Note: Descriptions are shown in the official language in which they were submitted.
CA 02448746 2003-11-26
WO 03/003007 PCT/US02/19736
METHODS AND COMPOSITIONS USEFUL FOR THE PREDICTION OF
BLOOD-BRAIN BARRIER PERMEATION
This invention relates to methods for predicting the
ability of compounds to cross the blood-brain barrier,
and to membrane compositions for use in such methods.
BACKGROUND OF THE INVENTION
To be effective as therapeutic agents, centrally
acting pharmaceuticals must cross the blood-brain barrier
(BBB). Conversely, to be devoid of unwanted central
nervous system (CNS) effects, peripherally acting
pharmaceuticals should demonstrate limited ability to
cross the BBB. In either case, the BBB permeability of a
pharmaceutical candidate needs to be known. However, the
experimental determination of blood-brain partitioning is
difficult, time-consuming, costly and unsuitable for
screening large collections of chemicals. A broadly
applicable method for predicting the BBB permeation of
pharmaceutical candidates at an early stage of discovery
would have a significant impact in pharmaceutical
research and development. Methods which produce reliably
predictable data related to BBB permeation for large
numbers of compounds at an early stage of the
discovery/development process are urgently needed.
Therefore, it is an object of this invention to
provide a robust, efficient and predictive method for the
in vitro determination of the BBB permeation capabilities
of a test compound such as a potential pharmaceutical
agent.
-1-
CA 02448746 2003-11-26
WO 03/003007 PCT/US02/19736
It is another object of this invention to provide a
membrane composition useful for the in vitro
determination of the BBB permeation capabilities of a
test compound.
It is a feature of this invention that the BBB
permeation determinative method and composition provide
high predictive value and high throughput efficiency.
It is another feature of this invention that said
method and composition are adaptable to standard
laboratory robotics platforms.
These and other objects and features of the
invention will become more apparent by the detailed
description set forth hereinbelow.
SUN~J'ARY OF THE INVENTION
The present invention provides a method for the
determination of the ability of a compound or mixture of
compounds to permeate the blood-brain barrier which
comprises measuring the rate said compound or mixture of
compounds passively diffuses through a porous filter
membrane impregnated with a mixture of brain polar lipid
extract and dodecane.
The present invention further provides a membrane
composition which comprises a porous solid support
impregnated with a mixture of brain polar lipid extract
and dodecane.
DETAILED DESCRIPTION OF THE INVENTION
Methods which can deliver reliable and predictive
data related to blood-brain barrier (BBB) permeation for
large numbers of compounds, i.e, 500-1,000 per day, at an
early stage of the discovery development process would
allow the rapid and inexpensive selection and
optimization of pharmaceutical candidates for desirable
brain penetration characteristics and may help in the
-2-
CA 02448746 2003-11-26
WO 03/003007 PCT/US02/19736
differentiation between active, paracellular and
transcellular processes. Known methods for predicting
BBB penetration include computational methods using
mathematical tools, cell culture methods using
endothelial cell cultures from animal origin, high
performance liquid chromotography (HPLC) using
immobilized. artificial membrane columns, measurement of
surface activity using critical micelle concentration
methodology, microdialysis techniques involving sampling
tissue from the brain of a living animal for external
HPLC analysis, the use of postmortem human brain
capillaries, and in vivo animal studies. None of these
known methods are entirely suitable for obtaining
economic, reliable and highly predictive BBB permeation
data for large numbers of test compounds.
Surprisingly, it has now been found that the ability
of a compound or a mixture of Compounds to permeate the
blood-brain barrier may be determined in a highly
predictive, reproducible, reliable, efficient and
economic manner in vitro by measuring the rate said
compound or mixture of compounds passively diffuses
through a porous filter membrane impregnated with a
mixture of brain polar lipid extract and dodecane.
Advantageously, the method of the invention is suitable
for use with standard laboratory robotics platforms.
The term passive diffusion, as used in the
specification and claims, designates a process of
transfer of individual molecules of a compound or mixture
of compounds across a semi-permeable membrane which is
brought about by random molecular motion and associated
with a concentration gradient. The modifier passive
refers both to the absence of external forces such as
increased pressure, reduced pressure, gravity or the like
and to the lack of active processes such as metabolism,
the use of transporters or the like.
-3-
CA 02448746 2003-11-26
WO 03/003007 PCT/US02/19736
The inventive method is a simple, high through-put
physico-chemical method which uses a very small amount of
sample, generally less than 0.1 mg, to accurately predict
the ability of said sample to permeate the blood-brain
barrier. In actual practice, a solution of known
concentration of a compound or mixture of compounds in a
buffer solution is separated from a buffer solution
containing 0% concentration of said compound or mixture
of compounds by a porous filter membrane impregnated with
a mixture of brain polar lipid extract and dodecane in
such a way that the surface of each buffer solution is in
contact with opposite sides of said impregnated membrane.
After a measured period of time, the concentration of
said compound or mixture of compounds is determined for
each buffer solution and the rate of diffusion is
calculated.
In one embodiment of the invention, a 96-well plate
filled with solutions of a known concentration of test
sample in a buffer solution (donor) may be covered with a
96-well filter plate wherein the porous filter membrane
is impregnated with a mixture of brain polar lipid
extract and dodecane and the wells are filled with a
buffer solution containing 0% concentration of test
sample (acceptor) such that the surface of each buffer
solution is in contact with opposite sides of said
impregnated filter membrane; after a measured period of
time, the donor plate~and acceptor plate are separated,
the concentration of sample in each buffer solution is
determined and the rate of permeation is calculated.
In another embodiment of the invention, a high
through-put permeability instrument such as the PSR4p
instrument manufactured by pION Inc., Woburn, MA may be
utilized. In this embodiment, a parallel artificial
membrane assay (PAMPA) technique is employed using as
-4-
CA 02448746 2003-11-26
WO 03/003007 PCT/US02/19736
artificial membrane a porous filter membrane impregnated
with a mixture of brain polar lipid extract and dodecane.
Buffer solutions suitable for use in the method of
invention include any conventional buffer solution of
about pH 6.0-8.0, preferably about pH 7.2-7.6 and more
preferably about pH 7.4.
Accordingly, the method of the invention
demonstrates higher predictability for blood-brain
barrier permeability with higher through-put capacity,
lower cost and without the sacrifice of living animals
than those methods currently known in the art.
The present invention also provides a membrane
composition which comprises a porous solid support
impregnated with a mixture of brain polar lipid extract
and dodecane. The mixture of brain polar lipid extract
and dodecane suitable for use in the inventive
composition may be about 10% wt/vol to 30o wt/vol,
preferably about 15o wt/vol to 25o wt/vol, more
preferably about 19% wt/vol to 21o wt/vol of brain polar
lipid extract in dodecane.
A porous solid support suitable for use in the
inventive composition includes any commonly used porous
material such as that used in the 96-well filter plates,
for example polyvinylideneflouride or an equivalent
thereof, preferably polyvinylideneflouride.
Brain polar lipid extracts suitable for use in the
composition of the invention may be those brain polar
lipid extracts, either synthetic or natural, which can be
found in the literature or which are commercially
available such as porcine, ovine, bovine or the like,
preferably porcine brain polar lipid extract.
The composition of the invention may be prepared by
impregnating the porous solid support with a mixture of
brain polar lipid extract and dodecane at a level of at
-5-
CA 02448746 2003-11-26
WO 03/003007 PCT/US02/19736
least 4~,L/38mm2, preferably about 4~,L/38mmz, of said
mixture per area of porous solid support.
For a more clear understanding of the invention, the
following examples are set forth below. These examples
are merely illustrative and are not understood to limit
the scope or underlying principles of the invention in
any way. Indeed, various modifications of the invention,
in addition to those shown and described herein, will
become apparent to those skilled in the art from the
following examples and the foregoing description. Such
modifications are also intended to fall within the scope
of the appended claims.
-6-
CA 02448746 2003-11-26
WO 03/003007 PCT/US02/19736
EXAMPLE 1
Comparative Evaluation of the Predictability for Blood
Brain Barrier Permeability of the Inventive Method Using
Known Compounds
In this evaluation, 30 literature compounds for
which the blood-brain barrier permeability is known (P.
Crivari, et al, Journal of Medicinal Chemistry, 2000, 43,
2204-2216) are each dissolved in DMSO at a Concentration
of 5 mg/mL to give a stock solution for each compound. A
100 ~.L volume of each stock solution is placed into one
well of a 96-well plate, and the well plate is placed in
a PSR4p permeability analyzer manufactured by pION, Inc.,
Woburn, MA. A 10 ~,L volume of stock solution is
robotically added to a deep 96-well plate containing 2.0
mL of pH 7.4 buffer solution. The resultant mixture is
mechanically stirred to form the donor solution. A 200
~.L volume of each donor solution is placed robotically
into 3 wells of a 96-well plate to afford the donor
plate. A vial containing 20 mg of porcine polar brain
lipid extract manufactured by Avanti Polar Lipids, InC.,
Alabaster, AL, dissolved in 1 mL of dodeCane is placed in
the reservoir of the PSR4p instrument. A 4 ~.L volume of
this brain lipid solution is placed on the filter surface
of each well of a 96-well microtiter filter well plate
manufactured by Millipore Corp., Bedford, MA, wherein the
the filter is a porous (0.45 ~,m) polyvinylidenefluoride
material approximately 104 ~,m in thickness. The
microtiter plate is then manually placed on an orbital
shaker for 1 minute to yield a 96-well microtiter plate
having a filter membrane permeated with a 20o wt/vol
solution of porcine polar brain lipid extract in
dodecane. A 200 ~.L volume of pH 7.4 buffer is
robotically inserted into the wells of the thus-prepared.
CA 02448746 2003-11-26
WO 03/003007 PCT/US02/19736
microtiter 96-well filter plate to give a receptor plate,
this receptor plate is placed on the donor plate to form
a sandwich and allowed to stand at ambient temperature
for 18 h.
A plate washer manufactured by Tecan, Hombrechin-
tikon, Switzerland is then used to prepare a UV
transparent 96-well plate (W plate). The receptor plate
is removed from the sandwich, a 150 ~.L volume of the
receptor solution is robotically placed in the W plate,
and the W absorption at 190-500 nm of each receptor well
is recorded. The UV plate is then replaced in the plate
washer, washed again and a 150 ~.L volume of the donor
solution is robotically placed in the UV plate and the UV
absorption at 190-500 nm of each donor well is recorded.
The rate of passive diffusion is measured as linear
velocity of permeation (Pe). The Pe is calculated for
each compound using version V1.4 of the PSR4p software
from pION Inc. Two standards, verapamil and
theophylline, are used for each microtiter plate. BBB
permeability is designated as CNS+ for Pe values >4 x 10-6
cm/sec and as CNS- for Pe values <2 x 10-6 cm/sec. The
data are averaged and the results are shown in Table I.
Table I
Test Pe CNS
Comt~ouad ( 10-scm/ s Eval Known2
)
Alprazolam 5.44 + +
Caffeine 1.30 - +
Chlorpromazine 6.36 + +
Clobazam 16.85 + +
Clonidine 5.31 + +
Desipramine 11.89 + +
_g_
CA 02448746 2003-11-26
WO 03/003007 PCT/US02/19736
Table I (cont'd)
Test Pe CNS
Compound ( 10-scm/ Knownz
s ) Eval
Diazepam 16.41 + +
beta-Estradiol 11.61 + +
Imipramine 13.12 + +
Oxazepam 10.12 + +
Progesterone 9.28 + +
Promazine 8.84 + +
Testosterone 16.75 + +
Thiopental 18.13 + +
Aldosterone 1.19 - -
Astemizolel 10.66 + -
Atenolol 0.84 - -
Hydrocortisone 1.99 - -
Dopamine 0.16 - -
Enoxin 0.90 - -
Isoxicam 0.25 - -
Lomefloxacin 1.19 - -
Loperamide 0.00 - -
Corticosteronel 5.13 + -
Norfloxacin 0.14 - -
Ofloxacin 0.80 - -
Piroxicam 2.53 - -
Terfenadine 0.00 - -
Tenoxicam 0.14 - -
Cimetidine 0.00 - -
1 These compounds were involved in active processes, i.e.,
carrier mediated transport(caffeine), Pgp efflux
(corticosterone) and rapidmetabolism (astemizole).
2 Crivori, P., et al., Journal of Medicinal Chemistry,
2000, 43, 2204-2216.
As can be seen by the data on Table I, the inventive
method demonstrates 90% accuracy for all 30 test
-9-
CA 02448746 2003-11-26
WO 03/003007 PCT/US02/19736
compounds including active transport processes. For
passive diffusion, the inventive method demonstrates 1000
accuracy.
EXAMPLE 2
Comparative Evaluation of the Predictability for Blood
Brain Barrier Permeability of the Inventive Method Using
Experimental Compounds
Using essentially the same procedure described in
Example 1 and substituting experimental compounds
obtained from three separate CNS projects, the Pe values
are determined and compared to the blood-brain barrier
permeability as determined by standard rat brain assay
methodology or biological endpoints from in vivo studies.
The results are shown in Table II.
Table II
Test Pe CNS CNS
Com~aound ( 10-scm/ s Eval In vivo
)
A 6.52 + +
B 6.07 + +
C 18.99 + +
D 11.61 + +
E 9.~8 + +
F 10.62 + +
G 15.11 + +
H 5.20 + +
I 13.75 + +
0.40 - -
-10-
CA 02448746 2003-11-26
WO 03/003007 PCT/US02/19736
Table II, Contd
Test Pe CNS CNS
Compound ( 10-scm/ s ) Eval . In vivo
K 0.13 - -
L 0.05 - -
M 0.03 - -
N 0.08 - -
As can be seen from the data on Table II, the
inventive method demonstrates 100% accuracy.
-11-
