Deracemization of bilirubin as the marker of the chirality of micellar aggregates

Article abstract of DOI:10.1002/chir.21026

The deracemization of bilirubin in micellar aggregates of structurally correlated chiral surfactants was studied by circular dichroism experiments and exploited as the marker of the expression of chirality of the aggregates. The obtained results suggest that the hydrophobic interactions control the transfer of chirality from the monomers to the aggregates, and that different regions of the same aggregate might feature opposite enantiorecognition capabilities.

Full text of DOI:10.1002/chir.21026

CHIRALITY 24:78–85 (2012)
Deracemization of Bilirubin as the Marker of the Chirality
of Micellar Aggregates
ALESSANDRO SORRENTI,1 BARBARA ALTIERI,2 FRANCESCA CECCACCI,2 PIETRO DI PROFIO,3,4 RAIMONDO GERMANI,3,4
LUISA GIANSANTI,1 GIANFRANCO SAVELLI,3,4 AND GIOVANNA MANCINI1,4
*
1CNR, Istituto di Metodologie Chimiche, Sezione Meccanismi di Reazione and Dipartimento di Chimica,
Universita` degli Studi di Roma ‘‘La Sapienza,’’ P.le A Moro 5, Roma, Italy
2Dipartimento di Chimica, Universita` degli Studi di Roma ‘‘La Sapienza,’’ P.le A Moro 5, Roma, Italy
3Dipartimento di Chimica, Universita` degli Studi di Perugia, Via Elce di Sotto 8, Perugia, Italy
4Centro di Eccellenza Materiali Innovativi Nanostrutturati per Applicazioni Chimiche, Fisiche e Biomediche (CEMIN),
Universita` degli Studi di Perugia, Via Elce di Sotto 8, Perugia, Italy
ABSTRACT
The deracemization of bilirubin in micellar aggregates of structurally correlated
chiral surfactants was studied by circular dichroism experiments and exploited as the marker of
the expression of chirality of the aggregates. The obtained results suggest that the hydrophobic
interactions control the transfer of chirality from the monomers to the aggregates, and that dif-
ferent regions of the same aggregate might feature opposite enantiorecognition capabilities.
C
VV
Chirality 24:78–85, 2012.
2011 Wiley Periodicals, Inc.
KEY WORDS: circular dichroism; micelle; aggregation; deracemization; bilirubin
INTRODUCTION
enantiomers (Scheme 1) interconverting with a barrier of
18–20 kcal mol21 22
A shift of the 1:1 enantiomeric equilib-
.
The lack of chiral symmetry is a widespread feature of
nature observed at each level of complexity, from the sub-
atomic and molecular scale to the macroscopic length scale
of some living organisms. Chiral homogeneity is also a trait
of the components of biological membranes, lipids and
proteins, that determine and control a number of complex
biological functions by self-organizing into the membrane
lipid bilayers according to the program encoded in their
molecular structure. As a facet of the molecular structure,
chirality affects the organization and functions of biomem-
branes through chiral recognition and diastereomeric inter-
actions. However, it is still poorly understood how the chiral
information encoded in the monomers is transferred into the
assembly and how the chiral function is translated in these
complex systems.
Because of the complexity of biomembranes, many investi-
gations, aimed at clarifying how chiral recognition works in
these systems, have been carried out on models such as
micelles,1–9 liposomes,10–12 and Langmuir monolayers.13–17
Micelles, although different from biological membranes in
their morphology and thermodynamics, are controlled by the
same type of noncovalent interactions that control the organi-
zation of biological membrane; further, differently from lipid
monolayers and bilayers they are stable and because of their
small size are more suitable than other models for spectro-
scopic investigations. Some investigations on chiral recogni-
tion in micellar aggregates have shown that the stereochemi-
cal information of chirality may influence the morphology
and the stability of the aggregates18,19 and may govern the
enantiodiscrimination in the interaction of chiral self-assem-
blies with chiral solutes.10,15,20,21 Nevertheless, the mecha-
nisms of chiral recognition and the sites of recognition in the
assembly have not been completely clarified yet.
rium toward one of the enantiomers has been observed in
the presence of chiral selectors such as proteins,23–26 chiral
amines,27 cyclodextrins,28 and chiral micelles,29 due to the
different stability of the diasteromeric complexes.
In a previous investigation, it was observed that the micel-
lar aggregates formed by the enantiopure N-alkyl-N,N-
dimethyl-N-(1-phenylethyl)ammonium bromides are able to
induce a deracemization of bilirubin racemic mixture.30 Inter-
estingly, the extent and the direction of the observed dera-
cemization was shown to depend strongly on the length of
the hydrophobic alkyl chain and on concentration conditions.
The work described here extended the above-mentioned
investigation to the aggregates formed by the structurally
correlated chiral cationic surfactants 1–4, exploring how dif-
ferent molecular features can influence the organization of
micellar aggregates and hence their expression of chirality.
MATERIALS AND METHODS
Chemicals (Sigma-Aldrich) were of the highest grade available and
were used without further purification. Solvents for the spectroscopic
studies were of spectroscopic grade and used as received. Thin Layer
Cromatography (TLC): Merck silica gel 60 F254; CC: Merck silica gel
60, 70–230 mesh American Society for Testing and Materials (ASTM).
1H Nuclear Magnetic Resonance (NMR) spectra were recorded on a
Bruker AC 300 operating at 300.13 and 75.47 MHz for 1H and 13C,
respectively, equipped with a sample tube thermostating apparatus. Sig-
nals were referenced with respect to tetramethylsilane (TMS) (d 5 0.000
ppm).
CD spectra were recorded on a Jasco spectropolarimeter J-715.
Contract grant sponsor: Dipartimento di Progettazione Molecolare.
*Correspondence to: Giovanna Mancini, CNR, Istituto di Metodologie
Chimiche, Sezione Meccanismi di Reazione and Dipartimento di Chimica,
Universita` degli Studi di Roma ‘‘La Sapienza,’’ P.le A Moro 5, 00185 Roma,
Italy. E-mail: giovanna.mancini@uniroma1.it
Received for publication 19 May 2011; Accepted 11 August 2011
DOI: 10.1002/chir.21026
Published online 26 November 2011 in Wiley Online Library
(wileyonlinelibrary.com).
Herein, a circular dichroism (CD) investigation on the chi-
ral recognition of bilirubin-IXa in micelles formed by the chi-
ral cationic surfactants 1–4 (Scheme 1) is reported. It is
known that bilirubin is a racemic mixture of conformational
C
VV
2011 Wiley Periodicals, Inc.

SORRENTI ET AL.
79
(2)-(1R,2S)-N-hexadecyl-N,N-dimethylephedrinium bromide,
1b. 0.51 g (2.8 mmol) of (1R,2S)-(2)-N-methylephedrine were quater-
nized with 0.92 g (3.0 mmol) of 1-bromohexadecane according to the
procedure described above for 1a. Crystallization by methanol/ether
yielded 0.61 g (45%) of a white powder (m.p. 117.0–118.58C).
1H NMR, d(CDCl3) ppm: 0.838 (3H, t, ÀÀ(CH2)11CH3); 1.130 (3H, d,
ÀÀCHCH3); 1.215 (24H, m, ÀÀ(CH2)12CH3); 1.297 (2H, m,
ÀÀN(CH2)2CH2ÀÀ); 1.681 (2H, m, ÀÀNCH2CH2ÀÀ); 3.265 (3H, s, ÀÀNCH3);
3.40–3.51 (1H, m, ÀÀNCH2); 3.424 (3H, s, ÀÀNCH3); 3.565 (1H, m,
ÀÀCHCH3); 3.770 (1H, m, ÀÀNCH2); 5.463 (1H, bs, ÀÀOH); 5.722 (1H, d,
ÀÀCHOHÀÀ); 7.196 (1H, m, ar); 7.286 (2H, t, ar); 7.454 (2H, d, ar).
13C NMR, d(CDCl3) ppm: 7.27; 14.09; 22.65; 22.87; 26.37; 29.29; 29.32;
29.40; 29.48; 29.59; 29.62; 29.67; 31.89; 49.47; 49.71; 64.22; 67.82; 72.83;
125.96; 127.56; 128.48; 141.17. aD5 210.6 (c 1.28, CH3OH).
Molar ellipticity (CH3OH, deg cm2 dmol21): 1633 (267 nm); 1805
(260 nm); 1566 (255 nm); 1300 (248 nm, shoulder); 1111 (241 nm,
shoulder).
(1)-(1S,2S)-N-dodecyl-N,N-dimethylpseudoephedrinium bromide,
2a. 0.51 g (2.8 mmol) of (1S,2S)-(2)-N-methylpseudoephedrine were
quaternized with 0.75 g (3.0 mmol) of 1-bromododecane. Purification by
chromatography on silica gel in gradient of polarity (CHCl3/MeOH from
95/5 to 80/20) yielded 0.48 g (40%) of a yellow oil.
1H NMR, d(CDCl3) ppm: 0.836 (3H, t, ÀÀ(CH2)11CH3); 1.120 (3H, d,
ÀÀCHCH3); 1.212 (16H, m, ÀÀ(CH2)8CH3); 1.261 (2H, m,
ÀÀN(CH2)2CH2ÀÀ); 1.684 (2H, m, ÀÀNCH2CH2ÀÀ); 2.560 (1H, bs, OH);
3.223 (3H, s, ÀÀNCH3); 3.321 (1H, m, ÀÀNCH2); 3.374 (3H, s, ÀÀNCH3);
3.683 (1H, m, ÀÀCH(CH3)ÀÀ); 3.800 (1H, m, ÀÀNCH2); 5.110 (1H, d,
ÀÀCHOHÀÀ); 7.244 (1H, m, ar); 7.316 (2H, t, ar); 7.402 (2H, d, ar).
13C NMR, d(CD3OD) ppm: 13.35; 14.06; 22.62; 22.86; 26.32; 29.19;
29.27; 29.37; 29.44; 29.55; 31.84; 50.18; 51.19; 65.96; 71.38; 73.73; 127.44;
128.42; 128.85; 141.69. aD 5 138.1 (c 1.92, CH3OH).
Molar ellipticity (CH3OH, deg cm2 dmol21): 2178 (267 nm); 2237 (262
nm); 2160 (255 nm); 274 (250 nm, shoulder); 224 (244 nm, shoulder).
(1)-(1S,2S)-N-hexadecyl-N,N-dimethylpseudoephedrinium bro-
mide, 2b. 0.51 g (2.8 mmol) of (2)-(1S,2S)-N-methylpseudoephedrine
were quaternized with 0.92 g (3.0 mmol) of 1-bromohexadecane. Crystal-
lization by methanol/ether yielded 0.54 g (40%) of a white powder (m.p.
90–928C).
Scheme 1. Chiral cationic surfactants 1–4 and probe of chirality bilirubin
IXa.
1H NMR, d(CDCl3) ppm: 0.836 (3H, t, ÀÀ(CH2)11CH3); 1.029 (3H, d,
ÀÀCHCH3); 1.214 (24H, m, ÀÀ(CH2)12CH3); 1.262 (2H, m,
ÀÀN(CH2)2CH2ÀÀ); 1.690 (2H, m, ÀÀNCH2CH2ÀÀ); 3.236 (3H, s, ÀÀNCH3);
3.30–3.42 (1H, m, ÀÀNCH2); 3.382 (3H, s, ÀÀNCH3); 3.712 (1H, m,
ÀÀCH(CH3)ÀÀ); 3.850 (1H, m, ÀÀNCH2); 5.085 (1H, d, ÀÀCHOHÀÀ); 7.269
(1H, m, ar); 7.310 (2H, t, ar); 7.391 (2H, d, ar).
Ultraviolet (UV) spectra were recorded on a Cary 300 UV-vis double
beam spectrophotometer (Varian Pty, Mulgrave, AU).
Optical rotations were determined on a DIP370-JASCO digital polarimeter.
Conductivity experiments were performed on a Hanna conductimeter
HI-9932, equipped with a thermostating apparatus.
13C NMR, d(CDCl3) ppm: 13.42; 14.07; 22.63; 22.90; 26.33; 29.21;
29.30; 29.38; 29.46; 29.58; 29.61; 29.65; 31.87; 50.19; 51.12; 65.91; 71.37;
73.78; 127.43; 128.43; 128.86; 141.67. aD 5 135.1 (c 1.18, CH3OH).
Molar ellipticity (CH3OH, deg cm2 dmol21): 2175 (267 nm); 2222 (261
nm); 2158 (255 nm); 274 (250 nm, shoulder); 230 (244 nm, shoulder).
Preparation of Surfactants
(2)-(1R,2S)-N-dodecyl-N,N-dimethylephedrinium bromide,
1a. To a solution of 0.51 g (2.8 mmol) of (1R,2S)-(2)-N-methylephe-
drine in 4 ml of acetonitrile, 0.75 g (3.0 mmol) of 1-bromododecane were
added. The reaction mixture was kept under reflux (ꢀ10 days) and
monitored by TLC (CHCl3/MeOH 5 80/20). After removal of the sol-
vent under reduced pressure, the residue was washed with Et2O to give
the crude product. Crystallization by methanol/ether yielded 0.48 g
(40%) of a white powder (m.p. 104–1068C).
(1)-(1S,2S)-2-(dimethylamino)-1-phenylpropane-1,3-diol. 50 ml
(0.67 mol) of formaldehyde 40% w/v were added to 31.7 g (0.19 mol) of
(1)-(1S,2S)-2-amino-1-phenylpropane-1,3-diol and cooled in an ice bath.
After the addition of 25 ml (0.66 mol) of formic acid 99% under cooling,
the mixture was heated to reflux for 24 h. After cooling, the addition of
solid NaOH to the reaction mixture, saturated with NaCl, induced phase
separation. The aqueous layer was extracted twice with 50 ml of CHCl3.
The collected organic layers were washed with brine and dried on anhy-
drous Na2SO4. After removal of the solvent under reduced pressure, pu-
rification on silica gel (eluent, CH2Cl2/CH3OH 8:2) yielded 32 g (88%) of
a white solid (m.p. 60–618C).
1H NMR, d(CDCl3) ppm: 0.844 (3H, t, ÀÀ(CH2)11CH3); 1.136 (3H, d,
ÀÀCHCH3); 1.220 (16H, m, ÀÀ(CH2)8CH3); 1.300 (2H, m,
ÀÀN(CH2)2CH2ÀÀ); 1.668 (2H, m, ÀÀNCH2CH2ÀÀ); 3.270 (3H, s, ÀÀNCH3);
3.3–3.5 (1H, m, ÀÀNCH2); 3.430 (3H, s, ÀÀNCH3); 3.584 (1H, m,
ÀÀCH(CH3)ÀÀ); 3.793 (1H, m, ÀÀNCH2); 5.482 (1H, d, ÀÀOH); 5.730 (1H,
m, ÀÀCHOHÀÀ); 7.214 (1H, m, ar); 7.292 (2H, t, ar); 7.457 (2H, d, ar).
13C NMR, d(CDCl3) ppm: 7.26; 14.09; 22.65; 22.87; 26.37; 29.29; 29.38;
29.46; 29.57; 31.87; 49.48; 49.71; 64.26; 67.85; 72.85; 125.95; 127.57;
128.48; 141.17. aD5 CH3OH, 211.2 (c 0.738, CH3OH)
1H NMR, d(CDCl3) ppm: 1.8 (2H, bs, 2OH); 2.50 (6H, s, 2CH3); 2.65
(1H, m, 1CH); 3.45 (2H, m, 1CH2); 4.4 (1H, d, 1CH); 7.4 (5H, m, Ar).
a
D 5 145.98 (c 7.76, CHCl3).
Molar ellipticity (deg cm2 dmol21): 1629 (267 nm); 1792 (260 nm);
1550 (255 nm); 1283 (248 nm, shoulder); 1133 (242 nm, shoulder);
154 (236 nm, shoulder).
(1)-(1S,2S)-N-(1,3-dihydroxy-1-phenylpropan-2-yl)-N,N-dime-
thylhexadecan-1-ammonium bromide, 3. 0.55 g (2.8 mmol) of (1)-
(1S,2S)-2-(dimethylamino)-1-phenylpropane-1,3-diol were quaternized
Chirality DOI 10.1002/chir

80
DERACEMIZATION OF BILIRUBIN IN MICELLES
with 0.92 g (3.0 mmol) of 1-bromohexadecane according to the proce-
dure described above for 1a. Crystallization by methanol/ether yielded
0.81 g (58%) of a white powder (m.p. 96–988C).
TABLE 1. Aggregation parameters of investigated surfactants
Krafft
Krafft
cmc
1H NMR, d(CD3OD) ppm: 0.894 (3H, t, ÀÀ (CH2)11CH3); 1.285 (24H,
m, (CH2)12CH3); 1.406 (2H, m, N(CH2)2CH2ÀÀ); 1.884 (2H, m,
NCH2CH2ÀÀ); 3.29–3.37 (m, 1H, ÀÀCH2OH); 3.377 (3H, s, NCH3); 3.462
(3H, s, NCH3); 3.680 (1H, m, ÀÀCH(CH2OH)); 3.70–3.98 (3H, m,
ÀÀCH2OH 1 ÀÀNCH2); 5.373 (1H, d, ÀÀCHOHÀÀ); 7.33–7.47 (3H, m, ar);
7.500 (2H, d, ar).
Surfactant
point (8C)
temperature (8C)
(mM)
1a
15
30 (10 mM)
4.0 (258C)
33 (20 mM)
1b
2a
2b
3
30
<4
17
44
42
45
<4
23
0.4 (338C)
3.5 (258C)
0.21 (258C)
0.23 (508C)
–
13C NMR, d(CD3OD) ppm: 14.45; 23.72; 23.93; 27.54; 30.18; 30.45;
30.54; 30.64; 30.74; 30.77; 33.05; 51.81; 51.96; 58.93; 68.29; 72.79; 76.24;
128.51; 129.83; 129.92; 143.23. aD 5 141.18 (c 4.83, CH3OH).
Molar ellipticity (CH3OH, deg cm2 dmol21): 2146 (267 nm); 2186
(261 nm); 2118 (255 nm); 249 (250 nm); 213 (244 nm, shoulder).
52
4a
ꢀ50b
aA clear break in the conductivity on increasing surfactant concentration and
temperature was not observed.
bTemperature at which a 2 mM solution of 4 becomes clear.
(1)-(2R,3S)-N-(3-hydroxy-2-methyl-3,4-diphenylbutyl)-N,N-dime-
thyldodecan-1-ammonium bromide, 4. 0.80 g (2.8 mmol) of Chirald
(99%) were quaternized with 0.74 g (3.0 mmol) of 1-bromododecane
according to the procedure described above for 1a. Crystallization by
methanol/ether yielded 0.81 g (65%) of a white powder (m.p. 154–
1558C).
The aggregation parameters, namely cmc, Krafft point,
and temperature, were measured by conductivity experi-
ments. Results are summarized in Table 1.
The cmc of surfactant 4 is very low (most probably <1027
M) and could not be determined by conductivity measure-
ments, being the values of conductivity in the range of con-
centration of cmc close to that of water.
1H NMR, d(CD3OD) ppm: 0.894 (3H, t, ÀÀ (CH2)11CH3); 1.15 (3H, d,
ÀÀCHCH3); 1.285 (18H, m, (CH2)9CH3); 1.730 (2H, m, NCH2CH2ÀÀ);
2.311 (1H, m, ÀÀCHCH3); 2.651 (1H, dd, NCH2CHCH3); 2.960 (3H, s,
NCH3); 2.974 (3H, s, NCH3); 3.186 (2H, m, ÀÀNCH2); 3.245 (1H, d,
ÀÀCH2Ph); 3.467 (1H, d, ÀÀCH2Ph); 3.878 (1H, dd, NCH2CHCH3); 7.19–
7.347 (6H, m, ar); 7.358 (2H, t, ar); 7.540 (2H, dd, ar).
Chiral Recognition Experiments
13C NMR, d(CD3OD) ppm: 14.45; 18.58; 23.59; 23.72; 27.37; 30.19;
30.45; 30.54; 30.62; 30.73; 33.06; 37.97; 45.95; 51.22; 51.59; 65.98; 68.86;
The interaction between bilirubin and the aggregates
formed by the chiral cationic surfactants 1–4 was studied by
means of CD and UV–vis spectroscopies in water solution at
the spontaneous pH (pH 6.0–6.5). The experiments were car-
ried out on samples of 10 and 100 lM bilirubin, at a surfac-
tant concentration above the cmc to ensure the presence of
micellar aggregates in solution.
78.54; 127.47; 127.86; 128.18; 128.79;129.02; 132.17;138.11; 144.43. aD
14.77 (c 4.28, CH3OH).
5
Molar ellipticity (CH3OH, deg cm2 dmol21): 1275 (266 nm); 1455
(260 nm); 1433 (255 nm); 1342 (251 nm, shoulder); 1289 (248 nm,
shoulder); 1205 (243 nm, shoulder).
CD and UV–vis experiments on 10 and 100 lM bilirubin in
aqueous solutions of 20 mM 1a and 1 mM 1b were carried
out at 35 and 508C, respectively (i.e., above the Krafft temper-
ature of the surfactant solutions). The obtained spectra are
reported in Figures 1 and 2. The CD spectra show a bisig-
nate band in the region of absorbance of bilirubin. Interest-
ingly, the intensity and the sign of the observed bands
depend on the concentration conditions and on length of
alkyl chain, as observed previously in aggregates formed by
N-alkyl-N,N-dimethyl-N-(1-phenylethyl)ammonium bromides.30
The CD spectrum of 10 lM bilirubin in aggregates of 1a
shows a negative bisignate band of modest intensity,
whereas the spectrum of 100 lM bilirubin in aggregates of
the same surfactant show a slightly more intense positive
bisignate band (Fig. 1a). Both CD spectra were fully repro-
ducible and, although modest, the intensity of the observed
bands was of the same order of magnitude of that observed
in the presence of other chiral selectors.27
On the other hand, both the CD spectra of 10 and 100 lM
bilirubin in aggregates of 1b show a positive bisignate band
(Fig. 2a) more intense than those observed in the experi-
ments with 1a. As reported previously, the occurrence of a
bisignate band (an exciton couplet) for bilirubin, in a system
lacking additional chromophores absorbing in the same
region of the spectrum as bilirubin, can be reasonably
ascribed to the deracemization of its racemic mixture.27,28
Actually, it has to be pointed out that in the investigation of
the diastereomeric interactions involving transparent chiral
hosts and enantiomeric chromophores, a mirror-image rela-
tionship between the diasteromeric complexes is usually im-
plicitly assumed and the possibility of differences in their CD
Determination of Aggregative Parameters of Surfactants
Krafft point and Krafft temperature of surfactants 2–4 were measured
by conductivity experiments according to a described procedure.31,32 Note
that the Krafft point is defined as the temperature at which the critical
micelle concentration (CMC) versus temperature plot intersects the solu-
bility versus temperature plot whereas the Krafft temperature is the tem-
perature, above the cmc, at which the surfactant is completely soluble,
hence is always higher than the Krafft point.33 For the determination of
Krafft point and Krafft temperature, a clear solution of surfactant in bidis-
tilled water, at a concentration estimated above the cmc, was obtained by
heating. The solution was kept at 48C for 24 h to allow the precipitation of
surfactant. When precipitation occurred, the conductivity of the solution at
increasing temperature (at a rate of 0.58C/min) was measured under stir-
ring. In the plot conductivity versus temperature, the Krafft point was esti-
mated by the first break whereas the Krafft temperature by the second
one. When precipitation did not occur, the Krafft temperature was referred
to as ꢁ 48C and not measured.
The cmc of surfactants 2–4 were measured by conductivity experiments.
Sample preparation. Micellar solutions were prepared by dissolv-
ing the proper amounts of surfactant in bidistilled water. Samples for
UV–vis and CD investigations were prepared by adding the necessary
volume (ll) of 1 or 10 mM stock solutions of bilirubin in ultrapure
dimethyl sulfoxide (DMSO) (dried on molecular sieves) to the aqueous
surfactants solutions (<2%, v/v). Handling of bilirubin samples was per-
formed in dim light. Spectra were recorded within 5 min since sample
preparation in quartz cuvettes of different path length (0.1, 1 cm).
RESULTS AND DISCUSSION
Preparation and Characterization of Amphiphiles
All amphiphiles have been prepared by quaternization of
the corresponding amines.
Chirality DOI 10.1002/chir

SORRENTI ET AL.
81
Fig. 1. CD (a) and UV–vis (b) spectra of 10 lM (solid line) and 100 lM (dashed line) bilirubin in aqueous 20 mM 1a.
spectra are not considered. This approximation is generally
feasible, although, if the enantiomeric excess is very modest,
small chirooptical differences between the diasteromeric
complexes come into play and may lead to no vanishing CD
spectra also for racemic mixtures. The shape of such CD
spectra may either be ‘‘normal,’’ as expected for the enan-
tiomers, if the CD spectra of diastereomers differ only in
magnitude, or more complex, if the spectra are shifted rela-
tive to each other34 (see below).
it is reasonable to expect similar conformations of the pig-
ment at the two concentration conditions, with y ꢀ 1008, as
predicted by the exciton model.35,36 Hence, it can be safely
claimed that the inversion of the Cotton effect sign observed
as a function of concentration of bilirubin in aggregates of 1a
is due to an inversion of molecular chirality. Evidently, the dif-
ferent concentration conditions induce either a different orga-
nization of the aggregates or a different site of binding of bili-
rubin, thus yielding an opposite enantioselection.
Given these considerations, it could be argued that the CD
bands observed when bilirubin is in the presence of aggre-
gates of 1a result from the different intensity of the CD spec-
tra of diastereomeric complexes rather than from a diastereo-
meric excess due to deracemization. However, the fact that
the sign of the bands depend on the concentration condition
suggests deracemization, although very modest.
As a matter of fact, a reversal of the Cotton effect sign of
bilirubin can be observed, without an inversion of molecular
chirality, in response to conformational changes, switching
from values of the dihedral angle between the dipyrrinone
planes of bilirubin y < 1408 to y > 1408. However, such a con-
formational change is accompanied by an absorbance
increase of the higher energy transition of exciton splitting in
the UV spectrum.35,36 As the UV–vis spectra of bilirubin 10
and 100 lM (Fig. 1b) do not show strong differences, featur-
ing the two transitions of exciton splitting of similar intensity,
On the other hand, in aggregates of 1b the inversion of
the Cotton effect sign as a function of concentration of biliru-
bin was not observed (Fig. 2a). The analysis of UV–vis spec-
tra in Figure 2b suggests a slightly more folded conformation
of bilirubin in aggregates of 1b with respect to 1a.35,36 The
lower values of the molar extinction coefficient at the higher
bilirubin concentration observed in Figure 2b are like to be
due to the aggregation of the pigment inside or at the surface
of the micellar aggregates because of the saturation of the
micellar pseudophase (actually a marked Tyndall effect is
observed at this concentration). Thus, the different intensity
of the CD bands in Figure 2a can be reasonably ascribed to a
similar extent of deracemization.
CD and UV–vis experiments on 10 and 100 lM bilirubin in
aqueous solutions of 50 mM 2a and 10 mM 2b were carried
out at 258C (Figs. 3 and 4). The CD spectra show bands
whose intensity and shape depend on the length of the alkyl
Fig. 2. CD (a) and UV–vis (b) spectra of 10 lM (solid line) and 100 lM (dashed line) bilirubin in aqueous 1 mM 1b.
Chirality DOI 10.1002/chir

82
DERACEMIZATION OF BILIRUBIN IN MICELLES
Fig. 3. CD (a) and UV–vis (b) spectra of 10 lM (solid line) and 100 lM (dashed line) bilirubin in aqueous 50 mM 2a.
chain of surfactant and on bilirubin concentration. In the
case of surfactant 2a (Fig. 3a), the CD spectrum shows a
negative bisignate band whose intensity strongly depends on
bilirubin concentration, it being higher at the lower concen-
tration of bilirubin (10 lM). Also in this case, the UV–vis
spectra suggest a bilirubin folded conformation with y ꢀ
100835,36 and the variation of the intensity of the CD band
would suggest a different extent of deracemization modu-
lated by the concentration conditions.
On the other hand, the CD spectra of bilirubin in the pres-
ence of surfactant 2b (Fig. 4) show a more complex pattern
and the shape of the CD bands strongly suggests the super-
imposition of two intense exciton couplets with inverted sign,
similar magnitude and slightly different position.35 Their ori-
gin can have different explanations: (i) the bilirubin enan-
tiomers are bound to two different types of aggregates, prob-
ably induced by the addition of the chromophore itself;
(ii) bilirubin is bound to different sites of the aggregate fea-
turing opposite enantioselecting capabilities (e.g., the enan-
tiomeric conformations of bilirubin could be, respectively,
bound to the hydrophobic core of the micelles and to their
polar region); (iii) bilirubin is bound to different sites of the
aggregate that stabilize two nonenantiomeric conformations
of bilirubin, with different dihedral angles, featuring inverted
and slightly shifted CD spectra.
The first two hypotheses imply that different sites of bind-
ing (either different aggregates or regions within the same
aggregate) can deracemize bilirubin in opposite direction,
nearly at the same extent, and that the diasteromeric com-
plexes of the enantiopure aggregates with the enantiomers
of bilirubin, i.e., M-bilirubin/2b and P-bilirubin/2b, feature a
nonperfect mirror relationship of their CD spectra. There-
fore, the results would indicate that the micellar aggregates
are able to discriminate the enantiomers, without an overall
deracemization of their mixture.
On the other hand, according to the third hypothesis the
observed CD pattern does not imply any enantiodiscriminat-
ing process. However, the analysis of UV–vis spectra show
the two exciton components of similar intensity which rea-
sonably rules out the presence of nonenantiomeric bilirubin
conformations featuring inverted CD spectra.35,36
To shed some light on the two discriminating sites and on
the origin of the pattern of the observed CD bands, we car-
ried out an annealing experiment, namely CD and UV experi-
ment were carried on an aqueous solution of 10 lM bilirubin
and 10 mM 2b heated to 508C and then cooled back to 258C.
Noteworthy, by heating the sample, the complex CD band
evolved irreversibly to a normal bisignate band (a negative
couplet), as shown in Figure 5a where we reported the spec-
trum obtained at 258C, before heating, and the spectrum
Fig. 4. CD (a) and UV–vis (b) spectra of 10 lM (solid line) and 100 lM (dashed line) bilirubin in aqueous 10 mM 2b.
Chirality DOI 10.1002/chir

SORRENTI ET AL.
83
Fig. 5. CD (a) and UV–vis (b) spectra of 10 lM bilirubin in aqueous 10 mM 2b at 258C (solid line) and 508C (dashed line).
Fig. 6. CD (a) and UV–vis (b) spectra of 10 lM (solid line) and 100 lM (dashed line) bilirubin in aqueous 5 mM 3.
obtained at 508C. The spectrum obtained at 258C after cool-
ing back (not shown) was identical to that obtained at 508C.
On the other hand, the corresponding UV spectra obtained
at 25 and 508C show negligible differences (Fig. 5b). These
results strongly suggest that in the aggregates of 2b, under
kinetic control, bilirubin is bound to different sites within the
aggregates with opposite enantioselecting capabilities,
whereas under equilibrium conditions all the pigment
features the same site of binding and in this case the overall
result is deracemization.
Also in the experiments relative to the aggregates of 2a
and 2b, the different behavior of bilirubin indicates a role of
Fig. 7. CD (a) and UV–vis (b) spectra of 10 lM bilirubin in aqueous 0.1 mM 4.
Chirality DOI 10.1002/chir

84
DERACEMIZATION OF BILIRUBIN IN MICELLES
LITERATURE CITED
the hydrophobic portion of the surfactant, and hence of
the hydrophobic interactions, in the enantiodiscriminating
processes.
1. Walde P, Blo¨chliger E, Morigaki K. Circular dichroic properties of phos-
phatidylcholine micelles. Langmuir 1999;15:2346–2350.
The CD and UV–vis spectra of samples relative to different
concentration of bilirubin in the presence of the aggregates
of surfactant 3, at 558C, are reported in Figure 6. The CD
spectra show an intense positive bisignate band likely
suggesting a high extent of deracemization. The intensity of
the band is almost independent from the surfactant/bilirubin
ratio. However, an analysis of the UV–vis spectra shows that
the shape of the UV band depends on the concentration
conditions, thus suggesting different conformations of biliru-
bin at the different surfactant/bilirubin ratios. In particular,
at 10 lM bilirubin (surfactant/bilirubin ratio 500:1), the
lower energy transition of exciton splitting is more intense,
thus suggesting a slightly higher value of the dihedral angle
between the dipyrrinone chromophores. This would
correspond to a smaller molar ellipticity as calculated by the
exciton coupling theory.36 Thus, the almost equal intensity
observed in the CD spectra at the two different concentra-
tions of bilirubin would be associated with a different
extent of deracemization (i.e., higher at lower bilirubin
concentration).
2. Dobashi A, Hamada M. Molecular recognition with micellar and mi-
celle-like aggregates in aqueous media. J Chromatogr A 1997;780:179–
189.
3. Tickle D, George A, Jennings K, Camilleri P, Kirby AJ. A study of the
structure and chiral selectivity of micelles of two isomeric D-glucopyrano-
side-based surfactants. J Chem Soc Perkin Trans 2 1998:467–474.
4. Ceccacci F, Mancini G, Sferrazza A, Villani C. pH variation as the switch
for chiral recognition in a biomembrane model. J Am Chem Soc 2005;
127:13762–13763.
5. Ceccacci F, Giansanti L, Mancini G, Mencarelli P, Sorrenti A. Discrimina-
tion of the enantiomers of new biphenylic derivatives in chiral micellar
aggregates. New J Chem 2007;31:86–92.
6. Alzalamira A, Ceccacci F, Monti D, Levi Mortera S, Mancini G, Sorrenti
A, Venanzi M, Villani C. Discrimination of the enantiomers of biphenylic
derivatives in micellar aggregates formed by chiral amidic surfactants.
Tetrahedron Asymmetry 2007;18:1868–1876.
7. Borocci S, Erba M, Mancini G, Scipioni A. Deracemization of an axially
chiral biphenylic structure in chiral micellar aggregates. Langmuir 1998;
14:1960–1962.
8. Bella J, Borocci S, Mancini G. Recognition in organized aggregates
formed by a chiral amidic surfactant. Langmuir 1999;15:8025–8031.
9. Borocci S, Ceccacci F, Galantini L, Mancini G, Monti D, Scipioni A,
Venanzi M. Deracemization of an axially chiral biphenylic derivative as a
tool for investigating chiral recognition in self-assemblies. Chirality 2003;
15:441–447.
The CD and UV–vis spectra of samples relative to 10 lM
bilirubin in the presence of the aggregates of surfactant 4, at
528C, are reported in Figure 7. The CD spectrum shows an
intense positive bisignate band corresponding to a relatively
high extent of deracemization, the UV–vis spectrum indicat-
ing a conformation with y ꢀ 1008.
10. Nakagawa H, Onoda M, Masuoka Y, Yamada K-I. Effect of phosphatidyl-
choline vesicle size on chirality induction and chiral discrimination.
Chirality 2006;18:212–216.
11. Nakagawa H, Kobori Y, Yoshida M, Yamada K-I. Chiral recognition by
single bilayered phosphatidylcholine vesicles using [5]thiaheterohelicene
as a probe. Chem Commun 2001:2692–2693.
It is worth of note that only a modest bleaching of the pig-
ment was observed when the experiments were carried out
at high temperature, thus indicating that the aggregates pro-
tect bilirubin from degradation.26,35
12. Cruciani O, Borocci S, Lamanna R, Mancini G, Segre AL. Chiral recogni-
tion of dipeptides in phosphatidylcholine aggregates. Tetrahedron
Asymmetry 2006;17:2731–2737.
13. Arnett EM, Gold JM. Chiral aggregation phenomena. 4. A search for
stereospecific interactions between highly purified enantiomeric and
racemic dipalmitoyl phosphatidylcholines and other chiral surfactants in
monolayers, vesicles, and gels. J Am Chem Soc 1982;104:636–639.
CONCLUSIONS
We have investigated by CD experiments the deracemiza-
tion of bilirubin in chiral micellar aggregates with the aim of
exploring the expression of chirality within the aggregates.
The chiral information of the monomer can be translated into
opposite enantiodiscriminating capability of the aggregates
as a function of bilirubin concentration and of the length of
the alkyl chain of the surfactant, demonstrating the funda-
mental role of hydrophobic interactions in the transfer of the
chiral information from the monomers to the aggregates. It
is possible that the observed deracemization processes also
result from the interaction of functional groups of bilirubin
with the functional groups of the surfactants. Actually an
interaction with the head group functional groups could be
responsible for the kinetic trapping observed in aggregates
of 2b; however, as deracemization is observed only in aggre-
gating condition, it is clear that these interactions are medi-
ated by aggregation.
14. Lalitha S, Kumar AS, Stine KJ, Covey DF. Chirality in membranes: first
evidence that enantioselective interactions between cholesterol and cell
membrane lipids can be a determinant of membrane physical properties.
J Supramol Chem 2001;1:53–61.
15. Pathirana S, Neely WC, Myers LJ, Vodyanoy V. Chiral recognition of
odorants (1)- and (2)-carvone by phospholipid monolayers. J Am Chem
Soc 1992;114:1404–1405.
16. Alakoskela J-M, Sabatini K, Jiang X, Laitala V, Covey DF, Kinnunen PKJ.
Enantiospecific interactions between cholesterol and phospholipids.
Langmuir 2008;24:830–836.
17. Sorrenti A, Diociaiuti M, Corvaglia V, Chistolini P, Mancini G. Chiral
recognition of dipeptides in Langmuir monolayers. Tetrahedron
Asymmetry 2009;20:2737–2741.
18. Fuhrhop JH, Schnieder P, Rosenberg J, Boekema E. The chiral
bilayer effect stabilizes micellar fibers. J Am Chem Soc 1987;109:
3387–3390.
19. Uragami M, Miyake Y, Regen SL. Influence of headgroup chirality on the
mixing behavior of phosphatidylglycerol mimics in fluid bilayers. Lang-
muir 2000;16:3491–3496.
Interestingly, in some cases the deracemization of biliru-
bin in chiral surfactant aggregates was observed at high tem-
perature, where chiral recognition phenomena are, in gen-
eral, less efficient due to less tight interactions between the
interacting species.
20. Andreani R, Bombelli C, Borocci S, Lah J, Mancini G, Mencarelli P,
Vesnaver G, Villani C. New biphenylic derivatives: synthesis, character-
isation and enantiodiscrimination in chiral aggregates. Tetrahedron
Asymmetry 2004;15:987–994.
21. Bombelli C, Borocci S, Lupi F, Mancini G, Mannina L, Segre AL, Viel S.
Chiral recognition of dipeptides in a biomembrane model. J Am Chem
Soc 2004;126:13354–13362; and references therein.
ACKNOWLEDGMENTS
22. Navon G, Frank S, Kaplan D. A nuclear magnetic resonance study of
the conformation and the interconversion between the enantiomeric
conformers of bilirubin and mesobilirubin in solution. J Chem Soc Perkin
Trans2 1984:1145–1149.
The authors thank Prof. David A. Lightner for helpful dis-
cussion. This work was carried out in the frame of CM0703
‘‘Systems Chemistry’’ COST Action.
Chirality DOI 10.1002/chir

SORRENTI ET AL.
85
23. Blauer G. Optical activity of bile pigments. Isr J Chem 1983;23:201–
209.
30. Bombelli C, Bernardini C, Elemento G, Mancini G, Sorrenti A, Villani C.
Concentration as the switch for chiral recognition in biomembrane mod-
els. J Am Chem Soc 2008;130:2732–2733.
24. Blauer G. Complexes of bilirubin with proteins. Biochim Biophys Acta
1986;884:602–604.
ˇ
´
´
´
ˇ ´
31. Filipovic-Vincekovic N, Babic-Ivancic V, Smit I. Interactions in aqueous
systems containing sodium cholate/calcium chloride mixtures. J Colloid
Interface Sci 1996;177:646–657.
25. Lightner DA, Wijekoon WMD, Zhang MH. Understanding bilirubin
conformation and binding. Circular dichroism of human serum albu-
min complexes with bilirubin and its esters. J Biol Chem 1988;263:
16669–16676.
32. Bales BL, Benrraou M, Zana R. Krafft temperature and micelle ionization
of aqueous solutions of cesium dodecyl sulfate.
2002;106:9033–9035.
J Phys Chem B
26. McDonagh AF. Bile Pigments: bilatrienes and 5,15-biladienes. In:
Dolphin D, editor. The porphyrins, Vol. 6. New York: Academic Press;
1979. p 293–491.
33. Rubunch DN, Holland PM, editors. Surfactant science series—cationic
surfactants, physical chemistry, Vol.37. New York and Basel: Marcel
Dekker; 1991. p. 171–177.
27. Lightner DA, Gawronski JK, Wijekoon WMD. Complementarity and chi-
ral recognition: enantioselective complexation of bilirubin. J Am Chem
Soc 1987;109:6354–6362.
34. Krois D, Lehner H. Induced circular dichroism and chiral discrimination
of racemates revisited: bilirubins as illustrative examples. J Chem Soc
Perkin Trans 2 1995;489–494.
28. Lightner DA, Gawronski JK, Gawronska K. Conformational enantiomer-
ism in bilirubin. Selection by cyclodextrins. J Am Chem Soc 1985;107:
2456–2461.
35. Trull FR, Person RV, Lightner DA. Conformational analysis of
symmetric bilirubin analogs with varying length alkanoic acids. Enantio-
selectivity by human serum albumin. J Chem Soc Perkin Trans 2 1997;
1241–1250.
29. Ceccacci F, Giansanti L, Levi Mortera S, Mancini G, Sorrenti A, Villani C.
Enantiodiscrimination of bilirubin-IXa enantiomers in biomembrane
models: has chirality a role in bilirubin toxicity? Biorg Chem 2008;36:
252–254.
36. Person RV, Peterson BR, Lightner DA. Bilirubin conformational analysis
and circular dichroism. J Am Chem Soc 1994;116:42–59.
Chirality DOI 10.1002/chir