Efficient synthesis of unsaturated 1-monoacyl glycerols for in meso crystallization of membrane proteins

Article abstract of DOI:10.1055/s-0030-1259912

A highly efficient synthesis of unsaturated 1-monoacyl glycerols was established to fulfill the pressing need for materials that form lipidic mesophases utilized in membrane protein crystallization. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-0030-1259912

LETTER
809
Efficient Synthesis of Unsaturated 1-Monoacyl Glycerols for in meso
Crystallization of Membrane Proteins
Synthesis of
U
nsat
u
urated 1-Monoacyl
F
G
lycerols u, Yue Weng, Wen-Xu Hong, Qinghai Zhang*
Department of Molecular Biology, The Scripps Research Institute, 10550 North Torrey Pines Road, CB265, La Jolla, CA 92037, USA
Fax 1(858)7849985; E-mail: qinghai@scripps.edu
Received 31 December 2010
developed by Coleman et al.5 Slight modification of this
Abstract: A highly efficient synthesis of unsaturated 1-monoacyl
synthesis procedure was reported later.6 However, a
broader test and application of this series of synthetic mol-
glycerols was established to fulfill the pressing need for materials
that form lipidic mesophases utilized in membrane protein crystal-
ecules has been hampered by their scarce availability.
Here we report a highly efficient procedure as an alterna-
tive solution for the synthesis of unsaturated 1-MAG lip-
ids in order to address some of the difficulties encountered
in the previous synthesis.5,6
lization.
Key words: hydrogenation, lipidic cubic phase, lipids, membrane
proteins, transesterification
For various 1-MAG lipids, a shorthand nomenclature,
N.T-MAG, was used to denote the carbon numbers for the
respective neck and tail of each single alkyl chain separat-
ed by the cis-double bond (Figure 1).4,5 Accordingly, mo-
noolein is abbreviated 9.9-MAG. Scheme 1 outlines our
new synthetic route to a series of N.T-MAG lipids. Brief-
ly, methyl alkynoates were first synthesized from simple
starting materials over four steps, followed by a highly
stereoselective cis-hydrogenation of the triple bond to Z-
alkenoates. The last step features a direct, regioselective
transesterification of methyl Z-alkenoates with unprotect-
ed glycerol to produce 1-MAG molecules. Table 1 lists
the isolated yield of each intermediate in the synthesis of
19 1-MAG molecules with the length of alkyl chains
(N+T) ranging from 14 to 22. Due to the close structural
similarity for all entries along each column, the respective
synthesis has little variation in the reaction conditions and
outcomes (within experimental error).
3D membrane protein crystals can be grown in a solution
of detergent micelles or in lipidic mesophases. The so
called in meso approach, that makes use of lipidic cubic or
sponge phase (LCP or LSP), has had tremendous success
in the crystallization of integral membrane proteins, such
as human G-protein coupled receptors over the last sever-
al years.1 A single molecule, monoolein (Figure 1), which
self-assembles into a stable cubic phase in aqueous solu-
tion, accounts for nearly all past success of in meso ap-
proach by acting as the hosting lipid for membrane
proteins.2 Prompted by the significant success of this nov-
el method, as well as the various membrane protein chal-
lenges ahead, synthesizing and testing molecules that
form lipidic mesophase with properties different from
monoolein are of great interest.3
neck
tail
head
O
In each synthesis, the respective w-Br-1-alkanol was first
reacted with 3,4-dihydro-2H-pyran (DHP), followed by
coupling with lithium alkynides to give tetrahydro-2H-
pyran (THP)-protected alkynols 2. Subsequent treatment
of 2 with three equivalents of acidic Jones reagent resulted
in THP deprotection and subsequent oxidation of the hy-
droxy group to carboxylic acid in one pot. Then the acids
3 were turned to methyl esters 4 by reacting with methanol
in the presence of sulfuric acid.
HO
O
N–2
T–2
OH
N.T-MAG (N,T = 7–11)
OH
monoolein
(9.9-MAG)
N = T = 9
HO
O
O
Figure 1
It has been noted by Coleman et al. that highly stereose-
lective cis-hydrogenation of alkyne fatty acids was chal-
lenging, which led the authors to take a detour in their
synthesis of 1-MAG lipids.5 We set out to overcome this
challenge by screening various transition-metal-based
catalysts for cis-hydrogenation of these long-chain fatty
acid derived esters (4). Indeed, most catalytic systems
gave unsatisfactory results in terms of the selectivity and
reproducibility. For example, using Lindlar catalyst in the
presence of quinoline gave 35% of E-alkene with 4l as the
substrate. The use of P-2 nickel catalyst gave 15% of
over-reduced alkane together with minor E product. We
Misquitta et al. have designed monoolein homologues (1-
monoacyl glycerols; 1-MAGs) that differ in the length of
alkyl chains and the position of cis-double bond, and a
few shorter-chain variants have been found useful in LCP-
based crystallization of membrane proteins.4 These 1-
MAG lipids were synthesized using a modular approach
SYNLETT 2011, No. 6, pp 0809–0812
x
x.
x
x.
2
0
1
1
Advanced online publication: 15.03.2011
DOI: 10.1055/s-0030-1259912; Art ID: S10210ST
© Georg Thieme Verlag Stuttgart · New York

810
Y. Fu et al.
LETTER
T–2
T–2
O
DHP, cat. PTSA
T–2
Jones reagent
0–4 °C
HO
Br
O
O
Br
n-BuLi, HMPA
–60 °C to r.t.
N–3
O
O
HO
N–3
CH2Cl2, r.t.
N–2
1
2
3 N–2
H2SO4, MeOH
r.t.
O
H2
P-2 Ni (10 mol%)
ethylenediamine (10 mol%)
HO
O
N–2
T–2
glycerol
P1-phosphazene (10 mol%)
O
T–2
O
OH
N.T-MAG
O
N–2
T–2
DMSO, r.t.
r.t.
O
+
N–2
HO
HO
5
4
O
O
N–2
T–2
2-MAG (3–8%)
Scheme 1
Table 1 Isolated Yields of Intermediates in 1-MAG Synthesis
noted that poisoning the P-2 nickel catalyst with 1,2-eth-
ylenediamine had led to stereoselective cis-hydrogenation
of long aliphatic chained alkynes.7,8 In our experiments,
hydrogenation of the above methyl alkynoates 4 using P-
2 nickel–1,2-ethylenediamine catalyst consistently gave
>95% of desired Z-alkenoates 5 in addition to ca. 2–3% of
over-reduced alkanoates and ca. 1–2% E-alkenoates, as
indicated by GC–MS analysis. Although the by-products
were not separable from Z-alkenoates on silica gel col-
umns, we found that the respective contaminants of the fi-
nal MAG lipids (after reaction with glycerol in the next
step) could be effectively removed by crystallization and/
or reversed-phase (RP) HPLC purification. Since removal
of other impurities (2-MAG isomer, vide infra) introduced
in the transesterification step requires similar purification
operations, we consider that >95% of cis-hydrogenation is
satisfactory given the overall substantially improved yield
and efficacy using the new procedure.
Entry N.T
1
2
3
4
5a
7.7 92 85 92 93 95 85 (72c)
MAGb
Mp (°C)
a
b
c
–
7.8
82 96 93 92 88 (71c)
89 95 94 93 84 (62c)
90 93 93 95 83 (54d)
85 96 92 93 81 (56d)
–
7.9
–
d
e
7.10
7.11
32–33
33–34
f
8.7 91 88 93 91 95 88 (70c)
33–35
–
g
h
i
8.8
83 95 98 94 86 (73c)
85 92 93 99 87 (61d)
92 95 94 95 82 (55d)
8.9
37–38
42–43
8.10
j
9.7 95 89 94 91 94 83 (68c)
–
Finally, transesterification of methyl Z-alkenoates 5 to 1-
MAGs was conducted by reaction with five equivalents of
glycerol in the presence of 10 mol% of P1 phosphazene
(2-tert-butylimino-2-diethylamino-1,3-dimethylperhydro-
1,3,2-diazaphosphorine, a strong nonionic base) in
DMSO at room temperature. This condition, optimized
based on a recent publication for the synthesis of other fat-
ty acid glycerides,9 gave good regioselectivity in our reac-
tions with a >92:8 ratio of desired 1-MAG products vs.
corresponding 2-MAG isomers (Scheme 1).10 Although
acetonitrile has been found as a better solvent for the syn-
thesis of saturated 1-monoacyl glycerols,9 it gave lower
yield (<60%) and similar regioselectivity compared to
DMSO in our case studies (35 °C). Another major by-
product of these reactions is 1,3-diacylated glycerols
(<10%), but it can be easily removed by silica gel chroma-
tography. We observed that the reaction suffered with rel-
atively slow and incomplete transformation when run on
gram scales. To drive the reaction to completion, we intro-
duced low vacuum (ca. 60 mmHg) to the reaction vessel
to remove methanol that was simultaneously produced as
the other reaction product. This set-up allows us to run the
reactions on gram scales with an isolation yield of 75–
85% MAG products (containing 3–8% 2-MAGs) after
k
l
9.8
88 94 91 94 80 (50d)
90 95 97 97 87 (61d)
89 92 95 96 81 (73c)
–
9.9
36–37
42–43
m
9.11
n
o
p
10.7 96 90 96 95 91 85 (60d)
43–44
45–47
50–51
10.8
89 96 95 93 82 (60d)
89 95 92 92 80 (75c)
10.10
q
r
s
11.7 87 81 92 98 91 81 (58d)
37–38
42–43
43–45
11.9
88 91 94 92 77 (75c)
84 98 93 92 75 (72c)
11.11
a Purity was >95%, containing ca. 2–3% of alkanotes and 1–2%
E-alkenoates.
b Comprised of 3–8% of 2-MAG and other E-alkenyl and alkyl impu-
rities, which can be readily purified by crystallization and/or re-
versed-phase HPLC.
c Percentage yield after purification by RP-HPLC.
d Percentage yield after purification by crystallization (the first crop).
Synlett 2011, No. 6, 809–812 © Thieme Stuttgart · New York

LETTER
Synthesis of Unsaturated 1-Monoacyl Glycerols
811
overnight reaction. Of note, this procedure avoided the
unnecessary glycerol protection and deprotection opera-
tions employed in previous synthesis in which acidic hy-
drolysis of acetoketal-protected 1-MAG still gave 7–9%
of 2-MAG by-products.5 It appears difficult to avoid the
formation of 2-MAG isomers, as they might be converted
from corresponding 1-MAGs via intramolecular transes-
terification that could occur under either basic or acidic
conditions.
References and Notes
(1) (a) Landau, E. M.; Rosenbusch, J. P. Proc. Natl. Acad. Sci.
U.S.A. 1996, 93, 14532. (b) Johansson, L. C.; Wohri, A. B.;
Katona, G.; Engstrom, S.; Neutze, R. Curr. Opin. Struct.
Biol. 2009, 19, 372. (c) Cherezov, V.; Rosenbaum, D. M.;
Hanson, M. A.; Rasmussen, S. G.; Thian, F. S.; Kobilka,
T. S.; Choi, H. J.; Kuhn, P.; Weis, W. I.; Kobilka, B. K.;
Stevens, R. C. Science 2007, 318, 1258. (d) Jaakola, V. P.;
Griffith, M. T.; Hanson, M. A.; Cherezov, V.; Chien, E. Y.;
Lane, J. R.; Ijzerman, A. P.; Stevens, R. C. Science 2008,
322, 1211. (e) Wu, B.; Chien, E. Y.; Mol, C. D.; Fenalti, G.;
Liu, W.; Katritch, V.; Abagyan, R.; Brooun, A.; Wells, P.;
Bi, F. C.; Hamel, D. J.; Kuhn, P.; Handel, T. M.; Cherezov,
V.; Stevens, R. C. Science 2010, 330, 1066. (f) Chien, E. Y.
T.; Liu, W.; Zhao, Q.; Katritch, V.; Han, G. W.; Hanson, M.
A.; Shi, L.; Newman, A. H.; Javitch, J. A.; Cherezov, V.;
Stevens, R. C. Science 2010, 330, 1091.
(2) (a) Caffrey, M. J. Struct. Biol. 2003, 142, 108. (b) Caffrey,
M.; Cherezov, V. Nat. Protoc. 2009, 4, 706.
(3) (a) Hato, M.; Yamashita, J.; Shiono, M. J. Phys. Chem. B
2009, 113, 10196. (b) Yamashita, J.; Shiono, M.; Hato, M.
J. Phys. Chem. B 2008, 112, 12286. (c) Borshchevskiy, V.;
Moiseeva, E.; Kuklin, A.; Bueldt, G.; Hato, M.; Gordeliy, V.
J. Cryst. Growth 2010, 312, 3326.
A majority of 1-MAGs in Table 1 could be purified to
>98% purity by low-temperature crystallization (–20 °C
→ 4 °C), similarly as previously described.5 We found
that acetonitrile worked as a good crystallization solvent
in which 2-MAG isomers could be effectively removed.
However, longer chained lipids (9.11-MAG, 11.9-MAG,
10.10-MAG and 11.11-MAG) precipitated quickly in ac-
etonitrile even at room temperature, which resulted in the
collection of less pure products by crystallization. In addi-
tion, the short-chained 7.7-MAG was more difficult to
crystallize and handle due to its low melting point. For
these exceptions, we have used reversed-phase (RP)
HPLC equipped with a preparative C18 column and an
evaporative light-scattering detector (for monitoring less
UV-sensitive compounds) to purify them to >98% purity
(eluting with mixed solvent of acetonitrile and water).
Crystallization and HPLC purification can also be com-
bined in other cases where higher purity is desired.
(4) (a) Misquitta, Y.; Cherezov, V.; Havas, F.; Patterson, S.;
Mohan, J. M.; Wells, A. J.; Hart, D. J.; Caffrey, M. J. Struct.
Biol. 2004, 148, 169. (b) Misquitta, L. V.; Misquitta, Y.;
Cherezov, V.; Slattery, O.; Mohan, J. M.; Hart, D.; Zhalnina,
M.; Cramer, W. A.; Caffrey, M. Structure 2004, 12, 2113.
(5) Coleman, B. E.; Cwynar, V.; Hart, D. J.; Havas, F.; Mohan,
J. M.; Patterson, S.; Ridenour, S.; Schmidt, M.; Smith, E.;
Wells, A. J. Synlett 2004, 1339.
The purified lipids were snowy white powders, displayed
low UV absorbance between l = 295–400 nm and low
fluorescence emission (lex = 280 nm) when dissolved in
methanol at a concentration of 100 mg/mL, meeting im-
portant criteria for being used in protein spectroscopic
analysis.2b A preliminary test of the synthetic 9.9 MAG in
this study has also yielded crystals of b2 adrenergic recep-
tor, demonstrating the applicability of the new synthetic
procedure. Systematic evaluation of the 1-MAG library is
ongoing in the Joint Center for Innovative Membrane Pro-
tein Technologies-Complexes at The Scripps Research
Institute.
(6) Caffrey, M.; Lyons, J.; Smyth, T.; Hart, D. J. In Current
Topics in Membranes, Vol. 63; DeLucas, L., Ed.; Elsevier:
San Diego, 2009, 83.
(7) Brown, C. A.; Ahuja, V. K. J. Chem. Soc., Chem. Commun.
1973, 553.
(8) Valicenti, A. J.; Pusch, F. J.; Holman, R. T. Lipids 1985, 20,
234.
(9) Kharchafi, G.; Jerome, F.; Douliez, J. P.; Barrault, J. Green
Chem. 2006, 8, 710.
(10) 2-MAG isomers can be clearly identified by examining their
signature 1H NMR and 13C NMR signals.5,6
(11) Representative Procedures for the Synthesis of 9.9-MAG
(Entry l)
In summary, we have reported a highly efficient method11
for the synthesis of 1-MAG lipids to address the pressing
need for materials that form lipidic mesophases utilized in
membrane protein crystallization. The N.T-MAG homo-
logues of monoolein are assumed to have different prop-
erties (e.g. varied thickness of lipid bilayers by altering
the chain length), thus providing valuable tools to chal-
lenging membrane protein studies.
2-[(8-Bromooctyl)oxy]tetrahydro-2H-pyran (1l): To a
stirred solution of 8-bromooctan-1-ol (10.5 g, 50 mmol) in
anhyd CH2Cl2 (50 mL), DHP (5.2 mL, 57 mmol) and p-
toluenesulfonic acid (PTSA, 430 mg, 2.5 mmol) were added
at r.t. The reaction was stirred overnight and then sat. aq
NaHCO3 (50 mL) was added. The aqueous layer was
extracted with CH2Cl2 (50 mL). The combined organic
extracts were dried over Na2SO4 and concentrated in vacuo.
Flash chromatography on silica gel using 2% EtOAc in
hexane as the eluent gave 1l12 as a colorless oil in 95% yield.
2-(Octadec-9-yn-1-yloxy)tetrahydro-2H-pyran (2l): To a
stirred solution of 1-decyne (2.76 g, 20 mmol) in anhyd THF
(50 mL), hexamethylphosphoramide (HMPA, 10 mL) and n-
BuLi (1.6 M solution in hexane, 11 mL) were slowly added
at –60 °C. After half an hour the reaction mixture was
warmed to –20 °C, and then 1l (4.1 g, 14 mmol) was added
dropwise. The reaction mixture was slowly warmed to r.t.
and stirred overnight. The reaction was quenched by
addition of sat. aq NH4Cl (50 mL) and extracted with EtOAc
(3 × 50 mL). The combined organic extracts were washed
with sat. aq NaHCO3 and brine, and then dried over Na2SO4
and filtered. Compound 2l13 was purified by silica gel
Acknowledgment
This work was supported by the NIH roadmap grant P50
GM073197. We thank Drs. W. Liu and V. Cherezov for quality test
of synthetic 1-MAG molecules for LCP applications, and Dr. M.G.
Finn for use of the GC–MS instrument.
Synlett 2011, No. 6, 809–812 © Thieme Stuttgart · New York

812
Y. Fu et al.
LETTER
chromatography (2% EtOAc in hexane) as a pale yellow oil
in 90% yield.
2,3-Dihydroxypropyl (Z)-Octadec-9-enoate (9.9-MAG):
Compound 5l (1.5 g, 5 mmol) was added to a two-necked
flask containing DMSO (10 mL). Glycerol (2.3 g, 25 mmol)
and P1 phosphazene (137 mg, 0.5 mmol) were subsequently
added. One neck of the reaction flask was connected to a
vacuum line (ca. 60 mmHg) installed in the fume hood. The
reaction mixture was stirred at r.t. overnight. Saturated aq
NaHCO3 (50 mL) was added; then the mixture was extracted
with EtOAc (3 × 50 mL). The combined organic extracts
were dried over Na2SO4, filtered, and concentrated in vacuo.
The residue was submitted to purification by silica gel
chromatography (20–60% EtOAc in hexane) to give a white
solid containing 1-MAG as the major product, accompanied
with 3% of corresponding 2-MAG, in a total yield of 87%.
The crude product was dissolved in MeCN at a concentration
of 100 mg/mL (slight heating if needed) and then left in the
refrigerator (4 °C) for crystallization. The precipitant was
filtered to give 9.9-MAG with over 99% purity. The mixture
can also be purified by preparative RP-HPLC (Atlantis dC18
column, 5 mm, 30 × 150 mm; MeCN–H2O). Three authentic
samples corresponding to the major by-products in this case:
1,3-dihydroxypropyl (Z)-octadec-9-enoate (major 2-MAG
by-product), 2,3-dihydroxypropyl (E)-octadec-9-enoate and
over-reduced 2,3-dihydroxypropyl octadecanoate have been
prepared and analyzed to reassure effective HPLC analysis
and purification. The chemical purity of final products were
also examined by 1H NMR and 13C NMR spectroscopic
analyses.5,6,15 1H NMR (400 MHz, CDCl3): d = 5.29–5.39
(m, 2 H), 4.14 (d, J = 5.5 Hz, 2 H), 3.87–3.95 (m, 1 H), 3.68
(dd, J = 11.6, 3.5 Hz, 1 H), 3.57 (dd, J = 11.6, 6.1 Hz, 1 H),
3.49 (br, 2 H), 2.34 (t, J = 7.6 Hz, 2 H), 1.95–2.07 (m, 4 H),
1.56–1.67 (m, 2 H), 1.19–1.38 (m, 20 H), 0.88 (t, J = 6.7 Hz,
3 H). 13C NMR (100 MHz, CDCl3): d = 174.51, 130.17,
129.84, 70.40, 65.18, 63.58, 34.30, 32.08, 29.94, 29.88,
29.70, 29.50 (2 × C), 29.37, 29.30 (2 × C), 27.39, 27.34,
25.05, 22.86, 14.29.
Octadec-9-ynoic Acid (3l): Freshly prepared Jones reagent
(2.7 M, 10 mL) was added dropwise to a solution of 2l (2.8
g, 8 mmol) in acetone (50 mL) while maintaining the
reaction temperature at 0–4 °C. Upon the disappearance of
2l, as monitored by TLC (ca. 4 h), i-PrOH (1 mL) was added
to quench excessive Jones oxidant. The chromium salt was
removed by filtration and washed thoroughly with acetone
(50 mL). The filtrate was concentrated in vacuo, and the
residue was submitted to silica gel chromatography (10–
40% EtOAc in hexane) to give 3l13 as a white solid (mp 47–
48 °C) in 95% yield.
Methyl Octadec-9-ynoate (4l): Concentrated H2SO4 (0.1
mL, ca. 96%) was added to a suspension of 3l (2.0 g, 7
mmol) in MeOH (30 mL). The solution was concentrated
after overnight reaction. Purification by silica gel
chromatography (2–5% EtOAc in hexane) gave compound
4l as a light yellow oil in 97% yield. 1H NMR (400 MHz,
CDCl3): d = 3.67 (s, 3 H), 2.30 (t, J = 7.5 Hz, 2 H), 2.13 (t,
J = 6.9 Hz, 4 H), 1.59–1.69 (m, 2 H), 1.42–1.54 (m, 4 H),
1.17–1.42 (m, 16 H), 0.88 (t, J = 6.7 Hz, 3 H).
Methyl (Z)-Octadec-9-enoate (5l): Ni(OAc)2·4H2O (250
mg, 1 mmol) was dissolved in degassed EtOH (10 mL),
followed by the addition of 1,2-ethylenediamine (67 mL,
1 mmol). Sodium borohydride (38 mg dissolved in 950 mL
EtOH and 50 mL of 2 N aq NaOH, 1 mmol) was added and
the solution turned from blue to black in color. The catalyst
was ready for use after cessation of hydrogen release.
A flask containing 4l (1.7 g, 6 mmol) was flushed with
hydrogen. The above freshly prepared P-2 Ni–ethylene-
diamine catalyst (6 mL) was introduced by a gas-tight
syringe. Upon the disappearance of 4l, as monitored by TLC,
the flask was flushed with air to stop the reaction. The
reaction mixture was diluted with EtOAc (6 mL), filtered
through a short pad of silica gel, and concentrated in vacuo.
The residue was purified by silica gel chromatography (2–
5% EtOAc in hexane) to provide 5l14 as a light yellow oil in
99% yield. The ratio of cis-5 to the over-reduced alkane and
the corresponding trans isomer was analyzed by GC–MS
(Agilent G1800C, HP-5MS column, 30 m × 0.25 mm × 0.25
mm). 1H NMR (400 MHz, CDCl3): d = 5.30–5.40 (m, 2 H),
3.67 (s, 3 H), 2.30 (t, J = 7.5 Hz, 2 H), 1.95–2.06 (m, 4 H),
1.56–1.70 (m, 2 H), 1.20–1.38 (m, 20 H), 0.88 (t, J = 6.8 Hz,
3 H).
(12) Fetcher, S.; Ahmad, A.; Perouzel, E.; Heron, A.; Miller,
A. D.; Jorgensen, M. R. J. Med. Chem. 2006, 49, 349.
(13) Gruiec, R.; Noiret, N.; Patin, H. J. Am. Oil Chem. Soc. 1995,
72, 1083.
(14) Crosignani, S.; White, P. D.; Linclau, B. J. Org. Chem. 2004,
69, 5897.
(15) The E-double bond isomers, if present, could be detected on
the 13C NMR spectra taken in CDCl3. Their olefinic carbons
appeared in the downfield region relative to those of Z-
configured 1-MAGs (ca. 0.4–0.6 ppm difference).
Synlett 2011, No. 6, 809–812 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not be copied or
emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.