+) skeletal cellular muscle cell filament

myosin⁽³⁾ (+) skeletal cellular muscle cell filament

Table 5 References

1. Le Douarin et aL, Dev. BioL 159: 24-49 (1993).

2. Naumann, K. and D. Pette, Differentiation 55: 203-211 (1994).

3. Ibid.

Thus, as summarized by Tables 2-5 respectively, neurons differentiated from HNC10 cells expressed neurofilament proteins, NF-L, NF-M and NF-H; and the vast majority of neurofilament-positive cells co-expressed β tubulin isotype III, a cell type specific marker for neurons as well as peripherin, a marker of mature peripheral nervous system neurons. These phenotypes are specifically and exclusively expressed by the human cells indicating that HNC10 cells can and do commit to and differentiate into nerve cells under the serum-containing culture condition.

HNC10 cells also differentiated into glial cells when grown in serum- containing medium, and expressed glial fibrillary acidic protein (GFAP), a cell type- specific marker for Schwann cells. These lineage cells also were immunoreaction- positive for S100, and PO cell type-specific markers for mature Schwann cells. These lineage cells were, however, immunoreaction-negative for 01 and 04, both cell type specific markers for oligodendrocytes. In addition, HNC10 cells grown in serum-containing medium for 2 weeks yielded lineage cells which were immunoreaction-positive for chromogranin and PNMT, two specific markers for adrenal chromaffin cells and were reaction positive for tyrosine hydroxylase, a marker for sympathetic neurons of the autonomic nervous system lineage. These results indicate that HNC10 cells are sympathoadrenal stem cells which are capable of differentiation into sympathetic neuroblasts and adrenal chromaffin cells.

Lastly, after 1-2 weeks in serum-containing medium, HNC10 stem cells gave rise to large, flat, elongated, multinucleated cells, and many of these multinucleated cells expressed desmin- and myosin-immunoreaction. These results indicate that HNC10 cells, human NCSCs, can commit to and differentiate into skeletal muscle cells. III. Promotion and Avoidance of Specific Cell Lineages by Individual Growth Factors

In the embryonic stage of human development, NCSCs migrate from the dorsal aspect of the neural tube and subsequently differentiate into a variety of cell types found in different embryonic locations. A variety of transplantation and cell culture studies support the view that the fate of pluripotent neural crest stem cells is determined by the local environment signals exerted in-vivo; and that at least some of such local environmental signals are created and provided by a range of growth factors such as fibroblast growth factor (FGF), transforming growth factor (TGF) and neuregulin/glial growth factor (GGF).

The human neural crest stem cells and their direct descendent progeny cells (the precursor progenitor cells and the primary progenitor cells) - both as non- transformed cells and stable immortalized cells maintained in culture - are demonstrably influenced in different ways by a diverse range of different growth factors. A representative listing of empirically evaluated growth factors and their actions is given by Table 6 below.

Table 6: Environmental Factors and Signals

Promotes Blocks/Avoids

Growth Differentiation Differentiation

Factors Of Of

FGF-2 (bFGF) neurons; Schwann cells; skeletal muscle cells

NGF Schwann cells;

(100 ng/ml) skeletal muscle cells

BDNF Schwann cells;

(10 ng/ml) skeletal muscle cells

NT-3 Schwann cells;

(10 ng/ml) skeletal muscle cells

PDGF-BB Schwann cells skeletal muscle cells

(10 ng/ml) neuregulin αl Schwann cells; neuronal cells

(10 ng/ml) skeletal muscle cells neuregulin β2 Schwann cells; neuronal cells

(10 ng/ml) skeletal muscle cells

TGF-βl Schwann cells; neuronal cells

(10 ng/ml) skeletal muscle cells

These findings indicate that the choice of each of several alternative fates available to NCSCs can be instructively promoted by different environmental signals including those growth factors previously reported by various investigators. In the present study, it was found that FGF-2 promotes differentiation of HNC10 cells into three cell types, i.e. , neurons, Schwann cells and skeletal muscle cells. Neuregulin and TGFβl induced HNC10 cells to differentiate into Schwann cells and skeletal muscle cells and those growth factors blocked neuronal differentiation of HNC10 cells. Members of the neurotrophin family, NGF, BDNF and NT3 also favored differentiation of HNC10 cells into glial and skeletal muscle cells. These findings also showed that trophic effect was not specific for a particular neurotrophin; and that the mechanism actually in effect involves activation of the non-selective neurotrophin receptor, p75NGFR, by one of these neurotrophins. The results further indicate that neurotrophins and other pertinent growth factors affect differentiation of HNC10 cells at multiple levels for each and different lineage of cells. Moreover, these results reveal that concerted action of combinations of growth factors is important in determining fate of the NCSCs rather than action of a single factor.

IV. In- Vivo Capabilities And Properties Of Human Neural Crest Stem Cells And Their Progeny

The isolated human neural crest stem cells maintained in culture as a cloned colony and their progeny cells - collectively, the precursor progenitor cells and primary progenitor cells - remain uncommitted and undifferentiated cells while passaged in-vitro. This collection of cells, both stem cells and progeny cells, are suitable for on-demand implantation in-vivo, preferably at a prechosen anatomic site, into a living host subject, human or animal.

It will be noted and appreciated that at the time of in-vivo implantation, these cells are uncommitted to a specific cell lineage of development and remain undifferentiated as to particular cell form or phenotype. All such cell lineage commitment and differentiation into cell form and phenotype occurs only after the cells are implanted in-vivo. Moreover, after implantation into a living host at a known anatomic site, the neural crest stem cells and their progenitor progeny cells optionally can migrate to one or more other anatomic sites. Such optional migration capabilities are empirically proven by the experimental data provided hereinafter. Subsequently, whether at the original implantation site or another anatomic location post-migration, the implanted cells will integrate with the particular cells and tissues then present and existing at that particular location. Such tissue integration brings the neural crest stem cells and their progenitor progeny cells into effective contact with and influence by the local environment - including such local ligands, activated cells, stimulating agents, immunlogically reactive substances, and enzymatically active compositions as are present in or may be conveyed to that local environment. Among such chemical entities are various growth factors, cytokines, circulating antibodies and transitory cells, as well as a range of fluids carrying nutrients, minerals, co-factors and other reactive compositions, compounds and ligands which typically serve as local environmental signals, stimulating agents, and/or modulating moieties.

As a consequence of integration at an identifiable anatomic site and the influence and effects of signals and interactions, caused or provided by the milieu and setting of the local environment, the now integrated neural crest stem cells and progenitor progeny cells become committed to one particular cell lineage and form of cellular development in-situ; and become differentiated cells in-situ consistent with a particular cell lineage. Thus, the local environment will control and dictate what differentiated cell form and phenotype will appear in functional form in the local anatomic area. This is clearly demonstrated and evidenced by the experiments and empirical data described hereinafter by Experimental Series II.

Accordingly, the implanted and integrated neural crest stem cells and their progenitor progeny cells become committed and differentiated in-situ as Schwann cells, peripheral nerve cells, skeletal muscle cells, adrenal chromaffin cells, as well as the other differentiated cell forms and phenotypes shown by Flow Scheme A previously herein. Each of these individual and different cell forms and phenotypes lies within the potential capacities and properties of the HNC10 series of cloned cells; and each clone of cells after implantation in-vivo will become integrated at and differentiated by the environmental and development signals, ligands, and cells presented at or conveyed to the local anatomic site.

V. Experiments and Empirical Data

The following is a summation presenting the sum and substance of many different in-vitro and in-vivo experiments. For ease of understanding and presentation clarity of information purposes, a lengthy recitation of the pertinent methods and materials background is given. It will be recognized that many of the individual in-vitro test and assay techniques and protocols are conventionally described in the scientific and medical/clinical published literature. The empirical studies are recited in substantive detail.

Materials and Methods:

UBC1 (a serum-free, chemically-defined medium):

HEPES 2900 - 6600 titanium oxide 0.001 - 0.003 zinc sulfate. 7H₂0 0.43 - 1.29 insulin/human³ 5 - 10 transferrin/human 5 - 10 hydrocortisone 0.2 - 0.4 progesterone 0.013 - 0.026 triiodothyronine 0.02 - 0.06

2. Pyruvate should be omitted when DMEM is used as a basal medium.

3. Insulin/ 10 m; g should be dissolved in 1 ml of 0.1 N Hcl and then add 1 ml H₂0.

Primary DRG cell culture

Thirty pairs of spinal dorsal root ganglia (DRG) were isolated from a 15 week gestational human embryo and dissociated into single cells by incubating in phosphate buffer saline (PBS) containing 0.25% collagenase (Type CLS, Wothington Biochemical, Lakewood, NJ) and 40 μg/ml DNase I (Sigma) for 1 hour at 37°C [Kim et al.. L Neurosci. Sci. 22: 50-59 (1989)]. Dissociated cells (5X10⁵ cells/ml) suspended in culture medium consisted of Dullbecco's modified Eagle medium (DMEM) containing 5% fetal bovine serum, 5% horse serum, 0.5% glucose, 20 μg/ml gentamicin and 2.5 μg/ml amphotericin B, were plated on poly- L-lysine (10 μg/ml)-coated 6 well dishes (Falcon). Two days later, medium was switched to serum-free culture medium consisted of DMEM containing UBC1 supplements (contain human insulin 5 μg/ml, human transferrin 10 μg/ml, sodium selenite 40 nM, hydrocortisone 40 nM, triiodothyronine 3 nM and other nutrients and anti-oxidants), 0.5% glucose, 20 μg/ml gentamicin, 2.5 μg/ml amphotericin B and FGF-2 (10 ng/ml, Peprotech) [Kim et aL , Brain Res. 275: 79-86 (1983)] . Culture medium was changed twice a week. DRG cultures grown for 1-3 weeks were consisted of DRG neurons, Schwann cells and fibroblasts and were used for gene transfer experiments. The permission to use embryonic tissue was granted by the Clinical Research Screening Committee involving Human Subjects of the University of British Columbia.

Retrovirus-mediated gene transfer into human DRG cells

An amphotropic replication-incompetent retroviral vector encoding vmyc (transcribed from the retrovirus LTR plus neo transcribed from an internal SV40 early promotor) not only permitted the propagation of human NCSC clones by genetic means, but also enabled confirmation of the monoclonal origin of all progeny. This amphotropic vector was generated using the ecotropic retroviral vector encoding vmyc, as described earlier for generating murine NSC clone C17-2 [Snyder et al.. Cell 68: 33-51 (1992)] to infect the GP+envAM12 amphotropic packaging line [Markowitz et al.. Virology 167: 400-406 (1988)]. Successful infectants were selected and expanded. Supernatants from these new producer cells contained replication-incompetent retroviral particles which efficiently infected the human neural cells as indicated by G418 resistance. Infection of human DRG cells in 6 well plates was performed three times by the following procedures: 2 ml of supernatant (4X10⁵ CFUs) from the packaging cell line and 8 μg/ml polybrene (Aldrich/Sigma) was added to target cells in 6 well plates and incubated for 4 hr at 37°C; the medium was then replaced with fresh growth medium; infection was repeated 24 hr and 48 hr later. Seventy two hr after the third infection, infected cells were selected with G418 (250 μg/ml, Sigma) in growth medium for 7-14 days and large clusters of clonal cells were individually isolated and grown in PL-coated 6 well plates. Individual clones were generated by limited dilution and propagated further. At this phase of isolation, individual clones were designated as human NCSC lines, HNC10. One of the HNC10 clones, HNC10/C2, was propagated further and investigated for its in-vitro characteristics. HNC10/C2 clone is referred to simply as an HNC10 line. One of the HNC10 clones was subsequently transduced with a retroviral vector encoding lacZ gene and puromycin-resistant gene, enabling transfected cells to produce β-galactosidase (β-gal) constitutively.

Determination of doubling time

HNC10 cells at the concentration of 5 X 10⁴ cells were plated in 30 mm dishes with 2.5 ml of medium. After 12-72 hr of growth, cells were exposed to 0.1 % trypsin in PBS for 5 min at 37°C, collected by centrifugation at 1200 rpm for 8 min, and resuspended in 0.5 ml PBS. Using a hemocytometer, cell number was counted 12, 24, 36, 48, 60 and 72 hr after plating under a Nikon inverted microscope.

Immunochemical characterization of human NCSCs

Immunochemical determination of cell type specific markers in HNC10 cells was performed as following: HNC10 cells were grown on PL-coated Aclar plastic coverslips (9 mm in diameter) with serum-free medium for 3-7 days, fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (PB, pH 7.4) for 3 min at room temperature (RT), washed twice with PBS, and incubated with antibodies specific for A2B5 (specific for GT1 ganglioside; mouse IgM monoclonal, American Type Culture Collection/ATCC, Rockville, MD), human p75NGFR (mouse IgG monoclonal, ATCC), FORSE-1 (mouse IgG monoclonal, ATCC), NG2 (specific for transmembrane chondroitin sulfate proteoglycan; rabbit polyclonal, Chemicon) and PSA-NCAM (specific for embryonic polysialylated form of neural cell adhesion molecule; mouse IgM monoclonal, Pharmingen). For cytoplasmic antigen staining, coverslips bearing cells were fixed in cold acid alcohol (5 % acetic acid/95 % ethanol) for 15 min at -20°C, incubated with heat-inactivated goat serum (1:10) for 30 min at room temperature before the primary antibodies were applied, in order to block any potential interaction between Fc receptors and Fc fragments. The cells were incubated in antibodies specific for nestin (rabbit polyclonal), and vimentin (mouse IgG monoclonal, Sigma) for 2 days at 4°C. Cells incubated with primary antibodies were followed by biotinylated secondary antibodies and avidin-biotin hydrogen peroxidase (ABC) immunochemical processing (Vector) and visualized with AEC (Sigma) chromogen development.

For immunochemical characterization of differentiated cell types the following antibodies were utilized: For neurons: neurofilament protein-L (NF-L, mouse IgG monoclonal, Sigma), NF-M (mouse IgG monoclonal, Sigma), NF-H (mouse IgG monoclonal, Sigma), microtuble associated protein-2 (MAP2, mouse IgG monoclonal, Sigma), neuron specific enolase (NSE, rabbit polyclonal, DAKO), β tubulin isotype III (mouse IgG monoclonal, Sigma) and peripherin (rabbit polyclonal, Chemicon). For Schwann cells: S-100 A/B (rabbit polyclonal, DAKO). For oligodendrocytes: CNPase (mouse IgG monoclonal, Sigma), galactocerebroside (GalC, mouse IgG monoclonal produced in our laboratory) and MBP (rabbit polyclonal, DAKO). For myoblast/ myotube: desmin (rabbit polyclonal, DAKO) and skeletal muscle myosin (rabbit polyclonal, Chemicon). For sympathicoadrenal lineage cells: tyrosine hydroxylase (rabbit polyclonal, pel-freeze), PNMT (rabbit polyclonal, Chemicon), and chromogranin (rabbit polyclonal, Dr. R. Angeletti).

For double-immunofluorescence of cell-surface and cytoplasmic antigens, the cells were fixed in 4% paraformaldehyde in 0.1 M PB for 3 min at RT, incubated with heat-inactivated normal goat serum (1:10) for 30 min at RT before the primary antibodies were applied as described above. Cells incubated with primary antibodies were followed by fluorescein isothiocyanate (FITC)-conjugated anti- mouse IgM, IgG or anti-rabbit IgG (Cappel, West Chester, PA). After several washes in PBS, cells were fixed in cold acid-alcohol for 15 min. Cells were washed in PBS and then incubated in second primary antibodies followed by rhodamine- conjugated second antibodies as described above. After several washes, coverslips were mounted on slides with gelvatol and examined under a Zeiss Universal microscope equipped with phase contrast, fluorescein and rhodamin optics.

Cellular differentiation of human NCSCs

To differentiate the immortalized cells, the cells were plated onto PL coated Aclar plastic coverslips (9 mm diameter) and then grown in serum-free medium (DMEM containing UBC1 supplements) for 4 days and then switched to differentiation medium. The differentiation medium consisted of DMEM containing 10% fetal bovine serum, 20 μg/ml gentamicin and 2.5 μg/ml amphoterine B. In order to study the effect of growth factors on differentiation, the following serum- free medium was utilized: DMEM containing UBC1 supplements, 20 μg/ml gentamicin, 2.5 μg/ml amphotericin B, and one of the following factors: NGF (100 ng/ml), BDNF (10 ng/ml), NT-3 (10 ng/ml), PDGF- BB (10 ng/ml), neuregulin αl (10 ng/ml), neuregulin β2 (10 ng/ml) and TGF-βl (10 ng/ml).

RT-PCR analysis Total RNA preparation was performed according to the acid guanidinium/ phenol/chloroform (AGPC) method previously described. Five ml of total RNA from each sample was subjected to DNase treatment and then processed for the first strand cDNA synthesis using Moloney murine leukemia virus (M-MLV) reverse transcriptase (GIBCO-BRL). Five ml of each cDNA products was amplified by PCR using the specific sense and antisense primers designed from the cDNA sequence for each marker gene for human CNS cells. The cDNA from HNC10 cells was amplified for human nestin and human p75NGFR as a marker gene for stem cells, human NF-L, -M, -H as marker genes for neurons, human MBP as a marker gene for oligodendrocytes, human GFAP as a marker for astrocytes, human P0 as a marker gene for a peripheral myelin protein [Lee et al., Mol. Cell Neurosci. 8: 336-350 (1997)], human B7-2 as a marker for microglia, and human trkA, B, C as markers for subtypes of neurotrophin receptors. The following primers were used with the expected PCR product length in base pair (bp): nestin sense: 5'-LCTCTGACCTG TC AGAAGAAT-3 ' antisense: 5'-ACGCTGACACTTACAGAT-3' (316 bp) first p75NGFR sense: 5'-LCTCACACCGGGGATGTG-3' antisense: 5'-LGTGGGCCTTGTGGCCTAC-3' nested primers for the second p75NGFR sense: 5'-TGTGGCCTACATAGCCTTC- 3' antisense: ATGTGGCAG- TGGACTCACT-3 ' (476 bp) NF-L sense : 5 ' -TCCTACTAC ACC AGCC ATGT-3 ' antisense: 5 * -TCCCC AGCACCTTCAACTTT-3 ' (284 bp)

NF-M sense: 5'-TGGG- AAATGGCTCGTCATTT-3 ' antisense: 5'-CTTCATGGAAGCGGCCAATT-3' (333 bp) NF-H sense: 5'-CTGGACGCTGAGCTGAGGAA-3' antisense: 5'-CAGTCACTTCTTCAGTCACT-3' (316 bp) MBP sense: 5 ' -ACACGGGCATCCTTGACTCCATCGG-3 ' antisense: 5'-TCCGGAACCAGGTGGGTTTTCAGCG-3' (510 bp) GFAP sense: 5'-GCAGAGATGATGGAGCTCAATGACC-3' antisense: 5',GTTTCATCCTGGAGCTTCTGCCTCA-3' (266 bp) PO sense: 5'-TTCTGGTCCAGTGAGTGGGTCTCAG-3' antisense: 5- TCACTGTAGTCTAGGTTGTGTATGA-3 ' (209 bp)

B7-2 sense: CTC- TTTGTGATGGCCTTCCTG-3' antisense: 5'-CTTAGGTTCTGGGTAACCGTG-3'(464 bp) first TrkA sense: 5'-CCA- TCGTGAAGAGTGGTCTC-3 antisense: 5 ' -GGTGACATTGGCCAGGGTCA-3 ' nested primers for the second TrkA sense: 5 ' -CCGTTTCGTGGCGCCAGATG-3 ' antisense: 5'-GCCCCAGGGATGGCAGACCC-3' (438 bp) first TrkB sense: 5'-AAGACCCTGAAGGATGCCAG-3' antisense: 5 ' -AGTAGTCAGTGCTGTACACG-3 ' nested primers for the second TrkB sense: 5'-CAATGCACGCAAGGACTTCC-3' antisense: 5'-TCCCGGGACATCCCAAAGTC-3' (364 bp) first TrkC sense: 5'-CTACAACCTCAGCCCGACCA-3' antisense: S'-GCTGTAG-ACATCTCTGGACA-S' nested primers for the second TrkC sense: 5'-AGGACAAGATGCTTGTGGCT-3' antisense: 5'-TGCCGAAGTCCCCAATCTTC-3' (308 bp)

PCR was carried out in a 50 μl of reaction mixture containing Taq DNA polymerase buffer (20 mM Tris-HCl, pH 8.4, 50 mM KCl, 200 M d NTP, 2.5 mM MgCl₂, 1 mM of each primer) and 2.5 U Taq DNA polymerase (GIBCO-BRL). The main amplification program consisted of a denaturation step at 94°C for 1 min followed by an annealing step at 55~57°C for 1 min, and a synthesis step at 74°C for 1 min, for 35-40 cycles. For the initial amplification, cDNA samples were denatured at 94°C for 5 min, annealed at 55~57°C for 1 min, and extended at 74°C for 3 min.

Brain transplantation study

Two different sets of these experiments (newborn mice or shaker rats) were performed. First sets of experiments, heads of cryoanaesthetized postnatal day 0 CD1 mice (n = 24) were transilluminated with a fiber optics and 2 μl volume containing 5X10⁴ cells/ml plus 0.05% w/v trypan blue in PBS were gently expelled into each lateral ventricle through a finely drawn glass micropipette. The distribution of trypan blue confirmed filling of the ventricular system from as far rostral as the olfactory bulbs to as far caudal as the IVth ventricle and the rostral central spinal canal [Snyder et aL. Nature 374: 367-370 (1995)]. All of these animals were sacrificed at various intervals between one and 4 weeks after transplantation. For second sets of experiments, Long Evans shaker rats (4 weeks old, n = 32) were anesthetized with ketamine hydrochloride (MTC Pharmaceuticals, 35 mg/kg) and xylazine (Bayer Canada, 5 mg/kg) and held in a stereotaxic apparatus with ear bars. A finely drawn glass micropipette containing five μl PBS with 5X10⁴ HNClO/lacZ cells/ml were slowly injected into the right caudate-putamen at 1.0 mm anterior, 3.0 mm lateral from bregma and 4.0 mm below the brain surface. Left side striatal tissues of implanted animals served as built-in negative controls. All implanted shaker rats received daily cyclosporin A (Novartis Canada, 10 mg/kg) given intraperitoneally. All of these animals were sacrificed between one and 4 weeks after transplantation. The operated mice or shaker rats were sacrificed at various intervals after transplantation. Under deep ketamine hydrochloride (35 mg/kg) and xylazine (5 mg/kg) anesthesia, each animal was perfused through the left cardiac ventricle with PBS followed by a fixative consisted of 0.1 M PB containing 4% paraformaldehyde, 0.1 % glutaraldehyde and 0.2% picric acid. The brains were removed from the skull and postfixed in the same fixative for 24 hours at 4°C. After cyroprotection with 0.1 M PB containing 15% sucrose, 20 μm sections were cut on a freezing microtome and collected in 0.1 M PBS and 0.1 % sodium azide. Sections were stored in the same solution until staining was conducted as free-floating sections. Engrafted cells were then detected by X-gal (5-bromo-4-chloro-3-indolyl-D- galactoside) histochemistry and/or an anti- β-gal antibody (mouse monoclonal IgG, Cappel) to demonstrate previously implanted β -gal-positive HNC10 (HNClO/lacZ) cells. For X-gal histochemistry cyrosections were stained by immersion in PBS containing 5 mM K₃Fe (CN)₆, 5 mM K₄Fe(CN)₆, 2 mM MgCl₂, 0.01 % sodium deoxycholate, 0.02% Nonidet P-40, and 2 mg/ml X-gal at 37°C overnight. Sections were washed with PBS and placed onto gelatin-coated slides. Both X-gal and β-gal staining methods detect lac-Z positive HNClO/lac-Z cells. Cell type identity of grafted HNC10 cells was also established by staining with following cell type specific markers: for neurons: MAP-2; for oligodendrocytes: CNPase and MBP; for astrocytes: GFAP. Immunohistochemistry was performed on coronal, free-floating sections. The sections were first incubated with 0.1 M PBS containing 5% normal goat serum and 0.3% Triton X-100. After washing, the sections were treated for 30 min at RT with 0.5% hydrogen peroxide to block endogenous peroxidase, and then rinsed three times for 10 min each in PBS. Afterwards, the sections were incubated either with the anti-MAP2, anti-MBP, anti-CNPase, or anti-GFAP antibodies for 3 days at 4°C. Sections were then incubated with biotinylated secondary antibody (Vector Laboratories, Burlingame, CA), diluted 1:1000 for 1 hour at RT, followed by incubation with ABC solution (Vector) diluted 1:1000 for 1 hour at RT. The peroxidase labeling was visualized by incubation for 5 min with a mixture containing 0.02% 3,3'diaminobenzidine tetrahydrochloride (DAB), 0.0045% hydrogen peroxide and 0.6% nickel ammonium sulfate in 0.05 M Tris-HCl buffer, pH 7.6. Sections were mounted on glass slides, dehydrated and coverslipped with Permount. Sections stained for anti-MAP2, MBP, CNPase and GFAP were counterstained with cresyl violet.

Experimental Series I A series of experiments were undertaken to isolate, characterize, and selectively differentiate in-vitro human neural crest stem cells and their descendent multipotent progeny cells. The isolated individual clones of neural crest stem cells are stable cell lines; maintained in culture as immortalized cell lines; and are suitable for subsequent in-vivo use in a variety of different medical/clinical applications.

Experiment 1 : Isolation and expansion of human immortalized neural crest stem cells (NCSCs) Dissociated cell cultures were established from human embryonic dorsal root ganglia (DRG) of 15 week gestation by collagenase treatment as described previously by Kim et al. [ Neurosci. Res. 22: 50-59 (1989)]. DRG cultures were grown for 1-3 weeks and consisted of small (10 μm in diameter) or larger (15 μm in diameter) nerve cells in singles or clusters, more numerous spindle shaped (15-20 μm in length) Schwann cells, flat polygonal fibroblasts and a small number of NCSCs. The result is shown by Fig. 1 A.

As seen in Fig. 1A, a dissociated cell culture of human dorsal root ganglia isolated from 15 week gestation embryo was grown in-vitro for 7 days. The culture contains a large number of nerve cells, Schwann cells and a small number of neural crest stem cells; and represents a starting material for the generation of human neural crest stem cell (NCSC) line. The bar indicates 20 μm.

Subsequently, DRG cultures with 80% confluency were subjected to retrovirally mediated transduction of vmyc oncogene and subsequent cloning. After 7-14 days of G418 selection, ten G418-resistant clones were isolated and expanded. One of these clones, HNC10/C2, was utilized for further study. The cloned NCSC was tripolar or multipolar in morphology with 10-15 μm in size. This is shown by Fig. IB. Fig. IB shows that a human neural crest stem cell line, HNCIO, can be grown in UBC1 serum-free medium which contains human insulin, human transferrin, sodium selenite and other nutrients and FGF-2. HNCIO cells can also be grown in serum-containing medium (10% fetal bovine serum) and differentiate into nerve cells, Schwann cells, adrenal chromaffin cells and skeletal muscle cells. The bar indicates 20 μm.

Afterwards, HNC10/C2, one of these clones, was subjected to further study. In the proliferative growth condition, the clone HNC10/C2 exhibited a doubling time of 24 hr as shown by Fig. lC. When calculated by a population analysis the doubling time for HNC10 cells was determined to be 23.67 hr.

Experiment 2: Characterization of HNC10 cells HNC10/C2 cell line grew in culture as single cells or clusters that could be subcultured and passaged weekly for 6 months. The cells grown on coverslips were processed for immunocytochemical staining of cell type specific markers.

As shown in Figs. 2 A and 2B, HNC10 cells carry two cell-type-specific markers for neural crest stem cells: nestin, a class of intermediate cytoskeletal protein only found in neural crest stem cells; and p75 neurotrophin receptor protein (p75NGFR), another unique cell type-specific marker for neural crest stem cells. HNC10 neural crest stem cells are shown to express a very strong nestin and p75NGFR immunoreactivity. The bar indicates 20 μm.

HNC10 cells were immunoreaction-positive with an antibody against human mitochondria and also for pan-myc oncoprotein, as shown by Figs. 2C and 2D, indicating that HNC10 cells are uniquely human origin and contain a copy or copies of ymyc-encoding retrovirus, as determined relative to positive and negative controls run in parallel. Fig. 2C demonstrates that HNC10 cells were of human donor origin in which HNC10 cells were reacted with a monoclonal antibody specific for human mitochondria antigen. Note that cytoplasmic mitochondria are intensely positive for the reaction. The bar indicates 20 μm. Also, since human dorsal root ganglia cells were transfected by the vmyc oncogene and transformed to become HNCIO cell lines, presence of myc antigen was determined. This is shown by Fig. 2D. Immunoreactivity was found in nuclei of HNCIO cells, indicating that these cells are indeed transformed by myc oncogene. The bar indicates 20 μm. Another cell-type marker for neural crest stem cells, vimentin, was also demonstrated in HNCIO cells (results not shown). In addition, HNCIO cells were also immunoreaction-positive for Musashi, a transcription factor uniquely detectable in neural progenitor cells [see for example, Sakakibara e l.. Neurosci. 17: 8300- 8312 (1997)]; FORSE-1, a surface antigen marker for neural progenitor cells [see for example, Tole S. and P.H. Patterson, Neurosci. 15: 970-980 (1995)]; PSA- NCAM, a surface antigen marker for oligodendrocyte-type 2 astrocyte (O-2A) pre- progenitor cells and neural progenitor cells [see for example, Grinspan, J.B. and B. Franceschini, Neurosci. Res. 41: 540-551 (1995)]; and NG2 proteoglycan, a surface antigen marker for glial progenitor cells and neural progenitor cells [see for example, Stallcup, W.B. and L. Beasley, Neurosci. 7: 2737-2744 (1987)].

Experiment 3: Neuronal and glial in-vitro differentiation of HNC10 cells To determine if HNC10 cells could differentiate into neurons and glial cells in the presence of serum, cultures grown under serum-containing medium with 10% fetal bovine serum were examined. The results are shown by Figs. 3A-3F.

When HNC10 cells are grown in serum-containing medium, many of these cells differentiate into nerve cells as shown by Fig. 3A. Note the highly-branched, well-differentiated live neurons in such conditions and the long and stout dendritic processes of nerve cells after 2 weeks in-vitro. The bar indicates 20 μm. Neurons differentiated from HNC10 cells also expressed neurofilament proteins, NF-L (10.5 ± 3.5%), NF-M (13.2 ± 4.1 %) and NF-H (13.8 ± 3.6%) as shown by Fig. 3B; and the vast majority of neurofilament-positive cells co- expressed β tubulin isotype III, a cell type specific marker for neurons [see for example, Ferreira, A. and A. Caceres, J. Neurosci. Res. 32: 516-529 (1992)]; and peripherin, a marker of mature peripheral nervous system (PNS) neurons [see for example, Parysek, L.M. and R.D. Goldman. J. Neurosci. 8: 555-563 (1988)]. These phenotypes are specifically and exclusively expressed by mammalian nerve cells including human nerve cells - indicating that HNCIO cells can differentiate into nerve cells under the serum-containing culture condition.

Moreover, as shown by Fig. 3C, HNCIO cells also differentiated into glial cells when grown in serum-containing medium, and expressed S100, a cell type- specific marker for mature Schwann cells [see for example, Jessen et al. , Neuron 12: 509-527 (1994)]; but also immunoreaction positive for glial fibrillary acidic protein (GFAP), a marker for mature Schwann cells [see for example, Jessen et al.. Development 109: 91-103 (1990)]. HNC10 cells were, however, immunoreaction- negative for 01 and 04, both cell type specific markers for oligodendrocytes [see for example, Summer I, and M. Schachner, Dev. BioL 83: 311-327 (1981); and Schachner et aL, Dev, BioL 83: 328-338 (1981)].

In addition, as shown by Fig. 3D, HNCIO cells grown in serum-containing medium for 2 weeks were also immunoreaction-positive for chromogranin, a specific cell type marker for adrenal chromaffin cells [see for example, Wilson, B.S. and R.V. Lloyd, Am, L Pathol. 115: 458-468 (1984)]; and tyrosine hydroxylase, a marker for sympathetic neurons of the autonomic nervous system lineage (data not shown). These results indicate that HNCIO cells are capable of differentiation into not only PNS sensory neurons and Schwann cells but also to sympathetic neurons and to adrenal gland primordia where they differentiate into adrenal chromaffin cells [see for example, Anderson, D. J. , Curr. Opin. Neurobiol. 3: 8-13 (1993)].

Equally important, as shown by Figs. 3E and 3F respectively, after 1-2 weeks in serum-containing medium, HNCIO cells gave rise to large flat, elongated, multinucleated cells, and many of these multinucleated cells expressed desmin- and myosin-immunoreaction [see for example, Naumann, K. and D. Pette,

Differentiation 55: 203-211 (1994)]. Note that Fig. 3E reveals that when HNCIO human neural crest stem cells were grown in culture in serum containing medium, the cells were seen to differentiate into myoblasts and myotubes indicating HNCIO cells' ability to develop into skeletal muscle cells (living/phase contrast microscopy). The bar indicates 20 μm. Fig. 3F supplements Fig. 3E and shows that HNCIO cells differentiate into myoblasts/myotubes and express irnmunoreactivity of human myosin. The bar indicates 20 μm. These results indicate that HNCIO cells, human NCSCs, can and do differentiate into skeletal muscle as reported earlier [see for example, N.M. Le Douarin, The Neural Crest, Cambridge University Press, Cambridge, U.K., 1982; Le Douarin et al., Dev. Biol. 159: 24-49 (1993)1.

Experiment 4: Reverse transcriptase-polymerase chain reaction (RT-PCR) analysis Reverse transcriptase-polymerase chain reaction (RT-PCR) analysis of genes expressed by the HNCIO cell line were undertaken. Fig. 4 shows RNA transcripts for nestin and p75NGFR (both cell type markers for neural crest stem cells) were demonstrated in HNCIO cell line growth in serum-free medium. Cell type markers for neurons (neurofilament/NF-L, NF-M and NF-H), astrocytes/ Schwann cells (GFAP), oligodendrocytes (MBP), Schwann cells (PO) or microglia (B7-2) were not detected in HNCIO human neural crest stem cells grown in serum-free medium. Fig. 4 reveals: Lane M: Molecular standards/bp; lane 1: nestin; lane 2: p75NGFR; lane 3: NF-L; lane 4: NF-M; lane 5: NF-H; lane 6: GFAP; lane 7: MBP; lane 8: PO (P-zero); lane 9: B7-2.

Results of RT-PCR analysis of cell type-specific markers for the HNCIO cell line grown in serum-free medium show that transcripts for nestin and p75NGFT, cell type-specific markers for NCSCs were clearly demonstrated. In addition transcripts for trkA, a subtype of neurotrophin receptors, were expressed by HNCIO cells (data not shown); but trkB and trkC, other neurotrophin receptors, were not detected. Transcripts for NF-L, NF-M and NF-H, cell type-specific markers for early and differentiated neurons, GFAP (a marker for astrocytes), MBP (a marker for oligodendrocytes), PO (a marker for Schwann cells) or B7-2 (a marker for microglia) were not demonstrated by the RT-PCR analysis.

This RT-PCR study of HNCIO cells indicates that the cells express transcripts for specific antigen markers of neural crest stem cells. It is evident from the RT-PCR results that HNCIO, human NCSCs, grown in serum-free medium, an environment in which NCSCs appear uncommitted as stem cells, expressed cell type-specific phenotypes of NCSCs, nestin and p75NGFR. Experiment 5: Effects of different growth factors in-vitro upon HNCIO cells HNCIO cells were grown in serum-free medium (DMEM containing UBC1 supplements) for 4 days and then supplemented with one of the following growth factors (differentiation medium): FGF-2, NGF, BDNF, NT-3, PDGF-BB, and neuregulin alpha and beta.

Differentiation of HNCIO cells, human neural crest stem cells, into neuronal and glial cells was determined by rmmunocytochemistry following stimulation by various neurotropic factors including FGF-2 (10 ng/ml), NGF (50 ng/ml), BDNF (50 ng/ml), NT-3 (50 ng/ml), PDGF-BB (20 ng/ml), neuregulin αl (20 ng/ml), neuregulin β2 (20 ng/ml) and TGFβl (20 ng/ml). HNC10 cells were grown in serum-free medium supplemented with one of the growth factors described above for 5-7 days and then processed for immunostaining. The results are shown by Figs. 5 A, 5B and 5C respectively.

Fig. 5A shows the differentiation of HNC10 cells into neurons as identified by immunostaining with neurofilament protein 150 (NF-M). Neuregulins, TGFβl and NT-3 are inhibitory to neuronogenesis, while others did not affect the neuronal induction of the HNC10 cells. Fig. 5B shows glial cells/Schwann cells which were identified by GFAP immunoreaction. All the growth factors except neuregulin β2 were effective in glial cell induction. It is noted that TGFβl is most effective as a glial cell inducer. Finally, Fig. 5C demonstrates the differentiation of HNC10 cells into myoblasts/myotubes by immunostaining with skeletal muscle myosin. All the growth factors except PDGF-BB are highly effective in myoblast/myotube induction of neural crest stem cells.

Overall therefore, there was a considerable degree of inhibition in neuronogenesis of HNC10 cells grown in neuregulins, TGFβ 1 , or NT-3 , while NGF, BDNF or PDGF showed much less inhibitory effect on neuronal differentiation (Fig. 5A). FGF-2 (bFGF) showed a small degree of neuron- inductive effect, although it was not statistically significant. TGFβl induced a greatly enhanced induction of Schwann cells (GFAP, 45.7 ± 5.8%; S100, 59.1 ± 9.2%). Similarly in HNC10 cells treated with FGF-2, NGF, BDNF, NT-3, PDGF or neuregulin for 5-7 days, there were large numbers of GFAP- or S100- immunoreactive cells indicating these growth factors favor differentiation of NCSCs 49

X-gal reaction positive blue cells migrated from the lumen of the ventricule into the neighboring paraenchymal region of the brain (the bar indicates 0.3 mm). Fig. 6C presents a higher magnification of Fig. 6B (the bar indicates 50 μm). Fig. 6D provides a schematic map of mouse brain as shown in Figs. 6E and 6F. Then, Fig. 6E shows a large number of X-gal-positive blue cells which are seen to migrate from the ventricular lumen into hippocampal formation (the bar indicates 0.2 μm). Fig. 6F provides a higher magnification of Fig. 6E (the bar indicates 50 μm).

The results evidenced by Figs. 6A-6F respectively show that LacZ- expressing HNC10 cells migrated throughout the mouse brain; and the pattern of spread indicates that they migrated a considerable distance into the brain parenchyma along the ventricular walls.

In several cases, also, a widespread presence of X-gal-positive cells along the neuronal layers of hippocampal fomtion was noted. This is shown by Figs. 6G and 6H. Fig. 6G shows the result two weeks post-operation. Beta-gal-positive HNC10 cells were found in brain parenchyma are indicated. The bar indicates

20 μm. Fig. 6H presents the same field as Fig. 6C. Beta-gal-positive HNC10 cells are doubly immunoreactive to galactocerebroside antibody, indicating that these cells differentiated into myelin-forming cells of Schwann cell lineage in response to a signal generated in the local environment (the bar indicates 20 μm). The empirical evidence and data thus show that these X-gal/ β-gal positive cells expressed galactocerebroside (Ol antibody) immunoreaction, a specific cell type marker for myelination, 2 weeks post-transplantation in the brain parenchyma. This indicates that HNC10 cells can and had differentiated into myelin forming glial cells, probably Schwann cells (Figs. 6G and 6H). These results also indicate that when HNC10 cells are implanted into the brains of the developing nervous systems, they generate progeny cells which become myelin-forming glial cells by responding to developmental signals originated in that area at the time the cells are grafted.

Experiment 8: Evaluation of the myelinating capacity of HNCIO cells in-vivo The next objective was to evaluate the myelinating capacity of HNC10 cells and to determine whether transplanted HNC10 cells could persist in the myelin- deficient mutant environment. HNC10 cells were therefore grafted into the 50 unilateral neostriatum of 4 week-old myelin mutant shaker rats. After 1-4 weeks post-transplantation, these shaker rats were sacrificed and processed for immunohistochemical investigation of CNPase (oligodendrocytes/Schwann cells/CNS and PNS myelin), GFAP (astrocytes/Schwann cells) and MAP2 (neurons). The results are illustrated by Figs. 7A-7D respectively.

Specifically, Fig. 7 provides a schematic map of myelin mutant shaker rat brain was made where HNCIO human neural crest stem cells were grafted earlier. Frontal sections through neostriatal region of shaker rat brain grafted with HNCIO cells. Four weeks post-implantation, CNPase (a specific marker for myelin) immunostaining was performed. Fig. 7 A shows the control side of the brain where the presence of CNPase-positive myelin is minimal (the bar indicates 0.3 mm). Fig. 7B shows a large number of CNP-immunoreaction positive patches representing myelinated axons are visible throughout the field. A brown colored strip indicates the track of HNC10 cell implantation (the bar indicates 0.3 mm). Fig. 7C provides a higher magnification of Fig. 7 A (the bar indicates 0.1 mm). Fig. 7D presents a higher magnification of Fig. 7B. Along the path of injection track, there are good numbers of myelin patches which represent remyelinated axons produced by Schwann cells differentiated from HNC10 human neural crest stem cells (the bar indicates 0.1 mm). Thus, as shown by Figs. 7A-7D, a widespread remyelination of previously unmyelinated axons is demonstrated as shown by an intense staining of myelinated axons by CNPase antibody, while CNPase staining was minimal in the control areas. Astrogliosis was not induced in the implanted sites as no GFAP-positive focus was found.

Conclusions Drawn From Experiments:

1. For the first time, immortalized cell lines of human NCSCs were generated. These were created via retrovirus-mediated vmyc transfer into cells initially obtained from human embryonic DRG. HNC10 cells, immortalized NCSCs, are self-renewing and multipotent in-vitro; and, in the proper setting, rise to four different lineages - i.e. , neurons, Schwann cells, adrenal chromaffin cells and 51 skeletal muscle cells. HNCIO cells express phenotypes characteristic for neural crest stem cells, i.e., nestin and p75NGFR, thereby indicating that HNCIO cells without any reservation are of neural crest stem cell origin.

2. Isolated as stable cell lines, human NCSCs can be distinguished from CNS stem cells by their morphology; by expression of low-affmity NGF receptor; by the multipotent progeny that they generate; and by their inability to generate CNS derivatives. NCSCs can differentiate into sensory, sympathetic, and enteric neurons; or Schwann cells; or adrenal chromaffin cells; as well as into non-neural derivatives such as melanocytes, cartilage, bone, and smooth/skeletal muscle.

HNCIO cells grown in serum-free medium expressed immunoreaction-positive for antibodies specific for A2B5, PSA-NCAM, NG2, Musashi and FORSE-1, cell type- specific markers known for neuronal and glial progenitors. When HNCIO cells were grown in serum-containing medium, a large number of these cells expressed immunoreaction positive for neurofilaments NF-L, NF-M, NF-H and MAP-2, tubulin βlll and neuron specific enolase, cell type-specific markers for neurons; GFAP and S-100, markers for Schwann cells; chromagranin and PNMT, markers for adrenal chromaffin cells; and tyrosine hydrolase (TH), a marker for synthetic neurons. In addition, HNCIO cells could differentiate into skeletal muscle cells as shown by myosin staining.

3. Since the myelin basic protein (MBP) gene defect in mutant shaker rats results in a severe dysmyelinating in white matter, shaker rats serve as an excellent animal model for human demylinating diseases (such as multiple sclerosis). HNCIO cells were implanted into neostriatal region of 4 week old shaker rats, and the brains were examined 1-4 weeks post-operation by irnmunocytochemical staining for MBP and CNPase, specific markers for myelin-forming glial cells and myelin. Implanted HNCIO cells survived well and differentiated into myelinating glial cells (Schwann cells) to remyelinate previously unmyelinated axons; and also migrated to other anatomical sites remote from the original grafted site. These results indicate that HNC10 cells, human immortalized NCSCs, are capable of survival, migration and remyelination of host axons in myelin deficient mutant brain tissue; and are the cells 52 of choice for cell replacement and repair of brain with demyelination/dysmyelination pathological lesions.

4. Mature NCSC-derived cells express phenotypic markers that can be used to distinguish them from related cells in the CNS. For example, Schwann cells can be distinguished from astrocytes and oligodendrocytes by the co-expression of GFAP and myelination antigens such as galactocerebroside and 04. Peripherin expression is characteristic of peripheral neurons such as DRG neurons, derivatives of NCSCs, but is generally seen in only a limited population of CNS derivatives. Although NCSCs are multipotent stem cells of the PNS, they seem incapable of generating CNS derivatives and transplantation of neural crest stem cells into the CNS results in primarily Schwann cell differentiation. Thus NSCs of the CNS and NCSCs represent two quite independent stem cells of two very different systems.

5. The choice of each of several alternative fates available to NCSCs can be instructively promoted by different environmental signals including those growth factors previously reported in the scientific literature. In the present study, it was found that FGF-2 promotes differentiation of HNCIO cells into three cell types, i.e., neurons, Schwann cells and skeletal muscle cells. Neuregulin and TGFβ-1 induced HNCIO cells to differentiate into Schwann cells and skeletal muscle cells and these growth factors blocked neuronal differentiation of HNCIO cells. Members of neurotrophin family, NGF, BDNF and NT3 also favored differentiation of HNCIO cells into glial and skeletal muscle cells. These findings showed that trophic effect was not specific for a particular neurotrophin; and that the differentiation mechanism involves activation of the non-selective neurotrophin receptor, p75NGFR, by one of these neurotrophins. These results indicate that neurotrophins and other pertinent growth factors affect differentiation of HNCIO cells at multiple levels for each and different cell lineage. Moreover, these results suggest that concerted action of combinations of growth factors is at least as important in determining the fate of NCSCs as the action of a single growth factor alone. 53

6. HNCIO cells possess a capability for myelin formation. As the empirical results show, when HNCIO cells were implanted into the neostriatum of 4-week old myelin mutant shaker rat brains, an extensive migration of HNCIO cells from the site of implantation to neighboring anatomical sites was found and a widespread remyelination of previously unmyelinated axons was demonstrated at 1-5 weeks post-implantation. These results clearly demonstrate that when HNCIO cells are implanted into pathological myelin-deficient areas of the CNS, their differential potential is strongly influenced by the environment signals at the site of implantation as they form myelin.

7. The present study shows that HNCIO cells, NCSCs, are useful for a variety of diseases characterized by profuse neural degeneration and cell death that might benefit from the replacement of lost neurons. Accordingly, when neural crest stem cells are implanted into different areas of the developing nervous system, they will provide progeny that would normally be generated in that area at the time the cells are grafted. Thus the same precursor cells can generate Purkinje cells when implanted into the cerebellum and hippocampal neurons when introduced into the hippocampus at the time that these cells are being generated from endogenous precursors. The present studies show that LacZ expressing HNCIO cells which were grafted into the cerebral ventricle of newborn mice migrated a great distance and became a part of hippocampal pyramidal cell neurons 2-3 weeks post- operationally. Thus, when multipotent neural crest stem cells are implanted into pathological areas of the CNS, their differentiation potential is influenced by the environment into which they are placed; and these become CNS neurons which are integrated into the cytoarchitecture of the host CNS.

8. The NCSC and their multipotent progeny are suitable for cell therapy purposes. Cell therapy has become a most promising strategy for the treatment of many human diseases including neurological disorders; and the objective of cell therapy is to replace lost cells and restore the function of damaged cells and tissues.

Transplantation of renewable, homogenous and well-characterized cells into the damaged target tissue or organ should replace lost cells and restore damaged 54 function. HNCIO cells, a human NCSC, can fulfill these criteria and serve as an ideal cell type as donor cells for cell therapy in various neurological diseases including motor and sensory neuropathy, multiple sclerosis, Parkinson disease, Huntington disease, spinal cord injury, stroke, Duchenne muscular dystrophy and pain control. In addition, human NCSCs can serve excellently as a vehicle to carry human genes by which to cure genetically incurred neurological diseases in an ever- expanding field of gene therapy. The stable immortalized human NCSC line described and characterized here can be expanded readily to provide a renewable and homogeneous population of glial, neuronal and muscle cells; and they will be most valuable for future research studies of fundamental questions in developmental neurobiology, cell and gene therapies, as well as the research and development of new drugs.

The present invention is not restricted in form nor limited in scope, except by the claims appended hereto.

Claims

55 What we claim is:

1. A primordial human neural crest stem cell suitable for on-demand implantation in-vivo into a living host subject, said primordial human neural crest stem cell comprising: a pluripotent and self-renewing neural crest stem cell of human origin which (i) carries native human genomic DNA which has not been genetically modified by human intervention means;

2. The living progeny of a primordial human neural crest stem cell suitable for on-demand implantation in-vivo into a living host subject, said living progeny comprising: multipotent descendent cells of human neural crest stem cell origin which (i) carries native human genomic DNA which has not been has not been genetically modified by human intervention means;

(iii) is implantable in-vivo as an uncommitted cell; (iv) optionally migrates in-vivo after implantation from the implantation site to another anatomic site for in-vivo integration within the living host subject; 56

(v) integrates in-situ after implantation into the body of the living host subject at a local anatomic site; and

3. A genetically modified human neural crest stem cell maintained as a stable cell line in-vitro and suitable for on-demand implantation in-vivo into a living host subject, said human neural crest stem cell comprising: a primordial neural crest stem cell of human origin which (i) remains uncommitted and undifferentiated while passaged in- vitro using as a mitotic, self-renewing cell line;

(ii) is implantable in-vivo as an uncommitted cell; (iii) optionally migrates in-vivo after implantation from the implantation site to other anatomic sites for integration within the body of the living host subject;

4. The genetically modified human neural crest stem cell as recited in claim 3 wherein said viral vector is an amphotrophic retroviral viral vector.

5. The genetically modified human neural crest stem cell as recited in claim 3 wherein said viral vector is an exogenous vmyc DNA sequence.

6. The genetically modified human neural crest stem cell as recited in claim 3 wherein said differentiated cell is a neuron. 57

7. The genetically modified human neural crest stem cell as recited in claim 3 wherein said differentiated cell is a Schwann cell.

8. The genetically modified human neural crest stem cell as recited in claim 3 wherein said differentiated cell is a skeletal muscle cell.

9. The genetically modified human neural crest stem cell as recited in claim 3 wherein said differentiated cell is an adrenal chromaffin cell.

10. The living progeny of a genetically modified human neural crest stem cell maintained in-vitro as a stable cell line and suitable for on-demand implantation in- vivo into a living host subject, said living progeny comprising: multipotent descendent cells of human neural crest stem cell origin which (i) remain undifferentiated while maintained in-vitro as mitotic cells;

(ii) are implantable in-vivo at a chosen implantation site as undifferentiated cells;

(iii) optionally migrate in-vivo after implantation from the implantation site to another anatomic site for integration within the body of the living host subject;

(iv) integrate in-situ after implantation into the body of the living host subject at a local anatomic site; and

(v) differentiate in-situ after integration into a recognized type of differentiated cell of neural crest origin; and human genomic DNA genetically modified to include a viral vector carrying at least one DNA segment comprised of an exogenous gene coding for a specific protein.

11. The living progeny of a genetically modified human neural crest stem cell maintained in-vitro as recited in claim 10 wherein said viral vector is an amphotrophic retroviral viral vector. 58

12. The living progeny of a genetically modified human neural crest stem cell as recited in claim 10 wherein said viral vector carries an exogenous vmyc DNA sequence.

13. The living progeny of a genetically modified human neural crest stem cell as recited in claim 10 wherein said differentiated cell is a neuron.

14. The living progeny of a genetically modified human neural crest stem cell as recited in claim 10 wherein said differentiated cell is a Schwann cell.

15. The living progeny of a genetically modified human neural crest stem cell as recited in claim 10 wherein said differentiated cell is a skeletal muscle cell.

16. The living progeny of a genetically modified human neural crest stem cell as recited in claim 10 wherein said differentiated cell is an adrenal chromaffin cell.

17. A clone of genetically modified human neural crest stem cells stably maintained in-vitro as a cell line and suitable for on-demand implantation in-vivo into a living host subject, said clone comprising: multiple primordial human neural crest stem cells which

(i) originate from and are the descendent stem cells of a single, genetically modified, human neural crest stem cell;

(ii) remain uncommitted and undifferentiated while passaged in- vitro as a mitotic, self-renewing cell line; (iii) are implantable in-vivo as uncommitted cells;

(iv) optionally migrate in-vivo after implantation from the implantation site to other anatomic sites for integration within the body of the living host subject;

(v) integrate in-situ after implantation into the body of the living host subject at a local anatomic site;

(vi) differentiate in-situ after integration into at least one recognized type of differentiated cell of neural crest origin; and 59

(vii) have the same genetically modified human genomic DNA in common, said DNA having been genetically modified to include a viral vector carrying at least one DNA segment comprised of an exogenous gene coding for a specific protein.

18. A clone of genetically modified human neural crest stem cell progeny stably maintained in-vitro as a cell line suitable for on-demand implantation in-vivo into a living host subject, said clone comprising: a plurality of multipotent living progeny cells which (i) originate from and are the descendent progeny of a single, genetically modified, human neural crest stem cell ancestor;

(ii) remain undifferentiated while passaged in-vitro as mitotic cells;

(iii) are implantable in-vivo as an undifferentiated cells; (iv) optionally migrate in-vivo after implantation from the implantation site to other anatomic sites for integration within the body of the living host subject;

(v) integrate in-situ after implantation into the body of the living host subject at a local anatomic site; (vi) differentiate in-situ after integration into at least one recognized type of differentiated cell of neural crest origin; and

(vii) have the same genetically modified human genomic DNA in common, said human genomic DNA being genetically modified to include a viral vector carrying at least one DNA segment comprised of an exogenous gene coding for a specific protein.

19. The clone of cells as recited in claim 17 or 18 wherein said viral vector is an amphotropic retroviral viral vector.

20. The clone of cells as recited in claim 17 or 19 wherein said viral vector includes an exogenous vmyc DNA sequence.