Current Drug Delivery, 2005, 2, 000-0001
1567-2018/05 $50.00+.00© 2005 Bentham Science Publishers Ltd.
Trevor P. Castor*
Aphios Corporation, 3-E Gill Street, Woburn, MA 01801, USA
Abstract: Phospholipid nanosomes are small, uniform liposomes manufactured utilizing supercritical fluid technologies.
Supercritical fluids are first used to solvate phospholipid raw materials, and then decompressed to form phospholipid
nanosomes that can encapsulate hydrophilic molecules such as proteins and nucleic acids. Hydrophobic therapeutics are
co-solvated with phospholipid raw materials in supercritical fluids that, when decompressed, form phospholipid nano-
somes encapsulating these drugs in their lipid bilayers. Mathematical modeling and semi-empirical experiments indicate
that the size and character of phospholipid nanosomes depend on the several process parameters and material properties
including the size and design of decompression nozzle, bubble size, pressure and the rate of decompression, interfacial
forces, charge distribution and the nature of compound being encapsulated. Examples are presented for the encapsulation
of a protein and hydrophobic drugs. In vitro and in vivo data on breast cancer cells and xenografts in nude mice indicate
that paclitaxel nanosomes are less toxic and much more effective than paclitaxel in Cremophor EL (Taxol). Camp-
tothecin nanosomes demonstrate that the normally very water-insoluble camptothecin can be formulated in a biocompati-
ble aqueous medium while retaining in vivo efficacy against lymphoma xenografts in nude mice. In vitro data for betulinic
acid nanosomes demonstrate enhanced efficacy against HIV-1 (EC50 of 1.01 µg/ml versus 6.72 µg/ml for neat betulinic
acid). Phospholipid nanosomes may find utility in the enhanced delivery of hydrophilic drugs such as recombinant pro-
teins and nucleic acid as well as hydrophobic anticancer and anti-HIV drugs.
Phospholipid nanosomes are small, uniform liposomes
manufactured utilizing supercritical fluid technologies. They
are nanometer-sized vesicles of phospholipid bilayers com-
prised of single or multiple lipid bilayers. As such, they are
non-toxic, non-antigenic and biodegradable in character
since they have the molecular characteristics of mammalian
cell membranes. Hydrophilic compounds and therapeutics
such as recombinant proteins are encapsulated in the aqueous
core while hydrophobic compounds and therapeutics such as
anticancer and other water-insoluble drugs are trapped within
the lipid bilayers. Encapsulation masks the hydrophobic
(water-insoluble) nature of drugs, and permits aqueous, bio-
compatible formulations to be administered. Phospholipid
nanosomes protect the encapsulated therapeutics prolonging
their circulation and increasing half-life. For cancer chemo-
therapy, this increases the likelihood that the drug will reach
and destroy cancer cells. Encapsulation also protects cells
from the circulating therapeutics and reduces their toxicities.
Supercritical fluids preferentially utilized for the manu-
facturing of phospholipid nanosomes are normally gases
such as carbon dioxide, fluorocarbons and alkanes at ambi-
ent conditions. When compressed, these gases become dense
phase fluids that exhibit enhanced thermodynamic properties
of solvation, selection, penetration and expansion. As shown
by the pressure-temperature diagram in (Fig. 1), a pure com-
pound enters its supercritical fluid region at conditions that
*Address correspondence to this author at the Aphios Corporation, 3-E Gill
Street, Woburn, MA 01801, USA; Tel: (001) 781-932-6933; Fax: (001)
781-932-6865; E-mail: firstname.lastname@example.org
equal or exceed both its critical temperature and critical pres-
sure. These critical parameters are intrinsic thermodynamic
properties of all sufficiently stable pure component com-
pounds. Carbon dioxide, for example, becomes supercritical
at conditions that equal or exceed its critical temperature of
31.1°C and its critical pressure of 7.38 Megapascals (MPa).
Fig. (1). Supercritical Fluid Phase Diagram
In the supercritical or near-critical fluid region, normally
gaseous substances, such as carbon dioxide, become dense
phase fluids that have been observed to exhibit greatly en-
2 Current Drug Delivery, 2005, Vol. 2, No. 4Trevor P. Castor
hanced solvation power as compared to the gaseous state. At
a pressure of 21 MPa and a temperature of 40°C, carbon di-
oxide has a density around 0.85 g/ml and behaves very much
like a nonpolar organic solvent such as hexane.
A supercritical fluid uniquely displays a wide spectrum
of solvation power because its density is strongly dependent
on both temperature and pressure - temperature changes of
tens of degrees or pressure changes by tens of atmospheres
can change solubility by an order of magnitude or more. This
unique feature facilitates solute recovery, the "fine-tuning"
of solvation power and the fractionation of mixed solutes.
The selectivity of nonpolar near-critical or supercritical fluid
solvents can be further enhanced by the use of small con-
centrations of polar entrainers or cosolvents such as ethanol,
methanol or acetone. In addition to its unique solubilization
characteristics, a supercritical fluid possesses other physico-
chemical properties that add to its attractiveness as a solvent.
A supercritical fluid solvent can exhibit a liquid-like density
and, at the same time, gas-like properties of viscosity and
diffusivity. The latter increases mass transfer rates, signifi-
cantly reducing processing times. Additionally, the ultra-low
surface tension of a supercritical fluid allows facile penetra-
tion into microporous materials, increasing extraction effi-
ciency and overall yields. Supercritical fluids, critical or
near-critical solvents with/without cosolvents are jointly re-
ferred to as SuperFluids [SFS].
In the SuperFluids phospholipid nanosomes (SFS-
CFN) manufacturing process [1-3], SFS at appropriate con-
ditions of pressure and temperature are utilized to solvate
phospholipids, cholesterol and other nanosomal raw materi-
als in an apparatus such as that shown in the (Fig. 2). The
ability of a selected SFS at certain conditions of temperature
and pressure to dissolve the nanosomal raw materials is a
key ingredient in the process. This selection is thus an im-
portant process parameter. A circulation pump is utilized to
ensure good mixing between the SFS and nanosomal raw
materials in an upper high-pressure loop. After a specific
residence time, the resulting mixture is decompressed via a
backpressure regulator (valve) though a dip tube with a noz-
zle into a decompression chamber (vessel B) that contains
phosphate-buffered saline or other biocompatible solution.
Bubbles will form at the injection nozzle tip because of
SFS depressurization and phase-conversion into a gas, and
the solvated phospholipids will deposit out at the phase
boundary of the aqueous bubble. As the bubbles detach from
the nozzle into the aqueous solution, they rupture causing
bilayers of phospholipids to peel off, encapsulating solute
molecules and spontaneously sealing themselves to form
phospholipid nanosomes. This SFS-CFN injection technique
is ideally suited for the nanosomal encapsulation of recom-
binant proteins for enhanced drug delivery, DNA used in
gene therapy, RNAi delivery, and other hydrophilic thera-
peutics and solutes.
In a second SFS-CFN technique, the phospholipids and
the target compound are solvated simultaneously in a SFS
"cocktail," which is dispersed continuously into an aqueous
environment. The process stream is decompressed, and the
unstable phospholipid bilayer fragments collide and rapidly
seal to form nanosomes, entrapping the target compound.
The controlling parameters for this process are pressure and
rate of decompression. The decompression technique is
readily scaled to larger production volumes. It is a "one-step"
process, and the SFS stream composition can be designed to
achieve concentrated phospholipid and target compound feed
streams. The results are high trapping efficiencies and con-
centrated product recovery streams in the SFS-CFN decom-
In a third technique, SFS-CFN evaporation can be
uniquely utilized to encapsulate very hydrophobic molecules
such as the potent anticancer drug paclitaxel, camptothecin, a
very effective topoisomerase-I inhibitor, and other anticancer
and anti-HIV therapeutics such as betulinic acid and bry-
ostatin 1. In this technique, the hydrophobic drug(s) and the
phospholipids are directly solvated in the SFS prior to injec-
tion into a phosphate-buffered saline or other biocompatible
solution. After decompression through a nozzle, the SFS
evaporates off leaving an aqueous solution of nanosomes
entrapping hydrophobic molecules within their lipid bilayers.
SUPERFLUIDS CFN EQUIPMENT
The SFS-CFN equipment shown in (Fig. 3) consists of
three major components or modules: (1) a high-pressure feed
system; (2) a high pressure circulation system for mixing and
solubilizing the nanosomal raw materials (and in some cases,
hydrophobic therapeutics) into the SFS; and (3) a decom-
pression module for reducing the pressure to form the nano-
somes and separate the SFS from the nanosomal product.
The first module consists of three high-pressure syringe
pumps, one each for the supercritical fluid, cosolvent and
Fig. (2). SuperFluids CFN Apparatus
Phospholipid NanosomesCurrent Drug Delivery, 2005, Vol. 2, No. 4 11
confidence limits) and, by HPLC analysis, contained 12.5
µg/ml betulinic acid. After freeze-drying and re-suspension
in de-ionized water, the mean diameter increased slightly to
202 nm and the betulinic acid content decreased slightly to
11.1 µg/ml. In a subsequent SFS-CFN experiment, betulinic
acid nanosomes were formed in a 1% sucrose solution in
order to increase the drug concentration by 10-fold after
freeze-drying and re-suspension. The initial product had an
average unimodal mean diameter of 186 nm and contained
18.6 µg/ml betulinic acid. After freeze-drying and re-
suspension in de-ionized water by a concentration of 12, the
average unimodal mean diameter increased to 235 nm and
the betulinic acid content increased significantly to 87.7
A nanosomal formulation run was performed with the
phospholipid raw materials but without betulinic acid. The
conditions for this run were identical to that of the betulinic
acid nanosomes run. This nanosomal formulation, also in a
1% sucrose solution, served as a vehicle control in subse-
quent bioactivity assays.
Anti-HIV Activity and Cytotoxicity
Several experiments were conducted to evaluate the anti-
HIV activity of betulinic acid nanosomes made utilizing
SFS-CFN and compare that with the anti-HIV activity of
non-encapsulated betulinic acid.
A stock solution of the betulinic acid standard at 10
mg/ml was prepared in dimethyl sulfoxide (DMSO). Serial
1:2 or 1:5 dilutions were tested in triplicate in the anti-HIV
assay, described below, beginning with 100 µg/ml. The same
batch of betulinic acid was used to prepare the nanosomal
formulation of betulinic acid in a 1% sucrose solution. The
nanosomes, with a betulinic acid concentration of 18.6
µg/ml, was filtered through a 0.2 µm low protein-binding
filter and serial 1:2 or 1:4 dilutions were tested in triplicate
beginning with 9.3 µg/ml of betulinic acid. A sample of the
empty nanosomes was also filtered through a 0.2 µm low
protein binding filter and serial 1:2 dilutions were tested in
triplicate beginning with the 1:2 dilution.
Viral reduction assays were performed to determine and
quantify the presence of anti-HIV activity in the samples.
Briefly, 100 TCID50 (50% tissue culture infectious dose) of
HIVIIIB was incubated with CEM-SS cells for 1 hour at 37oC.
Upon removal of the virus, serial dilutions of the sample,
3TC, or media were added and the cells incubated at 37oC.
On day 3, most of the media was replaced with fresh media
with or without the diluted sample or 3TC. On day 7, super-
natant was collected and the amount of p24 assayed by
ELISA (SAIC-Frederick). 3TC, known inhibitors of HIV,
were used as a positive control for p24 reduction. Wells
lacking virus served as a negative control for virus infection
while wells containing virus, but no sample/drug, served as a
positive control for virus infection. The percent p24 reduc-
tion was determined by the following equation: [((cells +
virus) – (cells + virus + test item))/(cells + virus)] * 100.
When possible, the 50% effective concentration (EC50) was
Additionally, a cytotoxicity assay was performed with
the samples. Briefly, serial 1:2 dilutions of samples was
added to 50,000 CEM-SS cells per well of a 96-well plate
and allowed to incubate at 37oC for 7 days. On day 3, 150
µl/well of spent media were replaced with media or diluted
samples. After the incubation a viability dye, WST-1, was
added and the plate incubated for 1 hour. After shaking the
plate for 1 minute, it was read in a microplate reader at 450
nm with a reference of 630 nm. Controls included media
alone, cells alone, as well as dilutions of DMSO and 1%
sucrose that were present in the betulinic acid and nanosomal
formulation samples, respectively. After subtraction of the
background (media alone wells), the percent toxicity was
calculated from the following: [((cells + media) – (cells +
sample))/(cells + media)] *100. The 50% cytotoxic concen-
tration (CC50) was then determined. Finally, the selective
index (SI), which is a ratio of CC50 to EC50, was determined.
In an in vitro cytoprotection assay, betulinic acid inhib-
ited p24 production by 76% at 20 µg/ml, (Fig. 13); the EC50
was 6.72 µg/ml or 14.7 µM. The positive control 3TC re-
sulted in 93% inhibition of p24 production at 0.2 µM, as
expected. The nanosomal formulation of betulinic acid in-
hibited p24 production by 73% at 2.3 µg/ml; the EC50 was
1.01 µg/ml or 2.2 µM (based on betulinic acid content). A
second assay confirmed these results, showing that betulinic
acid nanosomes inhibit HIV at lower concentrations than
betulinic acid alone, and that the empty nanosomes have a
minimal effect on HIV. Given the consistent results obtained
with the positive control, these experiments suggest that the
nanosomal formulation, produced by SFS-CFN, enhances
the natural properties of the nanoencapsulated betulinic acid.
Fig. (13). Inhibition of HIV p24 by Neat Betulinic Acid and Nano-
somal Betulinic Acid
Phospholipid nanosomes can be used for the encapsula-
tion and improved drug delivery of protein macromolecules,
nucleic acids and other hydrophilic drugs, as well as hydro-
phobic therapeutics such as anticancer and anti-HIV drugs.
The nanosomes are manufactured by a supercritical fluid
process that does not utilize organic solvents such as chloro-
form. In traditional rotary evaporation techniques, chloro-
form may be deleterious to the encapsulated therapeutics and
phospholipid raw materials. The phospholipid nanosomes
process substitutes supercritical, critical or near-critical fluid
solvents with or without polar cosolvents such as ethanol for
organic solvents utilized to solvate phospholipid raw materi-
The SFS-CFN process, for manufacturing phospholipid
nanosomes, also replaces the shear generated by sonication
12 Current Drug Delivery, 2005, Vol. 2, No. 4Trevor P. Castor
and high-pressure homogenization with decompressive
forces. For the SFS utilized, the maximum pressure require-
ments are around 3,000 to 5,000 psig because both effects of
decompressive energy and phospholipid solubility should
reach points of diminishing returns beyond a reduced pres-
sure (operating pressure divided by critical pressure) of
about 3.0. These pressure requirements are significantly
lower than those of high-pressure homogenization, reducing
equipment wear and operating costs.
The process does not generate excessive heat that can
damage thermally labile proteins and phospholipids. Unlike
high-pressure homogenization that can result in the genera-
tion of local hot spots by the dissipation of pressure and
shear energy within the microchannels of the homogenizing
valve, the SFS-CFN process produces a cooling effect as a
result of Joule-Thompson effects generated during decom-
pression. This cooling effect can be designed to freeze multi-
lamellar vesicles (MLVs), and the subsequent formation of
FATMLVs (freeze and thawed multi-lamellar vesicles) with
the added benefits of higher loading efficiencies and slower
drug release rates.
The process has several degrees of freedom such as noz-
zle diameter and size, operating pressure and temperature,
type and concentration of SFS solvents as well decompres-
sion rates and manufacturing technique. The decompression
technique, like high pressure homogenization, can be used to
form small unilamellar vesicles (SUVs). The injection tech-
nique, which is more like the rotary evaporation and sonica-
tion techniques, can be used to form SUVs, LUVs (large
unilamellar vesicles), MLVs and FATMLVs. The evapora-
tion technique can be used to readily encapsulate hydropho-
bic drugs. With all three techniques, the SFS mixture can be
decompressed through a filter to decrease nanosome size,
improve uniformity and, with 0.22-micron filters, enhance
The SFS-CFN process is a single-step one increasing
speed of processing versus the multiple pass requirements of
high-pressure homogenization. The process can be operated
in a continuous mode because it is a single step-one, and the
primary solvent can be readily separated, recovered and re-
used. The SFS-CFN manufacturing techniques are scalable
because they are essentially liquid-liquid unit operations, and
may impart an enhanced degree of end-product sterility since
supercritical fluid technologies will disrupt and inactive mi-
crobial and viral pathogens [23-27].
Phospholipid nanosomes could lead to an increase in a
drug’s therapeutic index resulting in: (i) enhanced therapeu-
tic efficacy; (ii) elimination of pre-medication to counteract
vehicles such as Cremophor El; (iii) reduction of drug
toxicity side effects; (iv) prolonged circulation time and in-
creased therapeutic effect; and (v) improved quality of life.
This research was, in part, supported by grants and con-
tracts from the United States National Science Foundation
(Grants No. ISTI-8961217 and ISTI-9112190) and the
United States National Cancer Institute, National Institutes
of Health (Grant No. 1R43-CA-58140, and Contracts No.
N43-CM-17019 and N44-CM-27122).
Castor, T.P. U.S. Patent No. 5,554,382, 1996; European Patent No.
Castor, T.P.; Chu, L. U.S. Patent No. 5,776,486, 1998; European
Patent No. 0,792,143, 2002.
Castor, T.P.: Supercritical Fluid Liposome Formulations presented
at a symposium entitled Liposomes and Vesicles: Fundamentals
and Applications at AIChE Annual Meeting, San Francisco, CA,
Reid, R.C.; Praustnitz, J.M.; Sherwood, T.K. The Properties of
Gases and Liquids, McGraw-Hill Book Company, New York,
Chu, L.; Laska, M.; Berrien, L.S.; Castor, T.P. Solubility of Phos-
pholipids in Supercritical Fluids presented at the 12th Symposium
on Thermophysical Properties, Boulder, CO, 1994.
Bird, R.B.; Stewart, W.E.; Lightfoot, E.N. Transport Phenomena,
John Wiley & Sons. Inc., New York, 1960.
Holmes, F.A.; Walters, R.S.; Theriault, R.L.; Forinan, A.D.; New-
ton, L.K.; Raber, M.N. Journal of National Cancer Institute, 1991,
Rowinsky, E.K.; Donehower, R.C. Journal of National Cancer
Institute, 1991, 83(24), 1778-81.
Arbuck, S.G.; Canetta, R.; Onetto, N.; Christian, M.C. Seminars in
Oncology, 1993, 20(4, suppl. 3), 31-9.
Castor, T.P. U.S. Patent 5,750,709, 1998.
Castor, T.P. U.S. Patent No. 5,440,055, 1995; European Patent No.
0,689,537, 1997; Chinese Patent No. ZL 94 1 91805.X, 2000; Ko-
rean Patent No. 10-311330, 2002; Japanese Patent No. 3,478,341,
Liebmann, J.E.; Cook, J.A.; Lipschultz, C.; Teague, D.; Fisher, J.;
Mitchell, J.B. Br. J. Cancer, 1993, 68, 104-9.
Hertzberg, R.P.; Caranfa, M.J.; Hecht, S.M. Biochemistry, 1989,
Burke, T.G.; Mi, Z.J. Medicinal Chemistry, 1994, 37, 40-6.
Burke, T.G.; Mishra, A.K.; Wani, M.C.; Wall, M.E. Biochemistry,
1993, 32, 5352-64.
Potmesil, M. Cancer Research, 1994, 54, 1431-9.
Emerson, D.L.; Besterman, J.M.; Brown, H.R.; Evans, M.G.; Leit-
ner, P.P.; Luzzio, M.J.; Shaffer, J.E.; Sternbach, D.D.; Uehling, D.;
Vuong, A. Cancer Research, 1995, 55, 603-9.
Pezzuto, J.M.; Das Gupta T.K.; Kim, D.S.H.L. U.S. Patent No.
Lee, K-H.; Kashiwada, Y.; Hashimoto, F.; Cosentino, L.M.; Ma-
nak, M. U.S. Patent No. 5,679,828, 1997.
Carlson, R.M.; Krasutsky, P.A.; Karim, R. U.S. Patent No.
Bringmann, G.; Saeb, W.; Assi, L.A.; Francois; G., Narayanan,
S.S.; Peters, K.; Peters, E-M. Planta Med., 1997, 63, 255.
Chowdhury, A.R.; Mandal, S.; Goswami, A.; Ghosh, M.; Mandal,
L.; Chakraborty, D.; Ganguly, A.; Tripathi, G.; Mukhopadhyay, S.;
Bandyopadhyay, S.; Majumder, H.K. Molecular Medicine, 2003,
Lin, H-M.; Yang, Z.; Chen, L-F. Biotechnology Progress, 1992, 8,
Castor, T.P.; Lander, A.D. U.S. Patent No. 6,465,168, 2002.
Castor, T.P.; Lander, A.D. European Patent No. 0,629,135, 2000;
Great Britain Patent No. 0,629,135, 2000; French Patent No.
0,629,135, 2000; Austrian Patent No. E194,921, 2001; German
Patent No. 69,329,108.7-08, 2001.
Castor, T.P.; Lander, A.D.; Cosman, M.D.; d’Entremont, P.R.;
Pelletier, M.R. U.S. Patent No. 5,877,005, 1999.
Castor, T.P.; Hong, G.T. U.S. Patent No. 5,380,826, 1995.
Received: February 16, 2005Accepted: March 15, 2005