CA2213780C - Pharmaceuticals containing multipotential precursor cells from tissues containing sensory receptors - Google Patents

Abstract

Current sources of neural stem and progenitor cells for neural transplantation are essentially inaccessible in living animals. This invention relates to neural precursor cells (stem cells, progenitor cells or a combination of both types of cells) isolated from the olfactory epithelium of mammals that can be passaged and expanded, and that will differentiate into cell types of the central nervous system (CNS), including astrocytes, oligodendrocytes, and tyrosine-hydroxylase-positive neurons. These precursor cells provide an accessible source for autologous transplantation in CNS, PNS, spinal cord and other damaged tissues.

Description

Title: Pharmaceuticals Containing Multipotential Precursor Cells From Tissues Containing Sensory Receptors Cross Reference to Related Applications This application claims the benefit of U.S. Provisional Application No. 60/024,590, filed August 26, 1996, and U.S. Provisional Application No. 60/024,456 filed August 27, 1996 which are incorporated by reference herein in their entirety.
- Field of the I~vention The present invention relates to multipotential precursor cells isolated from peripheral tissues cont~inin~ sensory receptors such as the olfactory epithelium and tongue. The invention 10 also relates to cells di~clcllliated from the precursor cells. The invention includes ph~rm~ce~ltical compositions co..'~ ing precursor cells. The invention also includes cells differentiated from precursor cells and uses for those cells.
R~rl~ound of the Ir~vention There are a number of diseases of the central nervous system ("CNS") which have a 15 dev~t~ting effect on patients. These ~ e~es are incurable and debilitating. They include Alzheimer's ~ e7 Huntington's ~i~e~ce, Parkinson's disease and Multiple Sclerosis, to name a few.
By way of example, Parkinson's disease is a progressive degenerative disorder ofunknown cause. In healthy brain tissue, dopaminergic neurons extend from the substantia nigra 20 of the brain into the striatum. Parkinson's disease occurs when these dopaminergic neurons die.
There are a number of methods to treat Parkinson's disease.
One method is to treat hllm~n~ having parkinsonism with L-DOPA. Another method is to transplant cells into the substantia nigra or striaturn. Transplanted cells replace endogenous cells that are lost as a consequence of damage. Transplanted cells may also be used as vectors for the 25 ex~lt;s~ion of thelalJeulic molecules. Another method is to implant fetal brain grafts co~ g dopaminergic neurons. This method is ~x~;lhllental (Widner et al., 1993; C~ h~n et al., 1992).
An animal model of Parkinson's disease is an MPTP-treated non-human prim~te. The animal models have been transplanted with dopamine-rich embryonic neurons with some success (Dunnett et al., 1991). (MPTP is a selective dopaminergic toxicant that produces parkinsonian symptoms in humans and in primates after a one-hit lesion to the neurons in the substantia nigra (Langston et al., 1983; Burns et al., 1983)).
Investigators studying other neurode~en~.dli~/e ~ e~ec, such as Alzheimer's disease and Huntington's ~ e~ee7 are exploring the possible usefulness of fetal-tissue implants in the - treatment of these ~ e~ces~
Current approaches to transplantation suffer from a number of serious limitations. First, many investigators are l~ti~ inp non-neural cells such as fibroblasts or transformed cell lines for transplantation. Second, the safety of transplantation of immortalized cell sources into the human brain is a concern. These cells may become unregulated and develop into tumors. Third, transplants of dopaminergic neuron fetal tissue to Parkinson's disease patients have a number of difficulties:
~ the fate of implanted dopalllinergic neurons in patients with Parkinson's disease is ullc~ in - whatever caused the loss of endogenous dop~ e gic neurons may also eventually injure the implanted ones, ~ in many cases, implants provide only telllpoldly relief as the ~ylll~lollls associated with the disease often return after a number of years, ~ the patient may reject foreign fetal tissue, ~ there are adverse reactions associated with immuno~ul l,lession (immuno~ul~le3sion is needed to try to help the patient accept the foreign fetal tissue, even though the brain is, to some degree, immunologically privileged), a sufficient number of cells in the fetal tissue being implanted are unable to survive during and after implantation, ~ the implants may not be regulated by the host brain, ~ other ~ice~es or disorders may be ll~ e~1 to the patient via the implant,

- 2 -~ the cost and effort associated with implanting fetal tissue may not be justified ~y the results, and ~ there are objections to the ethics associated with implanting fetal tissue.
Many of these problems are encountered with transplants used to treat other S neurodegenerative ~ e~es, disorders or abnormal physical states.
In some tissues, stem cells and progenitor cells are proposed as a source for alternative treatments of disease or injury to tissues. The proposed treatments involve transplants of healthy tissue or endogenous stimulation of stem cells or progenitor cells to produce healthy tissue.
Stem cells are undirr~.c~ ,Li~ted cells that exist in many tissues of embryos and adult 10 m~mm~l~. In embryos, blastocyst stem cells are the source of cells which differentiate to form the specialised tissues and organs of the developing fetus. In adults, specialised stem cells in individual tissues are the source of new cells which replace cells lost through cell death due to natural attrition, disease or injury. No stem cell is common to all tissues in adults. Rather, the term "stem cell" in adults describes different groups of cells in dirr~lcnt tissues and organs with 15 common ~h~r~cteristics.
Stem cells are capable of producing either new stem cells or cells called progenitor cells.
A progenitor cell differentiates to produce the mature specialized cells of m~mm~ n organs. In contrast, stem cells never tt?rmin~lly differentiate (i.e. they never dirr~lcnllate into specialized tissue cells). Progenitor cells and stem cells are referred to collectively as "l,rccul~or cells".
20 This term is often used when it is unclear whether a researcher is dealing with stem cells or progenitor cells or a combination of both cells.
Progenitor cells may differentiate in a manner which is unipotential or multipotential. A
unipotential progenitor cell is one which can form only one particular type of cell when it is t~rmin~lly dirrele~Li~tefl A multipotential progenitor cell has the potential to differentiate to 25 form more than one type of tissue cell. Which type of cell it llltim~tely becomes depends on conditions in the local enviro~ cnt such as the presence or absence of particular peptide growth factors, cell-cell co~ ication, amino acids and steroids. For example, it has been ~i~lr~ d that the hematopoietic stem cells of the bone marrow produce all of the mature Iymphocytes and erythrocytes present in fetuses and adult m~mm~l~ There are several well-studied progenitor cells produced by these stem cells, including three unipotential and one multipotential tissue cell.
The multipotential progenitor cell may divide to form one of several types of differenti~te~l cells S depending on circl.m~t~nces such as which hormones or factors act upon it and cell-cell contact.
Weiss et al, 1996, summarises the five defining characteristics of stem cells as the ability - to:
~ Proliferate: Stem cells are capable of dividing to produce tl~llghter cells.
~ Exhibit self-m~intPn~nce or renewal over the lifetime of the org~ni.~m Stem cells are capable of reproducing by dividing symrnetrically or asymmetrically to produce new stem cells. Symmetric division occurs where one stem cell divides into two f~ ghttqr stem cells. Asymrnetric division occurs where one stem cell forms one new stem cell and one progenitor cell. Symrnetric division is a source of renewal of stem cells. This permits stem cells to m~int~in a con~i~tent level of stem cells in an embryo or adult m~ nm~l ~ Gel~lale large nurnber of progeny: Stem cells may produce a large number of progeny through the transient arnplification of a population of progenitor cells.
~ Retain their mIlltilinP~ge potential over time: Stem cells are the ultimate source of di~ te~l tissue cells, so they retain their ability to produce multiple types ofprogenitor cells, which will in turn develop into speci~li7e~1 tissue cells.
~el,el~le new cells in response to injury or disease: T}~is is ec~.onti~l in tissues which have a high turnover rate or which are more likely to be subject to injury or tli~e~ce, such as the epitheliurn or blood cells.
Thus, the key fe~ s of stem cells are that they are multipotential cells which are 25 capable of long-terrn self-renewal over the lifetime of a m~mm~l There has been much effort to isolate stem cells and dçtçrmine which peptide growth factors, hormones and other metabolites influence stem cell renewal and production of progenitor cells~ which conditions control and influence the differentiation of progenitor cells into specialized tissue cells, and which conditions cause a multipotential progenitor cell to develop 5 into a particular type of cell.
Stem cells or progenitor cells may be used as substrates for producing healthy tissue where a ~lice~ce7 disorder or abnormal physical state has destroyed or damaged normal tissue.
For example, stem cells and progenitor cells may be used as a target for in vivo stim~ tion with growth factors or they may be used as a source of cells for transplantation. The stem cells or 10 progenitor cells may be transplanted or they may be in~luced to produce healthy differenti~ted cells for transplant.
In several tissues, stem cells have been isolated and characterised in an aKempt to develop new therapies to repair or replace damaged tissues. For example, neural stem cells have been isolated from the m~nnm~ n brain (Reynolds and Weiss, Science 255:107 (1992)) and 15 these cells were shown to be multipotential and able to dirr~,.cllliate into neurons, astrocytes and oligodendrocytes. WO 93/01275, WO 94/16718, WO 94/10292 and WO 94/09119 describe uses for these cells.
WO 95/13364 reports the delivery of growth factors to the ventricles ofthe CNS in order to stimulate neural stem cells to proliferate and produce neural progenitor cells which will 20 develop into neurons, oligodendrocytes or astrocytes. This procedure has many complications which must be addressed before it may be used clinically. Dirr~.ç..ti~ting the target neural stem cells or neural progenitor cells into a desired type of tissue which is functional is one complication. Another complication is choosing a growth factor which does not cause side effects in other areas of the brain.
These publications are limited to isolating or using adult stem cells from the brain (in particular, the tissue around the brain ventricles, the ventricle ependyma, which is the remnant of the embryonic brain germinal zone). Although these publications suggest that progenitor cells may be isolated from the adult peripheral nervous system ("PNS"), the publications define the PNS as the system which originates from the neural crest. There is no reported isolation of a stem cell from the PNS which does not originate from the neural crest.
There are no clinical treatments involving transplants of neural stem cells or neural progenitor cells isolated from the brain nor are there clinical treatments using differentiated cells 5 produced from the neural stem cells or neural progenitor stem cells isolated from the brain.
There are also no clinical tre~tm~ntc to endogenously stim~ te the neural stem cells or neural progenitor cells of the brain in vivo to produce differenti~te~l cells. Even if there were clinical procedures to transplant fetal neural stem cells or neural progenitor cells from the brain, or to transplant cells di~lellliated from these stem cells or progenitor cells (e.g. dopaminergic 10 neurons into Parkinson's disease patients), this would not overcome the many problems of transplants from one human to another. As mentioned above, the only current, accessible human source for these neural stem cells and neural progenitor cells is aborted human fetuses, raising serious ethical concerns. Heterologous transplants are also very risky and complicated because of problems with graft rejection, immlm~u~ e3sion~ and the potential for donor grafts 15 transferring diseases or disorders to a recipient. Fnc~rs~ tion of cells in microspheres has the potential to decrease the likelihood of graft rejection, but this effect is lost if the hlle~lily of the microsphere is disrupted. There is a clear need for safer tissue grafts which can be transplanted to a recipient without being rejected.
The safest type of tissue graft would be one that comes from self (an autologous tissue 20 source). Autologous tissue sources are widely used in procedures such as bone transplants and skin transplants because a source of healthy tissue is readily ~ccessible for transplant to a damaged tissue site. In brain tli~e~es, such as Parkinson's ~ e~e7 healthy dop~min~.gic neu.onal brain tissue may exist at other sites in the brain but alle~ ts to tr~n~pl~nt these neurons would harm the site where the healthy neurons originate. Neural stem cells or neural pre~;ul~r 25 cells that can be di~~ te~l into dop~minergic neurons may be available at the damaged site or at other sites from which they may be transplanted, but the CNS, particularly the brain, is physically difficult to access. It would be h,l~ ical or impossible to access brain or other CNS
tissue for biopsy and then again for transplant in patients with we~k~necl health. It would be very useful if there were accessible stem cells or progenitor cells that could be dirre~ te~l into CNS
cell types, such as dopaminergic neurons, tO provide a source of cells for autologous transplants.
It would be useful if neural stem cells or progenitor cells could be identified and isolated outside the CNS and outside the PNS which ori~in~t~s from the neural crest. Medical treatm~nte 5 could then be developed using those neural stem cells, neural progenitor cells or cells dirre,~"~i~te~ from those cells. It is clear that despite the work that has been done to attempt to treat neurodegenerative diseases by tissue transplant, a need still exists for a ph~ eutical composition in which (1) the composition is accepted by the patient, thus avoiding the difficulties associated with immuno~upplcs~ion, (2) the composition is safe and effective, thus 10 justifying the cost and effort associated with treatmçnt (3) the composition provides long term relief of the symptoms associated with the ~Ti~e~ce7 (4) the composition is efficacious during and after transplantation and (5) there are no objections to the ethics of the composition9s use.
Thus, there is a clear need to develop neural stem cell cultures or neural progenitor cell cultures from ~cces~ihle tissues of the PNS which can act as a source of cells that are 15 transplantable to the CNS, PNS, spinal cord or other tissues in vivo in order to replace ~1~m~gecl tissue.
Summary of the Invention This invention relates to the isolation of "~ ul~or cells" (which may be neural stem cells or neural progenitor cells or a combination of both types of cells) from p~;lipl~ldl tissue with 20 sensory lcccl)tol~, specifically olfactory epithelium and tongue, of the PNS. The olfactory epithelium is part of the PNS, but does not originate from the neural crest. Rather, it is of placodal origin. Hence, peripheral sensory neurons of the olfactory epithelium are developmentally distinct from the neurons of the neural crest derived PNS. Olfactory lllCCUI~
cells have been isolated, ~let~rmin~cl to be multipotential and capable of gellcl~ g CNS cell 25 types. Thus, they are a useful source of tissue for autologous or heterologous transplant to the CNS, PNS, spinal cord and other damaged tissues.

The invention also includes isolated and purified precursor cells of a m~mm~l from peripheral tissue co~ g sensory receptors, wherein the precursor cells are selected from a group consisting of neural stem cells, neural progenitor cells and a combination of neural stems cells and neural progenitor cells. The cells can be isolated from tongue.
The inventors have isolated precursor cells from the olfactory epithelium of m~mm~l~
(juvenile and adult mice, adult rat and humans). The precursor cells of the olfactory epithelium possess the two key char~ctt?ri~ing features of stem cells: they are mutipotential and are self-renewing. They can be passaged and differenti~tçd into cell types of the CNS, including astrocytes, oligodendrocytes, and dopaminergic neurons. Precursor cells isolated from the olfactory epithelium of neonatal mice express the immlmf)logical marker of neural stem and progenitor cells, nestin. These cells are not restricted to ~ g an olfactory phenotype, but instead can di~~ iate into astrocytes, oligodendrocytes, and dopaminergic neurons. This shows that the olfactory epithelium is a useful source of dop~min~rgic neurons for homotypic grafts into Parkinson's Disease p~ti~nt~ The plC~ O[ cells of the olfactory epithelium may also be used for autologous or homologous transplants to treat other neurodegenerative (li~ e~, disorders or ~hn~rm~l physical states.
Precursor cells were also isolated from tongue and these may also be used for autologous or homologous transplants to treat nt;ulol~ a or neur~dege~,- aLi~e ~ ç~es, disorders or abnormal physical states.
The stem cells or progenitor cells can be taken from an individual suffering from a neurodegellcld~ e disease and then di~lc.li~tçcl into neurons, astrocytes, oligodedrocytes for implantation into the nervous system of the individual. In a plcr~ ,d mode of the invention, cells may be t~ pl~ d into the CNS, PNS, spinal cord or other damaged tissues.
Thus, this invention overcomes the needs outlined above in that the precursor cells of this invention (1) are accepted by the patient becd lse they can be taken from the patient's own olfactory epithelium or tongue, (2) are safe in that the patient is not receiving cells or tissue from another source, (3) are effective in that the cells are of neural tissue origin and can be di~lç..li~te~ into neurons, astrocytes and oligodendrocytes for implantation and the cells survive during and after implantation, (4) offer the potential to provide long term relief of the symptoms associated with neurodegenerative ~ e~es, and (5) would not raise objections to the ethics of their use.
Therefore, this invention relates to isolated and purified precursor cells of peripheral tissues 5 with sensory lccelJtol~, such as the olfactory epithelium of a m~mm~l (juvenile or adult). Under applop,;ate conditions, the precursor cells can di~lcllliate into neurons, astrocytes or oligodendrocytes. The precursor cells may be transfected with a heterologous gene encoding, for example, a trophic factor. The precursor cells may then be implanted into the CNS, PNS, spinal cord or other damaged tissues of a patient and the heterologous gene c~lcssed.
This invention also relates to neurons, astrocytes and oligodendrocytes dirrele"li~ted from the precursor cells of this invention.
The invention also includes a ph~rm~ce~ltical co,ll~o~iLion for use in implant therapy. The composition inrhldes the precursor cells of this invention or neurons, astrocytes or oligodendrocytes di~~ ted from the precursor cells of this invention, in a ph~ cc;lll;~lly 15 acceptable carrier, auxiliary or excipient. The composition may include one or more types of cells selected from a group concieting of pl~ or cells, neurons, oligodendrocytes and astrocytes.
A method of treating an individual suffering from a neurodegt-nPrative disease is included within this invention. The method in~h1des implanting the precursor cells of this invention, or the neurons, astrocytes or oligodendrocytes derived from the p~ or cells of this invention, into the 20 CNS, PNS, spinal cord or other damaged tissues of the individual. Another method c.~ of treating an individual sllffPring from a neuro~ege~,.."l;~/e disease by ~lmini.ctering the ph~rm~ce~1tical composition of this invention to the individual.
This invention also includes a method for isolating and l~ul;rying ~ecul~o~ cells from the olfactory epithelium of a ...~...,..~I The method includes (1) taking a sample of the olfactory epithelium from the 111i1l11111~l, (2) dissociating the sample into single cells, (3) placing the cells in culture, (4) isolating the cells which survive in culture. These isolated cells may be dirrGl~ d into neurons, astrocytes or oligodendrocytes. The l,lGcul~or cells which survive in culture are spherical aggregates. The step of placing the cells in culture includes placing the cells in a tissue culture incubator in an applol,l;ate medium. We isolate precursor cells from the tongue and other peripheral tissues with sensory receptors using a similar technique.
In this method, the m~mm~l may be a human who is suffering from a neurodegenerative 5 disease, disorder (such as neurotrauma) or abnormal physical state. The method may further include implanting the precursor cells or the neurons, astrocytes or oligodendrocytes differçnti~ted from the neural stem cells, into the CNS, PNS, spinal cord or other damaged tissues of the human.
In another case, the m~n~m~l is a human and is not suffering from a neurodegenerative disease or neurotrauma. Then, the method includes implanting the l~le~ ol cells or the neurons, astrocytes 10 or oligodendrocytes di~~ ted from the precursor cells, into a second human who is snfff rin~
from the neurodeg~llelali~e disease or neurotrauma. The neurodegenerative disease may be one selected from a group consisting of Parkinson's disease, ~17hrimer's disease, Huntington's disease and Multiple Sclerosis, while types of neu~ ullla include stroke and spinal cord injury.
This invention also includes a kit for the ~ of a disease, disorder or abnormal 15 physical state. The kit inrl~ldes one or more types of cells inrhl~ing the ~lc~ Ol cells of this invention, or the neurons di~rc~. .1 ;~ted from these plc. u~or cells, the astrocytes di~el~ ted from these precursor cells and the neurons, astroycytes and oligodendrocytes di~c~ te~ from these precursor cells.
The invention also provides pl~ ol cell cultures which may be used in toxicity testing, 20 drug development testing or studies of genes and pl~t;hlS. Precursor cell cultures may also be in-lllced to produce healthy differenti~tecl cells which may be used for toxicity testing or drug development testing. Toxicity testing is done by culturing plC~ or cells or cells ~li~l . .1 ;~ted from precursor cells in a suitable medium and introducing a substance, such as a ph~rm~ce~ltic~l or chemical, to the culture. The ple~or cells or dirrelf ~ ted cells are exz~ ed to cl~e....i 25 if the substance has had an adverse effect on the culture. Drug development testing may be done by developing derivative cell lines, for example a pathogenic cell line, which may be used to test the efficacy of new drugs. Affinity assays for new drugs may also be developed from the precursor cells, differenti~ted cells or cell lines derived from the plec~sor cells or differrnti~ted cells. The methods of pelro~.l.ing toxicity testing and drug development testing are well known to those skilled in the art.
Precursor cells also provide a culture system from which genes, proteins and other metabolites involved in cell development can be isolated and identified. The composition of 5 stem cells may be col.lpared with that of progenitor cells and differenti~ted cells in order to determine the mPçh~ni~me and compounds which stimulate production of stem cells, progenitor cells or differenti~ted cells. Methods of isolating proteins and genes from cells are well known to those skilled in the art.
Other objects, features and advantages of the present invention will become appalelll from 10 the following detailed description. It should be understood, however, that the detailed description and the specific examples while in~iic~ting pl~r~ d embo~lim~nt~ of the invention are given by way of ~llplc only, since various cll~nges and modifications within the spirit and scope of the invention will become a~t;lll to those skilled in the art from this detailed description.
Brief Description of the Fi~ures The invention will now be described in relation to the figures:
Figure 1.
a) Bright field photograph of a small group of cells; 4 days in vitro ("DIV"). Scale bar = 40 ,um..
b) Bright field photograph of 3 floating olfballs; 12 DIV. Scale bar = 200 ,um..
c) Bright field photograph of 3 olfballs in the process of fusing; 12 DIV. Scale bar = 200 ,um.
20 d) Nestin ~ g of an olfball. 6 DIV and one day after plating down. Scale bar = 30 ,um.

Figure 2.
a) GFAP staining of di~.~ t.od olfballs. 16 days after plating down. Scale bar = 50 ,um.

b) GFAP st~ining of dirrel~.,l;~te~l cells derived from olfballs which had been passaged twice. 16 25 days after plating down. Scale bar = 50 ,um.

c) GC staining of di~~ iated olfballs. 16 days after plating down. Scale bar = 50 ~m.
d) Bright field of same field as shown in c). Scale bar = 50 Jlm.

e) GC staining of differenti~te(l olfballs derived from olfballs which had been passaged twice. 16 days after plating down. Scale bar = 200 ~lm.

Figure 3.
a) NF-160 staining of di~,~ te(l olfballs. 16 days after plating down. Scale bar = 50 ,um.
b) Bright field of same field as shown in c). Scale bar = 50 ,um.

c) LacZ staining of di~.~ 1 olfballs derived from T~ nlacZ mice (Gloster et al., 1994) that 10 express a neuron-specific E. coli ~-galactosidase marker gene. 16 days after plating down. Scale bar= 50 ~lm.
d) TH st~ining of di~le ,1 ;~te~l olfballs. 16 days after plating down. Scale bar = 50 ,~lm.
e) TH staining of di~,~ ted olfballs derived from olfballs which had been passaged twice. 16 days after plating down. Scale bar = 50 ~m.
15 f) Bright field of same field as shown in e). Scale bar = 50 ,um.
g) ~III tubulin st~ining of dif~,e- Il ;~te-l olfballs. 16 days after plating down. Scale bar = 100 ,um.
h) NeuN staining of di~.~ te~l olfballs. 16 days after plating down. Scale bar = 50 flm.

Figure 4 20 a) Bright field photograph of a small floating adult derived olfball; 8 DIV. Scale bar = 50 llm b) Bright field photograph of a larger adult derived floating olfball; 15 DIV. Scale bar = 50 llm c) nestin st~ining of a di~ tecl cell derived from an adult olfballs; 16 days after plating down.
Scale bar = 25 ~m Figure 5 r imitin~ dilution curve. Cells were plated at 700 to 7000 cells per well, cultured for 14 days in vitro, and then e~rninPd for the presence of olfballs. The fraction of wells without olfballs was 5 plotted against the number of cells plated. Based upon the Poisson distribution, the probability of a well not having an olfball at the 0.37 level (1/e) indicates that l of every 9000 cells plated has the capacity to generate an olfball. The correlation value of the line is r = -0.992.

Figure 6 10 Demonstration that the olfballs can be genetically modified. Olfballs were plated on polylysine in the presence of 2% FBS. CMV-~3-galactosidase adenovirus was added at an MOI (Multiplicity of infection) of 25. Three days later the cultures were X-gal stained (standard histochemical technique to reveal cells ~;Al.les~ g 13-g~l~ctosifl~ce), and 90% of cells were found to be ~Al~lCi,~ g J3-g~ tositl~ce Figure 7 Generation of TH-positive ll~uluns upon transplantation of olfballs into the adult rat slli~lulll.
The ~LliaLulll of adult rats was unilaterally dell~l~att;d using 6-hydroxydopalllille to elimin~te dop~ rgic fibers, and neonatal olfballs were transplanted into the ~Llialulll of the same 20 ~nim~lc. (a) With transplants of olfballs from Tal:nlacZ mice, B-galactosidase positive nuclei (arrows) are ~letec~ along the graft tract. (b) A complex TH-positive neuron (arrow) with multiple processes (arrowheads). (c) A cluster of morphologically simple TH-positive cèlls that are double-labelled with BrdU. Nûte the black speckled ap~eal~ue of the Brdu-l~hellin~
(arrow). (d) A TH-positive neuron (alluwhead) with a single process whose nl-rl~llc is double-25 labelled with BrdU (arrow). In this case, the BrdU staining fills the entire ml~lells. Scale bar:a = 100~m, b,c = 25~m, d = 5~m.

Detailed D~s.. ;"t;on of the Invention The inventors have isolated multipotential precursor cells from the olfactory epithelium of m~mm~l~ Guveniie and adult mice, adult rat and humans). The isolated cells proliferate in culture, so that large numbers of precursor cells can be generated. In culture, these cells form 5 floating spheres which are named "olfballs". These cells can be induced to differentiate into neurons, astrocytes, and oligodendrocytes by altering the culture conditions. The precursor cells can generate dirre.e"~i~te~l cells for use in autologous transplants for the tre~tm~nt of certain neurodegenerative disorders or neurotrauma. For example, precursor cells may be dirr~le."i~te~
into dopaminergic neurons and implanted in the substantia nigra or striatum of Parkinson's 10 disease patients. They can also be used to gen. lale oligodendrocytes for use in autologous transplants for multiple sclerosis. The p,c;.;u,~or cells are easily accessible by biopsy from the olfactory epithelium, so they are a ready source of cells for autologous transplants. Finally, they could be used as autologous cellular vectors to introduce growth factors into the ~li~e~ed or tr~-lm~ti7e~ CNS,-PNS, spinal cord and other damaged tissues.
The olfballs display some similarities to fo.~b.dill stem cells, but also possess some distinctive differences. In particular, (i) when olfballs dirrelellLiate in the p,ese"ce of serurn, almost half of the differ~nti~t~.i cells express neuronal m~rk~rs, whereas dirr. lÇ ~ tecl folel,.di stem cell neurospheres generate only a small p~ "Ldge of neurons, (ii) significant numbers of dopaminergic neurons are found in all dirr~,le"li~te~ cultures of olfballs. whereas they are never 20 found in cultures of fol~b,~ill stem cell neurospheres dirr~ te~ in serum, and (iii) many of the undifferçnti~tecl progenitor cells that are found in olfball cultures express glutamic acid-decarboxylase (GAD), a nculuLI~ enzyme that is expressed transiently in many neuroepithelial cells in vivo; in contrast, the only GAD-positive cells that derive from fol~b,di stem cell neurosphere cultures are neurons.
The pl~ ol cells of this invention may be used to prepare ph~. "~cc;~ 1 compositions which can be ~lmini~tçred to hllm~n.c or ~nim~lc Dosages to be ~rimini.ctPred depend on patient needs, on the desired effect and on the chosen route of ~ n;~l~dLion.

The invention also relates to the use of the cells of this invention to introduce growth factors into the (li~e~cetl damaged or physically abnormal CNS, PNS, spinal cord or other damaged tissues. The precursor cells act as a vector to transport a recombinant molecule, for example, or to transport a sense or ~nti~çn~e sequence of a nucleic acid molecule. In the case of a recombinant 5 molecule, the molecule would contain suitable ll~.s.;.;~ional or translational regulatory elements.
Suitable regulatory elements may be derived from a variety of sources, and they may be readily selected by one with ordinary skill in the art. If one were to upregulate the ~AI~lcssion of the gene, one would insert the sense sequence and the a~plo~.;ate promoter into the vehicle. If one were to downregulate the ~I,.ession of the gene, one would insert the ~nti~çn~e sequence and the 10 a~.u~.;ate promoter into the vehicle. These techniques are known to those skilled in the art.
Examples of regulatory el~m~nt~ include: a transcriptional promoter and enh~nc~r or RNA
polymerase binding sequence, a ribosomal binding sequence, including a translation initiation signal. Additionally, depending on the vector employed, other genetic elements, such as selectable markers, may be inco~o~aled into the recombinant molecule. The recombinant molecule may be 15 introduced into the p~e~or cells or the cells ~ c .li~ted from the precursor cells using in vitro delivery vehicles such as retroviral vectors, adenoviral vectors, DNA virus vectors and liposomes They may also be introduced into such cells in vivo using physical techniques such as microinjection and ele~ pGldLion or ~h~mic~l methods such as co~,recil,;ldlion and incorporation of DNA into liposomes. The genetically altered cells may be ~n~ps~ ted in microspheres and 20 implanted in the CNS, PNS, spinal cord and other damaged tissues.
The following examples describe (i) the derivation of olfballs from postnatal mouse and adult mouse tissue, (ii) the derivation of olfballs from rat and human tissue, (iii) the use of olfballs to gcn~,ldle endogenous CNS cell types in the transplanted adult mouse brain, (iv) methods for genetically manipulating olfballs for use as theldpeulic vectors, (vi) isolation of 25 ~.l.,CUl~Ol cells from other peripheral tissues with sensory receptors such as tongue We ~h~ract~rize and use these cells using procedures similar to those used with olfballs. These studies provide us with novel tools for the l1C~ I of the trallm~ti7~d or rli~e~ed adult nervous system.

Example l - Isolatin~ Multipotential Precursor Cells from Postnatal OlfactorY Epithelium of Mice Postnatal mice were stunned with a blow to the head and then decapitated. The heads were sagitally sectioned with a razor blade. The olfactory epithelium of about 6 postnatal (P1-9) 5 mouse pups were stripped from the conchae, nasal septum, and vomeronasal organs using watch-maker forceps. This tissue was placed into 3mls of media (DMEM/F-12 1:3 (Hyclone media) supplemented with 2% B-27 (Gibco), 20 ng/ml EGF (Collaborative Research), 0.1% fungizone, 0.5 ml/lOOml penicillin/streptomycin (Gibco). After epithelium from the postnatal pups was collected, the epithelium was teezed apart with watch maker forceps, releasing a large number of 10 single cells. The media was transferred to a 15 ml tube, and 7 ml more media was added. The cells were dissociated into single cells, by titration with 1 Oml plastic pipette (Falcon), and passed through a 60 micron filter (Gibco). Typically dissociated cells from the olfactory epithelium from 6 pups was plated into 2 SOml tissue culture flasks (Falcon). The dissociated cells were then placed in 50ml flasks in a 37~C, 5% C02 tissue culture incubator. Two days later most 15 cells in the cultures were dead or dying. However, a small number of large phase bright cells were present, most of which attach to the flask bottom. Over the next 2-6 days these cells divided and produced spherical ag~ gales which became larger over time. On day 4 (FIG lA) there were apl)roxilllately 500 clusters of dividing cells per pup used in the original isolation (n=2 independent isolations). Most of these cellular aggregations lifted from the flask surface 20 over the next few days (FIG IB). These floating spheres (olfballs) continlle~l to grow and fused together to become macroscopic (FIG 1C), re~rlling 100 microns in ~ meter if left for 10 days days in vitro. After 14 days in vitro, the diameter of the spheres was approximately 1 mm.
If EGF w~ not added to the media, small clusters of dividing cells were still seen by day 4, and some of these cells developed into olfballs, snggestin~ that the cells were producing trophic 25 factors themselves in q~l~ntitiP~ which in some c~es w~ sufficient for their proliferation.
The cells in these dividing clusters c;~ essed a marker for neural progenitor cells and neural stem cells, the int~rrnp~i~te fil~m~nt protein nestin; at six days, olfballs were transferred to polylysine coated 35mm dishes overnight in media co,.~ g 2% fetal bovine serum to facilitate the cells adhering to the sub~lldlulll, and were processed for indirect nestin immunohistochemisty. Filamentous antibody staining was observed in almost all the cells in the clusters (FIG lD).
These nestin positive cells could also be passaged. Six days after isolation, the media 5 (Sml) was removed from the flasks. This media contained many olfballs that had lifted from the flask surface. The media cont~inin~ olfballs was titturated with a fire polished pipette, thereby dissociating many of the cell clusters into single cells, and placed in a larger flask with an additional 1 5ml of fresh media (total volume now 20ml). After a further 6 days one quarter of the media was removed, the olfballs were again ll;Luldlcd, and put into a new flasks with 1 5ml 10 fresh media and EGF. These cells have been succe~sfully passaged four times.
Example 2 - Differenti~tin~ Precursor Cells Into Neurons. Astrocytes and Oligodendrocytes After the cellular clusters of Example 1 had been generated they could be differenti~
into neurons, astrocytes, and oligodendrocytes. Clusters from cultures 7 to 14 days after isolation were plated down onto polylysine coated 35mm culture dishes (Falcon) and 4 multiwell 15 culture dishes (NUNC), in DMEM/F12 media co..l;.i~-i.-g 2% fetal bovine serum (Hyclone) and 2% B-27 (no EGF). Media was changed every 3-4 days. Over the next 6-19 days cells migrated out of the olfballs onto the dish surface. Some of these cells had the morphology of neurons, astrocytes, or oligodendrocytes. We deterrnined the phenotype of these cells using marker antibodies to glial fibrillary acid protein (GFAP) (FIG 2A, B) for astrocytes, antibodies to 20 neurofil~nnent 160 (NF-160) (FIG 3A), ~ III tubulin (FIG 3G), NeuN (FIG 3H) for neurons, and antibodies to galactocerebroside (GC) (FIG 2C-E) for oligodendrocytes. Antibodies to tyrosine hydloxylase (TH) were used to identify dopalllinc~gic, noraLcnergic, and adlellergic neurons (FIG 3D-F). Dop~ine ~-dehydrogenase (DBH) was also used for nora~Lc".,l~,ic and a~l,el~ic neurons.
Immunohistochemical procedures. With the exception of GC immlmohistochPnni~try, culture dishes were washed twice with TBS (Tris Buffered Saline; 1 OmM Tris, 1 50mM NaCl, pH 8), then fixed with 4% pal~fol,llaldehyde, rinsed in three times with TBS, blocked with TBS
plus 2% goat serum (Jackson Tmmlm~Research), and 0.1% Triton-X (Sigma) for 30 min, then incubated with primary antibody in TBS plus 2% goat serurn, rinsed 3 times with TBS, inc~lb~tçcl in secondary antibody in TBS plus 2% goat serum, rinsed 3 times and then viewed under a Zeiss Axiovert 100 florescence inverted microscope. The antibodies to GFAP (BoehringerMannheim), ~III tubulin (Sigma and a gift from Dr. D. Brown, U. Ottawa), NeuN (Dr. R.
5 Mullen), NF-160 (American Tissue Culture Collection) were monoclonals used at concentrations of 1:200; 1:25; 1:10, and 1:1 respectively. Antibodies to nestin (a gift from Dr. Ron MacKay (Nation Institute of Health), TH (Eugenetech), and DBH (Eugenetech) were rabbit polyclonals used at concentrations of 1:1000, 1:200, and 1:200 respectively. Secondary antibodies were Cy3 conjugated goat anti-mouse (Jackson Tmml-noResearch) and Cy3 conjugated goat anti-rabbit (Jackson TmmllnoResearch), and were used at 1:1500. For double-labelling experiments FITC
goat anti-mouse (Jackson ImmunoResearch). GC imml-nohist~ch~nni.~try, living cultures were incubated with a DMEM media; HEPES; 5% HS (heat inactivated horse serum), and 1:10 GC
antibody (BRD1; a gift of Dr. B. Juurlink U. Sask.) for 30 min at 37~ C, rinsed 3 times with the medialHEPES/HSr fixed with 4% paraformaldehyde for 15 min, rinsed 3 times in TBS, inr~lb~ted in Cy3 conjugated goat anti-mouse antibody (1:1500) for 2 hr., finally rinsed 3 times in TBS. Cultures processed for immlln- histoch. nni~iry without primary antibodies revealed no st~imng.
Astrocytes, neurons, and oligodendrocytes were found. We also cultured olfballs from transgenic mice which express ~-galactosidase offof the neuron specific promoter Tal a-tubulin, 20 which allowed us to use st~ining with the ligand X-gal antibodies for ~ g~l~ctosidase as an additional neul~nal marker (FIG 3B, 3C).
Since the di~.r .,li~ted cells abutted each other and were piled up on top of each other in the center where the olfball originally ~tt~ it was not possible to count the number of cells eAlJle~ g each marker. The majority of cells that migrated out of the clusters were GFAP
25 positive while a large number of cells were either NeuN or lacZ positive. A lower number of cells were NF-160 positive, ,~ III tubulin, TH, GAD or GC positive. The.~ ~ole the olfballs could differentiate into neurons, astrocytes and oligodendrocytes. While a few of the ~III tubulin positive cells had complex morphologies (FIG 3A), most were simpler, pos~essing only a few neurites. The TH positive cells were the most morphologically complex cells in the cultures, with numerous neurites e~ten~ing from the cell body (FIG 3D, E). These TH positive cells are probably dopaminergic neurons and not noradrenergic or adrenergic neurons, since no cells were found to be DBH positive. Significantly, no TH, GFAP or GC positive cells have ever been 5 reported in vivo in the nasal epithelium. Therefore the olfactory derived nestin positive olfball cells could be differenti~ted into cell types never found in the olfactory epithelium-oligodendrocytes, astrocytes, GABAergic neurons, and dopaminergic neurons. The coexpression of astrocytic and neuronal markers has been reported for di~e,~ ;Ated cells derived from EGF-gellGId~Gd brain-derived progenitor cells (Peel et al., 1995). While most cells were either lacZ or 10 GFAP positive, there were a few cells which were both lacZ and GFAP positive, however none of the TH positive cells were also GFAP positive. Therefore while cells may transiently express both neuronal and glial markers during their di~G~GIlliation program, fully di~le. .l ;~tecl morphologically complex neurons express only neuronal markers.
Like the original olfballs, the passaged olfballs could also be di~le.,l;~tecl into neurons, 15 astrocytes, and oligodendrocytes. Olfballs which had been passaged twice were plated down on polylysine coated dishes. The olfballs cells migr~t~d out and spread out over the dish's surface, and after 16 days were immnno-positive for GC (FIG 2E), GFAP, ~III tubulin, NeuN, lacZ, and TH.
The plOpOl lion of cells positive for the various m~rk~r~ was similar to that seen in the di~Gle ~ 11 ;Z~ted cultures from the original cultures.
~0 Example 3 - Isolating Multipotential Precursor Cells from OlfactorY Epithelium of Adult Mice and Adult Rats Similar proliferating cells were also isolated from adult mouse and rat olfactory epithelium and vomeronasal organs. We developed techniques for reproducibly culturing, pacs~Ein~, and di~l~ tin~ the adult olfballs, on the basis of our experience with their juvenile 25 equivalents. As part of this aim, we (i) characterized the growth factor and media requirements for the adult cells to proliferate in culture, and (ii) characterized the growth factor and substrate e~luhGlllents for the di~le"liation of oligodendrocytes and do~i.. " i l .f . gic neurons from both adult and juvenile olfballs. We were informed in these studies by similar work on EGF- and FGF- dependent stems cells from the CNS, since olfballs likely respond to at least some of the same growth factors. The adult isolation procedures were essentially the same as for the postnatal olfballs (described in examples 1 and 2).
Adult mice and rats were anaesthetized with injected with an overdose of somnitol, and 5 then decapitated. The olfactory and vomoeronasal organ epithelia were stripped from the conchae and nasal septum and incllb~ted in F12/DMEM culture media for 1 or 2 days after their dissection and prior to the rest of the isolation procedure (B). After this incubation, the epithelia was dissociated in an identical manner as the postnatal epithelia. Two days after the isolation almost all the cells were dead with the exception of a very few large phase bright cells. These 10 cells divided over the next few days, forming small clusters of dividing cells similar to those seen in the postnatal cultures (FIG 4A, B) . These also grew to give rise to the large floating clusters which were routinely seen in the postnatal cultures. After 6 divisions some of these clusters began to dirre~ liate and spread out over the flask's surface, while some other clusters which had floated le~ h.od to the surface and then dirreie .l;~te~ (These cells multiplied to produce the small balls 15 or cells, but did not grow to form the large balls of cells like the postnatal cultures). We passaged these cells using the same procedure as that described above with respect to the cells isolated from postnatal olfactory epithelium.
These proliferating cells from the adult were also nestin positive. 10 days after their initial isolation the cells were transferred to polylysine coated dishes with 2% fetal bovine serum 20 (FBS). Two hours later the cells were processed for nestin immllnohistoch~?nti~try (FIG 4C).
After the cellular clusters of this Example had been gen~lal~d they could be dirre~ te(1 into neurons and oligodendrocytes. Clusters from cultures 7 day after isolation were plated down onto polylysine coated 35mm culture dishes and 4 multiwell culture dishes, in media co..l~it-il-g 2% fetal bovine serum and 2% B-27 (no EGF). Over the next month cells migrated out of the 25 olfballs onto the dish surface. We det~ minlod the phenotype of these cells using marker antibodies to glial fibrillary acid protein (GFAP) for astrocytes, antibodies to ~III tubulin for llt;UIOilS, antibodies to TH for dopa~ lergic neurons and antibodies to galactocerebroside (GC) for oligodendrocytes.

Neurons, and oligodendrocytes were found, although the number of these cells was much lower than the number obtained from the neonate. The phenotype of these adult derived dirrelc. ,li~tf ~ cells was ~c~es~e~l using indirect immlmohistochf mictry. The cells isolated from the adult were differenti~ted into ~III tubulin positive cells (neurons), tyrosine hydroxylase positive 5 cells (dopaminergic neurons), galactocerobroside positive cells (oligodendrocytes). No astrocytes (GFAP positive) cells were found. Therefore the adult derived olfballs could dirrerellliate into neurons and oligodendrocytes.
Example 4 - Precursor Cells Differentiate Into Neurons When Transplanted Into Adult Brain The major potential the,~eulic use for olfballs is autologous transplantation into the 10 injured or degenerating CNS, PNS, spinal cord and other damaged tissues, either to replace lost cell types and/or as vectors for expression of therapeutic molecules. Transplantation ex~efilllents determine the fate of transplanted olfactory-derived precursor cells. The precursor cells can dirrel~llliate into neurons when transplanted into the adult brain. To this end, we transplant mouse derived precursor cells into brains of immnn~ ~ul)pl~ssed adult rats and identify which of 15 the transplanted cells dirrelc;lltiate into neurons, using double labelling with the mouse specific and neuron specific antibodies (such as those which recognize neuron specific enolase and neuron specific ~ -tubulin). A similar approach has proved s~lccessful in the study of transplanted brain-derived stem cells (Winkler, ~mm~ng, and Bjorklund, 1996).
In order for these stem cells to be useful for transplantation to treat neurodegenerative 20 ~ e~es it is necessary to induce the ~lirr~l~llliation of the app,~ p,;ate neuronal phenotype, such as dopaminergic nc;ulolls in the case of Parkinson's disease. Therefore, initially we ~ e if the precursor stem cells transplanted into lesioned and unlesioned stri~tllm and s~lbst~nti~ nigra, differentiate into dop~minPrgic neurons in ,espol1se to signals from their new environment, as they do when they dirr.,l~,l,liate in vitro. Brain sections are double labelled with a mouse specific 25 antibody and antibodies to tyrosine hydroxylase to reveal dop~minf rgic neurons derived from the transplanted cells. Transplants into neonatal rat brains show that a more illlllli~ e host environment is able to induce do~ fl~,ic dirr~lel,liation.

We transplanted olfballs into the denervated and intact striatum of adult rats.
Specifically, we unilaterally destroyed the dopaminergic innervation of the adult striatum by a local infusion of 6-hydroxydopamine, under conditions where noradrenergic neurons are spared.
Several weeks following this lesion paradigm, olfballs were transplanted into both the intact and 5 lesioned striatum, and one week later, the fate of the transplanted olfballs was ~letç~min~d immunocytochemically. These studies demonsl~dled that transplanted olfballs can differentiate into tyrosine-hydroxylase-positive neurons in vivo, as they can in vitro. Given that the primary deficit in Parkinson's disease is a loss of dopaminergic innervation of the striatum due to neuronal loss, these preliminary studies raise the exciting possibility that olfballs provide an 10 autologous source of dopaminergic neurons with which to treat this disease.
We characterize the neuronal and glial cell types that are generated by olfballstransplanted into the adult striatum. In order to definitively identify the progeny of the transplanted olfballs, we (i) derive olfballs from transgenic mice ex~lessillg ~-galactosidase from either the Tal a-tubulin and/or myelin basic protein promoters, thereby m~rking both 15 transplanted neurons and oligodendrocytes, and (ii) rely upon the use of a mouse-specific antibody to distinguish all of the transplanted cells. ~It.?rn~tively, we tag the olfballs with a ~-galactosidase marker gene in vitro prior to their transplantation. We then double-label cells immunocytochemically with markers for the dirrele,ll cell types, such as tyrosine hydroxylase for dopaminergic neurons, galactocerebroside for oligodendrocytes, and GFAP for astrocytes. Using 20 this approach, we tagged the stem cells with BrdU, followed them and found double labelled GFAP and TH-positive cells.
Example S - P,~ or Cells Differentiate In Vitro We ~tt-.min~ whether there are in vitro conditions which promote the dirrerellliation of precursor cells into dirr~le,ll neural phenotypes. To identify such conditions the effect of a 25 variety of substrates, conditioned media, and growth factors are tested. We test the substrates l~minin, fil,lolle~;lill, heparin proteoglycans. Conditioned media we test includes media from cultures growing neonatal heart cells, astrocytes and stiatal neurons. We test the growth factors NGF, BDNF, NT-3, NT-4/5, bFGF, EGF, TGFa, TGF,~, LIF (le.lkemi~ inhibitory factor) and CNTF (ciliary neurotrophic factor), PDGF, GDNF and neurturin.
Exarnple 6 - The Extent To Which the Local Environrnent Determines Neural Cell Fate It is useful not only to generate neurons, but also oligodendrocyytes for transplantation 5 therapy. Another basic question that we address is the extent to which the local environment ~eterrnine neural cell fates. Outside of their norrnal environment, olfactory-derived progenitor -~ cells are not restricted to differçnti~ting into olfactory neurons. This ~ c~h~ent cleterrnines whether the converse is also true. We determine whether brain-derived stem cells co-cultured with olfactory epithelium differentiate into olfactory neurons. Co-cultures c~Jcfi~ents with 10 various explants and stem cells derived from various nervous system sources provide ExamPle 7 - Human Precursor Cells Derived From Human Nasal EPithelium much information on the role of the local environment in ~ g cell fate.
If olfactor,v-derived neural stem cells are to be used for autologous transplants for the tre~tment of neurodegel~ldlive disorders it is n~cess~. y to show that they can be gencldled from 15 human nasal epithelium. During certain neurosurgical operations nasal tissue is removed. We isolate human olfactory-derived neural stem cells from this tissue as it becomes available. We use the same procedures as we used to isolate the neural stem cells from the neonate and adult mice. As is known to those skilled in the art, work on primate and human neurospheres with regards to culture conditions provide information on o~)t;..,i7.;,-g proliferation of cell spheres and 20 specific cell types.
Example 8 - Transfection Of Olfactory-Derived Plecul~or Cells And Usin~ Precursor Cells In Autolo~ous Cellular Vectors To Introduce Growth Factors Into The Diseased Or Tra Im~ti7~cl CNS~ PNS~ Spinal Cord And Other Damaged Tissues As a therapy for neurodegell~,.d~i~e ~ e~es~ transplanted cells may have to be genetically 25 engineered so that they can survive the insults that caused the original neurons to die, and thercforc it would be advantageous to be able to transfect prc.;ul~ol cells. In addition, the transfected olfactory progenitor could be used as vectors for introducing biologically active molecules into the brain of patients with neurodegenerative disorders. If olfballs are to be used as autologous transplantation vectors for expression of theld~cuLic molecules, it is essential to develop mech~ni~m~ for genetically manipulating them. We determine whether recombinant adenovirus vectors can be used for this purpose. We have previously used recombinant 5 adenovirus to manipulate both postmitotic sympathetic neurons and cortical progenitor cells, with no cytotoxic effects when used under controlled conditions. We infect olfballs with a ,B-galactosidase-cx~lessing adenovirus, and clete mine (i? how long the tr~n~ cecl marker gene is expressed, and (ii) whether this manipulation affects the growth and/or differentiation of olfballs in vitro. Then we transplant ~-galactosidase-ex~lcs~ing olfballs in vivo, and measure the same 10 parameters. Finally, we use recombinant adenovirus to overcAy,css brain derived neulollophic factor ("BDNF") or GDNF in olfballs, and determine (i) whether autocrine BDNF or GDNF
causes increased survival and differentiation of olfball-generated neurons in vitro or in vivo, (ii) whether autocrine BDNF causes increases numbers of dop~minergic neurons to differentiate from olfballs in vitro or in vivo, and (iii) whether BDNF-cxylcs~hlg or GDNF-cAyiessing 15 olfballs transplanted into the substantia nigra or striatum can protect endogenous dopaminergic neurons from chemical lesions. We pursue this CAy~ Pnt to effectively manipulate these stem/progenitor cells using recombinant adenovirus.
For example, if a trophic factor was useful in treating a neurodegenerative disorder, then neural progenitor cells or neural stem cells transfected with genes coding for the trophic factors, 20 could be transplanted into a patient to provide a continuous source of the trophic factor at the transplantation sight. We use various strategies to transfect the stem cells including lipofe~ e and viral transfections including herpes and adenovirus. We use lipofect~mine and adenovirus me~i~te~l transfection strategies, which are well known to lcsearchers in the field, to transfect the stem/progenitor cells. For lipofe~ e transfections, we follow the standard protocol as outline 25 in the lipofect~mine product information sheet which comes with the lipofectAMINE Reagent purchased from Gibco. For adenovirus mediated transfections we follow the procedures outlined in Gage et al., 1995, Le Gal et al. 1993 and Slack and Miller, 1996.

Example 9 - Limitin~ Dilution Analysis of Isolation And Transplantation Of Accessible Multipotential Neural Pro~enitor Cells From The OlfactorY Epithelium To determine whether individual olfballs derived from single cells, we performedlimiting dilution analysis. This analysis demonstrated that the limiting dilution curve was linear, 5 displaying single hit kinetics, indicating that individual olfballs were clones of single cells.
Moreover, this analysis demonstrated that cells capable of giving rise to olfballs were present at a frequency of approximately one in 9000. Differentiation of these individual olfballs demonstrated that they were multipotent, as were olfballs in mass cultures. Specifically, double-label immunocytoch-?mi~try revealed that olfballs derived from single cells were capable of 10 generating astrocytes, oligodendocytes, and TH-positive neurons (Figure 5). Cells positive for i) GFAP and GC, ii) GFAP and TH, and iii) GC and TH were generated. Thus, approximately one in every 9000 cells from the dissociated olfactory epithelia is capable of gel~.a~ g an individual multipotent olfball.
To ~letçrmin.o whether individual olfballs derived from single cells, we p~.rulllled limiting 15 dilution analysis (Lefkowitz & Waldman, 1989). Specifically, diluted, dissociated cells from neonatal olfactory epithelia were cultured into 96 well plates, and the presellce or ~ksçnce of individual olfballs scored after 14 days (Bellows and Aubin, 1979). This analysis demonstrated that the limiting dilution curve was linear, displaying single hit kinetics, indicating that individual olfballs were clones of single cells.
For the limiting dilution ~l,. .;,llents, olfactory epithelia from P5 mouse pups were dissociated and cultured as described above, except that the cells were plated in 96 well dishes at cell ~l~mities ranging from 700 to 7000 cells per well. After 14 days in culture the wells were elr~min~cl for the presence or absence of olfballs. Dirrelellliation of the solitary olfballs was p~ilrolllled as for the mass cultures described above, and the wells labelled immlml)histochemically as described above, with primary antibodies to the following combinations of antigens; GFAP and TH, GC and TH, and GFAP and GC.

Example 10 - Genetic Modification of Olfactory Precursor Cells As discussed above, a number of studies have demonstrated that adenovirus-based vectors can effectively tr~nccluce postmitotic neurons of the central nervous system (CNS) in vivo, and cells derived from the CNS in vitro (Le Gal et al., 1993, for review see Slack and 5 Miller, 1996).
In some circ-lm~t~nces it would be important to modify neural progenitor cells prior to -~ their transplantation as therapy for neurodegenerative ~ e~es, since the transplanted neurons may have to be genetically engineered to survive the insults that caused the original neurons to die. Olfactory precursor cells can be transfected using the adenovirus gene transfer system. We 10 established that the olfactory epithelial-derived stem cells can be successfully transfected with high efficiency and low toxicity, using ~3-galactosidase as a marker gene (Figure 6). A
recombinant adenovirus carrying the J3-galactosidase reporter gene inserted in the deleted E1 region was used in transfection e~. ;,.,entc Multiplicity of infection (MOI) was calculated based on titration on cells for adenovirus-15 based vectors, and ~pl~ese~ the number of plaque-forming units added per cell.
Staining for ~ s~ion of the ~3-galactosidase marker gene was ~lro""ed. Cells were fixed with 0.2% glutaraldehyde in PBS (pH7.4) for 15 min at 4~C. After two washes with PBS, cells were inc~1batecl for 18 h in X-gal stain (2 mM MgCI2, 1 mg/ml X-gal, SmM K3Fe(CN)6, and S mM K4Fe(CN)6 in PBS (pH 7.4). To estim~te the pclc~ age of cells that were infected, the 20 total cell number and lacZ-positive cells were counted in five random fields. The data were expressed as the average of two separate t;xl.c. ;"~tqnt~ with error bars lep,~se l~ g the range.
Example 1 1 - Dt;l . " ~ I ;on of the intrinsic factors such as transcription factors that regulate cell fate ~ ,.,;"~tion.
To address whether candidate transcription factors, which have been identified in various 25 ~ . ;",ent~l systems, can regulate neural fate decisions we transfect these factors into the olfactory epithelium-derived stem cells using the adenovirus gene ~ rel system.

We transfect the olfactory epitheliurn derived stem cells with the candidate transcription factors including IsI1, en-1, en-2 and nurr, which have been implicated in regulating motoneuron and striatal phenotypes. With an understanding on how intrinsic and extrinsic factors regulate neural cell fate decisions, it will be possible to induce the differentiation of the specific neural 5 cell types required for neuronal transplant therapy.
Example 12 - Characterizing Olfactory Epithelial Derived Precursor Cells.
We characterize the growth and dirrelc,'liation of olfactory epithelia (OE) derived precursor cells, and their potential for use in transplantation therapy for neurodegenerative e~eee We previously isolated an EGF dependent population of multipotential neural 10 precursor cells from the olfactory epithelium, and demonstrated that they can dirr~ Liate into CNS phenotypes including astrocytes, oligodendrocytes, and neurons. We characterize these precursor cells, the intrinsic and extrinsic factors which regulate the neural cell fate these cells adopt upon differentiation, and the potential for using these cells in transplantation therapy. We isolate stem cells from a patients olfactory epithelium, expand these in culture, dirr~lellLiate them 15 into the desired neural phenotype, and then transplant these cells back into the patient to help reverse the functional deficit. This strategy avoids problems associated with immlml)rejection since the tr~nepl~ntec~ cells are derived from the p~tiPnte themselves.
We ~etermine whether i) there is more than one population of olfactory epithelium precursor cells ii) these precursor cells are in fact stem cells iii) there are any negative changes 20 associated with long term culture and iv) whether these cells can be isolated from human olfactory epithelium. We ~ e whether there is more than one population of precursor cells in these cultures analogous to the quiescent and relatively qnieSc~nt populations of brain derived stem cells, by limiting dilution analysis using trophic factors, which have been previously shown to support the proliferation of stem cells isolated from various regions of CNS such as EGF, 25 TGF, and bFGF. We look for additive and synergistic interactions between the trophic factors.
We demonstrated that these cells are multipotential and can be passaged up to 5 times, however, we propagate them over longer periods of time, while still ret~ining their multipotentiality. We also e~r~nnine whether there are any cytogenetic changes or changes in growth char~ctc-ri~tics after repeated pac~ging, to rule out possible neoplastic changes accompanying long term culture.
We isolated olfactory epithelium derived stem/progenitor cells from mice and rats. We generate olfactory epithelium derived stem cells from humans olfactory epithelium, to demonstrate the feasibility of using autographs in the treatment of neurodegenerative disease.
S Example 13 - Inducin~ Human Precursor Cells to Differentiate into Neurons The ~xl~cl;lllents to induce human cells to dirrelen~iate into neurons, astrocytes, and oligodendrocytes are done in the same manner as the mice ~x~ hnents (see example 2-3).
We induce precursor cells to differentiate into neurons by culturing them with growth factors. This is done similarly to the differentiation of precursor cells in example 2.
We demun~lldl~d that OE derived stem/progenitor cells can differentiate in the same culture into oligodendrocytes, astrocytes, and neurons. We analyze the roles of trophic factors.
Extracellular substrates are tested which we show are implicated in regulating differentiation of neural cell types in vitro and in vivo. Specifically, we demonslld~e the effects of brain-derived n~ulolluphic factor (BDNF), ciliary neul~o~ophic factor (CNTF), triiodolhylo~ e (T3), bone morphogenetic proteins (BMPs), platelet-derived growth factor (PDGF) and sonic hedgehog (SH). We identify growth factors that direct OE-derived stem cells to différentiate into specific cell types.
Example 14 - Inducing Human P~e~;ul~or Cells to Dirr~lcllliate into Astrocvtes We induce pl~-;ul~ol cells to differentiate into astrocytes by cnltllrin~ them with growth factors. This is done similarly to the dirr~LeLlliation of ple~,ul~or cells in example 2.
Example 15 - Inducin~ Human Precursor Cells to Dirr~lel,liate into Oligodendrocytes We induce ~lecul~or cells to dirr~iellliate into oligodendrocytes by c~lltllring them with growth factors. This is done simil~rly to the differentiation of precursor cells in example 2.
Example 16 - Using Neural Stem Cells or Neuronal Pro~enitor Cells in Autolo~ous Transplants In Tre~tment of Parkinson's Disease We grow olfactory epithelial derived stem cells in vitro and dirre~ iate these cells into specific neural cell types. These cells are useful therapeutically in the treatment of neurodegenerative di~e~es such as Parkinson's disease and multiple sclerosis. We implant dopamingeric neurons differenti~tecl from precursor cells into the substantia nigra or the striatum of patients having Parkinson's Disease.
The adult olfactory epithelium could be used as an autologous source of stem/progenitor cells for cellular replacement therapy in the diseased or tr~nm~ti7~ central nervous system. We focused on TH-positive neurons, which are lost in Parkinson's disease (17), and which could be differenti~tec~ from adult and neonatal olfballs, as shown here. To perform these ~ hllents, the dopaminergic innervation to the adult rat striatum was first unilaterally lesioned with the 10 chemotoxin 6-hydroxydopamine, and the efficacy of the lesions was tested two weeks later by amphetamine-in-l~lceA rotational behavior (18). Two days prior to transplantation, rats were immunosuppressed with cyclosporin. Olfballs were then stereotactically injected into the caudate-putamen complex on both the lesioned and unlesioned sides (18). Sixteen days following transplantation, ~nim~lc were sacrificed, and sections of the stri~hlm were analyzed 15 immlmocytochemically for nestin and TH (19). Five of 8 ~nim~l~ received sllccescful injections of olfballs in the stri~t lm Of these, 4 ~nim~l~ showed evidence of a nestin-positive tract on both the lesioned and unlesioned sides, although tracts on the lesioned side appeared to be more intensely nestin-imml.noreactive (data not shown). On ~djacent sections, TH-positive cells were observed confined to an area close to the transplant tract on both the lesioned and unlesioned side 20 (Fig. 7b-d). As many as 25-30 TH-positive cells were identified on each section. Cells varied in morphology from small round cells without processes, which may be neuroblasts or early postmitotic neurons, neurons with a single process, or a minority of neurons that were morphologically complex with multiple fine processes (Fig. 7b). In some cases, the processes of these TH-positive neurons ext~n(lecl into the striatum for rli~t~n~es of up to 300 llm. However, 25 the cell body size of even the morphologically-complex TH-positive neurons were small relative to adult dop~minergic neurons of the ~ub~ ia nigra.
To COI~ulll that these TH-positive neurons derived from the olfballs, we performed two sets of ~. .h~ents in which the transplanted cells were "tagged". In one set of e~e~ .ont~, transplanted olfballs were derived from Tal:nlacZ transgenic mice (21), in which the neuron-specific Tal a-tubulin promoter drives expression of a nuclear-localized ~3-galactosidase marker gene. Immunocytochemical analysis of ~nim~l~ receiving the transgenic olfballs (18) revealed the presence of 13-galactosidase-positive neurons within the transplant tract (Fig. 7a), confirming that the transplanted olfballs generated neurons in vivo as they did in vitro. In a second set of 5 ~ hllents, olfballs were labelled with BrdU for 18 hours, washed to remove the BrdU label, and the labeled cells transplanted unilaterally into the 6-hydroxydopamine-lesioned striatum of ~nim~l~ (10 rats/4 mice) prepared as described (18). Immunocytochemical analysis with anti-BrdU (22) revealed that all ~nim~l~ showed evidence of BrdU-positive transplant tracts. This BrdU-labelling took the form of a few blue-black nickel DAB labelled speckles (Fig. 7c), or a 10 coalescent solid nuclear pattern identified within a brown cytoplasmic background (Fig. 7d).
Tmm-lnocytochemictry with anti-GFAP revealed that, in both xenografts and allografts, GFAP-positive cells with heterogeneous morphology were concentrated at the transplant site, but were also present in moderate amounts over the entire ipsilateral h~ phere, with additional scaLL.,l~,d reactive astrocytes seen in the contralateral hemi~phPre. GFAP-BrdU double-labelled cells were 15 present mainly within or close to the transplant tract, and varied in morphology from small, round cells with only a few processes, to large polygonal or ru~iro~ cells with multiple processes. Tmml~nocytochPmi~try with anti-TH revealed that TH-BrdU double-labelled cells were also present, although these were few in number relative to GFAP-BrdU positive cells.
BrdU-TH double-labelled cells were mainly small to medium-sized without processes (Fig. 7c), 20 although some examples of double-labeled cells with processes were found within (Fig. 7d) and adjacent to, the transplant tract. Thus, olfballs genel~led astrocytes and neurons in vivo, and a subpopulation of the latter were TH-positive.Together, these fin~linEs show that multipotent stem/progenitor cells derived from the olfactory epithelium are capable of generating cell types never found within the olfactory epithelium, including oligodendrocytes and TH-positive 25 neurons. Moreover, TH-positive neurons can be generated not only in culture, but also in vivo in neural transplants. The implications of these fin-lings are two-fold. Similar stem/progenitor cells can be derived from biopsies of the olfactory epithelium of adult Parkinson's disease patients, and used as an autologous source of neurons for transplantation. Current sources of dopaminergic neurons for neural transplantation are derived from human fetal tissue, a nonautologous source that is limited by tissue availability, potential im mune rejection, and ethical issues (24). Moreover, although neural stem cells from the CNS are potential alternative sources, they have not yet been demonstrated to generate dopaminergic neurons, and are essentially inaccessible without invasive surgery. Similarly, olfballs are an autologous source for 5 transplantation in other neurodegenerative disorders, since they generate other neuronal phenotypes, as well as oligodendrocytes.
Second, these fin-ling~ demonstrate that a stem/progenitor cell from a peripheral, placodally-derived neural tissue is capable of generating cell types such as oligodendrocytes found only in the central nervous system. In fact, olfballs are similar, in many aspects, to 10 neurospheres, the previously-described EGF-dependent stem cell from the ."~."",~ n forebrain (5). There are, however, a number of major dirrelellces between olfballs and neurospheres.
Although olfballs and neurospheres both readily differentiate into GFAP and GC-positive nonneuronal cells, only olfballs spontaneously ~elleld~ TH-positive neurons. Moreover, olfballs can be derived in the absence of exogenous EGF, whereas proliferation of neurospheres is 15 dependent upon this growth factor. In spite of these differences, our fin-lings suggest that there may indeed be similar neural stem cell(s) throughout the peripheral and central nervous systems, and that the ultimate developmental outcome for the progeny of these cells is predomin~ntly a function of the local neural environment.
Olfactory epithelial -derived stem cells replace the dopaminergic input in the striatum in 20 the 6-OHDA animal model of Parkinson's disease. The gen~ldLion of differ~nti~te~ TH
immunopositive neurons from olfactory epithelial -derived stem cells permits these neurons to functionally co~l,pensate for, and restore the deficits caused by, the loss of dop~ ic input into the stri~tllm in P~kinsoll's ~ice~ce We show that 1) the TH positive cells which di~.~,lliate in culture are dopaminergic neurons ii) the number and neuritic complexity of TH
25 imrnunopositive cells increase in transplantation ~ lh.lents iii) the transplanted olfactory epithelial-derived stem cells ameliorates the functional deficit in animal models of Parkinson's disease iv) transplanted adult olfactory epithelial -derived stem cells also di~~ iate into TH

positive neurons. Finally, we transplant human olfactorv epithelial derived precursor cells into chemically lesioned rats to obtain TH positive cells and functional recovery.
Example 17 - Usin~ Neural Stem Cells or Neuronal Pro~enitor Cells in Autolo~ous Transplants in Treatment of Multiple Sclerosis The olfactory derived precursor cells or cells derived from these precursor cells are implanted into lesion sites of patients having Multiple Sclerosis.
Example 18 - Isolation of Precursor Cells From Other Peripheral Tissues We derived precursor cells from the tongue which is another peripheral tissue that contains sensory ~ecell~o,~. The tongue was fliccecte~ to remove the epithelial layer that contains the sensory receptors and their underlying basal cells. This layer of tissue is triturated to produce single cells and the single cells are plated in flasks co~ g DMEM/R12 media supplemPnted with B-27 (Gibco) and EGF, TGFa, and/or bFGF, as described for the olfactory epithelium.
After 2-3 days in a-37 degree Celsius, 5% carbon disoxide tissue culture incubator, most of the cells in the culture are dead or dying. However, a small number of large phase bright cells are present, most of which attach to the flask bottom. Over the next 2 to 6 days these cells divide and produce spherical aggregates that become larger over time and left from the flask surface.
The cells in these clusters produce a marker for neural progt;"i~o,~ and stem cells, nestin.
These nestin positive cells can be p~cc~ge~l using the same techniques as used for olfballs.
These nestin positive cells can be differenti~tec~ into llculOllS, astrocytes and oligodendrocytes using the same techniques as used for olfballs.
We isolate precursor cells from peripheral tissues co..l;~ g sensory receptors, other than the olfactory epitheliurn, using the above techniques. We passage these cells using the same techniques as used for olfballs. We di~l~il,Liate these cells into neurons, astrocytes and oligodendrocytes using the same techniques as used for olfballs. We induce precursor cells to 25 differentiate by culturing them with growth factors. This is done similarly to the dirrere"liation of precursor cells in example 2.

We grow precursor cells in vitro and differentiate these cells into specific neural cell types. We transfect the olfactory epithelium derived stem cells with the candidate transcription factors including IsI 1, en- 1, en-2 and nurr, which have been implicated in regulating motoneuron and striatal phenotypes. With an understanding on how intrinsic and extrinsic factors regulate 5 neural cell fate decisions, it is possible to induce the differentiation of the specific neural cell types required for neuronal transplant therapy.
We modify precursor cells as described above with respect to olfactory epithelial-derived cells using the adenovirus gene transfer system.
We isolate human-derived neural stem cells from peripheral tissues that contain sensory 10 receptors. We use the same procedures as we used to isolate the neural stem cells from the neonate and adult mice. As is known to those skilled in the art, work on primate and human neurospheres with regards to culture conditions provide information on oy~ proliferation of cell spheres and specific cell types. We char~ct~ri7~ the growth and dirre~cllliation of p~ or cells, and their potential for use in transplantation therapy for neurodegenerative 15 ~iceaces~
These precursor cells are useful thc.dyeulically in the 1~ l of neurodegene.~ /e~iceaces such as Parkinson's disease and multiple sclerosis. We implant dopamingeric neurons dirr~ ~ti~tecl from p~c~ or cells into the substantia nigra or the stri~tl~rn of patients having Parkinson's Disease. The olfactory derived precursor cells or cells derived from these yle~;ul~or 20 cells are implanted into lesion sites of patients having Multiple Sclerosis. We induce precursor cells to dirre.e..liate into astrocytes by cllltllring them with growth factors. This is done simil~rly to the dirre.~..liation of precursor cells in example 2.

The present invention has been described in terms of particular embol1im~ntc found or 25 proposed by the present inventors to comprise yre~ed modes for the practice of the invention.
It will be appreciated by those of skill in the art that, in light of the present disclosure, numerous modifications and changes can be made in the particular embo-lim~ntc exemplified without departing from the intent~e~l scope of the invention. All such modifications are inten~e~l to be included within the scope of the appended claims.
All publications, patents and patent applications are herein incorporated by reference in their entirety to the same extent as if each individual publication, patent or patent application was S specifically and individually indicated to be incorporated by reference in its entirety.

References l. Burns S, Chiueh CC, Markey SP, Ebert MH, Jacobowitz DM, Kopin IJ (1983) A primate model of parkinsonism: selective destruction of dopaminergic neurons in pars compacta of the substantia nigra by N-methyl-4-phenyl- 1,2,3,6-tetra-hydropyridine. Proc Natl Acad Sci (USA) 80:4546-4550 2. Fahn S (1992) Fetal-tissue transplants in Parkinson's disease. New Fn~l~n~l Journal of Medicine. 3271589-1590

3. Dunnett SB, Annett LE (1991) Nigral transplants in primate models of parkinsonism. In:
Lindvall O, Bjorklund A, Widner H, eds. Intracerebral transplantation in movement disorders. Restorative Neurology 4:27-51

4. T ~nE~tQn JW, Ballard P, Tetrud JW, Irwin I (1983) Chronic parkinsonism in hl-m~n~ due to a product of meperidine analog synthesis. Science 219:979-980

5. Remington's Ph~rm~ceutical Sciences, Mack Publishing Company, Easton, Pa., USA

6. Widner H, Tetrud J, Rehncrona S, Snow B, Brundin P, Gustavii B, Bjorklund A, Lindvall P, r ~n~ton JW (1993) Bilateral fetal m~s.oncephalic grafting in two patients with p~rkimonism in~ ced by l-methyl-4-phenyl-1,2-3,6-tetrahydropyridine (MPTP). New Fn~l~n-l Journal of Medicine 327:1556-1563

7. Winkler C, H~mm~ng JP, Bjorklund A (1995) EGF-responsive neural progenitor cells, survive, migrate and dirr~e..liate after transplantation into the adult rat striatum. Society for Neuroscience Abstracts 21 :2029

8. Gage FH, Coates PW, Palmer TD, Kuhn G, Fisher LJ, Suhonen JO, Peterson DA, Suhr ST, Ray J (1995) Survival and differentiation of adult neuronal progenitor cellstransplanted to the adult brain. Proc Natl Acad Sci (USA) 92:11870-11883

9. Reynolds B and Weiss S, (1992) Science 255:107

10. Weiss S, Reynolds B, Vescovi A, Morshead C, Craig C, van der Kooy D, (1996) Is there a neural stem cell in the m~mm~liAn forebrain? T.I.N.S. 19:9:1 I 1. Peel AL, Feldman DH, Reier PJ (1995) Co-localization of glial and neuronal markers in EGF-generated cultures of pluripotent CNS stem cells. Society for Neuroscience Abstract 21:285 12. Ruth S. Slack, Daniel J. Belliveau, Madelaine Rosenberg, Jasvinder Atwal, Hanns Lochmuller, Raquel Aloyz, Ali H~ghighi, B. Lach, Prem Seth, Ellis Cooper, and F. D.
Miller (1996) Adenovirus-mediated Gene Transfer of the Tumor Supple3sor, pS3, Induces Apoptosis in Postmitotic Neurons. The Journal of Cell Biology, Volume 135, No. 4 1085-1096 13. Le Gal La Salle, G., J.J. Roberts, S. Berrard, V. Ridoux, L.D. Stratford-Perricaudet, M.
Perricaudet, and J. Mallet, 1993. An adenovirus vector for gene transfer into neurons and glia in the brain, Science (Wash. DC), 259: 988-990.
14. 39. Slack, R.S., and F.D. Miller, 1996. Viral vectors for use in mo(3~ ting gene cA~l~,s~ion in neurons, Curr. Opin. Neurobiot, 6:576-583.
15. I. Lefkowitz, H. Waldman, T imiting Dilution Analysis of Cells in the Immune System.
Cambridge University Press, Cambridge, U.K. (1979) 16. C. G. Bellows, J.E. Aubin, Dev. Biol. 133, 8 (1989).
17. A. Carlsson, M. Lindqvist, T. Magnusson, Nature 180, 1200 (1957); H. Ehringer, O.
Hornykiewicz, Klin. Wschr. 38, 1236 (1960).
18. Female Sprague-Dawley rats or CD 1 albino mice (Charles River, Montreal, Quebec, Canada) weig~ung 180-200 g or 25-30 g les~eclively, were ~n~sth~ti7~cl with a llliA~UlC
of ket~mine (Ketaset, Ayerst, 90 mg/kg) and xylazine (Rompun, Haver, 10 mg/kg) (hllldp~ olleal) prior to stereotactic injections of 24 ug of 6-hydroxydop~llinehydrobromide (dissolved in S ul of 0.9% saline co~ i ng 0.2 mglml ascorbate) into the right medial folcbldil1 bundle (Tooth bdr:-2.4 mm; A:-4.4 mm; L: 1.0 mm; V:7.5 mm (19)). Two weeks after the lesion, ~nim~l~ were tested for rotational behavior as previously described (20). Animals were immunosuppressed with cyclosporine (Sandimmune, Sandoz, 40mg/kg, inlldp~ oneal) once a day and the immuno~upyies~ion was continued until the day of sacrifice. For olfball transplantation, ~n~Psthetized ;~nim~l~ were mounted in a Kopf stereotactic apl)~dlus, and 2 x 2.5 ~11 aliquots of olfballs were injected unilaterally into the lesioned caudate putamen or bilaterally in some ~nim~l~. The injections were made using a 5 ul Hamilton syringe at the followingcoordinates: Tooth bar, -2.4 mm; A: 0.2; L: 3.0; V: 5.5-6.0 (19). Injections were performed over a period of 3 minutes, a further 5 minutec was allowed for diffusion, and the needle was then retracted. These 5 ~,11 injections contained olfballs derived from one neonatal pup cultured for 7-14 days. For the BrdU ~x~.hllents, BrdU (lOIlM) was added to culture media for 18 hours, after which the olfballs were washed three times with fresh media to remove the BrdU, and then the olfballs tranplanted one day later. Sixteen days following transplantation, ~nim~lc were ~n~sthtoti71od with an overdose of i.p.
pentobarbital, and perfused transcardially sequentially with saline and 4%
p~dfo",l~l~ellyde in phosphate buffer (PB, O.lM, pH 7.4). The brains were post-fixed for 18 hours at 4~C, and then cryoprote~led for 48 hours in 30% sucrose dissolved in PB.
Brains were sectioned on a free_ing microtome in the coronal plane at 40 urn. Free-floating sections were collected in phosphate buffered saline (O.lM, PBS) and processed for TH, ~ ctocidase (Boehringer ~nnh~im), GFAP, or nestin immlmocytoçllPrni~try.Sections were initially incub~te~l in a PBS solution co~ inil~g 0.5% sodium borohydride for 20 mimltes, rapidly washed six times, and then incubated in PBS col,l~;"il-g 5% BSA.
Sections were then incubated in a PBS solution co~ g 0.1% Triton X-100, 2% BSA
and either anti-TH (1:1500), or monoclonal 13-g~l~ctositl~ce (1:500) or anti-nestin antibody (1:1000). After overnight inrllb~tion in the primary antibodies at 4~C, sections were rinsed in PBS (3X, S mins per wash), and in~ub~tecl for 1 hour at room telllp~,~alule in PBS co.,~ il-g biotinylated goat anti-rabbit IgG (1:200, Vector Laboldlolies), 0.1%
Triton X-100, and 2% BSA. After 3 brief washes in PBS, sections were inc~1b~te~1 for 1 hour at room ten~.dlu,e in PBS col.~ il-g an avidin-biotin complex (ABC, Vector Laboratories). Following 3 washes in PBS, the immunohistochemical reaction product was revealed by incubation in Tris buffer (0.05M, pH 7.6) contAining ~i~min~benzidine tetrahydrochloride (DAB, Sigma) (0.025 g/lOOml), 1% lM imidazole, and 0.3%
hydrogen peroxide. Sections were exposed to DAB for 15 minutes, rinsed 6x in PBS, mounted onto chrom-alum coated slides, air dried, dehydrated in graded alcohols, and coverslipped with Permount.
19. G. Paxinos and C. Watson, The Rat Brain in Stereotaxic Coorllin~tes. Ac~lçmic Press, SanDiego 1986.
20. U. Ungerstedt and G. W. Arbuthnott, Brain Res. 24, 485 (1970).
21. A. Gloster, W. Wu, A. Speellman, S. Weiss, C. C~ ing, C. Pomiak, B. Reynolds, E.
Chang, J.G. Toma, F.D. Miller, J. Neurosci. 14, 7319 (1994); S. Bamji, F.D. Miller, J.
Comp. Neurol. 374, 52 (1996).
22. BrdU immllnl histochPmi~try was pelrolllled as described (23), with some modifications.
Following preincubation (23), free-floating sections were in~lb~te~l for 16 hours in PBS
co..~ g anti-BrdU (1:40, Becton-Dickinson, San Jose, CA) and 2% BSA at 4 C.
Following 3 briefwashes in PBS, sections were incnb~tecl for 1 h in PBS co~ it-gsecondary antibody (biotinylated anti-mouse IgG, 1 :200, 1 :200, Vector, Burlingh~me, CA) and 2% BSA. Sections were rinsed in PBS (3 x 5 min), incubated in avidin-biotin complex (ABC, 1%, in PBS, Vector, Bllrlingh~m~ CA) for 1 hour, again rinsed in PBS
(3 x 5 min), and the immun~histochemical reaction product revealed by inrllb~ting the sections for 7-10 mimltes in a solution colll;1;llil~g 0.37 gm nickel ammonium sulfate, 25 mg DAB, and 2 ul of 30% hydrogen peroxide dissolved in 100 ml of Tris buffer (0.05 M, pH 7.6), yielding a blue-black reaction product. Sections were then thoroughly rinsed in PBS, and pl~ed for immllnohistoch~mi~try for the second antigen (18).
23. E. Soriano, J. A. Del Rio, J. Histochem. Cytochem. 39, 255 (1991).
24. S. Fahn, New Fng]~n(l J. Med. 327, 1589 (1992).

Claims (11)

1. A composition consisting of an isolated population of neural stem cells of a postnatal mammal and a carrier, wherein said neural stem cells form non-adherent clusters in culture, are self renewing, proliferate in an EGF-independent manner, express nestin, and differentiate, in the presence of serum, into neurons expressing tyrosine hydroxylase, said stem cells produced by a method comprising the steps of:

(a) providing a culture of peripheral tissue containing sensory receptors from said mammal;

(b) isolating neural stem cells from said peripheral tissue, based on the tendency of said neural stem cells to aggregate and form non-adherent clusters in culture, wherein said neural stem cells form non-adherent clusters in culture, are self renewing, proliferate in an EGF-independent manner, express nestin, and differentiate, in the presence of serum, into neurons expressing tyrosine hydroxylase.

2. A composition consisting of an isolated population of neural stem cells of a postnatal mammal and a carrier, wherein said neural stem cells form non-adherent clusters in culture, are self renewing, proliferate in an EGF-independent manner, express nestin, and differentiate, in the presence of serum, into neurons expressing tyrosine hydroxylase.

3. The composition of claim 1, wherein said peripheral tissue comprises olfactory epithelium.

4. The composition of claim 1, wherein said peripheral tissue comprises tongue.

5. The composition of claim 1, wherein said neural stem cells are transfected with a heterologous gene.

6. The composition of claim 5, wherein said gene encodes a trophic factor.

7. The composition of claim 1, wherein said neural stem cells are human stem cells.

8. The composition of claim 1, formulated in a pharmaceutically acceptable carrier, auxiliary or excipient.

9. The composition of claim 2, formulated in a pharmaceutically acceptable carrier, auxiliary or excipient.

10. The composition of claim 2, wherein said neural stem cells are human stem cells.

11. The composition of claim 2, formulated in a pharmaceutically acceptable carrier, auxiliary or excipient.