Synthesis and interfacial behavior of sulfur-containing analogs of lung surfactant dipalmitoyl phosphatidylcholine

Article abstract of DOI:10.1016/j.bmcl.2004.10.001

The synthesis and potential of lipids 2 and 3 to act as lung surfactants is reported. Synthesis methods and initial surface property characterizations are reported for two sulfur-containing phosphonolipids related structurally to dipalmitoyl phosphatidylc

Full text of DOI:10.1016/j.bmcl.2004.10.001

Bioorganic & Medicinal Chemistry Letters 14 (2004) 5983–5986
Synthesis and interfacial behavior of sulfur-containing analogs
of lung surfactant dipalmitoyl phosphatidylcholine
a
Yusuo Chang, Zhengdong Wang, Robert H. Notter, Zhongyi Wang,
a
a
b
b,
Long Qu and Adrian L. Schwan
b,
*
a
Department of Pediatrics, University of Rochester, Rochester, NY 14642, USA
Department of Chemistry, University of Guelph, Guelph, ON, Canada N1G 2W1
b
Received 18 August 2004; revised 28 September 2004; accepted 2 October 2004
Available online 26 October 2004
Abstract—Synthesis methods and initial surface property characterizations are reported for two sulfur-containing phosphonolipids
related structurally to dipalmitoyl phosphatidylcholine (DPPC), the major lung surfactant glycerophospholipid. Sulfur linkages in
these compounds affect molecular interactions relative to ester linkages, and are structurally resistant to cleavage by phospholipases.
The SO -linked analog synthesized here had increased adsorption and improved film respreading compared to DPPC, while reach-
2
ing very low surface tensions (61mN/m) in cycled interfacial films on both the Wilhelmy balance and the pulsating bubble surfac-
tometer. This compound appears to have potential utility as a component in future phospholipase-resistant synthetic exogenous
surfactants for treating clinical forms of inflammatory lung injury.
Ó 2004 Elsevier Ltd. All rights reserved.
1
. Introduction
During inflammatory lung injury, endogenous phospho-
lipases are released that can degrade lung surfactant
glycerophospholipids. One approach to designing novel
synthetic surfactants involves the use of phospholipase-
resistant structural analogs of dipalmitoyl phosphatidyl-
choline (DPPC), the major glycerophospholipid in
Pulmonary surfactant, a complex mixture of glycero-
phospholipids and biophysically active proteins in the
alveolar airsacs, is essential for normal respiratory func-
tion in air-breathing animals. Lung surfactant is defi-
cient in premature infants with the respiratory distress
syndrome (RDS), and can become inactivated or dys-
functional during inflammatory acute lung injury
ALI) and the acute respiratory distress syndrome
ARDS). Life saving therapy with animal-derived exo-
genous surfactant drugs is now available for premature
infants with RDS, and is being extended to patients with
1
,2
1
,3–6
mammalian pulmonary surfactant.
Analog com-
pounds can be synthesized not only to be resistant to
phospholipase activity, but also to have potentially im-
proved surface active properties compared to DPPC.
Our prior work has reported on a highly active diether
phosphonolipid analog of DPPC (designated DEPN-8,
(
(
3
,6
compound 1) and its use in synthetic surfactants.
1
clinical lung injury syndromes. However, significant
The present study reports the synthesis of two sulfur-
containing phosphonolipid analogs (compounds 2 and
3, Fig. 1), along with preliminary assessments of their
interest remains in developing new, active synthetic lung
surfactants with high inhibition resistance for therapeu-
tic applications, particularly for adult and pediatric pa-
tients with severe ALI/ARDS.
Keywords: Lung surfactant, Phospholipid; Surface tension; Sulfur;
Respread.
*
Corresponding author. Tel.: +1 519 8244120x58781; fax: +1 519 766
499; e-mail: schwan@uoguelph.ca
1
Permanent address: College of Chemistry & Chemical Engineering,
Central South University, Changsha 410083, China.
Figure 1. Structure of diether (1) and sulfur-containing phosphono-
lipid analogs (2 and 3) related to DPPC.
0
960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.10.001

5984
Y. Chang et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5983–5986
surface active properties in spread surface films at the
air–water interface and in aqueous dispersions. Com-
pared to the ester group, sulfur linkages have the poten-
tial to significantly alter molecular packing and
interactions at the interface and in the aqueous phase
in addition to providing resistance to endogenous
phospholipases.
2.2. Biophysical properties of sulfur-containing lipids 2
and 3
The biophysical properties of the S-lipid 2 and the SO2-
lipid 3 were examined in comparison to DPPC and to 1
(DEPN-8) by several methods. Surface films spread to
2
˚
a uniform high initial concentration of 15A /molecule
were studied on a Wilhelmy balance to highlight
1
,22,23
respreading behavior,
and adsorption to the air–
2
. Results and discussion
water interface was measured for surfactant dispersions
in a dish with a stirred subphase to minimize diffusion
resistance. Surface pressure–area isotherms on the Wil-
helmy balance showed that both sulfur-containing ana-
logs 2 and 3 had improved film respreading compared
2
4
2
.1. Synthetic chemistry
The sulfur-containing lipids were prepared as shown in
Schemes 1and 2 . Commercially available 1-thioglycerol
was converted to the 1-S-hexadecyl-rac-thioglycerol (4)
by the alkylation with hexadecyl bromide in alcoholic
to DPPC and diether 1 on cycle 2/1( Table 1). SO -lipid
2
3 also had improved film respreading compared to
DPPC and 1 on cycle 7/1(respreading for 2 could not
accurately be defined for cycle 7/1because of substantial
isotherm shape changes that occurred after cycle 3). If
cycle number was not considered, spread films of all four
compounds (1–3 and DPPC) had maximum surface pres-
sures of 72mN/m on the Wilhelmy balance at 23°C. In
studies on dispersed lipids, compounds 2 and 3 had in-
creased adsorption surface pressures compared to 1
and DPPC at 0.25 and 5min (Table 1).
KOH (95%). The primary hydroxyl group was selec-
7
tively tritylated (TrCl, Et N), producing known alcohol
3
5 in 93% yield. The remaining hydroxyl group of 5 was
alkylated with hexadecyl bromide to yield 6 in 94%
yield. The trityl group was cleaved with (pTSA) in
5% aq methanol to provide alcohol 7, a known com-
pound, which could be smoothly oxidized to the corre-
sponding sulfone 8 with MCPBA prior to installation of
the phosphono head group.
9
8
The dynamic surface activity of the sulfur lipids was
examined further at environmental conditions relevant
for the lungs in vivo on a pulsating bubble surfactometer
(37°C, 20cycles/min, 50% area compression). Meas-
The phosophonocholine head group was then installed
through a previously established protocol as shown in
Scheme 2. Final isolation and recrystallization afforded
lipids 2 and 3 in 48% and 59% yields, respectively. Se-
6
25
urements of minimum surface tension on this instrument
have been shown in multiple studies to provide a phys-
iologically relevant assessment of overall lung surfactant
activity that combines effects from both adsorption and
lected synthetic procedures and characterization data
9
are given in the References and notes section. Although
a number of phosphatidylcholines with sulfur-linked
8
,10–21
1
fatty alkyl chains are known,
pounds with thioether linkages,
including some com-
,20,21
lipids 2 and 3
dynamic film compression. Pulsating bubble measure-
8
ments indicated that lipid 3 had overall dynamic surface
activity approaching that of 1 at the surfactant concen-
tration studied (2.5mg/mL) (Fig. 2). The dynamic sur-
face activity of 2 on the pulsating bubble was
significantly less than that of 3 (Fig. 2). This reduced
activity on the bubble apparatus for 2 is presumably re-
lated to the lower maximum surface pressures found in
isotherms of spread surface excess films of this com-
pound during cycles 1–3 on the Wilhelmy balance
(Table 1 legend). Finally, as reported in a number of prior
studies (see Ref. 1 for review), the surface tension lower-
ing ability of DPPC was substantially worse on the pul-
sating bubble apparatus (Fig. 2) than on the Wilhelmy
balance (Table 1). This reduced activity for DPPC in
bubble experiments is due to its extremely poor adsorp-
tion when dispersed in the aqueous phase.
represent the first examples of thioether- or sulfone-con-
taining phosphonocholine derivatives.
OR2
On OC16H33
OH
iii
S
OR1
S
C16H33
C16H33
iv
, R1 = R2 = H
4
7, n = 0
8, n = 2
i
1
2
5
ii
6
, R = Tr; R = H
, R = Tr; R2 = C16H33
1
Scheme 1. Reagents and conditions: (i) TrCl, Et
3
3
N, CH CN/THF,
9
8
3%; (ii) KOH, DMSO, nC16H33Br, 94%; (iii) p-TSA, 95% aq MeOH,
5%; (iv) 2.3equiv MCPBA, CH
2
Cl
2
, 76%.
Although specific molecular mechanisms were not stud-
ied here, the fact that the newly synthesized SO -lipid 3
O
2
On OC16H33
P
Br ii
( )3
had interfacial activity approaching or exceeding that of
compound 1 is significant. A synthetic exogenous surf-
actant containing 1 plus 1.5% by weight of mixed bovine
lung surfactant proteins (SP)-B and C has recently been
shown by our group to have extremely high surface
7
or 8 +
Cl
Cl
S
O
C16H33
P
Br
O
OH
O
P
HO
i
iii
3Br
HO
( )
2
, n = 0; 3, n = 2
activity that is fully maintained in the presence of phosp-
6
holipase A . This synthetic surfactant (1+1.5% SP-B/C)
2
Scheme 2. Reagents and conditions: (i) PCl
iii) Me N, CHCl /CH
: 48%; 3: 59%.
5
, CHCl
3
; (ii) Et
3
N, CHCl
3
;
(
3
3
3
CN/iPrOH, 2 days, 60°C. Yields over all steps:
also had high resistance to biophysical inhibition by
plasma proteins and cellular lipids, which can be present
2

Y. Chang et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5983–5986
5985
Table 1. Respreading ratios, maximum film surface pressures and adsorption surface pressures for sulfur-containing lipids 2 and 3 relative to DPPC
and diether lipid 1
Compound
Wilhelmy balance measurements
Adsorption measurements
Film respreading
cycle 2/1
Film respreading
cycle 7/1
Maximum film
surface pressure
Surface pressure at
0.25min (mN/m)
Surface pressure at
5min (mN/m)
(
mN/m)
DPPC
DEPN-8, 1
28.3 ± 0.3
19.5 ± 1.0
14.3 ± 0.7
0.2 ± 0
48 ± 0.3
33.1 ± 1.5
21.6 ± 0.4
—
72
72
0
1.2 ± 0.2
7.2 ± 0.5
0
SO -Lipid, 3
S-Lipid, 2
2
72
72 (cycles 4–7)
20.0 ± 0.0
8.7 ± 1.3
30.3 ± 1.3
30.7 ± 3.2
˚
2
Data are mean ± standard error for n = 3–5 experiments. Films were spread to 15A /molecule on a Wilhelmy balance (compression ratio 4.35:1, rate
5
units) between compression curves 1, 2 or 1, 7.
min/cycle, 23°C). Maximum surface pressure for 2 was significantly less (ꢀ51–56mN/m) for cycles 6 3. Respreading is based on the area (arbitrary
22,23
An area of 0 between compressions indicates complete respreading, and larger areas indicate less
respreading. For adsorption, surfactants were added at time 0 to a dish with a stirred subphase, and surface pressure was measured with a hanging
1,24
Wilhelmy slide (2.5mg surfactant phospholipid/40mL of subphase, 37°C).
2
3
. Notter, R. H.; Wang, Z. Rev. Chem. Eng. 1997, 13, 1–118.
. Turcotte, J. G.; Lin, W. H.; Pivarnik, P. E.; Sacco, A. M.;
Bermel, M. S.; Lu, Z.; Notter, R. H. Biochim. Biophys.
Acta 1991, 1084, 1–12.
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
DPPC
DEPN-8, 1
SO -Lipid, 3
2
S-Lipid, 2
4
. Turcotte, J. G.; Sacco, A. M.; Steim, J. M.; Tabak, S. A.;
Notter, R. H. Biochim. Biophys. Acta 1977, 488, 235–
2
48.
5. Turcotte, J. G.; Lin, W. H.; Pivarnik, P. E.; Motola, N. C.;
Bhongle, N. N.; Heyman, H. R.; Notter, R. H. Chem.
Phys. Lipids 1991, 58, 81–95.
6. Wang, Z.; Schwan, A. L.; Lairson, L. L.; OÕDonnell, J. S.;
Byrne, G. F.; Foye, A.; Holm, B. A.; Notter, R. H. Am. J.
Physiol. 2003, 285, L550–L559.
2
3
1
0
5
10
Time (minutes)
15
20
7
. Hong, C. I.; Kirisits, A. J.; Nechaev, A.; Buchheit, D. J.;
West, C. R. J. Med. Chem. 1990, 33, 1380–1386.
Figure 2. Minimum surface tension as a function of time for sulfur-
containing lipids compared to DPPC and 1. Surface tension at
minimum bubble radius was measured as a function of time of
pulsation on a bubble surfactometer (37°C, 20cycles/min, 50% area
compression, 2.5mg phospholipid/mL) for surfactant dispersions in
8. Nali, M.; Rindone, B.; Bosone, E.; Farina, P.; Innocenti,
S.; Valcavi, U. Gazz. Chim. Ital. 1986, 116, 25–27.
9. Selected experimental procedures: Synthesis of 1-S-hexa-
decyl-2-O-hexadecyl-rac-sulfonylglycerol (8). 1-S-hexade-
8
cyl-2-O-hexadecyl-rac-thioglycerol (7) (2.0g, 3.7mmol)
0.15M NaCl. Data are mean ± standard error for n = 3–5 experiments.
Data for DPPC are adapted from Ref. 6.
was dissolved in dry CH
(1.93g, 8.6mmol) in CH
Cl
(35mL), and the MCPBA
(60mL) was added to the
2
2
2
Cl
2
sulfide, the reaction mixture was stirred at rt overnight.
Satd Na CO (40mL) was added and the mixture was
2
3
in the alveoli during inflammatory lung injury (ALI/
6
ARDS). Initial biophysical assessments of sulfur-con-
extracted with CH
4
over MgSO , filtered, and evaporated. The residue was
2 2
Cl , washed with water, brine, dried
purified by column chromatography (eluent: hexane/
EtOAc = 5:1) to give alcohol 8, 76% yield, mp: 60–61°C.
taining analogs in the present paper did not address
their properties and interactions with lung surfactant
proteins or related synthetic peptides. Studies are cur-
rently in progress to assess the surface active properties
1
Data for 8: H NMR (CDCl
3
, 400MHz): d 3.96 (m, 1H),
3
(
1
.88 (dt, J = 3.6 and 11.6Hz, 1H), 3.62–3.51 (m, 3H), 3.37
ddd, J = 1.2, 8.4, and 14.8Hz, 1H), 3.09–3.05 (m, 3H),
.84 (m, 3H), 1.58 (m, 2H), 1.41 (m, 2H), 1.34–1.26
of the SO -lipid 3 in combination with purified bovine
2
SP-B, SP-C, and mixed SP-B/C to more fully define its
potential utility as a component in synthetic exogenous
lung surfactants for possible use in ALI/ARDS.
13
(
3
m, 50H), 0.88 (m, 6H); C NMR (CDCl , 100MHz):
d 74.79, 70.20, 62.38, 54.94, 54.69, 31.89, 29.92, 29.67
(several CÕs), 29.52, 29.46, 29.33, 29.30, 29.08, 28.49, 26.16,
2
2.66, 21.84, 14.09. IR (CH
466, 1264cm . Anal. Calcd for C35
2
Cl
2
): m, 3443, 2925, 2853,
S: C, 75.47;
À1
1
H, 13.03. Found: C, 75.58; H, 12.72.
H
72
O
4
Acknowledgements
General procedure for installation of phosphonocholine head
group. Synthesis of Lipids 2 and 3. Using 800–1100mg of
alcohols 7 or 8, the procedure followed a previously
published method, with the exception that after flash
chromatography of the lipid, crystallization was from
The authors gratefully acknowledge the support of
grants HL-56176 and HL-66988 from the National
Institutes of Health.
6
CHCl
04°C (decomp.).
400MHz): d 4.02 (t, J = 5.6Hz), 3.64–3.52 (m, 3H), 3.44
m, 2H), 3.15 (s, 9H), 2.72 (ABX pattern, JAX = 5.6Hz;
BX = 6.0Hz; JAB = 13.6Hz, 2H), 2.59 (t, J = 7.4Hz, 2H),
2.04 (m, 2H), 1.70 (dt, J = 7.2 and 17.2Hz, 2H), 1.58 (m,
3
/acetone = 2:1. Yield of 2: 48% yield, mp: 203–
1
2
H
3 3
NMR (CDCl /CD OD = 1:1,
References and notes
(
J
1
. Notter, R. H. Lung Surfactants: Basic Science and Clinical
Applications; Marcel Dekker: New York, 2000.

5986
Y. Chang et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5983–5986
1
3
4
H), 1.27–1.41 (m, 52H), 0.89 (t, J = 6.8Hz, 6H);
C
12. Yu, L.; Dennis, E. A. Proc. Natl. Acad. Sci. U.S.A. 1991,
88, 9325–9329.
NMR (CDCl /CD OD = 1:1, 100MHz): d 78.16 (d,
3
3
J = 7.1Hz), 69.83, 66.09 (d, J = 13.7Hz), 63.88 (d,
J = 5.6Hz), 52.18, 32.78, 31.31, 29.42, 29.07 (several
CÕs), 28.97, 28.90, 28.74, 28.67, 28.28, 25.48, 22.61,
13. Bhatia, S. K.; Hajdu, J. Tetrahedron Lett. 1987, 28, 3767–
3770.
14. Sutton, L. D.; Froelich, S.; Hendrickson, S. H.; Quinn, D.
M. Biochemistry 1991, 30, 5888–5893.
15. Hendrickson, H. S.; Hendrickson, E. K. Chem. Phys.
Lipids 1990, 53, 115–120.
16. Murata, M.; Ikoma, S.; Achiwa, K. Chem. Pharm. Bull.
1991, 39, 1335–1336.
3
1
1
1
1
6.72,
13.17;
P
NMR
3 3
(CDCl /CD OD = 1:1,
62MHz): d 26.1; IR (Nujol): m, 2963, 2906, 1412, 1261,
096cm
À1
.
HRMS, + ESI (+ve), m/z: calcd for
PS [M+H] : 720.6093; found: 720.6126.
C
41
H87NO
4
1
Yield of 3: 59% yield, mp: 234–235°C (decomp.).
NMR (CDCl /CD
H), 3.83 (m, 1H), 3.70 (dt, J = 6.4 and 9.2Hz, 1H), 3.49
H
3
3
OD = 1:1, 400MHz): d 4.07–3.95 (m,
17. Murata, M.; Uchida, H.; Achiwa, K. Chem. Pharm. Bull.
1992, 40, 2849–2851.
2
(dt, J = 6.4 and 8.8Hz, 1H), 3.41–3.25 (m, 4H), 3.17–3.08
(m, 2H), 3.11 (s, 9H), 1.98 (m, 2H), 1.82 (m, 2H), 1.62–
18. Hajdu, J.; Sturtevant, J. M. Chem. Phys. Lipids 1990, 55,
323–330.
1.53 (m, 4H), 1.41 (m, 2H), 1.26 (m, 50H), 0.86 (t,
J = 6.8Hz, 6H);
19. Kan, C. C.; Bittman, R.; Hajdu, J. Biochim. Biophys. Acta
1991, 1066, 95–101.
1
3
C
3 3
NMR (CDCl /CD OD = 1:1,
1
00MHz): d 73.77 (d, J = 7.1Hz), 69.73, 66.29 (d,
20. Morris-Natschke, S.; Surles, J. R.; Daniel, L. W.; Berens,
M. E.; Modest, E. J.; Piantadosi, C. J. Med. Chem. 1986,
29, 2114–2117.
21. Letourneux, Y.; Bourass, J.; Boucrot, P.; Elkihel, L.; Petit,
J. Y. Pharmacol. Res. 1997, 35, 73–78.
22. Wang, Z.; Gurel, O.; Baatz, J. E.; Notter, R. H. J. Lipid
Res. 1996, 37, 1749–1760.
J = 13.5Hz), 61.84 (d, J = 5.5Hz), 54.48, 54.20, 52.12,
3
2
1.27, 29.45, 29.04 (several CÕs), 29.00, 28.93, 28.77, 28.70,
8.50, 27.89, 25.65, 23.11, 21.99, 21.75, 20.98, 17.06, 13.15;
P NMR (CDCl /CD OD = 1:1, 162MHz): d 23.2; IR
3
1
3
3
À1
(Nujol): m, 2963, 2907, 1412, 1262, 1085cm ; HRMS, ESI
TOF (+ve), m/z: calcd for
+
41 6
C H87NO PS [M+H] :
7
52.5992; found: 752.5974.
23. Wang, Z.; Hall, S. B.; Notter, R. H. J. Lipid Res. 1995, 36,
1283–1293.
1
0. Cox, J. W.; Horrocks, L. A. J. Lipid Res. 1981, 22, 496–
5
05.
24. Notter, R.; Taubold, R.; Finkelstein, J. Chem. Phys.
Lipids 1983, 33, 67–80.
25. Enhorning, G. J. Appl. Physiol. 1977, 43, 198–203.
1
1. Hendrickson, E. K.; Hendrickson, H. S. Chem. Phys.
Lipids 2001, 109, 203–207.