NMR-based molecular ruler for determining the depth of intercalants within the lipid bilayer. Part I. Discovering the guidelines

Chemistry and Physics of Lipids 155 (2008) 98–113

Contents lists available at ScienceDirect

Chemistry and Physics of Lipids

journal homepage: www.elsevier.com/locate/chemphyslip

NMR-based molecular ruler for determining the depth

of intercalants within the lipid bilayer

Part I. Discovering the guidelines

Yael Cohen, Efrat Bodner, Michal Richman, Michal Afri, Aryeh A. Frimer^∗

Ethel and David Resnick Chair in Active Oxygen Chemistry, Bar-Ilan University, Ramat Gan 52900, Israel

a r t i c l e i n f o

a b s t r a c t

Article history:

Received 3 April 2008

Received in revised form 6 July 2008

Accepted 8 July 2008

Available online 15 July 2008

The development of “molecular rulers” would allow one to quantitatively locate intercalants within the

liposomal bilayer. To this end, we have attempted to correlate the ¹³C NMR chemical shift of a polarizable

“reporter” carbon (e.g., carbonyl) of the intercalant—with the E_T(30) polarity it experiences, and with its

Angstrom distance from the interface. This requires families of molecules with the same two “reporter

carbons” separated by a ﬁxed distance, residing at various depths/polarities within the bilayer. The fami-

lies studied included 4,4-dialkylcyclohexa-2,5-dienones 1, benzenediacetic esters 15, benzenedipropionic

esters 17, 4-alkoxybenzaldehydes 19 and methyl 4-alkoxybenzoates 22. These compounds possessed

the following characteristics: (1) a planar backbone; (2) polar/hydrophilic “head” groups; (3) modular

hydrophobic tails; (4) large changes in the ¹³C NMR chemical shift (ꢀı) of the reporter atoms with sol-

vent polarity. These studies revealed a ﬁfth requirement, namely: (5) the reporter carbons must not be

strongly conjugated, lest it reﬂect the charge build-up at another site within the conjugated system.

Keywords:

Liposome

¹³C NMR

E_T(30)

DMPC

Membrane

Molecular ruler

1. Introduction

microenvironment polarity felt by the polar carbon in a particular

situation.

We have been exploring the active oxygen chemistry of sub-

strates intercalated within the phospholipid bilayer of model

liposomes (Frimer et al., 1983, 1996; Frimer, 1985; Strul et al.,

1993, 1994; Weitman et al., 2001; Afri et al., 2002, 2004a,b;

Bronshtein et al., 2004; Gamliel et al., 2008 ). Our studies indi-

cate that there is a clear correlation between the location of

a substrate within the bilayer and its biochemical activity. In

order to determine the location of an intercalant within the lipid

bilayer, we have used an NMR technique which is based on two

observations (Maciel and Ruben, 1963; Maciel and Natterstad,

1965; Ueji and Makamaura, 1976; Menger et al., 1978, 1988;

Menger, 1979; Janzen et al., 1989). Firstly, an excellent cor-

relation exists between the ¹³C NMR chemical shift (ı) of a

polarizable carbon (e.g., the carbonyl or nitronyl carbon) and

the Dimroth–Reichardt (Dimroth et al., 1963; Reichardt, 1965,

1988, 1994) E_T(30) polarity parameter of the solvent in which

the spectrum was obtained (Gamliel, 2005). Once this chemi-

cal shift–polarity correlation has been determined, one can work

backwards, utilizing the observed chemical shift to determine the

The second observation is that the liposomal bilayer con-

sists of a gradient of polarities ranging from that of water

(E_T(30) = 63.1 kcal/mol) at the lipid/water interface down to hexane

(E_T(30) = 31.0 kcal/mol) within the fatty acid chains. (Presumably,

there is no change in polarity within the lipid slab.) The chemical

shift–polarity correlation makes it possible to use chemical shifts as

a polarity gauge thereby yielding a qualitative estimate of the sub-

strate’s distance from the interface. This then allows us to correlate

a substrate’s average location within the bilayer with its reactivity.

We use the word “average” advisedly. The timescale of molecular

motion is faster than that of the NMR experiment; hence, the val-

ues we obtain by the above method are actually average locations.

The NMR chemical shift data will reﬂect polarity due to electro-

static ﬁelds from the time-averaged structure, as well as from water

penetration and perhaps partial exit of the surfactants.

Given this caveat, we nevertheless believe that this ¹³C NMR

chemical shift/polarity correlation technique is a useful tool for

approximating the location of substrates within lipid bilayers.

Indeed, this technique has enabled us to determine the location

of a variety of biologically active compounds including the natural

antioxidants vitamin E and ubiquinol (Afri et al., 2004a), the pho-

todynamic cancer therapy agent hypericin (Weitman et al., 2001)

and various derivatives of dichlorodihydroﬂuorescein (DCFH) (Afri

et al., 2004b) involved in oxyradical detection.

∗

Corresponding author at: Department of Chemistry, Bar-Ilan University, Ramat

Gan 52900, Israel. Tel.: +972 3 5318610; fax: +972 3 7384053.

E-mail address: frimea@mail.biu.ac.il (A.A. Frimer).

doi:10.1016/j.chemphyslip.2008.07.004

Y. Cohen et al. / Chemistry and Physics of Lipids 155 (2008) 98–113

99

Despite the above success, still lacking was a means for con-

verting this qualitative data into quantitative values, namely, the

vertical distance in Å from the water phase. To this end, we needed

to determine whether any correlation – linear, stepwise or other-

wise – existed between the polarity felt by a given carbon and its

depth. In order to map the bilayer, we required molecules of dif-

fering lipophilicities, residing at various depths within the bilayer,

with two or more “reporter carbons.” This would allow us to link

two or more vertical points within a bilayer by both distance (deter-

mined by ab initio or molecular mechanics calculations) and E_T(30)

polarity values. Such an assortment of molecules would eventually

enable us to develop a “molecular ruler” correlating E_T(30) polarity

data with Angstrom distances from the water phase.

The literature does record other methods for measuring

the depth of an organic molecule in non-homogeneous media,

notably London’s ﬂuorescence-based Parallax method (Abrams

and London, 1992; Chattopadhyay and London, 1987; Asuncion-

Punzalan and London, 1995), and Walker and Steel’s solvatochromic

UV technique (Walker and Steel, 2003; Steel et al., 2002). The major

drawback of these techniques is that the ﬂuorescence or UV absorp-

tion arises from a reporter chromophore or molecule, leading to

very low precision in locating the source of the spectrum. By con-

trast, the aforementioned NMR technique uses a single carbon as

the reporter element.

the various derivatives, and not necessarily according to the rules

of nomenclature. The NMR numbering for compounds 1, 2, 6, 9, 11,

15, 17, 19, 22 and 23 is exempliﬁed in Scheme 1.

2.2. General procedure for preparation of DMPC liposomal

solutions

All glassware was ﬁrst rinsed with conc. HCl to remove all traces

of detergents and then with doubly distilled water. In a typical lipo-

some preparation, the compounds to be intercalated (henceforth

dubbed “intercalants”) and DMPC in a molar ratio of 1:5 (except for

the case of compound 1, where the ratio was 1:4) were dissolved in

chloroform in a vial. The solvent was then evaporated with a gen-

tle steam of N₂while rotating in the palm of the hand, leaving a

uniformly thin layer of lipid on the walls of the vial. The vial was

ﬁnally set in a cotton-packed RB ﬂask and the solvent was removed

by rotary evaporation, leaving a uniformly thin ﬁlm layer of lipid on

the walls of the vial. The vial was then charged with 0.1 M phosphate

buffer with 10% D₂O (pH 7.8) yielding a cloudy solution in the exact

desired molar concentration. The lipid solution was vortexed for

10 min to obtain multilamellar liposomes. The liposomal solution

was then sonicated until a clear solution of essentially unilamellar

liposome was obtained. Sonication was effected with a MSE Tita-

nium Probe Ultrasonic Disintegrator model MK2 at 20 kHz output

frequency. Previous cryo-TEM work (Afri et al., 2004a) conﬁrms that

these conditions produce primarily unilamellar liposomes.

Our efforts towards developing such a molecular ruler are

described in this and the following companion paper (Cohen et al.,

2008a) The former describes our efforts in developing reliable cri-

teria for reporter molecules and selecting appropriate candidates,

while the latter describes our initial results in developing a molec-

ular ruler.

To verify that the various intercalants lie within the lipid bilayer

and not the aqueous phase, the liposomes were centrifuged down

(25,000 × g for 15 min) to a lipid pellet. The supernatant liquid was

decanted and replaced by buffer, and the pellet was redispersed by

vortexing the sample. NMR spectra of the starting and ﬁnal liposo-

mal and supernatant solutions revealed that the substrate indeed

resides within the lipid bilayer exclusively.

2. Materials and methods

2.1. General

2.3. Substrates—preparation and spectral data

Ab initio calculations were performed with Gaussian 03 per-

formed at the B3LYP level using 6–31G* basis set. Gaussian, Inc.,

Pittsburgh PA, 2003. Molecular modeling calculations were carried

out with PCMODEL version 7.50.00, Serena Software, Bloomington,

Indiana, USA—which uses the MMX force ﬁeld.

2.3.1. 4,4-Dialkylcyclohexa-2,5-dien-1-one (1)

The title dienones were prepared by a modiﬁcation of the

Zimmerman procedure (Zimmerman et al., 1971), as follows: 4,4-

dialkylcyclohexenone (6, 0.013 mol), DDQ (5.9 g 0.026 mol), and

puriﬁed dioxane (45 mL) were heated under nitrogen at 95–100 ^◦C

for 22 h. The cooled reaction solution was poured into a 1:1 mix-

ture of ether/pentane (40 mL) and washed with 5% aqueous sodium

hydroxide and water (40 mL). The organic phase was dried over

MgSO₄and rotary evaporated to dryness. The crude product was

chromatogr aphed on silica g el, eluting with petroleum ether/ethyl

acetate (90:10), yielding the desired products as oils in 75–85% yield

for 1a –d, but substantially lower for 1e–h. Dienones 1a (Bordwell

and Wellman, 1963; Zimmerman et al., 1967; Gramlich, 1979) 1c

(Dreiding, 1957) and 1d (Bordwell and Wellman, 1963; Gramlich,

1979; Zimmerman and Schuster, 1962) are known.

The NMR spectra were recorded on either 300 MHz (¹H) and

75 MHz (¹³C) or 50.3 MHz (¹³C). NMR spectra were generally taken

at 25 1 ^◦C, with the exception of the dimyristoylphosphatidyl-

choline (DMPC) liposome solutions which were run on 150.9 MHz

(

¹³C) at 45 ^◦C, above the phase transition temperature (T_C) of DMPC

(Zachariasse et al., 1981). Below this temperature, the mobility of

the intercalated molecules is low and the resulting NMR absorp-

tions are very broad. Raising the temperature sharpens the peaks

but does not seem to affect the chemical shifts. The NMR spec-

tra were generally recorded while locked on the deuterium signals

of the respective solvent. The chemical shifts were measured rela-

tive to internal tetramethylsilane (TMS) or solvent (Gottlieb et al.,

1997), except in the case of the aqueous liposome solutions in which

we calibrated the spectrum according to the trimethylammonium

peak at 54.6 ppm. KH₂PO₄and KOH were used in the preparation

of a 0.1 M phosphate buffer solution (pH 7.8 and containing 10⁻⁴M

EDTA). The general procedure for the preparation of DMPC lipo-

somal suspensions for NMR studies has been previously described

(Afri et al., 2002). The substrate concentration was generally 0.05 M

and the intercalant:lipid molar ratio was 1:5, except for the case

of compound 1, where the substrate concentration was 0.009 M

and the intercalant:lipid molar ratio was 1:4. The designation “t”

refers to a second order triplet resulting from virtual coupling. In

the assignments below, the carbons and attached hydrogens were

numbered to allow convenient comparison of the spectral data of

2.3.1.1. 4,4-Diethylcyclohexa-2,5-dien-1-one (1b). ¹H NMR (CDCl₃) ı

6.69 (AA^ꢀXX^ꢀ, 2H, H₃), 3.64 (d, J = 10.22 Hz, 2H, H₂), 1.69 (q, J = 7.5 Hz,

4H, H₅), 0.75 (t, J = 7.5 Hz, 6H, H₆); ¹³C NMR (CDCl₃) ı 186.3 (C₁),

154.4 (C₃), 130.2 (C₂), 46.7 (C₄), 31.8 (C₅), 8.5 (C₆); MS (EI) m/z 150.1

(M, 50.8%); HRMS (EI) m/z calcd for C₁₀H₁₄O (M⁺) 150.1044, found

150.1014; IR (KBr) 1665.24 (CO stretch) cm⁻¹

.

2.3.1.2. Spiro[5.5]undec-1,4-dien-3-one (1c). This compound is

known in the literature (Dreiding, 1957), but some of the spectral

data are lacking. ¹H NMR (CDCl₃) ı 7.08 (AA^ꢀXX^ꢀ, 2H, H₃), 6.25

(d, J = 10.14 Hz, 2H, H₂), 1.61 (m, 10H, H₅–H₇); ¹³C NMR (CDCl₃) ı

186.2 (C₁), 155.2 (C₃), 128.1 (C₂), 40.9 (C₄), 35.5 (C₅), 25.5 (C₆), 21.6

(C₇); MS (CI) m/z 163 (MH⁺, 81%); HRMS (CI) m/z calcd for C₁₁H₁₅

O

100

Y. Cohen et al. / Chemistry and Physics of Lipids 155 (2008) 98–113

Scheme 1. The NMR numbering for compounds 1, 2, 6, 9, 11, 15, 17, 19, 22 and 23.

(MH⁺) 163.1214, found 163.1223; IR (KBr) 1663.46 (CO stretch)

cm⁻¹

22.5(C₆), 22.7 (C₈), 14.2 (C₉); HRMS (EI) m/z calcd for C₁₆H₂₆O (M⁺)

.

234.1983, found 234.1988; IR (KBr) 1665.37 (CO stretch) 1622.90

(C C) cm⁻¹

.

2.3.1.3. 4,4-Dibutylcyclohexa-2,5-dien-1-one (1e). ¹H NMR (CDCl₃)

ı 6.7 (AA^ꢀXX^ꢀ, 2H, H₃), 6.34 (AA^ꢀXX^ꢀ, 2H, H₂), 1.617 (m, 4H, H₅), 1.24

(m, 4H, H₇) 1.09 (m, 4H, H₆), ı 0.84 (“t”, 6H, H₈); ¹³C NMR (CDCl₃) ı

187.0 (C₁), 155.6 (C₃), 130.2 (C₂), ı 46.5 (C₄), 39.8 (C₅), 26.8 (C₆), 23.1

(C₇), 13.9 (C₈); HRMS (EI) m/z calcd C₁₄H₂₂O (M⁺) 206.1670, found

2.3.1.5. 4,4-Dihexylcyclohexa-2,5-dien-1-one (1g). ¹H NMR (CDCl₃)

ı 6.69 (AA^ꢀXX^ꢀ, 2H, H₃), 6.34 (AA^ꢀXX^ꢀ, 2H, H₂), 1.63 (m, 4H, H₅),

1.21 (m, 12H, H₇, H₉) 1.18 (m, 4H, H₆), 0.86 (“t”, 6H, H₁₀); ¹³C

NMR (CDCl₃) ı 187.1(C₁), 155.6 (C₃), 130.3 (C₂), 46.7 (C₄), 40.2 (C₅),

31.3 (C₈), 29.8 (C₇), 24.7 (C₆), 22.7 (C₉), 14.1 (C₁₀); HRMS (EI) m/z

Calcd C₁₈H₃₀O (M⁺) 262.2296, found 262.2287; IR (KBr) 1665.77

206.1667; IR (KBr) 1665.87 (CO stretch), 1624.50 (C C) cm⁻¹

.

(CO stretch) 1623.70 (C C) cm⁻¹

.

2.3.1.4. 4,4-Dipentylcyclohexa-2,5-dien-1-one (1f). ¹H NMR (CDCl₃)

ı 6.74 (AA^ꢀXX^ꢀ, 2H, H₃), 6.34 (AA^ꢀXX^ꢀ, 2H, H₂), 1.63 (m, 4H, H₅), 1.22

(m, 8H, H₇, H₈) 1.18 (m, 4H, H₆), 0.86 (“t”, 6H, H₉); ¹³C NMR (CDCl₃)

ı 187.1(C₁), 155.6 (C₃), 130.3 (C₂), 46.7 (C₄), 40.1 (C₅), 32.3 (C₇),

2.3.1.6. 4-Dodecyl-4-decylcyclohexa-2,5-dien-1-one (1h). ¹H NMR

(CDCl₃) ı 6.72 (AA^ꢀXX^ꢀ, 2H, H₃), 6.34 (AA^ꢀXX^ꢀ, 2H, H₂), 1.63 (m, 4H,

Y. Cohen et al. / Chemistry and Physics of Lipids 155 (2008) 98–113

101

ꢀ

H₅), 1.21 (m, 32H, H₇–H₁₃, H₇–H₁₃, H₁₄and H₁₅), 1.19 (m, 4H, H₆

2.3.3. 4,4-Dialkylcyclohex-2-en-1-ones 6

and H₆), 0.86 (“t”, 6H, H₁₆and H₁₄); ¹³C NMR (CDCl₃) ı 186.7 (C₁),

Dialkylcylohexenones 6a–h were synthesized as precursors

for the preparation of 4,4-dialkylcyclohexa-2,5-dien-1-one (1),

described in the next section. The dialkylcylohexenones were pre-

pared from the abovementioned dialkylacetaldehydes 2 (0.1 mol

scale) and methyl vinyl ketone (3 ), following the general proce-

dure of Dauben and coworkers for the synthesis of 6a (Dauben

et al., 1968 ). In the case of enones 6a –d , the crude pr oduct was

chromatographed on silica gel, eluting with petroleum ether/ethyl

acetate (90:10), giving the desired product in a 70–85% yield. Cyclo-

hexenones 6a (Zimmerman and Schuster, 1962; Chan and Epstein,

1973; Gramlich, 1979), 6b (Chaki et al., 2002), 6c (Snider et al., 1993)

and 6d (Zimmerman et al., 1971; Chan and Epstein, 1973; Frimer et

al., 1989) are well known in the literature. In the case of the longer

chain enones 6e–h, the yields were substantially lower, presumably

due to steric hindrance. The product was used in its crude state for

the preparation of dienones 1; nevertheless, their identity was con-

ﬁrmed by the following characteristic peaks: ¹H NMR (CDCl₃) ı 6.7

(d, J = 10 Hz, 1H, H₃), 5.9 (d, J = 10 Hz, 1H, H₂); ¹³C NMR (CDCl₃) ı 200

(C₁), 159 (C₃), 128 (C₂). For convenience and clarity in the spectral

data, the carbons were numbered as shown in Scheme 1.

ꢀ

155.4 (C₃), 130.1 (C₂), 46.5 (C₄), 40.0 (C₅and C₅), 31.9 (C₁₄and C₁₂),

ꢀ

30.0, 29.62, 29.57 29.4 (C₇–C₁₁, C₇–C₁₁, C₁₂and C₁₃), 24.6 (C₆and

ꢀ

C₆), 22.7 (C₁₅and C₁₃), 14.1 (C₁₆and C₁₄); HRMS (DCI, CH₄) m/z

calcd for C₂₈H₅₀O (M⁺) 402.3861, found 402.3860; IR (KBr) 1660.48

(CO stretch), 1624.10 (C C) cm⁻¹

.

2.3.2. Dialkylacetaldehydes 2

The title compounds were required as precursors for the synthe-

sis of dialkylcylohexenones 6. Symmetrical aldehydes 2a–d (R R^ꢀ)

are commercially available. Symmetrical aldehydes 2e–g were pre-

pared (Taber et al., 1989; Hu and Mattern, 2000) via the acid

catalyzed hydrolysis of the below-synthesized enol ethers 9, as fol-

lows: Aqueous HCl (5 mL of a 10% solution) was added to the enol

ether 9 (0.05 mol) and reﬂuxed in 20 mL THF. After 12 h, the reaction

mixture was partitioned between ether and water. The combined

organic phases were dried with MgSO₄and concentrated in vacuo

to give the aldehydes 2 in a 70% yield. Asymmetrical aldehyde 2h

was essentially prepared as described by Corey and Suggs (Corey

and Suggs, 1975) beginning with the commercially available alcohol

2-decyltetradecan-1-ol 10. The desired product was isolated by ﬁl-

tration of the organic extracts through a silica sinter and the solvent

was removed by evaporation. The aldehydes underwent extremely

facile autoxidation to the corresponding carboxylic acids (11) and

had to be used soon after their preparation. In the HRMS the parent

peak was essentially absent; hence, for analysis purposes we relied

on the HRMS analysis of the derivative acids 11. We note that the

NMR spectra of the derived acids contain varying amounts of the

corresponding aldehyde.

2.3.3.1. 4,4-Diethylcyclohex-2-en-1-one (6b). This compound is

known in the literature (Chaki et al., 2002), but some of the spectral

data are lacking. ¹H NMR (CDCl₃) ı 6.72 (d, J = 10.28 Hz, 1H, H₃), 5.93

(d, J = 10.28 Hz, 1H, H₂), 2.44 (t, J = 7.12 Hz, 2H, H₆), 1.85 (t, J = 7.12 Hz,

2H, H₅), 1.51 (q, J = 7.52, 4H, H₇), 0.89 (t, J = 7.52 Hz, 6H, H₈); ¹³C NMR

(CDCl₃) ı 199.8 (C₁), 158.9 (C₃), 128.0 (C₂), 38.3 (C₄), 33.9 (C₆), 30.0

(C₅), 29.4 (C₇), 8.2 (C₈); MS (CI, CH₄) m/z 153 (CI, MH⁺, 90%); HRMS

(CI, CH₄) m/z calcd for C₁₀H₁₇O (MH⁺) 153.1279, found 153.1272.

2.3.4. Enol ethers 9e–g

2.3.2.1. 2-Butylhexanal (2e). ¹H NMR (CDCl₃) ı 9.519 (d, J = 2.4 Hz,

1H, H₁), 2.184 (ttd, J = 9.5, 7 and 3 Hz, 1H, H₂), 1.623 (m, 2H, H_3aand

The title compounds were prepared as precursors to the cor-

responding aldehydes 6, following procedures adapted from the

literature (Taber et al., 1989; Hu and Mattern, 2000). n-Butyllithium

(31 mL, 0.0502 mol, 1.6 M in hexanes) was added dropwise to

methoxymethyltriphenylphosphonium chloride (18.1 g, 0.053 mol)

in anhydrous ether (50 mL) at 0 ^◦C. The ice bath was removed, and

the deep red solution was stirred at room temperature for 4 h. The

solution was again cooled to 0 ^◦C and the commercially available

dialkyl ketone 7 (0.0176 mol) in ether (20 mL) was added drop-

wise. The solution was allowed to stir at room temperature for 48 h.

Following a water extraction, the precipitated phosphorous salts

were ﬁltered off, the organic ﬁltrate was concentrated in vacuo and

cooled to 10 ^◦C, at which point more phosphorous salts precipitated.

The supernatant liquid was chromatographed and eluted with hex-

ane to give the desired enol ether in 35–40% yield. For convenience

and clarity in the spectral data, the carbons were numbered as

shown below.

ꢀ

H_3a), 1.415 (m, 2H, H_3band H_3bꢀ ) 1.256 (m, 8H, H₄, H₄, H₅and H₅),

0.853 (“t”, 6H, H₆and H₆); ¹³C NMR (CDCl₃) ı 205.5 (C₁), 51.9 (C₂),

ꢀ

29.5, 28.6 (C₃, C₃, C₄and C₄), 22.7 (C₅and C₅), 13.8 (C₆and C₆);

HRMS m/z (EI) calcd for C₁₀H₂₀O (M⁺) 156.1514, found 156.1508 IR

(KBr) 1737.72 cm⁻¹(CO stretch).

2.3.2.2. 2-Pentylheptanal (2f). ¹H NMR (CDCl₃) ı 9.53 (d, J = 3 Hz,

ꢀ

1H, H₁), 2.19 (ttd, J = 9.5, 7 and 3 Hz, 1H, H₂), 1.58 (m, 2H, H_3a, H_3a),

ꢀ

1.42 (m, 2H, H_3b, H_3bꢀ ) 1.25 (m, 12H, H₄–H₆and H₄–H₆), 0.85 (“t”,

6H, H₇and H₇); ¹³C NMR (CDCl₃) ı 205.8 (C₁), 52.1 (C₂), 32.0 (C₅

ꢀ

and C₅), 29.0 and 26.9 (C₄, C₄, C₅and C₅), 22.6 (C₆and C₆), 14.1 (C₇

and C₇); HRMS (EI) m/z calcd for C₁₂H₂₃O (M⁺−1) 183.1749, found

ꢀ

183.1746. IR (KBr) 1727.41 cm⁻¹(CO stretch).

2.3.2.3. 2-Hexyloctanal (2g). ¹H NMR (CDCl₃) ı 9.54 (d, J = 3, 1H,

ꢀ

H₁), 2.20 (ttd, J = 9.5, 7 and 3 Hz, 1H, H₂), 1.58 (m, 2H, H_3aand H_3a),

2.3.4.1. 2-Butyl-1-methoxy-1-hexene (9e). ¹H NMR (CDCl₃) ı 5.75

ꢀ

1.43 (m, 2H, H_3band H_3bꢀ ) 1.26 (m, 16H, H₄–H₇and H₄–H₇), 0.86

ꢀ

(s, 1H, H₂), 3.51 (s, 3H, H₁), 2.05 (m, 2H, H₄), 1.86 (m, 2H, H₄), 1.33

(“t”, 6H, H₈and H₈); ¹³C NMR (CDCl₃) ı 205.8(C₁), 52.1 (C₂), 31.8

ꢀ

(m, 8H, H₅, H₅, H₆and H₆), 0.91 (“t”, 6H, H₇, H₇); ¹³C NMR (CDCl₃)

ꢀ

(C₆and C₆), 29.5, 29.1 (C₃–C₅and C₃–C₅), 27.2 (C₆and C₆), 22.7

ı 141.9 (C₂), 118.8 (C₃), 59.1 (C₁), 31.2, 30.5 and 30.0 (C₄, C₅and

C₅), 26.5 (C₄), 22.7 and 22.4 (C₆and C₆), 13.9 (C₇and C₇); IR (KBr)

+

ꢀ

(C₇and C₇), 14.2 (C₈and C₈); MS (CI, ammonia) m/z 213 (MH ); IR

ꢀ

(KBr) 1732.69 cm⁻¹(CO stretch).

1666 cm⁻¹(C C stretch).

2.3.2.4. 2-Decyltetradecanal (2h). ¹H NMR (CDCl₃) ı 9.54 (d, J = 3 Hz,

1H, H₁), 2.20 (ttd, J = 9.5 Hz, 7 and 3 Hz, 1H, H₂), 1.59 (m, 2H, H_3a

2.3.4.2. 1-Methoxy-2-pentyl-1-heptene (9f). ¹H NMR (CDCl₃) ı 5.73

ꢀ

(s, 1H, H₂), 3.50 (s, 3H, H₁), 2.03 (m, 2H, H₄), 1.85 (m, 2H, H₄), 1.29

and H_3a), 1.45 (m, 2H, H_3band H_3bꢀ ) 1.25 (m, 36H, H₄–H₁₁, H₄–H₁₁

,

(m, 12H, H₅–H₇, and H₅–H₇), 0.89 (“t”, 6H, H₈and H₈); ¹³C NMR

ꢀ

H₁₂and H₁₃), 0.86 (“t”, 6H, H₁₄and H₁₂); ¹³C NMR (CDCl₃) ı 206.0

(CDCl₃) ı 142.0 (C₂), 119.1 (C₃), 59.3 (C₁), 32.0, 31.8, 31.6 (C₄, C₆

ꢀ

(C₁), 52.2 (C₂), ı 32.1 (C₁₂and C₁₀), 29.9, 29.8, 29.7, 29.6, 29.5, 29.1

and C₆), 28.1, 27.6 and 26.5 (C₄, C₅and C₅), 22.7 (C₇and C₇), 13.9

and 27.3 (C₃–C₉, C₃–C₉, C₁₀and C₁₁), 27.2 (C₆and C₆), 22.9 (C₁₃

and C₁₁), 14.3 (C₁₄and C₁₂); IR (KBr) 1727.41 cm (CO stretch).

(C₈and C₈); HRMS m/z (EI) calcd for C₁₃H₂₆O (M⁺) 198.1983, found

ꢀ

198.1979. IR (KBr) 1666.77 cm⁻¹(C C stretch).

−1

ꢀ

102

Y. Cohen et al. / Chemistry and Physics of Lipids 155 (2008) 98–113

2.3.4.3. 2-Hexyl-1-methoxy-1-octene (9g). ¹H NMR (CDCl₃) ı 5.73

2.3.5.1. 4-(2-Methylbutoxycarbonylmethyl)phenylacetic acid methyl

ester (15a, n = 1, m = 0). 25.8% yield; R_f= 0.23; ¹H NMR (acetone-

d₆) ı 7.33 (s, 4H, H₄and H₅), 4.00, 4.08, 1.81 (ABX system

1H each, J_AB= 10, J_BX= 7, J_AX= 6 Hz, 2H₉, H₁₀), 3.71 (s, 4H, H₂

and H₇), 3.70 (s, 3H, H₁), 1.53 (dqd, J = 15, 7.5 and 6 Hz, 1H,

H₁₁), 1.28 (dqd, J = 15, 7.5 and 6 Hz, 1H, H₁₁), 1.01 (d, J = 6.5 Hz,

ꢀ

(s, 1H, H₂), 3.50 (s, 3H, H₁), 2.03 (m, 2H, H₄), 1.85 (m, 2H, H₄),

1.28 (m, 16H, H₅–H₈and H₅–H₈), ı 0.88 (“t”, 6H, H₉and H₉); ¹³C

NMR (CDCl₃) ı 141.9 (C₂), 118.9 (C₃), 59.2 (C₁), 27.1–31.9 (C₄–C₇

ꢀ

and C₄–C₇), 22.8, 22.7 (C₈and C₈), 14.2, 14.1 (C₉and C₉); HRMS

m/z (EI) calcd for C₁₅H₃₀O (M⁺) 226.2296, found 226.2296; IR (KBr)

1667 cm⁻¹(C C stretch).

3H, H₁₁or H₁₂), 0.99 (t, J = 7.5 Hz, 3H, H₁₂or H₁₁); ¹³C NMR

(acetone-d₆) ı 172.2 (C₁), 171.8 (C₈), 134.3 (C₃or C₆), 134.1

(C₆or C₃), 130.19 (C₄or C₅), 130.16 (C₅or C₄), 69.6 (C₉), 52.0

ꢀ

2.3.4.4. 2-Pentylheptanoic acid (11f). ¹H NMR (CDCl₃) ı 2.23 (m, 1H,

ꢀꢀ

(C₁), 41.2 (C₂or C₇), 40.9 (C₇or C₂), 35.0 (C₁₀), 26.6, 16.6 and

ꢀ

H₂), 1.58 (m, 2H, H_3a, H_3a), 1.42 (m, 2H, H_3b, H_3bꢀ ), 1.25 (m, 12H,

11.5 (C₁₁, C₁₁and C₁₂); MS (CI, CH₄) m/z 278 (M⁺, 12.54%),

ꢀ

H₄–H₆and H₄–H₆), 0.85 (“t”, 6H, H₇and H₇); ¹³C NMR (CDCl₃)

ꢀ

219 (M−COOCH₃, 22.4%), 163 (M−COOCH₂CH(CH₃)CH₂CH₃,

98.8%), 104 (M−COOCH₃,COOCH₂CH(CH₃)CH₂CH₃, 49.4%), 208

(MH−CH₂(CH₃)CH₂CH₃, 100%); HRMS (CI, CH₄): calcd (C₁₆H₂₂O₄,

M⁺) 278.1518, obsd 278.1503.

ꢀ

ı 182.3 (C₁), 45.7 (C₂), 32.0 (C₅and C₅), 29.0 and 26.9 (C₄, C₄, C₅

ꢀ

and C₅), 22.6 (C₆and C₆), 14.1 (C₇and C₇); HRMS (EI) m/z calcd for

C

₁₂H₂₅O₂(MH⁺) 201.1855, found 201.1851; IR (KBr) 1701.74 cm⁻¹

(CO stretch).

2.3.5.2. 4-(2-Ethylbutoxycarbonylmethyl)phenylacetic acid methyl

2.3.4.5. 2-Hexyloctanoic acid (11g). ¹H NMR (CDCl₃) ı 2.32 (m, 1H,

ester (15b, n = 1, m = 1). 13.33% yield; R_f= 0.2; ¹H NMR (CDCl₃) ı

ꢀ

H₂), 1.58 (m, 2H, H_3aand H_3a), 1.43 (m, 2H, H_3band H_3bꢀ ), 1.26 (m,

ꢀꢀ

7.23 (s, 4H, H₄and H₅), 4.00 (d, J = 6 Hz, 2H, H₉), 3.66 (s, 3H, H₁),

16H, H₄–H₇and H₄–H₇), 0.86 (“t”, 6H, H₈and H₈); ¹³C NMR (CDCl₃)

ꢀ

3.59 (s, 2H, H₂or H₇), 3.58 (s, 2H, H₇or H₂), 1.49 (sept, J = 6 Hz,

ꢀ

ı 182.5 (C₁), 45.5 (C₂), 31.8 (C₆and C₆), 29.5 and 29.1 (C₃–C₅and

ꢀ

1H, H₁₀), 1.31 (quint, J = 6.7 Hz, 4H, H₁₁and H₁₁), 0.85 (t, J = 7.3 Hz,

ꢀ ꢀ ꢀ

ꢀ ꢀ

₃–C₅), 27.2 (C₆and C₆), 22.7 (C₇and C₇), 14.2 (C₈and C₈); HRMS

C

6H, H₁₂and H₁₂); ¹³C NMR (CDCl₃) ı 171.9 (C₁), 171.6 (C₈), 133.1

ꢀ

(EI) m/z calcd for C₁₄H₂₈O₂(M⁺) 228.2089 found 228.2088; IR (KBr)

(C₃or C₆), 132.7 (C₆or C₃), 129.5 (C₄or C₅), 129.4 (C₅or C₄),

1706.93 cm⁻¹(CO stretch).

ꢀꢀ

66.9 (C₉), 52.0 (C₁), 41.1 (C₂or C₇), 40.8 (H₇or C₂), 40.3 (C₁₀),

+

ꢀ

23.3 (C₁₁and C₁₁), 11.0 (C₁₂and C₁₂); MS (CI, CH₄) m/z 292 (M ,

7.62%), 233 (M−COOCH₃, 26.81%), 163 (M−COOCH₂CH(CH₂CH₃)₂,

68.56%), 104 (M−COOCH₃–COOCH₂CH(CH₂CH₃)₂, 35.05%), 208

(MH−CH₂CH(CH₂CH₃)₂, 100%); HRMS (CI, CH₄): calcd (C₁₇H₂₄O₄,

M⁺) 292.1675, obsd 292.1660.

2.3.4.6. 2-Decyltetradecanoic acid (11h). ¹H NMR (CDCl₃) ı 2.35 (m,

ꢀ

1H, H₂), 1.59 (m, 2H, H_3aand H_3a), 1.45 (m, 2H, H_3band H_3bꢀ ), 1.25

ꢀ

(m, 36H, H₄–H₁₁, H₄–H₁₁, H₁₂and H₁₃), 0.86 (“t”, 6H, H₁₄. and H₁₂);

¹³C NMR (CDCl₃) ı 181.7 (C₁), 45.6 (C₂), ı 32.1 (C₁₂and C₁₀), 29.9,

ꢀ

29.8, 29.7, 29.6, 29.5, 29.1 and 27.3 (C₃–C₉, C₃–C₉, C₁₀and C₁₁),

ꢀ

27.2 (C₆and C₆), 22.9 (C₁₃and C₁₁), 14.3 (C₁₄and C₁₂); HRMS (EI)

m/z calcd for C₂₄H₄₈O₂(M⁺) 368.3654, found 368.3670; IR (KBr)

1707.03 cm⁻¹(CO stretch).

2.3.5.3. 4-(2-Ethylhexyloxycarbonylmethyl)phenylacetic acid methyl

ester (15c, n = 3, m = 1). 13.29% yield. R_f= 0.2; ¹H NMR (CDCl₃) ı

7.24 (s, 4H, H₄and H₅), 4.01 (d, J = 6 Hz, 2H, H₉), 3.68 (s, 3H,

ꢀꢀ

H₁), 3.6 (s, 2H, H₂or H₇), 3.59 (s, 2H, H₇or H₂), 1.56 (m,

2.3.5. 1,4-Phenylenediacetate diesters (15) and

1,4-phenylenedipropionate diesters (17)

1H, H₁₀), 1.31 (m, 8H, H₁₁, H₁₁, H₁₂and H₁₃), 0.88 (t, J = 7.5 Hz,

3H, H₁₄or H₁₂), 0.86 (t, J = 7.5 Hz, 3H, H₁₂or H₁₄); ¹³C NMR

(CDCl₃) ı 171.8 (C₁), 171.6 (C₈), 133.1 (C₃or C₆), 133.7 (C₆or

ꢀ

In a dry three-necked 100 mL ﬂask, equipped with a pressure-

equalizing funnel, an N₂inlet adapter, glass stopper and magnetic

stirring, was placed 1,4-phenylenediacetic acid (1.5 g, 7.7 mmol) or

1,4-phenylenedipropionic acid (1.7 g, 7.7 mmol) in dried distilled

CH₂Cl₂(25 mL). The mixture was stirred until the diacid dissolved.

Thionyl chloride (5.6 mL, 77 mmol) was then slowly added over the

course of 10 min to the reaction solution from a pressure-equalizing

funnel. During the addition, the glass stopper was brieﬂy removed

and pyridine (155 ␮L, 1.9 mmol) was added to the solution against

the increased N₂stream. The resulting clear solution was mag-

netically stirred at R.T. overnight. The solvent and excess thionyl

chloride were then evaporated in the hood with a high stream of N₂

through the inlet adapter leaving a solid residue of diacid chloride.

The three-necked ﬂask containing the diacid chloride (7.7 mmol)

was again equipped with an N₂inlet adapter and glass stoppers

and charged with dry distilled CH₂Cl₂(25 mL), the branched long

chain alcohol (R^ꢀOH in Scheme 6; 4.7 mmol) and pyridine (155 ␮L,

1.9 mmol). The clear solution was allowed to continue stirring for

a week at R.T. The short chain alcohol (methanol, ethanol or i-

propanol; ROH in Scheme 6; 10 mL) was then added to the solution

which was stirred for 30 min at R.T. The solvent was rotary evapo-

rated to dryness. The residue was taken up in CHCl₃and the organic

phase was extracted twice with brine, dried with MgSO₄, and rotary

evaporated to dryness. The crude material was puriﬁed by silica col-

umn chromatography (eluting with 15% ethyl acetate in hexane),

which yielded the desired product after solvent removal. The R_f

values given below are for TLC runs, eluting with the same solvent

mixture, unless otherwise indicated. The NMR carbon numbering

for diesters 15 and 17 is shown above (Scheme 1).

ꢀꢀ

C₃), 129.5 (C₄or C₅), 129.4 (C₅or C₄), 67.2 (C₉), 51.9 (C₁),

41.1 (C₂or C₇), 40.8 (C₇or C₂), 38.7 (C₁₀), 30.4, 28.8, 23.8,

ꢀ

22.9, 14.0, 11.0 (C₁₁–C₁₄and C₁₁–C₁₂); MS (CI, CH₄) m/z 320.197

(M⁺, 7.2%), 163 (M−COOCH₂CH(CH₂CH₃)(CH₂)₃CH₃, 46.55%),

104 (M−COOCH₃–COOCH₂CH(CH₂CH₃)(CH₂)₃CH₃, 42.22%), 207.94

(MH⁺−CH₂CH(CH₂CH₃)(CH₂)₃CH₃, 100%); HRMS (CI, CH₄): calcd

(C₁₉H₂₈O₄, M⁺) 320.1987, obsd 320.1967.

2.3.5.4. 4-(2-Butyloctoxycarbonylmethyl)phenylacetic acid methyl

ester (15d, n = 5, m = 3). 27.5% yield; R_f= 0.6; ¹H NMR (CDCl₃) ı

7.22 (s, 4H, H₄and H₅), 3.98 (d, J = 6 Hz, 2H, H₉), 3.65 (s, 3H,

ꢀꢀ

H₁), 3.583 (s, 2H, H₂or H₇), 3.579 (s, 2H, H₇or H₂), 1.6 (bs, 1H,

ꢀ

H₁₀), 1.24 bs, 16H, H₁₁–H₁₅and H₁₁–H₁₃), 0.88 (m, 3H, H₁₄or

H

₁₆), 0.86 (t, J = 6 Hz, 3H, H₁₆or H₁₆); ¹³C NMR (CDCl₃) ı 172.0

(C₁), 171.7 (C₈), 132.9 (C₃or C₆), 132.8 (C₆or C₃), 129.6 (C₄or

C₅), 129.5 (C₅or C₄), 67.7 (C₉), 52.1 (C₁), 41.2 (C₂or C₇), 40.9

(C₇or C₂), 37.3 (C₁₀), 31.9, 31.3, 31.0, 29.7, 28.9, 26.7, 23.0, 22.7,

14.2, 14.1 (C₁₁–C₁₆and C₁₁–C₁₄); MS (CI, CH₄), m/z 376.26 (M ,

0.53%), 163 (M−COOCH₂CH((CH₂)₃CH₃)(CH₂)₅CH₃, 66.65%), 104

(M−COOCH₃–COOCH₂CH((CH₂)₃CH₃)(CH₂)₅CH₃, 50.86%), 207.97

(M−CH₂CH((CH₂)₃CH₃)(CH₂)₅CH₃, 100%); HRMS (CI, CH₄): calcd

(C₂₃H₃₆O₄, M⁺) 376.2614, obsd 376.2611.

ꢀꢀ

+

ꢀ

2.3.5.5. 4-(2-Hexyldecyloxycarbonylmethyl)phenylacetic acid methyl

ester (15e, n = 7, m = 5). 14.26% yield; R_f= 0.37; ¹H NMR (CDCl₃)

ı 7.22 (s, 4H, H₄and H₅), 3.98 (d, J = 6 Hz, 2H, H₉), 3.65 (s,

3H, H₁), 3.581 (s, 2H, H₂or H₇), 3.576 (s, 2H, H₇or H₂), 1.6

(bs, 1H, H₁₀), 1.24 (bs, 24H, H₁₁–H₁₇and H₁₁–H₁₅), 0.88 (“t”,

ꢀꢀ

ꢀ

Y. Cohen et al. / Chemistry and Physics of Lipids 155 (2008) 98–113

103

6H, H₁₆and H₁₈); ¹³C NMR (CDCl₃) ı 171.8 (C₁), 171.5 (C₈),

(C₁), 171.0 (C₈), 133.1 (C₃or C₆), 132.9 (C₆or C₃), 129.4 (C₄or

ꢀ

ꢀꢀ

C₅), 129.4 (C₅or C₄), 68.2 (C₁), 67.2 (C₉), 41.3 (C₂or C₇), 41.2 (C₇

133.1 (C₃or C₆), 132.7 (C₆or C₃), 129.5 (C₄or C₅), 129.4 (C₅

or C₂), 38.7 (C₁₀), 30.4, 28.9, 23.8, 23.0, 14.1, 11.0 (C₁₁–C₁₄and

ꢀꢀ

or C₄), 67.6 (C₉), 51.9 (C₁), 41.1 (C₂or C₇), 40.7 (C₇or C₂),

+

ꢀ

ꢀꢀ

C

₁₁–C₁₂), 21.8 (2 C₂); MS (CI, CH₄), m/z 349.2401 (MH , 23.28%),

37.3 (C₁₀), 31.9, 31.8, 31.2, 29.9, 29.6, 29.5, 29.3, 26.75, 26.6, 22.7,

348.232 (M⁺, 17.41%), 307.185 (MH₂−CH(CH₃)₂, 36.35%), 237.108

+

ꢀ

22., 14.1, 14.1 (C₁₁–C₁₈and C₁₁–C₁₆); MS (CI, CH₄) m/z 433.33

(MH⁺, 13.57%), 163 (M−COOCH₂CH((CH₂)₅CH₃)(CH₂)₇CH₃, 12.4%),

104 (M−COOCH₃, COOCH₂CH((CH₂)₅CH₃)(CH₂)₇CH₃, 12.46%),

209 (MH−CH₂CH((CH₂)₅CH₃)(CH₂)₇CH₃, 100%), 865.57 (MMH⁺,

70.58%); HRMS (CI, CH₄): calcd (C₂₇H₄₅O₄, MH⁺) 433.3318, obsd

433.3311.

(MH⁺−CH₂CH(CH₂CH₃)(CH₂)₃CH₃, 80.59%), 195.065 (MH₃−CH₂

+

CH(CH₂CH₃)(CH₂)₃CH₃–CH(CH₃)₂, 100%), 150.069 (MH₂−CH₂CH

(CH₂CH₃)(CH₂)₃CH₃–COOCH(CH₃)₂, 49.11%), 105.067 (MH⁺−

COOCH₂CH(CH₂CH₃)(CH₂)₃CH₃–COOCH(CH₃)₂,42.44%);

(CI, CH₄): calcd (C₂₁H₃₃O₄, MH⁺) 349.2379, obsd 349.2401.

HRMS

2.3.5.10. 3-{4-[2-(2-Methylbutoxycarbonyl)ethyl]phenyl}propionic

acid methyl ester (17a, n = 1, m = 0). 10.65% yield; R_f= 0.28; ¹H

NMR (acetone-d₆) ı 7.15 (s, 4H, H₅and H₆), 3.89, 3.97, 1.66 (ABX

system 1H each, J_AB= 10.7, J_BX= 7, J_AX= 6 Hz, 2H₁₁, H₁₂), 3.66

2.3.5.6. 4-(2-Octyldodecyloxycarbonylmethyl)phenylacetic

acid

methyl ester (15f, n = 9, m = 7). 23.32% yield; R_f= 0.4; ¹H NMR

(CDCl₃) ı 7.23 (s, 4H, H₄and H₅), 3.99 (d, J = 5.7 Hz, 2H, H₉),

ꢀꢀ

3.66 (s, 3H, H₁), 3.592 (s, 2H, H₂or H₇), 3.586 (s, 2H, H₇or

ꢀꢀ

ꢀ

(s, 3H, H₁), 2.92 (t, J = 7.8 Hz, 2H, H₃or C₈), 2.91 (t, J = 7.8 Hz,

H₂), 1.61 (bs, 1H, H₁₀), 1.26 (m, 32H, H₁₁–H₁₉and H₁₁–H₁₇),

0.89 (m, 6H, H₁₈and H₂₀); ¹³C NMR (CDCl₃) ı 171.7 (C₁), 171.5

ꢀ

2H, H₈or C₃), 2.61 (t, J = 7.8 Hz, 2H, H₂or C₉), 2.6 (t, J = 7.8 Hz,

2H, H₉or C₂), 1.41 and 1.18 (AB system, 1H each, 2H₁₃), 0.88

(C₈), 133.1 (C₃or C₆), 132.7 (C₆or C₃), 129.4 (C₄or C₅), 129.4

ꢀꢀ

(t, J = 7.5 Hz, 3H, H₁₄), 0.87 (d, J = 6.5 Hz, 3H, H₁₃); ¹³C NMR

ꢀ

(C₅or C₄), 67.6 (C₉), 51.9 (C₁), 41.1 (C₂or C₇), 40.7 (C₇or C₂),

(acetone-d₆) ı 173.3 (C₁), 173.0 (C₁₀), 139.5 (C₄and C₇), 129.5

37.3 (C₁₀), 31.9, 31.2, 29.9, 29.7, 29.59, 29.55, 29.4, 29.3, 26.7,

ꢀꢀ

ꢀ

(C₅and C₆), 69.2 (C₁₁), 51.6 (C₁), 36.3 (C₂or C₉), 36.1 (C₉or

22.7, 14.1 (C₁₁–C₂₀and C₁₁–C₁₈); MS (CI, CH₄), m/z 489.3971

(MH⁺, 0.75%), 163.01 (M−COOCH₂CH((CH₂)₇CH₃)(CH₂)₉CH₃,

37.38%), 104.02 (M−COOCH₃–COOCH₂CH((CH₂)₇CH₃)(CH₂)₉CH₃,

29.09%), 208.98 (MH⁺−CH₂CH((CH₂)₇CH₃)(CH₂)₉CH₃, 100%);

HRMS (CI, CH₄): calcd (C₃₁H₅₃O₄, MH⁺) 489.3944, obsd

489.3971.

C₂), 35.0 (C₁₂), 31.2 (C₃or C₈), 31.1 (C₈or C₃), 26.6, 16.6, 11.5

(C₁₃and C₁₃–C₁₄); MS (CI, CH₄), m/z 306.1853 (M⁺, 91.43%),

ꢀ

307.1898 (MH⁺, 76.6%), 236.107 (MH−CH₂CH(CH₃)CH₂CH₃,

35.03%),

219.112

(M−CH₃–CH₂CH(CH₃)CH₂CH₃,

49.15%),

176.092 (MH⁺−CH₃–COOCH₂CH(CH₃)CH₂CH₃, 100%), 162.072

(M−CH₃–CH₂COOCH₂CH(CH₃)CH₂CH₃, 21.06%), 117.069 (M−

COOCH₃–CH₂COOCH₂CH(CH₃)CH₂CH₃, 39.13%); HRMS (CI, CH₄):

calcd (C₁₈H₂₇O₄, MH⁺) 307.1909, obsd 307.1898.

2.3.5.7. 4-(2-Decyltetradecyloxycarbonylmethyl)phenylacetic

acid

methyl ester (15g, n = 11, m = 9). 15.55% yield; R_f= 0.34; ¹H NMR

(CDCl₃) ı 7.23 (s, 4H, H₄and H₅), 3.98 (d, J = 6 Hz, 2H, H₉), 3.68 (s, 3H,

ꢀꢀ

2.3.5.11. 3-{4-[2-(2-Ethylbutoxycarbonyl)ethyl]phenyl}propionic

acid methyl ester (17b, n = 1, m = 1). 15.24% yield; R_f= 0.4; ¹H NMR

(CDCl₃) ı 7.12 (s, 4H, H₅and H₆), 3.99 (d, J = 6 Hz, 2H, H₁₁), 3.66 (s,

H₁), 3.604 (s, 2H, H₂or H₇), 3.594 (s, 2H, H₇or H₂), 1.6 (m, 1H, H₁₀),

ꢀ

1.26 (m, 32H, H₁₁–H₂₁and H₁₁–H₁₉), 0.89 (“t”, 6H, H₂₀and H₂₂);

¹³C NMR (CDCl₃) ı 172.0 (C₁), 171.8 (C₈), 133.2 (C₃or C₆), 132.8 (C₆

ꢀꢀ

3H, H₁), 2.92 (t, J = 8 Hz, 2H, H₃or C₈), 2.91 (t, J = 8 Hz, 2H, H₈or

or C₃), 129.6 (C₄or C₅), 129.5 (C₅or C₄), 67.8 (C₉), 52.1 (C₁), 41.3

C₃), 2.61 (t, J = 8 Hz, 2H, H₂or C₉), 2.6 (t, J = 8 Hz, 2H, H₉or C₂), 1.48

(C₂or C₇), 40.9 (C₇or C₂), 37.4 (C₁₀), 32.1, 31.3, 30.1, 29.8, 29.7, 29.5,

ꢀ

(quint, t, J = 7.5 and 6 Hz, 1H, H₁₂), 1.31 (quint, J = 7.5 Hz, 4H, H₁₃

26.8, 22.8, 14.2 (C₁₁–C₂₂and C₁₁–C₂₀); MS (CI, CH₄), m/z 544.4505

(M⁺, 2.48%), 163.07 (M−COOCH₂CH((CH₂)₉CH₃)(CH₂)₁₁CH₃,

25.31%), 104.03 (M−COOCH₃–COOCH₂CH((CH₂)₉CH₃)(CH₂)₁₁CH₃,

12.64%), 209.09 (MH⁺−CH₂CH((CH₂)₉CH₃)(CH₂)₁₁CH₃, 100%),

208.09 (M−CH₂CH(CH₂)₉CH₃)(CH₂)₁₁CH₃), 81.24%); HRMS (CI,

CH₄): calcd (C₃₅H₆₀O₄, M⁺) 544.4491, obsd 544.4504.

and H₁₃), 0.86 (t, J = 7.5 Hz, 6H, H₁₄and H₁₄); ¹³C NMR (CDCl₃) ı

ꢀ

173.5 (C₁), 173.2 (C₁₀), 138.6 (C₄or C₇), 138.5 (C₇or C₄), 128.5 (C₅

or C₆), 128.5 (C₆or C₅), 66.6 (C₁₁), 51.7 (C₁), 40.3 (C₁₂), 36.0 (C₂

or C₉), 35.8 (C₉or C₂), 30.7 (C₃or C₈), 30.6 (C₈or C₃), 23.3, 11.1

(C₁₃, C₁₃, C₁₄and C₁₄); MS (CI, CH₄), m/z 320.1994 (M , 33.49%),

321.207 (MH⁺, 11.47%), 236.104 (MH⁺−CH₂CH(CH₂CH₃)₂, 43.23%),

219.102 (M−CH₃–CH₂CH(CH₂CH₃)₂, 24.53%), 176.085 (M−CH₃–

COOCH₂CH(CH₂CH₃)₂, 100%), 162.068 (M−CH₃–CH₂COOCH₂

CH(CH₂CH₃)₂, 19.5%), 117.065(MH⁺−COOCH₃–CH₂COOCH₂CH(CH₂

CH₃)₂, 34.86%); HRMS (CI, CH₄): calcd (C₁₉H₂₉O₄, MH⁺) 320.2066,

obsd 321.2068.

ꢀꢀ

+

ꢀ

2.3.5.8. 4-(2-Ethylhexaloxycarbonylmethyl)phenylacetic acid ethyl

ester (15h, n = 3, m = 1). 17.46% yield; R_f= 0.58; ¹H NMR (CDCl₃)

ꢀꢀ

ı 7.24 (s, 4H, H₄and H₅), 4.14 (q, J = 7.13 Hz, 2H, H₁), 3.99

(d, J = 6 Hz, 2H, H₉), 4.00 (s, 2H, H₂or H₇), 3.98 (s, 2H, H₇or

H₂), 1.62 (sept, J = 6 Hz, 1H, H₁₀), 1.28 (m, 11H, H₁₁–H₁₃and

ꢀ

ꢀꢀ

ꢀ

H₁₁, H₂), 0.87 (t, J = 7.5 Hz, 3H, H₁₂or H₁₄), 0.85 (t, J = 7.5 Hz,

3H, H₁₄or H₁₂); ¹³C NMR (CDCl₃) ı 171.9 (C₁), 171.7 (C₈),

2.3.5.12. 3-{4-[2-(2-Ethylhexyloxycarbonyl)ethyl]phenyl}propionic

acid methyl ester (17c, n = 3, m = 1). 5.77% yield; R_f= 0.57 (30% ethyl

acetate in hexane); ¹H NMR (CDCl₃) ı 7.11 (4H, s, H₅and H₆),

3.99, 3.97, 1.54 (ABX system 1H each, J_AB= 12, J_BX= 6, J_AX= 5.4 Hz,

2H₁₁and H₁₂), 3.66 (s, 3H, H₁ꢀꢀ ), 2.92 (t, 2H, J = 8.2 Hz, H₃or C₈),

2.91 (t, 2H, J = 8.2 Hz, H₈or C₃), 2.61 (t, J = 8.2 Hz, 2H, H₂or C₉),

ꢀ

133.1 (C₃or C₆), 133.0 (C₆or C₃), 129.6 (C₄or C₅), 129.5 (C₅

ꢀꢀ

C₄), 67.4 (C₉), 61.0 (C₁), 41.3 (C₂

C₇), 41.2 (C₇

C₂), 38.8

or

ꢀ

(C₁₀), 30.5, 28.3, 23.9, 23.1, 14.3, 14.2, 11.1 (C₁₁-C₁₄, C₁₁–C₁₂

and C₂); MS (CI, CH₄), m/z 334.2161 (M⁺, 11.65%), 216.193

(M−CO₂Et), 10.21%, 222.088 (MH−CH₂CH(CH₂CH₃)(CH₂)₃CH₃),

100%, 177.093 (MH−CH₂CH(CH₂CH₃)(CH₂)₃CH₃)–OEt, 51.76%,

ꢀꢀ

ꢀ

2.6 (t, J = 8.2 Hz, 2H, H₉or C₂), 1.31 (m, 8H, H₁₃and H₁₃–H₁₅),

ꢀ

149.069

(MH−CH₂CH(CH₂CH₃)(CH₂)₃CH₃)–COOEt,

50.29%),

0.89 (t, J = 6 Hz, 3H, H₁₄or H₁₆), 0.86 (t, J = 7.5 Hz, 3H, H₁₆or

₁₄); ¹³C NMR (CDCl₃) ı 173.4 (C₁), 173.2 (C₁₀), 138.6 (C₄or C₇),

ꢀ

104.058 (M−COOCH₂CH(CH₂CH₃)(CH₂)₃CH₃)–COOEt, 48.23%);

H

HRMS (CI, CH₄): calcd (C₂₀H₃₀O₄, M⁺) 334.2144, obsd 334.2161.

138.5 (C₇or C₄), 128.5 (C₅or C₆), 128.5 (C₆or C₅), 67.0 (C₁₁),

ꢀꢀ

51.7 (C₁), 38.8 (C₁₂), 36.0 (C₂or C₉), 35.8 (C₉or C₂), 30.7 (C₃

2.3.5.9. 4-(2-Ethylhexaloxycarbonylmethyl)phenylacetic acid iso-

or C₈), 30.6 (C₈or C₃), 30.5, 29.0, 23.8, 23.1, 14.1, 11.1 (C₁₃–C₁₆

+

propyl ester (15i, n = 3, m = 1). 3.42% yield; R_f= 0.38; ¹H NMR

and C₁₃–C₁₄); MS (CI, CH₄), m/z 348.231 (M , 15.95%), 349.242

(MH⁺, 5.73%), 236.091 (MH−CH₂CH(CH₂)₃CH₃(CH₂CH₃), 58.63%),

219.127 (MH−CH₃–CH₂CH(CH₂)₃CH₃(CH₂CH₃), 13.23%), 176.076

(M−CH₃–COOCH₂CH(CH₂)₃CH₃(CH₂CH₃), 100%), 162.078 (M−

CH₃,CH₂COOCH₂CH(CH₂CH₃)₂, 18.47%), 117.07 (MH⁺−COOCH₃–

ꢀ

ꢀꢀ

(CDCl₃) ı 7.22 (s, 4H, H₄and H₅), 4.99 (sept, J = 6.2 Hz, 1H, H₁),

3.99 (d, J = 6 Hz, 2H, H₉), 3.58 (s, 2H, H₂or H₇), 3.54 (s, 2H, H₇

or H₂), 1.54 (sept, J = 6 Hz, 1H, H₁₀), 1.25 (m, 14H, H₁₁–H₁₃and

H₁₁, H₂), 0.84 (m, 6H, H₁₄and H₁₂); ¹³C NMR (CDCl₃) ı 171.7

ꢀ

ꢀꢀ

ꢀ

104

Y. Cohen et al. / Chemistry and Physics of Lipids 155 (2008) 98–113

CH₂COOCH₂CH(CH₂)₃CH₃(CH₂CH₃), 25.66%); HRMS (CI, CH₄):

calcd (C₂₁H₃₂O₄, M⁺) 348.23, obsd 348.231.

31.4, 30.7, 30.7, 30.1, 29.80, 29.75, 29.5, 26.8, 22.8, 14.3 (C₁₃–C₂₄

+

ꢀ

and C₁₃–C₂₂); MS (CI, CH₄), m/z 572.476 (M , 4.94%), 573.488

(MH⁺, 3.54%), 236.1 (MH−CH₂CH(CH₂)₁₁CH₃(CH₂)₉CH₃), 100%),

+

2.3.5.13. 3-{4-[2-(2-Butyloctoxycarbonyl)ethyl]phenyl}propionic

acid methyl ester (17d, n = 5, m = 3). 10.35% yield; R_f= 0.5 (30% ethyl

acetate in hexane); ¹H NMR (CDCl₃) ı 7.12 (s, 4H, H₅and H₆), 3.97

237.12(MH₂−CH₂CH(CH₂)₁₁CH₃(CH₂)₉CH₃), 36.96%), 176.09

(M−COOCH₃–CH₂CH(CH₂)₁₁CH₃(CH₂)₉CH₃, 53.9%); HRMS (CI,

CH₄): calcd (C₃₇H₆₅O₄, M⁺) 573.4883, obsd 573.4885.

ꢀꢀ

(d, J = 6 Hz, 2H, H₁₁), 3.66 (s, 3H, H₁), 2.91 (t, J = 8 Hz, 4H, H₃and

H₈), 2.61 (t, J = 7.1 Hz, 4H, H₂and H₉), 1.6 (m, 1H, H₁₂), 1.26 (m,

2.3.6. Branched 4-alkoxybenzaldehydes (19h–k) and branched

methyl 4-alkoxybenzoates (21g–k)

16H, H₁₃–H₁₅and H₁₃–H₁₇), 0.88 (“t”, 6H, H₁₆and H₁₈); ¹³C NMR

ꢀ

(CDCl₃) ı 173.4 (C₁), 173.1 (C₁₀), 138.6 (C₄or C₇), 138.4 (C₇or C₄),

The preparation of the title compounds was based on the

literature procedure of Hong-bin et al. (2001) with major modiﬁ-

cation of the workup. A 100 mL round bottom ﬂask was charged

with 4-hydroxybenzaldehyde (18) or 4-hydroxymethylbenzoate

(21) (24.6 mmol, 3 equiv.), K₂CO₃(2 equiv.), and the required alkyl

bromide (23 in Schemes 8 and 9; synthesized as described below),

dissolved in 25 mL of acetone (AR). The reaction ﬂask was then ﬁt-

ted with a reﬂux condenser and the clear solution was reﬂuxed for

24 h. The solution was evaporated to dryness, and the crude prod-

ucts were suspended in n-hexane. The unreacted starting materials

(aldehyde 18 or ester 21) were gravitationally ﬁltered from the solu-

tion. The ﬁltered solid was again suspended in n-hexane and the

solution ﬁltered. The combined ﬁltrates were rotary evaporated and

the crude product residue was silica column chromatographed. The

column was eluted ﬁrst with 670 mL of n-hexane per gram of crude

product to remove excess alkyl bromide, followed by 50% ethyl

acetate in hexane which yields the desired 4-alkoxybenzaldehyde

(19) or 4-alkoxymethylbenzoate (22). The R_fvalues given below are

for TLC runs eluting with 50% ethyl acetate in hexane, unless other-

wise indicated. The products were identiﬁed by their spectral data.

ꢀꢀ

128.4 (C₅and C₆), 67.3 (C₁₁), 51.6 (C₁), 37.3 (C₁₂), 36.0 (C₂or C₉),

35.8 (C₉or C₂), 31.9 (C₃or C₈), 31.3 (C₈or C₃), 31.0, 30.6, 30.6, 29.7,

ꢀ

29.0, 26.7, 23.0, 22.7, 14.2, 14.1 (C₁₃–C₁₈and C₁₃–C₁₆); m/z (CI, CH₄),

404.294 (M⁺, 7.72%), 236.099 (MH−CH₂CH(CH₂)₅CH₃(CH₂)₃CH₃),

78.13%), 190.115 (MH₂−OCH₃–OCH₂CH(CH₂)₅CH₃(CH₂)₃CH₃,

27.06), 176.088 (M−CH₃–COOCH₂CH(CH₂)₅CH₃(CH₂)₃CH₃, 100%),

162.022 (M−CH₃–CH₂COOCH₂CH(CH₂)₅CH₃(CH₂)₃CH₃, 17.53%),

117.05(MH−COOCH₃–CH₂COO

CH₂CH(CH₂)₅CH₃(CH₂)₃CH₃,

20.96%); HRMS (CI, CH₄): calcd (C₂₅H₄₀O₄, M⁺) 404.2927, obsd

404.2936.

2.3.5.14. 3-{4-[2-(2-Hexyldecyloxycarbonyl)ethyl]phenyl}propionic

acid methyl ester (17e, n = 7, m = 5). 2.37% yield; R_f= 0.26; ¹H NMR

(CDCl₃) ı 7.12 (s, 4H, H₅and H₆), 3.98 (d, J = 6 Hz, 2H, H₁₁), 3.66

(s, 3H, H₁), 2.92 (t, J = 8.2 Hz, 2H, H₃or C₈), 2.91 (t, J = 8.2 Hz, 2H,

H₈or C₃), 2.61 (t, J = 7.2 Hz, 2H, H₉or C₂), 2.6 (t, J = 7.2 Hz, 2H, H₂

ꢀꢀ

ꢀ

or C₉), 1.6 (bs, 1H, H₁₂), 1.26 (m, 24H, H₁₃–H₁₇and H₁₃–H₁₉),

0.883 (t, J = 7 Hz, 6H, H₁₈), 0.88 (t, J = 7 Hz, 6H, H₂₀); ¹³C NMR

ꢀ

(CDCl₃) ı 173.5 (C₁), 173.2 (C₁₀), 138.6 (C₄or C₇), 138.5 (C₇or

C₄), 128.5 (C₅or C₆), 128.5 (C₆or C₅), 67.4 (C₁₁), 51.7 (C₁), 37.4

ꢀꢀ

(C₁₂), 36.0 (C₂or C₉), 35.8 (C₉or C₂), 32.0 (C₃or C₈), 31.9 (C₈or

C₃), 31.3, 31.3, 30.7, 30.6, 30.1, 29.7, 29.7, 29.4, 26.79, 26.76, 22.8,

2.3.6.1. 4-(2-Butyloctoxy)benzaldehyde (19h). 9.5% yield; R_f= 0.65;

¹H NMR (CDCl₃) ı 9.87 (s, 1H, H₁), 7.83, 7.00 (AA^ꢀXX^ꢀsystem, 2H

+

ꢀ

22.8, 14.2 (C₁₃–C₂₀and C₁₃–C₁₈); MS (CI, CH₄), m/z 460.357 (M ,

+

ꢀ

each, H₃, H₃, H₄and H₄), 3.92 (d, J = 6 Hz, 2H, H₆), 1.81 (m, 1H,

5.44%), 237.100 (MH₂−CH₂CH(CH₂)₇CH₃(CH₂)₅CH₃), 23.08%),

ꢀ

H₇), 1.3 (m, 16H, H₈–H₁₀and H₈–H₁₂), 0.91 (t, J = 6.6 Hz, 3H, H₁₁or

236.099 (MH−CH₂CH(CH₂)₇CH₃(CH₂)₅CH₃), 100%), 176.087

(MH−CH₃COOH–CH₂CH(CH₂)₇CH₃(CH₂)₅CH₃, 97.46%), 162.022

(M−CH₂COOCH₃–CH₂CH(CH₂)₇CH₃(CH₂)₅CH₃, 16.1%), 117.05

(MH−COOCH₃–CH₂COO CH₂CH(CH₂)₇CH₃(CH₂)₅CH₃, 13.13%);

HRMS (CI, CH₄): calcd (C₂₉H₄₈O₄, M⁺) 460.355, obsd 460.357.

₁₃), 0.89 (t, J = 6.6 Hz, 3H, H₁₃or H₁₁); ¹³C NMR (CDCl₃) ı 190.8 (C₁),

ꢀ

H

ꢀ

164.6 (C₅), 132.0 (C₃and C₃), 129.8 (C₂), 114.9 (C₄and C₄), 71.4 (C₆),

37.9 (C₇), 31.9 31.4, 31.1, 29.7, 29.1, 26.9, 23.1, 22.7, 14.2, 14.1 (C₈–C₁₃

,

+

ꢀ

C₈–C₁₁); MS (CI, CH₄), m/z 290.2237 (M , 38.79%), 291.233 (MH ,

100%), 123.05 (MH−CH₂C(CH₂)₅CH₃(CH₂)₃CH₃, 99.24%); HRMS

(CI, CH₄): calcd (C₁₉H₃₁O₂, MH⁺) 291.232, obsd 291.233.

2.3.5.15. 3-{4-[2-(2-Octyldodecyloxycarbonyl)ethyl]phenyl}propionic

acid methyl ester (17f, n = 9, m = 7). 0.29% yield; R_f= 0.75 (30% ethyl

acetate in hexane); ¹H NMR (CDCl₃) ı 7.12 (s, 4H, H₅and H₆),

2.3.6.2. 4-(2-Hexyldecyloxy)benzaldehyde

(19i). 11.11%

yield;

ꢀꢀ

3.97 (d, J = 6 Hz, 2H, H₁₁), 3.67 (s, 3H, H₁), 2.91 (t, J = 7.8 Hz, 4H,

R_f= 0.85; ¹H NMR (CDCl₃) ı 9.88 (s, 1H, H₁), 7.82, 6.99 (AA^ꢀXX^ꢀ

H₃and H₈), 2.61 (t, J = 7.8 Hz, 4H, H₂and H₉), 1.6 (bs, 1H, H₁₂),

ꢀ

system, 2H each, H₃, H₃, H₄and H₄), 3.91 (d, J = 6 Hz, 2H, H₆),

ꢀ

1.26 (m, 32H, H₁₃–H₁₉and H₁₃–H₂₁), 0.88 (“t”, 6H, H₂₀and

ꢀ

1.8 (sept, J = 6 Hz, 1H, H₇), 1.3 (m, 24H, H₈–H₁₂and H₈–H₁₄),

H₂₂); ¹³C NMR (CDCl₃) ı 173.5 (C₁), 173.3 (C₁₀), 138.7 (C₄or C₇),

0.88 (m, 6H, H₁₃and H₁₅); ¹³C NMR (CDCl₃) ı 191.0 (C₁), 164.7

ꢀ

ꢀꢀ

138.5 (C₇or C₄), 128.5 (C₅and C₆), 67.5 (C₁₁), 51.7 (C₁), 37.4

ꢀ

(C₅), 132.1 (C₃and C₃), 129.9 (C₂), 114.9 (C₄and C₄), 71.5

(C₁₂), 36.1 (C₂or C₉), 35.8 (C₉or C₂), 32.05 (C₃or C₈), 32.05 (C₈

or C₃), 31.4, 30.7, 30.7, 30.1, 29.8, 29.7, 29.7, 29.49, 29.46, 26.8,

(C₆), 38.0 (C₇), 32.0, 31.4, 30.1, 29.8, 29.7, 29.5, 27.0, 22.8, 14.2

+

ꢀ

(C₈–C₁₅, C₈–C₁₃); MS (CI, CH₄), m/z 346.2923 (M , 23.77%),

347.295 (MH⁺, 59.76%), 224.24 (MH⁺−(CH₂)₇CH₃, 10.08%), 123.04

(MH−CH₂C(CH₂)₇CH₃(CH₂)₅CH₃, 100%); HRMS (CI, CH₄): calcd

(C₂₃H₃₉O₂, MH⁺) 347.295, obsd 347.295.

ꢀ

22.8, 14.3 (C₁₃–C₂₂and C₁₃–C₂₀); MS (CI, CH₄), m/z 516.4175

(M⁺, 2.59%), 236.10 (MH−CH₂CH(CH₂)₉CH₃(CH₂)₇CH₃), 98.68%),

176.08 (M−COOCH₃–CH₂CH(CH₂)₉CH₃(CH₂)₇CH₃, 91.75%), 117.07

(MH⁺−COOCH₃–CH₂COO–CH₂CH(CH₂)₉CH₃(CH₂)₇CH₃, 30.54%);

HRMS (CI, CH₄): calcd (C₃₃H₅₇O₄, MH⁺) 517.4257, obsd 517.4232.

2.3.6.3. 4-(2-Octyldodecyloxy)benzaldehyde (19j). 11.48% yield;

2.3.5.16. 3-{4-[2-(2-Decyltetradecyloxycarbonyl)ethyl]phenyl}propionic R_f= 0.74; ¹H NMR (CDCl₃) ı 9.88 (s, 1H, H₁), 7.83, 6.998 (AA^ꢀXX^ꢀ

ꢀ ꢀ

system, 2H each, H₃, H₃, H₄and H₄), 3.92 (d, J = 6 Hz, 2H, H₆), 1.82

acid methyl ester (17g, n = 11, m = 9). 1.97% yield; R_f= 0.65 (30% ethyl

acetate in hexane); ¹H NMR (CDCl₃) ı 7.12 (s, 4H, H₅and H₆),

ꢀ

(m, 1H, H₇), 1.27 (m, 32H, H₈–H₁₄and H₈–H₁₆), 0.89 (“t”, 6H, H₁₅

ꢀꢀ

3.97 (d, J = 5.7 Hz, 2H, H₁₁), 3.67 (s, 3H, H₁), 2.92 (t, J = 7.8 Hz,

and H₁₇); ¹³C NMR (CDCl₃) ı 190.8 (C₁), 164.6 (C₅), 132.0 (C₃and

ꢀ

4H, H₃and H₈), 2.61 (t, J = 7.8 Hz, 4H, H₂and H₉), 1.6 (bs, 1H,

C

₃), 129.8 (C₂), 114.9 (C₄and C₄), 71.4 (C₆), 37.9 (C₇), 32.0, 31.4,

ꢀ

H

₁₂), 1.26 (m, 40H, H₁₃–H₂₁and H₁₃–H₂₃), 0.88‘(“t”, 6H, H₂₂

30.2, 30.1, 29.8, 29.73, 29.69, 29.5, 27.0, 26.9, 22.8, 14.2 (C₈–C₁₇

and H₂₄); ¹³C NMR (CDCl₃) ı 173.5 (C₁), 173.3 (C₁₀), 138.7 (C₄

and C₈–C₁₅); MS (CI, CH₄), m/z 402.39 (M , 34.21%), 403.35 (MH ,

99.76%), 123.06 (MH−CH₂C(CH₂)₉CH₃(CH₂)₇CH₃, 76.04%); HRMS

(CI, CH₄): calcd (C₂₇H₄₇O₂, MH⁺) 403.3576, obsd 403.3558.

+

ꢀ

ꢀꢀ

or C₇), 138.5 (C₇or C₄), 128.5 (C₅and C₆), 67.5 (C₁₁), 51.7 (C₁),

37.4 (C₁₂), 36.1 (C₂or C₉), 35.8 (C₉or C₂), 32.1 (C₃and C₈),

Y. Cohen et al. / Chemistry and Physics of Lipids 155 (2008) 98–113

105

2.3.6.4. 4-(2-Decyltetradecyloxy)benzaldehyde (19k). 19.73% yield;

R_f= 0.78; ¹H NMR (CDCl₃) ı 9.88 (s, 1H, H₁), 7.83, 6.99 (AA^ꢀXX^ꢀ

153.047(MH−CH₂C(CH₂)₁₁CH₃(CH₂)₉CH₃, 100%); HRMS (CI,

CH₄): calcd (C₃₂H₅₇O₃, MH⁺) 489.4308, obsd 489.4294.

ꢀ

system, 2H each, H₃, H₃, H₄and H₄), 3.91 (d, J = 6 Hz, 2H, H₆),

ꢀ

1.8 (m, 1H, H₇), 1.26 (m, 40H, H₈–H₁₆and H₈–H₁₈), 0.88 (“t”, 6H,

2.3.7. Unbranched 4-alkoxybenzaldehydes (19a –f ) and branched

4-(2-ethylhexyloxy)benzaldehyde (19g)

H₁₇and H₁₉); ¹³C NMR (CDCl₃) ı 191.0 (C₁), 164.7 (C₅), 132.1 (C₃

ꢀ

and C₃), 129.8 (C₂), 114.9 (C₄and C₄), 71.4 (C₆), 38.0 (C₇), 32.1,

The unbranched 4-alkoxybenzaldehydes 19a–f are commer-

cially available, though we found it convenient to prepare 19b–f

(65–85% yield) and 4-(2-ethylhexyloxy)benzaldehyde 19g via the

classic Williamson procedure (Williamson, 1852), with major mod-

iﬁcations in the workup. A 100mL round ﬂask was charged with

4-hydroxybenzaldehyde (0.15 g, 1.23 mmol, 1 equiv.) and 0.25 M

NaOH solution (9.84 mL, 1 equiv.) was stirred under air for 15 min

at R.T. yielding a clear deep red solution. The water was removed by

rotary evaporation and the remaining crude solid was dissolved in

DMF (10 mL). The desired alkyl bromide (23, 1 equiv.) was added

and the solution was then stirred for 3 days at R.T. During this

period, the color of the solution gradually changed from deep red

to brown-yellow. The DMF was removed by rotary evaporation and

the oily liquid was dissolved in CHCl₃and extracted with an equal

volume of H₂O. The organic phase was dried over MgSO₄and evap-

orated to dryness, and the crude material was puriﬁed on a silica

column (eluting with 50% ethyl acetate in n-hexane), yielding the

desired product after solvent removal.

ꢀ

31.4, 30.1, 29.8, 29.5, 27.0, 22.8, 14.3 (C₈–C₁₉, C₈–C₁₇); MS (CI,

CH₄), m/z 458.4299 (M⁺, 72.66%), 459.419 (MH⁺, 100%), 123.053

(MH−CH₂C(CH₂)₁₁CH₃(CH₂)₉CH₃, 64.65%; HRMS (CI, CH₄): calcd

(C₃₁H₅₅O₂, MH⁺) 459.42, obsd 459.419.

2.3.6.5. Methyl-4-(2-ethylhexyloxy)benzoate (22g). 7.1% yield;

R_f= 0.7. The NMR spectral data of the title compound appear in the

literature (Zhi-Kuan et al., 1999; Seto et al., 1994). MS (CI, CH₄), m/z

264.1724 (M⁺, 38.65), 265.1764 (MH⁺, 11.83%), 233.098 (M−OCH₃),

152.054 (M−CH₂C(CH₂)₃CH₃CH₂CH₃,100%); HRMS (CI, CH₄):

calcd (C₁₆H₂₄O₃, M⁺) 264.1724, obsd 264.1724.

2.3.6.6. Methyl-4-(2-butyloctoxy)benzoate

(22h). 2.35%

yield;

R_f= 0.75; ¹H NMR (CDCl₃) ı 7.97 and 6.90 (AA^ꢀXX^ꢀsystem, 2H

ꢀ

each, H₃, H₃, H₄and H₄), 3.87 (d, J = 6 Hz, 2H, H₆), 3.87 (s, 3H,

H₁), 1.79 (m, 1H, H₇), 1.36 (m, 16H, H₈–H₁₀and H₈–H₁₂), 0.9

ꢀꢀ

ꢀ

(t, J = 6.6 Hz, 3H, H₁₁or H₁₃), 0.88 (t, J = 6.6 Hz, 3H, H₁₃or H₁₁);

¹³C NMR (CDCl₃) ı 167.0 (C₁), 163.3 (C₅), 131.6 (C₃and C₃),

ꢀ

2.3.7.1. 4-(2-Ethylhexyloxy)benzaldehyde (19g) (Woo et al., 1997).

The title compound was obtained in a 27% yield. R_f(50% ethyl

acetate in hexane) = 0.8; ¹H NMR (CDCl₃) ı 9.87 (1H, s, H₁), 7.82,

ꢀ

ꢀꢀ

122.4 (C₂), 114.2 (C₄and C₄), 71.2 (C₆), 51.9 (C₁), 38.0 (C₇),

32.0 31.4, 31.1, 29.8, 29.2, 26.9, 23.2, 22.8, 14.2, 14.8 (C₈–C₁₃and

+

ꢀ

C₈–C₁₁); MS (CI, CH₄), m/z 320.2362 (M , 36.78), 321.2486 (MH ,

15.99%), 152.067 (M−CH₂C(CH₂)₅CH₃(CH₂)₃CH₃, 100%), 121.023

(M−OCH₃CH₂,CH₂C(CH₂)₅CH₃(CH₂)₃CH₃, 60.79%); HRMS (CI,

CH₄): Calcd (C₂₀H₃₂O₃, M⁺) 320.2351, obsd 320.2362.

6.99 (AA^ꢀXX^ꢀsystem, 2H each, H₃, H₃, H₄and H₄), 3.92 (d, J = 6 Hz,

ꢀ

2H, H₆), 1.78 (m, 1H, H₇), 1.3 (m, 8H, H₈and H₈–H₁₀), 0.93 (t,

J = 7.4 Hz, 3H, H₉or H₁₁), 0.9 (t J = 7.4 Hz, 3H, H₁₁or H₉); ¹³C NMR

ꢀ

(CDCl₃) ı 191.0 (C₁), 164.6 (C₅), 132.1 (C₃and C₃), 129.8 (C₂), 114.9

ꢀ

(C₄and C₄), 71.0 (C₆), 37.4 (C₇), 30.6, 29.2, 23.9, 23.1, 14.2, 11.2

2.3.6.7. Methyl-4-(2-Hexyldecyloxy)benzoate (22i) (Moeller and

Wallat, 1986). 12.3% yield; R_f= 0.66; ¹H NMR (CDCl₃) ı 7.98 and

(C₈–C₁₁and C₈ꢀ –C₉ꢀ ); MS (CI, CH₄), m/z 234.1612 (M⁺, 43.02%),

235.1773 (MH⁺, 41.38%), 123.04 (MH−CH₂C(CH₂)₃CH₃CH₂CH₃,

94.13%); HRMS (CI, CH₄): calcd (C₁₅H₂₃O₂, MH⁺) 235.1698, obsd

235.1676.

6.91 (AA^ꢀXX^ꢀsystem, 2H each, H₃, H₃, H₄and H₄), 3.92 (s, 3H, H₁),

ꢀ

ꢀꢀ

ꢀ

3.87 (d, J = 5.4 Hz, 2H, H₆), 1.79 (m, 1H, H₇), 1.38 (m, 24H, H₈–H₁₂

and H₈–H₁₄), 0.88 (m, 6H, H₁₃, H₁₅); ¹³C NMR (CDCl₃) ı 167.1 (C₁),

ꢀ

163.4 (C₅), 131.7 (C₃and C₃), 122.4 (C₂), 114.3 (C₄and C₄), 71.3 (C₆),

2.3.8. Unbranched methyl 4-alkoxybenzoates (22a–f)

The unbranched esters, methyl 4-alkoxybenzoates 22a–f, are

commercially available; nevertheless, we found it convenient to

prepare 22b–f according to literature procedure (Fieser, 1957) in

20–45% yields.

ꢀꢀ

52.0 (C₁), 38.0 (C₇), 32.0, 32.0, 31.5, 30.1, 29.8, 29.7, 29.55, 27.0, 22.8,

+

ꢀ

14.2 (C₈–C₁₅and C₈–C₁₃); MS (CI, CH₄), m/z 376.304 (M , 44.22%),

377.304 (MH⁺, 100%), 153.025 (MH−CH₂C(CH₂)₇CH₃(CH₂)₅CH₃,

43.27%), 121.03 (MH−OCH₃,CH₂C(CH₂)₇CH₃(CH₂)₅CH₃, 24.66%);

HRMS (CI, CH₄): calcd (C₂₄H₄₁O₃, MH⁺) 377.3056, obsd

377.304.

2.3.9. Preparation of the branched alkyl bromides (23h–k)

The branched alkyl bromides 23h–k, required for the synthe-

sis of the branched derivatives of 4-alkoxybenzaldehydes 19 and

methyl 4-alkoxybenzoates 22 , were prepar ed by bromination of the

corresponding alcohols following the pr ocedure of Balachander and

Sukenik (1990). The unbranched bromides and branched bromides

23g and 23i (City Chemical) are commercially available, but the

latter was conveniently synthesized by this procedure as well. Bro-

mides 23h (Gaertner, 1965), j (Ellinger et al., 2007) and k (Pisula et

al., 2004) have been previously reported, though the spectral data

are incomplete for the ﬁrst and last and, therefore, reported below.

2.3.6.8. Methyl-4-(2-octyldodecyloxy)benzoate (22j). 9.64% yield;

R_f= 0.62; ¹H NMR (CDCl₃) ı 7.97, 6.90 (AA^ꢀXX^ꢀsystem, 2H each, H₃,

ꢀ

ꢀꢀ

H₃, H₄and H₄), 3.88 (s, 3H, H₁), 3.87 (d, J = 4.6 Hz, 2H, H₆), 1.78 (m,

ꢀ

1H, H₇), 1.26 (m, 32H, H₈–H₁₄and H₈-H₁₆), 0.88 (“t”, 6H, H₁₅and

₁₇); ¹³C NMR (CDCl₃) ı 167.1 (C₁), 163.4 (C₅), 131.7 (C₃and C₃),

ꢀ

H

ꢀ

ꢀꢀ

122.4 (C₂), 114.2 (C₄and C₄), 71.2 (C₆), 52.0 (C₁), 38.0 (C₇), 32.1,

31.4, 30.2, 30.1, 29.8, 29.7, 29.5, 29.5, 27.0, 22.8, 14.3 (C₈–C₁₇and

+

ꢀ

C₈–C₁₅); MS (CI, CH₄), m/z 432.41 (M , 62.42%), 433.3686 (MH ,

93.99%), 153.064 (MH−CH₂C(CH₂)₉CH₃(CH₂)₇CH₃100%); HRMS

(CI, CH₄): calcd (C₂₈H₄₉O₃, MH⁺) 433.3682, obsd 433.3626.

2.3.9.1. 2-Butyloctyl bromide (23h). 70.2% yield; ¹H NMR (CDCl₃) ı

3.45 (d, J = 4.8 Hz, 2H, H₁), 1.6 (m, 1H, H₂), 1.3 (m, 16H, H₃–H₇and

2.3.6.9. Methyl-4-(2-decyltetradecyloxy) benzoate (22k). 1.52%

yield; R_f= 0.44 (80% hexane in ethyl acetate); ¹H NMR (CDCl₃)

ꢀ

H

₃–H₅), 0.92 (m, 6H, H₈and H₆); ¹³C NMR (CDCl₃) ı 39.7 (C₁and

ı 7.98, 6.9 (AA^ꢀXX^ꢀsystem, 2H each, H₃, H₃, H₄and H₄), 3.88

ꢀ

C₂), 32.7, 32.4, 32.0, 29.6, 28.9, 26.7, 23.0, 22.8, 14.2, 14.16 (C₃–C₈,

ꢀꢀ

(d, J = 5.5 Hz, 2H, H₆), 3.87 (s, 3H, H₁), 1.79 (m, 1H, H₇), 1.26 (m,

40H, H₈–H₁₆and H₈–H₁₈), 0.88 (“t”, 6H, H₁₇and H₁₉); ¹³C NMR

ꢀ ꢀ

C₃–C₆).

ꢀ

(CDCl₃) ı 167.0 (C₁), 163.3 (C₅), 131.6 (C₃and C₃), 122.4 (C₂), 114.2

(C₄and C₄), 71.2 (C₆), 51.9 (C₁), 38.0 (C₇), 32.1, 31.4, 30.1, 29.82,

2.3.9.2. 2-Hexyldecyl bromide (23i). 62.3% yield; ¹H NMR (CDCl₃) ı

3.44 (d, J = 6 Hz, 2H, H₁), 1.58 (m, 1H, H₂), 1.28 (m, 24H, H₃–H₉and

H

ꢀ

ꢀꢀ

ꢀ

29.8, 29.75, 29.5, 29.49, 27.0, 22.8, 14.2 (C₈–C₁₉and C₈–C₁₇);

MS (CI, CH₄), m/z 488.4215 (M⁺, 31.35%), 489.429 (MH⁺, 28.15%),

₃–H₇), 0.89 (m, 6H, H₁₀and H₈); ¹³C NMR (CDCl₃) ı 39.7 (C2),

ꢀ ꢀ

ꢀ

106

Y. Cohen et al. / Chemistry and Physics of Lipids 155 (2008) 98–113

39.7 (C1), 32.7, 32.1, 32.0, 30.0, 29.7, 29.6, 29.5, 26.72, 26.7, 22.83,

ꢀ

22.81, 14.2 (C₃–C₁₀, C₃–C₈).

2.3.9.3. 2-Decyltetradecyl bromide (23k). 70% yield; ¹H NMR

(CDCl₃) ı 3.44 (d, J = 4.8 Hz, 2H, H₁), 1.6 (m, 1H, H₂), 1.27 (m, 40H,

H₃–H₁₃and H₃–H₁₁), 0.89 (m, 6H, H₁₄and H₁₂); ¹³C NMR (CDCl₃)

ꢀ

ı 39.7 (C2), 39.6 (C1), 32.8, 32.82, 30.0, 29.83, 29.79, 29.6, 26.7, 22.9,

ꢀ

14.3 (C₃–C₁₄, C₃–C₁₂).

3. Results and discussion

Scheme 3. Synthetic routes to ␣-disubstituted aldehydes 2.

3.1. Initial criteria for reporter molecules

Aldol cyclization leads to ␤-hydroxyketone 5 which yields enone

6 upon dehydration. Oxidation of the latter with DDQ generates

the desired cyclohexadienone 1. The vinyl hydrogens of dienones 1

have an AA^ꢀXX^ꢀsplitting pattern which is characteristically different

from that of the analogous hydrogens in enone system 6 .

Turning to the starting materials, the short-chain symmetri-

cal (where R = R^ꢀ) aldehydes 2a–d are commercially available. The

longer chain aldehydes 2e–g could be synthesized via the hydrol-

ysis of enol ethers 9 (Taber et al., 1989; Hu and Mattern, 2000 )

(Scheme 3). The latter were prepared in turn by treating com-

mercially available symmetrical ketones 7 with the Wittig reagent

methoxymethylenetriphenylphosphorane (8). The very long-chain

asymmetrical aldehyde 6h was ﬁnally prepared by the chromate

oxidation of the corresponding alcohol 10 (Scheme 3 ) (Corey and

Suggs, 1975).

Two comments are appropriate at this juncture. Firstly, this

route to the aldehyde has been previously carried out in a one-

pot method directly from the ketone, without the isolation of the

intermediate enol ether 9 (Taber et al., 1989; Hu and Mattern, 2000).

Nevertheless, it is preferable in our system to add this isolation step

because it allows for removal of starting ketone 7, whose polarity

is very similar to that of the ﬁnal aldehyde 2. Secondly, we note

that the aldehydes undergo very facile autoxidation to the cor-

responding carboxylic acids 11; hence, the aldehydes were used

immediately upon synthesis.

A priori, four properties would seem to be necessary for an

efﬁciently constructed “ruler”: (1) the compounds should have a

planar backbone—which simpliﬁes distance and dihedral angle cal-

culations. (2) The molecules should lie vertically within the lipid

bilayer, i.e., perpendicularly to the water/lipid interface. Putting

these ﬁrst two requirements together suggested molecules with

a hydrophilic/polar “head” group, a planar and symmetrical (or

near-symmetrical) body, and symmetrical (or near-symmetrical)

hydrophobic/lipophilic tails. (3) The hydrophobic tails should be

modular, permitting the ready synthesis of a family of compounds

with varying lipophilicity—and hence varying depths within the

lipid bilayer. (4) Our experience (Frimer et al., 1983, 1996; Frimer,

1985; Strul et al., 1993,1994; Weitman et al., 2001; Afri et al., 2002,

2004a,b; Bronshtein et al., 2004) has further taught us that accu-

rate ¹³C chemical shift/E_T(30) correlations require systems in which

the increase in the ¹³C NMR chemical shift (henceforth, ꢀı) of the

reporter carbons with solvent polarity is relatively large. In par-

ticular, we have found that results are reliable when the change

in ¹³C chemical shift is ≥1.8 ppm over ca. 20 kcal/mol E_Trange

(e.g., in going from methanol [E_T(30) = 55.4 kcal/mol] to benzene

[E_T(30) = 34.3]. The ꢀı is particularly large in conjugated systems

capped at alternate ends by a donor and an acceptor moiety, where

a large “push–pull” effect is present.

Based on the above criteria we began our search for appropriate

compounds with the preparation of cyclohexadienones 1a–h.

3.2.2. Locating cyclohexadienones 1a–h within

dimyristoylphosphatidylcholine (DMPC) liposomes

3.2. 4,4-Dialkylcyclohexa-2,5-dienones 1a –h

As noted above, it was our aim to determine the exact location

of speciﬁc carbons in each compound in the liposomal bilayer. As

we explained earlier, to do so requires developing a graph which

correlates between the ¹³C NMR chemical shift and solvent polar-

3.2.1. Synthesis

The title compounds were synthesized via a Robinson annula-

tion approach, as outlined in Scheme 2 (Frimer et al., 1989).

In the ﬁrst step, a Michael addition of the enolate of ␣-

disubstituted aldehyde 2 to methyl vinyl ketone 3 yields dione 4.

ity for the various carbons of each compound. To this end, the ¹³

C

NMR data were acquired for each of the compounds in four or ﬁve

solvents of varying E_T(30) polarity, generally carbon tetrachloride

(32.4), benzene (34.3), chloroform (39.1), acetone (42.2), acetoni-

trile (45.6) and methanol (55.4). In general, the correlations were

excellent (correlation coefﬁcient, r²> 0.9). From the accrued data,

it becomes clear that there are certain carbons which are most sen-

sitive to polarity change, i.e., had ꢀı = ≥1.8 ppm over an E_Trange

of ∼20 kcal/mol. We then prepared ¹³C NMR chemical shift/E_T(30)

polarity correlation graphs (henceforth, simply dubbed “correla-

tion graphs”) for each of these sensitive “reporter” carbons.

The next step was to intercalate each of the various derivatives

within the DMPC liposomal phospholipid bilayer and determine

the chemical shifts of the “reporter” carbons. Based on the corre-

lation graph, it was possible to calculate from the chemical shift

values the corresponding microenvironment polarity (expressed

as an E_T(30) value) presumably felt by a given reporter “carbon”.

The relative polarity values of these carbons give us a sense of the

molecules location and orientation. (We note, parenthetically, that

reporter moieties, such as carbonyls, might conceivably drag water

Scheme 2. Synthetic route to cyclohexadienones 1.

Y. Cohen et al. / Chemistry and Physics of Lipids 155 (2008) 98–113

107

Table 1

Calculated polarity (E_T(30) in kcal/mol) of carbons C-1, C-3 and C-4 of compounds

1a–h intercalated within the liposomal bilayer

Compound

C-1

C-3

C-4

1a

1b

1c

1d

1e

1f

58.80

48.58

46.45

44.48

42.03

39.67

38.14

31.60

57.70

48.48

45.61

44.07

41.62

39.70

38.49

35.74

49.18

46.25

43.64

42.63

40.14

38.36

36.38

36.22

1g

1h

with it the prominence of resonance structure 1B. The sizable ꢀı

of C-4 must result from an inductive effect.

Dienones 1a–h were then intercalated into DMPC liposomes and

the chemical shifts of carbons C-1, C-3 and C-4 were measured. The

calculated polarity (E_T(30) values) are listed in Table 1.

The following interim observations can be made about these

results:

Fig. 1. Plots of the ¹³C NMR chemical shift (ppm) for C-1, C-3 and C-4 of cyclo-

hexadienone 1f vs. E_T(30) solvent polarity (kcal/mol). (The solid symbols represent

chemical shift values in pure solvent, while hollow symbols are for those within

liposomes.)

(1) The data of Table 1 conﬁrm our assumption that long chain

derivatives impart greater lipophilicity to molecules. Looking

at the series 1a–h, we see that the longer the chains are at C-

4: the more lipophilic the molecule becomes; the deeper the

molecule penetrates into the bilayer toward the lipid slab; and,

as a result, the lower the E_T(30) values of the reporter carbons

become. Perhaps surprisingly in this regard, an n-butyl chain

(1e) imparts greater lipophilicity than a benzene ring (1d).

(2) Nevertheless, Table 1 contains seemingly irresolvable contra-

dictions. Thus, in derivative 1a, C-1 and C-3 have essentially

the same E_T(30) polarity value, suggesting that this section of

the molecule lies almost horizontally. However, these polarity

values are almost 10 E_T(30) units higher than C-4, suggesting

that the molecules is nigh vertical.

(3) In derivatives 1b–g, C-1 and C-3 have nearly the same polarity

value again suggesting that the dienones lie almost parallel to

the interface, horizontally in the bilayer—all this despite the

polar carbonyl head group and very lipophilic tails. What is

more, in the case of the most lipophilic dienone 1h, the E_T(30)

value of carbonyl carbon C-1 is actually lower than both C-3 and

the highly lipophilic C-4.

Scheme 4. Resonance structures of cyclohexadienones 1.

with them into normally hydrophobic regions and, thus, give a false

reading. We believe this to be highly unlikely, though. After all,

the carbonyl is essentially isolated, surround on all sides by highly

lipophilic groups which overwhelmingly determine the solvation

and polarity properties of the microenvironment.)

When we followed this procedure for dienones 1, we revealed

that carbons C-1, C-3 and C-4 were the most sensitive to polar-

ity change. Fig. 1 exempliﬁes this process for dienone 1f in which

the ꢀı values in going from benzene to methanol were 3.91, 5.25

and 2.53 for carbons C-1, C-3 and C-4, respectively, with correlation

factors (r²) ≥0.96.

3.2.3. Implications to the construction of a molecular ruler

The sensitivity of carbons C-1, C-3 to solvent polarity is not at

all surprising in light of the resonance structures of dienone sys-

tem 1 (Scheme 4). We see that both C-1 and C-3 have large charge

separation in structures 1A and 1B, respectively. The particularly

large change in the chemical shift of C-3 (ꢀı ≈ 5 ppm) suggests that

charge separation becomes more facile as polarity increases, and

As noted above, the calculated E_Tresults left us quite confused.

We had expected to see a correlation between the lipophilicity of

the substituents on the various intercalants and their depth in the

bilayer. This was in general observed, perhaps most clearly in the

case of dienones 1. Proceeding from 1a (where the total number

of carbons in R and R^ꢀis 2) to 1h (where this number has gone

Table 2

Calculated polarity of carbons C-I and C-II of methyl alkyl diesters 15a–g intercalated within the liposomal bilayer

c

Compound

Carbons

in R^ꢀ

Carbonyl-I^a

(kcal/mol)

Carbonyl-II^b

(kcal/mol)

ꢀE_T/ꢀÅ_I–II

(kcal/mol Å)

15a

15b

15c

15d

15e

15f

5

6

8

12

16

20

24

44.06

44.57

44.44

44.20

44.60

45.47

44.50

39.35

39.47

39.30

38.20

37.27

39.35

39.28

0.63

0.96

0.69

0.81

0.99

0.83

0.70

15g

Average 2a–g: 0.80

a

Experimental error = 1; average ꢀı = 2.62; average r²= 0.98.

Experimental error = 1; average ꢀı = 2.50; average r²= 0.98.

ꢀÅ (C_I–C_II): 7.34 Å.

b

c

108

Y. Cohen et al. / Chemistry and Physics of Lipids 155 (2008) 98–113

Table 3

Calculated polarity of carbons C-I and C-II of methyl alkyl diesters 17a–g intercalated within the liposomal bilayer

c

Compound

Carbons

in R^ꢀ

Carbonyl-I^a

(kcal/mol)

Carbonyl-II^b

(kcal/mol)

ꢀE_T/ꢀÅ_I–II

(kcal/mol Å)

17a

17b

17c

17d

17e

17f

5

6

8

12

16

20

24

42.40

42.57

43.48

43.67

43.79

43.54

43.33

31.61

34.89

35.28

35.04

33.54

34.80

n.d.^d

1.06

0.75

0.80

0.84

1.00

0.85

–

Average 4a–f: 0.88

17g

a

Experimental error = 1; average ꢀı = 2.40; average r²= 0.96.

Experimental error = 1; average ꢀı = 2.35; average r²= 0.95.

ꢀÅ (C_I–C_II): 10.23 Å.

b

c

d

Signal not detected.

up 11-fold to 22) causes C-1 of the intercalant to plummet from

E_T= 58.8 – not far from the water–lipid interface, down to E_T= 31.6

– presumably near or in the lipid slab.

The analysis regarding dienones 1 is in fact true for most conju-

gated systems where developing charge can be transmitted several

carbons away via the ␲-system. A related analysis resolves anal-

ogous results mutatis mutandis observed for two other systems

we studied brieﬂy: N-alkyl pyridones 12 and 4-alkoxypyridine N-

oxides 13 (Scheme 5). In these cases, the E_T(30) values calculated

for C-3 or C-4, respectively, reﬂect not the inherent polarity expe-

rienced at that carbon because of the microenvironment polarity,

but rather the charge build-up transferred to it in part from C-1 or

C-2, respectively (Cohen, 2005).

The above data lead us clearly to one important conclusion.

In developing a chemical ruler, we can only use molecules com-

posed of polar carbons which are not strongly conjugated, even

though they would seem preferable due to their large ꢀı. Instead,

we require systems which have at least two polar moieties probes

(e.g. carbonyls) – at a known distance from each other – whose

charge development in each case is dependant exclusively on its

own microenvironment. Therefore, an additional demand for the

molecular ruler is that the polar carbon bond must not be part of a

strongly conjugated system.

However, despite the amphiphilic nature of the various inter-

calants, at ﬁrst blush, the E_Tvalues of Tables 1–3 would seem to

strongly suggest that these molecules do not lie perpendicularly

to the interface as was expected. Indeed, in the case of dienones

1, for example, we anticipated that the E_Tvalues for C-1 would be

signiﬁcantly higher than those of C-3 and C-4. This would have indi-

cated that the C-1 carbonyl is situated closer to the polar interface,

while C-3 and the adjacent lipophilic C-4 would be substantially

deeper within the bilayer. Instead, the results of Table 1 show

that carbons C-1 and C-3 are very close to each other, while C-4

is only slightly lower, depending on the substituent. Unexpect-

edly, in compound 1h, C-1 is even lower than C-3, which is lower

still than C-4; this would suggest that the carbonyl is actually

deeper in the liposome than the long chain substituents! These

results suggest either that an unexpected phenomenon is occur-

ring, or alternatively that some of our fundamental premises are in

error.

All the above data lead us to the following conclusions regarding

the dienone 1 system:

3.3. Unconjugated diesters 15 and 17

(1) The longer the chains in the dienone system, the more the

molecule becomes lipophilic and, therefore, the deeper it

plunges into the bilayer.

3.3.1. Synthesis

In order to meet the above new criterion, we attempted to engi-

neer compounds that would have the following properties: (a) they

would contain two carbonyls as reporter carbons, separated by a

known distance one from the other, which are in no way conju-

gated to each other or any other moiety. (b) The molecules would

contain an aromatic ring to give it linearity and allow for easy detec-

tion by TLC. (c) The functional groups would permit the preparation

of derivatives of varying lipophilicity, for varying penetration of the

(2) It seems that the charge on C-1, which develops as the environ-

ment at C-1 becomes more polar, distributes itself between C-1

and C-3—this is a result of the resonance built into the conju-

gated enone system. Thus the amount of charge that develops

on C-3 is not totally or primarily a function of the microenvi-

ronmental polarity at C-3, but rather is highly dependant on

the polarity experienced at C-1! It is not completely surpris-

ing; therefore, that both C-1 and C-3 can give similar E_Tresults.

Hence, we conclude that in the case of dienones 1, the values

of C-3, are unreliable.

(3) As far as C-4 is concerned, the numbers here, too, are problem-

atic since no charge development is expected. The chemical

shifts and calculated E_Tvalues observed reﬂect not some

inherent charge developing at C-4—because of the microenvi-

ronmental polarity C-4 feels. Rather, they reﬂect the induction

transmitted to it from C-3, which in turn depends on the charge

developing at C-1.

Therefore, there is a fundamental problem in using the cyclo-

hexadienones in developing a chemical ruler. In fact, it is possible

that indeed some of the dienone derivatives lie perpendicular to

the interface, as we originally expected; however, we have no way

of proving or measuring this.

Scheme 5. Derivatives of heterocycles 12 and 13.

Y. Cohen et al. / Chemistry and Physics of Lipids 155 (2008) 98–113

109

Scheme 6. Synthetic route to dicarboxylic esters 15 and 17.

bilayer. To this end, we turned to benzenediacetic esters 15 and the

corresponding dipropionic esters 1 7 . Of the two independent car-

bonyls, one is a methyl ester which will serve as the polar head

group, while the other is a long chain, highly lipophilic, ester which

will hopefully be anchored in or near the lipid slab.

As outlined in Scheme 6, esters 15 and 17 were prepared from

the corresponding dicarboxylic acids 14 and 16, respectively.

The diacids were converted to diacyl chlorides with excess

thionyl chloride. Upon removal of the latter, the diacyl chlorides

were reacted with 0.5 equiv. of the long chain alcohol (R^ꢀOH) yield-

ing the corresponding monoester; the latter was then treated with

excess methanol (for analogs a–g), ethanol (for analog 15h) or

isopropanol (for analog 15i) creating the desired mixed ester. Iden-

tiﬁcation of the two carbonyls (dubbed C-I for the methyl ester

carbonyl, and C-II for that of the long chain ester) was accomplished

by ¹³C NMR 2D HMQC and HMBC techniques.

Fig. 3. Plots of the ¹³C NMR chemical shift (ppm) for C-I (diamond) and C-II (square)

of 17b vs. solvent polarity (E_T(30)).

Focusing on the polarizable carbonyls C-I and C-II alone, NMR

spectra of derivatives 15a–i and 17a–g were run in ﬁve pure deuter-

ated solvents [E_T(30) values in kcal/mol in parentheses]: methanol

(55.4), acetonitrile (45.6), acetone (42.2), chloroform (39.1), and

benzene (34.3). Correlation graphs (exempliﬁed by Figs. 2 and 3)

between solvent polarity (E_T(30) values) and chemical shift (ı) were

prepared. We then intercalated the diesters into dimyristoylphos-

phatidylcholine (DMPC) liposomes and observed the NMRs of C-I

and C-II. The correlation graphs were then used to determine the

calculated polarity (E_T(30)) values experienced by carbons C-I and

C-II within the liposomal bilayer for each derivative. These calcu-

lated E_T(30) values give us, in turn, insight as to where these carbons

are located along the continuum of polarities ranging from highly

polar at the interphase down to non-polar at the lipid slab. The

results are summarized in Tables 2 and 3, respectively. Table 4 com-

pares the results for 15c, h and i and gives us insight into the effect

of changing the head group lipophilicity.

We should note that, in the NMR studies described through-

out this paper, the molar ratio of intercalant:DMPC was generally

1:5. This rather high concentration of intercalant was necessary in

order to observe the desired reporter peaks via NMR. Interestingly,

however, in the ¹³C NMR spectra of the very long chain derivatives

2f, 2g, 4f and 4g of the unconjugated diesters, where the carbons

in R^ꢀrange from 20–24, we observed double carbonyl peaks. One

pair was sharp and located at relatively high ﬁeld (e.g., in 2g, they

appeared at 170.58 and 170.98 ppm), while the other pair was broad

and situated at lower ﬁeld (e.g., in 2g, they were observed at 171.60

Table 4

Calculated E_T(30) of carbonyls C-I, and C-II in compounds 15c, 15h and 15i interca-

lated within the liposomal bilayer

b

Compound

Head group

Carbonyl I^a

(kcal/mol)

Carbonyl II^a

(kcal/mol)

ꢀE_T/ꢀÅ_I–II

(kcal/mol Å)

15c

15h

15i

MeO–

EtO–

(CH₃)₂CHO–

44.44

40.80

38.31

39.30

39.15

40.81

0.69

0.22

−0.34

a

Experimental error = 1; average ꢀı: OMe = 2.58; OEt = 2.52; OCH(CH₃)₂= 2.45.

Average r²= 0.98.

Fig. 2. Plots of the ¹³C NMR chemical shift (ppm) for C-I (diamond) and C-II (square)

of 15b vs. solvent polarity (E_T(30)).

b

Experimental error = 1; ꢀÅ (C_I–C_II): 7.34 Å.

110

Y. Cohen et al. / Chemistry and Physics of Lipids 155 (2008) 98–113

under study, i.e., to what extent the intercalant is parallel to

the lipid chains and perpendicular to the water–lipid interface.

As the verticality increases, the polarity difference between

the reporting carbons (of ﬁxed ꢀÅ), as reﬂected in the ꢀE_T

observed, is expected to increase as well. Thus, when ꢀE_T/ꢀÅ

is maximal, we can assume that the intercalated compound lies

essentially vertical within the lipid bilayer.

The second factor affecting the size of ꢀE_T/ꢀÅ is the depth

of the reporter carbon within the bilayer. As noted above, a gra-

dient of polarities exists within the bilayer ranging from water

at the interface, down to hexane within the lipid slab. There

is no reason to assume, however, that the change in polarity

is linear with depth. As a result, ꢀE_T/ꢀÅ may well depend

on the region within the lipid bilayer that the measurements

are taken. For example, change in ꢀE_T/ꢀÅ may be moderate in

the upper polar region of the bilayer, increase rapidly and peak

in the lower polar region, and then slow dramatically as one

approaches or enters the lipid slab.

In Tables 2 and 3, we see that for diesters 15, which lie in

the relatively non-polar ꢀE_Tregion of 45–39, the ꢀE_T/ꢀÅ is

ca. 0.80 0.19. Interestingly, for diesters 17, lying in the ꢀE_T

region 44–32, the ꢀE_T/ꢀÅ value is 0.94 0.19, not substantially

different. The data suggest that in range of ꢀE_Tvalues between

45 and 32 there does not seem to be any dramatic changes in

alignment of these molecules or in the value of ꢀE_T/ꢀÅ with

depth.

Scheme 7. Calculated distance between the carbonyls of diesters 15 and 17.

and 172.49 ppm). The doubling results from two populations in

these samples: one group includes those molecules which are inter-

calated within the liposome and feel the polarity corresponding

to the liposomal microenvironment; while the other consists of

aggregates of these long chain derivatives, and, hence, they feel the

microenvironment polarity created by themselves, and should be

essentially phospholipid independent (Cohen et al., 2008b).

3.3.2. Location within the bilayer

From the data of Tables 2–4, we can glean several important

pieces of information.

(4) Table 4 lists the results obtained for two additional analogs of

15c, diesters 15h and 15i. The latter differ from 15c only in

that they possess the more lipophilic ethoxy and iso-propoxy

head groups, respectively, instead of the standard methoxy moi-

(1) Firstly, we see that within each diester family (diacetate 15a–g

or dipropionate 17a–g), the location of the C-I carbonyl for

each derivative is essentially the same, independent of the

length/lipophilicity of the tail. The same is true for the lower

lying C-II carbonyl for each derivative. The very lipophilic val-

ues of C-II (hexane has an E_T(30) value of 31 kcal/mol), suggests

that the lipophilic tail in all cases rapidly reaches into the lipid

slab. As a result, lengthening the fatty acid chain further does

not pull the molecule into a more lipophilic region solvent. With

the lipophilic tail well solvated in the lipophilic slab, the loca-

tion of C-I and C-II in compounds 15 and 17 is pretty much

determined by the solvation needs of these carbonyls and the

distance between them.

(2) In this light, we now compare between the families, and see that

the C-I carbonyl in both series are anchored close together—at

∼44.5 kcal/mol for the diacetates and a tad lower at 43.5 for the

dipropionates. Not surprisingly, though, the micropolarity felt

by the respective C-II carbonyls in these two families is substan-

tially different. Thus, in the diacetate family 15a–g, the lower

carbonyl C-II, which is 7 bonds away from, is 6.5 E_Tunits lower

than that of C-I, centered at E_T(30) = 38 kcal/mol. In the dipro-

pionate family 17a–g, C-II which is 9 bonds away from C-I, is 9

E_Tunits lower than C-I, centered at 33 kcal/mol, near or in the

lipid slab.

ety. Interestingly, Table 4, particularly the value of ꢀE_T/ꢀÅ ^b

,

I–II

reveals that the effect of changing the head group even slightly

is quite dramatic. As the head group becomes more lipophilic,

the lipophilicity difference between the head and the tail

rapidly decreases and the diesters quickly orient themselves

horizontally within the liposome. Thus, ꢀE_T/ꢀÅ ^bis 0.69 for

I–II

the methyl ester, but drops to 0.22 and −0.34 for the ethyl and

isopropyl analogs, respectively.

All the above results are quite consistent with these methyl

alkyl diester families lying vertically within the liposomal

bilayer—suggesting them as excellent candidates for the desired

molecular ruler. Unfortunately, since all the derivatives in each

family lie at the same location, the multiple derivatives are

redundant. Nevertheless, one example from each family can sup-

ply data points for our ruler, which will be discussed further

below.

3.4. 4-Alkoxybenzaldehydes 19 and methyl 4-alkoxybenzoates 22

as ruler candidates

(3) Let us introduce a quantity ꢀE_T/ꢀÅ_I–II, where ꢀÅ_I–IIis the dis-

tance in Å between two given “reporter” carbons in a molecule,

C-I and C-II, while ꢀE_Tis the difference in E_T(30) values

between these carbons. Thus, ꢀE_T/ꢀÅ_I–IIgives us a measure

of how much the E_T(30) changes in going from one point of

given depth to another. The latter was derived in each fam-

ily from the geometric relationships (bond length and angles)

between these carbonyls in the optimized (minimum energy)

conﬁguration with the aid of PCMODEL calculations. The dis-

tances between the reporter carbonyls in diesters 15 and 17 are

shown in Scheme 7 and are 7.34 and 10.23 Å, respectively.

The size of the ꢀE_T/ꢀÅ presumably depends on two factors.

The ﬁrst is the verticality of the lipid-intercalated compound

3.4.1. Synthesis

In an attempt to widen the number of substrates contribut-

ing to our ruler, we turned to two aromatic systems bearing polar

groups, 4-alkoxybenzaldehydes and methyl 4-alkoxybenzoates. A

preliminary study of the change of ¹³C NMR chemical shifts (ı) with

increasing solvent polarity was carried out on the carbons of the

commercially available methyl 4-methoxybenzaldehydes (19a in

Scheme 8) and 4-methoxybenzoates (22a in Scheme 9). In both

cases, the ꢀı values around the ring in going from benzene to

methanol (from E_T(30) 34.3 to 55.4 kcal/mol) were negligible except

for the exocyclic carbonyl carbon and ring carbon para to it, which

were large (>2.3). We thus explored the preparation of these fami-

lies of compounds.

Y. Cohen et al. / Chemistry and Physics of Lipids 155 (2008) 98–113

111

Fig. 5. Plots of the ¹³C NMR chemical shift (ppm) for C-1 and C-5 of methyl 4-(2-

octyldodecyloxy)benzoate (22j) vs. solvent polarity (E_T(30)).

Scheme 8. Synthetic route to 4-alkoxybenzaldehydes 19.

determine the calculated polarity (E_T(30)) values experienced by

carbons C-1 and C-5 within the liposomal bilayer for each aldehyde

or ester derivative. The results are summarized in Tables 5 and 6.

3.4.2. Location within the bilayer

Several important conclusions can be drawn from these data.

(1) The C-1 carbonyl of the benzoate esters 22, as a group, falls

substantially (8–13 E_Tunits) lower than the corresponding car-

bonyl in benzaldehydes 19. Within each of the families, the

derivatives tend to break themselves into three subgroups:

short-chain unbranched, long-chain unbranched and long-

chain branched derivatives. Thus, in the case of the aldehydes,

the E_T(30) values for C-1 and C-5 of the unbranched members

(19b–f) fall at approximately ca. 51 and 38 kcal/mol, respec-

tively, while these carbons in the branched analogs (19g–k) lie

2–3 E_Tunits deeper at values of 48 and 36 kcal/mol, respec-

tively. In the case of esters 5 the differences are much smaller.

The E_T(30) values C-1 and C-5 of the unbranched members fall

approximately at 38.5 and 35 kcal/mol, respectively; these car-

bons in the branched analogs lie 1–1.5 E_Tunits deeper at 37.5

and 33.5 kcal/mol. For the short-chained analogs 19a and 22a,

while C-1 is located essentially where the unbranched long-

chain analogs are, C-5 is situated ca. 3 E_Tunits higher.

(2) Tables 2 and 3 also records that the difference in E_T(ꢀE_T)

between carbonyls C-1 and C-5 is substantially smaller in the

benzoate esters 22 (3–5 kcal/mol) than it is in the benzaldehy-

des 19 (12–13 kcal/mol). This is despite the fact that the distance

in Angstroms (ꢀÅ) between C-1 and C-5 in both the benzoate

ester and benzaldehyde systems was determined by PCMODEL

calculations to be the same: 4.26 Å.

(3) Tables 2 and 3 reveals that the para carbon C-5 for the branched

benzaldehydes (19g–k), and that of the unbranched benzoates

(22a–f) lie approximately in the same region (E_T35–36). Thus,

a comparison of the ꢀE_T/ꢀÅ_1–5values between these families

is appropriate. The data reveal that average ꢀE_T/ꢀÅ_1–5for the

branched aldehydes (19g–k) is 2.74 (spanning a region from E_T

48 for C-1 to E_T36 for C-5), as compared to a much smaller value

of 0.87 (spanning a region from E_T= 38–35) for the unbranched

benzoates (22a–f). This despite the fact that the distance from

C-1 to C-5 (ꢀÅ_1–5) is the same in both systems.

Scheme 9. Synthetic routes to methyl 4-alkoxybenzoates 22.

The various benzaldehyde derivatives were synthesized using

the century and a half old Williamson reaction (Williamson, 1852 ),

either under the classic strongly basic conditions (Barton and Ollis,

1979) (Scheme 8, method I) or via the more convenient one-pot

Hong-bin variation (Hong-bin et al., 2001) (method II).

Methyl benzoates 22b–d were prepared via the esteriﬁcation

of the corresponding 4-alkoxybenzoic acids (Scheme 9, method I),

while analogs 22a and 22e–k were prepared from the correspond-

ing phenol 21, via the Hong-bin variation (Hong-bin et al., 2001) of

the Williamson reaction (method II).

In both the aldehyde 19 and methyl ester 22 systems, NMR

spectra of the various derivatives were run in four or ﬁve pure

deuterated solvents selected from methanol, acetonitrile, acetone,

chloroform and benzene. Focusing on the exocyclic carbonyl car-

bon (C-1) and ring carbon para to it (C-5), correlation graphs were

prepared for each of the derivatives (see typical Figs. 4 and 5 for

aldehyde 19h and ester 22j) and the compounds were intercalated

into the DMPC liposome. The correlation graphs were then used to

Two explanations can be suggested to resolve this discrepancy.

One is that benzaldehydes lie vertically in the lipid bilayer while

the benzoates lie at a substantial angle; thus the ꢀE_Tof the former

is larger. This explanation is unlikely, however, since the average

ꢀE_T/ꢀÅ values of the benzoates (0.87 for the unbranched analogs

and 0.96 for the branched) is consistent with that observed above

for diesters 15 (0.80) and 17 (0.94)—which are located in the same

Fig. 4. Plots of the ¹³C NMR chemical shift (ppm) for C-1 and C-5 of 4-(2-

butyloctoxy)benzaldehyde (19h) vs. solvent polarity (E_T(30)).

112

Y. Cohen et al. / Chemistry and Physics of Lipids 155 (2008) 98–113

Table 5

Calculated E_T(30) values of carbons C-1 and C-5 of benzaldehydes 19a–k intercalated within the liposomal bilayer

C-1^a(kcal/mol)

C-5^b(kcal/mol)

ꢀE_T/ꢀÅ_1–5

c

Compound

Carbons in R

(kcal/mol Å)

19a

19b

19c

19d

19e

19f

19g

19h

19i

1

8

49.42

49.70

49.50

51.64

51.58

50.44

47.39

47.28

47.60

48.56

41.76

38.01

38.34

37.23

38.55

38.37

35.45

36.61

35.84

36.06

36.77

1.80

2.74

2.62

3.38

3.06

2.83

2.80

2.50

2.68

2.71

2.77

19a: 1.80

10

12

15

18

8

12

16

20

24

Average 19b–f: 2.93

Average 19g–k: 2.70

19j

19k

a

Experimental error = 1; average ꢀı = 2.99; average r²= 0.95.

Experimental error = 1; average ꢀı = 1.72; average r²= 0.90.

ꢀÅ (C₁–C₅): 4.26 Å.

b

c

Table 6

Calculated E_T(30) values of carbons C-1 and C-5 of methyl benzoates 22a–k intercalated within the liposomal bilayer

c

Compound

Carbons in R

C-1^a(kcal/mol)

C-5^b(kcal/mol)

ꢀE_T/ꢀÅ_1–5

(kcal/mol Å)

22a

22b

22c

22d

22e

22f

22g

22h

22i

1

8

38.05

38.65

38.37

38.98

38.27

38.22

38.83

36.93

37.52

38.49

36.39

38.34

34.93

35.87

34.75

33.99

34.40

35.60

32.61

33.42

34.82

31.33

−0.07

0.87

0.60

0.99

1.00

0.90

0.76

1.01

0.96

0.86

1.19

22a: −0.07

10

12

15

18

8

12

16

20

24

Average 22b–f: 0.87

22j

22k

Average 22g–k: 0.96

Average 22b–k: 0.91

a

Experimental error = 1; average ꢀı = 1.96; average r²= 0.95.

Experimental error = 1; average ꢀı = 1.34; average r²= 0.80.

ꢀÅ (C1–C₅): 4.26 Å.

b

c

region (E_T= 44–34). A more likely explanation is that the large val-

ues in the benzaldehydes 19 are unreliable—and, hence need to be

discarded. This is because of resonance of the carbonyl moiety with

– and the dispersal of charge into – the aromatic ring. Such reso-

nance is much less pronounced in the case of esters because of the

swamping out effect of the alkoxy moiety (see, for example, Russell,

1973).

We conclude, therefore, that only the benzoate 22 data are

dependable. Unfortunately, all the members in this group lie in

essentially the same region and, hence, we will only be able to

use one or two representative (branched and unbranched) for con-

structing a ruler.

the construction of a molecular ruler, while 4,4-dialkylcyclohexa-

2,5-dienones 1, N-alkyl pyridones 12, 4-alkoxypyridine N-oxides

13 and 4-alkoxybenzaldehydes 19 are not. Time has come for the

actual construction of a molecular ruler. Our efforts in this regard

are described in the companion paper (Cohen et al., 2008a).

Acknowledgements

We acknowledge the kind and generous support of the Israel

Science Foundation (Grants Number 327/02 and 437/06)—founded

by The Israel Academy of Sciences and Humanities, and The Ethel

and David Resnick Chair in Active Oxygen Chemistry.

3.5. Conclusions

References

In this paper, we have developed and applied ﬁve criteria for the

selection of molecules for the use in the development of a molec-

ular ruler: (1) Planarity: the molecule should be close to planar in

structure in order to simplify geometric calculations. (2) Dipolarity:

it should bear a polar head group and a hydrophobic tail group, to

force the molecule to align itself in the bilayer perpendicularly to

the interface. (3) Modularity: the synthesis of the hydrophobic tails

of these rulers should be modular allowing varying lipophilicities.

(4) Reliability: the reporter carbons should show a large change

in chemical shift as we proceed to more polar solvents. Finally

(5) Charge localization: the reporter carbons should not be part of

a strongly conjugated system. Using these criteria, we have seen

that benzenediacetic esters 15, benzenedipropionic esters 17, and

methyl 4-alkoxybenzoates 22 are appropriate systems for use in

Abrams, F.S., London, E., 1992. Calibration of the parallax ﬂuorescence quenching

method for determination of membrane penetration depth: reﬁnement and

comparison of quenching by spin-labeled and brominated lipids. Biochemistry

31, 5312–5322.

Afri, M., Ehrenberg, B., Talmon, Y., Schmidt, J., Cohen, Y., Frimer, A.A., 2004a. Active

oxygen chemistry within the liposomal bilayer. Part III. Locating vitamin E,

ubiquinol and ubiquinone and their derivatives in the lipid bilayer. Chem. Phys.

Lipids 131, 107–121.

Afri, M., Frimer, A.A., Cohen, Y., 2004b. Active oxygen chemistry within

the liposomal bilayer. Part IV. Locating 2^ꢀ,7^ꢀ-dichloroﬂuorescein (DCF),

2^ꢀ,7^ꢀ-dichlorodihydroﬂuorescein (DCFH) and 2^ꢀ,7^ꢀ-dichlorodihydroﬂuorescein

diacetate (DCFH-DA) in the lipid bilayer. Chem. Phys. Lipids 131, 123–133.

Afri, M., Gottlieb, H.E., Frimer, A.A., 2002. Superoxide organic chemistry within the

liposomal bilayer. Part II. A correlation between location and chemistry. Free

Radic. Biol. Med. 32, 605–618.

Asuncion-Punzalan, E., London, E., 1995. Control of the depth of molecules within

membranes by polar groups: determination of the location of anthracene-

Y. Cohen et al. / Chemistry and Physics of Lipids 155 (2008) 98–113

113

labeled probes in model membranes by parallax analysis of nitroxide-labeled

phospholipid induced ﬂuorescence quenching. Biochemistry 34, 11460–11466.

Balachander, N., Sukenik, C.N., 1990. Monolayer transformation by nucleophilic sub-

stitution: Applications to the creation of new monolayer assemblies. Langmuir

6, 1621–1627.

Barton, D., Ollis, W.D., 1979. Comprehensive Organic Chemistry: The Synthesis and

Reactions of Organic Compounds, vol 1, 1st ed., Pergamon, Oxford, England, pp.

819–820, Part 4, Ch. 4.3.

Janzen, E.J., Haire, D.L., Coulter, G.A., Stronks, H.J., Krygsman, P.H., Tower, R.A.,

Hilborn, J.W., 1989. Locating spin traps in heterogeneous media by ¹³C NMR

spectroscopy. Investigations in SDS micelles, DMPC vesicles, and rat liver micro-

somes. J. Org. Chem. 54, 2915–2920.

Maciel, G.H., Natterstad, J.J., 1965. Carbon-13 chemical shifts of the carbonyl group.

III. Solvent effects. J. Chem. Phys. 42, 2752–2759.

Maciel, G.H., Ruben, Z.C., 1963. Solvent effects on the ¹³C chemical shift of the car-

bonyl group of acetone. J. Am. Chem. Soc. 85, 3903–3904.

Bordwell, F.G., Wellman, K.M., 1963. Dehydrobromination of cis-2,6-dibromo-4,4-

dimethylcyclohexanone and of cis-2,6-dibromo-4,4-diphenylcyclohexanone. J.

Org. Chem. 28, 2544–2550.

Menger, F.M., 1979. On the structure of micelles. Acc. Chem. Res. 12, 111–117.

Menger, F.M., Aikens, P., Wood Jr., M., 1988. The water content of phospholipid

bilayers. J. Chem. Soc., Chem. Commun. 1988, 180–182.

Bronshtein, I., Afri, M., Weitman, H., Frimer, A.A., Smith, K.M., Ehrenberg, B., 2004.

Porphyrin depth in lipid bilayers as determined by iodide and parallax ﬂuores-

cence quenching methods and its effect on photosensitizing efﬁciency. Biophys.

J. 87, 1155–1164.

Chaki, H., Aikawa, Y., Miyajima, M., Nishio, M., Kuroda, H., Tanaka, K., Hirono, S.,

Shiozawa, S., 2002. Spiro thiazolidine compounds or salts thereof and preven-

tives/remedies for autoimmune diseases and AP-1 inhibitors containing the

same, U.S. Patent No. 6,384,065.

Menger, F.M., Jercunica, J.M., Johnson, J.C., 1978. The water content of a micelle

interior. The Fjord vs. Reef models. J. Am. Chem. Soc. 100, 4676–4678.

Moeller, H., Wallat, S., 1986. Hydroxybenzoates, Patent No. US4684665.

Pisula, W., Kastler, M., Wasserfallen, D., Pakula, T., Mullen, K., 2004. Exceptionally

long-range self-assembly of hexa-peri-hexabenzocoronene with dove-tailed

alkyl substituents. J. Am. Chem. Soc. 126, 8074–8075.

Reichardt, C., 1965. Empirical parameters of the polarity of solvents. Angew. Chem.

Int. Edit. Eng. 4, 29–40.

Chan, Y., Epstein, W.W., 1973. 4-Dimethyl-2-cyclohexen-1-one. Org. Synth. 53,

48–52.

Reichardt, C., 1988. Solvents and Solvent Effects in Organic Chemistry, 2nd ed. VCH

Verlagsgesellschaft, Weinheim (Chap. 7).

Chattopadhyay, A., London, E., 1987. Parallax method for direct measurement of

membrane penetration depth utilizing ﬂuorescence quenching by spin-labeled

phospholipids. Biochemistry 26, 39–45.

Cohen, Y., 2005. Locating substrates intercalated within the liposomal phospholipid

bilayer, Ph.D. Thesis, Bar Ilan University, Ramat Gan, Israel.

Cohen, Y., Afri, M., Frimer, A.A., 2008a. NMR-based molecular ruler for determining

the depth of intercalants within the lipid bilayer. Part II. The preparation of a

molecular ruler. Chem. Phys. Lipids 155, 114–119.

Cohen, Y., Afri, M., Frimer, A.A., 2008b. Aggregate formation in the intercalation of

long chain fatty acid esters into liposomes. Chem. Phys. Lipids 155, 120–125.

Corey, E.J., Suggs, J.W., 1975. Pyridinium chlorochromate. An efﬁcient reagent for oxi-

dation of primary and secondary alcohols to carbonyl compounds. Tetrahedron

Lett. 16, 2647–2650.

Dauben, W.G., Shaffer, G.W., Vietmeyer, N.D., 1968. Alkyl-substitution effects in the

photochemistry of 2-cyclohexenenones. J. Org. Chem. 33, 4060–4069.

Dimroth, K., Reichardt, C., Siepmann, T., Bohlmann, F., 1963. Pyridinium N-

phenolbetaines and their use for the characterization of the polarity of solvents.

Justus Liebigs Ann. Chem. 661, 1–37.

Dreiding, A.S., 1957. Eine desaromatisierungsreaktion. synthese von spiro-

[5:5]-undecadien-(1,4)-on-(3). vorläuﬁge mitteilung. Helv. Chim. Acta 40,

1812–1814.

Reichardt, C., 1994. Solvatochromic dyes as solvent polarity indicators. Chem. Rev.

94, 2319–2358.

Russell, G.A., 1973. Free Radicals, I. In: Kochi, J.K. (ed.), Wiley Interscience, New York,

pp. 275–331, at p. 296 ff.

Seto N., Yoshioka Y., Morigaki M., 1994. Silver halide color photographic material

with excellent lightfastness, Patent No. US 5418122.

Snider, B.B., Vo Huu, N., Foxman, B.M., 1993. Mn(III)-based oxidative fragmentation–

cyclization reactions of unsaturated cyclobutanols. J. Org. Chem. 58, 7228–7237.

Steel, W.H., Damkaci, F., Nolan, R., Walker, R.A., 2002. Molecular rulers: a new

family of surfactants for measuring interfacial widths. J. Am. Chem. Soc. 124,

4824–4831.

Strul, G., Frimer, A.A., Weiner, L., 1993. Spin-trapping study of free radical penetration

into liposomal membrane. J. Chem. Soc., Perkin 2, 2057–2059.

Strul, G., Gottlieb, H.E., Frimer, A.A., Weiner, L., 1994. Determining the loca-

tion of hydrophobic spin-traps within liposomes. J. Chem. Soc., Perkin 2,

1229–1231.

Taber, D.F., Mack, J.F., Rheingold, A.L., Geib, S.J., 1989. Enantioselective Robin-

son annulation: synthesis of (+)-O-methyljoubertiamine. J. Org. Chem. 54,

3831–3836.

Ueji, S., Makamaura, M., 1976. The effects of solvent on carbon-13 chemical shifts of

the carbonyl system. Tetrahedron Lett., 2549–2552.

Ellinger, S., Ziener, U., Thewalt, U., Landfester, K., Moller, M., 2007. Synthesis and

self-organization of ␣,∞-substituted oligothiophenes with long, branched alkyl

substituents. Chem. Mater. 19, 1070–1075.

Fieser, L.F., 1957. In Experiments in Organic Chemistry, 3rd ed. D.C. Heath and Com-

pany Publication, Boston, pp. 77–78, Revised, Chap. 13.

Frimer, A.A., 1985. The semipermeability of biological membranes. J. Chem. Ed. 62,

89–90.

Frimer, A.A., Forman, A., Borg, D.B., 1983. H₂O₂-diffusion through liposomes. Israel

J. Chem. 23, 442–445.

Frimer, A.A., Gilinsky-Sharon, P., Aljadeff, G., Gottlieb, H.E., Hameiri-Buch, J., Marks,

V., Philosof, R., Rosental, Z., 1989. Superoxide anion radical (O₂^•−) mediated base

catalyzed autoxidation of enones. J. Org. Chem. 54, 4853–4866.

Frimer, A.A., Strul, G., Buch, J., Gottlieb, H.E., 1996. Can superoxide organic chemistry

be observed within the liposomal bilayer? Free Radic. Biol. Med. 20, 843–852.

Gaertner, Van R., 1965. Sucrose ethers. Patent No. US3170915.

Gamliel, A., 2005. Spin-trapping with new lipophilic spin-traps, Ph.D. Thesis. Bar

Ilan University, Ramat Gan, Israel.

Gamliel, A., Afri, M., Frimer, A.A., 2008. Determining radical penetration of lipid

bilayers with new lipophilic spin traps. Free Radic. Biol. Med. 44, 1394–1405.

Gottlieb, H.E., Kotlyar, V., Nudelman, A., 1997. NMR chemical shifts of common lab-

oratory solvents as trace impurities. J. Org. Chem. 62, 7512–7515.

Gramlich, W., 1979. ¹³C NMR-Untersuchungen an 2-cyclohexenonen und 2,5-

cyclohexadienonen. Justus Liebigs Ann. Chem., 121–132.

Walker, R.A., Steel, W.H., 2003. Measuring liquid/liquid interfacial width with molec-

ular rule. Nature 424, 296–299.

Weitman, H., Roslaniec, M., Frimer, A.A., Afri, M., Freeman, D., Mazur, Y., Ehrenberg,

B., 2001. Solvatochromic effects in the electronic absorption and NMR spectra

of hypericin in organic solvents and in lipid bilayers. Photochem. Photobiol. 73,

110–118.

Williamson, A.W., 1852. On etheriﬁcation. J. Chem. Soc. 106, 229–239.

Woo, E.P., Inbasekaran, M., Shiang, W., Roof, G.R., 1997. 9-Substituted ﬂu-

orenes and 9-substituted ﬂuorene oligomers and polymers, Patent No.

WO9705184.

Zachariasse, K.A., Van Phuc, N., Kozankiewicz, B., 1981. Investigation of

micelles, microemulsions, and phospholipid bilayers with the pyridinium N-

phenolbetaine E_T(30), a polarity probe for aqueous interfaces. J. Phys. Chem. 85,

2676–2683.

Zhi-Kuan, C., Hong, M., Yee-Hing, L., Wei, H., 1999. Photoluminescent poly(p-

phenylenevinylene)s with an aromatic oxadiazole moiety as the side chain:

synthesis, electrochemistry, and spectroscopy study. Macromolecules 32,

4351–4358.

Zimmerman, H., Binkley, R.W., McCullough, J.J., Zimmerman, G.A., 1967. Organic pho-

tochemistry. XXVII. Electronically excited state structures. J. Am. Chem. Soc. 89,

6589–6595.

Zimmerman, H., Schuster, D.I., 1962. A new approach to mechanistic organic photo-

chemistry. IV. Photochemical rearrangements of 4,4-diphenylcyclohexadienone.

J. Am. Chem. Soc. 84, 4527–4540.

Zimmerman, H.E., Hackett, R., Juers, D.F., McCall, J.M., Schoder, B., 1971. Competi-

tive photochemical pathways in the di-ꢁ-methane rearrangement. Exploratory

and mechanistic organic photochemistry. LXIII. J. Am. Chem. Soc. 93,

3653–3658.

Hong-bin, Z., Jing-heng, N., Jian, C., Jin-hue, Z., Ning, Z., Sheng-ming, P., 2001. Study

on synthesis of a series of alkoxybenzaldehydes. Huaxue Yanjiu 12, 19–22.

Hu, J., Mattern, J.L., 2000. Ferrocenyl derivatives with one, two, or three sulfur-

containing arms for self-assembled monolayer formation. J. Org. Chem. 65,

2277–2281.

Article Doi

NMR-based molecular ruler for determining the depth of intercalants within the lipid bilayer. Part I. Discovering the guidelines

DOI: 10.1016/j.chemphyslip.2008.07.004

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.chemphyslip.2008.07.004

Full text of DOI:10.1016/j.chemphyslip.2008.07.004

Products guided by the article

R&D Labs maybe for 1097948-47-4

Relevant to this article

Hot Product