A novel steroidal spin label for membrane structure studies: Synthesis and applications

Article abstract of DOI:10.1016/S0039-128X(99)00072-0

2,2,6,6-Tetramethyl piperidine-N-oxyl nitroxyls are known to partition between aqueous and lipid phases, thus serving as probes to study membrane dynamics. The synthesis of a novel steroidal spin label, 3α-hydroxycholan- 24-yl-(2'',2'',6'',6''-tetramethyl-N-oxyl)piperidyl butan-1',4'-dioate, containing 2,2,6,6-tetramethylpiperidine-N-oxyl moiety covalently bonded to the side chain in 3,24-caprostan-diol has been described. The localization of this spin label in model biomembranes has been studied by using electron spin resonance, differential scanning calorimetry, and ¹H and ³¹P NMR spectroscopic techniques. Its applicability in studying the phase transition properties of model membrane L-α-dipalmitoyl phosphatidyl choline in the presence and absence of drugs has been described by using electron spin resonance. The label has also been used to study the permeability of epinephrine into membrane. The results have shown the applicability of the spin label as a potential spin probe in the study of biomembranes.

Full text of DOI:10.1016/S0039-128X(99)00072-0

Steroids 64 (1999) 849–855
A novel steroidal spin label for membrane structure studies:
synthesis and applications
Rita Katocha, Girish K. Trivedia,*, Ratna S. Phadkeb
aDepartment of Chemistry, Indian Institute of Technology, Mumbai, 400076, India
bChemical Physics Group, Tata Institute of Fundamental Research, Mumbai, 400001, India
Received 14 April 1999; received in revised form 19 May 1999; accepted 21 May 1999
Abstract
2,2,6,6-Tetramethyl piperidine-N-oxyl nitroxyls are known to partition between aqueous and lipid phases, thus serving as probes
to study membrane dynamics. The synthesis of a novel steroidal spin label, 3␣-hydroxycholan-24-yl-(2Љ,2Љ,6Љ,6Љ-tetramethyl-N-
oxyl)piperidyl butan-1Ј,4Ј-dioate, containing 2,2,6,6-tetramethylpiperidine-N-oxyl moiety covalently bonded to the side chain in
3,24-caprostan-diol has been described. The localization of this spin label in model biomembranes has been studied by using electron
spin resonance, differential scanning calorimetry, and 1H and 31P NMR spectroscopic techniques. Its applicability in studying the
phase transition properties of model membrane L-␣-dipalmitoyl phosphatidyl choline in the presence and absence of drugs has been
described by using electron spin resonance. The label has also been used to study the permeability of epinephrine into membrane. The
results have shown the applicability of the spin label as a potential spin probe in the study of biomembranes. © 1999 Elsevier Science
Inc. All rights reserved.
Keywords: 3␣-hydroxycholan-24-yl-(2Љ,2Љ,6Љ,6Љ-tetramethyl-N-oxyl)piperidyl butan-1Ј,4Ј-dioate; TEMPO nitroxyl; Spin label; Model membrane;
Phase transition
1. Introduction
ture but the reports on the synthesis and applications of
steroidal TEMPO derivatives are relatively few [5–7].
Moreover, the steroids containing cis-fused A/B ring sys-
tems have also not been explored. SLs containing a steroid
molecule attached to TEMPO moiety are expected to have
higher mobility due to free rotation and are expected to be
a more versatile as TEMPO is known to partition between
aqueous and lipid phases [8].
The synthesis and applications of the first steroidal SL
having TEMPO moiety on the side chain was reported by
our group [9]. In continuation with our efforts to synthe-
size novel steroidal SLs for membranes and spin-immu-
noassays [10–13], it was thought worthwhile to synthe-
size and study the applications of some cis-fused
steroidal SLs containing TEMPO moiety at the side
chain. Our approach utilizes the introduction of the ni-
troxyl moiety in the substrate molecule by a covalent
bond formation between the substrate molecule and the
nitroxyl. For this purpose, a bile acid steroid, lithocholic
acid was chosen. Because the functional groups present
on the parent steroid molecule are, in most cases, respon-
sible for its biological activity, the site of attachment of
The spin-labeling technique is a well-established tool for
probing the structure and dynamic processes of model as
well as biomembranes. Various classes of molecules have
been labeled with the 2,2,6,6-tetramethylpiperidine-N-oxyl
(TEMPO) moiety to study membrane properties. These in-
clude small organic molecules like tempocholine, tem-
pophosphate, TEMPO labeled phosphatidylcholine, long-
chain acids and amines, etc. [1–4]. However, such labels are
not suitable for studying hydrophobic regions of membrane
and, hence, there was a need to synthesize spin labels (SLs)
that would bind more strongly to the hydrophobic regions of
membranes. Steroids are known for their affinity towards
membranes and can serve as good parent molecules for the
nitroxyls. Several derivatives of 4,4-dimethyloxazolidine-
N-oxyl (DOXYL) and 2,2,5,5-tetramethylpyrrolidine-N-
oxyl (PROXYL) steroids have been reported in the litera-
* Corresponding author. Tel.: ϩ91-22-5767173; fax: ϩ91-22-
5767152.
E-mail address: chgktia@ether.chem.iitb.ernet.in (G.K. Trivedi)
0039-128X/99/$ – see front matter © 1999 Elsevier Science Inc. All rights reserved.
PII: S0039-128X(99)00072-0

850
R. Katoch et al. / Steroids 64 (1999) 849–855
2. Experimental
Melting points are reported uncorrected. All solvents were
predried according to standard procedures. Petroleum ether
refers to the fraction having b.p. 60–80°C. DPPC and epi-
nephrine were purchased from Sigma Chemical Company (St.
Louis, MO, USA) whereas dilitazem was received as a gift
from Iswtituto di Science Fisiche, Unversity di Ancona, Italy.
4-Hydroxy TEMPO was prepared from 2,2,6,6-tetramethylpi-
peridine-4-one monohydrochloride that was purchased from
Fluka (Switzerland). 4-Oxo TEMPO was obtained by treating
2,2,6,6-tetramethylpiperidine-4-one monohydrochloride with
H2O2/Na2WO4 and was converted to the corresponding alco-
hol by reduction with lithium aluminum hydride. Column
chromatography was carried out by using 60–120 mesh silica
gel. Petroleum ether–ethyl acetate as the solvent system for
column chromatography as well as TLC. Elemental analysis
was carried out on CEST 1106. IR spectra were recorded on a
Nicolet Impact 400 Fourier transform IR spectrophotometer. A
Hewlett Packard MS Engine 5989-A spectrometer was used to
record the mass spectra. 1H NMR spectra were recorded on a
Varian VXR 300S spectrometer. About 5 mg of the sample
was dissolved in 0.6 ml of the solvent. The 1H NMR spectra of
nitroxyls were recorded after in situ reduction of their CDCl3
solutions with 1.5 equivalents of freshly distilled phenyl hydr-
1
azine. The H NMR and 31P NMR for determination of the
localization of the SL in DPPC (1:5 molar ratio) was taken in
0.6 ml of CDCl3. The differential scanning calorimetry exper-
iments were carried out on France Setaram instrument and the
volume of the sample taken was 0.85 ml. ESR spectra were
recorded at ambient temperature or at 50°C on Varian E-112
spectrometer operating in the X-band with tetracyanoethylene
as internal standard (g0 ϭ 2.00277). Deoxygenated chloroform
was used as the solvent for ESR measurements, the concen-
trations of the nitroxyls being ca. 10Ϫ5 M.
Scheme 1. Synthesis of 3␣-hydroxycholan-24-yl-(2Љ,2Љ,6Љ,6Љ-tetramethyl-
N-oxyl)piperidyl butan-1Ј,4Ј-dioate (7). Conditions: i) H2SO4, methanol,
8 h; ii) DHP, p-TSA, CH2Cl2, 12 h; iii) Lithium aluminum hydride, THF,
5 h; iv) Pyridine, succinic anhydride, 4 h; v) 4-Hydroxy TEMPO, DCC,
DMAP, THF, 16 h; vi) Methanol, p-TSA, 4 h.
the TEMPO moiety was chosen to be as far as possible
from its 3␣-hydroxy group. A hemisuccinate linker
would protrude the nitroxyl moiety away from the parent
steroid, thus increasing the flexibility of the SL [14].
The synthesized SL would be principally utilized to
study biomembrane properties. However, the lipid diversity
of biological membranes makes it experimentally, as well as
theoretically, difficult to characterize the thermodynamic
state of the lipid membrane and its associated phase equi-
libria. Within a reductionisitic approach, therefore, the stud-
ies of these properties of the model membranes constitutes
the first important step towards a description of complicated
multi-component membranes [15,16].
In the envisaged scheme, 4-hydroxy TEMPO has been
attached to a suitably modified C-24 atom of 3,24-capros-
tandiol via a succinyl group. The potential of the novel SL,
3␣-hydroxycholan-24-yl-(TEMPO) butan-1Ј,4Ј-dioate, thus
synthesized, has been exploited by incorporating it into a
model membrane L-␣-dipalmitoyl phosphatidyl choline
(DPPC), studying certain properties of DPPC and monitor-
ing the effect of various drugs on it with the help of electron
spin resonance (ESR) spectroscopy.
2.1. Preparation of multilamellar dispersions of DPPC
Chloroform solutions of the required SL and the lipid
were taken up in a small tube. The solvent was evaporated
slowly with a stream of nitrogen gas to obtain a thin film on
the walls of the tube. Traces of solvent were removed by
drying under vacuum for 3 h. The dried film was then
hydrated with required amount of 10 mM phosphate buffer
(pH ϭ 7.2) for 20 min and then vortexed at 50°C for 10 min
to get multilamellar vesicles. The concentration of the lipid
in the buffer solution was 80 mM and that of the SL was 0.8
mM. The samples were taken in 50 ␮l glass capillaries
sealed at both ends and mounted in the variable temperature
accessory of the spectrometer.
2.2. Preparation of unilamellar dispersions of DPPC
A lipid dispersion of DPPC in 10 mM phosphate buffer
with SL 7 was prepared according to the above mentioned
method. After a vortex of 10 min at 50°C, the dispersion

R. Katoch et al. / Steroids 64 (1999) 849–855
851
was subjected to sonification by using sonifier 1210 BRAN-
2.4. 3␣-Tetrahydropyranyloxycholan24-ol (4)
SON (Model 1210E-DTH, working frequency 47 KHz Ϯ
6%, HF-output power nom. 35 W) at 40°C. Sonification for
30 min produced clear homogeneous solution, that was used
for the ESR experiments.
For phase transition experiments with diltiazem, the mul-
tilamellar dispersion was hydrated as usual by using the
appropriate amount of 10 mM phosphate buffer (pH 7.2)
containing the drug. The system was allowed to equilibrate
for 20 min before vortexing. The molar ratios of the drug:
lipid:SL were maintained at 40:100:1.
For permeation studies with epinephrine, the temperature
was kept at 50°C, that is above the phase transition temper-
ature (41°C) of DPPC, to ensure that the lipid remained in
the liquid crystalline phase. The microwave power of the
instrument was set at a low value of 0.5 mW with a micro-
wave frequency of 9.1 GHz. The modulation amplitude was
set at 2.0 mW ϫ 1.0 G and the time constant of the detector
unit at 0.128 s. The scan time for each spectrum was 4 min.
The receiver gain at the start of the experiment was kept at
a fairly large value to give a strong signal whose decay
could be monitored with time. It was not altered throughout
the experiment. Only the first line in the ESR spectrum and
its decay with time were recorded.
Compound 3 (1 g, 2.1 mmol) was dissolved in dry
tetrahydrofuran (25 ml) under nitrogen atmosphere. Lithium
aluminum hydride (0.16 g, 4.2 mmol) was added to it slowly
at 0°C. The reaction mixture was then stirred at room
temperature for 5 h (monitored by TLC). After completion
of the reaction, the solvent was evaporated under reduced
pressure and the inorganic salts were filtered off followed
by extraction of the compound with ethyl acetate. After
evaporation of the solvent and purification over column, the
required compound 4 was obtained in 76% yield. m.p. ϭ
137–138°C. IR (KBr): v 3342, 2938, 2864, 1440, 1373,
1119, 1026 cmϪ1. MS: m/z 428 (M-18). Elemental analy-
sis: Calculated for C29H50O3: C, 77.97; H, 11.28%; found:
1
C, 77.85; H, 11.24%. H NMR (CDCl3): ␦ 4.73 (bs, 1H,
2ЉЈ-H), 3.91 (m, 1H, 6ЈЉ-H), 3.66 (m, 1H, 6ЈЉ-H), 3.61 (m,
2H, 24-H2), 3.51 (m, 1H, 3␤-H), 0.92 (d, J ϭ 6.3, 21-H3),
0.91 (s, 3H, 19-H3), 0.64 (s, 3H, 18-H3). 13C NMR (CDCl3):
␦ 98.8 (C-2ЈЉ), 71.8 (C-24), 68.0 (C-3), 62.7 (C-6ЈЉ), 23.3
(C-21), 18.6 (C-19), 12.0 (C-18).
2.5. Hemisuccinate 5 of 3␣-tetrahydropyranyloxycholan-
24-ol
In a typical experiment, to a 50 ␮l of the sonicated
preparation, required amount of drug solution in buffer was
added. This time, at which the drug was added, corre-
sponded to time t ϭ 0. Subsequently, the decay of the ESR
signal with time was monitored. The samples were taken in
50 ␮l glass capillaries sealed at both ends and mounted in
the variable temperature accessory of the spectrometer. In
all the permeation experiments, the concentration of the
lipid in the buffer solution was 100 mM and that of the SL
was 1.00 mM. The molar ratios of the drug:lipid:SL were
maintained at 20:100:1. The quencher solutions were pre-
pared in 10 mM phosphate buffer (pH 7.2).
Compound 4 (0.2 g, 0.45 mmol) was dissolved in pyri-
dine (2 ml) in a round bottom flask to which succinic
anhydride (0.045 g, 0.45 mmol) was added and stirred for
4 h. Ethyl acetate was then added and the reaction mixture
was filtered through celite. The product was extracted with
ethyl acetate, washed with dilute HCl, water and brine. The
organic extract was dried over anhydrous Na2SO4 and con-
centrated in vacuo. Silica gel column chromatography af-
forded the pure hemisuccinate 5 in 71% yield. m.p. ϭ
102–103°C. MS: m/z 546 (Mϩ). IR (KBr): v 3444, 2933,
2862, 1738, 1692, 1451, 1361, 1167, 1023 cmϪ1
.
Elemental analysis: Calculated for C33H54O6: C, 72.49; H,
1
2.3. Tetrahydropyranyl ether 3 of methyl lithocholate
9.95%; found: C, 72.44; H, 9.79%. H NMR (CDCl3): ␦
4.72 (bs, 1H, 2ЈЉ-H), 4.08 (m, 2H, 24-H2), 3.92 (m, 1H,
6ЈЉ-H), 3.64 (m, 1H, 6ЈЉ-H), 3.50 (m, 1H, 3␤-H), 2.64 (m, 4H,
2Ј-H2, 3Ј-H2), 0.91 (s, 3H, 19-H3), 0.64 (s, 3H, 18-H3). 13C
NMR (CDCl3): ␦ 177.5 (C-1Ј), 172.2 (C-4Ј), 96.7 and 96.4
(C-2ЈЉ, diastereomeric), 75.8 (C-24), 65.4 (C-3), 62.6 (C-6ЈЉ),
28.8 (C-2Ј, C-3Ј), 23.2 (C-21), 18.4 (C-19), 11.8 (C-18).
Methyl lithocholate (2) (1 g, 2.56 mmol) was dissolved
in dichloromethane (20 ml). Dihydropyran (0.43 ml, 5.12
mmol) and p-toluene sulphonic acid (0.015 g, 0.007 mmol)
were added to it. The reaction mixture was stirred for 14 h,
washed with 5% NaOH, water, brine and dried over anhy-
drous Na2SO4. The solvent was evaporated under reduced
pressure and the residue was purified over silica gel column
to obtain the required product 3 as a semisolid in 90% yield.
IR (KBr): v 3015, 2939, 2866, 1732, 1448, 1366,
1025 cmϪ1. MS: m/z 474 (Mϩ). Elemental analysis: Cal-
culated for C30H50O4: C, 75.90; H, 10.62%; found: C,
75.50; H, 10.40%. 1H NMR (CDCl3): ␦ 4.72 (bs, 1H, 2Ј-H),
3.91 (m, 1H, 6Ј-H), 3.66 (s, 3H, COOMe), 3.63 (m, 1H,
6Ј-H), 3.50 (m, 1H, 3␤-H), 2.36 (ddd, J ϭ 5.4, 10.2, 15.3,
1H, 23-H), 2.21 (ddd, J ϭ 6.6, 9.6, 15.3, 1H, 23-H), 0.91
(s, 3H, 19-H3), 0.40 (d, J ϭ 6.3, 3H, 21-H3), 0.63 (s, 3H,
18-H3).
2.6. TEMPO derivative 6 of 3␣-tetrahydropyranyloxy
cholan-24-hemisuccinate
Compound 5 (0.1 g, 0.18 mmol) was dissolved in dry
THF (20 ml) and to it 4-hydroxy-TEMPO (0.03 g, 0.18
mmol) in THF (5 ml), 1,3-dicyclohexylcarbodiimide (0.03
g, 0.18 mmol) and 4-dimethylaminopyridine (0.02 g, 0.18
mmol) were added under nitrogen atmosphere. The reaction
mixture was stirred for 16 h followed by filtration of insol-
uble urea and evaporation of the solvent under reduced
pressure. The crude product obtained was purified by silica

852
R. Katoch et al. / Steroids 64 (1999) 849–855
gel column chromatography to accrue compound 6 in 25%
yield. m.p. ϭ 44–45°C. MS: m/z 701 (M ϩ 1). IR (CHCl3):
v 2927, 2858, 1735, 1628, 1460, 1361, 1157, 1027
cmϪ1. Elemental analysis: Calculated for C42H70O7N: C,
71.96; H, 10.06; N, 2.00%; found: C, 71.86; H, 10.00; N,
2.00%. 1H NMR (CDCl3 with 1.5 equivalents of
PhNHNH2): ␦ 4.72 (bs, 1H, 2ЈЉ-H), 4.05–4.10 (m, 3H, 4Љ-H,
24-H2), 3.92 (m, 1H, 6ЈЉ-H), 3.64 (m, 1H, 6ЈЉ-H), 3.50 (m,
1H, 3␤-H), 2.62 (m, 4H succinyl (H2)2), 1.38 (s, 6H, gem-
dimethyls), 1.25 (s, 6H, gemdimethyls), 0.91 (s, 3H, 19-H3),
0.64 (s, 3H, 18-H3). ESR spectrum (10Ϫ5 M in CHCl3):
symmetrical triplet with g0 ϭ 2.0058 and A0 ϭ 15.81 G.
alcohol 4 that showed the -OH absorption band at 3342
cmϪ1 in the IR spectrum. The mass spectrum of compound
4 (C29H50O3) exhibited the molecular ion peak at m/z 428
(M-18). The absence of a three-proton singlet for the -OCH3
1
group was eminent in the H NMR spectrum of alcohol 4.
Instead, a two-proton multiplet, characteristic of -OCH2-
group was observed at ␦ 3.61 ppm, that was assigned to the
C-24 protons. The 13C NMR spectrum of compound 4 also
lacked in carbonyl resonances but showed a shielded C-24
peak at 71.8 ppm. The peaks at 98.8 (C-2ЈЉ) and 62.7 ppm
(C-6ЈЉ) were characteristic of the THP moiety. Alcohol 4
was then converted to its hemisuccinate derivative 5, that
was characterized by the presence of an ester and acid
carbonyl signal at 1738 and 1692 cmϪ1, respectively, in its
IR spectrum. Mass spectrum of compound 5 exhibited the
Mϩ peak at m/z 546 (C33H54O6). The 1H NMR spectrum of
compound 5 showed a multiplet at ␦ 2.64 ppm (4H) char-
acteristic of the hemisuccinate methylene protons in addi-
tion to the downfield shift of the 24-H2 resonance by ca. 0.5
ppm. The presence of an ester carbonyl peak at 177.5 ppm
along with a peak at 172.28 ppm for -COOH in its 13C NMR
spectrum confirmed its formation. The hemisuccinate 5
hence obtained was then condensed with 4-hydroxy-
TEMPO in the presence of dicyclohexyl carbodiimide and
N,N-dimethylamino pyridine to yield the nitroxyl 6. Product
formation was indicated by the absence of the acid carbonyl
peak and the appearance of ester carbonyl peaks at 1735
cmϪ1 in the IR spectrum. The mass spectrum of compound
2.7. 3␣-Hydroxycholan-24-yl-(TEMPO)piperidyl butan-
1Ј,4Ј-dioate (7)
The tetrahydropyranyl ether derivative (6; 50 mg, 0.071
mmol) was dissolved in aqueous methanol (20 ml) and
catalytic amount of p-toluene sulphonic acid was added to
it. The reaction mixture was stirred at room temperature for
4 h (monitored by TLC). After completion, the solvent was
removed under reduced pressure and the residue was ex-
tracted with ethyl acetate. The organic layer was washed
with sodium bicarbonate and dried over Na2SO4 and con-
centrated in vacuo. The crude product was column chro-
matographed to obtain the pure SL 7 as a pale yellow oil (31
mg, 70%). MS: m/z 617 (M ϩ 1). IR (CHCl3): v 3320,
2927, 2858, 1735, 1629, 1460, 1360, 1155, 1027
cmϪ1. Elemental analysis: Calculated for C37H62O6N: C,
72.04; H, 10.13; N, 2.27%; found: C, 72.00; H, 10.10; N,
2.05%. 1H NMR (CDCl3 with 1.5 equivalents of
PhNHNH2): ␦ 4.08 (m, 3H, 4Љ-H, 24-H2), 3.62 (m, 1H,
3␤-H), 2.62 (m, 4H, 2Ј-H2, 3Ј-H2), 1.36 (s, 6H, gemdi-
methyls), 1.28 (s, 6H, gemdimethyls), 0.91 (s, 3H, 19-H3),
0.68 (s, 3H, 18-H3). ESR spectrum (10Ϫ5 M in CHCl3):
symmetrical triplet with g0 ϭ 2.0060 and A0 ϭ 15.375 G.
1
6 showed M ϩ 1 peak at m/z 701 (C42H70O7N). The H
NMR spectrum after reduction of the nitroxyl with 1.5
equivalents of phenyl hydrazine showed additional singlets
at 1.38 and 1.25 ppm assigned to the gem-dimethyls of the
TEMPO moiety. The ESR spectrum of the nitroxyl showed
a symmetrical triplet with g0 ϭ 2.0058 and A0 ϭ 15.81 G.
The need for the deprotection of the 3␣-hydroxy group
of the TEMPO ester 6, stemmed from the fact that the SL
would thus possess a free 3␣-hydroxyl group similar to that
in the parent molecule, lithocholic acid and, hence, could
ensemble in the membrane layers like a bile steroid. The
cleavage of the tetrahydropyranyl ether group was achieved
in a yield of 70% by using p-toluenesulphonic acid in
methanol. The IR spectrum of product 7 showed the pres-
ence of a broad band at 3320 cmϪ1 indicating the desired
deprotection had occurred. Peaks corresponding to the THP
group were absent in the 1H NMR spectrum of phenyl
hydrazine reduced product 7 whereas the 3␣-H peak ap-
peared at 3.62 ppm. The mass spectrum of the nitroxyl 7
exhibited the M ϩ 1 peak at m/z 617 (C37H62O6N) and the
3. Results and discussion
The C-24 carboxy group of lithocholic acid (1) was
converted to its corresponding hydroxy group. For this,
methyl lithocholate (2), characterized by the presence of
1730 cmϪ1 peak for the ester carbonyl in the IR spectrum
and a three-proton singlet at ␦ 3.66 ppm for the ester OMe
1
in the H NMR spectrum, was prepared from compound 1.
Compound 2 was then converted to the corresponding 3␣-
tetrahydropyranyl (THP) ether 3. The product formed
showed the absence of -OH peak in its IR spectrum. The 1H
NMR spectrum showed resonances at 4.72, 3.91, and 3.63
ppm corresponding to the THP group. The multiplet for the
3␤-H was deshielded and appeared at 3.50 ppm. The mass
spectrum of compound 3 exhibited the molecular ion peak
at m/z 474 (C30H50O4) thus confirming it to be tetrahydro-
pyranyl methyl lithocholate. Lithium aluminum hydride re-
duction of the methyl ester 3 yielded the C-24 primary
ESR spectrum showed a symmetrical triplet with g0
ϭ
2.0060 and A0 ϭ 15.375 G. These data confirmed that
compound 7 is 3␣-hydroxycholan-24-yl-(2Љ,2Љ,6Љ,6Љ-tetra-
methyl-N-oxyl) piperidyl butan-1Ј,4Ј-dioate.
The TEMPO derivative 7 of 24-hemisuccinoyl capro-
stan-3␣-ol was then incorporated in the model membrane
DPPC liposomes to study its application in probing the

R. Katoch et al. / Steroids 64 (1999) 849–855
853
Fig. 1. ESR spectra of the nitroxyl 7 in (a) rapidly tumbling solution state (10Ϫ5 M in CHCl3) and (b) lipid matrix (DPPC liposomes), SL:DPPC 0.8:80.0
mM.
1
fluidity and permeability properties of the model membrane
as such and in presence of certain drugs.
(WLS) are tabulated (Table 1). Among the peaks in the H
NMR spectra, the broadening observed for the characteristic
-NMe3 constituting the polar head group of the lipid and the
terminal-CH3 group constituting the other extreme end viz.
nonpolar hydrocarbon region of the lipid were taken into
consideration. The -NMe3 peak showed more broadening
due to the proximity of the TEMPO group to the polar head
group of the lipid. The 31P resonances also indicated similar
possibility.
The incorporation of the nitroxyl 7 in liposomes was
ascertained by comparing the ESR spectra of the nitroxyl in
rapidly tumbling solution state (CHCl3) with that in the lipid
matrix, i.e. DPPC liposomes (Fig. 1). The hyperfine cou-
pling constants of the SLs in multi-lamellar vesicles was
found to be 15.26 G although in rapidly tumbling solution
phase was found to be 15.375 G. The differences observed
in the line shapes of the highest and lowest field lines in the
ESR spectrum of these SL in liposomes is due to the
anisotropy of the nuclear hyperfine coupling tensor and
g-tensor of the radicals in the lipid [13]. The absence of
composite peaks along with the changes in the line shape
and hyperfine coupling constant suggested the complete
incorporation of the SL 7 into the DPPC liposomes.
Differential scanning calorimetry of pure DPPC vesicles
showed the phase transition temperature at 41°C whereas
those incorporated with steroid 7 showed the same at 40°C
that is within the error limits of the instrument. This showed
the incorporation of the SL 7 did not cause any significant
perturbation to the membrane system.
Phase transition behavior of DPPC dispersions were
studied by using the SL 7 as ESR-sensitive probe. The
empirical parameter hϩ1/h0 (ratio of the signal heights of the
low field line to the central line in the ESR spectrum of the
SL) was plotted as a function of temperature to monitor the
phase transition characteristics (Fig. 2). This parameter
(hϩ1/h0) showed an initial gradual increase, followed by a
sudden large increase in the transition of the lipid from gel
to liquid crystalline state at 40°C, close to the values re-
ported earlier by using other techniques. The alterations in
Table 1
Line broadening of NMe3 and Terminal Me 1H and 31P NMR signals
caused by incorporation of spin label 7 in DPPC
Further evidences for the incorporation of the SL 7 in the
1
liposomes and the mode of localization came from the H
Half-line widths (Hz)
and 31P NMR spectroscopy [17,18]. It was observed that on
NϩMe3
Terminal Me
31P
1
incorporation of the nitroxyl 7, the H NMR resonances of
WL
WLS
Pure DPPC
7 ϩ DPPC
3.36505
5.50946
1.6372
1.3206
1.542
1.151
1.9362
2.7107
1.4
different regions underwent line broadening to different
1
extents. The half-line widths of the H and 31P NMR reso-
W
LS/WL
nances of pure DPPC (WL) and those incorporated with SL

854
R. Katoch et al. / Steroids 64 (1999) 849–855
Fig. 2. Plot of temperature (°C) v/s spectral parameter hϩ1/h0 of the SL 7
incorporated into pure DPPC liposomes, SL:DPPC 1:100 (F) and in
presence of diltiazem, SL:DPPC:Drug 1:100:40 (Œ).
Fig. 3. Plot of time (s) v/s the signal height S(t) of the ESR spectrum of the
SL 7 incorporated into DPPC liposomes in presence of epinephrine (SL:
DPPC:Drug 1:100:20).
the phase transition temperature induced by vasodilating
drug, diltiazem was also studied. In the presence of dilti-
azem, the transition temperature of DPPC was observed to
be lowered to 36°C (Fig. 2). However, the sigmoidal nature
of the curve is retained. This showed that the drug has an
affinity to bind loosely to the hydrophobic region of the
lipid [4,19].
ciple, decay with different rate constants, say k0 and ki,
respectively. Under these conditions the decay of the signal
intensity is expected to show a behavior as given by the
equation:
S͑t͒ ϭ S0͑0͒eϪk t ϩ Si͑0͒eϪk t
o
i
Finally, the permeation of epinephrine, an adrenergic
neurotransmitter, in model membrane DPPC was studied
utilizing ESR properties of the new TEMPO probe 7. Most
of the actions of this anti-anesthetic drug molecule, known
to inhibit glycogen breakdown and increase sodium trans-
port across the cell membrane of erythrocytes, are triggered
when it complexes with appropriate receptors located in the
plasma membrane with their binding sites oriented exter-
nally causing a local change in the membrane [20].
The nitroxyl spin labels are known to undergo reduction
by reducing agents such as ascorbic acid. Making use of this
fact, the rate of drug permeation in outer and inner mono-
layers has been determined earlier [21]. Epinephrine possess
the property of reacting with the spin label, thereby causing
loss of its paramagnetism. When this drug was introduced
into the DPPC system labeled with nitroxyl 7, it diffused
into the bilayer. The spin label molecules present in the
outer monolayer were readily accessible and therefore un-
derwent reduction at a faster rate as compared to those
residing in the inner region. As a result, the ESR signal
height decreased with time. The plot of signal height with
time showed an exponential decay (Fig. 3).
where S(t) is the ESR signal height due to the total SL
present at time t and S0(0) and Si(0) are signal heights due
to initial SL concentration in the outer and inner monolay-
ers, respectively [21,22]. From these results, half-life times
for the reduction of the SLs, that in turn reflect the rates of
permeation of the drugs, for the outer and inner monolayer
were calculated to be 19.05 s and 68.98 s, respectively.
These results showed that the drug molecules did not pro-
trude deep into the lipid hydrophobic core of the membrane
but were bound on the surface and diffused laterally at a
moderate rate [23].
In conclusion, a novel steroidal SL—3␣-hydroxycholan-
24-yl-(2Љ,2Љ,6Љ,6Љ-tetramethyl-N-oxyl) piperidyl butan-1Ј,4Ј-
dioate—was synthesized and was successfully incorporated in
DPPC liposomes. The paramagnetic properties of this SL
could be used to monitor phase transition in DPPC as such and
in the presence of diltiazem. The study involving the perme-
ation of epinephrine in DPPC by using ESR properties of the
synthesized spin labels showed that the results were congruent
with the literature reports. Thus, the new caprostanol TEMPO
derivative shows good potential as a spin probe in membrane
structure studies.
The ESR signal heights (S0) and (Si) due to the SL
present in the outer and the inner monolayers can, in prin-

R. Katoch et al. / Steroids 64 (1999) 849–855
855
[11] Banerjee S, Desai UR, Trivedi GK. Synthesis of 2Ј-(3␣-benzyloxy-
24-norcholan-23-yl)-2Ј,4Ј,4Ј-trimethyl-4Ј,5Ј-dihydrooxazoline-N-oxyl—
a new potential spin probe for biomembranes. Tetrahedron 1992;48:133–
48.
[12] Banerjee S, Trivedi GK, Srivastava S, Phadke RS. 2Ј-(3␣-benzyloxy-
24-norcholan-23-yl)-2Ј,4Ј,4Ј-trimethyl-4Ј,5Ј-dihydrooxazoline-N-
oxyl-as a potential spin probe for model membranes. Steroids 1994;
59:377–82.
[13] Banerjee S, Trivedi GK, Srivastava S, Phadke RS. Proxyl nitroxide of
lithocholic acid: a potential spin probe for model membranes. Bioorg
Med Chem 1993;1:341–7.
[14] Desai UR, Trivedi GK. New steroid haptens for radioimmunoassay;
synthesis of steroids with thio ether or ester linkage at the 2␣-
position. Steroids 1991;56:185–91.
Acknowledgment
We thank the Council for Scientific and Industrial Re-
search, New Delhi and Board of Research in Nuclear Sciences,
Mumbai for financial assistance. The spectral facilities pro-
vided by Regional Sophisticated Instrumentation Centre, In-
dian Institute of Technology, Mumbai and National Facility for
High Field NMR, Tata Institute of Fundamental Research,
Mumbai are gratefully acknowledged. We also wish to thank
Prof. Nand Kishore, Indian Institute of Technology, Mumbai
for providing the differential scanning calorimetry facility.
[15] Nayar R, Hope MJ, Cullis PR. Generation of large unilamellar ves-
icles from long-chain saturated phosphatidyl cholines by extrusion
techniques. Biochim Biophys Acta 1989;986:200–6.
[16] London E, Feigenson, GW. Fluorescence quenching in model mem-
branes, an analysis of local phospholipid environment of diphenyl-
hexatriene and Gramicidine A. Biochim Biophys Acta 1981;649:89–
97.
References
[1] Weiner H. Interaction of a spin-labeled analog of nicotinamide-
adenine dinucleotide with alcohol dehydrogenase. I. Synthesis, ki-
netic and electron paramagnetic resonance studies. Biochemistry
1969;8:526–33.
[2] Hisa JC, Piette LH, Noyes RW. Binding of spermatozoa with surface
active spin labels. J Reprod Fertil 1969;20:147–9.
[17] Srivastava S, Phadke RS, Govil G, Rao CNR. Fluidity, permeability
and antioxidant behaviour of model membranes incorporated with
␣-tocopherol and vitamin E acetate. Biochem Biophys Acta 1983;
734:353–62.
[18] Brockerhoff H. Model of the interaction of polar lipids, cholesterol,
and proteins in biological membranes. Lipids 1974;9:645–50.
[19] Rosenberg PH, Jansson JE, Gripenberg J. Effects of halothane,
thiopental, and lidocaine on fluidity of synaptic plasma mem-
branes and artificial phospholipid membranes. Anesthesiology
1977;46:322–6.
[20] Srivastava S, Phadke RS, Govil G. Magnetic resonance studies of
interaction of epinephrine and norepinephrine with model mem-
branes. Ind J Chem 1984;23B:1184–93.
[21] Marsh D, Phillips AD, Watts A, Knowles PF. A spin label study of
fractionated egg phosphatidyl choline vesicles. Biochem Biophys Res
Commun 1972;49:641–8.
[3] Waggoner AS, Keith AD, Griffith OH. Electron spin resonance of
solubilized long chain nitroxides. J Phys Chem 1968;72:4129–32.
[4] Hubbell WL, Metcalfe JC, Metcalfe SM, McConnell HM. The inter-
action of small molecules with spin-labeled erythrocyte membranes.
Biochim Biophys Acta 1970;219:415–27.
[5] Keana JFW. Newer aspects of the synthesis and chemistry of nitrox-
ide spin labels. Chem Rev 1978;78:37–64.
[6] Naik N, Braslau R. Synthesis and applications of optically active
nitroxides. Tetrahedron 1998;54:667–96.
[7] Banerjee S, Trivedi GK. Recent advances in the chemistry of nitroxides
and their applications in spin labeling. J Sci Ind Res 1995;54:623–36.
[8] McConnell HM. Molecular motions in biological membranes. In:
Berliner LJ, editor. Spin labeling I: Theory and applications. New
York, Academic Press, 1976; p. 526.
[9] Banerjee S, Trivedi GK. Synthesis of potential spin probes for bi-
omembranes—tempo and proxyl nitroxides of lithocholic acid. Tet-
rahedron 1992;48:9939–50.
[22] Kornberg RD, McConnell HM. Inside outside transitions of phospho-
lipids in vesicle membranes. Biochemistry 1971;10:1111–20.
[23] Schreier–Muccillo S, Marsh D, Smith ICP. Monitoring the perme-
ability profile of lipid membranes with spin probes. Arch Biochem
Biophys 1976;72:1–11.
[10] Katoch R, Korde SS, Deodhar KD, Trivedi GK. Synthesis of potential
spin probes of pregnenolone and progesterone for biomembranes and
immunoassays. Tetrahedron 1999;55:1741–54.