Facilitated phospholipid translocation in vesicles and nucleated cells using synthetic small molecule scramblases

Article abstract of DOI:10.1016/j.bmc.2008.11.011

A series of 16 synthetic scramblase candidates were prepared from a tris(aminoethyl)amine (TREN) scaffold and evaluated for ability to facilitate translocation of fluorescent phospholipid probes across vesicle membranes and endogenous phosphatidylserine a

Full text of DOI:10.1016/j.bmc.2008.11.011

Bioorganic & Medicinal Chemistry 17 (2009) 141–148
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Facilitated phospholipid translocation in vesicles and nucleated cells using
synthetic small molecule scramblases
*
Kristy M. DiVittorio, Frank T. Hofmann, James R. Johnson, Lica Abu-Esba, Bradley D. Smith
Department of Chemistry and Biochemistry and the Walther Cancer Research Center, University of Notre Dame, Notre Dame, IN 46556, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of 16 synthetic scramblase candidates were prepared from a tris(aminoethyl)amine (TREN) scaf-
fold and evaluated for ability to facilitate translocation of fluorescent phospholipid probes across vesicle
membranes and endogenous phosphatidylserine across the plasma membrane of nucleated cells. More
than half of the compounds were found to greatly accelerate phospholipid translocation in vesicles. How-
ever, they were generally unable to induce large increases in the amount of phosphatidylserine on the
surface of nucleated mammalian cells, which contrasts with previous results using erythrocytes. Fluores-
cence microscopy showed that the synthetic scramblases are rapidly trafficked out of the cell plasma
membrane and into the membranes of internal organelles. Future molecular designs of synthetic scramb-
lases should focus on structures that are more amphiphilic, a structural feature that is expected to
increase plasma membrane residence time.
Received 4 March 2008
Revised 6 November 2008
Accepted 7 November 2008
Available online 12 November 2008
Keywords:
Phosphatidylserine
Translocation
Molecular recognition
Flip-flop
Ó 2008 Elsevier Ltd. All rights reserved.
Membrane transport
Membrane scramblase
1. Introduction
brane structure and perhaps eventually as pharmaceutical leads.
We have reported previously that synthetic scramblases based on
The translocation of phospholipids across synthetic bilayer
membranes is known to be a very slow process with half-lives
on the order of hours to days; in comparison, lateral diffusion with-
in a phospholipid monolayer is much more rapid (Fig. 1).1,2 If the
membrane is not highly curved then the distribution is simply a
scrambled equilibrium. In contrast, most biological membranes
have a highly asymmetric distribution of phospholipids.3 In the
case of mammalian plasma membranes, the outer monolayer is en-
riched in zwitterionic choline-based phospholipids, such as phos-
phatidylcholine (PC), while the inner monolayer is enriched in
aminophospholipids, such as anionic phosphatidylserine (PS)
(Scheme 1).4 This asymmetric phospholipid distribution is main-
tained by the cumulative action of a family of endogenous mem-
brane transporters called scramblases, flippases and floppases.5
The asymmetric distribution is crucial for normal cellular activity,
but it becomes scrambled during certain pathological disorders
and also the early stages of cell death.6 The consequent exposure
of PS on the cell surface is signal for cell clearance by the innate im-
mune system.7
a TREN (tris(2-aminoethyl)amine) scaffold can scramble the
phospholipid distribution in vesicles and erythrocytes.8 Mechanis-
tic studies indicate that these TREN scramblases can associate with
the phospholipid head group and form lipophilic, hydrogen-
bonded phospholipid:scramblase complexes that diffuse through
the membrane (Scheme 1). Initially, we examined simple tris(sul-
fonamide) and tris(amide) derivatives, but a screening study found
that the urea-based scramblases 1 and 2 (Scheme 2) cannot only
promote translocation of fluorescently-labeled phospholipid
probes across vesicle membranes, but also induce surface exposure
of endogenous PS in erythrocytes.8f
These promising results motivated us to pursue an expanded
study of TREN scramblases. We report here the preparation of 16
For several years, our group has investigated the use of small
synthetic molecules to facilitate translocation of phospholipids
across bilayer membranes.8,9 The practical goal is to produce syn-
thetic scramblases for use as research tools to remodel cell mem-
* Corresponding author. Tel.: +1 574 631 8632; fax: +1 574 631 6652.
E-mail address: smith.115@nd.edu (B.D. Smith).
Figure 1.
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.11.011

142
K. M. DiVittorio et al. / Bioorg. Med. Chem. 17 (2009) 141–148
candidate compounds and their evaluation for facilitated translo-
cation activity in vesicles. The most active compounds were subse-
quently tested in live mammalian cells, to see if they increased the
amount of PS exposed on the cell surface. In contrast to our previ-
ous success with erythrocytes we find that the TREN scramblases
are less effective with nucleated cells. Microscopic imaging of a
fluorescent analog suggests that the scramblases are rapidly traf-
ficked away from the plasma membrane and distributed through-
out the organelle membranes inside the cell. It appears that
residence time in the plasma membrane is a key factor controlling
synthetic scramblase activity.
2. Results
2.1. Synthetic chemistry
As shown in Scheme 2, the central intermediate, TREN disulf-
onamide (22) was synthesized in three steps. It is worth noting
that improved reaction conditions were found to produce the
mono-N-Boc-protected TREN (20) in 92% yield. From 22 the urea
Scheme 1. (Top) Phospholipid structures, (Bottom) Scramblase hydrogen-bonded
to anionic residues within the phospholipid (PL) head group.
Scheme 2. Structures and syntheses of TREN scramblases.

K. M. DiVittorio et al. / Bioorg. Med. Chem. 17 (2009) 141–148
143
Table 1
tive PS scramblases were identified (Table 1). In most cases,
scramblase induced acceleration of NBD–PC was slightly greater
than NBD–PS.
Translocation half-lives for NBD–PC and NBD–PS in vesicles
Scramblase (12.5
l
M)
t1/2 (min)a
NBD–PC (2 mol%)
NBD–PS (2 mol%)
2.3. Effect of TREN scramblases on nucleated cells
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
60
8
0.5
90
0.5
20
0.5
0.5
0.5
20
0.5
0.5
>30
0.5
>60
>60
60
12
2
A major goal of the study was to evaluate if the TREN scramb-
lases could alter the PS distribution across the plasma membrane
of nucleated mammalian cells. Since PS exposure is a signature of
most if not all cell death processes, it was important to first deter-
mine if the scramblases induced any cell toxicity. Thus, they were
screened for effects on the viability of Jurkat, PC3 and LNCaP cells
using an MTT assay (Table 2). Briefly, 106 cells were plated out in
48-well plates and left to adhere and multiply for 48 h. The media
were exchanged with fresh media that were supplemented with
MTT. In healthy cells, the MTT is converted by mitochondrial
reductase enzymes into the purple formazan product. None of
the urea-based TREN scramblases 1–13 had any measurable effect
>90
0.25
60
a
on mitochondrial reductase activity (LD50 > 100
three scramblases 14, 15 and 16 exhibited modest toxicity
(LD50 > 15 M). Overall the TREN scramblases are not inherently
lM); however, the
Translocation half-life in 7:3 POPC/cholesterol vesicles, TES buffer (5 mM TES,
100 mM NaCl, pH 7.4; 25 °C); error 33%.
l
toxic to cells.
derivatives 1–13 and thiourea 14 were prepared in one step pro-
The next step was to determine if TREN scramblases can alter
the endogenous transmembrane distribution of PS in nucleated
cells. The exposed PS was detected by staining with the dye-la-
beled protein Annexin V-FITC.8f Briefly, separate samples of PC3
cells (1 ꢀ 105 cells/mL) in TES buffer were treated with TREN
cesses. Coupling of 22 with L-phenylalanine gave the amide 15,
and reductive benzylation of 22 gave 16. The fluorescent derivative
17 with an NBD (7-nitrobenz-2-oxa-1,3-diazol-4-yl) group was
prepared by aromatic substitution with commercially available
NBD–Cl.
scramblase (100 lM) and incubated at 37 °C for 2 h. Flow cytome-
try was used to identify cells that were stained by Annexin-FITC
and simultaneous 7AAD co-staining of the nucleus was used to
eliminate necrotic cells that had lost plasma membrane integrity.
Previously, we observed that treatment of erythrocytes with
scramblase 2 raised the fraction of Annexin-stained cells from 1%
to 34%. In comparison, the fraction of PC3 cells that were stained
by Annexin-FITC after scramblase treatment was low. The best re-
2.2. Phospholipid translocation in vesicles
The well-established NBD/dithionite quenching assay was used
to measure the translocation half-lives for fluorescent phospho-
lipid probes containing an NBD chromophore in one of its acyl
chains.8a,10,11 Exo-labeled vesicles were prepared by addition of a
small aliquot of NBD–PC or PBD–PS (2 mol.%) to a dispersion of
sult was obtained with scramblase 5 (100 lM) which increased the
unilamellar 7:3 POPC/cholesterol vesicles (25 lM). The NBD–lipid
fraction of Annexin-stained cells from a background value of 2–10%
(Fig. 2). Three other TREN scramblases were tested (1, 2 and 9) and
in each case the fraction of Annexin-stained cells was <5% of the
cell population (data not shown). The same outcome was observed
with PC3 cells that were pretreated with vanadate, a well-known
inhibitor of the ATP-dependent flippases.12
rapidly inserts into the vesicles. Bilayer membranes are effectively
impermeable to sodium dithionite; therefore, only NBD–phospho-
lipid located in the membrane outer leaflet is chemically quenched
upon dithionite addition. The background rate of phospholipid
translocation across vesicle membranes is many hours to days
and at final equilibrium there is 60% of the probe in the outer leaf-
let, reflecting the difference in surface area of inner and outer leaf-
lets. Thus, the translocation half-life is defined operationally as the
time required for 20% of the exo-NBD–phospholipid to be translo-
cated to the inner leaflet. Listed in Table 1 are the translocation
half-lives for NBD–PC observed upon treatment of the vesicles with
TREN scramblases 1–16. In general, the scramblases with with-
drawing groups on the 4-aryl urea ring produced shorter transloca-
tion half-lives. This agrees with earlier studies that withdrawing
groups enhance association with phospholipid head group due to
increased hydrogen bonding.8 Six of the scramblases were also
evaluated in a NBD–PS translocation assay and some highly effec-
2.4. Partitioning of fluorescent TREN probe into vesicles and
cells
To help understand why the TREN scramblases are unable to
greatly alter endogenous PS surface levels, a fluorescent NBD-ver-
sion, 17, was prepared and used for partitioning studies. A dithio-
nite protection study was employed to measure partitioning into
Table 2
Toxicity of TREN scramblasesa
Scramblase
LD50 (lM)
Jurkat cells
PC3 cells
LNCaP cells
14
15
16
15
25
25
25
15
25
15
40
40
Figure 2. Flow cytometry analysis of PC3 cells stained by Annexin V-FITC. (A)
a
As determined by the MTT assay; 10
M.
lM. The IC50 for compounds 1–16 was
Untreated cells. (B) Cells treated with scramblase 5 (100
lM). The data shown is a
>100
l
representative histogram from three independent experiments.

144
K. M. DiVittorio et al. / Bioorg. Med. Chem. 17 (2009) 141–148
(Fig. 3). The chemical quenching involved addition of sodium
dithionite at 50 s to quench the fluorescence of surface exposed
17 and subsequent vesicle lysis at 150 s with Triton X-100 allowed
complete quenching. With both vesicle compositions, there was
substantial protection of the probe from dithionite quenching even
0.5 min after the probe was added, indicating that the weakly lipo-
philic probe 17 equilibrates very rapidly across the vesicle
membrane.
This picture of rapid membrane transport was confirmed with a
fluorescence microscopy study of 17 partitioning into mammalian
cells. Briefly, separate samples of Jurkat cells were incubated for
30 min with 17 (12.5 lM) and either of the targeted fluorescent
probes, ER TrackerTM or Mito TrackerTM. Cells were then visualized
using epi-fluorescence microscopy. The large amount of co-local-
ized staining clearly indicates that probe 17 is localized primarily
in the ER and the mitochondria and that relatively little is in the
plasma membrane (Fig. 4). These membrane partitioning results
indicate that the TREN scramblases are trafficked rapidly away
from the plasma membrane of mammalian cells and they are
sequestered in the relatively large volume of intracellular organelle
membranes.13
3. Discussion
As expected from previous work,8 most of the urea-based TREN
scramblases with electron withdrawing groups on the urea aryl
ring are able to greatly accelerate the translocation of fluorescent
PC–NBD and PS–NBD probes across vesicle membranes. Most
likely, the electron withdrawing substituents enhance association
with phospholipid head group due to increased hydrogen bonding.
Although the TREN scramblases can greatly enhance the fraction of
erythrocytes that expose PS on their exterior surface, they are sub-
stantially less efficient with nucleated mammalian cells. In both
cases, the scramblases have to compete with the endogenous flip-
pase enzymes in the cell plasma membranes, and it may be that
flippase activity is inherently higher in nucleated cells. However,
evidence against this explanation is the fact that pretreatment of
the PC3 cells with vanadate, a well-known inhibitor of the ATP-
dependent flippases,12 does not help a TREN scamblase to increase
PS exposure on the cell surface. A more likely explanation for the
lack of scramblase activity in nucleated cells comes from the
microscopy results, which indicate that fluorescent TREN scramb-
lases are trafficked rapidly away from the plasma membrane and
Figure 3. Chemical quenching of 17, protection by vesicles. At t = 0, vesicles
composed of 7:3 POPC/cholesterol (A) or 3.5:3.5:3 POPC/POPS/cholesterol (B)
vesicles (25
lM) in TES buffer (5 mM TES, 100 mM NaCl, pH 7.4) were treated with
17 (12.5 M). After 0.5 min ( ) and 20 min ( ), samples (3 mL) were withdrawn
l
and assayed by adding sodium dithionite and Triton X-100 at times of 50 and 150 s,
respectively. The data shown are a representative plot from three independent
experiments. The error for each point is 5%.
vesicles composed of 7:3 POPC/cholesterol (mimic of zwitterionic
plasma membrane outer leaflet) and 3.5:3.5:3 POPC/POPS/choles-
terol (mimic of anionic plasma membrane inner leaflet). Briefly,
probe 17 was added to a solution of vesicles and the amount of
probe that had moved to the membrane interior surface (and thus
was protected from chemical quenching by the subsequent addi-
tion of sodium dithionite) was measured after 0.5 and 20 min
TM
Figure 4. Fluorescence microscopy images of two separate Jurkat cells incubated with 17 (12.5 lM) and the following co-stains. Top row: (A) 17, (B) ER Tracker , (C) color
TM
overlay. Bottom row: (D) 17, (E) Mito Tracker , (F) color overlay.

K. M. DiVittorio et al. / Bioorg. Med. Chem. 17 (2009) 141–148
145
into the membranes of internal organelles such as the ER and mito-
chondria. Erythrocytes, of course, do not have these internal mem-
branes. A shortened residence time in the plasma membrane
would diminish the inherent ability of the TREN scramblases to
promote PS flip-flop.
7.39 (s, 1H); 7.74 (d, J = 8.2 Hz, 4H); 13C NMR (CDCl3, TMS,
150 MHz) d 21.5, 37.7, 41.1, 53.7, 54.1, 55.5, 114.1, 121.7, 127.0,
129.8, 132.3, 136.6, 143.4, 155.5, 156.9; TLC (9:1 EtOAc/acetone)
Rf = 0.49.
Urea (5). According to the general procedure product 5 was ob-
tained as a colorless oil (84.4 mg, 58%); 1H NMR (CDCl3, TMS,
300 MHz) d 2.35 (s, 6H), 2.50 (m, 6H), 3.24 (m, 2H), 6.07 (t,
J = 4.8 Hz, 1H), 6.14 (unresolved t, 2H), 7.19 (d, J = 8.2 Hz), 7.44
(d, J = 8.8 Hz, 2H), 7.53 (d, J = 8.7 Hz, 2H), 7.72 (d, J = 8.2 Hz, 4H),
7.77 (s, 1H); 13C NMR (CDCl3, TMS, 75 MHz) d 21.4, 37.7, 41.1,
4. Conclusion
Although many of the TREN scramblases reported in this study
can accelerate the translocation of phospholipids in model mem-
brane systems, they are unable to significantly increase the expo-
sure of PS on the surface of nucleated cells because they are
rapidly trafficked out of the plasma membrane and into the mem-
branes of internal organelles. Future designs should focus on
scramblase structures that are more amphiphilic, a structural fea-
ture that is expected to increase plasma membrane residence
time.14 This structural strategy should also be considered when
designing other synthetic membrane transport systems that re-
quire the transporter to remain in the plasma membrane of nucle-
ated cells for extended periods (e.g., synthetic chloride transporters
for treatment of cystic fibrosis).15
2
1
53.5, 118.0, 123.54 (q, JCF = 32.6 Hz), 124.4 (q, JCF = 271.2 Hz),
3
126.0 (q, JCF = 3.51), 126.9, 129.8, 136.4, 143.0, 143.7, 155.6; TLC
(EtOAc) Rf = 0.46.
Urea (6). According to the general procedure product 6 was ob-
tained as a colorless oil (43.7 mg, 66%); 1H NMR (CDCl3, TMS,
300 MHz) d 2.27 (s, 3H), 2.34 (s, 6H), 2.43 (br m, 6H), 2.90 (br m,
4H), 3.19 (br m, 2H), 5.96 (unresolved t, 1H), 6.33 (unresolved t,
2H), 7.02 (d, J = 8.3 Hz, 2H), 7.19 (d, J = 8.3 Hz, 4H), 7.27 (d,
J = 8.3 Hz, 2H), 7.48 (s, 1H), 7.74 (d, J = 8.2 Hz, 4H); 13C NMR (CDCl3,
TMS, 75 MHz) d 20.7, 21.5, 37.6, 41.1, 53.8, 54.1, 119.6, 127.0,
129.4, 129.8, 131.9, 136.5, 136.8, 143.4, 156.6; TLC (EtOAc)
Rf = 0.50.
Urea (7). According to the general procedure product 7 was ob-
tained as a colorless oil (39.7 mg, 55%); 1H NMR (CDCl3, TMS,
300 MHz) d 2.36 (s, 6H), 2.49 (br m, 6H), 2.95 (br m, 4H), 3.21
(br m, 2H), 6.07 (unresolved t, 1H), 6.28 (unresolved t, 2H), 7.19
(d, J = 8.0 Hz, 4H), 7.30 (m, 4H), 7.67 (br s, 1H), 7.72 (d, J = 8.3 Hz,
4H); 13C NMR (CDCl3, TMS, 75 MHz) d 21.5, 37.5, 41.0, 53.6, 53.7,
114.4, 120.4, 126.9, 129.8, 131.6, 136.4, 138.8, 143.6, 156.0; TLC
(EtOAc) Rf = 0.57.
Urea (8). According to the general procedure product 8 was ob-
tained as a colorless oil (30.4 mg, 51%); 1H NMR (CDCl3, TMS,
300 MHz) d 2.35 (s, 6H), 2.49 (br m, 6H), 2.95 (br m, 4H), 3.22
(br m, 2H), 6.02 (unresolved t, 1H), 6.33 (unresolved t, 2H), 6.90
(d, J = 9.0 Hz, 2H), 6.94 (m, 2H), 7.04 (t, J = 7.4 Hz, 1H), 7.22 (d,
J = 8.1 Hz, 4H), 7.28 (m, 2H), 7.38 (d, J = 9.0 Hz, 2H), 7.57 (br s,
1H), 7.75 (d, J = 8.3 Hz, 4H); 13C NMR (CDCl3, TMS, 75 MHz) d
21.5, 37.6, 41.0, 53.7, 54.1, 117.9, 119.9, 120.9, 122.6, 127.0,
129.6, 129.8, 135.3, 136.5, 143.5, 151.8, 156.5, 158.0; TLC (EtOAc)
Rf = 0.43.
Urea (9). According to the general procedure product 9 was ob-
tained as a white solid (58%); 1H NMR (CDCl3, TMS, 300 MHz) d
1.21 (d, J = 6.86 Hz, 6H), 2.34 (s, 6H), 2.51 (br m, 6H), 2.84 (sep,
J = 6.9 Hz, 1H), 2.95 (br m, 4H), 3.21 (br m, 2H), 6.11 (unresolved
t, 1H), 6.48 (unresolved t, 2H), 7.08 (d, J = 8.5 Hz, 2H), 7.19 (d,
J = 8.1 Hz, 4H), 7.33 (d, J = 8.5 Hz, 2H), 7.67 (s, 1H), 7.76 (d,
J = 8.3 Hz, 4H); 13C NMR (CDCl3, TMS, 75 MHz) d 21.5, 24.1, 33.5,
37.4, 40.7, 40.8, 54.0, 119.6, 126.7, 127.0, 129.8, 136.5, 137.0,
143.1, 143.5, 156.7; TLC (EtOAc) Rf = 0.62; HPLC/MS RT = 9.55 min,
exact mass calcd for C30H41N5O5S2 [M+H]+ 616.26: found: 616.4.
Urea (10). According to the general procedure product 10 was
obtained as a colorless solid (38.5 mg, 76%); 1H NMR (CDCl3,
TMS, 300 MHz) d 2.37 (s, 6H), 2.55 (br m, 6H), 3.00 (br m, 4H),
3.25 (br m, 2H), 6.19 (br m, 3H), 7.22 (d, J = 8.0 Hz, 4H), 7.46 (d,
J = 8.9 Hz, 2H), 7.55 (d, J = 8.9 Hz, 2H), 7.72 (d, J = 8.3 Hz, 4H),
8.00 (br s, 1H); 13C NMR (CDCl3, TMS, 75 MHz) d 21.5, 37.5, 41.0,
53.4, 104.2, 118.2, 119.5, 126.9, 129.9, 133.0, 136.3, 143.8, 144.2,
155.3; TLC (EtOAc) Rf = 0.49.
5. Experimental
5.1. Synthetic chemistry
5.1.1. General procedure for 1–14
A solution of the p-substituted phenylisocyanate (1.1 equiv) in
dichloromethane (60 mL per mmol 22) was added drop wise to a
stirred solution of TREN disulfonamide (22) in dichloromethane
(120 mL per mmol 22) during 20 min at 0 °C. The reaction mixture
was allowed to stir for 16 h at room temperature. After removal of
the solvent in vacuo the crude product was purified by column
chromatography on silica gel using ethyl acetate/hexanes gradient
elution (60–100 vol% EtOAc).
Urea (1). According to the general procedure product 1 was ob-
tained as a colorless solid (34.5 mg, 62%); 1H NMR (CDCl3, TMS,
300 MHz) d 2.34 (s, 6H), 2.43 (br m, 6H), 2.90 (br m, 4H), 3.21
(br m, 2H), 6.00 (unresolved t, 1H), 6.29 (unresolved t, 2H), 6.96
(t, J = 7.4 Hz, 1H), 7.20 (d, J = 8.0 Hz, 4H), 7.23 (m, 2H), 7.40 (d,
J = 7.6 Hz, 2H), 7.57 (br s, 1H), 7.74 (d, J = 8.3 Hz, 4H); 13C NMR
(CDCl3, TMS, 75 MHz) d 21.5, 37.5, 41.1, 53.7, 119.2, 122.4, 127.0,
128.9, 129.8, 136.5, 139.6, 143.5, 156.3; TLC (9:1 EtOAc/acetone)
Rf = 0.47.
Urea (2). According to the general procedure product 2 was ob-
tained as yellow solid (38.4 mg, 34%); 1H NMR (CDCl3, TMS,
300 MHz) d 2.36 (s, 6H), 2.56 (br m, 6H), 3.01 (br m, 4H), 3.26
(br m, 2H), 6.19 (unresolved t, 2H), 6.26 (unresolved t, 1H), 7.22
(d, J = 8.2 Hz, 4H), 7.57 (d, J = 9.2 Hz, 2H), 7.72 (d, J = 8.3 Hz, 4H),
8.06 (d, J = 9.2 Hz, 2H), 8.18 (s, 1H); 13C NMR (CDCl3, TMS,
75 MHz) d 21.4, 37.5, 40.9, 53.3, 117.4, 125.0, 126.8, 129.8, 136.2,
141.6, 143.8, 146.2, 155.1; TLC (EtOAc) Rf = 0.57.
Urea (3). According to the general procedure product 3 was ob-
tained as a colorless solid (66.0 mg, 49%); 1H NMR (CDCl3, TMS,
600 MHz) d 2.36 (s, 6H), 2.45 (m, 6H), 2.93 (m, 4H), 3.20 (m, 2H),
5.99 (t, J = 5.0 Hz), 6.22 (unresolved t, 4H), 7.15 (d, J = 8.8 Hz, 2H),
7.20 (d, J = 7.9 Hz, 4H), 7.35 (d, J = 8.8 Hz, 2H), 7.54 (s, 1H), 7.72
(d, J = 8.3 Hz, 4H); 13C NMR (CDCl3, TMS, 150 MHz) d 21.5, 37.6,
41.2, 53.6, 53.6, 120.1, 126.9, 126.9, 128.7, 129.8, 136.4, 138.3,
143.6, 155.9; TLC (9:1 EtOAc/acetone) Rf = 0.51.
Urea (11). According to the general procedure product 11 was
obtained as a colorless oil (44.4 mg, 71%); 1H NMR (CDCl3, TMS,
300 MHz) d 1.27 (s, 9H), 2.33 (s, 6H), 2.41 (br m, 6H), 2.89 (br m,
4H), 3.19 (br m, 2H), 5.98 (unresolved t, 1H), 6.36 (unresolved t,
2H), 7.18 (d, J = 8.0 Hz, 4H), 7.23 (d, J = 8.9 Hz, 2H), 7.31 (d,
J = 8.8 Hz, 2H), 7.50 (s, 1H), 7.75 (d, J = 8.3 Hz, 4H); 13C NMR (CDCl3,
TMS, 75 MHz) d 21.5, 31.4, 34.2, 37.6, 41.1, 53.9, 54.2, 119.2, 125.6,
Urea (4). According to the general procedure product 4 was ob-
tained as a colorless solid (85.6 mg, 66%); 1H NMR (CDCl3, TMS,
600 MHz) d 2.34 (s, 6H), 2.39 (m, 6H), 2.87 (m, 4H), 3.17 (m, 2H),
3.74 (s, 3H), 5.91 (unresolved t, 1H), 6.34 (unresolved t, 2H), 6.78
(d, J = 8.9 Hz, 2H), 7.20 (d, J = 8.1 Hz, 4H), 7.28 (d, J = 9.0 Hz, 2H);

146
K. M. DiVittorio et al. / Bioorg. Med. Chem. 17 (2009) 141–148
127.0, 129.8, 136.5, 136.8, 143.4, 145.3, 156.5; TLC (EtOAc)
Rf = 0.57.
NBD disulfonamide (17). TREN disulfonamide 22 (35.1 mg,
0.0772 mmol) was dissolved in MeOH (5 mL) under argon at 0 °C.
K2HCO3 (32.0 mg, 3.0 equiv) was added. The resulting suspension
was allowed to stir for 5 min at 0 °C and for additional 10 min at
room temperature. After that a solution of NBD–Cl (15.4 mg,
1.0 equiv) in MeOH (5 mL) was added drop wise over 5 min. The
reaction was stirred at rt for 24 h after which time the solvent
was removed in vacuo. The residue was extracted with ethyl ace-
tate. After that the solvent was removed in vacuo and the crude
product was purified by column chromatography on Al2O3 using
0.5–1.0% MeOH:ethyl acetate gradient elution. The product 17
was obtained as a orange-green oil (19.6 mg, 41%); 1H NMR (CDCl3,
TMS, 300 MHz) d 2.36 (s, 6H), 2.52 (t, J = 6.1 Hz, 4H), 2.69 (t,
J = 6.3 Hz, 2H), 2.84 (q, J = 6.1 Hz, 4H), 3.43 (br s, 2H), 5.59 (unre-
solved t, 2H), 6.25 (br m, 1H), 7.30 (d, J = 8.0 Hz, 5H), 7.67 (d,
J = 8.3 Hz, 4H), 8.46 (d, J = 8.8 Hz, 1H); 13C NMR (CDCl3, TMS,
75 MHz) d 21.4, 41.9, 42.3, 52.6, 54.2, 99.9, 123.5, 127.7, 130.6,
138.2, 144.4, 145.5; TLC (EtOAc, SiO2) Rf = 0.72.
Mono-N-Boc TREN (20). To a stirred solution of tris-(2-amino-
ethyl)amine (19) (5.1 mL, 35 mmol) in dioxane (30 mL) under
nitrogen a solution of di-tert-butyl-dicarbonate (1.2 mL, 5.5 mmol)
in dioxane (30 mL) was added over 1 h at rt. The reaction mixture
was stirred for 17 h. The solvent was removed in vacuo and the res-
idue was dissolved in water (10 mL). The aqueous solution was ex-
tracted with dichloromethane (6ꢀ 15 mL). The organic phases
were combined. The removal of the solvent in vacuo gave the prod-
uct 20 (1.247 g, 92%) as a viscous oil; 1H NMR (CDCl3, 300 MHz) d
1.45 (s, 9H), 1.69 (s, 4H), 2.54 (t, J = 6.0 Hz, 4H), 2.57 (t, J = 5.8 Hz,
2H), 2.78 (t, J = 6.0 Hz, 4H), 3.20 (m, 2H), 5.32 (br s, 1H); TLC
(10:2:1 CH3CN/H2O/NH4OH) Rf = 0.20.
Urea (12). According to the general procedure product 12 was
obtained as a colorless oil (35.9 mg, 57%); 1H NMR (CDCl3, TMS,
300 MHz) d 2.34 (s, 6H), 2.52 (br m, 6H), 2.96 (br m, 4H), 3.22
(br m, 2H), 6.17 (unresolved t, 1H), 6.37 (unresolved t, 2H), 7.04
(d, J = 8.4 Hz, 2H), 7.19 (d, J = 8.1 Hz, 4H), 7.43 (d, J = 9.0 Hz, 2H),
7.73 (d, J = 8.3 Hz, 4H), 7.83 (s, 1H); 13C NMR (CDCl3, TMS,
1
75 MHz) d 21.4, 37.4, 40.9, 53.7, 54.0, 120.5 (q, JCF = 256.1 Hz),
119.7, 121.6, 127.0, 129.8, 136.4, 138.5, 143.7, 156.1; TLC (EtOAc)
Rf = 0.62; HPLC/MS RT = 9.42 min, exact mass calcd. for
C28H34F3N5O6S2 [M+H]+ 658.20: found: 658.3.
Urea (13). According to the general procedure product 13 was
obtained as a beige solid (42.4 mg, 62%); 1H NMR (CDCl3, TMS,
300 MHz) d 2.33 (s, 6H), 2.44 (br m, 6H), 2.91 (br m, 4H), 3.18
(br m, 2H), 4.99 (s, 2H), 6.00 (unresolved t, 1H), 6.44 (unresolved
t, 2H), 6.85 (d, J = 8.9 Hz, 2H), 7.20 (d, J = 8.1 Hz, 4H), 7.35 (m,
7H), 7.51 (s, 1H), 7.74 (d, J = 8.2 Hz, 4H); 13C NMR (CDCl3, TMS,
75 MHz) d 21.5, 29.7, 37.2, 40.7, 54.0, 70.2, 115.2, 121.5, 127.0,
127.5, 127.9, 128.5, 129.8, 132.7, 136.5, 137.1, 143.4, 154.6,
157.1; TLC (EtOAc) Rf = 0.54.
Thiourea (14). According to the general procedure product 14
was obtained as a white solid (23.2 mg, 67%); 1H NMR (CDCl3,
TMS, 300 MHz) d 2.39 (s, 6H), 2.60 (br m, 6H), 2.98 (br m, 4H),
3.62 (br m, 2H), 6.09 (unresolved t, 2H), 7.14 (unresolved t, 1H),
7.25 (m, 6H), 7.44 (d, J = 8.7 Hz, 2H), 7.74 (d, J = 8.3 Hz, 4H), 8.42
13
(br s, 1H);
NMR (CDCl3, TMS, 75 MHz) d 21.6, 41.0, 42.3, 52.9,
53.7, 125.6, 127.0, 128.9, 129.9, 130.6, 136.4, 137.1, 143.8, 181.1;
TLC (7:3 EtOAc/acetone) Rf = 0.46.
Phenylalanine disulfonamide (15). EDC (27.8 mg, 1.5 equiv), HOBt
(29.6 mg, 2.0 equiv), triethylamine (19.6 mg, 2.0 equiv) and N-Boc-
protected phenylalanine (30.8 mg, 1.2 equiv) were dissolved in
DMF (5 mL) and stirred at room temperature for 1 h. Next, com-
pound 22 (44.0 mg, 0.977 mmol) was dissolved in DMF (2 mL)
and added drop wise to the solution. The mixture was stirred for
16 h after which the solvent was removed in vacuo. Purification
of the crude product took place by column chromatography on
SiO2 using ethyl acetate as the eluent and yielding a colorless oil
which was dissolved in dichloromethane (10 mL). Dry HCl gas
was slowly bubbled through the stirred solution for 30 min. The
reaction mixture became cloudy and was stirred for additional
30 min and the solvent was removed in vacuo. The product 15
was obtained as a colorless oil (19.2 mg, 90%); 1H NMR (CD3OD,
300 MHz) d 2.41 (s, 6H), 3.08 (dd, J = 8.3 Hz, J = 13.9 Hz, 1H), 3.25
(m, 5H), 3.47 (br m, 7H), 3.71 (br m, 1H), 4.18 (m, 1H), 7.31 (m,
5H), 7.40 (d, J = 8.0 Hz, 4H), 7.78 (d, J = 8.3 Hz, 4H); 13C NMR
(CD3OD, 75 MHz) d 21.5, 36.0, 38.4, 38.8, 54.4, 55.2, 56.0, 128.4,
129.0, 130.3, 130.6, 131.1, 135.6, 137.4, 145.6, 171.7.
Benzylated disulfonamide (16). Under argon atmosphere, benzal-
dehyde (8.6 mg, 0.081 mmol) and 22 (40.5 mg, 1.1 equiv) were dis-
solved in 3 mL dichloroethane at room temperature. Sodium
triacetoxyborohydride (34.3 mg, 2.0 equiv) was added to the stir-
red solution, which then was allowed to stir for 24 h. The reaction
mixture was quenched by addition of saturated sodium hydro car-
bonate solution. The product was extracted with ethyl acetate
(5 ꢀ 5 mL). The combined organic phases were washed with brine
(15 mL) and dried over MgSO4. After removal of the solvent in va-
cuo the crude product was purified by column chromatography
using silica gel and ethyl acetate/methanol (8:2) elution. The prod-
uct was obtained as a colorless oil (22.2 g, 46%);7 1H NMR (CDCl3,
TMS, 300 MHz) d 2.39 (s, 6H), 2.56 (br m, 6H), 2.77 (br m, 2H),
2.89 (br m, 4H), 4.06 (s, 2H), 5.00 (br s, 3H), 7.24 (d, J = 8.3 Hz,
4H), 7.34 (m, 3H), 7.45 (d, J = 7.5 Hz, 2H), 7.75 (d, J = 8.3 Hz, 4H);
13C NMR (CDCl3, TMS, 75 MHz) d 21.5, 41.2, 45.5, 52.2, 52.4, 54.9,
127.1, 128.3, 128.9, 129.5, 129.6, 134.8, 137.3, 143.1; TLC (7:3
EtOAc/MeOH) Rf = 0.40.
Mono-N-Boc TREN disulfonamide (21). A solution of 20 (1.247 g,
5.062 mmol) and triethylamine (2.8 mL, 4.0 equiv) in acetonitrile
(80 mL) was prepared at room temperature under nitrogen. While
stirring,
a solution of toluenesulfonylchloride in acetonitrile
(40 mL) was added drop wise during 5 min. The reaction mixture
was allowed to stir as rt for 17 h after which time the solvent
was removed in vacuo. The residue was dissolved in dichlorometh-
ane (40 mL), washed with water (2ꢀ 10 mL) and with brine
(10 mL). After removal of the solvent the crude product was puri-
fied by column chromatography using silica gel and ethyl ace-
tate/hexanes gradient elution (30–60 vol% ethyl acetate). The
product 21 (2.091 g, 74%) was obtained as a viscous oil; 1H NMR
(CDCl3, 300 MHz) d 1.43 (s, 9H), 2.38 (t, J = 5.8 Hz, 2H), 2.41 (s,
6H), 2.47 (t, J = 5.2 Hz, 4H), 2.91 (m, 4H), 3.03 (m, 2 H), 5.07 (t,
J = 5.5 Hz, 1H), 5.80 (t, J = 5.0 Hz, 2H), 7.30 (d, J = 8.0 Hz, 4H), 7.78
(d, J = 8.0 Hz, 4H); TLC (6:4 EtOAc/cyclohexane) Rf = 0.31.
TREN disulfonamide (22). Over 15 min TFA (36 mL) was added
drop wise to a stirred solution of 21 (700 mg, 1.26 mmol) in dichlo-
romethane (36 mL) at 0 °C under nitrogen. The reaction mixture
was allowed to stir at 0 °C for 60 min after which the solvents were
removed in vacuo. The residue was dissolved in 60 mL ethyl ace-
tate and extracted with 1 M HCl (5ꢀ 10 mL). To the combined
aqueous phases saturated aqueous potassium carbonate solution
was added until the solution became basic. The basic solution
was extracted with ethyl acetate (4ꢀ 20 mL). The combined organ-
ic phases were washed with brine (20 mL) and dried with magne-
sium sulfate. Removal of the solvent in vacuo gave the product 22
(324 mg, 57%) as a viscous oil; 1H NMR (CD3OD, 300 MHz) d 2.38 (s,
6H), 2.48 (m, 6H), 2.71 (t, J = 5.3 Hz, 2H), 2.87 (t, J = 5.8 Hz, 4H),
7.34 (d, J = 8.2 Hz, 4H), 7.76 (d, J = 8.2, 4H).
5.2. Biochemical assays
5.2.1. Vesicle inward translocation assay
NBD–PC, NBD–PS and POPC were purchased from Avanti Polar
Lipids. Cholesterol was purchased from Sigma–Aldrich. All

K. M. DiVittorio et al. / Bioorg. Med. Chem. 17 (2009) 141–148
147
phospholipid stocks were stored in CHCl3 in a ꢁ20 °C freezer. An
appropriate mixture of cholesterol and POPC was dried in vacuo
for 1 h. A stock solution of 10 mM vesicles was made by rehydra-
tion at room temperature with TES buffer (5 mM TES, 100 mM
NaCl, pH 7.4). Multilamellar vesicles were extruded to form unila-
mellar vesicles with a Basic LiposoFast device purchased from
Avestin, Incorporated. The vesicles were extruded 29 times
through a 19-mm polycarbonate Nucleopore filter with 100-nm
diameter pores.
0.5 mL of the cell dispersion and incubated for 2 h at 37 °C, 5%
CO2. The samples were pelleted through centrifugation at
1500 rpm for 5 min and re-suspended in binding buffer. Annexin
V-FITC (5 lL) and the nuclear stain 7AAD (500 ng/mL) were
added to the samples and incubated for 15 min at 37 °C. Necrotic
cells, as well as those cells in the advanced stages of apoptosis,
have permeablized membranes and allow 7AAD to stain the cell
nucleus. Immediately after staining, flow cytometry was per-
formed on a Beckman Coulter Cytomics FC 500 MPL (Fullerton,
CA) flow cytometer. Data analysis was performed using CXP
Software (Fullerton, CA).
Another set of experiments used vanadate as a flippase inhibi-
tor. At 60% confluencey, PC3 cells were added to fresh media and
incubated with varying concentrations of sodium orthovanadate
for 2 h at 37 °C. The concentrations of vanadate ranged from 0 to
To a stock solution of vesicles (final concentration 25 lM) in TES
buffer, 0.5 mol% NBD–PC (in 200 proof ethanol) was added. From
this stock solution 3 mL samples were assayed using a Perkin-El-
mer Luminescence Spectrometer LS50B (excitation was set at
460 nm while the emission was measured at 530 nm using a
515 nm cut off filter). During this 200 s assay time, at 50 s 180
of sodium hydrosulfate (sodium dithionite; 60 mM in 1 M Tris,
L of 20% Triton X-100 (v/v)
lL
800
was added to the cells and incubated for 1 h at 37 °C. Cells were la-
beled with Annexin V-FITC (15 L), incubated at room temperature
lM. The TREN scramblase 2 (1 mM stock solution in DMSO)
pH ꢂ10) was added and at 150 s 20
l
was added. The amount of total NBD–phospholipid which was
translocated to the inside of the vesicles was calculated by the fol-
lowing equation:
l
for 15 min, washed with 1 mL of PBS, and then incubated at 37 °C
with trypsin to detach them. Samples were re-suspended in 1 mL
of Ham’s F-12K media and flow cytometry was performed as de-
scribed above.
% Exo NBD-PX ¼ ððFi ꢁ Ff Þ=FiÞ ꢃ 100
where Fi and Ff are the fluorescence intensities just prior to the
additions of sodium dithionite and Triton X-100. The values re-
ported are the averages of 2 or 3 repeated experiments. All percent
exo-probe values contain a 5% error and were normalized to the ini-
tial time point taken.
5.2.6. MTT assay
The VibrantÒ MTT cell proliferation assay kit was purchased
from Molecular Probes (Invitrogen), Inc. At the time of assay
200
l
L of cells (at least 5 ꢀ 104 cells/mL) in growth media were
dispensed in each well of a clear 96-well plate. TREN scramblases
(in a DMSO stock) were administered and incubated for a 24 h per-
iod. For the Jurkat cells the plate was centrifuged at 1500 rpm for
5.2.2. NBD protection assay
To a solution of vesicles (final concentration 25 lM) in TES buf-
fer was added 17 from a DMSO stock solution (25 mM). From this
vesicle solution, 3 mL samples were withdrawn and assayed using
a Perkin-Elmer Luminescence Spectrometer LS50B (excitation was
set at 460 nm while the emission was measured at 530 nm using a
5 min and the growth media was exchanged for 100
RPMI media and 10 L of a 12 mM MTT stock (5 mg MTT in 1 mL
filtered PBS buffer). For the PC3 cells or the LNCap cells the media
were exchanged for 100 L of fresh growth media and 10 L of a
12 mM MTT stock. After 4 h, 100 L of a SDS stock (1 g SDS in
lL of fresh
l
l
l
515 nm cut off filter). An aliquot of sodium dithionite (180
lL of a
l
60 mM stock in Tris buffer) was added at 50 s, and a 20
of 20% Triton X-100 (v/v) was added at 150 s.
lL aliquot
10 mL 0.01 M HCl) was added to each well and the plate was incu-
bated for 18 h at 37 °C, 5% CO2. The absorbance was measured
using a Labsystems Multiskan Ascent plate reader at the kmax is
570 nm. In order to determine the percent living cells in each sam-
ple the following equation was used:
5.2.3. Cell culture
Jurkat and LNCap cells were cultured separately in 75 cm2 tis-
sue culture flasks using RPMI media with 10% FBS. PC3 cells were
cultured in 75 cm2 tissue culture flasks using Ham’s F-12K media
supplemented with 10% FBS. All cell lines were incubated at
37 °C in a 5% CO2 atmosphere.
% Living cells ¼ ðA ꢁ ANegÞ=ðACon ꢁ ANegÞ ꢃ 100
where A is the absorbance of the well tested, ANeg is the absorbance
of the negative control (no cells added) and ACon is the absorbance
of the positive control (10 lL DMSO added).
5.2.4. Cell uptake of fluorescent TREN
TREN compound 17 (12.5 lM, 25 mM DMSO stock) was added
Acknowledgment
to 200
l
L of Jurkat cells (1 ꢀ 106 cells/mL) in growth media with
TM
a subsequent incubation at 37 °C, 5% CO2 for 30 min. ER Tracker
This work was supported by the NIH.
TM
and Mito Tracker were purchased from Molecular Probes and used
as directed by manufacture protocols. The sample was centrifuged
at 1500 rpm for 2 min and washed twice with TES buffer (5 mM
TES, 100 mM NaCl, pH 7.4). After washing, the cells were visualized
on an Axiovert S100 TV microscope (Carl Zeiss) equipped with fil-
ter sets DAPI/Hoechst/AMCA, FITC/RSGFP/Bodipy/Fluo3/DiO, Cy3
(Chroma). Images were acquired using a black and white digital
camera and colored upon acquisition using Metamorph software
version 6.2.
Supporting information
Supplementary data associated with this article can be found,
in the online version, at doi:10.1016/j.bmc.2008.11.011.
References and notes
1. Kornberg, R. D.; McConnell, H. M. Biochemistry 1971, 10, 1111.
2. Devaux, P.; McConnell, H. M. J. Am. Chem. Soc. 1972, 94, 4475.
3. Gordesky, S. E.; Marinetti, G. V. Biochem. Biophys. Res. Commun. 1973, 50, 1027.
4. Boon, J. M.; Smith, B. D. Med. Res. Rev. 2002, 22, 251.
5.2.5. Endogenous PS exposure
At 90% confluency the PC3 cells were washed twice with PBS
buffer then incubated at 37 °C, 5% CO2 with 2 mL of 2.5ꢀ trypsin
until the cells detached. Fresh media (8 mL) was added and the
cell sample was centrifuged for 5 min at 4 °C and re-suspended
in TES buffer to obtain a cell count of 1 ꢀ 105 cells/mL. The TREN
scramblase (25 mM stock solution in DMSO) was added to
5. (a) Bevers, E. M.; Comfurius, P.; Dekkers, D. W.; Zwaal, R. F. Biochim. Biophys.
Acta 1999, 1439, 317; (b) Daleke, D. L. J. Lipid Res. 2003, 44, 233; (c). Arch.
Biochem. Biophys. 2007, 462, 103.
6. (a) Yeung, T.; Gilbert, G. E.; Shi, J.; Silvius, J.; Kapus, A.; Grinstein, S. Science
2008, 319, 210; (b) Zwaal, R. F. A.; Comfurius, P.; Bevers, E. M. Cell. Mol. Life Sci.
2005, 62, 971.
7. Fadok, V. A.; Chimini, G. Immunology 2001, 13, 365.

148
K. M. DiVittorio et al. / Bioorg. Med. Chem. 17 (2009) 141–148
8. (a) Boon, J. M.; Smith, B. D. J. Am. Chem. Soc. 1999, 121, 11924; (b) Boon, J. M.;
10. Moss, R. A.; Bhattacharya, S. J. Am. Chem. Soc. 1995, 117, 8688.
11. McIntyre, J. C.; Sleight, R. G. Biochemistry 1991, 30, 11819.
12. Daleke, D. L.; Lyles, J. V. Biochim. Biophys. Acta 2000, 1486, 108.
13. Moderately lipophilic compounds such as organic dyes are known to rapidly
accumulate within the membranes of internal organelles. See, for example
Johnson, J. R.; Fu, N.; Arunkumar, E.; Leevy, W. M.; Gammon, S. T.; Piwnica-
Worms, D.; Smith, B. D. Angew. Chem. Int. Ed. 2007, 46, 5528.
14. (a) Peterson, B. R. Org. Biomol. Chem. 2005, 3, 3607; (b) Hao, M.; Maxfield, F. R. J.
Biol. Chem. 2000, 275, 15279.
Smith, B. D. J. Am. Chem. Soc. 2001, 123, 6221; (c) Boon, J. M.; Lambert, T. N.;
Smith, B. D.; Beatty, A. M.; Ugrinova, V.; Brown, S. N. J. Org. Chem. 2002, 67, 2168;
(d) Boon, J. M.; Shukla, R.; Smith, B. D.; Licini, G.; Scrimin, P. Chem. Comm. 2002,
260–261; (e) Shukla, R.; Sasaki, Y.; Krchnak, V.; Smith, B. D. J. Comb. Chem. 2004, 6,
703; (f) Sasaki, Y.; Shukla, R.; Smith, B. D. Org. Biomol. Chem. 2004, 2, 214.
9. (a) Lambert, T. N.; Boon, J. M.; Smith, B. D.; Perez-Payan, M. N.; Davis, A. P. J.
Am. Chem. Soc. 2002, 124, 5276; (b) Boon, J. M.; Lambert, T. N.; Sisson, A. L.;
Davis, A. P.; Smith, B. D. J. Am. Chem. Soc. 2003, 125, 8195; (c) Lambert, T. N.;
Smith, B. D. Coord. Chem. Rev. 2003, 240, 129; (d) Smith, B. D.; Lambert, T. N.
Chem. Comm. 2003, 2261; (e) DiVittorio, K. M.; Timothy, N.; Lambert, T. N.;
Smith, B. D. Bioorg. Med. Chem. 2005, 13, 4485.
15. Smith, B. D.; Davis, A. P.; Shephard, D. Chem. Soc. Rev. 2007, 36, 348.