HNature Biotechnology
18, 52 - 57 (2000)
doi:10.1038/71916
Stabilization of proteins encapsulated in injectable poly (lactide-
co-glycolide)Gaozhong Zhu1, 3, Susan R. Mallery2
& Steven P. Schwendeman11 Colleges of Pharmacy, The Ohio State University,
Columbus, OH 43210. 2 Colleges of Dentistry, The Ohio State University,
Columbus, OH 43210. 3 Current address: Department of Pharmaceutical Science
and Technology, Biogen Inc., 14 Cambridge Center,
Cambridge, MA 02142.
Correspondence should be addressed to Steven P. Schwendeman schwende@dendrite.pharmacy.ohio-state.eduControlled release from biodegradable polymers is a novel approach to replace
daily painful injections of protein drugs. A major obstacle to development
of these polymers is the need to retain the structure and biological activity
of encapsulated proteins during months of incubation under physiological conditions.
We encapsulated bovine serum albumin (BSA) in injectable poly(DL-lactide-
co-glycolide) (PLGA) 50/50 cylindrical implants and determined the mechanism
of BSA instability. Simulations of the polymer microclimate revealed that
moisture and acidic pH (<3) triggered unfolding of encapsulated BSA, resulting
in peptide bond hydrolysis and noncovalent aggregation. To neutralize the
acids liberated by the biodegradable lactic/glycolic acid-based polyester,
we coincorporated into the polymer an antacid, Mg(OH)2, which increased
microclimate pH and prevented BSA structural losses and aggregation for over
one month. We successfully applied this stabilization approach in both cylinder-
and microsphere-injectable configurations and for delivery of angiogenic basic
fibroblast growth factor and bone-regenerating bone morphogenetic protein-2.
protein deliveryaggregationstabilizationpoly(DL-lactide-co-glycolidebovine serum albuminSince protein delivery from polymers was first discovered1,
research has focused on developing polymer formulations widely applicable
for delivery of protein drugs. Efforts to this end have intensified recently
as hundreds of recombinant proteins are in the pipeline for US Food and Drug
Administration (Rockville, MD) approval2, and protein delivery
generally requires daily injections3. Frequent dosing is clinically
undesirable because of patient discomfort, psychological distress, and poor
compliance with self-injections. To reduce injection frequency, peptide and
protein drugs are encapsulated in biodegradable polymers, which are processed
into a form that is easily administered through a syringe needle. Current
preparations on the market for the delivery of small peptides can reduce injections
to once every one to three months depending on the polymer implant size4,
5,
6. During this incubation time, large globular proteins must
remain encapsulated at physiological temperature, which poses significant
challenges regarding retention of both the structural integrity and biological
activity of the protein7.
Two injectable polymer configurations are currently used to deliver peptides
and proteins: spherical particles approximately 1−100  m in diameter,
commonly referred to as "microspheres," and single cylindrical
implants approximately 0.8−1.5 mm in diameter, which we term "millicylinders."
Both configurations are prepared from the biocompatible polyesters formed
from lactic and glycolic acids (PLGAs) commonly used in resorbable sutures,
and each has distinct advantages and disadvantages8. Several
potential sources of irreversible inactivation of proteins encapsulated in
PLGAs have been identified7. These include: (1) elevated levels
of moisture, providing sufficient protein mobility for reactivity; (2) an
acidic microclimate induced by acidic degradation products and carboxylic
acid end groups of PLGA; and (3) adsorption of the protein to the polymer
surface, which may catalyze protein unfolding and aggregation.
Few have been successful in stabilizing proteins encapsulated in PLGAs9,
10,
11,
12. We report the mechanism of instability of a model protein,
bovine serum albumin (BSA), in PLGA millicylinders. This mechanistic analysis
suggested a general approach to stabilize proteins encapsulated in PLGA, which
was tested and found to work for BSA, basic fibroblast growth factor (bFGF),
and bone morphogenetic protein-2 (BMP-2), and for both millicylinder and microsphere
geometries.
Results and discussion
Irreversible instability of BSA in PLGA.
To investigate
BSA stability in the polymer, we encapsulated BSA in PLGA 50/50 millicylinders
(15% BSA wt/wt). Then, we monitored the release of BSA in a phosphate buffer
and the simultaneous loss of soluble BSA encapsulated in the polymer due to
aggregation. Bovine serum albumin release was incomplete (only 20%) with a
corresponding growth of insoluble aggregates in the polymer (
Fig. 1A and B, circles). By 28 days, most initially encapsulated BSA
( 80%) had become insoluble and unreleasable from the polymer. This insoluble
fraction was 98% soluble in 6 M urea (Table 1), indicating
that the aggregates were noncovalent instead of common disulfide-bonded aggregates,
which require reducing agent to dissolve13. The soluble BSA
recovered from the polymer (28 days of incubation) was examined by SDS−PAGE.
Several peptide fragments (e.g., 55, 40, and 25 kDa) accompanied the dimeric
and trimeric BSA (Fig. 1D), the precursors of insoluble
aggregates. Therefore, the analysis revealed two salient features of the instability
of encapsulated BSA: the formation of insoluble, noncovalent aggregates by
hydrophobic interactions and peptide bond hydrolysis.
 | | |
| | Table 1. Comparison of BSA instability under simulated and encapsulated conditions.
| | |  |
 Full Table |
| |
Simulating BSA instability in the polymer.
Primary
deleterious characteristics in the polymer microclimate are expected to consist
of one or more of the following: elevated moisture, acidic pH, and the polymer
surface7. To test this hypothesis, we simulated BSA instability
in the PLGA microclimate. If one or more of the chosen conditions adequately
describe the microclimate, we would expect to create a denatured state of
BSA similar to that observed for encapsulated BSA with instability occurring
over similar time scales.
Polymer acidity was simulated by exposing BSA, which had been lyophilized
from pH 2-5 solutions, to a humid environment (86% relative humidity [RH]
and 37°C) for one week. Of the conditions tested, only the highly acidic
simulation (pH 2) induced aggregation (43  6% insoluble BSA,
n = 3). For pH  3, no aggregation was observed (data not shown).
To probe for possible similarities with aggregation in the polymer, the kinetics
and mechanism of BSA aggregation under the pH 2 condition were examined. Acid-induced
aggregates of BSA formed over 10 days (Fig. 1C) and
were soluble in denaturing solvent (Table 1). Moreover,
the denatured BSA produced in the simulation and in the polymer microclimate
exhibited nearly identical peptide bond fragmentation, as shown by SDS−PAGE
(Fig. 1D). Hence, incubation of moist protein (lyophilized
at pH 2) produced a denatured state in BSA that was identical to that when
encapsulated (Table 1). The aggregate type (noncovalent),
peptide bond fragments (e.g., 25, 40, 55 kDa), and aggregation time scale
(e.g., 10 days) were equivalent.
Moisture affected the aggregation rate but not the mechanism. For example,
if little water was added to the acidic BSA powder, variable aggregation levels
were observed. For water:protein weight ratios of 0-0.2, no aggregates were
observed after one week at 37°C; for a 0.8-1.0 ratio, aggregates were
produced (about 70% insoluble BSA), which were >94% soluble in 6 M urea (data
not shown). To simulate BSA adsorption, BSA solutions (pH 2-7) were incubated
at 37°C for one week with blank PLGA microspheres or fine polymer powder
(as obtained from the manufacturer) to provide a polymer surface. Negligible
adsorption (<2%) was recorded in any of these simulations (data not shown).
Developing a hypothesis for the mechanism of instability of encapsulated
BSA.
Acidity commonly develops in PLGAs14,
15,
16,
17,
18,
19
because of accumulation of acidic degradation products upon polyester hydrolysis.
A highly acidic pH (<3) was also required for noncovalent aggregate formation
in our polymer microclimate simulations. Thus, the added 15% BSA did not significantly
alter the acidic pH (e.g., by creating water channels in the polymer to allow
buffer ions to diffuse in or by buffering pH itself). Instead, the encapsulated
BSA was rehydrated at a pH that caused the protein to unfold (BSA undergoes
a conformational transition from the F to E isoform at pH 2.7
[20]), providing the driving force for noncovalent
aggregation via hydrophobic interactions. Peptide bond hydrolysis is also
particularly rapid at acidic pH or when labile peptide bonds (e.g., Asp-Xaa
linkage21) become exposed during unfolding.
Developing a rational stabilization approach using Mg(OH)2
.
Our mechanistic observations suggested a rational approach
to stabilize encapsulated BSA. As BSA aggregation did not occur in microclimate
simulations at pH3, and peptide bond hydrolysis would be inhibited at
higher pH, we tried an established method of increasing the microclimate pH
in poly(ortho esters), a polymer cousin of PLGA. To stabilize the
ortho ester bond and to extend the release time of drugs, Mg(OH)
2 is routinely incorporated into poly(ortho esters)22.
Consequently, to test whether Mg(OH)2 could inhibit BSA instability,
we coencapsulated the salt with BSA at 0.5% and 3% by weight. Then, we compared
release and aggregation kinetics of these samples with those observed for
the 0% base control. As predicted, as more Mg(OH)2 was added, more
BSA was released and less encapsulated BSA became insoluble (
Fig. 1A and B). For the 3% Mg(OH)2 preparation, the aggregation
was virtually eliminated.
To examine hydrolysis extent and any other potential structural alterations
of BSA, SDS−PAGE, isoelectric focusing (IEF), circular dichroism (CD),
and fluorescence spectroscopy were employed to characterize the encapsulated
and released BSA from the Mg(OH)2/PLGA (Fig. 2).
The large degree of fragmentation that occurred in the absence of Mg(OH)
2 had largely disappeared (Fig. 2A). Some faint
low-molecular-weight bands were noticeable, particularly in the residual encapsulated
BSA after 28 days, but these were small relative to the monomeric band.
| | Figure 2. Structural analyses of BSA from incubated PLGA millicylinders by SDS−PAGE
(A), IEF (B), CD spectra (C), and fluorescence emission spectra (D). | | |  | In SDS−PAGE and IEF analyses, lanes 1−5 are standard BSA, released
BSA on day 1, days 5−7, and days 21−28, and residual BSA on day
28, respectively. In CD and fluorescence emission spectra, samples include:
standard BSA (—), BSA heat denatured in PBST (90°C for 30 min) (---),
released BSA samples from day 1 (— —), days 5−7 (−
− −), days 16−20 (---), days 21−28 (......),
and residual BSA (−.−.). The IEF of heat denatured BSA was distinguishable
from standard BSA (data not shown). Samples for fluorescence were 50 g/ml.
Full Figure and legend (32K) |
| |
To test for the most common route of protein degradation—deamidation—the
released and encapsulated BSA was examined by IEF. No alteration of the protein's
pI (4.7) was detected (Fig. 2B). Likewise, no alterations
in secondary or tertiary structure of BSA were noticeable in the CD (Fig. 2C) and fluorescence spectra (Fig. 2D
). Hence, the structure of BSA was retained (>90%) from the Mg(OH)
2/PLGA preparation for a period of one month.
To confirm that the Mg(OH)2 actually neutralized the acidic
microclimate in the polymer, we examined changes in polymer erosion of BSA-loaded
millicylinders on addition of Mg(OH)2. The following alterations
in polymer erosion would indicate that the salt induced a microclimate pH
increase: (1) a polymer water content increase due to increased osmotic pressure
generated by magnesium carboxylate salts and ionization of polymer end groups
(2) a polymer degradation half-time increase resulting from inhibition of
acid-catalyzed polyester hydrolysis23, and (3) an inhibited
release of sequestered acidic moieties (high-molecular-weight PLGA releases
degradation products into the erosion medium only after the molecular weight
declines to a critical value24). As seen in
Table 2, each trend was observed. When the base was added to 15% BSA/PLGA
millicylinders, polymer water content doubled, degradation half-time was extended
from 16 days to 25 days, and fewer acidic species were released (phosphate-buffered
saline, 0.02% Tween 80 [PBST] pH 7.4 dropped to pH 7.0 instead of 3.5 after
four weeks).
| | Table 2. Neutralization effect of Mg(OH)2 on the erosion behavior
of 15% BSA/PLGA millicylinders. | | | |
Full Table |
| |
The basicity of the encapsulated salt also affected microclimate pH. For
example, when a stronger base, Ca(OH)2, was substituted for Mg(OH)
2, after two weeks incubation disulfide-bonded aggregates of BSA began
to form (means.e.m.: 111% disulfide bonded; 3.90.1%
noncovalent bonded for 3% Ca[OH]2; n=3).
The formation of disulfide bonds requires a free thiolate, revealing a neutral
to basic microclimate pH. When a weaker base, ZnCO3, was used,
the noncovalent aggregation increased, suggesting a lower microclimate pH
(after two weeks, 101% noncovalent aggregates for 3% ZnCO3;
2.00.4% for 3% Mg[OH]2) with negligible disulfide-bonded
aggregation. (A weight basis was used instead of a molar basis for the salt
comparison, because pores in the polymer are saturated with salt and the weight
basis controls percolation of the salt throughout the polymer.) Finally, coencapsulation
of a neutral salt (e.g., NaCl) was also unsuccessful in preventing BSA aggregation
(after two weeks, 241% noncovalent aggregates for 3% NaCl), and any
partial stabilization was consistent with polymer water content increases
induced by the salt.
Generality of the use of basic salts to improve stability of proteins
encapsulated in PLGA.
To test the generality of our stabilization
approach, we encapsulated two growth factors, recombinant human bFGF and BMP-2,
in PLGA millicylinders, and formulated BSA in PLGA microspheres. Basic fibroblast
growth factor is currently being tested for wound healing, osteogenesis, and
diabetic ulcers25. Like most therapeutic proteins, bFGF's in
vivo serum half-life is very short (<3 min)26, making it
a good candidate for controlled release27,
28,
29,
30. Another
important application is to inject BMP-2 with a carrier matrix to induce bone
regeneration, which is a promising alternative to bone grafting31,
32.
PLGA is now considered a good candidate to overcome the difficulties with
conventional BMP-2 carriers, including inactivated collagenous bone matrix33,
34. Moreover, in contrast to millicylinders, microspheres have
several advantages, particularly to reduce pain of injections and to simplify
administration8. PLGA 50/50 microspheres also frequently develop
an acidic microclimate, according to recent findings15,
16,
17,
18.
Before encapsulating bFGF, we considered several instability pathways characteristic
of this protein. Basic fibroblast growth factor binds heparin, and when unbound,
loses activity rapidly35,
36. It adheres avidly to glass and
plastic surfaces29 and disulfide exchange of bFGF is catalyzed
by trace metals (e.g., those remaining from the polymerization of polymer7). Therefore, we selected five additives for encapsulation of bFGF.
The 3% Mg(OH)2/15% BSA combination was suitable for neutralizing
the acidic microclimate. High BSA concentrations may also inhibit adsorption
of bFGF to PLGA29. We added heparin at a weight ratio of 1:1
(heparin to bFGF) to enhance bFGF stability36, and EDTA to chelate
trace heavy metals. Sucrose was used to retain bFGF structure in the solid
state25.
As predicted, when bFGF was encapsulated (0.0025%) in the Mg(OH)
2/BSA/PLGA millicylinders, the growth factor was released (
Fig. 3A, filled circles) in a fashion similar to that observed for
BSA (Fig. 1A). Over 28 days, 71% of immunoreactive bFGF
was detected in the release medium and 21% remained in the polymer (Table 3), accounting for approximately 92% of initially encapsulated
bFGF. If millicylinders did not contain both heparin and Mg(OH)2/BSA,
bFGF lost immunoreactivity. For example, when heparin was removed from the
stabilized formulation, only 2% bFGF was released over one month with no immunoreactive
bFGF in the residual fraction (Table 3). Similarly,
when 20% arabic gum was substituted for 3% Mg(OH)2/15% BSA, no
bFGF was observed in the release medium after four days and only 38% was accounted
for in both the release and residual fraction. (A 0% Mg[OH]2 and
15% BSA control was not performed because of BSA aggregation.)
| | |
| | Table 3. Generality of the stabilization effect of basic salts for protein delivery
from PLGA. | | | |
Full Table |
| |
To increase the capacity of the polymer to deliver bFGF, we increased bFGF
loading to 0.01% and sucrose loading to 2.3%. The bFGF release initially was
much slower and later exhibited a linear release profile up to four weeks
(Fig. 3A, filled squares). The release of BSA from the
same preparation in PBST was retarded similarly (data not shown). Thus, sucrose
can be used to slow down release of both BSA and bFGF from the polymer, probably
by increasing viscosity of aqueous pores in the polymer.
Although >90% recovery by ELISA during release indicates retention of bFGF
immunoreactivity (Table 3), this fraction is not necessarily
bioactive. To test this, we examined whether encapsulated and released growth
factor induced cell proliferation (as indicated by [3H]thymidine
incorporation37). The encapsulation procedure did not affect
the biological activity of bFGF (time 0, Fig. 3B). Some
small inactivation apparently occurred during release, but 65−85% bioactivity
of bFGF was retained over the entire one-month release experiment.
To similarly stabilize encapsulated BMP-2, we again used 3% Mg(OH)
2 and 15% BSA to neutralize microclimate acidity in PLGA millicylinders
containing BMP-2 (0.25% loading). No other additives were used. We also included
a second positive control using the protein substitute, gum arabic (i.e.,
3% Mg[OH]2, 15% gum arabic, 0.25% BMP-2). In both cases, controlled
release was observed over 28 days (data not shown) resulting in a recovery
(released + soluble residue fraction) of >80% (Table 3).
In contrast, when base was removed from the formulation (18% gum arabic),
only 30% protein was recovered by immunoassay.
To test our approach in microspheres, we examined whether encapsulated
BSA undergoes the acid-induced mechanism of instability and if so, whether
basic additives inhibit the mechanism. As expected, BSA also forms noncovalent
aggregates (25−70%) when encapsulated in PLGA microspheres (Table 3), confirming that an acidic microclimate also commonly
develops in PLGA 50/50 microspheres. It has been suggested that BSA becomes
unstable in PLGA microspheres primarily by protein adsorption to the polymer38. This conclusion was strongly weighed on the SDS-induced liberation
of previously unreleasable BSA from the polymer. We found that the SDS buffer
used in the SDS−PAGE (Fig. 1D) dissolves encapsulated
noncovalent aggregates, which may explain the reported release of sequestered
BSA from the polymer. Therefore, we conclude that protein adsorption, consistent
with our simulations, is not the predominant source of BSA instability in
PLGA microspheres.
Whereas the BSA instability mechanism in microspheres was similar to that
observed in millicylinders, coencapsulation of Mg(OH)2 was only
moderately successful to inhibit BSA aggregation in microspheres. For example,
the soluble fraction of BSA in PLGA microspheres (0.64 dl/g) decreased from
68% without base to 26% with Mg(OH)2 (Table 3).
This modest BSA stability increase afforded by Mg(OH)2, and our
previous studies illustrating a heterogeneous pH distribution in Mg(OH)
2/PLGA microspheres (no protein)15, suggest that the basic
additive could not diffuse to all acidic protein pores in the polymer. To
overcome this problem, we turned to another base, MgCO3, which
has basicity equivalent to that of Mg(OH)2, but which is about
10-fold more water soluble, to facilitate base diffusion. The more soluble
salt inhibited BSA aggregation at a level similar to the inhibition attained
in millicylinders with Mg(OH)2. For the medium-molecular-weight
PLGA (0.64 dl/g), aggregation was held to just 13% over 28 days with 89% recovery
(Table 3). Remarkably, coencapsulation of MgCO
3 in low-molecular-weight PLGA (0.20 dl/g) resulted in a reduction of
BSA aggregation to just 1.5% over 51 days with 94% recovery (
Table 3). This latter preparation controlled release of BSA slowly
and continuously over the entire experiment after a 32% burst (data not shown).
We conclude that the acidic microclimate in PLGA delivery systems (including
microspheres) is a common source of instability of encapsulated proteins.
Poorly water-soluble basic salts can be used to neutralize the polymer microclimate
pH to levels necessary to retain the structure and biological activity of
encapsulated acid-labile proteins.
Experimental protocol
Reagents.
Poly(DL-lactide-co-glycolide) 50/50
(inherent viscosity of 0.20, 0.63, and 0.64 dl/g in hexafluoroisopropanol
solution) was from Birmingham Polymers (Birmingham, AL). Recombinant human
bFGF and BMP-2 were supplied by Scios (Sunnyvale, CA) and Orthogene (Fremont,
CA), respectively. Bovine serum albumin (A-3059), heparin (H-3393), and gum
arabic were from Sigma (St. Louis, MO). Fine Mg(OH)2, MgCO
3, and Ca(OH)2 (<5 m) powders were from Aldrich (Milwaukee,
WI) and ZnCO3 (<5 m) was from ICN Biomedical (Aurora, OH).
Cell culture reagents were from GIBCO Life Technologies (Rockville, MD). All
other chemicals were of analytical grade or purer and purchased from commercial
suppliers.
PLGA millicylinder preparation.
A suspension of sieved
BSA (<90 m) with or without basic salt in acetone-PLGA (0.63 dl/g)
solution (50% wt/wt) was loaded in a syringe and extruded into silicone rubber
tubing (0.8 mm i.d.) at approximately 0.1 ml/min (39). The solvent-extruded suspension was dried at room temperature
(24 h) and then under vacuum at 45°C (24 h). For encapsulation of bFGF
and BMP-2, each growth factor was combined with additives in 10 mM sodium
phosphate buffer (pH 7.4), lyophilized to a fine powder (4% moisture
determined by Fisher titration), and sieved before extrusion and drying. All
preparations had a loading efficiency invariably between 85% and 95%.
PLGA microsphere preparation.
According to a solvent
evaporation method40, 100 l of 150 mg/ml BSA in 10 mM phosphate
buffer (pH 7.4) were added to 1 ml of PLGA/CH2Cl2 solution
(30% for 0.64 dl/g; 70% for 0.2 dl/g) with or without basic salt. The mixture
was homogenized at 10,000 r.p.m. (1 min at 4°C), and transferred to a
2% polyvinyl alcohol (PVA) aqueous solution. The water-in-oil-in-water emulsion
was formed by vortexing the mixture for 20 s, and the particles were hardened
for 3 h in 100 ml of 0.5% PVA at room temperature. Microspheres were collected
by centrifugation, washed with water, and lyophilized. Microspheres from both
preparations were spherical with a mean diameter between 60 and 70 m.
The BSA loading was approximately 4% with an encapsulation efficiency between
70% and 80%.
Controlled release of proteins.
Bovine serum albumin
release was monitored in PBST. Millicylinders (10 mg) or microspheres
(20 mg) were placed in release medium (0.5 ml) and incubated under mild
agitation (37°C). To sample, the buffer was removed (by centrifugation
for microspheres) and replaced with new medium. Protein content in release
samples was determined by using a modified Bradford assay (Coomassie brilliant
blue plus protein assay, Pierce, Rockford, IL), which is also compatible with
denaturing/reducing agents used below. The release of bFGF and BMP-2 millicylinders
was examined similarly except that 1% BSA, 10 g/ml heparin, and 1 mM EDTA
were added to the release medium to prevent protein damage once released from
the polymer.
Evaluation of BSA aggregation.
Incubated polymers were
removed from release medium, dried, and dissolved in acetone. After centrifugation
and removal of the polymer solution, the remaining BSA pellet was reconstituted
in PBST and incubated (37°C) overnight before determining the protein
content; this gave a measure of the water-soluble protein encapsulated (also
used for protein loading measurement). Any aggregate was collected by centrifugation
and incubated (37°C for 30 min) in denaturing solvent (PBST, 6 M urea,
1 mM EDTA); analysis of protein concentration gave the amount of noncovalently
bonded BSA aggregates41. The same procedure was repeated with
reducing solvent (10 mM dithiothreitol in denaturing solvent) to determine
the amount of disulfide-bonded aggregates.
Simulations of BSA instability in the polymer microclimate.
For simulating pH, BSA (4 mg/ml) in a universal buffer (H3PO
4, HAc, and H3BO3; 40 mM each, titrated with NaOH)
was lyophilized from pH 2−5 and incubated (37°C and 86% RH)41. To sample, protein was reconstituted in PBST and examined for
the aggregate type. For simulating water content, water was added directly
to lyophilized BSA (from pH 2), sealed, and incubated (37°C for one week)42. To test protein adsorption, BSA (1 mg/ml) in universal buffer
(pH 2−7) was incubated for one week (37°C) with 20 mg PLGA powder
(0.20 dl/g, <100 m) or PLGA (0.63 dl/g) microspheres prepared by solvent
evaporation43. Losses of BSA from solution were used to determine
adsorption extent.
Structural analysis of encapsulated BSA.
Fluorescence
emission spectra of BSA (300−500 nm; 240 nm/min) were obtained with
a Perkin-Elmer (Norwalk, CT) LS50B luminescence spectrometer. Far ultraviolet
CD spectra (200−250 nm) were recorded with a J-500A Jasco (Tokyo, Japan)
spectropolarimeter at room temperature. The integrity of protein samples was
determined by both SDS−PAGE and IEF gel electrophoresis, which were
performed on a PhastSystem (Amersham Pharmacia Biotech AB, Uppsala, Sweden).
PhastGel gradient 10−15 gels and IEF 3−9 gels (Amersham Pharmacia
Biotech AB) were used for SDS−PAGE and IEF analyses, respectively. Coomassie
brilliant blue staining was performed after separation.
ELISA for bFGF.
A 96-well plate was coated with monoclonal
anti-bFGF (Upstate Biotechnology, Waltham, MA) overnight at 4°C (44 and 45). Samples
or standards of bFGF in release media were added to each well and incubated
at 4°C for 24 h. After washing, polyclonal rabbit anti-bFGF (Sigma) was
added and left at room temperature for 2 h, followed by 2 h with anti-rabbit
IgG−horseradish peroxidase (1:10,000, Sigma). The substrate o-phenylenediamine
with H2O2 (Sigma Fast OPD tablet sets) was added (30
min at room temperature) and the reaction was stopped with 3 M H2SO
4. The product was detected (490 nm) on a plate reader (Dynatech Laboratories,
Chantilly, VA).
BIAcore immunoassay for BMP-2.
BMP-2 was quantified
with a BIAcore 2000 biosensor (Biacore AB, Uppsala, Sweden). A monoclonal
antibody of BMP-2 (Orthogene) was immobilized onto a CM-5 sensor chip using
the amine coupling kit (Biacore AB). Bone morphogenetic protein-2 in release
medium was assayed over the immobilized antibody surface. Sample volume was
30 l and HEPES-buffered saline was the mobile phase (10 l/min). The
surface was regenerated (10 l 10 mM HCl) after each sample. The antibody
was stable on the chip surface for more than one month (standard curve range,
50-1,600 ng/ml BMP-2).
Cell proliferation assay for bFGF.
BALB/c 3T3 fibroblasts
(25,000/well, CCL-163, American Type Culture Collection) in Dulbecco's modified
Eagle's medium containing 10% bovine calf serum, 50 U/ml streptomycin, and
50 g/ml penicillin were seeded (200 l/well) on 96-well plates37 and grown to confluence (one week) without changing the medium.
Then, bFGF samples or standards (10 l) in release medium were added (20
h) followed by 1 Ci of [3H]thymidine (6.7 Ci/mmol, DuPont/NEN
Research Products, Boston, MA) per well (6−8 h). Cells were collected
on filter paper by using a PHD cell harvester (Cambridge Technology, Cambridge,
MA), resuspended in 3 ml scintillation cocktail 3a70B (Research Products International
Corporation, Mount Prospect, IL), and counted (Beckman scintillation counter,
Fullerton, CA).
Received 14 April 1999; Accepted 22 September 1999
REFERENCES
- Langer, R. & Folkman, J. Polymers for the sustained release of proteins and other macromolecules. Nature 263, 797−800 (1976). | PubMed | ISI | ChemPort |
- Struck, M.M. Biopharmaceutical R&D success rates and development times. A new analysis provides benchmarks for the future. Bio/Technology 12, 674−677 (1994). | Article | PubMed | ISI | ChemPort |
- Talmadge, J.E. The pharmaceutics and delivery of therapeutic polypeptides and proteins. Adv. Drug Delivery. Rev. 10, 247−299 (1993). | Article | ISI | ChemPort |
- Ogawa, Y., Okada, H., Heya, T. & Shimamota T. Controlled release of LHRH agonist, leuprolide acetate, from microcapsules: serum drug level profiles and pharmacological effects in animals. J. Pharm. Pharmcol. 41, 439−444 (1989). | ChemPort |
- Okada, H., Doken, Y., Ogawa, Y. & Toguchi, H. Preparation of three-month depot injectable microspheres of leuprorelin acetate using biodegradable polymers. Pharm. Res. 11, 1143 (1994). | Article | PubMed | ISI | ChemPort |
- Dutta, A.S., Furr, B.J.A. & Hutchinson, F.G. Discovery and development of goserelin (Zoladex). Pharm. Med. 7, 9−28 (1993).
- Schwendeman, S.P., Cardamone, M., Klibanov., A. & Langer, R. in Microparticulate systems for the delivery of proteins and vaccines (eds Cohen, S. & Bernstein, H.), 1−49 (Marcel Dekker, New York; 1996).
- Schwendeman, S.P., Costantino, H.R., Gupta, R.K. & Langer, R. in Controlled drug delivery. (ed. Park, K.), 229−267 (American Chemical Society, Washington, D.C; 1997).
- Johnson, O.L. et al. A month-long effect from a single injection of microencapsulated human growth hormone. Nat. Med. 2, 795−799 (1996). | Article | PubMed | ISI | ChemPort |
- Cleland, J.L. & Jones, A.J.S. Stable formulations of recombinant human growth hormone and interferon-gamma for microencapsulation in biodegradable microspheres. Pharm. Res. 13, 1464−1475 (1996). | Article | PubMed | ISI | ChemPort |
- Schwendeman, S.P., Tobio, M., Joworowicz, M., Alonso, M.J. & Langer, R. New strategies for the microencapsulation of tetanus vaccine. J. Microencapsulation 15, 299−318 (1998). | PubMed | ISI | ChemPort |
- Putney, S.D. & Burke, P.A. Improving protein therapeutics with sustained-release formulations. Nat. Biotechnol. 16, 153−157 (1998). | Article | PubMed | ISI | ChemPort |
- Costantino, H.R., Langer, R. & Klibanov, A.M. Solid-phase aggregation of proteins under pharmaceutically relevant conditions. J. Pharm. Sci. 83, 1662−1669 (1994). | PubMed | ISI | ChemPort |
- Herrlinger, M. Ph.D. In vitro polymerabbau und wirksttofffreigabe von poly-DL-laktidformlingen Thesis, Univ. Heidelberg, Germany (1994).
- Shenderova, A., Burke, T. & Schwendeman, S.P. The acidic microclimate in poly(lactide-co-glycolide) microspheres stabilizes camptothecins. Pharm. Res. 16, 241−248 (1999). | Article | PubMed | ISI | ChemPort |
- Brunner, A., Mäder, K. & Göpferich, A. pH and osmotic pressure inside biodegradable microspheres during erosion. Pharm. Res. 16, 847−853 (1999). | Article | PubMed | ISI | ChemPort |
- Mader, K., Bittner, B., Li, Y., Wohlauf, W. & Kissel, T. Monitoring microviscosity and microacidity of the albumin microenvironment inside degrading microparticles from poly(lactide-co-glycolide) (PLG) or ABA-triblock polymers containing hydrophobic poly(lactide-co-glycolide) A blocks and hydrophilic poly(ethyleneoxide) B blocks. Pharm. Res. 15, 787−793 (1998). | Article | PubMed | ISI | ChemPort |
- Fu, K., Pack, D.W., Laverdiere, A., Son, S. & Langer, R. Visualization of pH in degrading polymer microspheres. Proceedings 25th International Symposium on Controlled Resealse of Bioactive Materials 25, 150−151 (1998).
- Mader, K., Gallez, B., Liu, K.J. & Swartz, H.M. Non-invasive in vivo characterization of release processes in biodegradable polymers by low-frequency electron paramagnetic resonance spectroscopy. Biomaterials 17, 457−461 (1996). | Article | PubMed | ISI | ChemPort |
- Peters, T. Jr. Serum albumin. Adv. Protein Chem. 37, 161−245 (1985). | PubMed | ISI | ChemPort |
- Manning, M.C., Patella, L. & Borchardt, R.T. Stability of protein pharmaceuticals. Pharm. Res. 6, 903−917 (1989). | Article | PubMed | ISI | ChemPort |
- Heller, J. Development of poly(ortho esters): a historical overview. Biomaterials 11, 659−665 (1990). | Article | PubMed | ISI | ChemPort |
- Li, S.M., Garreau, H. & Vert, M. Structure-property relationships in the case of the degradation of massive poly(a-hydroxy acids) in aqueous media. Part 3. Influence of the morphology of poly(L-lactic acid). J. Mater. Sci. 1, 123−130 (1990). | ISI | ChemPort |
- Hutchinson, F.G. & Furr, B.J.A. Biodegradable polymer systems for the sustained release of polypeptides. J. Controlled Release 13, 279−294 (1990). | Article | ISI | ChemPort |
- Wang, Y.J. et al. in Formulation, characterization, and stability of protein drugs (eds Pearlman, R. & Wang, Y.J.) 141−180 (Plenum Press, New York; 1996).
- Edelman, E.R., Nugent, M.A. & Karnovsky, M.J. Perivascular and intravenous administration of basic fibroblast growth factor: vascular and solid organ deposition. Proc. Natl. Acad. Sci. USA 90, 1513−1517 (1993). | PubMed | ChemPort |
- Berscht, P.C., Nies, B., Liebendorfer, A. & Kreuter, J. Incorporation of basic fibroblast growth factor into methylpyrrolidinone chitosan fleeces and determination of the in vitro release characteristics. Biomaterials 15, 593−600 (1994). | Article | PubMed | ISI | ChemPort |
- Fukunaga, K. et al. Aluminium beta-cyclodextrin sulphate a stabilizer and sustained-release carrier for basic fibroblast growth factor. J. Pharm. Pharmacol. 46, 168−171 (1994). | PubMed | ISI | ChemPort |
- Edelman, E.R., Mathiowitz, E., Langer, R. & Klagsbrun, M. Controlled and modulated release of basic fibroblast growth factor. Biomaterials 12, 619−626 (1991). | Article | PubMed | ISI | ChemPort |
- Davies, M.J. et al. In vitro assessment of the biological activity of basic fibroblast growth factor released from various polymers and biomatrices. Journal of Biomaterial Applications 12, 31−56 (1997). | ISI | ChemPort |
- Wang, E.A. et al. Recombinant human bone morphogenetic protein induces bone formation. Proc. Natl. Acad. Sci. USA 87, 2220−2224 (1990). | PubMed | ChemPort |
- Welch, R.D. et al. Effect of recombinant human bone morphogenetic protein-2 on fracture healing in a goat tibial fracture model. J. Bone Miner. Res. 13, 1483−1490 (1998). | PubMed | ISI | ChemPort |
- Mayer, M., Hollinger, J., Ron, E. & Wozney, J. Maxillary alveolar cleft repair in dogs using recombinant human bone morphogenetic protein-2 and a polymer carrier. Plast. Reconstr. Surg. 98, 247−259 (1996). | Article | PubMed | ISI | ChemPort |
- Kenley, R.A. et al. Biotechnology and bone graft substitutes. Pharm. Res. 10, 1393−1401 (1993). | Article | PubMed | ISI | ChemPort |
- Gospodarowicz, D. & Cheng, J. Heparin protects basic and acidic FGF from inactivation. J. Cell. Physiol. 128, 475−484 (1986). | PubMed | ISI | ChemPort |
- Sommer, A. & Rifkin, D.B. Interaction of heparin with human basic fibroblast growth factor: protection of the angiogenic protein from proteolytic degradation by a glycosaminoglycan. J. Cell. Physiol. 138, 215−220 (1989). | PubMed | ISI | ChemPort |
- Sullivan, R. & Klagsbrun, M. Purification and assay of intact human basic fibroblast growth factor using heparin-sepharose chromatography. J. Tiss. Culture Methods 10, 125−132 (1986). | ChemPort |
- Crotts, G., Sah, H. & Park, T.G. Adsorption determines in-vitro protein release rate from biodegradable microspheres: quantitative analysis of surface area during degradation. J. Controlled Release. 47, 101−111 (1997). | Article | ISI | ChemPort |
- Zhang, X., Wyss, U.P, Amsden, B. & Goosen, M.F.A. Controlled release of albumin from biodegradable poly(DL-lactide) cylinders. J. Controlled Release 25, 61−69 (1993). | Article | ISI | ChemPort |
- Cohen, S., Yoshioka, T., Lucarelli, M., Hwang, L.H. & Langer, R. Controlled delivery systems for proteins based on poly(lactic/glycolic acid) microspheres. Pharm. Res. 8, 713−720 (1991). | Article | PubMed | ISI | ChemPort |
- Schwendeman, S.P. et al. Stabilization of tetanus and diphtheria toxoids against moisture-induced aggregation. Proc. Natl. Acad. Sci. USA 92, 11234−11238 (1995). | PubMed | ChemPort |
- Liu, W.R., Langer, R. & Klibanov, A.M. Moisture-induced aggregation of lyophilized proteins in the solid state. Biotechnol. Bioeng. 37, 177−184 (1991). | Article | ISI | ChemPort |
- Shenderova, A., Burke, T.G. & Schwendeman, S.P. Stabilization of 10-hydroxycamptothecin in poly(lactide-co-glycolide) microsphere delivery vehicles. Pharm. Res. 14, 1406−1413 (1997). | Article | PubMed | ISI | ChemPort |
- Watanabe, H. et al. A sensitive enzyme immunoassay for human basic fibroblast growth factor. Biochem. Biophys. Res. Commun. 175, 229−235 (1991). | PubMed | ISI | ChemPort |
- Gabra, N., Khiat, A. & Calabresi, P. Detection of elevated basic fibroblast growth factor during early hours of in vitro angiogenesis using a fast ELISA immunoassay. Biochem. Biophys. Res. Commun. 205, 1423−1430 (1994). | Article | PubMed | ISI | ChemPort |
Acknowledgments
We thank Dr. John Wang of Synzyme Technologies (Irvine, CA) for his helpful
suggestions. We are very grateful to Prof. Mark Coggeshall, from the Department
of Microbiology of the Ohio State University for his help in the bFGF bioassay
experiment. We thank Dr. Hanne Bentz from Orthogene and Prof. Jeffrey Hubbell
from the Swiss Federal Technology Institute for their aid in the BMP-2 studies.
We also thank Professor Gary Means from the Department of Biochemistry, and
Dr. Jichao Kang of the College of Pharmacy at the Ohio State University for
their helpful discussions and technical assistance, respectively. This work
was supported by NIH DE 12183 and a PhRMA grant to S.P.S.
|
), 0.5% (
),
and 3.0% (
) Mg(OH)2. (mean
100%.
