WO2008019434A1 - The use of growth factors in a method of improving fat-graft survival - Google Patents

Abstract

The present invention relates generally to the use of agents to enhance adipogenesis and to promote fat graft survival in mammalian subjects. More particularly, the present invention contemplates the use of growth factors delivered by local or sustained administration alone or in combination with a range of growth factors to enhance angiogenesis in association with adipogenesis and to promote fat graft survival to thereby improve weight maintenance and to preserve normal cellular architecture of tissue including free fat grafts in mammalian subjects. Compositions comprising the instant agents also form part of the present invention.

Description

THE USE OF GROWTH FACTORS IN A METHOD OF IMPROVING

FAT-GRAFT SURVIVAL

BACKGROUND

FIELD

The present invention relates generally to the use of agents to enhance adipogenesis and to promote fat graft survival in mammalian subjects. More particularly, the present invention contemplates the use of growth factors delivered by local or sustained administration alone or in combination with a range of growth factors to enhance angiogenesis in association with adipogenesis and to promote fat graft survival to thereby improve weight maintenance and to preserve normal cellular architecture of tissue including free fat grafts in mammalian subjects. Compositions comprising the instant agents also form part of the present invention.

DESCRIPTION OF THE PRIOR ART

Bibliographic details of the publications referred to by author in this specification are collected at the end of the description.

Reference to any prior art is not, and should not be taken as an acknowledgment or any form of suggestion that this prior art forms part of the common general knowledge in any country.

Autologous fat injection is widely used for the correction of acquired and congenital soft tissue defects, most commonly in the face. This technique has been described in the treatment of facial lipodystrophy, augmentation of nasolabial furrows, depressed scars, post-traumatic defects, and enlargement of the lips (Coleman, Dermatol CHn 17(4):S9l- 898, 1999). This method is also used in treatment of atrophic cutaneous tissue on the dorsum of the hand, as well as correction of contour defects resulting from liposuction. The abundance of adipose tissue in most individuals, combined with the low morbidity and versatility of the procedure, has made autologous fat grafting an attractive option for soft tissue augmentation for over a century (Neuber, CMr Kongr Verhandl Dsch Gesellsch:22, 1893).

The major drawback to fat transplantation is its variable longevity, with up to 70% of the injected volume ultimately absorbed (Illouz, Aesthetic Plast Surg 12(3):\75-8l, 1988; Horl et al, Ann Plast Surg 2(5^:248-258, 1991; Chajchir, Aesthetic Plast Surg 20(4):29l-296, 1996; Moscona et al, Ann Plast Surg 33(5):500-506, 1994). The high absorption rate results in the need for over-correction of the defect and repeat procedures, with each additional procedure further increasing the risk of graft failure and absorption by damaging the blood supply to the recipient area (Shoshani et al, Plast Reconstr Surg 775(3^:853-859, 2005). The inconsistency in long-term results led to a focus in the differences in the diverse methods of harvesting, treatment, and placement of the fat (Coleman, Aesthetic Plast Surg 19 '(5) :421-425, 1995). Subsequently, a multitude of methodological variations have been described including atraumatic harvesting of fat with low-pressure suction (Coleman 1995 supra, Kanchwala and Bucky, Facial Plast Surg 19(1):137-146, 2003), avoidance of graft contact with room air (Coleman 1995 supra), fat concentration procedures (Carraway and Mellow, Ann Plast Surg 24(3):293-296, 1990), and washing of the graft to remove inflammatory mediators (Mikus et al, Laryngoscope 105 (1):\1 -22, 1995). However, a recent animal model has demonstrated that none of the common methods of harvesting or preparing fat results in consistently increased graft viability (Smith et al, Plast Reconstr Surg 117(6): 1836- 1844, 2006), leaving unpredictable long- term success a major obstacle to autologous fat grafting (Chajchir and Benzaquen, Plast Reconstr Surg 84(6):921-934, 1989).

Recent years have brought a host of studies describing graft manipulations intended to affect preadipocytes, mature adipocytes, or the recipient bed in an attempt to use 'bioactivation' of harvested fat to improve long-term outcomes (Shoshani et al, 2005 supra; Eppley et al, Plast Reconstr Surg 90(6): 1022-1030, 1992; Shoshani et al, J Drugs Dermatol 4(3):3l 1-316, 2005; Yuksel et al, Plast Reconstr Surg 105 (5):1712-1120, 2000). Treatments have included addition of basic fibroblast growth factor (bFGF) attached to dextran beads (Eppley et al, 1992 supra), insulin, insulin-like growth factor, and bFGF delivered through a poly (lactic-co-glycolicacid)-polyethylene glycol (PLGA/PEG) microsphere delivery system (Yuksel et al, 2000 supra) and interleukin-8 (Shoshani et al, 2005 supra), all with varying degrees of success.

The donor morbidity inherent in the reconstruction of acquired and congenital soft tissue defects is well described. The emerging field of tissue engineering has the potential to address these issues through the prefabrication of mature, vascularized, autologous adipose tissue for use in human reconstruction (Langer and Vacanti, Science 260(5110):920-926, 1993; Thomas and Penington, International Journal of Adipose Tissue, 2005). Adipose tissue displays angiogenic properties (Silverman et al, Biochem Biophys Res Commun 153(l):347-352, 1988; Sierra-Honigmann et al, Science 281(5383):l683-\6%6, 1998) and has the capacity to continue growth throughout one's lifespan with the potential to acquire new fat cells from fat cell precursors (Gregoire et al, Physiol Rev 78(3):783-809, 1998). Such increases in fat mass occur in tandem with an increase in the microcirculation (Bouloumie et al, Ann Endocrinol (Paris) 63(2 Pt l):9\-95, 2002). The importance of neovascularization during adipose tissue growth is highlighted by findings that mice treated with angiogenesis inhibitors demonstrate dose dependent and reversible adipose tissue loss (Rupnick et al, Proc Natl Acad Sci USA 99(16): 10730-10735, 2002). Understanding the mechanistic interplay between angiogenesis and adipogenesis is therefore essential and has direct applications for de-novo adipose tissue engineering (Fukumura et al, Circ Res 93(9):8S-97, 2003).

Angiogenesis and adipogenesis are tightly coordinated in both time and space during embryonic development (Crandall et al, Microcirculation 42(2):2\ 1-232, 1997). Early observations in fetal pigs demonstrated a positive correlation between fat cell density and capillary density, as well as the finding that the largest developing fat cell clusters are consistently located at the entry points of large blood vessels (Hausman and Thomas, J Morphol 190(3):271-2S3, 1986). Recently, multiple studies have further delineated the developmental relationship between angiogenesis and adipogenesis. Crandall et al. showed that both human preadipocytes and endothelial cells express the αvβ3 integrin and express and secrete PAI-I, which regulates the migration of preadipocytes and endothelial cells in vitro (Crandall et al, JClin Endocrinol Metab 85 (7):2609-2614, 2000). This would suggest a developmental mechanism where preadipocytes would travel with endothelial cells during angiogenesis and adipogenesis, as well as a means for local control via secreted PAI-I (Crandall et al, 2000 supra). Neels et al, Faseb J 7S(9;:983-985, 2004 also demonstrated the close temporal and spatial relationship between these two processes. In their study, they injected preadipocytes in nude mice and revealed increased expression of endothelial cell marker genes (CD31, Tie-1, and Tie-2) in parallel with an increase in adipogenic marker genes (lipoprotein lipase and adipsin) in the developing fat pad (Neels et al, 2004 supra). In addition, the kinetics of angiogenesis in the fat pad paralleled the formation of lipid-filled vacuoles in the adipocytes (Neels et al, 2004 supra).

Using a similar methodology, Fukumura et al, 2003 supra, demonstrated the reciprocal regulation of angiogenesis and adipocyte differentiation. In their study, murine preadipocytes that were transplanted into a mouse dorsal skin fold chamber produced heavily vascularized fat pads (Fukumura et al, 2003 supra). Transfection of the preadipocytes with a dominant-negative PPARγ construct, which is known to inhibit adipogenesis (Ren et al, Genes Dev 16(l):27-32, 2002), not only abolished adipose formation but also reduced angiogenesis (Fukumura et al, 2003 supra). Reciprocally, inhibition of angiogenesis with a VEGFR2 blocking antibody reduced tissue growth and inhibited preadipocyte differentiation (Ren et al, 2002 supra). Although VEGF does not directly affect adipocyte differentiation, VEGFR2 signaling increases preadipocyte survival and differentiation in a paracrine manner (Fukumura et al, 2003 supra). The importance of growth factors in the molecular interplay between adipogenesis and angiogenesis has also been shown in earlier work. Kawaguchi et al, Proc Natl Acad Sci USA 95(3): 1062-1066, 1998, demonstrated that implanted Matrigel, supplemented with bFGF, induced neoangiogenesis followed by adipocyte precursor recruitment. Neoangiogenesis preceding adipogenesis is of interest since microvascular pericytes differentiate into adipocytes in vivo (Farrington-Rock et al, Circulation 110(15):2226- 2232, 2004) and recent findings that adipocytes and endothelial cells have a common progenitor (Planat-Benard et al, Circulation 109(5):656-663, 2004).

Delivery of different types of growth factors to free fat grafts in attempts to improve graft survival has been described in the literature. These have included the addition of bFGF complexed to dextran beads (Eppley et al, 1992 supra). Histologically, the treatment grafts had extensive intercellular collagen formation and small-sized adipocytes dispersed among the larger cells. It is unclear whether this resulted in better intergraft connective tissue in growth, persistence of the dextran beads, or bFGF-stimulated differentiation of preadipocytes or up-regulation of macrophages (Eppley et al, 1992 supra). Subsequent work suggested that the type of cytokine employed influenced cellular composition. Yuksel et al, 2000 supra evaluated insulin, insulin-like growth factor- 1 (IGF-I), and bFGF, alone and in combination, delivered by a PLGA/PEG microsphere system. Each growth factor treatment independently increased graft weight maintenance compared to controls at 12 weeks; however, there was no significant difference between the growth factors themselves, nor was there an additive or synergistic effect seen (Yuksel et al, 2000 supra).

There is a need to identify agents and in particular growth factors which alone or in combination with each other promote adipogenesis and the bioactivation of tissue including free fat grafts and to promote fat graft survival.

SUMMARY OF THE INVENTION

The present invention is predicated, in part, on the need to improve the integrity of grafted adipocytes including their precursors and mesenchymal stem cells. More particularly, the present invention provides the bioactivation of free fat grafts or seeded graft supports to promote adipogenesis and cellular migration and proliferation.

In one aspect, the present invention uses local or sustained delivery of platelet-derived growth factor (PDGF), vascular endothelial growth factor (VEGF) and/or fibroblast growth factor (FGF) or their homologs, splice variants or isoforms or the combined delivery of two or more of PDGF, VEGF and/or FGF to enhance adipogenesis of free fat grafts or seeded graft supports and/or to promote fat graft survival to thereby improve the weight maintenance and to preserve normal cellular architecture of tissue including free fat grafts in mammalian subjects. In particular, the present invention enables promotion of fat graft survival, enhanced adipogenesis and proliferation of adipocytes and their precursors. The methods of the present invention also promote angiogenesis in association with adipogenesis. Reference to adipogenesis includes a developmental process by which multipotent mesenchymal cells differentiate into a mature adipocyte (see Rosen, Prostaglandins Leukot Essent Fatty Acids 73:31-34, 2005).

Accordingly, one aspect of the present invention contemplates a method for facilitating tissue engineering in a mammalian subject the method comprising administering to a site of a free fat graft an adipocyte growth effective amount of one or more agents selected from PDGF, VEGF and FGF, such that the agent(s) is (are) provided locally or via sustained delivery or an adipocyte growth effective amount of two or more agents selected from PDGF, VEGF and FGF the administration being for a time and under conditions effective to promote weight maintenance and to preserve normal cellular architecture of free fat grafts in the mammalian subject. An "adipocyte growth effective amount" includes an amount of one or more growth factors to induce adipogenesis, proliferation of adipocytes or their precursors or mesenchymal stem cells or which promotes fat graft survival.

Another aspect of the present invention provides a method for promoting growth of adipocytes on or in a graft support implanted in a mammalian subject, said graft support comprising an agent selected from one or more of PDGF, VEGF and FGF in a sustained delivery form. Another aspect provides for a combination of two or more agents selected from PDGF, VEGF and FGF in amounts effective to promote adipogenesis in the graft support. A further aspect provides for a combination of PDGF, VEGF and FGF in amounts effectives to promote adipogenesis in the graft support. Further embodiments contemple the use of one or more of PDGF, VEGF and FGF in combination with one or more additional growth factors.

A "graft support" includes an extra-cellular matrix or extra-cellular matrix replacement, either synthetic or biological in nature.

Still another aspect of the present invention contemplates a method for grafting fat tissue in a mammalian subject the method comprising the steps of introducing adipocytes to a site in the mammalian subject together with or prior to the administration of one or more agents selected from PDGF, VEGF and FGF in a form for local or sustained delivery or two or more agents selected from PDGF, VEGF and/or FGF the administration being for a time and effective for weight maintenance and to preserve normal cellular architecture of the fat graft in the mammalian subject.

Still another aspect of the present invention provides a method for promoting angiogenesis in association with adipogenesis in a mammal the method comprising administering to a site of a free fat graft an adipocyte growth effective amount of one or more agents selected from PDGF, VEGF and FGF such that the agent is provided locally or via sustained delivery; or an adipocyte growth effective amount of two or more agents selected from PDGF, VEGF and FGF the administration being for a time and under conditions effective to promote weight maintenance and to preserve normal cellular architecture of free fat grafts in the mammalian subject.

Yet another aspect of the present invention is directed to the use of one or more of PDGF, VEGF and FGF in a sustained delivery form; or two or more of PDGF, VEGF and/or FGF in the manufacture of a medicament for the promotion of adipogenesis of a fat graft or seeded graft support in a mammalian subject and/or to promote fat graft survival.

Still another aspect of the present invention provides a sustained delivery composition comprising one or more of PDGF, VEGF and FGF and a matrix which facilitates sustained delivery.

The preferred mammal of the present invention is a primate such as a human. However, mouse and other animal models form part of the present invention.

The grafts may be allografts or xenografts although, for human use, allografts are preferred. The terms "allograft" and "autograft" may be used interchangeably throughout the subject specification.

Reference herein to "adipocytes" extends to adipocyte precursors including pre-adipocytes and mesenchymal stem cells. The growth factors of the present invention are proposed to promote proliferation of adipocytes and their precursors.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 is a graphical representation showing grafts mixed with PDGF bound to gelatin microspheres (Group 1) demonstrated superior weight maintenance compared to either at grafts alone (Group 2), free PDGF (Group 3), or blank microspheres (Group 4) (p=0.018).

Figure 2 is a graphical representation showing maintenance of normal adipocyte architecture. PDGF complexed to gelatin microspheres resulted in significantly greater maintenance of normal adipocyte architecture compared to free PDGF (p<0.0005).

Figure 3 is a photographic representation showing (a) superficial inferior epigastric pedicle (arrow)-, (b) silicone chamber sutured into the groin (arrow), (c) Bone wax sealing the base of the chamber, carefully avoiding injury to the vascular pedicle.

Figure 4 is a graphical representation showing the interaction of VEGF and PDGF in fat production.

Figure 5 is a graphical representation showing the interaction of FGF with VEGF in fat production.

Figures 6a is a graphical representation of adipose tissue volume at 6 weeks in the presence of different growth factors and combinations thereof.

Figure 6b is a graphical representation of triple growth factor group demonstrates highest percent vascular volume (PVV) at 2 weeks.

Figure 6c is a graphical representation of VEGF₁₂₀ + FGF-2, and the triple growth factor combination groups demonstrate increased new vascular cross sectional area at 2 weeks.

Figure 6d is a graphical representation of percent volume of connective tissue/inflammatory components at 2 weeks, demonstrating increases in all double and triple combinations. FGF-2 (p=0.008) and PDGF-BB (p=0.01) significantly increased this component.

Figure 7 is a diagrammatical representation of a mouse flow-through chamber model employing the superficial epigastric arteriovenous pedicle which supplies the inguinal fat pad.

Figure 8 is a graphical representation of mean chamber tissue volumes and standard error of the mean (SEM) bars for each group (in ml). Note TGF = triple growth factor, (F) = fat autograft (5 mg). ANOVA revealed a significant effect of the matrix (p<0.0005) and TGF (p<0.001) on the total tissue volume.

Figure 9 is a graphical representation of mean %ATVs and SEM bars for each group. ANOVA revealed a significant effect of the matrix (p<0.0005), TGF (p<0.0005) and autograft (p<0.0005) on the %ATV.

Figure 10 is a graphical representation of mean absolute adipose tissue volumes and SEM bars for each group (in μml). ANOVA revealed a significant effect of the matrix (p<0.0005), TGF (p<0.0005) and autograft (p=0.011) on the absolute adipose tissue volume.

DETAILED DESCRIPTION OF THE INVENTION

All scientific citations, patents, patent applications and manufacturer's technical specifications referred to hereinafter are incorporated herein by reference in their entirety.

The present invention defines the beneficial effects of the local or sustained application of one or more PDGF, VEGF and/or FGF; or the application of two or more of PDGF, VEGF and/or FGF on promoting adipogenesis and fat graft survival in mammalian subjects. This process of adipogenesis may be at the site of a free fat graft or in the promotion of cell or tissue growth in a seeded graft support implanted in a cavity or under the skin of a mammalian subject or the promotion of fat growth survival. Without limiting the present invention to any one theory or mode of action, the promotion of adipogenesis is proposed to be due to increased angiogenesis in the early period of tissue growth and promotion of proliferation of adipocyte precursors which migrate into the fat graft or tissue support during adipogenesis. There is a synergistic effect when two or more of PDGF, VEGF and/or FGF are applied or an efficacious effect with local high or sustained levels of one or more of PDGF, VEGF and/or FGF. It is proposed that the growth factors promote proliferation of adipocytes and their precursors.

In describing the present invention in detail, it is to be understood that unless otherwise indicated, the subject invention is not limited to specific agents including growth factors or combinations of growth, formulations of components, manufacturing methods, dosage or diagnostic regimes, or the like. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting. The present invention is directed to PDGF, VEGF and/or FGF in a sustained release formulation or a combination of two or more of PDGF, VEGF and/or FGF in a single formulation or in a sequentially provided or admixable formulation.

Reference to "two or more" includes PDGF + VEGF, PDGF + FGF, VEGF + FGF and PDGF + VEGF + FGF. The use of PDGF in a sustained delivery form or provided locally or PDGF in combination with one or both of VEGF and/or FGF are particularly preferred. Reference to "platelet-derived growth factor" or "PDGF" includes all its homologs, splice variants and isoforms such as PDGF-AA, PDGF-BB, PDGF-AB, PDGF-CC or PDGF-DD.

Reference to "vascular endothelial growth factor" or "VEGF" includes all its homologs, splice variants and isoforms and in particular VEGF-A-120, -121, -165, -189, -206, VEGF- B, VEGF-C, VEGF-D and VEGF-E.

Reference to "fibroblast growth factor" or "FGF" includes all its homologs, splice variants and isoforms and in particular FGF-2 (bFGF).

The PDGF, VEGF and FGF molecules may be in purified, naturally occurring or recombinant form. The preferred molecule is derived from the same species as the subject being treated. Hence, a homologous growth factor for use in humans, for example, is a human-derived growth factor. However, heterologous growth factors are also contemplated, such as a murine growth factor for use in humans. Humanized or other mammalianized forms are also contemplated by the present invention. In addition, the growth factors may be part of a formulation and further comprising one or more pharmaceutically acceptable carriers, diluents and/or excipients.

Accordingly, one aspect of the present invention contemplates a method for facilitating tissue engineering in a mammalian subject said method comprising administering to a site of a free fat graft an adipocyte growth effective amount of an agent selected from PDGF, VEGF and FGF such that the agent is provided locally or via sustained delivery or an adipocyte growth effective amount of two or more agents selected from PDGF, VEGF and FGF said administration being for a time and under conditions effective to promote weight maintenance and to preserve normal cellular architecture of free fat grafts in the mammalian subject.

Hence, the method of the present invention leads to one or more of promotion of adipogenesis, enhancement of adipocyte precursor proliferation, enhancement of graft survival and/or enhancement of angiogenesis. In one embodiment, two or more, in another embodiment, three or more and in a further embodiment, all four of the above features occur.

An "adipocyte growth effective amount" includes an amount of growth factor to induce adipogenesis, proliferation of adipocytes or their precursors or mesenchymal stem cells or which promotes fat graft survival. "Adipogenesis" refers to the differentiation of a mesenchymal cell to a mature adipocyte.

Another aspect of the present invention provides a method for promoting growth of adipocytes in a graft chamber implanted in a mammalian subject, said graft support comprising an agent selected from PDGF, VEGF and FGF in a sustained delivery form or a combination of two or more agents selected from PDGF, VEGF and FGF in amounts effective to promote adipogenesis in the graft support.

Still another aspect of the present invention contemplates a method for grafting fat tissue in a mammalian subject said method introducing adipocytes to a site in the mammalian subject together with or prior to the administration of an agent selected from PDGF, VEGF and FGF in a form for local or sustained delivery or two or more agents selected from PDGF, VEGF and/or FGF said administration being for a time and effective for weight maintenance and to preserve normal cellular architecture of said fat graft in said mammalian subject.

Reference herein to "adipocytes" extends to adipocytes precursors including pre- adipocytes and mesenchymal stem cells.

The singular forms "a", "an" and "the" include plural aspects unless the context clearly dictates otherwise. Thus, for example, reference to "a growth factor" includes a single growth factor as well as two or more growth factors; reference to "an agent" includes a single agent, as well as two or more agents; reference to "the formulation" includes a single formulation or multiple formulations; and so forth. In describing and claiming the present invention, the following terminology is used in accordance with the definitions set forth below.

The terms "agent", "chemical agent", "compound", "pharmacologically active agent", "medicament", "active" and "drug" are used interchangeably herein to refer to a growth factor or functional equivalent or analog that induces a desired pharmacological and/or physiological effect such as inducing angiogenesis in association with adipogenesis. In particular, it is proposed that the agent of the present invention promote adipogenesis, or have adipocyte precursor proliferation, enhanced fat graft survival and/or enhanced angiogenesis.

The terms also encompass pharmaceutically acceptable and pharmacologically active ingredients of those active agents specifically mentioned herein including but not limited to salts, esters, amides, prodrugs, active metabolites, analogs and the like. When the terms "agent", "chemical agent" "compound", "pharmacologically active agent", "medicament", "active" and "drug" are used, then it is to be understood that this includes the active agent per se as well as pharmaceutically acceptable, pharmacologically active salts, esters, amides, prodrugs, metabolites, analogs, etc. Hence, the growth factors contemplated herein are proposed to be useful in promoting fat graft survival and growing adipose tissue in seeded supports.

Reference to a "agent", "chemical agent", "compound", "pharmacologically active agent", "medicament", "active" and "drug" includes combinations of two or more active agents. A "combination" also includes multi-part such as a two-part or multi-part composition where the agents are provided separately and given or dispensed separately or admixed together prior to dispensation. For example, a multi-part pharmaceutical pack may have two or more agents separately maintained. Hence, this aspect of the present invention includes combination therapy in promoting adipogenesis of free fat grafts or of tissue growing in a seeded graft support. Reference to a "seeded support" includes a support in the form of a solid or hollow or partially hollow moulded chamber or matrix which is implanted surgically or via a catheter to a cavity in a mammalian body.

Generally the support is attached to a blood supply via a superficial inferior epigastric flow-through pedicle or anatomic neurovascular pedicles. The support is then seeded with cells such as stem cells, preadipocytes and/or adipocytes and then given sustained release single or multiple growth factors such as PDGF, VEGF and/or FGF or two or more of these growth factors are given locally or in close proximity to the seeded chamber. In particular, a "graft support" may be a synthetic or biological extra-cellular matrix or extracellular matrix replacement.

Reference to a "support" should not imply any structural limitation and it includes a silicone or other material in the form of a tube as well as a chamber.

The terms "effective amount" and "therapeutically effective amount" of an agent as used herein mean a sufficient amount of the agent to provide the desired therapeutic or physiological or genetic effect or outcome. Such an effect or outcome includes facilitating adipogenesis thereby promoting weight maintenance and preserving normal cellular architecture of tissue or free fat graft in mammalian subjects as well as promoting proliferation of adipocytes or their precursors which migrate into the fat graft or tissue support during adipogenesis. It similarly applies to the growth of tissue in a seeded graft support. Undesirable effects, e.g. changes in cell morphology are sometimes manifested along with the desired growth factor effect; hence, a practitioner balances the potential benefits against the potential risks in determining what is an appropriate "effective amount" of growth factor. The exact amount required will vary from subject to subject, depending on the species, age and general condition of the subject or cells, mode of administration and the like. Thus, it may not be possible to specify an exact "effective amount". However, an appropriate "effective amount" in any individual case may be determined by one of ordinary skill in the art using only routine experimentation. However, amounts of growth factors range from about 10 ng/ml to about 1 mg/ml such as 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 ng/ml, 200, 300, 400, 500, 600, 700, 800, 900 or 1000 ng/ml, 10, 20, 30, 40, 50, 60, 70, 80, 90 or 100 μg/ml and 200, 300, 400, 500, 600, 700, 800, 900 or 1000 μg/ml are contemplated by the present invention. The volume may be local or systemic volume. Alternatively, sustained release amounts of from about 10 ng/min to 1000 μg/min may also be applied. This may also be expressed as lng/ml/min to lOOμg/ml/min. The amounts are referred to as an "adipocyte growth effective amount" which includes an amount sufficient to include proliferation of adipocytes or their precursors and stem cells.

By "pharmaceutically acceptable" carrier, excipient or diluent is meant a pharmaceutical vehicle comprised of a material that is not biologically or otherwise undesirable, i.e. the material may be administered to a subject along with the selected active agent without causing any or a substantial adverse reaction. Carriers may include excipients and other additives such as diluents, detergents, coloring agents, wetting or emulsifying agents, pH buffering agents, preservatives, and the like.

Similarly, a "pharmacologically acceptable" salt, ester, emide, prodrug or derivative of a compound as provided herein is a salt, ester, amide, prodrug or derivative that this not biologically or otherwise undesirable.

"Treating" a subject may involve promoting the growth of fat cells or other tissue or preventing transformation of adipocytes to cancerous cells. It also involves promoting fat graft survival. In particular, the present invention contemplates promoting weight maintenance and preserving normal cellular architecture of free fat grafts and other tissue in mammalian subjects.

The present invention defines the beneficial effects of the local or sustained application of

PDGF, VEGF and/or FGF or the application of two or more of PDGF, VEGF and/or FGF on promoting adipogenesis and well as promoting adipocyte or adipocyte precursor cell proliferation and/or mesenchymal stem cell proliferation in mammalian subjects. This process of adipogenesis may be at the site of a free fat graft or in the promotion of cell or tissue growth in a seeded graft support implanted in a cavity or under the skin of a mammalian subject. Without limiting the present invention to any one theory or mode of action, the promotion of adipogenesis is proposed to be due to increased angiogenesis in the early period of tissue growth as well as the promotion of proliferation of adipocyte precursors which migrate into the fat graft or tissue support during adipogenesis. In particular, it is proposed that the beneficial effects include promotion of adipogenesis, enhancement of adipocyte precursor proliferation, enhancement of fat graft survival and/or enhancement of angiogenesis. There is a synergistic effect when two or more of PDGF, VEGF and/or FGF are applied or an efficacious effect with local high or sustained levels of one or more of PDGF, VEGF and/or FGF. As indicated above, PDGF is the preferred growth factor alone or in combination with other growth factors.

Another aspect of the present invention contemplates a method for grafting fat tissue in a mammalian subject said method introducing adipocytes or their precursors or mesenchymal stem cells to a site in the mammalian subject together with or prior to the administration of an agent selected from PDGF, VEGF and FGF in a form for local or sustained delivery or two or more agents selected from PDGF, VEGF and/or FGF said administration being for a time and effective for weight maintenance and to preserve normal cellular architecture of said fat graft in said mammalian subject.

The present invention is also directed to the use of PDGF, VEGF and/or FGF in a sustained delivery form or two or more of PDGF, VEGF and/or FGF in the manufacture of a medicament for the promotion of adipogenesis of a fat graft or seeded graft support in a mammalian subject.

Still another aspect of the present invention provides a sustained delivery composition comprising one or more of PDGF, VEGF and/or FGF and a matrix which facilitates sustained delivery.

Accordingly, the present invention provides compositions and methods useful in facilitating the success of fat grafting and growth of tissues in seeded supports.

The growth factors promote adipogenesis, adipose cell proliferation as well as angiogenesis associated with adipogenesis. After a suitable period of time, tissue grows and can be used in free fat grafts or other tissue grafts.

Reference to "graft" includes allografts and xenografts although allografts are preferred, especially when the intended recipient is a human. An allograft includes an autograft.

The term "fat graft" is useful to define the removal of adipocytes from one part of a body or from a body of another subject and transplanting them to the same or a different subject. Often this is done to fill depressions or to augment tissue (external or internal) or other trauma. Such grafting is useful for reconstructive, cosmetic and correctional reasons. Examples of reconstructive surgery includes oncologic resections, complex trauma and congenital abnormalities; cosmetic surgery includes augmentation therapy such as for breasts, cheeks, chins, jaws, lips and buttocks; correctional surgery includes implant removal and augmentation for breasts and other soft tissue.

A "depression" includes any hollow or sunken area in a body requiring cellular substance. These may arise following, for example, trauma or disease.

"Adipocytes" include fat storing cells taken from the same or a different subject for use in the graft or to seed the chamber. Generally, such cells are removed by suctioning although any of a range of techniques may be used. Common sources in humans and other mammals include the stomach, legs, buttocks or other areas where fat cells are located. As indicated above, an "adipocyte" as recited herein includes adipocyte precursors such as pre-adipocytes and mesenchymal stem cells.

For sustained release, any type of matrix may be employed such as gelatin capsules and gelatin microspheres. Gelatin microspheres are particularly preferred prepared using glutaraldehyde cross-linking of a gelatin aqueous solution using an emulsion method. The growth factor(s) is incorporated into microspheres by placing the aqueous solution of the growth factor on vacuum dried microspheres.

The present invention is further described by the following non-limiting Examples.

EXAMPLE 1 Effects of PDGF on fat graft survival

PDGF Purified human or recombinant human PDGF (PDGF-I and PDGF-2) are commercially available from R and D Systems, (Minneapolis, Minn.), UBI (Lake Placid, N.Y. USA), and Genzyme Corporation (Boston, Mass.).

Recombinant PDGF Platelet-derived growth factor (PDGF) derived from human platelets contains two polypeptide sequences (PDGF^-A) and PDGF²(-B) polypeptides (Antoniades and Hunkapiller, Science 220(4600):963-965, 1983). PDGF¹^A) is encoded by a gene localized in chromosome 7 (Betsholtz et al, Nature 320(6064):695-699, 1986) and PDGF²(-B) is encoded by the sis oncogene (Doolittle et al, Science 185(148):368-370, 1974; Waterfield et al, Nature 304(5921):35-39, 1983) localized in chromosome 22. The sis gene encodes the transforming protein of the Simian Sarcoma Virus (SSV) which is closely related to PDGF-B polypeptide. The human cellular c-sis also encodes the PDGF-B chain (Rees et al, Embo J 3(8): 1843-1847, 1984; Rao et al, Cold Spring Harb Symp Quant Biol 51 (2):959-966, 1986). Because the two polypeptide chains of PDGF are encoded by two different genes localized in separate chromosomes, the possibility exists that human PDGF consists of a disulfide-linked heterodimers defined as PDGF-A/B or a mixture of homodimers (homodimers PDGF-AA, -BB, -CC or -DD), or a mixture of the heterodimer and the two homodimers. Recombinant preparation of biologically active PDGF-AA, -BB, -AB, -CC or -DD dimers and of their anologs can be obtained by introduction of cDNA clones encoding csis/PDGF-AA, -AB- -BB, -CC or -DD genes into enkaryotic cells using appropriate expression systems (Institute of Molecular Biology, Inc., Boston, Mass.); U.S. Pat. No. 4776073 (Murray et al, I) [King et al, Proc Natl Acad Sd USA 82(16):5295-5259, 1985; Clarke et al, Nature 308(5958):464-467, 1984]. Expression of the biologically active dimeric v-sis protein product is ssv-infected NRK cells has been reported (Owen et al, Science 225(4657):54-46, 1984). Expression in procaryotes produced biologically inactive single chain protein product (Devare et al, Cell 36(1 ):43-49, 1984). Refolding of the single chain produced by procaryotes into its dimeric form produced biologically active PDGF preparations (Hoppe et al, Biochemistry 28(7):2956-2960, 1989).

Mammalian cells in culture infected with the Simian Sarcoma Virus which contains the gene encoding the PDGF chain were shown to synthesize the PDGF polypeptide and to process it into disulfide-linked homodimers with molecular weights of about 35,000 and 24,000 (Robbins et al, Nature 205(5935):605-608, 1983). In addition, PDGF homodimer reacts with antisera raised against human PDGF. Furthermore, the functional properties of the secreted PDGF homodimer are similar to those of platelet-derived PDGF in that it stimulates DNA synthesis in cultured fibroblasts, it induces phosphorylation at the tyrosine residue of a 185 kd cell membrane protein, and it is capable of competing with human PDGF for binding to specific cell surface PDGF receptors (Owen et al, 1984 supra). Similar properties were shown for the sis/PDGF gene product derived from cultured normal human cells (e.g. human arterial endothelial cells), or from human malignant cells expressing the sis/PDGF gene (Graves et al, Biochem Biophys Res Commun 129(l):56-62, 1985; Pantazis et al, Proc Natl Acad Sci USA §2^:2404-2408, 1985).

The identification and cloning of the gene encoding the PDGF chain (Betsholtz et al, 1986 supra) allowed the expression of its biologically active homodimer and the demonstration that the homodimer has functional activities similar to those of human PDGF. Receptor binding studies have shown that the PDGF-2 homodimer binds with high affinity and the human PDGF heterodimer with lower affinity to PDGF receptor beta (PDGF-Rβ) [Hart et al, Science 240(4858):l529-153l, 1988; Heldin et al, Embo J 7^:800-804, 1988].

The PDGF-Rβ did not recognize the PDGF homodimer; the latter was bound to a second receptor, the PDGF receptor alpha (PDGF-Rα). This receptor also bound the other two isoforms, the human PDGF and the PDGF homodimer with high affinity (Heldin et al, 1988 supra). The PDGF-Rα was cloned by Matsui et al, Science 243(4892):&00-804, 1989 and by Claesson- Welsh et al, J Biol Chem 264(3):1742-1747, 1989. This α receptor is structurally similar to the beta receptor sharing a 40% sequence identity, with an external binding domain and an intracellular kinase domain (Hart et al, 1988 supra; Heldin et al, 1988 supra; Matsui et al, 1989 supra; Claesson- Welsh et al, 1989 supra; Bryan and Hartl, Science 240(4849):215-217, 1988; Linette et al, Science 241(4865):573-576, 1988; Malhotra et al, Science 242(4879):! 55-159, 1988; McConkey et al, Science 242(4876):256-259, 1988; Pearson et al, Science 241 (4868)\91 '0-973, 1988; Thomas et al, Science 242(4881):\050-\053, 1988; Weinert and Hartwell, Science 241 (4863) :317-322, 1988; White and Hartzell, Science 239(4841 Pt l):778-780, 1988; Wilks et al, Science 242(4885):\541-\544, 1988; Yarden et al, Nature 323(6085):226-232, 1986).

Preparation of Gelatin Microspheres Gelatin microspheres were prepared through Gluteraldehyde (GA) (Sigma Aldrich Pty Ltd, Australia) cross-linking of a gelatin aqueous solution using an emulsion method. Immediately after mixing lOμL of GA aqueous solution (25% w/v) with 4 mL of 10% w/v gelatin aqueous solution preheated at 4O⁰C, the mixed solution was added drop wise to 150 ml of olive oil under stirring at 500 rpm and 4O⁰C in a standard 25OmL glass beaker to obtain a water-in-oil emulsion. After stirring the reaction, mixture for 15 minutes at 4O⁰C, the temperature was dropped to 25⁰C and stirring was continued for 24 hours to allow the gelatin to chemically crosslink. Acetone (4OmL) was added to the reaction mixture, and stirring was continued for another hour. The resulting microspheres were allowed to settle at the bottom of the centrifuge tubes for 1 hour. The oil phase was removed and the microspheres were washed five times with acetone by centrifugation [4⁰C, 3000 rpm, 5 minutes (3K30, Sigma Laboratory Centrifuges)]. The washed microspheres were placed in 100ml of 0.1 M glycine aqueous solution containing Tween 80 (0.1% w/v) and kept on stirring at 37⁰C and 1000 rpm for 1 hour to block the residual aldehyde groups of any unreacted GA. The cross-linked gelatin microspheres were then washed with MiUi-Q water and collected by centrifugation (4⁰C, 3000 rpm, and 5 minutes). The microspheres were disinfected by storing them in 70% v/v ethanol (Wissemann and Jacobson, In Vitro Cell Dev Biol 21 (7):391-40\, 1985). Finally, the microspheres were placed under vacuum for 24 hours to remove the ethanol before incorporating the PDGF. PDGF was incorporated into microspheres by dropping the aqueous solution of PDGF on vacuum dried microspheres as described below. Morphological Analysis of Gelatin Microspheres

Gelatin microspheres were coated with gold using a sputter coater and examined with a scanning electron microscope (Philips Model XL30 SEM, Netherlands) using an accelerating voltage of 12kV. For size distribution studies, particles were stained with food color and examined in a Beckman Coulter apparatus (LS- 130, Langley Ford Instruments). Cryo-SEM (CT 1500, Oxford Instruments) analysis was used to study the distribution and integration of the gelatin microspheres in collagen gel. The neutralized collagen solution (200μL) carrying the microspheres was placed onto an SEM planchett using a 21 -gauge needle. It was then plunged into liquid propane, and thereafter into liquid nitrogen. The sample was fractured at -185⁰C using a pre-cooled blade and sublimed at -8O⁰C for an hour prior to imaging. The image was taken at a working distance of 24mm using 2kV. SEM observations revealed that the gelatin particles were spherical in shape with smooth surface characteristics (Figure 2). The particles (70% v/v) were in the range of 20-160 μm in diameter in a swollen state according to the Coulter apparatus, which was acceptable for our experiments. The literature suggested that microspheres with diameters in the range of 20-100 μm can be effectively used for subcutaneous and intramuscular administration (Esposito et al, Biomaterials 17 (2) :2009-2020, 1996).

Preparation of Suspension of Gelatin Microspheres with PDGF The original PDGF-BB solution was diluted with Milli-Q water to a concentration of lμg/mL. A 50 μl aliquot of this solution was dropped onto 20 mg of vacuum dried gelatin particles (10 aliquots), which were left at room temperature for overnight impregnation of PDGF-BB into microspheres. For control group, 50 μl of phosphate buffered saline was dropped on to 20 mg of microspheres (10 aliquots). In general, a growth factor molecule possesses positively charged sites on its molecular surface under physiological conditions (Faham et al, Science 27/(15252): 116-1120, 1996). Gelatin microspheres prepared from acidic gelatin carry negative charge on their surface and can electrostatically interact with positively charged growth factors to allow physical immobilization. Gelatin microspheres have been used in number of studies for controlled delivery of positively charged growth factors such as FGF-2 21 PDGF-BB is a cationic protein with an isoelectric point of 9.8 (Heldin et al, Physiol Rev 79(4):1283-1316, 1999) and thus, it was hypothesized that it would bind to gelatin (isoelectric point 5.0) microspheres due to electrostatic interaction.

Improved fat graft weight maintenance and cellular architecure To determine the effectiveness of PDGF bound to gelatin microspheres in improving fat graft weight maintenance and cellular architecture, the following experiments were performed.

Animals Severe combined immune deficient (SCID) mice weighing between 20 and 25g were used for this experiment. The SCID mouse model allows for study of transplanted human fat without the immunologic sequelae of rejection (Bosma et al, Nature 201(5900):527-530, 1983).

Fat-Graft and Experimental Design

Fat was harvested by excision from the abdomen of an otherwise healthy 43-year-old woman during a breast reconstruction performed under general anesthesia. The tissue was washed twice with sterile phosphate-buffered saline, sectioned into small pieces with a scissors, and then passed repeatedly through a 3 -ml syringe attached to a 14-gauge needle until it reached a gelatinous consistency to simulate the mechanical disruption of cannular harvesting, as previously described in a related model (Eppley et al, 1992 supra).

The pericranial region of the mouse was chosen as the recipient site because of the absence of native fat in this area. Graft implantation was done by percutaneous injection (14-gauge needle) under general anesthesia (chloral hydrate administered intraperitoneally at 0.4 mg / g body weight). Prior to injection, the fat grafts were divided into 1 ml aliquots, mixed with microspheres or free PDGF depending on the experimental group, and weighed. The following experimental groups were thus created (minimum n=8 per group): (1) fat graft control, (2) fat graft with free PDGF, (3) fat graft with blank microspheres, and (4) fat graft with microspheres bound to PDGF. All grafts were harvested 12 weeks after graft insertion and evaluated for weight maintenance (Figure 1), histological changes, and histomorphometry (Table 2).

Histology and Histomorphometry Twelve weeks following injection, the grafts were harvested and fixed in 4% v/v paraformaldehyde (4% v/v PFA), sliced into 1-2 mm thick vertical slices, embedded in paraffin, sectioned to a width of 5 μm, and stained routinely with Masson's trichrome. Slices were embedded in a single block, and multiple fields from each section were viewed to evaluate the adipose tissue histology. For histomorphometry, complete histological sections of two slices per graft were counted from 6 animals in each group. The percentage of cellular infiltrate, adipocytes with normal architecture, fibrous tissue, cysts, and adipocytes with abnormal morphology were determined by point-counting using digital video imaging (JVC, TK C 1480E) and an automated, systematic random sampling point- counting system (CAST system, Olympus, Denmark) and the percent volume density determined (Cao et al, Biomaterials 27(7^:2854-2864, 2006). Shrinkage from histological processing was assumed constant for all tissue types (Beech et al, J Microsc 197(1). -26-45, 2000).

Statistical Analysis Groups were compared using one and two way Analysis of Variance (ANOVA) on SPSS vl2.0. In the two-way ANOVA, presence of spheres and addition of PDGF-BB were treated as the two independent variables. Alpha was set at 0.05.

Gross Observations There was no gross indication of acute inflection, inflammation, abscess formation, seroma, or necrosis of the graft itself or the adjacent host tissue in any animal. Macroscopic observation of the fat grafts in all animals revealed consolidation of the grafted tissue into a distinct mass, facilitating easy identification and harvest of all grafts. All fat-grafts were surrounded by a thin layer of loose connective tissue, which separated the donor adipose tissue from the recipient bed in all cases. Weight-Maintenance Analysis

In accordance with the literature, fat-graft weight maintenance was used as an index of volume stability (Eppley et al, 1992 supra; Yuksel et al, 2000 supra; Smith et al, 2006 supra; Eppley et al, J Oral Maxillofac Surg 6O(5)-Λ11-A%2, 1992; Eppley and Sadvoce, Aesthetic Plast Surg 15(3):215-21S, 1991). Weight maintenance was calculated as a percent by dividing graft weights at harvest by pre-grafting weights. The fat grafts mixed with PDGF bound to gelatin microspheres demonstrated a significant improvement in weight maintenance at 12 weeks compared to the control groups (p=0.018) (Figure 1).

Histology

Masson's trichrome-stained sections of the grafted fat at 3 months demonstrated qualitative differences among the treatment and three control groups. The PDGF bound to gelatin microspheres group contained relatively large areas of adipocytes organized into lobules with largely intact architecture. Small fat cysts surrounded by mild fibrous ingrowth and cellular infiltrate that included cuboidal cells surrounding the cysts was present at the periphery of all grafts in this group.

The three control groups demonstrated variable morphology. Areas of adipocytes organized into lobules and surrounded by small fat cysts and cellular infiltrate as described above were observed in some specimens. However the major portion of the control groups were disorganised areas of very large irregular confluent fat cysts intermingled with tracts of fibrous connective tissue.

Histomorphometric Analysis The experimental group (PDGF bound to gelatin microspheres) demonstrated a significantly greater area of adipocytes with intact architecture (44.63% + 6.97) when compared to the control groups: fat-graft control (3.4% + 3.15), free PDGF (8.98% + 9.06) and blank microspheres (7.33% + 8.33) [p<0.0005]. The groups had variable relative amounts of cellular infiltrate, adipocytes with disrupted architecture, fibrous tissue and cysts (Table 1). TABLE 1

Percentages of (1) cellular infiltrate, (2) adipocytes with normal architecture, (3) cysts, (4) fibrous tissue and (5) adipocytes with disrupted architectural in the four experimental groups

Medical Use

The results of these experiments indicate that treatment of the aging face, facial hemiatrophy, or lipodystrophy may be treated with the process described herein. Although free fat grafts are widely used in cosmetic surgery, their variable longevity requires frequent, repeat procedures. This method improves the weight maintenance of these grafts and therefore may lengthen time needed between treatments. Further, this method increases the area of normal adipocytes, and by inference, results in a superior graft.

The formulations of the present invention may be administered by pretreatment of the fat autografts or by local injection into the recipient bed prior to grafting. The compounds provided herein can be formulated into any pharmaceutically acceptable excipients or carrier, especially other microsphere delivery systems. EXAMPLE 2 Growth factor synergism in a murine tissue-engineering model

Anesthesia, Animals, and Chambers Wild-type male C57BL6 mice weighing between 20 and 25 grams were used for the experiments. All surgical procedures were performed under general anesthesia (chloral hydrate administered intraperitoneally at 0.4 mg/g body weight). The cylindrical noncollapsible silicone chambers were cut to a length of 5 mm with an internal diameter of 3.35mm (for a maximum volume of 44 μl) from standard laboratory tubing (Dow-Corning Corp., Midland, MI, USA). Hair on the abdominal wall and groin was removed with a depilatory cream and the skin cleansed with chlorhexidine and alcohol.

Matrigel and Growth Factors

Complete Matrigel was purchased from BD Biosciences (Bedford, MA, USA). Recombinant human Fibroblastic Growth Factor-basic (rh FGF -b) was purchased from CytoLab Ltd (Rehovot, Israel). Recombinant rat Platelet Derived Growth Factor-bb (rr PDGF-bb) and recombinant mouse Vascular Endothelial Growth Factor- 120 (rm VEGF- 120) were purchased from R&D Systems (Minneapolis, MN, USA). The growth factors were reconstituted to a final concentration of 50μg/ml according to the manufacturer's specifications. On the day of animal surgery, the growth factors were added to the Matrigel as follows: 100ng/ml of growth factors in combination or separately were mixed with 80U/ml of Heparin (Pharmacia & Upjohn, Kalamazoo, Michigan, USA) and added to the Matrigel. This solution was kept on ice prior to use to prevent the Matrigel from forming a gel. When the chamber was ready for seeding, 44μl of growth factor/Matrigel solution was injected into the chamber and allowed to set for 1-2 minutes before the chamber was sealed. Surgical Techniques

Chambers were placed bilaterally according to a method modified from Cronin et al. Plast Reconstr Surg 113(l):260-9, 2004. (Figure 3). Briefly, a transverse groin incision is made just above the groin fat pad. The superficial epigastric vessels are dissected from their origin at the femoral vessels to their insertion into the groin fat pad. The fat pad is then mobilized to create a space for chamber placement. The cylindrical chamber is longitudinally cut along one side and placed around the first 8 to 10 mm of the superficial epigastric vessels, where they have been freed from the fat pad. The chamber is then anchored to the underlying muscle near the origin of the superficial epigastric vessels with 10-0 nylon sutures. The proximal femoral end and the lateral split are then sealed with melted bone wax (Ethicon, Somerville, NJ, USA). The chamber is then filled with Matrigel containing the appropriate combination of growth factors depending on the animal's experimental group. The distal end of the chamber is sealed with melted wax, taking care not to damage the epigastric 7 artery and vein exiting the chamber. The construct is placed in the dissected plane in the groin lateral to the femoral vessels. The skin wound is then closed with metal clips.

Chamber Removal and Assessment

All animals were sacrificed at either 2 or 6 weeks post-chamber insertion and the construct exposed surgically. Vessel patency within the chamber was assessed by direct inspection and by presence or absence of bleeding from the cut ends of the pedicle as the chamber was removed. Contents of the chamber were removed, weighed, and the volume determined by displacement in a 0.9% w/v NaCl solution at room temperature (Weibel, Practical Methods for Biological Morphometry (1), 1979).

Morphology: Tissue was routinely fixed in 4% v/v paraformaldehyde (4% v/v PFA) and then embedded in paraffin. Serial histological sections were stained routinely with haematoxylin and eosin, and Masson's trichrome and slides examined light microscopically. Morphometry: The percentage of adipose tissue was determined for 6 week constructs on haematoxylin and eosin stained sections. In 2 week specimens the percent vascular - - volume (CD31 positive blood vessels) and associated percent of connective tissue components around the blood vessels was determined. In 2 and 6 week chambers tissue percent volumes of tissue components were determined by point-counting every 20th 5 μm section of each specimen using digital video imaging (JVC, TK C 1480E) and an automated, systematic random sampling point-counting system (CAST system, Olympus, Denmark). In 6 week assessments the proportion of adipose tissue was then multiplied by the total volume of the chamber to determine the absolute volume of adipose tissue. In the 2 week groups the proportion of blood vessels was multiplied by the total area of new tissue growth around the pedicle to determine the total cross sectional area of blood vessels in each group. Shrinkage from histological processing was assumed constant for all tissue types.

Immunohistochemistry (CD31 Staining)

Sections from 2 week chambers (3-5 chambers/group) were dewaxed and hydrated according to a standard lab protocol, and immunohistochemical staining for CD31 (a mouse endothelial cell marker) performed using a DAKO (Registered) Autostainer Universal Staining System (Dako Corporation, California, USA). In this protocol, sections were quenched with Dako peroxidase block (Dako Corporation, California, USA) for 5 minutes, followed by digestion with Dako proteinase K (Dako Corporation, California, USA) for 8 minutes. Non-specific binding was then blocked using Dako protein block (Dako Corporation, California, USA) for 30 minutes.

The primary antibody (rat anti-mouse CD31 ; BD Pharmingen, California, USA) was then applied at 1 :100 in Dako antibody diluent (Dako Corporation, California, USA) for 1 hour. This was followed by the secondary antibody (Dako rabbit anti-rat biotinylatedimmunoglobulin; Dako Corporation, California, USA) at 1 :300 for 30 minutes. An avidin-biotin-horse radish peroxidase detection system (ABC elite, Vector Laboratories, California, USA) was used for 30 minutes, followed by visualization with Dako DAB chromagen (Dako Corporation, California, USA) for 5 minutes. Slides were then removed from the autostainer, counter-stained with haematoxylin, dehydrated, and coverslipped using DPX mounting medium. In some cases counterstaining was performed with both haematoxylin and eosin to distinguish blood cells. AU washes were done with Tris buffered saline (TBS) pH 7.6, and all steps performed at room temperature.

Experimental Design and Statistical Analysis

In the 6 week groups the three treatment applications: rh FGF-b; rr PDGF-bb; and rm VEGF- 120 were combined factorially to give a total of six experimental groups including a control group with no treatment. There were six animals per group. All growth factors were used at a concentration of 100ng/ml for each growth factor. The factorial design allowed a comparison by 4 way ANOVA on SPSS for Windows (version 12.0.1, SPSS corporation). Comparisons were made for the main effect of each treatment across all animals in the study as well as assessing interactions between each treatment. Synergy between treatments was assessed by significance in the measures of interaction. Alpha was set at 0.05.

In the 2 week groups the factorial design allowed a comparison by 3 way ANOVA on SPSS for Windows (version 12.0.1, SPSS corporation). Comparisons were made for the main effect of each treatment across all animals in the study as well as assessing interactions between each treatment. Synergy between treatments was assessed by significance in the measures of interaction. Alpha was set at 0.05.

RESULTS

Adipose Tissue Volume

The addition of each of the growth factors resulted in an increase in the percentage and overall volume of fat in the construct at 6 weeks (Figure 6a,). These increases were in each case highly significant (p<0.0005 for each main effect on volume of adipose tissue). In addition to the individual effects of growth factors on fat production, there was also strong evidence of synergy between the three growth factors in fat production (p value for overall interaction of VEGF, PDGF and FGF is less than 0.0005). In particular VEGF and PDGF had a significant positive interaction (p < 0.0005) (Figure 4) as did VEGF with FGF (p = 0.001) (Figure 5). A positive interaction was also present between PDGF and FGF (p = 0.007) although to a lesser degree.

When interaction was sought between the presence of growth factors and the fat graft, only PDGF showed a significant interaction (p = 0.021) the presence of PDGF increasing the positive effect of the fat graft on the volume of fat produced.

The effect of growth factors administered either alone or in combination, including the combination of all three of PDGF, VEGF and FGF (i.e. triple growth factor or TGF), on adipose volume is shown in Figure 6a.

Vessel Volume Density at 2 weeks

The cellularity of the invading tissue increased around the pedicle and new capillary sprouts in double and triple growth factor combinations compared to control being particularly apparent in the VEGF₁₂₀ + FGF-2 and the triple growth factor combination groups.

Vascular tissue: The triple growth factor group demonstrated the highest percent vascular volume (9.5+/- 1.9%) compared to control (6.3+/-0.96%, Figure 6b). Total new blood vessel area was markedly increased in the VEGFi_2O + FGF-2 (0.146 +/-0.29 mm²) and triple growth factor groups (0.16 +/- 0.25 mm²), compared to control (0.06 +/- 0.00005 mm², Figure 6c) . FGF-2 was found to significantly increase the vasculature (p< 0.001), and there was a significant interaction (synergy) between FGF 2 and VEGF₁₂₀, (p=0.019).

Connective tissue: Two or more growth factors increased the connective tissue/inflammatory infiltrate, regardless of combination, compared to control and single growth factor administration (Figure 6d). Both FGF-2 (p=0.008) and PDGF-BB (p=0.01) significantly increased this component. There were no significant interactions.

EXAMPLE 3 The effect of Triple growth factor (TGF) on fat graft survival and adipogenesis in various extracellular matrices.

Triple growth factor (TGF) combination of bFGF, VEGF-A₁₂₀ and PDGF-bb (100 ng/ml each) significantly increased the adipose tissue volume formed at 6 weeks in growth factor reduced (GFR) Matrigel™ (Figure 6a). Therefore, this TGF regime was used to establish if this effect was translatable to other matrices. Adipogenesis was examined in a number of extracellular matrices (Myogel™, Cymetra™ and Puramatrix™) placed in an in vivo murine vascular flow-through chamber model for 6 weeks. Severe combined immunodeficient (SCID) mice were divided into 8 groups - GFR Matrigel™ (as a control), Myogel™, Cymetra™ and Puramatrix™ groups, with and without the TGF. All mice received bilateral chambers (one containing a fat autograft, the other without a fat autograft) and were sacrificed at 6 weeks for chamber removal and assessment. Morphometric assessment of percent and absolute adipose tissue volume was undertaken.

Cymetra™ constructs yielded the largest total chamber volume and absolute adipose tissue volume. Puramatrix™ constructs were small but yielded only slightly less adipose tissue than in Matrigel™ constructs. Myogel™ constructs did not produce an appreciable or consistent amount of fat. It was found that the TGF regime significantly increased percent and absolute adipose tissue volume (p<0.0005). Further, interaction analysis surprisingly found a lack of interaction between the matrix and TGF.

MATERIALS AND METHODS

Animals, Anaesthetic and Chambers

Adult male severe combined immunodeficient (SCID) mice weighing approximately 20- 25g each were used. Prior to all surgical procedures, mice were anaesthetised with chloral hydrate administered intraperitoneally at 0.4mg/g body weight, and kept under sedation by provision of a continuous supply of isofluorothane gas during surgery. Initial surgical procedures to insert chambers into mice for this study were performed in an isolated procedure room under aseptic conditions. Hair on the abdominal wall, groin and upper legs of the mice were removed using a depilatory cream and the skin decontaminated with chlorhexidine and alcohol. Tissue harvests were performed under aseptic conditions.

The cylindrical and non-collapsible silicone chambers used in these experiments were constructed from 5 mm long sections of standard laboratory tubing (Dow-Corning Corporation, Midland, MI, USA) with an internal diameter of 3.35mm (individual chamber volume 40-50μl).

Preparation of Extracellular Matrices (ECMS)

Growth factor reduced (GFR) Matrigel™ was purchased from BD Biosciences (Bedford, MA₃ USA) and stored at -20°C until a day before surgery, at which time it was transferred into 4°C storage. GFR Matrigel™ was kept on ice prior to its use to prevent significant gelation, and allowed to set for 1-2 minutes within the chamber at room temperature before chamber closure.

Mouse Myogel™, a hydrogel ECM extracted from mouse skeletal muscle, was stored at 4⁰C until the day of surgery, at which time it was kept on ice to prevent significant gelation.

Cymetra^® Micronized Alloderm^® Tissue was purchased from LifeCELL (Branchburg, NJ, USA) and stored at 4°C. On the day of mouse surgery, the packaged Cymetra^® powder (0.3ml) was rehydrated with a diluent (1.7ml normal saline) as per manufacturer's instructions. In brief, the rehydration process involved the connection of two syringes, one with the supplied powder and the other with the diluent containing normal saline and any growth factors. The Cymetra^® powder was first wetted with the diluent, and then thoroughly mixed, resulting in a thick paste ready for use.

BDTM Puramatrix™ Peptide Hydrogel was purchased from BD Biosciences (Bedford, MA, USA) and stored at 4°C. 500 μl aliquots of Puramatrix™ were covered with 1 ml of phosphate buffered saline (PBS) (pH 7.4) each, as per manufacturer protocols, for about 52 hours, with one change at 16 hours, to obtain a pH of about 7.2. The supplied Puramatrix™ was determined to have a pH of about 2.0-3.0 using Acilit^® pH strips (Merck, Darmstadt, Germany) and the final pH confirmed at about 7.0 using the same method. As a result of neutralisation, Puramatrix™ set into a soft gel (Puramatrix™ gels at pH > 4.5-5.0), which consequently fragmented when taken up into a syringe.

Triple Growth Factor (TGF) Combination

The TGF combination consisted of recombinant human Basic Fibroblast Growth Factor (rh bFGF) (CytoLab Ltd, Rehovot, Israel), recombinant mouse Vascular Endothelial Growth Factor-A₁₂o (rm VEGF-A₁₂₀) (R&D Systems, Minneapolis, MN, USA) and recombinant rat Platelet-Derived Growth Factor-BB (rr PDGF-BB) (R&D Systems, Minneapolis, MN, USA).

These growth factors were initially reconstituted to a concentration of 50μg/ml according to manufacturer specifications, and stored at -80°C. On the day of mouse surgery, the three growth factors were thawed and mixed together and with heparin (Pharmacia & Upjohn, Kalamazoo, Michigan, USA) (to sequester the growth factors in the construct). This mixture was then added to the matrix being tested to achieve a final concentration of 100ng/ml of each growth factor and 80U/ml of heparin, and the solution kept on ice to preserve growth factor function.

Mouse Flow-Through Chamber Model

Hollow silicon chambers were inserted bilaterally in each mouse. In brief, a transverse groin incision was made into the medial aspect of the upper leg of the mouse, exposing the inguinal fat pad. The superficial epigastric vessels were carefully dissected from their origin at the femoral vessels, along their length up to their insertion into the inguinal fat pad, which was also mobilised in the dissecting process. The cylindrical silicon tube used to form the chamber was cut from sections of standard laboratory tubing (Dow-Corning Corporation, Midland, MI, USA), split down the side and slipped around the superficial epigastric vessels (Figure 7). The chamber was then secured to the underlying muscle with 10-0 nylon sutures. The proximal femoral end, as well as the lateral split of the chamber, was then sealed with melted bone wax (Ethicon, Somerville, NJ, USA), taking care to maintain the integrity of the epigastric artery, vein and nerve where it entered the chamber.

In one chamber in each mouse, a 5 mg fat autograft (dissected from the inguinal fat pad) was placed in the chamber alongside the superficial epigastric vessels, and the rest of the chamber filled with the matrix being tested. The other chamber in the mouse was filled with matrix without a fat autograft. The distal end of the construct was then sealed with melted bone wax, taking care not to damage the epigastric vessels where they left the chamber. Finally, the chamber was placed in the dissected plane in the groin, lateral to the femoral vessels, and the wound sealed with metal clips.

Four extracellular matrices (Matrigel , Myogel , Cymetra and Puramatrix ) were assessed with and without a TGF combination, resulting in a total of 8 experimental groups. Each group included 6 SCID mice each with bilateral chambers, one chamber containing a 5 mg fat autograft from the inguinal fat pad in addition to the ECM. The incubation time for all groups was 42 days, and the experimental groups are summarised in Table 2.

Chamber Removal, Tissue Processing and Sectioning

All animals were sacrificed 42 days after chamber implantation and the construct extracted surgically. Vessel patency within the chamber was assessed by direct inspection prior to chamber removal and by the presence or absence of bleeding from the cut ends of the pedicle as the chamber was removed. Each construct was weighed and its volume determined by the amount (mg) of PBS displaced by the tissue. They were then immediately fixed overnight in 4% paraformaldehyde (PFA) solution at 4°C. Fixed tissues were stored in PBS at 4⁰C until processing.

Tissues to be embedded in paraffin were processed for 6 hours using an automated tissue processor (Shandon Hypercentre XP, Shandon Scientific Ltd., Cheshire, England) according to methods well known in the art. Especially delicate samples were set in small blocks of agarose prior to processing.

Serial histological sections (5 μm thick) were cut from each sample using a microtome (Biocut 1130, Reichert-Jung, Wein, Austria) with disposable blades (Reichert-Jung, Wein, Austria) and placed on AAS (aninopropyltriethoxylsilane) coated glass slides and incubated overnight at 37°C.

Histological staining Prior to staining, sections were dewaxed and hydrated in accordance with methods well known in the art. Serial sections were routinely stained with haematoxylin and eosin, toluidine blue or Masson's trichrome stains and coverslipped using DPX mounting medium.

Morphometric analysis

The percent adipose tissue volume (%ATV) were determined on haematoxylin and eosin stained slides using the Computer Assisted Stereological Toolbox (CAST) system (Olympus Denmark, Alberstslund, Denmark). For each chamber, two sections 500 μm apart were counted with a 2Ox objective. Using systematic random sampling, 12-point grids were superimposed on randomly selected fields representing a percentage of the total chamber cross-sectional area so that a minimum of 200 points would be counted for each chamber (for most chambers 400-600 points were counted).

To count a minimum of 200 points in a chamber, a percentage of 25% was initially selected. However, the small size of Puramatrix and Myogel samples meant that this percentage was not sufficient to produce 200 points, and thus a percentage of 100% was chosen for counting these. Prior to data collection, verification counts were performed on a number of samples which determined that the %ATV was not significantly different whether a 25% or 100% of a sample was counted.

Tissue Percentages and Volumes

The structure on which each point fell was recorded into one of five groups: a) 'normal' fat with a relatively healthy appearance (crisp and complete cell membrane), b) 'abnormal' or dying fat (lacking nuclei or featuring disrupted cell membranes), c) fat cysts d) connective tissue elements and blood vessels. Fat cysts were defined as large abnormal adipocytes representing a degenerative process, or slightly smaller structures surrounded by a cell infiltrate and cuboidal cells (preadipocytes) thought to pre-empt adipogenesis.

The percentage of points that fell on 'normal' fat was considered to be the percent adipose tissue volume (%ATV).

The absolute adipose tissue volume of each chamber was determined by multiplying the total displacement volume (in mg of PBS displaced) of each construct, measured at chamber harvest, by %ATV . These volumes were then converted into microlitres (μls) by multiplying by a correction factor of 1.002, using the density of PBS at 25°C (1.002 g/ml).

Statistical Analysis

A three way analysis of variance (ANOVA) was performed on all data fields using Statistical Package for the Social Sciences (SSPS) Version 12.0 (SPSS Incorporated, Chicago, USA). Total tissue volume, %ATV, and absolute adipose tissue volume were analysed. Further, the three way ANOVA was used to look for any interactions between the matrix, TGF and autograft in each of the analysed fields.

RESULTS

Total Tissue Volume Statistical analysis of total tissue volume by three way ANOVA revealed a significant difference between the matrices (p<0.0005) (Figure 8). Cymetra™ samples consistently showed the largest tissue volume, although morphological analysis suggests much of this was relatively unchanged Cymetra™ that had simply retained its volume over the 6 weeks. Myogel™ and Puramatrix samples regularly had quite small volumes.

The triple growth factor (TGF) combination was also found to significantly increase total tissue volume (p<0.001). Presence of a fat autograft had no measurable effect on the total volume of the tissue. No interactions were seen between the matrix, TGF and autograft in analysis of the total tissue volume.

Percent Adipose Tissue Volume (%A TV)

Statistical analysis revealed significant differences between the matrices (p<0.0005) as well as significant increases in the percentage of normal adipose tissue (excluding fat cysts and abnormal fat) counted due to the TGF combination (p<0.0005) and presence of a fat autograft (p<0.0005) (Figure 9). No interaction was seen between the matrix and TGF in analysis of the % ATV.

Absolute Adipose Tissue Volume

Statistical analysis revealed significant differences between the matrices (p<0.0005) and a significant increase in absolute adipose volume due to the TGF combination (p<0.0005) (Figure 10). The presence of a fat autograft has a small but measurable effect on the normal adipose tissue volume at 6 weeks (p=0.011). Myogel™ groups had the least absolute fat volume, whilst the other matrices were relatively comparable. No interaction was seen between the matrix and TGF in analysis of the absolute normal adipose tissue volume.

Those skilled in the art will appreciate that the invention described herein is susceptible to variations and modifications other than those specifically described. It is to be understood that the invention includes all such variations and modifications. The invention also includes all of the steps, features, compositions and compounds referred to or indicated in this specification, individually or collectively, and any and all combinations of any two or more of said steps or features. BIBLIOGRAPHY

Antoniades and Hunkapiller, Science 220(4600):963-965, 1983

Beech et al, JMicrosc 197(l):26-45, 2000

Betsholtz et al, Nature 320(6064):695-699, 1986

Bosnia et al, Nature 201 (5900):527 -530, 1983

Bouloumie et al, Ann Endocrinol (Paris) 63(2 Pt l):9l-95, 2002

Bryan and Hartl, Science 240(4849):2\5-2l7, 1988

Cao et al, Biomaterials 27(14):2854-2S64, 2006

Carraway and Mellow, Ann Plast Surg 24(3):293-296, 1990

Chajchir, Aesthetic Plast Surg 20(4):29\-296, 1996

Chajchir and Benzaquen, Plast Reconstr Surg 84(6):92\-934, 1989

Claesson-Welsh et al, J Biol Chem 264(3):\142-1141, 1989

Clarke et al, Nature 308 (5958) -.464-467, 1984

Coleman, Aesthetic Plast Surg 19(5):42l-425, 1995

Coleman, Dermatol Clin /7^:891-898, 1999

Crandall et al, Microcirculation 42(2):2l 1-232, 1997 Crandall et al, JClin Endocrinol Metab 85 (7):2609-2614, 2000

Cronin et al. Plast Recomtr Surg 113(l):260-9, 2004

Devare et al, Cell 36(l):43-49, 1984

Doolittle et al, Science 185(148):36S-370, 1974

Eppley and Sadvoce, Aesthetic Plast Surg 15(3):215-21S, 1991

Eppley et al, J Oral Maxillofac Surg 60(5):477-4S2, 1992

Eppley et al, Plast Reconstr Surg 90(6): 1022- 1030, 1992

Esposito et al, Biomaterials 17 (2) :2009-2020, 1996

Faham et al, Science 271(5252):U6-1120, 1996

Farrington-Rock et al, Circulation 110(15):2226-2232, 2004

Fukumura et al, Circ Res 93(9):S8-97, 2003

Graves et al, Biochem Biophys Res Commun 129(l):56-62, 1985

Gregoire et al, Physiol Rev 7Sf5J:783-809, 1998

Hart et al, Science 240(4858):\529-\53\, 1988

Hausman and Thomas, J Morphol 190(3)\27\-2?>3, 1986 Heldin et al, Embo J 7f5;:800-804, 1988 ^" Heldin et al, Physiol Rev 79^:1283-1316, 1999 ^• Hoppe et al, Biochemistry 28(7):2956-2960, 1989 Horl et al, Ann Plast Surg 2<5f5j:248-258, 1991 Illouz, Aesthetic Plast Surg 12(3):175-S\, 1988 Kanchwala and Bucky, Facial Plast Surg 19(1):137-146, 2003 Kawaguchi et al, Proc Natl Acad Sci USA 95(3):l062-\066, 1998 King et al, Proc Natl Acad Sci USA 82(16):5295-5259, 1985 Langer and Vacanti, Science 260(5110): 920-926, 1993 Linette et al, Science 241(4865):573-576, 1988 McConkey et al, Science 242(4876):256-259, 1988 Malhotra et al, Science 242(4879):755-759, 1988 Matsui et al, Science 243(4892):$00-804, 1989 Mikus et al, Laryngoscope 105 (1):17 -22, 1995 Moscona et al, Ann Plast Surg 53(5J:500-506, 1994 Neels et al, Faseb J 18(9):9%3-9%5, 2004 Neuber, Chir Kongr VerhandlDsch Gesellschll, 1893

Owen et al, Science 225(4657):54-46, 1984

Pantazis et al, Proc Natl Acad Sci USA 82 (8) -.2404-2408, 1985

Pearson et al, Science 241(4868):910-913, 1988

Planat-Benard et al, Circulation 109 (5) -.656-663, 2004

Rao et al, Cold Spring Harb Symp Quant Biol 51(2) -.959-966, 1986

Rees et al, Embo J 3 (8):1843-1841 , 1984

Ren et al, Genes Dev 16(l):27-32, 2002

Robbins et al, Nature 205 (5935) -.605-608, 1983

Rosen, Prostaglandins Leukot Essent Fatty Acids 73:31-34, 2005.

Rupnick et al, Proc Natl Acad Sci USA 99(16):10730-10735, 2002

Shoshani et al, Plast Reconstr Surg 115(3):853-859, 2005

Shoshani et al, J Drugs Dermatol 4(3):311-316, 2005

Sierra-Honigmann et al, Science 281(5383):1683-1686, 1998

Silverman et al, Biochem Biophys Res Commun 153(1): 347-352, 1988 Smith et al, Plast Reconstr Surg 117(6):l 836-1844, 2006 Thomas et al, Science 242(4881):1O5O-IO53, 1988 Thomas and Penington, International Journal of Adipose Tissue, 2005 • U.S. Pat. No. 4776073

Waterfield et al, Nature 304(5921):35-39, 1983 Weibel, Practical Methods for Biological Morphometry (1), 1979 Weinert and Hartwell, Science 241(4863):317-322, 1988 White and Hartzell, Science 239(4841 Pt l):77S-7S0, 1988 Wilks et al, Science 242(4885):l54l-1544, 1988 Wissemann and Jacobson, In Vitro Cell Dev Biol 21 (7):39l-401, 1985 Yarden et al, Nature 323(6085):226-232, 1986 Yuksel et al; Plast Reconstr Surg 105(5):1712-l720, 2000

Claims

CLAIMS:

1. A method for facilitating tissue engineering in a mammalian subject said method comprising administering to a site of a free fat graft an adipocyte growth effective amount of an agent selected from PDGF, VEGF and FGF such that the agent is provided locally or via sustained delivery or an adipocyte growth effective amount of two or more agents selected from PDGF, VEGF and FGF said administration being for a time and under conditions effective to promote weight maintenance and to preserve normal cellular architecture of free fat grafts and/or to induce adipogenesis and/or promote fat graft survival in said mammalian subject.

2. The method of Claim 1 wherein the PDGF is selected from PDGF-AA, -BB, -AB, - CC and -DD or a splice variant, homolog or isomer thereof.

3. The method of Claim 1 wherein the VEGF is selected from VEGF-A-120, -121, - 165, -189, -206, VEGF-B, VEGF-C, VEGF-D and VEGF-E or a splice variant, homolog or isomer thereof.

4. The method of Claim 1 wherein the FGF is FGF-b or a splice variant or homolog thereof.

5. The method of Claim 1 wherein the agent is PDGF.

6. The method of any one of Claims 1 to 5 wherein the mammalian subject is a human.

7. The method of Claim 6 to facilitate growth of a free fat graft or fat graft survival.

8. The method of Claim 6 to facilitate growth of tissue in or on a seeded support.

9. The method of any one of Claims 1 to 8 in the treatment of aging face, facial hemiatrophy or lipodystrophy.

10. A method for promoting growth of adipocytes or its precursors or mesenchymal stem cells on or in a graft support implanted in a mammalian subject, said graft support comprising an agent selected from PDGF, VEGF and FGF in a sustained delivery form or a combination of two or more agents selected from PDGF, VEGF and FGF in amounts effective to promote adipogenesis in the graft support as well as promoting proliferation of adipocytes or its use.

11. The method of Claim 10 wherein the PDGF is selected from PDGF-AA, -BB, -AB, -CC and -DD or a splice variant, homolog or isomer thereof.

12. The method of Claim 10 wherein the VEGF is selected from VEGF-A- 120, -121, - 165, -189, -206, VEGF-B, VEGF-C, VEGF-D and VEGF-E or a splice variant, homolog or isomer thereof.

13. The method of Claim 10 wherein the FGF is FGF -b or a splice variant or homolog thereof.

14. The method of Claim 10 wherein the agent is PDGF.

15. The method of any one of Claims 10 to 14 wherein the mammalian subject is a human.

16. The method of Claim 15 to facilitate growth of a free fat graft or fat graft survival.

17. The method of Claim 15 to facilitate growth of tissue in or on a seeded support.

18. The method of any one of Claims 10 to 17 in the treatment of aging face, facial hemiatrophy or lipodystrophy.

19. A method for grafting fat tissue in a mammalian subject said method introducing adipocytes to a site in the mammalian subject together with or prior to the administration of an agent selected from PDGF, VEGF and FGF in a form for local or sustained delivery or two or more agents selected from PDGF, VEGF and/or FGF said administration being for a time and effective for weight maintenance and to preserve normal cellular architecture of said fat graft in said mammalian subject.

20. The method of Claim 19 wherein the PDGF is selected from PDGF-AA, -BB, -AB, -CC and -DD or a splice variant, homolog or isomer thereof.

21. The method of Claim 19 wherein the VEGF is selected from VEGF-A-120, -121, - 165, -189, -206, VEGF-B, VEGF-C, VEGF-D and VEGF-E or a splice variant, homolog or isomer thereof.

22. The method of Claim 19 wherein the FGF is FGF-b or a splice variant or homolog thereof.

23. The method of Claim 19 wherein the agent is PDGF.

24. The method of any one of Claims 19 to 23 wherein the mammalian subject is a human.

25. The method of Claim 24 to facilitate growth of a free fat graft or fat graft survival.

26. The method of Claim 24 to facilitate growth of tissue in or on a seeded support.

27. The method of any one of Claims 19 to 26 in the treatment of aging face, facial hemiatrophy or lipodystrophy.

28. Use of PDGF, VEGF and/or FGF in a sustained delivery form or two or more of PDGF, VEGF and/or FGF in the manufacture of a medicament for the promotion of adipogenesis of a fat graft or seeded graft support in a mammalian subject.

29. A sustained delivery composition comprising one or more of PDGF, VEGF and/or FGF and a matrix which facilitates sustained delivery.