CA1341246C - Somatic cell gene therapy - Google Patents

Abstract

The present invention is a somatic cell gene therapy method that is especially useful for the treatment of certain diseases that are caused by gene defects. According to the invention, fibroblast cells are transduced so that they express a "replacement"
gene of interest. These transduced fibroblasts are preferably fixed in vitro in an extracellular matrix, and then implanted in the loose connective tissue of the skin of an individual or animal to be treated.
Because the fibroblasts are implanted in a highly vascularized compartment of the skin i.e., loose connective tissue of the dermis, the transduced cells, and thus their "replacement" gene products, have direct access to the circulatory system. As a result the needed replacement gene products can easily and efficiently be distributed to other parts of the body.
When the gene therapy is no longer needed, the implanted fibroblasts can be conveniently removed.

Description

t3~it~4rg-' SOMATIC CELL GENE THERAPY
ACKNOWLEDGMENT
This invention was made with government support under grants from the National Institutes of Health.
FIELD OF THE INVENTIOPT
The present invention relates generally to gene therapy. More specifically, the present invention relates to somatic cell gene therapy in humans and animals.
BACKGROUND OF THE INVENTION
More than 1500 human genetic diseases are caused by a single gene defect. It is possible that many of these diseases can alleviated, at least in part, if the deficient function can be supplied.
The concept of human gene therapy involves the introduction of a functionally active "replacement"
gene into somatic cells of an affected subject to correct the gene defect. Retroviral vectors, because of their unique structure, mode of replication, and ability to infect a wide variety of cells, including stem cells, are ideally suited to transfer genetic material into somatic cells (Verma, 1985).
To ensure a life long supply of the replacement gene product, it is essential to introduce and express the functionally active gene in cells that proliferate during the entire adult life of the recipient. Because pluripotent stem cells in bone marrow have both self renewal capacity as well as the ability to give rise to all hematopoietic lineages, they are a popular target for the introduction of functionally active genes (Miller, et al., 1984; Williams, et al., 1984; Keller, et al., 1985: Dick, et al., 1985). Recently, hepatocytes have been used as target cells for introducing functionally active genes (Ledley, et al., 1987; Wolfe, et al., 1987).
Although the number of stem cells in adult marrow is low (0.01-0.1%), the use of high-titer retrovirus has ensured infection and gene delivery into these cells. The problem however has been that neither the foreign genes nor the retroviral vector introduced into these stem cells, the progenitor cells, or the mature end cells are efficiently expressed (Williams, et al., 1984; Joyner, et al., 1985).
To overcome this problem of inefficient expression, the present invention discloses an alternative strategy for somatic cell gene transfer.
The new strategy uses skin fibroblasts that are infected with chimeric retrovirus containing a functionally active endogenous or foreign "replacement" gene. Once infected with the chimeric retrovirus, the transduced fibroblasts are preferably "fixed" in an extracellular collagen matrix, and then implanted in the loose connective tissue of the skin.
Since this compartment of the dermis is highly vascularized, the transduced fibroblasts, and thus their "replacement" gene products, have direct access to the circulatory system. As a result the needed replacement gene products can easily and efficiently be distributed to other parts of the body.
Recently two groups used mouse fibroblasts to introduce and express foreign genes in mice (Selden, et al., 1987; Carver, et al., 1987b). One group implanted mice with a DNA transfected cell line and showed that the recipient mice made the gene product (growth hormone) but. maintained the graft only if the mice were immunosuppressed (Selden, et al., 3987). The other group, using a chimeric retroviral vector containing the alphal-antitrypsin gene, produced a cell line from a transduced cell and then transplanted cells from the line into the peritoneal cavity of nude mice (Garver, et al., L987b). In both cases, cell lines were generated that would potentially be tumorigenic in mice. Neither study addresses the issue of cell maintenance in grafted mice without the use of harsh immunosuppressive agents.
The method described herein obviates the need for established cell. lines and uses instead, fibroblast cells from recipient subjects. Use of a subject's own cells minimizes the possibility of rejection. In addition, culturing the cells in an extracellular collagen matrix circumvents the problem of necrosis that would ensue following subcutaneous injection (Bell, et al., 1983). Finally, the high efficiency of retroviral infection and expression in fibroblasts (80%) essentially eliminates the need to identify transduced cells by means of selectable markers, thus greatly simplifying the overall endeavor of introduction of foreign genes.
Clinical disease states that are candidates for the gene therapy treatment method of the present invention include hemophilia, endocrine deficiency, alphal-antitrypsin, birth control, etc.
In addition to the work that has been done with fibroblasts, at least one group has shown that retroviral-mediated gene transfer can be used to introduce a recombinant human growth hormone gene into cultured human keratinocytes (Morgan, et al., 1987).
The transduced keratinocytes secreted biologically active growth hormone into the culture medium. When grafted as an epithelial sheet onto athymic mice, these cultured keratinocytes reconstituted an epidermis that was similar in appearance to that produced by normal 134 246 .
cells, but from which human growth hormone could be extracted. Unfortunately, it was not possible to determine rate of diffusion of human growth hormone from the graft site to the bloodstream. This may have been due to the fact that the surface skin graft does not vascularize as efficiently and as quickly as the embedded fibroblasts of the present invention.
BRIEF DESCRIPTION OF THE DRAWING
The following is a brief description of the drawings. More detailed descriptions of the figures are found in the Experimental Section of this specification.
The drawings comprise five Figures.
Figure 1 (A, B & C) is composed of a schematic drawing (A), and two photographs (B & C), all of which relate to an analysis of recombinant human factor IX
retrovirus. Fig. 1A shows the structural arrangement of the recombinant factor IX retrovirus pAFFIXSVNeo.
Figs. 1B and 1C are photographs of nitrocellulose filters that illustrate proviral DNA and RNA
hybridizations, respectively.
Figure 2 is a graph that illustrates secretion of human factor IX.
Figure 3 (A, B & C) is composed of a schematic drawing (A), and two photographs (B & C), all of which relate to embedding and implantation of transfected fibroblasts. Fig. 3A is a schematic representation of the protocol used to generate and graft the collagen implants into the loose connective tissue of the skin of the mouse model system. Figs. 3B & 3C are photographs of mice that show the MEF (primary fibroblast cells) and BL/6 (the tumor cell line) collagen implants at day 14.
Figure 4 is a graph that compares the amount of factor IX in human sera from both the MEF and BL/6 implants.
Figure 5 is a photograph of a nitrocellulose filter that illustrates detection of mouse anti-human factor IX antibodies in mice grafted with collagen implants.
DEFINITIONS
In the present specification and claims, reference will be made to phrases and terms of art which are expressly defined for use herein as follows:
As used herein, LTR means long terminal repeat.
As used herein, factor IX refers to the blood clotting factor gene or protein of the same name.
As used herein, pAFVXM refers to a retroviral construct generated by Kriegler, et al., (1984).
pAFVXM is a progenitor construct for the recombinant factor IX retrovirus, pAFFIXSVNeo. A replacement gene of interest (or a cDNA for such a gene) can be linked directly to the 5' LTR in the retrovirus by inserting a BamHI/HindIII fragment from the gene or clone between the BalII/HindIII sites of pAFVXM (Anson, et al., 1984). pKoNeo is a neomycin phosphotransferase expression plasmid.
As used herein, when reference is made to the Greek letter psi, the name psi is sometimes substituted for the symbol V.
As used herein, the letter g is sometimes used to signify the symbol for the Greek letter gamma, Y.
As used herein, MEF means primary mouse embryo fibroblasts.
As used herein, B1/6 refers to an immortalized skin cell line derived from x-ray irradiated skin fibroblasts obtained from C57BL/6J mice.
As used herein, psiFIXNeo means the cell line 1 3 ~ t 2 46 V~FIXNeo.
As used herein, moi means multiplicity-of-infection.
As used herein, POLYBRENE is the trademark of Sigma Chemical Company, St. Louis, MO for 1,5,-Dimethyl-1,5-diazaundecamethylene polymethobromide;
Hexadimet.hrine bromide.
As used herein, DMEM means Dulbecco's modified Eagle's medium.
As used herein, ELISA means enzyme linked immunoabsorbant assay.
As used herein, FGF means fibroblast growth factor. FGF is a angiogenic substance that can be used in the present invention to stimulate vascularization of the implanted fibroblasts.
As used herein, transduction refers to the process of conveying or carrying over, especially the carrying over of a gene from one cell to another by a virus or retrovirus. A retrovirus that carries a gene from one cell to another is referred to as a transducing chimeric retrovirus. A eukaryotic cell that has been transduced will contain new or foreign genetic material (e.g., a replacement gene) in its genome as a result of having been "infected" with the chimeric transducing retrovirus.
As used herein, transfection of eukaryotic cells is the acquisition of new genetic material by incorporation of added DNA.
As used herein, the term skin technically means to the body's largest organ. Skin consists of two components, the epidermis and the dermis. The dermis is a relatively inert structure which consists of collagen and other matri~c materials. The epidermis lies above the dermis and is separated from it by a !, 1 3 ~ fi~2~8 basement membrane.
As used herein, the term fibroblasts refers to flat, elongated connective tissue cells with cytoplasmic processes at each end and an oval, flat nucleus. Fibroblasts, which differentiate into chondroblasts, collagenoblasts, and osteoblasts, form the fibrous tissues in the body, e.g., tendons, aponeuroses, plus supporting and binding tissues of all sorts. Like other cells in the body, fibroblasts carry an entire complement of genetic material. However, only a small percentage of the genes contained in fibroblasts are biologically functional; that is, most of the genes in fibroblasts are not expressed at all or are expressed at such low levels that the proteins they encode are produced in undetectable amounts or at concentrations which are not biologically functional or significant. Using routine methods of molecular biology it is now possible to introduce exogenous genetic material (i.e., replacement genes) into mammalian cells, thus enabling them to express genetic materials not normally expressed. The transduced fibroblasts of the present invention incorporate exogenous genetic material, which they express, thereby producing the gene products encoded by the incorporated exogenous genetic material.
As used herein, a promoter is a specific nucleotide sequence recognized by RNA polymerase, the enzyme that initiates RNA synthesis. When exogenous genes are introduced into fibroblasts using a retroviral vector, the exogenous genes are subject to retroviral control; in such a case, the exogenous genes) is transcribed from an endogenous retroviral promoter. It is possible to make retroviral vectors that, in addition to their own endogenous promoters, 131246' have exogenous promoter elements which are responsible for the transcription of the exogenous gene(s). For example, it is possible to make a construct in which there is an additional promoter that is modulated by an external factor or cue, and in turn to control the level of exogenous protein being produced by the fibroblasts by activating the external factor or cue.
As an illustration, the promoter for the gene which encodes the metal-containing protein metallothionine is responsive to Cd++ ions. Incorporation of this promoter or another promoter influenced by external cues makes it possible to regulate the production of the proteins produced by the transduced fibroblasts of the invention.
As used herein, subcutaneously means below the basement membrane of the epidermis. Subcutaneous is abbreviated as "s.c.".
As used herein, "i.p." means intraperitoneally.
As used herein, skin fibroblasts are fibroblast cells that are normally found in the dermis portion of the skin.
As used herein, syngeneic means isogenic, i.e., having the same genetic constitution.
As used herein, exogenous genetic material means DNA or RNA, either natural or synthetic, that is not naturally found in cells of a particular type: or if it is naturally found in the cells, it is not expressed in these cells in biologically significant levels. For example, a synthetic or natural gene coding for human insulin would be exogenous genetic material to a yeast cell since yeast cells do not naturally contain insulin genes: a human insulin gene inserted into a skin fibroblast cell would also be an exogenous gene to that cell since skin fibroblasts do 134r'~246 not express human insulin in biologically significant levels.
As used herein, "exogenous" genetic material and "foreign" genetic material mean the same thing, and the terms "exogenous" and "foreign", when used to describe genes or genetic material, are used interchangeably.
A~; used herein, retroviral vectors are the vehicles used to introduce replacement genes into the skin fibroblasts. The following paragraphs contain some general background information about retroviruses.
Retrovirus are RNA viruses; that is, the viral genome is RNA. This genomic RNA is, however, reverse transcribed into a DNA intermediate which is integrated very efficiently into the chromosomal DNA of infected cells. This integrated DNA intermediate is referred to as a prom rus.
The retroviral genome and the proviral DNA have three genes: c~aq, pol and env, which are flanked by two long terminal repeat (LTR) sequences. The ,gag, gene encodes the internal structural (nucleocapsid) proteins; the pol gene encodes the RNA-directed DNA
polymerase (reverse transcriptase); and the env gene encodes viral envelope gylcoproteins. The 5'and 3' LTRs serve to promote transcription and polyadenylation of virion RNAs.
Adjacent to the 5' LTR are sequences necessary for reverse transcription of the genome (the tRNA
binding site) and for efficient encapsidation of viral RNA into particles (the ~ or psi site). (Mulligan, 1983; Mann et al., 1983; Verma, 1985.) The various elements required for replication of the retrovirus can be divided into cis- and trans-acting factors. The trans-acting factors include 13~t 2~g proteins encoded by the viral genome, which are required for encapsidation of viral RNA, entry of virions into cells, reverse transcription of the viral genome, and integration of the DNA form of the virus (i.e., the provirus) into host DNA. The cis-acing factors include signals present in the viral RNA which interact with the shove-described proteins and other factors during virus replication.
If the sequences necessary for encapsidation (i.e., packaging of retroviral RNA into infectious virions) are missing from the viral genome, the result is a cis defect which prevents encapsidation of genomic RNA. The resulting mutant is however still capable of directing synthesis of all virion proteins. When the packaging signals are removed, viral RNA and proteins are still synthesized, but no infectious particles are made because viral RNA cannot be packaged into virions.
Mann, et al., (1983) used this strategy to create ~2 cell lines which supported the generation of infectious transducing retroviruses without generating helper murine leukemia viruses. Unfortunately, murine leukemia virus env gene product is only able to infect rodent cells, which limits the utility of ~2 cell lines. On the other hand, amphotropic murine retroviruses are able to infect a wide variety of cell types, including human cells.
Using a strategy similar to the one described by Mann, et al., for the production of the ~2 cell lines, Verma and his colleagues generated a cell line using the env gene product of the amphotropic viruses (Verma, 1985; Miller, et al., 1985: Miller , et al., 1986). As a result of this work, a wide-host-range, packaging defective system was made available for the generation of high-titer retroviruses containing 1341 2~6 exogenous genes (Verma, 1985; Miller, et al., 1985;
WO 86 00922. Such retroviruses, or retroviral vectoxs, have general utility for high-efficiency transduction of genes in cultured cells, and specific utility for use in the method of the present invention.
SUMMARY OF THE INVENTION
The present invention discloses a new gene therapy method based on the use of transduced fibroblasts that are implanted in the loose connective tissue of tl-ie skin of the subject to be treated.
According to the invention, transduced fibroblasts are preferably created by infecting fibroblast cells in vitro with chimeric retroviruses that contain at least one functionally active "replacement gene", i.e., foreign or exogenous genetic material that does not normally occur in fibroblast cells, or if it does, is not expressed by the fibroblast cells in biologically significant concentrations. The transduced fibroblasts are then preferably fixed by culturing them in vitro in an extracellular matrix. Finally, the transduced fibroblasts are implanted subcutaneously in the loose connective tissue of the skin of the individual or animal being treated. To insure rapid vascularization of the implanted fibroblasts, an angiogenic substance such as fibroblast growth factor is preferably placed in the loose connective tissue along with the implant.
Because the fibrobiasts are implanted in a highly vascularized compartment of the skin i.e., loose connective tissue of the dermis, the transduced cells, and thus their "replacement" gene products, have direct access to the circulatory system. As a result, the needed replacement gene products can easily and efficiently be distributed to other parts of the body.

When the gene therapy is no longer needed, the implanted fibroblasts can be conveniently removed.
DETAILED DESCRIPTION OF THE INVENTION
The present invention is based on the discovery that transduced skin fibroblasts can be used for somatic cell gene therapy when the transduced fibroblasts are fixed in vitro in an extracellular collagen matrix and implanted in the loose connective tissue of the dermis of an subject to be treated. The discovery makes it possible to overcome several problems that have been encountered when prior art gene therapy methods were used to treat animals or individuals with genetic defects. Such problems include: (1) inefficient expression of the foreign "replacement" genes (Williams, et al., 1984; Joyner, et al., 1985); (2) use of transduced cells that had the potential to be tumorigenic to the animal or individual being treated (Selden, et al., 1987; Garver, et al., 1987b); (3) use of harsh immunosuppressive agents to avoid rejection by the animal or individual being treated (Selden, et al., 1987); (4) necrosis following subcutaneous injection of cells (Bell, et al., 1983); and (5) poor diffusion of the replacement gene product (Morgan, et al., 1987). As will be discussed more fully below, the present invention preferably employs chimeric retroviruses to introduce replacement genes into skin fibroblasts. Because of the nigh efficiency of retroviral proviral infection and expression in fibroblasts, the present invention invention essentially eliminates the need to use marker genes to identify transduced cells. This greatly simplifies the overall problem of introducing replacement genes into cells that will be used for gene therapy. Since the invention preferably uses ~3~~ z~s fibroblast cells from recipient individuals, it obviates the need to use potentially tumorigenic cell lines. Use of skin fibroblasts from the subject to be treated minimizes the possibility of rejection, which in turn lessons the need for harsh immunosuppressant drugs. In addition, since the invention uses transduced fibroblasts that preferably have been fixed in vitro in an extracellular collagen matrix, the problem of necrosis is also minimized. Finally, since the invention implants the transduced fibroblasts into the highly vascularized loose connective tissue of the dermis, the replacement gene products are easily and efficiently distributed to other parts of the body.
With regard to the fibroblasts employed herein, preferably they will be skin fibroblasts from the animal or individual to be treated with the gene therapy. Fibroblasts from these subjects can easily be obtained by skin biopsy, and then maintained in culture until it is convenient to transduce them. General methods for maintaining fibroblast cells in culture are well known to those skilled in the art of tissue culture. Such methods include culturing the cells in Dulbecco-Vogt modified Eagle's medium with loo fetal bovine serum. Such known methods can be used by the skilled artisan, without undue experimentation, to culture the fibroblast cells prior to transduction.
See generally, the Materials and Methods sections of Palmer, at al., (1987).
Exogenous genetic material or genes especially useful in the invention are those genes that encode secretory proteins. Such useful genes include, but are not limited to, genes that encode blood clotting factors such as human factors VIII and IX: hormone genes such as the genes encoding for insulin, 134 2~6 parathyroid hormone, lutenizing hormone releasing factor (LHRH), alpha and beta seminal inhibins, and human growth hormone; enzyme genes: genes encoding inhibitor substances such as alphal-antitrypsin, and genes encoding substances that function as drugs, e.g., genes encoding the diphtheria and cholera toxins.
(Genes that encode useful "gene therapy" proteins, e.g., many enzyme proteins, that are not normally secreted can be used in the invention if they are "functionally appended" to a signal protein sequence that will "transport" them across the fibroblasts' limiting membranes and into the extracellular space. A
variety of such signal sequences are known and can be used by those skilled in the art without undue experimentation.) It is possible to use vehicles other than retroviruses to genetically engineer the fibroblasts of the present invention. However, chimeric retroviruses are the preferred agents used to incorporate new genetic material into the skin fibroblasts.
Retroviruses and helper-free replication-defective viral vectors are well known and can be adapted for use in the present invention without undue experimentation.
Examples of such retroviruses are disclosed in the Experimental Section of present specification:
additional examples are disclosed and discussed in Palmer, et al., (1987): Miller, et al., (1986); St.
Louis and Verma (1987): Miller et al., (1985): and in WO 86 00922 which has been assigned to the Salk Institute for Biological Studies, San Diego, CA. Helper-free replication-defective viral vectors that have a dominant selectable marker such as the neomycin resistance gene or the mutant DHFR gene (Miller, et al., 1985) can be used in 13 ~ ~ 2 ~i ~
the present invention. However, since the efficiency of retroviral infection is about 80%, use of a dominant selectable marker to identify transfected cells is usually not necessary.
It is possible, through the use of a recombinant retrovirus to introduce new genetic material into fibroblasts without altering the functional characteristics of the recipient fibroblasts. Therefore, retroviral vectors useful in the method of the present invention will have a cloning site. The presence of such a site makes it possible to introduce exogenous genetic material into the vector and have it expressed by fibroblasts co-cultivated with the recombinant virus. Methods for introducing exogenous genetic material into the retroviral vectors are known and can be used by the skilled artisan without undue experimentation. For example, useful methods are disclosed in the Experimental Section of this specification and in Palmer, et al, (1987). (In Palmer, et al., 1987, see especially the Materials and Methods section of the publication.) Additional methods and helpful details are disclosed in Verma, (1985); Miller, et al., (1985); Miller, et al., (1986);
St. Louis and Verma, (1987), and in WO 86 00922 which has been assigned to the Salk Institute for Biological Studies, San Diego, CA.
A cell line that produces recombinant amphotropic chimeric retrovirus is used in co-cultivation with the fibroblasts to be "transduced".
The dam cell line, which can be modified using standard techniques to include chimeric retrovirus, is available from the American Type Culture Collection, Rockville, Maryland. For example, a dam line which produces a B

13~1~24s chimeric retrovirus can be constructed as follows: the exogenous gene or cDNA of interest is ligated into a cloning site in the retroviral vector. (Such a vector could also carry a selectable marker such as the Neo gene.) Chimeric retroviruses that carry the exogenous gene or DNA of interest are isolated and transfected into ~Uam, cells. ~Uam cells that produce the chimeric virus construct are isolated, e.g., as 6418 resistant colonies if the chimeric retrov:irus carried the Neo gene as a selectable marker.
Co-cultivation methods are well known to those skilled in the art and can be used in the present invention without undue experimentation. Generally, the methods can be summarized as follows: On day one, fibroblast cells to be "infected" with chimeric retrovirus are seeded in conventional culture medium at approx. 5 x 106 cells per 60-mm culture dish. On day two, the culture medium is replaced with medium from cells that produce chimeric retrovirus. On day three, the infected fibroblasts are suspended with an enzyme such as trypsin. (Although it would not usually be necessary due to the high efficiency of bulk infection, if the chimeric retrovirus carried a selectable marker, the fibroblast cells would be grown in culture dishes containing selective media. Resistant colonies (i.e., those formed from cells that have been transduced by the chimeric retrovirus) would then be scored after an appropriate amount of time (10-12 days). Fibroblasts from the resistant colonies would contain the new genetic material carried by the transducing chimeric retroviruses.) According to the invention, once the skin fibroblasts have been transduced with chimeric retrovirus, they are preferably "fixed" in vitro in an ~34~ 2~g extracellular matrix. See generally, Elsdale, et al., (1972) and Bell, et al., (1979). A preferred method for "'fixing" the transduced fibroblasts in vitro in an extracellular matrix is discussed in the Experimental Section of this specification. In summary, the fibroblasts are preferably fixed by culturing them in an extracellular matrix composed of collagen (either natural or synthetic) and culture medium. The cells are cultured at about 37° C. for about 3 days, during which time the collagen contracts to a tissue-like structure. Once contracted, the "artificial"
fibroblast tissue grafts can be implanted into the loose connective tissue in the dermis of the recipient subject. While the extracellular collagen matrix is preferred (since it is easy, inexpensive and effective), those skilled in the art will realize that other collagen-like materials, both natural and synthetic, could be used to generate the extracellular matrix into which the transduced fibroblasts become fixed.
To ensure rapid vascularization of the grafted implant, it is preferable to insert basic fibroblast growth factor along with each graft. The growth factor can be conveniently supplied by first applying it to a piece sterile sponge, e.g., as gelfoam (Upjohn), which is then implanted in the connective tissue along with each graft.
The present invention makes it possible to genetically engineer skin fibroblasts that can secrete a variety of useful gene products (e. g., clotting factors, immunoregulatable factors, hormones and drugs). when these transduced fibroblasts are implanted into the dermis of an individual or animal, the secreted gene products diffuse into the bloodstream, and thus are carried to various parts of the body.
The implanted transduced fibroblasts of the present invention can be used in a variety of applications. For example, the implanted fibroblasts can serve as a continuous drug delivery system to replace present regimes that require periodic administration (by ingestion, injection, etc.) of a needed substance. In another example, the transduced and implanted fibroblasts could be used to provide continuous delivery of insulin. This would be very useful since it would eliminate the need for daily injections of insulin. Genetically engineered fibroblasts can also be used for the production of clotting factors. Hemophiliacs lack a protein called Factor VIII, which is involved in blood clotting.
Factor VIII is now administered by injection. Like insulin, it could be made by transduced fibroblasts.
Transduced and implanted fibroblasts could also be used to deliver growth hormone.
Another application for transduced fibroblasts produced by the present invention is in fertility control. Several hormones, including lutenizing hormone releasing factor (LHRH) and the seminal and ovarian inhibins, are being studied for their ability to regulate fertility. Continuous administration of LHRH results in a sterile individual: yet when administration of the hormone is stopped, fertility returns. Rather than taking LHRH injections or oral medication, one could implant collagen fixed fibroblasts carrying the LHRH gene, and thus provide a continuous supply of the hormone.
Transduced fibroblasts having foreign or exogenous genetic material introduced according to the present invention can also be used for drug delivery, especially in animals. In this way certain drugs can be continuously delivered to the animals, thus eliminating the need to incorporate the drugs into the animals' food or water.
In each of the cited applications for the transduced fibroblasts of the present invention, the amount of replacement gene product delivered to the subject can be controlled by controlling such factors as: (1) the type of promoter used to regulate the replacement gene (e.g., use of a strong promoter or an weak one); (2) the nature of the promoter, i.e., whether the promoter is constitutive or inducible; (3) the number of transduced fibroblasts that are present in the implant; (4) the size of the implant; (5) the number of implants, (6) the length of time the implant is left in place, etc.
Without further elaboration, it is believed that one of ordinary skill in the art can, using the preceding description, and the following Experimental Section, utilize the present invention to its fullest extent. The material disclosed in the Experimental Section is disclosed for illustrative purposes and therefore should not be construed as limiting the appended claims in any way.
EXPERIMENTAL SECTION
ABSTRACT
Mouse primary skin fibroblasts were infected with a recombin~3nt retrovirus containing human factor IX cDNA. Bulk infected cells capable of synthesizing and secreting biologically active human factor IX
protein were embedded in collagen and the implant grafted under the epidermis. Sera from the transplanted mice contain human factor IX protein for at least 10-12 days. Loss of immunoreactive human factor IX protein in the mouse sera is not due to graft rejection. Instead, the mouse serum contains anti-human factor IX antibodies, which react with the protein.
EXPERIMENTAL METHODS
Construction and infection by recombinant factor IX retroviruses The recombinant pAFFIXSVNeo is based on a retroviral construct pAFVXM generated by Kriegler, et al. (Kriegler, et al., 1984). A human factor IX cDNA
was linked directly to the 5' long terminal repeat (LTR) by inserting a 1.6 kilobase kb BamHI/ HindIII
fragment from the clone CVI between the BctlII and HindIII sites of pAFVXM (Anson, et al., 1984). The entire expression unit from the neomycin phosphotransferase expression plasmid (pKoNeo) was excised by partial HindIII digestion and inserted into the HindIII site of the above factor IX viral construct (Fig. lA). "Helper free" recombinant ecotropic virus in ~2 cells was generated as described (Miller, et al., 1986; Mann, et al., 1983). The titres of recombinant retrovirus expressed from drug resistant clones were done essentially as described (Miller, et al., 1986).
Primary mouse embryo fibroblasts (MEF) were obtained from day 17 embryos of C578L/6J mice (Todaro, et al., 1963). The BL/6 line is an immortalized skin cell line derived from x-ray irradiated skin fibroblasts obtained from C57BL/6J mice. The skin fibroblast cell line BL/6, and NIH3T3 TK- cells were infected with recombinant retroviruses from the cell line, ~UFIXNeo 4, at a multiplicity-of-infection (moi) of 1-2 in the presence of POLYBRENE at 8 ug/ml: MEF

~~~~ ~~s cells were infected at a moi of 5.
Implantation of infected mouse fibroblasts in mouse Infected BL/6 and FIEF cells were cultured in vitro in an extracellular matrix composed of rat tail type I collagen (1 mg/ml: Sigma) and Dulbecco's modified Eagle's medium (DIEM) supplemented with 10%
fetal bovine serum in a 5-cm dish (Elsdale, et al., 1972; Bell, et al., 1979). The cells were cultured at 37°C for 3 days during which the collagen lattice contracted to a tissue-like structure (1/25th the area of the original gel). Once contracted, two artificial tissues containing approximately 4 x 106 infected fibroblasts were grafted into the loose connective tissue of the d'rmis in the midback of a recipient C57BL/6 mouse. To ensure rapid vascularization of the grafted tissue, a 2-mm2 piece of gelfoam (Upjohn) containing 2 ug of basic fibroblast growth factor was inserted into the loose connective tissue along with each graft. Serum samples were drawn at two day intervals and analyzed for the presence of human factor IX by ELISA.
Analysis of secreted factor IX
Levels of antigenic factor IX were assayed by ELISA a~ described by Anson, et al., (1987).
Biologically active human factor IX was immunoaffinity purified using A7 antibody (Anson, et al., 1987; Smith, et al., 1987). The amount of biologically active protein was determined by a one step clotting assay using canine fa~~tor IX deficient plasma (Goldsmith, et al., 1978). This assay is based on the ability of the sample to decrease the prolonged activated partial thromboplastin time of congenital factor IX-deficient plasma. Purified human factor IX was used as a control.

1341 ~~46 RESULTS
Transduction of Neomycin Resistance and Expression of Human Factor IX
The titres of helper-free ~ FIXNeo virus produced in the various cell lines ranged from 3 x 105 to 7 x 105 6418 resistant colony forming units per ml when assayed by NIH3T3 TK- cells. As measured by ELISA, all of the virus producing cell lines secreted essentially the same levels of factor IX into the culture media (approx. 200 ng/ml). All infected and drug resistant cell lines were also found to secrete factor IX into the culture media, albeit at different levels (see Fig. 2).
The organization of the integrated recombinant retrovirus in the virus producing cell line was determined by Southern blot analysis of SstI digested (SstI cleaves once in each LTR to generate a 5.1-kb DNA
fragment (Fig. 1B) genomic DNA. All infected cells displayed a single band of the expected size of approximately 5.1 kb which hybridizes to both the factor IX cDNA and the Neo probe, therefore ruling out any detectable rearrangements. Furthermore, the size of this band in infected NIH3T3 TK , BL/6, and MEF
cells is identical to that found in the virus producing cell line ~FIXNeo 4 (compare lanes 5 and 9 to other lanes).
The RNA blot analysis of the RNA isolated from ~FIXNeo 4, infected NIH3T3 TK , BL/6 and MEF is shown in Figure 1C. When hybridized to factor IX probe, only one major transcript of the expected size of 5.1 kb, corresponding to full length viral RNA could be detected in the infected cells. Hybridization with Neo probe reveals an additional 2.2 kb transcript that is the predicted size of the mRNA species, the synthesis 1341 24~
of which is initiated from the simian virus 40 early promoter and is terminated in the 3' LTR (Fig. 1C).
Ratios of the steady state levels of the 5.1 kb and the 2.2 kb transcripts varied within the different infected cell types. From results shown in Figure 1, it was concluded that the ~FIXNeo recombinant retrovirus is properly integrated and expressed in the infected cells.
Secretion of Factor IX Protein Because human factor IX is a secretory protein it was important to verify if it is secreted into the medium of the infected cells. Figure 2 shows that both rate and extent of antigenic factor IX released into the medium is dependent on the cell type rather than on the relative amounts of the factor IX transcripts. For instance, steady state levels of factor IX transcript in infected NIH3T3 TK- cells is much higher than in BL/6 cells (Fig. 1C); yet the rate and amount of factor IX secreted in the latter cell type is much higher.
Both the virus producing cell line y~F'IXNeo 4 and infected skin fibroblast cell line BL/6, secreted antigenic factor IX at similar rates, approximately 5.7 ng per ml/hr for 3 x 106 cells and 5.0 ng per ml/hr for 3 x 106 cells, respectively. This rate was almost 3 fold higher than the rate of factor IX secretion seen for infected MEF (1.75 ng per ml/hr for 3 x 106 cells) and infected NIH3T3 cells (1.65 ng per ml/hr for 3 x 106 cells). These results indicate that the rate of synthesis and/or secretion may be a property of the cell type, rather than the levels of expression.
Secreted Human Factor IX Protein Is Biologically Active The primary translation product of factor IX
gene undergoes extensive post-translational modification which include addition of sialic carbohydrates (Chavin, et al., 1984; Fournel, et al., 1985), vitamin K-dependent conversion of glutamic acid residues to Ycarboxy/glutamic acid (Suttie, 1980) and ~-hydroxylation of aspartic acid residue 64 (Ferlund, et al., 2983). The Ycarboxylation of factor IX is essential for clotting activity and this modification generally occurs in the liver, the primary source of factor IX synthesis in the body. Two different approaches were taken to assess biological activity of human factor IX secreted from cells in culture: (i) The infected mouse embryo fibroblasts were cultured in factor IX deficient canine serum obtained from hemophiliac dogs, supplemented with epidermal growth factor (10 ng/ml) and vitamin K (25 ng/ml). Media harvested after 48 hr incubation was monitored for activity by a one step assay (Goldsmith, 1978).
Conditioned media from MEF cells contained biologically active human factor IX at 210 ng/ml which is similar to the levels seen with ELISA assays (Fig. 4). (ii) Because BL/6 cells did not attach to the tissue culture dish in canine sera, a different approach had to be resorted to. Infected BL/6 cells were grown in 10%
total calf serum supplemented with vitamin K (25 ng/ml), and the media harvested after 48 hr incubation was applied to an immunoaffinity column containing human factor IX monoclonal antibody A-7 (Anson, et al., 1987; Smith, et al., 1987). This monoclonal antibody recognizes the calcium binding domain of human factor IX protein, thus discriminating between carboxyl-lacking factor IX and biologically active carboxyl human factor IX. One-hundred and sixty ml of the media obtained from BL/6 cells containing 32 ~,g of antigenic human factor IX (determined by ELISA) was passed ~3~~ 2~s through the column. Nearly 3.5 ug of the biologically active material was recovered from the column. This represents over 10% of the total antigenic factor IX in the starting sample. No biologically active factor IX
could be identified from uninfected MEF or BL/6 cells.
Despite lack of information on the extent of carboxylation or other post-translational modifications, it is concluded that the infected cells used for subsequent implantation studies synthesize biologically active human factor IX.
Detection of Human Factor IX In Mice Grafted With Infected Fibroblasts Infected MEF cells and BL/6 cells were cultured in an extracellular matrix, composed of collagen, before grafting. A tumor cell line, BL/6, was chosen in addition to MEF because it has an advantage in growth and vascularization and thus would increase our chances of detecting secreted factor IX in the sera.
Attachment of the cells to the collagen resulted in a 3 dimensional array of cells stacked on top of one another. After the primary fibroblast cells (MEF) or the tumor cell line BL/6 contracted in the collagen gel, the cells were grafted into the loose connective tissue of the mid-back dermis of a recipient syngeneic C57BL/6 mouse (Fig. 3A). Figure 3B shows that the inserted implants were extensively vascularized by day 14. A similar extent of vascularization was also detected in 2~ day implants (data not shown).
The serum levels of the human clotting factor were measured in engrafted mice by ELISA over a 34 day period. Figure 4 shows that the average levels of human factor IX in 3 mice progressively increased from 20 ng/ml at day 2 to a peak of 97 ng/ml 7 days after grafting the BL/6 cells into the mice. The 4 mice ~3~~ Z~s grafted with the infected MEF fibroblasts showed a similar pattern of increase in which an average peak of 25 ng/ml of factor IX was detected at day 9. This rise was followed by a rapid decline to near non detectable levels of serum human factor IX at day 16 in both the BL/6 and MEF grafts. A minor peak of factor IX was seen at day 20 in mice with either graft, which was followed by loss of any detectable factor IX antigen.
In parallel experiments, 10~ infected BL/6 or MEF cells were injected directly into the peritoneal cavity of the recipient C57BL/6 mice. Serum levels of human factor IX in th_= injected animals followed a similar profile as that seen with the grafts (data not shown).
EX~lanted Grafts Make Factor IX
The decline in serum levels of antigenic human factor IX in animals that were either grafted or injected i.p. was not associated with the necrosis of cells in the grafts. BL/6 cells in the collagen matrix grew as an aggressive tumor at the site of the graft.
The tumor continued to grow until the animals were sacrificed at day 32. Mice with grafts containing infected MEF were visibly vascularized upon gross inspection until day 28, however, by day 120 the extent of vascularization was reduced but the implant was still viable (data not shown). Additionally, cells explanted at various times during the course of the experiment praduced factor IX when grown in culture (Table I). The explanted BL/6 cells grew well in culture and secrete antigenic factor IX at levels similar to that before grafting. The MEF cells explanted from the grafts at days 14 and 21 grew well in culture, but produced slightly lower levels of factor IX. Ce?~.s explanted at day 28 did not grow well, and the low level of factor IX secreted from these cells is perhaps a consequence of this poor growth.
Detection of Serum Anti-Factor IX Antibodies To further investigate the decline of serum levels of human factor IX it was reasoned that the recipient animal mounted an immunological response against the highly immunogenic human factor IX protein.
To test whether mice bearing grafts with infected BL/6 or MEF cells are generating anti-factor IX IgG
antibodies, pooled serum samples were used to probe immunologic blots containing purified human factor IX
protein (Figure 5). The levels of anti-human factor IX
IgG antibodies were not detectable in mice with MEF
grafts at day 7 to day 21. Slightly higher levels of serum antibodies were detected in mice with BL/6 grafts during this period, presumably because they are releasing more factor IX. Maximum levels of anti-human factor I~; antibodies were detected at day 28 in mice with either graft. The mice with BL/6 grafts exhibited the highest level of xeno-antibodies. Pooled serum drawn from mice 28 days after i.p. injection with infected MEF also showed anti-factor IX IgG antibodies albeit at much lower levels. Naive animals which have not been exposed to infected BL/6 or MEF cells do not make anti-human factor IX antibodies. These observations world suggest that human factor IX is continuously produced in grafted mice but is not detectable due to a large pool of mouse anti-human factor IX antibodies.
DISCUSSION
This example presents the development and characterization of a different approach of gene product delivery into an animal model system. The BL/6 cells and MEF cells infected with a helper free 1347 24~
recombinant retroviral vector containing the human clotting factor IX cDNA secrete partially biologically active clotting factor at a rate 10 fold higher than seen with another retroviral vector containing the human clotting factor cDNA (Anson, et al., 1987). In addition, the example demonstrates that those genetically modified cells can be reintroduced into the loose connective tissue of the dermis of a syngeneic mouse. Grafts are quickly vascularized in the presence of angiogenic factor, fibroblast growth factor, and remain vascularized for at least 28 days. Grafts containing the BL/6 cells grow as aggressive tumors over this period while the size of the grafts containing the MEF cells does not increase over the same period. The clotting factor secreted from the infected cells in the graft is accessible to the circulatory compartment and can easily be detected in serum of the recipient. Functional status of the infected cells in the grafts can be measured by monitoring serum levels of human factor IX or by the ability of explanted cells to continue secreting the human protein. However, C57BL/6 mice recognize the human blood clotting factor as foreign and thus mount a strong humoral immune response against it. Although a humoral response against factor IX clearly exists, there does not appear to be a major cell mediated response against the cells in the grafts. The cells in the graft are still viable after 28 days of implantation and continue to synthesize factor IX
protein.
Even though the data presented here was obtained from mouse embryo fibroblasts, it should be noted that the observations have been extended by infecting adult hemophiliac dog fibroblasts with factor 1 3 4 ~ 2 46 IX retrovirus.
It should also be noted that in normal individuals, levels of factor IX protein are approximately 5 ug/ml of plasma. Although the levels reported here are lower by several orders of magnitude, it should be remembered that individuals containing 0.5 ~.g of biologically active factor IX per ml in plasma do not show the symptoms of hemophilia. The low levels of factor IX can be increased either by making improved vectors capable of generating large amounts of factor IX proteins or, alternatively, by grafting more cells.
According to the data presented here, up to 25 ng of factor IX per hr can be generated from an implant containing 4 x 106 cells (Fig. 4). In larger animals multiple grafts of up to 108 cells can be easily implanted, increasing the levels of factor IX protein to that required to alleviate the deficiency.
Culturing infected cells in a defined medium (without fetal bovine serum) and improved technology for reconstitution of living skin would also increase the efficiency of the system (Bell, et al., 1983).
Moreover, impro~,red surgical skills may ensure that the implant would lay flat in the dermal compartment of the mouse skin to allow more uniform vascular development and hence impro-~e cell viability during the brief period required for vascularization. Although the extent of cell viability has not yet been determined in grafts containing MEF cells, experiments in rats have shown that transplanted fibroblasts persist for at least 13 months (Bell, et al., 1983).
In conclusion, this example has shown that skin fibroblasts can be used as a viable mode of introduction and expression of foreign genes in mammals. The process of manipulation of genetically engineered fibroblasts appears to be both less complex and cumbersome than the widely accepted use of bone marrow transplantation far somatic cell gene therapy.
DETAILED DESCRIPTION OF THE FIGURE LEGENDS
Figure 1. Analysis of recombinant human factor IX re~rovirus. (a) pAFFIXSVNeo. Arrows indicate transcripts that initiate at either the promoter located in the 5' LTR, or the simian virus 40 early promoter located between the two LTRs, and terminate at the polyadenylation signal in the 3' LTR. Bars indicate the putative initiation site of transcription.
The restriction endonuclease cleavage sites SstI, HindIII, BamHI, BalII and Clal are diagnostic sites used during the construction of the vector or subsequent characterization of the provirus in the genome of infected cell lines. (b) Proviral DNA.
Genomic DNA isolated from either uninfected or infected ~FIXNeo ~, NIH3T3 TK , BL/6, MEF cells, and from the virus producing cell line ~FIXNeo 4 was digested with Sstl fractionation by agarose gel electrophoresis, transferred onto a nitrocellulose membrane and hybridized to either a nicktranslated 1.6 kb factor IX
cDNA probe (lanes 1-8) or 1.4 kb HindIII to BamHI Neo DNA probe (lanes 9-12). Under these conditions of hybridization, human factor IX cDNA does not hybridize to mouse DNA. (c) RNA transcripts. Total RNA (10 ug) isolated from uninfected and infected cells and cell lines was subjected to RNA blot analysis.
Figure 2. Secretion of Human Factor IX. Rate of secretion of human factor IX by the virus producing cell line ~FIXNeo 4 (shown by open squares) and by infected NIH3T3 TK- cells (shown by solid squares), BL/6 cells (shorn by solid diamonds), and MEF cells (shown by open diamonds). Cells were seeded at 3 x 106 13 4 '~ ~ ~i !~
cells per 5 cm dish in 4 ml of medium. At each indicated time point 100 ul of medium was removed and assayed for human factor IX by enzyme linked immunoabsorbant assay (ELISA) (Anson, et al., 1987).
The mouse anti-human monoclonal antibody, FXC008, generated by Bajaj, et al. (Bajaj, et al., 1985) was used as the primary antibody, whereas pooled normal human sera were used as a standard. Each time point was done in triplicate and thus represE:nts an average amount of factor IX secreted over a 48 hr period.
Curves were corrected for the slight increase in cell number over this period.
Figure 3. Embedding and Implantation. (A) Schematic representation of the protocol used to generate and graft the collagen implants into the loose connective tissue of the skin of the mouse model system. (B and C) View of both MEF (B) and BL/6 (C) collagen implants at day 14. The grafts (white area) and the high degree of vascularization are clearly visible. The implant is 0.75 cm in diameter. FGF, fibroblast growth factor.
Figure .~. Factor IX in Human Sera. Average (four mice for IvEF; three mice for BL/6) amount of human factor IX detected in the sera of mice that received two collagen implants containing approximately 4 x 106 cells each. Sera was drawn from each animal at the indicated times. Levels of circulating human factor IX were determined by ELISA: levels of factor IX
fluctuated 2- to 3-fold between experiments.
Figure 5. Detection of mouse anti-human factor IX antibodies in mice grafted with collagen implants.
Purified human factor IX was subjected to PAGE under denaturing conditions and then transferred onto nitrocellulose as described (Towbin, et al., 1979).

The nitrocellulose strips were treated with blocking solution for 2 hr followed by 1:100 dilution of naive normal mouse serum; 1/100 dilution mouse monoclonal anti-factor IX antibody FXC008 (lane 2); 1:100 dilution of serum from mouse harboring grafts containing infected MEF cells drawn at day 7, day 14, day 20, and day 28 (lanes 3-6); 1:100 dilution of serum from mouse harboring grafts containing infected BL/6 cells drawn at day 7, day 15, day 21, and day 29 (lanes 7-10).
After overnight incubation at 37°C the strips were washed, incubated with 1251-labeled goat anti-mouse IgG
antibody, and then subjected to autoradiography as described (Glenney, 1986).

1341 2~6 Table I. Amount of antigenic factor IX secreted from cells explanted from grafts. Tissue explanted from the grafts at times indicated after implantation were cultured in vitro. When cells were confluent medium was replaced; after 48 hr, levEls of secreted factor IX
secreted into the culture were assayed by ELISA.
Collagen Days after explants,, n Implantation BL/6 MEF

1341 24g _ REFERENCES CITED IN THE SPECIFICATION
The following patent and journal publications are referred to in the specification.
JOURNAL PUBLICATIONS
1. Anson, D.S., Choo, K.H., Rees, D.J.G., Gianelli, F., Goald, K., Huddleston, J.A., & Brownlee, G.G. (1984), EMBO J. 3:1053-1060.

2. Anson, D.S., Hock, R.A., Austen, D., Smith, K.J., Brownlee, G.G., Verma, I.M., & Miller, A.D. (1987), Mol. Biol. Med.

4:11-20.

3. Bajaj, S.P., Rapaport, S.I., & Maki, S.L.

(1985), J. Biol. Chem. 260:11574-11580.

4. Bell, E., Ivarsson, B., & Merrill, C.

(1979), Proc. Natl. Acad. Sci. USA

76:1274-1278.

5. Bell, S., Sher, S., Hall, B., Merrill, C., Rosy=_n, S., Chamson, A., Asselinean, D., Dubertret, L., Coulomb, B., Capiere, C., Nusgens, B., and Nevewe, Y. (1983), J.

Invest. Dermatol. 81:25-103.

6. Cha~~in, S.I. & Weidner, S.M. (1984), J. Biol. Chem. 259:3387-3390.

7. Dick, J.E., Magli, M.C., Huszar, D.H., Phillips, R.A., & Bernstein, A. (1985), Cel:L 42:71-79.

8. Elsdale, T. & Bard, J. (1972), J. Cell.

Biol. 54:626-637.

9. Fer:Lund, P. & Stenflo, J. (1983), J. Biol.

Chem: 258:12509-12512.

X341 ~~~

10. Fournel, M.A., Newgren, J., Madanat, M. &
Pancham, N. (1985), Thromb. Haemostasis 54:147.
1.1. Carver, R.I., Jr., Chytil, A., Courtney, M., & Crystal, R.G. (1987b), Science 237:762-764.
12.. Carver, R.I., Jr., Chytil, A., Karlsson, S., Fells, G.A., Brantley, M.L., Courtney, M., Kantoff, P.W., Nienhuis, A.W., Ander-son, W.F., & Crystal, R.G. (1987a), Proc.
Natl. Acad. Sci. USA 84:1050-1054.
13. Glenney, J. (1986), J. Anal. Biochem.
156:315-319.
14. Goldsmith, J.C., Chung, K.S., and Roberts, H.R. (1978), Thromb. Res. 12:497-502.
15. Kelley, G., Paige, C., Gilboa, E., & Wag-ner, E.F. (1985), Nature jLondon) 318:149-154.
16. Kriegler, M., Perez, C.F., Hardy, C., &
Botchan, M. (1984), Cell 38:483-491.
17. Joyner, A., Kelley, G., Phillips, R.A., &
Bernstein, A. (1985), Retrovirus transfer of a bacterial gene into haematopoietic progenizot cells. Nature 305:556.
18. Ledley, F.D., Darlington, G.J., Hahn, T., &
Woo, S.L.C. (1987), Proc. Natl. Acad. Sci.
USA 84:5335-5339.
19. Ledley, F.D., Grenett, H.E., McGinnis-Shelnutt, M., ~r WOO, S.L.C. (1986), PrOC.
Natl. Acad. Sci. USA 83:409-413.
20. Mann, R., Mulligan, R.C., & Baltimore, D.
(1983), Cell 33:153-159.

~3412'~8~::
21. Miller, A.D. & Buttimore, C. (1986), Mol.
Cell. Biol. 6:2895-2902.
22. Miller, A.D., Eckner, R.J., Jolly, D.J., Fri~dmann, T., & Verma, I.M. (1984), Science 225:630-632.
23. Miller, A.D., Law, M-F., & Verma, I.M.
(1985}, "Generation of Helper-Free Am-photropoic Retroviruses That Transduce a Dominant-Acting, Methotrexate-Resistant Dihydrofolate Reductase Gene", Molec. Cell.
Biol-, 5 (3):431-437.
24. Mor~an, J.R., Barrandon, Y., Green, H., &
PMulligan, R.C. (1987), Science 237:1476-1479.
25. Mulligan, R.C. (1983), "Construction of Highly Transmissible Mammalian Cloning Vehicles Derived from Murine Retroviruses, in Experimental Manipulation of Gene Expression, M. Inouye (ed), p. 155-173.
26. Palmer, T.D., Hock, R.A., Osborne, W.R.A., b Miller, A.D. (1987), Proc. Natl. Acad.
Sci. USA 84:1055-1059.
27. Selden, R.F., Skoskiewicz, M.J., Howie, K.B., Russell, P.S., & Goodman, H.M.
(1987), Science 236:714-718.
28. Smith, K.J., Singaraju, C., & Smith, L.F.
(1987), Am. J. Clin. Pathol. 87:370-376.
29. Sorge, J., Kuhl, W., West, C., & Beutler, E. (1987), Proc. Natl. Acad. Sci. USA
84:906-909.
30. Suttie, J.W. (1980), CRC Crit. Rev.
Bio~~hem. 8:191-223.
31. Todaro, G.J. & Green, H.J. (1963), J. Cell.
Biol. 17:299-313.

1341 2~6 32. Towbin, H., Staehelin, T., & Gordan, J.
(1979), Proc. Natl. Acad. Sci.. USA
76:4350-4354.
33. Verma, I.M. (1985), Retroviruses for gene transfer. In Microbioloq_y-1985 (ed.
L.Leive et al.), p. 229. American Society for Microbiology, Washington, D.C.
34. Williams, D.A., Lemischka, I.R., Nathans, D.G., & Mulligan, R.C. (1984), Nature ~~London) 310:476-480.
35. Wolfe, J.A., Yee, J.K., Skelly, H.F., Moores, J.C., Respess, J.G., Friedmann, T., & Leffert, H. (1987), Proc. Natl. Acad.
Sci. USA 84:3344-3348.
PATENT PUBLICATIONS
1. WO 86 00922, Applicant: The Salk Institute For Biological Studies; Inventors: I.M.
Verma, A.G. Miller, and R.M. Evans; Title:
Retroviral Gene Transfer Vectors.
2. United States Patent 4,624,944, issued November 25, 1986 to the Regents of the University of California for "Human Seminal Alpha Inhibins".
SPECIFICATION SUMMARY
From the foregoing description, one of ordinary skill in the art can understand that the present invention is a new somatic cell gene therapy method. According to the invention, transduced fibroblasts are preferably created by infecting fibroblast cells in vitro with chimeric retroviruses that contain at least one functionally active "replacement gene". Such replacement genes can be 1341 24~~
either foreign genetic material that is not found in fibroblast cells, or native genetic material that is found in fibroblast cells but not normally expressed in biologically significant concentrations in these cells.
Since the invention uses transduced fibroblasts from the individual or animal to be treated, the possibility of rejection is minimized. In addition, since the invention implants the transduced fibroblasts in the highly vascularized loose connective tissue of the dermis, the transduced cells, and thus their "replacement" gene products, have direct access to the circulatory system. As a result the needed replacement gene products can easily and efficiently be distributed to other parts of the body. When the gene therapy is no longer needed, the implanted fibroblasts can be conveniently removed.
Since the fibroblasts can be transduced to express a variety of replacement genes, the method of the invention has many important applications for both humans and animals. For example, the method can be used to treat diseases caused by genetic defects, to deliver drugs to individuals and animals, and to ad-minister birth control hormones.
without departing from the spirit and scope of this invention, one of ordinary skill can make various changes and modifications to the invention to adapt it to various usages and conditions. As such, these changes and modifications are properly, equitably, and intended to be, within the full range of equivalence of the following claims.

Claims (24)

1. For implantation in the loose connective tissue of the dermis of a subject, a collagen matrix containing transduced subject-derived primary fibroblasts, wherein said transduced fibroblasts are infected with a recombinant retroviral vector that comprises exogenous genetic material encoding a gene product, and wherein said transduced fibroblasts express said gene product.

2. For implantation in the loose connective tissue of the dermis of a subject, the matrix-bound fibroblasts of claim 1 and an angiogenic substance.

3. For implantation in the loose connective tissue of the dermis of a subject, the matrix-bound fibroblasts and angiogenic substance of claim 2, wherein said angiogenic substance is fibroblast growth factor.

4. For implantation in the loose connective tissue of the dermis of a subject, the matrix-bound fibroblasts of claim 1, wherein said exogenous genetic material comprises at least one functionally active replacement gene.

5. For implantation in the loose connective tissue of the dermis of a subject, the matrix-bound fibroblasts of claim 4, wherein said functionally active replacement gene encodes at least one protein selected from blood clotting factors, hormones, enzymes, inhibitors or drugs.

6. For implantation in the loose connective tissue of the dermis of a subject, the matrix-bound fibroblasts of claim 5, wherein said blood clotting factor is factor VIII or factor IX.

7. For implantation in the loose connective tissue of the dermis of a subject, the matrix-bound fibroblasts of claim 5, wherein said hormone is selected from insulin, parathyroid hormone, luteinizing hormone releasing hormone (LHRH), human seminal and ovarian inhibins, or human growth hormone.

8. For implantation in the loose connective tissue of the dermis of a subject, the matrix-bound fibroblasts of claim 5, wherein said inhibitor is alphas-antitrypsin.

9. For implantation in the loose connective tissue of the dermis of a human subject, a collagen matrix containing transduced subject-derived primary fibroblasts, wherein said transduced fibroblasts are infected with a recombinant retroviral vector that comprises exogenous genetic material encoding a gene product, and wherein said transduced fibroblasts express said gene product.

10. For implantation in the loose connective tissue of the dermis of a human subject, the matrix-bound fi.broblasts of claim 9 and an angiogenic substance.

11. For implantation in the loose connective tissue of the dermis of a human subject, the matrix-bound fibroblasts and angiogenic substance of claim 10, wherein said angiogenic substance is fibroblast growth factor.

12. For implantation in the loose connective tissue of the dermis of a human subject, the matrix-bound fibroblasts of claim 9, wherein said exogenous genetic material comprises at least one functional replacement gene.

13. For implantation in the loose connective tissue of the dermis of a human subject, the matrix-bound fibroblasts of claim 12, wherein said functional replacement gene encodes at least one protein selected from blood clotting factors, hormones, enzymes, inhibitors or drugs.

14. For implantation in the loose connective tissue of the dermis of a human subject, the matrix-bound fibroblasts of claim 13, wherein said blood clotting factor is factor VIII or factor IX.

15. For implantation in the loose connective tissue of the dermis of a human subject, the matrix-bound fibroblasts of claim 13, wherein said hormone is selected from insulin, parathyroid hormone, luteinizing hormone releasing hormone (LHRH), human seminal and ovarian inhibins, or human growth hormone.

16. For implantation in the loose connective tissue of the dermis of a human subject, the matrix-bound fibroblasts of claim 13, wherein said inhibitor is alphal-antitrypsin.

17. For gene therapy, transduced, subject-derived primary fibroblasts contained in collagen matrix suitable for implantation in the loose connective tissue of the dermis of a subject, wherein said transduced fibroblasts are infected with a recombinant retroviral vector that comprises exogenous genetic material encoding a gene product, wherein said transduced fibroblasts express said gene product, and wherein expression of said gene product is under the control of a constitutive promoter.

18. For gene therapy, the matrix-bound fibroblasts of claim 17 and an angiogenic substance.

19. For gene therapy, the matrix-bound fibroblasts and angiogenic substance of claim 18, wherein said angiogenic substance is fibroblast growth factor.

20. For gene therapy, the matrix-bound fibroblasts of claim 17, wherein said exogenous genetic material comprises at least one functionally active replacement gene.

21. For gene therapy, the matrix-bound fibroblasts of claim 20, wherein said functionally active replacement gene encodes at least one protein selected from blood clotting factors, hormones, enzymes, inhibitors or drugs.

22. For gene therapy, the matrix-bound fibroblasts of claim 21, wherein said blood clotting factor is factor VIII or factor IX.

23. For gene therapy, the matrix-bound fibroblasts of claim 21, wherein said hormone is selected from insulin, parathyroid hormone, luteinizing hormone releasing hormone (LHRH), human seminal and ovarian inhibins, or human growth hormone .

24. For gene therapy, the matrix-bound fibroblasts of claim 21, wherein said inhibitor is alphal-antitrypsin.