Immunosuppressive cyclolignans

Article abstract of DOI:10.1021/jm960023h

The immunosuppressive activity of several lactonic, nonlactonic, and heterocycle-fused cyclolignans has been demonstrated for the first time by use of a T-cell-mediated immune response. Of the compounds tested, 4'- demethyldeoxypodophyllotoxin (8), β-apopicropodophyllin (6), and the isoxazoline-fused cyclolignan 15 are the most potent with respect to their suppression of activated splenocytes.

Full text of DOI:10.1021/jm960023h

J . Med. Chem. 1996, 39, 2865-2868
2865
Im m u n osu p p r essive Cyclolign a n s
Marina Gordaliza,*,† Glynn T. Faircloth,‡ Ma Angeles Castro,† J ose´ M. Miguel del Corral,†
Ma Luisa Lo´pez-Va´zquez,† and Arturo San Feliciano†
Laboratorio de Quı´mica Farmace´utica, Facultad de Farmacia, Universidad de Salamanca, 37007-Salamanca, Spain, and
PharmaMar USA Inc., 26 Landsdowne St., J ackson Building, Suite 420, Cambridge, Massachusetts 02139
Received December 27, 1995X
The immunosuppressive activity of several lactonic, nonlactonic, and heterocycle-fused cyclo-
lignans has been demonstrated for the first time by use of a T-cell-mediated immune response.
Of the compounds tested, 4′-demethyldeoxypodophyllotoxin (8), â-apopicropodophyllin (6), and
the isoxazoline-fused cyclolignan 15 are the most potent with respect to their suppression of
activated splenocytes.
In tr od u ction
As a continuation of our research in this area, a
number of lignan derivatives, which are representative
of several classes of natural and semisynthetic cyclo-
lignans related to podophyllotoxin and â-peltatin, have
been subjected to in vitro evaluation of their immuno-
modulatory (IM) activity. As the standard for charac-
terizing immunossupression CyA was used.
Immunomodulators are natural or synthetic products
capable of modifying the normal or aberrant immune
system through stimulation or suppression. Currently,
they constitute an important category of therapeutic
agents in a rather wide range of conditions from allergy
and anaphylaxis to psoriasis, rheumatoid arthritis, and
cancer.1 Compounds with established immunomodu-
latory properties have discrete or polymeric structures,
such as steroids, purine derivatives, polypeptides, and
glycosides. As a common feature, most of these materi-
als have inherent toxicity and adverse side effects which
direct the search for new and safer agents. Conse-
quently, apart from the demonstration of their thera-
peutic efficacy, the search for new useful immunosup-
pressants is mainly guided by the absence of cytotoxicity
and other adverse side effects of these putative drugs.2
In common with the majority of chemotherapeutics
used against malignancy, AIDS, and other viral dis-
eases, lignan derivatives such as etoposide, teniposide,
and podophyllotoxin have a number of undesirable side
effects (anemia, alopecia, carcinogenicity, gastrointes-
tinal disturbances, etc.). These agents also diminish the
natural defenses and immune responses of treated
patients, making them susceptible to opportunistic
attacks by bacteria, fungi, and other microorganisms.
This immunosuppressive effect can be put to good use
as it can be used to prevent the acute rejection of
transplanted organs. Thus, the alkylating antineoplas-
tic cyclophosphamide, the macrolide FK-506, the purine
antimetabolite azathioprine, and, principally, the polypep-
tide antibiotic cyclosporine A (CyA), although cytotoxic
immunosuppressants, are used clinically to prolong the
survival of transplanted organs.2
Ch em istr y
Some of the compounds tested or used as starting
materials were isolated from Podophyllum resin by
means of chromatographic procedures; that is the case
of podophyllotoxin (1), deoxypodophyllotoxin (2), â-pelta-
tin (3), and R-peltatin (4) (Scheme 1). 1 was trans-
formed into 5,6 6,7 and 78 by described procedures.
Compounds 8-11 have a phenolic hydroxyl group
instead a methoxyl group at position C-4′ of the phenyl
ring. They were obtained from the corresponding tri-
methoxyphenyl analogues by demethylation with HBr-
AcOH 33% and futher acetylation in the case of 9 and
10.
The heterocyclic-fused cyclolignans 12-14 were pre-
pared from podophyllotoxone (7). 7 was obtained from
1 by oxidation with pyridinium dichromate and then
condensed with the appropriate hydrazine derivative in
glacial acetic acid at room temperature. The resulting
carboxylic acid was transformed into the corresponding
derivatives 12-14 as described previously.5 Similarly,
by reaction of 7 with hydroxylamine in ethanol, com-
pound 15 was obtained.
Compound 16 was obtained from 1 after opening the
lactone ring in basic conditions, followed by oxidation
of the benzylic hydroxyl group with J ones’ reagent at 0
°C.
During the past few years, we had the aim of finding
more potent and/or less toxic antineoplastic compounds
and we prepared a considerable number of cyclolignans
related to podophyllotoxin. One of the main conclusions
of our qualitative SAR analysis was that the presence
of the trans-fused γ-lactone is a very important feature
in relation with the cytotoxicity levels of this type of
lignan.3,4 Recently we have reported some chemical
results which permitted us to propose a mechanistic
explanation for the irreversible interaction of the podo-
phyllotoxin type of lignan and receptor biomolecules.5
Biologica l Resu lts a n d Discu ssion
The method used for IM evaluation, the mixed
lymphocyte reaction (MLR) method, is a radioisotopic
assay involving cell mediated immune reactions.9 This
is accomplished using a two-way MLR derived from
murine splenocytes of genetically dissimilar strains of
mice. The MLR data are calculated as a percentage of
immune cell proliferative activity relative to control, and
IC50 values are interpolated for each test compound. In
parallel, the cytotoxic properties of the samples in
evaluation are established by the LcV (lymphocyte
viability) assay10 and their IC50 values interpolated from
the cell viability percentages relative to control. MLR
† Universidad de Salamanca.
‡ PharmaMar USA Inc.
X Abstract published in Advance ACS Abstracts, J une 15, 1996.
S0022-2623(96)00023-4 CCC: $12.00 © 1996 American Chemical Society

2866 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 14
Notes
Sch em e 1a
a
(a) Pyridinium dichromate, Cl2CH2 room temperature; (b) H2N-NHR1, AcOH, room temperature; (c) H2NOH, EtOH-Py, room
temperature; (d) KOH/MeOH 1%, room temperature; (e) CH2N2-ether; (f) CrO3, H2SO4, acetone, 0 °C.
Ta ble 1. Immunosuppressive Activity and Cytotoxicity Data
on the C-4′ phenyl ring has also been observed to be
for Several Cyclolignan Derivatives
important for the inhibitory activity of 7-epipodophyl-
lotoxin derivatives against DNA topoisomerase-II.6
compd
MLRa
LcVb
nCIMIc
1
3
4
5
6
8
9
10
11
12
13
14
15
16
CyA
<0.128
<0.00945
<0.00495
0.229
0.00825
0.00281
0.0718
0.453
11.8
9.29
0.848
7.68
280
19.7
12.8
>65.0
27.5
11.5
15.8
22.8
>25.0
>100
>26.0
>100
30.0
>2190
>2085
>2586
>284
>3333
4097
220
Compounds having a pyrazoline ring fused to the
cyclolignan are among the least potent of the series. This
decrease could be due either to the presence of the
substituent R1 attached to the saturated nitrogen atom
or to the methylation of the carboxyl group, because the
oxazoline analog 15 is as potent as those lignans having
the γ-lactone ring.
All the compounds exhibited similar activities in the
LcV assay within a 10-fold range. Therefore the com-
parison of relative potency of these cyclolignan deriva-
tives as immunosuppressants, through their noncyto-
toxic IM activity indexes (nCIMI), follows a pattern
similar to that found when the MLR results were
considered alone. However, while 4′-demethyldeoxy-
podophyllotoxin (8) was the most potent immunosup-
pressive compound of the series, â-apopicropodophyllin
(6) and the isoxazoline derivative (15) showed better
nCIMI than 3 and 4 and as a result are better candi-
dates for further in vivo evaluation.
50
>2
>11
>31
>13
<0.00900
1.41
0.0518
>3333
>18
>25.0
2.75
53
a
b
IM activity, IC50 (µg/mL). Cytotoxic activity, IC50 (µg/mL).
c Noncytotoxic IM activity (ratio LcV/MLR).
and LcV values are then compared to provide a measure
of the noncytotoxic immunosuppressive activity (nCIM
Index, nCIMI) for each substance.
From the MLR data in Table 1, considered separately
and within the non-nitrogenated derivatives, it can be
seen that the most potent IM compounds, which are in
the order 8 > 4 > 6 > 3, have no substituent at position
C-7. Thus, a reasonable analogy between IM and
antineoplastic activities3,4 is observed. The potency for
compounds having an oxygenated function at the same
position decreases in the order R-OH (1) > â-OH (5) >
OAc (10) > CO (11). The opening of the lactone
fragment, as in compound 16, also decreases the IM
potency.
Another structural feature that could be important
for the activity is the nature of the substitution on the
aryl ring at the C-4′ position. Compounds 4 and 8,
which have a free phenolic hydroxyl at the 4′-position
of the phenyl ring, were more potent than the corre-
sponding compounds 3 and 6, which have a methoxy
group at that position. In addition, acetylation of the
phenolic hydroxyl in 8 afforded 9, which proved to be
about 20 times less potent than 8. A free phenolic group
The most potent of the immunossupressive activities
of the cyclolignans tested occurred with IC50 values
between 3 and 9 ng/mL, wich could be consider in the
same range of CyA (51.8 ng/mL). However, CyA is
cytotoxic, thereby effectively reducing its therapeutic
index (nCIMI of 53) compared to the cyclolignans
(nCIMI ranging 2000-3000). In conclusion, it can be
reported that several cyclolignans (1,3, 4, 6, 8, and 15)
may act to suppress a T-cell-mediated immune response
by a noncytotoxic mechanism of action.
Although the number of compounds tested is small,
from these preliminary results, it could be concluded
that the chemical transformation of podophyllotoxin-
related cyclolignans has led to several derivatives which
show promising IM activity that merit further attention.
This is the first report of direct experimental evaluation
of this property in cyclolignans, although there are some
references to the potential utility of lignans.11,12

Notes
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 14 2867
CHCl3) 3400-2800, 1710, 1600, 1510, 1490, 1470, 1230, 1040,
940, 880, 860; UV (ꢀ) (nm, EtOH) 215 (26 500), 274 (1200),
313 (8900); H NMR (200 MHz, CDCl3) δ 8.38 (1H, bs), 7.43
Exp er im en ta l Section
Biological Assays. Mixed Lym ph ocyte Reaction (MLR).
Each compound was solved in absolute EtOH, and eight
dilutions (concentrations from 10 to 3.33 × 10-6 mg/mL) were
prepared. Duplicate volumes (10 mL) of each dilution were
added into the wells of a 96-well microtiter plate and then
evaporated to dryness at room temperature. Controls were
prepared similarly. Splenocytes from Balb/c (H-2d) and C57Bl/6
(H-2b) mice (100 mL of each cell suspension) were added
together to test and control wells. Negative control wells were
prepared by adding only 200 mL of each splenocyte suspension
and nonspecific control wells by adding only 200 mL of culture
media. Plates were incubated in a 5% CO2-humidified incuba-
tor at 37 °C for 96 h and then pulsed for 15 h with 1 mCi of
[3H]thymidine (20 Ci/mmol) per well. The contents of each
well were filtered through a glass fiber strip and the tritiated-
thymidine incorporated into newly synthesized DNA measured
using a dry scintillant counter.
Lym p h ocytes Via bility (LcV) Assa y. A 200 mL portion
of Balb/c splenocyte suspension was added to one set of test
compounds and control wells. A 200 mL sample of culture
media was added to another set of compounds to serve as
nonspecific control. The plates were incubated as above, then
pulsed with 75 mL/well of MTT-thiazolyl blue solution (150
mg), and decanted. The resulting insoluble formazan crystals,
if formed, were dissolved in 200 mL of 2-propanol and read at
570 nm with a plate reader.
Ch em istr y: Gen er a l P r oced u r es. Melting points were
determined by heating in an external silicone bath and were
uncorrected. Optical rotations were recorded on a Perkin-
Elmer 241 polarimeter in chloroform solution and UV spectra
on a Hitachi 100-60 spectrophotometer in ethanol solution. IR
spectra were obtained on a Beckmann (Acculab VIII) spectro-
photometer in chloroform solution. EIMS were run in a VG-
TS-250 spectrometer working at 70 eV. NMR spectra were
recorded at 200 MHz for 1H and 50.3 for 13C in deuteriochlo-
roform using TMS as internal reference, on a Bruker WP 200
SY. Chemical shift values are expressed in ppm followed by
multiplicity and coupling constants (J ) in hertz. Flash chro-
matography was performed on silica gel (Merck No 9385).
Elemental analysis were carried out on a Perkin-Elmer 2400
CHN elemental analyzer.
Isola tion of Lign a n s 1, 2, 3, a n d 4. The resin (50 g) of
Podophyllum peltatum was extracted with hot CHCl3. The
soluble fraction was chromatographed on neutral alumina
(activity II), and the following cyclolignans were eluted with
CHCl3: deoxypodophyllotoxin (2, 1%), podophyllotoxin (1, 8%),
â-peltatin (3, 9%), and R-peltatin (4, 7%).
1
(1H, s, H-2), 6.56 (1H, s, H-5), 6.20 (2H, s, H-2′,6′), 5.99 (2H,
s, OCH2O), 4.82 (1H, t, J ) 12.3 Hz, H-9a), 4.67 (1H, d, J )
7.0 Hz, H-7′), 3.80 (2H, m, H-8, H-9b), 3.76 (3H, s, MeO-4′),
3.66 (6H, s, MeO-3′, 5′), 3.23 (1H, dd, J ) 11.4 and 5.0 Hz,
H-8′); 13C NMR (50.3 MHz, CD3OD) δ 178.8 (C-9′), 159.5 (C-
7), 154.0 (C-3′,5′), 151.9 (C-4), 148.8 (C-3), 138.5 (C-6), 138.2
(C-1′), 138.0 (C-4′), 120.2 (C-1), 110.4 (C-5), 109.2 (C-2′,6′),
104.2 (C-2), 103.1 (OCH2O), 76.2 (C-9), 61.0 (MeO-4′), 56.4
(MeO-3′,5′), 53.8 (C-8′), 49.0 (C-7′), 46.0 (C-8). Anal.
(C22H21O8N) C, H, N.
Meth yl 7-Oxod eoxyp icr op od op h ylla te (16): 1 (1000 mg)
was added to a solution of 5% KOH/MeOH and stirred for 30
min at room temperature. After partial evaporation of the
solvent, water and HCl (2 N) until pH ) 5 were added and
extracted with EtOAc. The reaction product was treated with
an ethereal solution of CH2N2 to yield the corresponding
dihydroxy ester (90%). J ones’ reagent (1 mL) was added to a
solution of the dihydroxy ester (220 mg) in acetone (4 mL) and
stirred at 0 °C for 90 min. Excess oxidant was destroyed by
addition of saturated sodium bisulfite solution. The organic
layer was diluted with EtOAc, washed with brine, and dried
over Na2SO4, and the solvent was evaporated in vacuo. The
reaction product was chromatographed to afford compound 16
(50%): mp 76-78 °C (H/EtOAc); [R]22 -46.8° (c 1, CHCl3);
D
IR (cm-1, CHCl3) 3500, 1740, 1680, 1600, 1510, 1490, 1470,
1260, 1045, 945, 895, 850. UV (ꢀ) (nm, EtOH) 212 (12 000),
233 (12 000), 273 (7600), 314 (6300); 1H NMR (200 MHz,
CDCl3) δ 7.55 (1H, s, H-2), 6.53 (1H, s, H-5), 6.24 (2H, s,
H-2′,6′), 6.06 (1H, s, OCH2O), 6.03 (1H, s, OCH2O), 4.62 (1H,
d, J ) 2.8 Hz, H-7′), 4.24 (1H, dd, J ) 11.2 and 7.8 Hz, H-9a),
3.83 (3H, s, MeO-4′), 3.77 (6H, s, MeO-3′,5′), 3.65 (3H, s,
COOMe), 3.58 (1H, dd, J ) 11.2 and 4.4 Hz, H-9b), 3.28 (1H,
dd, J ) 2.8 and 3.8 Hz, H-8′), 2.95 (1H, m, H-8); 13C NMR
(50.3 MHz, CD3OD) δ 188.8 (C-7), 172.7 (C-9′), 153.7 (C-3′,5′),
152.8 (C-4), 147.7 (C-3), 137.9 (C-4′), 137.8 (C-1′), 137.1 (C-6),
127.8 (C-1), 109.3 (C-5), 106.6 (C-2′,6′), 105.7 (C-2), 101.9
(OCH2O), 62.1 (C-9), 60.7 (MeO-4′), 56.5 (MeO-3′,5′), 52.1
(COOMe), 50.9 (C-8′), 47.8 (C-7′), 45.9 (C-8).
Ack n ow led gm en t. Financial support for this work
came from Spanish DGICYT (PB 93/616) and J unta de
Castilla y Leo´n (Consejer´ıa de Educacio´n y Cultura, SA-
35/94). The authors thank Dr. B. Macias (Salamanca
University) for the elemental analysis.
Refer en ces
4′-Dem eth yldeoxypodoph yllotoxin (8). HBr-AcOH 33%
(2.5 mL) was added to a solution of 2 (208 mg) in 1,2-
dichloroethane (10 mL), and the mixture was stirred for 15 h
at room temperature. The reaction mixture was added to a
saturated solution of NaHCO3 mixed with ice and extracted
with EtOAc. The extract was washed with brine, and dried
over Na2SO4, and the solvent was evaporated off. The reaction
product was chromatographed to yield 155 mg of 8 (72%).
Physical data is in accord with those described previously.13
Acetylation of 8 with acetic anhydride in pyridine afforded 9.
Similarly, the following 4′-demethylated cyclolignans were
obtained.
4′-Dem eth ylisopicr opoph yllon e (11): from 7 (46%). Physi-
cal data is in accord with those described previously.13
4′-Dem eth ylep ip od op h yllotoxin Dia ceta te (10): from 1
(49%) after acetylation and flash chromatography of the
reaction product. Physical data is according with those
described previously.6
(1) Gilman, S. C.; Lewis, A. J .; Wooley, P. H. The Immune System.
In Comprehensive Medicinal Chemistry; Hansch, C., Ed.; Per-
gamon Press: Oxford, 1990; Vol. 1, Chapter 2.5.
(2) Wong, S. Immunotherapeutic Agents. In Kirk-Othmer Encyclo-
pedia of Chemical Technology; Kroschwitz, J . I., Howe-Grant,
M., Eds.; J ohn Wiley & Sons: New York, 1995; Vol. 14, p 64.
(3) San Feliciano, A.; Gordaliza, M.; Miguel del Corral, J . M.; Castro,
M. A.; Garc´ıa-Gra´valos, M. D.; Ruiz-La´zaro, P. Antineoplastic
and Antiviral Activities of Some Cyclolignans. Planta Med. 1993,
59, 246-249.
(4) Gordaliza, M.; Castro, M. A.; Garc´ıa-Gra´valos, M. D.; Ruiz-
La´zaro, P.; Miguel del Corral, J . M.; San Feliciano, A. Antineo-
plastic and Antiviral Activities of Podophyllotoxin Related
Lignans. Arch. Pharm. (Weinheim) 1994, 327, 175-179.
(5) Gordaliza, M.; Miguel del Corral, J . M.; Castro, M. A.; Lo´pez-
Va´zquez, M. L.; San Feliciano, A.; Garc´ıa-Gra´valos, M. D.; Carpy,
A. Synthesis and Evaluation of Pyrazolignans. A New Class of
Cytotoxic Agents. Bioorg. Med. Chem. 1995, 3, 1203-1210.
(6) Thurston, L. S.; Irie, H.; Tam, S.; Han, F. S.; Liu, Z. C.; Cheng,
Y. C.; Lee, K. H. Antitumor Agents. 78. Inhibition of Human
DNA Topoisomerase II by Podophyllotoxin and R-Peltatin
Analogues. J . Med. Chem. 1986, 29, 1547-1550.
(7) Buchardt, O.; J ensen, R. B.; Hansen, H. F.; Nielsen, P. E.;
Andersen, D.; Chinoin, I. Thermal Chemistry of Podophyllotoxin
in Ethanol and a Comparison of the Cytostatic Activity of the
Thermolysis Products. J . Pharm. Sci. 1986, 75, 1076-1080.
(8) Castro, M. A.; Gordaliza, M.; Miguel del Corral, J . M.; San
Feliciano, A. Preparation of Triols and Ethers Related to
Podophyllotoxin. Org. Prep. Proc. Int. 1994, 26, 539-547.
(9) Schendel, D. J .; Alter, B. J .; Bach, F. H. The Involvement of LD-
and SD- Regional Differences in MLR and CMC: A Three Cell
Experiment. Transplant. Proc. 1973, 5, 1651-1655.
Isoxa zop od op h yllic Acid (15): Pyridine (0.2 mL) and
hydroxylamine chlorhydrate (64 mg, 0.93 mmol) were added
to a solution of 7 (300 mg, 0.73 mmol) in ethanol (15 mL). The
mixture was kept at 95-100 °C for 72 h. Then the ethanol
was evaporated off, diluted with EtOAc, and washed with HCl
(2 N) and brine to afford 310 mg of reaction product, from
which 250 mg (81%) of 15 were separated after crystallization
with CH2Cl2: mp 246-248 °C; MS m/ z 483 (M+), 329, 286,
260, 176, 133, 89; [R]22 -152.6° (c 0.6, CHCl3); IR (cm-1
,
D

2868 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 14
Notes
(10) Faircloth, G. T.; Stewart, D.; Clement, J . J . A Simple Screening
Procedure for the Quantitative Measurement of Cytotoxicity to
Resting Primary Lymphocyte Cultures. J . Tissue Cult. Methods
1988, 11, 201-205.
(11) Hirano, T.; Wakasugi, A.; Oohara, M.; Oka, K.; Sashida, Y.
Suppression of Mitogen-Induced Proliferation of Human Periph-
eral Blood Lymphocytes by Plant Lignans. Planta Med. 1991,
57, 331-334.
(12) Rantapaa-Dahlqvist, S.; Lofvenbert, E.; Norberg, B. Effect of
CPH82 in Rheumatoid Arthritis. Accumulation of Bone Marrow
Cells in Mitosis and Clinical Response. Clin. Rheumatol. 1993,
12, 541-542.
(13) J ackson, D. E.; Dewick, P. M. Aryltetralin Lignans from Podo-
phyllum hexandrum and Podophyllum peltatum. Phytochemistry
1984, 23, 1147-1152.
J M960023H