A Localizable, Biological-based System for the Delivery of Bioactive IGF-1 Utilizing Microencapsulated Genetically Modified Human Fibroblasts

Abstract

Insulin-like growth factor 1 (IGF-1) is a potent mitogen and differentiation factor with particular relevance to orthopedic tissue engineering. A biologically based Ca²⁺-alginate microcapsule vehicle, utilizing genetically modified primary normal human fibroblasts (NHFs), was developed and characterized for localized synthesis and delivery of human IGF-1 (hIGF-1). Normal human fibroblasts were transfected to overexpress the hIGF-1 gene, leading to cells that expressed 4 ng of hIGF-1 per 10⁶ cells per 24 hours. Encapsulation within alginate led to a six-fold enhancement in the generation and release of hIGF-1 to 22 ng of hIGF-1 per 10⁶ cells per 24 hours. Release was constitutive, predictable, and exhibited highly repeatable first-order kinetics with no initial burst. Released growth factor was biologically active and exhibited a proliferative effect comparable to commercially available recombinant hIGF-1. The magnitude of hIGF-1 release met the requirements of orthopedic tissue generation, and this approach is considered an attractive alternative to other proposed methods of growth factor delivery.

Growth factors provide progenitor cells with the biochemical signals necessary for tissue development, and thus delivery of such bioactive molecules is increasingly recognized as a necessary component in many tissue engineering strategies.^1–3 For many in situ applications, localizable growth factor delivery could permit cytokines to stimulate target cells without adversely affecting other sites of the body. Delivery must also be consistent and predictable over the time course of therapeutic need.

Microencapsulation, a process originally developed for biohybrid organs, offers a promising technology for the controlled delivery of growth factors. Allogeneic or xenogeneic cells, genetically engineered to express or overexpress a therapeutic factor of interest, can be encapsulated within semi-permeable membranes and implanted into a host at the site of tissue repair. Barrier materials are formulated with a permeability that allows for the bidirectional diffusion of oxygen, nutrients, and cytokines, while isolating encapsulated donor cells from direct cell-cell contact with the host, thereby lessening immunorejection.⁴

Cell encapsulation has been explored as a potential therapeutic platform for a variety of applications, many of which utilize semi-permeable hollow fiber membranes to separate grafted cells from the host.^5–7 Such approaches have gone on to be successfully evaluated in humans through Phase I-III clinical trials in the past, notably for treatments of chronic pain,⁸ Huntington's disease,⁹ amyotropic lateral sclerosis,⁸ and more recently, for retinal degeneration.¹⁰

An alternative encapsulation material is alginate.^11–13 This marine- derived polysaccharide is composed of chains of β-D-mannuronic acid and α-L-guluronic acid residues that can crosslink through interactions with divalent metal ions such as Ca²⁺ or Ba²⁺. Cells synthesizing therapeutic molecules can be entrapped within an alginate hydrogel undergoing cross-linking, and the release of therapeutics from the hydrogel can be adjusted by varying the encapsulated cell density and gelling and cross-linking properties of alginate.^14–16 Alginate microcapsules can be maintained stable for long periods of time, and when implanted, alginate hydrogels provide a highly hydrated three-dimensional matrix structure similar to natural extracellular matrix which can serve as scaffolding to progenitor cells.¹⁵ Recent studies in animal models have explored the use of alginate-based encapsulated cells as potential therapies for obesity,¹⁷ cardiovascular disease,¹⁸ and liver failure.¹⁹In vitro, the utility of encapsulated cells in alginate has been recently examined through studies utilizing Schwann cells,²⁰ retinal pigment epithelial cells,¹³ and hepatocytes,²¹ among others.

Insulin-like growth factor (IGF-1), also known as somatomedin C, is a potent mitogen and differentiation factor known to be present in most tissues in vivo.^22,23 Insulin-like growth factor 1, which has a molecular weight of 7.65 kDa, plays a significant role in many cartilage tissue engineering applications with the ability to increase matrix synthesis^24–27 and stimulate cellular proliferation in chondrocytes,^{24–26,28,29} as well as stimulate chondrogeneic differentiation.²⁹ For bone, IGF-1 has also been shown to have a proliferative effect on osteoblasts^23,30 and stimulate osteogeneic differentiation.³¹

This report describes the development and characterization of a biologically based Ca²⁺-alginate microcapsule system employing genetically modified human fibroblasts for the local synthesis and delivery of bioactive human IGF-1 (hIGF-1) for use in orthopedically relevant applications. Measurements of morphology, release kinetics, released growth factor bioactivity, and cell viability are presented and compared with other delivery systems described in the literature.

Materials and Methods

Cell Culture

Primary normal human fibroblasts (NHFs) were harvested from the dermis of neonatal foreskins (Women and Infants' Hospital, Providence, RI) with appropriate institutional review board approval and cultured as previously described.³² Except where stated otherwise, NHFs were cultured at 37°C and 10% CO₂ in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS, Invitrogen, Carlsbad, CA) and 1% penicillin/streptomycin (Invitrogen, Carlsbad, CA).

Cell Transfection

Normal human fibroblasts were genetically modified using a retroviral vector (MFG-hIGF-1) containing the human gene encoding hIGF-1 as established by Eming et al.²² and the retroviral gene transfer technique established by Morgan et al.³³ Briefly, the vector was transfected into a packaging cell line, and the resulting stable virus-producing cells were grown in a 10-cm Petri dish. Upon reaching confluence, culture medium was aspirated and fresh medium was added to the dish (10 ml/dish). After 24 hours, culture medium was removed, filtered using a 0.45-μm filter, and stored at −80°C. For transduction, 1.15 × 10⁵ NHFs (passage 4) were seeded in a 35-mm petri dish and cultured overnight. A thawed stock of retrovirus was incubated with 80 μg/ml of chondrotin sulfate C from shark cartilage (Sigma Aldrich, St. Louis, MO) at 37°C for 10 minutes, followed by incubation with 80 μg/ml of polybrene (1,5-dimethyl-1,5-diazaundecamethylene polymethobromide, hexadimethrine bromide, Sigma Aldrich, St. Louis, MO) for an additional 10 minutes at 37°C. The mixture was centrifuged at 10,000 rpm for 5 minutes at room temperature, and the visible pelleted virus was isolated and resuspended to one-tenth of its original volume in cell culture medium. In the dish containing the NHFs, the culture medium was completely removed. The virus-containing culture medium was added and cells were incubated at 37°C for 24 hours. Normal human fibroblasts were then washed and incubated with fresh culture medium for an additional 24 hours. Cells were trypsinized and plated in 10-cm tissue culture dishes. To assess the success of transfection, modified and unmodified cells were then grown to confluence in culture medium containing 1% FBS. For 48 hours postconfluence, conditioned medium was collected and assayed for concentration of hIGF-1 using an enzyme-linked immunosorbent assay (ELISA, R&D Systems, Minneapolis, MN) and Spectromax Absorbance plate reader (Molecular Devices, Sunnydale, CA).

Transfection was confirmed by the overexpression of hIGF-1 from modified cells compared with unmodified controls. To determine if hIGF-1 production was stable with cell passage, conditioned medium from cells from passages 6–10 was collected and assessed for concentration of hIGF-1 using ELISA. Upon confirmation of a stable transfection, cells were frozen in liquid nitrogen at passages 6–7 to create a working cell bank.

Microencapsulation

After thawing, cells were expanded in tissue culture flasks before use. For all studies, passage 9–11 polyclonal cells were used. After expansion, cells were trypsinized and counted using a hemacytometer. Cells were then suspended in sterile-filtered 1.8% alginate solution (Sigma Aldrich, St. Louis, MO) containing 0.9% sodium chloride (Sigma Aldrich, St. Louis, MO) at a target density of 1 × 10⁶ cells per ml of alginate. Using a commercially available encapsulator (Inotech, Dottikon, Switzerland), the cell suspension was extruded through a 100-μm nozzle at a vibration of approximately 5,000 Hz and a flow rate of 1.8 ml per minute. Individual droplets were collected in a stirring bath of 0.15 M calcium chloride (Sigma Aldrich, St. Louis, MO) solution with 5 mM HEPES (pH = 7.4, Sigma Aldrich, St. Louis, MO); calcium ions cross-linked the alginate co-polymer chains, leading to the formation of Ca²⁺-alginate microcapsules. Capsules were allowed to gel and harden for 10 minutes, and then filtered from the bath using a 40-μm mesh strainer (BD Biosciences, San Jose, CA). Capsules were then washed three times with sterile DMEM and resuspended in culture medium.

Postencapsulation Capsule Characterization

Microcapsule shape was determined using a phase contrast microscope. Microcapsule size was determined using a Coulter LS230 particle size analyzer (Beckman Coulter, Fullerton, CA), and is presented as mean volumetric diameter ± standard deviation (μm). This diameter was used to determine the volume of a capsule with an assumed geometry of a uniform sphere.

The number of capsules in a sample was calculated visually. Briefly, a 0.1 ml aliquot was removed from the sample and placed in a tissue culture dish. The number of capsules in this aliquot was then counted using light microscopy. The capsule density (capsules per ml) was multiplied by the total volume of the sample to obtain the total number of capsules in the sample.

Encapsulated Cell Count and Viability

Encapsulated cell viability was determined immediately after capsule fabrication, at the end of the 10-day release study, and at every time point in the proliferation studies. To calculate the cell viability, capsules were incubated in 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, Sigma Aldrich, St. Louis, MO) at 10 mg/ml in phosphate buffer solution (PBS, Invitrogen, Carlsbad, CA) for 4 hours. Alginate capsules were then solubilized via incubation in 55 mM sodium citrate (Sigma Aldrich, St. Louis, MO) solution containing 0.45% NaCl and 10 mM HEPES (pH = 7.4) for 5 minutes. Cells were collected via centrifugation at 400g for 5 minutes, resuspended in DMEM, and then counted using a hemocytometer. Live cells, which reduced the MTT dye to a dark purple formazan precipitate within their mitochondria, were stained purple, while necrotic cells remained unstained. The number of live and dead cells was counted, with viability represented as the ratio of live cells to the total number of cells encapsulated. The mean number of live cells per capsule was determined by dividing the number of live cells in an aliquot by the number of capsules in the aliquot.

IGF-1 Release Rates

For unencapsulated cell release studies, NHFs were grown to confluence in 6-well tissue culture plates containing DMEM with 10% FBS. At hours 12, 24, 36, and 48, the conditioned culture medium in each well was completely removed and fresh culture medium was added to the well. Conditioned medium samples were stored at −80°C for later analysis. At the end of the study, cells were detached from the wells using 0.25% trypsin with EDTA (Invitrogen, Carlsbad, CA) and counted using a hemocytometer.

For encapsulated studies, alginate capsules containing NHFs were incubated in 6-well tissue culture plates containing DMEM with 10% FBS. Each well was plated with approximately 50,000–100,000 encapsulated cells based on postencapsulation cell count. At days 2, 4, 6, 8, and 10, each sample was transferred to a sterile 15-ml conical tube and centrifuged at 400g for 5 minutes to isolate capsules from supernatant. The supernatant was then completely removed using care to not remove any capsules and fresh culture medium was added to the conical tube. Supernatant samples were stored at −80°C for later analysis of hIGF-1 content. At the end of the study, cell count and viability was assessed by incubating capsules with MTT dye before solubilization and counting with a hemocytometer.

The quantity of human IGF-1 (hIGF-1) from each supernatant sample was determined using ELISA. All samples were run in triplicate, and amount released was normalized to one million viable cells per sample as counted at the end of each release kinetics study. Fresh culture medium containing FBS was also measured for hIGF-1 content to account for any cross-reactivity of bovine-derived IGF-1 with the ELISA. Profiles were plotted as cumulative mass release versus time, with all data points shown as mean ± standard deviation. Release rates were determined by obtaining the slope from first-order linear regression forced through the axis on the profiles.

Cell Proliferation and Viability

Encapsulated NHFs were incubated in 6-well tissue culture plates containing DMEM with 10% FBS. Each well was plated with approximately 25,000–50,000 cells based on the postencapsulation cell count. Cell proliferation and viability within the capsules was assessed over the course of 7-days with culture medium changes occurring every 3 days. On days 0, 1, 3, 5, and 7, six wells of capsules were incubated with MTT dye before solubilization and cells were counting with a hemocytometer to determine the total number of cells in each well. The percent cell viability for the cells in each well was determined as the ratio of live cells to the total number of cells encapsulated. Cell viability is reported as mean ± standard deviation; significance (p < 0.01) was determined using a t-test assuming unequal variances.

IGF-1 Bioactivity Assay

A release study was completed using assay medium consisting of DMEM/F-12 (1:1; Invitrogen, Carlsbad, CA) supplemented with 1% penicillin/streptomycin, 0.2% bovine serum albumin (Sigma Aldrich, St. Louis, MO), and 10 μg/ml transferrin (Invitrogen, Carlsbad, CA). Alginate capsules containing NHFs were incubated in 6-well tissue culture plates containing bioactivity assay medium. At days 2, 4, 6, 8, and 10, all supernatant was completely removed in each well and replenished with fresh assay medium. Supernatant samples were stored at −80°C for later analysis of hIGF-1 content.

The biological activity of hIGF-1 from supernatant samples was then measured using a protocol adapted from Karey et al.³⁴ and R&D Systems (Minneapolis, MN). Briefly, MCF-7 cells (HTB-22; ATCC, Manassas, VA) were washed three times and resuspended in assay medium at a density of 1 × 10⁵ cells/ml. 5,000 cells were then plated per well, followed by 75 μl of assay medium containing commercially available recombinant hIGF-1 diluted in 0.1% BSA (0–48 ng/ml; R&D Systems, Minneapolis, MN) or hIGF-1 released from encapsulated NHFs (supernatant from release study conducted in assay medium). All assay medium used on cells was analyzed via ELISA for hIGF-1 concentration. Plates were incubated at 37°C and 5% CO₂. After five days, medium was completely aspirated and 100 μL of fresh assay medium along with 20 μL of CellTiter-Blue reagent (Promega, Madison, WI) was added to each well. An eight-point MCF-7 cell number standard curve (0–5 × 10⁴ cells) was also plated using fresh assay medium and CellTiter-Blue reagent. Plates were incubated for 4 hours. Cell number in each well was determined using a Gemini XPS spectrofluorometer (Molecular Devices, Sunnyvale, CA) at 560_EX/590_EM. All standard curves and samples were run in triplicate and a cell number standard curve was generated using a four-parameter fit.

Statistical Analysis

Data is expressed as means ± standard deviations. The standard deviation for the slope obtained from linear regression was calculated as the ratio of the standard deviation of the residuals and the square root of the sum of the residuals.³⁵ Significance (p < 0.01) was determined using a two-tailed t-test assuming unequal variances. All statistical analysis was performed using SigmaPlot software (Systat Software Inc., San Jose, CA) and Microsoft Excel (Microsoft Corporation, Redmond, WA).

Results

Microencapsulation

Figure 1 is a phase-contrast microscopy image of NHFs immobilized within Ca²⁺-alginate microcapsules. Cells, referenced by the arrows, were distributed at low density throughout approximately spherical capsules in a contracted form within the Ca²⁺-alginate matrix. Cells remained in this contracted state for the duration of culture. The cell packing density for the capsules (see Methods) was approximately 30–40 cells per capsule. Using Coulter particle sizing techniques, volumetric diameter was measured to be 457 ± 172 μm. Capsule volume was calculated to be approximately 0.05 mm³ assuming spherical capsule geometry. Capsule size and loading cell density could be increased or decreased by adjusting the frequency of the vibration and adjusting the number of cells suspended in alginate prior to encapsulation (data not shown). Assuming a nominal cell diameter of 20 μm, the total volume of a capsule occupied by the cells was approximately 0.2%–0.3%.

Figure 1.: Normal human fibroblasts (NHFs) are uniformly immobilized within Ca²⁺-alginate microcapsules. Phase contrast microscopy indicates that NHFs are uniformly immobilized at low density throughout Ca²⁺-alginate microcapsules. Image shows a cross-sectional view of only one plane in the capsules. Capsule solubilization and cell counting within capsules postencapsulation yields approximately 30–40 cells per capsule; volumetric diameter is approximately 500 μm. Arrows reference cells within microcapsules in a contracted morphology. Scale bar denotes 200 μm.

Release of hIGF-1

No hIGF-1 was detected by ELISA in fresh culture medium containing 10% FBS. The cumulative release profiles of hIGF-1 from the culture medium in which encapsulated and unencapsulated NHFs were cultured are shown in Figure 2 (unencapsulated) and Figure 3 (encapsulated). Release is presented as cumulative mass released in culture medium as a function of time in culture, and this quantity is normalized to the number of cells counted in the culture at the end of the study. Table 1 summarizes the rate of release obtained as the slope from linear regression. In unencapsulated culture, modified NHFs released 3.9 ± 0.0 ng hIGF-1 per 10⁶ cells per 24-hours, while unmodified NHFs released no hIGF-1 as detected by ELISA. When encapsulated, modified and unmodified NHFs secreted 21.4 ± 1.0 and 0.6 ± 0.1 ng hIGF-1/(10⁶ cells · day), respectively.

Figure 2.: Unencapsulated genetically modified normal human fibroblasts (NHFs) expressing the human insulin-like growth factor 1 (hIGF-1) gene have higher cumulative release of hIGF-1 than unmodified control cells. Genetically modified NHFs (•) and unmodified controls (○) were grown to confluence in 6-well plates and conditioned medium was analyzed at various time points for 2-day cultures postconfluence using ELISA. Mass released was normalized to one million viable cells counted at the completion of each study. All data points representing mean ± standard deviation. Profiles are fit to first-order linear regression forced through the axis; see Table 1 for release rates (n = 6).

Figure 3.: Encapsulated modified normal human fibroblasts (NHFs) have increased synthesis of human insulin-like growth factor 1 (hIGF-1) relative to unmodified control cells. Mass released from encapsulated genetically modified NHFs (•) and unmodified controls (○), also encapsulated, is plotted versus time in culture. Data is shown as mean ± standard deviation. Comparison of the release rate of hIGF-1 (slope of regressed line) from encapsulated cells to that of unencapsulated cells as shown in Figure 2 (different time scale for x-axis) demonstrates encapsulation significantly increases the release of hIGF-1 from genetically modified NHFs (n = 12).

Table 1: Release Rates of hIGF-1 From Encapsulated or Unencapsulated NHFs

Cell Proliferation and Viability

No significant changes in the number of both live and dead cells within capsules (no proliferation) were observed when measured daily over 7-days in vitro. In addition, initial and final counts remained the same for the 10-day release study. Figure 4 presents the viability of encapsulated cells immediately postencapsulation (n = 2) and at 2-day intervals over 7-days in culture (n = 6). Results, shown as percentage cell viability versus time, demonstrate that percent viability remained constant, and averaged to be 75.4% ± 5.8% for the duration of the study. At the end of the 10-day release studies, the average value of cell viability was 77.2% ± 5.8%.

Figure 4.: Encapsulated cell viability is not compromised significantly after seven days in culture. Viability of encapsulated unmodified and modified normal human fibroblasts (NHFs) was measured immediately following the encapsulation (n = 2), and at various time-points during a 7-day viability study (n = 6). Results are presented as percentage cell viability versus time, and data points are shown as mean ± standard deviation. Percent viability remains constant and >70% for the duration of the study.

IGF-1 Bioactivity

As shown in Figure 5, the presence of recombinant hIGF-1, both commercially available and released by encapsulated NHFs, resulted in a significant increase in growth rate of MCF-7 cells as compared with cells grown in medium containing no hIGF-1. Furthermore, in the presence of medium conditioned by encapsulated unmodified NHFs releasing no hIGF-1, MCF-7 cells exhibited no significant increase in growth rate as compared with cells grown in the assay medium containing no hIGF-1. Overall, hIGF-1 released by encapsulated modified NHFs resulted in a proliferative response equivalent to that achieved with commercially available hIGF-1 at similar dosages.

Figure 5.: Human insulin-like growth factor 1 (hIGF-1) released by encapsulated modified normal human fibroblasts (NHFs) causes increased proliferation in MCF-7 cells, a response which is equivalent to that achieved with commercially available hIGF-1 at similar dosages. Cell population doublings for MCF-7 cells are shown as a function of the medium in which cells were cultured in for 5 days. Cells cultured in assay medium containing no hIGF-1 and assay medium containing 1.4 ng/ml of commercially available recombinant hIGF-1 (R&D Systems, Minneapolis, MN) serve as negative and positive controls, respectively. Cells cultured in assay medium that was conditioned by encapsulated unmodified cells (0 ng/ml of hIGF-1) and encapsulated modified cells (1.5 ng/ml of hIGF-1) serve as the experimental study arm. Rate of cell population doubling in the presence of hIGF-1 (rightmost bar) is equivalent to that of commercial hIGF-1 at equivalent concentrations measured via enzyme-linked immunosorbent assay (ELISA).

Discussion

Retroviral transfection of the hIGF-1 gene into NHFs yielded a cell phenotype which overexpressed biologically active hIGF-1 and which could be banked as a polyclonal working cell population for further studies. In modified NHFs, hIGF-1 expression was stable and linear over time with a release rate of 4 ng of hIGF-1 per million cells per day as compared to unmodified cells, which released no hIGF-1.

Encapsulation resulted in the entrapment of cells within Ca²⁺-alginate microcapsules with spherical morphology and cell packing density, as observed visually. Capsule dimensions facilitated adequate diffusional transport for maintenance of cell viability. Synthesis and subsequent release of hIGF-1 from encapsulated cells exhibited well-behaved first-order kinetics, with no initial burst or starburst effect present. Bovine IGF-1, derived from FBS present in fresh culture medium, was not cross-reactive with the hIGF-1 ELISA. hIGF-1 released from encapsulated cells displayed biological activity in stimulating MCF-7 cells (Figure 5) equivalent to commercially available recombinant hIGF-1.

Microencapsulation yielded an absolute increase in the release rate of hIGF-1 for unmodified cells from 0 to 0.6 ng hIGF-1 per 10⁶ cells per 24-hours. Additionally, daily release rates from microcapsules increased six-fold for modified cells as compared to unencapsulated modified cells. Initial and final counts of cells per capsule indicated no cell proliferation within the capsules over time. Thus, the higher release is not a consequence of intracapsule expansion of cells. Comparison of release profiles unequivocally demonstrates higher expression and release of hIGF-1 upon encapsulation for both modified and unmodified cells. Significantly, this finding stands in contrast to other growth factor studies that have shown a decrease in cellular product release with encapsulation. For example, encapsulated modified NHFs showed decreased production of transforming growth factor beta 1.³⁶ Utilizing the same cell type as is presented in this report, this study showed on average, a two-fold increase in the number of encapsulated cells over the course of 7 days of in vitro culture.³⁶ The reason for this increase has not been established with certainty. The encapsulated NHFs in our 10-day study did not proliferate to any significant extent so increased IGF-1 release is not due to a greater number of cells. More likely, some aspect of encapsulation may be upregulating IGF-1 expression from the endogenous IGF-1 gene as well as the IGF-1 transgene. Upregulation would occur at the transcriptional and/or translational levels and more experimentation is needed to determine this mechanism. This study of the effect of encapsulation on cellular metabolism may provide valuable insight to both this IGF-1 delivery system as well as other encapsulation technologies, and thus, might allow further optimization of encapsulated cells as a platform technology for the delivery of cellular therapeutics.

Both the rate and stability of release of hIGF-1 achieved by microencapsulated cells are consistent with the levels required in orthopedic tissue engineering and regenerative medicine applications. For example, increased matrix synthesis has been reported in chondrocytes via bolus application of 25–100 ng of IGF-1 per ml of culture medium every 1–3 days for one to 3 weeks in vitro.^25–27 This dosage is well-within the range achievable by our alginate microcapsules. The capsules have shown constant first-order release through at least 10-days in culture, with no indication of diminished activity toward the end of that time frame and thus have utility for such durations of application. With respect to this concentration dose, with a loading of 35 cells per capsule, approximately 31,000 to 126,000 capsules (volume of 1.6–6.4 cm³, respectively) would release 25–100 ng of hIGF-1 per day, respectively. Although this volume may be suitable for in vitro studies, it may be too large for in vivo applications, where a defect site may be smaller than 1 cm³. In such cases, cell packing density within the microcapsules could likely be increased simply by adding more cells to the alginate prior to extrusion. In addition, our laboratories are developing alternative fabrication technologies that yield much higher cell packing densities as well as examining the impact of higher loading on intracapsule cell viability. The alginate microcapsules delivering hIGF-1 provide a slow continuous release of growth factor over time as compared with the high burst of delivery from bolus injection application,^25,27 and therefore continuous dosage required to reach the same efficacy as the bolus delivery might be well lower, requiring fewer capsules than anticipated. Finally, in vitro bolus delivery system employ growth factors stored for long intervals postsynthesis.^25,27 The microcapsule-based system delivers freshly synthesized growth factor in situ. Comparison of continuous delivery of freshly synthesized growth factor with intermittent (bolus) delivery of off-the-shelf moieties represents another potentially fruitful area for future study.

The approach described here relies upon allogeneic rather than autologous cells. Autologous cells are certainly the ideal transplant candidate in terms of immuno-compatibility, but are difficult to harvest and require ex-vivo expansion, accompanied by significant cost and regulatory complexities, prior to transfection and re-transplantation into the host. Some investigators have skirted this lengthy ex-vivo processing by in situ transfection of autologous cells with adenoviral vectors.^28,37 Although adenoviruses have been approved for use in human gene transfer clinical trials targeting a variety of monogeneic diseases and carcinomas, the possibility of illiciting an immune response upon in vivo viral transfection remains a concern.³⁷ Allogeneic cells, which are abundant in supply, provide an alternative, less-costly cell source. These cells can be harvested from a donor, transfected ex-vivo, banked, characterized, and then implanted into the host at the site of repair. This method decreases the invasiveness of gene therapy and cell transplantation to the host in two ways: first, by eliminating the use of viral vectors directly in the host, and second, by requiring the host to undergo only a single operation. For cartilage repair, the transplantation of genetically modified allogeneic cells has been studied by Madry et al.³⁸ through the transplant of IGF-1 producing rabbit chondrocytes into an osteochondral defect in a rabbit model and Goodrich et al.³⁹ through the transplant of IGF-1 producing equine chondrocytes for delivery to an equine joint model. In both studies, primary allogeneic cells were transfected with adenoviral methods ex vivo, encapsulated within a biomaterial matrix, and transplanted to the site of repair within 1-day posttransfection.^38,39 Encouragingly, both studies suggest in situ cartilage repair is enhanced in the presence of transplanted genetically modified IGF-1 releasing cells after 36 weeks, demonstrating the utility of localized IGF-1 delivery for orthopedic tissue generation.^38,39 In both studies, after transfection and encapsulation, the modified cells were shown to have high expression of IGF-1 up to 4 days posttransfection followed by a significant decrease in genetic expression over time,^38,40 a behavior which is typical to adenoviral transfection methods.^28,37 In contrast, modified NHFs, reported here and which were prepared by retroviral transfection, have shown constant, stable expression of IGF-1 for at least 10 days in culture and over the course of multiple passages, permitting the creation of cell bank useful for future testing. Although the encapsulated modified NHFs released IGF-1 at lower concentrations than the initial burst released by these adenoviral-transfected cells, the overall stability of the continuous delivery system may well make this approach more effective in cartilage repair. Additionally, allogeneic NHFs derived from neonates as demonstrated in this paper may have more advantageous cellular metabolism and gene expression as compared to autologous cells derived from the elderly, the patient population to benefit the most from cartilage repair techniques.

Several further areas of research, some basic and some translational, are identified and recommended. The significant increase in release of IGF-1 as a result of encapsulation with Ca²⁺-alginate warrants elaboration of the cellular-biomaterial mechanism(s) involved. Continuous delivery of freshly synthesized IGF-1 should be compared with bolus delivery of off-the-shelf growth factor. In vitro release studies, reported here, should be extended to in vitro and in vivo models of orthopedic efficacy. Translational studies facilitating the use of this system in vivo include maximizing cell loading per capsule and examining encapsulated cell viability, bioactivity of the released growth factor, and host immune response both postimplantation and after extended culture in vivo.

Conclusion

Retroviral transfection can be employed to create bankable NHF cells that overexpress bioactive hIGF-1. This overexpression increases by more than an order of magnitude when the cells are encapsulated within Ca²⁺-alginate matrices, and encapsulation has no significant impact on the biological activity of the growth factor as compared with commercially available hIGF-1. Release rates of approximately 21 ng of hIGF-1 per 10⁶ cells per 24-hours are achievable, which is consistent with the requirement for applications in orthopedic tissue engineering and regenerative medicine. Identified advantages of this delivery system include highly repeatable constitutive delivery with no initial starburst and cost-effective generation of therapeutic qualities of bioactive IGF-1.

Acknowledgment

Supported, in part, by the Office of Research and Development Rehabilitation R&D Service, Department of Veterans Affairs, and the Center for Restorative and Regenerative Medicine, Providence VA Medical Center.