Journal of Controlled Release 64 (2000) 155166

www.elsevier.com / locate / jconrel

DepoFoamE technology: a vehicle for controlled delivery of

protein and peptide drugs
Qiang Ye, John Asherman, Mark Stevenson, Elizabeth Brownson,
Nandini V. Katre*
DepoTech Corporation 1 , 10450 Science Center Drive, San Diego, CA 92121, USA
Accepted 15 April 1999

Abstract
A major challenge in the development of sustained-release formulations for protein and peptide drugs is to achieve high
drug loading sufficient for prolonged therapeutic effect coupled with a high recovery of the protein / peptide. This challenge
has been successfully met in the formulation of several peptide and protein drugs using the DepoFoamE, multivesicular
lipid-based drug delivery system. DepoFoam technology consists of novel multivesicular liposomes characterized by their
unique structure of multiple non-concentric aqueous chambers surrounded by a network of lipid membranes. The objective
of this paper is to demonstrate that DepoFoam technology can be used to develop sustained-release formulations of
therapeutic proteins and peptides with high loading. DepoFoam formulations of a protein such as insulin, and peptides such
as leuprolide, enkephalin and octreotide have been developed and characterized. The data show that these formulations have
high drug loading, high encapsulation efficiency, low content of free drug in the suspension, little chemical change in the
drug caused by the formulation process, narrow particle size distribution, and spherical particle morphology. Drug release
assays conducted in vitro in biological suspending media such as human plasma indicate that these formulations provide
sustained release of encapsulated drug over a period from a few days to several weeks, and that the rate of release can be
modulated. In vivo pharmacodynamic studies in rats also show a sustained therapeutic effect over a prolonged period. These
results demonstrate that the DepoFoam system is capable of efficiently encapsulating therapeutic proteins and peptides and
effectively providing controlled delivery of these biologically active macromolecules. 2000 Elsevier Science B.V. All
rights reserved.
Keywords: Multivesicular liposome; Sustained delivery; High loading liposome; Protein and peptide encapsulation; Controlled release

1. Introduction
Therapeutic proteins and peptides administered
intravenously or subcutaneously are often cleared
*Corresponding author. Tel.: 11-858-625-2424; fax: 11-858625-2439.
E-mail address: nandini katre@skyepharma.com (N.V. Katre)
]
1
Presently SkyePharma, Inc.

rapidly from the circulation, and therefore need to be

injected frequently in order to maintain therapeutic
levels in the blood. Various types of liposomal
formulations have been utilized as drug delivery
vehicles for sustained release of proteins and peptides [14], and some have been evaluated for
clinical applications. Most of the reported studies
involve traditional liposomes, such as unilamellar
(ULV) [57] or multilamellar vesicular (MLV) [8,9]

0168-3659 / 00 / $ see front matter 2000 Elsevier Science B.V. All rights reserved.
PII: S0168-3659( 99 )00146-7

156

Q. Ye et al. / Journal of Controlled Release 64 (2000) 155 166

systems, while a few deal with multivesicular liposomes, or DepoFoamE particles [1013]. A unique
feature of the DepoFoam system is that inside each
DepoFoam particle, discontinuous internal aqueous
chambers, bounded by a continuous, non-concentric
network of lipid membranes render a higher aqueous
volume-to-lipid ratio and much larger particle diameters compared with MLV [10]. Fig. 1 illustrates the
differences between DepoFoam particles, ULV and
MLV. This multivesicular nature also provides for

sustained release of encapsulated drug since, unlike

ULV, a single breach in the external membrane of a
DepoFoam particle will not result in a total release of
the internal aqueous contents.
This paper addresses the major challenge in the
development of sustained release formulations for
therapeutic proteins and peptides, namely, achieving
high drug loading with high drug recovery, and
controlling its delivery over a prolonged period. With
the use of DepoFoam technology, we have de-

Fig. 1. Illustrative electron micrographs of (A) unilamellar vesicles (ULV) (adapted from Lasic [14]) which contain a single internal aqueous
compartment and typically have a diameter of 0.020.5 mm; (B) multilamellar vesicles (MLV) (adapted from Lasic [14]) which contain
multiple concentric internal aqueous compartments and typically have a diameter of 0.25 mm; and (C) multivesicular liposomes
(DepoFoamE particles) (adapted from Ref. [15]), which contain multiple non-concentric internal aqueous compartments and typically have
a diameter of 1100 mm. These electron micrographs show that DepoFoam particles are structurally distinct from MLV and ULV.

Q. Ye et al. / Journal of Controlled Release 64 (2000) 155 166

veloped sustained-release protein and peptide formulations which meet this challenge. In particular,
DepoFoam formulations of the following protein and
peptides are evaluated: (1) recombinant human insulin [16,17], which also has been encapsulated into
traditional liposomes [18]; (2) leuprolide: a nine
amino acid peptide analogue of LHRH [1921]; (3)
met-enkephalin: an endogenous opioid peptide of
five amino acids [22,23]; and (4) octreotide: a cyclic
octapeptide analog of somatostatin [24,25].

2. Materials and methods

2.1. Materials
Recombinant human insulin, produced in E. coli,
FITC-labeled insulin, and met-enkephalin were obtained from Sigma Chemical Co. (St. Louis, MO).
Leuprolide acetate was obtained from Bachem Bioscience Inc. (King of Prussia, PA), and octreotide
was custom-synthesized by Multiple Peptide Systems
Inc. (San Diego, CA). BODIPY 665 / 676 (B-3932)
was obtained from Molecular Probes Inc. (Eugene,
OR). The phospholipids, 1,2-dioleoyl-sn-glycero-3phosphocholine (DOPC or DC18:1 PC), 1,2dierucoyl-sn-glycero-3-phosphocholine (DEPC or
DC22:1 PC), 2-dinervonoyl-sn-glycero-3-phosphocholine (DNPC or DC24:1 PC), 1,2-dipalmitoyl-snglycero-3-phosphoglycerol (DPPG) and triolein were
from Avanti Polar Lipids Inc. (Alabaster, AL), and
tricaprylin from Sigma (St. Louis). Cholesterol was
from Spectrum Chemical Manufacturing Corp. (Gardena, CA). L-Lysine was from Degussa Corp. (Marceau, France), normal saline (0.9% sodium chloride)
and 50% glucose from McGaw Inc. (Irvine, CA),
citric acid from Sigma (St. Louis, MO), 1 M
hydrochloric acid from Fisher Chemical (Fair Lawn,
NJ). UV absorbance was determined with a U2000
UV/ VIS spectrophotometer from Hitachi Instruments
Inc. (San Jose, CA). Reverse phase-HPLC was
performed with an Hewlett-Packard Model 1100
HPLC system equipped with a Diode Array Detector
(Hewlett-Packard GmbH, Waldborn, Germany) and a
Waters C18 symmetric column (Millipore, Milford,
MA).

157

2.2. Preparation of multivesicular liposome

formulations ( DepoFoam formulations)
The multivesicular DepoFoam particles containing
protein or peptide drugs were prepared by a two-step
water-in-oil-in-water double emulsification process,
described in Refs. [10,13]. Briefly, the first step is
the formation of a water-in-oil emulsion. A lipid
combination solution made of 13.20 mM DCn:1PC
(n518, 22, or 24), 19.88 mM cholesterol, 2.79 mM
DPPG, and 2.44 mM triolein or tricaprylin in
chloroform was emulsified with an equal volume of
an aqueous solution (the first aqueous solution)
containing drugs (insulin, met-enkephalin, octreotide, or leuprolide,) in 50 mM HCl (for insulin),
or 25 mM citric acid (for enkephalin and octreotide),
or 100 mM phosphoric acid (for leuprolide), and
varying amounts of sucrose (ranging from 2.5
7.5%), to produce a water-in-oil emulsion (the first
emulsion). A subsequent emulsification with a second aqueous solution, such as 1.5% glycine containing 40 mM lysine, resulted in a water-in-oil-inwater double emulsion (the second emulsion). The
second emulsion was transferred and added to
another aliquot of the second aqueous solution.
Chloroform was removed by flushing nitrogen over
the surface of the mixture at 378C. The resulting
multivesicular liposomes were washed to remove
unencapsulated drug and harvested by centrifugation
for 10 min at 6003g and then resuspended in
buffered saline solution. After harvesting the DepoFoam particles, the amount of free drug in the
supernatant was determined, which reflects the
amount of drug released subsequent to the wash
steps, and is an indication of the stability of the
DepoFoam matrix.
DepoFoam-encapsulated insulin (DepoInsulin)
formulations 1 and 2 were prepared under the
following first aqueous conditions: (1) 10 mg / ml
insulin, 5% sucrose, 50 mM HCl, and (2) 30 mg / ml
insulin, 2.5% sucrose, 50 mM HCl. A DepoInsulin
formulation similar to (1) containing fluorescent
labels, both on the protein (FITC) and on the lipids
(BODIPY) was prepared for confocal microscopy
using a first aqueous solution containing 50% FITCinsulin and 50% unlabeled insulin, and BODIPYlabeled lipid solution. The lipid solution was labeled
with 10% (v / v) of a BODIPY 665 / 676 solution (0.2

158

Q. Ye et al. / Journal of Controlled Release 64 (2000) 155 166

mg / ml). DepoLeuprolide formulation used in the in

vitro and in vivo studies was prepared with first
aqueous conditions of 15 mg / ml leuprolide, 6%
sucrose and 100 mM phosphoric acid. DepoOctreotide and DepoEnkephalin formulations used in
the in vitro studies were prepared with first aqueous
conditions of 10 mg / ml of appropriate peptide, 5%
sucrose and 25 mM citric acid, and with DepoEnkephalin the tricaprylin:triolein ratio in the lipid
solution was varied as follows: 9:1 and 14:1. DepoFoam suspensions were typically suspended at
4050% lipocrit.

2.3. Determination of percent encapsulation (or

percent yield), lipocrit, percent free drug, particle
size distribution, and drug loading
Percent encapsulation (or percent yield) of drug is
defined as the percent ratio of the amount of drug in
the final liposome suspension to the total amount of
drug used in the first aqueous solution. Lipocrit is
defined, in analogy to hematocrit, as the percent ratio
of the pellet volume to the suspension volume (see
conditions below for obtaining the pellet volume).
Percent free drug is defined as the percent ratio of
the amount of drug in the supernatant to the amount
of drug in the final suspension times (1-lipocrit).
Drug loading is defined as the amount of drug
encapsulated in each unit of the encapsulated volume, and can be estimated as the ratio of the drug
concentration of the final DepoFoam suspension to
the lipocrit.
Lipocrit was calculated by the hematocrit method.
In this method, about 50 ml of the multivesicular
liposome suspension was taken up into a capillary
tube. One end of the tube was sealed with Critoseal
(Oxford Labware, St. Louis, MO) while ensuring
that there were no air bubbles. Upon centrifugation
at 6003g for 10 min, the suspension was separated
into a pellet layer and a supernatant layer. The
percent ratio of the length occupied by the pellet to
that by the supernatant plus pellet gave the lipocrit.
To determine the free drug concentration in the
formulation product, the supernatant was obtained by
centrifugation for 10 min at 6003g and then extracted with acidified isopropyl alcohol (IPA) solution, typically 24-fold, followed by rigorous mixing
to obtain a clear solution. For determination of the

total concentration, the DepoFoam suspension was

extracted with the same extraction solution as for the
supernatant except that a higher extraction dilution,
typically 10-fold with the acidified IPA was used.
The absorbance at 275 nm for enkephalin, and 280
nm for insulin, leuprolide and octreotide was measured with a UV spectrophotometer. For reversephase HPLC, a typical linear gradient, as described
in Ref. [13], made of mobile phase A (10%
acetonitrile / 0.2% trifluoroacetic acid) and mobile
phase B (90% acetonitrile / 0.2% trifluoroacetic acid)
was used for elution, with UV detection at 275 and
214 nm for enkephalin, and 280 and 214 nm for
insulin, leuprolide and octreotide. The drug concentrations in the supernatant and suspension were
determined using a standard curve generated by
known concentrations of appropriate drug.
Particle size distribution and the median diameter
were determined by the method of laser light diffraction using a LA-910 particle size analyzer from
Horiba Inc. (Irvine, CA). Confocal microscopy was
performed using a BioRad MRC 1024 laser scanning
confocal microscope (BioRad Corporation). Lipid
probes are shown in the images as red, while the
aqueous probes are green.

2.4. In vitro release

Briefly, in vitro release experiments were performed as follows: the DepoFoam suspensions were
diluted 5-fold into human plasma containing 0.01%
NaN 3 ; a 0.5-ml sample of the diluted suspension in
screw-cap Eppendorf tube was used for each timepoint; samples were incubated under gentle rotating
conditions (12 cycles / min) at 378C. Duplicate samples were taken for analyses according to the planned
schedule to which 0.9 ml of normal saline was
added, and particle pellets were then obtained by
centrifugation in a micro-centrifuge at 16 0003g for
4 min and stored at 2208C until assayed by RPHPLC. Particle pellets were extracted with acidified
IPA as described above, and the extracts analyzed by
RP-HPLC.

2.5. In vivo study

Pharmacodynamic studies were conducted in rats
to investigate the in vivo behavior of DepoFoam

Q. Ye et al. / Journal of Controlled Release 64 (2000) 155 166

encapsulated leuprolide (DepoLeuprolide) formulation. A DepoLeuprolide formulation described in

Section 2.2 was compared to a 1-month sustained
release polymer microsphere formulation (Lupron
Depot, [19]). Their in vivo efficacy was evaluated in
terms of the suppression of serum testosterone level.
Blood samples were collected according to schedule
after single intramuscular injections of DepoLeuprolide or Lupron Depot into male SpragueDawley
rats, and sera were analyzed for testosterone concentration using a standard RIA assay conducted in
Dr David Hesss laboratory at the Oregon Regional
Primate Research Center (Beaverton, OR). The same
dose (3.75 mg) of DepoLeuprolide or Lupron Depot
was given to each of the three rats in each group.

3. Results

3.1. Characterization of DepoFoam encapsulated

protein and peptide formulations
Table 1 presents a summary of some major
characteristics of DepoFoam formulations of insulin,
leuprolide, met-enkephalin, and octreotide, and
shows that all four therapeutic molecules have been
encapsulated with good efficiency (5085%) into
DepoFoam particles. The solubilization medium of
acidified IPA used to extract all four drugs prior to
RP-HPLC analyses was satisfactory for quantitation,

Table 1
Representative DepoFoam encapsulated protein and peptide
formulations a
DepoFoam formulations
Drug
(protein or
peptide)

%Yield
(recovery)

Drug
loading
(mg / ml)

Drug integrity
by HPLC b (% of
standard drug)

Insulin
Leuprolide
Enkephalin
Octreotide

7585
5060
5570
5060

850
818
2535
1626

99.8
99.1
99.5
96.7

The formulation conditions have been described in Section

2.2. Drug loading and % yield are as defined in Section 2.3. Drug
integrity of the DepoFoam encapsulated drug was determined by
RP-HPLC, and expressed as the percent purity with respect to the
standard free form of the drug.
b
Percent purity relative to the free form of protein or peptide.

159

and gave 100% recovery of the drug, which was

determined by spiked recovery of the drug. The data
also show that a high drug loading up to 50 mg / ml
has been achieved for DepoInsulin, with a high
encapsulation yield of 7585%. Relatively high drug
loading and encapsulation yield have also been
achieved for peptide drugs, leuprolide, enkephalin
and octreotide, upon formulation in DepoFoam. The
levels of drug loading for these peptide formulations
are appropriate for their potential therapeutic applications. The concentration of total lipids in all the
DepoFoam formulations in Table 1 was 10 mg / ml.
Therefore, the drug to lipid ratios in the DepoFoam
formulations ranged from 1 to 5.
The recoveries shown in Table 1 depend on the
formulation conditions, typically the concentration of
drug in the first aqueous solution, where lower
recoveries are obtained with higher concentration of
drug. The drug loading shown in Table 1 can be
modulated by the osmolarity of the first aqueous
solution, for example, with leuprolide concentration
at 15 mg / ml in 100 mM phosphoric acid, the peptide
loading of 11 mg / ml with 6% sucrose (|330 mOsM)
was increased to 18 mg / ml with a decrease in
sucrose concentration to 4% and a corresponding
decrease in osmolarity to |260 mOsM. With insulin
at 10 mg / ml in 50 mM HCl, the loading in DepoFoam particles increased two-fold, from 8 to 16
mg / ml with the decrease in sucrose concentration
from 7.5% (|320 mOsM) to 2.5% (|150 mOsM).
Fig. 1C depicts the DepoFoam structure both
schematically and in an electron micrograph in
comparison with traditional liposome structures (Fig.
1A and B). Fig. 2 shows the structure and morphology at the light microscope level of multivesicular
liposome particles (DepoFoam) containing a protein
drug (insulin) illustrated in a confocal image of a
fluorescent double-labeled DepoInsulin preparation.
The BODIPY labeled lipids are show in red on the
far left, with the FITC-insulin label in green in the
middle, and the merged image showing both fluorescent labels in the far right. The even distribution of
lipid and protein in these particles illustrates no
preferential segregation of the protein molecules in
the DepoFoam particles. Furthermore, the unique
multivesicular internal structure of the DepoInsulin
particle is clearly illustrated in this figure, which is
distinct from structures of traditional liposomes.

160

Q. Ye et al. / Journal of Controlled Release 64 (2000) 155 166

Fig. 2. Confocal images of double-labeled DepoInsulin particle. Confocal microscopy was performed using a BioRad MRC 1024
microscope. DepoInsulin formulation was prepared using FITC-insulin and BODIPY-labeled lipid solution, as described in Section 2. The
far left image shows the red-fluorescent lipid label, the middle image is the green-fluorescent FITC label showing the encapsulated insulin,
and the far right image shows both lipid and protein in the DepoFoam (MVL) particle, illustrating the multivesicular structure of the particle,
and the even distribution of the protein in the DepoFoam particle.

Fig. 3A and B show representative particle size

distributions of a DepoInsulin and a DepoEnkephalin
formulation, with volume-weighted median diameters of 13 and 15 mm, respectively, and 95% of the
particles sizing between 5 and 50 mm for DepoInsulin and 96% of the particles ranging from 5 to 30
mm for DepoEnkephalin. In general, monomodal and
relatively narrow particle size distributions were
achieved for DepoFoam encapsulated protein and
peptide formulations, with a volume-weighted mean
diameter usually ranging from about 6 to 20 mm and
with no particles smaller than 1 mm or greater than
100 mm. Particle size can be modulated by the
process parameters, mostly the emulsification conditions of the second emulsification, which results in
the water-in-oil-in-water emulsion, and this size is
not dependent on the first aqueous conditions of
sucrose or protein concentrations.
The RP-HPLC method is sensitive to some modifications of proteins, and to fragmentation of the
peptide / protein drugs. No such changes were observed by RP-HPLC analysis, as shown in Fig. 4,
comparing the HPLC profile of insulin with that of
DepoInsulin. Earlier work with DepoIGF-I [13] has

also shown that there are no chemical changes in the

protein upon encapsulation into DepoFoam. RPHPLC analyses for the peptide formulations have
indicated little or no drug degradation caused by the
DepoFoam encapsulation process. As an example,
Fig. 5 shows a comparison of the RP-HPLC profile
of DepoLeuprolide with that of leuprolide, which
indicates that the peptide is unchanged upon encapsulation into DepoFoam. In these RP-HPLC
assays, UV absorbance at both 280 / 275 nm and 214
nm were monitored, in order to ensure that there
were no peptide fragments generated due to degradation of the drugs during the encapsulation process.
The lack of adverse effects on the encapsulated
proteins and peptides was also demonstrated by the
following: (i) the retention of full biological activity
of DepoInsulin, measured by receptor binding assay
[26], (ii) pharmacodynamic study with DepoLeuprolide in rats (data shown below), (iii) pharmacodynamic study with DepoEnkephalin in rats showed
that analgesia could be sustained for up to 6 h (data
not shown) with the hot-plate method described in
Ref. [12], (iv) our earlier work with DepoFoamencapsulated IL-2 (interleukin-2) [27] and IGF-I

Q. Ye et al. / Journal of Controlled Release 64 (2000) 155 166

161

Fig. 4. Comparison of the RP-HPLC profile of unencapsulated

(free form) of recombinant human insulin to that of recombinant
human insulin extracted from DepoFoam particles, taken using a
HP1100 HPLC. The profiles show the UV absorbance detected at
280 nm versus retention time in minutes.

Fig. 3. Representative volume-weighted particle size distributions

for DepoFoam formulations. (A) DepoInsulin (median size 13
mm); and (B) DepoEnkephalin (median size 15 mm), showing that
the size distribution is narrow and monomodal, and there are no
particles with diameters smaller than 1 mm or greater than 100
mm.

(insulin-like growth factor-I) [13,27], showed that

both proteins retained their full biological activities
(in vitro cell proliferation assays) after encapsulation
into DepoFoam particles.
The percent free drug in all the DepoFoam
suspensions of insulin, enkephalin, octreotide and
leuprolide was ,2%, after preparation of the DepoFoam formulations. There was no leakage of any
of the drugs from the DepoFoam particles over a
3-month period when stored at 48C, indicating that
the DepoFoam matrix is stable to storage at 48C over
the indicated time period. DepoIGF-I has been
evaluated in a long-term storage stability study,

Fig. 5. Comparison of the RP-HPLC profile of unencapsulated

(free form) leuprolide to that of leuprolide extracted from DepoFoam particles, taken using a HP1100 HPLC. The profiles show
the UV absorbance detected at 280 nm versus retention time in
minutes.

162

Q. Ye et al. / Journal of Controlled Release 64 (2000) 155 166

where it was demonstrated that the DepoFoam

particles were stable over 1 year at 48C [13].

3.2.1. In vitro release

Fig. 6 shows in vitro release profiles for DepoInsulin and DepoLeuprolide formulations in
human plasma. The percentage of drug retained by
the DepoFoam particles as a function of time relative
to that at time zero indicates a sustained release of

the encapsulated protein and peptide over a period of

about a week for the DepoInsulin formulation (Fig.
6A), and over a period of several weeks for the
DepoLeuprolide formulation (Fig. 6B). Fig. 6A
illustrates the modulation of the release rate upon
variation in the first aqueous conditions during the
DepoInsulin manufacturing process, as described in
Section 2.2, which results in a decrease in insulin
released by day 8 from 70% (formulation 1) to 40%
(formulation 2).
Fig. 7 shows sustained in vitro release over a
period of several days for DepoOctreotide (Fig. 7A),
and several hours for DepoEnkephalin (Fig. 7B).

Fig. 6. In vitro release characteristics of (A) DepoInsulin and (B)

DepoLeuprolide in human plasma under gentle rotating conditions
(12 cycles / min) at 378C. The percent drug retained in the particle
fractions over time were measured by RP-HPLC, and normalized
to time 0. Notice the modulation of protein release in DepoInsulin
(A) by variation in the first aqueous conditions (described in
Section 2.2) during the formulation process. The data represent
the mean of duplicates, with the duplicates matching within 5%.

Fig. 7. In vitro release characteristics of (A) DepoOctreotide, and

(B) DepoEnkephalin in human plasma, under gentle rotating
conditions (12 cycles / min) at 378C. Notice the modulation of drug
release rates by variation of the tricaprylin:triolein (C8:C18) ratio
in the DepoEnkephalin formulations (B). The percent drug
retained in the particle fractions over time were measured by
RP-HPLC, and normalized to time 0. The data represent the mean
of duplicates, with the duplicates matching within 5%.

3.2. In vitro release and in vivo behavior of

DepoFoam encapsulated protein and peptide
formulations

Q. Ye et al. / Journal of Controlled Release 64 (2000) 155 166

The rate of drug release from the DepoFoam particles can also be modulated by varying the lipid
composition in the DepoFoam formulation. As
shown for DepoEnkephalin formulations in Fig. 7B,
an increase of tricaprylin:triolein (C8:C18) ratio
from 9:1 to 14:1 resulted in an increase of enkephalin release from about 5% release to 60%
release on day 3. The impact of the triglycerides on
controlling the release rates of DepoFoam-encapsulated proteins, such as GM-CSF, G-CSF and IGF-I,
has been described earlier [28].
It should be noted that the in vitro release profile
obtained may not be quantitatively predictive of the
in vivo behavior. Nevertheless, in vitro release
studies allow for screening and rank-ordering various
formulations by the rates of drug release, and
provide a basis for in vivo studies. The in vitro
release studies shown in Figs. 6 and 7 were performed after dilution of the DepoFoam formulations
in human plasma and incubation at 378C under
dynamic conditions, and therefore do not relate to
storage stability of the formulations (stored undiluted
at 48C). In Figs. 6 and 7, the release at time zero was
normalized to 100%, since there was no drug release
at time zero, and the absolute values corresponded to
the initial drug concentration.

163

Fig. 8. In vivo pharmacodynamic profiles of a DepoLeuprolide

formulation (h) compared to the normal saline control () and
Lupron Depot (n) after single intramuscular injections of the
same dose (3.75 mg / rat) into rats. The prolonged suppression of
testosterone level in serum (mean of three rats) by DepoLeuprolide, similar to that by Lupron Depot, demonstrates a sustained
release of leuprolide from the DepoFoam formulation. The initial
spike in testosterone level obtained with leuprolide, using both
Lupron depot and DepoLeuprolide injections is not shown in the
figure.

and proteins, such as, IGF-I [13,28], GM-CSF and

G-CSF [28], and interferon alpha-2b [31].

4. Discussion

3.2.2. In vivo release

Fig. 8 shows a pharmacodynamic profile obtained
for a DepoLeuprolide formulation after a single
intramuscular injection into rats. The time course
represents an average serum testosterone level of
three rats after each received the same dose (3.75
mg / rat) of DepoFoam encapsulated leuprolide. The
data indicate that a sustained suppression of testosterone level in rats is achieved over a period of a
month, with a pharmacodynamic profile similar to
that generated by a microsphere encapsulated leuprolide formulation (Lupron Depot, [19]). These data
indicate that a sustained biological effect can be
achieved over a 1-month period with DepoLeuprolide formulations. The in vitro to in vivo correlation showing that the sustained release in vitro
corresponds to sustained release in vivo has been
demonstrated for several DepoFoam-encapsulated
drugs, for example, small molecules, such as, morphine [29], methotrexate [30] and cytarabine [11],

The versatility of DepoFoam drug delivery system

as a vehicle for controlled delivery of protein and
peptide drugs has been demonstrated. The rate of
drug release in these formulations can be modulated
to cover a relatively wide range of time courses,
from one to two days up to several weeks. The data
also indicate that the integrity of the encapsulated
proteins and peptides has been maintained during the
formulation process. The achievement of several
important formulation objectives, i.e. high encapsulation yield and high loading combined with controlled
release, described in this paper illustrates the advantage of the DepoFoam technology for sustained
delivery of proteins and peptides.
Using proteins and peptides as model systems, we
have achieved encapsulation yields of up to 85% and
drug loading of up to 50 mg / ml. The DepoFoam
formulations also have a low content of free drug
and little chemical degradation due to the DepoFoam

164

Q. Ye et al. / Journal of Controlled Release 64 (2000) 155 166

encapsulation process, shown here and earlier [13].

However, for some drugs of high potency, a high
drug loading may not be required or essential. The
major factors affecting drug loading in the
DepoFoam particles are usually the aqueous solubility, the membrane permeability of the drug, and
the osmolarity of the drug-containing first aqueous
solution. With a typical percent encapsulated aqueous
volume of 95% for the DepoFoam particles, a much
higher drug loading has been achieved for the
DepoFoam system than that reported for traditional
unilamellar or multilamellar liposomes [14].
Various factors in the DepoFoam manufacturing
process can influence the rate of drug release. For
example, conditions of the aqueous solutions containing the drug (such as pH, osmolarity, drug
concentration and excipients), described in Refs.
[32,33], lipid composition [28], process parameters,
and drug / lipid interactions all impact on the rate of
drug release from the DepoFoam particles. We have
demonstrated here the impact of the aqueous conditions, such as sucrose and protein concentrations,
and lipid composition (variation in triglyceride content) on controlling the in vitro release rate of
DepoFoam-encapsulated protein and peptide formulations. Although the in vitro release profiles do not
necessarily predict in vivo behavior, they allow one
to assess the formulations with respect to relative
release rates and sustained release, as shown for
several proteins [13,28,31,33].
In the literature, various delivery systems have
been evaluated to address one or more of the aspects
of drug loading and controlled release. For instance,
Katre and co-workers [3436] have shown that
derivatization of therapeutic proteins with polymers
such as polyethylene glycol prolongs the in vivo
half-life of the protein drugs. Protein and peptide
encapsulation using other liposome systems [14] or
polymeric materials such as polylactide / polyglycolide copolymers [1921,3739] have also been
utilized to provide sustained delivery. However,
these approaches may not be able to address all the
important formulation objectives described above.
The biocompatibility of the formulation components is also an important factor for a parenteral
dosage system. The lipids used in the manufacture of
DepoFoam particles are synthetic versions of naturally occurring lipids and are, therefore, biocompatible. The DepoFoam particle itself does not cause

any local or systemic toxicity in humans or animals,

and there is no foreign body response at the
injection site after subcutaneous injections [40].
DepoFoam technology may, therefore, provide a
unique solution for controlled delivery of therapeutic
proteins and peptides.
For pharmaceutical application, a sustained release
DepoFoam protein and peptide formulations will
have obvious advantages in optimizing the pharmacokinetic profiles, and also be useful in the clinic
for developing controlled release dosage forms to
cover a wide range of dosing schedules. Finally, the
biocompatible DepoFoam matrix allows for delivery
into sensitive areas, such as ocular, intrathecal,
epidural, in addition to subcutaneous, intramuscular
and intraarticular.

5. Conclusion
In conclusion, the results demonstrate that DepoFoam technology is capable of: (a) encapsulating
therapeutic proteins and peptides with high yields;
(b) accommodating high drug loading; and (c) controlling the sustained delivery of the encapsulated
drugs. In addition to proteins and peptides, the
DepoFoam system is applicable for encapsulation of
a wide range of therapeutic agents, from small
molecule drugs, one of which (cytarabine) has already been demonstrated to be clinically beneficial
[41] and approved for clinical use, to other macromolecule drugs such as nucleic acids and vaccines.

Acknowledgements
We thank Beth Gualdoni and Mario Flores of
DepoTech Corporation for the animal work, Melissa
Langston for the confocal images, Dr Jerrold M.
Olefsky at Department of Endocrinology and Metabolism, UCSD for the insulin receptor assays, and Dr
David Hess at the Oregon Regional Primate Research Center for RIA assay of testosterone.

References
[1] D.D. Lasic, Liposomes: From Physics To Applications,
Elsevier, Amsterdam, 1993.

Q. Ye et al. / Journal of Controlled Release 64 (2000) 155 166

[2] D.D. Lasic, D. Papahadjopoulos, Liposomes revisited, Science 267 (1995) 12751276.
[3] P.M. Anderson, D.C. Hanson, D.E. Hasz, M.R. Halet, B.R.
Blazar, A.C. Ochoa, Cytokines in liposomes: preliminary
studies with IL-1, IL-2, IL-6, GM-CSF and Interferongamma, Cytokine 6 (1994) 92101.
[4] J. Meyer, L. Whitcomb, D. Collins, Efficient encapsulation of
proteins within liposomes for slow release in vivo, Biochem.
Biophys. Res. Commun. 199 (1994) 433438.
[5] L.D. Mayer, M.J. Hope, P.R. Cullis, Vesicles of variable sizes
produced by a rapid extrusion procedure, Biochim. Biophys.
Acta 858 (1) (1986) 161168.
[6] P.R. Cullis, M.J. Hope, M.B. Bally, T.D. Madden, L.D.
Mayer, D.B. Fenske, Influence of pH gradients on the
transbilayer transport of drugs, lipids, peptides and metal
ions into large unilamellar vesicles, Biochim. Biophys. Acta
1331 (2) (1997) 187211.
[7] C. Huang, Studies on phosphatidylcholine vesicles. Formation and physical characteristics, Biochemistry 8 (1) (1969)
344352.
[8] A.D. Bangham, N.M. Standish, J.C. Watkins, Diffusion of
univalent ions across the lamellae of swollen phospholipids,
J. Mol. Biol. 13 (1965) 238252.
[9] A.D. Bangham, Lipid bilayers and biomembranes, Annu.
Rev. Biochem. 41 (1972) 753776.
[10] S. Kim, M.S. Turker, E.Y. Chi, S. Sela, G.M. Martin,
Preparation of multivesicular liposomes, Biochim. Biophys.
Acta 728 (3) (1983) 339348.
[11] S. Kim, DepoFoam-mediated drug delivery into cerebrospinal fluid, Methods Neurosci. 21 (1994) 118131.
[12] T. Kim, S. Murdande, A. Gruber, S. Kim, Sustained-release
morphine for epidural analgesia in rats, Anesthesiology 85
(2) (1996) 331338.
[13] N.V. Katre, J. Asherman, H. Schaefer, M. Hora, Multivesicular liposome (DepoFoamE) technology for the sustained
delivery of insulin-like growth factor-I (IGF-I), J. Pharm.
Sci. 87 (11) (1998) 13411346.
[14] D.D. Lasic, Liposomes: synthetic lipid microspheres serve as
multipurpose vehicles for the delivery of drugs, genetic
material and cosmetics, Am. Sci. 80 (1992) 2031.
[15] M.S. Spector, J.A. Zasadzinski, M.B. Sankaram, Topology
of multivesicular liposomes: a model biliquid foam, Langmuir 12 (1996) 47044708.
[16] L. Heinemann, B. Richter, Clinical pharmacology of human
insulin, Diabetes Care 16 (3) (1993) 90100.
[17] A. Hoffman, F. Ziv, Pharmacokinetic Considerations of new
insulin formulations and routes of administration, Clin.
Pharmacokinet. 33 (4) (1997) 285301.
[18] R.S. Spangler, Insulin administration via liposomes, Diabetes
Care 13 (9) (1990) 911922.
[19] H. Okada, T. Heya, Y. Ogawa, T. Shimamoto, One-month
release injectable microcapsules of a luteinizing hormonereleasing hormone agonist (leuprolide acetate) for treating
experimental endometriosis in rats, J. Pharmacol. Exp. Ther.
244 (1988) 744750.
[20] A.M. Dlugi, J.D. Miller, J. Knittle, Lupron depot (leuprolide
acetate for depot suspension) in the treatment of endometriosis: a randomized, placebo-controlled, double-blind
study, Fertil. Steril. 54 (1990) 419427.

165

[21] H. Okada, Y. Doken, Y. Ogawa, H. Toguchi, Preparation of

three-month depot injectable microspheres of leuprorelin
acetate using biodegradable polymers, Pharm. Res. 2 (8)
(1994) 11431147.
[22] J. Hughes, H.W. Kosterlitz, T.W. Smith, The distribution of
methionine-enkephalin and leucine-enkephalin in the brain
and peripheral tissues, Br. J. Pharmacol. 120 (4) (1997)
428436.
[23] Y. Kuraishi, N. Hirota, M. Satoh, H. Takagi, Antinociceptive
effects of intrathecal opioids, noradrenaline and serotonin in
rats: mechanical and thermal algesic tests, Brain Res. 326 (1)
(1985) 168171.
[24] S.W. Lamberts, Acromegaly and its treatment, J. Endocrinol.
155 (1997) S49S51.
[25] M.N. Pollak, A.V. Schally, Mechanisms of antineoplastic
action of somatostatin analogs, Proc. Soc. Exp. Biol. Med.
217 (2) (1998) 143152.
[26] J.M. Olefsky, Personal communication, 1995.
[27] N.V. Katre, J. Asherman, H. Schaefer, M. Hora, A multivesicular lipid-based sustained-release system for the delivery of therapeutic proteins, in: Proc. 8th Int. Pharm. Technol.
Symp., Turkey, 1996, pp. 2021.
[28] R.C. Willis, Method for utilizing neutral lipids to modify
in-vivo release from multivesicular liposomes, US Patent
5,891,467, 1998.
[29] T. Kim, J. Kim, S. Kim, Extended-release formulation of
morphine for subcutaneous administration, Cancer
Chemother. Pharmacol. 33 (1993) 187190.
[30] E. Chatelut, P. Suh, S. Kim, Sustained-release methotrexate
for intracavitary chemotherapy, J. Pharm. Sci. 83 (3) (1994)
429432.
[31] A. Bonetti, S. Kim, Pharmacokinetics of an extended-release
human interferon alpha-2b formulation, Cancer Chemother.
Pharmacol. 33 (1993) 258261.
[32] M.B. Sankaram, S. Kim, Multivesicular liposomes with
controlled release of encapsulated biologically active substances, US Patent 5,766,627, 1998.
[33] S. Kim, S.B. Howell, Multivesicular liposomes having a
biologically active substance encapsulated therein in the
presence of a hydrochloride, US Patent, 5,807,572 1998.
[34] N.V. Katre, M.J. Knauf, W.J. Laird, Chemical modification of
recombinant interleukin-2 by polyethylene glycol increases
its potency in the murine Meth A sarcoma model, Proc. Natl.
Acad. Sci. USA 84 (6) (1987) 14871491.
[35] M. Knauf, D. Bell, P. Hirtzer, Z. Luo, Y. Young, N.V. Katre,
Relationship of effective molecular size to systemic clearance in rats of recombinant interleukin-2 chemically modified with water-soluble polymers, J. Biol. Chem. 263 (29)
(1988) 1506415070.
[36] N.V. Katre, The conjugation of proteins with polyethylene
glycol and other polymers, Adv. Drug Deliv. Rev. 10 (1993)
91114.
[37] J.L. Cleland, A.J.S. Jones, Stable formulations of recombinant human growth hormone and interferon-gamma for
microencapsulation in biodegradable microspheres, Pharm.
Res. 13 (10) (1996) 14641475.
[38] O. Johnson, W. Jaworowicz, J. Cleland, L. Bailey, M.
Charnis, E. Duenas, C. Wu, D. Shepard, S. Magil, T. Last, A.
Jones, The stabilization and encapsulation of human growth

166

Q. Ye et al. / Journal of Controlled Release 64 (2000) 155 166

hormone into biodegradable microspheres, Pharm. Res. 14

(6) (1997) 730735.
[39] L. Brown, L. Siemer, C. Munoz, R. Langer, Controlled
release of insulin from polymer matrices: In vitro kinetics,
Diabetes 35 (1986) 684691.

[40] J.M. Anderson, Personal communications, 1995.

[41] M.C. Chamberlain, S. Khatibi, J.C. Kim, S.B. Howell, E.
Chatelut, S. Kim, Treatment of leptomeningeal metastasis
with intraventricular administration of depot cytarabine
(DTC 101), Arch. Neurol. 50 (1993) 261264.