Efficient synthesis and biological evaluation of ω-oxygenated analogues of vitamin K2: Study of modification and structure-activity relationship of vitamin K2 metabolites

Article abstract of DOI:10.1016/j.bmcl.2006.12.082

Novel ω-oxygenated vitamin K₂ analogues, which are candidates for metabolites of vitamin K₂ homologues, were efficiently synthesized and their apoptosis-inducing activity was evaluated. We revealed that some of those analogues were biologically active and the side-chain part played an important role in apoptosis-inducing activity. Our results can provide useful information to develop the structure-activity relationship of vitamin K₂ analogues for new drugs based on vitamin K.

Full text of DOI:10.1016/j.bmcl.2006.12.082

Bioorganic & Medicinal Chemistry Letters 17 (2007) 1622–1625
Efficient synthesis and biological evaluation of x-oxygenated
analogues of vitamin K2: Study of modification and structure–
activity relationship of vitamin K2 metabolites
Yoshitomo Suhara, Aya Murakami, Maya Kamao, Shino Mimatsu,
Kimie Nakagawa, Naoko Tsugawa and Toshio Okano*
Department of Hygienic Sciences, Kobe Pharmaceutical University, 4-19-1 Motoyamakita-machi,
Higashinada-ku, Kobe 658-8558, Japan
Received 2 November 2006; revised 18 December 2006; accepted 23 December 2006
Available online 4 January 2007
Abstract—Novel x-oxygenated vitamin K2 analogues, which are candidates for metabolites of vitamin K2 homologues, were effi-
ciently synthesized and their apoptosis-inducing activity was evaluated. We revealed that some of those analogues were biologically
active and the side-chain part played an important role in apoptosis-inducing activity. Our results can provide useful information to
develop the structure–activity relationship of vitamin K2 analogues for new drugs based on vitamin K.
Ó 2007 Elsevier Ltd. All rights reserved.
Vitamin K is an essential nutrient and a cofactor for the
carboxylation of specific glutamyl residues of proteins to
c-glutamyl residues, which are vital to the blood coagu-
lation system.1,2 Vitamin K-dependent carboxylase was
originally observed in the liver, and it has subsequently
been detected in other tissues.3 In bone, c-carboxylated
osteocalcin is known to be involved in the bone remod-
eling system, and consequently, vitamin K insufficiency
may cause related diseases such as osteoporosis.4–6
There are two major forms of vitamin K analogues.
Examples include the plant-derived vitamin K1 (1)
(phylloquinone, PK) and the bacterium-derived vitamin
K2 (2) (menaquinone, MK) (Fig. 1). Among these
homologues, MK-4 is the most potent coagulation
cofactor for c-carboxylase with hemostatic activity in
comparison with PK and other menaquinones.
excreted glucuronides of 10 and 11 as well as those of
x-carboxylic acid (9) (Fig. 2).8,9 However, x-alcohol
and x-aldehyde metabolites, which are the first step to
glucronide conversion, have not been detected yet,
although it was suggested that metabolites were formed
by side-chain degradation via x- and b-oxidation as
aforementioned.
In this study, we focused on metabolites of vitamin K2
homologues because MK-4 has unique biological activ-
ities among menaquinones. In fact, some groups recent-
ly reported that MK-4 showed anti-arteriosclerosis
activity10 and anti-tumor actions in various cancer
cells.11–13 Furthermore, it was revealed that MK-4
works as an SXR-specific ligand and regulates gene
transcription.14 Quite recently, MK-4 binding protein
was identified; in short, 17b-hydroxysteroid dehydroge-
nase 4 concerning estradiol production might be a
Until recently, the metabolism of vitamin K has been
studied in humans and rats. For example, the adminis-
tration of radiolabeled phylloquinone to healthy male
volunteers led to the detection of urinary metabolites
as K acid I (10) and K acid II (11).7 In addition, rats
fed a diet containing [14C]-labeled menaquinone-4
O
O
O
O
H
3
H
n
Keywords: Vitamin K; Metabolite; Apoptosis; Analogue; Menaquin-
one-4.
1
2
*
Figure 1. Structure of vitamin K homologues: phylloquinone (1) and
menaquinones (2).
Corresponding author. Tel.: +81 78 441 7563; fax: +81 78 441
7565; e-mail: t-okano@kobepharma-u.ac.jp
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.12.082

Y. Suhara et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1622–1625
1623
O
The primary hydroxyl group of 12a–c was first protected
with an acetyl group to give 13a–c in quantitative yield.
Then, the selective x-oxygenation of 13a and 13b with
SeO2, 70% tert-BuOOH solution, and salicylic acid in
CH2Cl2 afforded x-oxygenated alcohol 14a and 14b with
x-aldehyde analogue in good yield.16 Hence, 13c in the
same condition was converted to x-oxygenated 14c as
well as x-1-oxygenated and diol derivatives. A signifi-
cant reason for the low chemical yield was that selective
reactivity of SeO2 was reduced due to a longer alkyl
chain length. Then, separation of the mixture by silica
gel column chromatography gave the desired 14c in
20% yield. Protection of the hydroxyl group of 14a–c
with 3,4-dihydro-2H-pyran (DHP) and TsOH in
CH2Cl2 gave 15a–c in excellent yield. Removal of the
protecting group in 15a–c gave the corresponding alco-
hol 16a–c in 78–82%.
O
O
COOH
OH
n
O
3
9
3: (n= 2)
4: (n= 3)
5: (n= 4)
O
COOH
O
O
O
O
10
11
CHO
n-1
COOH
6: (n= 2)
7: (n= 3)
8: (n= 4)
O
Figure 2. Our synthesized compounds 3–8 and chemical structures of
the known metabolites of menaquinone-4, x-COOH (9), K acid I (10),
K acid II (11).
Scheme 2 shows the synthesis of x-oxygenated vitamin
K analogues. Vitamin K4 monoacetate (17)17 was used
as a naphthoquinone synthon, which condensed with
side-chain analogues. Treatment of the monoacetate 17
with an alkyl side-chain alcohol 16a–c in the presence
of boron trifluoride etherate yielded 18a–c in 45–51%
yield. We first tried to obtain x-alcohol compounds
3–5 directly by alkaline hydrolysis, but the chemical
yield was extremely low. Presumably, the quinone form
of 18a–c might be degraded under alkaline conditions;
therefore, we employed a two-step synthesis by way of
dimethyl ether analogues 19a–c. In short, monoacetate
derivatives 18a–c were treated with an excess amount
of potassium hydride in anhydrous conditions followed
by the addition of methyl iodide to give 19a–c in good
yield. Finally, oxidation and deprotection of the THP
group of naphthohydroquinone methyl ether 19a–c with
Ce(NH4)2(NO3)6 in water gave the desired x-oxygenat-
ed vitamin K homologues 3–5 in good yield. These
analogues were further converted to x-aldehyde
analogues 6–8 with PDC treatment in good yield. Thus,
six kinds of x-oxygenated analogues were prepared.18
candidate;15 however, the biological activities of these
metabolites have hardly been evaluated so far. We antic-
ipated that if those metabolites were studied in detail,
they would provide insight into the biological signifi-
cance of vitamin K and valuable information for the
development of new drugs. We were interested in x-car-
boxylate 9 as a metabolite of MK-4 at first, because the
terminal functional group of an unsaturated alkyl side-
chain would induce various biological functions such
as geranylgeraniol, farnesol, and acyclic retinoid, but
we found that 9 was quite unstable and degraded unless
the conjugate was formed. For this reason, we took par-
ticular note of ‘stable’ x-alcohol and x-aldehyde men-
aquinones, which were precursors of x-carboxylated
menaquinones. In this paper, we present the synthesis
of six kinds of x-oxygenated menaquinone analogues
3–8 and initial studies on the metabolite of vitamin K
homologues.
We first planned the synthesis of x-oxygenated men-
aquinones. The requisite analogues were synthesized
by coupling the naphthoquinone derivative and side-
chain moiety. Scheme 1 outlines the synthesis of side-
chain synthons of x-oxygenated MK-2, MK-3, and
MK-4 analogues. We chose geraniol (12a), farnesol
(12b), and geranylgeraniol (12c) as the starting material.
To investigate the biological activities of those ana-
logues, we especially focused on the apoptosis-inducing
activity against cancer cell lines among their various
biological activities. MK-4 is used as an anti-hepatic
cancer drug in clinical practice; therefore, we believe
OAc
OAc
OMe
a
b
AcO
a
HO
AcO
OTHP
n
OTHP
n
b
OH
n
H
n
H
n
OH
OH
OMe
17a: (n= 2)
17b: (n= 3)
17c: (n= 4)
18a: (n= 2)
18b: (n= 3)
18c: (n= 4)
19a: (n= 2)
19b: (n= 3)
19c: (n= 4)
12a (n= 2): Geraniol
12b (n= 3): Farnesol
12c (n= 4): Geranylgeraniol
13a: (n= 2)
13b: (n= 3)
13c: (n= 4)
14a: (n= 2)
14b: (n= 3)
14c: (n= 4)
O
O
O
O
AcO
c
HO
d
d
c
OTHP
OTHP
CHO
OH
n
n
n
n-1
15a: (n= 2)
15b: (n= 3)
15c: (n= 4)
16a: (n= 2)
16b: (n= 3)
16c: (n= 4)
6: (n= 2)
7: (n= 3)
8: (n= 4)
3: (n= 2)
4: (n= 3)
5: (n= 4)
Scheme 1. Reagents: (a) pyridine, Ac2O, 92–95%; (b) SeO2, 70%
tert-BuOOH, salicylic acid, CH2Cl2, 40–75%; (c) DHP, TsOH,
CH2Cl2, 75–81%; (d) 1 N NaOH aq, MeOH, 78–82%.
Scheme 2. Reagents: (a) 16a–c, BF3ÆEt2O, EtOAc/dioxane (1:1),
45–51%; (b) KH, CH3I, THF, 60–65%; (c) CAN, CH3CN–H2O,
71–76%, (d) PDC, CH2Cl2 35–55%.

1624
Y. Suhara et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1622–1625
preserve the nature of vitamin K analogues rather than
alkyl side-chain analogues. We previously reported the
synthesis of new vitamin D analogues, which introduced
the x-hydroxyl alkyl group to the molecule.21 Some had
potent binding affinity to vitamin D receptor and this
depended on the length of the alkyl group. Similar to
this case, the interaction of compound 5 with the target
protein increased. Our results indicated that metabolites
of vitamin K might have potential as biologically active
compounds and can provide useful information to
develop new drugs based on vitamin K with modifica-
tion of the terminal alkyl group.
Acknowledgments
Figure 3. Apoptosis-inducing activity of menaquinone derivatives in
HL-60 cells. HL-60 cells were treated with menaquinone analogues at
10 lM (j) and 5 lM ( ) for 3 days. Apoptosis-inducing activity was
identified with flow cytometric analysis of DNA content. Results
represent the mean and standard errors of three separate experiments.
x-Aldehyde analogues 6–8 did not exhibit apoptosis-inducing activity.
We are grateful to Dr. Makiko Sugiura and Dr. Atsuko
Takeuchi for the spectroscopic measurements. We are
also grateful to Eisai Co., Ltd. for the gift of vitamin
K standards. This study was supported in part by a
Grant-in-Aid for Scientific Research <KAKENHI>
from the Japan Society for the Promotion of Science.
that vitamin K analogues can be applied to anti-cancer
drugs. Regarding the cancer cell lines, we chose HL-60
cells (human leukemia cells) for preliminary study since
they are the most common and convenient to evaluate
the apoptosis-inducing activity of biologically active
agents.19 Samples of 5 and 10 lM analogues were added
to HL-60 cells. After 72 h, the cells were collected and
treated with pyridinium iodide, and then apoptosis
activity was determined with a fluorescence-activated
cell sorter (FACS). As shown in Figure 3, the activity
of most analogues increased in a dose-dependent man-
ner. MK-3 particularly showed the most potent activity,
around 75%, among vitamin K2 homologues at 10 lM.
On the other hand, the potency of MK-4 was approxi-
mately 15%, which equaled 20% of MK-3 at the same
dose; however, MK-2 and x-aldehyde analogues 6–8
did not exhibit such activity. In terms of x-oxygenated
alcohol 3–5, x-oxygenated MK-3 (4) decreased its
potency compared to MK-3; however, it retained about
50% potency of MK-3. Interestingly, only x-oxygenated
MK-4 (5) increased the activity more than substrate
MK-4 at 10 lM.
References and notes
1. Stenflo, J.; Fernlund, P.; Egen, W.; Roepstorff, P. Proc.
Natl. Acad. Sci. U.S.A. 1974, 71, 2730.
2. Nelsestuen, G. L.; Zytkovicz, T. H.; Howard, J. B. J. Biol.
Chem. 1974, 249, 6347.
3. Vermeer, C.; Hendrix, H.; Daemen, M. FEBS Lett. 1982,
148, 317.
4. Suttie, J. W. Hepatology 1987, 7, 367.
5. Vermeer, C. Biochem. J. 1990, 266, 625.
6. Shearer, M. J. Blood Rev. 1992, 6, 92.
7. Shearer, M. J.; Barkhan, P. Biochim. Biophys. Acta 1973,
297, 300.
8. Tadano, K.; Yuzuriha, T.; Sato, T.; Fujita, T.; Shimada,
K.; Hashimoto, K.; Satoh, T. J. Pharmacobiodyn. 1989,
12, 640.
9. Sano, Y.; Tadano, K.; Kaneko, K.; Kikuchi, K.;
Yuzuriha, T. Jpn. Pharmacol. Ther. 1995, 23, 207.
10. Spronk, H. M. H.; Soute, B. A. M.; Schurgers, L. J.;
Thijssen, H. H. W.; DeMey, J. G. R.; Vermeer, C. J. Vasc.
Res. 2003, 40, 531.
11. Habu, D.; Shiomi, S.; Tamori, A.; Takeda, T.; Tanaka, T.;
Kubo, S.; Nishiguchi, S. JAMA 2004, 292, 358.
12. Otsuka, M.; Kato, N.; Shao, R. X.; Hoshida, Y.; Ijichi,
H.; Koike, Y.; Taniguchi, H.; Moriyama, M.; Shiratori,
Y.; Kawabe, T.; Omata, M. Hepatology 2004, 40, 243.
13. Lamson, D. W.; Plaza, S. M. Altern. Med. Rev. 2003, 8,
303.
We synthesized six kinds of vitamin K metabolites and
evaluated their biological activities in vitro. Regarding
MK-4 analogues, x-hydroxylation of the side-chain part
increased apoptosis-inducing activity for HL-60 cells
compared to MK-4. Until recently, only the apoptosis-
inducing activity of vitamin K analogues as well as their
unsaturated alkyl side-chain moiety such as geranylger-
aniol and farnesol had been investigated in vitro. Both
action mechanisms of vitamin K and alkyl side-chain
analogues are similar in the point of inducing caspase
3 through mitochondria;20 however, the activity of the
unsaturated alkyl side-chain moiety appeared within a
few hours, whereas that of vitamin K took 3 days.
Presumably, the time of their uptake to cells and
transfer to mitochondria were affected. Our x-hydroxyl-
ated compound 5 was exhibited about 2–3 days after the
addition to cells; therefore, our compound would
14. Tabb, M. M.; Sun, A.; Zhou, C.; Grun, F.; Errandi, J.;
¨
Romero, K.; Pham, H.; Inoue, S.; Mallick, S.; Lin, M.;
Forman, B. M.; Blumberg, B. J. Biol. Chem. 2003, 278,
43919.
15. Otsuka, M.; Kato, N.; Ichimura, T.; Abe, S.; Tanaka, Y.;
Taniguchi, H.; Hoshida, Y.; Moriyama, M.; Wang, Y.;
Shao, R.-X.; Narayan, D.; Muroyama, R.; Kanai, F.;
Kawabe, T.; Isobe, T.; Omata, M. Life Sci. 2005, 76, 2473.
16. Chehade, K. A. H.; Andres, D. A.; Morimoto, H.;
Spielmann, H. P. J. Org. Chem. 2000, 65, 3027.
17. Flowers, R. A., II; Naganathan, S.; Dowd, P.; Arnett, E.
M.; Ham, S. W. J. Am. Chem. Soc. 1993, 115, 9409.
18. Compound 3: 1H NMR (500 Mz, CDCl3):
(3H, s), 1.79 (3H, s), 2.03 (2H, t, J = 7.5 Hz), 2.12
d
1.63

Y. Suhara et al. / Bioorg. Med. Chem. Lett. 17 (2007) 1622–1625
1625
(2H, t, J = 7.5 Hz), 2.19 (3H, s), 3.36 (2H, d, J = 6.5 Hz),
3.94 (2H, br s), 5.01 (1H, t, J = 7.0 Hz), 5.32 (1H, t,
J = 7.0 Hz), 7.68–7.70 (2H, m), 8.07–8.09 (2H, m); 13C
NMR (125 Mz, CDCl3): d 12.7, 13.7, 16.3, 25.9, 26.1,
39.2, 68.9, 119.6, 125.6, 126.2, 126.3, 132.1, 133.3, 133.4,
135.0, 137.0, 143.4, 146.1, 184.6, 185.4; EI-LRMS m/z
324 (M+). EI-HRMS calcd for C21H24O3 324.1725.
Found 324.1748. Compound 4: 1H NMR (500 Mz,
CDCl3): d 1.57 (3H, s), 1.63 (3H, s), 1.79 (3H, s),
1.93–2.09 (8H, m), 2.17 (3H, s), 3.67 (2H, d, J = 7.5 Hz),
3.97 (2H, br s), 5.00–5.07 (2H, m), 5.32–5.35 (1H, m),
7.68–7.70 (2H, m), 8.07–8.09 (2H, m); 13C NMR
(125 Mz, CDCl3): d 12.7, 13.6, 16.0, 16.4, 26.0, 26.2,
26.3, 39.2, 39.6, 69.0, 119.2, 124.1, 126.0, 126.2, 126.3,
132.1, 132.2, 133.28, 133.34, 134.7, 134.8, 137.4, 143.3,
146.1, 184.5, 185.5; EI-LRMS m/z 392 (M+). EI-HRMS
calcd for C26H32O3 392.2351. Found 392.2376. Com-
pound 5: 1H NMR (500 Mz, CDCl3): d 1.53 (3H, s),
1.56 (3H, s), 1.66 (3H, s), 1.79 (3H, s), 1.91–2.13 (12H,
m), 2.19 (3H, s), 3.37 (2H, d, J = 7.5 Hz), 3.99 (2H, br
s), 5.00–5.08 (3H, m), 5.38 (1H, dt, J = 1.5, 7.5 Hz),
7.68–7.70 (2H, m), 8.07–8.09 (2H, m); 13C NMR (125
Mz, CDCl3): d 12.7, 13.7, 15.96, 16.0, 16.4, 26.0, 26.2,
26.5, 26.6, 39.3, 39.6, 39.7, 69.0, 119.1, 123.9, 124.5,
126.16, 126.22, 126.3, 132.2, 133.3, 133.4, 134.5, 135.1,
137.5, 143.4, 146.2, 184.5, 185.4; EI-LRMS m/z 460
(M+). EI-HRMS calcd for C31H40O3 460.2977. Found
460.2965. Compound 6: 1H NMR (500 Mz, CDCl3): d
1.72 (3H, s), 1.83 (3H, s), 2.18 (2H, dd, J = 7.5, 15.0 Hz),
2.19 (3H, s), 2.45 (2H, dd, J = 7.5, 15.0 Hz), 3.39 (2H, d,
J = 7.0 Hz), 5.08 5.09 (1H, dt, J = 1.5, 7.5 Hz), 6.41 (1H,
dt, J = 1.5, 7.5 Hz), 7.69–7.71 (2H, m), 8.07–8.10 (2H,
m), 9.33 (1H, s); 13C NMR (125 Mz, CDCl3): d 9.2, 12.7,
16.3, 26.1, 7.3, 38.0, 120.5, 121.3, 126.27, 126.31, 132.1,
133.41, 133.44, 136.0, 143.5, 145.7, 153.8, 184.5, 185.4,
195.1; EI-LRMS m/z 322 (M+). EI-HRMS calcd for
C21H22O3 322.1569. Found 322.1554. Compound 7: 1H
NMR (500 Mz, CDCl3): d 1.59 (3H, s), 1.70 (3H, s), 1.79
(3H, s), 2.00 (2H, t, J = 7.5 Hz), 2.06–2.11 (4H, m), 2.19
(3H, s), 2.39(2H, dd, J = 7.5, 15.0 Hz), 3.37 (2H, d,
J = 7.5 Hz), 5.02 (1H, dt, J = 1.5, 7.5 Hz), 5.09 (1H, dt,
J = 1.5, 7.5 Hz), 6.40 (1H, dt, J = 1.5, 7.5 Hz), 7.68–7.70
(2H, m), 8.07–8.09 (2H, m), 9.35 (1H, s); 13C NMR
(125 Mz, CDCl3): d 9.19, 12.7, 15.9, 16.4, 26.1, 26.4,
27.4, 37.9, 39.5, 119.4, 125.2, 126.3, 126.3, 132.17,
132.21, 133.37, 133.41, 133.6, 137.3, 139.4, 143.4, 146.2,
154.4, 184.6, 185.5, 195.3; EI-LRMS m/z 390 (M+). EI-
HRMS calcd for C26H30O3 390.2195. Found 390.2201.
Compound 8: 1H NMR (500 Mz, CDCl3): d 1.56 (3H,
s), 1.59 (3H, s), 1.74 (3H, s), 1.79 (3H, s), 1.92 (2H, t,
J = 7.0 Hz), 1.98–2.07 (6H, m), 2.14 (2H, t, J = 7.5 Hz),
2.19 (3H, s), 2.43 (2H, dd, J = 7.0, 14.5 Hz), 3.37 (2H, d,
J = 7.5 Hz), 5.00–5.12 (3H, m), 6.45 (1H, dt, J = 1.5,
7.5 Hz), 7.68–7.70 (2H, m), 8.07–8.09 (2H, m), 9.37 (1H,
s); 13C NMR (125 Mz, CDCl3): d 9.23, 12.7, 15.9, 16.4,
26.0, 26.5, 26.6, 27.4, 38.0, 39.5, 39.7, 119.2, 124.1,
125.6, 126.2, 126.3, 132.2, 133.3, 133.4, 135.0, 137.5,
139.4, 143.4, 146.2, 154.5, 184.6, 185.5, 195.3; EI-LRMS
m/z 458 (M+). EI-HRMS calcd for C31H38O3 458.2821.
Found 458.2812.
19. Mizuguchi, Y.; Wada, A.; Nakagawa, K.; Ito, M.; Okano,
T. Biol. Pharm. Bull. 2006, 29, 1803.
20. Shibayama-Imazu, T.; Sakairi, S.; Watanabe, A.; Aiuchi,
T.; Nakajo, S.; Nakaya, K. J. Cancer Res. Clin. 2003,
129, 1.
21. Suhara, Y.; Nihei, K.; Kurihara, M.; Kittaka, A.; Yam-
aguchi, K.; Fujishima, T.; Konno, K.; Miyata, N.;
Takayama, H. J. Org. Chem. 2001, 66, 8760.