anti-cancer

ure Appl. Chem., Vol. 71, No. 6, pp. 1101±1104, 1999.
Printed in Great Britain.
q 1999 IUPAC
1101
Oncogene signal transduction inhibitors from
Chinese medicinal plants*
Ching-jer Chang,² Curtis L. Ashendel, Thomas C. K. Chan,
Robert L. Geahlen, Jerry McLaughlin and David J. Waters
Department of Medicinal Chemistry and Molecular Pharmacology Department of
Veterinary Pharmacology and Physiology Department of Veterinary Clinical Sciences
Purdue University, West Lafayette, Indiana 47907, USA
Abstract: Oncogene modulated signal transduction based on intracellular phosphorylation of
protein tyrosine or serine/threonine has been utilized as a target for oncogene-based anti-cancer
drug discovery. Inhibition of protein-tyrosine kinase and protein kinase C directed prescreen
has identi®ed numerous potential anti-tumor Chinese medicinal plants. Further bioassayguided
fractionation and separation have led to the discovery of novel protein kinase inhibitors,
anthraquinones, stilbenes and polythiophenes as potential anti-tumor agents.
Plants have been important sources for providing many anti-tumor agents with novel structures and
unique mechanisms for the control and cure of cancer. The key to the success of the plant natural product
drug discovery program resides in bioactivity-directed isolation procedures. The systematic screening for
anti-tumor agents from natural sources developed by the US National Cancer Institute was initially
guided by the activity in the mouse L1210 and P388 leukemia assay. In the last decade, a `disease
oriented' approach has been employed for screening anti-tumor activity [1,2]. Extracts or compounds are
tested directly against human tumor cell panels consisting of 60 cell lines of major human tumors
(leukemia, lung, colon, central nerve system, skin, ovary, kidney, prostate and breast cancers). The end
point for all these bioassays is cytotoxic effect toward tumor cells. However, the in vitro cytotoxic
potency is often not a good indicator for the in vivo anti-tumor ef®cacy. Therefore, an alternative
approach must be envisioned for discovery of novel anti-tumor agents with unique mechanisms.
It has now been well established that the differentiation and growth of cancer cells are tightly
controlled by oncogenic protein induced signaling processes [3,4]. These oncogene-modulated signal
transduction pathways therefore offer attractive targets for oncogene-based anti-tumor drug discovery.
Our natural product drug discovery group has designed two signal transduction-based bioassays to inhibit
the intracellular phosphorylation of the tyrosine or serine/threonine unit of signaling proteins for the
bioactivity-directed isolation of anti-tumor agents from Chinese medicinal plants, which have displayed
only marginal cytotoxic or noncytotoxic effect against human tumor cell lines.
PROTEIN-TYROSINE KINASE (PTK) INHIBITORS [5±8]
We have focused our research effort on the search for the inhibitors of Src-family kinases because of their
involvement in many src-oncogene modulated signal transduction pathways. Lck ( p56lck ) proteintyrosine
kinase is selected as our initial target for the identi®cation of Src-family kinase inhibitors. The
crude extracts of many Chinese anti-tumor medicinal plants [9,10] Polygonum cuspidatum (Hu Zhang),
Rheum palmaturn (Da Huang), Scutellaria baicalensis (Huang Qin), Polygonum multiforum (He Shou
Wu), Ganoderma luidum (Ling Zhi) and Cassia occidentalis (Wang Jiang Nan) were shown to be active in
this bioassay.
*Invited Lecture presented at the 21st IUPAC International Symposium on The Chemistry of Natural Products
(ISCNP-21), Beijing, China, 11±16 October 1998, pp. 1025±1166.
²Corresponding author.
Emodin
Extracts from the roots of Polygonum cuspidatum, contained inhibitory activity as detected by an in
vitro peptide phosphorylation assay. Bioassay-directed fractionation of this extract yielded the
anthraquinoid, emodin (IC50: 5mg/mL) [11]. Kinetic analyses indicated that emodin was a competitive
inhibitor of Lck with respect to ATP and was noncompetitive with respect to the peptide substrate.
Anti-oncogene activity and selective cytotoxicity
The cytotoxicity pro®le of emodin as evaluated in the NCI human tumor cell line panels indicated that the
compound was only moderately cytotoxic [12]. We sought therefore alternative techniques for the
evaluation of noncytotoxic agents such as emodin. One such method is to examine the capacity of
agents to selectively alter the growth properties of cells transformed due to the expression of a speci®c
oncogene. Interestingly, we found that emodin was a selective inhibitor of the growth of oncogenetransfected
or -overexpressed cells that had been transformed by transfection with an activated oncogene.
Anti-ras
Treatment of transformed bronchial epithelial (TBE) cells with emodin resulted in a dose-dependent
inhibition of cell growth at concentrations that had little or no effect on the growth of normal human
primary bronchial epithelial (HBE) cells [13]. To explore the mechanism of action of emodin in these
cells, we probed TBE and HBE cell lysates with anti-phosphotyrosine antibodies to estimate the relative
levels of tyrosine-phosphorylated proteins present in each cell type. We found that TBE cells exhibited
elevated levels of phosphotyrosine-containing proteins relative to those found in HBE cells, even though
the transforming principal (activated Ras) lacks protein-tyrosine kinase activity. Treatment of intact TBE
cells with emodin resulted in a marked decrease in the concentration of cellular protein-tyrosine
phosphorylation [13].
Anti-Her-2/neu
Emodin was also found to be a selective inhibitor of the growth of breast cancer cells that overexpress the
HER-2/neu proto-oncogene product [14]. These studies showed that emodin was an effective inhibitor of
the tyrosine kinase activity of the immunoprecipitated Her-2/neu receptor. Treatment of breast cancer
cells (MDA-MB453, AU-565 and BT-483), which overexpress the HER-2/neu protein-tyrosine kinase
receptor, with emodin inhibited the autophosphorylation of the receptor and, as a consequence, inhibited
the intrinsic kinase activity of the receptor (as assayed by the phosphorylation of enolase in anti-Her-2/neu
immunoprecipitates). Treatment of MDA-MB453 breast cancer cells with 40mM emodin resulted in a 78%
inhibition of cell growth. In contrast, the treatment of MCF-7 and MDA-MB231 cells, which express
normal levels of HER-2/neu receptor, resulted in only a 37% inhibition of growth. Interestingly, HBL-100
cells, which were derived from normal human breast tissue, were insensitive to the growth inhibitory
effects of emodin, even at concentrations as high as 80mM [14]. Cell cycle analyses indicated that emodin
blocked the entry of MDA-MB453 cells into the S phase of the cell cycle. MDA-MB453 cells assume a
rounded morphology characteristic of cells transformed by oncogenic PTKs. Treatment with emodin
resulted in a change in morphology from a rounded to a ¯at, more normal phenotype [14] and regained
contact inhibition. Thus emodin, or related analogs, may prove to be speci®c chemotherapeutic agents
that exhibit selectivity for breast cancer cells in which the HER-2/neu proto-oncogene is overexpressed.
Recently, Zhang & Hung [15] demonstrated that emodin selectively suppresses the proliferation of Her-
2/neu-overexpressing nonsmall cell lung cancer (NSCLC) cells (NCI-H1435 and Her-2/neu-transfected
1102 C.-j. CHANG et al.
q 1999 IUPAC, Pure Appl. Chem. 71, 1101±1104
Scheme 1
NCI-H460). Furthermore, the combination of emodin with another anti-cancer drug (doxorubicin,
etoposide cisplatin) induced synergistic inhibition of the growth of Her-2/neu-overexpressing NSCLC
cells.
Other anthraquinones
A series of natural and semisynthetic emodin derivatives were recently selected for evaluation of their
inhibition of three different kinases (Table 1) [16]. It appears that emodin is the best PTK inhibitor.
Deletion or modi®cation of the 6-OH group abolishes all kinase inhibitory activity. Further oxidation of
the 3-CH3 group results in a reduction of activity except for the 3-CHO group. We also examined the
requirement for the 10-keto functional group by preparing the anthrone analog of emodin.
This modi®cation results in the loss of kinase inhibitory activitiy and the retention of cytotoxicity. We
therefore, explored the carbon analog of the carbonyl group by preparing the quinone methide
derivatives, benzylideneanthrone compounds. We have shown recently that 10-(4-acetamidobenzylidene)-
9-anthrone (R: NHCOCH3) was more effective than emodin in repressing the tyrosine
phosphorylation of p185Her-2/neu, selectively inhibiting the proliferation of Her-2/neu-transformed
NIH3T3 and Her-2/neu-overexpressing human breast tumor cells, and suppressing the metastasisassociated
secretion of membrane-degrading gelatinase and invasion through a Matrigel basement
membrane preparation [17].
PROTEIN KINASE C (PKC) INHIBITORS [18±20]
Protein kinase C is a family of serine/threonine protein kinases and is endogenously activated by
diacylglycerol, which is produced from mitogen induced hydrolysis of inositol phosopholipids by
phosphlipase C. It is a phospholipid-dependent, calcium activated protein kinase. Tumor promoting
phorbol esters can also activate this protein in a manner similar to diacylglycerol. The crude extracts of
Chinese anti-tumor medicinal plants [9,10], Eclipta prostrata (Li Chang), Echinop grisijii (Lou Lu), and
Echinop latifolius (Lou Lu), were shown to contain protein kinase C inhibitory activity. On the basis of
this inhibitory activity. we have discovered a series of polythiophenes as novel PKC inhibitors (Table 2)
Chinese medicinal plants 1103
q1999 IUPAC, Pure Appl. Chem. 71, 1101±1104
Table 1 Inhibition of protein kinases by emodin and its derivatives
IC50 (mg/mL)
Compound R3 R6 p56lck p180Her-2/neu PKC
emodin CH3 OH 5 6 40
citreorosein CH2OH OH 30 > 30 > 200
CH2Br OH 30 10
CHO OH 6 30 200
emodic acid CO2H OH 30 > 30 200
CONH2 OH 30 > 30 40
physcion CH3 OCH3 > 800 > 200
fallacinal CHO OCH3 > 800 > 200
chrysophanic acid CH3 H > 800 > 30 > 200
rhein CO2H H 300 > 30 > 200
Scheme 2
[21].We observed that the PKC inhibitory effect increased as the number of thiophene ring increased in the
aldehyde series. 2-Formyl-a-terthiophene is 10 times more potent than the corresponding hydroxymethyl or
cyano analogs.Most of the functionalized a-terthiophenes exhibited highly differential cytotoxicity against
the renal and ovarian cancer panels in the NCI human tumor cell panels cytotoxocity pro®les.
ACKNOWLEDGEMENTS
We gratefully acknowledge ®nancial support from the National Cancer Institute (U01 CA50743).
REFERENCES
1 G.M. Cragg, M. R. Boyd, M. A. Christini, R. Kneller, T. D. Mays, K. D. Mazan, D. J. Newman, E. A. Sausville.
In Phytochemical Diversity (S. Wrigley, M. Hayes, R. Thomas, E. Chrystal, eds), pp. 1±29. Royal Chemistry
Society, London (1997).
2 M. R. Grever, B. A. Chabner. In Cancer: Principles and Practice of Oncology (V. T. Devita Jr, S. Hellman, S. A.
Rosenberg, eds), pp. 385±394. Lippincott-Raven, Hagerstown, MD (1996).
3 P. Blume-Jensen. In Encyclopedia of Cancer (J. D. Bertino, ed.), pp. 1626±1656. Academic Press, San Diego (1996).
4 D. C. Heimbrook, A. Oliff, J. B. Gibbs. In Cancer: Principles and Practice of Oncology (V. T. Devita Jr,
S. Hellman, S. A. Rosenberg, eds), pp. 35±46. Lippincott-Raven, Hagerstown, MD (1996).
5 E. M. Dobrusin, D. W. Fry. Annu. Repts. Med. Chem. 27, 169±178 (1993).
6 M. Frame. In: Encyclopedia of Cancer (J. D. Bertino, ed.), pp. 1172±1192. Academic Press, San Diego (1996).
7 S. Kellie. Tyrosine Kinase and Neoplastic Transformation. R. G. Landes, Austin, TX (1994).
8 C.-j. Chang, R. L. Geahlen. J. Nat. Prod. 55, 1529±1560 (1992).
9 H.-y. Hsu. Treating Cancer with Chinese Herbs. Oriental Healing Art Institute, Los Angeles, CA (1982).
10 M.-Y. Chang. Anticancer Chinese Medicine, Hunan Science and Technology, Changsha (1996).
11 H. Jayasuriya, N.M. Koonchanok, R. L. Geahlen, J. L. McLaughlin, C.-j. Chang. J. Nat. Prod. 55, 696±698 (1992).
12 C.-j. Chang, C. L. Ashendel, R. L. Geahlen, J. L. McLaughlin, D. J. Waters. In Vivo 10, 185±190 (1996).
13 T. C. K. Chan, C.-j. Chang, N. M. Koonchanok, R. L. Geahlen. Biochem. Biophys. Res. Commun. 193,
1152±1158 (1993).
14 L. Zhang, C.-j. Chang, S. S. Bacus, M.-C. Hung. Cancer Res. 55, 3890±3896 (1995).
15 L. Zhang, M.-C. Hung. Oncogene 12, 571±576 (1996).
16 D. S. Kim, N. M. Koonchanok, R. L. Geahlen, C. L. Ashendel, C.-j. Chang. Nat. Prod. Lett. 10, 173±180 (1997).
17 L. Zhang, Y.-K. Lau, L. Xi, R.-L. Hong, D. S. Kim, C.-F. Chen, G. N. Hortobagyi, C.-j. Chang, M.-C. Hung.
Oncogene 16, 2855±2863 (1998).
18 G. Hardie, S. Hanks. The Protein Kinase Facts BookÐI: Protein Serine Kinase. Academic Press, New York (1995).
19 A. Basu. Pharmac. Ther. 59, 257±280 (1993).
20 A. Gescher. Br. J. Cancer 66, 10±19 (1992).
21 D. S. Kim, C. Ashendel, Q. Zhou, C.-t. Chang, E.-S. Lee, C.-j. Chang. Bioorg. Med. Chem. Lett. 8, 2695±2698
(1998).
1104 C.-j. CHANG et al.
q 1999 IUPAC, Pure Appl. Chem. 71, 1101±1104
Table 2 Inhibition of protein kinase C
Compounds IC50 (M) Compounds IC50 (M)
>4´ 10ÿ3 7 ´ 10ÿ7
7 ´ 10ÿ4 4 ´ 10ÿ6
1 ´ 10ÿ5 2 ´ 10ÿ6