Homologation of polyamines in the rapid synthesis of lipospermine conjugates and related lipoplexes

Article abstract of DOI:10.1016/S0040-4020(99)01082-0

Lipopolyamine amides are useful cationic lipids, synthetic vectors for non-viral gene delivery. Desymmetrisation of readily available symmetrical polyamines is an important first step in the synthesis of such compounds. The application of trifluoroacetyl

Full text of DOI:10.1016/S0040-4020(99)01082-0

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 2449–2460
Homologation of Polyamines in the Rapid Synthesis of
Lipospermine Conjugates and Related Lipoplexes
*
Andrew J. Geall and Ian S. Blagbrough
Department of Pharmacy and Pharmacology, University of Bath, Bath BA2 7AY, UK
Received 20 August 1999; revised 24 November 1999; accepted 9 December 1999
Abstract—Lipopolyamine amides are useful cationic lipids, synthetic vectors for non-viral gene delivery. Desymmetrisation of readily
available symmetrical polyamines is an important first step in the synthesis of such compounds. The application of trifluoroacetyl as a
protecting group allows unsymmetrical polyamine amides to be rapidly prepared. A reductive alkylation homologation strategy allows the
sequential, regiocontrolled introduction of additional positive charges. Tetraamine spermine and other polyamine derivatives have been
N1-acylated with various single alkyl chains, and their relative binding affinities for DNA determined using an ethidium bromide displace-
ment assay. The important effects on DNA binding affinity of the number of positive charges on the polyamine moiety and also the nature
(chain length and degree of unsaturation) of the covalently attached lipid are demonstrated. ᭧ 2000 Elsevier Science Ltd. All rights reserved.
In our studies of polyamines and polyamine amides,1–11 we
are investigating the triamine spermidine 1 and the tetra-
amine spermine 2. These naturally occurring linear amines
are found in most living cells and play important roles in
vivo. Maintaining the 3D structure of DNA12–16 by conden-
sation11,17,18 is one of these crucial biological roles. Spermi-
dine 1 and spermine 2 both contain a 3–4 methylene spacing
between the amino functional groups that means that these
molecules are essentially fully protonated at physiological
pH (i.e. ammonium ions at pHˆ7.4).19 Therefore, as
polycations they interact readily with the DNA phosphate
anionic backbone, causing condensation by charge neutra-
lisation and this effect is a key first step in minimising the
size of foreign DNA for gene therapy.13–18
As part of our continuing studies on polyamine-mediated
However, these polyamine–DNA interactions are readily
reversible under physiological conditions20 and form one
of the plethora of roles played by spermidine 1 and spermine
2 in vivo, together with polycationic histones (natural
polymers rich in the basic amino acids arginine and/or
lysine).21–23 Structure–activity relationship (SAR) studies
with polyamines (for a review, see: Blagbrough et al.11)
have shown that these molecules are ideally suited to bind
to and then condense DNA.24 In order to reinforce these
effects, it is apparently beneficial if a lipid moiety is
covalently bound to the polyamine, such a lipid can be
cholesterol,24,25 a bile acid,26 or an aliphatic chain.27–29
DNA condensation,30–33 we have developed a rapid
synthetic route to unsymmetrically protected spermine,30
then homologated this compound31 in order to allow the
introduction of another secondary amine and hence an
additional positive charge. Covalent binding of different
lipids, palmitic (hexadecanoic), stearic (octadecanoic),
oleic (cis-9-octadecenoic) and elaidic (trans-9-octa-
decenoic) acids, then allows SAR to be determined for
their binding to DNA.
Recently, we and others have shown that polyamines and
polyamine amides can be prepared by reductive alkyl-
ation,7–9,34,35 consecutive Michael additions to acrylo-
nitrile,34,36 or regioselective acylation of unsymmetrically
protected polyamines.1–4,34–37 Tetraamine spermine 2 is
readily available, an ideal starting material to incorporate
three (or four) positive charges into a target molecule.
However, the desymmetrisation protocol is by nature low
yielding and often involves laborious chromatographic
Keywords: polyamine amides; spermine; DNA condensation; DNA binding
affinity; reductive alkylation; lipoplex.
* Corresponding author. Tel.: ϩ1225-826795; fax: ϩ1225-826114;
e-mail: prsisb@bath.ac.uk
0040–4020/00/$ - see front matter ᭧ 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(99)01082-0

2450
A. J. Geall, I. S. Blagbrough / Tetrahedron 56 (2000) 2449–2460
purifications. Such low yielding and repetitive steps are not
efficient on a gram scale. There are problems with efficient
syntheses of N1-mono-Z- and N1-mono-Boc-spermine.
Using either ZCl together with sensitive pH control, or
(Boc)2O with the polyamine in large excess, was either
not practical or required time-consuming chromatographic
purification from the excess of unreacted polyamine.38–40 In
this paper, we report the practical synthesis of unsymmetri-
cal polyamine amides using trifluoroacetyl as a protecting
group whose introduction and removal can be controlled
under facile conditions and on a gram scale.
by treatment with trifluoroacetic acid gave the poly-trifluoro-
acetate salt of polyamine amides 8, 10, 12 and 14, respect-
ively. 3-Amino-propan-1-ol was Z-protected under
Schotten–Baumann conditions to afford alcohol 15. Swern
oxidation of the primary alcohol 15 with oxalyl chloride
activated DMSO gave aldehyde 16. Reductive alkylation
of the primary amine in 6 with aldehyde 16 afforded
protected polyamine 17. Protection of the newly introduced
secondary amine (N 4) was achieved with di-tert-butyl
dicarbonate to form fully protected polyamine 18. Hydro-
genation of the Z carbamate, in the presence of Pearlman’s
catalyst (Pd(OH)2), afforded protected unsymmetrical poly-
amine 19. N-Acylation of protected homologated spermine
19 with hexadecanoic acid, mediated by DCC and catalytic
HOBt afforded tetra-Boc protected lipospermine 20. Depro-
tection by treatment with trifluoroacetic acid gave the poly-
trifluoroacetate salt of polyamine amide 21. In this paper,
we report detailed NMR assignments of spermine (TFA
salt) and N1-hexadecanoyl-1,16-diamino-4,8,13-triazahexa-
decane 21. The target compounds have been named as their
corresponding spermine (1,12-diamino-4,9-diazadodecane)
derivatives (Scheme 1).
The ratio of primary amine to protecting group reagent is
critical in order to avoid di-protection (of both primary
amines) and poly-protection (including secondary
amines).38 Presumably, the higher nucleophilicity of the
secondary amines is masked by corresponding steric
effects.34 The facile and specific (for primary over second-
ary amines) introduction of trifluoroacetyl using ethyl
trifluoroacetate, as reported recently,41 and its ready
removal with aqueous ammonia42 (pHˆ11) or with methan-
olic aqueous K2CO3 solution43 makes it a superior protect-
ing group to carbobenzoxy (Z, CBZ) and to tert-
butoxycarbonyl (Boc) for the purpose of gram scale
protection of polyamines. Thus, trifluoroacetyl is the
protecting group of choice, over Z and Boc, for practical
routes to unsymmetrical polyamine amides and carbamates.
Therefore, using this strategy, we prepared N1,N2,N3-tri-
Boc-spermine 6. We have prepared unsymmetrical poly-
amine amides, which are charged at physiological pH and
therefore interact with DNA. The syntheses of target lipo-
spermidines 8, 10, 12 and 14 that mimic the charge distri-
bution of spermidine 1, but are covalently attached to
different lipids, are outlined. The charge distribution of
spermine 2 is mimicked using reductive alkylation on the
poly-protected spermine 6, to form pentaamine 19 intro-
ducing a secondary amine and hence an additional charge.
Covalent attachment of palmitic acid, leads to target
compound 21. We are utilising the charge distribution
found in the natural polyamines spermidine 1 and spermine
2 as biomimetic warheads for the efficient condensation of
DNA, an essential first step in non-viral gene delivery.
Lipopolyamine conjugates 8, 10, 12, 14 and 21 interact with
DNA (forming lipoplexes44) as demonstrated by an
ethidium bromide (Eth Br) fluorescence quenching
assay.45 Prevention of Eth Br binding to DNA is one method
of studying the DNA binding behaviour of small molecule
ligands, e.g. polyamines,45–52 though the DNA binding
modes of aliphatic polyamines and Eth Br, a polyaromatic
intercalator dye, are certainly different, however a quali-
tative comparison of DNA binding affinity between related
chemical structures is possible.50–52 Therefore, lipopoly-
amine amides 8, 10, 12, 14 and 21 can be critically
compared as a function of both the concentration and charge
ratio44 required to displace Eth Br binding to DNA. We have
previously determined the positive charge on the spermidine
headgroup of 3-cholesteryl carbamate analogues to be 2.4.32
Further, we have determined the positive charge on the
spermine headgroup potentiometrically as 3.4.
NMR spectroscopic structural assignments
In order to make complete and unambiguous H and 13C
1
Results and Discussion
Synthesis
NMR assignments for the polyamine headgroups, we have
first conducted NMR experiments on spermine 2 (Fig. 1). To
establish confidence in our analysis of spermine, we have
compared these data to the literature values53 and also to
those calculated using additivity rule calculations (Fig. 1).54
The techniques used for spermine 2 were then used for the
assignments of the polyamine amides. The assignment of
the spermidine headgroup in amides 8, 10, 12 and 14 is also
compared to a synthetic wasp toxin analogue, an amide
containing the same polyamine moiety (Fig. 2).55
Spermine 2 was selectively protected on one of the primary
amines with ethyl trifluoroacetate in order to afford mono-
trifluoroacetamide 3, but also affording di-trifluoroacet-
amide 4. Immediately, in this solution, the remaining free
amines were Boc protected with di-tert-butyl dicarbonate to
afford compound 5. Selective deprotection of the trifluoro-
acetamide was then achieved by increasing the pH of the
solution above 11 with conc. aq. ammonia to afford poly-
amine 6 with a free primary amine unmasked. N-Acylation
of this selectively protected spermine 6 with palmitic
(hexadecanoic) acid, stearic (octadecanoic), oleic (cis-9-
octadecenoic) and elaidic (trans-9-octadecenoic) acids,
mediated by DCC and catalytic 1-hydroxybenzotriazole
(HOBt) afforded tri-Boc protected lipopolyamine 7 (and
lipopolyamines 9, 11 and 13, respectively). Deprotection
The resonance of the methylene backbone of the free base of
spermine 2 can be found in three distinct regions,55–57
around 50 ppm resonate the methylene groups adjacent to
a secondary amine (C3, C5, C8, C10), around 40 ppm
methylene groups adjacent to a primary amino group (1
and 12) and around 30 ppm methylene groups separated
from nitrogen by at least one carbon on each side (C2,
C6, C7, C11). The protonation of amines causes a shielding

A. J. Geall, I. S. Blagbrough / Tetrahedron 56 (2000) 2449–2460
2451
Scheme 1.
of the carbon atoms in the vicinity of the nitrogen resulting
in an upfield shift in their signals. Methylene groups
positioned a to an amine are deshielded more, and therefore
have larger downfield 13C shifts than those positioned
further away.58,59 Methylene groups a to a secondary
amine have larger downfield chemical shifts than those a
to a primary amine.58 The upfield shift on protonation of
amines is detectable as far as five carbon atoms away, the
greatest effect being at the b-position.58,60 Thus, in the fully
protonated spermine species, carbons 5 and 8 have the
furthest downfield signals as they are both a and d to proto-
nated secondary amines (Fig. 1). C3 and C10 have signals
that are upfield from C5 and C8 as they are both located a to
a secondary and g to a primary protonated amines. C1 and
C12 come into resonance the furthest upfield of the methyl-
enes attached directly to an amine because they are a to a
primary and d to a secondary protonated amines (Fig. 1). C2
and C11 are influenced by two b protonated amines (one
primary and one secondary) and therefore they come into
resonance further downfield from C6 and C7, which are

2452
A. J. Geall, I. S. Blagbrough / Tetrahedron 56 (2000) 2449–2460
Figure 1. 13C NMR assignment for spermine 2. Literature values53 are in D2O at 40ЊC (tetrahydrochloride salt). The observed values are in D2O at 22ЊC
(tetratrifluoroacetate salt). Calculated values are estimates based on additivity rule calculations of 13C chemical shifts in aliphatic compounds.54
influenced by a b and a g protonated secondary amines (Fig.
1).
chemical shift predictions54 for fully protonated spermine
2 are 1.7 (C6, C7), 2.2 (C2, C11) and 2.94 ppm (C1, C3, C5,
C8, C10, C12). These are in good agreement with the
1
The chemical shifts calculated for spermine (Fig. 1) on the
basis of the additivity rules54 differ in general by about
5 ppm from the experimentally determined values. This
calculation method only claims to be within ϳ5 ppm of
the observed values and our findings are in agreement
with this. More importantly, these predicted values agree
with the order of the observed assignments, with C6 and
C7 coming into resonance nearest to TMS. 1H NMR
observed values (see Fig. 1), except that the H on C3 and
C10 are magnetically distinct from C1, C5, C8, C12 and
come into resonance further downfield. This can be
accounted for by the small deshielding effect of a g
protonated primary amine, not allowed for in the calcu-
lation. 1H, 13C NMR correlation spectroscopy confirms
these assignments of spermine 2 and supports the calcu-
lation methods.
Figure 2. 13C NMR assignment for philanthotoxin-3.4.3 22 in D2O at 25ЊC (trihydrochloride salt).55 The observed values for compound 8 are in D2O at 22ЊC
(tritrifluoroacetate salt). Calculated values are estimates based on additivity rule calculations of 13C chemical shifts in aliphatic compounds.54

A. J. Geall, I. S. Blagbrough / Tetrahedron 56 (2000) 2449–2460
2453
Figure 3. 13C NMR assignment for compound 21 are in DMSO at 22ЊC (tetratrifluoroacetate salt). Calculated values are estimates based on additivity rule
calculations of 13C chemical shifts in aliphatic compounds.54
1
N-Acylation of one of the primary amines of spermine leads
to an unsymmetrical polyamine and therefore loss of
symmetry of the chemical shifts in the propylene chains.
C1, C2 and C3 are influenced by an amide rather than a
protonated primary amine, and therefore are more shielded,
resonating further upfield than their counterparts, C10, C11
and C12, on the other propylene chain. The 13C assignments
for compound 8 (Fig. 2) compare favourably with the
assignments of the polyamine moiety in philanthotoxin-
3.4.3 2255 and the calculated values are within 5 ppm.
Using the additivity rules,54 we calculated that C12 will
resonate further upfield than C1, which is clearly not the
spectrum confirms the assignments. Calculation of the H
chemical shifts of a methylene group adjacent to an alkyl
amide (2.99 ppm) and
a protonated primary amine
(2.67 ppm) are also in agreement with this observation.
This allows the unequivocal identification of C1, which
couples to a signal at 35.6 ppm and allows the signal at
36.1 to be assigned to C16. C14 is chemically distinct
from C3, C5 and C7 because it is g to a protonated primary
amine, based on this fact and also on the corresponding
calculated values, C14 is assigned to the downfield signal
at 44.7. The assignments for C2, C15, C10, C11 and C6 are
based on H, 13C NMR chemical shift correlation spectro-
1
case experimentally. Comparisons of the H, 13C chemical
scopy. Thus, unambiguous total 13C NMR spectroscopic
assignments of the spermine headgroup are therefore
1
shift correlation spectra of spermine 2 and polyamine amide
8, together with the assignments of philanthotoxin-3.4.3
22,55 show that, in an unsymmetrical polyamine amide,
the protons on C1 resonate further upfield from those
protons adjacent to a protonated amine, C12.
based on calculations and by H, 13C NMR chemical shift
1
correlation spectroscopy.
Calculations of the 1H chemical shifts of a methylene group
adjacent to an alkyl amide (2.99 ppm) and a protonated
primary amine (2.67 ppm) are also in agreement with the
NMR correlation spectroscopy assignments. This allows us
to make an unequivocal identification of C1 as coupled to
the signal at 36.8 ppm. This therefore allows the signal at
37.3 to be assigned to C12. Thus, unambiguous total 13C
NMR spectroscopic assignments of the spermidine head-
group are based on the comparison with a literature
compound, calculations using additivity rules, and also by
1H, 13C NMR chemical shift correlation spectroscopy.
The 13C NMR assignment of compound 21 (Fig. 3) are in
good agreement with the calculated values. Even though the
calculated chemical shifts of C16 and C1 are not particularly
accurate (they only claim to be within an acceptable error of
^5 ppm), the 1H, 13C NMR chemical shift correlation
Figure 4. Eth Br fluorescence assay at pH 7.4, 20 mM NaCl, as a function
of charge ratio.

2454
A. J. Geall, I. S. Blagbrough / Tetrahedron 56 (2000) 2449–2460
Figure 5. Eth Br fluorescence assay at pH 7.4, 20 mM NaCl, as a function
of concentration.
Figure 6. Eth Br fluorescence assay at pH 7.4, 150 mM NaCl, as a function
of charge ratio.
DNA binding affinities
are a function of both the number of positive charge and the
structure of the lipid covalently attached to the polyamine.
Polyamine amides 10 and 21 displace 50% of the Eth Br at
much lower concentrations (1.40 and 1.85 mM) compared to
compounds 8, 12 and 14 (4.27, 2.85 and 2.56 mM, respec-
tively). Conjugate 21 has 3.4 positive charges distributed
along the polyamine headgroup compared to only 2.4 on
the other polyamine amides.32 Conjugate 10 has a saturated
C18 alkyl chain compared to the C16 alkyl chain on 21 and
8 and the unsaturated C18 alkyl chains on 12 and 14. In this
series, polyamine amide 10 has the greatest DNA binding
affinity. It is clear (see Figs. 6 and 7) that the C18 saturated
chain, rather than the C16 chain (10 cf. 8 and 21) affords
higher DNA binding affinity. We therefore conclude that the
nature of the lipid chain moiety attached to the polyamine is
a more critical function at higher salt concentrations than
even the number of positive charges. Higher concentrations
of lipopolyamines, at elevated salt concentrations, are also
required to displace the Eth Br, reflecting the salt dependent
DNA binding character of this type of compound.
As one part of our SAR studies, we have measured relative
DNA binding affinities using a fluorescence quenching
assay with Eth Br.31–33,49,52 Wilson and Bloomfield
predicted,17 using the polyelectrolyte theory of Manning,61
that when ϳ90% of the negative charges on DNA (from the
phosphate groups) are neutralised by the positive charges
along the polyammonium ion moiety, DNA condensation
will occur.17,18,62,63 Therefore, we have measured the
percentage fluorescence of intercalated Eth Br as a function
of charge ratio (positive/negative charges). DNA conden-
sation is clearly an efficient process with lipopolyamine
amides 8, 10, 12, 14 and 21 (Fig. 4), as nearly complete
exclusion of Eth Br occurs before the charge ratio of the
complex reaches one. The charge ratios at which 50% of the
Eth Br (CR50) are displaced are 0.52, 0.49, 0.50, 0.50 and
0.51 for 8, 10, 12, 14 and 21, respectively (Fig. 4). Aggre-
gation of the DNA probably accounts for the incomplete
exclusion of Eth Br.64
When these Eth Br fluorescence assay data for polyamine
amides 8, 10, 12, 14 and 21, at pH 7.4 (at 20 mM NaCl, Fig.
4) are represented as a function of concentration (mM, Fig.
5), conjugate 21 displaces 50% of the Eth Br at a much
lower concentration (0.90 mM) compared to conjugates
8 (1.35 mM), 10 (1.27 mM), 12 (1.29 mM) and 14
(1.29 mM). DNA binding affinity is a function of charge,
and therefore we expect that tetraamine 21 should bind
with a higher affinity than triamines 8, 10, 12 and 14 (at
low salt concentrations).
These data support our hypothesis that DNA binding affinity
and condensation are a sensitive function of both the
charge32 and hydrophobicity33 of this type of ligand. We
have used an adaptation of an Eth Br displacement assay
based on the work of Cain et al.52 Previously31–33 we used an
At 150 mM NaCl DNA condensation is still clearly an
efficient process with lipopolyamine amide 10 (CR50ˆ0.55,
Fig. 6). However, for analogous polyamine amides 12, 14
and 21, the process is less efficient (CR50ˆ1.12, 1.00 and
1.05, respectively). Lower DNA binding affinity (cf 10)
achieved by incorporation of a CyC double bond is
shown by conjugates 12 and 14. Conjugate 8 contains the
same spermidine headgroup as conjugate 10, but two less
methylenes in the alkyl chain and yet has a profoundly
different (weaker) binding affinity for DNA (CR50ˆ1.67
compared to 0.55). These data (at 150 mM NaCl), repre-
sented as a function of concentration (mM, Fig. 7), show
that the binding affinities to DNA of these lipopolyamines
Figure 7. Eth Br fluorescence assay at pH 7.4, 150 mM NaCl, as a function
of concentration.

A. J. Geall, I. S. Blagbrough / Tetrahedron 56 (2000) 2449–2460
2455
Eth Br exclusion assay based on the work of Gershon et al.49
This was both time consuming for the many data points
required, and produced variable results at key intermediate
polyamine amide to DNA charge ratios. Gershon et al.49
have demonstrated that at high and low charge ratios,
where the fluorescence intensity is at its extreme values,
no time dependence is observed. However, at the key inter-
mediate charge ratios during the DNA condensation
process, fluorescence is time dependent.49 The displacement
assay of Cain et al.52 has previously been used to compare
the binding affinity of both intercalating and non-inter-
calating drugs and provides rapid and comparable results
without any significant variability in fluorescence measure-
ment at intermediate concentrations. In the assay of Cain et
al.,52 the fluorescence enhancement was due to direct
excitation of the intercalated Eth Br (lexcitˆ546 nm,
lemissˆ595 nm). In our adaptation, we have indirectly
excited the Eth Br by energy transfer from the DNA, in a
similar manner to that used by Gershon, Minsky and
co-workers,49 lexcitˆ260 nm, this produces a much greater
(10-fold) fluorescence enhancement.
Coupling constants (J, absolute values) are expressed in
Hertz and the multiplicities are recorded as s (singlet), d
(doublet), t (triplet), q (quartet), quin. (quintet), m (multi-
plet) and br (broad). High and low resolution fast atom
bombardment (FAB) mass spectra were recorded on a
Fisons VG AutoSpec Q spectrometer, with m-nitrobenzyl
alcohol (mNBA) as the matrix. Microanalyses were
performed by the Microanalysis Laboratory, University of
Bath, but the presence of polyamines in the cationic lipids
makes elemental analysis inadequate as a criterion of purity.
1
Structures were unambiguously assigned using IR, H and
13C NMR and accurate MS although the elemental analysis
was not within ^0.4%.
Anhydrous methanol was prepared by distillation from
˚
magnesium turnings and iodine, and was stored over 3 A
molecular sieves under anhydrous nitrogen. Anhydrous
CH2Cl2 and DMSO were prepared by distillation from
˚
calcium hydride (5% w/v) and were stored over 4 A mole-
cular sieves. Analytical and semi-preparative RP–HPLC
were performed with a Jasco PU-980 pump equipped with
a Jasco UV-975 detector (lˆ220 nm). The column
stationary phase was Supelcosil ABZϩPlus, 5 mm
(15 cm×4.6 mm for the analytical and 25 cm×10 mm for
the semi-preparative columns). The mobile phase was an
isocratic mixture of methanol and 0.1% aqueous trifluoro-
acetic acid, with 1.5 and 4.0 ml/min flow rates for the
analytical and semi-preparative columns, respectively.
In this paper, we have designed and developed a novel
strategy, using trifluoroacetyl as a protecting group, to
allow the rapid synthesis of unsymmetrical lipopolyamines
starting from spermine 2 whose products incorporate the
positive charge distribution of spermidine 1. The application
of a homologation strategy, based on reductive alkylation, in
the synthesis of unsymmetrical lipopolyamines has also
been demonstrated to afford the four positive charges
distributed as in spermine 2. Using a modified Eth Br dis-
placement assay, we have established that the relative
binding affinity to DNA of these compounds is subtly
dependent upon the lipid covalently attached to the poly-
amine, the positive charge distributed along the polyamine
and also the salt concentration.
DNA binding affinities of lipopolyamine amides were
measured with an Eth Br fluorescence quenching assay
based upon the displacement of Eth Br (1.3 mM) from calf
thymus DNA (SAF). This assay is an adaptation of Cain et
al.,52 the method is rapid and involves the addition of micro-
litre aliquots of polyamine conjugate to a 3 ml solution of
Eth Br (1.3 mM) and calf thymus DNA (serially diluted and
determined spectroscopically44 as 6 mg, [DNA base-
pair]ˆ3.0 mM) in buffer (20 mM NaCl, 2 mM HEPES, pH
7.4) with the decrease in fluorescence (lexcitˆ260 nm,
lemissˆ600 nm; 1 cm path length glass cuvette) recorded
after 1 min equilibration time following each addition.
The decrease in fluorescence was critically compared for
compounds 8, 10, 12, 14 and 21 using both charge ratio44
(Fig. 4) and concentration (Fig. 5). Salt dependence of the
binding affinities of the conjugates has also been investi-
gated using this assay at physiological salt concentration
(150 mM NaCl, 2 mM HEPES, pH 7.4) and the decrease
in fluorescence was critically compared for both charge ratio
(Fig. 6) and concentration (Fig. 7). All chemicals, reagents
and buffers were purchased from SAF; solvents (HPLC
grade) were purchased from Fisons.
Experimental
Column chromatography was performed over silica gel 60
(35–75 mm) (Merck) purchased from Prolabo. Analytical
TLC was performed using aluminium-backed plates coated
with Kieselgel 60 F254, purchased from Merck. The
chromatograms were visualised with either potassium per-
manganate (basic aqueous) or ninhydrin (acidic butanolic).
Concentrated in vacuo means the use of a Bu¨chi Rotavapor
R-114 at water aspirator pressure. Melting point determi-
nations were carried out using a Reichert–Jung Thermo
Gfalen Kopfler block and are uncorrected. Polyamines are
highly hygroscopic and can adopt a different salt degree;29
determination of the melting points of their polytrifluoro-
acetic acid salts was not useful. IR spectra were recorded
(n cmϪ1) either as liquid films or as KBr discs using a
Perkin–Elmer 782 spectrometer, selected strong bands are
General procedure A: amine acylation
1
reported. H and 13C NMR spectra were recorded using
To 1.0 mmol of poly-Boc-protected polyamine dissolved in
DMF (5 ml), fatty acid (e.g. palmitic acid) (1.2 mmol),
1-HOBt (27 mg, 0.2 mmol) and DCC (308 mg, 1.5 mmol)
were added and the mixture was heated to 40ЊC and stirred
under nitrogen for 17 h. The solution was then concentrated
in vacuo (40ЊC) and the residue dissolved in CH2Cl2
(10 ml). The precipitate of DCU was removed by filtration,
the filtrate was concentrated in vacuo and the residue
1
JEOL 270 (operating at 270 MHz for H and 67.8 MHz
1
for 13C) or JEOL EX 400 (operating at 400 MHz for H
and 100.8 MHz for 13C) spectrometers. Chemical shift
values are recorded in parts per million (ppm) on the d
scale. Spectra were referenced internally to TMS, or using
the residual solvent resonance for 1H and 13C, or to
3-(trimethylsilyl)-propionic-2,2,3,3-d4 acid (sodium salt).

2456
A. J. Geall, I. S. Blagbrough / Tetrahedron 56 (2000) 2449–2460
purified over silica gel (EtOAc–hexane 50:50 to 60:40 v/v)
to afford the title compound as a colourless oil.
25.5, 25.5, 25.6, 25.8, 26.0 (6-CH2, 7-CH2, 30-CH2); 27.2,
27.7, 28.4, 28.8, 29.0 [2-CH2, 11-CH2, O–C–(CH3)3, over-
lapping]; 29.4, 29.5, 29.5, 29.7, 29.7 (40-CH2 to 130-CH2,
overlapping); 31.9 (140-CH2); 35.4, 35.9 (12-CH2); 37.0 (20-
CH2); 37.4, 37.7 (1-CH2); 43.1, 43.3, 43.8, 44.2 (3-CH2, 10-
CH2); 46.2, 46.7 (5-CH2, 8-CH2, overlapping); 79.0, 79.6,
79.6, 79.8 (quat. C×3); 155.5, 155.5, 156.1, 156.5 [N–CO–
O–C–(CH3)3, overlapping]; 173.5 (N–CO–CH2). MS,
FABϩ found 741, 70% (Mϩϩ1), C41H80N4O7 requires
Mϩˆ740. High-resolution MS: m/z, FABϩ found 741.6109,
(Mϩϩ1), C41H81N4O7 requires Mϩϩ1ˆ741.6105.
General procedure B: Boc removal
To a stirring solution of lipopolyamine dissolved in CH2Cl2
(3 ml), under nitrogen, at 25ЊC was added TFA (3 ml). After
2 h, the solution was concentrated in vacuo, lyophilised and
the residue was then purified by semi-preparative
RP–HPLC over Supelcosil ABZϩPlus (MeOH–0.1% aq.
TFA) to afford the title compound as a white solid, its poly-
trifluoroacetate salt.
N1-Hexadecanoyl-1,12-diamino-4,9-diazadodecane 8. Pro-
tected polyamine amide 7 (381 mg, 0.51 mmol) was depro-
tected according to general procedure B to afford the title
compound 8 as a white solid (polytrifluoroacetate salt
267 mg, 66%), tR 3.7 min by RP–HPLC (Supelcosil
ABZϩPlus, 5 mm, 15 cm×4.6 mm, MeOH–0.1% aq. TFA
(N1,N4,N9-Tri-tert-butoxycarbonyl)-1,12-diamino-4,9-
diazadodecane 6. To a solution of spermine 2 (1.0 g,
4.95 mmol) in methanol (70 ml), at Ϫ78ЊC under nitrogen,
ethyl trifluoroacetate (703 mg, 4.95 mmol) was added drop-
wise over 30 min. Stirring was continued for a further
30 min, then the temperature was increased to 0ЊC to afford
predominantly the mono-trifluoroacetamide 3. Without
isolation, the remaining amino functional groups were
quantitatively protected by dropwise addition of an excess
of di-tert-butyldicarbonate (4.23 g, 19.80 mmol, 4.0 equiv.)
in methanol (10 ml) over 3 min. The reaction was then
warmed to 25ЊC and stirred for a further 15 h to afford the
fully protected polyamine 5, Rf 0.6 (EtOAc). The trifluoro-
acetate protecting group was then removed in situ by
increasing the pH of the solution to above 11 with conc.
aq. ammonia and then stirring at 25ЊC for 15 h. The solution
was concentrated in vacuo and the residue purified over
silica gel (CH2Cl2–MeOH–conc. aq. NH3 70:10:1 to
50:10:1 v/v/v) to afford the title compound 6 as a colourless
homogeneous oil (1.24 g, 50%), Rf 0.5 (CH2Cl2–MeOH–
conc. aq. NH3 50:10:1 v/v/v). IR (film) 1670 (O–CO–N).
1H NMR, 400 MHz, CDCl3: 1.42–1.55 [m, 31H, 6-CH2, 7-
CH2, O–C–(CH3)3×3, overlapping]; 1.60–1.72 (m, 6H, 2-
CH2, 11-CH2, NH2); 2.70 (t, 2H, Jˆ7 Hz, 12-CH2); 3.05–
3.38 (m, 10H, 1-CH2, 3-CH2, 5-CH2, 8-CH2, 10-CH2); 5.29–
5.44 (br s, 1H, CO–NH–CH2). 13C NMR, 100 MHz, CDCl3:
25.4, 25.8, 25.9, 26.3 (6-CH2, 7-CH2); 28.35, 28.5, 28.7 [2-
CH2, O–C–(CH3)3]; 31.3, 32.5 (11-CH2); 37.3, 37.6 (1-
CH2); 38.8, 39.3 (12-CH2); 43.7, 44.1, 44.2, 44.4 (3-CH2,
10-CH2); 46.3, 46.7 (5-CH2, 8-CH2, overlapping); 78.8,
78.9, 79.1, 79.3 (quat. C×3); 155.3, 155.5, 155.6, 156.0
[N–CO–O–C–(CH3)3, overlapping]. MS, FABϩ found
503, 21% (Mϩϩ1), C25H50N4O6 requires Mϩˆ502. High-
resolution MS: m/z, FABϩ found 503.3823, (Mϩϩ1),
C25H51N4O6 requires Mϩϩ1ˆ503.3808.
1
70:30). IR (KBr) 1670 (CO–N). H NMR 400 MHz, D2O:
0.80–0.92 (m, 3H, 160-CH3); 1.15–1.38 (m, 24H, 40-CH2 to
150-CH2); 1.47–1.61 (m, 2H, 30-CH2); 1.70–1.84 (m, 4H, 6-
CH2, 7-CH2); 1.84–1.95 (m, 2H, 11-CH2); 2.09 (q, 2H,
Jˆ8 Hz, 2-CH2); 2.19 (t, 2H, Jˆ7, 20-CH2); 2.95–3.19
(m, 10H, 3-CH2, 5-CH2, 8-CH2, 10-CH2, 12-CH2); 3.19–
3.30 (m, 2H, 1-CH2). 13C NMR, 100 MHz, D2O: 14.6
(160-CH3); 23.4 (150-CH2); 23.6, 23.7 (6-CH2, 7-CH2);
24.6 (2-CH2); 26.5 (11-CH2, 30-CH2, overlapping); 30.0,
30.2, 30.3, 30.5, 30.6, 30.6, 30.7 (40-CH2 to 130-CH2, over-
lapping); 32.8 (140-CH2); 36.7 (20-CH2); 36.8 (1-CH2); 37.3
(12-CH2); 45.4 (3-CH2); 46.0 (10-CH2); 47.8, 47.8 (5-CH2,
8-CH2); 177.2 (NH–CO–CH2). MS, FABϩ found 441,
100% (Mϩϩ1), C26H56N4O requires Mϩˆ440. High-reso-
lution MS: m/z, FABϩ found 441.4542, (Mϩϩ1), C26H57N4O
requires Mϩϩ1ˆ441.4532.
N1-Octadecanoyl-(N4,N9,N12-tri-tert-butoxycarbonyl)-1,12-
diamino-4,9-diazadodecane 9. Protected tetraamine 6 (508
mg, 1.0 mmol) was reacted with stearic acid (345 mg,
1.2 mmol) according to general procedure A to afford the
title compound 9 as a colourless oil (659 mg, 85%), Rf 0.2
(EtOAc–hexane 50:50 v/v). IR (film) 1680, 1640 and 1520
(CO). 1H NMR, 400 MHz, CDCl3: 0.88 (t, 3H, Jˆ7 Hz, 180-
CH3); 1.23–1.35 (m, 28H, 40-CH2 to 170-CH2); 1.43–1.52
[m, 31H, 6-CH2, 7-CH2, O–C–(CH3)3×3]; 1.60–1.70 (m,
6H, 2-CH2, 11-CH2, 30-CH2); 2.18 (t, 2H, Jˆ7 Hz, 20-CH2);
3.05–3.34 (m, 12H, 1-CH2, 3-CH2, 5-CH2, 8-CH2, 10-CH2,
12-CH2); 5.24–5.40 (br s, 1H, CH2–NH–CO–O); 6.70–
6.85 (br s, 1H, CH2–CO–NH–CH2). 13C NMR, 100 MHz,
CDCl3: 14.1 (180-CH3); 22.6 (170-CH2); 25.4, 25.5, 25.6,
25.8, 25.9 (6-CH2, 7-CH2, 30-CH2); 27.6, 28.4, 28.7, 28.9
[2-CH2, 11-CH2, O–C–(CH3)3, overlapping]; 29.1, 29.3,
29.5, 29.6 (40-CH2 to 150-CH2, overlapping); 31.9 (160-
CH2); 33.9, 35.3 (12-CH2); 36.9, 37.3 (1-CH2, 20-CH2, over-
lapping); 43.2, 43.7, 44.1, 44.1 (3-CH2, 10-CH2); 46.2, 46.6
(5-CH2, 8-CH2, overlapping); 79.5, 79.7 (quat. C×3, over-
lapping); 156.0, 156.4 [N–CO–O–C–(CH3)3, overlapping];
173.3 (N–CO–CH2). MS, FABϩ found 769, 15% (Mϩϩ1),
C43H84N4O7 requires Mϩˆ768. High-resolution MS: m/z,
FABϩ found 769.6427, (Mϩϩ1), C43H85N4O7 requires
Mϩϩ1ˆ769.6418.
N1-Hexadecanoyl-(N4,N9,N12-tri-tert-butoxycarbonyl)-
1,12-diamino-4,9-diazadodecane 7. Protected tetraamine 6
(500 mg, 1.0 mmol) was reacted with palmitic acid
(340 mg, 1.2 mmol) according to general procedure A to
afford the title compound 7 as a colourless oil (663 mg,
96%), Rf 0.3 (EtOAc–hexane 60:40 v/v). IR (film) 1690,
1
1670 and 1530 (CO). H NMR, 400 MHz, CDCl3: 0.88 (t,
3H, Jˆ7 Hz, 160-CH3); 1.24–1.35 (m, 24H, 40-CH2 to 150-
CH2); 1.43–1.52 [m, 31H, 6-CH2, 7-CH2, O–C–(CH3)3×3];
1.57–1.60 (m, 6H, 2-CH2, 11-CH2, 30-CH2); 2.18 (t, 2H,
Jˆ7 Hz, 20-CH2); 3.20–3.40 (m, 12H, 1-CH2, 3-CH2, 5-
CH2, 8-CH2, 10-CH2, 12-CH2); 5.25–5.40 (br s, 1H, CH2–
N1-Octadecanoyl-1,12-diamino-4,9-diazadodecane 10. Pro-
tected polyamine amide 9 (400 mg, 0.52 mmol) was
NH–CO–O); 6.70–6.85 (br s, 1H, CH2 CO–NH–CH2. 13
C
NMR, 100 MHz, CDCl3: 14.1 (160-CH3); 22.7 (150-CH2);

A. J. Geall, I. S. Blagbrough / Tetrahedron 56 (2000) 2449–2460
2457
deprotected according to general procedure B to afford the
title compound 10 as a white solid (polytrifluoroacetate salt,
257 mg, 61%), tR 10.35 min by RP–HPLC (Supelcosil
ABZϩPlus, 5 mm, 15 cm×4.6 mm, MeOH–0.1% aq. TFA
60:40). IR (KBr) 1660 and 1560 (CO–N). 1H NMR,
400 MHz, D2O: 0.83–0.90 (m, 3H, 180-CH3); 1.17–1.45
(m, 28H, 40-CH2 to 170-CH2); 1.47–1.63 (m, 2H, 30-CH2);
1.72–1.83 (m, 4H, 6-CH2, 7-CH2); 1.83–1.94 (m, 2H, 11-
CH2); 2.09 (q, 2H, Jˆ8 Hz, 2-CH2); 2.19 (t, 2H, Jˆ7 Hz, 20-
CH2); 2.95–3.17 (m, 10H, 3-CH2, 5-CH2, 8-CH2, 10-CH2,
12-CH2); 3.17–3.30 (m, 2H, 1-CH2). 13C NMR, 100 MHz,
D2O: 17.4 (180-CH3); 26.2 (170-CH2); 26.3, 26.4 (6-CH2, 7-
CH2); 27.3 (2-CH2); 29.3 (11-CH2, 30-CH2, overlapping);
32.7, 33.0, 33.0, 33.2, 33.3, 33.5 (40-CH2 to 150-CH2, over-
lapping); 35.5 (160-CH2); 39.5 (20-CH2); 39.5 (1-CH2); 40.0
(12-CH2); 48.1 (3-CH2); 48.8 (10-CH2); 50.6 (5-CH2, 8-
CH2, overlapping); 179.9 (NH–CO–CH2). MS, FABϩ
found 469, 100% (Mϩϩ1), C28H60N4O requires Mϩˆ468.
High-resolution MS: m/z, FABϩ found 469.4845, (Mϩϩ1),
C28H61N4O requires Mϩϩ1ˆ469.4845.
12-CH2); 3.20–3.28 (m, 2H, 1-CH2); 5.27–5.38 (m, 2H, 90-
CH2, 100-CH2). 13C NMR, 100 MHz, D2O: 14.7 (180-CH3);
23.4 (170-CH2); 23.6, 23.7 (6-CH2, 7-CH2); 24.6 (2-CH2);
26.5 (11-CH2, 30-CH2, overlapping); 27.9, 28.0 (80-CH2,
110-CH2); 29.9, 30.0, 30.1, 30.1, 30.3, 30.5 (40-CH2 to 70-
CH2 and 120-CH2 to 150-CH2, overlapping); 32.7 (160-CH2);
36.7 (20-CH2); 36.8 (1-CH2); 37.3 (12-CH2); 45.4, 46.0 (3-
CH2, 10-CH2); 47.8 (5-CH2, 8-CH2, overlapping); 130.4,
130.5 (90-CH, 100-CH); 177.2 (NH–CO–CH2). MS, FABϩ
found 467, 100% (Mϩϩ1), C28H58N4O requires Mϩˆ466.
High-resolution MS: m/z, FABϩ found 467.4693, (Mϩϩ1),
C28H59N4O requires Mϩϩ1ˆ467.4689.
N1-trans-9-Octadecenoyl-(N 4, N 9, N 12-tri-tert-butoxy-
carbonyl)-1,12-diamino-4,9-diazadodecane 13. Protected
tetraamine 6 (471 mg, 0.9 mmol) was reacted with elaidic
acid (318 mg, 1.1 mmol) according to general procedure A
to afford the title compound 13 as a white waxy solid
(644 mg, 90%), Rf 0.1 (EtOAc–hexane 50:50 v/v). IR
(film) 1690, 1670 and 1540 (CO) and 970 (trans CyC).
1H NMR, 400 MHz, CDCl3: 0.88 (t, 3H, Jˆ7 Hz, 180-
CH3); 1.23–1.38 (m, 20H, 40-CH2 to 70-CH2, 120-CH2 to
170-CH2); 1.40–1.55 [m, 31H, 6-CH2, 7-CH2, O–C–
(CH3)3×3]; 1.58–1.62 (m, 6H, 2-CH2, 11-CH2, 30-CH2);
1.91–2.00 (m, 4H, 80-CH2, 110-CH2); 2.18 (t, 2H, Jˆ7 Hz,
20-CH2); 3.05–3.33 (m, 12H, 1-CH2, 3-CH2, 5-CH2, 8-CH2,
10-CH2, 12-CH2); 5.25–5.46 (m, 3H, 90-CH, 100-CH, CH2–
N1-cis-9-Octadecenoyl-(N4,N9,N12-tri-tert-butoxycarbon-
yl)-1,12-diamino-4,9-diazadodecane 11. Protected tetra-
amine 6 (471 mg, 0.9 mmol) was reacted with oleic acid
(318 mg, 1.1 mmol) according to general procedure A to
afford the title compound 11 as a white waxy solid
(550 mg, 77%), Rf 0.1 (EtOAc–hexane 50:50 v/v). IR
1
(film) 1690, 1670 and 1530 (CO). H NMR, 400 MHz,
NH–CO–O); 6.72–6.88 (br s, 1H, CH2 CO–NH–CH2). 13
C
CDCl3: 0.88 (t, 3H, Jˆ7 Hz, 180-CH3); 1.24–1.37 (m,
20H, 40-CH2 to 70-CH2, 120-CH2 to 170-CH2); 1.42–1.55
[m, 31H, 6-CH2, 7-CH2, O–C–(CH3)3×3]; 1.59–1.62 (m,
6H, 2-CH2, 11-CH2, 30-CH2); 1.88–2.05 (m, 4H, 80-CH2,
110-CH2); 2.18 (t, 2H, Jˆ7 Hz, 20-CH2); 3.00–3.35 (m,
12H, 1-CH2, 3-CH2, 5-CH2, 8-CH2, 10-CH2, 12-CH2);
5.30–5.42 (m, 2H, 90-CH, 100-CH); 6.64–6.82 (br s, 1H,
CH2 CO–NH–CH2). 13C NMR, 100 MHz, CDCl3: 14.1
(180-CH3); 22.6 (170-CH2); 24.9, 25.5, 25.6, 25.7, 25.9 (6-
CH2, 7-CH2, 30-CH2); 27.2 (80-CH2, 110-CH2, overlapping);
27.6, 28.4, 28.7 [2-CH2, 11-CH2, O–C–(CH3)3, overlap-
ping]; 28.9, 29.1, 29.3, 29.5, 29.7 (40-CH2 to 70-CH2, 120-
CH2 to 150-CH2, overlapping); 31.8 (160-CH2); 35.3 (12-
CH2, overlapping); 36.9 (20-CH2); 37.3, 37.7 (1-CH2);
43.2, 43.7, 44.0 (3-CH2, 10-CH2, overlapping); 46.6, 46.8
(5-CH2, 8-CH2, overlapping); 79.5, 79.7 (quat. C×3, over-
lapping); 129.7, 129.9 (90-CH, 100-CH); 156.0 [N–CO–O–
C–(CH3)3, overlapping]; 173.2 (N–CO–CH2). MS, FABϩ
found 767, 20% (Mϩϩ1), C43H82N4O7 requires Mϩˆ766.
High-resolution MS: m/z, FABϩ found 767.6277, (Mϩϩ1),
C43H83N4O7 requires Mϩϩ1ˆ767.6262.
N1-cis-9-Octadecenoyl-1,12-diamino-4,9-diazadodecane
12. Protected polyamine amide 11 (519 mg, 0.52 mmol)
was deprotected according to general procedure B to afford
the title compound 12 as a white solid (polytrifluoroacetate
salt, 257 mg, 61%), tR 7.40 min by RP–HPLC (Supelcosil
ABZϩPlus, 5 mm, 15 cm×4.6 mm, MeOH–0.1% aq. TFA
65:35). IR (KBr) 1660 (CO–N). 1H NMR, 400 MHz, D2O:
0.86 (t, 3H, Jˆ7 Hz, 180-CH3); 1.20–1.37 (m, 20H, 40-CH2
to 70-CH2 and 120-CH2 to 170-CH2); 1.51–1.59 (m, 2H, 30-
CH2); 1.63–1.80 (m, 4H, 6-CH2, 7-CH2); 1.88 (q, 2H,
Jˆ7 Hz, 11-CH2); 1.95–2.04 (m, 4H, 80-CH2, 110-CH2);
2.09 (quin., 2H, Jˆ7 Hz, 2-CH2); 2.19 (t, 2H, Jˆ7 Hz, 20-
CH2); 2.95–3.10 (m, 10H, 3-CH2, 5-CH2, 8-CH2, 10-CH2,
NMR, 100 MHz, CDCl3: 14.1 (180-CH3); 22.6 (170-CH2);
25.4, 25.5, 25.6, 25.7, 25.9 (6-CH2, 7-CH2, 30-CH2); 27.6,
28.4, 28.7, 28.8 [2-CH2, 11-CH2, O–C–(CH3)3, over-
lapping]; 29.0, 29.1, 29.2, 29.3, 29.4, 29.5, 29.6 (40-CH2
to 70-CH2, 120-CH2 to 150-CH2, overlapping); 32.5 (160-
CH2); 33.9, 34.0 (80-CH2, 110-CH2); 35.3, 35.9 (12-CH2);
36.9 (20-CH2); 37.3, 37.6 (1-CH2); 43.2, 43.7, 44.1 (3-CH2,
10-CH2, overlapping); 46.6 (5-CH2, 8-CH2, overlapping);
79.5, 79.7 (quat. C×3, overlapping); 130.2, 130.3 (90-CH,
100-CH); 156.1, 156.4 [N–CO–O–C–(CH3)3, overlapping];
173.3 (N–CO–CH2). MS, FABϩ found 767, 15% (Mϩϩ1),
C43H82N4O7 requires Mϩˆ766. High-resolution MS: m/z,
FABϩ found 767.6255, (Mϩϩ1), C43H83N4O7 requires
Mϩϩ1ˆ767.6262.
N1-trans-9-Octadecenoyl-1,12-diamino-4,9-diazadodecane
14. Protected polyamine amide 13 (400 mg, 0.52 mmol)
was deprotected according to general procedure B to afford
the title compound 14 as a white solid (polytrifluoroacetate
salt, 257 mg, 61%), tR 10.65 min by RP–HPLC (Supelcosil
ABZϩPlus, 5 mm, 15 cm×4.6 mm, MeOH–0.1% aq. TFA
60:40). IR (KBr) 1670 (CO–N). 1H NMR, 400 MHz, D2O:
0.83–0.90 (m, 3H, 180-CH3); 1.17–1.45 (m, 20H, 40-CH2 to
70-CH2, 120-CH2 to 170-CH2); 1.47–1.63 (m, 2H, 30-CH2);
1.72–1.83 (m, 4H, 6-CH2, 7-CH2); 1.85–2.03 (m, 6H, 11-
CH2, 80-CH2, 110-CH2); 2.03–2.15 (m, 2H, 2-CH2); 2.19 (t,
2H, Jˆ7 Hz, 20-CH2); 2.95–3.17 (m, 10H, 3-CH2, 5-CH2, 8-
CH2, 10-CH2, 12-CH2); 3.17–3.30 (m, 2H, 1-CH2); 5.35–
5.46 (m, 2H, 90-CH2, 100-CH2). 13C NMR, 100 MHz, D2O:
14.7 (180-CH3); 23.4 (170-CH2); 23.6, 23.7 (6-CH2, 7-CH2);
24.6 (2-CH2); 26.5 (11-CH2, 30-CH2, overlapping); 30.0,
30.1, 30.2, 30.4, 30.5 (40-CH2 to 70-CH2, 120-CH2 to 150-
CH2, overlapping); 32.7 (160-CH2); 33.4 (80-CH2, 110-CH2);
36.7 (20-CH2); 36.8 (1-CH2); 37.3 (12-CH2); 45.4, 46.0 (3-
CH2, 10-CH2); 47.8 (5-CH2, 8-CH2, overlapping); 130.9,

2458
A. J. Geall, I. S. Blagbrough / Tetrahedron 56 (2000) 2449–2460
131.0 (90-CH, 100-CH); 177.1 (NH–CO–CH2). MS, FABϩ
found 467, 100% (Mϩϩ1), C28H58N4O requires Mϩˆ466.
High-resolution MS: m/z, FABϩ found 467.4679, (Mϩϩ1),
C28H59N4O requires Mϩϩ1ˆ467.4689.
in vacuo and the residue purified over silica gel (CH2Cl2–
MeOH–conc. aq. NH3 100:10:1 v/v/v) to afford homo-
logated polyamine 17; Rf 0.25 (CH2Cl2–MeOH–conc. aq.
NH3 100:10:1 v/v/v), as a yellow oil. Amine 17 was then
dissolved in DMF (10 ml) at 25ЊC, under nitrogen, and
di-tert-butyl dicarbonate (445 mg, 2.0 mmol) was added
dropwise over 3 min to the stirring solution to afford poly-
protected polyamine 18. After 1 h, conc. aq. NH3 (1 ml) was
added to quench the excess of reagent, the solution was
stirred for a further 30 min and then concentrated in vacuo
(40ЊC). The residue was then dissolved in MeOH (10 ml),
Pearlman’s catalyst [500 mg, Pd(OH)2 on carbon 20%] was
added and the flask and contents evacuated and flushed
twice with hydrogen. The solution was then stirred for 4 h
at 25ЊC under an atmosphere of hydrogen. The catalyst was
filtered through a bed of celite and the filtrate concentrated
in vacuo and the residue purified over silica gel (CH2Cl2–
MeOH–conc. aq. NH3 100:10:1 to 75:10:1 v/v/v) to afford
the title compound 19 as a colourless oil (518 mg, 45%), Rf
0.15 (CH2Cl2–MeOH–conc. aq. NH3 100:10:1 v/v/v). IR
(film) 1670 (O–CO–N). 1H NMR, 400 MHz, CDCl3:
1.41–1.60 [m, 40H, 6-CH2, 7-CH2, C–(CH3)3×4]; 1.60–
1.80 (m, 6H, 2-CH2, 11-CH2, 15-CH2); 2.18–2.24 (br s, 2H,
NH2); 2.71 (t, 2H, Jˆ7 Hz, 16-CH2); 3.05–3.35 (m, 14H, 1-
CH2, 3-CH2, 5-CH2, 8-CH2, 10-CH2, 12-CH2, 14-CH2);
5.30–5.50 (br s, 1H, CO–NH–CH2). 13C NMR, 100 MHz,
CDCl3: 25.3, 25.4, 25.5, 25.8, 25.8, 25.9 (6-CH2, 7-CH2, 11-
CH2); 28.3, 28.7, 28.8 [2-CH2, O–C–(CH3)3×4, overlap-
ping]; 31.1, 32.3 (15-CH2); 37.2, 37.6 (1-CH2); 38.7, 39.3
(16-CH2); 43.7, 43.9, 44.0, 44.1, 44.1, 44.2, 44.7 (3-CH2,
10-CH2, 12-CH2, 14-CH2, overlapping); 46.4, 46.7 (5-CH2,
8-CH2, overlapping); 155.3, 160.0 (N–CO–O×4). MS,
FABϩ found 660, 95% (Mϩϩ1), C33H65N5O8 requires
Mϩˆ659. High-resolution MS: m/z, FABϩ found
660.4906, (Mϩϩ1), C33H66N5O8 requires Mϩϩ1ˆ
660.4911.
3-Benzyloxycarbonylaminopropan-1-ol 15. To a stirring
solution of 3-aminopropan-1-ol (30.0 g, 40 mmol) in aq.
NaOH (1 M, 44 ml) at 0ЊC, benzyl chloroformate (7.51 g,
44 mmol) was added dropwise over 3 min. The solution was
then allowed to warm to 25ЊC, stirred for 1 h and then
CH2Cl2 (30 ml) was added. After 3 h, the solution was
concentrated in vacuo and the residue purified over silica
gel (CH2Cl2–MeOH–conc. aq. NH3 300:10:1 to 200:10:1
v/v/v) to afford the title compound 15 as a white solid
(8.26 g, 99%), Rf 0.3 (CH2Cl2–MeOH–conc. aq. NH3
200:10:1 v/v/v), mp: 50–51ЊC. IR (KBr) 3320 and 1070
1
(OH) and 1690 (O–CO–N). H NMR, 270 MHz, CDCl3:
1.67 (quin, 2H, Jˆ6 Hz, 2-CH2); 3.31 (t, 2H, Jˆ6 Hz, 3-
CH2); 3.64 (t, 2H, Jˆ6 Hz, 1-CH2); 5.09 (s, 2H, CO–O–
CH2–Ph); 7.26–7.34 (m, 5H, Ph). 13C NMR, 67.5 MHz,
CDCl3: 32.4 (2-CH2); 37.9 (1-CH2); 59.6 (3-CH2); 66.7
(O–CH2–Ph); 128.0, 128.1, 128.4, (Ph); 136.4 (quat. C
Ph); 157.2 (N–CO–O). MS, FABϩ found 210, 75%
(Mϩϩ1), C11H15NO3 requires Mϩˆ209. Anal. calcd for
C11H15NO3: C 63.14; H 7.23; N 6.69. Found: C 63.10; H
7.25; N 6.64.
3-Benzyloxycarbonylaminopropanal 16. Oxalyl chloride
(1.91 g, 15.0 mmol) was dissolved in freshly distilled anhy-
drous CH2Cl2 (from CaH2) and stirred at Ϫ78ЊC under nitro-
gen. Then anhydrous DMSO (1.94 ml, 27.0 mmol) was
added dropwise over 3 min and the mixture stirred for a
further 10 min at Ϫ78ЊC. Alcohol 15 (2 g, 13.7 mmol) in
anhydrous CH2Cl2 (10 ml) was then added dropwise over
3 min. The resultant cloudy suspension was then warmed
(ϳϪ40ЊC) until the solution cleared and then cooled again
to Ϫ78ЊC for 10 min. Triethylamine (9.5 ml, 68 mmol) was
added, the solution was warmed to 25ЊC and water (50 ml)
was then added. The layers were separated and the aqueous
layer extracted with CH2Cl2 ꢀ2 × 35 ml†: The combined
organic extracts were dried (MgSO4) and the solution was
concentrated in vacuo and the residue purified over silica gel
(EtOAc) to afford the title compound 16 as a white solid
(1.63 g, 82%), Rf 0.5 (EtOAc), mp: 57–58ЊC. IR (KBr) 2840
and 1710 (CHO) and 1680 (O–CO–N). 1H NMR, 270 MHz,
CDCl3: 2.71 (t, 2H, Jˆ6 Hz, 2-CH2); 3.43–3.56 (m, 2H, 3-
CH2); 5.07 (s, 2H, CO–O–CH2–Ph); 5.22–5.34 (br s, 1H,
CH2–NH–CO–O); 7.30–7.42 (m, 5H, Ph); 9.70–9.84 (m,
1H, 1-CHO). 13C NMR, 67.5 MHz, CDCl3: 34.4 (2-CH2);
44.0 (3-CH2); 66.7 (O–CH2–Ph); 128.0, 128.1, 128.4, (Ph);
136.3 (quat. C); 156.3 (N–CO–O); 201.1 (1-CHO). MS,
FABϩ found 208, 44% (Mϩϩ1), C11H13NO3 requires
Mϩˆ207. Anal. calcd for C11H13NO3: C 63.76; H 6.32; N
6.76. Found: C 63.70; H 6.41; N 6.73.
N1-Hexadecanoyl-(N4,N9,N13,N16-tetra-tert-butoxycarbon-
yl)-1,16-diamino-4,9,13-triazahexadecane 20. Protected
pentaamine 19 (478 mg, 0.73 mmol) was reacted with
palmitic acid (223 mg, 0.87 mmol) according to general
procedure A to afford the title compound 20 as a colourless
oil (570 mg, 88%), Rf 0.2 (EtOAc–hexane 60:40 v/v). IR
1
(film) 1690, 1670 and 1530 (CO). H NMR, 400 MHz,
CDCl3: 0.88 (t, 3H, Jˆ7 Hz, 160-CH3); 1.20–1.35 (m,
24H, 40-CH2 to 150-CH2); 1.35–1.58 [m, 42H, 2-CH2, 10-
CH2, 11-CH2, O–C–(CH3)3×4]; 1.58–1.83 (m, 6H, 6-CH2,
15-CH2, 30-CH2); 2.18 (t, 2H, Jˆ7 Hz, 20-CH2); 3.05–3.40
(m, 16H, 1-CH2, 3-CH2, 5-CH2, 7-CH2, 9-CH2, 12-CH2, 14-
CH2, 16-CH2); 5.24–5.40 (br s, 1H, CH2–NH–CO–O);
6.70–6.85 (br s, 1H, CH2–CO–NH–CH2). 13C NMR,
100 MHz, CDCl3: 14.1 (160-CH3); 22.6 (150-CH2); 25.4,
25.5, 25.7, 25.8, 25.9 (6-CH2, 10-CH2, 11-CH2, 30-CH2,
overlapping); 27.6, 27.7, 27.8, 28.4 [2-CH2, 15-CH2, O–
C–(CH3)3, overlapping]; 29.3, 29.5, 29.6, 29.6 (40-CH2 to
130-CH2, overlapping); 31.9 (140-CH2); 33.9, 35.3 (16-
CH2); 36.9, 37.3 (1-CH2, 20-CH2, overlapping); 43.2, 43.7,
44.1, 44.7, 46.5, 46.8 (3-CH2, 5-CH2, 7-CH2, 9-CH2, 12-
CH2, 14-CH2, overlapping); 79.3, 79.4, 79.7 (quat. C×4,
overlapping); 155.4, 156.1, 156.3 [N–CO–O–C–
(CH3)3×4, overlapping]; 173.3 (N–CO–CH2). MS,
FABϩ found 898, 20% (Mϩϩ1), C49H95N5O9 requires
Mϩˆ897. High-resolution MS: m/z, FABϩ found
(N1,N4,N9,N13-Tetra-tert-butoxycarbonyl)-1,16-diamino-
4,9,13-triazahexadecane 19. Protected tetraamine 6 (1.06
˚
g, 2.10 mmol), over 4 A molecular sieves (ϳ4 g), was
dissolved in freshly distilled anhydrous MeOH (20 ml)
under nitrogen. Aldehyde 16 (366 mg, 1.76 mmol),
NaCNBH3 (166 mg, 2.64 mmol) and cat. glacial HOAc
were then added and the reaction mixture stirred at 25ЊC,
under nitrogen, for 24 h. The solution was then concentrated

A. J. Geall, I. S. Blagbrough / Tetrahedron 56 (2000) 2449–2460
2459
898.7207, (Mϩϩ1), C49H96N5O9 requires Mϩϩ1ˆ
11. Blagbrough, I. S.; Carrington, S.; Geall, A. J. Pharm. Sci.
1997, 3, 223.
898.7208.
12. Feuerstein, B. G.; Pattabiraman, N.; Marton, L. J. Nucleic
Acids Res. 1990, 18, 1271.
13. Behr, J.-P.; Demeneix, B.; Loeffler, J.-P.; Perez-Mutul, J.
Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 6982.
14. Rowatt, E.; Williams, R. J. P. J. Inorg. Biochem. 1992, 46, 87.
15. Stewart, K. D.; Gray, T. A. J. Phys. Org. Chem. 1992, 5, 461.
16. Remy, J.-S.; Sirlin, C.; Vierling, P.; Behr, J.-P. Bioconjugate
Chem. 1994, 5, 647.
17. Wilson, R. W.; Bloomfield, V. A. Biochemistry 1979, 18,
2192.
18. Bloomfield, V. A. Curr. Opin. Struct. Biol. 1996, 6, 334.
19. Aikens, D.; Bunce, S.; Onasch, F.; Parker III, R.; Hurwitz, C.;
Clemans, S. Biophys. Chem. 1983, 17, 67.
20. Behr, J.-P. Acc. Chem. Res. 1993, 26, 274.
21. Olins, D. E.; Olins, A. L. J. Mol. Biol. 1971, 57, 437.
22. McArthur, M.; Thomas, J. O. EMBO J. 1996, 15, 1705.
N1-Hexadecanoyl-1,16-diamino-4,8,13-triazahexadecane
21. Protected polyamine amide 20 was deprotected accord-
ing to general procedure B to afford the title compound 21 as
the polytrifluoroacetate salt (104 mg, 49%), tR 5.0 min by
RP–HPLC (Supelcosil ABZϩPlus, 5 mm, 15 cm×4.6 mm,
1
MeOH–0.1% aq. TFA 65:35). IR (KBr) 1660 (CO–N). H
NMR, 400 MHz, [2H]6 DMSO: 0.86 (t, 3H, Jˆ7 Hz, 160-
CH3); 1.15–1.35 (m, 24H, 40-CH2 to 150-CH2); 1.42–1.53
(m, 2H, 30-CH2); 1.60–1.69 (m, 4H, 10-CH2, 11-CH2); 1.72
(quin, 2H, Jˆ7 Hz, 2-CH2); 1.86–2.00 (m, 4H, 6-CH2 15-
CH2); 2.06 (t, 2H, Jˆ7 Hz, 20-CH2); 2.83–2.94 (m, 8H, 9-
CH2, 12-CH2, 14-CH2, 16-CH2); 2.94–3.05 (m, 6H, 3-CH2,
5-CH2, 7-CH2); 3.05–3.12 (m, 2H, 1-CH2). 13C NMR,
100 MHz, [2H]6 DMSO: 14.0 (160-CH3); 22.1 (150-CH2);
22.5 (6-CH2); 22.7 (10-CH2, 11-CH2, overlapping); 23.8
(15-CH2); 25.3 (30-CH2); 26.1 (2-CH2); 28.7, 28.8, 29.0,
29.1 (40-CH2 to 130-CH2, overlapping); 31.3 (140-CH2);
35.4 (20-CH2); 35.6 (1-CH2); 36.1 (16-CH2); 43.8, 43.9,
44.0 (3-CH2, 5-CH2, 7-CH2); 44.7 (14-CH2); 46.1 (9-CH2,
12-CH2, overlapping); 172.7 (NH–CO–CH2). MS, FABϩ
found 498, 100% (Mϩϩ1), C29H63N5O1 requires
Mϩˆ497. High-resolution MS: m/z, FABϩ found
498.5114, (Mϩϩ1), C28H59N4O1 requires Mϩϩ1ˆ
498.5111.
¨
23. Luger, K.; Mader, A. W.; Richmond, R. K.; Sargent, D. F.;
Richmond, T. J. Nature 1997, 389, 251.
24. Lee, E. R.; Marshall, J.; Siegel, C. S.; Jiang, C.; Yew, N. S.;
Nichols, M. R.; Nietupski, J. B.; Ziegler, R. J.; Lane, M. B.; Wang,
K. X.; Wan, N. C.; Scheule, R. K.; Harris, D. J.; Smith, A. E.;
Cheng, S. H. Hum. Gene Therapy 1996, 7, 1701.
25. Moradpour, D.; Schauer, J. I.; Zurawski, V. R., Jr.; Wands,
J. R.; Boutin, R. H. Biochem. Biophys. Res. Commun. 1996, 221,
82; Guy-Caffey, J. K.; Bodepudi, V.; Bishop, J. S.; Jayaraman, K.;
Chaudhary, N. J. Biol. Chem. 1995, 270, 31391.
26. Walker, S.; Sofia, M. J.; Kakarla, R.; Kogan, N. A.; Wierichs,
L.; Longley, C. B.; Bruker, K.; Axelrod, H. R.; Midha, S. I.; Babu,
S.; Kahne, D. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 1585.
27. Behr, J.-P.; Demeneix, B.; Loeffler, J.-P.; Perez-Mutul, J.
Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 6982.
28. Remy, J.-S.; Sirlin, C.; Vierling, P.; Behr, J.-P. Bioconjugate
Chem. 1994, 5, 647.
29. Byk, G.; Dubertret, C.; Escriou, V.; Frederic, M.; Jaslin, G.;
Rangara, R.; Pitard, B.; Crouzet, J.; Wils, P.; Schwartz, B.;
Scherman, D. J. Med. Chem. 1998, 41, 224.
Acknowledgements
We thank the EPSRC and Celltech Chiroscience (CASE
award to A. J. G.) for financial support. We also thank Dr
Michael A. W. Eaton (Celltech Chiroscience), Dr Eduardo
Moya (University of Bath) and Dr Ian S. Haworth (Uni-
versity of Southern California) for useful discussions and
for their interest in these studies. I. S. B. and I. S. H. are joint
holders of a NATO grant (CRG 970290).
30. Blagbrough, I. S.; Geall, A. J. Tetrahedron Lett. 1998, 39, 439.
31. Geall, A. J.; Blagbrough, I. S. Tetrahedron Lett. 1998, 39, 443.
32. Geall, A. J.; Taylor, R. J.; Earll, M. E.; Eaton, M. A. W.;
Blagbrough, I. S. Chem. Commun. 1998, 1403.
References
33. Geall, A. J.; Al-Hadithi, D.; Blagbrough, I. S. Chem. Commun.
1998, 2035.
1. Usherwood, P. N. R.; Blagbrough, I. S. Pharmacol. Ther. 1991,
52, 245.
34. Ganem, B. Acc. Chem. Res. 1982, 15, 290.
35. Huang, D.; Jiang, H.; Nakanishi, K.; Usherwood, P. N. R.
Tetrahedron 1997, 53, 12391.
36. Bergeron, R. J. Acc. Chem. Res. 1986, 19, 105.
37. McCormick, K. D.; Meinwald, J. J. Chem. Ecol. 1993, 19,
2. Blagbrough, I. S.; Bruce, M.; Bycroft, B. W.; Mather, A. J.;
Usherwood, P. N. R. Pestic. Sci. 1990, 30, 397; Blagbrough, I. S.;
Brackley, P. T. H.; Bruce, M.; Bycroft, B. W.; Mather, A. J.;
Millington, S.; Sudan, H. L.; Usherwood, P. N. R. Toxicon 1992,
30, 303.
¨
2411; Schafer, A.; Benz, H.; Fiedler, W.; Guggisberg, A.;
3. Blagbrough, I. S.; Usherwood, P. N. R. Proc. Roy. Soc. Edin.
1992, 99B, 67.
Bienz, S.; Hesse, M. Alkaloids 1994, 45, 1.
38. Atwell, G. J.; Denny, W. A. Synthesis 1984, 1032.
39. Saari, W. S.; Schwering, J. E.; Lyle, P. A.; Smith, S. J.;
Engelhardt, E. L. J. Med. Chem. 1990, 33, 97.
40. Krapcho, A. P.; Kuell, C. S. Synth. Commun. 1994, 20, 2559.
41. O’Sullivan, M. C.; Dalrymple, D. M. Tetrahedron Lett. 1995,
36, 3451; Xu, D.; Prasad, K.; Repic, O.; Blacklock, T. J.
Tetrahedron Lett. 1995, 36, 7357.
42. Imazawa, M.; Eckstein, F. J. Org. Chem. 1979, 44, 2039.
43. Bergeron, R. J.; McManis, J. S. J. Org. Chem. 1988, 53, 3108.
44. Felgner, P. L.; Barenholz, Y.; Behr, J. P.; Cheng, S. H.; Cullis,
P.; Huang, L.; Jessee, J. A.; Seymour, L.; Szoka, F.; Thierry, A. R.;
Wagner, E.; Wu, G. Hum. Gene Therapy 1997, 8, 511.
4. Adlam, G.; Blagbrough, I. S.; Taylor, S.; Latham, H. C.;
Haworth, I. S.; Rodger, A. Bioorg. Med. Chem. Lett. 1994, 4, 2435.
5. Rodger, A.; Blagbrough, I. S.; Adlam, G.; Carpenter, M. L.
Biopolymers 1994, 34, 1583.
6. Rodger, A.; Taylor, S.; Adlam, G.; Blagbrough, I. S.; Haworth,
I. S. Bioorg. Med. Chem. 1995, 3, 861.
7. Blagbrough, I. S.; Moya, E. Tetrahedron Lett. 1995, 36, 9393.
8. Ashton, M. R.; Moya, E.; Blagbrough, I. S. Tetrahedron Lett.
1995, 36, 9397.
9. Moya, E.; Blagbrough, I. S. Tetrahedron Lett. 1995, 36, 9401.
10. Blagbrough, I. S.; Moya, E.; Walford, S. P. Tetrahedron Lett.
1996, 37, 551.

2460
A. J. Geall, I. S. Blagbrough / Tetrahedron 56 (2000) 2449–2460
45. LePecq, J.-B.; Paoletti, C. J. Mol. Biol. 1967, 27, 87.
46. Waring, M. J. J. Mol. Biol. 1970, 54, 247.
47. Morgan, A. R.; Lee, J. S.; Pulleyblank, D. F.; Murray, N. L.;
Evans, D. H. Nucleic Acids Res. 1979, 7, 547.
48. Stewart, K. D.; Gray, T. A. J. Phys. Org. Chem. 1992, 5, 461.
49. Gershon, H.; Ghirlando, R.; Guttman, S. B.; Minsky, A.
Biochemistry 1993, 32, 7143.
55. Jaroszewski, J. W.; Matzen, L.; Frølund, B.; Krogsgaard-
Larsen, P. J. Med. Chem. 1996, 39, 515.
56. Kimberly, M. M.; Goldstein, J. H. Anal. Chem. 1981, 53, 789.
57. Aikens, D. A.; Bunce, S. C.; Onasch, O. F.; Schwartz, H. M.;
Hurwitz, C. J. Chem. Soc., Chem. Commun. 1983, 43.
58. Sarneski, J. E.; Surprenant, H. L.; Molen, F. K.; Reilly, C. N.
Anal. Chem. 1975, 47, 2116.
50. Hsieh, H.-P.; Muller, J. G.; Burrows, C. J. J. Am. Chem. Soc.
1994, 116, 12077.
51. Hsieh, H.-P.; Muller, J. G.; Burrows, C. J. Bioorg. Med. Chem.
1995, 3, 823.
52. Cain, B. F.; Baguley, B. C.; Denny, W. A. J. Med. Chem. 1978,
21, 658.
53. Frassineti, C.; Ghelli, S.; Gans, P.; Sabatini, A.; Moruzzi, M. S.;
Vacca, A. Anal. Biochem. 1995, 231, 374.
59. Rabenstein, D. L.; Sayer, T. L. J. Magn. Res. 1976, 24, 27.
60. Batchelor, J. G.; Feeney, J.; Roberts, G. C. K. J. Magn. Res.
1975, 20, 19.
61. Manning, G. S. Quart. Rev. Biophys. 1978, 11, 179.
62. Bloomfield, V. A. Biopolymers 1991, 31, 1471.
63. Bloomfield, V. A. Biopolymers 1997, 44, 269.
64. Basu, H. S.; Schwietert, H. C. A.; Feuerstein, B. G.; Marton, L. J.
Biochem. J. 1990, 269, 329.
54. Tables of Spectral Data for Structure Determination of
Organic Compounds, Springer: Berlin, 1989, pp C5–C47.