Synthesis of Isotopically Labelled Disparlure Enantiomers and Application to the Study of Enantiomer Discrimination in Gypsy Moth Pheromone-Binding Proteins

DOI: 10.1002/ejoc.201901164

Full Paper

Enantioselective Recognition

Synthesis of Isotopically Labelled Disparlure Enantiomers and

Application to the Study of Enantiomer Discrimination in Gypsy

Moth Pheromone-Binding Proteins

Govardhana R. Pinnelli,^[a]Mailyn Terrado,^[a]N. Kirk Hillier,^[b]David R. Lance,^[c]and

Erika Plettner*^[a]

Abstract: To study the binding mechanism of disparlure (7,8)- disparlure enantiomers in ²H NMR show that the binding of

epoxy-2-methyloctadecane enantiomers with pheromone-bind- disparlure enantiomers to PBP1 differs from binding to PBP2, as

ing proteins (PBPs) of the gypsy moth, oxygen-17 or 18 and expected from their opposite binding preferences (PBP1 binds

5,5,6,6-deuterium labelled disparlure enantiomers were pre- (–)-disparlure, and PBP2 binds (+)-disparlure more strongly).

pared in an efficient, enantioselective route. Key steps involve Models of the disparlure enantiomers bound to one internal

the asymmetric α-chlorination of dodecanal by SOMO catalysis binding site and two external binding sites of both PBPs were

and Mitsunobu inversion of a 1,2-chlorohydrin. The pheromone, constructed. The observed chemical shift changes of deuter-

(+)-disparlure (7R, 8S), was tested in two infested zones, demon- ated ligand signals, from non-bound to bound, T₁and T₂values

strating that it is very attractive towards male gypsy moths. are correlated with results from the simulations. Together these

Studies of the binding of (+)-disparlure and its antipode to results suggest that the disparlure enantiomers adopt distinct

gypsy moth PBPs by ²H &¹⁷O NMR at 600 MHz are reported. conformations within the binding sites of the two PBPs and

Chemical shifts, spin-lattice relaxation decay times (T₁) and interact with residues that line the sites.

transverse relaxation decay times (T₂) of deuterium atoms of

Introduction

ing mechanism of pheromone with PBPs offers some unan-

swered questions. Our group proposed previously a two-step

association mechanism, based on kinetic studies with ligand

and PBPs. The kinetic data suggested that a ligand will interact

with the protein on an external site before entering the internal

The gypsy moth, Lymantria dispar, is a harmful pest of a large

variety of trees (oak, apple, maple, mountain ash, pine, and

spruce), causing serious forest losses during outbreaks in Eu-

rope, Asia and North America. (+)-Disparlure (+)-1 (Figure 1),

(7R, 8S)-epoxy-2-methyloctadecane is the sex attractant phe-

romone, produced and released by female gypsy moths and

detected by the male moths. Initially, disparlure was identified

as cis-(7, 8)-epoxy-2-methyloctadecane by Bierl and co-workers

in 1970.^[1]Later, Iwaki et al. demonstrated the absolute stereo-

chemistry of disparlure by synthesizing both enantiomers, test-

ing them in the field, and reporting that (+)-disparlure (+)-1 is

more active than (–)-disparlure (–)-1 (Figure 1).^[2]It has been

shown that the highest number of moths can be caught in

traps when the most enantiopure (+)-disparlure (> 99 % ee) is

used.^[2]Studies have shown that the disparlure enantiomers

bind strongly to pheromone binding proteins 1 and 2 (PBP1 &

PBP2), with opposite enantioselectivity.^[3,4]However, the bind-

[a] Dept. of Chemistry, Simon Fraser University, 8888 University Dr., Burnaby

B. C. V5A 1S6, Canada

E-mail: plettner@sfu.ca

[b] Dept. of Biology, Acadia University,

33 Westwood Ave., Wolfville, NS B4P 2R6, Canada

[c] USDA-APHIS-PPQ CPHST Otis Laboratory,

1398 W Truck Rd, Buzzards Bay, MA 02542, USA

Supporting information and ORCID(s) from the author(s) for this article are

available on the WWW under https://doi.org/10.1002/ejoc .201901164.

Figure 1. Structures of disparlure enantiomers and (+)-monachalure, prepared

during optimization of the (+)-disparlure synthesis.

Eur. J. Org. Chem. 0000, 0–0

1

Full Paper

binding pocket.^[3]Once the ligand reaches the internal pocket in their ligand binding and dissociation kinetics: PBP2 binds

we know that the two gypsy moth PBPs differ in their affinity ligands at its internal binding site very slowly,^[4]whereas PBP1

for the disparlure enantiomers, but there only are in-silico dock- has much faster association and dissociation kinetics. The inter-

ing models of the PBP-bound pheromones.^[5,6]The challenge of action and discrimination of these two PBPs towards disparlure

binding studies of PBP-ligand interactions using already known enantiomers are not completely understood due to lack of crys-

assays is to probe the equilibrating PBP-ligand mixture without tal or NMR-based structures.

excessively disturbing it and to detect the bound ligand.

Here we describe a short synthesis of enantiopure, labelled

and non-labelled disparlure, followed by experiments in which

we studied the kinetics of non-labelled disparlure enantiomer

association and dissociation with PBP1, the binding of ¹⁷O-dis-

parlure to PBP1 and of ¹⁸O/5,5,6,6-deuterium labelled disparlure

enantiomers to both PBPs. With the latter, two types of analysis

were done on the data: 1) chemical shift changes between dis-

parlure in buffer vs. bound to PBP were correlated with the

2

Therefore, we aimed to prepare ¹⁷O or ¹⁸O & H labelled (+)-

disparlure (+)-1b or (+)-1c and (–)-disparlure (–)-1b or (–)-1c

(Figure 1) and study their binding orientation with pheromone

binding proteins using ¹⁷O and ²H NMR spectroscopy. We corre-

lated results from ²H NMR experiments with in-silico docking

models at one internal and two external binding sites of PBP1

and PBP2.

To date, more than twenty synthetic routes to synthesize (+)- location of the deuterium labels in PBP-ligand structures ob-

disparlure have been reported.^[7]These methods are based on tained through docking simulations, and 2) the measurement of

either asymmetric synthesis or chiral pool precursors. Most of longitudinal relaxation (T₁) and transverse relaxation (T₂) times.

the synthetic routes have utilized Sharpless asymmetric epoxid- Docking simulations of both enantiomers were done with

ation^[8]and dihydroxylation,^[9,7d,7e]whereas other synthetic homology models of L. dispar PBP1 and PBP2, at one internal

routes used chiral stannanes,^[10]enantiopure sulfoxides,^[11]and two external binding sites. These were used to interpret

asymmetric chloroallylboration,^[12a]asymmetric MacMillan's the position of disparlure ligands relative to aromatic residues

self-aldol reaction^[12b]and chiral pool starting materials such as in the protein and the level of mobility of the labeled positions

L

-glutamic acid,^[13]-tartaric acid^[14]and -glucose-δ-lactone.^[7m]of disparlure under various conditions (in organic solvent, in

L D

The drawbacks of previously reported approaches include the buffer and bound to the PBPs). We then correlated the simula-

use of uncommon starting materials, low enantiomeric excess tions with the observed NMR parameters. We detected differen-

of the final (+)-disparlure, many steps and low overall yield. For ces between the enantiomers of disparlure interacting with the

example, disparlure currently used for monitoring programs in two proteins. This is discussed in the context of differences de-

the field is prepared by Sharpless epoxidation of (Z)-tridec-2- tected in other assays.

en-1-ol, to access the chiral epoxide ring, and Wittig olefination

of the epoxy aldehyde to install the branched alkyl side chain.

Typically, the overall yield of this route is 33 % from non-com-

Results and Discussion

mercially available starting material and enantiomeric excess

values range from 91–95 %.^[8b]Here we describe a route to (+)-

Synthesis

disparlure that is one step shorter (counting from commercial

Herein we report an efficient total synthesis of (+)-disparlure,

starting material), has an overall yield in the range of 42–53 %

(–)-disparlure, isotope labelled disparlure enantiomers and

and an enantiomeric excess of > 99 %. Apart from that, this

(+)-monachalure (7R, 8S)-epoxyoctadecane, the pheromone of

the nun moth, Lymantria monacha (Figure 1), in five steps. This

work has resulted in preparation of optically pure (+)-disparlure

route provides easy access to isotope (¹⁷O, ¹⁸O and ²H) labelled

disparlure enantiomers for studies with pheromone binding

proteins (PBPs) and degrading enzymes that are present in the

in large quantities (>1 g). The (+)-disparlure was tested in two

male gypsy moth antennae.

infested zones, demonstrating that it is very attractive towards

male gypsy moths (see below).

We explored the use of MacMillan's procedure of SOMO-acti-

Disparlure Enantiomer and PBP Interaction

vated α-chlorination of an aldehyde to prepare enantiopure α-

chloroaldehyde^[20,21]Chlorination of dodecanal 2 by LiCl as the

The remarkable enantiomer discrimination between (+)-dispar-

chlorine source, in the presence of an imidazolidinone catalyst

afforded the enantiopure α-chlorododecanal (+)-3 (Scheme 1)

in 87 % yield. The enantiopurity of α-chlorododecanal was de-

termined to be >99 % by chiral HPLC (see Supplementary Infor-

mation).

lure and (–)-disparlure is governed by the selectivity and sensi-

tivity of neurons of the olfactory receptors, which are located

in the sensory hairs of male moth antennae.^[3,4,15]Sensory hairs

are hollow, and the interior space of the hairs is filled with

sensillar lymph, which contains the abundant, soluble PBPs.^[16]

The sensory hairs also house the dendrites of 2–3 chemosen-

With the highly enantiopure α-chloroaldehyde in hand, we

sory neurons, which are responsible for detecting any odorants next turned our attention to the diastereoselective addition of

that enter the lymph via pores on the sensillar cuticle.^[17]The organometallic reagents.^[22–25]For the optimization of this step

PBPs are believed to control the diffusion of pheromones and we used the non-branched building block that results in (+)-

other odors to olfactory neurons in the dendritic membrane. monachalure (+)-1a (Scheme 1). The addition of either n-hexyl-

The gypsy moth has two known PBPs: LdisPBP1 and LdisPBP2, lithium or Grignard reagent (hexylmagnesium bromide) to

and these proteins are known to have opposite affinities for α-chloroaldehyde (+)-3 afforded chlorohydrins in good yield

(+)- and (–)-disparlure.^[18,19]Furthermore, the two proteins differ but in moderate diastereomeric excess (Table 1, entries 3 & 4).

Eur. J. Org. Chem. 0000, 0–0

www.eurjoc.org

2

Full Paper

Scheme 1. The synthetic route to (+)-monachalure, to non-labelled and isotope labelled (+)-disparlure developed here.

A brief survey of solvents (Et₂O, THP) failed to significantly im- reaction, but the reaction was unsuccessful, and the starting

prove diastereoselectivity of carbon nucleophile addition. Fi- material was recovered.

nally, we treated α-chloroaldehyde (+)-3 with the alkyne lithium

anion derived from 1-hexyne or 5-methyl 1-hexyne: the reac-

tion proceeded cleanly and the product 1,2-anti-chlorohydrin

(+)-4 was isolated in good yield with a 20:1 diastereoselectivity.

At this point, the diastereomers were separated by flash column

chromatography.

In order to prepare 1,2-syn product, the stereochemistry at

the C-8 carbon of compound (+)-4 was inverted by a Mitsunobu

reaction. We treated compound (+)-4 with TPP (triphenylphos-

phane) DIAD (diisopropyl azadicarboxylate) and benzoic acid

in dry THF under an inert atmosphere. This afforded complete

inversion product (+)-5, which was isolated in 80 % yield. Hyd-

rogenation of compound (+)-5 with H₂and catalytic Pd/C in

methanol afforded reduced product in 95 % yield, which was

treated with 4 N NaOH solution in methanol, to produce the

final product (+)-disparlure (+)-1 in 90 % yield. The spectro-

scopic data (¹H NMR, ¹³C NMR, IR, [α]_Dand MS) of (+)-1 are

identical to those reported in the literature.^[2,7b]

Table 1. Addition of organometallic reagents to enantiopure α-chloroalde-

hyde (+)-3.

For the synthesis of (–)-disparlure (Scheme S1, Supporting

Information), D-SOMO (2R, 5S)-2-(tert-butyl)-3,5-dimethylimid-

azolidin-4-one) was used to prepare enantiopure α-chloro-

dodecanal (–)-3. Diastereoselective addition of (–)-3 with lith-

ium alkyne anion resulted in (–)-4 in 85 % yield. Reaction of

(–)-4 with benzoic acid and DIAD produced the ester, which

was hydrogenated with H₂and Pd/C, then treated with NaOH

to close the epoxide and furnish (–)-disparlure (–)-1 in 90 %

yield.

[a] Ratio of diastereomers determined by ¹H NMR analysis of crude product.

If the 1,2-anti chlorohydrin (+)-4 (Scheme 1) is treated with

base, then a trans epoxide is formed. However, we require a

1,2-syn product to obtain the cis epoxide. Therefore, we need

to invert the stereochemistry either at the C-8 or C-9 position

Synthesis of ¹⁷O or ¹⁸O enriched compounds usually involves

of the 1,2-anti-chlorohydrin (+)-4. An attempt was made to in- water, carbon monoxide and [^17/18O] enriched starting materi-

vert the stereochemistry at the C-9 position by a Finkelstein als,^[26,27]which are expensive. The most convenient precursor

Eur. J. Org. Chem. 0000, 0–0

www.eurjoc.org

3

Full Paper

to prepare ^17/18O enriched compounds is labelled water, which

is commercially available with various enrichment levels start-

ing from 10 to 97 %. To prepare isotope labelled disparlure

enantiomers, we obtained ¹⁷O or ¹⁸O labelled benzoic acid by

hydrolysis of benzoyl chloride with ¹⁷O or ¹⁸O water (20 atom-

Table 2. Oxygen and hydrogen isotope enrichment of (+)-and (–)-disparlure.

Isotope enrichment [%]

Entry

Compound

¹⁷O

¹⁸O

²H (C-5 & 6)

1

2

3

4

(+)-1b

(+)-1c

(–)-1b

(–)-1c

8–10

> 99

50

%

¹⁷O or 97 atom-% ¹⁸O). Further, the ¹⁷O or ¹⁸O labelled inter-

8–10

mediates (+)-5b, (+)-5c, (–)-5b and (–)-5c (Scheme 1) were pre-

pared by Mitsunobu reaction of intermediates (+)-4 and (–)-4

with ¹⁷O or ¹⁸O labelled benzoic acid. The labelled intermedi-

ates were obtained as liquids, which contained approximately

50 % ¹⁸O and 15–20 % ¹⁷O at either one of the benzoate oxy-

gens. Through the subsequent deuteration of (+)-5b, (+)-5c,

(–)-5b and (–)-5c with D₂in the presence of Wilkinson's catalyst,

the products (+)-6b, (+)-6c, (–)-6b and (–)-6c were obtained.

Finally, the target compounds (+)-1b, (+)-1c, (–)-1b and (–)-1c

were obtained by basic epoxide ring closure of (+)-6b, (+)-6c,

(–)-6b and (–)-6c with NaOH. The percentages of ¹⁷O or ¹⁸O at

the epoxide oxygen of the target compounds were calculated

from the area of the shifted carbon signal in their ¹³C NMR

spectra and from HR-MS (Figure 2 and Table 2). The ¹⁷O NMR

spectrum was obtained for the ¹⁷O labelled (+)-disparlure (Fig-

ure 2d). In ¹⁸O labelled (+)-disparlure, the signals for the oxirane

ring carbon atoms were seen upfield from those of the non-

labelled variant due to the isotope effect on the chemical shift.

Determination of Enantiopurity

The common method for finding enantiopurity of chiral organic

molecules is the conversion of enantiomers to diastereomers

by introducing a second defined chiral center.^[28]To optimize

the method used to determine enantiomeric excess of (+)-dis-

parlure, we converted a scalemic sample of disparlure 7 (Sup-

porting Information, Scheme S3) to the corresponding cis-N-

(α-methylbenzyl)aziridines with inversion of configuration at

both, the 7 and 8 positions.^[29]The cis epoxide of scalemic dis-

parlure was opened with R-(+)-α-methylbenzylamine in the

presence of trimethylaluminium, which produced a mixture of

amino alcohols. The amino alcohols were treated with meth-

anesulfonyl chloride and triethylamine, to afford a diastereo-

meric mixture of cis-N-(α-methylbenzyl) aziridines 16 and 17.

Then the diastereomeric aziridines were resolved by gas chro-

matography, and we observed nearly baseline separation of the

peaks. The diastereomers of cis-N-(α-methylbenzyl)aziridine 16

and 17 were observed at 53.05 and 53.40 min, respectively.

Similarly, (+) and (–)-disparlure (+)-1 and (–)-1 were converted

to corresponding cis-N-(α-methylbenzyl)aziridine and found to

be in enantiomeric excess (> 99 % ) (see Supporting Informa-

tion, Scheme S4 and Scheme S5, Figure S2).

Field Test of (+)-1

(+)-Disparlure (+)-1 prepared herein was tested in two infested

zones, located in Massachusetts (USA) and Nova Scotia (Ca-

nada), to demonstrate activity against male gypsy moths. We

used delta sticky traps baited with either white rubber septa

(Nova Scotia) or milk carton traps baited with impregnated den-

tal cotton (Massachusetts), in both cases dosed with 500 μg of

(+)-disparlure or 500 μg of paraffin oil in hexanes as a control.

In Massachusetts, milk carton traps with cotton wick dispensers

were used and left for 4 days. Three different doses (100, 500

and 1000 μg) were tested against a control with mineral oil.

The 100 and 1000 μg treatments were from a commercial stan-

dard, and the 500 μg dose was from this study. The control did

Figure 2. ¹³C NMR signals of epoxy carbons of a) (+)-disparlure, b) ¹⁸O (+)-

disparlure c) ¹⁷O (+)-disparlure, and d) ¹⁷O NMR spectrum of ¹⁷O (+)-dispar-

lure in CDCl₃.

Table 3. Summary of field trails.

Location

Trap

type

Dispenser

Treatment

Dose

[μg]

Moths caught

(mean SEM,

n = 10)

Massachusetts

Milk carton

Delta

Cotton wick

Control (mineral oil)

500

100

500

1000

500

0

(+)-disparlure commercial standard

(+)-disparlure (this study)

(+)-disparlure commercial standard

Control (mineral oil)

52 15

70

82 17

0.9 0.4

22

8

Nova Scotia

Rubber septum

(+)-disparlure (this study)

2

Eur. J. Org. Chem. 0000, 0–0

www.eurjoc.org

4

Full Paper

not catch any moths, and the pheromone-baited traps caught acid salts as emulsifier and a bridging solvent (acetonitrile). We

the same number of moths within limits of error (F = 2.27; d.f. = chose to use fatty acid micelles to disperse disparlure in the

2, 8; P = 0.17) (Table 3). Significant contamination with (–)-dis- aqueous buffer, because previous studies from our laboratory

parlure (due to insufficiently high ee) on a 500 μg wick would

normally suppress trap catch to levels significantly below those

obtained with highly pure (+)-disparlure at either 100 μg or

1000 μg loadings.^[41]In Nova Scotia, red delta traps with rubber

septum dispensers were used and left for 10 days. Three of the

treatment traps were lost due to predation. Nonetheless, the

treatment traps caught significantly more moths than the con-

trol (Table 3). Observation on the first day suggested that traps

saturated with moths within a matter of hours. The traps at-

tracted moths within minutes of being deployed. Traps in the

two locations caught different total numbers of moths due to

two factors: 1) different infestation levels and 2) different dis-

pensers and traps.

have shown that gypsy moth sensilla contain high quantities of

fatty acids in the lymph.^[30]The result suggests that disparlure

mobility would have decreased significantly in the fatty acid

micelles and, therefore, the ¹⁷O signal is so broad that it cannot

be detected.

Because ¹⁷O was difficult to detect, we decided to use ²H

labelled disparlure enantiomers (+)-1c and (–)-1c (Scheme 1

and Scheme S1) in our binding studies. We used 600 MHz deu-

terium NMR to observe binding-induced chemical shift change

of the deuterium signals of the deuterated disparlure enantio-

mers. The ²H chemical shift of the deuterium labelled disparlure

enantiomers (+)-1c or (–)-1c resonance in the presence and ab-

sence of LdisPBPs was monitored. It was possible to distinguish

the more shielded 5-D signals from the 6-D signals (Figure 3).

We have named these signals 5-D_a, 5-D_b, 6-D_aand 6-D_b. In a

solvent (CDCl₃, toluene or phosphate buffer) the majority of

disparlure molecules will adopt a conformation around the ep-

oxide at or near the global energy minimum.^[5]In that confor-

mation, (+)-1c has the 5-pro-R H atom in the shielding zone

above the epoxide ring and the 5-pro-S H atom in the deshield-

Heteronuclear NMR Studies of Disparlure Binding to PBPs

The binding of disparlure enantiomers to pheromone binding

proteins of male gypsy moth was studied by NMR spectroscopy.

First, we studied binding of ¹⁷O disparlure to LdisPBP1, as well

as ¹⁷O disparlure by itself in CDCl₃(Figure 2d) and in buffer.

The ¹⁷O signal was broad, as expected, in CDCl₃, but completely ing zone at the edge of the epoxide ring, along the C-O

disappeared when disparlure was placed in buffer with fatty bond.^[31]In our data, 5-D_ais more shielded than 5-D_b(Table 4),

Figure 3. The 600 MHz ²H NMR spectra recorded for deuterium labelled disparlure enantiomers with and without LdisPBPs in phosphate buffer at pH 8. The

chemical shift scales of the above spectra are aligned on the basis of the deuterium resonance (2.10 ppm) of internal standard CH₃CN. a) The region of the

²H NMR spectra of deuterium labelled disparlure enantiomers and acetonitrile resonances are shown in the absence of LdisPBPs, b) & c) downfield shifting of

the deuterium signals of the disparlure enantiomers by LdisPBP1 and LdisPBP2, respectively. d) superimposed spectra of a), b) and c). The same buffer (50 m

M

sodium phosphate pH 8.0) was used for all these spectra.

Eur. J. Org. Chem. 0000, 0–0

www.eurjoc.org

5

Full Paper

Table 4. The ²H chemical shifts of 5,6-D₄(+)-1c and (–)-1c in the presence and absence of LdisPBPs.

Compound

Position signal &

assignment^[a]

δ CDCl₃

δ Toluene

δ Phosphate buffer

(pH 8)

δ Phosphate buffer

(pH 8) + LdisPBP1^[b]

δ Phosphate buffer

(pH 8) + LdisPBP2^[b]

1

(+)-1c

5-D_a

pro-R

pro-S

pro-R

pro-S

pro-R

pro-S

0.8

0.94

1.26

1.46

1.56

0.94

1.26

1.46

1.56

0.73

0.98

1.16

1.38

0.73

0.98

1.16

1.38

0.96 (pro-R)

1.30 (pro-S)

1.45 (pro-S)

1.52 (pro-R)

0.91 (pro-R)

1.21 (pro-S)

1.43 (pro-R)

1.49 (pro-S)

0.89 (pro-S)

1.17 (pro-R)

1.41 (pro-S)

1.45 (pro-R)

0.85 (pro-R)

1.17 (pro-S)

1.37 (pro-S)

1.41 (pro-R)

1

5-D_b

6-D_a

1.07

1.25

1.34

0.8

1.07

1.25

1.34

1

6-D_b

1

(–)-1c

5-D_a

1

5-D_b

1

6-D_a

1

6-D_b

[a] Assignments of the 5 and 6 deuterium atom signals for disparlure in solvent, based on the global energy minimum around the epoxide found in ab initio

calculations in^[3]and the anisotropic effect of the oxirane ring.^[31][b] Chemical shifts and assignment, based on averaged protein-bound conformations of

the disparlure enantiomers, chemical shift differences between protein-bound and solvent (Table 5) and anisotropic effects of the oxirane and the proteins

(see Figure 5).

so 5-D_amust correspond to the pro-R D atom and 5-D_bmust field (Figure 3b and Figure 3c, Table 4, Table 5 and Table 6). The

be the pro-S. For (–)-1c, the 5-pro-S H atom is in the epoxide ²H chemical shift changes of the bound (+)-1c and (–)-1c no-

shielding zone (above the oxirane ring) and the 5-pro-R H is in ticed were typically of the order of 0.07 to 0.29 ppm (Table 6).

the deshielding zone at the global minimum.^[5]Therefore, for The deuterium signals of (+)-1c were more deshielded than sig-

this compound, 5-D_amust correspond to the pro-S D, whereas nals of (–)-1c by LdisPBP1 and LdisPBP2. The apparent bound

5-D_bmust correspond to the pro-R D. Using the same approach, chemical shift difference of the compounds (+)-1c and (–)-1c

we assign the deuterium atoms at the 6 position as 6-D_a= 6-D_Sdeuterium signals shows that the binding (and therefore local

and 6-D_b= 6-D_Rfor (+)-1, and 6-D_a= 6-D_Rand 6-D_b= 6-D_Svariations in the magnetic field) of disparlure enantiomers to

for (–)-1. In the chiral protein binding pockets this assignment LdisPBP1 is different from binding to LdisPBP2.

can change, because there are numerous anisotropic effects

(from aromatic side chains and carbonyl groups), apart from

Both PBPs have one internal binding site and multiple exter-

nal binding sites,^[5]which interact with a ligand in a stepwise

manner, fast external binding, followed by slow internaliza-

the anisotropic effect of the oxirane ring (see below).

2

The H NMR spectrum of the compounds (+)-1c and (–)-1c tion.^[4]The apparent equilibrium constant for the external sites,

in buffer, in the absence of LdisPBPs showed two peaks, but K_d′, is the ratio of the dissociation and association rate con-

these were not well resolved due to poor solubility of the com- stants, k_off/k_on, for the first external binding step (Figure 4).

pounds in the phosphate buffer solution (Figure 3a). The most From^[4]we had data for PBP2, but for PBP1 we lacked data

shielded 5-D signal was sharper than the 6-D and the second because this protein interacts with ligands much faster than

5-D signals. One explanation is that disparlure forms micelles in PBP2.^[18]Therefore, we used stopped-flow experiments to ob-

the aqueous environment and the mobility of the 6-position is tain k_offand k_onfor PBP1 and the disparlure enantiomers from

more restricted than that of the 5-position. Furthermore, the competitive kinetics between a fluorescent reporter ligand

hydrophobic disparlure molecule adsorbs on glass surfaces,^[30](1-phenylnaphthylamine, NPN) and (+)-1 or (–)-1. This informa-

and the adsorbed molecules will have very restricted mobility tion was required to process the in silico ligand docking data

and, therefore, broad signals. Addition of LdisPBPs causes an (see below). For the measurement of association, NPN bound

increase in solubility of the (+)-1c and (–)-1c in the phosphate to PBP is displaced with disparlure (which causes NPN fluores-

buffer solution,^[3,30]and deuterium resonances to shift down- cence to decrease). For measurement of disparlure dissociation,

Table 5. Chemical shift differences, relative to the CD₃CN signal, used as an internal standard and net shifts of (+)-1c and (–)-1c in buffer and PBP-bound.

Condition

δδ (CD₃CN-5-D_a), ppm

δδ (CD₃CN-5-D_b), ppm

δδ (CD₃CN-6-D_a), ppm

δδ (CD₃CN-6-D_b), ppm

CDCl₃

Toluene

1.3

1.03

0.84

1.12

0.85

0.93

0.89

0.93

0.85

0.64

0.94

0.65

0.69

0.67

0.73

0.76

0.54

0.72

0.58

0.65

0.61

0.69

1.16

1.37

1.14

1.21

1.19

1.25

Phosphate buffer

LdisPBP1 (+)-1c

LdisPBP2 (+)-1c

LdisPBP1 (–)-1c

LdisPBP2 (–)-1c

Table 6. Net shifts of (+)-1c and (–)-1c in buffer and PBP-bound.

Condition

δδ (LdisPBP1-buffer), ppm

δδ (LdisPBP2-buffer), ppm

5-D_a

5-D_b

6-D_a

6-D_b

5-D_a

5-D_b

6-D_a

6-D_b

LdisPBP + Buffer + (+)-1c

LdisPBP + Buffer + (–)-1c

0.23

0.18

0.32

0.2

0.29

0.27

0.14

0.11

0.16

0.12

0.19

0.25

0.21

0.1

0.07

Eur. J. Org. Chem. 0000, 0–0

www.eurjoc.org

6

Full Paper

Figure 4. Association and dissociation kinetics of PBPs and cognate ligands. a) Kinetic scheme for ligand-PBP interaction. The ligand first binds externally in a

rapid, reversible manner, and then slowly diffuses into the internal binding site of the PBP. b) Structure of NPN, the competitive fluorescent reporter ligand

used in PBP1-ligand association kinetics experiments in a stopped-flow apparatus. c) Stopped-flow kinetics with fluorescence reporting of PBP1 with either

(+)-1 or (–)-1 and NPN. Both association (k_on) and dissociation (k_off) rate constants can be obtained from non-linear fitting of the traces at different concentra-

tions of ligand and constant concentration of NPN reporter.

Table 7. Dissociation constants used in the calculation of ligand distribution between internally, externally bound and free states. Distributions between states

obtained for each PBP with both disparlure enantiomers.

K_d′^[b]

[μ

Ref. K_d

% bound

internal^[c]

% bound

external^[d]

% free

[a]

′

Protein

Ligand

K_d

[μ

M]

M

]

PBP1

PBP2

(+)-1

(–)-1

(+)-1

(–)-1

6.8

6.0

2.2

4.7

56.5

11.8

1.1

This work

48.9

49.0

49.6

49.2

43.8

49.1

50.2

50.5

7.3

1.9

0.2

0.3

This work

[4]

1.6

[a] From Terrado et al.^[33][b] Calculated as k_off/k_on, the ratio of association and dissociation rate constants. [c] [PBP]_total= 300 μ

there are two external binding sites, their combined concentration is 600 μ

M; [L]_total= 600 μM. [d] Since

M.

disparlure bound to PBP is displaced with NPN (which causes relative to those at the global minima. Specifically, for (–)-1

NPN fluorescence to increase. Both measurements can be done bound to PBP1,5-D_ais pro-R and 5-D_bis pro-S. For (+)-1 bound

simultaneously in experiments wherein NPN and disparlure to PBP2, 5D_ais pro-S and 5-D_bis pro-R. Finally, for (–)-1 bound

compete for the binding sites on the protein (Figure 4).

to PBP2,5-D_a= pro-R, 5-D_b= pro-S, 6-D_a= pro-S and 6-D_b= pro-

R. There are two factors which influence the chemical shifts of

the 5 and 6 protons of disparlure when bound to the PBPs: 1)

the conformation of the pheromone and 2) the shielding and

deshielding effects of nearby groups on the protein, particularly

carbonyl or phenyl groups. For example, Figure 6 shows the

most populated pose at the internal binding site for the four

cases studied here. The disparlure enantiomers are clearly not

bound at a minimum conformation calculated in.^[5]For the

PBP1 and (–)-1 case, one can see that 5-H_Ris in the weak shield-

ing zone near the oxirane, and 5-H_Sis in the deshielding zone

around the oxirane C-O bond, the reverse of the arrangement

in the global minimum. Also, 5-H_Sis in the deshielding zones

of Phe 119 and Trp 37, which reinforce the oxirane deshielding

effect.

To understand the chemical shift changes observed, we per-

formed in-silico docking simulations of (+)-1 and (–)-1 into ho-

mology models of PBP1 and PBP2. We docked both ligands into

the internal binding site (site 1) and the two largest external

sites (sites 2 and 3). The two external sites were assumed to

contribute randomly (i.e. equally). Structures that were calcu-

lated to be populated > 5 % were examined with regard to the

interactions of the focal hydrogen atoms, at the 5 and 6 posi-

tions of disparlure, with the protein. Anisotropic effects from

aromatic side chains and carbonyl groups within a range of

4.5 Å around the focal H atoms were scored (see experimental

section), along with the anisotropic effect from the oxirane ring.

Scores were weighted according to their proportion at the

docking site, and internal vs. external sites were weighted ac-

cording to their proportion (Table 7). The patterns obtained for

the pro-R and pro-S hydrogen atoms at each position allowed

us to assign signals in the PBP-ligand treatments (Figure 5).

We also determined the spin-lattice relaxation time, T₁, and

the transverse relaxation time T₂for the two 5-D signals and

the combined 6-D signals (Figure 7), to better understand how

As shown in Figure 5, the assignments for D_aand D_bwere the hydrogen atoms at the 5 and 6 positions of the disparlure

reversed in some cases of PBP-bound disparlure enantiomers, enantiomers interact locally with the two binding proteins.

Eur. J. Org. Chem. 0000, 0–0

www.eurjoc.org

7

Full Paper

Figure 6. Interactions at the internal binding site of the most populated pose

for each combination of protein and ligand. Top row: PBP1, bottom row:

PBP2. The disparlure chain is shown in ball-and-stick format, with the epoxide

colored red, 6-H_S= light green, 6-H_R= dark green, 5-H_Slight pink, 5-H_R

=

dark pink. Residues that have a part interacting with any of the focal

hydrogen atoms within a 4.5 Å radius are shown in space-filling format, with-

out their hydrogen atoms.

T₁reflects the longitudinal relaxation of the magnetic field in

the z direction to its equilibrium value. This process is influ-

enced by the local fluctuations in the magnetic field around the

nucleus being observed. The frequencies of these fluctuations

depend on the local density of the medium around the nucleus:

the greater the density and viscosity (i.e. the lower the mobil-

ity), the larger T₁.^[32]The transverse relaxation process reflects

the decay of the phasing of nuclei in the xy plane. It depends

on the fluctuations in the local magnetic field and on spin ex-

change between neighboring nuclei of opposite spin. In the gas

phase, the local fluctuations average and 2T₁≅ T₂. At increasing

viscosity in a condensed phase, the local magnetic fluctuations

do not average, spin exchange occurs and T₂decreases.^[32]

Figure 5. Left: weighted scores for shielding (< 0) or deshielding (> 0) effects,

from docking simulations, weighted for distribution of poses at the internal

and the two external sites and weighted for the distribution of the ligand

between external and internal sites as shown in Table 7. Right: chemical shift

differences (Δδ) between disparlure bound to PBP and disparlure in buffer.

The hatched bars correspond to the signal labelled D_aand the dark gray bars

correspond to the signal labelled D_b. Bars labelled “rev” exhibited a reversed

assignment of prochirality for the D_aand D_bsignals, relative to the assign-

ment in an isotropic medium (Table 4).

Figure 7. Determination of relaxation times T₁and T₂. a) Example of a T₁determination for 5-D_aof (–)-1c with Ldis PBP1 (phosphate buffer, pH 8) by following

the re-establishment of the magnetization after a 90° pulse (see Figure S9 for traces) b) Example of a T₂determination for 5D_aof (–)-1c with Ldis PBP1

(phosphate buffer, pH 8) by following the decay of the signal intensity after a 180° pulse. c) Example of signal deconvolution for 5-D_a, 5-D_band 6-D_a/b

.

Eur. J. Org. Chem. 0000, 0–0 www.eurjoc.org © 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

8

Full Paper

Thus, the two relaxation times give insight into the mobility is a measure of the local viscosity: the more interactions, the

around the focal nucleus. In our molecular models, we checked lower the mobility and the higher the local viscosity.

the number of interactions between the focal hydrogen atoms

and groups either on the protein or the twisted chain of dispar-

lure. All interactions within a 4 Å radius were counted, and an

estimate of the local mobility was obtained as the inverse of

that number of interactions. This was done for each retained

pose at each of the three sites, and mobility was averaged for

each site and across sites, as described in the experimental and

supplemental information. The inverse of the averaged mobility

Figure 7 shows an example of T₁and T₂determination for

5-D_aof (–)-1c in the presence of Ldis PBP1 (phosphate buffer,

pH 8). The T₁values for the four combinations of PBP and dis-

parlure enantiomers are shown in Figure 8a and Table 8. Be-

cause CDCl₃and buffer are non-chiral, isotropic media, we only

performed the measurements with one of the two enantiomers,

(–)-1c. It is interesting to note that both solvent systems gave

the same T₁values within limits of error. In buffer, the signals

were much weaker than in CDCl₃or in the protein-bound cases

(Figure 3), because in buffer disparlure has a strong tendency

to adsorb on the walls of the vial. Signal from adsorbed dispar-

lure is not detectable, since its mobility is completely restricted

and T₁is very long, T₂is very short, both outside of the range

of this experiment. Thus, in buffer, we detected only the dispar-

lure that was dissolved in the aqueous phase.

Comparing T₁values obtained for PBP-bound disparlure

enantiomers and the value obtained for the free disparlure in

solvent (Figure 8a) we see that, in general, there was an increase

in T₁upon binding of disparlure to the PBPs. This suggests that,

generally, the mobility of the focal hydrogen atoms decreased

upon binding to the protein. However, the signal intensity in-

creased relative to buffer (Figure 3), and this is because PBPs

help to desorb pheromones from solid surfaces, thus brining

the hydrophobic pheromone into solution.^[3,30]Exceptions to

the general increase in T₁upon protein binding were: 5-H_Sin

PBP2 with (+)-1c, 6-H_S/Rin PBP1 with (+)-1c and 6-H_S/Rin PBP2

with (–)-1c, all of which did not differ from the T₁value seen

in buffer. Thus, for these exceptional cases the local viscosity

around the focal hydrogen atom in the protein-bound ligand

was the same as that of free ligand.

The site-averaged inverted mobility patterns obtained from

modelling (Figure 8c) parallel the patterns seen for T₁with the

exception of 5-H_Sin PBP2 with (–)-1c, 5-H_Rin both PBP1 with

(–)-1c and PBP2 with (+)-1c. For 5-H_Sin PBP2 with (–)-1c the

inverted mobility (local viscosity) was lower than reflected in

the T₁. Interaction at the internal binding site between 5-H_Sand

the two methyl groups of the side chain of Val 94 (both within

van der Waals radius) contributes to restriction in the mobility

of 5-H_Smuch more than is estimated by simple counting of

interactions. Similarly, for 5-H_Rin PBP2 with (+)-1c, the methyl

groups of Val 94 or of Leu 52 clamp around that hydrogen atom

in the three most populated poses at the internal binding site.

Figure 8. Left: spin relaxation data from ²H NMR experiments with PBP1 and

PBP2 with (+)-1 or (–)-1 bound. Right: measurements obtained from docking

simulations of the ligands into one internal and two external sites on PBP1

and PBP2. Dark gray = PBP1 and (+)-1, stippled = PBP1 and (–)-1, medium

gray = PBP2 and (+)-1, hatched = PBP2 and (–)-1. a) Top: Spin lattice relaxa-

tion time,T₁, in ms. Bars represent the average value from fitting of decay

data, and the error bars denote the fitting error. Bottom: difference in T₁

between PBP-bound ligand and the ligand in buffer. Error bars are the sum

of both errors. b) Top: transverse relaxation time,T₂, in ms. Bars represent the

average value from fitting of decay data, and the error bars denote the fitting

error. Bottom: difference in T₂between PBP-bound ligand and the ligand in

buffer. Error bars are the sum of both errors. c) Inverse mobility (which is

approximately the average number of interactions and is proportional to the

local viscosity) detected at all three sites (weighted for population of the sites

– see text), at the internal site and at the two external sites (average of

both sites). d) Average dihedral angles around C8–7–6–5 and C7–8–9–10, i.e.

around the epoxide moiety. For the PBP-bound disparlure enantiomers the

color scheme is the same as for the other graphs. Bars represent the weighted

averages of all dihedral angles of retained poses, weighted by pose at each

site and by site distribution (see text). Error bars represent the S. D. The light

gray bars represent the dihedral angles at the global minimum for (+)-1 and

the horizontally striped bars represent the dihedral angles at the global mini-

mum for (–)-1 as calculated in.^[5]

Additionally, methyl groups or Ile 68 and Thr 73 also surround

that hydrogen atom in those internal poses. Conversely, 5-H_Rof

PBP1 with (–)-1c only has weak interactions with Phe 36 and

Phe 119 on the protein and intramolecular interactions with

7-H, 3-H_Sand 2-H. Counting these interactions overestimates

their contribution to the relative local viscosity.

Comparing T₂values obtained for PBP-bound disparlure

enantiomers and the value obtained for the free disparlure in

solvent (Figure 8b, Table 9) we see that, in general, there was a

decrease in T₂upon binding of disparlure to the PBPs. This

suggests that, generally, the mobility of the focal hydrogen

atoms decreased upon binding to the protein, consistent with

the conclusion from T₁data. The only case in which T₂in the

Eur. J. Org. Chem. 0000, 0–0

www.eurjoc.org

9

Full Paper

Table 8. Spin-lattice relaxation (T₁) of 5,6-D₄(+)-1c and (–)-1c in CDCl₃, buffer and PBP-bound.

Condition

5-D_a

5-D_b

6-D_a& 6-D_b

(+)-1c

(–)-1c

(+)-1c

(–)-1c

(+)-1c

(–)-1c

CDCl₃

13.6 2.6

11.5 0.1

10.0 0.2

13.2 0.7

12.7 0.7

11.6 1.5

12.9 1.0

18.0 2.1

10.6 1.9

Phosphate buffer

Buffer+ LdisPBP1

Buffer+ LdisPBP2

11.7 0.5

20.05 1.42

23.0 1.9

16.5 0.7

12.2 0.7

17.9 0.9

25.5 1.6

13.6 2.4

19.8 0.5

Table 9. Spin-spin (transverse) relaxation (T₂) of 5,6-D₄(+)-1c and (–)-1c in the presence and absence of LdisPBPs.

Condition

5-Da

5-Db

6-Da & 6-Db

(+)-1c

(–)-1c

(+)-1c

(–)-1c

(+)-1c

(–)-1c

CDCl₃

12.0 0.1

10.3 0.5

6.5 0.8

7.2 0.5

11.4 0.6

9.1 0.2

5.2 1.3

7.8 0.5

10.2 0.1

11.7 0.3

7.2 0.9

3.2 0.1

Phosphate buffer

Buffer+ LdisPBP1

Buffer+ LdisPBP2

8.3 0.9

3.6 0.5

6.5 0.6

9.9 0.6

4.6 1.1

7.5 0.6

protein-bound disparlure and the solvent did not differ was is consistent with the body of previous studies on PBP1 and

5-H_Rin PBP2 with (+)-1. This is interesting, because we know PBP2 interactions with the enantiomers of disparlure. PBP1

from T₁and the modelling that 5-H_Rin this case is very strongly interacts more weakly overall but with greater differences in

restricted in its mobility at the internal binding site, so T₂equilibrium dissociation constants than PBP2,^[3,5,6,33]and PBP1

should be short and not detected. From kinetic data we know associates with and dissociates from disparlure enantiomers

that, for PBP2, once disparlure is bound internally it dissociates faster and more selectively than PBP2.^[4,18]The slower kinetic

only very slowly (with a rate constant of 4.7 0.4 × 10^–4s^–1),^[4]regime followed by PBP2 is consistent with the greater number

too slow for T₂to be detectable in the current experiment. At of significant differences between enantiomers in T₁and T₂de-

the external sites in this case, interactions with 5-H_Rin PBP2 tected here. PBP1 binds (–)-1 more strongly at equilibrium than

with (+)-1 are weak, and anisotropic effects cancel each other, (+)-1, and PBP2 is the opposite.^[3,5,33]Taken together, the chem-

so fluctuations in the magnetic field would be averaged as they ical shift, T₁and T₂data, along with the models, demonstrate

would in a solvent.

that the PBPs discriminate the disparlure enantiomers not only

by binding differently to the epoxide moiety (Figure 8d), but

also by interacting with subtle differences at positions 5 and 6,

which were identified in 1977 as important general odotopes of

disparlure in an electrophysiology structure-activity study with

racemic disparlure analogs.^[34]

Another interesting case is 6-H_S/Rin PBP1 with (+)-1, in which

T₂decreased significantly relative to solvent but T₁was no dif-

ferent than in solvent. This pattern can be explained by a high

number of interactions with 6-H_S/Rat the internal binding site.

PBP1 associates with and dissociates from ligand much faster

than PBP2, so that internal binding contributions would be de-

tectable in the pattern of T₂. PBP1 binds (+)-1 more weakly

than (–)-1, at internal^[3,33]and external (this study) binding sites.

Thus, for PBP1 and the two enantiomers, the T₂pattern is likely

dominated by contributions from internal binding, and the in-

verted mobility pattern (or local viscosity) (Figure 8c) perfectly

explains the T₂pattern: more viscosity (lower T₂) for (+)-1 and

lower local viscosity (higher T₂) for (–)-1.

Conclusion

In conclusion, a concise and efficient synthetic route to bioac-

tive (+)-disparlure, (–)-disparlure and isotope labelled disaprlure

enantiomers was completed in high overall yields from com-

mercially available dodecanal. Key steps in the synthesis in-

volved MacMillan's SOMO-activated α-chlorination of the alde-

hyde, which was shown to be a highly effective means for pre-

paring enantiopure α-chloroaldehyde (>99 % ee), and Mitsu-

nobu inversion. Field tests showed that (+)-disparlure was very

attractive. This is the first report on a study of interactions be-

tween isotope labelled disparlure enantiomers and LdisPBPs of

Finally, the case of 6-H_S/Rin PBP2 with (–)-1 is also interest-

ing. The T₁for this case did not change significantly from sol-

vent (Figure 8a), but T₂significantly decreased upon protein

+

binding (Figure 8b). In external site 2 of PBP2, the Lys 2 ε-NH₃

is H-bonded to the epoxide of (–)-1 and the δ-CH₂of Lys 2

packs near 6-H_Sin the model. Both 6-H_Sand 6-H_Rpack with the

ꢀ-CH₂of Asp 132. These interactions could restrict mobility of

the 6-H_S/Rof (–)-1, such that the T₂value detected is shorter

than the one detected with (+)-1, which does not interact in

that way with site 2 of PBP2.

2

gypsy moth by using H NMR spectroscopy. The present study

confirms that the 5 and 6 positions on the hydrophobic short

side chain of the disparlure enantiomers play an important role

in recognition of the disparlure enantiomers in the binding site

of the LdisPBPs.

PBP1 only had a significant difference in T₁between bound

enantiomers of disparlure at 5-H_R, whereas PBP2 showed

enantiomer differences at 5-H_S, 5-H_Rand 6-H_S/R(Figure 8a). For

T₂, PBP1 showed significant differences between enantiomers

for 5-H_Rand 6-H_S/R, and PBP2 again showed significant differen-

Experimental Section

General

All reactions were carried out in the presence of a nitrogen atmos-

ces between enantiomers for all three signals (Figure 8b). This phere and at room temperature (22 °C), unless the reactions were

Eur. J. Org. Chem. 0000, 0–0

www.eurjoc.org

10 © 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

performed in aqueous media or unless otherwise specified. Reac-

tions carried out at –78 °C used a bath of dry ice in acetone. Reac-

tions undertaken at 0 °C utilized a bath of water and ice. Hexanes

and ethyl acetate were distilled prior to use. THF was distilled from

sodium benzophenone. Chemicals and Reagents were used without

further purification. Syringes and cannulas were used to transfer

reagents. Reactions were monitored by thin layer chromatography

(TLC) on aluminum baked silica plates (Merck Silica Gel 60 F254)

and products were visualized under UV (λ = 254 nm) or stained

with phosphomolybdic acid (PMA), anisaldehyde or potassium per-

manganate, followed by exposure of the stained plates to heat.

Silica flash chromatography (Fisher Silica Gel 60 40–63 μm) was

undertaken to purify crude reaction mixtures using hexanes/ethyl

acetate mixture. The enantiomeric excess (ee) of 2-chlorododecyl

benzoates 8 was analyzed on an Agilent 1100 HPLC equipped with

a chiral Lux 5μm Cellulose-2 column (Phenomenex) and variable

wavelength detector (VWD). The HPLC chromatograph was pro-

grammed isocratically with hexanes/2-propanol (99:1). Optical rota-

Sample preparation for T₁and T₂relaxation experiments

The stock solutions of poorly water soluble deuterated disparlure

enantiomers were prepared by dissolving in acetonitrile. 50 m

phosphate buffer of pH 8 was prepared by using mono and dibasic

sodium phosphate. The protein (LdisPBPs) solutions (0.3 m ) were

incubated with deuterium labelled disparlure enantiomers (0.6 m

M

M)

in ice cold water for 3 h. The protein-ligand interaction process

was monitored through acquisition of ²H NMR spectra. The ²H NMR

relaxation spectra of the deuterated disparlure enantiomers/

LdisPBPs system were obtained using a Bruker –600 NMR spectrom-

eter equipped with QCI probe. The T₁(relaxation) times were meas-

ured by using with the standard inversion-recovery pulse se-

quence^[35](180° – τ - 90°). The T₂(transverse relaxation) times were

measured using with Carr-Purcell-Meiboom-Gill pulse sequence^[36]

(90° (τ-180° – τ) _n). The delay times in between 180° were used as

0.42 ms. T₁and T₂relaxation curves were fitted to Equation 1 and

Equation 2, respectively.

tions were recorded on a Perkin–Elmer Polarimeter 340 thermostat- M_z(t) = M_z(0) (1 – exp (–t/T₁)

ted to 20 °C, using the sodium D line.

The ¹H NMR spectra were obtained on Bruker DRX 400 and 500 MHz M_xy(t) = M_xy(0) (exp (–t/T₂)

(1)

(2)

spectrometers in CDCl₃. Chemical shifts and coupling constants are

1

reported in parts per million (ppm) and Hertz (Hz) respectively. H

NMR data was reported as follows: chemical shift values (ppm), mul-

tiplicity (s = singlet, d = doublet, t = triplet, q = quartet, m = multi-

plet). ¹³C NMR spectra were recorded in CDCl₃by using a Bruker

DRX 400 or DRX 500. ¹³C NMR data was reported as chemical shift

values (ppm). IR spectra were obtained with a Perkin–Elmer Spec-

trum One FT-IR spectrometer and samples were directly placed on

the KBr plates. High-resolution mass spectra (HRMS) were obtained

by using positive electrospray ionization and by TOF method.

Syringes and cannulas were used to transfer reagents. The GC–MS

analysis was performed on GC–MS (Varian CP3¯800 GC, interfaced

with a Varian Saturn 2000 MS ) using a SPB-5 fused silica capillary

column (30 m × 0.25 mm i.d., film thickness 0.25 μm, Supelco, Belle-

fonte, PA, USA) with positive electron ionization (EI), and final

GC–MS analysis was done on a Perkin Elmer Clarus 690 GC with a

Elite-5MS (30 m × 0.25 mm i.d., film thickness 0.25 μm) column,

interfaced with a SQ8T quadrupole mass spectrometer. EI was the

ionization mode, and He was the carrier gas.

Homology models of LdisPBP1 and LdisPBP2

Gypsy moth PBP1 and PBP2 share 61.7 % and 48.9 % similarity with

Bombyx mori PBP, respectively. The homology models of LdisPBP1

and LdisPBP2 were created based on the crystal structure of Bom-

byx mori PBP (PDB ID: 1LS8.1. A) by using SWISS MODEL (ex-

pasy.org).^[37,38]The models were corrected for lacking hydrogen at-

oms, using the “Protonate 3D” function in Molecular Operating En-

vironment (MOE, Chemical Computing Group, Montreal, Canada) at

pH 8 and with 0.10

M salt concentration. Using the MMFF94X force

field, the “protonate 3D” function assigns the location of hydrogen

atoms and ionization states in the protein structure. After protona-

tion, energy minimization was applied to the protein structures.

Molecular docking simulations

The homology models were docked with disparlure enantiomers by

using an induced fit protocol. Docking stimulations were performed

with the MMFF94X force field in MOE, using default parameters

(Refinement: Force field, Placement: Triangle Matcher). Two scoring

functions (Rescoring 1: London dG and Rescoring 2: GBVI/WSA dG)

were used for this docking protocol. The ligand binding sites of

protein were detected in MOE by using geometric algorithm based

on Edelsbrunner's Alpha Shapes called “Site Finder” in MOE.^[39]

Stopped-flow kinetics

Kinetics experiments were performed in a Chirascan stopped-flow

instrument (Applied Photophysics Ltd, UK). For the competition ki-

netics, N-phenyl-1-naphthylamine (1-NPN) was used as the fluores-

cent reporter. First, the association and dissociation rate constants

of 1-NPN were determined by association and the values were used

as constraints for calculating the k_onand k_offof the ligands. In the

competitive assay, one reservoir was filled with the PBP1 solution

Evaluation of molecular docking simulation data

The simulations were programmed such that the fit of the ligand

into the binding site was induced and up to 30 poses of the ligand

were retained. We obtained between 18 and 30 poses, which were

evaluated for the total potential energy. Using the assumption that

all the poses can equilibrate with the pose of lowest energy, a distri-

bution of poses was calculated for each case (see supplemental

information).

(0.50 μ

25 m

tained the solution of NPN (0.50 μ

tions of (+)-1c or (–)-1c in the same buffer. The NPN was added

from a 0.50 m NPN in methanol stock. Six microliters of (+)-1c or

(–)-1c was added from stock solutions in ethanol (0.20 m to

1.6 m ). The stopped-flow instrument mixed the solutions in 1:1

volume ratio from the two reservoirs. The final concentrations of

the participating components were the following: (1) PBP1 0.25 μ

(2) NPN 0.25 μ , and (3) (+)-1c or (–)-1c 0 to 9.4 μM. NPN fluores- of the focal atoms were selected. Distances of nearby hydrogen

M

PBP1 in 20 m

NaCl, and 0.1 % ethanol) while the second reservoir con-

) mixed with various concentra-

M Tris/HCl buffer pH 8.0 with 180 mM KCl,

M

Next, all poses that were found to be populated above 5 % were

examined with regard to molecular interactions with the epoxide

and the 5_S, 5_R, 6_Sand 6_Rhydrogen atoms. To perform this inspec-

tion, binding pocket residues and functional groups within a 4.5 Å

M

M,

M

cence was observed using 337 nm excitation and the emission de-

atoms to the 5 and 6 H atoms were measured, both on the protein

tected using longpass filter disc with 395 nm cutoff. The experiment

and on disparlure itself. We also checked whether the focal

was performed at constant temperature, 22 °C. The fluorescent tra- hydrogen atoms are within the shielding or deshielding zones of

ces were fitted using kinetics of competitive binding model in

GraphPad Prism 5 (GraphPad Software Inc., California).

nearby phenyl or indole systems of phenylalanine or tryptophan

residues or of carbonyl groups (in the backbone or aspartate,

Eur. J. Org. Chem. 0000, 0–0

www.eurjoc.org

11

Full Paper

glutamate, asparagine or glutamine side chains). We also assessed

every focal hydrogen atom with regard to its position relative to

the oxirane ring's shielding and deshielding zones.^[40]The interac-

tions found were tabulated and scored according to the following

scale: 0 = neither shielding nor deshielding; –1 = weak shielding;

1 = weak deshielding; –2 moderate shielding; 2 = moderate de-

shielding; –3 = strong shielding; 3 = strong deshielding. This assess-

ment was performed for each enantiomer in each relevant retained

pose (populated > 5 %). Weighted averages were calculated for

each enantiomer at each site.

¹³C NMR (500 MHz, CDCl₃) δ: 195.3, 63.8, 31.9, 31.7, 29.4, 29.3, 29.1,

28.8, 25.4, 22.6, 14.0. IR (neat): ν = 2957, 2928, 2855, 2718, 1738,

˜

1466, 759 cm^–1

.

HRMS (ESI) m/z calculated for C₁₂H₂₃ClO

[M – H] 217.1359, found 217.1362. [α]²⁵_D:–9.9 (c 0.5, CHCl₃).

General procedure for Preparation of (+)-4 and (–)-4

To a cold (–78 °C), stirred solution of 1-hexyne (0.015 mol) in THF

(20 mL), a solution of n-butyllithium (2.5

M in hexanes, 0.014 mmol)

was added dropwise. The resulting mixture was stirred at –78 °C for

1 h. After this time, a solution of (+)-3 or (–)-3 (0.008 mol) in THF

(10 mL) was added dropwise and the reaction mixture was stirred

for an additional 1 h. Then the reaction mixture was quenched with

a saturated aqueous solution of NH₄Cl (10 mL), diluted with ethyl

acetate (50 mL) and water (20 mL). The phases were separated, and

the aqueous phase was extracted with ethyl acetate (3×50 mL). The

combined organic layers were washed with brine (30 mL), dried

with MgSO₄, and concentrated under reduced pressure, which af-

forded crude product (dr. 20:1 determined by ¹H NMR analysis of

the crude reaction mixture). Purification of the crude product by

flash chromatography afforded (silica gel, 95:5 hexanes/ethyl acet-

ate) yielded (+)-4 or (–)-4.

Finally, a global weighting was done for internal and external sites,

taking into consideration that the ligand can partition between ex-

ternal sites and the internal one, according to previous kinetic bind-

ing experiments,^[4]and overall equilibrium binding constants deter-

mined recently at pH 8.0.^[33]The overall binding constants deter-

mined at equilibrium were taken to reflect the affinity of the pro-

teins for internally bound ligands. Affinity of the external sites for

the ligands was determined here for PBP1, using a fluorescent dis-

placement assay in a stopped-flow apparatus and from our previous

study^[4]for PBP2. Dissociation constants for external binding were

estimated as k_off/k_on, the ratio of the dissociation and association

rate constants. Shielding and deshielding contributions from the

three sites were weighted according to the percentages shown in

Table 7.

(7S, 8R)-8-Chloro-2-methyloctadec-5-yn-7-ol (+)-4

(Eluent: 5 % EA/hexane), 1.94 g, 89 %, pale yellow oil: ¹H NMR

(500 MHz, CDCl₃) δ: 4.50 (dt, 1H, J = 3.6, 1.9 Hz), 4.0 (ddd, 1H, J =

9.3, 4.4, 3.5 Hz), 2.24 (td, 2H, J = 7.4 Hz, 2.0 Hz),1.82 (m, 2H),1.70 (m,

1H),1.56 (m, 2H),1.27 (m, 17H), 0.90 (d, 6H, J = 6.6 Hz), 0.89 (t, 3H,

J = 7 Hz). ¹³C NMR (500 MHz, CDCl₃) δ: 87.9, 76.8, 67.6, 66.4, 37.5.6,

33.7, 32.0, 29.7, 29.5, 29.4, 29.2, 27.3, 26.6, 22.8, 22.2, 16.8, 14.2. IR

Synthetic procedures

General procedure for Preparation of (+)-3 and (–)-3

To a cold (0 °C), stirred suspension of lithium chloride (0.016 mol) in

acetonitrile (50 mL), was added copper (II) trifluoroacetate hydrate

(0.004 mol) and sodium persulfate (0.004 mol) followed by H₂O

(0.017 mol). The reaction mixture was stirred for 5 min and then

the SOMO catalyst ((2S, 5R)-2-(tert-butyl)-3,5-dimethylimidazolidin-

4-one or (2R, 5S)-2-(tert-butyl)-3,5-dimethylimidazolidin-4-one

0.0016 mol) was added. After stirring for 5 min at 0 °C, dodecanal

(0.008 mmol) was added. The reaction mixture was stirred for 1 h

and then allowed to slowly warm to 5 °C over the course of 24 h.

The mixture was stirred at 5 °C until dodecanal had been com-

pletely consumed (as determined by ¹H NMR spectroscopy). After

this time, the reaction mixture was treated with water (20 mL) and

diluted with ethyl acetate (50 mL), and the phases were separated.

The aqueous phase was extracted with ethyl acetate (3×50 mL) and

the combined organic phases were washed with brine (20 mL),

dried with MgSO₄, and concentrated to give the crude chloroalde-

hyde. Purification of the crude product by flash chromatography

afforded the desired products.

(neat): ν = 3416, 2956, 2926, 2855, 2237, 1466, 1042, 759 cm^–1

HRMS (ESI) m/z calculated for C₁₈H₃₃ClONa [M + Na]: 337.2274,

.

˜

found 337.1042.

(7R, 8S)-8-Chloro-2-methyloctadec-5-yn-7-ol (–)-4

(Eluent: 5 % EA/hexane), 1.88 g, 86 %, pale yellow oil: ¹H NMR

(500 MHz, CDCl₃) δ: 4.50 (dt, 1H, J = 3.6, 1.9 Hz), 4.03–3.99 (ddd,

1H, J = 9.3, 4.4, 3.5 Hz), 2.26–2.22 (td, 2H, J = 7.4 Hz, 2.0 Hz),1.88–

1.75 (m, 2H),1.72–1.66 (m, 1H),1.63–1.52 (m, 2H),1.26 (m, 17H), 0.90

(d, 6H, J = 6.6 Hz), 0.86 (t, 3H, J = 7 Hz). ¹³C NMR (500 MHz, CDCl₃)

δ: 87.7, 76.8, 67.4, 66.2, 37.5.6, 33.5, 31.8, 30.4, 29.5, 29.3, 29.2, 29.0,

26.3, 22.6, 21.8, 18.2, 14.0, 13.4. IR (neat): ν = 3417, 2957, 2926,

˜

2855, 2237, 1466, 1042, 759 cm^–1. HRMS (ESI) m/z calculated for

C₁₈H₃₃ClONa [M + Na]: 337.2274, found 337.1042.

General procedure for Preparation of (+)-5, (+)-5a, (+)-5b,

(+)-5c, (–)-5, (–)-5b and (–)-5c

(+)-(2R)-2-Chlorododecanal (+)- 3

(Eluent: 1 % EA/hexane), 1.03 g, 87 %, pale yellow oil: ¹H NMR

(500 MHz, CDCl₃) δ: 9.48 (d, 1H, J = 2.5 Hz), 4.15 (ddd, 1H, J = 8.5 Hz,

5.5 Hz, 3 Hz), 1.99–1.92 (dddd, 1H, J = 16 Hz, 11.5 Hz, 10.5 Hz, 6Hz),

1.84–1.78 (dddd, 1H, J = 18.5 Hz, 14 Hz, 8.5 Hz, 5 Hz), 1.54–1.48 (m,

1H), 1.44–1.37 (m, 1H), 1.25 (m, 14H), 0.88 (t, 3H, J = 7 Hz). ¹³C NMR

(500 MHz, CDCl₃) δ: 194.7, 63.4, 31.4, 31.3, 28.9, 28.8, 28.7, 28.3,

To an ice cold, stirred solution of triphenylphosphane (0.0082 mol)

in dry THF (20 mL), benzoic acid or ¹⁷O or ¹⁸O benzoic acid

(0.0082 mol) and the solution of (+)-4 or (–)-4 (0.0041 mol) in dry

THF (10 mL) were added slowly to the reaction flask under an inert

N₂atmosphere. The resulting mixture was stirred at 0 °C for 5 min-

utes. After this time, a solution of diisopropyl azodicarboxylate

(0.0082 mol) in dry THF (5 mL) was added dropwise and the reac-

tion mixture was stirred at room temperature for 12 h. The reaction

solvent was removed by evaporation under reduced pressure and

diluted with ethyl acetate (50 mL) and water (20 mL). The phases

were separated, and the aqueous phase was extracted with ethyl

acetate (3 × 50 mL). The combined organic layers were washed with

brine (20 mL), dried with MgSO₄, and concentrated under reduced

pressure, which afforded crude product. Purification of the crude

product by flash chromatography (silica gel, 99:1 hexanes/ethyl

acetate) afforded the desired target compounds.

24.9, 22.0, 13.5. IR (neat): ν = 2955, 2925, 2855, 2718, 1736, 1466,

˜

759 cm^–1. HRMS (ESI) m/z calculated for C₁₂H₂₃ClO [M – H] 217.1359,

found 217.1362. [α]²⁵_D:+10 (c 0.5, CHCl₃).

(–)-(2S)-2-Chlorododecanal (–)-3

(Eluent: 1 % EA/hexane), 1.0 g, 84 %, pale yellow oil: ¹H NMR

(500 MHz, CDCl₃) δ: 9.48 (d, 1H, J = 2.5 Hz), 4.17–4.14 (ddd, 1H, J =

8.5 Hz, 5.5 Hz, 3 Hz), 2.02–1.93 (dddd, 1H, J = 16 Hz, 11.5 Hz, 10.5 Hz,

6Hz), 1.86–1.77 (dddd, 1H, J = 18.5 Hz, 14 Hz, 8.5 Hz, 5 Hz), 1.68–

1.57 (m, 1H), 1.53–1.39 (m, 1H), 1.26 (m, 14H), 0.87 (t, 3H, J = 7 Hz).

Eur. J. Org. Chem. 0000, 0–0

www.eurjoc.org

12

Full Paper

(7R, 8R)-8-Chloro-2-methyloctadec-5-yn-7-yl benzoate (+)-5

(m, 2H), 1.43 (dd, 2H, J = 14.5 Hz, 7.3 Hz), 1.26 (m, 14H), 0.89 (t, 3H,

J = 6.3 Hz), 0.88 (d, 6H, J = 6.6 Hz ). ¹³C NMR (500 MHz, CDCl₃) δ:

165.1, 133.2, 130.0, 129.5, 128.3, 88.4, 77.1, 74.2, 67.7, 62.4, 37.1,

33.7, 32.0, 29.5, 29.4, 29.3 29.2, 29.0, 27.1, 26.0 22.6, 22.0, 16.7, 14.0.

(Eluent: 1 % EA/hexane), 1.37 g, 80 %, pale yellow oil: ¹H NMR

(500 MHz, CDCl₃) δ: 8.09 (d, 2H, J = 7.2 Hz), 7.58 (t, 1H, J = 7.4 Hz),

7.46 (t, 2H, J = 7.7 Hz), 5.74 (dt, 1H, J = 6.1 Hz, 1.9 Hz), 4.09 (ddd,

1H, J = 9.6 Hz, 6.2 Hz, 3.4 Hz), 2.24 (td, 2H, J = 7.4 Hz, 1.9 Hz), 2.02

(m, 1H), 1.84 (m, 1H), 1.66 (m, 2H), 1.43 (dd, 2H, J = 14.5 Hz, 7.3 Hz),

1.26 (m, 14H), 1.16 (m, 2H), 0.89 (t, 3H, J = 6.3 Hz), 0.88 (d, 6H, J =

6.6 Hz ). ¹³C NMR (500 MHz, CDCl₃) δ: 165.1, 133.3, 130.0, 129.5,

128.5, 88.7, 77.1, 74.4, 67.8, 62.5, 37.2, 33.8, 32.0, 29.7, 29.6, 29.5

IR (neat): ν = 2941, 2879, 2238, 1727, 1261, 1103, 1093, 956, 708 cm^–1

.

˜

HRMS (ESI) m/z calculated for C₂₆H₃₉ClO₂[M + H]: 419.2701, found

419.2696. [α]²_D⁵: - 5.67 (c 0.8, CCl₄).

(7S, 8S)-8-Chloro-octadec-5-yn-7-yl benzoate-¹⁷O₂(–)-5b

Triphenylphosphane (0.00127 mol), ¹⁷O benzoic acid (0.00127 mol)

and (–)-4 (0.000636 mol). (Eluent: 1 % EA/hexane), 0.208 g, 78 %,

pale yellow oil: ¹H NMR (500 MHz, CDCl₃) δ: 8.09–8.08 (d, 2H, J =

7.2 Hz), 7.60–7.56 (t, 1H, J = 7.4 Hz), 7.46–7.44 (t, 2H, J = 7.7 Hz),

5.76–5.73 (dt, 1H, J = 6.1 Hz, 1.9 Hz), 4.12–4.07 (ddd, 1H, J = 9.6 Hz,

6.2 Hz, 3.4 Hz), 2.26–2.23 (td, 2H, J = 7.4 Hz, 1.9 Hz), 2.07–1.99 (m,

1H), 1.89–1.79 (m, 1H), 1.71–1.63 (m, 1H), 1.26 (m, 18H), 0.89 (t, 3H,

J = 6.3 Hz), 0.87 (d, 6H, J = 6.6 Hz ). ¹³C NMR (500 MHz, CDCl₃) δ:

29.4, 29.1, 27.3, 26.1 22.7, 22.2, 16.8, 14.2. IR (neat): ν = 2941, 2879,

˜

2238, 1727, 1261, 1103, 1093, 956, 708 cm^–1. HRMS (ESI) m/z calcu-

lated for C₂₆H₃₉ClO₂[M + H]: 419.2701, found 419.2696. [α]²_D⁵: + 5.7

(c 0.8, CCl₄).

(7R, 8R)-8-Chloro-octadec-5-yn-7-yl benzoate (+)-5a

(Eluent: 1 % EA/hexane), 1.25 g, 73 %, pale yellow oil: ¹H NMR

(500 MHz, CDCl₃) δ: 8.09 (d, 2H, J = 7.2 Hz), 7.58 (t, 1H, J = 7.4 Hz),

7.46 (t, 2H, J = 7.7 Hz), 5.74 (dt, 1H, J = 6.1 Hz, 1.9 Hz), 4.09 (ddd,

1H, J = 9.6 Hz, 6.2 Hz, 3.4 Hz), 2.24 (td, 2H, J = 7.4 Hz, 1.9 Hz), 2.02

(m, 1H), 1.84 (m, 1H), 1.66 (m, 2H), 1.43 (dd, 2H, J = 14.5 Hz, 7.3 Hz),

1.26 (m, 14H), 1.16 (m, 2H), 0.89 (t, 3H, J = 6.3 Hz), 0.87 (t, 3H, J =

5.9 Hz ). ¹³C NMR (500 MHz, CDCl₃) δ: 165.1, 133.3, 130.0, 128.5,

88.7, 77.1, 74.4, 67.8, 62.5, 37.2, 33.8, 32.0, 29.7, 29.6, 29.5 29.4, 29.1,

165.0, 164.9 (¹³C_benzoate

-

¹⁷O), 133.1, 129.7, 129.5, 128.2, 88.3, 77.1,

¹⁷O), 62.4, 37.0, 33.6, 31.8, 29.5, 29.4, 29.3

74.2, 67.7, 67.6 (¹³C_benzoate

-

29.2, 28.9, 27.1, 25.9 22.6, 22.0, 16.7, 13.9. IR (neat): ν = 2947, 2882,

˜

2238, 1727, 1261, 1103, 1093, 956, 708 cm^–1. [α]_D²⁵: - 5.67 (c 0.8,

CCl₄).

(7S, 8S)-8-Chloro-octadec-5-yn-7-yl benzoate-¹⁸O₂(–)-5c

27.3, 26.1 22.7, 22.2, 16.8, 14.2. IR (neat): ν = 2941, 2879, 2238, 1727,

˜

Triphenylphosphane (0.00127 mol), ¹⁸O benzoic acid (0.00127 mol)

and (–)-4 (0.000636 mol). (Eluent: 1 % EA/hexane), 0.213 g, 80 %,

pale yellow oil: ¹H NMR (500 MHz, CDCl₃) δ: 8.10–8.07 (d, 2H, J =

7.2 Hz), 7.60–7.56 (t, 1H, J = 7.4 Hz), 7.48–7.44 (t, 2H, J = 7.7 Hz),

5.75–5.74 (dt, 1H, J = 6.1 Hz, 1.9 Hz), 4.12–4.07 (ddd, 1H, J = 9.6 Hz,

6.2 Hz, 3.4 Hz), 2.26–2.22 (td, 2H, J = 7.4 Hz, 1.9 Hz), 2.07–1.99 (m,

1H), 1.89–1.79 (m, 1H), 1.71–1.64 (m, 1H), 1.30–1.26 (m, 18H), 0.89

(t, 3H, J = 6.3 Hz), 0.87 (d, 6H, J = 6.6 Hz ). ¹³C NMR (500 MHz,

1261, 1103, 1093, 956, 708 cm^–1. [α]_D²⁵: + 5.7 (c 0.8, CCl₄).

(7R, 8R)-8-Chloro-2-methyloctadec-5-yn-7-yl benzoate-¹⁷O₂(+)-

5b

Triphenylphosphane (0.00127 mol), ¹⁷O benzoic acid (0.00127 mol)

and (+)-4 (0.000636 mol). (Eluent: 1 % EA/hexane), 0.208 g, 78 %,

pale yellow oil: ¹H NMR (500 MHz, CDCl₃) δ: 8.09–8.08 (d, 2H, J =

7.2 Hz), 7.60–7.57 (t, 1H, J = 7.4 Hz), 7.47–7.44 (t, 2H, J = 7.7 Hz),

5.75–5.74 (dt, 1H, J = 6.1 Hz, 1.9 Hz), 4.12–4.08 (ddd, 1H, J = 9.6 Hz,

6.2 Hz, 3.4 Hz), 2.26–2.23 (td, 2H, J = 7.4 Hz, 1.9 Hz), 2.07–2.00 (m,

1H), 1.88–1.80 (m, 1H), 1.70–1.65 (m, 1H), 1.26 (m, 18H), 0.89 (t, 3H,

J = 6.3 Hz), 0.87 (d, 6H, J = 6.6 Hz ). ¹³C NMR (500 MHz, CDCl₃) δ:

CDCl₃) δ: 165.05, 165.03 (¹³C_benzoate

-

¹⁸O), 133.2, 129.8, 128.3, 88.4,

77.1, 74.2, 67.7, 67.6 (¹³C_benzoate¹⁸O), 62.4, 37.1, 33.6, 31.8, 29.5,

-

29.4, 29.3 29.2, 28.9, 27.1, 26.0 22.6, 22.0, 16.7, 14.0. IR (neat):

ν = 2945, 2879, 2238, 1727, 1261, 1103, 1093, 956, 708 cm^–1. HRMS

˜

(ESI) m/z calculated for C₂₆H₃₉ClNa¹⁸O₂[M + Na]: 443.2573, found

443.2574. [α]²_D⁵: - 5.67 (c 0.8, CCl₄).

165.05, 165.02 (¹³C_benzoate

74.2, 67.7, 67.6 (¹³C_benzoate

-

¹⁷O), 133.2, 129.8, 129.5, 128.3, 88.4, 77.1,

¹⁷O), 62.4, 37.1, 33.6, 31.8, 29.5, 29.4, 29.3

29.2, 28.9, 27.1, 25.9 22.6, 22.0, 16.7, 14.0. IR (neat): ν = 2947, 2882,

˜

General procedure for Preparation of (+)-6, (+)-6a, (+)-6b,

(+)-6c, (–)-6, (–)-6b and (–)-6c

2238, 1727, 1261, 1103, 1093, 956, 708 cm^–1. [α]_D²⁵: + 5.7 (c 0.8, CCl₄).

(7R, 8R)-8-Chloro-2-methyloctadec-5-yn-7-yl benzoate-¹⁸O₂(+)-

5c

A 100 mL round-bottomed flask was charged with methanol or

benzene. The flask was evacuated and back filled with nitrogen gas.

After two vacuum/nitrogen cycles to replace air with nitrogen inside

the reaction flask, the compound (+)-6 or (–)-6 (0.0028 mol), 20 %

Pd/C (240 mg, 20 wt.-% of compound) or RhCl(PPh3)₃(10 mol-%)

were added. The reaction mixture was vigorously stirred at room

temperature under atmospheric hydrogen or deuterium pressure

(balloon) for 3 h. After that the reaction mixture was filtered

through a celite pad, and the filtrate was concentrated on a rotary

evaporator to afford crude product. The crude product was carried

to the next step without further purification.

Triphenylphosphane (0.00127 mol), ¹⁸O benzoic acid (0.00127 mol)

and (+)-4 (0.000636 mol). (Eluent: 1 % EA/hexane), 0.213 g, 80 %,

pale yellow oil: ¹H NMR (500 MHz, CDCl₃) δ: 8.09–8.08 (d, 2H, J =

7.2 Hz), 7.60–7.57 (t, 1H, J = 7.4 Hz), 7.47–7.44 (t, 2H, J = 7.7 Hz),

5.75–5.73 (dt, 1H, J = 6.1 Hz, 1.9 Hz), 4.12–4.08 (ddd, 1H, J = 9.6 Hz,

6.2 Hz, 3.4 Hz), 2.25–2.22 (td, 2H, J = 7.4 Hz, 1.9 Hz), 2.06–2.02 (m,

1H), 1.88–1.80 (m, 1H), 1.70–1.65 (m, 1H), 1.30–1.26 (m, 18H), 0.89

(t, 3H, J = 6.3 Hz), 0.87 (d, 6H, J = 6.6 Hz ). ¹³C NMR (500 MHz,

CDCl₃) δ: 165.05, 165.03 (¹³C_benzoate

-

¹⁸O), 133.2, 129.8, 128.3, 88.4,

77.1, 74.2, 67.7, 67.6 (¹³C_benzoate¹⁸O), 62.4, 37.1, 33.6, 31.8, 29.5,

-

(7R, 8R)-8-Chloro-2-methyloctadecan-7-yl benzoate (+)-6

29.4, 29.3 29.2, 28.9, 27.1, 26.0 22.6, 22.0, 16.7, 14.0. IR (neat): ν =

˜

Yield 1.14 g, 95 %, colorless oil: ¹H NMR (500 MHz, CDCl₃) δ: 8.09

(d, 2H, J = 7.2 Hz), 7.58 (t, 1H, J = 7.4 Hz), 7.45 (t, 2H, J = 7.7 Hz),

5.28 (dt, 1H, J = 7.2 Hz, 3.2 Hz), 4.06 (dt, 1H, J = 9.1 Hz, 4.0 Hz), 1.82

(m, 4H), 1.50 (m, 1H), 1.41 (m, 1H), 1.34 (m, 4H), 1.23 (m, 15H), 1.16

(m, 2H),0.87 (t, 3H, J = 6.3 Hz), 0.84 (d, 6H, J = 6.6 Hz ). ¹³C NMR

(500 MHz, CDCl₃) δ: 166.0, 133.1, 130.0, 128.5, 75.7, 63.9, 38.7, 34.6,

32.0, 31.4, 29.6, 29.5, 29.4 29.3, 29.0, 27.9, 27.2, 26.7, 25.7, 22.7, 22.6,

2945, 2879, 2238, 1727, 1261, 1103, 1093, 956, 708 cm^–1. HRMS

(ESI) m/z calculated for C₂₆H₃₉ClNa¹⁸O₂[M + Na]: 443.2573, found

443.2574. [α]²_D⁵: + 5.7 (c 0.8, CCl₄).

(7S, 8S)-8-Chloro-octadec-5-yn-7-yl benzoate (–)-5

(Eluent: 1 % EA/hexane), 1.30 g, 76 %, pale yellow oil: ¹H NMR

(500 MHz, CDCl₃) δ: 8.09–8.07 (d, 2H, J = 7.2 Hz), 7.60–7.56 (t, 1H,

J = 7.4 Hz), 7.46–7.44 (t, 2H, J = 7.7 Hz), 5.75–5.74 (dt, 1H, J = 6.1 Hz,

1.9 Hz), 4.12–4.07 (ddd, 1H, J = 9.6 Hz, 6.2 Hz, 3.4 Hz), 2.26–2.22 (td,

2H, J = 7.4 Hz, 1.9 Hz), 2.07–1.99 (m, 1H), 1.89–1.79 (m, 1H), 1.66

22.5, 14.1. IR (neat): ν = 2945, 2880, 1721, 1266, 1108, 1069, 709 cm^–1

.

˜

HRMS (ESI) m/z calcd. for C₂₆H₄₃ClO₂[M + H]: 424.0792, found

424.2289. [α]²_D⁵: +6.4 (c 0.68, CCl₄).

Eur. J. Org. Chem. 0000, 0–0

www.eurjoc.org

13

Full Paper

(7R, 8R)-8-Chloro-octadecan-7-yl benzoate (+)-6a

(7S, 8S)-8-Chloro-2-methyloctadecan-7-yl benzoate-d₄, ¹⁸O₂(–)-6c

Yield 1.03 g, 86 %, colorless oil: ¹H NMR (500 MHz, CDCl₃) δ: 8.09

(d, 2H, J = 7.2 Hz), 7.58 (t, 1H, J = 7.4 Hz), 7.45 (t, 2H, J = 7.7 Hz),

5.28 (dt, 1H, J = 7.2 Hz, 3.2 Hz), 4.06 (dt, 1H, J = 9.1 Hz, 4.0 Hz), 1.82

(m, 4H), 1.50 (m, 1H), 1.41 (m, 1H), 1.34 (m, 4H), 1.23 (m, 15H), 1.16

(m, 2H), 0.89 (t, 3H, J = 6.3 Hz), 0.87 (t, 3H, J = 5.9 Hz ). ¹³C NMR

(500 MHz, CDCl₃) δ: 166.0, 133.1, 130.0, 128.5, 75.7, 63.9, 38.7, 34.6,

32.0, 31.4, 29.6, 29.5, 29.4 29.3, 29.0, 27.9, 27.2, 26.7, 25.7, 22.7, 22.6,

The compound (–)-6 (0.00024 mol) and 20 % Pd/C. 0.093 g, 92 %,

colorless oil: ¹H NMR (500 MHz, CDCl₃) δ: 8.10–8.08 (d, 2H, J =

7.2 Hz), 7.60–7.56 (t, 1H, J = 7.4 Hz), 7.48–7.45 (t, 2H, J = 7.7 Hz),

5.31–5.28 (m, 1H), 4.08–4.04 (dt, 1H, J = 9.1 Hz, 4.0 Hz), 1.83–1.71

(m, 2H), 1.53–1.47 (m, 1H), 1.30–1.23 (m, 20H), 0.87 (t, 3H, J = 6.3 Hz),

0.85 (d, 6H, J = 6.6 Hz ). ¹³C NMR (500 MHz, CDCl₃) δ: 165.9, 165.8

13

(

C

- -

¹⁸O), 133.0, 129.9, 129.7, 128.3, 75.5, 75.4 (¹³C_benzoate

benzoate

¹⁸O), 63.7, 38.6, 34.4, 31.8, 29.5, 29.4, 29.3 29.2, 28.9, 27.8, 26.5, 22.6,

22.5, 14.1. IR (neat): ν = 2945, 2880, 1721, 1266, 1108, 1069, 709 cm^–1

.

˜

[α]²_D⁵: +6.4 (c 0.68, CCl₄).

22.5, 22.4, 14.0. IR (neat): ν = 2955, 2888, 1727, 1266, 1110, 1080,

˜

709 cm^–1. HRMS (ESI) m/z calculated for C₂₆H₃₉ClKD₄¹⁸O₂[M + K]:

467.2888, found 467.2876. [α]²_D⁵: –6.4 (c 0.68, CCl₄).

(7R, 8R)-8-Chloro-2-methyloctadecan-7-yl benzoate-d_4,¹⁷O₂

(+)-6b

General procedure for Preparation of (+)-1, (+)-1a, (+)-1b, (+)-

1c, (–)-1, (–)-1b and (–)-1c

The compound (+)-6 (0.00024 mol) and 20 % Pd/C. 0.093 g, 92 %,

colorless oil: ¹H NMR (500 MHz, CDCl₃) δ: 8.10–8.08 (d, 2H, J =

7.2 Hz), 7.59–7.56 (t, 1H, J = 7.4 Hz), 7.48–7.45 (t, 2H, J = 7.7 Hz),

5.30–5.28 (m, 1H), 4.08–4.05 (dt, 1H, J = 9.1 Hz, 4.0 Hz), 1.83–1.69

(m, 3H), 1.53–148 (m, 1H), 1.33–1.24 (m, 19H), 0.87 (t, 3H, J = 6.3 Hz),

To an ice cold, stirred solution of (+)-6 or (–)-6 (0.9456 mmol) in

methanol (6 mL), 4 N NaOH in methanol (4 mL) was added slowly

to the reaction flask. The resulting mixture was stirred at room tem-

perature for 2 h. The reaction solvent was removed by evaporation

under reduced pressure and diluted with ethyl acetate (20 mL) and

water (10 mL). The phases were separated, and the aqueous phase

was extracted with ethyl acetate (3 × 20 mL). The combined organic

layers were washed with brine (10 mL), dried with MgSO₄, and con-

centrated under reduced pressure. Purification of the crude product

by flash chromatography (silica gel, 100 % hexanes) afforded (+)-

disparlure or (–)-disparlure.

0.84 (d, 6H, J = 6.6 Hz ). ¹³C NMR (500 MHz, CDCl₃) δ: 165.9, 165.8

13

(

C

-

¹⁷O), 133.0, 129.9, 129.7, 129.6, 128.3, 75.5, 75.4

¹⁷O), 63.7, 38.6, 34.4, 31.8, 29.5, 29.4, 29.3, 29.2 28.9,

benzoate

13

benzoate

27.7, 26.5, 22.6, 22.5, 22.4, 14.0. IR (neat): ν = 2945, 2880, 1721, 1266,

˜

1108, 1069, 709 cm^–1. [α]_D²⁵: +6.4 (c 0.68, CCl₄).

(7R, 8R)-8-Chloro-2-methyloctadecan-7-yl benzoate-d₄, ¹⁸O₂

(+)-6c

(7R, 8S)-7, 8-Epoxy-2-methyloctadecane (+)-1

The compound (+)-6 (0.00024 mol) and 20 % Pd/C. 0.093 g, 92 %,

colorless oil: ¹H NMR (500 MHz, CDCl₃) δ: 8.09–8.08 (d, 2H, J =

7.2 Hz), 7.59–7.56 (t, 1H, J = 7.4 Hz), 7.48–7.45 (t, 2H, J = 7.7 Hz),

5.29–5.27 (m, 1H), 4.07–4.04 (dt, 1H, J = 9.1 Hz, 4.0 Hz), 1.81–1.69

(m, 2H), 1.53–1.47 (m, 1H), 1.30–1.23 (m, 20H), 0.87 (t, 3H, J = 6.3 Hz),

1

(Eluent: 100 % hexane), 0.240 g, 90 %, clear oil: H NMR (500 MHz,

CDCl₃) δ: 2.92–2.88 (m, 2H), 1.56–1.18 (m, 27H), 0.90–0.85 (m, 9H).

¹³C NMR (500 MHz, CDCl₃) δ: 57.4, 39.0, 32.0, 29.7, 29.7, 28.0, 27.9,

27.0, 26.7, 22.8, 22.7, 22.7, and 14.2. IR (neat): ν = 2954, 2923, 2854,

˜

0.85 (d, 6H, J = 6.6 Hz ). ¹³C NMR (500 MHz, CDCl₃) δ: 165.9, 165.8

1466, 1385, 1029, 721 cm^–1. HRMS (ESI) m/z calculated for C₁₉H₃₈

[M + H]: 283.3001, found 283.2985. [α]²_D⁵: +0.54 (c 0.56, CCl₄).

O

13

(

C

- -

¹⁸O), 133.0, 129.9, 129.7, 128.3, 75.5, 75.4 (¹³C_benzoate

benzoate

¹⁸O), 63.7, 38.6, 34.4, 31.8, 29.5, 29.4, 29.3 29.2, 28.9, 27.8, 26.5, 22.6,

(7R, 8S)-7, 8-Epoxy-2-octadecane (+)-1a

22.5, 22.4, 14.0. IR (neat): ν = 2955, 2888, 1727, 1266, 1110, 1080,

˜

1

709 cm^–1. HRMS (ESI) m/z calculated for C₂₆H₃₉ClKD₄¹⁸O₂[M + K]:

467.2888, found 467.2876. [α]²_D⁵: +6.4 (c 0.68, CCl₄).

(Eluent: 100 % hexane), 0.242 g, 91 %, clear oil: H NMR (500 MHz,

CDCl₃) δ: 2.92–2.88 (m, 2H), 1.56–1.18 (m, 27H), 0.90–0.85 (m, 6H).

¹³C NMR (500 MHz, CDCl₃) δ: 57.4, 39.0, 32.0, 29.7, 29.7, 28.0, 27.9,

(7S, 8S)-8-Chloro-2-methyloctadecan-7-yl benzoate (–)-6

27.0, 26.7, 22.8, 22.7, 22.7, and 14.2. IR (neat): ν = 2954, 2923, 2854,

˜

1466, 1385, 1029, 721 cm^–1. [α]_D²⁵: +0.55 (c 0.56, CCl₄).

(7R, 8S)-7, 8-Epoxy-2-methyloctadecane-d₄,¹⁷O (+)-1b

The compound (+)-6 (0.000163 mol) and 4 N NaOH. (Eluent: 100 %

1

Yield 1.10 g, 90 %, colorless oil: H NMR (500 MHz, CDCl₃) δ: 8.09–

8.08 (d, 2H, J = 7.2 Hz), 7.57 (t, 1H, J = 7.4 Hz), 7.48–7.45 (t, 2H, J =

7.7 Hz), 5.31–5.28 (dt, 1H, J = 7.2 Hz, 3.2 Hz), 4.08–4.04 (dt, 1H, J =

9.1 Hz, 4.0 Hz), 1.88–1.80 (m, 4H), 1.69–1.53 (m, 1H), 1.41 (m, 1H),

1.34 (m, 4H), 1.24 (m, 15H), 0.89 (t, 3H, J = 6.3 Hz), 0.85 (d, 6H, J =

6.6 Hz ). ¹³C NMR (500 MHz, CDCl₃) δ: 165.9, 133.0, 130.0, 129.7,

128.3, 75.6, 63.7, 38.7, 34.4, 31.8, 31.3, 29.6, 29.5, 29.4 29.3, 29.2,

1

hexane), 0.042 g, 90 %, clear oil: H NMR (500 MHz, CDCl₃) δ: 2.92–

2.88 (m, 2H), 1.56–1.16 (m, 23H), 0.89–0.86 (m, 9H). ¹³C NMR

13

(500 MHz, CDCl₃) δ: 57.2–57.0 (¹³C_epo-²H &

C

epo

-

¹⁷O), 38.8, 31.8,

29.5, 29.4, 29.2, 27.8, 27.7, 26.5, 22.6, 22.6, 22.5, and 14.0. IR (neat):

29.0, 27.8, 27.1, 26.5, 25.5, 22.6, 22.5, 22.4, 14.0. IR (neat): ν = 2945,

˜

ν = 2954, 2923, 2854, 1466, 1385, 1029, 721 cm^–1. HRMS (ESI) m/z

2880, 1721, 1266, 1108, 1069, 709 cm^–1. HRMS (ESI) m/z calculated

for C₂₆H₄₃ClO₂[M + H]: 424.0792, found 424.2289. [α]²_D⁵: –6.35 (c

0.68, CCl₄).

˜

calculated for C₁₉H₃₅D₄¹⁷O [M + H]: 288.3281, found 288.3288.

[α]²_D⁵: +0.54 (c 0.56, CCl₄).

(7R, 8S)-7, 8-Epoxy-2-methyloctadecane-d₄,¹⁸O (+)-1c

(7S, 8S)-8-Chloro-2-methyloctadecan-7-yl benzoate-d₄, ¹⁷O₂(–)-

6b

The compound (+)-6 (0.000163 mol) and 4 N NaOH. (Eluent: 100 %

hexane), 0.042 g, 89.5 %, clear oil: ¹H NMR (500 MHz, CDCl₃) δ:

The compound (–)-6 (0.00024 mol) and 20 % Pd/C. 0.093 g, 92 %,

colorless oil: ¹H NMR (500 MHz, CDCl₃) δ: 8.09–8.08 (d, 2H, J =

7.2 Hz), 7.60–7.56(t, 1H, J = 7.4 Hz), 7.48–7.44 (t, 2H, J = 7.7 Hz),

5.31–5.27 (m, 1H), 4.08–4.04 (dt, 1H, J = 9.1 Hz, 4.0 Hz), 1.87–1.71

(m, 3H), 1.54–147 (m, 1H), 1.34–1.23 (m, 19H), 0.87 (t, 3H, J = 6.3 Hz),

2.92–2.89 (m, 2H), 1.56–1.18 (m, 23H), 0.90–0.85 (m, 9H). ¹³C NMR

13

(500 MHz, CDCl₃) δ: 57.2–57.0 (¹³C_epo-²H &

C

epo

-

¹⁸O), 39.0, 32.0,

29.7, 29.5, 29.4, 29.2, 27.8, 27.7, 26.5, 22.6, 22.5, 22.4, and 14.0. IR

(neat): ν = 2954, 2923, 2854, 1466, 1385, 1029, 721 cm^–1. HRMS

˜

(ESI) m/z calculated for C₁₉H₃₄D₄Na¹⁸O [M + Na]: 311.3120, found

311.3108. [α]²_D⁵: +0.54 (c 0.56, CCl₄).

0.84 (d, 6H, J = 6.6 Hz ). ¹³C NMR (500 MHz, CDCl₃) δ: 165.9, 165.8

13

(

C

-

¹⁷O), 133.0, 129.9, 129.7, 129.6, 128.3, 75.5, 75.4

¹⁷O), 63.7, 38.6, 34.4, 31.8, 29.5, 29.4, 29.3, 29.2 28.9,

benzoate

(7S, 8R)-7, 8-Epoxy-2-methyloctadecane (–)-1

13

benzoate

1

27.7, 26.5, 22.6, 22.5, 22.4, 14.0. IR (neat): ν = 2945, 2880, 1721, 1266, (Eluent: 100 % hexane), 0.240 g, 90 %, clear oil: H NMR (500 MHz,

˜

1108, 1069, 709 cm^–1. [α]_D²⁵: –6.37 (c 0.68, CCl₄).

CDCl₃) δ: 2.91–2.87 (m, 2H), 1.53–1.19 (m, 27H), 0.89–0.86 (m, 9H).

Eur. J. Org. Chem. 0000, 0–0

www.eurjoc.org

14

Full Paper

¹³C NMR (500 MHz, CDCl₃) δ: 57.2, 57.0, 38.8, 31.8, 29.7, 29.1, 27.9,

Russ. J. Org. Chem. 2007, 72, 3155–3157; j) V. N. Kovalenko et al., Russ.

J. Org. Chem. 2009, 45, 1318–1324; k) A. K. Dudey, A. Chattopadhyay,

Tetrahedron: Asymmetry 2011, 22, 1516–1521; l) W. Zhigang, Z. Jianfeng,

H. Peiqiang, Chin. J. Chem. 2012, 30, 23–28; m) V. Bethi, P. Kattanguru,

R. A. Fernandes, Eur. J. Org. Chem. 2014, 3249–3255.

27.8, 27.7, 27.6, 27.2, 26.7, 26.4, 22.6, 22.5, 22.5, 22.4, 14.2, 14.01. IR

(neat): ν = 2954, 2923, 2854, 1466, 1385, 1029, 721 cm^–1. HRMS (ESI)

˜

m/z calculated for C₁₉H₃₈O [M + H]: 283.3001, found 283.2985.

[α]²_D⁵: –0.537 (c 0.56, CCl₄).

[8] a) L. H. Li, D. Wang, T. H. Chan, Tetrahedron Lett. 1981, 103, 464–466.

[9] E. Keinan, S. C. Sinha, A. Sinhabagchi, Z. M. Wang, X. L. Zhang, K. B.

Sharpless, Tetrahedron Lett. 1992, 33, 6411–6414.

(7S, 8R)-7, 8-Epoxy-2-methyloctadecane- d₄,¹⁷O (–)-1b

[10] See ref.^[7f]

.

The compound (–)-6 (0.000163 mol) and 4 N NaOH. (Eluent: 100 %

hexane), 0.039 mg, 83 %, clear oil: ¹H NMR (500 MHz, CDCl₃) δ:

[11] T. Sato, T. Itoh, T. Fujisawa, Tetrahedron Lett. 1987, 28, 5677–5680.

[12] a) S. J. Hu, S. Jayaraman, A. C. Oehlschlager, J. Org. Chem. 1999, 64, 3719–

3721; b) Y. Garg, A. K. Tiwari, S. K. Pandey, Tetrahedron Lett. 2017, 58,

3344–3346.

2.92–2.89 (m, 2H), 1.56–1.16 (m, 23H), 0.89–0.86 (m, 9H). ¹³C NMR

13

(500 MHz, CDCl₃) δ: 57.2–57.0 (¹³C_epo-²H &

C

epo

-

¹⁷O), 38.8, 31.8,

29.5, 29.4, 29.2, 27.8, 27.7, 26.5, 22.6, 22.6, 22.5, and 14.0. IR (neat):

[13] See ref.^[2]

.

ν = 2954, 2923, 2854, 1466, 1385, 1029, 721 cm^–1. HRMS (ESI) m/z

˜

[14] a) K. Mori, T. Takigawa, M. Matsui, Tetrahedron 1979, 35, 833–837; b) K. R.

Prasad, P. Anbarasan, J. Org. Chem. 2007, 72, 3155–3157.

[15] F. Grater, W. Xu, W. Leal, H. Grubmuller, Structure 2006, 14, 1577–1586.

[16] R. Maida, A. Steinbrecht, G. Ziegelberger, P. Pelosi, Insect Biochem. Mol.

Biol. 1993, 23, 243–253.

calculated for C₁₉H₃₅D₄¹⁷O [M + H]: 288.3281, found 288.3288.

[α]²_D⁵: –0.535 (c 0.56, CCl₄).

(7S, 8R)-7, 8-Epoxy-2-methyloctadecane d₄,¹⁸O (–)-1c

The compound (–)-6 (0.000163 mol) and 4 N NaOH. (Eluent: 100 %

[17] J. Krieger, H. Breer, Science 1999, 286, 720–723.

1

[18] Y. Gong, H. Tang, C. Bohne, E. Plettner, Biochemistry 2010, 49, 793–801.

[19] R. G. Vogt, A. C. Kohne, J. T. Dubnau, G. D. Prestwich, J. Neurosci. 1989,

9, 3332–3346.

[20] M. P. Brochu, S. P. Brown, D. W. C. MacMillan, J. Am. Chem. Soc. 2004,

126, 4108–4109.

[21] M. Amatore, T. D. Beeson, S. P. Brown, D. W. C. MacMillan, Angew. Chem.

Int. Ed. 2009, 48, 5121–5124; Angew. Chem. 2009, 121, 5223.

[22] D. Bernard, A. Doutheau, J. Gore, J. Moulinoux, V. Quemener, J. Chante-

pie, G. Quash, Tetrahedron 1989, 45, 1429–1439.

hexane), 0.041 g, 87 %, clear oil: H NMR (500 MHz, CDCl₃) δ: 2.92–

2.89 (m, 2H), 1.56–1.18 (m, 23H), 0.90–0.85 (m, 9H). ¹³C NMR

13

(500 MHz, CDCl₃) δ: 57.2–57.0 (¹³C_epo-²H &

C

epo

-

¹⁸O), 39.0, 32.0,

29.7, 29.5, 29.4, 29.2, 27.8, 27.7, 26.5, 22.6, 22.5, 22.4, and 14.0. IR

(neat): ν = 2954, 2923, 2854, 1466, 1385, 1029, 721 cm^–1. HRMS

˜

(ESI) m/z calculated for C₁₉H₃₄D₄Na¹⁸O [M + Na]: 311.3120, found

311.3108. [α]²_D⁵: –0.535 (c 0.56, CCl₄).

[23] H. Brinkmann, R. W. Hoffmann, Chemische Ber. 1990, 123, 2395–2401.

[24] J. W. Concellon, H. Rodriguez-Solla, C. Simal, C. Gomez, Synlett 2007,

75–78.

Acknowledgments

This research was supported by Discovery grants from the Natu-

ral Sciences and Engineering Research Council (NSERC) of

Canada # 22923-2010 and 2015-06088, as well as by Discovery

Accelerator Supplement # 477793-2015. We thank Ms. Nathalie

Leva (USDA) for assistance with the field trial in Massachusetts.

[25] B. Kang, R. Britton, Org. Lett. 2007, 9, 5083–5086.

[26] D. G. Ott, In Synthesis with stable isotopes of carbon, nitrogen and oxygen,

John & Wiley Sons, New York, 1981.

[27] G. W. Kabalaka, N. M. Goudgaon, ¹⁷O-enriched methods. In ¹⁷O NMR spec-

troscopy in organic chemistry, CRC Press, Boca Raton, Flo, 1991, p. 21–37.

[28] M. W. Skidmore, Handbook of Derivatives for Chromatography, 2^nded.

John Wiley & Sons, 1993.

[29] J. E. Oliver, R. M. Waters, J. Chem. Ecol. 1995, 2, 199–211.

[30] J. Nardella, M. Terrado, N. Honson, E. Plettner, Arch. Biochem. Biophys.

2015, 579, 73–84.

Keywords: Gypsy moth · Disparlure enantiomers ·

LdisPBPs · Isotope labelling · Mitsunobu inversion

[31] K. Tori, K. Aono, K. Kitahonoki, R. Muneyuki, Y. Takano, H. Tanida, T. Tsuji,

Tetrahedron Lett. 1966, 25, 2921–2926.

[1] B. A. Bierl, M. Beroza, C. W. Collier, Science 1970, 170, 87–89.

[2] S. Iwaki, S. Marumo, T. Satio, M. Yamada, K. Katagiri, J. Am. Chem. Soc.

1974, 96, 7842–7844.

[3] E. Plettner, J. Lazar, E. G. Prestwich, G. D. Prestwich, Biochemistry 2000,

39, 8953–8962.

[32] C. R. Cantor, P. R. Schimmel, Biophysical Chemistry, Part II: Techniques for

the study of biological structure and function. (Freeman, W. H & Co. 1980).

[33] M. Terrado, Y. Yu, E. Plettner, Can. J. Chem. 2018, 96, 168–177.

[34] D. Schneider, W. A. Kafka, M. Beroza, B. A. Bierl, J. Comp. Physiol. 1977,

113, 1–15.

[4] Y. Gong, T. C. S. Pace, C. Castillo, C. Bohne, M. A. O′Neill, E. Plettner, Chem.

Biol. 2009, 16, 162–172.

[35] R. L. Vold, J. S. Waugh, M. P. Kelvin, D. E. Phelps, J. Chem. Phys. 1968, 48,

3831–3832.

[5] Y. Yu, E. Plettner, Bioorg. Med. Chem. 2013, 21, 1811–1822.

[6] J. T. Sanes, E. Plettner, Arch. Biochem. Biophys. 2016, 606, 53–63.

[7] For synthesis of (+)-disparlure over last two decades, see: a) E. Fukusaki,

S. Senda, Y. Nakazono, H. Yuasa, T. Omata, J. Ferment. Bioeng. 1992, 73,

284–286; b) J. L. Brevet, K. Mori, Synthesis 1992, 1007–1012; c) S. Tsuboi,

H. Furutani, M. H. Ansari, T. Sakai, M. Utaka, A. Takeda, J. Org. Chem.

1993, 58, 486–492; d) S. Y. Ko, Tetrahedron Lett. 1994, 35, 3601–3604; e)

A. Sinha-Bagchi, S. C. Sinha, E. Keinan, Tetrahedron: Asymmetry 1995, 6,

2889–2892; f) J. A. Marshall, J. A. Jablonowski, H. J. Jiang, J. Org. Chem.

1999, 64, 2152–2154; g) A. E. Koumbis, D. D. Chronopoulos, Tetrahedron

Lett. 2005, 46, 4353–4355; h) C. X. Zhang, S. J. Da, H. B. Zhang, B. Sun, Y.

Li, Acta Chim. Sinica 2007, 65, 2433–2436; i) K. R. Prasad, P. Anbarasan,

[36] S. Meiboom, D. Gill, Rev. Sci. Instrum. 1958, 29, 688–691.

[37] M. Biasini, S. Bertoni, A. Waterhouse, K. Arnold, G. Studer, T. Schmidt, F.

Kiefer, T. G. Cassarino, M. Bertoni, L. Bordoli, T. Schwede, Nucleic Acids

Res. 2014, 42, 252–258.

[38] T. J. S. Merritt, S. Laforest, G. D. Prestwich, J. M. Quattro, R. G. Vogt,

J. Mol. Evol. 1998, 46, 272–276.

[39] Chemical Computing Group, Molecular Operating Environment, Canada,

Quebec city, C 2011.

[40] E. Kleinpeter, S. Kruger, A. Koch, J. Phys. Chem. A 2015, 119, 4268–4276.

[41] Plimmer et al., Environ. Entomol. 1977, 6, 518–522.

Received: August 6, 2019

Eur. J. Org. Chem. 0000, 0–0

www.eurjoc.org

15

Full Paper

Enantioselective Recognition

G. R. Pinnelli, M. Terrado, N. K. Hillier,

D. R. Lance, E. Plettner* ................. 1–16

Synthesis of Isotopically Labelled

Disparlure Enantiomers and Appli-

cation to the Study of Enantiomer

Discrimination in Gypsy Moth Phe-

romone-Binding Proteins

Disparlure, (7R, 8S)-epoxy-2-methyl- parlure and studies by ²H NMR of

oxtadecane, is the sex pheromone of the interaction between disparlure

the gypsy moth. Its shortest synthesis, enantiomers and the two pheromone-

in high (> 99 %) enantiomeric excess binding proteins found in the anten-

is reported. Furthermore, the incorpo- nae of male gypsy moths in described.

ration of stable isotope labels into dis-

DOI: 10.1002/ejoc.201901164

Eur. J. Org. Chem. 0000, 0–0

www.eurjoc.org

16

Article Doi

Synthesis of Isotopically Labelled Disparlure Enantiomers and Application to the Study of Enantiomer Discrimination in Gypsy Moth Pheromone-Binding Proteins

DOI: 10.1002/ejoc.201901164

Source and publish data:

Authors:

Article abstract of DOI:10.1002/ejoc.201901164

Full text of DOI:10.1002/ejoc.201901164

Products guided by the article

R&D Labs maybe for 54910-51-9

Relevant to this article

Hot Product