Solid state molecular motion in sucrose octapalmitate as studied by deuterium NMR spectroscopy

Article abstract of DOI:10.1016/S0009-3084(99)00114-0

Sucrose octapalmitate-d₁₁, d₂₄ and d₂₄₈ have been synthesized. Using ²H NMR T₁ and analyses of the temperature dependence of the lineshapes, a detailed description of the solid state molecular motional modes is presented. Activation energies for methyl and methylene group rotation in the fatty acyl chains have been determined. The sucrose moiety is found to be static on the solid state deuterium NMR timescale. Copyright (C) 2000 Elsevier Science Ireland Ltd.

Full text of DOI:10.1016/S0009-3084(99)00114-0

Chemistry and Physics of Lipids
04 (2000) 23–34
1
www.elsevier.com/locate/chemphyslip
Solid state molecular motion in sucrose octapalmitate as
studied by deuterium NMR spectroscopy
a,
a
b
G.W. Buchanan *, G. McManus , H.C. Jarrell
a
Department of Chemistry, Ottawa-Carleton Chemistry Institute, Carleton Uni6ersity, Carleton, Ottawa, Ont., Canada, K1S 5B6
b
Di6ision of Biological Sciences, National Research Council of Canada, Carleton, Ottawa, Ont., Canada, K1A 0R6
Received 7 April 1999; received in revised form 1 September 1999; accepted 1 September 1999
Abstract
Sucrose octapalmitate-d , d and d248 have been synthesized. Using H NMR T and analyses of the temperature
2
11
24
1
dependence of the lineshapes, a detailed description of the solid state molecular motional modes is presented.
Activation energies for methyl and methylene group rotation in the fatty acyl chains have been determined. The
sucrose moiety is found to be static on the solid state deuterium NMR timescale. © 2000 Elsevier Science Ireland Ltd.
All rights reserved.
Keywords: Sucrose octapalmitate; Solid state molecular motion; Deuterium NMR spectroscopy
1
. Introduction
Brandenburg et al., 1993, 1998). Also certain
members of this class of molecules function as
outer membrane glycolipids (Jarrell et al., 1987;
Carrier et al., 1989; Renou et al., 1989) in cells.
The present study examines some members of
the class of materials known as sucrose polyesters
(SPE). In recent years, extensive research on SPE
has led to the discovery of a large range of
applications centered around SPE’s physicochemi-
cal properties (Jandacek and Webb, 1978; Akoh
and Swanson, 1989) as they relate to its ability to
serve as a lipid-based fat substitute. SPE in the
form of a mixture mainly composed of sucrose
octaesters with a low amount of sucrose hep-
taesters of palmitic, stearic, oleic, and linoleic
Lipidic carbohydrate polyesters have been the
focus of a broad spectrum of scientific research.
Structural variables in these systems are the na-
ture of the esterified fatty acyl chains, the degree
of esterification and the type of the carbohydrate
moiety. Some of the research interest in these
molecules has been based on their properties as
bacterial lipopolysaccharide (LPS) endotoxins
during bacterial infections (Rietschel et al., 1985;
*
Corresponding author. Tel.: +1-613-520-3841; fax:+1-
13-520-3749.
E-mail address:
Buchanan)
6
gwbuchan@ccs.carleton.ca
(G.W.
0
009-3084/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved.
PII: S0009-3084(99)00114-0

2
4
G.W. Buchanan et al. / Chemistry and Physics of Lipids 104 (2000) 23–34
acids is known as (Olestra*), a low-calorie fat
substitute (Akoh, 1995). The ability of SPE to
behave as non-caloric lipid-like molecule is due to
this high degree of esterification with long chain
fatty acids which prevents the hydrolytic action of
pancreatic lipase. Thus, esters of this type are not
absorbed by the small intestine and pass through
the human body unmetabolized (Matteson and
Volpenhein, 1972).
With respect to beneficial applications of these
systems, it has been demonstrated that SPE is
effective as a non-absorbable lipophilic binding
agent that can reduce intestinal absorption of
certain lipophilic molecules. The presence of SPE
as substitute for dietary triacylglycerides has been
shown to decrease the absorption of LDL choles-
terol (Crouse and Grundy, 1979; Jandacek et al.,
1990), in the present report we have focused on
sucrose octapalmitate (SOP) as a representative
compound of SPE. Our hypothesis is that there is
likely to be a substantial interaction between the
fat-soluble vitamins and SOP in the hydrophobic
portion of the molecule — i.e. in the long chain
fatty acid region. This should influence dramati-
cally the motional characteristics in these side
chains of the SOP system. Since these interactions
in vivo are not strictly solution phase ones, we
have decided to employ solid phase techniques to
examine them. Probably the most powerful
method to study motional details of molecules in
2
the solid state is H NMR. Accordingly, we have
synthesized three different isotopically labeled
forms of SOP. In the first compound, all eight
palmitoyl chains are perdeuterated, giving rise to
SOP-d248. In the second molecule, 11-deuterium
atoms are located on the disaccharide rings giving
1
990). Also, SPE has been used to detoxify
lipophilic environmental contaminants such as
organochlorine pesticides and other halogenated
hydrocarbons (Mutter et al., 1988). A more recent
application of SPE is as an oral contrast agent for
human abdominal MRI studies (Ballinger et al.,
rise to SOP-d . The third molecule contains all
11
eight palmitoyl methyl groups deuterated, thus
producing SOP-d24
The present work uses H solid state NMR T
1
.
2
1
991).
There are, however, some negative aspects of
relaxation times and lineshape analysis to probe
the dynamics and structural properties of the
disaccharide rings and of the saturated fatty acyl
the use of SPE as a non-caloric fat substitute. A
source of concern with the use of SPE is that it
absorbs the fat-soluble vitamins (Crouse and
Grundy, 1979; Mellies et al., 1985), thus depleting
the availability of these important micro-nutri-
ents. As a result, food containing SPE must be
supplemented with these vitamins (Akoh, 1995).
Of further interest is the potential of SPE to
interact with and insert into biological mem-
branes. In the latter case, SPE may exhibit simi-
larities to bacterial lipidic carbohydrate polyesters
such as lipid A.
In view of the aforementioned importance of
SPE, we have begun a study of its interactions
with a variety of fat-soluble materials at the
molecular level, as well as studies of the insertion
of SPE into biological membranes. Initially, how-
ever, we are investigating the properties of indi-
vidual SPE components in order to determine
their detailed motional characteristics.
2
chains in polycrystalline SOP. H NMR offers
two principle advantages for studying molecular
2
dynamics. Firstly, H NMR line shapes and relax-
ation are dominated by the intramolecular
quadrupolar interaction. The interpretation of the
experimental results is usually not complicated by
consideration of other intra or intermolecular spin
2
interactions. Secondly, H NMR affords a tech-
nique that permits molecular dynamics to be
0
10
probed over a broad dynamic range (10 –10
Hz). In order to interpret the experimental data
and to correlate them to a molecular motion, a
model of the motion must be formulated. The
motional model can be built by taking into con-
sideration the possible molecular motions that are
consistent with the known molecular structure.
Through computer simulations, the analysis of the
experimental line shapes for fast and slow mo-
tional modes provide insight into detailed aspects
of the molecular dynamics.
Since a substantial amount of the commercial
product Olestra is present as a sucrose octaester
of long saturated acyl chains (Jandacek et al.,

G.W. Buchanan et al. / Chemistry and Physics of Lipids 104 (2000) 23–34
25
2
. Experimental procedure
the resulting heterogeneous mixture was stirred
overnight. The next day the methanol was re-
moved by gravity filtration. The precipitate was
dissolved in a minimum amount of hexane and
washed with methanol until the methanol layer
was colorless. After drying over anhydrous
sodium sulfate and filtration, the hexane was re-
moved from the crude product by rotary evapora-
tion. Subsequent column chromatography (silica
gel with hexane:ether:acetic acid (70:30:1) as elu-
tion solvent, gave as the first eluted product,
sucrose octapalmitate-d248 (SOP-d248), 490 mg
2
.1. Synthesis of methyl hexadecanoate-d31
Methyl hexadecanoate-d31 was synthesized
starting with hexadecanoic-d31 acid (lot No,
A183AP14) purchased from CDN isotopes Ltd.
Hexadecanoic-d31 acid (2.00 g, 7.38 mmol) was
refluxed for 1 h with excess methanol and 1 ml of
hydrochloric acid. After cooling to ambient tem-
perature, the product was extracted into
dichloromethane (3×50 ml). The organic extracts
were washed with water followed by a 20%
aqueous NaCl solution and then dried over anhy-
drous sodium sulfate. After filtration, the solvent
was removed by rotary evaporation to give crude
methyl hexadecanoate-d . Further solvent re-
7
(24% yield), m.p. 43.8°C from DSC, literature
47–48°C. Thermal decomposition temperature
was 300°C from thermal gravimetric analysis.
2.3. Synthesis of sucrose octapalmitate-d24
(SOP-d )
31
moval under vacuum for 1 h gave (1.84 g), 92%
yield of the ester which was used without further
purification in the next step.
24
SOP-d24 was prepared via an identical proce-
dure to that for SOP-d248, except that methyl-
2
.2. Synthesis of deuterated sucrose octapalmitate
palmitate-d was employed as a starting material.
3
(SOP-d248
)
For SOP-d , mp is 47–48°C.
24
This synthesis was based on a procedure devel-
2.4. Synthesis of C-deuterated sucrose (Sd )
11
oped by Akoh et al. (Akoh, 1990). A 100 ml
three-necked round bottom flask equipped with a
magnetic stirrer, thermometer, fractional distilla-
tion column connected to a condenser, with a
vacuum take-off line leading to a liquid nitrogen
cold trap, manometer, and a vacuum pump.
Methyl hexadecanoate-d31 (1.84 g, 6.78 mmol)
was added to the reaction flask and flushed with
This synthesis was based on a procedure devel-
oped by Koch et al. (Koch and Stuart, 1978)
Sucrose (3.0 g, 8.77 mmol) was dissolved in 30 ml
of D O and rotoevaporated to dryness to give
2
sucrose OD . Raney Nickel (20 ml settled volume)
8
was deuterated by exchange by first centrifuging
the aqueous metal dispersion and decanting the
supernatant H O. Subsequently, 10 ml D O was
dry N gas for 30 min. Sucrose octaacetate (0.55
2
2
2
g, 0.81 mmol) from Aldrich was added and the
system was flushed for an additional 15 min.
Sodium metal (0.05 g, 2% of the reactants w/w)
was added, the reaction flask was then immersed
in a preheated silicon oil bath at 80°C. The reac-
tion mixture was gradually heated to 110°C and
vacuum applied to obtain a pressure of about 5
added and the solution was vigorously mixed on a
vortex followed by centrifugation and decantation
of the D O. This procedure was repeated three
2
times. D O (80 ml) was added to a 250 ml round
2
bottom flask containing 3.0 g dry sucrose-OD8
and the Raney Nickel. The reaction mixture was
refluxed at 100°C for 15 h. After cooling, the
reaction mixture was decanted by gravity through
a paper filter to remove the Raney Nickel. The
mm Hg. N gas flow was input every 20 min for a
2
1
min period. The reaction was carried out under
these conditions for 2 h at which point heat and
pressure were removed. After cooling to room
temperature, 1 ml of acetic acid in 100 ml hexane
was added to the solution. Subsequent addition of
solid was washed three times with 40 ml of H O
2
and the filtrate was then passed through an Am-
berite IR-120 cation-exchange resin to eliminate
any suspended Raney Nickel. Rotary evaporation
was used to remove the aqueous solvent to give
1
l of methanol caused a precipitate to form and

2
6
G.W. Buchanan et al. / Chemistry and Physics of Lipids 104 (2000) 23–34
2
.62 g (7.39 mmol), 84% yield of a clear syrup of
2.8. Computer simulations
C-deuterated sucrose. The identity of the product
1
was confirmed by H NMR spectroscopy.
Simulations were performed on a Silcon Graph-
ics 4D/280S computer using a modified line shape
simulation program (Wittebort et al., 1987) based
on the general formalism of Torchia and Szabo
2
.5. Synthesis of C-deuterated sucrose
octaactetate (SOAc-d )
11
(
Torchia and Szabo, 1982). Simulated line shapes
were corrected for the finite width of the experi-
mental 90° pulse in the quadrupolar echo se-
C-deuterated sucrose (2.62 g, 7.39 mmol) was
dissolved in 50 ml of dry pyridine and 5 ml of
acetic anhydride was added. The solution was
stirred at room temperature for 11 h. The reaction
was quenched by pouring it into ice water (100
ml). Extraction with CH Cl (3×50 ml) was car-
ried out, followed by drying over anhydrous
sodium sulfate. Solvent removal via rotary evapo-
ration gave 3.7 g of a clear syrup. Thin layer
chromatography showed the presence of a mix-
ture of six compounds. This mixture was sub-
jected to column chromatography (silica gel,
quence (Bloom et al., 1980). The CD and CD3
2
line shapes were simulated separately prior to
adding them to each other. In addition to
parameters associated with details of the motional
model (see below), input parameters for the simu-
lations were as follows: line-broadening 1.5 kHz,
half width of plot 2.0×105 kHz, spectrometer
frequency 61.1 MHz and pulse interval 35 ms.
2
2
3
. Results
1
30–270 mesh) with 5:1 ether: light petroleum
ether as eluant. The fourth eluted product was
C-deuterated sucrose octaacetate (300 mg, 0.436
mmol) (59% yield).
2
H spectra of SOP deuterated in each of the
palmitoyl chains are shown as a function of tem-
perature in Fig. 1. At low temperature (135 K) the
spectra exhibit two sub-spectral features corre-
sponding to H–C bonds associated with static
methylene residues with quadrupolar splitting of
2
2
.6. Synthesis of C-deuterated sucrose
octapalmitate (SOP-Sd )
11
1
26 kHz and to methyl groups undergoing rapid
reorientation about their C axis with quadrupo-
3
C-deuterated sucrose octaactetate (300 mg,
.436 mmol) was reacted for 2 h with 0.95 g
3.518 mmol) of methylhexadecanoate in the same
lar splitting of 40.1 kHz. As the temperature is
0
(
2
elevated both the H spectral line shape and inte-
grated intensity changes indicate the onset of
chain motion. At 298 K, the spectral lineshape is
reminiscent of a motionally averaged quadrupolar
interaction characterized by an asymmetry
parameter p of approximately 1 and full spectral
intensity. The latter suggests that molecular mo-
tion is occurring on a timescale that is fast relative
to the reciprocal of the residual quadrupolar in-
teraction. At temperatures intermediate between
200 and 298 K, molecular motion is on the
timescale of the quadrupolar interaction. Such
behavior has been seen previously for structurally
related molecules having long polymethylene
chains such as phospholipids (Huang et al., 1980;
manner as previously described for the synthesis
of SOP-d248. From this reaction, 0.124 g (12.6%
yield) of pure SOP-Sd11 (mp 46–47°C) was
isolated.
2
.7. NMR
2
H NMR spectra were obtained at 61.1 MHz
on a VARIAN Unity 400 NMR spectrometer.
Spectra were acquired using the quadrupolar echo
sequence as described previously (Fenske et al.,
1
991) using a p/2 pulse length of 3.5 ms (5 mm
solenoid coil). Samples were enclosed in a glass
jacket in which the temperature was regulated to
2
Speyer et al., 1989). In such cases, H lineshape
9
1°C. The recycling delay time was set to be
changes were consistent with a two-site jump
approximately equal to 5T1.
trans-gauche isomerization for the methylene

G.W. Buchanan et al. / Chemistry and Physics of Lipids 104 (2000) 23–34
27
Fig. 1. Experimental (a) and simulated (b) 2H NMR spectra of SOP-d248 recorded as a function of temperature. Simulation jump
rate and trans-gauche population ratio parameters are presented in Table 1, with their respective error values. The spectral
singularities DVZZ, DVYY, and DVXX described in the text are labeled on the bottom right hand side spectrum.

2
8
G.W. Buchanan et al. / Chemistry and Physics of Lipids 104 (2000) 23–34
groups. Thus the spectral changes associated
with SOP-d248 were analyzed in terms of a simi-
lar motional model.
the acyl chain. In order to define the orientation
VXX and VYY the separation between the spectral
singularities DVYY, DVXX and DVZZ were plotted
as a function of temperature in Fig. 3. It is quite
clear that the DVYY singularity for the CD2
groups remains nearly constant throughout the
entire temperature range. This indicates that the
VYY tensor must lie parallel with the axis of the
molecular motion and perpendicular to the CD2
3
.1. Motional model
Fig. 2 depicts the well known intramolecular
gauche-trans isomerization of polymethylene
chains with the principal axis system of the elec-
tric field gradient tensor VZZ, VYY, and VXX hav-
ing the polar angles q and ƒ relative to the static
bond plane, leaving VXX bisecting the CD bond
2
plane. Gauche-(+) or gauche-(−) isomerisation
around a C–C bond (Fig. 2) displaces the CD2
by an angle of 120° about the chain long molec-
magnetic field (B ) direction. In the treatment of
0
the gauche-trans isomerization model the gauche-
(
+) and gauche-(−) conformations are consid-
ular axis. For the CD groups, the DV
singu-
3
YY
ered as having an equal probability of
occurrence. Thus, the conformational jump can
be taken as a two-site process which known to
generate an p=1 spectrum with spectral breadth
of ca. 120 kHz (Huang et al., 1980; Speyer et al.,
larity shows a similar behavior up to about 217
K. Above this temperature, however, DVYY starts
to deviate from linearity due to the onset of
reorientational motion of the terminal CD3
groups. To analyze the powder lineshape, spectra
spanning the temperature range of 298–135 K
were simulated using the model shown in Fig. 2
and a lineshape simulation program developed
by Wittebort et al. (Wittebort et al., 1987) based
on the general formalism of Torchia and Szabo
(Torchia and Szabo, 1982).
1
989). For an acyl chain in the all trans confor-
mation the principle axis VZZ is taken to be
parallel to the C–D bond and the VYY tensor
component is perpendicular to the CD2 plane
and parallel to the axis of molecular motion
which is chosen to be the long molecular axis of
Fig. 2. Model of a twofold jump isomerization in a polymethene chain segment: (a) trans conformation; and (b) gauche
conformation. Also shown are the static magnetic field B , and the polar angles q and ƒ made by the magnetic field vector in the
0
principal axis system of the electric field gradient tensor VZZ, VYY, and VXX
.

G.W. Buchanan et al. / Chemistry and Physics of Lipids 104 (2000) 23–34
29
polar coupling constant of 171 kHz which de-
scribed a two site exchange over a tetrahedral
geometry. In order to effectively simulate the
methyl groups in terms of rates of trans-gauche
isomerization while accounting for the rapid rota-
tion about a three fold axis and the residual
motion present at the chains ends, a variety of
quadrupolar coupling constants were used. The
results shown in Fig. 1b represent the best fit to
the experimental data and were obtained using a
quadrupolar coupling constant of 50 kHz instead
of the predicted 57 kHz. The lineshape simulated
spectra for the methyl and the methylene groups
were combined after using a fixed 3:28 deuterium
weighting ratio, corresponding to the number of
methyl and methylene sites, respectively, through-
out the entire temperature range. Inspection of
Fig. 1 shows that a simple trans-gauche jump
motional model adequately reproduces both line
shapes and spectral intensity changes at all tem-
peratures. The spectral simulations indicate that
varying both the isomerization rate and popula-
tions can satisfactorily reproduce the experimental
temperature dependent lineshape.
3.3. Analysis of experimental lineshape of the
acyl chains
2
The H NMR experimental spectra and com-
puter simulations of SOP-d248 as a function of
temperature are shown in Fig. 1. The experimen-
tal powder line shapes for varying temperatures
2
span the three characteristic H NMR motional
rates from which structural and motional infor-
mation can be obtained. The experimental line
shape at 298 K is in the fast motion limit where
the molecular motion occurs on a timescale much
shorter than the reciprocal line width of the static
Fig. 3. Plot of the frequency differences, (DV ) between the
ii
2
features of the solid state H NMR spectra of Fig. 1a, as a
function of temperature. (a) CD2 and (b) CD . (ꢀ, VZZ; ꢁ,
3
DVYY; ꢂ, DVXX). Uncertainties for DVii are 91 kHz.
7
−1
spectrum (\10 s ). The line shape has axial
asymmetry (p=1) and a spectrum breadth ca 120
kHz suggesting the motional process responsible
for the collapse is characteristic of a two site
exchange between equally populated sites. As ex-
pected at lower temperature, the powder line
shape seen at 135 K has axial symmetry (p=0)
which is consistent with motion occurring on a
timescale much longer than the reciprocal line
3
.2. Computer simulation
The powder spectral line shapes for CD and
2
CD were individually simulated as two-site jump
3
exchanges between non-equally populated sites
with the C–D bond maintaining the tetrahedral
angle over the course of the motion. The
methylene groups were simulated using a quadru-
−
4
−1
width of the static spectrum (B10
s
). Also

3
0
G.W. Buchanan et al. / Chemistry and Physics of Lipids 104 (2000) 23–34
noteworthy, is the reduction of the average
quadrupolar splitting of the methyl groups (40.1
kHz) which under rapid rotation about a C3
axis would be expected to have an average
quadrupolar splitting of 42.7 kHz (Batchelder et
al., 1983). This discrepancy is consistent with a
further motional averaging mechanism operating
at the acyl chain ends which has been previously
described for various methyl bearing compounds
(Batchelder et al., 1983; Taylor et al., 1983) as
either a slight departure from the tetrahedral ge-
ometry or as the fast torsion fluctuation of the
penultimate C–C bond of the –CD CD3
2
groups. Batchelder et al. (Batchelder et al.,
1
983) have shown that the former and latter
description of polycrystalline methyl reorienta-
tion could be differentiated by a simple analysis
of the observed average quadrupolar splitting as
a function of temperature. Inspection of Fig. 3
shows that there is a significant reduction of the
values of the observed average quadrupolar
splitting as temperature increases indicating that
the additional motional averaging mechanism
operating at the acyl chain ends is due to large-
amplitude asymmetric reorientational motion of
the methyl C axis.
3
3.4. Analysis of chain end packing
Fig. 4 shows a graphical representation of the
simulation values from Table 1 and represents a
comparative analysis of CD2 and CD3 motion
and energetics as a function of temperature. Up
to about 217 K, the CD groups have a higher
2
jump rate and gauche defect population than do
the CD . Then above 217 K the situation is
3
reversed. Thus it appears that below 217 K
there is insufficient thermal energy to promote
large amplitude reorientation of the chain ends.
It is to be noted that only by using different
jump rates for CD and CD groups as a func-
2
3
tion of temperature was it possible to simulate
2
the experimentally observed H solid state NMR
spectra.
One possible explanation for the decreased de-
gree of motional freedom in the CD3 groups
versus the CD is the existence of intermolecular
2
intercalation in the long chains. This would pro-
duce increased steric hindrance, especially at the
chain termini, thereby reducing the degrees of
motional freedom for the CD3 moieties. Some
X-ray crystallographic evidence for this phe-
nomenon in long chain hydrocarbons has been
found (Mnyukh, 1960).
Fig. 4. Comparative representation of Table 1 methylene and
methyl group simulation parameters as a function of tempera-
ture. (a) Jump rates. (b) Trans-gauche isomerization popula-
tion. Jump rate error values are indicated in Table 1. ꢂ, CD3;
ꢁ
, CD2.

G.W. Buchanan et al. / Chemistry and Physics of Lipids 104 (2000) 23–34
31
Table 1
Parameters used for the temperature dependent 2H NMR lineshape simulations of SOP-d248 represented in Fig. 1b
Temperature (K) CD2 jump rate (s−1)
Population trans/gauchea
CD3 jump rate (s−1
)
Population trans/gauchea
7
298
250
225
200
150
135
1.0(92)×10
3.0(91)×10
0.50/0.50
0.68/0.32
0.78/0.22
0.80/0.20
0.94/0.06
0.99/0.01
1.0(92)×107
1.0(91)×106
3.0(90.5)×105
1.0(90.5)×105
5.0(91)×103
1.0(92)×103
0.50/0.50
0.64/0.36
0.74/0.26
0.85/0.15
0.95/0.05
0.99/0.01
6
6
5
1.0(90.5)×10
5.0(90.5)×10
4
1.0(91)×10
3
1.0(92)×10
a
Uncertainties in populations are 90.01.
3
.5. Sucrose motion
While a gauche-trans segmental motion of the
patterns were found to be identical to the SOP-
d248 CD group powder patterns measured over
3
the same temperature range. From the simulated
jump rates as a function of temperature (Fig. 6a)
we have measured the activation energies for the
CD and CD motion using the Arrhenius equa-
2
acyl chains adequately describes the H lineshapes
for SOP-d248, the possibility of other molecular
motions must also be considered. Considering the
molecular structure similarities of SOP with li-
pidic molecules, other distinct motional modes
could be postulated, namely axial diffusion of the
acyl chains, rotation of the entire molecule about
an axis or conformational fluctuations of the dis-
accharide moiety. These motions, occurring either
independently or simultaneously, would be ob-
2
3
tion. Fig. 6b shows the plot of SOP-d24 CD T
3
1z
as a function of temperature from which an acti-
vation energy was measured using the Arrhenius
equation fitted to the high temperature linear part
of the curve. Results are shown in Table 2. Acti-
vation energies obtained were found to be in good
agreement with those reported for trans–gauche
isomerization or ‘kink defect’ in solid long chain
alkanes (Stohrer and Noack, 1977), and those
2
servable as a temperature-dependent H NMR
lineshape change of SOP-d11 where the deuterons
are located on the disaccharide moiety. H spectra
2
calculated for fast C axial reorientation of methyl
3
of SOP-d11 were acquired over the same tempera-
ture range as was used for SOP-d248. Over the
entire temperature range the resulting powder
lineshapes did not change and were similar to that
shown in Fig. 5 (SOP-d11 at 300 K). The quadru-
polar splitting of 123 kHz is consistent with a
rigid disaccharide moiety with a quadrupolar cou-
pling constant of 164 kHz (Renou et al., 1989).
The results confirm the absence of conformational
exchange about the glycosidic bond, ring pucker-
ing of disaccharide, as well as overall molecular
groups in alkanes (Smith and Karplus, 1992). The
T1z minimum occurs at 150 K. Using a spectrom-
eter frequency of 61 MHz, this translates into a
correlation time of 1.6 ns, a value that is approxi-
mately an order of magnitude slower than that
found for hydrated phospholipids.
−
5
reorientation on a timescale of 10
s.
3
.6. Temperature dependence of T1
2
The H spin lattice relaxation times, T , at the
1z
i=90° spectral singularity associated with the
deuteromethyl residues of SOP-d24 were measured
as a function of temperature using the inversion-
recovery technique. The SOP-d24 CD3 powder
2
Fig. 5. H solid state NMR of SOP-d at 300 K. The powder
1
1
pattern is characteristic of a rigid structure.

3
2
G.W. Buchanan et al. / Chemistry and Physics of Lipids 104 (2000) 23–34
comparison with results of structurally related
compounds will certainly help assert our pro-
posed molecular structural and dynamics de-
scription of SOP.
First, a conformational dynamic representa-
tion of the acyl chains can be elaborated on the
basis of activation energies. Results show that
our experimentally determined activation energy
values are in good agreement with activation en-
ergies previously found for methylene trans-
gauche isomerisation and for methyl group
reorientation measured for alkanes (Stohrer and
Noack, 1977; Smith and Karplus, 1992). This
indicates that the motional model described
herein can be quite well justified based on the
energetics of the system.
Secondly, additional information about the
acyl chain motion and packing can be gained
through comparison of the literature results on
structurally related compounds such as phospho-
lipids (Huang et al., 1980; Speyer et al., 1989),
long alkane chains (Mnyukh, 1960; Stohrer and
Noack, 1977; Taylor et al., 1983) and triacyglyc-
erol (Jensen and Mabis, 1966). The comparison
can be made based on the fact that these com-
pounds have a molecular structural similarity to
SOP in that they possess long fatty acyl chains.
2
It was shown through H solid state NMR that
for the above mentioned compounds the acyl
chain motions could be characterized as ether a
trans-gauche isomerisation, axial diffusion and
or translational motion. The observation of ax-
Table 2
Comparison of experimental and literature values activation
energy (Ea) for methylene and methyl groups found in
aliphatic alkanes
Fig. 6. Representations of Arrhenius plots of (a) the lineshape
simulation jump rate values from Table 2 plotted as a function
of temperature, (b) Natural logarithm (ln) of SOP-d24, CD3,
T1z plotted as a function of temperature. SOP-d24, CD , T
Molecular mo- E , experimental E , literature values
a
a
tion
(kJ/mol)
(kJ/mol)
3
1z
were measured from i=90° spectral singularities. ꢂ, CD3;
, CD2.
CD2 trans-
gauche
18.8a
17.9a
20.1 (Stohrer and
Noack, 1977)
ꢁ
CD3 trans-
gauche
4
. Discussion
CD3 reorienta- 11.3b
tion
12.5 (Smith and
Karplus, 1992)
From our NMR results alone a relatively
good description of SOP molecular structure
and dynamics can be inferred. However, the
a
Calculated from jump rates with error of 915%
Calculated from T1z slope with error of 910%
b

G.W. Buchanan et al. / Chemistry and Physics of Lipids 104 (2000) 23–34
33
ial asymmetry (p=1) spectra at 298 K is charac-
teristic of a motional process proceeding with a
two-site exchange between equally populated sites
with the C–D bond subtending the tetrahedral
chain ends (Barnes, 1974; Huang et al., 1980).
This effect could be a consequence of having a
high degree of order in the packing of the acyl
chain generating strong restrictions angle over the
course of the motion. Thus, based on the knowl-
edge that axial diffusion and translation motion
have more than two-site exchange in their motion
it is clear that the only possible motional mecha-
nism operating for the acyl chains of SOP is a
trans-gauche isomerisation. These observations,
combined with the results of the lineshape simu-
lation, provide further evidence of the validity of
the proposed trans-gauche isomerisation model.
Thirdly, it was shown that the disaccharide
moiety is motionless on the NMR time scale and
Based on the above discussion we can now
elaborate an accurate description of SOP molecu-
lar structure in which all eight acyl chains have
their long axes closely packed and parallel in
both inter- and intramolecular fashions. This
packing generates strong motional restriction on
the acyl chains which can be described as extend-
ing both above and below the plane of a rigid
disaccharide unit. This type of molecular struc-
tural description was previously implied by X-ray
powder diffraction (Jandacek and Webb, 1978)
2
which is now corroborated by our H NMR
results. Furthermore, a characterization of SOP
dynamics is now available for the acyl chain
as well as for the disaccharide unit, which can
be used as a template for structural and dyn-
amical analysis of structurally similar mole-
cules, such as for bacterial lipidic carbohydrate
polyesters.
that simulation of the acyl chain CD spectra can
2
be well described by a simple trans-gauche jump
with a single trans-gauche ratio without any axial
diffusion of the acyl chain, it would appear that
the local acyl chain environment is not signifi-
5
. Conclusions
The present molecular motional model derived
cantly perturbing this ratio at each CD site and
2
from NMR data is the first one available in the
literature for solid sucrose octapalmitate. Com-
bined with the previously determined X-ray pow-
der diffraction results (Jandacek and Webb, 1978)
a detailed molecular dynamic and structural de-
scription of SOP-d248 has now been made possi-
ble. Future endeavors in this field will involve the
study of the influence of various fat soluble com-
pounds (such as Vitamins A and E) on the de-
that there is no large distribution of conformers
arising from attachment of the acyl chains to the
eight sucrose positions. That is, there does not
appear to be a substantial local conformational
effect at each site, except for the fast reorienta-
tion of the chain ends. This effect could be a
consequence of having a high degree of order in
the packing of the acyl chain generating strong
restrictions on mobility. Such a restriction may
be attributed to both inter- and intramolecular
2
tailed H NMR lineshapes of the deuterated SOP
systems examined in this work.
interaction between adjacent CD groups. X-ray
2
crystal structure determination of triacylglycerol
(
Jensen and Mabis, 1966) has shown that similar
Acknowledgements
interactions occur between inter- and intramolec-
ular CD groups due to a close molecular pack-
2
The authors gratefully acknowledge Dr John
Katsaras for the gift of methyl deuterated
ing of triacylglycerol molecules in which the acyl
chains axes are aligned parallel to the long repeat
distance of the lattice, extending both above and
below, from a perpendicular glycerol backbone.
In view of structural and physical similarities of
SOP with triacylglycerol it would be reasonable
to assume that SOP would have similar molecular
packing.
palmitic acid-d . G. McManus thanks Dr David
3
J. Siminovitch for several stimulating conversa-
tions. G.W. Buchanan gratefully acknowledges
continuing financial support from the Natural
Sciences and Engineering Research Council of
Canada (NSERC).

3
4
G.W. Buchanan et al. / Chemistry and Physics of Lipids 104 (2000) 23–34
Jandacek, R.J., Ramirez, M.M., Crouse, J.R., 1990.
References
Metabolism 39, 848–852.
Jarrell, H.C., Jovall, P.A., Giziewicz, J.B., Turner, L.A.,
Smith, I.C.P., 1987. Biochemistry 26, 1805–1811.
Jensen, L.H., Mabis, A.J., 1966. Acta Cryst. 21, 770–781.
Koch, H.J., Stuart, R.S., 1978. Carbohydr. Res. 67, 341–348.
Matteson, F.H., Volpenhein, R.A., 1972. J. Lipid. Res. 13,
325–328.
Akoh, C.C., Swanson, B.G., 1989. J. Am. Oil. Chem. Soc. 66,
1
581–1587.
Akoh, C.C., Swanson, B.G., 1990. J. Food Sci. 55, 236–243.
Akoh, C.C., 1995. J. Food Sci. 35, 405–430.
Ballinger, R., Malin, R.L., Webb, A.G., 1991. Magn. Reson.
Med. 19, 199–202.
Barnes, R.G., 1974. Adv. Nucl. Quadrupole Reson. 1, 335–
Mellies, M.J., Vitale, C., Jandacek, R.J., Lamkin, G.E.,
Glueck, C., 1985. J. Am. J. Clin. Nutr. 41, 1–12.
Mnyukh, Y.V., 1960. J. Struct. Chem. (USSR) 1, 346–364.
Mutter, L.C., Blanke, R.V., Jandacek, R.J., Guzelian, P.S.,
1988. Toxicol. Appl. Pharmacol. 92, 428–435.
Renou, J.P., Giziewicz, J.B., Smith, I.C.P., Jarrell, H.C., 1989.
Biochemistry 28, 1804–1814.
Rietschel, E.T., Brade, L., Kaca, W., Kawahara, K., Lindner,
B., Luderitz, T., Tomita, T., Schade, U., Seydel, U., 1985.
Prog. Clin. Biol. Res. 189, 31–51.
Smith, J.C., Karplus, M., 1992. J. Am. Chem. Soc. 114,
801–812.
Speyer, J.B., Weber, R.T., Das Gupta, S.K., Griffin, R.G.,
1989. Biochemistry 28, 9569–9574.
Stohrer, M., Noack, M., 1977. J. Chem. Phys. 67, 3729–3738.
Taylor, M.G., Kelusky, E.C., Smith, I.C.P., 1983. J. Chem.
Phy. 78, 5108–5112.
Torchia, D.A., Szabo, A., 1982. J. Magn. Reson. 49, 107–121.
Wittebort, R.J., Olejniczak, E.T., Griffin, R.G., 1987. J.
Chem. Phys. 86, 5411–5420.
3
39.
Batchelder, L.S., Niu, C.H., Torchia, D.A., 1983. J. Am.
Chem. Soc. 105, 2228–2231.
Bloom, M., Davis, J.H., Valic, M.I., 1980. Can. J. Phys. 58,
1
50–157.
Brandenburg, K., Meyer, H.W., Koch, M.H., Weckesser, J.,
Rietschel, E.T., Seydel, U., 1993. Eur. J. Biochem. 218,
5
55–563.
Brandenburg, K., Ritcher, W., Koch, M.H., Meyer, H.W.,
Seydel, U., 1998. Chem. Phys. Lipids. 91, 53–69.
Carrier, D., Giziewicz, J.B., Moir, D., Smith, I.C.P., Jarrell,
H.C., 1989. Biochem. Biophys. Acta 983, 100–108.
Crouse, J.R., Grundy, S.M., 1979. Metabolism 28, 994–1000.
Fenske, D.B., Letellier, M., Roy, R., Smith, I.C.P., Jarrell,
H.C., 1991. Biochemistry 30, 542–550.
Huang, T.H., Skarjune, R.P., Wittebort, R.J., Griffin, R.G.,
Oldfield, E., 1980. J. Am. Chem. Soc 102, 7377–7379.
Jandacek, R.J., Webb, M.R., 1978. Chem. Phys. Lipids. 22,
1
63–176.
.
.