Synthesis and purification of [1,2-13C2]coniferin

Article abstract of DOI:10.1002/jlcr.1068

The double labelled lignan precursors [1,2-¹³C ₂]coniferin and the glucoside of [1,2-¹³C ₂]ferulic acid were prepared by classical synthetic methods. Pure double labelled lignan precursors could only be obtained

Full text of DOI:10.1002/jlcr.1068

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2006; 49: 463–470.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.1068
Research Article
Synthesis and purification of [1,2-13C2]coniferin
Vickram Beejmohun1,2, Eric Grand1, David Lesur1, Franc
Marc-Andre
Fliniaux2 and Jose Kovensky1,*
Laboratoire des Glucides CNRS UMR 6219, Faculte des Sciences, Universite de
Picardie Jules Verne, 33 rue Saint-Leu, 80039 Amiens, France
¸
ois Mesnard2,
´
´
1
´
´
2
´ ´
Laboratoire de Phytotechnologie, Faculte de Pharmacie, Universite de Picardie Jules
Verne, 1 rue des Louvels, 80037 Amiens, France
Summary
The double labelled lignan precursors [1,2-13C2]coniferin and the glucoside of
[1,2-13C2]ferulic acid were prepared by classical synthetic methods. Pure double
labelled lignan precursors could only be obtained after separation from their
contaminating Z-isomers and dihydro by-products by high-performance liquid
chromatography. Copyright # 2006 John Wiley & Sons, Ltd.
Key Words: monolignols; synthesis; isotopic labelling; carbon 13
Introduction
Lignans are widely distributed plant metabolites associated with a range of
biological activities. Podophylotoxin is used as an antiviral agent in the
treatment of genital warts,1 and secoisolariciresinol diglucoside is a natural
cancer chemopreventive agent effective against the onset of breast, prostate,
and colon cancers.2 The biosynthetic pathway of lignans is not completely
elucidated, as it is specific to each species of plants.3 Most of the present
knowledge about the conversion of monolignols to their dimers lignans and
the formation of cell wall lignins in plants comes from incorporation
experiments using specifically labelled precursors or substrates.4,5
Monolignol glucosides (p-glucocoumaryl alcohol, coniferin, syringin)
are considered to be the storage or excretion form of monolignols.6
The synthesis of monolignol glucosides enriched with 13C at specific positions
has been used to prove their effective incorporation in lignin.7 However,
*Correspondence to: Jose
Universite de Picardie Jules Verne, 33 rue Saint-Leu, 80039 Amiens, France. E-mail: jose.kovensky@
u-picardie.fr
Kovensky, Laboratoire des Glucides CNRS UMR 6219, Faculte des Sciences,
´ ´
´
Contract/grant sponsor: Conseil Regional de Picardie (Pole IBFBio)
´
Copyright # 2006 John Wiley & Sons, Ltd.
Received 28 February 2006
Revised 7 March 2006
Accepted 7 March 2006

464
V. BEEJMOHUN ET AL.
to obtain detailed information about the biosynthetic route of lignans in vivo,
the synthesis of multiple labelled precursors becomes necessary.4 The
presence of two consecutive 13C atoms in the monolignol molecules
allows their conversion to dimers to be followed, due to the characteristic
signals observed in the NMR spectra of intermediates involved. We chose the
two atom carbons mostly involved in the dimerization process leading to
lignans.
In this paper, we report on the synthesis, chromatographic purification and
complete characterization of the labelled monolignol glucoside [1,2-13C2]con-
iferin, and its precursor, the glucoside of [1,2-13C2]ferulic acid.
Results and discussion
The synthetic strategy8 is shown in Scheme 1. Conversion of vanillin (1)
to its corresponding b-glucoside was performed with tetra-O-acetyl-a-d-
glucopyranosyl bromide and silver oxide in quinoline to give 2 in 73% yield.
The introduction of the double 13C labelling was accomplished by condensa-
tion of 2 with readily available [1,2,3-13C3]malonic acid leading, after
decarboxylation, to compound 3 in 70% yield. The typical pattern of
consecutive labelling was observed in the 13C NMR spectrum of compound 3
in CDCl3, the two doublets appear at d 171.5 (C-1) and d 116.6 (C-2) with a
J1;2 ¼ 74 Hz:
Since the glucoside of ferulic acid is also an interesting substrate for
biosynthetic studies, we decided to perform the deprotection of the sugar
moiety in NaOMe–methanol. Deacetylation and purification by flash
chromatography on silica gel lead to the acid 4. Disappointingly, the
13C NMR spectrum of deacetylated glucoside of labelled acid 4 in CD3OD
showed the presence of a second product, its Z-isomer, which had not been
detected on the NMR spectrum of the protected acid 3. Previous synthesis of
unlabelled coniferin using the same synthetic route9 did not report about the
formation of the Z-isomer, however in our products the presence of the double
labelling allows its easy detection by 13C NMR. On the other hand, as the Z-
isomer has been previously obtained by other methods,10,11 the additional
NMR doublets could be undoubtedly assigned. For the major E-acid two
doublets appear at d 170.7 (C-1) and d 117.9 (C-2) with a J1;2 ¼ 73:2 Hz
whereas the corresponding resonances of the minor Z-acid are d 170.8 (C-1)
and d 120.3 (C-2) with a J1;2 ¼ 71:2 Hz:
It is important at this stage to separate the two isomers to avoid
contamination of our biosynthetic studies by the less common Z-isomer. All
attempts to perform separation of E- and Z-isomers before deacetylation
(compound 3) were unsuccessful. We decided to purify the deacetylated acid 4.
Using a Prevail C-18 column and a gradient of water (containing 0.2% of
acetic acid)–acetonitrile (Figure 1(a)), first was eluted the natural glucoside of
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 463–470

SYNTHESIS AND PURIFICATION OF [1,2-13C2]CONIFERIN
465
Scheme 1. Synthesis of labelled ferulic acid and coniferin. Reagents: (i) Tetra-O-
acetyl-a-d-glucopyranosyl bromide, Ag2O, quinoline, r.t., 2 h; (ii)
13CH2(13COOH)2, pyridine, piperidine, 1 h at 558C and 2 h at 858C; (iii)
MeONa, MeOH, r.t., overnight; (iv) (COCl)2, CH2Cl2, r.t., overnight; (v)
NaBH4, THF, r.t., 4 h; (vi) NH3, MeOH, r.t., overnight
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 463–470

466
V. BEEJMOHUN ET AL.
Figure 1. (a) HPLC elution profile of deacetylated glucoside of labelled ferulic
acid 4. Prevail C-18 column, gradient water (containing 0.2% of acetic acid)–
acetonitrile. tR 14.40 min=E-ferulic acid glucoside, tR 16.23 min=Z-isomer,
k=280 nm. (b) HPLC elution profile of labelled coniferin 5. Kromasyl C-18
column, gradient water (containing 0.2% of acetic acid)-acetonitrile. tR
39.65 min=E-coniferin, tR 41.28 min=dihydroconiferin, tR 43.67 min=Z-
isomer, k=280 nm
E-ferulic acid, then the contaminant Z-isomer (tR 14.40 and 16.23 min,
respectively). After purification, the 13C NMR spectra of 4 showed only the
doublets corresponding to the desired E-isomer.
The synthesis of the labelled monolignol from the protected acid 3 was
performed through the acyl chloride derivative. As we have previously
reported,8 after deacetylation, the desired alcohol 5 was obtained, together
with the reduced double bond by-product 6.
As the pure E-acid could only be purified by HPLC in their deacetylated
form, we needed to perform a reacetylation step before its conversion into the
acyl chloride and reduction to alcohol. Unfortunately, some degradation was
observed during the reacetylation step, and in addition, traces of pyridine or
acetic anhydride resulted in low yields of the following borohydride reduction
step. Therefore, the acetylated acid 3 (E/Z mixture) obtained before was
directly used for reduction. In spite of different chromatographic conditions,
only two peaks were observed for the acetylated alcohol prepared. MS and
NMR analysis showed that they corresponded to the Z-alcohol and an
inseparable mixture of the E-alcohol and the dihydro derivative, respectively.
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 463–470

SYNTHESIS AND PURIFICATION OF [1,2-13C2]CONIFERIN
467
As the complete separation was not possible at this stage, the crude alcohol
was deacetylated. Separation of E- and Z-isomers of some monolignol
glucosides has been previously reported using two columns in series,12 but
these conditions were unsuccessful for obtaining pure E-isomers without the
dihydrogenated by-products. Reversed-phase HPLC on a Kromasyl C-18
column allowed us to obtain the three components on different peaks of the
elution profile, as it is shown for the labelled coniferin 5 (Figure 1(b)): first was
eluted the E-isomer, then the dihydrogenated by-product, and finally the
Z-isomer (tR 39.65, 41.28 and 43.67 min, respectively).
In the 13C NMR spectra of the mixture obtained, the major product (E)
presents two doublets at d 128.9 (C-2) and 63.7 (C-1) with a J1;2 ¼ 47:4 Hz:
The Z-isomer, at d 131.9 (C-2) and 59.9 (C-1) with a J1;2 ¼ 46:3 Hz; and the
dihydrogenated by-product showed two resonances at d 62.2 (C-1) and 35.5
(C-2) with a J1;2 ¼ 37:6 Hz: The 13C NMR spectra of the pure [1,2-13C2]
coniferin 5 showed only the expected two typical doublets.
Experimental
Melting points were determined on a Buchi 535. Optical rotations were
¨
measured with a Perkin Elmer Model 343 polarimeter using a sodium lamp at
208C. High-resolution electrospray mass spectra in the positive ion mode were
obtained on a Q-TOF Ultima Global hybrid quadrupole/time-of-flight
instrument (Waters-Micromass, Manchester, UK), equipped with a pneuma-
tically assisted electrospray (Z-spray) ion source and an additional sprayer
1
(Lock Spray) for the reference compound. H NMR spectra were recorded
with a Bruker AC 500 spectrometer, 13C NMR spectra were recorded with a
Bruker AC 300 spectrometer.
HPLC
The instrumentation in these purifications employed a Waters Model Prep LC
4000 solvent delivery system and a Waters Model 2487 dual l absorbance.
Preparative chromatographic separation of deacetylated acids used a Prevail
C18 (250 mm ꢀ 22 mm, 5 mm, stainless steel) column. Elution was performed
with water (containing 0.2% of acetic acid)–acetonitrile, linear gradient
100:0 – 82:18 (10 min) and kept for additional 20 min. The flow rate was
18 ml minꢁ1, and detection at 254 and 280 nm. Preparative chromatographic
separation of deacetylated alcohols used a Kromasyl C18 (250 mm ꢀ 10 mm,
5 mm, stainless steel) column. Elution was performed with water (containing
0.2% of acetic acid)–acetonitrile, linear gradient 100:0–86:14 (54 min). The
flow rate was 4.7 ml minꢁ1, and detection at 254 and 280 nm.
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 463–470

468
V. BEEJMOHUN ET AL.
40-O-(200,300,400,600-Tetra-O-acetyl-b-d-glucopyranosyl)vanillin (2)
Vanillin (0.222 g, 1.46 mmol) and acetobromoglucose (1.026 g, 2.5 mmol) were
dissolved in distilled quinoleine (10 ml) and silver oxide (0.336 g, 1.7 mmol)
was added in portions at 08C. The reaction mixture was stirred for 2 h in the
dark at room temperature. After the addition of acetic acid (20 ml), the whole
mixture was poured into water (200 ml) to form a precipitate, which was
collected by filtration. Purification by silica gel column chromatography (3:7
cyclohexane–ethyl acetate) gave 2 (0.513 g, 73% yield) as a white solid: m.p.
143.2–144.98C (lit.13 143–1448C).
[1,2-13C2]-40-O-b-d-glucopyranosylferulic acid (4)
A mixture of 2 (0.5 g, 1 mmol), [1,2,3-13C3]malonic acid (0.155 g, 1.3 mmol),
pyridine (3.5 ml) and piperidine (four drops) was heated at 558C for 1 h then at
858C for 2 h. After cooling, water (5 ml) was added and crystals of [1,2-13C2]-
40-O-(200,300,400,600-tetra-O-acetyl-b-d-glucopyranosyl)-ferulic acid (3) were
obtained by acidification with HCl. Pyridine and piperidin were co-evaporated
with toluene. Purification by silica gel column chromatography (6:4–3:7
cyclohexane–ethyl acetate) gave 3 (0.398 g, 73% yield) as a white solid.
Compound 3 was dissolved in a NH3 solution (7 N in methanol) and was
stirred for 12 h. The solvent was removed under reduced pressure and the
residue was dissolved in water and then lyophilized. A mixture of two isomers
of 4, E/Z was obtained (0.252 g, 93% yield), which were separated by reverse
phase HPLC (Prevail C18, 5 m, 250 ꢀ 22 mm, water 0.2% acetic acide:acetoni-
trile), to give pure 4, as a white solid: m.p. 141.5–143.58C, [a]D-7.8 (c 0.62,
1
00
00 00
CH3OH). H NMR (500 MHz, CD3OD): d (ppm) 3.40 (dd, 1 H, H4 , J4 ;3
¼
00
00 00
00 00
00 00
8:3 Hz; J4 ;5 ¼ 9:6 Hz); 3.45 (ddd, 1 H, H5 , J4 ;5 ¼ 9:6 Hz; J5 ;6 a ¼ 5:3 Hz;
00
00 00
00 00
00 00
J5 ;6 b ¼ 2:2 Hz); 3.48 (dd, 1 H, H3 , J3 ;2 ¼ 9:3 Hz; J3 ;4 ¼ 8:3 Hz); 3.52
00
00
00 00
00 00
(dd, 1 H, H2 , J2 ;1 ¼ 7:4 Hz; J2 ;3 ¼ 9:3 Hz); 3.70 (dd, 1 H, H6 a
,
00
00
00
00
00
00
00
J6 a;6 b ¼ 12:1 Hz; J6 a;5 ¼ 5:3 Hz); 3.88 (dd, 1 H, H6 b, J6 b;5 ¼ 2:2 Hz;
00
00
00
00 00
J6 b;6 a ¼ 12:1 Hz); 3.90 (s, 3 H, OCH3); 4.97 (d, 1 H, H1 , J1 ;2 ¼ 7:4 Hz); 6.39
(ddd, 1 H, H2, J2;3 ¼ 15:9 Hz; J2;C2 ¼ 161:0 Hz; J2;C1 ¼ 2:6 Hz); 7.15 (dd, 1 H,
0
0
0
0
0
0
0
0
H6 , J6 ;5 ¼ 8:4 Hz; J6 ;2 ¼ 1:7 Hz); 7.18 (d, 1 H, H5 , J5 ;6 ¼ 8:4 Hz); 7.24 (d,
0
0
0
1 H, H2 , J2 6 ¼ 1:7 Hz); 7.61 (ddd, 1 H, H3, J3;2 ¼ 15:9 Hz; J3;C2 ¼ 2:8 Hz;
J3;C1 ¼ 6:7 Hz). 13C NMR (75.5 MHz, CD3OD): d (ppm) 56.8 (CH3O); 62.5
00
00
00
00
00
00
(C6 ); 71.3 (C4 ); 74.8 (C2 ); 77.9 (C3 ); 78.3 (C5 ); 102.2 (C1 ); 112.5 (d, C2 ,
0
0
0
0
JC2 ;C2 ¼ 4:5 Hz); 117.9 (d, C2, C5 , JC2;C1 ¼ 73:2 Hz); 123.4 (d, C6 ,
0
0
0
JC6 ;C2 ¼ 5:3 Hz); 130.6 (d, C1 , JC1 ;C1 ¼ 7:0 Hz); 146.0 (d, C3,
0
0
0
JC3;C2 ¼ 70:6 Hz); 151.0 (d, C4 , JC4 ;C2 ¼ 1:2 Hz); 151.0 (C3 ); 170.7 (d, C1,
13C2H20O9Na :
12
JC1;C2 ¼ 73:2 Hz). ESI-HRMS ½M þ Naꢂ: Calculated for
m/z 381.1072. Found: 381.1075.
C
14
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 463–470

SYNTHESIS AND PURIFICATION OF [1,2-13C2]CONIFERIN
469
[1,2-13C2]Coniferin (5)
Compound 3 (0.300 g, 0.57 mmol) was dissolved in dried dichlorome-
thane (25 ml) under argon and cooled in an ice water bath. Oxalyl chloride
(0.15 ml) was added dropwise. After addition, the ice water bath was
removed and the reaction was left for 12 h at room temperature. The solvent
was removed under reduced pressure to give a paleyellow precipitate.
Sodium borohydride (96 mg) was dissolved in dried THF (20 ml) and
[1,2-13C2]-40-O-(200,300,400,600-tetra-O-acetyl-b-d-glucopyranosyl)feruloyl chlor-
ide was slowly added dissolved in dried THF (20 ml) under stirring
under argon. The reaction mixture was left at room temperature for 5 h.
The completion of the reaction was confirmed by mass spectrometry.
Methanol was added to react with the excess of sodium borohydride and
then the solvents were removed under reduced pressure. Several evaporations
with methanol gave a residue which was purified by column chromatography
on silica gel (4:6 cyclohexane–ethylacetate). A mixture of [1,2-13C2]-
200,300,400,600-tetra-O-acetylconiferin (0.271 g, 70% yield), the isomer Z and
[1,2-13C2]-200,300,400,600-tetra-O-acetyldihydroconiferin was obtained. The mix-
ture (0.21 g) was dissolved in a NH3 solution (7 N in methanol) (10 ml) and was
stirred for 12 h. The solvent was removed under reduced pressure without
heating in absence of light, then the residue was redissolved in water and
lyophilized. A mixture of 5 (0.178 g, 92% yield) and two isomers: the isomer Z
and dihydroconiferin 6 was obtained. The three compounds could be
separated by reverse-phase HPLC: (Prevail C18, 5 m, 250 ꢀ 22 mm, water
0.2% acetic acide:acetonitrile) to give pure 5, as a white solid: m.p. 189.3–
1
190.58C, [a]D-1.4 (c 0.7, CH3OH). H NMR (500 MHz, CD3OD): d (ppm)
00
00
00
00 00
3.43–3.37 (m, 2 H, H4 , H5 ); 3.46 (dd, 1 H, H3 , J3 ;2 ¼ 9:1 Hz;
00
00 00
00 00
00 00
J3 ;4 ¼ 10:1 Hz); 3.50 (dd, 1 H, H2 , J2 ;1 ¼ 7:3 Hz; J2 ;3 ¼ 9:1 Hz); 3.69
00
00
00
00
00
(m, 1 H, H6 a, J6 a;6 b ¼ 12:1 Hz; J6 a;5 ¼ 5:1 Hz); 3.87 (s +dd, 4 H, OMe,
00
H6 b
00
00
00
00
,
J6 b;5 ¼ 1:6 Hz; J6 b;6 a ¼ 12:1 Hz); 4.21 (dddd,
2
H,
2
H1,
J1;C1 ¼ 141:5 Hz; J1;3 ¼ 1:4 Hz; J1;2 ¼ 5:7 Hz; J1;C2 ¼ 4:5 Hz); 4.89 (d, 1 H,
00
00 00
H1 , J1 ;2 ¼ 7:3 Hz); 6.28 (dddt, 1 H, H2, J2;3 ¼ 16:1 Hz; J2;C2 ¼ 151:5 Hz;
J2;1 ¼ 5:7 Hz; J2;C1 ¼ 3:9 Hz); 6.55 (ddl, H, H3, J3;2 ¼ 16:1 Hz;
1
0
0
0
0
0
J3;C1 ¼ 7:0 Hz); 6.95 (dd, 1 H, H6 , J6 ;5 ¼ 8:4 Hz; J6 ;2 ¼ 1:9 Hz); 7.07 (d,
1 H, H2 , J2 6 ¼ 1:9 Hz); 7.12 (d, 1 H, H5 , J5 ;6 ¼ 8:4 Hz). 13C NMR
0
0
0
0
0
0
00
(75.5 MHz, CD3OD): d (ppm) 56.7 (CH3O); 62.5 (C6 ); 63.7 (d, C1,
00
00
00
00
00
JC1;C2 ¼ 47:4 Hz); 71.3 (C4 ); 74.9 (C2 ); 77.8 (C3 ); 78.2 (C5 ); 102.7 (C1 );
0
0
0
0
0
111.3 (d, C2 , JC2 ;C2 ¼ 4:5 Hz); 117.8 (C5 ); 120.7 (d, C6 , JC6 ;C2 ¼ 5:0 Hz);
128.9 (d, C2, JC2;C1 ¼ 47:4 Hz); 131.3 (dd, C3, JC3;C2 ¼ 72:8 Hz;
0
0
0
JC3;C1 ¼ 3:5 Hz); 133.6 (d, C1 , JC1 ;C1 ¼ 6:3 Hz); 147.6 (d, C4 ,
12
13
0
0
JC4 ;C2 ¼ 1:1 Hz); 150.8 (C3 ). ESI-HRMS ½M þ Naꢂ: Calculated for
C2H22O8Na : m/z 367.1279. Found: 367.1265.
C
14
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 463–470

470
V. BEEJMOHUN ET AL.
Conclusion
During our synthesis of the double labelled glucoside of [1,2-13C2]ferulic acid
and [1,2-13C2]coniferin, we could detect the formation of the Z-isomer in the
condensation step of vanillin with labelled malonic acid, and the partial
hydrogenation of the double bond as a side reaction in the reduction step.
For both lignans precursors, a purification by reversed-phase chromatography
was necessary to avoid contamination by their double labelled synthetic by-
products during plant inoculation.
Acknowledgements
This work was carried out with the financial contribution of the Conseil
Regional de Picardie (Pole IBFBio). V.B. thanks the Conseil Regional de
´ ´
Picardie for a fellowship.
References
1. Beutner KR, von Krogh G. Semin Dermatol 1990; 9: 148–151.
2. Thompson LU, Seidl MM, Rickard SE, Orcheson LJ, Fong HH. Nutr Cancer
1996; 26: 159–165.
3. Croteau R, Kutchan TM, Lewis NG. Natural products (secondary metabolites).
In Biochemistry & Molecular Biology of Plants, Buchanan B, Gruissem W,
Jones R (eds). American Society of Plant Physiologists: Rockville, Maryland,
USA, 2000; 1250–1302.
4. Seidel V, Windhovel J, Eaton G, Alfermann AW, Arroo RRJ, Medarde M,
¨
Petersen M, Woolley JG. Planta 2002; 215: 1031–1039.
5. Ford JD, Huang K-S, Wang H-B, Davin LB, Lewis NG. J Nat Prod 2001; 64:
1388–1397.
6. Gross GG. Biosynthesis and metabolism of phenolic acids and monolignols. In
Biosynthesis and Biodegradation of Wood Components, Higuchi T (ed.). Academic
Press: New York, 1985; 229–271.
7. Terashima N, Seguchi Y, Robert D. Holzforschung 1991; 45: 35–39.
8. Beejmohun V, Grand E, Mesnard F, Fliniaux M-A, Kovensky J. Tetrahedron
Lett 2004; 45: 8745–8747.
9. Terashima N, Ralph SA, Landucci LL. Holzforschung 1995; 50: 151–155.
10. Rasmussen S, Wolff C, Rudolph H. Phytochemistry 1996; 42: 81–87.
11. Wu J, Zhang S, Huang J, Xiao Q, Li Q, Long L, Huang L. Chem Pharm Bull
2003; 51: 1201–1203.
12. Lewis NG, Inciong MEJ, Dhara KP, Yamamoto E. J Chromatogr 1989; 479:
345–352.
13. Fischer E, Raske K. Ber. Deutsch Chem.Ges. 1909; 42: 1465–1476.
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 463–470