In vitro release of vascular endothelial growth factor from gadolinium-doped biodegradable microspheres

Abstract

A drug delivery vehicle was constructed that could be visualized noninvasively with MRI. The biodegradable polymer poly(DL-lactic-co-glycolic acid) (PLGA) was used to fabricate microspheres containing vascular endothelial growth factor (VEGF) and the MRI contrast agent gadolinium diethylenetriamine pentaacetic acid (Gd-DTPA). The microspheres were characterized in terms of size, drug and contrast agent encapsulation, and degradation rate. The PLGA microspheres had a mean diameter of 48 +/- 18 microm. The gadolinium loading was 17 +/- 3 microg/mg polymer and the VEGF loading was 163 +/- 22 ng/mg polymer. Electron microscopy revealed that the Gd was dispersed throughout the microspheres and it was confirmed that the Gd loading was sufficient to visualize the microspheres under MRI. VEGF and Gd-DTPA were released from the microspheres in vitro over a period of approximately 6 weeks in three phases: a burst, followed by a slow steady-state, then a rapid steady-state. Biodegradable Gd-doped microspheres can be effectively used to deliver drugs in a sustained manner, while being monitored noninvasively with MRI.

FIG. 1

Differential interference contrast (DIC) light micrograph of VEGF/Gd-DTPA PLGA microspheres. Scale bar = 100 μm.

FIG. 2

Size distribution of VEGF/Gd-DTPA PLGA microspheres. The graph shows the contribution to the total volume of microspheres vs. microsphere diameter. The mean ± SD of this volume-weighted distribution is 48 ± 18 μm.

FIG. 3

Electron micrographs of VEGF/Gd-DTPA PLGA microsphere. a: Low-loss transmission electron micrograph of a single sphere. b: Energy-filtered transmission electron micrograph (EFTEM) of same sphere in a showing localization of Gd. Dark regions in a indicating regions of high electron density correspond to bright regions in b which contain Gd. Scale bar = 2 μm.

FIG. 4

R₁ relaxation rates for 8% gelatin, empty microspheres, filtrate, and Gd-DTPA microspheres. The empty microspheres did not have an effect on the relaxation rate, indicating that the PLGA by itself had no T₁ effect. The result for the filtrate is a measure of the contribution of nonencapsulated Gd-DTPA to the overall relaxation rate enhancement of the Gd-DTPA microspheres.

FIG. 5

Fast spin-echo image of VEGF/Gd-DTPA PLGA micro-spheres after 30 days of degradation. The three samples have been centrifuged and the microsphere pellets appear as a bright region at the bottom of the tubes. Imaging parameters were TR/TE = 10 sec / 10 ms, echo train length = 4, inversion time = 1600 ms, resolution = 1.25 × 1.25 mm.

FIG. 6

Release (top) and relaxation rate (bottom) kinetics of VEGF/Gd-DTPA PLGA microspheres. In the top figure the amounts of VEGF and Gd-DTPA are expressed as the percentage of the total released. The bottom graph shows the relaxation rate of the microsphere pellet. The release takes place in three phases: an initial burst, followed by a slow steady-state, followed by a rapid steady-state. The relaxation rate of the microspheres reaches a maximum during the maximum rate of Gd-DTPA release and resembles the derivative of the release curve. Points on the graphs represent the mean of three samples and the bars represent the SD.

FIG. 7

Bioactivity measured by human umbilical vein cell (HUVEC) proliferation. Each bar represents the cell count at the end of 4 days of incubation with a particular media formation. VEGF = exogenously supplied VEGF (10 ng/ml), VEGF microspheres = VEGF released from VEGF/Gd-DTPA microspheres, empty = supernatant from empty (BSA/Gd-DTPA) microspheres, media = modified media without VEGF. The VEGF group served as a positive control and the empty and media groups served as negative controls. The counts for the VEGF microspheres group was not significantly different than the counts for the VEGF group, and were significantly (P ≤ 0.05) higher than the two negative controls. The data are expressed as mean ± SD of measurements made in triplicate.