- Email: [email protected]
New research in breast reconstruction: Adipose tissue engineering
NEW DIRECTIONS IN PLASTIC SURGERY, PART II
0094–1298/02 $15.00 .00
NEW RESEARCH IN BREAST RECONSTRUCTION Adipose Tissue Engineering Saleh M. Shenaq, MD, and Eser Yuksel, MD
In the human species, a major function of the breast is aesthetic. The soft-tissue volume within the breast displaces the overlying skin to create the protuberant contour of the female thorax, which is associated solidly with, and to some extent, definitive of, femininity in modern culture. Adipose tissue is the major contributor to the volume of the breast. Despite the high emotional, physical, and aesthetic cost, partial or complete mastectomy, with or without radiation therapy, remains the cornerstone of the modern management for breast cancer. 1 This surgical procedure removes the volume element from the breast obliterating the feminine contour. Currently, techniques for postmastectomy reconstruction include the transverse rectus abdominis myocutaneous (TRAM) flap, latissimus dorsi flap with or without alloplastic implants, and implants with or without skin expansion.2 Although effective, these meth-
ods frequently are limited by high donor-site morbidity and difficulty in restoring the specific volume, shape, size, texture, and sensation characteristics of a normal breast. Use of implants is avoided in prospective radiation treatment cases because of possible implant extrusion and wound healing problems. Additionally, implant application has a risk for developing capsule contracture, which may require further surgical interventions to correct the deformity and to eliminate the pain. Based on these facts, autologous breast reconstruction remains the primary choice in most cases. A major donor-site morbidity of the TRAM flap is associated with the weakness of the anterior abdominal wall musculofascial system, however, and is well recognized by patients and physicians and has led plastic surgeons to search for less morbid operations, such as perforator-based muscle-sparing free TRAM flaps.3, 4 The mor-
Financial Support Parts of this work were funded by: A developmental grant from NIH Breast Spore Grant A PSEF Research Grant An ASERF Research Grant Samuel Spira Research Fund
From the Division of Plastic Surgery, De Bakey Department of Surgery, Baylor College of Medicine, Houston, Texas
CLINICS IN PLASTIC SURGERY VOLUME 29 • NUMBER 1 • JANUARY 2002
111
112
SHENAQ & YUKSEL
bidity of latissimus dorsi musculocutaneous pedicled flaps includes unsightly donor-site scars and the alteration of the shoulder functional integrity. This morbidity is in addition to the inability to avoid use of implants in most of the cases. An ideal solution to the problem of the absent breast after mastectomy would be a method that restores or adds soft-tissue volume to the breast without donor-site morbidity or the risks associated with allogenic implants. Considering the trend to perform skin-sparing mastectomy modalities without altering the principles and defined frame of the ideal oncologic treatment for breast cancer, missing volume component at the mastectomy site can be considered the main denominator of the reconstructive surgical design. Adipose tissue is the most appropriate tissue to be used for this role because not only does it remain as the main fraction of the removed volume but it also is the most compatible tissue for missing glandular component. TISSUE ENGINEERING: GENERAL PRINCIPLES Tissue engineering is the science of generating tissue by using the principles of molecular biology and material engineering. The main elements to be optimized in tissue engineering are the cell and extracellular matrix and cross-talk between these elements. Other than cross-talk of cell and matrix, the generation of the vascular network of the tissue is an indispensable component of this dynamic process. Interestingly, the vascular phase itself is dependent on the complex management of cell and extracellular matrix interaction within the vascular compartment by the organized and adequately reinforced release of angiogenic signals released from the generating tissue. The main principles of tissue engineering research are discussed in a general descriptive level in the following section. The Cell The Cell Source The cell is the center of all events at the molecular level and is a prerequisite player
in the regeneration and maintenance of the tissue. Any organ in the body consists of more than one type of cell; however, a specific cell type exists in all tissue that structurally and functionally dominates the medium. To simplify the engineering process, a single-cell type usually is chosen to be manipulated to control the matrix and generative events. The cell used for this purpose can be autogenous or allogenous in origin. Allogenic cells do not possess a long-term or permanent cycle in living systems, unless an immune modulation of the donor or recipient is executed or embryonic stem cells are used. Despite its high differentiative and proliferative potential for diverse types of tissue, the use of embryonic stem cells currently is being evaluated because of the associated major ethical and legal issues that are brought by the opponents. Therefore, the conventional direction is toward the use autologous cells. The next question is the use of committed and end-stage differentiated cells or pluripotential noncommitted cells. Use of pluripotential mesenchymal stem cells or noncommitted precursor cells is favored because of their high capacity for prospective differentiation and multiplication. The third level dilemma in the cell source is whether to use the cells in vitro or to stimulate them in vivo. In vitro use is limited to low volume-surface ratio tissue types (such as skin) or to expand research information for isolated manipulations in controlled parameter systems. In vivo applications are discussed in the research background section. The Manipulation of the Cell Regardless of the cell source chosen, the cell has to be informed, stimulated, and maintained for accelerated and directed behavioral pattern because it is silent or reluctant without the right stimulus vector. This goal can be achieved by extracellular matrix modification such as long-term delivery of bioactive factors or the configurational effect of extracellular matrix structural elements. This stimulation is converted to intracellular messenger events, which in turn alter the gene transcription profile or post-transcriptional modifications for a period as long as the extracellular vector ex-
NEW RESEARCH IN BREAST RECONSTRUCTION
113
ists. The other method is presenting the selfcopying gene vectors to the cell for the synthesis of a particular protein to trigger the cascade of events for targeted behavior. Transcription factors also can be used in a selective fashion for targeted behavioral manipulation.
check systems for adequacy of the growth or by the limiting effects of extracellular matrix composition, which also includes the vascular support, and tissue barriers for repelling signals to counteract the forwarding vector. This task requires use of different matrices in specialized prefabrication models.
Tissue Transformers and Stimulators
Maintenance Phase
Differentiative Phase The noncommitted precursor cells can be stimulated to differentiate to a specialized cell type, which declares the main constitution for the targeted tissue and acts as the primary cell structurally and functionally. There are specific transcription factors serving for different tissue transformation processes. The adipose tissue transformers are discussed in the following sections. Targeting differentiated cells for transformation involves complex signaling pathways, which enables dedifferentiation and differentiation processes consecutively. Proliferative and Migratory Phase To fill a targeted volume for regeneration, precursor and differentiated cells should be under the influence of the right vector in magnitude and in direction. This influence is achieved by the organization of cell to extracellular matrix protein cross-talk, and cell-tocell cross-talk, which in turn dictates the cell dynamics. The paracrine form of this interaction has an exclusive and mandatory task of stimulating angiogenesis. This task can be reinforced independently in the design of the tissue regeneration model by delivery of angiogenic factors, genes, or stimulatory vectors. The synchronized orchestration of tissue migration and angiogenesis in a three-dimensional pattern is well performed in in vivo prefabrication systems, unlike the inadequacy of this partnership in in vitro systems. Maturation and Control of GrowthFruition Phase The tissue growth may be controlled by manipulating the intracellular diagnostic and
The signal for the cell should be stable in the long term to maintain the trophism at the plateau level. Fortunately, in natural regeneration models, the cell is capable of modifying the extracellular matrix to provide the external support for its existence. This cellular decision is highly affected by external vectors, however, which alters the oxidation-reduction equilibrium.
Extracellular Matrix Optimization Engineering of material can be evaluated in two disciplines: one is three-dimensional design and prefabrication of scaffold systems, and the other area is based on optimizing components of the biomaterial, which involves cross-linking and stabilization of different matrix elements. Macrodesign The computer-aided engineering modalities currently are developed for different tissue and organ templates. Porosity, or organization of matrix, acts as a major denominator for tissue migration. This task can be achieved by different processing techniques.5–8 Micropatterning The cell’s behavior is affected by the extracellular matrix pattern in nanometer scale. The progress in nanotechnology will enable manufacturing the nano-patterns for different tissue models. Currently, it is possible to manufacture micropatterns at 0.20 micrometer scale, and in the year 2005, it is predicted that this will be achieved at 0.10 micrometer scale.
114
SHENAQ & YUKSEL
Optimization of Signaling Properties of Extracellular Matrix Cell-to-extracellular matrix cross-talk can be directed with the sustained release of bioactive proteins.9, 10 The release system might involve use of biodegradable microspheres or chemical stabilization of factors in extracellular matrix back bone. Additionally, the threedimensional configuration of structural matrix proteins and their composition are effective in altering the cellular behavior. The use of strong signaling molecules should be fashioned within the principles of gradient difference to provide the maintenance of the migration vector toward the targeted direction in adequate magnitude.
In Vitro/In Vivo Prefabrication Models In the authors’ laboratory, the concept of in vivo tissue engineering for high volume surface ratio regeneration models is emphasized. This concept describes the generation of tissue by processes occurring completely inside the living body. In vivo tissue engineering, like traditional in vitro models of tissue engineering, involves the use of cells to generate tissues. Unlike traditional tissue engineering, however, there is no ex vivo cell seeding and therefore no need for cell harvesting and culturing procedures. The cells that are used to generate the desired tissues are mesenchymal precursor cells already present in the region requiring reconstruction. Therefore, the optimum guidance of these cells with the stimulant vectors within the principles discussed above can lead to selfsupporting tissue growth in terms of vascular component. The in vivo models can be developed as orthotopic or heterotopic prefabrication models.
tems in the presence of DNA damage. Furthermore, local regeneration designs should be exclusive for systemic effects and should not alter the treatment of primary disease if any exists. Another challenge in translation of tissue engineering modalities for human trials is the difference in behavioral patterns of organisms. Particularly, the template size in animal models is smaller in dimensions whereas the cell size is approximately equal when compared with human models. DIFFERENTIATION OF ADIPOSE TISSUE Adipose tissue is formed by terminally differentiated adipocytes and their committed precursor cells called preadipocytes. The earlier members of adipocyte lineage include noncommitted stem cells and adipoblasts, which are speculated to exist in various compartments. Multiple steps of adipose cell differentiation involve fatty acid activated receptor, peroxisome proliferator–activated receptor ␥, insulin-like growth factor-1 (IGF-1), insulin, retinoids, triiodothyronine, and prostaglandins (I2, D, and J series).11 The differentiation from pluripotential fibroblastic cell to the next step is not well identified, however.12 The differentiation cascade may be initiated by several growth factors. These factors include basic fibroblastic growth factor, IGF-1, and tumor necrosis factor ␣ (TNF ␣).12–14 TNF ␣ also is demonstrated as an inhibitor of adipogenesis. 15 Transforming growth factor  (TGF ) and epidermal growth factor also have been shown to be inhibitory within this process.16, 17 Adipocytic differentiation of preadipocytes is enhanced when Preadipocyte Factor 1 (an epidermal growth factor–like domain containing transmembrane protein) is underexpressed (such as with dexamethasone administration).18 Glucocorticoids and insulin can increase the differentiation rate by 30- to 70- fold.19
Safety Issues Manipulating the cell cycle toward the accelerated proliferation rate opposes the primary goal on which the cancer research is focused. Thus, performing this task should exclude the alteration of cell cycle check sys-
Insulin, Insulin-like Growth Factor-1, and Intracellular Seconder Messanger Pathways Insulin and GF-1 have been demonstrated to promote adipocytic differentiation in sev-
NEW RESEARCH IN BREAST RECONSTRUCTION
eral steps.11, 20–22 Adipogenic action of the insulin starts with tyrosine kinase receptors and proceeds with the involvement of the Ras pathway,23 while glucose transport regulation is performed through PI 3-kinase pathway, which is the other option for insulin after insulin receptor substrate (IRS) binding proteins are activated.4a This is particularly important to developing a hypothesis around the postreceptor defects in type 2 diabetes and to understanding the obesity, insulin, diabetes mellitus (DM) triangle.
Nuclear Receptors and Transcription Factors Peraxisome proliferator activated receptor (PPAR) ␥ receptors are involved primarily in terminal differentiation stages.12, 24, 25 Based on this fact, targeting the PPAR-induced transcription factors or activators of PPAR is the main interest of many research centers focused on obesity or adipogenesis.26–31
Role of Frizzled Related Protein (Frzbs) and Wnt in Adipocyte Differentiation Frzb is a newly discovered family of glycoproteins that modulate (antagonistic) signaling activity of Wnt.32 Wnts are secreted signaling proteins that are involved in the developmental processes. Inhibition of their signal leads to terminal differentiation of adipocytes.33 Thus frzbs might be new targets for reversing the suppression of adipocytic differentiation.
Stem Cell Differentiation When the focus of interest is to develop adipose tissue from other sources, as indicated in orthotopic in vivo models,34 the availability of earlier members of adipocytic lineage and their identification become more imperative. The stem cells from nonadipocytic sources have been identified and differentiated into adipocytic-committed members. 35–37 An interesting finding in these reports is the involvement of mitogen-acti-
115
vated kinases, which puts insulin in a central or dominant location in induction of stem cells. These findings support the observations in the authors’ research.34 In these cells, osteogenic differentiation is regulated by retinoic acid receptor ␣, which can be targeted for inhibition to provide the shift toward adipocytic differentiation.37
SOFT-TISSUE ENGINEERING EXPERIENCE The authors’ current research objectives are to develop a noninvasive breast reconstruction method for postmastectomy defects using in vivo tissue-engineering methods and to use a totally biodegradable system that enables the sustained release of adipogenic factors that in turn stimulate the adipogenic differentiation of recipient cells in the pectoral fascia. When the research objectives are achieved, in clinical perspective the system will be incorporated with an endoscopic delivery technique to the pectoralis muscle and fascia without interfering with the operative site and remaining breast tissue, particularly for delayed and skin sparing mastectomy reconstructions. The authors’ laboratory has worked to develop a biodegradable system that stimulates the growth of adipose tissue to be used in an orthotopic in vivo tissue-engineering approach to breast reconstruction and augmentation mammaplasty. The experimental phases the authors have followed in their laboratory’s soft-tissue engineering research are listed in Table 1.
Adipogenic Factors and Long-term Delivery Method Adipocyte precursor cells (preadipocytes) contained in postnatal human tissues have the capability to proliferate and differentiate into mature lipid-containing adipocytes. In vitro experiments have demonstrated that insulin and IGF-1 can stimulate these preadipocyte activities.38 In vitro experiments demonstrate that insulin and IGF-1 stimulate adipocyte proliferation, differentiation, and
116
SHENAQ & YUKSEL
Table 1. EXPERIMENTAL PHASES OF IN VIVO SOFTTISSUE PREFABRICATION Adipogenic Factors Long-term Delivery Method: Biodegradable Microsphere Processing Incorporation with Growth Factors Controlled Release Kinetics In Vivo Application: Improvement of Fat Graft Survival Rates Augmentation of Adiposo-fascial Flaps Augmentation of Vascularity in Graft Prefabrication De Novo Fat Generation Biodegradable Scaffolds: Template Processing Degradation Kinetics Combination with microspheres Breast Prefabrication: Template Processing Flap Models In Vivo Regeneration
trophic activities. Additionally, lipohypertrophy is reported as a side effect of repeated subcutaneous insulin injections. Other researchers have demonstrated that a subpopulation of preadipocytes with the potential to differentiate into mature, lipid-containing adipocytes exists within postnatal adipose tissue. IGF-1 has a negative effect on intracellular lipid content, and insulin’s action is to increase the intracytoplasmic lipid content despite the fact that they act in the same direction through similar tyrosine kinase receptors to stimulate the differentiation and proliferation of the preadipocytes. Numerous biodegradable drug delivery systems are in various stages of development. These systems have been employed experimentally in a variety of tissues39 and a variety of species40 to deliver a variety of pharmacologic compounds (cytotoxic agents, steroids, peptide/protein hormones, or cytokines). Various degrees of clinically relevant results have been obtained.41, 42 The theoretic advantages of biodegradable drug delivery systems are the ability to deliver an active agent for a long duration with the effects potentially limited to a local region without multiple surgical procedures and without the implantation of a permanent foreign body.43 Poly(lactic–co–glycolic–acid)–polyethylene glycol (PLGA–PEG) microspheres are a biodegradable drug delivery system that currently is undergoing experimentation and further development (Cleek). 9 PLGA-PEG micro-
spheres are capable of delivering proteins and cDNA for long durations,9 and molded PLGA scaffolds serve well for guided tissue regeneration.5 Microsphere Preparation Microspheres were prepared using the double solvent emulsion method.9, 34, 44 A Coulter multisizer (Coulter Electronics, Miami, FL) was used to confirm a mean particle size of 12 m. Scanning electron microscopy was employed to demonstrate normal microsphere morphology (Fig. 1). During the manufacture process, the ‘‘drug’’ to be delivered is added as an aqueous solution to the polymer solution. This biphasic mixture is mechanically homogenized so that small aqueous droplets are suspended in the polymer solution. The polymer is precipitated into a solid phase trapping the aqueous droplets. The water then is sublimated away leaving the ‘‘drug’’ behind in numerous hollow cavities. The inclusion of PEG in the polymer solution leads to the formation of multiple microchannels. The result is closed porous microspheres with a network of microchannels with the ‘‘drug’’ trapped in the pores. As the microspheres undergo slow biodegradation, the eroding surface meets the cavities, and the growth factors are released. Incorporation with Growth Factors and In Vitro Release Kinetics An in vitro experiment was designed to confirm the microsphere’s ability to accomplish long-term peptide delivery. In this experiment, only insulin-containing microspheres were studied. Equal weights of microspheres were placed in four equal volumes of sterile and neutral phosphate buffered saline. The microsphere solutions were permitted to incubate at physiologic temperatures for four separate time periods each; 7, 14, 21, and 28 days. Measurements of optical density and PAGE-electrophoresis were employed to provide a quantitative and qualitative analysis (respectively) demonstrating increased insulin concentration with increased incubation/degradation time thus confirming release for extended duration.
NEW RESEARCH IN BREAST RECONSTRUCTION
117
Figure 1. Scanning electron microscopic images for PLGA/ PEG microspheres of 12–15 m average diameter (original magnification 1000).
In Vivo Applications Improvement of Fat Graft Survival Rates The first model the authors have used to test the efficacy of the adipogenic factors within a sustained release system was the fatgraft model. In this investigation, the effects of long-term, local delivery of insulin, IGF-1, and bFGF on fat graft survival have been evaluated using a poly (lactic-co-glycolicacid)—polyethylene glycol (PLGA-PEG) microsphere delivery system. Twelve-micrometer PLGA-PEG microspheres incorporated separately with insulin, IGF-1, and basic fibroblast growth factor (bFGF) were manufactured using a double-emulsion-solvent-extraction technique. Inguinal fat from Sprague Dawley rats was harvested, diced, washed, and mixed with 1) insulin microspheres, 2) IGF-1 microspheres, 3) basic fibroblast growth factor microspheres, 4) a combination of the insulin and IGF-1 microspheres, and 5) a combination of insulin, IGF-1, and bFGF microspheres. The treated fat grafts were implanted autologously into subdermal pockets in six
animals for each group. Animals receiving untreated fat grafts and fat grafts treated with blank microspheres constituted two external control groups (six animals per external control group). At 12 weeks all fat graft groups were compared on the basis of weight maintenance and a histomorphometric analysis of ‘‘adipocyte area percentage’’ (AAP); indices of volume retention and cell composition, respectively. Weight maintenance was defined as the final graft weight as a percent of the implanted graft weight. All growth factor treatments significantly increased fat graft weight maintenance objectively and volume maintenance grossly in comparison with the untreated and blank microsphere treated controls. Treatment with insulin and IGF-1, alone or in combination, was found to increase the AAP in comparison with fat grafts treated with bFGF alone or in combination with other growth factors. In conclusion, the findings of this study indicated that long-term, local delivery of growth factors with PLGA-PEG microspheres has the potential to increase fat graft survival rates. Furthermore, the type of growth factor delivered may influence the cellular/stromal composition of the grafted tissue.43
118
SHENAQ & YUKSEL
Augmentation of Adiposofascial Flaps The adiposofascial flaps currently described in the literature frequently lack the volume requirements for reconstructive goals. In this study, the authors examined the use of long-term local delivery of insulin and IGF1 together by PLGA-PEG microspheres to augment inguinal adiposofascial flaps based on inferior epigastric vessels in the rat. In this experiment, a group receiving only an operation and a group treated with blank (no growth factor) microspheres served as external controls for the surgical procedure and the drug delivery device, respectively. In all groups (n 5), the contralateral flap served as internal controls. Upon harvest on the twenty-eighth postoperative day, the insulin and IGF-1 treated flaps in both models weighed statistically more than the internal control flaps and the two external control flaps. Likewise, on gross inspection, the adipogenic growth factor–treated flaps had greater volumes than their internal control flap groups and both of the external control flap groups (operation only and blank microspheres). Other intergroup comparisons suggested the absence of a systemic insulin and IGF-1 effect on adiposity. A histomorphometric analysis suggested that 1) insulin and IGF1 treatment do not alter flap cell composition and 2) flap augmentation is secondary to the stimulation of cell proliferation and adipocytic differentiation rather than hypertrophy of mature adipocytes. Further evidence in favor of cell proliferation and differentiation was the finding of nonanatomic, ectopic fat islands on the treated flaps’ pedicle sheath and no variation in cell size distribution among groups. It was concluded that the long-term local delivery of insulin and IGF-1 with PLGA-PEG microspheres is an effective method for adiposofascial flap augmentation through the increase in mature adipocyte number rather than an increase in the cell size of pre-existing cells.44 Augmentation of Vascularity in a Fat Graft Prefabrication Model In this phase the authors have observed the effects of long-term delivery of bFGF to the
base (predicted pedicle of the prefabricated flap) area of fat grafts. The vascular density was measured using color distraction of microphotographs and Image Pro software. The results revealed that vascularity was increased significantly in the bFGF treated group (Fig. 2). De Novo Fat Generation This study was undertaken to characterize the duration of y and to evaluate the potential of the long-term delivery of insulin and IGF1 for the de novo generation of adipose tissue in vivo. Insulin and IGF-1–containing microspheres were administered directly to the deep muscular fascia of the rat abdominal wall. A group of animals treated with blank microspheres served as an external control. At the 4-week harvest period, multiple ectopic islands of adipose tissue were observed on the abdominal wall of the animals treated with insulin, IGF-1, and insulin plus IGF-1 microspheres (Fig. 3). Such islands were not seen in the blank microsphere group. Hematoxylin and eosin–stained sections of the growth factor groups demonstrated mature adipocytes interspersed with fibrous tissue superficial to the abdominal wall musculature and continuous with the fascia. Oil Red O– stained sections demonstrated that these cells contained lipid. Computer-aided image analysis of histologic sections confirmed that there were statistically significant increases in the amount of ‘‘ectopic’’ adipose neotissue developed on the abdominal wall of animals treated with growth factor microspheres. In conclusion, this study confirmed the longterm release of proteins from PLGA-PEG microspheres and demonstrated the potential of long-term, local insulin and IGF-1 to induce adipogenic differentiation to mature, lipidcontaining adipocytes from nonadipocyte cell pools in vivo.34 Scanning electron microscopy of the anterior abdominal fascia sections demonstrated the predominancy of extracellular elements and scarcity of cells in the blank microsphere group while the entire surface was covered by adipocytes in the insulin and IGF-1–treated group (Fig. 4). There were three possible explanations for the adipogenic phenomenon:
NEW RESEARCH IN BREAST RECONSTRUCTION
Figure 2. A, Prefabrication of fat-fascia flap using free fat grafts and angiogenic factors. B, Vascular density is measured using Image Pro software (Media Cybernetics, Del Mar, CA) for the images after green extraction and Sobel filtering has been applied. bFGF delivery (microsphere) group has higher vascular density values (statistically significant, P ⬍.01)
119
120
SHENAQ & YUKSEL
Figure 3. Representative gross image demonstrating the de novo neotissue as ectopic islands well integrated on the surface of the abdominal wall in an IGF-1 insulin treated animal.
Figure 4. Scanning electron microscopy of the anterior abdominal fascia. A, Blank microsphere group: predominancy of extracellular elements and scarcity of cells. B, Insulin and IGF-1–treated group: the entire surface is covered by adipocytes. (original magnification 1200.)
NEW RESEARCH IN BREAST RECONSTRUCTION
1. Mesenchymal stem cells (MSCs) might be present in the abdominal wall fascia and could have been stimulated to differentiate into preadipocytes and then differentiate into mature adipocytes. 2. A preadipocyte pool may exist within the abdominal wall fascia that might have been stimulated to differentiate into adipocytes. 3. A process of dedifferentiation might have occurred for a subset of the fibroblasts normally contained in the abdominal wall fascia. These cells then could have differentiated into preadipocytes followed by subsequent adipocytic differentiation.
Biodegradable Scaffold Engineering In this phase of the study, the authors have investigated the role of a rigid scaffold system for guided tissue regeneration in the previously described stimulatory system.
121
chymal cells in the recipient tissue bed. A breast-shaped implant was engineered from biodegradable PLGA foam (Fig. 5). This implant was designed as a porous scaffold to allow the infusion of microspheres and to serve as sources of support and guidance for new adipose tissue. The implant was used in conjunction with insulin and IGF-1 to create an in vivo tissue engineering system that generated shaped adipose tissue of a predetermined shape in an in vivo rat model (Fig. 6). The evaluation of the scaffolds at 5 weeks revealed migration of tissue towards the center of the template (Fig. 7). At 12 weeks, the entire scaffold was replaced by invading tissue, which was fibroelastic in gross examination (Fig. 8). The histologic examination suggested that adipose component dominated in the regenerated tissue (Fig. 8). The results demonstrated the potential power of in vivo tissue engineering for producing adipose tissue in situ for breast reconstruction and augmentation without flap transfer or the use of allogenic expanders or implants.
Template Processing PLGA 75:25 foams were produced by a particulate leaching technique using 300 to 500 m salt (NaCl) particles as a porogen.6 Salt particles were dispersed in a PLGA-methylene chloride solution yielding a 90% fraction of salt. Hemisphere-shaped constructs (2.5 cm in diameter) were placed in 70% ethyl alcohol for 12 hours to induce polymer cross-linking. The salt was leached from the PLGA constructs with distilled deionized water, creating a vast array of pores to convert the material to a foam.
Scaffold Implantation Breast Prefabrication—In Vivo Regeneration The authors’ laboratory has developed an in vivo approach to breast reconstruction and augmentation using a biodegradable implant in an animal model. This implant system is designed to stimulate its own replacement by new adipose tissue generated by the mesen-
Current and Future Phases In past experiments, the authors have used biodegradable PLGA breast-shaped scaffolds and biodegradable PLGA-PEG microspheres delivery of adipogenic (insulin and IGF) growth factors to achieve guided adipose tissue generation in vivo. Although this was the first report of adipose tissue generation in vivo, scaffold replacement fell short by 30%. Currently, the authors aim at testing the hypothesis that the introduction of the extracellular matrix molecule hyaluronan to the scaffold system will enhance the tissue generation rates and allow total tissue replacement of the breast-shaped biodegradable scaffold. Another work is focused on employing gel matrices within a silicon spacer induced capsule model (Fig. 9). This model not only enables the objective measurement of the regenerating tissue within the capsule but also serves as a model for pre-existing silicon implant carriers. The future phases of the authors’ research are directed towards the following aims:
122
SHENAQ & YUKSEL
Figure 5. Manufacturing 75:25 poly-lacto glycolic acid (PLGA) scaffolds. Process: particulate leaching method using NaCl as a porogen. Vulcanization into hemispheric form with a force of 250 lb at 160C. Porosity: 90%, pore size: 300–500 micron. Scanning electron microscopy was applied after freeze fracturing of the scaffolds (original magnification 300).
Figure 6. PLGA/PEG microspheres of 12–15 m average diameter are dispersed within the hemispheric scaffold (75:25 PLGA), porosity: 90%, pore size: 300–500 micron using specialized closed injection system with pluronic acid gel. Scanning electron microscopy was applied after freeze fracturing of the scaffolds (original magnification 600).
NEW RESEARCH IN BREAST RECONSTRUCTION
Figure 7. Harvest of 75:25 PLGA scaffolds, implanted beneath the muscle. At 5 weeks, tissue growth to the center can be observed.
Figure 8. Harvest of scaffold at 12 weeks. A, The entire scaffold is replaced by soft tissue. B, Hematoxylin-eosin stained cross-section shows that adipocytes predominate within the regenerated tissue.
123
124
SHENAQ & YUKSEL
Figure 9. Use of silicon spacers for capsule prefabrication. Gel matrices are applied within this isolated compartment model.
1. Incorporation of biodegradable extracellular matrix gel system (in vivo). 2. Identification of precursor cell lines for fascia and muscle. Demonstration of involvement of PPAR ␥ nuclear receptors and transcription factors within the differentiation process of fascia derived stem cells. 3. Optimization of long-term delivery of IGF-1 plasmid through stabilization in biodegradable scaffold for extended availability and transfection of precursor cells. 4. Quantitate and correlate the amount of IGF-1 and insulin required per unit volume of tissue and study local/systemic effects of IGF-1 and insulin. 5. In vivo demonstration of autologous soft-tissue generation within biodegradable scaffold gel systems in rabbit pectoral model. This phase will involve optimization of volume/surface ratio of the scaffold and controlling the safety measures of the induction process. Confirmation of long-term results in animal models is an essential step for the translational human applications. 6. Safety and maintenance of the trophism concepts should be investigated further as well. References 1. Ailhaud G: Adipose cell differentiation: A long way to Tipperary. In Angel A, Anderson H, Bouchard C,
et al (eds): Progress of Obesity Research. London, John Libbey and Company, 1996, pp 3–11 2. Alper J: New insights into type 2 diabetes. Science 289:37–38, 2000 3. Arnez ZM, Khan U, Pogorelec D, et al: Rational selection of flaps from the abdomen in breast reconstruction to reduce donor site morbidity. Br J Plast Surg 52:351–354, 1999 4. Aubert J, Marc PS, Belmonte N, et al: Prostacyclin IP receptor up-regulates the early expression of C/ EBPBeta and C/EBPGamma in preadipose cells. Mol Cell Endometab 160:149–156, 2000 4a. Bland KI, Copeland EM III (eds): The Breast: Comprehensive Management of Benign and Malignant Diseases. Philadelphia, W. B. Saunders, 1991, pp 18– 116 5. Boney CM, Moats-Staats BM, Stiles AD, et al: Expression of insulin-like growth factor-1 (IGF-1) and IGFbinding proteins during adipogensis. Endocrinology 135:1863, 1994 6. Bostwick J III: Plastic and Reconstructive Breast Surgery. St. Louis, Quality Medical Publishing, 1990, pp 555–577 7. Chawla A, Lazar MA: Peroxisome proliferator and retinoid signaling pathways co-regulate preadipocyte phenotype and survival. Proc Natl Acad Sci U S A 91:1786–1790, 1994 8. Cinti S, Eberbach S, Castellucci M, et al: Lack of insulin receptors affects the formation of white adipose tissue in mice. Diabetologica 41:171–177, 1998 9. Cleek RL, Rege AA, Denner LA, et al: Inhibition of smooth muscle cell growth in vitro by antisense oligodeoxynucleotide released from poly(DL-lacticco-glycolic acid) microparticles. J Biomed Materials Res 35:525, 1997 10. Dani C: Embryonic stem cell-derived adipogenesis. Cells, Tissues, Organs 165:173–180, 1999 11. Fan LT, Singh SK: Controlled release—a quantitative treatment. Berlin, Springer Verlag, 1989, pp 105–106 12. Feller AM, Galla TJ: The deep inferior epigastric artery perforator flap. Clin Plast Surg 25:197–206, 1998 13. Gradus-Pizlo I, Wilensky RL, March KL, et al: Local delivery of biodegradable microparticles containing colchicine or a colchicine analogue: Effects on restenosis and implications for catheter-based drug delivery. J Ame Coll Cardiol 26:1549–1557, 1995 14. Gregorie FM, Smas CM, Sul HS: Understanding adipocyte differentiation. Phys Rev 78:783–809, 1998 15. Hauner H, Petruschke TH, Rohrig K: Effects of epidermal growth factor, platelet growth factor and fibroblast growth factor on human adipocyte development and function. Eur J Clin Invest 25:90–96, 1995 16. Hauner H, Schmid P, Pfeiffer EF: Glucocorticoids and insulin promote the differentiation of human adipocyte precursor cells into fat cells. J Clin Endometab Metab 64:832–835, 1987 17. Heinrich N, Fechner K, Berger H, et al: In-vivo release of a GnRH agonist from a slow-release poly(lactide-glycolide) copolymer preparation: Comparison in rat, rabbit and guinea-pig. J Pharm Pharmacol 43:762–765, 1991 18. Hinton JL Jr, Warejcka DJ, Mei Y, et al: Inhibition of epidural scar formation after lumbar laminectomy in the rat. Spine 20:564–570, 1995 19. Hu E, Zhu Y, Fredrickson T, et al: Tissue restricted expression of two human frzbs in preadipocytes and pancreas. Biochem Biophys Res Comm 247:287–293, 1998 20. Jaiswal RK, Jaiswal N, Bruder SP, et al: Adult human
NEW RESEARCH IN BREAST RECONSTRUCTION
21. 22.
23.
24.
25.
26. 27.
28.
29.
30. 31. 32.
mesenchymal stem cell differentiation to the osteogenic or adipogenic lineage is regulated by mitogenactivated protein kinase. J Biol Chem 275:9645–9652, 2000 Katz AJ, Llull R, Hedrick MH, et al: Emerging approaches to the tissue engineering of fat. Clin Plast Surg 26:587–603, 1999 Kawaguchi N, Toriyama K, Lena EN, et al: De novo adipogenesis in mice at the site of injection of basement membrane and fibroblast growth factor. Proc Natl Acad Sci U S A 95:1062–1066, 1998 Kim HS, Hausman GJ, Hausman DB, et al: The expression of PPAR gamma in pig fetal tissue and primary stromal-vascular cultures. Obes Res 8:83– 88, 2000 Kruisbrink J, Boer GJ: Controlled long-term release of small peptide hormones using a new microporous polypeptide polymer: Its application for vasopressin in the Brattleboro rat and potential perinatal use. J Pharmaceut Sci 73:1713–1718, 1984 Kras KM, Hausman DB, Martin RJ: Tumor necrosis factor-alpha stimulates cell proliferation in adipose tissue-derived stromal-vascular cell culture. Obesity Res 8:186–193, 2000 Mooney DJ, Breuer C, McNamara K, et al: Fabricating tubular devices from polymers of lactic and glycolic acid for tissue engineering. Tissue Eng 1:107, 1995 Ohgushi MM, Yanagida Y, Haruyama T, et al: Osteogenesis coordinated in C3H10T1/2 cells by adipogenesis-dependent BMP-2 expression system. Tissue Eng 6:9–18, 2000 Petruschke TH, Hauner H: Tumor necrosis factoralpha prevents the differentiation of human adipocyte precursor cells and causes delipidation of newly developed fat cells. J Clin Endometab Metab 76:742– 747, 1993 Petruschke TH, Rohrig K, Hauner H: TGF-beta inhibits the differentiation of human adipocyte precursor cells in primary culture. Int J Obes Relat Metab Disord 18:532–536, 1994 Porras A, Santos E: The insulin/Ras pathway of adipocytic differentiation of 3T3 L1 cells. Int J Obes 20 (suppl3):S43–S51, 1996 Puigserver P, Adelmant G, Wu Z, et al: Activation of PPARgamma coactivator-1 through transcription factor docking. Science 286:1368–1371, 1999 Ross SE, Hemati N, Longo KA, et al: Inhibition of
33.
34. 35.
36.
37. 38.
39. 40.
41.
42.
43.
44.
125
adipogenesis by wnt signaling. Science 289:950–953, 2000 Saladin R, Fajas L, Halvorsen YD, et al: Differential regulation of peroxisome proliferator activated receptor gammal and PPAR gamma2 messenger RNA expression in the early stages of adipogenesis. Cell Growth Differ 10:43–48, 1999 Shih C, Higuchi T, Himmelstein KJ: Drug delivery from catalysed erodible polymeric matrices of poly– ortho)esters. Biomaterials 5:237, 1984 Smas CM, Chen L, Zhao L, et al: Transcriptional repression of pref-1 by glucocorticoids promotes 3T3L1 adipocyte differentiation. J Biol Chem 274:12632– 12641, 1999 Teruel T, Valverde AM, Benito A, et al: IGF-1 and insulin induce adipogenic related gene expression in fetal brown adipocyte primary cultures. Biochem J 319:627–632, 1996 Thomson RC, Wake MC, Yaszemski MJ, et al: Biodegradable polymer scaffolds to regenerate organs. Adv Polym Sci 122:245–274, 1995 Thuillier P, Baillie R, Sha X, et al: Cytosolic and nuclear distribution of PPARgamma2 in differentiating 3T3-L1 preadipocytes. J Lipid Res 39:2329–2338, 1998 Wake MC, Gupta PK, Mikos AG: Fabrication of pliable biodegradable polymer foams to engineer soft tissues. Cell Transplantation 5:465–473, 1996 Wang Y, Lee-Kwon W, Martindale JL, et al: Modulation of CCAAT/enhancer-binding protein-alpha gene expression by metabolic signals in rodent adipocytes. Endocrinology 140:2938–2947, 1999 Webitsch M, Henze E, Hauner H, et al: Biological effects of human growth hormone in rat adipocyte precursor cells and newly differentiated adipocytes in primary culture. Metabolism 45:34–42, 1996 Yuksel E, Weinfeld AB, Cleek R, et al: Augmentation of adiposo-fascial flaps using the long-term local delivery of insulin and insulin-like growth factor-1. Plast Rec Surg 106:373–382, 2000 Yuksel E, Weinfeld AB, Cleek R, et al: Adipose tissue generation through long-term, local delivery of insulin and IGF-1 by PLGA/PEG microspheres in an in vivo rat model. Plast Reconstr Surg 105:1721–1729, 2000 Yuksel E, Weinfeld AB, Wamsley S, et al: Increased free fat graft survival with the long-term, local delivery of insulin, IGF-1, and bFGF by PLGA/PEG microspheres. Plast Rec Surg 105:1712–1720, 2000 Address reprint requests to Eser Yuksel, MD 6560 Fannin, Suite 800 Houston, TX 77030 e-mail:[email protected]
0094–1298/02 $15.00 .00
NEW RESEARCH IN BREAST RECONSTRUCTION Adipose Tissue Engineering Saleh M. Shenaq, MD, and Eser Yuksel, MD
In the human species, a major function of the breast is aesthetic. The soft-tissue volume within the breast displaces the overlying skin to create the protuberant contour of the female thorax, which is associated solidly with, and to some extent, definitive of, femininity in modern culture. Adipose tissue is the major contributor to the volume of the breast. Despite the high emotional, physical, and aesthetic cost, partial or complete mastectomy, with or without radiation therapy, remains the cornerstone of the modern management for breast cancer. 1 This surgical procedure removes the volume element from the breast obliterating the feminine contour. Currently, techniques for postmastectomy reconstruction include the transverse rectus abdominis myocutaneous (TRAM) flap, latissimus dorsi flap with or without alloplastic implants, and implants with or without skin expansion.2 Although effective, these meth-
ods frequently are limited by high donor-site morbidity and difficulty in restoring the specific volume, shape, size, texture, and sensation characteristics of a normal breast. Use of implants is avoided in prospective radiation treatment cases because of possible implant extrusion and wound healing problems. Additionally, implant application has a risk for developing capsule contracture, which may require further surgical interventions to correct the deformity and to eliminate the pain. Based on these facts, autologous breast reconstruction remains the primary choice in most cases. A major donor-site morbidity of the TRAM flap is associated with the weakness of the anterior abdominal wall musculofascial system, however, and is well recognized by patients and physicians and has led plastic surgeons to search for less morbid operations, such as perforator-based muscle-sparing free TRAM flaps.3, 4 The mor-
Financial Support Parts of this work were funded by: A developmental grant from NIH Breast Spore Grant A PSEF Research Grant An ASERF Research Grant Samuel Spira Research Fund
From the Division of Plastic Surgery, De Bakey Department of Surgery, Baylor College of Medicine, Houston, Texas
CLINICS IN PLASTIC SURGERY VOLUME 29 • NUMBER 1 • JANUARY 2002
111
112
SHENAQ & YUKSEL
bidity of latissimus dorsi musculocutaneous pedicled flaps includes unsightly donor-site scars and the alteration of the shoulder functional integrity. This morbidity is in addition to the inability to avoid use of implants in most of the cases. An ideal solution to the problem of the absent breast after mastectomy would be a method that restores or adds soft-tissue volume to the breast without donor-site morbidity or the risks associated with allogenic implants. Considering the trend to perform skin-sparing mastectomy modalities without altering the principles and defined frame of the ideal oncologic treatment for breast cancer, missing volume component at the mastectomy site can be considered the main denominator of the reconstructive surgical design. Adipose tissue is the most appropriate tissue to be used for this role because not only does it remain as the main fraction of the removed volume but it also is the most compatible tissue for missing glandular component. TISSUE ENGINEERING: GENERAL PRINCIPLES Tissue engineering is the science of generating tissue by using the principles of molecular biology and material engineering. The main elements to be optimized in tissue engineering are the cell and extracellular matrix and cross-talk between these elements. Other than cross-talk of cell and matrix, the generation of the vascular network of the tissue is an indispensable component of this dynamic process. Interestingly, the vascular phase itself is dependent on the complex management of cell and extracellular matrix interaction within the vascular compartment by the organized and adequately reinforced release of angiogenic signals released from the generating tissue. The main principles of tissue engineering research are discussed in a general descriptive level in the following section. The Cell The Cell Source The cell is the center of all events at the molecular level and is a prerequisite player
in the regeneration and maintenance of the tissue. Any organ in the body consists of more than one type of cell; however, a specific cell type exists in all tissue that structurally and functionally dominates the medium. To simplify the engineering process, a single-cell type usually is chosen to be manipulated to control the matrix and generative events. The cell used for this purpose can be autogenous or allogenous in origin. Allogenic cells do not possess a long-term or permanent cycle in living systems, unless an immune modulation of the donor or recipient is executed or embryonic stem cells are used. Despite its high differentiative and proliferative potential for diverse types of tissue, the use of embryonic stem cells currently is being evaluated because of the associated major ethical and legal issues that are brought by the opponents. Therefore, the conventional direction is toward the use autologous cells. The next question is the use of committed and end-stage differentiated cells or pluripotential noncommitted cells. Use of pluripotential mesenchymal stem cells or noncommitted precursor cells is favored because of their high capacity for prospective differentiation and multiplication. The third level dilemma in the cell source is whether to use the cells in vitro or to stimulate them in vivo. In vitro use is limited to low volume-surface ratio tissue types (such as skin) or to expand research information for isolated manipulations in controlled parameter systems. In vivo applications are discussed in the research background section. The Manipulation of the Cell Regardless of the cell source chosen, the cell has to be informed, stimulated, and maintained for accelerated and directed behavioral pattern because it is silent or reluctant without the right stimulus vector. This goal can be achieved by extracellular matrix modification such as long-term delivery of bioactive factors or the configurational effect of extracellular matrix structural elements. This stimulation is converted to intracellular messenger events, which in turn alter the gene transcription profile or post-transcriptional modifications for a period as long as the extracellular vector ex-
NEW RESEARCH IN BREAST RECONSTRUCTION
113
ists. The other method is presenting the selfcopying gene vectors to the cell for the synthesis of a particular protein to trigger the cascade of events for targeted behavior. Transcription factors also can be used in a selective fashion for targeted behavioral manipulation.
check systems for adequacy of the growth or by the limiting effects of extracellular matrix composition, which also includes the vascular support, and tissue barriers for repelling signals to counteract the forwarding vector. This task requires use of different matrices in specialized prefabrication models.
Tissue Transformers and Stimulators
Maintenance Phase
Differentiative Phase The noncommitted precursor cells can be stimulated to differentiate to a specialized cell type, which declares the main constitution for the targeted tissue and acts as the primary cell structurally and functionally. There are specific transcription factors serving for different tissue transformation processes. The adipose tissue transformers are discussed in the following sections. Targeting differentiated cells for transformation involves complex signaling pathways, which enables dedifferentiation and differentiation processes consecutively. Proliferative and Migratory Phase To fill a targeted volume for regeneration, precursor and differentiated cells should be under the influence of the right vector in magnitude and in direction. This influence is achieved by the organization of cell to extracellular matrix protein cross-talk, and cell-tocell cross-talk, which in turn dictates the cell dynamics. The paracrine form of this interaction has an exclusive and mandatory task of stimulating angiogenesis. This task can be reinforced independently in the design of the tissue regeneration model by delivery of angiogenic factors, genes, or stimulatory vectors. The synchronized orchestration of tissue migration and angiogenesis in a three-dimensional pattern is well performed in in vivo prefabrication systems, unlike the inadequacy of this partnership in in vitro systems. Maturation and Control of GrowthFruition Phase The tissue growth may be controlled by manipulating the intracellular diagnostic and
The signal for the cell should be stable in the long term to maintain the trophism at the plateau level. Fortunately, in natural regeneration models, the cell is capable of modifying the extracellular matrix to provide the external support for its existence. This cellular decision is highly affected by external vectors, however, which alters the oxidation-reduction equilibrium.
Extracellular Matrix Optimization Engineering of material can be evaluated in two disciplines: one is three-dimensional design and prefabrication of scaffold systems, and the other area is based on optimizing components of the biomaterial, which involves cross-linking and stabilization of different matrix elements. Macrodesign The computer-aided engineering modalities currently are developed for different tissue and organ templates. Porosity, or organization of matrix, acts as a major denominator for tissue migration. This task can be achieved by different processing techniques.5–8 Micropatterning The cell’s behavior is affected by the extracellular matrix pattern in nanometer scale. The progress in nanotechnology will enable manufacturing the nano-patterns for different tissue models. Currently, it is possible to manufacture micropatterns at 0.20 micrometer scale, and in the year 2005, it is predicted that this will be achieved at 0.10 micrometer scale.
114
SHENAQ & YUKSEL
Optimization of Signaling Properties of Extracellular Matrix Cell-to-extracellular matrix cross-talk can be directed with the sustained release of bioactive proteins.9, 10 The release system might involve use of biodegradable microspheres or chemical stabilization of factors in extracellular matrix back bone. Additionally, the threedimensional configuration of structural matrix proteins and their composition are effective in altering the cellular behavior. The use of strong signaling molecules should be fashioned within the principles of gradient difference to provide the maintenance of the migration vector toward the targeted direction in adequate magnitude.
In Vitro/In Vivo Prefabrication Models In the authors’ laboratory, the concept of in vivo tissue engineering for high volume surface ratio regeneration models is emphasized. This concept describes the generation of tissue by processes occurring completely inside the living body. In vivo tissue engineering, like traditional in vitro models of tissue engineering, involves the use of cells to generate tissues. Unlike traditional tissue engineering, however, there is no ex vivo cell seeding and therefore no need for cell harvesting and culturing procedures. The cells that are used to generate the desired tissues are mesenchymal precursor cells already present in the region requiring reconstruction. Therefore, the optimum guidance of these cells with the stimulant vectors within the principles discussed above can lead to selfsupporting tissue growth in terms of vascular component. The in vivo models can be developed as orthotopic or heterotopic prefabrication models.
tems in the presence of DNA damage. Furthermore, local regeneration designs should be exclusive for systemic effects and should not alter the treatment of primary disease if any exists. Another challenge in translation of tissue engineering modalities for human trials is the difference in behavioral patterns of organisms. Particularly, the template size in animal models is smaller in dimensions whereas the cell size is approximately equal when compared with human models. DIFFERENTIATION OF ADIPOSE TISSUE Adipose tissue is formed by terminally differentiated adipocytes and their committed precursor cells called preadipocytes. The earlier members of adipocyte lineage include noncommitted stem cells and adipoblasts, which are speculated to exist in various compartments. Multiple steps of adipose cell differentiation involve fatty acid activated receptor, peroxisome proliferator–activated receptor ␥, insulin-like growth factor-1 (IGF-1), insulin, retinoids, triiodothyronine, and prostaglandins (I2, D, and J series).11 The differentiation from pluripotential fibroblastic cell to the next step is not well identified, however.12 The differentiation cascade may be initiated by several growth factors. These factors include basic fibroblastic growth factor, IGF-1, and tumor necrosis factor ␣ (TNF ␣).12–14 TNF ␣ also is demonstrated as an inhibitor of adipogenesis. 15 Transforming growth factor  (TGF ) and epidermal growth factor also have been shown to be inhibitory within this process.16, 17 Adipocytic differentiation of preadipocytes is enhanced when Preadipocyte Factor 1 (an epidermal growth factor–like domain containing transmembrane protein) is underexpressed (such as with dexamethasone administration).18 Glucocorticoids and insulin can increase the differentiation rate by 30- to 70- fold.19
Safety Issues Manipulating the cell cycle toward the accelerated proliferation rate opposes the primary goal on which the cancer research is focused. Thus, performing this task should exclude the alteration of cell cycle check sys-
Insulin, Insulin-like Growth Factor-1, and Intracellular Seconder Messanger Pathways Insulin and GF-1 have been demonstrated to promote adipocytic differentiation in sev-
NEW RESEARCH IN BREAST RECONSTRUCTION
eral steps.11, 20–22 Adipogenic action of the insulin starts with tyrosine kinase receptors and proceeds with the involvement of the Ras pathway,23 while glucose transport regulation is performed through PI 3-kinase pathway, which is the other option for insulin after insulin receptor substrate (IRS) binding proteins are activated.4a This is particularly important to developing a hypothesis around the postreceptor defects in type 2 diabetes and to understanding the obesity, insulin, diabetes mellitus (DM) triangle.
Nuclear Receptors and Transcription Factors Peraxisome proliferator activated receptor (PPAR) ␥ receptors are involved primarily in terminal differentiation stages.12, 24, 25 Based on this fact, targeting the PPAR-induced transcription factors or activators of PPAR is the main interest of many research centers focused on obesity or adipogenesis.26–31
Role of Frizzled Related Protein (Frzbs) and Wnt in Adipocyte Differentiation Frzb is a newly discovered family of glycoproteins that modulate (antagonistic) signaling activity of Wnt.32 Wnts are secreted signaling proteins that are involved in the developmental processes. Inhibition of their signal leads to terminal differentiation of adipocytes.33 Thus frzbs might be new targets for reversing the suppression of adipocytic differentiation.
Stem Cell Differentiation When the focus of interest is to develop adipose tissue from other sources, as indicated in orthotopic in vivo models,34 the availability of earlier members of adipocytic lineage and their identification become more imperative. The stem cells from nonadipocytic sources have been identified and differentiated into adipocytic-committed members. 35–37 An interesting finding in these reports is the involvement of mitogen-acti-
115
vated kinases, which puts insulin in a central or dominant location in induction of stem cells. These findings support the observations in the authors’ research.34 In these cells, osteogenic differentiation is regulated by retinoic acid receptor ␣, which can be targeted for inhibition to provide the shift toward adipocytic differentiation.37
SOFT-TISSUE ENGINEERING EXPERIENCE The authors’ current research objectives are to develop a noninvasive breast reconstruction method for postmastectomy defects using in vivo tissue-engineering methods and to use a totally biodegradable system that enables the sustained release of adipogenic factors that in turn stimulate the adipogenic differentiation of recipient cells in the pectoral fascia. When the research objectives are achieved, in clinical perspective the system will be incorporated with an endoscopic delivery technique to the pectoralis muscle and fascia without interfering with the operative site and remaining breast tissue, particularly for delayed and skin sparing mastectomy reconstructions. The authors’ laboratory has worked to develop a biodegradable system that stimulates the growth of adipose tissue to be used in an orthotopic in vivo tissue-engineering approach to breast reconstruction and augmentation mammaplasty. The experimental phases the authors have followed in their laboratory’s soft-tissue engineering research are listed in Table 1.
Adipogenic Factors and Long-term Delivery Method Adipocyte precursor cells (preadipocytes) contained in postnatal human tissues have the capability to proliferate and differentiate into mature lipid-containing adipocytes. In vitro experiments have demonstrated that insulin and IGF-1 can stimulate these preadipocyte activities.38 In vitro experiments demonstrate that insulin and IGF-1 stimulate adipocyte proliferation, differentiation, and
116
SHENAQ & YUKSEL
Table 1. EXPERIMENTAL PHASES OF IN VIVO SOFTTISSUE PREFABRICATION Adipogenic Factors Long-term Delivery Method: Biodegradable Microsphere Processing Incorporation with Growth Factors Controlled Release Kinetics In Vivo Application: Improvement of Fat Graft Survival Rates Augmentation of Adiposo-fascial Flaps Augmentation of Vascularity in Graft Prefabrication De Novo Fat Generation Biodegradable Scaffolds: Template Processing Degradation Kinetics Combination with microspheres Breast Prefabrication: Template Processing Flap Models In Vivo Regeneration
trophic activities. Additionally, lipohypertrophy is reported as a side effect of repeated subcutaneous insulin injections. Other researchers have demonstrated that a subpopulation of preadipocytes with the potential to differentiate into mature, lipid-containing adipocytes exists within postnatal adipose tissue. IGF-1 has a negative effect on intracellular lipid content, and insulin’s action is to increase the intracytoplasmic lipid content despite the fact that they act in the same direction through similar tyrosine kinase receptors to stimulate the differentiation and proliferation of the preadipocytes. Numerous biodegradable drug delivery systems are in various stages of development. These systems have been employed experimentally in a variety of tissues39 and a variety of species40 to deliver a variety of pharmacologic compounds (cytotoxic agents, steroids, peptide/protein hormones, or cytokines). Various degrees of clinically relevant results have been obtained.41, 42 The theoretic advantages of biodegradable drug delivery systems are the ability to deliver an active agent for a long duration with the effects potentially limited to a local region without multiple surgical procedures and without the implantation of a permanent foreign body.43 Poly(lactic–co–glycolic–acid)–polyethylene glycol (PLGA–PEG) microspheres are a biodegradable drug delivery system that currently is undergoing experimentation and further development (Cleek). 9 PLGA-PEG micro-
spheres are capable of delivering proteins and cDNA for long durations,9 and molded PLGA scaffolds serve well for guided tissue regeneration.5 Microsphere Preparation Microspheres were prepared using the double solvent emulsion method.9, 34, 44 A Coulter multisizer (Coulter Electronics, Miami, FL) was used to confirm a mean particle size of 12 m. Scanning electron microscopy was employed to demonstrate normal microsphere morphology (Fig. 1). During the manufacture process, the ‘‘drug’’ to be delivered is added as an aqueous solution to the polymer solution. This biphasic mixture is mechanically homogenized so that small aqueous droplets are suspended in the polymer solution. The polymer is precipitated into a solid phase trapping the aqueous droplets. The water then is sublimated away leaving the ‘‘drug’’ behind in numerous hollow cavities. The inclusion of PEG in the polymer solution leads to the formation of multiple microchannels. The result is closed porous microspheres with a network of microchannels with the ‘‘drug’’ trapped in the pores. As the microspheres undergo slow biodegradation, the eroding surface meets the cavities, and the growth factors are released. Incorporation with Growth Factors and In Vitro Release Kinetics An in vitro experiment was designed to confirm the microsphere’s ability to accomplish long-term peptide delivery. In this experiment, only insulin-containing microspheres were studied. Equal weights of microspheres were placed in four equal volumes of sterile and neutral phosphate buffered saline. The microsphere solutions were permitted to incubate at physiologic temperatures for four separate time periods each; 7, 14, 21, and 28 days. Measurements of optical density and PAGE-electrophoresis were employed to provide a quantitative and qualitative analysis (respectively) demonstrating increased insulin concentration with increased incubation/degradation time thus confirming release for extended duration.
NEW RESEARCH IN BREAST RECONSTRUCTION
117
Figure 1. Scanning electron microscopic images for PLGA/ PEG microspheres of 12–15 m average diameter (original magnification 1000).
In Vivo Applications Improvement of Fat Graft Survival Rates The first model the authors have used to test the efficacy of the adipogenic factors within a sustained release system was the fatgraft model. In this investigation, the effects of long-term, local delivery of insulin, IGF-1, and bFGF on fat graft survival have been evaluated using a poly (lactic-co-glycolicacid)—polyethylene glycol (PLGA-PEG) microsphere delivery system. Twelve-micrometer PLGA-PEG microspheres incorporated separately with insulin, IGF-1, and basic fibroblast growth factor (bFGF) were manufactured using a double-emulsion-solvent-extraction technique. Inguinal fat from Sprague Dawley rats was harvested, diced, washed, and mixed with 1) insulin microspheres, 2) IGF-1 microspheres, 3) basic fibroblast growth factor microspheres, 4) a combination of the insulin and IGF-1 microspheres, and 5) a combination of insulin, IGF-1, and bFGF microspheres. The treated fat grafts were implanted autologously into subdermal pockets in six
animals for each group. Animals receiving untreated fat grafts and fat grafts treated with blank microspheres constituted two external control groups (six animals per external control group). At 12 weeks all fat graft groups were compared on the basis of weight maintenance and a histomorphometric analysis of ‘‘adipocyte area percentage’’ (AAP); indices of volume retention and cell composition, respectively. Weight maintenance was defined as the final graft weight as a percent of the implanted graft weight. All growth factor treatments significantly increased fat graft weight maintenance objectively and volume maintenance grossly in comparison with the untreated and blank microsphere treated controls. Treatment with insulin and IGF-1, alone or in combination, was found to increase the AAP in comparison with fat grafts treated with bFGF alone or in combination with other growth factors. In conclusion, the findings of this study indicated that long-term, local delivery of growth factors with PLGA-PEG microspheres has the potential to increase fat graft survival rates. Furthermore, the type of growth factor delivered may influence the cellular/stromal composition of the grafted tissue.43
118
SHENAQ & YUKSEL
Augmentation of Adiposofascial Flaps The adiposofascial flaps currently described in the literature frequently lack the volume requirements for reconstructive goals. In this study, the authors examined the use of long-term local delivery of insulin and IGF1 together by PLGA-PEG microspheres to augment inguinal adiposofascial flaps based on inferior epigastric vessels in the rat. In this experiment, a group receiving only an operation and a group treated with blank (no growth factor) microspheres served as external controls for the surgical procedure and the drug delivery device, respectively. In all groups (n 5), the contralateral flap served as internal controls. Upon harvest on the twenty-eighth postoperative day, the insulin and IGF-1 treated flaps in both models weighed statistically more than the internal control flaps and the two external control flaps. Likewise, on gross inspection, the adipogenic growth factor–treated flaps had greater volumes than their internal control flap groups and both of the external control flap groups (operation only and blank microspheres). Other intergroup comparisons suggested the absence of a systemic insulin and IGF-1 effect on adiposity. A histomorphometric analysis suggested that 1) insulin and IGF1 treatment do not alter flap cell composition and 2) flap augmentation is secondary to the stimulation of cell proliferation and adipocytic differentiation rather than hypertrophy of mature adipocytes. Further evidence in favor of cell proliferation and differentiation was the finding of nonanatomic, ectopic fat islands on the treated flaps’ pedicle sheath and no variation in cell size distribution among groups. It was concluded that the long-term local delivery of insulin and IGF-1 with PLGA-PEG microspheres is an effective method for adiposofascial flap augmentation through the increase in mature adipocyte number rather than an increase in the cell size of pre-existing cells.44 Augmentation of Vascularity in a Fat Graft Prefabrication Model In this phase the authors have observed the effects of long-term delivery of bFGF to the
base (predicted pedicle of the prefabricated flap) area of fat grafts. The vascular density was measured using color distraction of microphotographs and Image Pro software. The results revealed that vascularity was increased significantly in the bFGF treated group (Fig. 2). De Novo Fat Generation This study was undertaken to characterize the duration of y and to evaluate the potential of the long-term delivery of insulin and IGF1 for the de novo generation of adipose tissue in vivo. Insulin and IGF-1–containing microspheres were administered directly to the deep muscular fascia of the rat abdominal wall. A group of animals treated with blank microspheres served as an external control. At the 4-week harvest period, multiple ectopic islands of adipose tissue were observed on the abdominal wall of the animals treated with insulin, IGF-1, and insulin plus IGF-1 microspheres (Fig. 3). Such islands were not seen in the blank microsphere group. Hematoxylin and eosin–stained sections of the growth factor groups demonstrated mature adipocytes interspersed with fibrous tissue superficial to the abdominal wall musculature and continuous with the fascia. Oil Red O– stained sections demonstrated that these cells contained lipid. Computer-aided image analysis of histologic sections confirmed that there were statistically significant increases in the amount of ‘‘ectopic’’ adipose neotissue developed on the abdominal wall of animals treated with growth factor microspheres. In conclusion, this study confirmed the longterm release of proteins from PLGA-PEG microspheres and demonstrated the potential of long-term, local insulin and IGF-1 to induce adipogenic differentiation to mature, lipidcontaining adipocytes from nonadipocyte cell pools in vivo.34 Scanning electron microscopy of the anterior abdominal fascia sections demonstrated the predominancy of extracellular elements and scarcity of cells in the blank microsphere group while the entire surface was covered by adipocytes in the insulin and IGF-1–treated group (Fig. 4). There were three possible explanations for the adipogenic phenomenon:
NEW RESEARCH IN BREAST RECONSTRUCTION
Figure 2. A, Prefabrication of fat-fascia flap using free fat grafts and angiogenic factors. B, Vascular density is measured using Image Pro software (Media Cybernetics, Del Mar, CA) for the images after green extraction and Sobel filtering has been applied. bFGF delivery (microsphere) group has higher vascular density values (statistically significant, P ⬍.01)
119
120
SHENAQ & YUKSEL
Figure 3. Representative gross image demonstrating the de novo neotissue as ectopic islands well integrated on the surface of the abdominal wall in an IGF-1 insulin treated animal.
Figure 4. Scanning electron microscopy of the anterior abdominal fascia. A, Blank microsphere group: predominancy of extracellular elements and scarcity of cells. B, Insulin and IGF-1–treated group: the entire surface is covered by adipocytes. (original magnification 1200.)
NEW RESEARCH IN BREAST RECONSTRUCTION
1. Mesenchymal stem cells (MSCs) might be present in the abdominal wall fascia and could have been stimulated to differentiate into preadipocytes and then differentiate into mature adipocytes. 2. A preadipocyte pool may exist within the abdominal wall fascia that might have been stimulated to differentiate into adipocytes. 3. A process of dedifferentiation might have occurred for a subset of the fibroblasts normally contained in the abdominal wall fascia. These cells then could have differentiated into preadipocytes followed by subsequent adipocytic differentiation.
Biodegradable Scaffold Engineering In this phase of the study, the authors have investigated the role of a rigid scaffold system for guided tissue regeneration in the previously described stimulatory system.
121
chymal cells in the recipient tissue bed. A breast-shaped implant was engineered from biodegradable PLGA foam (Fig. 5). This implant was designed as a porous scaffold to allow the infusion of microspheres and to serve as sources of support and guidance for new adipose tissue. The implant was used in conjunction with insulin and IGF-1 to create an in vivo tissue engineering system that generated shaped adipose tissue of a predetermined shape in an in vivo rat model (Fig. 6). The evaluation of the scaffolds at 5 weeks revealed migration of tissue towards the center of the template (Fig. 7). At 12 weeks, the entire scaffold was replaced by invading tissue, which was fibroelastic in gross examination (Fig. 8). The histologic examination suggested that adipose component dominated in the regenerated tissue (Fig. 8). The results demonstrated the potential power of in vivo tissue engineering for producing adipose tissue in situ for breast reconstruction and augmentation without flap transfer or the use of allogenic expanders or implants.
Template Processing PLGA 75:25 foams were produced by a particulate leaching technique using 300 to 500 m salt (NaCl) particles as a porogen.6 Salt particles were dispersed in a PLGA-methylene chloride solution yielding a 90% fraction of salt. Hemisphere-shaped constructs (2.5 cm in diameter) were placed in 70% ethyl alcohol for 12 hours to induce polymer cross-linking. The salt was leached from the PLGA constructs with distilled deionized water, creating a vast array of pores to convert the material to a foam.
Scaffold Implantation Breast Prefabrication—In Vivo Regeneration The authors’ laboratory has developed an in vivo approach to breast reconstruction and augmentation using a biodegradable implant in an animal model. This implant system is designed to stimulate its own replacement by new adipose tissue generated by the mesen-
Current and Future Phases In past experiments, the authors have used biodegradable PLGA breast-shaped scaffolds and biodegradable PLGA-PEG microspheres delivery of adipogenic (insulin and IGF) growth factors to achieve guided adipose tissue generation in vivo. Although this was the first report of adipose tissue generation in vivo, scaffold replacement fell short by 30%. Currently, the authors aim at testing the hypothesis that the introduction of the extracellular matrix molecule hyaluronan to the scaffold system will enhance the tissue generation rates and allow total tissue replacement of the breast-shaped biodegradable scaffold. Another work is focused on employing gel matrices within a silicon spacer induced capsule model (Fig. 9). This model not only enables the objective measurement of the regenerating tissue within the capsule but also serves as a model for pre-existing silicon implant carriers. The future phases of the authors’ research are directed towards the following aims:
122
SHENAQ & YUKSEL
Figure 5. Manufacturing 75:25 poly-lacto glycolic acid (PLGA) scaffolds. Process: particulate leaching method using NaCl as a porogen. Vulcanization into hemispheric form with a force of 250 lb at 160C. Porosity: 90%, pore size: 300–500 micron. Scanning electron microscopy was applied after freeze fracturing of the scaffolds (original magnification 300).
Figure 6. PLGA/PEG microspheres of 12–15 m average diameter are dispersed within the hemispheric scaffold (75:25 PLGA), porosity: 90%, pore size: 300–500 micron using specialized closed injection system with pluronic acid gel. Scanning electron microscopy was applied after freeze fracturing of the scaffolds (original magnification 600).
NEW RESEARCH IN BREAST RECONSTRUCTION
Figure 7. Harvest of 75:25 PLGA scaffolds, implanted beneath the muscle. At 5 weeks, tissue growth to the center can be observed.
Figure 8. Harvest of scaffold at 12 weeks. A, The entire scaffold is replaced by soft tissue. B, Hematoxylin-eosin stained cross-section shows that adipocytes predominate within the regenerated tissue.
123
124
SHENAQ & YUKSEL
Figure 9. Use of silicon spacers for capsule prefabrication. Gel matrices are applied within this isolated compartment model.
1. Incorporation of biodegradable extracellular matrix gel system (in vivo). 2. Identification of precursor cell lines for fascia and muscle. Demonstration of involvement of PPAR ␥ nuclear receptors and transcription factors within the differentiation process of fascia derived stem cells. 3. Optimization of long-term delivery of IGF-1 plasmid through stabilization in biodegradable scaffold for extended availability and transfection of precursor cells. 4. Quantitate and correlate the amount of IGF-1 and insulin required per unit volume of tissue and study local/systemic effects of IGF-1 and insulin. 5. In vivo demonstration of autologous soft-tissue generation within biodegradable scaffold gel systems in rabbit pectoral model. This phase will involve optimization of volume/surface ratio of the scaffold and controlling the safety measures of the induction process. Confirmation of long-term results in animal models is an essential step for the translational human applications. 6. Safety and maintenance of the trophism concepts should be investigated further as well. References 1. Ailhaud G: Adipose cell differentiation: A long way to Tipperary. In Angel A, Anderson H, Bouchard C,
et al (eds): Progress of Obesity Research. London, John Libbey and Company, 1996, pp 3–11 2. Alper J: New insights into type 2 diabetes. Science 289:37–38, 2000 3. Arnez ZM, Khan U, Pogorelec D, et al: Rational selection of flaps from the abdomen in breast reconstruction to reduce donor site morbidity. Br J Plast Surg 52:351–354, 1999 4. Aubert J, Marc PS, Belmonte N, et al: Prostacyclin IP receptor up-regulates the early expression of C/ EBPBeta and C/EBPGamma in preadipose cells. Mol Cell Endometab 160:149–156, 2000 4a. Bland KI, Copeland EM III (eds): The Breast: Comprehensive Management of Benign and Malignant Diseases. Philadelphia, W. B. Saunders, 1991, pp 18– 116 5. Boney CM, Moats-Staats BM, Stiles AD, et al: Expression of insulin-like growth factor-1 (IGF-1) and IGFbinding proteins during adipogensis. Endocrinology 135:1863, 1994 6. Bostwick J III: Plastic and Reconstructive Breast Surgery. St. Louis, Quality Medical Publishing, 1990, pp 555–577 7. Chawla A, Lazar MA: Peroxisome proliferator and retinoid signaling pathways co-regulate preadipocyte phenotype and survival. Proc Natl Acad Sci U S A 91:1786–1790, 1994 8. Cinti S, Eberbach S, Castellucci M, et al: Lack of insulin receptors affects the formation of white adipose tissue in mice. Diabetologica 41:171–177, 1998 9. Cleek RL, Rege AA, Denner LA, et al: Inhibition of smooth muscle cell growth in vitro by antisense oligodeoxynucleotide released from poly(DL-lacticco-glycolic acid) microparticles. J Biomed Materials Res 35:525, 1997 10. Dani C: Embryonic stem cell-derived adipogenesis. Cells, Tissues, Organs 165:173–180, 1999 11. Fan LT, Singh SK: Controlled release—a quantitative treatment. Berlin, Springer Verlag, 1989, pp 105–106 12. Feller AM, Galla TJ: The deep inferior epigastric artery perforator flap. Clin Plast Surg 25:197–206, 1998 13. Gradus-Pizlo I, Wilensky RL, March KL, et al: Local delivery of biodegradable microparticles containing colchicine or a colchicine analogue: Effects on restenosis and implications for catheter-based drug delivery. J Ame Coll Cardiol 26:1549–1557, 1995 14. Gregorie FM, Smas CM, Sul HS: Understanding adipocyte differentiation. Phys Rev 78:783–809, 1998 15. Hauner H, Petruschke TH, Rohrig K: Effects of epidermal growth factor, platelet growth factor and fibroblast growth factor on human adipocyte development and function. Eur J Clin Invest 25:90–96, 1995 16. Hauner H, Schmid P, Pfeiffer EF: Glucocorticoids and insulin promote the differentiation of human adipocyte precursor cells into fat cells. J Clin Endometab Metab 64:832–835, 1987 17. Heinrich N, Fechner K, Berger H, et al: In-vivo release of a GnRH agonist from a slow-release poly(lactide-glycolide) copolymer preparation: Comparison in rat, rabbit and guinea-pig. J Pharm Pharmacol 43:762–765, 1991 18. Hinton JL Jr, Warejcka DJ, Mei Y, et al: Inhibition of epidural scar formation after lumbar laminectomy in the rat. Spine 20:564–570, 1995 19. Hu E, Zhu Y, Fredrickson T, et al: Tissue restricted expression of two human frzbs in preadipocytes and pancreas. Biochem Biophys Res Comm 247:287–293, 1998 20. Jaiswal RK, Jaiswal N, Bruder SP, et al: Adult human
NEW RESEARCH IN BREAST RECONSTRUCTION
21. 22.
23.
24.
25.
26. 27.
28.
29.
30. 31. 32.
mesenchymal stem cell differentiation to the osteogenic or adipogenic lineage is regulated by mitogenactivated protein kinase. J Biol Chem 275:9645–9652, 2000 Katz AJ, Llull R, Hedrick MH, et al: Emerging approaches to the tissue engineering of fat. Clin Plast Surg 26:587–603, 1999 Kawaguchi N, Toriyama K, Lena EN, et al: De novo adipogenesis in mice at the site of injection of basement membrane and fibroblast growth factor. Proc Natl Acad Sci U S A 95:1062–1066, 1998 Kim HS, Hausman GJ, Hausman DB, et al: The expression of PPAR gamma in pig fetal tissue and primary stromal-vascular cultures. Obes Res 8:83– 88, 2000 Kruisbrink J, Boer GJ: Controlled long-term release of small peptide hormones using a new microporous polypeptide polymer: Its application for vasopressin in the Brattleboro rat and potential perinatal use. J Pharmaceut Sci 73:1713–1718, 1984 Kras KM, Hausman DB, Martin RJ: Tumor necrosis factor-alpha stimulates cell proliferation in adipose tissue-derived stromal-vascular cell culture. Obesity Res 8:186–193, 2000 Mooney DJ, Breuer C, McNamara K, et al: Fabricating tubular devices from polymers of lactic and glycolic acid for tissue engineering. Tissue Eng 1:107, 1995 Ohgushi MM, Yanagida Y, Haruyama T, et al: Osteogenesis coordinated in C3H10T1/2 cells by adipogenesis-dependent BMP-2 expression system. Tissue Eng 6:9–18, 2000 Petruschke TH, Hauner H: Tumor necrosis factoralpha prevents the differentiation of human adipocyte precursor cells and causes delipidation of newly developed fat cells. J Clin Endometab Metab 76:742– 747, 1993 Petruschke TH, Rohrig K, Hauner H: TGF-beta inhibits the differentiation of human adipocyte precursor cells in primary culture. Int J Obes Relat Metab Disord 18:532–536, 1994 Porras A, Santos E: The insulin/Ras pathway of adipocytic differentiation of 3T3 L1 cells. Int J Obes 20 (suppl3):S43–S51, 1996 Puigserver P, Adelmant G, Wu Z, et al: Activation of PPARgamma coactivator-1 through transcription factor docking. Science 286:1368–1371, 1999 Ross SE, Hemati N, Longo KA, et al: Inhibition of
33.
34. 35.
36.
37. 38.
39. 40.
41.
42.
43.
44.
125
adipogenesis by wnt signaling. Science 289:950–953, 2000 Saladin R, Fajas L, Halvorsen YD, et al: Differential regulation of peroxisome proliferator activated receptor gammal and PPAR gamma2 messenger RNA expression in the early stages of adipogenesis. Cell Growth Differ 10:43–48, 1999 Shih C, Higuchi T, Himmelstein KJ: Drug delivery from catalysed erodible polymeric matrices of poly– ortho)esters. Biomaterials 5:237, 1984 Smas CM, Chen L, Zhao L, et al: Transcriptional repression of pref-1 by glucocorticoids promotes 3T3L1 adipocyte differentiation. J Biol Chem 274:12632– 12641, 1999 Teruel T, Valverde AM, Benito A, et al: IGF-1 and insulin induce adipogenic related gene expression in fetal brown adipocyte primary cultures. Biochem J 319:627–632, 1996 Thomson RC, Wake MC, Yaszemski MJ, et al: Biodegradable polymer scaffolds to regenerate organs. Adv Polym Sci 122:245–274, 1995 Thuillier P, Baillie R, Sha X, et al: Cytosolic and nuclear distribution of PPARgamma2 in differentiating 3T3-L1 preadipocytes. J Lipid Res 39:2329–2338, 1998 Wake MC, Gupta PK, Mikos AG: Fabrication of pliable biodegradable polymer foams to engineer soft tissues. Cell Transplantation 5:465–473, 1996 Wang Y, Lee-Kwon W, Martindale JL, et al: Modulation of CCAAT/enhancer-binding protein-alpha gene expression by metabolic signals in rodent adipocytes. Endocrinology 140:2938–2947, 1999 Webitsch M, Henze E, Hauner H, et al: Biological effects of human growth hormone in rat adipocyte precursor cells and newly differentiated adipocytes in primary culture. Metabolism 45:34–42, 1996 Yuksel E, Weinfeld AB, Cleek R, et al: Augmentation of adiposo-fascial flaps using the long-term local delivery of insulin and insulin-like growth factor-1. Plast Rec Surg 106:373–382, 2000 Yuksel E, Weinfeld AB, Cleek R, et al: Adipose tissue generation through long-term, local delivery of insulin and IGF-1 by PLGA/PEG microspheres in an in vivo rat model. Plast Reconstr Surg 105:1721–1729, 2000 Yuksel E, Weinfeld AB, Wamsley S, et al: Increased free fat graft survival with the long-term, local delivery of insulin, IGF-1, and bFGF by PLGA/PEG microspheres. Plast Rec Surg 105:1712–1720, 2000 Address reprint requests to Eser Yuksel, MD 6560 Fannin, Suite 800 Houston, TX 77030 e-mail:[email protected]