Non-depolarizing neuromuscular blocking activity of bisquaternary amino di- and tripeptide derivatives

Article abstract of DOI:10.1021/jm0010062

Herein we describe the synthesis of novel di- and tripeptide derivatives with two quaternary nitrogen groups attached and the biological testing of these compounds for neuromuscular blocking (NMB) activity in vitro and in vivo. The short peptide scaffold

Full text of DOI:10.1021/jm0010062

4822
J . Med. Chem. 2000, 43, 4822-4833
Non -d ep ola r izin g Neu r om u scu la r Block in g Activity of Bisqu a ter n a r y Am in o Di-
a n d Tr ip ep tid e Der iva tives
Leo H. D. J . Booij,‡ Leon A. G. M. van der Broek,† Wilson Caulfield,# Brigitte M. G. Dommerholt-Caris,†
J ohn K. Clark,# J an van Egmond,‡ Ross McGuire,# Alan W. Muir,# Harry C. J . Ottenheijm,† and
David C. Rees*,#
Departments of Medicinal Chemistry and Pharmacology, Organon Laboratories, Newhouse, Scotland ML1 5SH, U.K.,
Scientific Development Group, N V Organon, Molenstraat 110, 5342 CC Oss, The Netherlands, and Department of
Anaesthesiology, University Medical Center, University of Nijmegen, Nijmegen, The Netherlands
Received J une 8, 2000
Herein we describe the synthesis of novel di- and tripeptide derivatives with two quaternary
nitrogen groups attached and the biological testing of these compounds for neuromuscular
blocking (NMB) activity in vitro and in vivo. The short peptide scaffold was selected because
it offers potential for desired distance between the two pharmacophoric quaternary nitrogen
groups, short duration of action, straightforward synthesis, and compatibility with an injectable
formulation. From a small series of compounds 20c,e are identified as effective non-depolarizing
NMB agents in vitro and in vivo in anesthetized cats and Rhesus monkeys with potencies
similar to those of the clinical reference compounds rocuronium (4) and suxamethonium (2)
(monkey ED90 ) 0.68, 0.23, 0.16, 5.04 µmol/kg, respectively). These new peptide derivatives
20c,e have similar potency and onset time but longer duration and slower recovery than the
clinically used reference compounds. The structure-activity relationships described for this
chemical series lead to the conclusion that the di- or tripeptide fragment can be regarded as
an alternative template to the steroid or aliphatic ester of previously reported NMBs and within
this tripeptide-derived series clog P correlates well with in vitro NMB activity.
In tr od u ction
of action: the depolarizing agents, e.g. suxamethonium
(2), and the non-depolarizing agents, e.g. pancuronium
(3), rocuronium (4), and atracurium (5). The non-
depolarizing agents act as competitive antagonists for
acetylcholine, and their action may be reversed by an
anticholinesterase agent, e.g. neostigmine.2 The depo-
larizing NMB suxamethonium (2) produces rapid pe-
ripheral skeletal muscle relaxation by nicotinic acetyl-
choline receptor (nAChR) desensitization (tachyphylaxis)
but in addition causes (mechanism-related) undesired
side effects such as potassium release and autonomic
activation (through ganglion receptors).
The chemical structures of both classes of the clini-
cally used NMBs possess two or three alkylated amino
nitrogen atoms with at least one of these quaternized
leading to a permanent charge (hence distinguishing
these drugs from the widely researched centrally acting
nAChR modulators). NMB activity is loosely related to
the inter-nitrogen distance: approximately 1.0-1.4 nm
is optimal and non-depolarizing agents tend to be more
sterically bulky and less conformationally flexible than
depolarizers.3,4 This interonium distance can be main-
tained by various chemical moieties. Two extensively
studied chemical series of non-depolarizing NMBs are
the amino steroids (e.g. 3, 4)3,5 where the distance is
maintained by an androstane skeleton and the bisben-
zylisoquinolinium diesters (e.g. 5)6 where a linear ester
structure determines the distance. The conotoxin pep-
tide muscle relaxants are chemically unrelated to the
series described above but have been reported to dem-
onstrate neuromuscular block in vitro.7
General anesthesia is a state of reversible uncon-
sciousness and pain suppression, which is induced to
cause amnesia and prevent autonomic responses and
to modulate muscle tone and cardiovascular reflexes
during surgery. It consists of three separate compo-
nents: hypnosis with amnesia, analgesia, and skeletal
(peripheral) muscle relaxation, all of which are con-
trolled by three different classes of drugs.1 These drugs
are administered by injection or inhalation and are used
in combination with each other to achieve a fast onset
of action (less than 2 min) followed by rapid and
predictable recovery without serious side effects.
Neuromuscular blockers (NMBs) are one of the three
classes of drugs used widely during surgical anesthesia.
They are peripherally acting skeletal muscle relaxants
that are used to paralyze laryngeal muscles prior to
airway intubation and paralyze other skeletal muscles,
predominantly those of the thorax and abdomen, during
surgery. All currently used NMB drugs achieve this
effect by blocking the pharmacological effects of acetyl-
choline (1), CH3CO2CH2CH2N+Me3, at membrane-
bound ligand-gated nicotinic acetylcholine receptors on
skeletal muscle.
Two classes of clinically used NMB drugs can be
distinguished based upon the compounds’ mechanisms
* To whom correspondence should be addressed. Current address:
Medicinal Chemistry Department, AstraZeneca R&D Mo¨lndal, S-431
83 Mo¨lndal, Sweden. Tel: +46 31 7761350. Fax: +46 31 7763868.
E-mail: David.C.Rees@astrazeneca.com.
#
Organon Laboratories, Scotland.
One of the challenges facing medicinal chemists
researching new anesthetic agents is to achieve phar-
† N V Organon, The Netherlands.
‡ University of Nijmegen.
10.1021/jm0010062 CCC: $19.00 © 2000 American Chemical Society
Published on Web 11/08/2000

NMB of Bisquaternary Amino Di- and Tripeptides
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 25 4823
Sch em e 1a
a
Reagents and conditions: (a) DCC/HOBT/H2NCH2CH2NR1R2/DMF; (b) TFA or HCl/EtOAc; (c) Dowex or KOH/MeOH; (d) R3X/NaHCO3/
Sephadex/Dowex; (e) DCC/HOBT/piperidineacetic acid/DCM; (f) R3X/MeCN.
macokinetic properties leading to rapid onset and short
duration of action. As a result anesthesia will become
more controllable and can be rapidly adjusted to the
clinical requirements. This challenge is created by the
increasing trend of day care surgery and by the fact that
nowadays more severely sick patients are operated upon
and thus anesthetized. Attempts to fulfill this require-
ment have included: a soft-drug deactivation approach,
e.g. incorporating metabolically labile ester groups into
the opioid analgesic remifentanil,8 incorporating chemi-
cally labile esters into the bistetrahydroisoquinolinium
neuromuscular blockers,9 and decreasing the potency
of muscle relaxants to increase their speed of onset.10,11
Another challenge in the search for new NMBs is the
requirement for selectivity between blocking effects at
the receptors on skeletal muscle and at other acetyl-
choline receptors, particularly those in the autonomic
ganglia and the heart in order to avoid undesirable
cardiovascular side effects.12
the structure-activity relationship (SAR) of such a
series. It was envisaged that a peptide scaffold could
confer a number of desirable properties in this context:
e.g. rapid in vivo metabolism to nontoxic metabolites
(the soft drug concept), low propensity to cross the
blood-brain barrier, and straightforward synthesis from
inexpensive starting materials. In designing which
peptide fragments and which quaternary side chains to
synthesize the following aspects were considered: mimic
approximately the interonium distance of the steroidal
NMBs, allow for some steric bulk/conformational con-
straints associated with non-depolarizing activity, and
incorporate some quaternary side chains previously
reported to confer NMB activity. It has been reported
in the patent literature that certain bisquaternary
peptide amines exhibit non-depolarizing NMB activity,
e.g. Me3N+PheNH(CH2)10NHPheN+Me3, but the SARs
of these compounds have not been reported.13,14
Ch em istr y
The aim of the research described in this manuscript
is to investigate whether a short peptide fragment,
decorated with two alkylated amino groups, would
exhibit non-depolarizing NMB activity and to explore
The di- and tripeptide derivatives linked to various
diaminoethanes described in Schemes 1-3 were pre-
pared by standard peptide synthesis techniques, es-

4824 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 25
Booij et al.
Sch em e 2a
a
Reagents and conditions: (g) DCC/HOBT/H2NCH(R3)CO2R‚HCl/Et3N/DCM; (h) NaOH/MeOH; (i) DCC/HOBT/H2NCH2CH2NR1R2/
DMF; (f) R6X/MeCN.
sentially following the methods employed by Konteatis
et al.15 Four of the substituted diaminoethanes were
previously prepared according to known literature
methods16-18 as were piperidineacetic acid19 and diben-
zylglycine.20 Commencing with a simple N-BOC-pro-
tected dipeptide 6, as outlined in Scheme 1, couplings
with various N,N-disubstituted ethylenediamine deriva-
tives were carried out in DMF using 1-hydroxybenzot-
riazole hydrate (HOBT) and dicyclohexylcarbodiimide
(DCC) to afford the ethylamides 7a -e. The ethylamides
7a -d could be deprotected using excess trifluoroacetic
acid (TFA) giving the corresponding diamines 8a -d
after liberation from their TFA salts using ion-exchange
resin, as highly water-soluble products. In the case of
the benzylmethylamino derivative 7e the deprotection
was carried out by the method of Stahl et al.21 which
uses HCl in ethyl acetate, and the free base 8e was
obtained by standard extraction. The quaternary salts
9a -d were then prepared from diamines 8a -d by
reaction with a large excess of iodomethane in the
presence of sodium bicarbonate, a technique used in
peptide quaternary salt preparation,15,22 but the prod-
ucts proved insoluble in solvents such as ethyl acetate
or DCM so the workup method in the above reference22
could not be employed. Instead the products were
isolated by an aqueous chromatographic method using
Sephadex G-10 as adsorbent, which provided the qua-
ternaries as the extremely hygroscopic ammonium
hydroxides, which required treatment with ion-ex-
change resin to yield the corresponding chlorides.
Although NMR spectra were consistent with the struc-
tures 9a -d , elemental analysis values for the com-
pounds indicated high water content and the possibility
of some sodium ion contamination. Realizing this dif-
ficulty, the diamine 8e was further coupled to pip-
eridineacetic acid 12b19 giving the more lipophilic
tripeptide derivative 10, which could then be readily
converted to quaternary compounds 11a -c using the
appropriate alkyl halides without the need for any
scavenger, and the products could be obtained after
isolation by precipitation from acetone/ether mixtures.
This finding significantly altered the methodology
used in subsequent synthetic procedures. Beginning
with an N,N-dialkylated terminal amino acid and
coupling to amino acid ester derivatives followed by
hydrolysis generated intermediates which were easier
to isolate and handle than those of Scheme 1. Thus N,N-
dibenzylglycine 12a 20 (Scheme 2) was coupled to glycine
ethyl ester and then hydrolyzed using a slight excess
of sodium hydroxide to give the dipeptide 13a . Further
coupling to 2-aminoethylpiperidine gave ethylamide
14a , which could be readily converted to quaternary
dimethiodide 15a again without contamination by scav-
enger. In a similar process, piperidineacetic acid 12b19
was reacted with L-phenylalanine methyl ester followed
by hydrolysis to provide dipeptide 13b. Coupling of 13b
to 4-(2-aminoethyl)morpholine gave ethylamide 14b,
which was alkylated to afford quaternary salts 15b,c.
Scheme 3 indicates first an attempt to introduce some
rigidity into the already interesting compounds 11a -c
of Scheme 1 by using sarcosine units in the backbone
instead of glycine. This had two immediate synthetic
effects, in that the coupling reactions using a secondary
amine proved to be less facile and due to rotameric
isomerization about the hindered amide bonds the NMR
spectra of all the compounds proved to be difficult to
interpret. Second, Scheme 3 shows an attempt to test
the scope by the use of a chain-extended amino acid 16.
Piperidineacetic acid 12b19 and the corresponding pro-
pionic acid 16 were reacted with sarcosine ethyl ester

NMB of Bisquaternary Amino Di- and Tripeptides
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 25 4825
Sch em e 3a
a
Reagents and conditions: (j) DCC/HOBT/CH3NHCH2CO2Et‚HCl/DIPEA/DCM; (k) NaOH/MeOH or dioxane; (l) HBTU/R3NHCH2CH2NR1R2/
DIPEA/DMF; (f) R4X/MeCN.
using standard procedures mentioned above and hy-
drolyzed to give dipeptide 17a and amino acid 17b.
Further coupling to a second unit of sarcosine and
hydrolysis provided 18a ,b, and finally addition of 2-(me-
thylaminoethyl)benzylmethylamine18 was accomplished
using 2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluro-
nium hexafluorophosphate (HBTU), a reagent described
by Knorr et al.23 as being an excellent activating agent
superior to DCC. The diamines 19a -c could be chro-
matographically purified before alkylation to give the
quaternary products 20a -e.
will be 1.15-1.49 nm, while the corresponding distance
in the dipeptide derivative 9a will be 0.91-1.13 nm.
Thus both the di- and tripeptide derivatives have the
potential to exhibit the favored interonium distance of
the order of 1.1 nm.
In addition to this conformational guideline, an ele-
ment of bulk and/or lipophilicity has also been postu-
lated as important for non-depolarizing ability.3,5 Struc-
tural variations have been made to test this hypothesis
by making changes to the lipophilicity of the molecules
in the series of tripeptides. The initial biological screen
was in an in vitro isolated chick biventer muscle
preparation. Compounds were tested to determine the
dose required to block the indirectly elicited electrically
stimulated muscle twitch. The doses of the dipeptide
bisquaternary compounds, tripeptide bisquaternary com-
pounds, and reference compounds producing 50% twitch
block are shown in Tables 1-3, respectively.
The most active compounds identified in the chick
biventer assay were then evaluated in vivo in anesthe-
tized cats. These experiments determined the potency
(ED50) of the inhibition of the twitch response of the
electrically indirectly stimulated tibialis anterior muscle
and the time course in terms of onset, recovery, and
duration of action. The selectivity of test compounds for
undesirable effects on autonomic system was investi-
gated by measuring inhibition of the stimulated vagus
Resu lts a n d Discu ssion
In selecting which length of peptide fragment to use,
the aim was to approximately mimic the interonium
distance of the steroidal NMBs. Analysis of the X-ray
crystal structure24 of pancuronium (3) shows that the
interquaternary nitrogen distance is 1.11 nm. In this
molecule, as in the close structural analogue rocuronium
(5), there are no freely rotatable bonds between the two
nitrogens. Thus, even allowing for some ring flexibility,
the N-N distance will not alter significantly. X-ray data
for various suxamethonium (2) crystals also exist, and
these show that in the solid state the N-N distance can
vary from 0.76 nm25 to 1.2 nm.26 Conformational
analyses suggest that the likely distance between the
quaternary nitrogens for the tripeptide derivative 11a

4826 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 25
Booij et al.
Ta ble 2. Structure and in Vitro NMB Activity of Tripeptide
Ta ble 1. Structure and in Vitro NMB Activity of Dipeptide
Bisquaternary Derivatives
Bisquaternary Derivatives
a
Dose causing 50% block of muscle twitch (µM); number of
determinations in parentheses. B Developed slow increase in base-
line tension. Decrease in baseline tension. Poor recovery or no
recovery following washing.
b
r
nerve-induced bradycardia and superior cervical nerve-
induced contractions of the nictitating membrane. These
results are shown in Table 4, also in comparison with
reference NMBs. Two of the tripeptide derivatives are
evaluated in anesthetized Rhesus monkeys (Table 5).
The first compound in Table 1 is the parent Gly-Gly
dipeptide derivative 9a possessing an ethylenediamine
moiety which confers potential for similar ‘through-bond’
distance between the two quaternary nitrogens as in
the reference NMB steroid, pancuronium. This com-
pound, 9a , is inactive in the chick biventer assay (>390
µM) so the next three compounds introduce cyclic
quaternary groups present in known NMBs from chemi-
cal series with different scaffolds (e.g. steroids and the
bisbenzylisoquinolinium diesters): i.e. the morpholine
9b, the pyrrolidine 9c, and the tetrahydroisoquinoline
9d . Modification of these structures failed to increase
NMB activity (>390 µM); however when both of the
quaternary groups of 9a are changed, as in 15a , the
NMB activity (134 µM) was modestly increased. In-
creasing the lipophilicity and conformational restraint
within the dipeptide scaffold by incorporating a Phe
residue and introducing cyclic quaternary groups at
both nitrogens leads to 15b (407 µM) and 15c (19.8 µM)
which is the most active compound prepared in this
series. The p-nitrobenzyl quaternary substituent in 15c
has previously been incorporated into a different chemi-
cal series where it is reported to convert depolarizing
NMB activity into non-depolarizing NMB activity.27
Compound 15c was tested in the anesthetized cat model
where it demonstrates NMB activity ED50 ) 10 µmol/
kg (Table 4) albeit at approximately 2 orders of mag-
nitude higher doses than the clinically used reference
compounds (2-5). For comparison, an inactive repre-
a
Dose causing 50% block of the muscle twitch (µM).
Ta ble 3. Structure and in Vitro NMB Activity of Reference
Compounds
*Immediate concentration-dependent increase in baseline ten-
sion showing a depolarizing action.
sentative from the dipeptide series, 9c, was tested in
the cat and, as expected, is inactive (ED50 > 10 µmol/
kg).
Homologation of the Gly backbone gives the tripeptide
series in Table 2. The piperidinobenzyl bisquaternary
11a displayed only weak NMB activity (ED50 ) 345 µM),
but increasing the bulk of the quaternary group with
bisallyl or p-nitrobenzyl slightly increases activity, in
line with the SAR of previously reported NMB chemical

NMB of Bisquaternary Amino Di- and Tripeptides
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 25 4827
F igu r e 1. Plot of pED50 (chick biventer data from Tables 1 and 2) versus predicted pED50 according to eq 1.
Ta ble 4. Potency and Time Course of NMB Activity in Anesthetized Cats
potency
ED50 (µmol/kg)
onset
(min)
recovery25-75%
(min)
duration90%
(min)
vagal
ED50 (µmol/kg)
nic memb
ED50 (µmol/kg)
compd (n ) 1)
9c
15a
15c
11c
>10.0
>19.2
10.1
7.3
>10.0
∼12.8
∼0.8
0.7
>10.0
∼19.2
>11.2
5.3
20a
20c
20e
>12.8
∼1.6
<0.4
∼0.2
>12.8
>1.8
.0.2
a
1.2
3.3
2.6
1.7
4.1
2.0
5.8
3.8
3.9
1.9
4.0
2.9
5.9
12.7
10.4
5.7
11.7
8.3
∼0.1
0.11
0.03
0.25
0.10
suxamethonium, 2
pancuronium, 3
rocuronium, 4
atracurium, 5
0.07
2.17
1.6
19.4
a
Pre- and post-block increases in twitch tension and short-lived muscle fasciculations 20-30 s after dosing.
Ta ble 5. Potency and Time Course of NMB Profile in
Anesthetized Monkeys
This result encouraged us to make additional analogues.
Homologation of the backbone with a single methylene
unit at the N-terminus to give the â-Ala-Gly(Me)-Gly-
(Me) derivative 20d decreases activity in vitro compared
to 20c (ED50 ) 4.2, 0.4 µM); however, rigidifying the
C-terminal region by incorporating a piperazine group
retains in vitro activity at similar dose levels as the
reference compounds (20e: ED50 ) 0.7 µM) and gives
approximately 10-fold improved potency with a similar
or possibly improved onset time in vivo (cat ED50 ) ∼0.1
µmol/kg; onset time 2.6 min). Although selectivity with
respect to the autonomic system is improved with 20e
compared to other peptides, it is less attractive in this
respect than the reference compounds (Table 4).
The initial working hypothesis that lipophilicity might
be an important factor in determining non-depolarizing
NMB activity was quantified retrospectively by a QSAR
analysis. Figure 1 and eq 1 show that in the tripeptide
series (the first eight compounds in Table 6) activity at
the chick biventer varies with calculated lipophilicity28
with an r2 of 0.86:
recovery
25-75%
(min)
duration
90%
(min)
potency
ED90 (µmol/kg)
onset
(min)
compd
20c (n ) 2)
20e (n ) 2)
rocuronium,
4 (n ) 5)
suxamethonium,
2 (n ) 5)
0.68
0.23
0.16
1.7
2.9
17.6
15.6
27.7
42.5
3.1 ( 0.5 5.0 ( 0.9 13.6 ( 2.4
5.04
1.5 ( 0.2 2.9 ( 0.6 9.7 ( 2.0a
a
Data from Muir et al. Eur. J . Anaesthesiol. 1998, 15, 467-
479.
series, to within 2 orders of magnitude of the clinically
used reference compounds in the same assay (Table 3)
(11b,c: ED50 ) 283, 47 µM, respectively).
In the cat model 11c shows weak NMB potency (ED50
) 7.3 µmol/kg) but is more active at inhibiting the vagal
nerve (ED50 ) 0.7 µmol/kg) indicating undesirable
selectivity for the autonomic system. N-Methylation of
the glycine backbone of 11a gives a modest increase in
NMB activity in vitro (11a , 20a : ED50 ) 345, 140 µM,
respectively) which is further enhanced by changing the
quaternary groups from methyl to allyl and p-nitroben-
zyl (20a -c: ED50 ) 140, 69, 0.4 µM respectively, Table
2). This latter compound 20c is potent in vivo (cat ED50
) 1.2 µmol/kg, Table 4) and has a time course that
compares favorably with some of the reference com-
pounds, being faster in onset (3.3 min), recovery (3.8
min), and duration (12.7 min) than atracurium (5) but
slower than suxamethonium (2) and rocuronium (4).
predicted pED50 ) 0.56(clog P) + 5.40
n ) 8; r2 ) 0.86; F ) 38.8; s ) 0.46
(1)
In eq 1, n is the number of observations, r is the
correlation coefficient, F is the F-value, and s is the
standard error.
The dataset used in Figure 1 may be extended to
include the non-depolarizing reference compounds pan-

4828 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 25
Booij et al.
F igu r e 2. Plot of pED50 (chick biventer) versus predicted pED50 according to eq 2.
Ta ble 6. Chick Biventer Activity Data and Clog P Values for
these two cases, and it could be that the presence of
bulky benzyl groups in the center of the last two
dipeptides causes unexpected effects on activity or
perhaps the effects of the morpholine fragments are
being poorly modeled by the clog P descriptor.
Regression Equations
chick biventer
ED50 (µM)
compd
pCBa
clog Pb
11a
11b
11c
20a
20c
20d
20e
20b
345
283
47
140
0.4
4.2
0.7
69
134
407
19.8
0.2
0.76
0.41
3.46
3.55
4.33
3.85
6.40
5.38
6.15
4.16
3.87
3.39
4.70
6.70
6.12
6.39
-4.12
-2.97
-1.41
-2.4
0.31
-0.02
1.57
-1.25
-2.93
-1.81
2.01
The mechanism of the NMB activity of the new
peptide-derived compounds described above appears to
be non-depolarizing (competitive). In contrast to the
depolarizing agent, suxamethonium (2), the new com-
pounds do not produce either muscle fasciculation or
pre-/post-twitch augmentation in the cat or monkey in
vivo. In the chick biventer preparation some compounds
produced slowly developing small increases or decreases
in the baseline tension, the mechanism of which is
unknown, but which indicates a change in basic muscle
tone. In contrast, suxamethonium (2) produced an
immediate concentration-dependent increase in the
baseline tension which in a multifocally innervated
chick muscle is indicative of a depolarizing action.
Upon the basis of the above biological data, two
tripeptide derivatives were selected for evaluation in
anesthetized Rhesus monkeys. Both 20c,e are effective
NMBs (ED50 ) 0.68, 0.23 µM/kg) with a rapid onset (1.7,
2.9 min, Table 5). The duration of action of NMBs is a
critical factor in determining their clinical applications,
and in comparison with the reference compounds rocu-
ronium (4) and suxamethonium (2) these peptides are
of similar potency and onset time but have longer
duration and slower recovery, which may be a less
attractive profile for short-duration surgery.
15a
15b
15c
pancuronium, 3
rocuronium, 4
atracurium, 5
1.21
0.6
2.36
a
b
-log(ED50). Calculated log P (ref 28).
curonium (3), rocuronium (4), and atracurium (5).
Regression analysis on this dataset gives the QSAR
shown in Figure 2 and eq 2:
predicted pED50 ) 0.59(clog P) + 5.46
n ) 11; r2 ) 0.88; F ) 69.2; s ) 0.46
(2)
In eq 2, n is the number of observations, r is the
correlation coefficient, F is the F-value, and s is the
standard error.
Equation 1 shows that in a homologous series of
NMBs, increasing the overall lipophilicity results in
increased activity in vitro. Equation 2 demonstrates that
this relationship between clog P and activity is a more
general trend, which applies to different structural
classes of non-depolarizing NMBs.
Applying eq 2 to the dipeptide compounds 15a -c
gives mixed results. In the case of 15c, the data are
complicated by an increase in baseline tension. Equation
2 would predict a pED50 of 3.73 for 15a , in good
agreement with the experimentally determined value
of 3.87. The predicted pED50 values for 15b,c would be
4.39 and 6.64, compared to experimental values of 3.39
and 4.7, respectively. Thus eq 2 is less predictive in
Con clu sion s
Di- and tripeptide moieties are shown to be suitable
scaffolds for designing compounds with potent in vitro
and in vivo NMB activity by appending two quaternary
nitrogen atoms. In this context the peptide scaffold can
be regarded as an alternative to previously reported
ones based, for example, on steroids, alkaloids, or
aliphatic esters. The dipeptide based bisquaternary
derivative 15c demonstrates NMB activity in vitro and
in an anesthetized cat assay at doses approximately
100-fold higher than clinically used reference com-

NMB of Bisquaternary Amino Di- and Tripeptides
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 25 4829
Hz), 3.55 (t, 2H, J ) 6 Hz), 3.36 (s, 9H), 3.2 (s, 9H); MS M -
H 259.7; HPLC >98%. Anal. (C12H28N4Cl2O2‚2.5H2O‚1.5NaCl)
C, H, N; Cl: calcd, 26.74; found, 29.85.
pounds. The tripeptide derivative 20e is more potent
than 15c in vitro and in vivo in cats, and in Rhesus
monkeys 20e exhibits comparable potency and onset/
time course profile as atracurium (5) and rocuronium
(4). Within this series clog P appears to correlate well
with the observed in vitro NMB activity in the chick
biventer assay. Taken together these results contribute
to the design of potentially short-acting NMB drugs.
N-ter t-Bu tyloxyca r bon yl-glycyl-glycin e-2-(4-m or p h oli-
n o)eth yla m id e (7b). Compound 6 (2.3 g, 10 mmol) was
reacted with 4-(2-aminoethyl)morpholine (1.9 mL, 10 mmol)
using conditions and reagent quantities similar to those
described for the synthesis of 7a , giving 2.1 g of the product
(61% yield): 1H NMR (CDCl3) δ 7.47 (m, 1H), 6.95 (m, 1H),
5.74 (m, 1H), 3.95 (m, 4H), 3.85 (d, 2H, J ) 7 Hz), 3.38 (m,-
4H), 2.45 (m, 6H), 1.45 (s, 9H).
Exp er im en ta l Section
N-Glycyl-glycin e-2-(4-m or ph olin o)eth ylam ide (8b). Com-
pound 7b (2 g, 5.8 mmol) was deprotected using conditions
and reagent quantities similar to those described during the
synthesis of 8a , giving 1.4 g of the product (99% yield): 1H
NMR (D2O) δ 3.96 (s, 2H), 3.82 (m, 6H), 3.45 (t, 2H, J ) 6
Hz), 2.85 (m, 6H); MS M + H 245.1.
Gen er a l Exp er im en ta l - Ch em istr y. All reagent chemi-
cals were obtained from commercial suppliers and used
without further purification except those described in the text
which were prepared by published procedures. Elemental
analyses of the final products were performed using a Perkin-
Elmer 2400 CHN analyzer and counterions were evaluated
by Dionex ion chromatography. All 1H NMR spectra were
recorded at 200 or 400 MHz on Bruker AM200 or 400
spectrometers, and chemical shifts are reported in ppm relative
to TMS or a soluble sodium derivative. Mass spectra, where
quoted, were recorded with a Finnigan MAT 90 mass spec-
trometer for fast atom bombardment spectra using glycerol
or thioglycerol as matrix and an HP 5989 MS engine in positive
mode for electrospray spectra, more suitable for the bisqua-
ternary products. HPLC data were obtained using a Perkin-
Elmer Integral 4000LC system run on Turbochrom Worksta-
tion software version 6.1.1 and a Phenomenex Aqua C18 25
× 0.46 cm reverse-phase column. Elution solvents were 0.1%
formic acid (or trifluoroacetic acid) in water (A) and 0.1% acid
in acetonitrile (B) using a gradient elution method of 95/5 A/B
for 5 min, then up to 10/90 over 15 min, and finally 10/90 for
10 min. Detection was carried out by a Sedex 55 evaporative
light scattering detector or Perkin-Elmer UV detector at 230
nM.
N,N,N-Tr im eth yl-glycyl-glycin e-2-[4-(4-m eth ylm or ph oli-
n o)]eth yla m id e Dich lor id e (9b). Prepared from compound
8b (250 mg, 1.02 mmol) using conditions and reagent quanti-
ties similar to those described for the synthesis of 9a , giving
209 mg of the product (55% yield): 1H NMR (D2O) δ 4.25 (s,
2H), 4.08 (m, 4H), 4.04 (s, 2H), 3.75 (m, 2H), 3.69 (m, 2H),
3.60 (m, 4H), 3.35 (s, 9H), 3.29 (s, 3H); MS M - H 301.7; HPLC
>95%. Anal. (C14H30N4Cl2O3‚2H2O‚NaCl) C, H, N, Cl.
N-ter t-Bu tyloxyca r bon yl-glycyl-glycin e-2-(1-p yr r olid i-
n o)eth yla m id e (7c). Compound 6 (2.3 g, 10 mmol) was
reacted with 1-(2-aminoethyl)pyrrolidine (1.14 g, 10 mmol)
using conditions and reagent quantities similar to those
described for the synthesis of 7a . The product proved to be
soluble in water, and the aqueous layer was basified with 4 N
NaOH then extracted with DCM giving after drying over Na2-
SO4 and evaporation 1.3 g of the product (40% yield): 1H NMR
(CDCl3) δ 7.27 (s, 1H), 7.05 (s, 1H), 5.72 (s, 1H), 3.96 (d, 2H,
J ) 6 Hz), 3.85 (d, 2H, J ) 6 Hz), 3.42 (q, 2H, J ) 6 Hz), 2.65
(t, 2H, J ) 6 Hz), 2.56 (m, 4H), 1.78 (m, 4H), 1.45 (s, 9H).
N-Glycyl-glycin e-2-(1-pyr r olidin o)eth ylam ide (8c). Com-
pound 7c (1.3 g, 4 mmol) was deprotected using conditions and
reagent quantities similar to those described for the synthesis
of 8a , giving 700 mg of the product (78% yield): 1H NMR (D2O)
δ 3.88 (s, 2H), 3.35 (m, 4H), 2.6 (m, 6H), 1.74 (m, 4H).
N,N,N-Tr im eth yl-glycyl-glycin e-2-[1-(1-m eth ylp yr r oli-
d in o)]eth yla m id e Dich lor id e (9c). Prepared from compound
8c (170 mg, 0.75 mmol) using conditions and reagent quanti-
ties similar to those described for the synthesis of 9a , giving
101 mg of the product (38% yield): 1H NMR (D2O) δ 4.24 (s,
2H), 4.0 (s, 2H), 3.75 (t, 2H, J ) 8 Hz), 3.5 (m, 6H), 3.35 (s,
9H), 3.11 (s, 3H), 2.2 (m, 4H); MS M - H 285.7; HPLC >97%.
Anal. (C14H30N4Cl2O2‚3H2O‚NaCl) C, H, N; Cl: calcd, 22.63;
found, 21.58.
N-ter t-Bu t yloxyca r bon yl-glycyl-glycin e-2-(1,2,3,4-t et -
r a h yd r o-2-isoqu in olin o)eth yla m id e (7d ). This compound
was prepared by reacting 6 (2.3 g, 10 mmol) in dry DMF (50
mL) with 2-(2-aminoethyl)-1,2,3,4-tetrahydroisoquinoline17 (1.72
g, 9.75 mmol) in the presence of HOBT (1.35 g, 10 mmol) and
DCC (2 g, 10 mmol) stirring for 24 h at room temperature.
DCU was removed by filtration, and the DMF was evaporated
under reduced pressure. The residue was partitioned between
ethyl acetate and 5% NaHCO3 and finally brine. The product
proved to be soluble in water, and the aqueous layer was
basified with 4 N NaOH then extracted with DCM. The organic
layers were combined and dried over Na2SO4 then evaporated
to give 2.44 g of the product (63% yield): 1H NMR (CDCl3) δ
7.14-7.05 (m, 5H), 6.88 (s, 1H), 5.38 (s, 1H), 3.9 (d, 2H, J ) 7
Hz), 3.74 (d, 2H, J ) 8 Hz), 3.63 (s, 2H,), 3.47 (q, 2H, J ) 6
Hz), 2.9 (m, 2H), 2.76 (m, 2H), 2.55 (t, 2H, J ) 6 Hz), 1.45 (s,
9H).
N-ter t-Bu t yloxyca r b on yl-glycyl-glycin e-2-(ter t-b u t yl-
oxyca r bon yla m in o)eth yla m id e (7a ). This compound was
prepared by reacting N-tert-butyloxycarbonyl-glycyl-glycine (6)
(3.3 g, 14 mmol) in dry DMF (50 mL) with N-tert-butyloxy-
carbonylethylenediamine16 (1.6 g, 10 mmol) in the presence
of HOBT (2 g, 15 mmol) and DCC (ca u tion ! ir r ita n t,
sen sitizer , a n d p oten tia l ca r cin ogen ; 2 g, 10 mmol) stirring
for 24 h at room temperature. Dicyclohexylurea (DCU) was
removed by filtration, and the DMF was evaporated under
reduced pressure. The residue was partitioned between ethyl
acetate and 0.4 N HCl then 5% NaHCO3 and finally brine.
The product was isolated by drying over Na2SO4 and evapora-
1
tion to give an oil: 2.8 g (76% yield; H NMR (CDCl3) δ 7.37,
7.35 (2s, 2H), 5.75 (s, 1H), 5.45 (s, 1H), 3.95 (d, 2H, J ) 8 Hz),
3.84 (d, 2H, J ) 8 Hz), 3.35 (m, 2H), 3.26 (m, 2H), and 1.43 (s,
9H); MS M + H 375.2.
N-Glycyl-glycin e-2-a m in oeth yla m id e (8a ). Compound
7a (2.2 g, 5.9 mmol) was dissolved in TFA (22 mL). After 30
min at room temperature the excess TFA was evaporated and
the residue was partitioned between diethyl ether and water.
The aqueous layer contains the product as the TFA salt, which
can be liberated by treatment with Dowex Ira 400 resin, 20-
50 mesh. The product was obtained by evaporation as a gum:
1
1 g (98% yield); H NMR (D2O) δ 3.9 (s, 2H), 3.4 (s, 2H), 3.35
(m, 2H), 2.86 (t, 2H, J ) 8 Hz); MS M + H 175.1.
N,N,N-Tr im eth yl-glycyl-glycin e-2-(N′,N′,N′-tr im eth yl-
a m m on io)eth yla m id e Dich lor id e (9a ). Compound 8a (250
mg, 1.44 mmol) was dissolved in MeOH (10 mL) and reacted
with iodomethane (1.4 mL, 23 mmol) in the presence of
NaHCO3 (1.6 g, 9.5 mmol), stirring the mixture at room
temperature for 72 h. The reaction was filtered, the filtrate
evaporated and the residue dissolved in the minimum of water
and chromatographed on a column of Sephadex G-10 (170 g,
previously swollen in water). The fractions containing the
product were evaporated under reduced pressure giving the
very hygroscopic quaternary ammonium hydroxide, which was
converted to the corresponding dichloride by ion exchange
using Dowex 1 × 8 200-400 mesh resin: 165 mg (34% yield);
1H NMR (D2O) δ 4.25 (s, 2H), 4.03 (s, 2H), 3.78 (t, 2H, J ) 6
N-Glycyl-glycin e-2-(1,2,3,4-tetr a h yd r o-2-isoqu in olin o)-
eth yla m id e (8d ). Compound 7d (500 mg, 1.28 mmol) was
deprotected using conditions and reagent quantities similar
to those described during the synthesis of 8a , giving 322 mg
of the product (87% yield): 1H NMR (D2O) δ 7.18 (m, 3H), 7.1
(m, 1H), 3.9 (s, 2H), 3.7 (s, 2H), 3.45 (t, 2H, J ) 8 Hz), 3.44 (s,
2H), 2.85 (m, 4H), 2.73 (t, 2H, J ) 8 Hz).

4830 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 25
Booij et al.
N,N,N-Tr im eth yl-glycyl-glycin e-2-(2-m eth yl-1,2,3,4-tet-
r a h yd r o-2-isoqu in olin o)eth yla m id e Dich lor id e (9d ). Pre-
pared from compound 8d 150 mg, 0.52 mmol) using conditions
and reagent quantities similar to those described for the
synthesis of 9a , giving 101 mg of the product (83% yield): 1H
NMR (D2O) δ 7.46 (m, 3H), 7.2 (m, 1H), 4.81 (s, 2H), 4.62 (d,
2H, J ) 6 Hz), 4.17 (s, 2H), 3.95 (s, 2H), 3.77 (2t, 4H, J ) 6
Hz), 3.55 (t, 2H, J ) 6 Hz), 3.27 (s, 9H), 3.23 (s, 3H); MS M -
H 347.2; HPLC >89%. Anal. (C19H32N4Cl2O2‚7H2O‚NaCl) C,
Cl; H: calcd, 8.07; found, 6.22; N: calcd, 9.75; found, 9.26.
N-ter t-Bu tyloxycar bon yl-glycyl-glycin e-2-[(m eth ylph en -
ylm eth yl)a m in o]eth yla m id e (7e). Compound 6 (2.3 g, 10
mmol) was reacted with (2-aminoethyl)benzylmethylamine17
(1.6 g, 9.75 mmol) using conditions and reagent quantities
similar to those described for the synthesis of 7d , giving 3.5 g
of the product (92% yield): 1H NMR (CDCl3) δ 7.29 (m, 5H),
7.2 (s, 1H), 6.7 (s, 1H), 5.55 (s, 1H), 3.92 (d, 2H, J ) 7 Hz),
3.83 (d, 2H, J ) 7 Hz), 3.5 (s, 2H), 3.35 (q, 2H, J ) 8 Hz), 2.5
(t, 2H, J ) 8 Hz), 2.2 (s, 3H), 1.45 (s, 9H).
N-Glycyl-glycin e-2-[(m eth ylp h en ylm eth yl)a m in o]eth -
yla m id e (8e). HCl gas (1.1 g, 30.2 mmol) was dissolved in
dry ethyl acetate (25 mL) and this solution was added to
compound 7e (3.5 g, 9.25 mmol) in ethyl acetate (35 mL). The
solution was stirred for 24 h at room temperature after which
the product as its HCl salt could be removed by filtration. The
salt (2.04 g, 5.8 mmol) was dissolved in MeOH (20 mL) and
treated with a solution of KOH (766 mg, 5.8 mmol) in MeOH
(20 mL) with stirring at room temperature for 15 min, then
the solvent was removed under reduced pressure. The residue
was extracted with DCM and filtered, giving on evaporation
1.3 g of product (81% yield): 1H NMR (CDCl3) δ 7.83 (s, 1H),
7.29 (m, 5H), 6.48 (s, 1H), 3.94 (d, 2H, J ) 5 Hz), 3.51 (s, 2H),
3.41 (s, 2H), 3 37 (q, 2H, J ) 5 Hz), 2.5 (t, 2H, J ) 5 Hz), 2.21
(s, 3H), 1.8 (s, 2H).
N -(1-P ip e r id in o m e t h y lc a r b o n y l)g ly c y l-g ly c in e -2-
[(m et h ylp h en ylm et h yl)a m in o]et h yla m id e (10). Piperi-
dineacetic acid 12b19 (700 mg, 4.9 mmol) was dissolved with
slight heating in DCM (50 mL) and treated with compound
8e (1.3 g, 4.7 mmol) followed by HOBT (635 mg, 4.7 mmol)
and DCC (968 mg, 4.7 mmol). The mixture was stirred at room
temperature for 60 h, then the solvent was reduced to low
volume and DCU was removed by filtration. The filtrate was
shaken with 5% Na2CO3 (3 × 20 mL) dried over Na2SO4 and
evaporated, giving 1.5 g of the product (80% yield): 1H NMR
(CDCl3) δ 7.9 (m, 1H), 7.29 (m, 5H), 6.85 (m, 1H), 6.45 (m,
1H), 4.0 (d, 2H, J ) 5 Hz), 3.9 (d, 2H, J ) 4 Hz), 3.54 (s, 2H),
3.45 (q, 2H, J ) 5 Hz), 3.0 (s, 2H), 2.5 (m, 6H), 2.22 (s, 3H),
1.6 (m, 5H), 1.5 (m, 1H).
1-[1-[2-[N-(P h en ylm eth yl)-N,N-d im eth yla m m on io]eth -
yla m in o]-N-(1-m eth yl p ip er id in om eth ylca r bon yl)glycyl]-
glycin e Diiod id e (11a ). Compound 10 (300 mg, 0.74 mmol)
was dissolved in acetonitrile (15 mL) and stirred at room
temperature for 60 h with iodomethane (0.6 mL, 9.6 mmol).
The solvent was removed under reduced pressure and the
residue was triturated with acetone and diethyl ether giving
370 mg of the product as an amorphous solid (73% yield): 1H
NMR (DMSO) δ 8.85 (m, 1H), 8.44 (m, 1H), 8.24 (m, 1H), 7.55
(s, 5H), 4.6 (s, 2H), 4.15 (s, 2H), 3.88 (d, 2H, J ) 4 Hz), 3.75
(d, 2H, J ) 4 Hz), 3.64 (m, 2H), 3.5 (m, 6H), 3.25 (s, 3H), 3.02
(s, 6H), 1.85 (m, 4H), 1.57 (m, 2H); MS M - H 433.4; HPLC
>98%. Anal. (C23H39N5I2O3‚0.5H2O) C, H, I; N: calcd, 10.34;
found, 9.64.
1-[1-[2-[N-Met h yl-N-(p h en ylm et h yl)-N-(2-p r op en yl)-
a m m on io]eth yla m in o]-N-[1-(2-p r op en yl)p ip er id in om eth -
ylca r bon yl]glycyl]glycin e Dibr om id e (11b). Compound 10
(300 mg, 0.74 mmol) was dissolved in acetonitrile (15 mL) and
stirred at room temperature for 60 h with allyl bromide (0.3
mL, 3.5 mmol), excluding light from the reaction. The solvent
was removed under reduced pressure and the residue was
triturated with acetone and diethyl ether giving 400 mg of the
product as an amorphous solid (83% yield): 1H NMR (DMSO)
δ 8.94 (m, 1H), 8.45 (m, 1H), 8.32 (m, 1H), 7.54 (m, 5H), 6.13
(m, 2H), 5.65 (m, 4H), 4.6 (s, 2H), 4.25 (d, 2H, J ) 6 Hz), 4.15
(m, 3H), 3.95 (d, 1H, J ) 6 Hz), 3.88 (d, 2H, J ) 6 Hz), 3.72 (d,
2H, J ) 6 Hz), 3.6 (m, 4H), 3.45 (m, 4H), 2.95 (s, 3H), 1.85 (m,
4H), 1.55 (m, 2H); MS M - H 484.6; HPLC >73%. Anal.
(C27H43N5Br2O3‚0.5H2O) C, H, Br; N: calcd, 10.97; found,
10.45.
1-[1-[2-[N-Meth yl-N-(4-n itr op h en ylm eth yl)-N-(p h en yl-
m eth yl)a m m on io]eth yla m in o]-N-[1-(4-n itr op h en ylm eth -
yl)p ip er id in om eth ylca r bon yl]glycyl]glycin e Dibr om id e
(11c). Compound 10 (300 mg, 0.74 mmol) was reacted with
4-nitrobenzyl bromide (321 mg, 1.49 mmol) using conditions
similar to those described for the synthesis of 11a , giving 460
mg of the product (77% yield): 1H NMR (DMSO) δ 8.98 (m,
1H), 8.51 (m, 1H), 8.35 (m, 5H), 7.96 (m, 2H), 7.88 (m, 2H),
7.62 (m, 2H), 7.54 (m, 3H), 5.05 (s, 2H), 5.0-4.5 (m, 4H), 4.05
(s, 2H), 3.95 (d, 2H, J ) 4 Hz), 3.75 (m, 4H), 3.58 (m, 4H),
3.28 (m, 2H), 2.98 (s, 3H), 1.9 (m, 4H), 1.61 (m, 2H); MS M -
H 674.2; HPLC >97%. Anal. (C35H45N7Br2O7‚H2O) H, N; C:
calcd, 49.25; found, 50.63; Br: calcd, 18.72; found, 17.86.
N,N-(Dip h en ylm eth yl)glycyl-glycin e (13a ). N,N-Diben-
zylglycine (12a )20 (2.55 g, 10 mmol) was suspended in DCM
(100 mL) then glycine ethyl ester HCl salt (1.39 g, 10 mmol)
was added followed by HOBT (1.35 g, 10 mmol), DCC (2 g, 10
mmol) and finally triethylamine (1.39 mL, 10 mmol). The
resulting solution was stirred at room temperature for 60 h,
the volume was reduced and DCU was removed by filtration.
The solvent was evaporated and the residue was dissolved in
ethyl acetate then shaken with 5% Na2CO3 (3 × 20 mL), dried
and evaporated to give 3.2 g of the ester as a gum (94% yield):
1H NMR (CDCl3) δ 7.72 (m, 1H), 7.35 (m, 10H), 4.25 (q, 2H, J
) 6 Hz), 4.0 (d, 2H, J ) 4 Hz), 3.62 (s, 4H), 3.15 (s, 2H), 1.29
(t, 3H, J ) 6 Hz). The ester (3.2 g, 9.4 mmol) was dissolved in
MeOH (30 mL) and treated with 4 N NaOH (2.49 mL. 9.9
mmol) for 2 h at room temperature. 5 N HCl (1.99 mL, 9.9
mmol) was added and the solvent was removed under reduced
pressure. The residue was extracted with DCM, filtered
through Celite and evaporated to give 2.8 g of the product (95%
yield): 1H NMR (CDCl3) δ 7.78 (m, 1H), 7.32 (m, 10H), 4.5 (s,
1H), 3.91 (d, 2H, J ) 2 Hz), 3.67 (s, 4H), 3.19 (s, 2H).
N,N-(Dip h en ylm eth yl)glycyl-glycin e-2-(1-p ip er id in o)-
eth yla m id e (14a ). Compound 13a (1 g, 3.2 mmol) was reacted
with 2-aminoethylpiperidine (0.45 mL, 3.2 mmol) in the
presence of HOBT (430 mg, 3.2 mmol) and DCC (660 mg, 3.2
mmol) in DCM (20 mL) using conditions similar to those
described for the synthesis of 10, giving 1.3 g of the product
(98% yield): 1H NMR (as di-HCl salt in D2O) δ 7.52 (s, 10H),
4.5 (s, 4H), 3.95 (s, 2H), 3.76 (s, 2H), 3.55 (m, 4H), 3.22 (t, 2H,
J ) 6 Hz), 2.95 (t, 2H, J ) 8 Hz), 1.95-1.55 (m, 6H).
[2-(N-Met h ylp ip er id in o)et h yla m in o][N-m et h yl-N,N-
bis(p h en ylm eth yl)glycyl]glycin e Diiod id e (15a ). Prepared
from compound 14a using conditions and reagent quantities
similar to those described during the synthesis of 11a , giving
300 mg of the product (60% yield): 1H NMR (DMSO) δ 8.95
(m, 1H), 8.35 (m, 1H), 7.58 (m, 10H), 4.85 (q, 4H, J ) 10 Hz),
3.85 (m, 4H), 3.56 (m, 2H), 3.38 (m, 6H), 3.07 (s, 3H), 3.02 (s,
3H), 1.78 (m, 4H), 1.54 (m, 2H); MS M - H 451.0; HPLC >97%.
Anal. (C27H40N4I2O2‚0.25H2O) C, H, N, I.
N-(1-P iper idin om eth ylcar bon yl)-L-ph en ylalan in e (13b).
Compound 12b19 (700 mg, 5 mmol) was reacted with L-
phenylalanine methyl ester HCl salt (1.08 g, 5 mmol) using
conditions and reagent quantities similar to those described
for the synthesis of 13a giving 1.5 g of the ester (99% yield):
1H NMR (CDCl3) δ 7.71 (d, 1H, J ) 6 Hz), 7.28 (m, 2H), 7.15
(m, 3H), 4.9 (m, 1H), 3.74 (s, 3H), 3.15 (t, 2H, J ) 6 Hz), 2.9
(d, 2H, J ) 4 Hz), 1.85 (m, 4H), 1.45 (m, 6H). The ester (1.4 g,
4.6 mmol) was hydrolyzed in dioxane (14 mL) using conditions
and reagent quantities similar to those described during the
synthesis of 13a , giving 540 mg of the product (41% yield):
1H NMR (CDCl3) δ 8.45 (d, 1H, J ) 4 Hz), 7.25 (m, 5H), 5.0 (s,
1H), 4.69 (m, 1H), 3.55 (d, 1H, J ) 8 Hz), 3.29 (m, 2H), 2.95
(m, 1H), 2.46 (m, 4H), 1.66 (m, 4H), 1.46 (m, 2H).
N-(1-P ip er id in om eth ylca r bon yl)-L-p h en yla la n in e-2-(4-
m or p h olin o)eth yla m id e (14b). Compound 13b (540 mg, 1.9
mmol) was reacted with 4-(2-aminoethyl)morpholine (0.24 mL,
1.9 mmol) using conditions and reagent quantities similar to
those described for the synthesis of 14a giving 700 mg of the

NMB of Bisquaternary Amino Di- and Tripeptides
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 25 4831
product (93% yield): 1H NMR (CDCl3) δ 7.56 (d, 1H, J ) 6
Hz), 7.25 (m, 5H), 6.25 (m, 1H), 4.65 (q, 1H, J ) 6 Hz), 3.61
(m, 4H), 3.26 (q, 2H, J ) 4 Hz), 3.08 (d, 2H, J ) 6 Hz), 2.89 (d,
2H, J ) 6 Hz), 2.31 (m, 10H), 1.5 (m, 6H).
2-[N-(4-Meth ylm or p h olin o)]eth yla m in o-N-(1-m eth ylp i-
per idin om eth ylcar bon yl)-L-ph en ylalan in e Diiodide (15b).
Prepared from compound 14b using conditions and reagent
quantities similar to those described during the synthesis of
15a , giving 370 mg of the product (73% yield): 1H NMR (D2O)
δ 7.35 (m, 5H), 4.71 (t, 1H, J ) 6 Hz), 4.05 (m, 6H), 3.65 (m,
2H), 3.42 (m, 10H), 3.15 (s, 3H), 3.13 (m, 2H), 3.08 (s. 3H),
1.87 (m, 4H), 1.55 (m, 2H); MS M - H 431.5; HPLC >98%.
Anal. (C24H40N4I2O3‚0.75H2O) C, H, N, I.
2-N-[4-(4-Nitr op h en ylm eth yl)m or p h olin o]eth yla m in o-
N-[1-(4-n itr op h en ylm eth yl)p ip er id in om eth ylca r bon yl]-
L-p h en yla la n in e Dibr om id e (15c). Prepared from compound
14b using conditions and reagent quantities similar to those
described during the synthesis of 11c, giving 460 mg of the
product (58% yield): 1H NMR (DMSO) δ 9.17 (d, 1H, J ) 4
Hz), 8.78 (m, 1H), 8.35 (d, 2H, J ) 6 Hz), 8.29 (d, 2H, J ) 6
Hz), 7.97 (d, 2H, J ) 6 Hz), 7.77 (d, 2H, J ) 6 Hz), 7.28 (m,
4H), 7.22 (m, 1H), 5.0 (s, 2H), 4.88 (m, 2H), 4.75 (m, 1H), 3.98
(m, 6H), 3.75 (m, 2H), 3.53 (m, 4H), 3.43 (m, 2H), 3.34 (m,
2H), 3.25 (m, 2H), 3.15 (m, 1H), 2.82 (m, 1H), 1.85 (m, 2H),
1.75 (m, 2H), 1.55 (m, 1H), 1.46 (m, 1H); MS M - H 673.6;
HPLC >98%. Anal. (C36H46N6Br2O7‚0.75H2O) C, H, N; Br:
calcd, 18.21; found, 17.74.
N-Meth yl-N-(1-piper idin om eth ylcar bon yl)glycin e (17a).
Compound 12b19 (1.29 g, 9 mmol) was reacted with sarcosine
ethyl ester HCl salt (1.31 g, 9 mmol) using conditions and
reagent quantities similar to those described for the synthesis
of 13a giving 1.7 g of the ester (83% yield): 1H NMR (CDCl3)
δ 4.22 (m, 4H), 3.19 (s, 1H), 3.17 (s, 1H), 3.08, 2.99 (2s, 3H),
2.44 (m, 4H), 1.5 (m, 6H), 1.3 (m, 3H). The ester (1.7 g, 7 mmol)
was hydrolyzed using conditions and reagent quantities
similar to those described for the synthesis of 13a , giving 1.5
g of the product (99% yield): 1H NMR (CDCl3) δ 6.25 (s, 1H),
3.91-3.78 (3s, 4H), 3.35 (m, 3H), 3.21 (m, 1H), 3.06-3.0 (2s,
3H), 1.88 (m, 4H), 1.62 (m, 2H).
during the synthesis of 11a , giving 450 mg of the product (92%
yield): 1H NMR (DMSO) δ 7.56 (m, 5H), 4.68 (m, 2H), 4.52
(m, 2H), 4.32 (m, 2H), 4.22 (m, 2H), 3.85 (m, 2H), 3.72 (m,
2H), 3.48 (m, 4H), 3.25 (m, 2H), 3.02 (m, 14H), 2.85 (m, 2H),
1.85 (m, 4H), 1.55 (m, 2H); MS M + I 602.2 (M - H not found);
HPLC >99%. Anal. (C26H45N5I2O3‚H2O) C, H, N; I: calcd,
33.95; found, 36.50.
1-[1-[2-[N-Met h yl-N-(p h en ylm et h yl)-N-(2-p r op en yl)-
a m m on io]eth yl-N-m eth yla m in o]-N-[1-(2-p r op en yl)p ip e-
r id in om et h ylca r b on yl]-N-m et h ylglycyl]-N-m et h ylgly-
cin e Dibr om id e (20b). Prepared from compound 19a (300
mg, 0.67 mmol) using conditions and reagent quantities similar
to those described during the synthesis of 11b, giving 340 mg
of the product (73% yield): 1H NMR (CDCl3) δ 7.28 (m, 2H),
7.47 (m, 3H), 6.05 (m, 2H), 5.75 (m, 4H), 5.18 (m, 2H), 4.85
(m, 2H), 4.65 (m, 2H), 4.45 (m, 4H), 4.28 (m, 2H), 4.15 (m,
6H), 3.52 (m, 2H), 3.25 (m, 6H), 2.95 (m, 6H), 1.95 (m, 6H);
MS M - H 526.7; HPLC >98%. Anal. (C30H49N5Br2O3‚2H2O)
H, N, Br; C: calcd, 49.80; found, 50.50.
1-[1-[2-[N-Meth yl-N-(4-n itr op h en ylm eth yl)-N-(p h en yl-
m et h yl)a m m on io]et h yl-N-m et h yla m in o]-N-[1-(4-n it r o-
p h en ylm eth yl)p ip er id in om eth ylca r bon yl]-N-m eth ylgly-
cyl]-N-m eth ylglycin e Dibr om id e (20c). Prepared from
compound 19a (300 mg, 0.67 mmol) using conditions and
reagent quantities similar to those described during the
synthesis of 11c, giving 450 mg of the product (76% yield):
1H NMR (D2O) δ 8.38 (m, 1H), 8.25 (m, 3H), 7.65 (m, 9H), 4.97
(s, 2H), 4.69 (m, 4H), 4.53 (s, 2H), 4.25 (m, 2H), 4.22 (s, 2H),
4.11 (m, 2H), 3.88 (m, 2H), 3.68 (m, 2H), 3.54 (m, 2H), 3.15
(m, 4H), 3.05 (m, 4H), 2.95 (m, 4H), 2.05 (m, 4H), 1.78 (m,
2H); HPLC >97%. Anal. (C38H51N7Br2O7‚3.75H2O) C, H, N, Br.
N-Meth yl-N-(1-p ip er id in oeth ylca r bon yl)glycin e (17b).
1-Piperidinepropionic acid (16) (7.68 g, 50 mmol) was reacted
with sarcosine ethyl ester HCl salt (7.3 g, 47.5 mmol) and
DIPEA (8.28 mL, 47.5 mmol) using conditions and reagent
quantities similar to those described for the synthesis of 13a
giving 4.35 g of the ester (36% yield): 1H NMR (CDCl3) δ 4.22
(m, 4H), 3.09, 2.98 (2s, 3H), 2.68 (m, 4H), 2.45 (m, 4H), 1.61
(m, 4H), 1.45 (m, 2H), 1.29 (m, 3H). The ester (4.35 g, 17 mmol)
was hydrolyzed using conditions and reagent quantities
similar to those described for the synthesis of 13a , giving 3 g
of the product (67% yield), as the HCl salt: 1H NMR (D2O) δ
4.38 (m, 1H), 3.95 (m, 1H), 3.71 (m, 4H), 3.18 (m, 7H), 2.11
(m, 4H), 1.81 (m, 2H).
N-Meth yl-N-[N-m eth yl-N-(1-p ip er id in oeth ylca r bon yl)-
glycyl]glycin e (18b). Compound (17b) (3 g, 11.3 mmol) was
reacted with sarcosine ethyl ester HCl salt (1.77 g, 11.5 mmol)
and DIPEA (4.2 mL, 22 mmol) using conditions and reagent
quantities similar to those described for the synthesis of 17a
giving 3.7 g of the ester, which was purified by chromatography
over silica (40 g), using DCM/MeOH/NH3 (500/10/1) as eluant,
which gave 2.3 g (62% yield): 1H NMR (CDCl3) δ 4.17 (m, 6H),
3.10, 3.07, 2.96, 2.95 (4s, 6H), 2.56 (m, 4H), 2.45 (m, 4H), 1.58
(m, 4H), 1.45 (m, 2H), 1.28 (m, 3H). The ester (2.3 g, 7 mmol)
was hydrolyzed in dioxane (20 mL), using conditions and
reagent quantities similar to those described for the synthesis
of 17a , giving 2.15 g of the product (89% yield), as the HCl
salt: 1H NMR (D2O) δ 4.45 (m, 4H), 3.92 (2s, 1H), 3.68 (m,
2H), 3.55 (m, 2H), 3.15 (m, 9H), 2.05 (m, 4H), 1.85 (m, 2H).
N-Meth yl-N-[N-m eth yl-N-(1-p ip er id in oeth ylca r bon yl)-
glycyl]glycin e-N-m eth yl-2-[(m eth yl-1-p h en ylm eth yl)a m i-
n o]eth yla m id e (19b). Compound 18b (2.15 g, 6.4 mmol) was
reacted with 2-(methylaminoethyl)benzylmethylamine18 (1.08
g, 6.1 mmol) using conditions and reagent quantities similar
to those described for the synthesis of 19a , giving on extraction
2.8 g of the product, which was chromatographed using similar
elution techniques giving 1.55 g (55% yield): 1H NMR (CDCl3)
δ 7.28 (s, 5H), 4.17 (m, 3H), 3.51 (m, 3H), 3.28 (m, 1H), 3.05
(m, 10H), 2.81 (m, 3H), 2.53 (m, 7H), 2.24 (m, 3H), 1.62 (m,
4H), 1.49 (m, 2H).
N-Meth yl-N-[N-m eth yl-N-(1-p ip er id in om eth ylca r bon -
yl)glycyl]glycin e (18a ). Compound 17a (1.5 g, 7 mmol) was
reacted with sarcosine ethyl ester HCl salt (1.02 g, 7 mmol)
using conditions and reagent quantities similar to those
described for the synthesis of 17a giving 1.4 g of the ester (64%
yield): 1H NMR (CDCl3) δ 4.28-4.13 (m, 6H), 3.19 (s, 2H),
3.17 (s, 2H), 3.08 (m, 6H), 2.44 (m, 3H), 2.38 (m, 1H), 1.62 (m,
2H), 1.42 (m, 2H), 1.25 (m, 3H). The ester (1.4 g, 4.5 mmol)
was hydrolyzed using conditions and reagent quantities
similar to those described for the synthesis of 17a , giving 1.3
1
g of the product (99% yield): H NMR (CDCl3) δ 6.75 (s, 1H),
4.24 (m, 2H), 3.92 (m, 4H), 3.32 (m, 4H), 2.99 (m, 6H), 1.88
(m, 4H), 1.62 (m, 2H).
N-Meth yl-N-[N-m eth yl-N-(1-p ip er id in om eth ylca r bon -
yl)glycyl]glycin e-N-m eth yl-2-[(m eth yl-1-p h en ylm eth yl)-
a m in o]eth yla m id e (19a ). Compound 18a (3.5 g, 12.27 mmol)
was reacted with 2-(methylaminoethyl)benzylmethylamine18
(1.97 g, 11.05 mmol) in DMF (30 mL) in the presence of DIPEA
(4.3 mL, 22.1 mmol) and HBTU (4.65 g, 12.27 mmol), stirring
at room temperature for 24 h. The solution was added to water
(200 mL) and extracted with DCM (100 mL + 50 mL) and the
combined extracts were washed with brine then evaporated
to give a gum, which was dried under high vacuum to remove
residual DMF. The product was purified using flash chroma-
tography over fine silica (30 g) and gradient elution (DCM-
DCM/MeOH 90/10-DCM/MeOH/NH3 190/10/2) giving 3.5 g
(71% yield): 1H NMR (CDCl3) δ 7.28 (m, 5H), 4.2 (m, 2H), 3.52
(m, 3H), 3.3 (m, 1H), 3.18 (m, 1H), 3.15 (m, 2H), 3.05 (m, 3H),
2.96 (m, 4H), 2.8 (m, 1H), 2.5 (m, 6H), 2.24 (m, 3H), 1.85 (s,
2H), 1.58 (m, 4H), 1.44 (m, 2H).
1-[1-[2-[N,N-Dim eth yl-N-(p h en ylm eth yl)a m m on io]eth -
yl-N-m eth yla m in o]-N-(1-m eth ylp ip er id in om eth ylca r bo-
n yl)-N-m et h ylglycyl]-N-m et h ylglycin e Diiod id e (20a ).
Prepared from compound 19a (300 mg, 0.67 mmol) using
conditions and reagent quantities similar to those described
1-[1-[2-[N-Meth yl-N-(4-n itr op h en ylm eth yl)-N-(p h en yl-
m et h yl)a m m on io]et h yl-N-m et h yla m in o]-N-[1-(4-n it r o-
ph en ylm eth yl)piper idin oeth ylcar bon yl]-N-m eth ylglycyl]-
N-m eth ylglycin e Dibr om id e (20d ). Compound 19b (600 mg,

4832 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 25
Booij et al.
1.31 mmol) was reacted with 4-nitrobenzyl bromide (680 mg,
3.1 mmol) in acetonitrile (8 mL) for 24 h at room temperature.
The solvent was removed and the residue was chromato-
graphed over alumina (18 g) using DCM/2-propanol (5/1) as
eluant, followed by precipitation of the product from acetone
and ether, giving 200 mg (17% yield): 1H NMR (DMSO) δ 8.31
(m, 4H), 7.95 (m, 2H), 7.81 (m, 2H), 7.64 (m, 2H), 7.51 (m,
3H), 4.8 (m, 4H), 4.55 (m, 2H), 4.29 (m, 3H), 3.93 (m, 2H), 3.48
(m, 4H), 3.32 (m, 4H), 2.99 (m, 11H), 2.81 (m, 4H), 1.86 (m,
4H), 1.55 (m, 2H); MS M - H 730.8; HPLC >91%. Anal.
(C39H53N7Br2O7‚4H2O) C, H, N, Br.
N-Meth yl-N-[N-m eth yl-N-(1-p ip er id in om eth ylca r bon -
yl)glycyl]glycin e-1-(4-p h en ylm et h yl)p ip er a zin e Am id e
(19c). Compound 18a (3 g, 10.5 mmol) was reacted with
1-benzylpiperazine (1.66 g, 9.5 mmol) using conditions and
reagent quantities as described for the synthesis of 19a , giving
3.05 g of the product after chromatography (65% yield): 1H
NMR (CDCl3) δ 7.31 (s, 5H), 4.21 (m, 4H), 3.51 (m, 8H), 3.15
(m, 4H), 3.07 (m, 4H), 2.45 (m, 6H), 1.55 (m, 4H), 1.42 (m,
2H).
1-[[N-(4-Nitr op h en ylm eth yl)-N-(p h en ylm eth yl)p ip er a -
zin -4-yl]-N-[1-(4-n it r op h en ylm et h yl)p ip er id in om et h yl-
ca r bon yl]-N-m eth ylglycyl]-N-m eth ylglycin e Dibr om id e
(20e). Compound 19c (1.5 g, 3.38 mmol) was reacted with
4-nitrobenzyl bromide (1.9 g, 8.79 mmol) using conditions and
reagent quantities as described for the synthesis of 20d , giving
after chromatography 520 mg of the product (18% yield): 1H
NMR (D2O) δ 8.36 (m, 4H), 7.75 (m, 4H), 7.56 (m, 5H), 4.86
(m, 6H), 4.38 (m, 4H), 4.15 (m, 6H), 3.85 (m, 2H), 3.62 (m,
6H), 3.08, 2.98, 2.93, 2.89 (4s, 6H), 2.02 (m, 4H), 1.75 (m, 2H);
MS M - H 714.7; HPLC >95%. Anal. (C38H49N7Br2O7‚4H2O)
C, H, N; Br: calcd, 16.86; found, 16.41.
of twitch tension were made on an ink writing chart recorder,
via Grass FT03 force displacement transducers. After an initial
30-min stabilization period, drugs were added cumulatively
to the tissue bath (normally 4-5 concentrations) to produce
85%-95% twitch block. Concentration-twitch inhibition re-
sponse lines were constructed using the method of least
squares, and concentrations of the compounds producing 50%
twitch block were determined.
b. An esth etized ca t: Neuromuscular blocking experiments
in cats were carried out using similar methodology to that
described previously.12 Cats of either sex (1.9-2.6 kg) were
anesthetized with a mixture of R-chloralose (80 mg/kg) and
sodium pentobarbitone (5 mg/kg) injected intraperitoneally.
The lungs were ventilated with room air at a rate of 28 breaths
min-1 and at a stroke volume (12-15 mL/kg) to maintain
arterial blood pH between 7.3 and 7.4. Twitch responses of
the tibialis anterior muscle were elicited every 10 s (0.1 Hz),
by stimulating the sciatic nerve with rectangular pulses of
0.25-ms duration and at supramaximal voltage, and quanti-
tated using a Grass FT03 force displacement transducer, and
recorded on a Grass chart recorder. Arterial pressure was
measured via a catheter placed in the right carotid artery and
connected to a Gould-Statham pressure transducer. Heart rate
was monitored continuously using the arterial pressure pulse
to trigger a Grass cardiotachometer. Signals were amplified
and recorded on a Grass model 7 chart-recorder. Rectal
temperature was monitored and maintained at 37 ( 1 °C.
Drugs were administered through a catheter in the right
external jugular vein.
Effects on autonomic mechanisms were measured simulta-
neously every 100 s, by recording the effect of the drugs on
decreases in heart rate, produced by stimulating the vagus
nerve for 10 s at 2-5 Hz, and on nictitating membrane
contractions, induced by stimulating the cervical sympathetic
nerve. Toward the end of some experiments, logarithmically
increasing doses of the compounds were injected at 100-s
intervals to obtain blocks of the vagus nerve-induced brady-
cardia and cervical sympathetic nerve-induced contractions of
the nictitating membrane.
c. An esth etized m on k eys: Experiments were carried out
in fasted Rhesus monkeys (5.0-6.5 kg) premedicated with
intramuscular atropine sulfate (0.25 mg) to prevent excessive
salivation. Anesthesia was induced with 10 mg/kg ketamine
intramuscular. The trachea was intubated without the use of
a muscle relaxant. Anesthesia was maintained using a loading
iv bolus dose of pentobarbitone sodium (15-25 mg/kg) followed
by an infusion, which was adjusted to produce a clinically
steady depth of anesthesia. Artificial ventilation was applied
(60% N2O in oxygen) with an Ohmeda ventilator and end-tidal
carbon dioxide concentration was measured and maintained
at 4-4.5%. Heart rate was measured continuously from the
ECG signal (automatic counting of QRS-complexes). A catheter
was introduced percutaneously into a superficial vein in one
of the hind limbs for drug administrations. Arterial blood
pressure was recorded continuously using a Finapress nonin-
vasive automated blood pressure monitor. The ulnar nerve was
stimulated with rectangular electrical pulses (0.1 Hz) via
bipolar subcutaneous needle electrodes. The resulting twitches
of the adductor pollicis muscle were recorded using a Statham
UC3 transducer, attached by a wire to a U-clamp, which was
fixed to the basal phalanx of the thumb. All parameters were
recorded on a Nihon-Kohden chart recorder and on a computer
hard-disk for further analysis.
P r otocol. The effects of the drugs on neuromuscular
transmission were determined in cats and monkeys, by inject-
ing increasing doses of the neuromuscular blocking drugs at
30-60-min intervals to produce a range of levels of neuro-
muscular block, i.e. between 10% and 95% inhibition of twitch
tension. Dose-response lines for the data were plotted from
regression analysis of points lying between 15% and 85% of
maximum block, to confine the analysis to the linear portion
of the log dose-inhibition line. Doses producing 50% inhibition
of electrically induced contractions of the muscles were
calculated by linear interpolation of these lines. Doses of the
Molecu la r Mod elin g - Exp er im en ta l. X-r a y d a ta : Com-
pounds were extracted from the Cambridge Crystallographic
Structure Database29 as Fdat files and read into Chem-X,30
where the interquaternary nitrogen distances were measured.
Con for m a tion a l a n a lyses: Crude range-finding confor-
mational analyses were performed using Chem-X. For the
tripeptide derivative 11a , a grid search through all the
nonterminal freely rotatable torsion angles was carried out.
Amide bonds were excluded and remained trans throughout.
All non-amide bonds between the two keto groups nearest the
two quaternary nitrogens were rotated through 360° in 60°
degree increments. All other non-amide bonds between the two
quaternary nitrogens were rotated through 360° in 120°
increments. The default Chem-X parameter files, Gasteiger
charges, and molecular mechanics force field were used
throughout. The minimum energy conformation had an inter-
quaternary distance of 1.43 nm. This distance varied between
1.15 and 1.49 nm in conformations within 10 kcal/mol of the
search minimum.
For the dipeptide derivative 9a , a similar crude conforma-
tional analysis was performed. Again all amide bonds were
held trans throughout. All other nonterminal bonds between
the two quaternary nitrogens were rotated through 360° in
60° degree increments, with the exception of the only CH2-
CH2 bond, where 120° increments were used. The conforma-
tional search energy minimum conformation had an interqua-
ternary distance of 1.08 nm. This distance varied between 0.91
and 1.13 nm in conformations within 10 kcal/mol of the search
28
minimum. Clog P values were calculated using PCModels.
Plots were produced using Microsoft Excel.31
P h a r m a cologica l - E xp er im en t a l. Neu r om u scu la r
Stu d ies. a . Isola ted ch ick biven ter cer vicis m u scle:
Biventer cervicis nerve muscle preparations from young
chickens (3-12 day old) were dissected and mounted in Krebs
Henseleit solution of the following composition: NaCl 118, KCl
5, KH2PO4 1, MgSO4 1, NaHCO3 30, CaCl2 2.5, and glucose
20 mmol/L. The bathing fluid was bubbled with 5% carbon
dioxide in oxygen to maintain a pH of 7.4. The methodology
for setting up this preparation was essentially similar to that
described by Ginsborg and Warriner.32 The motor nerves were
stimulated at 10-s intervals (0.1 Hz) with rectangular pulses
of 0.2-ms duration and at a supramaximal voltage. Recordings

NMB of Bisquaternary Amino Di- and Tripeptides
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 25 4833
(13) Cheng, Y. S.; Konteatis, Z. D.; Macielag, M. J .; Palmer, D. C.
Peptide Skeletal Muscle Relaxants. Eur. Pat. Appl. EP 498,941,
Aug 19, 1992; Chem. Abstr. 1993, 118, 81439e.
(14) Konteatis, Z. D.; Palmer, D. C. Peptide amides and amide dimers.
Eur. Pat. Appl. EP 498,940, Aug 19, 1992; Chem. Abstr. 1993,
118, 39426y.
(15) Konteatis, Z. D.; Palmer, D. C. Eur. Pat. Appl. 0 498 940 A2.
Cheng, Y. S.; Konteatis, Z. D.; MacLelag, M. J .; Palmer, D. C.
Eur. Pat. Appl. 0 498 941 A2.
(16) N-tert-Butyloxycarbonylethylenediamine was prepared in 82%
yield by the method of: Krapcho, A. P.; Kuell, C. S. Monopro-
tected Diamines. N-tert-Butyloxycarbonyl-ω,ω-Alkanediamines
from ω,ω- alkanediamines. Synth. Commun. 1990, 20, 2559-
2564.
(17) (2-Aminoethyl)-1,2,3,4-tetrahydroisoquinoline was prepared in
25% yield by the method of: Mull, R. P. Dibenzopolymethylen-
iminoalkyleneguanidines. U.S. Pat. 3,093,632; Chem. Abstr.
1963, 59, 12771f. (2-Aminoethyl)benzylmethylamine was pre-
pared by the same method in 80% yield.
(18) (Methylaminoethyl)benzylmethylamine was prepared from (2-
aminoethyl)benzylmethylamine (prepared as in ref 17 above) by
formylation then reduction of the N-formyl intermediate with
LiAlH4 in 60% yield.
neuromuscular blocking compounds producing 50% block of
the induced bradycardia and nictitating membrane contrac-
tions were calculated similarly.
Time course measurements were made using doses of the
drugs producing 80-95% block (cats) and also at 3 × 90%
blocking doses (monkeys). Onset time was measured from drug
injection to the first maximally depressed twitch, recovery25-75%
time was time taken for the twitches to recover from 25-75%
of control, and duration times were from injection to 90%
spontaneous recovery compared to predrug twitches.
Ack n ow led gm en t. We acknowledge and thank the
Departments of Analytical Chemistry (Organon, Scot-
land, and N V Organon, The Netherlands) for generat-
ing spectroscopic and elemental analytical results, and
we acknowledge and thank Rona Mason, Kenneth
Anderson, and Francien van der Pol for their expert
technical assistance in performing pharmacology ex-
periments. We thank Margaret Logue for assistance in
preparing the manuscript.
(19) Ursy, Y.; Paty, M. The action of haloacetates on piperidine-the
methyl chloroacetates. Compt. Rend. 1961, 252, 3812-3814;
Chem. Abstr. 1961, 55, 23524i.
(20) N,N-Dibenzylglycine was prepared in 55% yield by the method
of: Birkofer, L. Regularities in the Hydrogenative Fission of
N-Benzyl compounds. Chem. Ber. 75B, 429-441; Chem. Abstr.
1943, 37, 3067(9).
(21) Stahl, G. L.; Walter, R.; Smith, C. W. General Procedure for the
Synthesis of Mono-N-acylated 1,6-Diaminohexanes. J . Org.
Chem. 1978, 43, 2285-2286.
(22) Chen, F. C. M.; Benoiton, N. L. A New Method of Quaternising
Amines and its use in Amino Acid and Peptide Chemistry. Can.
J . Chem. 1976, 54, 3310-3311.
Refer en ces
(1) Rees, D. C.; Hill, D. R. Drugs in Anesthetic Practice. In Annual
Reports in Medicinal Chemistry; Bristol, J ., Ed.; Academic
Press: New York, 1996; Vol. 31, Chapter 5, pp 41-50.
(2) Standaert, F. G. Basic physiology and pharmacology of the
neuromuscular junction. In Anesthesia, 2nd ed.; Miller, R. D.,
Ed.; Churchill Livingstone: New York, 1986; pp 835-870.
(3) Martin-Smith, M. Rational elements in the development of
superior neuromuscular blocking drugs. In Drug Design; Ariens,
E. J ., Ed.; Academic Press: New York, 1971; pp 453-530.
(4) Kimura, M.; Kimura, I.; Muroi, M.; Tanaka, K.; Nojima, H.;
Uwano, T.; Koizumi, T. The Structure Activity Relationship
between phenylene-polymethylene Bis-ammonium derivatives
and their Neuromuscular Blocking Action on Mouse Phrenic
Nerve-Diaphragm Muscle. Biol. Pharm. Bull. 1994, 17, 1224-
1231.
(23) Knorr, R.; Trzeciak, A.; Bannwarth, W.; Gillessen, D., New
Coupling Reagents in Peptide Chemistry, Tetrahedron Lett.
1989, 30, 1927-1930.
(5) Miller, R. D.; Savarese, J . J . Pharmacology of muscle relaxants
and their antagonists. In Anesthesia, 2nd ed.; Miller, R. D., Ed.;
Churchill Livingstone: New York, 1986; pp 889-943.
(6) Dhar, N. C.; Maehr, R. B.; Masterson, L. A.; Midgley, J . M.;
Stenlake, J . B.; Wastila, W. B. Approaches to short acting
neuromuscular blocking agents: nonsymmetrical bis-tetrahy-
droisoquinolinium mono- and diesters. J . Med. Chem. 1996, 39,
556-561.
(24) Savage, D. S.; Cameron, A. Forbes; Ferguson, George; Han-
naway, C.; Mackay, I. R. Molecular structure of pancuronium
bromide (3,17â-diacetoxy-2â,16â-dipiperidino-5R-androstane
dimethobromide), a neuromuscular blocking agent. Crystal and
molecular structure of the water: methylene chloride solvate.
J . Chem. Soc. B 1971, 3, 410-415.
(25) Mndzhoyan, A. L.; Avoyan, R. L.; Avetisyan, A. A.; Arutyunyan,
E. G. Conformation and physiological activity of molecules. II.
X-ray structural analysis of ditilin. Arm. Khim. Zh. 1972, 25,
710-717.
(26) J ensen, B. The crystal structure of succinylcholine chloride
dihydrate. Acta Chem. Scand., Ser. B 1976, B30, 1002-1004.
(27) Randall, L. O. The conversion of decamethonium-like agents to
tensilon-reversible agents by aromatic substituents J . Pharma-
col. Exp. Ther. 1952, 105,16-26.
(7) Marshall, I. G.; Harvey, A. L. Selective neuromuscular blocking
properties of alpha-conotoxins in vivo. Toxicon 1990, 28, 231-
234.
(8) Dershwitz, M.; Roscow, C. E. Remifentanil: an opioid metabo-
lised by esterases. Exp. Opin. Invest. Drugs 1996, 5, 1361-1376.
(9) Boros, E. E.; Bigham, E. C.; Boswell, G. E.; Mook, R. A., J r.;
Patel, S. J .; Savarese, J . J .; Ray, J . A.; Thompson, J . B.; Hashim,
M. A.; Wisowaty, J . C.; Feldman, P. L.; Samano, V. Bis- and
Mixed-Tetrahydroisoquinolinium Chlorofumarates: New Ultra-
Short-Acting Nondepolarising Neuromuscular Blockers. J . Med.
Chem. 1999, 42, 206-209.
(28) CLOGP PCModels, version 4.61; Daylight Chemical Information
Systems Inc., Irvine, CA.
(10) Bowman, W. C.; Rodger, I. W.; Houston, J .; Marshall, R. J .
Structure: action relationships among some analogues of pan-
curonium and vecuronium in the anesthesized cat. Anesthesiol-
ogy 1988, 69, 57-62.
(29) CSD, version 512; Cambridge Crystallographic Data Centre,
Cambridge, U.K.
(30) The Chem-X suite of Programs; developed by CDL, Oxford
Molecular Group, Oxford, U.K.
(31) Microsoft Excel 97; Microsoft Corp.
(32) Ginsborg, B. L.; Warriner, J . The isolated chick biventer cervicis
nerve-muscle preparation. Br. J . Pharmacol. 1960, 15, 410-411.
(11) Kopman, A. F. Pancuronium, gallamine, and d-tubocurarine
compared: is speed of onset inversely related to drug potency?
Anesthesiology 1989, 70, 915-920.
(12) Marshall, I. G.; Gibb, A. J .; Durant, N. N. Neuromuscular and
vagal blocking actions of pancuronium bromide, its metabolites,
and vecuronium bromide (ORG NC 45) and its potential me-
tabolites in the anesthesised cat. Br. J . Anaesthesia 1983, 55,
703-714.
J M0010062