Cytotoxic responses to aromatic ring and configurational variations in α-conidendrin, podophyllotoxin, and sikkimotoxin derivatives

Article abstract of DOI:10.1021/jm990563p

Derivatives of α-conidendrin, podophyllotoxin, and sikkimotoxin were prepared to evaluate the cytotoxic contributions of C-4 configuration and pendant and fused arene substitutions. Dimethyl-α-conidendryl alcohol (5), 9-deoxypodophyllol (6), and 9-deoxysikkimol (17) were dehydrated to their respective oxolane derivatives 4, 3, and 9. Diols 5 and 6 were converted via oxabicyclo[3.2.1]octanols 10 and 14 to target oxolanes 8 and 7 where C-4 had been inverted relative to that in 3 and 4. Cytotoxicities of the five oxolanes were determined in two drug-sensitive human leukemia and two multidrug-resistant cell lines expressing P-glycoprotein or multidrug-resistance associated protein (MRP). Changing the pendant arene configuration or replacing a m-methoxy by hydrogen resulted in a 100-fold cytotoxicity loss. Replacing a methylenedioxy group in the fused arene by two methoxy substituents reduced cytotoxicity by 10-fold. Drug-resistant cell lines were equally resistant to compounds 3, 4, 8, and 9 indicating that these four compounds do not serve as substrates of the transport proteins P-glycoprotein and MRP.

Full text of DOI:10.1021/jm990563p

180
J . Med. Chem. 2001, 44, 180-185
Cytotoxic Resp on ses to Ar om a tic Rin g a n d Con figu r a tion a l Va r ia tion s in
r-Con id en d r in , P od op h yllotoxin , a n d Sik k im otoxin Der iva tives
Anne Dantzig,† Robert T. LaLonde,*,‡ Frank Ramdayal,‡ Robert L. Shepard,† Koichi Yanai,‡,§ and Mianji Zhang‡
Department of Chemistry, State University of New York, College of Environmental Science and Forestry,
Syracuse, New York 13210-2726, and Lilly Research Laboratories, Lilly Corporate Center, Indianapolis, Indiana 46285-0424
Received November 10, 1999
Derivatives of R-conidendrin, podophyllotoxin, and sikkimotoxin were prepared to evaluate
the cytotoxic contributions of C-4 configuration and pendant and fused arene substitutions.
Dimethyl-R-conidendryl alcohol (5), 9-deoxypodophyllol (6), and 9-deoxysikkimol (17) were
dehydrated to their respective oxolane derivatives 4, 3, and 9. Diols 5 and 6 were converted
via oxabicyclo[3.2.1]octanols 10 and 14 to target oxolanes 8 and 7 where C-4 had been inverted
relative to that in 3 and 4. Cytotoxicities of the five oxolanes were determined in two drug-
sensitive human leukemia and two multidrug-resistant cell lines expressing P-glycoprotein or
multidrug-resistance associated protein (MRP). Changing the pendant arene configuration or
replacing a m-methoxy by hydrogen resulted in a 100-fold cytotoxicity loss. Replacing a
methylenedioxy group in the fused arene by two methoxy substituents reduced cytotoxicity by
10-fold. Drug-resistant cell lines were equally resistant to compounds 3, 4, 8, and 9 indicating
that these four compounds do not serve as substrates of the transport proteins P-glycoprotein
and MRP.
Ch a r t 1. Structures of γ-Lactone,
Tetrahydronaphthalene Lignans, and Their Oxolane
Analogues
In tr od u ction
R-Conidendrin (ACON, 1) (Chart 1) and podophyllo-
toxin (PT, 2) are tetrahydronaphthalene (THN) lignans.
ACON was once reclaimed on a large scale from sulfite
pulping1 and offered for commercial application. ACON
seems to have no cytotoxic properties of current medical
significance. In contrast, PT is a well-known cytotoxin
isolated from Podophyllum species for manufacture of
the oncolytic etoposide. Continued supply of PT from
its source has been questioned.2,3 In view of this
background, we investigated the relative contributions
of three structural differences between PT and ACON
analogues that may account for dissimilar cytotoxicities
of these lignans. In implementing this comparison,
methylation of the two hydroxyl groups of ACON,
hydrogenolysis of the C-9 hydroxyl group of PT, and
conversion of the carbonyl groups of both ACON and
PT to methylene groups were effected. The replacement
of the hydroxyl group by hydrogen and the conversion
of the carbonyl to a methylene group had transformed
PT to the oxolane 9-deoxyanhydropodopodophyllol (3)
with little bioactivity change.4
The remaining structural differences relate to the two
arenes. The fused arene is substituted at C-6 and C-7
in both ACON and PT, by hydroxy and methoxy groups
in ACON but by a methylenedioxy group in PT. The
stereochemical configuration of the pendant arene at
C-4 is â in ACON but R in PT. Also, ACON lacks the
C-5′ methoxy group of PT. A group of PT derivatives
bearing methoxy groups at C-6, C-7, and C-4′ showed
no activity for DNA breakage or inhibiting topoi-
* To whom correspondence should be sent. Tel: 315-470-6823.
Fax: 315-470-6856. E-mail: rtlalond@mailbox.syr.edu.
† Lilly Research Laboratories and to whom inquiries specifically
regarding the cytotoxicities should be addressed.
somerase II.5 However, replacement of the C-5′ methoxy
group of etoposide by hydrogen resulted in a cytotoxic
derivative.6
‡ State University of New York.
§ Present address: Nippon Paper Industries Co. Ltd.
10.1021/jm990563p CCC: $20.00 © 2001 American Chemical Society
Published on Web 12/12/2000

R-Conidendrin/ Podophyllotoxin/ Sikkimotoxin Derivatives
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 2 181
Sch em e 1
Sch em e 2
Two of the five targets required for our study required
inversion of C-4, from â to R in ACON and from R to â
in PT. Anticipating these adjustments, we had investi-
gated the influence of fused arene ring substituents that
favored benzhydrylic (C-4) as opposed to benzylic (C-9)
carbon oxygenation using simple THNs and â-coniden-
drol as models.7 Here we report the preparation of the
five PT and ACON targets required to separate the
three structural variables and the results of the cyto-
toxicity assays required for the SAR analysis. In addi-
tion to our interest in cytotoxicity-SAR, the question
of cell removal of the same PT and ACON analogues by
P-glycoprotein (Pgp) and multidrug-resistance protein
(MRP) pumps was addressed.
Ch em ica l Tr a n sfor m a tion s
ACON (1) and PT (2) were the source materials for
five compounds required for comparing cytotoxic prop-
erties. The oxolane, dimethylanhydro-R-conidendrol (4)
(Scheme 1), was obtained in three steps from 1. Steps
included methylation of the two phenolic hydroxyl
groups of 1, lithium aluminum hydride (LAH) reduction
of the lactone to dimethyl-R-conidendrol (5) (Scheme 1),
and tosyl chloride (TsCl)-promoted dehydration of the
diol to 4. Similarly, 9-deoxanhydropodophyllol (3) was
obtained from 2 by Pd/C hydrogenolysis of the C-9
hydroxyl group of 2, LAH reduction of the lactone to
diol, 6, and TsCl-promoted dehydration. Properties of
oxolanes 3 and 4 were consistent with those previously
reported4,8 and the newly obtained spectral data.
Inverting C-4 from its configuration found in the two
source materials or replacement of the methylenedioxy
group by two methoxy groups required additional
intermediate steps to obtain the remaining three oxo-
lanes 7-9 (Schemes 2-4). Inversion of C-4 in dimethyl-
R-conidendryl alcohol (5) took advantage of the C-9a
hydroxymethylene group’s position and configuration.
Treatment of 5 with 2,3-dichloro-5,6-dicyano-1,4-quino-
ne (DDQ) in a mixture of CH2Cl2 and THF resulted in
the intramolecular oxygenation of C-4 giving the
oxabicyclo[3.2.1]octanol 10 (Scheme 2) as evidenced by
DEPT, COSY, HETCOR, and HMBC analyses.
Sch em e 3
amounts of esters diastereomeric at C-4. However,
hydrogenolysis of the tert-butyldimethylsilyl (TBS) ether
13 occurred with simultaneous loss of the silyl group
and the required hydrogenolysis, giving nearly complete
retention (93:7) favoring 11. Recrystallization of the 11-
enriched mixture from acetone/hexane afforded pure 11,
which when treated with TsCl was converted to the
oxolane 7. Since hydrogenolyses of alcohol 10 and its
ester 12 had occurred with undesired inversion, the
bulky and less polar TBS derivative was chosen to shield
the R-face of the molecule and favor catalyst contact
from the less hindered â-face to give required retention
of hydrogenolysis.
Since formation of oxabicyclooctanol 10 proceeded
with inversion of configuration at C-4, succeeding hy-
drogenolysis of 10 required retention of configuration
at C-4 for overall transformation of 5 to diastereomer
11 (Scheme 2), and ultimately to target 7. However, the
Pd/C-catalyzed hydrogenolysis of bicyclooctanol 10 pro-
duced diols 5 and 11 in a ratio of 58:42, as determined
Preparation of the PT-derived oxolane 8 (Chart 1)
involved DDQ-promoted conversion of diol 6 to oxabi-
cyclooctanol 14 (Scheme 3). The structure assigned to
1
by HPLC and H NMR. Likewise, Pd/C hydrogenolysis
of the acetate ester 12 (Scheme 2) gave nearly equal

182 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 2
Dantzig et al.
Sch em e 4
tance to a wide array of natural product drugs due to
decreased accumulation of the cytotoxic agent intrac-
ellularly. The most cytotoxic of the six compounds to
these cell lines was PT oxolane (3), which was ap-
proximately 30-50 times more active than etoposide.
With respect to the remaining compounds, 3 was
consistently 10-20 times more active than sikkimotoxin
oxolane (9). Also, 3 was more active than the PT and
ACON oxolanes 4, 7, and 8 by a factor of 102-103.
Each of the five oxolanes 3, 4, and 7-9 is distin-
guished by a unique set of three structural variables,
which are: the configuration (R or â) of C-4, the number
of methoxy groups (2 or 3) attached to the pendant
arene, and the substitution of the fused arene by one
methylenedioxy group or two methoxy groups. Oxolanes
selected for pairwise comparisons of cytotoxicity were
limited to those differing by a single variable. The mean
cytotoxicities (Tables 1 and 2) for 8 and 3 show that a
C-4 R- to â-configurational change results in an ap-
proximate activity loss of 100 in all four cell lines.
Diminishing the number of pendant arene methoxy
groups from three in 9 to two in 7 results in an activity
loss of 100 as well. Replacement of the methylenedioxy
group in the fused arene of 3 by the two methoxy groups
in 9 produces a smaller 10-fold activity loss in all four
cell lines. Although oxolanes 4 and 7 differ structurally
by only the C-4 configuration, cytotoxicity for both has
diminished substantially compared to 3 and to such a
low level that cytotoxicities appear virtually the same.
As shown in Table 2, the HL60/ADR and HL60/Vinc
cell lines are, respectively, 33- and 11-fold resistant to
etoposide relative to HL60/S. In contrast, the resistant
cell lines were more sensitive to the oxolanes by 2-5-
fold. These data indicate that these oxolanes cannot be
removed from the cell by either Pgp- or MRP-mediated
efflux.
Ta ble 1. Cytotoxicity of Tetrahydronaphthalene Lignan
Derivatives to CCRF-CEM Cells
compd
IC50 (ng/mL)a
rel potb
etoposide
1146 ( 83
43 ( 8
21862 ( 2511
38450 ( 3882
8147 ( 974
470 ( 38
1
3
4
7
8
9
27.0
0.05
0.03
0.14
2.4
a
Data represent the mean ( SE of 2-4 independent experi-
b
ments determined in duplicate. Rel pot is the relative potency
of the indicated compound compared to etoposide.
14 was consistent with H and 13C NMR, DEPT, COSY,
1
HETCOR, and HMBC as well as assignments made for
the similarly bridged structure 10. The Pd/C hydro-
genolysis of 14 produced a mixture of diastereomeric
diols 6 and 15. This mixture was transformed by TsCl
dehydration to a mixture of oxolanes 3 and 8, which on
recrystallization from ether provided pure diastereomer
8. The fifth oxolane required was 9 (Scheme 4). Its
preparation began also with PT (2), which was con-
verted in three steps to the 6,7-dimethoxy lactone
analogue 16 of 9-deoxysikkimotoxin.9 The lactone 16
was reduced to the diol 17, which on TsCl dehydration
provided oxolane 9.
Con clu sion s
The configurations at C-4 in tetrahydronaphthalene
lignans ACON and PT were inverted in three steps
giving, respectively, two diols: 11 and 15. These were
the C-4 diastereomers of the corresponding two (5 and
6) obtained more directly by reduction of the lactones
ACON and 9-deoxy-PT. Dehydration of the four diols
resulted in four (3, 4, 7, and 8) of the five lignan
oxolanes required for the determination of cytotoxicities.
The fifth resulted directly by replacement of the PT
dioxymethylene group by two methoxy groups followed
by routine reduction of lactone to diol and dehydration
of the latter to the sikkimotoxin oxolane (9). Comparison
of the cytotoxicities indicated that the greatest reduc-
Cytotoxicity a n d SAR
Cytotoxicity was assessed using two human leukemic
cells lines: CCRF-CEM (Table 1) and HL60 (Table 2).
Table 2 also includes data for two multidrug-resistant
variants: HL60/ADR and HL60/Vinc. These cells were
selected for their resistance to vincristine or adriamycin
and respectively overexpress either Pgp or MRP (Table
2). Both proteins are members of the superfamily of the
ATP-binding cassette transport proteins. Overexpres-
sion of these proteins has been shown to confer resis-
Ta ble 2. Cytotoxicities of Tetrahydronaphthalene Derivatives to Drug-Sensitive and -Resistant Cell Lines (HL60/S, HL60/ADR, and
HL60/Vinc)a,b
IC50 (ng/mL)
compd
HL60/S
rel potc
HL60/ADR
RFd
HL60/Vinc
RFd
etoposide
1643 ( 419
30 ( 1
54940 ( 780
19 ( 2
26054 ( 2763
23336 ( 2268
4863 ( 225
362 ( 5
33
18088 ( 29
15 ( 0
14741 ( 725
36083 ( 4120
3059 ( 467
314 ( 66
11
3
4
7
8
9
54.8
0.05
0.01
0.23
3.08
0.6
0.7
0.2
0.7
0.7
0.5
0.4
0.3
0.4
0.6
36251 ( 8438
117178 ( 771
7094 ( 75
533 ( 29
a
b
HL60/ADR cells overexpress MRP and not Pgp; HL60/Vinc cells overexpress Pgp and not MRP. Data represent averages of 2
determinations from a single experiment. c Rel pot is the relative potency of the compound compared to etoposide. RF is the resistance
d
factor of the cells to the compound where: RF ) IC50(drug-resistant cells)/IC50(drug-sensitive cells).

R-Conidendrin/ Podophyllotoxin/ Sikkimotoxin Derivatives
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 2 183
drated to dimethylanhydro-R-conidendrol, 4. 5: mp 168-171
tions in activity, relative to the PT oxolane 3, resulted
from two changes at the C-4 pendant arene. Inversion
of C-4 from R to â and removal of one of two m-methoxy
groups each reduced cytotoxicity by approximately 100-
fold. In contrast, replacement of the dioxymethylene
group by two methoxy groups in the fused arene
diminished activity by 10-fold. We conclude that changes
to the pendant arene attached to a rigid oxolane scaffold
dramatically affect the potency of these compounds. In
addition, these changes in structure also prevented their
removal from the cell by two multidrug-resistant pumps,
Pgp and MRP, and therefore may be potentially useful
in the treatment of multidrug-resistant tumors. Since
the five PT and ACON derivatives lack the glucoside
and pendant ring phenolic hydroxy groups found in
etoposide, it may be that the absence of these hydro-
philic groups suppresses pumping of the five derivatives
by Pgp and MRP.
25
°C (lit.8 168-172 °C); [R]D25 +33.2° (c 1.34, acetone) (lit.8 [R]D
+21° (c 0.5, 95% EtOH)). 4: mp 148-149 °C (lit.8 149-150
°C); [R]D -36.8° (c 1.31, acetone) (lit.8 [R]D -52° (c 2.1,
CHCl3)); HPLC tRI 12.1, tRG 9.8. Anal. (C22H26O5) H; C: calcd,
71.33; found, 70.73.
22
25
Podophyllotoxin was converted in three steps involving the
successive intermediates 9-deoxypodophyllotoxin4 and 9-deox-
ypodophyllol (6) leading to 9-deoxyanhydropodophyllol (3). 6:
mp 150-151 °C (lit.4 148-149 °C). 3: mp 68-75 °C (amor-
25
phous solid) (lit.9,10 65-85 °C); [R]D -59.5° (c 2.0, CHCl3)
(lit.9,10 [R]D -71° (CHCl3)); HPLC tRI 18.8, tRG 12.0. Also,
25
podophyllotoxin was converted in three steps through the
known 9-deoxypodophyllotoxin4 and 6,7-O-demethylene-9-
deoxypodophyllotoxin11 to the known 9-deoxysikkimotoxin
(16),12,13 which was reduced to 9-deoxysikkimol (17), and then
dehydrated to obtain 9-deoxyanhydrosikkimol (9). Preparation
procedures and properties of 17 and 9 are given further below.
16: mp 151-153 °C (lit.12 162-163 °C (amorphous solid));
[R]D -105.9° (c 1.7, CHCl3) (lit. [R]D -127° (CHCl3),12 [R]D
25
20
-85.8° (c 0.033, CHCl3)13).
Con ver sion of Dim eth yl-r-con iden r ol (5) to Oxabicyclo-
octa n ol 10. To a stirred solution of 360 mg (0.93 mmol) of 5
in 54 mL of CH2Cl2/THF (39:15) at 25 °C was added 246 mg
(1.08 mmol) of DDQ in 8.1 mL of CH2Cl2. Stirring continued 3
h and the solvents were removed. The residue was stirred with
a mixture of 5% aqueous NaHCO3 (33 mL) and EtOAc (30 mL),
the phases were separated, and the EtOAc phase was washed
with water then brine. The EtOAc phase was evaporated to
dryness. Preparative TLC (CH2Cl2/EtOAc, 1:1) gave 171 mg
Exp er im en ta l Section
Gen er a l. NMR data were obtained from Bruker 300 and
600 spectrometers and recorded in CDCl3 solution, unless
indicated otherwise. Chemical shift values (δ) are reported in
1
ppm and in relation to TMS (δ 0.00) and CDCl3 (δ 77.0) for H
and 13C NMR, respectively; J values are in hertz (Hz).
Quaternary, methine, methylene, and methyl carbons were
differentiated by DEPT and when unassigned to a specific
carbon are designated within parentheses as 0, 1, 2, and 3 in
association with 13C NMR δ values. 1H-1H correlation and one-
bond 1H-13C connectivity were determined by COSY and
HMQC or HETCOR experiments, respectively, while multiple-
25
(48%) of 10: mp 65-72 °C; [R]D +72.5° (c 1.71, acetone); IR
3500-3200; 1H NMR (CDCl3) (298 K) 6.87-7.4 (very broad
singlet, 2, H-2′ and H-6′), 6.87 (d, J ) 8.27, 1, H-5′), 6.66 (s, 1,
H-5 or H-8), 6.45 (s, 1, H-8 or H-5), 4.24 (ddd, J ) 8.10, 5.67,
2.24, 1, H-1), 3.89 (s, 3, OCH3), 3.85 (s, 3, OCH3), 3.84 (s, 3,
OCH3), 3.78-3.90 (m, 2, H-1, H-3), 3.69 (t, J ) 10.76, 1, H-3),
3.59 (s, 3, OCH3), 3.28 (m, 1, H-9), 2.89 (m, 1, H-9a), 2.78 (dd,
1
bond H-13C connectivity was established by HMBC. IR were
obtained from deposited films on NaCl disks and are reported
as absorbance in cm-1. MS (low-resolution) were obtained by
EI. HRMS were determined by the Midwest Center for Mass
Spectrometry, University of Nebraska-Lincoln, NE. Prepara-
tive TLC was performed using 0.5- or 1.0-mm thickness silica
gel plates containing fluorescent indicator and were viewed
under 254-nm irradiation. Isocratic and gradient HPLC of 3,
4, and 7-9 submitted for cytotoxicity assays employed a 5-µm,
C18(2), 250- × 4.60-mm column, with MeOH/H2O (65/35, 1 mL/
min) for isocratic analyses and CH3CN/H2O (50/50-95/5, 1 mL/
min) in 25-min linear gradient analyses. Retention times (in
min) are designated respectively tRI for isocratic and tRG for
gradient analyses. Detection wavelength and temperature for
all HPLC were 254 nm and 25 °C. Variation from these
conditions is noted.
1
J ) 16.84, 2.11, 1, H-9), 2.29 (m, 1, H-3a); H NMR (CDCl3)
(333 K) 7.13 (brs, 1, H-2′ or H-6′), 6.96 (brs, 1, H-6′ or H-2′)
with the remaining signals being identical to those observed
at 298 K; 13C NMR (CDCl3) 148.55 (0), 148.03 (0), 146.87 (0),
133.78 (C-1′), 131.40 (C-4a), 127.90 (C-8a), 119.12 (0), 112.00
(C-5 or C-8), 110.85 (C-8 or C-5), 110.45 (brs, C-5′), 83.78 (C-
4), 72.55 (C-1), 59.91 (C-3), 55.92 (OCH3), 55.87 (OCH3), 55.76
(OCH3), 53.34 (C-3a), 36.73 (C-9a), 32.75 (C-9). HRMS Calcd
for C22H26O6: 386.1729. Found: 386.1738.
Con ver sion of 9-Deoxyp od op h yllol (6) to Oxa bicyclo-
octa n ol 14. 9-Deoxypodophyllol (123 mg, 0.31 mmol) in CH2-
Cl2/THF was treated with DDQ in CH2Cl2, using the procedure
described for the preparation of 10. Preparative TLC (CH3Cl/
23
Ma ter ia ls. Growth medium and cell culture reagents were
obtained from GIBCO (Grand Island, NY); FBS was purchased
from Hyclone (Logan, UT). R-Conidendrin and podophyllotoxin
were obtained from Crown Zellerbach (Camas, WA) and
Bristol-Myers Squibb (Syracuse, NY) respectively, and etopo-
side was purchased from Sigma Chemical (St. Louis, MO).
Gen er a l Rea ction a n d Extr a ction P r oced u r es. Unless
indicated otherwise, reactions were conducted under dry N2.
Reaction solvents were dry and removed under vacuum after
use. Reaction product extracts were dried over anhydrous
NaSO4 or MgSO4 and the solvent was removed under vacuum.
Gen er a l P r oced u r e for Deh yd r a tin g th e 1,4-Bu ta n e-
d iol Lign a n s to th e Cor r esp on d in g Tetr a h yd r ofu r a n s.
Following a known procedure4 p-toluenesulfonyl chloride
(TsCl) in pyridine was added to the 1,4-diol in pyridine, and
the resulting mixture was heated to reflux for 3 h. Water was
added, and the solution was extracted repeatedly with EtOAc.
The combined EtOAc solution was washed successively with
1 M aqueous HCl, 5% aqueous NaHCO3, water, and brine.
From the organic layer resulted a residue that was purified
by MPLC.
EtOAc, 1:1) gave 51 mg (42%) of 14: mp 148-149 °C; [R]D
+44.2° (c 1.19, acetone); IR, 3250-3550; 1H NMR (CDCl3) 7.16
(brs, 1), 6.64 (s, 1), 6.39 (s, 1), 6.18 (brs, 1), 5.88 (d, J ) 1.39,
1, OCH2O), 5.83 (d, J ) 1.40, 1, OCH2O), 4.23 (ddd, J ) 8.14,
5.60, 2.46, 1, H-1), 3.97 (dd, 10.81, 5.48, 1, H-3), 3.70-3.89
(m, 11, H-1, H-3, 3(OCH3)), 3.27 (td, J ) 17.19, 2.76, 1, H-9),
2.85-2.90 (m, 1, H-9a), 2.75 (dd, J ) 17.24, 2.23, 1, H-9), 2.32
(m, 1, H-3a), 1.72 (bs, 1); 13C NMR (CDCl3) 147.1 (0), 145.7
(0), 137.1 (0), 136.8 (0), 132.6 (0), 129.2 (0), 109.0 (C-5 or C-8),
107.7 (C-8 or C-5), 104.3 (brs, C-2′ and C-6′), 100.8 (OCH2O),
84.1 (C-4), 72.6 (C-1), 60.8 (OCH3), 60.0 (C-3), 56.2 (OCH3),
52.8 (C-3a), 36.8 (C-9a), 33.1 (C-9). HRMS Calcd for C22H24O7:
400.1522. Found: 400.1526.
Con ver sion of Oxa bicycloocta n ol 10 to Aceta te Ester
12. To a solution of 10 (100 mg, 0.26 mmol) in THF (100 mL)
were added DMAP (65 mg, 0.53 mmol) and acetic anhydride
(56 mg, 0.55 mmol) in THF (0.5 mL). The solution was stirred
at 25 °C for 2 h. The THF was removed, and the residue was
dissolved in ether, which was washed successively with 5%
aqueous NaHCO3 (2 × 6 mL), 1 M aqueous HCl (5 × 6 mL),
water, and brine. The ether phase was dried, and the ether
was removed giving 95 mg (85%) of 12: mp 143.5-144.5 °C;
P r ep a r a tion of 3 a n d 4 fr om Lign a n Sou r ces. R-Coni-
dendrin was converted to dimethyl-R-conidendrin, which was
reduced to dimethyl-R-conidendrol, 5. Compound 5 was dehy-
[R]D 101.5° (c 2.05, acetone); IR 1735; 1H NMR (298 K)
25
(CDCl3) a very broad singlet (δ 6.9-7.4) appearing for 2

184 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 2
Dantzig et al.
protons in the aromatic region; this signal sharpened some-
what when the NMR was recorded at 333 K; 1H NMR (333 K)
(CDCl3) 7.14 (brs, 1), 6.98 (brs, 1), 6.87 (d, J ) 8.37, 1), 6.67
(s, 1), 6.48 (s, 1), 4.32 (dd, J ) 11.34, 5.45, 1), 4.23-4.28 (m,
1), 4.17 (dd, 11.24, 9.82, 1), 3.90 (s, 3), 3.87 (s, 3), 3.84 (s, 3),
3.83 (m, 1), 3.60 (s, 3), 3.21-3.28 (m, 1), 2.74-2.83 (m, 2),
2.41-2.48 (m, 1), 1.96 (s, 3); 13C NMR (CDCl3) 170.8 (0), 148.7
(0), 148.1 (0), 147.0 (0), 133.1 (0), 131.0 (0), 127.4 (0), 119.3
(0), 112.1 (1), 110.8 (1), 110.4 (1, brs), 83.8 (0), 72.5 (2), 61.9
(2), 55.91 (3), 55.87 (3), 55.76 (3), 49.6 (1), 36.9 (1), 32.8 (2),
20.7 (3). HRMS Calcd for C24H28O7: 428.1835. Found: 428.1860.
Con ver sion of Oxa bicycloocta n ol 10 to ter t-Bu tyld i-
m eth ylsilyl Eth er 13. tert-Butyldimethylsilyl chloride (413
mg, 2.74 mmol) in 3.6 mL of dry THF was added to a solution
of 10 (288 mg, 0.75 mmol) and DMAP (378 mg, 3.09 mmol) in
6 mL of THF. The mixture was stirred at 25 °C for 24 h.
Thereafter, 40 mL of 5% aqueous NaHCO3 was added and the
aqueous layer was extracted with EtOAc (5 × 20 mL). The
combined EtOAc extracts were washed with water (2 × 70 mL)
and brine (2 × 20 mL). Removal of EtOAc gave a residue that
by MPLC (hexane/EtOAc, 4:1.5) yielded 349 mg (94%) of 13
as a viscous oil: [R]D25 + 31.2° (c 1.25, CHCl3); 1H NMR (CDCl3)
6.88 (d, J ) 8.40, 1), 6.68 (s, 1), 6.42(s, 1), 6.43-7.1 (brs of
very low intensity, 2), 4.26 (ddd, J ) 8.18, 5.60, 2.37, 1), 3.92
(s, 3), 3.87 (s, 1), 3.78-3.84 (m, 2), 3.63 (t, J ) 10.31, 1), 3.60,
(s, 3), 3.31 (m, 1), 2.89 (m, 1), 2.75 (dd, J ) 17.20, 2.07, 1),
2.32 (m, 1), 0.84 (s, 9), - 0.02 (s, 3), - 0.04 (s, 3); 13C NMR
(CDCl3) 148.4 (0), 147.88 (0), 146.77 (0), 133.91 (0), 131.77 (0),
128.26 (0), 119.26 (0), 111.95 (1), 110.67 (1), 110.42 (broadened,
1), 83.57 (0), 72.65 (2), 60.06 (2), 55.89 (3), 55.76 (3), 53.12 (1),
36.78 (1), 32.79 (2), 25.80 (3), 18.12(0), -5.39 (3), -5.52 (3).
HRMS Calcd For C28H40SiO6: 500.2594. Found: 500.25868.
and then filtered. Removal of the solvent gave 183 mg of crude
product (74%). TLC indicated no 13. HPLC (MeOH/H2O, 1:1
at 1 mL/min) showed 11 (tR 39.5) to 5 (tR 44.7) ratio was 93:7.
Recrystallization from acetone/hexanes gave 151 mg (61%) of
25
11: mp 183-184 °C; [R]D -276.7° (c 0.3, CHCl3); IR 3500-
3200; 1H NMR 6.74 (d, J ) 8.09, 1, H-5′), 6.65 (s, 1, H-8), 6.59-
6.55 (m, 2, H-6′, H-2′), 6.37 (s, 1, H-5), 4.15 (brs, 1, H-4), 3.86
(s, 3, OCH3), 3.83 (s, 3, OCH3), 3.78 (s, 3, OCH3), 3.83-3.76
(m, 2, H-3 and H-1), 3.69 (s, 3, OCH3), 3.67 (m, 1, H-1), 3.53
(dd, J ) 13.59, 5.08, 1, H-3), 2.89 (dd, J ) 17.21, 4.38, 1, H-9),
2.76 (dd, J ) 17.02, 9.90, 1, H-9), 2.13 (m, 2, H-9a and H-3a);
13C NMR 148.52 (0), 147.81 (0), 147.56 (0), 135.71 (C-1′), 130.91
(C-8a or C-4a), 128.03 (C-4a or C-8a), 122.00 (C-6′), 113.38 (C-
2′), 112.38 (C-5), 110.98 (C-8), 110.94 (C-5′), 65.44 (C-1), 65.08
(C-3); 55.97 (OCH3), 55.85 (OCH3), 47.71 (C-4), 43.68 (C-3a),
34.88 (C-9a), 32.58 (C-9). HRMS Calcd for C22H28O6: 388.1886.
Found: 388.18953.
Hyd r ogen olysis of Oxa bicycloocta n ol 14 to a Mixtu r e
of 4-Isod eoxyp od op h yllol (15) a n d Deoxyp od op h yllol (6).
To a mixture of the benzoxabicyclooctane 14 (680 mg, 1.70
mmol) and 340 mg of 10% Pd/C were added 122 mL of ethanol
and 14 mL of acetic acid. The resulting mixture was stirred
under H2 at a constant 50 °C for 8.5 h, cooled to 25 °C, and
filtered. The filtrate was concentrated in vacuo to near dryness.
The residue was extracted with EtOAc and the extract was
washed with water then brine and dried. Removal of EtOAc
gave a residue that was chromatographed successively with
solvent CH2Cl2/EtOAc in ratios of 4:1 and 1:1, and finally with
EtOAc. Removal of the EtOAc gave a white solid (523 mg,
76%), a mixture of 6 and 15 proving to be inseparable by HPLC
using ODS columns and solvents of water/CH3OH or water/
CH3CN. Therefore, the mixture was used in the dehydration
step without separation.
Hyd r ogen olysis of Oxa bicycloocta n ol 10 to a Mixtu r e
of Dim eth yl-r-con id en d r ol (5) a n d Dim eth yl-4-iso-r-
con id en d r ol (11). Ca ta lyst 10% P d /C: A solution of 60 mg
of 10 in 10 mL of 95% EtOH was added to 30 mg of Pd/C, and
the resulting suspension was stirred at 25 °C under H2 for 6
h. The mixture was filtered and the EtOH was removed. TLC
of the residue showed no 10. HPLC revealed a ratio of 5 (tR
19.97) to 11 (tR 17.85) of 58:41, which was confirmed by 1H
NMR peak integrations. HPLC conditions were the same as
those described in the general experimental conditions except
the flow rate was 2 mL/min and the ratio of MeOH/H2O )
1:1.
Red u ction of 9-Deoxysik k im otoxin (16) to 9-Deoxysik -
k im ol (17). 16 (77.0 mg, 0.19 mmol) in 4 mL of THF was
added dropwise to LAH (43.7 mg, 11.5 mmol) stirred in 2 mL
of THF at 0 °C. Stirring was continued 3 h. Thereafter, EtOAc
was added dropwise and then saturated aqueous NH4Cl (3 mL)
and water (1 mL). The mixture was filtered through sintered
glass. The filtrate was extracted with EtOAc. The organic layer
was washed with water and then brine. The EtOAc extract
was dried, and the solvent was removed giving 63 mg (81%)
of 17: mp 125-127 °C; [R]D -172° (c 1.45, CHCl3); 1H NMR
25
(CDCl3) 6.64 (s, 1, H-8), 6.38 (s, 1, H-5), 6.24 (s, 2, H-2′, 6′),
4.13 (d, J ) 3.57, 1, H-4), 3.86 (s, 3, OCH3), 3.80 (s, 3, OCH3),
3.74 (s, 6, OCH3), 3.71 (s, 3, OCH3), 3.90-3.70 (m, 2, H-1, 3),
3.64 (dd, J ) 3.60, 10.50, 1, H-1), 3.51 (dd, J ) 10.56, 6.45, 1,
H-3), 2.87 (dd, J ) 16.94, 4.87, 1, H-9), 2.75 (dd, J ) 16.94,
10.41, 1, H-9), 2.40-2.20 (br s, 1, OH), 2.20-2.05 (m, 2, H-3a,
9a), 1.90-1.50 (br s, 1, OH); 13C NMR (CDCl3) 152.7 (C-3′, 5′),
147.8 (C-6), 147.5 (C-7), 138.6 (C-1′), 136.8 (C-4′), 130.5 (C-
8a), 128.0 (C-4a), 112.4 (C-5), 111.0 (C-8), 107.3 (C-2′, 6′), 65.3
(2, C-1), 64.9 (C-3), 60.8 (OCH3), 56.1 (OCH3), 55.9 (OCH3),
55.8 (OCH3), 48.2 (C-3a), 43.6 (C-4), 34.9 (C-9a), 32.4 (C-9).
HRMS Calcd for C23H30O7: 418.1992. Found: 418.1977.
Ca ta lyst Ra n ey Ni: Similarly, 44 mg of catalyst, 16 mg
(0.04 mmol) of 10 in 15 mL of ethanol stirred under H2 for 72
h gave quantitative conversion to 5 and 11, which appeared
in a ratio of 78:21 by HPLC.
Hyd r ogen olysis of Aceta te 12 to a Mixtu r e of Di-
m eth yl-r-con id en d r ol (5) a n d 4-Isod im eth yl-r-con id en -
d r ol (11). Ca ta lyst P d /C: A solution of 63 mg (0.15 mmol) of
12 in 14 mL of 95% EtOH was stirred with 29 mg of Pd/C
under H2 for 24 h at 25 °C. Thereafter, TLC showed no starting
ester. The suspension was filtered and EtOH was removed.
The dry residue was chromatographed (CH2Cl2:EtOAc, 5:1) to
obtain 55 mg (87%) of a mixture of two esters present in a
ratio of 48:52 as indicated by HPLC (tR 36.7 and 44.7).
Methanolysis of the mixture in the presence of K2CO3 at 25
°C for 1.9 h produced, after workup, a mixture of 5 and 11 in
a ratio of 46:54 by HPLC (tR 5, 20.0; 11, 17.85). Conditions for
HPLC were the same as those for the Pd/C hydrogenolysis of
10.
Ca ta lyst Ra n ey Ni: Hydrogenolysis (1 atm) for 24 h of 12
(15 mg, 0.04 mmol) in dry EtOH (10 mL) in the presence of 11
mg of stirred catalyst resulted in complete conversion to a
mixture of esters in a ratio of 44:56 by HPLC. Methanolysis
of the mixture in the presence of K2CO3 gave 5 and 11 in a
ratio of 44:56. Conditions for all the HPLC analyses were the
same as those given for the Pd/C hydrogenolysis of 10, above.
Dim eth yla n h yd r o-4-iso-r-con id en d r ol (7). TsCl (59 mg,
0.31 mmol) in 1 mL of pyridine was added to 63 mg (0.16
mmol) of 11 in 1 mL of pyridine. Heating the mixture,
processing in the general manner, and MPLC (CH2Cl2/EtOAc,
5:1), gave 43 mg (72% yield) of 7 (white amorphous solid):
25
[R]D -156° (c 0.5, acetone); HPLC tRI 10.0, tRG 9.0; IR 1608;
1H NMR 2.26-2.34 (m, 2, H-9a, H-3a), 2.59 (dd, J ) 15.69,
10.91, 1, H-9), 3.00-3.08 (m, 2, H-3, H-9), 3.43 (dd, J ) 9.46,
7.79, 1, H-1), 3.72 (s, 3, OCH3), 3.77 (s, 3, OCH3), 3.83 (s, 3,
OCH3), 3.88 (s, 3, OCH3), 4.03 (dd, J ) 15.28, 7.72, 2, H-1,
H-3), 4.30 (d, J ) 5.24, 1, H-4), 6.39 (dd, J ) 8.16, 1.99, 1,
H-6′), 6.43 (d, J ) 1.97, 1, H-2′) 6.46 (s, 1, H-5), 6.67 (s, 1,
H-8), 6.73 (d, J ) 8.14, 1, H-5′); 13C NMR 32.32 (C-9), 35.06
(C-9a), 44.84 (C-4), 46.07 (C-3a), 55.85 (OCH3), 55.87 (OCH3),
55.97 (OCH3), 70.01 (C-3), 72.71 (C-1), 110.85 (C-5′), 111.46
(C-8), 113.50 (C-5), 113.60 (C-2′), 122.28 (C-6′), 128.54 (C-4a
or C-8a), 130.65 (C-4a or C-8a), 134.68 (C-1′), 147.72 (0), 147.80
(0), 148.47 (0). HRMS Calcd for C22H26O5: 370.17802. Found:
370.17863.
Hyd r ogen olysis of ter t-Bu tyld im eth ylsilyl Eth er 13 to
Dim eth yl-r-con id en r ol (5) a n d Dim eth yl-4-iso-r-con i-
d en d r ol (11). tert-Butyldimethylsilyl ether 13 (318 mg, 0.64
mmol) in 91 mL of CH3OH/hexanes (2.5:4) was added to 212
mg of 10% Pd/C. The suspension was stirred under H2 for 6 h,

R-Conidendrin/ Podophyllotoxin/ Sikkimotoxin Derivatives
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 2 185
Deoxya n h yd r o-4-isop od op h yllol (8). TsCl (710 mg, 3.72
mmol) and 500 mg (1.24 mmol) of a mixture of 6 and 15 in 10
mL of pyridine were heated. Workup in the general manner
gave 285 mg of crude 3 and 8 which by a single recrystalli-
zation from ether gave 94 mg of 8 (24%): mp 171-172 °C;
[R]D25 -66.8° (c 0.66, acetone); HPLC tRI 20.7, tRG 12.2; 1H NMR
2.24 (m, 2, H-3a and 9a), 2.74 (dd, J ) 15.92, 9.65, 1, H-9),
2.98 (dd, J ) 15.92, 3.97, 1, H-9), 3.48-3.56 (m, 2H), 3.68 (d,
J ) 9.53, 1), 3.81 (s, 6H, OCH3), 3.82 (m, 1), 3.85 (s, 3, OCH3),
4.19 (t, J ) 7.26, 1, H-1), 5.87 (d, J ) 1.29, 1, OCH2O), 5.88
(d, J ) 1.30, 1, OCH2O), 6.30 (s, 1, H-5 or H-8), 6.31 (s, 2, H-2′
and H-6′), 6.61 (s, 1, H-8 or H-5); 13C NMR 32.82 (C-9), 42.27
(C-9a), 50.56 (C-3a), 50.76 (C-4), 56.16 (OCH3), 60.84 (OCH3),
72.25 (C-3), 73.10 (C-1), 100.87 (OCH2O), 105.30 (C-2′ and
C-6′), 108.66 (C-8), 109.27 (C-5), 129.37 (C-4a), 132.80 (C-8a),
136.80 (C-4′), 140.27 (C-1′), 146.10 (C-6), 146.13 (C-7), 153.36
(C-3′ and C-5′). HRMS Calcd for C22H24O6: 384.1573. Found:
384.1574. Anal. (C22H24O6) C, H.
ESF. Determination of high-resolution mass spectra by
the Midwest Center for Mass Spectrometry, University
of Nebraska-Lincoln, is also acknowledged. Dr. Pietro
Bollinger, Novartis Pharma A.G., supplied a reference
sample of sikkimotoxin, for which we are grateful.
Su p p or tin g In for m a tion Ava ila ble: HPLC for 3, 4, and
7-9; elemental analyses for 4 and 8. This material is available
free of charge via the Internet at http://pubs.acs.org.
Refer en ces
(1) Herrick, F. W.; Hergert, H. L. Recent Advances in Photochem-
istry; Plenum Press: New York, 1977; pp 443-515.
(2) Canal, C.; Moraes, R. M.; Dayan, F. A.; Ferreira, D. Podophyl-
lotoxin. Phytochemistry 2000, 54, 115-120.
(3) Moraes-Cerdeira, R. M.; Burant, C. L., J r.; Bastos, J . K.;
Nanayakkara, N. P. D.; McChesney, J . D. In vitro propagation
of P. peltatum. Planta Med. 1998, 64, 42.
(4) Gensler, W. J .; Murthy, C. D.; Trammell, M. H. Nonenolizable
podophyllotoxin derivatives. J . Med. Chem. 1977, 20, 63.
(5) Wang, Z.-Q.; Hu, H.; Chen, H.-X.; Cheng, Y.-C.; Lee, K. H.
Antitumor agents. 124. New 4â-substituted aniline derivatives
of 6,7-O,O-demethylene-4′-O-demethylpodophyllotoxin and re-
lated compounds as potent inhibitors of human DNA topoi-
somerase II. J . Med. Chem. 1992, 35, 871-877.
Deh yd r a tion of 9-Deoxysik k im ol (17) to 9-Deoxya n -
h yd r osik k im ol (9). TsCl (54.7 mg, 0.29 mmol) in 3 mL of
pyridine was added to a solution of 17 (60 mg, 0.14 mmol) in
3 mL of pyridine. Heating the mixture, workup in the general
manner, and MPLC (CH2Cl2/EtOAc, 5:1) gave 37.0 mg (64.4%)
25
of 9, a glasslike solid liquifying at 100-101 °C: [R]D -73.6°
(6) Saulnier, M. G.; Vyas, D. M.; Langley, D. R.; Doyle, T. W.; Rose,
W. C.; Crosswell, A. R.; Lang, B. H. E-ring desoxyanalogues of
etoposide. J . Med. Chem. 1989, 32, 1420-1425.
(c 2.7, CHCl3); HPLC tRI 8.7, tRG 8.8; 1H NMR (CDCl3) 6.67 (s,
1, H-8), 6.47 (s, 1, H-5), 6.09 (s, 2, H-2′, 6′), 4.29 (d, J ) 5.22,
1, H-4), 4.07 (dd, J ) 7.70, 3.84, 1, H-1), 4.03 (dd, J ) 15.12,
7.49, 1, H-3), 3.89 (s, 3, OCH3), 3.81 (s, 3, OCH3), 3.74 (s, 3,
OCH3), 3.73 (s, 6, OCH3), 3.44 (dd, J ) 9.35, 7.70, 1, H-1), 3.08
(dd, J ) 10.58, 7.49, 1, H-3), 3.04 (dd, J ) 15.94, 5.09, 1, H-9),
2.59 (dd, J ) 15.94, 10.85, 1, H-9), 2.32 (m, 1, H-3a), 2.29 (m,
1, H-9a); 13C NMR (CDCl3) 152.8 (C-3′, 5′), 147.9 (C-6), 147.8
(C-7), 137.7 (C-1′), 136.9 (C-4′), 130.3 (C-8a), 128.6 (C-4a), 113.6
(C-5), 111.5 (C-8), 107.6 (C-2′, 6′), 72.8 (C-1), 70.1 (C-3), 60.8
(OCH3), 56.2 (OCH3), 56.0 (OCH3), 55.9 (OCH3), 46.0 (C-3a),
45.5 (C-4), 35.3 (C-9a), 32.3 (C-9). HRMS Calcd for C23H28O6:
400.1886. Found: 400.1876.
(7) Ramdayal, F. D.; Kiemle, D. J .; LaLonde, R. T. Directed DDQ-
promoted benzylic oxygenations of tetrahydronaphthalenes. J .
Org. Chem. 1999, 64, 4607-4609.
(8) Hearon, W. M.; MacGregor, W. S. The naturally occurring
lignans. Chem. Rev. 1955, 55, 957-1068.
(9) Hartwell, J . L.; Schrecker, A. W. The Naturally Occurring
Lignans. In Progress in the Chemistry of Organic Natural
Products; Zechmeister, L., Ed.; Springer-Verlag: Vienna, 1958;
Vol. 15, pp 83-166.
(10) Schrecker, A. W.; Hartwell, J . L. Application of tosylate reduc-
tions and molecular rotations to the stereochemistry of lignans.
J . Am. Chem. Soc. 1955, 77, 432-437.
(11) Terada, T.; Fijimoto, K.; Nomura, M.; Yamashita, J .; Kobunai,
T.; Takeda, S.; Wierzba, K.; Yamada, Y.; Yamaguchi, H. Anti-
tumor agents. I. DNA topoisomerase II inhibitory activity and
the structural relationship of podophyllotoxin derivatives as
antitumor agents. Chem. Pharm. Bull. 1992, 40, 2720-2727.
(12) von Schreier, E. Zur structure des sikkimotoxins II. Partial-
synthese der 6,7-dimethoxy-analogen von podophyllotoxin. (To-
ward the structure of the sikkimotoxins II. Partial synthesis of
the 6,7-dimethoxy-analogues of podophyllotoxin.) Helv. Chim.
Acta 1964, 47, 1529-1555.
Cell Cu ltu r e. The human lymphoblastic leukemia cell line
CCRF-CEM was obtained from Dr. W. T. Beck (Cancer Center,
University of Illinois at Chicago, Chicago, IL). Cells were
maintained in minimum essential media for suspension cul-
tures containing Earle’s salts, 2 mM L-glutamine, and 10%
FBS.14 The HL60 cells, an acute human myeloblastic leukemia,
were also used in this study. The drug-sensitive parental
HL60, the drug-resistant HL60/ADR which overexpresses
MRP1, and HL60/Vinc which overexpresses Pgp were obtained
from Dr. Melvin Center (Division of Biology, Kansas State
University, Manhattan, KS). These cell lines were grown in
RPMI 1640 media, containing L-glutamine and 10% FBS.
Media for the HL60/ADR cells also contained 50 ng/mL
doxorubicin.
(13) Coltart, D. M.; Charlton, J . L. The asymmetric synthesis of
aryltetralin lignans: (-)-isolariciresinol dimethyl ether and (-)-
deoxysikkimotoxin. Can. J . Chem. 1996, 74, 88-94.
(14) Beck, W. T.; Mueller, J . J .; Tanzer, L. R. Altered surface
membrane glycoprotiens in Vinca alkaloid-resistant human
leukemic lymphoblast. Cancer Res. 1979, 39, 2070-2076.
(15) Denizot, F.; Lang, R. Rapid colorimetric assay for cell growth
and survival. Modifications to the tetrazolium dye procedure
giving improved sensitivity and reliability. J . Immunol. Methods
1989, 89, 271-277.
(16) Dantzig, A. H.; Shepard, R. L.; Cao, J .; Law, K. L.; Ehlhardt,
W. J .; Baughman, T. M.; Bumol, T. F.; Starling, J . J . Reversal
of P-glycoprotein-mediated multidrug resistance by a potent
cyclopropyldibenzosuberan modulator, LY335979. Cancer Res.
1996, 56, 4171-4179.
Cytotoxicity Assa ys. Cell viability was determined using
a modified tetrazolium dye reduction method.15 Cells were
harvested during logarithmic growth phase, seeded in 96-well
plates (Costar) at 6.7 × 104 cells/well, and cultured in the
presence of oncolytics as previously described.16
Ack n ow led gm en t. This work was supported by the
McIntire-Stennis USDA-CREES program and SUNY-
J M990563P