Complete 1H NMR assignment of cholesteryl benzoate

Article abstract of DOI:10.1016/j.steroids.2018.06.010

The 750 MHz ¹H NMR spectrum of cholesteryl benzoate (1b) could be assigned completely, which means all chemical shifts and all coupling constants, including some long-range values, were established. This task was possible by extracting many approximate coupling constant values in the overlapped spectrum region from an HSQC experiment, and using these values in the ¹H iterative full spin analysis integrated in the PERCH NMR software. The task was facilitated using our published data for 3β-acetoxypregna-5,16-dien-20-one (3), the assignment data of the sesquiterpene benzoquinone dihydroperezone (2), also performed in the present study, which contains the same carbon atoms chain than cholesterol (1a), and an HSQC study of (25R)-27-deuteriocholesterol (1c) we prepared some 40 years ago. The HSQC values of 1c in combination with the coupling constants of 1b also allowed to completely assigning the spectrum of 1c. The complete assignment of 1b and 1c further provided the opportunity to estimate the hydrogen shifts induced upon benzoylation of cholesterol. Comparison of the experimental vicinal coupling constants of 1b with the values calculated using the Altona software provides an excellent correlation. In addition, a single crystal X-ray diffraction study of 1b provided the molecular conformation in the solid state, which revealed the side chain adopts an extended conformation.

Full text of DOI:10.1016/j.steroids.2018.06.010

Accepted Manuscript
Complete 1H NMR assignment of cholesteryl benzoate
Nury Pérez-Hernández, Elvia Becerra-Martínez, Pedro Joseph-Nathan
PII:
S0039-128X(18)30120-X
https://doi.org/10.1016/j.steroids.2018.06.010
STE 8281
DOI:
Reference:
Steroids
To appear in:
Received Date:
Revised Date:
Accepted Date:
3 May 2018
25 June 2018
29 June 2018
1
Please cite this article as: Pérez-Hernández, N., Becerra-Martínez, E., Joseph-Nathan, P., Complete H NMR
assignment of cholesteryl benzoate, Steroids (2018), doi: https://doi.org/10.1016/j.steroids.2018.06.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Complete 1H NMR assignment of cholesteryl benzoate
Nury Pérez-Hernández,a Elvia Becerra-Martínez,b Pedro Joseph-Nathanc,
a
Escuela Nacional de Medicina y Homeopatía, Instituto Politécnico Nacional, Mexico City,
07320 Mexico
b
Centro de Nanociencias y Micro y Nanotecnologías, Instituto Politécnico Nacional,
Mexico City, 07738 Mexico.
c Departamento de Química, Centro de Investigación y de Estudios Avanzados del Instituto
Politécnico Nacional, Apartado 14-740, Mexico City, 07000 Mexico.
* Corresponding author at: Departamento de Química, Centro de Investigación y de
Estudios Avanzados del Instituto Politécnico Nacional, Apartado 14-740, Mexico City,
07000 Mexico.
Tel.: +52 55 5747 7112; Fax.: 52 55 5747 7137
E-mail address: pjoseph@nathan.cinvestav.mx

ABSTRACT
1
The 750 MHz H NMR spectrum of cholesteryl benzoate (1b) could be assigned
completely, which means all chemical shifts and all coupling constants, including some
long-range values, were established. This task was possible by extracting many
approximate coupling constant values in the overlapped spectrum region from an HSQC
1
experiment, and using these values in the H iterative full spin analysis integrated in the
PERCH NMR software. The task was facilitated using our published data for 3β-
acetoxypregna-5,16-dien-20-one (3), the assignment data of the sesquiterpene
benzoquinone dihydroperezone (2), also performed in the present study, which contains the
same carbon atoms chain than cholesterol (1a), and an HSQC study of (25R)-27-
deuteriocholesterol (1c) we prepared some 40 years ago. The HSQC values of 1c in
combination with the coupling constants of 1b also allowed to completely assigning the
spectrum of 1c. The complete assignment of 1b and 1c further provided the opportunity to
estimate the hydrogen shifts induced upon benzoylation of cholesterol. Comparison of the
experimental vicinal coupling constants of 1b with the values calculated using the Altona
software provides an excellent correlation. In addition, a single crystal X-ray diffraction
study of 1b provided the molecular conformation in the solid state, which revealed the side
chain adopts an extended conformation.
Keywords:
Cholesteryl benzoate; Iterative 1H NMR analysis; X-ray structure; Dihydroperezone

1. Introduction
Familiar and social conversations dealing with diseases, health, and blood levels of
cholesterol (1a) are relatively frequent, suggesting this molecule, the sterol of the animal
kingdom, seems nowadays to be very popular. This appears to be a natural consequence of
almost three centuries in the study of 1a since literature can be traced back to contributions
published in 1733, 1775 and 1789, the first one when it was recognized that gallstones were
soluble in alcohol or turpentine, as quoted in a superb book on cholesterol [1] published
over half a century ago. It is extremely difficult to completely trace the scientific literature
regarding cholesterol, since literally thousands of publications deal with almost any aspect
of the still intriguing behavior of the molecule. Therefore, to this end it suffices here to say
that despite many publications that continue to appear, at least four books published during
the last 15 years testify the current interest on cholesterol [2-5].
The structure of cholesterol (1a) (Scheme 1) followed after extensive chemical studies
that included degradation processes, as well as synthesis and preparation of derivatives of
some degradation molecules. The information is described in some 25 publications that
appeared during the last decade of the nineteenth century and the initial four decades of the
twentieth century [1].
The complete biosynthesis of cholesterol (1a) has also been studied extensively [6]
starting from its most basic precursor using [14C]acetate. Although this required the tedious
time-consuming carbon-by-carbon degradation processes, it allowed to completely
13
mapping the two atoms of acetate in the sterol. The advent of continuous-wave C NMR
spectroscopy for studies of natural products and some of their derivatives in the late sixties
of the last century provided us [7] the opportunity to study dihydroperezone (2), which

13
possesses the same side chain than sterol 1a. Since C NMR spectroscopy suffered a very
fast evolution by the introduction of the pulse Fourier transform methodologies, the 13C
NMR spectrum of cholesterol (1a) was almost completely assigned quite soon [8],
excepting the distinction of the C-26 and C-27 signals. This spectroscopic technique
changed dramatically the way biosynthetic studies were performed since years of efforts
consumed by degradation processes of 14C radioactive materials could be replaced by
weeks of work [9] using 13C NMR. The assignment of the spectrum of cholesterol (1a) was
finally completed [10] when we prepared a sample of (25S)-27-deuteriocholesterol (1c) in a
four steps sequence using an authentic sample of kryptogenin (cholest-5-ene-3β,26-diol-
16,22-dione) isolated by Russel Marker around 1940.
1
As far as we know, the assignment of H resonances of cholesterol (1a) has been
attempted twice. In the first trial [11], the incomplete assignment of the hydrogen atoms of
the A and B rings was accomplished by means of decoupling experiments and comparison
1
of the H spectrum of a partly deuterated sample. In the second case all chemical shifts of
1a were described combining HSQC and HMBC correlations of the digitonin-cholesterol
complex dissolved in pyridine [12]. Even though this study provided all 1H chemical shifts
of 1a, most of them cannot be used as reference since some suffer strong solvent effects
due to the aromatic solvent, as for instance the signals owing to the C-11 methylene. To
complete the NMR characterization of cholesterol (1a) in a classic solvent like CDCl3 it
1
thus remains to have a detailed assignment of its H spectrum, which must include the
individual chemical shifts of all hydrogen atoms, as well as all homonuclear coupling
constants. Although this is not an easy task due to the severe overlapping spin multiplets in
the 1.54–1.90 ppm region, we took advantage of our recently published work [13] on the

total assignment of 3-acetoxypregna-5,16-dien-20-one (3) and of the availability of a
sample of dihydroperezone (2) [7,10]. In addition, due to the difficulty to obtain highly
purified samples of cholesterol (1a), which in this case is very important since minor
impurities generating NMR signals in the observed spectra region can severely complicate
spin-spin iteration processes, the molecule of choice was cholesteryl benzoate (1b) which
eventually could be purified, after extensive recrystallizations, to afford a sample that was
even suitable for a single crystal X-ray diffraction analysis.
2. Materials and methods
2.1 General
Cholesteryl benzoate (1b) was prepared by treatment of cholesterol (1a) with benzoyl
chloride in pyridine. A sample of approximately 10 mg was placed in a 5 mm NMR sample
tube, dissolved in 0.9 mL of CDCl3 and the solution was degassed by slow bubbling of an
argon stream under ultrasound for 5 min. This rendered a final volume of 0.5 mL to which
a small amount of TMS in CDCl3 was added. In turn, (25R)-27-deuteriocholesteol (1c) and
dihydroperezone (2) were available from previous 13C NMR studies [7,10].
2.2 NMR Spectra
All spectra were obtained on a Bruker Ascend 750 spectrometer equipped with a
cryprobe. The 1H NMR spectra were acquired with number of scans (NS) = 16, acquisition
time (AQ) = 2.18 s, relaxation delay (RD) = 1.0 s, 90o pulse width (P1) = 8.24 ms, spectral
width (SW) = 15 000 Hz and FT size = 65536. The acquisition data for the 2D HSQC of 1b
1
13
were SW H = 887.8 Hz with 2048 increments, and C = 10352 Hz with 512 increments,
AQ = 2.3 s, NS = 16 and RD = 2.0 s.

2.3 1H NMR full spin analysis
1
Complete H NMR spectra analysis of 1b, 1c, and 2 was achieved using the iterative
full spin analysis (HiFSA) method [14] integrated in the PERCH v.2011.1 software
(PERCH Solutions Ltd., Kuopio, Finland). The 750 MHz experimentaldata of 1b, 1c, and 2
were imported into the preparation unit (PAC) of the PERCH shell and subjected to phase
and baseline correction, peak picking, and integration. Molecular models for 1b, 1c, and 2
were built using the molecular modeling module (MMS) and after geometry optimization a
Monte Carlo conformers distribution was done. The most stable conformer for each
molecule was used to obtain the initial calculated spectra.
In the case of dihydroperezone (2) the published [7,10] 13C NMR chemical shifts in
combination with a HSQC plot provided the 1H chemical shifts and revealed which pairs of
hydrogen atoms belong to the same methylene group. In addition, some approximate
coupling constant values were taken from the published values for perezone (4a), O-
methylperezone (4b), and 6-hydroxyperezone (4c) [15]. For the individual assignment of
the pro-R and pro-S hydrogen atoms of the C-10, C-11, and C-12 methylene groups,
several values were exchanged until the convergence was reached at the lowest total root-
mean-square deviation (RMS) of 0.040.
In a similar but not identical approach, in the case of cholesteryl benzoate (1b) the
initial chemical shift values were defined after extraction of the clean multiplets from a 2D
HSQC experiment, while the initial coupling constant values were determined using the
parameters of the two model compounds, 3-acetoxypregna-5,16-dien-20-one (3) [13] and
dihydroperezone (2) as is detailed in the Results and Discussion Section. The vicinal

coupling constant values of 1b were further tested using dihedral angle values obtained
from the optimized conformer by means of the Altona software [16].
2.4 Single crystal X-ray diffraction analysis of cholesteryl benzoate (1b)
The data were collected on an Enraf-Nonius CAD4 diffractometer using Cu Kα radiation (λ
= 1.54184 Å) at 293(2) K in the /2θ scan mode. Crystal data were C34H50O2, M = 490.74,
tetragonal, space group P41212, a = b = 10.513(1) Å, c = 54.436(8) Å, V = 6016.8(1) Å3, Z
= 8,  = 1.083 mg/mm3, μ = 0.492 mm-1, total reflections = 4694, unique reflections 3987
(Rint 0.01%), observed reflections 2727. The structure was solved by direct methods using
the SIR2002 program included in the WinGX v1.70.01 crystallographic software package.
For the structural refinement, the non-hydrogen atoms were treated anisotropically, and the
hydrogen atoms, included in the structure factor calculation, were refined isotropically. The
final R indices were [I > 2(I)] R1 = 5.2% and wR2 = 13.4%. Largest difference peak and
hole, 0.228 and -0.147 e.Å3. The Olex2 v1.1.5 software1 allowed to calculate the Flack
parameter2 x = 0.0(6) and Hooft parameter3 y = -0.4(3). For the inverted structure these
parameters were x = 1.0(6) and y = 1.4(3), respectively. A PLUTO plot of the X-ray
structure is shown in Fig. 1. Crystallographic data (excluding structure factors) have been
deposited at the Cambridge Crystallographic Data Centre (CCDC No. 1848799). Copies of
the data can be obtained free of charge on application to the CCDC, 12 Union Road,
Cambridge CB2 IEZ, UK. Fax: +44-(0)1223-336033 or e-mail: deposit@ccdc.cam.ac.uk.
3. Results and discussion
In order to perform the detailed complete 1H NMR assignment of cholesteryl benzoate (1b)
through iterative full spin analysis (HiFSA) [14], the methodology used to determine the

complete spectra parameters of several natural products [17-21], we relied on the previous
knowdelge of the spectra parameters of two simpler related compounds: 3-acetoxypregna-
5,6-dien-20-one (3) we published a few years ago [13], which contains the
cyclopentanoperhydrophenantrene core including the two angular methyl groups (Scheme
1), and dihydroperezone (2), which was assigned in the present work, containing the same
saturated carbon atoms chain than cholesterol (1a). We used benzoate 1b since it can be
purified easierly than 1a, by extensive recrystallizations, thus allowing to get a high quality
1H NMR spectrum and a sample suitable for single crystal X-ray diffraction analysis (Fig.
1). Due to the iteration capability of the PERCH software, which compares the peaks of the
experimental and calculated spectra, the high purity of a sample for HiFSA is mandatory,
since experience dictates that small impurities that originate NMR signals in the studied
spectra regions can severely complicate a good detailed assignment.
Thus, the 750 MHz free induction decay of 1b was edited in the PAC module of the
PERCH shell. Briefly, the atom coordinates obtained from the single crystal X-ray
diffraction study were introduced in the molecular modeling module (MMS) and, after
geometry optimization; the Monte Carlo conformer distribution analysis provided the most
stable conformer which was used to obtain the initial calculated spectrum. The hydrogen
atoms at positions C-3, C-4, C-6, C-18, C-19, and C-21 were easily recognized at this point.
The positions of all remaining hydrogen atoms were determined using the HSQC
13
correlations based on the reported C assignments of 1a. For this task, several sets of 13C
NMR data are available, although we selected those of the classic works that appeared in
the seventies last century [8,9], which we know since then as correct. In a recent data set
some discrepancies for the chemical shifts of C-16 and of the C-22 to C-25 atoms chain

were suggested [22] but they should be ignored. In addition, to determine the heteronuclear
1
correlations, we decided to disentangle the overlapped spin pattern by extraction of H
multiplicities (Fig. 2) from the HSQC experiment [23]. This method was very useful for the
total assignment of 3β-acetoxypregna-5,16-dien-20-one (3) [13] and was originally
developed by one of us for an early study of α-cedrene [24]. In the present study of 1b, a
SW of 887.8 Hz and AQ of 2.307 s gave a 4k data table which after zero filling to 64k and
1
FT provided an H NMR digital resolution around 0.1 Hz/point. Having the extraction of
1
the traces from the HSQC plot at the carbon frequency, the precise recognition of the H
1
shifts of 1b was possible, (Fig. 2). These H chemical shifts were used to feed and update
the initial values in the table editor of the PMS module. At this point it was also possible to
define the α- and β- positions for the hydrogen atoms at C-1, C-2, C-7, C-11 and C-12
according to their signals shape, since for broader signals at least one additional large
coupling constant, corresponding to the axial hydrogen atoms, were expected. Taking into
account that the coupling constants of the A-, B-, and C-ring hydrogren atoms of 1b keep
similarity with those of 3-acetoxypregna-5,6-dien-20-one (3) [13], since the
conformational backbone should be similar in both compounds, we took advantage of the
reported coupling constants for 3 and substituted them in the 1b spectrum file. Using these
parameters, the simulation of the less crowded spectrum region, at higher frequencies than
1.70 ppm, evidenced a high similarity with the experimental spectrum, and after a few
iterations the simulation of this region was accomplished reaching a RMS of 0.052%.
For the most crowded window, in the 0.90–1.62 ppm region, in which the hydrogen
atoms of the C-22 to C-25 side chain were spliced with many sterane hydrogen atoms, we
took advantage of a sample of dihydroperezone (2) that was available from previous NMR

studies [7,10]. The HSQC experiment of 2, performed as in the case of 1b, provided the
plots depicted in Fig. 3, which for the C-10 to C-13 atoms, that are equivalent to the C-22
to C-25 atoms of 1b, show that H-13 appears at 1.48 ppm, the two multiplets at C-10
provide signals near 1.52 and 1.71 ppm, those assigned to the two hydrogen atoms at C-11
appear at 1.24 and 1.14 ppm, the later being overlapped in the same multiplet than the two
hydrogen atoms at C-12. Consequently, with this information and after several cycles of
iteration, and the interchange of the hydrogen atoms of a given methylene group, a good
agreement for the spectrum of 2, providing a RMS value of 0.040%, could be obtained, as
is shown in Fig. 3.
The coupling constant values obtained for dihydroperezone (2), which are
summarized in Table 1, were used to complete the assignment of the spectrum of 1b. After
feeding the table editor of the PMS module and performing some additional manual
adjustments for these values, a reduced number of iteration cycles finally provided an
excellent concordance between the experimental and calculated spectra of 1b reaching a
RMS of 0.037%.
The complete data for cholesteryl benzoate (1b) are shown in Table 2, while the
simulated spectrum is shown in Fig. 4. HiFSA calculations provide by default chemical
shift and coupling constant values with six and four decimal places, respectively. Since the
experimental 750 MHz spectra of 1b and 2 were acquired with magnet homogeneities
better than 0.8 Hz, chemical shift and coupling constant values in Tables 1 and 2 are
presented with three and two digits, respectively, after the decimal point, as done for
previous cases. [17-21]

It is well known that in cyclohexanes the chemical shifts of axial hydrogen atoms
appear at lower frequencies than the equatorial hydrogen atoms, except in the vicinity of
substituents [25]. The axial hydrogen atoms at positions 1α, 2β, 7α, 11β, and 12α were not
the exception for this trend, resonating in the 0.9–1.7 ppm range, in contrast to the
equatorial hydrogen atoms which appeared in the 1.5–2.0 ppm range. The axial/equatorial
distinction for the two methylene groups on ring D was opposite than that for the six
member ring, the quasi-axial atoms being localized at higher frequencies than the quasi-
equatorial atoms, rendering H-15β and H-16α at 1.078 and 1.840 ppm, respectively, and H-
15α and H-16β at 1.589 and 1.266 ppm, respectively. This is in accordance with the
assignments for some bile acids and some cholestane derivatives [26], although it is
difficult to make a generalization due to the high conformational mobility of the five
member ring and the nature of the substituent at C-17. In turn, the chemical shifts of the
cholestane side chain C-20 to C-25 are in agreement with the values described for some
cholestane steroids [26] having the same side chain. Finally, the methyl groups at C-18, C-
19 and C-21 were observed at 0.692, 1.071 and 0.924 ppm, respectively, and it was
possible for the first time to distinguish CH3-26 and CH3-27 (Fig. 5), at 0.871 and 0.867
ppm, respectively, taking advantage of our stereospecific C-27 deuterated sample prepared
for the 13C NMR assignment of cholesterol [10], in combination with an HSQC correlation
measurement.
1
The main difficulty for the complete assignment of the H NMR spectrum of 1b
resides on the precise knowledge of all homonuclear coupling constant values. Therefore, it
is of relevance to evaluate these values in some detail, which means the J values of the
rings and of the side chain require separate comments. Regarding the rings, the Jgem values

of the methylene groups are consistent with strain-free systems with values between -12.40
and -13.73 Hz, although the geminal coupling constant of the C-7 methylene groups is
much larger, -17.65 Hz, since it is influenced by the adjacent π-bond, as is well documented
[27], and is also observed (Table 3) for the C-11 methylene in compounds 4a–c [15]. The
magnitude of vicinal coupling constants of rings A and C are in good agreement with chair
conformations, while for ring B they agree with a half chair conformation. The axial-axial
values for J1α,2β, J2β,3, J3,4β, J7,8, J8,9, J8,14, J9,11β and J11β,12α, are in the 10.41–14.08 Hz
range, the axial-equatorial values for J1α,2α, J1β,2β, J2,3, J3,4, J7β,8, J9,11α, J11α,12α, and J11β,12β
are in the 3.72–5.37 Hz range, and the equatorial-equatorial values for J1β,2α and J11α,12β are
3
3.45 and 2.94 Hz, respectively. Regarding ring D, the JHH values are variable, since Jcis
corresponding to J14,15β, J15α,16α, and J15β,16β show values of 12.80, 9.81, and 11.64 Hz,
respectively, which are larger than the Jtrans values corresponding to J14,15α, J15α,16β, and
J15β,16α of 7.38, 3.14, and 6.23 Hz, respectively, according to the tendency for an envelope
conformation [28]. The remaining vicinal coupling constants, J6,7α and J6,7β in ring B, and
J16α,17 and J16β,17 in ring D show values following a Karplus type relationship [29] for the
first pair, and close to 9 Hz for the second pair.
In addition to the vicinal coupling constants, long range couplings can usually be
1
4
observed in the H NMR spectra of steroids. In this sense, the long-range JHH coupling
constants observed in 1b corresponding to the well-known W arrangements between H-1α
and CH3-19, of H-2 and H-4, of H-9 and CH3-19, and of H-6 and H-8 showed values
smaller than 1 Hz [25], while the well-known 4JHH of H-12α and CH3-18 shows a value of -
0.63 Hz.

Allylic coupling constants were observed for hydrogen atoms on either side of the
C-5–C-6 double bond, as expected from some described steroids [25]. The magnitude of
the allylic coupling between H-4α and H-6 (0.60 Hz) was smaller than those of H-4β and
H-6 (2.07 Hz) in line with theoretical correlations of the torsion angles between the plane of
the double bond and the adjacent methylene hydrogen atoms [25]. Homoallylic couplings
were also seen between the methylene groups at C-4 and C-7, specifically for H-4β with H-
7α and H-7β, of 3.51 and 2.62 Hz, respectively.
For the vicinal coupling constant values of the carbon atoms chain of 1b, the values
for J22R,23R, J22R,23S, J22S,23R, J22S,23S, J23R,24R, J23R,24S, J23S,24R, and J23S,24S are in excellent
agreement with those of dihydroperezone (2) and it is also interesting to note they provide
similar values than those observed for J2,3, J2,3´, J3,4 and J3,4´, in n-hexane when measured in
3
3
choloform [30]. The J2,3/3J2,3’ and J3,4/3J3,4 values in this n-alkane are 1.58 and 1.67
respectively, which are related to a trans-gauche equilibrium preference of C-2 with C-3
and of C-3 with C-4, being values close to 2 those which denote a major proportion of the t
3
conformer [30]. In the case of 1b the ratio J22R,23S/3J22R,23S ´is 2.2, while the ratio
3J23R,24S/3J23R,24S is 1.9, both values being comparable with those of n-hexane suggesting
likewise a preference for a t conformation. The remaining vicinal coupling constants of the
carbon atoms chain, J24R,25 and J24s,25, as well as those with the methyl groups J20,21, J25,26
,
and J25,27 show traditional values in the 6-7 Hz range, also similar than those found in
dihydroperezone (2).
The similarity of the vicinal coupling constant values of the side chains of
cholesteryl benzoate (1b) and of dihydroperezone (2) suggests the spatial atom arrangement
is similar in both molecules. Close inspection of the values shown in Table 3 reveals the

only noticeable difference is observed between J20,22S = 2.58 Hz for 1b as compared to J8,10S
= 6.67 for 2, which is attributed to the fact that in the former compound the side chain is
appendage of a cyclopentanone moiety, while in the later it is appendage of a 2-hydroxy-
1,4-benzoquinone. This is further evidenced from J17,20 = 10.27 Hz in 1b, which according
to Altona [16] corresponds to an H(17)-C(17)-C(20)-H(20) dihedral angle of 157° in
reasonable good agreement with the torsion angle of 176.9° derived from the X-ray study.
In turn, the coupling constants of H-20 and the two hydrogen atoms at C-22 are 8.78 and
2.58 Hz, which according to Altona provide dihedral angle values of 146 and 63°,
respectively, in reasonable agreement with the values (169.7 and 88.1°) provied by the
torsion angles of the crystal structure. In the case of dihydroperezone (2) there is
considerable less steric hindrance between the quinone ring and the sustituents at C-8,
which results in an overall higher conformational mobility and therefore the coupling
constants of H-8 and the two hydrogen atoms at C-10 are more similar one respect to the
other one, providing values of 8.70 and 6.67 Hz. Furthermore, comparison of the vicinal
coupling constant values of the C-8–C-10–C-11 fragment of 2 and those of 4a–c (Table 3)
reveals these values are quite similar, also suggesting the molecules share similar
conformational space distributions in agreement with vibrational circular dichroism results
obtained for 2 and 4a [31].
In order to approximately validate the coupling constant values obtained by PERCH
3
analysis of 1b, the JH–H values of hydrogen atoms on the rings were compared with those
calculated using the Altona software [16]. For this purpose, the molecular model of 1b was
constructed using the Spartan 04 (Wavefunction, Inc., Irvine, CA) software, and the
conformational space was evaluated using the Monte Carlo protocol with molecular

mechanics force field (MMFF94) in an energy window of 10 kcal/mol. This provided 20
conformers in the initial 3.14 kcal/mol energy gap which were subjected to single point
calculations using DFT with the B3LYP functional and the 6-31G(d) basis set. The
resulting 20 conformers in a 3.02 kcal/mol energy range, covering a Boltzmann population
of 99.8% showed the D-ring in an envelope conformation. Comparison of the observed
vicinal coupling constants with the values calculated by Altona show a good congruency as
can be seen in Table 4. The R-squared (R2) value derived from linear regression of
experimental and calculated values was 0.9849, indicating an excellent correlation. This
1
fact reveals the importance of a second methodology for verification of simulated H NMR
data.
1
The comparison of the H NMR spectra of cholesteryl benzoate (1b) and (25R)-27-
deuteriocholesterol (1c), shown in Fig. 5, in addition of allowing distinction of the methyl
signals at positions 26 and 27, provides the opportunity to evaluate the H-3–H2-4 spin
system of both compounds. In the case of 1b the pair of hydrogen atoms at C-4 have almost
the same chemical shift and are coupled one with the other one while being differently
coupled with H-3. The resulting ABX spin system shows a doublet (J = 7.62 Hz) at 2.467
ppm for the AB portion, which collapses to a singlet upon irradiation of H-3 at 4.863 ppm,
revealing the existence of an uncommon virtual coupling spin system [32] for which it is
quite difficult to accurate obtain coupling constant values beyond those estimated by the
PERCH software. In contrast for 1c the two H-4 signals have distinct chemical shifts as
seen in Fig. 5.
Since the conformation of the rings system of 1b and 1c might essentially be the
same, one would expect quite similar coupling constant values in both compounds.

Therefore an HSQC experiment of 1c provided the approximate chemical shifts of all
hydrogen atoms, which in combination with the coupling constants of 1b permitted to
simulate the spectrum of 1c to a nice RMS value of 0.042%. Comparison of the
experimental and calculated spectra of 1c is shown in Fig. 6, while the pertinent parameters
are summarized in Table 5. In turn, comparison of the chemical shift values given in Table
5 for 1c and those of 1b shown in Table 2 reveals that benzoylation of cholesterol shifts the
hydrogen atoms on the rings system in the magnitudes shown in Fig. 7. It can further be
observed that the most influenced signals are those corresponding to the atoms on ring A,
and as a general trend the magnitude of the induced shifts decrease for rings B, C, and D. In
addition, all axial hydrogen atoms are more shifted than their equatorial counterparts on a
given carbon atom, excepting the two signals owing to H2-11 which are equally shifted.
4. Conclusions
1
All H NMR chemical shifts and all homonuclear coupling constant values of the
steroidal portion of cholesteryl benzoate (1b) were obtained by full spin analysis integrated
in the PERCH NMR software in combination with a high resolution spectrum measured at
750 MHz under high magnet homogeneity conditions. For this goal it was necessary to
resort to the 1H NMR data of dihydroperezone (2), which possesses the same carbon atoms
chain than cholesterol (1a), of 3β-acetoxypregna-5,16-dien-20-one (3), which possesses the
sterol ring system including the two angular methyl groups, and of (25S)-27-
deuteriocholesterol (1c) which allowed to ascribe the two terminal methyl groups which
differ by only 0.004 ppm. The homonuclear hydrogen coupling constants of the C-8-C-10–

C-11 fragment of dihydroperezone (2), are quite similar to those of perezone (4a), O-
methylperezone (4b), and 6-hydroxyperezone (4c), suggesting they occupy similar
conformational spaces. In turn, the complete side chains of cholesteryl benzoate (1b) and
dihydroperezone (2) also show quite similar coupling constant values, suggesting extended
chain conformers.
Acknowledgements
We thank CONACYT-Mexico Grant No. 284194 for partial financial support and to a
reviewer for pointing our attention to the virtual coupling of the H-3–H2-4 spin system in
1b.
References
[1] D. Kritchevsky, Cholesterol, Wiley and Sons Inc., New York and Chapman & Hall,
Ltd., London, 1958.
[2] D. Graveline, Lipitor, thief of memory: Statin drugs and the misguided war on
cholesterol, published by D. Graveline, www.spacedoc.net, 2006.
[3] B.H. Roberts, The truth about statins: Risks and alternatives to cholesterol-lowering
drugs, Pocket Books, New York, USA, 2012.
[4] F.A. Kummerow, J.M. Kummerow, Cholesterol is not the culprit: A guide to preventing
heart disease, Spacedoc Media, LLC, no city given, 2014.

[5] J. Bowden, S. Sinatra, The great cholesterol myth: Why lowering your cholesterol won't
prevent heart disease, Fair Winds Press, Beverly, MA, USA, 2015.
[6] a) J. Wüersch, R.L. Huang, K. Bloch, The origin of the isooctyl side chain of
cholesterol, J. Biol. Chem. 195 (1952) 439-446; b) K. Bloch, Über die herkunft des
kohlenstoff-Atoms 7 in Cholesterin. Ein beitrag zur kenntnis der Biosynthese der
Steroide, Helv. Chim. Acta 36 (1953) 1611-1614; c) W.G. Dauben, K.H. Takemura, A
Study of the mechanism of conversion of acetate to cholesterol via squalene, J. Am.
Chem. Soc. 75 (1953) 6302-6304. d) J.W. Cornforth, G.D. Hunter, G. Popják, Studies
of cholesterol biosynthesis. 1. A new chemical degradation of cholesterol, Biochem. J.
54 (1953) 590-597; e) J.W. Cornforth, G.D. Hunter, G. Popják, Studies of cholesterol
biosynthesis. 2. Distribution of acetone carbon in the ring structure, Biochem. J. 54
(1953) 597-601. f) J.W. Cornforth, I.Y. Gore, G. Popják, Studies on the biosynthesis of
cholesterol. 4. Degradation of rings C and D, Biochem. J. 65 (1957) 94-109.
13
[7] P. Joseph-Nathan, M.P. González, L.F. Johnson, J.N. Shoolery, Natural abundance C
NMR studies of perezone and derivatives, Org. Magn. Reson. 3 (1971) 23-29.
[8] a) H.J. Reich, M. Jautelat, M.T. Messe, F.J. Weigert, J.D. Roberts, Nuclear magnetic
resonance spectroscopy. Carbon-13 spectra of steroids, J. Am. Chem. Soc. 91 (1969)
7445-7454; b) H.H. Mantsch, I.C.P. Smith, A study of solvent effects on the 13C nuclear
magnetic resonance spectra of cholesterol, pyridine, and uridine, Can. J. Chem. 51
(1973) 1384-1391; c) W.B. Smith, D.L. Deavenport, J.A. Swanzy, G.A. Pate, Steroid
carbon-13 chemical shifts. Assignments via shift reagents, J. Magn. Reson. 12 (1973)
15-19. d) J.W. ApSimon, H. Beierbeck, J.K. Saunders, Lanthanide shift reagents in 13C

nuclear magnetic resonance: Quantitative determination of pseudocontact shifts and
assignment of 13C chemical shifts of steroids, Can. J. Chem. 51 (1973) 3874-3881.
[9] G. Popják, J. Edmond, F.A.L. Anet, N.R. Eaton Jr, Carbon-13 NMR studies of
cholesterol biosynthesized from [13C]mevalonates, J. Am. Chem. Soc. 99 (1977) 931-
935.
[10] P. Joseph-Nathan, G. Mejía, D. Abramo-Bruno, 13C NMR assignment of the side-
chain methyls of C27 steroids, J. Am. Chem. Soc. 101 (1979) 1289-1291.
[11] S.P. Sawan, T.L. James, L.D. Gruenke, J.C. Craig, Proton NMR assignments for
cholesterol. Use of deuterium NMR as an assignment aid, J. Magn. Reson. 35 (1979)
409-413.
[12] P. Muhr, W. Likussar, M. Schubert-Zsilavecz, Structure investigation and proton and
carbon-13 assignments of digitonin and cholesterol using multidimensional NMR
techniques, Magn. Reson. Chem. 34 (1996) 137-142.
[13] E. Becerra-Martínez, K.E. Ramírez-Gualito, N. Pérez-Hernández, P. Joseph-Nathan,
Total 1H NMR assignment of 3β-acetoxypregna-5,16-dien-20-one, Steroids 104 (2015)
208-213.
[14] a) R. Laatikainen, M. Niemitz, U. Weber, J. Sundelin, T. Hassinen, J. Vepsäläinen,
General strategies for total-line-shape-type spectral analysis of NMR spectra using
integral-transform iterator, J. Magn. Reson. Ser. A 120 (1996) 1-10. b) J.G. Napolitano,
D.C. Lankin, J.B. McAlpine, M. Niemitz, S.P. Korhonen, S.N. Chen, G.F. Pauli, Proton
1
fingerprints portray molecular structures: Enhanced description of the H NMR spectra

of small molecules, J. Org. Chem. 78 (2013) 9963-9968. c) R. Laatikainen, M. Tiainen,
S.P. Korhonen, M. Niemitz, Computerized analysis of high-resolution solution-state
spectra. In: Harris RK, Wasylishen RE, editors. Encyclopedia of Magnetic Resonance,
Wiley, Chichester, UK, 2012.
[15] G. Zepeda, E. Burgueño-Tapia, N. Pérez-Hernández, G. Cuevas, P. Joseph-Nathan,
NMR-based conformational analysis of perezone and analogues, Magn. Reson. Chem.
51 (2013) 245–250.
[16] C.M. Cerda-García-Rojas, L.G. Zepeda, P. Joseph-Nathan, A PC program for
1
calculation of dihedral angles from H NMR data, Tetrahedron Comput. Methodol. 3
(1990) 113–118.
[17] L.F. Julio, E. Burgueño-Tapia, C.E. Díaz, N. Pérez-Hernández, A. González-Coloma,
P. Joseph-Nathan, Absolute configuration of the ocimene monoterpenoids from
Artemisia absinthium, Chirality 29 (2017) 716-725.
[18] N. Pérez-Hernández, B. Gordillo-Roman, D. Arrieta-Baez, C.M. Cerda-Garcia-Rojas,
1
P. Joseph-Nathan, Complete H NMR assignment of cedranolides, Magn. Reson.
Chem. 55 (2017) 169–176.
[19] A.R. Ortega, N. Ortiz-Pastrana, L. Quijano, E. Becerra-Martínez, J.O. Olmedo-
Aguirre, P. Joseph-Nathan, Structure and absolute configuration of hydroxy-bis-
dihydrofarinosin from Encelia farinosa, Magn. Reson. Chem. 55 (2017) 530-539.

[20] B. Esquivel, E. Burgueño-Tapia, C. Bustos-Brito, N. Pérez-Hernández, L. Quijano, P.
Joseph-Nathan, Absolute configuraton of the diterpenoids icetexone and conacytone
from Salvia ballotaeflora, Chirality 30 (2018) 177-188.
[21] M.A. Muñoz, N. Pérez-Hernández, M.W. Pertino, G. Schmeda-Hirschmann, P.
Joseph-Nathan, Absolute configuration and 1H NMR characterization of
rosmaridiphenol diacetate, J. Nat. Prod. 75 (2012) 779−783.
[22] M. Tesemma, L. Adane, Y. Tariku, D. Muleta, S. Demise, Isolation of compounds
from acetone extract of root wood of Moringa stenopetala and evaluation of their
antibacterial activities, Res. J. Med. Plants 7 (2013) 32-47.
[23] S. Simova, Application of HSQC to the measurement of homonuclear coupling
constants, J(H,H), Magn. Res. Chem. 36 (1998) 505-510.
[24] P. Joseph-Nathan, Resonancia magnética nuclear de cedranólidos estudiados en una y
en dos dimensiones, Rev. Soc. Quím. Méx. 32 (1988) 106–119.
[25] N.S. Bhacca and D.H. Williams, Applications of NMR Spectroscopy in Organic
Chemistry, Holden-Day, San Francisco, 1964.
[26] D.N. Kirk, H.C. Toms, C. Douglas, K.A. White, K.E. Smith, S. Latif, R.W.P.
1
Hubbard. A survey of the high-field H NMR spectra of the steroid hormones, theirs
hydroxylated derivatives, and related compounds, J. Chem. Soc. Perkin Trans 2 (1990)
1567-1594.

[27] V. Rutar, J.R. Clark, Measurement of proton geminal coupling constants via selective
two-dimensional indirect J spectroscopy. Application to the study of the conformation of
steroids, J. Chem. Soc. Chem. Commun. (1984) 1518-1520.
[28] K.R. Gustafson, S.T.S. Chan, Y. Liu, Assigning molecular configuration by nuclear
magnetic resonance in modern NMR approaches to the structure elucidation of natural
products: Volume 2: Data Acquisition and Applications to Compound Classes, Edited by
Antony Williams Gary, Gary Martin and David Rovnyak, Royal Society of Chemistry,
Cambridge, UK, (2017) 42-43.
[29] M. Karplus, Proton spin coupling by Pi electrons, J. Chem. Phys. 33 (1960) 1842-
1849.
1
[30] T. Tynkkynen, T. Hassinen, M. Tiainen, P. Soininen, R. Laatikainen, H NMR spectral
analysis and conformational behavior of n-alkanes in different chemical environments,
Magn. Reson. Chem. 50 (2012) 598-607.
[31] E. Burgueño-Tapia, C.M. Cerda-García-Rojas, P. Joseph-Nathan. Conformational
analysis of perezone and dihydroperezone using vibrational circular dichroism,
Phytochemistry 74 (2012) 190-195.
[32] H. Günther, NMR Spektroskopie, pp 172-184, Thieme, Stuttgart 1973.

SCHEME AND FIGURE LEGENDS
Scheme 1. Molecular formulas
Fig. 1. PLUTO plot of X-ray diffracted cholesteryl benzoate (1b)
Fig. 2. Partial 1H NMR spectrum of 1b showing the lower frequencies region. Multiplets
extracted at the 13C chemical shifts from HSQC are: C-9 at 50.1 ppm, C-14 at 56.8 ppm, C-
17 at 56.2 ppm, C-24 at 39.5 ppm, C-23 at 23.9 ppm, C-22 at 36.2 ppm, C.-20 at 35.8 ppm,
C-25 at 28.0 ppm, C-11 at 21.1 ppm, C-15 at 24.3 ppm, C-16 at 28.5 ppm, C-1 at 37.1 ppm,
C-7 at 32.0 ppm, C-8 at 31.9 ppm, C-2 at 27.9 ppm and C-12 at 39.8 ppm
Fig. 3. Partial 1H NMR spectrum of 2 showing the lower frequencies region. Multiplets
extracted at the 13C chemical shifts from HSQC are: C-11 at 25.9 ppm, C-13 at 27.9 ppm,
C-10 at 34.3 ppm, and C-12 at 38.9 ppm, The labeled peak (*) is due to moisture
1
Fig. 4. Comparison of the PERCH calculated (top) and the experimental (bottom) H NMR
spectra of 1b in CDCl3 at 750 MHz
Fig. 5. Comparison of the experimental 750 MHz 1H NMR spectra of cholesteryl benzoate
(1b, bottom) and (25R)-27-deuteriocholesterol (1c, top). Signal expansions for 1b are
shown in Fig. 4. The labeled peak (*) is due to moisture
Fig. 6. Comparison of the PERCH calculated (top) and the experimental (bottom) 1H NMR
spectra of 1c in CDCl3 at 750 MHz
Fig. 7. Benzoylation induced 1H NMR chemical shifts (in ppm) for cholesterol

Table 1. Chemical shifts (in ppm), coupling orders, and J-values (in Hz) of
dihydroperezone (2)
1H
6
δ
multiplicity
J
4J6,7 = 1.73
6.481
2.059
3.035
1.197
1.518
1.711
1.237
1.136
1.140
1.129
1.480
0.822
0.832
d
d
7
4J6,7 = 1.73
8
ddd
d
3J8,9 = 7.08, 3J8,10R = 6.67, 3J8,10S = 8.70
9
3J8,9 = 7.08
10R
10S
11R
11S
12R
12S
13
dddd
dddd
ddddd
ddddd
dddd
dddd
ddqq
d
3J8,10R = 6.67, 2J10R,10S = -13.37, 3J10R,11R = 5.72, 3J10R,11S = 10.39
3J8,10S = 8.70, 2J10R,10S = -13.37, 3J10R,11S = 10.42, 3J10S,11S = 4.80
3J10R,11R = 5.72, 3J10R,11S = 10.42, 2J11R,11S = -12.91, 3J11R,12R = 5.26, 3J11R,12S = 10.27
3J10R,11S = 10.39, 3J10S,11S = 4.80, 2J11R,11S = -12.91, 3J11S,12R = 10.49, 3J11S,12S = 5.43
3J11R,12R = 5.26, 3J11S,12R = 10.49, 3J12R,12S = -13.52, 3J12R,13 = 7.00
3J11R,12S = 10.50, 3J11S,12S = 5.42, 3J12R,12S = -13.52, 3J12S,13 = 6.80
3J12R,13 = 7.00, 3J12S,13 = 6.80, 3J13,14 = 6.57, 3J13,15 = 6.63
3J13,14 = 6.57
14
15
d
3J13,15 = 6.63

Table 2. Chemical shifts (in ppm), coupling orders, and J-values (in Hz) of cholesteryl
benzoate (1b)
1H
1α
1β
2α
2β
3
δ
multiplicity
dddq
ddd
J
2J1,1β = -13.53, 3J1α,2α = 3.77, 3J1α,2β = 14.08, 4J1,19 = -0.62
2J1,1β = -13.53, 3J1β,2α = 3.45, 3J1β,2β = 3.72
1.217
1.918
2.000
1.739
4.863
2.468
2.466
5.421
1.568
1.992
1.475
0.995
1.530
1.479
1.184
2.025
1.018
1.589
1.078
1.840
1.266
1.107
0.692
1.071
1.381
3J1α,2α = 3.77, 3J1β,2α = 3.45, 2J2,2β = -12.64, 3J2,3 = 4.42, 4J2,4 = -1.89
3J1α,2β = 14.08, 3J1β,2β = 3.72, 2J2,2β = -12.64, 3J2β,3 = 11.54
3J2,3 = 4.42, 3J2β,3 = 11.54, 3J3,4 = 5.18, 3J3,4β = 11.50
4J2,4 = -1.89, 3J3,4 = 5.18, 2J4,4β = -12.78, 4J4,6 = 0.60
3J3,4β = 11.50, 2J4,4β = -12.78, 4J4β,6 = 2.07, 4J4,7 = 3.51, 4J4β,7β = 2.62
4J4,6 = 0.60, 4J4β,6 = 2.07, 3J6,7α = 1.98, 3J6,7β = 5.11, 4J6,8 = -0.49
4J4β,7α = 3.51, 3J6,7α = 1.98, 2J7,7β = -17.65, 3J7,8 = 10.41
4J4β,7β = 2.62, 3J6,7β = 5.11, 2J7,7β = -17.65, 3J7β,8 = 5.37
4J6,8 = -0.49, 3J7,8 = 10.41, 3J7β,8 = 5.37, 3J8,9 = 11.09, 3J8,14 = 10.88
3J8,9 = 11.09, 3J9,11α = 4.61, 3J9,11β = 12.56, 4J9,19 = -0.37
3J9,11α = 4.61, 2J11,11β = -13.73, 3J11α,12α = 3.97, 3J11α,12β = 2.94
3J9,11β = 12.56, 2J11,11β = -13.73, 3J11β,12α =13.44, 3J11β,12β = 4.06
3J11α,12α = 3.97, 3J11β ,12α = 13.44, 2J12,12β = -12.74, 4J12,18 = -0.63
3J11α,12β = 2.94, 3J11β,12β = 4.06, 2J12,12β = -12.74
ddddd
dddd
dddd
ddd
4α
4β
6
ddddd
dddd
ddd
7α
7β
8
ddd
ddddd
dddd
dddd
dddd
dddd
ddd
9
11α
11β
12α
12β
14
15α
15β
16α
16β
17
18
19
20
dddd
dddd
dddd
dddd
dddd
ddd
3J8,14 = 10.88, 3J14,15α = 7.38, 3J14,15β = 12.80
3J14,15α = 7.38, 2J15,15β = -12.40, 3J15α,16α = 9.81, 3J15α,16β = 3.14
3J14,15β = 12.80, 2J15,15β = -12.40, 3J15β,16α = 6.23, 3J15β,16β = 11.64
3J15α,16α = 9.81, 3J15β,16α = 6.23, 2J16,16β = -13.47, 3J16,17 = 9.17
3J15α,16β = 3.14, 3J15β,16β = 11.64, 2J16,16β = -13.47, 3J16β,17 = 9.64
3J16,17 = 9.17, 3J16β,17 = 9.64, 3J17,20 = 10.27
4J12,18 = -0.63
d
4J1,19 = -0.62, 4J9,19 = -0.37
dd
dddq
3J17,20 = 10.27, 3J20,21 = 6.60, 3J20,22R = 8.74, 3J20,22S = 2.58

Table 2. Continuation …
1H
21
δ
multiplicity
d
J
0.924
1.004
1.344
1.339
1.153
1.005
1.144
1.521
0.871
0.867
3J20,21 = 6.60
22R
dddd
dddd
ddddd
ddddd
dddd
dddd
ddqq
d
3J20,22R = 8.74, 2J22R,22S = -13.34, 3J22R,23R = 4.77, 3J22R,23S = 10.51
3J20,22S = 2.58, 2J22R,22S = -13.34, 3J22S,23R = 10.78, 3J22S,23S = 5.43
3J22R,23R = 4.77, 3J22S,23R = 10.78, 2J23R,23S = -12.65, 3J23R,24R = 5.25, 3J23R,24S = 10.11
3J22R,23S = 10.51, 3J22S,23S = 5.43, 2J23R,23S = -12.65, 3J23S,24R = 10.47, 3J23S,24S = 5.28
3J23R,24R = 5.25, 3J23S,24R = 10.47, 2J24R,24S = -13.17, 3J24R,25 = 6.60
2J23R,24S = 10.11, 3J23S,24S = 5.28, 2J24R,24S = -13.17, 3J24S,25 = 7.12
3J24R,25 = 6.60, 3J24S,25 = 7.12, 3J25,26 = 6.64, 3J25,27 = 6.65
3J25,26 = 6.64
22S
23R
23S
24R
24S
25
26 (pro-R)
27 (pro-S)
d
3J25,27 = 6.65

Table 3. Side chain coupling constant values (in Hz) for cholesteryl benzoate (1b),
dihydroperezone (2), perezone (4a), O-methylperezone (4b), and 6-hydroxyperezone (4c)
J
4aa
8.87
6.50
-13.21
5.97
9.37
9.41
6.23
-14.09
-
4ba
8.63
6.68
-13.25
6.07
9.37
9.47
6.17
-14.33
-
4ca
9.29
6.67
-13.49
5.52
9.40
9.05
6.20
-14.72
-
J
2
8.70
1b
8.78
8,10R
20,22R
20,22S
8,10S
6.67
2.58
10R,10S
10R,11R
10R,11S
10S,11R
10S,11S
11R,11S
11R,12R
11R,12S
11S,12R
11S,12S
12R,12S
-13.37
5.72
-13.34
4.47
22R,22S
22R,23R
22R,23S
22S,23R
22S,23S
23R,23S
10.39
10.42
4.80
10.51
10.78
5.43
-12.91
5.26
-12.66
5.25 23R,24R
-
-
-
10.27
10.49
5.43
10.11
23R,24S
23S,24R
23S,24S
24R,24S
-
-
-
10.47
5.28
-
-
-
-
-
-
-13.52
-13.17
a From reference [15].

Table 4. Comparison of HiFSA determined coupling constants for cholesteryl benzoate
(1b) with the vicinal coupling constants calculated using Altona.
J
Exp
3.77
Cal
3.94

1α,2α
1α,2β
1β,2α
1β,2β
-0.17
+0.73
+0.35
+0.28
-0.06
+0.65
+0.46
+0.61
-1.00
-0.25
-0.97
+0.52
+0.42
-0.25
+0.20
+0.49
-0.20
-1.31
+1.30
-1.68
+1.39
+0.31
+0.25
+0.66
+0.53
14.08
3.45
13.35
3.10
4.06
3.78
4.42
4.48
2,3
2β,3
11.54
5.18
10.89
4.72
3,4
3,4β
11.50
10.41
5.37
10.89
11.41
5.62
7,8
7β,8
8,9
11.09
4.61
12.06
4.09
9,11α
9,11β
12.56
3.97
12.14
4.22
11α,12α
11α,12β
11β,12α
11β,12β
14,15α
14,15β
15α,16α
15α,16β
15β,16α
15β,16β
2.94
2.74
13.44
4.06
12.95
4.26
7.38
6.07
12.80
9.81
11.50
11.49
1.75
3.14
6.23
5.92
11.64
9.17
11.39
8.51
16,17
16β,17
9.64
9.11

Table 5. Chemical shifts (in ppm), coupling orders, and J-values (in Hz) of (25R)-27-
deuteriocholesterol (1c).
1H
1α
1β
2α
2β
3
δ
multiplicity
dddq
ddd
J
2J1,1β = -13.40, 3J1α,2α = 3.79, 3J1α,2β = 13.90, 4J1,19 = -0.64
2J1,1β = -13.93, 3J1β,2α = 3.38, 3J1β,2β = 3.79
1.079
1.848
1.835
1.501
3.525
2.294
2.236
5.352
1.528
1.972
1.455
0.928
1.509
1.458
1.156
2.012
0.988
1.574
1.069
1.828
1.258
1.089
0.679
1.009
1.373
3J1α,2α = 3.79, 3J1β,2α = 3.38, 2J2,2β = -12.31, 3J2,3 = 4.41, 4J2,4 = -2.43
3J1α,2β = 13.90, 3J1β,2β = 3.79, 2J2,2β = -12.31, 3J2β,3 = 11.58
3J2,3 = 4.41, 3J2β,3 = 11.58, 3J3,4 = 4.84, 3J3,4β = 11.20
4J2,4 = -2.43, 3J3,4 = 4.84, 2J4,4β = -13.03, 4J4,6 = 0.67
3J3,4β = 11.20, 2J4,4β = -13.03, 4J4β,6 = 2.00, 4J4β,7 = 3.47, 4J4β,7β = 2.78
4J4,6 = 0.67, 4J4β,6 = 2.00, 3J6,7α = 2.11, 3J6,7β = 5.20, 4J6,8 = -0.47
4J4β,7α = 3.47, 3J6,7α = 2.11, 2J7,7β = -17.78, 3J7,8 = 10.52
4J4β,7β = 2.78, 3J6,7β = 5.20, 2J7,7β = -17.78, 3J7β,8 = 5.38
4J6,8 = -0.47, 3J7,8 = 10.52, 3J7β,8 = 5.38, 3J8,9 = 11.19, 3J8,14 = 10.45
3J8,9 = 11.19, 3J9,11α = 4.70, 3J9,11β = 12.53, 4J9,19 = -0.36
3J9,11α = 4.70, 2J11,11β = -13.68, 3J11α,12α = 4.06, 3J11α,12β = 3.10
3J9,11β = 12.53, 2J11,11β = -13.68, 3J11β,12α =13.59, 3J11β,12β = 4.00
3J11α,12α = 4.06, 3J11β ,12α = 13.59, 2J12,12β = -12.73, 4J12,18 = -0.54
3J11α,12β = 3.10, 3J11β,12β = 4.00, 2J12,12β = -12.74
ddddd
dddd
dddd
ddd
4α
4β
6
ddddd
dddd
ddd
7α
7β
8
ddd
ddddd
dddd
dddd
dddd
dddd
ddd
9
11α
11β
12α
12β
14
15α
15β
16α
16β
17
18
19
20
dddd
dddd
dddd
dddd
dddd
ddd
3J8,14 = 10.45, 3J14,15α = 6.75, 3J14,15β = 12.72
3J14,15α = 6.75, 2J15,15β = -12.30, 3J15α,16α = 9.95, 3J15α,16β = 3.04
3J14,15β = 12.72, 2J15,15β = -12.30, 3J15β,16α = 6.26, 3J15β,16β = 11.73
3J15α,16α = 9.95, 3J15β,16α = 6.26, 2J16,16β = -13.47, 3J16,17 = 9.17
3J15α,16β = 3.04, 3J15β,16β = 11.73, 2J16,16β = -13.47, 3J16β,17 = 9.54
3J16,17 = 9.17, 3J16β,17 = 9.54, 3J17,20 = 10.22
4J12,18 = -0.54
d
4J1,19 = -0.64, 4J9,19 = -0.36
dd
dddq
3J17,20 = 10.22, 3J20,21 = 6.58, 3J20,22R = 8.54, 3J20,22S = 2.76