NO20160498L - Primary, cultured adipocytes for gene therapy - Google Patents

Abstract

The present invention provides primary, cultured gene therapy adipocytes that stably carry a foreign gene encoding a protein secreted into the cell environment. More specifically, cells suitable for gene therapy are provided which can be used as a replacement for bone marrow cells or hepatocytes which have been used for ex vivo gene therapy. Thus, a method is established for transferring a foreign gene to primary cultured adipocytes suitable for ex vivo gene therapy, which can be easily collected and transplanted and which can be readily removed after transplantation. Specifically, primary, cultured adipocytes are established which are suitable for ex vivo gene therapy and which stably carry a foreign gene encoding a protein secreted into the cell environment.

Description

Field of the invention

The present invention relates to primary, cultured adipocytes for gene therapy to which one or more foreign genes have been transferred.

State of the art

Today's gene therapy (Toyooka et al., Folia Pharmacol. Jpn. 116:158-162, 2000) can be classified into two groups: (1) methods for transferring therapeutic genes to patients by direct administration of viral vectors, naked plasmids or the like as codes for the gene (in vivo), and (2) methods of temporarily removing cells from patients, transferring a gene to those cells, and then returning those cells to the patient (ex vivo).

In the in wVo methods, important issues still remain to be resolved, such as transfection efficiency, continuous expression and selective gene transfer to target cells. Ex wVo methods, on the other hand, can potentially overcome these problems. The majority of examples of ex wVo procedures have been performed using cells from the blood system (peripheral lymphocytes and bone marrow cells), as their collection and transplantation is relatively simple and the burden on the patients is reduced (Tani et al., Saishin Igaku, 56:258-267, 2001). In the case of cells other than cells from the blood system, methods that transfer genes to hepatocytes and then return these cells to the patient have been performed (Raper, S.E. et al., Cell Transplant 2 (5): 381-400, 1993), but most of these methods focus on the recovery, maintenance and enhancement of the function of the transfected cells themselves.

Description of the invention

While searching for cells suitable for ex vivo gene therapy, the present inventors developed the idea of using primary, cultured adipocytes. The use of adipocytes has the following advantages: (1) there are many reports of humoral factors secreted from adipocytes, and adipocytes include the functions associated with hormone formation and can function as secretory organs (Bradley R.D., et al., Recent Prog. Horm. Res., 2001, 56, 329-358), (2) adipocytes can be easily collected, since they are also present subcutaneously, and techniques associated with their extirpation have been developed in the fields of plastic and cosmetic surgery, moreover, these cells are not heterotropic even when transplanted into subcutaneous tissue, which easily allows implantation, since they originally belonged to this area, (3) since isolated, primary, cultured adipocytes actively proliferate, even in vitro, they are suitable for procedures such as gene transfer, (4) since adipocytes are likely to remain in a limited area after implantation, the transplanted cells can be extirpated if desired after implantation (especially when one wishes to eliminate the gene expression), (5) since adipocytes themselves form angiogenic factors (Mick, G.J., et al., Endocrinology 2002, 143 (3): 948-53), a high degree of engraftment can be expected after implantation, (6) extirpation or implantation of adipocytes has little effect on the human body, since the weight of this organ varies widely in adults, and (7) adipocytes are widely perceived as redundant and obstructive, and approval of collection of these can be easily obtained.

Although investigations with a similar purpose are currently underway using keratinocytes (J. Gene. Med. 2001 Jan-Feb, 3(1) : 21-31; Histochem. Cell. Biol. 2001 Jan, 115 (1) : 73-82), then removing the biological barrier that the skin forms in the process of isolating the primary culture is problematic in light of the risk of infection. The pain to the patient during the extirpation and implantation is expected to be severe, and re-extirpation (4, mentioned above) to remove the expression is not easy. Furthermore, when using keratinocytes or skin, which can only be transplanted two-dimensionally, the amount of graft can only be increased by increasing the surface area of the graft. For that reason, adipocytes, which allow three-dimensional transplantation, are considered more applicable.

The present inventors devised methods for the efficient transfer of genes into primary, cultured adipocytes. They also confirmed that the transferred genes are functional after implantation and found that adipocytes can be effectively used in gene therapy. Moreover, adipocytes which stably express the transferred foreign gene in vivo over a longer period of time can be obtained by the methods according to the present invention. The implanted, mature adipocytes can continue to express foreign genes for a year or longer. If expression of the foreign gene becomes unnecessary after implantation of adipocytes, the expression can also be stopped by removing the graft.

More specifically, the present invention relates to primary, cultured adipocytes for gene therapy which stably carry one or more foreign genes that code for one or more proteins secreted from the cell, methods for producing these cells, implant compositions comprising these cells, use of these cells and similar, and specifically applies to:

[1] a primary, cultured adipocyte for gene therapy, where the adipocyte stably carries a foreign gene that codes for a protein secreted from the cell,

[2] the adipocyte according to [1], where the gene has been transferred to the cell with a retrovirus vector or an adeno-associated virus vector,

[3] the adipocyte according to [1] capable of significant expression of the protein in vivo for at least 20 days,

[4] the adipocyte according to [1], which is used for releasing the protein into the bloodstream,

[5] the adipocyte according to [1], where the protein is insulin or glucagon-like peptide 1 (GLP-1),

[6] a method for producing an adipocyte for gene therapy, where the method comprises the steps:

(i) primary culture of an adipocyte, and

(ii) transfer and then stable maintenance of a foreign gene encoding a protein secreted from the cell,

[7] the method according to [6], where the foreign gene is transferred with a retrovirus vector or an adeno-associated virus vector,

[8] an adipocyte for gene therapy which is produced by the method according to [6] or

[7],

[9] an implant composition for gene therapy, where the composition comprises a primary, cultured adipocyte stably carrying a foreign gene that codes for a protein secreted from the cell and a pharmaceutically acceptable carrier,

[10] implant composition according to [9] which further comprises a component of the extracellular matrix,

[11] implant composition according to [9] which further comprises an angiogenesis factor,

[12] a gene therapy method comprising the step of introducing into a body a primary cultured adipocyte stably carrying a foreign gene encoding a desired therapeutic protein secreted from the cell,

[13] a method for releasing a protein into the blood stream, wherein the method comprises the step of supplying to a body a primary, cultured adipocyte which stably carries a foreign gene which codes for a protein which is secreted from the cell,

[14] the method according to [13], which is a method for releasing the protein into the bloodstream over a period of 20 days or more,

[15] a method for reducing the glucose concentration in the blood, where the method comprises the step of adding to a body a primary, cultured adipocyte which stably carries a gene encoding insulin or glucagon-like peptide 1 (GLP-1), and

[16] an animal whose body is implanted with a primary, cultured adipocyte that stably carries a foreign gene encoding a protein secreted from the cell.

In what follows, the embodiment of the present invention will be described.

Firstly, the present invention provides primary, cultured adipocytes for gene therapy where the adipocytes stably carry one or more foreign genes that code for one or more proteins that are secreted into the cell's surroundings.

Here, a foreign gene refers to a gene that has been transferred to primary, cultured adipocytes from the outside, and includes genes that code for proteins that are not produced by the primary, cultured adipocytes. Furthermore, primary, cultured cells refer to non-established cells that are grown from tissue removed from a living body. Adipocytes refer to mature adipocytes and cells that have the ability to differentiate into adipose tissue, such as preadipocytes. More specifically, unless the adipocytes are specifically said to be "mature" adipocytes, the term also includes preadipocytes. Mature adipocytes are spherical cells that store fat and contain lipid droplets. Fat stored in mature adipocytes can be detected by staining with "oil red 0". Mature adipocytes normally secrete leptin in response to insulin. Preadipocytes normally exist as stromal cells that have not yet differentiated into mature adipocytes. Preadipocytes can be isolated by treating adipose tissue with collagenase, or can be isolated from division of mature adipocytes using the "ceiling" culture method described below (Sugihara, et al. Nippon Rinsho 1995, 53, 115-120; Sugihara, H. , et al. J. Lipid Res. 1987, 28, 1038-1045; Zhang H.H., et al. J. Endcriniol. 2000, 164, 119-128). Although the presence of adipocyte-specific surface antigens has not been confirmed, a high level of CD36 expression and the like has been found in mature adipocytes (Abumrad N.A., et al., J. Biol. Chem. 1993 Aug. 25, 268 (24): 17665-8). For that reason, extremely pure adipocytes can be collected using such molecules as markers. By inducing differentiation as described below, preadipocytes can differentiate into mature adipocytes within a few days to a few weeks (Hauner H., et al., J. Clin. Invest. 84, 1663-1670, 1989; Marko, et al ., Endocrinology 136, 4582-4588, 1994). Primary cultured adipocytes can be isolated from a desired tissue, such as subcutaneous adipose tissue or visceral adipose tissue, such as tissue surrounding the epididymis or mesenteric tissue.

The term "for gene therapy" refers to the use of in vivo expression of one or more proteins encoded by one or more foreign genes in anticipation of a therapeutic effect. Furthermore, cells for gene therapy refer to cells that carry one or more foreign genes, where the cells are used for administration of the foreign gene to a body by ex wVo delivery and where the cells have the ability to express the protein in the body. Ex wVo administration refers to removing adipose tissue or adipocytes from an individual, performing gene transfer in vitro and then implanting the cells into the same or another individual.

Cells for gene therapy preferably refer to cells used for the treatment of disorders, which are cells that are implanted so that a specific protein is formed. Preferably, treatment with a specific protein comprises replacement therapy, which uses a protein whose physical or functional deficiency or absence causes a disorder, or neutralization therapy, which uses a protein that has an effect that can neutralize factors that are the cause of triggering and worsening of a given disease development . The specific protein is preferably a protein which shows activity in the blood stream or which is carried to a target tissue via the blood stream and which functions on the cell surface of this tissue. A continuous supply of the specific protein is also preferably required over a given period of time (eg from a few days to a few weeks or more). Factors or interferences for which protein replacement therapy is already performed or where such treatment is expected to be effective can all be targets.

In what follows, representative measures are listed based on their classification, but their application should not be understood to be limited to these examples, and the application of similar factors for similar purposes is encompassed by the invention.

Replacement therapy includes supplementation against disorders that develop or worsen due to lack or reduced function of a hormone or cytokine, supplementation against disorders due to a congenital genetic defect and supplementation of a factor to improve a disease state: insulin/diabetes, glucagon-like peptide 1 (GLP -l)/diabetes, obesity, eating disorders, GLP-2/inflammatory enteropathy, gastrointestinal disorders resulting from cancer chemotherapy, leptin/obesity, lipodystrophic diabetes, adiponectin/diabetes, angiopathy, blood coagulation factors VIII and IX/hemophilia, lipoprotein lipase (LPL )/LPL deficiency, hypertriglyceridemia, lecithin cholesterol acyl transferase (LCAT)/LCAT deficiency, erythropoietin/erythropenia, apo A-I/hypo-HDL-cholesterolemia, albumin/hypoproteinemia, atrial natriuretic peptide (ANP)/hypertension, heart failure, luteinizing hormone- releasing hormone (LH RH)/breast cancer, prostate cancer, angiostatin, endostatin/angiogenesis, metastasis inhibition, morphine prescription r-agonist peptide (endogenous opioid peptide (e.g. enkephalin), dynorphin, etc.)/pain relief, calcitonin and bone morphogenic factor (BMP)/osteoporosis, interferon-cc and p/malignant tumor, interferon-y/malignant tumor, hepatitis, allergy, interferon-p l/multiple sclerosis, interleukin- lcc and -ip malignant tumor, interleukin 4/psoriasis, interleukin 10/autoimmune disease, interleukin 12/malignant tumor, pancreatic secretory trypsin inhibitor/pancreatitis, superoxide dismutase/ischemic heart disease, angiopathy and the like.

Neutralization treatment against disease-causing factors or malignant transformation factors involves the production of partial peptides of solubilized receptors or neutralizing antibodies, or dominant negative proteins. Tumor necrosis factor-cc (TNF-a)-solubilized receptor/rheumatoid arthritis, solubilized IgE receptor/allergies, solubilized IgA receptor/food allergies, solubilized cytotoxic T-lymphocyte antigen 4 (CTLA4)/autoimmune diseases, solubilized CD40 ligand/immunological disorders, dominant negative blood coagulation factor Vila/thrombosis, solubilized fibroblast growth factor (FGF) receptor/thickening of the intima of blood vessels, etc.

Furthermore, the adipocytes according to the present invention are not limited to adipocytes used for so-called "therapy", but include cells used for in vivo expression of a desired secretory protein. For example, the methods according to the present invention allow the production of model animals by a posteriori expression of a given protein. By using these methods, disease model animals with a postenone expression of pathogenesis factors or exacerbating factors can be produced, and these animals can be used for drug screening. In the case of expression of disease-improving factors, these methods can also be used as evidence for working hypotheses for the discovery of new drugs where a given factor improves a pathological condition. The animals used include desired non-human animals, and preferably non-human mammals (including rodents and primates).

The primary cultured adipocytes for gene therapy according to the present invention stably carry one or more foreign genes that code for one or more proteins that are secreted from the cell. The term "stable carrier" means that the foreign gene is transferred to daughter cells during cell division, and the term refers more specifically to the introduction of the foreign gene into a chromosome of a cell. The adipocytes for gene therapy according to the present invention preferably comprise one or more foreign genes which have been stably transferred by means of a chromosome-integrating virus vector. More preferably, the foreign gene is transferred with a retroviral vector.

Ret rovi rus vector is stably integrated into a cell chromosome and has the ability to express a transferred gene for a long period. The vector's transfer efficiency and continued expression of the transferred gene depends on the cell type. For example, a gene transferred with a retroviral vector may show continued expression as long as cells grow, but expression may stop when cell growth stops (Lund, A.H., et al., J. Biomed. Sei. 1996, 3:365-378; Niwa, O. et al., 1983, Cell, 32:1105-1113). The expression of foreign genes is often observed to be suppressed, particularly after introducing the gene into a body by in vivo or ex vivo methods. This repression of expression is said to involve de n<y>o-methylation of the promoter or coding sequence of the transferred gene (Jahner, D. and Jaenisch, R., Nature 315:594-597, 1985; Challita, P.- M. and Kohn, D.B., Proe. Nat. Acad. Sci. USA 91:2567-2571, 1994; Hoeben, R.C. et al., J. Virol. 65:904-912, 1991). Moreover, deacetylation of histones is involved in the silencing of the transferred gene (Chen, W.Y. et

al., Proe. Nati. Acad. Pollock. USA 97:377-382, 2000; Chen, W.Y. et al., Proe. Nati. Acad. Pollock. USA 94:5798-5803, 1997). However, when the present inventors transferred a foreign gene into primary, cultured adipocytes using a retroviral vector, surprisingly, the expression of the transferred gene was found to persist in an extremely stable manner, both in vitro and in vivo. The expression of transferred genes is stable in adipocytes before differentiation, and also in mature adipocytes. The expression of the transferred gene was confirmed to persist throughout the duration of the experiment for in wtro cultures (80 days or more) and for the entire duration of the experiment when implanted in the body (360 days or more). Primary, cultured adipocytes into which one or more foreign genes have been stably transferred can therefore be used as implants that stably express one or more genes for a long period.

The adipocytes for gene therapy according to the present invention have the ability to significantly express one or more proteins encoded by one or more foreign genes for at least 20 days or more in vitro, more preferably in vivo. The term "significant expression" means, for example, that expression is detected at a statistically significant level, compared to the situation when the foreign gene is not transferred (for example, with a significance level of 5% or with higher significance). More preferably, the adipocytes according to the present invention, when transplanted into a body, have the ability to significantly express one or more proteins encoded by one or more foreign genes in the body for at least 30 days or more, preferably 40 days or more, more preferably 50 days or more, still more preferably 60 days or more, still more preferably 80 days or more, still more preferably 100 days or more, still more preferably 150 days or more, still more preferably 200 days or more, even more preferably 250 days or more , still more preferably 300 days or more and still more preferably 350 days or more.

The adipocytes for gene therapy according to the present invention are particularly useful as cells for the release of proteins, which are encoded by foreign genes carried by the cells into the blood stream. The proteins released into bloodstreams include desired secretory proteins that exhibit activity in the bloodstream or on the surface of cells in target tissues, and examples include desired humoral factors such as hormones and cytokines, and antibodies. More specific examples are, as mentioned above, hypoglycemic hormones such as insulin and/or glucagon-like peptide 1 (GLP-1) for the treatment of diabetes and the like, blood coagulation factors for the treatment of hemophilia and the like, and solubilized fragments of TNF-α receptor or anti- TNF-a antibody (including antibody fragments comprising a variable region from an antibody, such as Fab and scFv) in the treatment of diseases showing elevated TNF-a levels, such as rheumatoid arthritis. For example, for insulin, the cleavage sites (site 1 and site 2) can be replaced with the cleavage sequence of a protease expressed in adipocytes, so that mature insulin can be efficiently produced (eg, Groskreutz, D.J., et al. JBC, 1994, 269(8) , 6241). An insulin analog that is modified so that it exists in a single chain can also be used (Lee, H.C., et al., Nature. 2000 Nov. 23, 408 (6811) : 483-8). For GLP-1, a desired peptide that acts as a ligand for the GLP-1 receptor can be used (NP.002053; Thorens, B. et al., Diabetes 42, 1678-1682 (1993); Dillon, J.S. et al., Endocrinology 133, 1907-1910 (1993); Graziano, M.P. et al., Biochem. Biophys. Res. Commun. 196, 141-146 (1993); Stoffel, M. et al., Diabetes 42, 1215-1218 (1993) ). An example is GLP-1 (7-37) (Diabetes, 1998, 47:159-69; Endocrinology, 2001, 142: 521-7; Curr. Pharm, Dec., 2001, 7:1399-412; Gastroenterology, 2002 , 122:531-44).

The present invention also applies to methods for the production of adipocytes for gene therapy, where the methods include the steps:

(1) primary culture of adipocytes, and

(2) transfer to the cells of one or more foreign genes that code for one or more proteins that are secreted into the cell's environment, preferably using a retrovirus vector or an adeno-associated virus vector, so that the gene is stably maintained.

The present invention also applies to the adipocytes for gene therapy which are produced by this method. "Maintained stably" means the transfer of one or more foreign genes so that it is transferred to daughter cells when the cell divides, and refers more specifically to the integration of the foreign gene into a chromosome in the cells. Southern blotting or PCR with genomic DNA can molecular-biologically demonstrate that the foreign gene has achieved stable expression by being integrated into a chromosome. For concentrating the stably transfected cells, for example, a method using fluorescence-activated cell sorting (FACS), which concentrates cells by recognizing GFP expressed by the cells together with the target gene, can be used.

1. Procedures for collecting primary, cultured adipocytes.

Primary cultured adipocytes can be collected by methods described in the report by Sugihara et al. (Sugihara, H. et al., Differentiation, 31:42-49, 1986). More specifically, adipose tissue, preferably subcutaneous adipose tissue or visceral adipose tissue, for example tissue surrounding the epididymis or mesenteric tissue, is excised from the implant recipient and then finely divided using, for example, scissors or a scalpel after washing with PBS. This finely divided tissue is then treated by shaking at 37°C in a medium comprising a suitable amount of collagenase, preferably from 1 to 3 mg/ml, over a suitable period of time, preferably from 20 to 60 minutes, and then separated into a precipitated residual material and a liquid layer by centrifugation.

The liquid layer is preferably washed once or twice further by centrifugation and then transferred to a culture bottle filled with medium. Bubbles are removed and the flask is incubated in a CO 2 incubator to grow the cells so that the conventional culture surface is a ceiling ("ceiling" culture). After cultivation for a suitable period of time, preferably from 10 to 14 days, cells adhering to the roof surface are collected by trypsin treatment. These cells are then cultivated in a conventional culture system.

Primary, cultured adipocytes can be stored by freezing before or after gene transfer. This method allows multiple use of adipocytes after a single collection.

2. Gene transfer to adipocytes

Gene transfer can be performed using gene transfer reagents (Fugene 6, Roche; Lipofectamin, Invitrogen; Cellphect transfection kit (calcium phosphate method), Amersham, etc.), electroporation methods (Chen, H. et al., J. Biol. Chem. 1997, 272( 12), 8026-31), or viral vectors (Kay, M.A., et al., Nat. Med. 2001, 7, 33-40). The transmission is preferably carried out using viral vectors, more preferably using retroviral vectors (see e.g. Arai, T. et al., J. Virol., 1998, 72, pp. 1115-21).

If gene transfer is carried out using a plasmid, the plasmid is transfected into adipocytes, and the adipocytes that maintain the transferred, foreign gene in a stable manner can be selected. Such adipocytes can for example be selected by expressing the plasmid that codes for the foreign gene with a drug resistance gene or by carrying out the transfection together with a plasmid carrying a drug resistance gene, and then selecting the transfected cells using this drug. Alternatively, the cells can be obtained by cloning the transfected cells by limiting dilution techniques. Furthermore, when gene transfer is carried out using a plasmid, a method for transient expression of a phage-derived integrase can be used to increase the efficiency of chromosomal insertion (Mol. Cell Biol. 2001 June, 21 (12) : 3926-34).

An example of gene transfer to adipocytes using a viral vector is a method comprising an adeno-associated virus (AAV). AAV is a virus belonging to the genus Dependovirus from the Parvoviridae family and is characterized by chromosomal integration of the transferred gene. A recombinant AVV vector into which a foreign gene has been integrated can be prepared by integrating the foreign gene between two inverted end repeats

(ITR) and express the AAV packaging proteins (the products of the rep and cap genes) in the presence of the adenovirus proteins E1, E2A and E4 (Muzyczka, N. (1992) Curr. Top. Microbiol. Immunol. 158, 97-29 ; Kaplit, M.G. et al (1994) Nat. Genet. 8, 148-54; Xiao, X. et al.

(1996) J. Virol. 70, 8098-108; Kessler, P.D. et al. (1996) Proe. Nati. Acad. Pollock. US 93, 14082-4087; Xiao, W. et al. (1998) J. Virol. 72, 10222-0226).

In the present invention, the foreign gene is preferably transferred to adipocytes using a retrovirus vector. Retroviruses refer to viruses belonging to the Retroviridae family and include oncoviruses, foamy viruses (Russell, D. W. and Miller, A. D., J. Virol. 1996, 70:217-222; Wu, M. et al., J. Virol. 1999, 73:4498-4501), and lentiviruses (eg, HIV-1 (Naldini, L. et al., Science 1996, 272:263-267; Poeschla, E. et al., Proe. Nati. Acad. Sei. USA 1996, 93:11395-11399; Srinvasakumar, N. et al., J. Virol. 1997, 71:5841-5848; Zufferey, R., et al. Nat. Biotechnol. 1997, 15:871-875; Kim, V.N., et al., J. Virol. 1998, 72:811-816) and feline immunodeficiency virus (Johnston, J.C. et al., J. Virol. 1999, 73:4991-5000; Johnston, J. and Power, C , J. Virol. 1999, 73-2491-2498; Poeschla, E.M. et al., Nat. Med. 1998, 4: 354-357)). A preferred retroviral vector for use in the present invention is a Moloney murine leukemia virus (MoMLV) vector (T.M. Shinnick, R.A. Lerner and J.G. Sutcliffe, Nature 293, 543-548, 1981).

The retroviruses may be self-inactivating vectors (SIN) vectors. A SIN vector can be prepared by removing part of the 3' LTR during the packaging of the virus (Yu S.F. et al. (1986) Proe. Nati. Acad. Sei. USA 83:3194; Yee, J.K. et al., 1987 , Proe. Nat. Acad. Sci. USA 84:5197-5201; Zufferey, R. et al., 1998, J. Virology, 72, 9873-9880). The foreign gene in the retrovirus can be transcribed with the LTR as promoter, or it can be expressed from another promoter inside the vector. For example, a constitutive expression promoter such as the CMV promoter, the EF-1a promoter or the CAG promoter or any desired inducible promoter may be used. Furthermore, a chimeric promoter in which part of the LTR is replaced with another promoter can be used.

In particular, for the transfer of genes using retroviruses, a plasmid carrying a gene to be transferred, such as pBabe CL-SEAP-IRES-GFP, is gene-transferred into packaging cells, such as 293-EBNA cells (Invitrogen) using a gene transfer reagent or the like. These are then cultured for a suitable period of time, preferably one to three days, and the recombinant viruses formed in the supernatant are collected. These viruses are then used to infect the adipocytes to be transfected.

The retroviral vectors preferably comprise an envelope protein with broad tropism so that they can infect a wide variety of mammalian adipocytes, including human adipocytes. For example, amphotropic coat protein can be used (eg 4070A)

(accession K02729; Sorge, J. et al., Mol. Cell. Biol. 4 (9), 1730-1737 (1984)). In the present invention, the retroviruses are preferably pseudotyped (Emi, T. Friedman and J.K. Yee, J. Virol., 65 (3), 1202-1207 (1991); Yee, J.K. et al. (1994) Methods Cell Biol. 43 43: 99-112; Burns, J.C. et al. (1993) Proe. Nati. Acad. Sei. USA 90 90:8033-8037) with vesicular stomatitis virus G protein (VSV-G) (Rose, J.K. and Gallione, C.J., J . Virol. 39 (2), 519-528 (1981)). Pseudotyping with VSV-G allows highly efficient transfer of genes into adipocytes. VSV-G pseudotyped vector can be prepared by expressing VSV-G in packaging cells. More specifically, for example, packaging cells which can be induced to express VSV-G can advantageously be used (for example Arai T. et al., J. Virol., 1998: 72, pp. 1115-21).

The titer of the viruses produced can be determined by infecting cells with virus solutions that have been stepwise diluted and counting the number of colonies of infected cells (for details, see Ausubel et al). (Ausubel, F.M. et al. eds. (1995) Current Protocols in Molecular Biology. (John Wiley & Sons, NY)). Alternatively, the titer can be determined by the method of Byun et al., (Byun, J. et al. (1996) Gene Ther. 33333:1018-1020), Tafuro et al. (Tafuro, S. et al. (1996) Gene Ther. 33333:679-684). Miyao et al. (Miyao, Y. et al (1995) Cell Struct. Funct. 20 20:177-183), Claudio et al. (Claudio, P.P. et al. (2001) Anal Biochem. 291: 96-101) or Cashion et al. (Cashion, L.M. et al. (1999) Biotechniques 26: 924-930).

Primary, cultured adipocytes can be supplied with viral vectors by contacting the vectors with the cells. For example, primary cultured adipocytes are incubated in a culture solution comprising viral vectors. Adipocytes are preferably infected in the form of preadipocytes. The infection efficiency can be increased by adding 0.5 to 8 µg/ml or so of polybrene. The multiplicity of infection (MOI) is not particularly limited, but can be suitably adjusted within the range from 0.1 to 100. Gene-added cells can be selected by using, for example, a marker gene. However, if the infection is carried out at an MOI of about 2 or more, more preferably about 3, 4, 5 or more, the gene can be transferred to most cells, even without selection. The gene-added adipocytes can be used for implantation without further treatment, or they can in certain cases be transformed into mature adipocytes by cultivation in a medium comprising 3-isobutyl-1-methylxanthine (IBMX), dexamethasone and insulin. In such cases, since IBMX and dexamethasone are mainly used to activate the adipocyte peroxisome proliferator-activated receptor-γ (PPAR-γ), drugs that directly activate this receptor (for example the thiazolidine derivatives, pioglitazone/Takeda Pharmaceutical Company Limited and rosiglitazone/GlaxoSmithKline) can be added at the same time .

The primary, cultured adipocytes according to the present invention carrying a desired therapeutic gene can be implanted into the body of an immunologically matched recipient, so that gene therapy is made possible by in wVo expression of the secretory protein encoded by the therapeutic gene. The primary, cultured adipocytes to be implanted are preferably cells from the same host as recipient. The gene therapy methods in which the primary, cultured adipocytes according to the present invention are implanted can be used for expression of a desired secretory protein in a body in anticipation of that protein's effects. For example, a disorder can be treated or prevented by implanting the adipocytes according to the present invention that maintain one or more foreign genes that code for one or more proteins that have a therapeutic or preventive effect on the disorder. Furthermore, the present invention relates to methods for releasing proteins into the blood stream where the methods of advancement include the step of administering the primary, cultured adipocytes according to the present invention to a body. Using these methods, the protein encoded by a foreign gene can be significantly secreted into the blood stream for at least 20 days or more, preferably 30 days or more, more preferably 40 days or more, even more preferably 50 days or more, even more preferably 60 days or more, even more preferably 80 days or more, even more preferably 100 days or more even more preferably 150 days or more, even more preferably 200 days or more, even more preferably 250 days or more, even more preferably 300 days or more and even more preferably 350 days or more. The foreign gene that is expressed in the body can be detected and/or quantified by, for example, immunoassays such as EIA. Removal of the transplanted cells can stop the expression of the administered foreign gene at any time. In certain cases, upon transfer of an inducible suicide gene (eg HSV-tk) to the transplanted cells, the transplanted cells can be eliminated by administration of eg ganciclovir.

The present invention also provides implant compositions for gene therapy where the compositions comprise primary, cultured adipocytes that stably carry one or more foreign genes that code for one or more proteins that are secreted into the cell's surroundings and pharmaceutically acceptable carriers. Examples of such carriers are physiological saline, phosphate buffer, culture solutions, serum and body fluids. These can also be combined with a solid or gel-shaped support that becomes a platform for the cells.

The implant compositions according to the present invention preferably comprise an extracellular matrix (ECM) component. An extracellular matrix component refers to a component such as a protein or mucopolysaccharide, which forms part of an insoluble network or a fibrous structure that accumulates between cells. They can be isolated from organisms or artificially reconstructed. ECM components that are preferably used in the present invention are collagen, fibronectin, vitronectin, laminin, heparan sulfate, proteoglycan, glycosaminoglycan, chondroitin sulfate, hyaluronate, dermatan sulfate, keratin sulfate, elastin, or combinations of two or more of the above. These ECM components are preferably formed into a gel and then mixed with adipocytes. ECM gels used in the present invention are not particularly limited as long as at least one or more of the above-mentioned components are included, but they preferably include at least type IV collagen, laminin and neparan sulfate. Such ECM includes a substrate extracted from Engelbreth-Holm-Swarm mouse tumor ("Matrigel") (Becton Dickinson Labware) (US Patent No. 4,829,000). The structure of the compositions comprising the ECM component and the adipocytes used in the present invention is not particularly limited and can, for example, be a gel or paste network structure, a fibrous structure, a flat (disc) structure, a honeycomb structure and a sponge-like structure. The ECM components can be transferred to gels using conventional methods. The gel formation can be carried out, for example, by incubating an aqueous solution comprising approximately 0.3 to 0.5% collagen at 37°C for ten to 30 minutes. Otherwise, ECM components can be transferred into gels using a gelling agent.

Furthermore, the implant composition according to the present invention preferably comprises an angiogenesis factor. The implant compositions according to the present invention which comprise an angiogenesis factor cause blood vessels to form around them after implantation and can secrete a foreign protein into the blood stream at higher efficiency. The angiogenesis factors are not particularly limited as long as they are factors that can induce angiogenesis in vivo, and examples are vascular endothelial cell growth factor (VEGF), basic fibroblast growth factor (bFGF), acidic fibroblast growth factor (aFGF), platelet-derived growth factor, transforming growth factor-p (TGF-p ), osteonectin, angiopoietin and hepatocyte growth factor (HGF). The most preferred example is bFGF. bFGFs, which are also called FGF2, are not only fibroblast growth factors, but also have activity that promotes the growth of several cells such as vascular endothelial cells, cartilage, osteoblasts and epidermal cells (Abraham et al., EMBO J., 5, 2523-2528, 1986; Prats et al., Proe. Nati. Acad. Sei. USA, 86, 1836-1840, 1989). The bFGFs used in the present invention are not only natural proteins, but can also be produced by gene manipulation by recombinant DNA technology and modified forms thereof. Examples of bFGFs are those described in WO87/01728, WO89/04832, WO86/07595, WO87/03885, European patent applications with publication no. 237966, 281822, 326907, 394951 and 493737. Alternatively, another expression vector that transiently expresses an angiogenesis factor can be introduced into the adipocytes (see WO97/49827). The main purpose of angiogenesis factors used in this way is to form blood vessels around the transplanted cells so that the foreign protein can be efficiently secreted in the blood stream from the adipocytes according to the present invention. When using a vector encoding a vasculature-inducing factor for expression of this vasculature-inducing factor from adipocytes, the use of a transient expression vector (more specifically a vector which is not incorporated into the chromosome) is therefore preferred. When the adipocytes express a vasculature-inducing factor for a long period, excess amounts of blood vessels are formed around the implanted adipocytes, which can cause systemic side effects. It is therefore preferred that the foreign gene which codes for an angiogenesis factor is not stably transferred to the primary, cultured adipocytes according to the present invention.

3. Implantation of adipocytes

Gene-added adipocytes are prepared in a suitable cell concentration, preferably 0.2 x 10<7> to 2 x 107 cells/ml or 0.2 x 106 to 5 x 106 cells/ml if transfected with a retrovirus. They are then infused into the subcutaneous tissue or adipose tissue, preferably the subcutaneous tissue, or by mixing with an effective medium, preferably a solution comprising an extracellular matrix such as collagen. Injection into fatty tissue can be performed by making an incision and exposing the fatty tissue. Cells that have terminally differentiated into mature adipocytes will not proliferate after transplantation and will express the foreign gene for a long period at a constant level. The expression level of a foreign gene in a body receiving an implant is proportional to the number of implanted cells. When carrying out an implantation, a desired expression level can therefore be maintained for a long period in a body that receives an implant by adjusting the amount of adipocytes that are implanted, to adapt to a pre-measured expression level in vitro of the foreign gene.

Brief description of the figures

Fig. 1 is a set of photomicrographs of primary, cultured adipocytes isolated from subcutaneous fat of three-week-old ICR mice. (A) shows adipocytes that adhered to the roof-side culture surface after 14 days of "roof" culture, (B) shows primary cultured adipocytes grown in normal culture, (C) shows mature adipocytes that have stored lipid droplets due to induction of differentiation, and ( D) shows an oil red 0-stained image of cells induced to differentiate. Fig. 2 shows the activity of alkaline phosphatase (AP) in plasma obtained by implanting ICR nude mice with primary, cultured adipocytes (derived from subcutaneous fat of ICR mice) transiently transfected with AP-expressing plasmid pcDNA3.1-SEAPmh. Fig. 3 shows a comparison of gene transfer efficiency when retroviral vector MLV(VSV)/pBabeCL (PLAP) IP is transduced into primary, cultured adipocytes derived from different adipose tissues. Fig. 4 is a set of photomicrographs showing images of the induction of differentiation of primary cultured adipocytes transduced with MLV(VSV)/pBabeCL (GFP)IP. (A) and (B) show a light micrograph and a GFP fluorescence photograph, respectively, of the same field of view. Fig. 5 shows the duration of AP expression in subcultures of primary cultured adipocytes transduced with an AP-expressing viral vector. (A) shows the result of transfer of SEAP gene (MLV/VSV)/pBabeCL (SEAPmh) I2G) or PLAP gene (MLV(VSV)/pBabeCL(PLAP)IP into cells derived from subcutaneous fat of C57BL/6 mice (B) shows the result of transfer of PLAP gene (MLV(VSV)/pBabeCL (PLAP)IP) or GFP gene (MLV(VSV)/pBabeCL (GFP) IP) into adipocytes derived from ICR mice. 6 is a set of photographs and a curve showing the change in expression in differentiation-induced gene-injected adipocytes.(A) shows a GFP light microscope image of primary, cultured adipocytes under non-differentiation-inducing conditions where the adipocytes, which are transfected with MLV(VSV) /pBabeCL (GFP) IP, is derived from ICR subcutaneous fat. (B) shows a corresponding GFP microscope image taken under differentiation-inducing conditions. (C) shows the AP formation from MLV(VSV)/pBabeCL (PLAP) IP-transfected primary, cultured adipocytes (derived from ICR subcutaneous fat) under non-differentiation-inducing conditions (non-differentiation) and

differentiation-inducing conditions (differentiation).

Fig. 7 shows the (pro)-insulin production by plasmid transfection into primary, cultured adipocytes. Fig. 8 shows the stable expression of AP in primary, cultured adipocytes (derived from subcutaneous fat of C57BL/6 mice) transfected with AP-expressing AAV. Fig. 9 shows insulin expression at the time of induction of differentiation in primary, cultured adipocytes transfected with the insulin-expressing retroviral vector sls2B10. (A) shows the results using an EIA produced by Morinaga and (B) shows the results using an EIA produced by IBL. Fig. 10 shows the expression of GLP-1 (7-37) in primary, cultured adipocytes transfected with a GLP-1 (7-37)-expressing retroviral vector. The measurements were made in triplicate, and their mean values and standard deviation are shown. Fig. 11 shows the effect of the presence or absence of pre-implantation induction of differentiation stimulation on AP expression in vivo upon implantation of AP-expressing primary cultured adipocytes. Fig. 12 is a set of curves and a photograph. (A) shows the change in plasma AP activity when AP-expressing primary cultured adipocytes are implanted in the presence of stimulation of differentiation using a basic FGF-supplemented Matrigel. (B) shows loss of AP activity in plasma after extirpation of the implanted Matrigel (individual A). (C) shows a GFP light microscopy image of Matrigel excised from the control group, which was fed with GFP-transfected cells. For the PLAP-implanted group shown in (A), the values shown are the group mean and standard deviation of values measured for each individual up to day 32. The remaining values are mean values. Fig. 13 shows the results of the long-term investigation of the AP activity in blood in mice supplied with an implant by the method according to Fig. 12 (A) and by a number of other methods. Fig. 14 shows the results of carrying out an extirpation test corresponding to the one in

Fig. 12 (B) in late stage of the transplant.

Fig. 15 shows the dependence of blood AP activity on the number of implanted cells upon implantation of AP-expressing adipocytes. The stated values are the average value for the group and standard deviation for the measurements in each individual. Fig. 16 shows the effect of implantation of sls2B10 insulin-expressing adipocytes into STZ-induced diabetic mice. (A) shows the effect on fasting plasma glucose levels, and (B) shows the effect on body weight. The stated values are the average for the group and standard deviation for the measurements in each individual.

The preferred embodiment of the invention

The present invention will now be described in detail below with reference to examples.

[Example 1] Primary culture of mouse adipocytes

[Methods]

Three-week-old male ICR mice or four- to five-week-old male C57BL/6 mice (both from Charles River) were anesthetized with diethyl ether and sacrificed by collection of whole blood from the heart. Subsequently, inguinal subcutaneous fat or fat surrounding the epididymis and mesenteric adipose tissue were excised individually under sterile conditions. The excised tissues were washed with PBS and then finely divided using scissors or a scalpel. This minced tissue was digested with shaking at 37°C for 20 to 60 min in normal medium (DMEM-high glucose/SIGMA, 10% FCS) supplemented with 1 mg/ml collagenase (Si fraction/Nitta gelatin) and then separated in precipitated material and suspended layer by centrifugation (300 g, five minutes).

The liquid layer was further centrifuged once or twice to remove the collagenase by dilution, and then transferred to a T-25 bottle (IWAKI) filled with medium. Bubbles were removed and the cells were grown under a 5% CO 2 atmosphere in a CO 2 incubator at 37°C with the conventional culture surface uppermost ("ceiling" culture). After ten to 14 days of culture, the cells that adhered to the roof surface were collected by trypsin treatment and transferred to a normal culture system. Subculture was then carried out in a ratio of 1:3 to 1:10.

To induce differentiation, medium with cells grown to confluence in a 6-well plate was transferred to an induction medium (normal medium supplemented with 0.5 mM IBMX, 0.25 nM dexamethasone, and 10 µg/ml insulin). This stimulation continued for 48 hours. The cells were then differentiated in a maturation medium (normal medium with 10 ug/ml insulin added). The maturation medium was changed every three days.

"Oil red 0" color solution was prepared by mixing a stock solution, prepared by dissolving 0.3 g of "oil red O" in 100 ml of isopropanol (99%), with distilled water in a 3:2 ratio just before use. The cells were washed with PBS and then fixed with a 10% neutral formalin solution (WAKO). After another wash with PBS, the cells were stained with "oil red 0" dye solution at room temperature for ten minutes. The cells were again washed with PBS and then examined with a microscope.

[Results]

Fig. 1 is a set of photomicrographs of primary, cultured adipocytes isolated from subcutaneous fat of three-week-old ICR mice. After 14 days of "roof" culture, adhesion of adipocytes containing lipid droplets was observed on the roof side of the culture surface (A). When these cells were transferred to a normal culture system, they showed fibroblast-like growth, as shown in (B). However, when differentiation was induced with IBMX, dexamethasone, and insulin, the cells differentiated again into mature adipocytes containing lipid droplets (C). Stored fat was stained red with "oil red 0" stain (D). Cells isolated by this method were shown to be primary, cultured adipocytes with the ability to differentiate.

[Example 2] Transient transfer of a heat-stable secretory alkaline phosphatase (AP) gene into primary cultured adipocytes and implantation of transfected adipocytes into mice.

As a model system for gene expression, the AP gene, specifically the SEAP gene (Clontech) or the PLAP gene (Goto, M. et al. Mol. Pharmacol, Vol. 49860-873

(1996)) transferred to primary, cultured adipocytes, and changes in AP activity were examined. (Both AP gene products are heat stable and can be easily separated from endogenous alkaline phosphatases by heat treatment).

[Procedures]

(1) Preparation of primary cultured adipocytes transiently transfected with the SEAP gene

The AP-expressing plasmid (pcDNA3.1-SEAPmh) was constructed by inserting the SEAP sequence, obtained by double cutting pSEAP2 base vector (Clontech) with the restriction enzymes HindIII - XbaI, into the W/77dIII-XZ>aI site of pcDNA3 .1Myc-HisA (Invitrogen), which is a vector for expression in mammalian cells.

For each gene transfer, 500 µl FCS-free DMEM medium and 15 µl Fugene 6 reagent (Roche) were mixed in a 10 cm dish, after which 5 µg pcDNA3.1-SEAPmh was added. This mixture was left at room temperature for 15 minutes. This mixture was added to primary, cultured cells (derived from ICR subcutaneous fat) grown to 70 to 80% confluence in a 10 cm dish. This was then cultured for 24 hours in a CO2 incubator.

(2) Implantation of mice with primary cultured adipocytes transfected with the gene for alkaline phosphatase

Transfected cells were collected by trypsinization and washed twice with PBS by centrifugation. The cells were then suspended in PBS as 1 x 10 cells/ml. The animals (ICR nude mice, five weeks old at the time of surgery) were anesthetized by intraperitoneal administration of 50 mg/kg sodium pentobarbital (Nembutal; Dainippon Pharmaceutical). After disinfecting the area to be operated on with diluted Hibitan solution (Sumitomo Pharmaceuticals), an approximately 3 mm to 5 mm incision was made in the skin near the root of the right hind leg, and the inguinal subcutaneous fat was exposed. 0.55 ml of the prepared cell suspension (5.5 x 10 cells/head) was transferred to a 1 ml syringe and injected into the subcutaneous fat using a 22 G injection needle. As a control, PBS was injected into the same seat. To compare this with the protein supplementation procedure, 1 µg of purified AP (Roche) was dissolved in PBS under sterile conditions and injected in a similar manner. The incision in the skin was sutured together and the surgical site disinfected with surgical Isodine (Meiji Seika).

Blood was collected from the postorbital plexus vein using a heparin-coated capillary (Dramond) before implantation (day 0) and at various times after implantation. Plasma was obtained from whole blood by centrifugation at 2000 g for 15 minutes. The AP activity in this plasma was measured using an assay kit (SEAP reporter gene assay kit, Roche) following the enclosed instructions.

[Results]

Fig. 2 shows the AP activity in plasma obtained by implanting mice with primary cultured cells transiently transfected with alkaline phosphatase (AP)-expressing plasmid pcDNA3.1-SEAPmh. For comparison purposes, mice were administered 1 µg of purified AP protein (Roche) by injection. Seven days after administration, AP activity in the blood of these mice had decreased to control levels. On the other hand, AP activity in blood of mice given an implant of cells transiently transfected with genes was shown to reach a peak at day 4 after implantation, and the duration of expression was 14 days. The duration of expression in vivo by the implantation of cells containing transiently transfected genes was short, and the concentration in the blood was found to vary widely, although the activity lasted longer than by injection of protein.

[Example 3] Preparation of adipocytes stably expressing AP using a viral vector

[Procedures]

(1) Construction of AP expression vectors and control GFP expression vectors

The PLAP gene was excised from pTK-PLAP using HindIII and BglII, as described in the literature (Goto, M. et al. Mol. Pharmacol. vol. 49, 860-873 (1996)). The SEAP gene was obtained by cutting pcDNA 3.1.SEAPmh with Hincflll/Pmel. The GFP gene was excised from pEGFP-N2 using NotI-Ncol.

The plasmid, pBabeCLXI2G, used for viral vector production was prepared based on pBabePuro (Morgenstern, J.P. et al. Nucleic Acids Res. Vol. 18, 3587-3596 (1990)) by cutting out the SV40 promoter and neomycin resistance genes using SalI - Clal, blunting these ends with Klenow fragments and then replacing these with the internal ribosome binding site (IRES) from encephalomyocarditis virus (EMCV), which was excised from pIRES2-EGFP by Hincll-Hincll, and the green fluorescent protein (GFP), and then replace the part from the long terminal repeat (LTR) to the insertion site of the foreign gene (multicloning site) (SspI-BamHl) with a sequence corresponding to (SspI-BamHl) from pCLXSN (IMGENEX). In addition, pBabeCLXIP, in which the IRES-GFP part of pBabeCLXI2G was replaced with the IRES-puromycin resistance gene, was also used.

Each of the DNA fragments with the above-mentioned PLAP, SEAP, and GFP was blunt-ended with Klenow fragments and then inserted into the vector pBabeCLXIP or pBabeCLXI2G cut with Hpa I, to obtain pBabeCL (PLAP) IP, pBabeCL (SEAPmh) I2G, and pBabeCL (GFP) IP.

(2) Preparation of viral vectors

Each gene transfer to a 10 cm dish was performed as follows: 30 µl of plasmid transfection reagent TransIT (MIRUS) was mixed with 500 µl of FCS-free DMEM medium and left at room temperature for five minutes (mixed DMEM/TransIT solution). In a separate tube, 3.3 µg of a vector encoding VSV-G (pCALG, modified according to Arai, T. et al., J. Virol., 1998, 72, pp. 1115-21), 3.3 ug of a vector encoding Gag-Pol (pCLampho/RetroMax system (IMGENEX)), and 3.3 ug of a vector comprising a packaging signal and the gene to be transferred (pBabeCL (PLAP) IP, pBabeCL (SEAPmh)I2G or pBabeCL ( GFP) IP, was mixed together, totaling 9.9 µg (plasmid solution). The plasmid solution was added to the mixed DMEM/TransIT solution, mixed thoroughly and then left at room temperature for 15 minutes. This was then added to 293-EBNA cells (Invitrogen), cultured overnight from 2x 10<6> cells/10 cm dish from the previous day.

The medium was changed eight hours after addition, and the culture supernatant was collected after another two days of cultivation. The collected culture supernatant was centrifuged (300 g, 5 minutes) or filtered through a 0.45 µl syringe filter (Millipore) to remove contaminants, and this supernatant was used as virus solution for (MLV(VSV)/pBabeCL (PLAP) IP, respectively , MLV(VSV)/pBabeCL (SEAPmh)I2G, and MLV(VSV)/pBabeCL (GFP/ IP. Some of the virus suspension was concentrated by ultracentrifugation (19500 rpm, 100 minutes) and then used.

(3) Gene transfer to and cultivation of primary, cultured adipocytes.

Adipocytes to be used for gene transfer (derived from subcutaneous fat, fat surrounding the epididymis and mesenteric fat from ICR mice and subcutaneous fat from C57BL/6 mice) were cultured in 6-well or 96-well plates so that they were 50 to 80% confluent the day before transfection. The medium was removed and equal amounts of 4 µg/ml Polybrene (SIGMA) solution and virus suspension were added to the cells for transduction of the viral vector. Eight hours after the transduction, the medium was replaced with normal medium, and further culturing and subculturing were carried out. The AP activity in a portion of the cells was measured by collecting the 24-hour culture supernatant on day four after the transfection (Fig. 3).

The subcultivation was carried out according to the procedure in Example 1 in a 10 cm dish scale. The cells were cultured for four to seven days, and the medium was changed when they reached confluence. AP activity was measured in the culture supernatant 17 hours later. These cells were continuously sub-cultured, and by performing corresponding manipulations in a suitable manner, the maintenance of the expression was investigated (Figs. 5 and 6). The AP activity was not measured each time the cells were sub-cultured.

Differentiation was induced in 6-well plates according to the method in Example 1. However, the treatment was carried out over three days with induction medium which was replaced with maturation medium every three days thereafter. The AP activity in the culture supernatant was measured using culture supernatant obtained every third day, and the x-axes in the figures show the day on which the supernatant was collected. For the GFP-transfected cells, photomicrographs were taken under appropriate GFP light (Figs. 4 and 6). Non-differentiation-inducing conditions refer to conditions where cultivation is continued in normal medium instead of in induction medium or maturation medium.

[Results]

Fig. 3 is a comparison of gene transfer efficiency for each type of tissue-derived cell using retroviral vectors. AP activity was confirmed in the culture supernatant of all cells when the gene transfer was performed on the primary, cultured adipocytes isolated from each of the adipose tissues present in the inguinal, subcutaneous tissue, epididymis and mesentery of ICR mice. This showed that the retroviral vectors can transfer genes regardless of the place of origin of the cells. Fig. 4 is a set of photomicrographs showing images of the induction of differentiation of cells transduced with a GFP-expressing retroviral vector. Induction of differentiation was initiated 13 days after the gene transfer, and photographs were taken three weeks later. GFP fluorescence was observed in cells containing lipid droplets, which shows that the viral vector can transfer genes to preadipocytes that have the ability to differentiate and that gene transfer by the vector does not affect the cells' ability to differentiate. Fig. 5 shows the course of expression in the subcultures of primary, cultured adipocytes transfected with an AP-expressing virus vector. AP activity was measured in culture supernatant taken 17 hours after the cells reached confluence in a 10 cm dish. Continuous AP production was confirmed over the 87 days in which primary cultured adipocytes derived from subcutaneous fat from C57BL/6 mice were examined (A), and over the 63 days in which primary cultured adipocytes derived from subcutaneous fat from ICR mice were examined (B). These results showed that transduction of the viral vector into primary, cultured adipocytes can provide cells with stable expression that can maintain the foreign genes in the daughter cells that are formed after cell division. Fig. 6 is a set of photographs and a graph showing changes in expression in differentiation-induced adipocytes with added genes. GFP-expressing adipocytes derived from subcutaneous fat of ICR mice showed strong GFP expression under both normal culture conditions (A) and differentiation-inducing conditions (B). Furthermore, AP-expressing adipocytes derived from subcutaneous fat of ICR mice showed continuous expression of AP under both non-differentiation-inducing conditions (non-differentiation) and differentiation-inducing conditions (differentiation) (C). The primary, cultured adipocytes that had been transfected with genes via the viral vectors were found to stably express genes in any phase, not only under the proliferative conditions described in Fig. 5, but also under non-differentiation-inducing conditions, or more precisely under non-proliferative conditions or mature conditions.

[Example 4] Preparation using a plasmid vector of adipocytes stably expressing insulin.

The methods of gene transfer include methods using plasmid vectors.

[Procedures]

(1) Isolation and modification of the human insulin gene

PCR was performed on a human pancreatic cDNA library (Stratagene) using the primers shown in Table 1 (Insulin Fw and Rv). A fragment of the human insulin gene was obtained. The nucleotide sequence of this obtained fragment of 354 bp was determined and the fragment was subcloned into the vector pCR2.1TOPO (Invitrogen) as native insulin.

(Bold letters indicate the initiation codon in Fw and the "antisense" sequence to the stop codon in Rv. Underlined letters indicate mutated parts).

To express mature insulin in the adipocytes, genetic modification was then performed based on the literature (JBC, 1994, 269 (8), 6241-). Specifically, primers in both directions were synthesized separately so that they contained mutations in each transition site between the human insulin B-chain and the C-peptide (site 1), between the C-peptide and the A-chain (site 2) and histidine residue no. 10 in the B-chain B10) (Table 1). The mutants were obtained using a "Quickchange" mutagenesis kit (Stratagene). Performing this reaction on site 1 and site 2 produced the sls2 mutant. Carrying out the reaction on site 1, site 2 and B10 gave the sls2B10 mutant of insulin. After confirmation of the nucleotide sequence of the obtained, modified human insulin gene, the gene was introduced into the vector pcDNA3.1 and then used for gene transfer.

(2) Gene transfer to primary, cultured adipocytes

After mixing 500 µl of FCS-free DMEM medium and 15 µl of Fugene 6 reagent (Roche), 5 µg of transfection plasmid was added, and the mixture was left at room temperature for 15 min. The mixed solution was added to primary cultured adipocytes (derived from periepididymal adipose tissue of C57BL/6 mice) grown to 70 to 80% confluence in a 10 cm dish. This was cultured for 24 hours in a CO2 incubator. Four days after the gene transfer, the cells were subcultured into a T225 flask and grown overnight. The medium was then changed to a medium comprising 0.2 mgU/ml G418 (SIGMA), and culture was continued for three weeks, after which gene-injected cells were selected. The obtained G418-resistant cells were seeded in a 10 cm dish, and the amount of insulin in the culture supernatant was measured using an ultrasensitive insulin EIA kit (Morinaga). This EIA kit detects both proinsulin that has not yet been processed and mature insulin.

[Results]

Fig. 7 shows the (pro)insulin production by plasmid transfection of primary, cultured adipocytes. Both pcDNA3.1Myc-His vectors containing the native human insulin gene (native), the setel/sete2/B10 modified form (sls2B10) and empty vector (mock transfection) were transfected into adipocytes derived from adipose tissue surrounding the epididymis of C57BL/6 mice. Human (pro)insulin was detected in the culture supernatant of resistant cells obtained by G418 selection. This showed that stable gene transfer to primary, cultured adipocytes is also possible using a plasmid vector.

[Example 5] Preparation of adipocytes stably expressing AP using adeno-associated virus

Methods of gene transfer include methods using adeno-associated virus (AAV).

[Procedures]

The examination was performed using the AAV Helper-Free System (Stratagene). The PLAP fragment from Example 2 (a fragment excised using HindIII and Sg/II) was inserted into the same restriction enzyme site in the vector pAAV-MCS to obtain pAAV-PLAP.

A AV vector production was performed as follows: 1.75 ml of OPTI-MEM (Invitrogen) was mixed with 220 µl of the plasmid transfection reagent Fugene, then 25 µg of each of pAAV-PLAP, pAAV-RC, and pHelper were mixed in, and these were left at room temperature for 15 minutes (Fugene/plasmid solutions). Meanwhile, 293-EBNA cells grown to 60 to 70% confluence in a 15 cm dish were prepared. The culture medium was changed to FCS-free DMEM, after which the Fugene/plasmid solution was dripped in evenly, and this was cultured for two to three hours. FCS was then added to a final concentration of 10% and this was cultured for a further two days. The cells were harvested by trypsinization and centrifugation, and then suspended in 50 mM tris-HCl and 150 mM NaCl solution so that the final volume was 3 ml. The cells were disrupted by performing three cycles consisting of freezing in dry ice-ethanol and thawing at 37°C of this suspension solution. After digestion of host genomic DNA using Benzonase (SIGMA), the virus solution was then prepared by centrifugation at 9000 rpm for 30 minutes, followed by filtration of the supernatant.

Primary, cultured adipocytes (derived from subcutaneous fat of C57BL/6 mice) were seeded in a 12-well plate at lx 10<4> cells/well the day before gene transfer and cultured. They were then treated for six hours in a medium containing 40 mM hydroxyurea and 1 mM butyric acid (both from SIGMA). After removal of this medium, 0.5 ml/well of the virus solution diluted 1/100 with FCS-free DMEM was added. After culturing for one hour, FCS-containing medium was added to a final concentration of 10% and this was cultured overnight. The medium was then changed in the normal way, and continuation of the culture was carried out on day 24.

The medium was changed on the first, seventh and 25th day of transfection, and the culture supernatant harvested two days after each change was used for the AP assays. The 10 µl supernatant, which was diluted as needed, was heated to 65 °C for 20 min, then 50 µl assay buffer (16 mM NaHCO 3 , 12 mM Na 2 CO 3 , 0.8 mM MgSO 4 ), and 50 µl luminescence dye reagent (CDP-Star Ready to Use with Sapphire II, TROPIX), added, allowed to react in the dark for 30 minutes and then measured with a luminometer.

[Results]

Fig. 8 shows stable expression of AP in primary, cultured adipocytes (derived from subcutaneous fat of C57BL/6 mice) transfected with AP-expressing AAV. AP activity was detected in the culture supernatant throughout the investigation period. This demonstrated that stable gene transfer into primary, cultured adipocytes can be achieved using an AAV vector.

[Example 6] Preparation of a retrovirus vector expressing human insulin and transduction thereof into adipocytes

[Procedures]

The modified human insulin gene prepared in Example 4 (sls2B10Ins) was inserted into the vector pBabeCLXI2G following the procedure of Example 3 (pBabeCI (sls2B10Ins) I2G). This plasmid, together with a VSV-G encoding vector (pVPack-VSV-G/Stratagene), and a Gag-Pol encoding vector (modified from pVPack-gp/Stratagene) was inserted into 293-EBNA cells according to the procedure in Example 3 to obtain the modified insulin-expressing retrovirus vector (MLV(VSV)/pBabeCL (sls2B10Ins) I2G). The culture supernatant (approximately 200 ml) of 293-EBNA cells from twenty-two 10 cm dishes was collected, insoluble material was removed by centrifugation/filtration treatment and then a concentrated virus solution was obtained by ultracentrifugation (19,500 rpm, 100 minutes). This was transferred to primary, cultured adipocytes (derived from subcutaneous fat of C57BL/6 mice) that had been seeded in a 6-well plate the day before.

The reintroduced cells were transferred to a 6-well plate, and differentiation was induced according to the procedure in Example 1. Culture supernatants were collected for three days, from three days before induction to the day of induction start (pre-induction) and for three days from 14 to day 17 of the induction (post-induction). The amount of insulin was analyzed by the same procedure as in Example 4. To confirm that processing occurred in the desired sites and that mature insulin was produced, further measurements were carried out using an insulin EIA kit (IBL) which only recognizes mature insulin . The culture supernatant from non-gene-introduced cells, which were simultaneously subjected to differentiation induction, was used as a control.

[Results]

Fig. 9 shows insulin expression at the time of induction of differentiation in primary cultured adipocytes transduced with retrovirus vector expressing sls2B10 insulin. (A) shows the results using the EIA kit produced by Morinaga, and (B) shows the results using the EIA kit produced by IBL. These results show that insulin is secreted stably both before and after the induction of differentiation and that transfer of a mutant insulin gene can result in the formation of mature insulin from adipocytes.

[Example 7] Construction of a retrovirus vector expressing human glucagon-like peptide-1 (GLP-1) and transduction thereof into adipocytes.

GLP-1 is a peptide that is formed from small L-cells in the intestine during food intake and includes the ability to stimulate insulin secretion by acting on pancreatic β-cells. GLP-1 is also known to have a number of other anti-diabetic effects and anti-obesity effects, for example regenerative effect on pancreatic β-cells, appetite suppressant effect and inhibitory effect on gastric emptying (Meier, J. J. et al. Eur. J. Pharmacol. 2002, 12 ; 440 (2-3) : 269-79; Drucker, DJ. Gastroenterology 2002; 122 (2) : 531-544. A peptide comprising the sequence position 7 to 37 of the amino acid sequence of GLP-1 (or up to position 36 in the amide form) is formed by tissue-specific processing of the polypeptide formed from the preproglucagon gene and is known to possess the most important pharmacological activity (Drucker, D. J. et al. Proe. Nati. Acad. Sei. USA. 1987 May; 84 (10) : 3434 -3438; Kreymann, B. et al. Lancet. 1987, 5; 2 (8571): 1300-1304; Mojsov, S. et al. J. Clin. Invest. 1987 Feb. 79 (2) : 616-619) .The subsequent investigation was carried out to produce this factor from adipocytes.

[Procedures]

A nucleotide sequence with a total of 156 base pairs was designed comprising a sequence (SEQ ID NO: 10 shows the coding sequence) where human GLP-1 (7-37) and a stop codon are linked to the signal peptide (17 amino acids) from the PLAP gene which was used in Example 3. Nucleotides were synthesized so that a 22-mer overlap was present in the middle (sPL-GLP-1Fw and sPL-GLP-1Rv in Table 1). These were hybridized and a double strand was formed using Pfu polymerase (Stratagene). The target fragment was then amplified by PCR using 5'-end and 3'-end primers (GLP-5' and GLP-3' in Table 1). This fragment was subcloned into the vector pCR2.1, cut out with restriction enzymes and then inserted into the vector pBabeCLXI2G in the same way as in Example 3 (pBabeCL (sPL-GLP1) I2G). This plasmid was transfected into 293-EBNA cells by a procedure similar to that of Example 6 for the preparation of the GLP-1 expressing retrovirus vector (MLV(VSV)/pBabeCL (sPL-GLP-1) I2G). Approximately 90 ml of culture supernatant from 293-EBNA cells from nine 10 cm dishes was collected. Insoluble material was removed by centrifugation/filtration, and the supernatant was then ultracentrifuged (19500 rpm, 100 minutes) to obtain a concentrated virus solution. This was transduced into primary, cultured adipocytes (derived from subcutaneous fat from C57BL/6 cells) which the day before had been seeded in a 6-well plate. The transfected adipocytes were again plated in a twelve-well plate, and induction of differentiation was performed according to the procedure in Example 1. "Non-induced" refers to a condition in which the culture was continued in normal medium instead of in induction medium or maturation medium. Seven days later, the medium was changed to FCS-free DMEM medium supplemented with 1 mM Valine-pyrrolidine (GLP-1 degradation enzyme inhibitor synthesized by Eisai). The culture supernatant was collected 18 hours later and the amount of active GLP-1 (7-37) was measured using ELISA (LINCO).

[Results]

Fig. 10 shows the expression level in primary, cultured adipocytes transfected with retrovirus vector expressing GLP-1 (7-37). Expression of the active form GLP-1 (7-37) was observed in culture supernatant from both non-differentiation-induced and differentiation-induced adipocytes. This showed that even if a factor is formed as a prepro-type and then cut out by processing, this method allows the production of only that factor from adipocytes.

[Example 8] Implantation of mice with cells stably expressing AP (Analysis 1)

[Procedures]

After culturing the AP-expressing adipocytes (transduced with MLV(VSV)/pBabeCL (PLAP) IP; derived from subcutaneous fat of C57BL/6 cells), prepared by the method of Example 3, to confluence, the cells were harvested by trypsinization, washed with PBS and suspended with 5x 10<6> cells/ml in ice-cold Matrigel (Becton Dickinson). The implantation was carried out by injecting this into the dorsal, subcutaneous area (Sc) of C57BL/6 mice (eight weeks old at the time of treatment, Charles River) in a dose of 0.2 ml per mice (lx 10<6>cells/animal) (without induction of differentiation). On the other hand, the same cells were grown to confluence and then cultured for three days in induction medium according to Example 1, and then implanted by corresponding methods (with induction of differentiation). Blood was collected overtime by the method set forth in Example 2, and AP activity in plasma was measured.

[Results]

Fig. 11 shows the change in AP activity in plasma from mice implanted with AP-expressing primary cultured adipocytes. Individuals given an implant of cells that had been subjected to differentiation-inducing stimulation for three days prior to implantation ("with induction of differentiation") showed less fluctuation in the continued expression than individuals given an implant of cells that had not been induced. However, both implantation procedures showed continuous expression over the entire 50 days or so of the study. This shows that the survival rate of cells after transplantation can be improved by providing differentiation-inducing stimulation.

[Example 9] Implantation of mice with cells stably expressing AP (Analysis 2)

[Procedures]

(1) Implantation

The AP-expressing adipocytes (transduced with MLV(VSV)/pBabe(PLAP)IP; derived from subcutaneous fat of ICR mice) prepared in Example 3 were grown to confluence. The cells were harvested by trypsinization, washed with PBS and suspended at 5x 10<6> cells/ml in ice-cold Matrigel (Becton Dickinson) to which was added 1ug/ml bFGF (Genzyme Techne). The implantation was carried out by injecting this in a dose of 0.2 ml per mice (lx 10<6>cells/animal) in each site (the dorsal subcutaneous area (Sc), inguinal subcutaneous fat (fat) and the intraperitoneal area (ip)) in ICR nude mice (six weeks old at the time of surgery, Charles River). As a control, GFP-expressing adipocytes were similarly treated and implanted into the subcutaneous tissue.

Some of the AP-expressing cells were cultured for three days in the induction medium according to Example 1 and then collected and implanted in the same way (Dif). After application of the induction medium, some of these cells were cultured for 4 days in a maturation medium and then collected and implanted in the same manner (Mat).

Furthermore, some of the AP-expressing cells were seeded in an 8-well Labteck chamber (Nunc) under the same conditions used for the implantation (x 10<6>/0.2 ml Matrigel added with bFGF), and the cells were solidified by heating to 37 °C. Implantation was achieved by inserting this solidified gel into the subcutaneous area of the mouse. Cells grown in normal medium after solidification are referred to here as prefixed (pf)/gr, and cells grown in a differentiation-inducing medium are referred to as pf/dif. The implantation was performed after seven days of cultivation.

The AP activity in plasma was measured over time, before implantation (day 0) and after implantation according to the procedure in Example 2.

(2) Extirpation

In the group that was implanted after induction of differentiation (Dif/Sc), the implanted cell mass was extirpated together with Matrigel from individuals A and B five and 43 weeks respectively after implantation. Extirpation was performed on the control sample in the fifth week after implantation. Each subject was administered intraperitoneally with 50 mg/kg Nembutal as anaesthesia. An incision was then made in their skin and a visually confirmed implanted piece of Matrigel was extirpated. The surgical site was sutured and disinfected with Isodine (Meiji). The animals were then housed in the same way, and blood was collected overtime.

[Results]

Fig. 12 (A) shows the result of the investigation of the change in AP activity in plasma over 50 days after implantation of AP-expressing primary, cultured adipocytes using Matrigel added with basic FGF in the presence of differentiation stimulation (Dif/Sc group). The change of AP activity in blood was stable for 49 days over a range of approximately 5-fold. This showed that the addition of bFGF at the time of implantation can further improve the engraftment rate after implantation. (B) shows the decrease in AP activity in plasma due to extirpation of the implanted Matrigel (individual A) over the same time period. The AP activity in the excised individuals was significantly reduced compared to the mean value for the PLAP-transduced group. This showed that AP in the blood is derived from the implanted cells and that extirpation of the graft can quickly remove the gene expression. At this time, extirpation was also performed on animals from the control group implanted with GFP-transfected cells. GFP-positive cells were found in the excised Matrigel, and many of these showed a vacuolar image (C) corresponding to that shown in Fig. 6 (B). This showed that primary, cultured adipocytes implanted by this method can be engrafted as mature adipose tissue in vivo. Fig. 13 shows the result from a long-term investigation of the AP activity in blood from the implanted mice according to Fig. 12 (A) and in mice that were given an implant by a number of other methods. In the group implanted with PLAP-transfected cells, a clear increase of AP activity in blood was confirmed for all implantation sites and implantation procedures. The AP activity in the blood was maintained over a long period of time, and more specifically, stable AP expression was observed over one year when analyzing the Dif/Sc group (the group described in Fig. 12 (A)). Continuous AP production was also confirmed for the other implantation procedures over the period of the study (316 days for the ip group, 54 days for the fat group, 225 for the Sc group, 317 days for the Mat/Sc group and 314 days for the two prefix the groups). The activity peak observed during one week of implantation was highest in the ip group. The highest values then had the order Sc>fat>Dif/Sc~pf-dif>pf-gr~Mat/Sc. The range of variation after implantation was observed as the ratio between the activity after 13 weeks and the peak activity, a ratio that can be compared in all groups. The variance was lowest, approximately three times, in the two prefix groups, approximately five times in the ip group, the Dif/Sc group and the Mat/Sc group and approximately ten times in the Sc group and the fat group. The peak value immediately after implantation and the range of variation after implantation varied for each implantation procedure. Any one of these methods can thus be used depending on the characteristics of the gene product used, the characteristics of the disease state and how easy the technique is to perform. This showed that implantation of primary, cultured adipocytes in which genes were stably introduced ex vivo can be carried out by a number of methods and that long-term stable wVo gene expression is possible after implantation.

Fig. 14 shows the result from carrying out an extirpation experiment corresponding to that described in Fig. 12 (B) in the later stages of implantation. The AP activity in the blood after the extirpation was confirmed to disappear rapidly, not only in individuals where the extirpation was performed in the early stages of implantation (individual A), but also in individuals where the extirpation was performed in the later stages of implantation (individual B). This showed that the adipocytes implanted by this method remain in the implantation site for a long period after implantation and that their extirpation as needed can remove the gene expression regardless of the time.

[Example 10] Transplantation of mice with cells stably expressing AP (Analysis 3)

The subsequent investigations were carried out to confirm the "dose" dependence of the number of implanted cells

[Procedures]

The AP-expressing adipocytes prepared in Example 3 (transfected with MLV(VSV)/pBabeCL(PLAP)IP; derived from subcutaneous fat of ICR mice) were grown to confluence. The cells were cultured for three days in the induction medium indicated in Example 1 and then harvested by trypsinization. After washing with PBS, the cells were suspended at 5x 10 cells/ml in Matrigel. A five-fold stepwise dilution was performed on the AP cell suspension using Matrigel, and suspensions of 1x 10<6> cells/ml and 2x 10<5> cells/ml were prepared. bFGF was added to these solutions at a final concentration of 1 µg/ml, and the solutions were then transplanted into the dorsal, subcutaneous area of ICR nude mice at a dose of 0.2 ml per mice (the high dose: lx 10<6>cells/animal, medium dose: 2x 10<5>cells/animal, low dose: 4x 10<4>cells/animal). As a control, GFP-expressing adipocytes were similarly treated and implanted into the subcutaneous tissue under the same conditions as for the high-dose conditions (lx 10<6> cells/animal).

[Results]

Fig. 15 shows how the AP activity in the blood depends on the number of implanted cells when implanting AP-expressing adipocytes. A dose-dependent AP activity in the blood was observed when the number of implanted cells was changed, and this was not affected by the duration. Specifically, the medium or low dose groups did not show a peak in the earlier stage of implantation, which was observed for the high dose group, and the range of variation was narrower. This showed that the in wVo expression level can be easily adjusted by using the number of implanted cells and that by adjusting the optimal cell number the blood concentration (expression level) after implantation can be stabilized.

[Example 11] Hypoglycemic effect in a mouse model of diabetes by implantation of insulin-expressing adipocytes

[Procedures]

Diabetic mice were produced by intravenous administration to eight-week-old male C57BL/6 mice of 10 ml/kg of 170 mg/kg streptozotocin (STZ, SIGMA). Fasting blood glucose (FBG) levels were measured individually one and two weeks after STZ administration, and individuals with an FBG of 300 mg/dl or more were determined to have diabetes. The blood sugar level was measured by performing a perchlorate treatment immediately after collection of whole blood and then using Glucosetest-II (WAKO).

MLV(VSV)/pBabeCL (sls2B10Ins)I2G-transfected adipocytes prepared in Example 6 were subjected to differentiation induction stimulation using the same procedure as in Example 10 and then suspended with 5x 10 cells/ml in Matrigel to which 1 µg had been added /ml bFGF. This suspension solution was implanted in the dorsal, subcutaneous area of the diabetic mice at 0.2 ml per seat for a total of 4 seats (4x 10<6>/animal). For the control group, non-transgenic adipocytes were implanted by the same procedure. The implantation was performed 19 days after the STZ treatment, after which the FBG level was measured overtime. Statistical analysis was performed by comparison with the control group (unpaired t-test).

[Results]

Fig. 16 shows the effect of implantation of sls2B10-insulin-expressing adipocytes

in STZ-induced diabetic mice. Non-transfected cells were implanted as a control. The blood glucose level of the group implanted with insulin-expressing cells tended to decrease from day seven of implantation, and a significant hypoglycemic effect could be observed on days 13 and 21 of implantation (A). Body weight 20 days after implantation was significantly higher in the group implanted with insulin-expressing cells than in the control group, and weight loss due to diabetes was therefore improved (B). The results of the investigation with AP suggest that this hypoglycaemic effect will be maintained for a long period. The foreign gene product that is formed from the implanted primary, cultured adipocytes was therefore shown to be able to contribute to changing the disease state of the recipient, which suggests that this method can be used for the treatment of diabetes.

Industrial applicability

The present invention established methods for ex wVo transfer of a foreign gene to primary, cultured adipocytes which are suitable for gene therapy and established primary, cultured adipocytes which stably maintained a foreign gene.

Claims (12)

1. Pharmaceutical composition comprising a primary cultured pre-adipocyte characterized in that the pre-adipocyte stably maintains a foreign gene that codes for a protein that is secreted into the cell's surroundings, and where the gene has been inserted into a retrovirus vector and has been transferred to the cell of the retrovirus vector, where the protein is insulin or glucagon-like peptide 1 (GLP-1). 2. Primary cultured pre-adipocyte characterized in that the pre-adipocyte stably maintains a foreign gene that codes for a protein that is secreted into the cell's environment, and where the gene has been inserted into a retroviral vector and has been transferred to the cell by the retroviral vector, for use in gene therapy, where the protein is insulin or GLP-1. 3. Pharmaceutical composition according to claim 1, or pre-adipocyte for use according to claim 2, where the pre-adipocyte has the ability to significantly express the protein in vivo for at least 20 days. 4. Pharmaceutical composition according to claim 3, or pre-adipocyte according to claim 2 or 3, where the pre-adipocyte is used to release the protein into the blood stream. 5. In vitro method for producing a pre-adipocyte for use in gene therapy, characterized in that the method comprises the steps: (1) primary culture of a pre-adipocyte, and (2) transfer, and then stable carrying of a foreign gene that codes for a protein that is secreted into the cell's surroundings, where the protein is insulin or GLP-1. 6. Method according to claim 5, where the foreign gene is transferred by means of a retrovirus vector. 7. Pre-adipocyte for use according to claim 2, where it is produced by the method according to claim 5 or 6. 8. Implant composition, characterized in that it comprises a primary cultured pre-adipocyte which stably carries a foreign gene which codes for a protein which is secreted into the cell's environment and where the gene has been inserted into a retroviral vector and has been transferred to the cell by the retroviral vector, and a pharmaceutically acceptable carrier, for use in gene therapy, where the protein is insulin or GLP-1. 9. Implant composition for use according to claim 8, or the pharmaceutical composition according to any one of claims 1, 3 and 4, where it further comprises an extracellular matrix component. 10. Implant composition for use according to claim 8 or 9, or the pharmaceutical composition according to any one of claims 1, 3, 4 and 9 where it further comprises an angiogenesis factor. 11. A primary cultured pre-adipocyte stably carrying a gene encoding insulin or glucagon-like peptide-1 (GLP-1) for use in lowering blood glucose, characterized in that the gene has been inserted into a retroviral vector and has been transferred to the cell by the retrovirus vector. 12. Non-human animal, characterized in that its body is implanted with the primary cultured pre-adipocyte according to claim 2.