JP2005239721A - Composition containing genetically modified donor cell and used for treating cell having defect, disease, or injury in central nervous system, and method for preparing the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain a composition for treating a defect, a disease, or an injury of the central nervous system, by transplanting a genetically modified cell, and to provide a method for preparing the same. <P>SOLUTION: The modified donor cell which produces a functional molecule having an effect of recovering or improving a function of a cell in the central nervous system is formed by gene manipulation. The formed donor cell is injected into and transplanted to the system for the purpose of treating the defect, the disease, or the injury of the central nervous system. The transplanted cell expresses the functional molecules in an enough amount for restoring the cell in the system. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to the use of recombinant techniques for genetic modification of donor cells to treat CNS defects, diseases or injuries by transplantation into the central nervous system (CNS) of a specimen. More specifically, this invention relates to inserting into a donor cell a gene encoding a molecule that has an improving effect on cells in the CNS, which involves inserting a neuronal cell into the donor cell. , When the donor cells are transplanted into the CNS, the molecules are expressed so that the effect of those molecules on the diseased or damaged cells is exerted.

  This invention includes N00014-86-K-0347 grant contract approved by the Office of Naval Research, as well as HD-20034 and NIA-06088 approved by NIH. It was made with the support of the US government under HD-00699 and the NIA R01 AG08514 grant subcontract approved by the Department of Health and Human Services. The US government has certain rights in this invention.

  Attempts to repair mammalian brain or replace CNS function due to CNS deficiencies or as a result of disease or injury have been hampered by a poor understanding of the complex structure-function relationship in the CNS. In recent years, knowledge of some basic principles of cell function in the brain has greatly advanced, but the interaction and behavioral and neurological functions between cell clusters or systems and cell circuits in different regions of the brain Understanding of the relationship of these outward manifestations to these has been far behind. The difficulty in approaching these problems is caused in part by the many different cell types and the number and complexity of their connections in the mammalian CNS. In addition, the blood-brain barrier makes access to the brain more difficult for diagnosis, treatment, and design of new therapies.

  Despite the lack of advanced knowledge of pathophysiology regarding most normal or abnormal brain functions, several attempts to pharmacological treatment for CNS dysfunction have already become useful and effective. These include the use of drugs that affect the psychiatric disorders, such as schizophrenia, and in rare cases such as Parkinson's disease where the biochemical cellular basis of CNS disorders is relatively well understood. Specific replacement therapy is included. At the heart of most therapeutic approaches is the purpose of replacing, or restoring, specific chemical functions of the brain that are lost as a result of disease or injury.

  Intracranial nerve transplantation has recently emerged as an additional potential approach to CNS treatment. Replacement or addition to the CNS of cells that can produce and secrete therapeutically useful metabolites has the advantage of avoiding repeated dosing and avoiding drug delivery complications by the blood-brain barrier. (Rosenstein, Science 235: 772-774 (1987)). The concept of brain transplantation and basic techniques have been known for decades, but most of the factors for maximizing the survival of transplanted cells have recently been studied and partially understood (Bjorklund et al., In Neural Grafting in the Mammalian CNS, p. 709, Elsevier, Amsterdam (1985); Sladek et al., In Neural Transplants: Development and Fuction, Plenum Press, NY (1984 )). Several factors have been identified that are decisive for reliable and effective transplant survival, including:

That is,
(1) Donor age: The efficiency of transplantation decreases with increasing donor cell age.
(2) Host age: Younger recipients accept transplants more easily than older ones.
(3) Availability of neurotrophic factors in host and donor tissues: Wound-induced neurotrophic factors enhance transplant survival.
(4) Immunological response: The brain is not at all an immunologically privileged site.
(5) Importance of target-donor matching: Neurons survive better when transplanted to sites where they are normally distributed.
(6) Angiogenesis: The transplant should be rapidly vascularized or it is important that the surroundings be well-fed.

  As these critical factors are recognized and complete, brain transplantation is effective for neurophysiologists to study CNS function and possibly for clinicians to design treatments for CNS diseases. And has become a reliable tool. This approach has reached the level of experimental clinical application in Parkinson's disease.

  Parkinson's disease is an age-related injury and is characterized by a loss of dopamine neurons in the substantia nigra of the midbrain with the basal ganglia as the main target organ. Symptoms include tremors, stiffness and ataxia. Although the disease is progressive, it can be treated by replacement of dopamine via administration of a pharmacological dose of L-dopa, a precursor to dopamine (Marsden, Trends Neurosci. 9: 512 (1986); Vinken et al., In Handbook of Clinical Neurology p. 185, Elsevier, Amsterdam (1986)), chronic use of pharmacotherapy often renders patients unresponsive to the continuous effects of L-dopa. Many mechanisms have been suggested for the development of this unresponsive state, but the simplest is that the patient reaches the threshold for cell deletion, in which case the remaining cells are sufficiently removed from the precursor. Dopamine cannot be synthesized. Parkinson's disease is the first brain disease for which therapeutic intracerebral transplantation has been used in humans. Several attempts have been made to supply the neurotransmitter dopamine to the diseased basal ganglion cells of Parkinson's patients by allogeneic transplantation of adrenal medullary cells into the patient's brain (Backlund et al., J. Neurosurg. 62: 169-173 (1985); Madrazo et al. New Eng. J. Med. 316: 831-836 (1987)).

  Other donor cells, such as fetal brain cells from the substantia nigra, which is a region of the brain rich in dopamine-containing cell bodies and the most affected brain region in Parkinson's disease, but selected by transplantation of this cell Have been shown to be effective in reversing behavioral deficits induced by experimental dopamine neurotoxin (Bjorklund et al., Ann. NYAcad. Sci. 457: 53-81 (1986); Dunnett et al. , Trends Neurosci. 6: 266-270 (1983)). These experiments suggest that synaptic connectivity is not a requirement for functional transplantation, and having cells that constitutively produce and secrete dopamine in the vicinity of defective cells is sufficient.

  In this background, Parkinson's disease is a candidate disease for transplantation of genetically engineered cells because (1) chemical defects are well known (dopamine), (2 ) Human and rat genes for rate-limiting enzymes in dopamine production have been cloned (tyrosin hydroxylase), (3) anatomical location of affected areas has been confirmed (basal ganglia), (4) This is because it seems that synaptic connectivity is not required for complete functional repair.

  Other CNS diseases, including Huntington's disease (Gusella et al., Nature 306: 234-238 (1983)), Alzheimer's disease (Delabar et al., Science, NY 235, 1390-1392 (1987); Goldgaber et al ., Science, NY 235: 877-880 (1987); St. George-Hyslop et al., Science, NY 235: 885-890 (1987); Tanzi et al., Science, NY 235: 880-884 (1987) ), Bipolar disease (Baron et al., Nature 326: 289-292 (1987)), schizophrenia (Sherrington et al., Nature 336: 164-167 (1988)), and many other human Although disease is included, recent demonstration of genetic components in the rapidly growing list of these diseases suggests that gene therapy is a useful approach to address these CNS diseases.

  In parallel with advances in psychobiology over the past few decades, advances in understanding molecular biology and the development of advanced molecular genetic techniques have provided new insights into human disease in general. As a result, medical scientists and geneticists have developed a profound understanding of many human diseases at the biochemical and genetic levels. The normal and abnormal biochemical characteristics of many human genetic diseases are understood, related genes are isolated and characterized, and early model systems introduce functional wild-type genes into mutant cells. Developed to correct disease phenotypes (Anderson, Science 226: 401-409 (1984)). The extension of this approach to all animals, ie the modification of the disease phenotype in vivo through the use of functional genes as pharmacological agents, has come to be referred to as “gene therapy” (Friedmann et al., Science 175: 949-955 (1972); Friedmann, Gene Therapy Fact and Fiction, Cold Spring Harbor Laboratory, New York (1983)). Gene therapy is based on the assumption that disease phenotype correction can be achieved by modifying the phenotype of the underlying mutant gene or by introducing new genetic information into defective or damaged cells in vivo. ing.

  At present, techniques for the ideal version of gene therapy, ie in vivo modification or replacement via site-specific gene sequences, have just begun to be considered and are not well developed. Thus, many current models of gene therapy are actually gene augmentations over replacement models, and functional wild-type gene information can be transferred in vitro and in vivo, depending on the development of an efficient gene transfer system. It is to introduce into a defective cell. In order to be clinically useful, the availability of efficient delivery vectors for functional DNA sequences (transgenes) is easy to access and stable to appropriate disease-related target cells or organs. And it must be combined with the development of technology for safe vector introduction.

  A model system for the genetic phenotypic correction of simple enzyme deficiencies is currently under development, as well as the identification of appropriate potential recipient cells and organs associated with specific metabolic genetic diseases. Evidence has recently been shown that heterologous genes introduced into the fertilized mouse ovum can correct the disease phenotype (Constantini et al., Science 233: 1192-1394 (1986); Mason et al., Science 234 : 1372-1378 (1986); and Readhead et al., Cell 48: 703-712 (1987)).

  Much attention has been focused recently on the use of gene delivery vectors derived from murine retroviruses for gene transfer (Anderson, Science 226: 401-409 (1984); Gilboa et al., Biotechniques 4: 504-512 (1986)). In vitro transfer using such retroviral vectors is extremely efficient for a wide range of recipient cells, and this vector has a reasonably large capacity for the gene being added, and infection by these vectors is susceptible to recipient cells. Causes only minor metabolic or genetic damage (Shimotohno et al., Cell 26: 67-77 (1981); Wei et al., J. Virol. 39: 934-944 (1981); Tabin et al., Molec. Cell. Biol. 2: 426-436 (1982)). Several useful systems have demonstrated that expression of genes introduced into cells by retroviral vectors can correct metabolic abnormalities in vitro in several human genetic diseases associated with single gene enzyme deficiency (Willis et al., J. Biol. Chem. 259: 7842-7849 (1984); Kantoff et al., Proc. Natl. Acad. Sci. USA 83: 6563-6567 (1986)). Of particular interest is the bone marrow as a potential target organ for this approach to gene therapy. This includes Mediterranean anemia, sickle cell anemia, Gaucher disease, chronic granulomatous disease (CGD), and immune deficiency as a result of purine pathway enzymes, adenosine deaminase (ADA) and purine nucleoside phosphorylase (PNP) deficiencies Because of the prevalence and importance of bone marrow-derived cell line injury in various major human diseases (Kantoff, supra; McIvor et al., Molec. Cell. Biol. 7: 838-846 (1987); Soriano et al. al., Science 234: 1409-1413 (1986); Willis et al., supra)). Other metabolically important target organs, such as the liver, have also recently become theoretically accepted for genetic treatment through the demonstration of viral vector infection of cells from these organs (Wolff et al. al., Proc. Natl. Acad. Sci. USA 84: 3344-3348 (1987)). In addition, the discovery of many cell-specific regulatory signals, such as cis-acting enhancers, tissue-specific promoters, and other sequences, can be found in other, even after general non-specific infection and gene transfer in vivo. It may be possible to provide tissue specific gene expression in many organs (Khoury et al., Cell 33: 313-314 (1983); Serflin et al., Trends Genet. 1: 224-230 (1985)).

  A recently developed model of gene therapy uses target cells taken from a patient, which is cultured, genetically modified in vitro and then reimplanted into the patient (Wolff et al., Rheumatic Dis. Clin. N Amer. 14 (2) 459-477 (1988); Eglitis et al., Biotechniques 6: 608-614 (1988)). Target cells include bone marrow stem cells (Joyner et al., Nature 305: 556-558 (1983); Miller et al., Science 225: 630-632 (1984); Williams et al., Nature 310: 476-480 ( 1984)), fibroblasts (Selden et al., Science 236: 714-718 (1982); Garver et al., Proc. Natl. Acad. Sci. USA 84: 1050-1054 (1987) and St. Louis et al. al., Proc. Natl. Acad. Sci. USA 85: 3150-3154 (1988)), keratinocytes (Morgan et al., Science 237: 1476-1479 (1987)), and hepatocytes (Wolff et al., Proc. Natl. Acad. Sci. USA 84: 3344-3348 (1987)). This indirect approach of in vivo gene transfer is required by the inability to efficiently transfer genes directly to cells in vivo. Although some recent progress towards genetic modification of neurons by culture is seen (Geller et al., Science 241: 1667-1669 (1988)), this indirect approach of in vivo gene transfer has been introduced to the CNS. Has not been applied yet.

  There are several ways to introduce new functions into target cells in the CNS in a phenotypically useful manner, ie to treat defects, diseases or dysfunctions (FIG. 1). The most direct approach, which altogether avoids the need for cell transplantation, but cells that are abnormal in function as a result of developmental or genetic defects, i.e., Taes-Sachs disease, and possibly Lesch-Nihan disease, and In the case of Parkinson's disease, this is the direct introduction of the transgene into nerve cells (1 in FIG. 1). Alternatively, new functions are expressed in defective target cells by introducing genetically modified donor cells that can make gap junctions or other contacts with the target cells (2 in FIG. 1). Some of these contacts are known to allow efficient diffusion of metabolically important small molecules from one cell to another, leading to phenotypic changes in recipient cells ( Lowenstein, Biochem. Biophys. Acta. 560: 1-66 (1979)). This process is called “metabolic cooperation” and is known to occur between fibroblasts and glial cells (Gruber et al., Proc. Natl. Acad. Sci. USA 82: 6662-6666 ( 1985)), in conclusion, has not been proven in neurons yet. Still other donor cells can express and secrete diffusible gene products that are taken up and used by nearby defective target cells (3 in FIG. 1). Donor cells are genetically modified in vitro or directly infected in vivo (4 in FIG. 1). This type of “cooperativity” has been demonstrated for CNS cells as in the case of NGF-mediated protection of cholinergic neuronal death following CNS injury. (Hefti, J. Neurosci. 6: 2155 (1986); Williams et al., Proc. Natl. Acad. Sci. USA 83: 9231-9235 (1986)) Introduced donor cells that are also infected with a tent helper virus locally produce high-titer progeny virus that in turn infects nearby target cells to produce a functional new transgene. Will be possible (5 in FIG. 1).

  There are several types of neurons in the mammalian brain. Cholinergic neurons are found in the mammalian brain and project from the medial septum and vertical limbs of the broker's diagonal band into the hippocampus of the basal forebrain. The medial membrane of the diagonal band and the short nerve-like part of the brain that connects the vertical limb and the hippocampus are called "fimbria fornix". The foramen fold includes the axons of neurons located in the interstitial membrane and the diagonal zone. A commonly accepted model of neuronal survival in vivo is the survival of cholinergic neurons in the diaphragm following a transection or lesion (also referred to as an “axotomy”) of the arch. Axotomi cuts cholinergic neurons in the diaphragm and diagonal zone, resulting in death of up to half of the cholinergic neurons (Gage et al., Neuroscience 19: 241-256 (1986)). This degenerative response is attributed to a lack of nutritional support from nerve growth factor (NGF), which is normally carried retrogradely from the hippocampus to the diaphragm cholinergic cell body to the intact brain ( Korsching et al., Proc. Natl. Acad. Sci. USA 80: 3513 (1983); Whittemore et al., Proc. Natl. Acad. Sci. USA 83: 817 (1986); Shelton et al., Proc. Natl Acad. Sci. USA 83: 2714 (1986); Larkfors et al., J. Neurosci. Res. 18: 525 (1987); and Seilor et al., Brain Res. 300: 33 (1984)).

  Studies have shown that chronic indoor administration of NGF prior to Axotomi will prevent cholinergic neuron death in the diaphragm (Hefti, J. Neurosci. 8: 2155-2162 (1986); Williams et al., Proc. Natl. Acad. Sci. USA 83: 9231 (1986); Kromer, Science 235: 214 (1987); Gage et al., J. Comp. Neurol. 269: 147 (1988)). Thus, Axotomi provides one in vivo model for determining the ability of various therapies at various points in time to prevent retrograde neuronal death.

  One characteristic of the adult mammalian CNS is that new axons generated following perturbation can grow a relatively short distance in the brain (Cajal, in Degeneration and Regeneration of the Nervous System, Oxford University Press, London, England, (1928)). The reason why regeneration in response to the injury of adult neurons is impossible is as follows. 1) Activation of astrocytes, a supportive, nutrient cell found throughout the CNS that results in the formation of scar tissue following injury, as a physical barrier to axonal regeneration Acts and the fibers cannot cross the scar tissue, and thus scars from astrocytes cannot provide a substrate that aids in axonal elongation (Cajal, supra; Brown, J. Comp. Neurol. 87: 131 (1947); Clemente, in Regeneration in the Central Nervous System, ed. Windle, pp. 147-161, Thomas, Springfield, (1955); Windle, Physiol. Rev. 36: 426 (1956); and Reier et al., in Spinal Cord Reconstruction, eds. Kao et al., pp. 163-195, Raven, New York, (1983)). On the other hand, astrocytes in the immature CNS play an important role in leading axonal outgrowth (Silver et al., J. Comp. Neurol. 210: 10-29 (1982)). 2) Myelin-related substances released following brain injury have been shown to inhibit axon growth in vitro (Schwab and Croni, J. Neurosci. 8: 2381-2393 (1988)) . 3) The lack of axonal regeneration in the adult CNS may be due in part to inappropriate levels of trophic molecules that induce neuron regeneration and promote axonal growth (Reier et al., In Neural Regeneration). and Transplantation, pp. 183-209, Alan R. Liss, New York, (1989) and Schwartz et al., FASEB J. 3: 2371-2378 (1989)).

Despite many factors that prevent axonal regrowth in the adult brain, certain neurons in the rat CNS, particularly retinal ganglion neurons and septal cholinergic neurons, are responsible for the new axon to various types of substrates. Shows remarkable potential for elongation. These are so-called “neural bridges”, segments of autologous peripheral nerves (David and Aguayo, Science 214: 931 (1981); Benfry and Aguayo, Nature 296: 150 (1982); Wendt et al., Exp. Neurol. 79: 452 (1983); and Hagg et al., Exp. Neurol. 109: 153 (1990)), cultured Schwann cells in a collagen matrix (Kromer and Cornbrooks, Proc. Natl. Acad Sci. USA 82: 6330 (1985)), embryonic rat hippocampus (Kromer et al., Brain Res. 210: 153 (1981)), and human amniotic membrane (Gage et al., Exp. Brain Res. 72: 371 ( 1988)). The degree of connectivity between the brain diaphragm and the hippocampus has also been demonstrated using a peripheral homogenate graft of neurons (Wendt, Brain Res. Bull. 15: 13-18 (1985)).
Rosenstein, Science 235: 772-774 (1987) Bjorklund et al., In Neural Grafting in the Mammalian CNS, p. 709, Elsevier, Amsterdam (1985) Sladek et al., In Neural Transplants: Development and Fuction, Plenum Press, NY (1984) Marsden, Trends Neurosci. 9: 512 (1986) Vinken et al., In Handbook of Clinical Neurologyp. 185, Elsevier, Amsterdam (1986) Backlund et al., J. Neurosurg. 62: 169-173 (1985) Madrazo et al. New Eng. J. Med. 316: 831-836 (1987) Bjorklund et al., Ann. NYAcad. Sci. 457: 53-81 (1986) Dunnett et al., Trends Neurosci. 6: 266-270 (1983) Gusella et al., Nature 306: 234-238 (1983) Delabar et al., Science, NY 235, 1390-1392 (1987) Goldgaber et al., Science, NY 235: 877-880 (1987) St. George-Hyslop et al., Science, NY 235: 885-890 (1987) Tanzi et al., Science, NY 235: 880-884 (1987) Baron et al., Nature 326: 289-292 (1987) Sherrington et al., Nature 336: 164-167 (1988) Anderson, Science 226: 401-409 (1984) Friedmann et al., Science 175: 949-955 (1972) Friedmann, Gene Therapy Fact and Fiction, Cold Spring Harbor Laboratory, New York (1983) Constantiniet al., Science 233: 1192-1394 (1986) Mason et al., Science 234: 1372-1378 (1986) Readhead et al., Cell 48: 703-712 (1987) Anderson, Science 226: 401-409 (1984) Gilboa et al., Biotechniques 4: 504-512 (1986) Shimotohno et al., Cell 26: 67-77 (1981) Wei et al., J. Virol. 39: 934-944 (1981) Tabin et al., Molec. Cell. Biol. 2: 426-436 (1982) Willis et al., J. Biol. Chem. 259: 7842-7849 (1984) Kantoff et al., Proc. Natl. Acad. Sci. USA 83: 6563-6567 (1986) McIvor et al., Molec. Cell. Biol. 7: 838-846 (1987) Soriano et al., Science 234: 1409-1413 (1986) Wolff et al., Proc. Natl. Acad. Sci. USA 84: 3344-3348 (1987) Khoury et al., Cell 33: 313-314 (1983) Serflin et al., TrendsGenet. 1: 224-230 (1985) Wolff et al., Rheumatic Dis. Clin. N. Amer. 14 (2) 459-477 (1988) Eglitis et al., Biotechniques 6: 608-614 (1988) Joyner et al., Nature 305: 556-558 (1983) Miller et al., Science 225: 630-632 (1984) Williams et al., Nature 310: 476-480 (1984) Selden et al., Science 236: 714-718 (1982) Garver et al., Proc. Natl. Acad. Sci. USA 84: 1050-1054 (1987) St. Louis et al., Proc. Natl. Acad. Sci. USA 85: 3150-3154 (1988) Morgan et al., Science 237: 1476-1479 (1987) Wolff et al., Proc. Natl. Acad. Sci. USA 84: 3344-3348 (1987) Geller et al., Science 241: 1667-1669 (1988) Lowenstein, Biochem. Biophys. Acta. 560: 1-66 (1979) Gruber et al., Proc. Natl. Acad. Sci. USA 82: 6662-6666 (1985) Hefti, J. Neurosci. 6: 2155 (1986) Williams et al., Proc. Natl. Acad. Sci. USA 83: 9231-9235 (1986) Gage et al., Neuroscience 19: 241-256 (1986) Korsching et al., Proc. Natl. Acad. Sci. USA 80: 3513 (1983) Whittemore et al., Proc. Natl. Acad. Sci. USA 83: 817 (1986) Shelton et al., Proc. Natl. Acad. Sci. USA 83: 2714 (1986) Larkfors et al., J. Neurosci. Res. 18: 525 (1987) Seilor et al., Brain Res. 300: 33 (1984) Hefti, J. Neurosci. 8: 2155-2162 (1986) Williams et al., Proc. Natl. Acad. Sci. USA 83: 9231 (1986) Kromer, Science 235: 214 (1987) Gage et al., J. Comp. Neurol. 269: 147 (1988) Cajal, in Degeneration and Regeneration of the Nervous System, Oxford University Press, London, England, (1928) Brown, J. Comp. Neurol. 87: 131 (1947) Clemente, in Regeneration in the Central Nervous System, ed. Windle, pp. 147-161, Thomas, Springfield, (1955) Windle, Physiol. Rev. 36: 426 (1956) Reier et al., In Spinal Cord Reconstruction, eds.Kao et al., Pp. 163-195, Raven, New York, (1983) Silver et al., J. Comp. Neurol. 210: 10-29 (1982) Schwab and Croni, J. Neurosci. 8: 2381-2393 (1988) Reier et al., In Neural Regeneration and Transplantation, pp. 183-209, Alan R. Liss, New York, (1989) Schwartz et al., FASEB J. 3: 2371-2378 (1989) David and Aguayo, Science 214: 931 (1981) Benfry and Aguayo, Nature 296: 150 (1982) Wendt et al., Exp. Neurol. 79: 452 (1983) Hagg et al., Exp. Neurol. 109: 153 (1990) Kromer and Cornbrooks, Proc. Natl. Acad. Sci. USA 82: 6330 (1985) Kromer et al., Brain Res. 210: 153 (1981) Gage et al., Exp. Brain Res. 72: 371 (1988) Wendt, Brain Res. Bull. 15: 13-18 (1985)

  Therefore, we developed a therapy that transfers genes through efficient vectors and subsequently transplants genetically modified cells into the brain in vivo, thereby improving neuronal diseases, defects or dysfunction, Thus, it would be beneficial to promote axonal regeneration to treat CNS injuries such as Alzheimer's disease, Parkinson's disease and Huntington's disease.

  The present invention provides a method for treating cells suffering from a defect, disease, or injury in a mammalian central nervous system, the method comprising a genetically modified donor in the central nervous system By transplanting the cells, it is possible to produce a sufficient amount of functional molecules to repair the cells suffering from the aforementioned defects, diseases or injuries in the central nervous system. The cells are pLN. Viral or retroviral vectors containing 8RNL and pLThRNL, ie, one or more of the inserted therapeutics encoding one or more products that act directly or indirectly on the cell Or by other methods of introducing functional DNA into cells. The cells are cultured and injected as a suspension into the central nervous system. It is also administered with a therapeutic agent to treat cells that are damaged, diseased, or damaged in the central nervous system. The method of the present invention involves selection and preparation of donor cells, transfection, and transplantation.

  The donor cells of the present invention are mixed cells of a cell type obtained from different anatomical structures, and can be primary cells or immortalized cells.

  Donor cell transplantation is performed by multiple transplantation of donor cells at several different sites within the mammalian CNS. The cells to be transplanted in combination may be a mixture of different types of donor cells, the same type of cells containing the same transgene or different transgenes, or different types of cells. Good.

  Transplantation can be done by implanting cells in a matrix material such as a collagen matrix and allowing the transplant species to survive, or by using such materials, improving the connectivity or interrelationship between damaged nerves. Can be made easier.

  Vectors used in this invention include vectors that carry a promoter, such as a collagen promoter, that can enhance the expression of one or more transgenes.

  The secretion of functional molecules is regulated by using precursors to the molecules. In the method of the present invention, a step of transplanting fibroblasts modified to express acetylcholine into the CNS, a step of oral administration of choline chloride for maintaining and enhancing secretion of acetylcholine from the transplanted fibroblasts, including.

  This invention involves enhancing expression of a therapeutic transgene product from resting donor cells transplanted into the mammalian central nervous system. It is done by inserting a resting promoter into the vector that is used to genetically modify the donor cell. The promoter can be a non-viral promoter such as, for example, a collagen promoter. Cytokines can be administered to regulate the activity of promoters to enhance transgene expression.

  Immunosuppressants will be used to suppress the production of interferon-γ and increase transgene expression. Alternatively, anti-inflammatory drugs are used for the purpose of reducing cytokine production to increase the expression of therapeutic transgenes. Growth factors can also be used to maintain the survival of donor cells transplanted into the CNS.

  The present invention relates to a method capable of treating a disease or injury caused in the CNS by transplanting genetically modified donor cells into the mammalian central nervous system (CNS). More specifically, the invention is modified to produce molecules that have a direct or indirect effect on cells within the CNS to repair damage incurred by cells derived from a defect, disease or disorder. It relates to the use of a vector carrying a functional gene (transgene) inserted into a given donor cell. Preferably, in order to treat a cell defect, disease or injury in the CNS, a donor cell such as a fibroblast, for example, is a transgene or a transgene such as a gene encoding nerve growth factor (NGF). It is modified by introducing a retroviral vector containing the gene. Genetically modified fibroblasts can be transplanted into the central nervous system such as the brain, for example, so that they can treat defects, diseases such as Alzheimer's disease and Parkinson's disease, or damage resulting from physical disorders . This treatment is due to the restoration and improvement of function in the damaged nerve as a result of expressing a transgene product that can be expressed from the genetically modified donor cell. is there. Donor cells are also used to introduce one or more transgene products into the CNS, which transgene products produce endogenous molecules that have improved effects in vivo. Is to increase.

  As used herein, the term “transgene” or “therapeutic transgene” refers to DNA inserted into a donor cell that encodes an amino acid sequence corresponding to a functional protein. This functional protein can exert a therapeutic effect on CNS cells and has an effect capable of regulating the expression of functions in CNS cells.

(Donor cell in vitro gene transfer)
The donor cell in vitro gene strategy is outlined in FIG. 2 and includes the following basic steps: (1) A step of selecting an appropriate transgene or a transgene expressed in association with CNS disease or dysfunction. (2) Stages of selection and development of appropriate and efficient vectors for gene transfer. (3) preparing donor cells from a primary culture or an established cell culture system. (4) Proving that transplanted donor cells expressing new functions are viable and can stably and efficiently produce transgene products. (5) Prove that transplantation does not cause harmful effects on the mind and body, and (6) Appearance of desirable expression system within the host.

  Functional molecules produced by the transgene useful in this invention are not limited, but include growth factors, enzymes, gangliosides, antibiotics, neurotransmitters, neurohormones, toxins, neurite growth promoting molecules, Including antimetabolites, precursors of these molecules. In particular, transgenes inserted into donor cells are not limited to tyrosine hydroxylase, tryptophan hydroxylase, NGF, ChAT, GABA-decarboxylase, dopa decarboxylase (AADC), enkephalin, hair-like It includes somatic nerve / nutrient factor (CNTF), brain-derived nerve / nutrient factor, neurotrophin (NT) -3, NT-4, and basic fibroblast growth factor (bFGF).

(Genetic modification of donor cells)
The methods described below for modifying donor cells by using retroviral vectors and fusing to the CNS are merely illustrative and typical examples that have already been used. However, other procedures have been understood and used as the technology.

  Most of the techniques that have been used to transform cells and make vectors and the like are widely practiced as techniques, and many practitioners use standardized techniques to describe special conditions and procedures. I am familiar with new ingredients. However, for convenience, the following paragraphs will serve as indicators.

(Selection of donor cells)
The choice of donor cells used for transplantation is highly dependent on the gene being expressed, the characteristics of the vector, and the characteristics of the desired phenotypic result. Retroviral vectors require cell division and DNA synthesis for efficient infection, integration, and gene expression (Weiss et al., RNA Tumor viruses, 2nd Ed., Weiss et al., Eds., Cold Spring Harbor Press , New York (1985)), if such a vector is used, the donor cell can be the olfactory mucosa or the developing or reactivating glia, such as the primary fibroblast or established cell line. It is a fairly actively growing cell line that replicates early neurons from selected areas or replicates grown neurons.

  Primary cells, ie cells that are freshly obtained from a living body such as fibroblasts, or cells that are not in a transformed state are suitable for use in the present invention. Other suitable donor cells are immortalized (transformed cells that repeat cell division) fibroblasts, glial cells, adrenal cells, hippocampal cells, keratinocytes, hepatocytes, connective tissue cells, lining cells , Bone marrow cells, stem cells, leukocytes, chromaffin cells, and other mammalian cells that can be connected by genetic techniques and methods currently invented.

  Application of methods to induce infectivity of target cells that stop and do not replicate will transform many other cell types into a state suitable for viral transduction. For example, methods have been developed for infecting primary cultures of adult rat hepatocytes that are normally immunized to such vector infections with complete retroviral vectors. Similar methods may be useful for many other cells (Wolff et al., Proc. Natl. Acad / Sci. USA 84: 3344-3348 (1987)). In addition, donor cells, such as electroporation (Toneguezzo et al., Molec. Cell. Biol. 6: 703-706 (1986)), lipofection, or direct gene insertion, Similar to efficient non-viral methods of introducing DNA, the development of many other vectors derived from herpes, varicella, adenovirus, and many others not currently infectious to retroviral infections Will be used to introduce genes into cells.

  Another characteristic of donor cells related to the success or failure of transplantation is the number of passages of donor cells. The results presented here indicate the possibility that human cells that have been subcultured will be used for transfection with transgenes for transplantation.

(Select vector)
Although other vectors are used, the optimal vector used in the presently invented method is a viral vector including a retrovirus. The chosen viral vector should meet the following criteria: 1) The vector must be able to infect donor cells. That is, a virus vector with an appropriate host must be selected. 2) The introduced gene has the ability to remain in the cell in terms of stable maintenance and expression in the cell, and is expressed in the cell unless cell death occurs. 3) The vector should do little damage to the target cell if possible. Murine retroviral vectors have become the best method for efficient foreign gene transfer and expression in mammalian cells. These vectors have a very broad host and cell type range, bind to random sites in the host genome with a well-understood mechanism, and allow stable and efficient expression of genes, under most circumstances. Does not kill or damage host cells.

(General method of vector construction)
The construction of a suitable vector containing the desired therapeutic genetic code and regulatory sequences is well understood in the art (Maniatis et al., In Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, New york ( (1982) use standard combining and reconstruction techniques. Isolated plasmids, DNA sequences, synthesized oligonucleotides are cleaved, trimmed and recombined into the desired form.

  Site-specific DNA cleavage is a suitable restriction enzyme under conditions that are widely understood in the art, or a special one specialized by the manufacturer of the restriction enzyme to make it commercially available (eg New England (See Biolabs Product Catalog). In general, approximately 1 μg of plasmid or DNA sequence is cleaved by 1 unit of enzyme in approximately 20 μl of buffer. Typically, use of excess restriction enzyme ensures complete cleavage of the DNA substrate.

  If the mutation is prominent, the cleavage is done by incubating at 37 ° C for 1-2 hours. After incubation, the protein is extracted with a phenol / chloroform extract and extracted again with an ether extract. The nucleic acid is precipitated with ether and protected in the aqueous layer. If desired, the standard technique is to separate by size of the cleaved fragments by polyacrylamide gel or agarose gel electrophoresis. A general description of size separation can be found in Methods of Enzymology 65: 499-560 (1980).

Fragments by restriction enzymes were prepared in the presence of 4 types of deoxynucleotide triphosphates (dNTPs), 50 mM Tris (pH 7.6) 50 mM NaCl, 6 mM MgCl ₂ , 6 mM DTT, 5-10 μM dNTPs, 15 ° C. at 20 ° C.-25 ° C. It may be a blunt fragment treated with a large fragment of E. coli DNA polymerase I (Klenow), treated with an incubation time of 20 minutes. The Klenow fragment has only a sticky end at the 5 ′ end and meshes with the 3 ′ end from which a single strand protrudes even in the presence of four types of dNTPs. If desired, selective pairing is done by giving only one dNTPs selected within the limits that are not found in the nature of the sticky ends. After Klenow treatment, the mixture is extracted with phenol / chloroform and ethanol precipitated. When treated with S1 nuclease or Bal-31 in an appropriate environment, any single-stranded portion is hydrolyzed.

Binding is done in a 15-50 μl volume at the standard conditions or temperatures indicated below. In addition to the reaction solution of 20 mM Tris-Cl pH 7.5, 10 mM MgCl ₂ , 10 mM DTT, 33 mg / ml BSA, 10 mM-50 mM NaCl, 40 μM ATP, 0.01-0.02 (Weiss) unit T4 DNA ligase at 15 ° C. (When binding sticky ends) or 1 mM ATP, 0.3-0.6 (Weiss) unit T4 DNA ligase at 25 ° C. (when binding blunt ends). Intermolecular sticky end bonding is usually performed at a total DNA concentration of 33-100 μg / ml (5-100 nM total end concentration). Intermolecular blunt end bonding (usually 10-30 times the linker) is done at a total end concentration of 1 μM.

  In vector construction using vector fragments, the vector fragment usually removes the 5'-terminal phosphate and inhibits the recombination of the vector to prevent bacterial alkaline phosphatase (BAP) or calf intestine-derived alkali. Treated with phosphatase (CIP). Cleavage is done in the presence of pH 8, Na +, Mg + 2, approximately 1 unit of BAP or CIP per mg of vector at 37 ° C. for about 1 hour. To recover the nucleic acid fragment, it is prepared by extraction with a phenol / chloroform mixture and ethanol precipitation. Alternatively, vector recombination can be inhibited by cleavage of restriction enzymes added to cleave fragments that are not needed.

  Methods for retroviral vector preparation have been described (Yee et al., Cold Spring Harbor Symp. On Quant. Biol. Vol. LI, pp. 1021-1026 (1986); Wolff et al., Proc. Natl. Sci. USA 84: 3344-3348 (1987); Jolly et al., Meth. In Enzymol. 149: 10-25 (1987); Miller et al., Mol. Cell. Biol. 5: 431-437 ( 1985); and Miller et al., Mol. Cell. Biol. 6: 2895-2902 (1986), and Eglitis et al., Biotechniques 6: 608-614 (1988)) and now widely used in many laboratories. Has been. Retroviruses contain long terminal repeats (LTRs) and packaging (psi, Ψ) sequences of retroviruses, as well as plasmid sequences for bacterial replication, for replication of eukaryotic cells. Other sequences such as the early promoter and enhancer of SV40 are also included. Vectors are housed as RNA in viruses and transfect DNA constructs into cell lines. These include psi (2), which produces the outer membrane of viruses that can infect rodent cells, and psi AM and PA 12, which produce the outer membrane that can infect a wide range of species.

  In the preferred viral vector, the transgene is handled under the control of the viral LTR promoter-enhancer signal or an internal promoter. The signal retained in the retroviral LTR is then efficiently bound to the genome of the host cell. To regulate hereditary viruses, the incomplete vector recombinant DNA molecule expresses all of the retroviral functions necessary to wrap viral transcripts in the hereditary viral outer membrane, but RNA transcription It is transfected into a “producer” cell line containing a provirus that lacks the critical signal necessary to encapsulate the product with the mature viral outer membrane. These include a group specific antibody (GAG) and an envelope (ENV) gene encoding a capsid protein and reverse transcriptase (POL). Due to this lack, transcripts from the helper cannot be encased in the viral outer membrane, so the producer cell produces only an empty virus. However, a combined defective retrovirus introduced into the same cell using calcium phosphate-regulated transfection, in which the GAG, ENV and POL genes are transferred by a transgene with a complete psi sequence Vectors produce transcripts wrapped in trans because they have packaging sequences. The cell contains two proviral sequences joined at different sites in the host cell genome. Since RNA transcripts from newly introduced proviruses have packaging sequences, they were efficiently encapsulated in the viral outer membrane by viral functions made in trans. Ideally, the result is production by infectious cells that liberate the transgene from the replication-competent wild-type helper virus. In most cases it is not necessary for all models of gene therapy, but the production of helper viruses is far from desirable. This is because it may lead to the spread of infection and disease to the lymphocytes and other tissues of the host animal.

  Since herpesviruses have the ability to establish a potential infection and a clear pathogenic relationship between certain neurons, herpes-derived vectors, ie HSV-HSV-1, are used. Similar, but useful transport and expression vectors can be efficiently applied to CNS cells such as rabies virus, measles, other paramyxoviruses, and human immunodeficiency retroviruses (HIV). You will gain benefits in the ultimate understanding of other human and animal viruses that are infected. In most cases, with the exception of rabies viruses, infection with these viruses is not truly neurotrophic. Rather, it is due to the fairly high infectivity of host cells. They appear to be neurotrophic because the metabolic and physiological effects of infection appear in the cells of the CNS. Therefore, it seems that many vectors derived from these viruses are related to various cells in the cell domain, and the CNS specificity for expression depends on the oligodendrologlia specific expression of JC virus, lipid protein complex Appropriate cell-specific enhancers, promoters, and other sequences must be provided to control glial cell-specific acid protein (GFAP) gene glial cell-specific expression and other CNS-specific functions in mice .

  Other viral vectors used to transfect cells to modify CNS disease are retroviruses such as Moloney murine leukemia virus (MoMuLV), JC, SV40, polyoma virus, adeno Includes papovaviruses such as viruses, Esptein-Barr (EBV), papilloma viruses, ie bovine virus type I (BPV), varicella and poliovirus, and other human and animal viruses .

  Problems arising from the use of defective viral vectors are the possibility of pathogenicity, anti-replication, the final appearance of wild-type virus and “rescue” by rearranging virus-like or other sequences in the cell It is sex. This possibility is reduced by the removal of all viral control sequences that are not necessary for vector infection, stabilization and expression.

  In addition to the methods described above for introducing functional DNA transgenes into donor cells, other methods could be used. For example, methods that do not use vectors include methods that physically transfect DNA into cells non-virally. For example, microinjection (DePamphilis et al., BioTechnique 6: 662-680 (1988)), electroporation (Toneguzzo et al., Molec. Cell. Biol. 6: 703-706 (1986), Potter, Anal. Biochem. 174: 361-373 (1988)), calcium phosphate transfection (Graham and van der EB, ibid., Chen and Okayama, Mol. Cell. Biol. 7: 2745-2752 (1987), Chen and Okayama Biotechnique. 6: 632-638 (1988)), DEAE-dextran modified transition (McCutchan and Pagano, J. Natl. Cancer Inst. 41: 351-357 (1968)) Methods, Transfection by modifying liposomes with cations (Felgner et al., Proc. Natl. Acad. Sci. USA, 84: 7413-7417 (1987), Felgner and Holm, Focus 11: 21-25 (1989) And Felgner et al., Proc. West. Pharmacol. Soc. 32: 115-121 (1989)), and other methods known in the art.

(Mechanism of phenotypic correction by donor cells)
(Donor cell preparation)
Donor cells must be properly processed for transplantation. For example, for the injection of genetically modified donor cells according to the present invention, cells such as fibroblasts obtained from skin samples are grown in a culture medium suitable for cell growth and storage, eg in contact. Place in fetal calf serum (FCS), as possible. Cells are released by culture medium culture such as phosphate buffered saline (PBS) containing 0.05% trypsin, glucose 1 mg / ml; MgCl 0.1 mg / ml; CaCl ₂ 0.1 mg / ml Store in a buffer such as PBS plus 5% serum to inactivate trypsin. Cells are washed by centrifugation with PBS and then resuspended in PBS without trypsin to a concentration suitable for injection. In addition to PBS, an osmotic equilibrium solution that is physiologically inert to the substance of the host may be used for suspension or injection of donor cells into the host.

  The long-term survival of the transplanted cells is the infection of the cells with the virus, the damage of the cells caused by the culture conditions, the method of cell transplantation and proper vascularization, the immune response to foreign cells and transgene products Depends on the effect. Although the mammalian brain has traditionally been considered an immunologically special organ, recent studies have finally revealed the immune response to foreign antigens in the rat brain. Use of vectors that do not change the antigen on the cell surface beyond the phenotypic correction, or the host cells are immune to the possibility of rejection and the reaction of the transplanted cells with the host by the transplanted cells. Even if it is possible to introduce cells at a time that are resistant to, it is important to minimize by using autologous cells.

  In the present invention, the most effective method and timing for transplanting donor cells having a mutated gene to treat disease, trauma or injury in a patient's CNS is the damage of cells such as neurons in the CNS. It depends on the severity and course of the severity, illness and trauma, the patient's health and response to treatment, and the judgment of the treating health professional.

  Of course, as with all other gene transplant systems, accurate gene expression is guaranteed to be sufficient to achieve the expected phenotypic effect and not so high as to be toxic to the cells. To be determined.

  A problem associated with using genetically engineered cells, such as those that are transplanted for gene therapy, is that gene expression (including gene transfer) is reduced when the cells are in quiescence (not dividing). This is observed ("down-regulation") (Palmer et al., Proc. Natl. Acad. Sci. USA 88: 1330-1334 (1991)). Fibroblasts initially transplanted into the brain do not continue to divide if they are transformed and become non-tumorigenic. So they exist in the resting state in the brain. Therefore, providing a means to increase the maintenance or expression of mutated genes is useful to promote long-term stable expression of therapeutic genes in fibroblasts for gene therapy without cell division It is.

  Gene expression is regulated at the transcription, translation and post-transcriptional stages. Transcription initiation is an important early event in gene expression. This depends on the promoter and enhancer sequences and is influenced by specific factors in the cell that interact with these sequences. Many prokaryotic transcription units consist of a promoter and, in some cases, components such as enhancers and regulators (Banerji et al., Cell 27: 299 (1981); Corden et al., Science 209: 1406 (1980) and Breathnach and Chambon, Ann. Rev. Biochem. 50: 349 (1981)). In the case of retroviruses, regulatory factors required for retroviral genome replication are present in long-range repeats (LTRs) (Weiss et al., Eds., In: The molecular biology of tumor viruses: RNA tumor viruses). Cold Spring Harbor Laboratory, Cold Spring Harbor, New York (1982)). Molony murine leukemia virus (MLV) and Rous sarcoma virus (RSV) contain promoter and enhancer sequences (Jolly et al. Nucleic Asid Res. 11: 1855 (1983); Capecchi et al., In: Enhancer and eukaryotic gene expression, Gulzman and Shenk, eds., Pp. 101-102, Cold Spring Harbor Laboratories, Cold Spring Harbor, New York). Promoters and enhancer regions of several non-viral promoters have also been described (Schmidt et al., Nature 314: 285 (1985), Rossi and de Crombrugghe, Proc. Natl. Acad. Sci. USA 84: 5590- 5594 (1987)).

  This invention relates to collagen type I (α1 and α2) (Prockop and Kivirikko, N. Eng. J. Med. 311: 376 (1984); Smith and Niles, Biochem. 19: 1820 (1980); deWet et al. , J. Biol. Chem. 258: 14385 (1983)), proposed a method for maintaining and increasing the expression of transgenes in quiescent cells using promoters including the SV40 and LTR promoters.

  In addition to the use of viral and non-viral promoters to regulate transgene expression in donor cells such as primary fibroblasts, it has been used to increase the level of enhancer sequence transgene expression. Enhancers can increase the transcriptional activity of some foreign genes as well as their native genes (Armelor, Proc. Natl. Acad. Sci. USA 70: 2702 (1973)). For example, in this invention, a collagen enhancer sequence was used with a collagen promoter α2 (I) to increase transgene expression.

  In addition, enhancer sequence elements found in SV40 virus have been used to increase transgene expression. This enhancer sequence is described in Grusset al., Proc. Natl. Acad. Sci. USA 78: 943 (1981), Benoist and Chambon, Nature 290: 304 (1981), and Fromm and Berg, J. Mol. Appl / Genetics, It consists of 72 base pairs that repeat as described in 1: 457 (1982), all of which are in this reference. This repetitive sequence can increase transcription of many different viral and cellular genes when present in sequence with various promoters (Moreau et al., Nucleic Asid Res. 9: 6047 (1981) ). In most cases, the SV40 enhancer element is present in the plasmid DNA, either oriented or oriented in various directions, or present either oriented or oriented in various directions. Fully functional (Fromm and Berg above). The ability of the SV40 enhancer to up-regulate transcription from the α2 (I) collagen promoter (Col) has been confirmed in chicken fibroblasts and monkey kidney CV-1 cells (Xu et al., In Enhancer and Eukaryotic Gene). Expression, Gulzman and Shenk, eds., Cold Spring Harbor Laboratories, Cold Spring Harbor, New York, pp. 51-54 (1983)). In these cells, vectors containing the SV40 enhancer region downstream of the α2 (I) collagen promoter show chloramphenicol acetyltransferase activity in CV-1 and CEF cells, respectively, using a reporter gene. It increased to 64 times and 18 times.

  Transgene expression also appears to enhance long-term stable expression after transplantation when cytokines are used to regulate promoter activity. Several cytokines have been reported to regulate the expression of transgenes from collagen α2 (I) and the LTR promoter (Chua et al., Connective Tissue Res. 25: 161-170 (1990), Elias et al. , Annals NY Acad. Sci. 580: 233-244 (1990), Seliger et al., J. Immunol. 141: 2138-2144 (1988) and Seliger et al., J. Virology 62: 619-621 (1988) ). For example, transforming growth factor (TGF) β, interleukin (IL) -1β, interferon (Inf) α or γ down-regulates the expression of transgenes controlled by viral promoters such as LTR. Tumor necrosis factor (TNF) α, TGFβ1, and IF-1β up-regulate the expression of transgenes controlled by the α2 (I) collagen promoter (Coll). The results presented here show that several cytokines such as TNFα, TGFβ, and IL-1β are used to regulate the expression of transgenes controlled by promoters in donor cells such as fibroblasts. It shows that it has come. Other cytokines that have proven useful include basic fibroblast growth factor (bFGF) and epidermal growth factor (EGF).

  A collagen promoter with a collagen enhancer sequence (Coll (E)) should be used to utilize high levels of cytokines to increase transgene expression in the modified donor cells present in the brain after transplantation Can do. In addition, anti-inflammatory agents, including steroids, such as dexamethasone, may be administered to transplanted patients immediately after fibroblast transplantation and subsequently until the inflammatory response is subsided by cytokines. In some cases, an immunosuppressant such as cyclosporine may be administered to reduce the production of interferon-γ that down-regulates the LTR promoter and Coll (E) promoter-enhancer to suppress transgene expression. Good.

  For in vivo use, a transgene controlled by a collagen promoter is introduced into a cell and then transplanted directly into the brain so that it is not affected by others. Cytokines released after transplantation as a natural response of the patient stimulate the collagen promoter controlled by transcription of the selected transgene.

  Cytokines including growth factors bFGF and EGF may also be administered before, during or after transplantation to increase the survival of donor cells transplanted into the CNS.

  It would be useful to be able to regulate the secretion of genetically enhanced gene products after transplantation. Similar to the specific example shown here, the release of gene products such as the transmitter acetylcholine (Ach) is very low in Alzheimer's disease. Expression of acetyltransferase cDNA in cultured cell choline infected with MLV vector could be increased using acetylcholine precursor choline. This suggests a method for dietary restriction of gene therapy in the cerebrum.

  Some or many CNS disease genetic modifications appear to require the proper organization or reconfiguration of synaptic connections between cells. No model system has been developed to study these possibilities. This is because there are few cell culture systems that have not been replicated and transformed, and even neurons that do not replicate do not respond to retroviral infection. However, according to recent studies, including those that require the persistence of embryonic hippocampal neurons, replicating neuronal cell culture systems are readily available for in vivo gene transfer and even in vivo transplantation. It seems to be available (Caettano and MacKay, Nature 347: 762-765 (1990)).

  Such neurons allow efficient introduction by vectors from retroviruses and other viruses, and if they can retain their neuronal characteristics, they can be transferred to other cells after transplantation into the brain. A synaptic junction could be formed. As an alternative, there are cells existing in the CNS that are slow-growing like the dentate gyrus lobe of the hippocampus and those that continue to divide in the living body like the olfactory mucosa and the dentate gyrus.

  The use of non-neuronal cells for transplantation prevents the development of specific neural junctions with target cells inside the host. So, the effect on fibroblasts and other non-neural donor cells and the target cell phenotype in vivo is either the diffusion of required gene products and metabolites, or gap junctions ("metabolic co-operation" ") Or uptake by target cells of genetic products and metabolites secreted from donor cells. Donor cells also remove neurotoxins as “sinks” of poisons by the expression of new gene products and metabolites.

  Instead, neural bridges promote reconnection between neurons in damaged CNS tissue. As noted above, donor cells suspended and transplanted in a culture medium such as a collagen matrix can act as nerve bridges to promote axonal regeneration and rejoining damaged neurons, and to produce synthetic or in vivo substances such as nerves. It seems to be used to communicate with nerve bridges formed from disrupted suspensions of cells and placenta or extracellular matrix that extends neurites.

(Transplant)
The method of the invention contemplates transplantation of donor cells containing a transgene for insertion into a region of the CNS that has undergone defects, disease, and trauma into the brain.

  Nerve transplantation, or “transplantation”, includes cell implantation into the central nervous system, or cell implantation into the ventricular cavity or subdura on the surface of the host brain. Conditions for successful planting are (1) transplant survival (2) graft retention at the site of implantation (3) minimum amount of pathological response at the site of implantation.

  For example, methods for implanting various neural tissues, such as fetal brain tissue, into the host brain have been described in Neural Grafting in the Mammalian CNS. Bjorklund and Stenevi, eds., (1985) Das, Ch. 3 pp. 23-30; Freed, Ch. 4, pp. 31-40; Stenevi et al., Ch. 5, pp. 41-50; Brundin et al., Ch. 6, pp. 51-60; David et al., Ch. 7, pp. 61-70; Seiger, Ch. 8, pp. 71-77 (1985), which are hereby incorporated by reference. Has been. These procedures include intraparenchymal planting. That is, in the host brain that is achieved by injection or replacement into the host brain to be placed in the brain parenchyma at the time of implantation (this is compared to extraparenchymal implantation). (See Das, ibid.).

  The two main procedures of endothelium implantation are (1) injecting donor cells into the host brain parenchyma, (2) preparing a cavity with surgical means to expose the host brain parenchyma, and transplanting into the cavity Replacing a piece (see Das, ibid.). Both methods provide substantial juxtaposition between the graft and host brain tissue at the time of transplantation, and both methods facilitate anatomical integration between the graft and host brain tissue. This is important when it is necessary for the graft to become an integral part of the host brain and for the survival of the host's life.

  Instead, the graft will be placed in the ventricle. For example, it is provided in the cerebral ventricle or the dura mater, or provided on the surface of the host brain that is separated from the host brain by incorporating the buffy coat, the capsule and the buffy coat. Transplantation into the ventricle can be accomplished by infusion of donor cells or can be accomplished by growing the cells in a culture medium such as 3% collagen to form a solid tissue plug. This solid tissue plug is implanted into the ventricle to prevent graft metastasis. For subdural implantation, cells are injected around the surface of the brain after creating a slit in the dura mater. Injection into selected areas of the host brain is done by piercing through the dura to allow the needle of the microsyringe to be inserted. The microsyringe is preferably placed in a stereotaxic frame and the three-dimensional stereotaxic coordinates are chosen to place the needle in the desired location of the brain and spinal cord.

  Donor cells are also fed into the nucleus, nucleoplasm, hippocampal cortex, striatum, and caudate region of the brain as well as the spinal cord.

  For donor cells that have been passaged, preferably the cells are passaged by about 2 to about 20 passages.

  For transplantation, the cell suspension is drawn up into a syringe and administered to an anesthetized transplant recipient. Multiple injections are made using this procedure. The generation of donor tissue, ie the stage of development, affects the good outcome of cell survival after transplantation.

Thus, the cell suspension procedure allows for the transplantation of genetically modified donor cells to a given site in the brain and spinal cord that is relatively intact, and several different sites, It allows simultaneous compound transplantation at the same type of site using the same cell suspension, and also allows a mixture of cells from different anatomical regions. A compound transplant consists of a mixture of cell types or a mixture of transgenes inserted into cells. Preferably, about 10 ⁴ to about 10 ⁸ cells are provided for a unit transplant. For implantation in a cavity convenient for spinal cord transplantation, tissue is removed from the area near the outer surface of the CNS to form the implantation cavity. This is done, for example, by removing bone on the brain and stopping the bleeding with something like a foamy gel, as described by Stenevi et al. Suction is used to create a cavity. One or more transplants are made in the same cavity using infusion of cells or solid tissue transplants.

  Transplanting donor cells into the traumatic brain requires a different procedure. For example, before attempting a transplant, the damaged site must be cleaned and bleeding stopped. In addition, the donor cells must have sufficient growth potential to fill host brain damage and cavities to prevent transplant translocation in the pathological environment of the traumatic brain.

  Thus, the present invention provides a method for genetically modified donor cells into the transplanted spinal cord to handle CNS defective, diseased or damaged cells.

  The method of the invention is also intended for use in transgene donor transplantation in conjunction with other therapeutic procedures to treat diseases and trauma in the CNS. Thus, the genetically modified donor cells of this invention are co-transplanted with other cells, both beneficially working genetically modified cells and non-genetically modified cells from the adrenal gland. It affects cells in the CNS, such as chromaffin cells, fetal brain cells, and placental cells. This genetically modified donor cell is provided to support the viability and function of co-transplanted genetically unmodified cells. For example, there are fibroblasts modified to supply in vivo nerve growth factor (NGF) as described in the examples below.

  Furthermore, the genetically modified donor cells of this invention are effectively co-administered with therapeutic drugs in the handling of CNS defects and injuries and diseases. This CNS is like a growth factor: a nerve growth factor; a ganglioside; an antibiotic, a neurotransmitter, a neurohormone, a toxicant, an antimetabolite, a neurite promoting molecules, and a precursor of dopamine It is a precursor of these drugs, such as body L-DOPA.

  The method of this invention is illustrated by a preferred embodiment. There, donor cells containing a vector carrying a therapeutic transgene are transplanted into the cortex within the specimen. In the first preferred embodiment, the established HPRT-deficient rat fibroblast cell line 208F, primary rat fibroblasts and primary rat astrocytes one day after birth are genetically modified using retroviral vectors. It has been used to show that cultured cells survive survival during transplantation in the mammalian brain and continue to express the transgene product.

  In a second preferred embodiment, fibroblasts are genetically modified to secrete NGF upon infection with a retroviral vector, and the modified fibroblasts cause surgical damage to the fimbriae foramen region. It is transplanted into the brain of a rat. Transplanted cells survived and provided sufficient NGF to prevent alteration of cholinergic neurons that would die without treatment. In addition, protected cholinergic cells germinate axons. This protruded in the direction of NGF producing cells.

  In a third preferred embodiment, fibroblasts were genetically modified to express and secrete L-dopa upon infection with a retroviral vector. The modified fibroblasts were also transplanted into rat tails that model Parkinson's disease as a result of unilateral dopamine depletion. The cells supply enough L-dopa to survive and reduce the circulatory movement caused by dopamine depletion.

  In a fourth preferred embodiment, skin primary fibroblasts isolated from skin biopsy and maintained in culture were used as autologous cells for intracortical transplantation into adult rat striatum. Once confluent, fibroblasts cease to proliferate and have in vitro repressed contact. Fibroblasts survive for at least 8 weeks with the following intracortical transplantation and continue to synthesize collagen and fiboronectin in vivo. The transplant also maintains the same volume for 3 to 8 weeks. This indicates that the skin primary fibroblasts are not swollen or dead. It was observed that dynamic host-transplant interactions, including phagocytic cell migration, astrocyte hypertrophy, and transplantation wetting, show structural integration of skin primary fibroblast transplantation in the adult rat CNS.

  In a fifth preferred embodiment, skin primary fibroblasts obtained from skin biopsies from rat inbred lines were used as donor cells for genetic modification and transplantation. When implanted into the brain of a rat of the same genetic line as a donor rat, fibroblasts containing a transgene for either tyrosine hydrokinase (TH) or β-galactosidase survived 10 weeks and continued to express the transgene. . TH synthesized by transplanted fibroblasts appeared to convert tyrosine to L-dopa as observed in vitro and to affect the host brain as assessed through behavioral measurements. It has been shown that supplying L-dopa locally to the striatum is sufficient to partially compensate for the loss of striatal dopaminergic input.

  In the sixth preferred embodiment, the ability of senile fibroblasts to serve as donor cells for human NGF was demonstrated.

  In a seventh preferred embodiment, rat fibroblasts are genetically modified to express and secrete choline acetyltransferase (dChAT). Intracortical and extracortical levels of ACh have been shown to increase with the addition of exogenous choline chloride. In addition, serum deficiency and induced confluence caused a 80% reduction in rebound ChAT activity. ACh release (compared with active growing cells) was also suppressed in resting fibroblast cultures. Exogenous choline mitigates the decrease in ACh secretion. These results indicate that fibroblasts can be genetically modified to supply ACh and that ACh release can be regulated, eg, increased by supplying choline into the culture medium.

  In a ninth preferred embodiment, cytokines, anti-inflammatory drugs and dexamethasone are evaluated for their efficacy on LTR promoter activity and enhance the expression of CAT and dChAT genes in skin primary fibroblasts. In addition, a collagen promoter α2 (I) with a collagen enhancer (Coll (E)) was used to assess the efficacy of cytokines in the expression of CAT in resting fibroblasts.

  In a tenth preferred embodiment, skin primary fibroblasts were modified to express and secrete NGF and were transplanted into the striatum of adult female rats. From one to three weeks, the following striatal implants and NGF receptor-immunoreactive axons surrounded the NGF-supplied fibroblast grafts. From 3 weeks to 8 weeks, NGF receptor-positive profiles were found inside the graft. Control of transplantation of normal primary fibroblasts depletes immunoreactive axons. By ultrastructural testing, unmyelinated axons within the extracellular matrix of the graft are surrounded by the progression of active astrocytes, and different basal layers surround the axon glial bundles It was shown. The progression process of active astrocytes was evident in both control and transplantation involving genetically modified primary fibroblasts. This example describes an in vivo model, where the release of NGF from the graft induces the growth direction of NGF receptor-positive axons and the mature active astrocytes migrate the axons Of an acceptable substrate. In this example, genetically modified fibroblasts were also placed in collagen matrix grafts to assess the reproductive capacity of adult rat media septa. The graft supplied NGF and also promoted regeneration of the diaphragm axon; the graft possessed a number of unmyelinated axons compared to control fibroblast transplantation. Regenerating phrenic axons distributed to the hippocampus with afferent block; topographical and synaptic organization of diaphragm input in the hippocampal dentate gyrus is a normal cholinergic nerve distribution caused by the diaphragm nucleus It was the same as that.

  In order to better understand the invention described herein, the following experimental examples are described. It should be understood that these experimental examples are given for illustrative purposes only and are not intended to limit the scope of the invention.

[Experimental Example I]
“Transplantation of cells that have been genetically modified to express the HPRT transgene into the brain”
(Infecting cells)
Expression of HPRT cDNA in 208F rat fibroblasts (Jolly et al., Proc. Natl. Acad. Sci, USA 80: 477-481 (1983)) defective in donor hypoxanthine guanine phosphoribosyltransferase (HPRT) Prototype HPRT vector pLS P LM (Miller et al., Science 225: 630-632 (1984); Yee et al., Gene 53: 97-104 (1987)) or Tn5 transposon neomycin-resistance gene (neo ^R ) and Neo ^R- luciferase vector pLNHL2 (Eglitis et al., Sience 230: 1395-1398) expressing both firefly luciferase cDNA (de Wet et al., Molec. Cell. Biol. 7: 725-737 (1987)). (1985); Wolff et al., Proc. Natl. Acad. Sci. (USA)) (FIG. 4). The pLSP LM vector is derived from the vector pLPL2 (Miller et al., Molec, Cell. Biol. 6: 2895-2902 (1986)) and is added with a new C-terminal hexagon added by in vitro mutagenesis of the translation stop codon. Contains human HPRT cDNA encoding a protein with a peptide. The vector pLpLM (Yee et al., Gene 53: 97-104 (1987)) was digested with Xhol and BamHl into a 1.3 Kb long fragment, and the plasmid pLpL2 (Miller, ibid.) Already cleaved with Xhol and BamHl. Reference). The resulting vector is pLS P LM.

The vector pNLHL2 carries the cDNA encoding firefly luciferase (LUX), the Tn5 neomycin-resistance gene (neo ^R ) and the human cytomegalovirus immediate early gene (HCMV) promoter and enhancer. 5 'and 3' LTRs were obtained from the replicating murine leukemia gene (MLV) described by Mason et al. (Science 234: 1372-1378 (1986)). The vector pNLHL2 was prepared as follows: Plasmid pLNHPL2 (also known as PNHP-1 by Yee et al., Proc. Natl. Acad. Sci. USA 84: 5197-5209 (1987)) was cut with BamHI and HPRT Remove the DNA sequence. The ends were repaired with Klenow polymerase. Plasmid pLV2A (deWet et al., Molec. Cell. Biol., Supra) (University of California, San Diego, Calif., Provided by Dr. Subramani) was cut with HindIII and Sspl to isolate the luciferase fragment. The ends were repaired by the procedure described above. PNLHL2 cleaved with BamHl and pSV2A cleaved with HindIII-Sspl were combined to synthesize a vector pNLHL2 (FIG. 5).

  To ensure that only infected cells are used, the cells are selected media for cells expressing HPRT containing hypoxanthine, aminopterin and thymidine (HAT), and cells expressing neoR containing the neomycin analog G418. Incubated with both selective media for.

  Primary fibroblasts and astrocytes were infected with the neoR-luciferase vector only. Subsequently, HAT-resistant and G418-resistant cells were incubated overnight in serum-free medium or medium containing rat serum to suppress immune responses in the rat brain.

(Transplant)
The cells were re-stirred in an equilibrated glucose-saline solution and injected stereotactically into several regions of the rat brain using a sterile syringe. A solution containing 10,000 to 100,000 cells per 1 μl was injected at a rate of 1 μl / min for a total volume of 3-5 μl. After 1 week to 3 months, the animals died and the areas containing the transplanted cells were identified and excised and then examined histologically and biochemically.

(Histological analysis)
To describe histologically the transplanted cells, to describe general morphological features, with Nissl colorant and cresyl violet, and for specific cell antigen markers (fibronectin for fibroblasts, for glial cells To establish the presence of fibrillar acidic protein (GFAP), the brain was colored and sectioned by perfusion of each reagent into the rat body using immunochemical methods. In summary, the cut pieces were rinsed in Tris-buffered saline (pH 7.4) containing 0.25% Triton-X. The cut pieces were rabbit polyclonal antibodies to fibronectin (diluted 1: 2000; baralle, University of Oxford, England) and GFAP (Gage et al., Exp. Neurol. 102: 2-13 (1989); Dakopatt, Glostryp, Denmark A 1: 1000 dilution of TBS with 0.25% Triton-X and 3% goat serum, or rat macrophage membrane polypeptides, granular leukocytes and dendritic cells (MRC OX-42, Serotec Incubation was carried out at 4 ° C. for 24 hours in a 1: 100 dilution of 0.1 M TBS containing 0.25% Triton-X of mouse IgG2a and 1% bovine serum, which are monoclonal antibodies against After thorough rinsing, the cut pieces are in a 1: 200 dilution of biotinylated goat anti-rabbit IgG (Vectastain) in 0.25% Triton-X and 0.1M TBS containing 15% horse serum. Incubation was followed by rinsing in TBS containing 0.25% Triton-X and 1% goat serum or 1% horse serum. The cuts contained 0.25% Triton-X in complex with avidin and biotinylated horseradish peroxidase (Vectastain, ABCkit, Vector Labs, Burlingame, Calif.) And 1% goat serum or 1% horse serum. Incubate for 1 hour at room temperature in 1: 100 diluted solution with 1M TBS, followed by thorough rinsing. Peroxidase was reacted in a TBS solution of 0.05% 3,3-diaminobenzidine tetrahydrochloride (DBA) (Sigma Chemical Co., St. Louis, MO), 0.05% NiCl2 and 0.01% H2 O2 for 15 minutes at room temperature. I made it clear.

  FIG. 6 shows primary rat fibroblasts transplanted into rat neostriatum 7 weeks ago. A series of 40 μm thick pieces are antifibronectin (FIGS. 6 (a) and 6 (d)), cresyl violet (Disbrey et al., Historogical Laboratory Methods, E. & S. Livingstone). , Edinburgh and London (1970)) (FIGS. 6B and 6E) and anti-GFAP (FIGS. 6C and 6F). Surviving cells appeared intact and appeared to aggregate or aggregate around the injection area. Similar to what can be seen only in the cannula tube, the cells show intense staining with fibronectin at the center of the graft and distinct staining with GFAP-colorant at the periphery of the graft due to reactive gliomatosis showed that. However, slight GFAP staining was also observed in the graft itself. Small round cells darkly colored by cresyl violet were observed in the area of the graft, which may be microglia or lymphocytes that have penetrated due to injury. Macrophages were also observed in many grafts. In many fibroblasts, it could be identified by the elongated shape seen by the coloration of cresyl violet and the pink color knitted by the sheet of collagen material surrounding the cells. The appearance of 208F fibroblasts is similar to primary fibroblasts (not shown). Astrocyte grafts also showed the same appearance, but were not fibronectin positive and stained with GFAP through the center of the graft. For these three cell types, no difference was observed between retrovirus-infected cells and control cells. An important property of these cell suspension grafts is that most cells collect and remain around the injection site, and in such an environment, they move from the injection site to a distance in the host brain. There wasn't. This apparent lack of mobility is quite different from other donor cell types and transplant sites, so the area in the brain where the cells are transplanted, the nature of the donor cells, the phenotype of the target cell for the transgene, etc. This may be an important factor in selecting donor cells.

(Characteristics of transplanted cells)
The transplanted cells were dissected and prepared for re-culture, and the cells were thawed with trypsin and ready for biochemical and molecular characterization. For the detection of human HPRT activity, cell extracts were prepared from the majority of each sample according to the procedure previously shown, and HPRT analysis focused on the isoelectric point of the polyacrylamide gel (Jolly et al., Proc. Natl. Acad. Sci. (USA) 80: 477-481 (1983); Miller et al., Proc. Natl. Acad. Sci (USA) 80: 4709-4713 (1983); Willis et al., J. Biol Chem. 259: 7842-7849 (1984); Miller et al., Sience 255: 630-632 (1984); Gruber et al., Science 230: 1057-1061 (1985), and Yee et al., Gene 53 : 97-104 (1987)). The residue of each sample was placed on the medium. Results of HPRT gel analysis of rat 208F cell HAT resistance after infection with HPRT vector were transplanted into one of the rat brainstem ganglia. The state after 3 and 7 weeks after transplantation and the state before analysis are shown in FIG.

  The presence of human HPRT enzyme activity means that infected and genetically altered rat 208F cells transplanted into the brain survive and express HPRT transgenes at a level that can be easily detected for at least 7 weeks. Indicates to continue. In addition, the transplanted cells were successfully recultured and produced cells that were morphologically identical to the original cultured tissue. Infecting these cells with helper virus produced HPRT virus, which establishes cellular identity and indicates that the provirus is intact. Studies with the neoR-luciferase vector establish the survival and expression of luciferase infected cells.

[Experimental Example II]
“Transplantation of cells genetically modified to express NGF into damaged brain”
The above experimental examples demonstrated that cultured cells genetically modified using retroviral vectors remain viable and express transgene products when transplanted into mammalian brain. This experimental example was conducted to determine whether genetically modified cells could supplement or repair the function of brain that was deficient in vivo or previously damaged.

(Construction of NGF vector pLN.8RNL)
According to a previously reported method (Wolf et al., Mol. Biol. Med. 5: 43-59 (1988)) the retrovirus was transformed into Moloney murine leukemia virus (MoMuLV) (Varmus et al., RNA Tumor Viruses; Weiss et al., Cold Spring Harbor, NY, p. 233 (1982)). The pLN.8RNL vector contains the 777 base pair Hgal-Pstl fragment of mouse NGF cDNA (Scott et al., Nature 302: 538 (1983); Ullrich et al., Nature 303: 821 (1983)) at the 5 ′ end of the virus. Included under LTR restrictions. This insertion corresponds to a shorter NGF transcript (Edwards et al., Nature 319: 784 (1986)) that is common in mouse tissues that have undergone exchange inhibition and is believed to encode a constitutively secreted NGF precursor. It is done. The vector also controls the neomycin resistance function of transposon Tn5 (Southern et al., J. Mol. Appl. Genet. 1: 327 (1982)) under the restriction of the internal Rous sarcoma virus promoter. Included a visual marker.

  The plasmid pSPN15 '(Wolf et al., Mol. Biol. Med. 5: 43-59 (1988), Harvard Medical School Boston, MA, provided by Dr. Blakefield) is a known method (Maniatis et al., Cold Spring Harbor). Press, Cold Spring Harbor, New York, (1982)) and a 777 base pair (bp) DNA fragment containing the NGF sequence, digested with the restriction enzymes Pstl and Hgal, was purified using standard purification methods (Maniatis, et al., Isolated by the above literature). 777 bp was then blunt-ended to the plasmid pLMTPL by the method described by Yee et al. In pro. Natl. Aca. Sci. (USA) 84: 5197-5201 (1987). The plasmid pLMPTL was digested with HindIII to remove the metallothionein promoter and most of the HPRT cDNA and overhanging 5 'ends were repaired with the method of Maniatis et al. (Supra) using Klenow polymerase. The overhanging end of the 777 bp fragment isolated as described above was similarly repaired. The 777 bp fragment was then blunt-ended into the digested plasmid pLMPTL. The resulting plasmid was PLN. It is called 8L. E. coli DH1 strains were transfected, cultured and purified by established methods for plasmid purification including cesium chloride centrifugation (Maniatis et al., supra). The purified plasmid pLN. 8L was digested with the restriction enzymes BamHI and Clal, and the resulting 6.1 kilobase fragment was ligated to the 2.1 k base BamHI-Clal isolated from plasmid pLLLRNL. Plasmid pLLRNL is derived from plasmid JD204 described in de Wet et al., Mol. Cell. Biol. 7: 725-737 (1987) as follows: 1717 bp HindIII- from the firefly luciferase gene derived from plasmid JD204 The Ssp1 fragment and the 1321 bp HindIII-Smal fragment of the plasmid pSV2NeoR described in Mol. Appl. Genet. 1: 327-341 (1982) by Southern and Berg contain a mutant RSV promoter in the 300 bp BamHI-HindIII fragment from plasmid pUCRH I was connected. Plasmid pLLLRNL is depicted in FIG.

  Plasmid pUCRH was generated as follows. Plasmid pRSVneoR was restricted using HindIII and the linearized plasmid was obtained by restriction using HindIII to remove the HPRT sequence (FIG. 9) (University of Carifornia, San Diego, CA Dr. Friedmann From the donation) to the remaining fragment. Plasmid pPR1 was obtained from p4aA8 (Jolly et al., Pro. Natl. Acad. Sci. 80: 477-481 (1983)) using Pst and Rsa. The resulting plasmid is called pRHN and is restricted by PvuII and recombined to form plasmids PN (+) and PAC (−). PN (+) is limited using BamHl. The resulting linearized plasmid was ligated with the fragment obtained from the pSVori plasmid restricted using BamHI. Plasmid pSVori was obtained by restriction of plasmid p4aA8 with Sall and Pstl and subcloning of the resulting fragment with plasmid pUCl8 (Bethesda Research Laboratories, Gaithersburg, MD) restricted with Sall and Pstl. The resulting plasmid is called pSVori.

  The plasmid pRH + S + obtained by ligating the BamHL restriction plasmid PN (+) and the BamHl restriction plasmid pSVori was restricted by MstII and the overhanging 5 'end was repaired as described above using Klenow polymerase. This fragment was ligated to M13mp18 (Bethesda Research Laboratories) linearized by Smal and phosphate hydrolyzed by calf intestinal alkaline phosphatase (Boeringer Mannheim, Mannheim, W. Germany). The resulting plasmid was called pmpRH and contained HPRT cDNA expressed from the RSV promoter.

  Plasmid pmpRH was subjected to position-directed mutagenesis by Kunkel et al. Proc. Natl. Acad. Sci. USA 82: 488-492 (1985) to change the polyadenylation signal from AATAAA to AGCAAA. After mutagenesis, the resulting plasmid was restricted by HindIII, and the resulting fragment was ligated to a HindIII fragment obtained by restriction of plasmid pUC19 (Bethesda Research Laboratories) to yield plasmid pUCRSV. Plasmid CRSV was restricted using BamHl and Pstl to yield a fragment containing the RSV promoter. This fragment is divided into a Pstl-Sstl fragment containing the gene encoding plasmid HPRT obtained by restriction of plasmid pLS P LM by the method described in Experimental Example I above, and a BamH1-Hindlll fragment obtained from plasmid pUC19. Ligated to generate plasmid pUCRC. The binding product between the BamHl-Hindlll fragment from pUCRH, the 1717 bp Hindlll-Sspl fragment from pJD204, and the 1321 bp Hindlll-Small fragment from pSV2NeoR was transfected and purified in the established manner as described above. It was called pLLRNL (FIG. 8).

  After transfection and purification, plasmid pLN. The plasmid resulting from the ligation of the 6.1 BamH1-Clal kilobase fragment from 8L and the 2.1 kilobase BamHl-Clal fragment from plasmid pLLRNL is pLN. It is called 8RNL.

(Adjustment of infectious retrovirus)
The infectious retrovirus is pLN. 8RNL was added to PA317 amphotropic producer cells (Miller et al., Mol. Cell. Biol. 6: 2895 (1986), provided by Dr. Miller, Fred Hutchinson Cancer Research Center) and calcium phosphate coprecipitation (Graham et al. al., Virology 52: 456 (1973)) using the medium of these cells to produce Ψ-2 ecotropic producer cells (Mann et al., Cell 33: 153 (1983)) with 4 μg / ml polybrane. (Sigma Chemical, St. Louis, Mo). A virus from a Ψ-2 clone showing the highest titer of 4 × 10 ⁵ colony forming units / ml was introduced into the established rat fibroblast cell line 208F (Quade, Virology 98: 461 (1979)) by Proc by Miyanohara et al. Used to infect in the manner reported by Natl. Acad. Sci. USA (1988).

(NGF production and secretion assay)
Individual neomycin resistant 208F colonies selected from media containing the neomycin analog G418 were grown and two-site (ELISA) enzyme immunoassay (Korsching et al., Proc. Natl. Acad. Sci. (USA) 80: 3513- 3516 (1983)) tested NGF production and secretion using commercially available reagents according to the manufacturer's protocol (Boehringer Mannheim). Clone producing the highest levels of NGF includes a NGF of total cellular protein per 1.7ng of 1 mg, secreted NGF into the medium at a rate of 50 pg / HR10 ⁵ cells. NGF secreted by this clone is determined by its activity inducing axonal derivation from pc12 rat chromocytoma (Greene, et al, Pro. Natl. Acad. Sci. USA 73: 2424 (1976)). As biologically active. Uninfected 208F cells, in contrast, did not produce detectable levels of NGF in either assay.

(Cutting the oviduct's fimbriae cap)
Axotomy of oviductal fimbria was performed as described in Gage et al., Brain Res. 268: 27-37 (1983) and Neuroscience 19 (1) 241-255 (1986). Briefly, female adult Sprague-Dawley rats (Bantin and Kingman, San Francisco, Calif.) Weighing 200 g-225 g at the start of the experiment were used. The animals were anesthetized by intraperitoneal injection of a ketamine-xylazine mixture (10 μg / kg Ketalar, Parke-Davis AnnArbor, MI and 5 μg / kg Rompun, Hoechst, Frnnkfurt, W. Germany). Unabsorbed lesions were completely removed on one side of the tubal fimbriae by aspiration through the zonal cortex and the method described in Gage et al, Brain Res. 268: 27-37 (1983). And a partial denervation on the hippocampal target was made by removing the dorsal apex of the hippocampus and overlying cingulate cortex. All animals in each experimental group received the same unilateral suction lesion. Oviductal fimbriae lesions were made in 16 rats as described above; 8 rats were transplanted with infected cells and the remaining 8 rats were uninfected control rats.

  Although the portion stained with fibronectin is a fibroblast-specific marker, it was found that both groups showed similar transplant survival (FIGS. 13 (a) and 13 (b)). The amount of choline acetyltransferase (ChAT) stained to assess the survival rate of cholinergic cell bodies is greater in the lesion of the inner septum of the animal transplanted with the infected cells than in the animals that received the control transplant. These neurons were found to remain (FIGS. 13 (c) and 13 (f)).

Retrovirus-infected cells (NGF secretion) and control 208F cells were removed from confluent plates using Dulbecco's phosphate buffer saline (PBS) containing 0.05% trypsin and 1 mM EDTA. Extraction was performed using MgCl and CaCl ₂ (complete PBS) and 5% rat serum to inactivate trypsin. Cells were pelleted by centrifugation at 1000 × g for 4 minutes at 4 ° C., washed twice with complete PBS, and resuspended in complete PBS at 10 ⁵ cells / μl. 4 μl of suspended cells were injected free of charge into the animal's body cavity and the lateral ventricle on the same side of the body cavity using a Hamilton syringe. A gelatin foam was gently placed on the surface of the body cavity and the animal was sutured.
(Immunohistochemistry) Two weeks after surgery, rats were perfused, brains were removed, fixed overnight and placed in 30% sucrose with phosphate buffer for 24 hours at 4 ° C. The 40 μm thick section was cut on a frozen slide microtome and stored at −20 ° C. in cryoprotectant (phosphate buffer glycerol and ethylene glycol). All fifth parts were immunohistochemically labeled by standard methods using polyclonal antibodies against fibronectin to assess fibroblast viability. A polyclonal antibody against choline acetyltransferase (anti-ChAT antiserum) was made and the viability of cholinergic cell bodies as described in Gage et al., J. of Comparative Neurol. 269: 147-155 It was evaluated. The tissue parts were processed immunohistologically according to a modification of the Havi et al., 29: 1349-1353 (1981) and the literature cited therein, avidin-biotin labeling. The method consists of the following steps: 1) overnight incubation with ChAT antibody or control antibody (ie preimmune or absorbed anti-serum). The ChAT antibody was diluted 1: 1500 using 0.1M Trissaline containing 1% goat sea ram and 0.25% Triton X-100; 2) 1: 1 hour incubation with biotinylated goat anti-rabbit antibody IgG (Vector Laboratories, Burlingame, Calif.) Diluted to 200; 3) ABC complex diluted 1: 100 with Trissarum containing 1% goat salem ( 1) incubation with Vector Laboratories); 4) treatment with 0.05% 3, 3 'diaminobenzidine (DAB), 0.01% hydrogen peroxide and 0.04% nickel chloride in 0.1M Tris buffer for 15 minutes . The immunologically labeled tissue portion was placed on a glass slide, air-dried, and covered with a parmount and a glass cover slip. Two parts stained with ChAT separated by 200 μm with a septum were used to assess the extent of cholinergic cell viability. All ChAT positive cells in the ipsilateral and contralateral septum were sorted according to the area of the plane using the Olympus Que-2 image analysis system and counted respectively. The tissue was also stained with acetylcholinesterase according to the method of Herdreen et al., J. Histochem. 33: 134-140 (1985) and the literature cited therein, and whether the tubal fimbriae cap was completely cut or not. Was evaluated.

  Neuronal viability is quantified (FIG. 14) and expressed as a percentage of cholinergic cells remaining in the septum on the lesioned side compared to the septum on the opposite side doing nothing, NGF secreting cells Was 92% in animals transplanted with, but only 49% in animals transplanted with control cells. The results of the control group corresponded to previous observations in animals that did not receive transplants but had lesions (Gage et al., Neurosci. 8: 2155 (1986); Williams et al., Proc. Natl ACad. Sci. USA 83: 9231 (1986); Kromer, Science 235: 214 (1987); Gage et al., J. Comp. Neurol. 369: 147 (1988)).

  In addition to a significant increase in the percentage of ChAT positive cells in the NGF group, it was shown that AChE positive fiber and cell staining also increased (FIG. 15). The most striking finding was a strong and rapid growth reaction in the quarter circle biased to the back of the septum, with the area adjacent to the cavity containing the transplanted cells being the most intensely stained (FIG. 15).

  The above results demonstrate the ease of transgene expression by cells transplanted into the CNS, and for the first time phenotypic correction in animals as a result of transplantation of genetically modified cells. It is shown.

[Experimental Example III]
“Transplantation of fibroblasts genetically modified to express L-dopa in the CNS of rats infected with Parkinson's disease”
This example demonstrates that the presently invented method of genetically modifying donor cells and transplanting the cells into the CNS significantly implicated disease symptoms in animal models, such as the rat model of Parkinson's disease. It has been proven and accepted to improve.

The method for imparting the ability to produce L-dopa to the fibroblasts used in this experimental example is the step of limiting the reaction rate of catecholamine synthesis to tyrosine hydroxylase (TH), and tyrosine to L-dopa. It is based on the ability to catalyze the action of conversion. The cofactor of TH, tetrahydrobiopterine (H ₄ -B), requires TH enzyme activity. Brain holds biopterin in a high level, and fibroblasts biopterin H ₂ - because it reduces the biopterin, TH seems encouraging to fibroblast cells located in the brain.

(Construction of retroviral vector pLThRNL)
The vector pLThRNL is a retroviral vector derived from Moloney leukemia virus, but is constructed to express a rat cDNA of tyrosine hydroxylase (TH) derived from the 5 ′ LTR sequence and transcribed from the position of the intracellular RSV promoter. It contained a resistance gene. Fragments from these three plasmids, pLRbL, pTH54, and pLHRNL were joined together from pLMTPL. Plasmid pLRbL was obtained by cutting plasmid pLMTPL with restriction enzymes Hind III and Hpal (as already described in Experimental Example II) and removing the fragment containing the HPRT gene. The remaining plasmid DNA was constructed by inserting DNA encoding the retinoblastoma gene (Rb) into plasmid pGEM1 obtained from Promega, Madison, Wis. (Provided by Dr. Lee at the University of California, San Diego, USA). The resulting plasmid was named pLRbL.

  1688 bp fragments containing rat TH cDNA were obtained by digesting plasmid pTH54 with restriction enzymes BamIII and Sph1 (O'Malley, J. Neurosci. Res. 60: 3-10 (1986), suppied). by Dr. O'Malley, Washington University, St. Louis, Mo).

  Producer producing a virus carrying the gene encoding tyrosine hydroxylase by ligating the fragments from plasmid pTH54 and plasmid pLRbL with the fragments of Cla1 and Sma1 obtained from plasmid pL2RNL (described in Experimental Example I) The vector pLThRNL containing the retroviral provirus was obtained for transfection into cells. The origin of pLThRNL and the circular restriction enzyme map are shown in FIG.

As shown in Experimental Example II, a helper-free retrovirus was produced and infected with the retrovirus. The plasmid DNA containing the above LThRNL provirus is introduced here by reference, but as described by Wigler et al., Cell 11: 223-232 (1977), Fred Hutchington Cancer Research in Seattle, Washington. CaPO _{4 was} transfected into the amphotropic PA317 helper cell line supplied by the center mirror. Two days after transfection, the media of these cells was filtered and transferred to the ecotropic Ψ2 helper cell line (Miller et al., Mol. Cell. Biol. 6: 2895-2902 (1986), supplied by Dr. Miller). Used to infect. The G418-resistant Ψ2 clone (Ψ2 / TH) was selected that contained TH activity at the highest level and produced the most viral load (5 × 10 ⁵ / ml). Immortalized rat fibroblasts (208F) (Quade, Virology 98: 461-465 (1979)) have a multiplicity of infection (MOI) of ^10-4 or less of the LThRNL virus produced by Ψ2 / TH producer cells. I was infected. G418 resistant clones were established in further studies. All retroviral infections were made in the presence of 4 μg / ml polybrene (Sigma). Cells were selected by increasing G-418 to 400 μg / ml for expression of the neomycin-resistance gene.

(Quantitative analysis of tyrosine hydroxylase activity)
Cells confluent on 10 cm plates are washed twice with calcium / magnesium-free Dulberco's phosphate buffer (PBS) and then detached from the plates. Cells were homogenized with 0.15 ml of 50 mM Tris / 50 mM sodium pyrophosphate / 0.2% Triton X-100, ice-cooled and adjusted to pH 8.4 with acetic acid. Thereafter, the mixture was centrifuged at 32000 g for 15 minutes at 2-4 ° C. The supernatant was used for both TH and protein measurements. TH activity was determined as previously described (Iuvone et al., J. Neurochem, 43: 1359-1368 (1986)), 20 C tyrosine labeled with ¹⁴ C, 1 mM 6-methyl-5,6,7, Measurements were made by decarboxylase coupling reaction using 8-tetrahydropterin (6MPH4) (Calbiochem, La Jolla, Calif.) Magnesium phosphate buffer (pH 6). The protein was determined by the method of Lowry (J. Biol. Chem. 193: 265 (1951)) using bovine serum albumin although it was a standard method.

(Quantitative analysis of catecholamines and their metabolites)
Cultured cells were detached from the plate as in quantitative analysis of tyrosine hydroxylase activity, except that ascorbic acid was added to the cell pellet before freezing (to a final concentration of 50 nM). The cells were cultured in a medium in which 10% fetal bovine serum was added to DME. To the culture, 0.1 mM 6MPH4 was added 16 hours before peeling from the plate. Cells were homogenized in 0.1N HCl / 0.1% sodium metabisulfite / 0.2% Na ₂ EDTA containing 250 μl of 5 ng / ml 3,4-dihydroxybenzylamine (DHBA). Dopamine, dopa, 3,4-hydroxyphenylacetic acid (DOPAC), DHBA, 3-methoxy-4-histoxyphenylglycol (MHPG) are extracted from the supernatant of alumina absorption (Anton et al., J. Pharmacol. Exp Ther. 138: 360-375 (1962)), eluted with 150 μl of 0.1 NH ₃ PO ₄ . They are adjusted to contain high concentrations of sodium octyl sulfate (0.45 mM) and low pH (2.8) as shown in Iovone et al., Brain Res. 418: 314-324 (1987) The analysis was performed by HPLC, which is an electrochemical detection method using the prepared mobile phase. Homovanic acid (HVA) was analyzed by HPLC. The concentration of catecholamines and metabolites in the medium was also determined by another method, HPLC-EC. The above procedure for alumina extract is omitted. The medium was adjusted to 0.1 perchloric acid / 0.01 M EDTA to remove precipitated material and centrifuged at 10000 g for 10 minutes so that it could be used directly for HPLC-EC. In this system, the two phases consist of a pH 3.2 0.1 M phosphate buffer (buffer A) containing 0.137% SDS and a pH 3.35 phosphate buffer (buffer B) containing 40% ethanol. It is made up. The sample complex is eluted for 12 minutes in the state of buffer A 100%, and then buffer B is gradually mixed with a linear gradient to elute so that buffer B becomes 100% over 30 minutes. did. The eluate then passed between 16 pairs of coulometric electrodes with a voltage of 60 mV.

(Parkinson's disease rat model)
At the center of the forebrain bundle of female Sprague-Dawley rats (AP = −4.4, ML = 1.1, DV = 7.5) in relation to 12 μl of physiological saline 2 μg of ascorbic acid 6-hydroxydopamine (6-OHDA) was injected unilaterally. Completion of the lesion created was assessed by injection of apomorphine (0.1 mg / kg, subcutaneous injection) or amphetamine (5 mg / kg, subcutaneous injection) that causes rolling behavior after 10-20 days. (Ungerstedt and Arbuthnott, Brain Res. 24: 485-493 (1970)). Prior to transplantation, each animal was tested for at least twice a day to establish a turning behavior that was a criterion for apomorphine or amphetamine. Animals that began to roll more than seven times per minute (Schmidt et al., J. Neurochem. 38: 737-748 (1982)) were used in the study (one minute on the reverse side after administration of apomorphine). And at least 7 turns per minute on the same side after administration of amphetamine, 19 times for apomorphine and 14 times for amphetamine). The change in average number of turns from baseline to post-transplant was compared in four animal groups.

(Transplantation of fibroblasts)
A 10 cm plate confluent with TH-infected fibroblasts and uninfected ones were trypsinized with PBS containing 0.05% trypsin, and then 1 mg / ml glucose, 0.1 mg while suppressing the activity of trypsin / Ml MgCl ₂ , 0.1 mg / ml CaCl ₂ , and 5% rat serum and pipetted with PBS. The cells were washed twice by centrifugation at 1000 g in PBS and suspended in PBS to a density of 80000 in 1 μl. Because the transplantation location seems to be crucial to recover from the rolling asymmetry (Herrera etal., Brain Res. 297: 53-61 (1984); Dunnett et al., Scand. Suppl. 522: 29-37 (1983) ), The foremost beak in the brain where the suspended cells are paralyzed (AP = 1.4, ML = 2.0 in terms of coordinates, DV = 3.5-5.5 to AP = 2.5, ML = 1.5, DV = 3.5 / 4.5 position) and caudate part (AP = 0.4, ML = 1.5, DV = 3.5 / 4. It was injected in a stereotaxic position at positions 2 and 3 away from 5). Each position was divided into 1 to 2 mm and divided into 2 times for a total injection of 4 μl. Unaffected fibroblasts were injected into the control affected animals.

(Post-transplant behavioral test)
The transplanted rats were tested for rolling asymmetry 1-2 weeks after transplantation.

(Histological method)
After the final behavioral test, the rats were anesthetized and sprayed with 10% formalin. The brains were fixed overnight, placed in 30% sucrose for 48 hours and sectioned (40 μm) with a freezing microtome. Each section was stained with a polyclonal anti-tyrosine hydroxylase antibody (Eugenetech, NJ) of cresol violet, fibronectin (FB), and TH. Briefly, sections were rinsed with Tris buffer (TBS) (pH 7.4) containing 0.25% Triton-X. Sections were either rabbit polyclonal antibody of tyrosine hydroxylase diluted 1: 600 in TBS containing 0.25% Triton-X and 3% goat serum at 4 ° C for 24 hours, or diluted 1: 2000 in TBS. Incubated with polyclonal antifibronectin. After rinsing, the sections were incubated for 1 hour with biotinylated goat anti-rabbit IgG (Vectastain) diluted 200-fold in 0.1M TBS containing 0.25% Triton-X and 1% goat serum. Thereafter, it was rinsed several times with 0.1M TBS containing 0.25% Triton-X and 1% goat serum, and the sections were then 0.1M containing 0.25% Triton-X and 1% goat serum. Incubate for 1 hour at room temperature with a complex of avidin and biotinylated horseradish peroxidase (Vectorstein, Vector Laboratory ABC Kit, Burlingham, Calif.) Diluted 100-fold with TBS. Peroxidase was 0.05% 3,3-diaminobenzine tetrahydroxychloride DAB) (obtained from Sigma Chemical in St. Louis, MO), 0.05% NiCl _2, made visible by reacting at room temperature for 15 minutes in TBS containing 0.01% H ₂ O _2. Installation The sections were evaluated for the size and location of active fibroblast transplants.

(Establishment of fibroblast clones that express TH at high levels)
Immortalized rat fibroblasts (208F) were infected by the LThRNL virus made by Ψ2 / TH producer cells and 12 G418 resistant clones were established. [Table 1] shows three TH activities having the highest activity among these 12 G418 resistant clones and the TH activity of the Ψ2 producer cell line.

  The highest activity clones (clones 208F / TH-8 and 208F / TH-11) have TH activity approximately one-fourth that of rat striatal TH activity. The highest activity clone 208F / TH-8 was selected for use in future studies.

(Fibroblasts expressing TH produce and secrete L-dopa.)
Cell extracts of fibroblasts 208F / TH-8 expressing TH and control 208F were used for L-dopa quantification [Table 2]. Only 208F / TH-8 cells cultured in medium supplemented with 6MPH4 produced L-dopa. Control cells did not produce enough to detect L-dopa. Dopamine and its metabolites DOPAC and HVA could not be detected in either 208F / TH-8 cells or 208F control cells.

As shown by Table 2], L-dopa, 208F / TH-8 in the cell culture medium, the control 208F media 63ng / h / 10 ⁶ cells in, 239ng / h / 10 in medium TH-infected ⁶ cells were detected. There was no undetectable level of L-dopa metabolite contained or released in 208F / TH-8 or control fibroblasts.

(Transplant histology)
The transplanted fibroblasts survived in the parenchyma in many locations within the paralyzed caudate. In order to survive the aggressive transplantation of fibronectin, it seems that the graft should be of an appropriate size regardless of the position (FIGS. 18 (a) and (b)). Based on the limited condition of fibronectin staining in syringe tubes, only 4 out of 31 transplants failed to survive (FIGS. 18 (c) and (d)). Behavioral data for rats whose grafts could not survive were excluded from statistical processing. Fibroblast TH immunoreactivity was not observed in vitro or in vivo.

(Transplantation effect on turning asymmetry)
The number of individual animal turnovers due to drug induction was compared before and 2 weeks after transplantation. Since there was no difference between rats tested with apomorphine and rats tested with amphetamine, we decided to treat each turn in the same way.

  The improvement in rolling asymmetry was dependent on the location of implantation. Rats with the fibroblast graft at the caudal site of the caudate organ (the top of FIG. 19) (AP = 0 to 0.4) had little change in the turning behavior. Rats (AP = 1.4-2.2) that had TH infection in the parenchymal parenchyma and had surviving fibroblasts (AP = 1.4-2.2) had an average of 33% reduction in drug-induced turnover two weeks after transplantation (FIG. 19 bottom).

  From these results, it can be seen that the rat cDNA encoding the TH gene exhibited functional TH enzyme activity when stably transformed into fibroblasts. Fibroblasts expressing the TH gene can produce and secrete L-dopa in vitro. When these dopa-producing fibroblasts were transplanted at the beak-like site, the rolling asymmetry of the rat model of Parkinson's disease was significantly reduced. Since the control fibroblasts do not produce much L-dopa in vitro and do not reduce the turning asymmetry of these rats, the ability of these dopa-producing cells to reduce the turning asymmetry of the rat is simply cell It must depend on the presence of the TH gene. These data show the effect of turning behavior for at least 2 weeks.

  The effect of dopa-producing fibroblasts on the turning behavior depends on the position of the beak-like site. Data from previous studies of transplanting fetal nerves into rats have shown that the rotational asymmetry is most effectively reduced when the transplant is performed at the beak site (Dunnett, ibid.). Since the fibroblasts used in the experiment are unable to extend axons, the location of the transplant is highly dependent on proper transplant location.

  The exact mechanism by which dopa-producing fibroblasts reduce turning asymmetry has been determined. Perhaps once dopa is secreted, the dopa decarboxylase activity at the caudate site remains sufficient even in animals with general nerve anesthesia (Lloyd etal., Science 170: 1212-1213 (1970); Hornykiewicz , British Med. J. 29: 172-178 (1973)). As a result, L-dopa is converted to dopamine that modifies drug-induced turning behavior. This hypothetical mechanism for the activity of these dopa-producing cells would be consistent with the highly established effect of L-Dopa's therapeutic system on Parkinson's disease (Calne, N. Eng. J. Med. 310: 523- 524 (1984)). These dopa-producing fibroblasts essentially play the role of a locally small pump of L-dopa.

  As shown in this experimental example, the ability to modify cells to produce L-dopa has evolved into studies looking for ideal cells for Parkinson's disease transplantation therapy. Any cell that is modified to express the TH gene and can survive long in the brain without being transformed into a tumor or causing other illnesses may be used. These specially immortalized rat fibroblasts do not turn into tumors even after 3 months, and primary cells such as primary fibroblasts and primary glial cells have a theoretical tendency to become neoplastic. It is low and in some instances preferred for transplantation. In addition, the patient's own primary cells can be used for autologous tissue transplantation to reduce transplant rejection. However, the use of immortalized cells such as the 208F cells used in this experimental example shows that it is advantageous to prepare a large number of well-known and well-characterized cells. Yes.

  These results indicate that fibroblasts can be genetically modified so that they can provide the functions normally given by nerves, and therefore there is no need to utilize fetal tissue for nerve transplantation Is shown. Combining physical therapy called transplantation and gene transfer is a powerful method for treating CNS dysfunction. The presently invented method may also be used to treat other models of animal and human brain disease.

[Experimental Example IV]
"Transplantation of cultured autologous primary dermal fibroblasts"
As mentioned above, it is desirable in certain circumstances to minimize rejection during transplantation into the cerebrum and to prevent tumor formation using autologous cells such as primary fibroblasts. In this experimental example, primary fibroblasts evaluated this cell population for intracerebral transplantation by assessing the growth of cells in vitro and in vivo, and assessing colonization of transplanted cells into the donor striatum.

  In these experiments, the growth rate of primary dermal fibroblasts in the medium was measured. This growth type was compared to that of rat-1 fibroblasts fixed in vitro in order to determine whether the growth and characteristics of the cells in the medium could predict the following transplantation for them. Variables that were likely to affect primary fibroblast survival were also tested, including the number of passages the cells could receive in the medium prior to transplantation and the density of transplanted cells. The introduction of new genetic material into cultured cells requires several pathways for selection and expansion of altered populations, so the former evaluation is important. The density of transplanted cells was changed in determining the optimal transplant size. Morphological and neurochemical characterization of the graft was also performed, indicating cell interactions between the autofibroblast graft and the host striatum.

  Skin biopsies were collected from the ventral abdominal wall of 23 female Sprague-Dawley rats (200-250 g). Pure fibroblasts from each biopsy were maintained under normal conditions and given DMEM containing 10% fetal calf serum three times a week. Fibroblasts that grew into aggregates were split 1: 2. Primary dermal fibroblast growth rate was determined using a Coulter Counter (Coulter Electronics Inc). Three-part samples of these cells were measured daily and compared to those of fixed rat-1 fibroblasts grown and collected under the same conditions.

For intracerebral transplantation, primary dermal fibroblasts from each animal were inoculated once or four times (approximately 5 cell divisions occurred during the first to fourth passages). . These cells grew into aggregates and were harvested in a quiescent state. The cultured fibroblasts were suspended in phosphate buffered saline for transplantation supplemented with 1 mg / kg MgCl ₂ and CaCl ₂ and 0.1% glucose. The density of the suspension ranged from 10 ⁴ to 10 ⁵ cells / μl. Ketamine-xylazine (10 mg / kg Ketalar, Parke-Davis, Ann Arbor, MI); 5 mg / kg Lampang, Hoechst, Frankfurt, Jar Manny (Rumpun, Hoechst, Frankfurt, Germany); and 6 mg / kg acepromazine maleate, acepromazine maleate, TechAmerica) were anesthetized by intravascular injection and placed in place in the frame. Each animal was implanted with a cultured skin fibroblast graft obtained from its own skin biopsies. All four cell suspensions (3 μl / site) were in a Hamilton syringe and injected into the striatum stereotaxy with two rostral and two caudal sides. The animals survived 3 to 8 weeks after transplantation.

In order to evaluate the parameters for infecting primary fibroblast grafts in primary skin in the cerebrum, two experiments resulted in cell inoculation (n = 12) and cell density (n = 9). Planned to be able to show. In the first experiment, groups of 3 rats were injected with autologous primary skin fibroblasts (10 ⁵ cells / μl) that had been inoculated from 1 to 4 passages. These animals were sacrificed and perfused 3-8 weeks after surgery. In the second experiment, groups of 3 animals were injected with a suspension of 10 ⁴ , 2.5 × 10 ⁴ , 5.0 × 10 ⁴ , or 10 ⁵ cells / μl. The number of passages (4th generation) and survival (8 weeks) were constant.

(Organizational preparation and histology)
To use an optical microscope, animals were perfused with phosphate buffer (pH 7.3) containing 4% paraformaldehyde through the heart. The brain was removed, fixed overnight, and left in 30% sucrose for 3 days. The top of the head through which the striatum passed was cut into a thickness of 40 μm on a frozen microtome and collected in a cryoprotectant. A series of cut pieces was colored with cresyl violet. The remaining fragments were divided into three sets for immunohistochemical detection of fibronectin, glial fibrillary acidic protein (GFAP), and microglia and / or macrophages. This cut piece was first treated in Tris buffer (pH 7.4) containing 0.6% hydrogen peroxide for 30 minutes to stop endogenous peroxidase activity. The slices are washed with buffer and 3% standard goat in rabbit anti-human fibroblast immunoglobulin (IgG) (diluted 1: 2000) or rabbit anti-human GFAPIgG (diluted 1: 1000). Incubate overnight at room temperature in Tris buffer with serum and 0.25% Triton-X. They were then rinsed with buffer and incubated for 1 hour in biotinylated goat anti-rabbit IgG (diluted 1: 250). After further washing, they were cultured for 1 hour in an avidin-biotin complex (Vector Laboratories) and washed. These were treated with 0.025% diaminobenzidine (DAB) tetrahydrochloride, 0.5% nickel chloride and 0.03% hydrogen peroxide in Tris buffer for 5 minutes.

  Immunohistochemical localization of reactive microglia and / or macrophages was performed using the mouse monoclonal antibody MRC OX-42 (Serotec). Slices were washed with buffer and incubated overnight at room temperature in Tris buffer with 3% standard horse serum and 0.25% Triton-X at a concentration of 10 μl / ml. These were washed and cultured in biotinylated horse anti-mouse IgG (1: 160 dilution) for 1 hour. The cut pieces were washed and cultured in avidin-biotin complex, washed again and reacted in DAB solution (as described above). After the immunoperoxidase reaction all the cut pieces were washed with Tris buffer followed by phosphate buffer. These were placed on a slide coated with chrome alum-gelatin, dehydrated with ethanol, and then covered with a cover glass.

  To use an electron microscope, animals (n = 2, obtained from cell density studies) were perfused through the heart with Tris buffer (pH 7.4) containing 4% paraformaldehyde and 0.1% glutaraldehyde. I let you. The brain was removed and fixed for 3 hours, and the parietal portion through which the striatum passed was cut into a thickness of 50 μm on an Oxford vibratome and collected in a Tris buffer. The cut pieces were fixed in 1% osmium tetraoxide immunohistochemically colored with fibronectin or treated with buffer for 1 hour. The latter piece was dehydrated with rinse methanol, clarified in propylene oxide, and fixed in a poly / Bed and Araldite mixture. In order to evaluate the graft, a slightly thinner piece (1-2 μm) was collected and colored with 1% toluidine blue. Very thin cuts were cut on a Reichert Om U3 ultramicrotome, collected on a copper grid and colored with uranyl acetate and lead citrate. The cut piece was observed with a JEOL 100 CXII transmission electron microscope.

(Graft analysis)
Since fibronectin is not normally detected in the rat brain, the fibroblast graft is defined by an area that is immunohistochemically colored with fibronectin. Slices showing immunoreactivity were analyzed using a Cue-2 image analysis system (Olympus Corp). The immunostaining was followed according to the manual (magnification (mag.) = 4 ×) and the area of the graft was calculated for each cut within a given series. Graft volume (mm ³ ) was calculated from this series by Uylings et al. (J. Neurosci. Meth. 18: 18-37 (1986)) and is described in this reference. This document is incorporated herein by reference. The cut piece was colored with cresyl violet, GFAP and MRC OX-42 in the same manner as the sample observed with an electron microscope, and quantitatively evaluated.

(Growth curves for rat-1 and primary fibroblasts)
The growth rate of rat-1 cells, fixed fibroblasts presumed to have undergone contact inhibition, was compared to that of primary fibroblasts cultured under similar conditions. The growth curve for all cell types was similar for the first week, but dramatically different growth patterns were observed after 10 days (FIG. 20). Rat-1 cells seemed not contact-inhibited because they continued to grow once they reached confluence. On the other hand, the growth of primary skin fibroblasts was stable once the cells reached confluence. These modes of sustained growth (rat-1 fibroblasts) versus quiescence (primary skin fibroblasts) continued for 3 weeks.

(Cell passage inoculation)
A parameter that appears to affect the size of primary fibroblasts in the striatum is the number of in vitro passages performed prior to transplantation. In this study, the cell density of the graft is constant and 10 ⁵ per μl. Six animals infused with the fibroblast suspension received one passage in the medium and the other six animals that received the fibroblast inoculation received four passages. I received it. Three animals from each group were killed and perfused for 3-8 weeks after receiving the transplant. 2-Factor ANOVA was used over 3 to 8 weeks after transplantation to assess the difference in volume of the first and fourth passages (FIG. 21). Graft volume determined by fibronectin immunostaining did not differ as a function of the number of passages on the medium performed prior to transplantation (1: 4). It was also clear that cultured autologous fibroblasts transplanted into the brain did not form tumors or decline and die. In fact, the size of the grafts clearly did not differ between those that survived 3 and 8 weeks in vivo.

(Cell density)
Another parameter of graft size measured is the density of cells transplanted into the striatum. Four densities (10 ⁴ , 2.5 × 10 ⁴ , 5.0 × 10 ⁴ or 10 ⁵ per μl) were evaluated and the volume of the graft was determined immunohistochemically using fibronectin. Fibroblasts for transplantation into the cerebrum were harvested after 4 passages in culture and the animals were killed 8 weeks after transplantation. Using 2-factor ANOVA, a clear difference in graft volume was observed between the four groups; high cell density suspension grafts produced larger grafts with lower cell density ( FIG. 22). From this, the relationship between the volume of the graft and the density of the transplanted cells is clear.

(Histology)
Immunostaining with fibronectin, a structurally related glycoprotein family, showed the size and extent of intracerebral transplantation. At 3 weeks, the striatum was somewhat slightly expanded into the neuroropil, while after 8 weeks the graft appeared to shrink more and have fewer protrusions (FIG. 23). However, immunostaining with fibronectin failed to separate individual fibroblasts and other cellular components in and near the graft. In coloring with cresyl violet, long nuclei characteristic of fibroblasts were observed in the graft (FIGS. 24 (a) and (b)). There were also blood vessels passing through the graft, with cells debris-filled preferentially in the center. Astrocytes stained immunohistochemically with GFAP were prominent in the host striatum surrounding the graft at 3 weeks, but immunostaining within the graft was dilute. GFAP-immunoreactive effects were observed both inside and around the 8 week implant (FIGS. 24 (c), (d)). Reactive microglia and / or macrophages visualized by MRC OX-42 immunoreactivity surrounded the graft at 3 weeks, but declined and became more mutated at 8 weeks (Fig. 24 (e), (f)).

(Plasma ultrastructure)
The structural organization of primary fibroblast autografts at 8 weeks after transplantation was investigated at the electron microscope level. The most prominent feature of the graft is the presence of large amounts of fibroblasts and collagen. Fibroblasts had elliptical nuclei that were often aggregated; nucleoli were rarely observed (FIGS. 25 (a), (b)). The cytoplasm of these cells included a rough inner protoplasmic network, elongated mitochondria, and secretory vesicles. A small amount of Golgi was evident. The elongated fibroblast protrusions seen throughout the graft also contained a coarse inner protoplasmic network and secretory vesicles (FIG. 25 (c)). Collagen surrounded the fibroblast body and its protrusions. Neither fibroblasts nor collagen was observed in the striatum nerve network. In the extracellular matrix of the graft, it was observed that enlarged astrocytes expanded between the bundles of collagen; the astrocyte cell body was not found in the fibroblast transplanted tissue. (FIG. 26 (a)). The morphology of the phagocytic cells in the graft is variable; the cytoplasm has few organelles and is mostly filled with small pieces of electrons (FIG. 26 (b)). The nuclei of these cells are irregular and the chromatin often aggregates in the nuclear envelope. Eight weeks after transplantation, few lymphocytes were also observed around and inside the graft.

  Another significant feature of the primary fibroblasts was the presence of continuous capillaries; the inner surface of the endothelium of the tubule was continuous and the fenestrations were not clear (FIG. 27 (a)). There was a basal layer surrounding these tubes. Endothelial cells surrounding the lumen had nuclei with chromatin condensed in the nuclear envelope. Many vesicle invades in both the tubular and non-tubular plasma membranes of endothelial cells were evident, and how vesicles in the cytoplasm were also evident (FIG. 27 (b)). Intercellular junctions were observed at sites in close proximity between endothelial cells (FIG. 27 (c)). Collagen and / or reactive astrocytes were closely associated with the graft capillaries. The nature of these reactive living astrocytes in proximity to the capillaries did not resemble those of amyelinated endfeet astrocytes filled with monofilaments and enclosing normal cerebral ducts (Fig. 27 (d)). Some capillaries also had several reactive astrocyte protrusions that completely enclose the tube; the space between the endothelium and the astrocytes usually had collagen, and its size changed.

  Primary fibroblasts stopped growing once they reached contact and reached confluence. This indicates contact inhibition in vitro. As shown here, after transplantation into the striatum, the volume of primary dermal fibroblast grafts remained constant for 3 to 8 weeks; Of dermal fibroblasts was not characteristic. Characteristics of primary fibroblasts in the medium, such as contact inhibition, may be reflected in graft growth and survival after transplantation into the CNS. Data from this research paper also show that cell suspension density is another factor that determines the size of primary skin fibroblast grafts: dense cell suspensions Produces larger grafts than those of low density. The number of cell passages in the medium performed prior to transplantation and the period after transplantation do not appear to be a factor influencing the size of the graft.

  Another advantage of using primary skin fibroblasts for transplantation is that the potential for autoimmune rejection of autologous cells can be reduced. The immunological response to allografts may be one factor that kills cells after transplantation, especially in 208F fibroblasts. In this example, fibroblasts from skin biopsy were grown under culture conditions and transplanted into the donor striatum. In this experimental example, fibroblasts from skin biopsy were grown under culture conditions and transplanted into donor striatum. After fibroblasts are placed in the striatum, the blood-brain barrier is broken and several cell populations containing lymphocytes enter from the damaged site. Despite the presence of lymphocytes near the graft, the immune response against these autologous cells appeared to be minimal. Thus, the histocompatibility of primary fibroblasts will also promote graft survival in vivo.

  This experimental example provides direct evidence for maintaining survival of grown rat CNS primary dermal fibroblasts. Analysis of protoplasmic ultrastructure indicates that these grafts are initially composed of fibroblasts and collagen in addition to other cell populations (eg, phagocytic cells and lymphocytes). The morphological characteristics of fibroblasts 8 weeks after transplantation into striatal cells are the same as those normally found in the skin, with an elliptical nucleus with aggregated chromatin, elongated mitochondria and a few Has a Golgi apparatus. The presence of internal protoplasmic networks and secretory vesicles within the cytoplasm and fibroblast projections strongly suggests active protein production. Furthermore, the immense staining with fibronectin and the presence of collagen provide evidence that fibroblasts synthesize two different cell products in the graft. Despite the production of collagen, it remains approximately the same size for 3 to 8 weeks after transplantation. One plausible explanation for this observation is that cultured fibroblasts are known to produce collagenase (Millis et al., Exp. Geront. 24: 559-575 (1989) Therefore, fibroblasts may regulate the level of collagen production and suppression in the graft. Optical and electron microscopic observations concluded that primary fibroblasts transplanted into rat striatum maintain their morphological characteristics, produce and secrete collagen and fibronectin, and survive for up to 8 weeks. . Although the total number of viable cells or the proportion of surviving cells was not determined, there is little chance that cells will survive at a high rate in the graft 8 weeks after transplantation. Nutrients such as fibroblast growth factor (FGF) will increase the viability of primary skin fibroblasts transplanted into the rat CNS.

(Host-graft interaction)
An important issue for transplantation of cultured autologous fibroblasts is whether these grafts bind to the host nervous system. After transplanting cultured primary dermal fibroblasts into rat striatum, three prominent cellular events were observed due to the dynamic interaction between the host and the graft. These were transient events that occurred as follows: 1) activation and migration of microglia and / or macrophages to the graft; 2) neovascularization; 3) hypertrophy of astrocytes and Invasion into the graft. Based on data from previous studies of the initial response to the CNS in the penetration of blood-brain barrier wounds, the model system described here is incorporated herein. Mononuclear phagocytes have been shown to be the first cells to respond to wounds in the CNS (Imamoto and Leblond, J. Comp. Neurol. 174: 225-280 (1977)). Macrophages, microglia and granular leukocytes are immunized with MRC OX-42 which recognizes rat homologues of human and mouse complement type 3 (Robinson et al., Immunol. 57: 239-247 (1986)). Will be shown histochemically. The results presented here indicate that the cells that retained OX-42 immunoreactivity were present around the graft 3 to 8 weeks later. An electron microscopic examination of the tissue at 8 weeks ensured the presence of phagocytic cells in the border region between the graft and the intact neural network, and these cells were immunohistochemically stained with OX-42. Would have been. Phagocytic cells are also present throughout the transplanted cells, but OX-42 immunoreactivity was not evident within the graft itself. Non-immunoreactive phagocytes therefore constitute other cell populations or granular leukocytes lacking macrophages, microglia and complement. The cells move to the damaged site and enter the graft.

  An important host-graft cell interaction that determines the fate of fibroblasts after transplantation is the construction of the vascular system into the graft. The vascular system builds into the graft and relies on angiogenic substances, such as parts, fibroblast growth factor. Production and release of fibroblast growth factor by transplanted dermal fibroblasts promotes angiogenesis. Neovascularization in the graft is also promoted by reactive microglia and macrophages that produce and release interleukin-1.

  Transplantation of vascular tissue into the CNS depends entirely on the surrounding host tissue for the generation of new capillaries. The capillaries inside the dermal fibroblast graft consist of windowless endothelial cells that form a series of ducts and retain intercellular junctions (not necessarily strong junctions) where the cells are located in parallel. These tubes imply tubes found in both connective and neural tissue; however, several features distinguish the graft capillaries from those of the adjacent striatum. First, the endothelial cells in the graft have many intracellular vesicles and vesicle intrusions into the plasma membrane, whereas in neurocapillaries, there are few vesicle invades on the tubal and antitubular sides Was recognized. Secondly, the reactive astrocytes associated with the graft tubes do not resemble the astrocytes of the terminal spheres of unmyelinated nerves surrounding the nerve capillaries, where the former is a streak It is not always covered and enlarged and always wraps the capillaries of the graft. Third, the space near the tube between the endothelium and the side of the reactive astrocytes of the graft capillary is variable and often there are intervening elements (including collagen). The neural tube has a space close to the narrow canal including the basal layer.

  The third interaction observed between the striatum neural network and primary fibroblasts was astrocyte responsiveness to autografts. Intracerebral transplantation usually causes a glial reaction that involves the migration of reactive astrocyte processes and / or the formation of glue limitans. In these studies, GFAP-immunoreactive astrocytes perkaraya were abundant around the 3 and 8 week grafts, and immunoreactive processes were found inside the 8 week grafts. Protoplasmic ultrastructural data confirms the latter observation and indicates that reactive astrocyte protrusions are present throughout the graft, but the cell body remains outside the graft itself. . The astrocyte response observed in this and other studies is that of reactive microglia and macrophages that are located near some partially injured sites (these cells produce and release interleukin-1 in vitro) It will be because of the existence).

  These results provide direct evidence of the survival of cultured fibroblasts transplanted into the grown rat CNS. These grafts did not form tumors in the striatum and the possible immunological response appeared minimal due to the self-organizing properties of these cells. Furthermore, the fibroblast grafts were structurally and functionally bound to the host brain. Dynamic interaction between the host striatum and fibroblasts showing structural and functional binding of fibroblasts to grown rat CNS cells is responsible for the construction of the tubule system in the graft, phagocytic cell migration And invasion and hypertrophy of reactive astrocytes into the graft. In summary, these features are that primary fibroblasts are suitable donor cell candidates for transplantation into the cerebrum, and therefore cells that have undergone genetic modification for the treatment of diseased or damaged CNS. It will be used as

[Experiment V]
"Transplantation of L-dopa-producing primary fibroblasts"
This experimental example confirms that genetically modified primary fibroblasts survive for a long time after being transplanted into the brain. Developing the above experiments, the rat TH gene was inserted into primary fibroblasts obtained by Fisher rat skin biopsy. The viability of TH-containing primary fibroblasts was examined over a two-month period using 6-OHDA lesions before the nigrostriatal pathway after transplantation into Fisher rat brain. In addition, the presence of TH-transgene and transgene product in the transplanted fibroblasts was examined. Finally, the in vivo extended function of genetically modified fibroblasts was studied by evaluating the turning behavior of transplanted rats every two weeks for eight weeks after transplantation.

(Retroviral vector preparation)
The vector pLThRNL used in this experiment was prepared as in Experimental Example III (see FIGS. 16 and 17).

(Preparation of fibroblasts)
Primary fibroblasts are inbred by the method of Sly and Grubb (Methods in Enzymology, Vol. LVIII, Colowick and Kaplan, Eds., Academic Press, New York, pp. 444-450 (1979)). Obtained from abdominal skin biopsy of Fisher 344 rats. Briefly, rats were anesthetized with a mixture of ketamine, Aceptomazine, and Rompun. The abdomen was shaved and appeared with alcohol. Approximately 1-2 square centimeters of skin was removed, rinsed briefly with alcohol, and immediately placed in a vial containing sterile DMEM / S. The biopsy was usually transferred to a cover glass within the second period after collection and cultured in DMEM / S for 21 days. Cells are infected with ecotropic LThRNL or BAG virus (Price et al., Proc. Natl. Acad. Sci. USA 84: 156-160 (1987)) at a multiplicity of infection of 10 in the presence of polybrene (4 μg / ml). It was. G-418 resistant cells were grown to confluence and the expression of TH activity was tested by decarboxylase binding assay (Iuvone, J. Neurochem. 43: 1359-1368 (1984)). A subclone (FF2 / TH) showing TH activity and TH immunoreactivity in vitro was selected for transplantation. It was used as a fibroblast control cell (FF2 / βGal) infected with BAG virus and expressing β-galactosidase activity (β-Gal) in vitro.

(In vitro biochemical analysis)
Production of L-dopa and its metabolites and secretion from infected cells was examined after 24 hours of growth in DMEM / S supplemented with 100 μm tetetrahydrobiopterin, an active cofactor of TH. Conditioned media and cells are collected, adjusted to 0.1 M perchloric acid (PCA) and 0.05 M EDTA, centrifuged at 10,000 × g for 15 minutes at 4 ° C. to remove the precipitate. It was. As for the sample, the presence of catecholamine and catecholamine metabolite was determined by using 16 electrochemical sensors (Matson et al., Clin. Chem. 30: 1477-1488 (1984); Langlais et al , Monitoring Neurotransmitter Release During Behavior, Fillenz, et al., Eds., Horwood, Chichester, UK, pp. 224-232 (1986); Matson et al., Life Sci. 41: 905-908 (1987)) Analysis was performed by injection into a coulometric electrode array equipped with a gradient liquid chromatography (CEAS model 55-0650, ESA, Bedfore, MA).

(Lesions)
The dopaminergic nigrostriatal pathway was unilaterally damaged by neurotoxin 6-hydroxydopamine (6-OHDA). 6-OHDA was dissolved in saline at a concentration of 6 μg / μl and 0.1% ascorbine sun was added to retard oxidation. Female Fischer 344 rats (130-160 g) were anesthetized as described above and placed in a tactile frame. 2 μl of 6-OHDA to the left forebrain center (AP = −4.4 mm, ML = 1.1 mm, DV = 7.5 mm) at 1 μl / min Paxinos and Watson's The Rat Brainin Stereotaxic Coordinates , San Diego, Academic Press, (1986). It was lifted 2 mm after injection and rested for 2 minutes to diffuse neurotoxin. After a 10-14 day recovery period, the damaged rats were scaled for damage using apomorphine.

  In order to measure the contribution of non-specific striatal damage to changes in turning behavior, the method of Experimental Example III was used using 6-OHDA-damaged rats that had kainic acid (KA) damage on only one side of the striatum. Was done. For this experiment female Sprague-Dawley rats were anesthetized and damaged using 6-OHDA as described above. Rats that showed 7 or more turnovers per minute against apomorphine (0.05 mg / kg) were selected for KA injury. Kainic acid was dissolved in saline at a concentration of 5 μg / μl and injected into the anesthetized rat striatum (AP = 0.3 mm, ML = 2 mm, DV = 4 mm) at various doses. Three rats were injected with 1.0 μg, another 3 with 2.0 μg, 3 with 4.0 μg, and 2 with 5.0 μg. KA was added at 0.1 μl / min, then the syringe was slowly lifted 2 mm and left in place for 2 minutes to diffuse the neurotoxin. Diazepam (2 mg / kg) was given immediately after injection, and seizures that could occur with KA were suppressed.

(Behavior test)
The sensitivity of the posterior nerve dopamine receptors in the striated and anesthetized striatum was evaluated as in Example III, 7-10 days after 6-OHDA injury by administration of apomorphine (0.1 mg / kg). . Rats were placed in an automatic rotometer ball and the total number of turns was counted every 10 minutes for 60 minutes. After two apomorphine tests, rats with a minimum of 7 revolutions per minute were selected for the next test. Rats were tested at least three times with apomorphine or until no change in turn was observed since the previous measurement (Bevan, Neurosci. Letter 35: 185-189 (1983); Castro et al. , Psychopharm. 85: 333-339 (1985)). Ten days passed after each apomorphine test. Tested with apomorphine (0.1 mg / kg) every 2 weeks for 8 weeks after transplantation. Rats with KA damage were evaluated using apomorphine (0.05 mg / kg) at 2, 4, and 6 weeks after being damaged. For statistical comparison, the net number of revolutions per minute for 60 minutes was determined after transplantation and compared to pre-transplant values using the least squares method.

(Transplant)
Fibroblasts were prepared for transplantation as in Example II above. Briefly, cells were removed from tissue culture plates with 0.05% trypsin and 1 mg EDTA in Dulbecco's PBS and suspended in PBS with MgCl2, CaCl2, 0.1% glucose (complete PBS) and 5% rat serum. It was cloudy. Cells were collected by centrifugation, washed twice with complete PBS, and resuspended at a concentration of 100,000 cells / μl in complete PBS.

  Fifteen anesthetized rats administered a total of 10 μl FF2 / TH (N = 10) or FF2 / βGal (N = 5) had fibroblasts in the striatum (AP = 0.3 mm, ML = 2. 0 mm, DV = 4.5 mm, AP = 1.5 mm, ML = 2.0, DV = 4.5 mm). At each location 2.5 μl of cells were dispensed, the syringe was lifted 1 mm, and another 2.5 μl of cells was pushed out of the cannula tube. Cells were added at a rate of 1 μl per minute. After the injection to the backmost side, the syringe was left in the position for 2 minutes to allow the cells to diffuse. The other 5 rats that did not receive the transplant were used as controls.

(Dissection method)
For in vitro immunohistochemical analysis, infected fibroblasts were fixed on slides with 4% paraformaldehyde, permeabilized with 0.2% Triton X-100, and against TH (Boehringer Mannheim Biochemicals). Incubated overnight with monoclonal antibody. Staining was performed by avidin-biotin method (Vector Laboratories, Elite Kit) using nickel-enhanced 3,3′-diethylaminobenzidine (DAB) as chromagen.

  For in vivo analysis, transplanted fibroblasts were examined for TH- and fibronectin immunoreactivity 10 weeks after transplantation. Rats were deeply anesthetized, rinsed briefly with saline and then perfused with 4% paraformaldehyde. The brain was removed and placed in the same fixative for one night and transferred to 30% sucrose to reach equilibrium. The frozen part was cut on slide microsomes (40 μm), collected in cryoprotectant and stored for processing. Free-floating sites were incubated with antibodies against one night fibroblast marker (Organon Technica) or monoclonal antibodies against TH. Fibronectin staining was performed using avidin-biotin followed by nickel enriched DAB. The TH-labeled portion was incubated with a secondary antibody conjugated with rhodamine for fluorescent microscopy (Jackson ImmunoResearch Labs., Inc., Bar Harbor, Maine) against mouse antigen. Some parts were processed using OX-42 (Serotec) antibody or ED-1 (Chemicon) antibody to identify microglia or monocytes / macrophages. Portions were incubated with OX-42 antibody for 48 hours or with ED-1 antibody for 24 hours at 4 ° C. At a later date OX-42 was developed using avidin-biotin and nickel-enhanced DAB, and the ED-1 labeled portion was incubated with a rhodamine labeled secondary antibody as described above.

(Hybridization in situ)
In situ hybridization to TH mRNA was performed using a probe labeled with [ ³⁵ S]. The method is described in Higgins et al., Proc. Natl. Acad. Sci. (USA) 85: 1297-1301 (1988), the description of which is incorporated herein by reference. The 63 base pair template of the TH probe contains a 311 nucleotide cDNA complementary to rat TH mRNA (provided by Dr. Dona Chikaraishi, Tufts Medical School, Boston, MA). Plasmid pSP65 was generated by cutting with AvaII and transcribed with SP6 polymerase. The part mounted on the slide was immobilized with 4% paraformaldehyde for 1 minute, cleaved with proteinase K for 5 minutes, and then triethanolamine (0.1 M 0.9% NaCl and (0.1% acetic anhydride) at room temperature for 10 minutes. The portion was subsequently dehydrated with alcohol and then treated with chloroform for 5 minutes at room temperature, returned to 100% ethanol for 2 minutes, and then treated with 95% ethanol for another 2 minutes, Air drying was done. The dehydrated portion was 50% formamide, 0.75 M NaCl, 20 mM Pipes (pH 6.8), 10 mM EDTA, 250 mM DTT, 5X Denhardt's solution (0.02% BSA, 0. In a buffer containing 02% Ficoll, 0.02% polyvinylpyrrolidone), 0.2% SDS, 10% dextran sulfate, and 500 μg / ml denatured yeast tRNA. Pre-hybridization for 3 hours. After the prehybridization, excess buffer was removed and the portion was subjected to 10-15 ng labeled probe contained in hybridization buffer with a total volume of 75 μl for 18 hours at 55 ° C. Hybridization. The slides were rinsed and hybridized in a solution containing 0.6 M NaCl, 0.06 M Na citrate with or without RNase. After a subsequent hybridization rinse, the part was air-dried and irradiated with X-ray film (Kodak XAR) for 24 hours. The part was then immersed in Kodak NTB-2 emulsion and left to stand at 4 ° C. for 4 days for growth.

  When large amounts of TH-expressing fibroblasts were stained for TH-immunoreactivity, differences in TH-label strength were observed. In an effort to isolate clones that stably express high levels of TH, TH-expressing cells were subcloned. The selected clone, FF / TH, showed relatively uniform TH-immunity activity (see FIG. 28). This clone has an apparent morphological morphology similar to uninfected fibroblasts or fibroblasts infected with vectors carrying βgal. Cells were usually extended in a multifaceted process spreading from the body. Cells expressing βGal did not show TH-immunity activity in vitro.

(In vitro analysis)
Primary fibroblasts injected with a vector carrying a gene for TH or βGal (FF2 / βGal) were tested for TH activity using a decarboxylase-linked assay. TH activity was found only in FF2 / TH cells, and the level was 1.1 ± 0.4 pmole L-dopa / min / mg protein. This level of TH in the primary fibroblasts of FF2 / TH was compared to the level observed previously for immunized 208F fibroblasts infected with the same LThRNL vector (see Example III above) .

Infected fibroblasts were analyzed for production and release of L-dopa and other catecholamines after 24 hours of contact with 100 μM BH ₄ , a natural cofactor for TH. The medium collected from cells containing the TH gene was found to contain 17.2 μg L-dopa per 10 ⁶ cells. As assessed through analysis of the FF2 / TH cell pellet, the intracellular quantification of L-dopa was 0.2 μg L-dopa per 10 ⁶ cells. Control FF2 / βGal cells contained no or no L-dopa. Detectable metabolites of L-dopa, such as metabolites such as dopamine, 3,4-dihydroxyphenylacetic acid, or homovanic acid, are also available in the medium of either FF2 / TH or FF2 / βGal cells. Or in any sample collected from either cell pellet.

(Behavioral analysis of rats implanted with fibroblasts)
All 6-OHDA-regulated rats transplanted with FF2 / TH cells or FF2 / βGal cells normally eat and drink within 24 hours after transplantation, and 2 months after transplantation. During the test period, the body weight was found to be unchanged or slightly increased. There was no obvious change in the spontaneous behavior of the transplanted rats, and the behavior of the personal and motor skills were compared with those appearing in the non-transplanted control rats.

  Behavioral tests were performed to assess L-dopa production and release from cells transplanted in vivo. Rats received apomorphine. Apomorphine is a drug that causes activation of the striatal receptors in a dopamine-depleted state. This drug-induced excitement leads the rat to the behavior of walking in a quantifiable circuit or turn pattern (Ungerstedt and Arbuthnott, Brain Res. 24: 485-493 (1970)). After transplantation, the average number of turns per minute for 60 minutes after administration of apomorphine was measured for transplanted rats and control non-transplanted rats, and statistical analysis was performed. Compared to the previous number of turns. In control animals, there was a slight level increase in the number of turns after transplantation (see control in FIG. 29), compared to the number of turn levels observed before transplantation (p> 0.05). There was no statistical difference. The number of turns induced by apomorphine caused by rats transplanted with cells containing βGal is also compared to the number of turns before transplantation (p> 0.05; FIG. 29, FF2 / βGal) throughout the 2 month period after transplantation. There was no difference between them. In contrast, in FF2 / TH rats, there was a significant decrease in the rolling exercise 2 weeks after transplantation, which was observed throughout 8 weeks after transplantation (see FIG. 29, FF2 / TH). In the turn of FF2 / TH rats, there was an initial 65% reduction, ie, 11.7 ± 3.0 to 4.1 ± 2.9 (mean ± mean of standard error), 6 and 8 after transplantation. During the week, the plateau was reduced to almost 30% below the pre-transplant level.

(Anatomical analysis)
Fibroblast grafts were present in the brains of all rats 10 weeks after transplantation. However, the grafts disappeared from the two animals during that period. In general, the placement of the FF2 / TH and FF2 / βGal cell grafts was found to be comparable to both the relatively rostral-caudal site and the central site. . Typically, however, FF2 / TH grafts were larger than FF2 / βGal grafts. Among the FF2 / TH groups, there was no correlation between graft size and reduced turning behavior.

  All grafts contained elongated fibroblast-like cells interspersed in large amounts of collagen fibers. In addition to typical fibroblasts, large, yellow cells were found in the core of the graft, apparently filled with hemasiderin. Because these cells often contain debris, they pass through the graft using OX-42 and ED-1, markers that are specifically stained for microglia and monosite / macrophages. The area to be stained is stained. However, although immunoreactive macrophages were found around the graft, the core portion of the cells filled with a large amount of debris was not labeled with OX-42 or ED-1.

  To assess whether the transgene continues to be expressed in the transplanted fibroblasts, the region that passed through the graft was processed for in situ hybridization to TH mRNA. Only positive hybridization to TH mRNA was found to be present in cells in grafts containing FF2 / TH fibroblasts (see FIG. 30 (a)). In grafts of FF2 / TH cells and FF2 / βGal cells (see FIG. 30 (b), control), the graft region pre-occupied with collagen fibers is significantly more grain-free than the surrounding background. The density was found to be small. There were various morphological cells showing increased grain density, and these cells exhibited many morphological features of elongated fibroblasts (FIG. 31 (c)). However, some cells with aggressive TH hybridization were observed as large and round and were sometimes filled with hemaciderin and granular microparticles. The presence of cells was not observed in βGal grafts that had been specifically hybridized to TH mRNA (see FIG. 31 (d)).

  Experiments were performed on TH-immunoreactivity and βGal histochemistry on the grafts to see if the transgene product was synthesized and present in fibroblasts. In the FF2 / TH graft, it was observed that cells that were TH-labeled with elongated fibroblasts were scattered in the collagen fibers (see FIG. 32 (a)). Autofluorescent material was present in the graft, but rhodamine-specific labeling (emission at 582 nm) due to the shape of the material and broad fluorescence through 528 nm (FIGS. 32 (a) and (b), asterisk). ) Was easily identifiable. As observed in vitro, there was heterogeneity in the intensity of TH labeling. Autofluorescent cells were present in the FF2 / βGal graft, but there were no distinguishable TH-immunoreactive fibroblasts. The portion stained for βGal histochemistry revealed diffuse cytoplasmic labeling of cells within the FF2 / βGal graft. FF2 / βGal staining was observed in the large round cells within the core of the graft and in the elongated cells typical of fibroblasts (FIG. 31 (b)). ΒGal reactivity within FF2 / TH grafts was limited to a small number of round cells with the blue granular staining previously described for endogenous macrophages (Shimohara et al., Mol. Brain Res. 5: 271- 278 (1989)).

  These results confirm that primary fibroblasts obtained from skin biopsies can be grown in culture and genetically manipulated to express the transgene. Evidence that the inserted TH gene production was synthesized and biologically activated in vitro was provided by histological and biochemical measurements. Cells containing the TH gene showed positive immunoreactivity to TH in vitro. And in the presence of pterin cofactor for TH, the presence of tyrosine in the cell culture medium was converted to L-dopa through TH enzyme activity.

  The primary advantage of primary fibroblasts for genetic manipulation is that these modified cells are used as donor material for autotransplantation and thus minimize the problem of immunological response from the host. It is a possibility. The primary fibroblasts used in this study were not transplanted into the same donor animal, but rather genetically inbred Fischer 344 rats were degenerated as neuroblast donors. Used as both. This allograft approach was chosen because it simplifies the maintenance of cells in vitro while giving them closer conditions of autotransplantation. Following implantation of the 6-OHDA region into the rat brain, primary fibroblasts containing genes for βGal or TH were found to survive for at least 10 weeks. Furthermore, the transplanted fibroblasts continued to express the βGal and TH transgenes during the post-transplant life every 2 months. Evidence that TH-containing cells continue to produce and secrete L-dopa in vivo is the observational observation regarding behavior, that is, a considerable decrease in the asymmetry of rotation was observed only for rats with TH-containing fibroblast grafts. It was provided from that.

(Fibroblast survival in the brain)
After an experimental period of 2 months, the FF2 / TH graft was observed to be larger than the FF2 / βGal graft. The reason for this difference is unknown. There are many factors that keep primary fibroblasts alive in the brain, including the state of the cells prior to planting and the time that the cells are maintained in the medium prior to transplantation. FF2 / TH used in the current study is a subclone of TH-expressing fibroblasts, while FF2 / βGal cells were a large population of βGal-expressing fibroblasts. Subcloning methods that synthesize FF2 / TH cells via dimerization further than large amounts of FF2 / βGal cells may affect the ability of fibroblasts to survive in vivo. In early passages, no correlation was observed between the size of the fibroblast grafts and the number of cell divisions in the culture prior to transplantation as shown in Experimental Example IV. Fibroblasts that have been maintained in culture for longer periods of time, for example, are FF2 cells that are used in today's studies, and such fibroblasts are the size of the graft after 4-6 weeks. Or, there was no substantial change in the composition, and it was well alive in the brain for 6 months.

  Other factors that affect fibroblasts surviving in the brain include the availability of transplanted cells to provide the appropriate nutrients. Vascular generation was evident in the FF2 / TH graft, but less so in the smaller FF2 / βGal graft. The time course for establishing a vascular supply to the graft also affects the long term survival of fibroblasts. Grafts containing other cell types, eg cell types such as fetal nerve cells, have been found to be subject to rapid vascularization for survival following transplantation (Stenevi et al., Brain Res. 114: 1-20 (1976)). The blood vessels in the graft have established a barrier to macromolecules like typical brain blood vessels, or leakage remains as seen in the original tissue around the typical graft (Rosenstein and Brightman, J. Comp. Neurol 250: 339-351 (1986)) also affects the survival of transplanted fibroblasts. Incision of blood vessels in the brain after transplantation suggests opening a “window” in the brain against factors that are typically useless for cells in the brain (Rosenstein and Bridhtman, ibid. Literature; see Sandberg et al., Exp. Neurol. 102: 149-152 (1988)). Such factors are maintained in a serum-enriched culture medium (see Gibbs et al., Brain Res. 382: 409-415 (1986); Ezerman, Brain Res. 469: 253-261 (1988)). May be suitable for the cells in question. On the contrary, the entrance of the blood vessel is attacked by circulating macrophages by arranging the transplanted cells at a vulnerable position.

(Construction of graft)
Transplanted cells are known to survive for 2 months, but all grafts are characterized by a large, hemasiderin and cell core containing debris, which is the first Has been suggested to have been attacked by macrophages. However, due to the limited number of ED-1-positive macrophages, it was observed that they were scattered around some grafts, but the cells in the core part of the graft were Alternatively, no immune reaction occurred with either OX-42, ie, either ED-1 or OX-42, which are specific markers for microglia and mononuclear cells / macrophages. Furthermore, when the graft was stained with βGal, it was found that the enzyme previously possessed the appearance of distinct granules in foreign macrophages (Shimohara et al., Ibid., 1989), FF2 / None of the large core cells showed granule staining in either TH or FF2 / βGal grafts. However, some core cells within the FF2 / βGal graft clearly show diffuse and cytoplasmic βGal, respectively, and by pre-labeling are characteristic of immunized fibroblasts containing the βGal transgene. It has been shown (Shimohara et al., Ibid., 1989). Finally, large core cells within the FF2 / TH grafts have been shown to hybridize positively to TH mRNA. These observations strongly suggested that phagocytic cells within the core portion of the fibroblast graft were derived from the original donor cell suspension.

(Mechanism of graft function)
The potential for fibroblast implantation to produce post-synaptic dopamine receptor-related deficiencies and artificial “recovery” in behavior due to loss of striatal tissue. However, there are two pieces of evidence suggesting that this is not promising. First, FF2 / βGal rats transplanted with the same volume and the same number of cells as FF2 / TH rats did not show that the turning behavior after transplantation changed significantly. Second, destruction of endogenous nerves and receptors with kainic acid caused a post-injury turning behavior that was clearly different from the turning behavior observed in FF2 / TH rats. Rats with minimal KA region, or with a region that affects a much smaller area than the area surrounded by the transplant, range from 19% at 2 weeks after injury to 50% at 6 weeks after injury. This shows a definite reduction in turnover, consistent with the degenerative changes known to occur in the striatum after KA administration (Coyleand Schwarcz, Nature 263: 244-246 ( 1976)). In contrast, the apparent decrease in turning behavior observed for FF2 / TH occurs 2 weeks after transplantation (65%), followed by a gradual increase and stabilization, but for 6-8 weeks Then it is higher and the number of turns gradually increases (30% below baseline). The apparent reduction in transplant efficacy between 2 and 8 weeks may result from several different mechanisms. First, some transplanted cells will disappear within the first week after transplantation. In other studies, fibroblast grafts were found to shrink in size in the brain 3 weeks after implantation. After this time, the stable graft size persisted for at least 6 months. If the period after transplantation is long, the second reason for reducing the effect of transplantation is related to changes in the properties of fibroblast metabolism. The constructed fibroblasts will have the properties of stationary cells with reduced metabolic activity (Dean et al., J. Biol. Chem. 261: 9161-9166 (1986); Rao and Church Exp. Cell Res. 178: 449-456 (1988)). The fibroblasts will also have the property of reducing L-dopa production. Finally, the effect of the graft is due to genetic events such as deletion or rearrangement of TH cDNA inserted into fibroblasts, or epigenetic termination of transcription from the LTR promoter that promotes transcription of TH cDNA. Is interpreted concessively. Such facts have also been recorded for other retroviral vectors (Xu et al., Ibid., 1989).

  The last possibility to explain the reduced apomorphine turnover in FF2 / TH-transplanted rats is that the graft will disrupt cortical-striatal interactions through non-specific damage to cortical tissue is there. Such damage has been reported to reduce the turning behavior of 6-OHDA-lesioned rats (see Freed and Canon-Spoor, Behav. Brain Res. 32: 279-288 (1989)). ). However, in our previous study using immortalized rat fibroblasts that showed uncontrolled growth in the brain, extensive damage limited to cortical tissue was observed in apomorphine-induced turning behavior. It was not related to drop. The grafts tested in the current study show that the expansion of transplantation into cortical tissue is much more limited than that seen in rat-1 grafts, thus reducing the turning behavior of FF2 / TH rats However, it does not appear to reflect non-specific damage to the cortical-striatal pathway.

  Currently, L-dopa is the most effective treatment for Parkinson's disease (PD) symptoms. However, long-term planned administration of L-dopa causes undesirable side effects. Side effects include voluntary movement disorders and “on-off” phenomena (see Marsden amd Parks, Lancet 7: 292-296 (1976)). Some of these problems have been suggested to arise from fluctuations in cytoplasmic L-dopa levels following oral administration of L-dopa, which can be alleviated by continuous infusion of L-dopa ( Nutt et al., New Eng. J. Med. 310: 483-488 (1984)). Transplanting cells genetically modified to produce L-dopa or dopamine into the brain provides an alternative method for controlling the symptoms of PD, which is It also provides an alternative method for reducing the side effects of extensive administration of L-dopa by continuous and local infusion of dopamine directly into the target area.

  These results show that primary fibroblasts genetically modified to express the transgene provide viable donor cells for intracerebral transplantation and human neurological such as PD. It shows that it can provide an effective strategy for treating diseases.

[Experimental Example VI]
“Ability of senescent fibroblasts to be supplied as donor cells”
This experimental example demonstrates whether senile fibroblasts are effective as donor cells for intracortical transplantation.

(Retrovirus construction for human nerve growth factor (NGF))
Moloney murine leukemia virus, pLhNRNL, containing human NGF cDNA (obtained from Syntex Research, Palo Alto, Calif.) Was used. This vector was prepared by inserting NGF cDNA into the plasmid pLChRNL at the location of the dChAT gene (see Experimental Example VII, FIG. 34). The 5 ′ long terminal repeat (LTR) promoted expression of human NGF (hNGF) cDNA. The internal Rous sarcoma virus LTR promoter promoted the expression of a selectable bacterial neomycin resistant gene (neo ^R ).

(Production of virus stock)
Production of infectious virus stock is achieved by first transferring the amphotropic cell line PA317 with plasmid DNA by calcium phosphate coprecipitation, which is Miller and Buttimore, Mol. Cell. Biol. 6: 2895. -2902 (1986) (supplied by Miller, Fred Hutchinson Cancer Research Center, Seattle, WA). Conditioned media from these cells was collected, filtered, and used to infect the ecotropic helper cell line psi (Ψ) 2. This is described by Mann et al. In Cell 33: 153-159 (1983). Conditioned media from these cells was then used to infect amphotropic helper cells PA317. Producer cells from this line are selected for their expression of the neo ^R gene by growing cells in media containing 400 μg / ml of the neomycin analog G418 (Gibco / BRL, Gaithersburg, MD). It was done. Virus from the PA317 clone producing the highest titer (about 4 × 10 ⁴ cfu / ml) was used to infect various cell lines as described below.

(Cell infection)
Primary fibroblasts obtained from skin biopsies and defined as cell lines were infected with amphotropic retroviruses. Primary fibroblasts of young rats (about 3 months) were infected with ecotropic virus expressing mouse NGF or amphotropic virus expressing human NGF. Skin primary fibroblasts from the elderly (66-year-old male with Alzheimer's disease and 69-year-old male without Alzheimer's disease) were obtained from skin biopsy by Dr. R. Katzman, University of California, San Diego, CA. Young (newborn) human skin fibroblasts were obtained from the Core Tissue Culture Facility at the University of Cakifornia, San Diego, CA. Rhesus monkey skin fibroblasts (10-year-old male and 11-year-old female) were obtained by biopsy from the Primate Center of the University of California, Davis, CA.

  Fibroblasts that grew from 70% to 80% confluent were cultured overnight with 10 ml of medium containing retrovirus and 4 μg / ml polybrene. This infection was achieved in 3 consecutive days. As a result, the medium was aspirated and replaced with fresh medium containing 400 μg / ml of G418. Following infection, cells were maintained intermittently with G418 supplemented with media.

(NGF activity assay)
NGF production and secretion by infected fibroblasts was assayed by two-site ELISA analysis using anti-βNGF antibody (Boehringer-Mannheim, Germany). This is described in Experimental Example II above. NGF activity was measured in the medium on the day shown in FIG. FIG. 33 shows that the amount of NGF was secreted at different confluences from fibroblasts of different origins expressing either human NGF (hNGF) or mouse NGF (rat fibroblasts). In all cases, NGF production increases between confluences, or at least remains constant (relative to monkey cells). In the logarithmic or confluent period, there was almost no difference in NGF production depending on the youth and aging of human fibroblasts.

  These results indicate that non-human primates (eg, monkeys) with human disease can be used to test the efficacy of genetically modified transgene transplantation. In addition, these results indicate that senile fibroblasts are successfully infected with vectors carrying the transgene, and suggest that senile fibroblasts can be used as donor cells. .

[Experimental Example VII]
"Regulation of the release of transgene products arising from donor cells"
This example demonstrates the regulation of secretion of the transgene product acetylcholine (ACh). Acetylcholine can be obtained from cells transfected with choline.

(Cell culture)
Rat-1 fibroblasts were grown in DMEM suspended in 0.1% glutamine and 10% FBS. The cultures were incubated under conditions of 10% CO2, 37 ° C.

(Construction of retrovirus)
A Moloney murine leukemia viral vector (MoMLV) (obtained from Dr. P. Salvaterra, Duarte, CA), pLChRNL, shown in FIG. 34. Built to be. HindIII consisting of 2519 base pairs (bp) containing the dChAT cDNA obtained from the plasmid pCha-2 (Itoh et al., Proc. Natl. Acad. Sci. USA 83: 4081-4085 (1986)) An EcoRI / BamHI fragment consisting of 377 bp from pBR322 and 6500 bp which is a retroviral construct pLLRNL (Xu et al., Virology 171: 331-341 (1989)) The resulting fragment was incorporated by known methods (see Sambrook et al., In: Molecular Cloning, A Laboratory Manual, ed. C. Nolan, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY (1989)). The 5 ′ LTR in pLLRNL advances the expression of the dChAT gene. Similarly, the internal RSV promoter drives the expression of a specific bacterial neomycin resistance gene (neo ^R ).

(Infection)
Production of a vector strain capable of infecting is carried out by using an amphotropic helper cell line PA317 (Miller and Buttimore, Mol. Cell. Biol. 6: 2895-2902 (1986)) using plasmid DNA, Done by transfecting. The medium conditioned by these cells is collected and applied with a phyllator before the ecotropic helper cell line Ψ-2 (Mann et al., Cell 33: 153-159 (1983)). Used to infect. The production cells forming this line were selected for neo ^R expression by growing the cells in a culture medium containing 400 μg / ml neomycin analog G418 (Gibco). 2 G418-resistant colonies selected from a total of 16 colonies, with highest levels of ChAT activity and highest titer against virus (approximately 10 ⁴ cfu / ml) ) Were used to infect rat-1 cells.

  Rat-1 fibroblasts grown to confluence on 10 cm tissue culture plates were split 1: 5 and allowed to grow for 1 day before several retroviruses and 4 μg / ml polybrene. Incubate in 5 ml of medium containing Cells were allowed to incubate overnight in this medium. The medium was then aspirated and replanted into fresh medium containing 400 μg / ml G418. After infection, the cells were continuously maintained in culture medium supplemented with G418.

(ChAT activity assay)
The method used to perform the assay for ChAT activity is a modification of the method described by Fonnum in the following document J. Neurochem. 24: 407-409 (1975). This document is incorporated herein as a reference document. Fibroblasts were seeded in 35 mm tissue culture dishes and cultured to 70-80% confluence. The cells were subsequently collected by scraping the cells in PBS and homogenized buffer (0.87 mM EDTA solution containing 0.1% Triton X-100). The resuspension was transferred to a 1.5 ml Eppendorf tube and sonicated (3X, 10 minutes). Of the homogenate that was sonicated, 10 microliters were transferred to another Eppendorf tube. 10 microliters of radioactive solution (0.2 mM ¹⁴ C-acetylcoenzyme A, 1 mg / ml BSA, 0.2 mM eserine salicylate, 4 mM choline chloride, 18 mM EDTA, 0.6 mM NaCl , And 0.1 mM NaH ₂ PO ₄ ). This mixture was incubated at 37 ° C. for 5 minutes. The reaction was stopped by adding 100 μl of ice-cold distilled water. The sample was extracted with 1.0 ml extraction buffer (a solution of 850 ml toluene and 150 ml acetonitrile mixed with 15 g sodium tetraphenylboron) and centrifuged for 2 minutes in an Eppendorf microcentrifuge tube. It was. Thereafter, 0.65 ml of the organic layer was collected and added to 5.0 ml of scintillation cocktail (National Diagnostics, Manville, NJ). Samples were counted in a scintillation counter for 10 minutes. The amount of protein for the culture was determined by Coomassie protein analysis (Pierce Rockville, IL).

(Northern blot analysis)
Fibroblasts were seeded in 10 cm tissue culture plates and cultured until they reached 70-80% confluence. After washing with PBS, total RNA from each culture was isolated by extracting the cells in 0.5 ml guanidinium isothiocyanate solution. (See description in Chomczynski and Sachhi, Anal. Biochem. 162: 156-159 (1987), which is incorporated herein by reference). The amount of RNA is quantified by measuring absorbance at 260 nm. It was. 10 μg to 15 μg of total RNA was loaded on a 1% formaldehyde-agarose gel. The isolated RNA was obtained from Sunbrook et al. (Sambrook et al., In “Molecular Cloning, A Laboratory Manual”, (Ed. Nolan), Cold Spring Harbor Lab. Press, Cold Spring Harbor. , NY (1989): This document was blotted onto a nylon membrane (MSI), as described in (herein incorporated by reference). Blots were prehybridized (50% formamide, 5X SSPE, 0.5% SDS, 100 μg / ml denatured herring sperm DNA) for 1-4 hours at 42 ° C. It was. Probes are prepared by activating an approximately 2.5 Kb EcoR1 fragment obtained from pLChRNL, or a 953 bp BamHI fragment obtained from pGCc (available from H. Jinnah, University of California, San Diego, Calif.). , Used to detect dChAT or cyclophilin mRNA, respectively. The fragment was isolated on a 1.0% low melting agarose gel, purified using Gene Clean (Bio 101, San Diego, Calif.) And then using ³² P-dCTP. Labeled by random priming (Boerhinger Mannheim, Indianapolis, IN). Approximately 1 × 10 ⁶ cpm / 10 ml of hybridization buffer was used per blot. The radioactively labeled probe was provided directly to the prehybridization solution. Hybridization was performed at 42 ° C. for 12-20 hours. The blot was then washed twice with 6X SSC, 0.1% SDS at room temperature, then twice with 1X SCC, 0.1% SDS at 42 ° C and finally 0.1X Washed once at 65 ° C. using SSC, 0.1% SDS. All were washed in 10 minutes. The washed blot was wrapped with plastic film and autoradiographed (XAR film, Kodak, Rochester, NY). The probe was removed for subsequent hybridization by incubating the blot in 50% formamide, 10 × SSPE at 65 ° C. or higher for 1 hour. The blot was rinsed in sX SSPE before the next prehybridization. The density scan of the autoradiograph was controlled by the LKB fast scan XL laser densitometer.

(Immunohistochemistry)
Fibroblasts are seeded in 2 well tissue culture slides (Tissue Tek) at a density of 1 × 10 ⁴ cells and grown for 2 to 3 days with 4% PBS-buffer paraformaldehyde. Immobilized. Cultures were permeabilized with 0.25% Triton-PRS for 15 minutes, washed twice with PBS, and incubated with primary antibody overnight at room temperature. Primary antibodies used were rabbit-anti-dChAT (1: 100, obtained from Dr. P. Salvaterra, Duarte, Calif.), Mouse-anti-vimentin (1: 250, from Vector Laboratories). Available), or rabbit-anti-rat ChAT (1: 7500, obtained from Boehringer Mannheim). In most cases, the cultures were co-incubated with antibodies against vimentin and one or other ChAT antibodies. After the initial incubation, the culture was again washed twice with PBS and further labeled with fluorescent dye for 1 hour at 37 ° C. (1: 100, obtained from Vector Laboratories), And goat-anti-mouse (1: 100, obtained from Chemicon, Temecular, Calif.), The second antibody. Slides were mounted with Hydromount (National Diagnostics) and viewed with an Olympus fluorescence microscope.

(Stationary phase caused by serum deficiency and confluence)
Rat-1 cells were plated at a density of 2.5 × 10 ⁵ cells per 60 mm plate. In experiments where serum was depleted, the cells were allowed to reach 70-80% confluence, and then the culture medium was aspirated and replaced with fresh DMEM only. Control cells were seeded in medium supplemented with normal serum. Incubation was carried out for 20 hours.

  In the stationary phase, which occurs in a confluent state, the cells are planted as described above. The cultures were then maintained for 3, 7, and 14 days after they reached confluence. To plant these cultures, an equal volume of culture medium was aspirated and replaced with an equal volume of fresh medium. Three days after the last planting, the cells were analyzed. For Northern blot analysis, cells were seeded in 10 cm plates and treated as described above.

([Methyl] - ³ H- introduction thymidine)
The introduction of thymidine was used to monitor the proliferation rate of fibroblasts during the logarithmic or stationary phase of growth. Cells were seeded in 10 cm culture dishes and incubated for 5 hours with ³ H-thymidine (25 Ci / mmol; 0.5 μCi / ml; Boehinger Mannheim). The cultures were then trypsinized, counted using a hemacytometer (cytometer), pelleted and then suspended in 1 ml of 0.3 M NaOH. Incubated for 30 minutes in NaOH solution. The total lysate was then pipetted into scintillation vials containing 10 ml of scintillation solution and counted on a Mark III liquid phase scintillation system model 6881 (Tracor Analytic, Austin, TX).

(Measurement of ACh by HPLC)
ACh was measured by HPLC using an electrochemical assay. ACh was separated from choline on a column charged with weak cations. The column was prepared by running on a Chrompher Cartridge C18 column (100 mm × 3 mm, Chompack) using laurylsulphate. Enzymatic next column reactor containing immunized acetylcholinesterase (type VI-S) and choline oxidase (Sigma, St. Louis, MO) (10 × 2.1 mm, Bioanalytical Systems, West Lafayette, IN) Was used to convert ACh to hydrogen peroxide. Subsequently, hydrogen peroxide was measured with an electrochemical detector (biological analysis system) using a platinum electrode maintained at a potential of +0.5 V with respect to a reference electrode made of Ag / AgCl. A mobile phase containing 0.1 M calcium phosphate and 0.1 mM tetramethylammonium hydroxide, adjusted to pH 8.1, was prepared using an isocratic pump (Pharmacia, Piscataway, NJ). A flow rate of 8 ml / min was supplied.

  The medium conditioned by fibroblasts was filtered through a 0.22 μm Millipore filter and directly injected into the analytical column without further sample preparation. The level of ACh inside the cells was determined by resuspending the pellet, sonicating the suspension, then removing the cellular debris by centrifugation, followed by filtration through a 0.22 μm Millipore filter. Measured. The remaining filtrate was injected into the analytical column.

(Effects of choline chloride and acetyl-dl-carnitine on ACh production)
Fibroblasts were seeded at 1 × 10 ⁵ cells per 35 mm plate and grown to 70-80% confluence. The medium was then aspirated and replaced with fresh medium containing the appropriate amount of choline chloride (Sigma) or acetyl-dl-carnitine (Sigma). Cells were incubated in this medium for 20 hours. Fibroblasts were harvested and prepared for use in either ChAT analysis or HPLC measurement of ACh. The culture for analysis by HPLC had 0.1 mM eserine contained in the medium.

(Statistical analysis)
Student's t-test and ANOVA statistical analysis (STATView, Palo Alto, CA) were used to detect differences between experiments. A post-hoc Dunnet t test was used to examine individual differences between groups.

(Confirmation of dChAT expression by rat-1 / dChAT fibroblasts)
Rat-1 fibroblasts were infected with retrovirus (FIG. 34) and selected on media containing G418-. The presence of dChAT in infected cells was first confirmed by immunocytochemistry. Uninfected rat-1 (control) and rat-1 / dChAT fibroblasts were stained simultaneously with antibodies against vimentin and dChAT or vimentin and rat-ChAT (FIG. 35 (a)-( f)). Vimentin's immune activity effectively reveals fibroblast morphology (see FIGS. 35 (a)-(c)). Most rat-dChAT cells were morphologically indistinguishable from controls and had a flat, epithelial phenotype. Rat-1 / dChAT fibroblasts were not observed morphologically in distinction from controls, and exhibited an elongated morphology.

  As expected, the anti-dChAT antibody was labeled only with rat-1 / dChAT fibroblasts (see comparison between FIG. 35 (a) and FIG. 35 (e)). The staining for dChAT distributed on average through the cytoplasm. However, there was a region with greater intensity of immune activity adjacent to the periphery of the nucleus. As such, it was not stained. In addition to staining for the presence of rebinding dChAT, rat-1 cells and rat-1 / dChAT cells also stained for the presence of rat-ChAT. There was no sign of rat-ChAT immunostaining in either control cells or rat-1 / dChAT cells (see FIG. 35 (f)). The lack of rat-ChAT immune activity indicates very little, if any, endogenous ChAT activity in rat-1 cells. Furthermore, these observations indicate that dChAT molecules are immunologically distinct from rat-ChAT. Previous biochemical and molecular studies have reported that the sequence between dChAT and other mammalian cloned ChAT is different (Berrard et al., Proc. Natl. Acad). Sci. USA 84: 9280-9284 (1987); Itoh et al., Proc. Natl. Acad. Sci. USA 83: 4081-4085 (1987)). These data indicate that in addition to catalyzing the formation of ACh, dChAT can be used as a specific marker for dChAT-expressing fibroblasts transplanted into rat CNS.

  Total RNA from rat-1 fibroblasts and rat-1 / dChAT fibroblasts can also be compared by Northern blot analysis to determine the specificity of transgene expression. The results obtained from this study clearly show that mRNA encoding dChAT is present only in rat-1 / dChAT fibroblasts (FIG. 36 (a)). The size of dChAT mRNA isolated from rat-1 / dChAT was approximately 6 kb. The size of the mRNA is due to the lack of a preadenylation sequence located between the dChAT cistron and the RSV promoter. This allows transcription from the 5'LTR to the 3'LTR to proceed without interruption.

  Finally, ChAT activity was analyzed for control and rat-1 / dChAT fibroblasts. The levels of ChAT detected in rat-1 cells varied widely, ranging from those with no activity to those with a level of about 6.4 nmolACh / hour / mg protein. The average activity exhibited by control cells was approximately 1.9 ± 1.1 nmol ACh / hour / mg protein (see FIG. 36 (b)). In contrast to the low level of ChAT detected in rat-1 cells, the ChAT activity of rat-1 / dChAT cells was over 1200 times higher. The average activity of the rat-1 / dChAT culture was 2397 ± 149 nmolACh / hour / mg protein (FIG. 36 (b)). Furthermore, when dChAT was expressed by transformed fibroblasts, it was stable. The activity of ChAT in fibroblasts that were continuously passaged for one year after infection was the same as the activity of ChAT in fibroblasts analyzed one week after infection.

(Production and secretion of ACh in rat-1 / dChAT cells)
In experiments demonstrating that transformed rat-1 cells express an active recombination enzyme with the ability to acetylate choline in an in vitro assay, the ability of the cell to produce and release ACh from its cytoplasm Is not proved to have Therefore, the level of ACh found in the culture medium and in the cells was analyzed by HPLC with electrochemical detection. ACh was not found either in the cells or in the nutrient medium of uninfected rat-1 cells (see FIG. 36 (c)). This result suggests that the cDNA activity detected in rat-1 cells is due to contamination with an enzyme capable of acetylating choline. Carnitine acetyltransferase is an enzyme found in most cells, which can transfer an acetyl group from acetyl-CoA to choline (White and Wu, Biochemistry 12: 841-846 (1973)). In contrast to the absence of ACh in control cultures, significant levels of ACh were measured in rat-1 / cDNA cultures both intracellularly and extracellularly (FIG. 36 ( c)). The average intracellular volume of ACh was 0.78 ± 0.23 nmol. Compared to this, about 1.7 ± 0.10 nmol of ACh was found in the culture medium.

  These results indicate that ACh has the activity produced by rat-1 / dChAT fibroblasts and that ACh is released from the cells into the surrounding culture medium.

(Effects of choline chloride and acetyl-dl-carnitine on ChAT and ACh expression)
Choline chloride and acetylcarnitine have been reported to enhance secretion of ACh from nerves or synaptosomes in various model systems (Gibson and Shimada, Biochem. Pharmacol. 29: 167-174 (1980); Imperato , neurosci. Lett. 107: 251-255 (1989); Jenden et al., Science 19: 635-637 (1976); Tucek, J. Neurochem. 44: 1-24 (1985)). These drugs were tested for their ability to modulate ACh secretion and ChAT activity in rat-1 / dChAT fibroblasts. Choline chloride was used at concentrations ranging from 0.25 to 500 μM (see FIG. 37 (a)). These concentrations were adjusted by adding to the concentrations usually contained in the medium (about 28 μM) and serum (less than 2 μm). The exogenous choline concentration was 0.25-5.0 μM and did not affect ACh levels. At 10 μM, both intracellular ACh concentration and secreted ACh concentration increased (FIG. 37 (a)). When the intracellular level of choline was increased approximately 2-fold, the concentration of ACh found in the culture medium was increased approximately 1.8-fold. As the concentration of choline continues to increase, intracellular ACh levels continue to increase. Even when choline was added to a concentration of 500 μM or more, the ACh content did not increase further.

  In contrast to the steady increase in intracellular ACh, secreted ACh increased by 80% with 10 μM choline and then decreased back to basal levels with 50 μM and 100 μM choline (FIG. 37 (a )reference). At 200 μM choline, the level of released ACh increased nearly 2.0-fold. ACh continued to increase to 500 μM choline, in which case it reached 2.6 times the level observed in control (no choline added) cultures. At concentrations of choline above 500 μM, ACh release did not increase further. Addition of choline (until 10 mM was reached) to the medium of uninfected rat-1 cells did not result in detection of ACh as measured by HPLC.

  The effect of choline on ChAT activity was also measured (see FIG. 37 (b)). Concentrations of 10, 200, and 500 μM choline were tested. Addition of choline to the culture medium did not significantly affect the ChAT activity of rat-1 / dChAT fibroblasts.

  The effect of acetyl-dl-carnitine on ACh was also tested. Acetyl-dl-carnitine had no effect on the amount of ACh produced and released from rat-1 / dChAT fibroblasts, nor did it affect ChAT activity.

(Effects of serum deficiency and confluence on dChAT expression)
Most of the characteristics of rat-1 / dChAT cells are directed to growing fibroblast cultures. However, the ultimate goal is to use fibroblasts for transplantation. Fibroblasts transplanted into the brain rarely proliferate actively. Thus, the activity of fibroblasts in quiescence in vitro can more accurately reflect the conditions found in vivo, and it should be evaluated. This assessment provides predictive information about the activity of non-dividing fibroblasts implanted in the brain. Rat-1 / dChAT fibroblasts are maintained for several days in either serum-free medium or high density to guide them to stationary phase in fibroblast culture. However, in both methods, thymidine incorporation is significantly reduced, which is that the fibroblasts are maintained for an extended period of time after reaching the confluent stage until the resulting thymidine introduction is much reduced. (See Table 3).

  FIG. 38 (a) shows the effect of serum deficiency on ChAT activity. This treatment did not significantly affect fibroblast morphology or growth. In contrast, when rat-1 / dChAT fibroblasts were maintained in media lacking serum, a significant decrease in transgene expression was caused. Serum deficiency caused an approximately 80% decrease in ChAT activity.

  Northern blot analysis was performed to see if the steady state level of dChAT mRNA decreased with serum deprivation. When serum was removed from the culture, the level of dChAT mRNA was significantly reduced (approximately 46%) (see FIG. 38 (b)). Similar blots were also examined using a cDNA probe for cyclophilin used as an internal control (see McKinnon et al., Mol. Cell. Biol. 7: 2148-2154 (1987)). Contrary to expectations, the steady state level of cyclosphyrin mRNA was also decreased (about 43%) in serum-deficient fibroblasts. By measuring the absorbance of dChAT to cyclophilin with a light densitometer, it was convinced that the reduction seen at the steady state level of dChAT and cyclophilin mRNA was in a proportional condition. Indeed, the ratio of relative absorbance of dChAT to cyclophilin was similar to the ratio of control and serum-deficient culture (see FIG. 38 (c)). Actin and HPRT mRNA were also decreased in serum-depleted fibroblasts.

  Other alternative methods have been used because it is clear that serum deficiency affects the general metabolites of fibroblasts. In that method, after the fibroblasts reached confluence (the number of days after confluence was defined as DPC), the quiescent phase was induced by continuing to maintain the fibroblasts for 3, 7, and 14 days. 3DPC reduced the level of ChAT to 80% of the activity observed in the growing control culture (see FIG. 39 (a)). As DPC increased, ChAT activity continued to decline, with 7DPC and 14DPC each decreasing from 50% to 20% of the control level (see FIG. 39 (a)). Maintaining cells beyond 14 DPC did not result in further reduction in ChAT activity.

  The amount of dChAT mRNA in confluent rat-1 / dChAT was determined by Northern blot analysis (see FIG. 39 (b)). The level of dChAT mRNA isolated from growing cultures (control) or confluent rat-1 cultures did not show a clear decrease, but according to analysis with a light densitometer, dChAT Was found to decrease by about 29% and about 35% at 3DPC and 14DPC, respectively. Interestingly, as the level of dChAT mRNA decreased relatively, the level of cyclophilin mRNA increased. Compared to controls, the relative amount of cyclophilin mRNA isolated from 3DPC and 14DPC cultures increased by approximately 84%. Measurement of dChAT versus cyclophilin ratio showed that a 60.5% reduction occurred at 3DPC and a 64% reduction occurred at 14DPC (see FIG. 39 (c)). All numbers are relative to the dChAT / cyclophilin ratio determined for growing cells.

(Effects of choline when ACh is released from confluent fibroblasts)
In order to evaluate whether choline can enhance the release of ACh from confluent fibroblasts, rat-1 / dChAT cells were treated with confluent medium or choline during 7DPC (days after confluence). Maintained in any of the supplemented (500 μM) media. FIG. 40 shows the effect of maintaining rat-1 / dChAT fibroblasts on ACh production and release in the confluent stage. The total content of ACh (intracellular and extracellular) in 7DPC rat-1 / dChAT cultures was 18.9 of the contents measured in cultures containing mitotic cells. %Met. The decrease in total ACh was apparent to be 50.1% and 82.1% of the reduction and release of ACh in the molecule, respectively. In contrast to confluent cultures maintained in normal medium, stationary phase rat-1 / dChAT fibroblasts maintained in medium supplemented with choline are growing rat-1 / dChAT cultures. The ACh level was comparable to the ACh level measured in (see FIG. 40). Intracellular ACh levels increased to 138% when external choline was added to the nutrient medium, which was comparable to confluent fibroblasts maintained in normal medium. Similarly, the amount of ACh inside the cell increased to 473%.

  The above results described in this experimental example show that rat-1 fibroblasts can be successfully infected with a retroviral vector containing cDNA encoding dChAT. Furthermore, these cells can be produced industrially, and more importantly, ChAT catalyst and ACh can be produced freely. ACh production and secretion from proliferating fibroblasts can be regulated by whether additional choline chloride is added to the culture medium. These data also show that transgene expression is reduced in quiescent fibroblasts, but choline can enhance the release of ACh from these cells.

  Rat-1 / dChAT cells have been developed because several factors that could affect ACh production could be studied in vitro. Rat-1 fibroblasts were selected because they could continue to maintain cultures with little if any change in their genotype or phenotype. The stability in transgene expression in these cells is reflected by the observation that ChAT activity is the same in cells analyzed one week or one year after retroviral infection. The goal in the short-term view in developing ACh-secreting fibroblasts is to analyze the fraction of these cells and transplant them into the CNS. However, rat-1 and other fibroblast cell lines implanted in rat CNS sometimes form tumors (Horellou et al., Neuron 5: 393-402 (1990); Wolff et al., Proc. Natl. Acad. Sci. USA 86: 9011-9014 (1989)). In contrast, primary skin fibroblasts can survive an extended period (6 mos.) In the brain without inducing tumor formation (Fisher et al., Neuron 6: 371-380). (1991); Kajiwa et al., J. Comp. Neurol. 308: 2-13 (1991)). Thus, if ACh-secreting fibroblasts are used for typical brain transplantation, primary cells should be used. Primary skin fibroblasts appear to react in the same manner as rat-1 / dChAT cells, regardless of choline or quiescence (ie, each increases ACh secretion and decreases ChAT activity). .) Therefore, the rat-1 fibroblasts used in this example are used in a strict in vitro format for the reaction of primary fibroblasts grown under similar culture conditions.

  Manipulating the humoral environment within the CNS can increase the ability of cells transplanted into the CNS to produce functional effects if the secretion of the molecule of interest can be regulated. It is effective in. In the in vitro experiments described herein, choline increases the production and secretion of ACh by rat-1 / dChAT fibroblasts. Importantly, choline also induces increased production and release of ACh from quiescent fibroblasts. According to this latter observation, it is possible to provide a means capable of regulating the production of ACh from ChAT-producing cells transplanted into the CNS. At present, no mechanism is known to support the increase in intracellular and extracellular ACh caused by exogenous choline. However, as assessed by measuring ChAT activity, the amount of enzyme is not affected by increasing the extracellular concentration of choline. These results indicate that the reaction rate, ie, the reaction in the direction of ACh formation and not involving the enzyme, is due to the increased level of ACh observed in fibroblast cultures supplemented with choline. ing.

  These results suggest that, for example, administration of choline as an additional diet may increase the release of ACh in dChAT-expressing fibroblasts transplanted into the brain. Choline has been reported to cross the blood-brain barrier (Cohen and Wurtman, Life Sci. 16: 1095-1102 (1975); Cohen and Wurtman, Science 191: 561-562 (1976); Wecker and Schmidt, Brain Res. 184: 234-238 (1980)).

  In addition to the feasible regulation of secretion of the transgene product, another desirable property of genetically modified cells is that the modified cells are stable after transplantation and the recombined gene is long. It is to develop for a period. Fibroblasts transplanted into the brain are surrounded by a copyable amount of collagen and apparently finish growing (Kawaja et al., Ibid., 1991). Current studies strongly suggest that transgene expression from MoMLVLTR is dramatically reduced in serum-deficient and contact-inhibited quiescent fibroblasts. The mechanism of kinetics behind the stationary phase induced by serum deprivation and confluence is unclear, but they are clearly different. This is evidenced by the observation that dChAT mRNA and ChAT activity decrease faster in serum-deficient fibroblasts than in media containing cells with contact inhibition.

  Current data suggests that once primary fibroblasts are transplanted into the brain, expression from proviral transgenes decreases in these cells. This hypothesis is supported by previous studies. This study shows that TH-expressing 208F fibroblasts can be immunolabeled for TH in vitro but are not effectively stained once transplanted into the brain (Wolff et al. Reference 1989). Primary fibroblasts expressing genes for GAD also show a marked decrease in GAD activity when they are implanted into the brain (Chen et al., J. Cell. Biochem., 45-252-257 (1991)). More recent in situ hybridization and immunocytochemical studies have shown that residual amounts of TH mRNA and protein can be seen after subsequent implantation of primary fibroblasts (10 weeks) (Fisher et al. , Ibid., 1991). The results linking the results presented in this experimental example show that although the amount of recombined product is reduced in stationary phase fibroblasts, there is still a residual amount in the cell. The results indicate that a baseline level of dChAT activity (approximately 20% of that found in proliferating cells) is present in quiescent rat-1 / dChAT fibroblasts. This amount of activity is not reduced by maintaining fibroblasts in a confluent state with a period longer than 14 DPC. Thus, even if the activity of the transgene is sufficiently enhanced (as measured in confluent cells in vitro), the amount of activity that remains once the cells have been implanted can elicit functional efficacy. Will be enough. This hypothesis is supported by the observation that the amount of substrate (choline) increases the release of ACh from quiescent rat-1 / dChAT fibroblasts. Therefore, that small amount in the expression of a transgene will be compensated by defined epigenetic factors.

  In conclusion, these results indicate that fibroblasts can be genetically altered to produce and secrete ACh. ACh release can be increased by manipulating the extracellular choline concentration and, if possible, can be increased by administering dietary choline. These data also indicate that transgene expression decreases significantly with growth arrest in vitro.

[Experimental Example VIII]
“Utilization of quiescent fibroblast promoters”
In this experimental example, growth phase (log) and confluent (stationary phase) fibroblast promoters were compared. The reporter transgene used in this study is chloramphenicol acetyltransferase (CAT).

(Vector expression)
Expression of vectors containing the CAT gene has been obtained under the control of various promoters that determine transgene expression in proliferating or quiescent primary fibroblasts. Plasmid pSV2CAT in which the CAT gene is expressed under the control of the SV40 enhancer sequence and the SV40 promoter (Gorman et al., Mol. Cell. Biol. 2: 1044-1051 (1982); Subramani and Southern, Anal. Biochem. 135 : 1 (1983); ATCC No. 37155) and plasmid pSV232ACAT (Fig. 41, provided by Dr. Jolly and Theodore, Dr. Friedman, University of California, D, San Diego, CA, J. Mol. Appl, From and Berg. Genet. 2: 127-135 (1983)) was used. Plasmid pRSVCAT (Rous sarcoma virus long terminal repeat promoter and enhancer, Gorman et al., Proc. Natl. Acad. Sci. USA 79: 6777-6781 (1982); ATCC No. 37152), pMLVCAT (FIG. 42, pSV232ACAT and pGEM) -3 (purchased from Promega, Madison, Wis.) And pN2 (Eglitis et al., Science 230: 1345-1398 (1985)) and provided by Dr. Rosenberg, University of California, San Diego, California Constructed by Dr. Rosenberg from the mouse Maloney Leukemia Virus long terminal repeat promoter and enhancer) and pCMVCAT (Figure 43, pSV232ACAT and pON249 (provided by ES Mocalsky, Stanford University School of Medicine, Palo Alto, Calif.)) Of human cytomegalovirus Early promoters and enhancers) were provided by Dr. Friedman (University of California, San Diego, Calif.). The collagen promoter is a product of M.C. D. Given by Dr. De Kromburg of Anderson Cancer Center. Plasmids pG100 and pAZ1003 (de Crombrugghe and Schmidt, Methods in Enzymology 266: 61-76 (1987)) were given by Dr. De Kromburg, Houston, Texas, and the mouse α1 (I) and α2 (I) collagen promoters (de Crombrugghe, ibid., Α2 (I); Thompson et al., Annals New York Acad. Sci. 580: 454-458 (1990), α1 (I)).

  Since CAT is a bacterial gene that has no analog in mammals, the CAT activity found in transfected cells depends on the presence of the transfected plasmid. The level of CAT enzyme in the transfected cells is balanced with the strength of the promoter (Gorman et al., Proc. Natl. Acad. Sci. USA 79: 6772 (1982)).

(Establishment of fibroblast culture system)
To take the primary fibroblasts of rat skin, the abdomen of 20 young anesthetized Fisher 344 rats was washed with 70% ethanol, taking 25 x 25 mm biomaterial from any rat, Immediately soak in 70% alcohol. Transfer to phosphate buffer (PBS, pH 7.4) to remove alcohol from the tissue. The biomaterial was then transferred to PBS containing 10 mg / ml collagenase and incubated at 37 ° C. for 15 minutes. The skin layer was peeled off from the epidermis layer by cutting with good quality tweezers and enzymes for 30 minutes. Cells were pelleted and transferred to 35 mm ² culture dishes and maintained in minimal medium (DMEM) in Dulbecco (purchased from Gibco, Grand Island, NY) containing 10% fetal bovine serum (FBS). All cultures were maintained at 37 ° C. under 10% CO 2, with the ratio of dividing cells being one third of the total and 80-90% confluent. Transfected fibroblasts were assayed for CAT activity during the log phase when they grew to 70-80% confluence. When quantifying CAT activity in a confluent state, cells grown to 100% confluence were left as they were for 7 days until quantification.

(Fibroblast transfection)
Fibroblasts were prepared using Lipofectin reagent (BRL, Inc., Gaithersburg, MD) using Fergner et al., Proc. Natl. Acad. Sci. USA 84: 7413-7417 (1987) and Felgner and Holm, Focus 11: Plasmid DNA vs. pRSVneo ^R plasmid was transfected with the selected plasmid in a ratio of 10: 1 as described in 21-25 (1988). Both are built into the reference. For each 60 mm culture dish (5 × 10 ⁵ cells / culture dish), 3 ml of Opti-MEM medium (BRL, Inc.) containing different amounts of DNA and 30 μg Lipofectin reagent was used.

In order to select appropriately transformed plasmid DNA containing the neo ^R gene under RSV / LTR control, vector DNA containing the reporter gene (CAT) and under the control of various types of promoters, Mixed in a 1:10 ratio. Once washed with Opti-MEM, the fibroblasts were cultured at 37 ° C. for 5 hours in the presence of a DNA / lipofectin mixture. Thereafter, the medium was changed to DMEM containing 10% FCS and incubated for 24-36 hours before selection.

  Stable transformed cells were selected in the presence of 200 μg / ml G418. Transfected cells were passaged when the culture dish was at least 50% confluent. When the transfected fibroblasts became more confluent than 80%, they were split at a ratio of 1: 2.

  Transfected cells that expressed CAT were analyzed as shown below, both in logarithmic growth phase and after confluence. The results were compared to determine whether the expression of stably surviving CAT transgene decreased or increased after the cells became confluent.

(Analysis of CAT activity)
CAT activity was analyzed using a modification of the method of Sley et al. (Sleigh, Anal. Biochem. 162: 156-159 (1987)). The description of this document is incorporated herein. The dish containing the transfected fibroblasts was washed with PBS and the cells were collected by scraping. Cells are resuspended in 0.25 M Tris-HCl, pH 7.8 buffer (100 μl / plate), lysed by sonication and then centrifuged to remove cell debris. It was. Protein concentrates obtained from the soluble fraction were detected using Cummery's protein reagent (Pierce Chemical Co., Rockford, IL). For analysis, an equal amount of protein from each transfection was used. When added to the cell extract, 100 μl of the reaction mixture contained 20 μl of 8 mM chloramphenicol and 5 μl of 0.5 mM cold [ ¹⁴ C] acetyl-CoA. After incubation at 37 ° C. for 1 hour, chloramphenicol and its acetylated form were extracted with 0.12 ml of ethyl acetate. Approximately 80 μl of the organic layer was collected after centrifugation (10,000 × g) at room temperature. Subsequently, the reaction mixture was re-extracted with ethyl acetate. Approximately 100 μl of the organic layer was collected and mixed with the previously collected 80 μl of the organic layer. After adding 1 ml of Packard Instagel (Packaard, Downers Grove, IL) to the organic layer, the resulting samples were counted in a scintillation counter.

(Northern blot analysis)
During the logarithmic or confluent phase of fibroblast growth, Northern blot analysis was performed to examine whether various expression of the transgene occurred at the transcriptional level, and as described below, The amount was quantified. Cells grown on 10 cm plates are subjected to a process for isolation of total RNA by performing Northern blot analysis at either stage after maintaining 70-80% confluence or for 7 days. Was done. Total RNA obtained from each culture was isolated by extracting the cells in 0.5 ml guanididine isothiocyanate solution. This is done according to the method of Komchinsky et al. (Chomczynski and Sacchi, in the Anal. Biochem, 162: 156-159 (1987)), the description of which is incorporated herein. Protein and DNA were separated from DNA by extraction with phenol-chloroform. The amount of total RNA was quantified by measuring the absorbance at 260 nm. Subsequently, 10-15 μg of total RNA was run on a 1.2% formaldehyde-agarose gel. The isolated RNA was blotted by spreading on a nylon membrane (Hybridization Transfer Membrane, MSI, Inc.) with a capillary using 10X SSC. Prehybridization of this blot (50% formamide, 5X Denhardt's solution, 5X SSPE, 0.5% SDS, 100 μg / ml denatured Hering sperm DNA) It took 2 hours.

To obtain the probe, dChAT cDNA was activated from pcMVCAT using the pstI enzyme obtained from pLChRNL and purified on a low melting agarose gel by the method described below by Maniatis et al. (Maniatis et al., In Molecular Cloning-A LaboratoryManual, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, (1982), Note that the descriptions of which are incorporated herein by reference.) Using a ³² P random Labeled by random priming (MAniatis et al., Ibid.). Denatured radiolabeled probe (boiling for 2 minutes followed by ice cooling) was added directly to the prehybridization solution. Hybridization was performed at 42 ° C. for 12-20 hours. The blot was then washed twice with 6X SSC, 0.1% SDS at room temperature, then twice with 1X SCC, 0.1% SDS at 42 ° C and finally 0.1X Washed once at 65 ° C. using SSC, 0.1% SDS. All were washed in 10 minutes. The washed gel was wrapped with plastic XAR film and autoradiographed (XAR film, Kodak, Rochester, NY). The intensity of each RNA band was measured by scanning with a LKB fast scan XL laser densitometer (LKB Sweden). The blot was reprobed with cyclophilin so that all numbers were removed by washing and could be used as an internal standard. The probe was removed by washing in 50% formamide, 6X SSPE for 30 minutes at 65 ° C or higher for subsequent hybridization. The blot was rinsed in 2X SSPE for successful prehybridization.

  In this experimental example, both the SV40 and LTR promoters and the α2 (I) collagen promoter, a promoter with and without a collagen enhancer, were tested on log phase fibroblasts and stationary phase fibroblasts. The result is shown in FIG.

  The level of CAT activity in fibroblasts containing the SV40 promoter (SV40) was about 3247 and about 2712 DPM / μg of protein for growing and quiescent cells, respectively. This is about 17% attenuation. Similarly, CAT activity in LTR-containing fibroblasts is approximately 24% confluent (3291 DPM / μg protein for log phase cells and 2494 DPM / μg protein for stationary phase cells). It was reduced by. In contrast, the level of CAT measured in fibroblasts containing the α2 (I) collagen promoter increased in the quiescent phase, but overall expression from these cells decreased. The CAT activity of fibroblasts containing the α2 (I) collagen promoter (Col) without an enhancer is about 13 DPM / μg protein and about 75 DPM / μg for log phase cells and quiescent cells, respectively. The amount of protein. This indicates an increase of about 576%. In fibroblasts containing an α2 (I) collagen promoter with a collagen enhancer (Coll (E): Rossi and de Crombrugghe, Proc. Natl. Acad. Sci. (USA) 84: 5590-5594 (1981)) Levels of 405 DPM / μg protein in log phase fibroblasts and 606 DPM / μg protein in stationary phase fibroblasts. This can be said to have increased by about 149%.

  From these results, it has been found that transgene expression obtained from a viral promoter is attenuated in quiescent fibroblasts. In contrast, transgene expression from the α2 (I) collagen promoter increases in quiescent cells. Since fibroblasts transplanted into the brain do not divide, one method for maximizing stable transgene expression is to use a promoter with normal activity in quiescent fibroblasts. can give. The α2 (I) collagen promoter is such a promoter. Furthermore, the enhancer of the Coll (E) promoter has the same effect as the α2 (I) promoter in allowing transgene expression under suitable conditions.

  These results show that when a cell that has been split has been transplanted into the body, it increases the expression of the transgene resulting from that cell, thereby promoting stable expression of the therapeutic transgene over time. It shows how you can do it.

[Experimental Example IX]
"Promoter regulation by cytokines and anti-inflammatory drugs"
Many mononuclear phagocytes and lymphocytes invade the CNS after crossing the blood-brain barrier. This occurs, for example, after transplanting cells into the brain. These cells are known to secrete cytokines into areas around the outside of the cell. Invasion and accumulation of these cells around the transplanted material was observed within one week after transplantation. Since fibroblasts are known to respond to various cytokines, this experimental example was performed to evaluate the role of cytokines in regulating the expression of LTR-acting transgenes in primary rat fibroblasts. It was broken.

  This experimental example is described for the following two uses. 1) Utilizing dexamethasone, an anti-inflammatory drug to eliminate the effects of cytokines at the steady state level of LTR-acting proviral mRNA; and 2) present in the post-transplant brain In order to take advantage of the high level of cytokines, there are two types: utilizing the α2 (I) collagen (Col) promoter with a collagen enhancer (Col (E)). In this experimental example, the transgenes utilized were Drosophila choline transferase (dChAT) and chloramphenicol acetyltransferase (CAT).

(Method)
The method used in this experimental example is as described in Experimental Example VIII above, with the following exceptions:

(Lipofection)
The method of lipofection was as described in Experimental Example VIII above. The plasmids used to transfect cells with fibers in this example were pMLVCAT (the LTR promoter was provided by Dr. Freeman, University of California, San Diego, Calif.), And pR40 (Coll (E)). Promoters and enhancers (Rossi and de Chombrugghe, Proc. Natl. Acad. Sci. USA 84: 5590-5594 (1987)) and Dr. de Crombrugghe, MD Anderson Cancer Center, Huston, Texas DChAT-fibroblasts were transduced by infection with a retrovirus as described above in Experimental Example VII above.

(Fibroblast culture)
Primary skin fibroblast cultures were constructed as described in Experimental Example VIII above, with the following differences. dChAT-fibroblasts were grown to confluence in 35 mm tissue culture dishes using DMEM supplemented with 10% FBS and 200 μg / ml G418. When confluence was reached, the medium was transferred to medium containing 2% serum and maintained for an additional 7 days. Fibroblasts lipofected with pMLVCAT or pR40 plasmids were grown to confluence as described for dChAT-fibroblasts. However, after reaching confluence, the cells were maintained for 4 days in medium containing 10% serum. The cells were then transferred to medium containing 2% serum for the last 3 days (total 7 days after reaching confluence).

(Treatment of transfected fibroblasts with cytokines or anti-inflammatory drugs)
Cytokines were added on day 7 after reaching confluence. TGFβ1 (10 ng / ml), IL-1β (30 μ / ml), and THFα (150 ng / ml) were incubated with the culture for 24 hours. In addition, dexamethasone (25 μM), an anti-inflammatory drug, was added only to fibroblasts and in the presence of TGFβ1 and IL-1β1. The cells were then subjected to Northern blot analysis (for dChAT-fibroblasts) and CAT analysis (for cells containing the LTR or Coll (E) promoter). Infγ was incubated as described above for other cytokines. However, in some experiments, media containing Infγ was aspirated and replaced with media containing fresh 2% serum. The cells were thus maintained for an additional 48 hours, after which Northern blot analysis or CAT analysis was performed. This technique was used to measure that Infγ requires this much time to function.

  The dChAT-expressing fibroblasts in a confluent state were incubated with TGFβ1, TNFα, IL-1β or Infγ described above. Total mRNA from these cultures was isolated and probed using a cDNA fragment that selectively recognizes proviral mRNA by Northern blot analysis. FIG. 45 (a) shows that both TGFβ1 (lane 2) and IL-1β (lane 3) significantly reduce the steady state level of proviral mRNA. On the other hand, TNFα (lane 4) did not produce a significant effect on the level of proviral mRNA. FIG. 45 (b) shows the effect of Infγ on dChAT mRNA capacity. After 24 hours, no change was seen in the steady state of proviral mRNA, but mRNA levels were significantly reduced 48 hours after Infγ was removed and replaced with fresh medium.

  These results indicate that cytokines have an effect on the expression of the LTR agonist gene dChAT in vitro.

  FIG. 46 shows the results of Northern blot analysis in which fibroblasts in the fibroblasts were administered by a combined administration of TGFβ1 and IL-1β that led the LTR-acting dChAT-producing fibroblasts to a confluent state. It has been shown to reduce the amount of proviral mRNA contained in. In contrast, dexamethasone increased the amount of dChAT mRNA detected by Northern blot analysis. When dChAT-fibroblasts were contacted simultaneously with dexamethasone, TGFβ1, and IL-1β for 24 hours, the levels of proviral mRNA were the same as those detected in control cultures. These results show that anti-inflammatory drugs, ie anti-inflammatory drugs including steroids such as dexamethasone, add negative effects of TGFβ1 and IL-1β at the steady state level of proviral mRNA in vitro. Suggests that it can hinder.

  As described above, several cytokines regulate to reduce transgene expression from the LTR. One way to reduce the effects of these cytokines after transplanting fibroblasts into the brain is to suppress the inflammatory reaction using an anti-inflammatory drug such as dexamethasone, for example. Another method that does not require the administration of anti-inflammatory drugs is to use one or more promoters that are not affected by the presence of cytokines or are as high as possible. One such promoter is the α2 (I) collagen promoter. In these experimental examples, the CAT gene was used as a receptor.

  FIG. 47 shows the effects of TGFβ1, TNFα, IL-1β and Infγ on CAT expression in quiescent fibroblast cultures. Here, the Coll (E) promoter-enhancer was used to drive transgene expression (Col (E) -fibroblasts). In these experiments, the culture was maintained in medium containing 2% serum for the last 3 days out of 7 days after reaching confluent culture conditions. Surprisingly, the CAT activity of a Coll (E) -fibroblast culture maintained in a medium containing 2% serum was compared to a control culture maintained in a medium containing 10% serum. It was 440% larger than the CAT activity (Fig. 47 (a)). This activity reached the activity measured against LTR (compared with FIG. 44 (Experimental Example VIII) and FIG. 47 (a)).

  When the cytokine test was performed, TGFβ and IL-1β did not significantly affect CAT activity (see FIG. 47 (b)). On the other hand, the results showed that when contacted with TNFα, the CAT activity increased by about 32%. In contrast, Coll (E) -fibroblasts incubated with INFγ showed a nearly 53% decrease in CAT activity. This measurement was made 48 hours after Infγ was removed and replaced with fresh medium containing 2% serum.

(In vivo expression using Coll (E))
LTR-CAT cells (fibroblasts, where the CAT gene is activated by the LTR) and Coll (E) -CAT cells (fibroblasts, where the CAT gene is α2 (I) collagen Actuated by a promoter-collagen enhancer) was implanted into the striatum of 344 adult Fisher rats as described in Experimental Example VIII above. Three (3) μl LTR-CAT or Coll (E) -CAT cells (200,000 cells / μl: 600,000 cells in total) were injected. LTR-CAT fibroblasts were injected on the right side and Coll (E) -CAT cells were injected on the left side. At 1 to 4 weeks after transplantation, the animals were perfused and the brains were cut and processed to be examined immunohistochemically. The antibody anti-chloramphenicol was used at a dilution of 1: 2000 and was purchased from 5 Prime-3-Prime, Boulder, Co.

  As shown in FIG. 48, one week after transplantation, many strongly stained CAT-positive fibroblasts contain either LTR-fibroblasts or Coll (E) -CAT fibroblasts. (See white arrows in FIGS. 48 (a) and 48 (b)). The intensity of staining was compared between the two transplant types. Staining of CAT at 4 weeks (see white arrows in FIGS. 48 (c) and 48 (d)) appeared more strongly in Coll (E) -CAT fibroblast grafts. At this point, more phagocytic cells were found associated with all grafts (arrowheads indicate the brain interface of the transplanted host). These results indicate that the Coll (E) promoter -Suggests that enhancers can make transgene expression more stable and stronger after transplantation of donor cells into the brain. The presence of phagocytes in the center of the 4 week implant indicates that cytokines may still be present at high levels. Thus, one of the reasons for the stronger expression from the Coll (E) promoter-enhancer compared to the LTR may be that the cytokine works to increase expression.

  The data obtained in this example is that 1) certain cytokines (eg TGFβ1, IL-1β, and Infγ) have a negative effect on the level of proviral mRNA, 2) dexamethasone is Can counteract the negative effects of TGFβ and IL-1β on steady-state levels of proviral mRNA, 3) Coll (E) promoter-enhancers enhance levels of cytokines found in the brain after transplantation It shows that it can be used.

  The observation that cytokines can suppressively regulate expression from the LTR suggests that fibroblasts implanted in the brain may be affected by cytokines. This cytokine is produced by phagocytes and lymphocytes, which are blood-derived mononuclear cells that penetrate the brain after transplantation. Microglial cells and endothelial cells are another source of cytokines in the brain and can contribute to this scenario. Thus, the expression of LTR-operated transgenes in transplanted fibroblasts will result in the time elapsed after transplantation. One method for counteracting the undesirable inhibitory action of cytokines in the LTR is to use molecules such as anti-inflammatory drugs that reduce cytokine production. This approach not only increases the stability of transgene expression in transplanted fibroblasts, but can also contribute to cell survival by decreasing the immune response.

  Alternative approaches include methods that take advantage of the effects of cytokines by using a promoter that has a function that is regulated to enhance the effect by the cytokine or that is at least unaffected by the cytokine. This experimental example shows that TGFβ1 and TNFα do not negatively affect transgene expression in Coll (E) -fibroblasts (as assessed by the activity of CAT). In contrast, TNFα increases expression. Infγ has an effect on the Coll (E) promoter-enhancer as well as on the LTR (ie, reducing transgene expression). Therefore, it is further required to regulate the release of cytokines such as Infγ obtained from T-lymphocytes using immunosuppressive agents such as cyclosporine.

  This data is also preferred in cells in which expression from the Coll (E) promoter-enhancer is maintained in media containing 2% serum compared to media containing 10% serum. It shows an increase without any doubt. The expression level is compared to the observed level observed in fibroblasts containing LTR. Fibroblasts implanted in the body will be exposed to fewer serum elements than during culture. Thus, under certain conditions, expression from the Coll (E) promoter-enhancer could be equivalent to the strongest viral promoter currently used to drive transgene expression.

[Experiment X]
"Regeneration of adult axotomized nerve"
In this example, a new in vivo model is provided that evaluates a substrate that maintains abnormal axon growth in response to trophic factors in the adult rat CNS. In addition, the importance of NGF for the ability of the rat septal nerve to regenerate after axotomy is assessed.

  This experimental example should allow for axonal growth when adult astrocytes are supplied by NGF from a genetically modified primary fibroblast linear graft. Point out that it can serve as a substrate. In addition, the effect of a modified primary fibroblast graft secreting NGF on the regenerative capacity of the rat septal nerve after axotomy is evaluated.

A graft consisting of genetically modified primary skin fibroblasts that express and secrete NGF prepared as detailed in Experimental Example II above was transplanted into the striatum of an adult rat. Primary fibroblasts were obtained from a skin biopsy of Female Fischer 344 rat. Cells were fed under standard culture conditions and fed 3 times a week with DMEM containing 3% fetal calf serum. The primary cells were mouse β-NGF, pLN. Infected with a mulan retroviral vector containing cDNA resulting from 8RLN. Infected cells were selected with the neomyosin analog G418. Prior to isolation of cells for transplantation, culture media samples were removed from control and infected cells. A sensitive two-site immunoassay (Böhlinger-Mannheim) was employed to determine that infected cells produced and secreted 154-173 pg NGF / hr / 105 cells. NGF levels could not be detected in the control medium. Female Fisher rats (weighing about 175-200 g) were anesthetized with a ketamine-xylamine mixture (ketamine 25 mg / ml; lompan 1.3 mg / ml; acepromazine 0.25 mg / ml). The head was shaved and placed in a stereotaxic frame. Prior to the start of the operation, a bactericidal agent was applied to the head. Each animal received 3 × 10 5 cells in 3 μm of grafting solution (Phosphate Buffered Saline supplemented with 1 μg / ml of MgCl ₂ and CaCl ₂ and 0.1% glucose) into the striatum. : NGF-producing cells on the right, non-infected cells on the left. After 1, 3 and 8 weeks survival, the animals were anesthetized and perfused with 0.1 M phosphate buffer 4% paraformaldehyde and 0.1% glutaraldehyde through the cardia. Horizontal or sagittal suture sections through the graft were cut with a freezing microtome and processed for NGF receptor immune response, or cut with a vibratome and processed for electron microscopy experiments.

  Genetically modified cells were injected into the right striatum, and uninfected control cells were injected into the left striatum. One week after transplantation, a network of NGF receptor-immunoreactive axons is evident at the caudal pole of the striatum graft that makes up the NGF-producing fibroblasts, and therefore only a few large and swollen An immune response axon profile was observed near the graft of uninfected control cells (FIGS. 49 (a) and 49 (b)). In three weeks, the density of immune response axons surrounding the NGF-producing graft increased dramatically. The few NGF receptor-immune response profiles completely filled the grafts that made up the NGF-producing cells, and there was no longer a network of axon profiles surrounding the grafts (FIG. 49 (d)). These immune response profiles are larger and clearer than single axons and usually spread vertically. Although uninfected primary fibroblast grafts lack such a pattern of immunostained axon profiles 3-8 weeks after transplantation, the NGF receptor immune response was restricted to few blood vessels (Fig. 49 (c)).

  Ultrastructural experiments showed that all striatum grafts consisted of primary fibroblasts with extensive rough endoplasmic reticulum and an extracellular matrix filled with collagen ( FIG. 50). The capillaries constituted non-perforated endothelial cells that provided an extensive vascular network within these grafts (FIG. 50). At 3 and 8 weeks after transplantation, reactive astrocytes completely encapsulated the graft. These astrocyte profiles had a clear basal lamina on these surfaces juxtaposed around the graft. This barrier between the CNS neural network and the non-CNS graft showed the characteristics of a “gleal scar” that is normally found following wounds penetrating the neural network. Reactive astrocyte processes also spread within the graft material and were often strongly bound to capillaries. Only in the extracellular matrix of NGF-producing grafts, bundles of unmyelinated axons were scattered throughout the graft. Clear and spherical water jars were often observed. Reactive astrocyte processes with many filaments were found to wrap around these axon profiles in a convoluted form (FIGS. 51 (a), 51 (b)). Basal lamina surrounds the entire axon-glial complex and was not evident in the opposing plasma membrane between axons and astrocytes. The occurrence of these axon-glia sequences was greater at 8 weeks than at 3 weeks after transplantation. No axon profile was found in grafts of uninfected control cells.

  These data indicate that NGF receptor-positive axons grow against NGF-producing primary skin fibroblast grafts. On the other hand, grafts of uninfected cells do not reveal similar neural responses. However, both types of cell transplants induce reactive astrocytes to surround the graft and encapsulate it before 3 weeks. Despite the presence of this glial barrier, unmyelinated axons extending along the elaborate array of reactive astrocyte processes were found only in NGF-producing grafts. These axons were confined exclusively to the surface of astrocytes lacking a clear basal lamina. Although the extracellular matrix of the primary fibroblast transplant tissue consists of other conductive substrates for axon growth, such as collagen, laminin, and fibronectin, axons growing within the graft are Did not combine. All of these have been previously pointed out as conductive substrates for axonal growth in vitro. Thus, for NGF-sensitive axons that penetrate a genetically modified primary skin fibroblast graft, reactive astrocytes are the preferred substrate for axon extension.

  These results indicate that mature reactive astrocytes in the adult CNS can act as a substrate that promotes rather than inhibits axonal growth. These data indicate that in order to employ genetically modified cell striatum grafts that produce NGF, successful axonal elongation requires the necessary tropism and nutritional maintenance. Show. Thus, it is clear that the effectiveness of growth-promoting factors has a significant effect on rearrangement, not a substrate for growth.

  To test the effect of grafts on axonal rearrangement, a genetically modified primary cell to express NGF in vivo was established and axotomized cholinergic septum Indicated to provide neuronal nutritional maintenance. To provide a cell source of NGF in vivo, primary skin fibroblasts genetically modified to express NGF prepared as detailed in Experimental Example II are suspended in a collagen matrix. , Placed in a cavity formed by excision of only one of the fibrier fornix (FF) passages, the primary route for septal axons protruding into the rostral hippocampal ridge.

  Under anesthesia, primary skin fibroblasts were obtained from a biopsy of the abdominal abdominal wall of Fimel Fischer 344 rat, and these cells were maintained under standard culture conditions. Mouse β-NGF, pLN. Using retroviral vectors containing the cDNA resulting from 8RNL, primary cells are infected in culture and the NGF product by these cells is then evaluated using a two-site immunoassay (detailed in Experimental Example II). It was done. To suspend the cells in the collagen matrix, cultures of NGF-producing primary cells and control uninfected primary cells are rinsed with phosphate-buffered sarin, trypsinized, and suspended in DMEM. It was done. Cells were then counted and a total of 10 6 cells were equally differentiated in the medium. Type I collagen from rat tail (Sigma) was dissolved in 0.1% acetic acid to a final concentration of 3%. To prepare a more alkaline medium, sodium hydroxide (0.1 N) is added to the cell suspension, followed by collagen (final concentration 1%), mixed, and etc. into a centrifuge tube Differentiated. The collagen / fibroblast matrix was then incubated for 48 hours at 37 ° C. Forty Fimel Fischer 344 rats (weight approx. 175 g) were anesthetized using a mixture of ketamine (75 mg / kg), rompas (4.0 mg / kg) and acepromazine (5.6 mg / kg) for surgery. It was over. These heads were shaved and the animals were placed in a Coff stereotaxic frame. The operation was a two step procedure as detailed in Example II. First, only one suction injury through the cingulate cortex and the FF pathway was formed under stereoscopic viewing. The collagen / fibroblast matrix was then cut into small pieces (approximately 2-3 mm 3) and placed in the damaged cavity. Other animals received only FF damage. Animals recovered from surgery and survived for 3, 4 or 8 weeks. After these survival periods, the animals were deeply anesthetized as described above and perfused through the cardia with 4% paraformaldehyde in phosphate buffer (0.1% glutaraldehyde added for ultrastructure). It was done. The brain is removed, post fixed, cut with either a freezing microtome or vibratome, and each piece (thickness 40-50 μm) is processed for acetylchlorin esterase histochemistry, immunohistochemistry and ultramicrotomy It was done.

  Only animals with complete FF damage and grafts in the cavity received the following evaluation.

  1) Relief of the septal nerve following axotomy, 2) Acetylchlorine esterase (AChE) histochemical and NGF receptor immunochemical staining of grafts and hippocampus, 3) Nitsl staining of grafts, And tyrosine hydroxylase (TH), glial fiber acidic protein (GFAP) and lamina immunochemical staining, 4) hippocampal ridges on the ipsilateral side of the lesion and either NGF production or control uninfected fibroblasts Retrograde transport of fluorescent markers from the graft, 5) Ultrastructural tissue of the graft, 6) Ultrafine AChE staining in the dentate cerebral gyrus with the import path blocked on the same side as the graft of the NGF-producing graft Structure distribution.

  Histochemical and immunochemical detection is done as follows. Sections resulting from collagen / fibroblast grafts were stained with Nissle substance using a 0.5% aqueous solution of thionine, and modified by Hedrien et al. (J. Histochem. Cytochem. 33: 134-140 (1985) ) And histochemically stained with AChE and immunohistochemically with laminin, glial fibrillary acid protein (GFAP) and tyrosine hydroxylase (TH). Sections resulting from the septum are stained immunohistochemically with NGF receptor, and sections resulting from hippocampal ridge are stained histochemically with AChE and immunohistochemically with NGF receptor and TH. It was done.

  The same protocol was used for immunohistochemical detection of NGF receptors, TH and GFAP. The slices are first treated with an aqueous solution of 0.6% hydrogen peroxide for 30 minutes, rinsed briefly with TBS, and contain TBS, 3% defined horse serum and 0.25% Triton-X. Incubated in solution for 1 hour (this solution was used to dilute all antibodies). The sections were then incubated for 48 hours at 4 ° C. in a solution containing one of the following antibodies: Monoclonal 192 IgG against NGF (Chandler et al., J. Biol. Chem. 259: 6882-6889 (1984), 1: 100 dilution), Monoclonal IgG against TH (Boehringer-Mannheim; 1: 250), and GFAP Monoclonal IgG against (Amersham, 1: 100). Control sections are incubated in a solution lacking primary antiserum. All sections were then rinsed with buffer and incubated for 1 hour at room temperature in biotinylated horse anti-mouse IgG (1: 167). After another rinse, the sections were incubated for 1 hour in avidin-biotin complex (ABC; VectorLaboratories). These were then rinsed again and reacted for 5 minutes in a solution containing 0.025% diaminobenzidine tetrahydrochloride, 0.5% nickel chloride, 0.18% hydrogen peroxide in TBS. The reaction was stopped by rinsing the sections in buffer.

  Sections resulting from the graft were also immunohistochemically stained with laminin. Again, the sections were first treated with 0.6% aqueous hydrogen peroxide solution, rinsed, and incubated in a solution of TBS containing 3% normal goat serum and 0.25% Triton-X. It was. Sections were then incubated for 48 hours at 4 ° C. in the same solution as above, supplemented with rabbit anti-human Riminin IgG (1: 800 dilution). Control sections were incubated in a solution lacking primary antiserum. They were then rinsed and incubated in biotinylated goat anti-rabbit IgG (1: 220) for 1 hour at room temperature. After other rinses, the sections were incubated in ABC and reacted as above.

  A marked decrease in the NGF receptor-immunoreactive neuronal population was observed on the same side 3 and 8 weeks after excision of only one of the FFs and placement of the graft comprising collagen and primary fibroblasts. Was evident in the inner septum (FIGS. 52a and 52b). Lines drawn through the medial septum (MS) and graft (G) and hippocampal dentate gyrus (DG) represent the level of the coronal section of the micrograph in this figure. The injection location for fluorescent microspheres within the hippocampus is also shown. Moderate relief of NGF receptor-positive nerves in the same side (right) septum is evident in NGF-producing grafts (arrowheads indicate the central line of the septum).

  NGF receptor rescued by NGF-producing fibroblast grafts-For determination of the septal nerve of the immune response, 120 μm of the septal region immunohistochemically stained with NGF receptor Five sections were selected for counting nerves. A counting grid (0.5 × 0.5 mm) was used to count all immune response nerves in the same and opposite medial septum. The difference between the percentage of NGF receptor-positive viable cells per animal for each group of animals was assessed by one-way ANOVA. Septal rescued with grafts of NGF-producing fibroblasts or control uninfected fibroblasts in the injured cavity 3 or 4 weeks and 8 weeks after FF resection of only one FF Results showing the percentage of nerves (opposite vs. same side) are displayed in Table 4. Coronal sections from the septum were stained immunohistochemically with NGF to determine the value of septal nerve survival. Sections resulting from fibroblast grafts were taken with TH to assess the inward extension of the sympathetic axon, and with GFAP to assess the astrocyte responsiveness to the graft, It was also stained immunohistochemically with laminin to show neovascularity. Sections resulting from the graft and the ipsilateral hippocampus were stained histochemically with AChE and NGF receptors to examine regeneration of septal axons.

  [Table 4] shows that NGF-producing primary fibroblast grafts have NGF receptors at 3 (or 4) weeks and 8 weeks more than collagen grafts with control uninfected primary fibroblasts in the same period. -Significantly maintained a higher percentage of positive septal nerves. The percentage of NGF receptor-immunoreactive nerves is compared between control uninfected grafts and no grafts. Thus, while the degree of rescue in NGF-producing cells is modest, these results are indirect for in vivo expression of the transgene for NGF by genetically modified primary fibroblasts. Provide proof. This expression allows the cells to provide nutrient maintenance for axotomized cholinergic septal nerves.

  Rearrangement of septal axons in the dentate cerebral gyrus of grafts and blocked import tracts was assessed by detection of AChE and NGF receptor staining. NGF-producing fibroblast grafts were vigorously and moderately stained with AChE (FIG. 52 (c)). On the other hand, control uninfected primary fibroblast grafts often lacked AChE reactivity (FIG. 52 (d)). Note that because the NGF-producing graft has a densely stained perimeter, the control graft lacks AChE reactivity (dotted lines indicate adjacent borders of the graft).

  Three weeks after unilaterally-biased FF transection and placement of control non-infected fibroblast / collagen grafts in the cavity, the dentate cerebral tract blocked from the import tract was stained with AChE and NGF receptors. There was no axon. This was in stark contrast to the marked staining with AChE (FIG. 52 (e)) and NGF receptor seen in the normal dentate gyrus. The normal pattern of AChE staining shows strong input for all layers of the dentate gyrus. During the 8 weeks following injury / transplantation, the dentate cerebral gyrus that was blocked from import lanes still lack AChE-positive axons (FIG. 52 (g)), but are now immunohistochemically stained with the NGF receptor. Large diameter axons were provided, especially in the polymorphic layer of the dentate gyrus. These axons imply sympathetic axons that occur within hippocampal ridges after FF injury (see Batcheler et al., J. Comp. Neurol. 284: 187 (1989)). In two cases, AChE-positive axons were observed across the control graft from the opposite side to the hippocampal protuberance that was blocked from the import tract. Inside and above the dentate gyrus where a moderate cluster of AChE-positive axons was blocked from the import tract three weeks after transversal unilateral transection of FF and placement of NGF-producing fibroblasts into collagen The shape of the was obvious. During 8 weeks, AChE-positive fibers increased dramatically, but many axons were still restricted to the superior medial dentate gyrus. However, in some cases, AChE-positive axons topographically organized in the dentate gyrus and CA2-3 range were observed (FIG. 52 (f)). Although the pattern of reinnervation was compared to AChE-positive septal axons in normal, uninjured hippocampal ridges (FIG. 52 (e)), axon density was usually low. In the case of NGF-producing grafts, large axons stained with the NGF receptor are also polymorphic layers of the dentate gyrus blocked by the import tract, usually on the caudal side of the hippocampal ridge. Observed in position.

  Although the pattern of AChE staining in NGF-producing and control grafts was significantly different, both types of grafts had similar cell populations and immunostaining with TH, GFAP and laminin. Nissle staining is a dense form of collagen with a variety of different cell types in and around all grafts, including fibroblasts, lymphocytes, mast cells, plasma cells, capillary stars and endothelial cells. It was shown to be composed of tube bundles. Both types of grafts also had a small number of TH-immunoreactive axons, especially in dense collagen tube bundles. Astrocyte processes immunohistochemically stained with GFAP were found to spread from the damaged nerve tissue around the injury cavity and penetrate into the lateral surface of the graft. Occasionally, GFAP-positive cells were found in the graft. Laminin immunostaining of NGF-producing and uninfected primary fibroblast grafts showed a wide range of host-induced vascular networks. Elongated laminin-immunoreactive cell bodies, possibly Schwann cells, were also found in and around dense collagen tube bundles of both types of grafts.

  To determine the origin of these septal nerves that regenerate new axons after resection of FF and reinnervate the hippocampal protuberances that have been blocked from import tracts, the stereotactic orientation of fluorescent microwear is on the same side. Performed in the dentate gyrus and CA2-3 area. Nine deeply anesthetized animals with NGF-producing (n = 5) or control uninfected (n = 4) fibroblast grafts were used for the dentate gyrus of the hippocampal ridge on the same side as the injury And received two stereotactic arrangements of rhodamine-fluorescence microware (200 nl / region) within the CA2-3 region. Following 48 hours recovery, the animals were re-anaesthetized and perfused through the cardia with 4% paraformaldehyde. The brain was coronally cut with a freezing microtome, and the nerves retrogradely targeted in the septal region were observed with a fluorescence microscope. The schematic diagram in FIG. 52 (h) shows the injection position.

  The most retrograde fluorescently labeled stretched nerves were found in the medial septum and the oblique band region on the same side. On the other hand, a few were observed in the septal medulla and opposite septal nuclei (FIG. 53). With control uninfected fibroblasts, the septal region did not have retrogradely labeled nerves.

At the ultrastructural level, the graft consisted of dense collagen tube bundles with outer reticulated arrays released 8 weeks after implantation. Fibroblasts were observed within a dense and reticulated collagen configuration, and the weakened processes of these cells wrapped up dense accumulation of collagen in both regions. Capillaries comprising non-perforated endothelial cells surrounded by basal lamina were found throughout the graft. Several types of cells were also observed in the graft, including lymphocytes, plasma cells, mast cells, and phagocytic cells. Astrocytes and their processes were widely found within the free reticulum around the central part of dense collagen. The most striking difference between NGF-generated and uninfected fibroblast grafts was the number of unmyelinated axons. That is, 1625 axons were found in the NGF-producing grafts in the 960 μm ² region, whereas 329 axons in the control grafts. Ultra-thin sections of NGF-produced (n = 3 different cases) or control uninfected (n = 3 different cases) fibroblast / collagen grafts were examined and the total number of axons counted within 10 grids (5 grids of 2 parallel columns, total area 960 μm ² , with dense and free collagen area graft tissue). In three cases of NGF-producing grafts, 1625 axons were found compared to 329 in the three cases of control non-infected grafts.

  As shown in FIGS. 54 (a) to 54 (c), a large number of axons surrounded or entered the narrow process of an astrocyte (*) containing an intermediate filament. These astrocyte profiles have a basal lamina (arrowhead) at the surface of these cells facing the matrix outside the transplanted cell. Axons are usually found alongside the surface of these astrocytes that lack basal lamina. Such axons also contain fibroblasts (F) (FIG. 54 (d)), cells similar to other astrocytes, and cells of astrocytes or other cells containing their processes. Found near the body. It was noted that unmyelinated axons bound strongly to the basal lamina of capillary endothelial cells (C) or were found in the extracellular matrix of NGF-producing grafts. Collagen fibrils are distributed between axons (FIG. 54 (e)). Within the NGF-producing or control fibroblast grafts, a small number of unmyelinated axons wrapped by swan cells (S) and their processes were observed in the reticulum region. (Capillary lumen = L) (FIG. 54 (f)). Axons observed in the dense collagen areas of both types of grafts are usually covered in reduced glia. Sometimes control grafts also had a very small population of axons in the extracellular matrix.

  Ultrastructural experiments of hippocampal protuberances blocked by import pathways after placement of NGF-producing grafts show a sparse distribution of AChE responses within the dentate glia. The electron-dense reaction was concentrated at the ends containing small-diameter plasma membranes, unmyelinated axons, and distinct spherical vesicles. These stained axon profiles were widely found in the molecular layer of the dentate gyrus (FIG. 55). The dentate neural network was distributed throughout a population of AChE-positive unmyelinated axons, and aggregates of these axons were usually found near astrocytes. Occasionally, axon ends reactive for AChE formed synapses that contact the dendritic shaft and spine. However, the symmetry of contact between the AChE-positive end and the granular neural dendrite profile was usually unclear due to the localization of the reaction product at close positions. Since the sections of the dentate gyrus of these animals with control uninfected cells that were blocked from import lanes lack AChE reactivity, subsequent analysis could not be derived at the ultrastructural level.

  In particular, FIG. 55 shows a control section (40 μm) of the dentate cerebral gyrus stained with AChE, obtained immediately near the tissue studied at the ultrastructural level. AChE-positive fibers are evident in the inner (IM) and outer (OM) molecular layers and the granular layer (G). All AChE-positive tissues and fibers are also found in the polymorphic layer (P). In FIGS. 55 (b) and 55 (c), a population of AChE-positive unmyelinated axons is found throughout the granular (B) and molecular (C) layers. Figures 55 (d), 55 (e) and 55 (f) show that AChE-positive end-forming snaps contact (arrowhead) the dendritic shaft (Sh) and spine (Sp) in the molecular layer. Show. Astrocyte processes (As) are usually observed near the AChE-positive axons and terminal populations in the dentate gyrus.

  These results point to the importance of nutrient / tropic substances such as NGF in the ability of the septal cholinergic nerve to regenerate. This is the percentage of NGF receptor-positive nerves rescued in the inner septum, and the placement of the graft that makes up the collagen matrix with unilaterally biased FF damage and NGF-producing primary fibroblasts Assessed by reinnervation of the dentate gyrus of later hippocampal ridges. Therefore, an implant in the cerebrum of an NGF-producing primary fibroblast is obtained after axotomy, such as an NGF-producing immortalized fibroblast, as shown in Experimental Example II, and in the lateral ventricle. Exogenous NGF injection (Hefti, J. Neurosci. 6: 2155 (1986); Kromer, Science 235: 214 (1987); Williams et al., Proc. Natl. Acad. Sci. USA 83: 9231 (1986) ) Can later prevent retrograde degeneration of cholinergic septal nerves. The percentage of cells rescued with NGF-producing primary fibroblasts in the collagen matrix, however, is lower than that achieved with injection of NGF or NGF-producing immortalized fibroblasts into the ventricle of the transplanted tissue . This is due to 1) the small number of cells actually transplanted into the collagen matrix due to low levels of NGF product; 2) possible down-regulation of the transgene after transplantation.

  These results show that NGF-producing fibroblast grafts within the collagen matrix, as well as other different types of tissues that maintain new axon growth from septal nerves, also develop septal fiber growth. As a result, it shows that sparse reinnervation of hippocampal uplift is evident in 3 weeks. However, up to 8 weeks, the density of axons within the hippocampal ridge is clearly increased over that seen in 3 weeks. As the septal AChE-positive fibers extend very little into the hippocampal ridge, and labeled cells are not found in the septal region on the same side as the dentate gyrus after placement of the retrograde tracer, Transplantation of control cells within collagen exhibits little or no maintenance. Similarly, grafts of peripheral nerves without cells (ie those lacking Schwann cells) cannot maintain septal axon growth (Hagg et al., Exp. Neurol. 112: 79 (1991) ). At the same time, these data indicate that septal axons can use many different transplant environments consisting of cells for growth and extracellular matrix.

  These data also show that collagen grafts containing primary dermal fibroblasts are basal lamina, stellate, regardless of whether the primary cells are genetically modified to produce NGF. It is shown to promote similar invasion of capillaries with cells and Schwann cells. Moreover, because the extracellular matrix of both types of grafts is similar, the pattern of laminin immunostaining and the distribution of collagen are compared. The actual difference between the NGF-producing graft and the control graft is the intense staining for AChE and the amount of unmyelinated axons within the NGF-producing graft. First of all, significant AChE reactivity is evident in NGF-producing grafts. Control grafts usually lack AChE staining. However, both types of grafts have a sparse population of TH-immunoreactive fibers, as well as Schwann cells and axons that extend into their processes. The largest number of axons connected by these glial elements shows extension into TH-positive sympathetic nerves. Second, NGF-producing grafts have numerous axons that are covered with astrocyte projections, pass through capillaries, or spread into a loose array of collagen. In contrast, control grafts only have a very small population of axons in the extracellular environment. From these collected data, NGF-sensitive axons generated from destabilized septal nerves require the utility of NGF and an acceptable transplantation environment for new growth. By using genetically modified cell grafts that produce NGF in vivo, rearranging septal axons now grow only on their presence on a variety of different substrates showed that. Without increasing levels of NGF, axons do not regenerate in response to grafts consisting of collagen and control uninfected fibroblasts, even though the same cells and extracellular matrix serve.

  Previous studies investigating the ability of septal cholinergic neuronal rearrangement in rats with axotomy and transplantation have shown that "reinnervation of an imported pathway blocked hippocampal ridge with a bridging environment. (Hagg et al., Exp. Neurol. 112: 79 (1991)). However, previous studies have not provided direct morphological evidence that AChE-positive septal axons within the hippocampal ridge actually innervate the postsynaptic target. The ultrastructural data provided in this experimental example show that the axons stained for the activity of AChE were cerebral circulatory in which the import tract was blocked for 8 weeks after transplantation of FF damaged and NGF-producing fibroblasts. It shows that it is distributed significantly. These axons are often found in two or more populations. The ultrastructural tissue of AChE-positive septal axons is also found in septal axons immunohistochemically stained with NGF receptor, which also forms small aggregates in the dentate cerebral gyrus of normal rats. Similar to the recent observation of formation (Kawaja and Gage, J. Comp. Neurol. 307: 517 (1991)). In addition, these newly formed AChE-positive septal axons give rise to synaptic terminals that contact dendritic shafts and spines in the dentate gyrus that are blocked from import tracts. . The pattern of synaptic contacts between the septal cholinergic axon and granule dendritic profiles is compared to that normally found in the rat dentate gyrus. Axotomy contacts between cholinergic axons and granule nerves are rare (Shute and Lewis, Zellforsch. Mikrosk. 69: 334 (1966); and Clarke, Brain Res. 360: 349 ( 1985)). When embryonic septal grafts are transplanted into the dentate cerebral gyrus of adult rats after FF resection, cholinergic axons from the graft are more frequent than normal in the whole tissue of the granule nerve Stimulating with is a valuable reference (Clarke et al., Brain Res. 369: 151 (1986)). Furthermore, the number of transplant-induced cholinergic axons in contact with the dendritic profile is dramatically reduced (Id). From these data, cholinergic axons from embryonic septal grafts in the dentate cerebral gyrus that are blocked from import tracts form a unique synaptic array with granular nerves, whereas collagen and fibroblasts Axons that regenerate across NGF-rich grafts of cells repeat normal synaptic organization within the dentate gyrus (ie, AChE-positive septal axons are trees) Preferentially terminates on the shaft and spine of the projections).

  The results presented in this experimental example show that NGF-producing cholinergic nerves in the collagen matrix maintain damaged cholinergic nerves in the inner septum of rats, and NGF-generated from these nerves. Providing a guided environment for regenerating sensitive axons, and affecting the reinnervation of nerves in the dentate gyrus import pathways blocked by septal axons, Point out that. In particular, these grafts promote axon elongation and synaptic contact. To regenerate NGF-sensitive axons, both cellular and matrix substrates are used for growth in grafts consisting of NGF-producing primary fibroblasts and collagen. Without increasing the level of NGF (ie, using a control fibroblast / collagen graft), a higher percentage of septal nerves suffers from degeneration, despite the presence of a conductive substrate, and Axons fail to grow.

  In addition to NGF, other neurotrophic factors such as neurotrophin (NT) -3, NT-4, and ciliary trophic factor (CNTF) are also used to maintain axotomized nerves and Used to enhance axonal regrowth in other neural populations in the CNS.

  Obviously, modifications and variations of the invention as described above may be made without departing from the spirit and scope of the invention. The specific embodiments described in detail are only given as examples, and the present invention is limited only to the language of the claims.

It is the figure which showed typically the mode for outlining and analyzing the effect | action of a new function in a target cell. It is the schematic which showed the method for introduce | transducing a new function into the target cell in CNS using the genetically engineered donor cell. It is the schematic for demonstrating the synthesis | combination of the inherited retroviral vector containing a transgene. (GAG = specific antigen group; ENV = envelope; POL = reverse transcriptase) FIG. 2 is a schematic showing a linear restriction enzyme map of the complete vectors LSΔPΔLM and pLNHI 2 as described in Experimental Example I. (The arrows indicate the location of the promoter and the direction of transcription. The hatched box of LSΔPΔLM encodes a protein with a novel terminal hexapeptide incorporated by mutagenesis in vitro. Represents human HPRT cDNA, LTR = long terminal repeat) FIG. 5 shows a circular restriction enzyme map of the vector pLNL2 as described in Experimental Example I. (A)-(f) are fibers of primary rats previously infected with phosphoribosyltransferase (HPRT) implanted in rat basal ganglia, as described in Experimental Example I. It is a microscope picture of a blast cell. (FIG. 6 (a), (d) = anti-fibronectin; FIG. 6 (b), (e) = Cresyl violet; FIG. 6 (c), (f) = GFAP; (Magnification: FIGS. 6 (a) to (c) = 88X; (d) to (f) = 440X) FIG. 4 is a photograph of a gel when focused on the isoelectric point performed on the HPRT enzyme activity of brain extracts from basal ganglia as described in Experimental Example I. FIG. 5 shows a circular restriction enzyme map of the vector pLLRNL as described in Experimental Example II. FIG. 4 shows a circular restriction enzyme map of the vector pPR1 as described in Experimental Example II. FIG. 4 shows a circular restriction enzyme map of the vector pUCRH as described in Experimental Example II. As shown in Example II, 777 base pairs of mouse nerve growth factor (NGF) cDNA, Hgal-Pstl, under the control of a viral 5 'long terminal repeat (LTR). The complete NGF retroviral vector PLN. It is a figure which shows the linear restriction enzyme map of 8RNL. (Arrows indicate the start site of transcription; psi (ψ) = retroviral packaged signal; RSV = loose sarcoma virus promoter; neo ^R = neomycin-resistance Genetic marker) As shown in FIG. 11 and Experimental Example II, the vector pLN. It is a figure which shows the cyclic | annular restriction enzyme map of 8RNL. It is a microscope picture at the time of performing an immunohistochemical dyeing | staining with respect to fibronectin and ChAT which are demonstrated in Experimental example II. (FIGS. 13 (a), (b) = the fimbria fornix cavity; FIGS. 13 (c)-(f) = inside the tissue stained for ChAT. FIG. 13 (a), (c), (e) = animal transplanted with retrovirus-infected cells; FIG. 13 (b), ( d), (f) = animal transplanted with control cells; magnification (Mag.): FIGS. 13 (a) and (b) = 20X; FIGS. 13 (c) and (d) = 70X; ) And (f) = 220X) As described in Experimental Example II, it is a graph showing the results of examining the survival of ChAT-immunoreactive cells in the rat diaphragm in the presence and absence of NGF. (A)-(f) is a microscope picture which shows the histochemistry of acetylcholinesterase demonstrated in Experimental example II. (FIG. 15 (a) = low magnification of an animal transplanted with NGF-infected donor cells; FIG. 15 (c) = high magnification of (a) through the inner diaphragm; (E) = high magnification of (a) through the lateral quadrant on the dorsal side; FIGS. 15 (b), (d), (f) = animals implanted with control cells, FIG. ), (C), (e); performed under the same conditions as described for; magnification (Mag.): FIGS. 15 (a), (b) = 20X; FIGS. ) = 220X) FIG. 4 is a schematic diagram showing a linear restriction enzyme map of the pLThRNL retroviral complete vector as described in Experimental Example III. (Arrows indicate promoter location and transcription direction; LTR = long terminal repeat; RSV = modified RSV promoter; neo ^R = neomycin-resistant gene) FIG. 17 is a diagram for explaining the induction of the vector pLThRNL as shown in FIG. 16 and described in Experimental Example II. (A)-(d) is a microscope picture which shows the fibroblast transplant to the caudate which shows fibronectin immunoactivity as demonstrated in Experimental example III. (Magnification: FIG. 18 (a), (c) = 10X; FIG. 18 (b), (d) = 20X) As described in Experimental Example III, it is a graph showing the average rate of change when the number of rotations from the reference point to the next transplantation is examined for a group of four experimental animals. (Upper part of the drawing: transplantation by control and transplantation for TH-infection in tail striatum; Lower part: substitution by control and transplantation by TH-infection in rostral striatum; horizontal axis Is the standard deviation) It is the graph which showed the growth curve of the fibroblast of the primary skin (primaryskin) and the fibroblast of the rat-1 according to description in Experimental example IV. As described in Experimental Example IV, after one or four passages of inoculation in the culture medium prior to transplantation, self-fibroblasts after survival for 3 to 8 weeks while observing the course It is a histogram which shows the capacity | capacitance (mm < ³ >) of a cell. As described in Experiment IV, 10 ⁴ per 1 [mu] l, 2.5 × 10 ⁴ cells, and 10 volumes of ⁵ cells were obtained by transplanting autologous fibroblast grafts (mm ³ ). (A) to (d), as described in Experimental Example IV, after 3 weeks (FIGS. 23 (a) and (c)) and 8 weeks (FIGS. 23 (b) and (d)) after transplantation. It is a photograph which shows the graft (G) of the striatal primary fibroblast obtained by dyeing fibroblasts immunohistochemically. (Magnification: FIGS. 23 (a), (b) = 12X; FIGS. 23 (c), (d) = 120X) (A)-(f), as described in Experimental Example IV, 3 weeks after transplantation (FIGS. 24 (a), (c), (e)) and 8 weeks (FIG. 24 (b), (D), (f)) At the same time, photographs of primary fibroblasts in the striatum stained with cresyl violet (FIGS. 24 (a), (b)), and at the same time, about GFAP (FIG. 24). (C), (d)), and photographs of primary fibroblasts in the striatum immunologically stained for OX-42 (FIGS. 24 (e), (f)), fibroblasts immunohistochemically It is a photograph which shows the graft | graft (G) of the striatal primary fibroblast obtained by dyeing | staining to. (Magnification = 120X) (A) to (c) are self-transplantation superstructures 8 weeks after transplantation (F = fibroblasts themselves; solid points = existing portions of fibroblasts) as described in Experimental Example IV. It is a photograph which shows (ultrastructure). (Magnification: FIG. 25 (a) = 6,600X; FIG. 25 (b) = 9,100X; FIG. 25 (c) = 16,600X) (A), (b) is a photograph showing the astrocytes present in the graft and phagocytic cells as described in Experimental Example IV. (Magnification: Fig. 26 (a) = 16,600X; Fig. 26 (b) = 15,000X) (A), (d) is a photograph showing a transplanted blood vessel as described in Experimental Example IV. (FIG. 27 (a), E = fenestrated endothelial cells without window); FIG. 27 (b), arrowhead portion = inserted portion; FIG. 27 (c), arrow = injection into the cell; d), As = Astrocyte present portion; Magnification: FIG. 27 (a) = 13,200X; FIG. 27 (b) = 25,400X; FIG. 27 (c) = 25,000X; FIG. 16,600X) As described in Experimental Example V, the photographs show the results of in vitro labeling of tyrosine hydroxylase of primary fibroblasts containing a TH transgene. (Magnification = 25X) FIG. 6 is a graph showing the results of transplantation for turning behavior in 6-OHDA-lesioned rats introduced with apomorphine as described in Experimental Example V. FIG. . (A)-(b) is a microscope picture which shows the result of having hybridized in situ with the TH primary fibroblast transplanted to the rat striatum, as demonstrated in Experimental example V. FIG. (FIG. 30 (a) is a graft of FF2 / TH cells; FIG. 30 (b) is a graft of FF2 / βGal cells (control); G = graft; V = ventricle) (A)-(d) are photographs showing the results of in vivo labeling of transplanted cells against either βGal histochemistry or in situ hybridization to TH mRNA as described in Experimental Example V. It is. (FIG. 31 (a) is a graft of FF2 / βGal histochemically stained βGal; FIG. 31 (b) is a high magnification of the FF2 / Gal graft; FIG. 31 (c) is FF2 Hybridization in situ to TH mRNA in / TH transplant, arrow = fibroblast; FIG. 31 (d) shows in situ hybridization to TH mRNA in FF2 / βGal transplant; magnification: (a) = 10X, (B), (c) and (d) = 40X) (A), (b) are photographs showing the TH-immunoactivity of FF2 / TH fibroblasts in vivo, as shown in Experimental Example V. (ASTRICK = rhodamine fluorescent dye; magnification = 20 ×) FIG. 4 is a graph illustrating human NGF secretion obtained from genes and cell types as shown and described in Experimental Example VI. As shown in Experimental Example VII, it is a schematic diagram of the pLChRNL vector. FIG. 4 is a photograph of an immunostained fibroblast as described in Experimental Example VII. (A), (c) follows the description in Experimental Example VII and shows the following results. FIG. 36 (a) is a Northern blot analysis of total RNA obtained from fibroblasts; FIG. 36 (b) is an analysis of ChAT activity in fibroblasts, and FIG. Analysis of ACh secreted into the medium. (A)-(b) are graphs illustrating the results of adding choline chloride to a culture of rat-1 / dChAT as measured by HPLC and as described in Experimental Example VII. It is. (FIG. 37 (a) = effect on intracellular ACh and extracellular ACh; FIG. 37 (b) = effect on ChAT activity) (A)-(c) show ChAT activity (FIG. 38 (a)); steady level dChAT mRNA (FIG. 38 (b)); and dChAT / cyclophilin (cyc.) As described in Experimental Example VII. As can be seen by the effect on the mRNA ratio (FIG. 38 (c)), the results show a quiescent state led by removal of serum in the expression of dChAT. (A) to (c) show ChAT activity (FIG. 39 (a)); dChAT mRNA level (FIG. 39 (b)); and dChAT / cyclophilin (cyc.) As described in Experimental Example VII. As can be seen by the effect on the ratio (FIG. 39 (c)), the result shows a quiescent state led by contact with an inhibitor in the expression of dChAT. FIG. 4 is a bar graph showing the effect of choline on the release of ACh obtained from confluent fibroblasts as described in Experimental Example VII. FIG. 5 is a diagram showing a circular restriction enzyme map of vector pSVS32ACAT as described in Experimental Example VIII. FIG. 4 is a diagram showing a circular restriction enzyme map of vector pMLVCAT, as described in Experimental Example VIII. FIG. 4 is a diagram showing a circular restriction enzyme map of vector pCMVCAT, as described in Experimental Example VIII. FIG. 5 is a bar graph showing the relative CAT expression of SV40, LTR and collagen promoter (Col1 and Coll (E) promoter-enhancer) as described in Experimental Example VIII. (A)-(b) are Northern blot photographs showing the effects of TGFβ, IL-1β and TNFα on dChAT-expressing fibroblasts as described in Experimental Example IX. (FIG. 45 (a) shows the effect of TGF-β (lane 2) and IL-1β (lane 3). FIG. 45 (b) shows the effect of Infγ. Lane 1 = control; cyclophilin ( Cyc) = substantial control; lane 2 = effect after 24 hours; lane 3 = effect after 48 hours; dChAT = dChAT proviral mRNA.) As described in Experimental Example IX, Northern blot photographs showing the effect of combined administration of TGFγ and IL-1β, and further combined addition of dexamethasone, on confluent dChAT-producing fibroblasts. (Lane 1 = control; lane 2 = TGF-β / IL-1β; lane 3 = administration of dexamethasone only; lane 4 = TGFβ / IL-1β and dexamethasone; strong signal = dChAT mRNA; weak signal = cyclophilin mRNA) (A)-(b) are about the activity about the collagen promoter contained in 10% with respect to 2% serum, and the effect of various cytokines on the collagen promoter as described in Experimental Example IX. It is the bar graph shown. (A)-(d) shows LTR-CAT transplanted into the striatum of adult Fischer rats (FIGS. 48 (a), 48 (b) as described in Experimental Example IX. )) A micrograph showing a cell or a graft of Coll (E) -CAT (FIG. 48 (c), FIG. 48 (d)) cell. (White arrow = fibroblast) (A)-(d) are transplants (cells genetically modified by injection into the right striatum and infection injected into the left striatum, as described in Experimental Example X) Left and right striatum at 1 week (FIG. 49 (a), FIG. 49 (b)) and 8 weeks (FIG. 49 (c), FIG. 49 (d)) after the uncontrolled) It is a microscope picture which shows the state which passed through and was cut | disconnected horizontally. (G = graft; NG = nucleus basalis); RT = reticular thalamic nucleus; scale bar: 0.45 mm for FIGS. 49 (a) and 49 (b). 49 (c) and 49 (d) are 0.22 mm) As described in Experimental Example X, an electron micrograph of a graft with NGF-producing cells bound for 8 weeks after transplantation. (F = fibroblasts; E = endothelial cells; arrow = existing part of astrocytes; scale bar = 1.0 μm) As described in Experimental Example X, a photograph and schematic view of an axo-glial arrangement in an NGF-producing cell graft after 8 weeks, as viewed under a microscope. is there. (Ax = axon without myelin sheath; As = represented part of reactive astrocytes; arrowhead = basal lamina; scale bar = 1.0 μm) (H), as described in Experimental Example X, peels off one side of the fibrier-envelope (FF) and uses either NGF-productive or control non-infected primary fibroblasts. It is the figure which showed roughly the sagittal suture part of the frontal part of an adult rat after performing the replacement | exchange of the graft | graft which comprises collagen, and (a)-(g) is the photograph of the applicable part. (MS = diaphragm of medium, G = graft, and DG = hippocampal gyrus of the hippocampus, FIGS. 52 (a), (b) = unilateral FF aspirative lesion)) and NGF− After replacement of either the production (FIG. 52 (a)) or control (FIG. 52 (b)) grafts (arrowheads are the septal centerline), immunohistochemically directed against the NGF receptor. The parietal region seen at a comparable level through the stained septum; FIG. 52 (c), (d) = NGF-stained with AChE-productive or control non-infected (FIG. 52 (d) ) Through the graft of any of the fibroblasts (dotted lines indicate the approximate boundaries of the transplant, scale bars are 100 μm in FIGS. 52 (b) and 52 (c)); ) Scale bar is 200 μm); FIG. e), (f), (g)) = the parietal region when viewed through the dentate gyrus of the hippocampus stained with AChE (G = granular layer; IM = inner molecular layer; OM = outer molecule) Layer, and P = polymorpholine layer) (A)-(d), as described in Experimental Example X, the septal nerve was labeled with declining fluorescence, followed by surgery on animals that received NGF-producing grafts. It is a microscope picture which shows being arrange | positioned at the fixed position of the fluorescent microspheres (fluorescent microspheres) in a later toothed cerebral gyrus. (FIG. 53 (a) = diaphragm of the central part on the same side; FIG. 53 (b) = an oblique band on the opposite side; FIG. 53 (c) = centerline of the septum; and FIG. Diagonal band; scale bar = 50 μm) (A)-(f) is an axon that does not have a myelin sheath in a collagen and NGF-producing fibroblast graft after surgery, as described in Experimental Example X. It is an electron micrograph which shows a superstructure dispersion arrangement. (* = Existing portion of astrocytes; S = Schwann cells; L = capillary lumen) (A)-(f), as explained in Experimental Example X, in the dentate gyrus after exfoliation of FF and transplantation of NGF-producing fibroblasts into a collagen matrix It is the figure (FIG. 55 (a)) which showed AChE activity as a local distribution arrangement | positioning, and the figure (FIGS. 55 (b)-(f)) shown as a distribution arrangement | distribution in a synapse. (FIG. 55 (a) = A portion immediately adjacent to the tissue to be detected at the ultrastructural level, the parietal portion of the dentate gyrus stained with AChE (40 μm), IM = inside G = granular layer; P = polymorpholine layer; FIG. 55 (b) = granular layer; FIG. 55 (c) = molecular layer; FIG. 55 (d), (e), (F) Arrowhead = synaptic contact, Sh = dendritic shape; Sp = spinal cord, As = astrocyte existing part, scale bar: FIG. 55 (a) = 25 μm; FIG. 55 (b) = 0 .5 μm, FIG. 55 (c) and FIGS. 55 (d) to (f) = 0.25 μm)

Claims (3)

A composition for administration in conjunction with oral administration of choline chloride to treat a mammal having cells with defective, diseased or damaged central nervous system comprising:
A composition comprising, as an active ingredient, donor cells that have been genetically modified by inserting a vector containing choline acetyltransferase (ChAT) cDNA. A method of preparing a composition for administration in conjunction with oral administration of choline chloride to treat a mammal having cells with defective, diseased or damaged central nervous system comprising:
A method comprising genetically modifying a donor cell by inserting a vector containing choline acetyltransferase (ChAT) cDNA into the donor cell. A method of preparing a composition to be administered to treat a defective, diseased or damaged cell in a mammalian central nervous system comprising:
The donor cell is genetically modified by inserting into the donor cell a vector having a therapeutic transgene encoding a functional molecule and a promoter that enhances the expression of the transgene when the donor cell is at rest. A method comprising the step of:

Images