Lipopolysaccharide sequestrants: Structural correlates of activity and toxicity in novel acylhomospermines

Article abstract of DOI:10.1021/jm049449j

Lipopolysaccharides (LPS), otherwise termed "endotoxins", are outer membrane constituents of Gram-negative bacteria. Lipopolysaccharides play a key role in the pathogenesis of "septic shock", a major cause of mortality in the critically ill patient. Therapeutic options aimed at limiting downstream systemic inflammatory processes by targeting lipopolysaccharide do not exist at the present time. We have defined the pharmacophore necessary for small molecules to specifically bind and neutralize LPS and, using animal models of sepsis, have shown that the sequestration of circulatory LPS by small molecules is a therapeutically viable strategy. In this paper, the interactions of a series of acylated homologated spermine compounds with LPS have been characterized. The optimal acyl chain length for effective sequestration of LPS was identified to be C₁₆ for the monoacyl compounds. The most promising of these compounds, 4e, binds LPS with an ED₅₀ of 1.37 μM. Nitric oxide production in murine J774A.1 cells, as well as TNF-α in human blood, is inhibited in a dose-dependent manner by 4e at concentrations orders of magnitude lower than toxic doses. Administration of 4e to D-galactosamine-sensitized mice challenged with supralethal doses of LPS provided significant protection against lethality. Potent antiendotoxic activity, low toxicity, and ease of synthesis render this class of compounds candidate endotoxin-sequestering agents of potential significant therapeutic value.

Full text of DOI:10.1021/jm049449j

J. Med. Chem. 2005, 48, 2589-2599
2589
Lipopolysaccharide Sequestrants: Structural Correlates of Activity and
Toxicity in Novel Acylhomospermines
Kelly A. Miller,† E. V. K. Suresh Kumar,‡ Stewart J. Wood,† Jens R. Cromer,† Apurba Datta,‡ and
Sunil A. David*,†
Department of Medicinal Chemistry, Life Sciences Research Laboratories, 1501 Wakarusa Drive, University of Kansas,
Lawrence, Kansas 66049, and Malott Hall, 1251 Wescoe Hall Drive, University of Kansas, Lawrence, Kansas 66045
Received July 12, 2004
Lipopolysaccharides (LPS), otherwise termed “endotoxins”, are outer membrane constituents
of Gram-negative bacteria. Lipopolysaccharides play a key role in the pathogenesis of “septic
shock”, a major cause of mortality in the critically ill patient. Therapeutic options aimed at
limiting downstream systemic inflammatory processes by targeting lipopolysaccharide do not
exist at the present time. We have defined the pharmacophore necessary for small molecules
to specifically bind and neutralize LPS and, using animal models of sepsis, have shown that
the sequestration of circulatory LPS by small molecules is a therapeutically viable strategy.
In this paper, the interactions of a series of acylated homologated spermine compounds with
LPS have been characterized. The optimal acyl chain length for effective sequestration of LPS
was identified to be C16 for the monoacyl compounds. The most promising of these compounds,
4e, binds LPS with an ED50 of 1.37 µM. Nitric oxide production in murine J774A.1 cells, as
well as TNF-R in human blood, is inhibited in a dose-dependent manner by 4e at concentrations
orders of magnitude lower than toxic doses. Administration of 4e to D-galactosamine-sensitized
mice challenged with supralethal doses of LPS provided significant protection against lethality.
Potent antiendotoxic activity, low toxicity, and ease of synthesis render this class of compounds
candidate endotoxin-sequestering agents of potential significant therapeutic value.
Introduction
two protonatable positive charges separated by a dis-
tance of ∼14 Å, enabling ionic H-bonds between the
cationic groups and the lipid A phosphates; in addition,
appropriately positioned pendant hydrophobic function-
alities are required to further stabilize the resultant
complexes via hydrophobic interactions with the poly-
acyl domain of lipid A (for a recent review, see ref 14).
These structural requisites are exemplified in the lipo-
polyamines, which are of particular interest because
they are active in vitro and afford protection in animal
models of Gram-negative sepsis, are synthetically easily
accessible, and, importantly, are nontoxic, on account
of their degradation to physiological substituents (sper-
mine and fatty acid).15,16 A careful evaluation of struc-
ture-activity relationships in these compounds would
be crucial in further iterations of designing potent
analogues and in their preclinical development as
potential LPS-sequestering agents. The detailed studies
on acylhomospermines that we report in this paper
address two questions: (i) what is the optimal hydro-
phobic chain length for effective antiendotoxic activity,
and (ii) are symmetrical bis-acyl spermines more effec-
tive than monoacyl compounds? Our results indicate
that a carbon number of 14-16 is optimal in monoacyl
spermines which are, in general, as potent as their bis-
homologues and, in addition, show less surface activity
(lower nonspecific cytotoxicity) and possess physical
properties that are better suited for parenteral admin-
istration.
Endotoxins, or lipopolysaccharides (LPS), the pre-
dominant structural component of the outer membrane
of Gram-negative bacteria,1 play a pivotal role in “septic
shock”, a syndrome of systemic toxicity that occurs when
the body’s defense mechanisms are compromised or
overwhelmed or as a consequence of antibiotic chemo-
therapy of serious systemic infections (Gram-negative
sepsis).2 Gram-negative sepsis is the number one cause
of deaths in the intensive care unit,3 accounting for more
than 200 000 fatalities in the US annually.4
The presence of LPS in the systemic circulation causes
a widespread activation of the innate immune re-
sponse5,6 leading to the uncontrolled production of
numerous inflammatory mediators, including tumor
necrosis factor-R (TNF-R), interleukin-1 â (IL-1â), and
interleukin-6 (IL-6), primarily by cells of the monocyte/
macrophage lineage,7,8 as well as others, such as nitric
oxide produced by the endothelial cell,9,10 which, in
concert, act to cause a frequently fatal systemic inflam-
matory response,11 termed septic shock. The toxic
moiety of LPS is its structurally conserved glycolipid
component called lipid A,12 which is composed of a
hydrophilic, bis-phosphorylated diglucosamine backbone
and a hydrophobic domain of six (Esherichia coli) or
seven (Salmonella) acyl chains12 (Figure 1). We have
determined that the pharmacophore necessary for the
neutralization of lipid A13 by small molecules requires
Results and Discussion
* Corresponding author. Tel: 785-330-4316. Fax: 785-330-4332.
E-mail: sdavid@ku.edu.
Synthesis of Acylhomospermines. The synthesis
of the desired monoacylhomospermine derivatives under
† Life Sciences Research Laboratories.
‡ Malott Hall.
10.1021/jm049449j CCC: $30.25 © 2005 American Chemical Society
Published on Web 02/19/2005

2590 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7
Miller et al.
Scheme 1a
less soluble, with 8d-f being practically insoluble in
physiological buffers. Stock solutions (5 mM) of all
compounds were prepared in neat DMSO and diluted
in buffer in the assays that are described below.
Affinity of Binding to Lipopolysaccharide. We
had previously investigated a number of classes of
hydrophobic polyamines,14-16,20-22 including DOSPER,
a lipospermine analogue23,24 with two centrally placed
pendant oleoyl chains,16 as well as a C15 analogue of 6.15
From these initial, exploratory studies we had noted
that homologated monoacylhomospermine binds lipid A
with higher affinity and antagonizes the toxicity of LPS
with greater potency than DOSPER, and we hypoth-
esized that this could be attributable to steric hindrance
in the case of DOSPER and that terminally placed
hydrophobic group(s) may obviate steric problems. This
class of compounds is synthetically easily accessible and,
importantly, expected to be nontoxic on account of being
comprised of metabolically inert polyamine and fatty
acid fragments and are, therefore, attractive drug
candidates. It was thus of interest to systematically
examine the effect of the number (mono versus bis) and
length of the acyl chains on binding affinity and biologi-
cal activity. We first examined quantitatively the bind-
ing affinities of the homologated acylhomospermines to
LPS using a rapid and robust high-throughput fluores-
cence displacement assay described recently.25 The
assay, in its entirety, consists of three automated
steps: (i) in-plate serial dilution of compounds; (ii)
addition of a mixture of LPS and BODIPY-TR-cadav-
erine; and (iii) recording end-point fluorescence intensity
at a fixed wavelength (620 nm) in a plate reader. The
addition of test compounds results in dequenching of
LPS-bound BC fluorescence, manifesting in emission
intensity enhancements. The apparent binding affinities
of the compounds, computed from displacement curves,
are shown in Table 1. We confirmed the validity of
binding affinity (ED50) obtained from fluorescence dis-
placement data by performing isothermal titration
calorimetry (ITC) experiments on several of the com-
pounds that were freely soluble in aqueous buffers (data
not shown). For instance, the interaction of 4e with LPS
is exothermic and enthalpically driven (∆H ) -17.02
kcal mol-1, ∆S ) -27.8 cal mol-1 K-1), yielding a
dissociation constant (KD) of 1.21 µM, which is in
excellent agreement with the ED50 value of 1.37 µM
obtained from fluorescence experiments (Table 1). A
striking difference in binding affinities was observed
between the mono- and bis-acyl compounds (Table 1).
4a, with an acetyl substituent, binds very weakly,
whereas all other (C8-C18) compounds bind LPS with
virtually identical affinities. The observation that 4a
would be a weak binder was anticipated, given that we
had earlier established that a hydrophobic group was
necessary for optimal binding to the lipid A moiety of
LPS.14-16,20-22,25-27 In contrast, a clear inverse correla-
tion between binding affinity and acyl chain length is
observed with the bis-acyl homologated acylhomosper-
mines, with 8a (bis-C8) displaying an ED50 of 0.325 µM,
a value that was even more favorable than that of
polymyxin B (1.2 µM), the reference compound in all
experiments. Although we were initially surprised by
this somewhat counterintuitive result, as pointed out
earlier, we observed that the longer homologues (8c-f)
were progressively less water-soluble and were precipi-
a Reagents: (a) Ac2O, py, DMAP, rt (for 3a), or RCOCl, DMAP,
py, rt (for 3b,c), or RCOOH, EDCl, THF, 10 h (for 3d-f). (b) TFA,
rt, 8 h.
study is depicted in Scheme 1. Following reported
procedures, commercially available spermine 1 was
converted to the corresponding strategically protected
homospermine derivative 2 in high yield.17-19 The free
amino functionality of the compound 2 was acylated
with various commercially available long chain aliphatic
acids/acid derivatives under standard conditions, pro-
viding the corresponding mono-N-acylated pentamine
derivatives 3a-f. While the compounds 3b,c were
acylated using the corresponding acid chlorides, the
higher chain length analogues 3d-f were synthesized
via EDCI-mediated coupling between 2 and the corre-
sponding carboxylic acids. Finally, removal of the Boc-
protecting groups with excess trifluoroacetic acid, con-
centration of the reaction mixture under vacuum, and
trituration of the resulting residue with diethyl ether
resulted in the desired tetrakis(trifluoroacetyl)ammo-
nium salts of the N-monoacylhomospermine derivatives
4a-f as white/off white amorphous solids in moderate
to good yields. The assigned structure and purity of the
products were confirmed by NMR (1H and 13C), mass
spectroscopy, and elemental analyses.
Following a similar reaction strategy as in Scheme
1, selective protection of both the terminal primary
amino groups of spermine 1 as the corresponding bis-
trifluoroacetamide, followed by Boc-protection of the two
secondary amino groups and subsequent deblocking of
the primary amines, resulted in the N2,N3-di-Boc-
spermine 5 in near quantitative yield (Scheme 2).
Homologation at both the free amino terminals was
accomplished via Michael addition of the amines with
acrylonitrile followed by Boc-protection of the resulting
secondary amines to give the tetra-Boc-protected amino
nitrile derivative 6. Hydrogenation of the nitrile groups
of 6 under standard conditions provided the correspond-
ing hexaamine derivative 7 in good yield. The free
primary amino groups of 7 were subjected to acylation
with various long chain aliphatic acids, and the adducts
were treated with excess trifluoroacetic acid to remove
the Boc-protecting groups. Removal of excess reagent
under high vacuum and trituration of the residue with
diethyl ether afforded the desired trifluoroacetic acid
salts of N1,N6-diacyl-bis-homospermine derivatives 8a-f
as white/off white amorphous solids. The identity and
purity of the products were verified by NMR (1H and
13C), mass spectroscopy, and elemental analyses.
The monoacyl compounds 4a-f were all freely water
soluble, but the bis-compounds 8a-f were significantly

Acylhomospermine Endotoxin-Sequestering Compounds
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7 2591
Scheme 2a
a Reagents: (a) (i) F3CCOOEt (2 equiv), MeOH, -78 to 0 °C, 1 h; (ii) Boc2O (excess), 0 °C to rt, 1 h; (iii) aq MeOH, NH3, rt, 25 h. (b)
(i) H2CdCHCN, MeOH, rt, 15 h; (ii) Boc2O, CH2Cl2, 90 min. (c) Pd(OH)2/C, H2, AcOH, 50 psi. (d) (i) RCOOH, EDCl, THF, 10 h; (ii) TFA,
rt, 8 h.
Table 1. Summary of Binding Affinities and in Vitro Biological Activity (NO and TNF-R Inhibition)a
a The Z′ factor of the HTS assay for quantifying ED50 (relative binding affinity) is 0.82, and the CVs at 0% and 100% probe displacement
are 4.1% and 6.2%, respectively.56 The CVs for NO and TNF-R inhibition assays, are, respectively, 3.2% and 4.7%.
tating out of solution when the compound stocks in
DMSO were being diluted in aqueous buffer. The
consequences of poor solubility were also manifested in
in vivo experiments (see below).
In Vitro Endotoxin-Neutralizing Activity. Mu-
rine, but not human monocytes, produce measurable
quantities of nitric oxide (NO), an important surrogate
marker of immune activation by bacterial products.28

2592 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7
Miller et al.
Figure 1. Structure of lipid A, the toxic moiety of bacterial
lipopolysaccharide. Numbers indicate acyl chain length.
Figure 2. Correlation between carbon number of the hydro-
carbon group in mono- (open stars, dotted line) and bis-acyl
(closed stars, solid line) and TNF-R inhibition in human blood
(top) and NO inhibition in murine J774A.1 cells (bottom). Data
represent means of duplicate values obtained from a repre-
sentative experiment.
In vitro NO inhibition assays have proven to be a
reliable and rapid primary biological screen for identify-
ing anti-LPS compounds.14 Mouse J774A.1 cells were
exposed to 10 ng/mL LPS, with or without graded
concentrations of the compounds. The dose-dependent
inhibition of NO (measured as nitrite) and 50% inhibi-
tory concentration (IC50) values for NO release are listed
in Table 1. For the monoacyl compounds, the shortest
(C1) analogue, 4a, is practically bereft of inhibitory
activity, as expected, with an IC50 of 171.4 µM. The C8
(4b) and C9 analogues (4c) are weak, with IC50s of 52.5
and 23.5 µM, respectively. However, the higher (C14-
C18) monoacyl compounds (4d-f) are all active with
comparable potencies (3.7-4.3 µM), suggesting that a
long-chain acyl group with a carbon number of 14-18
is necessary for optimal neutralization (Figure 2).
Among the bis-acyl compounds, the bis-C11 analogue
(8c) is most active (IC50 ) 1.44 µM), and longer chain
lengths are correlated with progressively lower affinities
(Figure 2), probably a consequence of poor solubility. In
all of these experiments, care was taken to verify that
the inhibition of NO was not due to cytotoxic effects of
the test compounds, using the XTT assay (data not
shown).
The uncontrolled production of proinflammatory cy-
tokines such as TNF-R, IL-1â, and IL-6 in response to
circulatory LPS plays a central role in the pathogenesis
of human septic shock.7 We therefore also examined the
activity of these compounds in a whole-blood cytokine
release assay.29,30 The multiplexed cytokine detection
system that we used measures six cytokines concur-
rently from each sample. As controls, we used phorbol
myristate acetate (PMA; 100 ng/mL) plus ionomycin (1
µM), or PMA plus phytohemagglutinin (2 µg/mL) as non-
LPS stimuli. None of the test compounds inhibited TNF-
R, IL-6, or IL-8 appreciably up to concentrations of 20
µM, verifying that the inhibition observed with LPS
stimulation was a consequence of sequestration by the
compounds. The IC50 values for LPS-induced TNF-R
inhibition listed in Table 1 indicate that the dependence
of TNF-R inhibition on the carbon number for both the
mono- and bis-acyl compounds is very similar to that
observed with NO inhibition (Figure 2). However, unlike
in the NO inhibition data, the long-chain monoacyl-
homospermine compounds (4d-f) show higher activity
in inhibiting TNF-R in human whole blood than 8a, the
most potent of the bis-series (Figure 2). We surmised
that this discrepancy could also be a result of differential
bioavailability arising from disparities in aqueous solu-
bilities of the mono- and bis-compounds. Serum albumin
is a known carrier protein for otherwise insoluble
ligands.31,32 It appeared possible that a greater fraction
of the more insoluble bis-analogues would be protein-
bound, thus reducing the concentration of free com-
pound to sequester LPS. We therefore wished to test
whether differences in in vivo bioavailability may alter
the biological outcome. Consequently, we tested 8a and
8c in addition to 4e in the mouse model of LPS-induced
lethality (see below).
Correlation of Binding Affinity and in Vitro
Biological Activity. Binding to LPS is a necessary, but
not sufficient, requisite for biological neutralization of
endotoxic activity. For example, polymyxin B (PMB;
decapeptide) and its nonapeptide derivative (PMBN)
bind LPS with comparable affinities, but PMBN is
virtually devoid of neutralizing activity.14 Furthermore,
although the BC fluorescent probe displacement assay
is an excellent rapid-throughput method to screen for
LPS-binding molecules, ED50 values obtained from this
assay are heavily weighted for electrostatic interactions
between LPS and its ligand,21 and the assay is not an

Acylhomospermine Endotoxin-Sequestering Compounds
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7 2593
account early in the rational design and development
of subsequent generations of potential antisepsis drugs.
Toxicity. The affinity of binding of a drug to its target
becomes of crucial importance when nonspecific binding
to pharmacologically irrelevant targets results in ad-
verse effects. However, low affinity can be overcome
simply by administering larger doses of the drug, by
virtue of mass action affects. This is clinically realizable
only when the toxicity profile (therapeutic index) is
favorable. Considerations of toxicity therefore super-
vene, and sometimes supplant, those of other pharma-
cological properties, particularly in the therapy of the
critically ill patient. A key consideration in the design
of the compounds we have presented in this paper is
the minimization of toxicity by specifically incorporating
hydrolytic lability (acyl linkage) as well as the use of
physiological, metabolically inert building blocks (sper-
mine and fatty acid), a classical strategy in “soft drug”
design.34,35 Members of the lipopolyamine class are of
low toxicity to mammalian cells and are being approved
by the FDA for human use for gene transfection as safe
alternatives to viral vectors.36,37 Indeed, our earlier work
on DOSPER demonstrated that this compound was
nontoxic to mice at concentrations well above therapeu-
tic doses.16 However, both the 4 and 8 series are cationic
amphipaths; in particular, the 8 series are analogous
to “Gemini surfactants”, so named after their twin-
headed structures,38 and could, possibly, display non-
specific cytotoxicity because of membrane-perturbing
activity. We thought it prudent to characterize the
surface activity of these compounds, correlate it to in
vitro cytotoxicity, and carefully examine if these would
have adverse consequences in vivo, prior to commencing
animal experiments. As expected, the ‘Gemini’-like 8a
and 8b (measured in 5% DMSO to ensure solubility; the
higher homologues were insoluble and could not be
tested), are indeed considerably surface active (data not
shown). For the 4 series (all of which were freely soluble
in 5% DMSO), there is a distinct correlation between
acyl chain length and surface tension lowering activity,
as could be expected, with homologues with longer acyl
chains being more surface active. We sought to quan-
titatively correlate surface activity and cell lysis first
in highly diluted (1:1000; diluted in physiological saline),
aged human whole blood. In this assay, erythrocytes
become exquisitely susceptible to membrane damage
and lysis, not only because of the increased osmotic
fragility of the erythrocytes due to depleted Na+ K+
ATPase activity39 but also due to the absence of “buffer-
ing” effects of plasma proteins. The hemolytic activity
of the 4 series monoacyl compounds were found to
correlate closely with their surface activity, a relation-
ship that was not quite so straightforward for the bis-
acyl 8 series because of the inferior solubility of the
latter compounds (data not shown). The pronounced
hemolytic activity of 4f and 4e (100% hemolysis at 1-5
µM) occasioned concern, and we questioned if the results
of this assay employing deliberately exaggerated eryth-
rocytic fragility were physiologically relevant, that is,
if these compounds would likely cause intravascular
hemolysis in vivo if administered parenterally. The
hemolytic activities of these compounds were therefore
reexamined using human whole blood and, to our relief,
we observed significant hemolysis starting to occur only
Figure 3. Correlation of binding affinity of the acylhomo-
spermines determined by BC fluorescent probe displacement
with NO inhibition in murine J774A.1 cells (bottom) and
TNF-R inhibition in human blood (top). Data represent means
of duplicate values obtained from a representative experiment.
accurate reporter of hydrophobic interactions.14 We have
previously shown this to be a crucial determinant of
biological neutralization.14 The identification of provi-
sional “hits” with the BC high-throughput screen in our
laboratory is therefore always confirmed in the murine
NO assay. With this caveat in mind, although we did
not expect a perfect correlation, it was of interest to
compare the NO and TNF-R inhibition profiles with
binding affinities observed with this congeneric series
of compounds. As can be seen in Figure 3, with the
exception of 4a, all of the other monoacyl compounds
bind LPS with ED50 values between ∼1 and 2 µM, while
only the longer acyl chain compounds (4d-f) are
biologically active, indicating that the fluorescence assay
does not discriminate adequately between high-affinity
binders and high-potency neutralizers. Similar trends
had also been observed with acyl chain lengths in
cationic lipopeptides.33 This result emphasizes the
necessity of employing a biological primary screen in
tandem with the displacement assay in order to derive
reliable structure-activity relationships in LPS-seques-
tering compounds. In contrast, there is an apparent
linear correlation between ED50 and neutralization
potency for the bis-acyl 8 series. This is simply a
consequence of the poor solubility of the 8 homologues.
Free aqueous concentrations of 8 are progressively
retarded with increasing chain length, diminishing
binding and, consequently, neutralization. It is note-
worthy that these differences are clearly manifested in
the outcomes of in vivo studies (see below). Physical
properties such as solubility will likely affect the phar-
macokinetics and pharmacodynamics of these com-
pounds significantly and will have to be taken into

2594 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7
Miller et al.
Table 2. Dose-Dependent Protection of CF-1 Mice Challenged
with a Supralethal Dose of 200 ng/Mouse (LD100 ) 100 ng) by
the Acylhomospermines in Cohorts of Five Animalsa
lethality
lethality
compd
dose
(µg/mouse)
compd
dose
(dead/total)
(dead/total)
4e
8a
8c
(µg/mouse) 4e 8a 8c
200
100
50
0/5* 1/5* 1/5*
10
0
5/5 5/5 4/5
5/5 5/5 5/5
0/5* 4/5
2/5 5/5
2/5
4/5
a Lethality was recorded at 24 h postchallenge. Asterisks
indicate statistically significant values, p < 0.05.
Table 3. Time Course of Protection Afforded by 4e in the
D-Galactosamine-Sensitized CF-1 Mouse Lethality Modela
time of 4e
administration
(h)
time of 4e
administration
(h)
lethality
(dead/total)
lethality
(dead/total)
-6
-4
-2
1/5*
1/5*
0/5*
0
+1
+2
0/5*
5/5
4/5
a Animals were injected with 200 µg of 4e ip at times noted with
respect to LPS challenge (200 ng/mouse). Lethality was recorded
at 24 h following LPS challenge. Asterisks indicate statistically
significant values, p < 0.05.
necessitated administration in 50% DMSO. As is evi-
dent from Table 2, a clear dose-response is observed,
with 4e affording complete (statistically significant)
protection at the 100 or 200 µg/mouse dose and partial
protection at the 50 µg dose. Both the bis-compounds
are inferior to 4e (Table 2), underlining the importance
of favorable pharmacodynamics properties enabling
adequate plasma concentrations of free drug to ef-
fectively sequester LPS.
Figure 4. Hemolytic activity of mono- and bis-acyl homolo-
gated spermines in whole human blood (top) and the effect of
addition of human serum albumin (650 µM) on the hemolysis
of 1:1000 diluted, washed, human erythrocytes, determined
by automated video microscopy (bottom).
at millimolar concentrations (Figure 4, top panel). In
these latter experiments, melittin, an R-helical 26-
residue hemolytic peptide isolated from bee venom,40,41
caused hemolysis at low micromolar concentrations.
Hypothesizing that the abrogation of hemolysis in whole
blood was a consequence of binding of the otherwise
surface active compounds to plasma proteins, we re-
tested the effects of 4e and 4f on diluted, washed human
erythrocytes, as described previously, in the absence and
presence of physiological concentrations of human se-
rum albumin. The presence of 650 µM albumin com-
pletely abolished the hemolytic activity of both these
compounds (Figure 4, bottom panel), indicating that a
large fraction of these compounds was bound to albumin
and that the protein-bound form was unlikely to exert
toxicity. Indeed, ongoing experiments with related
compounds show no demonstrable dermal toxicity,
neither acute nor chronic, when administered subcuta-
neously to mice at concentrations as high as 10 mg/mL
(manuscripts in preparation).
Protection against Endotoxin-Induced Lethality
in Mice. We elected to characterize the protective
effects of 4e, the most potent compound in the human
TNF-R inhibition assay; 8a, which was most active in
inhibiting NO release in murine J774A.1 cells; and 8c,
which was of lower potency than either 4e or 8a in both
assays. A well-established animal model of LPS-induced
lethality in mice sensitized by D-galactosamine was
used.42,43 A supralethal dose (twice the dose causing
100% lethality) of 200 ng/mouse was administered
intraperitoneally (ip) to groups of five mice sensitized
with D-galactosamine, along with concurrent, separate
ip injections of graded doses of compound, and lethality
was observed at 24 h. The highly soluble 4e was
dissolved in saline. The poor solubility of 8a and 8c
In our earlier studies on DOSPER,16 we had observed
that the temporal window of protection was very short.
DOSPER had to be administered concurrent with the
LPS challenge, and significant mortality resulted when
the compound was given even 15 min prior to LPS. This
was attributed to the extreme hydrolytic susceptibility
to serum esterases of the o-ester linkages of the oleoyl
groups,16 and it was of interest to explore if the amide-
linked acylhomospermines would display a longer plasma
half-life, thus affording a longer window of protection.
It is evident from the time-course experiment (Table 3)
that 4e is indeed apparently much longer lived, with
near-complete protection evident when the compound
is administered 6 h prior to LPS challenge (Table 3). In
a few cohorts, we observed that this time window of
protection was 4 h when 4e was administered intrave-
nously (iv) and was 8 h when given subcutaneously,
signifying a depot effect and prolonged release into
systemic circulation in the latter experiments (data not
shown). Mention was made earlier that one of the
heuristics that we have used in developing potential
anti-LPS compounds is to “design in” metabolic lability.
The consequence of lability of a molecule is, of course,
poor pharmacokinetic behavior because of rapid elimi-
nation from circulation, a situation that would be
unacceptable for an orally bioavailable drug for the
treatment of a chronic condition. However, rapid clear-
ance, per se, is not of concern in the setting of an
intensive care unit. For instance, sodium nitroprusside,
used in the management of malignant hypertension has
a t1/2 of a few seconds. Administered iv, the low t1/2 is
actually useful in titrating the hypotensive effect on a

Acylhomospermine Endotoxin-Sequestering Compounds
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7 2595
35.7, 37.4, 43.4, 43.8, 44.3, 44.9, 46.9, 78.9, 79.5, 79.9, 155.5,
156.2, 156.4, 170.37; MS (FAB) calcd for C35H67N5O9 m/z 701.4,
found 702.4 (MH)+.
minute-by-minute basis. In the therapy of a critically
ill patient in septic shock, an ideal regimen would be a
rapid iv bolus infusion to achieve rapid plasma levels
followed by a constant infusion to maintain an optimal
drug concentration. We are presently examining ana-
logues with alkylated spermines, as well as compounds
with the hydrophobic functionalities in ester, urea, and
carbamate linkages, to optimize pharmacokinetics, while
low toxicity is retained. It is to be noted that 4e is
without effect if administered 1 h following LPS chal-
lenge. The result of this time-course experiment is most
instructive in that it suggests that if LPS-sequestering
compounds, such as the acylhomospermines, are ever
to find utility in the clinic, they will have to be used as
prophylactic agents, rather than to treat sepsis once the
inflammatory cascades are already set in motion. This
may indeed be feasible and, indeed, desirable, since not
only have many of the therapeutic strategies that target
downstream processes such as blockade of TNF-R or IL-
1â failed44,45 but also because the predisposing factors
for septic shock are very well recognized.46,47
In summary, we have described in this paper a
detailed characterization of the endotoxin-binding and
-neutralizing properties of a synthetically easily acces-
sible class of nontoxic acylhomospermines. It is gratify-
ing that considerably potent and yet nontoxic analogues
have been identified in the early phases of development.
Can greater binding affinity and specificity be achieved
without incurring the cost of additional toxicity or
unfavorable physical properties? One possible approach
to addressing the question of augmenting affinity is to
test the hypothesis whether incorporating optimally
placed H-bond/donor atoms would serve to maximize
interactions with the lipid A backbone.14,21 We are
currently examining these questions in compounds with
novel nonpolyamine scaffolds.
General Procedure for the Synthesis of Compounds
3b,c. To a solution of 2 (0.24 g, 0.36 mmol) in anhydrous
pyridine (5 mL) at 0 °C were added the respective acid
chlorides (5 equiv) [RCOCl: R ) C8H17, C9H19, and C11H23,
respectively] and a catalytic amount of DMAP. The reaction
mixture was stirred well for 10 h at room temperature followed
by quenching the reaction by addition of water (15 mL). The
resulting solution was diluted with ethyl acetate (50 mL), and
the organic layer was separated and washed sequentially with
ice-cold 10% HCl solution (20 mL × 2), satd aq NaHCO3
solution (20 mL × 3), water (20 mL), and brine (20 mL). The
organic layer was dried over Na2SO4 and concentrated. The
resulting Boc-protected monoacylated derivatives 3b,c (62%
and 66% crude yield) were dried under high vacuum overnight
and used as such for the next reaction.
General Procedure for the Synthesis of Compounds
3d-f. To a solution of compound 2 (0.30 mmol) in anhydrous
THF (5 mL) at 0 °C were added the respective carboxylic acid
(4 equiv) [RCOOH: R ) C14H29, C16H33, C18H37, respectively]
and EDCI (4 equiv). The resulting mixture was stirred at room
temperature for 10 h. After removal of solvent, the residue
was taken up in ethyl acetate (25 mL) and washed sequentially
with water (10 mL × 3), satd aq NaHCO3 solution (20 mL ×
2), and brine (10 mL). The organic layer was dried over Na2-
SO4 and solvent removed under reduced pressure. The result-
ing viscous liquids were purified by flash column chromatog-
raphy (hexanes/EtOAc ) 1:1) to obtain the Boc-protected
monoacylated polyamines 3d-f.
N1-Pentadecanoyl-N4,N9,N13,N16-tetrakis(tert-butoxy-
carbonyl)-1,16-diamino-4,9,13-triazahexadecane (3d): vis-
cous oil; yield 82%; 1H NMR (400 MHz, CDCl3) δ 0.87 (br t, J
) 6.8 Hz, 3H), 1.22-1.31 (m, 24H), 1.4-1.51 (m, 39H), 1.59-
1.78 (br m, 7H), 2.17 (t, 2H, J ) 6.8 Hz), 3.08-3.27 (br m,
16H); 13C NMR (100.6 MHz, CDCl3) δ 14.3, 22.9, 25.7, 26.0,
27.8, 28.6, 28.7, 29.1, 29.5, 29.6, 29.7, 29.8, 29.9, 32.1, 35.5,
37.1, 43.4, 44.9, 46.9, 79.3, 79.6, 80.0, 155.6, 156.3, 173.5; MS
(FAB) calcd for C48H93N5O9 m/z 883.7, found 884.6 (MH)+.
N1-Heptadecanoyl-N4,N9,N13,N16-tetrakis(tert-butoxy-
carbonyl)-1,16-diamino-4,9,13-triazahexadecane (3e): vis-
1
cous oil; yield 86%; H NMR (400 MHz, CDCl3) δ 0.87 (t, J )
Experimental Section
7.3 Hz, 3H), 1.24-1.28 (m, 29H), 1.43-1.46 (m, 38H), 1.62-
1.85 (br m, 7H), 2.15-2.17 (m, 2H), 3.05-3.32 (br m, 16H);
13C NMR (100.6 MHz, CDCl3) δ 14.3, 22.8, 26.0, 27.7, 28.6,
28.7, 29.1, 29.5, 29.6, 29.7, 29.8, 29.9, 32.1, 35.5, 37.1, 43.4,
44.9, 46.9, 79.3, 79.6, 80.0, 155.6, 156.3, 173.5; MS (FAB) calcd
for C50H97N5O9 m/z 911.7, found 912.4 (MH)+.
Chemistry. All of the solvents and reagents used were
obtained commercially and used as such unless noted other-
wise. Moisture- or air-sensitive reactions were conducted under
argon atmosphere in oven-dried (120 °C) glass apparatus. THF
was distilled from sodium benzophenone ketyl, while dichlo-
romethane was distilled over calcium hydride, prior to use.
Solvents were removed under reduced pressure using standard
rotary evaporators. Flash column chromatography was carried
out using silica gel 60 (230-400 mesh), while thin-layer
chromatography (TLC) was carried out on silica gel HLF,
precoated glass plates. All yields reported refer to isolated
material judged to be homogeneous by TLC and NMR spec-
troscopy. Unless noted otherwise, NMR spectra were recorded
with the chemical shifts (δ) reported in ppm relative to Me4Si
N1-Nonadecanoyl-N4,N9,N13,N16-tetrakis(tert-butoxy-
carbonyl)-1,16-diamino-4,9,13-triazahexadecane (3f): vis-
1
cous liquid; yield 82%; H NMR (400 MHz, CDCl3) δ 0.87 (br
t, J ) 6.8 Hz, 3H), 1.24-1.28 (m, 34H), 1.43-1.46 (m, 36H),
1.62-1.80 (br m, 8H), 2.15-2.17 (m, 2H), 3.05-3.32 (br m,
16H); 13C NMR (100.6 MHz, CDCl3) δ 14.3, 22.8, 26.0, 27.8,
28.6, 28.7, 29.5, 29.6, 29.7, 29.8, 29.9, 32.1, 35.5, 37.2, 43.4,
44.9, 46.9, 76.9, 79.1, 79.6, 80.0, 155.6, 156.3, 156.6, 173.5; MS
(FAB) calcd for C52H101N5O9 m/z 939.7, found 940.9 (MH)+.
1
(for H) and CDCl3 (for 13C) or DMSO-d6 (for 13C) as internal
General Procedure for the Synthesis of Compounds
4a-f. Boc-protected monoacylated derivatives 3a-f were
dissolved in 8 mL of trifluoroacetic acid and stirred for 8 h at
ambient temperature. Excess solvent was removed under
reduced pressure and the residue dried under high vacuum
overnight. The resulting sticky residue was washed thoroughly
with diethyl ether to obtain the desired compounds 4a-f as
off white flaky solids.
standards, respectively.
N1-Acetyl-N4,N9,N13,N16-tetrakis(tert-butoxycarbonyl)-
1,16-diamino-4,9,13-triazahexadecane (3a). To a solution
of compound 217-19 (0.24 g, 0.36 mmol) in anhydrous methylene
chloride (8 mL) at 0 °C were added pyridine (0.14 g, 1.8 mmol),
acetic anhydride (0.18 g, 0.18 mmol), and a catalytic amount
of DMAP. The reaction mixture was stirred well for 10 h at
room temperature. After removal of solvent, the residue was
dissolved in ethyl acetate (25 mL) and washed with water (10
mL × 3) followed by brine (10 mL). The organic layer was dried
over Na2SO4, filtered, and concentrated under reduced pres-
sure. The resulting viscous liquid was purified by flash column
chromatography (hexanes/EtOAc ) 1:1) to afford the Boc-
protected monoacylated derivative 3a (0.21 g, 85%) as a viscous
liquid: 1H NMR (400 MHz, CDCl3) δ 1.36-1.42 ((m, 40H),
1.60-1.69 (br m, 6H), 1.94 (s, 3H), 3.1-3.23 (br m, 16H); 13C
NMR (100.6 MHz, CDCl3) δ 23.5, 26.0, 26.1, 27.6, 28.5, 28.6,
N1-Acetyl-1,16-diamino-4,8,13-triazahexadecane tet-
1
rakis(trifluoroacetic acid) salt (4a): yield 82%; H NMR
(400 MHz, DMSO) δ 1.58-1.75 (m, 4H), 1.80 (s, 3H), 1.83-
2.01 (m, 4H), 2.82-3.04 (br m, 16H). 3.08-3.15 (m, 2H), 8.07
(br s, 4H), 8.87 (br s, 6H); 13C NMR (100.6, MHz, DMSO-d6) δ
22.5, 22.6, 23.8, 36.1, 43.8, 46.1, 158.8, 169.7; MS (FAB) calcd
for C15H35N5O m/z 301.4, found 302.4 (MH)+ (free base).
N1-Nonanoyl-1,16-diamino-4,8,13-triazahexadecane tet-
rakis(trifluoroacetic acid) salt (4b): yield 70%; 1H NMR

2596 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7
Miller et al.
(400 MHz, D2O) δ 0.85 (br s, 3H), 1.17 (br s, 10H), 1.55-1.65
(m, 2H), 1.76 (br s, 4H), 1.85-1.95 (m, 2H), 2.05-2.18 (m, 4H),
2.22-2.26 (m, 2H), 3.05-3.20 (bm, 14H), 3.26-3.29 (m, 2H);
13C NMR (100.6 MHz, DMSO-d6) δ 14.4, 22.5, 22.8, 23.0, 24.2,
25.7, 26.4, 28.9, 29.0, 29.1, 31.6, 35.8, 36.0, 36.7, 44.4, 44.5,
45.2, 46.7, 159.3, 174.2; MS (FAB) calcd for C22H49N5O m/z
301.4, found 300.7 (MH)+ (free base).
N1-Decanoyl-1,16-diamino-4,8,13-triazahexadecane tet-
rakis(trifluoroacetic acid) salt (4c): yield 56%; 1H NMR
(400 MHz, DMSO-d6) δ 0.85 (br t, J ) 6.8 Hz, 3H), 1.23 (bs,
12H, CH2), 1.45-1.53 (m, 2H), 1.61 (m, 4H), 1.68-1.74 (m,
2H), 1.85-1.96(m, 4H), 2.06 (t, J ) 7.4 Hz, 2H), 2.85-2.98
(br m, 14H), 3.08-3.15 (m, 2H), 7.92 (br m, 4H), 8.60-8.74
(br m, 6H); 13C NMR (100.6 MHz, DMSO-d6) δ 13.9, 22.1, 22.4,
22.6, 24.3, 25.2, 26.4, 28.7, 28.8, 28.9, 31.2, 35.2, 35.5, 36.1,
43.8, 43.9, 44.9, 46.5, 159.3, 159.4, 174.5; MS (FAB) calcd for
C23H51N5O m/z 413.7, found 414.5 (MH)+ (free base).
N1-Pentadecanoyl-1,16-diamino-4,8,13-triazahexadec-
ane tetrakis(trifluoroacetic acid) salt (4d): yield 81%; 1H
NMR (400 MHz, DMSO) δ 0.85 (t, J ) 6.8 Hz, 3H), 1.22 (br s,
22H), 1.42-1.52 (m, 2H), 1.60-1.75 (m, 6H), 1.87-1.95 (m,
4H), 2.05 (t, J ) 7.4 Hz, 2H), 2.85-3.02 (br m, 14H), 3.07-
3.11 (m, 2H), 7.99 (br s, 4H), 8.89 (br m, 6H); 13C NMR (100.6
MHz, DMSO-d6) δ 13.9, 22.1, 22.6, 25.2, 28.7, 28.8, 28.9, 29.0,
32.3, 43.8, 158.9, 159.2, 173.2; MS (FAB) calcd for C28H61N5O
m/z 483.7, found 484.6 (MH)+ (free base).
N1-Heptadecanoyl-1,16-diamino-4,8,13-triazahexadec-
ane tetrakis(trifluoroacetic acid) salt (4e): yield 78%; 1H
NMR (400 MHz, DMSO-d6) δ 0.85 (t, J ) 6.8 Hz, 3H), 1.23 (br
s, 26H), 1.45-1.50 (m, 2H), 1.61-1.69 (br s, 4H), 1.70-1.77
(m, 3H), 1.88-1.92 (m, 4H), 2.05 (t, J ) 7.4 Hz, 2H), 2.80-
3.01 (br m, 14H), 3.07-3.11 (m, 2H), 7.91-7.99 (br m, 4H),
8.66-8.79 (br m, 6H); 13C NMR (100.6 MHz, DMSO-d6) δ 13.9,
22.1, 22.4, 22.6, 23.8, 25.2, 26.1, 28.7, 28.8, 28.9, 29.0, 31.3,
35.3, 35.5, 36.2, 43.8, 43.9, 44.7, 46.1, 158.7, 172.7; MS (FAB)
calcd for C30H65N5O m/z 511.7, found 512.7 (MH)+ (free base).
of di-tert-butyl dicarbonate (1.05 g, 4.8 mmol) in CH2Cl2 (10
mL). The resulting solution was stirred for 90 min at ambient
temperature, concentrated in vacuo, and purified by flash
column chromatography (hexanes-EtOAc ) 3:2) to give
compound 6 (1.12 g, 64%) as a viscous oil: 1H NMR (400 MHz,
CDCl3) δ 1.45 and 1.47 (2s, 42H), 1.72-1.79 (m, 4H), 2.55-
2.68 (m, 4H), 3.08-3.20 (br s, 6H), 3.21-3.30 (m, 4H), 3.46-
3.51 (m, 4H); 13C NMR (100.6 MHz, CDCl3) δ 17.0, 17.6, 25.6,
26.0, 28.4, 28.5, 43.5, 44.0, 44.6, 45.5, 46.6, 47.0, 79.5, 80.6,
154.7, 155.2, 155.5; MS (FAB) calcd for C36H64N6O8 m/z 708.4,
found 709.5 (MH)+.
N4,N8,N13,N17-Tetrakis(tert-butoxycarbonyl)-1,20-di-
amino-4,8,13,17-tetrazaicosane (7). A solution of bis-nitrile
6 (0.9 g, 1.26 mmol) in 30 mL of glacial acetic acid was
hydrogenated over Pd(OH)2/C (0.9 g) at 50 psi hydrogen
pressure for 2 h. The catalyst was removed by filtration and
the residue washed thoroughly with methanol. After concen-
trating the combined filtrate under vacuum, the residual oil
was dissolved in ethyl acetate (100 mL) and washed sequen-
tially with 1 N NaOH (50 mL × 2) and water. After drying
over MgSO4, the solution was concentrated and left under high
vacuum overnight to afford compound 7 (0.80 g, 97%) as a
viscous solid: 1H NMR (400 MHz, CDCl3) δ 1.45 (s, 42H),
1.62-1.73 (m, 8H), 2.66 (br s, 4H,), 3.15-3.26 (m, 18H); 13C
NMR (100.6 MHz, CDCl3) 25.6, 28.5, 28.6, 38.7, 43.8, 44.9,
46.7, 79.4, 79.6, 155.5; MS (FAB) calcd for C36H72N6O8 m/z
716.5, found 717.6 (MH)+.
General Procedure for the Synthesis of Compounds
8a-f. To a solution of compound 7 (0.10 g, 0.15 mmol) in
anhydrous THF (5 mL) at 0 °C was added the respective
carboxylic acid (8 equiv) [RCOOH: R ) C8H17, C9H19, C10H21,
C13H27, C15H31, C17H35, respectively) and EDCI (8 equiv)
followed by stirring of the mixture for 10 h at room temper-
ature. After removal of solvent, the residue was taken up in
ethyl acetate (25 mL) and washed sequentially with water (10
mL × 3), ice-cold 10% HCl solution (10 mL × 2), saturated aq
NaHCO3 solution (20 mL × 2), and brine (10 mL). The organic
layer was dried over Na2SO4 and filtered and solvent evapo-
rated under reduced pressure. The resulting Boc-protected bis-
acylated polyamines were dried under high vacuum and
dissolved in 8 mL of dry trifluoroacetic acid and stirred at room
temperature for 8 h. Excess solvent was removed under
reduced pressure and the residue left under high vacuum
overnight. The resulting sticky residue was thoroughly washed
with diethyl ether to obtain the desired compounds 8a-f as
off white flaky solids.
N1-Nonadecanoyl-1,16-diamino-4,8,13-triazahexadec-
ane tetrakis(trifluoroacetic acid) salt (4f): yield 82%; 1H
NMR (400 MHz, DMSO-d6) δ 0.85 (t, J ) 6.6 Hz, 3H), 1.22 (br
s, 32H), 1.45-1.51 (m, 2H), 1.60-1.78 (br m, 4H), 1.88-1.98
(m, 4H), 2.05 (t J ) 7.4 Hz, 2H), 2.85-3.06 (br m, 14H), 3.09-
3.12 (m, 2H), 7.99 (br s, 4H), 8.73-8.86 (br m, 6H); 13C NMR
(100.6 MHz, DMSO-d6) δ 13.9, 22.1, 22.5, 24.3, 22.6, 25.28,
26.4, 28.7, 28.8, 28.9, 29.0, 31.3, 35.4, 35.5, 36.2, 43.8, 43.9,
44.9, 46.1, 158.4, 156.5, 172.7; MS (FAB) calcd for C32H69N5O
m/z 539.7, found 540.8 (MH)+ (free base).
N1,N20-Dinonanoyl-1,20-diamino-4,8,13,17-tetraza-
icosane tetrakis(trifluoroacetic acid) salt (8a): yield 60%;
1H NMR (400 MHz, DMSO-d6) δ 0.85 (t, J ) 7.2 Hz, 6H), 1.23
(br s, 20H), 1.46-1.51, (m, 4H), 1.62-1.68 (b s, 4H), 1.69-
1.74 (m, 4H), 1.90-1.98 (m, 4H), 2.05 (t, J ) 7.3 Hz, 4H), 2.85-
3.01 (br m, 16H), 3.08-3.12 (m, 4H), 7.98 (t, J ) 5.6 Hz, 2H),
8.78 and 8.93 (2s, 8H); 13C NMR (125.7 MHz, DMSO-d6) δ 14.3,
22.4, 22.8, 25.6, 23.0, 25.6, 26.5, 29.0, 29.1, 31.6, 35.7, 35.9,
44.3, 46.5, 159.2, 173.1; MS (FAB) calcd for C34H72N6O2 m/z
596.5, found 597.7 (MH)+ (free base).
N1,N20-Didecanoyl-1,20-diamino-4,8,13,17-tetraza-
icosane tetrakis(trifluoroacetic acid) salt (8b): yield 76%;
1H NMR (500 MHz, DMSO-d6) δ 0.85 (t, J ) 6.8 Hz, 6H), 1.23
(br s, 24H), 1.46-1.51 (m, 4H), 1.63-1.75 (m, 8H), 1.90-2.00
(m, 4H), 2.06 (t, J ) 7.40 Hz, 4H), 2.86-2.98 (br m, 16H), 3.09-
3.12 (m, 4H), 7.98 (t, J ) 5.7 Hz, 2H), 8.75 and 8.89, 2s, 8H);
13C NMR (125.7 MHz, DMSO-d6) δ 14.3, 22.5, 22.8, 23.0, 25.6,
26.5, 29.1, 29.2, 29.3, 31.6, 35.7, 35.9, 44.3, 44.4, 45.1, 46.5,
159.2, 173.2; (FAB) calcd for C36H76N6O2 m/z 624.5, found 625.7
(MH)+ (free base).
N1,N20-Didodecanoyl-1,20-diamino-4,8,13,17-tetraza-
icosane tetrakis(trifluoroacetic acid) salt (8c): yield 72%;
1H NMR (500 MHz, DMSO-d6) δ 0.85 (t, J ) 6.8 Hz, 6H),1.23,
(br s, 32H), 1.45-1.51 (m, 4H), 1.65 (b s, 4H), 1.68-1.75 (m,
4H), 1.91-1.98 (m, 4H), 2.05 (t, J ) 7.3 Hz, 4H), 2.86-3.01
(3m, 16H), 3.08-3.12 (m, 4H), 7.99 (t, J ) 5.7 Hz, 2H), 8.76
and 8.90 (2s, 8H); 13C NMR (125.7 MHz, DMSO-d6) δ 14.3,
22.5, 22.8, 23.0, 25.6, 26.5, 29.1, 29.2, 29.3, 29.4, 29.4, 31.7,
N4,N9-Bis(tert-butoxycarbonyl)-1,12-diamino-4,9-diaza-
dodecane (5). To a solution of spermine 1 (1.0 g, 4.95 mmol)
in methanol (70 mL) at -78 °C was added dropwise ethyl
trifluoroacetate (1.40 g, 9.90 mmol) over 30 min and the
solution stirred for another 30 min. The temperature was
increased to 0 °C and an excess of di-tert-butyl dicarbonate
(3.17 g, 14.85 mmol) in methanol (10 mL) was added over 10
min. The reaction was then warmed to 25 °C and stirred for a
further 17 h. The trifluoroacetate protecting groups were then
removed in situ by increasing the pH of the solution to above
11 with concentrated aqueous ammonia and then stirred at
25 °C for 15 h. After removal of solvent under vacuum, the
residue was purified by flash column chromatography (CH2-
Cl2-MeOH-concd aq NH3 50:10:1) to afford the title com-
pound 7 as a colorless viscous oil (1.92 g, 97%): 1H NMR (400
MHz, CDCl3) δ 1.43 and 1.49 (2s, 22H), 1.66 (br s, 4H), 2.89
(br s, 4H), 3.17-3.29 (2 br s, 8H), 6.27 (br s, 4H, exchangeable
with D2O); 13C NMR (100.6 MHz, CDCl3) δ 23.3, 25.3, 25.7,
27.4, 28.0, 28.3, 28.5, 28.7, 29.5, 31.7, 32.4, 33.2, 38.9, 39.3,
43.7, 44.2, 46.2, 46.5, 56.4, 76.9, 79.1, 81.0, 82.4, 155.4, 155.6;
MS (FAB) calcd for C20H42N4O4 m/z 402.2, found 403.1 (MH)+.
N3,N7,N12,N16-Tetrakis(tert-butoxycarbonyl)-1,18-dicy-
ano-3,7,12,16-tetrazaoctadecane (6). To a solution of com-
pound 5 (1.0 g, 2.48 mmol) in methanol (50 mL) was added
acrylonitrile (0.26 g, 5 mmol) and the mixture stirred at room
temperature for 15 h. After removal of solvent under high
vacuum, the crude bis-nitrile derivative (1.2 g, 95%) was
dissolved in CH2Cl2 (50 mL) followed by addition of a solution

Acylhomospermine Endotoxin-Sequestering Compounds
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7 2597
35.7, 35.9, 44.3, 44.3, 45.1, 46.5, 159.1, 173.1; MS (FAB) calcd
were added to the cell medium and left to incubate overnight
for 16 h. Polymyxin B was used as reference compound in each
plate. Positive (LPS stimulation only) and negative controls
(J774A.1 medium only) were included in each experiment.
Nitrite concentrations were measured by adding 30 µL of equal
volumes of Griess reagents (0.1% NED solution in ddH2O and
1% sulfanilamide, 5% phosphoric acid solution in ddH2O) to
the wells and incubating for 15 min at room temperature in
the dark. Absorbance at 535 nm was measured using a
Molecular Devices Spectramax M2 multifunction plate reader
(Sunnyvale, CA). Nitrite concentrations were interpolated from
standard curves obtained from serially diluted sodium nitrite
standards.
In Vitro XTT Cytotoxicity Assay. The determination of
cell viability was accomplished by the addition of an XTT50
solution to J774A.1 cultures treated with graded concentra-
tions of the test compounds. Cell culture and plating proce-
dures were performed as described previously for nitric oxide
measurement. Cytotoxicity was measured the following day
by the addition of 30 µL/well of XTT/phenazine methosulfate
(PMS) solution (XTT solution, 2 mM in PBS, pH 7.4, pH
adjusted to 6.0-6.5; PMS solution, 0.92 mg/mL in PBS, pH
7.4; solutions mixed at a ratio of 8 mL of XTT solution to 200
µL PMS solution) followed by an incubation time of 1.5 h at
37 °C. Absorbance was read at 490 nm with scatter correction
at 690 nm.
for C40H84N6O2 m/z 680.5, found 681.9 (MH)+ (free base).
N1,N20-Dipentdecanoyl-1,20-diamino-4,8,13,17-tetraza-
icosane tetrakis(trifluoroacetic acid) salt (8d): yield 48%;
1H NMR (400 MHz, DMSO-d6) δ 0.85 (t, J ) 7.0 Hz, 6H), 1.23
(br s, 44H), 1.40-1.51 (m, 4H), 1.62-1.75 (m, 8H), 1.89-1.99
(m, 4H) 2.05 (t, J ) 7.4 Hz, 4H), 2.85-3.02 (br m, 16H), 3.08-
3.12 (m, 4H), 7.99 (t, J ) 6.4 Hz, 2H), 8.68 and 8.84 (2br s,
8H); 13C NMR (100.6 MHz, DMSO-d6) δ 14.0, 22.1, 22.4, 22.7,
25.3, 26.2, 28.7, 28.8, 28.9, 29.0, 29.1, 31.3, 35.3, 35.5, 43.9,
44.7, 46.1, 158.5, 172.8; MS (FAB) calcd for C46H96N6O2 m/z
764.5, found 765.8 (MH)+ (free base).
N1,N20-Diheptadecanoyl-1,20-diamino-4,8,13,17-tetraza-
icosane tetrakis(trifluoroacetic acid) salt (8e): yield 81%;
1H NMR (400 MHz, DMSO-d6) δ 0.85 (t, J ) 6.9 Hz, 6H), 1.23
(br s, 56H), 1.44-1.5 (m, 4H), 1.60-1.72, m, 6H), 1.9-2.1 (br
m, 8H), 2.82-3.15 (br m, 18H), 7.99 (m, 2H), 8.65 and 8.80
(2br s, 8H); 13C NMR (125.7 MHz, DMSO-d6) δ 13.9, 22.0, 22.4,
22.6, 25.2, 26.1, 28.6, 28.7, 29.0, 31.2, 35.3, 35.4, 43.8, 43.9,
44.7, 46.0, 172.1; MS (FAB) calcd for C50H104N6O2 m/z 820.7,
found 821.9 (MH)+ (free base).
N1,N20-Dinonadecanoyl-1,20-diamino-4,8,13,17-tetraza-
icosane tetrakis(trifluoroacetic acid) salt (8f): yield 48%;
1H NMR (400 MHz, DMSO-d6) δ 0.85 (t, J ) 6.9 Hz, 6H), 1.22
(s, 62H), 1.42-1.51 (m, 4H), 1.61-1.75 (m, 8H), 1.90-2.00 (m,
4H), 2.05 (t, J ) 7.4 Hz, 4H), 2.81-3.0 (br m, 14H), 3.05-3.13
(m, 4H), 7.99 (t, J ) 5.6 Hz, 2H), 8.75 and 8.87 (br m, 8H); 13
C
Multiplexed Cytokine Assay ex Vivo in Human Blood.
Aliquots (100 µL) of fresh whole blood, anticoagulated with
EDTA, obtained by venipuncture from healthy human volun-
teers with informed consent and as per guidelines approved
by the Human Subjects Experimentation Committee, was
exposed to an equal volume of 50 ng/mL of E. coli 0111:B4
LPS, with graded concentrations of test compounds diluted in
saline for 4 h in a 96-well microtiter plate.29,51 The effect of
the compounds on modulating cytokine production was exam-
ined using a FACSArray multiplexed flow-cytometric bead
array (CBA) system (Becton-Dickinson-Pharmingen, San Jose,
CA). The system uses a sandwich ELISA-on-a-bead prin-
ciple52,53 and is comprised of six populations of microbeads that
are spectrally unique in terms of their intrinsic fluorescence
emission intensities (detected in the FL3 channel of a standard
flow cytometer). Each bead population is coated with a distinct
capture antibody to detect six different cytokines concurrently
from biological samples (the human inflammation CBA kit
includes TNF-R, IL-1â, IL-6, IL-8, IL-10, and IL-12p70). The
beads are incubated with 30 µL of sample, and the cytokines
of interest are first captured on the bead. After washing the
beads, a mixture of optimally paired second antibodies con-
jugated to phycoerythrin is added which then forms a fluo-
rescent ternary complex with the immobilized cytokine, the
intensity (measured in the FL2 channel) of which is propor-
tional to the cytokine concentration on the bead. The assay
was performed according to protocols provided by the vendor.
Standard curves were generated using recombinant cytokines
provided in the kit. The data were analyzed with the CBA
software suite that is integral to the FACSArray system.
Surface Tension Measurements and Hemolytic Activ-
ity. Because the acylhomospermines are cationic amphipaths,
these compounds are likely to have membrane-perturbing
activity, which may be a principal manifestation of their
toxicity. We therefore attempted to correlate their surface
activity with hemolytic activity as has been reported recently.54
The surface activity of the compounds was measured via
dynamic bubble pressure and surface age tensiometry55 using
a Kru¨ss PocketDyne instrument (Kru¨ss GmbH, Hamburg,
Germany). Samples were at 500 µM concentration in 50 mM
Tris buffer, pH 7.4 containing 5% DMSO. The instrument was
calibrated with water at 25 °C (72 mN/m), and surface tension
values were recorded over a range of bubble surface ages from
100 to 1500 ms at 25 °C. Erythrocyte damage was measured
using two different techniques. In the first, hemolysis was
quantified using extremely diluted, aged human whole blood
such that the effects of the compounds binding to plasma
proteins would be negligible and the hemolytic activity would
NMR (100.6 MHz, DMSO-d6) 13.9, 22.1, 22.7, 23.0, 23.6, 25.2,
26.3, 27.5, 28.7, 28.8, 28.9, 29.0, 31.3, 35.3, 35.5, 36.7, 44.2,
44.81, 46.3, 53.3, 172.7; MS (FAB) calcd for C54H112N6O2 m/z
877.1, found 878.0 (MH)+ (free base).
Rapid-Throughput Fluorescence Displacement Assay
for Quantifying Binding Affinities to LPS. The BODIPY-
TR-cadaverine [BC; (5-(((4-(4,4-difluoro-5-(2-thienyl)-4-bora-
3a,4a-diaza-s-indacen-3-yl)phenoxy)acetyl)amino)pentyl-
amine hydrochloride; obtained from Molecular Probes, Inc.,
Eugene, OR] displacement assay to quantify the affinities of
binding of compounds to LPS has been described in detail
recently.25 This assay was performed in a rapid-throughput
format as follows: the first column (16 wells) of a Corning
Nonbinding Surface 384-well flat-bottom black fluorescence
microplate contained 15 test compounds plus polymyxin B, all
at 5 mM in DMSO, and were serially diluted 2-fold in 50 mM
Tris buffer, pH 7.4, across the remaining 23 columns, achieving
a final dilution of 0.596 nM in a volume of 40 µL. Polymyxin
B (PMB), a peptide antibiotic known to bind and neutralize
LPS,48 served as the positive control and reference compound
for every plate, enabling the quantitative assessment of
repeatability and reproducibility (CV and Z′ factors) for the
assay. Automated liquid handling was performed on a Preci-
sion 2000 automated microplate pipetting system, pro-
grammed using the Precision Power software, Bio-Tek Instru-
ments Inc., VT.
Isothermal Calorimetry (ITC). ITC experiments were
performed using a VP-ITC Microcalorimeter (Microcal Inc.).
A typical titration experiment involved 35 consecutive injec-
tions at 360 s intervals consisting of 3 mL injections of E. coli
0111:B4 LPS into the sample cell (cell volume 1.4119 mL)
containing the acylhomospermine compound, at 37 °C in Tris
buffer (pH 7.4, 50 mM). The titration cell was stirred continu-
ously at 310 rpm. Care was taken to ensure that both LPS
and ligand were dissolved in the same buffer, and appropriate
control experiments (LPS injected into buffer, BC injected into
buffer) were performed. The resulting data were then analyzed
using Microcal’s ITC data analysis package VP Viewer 2000,
which uses the scientific plotting software Origin 7 (Origin
Lab. Corp.).
Nitric Oxide Assay. Nitric oxide production was measured
as total nitrite in murine macrophage J774A.1 cells using the
Griess assay49 as described previously.16 J774A.1 cells were
plated at ∼2 × 106/mL in a volume of 40 µL/well, in 384-well,
flat-bottomed, cell culture treated microtiter plates and sub-
sequently stimulated with 100 ng/mL LPS. Concurrent to LPS
stimulation, serially diluted concentrations of test compounds

2598 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7
Miller et al.
be magnified because of increased osmotic fragility of the
erythrocytes as a consequence of depleted Na+-K+ ATPase
activity.39 Dilute erythrocyte suspensions were prepared by
diluting 1-week-old whole blood obtained by venipuncture from
healthy human volunteers 1:1000 in isotonic (0.9 g/100 mL)
saline solution to which was added graded doses of compound.
Absorptimetric determination of hemoglobin released from
such a dilute erythrocyte suspension was not reliable. The
samples were therefore examined with a Beckman-Coulter Vi-
Cell cell viability analyzer (Hialeah, FL). This instrument
implements an automated intravital trypan blue exclusion
method using real-time automated video microscopy. Measure-
ment parameters for erythrocytes were gated appropriately
on control erythrocytes to specify thresholds of cell recognition
and viability. Data on the total number of cells/mL and viable
cells/mL were collected through 50 captured images per sample
with a counting accuracy of (3%. To examine the effect of
plasma proteins on the surface activity, some of the experi-
ments were repeated in the presence of near-physiological
concentrations of human serum albumin. Because it became
apparent that the compounds were binding strongly to albu-
min, thereby resulting in an almost complete abrogation of
hemolytic activity, it was of interest to examine the compounds
under physiological conditions. The second method, conse-
quently, was designed to examine the effects of the compounds
on whole blood. A 100 µL portion of serially diluted compounds
was mixed with an equal volume of fresh, undiluted, EDTA-
anticoagulated human blood in a 96-well microplate using an
automated liquid handler. After incubation at 37 °C for 30 min,
the plates were centrifuged at 3000 rpm for 10 min, 80 µL of
supernatant was transferred to a fresh plate, and the amount
of free hemoglobin released into the supernatant was quanti-
fied using absorptimetry at 570 nm. In the latter assay,
melittin, a potently hemolytic R-helical bee venom peptide,41
was used as positive control.
This material is available free of charge via the Internet at
http://pubs.acs.org.
References
(1) Rietschel, E. T.; Kirikae, T.; Schade, F. U.; Mamat, U.; Schmidt,
G.; Loppnow, H.; Ulmer, A. J.; Za¨hringer, U.; Seydel, U.; Di
Padova, F.; et al, a. Bacterial endotoxin: Molecular relationships
of structure to activity and function. FASEB J. 1994, 8, 217-
225.
(2) Hurley, J. C. Antibiotic-induced release of endotoxin. A thera-
peutic paradox. Drug Saf. 1995, 12, 183-195.
(3) Gasche, Y.; Pittet, D.; Sutter, P. Outcome and prognostic factors
in bacteremic sepsis. In Clinical trials for treatment of sepsis.
Sibbald, W. J., Vincent, J. L., Eds.; Springer-Verlag: Berlin,
1995; pp 35-51.
(4) Centers for Diseases Control. Increases in national hospital
discharge survey rates for septicemia-United States, 1979-
1987. MMWR 1990, 39, 31-34.
(5) Ulevitch, R. J. Molecular mechanisms of innate immunity.
Immunol. Res. 2000, 21, 49-54.
(6) Ulevitch, R. J.; Tobias, P. Recognition of Gram-negative bacteria
and endotoxin by the innate immune system. Curr. Opin.
Immunol. 1999, 11, 19-23.
(7) Dinarello, C. A. Cytokines as mediators in the pathogenesis of
septic shock. Curr. Top. Microbiol. Immunol. 1996, 216, 133-
165.
(8) Michie, H. R.; Manogue, K. R.; Spriggs, D. R.; Revhaug, A.;
O’Dwyer, S.; Dinarello, C. A.; Cerami, A.; Wolff, S. M.; Wilmore,
D. W. Detection of circulating tumor necrosis factor after
endotoxin administration. N. Engl. J. Med. 1988, 318, 1481-
1486.
(9) Meyer, J.; Traber, D. L. Nitric oxide and endotoxin shock.
Cardiovasc. Res. 1992, 26, 558.
(10) Wright, C. E.; Rees, D. D.; Moncada, S. Protective and pathologi-
cal roles of nitric oxide in endotoxin shock. Cardiovasc. Res.
1992, 26, 48-57.
(11) Bone, R. C. The sepsis syndrome. Definition and general
approach to management. Clin. Chest Med. 1996, 17, 175-181.
(12) Raetz, C. R. H.; Whitfield, C. Lipopolysaccharide endotoxins.
Annu. Rev. Biochem. 2002, 71, 635-700.
Mouse Lethality Experiments. Female, outbred, 9-11-
week-old CF-1 mice (Charles River, Wilmington, MA) weighing
22-28 g were used as described elsewhere.16 The animals were
sensitized to the lethal effects of LPS by D-galactosamine.42,43,52
The lethal dose causing 100% mortality (LD100) for the batch
of LPS used (E. coli 0111:B4; procured from Sigma) was first
determined by administering D-galactosamine (800 mg/kg) and
LPS (0, 10, 20, 50, 100, 200 ng/mouse) as a single injection
intraperitoneally (ip) in freshly prepared saline to cohorts of
five animals in a volume of 0.2 mL. The expected dose-
response profile was observed in two independent experiments
with all five mice receiving 100 ng succumbing within 24 h,
establishing the LD100 dose to be 100 ng/mouse. In experiments
designed to test dose-response effects of the acylhomosper-
mines in affording protection against LPS-induced lethality,
mice received graded doses of compound diluted in saline ip
in one flank, immediately before a supralethal (200 ng) LPS
challenge, which was administered as a separate ip injection
into the other flank. Some animals also received test com-
pounds subcutaneously or iv to observe differences in the
degree and duration of protection; these experiments also
served to verify that LPS was being sequestered systemically,
and not just in situ, in the peritoneal cavity. In experiments
in which the temporal window of protection was to be
examined, a fixed dose of 200 µg/mouse of compound was
administered at various times, before or after supralethal (200
ng/mouse) LPS challenge. Lethality was determined at 24 h
after LPS challenge.
(13) David, S. A.; Mathan, V. I.; Balaram, P. Interactions of linear
dicationic molecules with lipid A: Structural requisites for
optimal binding affinity. J. Endotoxin. Res. 1995, 2, 325-336.
(14) David, S. A. Towards a rational development of anti-endotoxin
agents: Novel approaches to sequestration of bacterial endo-
toxins with small molecules (Invited Review). J. Mol. Recognit.
2001, 14, 370-387.
(15) Blagbrough, I. S.; Geall, A. J.; David, S. A. Lipopolyamines
incorportaing the teraamine spermine bound to an alkyl chain,
sequester bacterial lipopolysaccharide. Bioorg. Med. Chem. Lett.
2000, 10, 1959-1962.
(16) David, S. A.; Silverstein, R.; Amura, C. R.; Kielian, T.; Morrison,
D. C. Lipopolyamines: Novel antiendotoxin compounds that
reduce mortality in experimental sepsis caused by Gram-
negative bacteria. Antimicrob. Agents Chemother. 1999, 43, 912-
919.
(17) Blagbrough, I. S.; Geall, A. J. Practical synthesis of unsym-
metrical polyamine amides. Tetrahedron Lett. 1998, 39, 439-
442.
(18) Geall, A. J.; Blagbrough, I. S. Homologation of polyamines in
the synthesis of lipospermine conjugates and related lipoplexes.
Tetrahedron Lett. 1998, 39, 443-446.
(19) Jasys, V. J.; Kelbaugh, P. R.; Nason, D. M.; Phillips, D.; Rosnack,
K. J.; Saccomano, N. A.; Sroh, J. G.; Volkmann, R. A. Isolation,
structure elucidation, and synthesis of novel hydroxylamine-
containing polyamines from the venom of Agelenopsis aperta
spider. J. Am. Chem. Soc. 1990, 112, 6696-6704.
(20) David, S. A.; Bechtel, B.; Annaiah, C.; Mathan, V. I.; Balaram,
P. Interaction of cationic amphiphilic drugs with lipid A:
Implications for development of endotoxin antagonists. Biochim.
Biophys. Acta 1994, 1212, 167-175.
(21) David, S. A.; Mathan, V. I.; Balaram, P. Interactions of linear
dicationic molecules with lipid A: Structural features that
correspond to optimal binding affinity. J. Endotoxin Res. 1995,
2, 325-336.
(22) David, S. A.; Perez, L.; Infante, M. R. Sequestration of bacterial
lipopolysaccharide by bis(args) gemini compounds. Bioorg. Med.
Chem. Lett. 2002, 12, 357-360.
(23) Behr, J. P.; Demeneix, B.; Loeffler, J. P.; Perez-Mutul, J. Efficient
gene transfer into mammalian primary endocrine cells with
lipopolyamine-coated DNA. Proc. Natl. Acad. Sci. U.S.A. 1989,
86, 6982-6986.
Acknowledgment. This work was supported from
NIH grant 1R01 AI50107, and a First Award Grant
from P20 RR015563 from the COBRE Program of the
National Center for Research Resources and matching
support from the State of Kansas, and the University
of Kansas.
Supporting Information Available: 1H and 13C NMR
spectra of key intermediate compounds, elemental analyses
of all key target compounds, and additional experimental data.
(24) Behr, J. P. Gene transfer with synthetic cationic amphiphiles:
Prospects for gene therapy. Bioconjugate Chem. 1994, 5, 382-
389.

Acylhomospermine Endotoxin-Sequestering Compounds
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 7 2599
(25) Wood, S. J.; Miller, K. A.; David, S. A. Anti-endotoxin agents. 1.
Development of a fluorescent probe displacement method for the
rapid identification of lipopolysaccharide-binding agents. Comb.
Chem. High. Throughput. Screen. 2004, 7, 239-249.
(26) Bhattacharjya, S.; David, S. A.; Mathan, V. I.; Balaram, P.
Polymyxin B nonapeptide: Conformations in water and in the
lipopolysaccharide-bound state determined by two-dimensional
NMR and molecular dynamics. Biopolymers 1997, 41, 251-265.
(27) David, S. A.; Awasthi, S. K.; Wiese, A.; Ulmer, A. J.; Lindner,
B.; Brandenburg, K.; Seydel, U.; Rietschel, E. T.; Sonesson, A.;
Balaram, P. Characterization of the interactions of a polyca-
tionic, amphiphilic, terminally branched oligopeptide with lipid
A and lipopolysaccharide from the deep rough mutant of
Salmonella minnesota. J. Endotoxin Res. 1996, 3, 369-379.
(28) Weisz, A.; Oguchi, S.; Cicatiello, L.; Esumi, H. Dual mechanism
for the control of inducible-type NO synthase gene expression
in macrophages during activation by interferon-gamma and
bacterial lipopolysaccharide. Transcriptional and post-transcrip-
tional regulation. J. Biol. Chem. 1994, 269, 8324-8333.
(29) Remick, D. G.; Newcomb, D. E.; Friedland, J. S. Whole-blood
assays for cytokine production. In Septic shock. Methods and
protocols; Evans, T. J., Ed.; Humana Press: Totowa, NJ, 2000;
pp 101-114.
(30) Fennrich, S.; Fischer, M.; Hartung, T.; Lexa, P.; Montag-Lessing,
T.; Sonntag, H.-G.; Weigandt, M.; Wendel, A. Detection of
endotoxins and other pyrogens using human whole blood. Dev.
Biol. Standards 1999, 101, 131-139.
(31) David, S. A.; Balaram, P.; Mathan, V. I. Characterization of the
interaction of lipid A and lipopolysaccharide with human serum
albumin: Implications for an endotoxin-carrier function for
albumin. J. Endotoxin Res. 1995, 2, 99-106.
(32) He, X. M.; Carter, D. C. Atomic structure and chemistry of
human serum albumin. Nature 1992, 358, 209-215.
(33) Majerle, A.; Kidric, J.; Jerala, R. Enhancement of antibacterial
and lipopolysaccharide binding activities of a human lactoferrin
peptide fragment by the addition of acyl chain. J. Antimicrob.
Chemother. 2003, 51, 1159-1165.
(34) Bodor, N. Novel approaches to the design of safer drugs: Soft
drugs and site-specific chemical delivery systems. Adv. Drug Res.
1984, 13, 255-331.
(35) Bodor, N.; Buchwald, P. Soft drug design: General principles
and recent applications. Med. Res. Rev. 2000, 20, 58-101.
(36) Nabel, E. G.; Simari, R.; Yang, Z.; San, H.; Nabel, G. J. In vivo
gene transfer: A biological tool. Ann. N. Y. Acad. Sci 1997, 811,
289-292.
(37) Stephan, D. J.; Yang, Z. Y.; San, H.; Simari, R. D.; Wheeler, C.
J.; Felgner, P. L.; Gordon, D.; Nabel, G. J.; Nabel, E. G. A new
cationic liposome DNA complex enhances the efficiency of
arterial gene transfer in vivo. Hum. Gene Ther. 1996, 7, 1803-
1812.
(38) Menger, F. M.; Keiper, J. S. Gemini surfactants. Angew. Chem.,
Int. Ed. 2000, 39, 1906-1920.
(39) Nagini, S.; Selvam, S. Biochemical indicators of membrane
damage in the plasma and erythrocytes of rats fed the peroxi-
some proliferator di(2-ethylhexyl)phthalate. Med. Sci. Res. 1997,
25, 119-121.
(40) Bernard, E.; Faucon, J. F.; Dufourcq, J. Phase separations
induced by melittin in negatively charged phospholipid bilayers
as detected by fluorescence polarization and differential scanning
calorimetry. Biochim. Biophys. Acta 1982, 688, 152-162.
(41) David, S. A.; Mathan, V. I.; Balaram, P. Interaction of melittin
with endotoxic lipid A. Biochim. Biophys. Acta 1992, 1123, 269-
274.
(42) Freudenberg, M. A.; Galanos, C. Tumor necrosis factor alpha
mediates lethal activity of killed Gram-negative and Gram-
positive bacteria in d-galactosamine-treated mice. Infect. Immun.
1991, 59, 2110-2115.
(43) Tracey, K. J.; Cerami, A. Tumor necrosis factor, other cytokines
and disease. Annu. Rev. Cell Biol. 1993, 9, 317-343.
(44) Quezado, Z. M. N.; Banks, S. M.; Natanson, C. New strategies
for combatting sepsis: The magic bullets missed the mark... but
the search continues. Trends Biotechnol. 1995, 13, 56-63.
(45) Zeni, F.; Freeman, B.; Natanson, C. Anti-inflammatory therapies
to treat sepsis and septic shock: A reassessment. Crit. Care Med.
1997, 25, 1097-1100.
(46) Bone, R. C. Gram-negative sepsis: A dilemma of modern
medicine. Clin. Microbiol. Rev. 1993, 6, 57-68.
(47) Opal, S. M.; Yu, R. L. J. Antiendotoxin strategies for the
prevetion and treatment of septic shock. New approaches and
future directions. Drugs 1998, 55, 497-508.
(48) Morrison, D. C.; Jacobs, D. M. Binding of polymyxin B to the
lipid A portion of bacterial lipopolysaccharides. Immunochem-
istry 1976, 13, 813-818.
(49) Green, L. C.; Wagner, D. A.; Glogowski, J.; Skipper, P. L.;
Wishnok, J. S.; Tannenbaum, S. R. Analysis of nitrate, nitrite
and [15-N] nitrate in biological fluids. Anal. Biochem. 1982, 126,
131.
(50) Roehm, N. W.; Rodgers, G. H.; Hatfield, S. M.; Glasebrook, A.
L. An improved colorimetric assay for cell proliferation and
viability utilizing the tetrazolium salt XTT. J. Immunol. Methods
1991, 142, 257-265.
(51) Fennrich, S.; Fischer, M.; Hartung, T.; Lexa, P.; Montag-Lessing,
T.; Sonntag, H.-G.; Weigandt, M.; Wendel, A. Detection of
endotoxins and other pyrogens using human whole blood. Dev.
Biol. Standards 1999, 101, 131-139.
(52) Cook, E. B.; Stahl, J. L.; Lowe, L.; Chen, R.; Morgan, E.; Wilson,
J.; Varro, R.; Chan, A.; Graziano, F. M.; Barney, N. P. Simul-
taneous measurement of six cytokines in a single sample of
human tears using microparticle-based flow cytometry: Allergics
vs nonallergics. J. Immunol. Methods 2001, 254, 109-116.
(53) Funato, Y.; Baumhover, H.; Grantham-Wright, D.; Wilson, J.;
Ernst, D.; Sepulveda, H. Simultaneous measurement of six
human cytokines using the Cytometric Bead Array System, a
multiparameter immunoassay system for flow cytometry. Cy-
tometry Res. 2002, 12, 93-103.
(54) Ross, B. P.; Braddy, A. C.; McGeary, R. P.; Blanchfield, J. T.;
Prozkai, L.; Toth, I. Micellar aggregation and membrane par-
titioning of bile salts, fatty acids, sodium dodecyl sulfate, and
sugar-conjugated fatty acids: Correlation with hemolytic potency
and implications for drug delivery. Mol. Pharm. 2004, 1, 233-
245.
(55) Fainerman, V. B.; Miller, R. Maximum bubble pressure
tensiometrysan analysis of experimental constraints. Adv.
Colloids Interface Sci. 2004, 108-109, 287-301.
(56) Wood, S. J.; Miller, K. A.; David, S. A. Anti-endotoxin agents. 2.
Pilot high-throughput screening for novel lipopolysaccharide-
recognizing motifs in small molecules. Comb. Chem. High
Throughput Screen. 2004, 7, 733-743.
JM049449J