Note: Descriptions are shown in the official language in which they were submitted.
W O 94/11021 2 1 4 7 ~ ~ 9 PC~r/US93/10222
BRADYK~NIN ANTAGONISTS
The present invention relates to
pharmaceutically effective heterodimers comprising a
bradykinin antagonist (BKAn) component covalently
linked to another different pharmacophore component.
Related A~lications
In prior application WO 92/17201 published
October 15, 1992, there i9 described bradykinin
antagonist dimers of the type:
X(BKAn)2
where BKAn represents a bradykinin antagonist
peptide and X is a linking group which joins the two
BKAn components at points intermediate to their
ends. The BKAn substituents may be the same or
different. However, also described are certain
heterodimers involving the linkage of a BKAn peptide
and another peptide of different receptor activity
through the linking group X, e.g. an NKl or NK2
antagonist peptide or a mu-opioid receptor agonist
peptide. Such heterodimers are particularly useful
w-here there is a close relationship between the
activities of concern. Thus, it is known that in a
number of pathophysiologically important processes,
there is an intimate interaction of inflammatory and
neurogenic mediators. This occurs, for example, in
both pain secondary to tissue trauma (accidental and
post-operative) as well as in asthma. In both
situations, there is a complex interplay of tissue
and plasma derived media~ors (such as kinins acting
at BK2 receptors) and neuronally derived factors
such as substance P (NKl receptors) and neurokinin A
(NK2 receptors). There are also locally acting
neuronal receptors of the mu-opiate class that when
WO94/11021 PCT/US93/1022~
21~78~9
stimulated can inhibit the release of the neurogenic
peptides regardless of type (substance P, neurokinin
A, neurokinin B, cholecystokinin, CGRP, etc.).
Given the interaction of these as well as other
inflammatory and neurogenic mediators, no one agent
is likely to be universally efficacious in
ameliorating the symptoms attendant to the
pathophysiology. The heterodimers described in the
above-noted applications are directed towards
addressing these problems with single agents
possessing dual selectivity. Other advantages of
such heterodimers will also be appreciated by those
in the art
Brief Description of the Invention
The pre~ent invention, in it~ broadest a~pect,
is concerned with heterodimers obtained by linking a
BKAn peptide to another pharmacophore which is not a
bradykinin antagonist, i.e. which may be either a
peptide as described in WO 92/17201, or a non-
peptide, effective against a different, non-
bradykinin component responsible, for example, for
pain and/or the inflammatory process, or other
problems related to or occurring in concert with the
activity of the k; ni n~ . The resulting compounds are
~dual action" compounds that are capable of
interacting with two receptor populations or,
alternatively, with a receptor and an enzyme. This
is not intended to suggest that the single molecule
will engage two receptors or a receptor and an
enzyme simultaneously; only that the molecule is
capable of interacting with either one of two
receptor types or with a single class of receptors
and/or an enzyme. The overall pharmacological
effect of administering such a compound in an
~ WO94/11021 21 ~ 7 8 6 9 PCT/US93/10222
appropriate dose, however, is at least the summation
of the two types of activities. The compounds can
be designed to remain intact or they can be designed
to be dissociated into two separate molecules each
retaining its own identifiable activity.
The heterodimers o the invention can be
structurally represented as follows:
(Y)(X)(BKAn)
where BKAn is a bradykinin antagonist peptide; X is
a linking group and Y is a peptide or non-peptide
pharmacophore which is not a bradykinin antagonist
and demonstrates activity towards a different
receptor or enzyme than the BKAn component,
preferably one related to pain or the inflammatory
lS process.
The present heterodimers offer the possibility
of providing a wider spectrum of treatment for pain
and inflammation. It is a generally held opinion
that in inflammatory states, regardless of severity,
the likelihood that a single agent or mediator is
completely responsible for all of the clinical
manifestations of the syndrome being addressed is
extraordinarily small. A corollary to this is that,
given the role of bradykinin in inflammatory
pathophysiology, any combination therapy used in the
treatment of inflammatory disorders should include
bradykinin antagonism as part of its overall profile
of action. Broad spectrum and potent non-specific
therapies (such as the use of steroids in asthma)
while perhaps efficacious, carry with them the
burdens of undesired and potentially serious side
effects and toxicities.
In many cases, two discrete mediators are known
to act synergistically and to account for ah
overwhelming proportion of the clinically important
WO94/11021 PCT/US93/10222
21~78~9 ~
manifestations of the disease being treated. Such
is the case, for example, with substance-P acting at
NKl receptors and bradykinin acting at BK2 receptors
in the contexts of asthma and post-traumatic or
post-operative pain. Similarly, neutr~hil elastase
as one of the more important down stream effectors
of inflammation and bradykinin as one of the more
important initiating and sustaining inflammatory
mediators also can be viewed as being synergistic in
their actions.
The concept of providing homodimers of
pharmaceutically active materials to improve such
characteristics as metabolic stability, selectivity
and receptor binding has previously been described
~or other systems. This prior work has included t~e
dimerization o~ peptide agonists and antagonists in
order to increase potency and/or duration of action.
See, Caporale et al, Proc. 1Oth American Peptide
SYm~., Pierce Chemical Co., Rockford, IL 449-451
(1988) and Rosenblatt et al, European Patent
Application No. EP 293130A2. Thus, dimerization of
peptide agonists has been disclosed for
enkephalins/endorphins (Shimohigashi, Y., et al,
BBRC, 146, 1109-1115, 1987); substance P (Higuchi,
Y., et al, E.J.P., 160, 413-416, 1989); bradykinin
(Vavrek, R. and Stewart, J., J. Proc. 8th Amer.
Pe~t. SVmD., 381-384, 1983); neurokinin A ~ B,
(Kodama, H., et al, E.J.P., 151, 317-320, 1988);
insulin (Roth, R.A., et al, FEBS, 170, 360-364,
1984) and atrial natriuretic peptide (Chino, N., et
al, BBRC, 141, 665-672, 1986). Dimerization of
antagonists has been shown for parathyroid hormone
(Caproale, L.H., et al, Proc. 10th Amer. Pe~t.
Symp., 449-451, 1987)). However, the literature has
not disclosed heterodimers comprised of a bradykinin
WO94/11021 PCT/US93/10222
~ 21~786~ .
antagonist and a different pharmacophore as
contemplated herein.
Detailed Descri~tion of the Invention
Numerous bradykinin antagonist peptides are
known in the art and any of these may be used for
present purposes to provide the BKAn substituent of
the present dimers. One of the more potent
bradykinin antagonists in ~itro is the peptide
having the formula:
io D_ARG0-Argl-pro2-Hyp3-Gly4-Phe5-Ser6-D-Phe'-Leu8-
Arg9
See Regoli et al, Trends in Pharmacoloqical Science,
11:156-161 (1990). This peptide is referred to
herein for convenience as CP-0088.
While CP-0088 is a convenient BKAn to use,
those in the art will appreciate that other
available or known bradykinin antagonist peptides
can also be used for present purposes. A wide
variety of such bradykinin antagonist peptides have
been disc-losed in the recent patent literature and
any of these can be used for present purposes. See,
for example, EP-A-0334244 (Procter and Gamble) which
discloses nona- and larger bradykinin antagonist
peptides in which certain amino acid residues are
modified. EP-A-0370453 (Hoechst) and WO 89/01780
and WO 89/01781 (Stewart et al) also describe
bradykinin antagonist peptides. ~one of these
patent publications appears to show dimers as
contemplated herein. However, as noted, the
peptides of these publications can be used in the
practice of the present invention.
WO94/l1021 PCT/US93/1022~
21~78~9
Any linking group X may be used for present
purposes to chemically or covalently link together
the BKAn and Y components provided this does not
lnterfere with the activity of the components BKAn
and Y. The linking group may be inorganic (e.g. -S-
) or organic and may be selected so as to hydrolyze
or otherwise dissociate in order to liber~ate the two
active components BKAn and Y in ~ivo. -~
Alternatively, the linking group may be such that
the heterodimer remains intact when used.
Conveniently the linking group X can include an
-S- atom derived by reacting a sulfhydryl group on
the BKAn peptide chain with the other pharmacophore
component. This can be accomplished by reaction
involving a cysteine (Cys) sul~hydryl group within
the peptide chain, i.e. intermediate the ends o~ the
peptide. This may require initially modifying the
starting BKAn peptide so that it includes a Cys
group in the appropriate position in the peptide
chain. For example, CP-0088 may be modified by
replacing the Ser in the 6-position with Cys (such
modified CP-0088 being called CP-0126 hereinafter)
to provide for convenient linking to the other
pharmacophore through the Cys sulfhydryl.
CP-0126 can be structurally illustrated as
follows:
D-ARG-Arg-Pro-Hyp-Gly-Phe-Cys-D-Phe-Leu-Arg
SH
In abbreviated fashion, the formula may be stated
as:
DR-R-P-J-G-F-C-DF-L-R
WO94/11021 21 ~ 78 6 g PCT/US93/10222
Using Cys as the position of attachment, the
linking group X then includes the -S- of the
cysteine sulfhydryl. This may be the entire linking
group X (as in a disulfide based dimer) or only a
part thereof. Thus, for example, the linking group
may comprise a bissuccinimidoalkane such as
bissuccinimidoh~x~ne joined at its end to the BKAn
and Y components. These and other linking groups
are disclosed in applicant's related application and
any of these may be used for present purposes.
Other linking groups X, some of which do not require
or contain an -S- atom, can be derived from the six
families of compounds listed below which can be
generically categorized as amino acid analog linkers
or maleimide-based linkers. These linkers are
included as examples only and are not intended to be
totally inclusive of all potential linking moieties:
PCI/US93/1022
WO 94/11021
2 8
CLASS I (Amino Acid Analogs)
* - "D" or "L" Configur~tion
A) NH2 Rl ' Rl & R2 = -H, -CH3, -CH,-CH3
or --Rl/R2-- = CYCLO ALK~
* C~I--(CH2)~ C\ R3 = -OH, -CO2H, or -NH2
CO2H R3 x=1-12
y= 1 -4
B)1 2 f~
* CH (CH2)y N~<N (CH2)~ C
O R3
CO2H
C)~ H2 0~,
* C'~I (CH2)yN N--(CH2)~ C\
CO2H R3
D)Nl~2 ~O R\~ / R2
* CH (CH2)y~ N N (CH2),C C\
CO2H O R3
E)IH2 ~ 4 R~ / R2
* CH (CH2)y ,N~N--(CH2)~ C\
CO2H R3
CLASS 11 (Maleimide B~sed Linlcers)
~y R I R
F) [~N (CH~)~ C/
O R~
~ WO94/11021 PCT/US93/10222
21478~
The amino acid analog linkers (Class I) can be
directly incorporated into the peptide chain of the
BKAn and then used to form esterase stable or labile
heterodimers with the geminal pharmacophore
(component Y). Alternatively, the maleimide based
linker can be reacted with the desired pharmacophore
and then conjugated to a sulfhydryl containing
peptide. Finally, linkers from any of these
families of compounds which contain -CO2H as the R3
moiety can be reacted with another linker from these
classes of compounds to form esterase labile (R3= -OH
containing linkers) or esterase stable (R3= -NH2
containing linkers) which can then be used to form
the desired peptide/non-peptide heterodimer. R~ and
R2 can be varied so as to provide for completely
l~nhln~ered or significantly hln~ered access to the
carbonyl carbon of an ester based linking element so
that the rate of i~ vivo hydrolysis of said ester
can be controlled as desired.
Certain of the linker-modified BKAns or
pharmacophores used herein to prepare the dimers of
the invention are themselves novel and constitute a
further aspect of the invention.
The component Y of the present heterodimers may
be any peptide or non-peptide pharmacophore, other
than a bradykinin antagonist, which demonstrates
activity towards a different (non-bradykinin)
component related to pain and/or the inflammatory
process so as to provide dual action compounds that
- 30 are capable of interacting independently with two
different receptor populations or a receptor and an
enzyme. Thus, for example, the component Y may be a
non-peptide mu-opioid receptor agonist, e.g.
morphine or one of its derivatives such as oxycodone
or oxymorphone.
WO94/11021 PCT/US93/10 ~
21~ ~8~9
Indomethacin is a useful choice for the Y
component when cyclooxygenase inhibition (COI) is
desired. However, any of the ~ther conventional
non-steroidal anti-inflammatory agents, such as
aspirin, ibuprofen, naproxyn or the like can be.
used. In this case, the BKAn/COI heterodimer may
need to be hydrolyzed in order to obtain in vivo COI
activity as cyclooxygenase is generally considered
to be an intra-cellular enzyme.
Where neutrophil elastase inhibition is
required, this may be an active ester, e.g. 2-
phenyl-alkanoate ester. There may also be used a
heteroaryl alkanoate esterase inhibitor. A
preferred neutrophil elastase inhibitor for use
herein as component Y is identi~ied ~elow as CE-
1218. This is believed to be a new compound and
constitutes a further feature of the present
invention.
Other types of elastase inhibitors which may
also be used as component Y, include fluoromethyl
ketones, phosphonates, benzoxazoles, beta-lactams,
etc.
As n~ted earlier, component Y may comprise a
peptide or non-peptide inhibitor having a desired
activity other than bradykinin antagonist activity.
However, the Y component is preferably selected to
provide activity against receptors or enzymes which
have a common or close relationship to the activity
of bradykinin, e.g. the treatment of pain or
inflammation. The rationale for using combinations
of a BKAn with a mu-opioid receptor agonist,
neutrophil elastase inhibitor, cyclooxygenase
inhibitor or NK1 or NK2 receptor antagonist in
various conditions is discussed below for purposes
of illustration.
~ W O 94/11021 21 ~ 78 6 9 PC~r/US93/10222
BKAn/mu-oPioid recePtor aqonists
C-Fiber afferents are known to mediate both the
f sensation of pain as well as the neurogenic
component of inflammation. These afferent neurons
release a variety of neuropeptides in response to
specific and non-specific stimuli in both the
central nervous system (CNS) as well as in the
peripherally innervated tissues. Some of these
neuropeptides include: substance-P, neurokinin A,
neurokinin B, calcitonin gene related peptide
(CGRP), cholecystokinin (CCK), vasoactive intestinal
polypeptide (VIP), and neuropeptide Y, among other
neurotransmitters. To add to this complexity,
different C-fibers appear to contain different
amounts and/or ratios of these neuropeptides
depending on the tissue innervated. All of these
peptides have been shown to play contributory roles
in the various neurogenic processes that have been
impIicated in numerous diseases and clinical
syndromes. In fact, specific antagonists to these
peptides are being developed as potential
therapeutics by a variety of pharmaceutical
companies.and independent research laboratories.
One apparently common feature among this
otherwise diverse group of neurons is that they all
have mu-opioid receptors that modulate the release
of these neuropeptides. Both the endogenous
enkephalins as well as other exogenously
administered small molecular weight compounds such
as morphine, oxymorphone, fentanyl and their
derivatives will inhibit the release of the
neuropeptides from peripheral C-fibers by acting as
mu-opioid receptor agonists locally (at terminal mu-
opioid receptors in the periphery) and in the CNS.
This inhibition is independent of both the
WO94/11021 PCT/US93/10 ~
2 1 ~ 7 8 ~ 9
constellation of peptides contained in the specific
C-fiber as well as the stimulus causing their
release.
As a result, one important class of compounds
considered to have a particularly good profile of
activities for the treatment of conditions that are
produced by combined humoral and neurogenic
processes are BKAn/mu-opioid receptor agonist
heterodimers. These compounds would be expected to
lO attenuate or block both the humoral component of the
inflammatory process as represented by the k;n;n.c as
well as the neurogenic aspects of inflammation
produced by the.release of the neuropeptides. In
addition, one of the limiting aspects of the use of
existing mu opioid agonists iq their propensity to
produce qedation, con~usion, and a depressed
respiratory drive, not to mention their potential
for the development of addiction and/or tolerance in
the.patients being treated with these agents. These
undesirable aspects of mu-opioid receptor agonists
are due to their ability to easily penetrate the
CNS. BKAn/mu-opoid receptor agonist heterodimers,
however, should not penetrate the CNS due to the
highly cationic nature of the BKAn. Consequently,
mu-opoid receptor agonist activity should be limited
to the periphery and should result in a
substantially reduced side effect/toxicity profile
for these types of compounds.
BKAn/Neutro~hil Elastase Inhibitor (NEI)
As previously mentioned the control of both
systemic and local inflammatory responses may
require lnterventions in more than one inflammatory
pathway. In particular, the ability to block the
activity of a primary mediator responsible for the
~ WO94/11021 2 1 ~ 7 8 6 9 PCT/US93/10222
initiation and maintenance of the inflammatory
process (such as bradykinin) and a primary final
pathway effector responsible for actual tissue
degradation and injury (such as neutrophil elastase)
may be a key to "single drug therapy" of sepsis or
other severe inflammatory conditions requiring
parenteral therapy or for the treatment of
inflammatory dermatologic or dental/periodontal
conditions.
Heterodimers containing combined BKAn/NEI
activities can be designed to remain intact or to be
dissociable as both targets (bradykinin receptors
and neutrophil elastase) are extracellular in
nature. However, should dissociation of the two
active pharmacophores be desired, linking moieties
tethering the two active components of the
heterodimer can be designed to be hydrolyzed by, for
example, plasma hydrolases. These types of
dissociable or hydrolyzable heterodimers are
discussed herein.
BKAn/Cvcloox~enase Inhibitor (COI)
A large proportion of the biological activity
of bradykinin is interwoven with the generation of
prostagl~n~i~q. For example, much of the
hyperalgesia associated with inflammatory pain
appears to be dependent on the generation of certain
prostagl~n~inq both by the injured tissues and by
the C-fibers themselves. In the latter case,
bradykinin and substance-P appear to be the primary
stimuli for these "second messengers". The local
generation of prostaglandins by the injured tissues
is bradykinin independent. This interaction of
peptide pro-inflammatory mediators and
prostaglandins occurs in other settings as well and
W094/11021 2 1 ~ 7 8 6 9 PCT/US93/10 ~ ~
can also be considered a target for dual action
compounds. Heterodimers cont~Lning combined
BKAn/COI activities may need to undergo in vivo
dissociation of the respective pharmacophores as
S cyclooxygenase is an intraGellular enzyme and
functional bradykinin receptors are limited to the
external plasma membrane.
BKAn/NKl-ReceDtor Antaaonist (NKlAn)
Bradykinin and substance-P are known to act
synergistically in the initiation and maintenance of
the inflammatory and neurogenic components of both
asthma and a variety of painful conditions. In both
of these situations, bradykinin is one of the more
potent, if not the most potent, agents capable o~
stim~lating C-~ibe~ sensory a~ferents that mediate
peripheral pain and/or the sensation of cough and
dyspnea in asthma. These neurons, regardless of the
primary stimulus, will release substance-P which
amplifies and augments the activity of bradykinin
and other stimuli at the sensory nerve endings where
these stimuli are acting. This "one/two punch'~ of
initial &timulus followed by local amplification is
well documented and has significant implications for
the success or failure of any single intervention.
By targeting both components of these processes with
a single compound, it is possible to provide a
dually-specific agent which is superior than mono-
specific agents used alone and both easier and
cheaper to use than combination therapies.
BKAn/NK2 Antaaonist (NK2An)
sradykinin~s ability to produce acute bronchial
smooth muscle constriction is at least partially
dependent on the release of neurokinin A by the same
~ WO94/11021 21 ~ ~8 69 PCT/US93/10222
C-fibers that release substance-P. Neurokinin A
exerts its effect via ~K2 receptors on the bronchial
smooth muscle: However, more than just bradykinin
can release neurokinin A from these neurons and, as
a result, a dually-specific antagonist with com~ined
BK2/NK2 antagonist activity should provide better
overall amelioration of bronchoconstriction in the
asthmatic patient than any other single agent.
The heterodimers of the invention may be
io prepared in generally the same manner as described
in WO 92/17201. Normally this involves adding the
linking group X to the BKAn component at an
appropriate position along the peptide chain
followed by joining the non-peptide pharmacophore to
the BKAn through the linking group. Alternatively,
the linking group may be added to the non-peptide
pharmacophore and the BKAn thereafter joined to the
linker-modified pharmacophore. Representative
procedures are described below although it will be
recognized that various modifications may be used.
The invention is illustrated but not limited by
the following examples:
Examples 1-4 (BRAn/mu-opioid agonist~
Four different peptide/opiate heterodimers
(designated CP-0477, CP-0488, CP-0494 and CP-0499)
were made in order to illustrate the invention.
Three of these compounds were made using CP-0126
(DR-R-P-J-G-F-C-DF-L-R) and the fourth used CP-0347
(DR-R-P-J-G-Thi-C-DTic-Oic-R). Similarly, two
different opiates (oxycodone and oxymorphone) and
two different linker chemistries were used to
provide the respective heterodimers as follows:
,
WO94/11021 PCT/US93/103Z~
21~786~ ~
.
ExamPle Compound # PePtide OPiate
CP-0477 CP-0126 Oxycodone
2 CP-0488 CP-01 ~6 Oxycodone
3 CP-0494 CP.-.~126 Oxymorphone
4 CP-0499 CP-0347 Oxymorphone
The heterodimers CP-0477, CP-0488, CP-0494 and
CP-0499 were prepared as detailed hereinafter with
reference to the accompanying Figure l:
Preparation of Com~ound I:
Oxycodone hydrochloride (0.182g, 0.52 mmol),
acetic acid (0.475 ml, 8.3 mmol), S-benzyl
cyst~m;n~ (0.174g, 1.04 mmol) and methanol (5 ml)
were combined and stirred at room temperature ~or an
hour. Sodium cyanoborohydride (95~, 0.033g, 0.52
mmol) was added, and the reaction stirred at room
temperature for 24 h. The mixture wa~ ~ôn~entrated
in vacuo. The resulting oil was dissolved in ethyl
acetate and the ethyl acetate fraction was washed
with saturated sodium bicarbonate solution, dried
over magnesium sulfate and evaporated in vacuo. The
crude material was chromatographed on a silica
column and eluted with EtOAc ! EtOAc-MeOH t9:l, v/v)
and EtOAcMeOH-Et3N (9:l:0.2, v/v/v) successively
Compound I was isolated as an oil in 25.0~ (59.0 mg)
yield.
Pre~aration of Com~ound II (CP-0477):
I (0.059g, 0.127 mmol) was dissolved in 2 mL
dry tetrahydrofuran, and was transferred to a oven
dried three-necked l00 ml flask. The flask was
fitted with a dewar condensor, a nitrogen source and
an ammonia inlet. Approximately l0 ml of ammonia was
condensed into the flask maintained at -78C. Small
~ W094/11021 2 I 4 78 6 9 PCT/US93/10222
pieces of sodium were added until the intense blue
color was maintained and then quenched after 40
seconds with ~olid ~mmonium chloride. The reaction
mixture was allowed to warm to room temperature and
the ammonia boiled off through a bubbler, methanol
(25 ml x 3) was added and evaporated in vacuo. The
thiol isolated was dissolved in a minimum quantity
of DMF (N,N-dimethyl formamide, 2 ml). Compound X
(approximately 0.3 equiv) was dissolved in tris
buffer (0.5 M, 4.0 ml) and added to the DMF solution
and then stirred for 17h. The crude mixture was
purified on a reverse phase Vydac C-18 HPLC column
using the gradient 15 - 40~ CH3CN in water, 0.1~
constant TFA, over 20 minutes. Retention time was
16.0 minutes. 26.4 mg of II was isolated as a white
powder on lyophillization.
AnalYsis:
The mass spectra was run on a Finnigan Lasermat
Mass Analyzer.
calulated molecular weight--1916
observed molecular weight--1918
Amino Acid Analvsis:
Gly 1.02 (1), Arg 3.14 (3), Pro 1.01 (1), Leu 0.97
(1), Phe 1.92 (2) and Hyp 0.94 (1).
PreDaration of Com~ound III:
To the mixture of oxycodone hydrochloride (1.0
g, 2.84 mmol) and ammonium acetate (2.2 g, 28.4
mmol) dissolved in methanol (10.0 ml) was added a
methanolic (4.0 ml) solution of NaCNPH3 (0 18 g, 2.84
mmol). The resulting solution was adjusted to pH 7.0
with concentrated hydrochloric acid, stirred for
W O 94/11021 PC~r/US93/102 ~
214~869
18
17h, and acidified to pH 1.0~with concentrated
hydrochloric acid. The solYent was removed in vacuo
and the rem~; n; ng material was dissolved in water.
The aqueous layer was extracted with chloroform,
adjusted to pH 9.0 with 10~ sodium carbonate
solution, saturated with NaCl and extracted with
chloroform. The chloroform layer was dried over
magnesium sulfate and evaporated in vacuo. The crude
oil was purified by silica gel chromatography and
eluted with EtOAc, EtOAc-MeOH (9:1, v/v), EtOAc-
MeOH-Et3N (9:1:0.3, v/v~v) successively. Compound III
was isolated as an oil in 47 0~ (0.42g) yield.
PreDaration of Compound IV:
BOC-Glycine (0.16g, 0.91 mmol), HOBt (0.125g,
o.sl mmol) and 1-(3-dimethylaminopropyl)
-3-ethylcarbodiimide hydrochloride (EDC) ( 98 0~,
0.18g, 0.91 mmol) were dissolved in DMF (2.0 ml) and
stirred at 0C for an hour. The amine III (0.24g,
0.76 mmol) dissolved in DMF (3.0 ml) was added to
the reaction mixture, the reaction mixture was
warmed to room temperature and stirred for 17h. DMF
was removed in vacuo and the resulting material was
dissolved in ethyl acetate. The ethyl acetate layer
was washed with saturated sodium bicarbonate
solution, brine and dried over magnesium sulfate.
The organic layer was evaporated in vacuo and the
crude mixture was flash chromatograhed on a silica
gel column and eluted with EtOAc-MeOH-Et3N
(9.5:0.5:0.3, v/v/v). Compound IV was isolated as an
oil in 82.0~ (0.29g) yield.
Pre~aration of Compound V:
The BOC protecting group was removed off the
compound IV with TFA (5.0 ml) in methylene chloride
~ WO94/11021 21~7869 PCT/US93/10222
19
(5.0 ml). Methylene chloride was removed in vacuo
and the residue was stripped with methylene chloride
(20 ml x 3) and then with triethyl amine (3 ml x 3).
3-S-benzyl mercapto propionic acid (0.15g, 0.75
mmol), EDC (0.lSg, 0.75 mmol), HOBt (0.103g, 0.75
mmol) and Et3N (0.35 ml, 2.48 mmol)were dissolved in
DMF (5.0 ml) and stirred at 0C for an hour. The
solution of the amine (0.23g, 0.62 mmol) in DMF (3.0
ml) was added to the reaction mixture. The reaction
mixture was warmed to room temperature and stirred
for lih. DMF was evaporated in vacuo and the residue
was dissolved in EtOAc. The EtOAc layer was washed
with 10~ Na2CO3, brine, dried (over MgSO4) and
evaporated in vacuo. The crude material was purified
on a flash silica gel column and eluted with
EtOAc-MeOH-Et3N (9:1:0.3, v/v/v). Compound V was
isolated as an oil in 60.0~ (0.205g) yield.
Preparation of Com~ound VI (CP-0488):
32.0 mg (0.057 mmol) of V was deprotected using
the procedure described for II and the thiol
isolated was reacted with compound X (0.073g, 0.048
mmol) in tris buffer. The crude mixture was purified
using the procedure for II. Retention time of the
product was 16.82 minutes. 9.5 mg (10~) of VI was
obtained as a white powder on lyophillization.
Mass sDectral Data:
calulated molecular weight 2002
observed molecular weight 2004
Amino Acid AnalYsis:
Gly 1.76 (2), Arg 3.19 (3), Pro 1.06 (l), Leu o.99
(1), Phe 2.06 (2), Hyp 0.95 (1).
,
WO94/11021 PCT/US93/102~
2~478~9
Pre~aration of Compound VII:
Oxymorphone hydrochloride (0.56g, 1.66 mmol) ,
S-benzyl cysteamine (0.69g, 4.15 mmol), acetic acid
(1.52 ml, 26.5 mmol) and sodium cyanoborohydride
(0.11g, 1.66 mmol) were used according to the
procedure for I. The composition of the third
eluant, EtOAc-MeOH-Et3N, used in the purification of
the crude mixture was 9:1:0.3, v/v/v. On
purification, 0.213g of compound VII (29.0~) was
isolated as an oil.
Pre~aration o~ Com~ound VIII (CP-0494):
VII (0.063g, 0.14 mmol) was deprotected
following the procedure for II. The thiol was then
treated with X (0.~5g, 0.152 mmol) in tris buffer.
The crude mixture was purified on a reverse phase
Vydac C-18 HPLC column using the gradient 15 - 70
CH3CN in water, 0.1~ constant TFA over 35 minutes.
VIII had a re~ention time of 15.0 minutes. 119.0 mg
t45.0~) of VII was isolated as a white powder on
lyophillization.
Mass S~ectral Data:
calculated molecular weight 1902
observed molecular weight 1904
Amino Acid Analysis:
Gly 0.81 (1), Arg 3.12 (3), Pro 1.07 (1), Leu 0.99
(1), Phe 2.04 (2), Hyp 0.98 ~1).
Pre~aration of Com~ound IX (CP-0499):
VII (0.009g, 0.02 mmol) was deprotected
following the procedure for II and the thiol was
then reacted with XI (0.026g, 0.016 mmol) in tris
buffer. Crude mixture was purified on a Vydac C-18
-
WO94/11021 PCT/US93/10222
21~7~6~
.
reverse phase HPLC uslng the gradient 15 - 70~ CH3CN
in water, 0.l~ constant TFA, flow rate of 8.0 ml/min
over 40 minutes. Retention time of IX was 14.22
minutes. IX (6.4 mg, 20.0~) was isolated as white
powder on lyophillization.
Mass S~ectral Da~a:
Calculatedmolecularweight 1957
Observed molecular weight 1958
,
I~ Vi tro Testing
The BKAn/mu-opioid receptor agonist
heterodimers were evaluated in vitro using the rat
uterus (BK2-receptor activity) and the electrically
stimulated guinea pig ileum (mu-opiate receptor
activity) assays. These assays are well known in
the art. The results obtained are shown in Table I:
TABLE I
Compound pA2--Rat Ut~rus ICso Guinea Pig lleum
(nmolar)
CP-0126 . 7.1 . inactive
2 o CP-0347 9.5 inactive
oxycodone inactive inactive
oxy.. ,Gr~hone inactive 21.7
CP-0477 7.9 i~
CP-0488 8.2 inactive
CP-0494 8.4 24.0
CP-0499 8.9 17.0
It should be noted that neither oxycodone nor
the heterodimers derived from oxycodone (CP-0477 and
CP-0488) showed any activity in the in vitro guinea
pig ileum assay of mu-opiate receptor agonist
activity. This is probably due ~o the fact that for
WO94/11021 PCT/US93/10~L
2~478~9 ~.
complete activity, oxycodone apparently needs to be
demethylated in vivo. As a result, oxycoàone and
oxycodone-based compounds would~ot be expected to
show activity in an assay wherein the appropriate
demethylating enzymes were m~ssing.
More important, however, are the data regarding
the activity of the BKAn component of these
heterodimers on the rat uterus and the data
regarding the activity of the oxymorphone containing
compounds. As can be seen from the data outlined in
Table I, full ~3KAn activity was retained in all of
these heterodimers and in those compounds utilizing
oxymorphone as the opiate, full mu-opiate receptor
agonist activity was also retained. From these
data, it is evident that sKAn/mu-opioid receptor
agonist heterodimers can interact with their
respective receptor populations in in vitro systems.
In V~ro Testing
In order to test the activity of these
compounds in vi~o, a model of inflammatory and
neurogenic pain was used. This model measures the
behavioral responses of mice injected in the hind
limb foot pad with 50 ul of formalin. The data from
these studies are summarized in Figures 2, 3 and 4.
Control mice (open circles) show a characteristic
bi-phasic response to the injected formalin wherein
there is a short lasting initial response followed
by a quiescent period which is then followed by a
sustained period of hind limb licking. The licking
behavior is interpreted to mean that the limb is
irritated and painful. The greater the time spent
licking, the more painful the stimulus.
Oxymorphone (Fig. 2 A and B) reduces both
phases of the licking behavior but with significant
~ WO94/11021 2 1 ~ 7 8 6 9 PCT/US93/10222
behavioral obtundation resulting in catalepsy and
frank respiratory depression at the highest doses
(0.9 and 3.0 umoles/kg). The bradykinin antagonist
CP-0127 (a potent BK2 selective antagonist--Fig. 3 A
and B) will reduce the time spent licking in both
phases of the formalin test but at doses that are
substantially greater than would be practical in a
clinical setting. CP-0494 (Fig. 4 A and B),
however, not only blocks both phases of the pain
response, but does so at doses substantially lower
(O.1 umoles/kg) than for either oxymorphone (0.9
umoles/kg) or CP-0127 ~12.6 umoles/kg) alone and, of
equal or greater importance, with no observable
narcotic effects over several hours. These data
indicate that BKAn/mu-opioid receptor agonist
heterodimers are pharmacologically qualitatively
superior to either of the parent pharmacophores as
would be expected from the theoretical
considerations outlined above.
One skilled in the art will appreciate that the
compounds described are representative of a wide
variety of compounds in which each of the components
of the heterodimer (BKAn, linker and/or mu-opioid
receptor agonist) can be varied to produce the
optimal effect desired.
Example 5 (BKAn/NEI)
A BKAn/NEI type of compound (CP-0502) of the
structure shown in Synthetlc Scheme 3, was
synthesized to illustrate that this class of
compounds can be used as a potent topical and/or
systemic anti-inflammatory agent. This compound is
derived from CP-0126 and the prototype elastase
inhibitor, CE-1218 (see Compound (6), Synthetic
Scheme l below).
WO 94/11021 ' . PCI/US93/1022.2
21~78g9
24
SYNTHETIC SCHEME 1
~ AcCI. CS2 ~ Pb(OAc)4, .;leOH ~ OMe
X~ AlCI3 X~ BF3-OEt2. C6H6 X~
(1) (2)
THF, NaH
Mel
1. SOCI2.CH2Ck ~OH KOH,EtOH j ~fOMe
OH >~ H20. ~ X~
o
S~~
TFA, CH2CI2
X~ S ~~ O~butyl
(4) O
H202, AcOH >~,~S ~OH
(S) C~
~, o--l'
C ~
~ W094/11021 21 ~ 7869 PCT/US93/10222
The linking element used in this heterodimer
was chosen so as to allow for l-nh~ n~ered hydrolysis
of the joining ester bond by serum esterases. Those
skilled in the art will recognize that the linker
can be modified so as to provide different rates of
hydrolysis varying from rapid to practically ~ero by
altering the steric accessibility of the ester
carbonyl carbon or by changing the chemistry to an
amide linkage. Completely stable linker moieties
can also be used which are free from potential
hydrolytic degradation.
Synthesi~ and analysis of BRAn/NEI heterodimer(s)
- The synthesis of these compounds is illustrated
by reference to Synthetic Schemes 2 and 3 and the
following detailed synthesis description which
includes the preparation of the elastase inhibitor
CE-1218 according to Synthetic Scheme 1:
~Nl~lIC SCHEME 1:
SYnthesis of 4-tert-Butvlaceto~henone (1)
To a dry l-L flask was added CS2 (250 mL) and
AlCl3 (133.34 g, 0.56 mol) with stirring. The
suspension was cooled in an ice bath and a solution
of tert-butylbenzene (50.00 g, 0.37 mol) and acetyl
chloride (78.50 g, 0.41 mol) was added dropwise over
2 hr (not allowing the temperature to rise above
25C). The reaction was allowed to stir at room
temperature overnight and then poured into a 2 L
beaker filled with ice. After quenching with 200 mL
of 6 N HCl the solution was saturated with NaC1 and
separated. The aqueous layer was washed with ether
(2 x 100 mL) and combined with previous organics.
This new organic solution was washed with water (100
mL), dried (MgSO4) and evaporated to give an oil
-
WO94/11021 PCT/US93/10 ~
21~78~9
26
which was distilled to give 52.1 g (79.3~) of 4-
tert-acetophenone as a clear colorless oil (bpo~5 mm
70-76C). lH NMR (CDCl3) ~ 1.35 (s, 9 H), 2.58(s, 3
H), 7.48 (d, J = 8.5 Hz, 2 H), 7.91 (d, J = 8.5 Hz,
2 H). 13C NMR (CDC13) ~ 26.42, 30.96, 34.95, 125.36,
128.16, 134.49, 156.64, 197.61.
S~nthesis of Methvl 4-tert-butvlphenYlacetate (2)
A dry 1 L flask equipped with a mechanical
stirrer containing Pb(OAc) 4 (132.06 g, 0.298 mol) and
250 mL of benzene was purged with nitrogen and
cooled in an ice bath. To this cooled slurry was
added dropwise a solution o~ BF30Et2 (137.8 mL, 1.12
mol), 4- tert-butylacetophenone (50 . 00 g, 0 . 284 mol)
in 70 mL of methanol over 1 hr. This mixture was
allowed to stir overnight, quench with water (500
mL), diluted with 250 mL ether and the layers
separated. The organic layer was washed with water,
diluted NaHCO3 (carefully) and dried over MgSO4. The
mixture was filtered, evaporated and distilled to
give 31.2 g (53.4~) of methyl 4-tert-
butylphenylacetate as clear colorless oil (bpo 04 mm
75-80C).. lH NMR (CDC13) 1.32 (s, 9 H), 3.62 (s, 2
H), 3.71 (s, 3 H), 7.23 (d, J = 8.4 Hz, 2 H), 7.37
(d, J = 8.4 Hz, 2 H). l3C (CDC13) 31.33, 34.46,
40.67, 52.04, 125.53, 128.88, 130.9~, 149.94,
172.26.
Svnthesis of Methyl 4-tert-butYlphenYlisobutyrate
(3)
A solution of methyl 4-tert-butylphenylacetate
(30.00 g, 0.145 mol) and iodomethane (45.41 g, 0.320
mol) in 125 mL of dry THF was added to a slurry of
NaH (8.72 g, 0.363 mol) in 200 mL of THF dropwise
over 30 minutes. Afte_ completion of the addition,
~ WO94/11021 21~ 7869 PCT/US93/10222
the reaction mixture was heated at reflux for 1.5
hr. The rçaction was allowed to cool to room
temperature, filtered through Celite and
concentrated. The residue was diluted in ether,
washed with H2O, and dried over MgSO4. Evaporation
of the solvent afforded the desired product as an
oil. A mixture of the crude methyl 4-tert-
butylphenylisobutyrate and 4:1 EtOH/H2O containing
KOH (10.07 g, 0.179 mol) were heated to reflux for 4
hr. The EtOH was evaporated in vacuo, the residual
solution was acidified to pH 2 with 2 N HCl, and the
precipitated solid filtered. The white solid was
the dried (60C, 1 mm Hg, 24 hr) to give the desired
product (23.45 g, 73.2~ from methyl 4-tert-
butylphenylacetate). lH NMR (CDCl3) l.34 (s, 9 H),1.62 (s, 6 H), 7.37 (s, 4 H), 11.4-12.4 (brs, 1 H).
13C NMR (CDCl3) 26.16, 31.30, 34.35, 45.81, 125.31,
125.48, 140.64, 149.66, 183.57.
S~nthesis of 4-(3'-carbo-tert-butoxv-~roDYl
merca~to)Dhenvl 4-tert-butYl~henvlisobutYrate (4):
A mixture of 4-tert-butylphenylisobutyric acid
(2.00g, 0~0091mol) and thionyl chloride (1.62 g,
0.0136 mol) in 16 ml of CH2Cl2 was allowed to stir
overnight under Argon. The volatiles were removed
under vacuum and the resulting solid was dissolved
into THF (15 mL) and a solution of tert-butyl-4-(4'-
hydroxyphenyl)mercaptobutyrate (2.44 g, 0.0091 mol),
TEA (2.5 mL) in THF (15 mL) was added dropwise over
10 min. The mixture was stirred for 3 days, diluted
with Et2O and extracted with 5~ NaHCO3. The organics
were washed with H2O, brine and dried (MgSO4). After
evaporization the colored oil was separated tHPLC,
silica gel 70:30 CH2Cl2/hexane to CH2C12 linear
graduent) to give the desired product as an oil
WO94/11021 PCT/US93/10 ~
21478~9
28
(2.20 g, 51.5~). lH NMR (CDC13) ~ 1.33 (s, 9 H), 1.43
(s, 9 H), 1.70 (s, 6 H), 1.88 ~tt, J = 7.2 Hz, 2 H),
2.35 (t, ~ = 7.2 Hz, 2 H), 2.90 (t, ~ = 7.2, 2 H),
. 6.92 (d, ~ = 8.6 Hz, 2 H), 7.32 (d, J = 8.6 Hz, 2
H), 7.34 - 7.40 (m, 4 H). l~C NMR (CDCl3) ~ 24.50,
26.42, 28.08, 31.31, 33.70, 34.11, 34.39, 46.40,
80.40, 121.92, 125.23, 125.46, 130.82, 133.03,
140.86, 149.58, 149.69, 172.21, 175.34.
SYnthesis of 4-(3'-carboxY-~roPYlmerca~to)~henyl
4-tert-butvlDhenYlisobutyrate (5)
Trifluoroacetic acid (25 mL) was added to a
stirred solution of 4-(3~-carbo-tert-butoxy-
propylmercapto)phenyl 4- tert-butylphenylisobutrate
(2 .40 g, 0 . 00510 mol) in 20 mL o~ CH2Cl2 over 15 min.
After an additional 15 min the volatiles were
removed and the oll crystalllzed ~hexane) to glve
1.94 g ~91.8~) of desired product as a white solid,
m.p. 86.0-87.0C, lH(CDCl3). 1.33 (s, 9 H), 1.70 (s, 6
H), 1.92 (tt, ~ = 7.0 Hz, 2 H), 2.50 (t, ~ = 7.0 Hz,
2 H), 2.93 (t, ~ = 7.0 Hz, 2 H), 6.93 (d, J = 8.7
Hz, 2 H), 7.33 (d, ~ = 8.7 Hz, 2 H), 7.35-7.39 (m, 4
H). 13C NMR (CDCl3) 23.89, 26.4S, 31.32, 32.34,
33.63, 34.42, 46.42, 122.03, 125.24, 125.48, 131.12,
132.~3, 140.85, 149.74, 175.40, 178.65.
SYnthesis of 4-(3'-carboxv-~ropylsulfonyl)DhenYl
4-tert-butylphenYlisobutYrate (6)
To a 50 mL flask was added 4-(3'-carboxy-
propylsulfonyl)phenyl 4-tert-butylphenylisobutyrate
(1.64 g, 0.00396 mol), HDAc (25 mL) and 15 mL of 30
H~O2. The reaction was allowed to stir overnight,
diluted with H2O (50 mL) and the resulting solid
filtered. After drying (12 hr, 1 mmHg) the solid
was recrystallized (CH2Cl2/hexane) to give 1.54 g
WO94/11021 PCT/US93/10222
21~786~ '
29
(87.1~) of the desired product as a white powder.
mp 107-108.5C. ~H (CDCl3) 1.33 (s, 9 H), 172. (s, 6
H), 2.02 (p, ~ = 7.0 Hz, 2 H), 2.52 (t, ~ = 7.0 Hz,
2 H), 3.17 (t, J = 7.0 Hz, 2 H), 7.20 (d, ~ = 8.7
Hz, 2 H), 7.36 (d, ~ = 8.7 Hz, 2 H), 7.41 (d, J =
8.6 Hz, 2 H), 7.90 (d, ~ = 8.6 Hz, 2 H). 13C NMR
(CDCl3) 17.88, 26.29, 31.28, 31.79, 34.42, 46.53,
55.04, 122.56, 125.16, 125.61, 129.70, 135.81,
140.22, 150.02, 155.29, 174.73, 177.71.
Synthesis of 6-Maleimidohexanol (7):
The synthesis of 6-maleimidohex~nol to be used
for linking is illustrated in Synthetic Scheme 2
below:
SY~ lCSCHE~IE2 `
HO ~~~ NH2 NaHC03, aq ~ ~ N
O J~ (7) o
~N _~
MeO
To a 100 mL flask was added 6-aminoh~nol
(0.76 g, 0.0064 mol) and 25 mL of saturated NaHCO3.
The homogenous was allowed to stir a RT and N-
methoxycarbonylmaleimide (1.00 g, 0.0064 mol) was
added as a solid. The mixture cleared shortly after
the addition and was all-owed to stir for 1 hr. The
mixture was extracted by EtOAc, dried (MgS04) and
evaporated. The resulting mixture was separated on
silica gel (CH2Cl2 to EtOAC). To give the product as
a white solid 0.32 g (25.2~), used without further
W094/11021 ' PCT/US93/10 ~
214 7 8 6 ~
purification. lH (CDCl3) ~ 1.25 - 1.45 (m, 4 H),
1.45 - l. 70 (m, 4 H), 3.53 ~t, J = 7.3 Hz, 2 H),
3.63 (t, J = 6.0 Hz), 6.71 (s, 2 H).
SYnthesis of CP-05Q2: .
Compound (6) (CE-1218) was esterified with
compound (7) to form compound (8) which was then
conjugated to CP-0126 to form the dimer CP-0502.
These latter reactions are illustrated in Synthetic
Scheme 3:
WO 94~11021 2 1 ~ 78 ~ 9 PCI'/US93/10222
3 l
SYNTHETIC SCHEME 3
~5 OH
HO ~ N~
Bop-CI (7) o
TEA
CH2C12 Q
~0~5 ~N~D
(8) 0
- . CP-0126
DIEA
DMF
H2N
~= NH-TFA NH2
o. ~>~ HO ~ ~,~A-HN =~,
TFA-H2N ~ ~, H O I H O , H O
S ~ H ~ o S
,~= NH-TFA 4N
H2N O
~~0 o
cr~-0~02
wo94/ll2~ ~7 8 ~ 9 PCT/US93/10
Referring more specifically to Synthetic Scheme
3, compound (6) (200 mg, ~.448 mmol), triethylamine
(0.124 ml, 2 eqv), 6-mal~imidohexanol (7) (97 mg.
1.1 eqv) were dissolved in 2 ml methylene chloride.
Bis(2-oxo-3- oxazolidinyl)phosphinic chloride (122
mg, 1.0 eqv) was added to the stirred solution. The
resulting suspension was stirred at room temperature
for four hours. The reaction mixture was diluted
with 25 ml methylene chloride and washed with
saturated NaHC03. The organic solution was dried over
anhydrous MgS04 and the solvent was removed in vacuo.
Silica gel chromatography (2X18 cm column);
3s/65:acetone/hexane (R~=O . 4 ) a~ eluent pro~ided the
compound ( 8 ) as a colorless oil .
Compound (8) (50 mg, 0.08mmol) was dissolved in
lo ml DMF containing 100 ul diisopropylethylamine.
lo~ mg (~.08mmols) o~ CP-0126 was added and reaction
proceeded ~or 30 minutes, with occasional mixing.
The reaction mixture was injected on a Vydac 1" C-18
reverse phase column, and eluted at lOml/min,
15~-90~ acetonitrile in H20 over 35 minutes
(Constant .1~ TFA). The appropriate fractions were
lyophilized to yield 52 mg (35~) of a colorless
white powder (CP-0502) Laser desorption mass
spectrometry M/Z =1890 (M+H), calculated 1890.
Automated amino acid sequence results confirmed the
correct peptide sequence with no altered amino
acids.
I~ vitro acti~ity of BRAn/NEI heterodimer (8)
In vi tro evaluation of BKAn and NEI activity of
the following compounds was carried out according to
standard protocols well known to those in the art.
BKAn activity (PA2) was assessed using the rat uterus
preparatlon and NEI (Klss) activity was evaluated
~ WO94/11021 21 ~ 7 8 6 9 PCT/US93/10222
using purified human neutrophil elastase (HNE) and a
synthetic soluble chromogenic substrate,
methoxysuccinyl-alanyl-alanyl-prolyl-valyl-
paranitroaniline (MOS-AAPV-pNA). The inhibitor was
mixed with MOS-AAPV-pNA (0.5 mM) in 0.05 M sodium
phosphate, 0.1 M NaCl, 0.005~ Triton X-100, 5~ DMSO,
pH 7.5. HNE (10-20 nM) is then added. The
production of nitroaniline was monitored
spectrophotometrically at a wavelength of 400-410 nm
at 25 C. An ENZFITTER program then automatically
calculated standard enzyme kinetic parameters
including Ki~5.
The following results were obtained:
TABLE II
Compound pA2--Rat Uterus Ki's (HNE)
nM
CP-0126 7.1 inactive
CE-1218 inactive 10.5
CP-0487 8.4 inactive
CP-0502 7.5 6.6
The data in Table II indicate that for NEI
activity there is little difference between the
intact heterodimer (CP-0502) and the free monomeric
NEI moiety (CE-1218) as far as their respective Ki's
are concerned. This is not true for the activity of
the BKAn portion of the intact heterodimer relative
to its hydrolysis product, CP-0487 (the
succinimidohexanol derivative of CP-0126) wherein
the intact compound is almost a full log less potent
than the monomeric BKAn. Interestingly, the
activity of the intact compound displayed a type of
irreversible bradykinin antagonism and an apparently
enhanced antagonist activity of bradykinin induced
WO94/11021 PCT/US93/10 ~
21~
uterine contractions at longer~-incubation times.
These types of receptor in~e~actions are not well
measured by standard pA2 an~lyses so the differences
in activity observed between CP-0487 and CP-0502
with respect to BKAn activity may be more apparent
than real. Regardless of the molecular
pharmacologic mechanisms underlying these data it is
clear that combined BKAn and NEI activity can be
incorporated into a single molecule.
The above data suggest that allowing for in
vivo hydrolysis of the intact compound may alter the
behavior of the two moieties so as to enhance the
overall in vivo activity of the primary compound.
- Un~ortunately, there are no established An; m~ 1
models that can be employed to assess combined BKAn
and N~I acti~rity in viw. Therefore, in ~rdex to
assess the potential for in vivo hydrolysis of the
intact heterodimer, an in vi tro " surrogate" system
was employed wherein the parent heterodimer (CP-
0502) was incubated with human plasma and theresulting metabolites analyzed by reverse phase
HPLC .
CP-0502-was added to freshly obtained normal
human plasma and allowed to incubate at 37C for
varying amounts of time. At the designated time the
samples were treated with acidified (O.l N HCl)
acetonitrile in order to precipitate the plasma
proteins. Aliquots (75 ul) of the supernatents were
then analyzed on a Vydac C-18 reverse phase HPLC
column using 24~ to 80~ acetonitrile gradient in
O.l~ TFA. The eluent was monitored at 214 nm.
Figures 5 a and b are representative reverse
phase HPLC chromatograms illustrative of this type
of analysis. As can be seen from these
chromatograms, the parent compound appears to be
WO94/11021 ~1 q 78 6 9 PCT/US93/10222
readily hydrolyzed to the succinimidohexanol
modified monomer, CP-0487 and its des-Arg9 derivative
(Plasma carboxypeptidase will cleave the terminal
arginine residue from both the intact heterodimer,
CP-0502, as well as CP-0487.) The apparent Tl/2 of
this hydrolysis reaction is approximately 113
minutes. The NEI component of the heterodimer is an
active ester and undergoes hydrolysis as well.
However, the intact NEI monomer as well as its
hydrolysis products are obscured by the plasma
derived peaks seen in the middle of this tracing and
cannot be visualized using this system. Since the
NEI is equally active as a component of the
heterodimer as it is as a monomer, the dissociation
of the heterodimer into its two component parts will
have less of an effect on its activity than that for
the BKAn component.
Those skilled in the art will appreciate that
the hydrolysis rate of the heterodimer can be
influenced by the steric and electronic environment
of the "linking" ester moiety and that the type of
chemistry used is only a single example of the types
of chemistry.that can be employed to adjust the rate
of dissociation (or lack thereof) of the two
components of the heterodimer.
Example 6 (BRAn/COI)
Synthesis and Analysis of BRAn/COI Hetero~ ~s
A representative BKAn/COI heterodimer (CP-0460)
was synthesized according to Synthetic Schemes 4 and
5 below:
PCI~/US93/102
WO 94/11021
2~8~9
.~ .
36
SYN l ~l~; l IC SCHEME 4
~J~OH ~0
CH30 ~ I . DCC, C1-{7CI~ CH30
~--N 2. Na~C03. 7 ~--N
c~ ~CI
~ WO94/11021 21 4 78 ~ 9 PCT/US93/10222
S~nthesis of 6-Maleimidohexan~l 1-(4-chlorobenzovl)
-5-methoxv-2-methY1-3-1ndohYlacetate (9~ .
To a 100 mL flask was added indomethacin (1.90
g, 0.00532 mol), 25 mL of CH2Cl2 and DCC (0.55 g,
0.00266 mol). After 2 hr, the mixture filtered, the
DCU washed with 15 mL of CH2Cl2 and to this new
solution was added 6-maleimidoh~nol (0.50 g,
0.00253 mol) as a solid followed by anhydrous Na2CO3
(0.32 g, 0.00304 mol). After 4 days the mixture was
filtered diluted with Et2O and washed with 5~ NaHCO3,
H2O and dried (MgSO4). The resulting yellow oil was
purified in a HPLC (silica gel; CH2Cl2 to 80:20
CH2Cl2/EtOAc, linear gradient 60 min.) to give 0.79 g
(58.0~) of the desired product as a yellow oil. lH
(CDCl3) 1.20 - 1.35 (m, 4 H), 2.38 (s, 3 H), 3.47 (t,
J = 7.3 Hz, 2 H), 3.66 (s, 2 H), 3.83 (s, 3 H), 4.08
(t, J = 6.6 Hz, 2 H), 6.66 (J = 9.0 Hz, J = 2.5 Hz,
1 H), 6.68 (s, 1 H), 6.87 (d, J = 9.0 Hz, 1 H), 6.96
(d, J = 2.5 Hz, 1 H), 7.47 (d, J = 8.5 Hz, 2 H),
7.66 (d, J = 8.5 Hz, 2 H).
Conjugation of compound (9) with CP-0126 to
form CP-0460 is illustrated in Synthetic Scheme 5
and described thereafter:
PCI~/US93/102,~
W0 94/11021
2~ ~78~
38
SYNTHETIC SCHEME 5
H2N NH2
~= NH-TFA
~A-HN =~ NH
TFA-H2N ~ N ~ ~ N ~JI~ N ~ N ~ N ~ N ~J~ N
S~ H ~o SH ~--
,~ NH-TFA CP-0126
H2N
o o~ ,
CH30 ~ (9~ o DMF
~ N NH4HCO3
.' ~cl
NH-TFA NH2
HO ~ ~A-HN =(
0 ~ N J~ N ~ N ~ N ~J~ N
H ~o ~ H O ~H O
HN )~
,~ NH-TFA N ~
H2N CH30 . / o
~O~ ~o ~,
~\ CP-0~60
o~
Cl
~ WO94/11021 21~7869 PCT/US93/10222
CP-0126 (100 mg, 0.08 mmol) was reacted with
compound (9) (0.12 mmol, 1.5 eqv) in 2 mL 95~ DMF/5
0.lM ammonium bicarbonate containing 50 ul
diisopropylethylamine, for 30 minutes, with
occasional mixing. The reaction mixture was purified
in 1 injection on a Vydac 1" C-18 reverse phase
column at lOml/min, using a gradient running from
15~ acetonitrile/0.1~ TFA to 40~ acetonitrile/0.1
TFA in 20 minutes. Appropriate fractions were
lyophilized to yield 64 mg (45~) of a colorless
powder (CP-0460). Laser desorption mass
spectrometry: M/Z= 1802 (M+H), calculated 1802.
As mentioned previously, for the COI to work it
may need to be dissociated from the BKAn so as to
allow for its intracellular penetration.
Therefore, in order to evaluate the functional
activity of a BKAn/COI heterodimer, CP-0460 was
exposed to rat lung parenchymal strips which were
then challenged with arachidonic acid. This tissue
is known to contain both non-specific esterase
activity as well as to convert arachidonic acid to
thromboxane (via a cyclooxygenase dependent pathway)
which is then ultimately responsible for the smooth
muscle contraction observed in this assay.
Using this system, the log dose ratio shifts
for indomethacin and CP-0460 were found to be 0.998
+/- 0.425 and 1.029 +/- 0.042 respectively
indicating that both indomethacin alone and CP-0460
will prevent the contraction produced in response to
exogenously applied arachidonic acid with equal
potency. BKAn's have no effect on this system in
and of themselves. These data indicate that the COI
component of BKAn/COI heterodimer is functionally
active in a tissue containing both esterolytic and
cyclooxygenase activities.
WO94/11021 PCT/US93/10 ~
2~ ~78~
Intact CP-0460 was also tested for BKAn
activity using the standard rat uterus assay and the
PA2 of the CP-0460 was found to be approximately 7.8.
Again, CP-0460 (similarly to CP-0502) did not behave
as a classical competitive antagonist of bradykinin
induced uterine contraction but rather as a type of
"pseudo-non-competitive" antagonist, particularly at
higher concentrations. This atypical behavior
cannot be attributed to COI activity per se as free
indomethacin has no effect on this assay at any
concentration.
Regardless of the explanation for the observed
data, one skilled in the art will appreciate that,
as in the other two classes o~ compounds illustrated
herein, pharmacologically important sKAn/coI
heterodimers can be made using a variety of
appr~pria~e llnking moietie~ to pr~vide a free
hydroxyl and the car~oxyl group (a common feature of
many COIs) of the COI monomer to form a hydrolyzable
ester based heterodimer. Compounds such as these
may be used in the treatment of a variety of
inflammatory or painful conditions as well as in the
treatment of dysfunctional uterine smooth muscle
activity.
While the invent.ion has been exemplified above
by the use of Y components which are non-peptides,
this component may equally comprise in whole or part
a pep~ide as exemplified in the afore-mentioned WO
92/17201, including the heterodimers therein
described.
The dimers of the invention may be used n the
form of conventional pharmaceutical compositions
comprising the active component and a
pharmaceutically acceptable carrier. Such
^~ compositions may be adapted ~or topical, oral,
WO94/11021 21 4 78 6 9 PCT/US93/10222
aerosolized, intramuscular, subcutaneous or
- intravenous administration. The amount of active
component present in such compositions will range
from, for example, about 0.001 to 90.0~ by weight
depending on the application and mode of
administration although more or less of the active
component may be used. Conventional dosages will
vary considerably on the basis of the intended
application and mode of administration. Usually,
however, an effective dose is in the order of 0.1 to
1000 micrograms per kg body weight.
The scope of the invention is defined in the
following claims wherein:
