Synthesis and cytotoxic activity of new 9-substituted camptothecins

Article abstract of DOI:10.1016/j.bmcl.2008.04.016

A series of novel 9-substituted camptothecins derived from 9-formylcamptothecin were synthesized. The aldehyde was obtained from 10-hydroxycamptothecin or, better, by total synthesis. The compounds showed antiproliferative activity higher than that of the

Full text of DOI:10.1016/j.bmcl.2008.04.016

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2781–2787
Synthesis and cytotoxic activity of new 9-substituted camptothecins
Sabrina Dallavalle,a,* Daniela Granza Rocchetta,a Loana Musso,a Lucio Merlini,a
Gabriella Morini,b Sergio Penco,a Stella Tinelli,c
Giovanni Luca Berettac and Franco Zuninoc
a
`
Dipartimento di Scienze Molecolari Agroalimentari, Universita di Milano, Via Celoria 2, 20133 Milano, Italy
b
`
Universita di Scienze Gastronomiche, Pollenzo, Italy
Unita di Chemioterapia e Farmacologia Antitumorale Preclinica, Dipartimento di Oncologia Sperimentale e Laboratori,
Fondazione IRCCS Istituto Nazionale dei Tumori, Via Venezian 1, 20133 Milano, Italy
c
`
Received 14 March 2008; revised 4 April 2008; accepted 6 April 2008
Available online 10 April 2008
Abstract—A series of novel 9-substituted camptothecins derived from 9-formylcamptothecin were synthesized. The aldehyde was
obtained from 10-hydroxycamptothecin or, better, by total synthesis. The compounds showed antiproliferative activity higher than
that of the reference compound topotecan. Modelling suggested the possibility of a favourable interaction of small and polar 9-sub-
stituents with the topoisomerase I–DNA complex, which is consistent with the higher activity of these derivatives with respect to the
corresponding 7-substituted camptothecins.
Ó 2008 Elsevier Ltd. All rights reserved.
The natural product camptothecin (CPT)1 and its syn-
thetic analogues are among the most promising new
agents for the treatment of human cancers. CPT is a
pentacyclic alkaloid, which was first isolated in 1966
from the extracts of a Chinese plant, Camptotheca
acuminata, by Wall et al.2 Although CPT is a potent
cytotoxic agent, early clinical studies with this com-
pound in the 1970s were unsuccessful as a consequence
of severe and unpredictable toxicities. Interest in CPT
derivatives was revitalized in 1985 by the discovery that
CPT exhibits a unique mechanism of action because it
(and in some cases also 12) are tolerated or give substan-
tial increase of the activity.6 In particular, both topotec-
an and SN-38 have an OH substituent in position 10,
which seems to be important either to decrease the un-
wanted stabilization of the open hydroxyacid form by
human albumin7 or to increase the solubility. A recent
X-ray crystallographic analysis of a ternary complex be-
tween a topoisomerase I construct, a DNA oligonucleo-
tide and topotecan indicated that modifications at the
7- and 9-position of CPT would not interfere with
drug–protein interactions.8 To date most of the sec-
ond-generation CPT that have reached preclinical or
clinical development studies are 7-substituted deriva-
tives, for example, silatecans, karenitecins, gimatecan,
and belotecan, all bearing highly lipophilic substituents,
deemed to increase the stability of the lactone form in
ring E of CPT with respect to the open hydroxyacid
form (Chart 2).7,9
3
targets the nuclear enzyme topoisomerase I (Topo I).
CPT forms a ternary complex with topoisomerase I
and DNA, and the stabilization of this complex results
in DNA breaks by preventing DNA religation.4
Intensive efforts in medicinal chemistry over the past dec-
ades have provided a large number of CPT analogues, of
which topotecan and irinotecan (prodrug of SN-38) (Chart
1) are those so far approved for the clinical therapy.5
In our laboratory a series of camptothecins substituted
in position 7 with lipophilic moieties have been synthe-
sized using as a key intermediate CPT-7-aldehyde. Var-
ious groups were linked to the CPT scaffold via carbon-
carbon double bonds, iminomethyl or oxyiminomethyl
moieties. These compounds exhibited potent cytotoxic
activity in vitro and in vivo comparable or superior to
topotecan.10–12 The highest activity was shown by oxy-
iminomethyl derivatives: gimatecan (Chart 2), the most
SAR studies on the numerous compounds prepared
have shown that substitutions in positions 7, 9, 10, 11
Keywords: Camptothecin; Antitumor agents; Topoisomerase I; DNA
cleavage; Synthesis.
*
Corresponding author. Tel.: +39
dallavalle@unimi.it
2 50316816.; e-mail: sabrina.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.04.016

2782
S. Dallavalle et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2781–2787
N(CH3)2
C2H5
RO
HO
O
O
O
A
N
N
N
N
N
N
E
O
O
O
OH
OH
O
O
OH
O
N
N CO
Irinotecan R =
Topotecan
Camptothecin
SN 38
R = H
Chart 1.
One of the reasons for the limited number of 9-substi-
tuted CPT studied may be the scarce accessibility of this
position. In fact, in accordance with the reactivity of the
quinoline nucleus, substitution of CPT in position 7 oc-
curs easily via radical reactions.1 On the contrary, the
functionalization of position 9, that requires an electro-
philic substitution on a deactivated position, usually
occurs with low yields and gives mixtures with 12-substi-
tuted isomers.1 This difficulty has been overcome, in
most cases, starting from the naturally occurring 10-hy-
droxy-CPT, where the OH group confers increased
reactivity to the A ring.15–17
R
N
7
O
R = Si(alkyl)3
Silatecans
Karenitecins
N
R = Alkyl-Si(alkyl)3
R = CH=NOC(CH3)3
O
OH
O
Chart 2.
active compound of the series, is in phase I–II of clinical
studies in patients with advanced solid tumours no more
sensitive to standard therapies.13
The high cytotoxic activity of the few 9-substituted CPT
reported in the literature prompted us to investigate the
introduction of other moieties in this quite unexplored
position. In particular, the goal of the present study
was to synthesize analogues in position 9 of the most ac-
tive examples of 7-substituted CPT (1a–n) previously ob-
tained by us (Table 1).
While most researches have focused on the synthesis of
CPT derivatives bearing substituents in position 7, the
literature reports only a few examples of camptothecins
substituted in position 9, most relevant being 10-hydro-
xy-9-dimethylaminomethylcamptothecin (topotecan), a
drug in current clinical use, 10-hydroxy-9-allyl-CPT
(chimmitecan),14 9-nitrocamptothecin (rubitecan) and
9-aminocamptothecin, which have been submitted to
preclinical and clinical studies.5
Prior to starting the synthesis, we modelled the binding
of the designed compounds to the Topo I covalent com-
Table 1. Cytotoxic activity (IC50 (lM) on H460 cell line)
R1
R
N
R2
O
N
O
OH
O
Entry
R or R1
R2
IC50 (lM)
7-Substituted (R 1 = H)
7-Substituted
9-Substituted
Topotecan
9-Substituted (R = H)
1.18 0.24
CH2–N(CH3)2
CH2CH3
CH2OH
OH
OH
H
SN-38
1a
1b
1c
0.22 0.013
0.49 0.093
0.388 0.29
1.78 0.5
16
7
0.21 0.074
0.058 0.029
6.96 3.13
6.00 1.49
0.428 0.149
9.86 0.18
0.67 0.025
0.23 0.11
0.36 0.015
0.32 0.044
2.53 0.97
0.23 0.143
CHO
CHO
H
OH
3
1d
1e
5
CHO
CN
OCH3
H
0.18 0.05
1.04 0.5
5.65 2.5
18
4b
8b
17c
4a
8a
17b
17a
1f
CN
CN
OH
OCH3
H
1g
1h
1i
0.66 0.02
0.015 0.006
0.13 0.03
0.07 0.004
0.51 0.15
0.03 0.0076
CH = NOC(CH3)3
CH = NOC(CH3)3
CH = NOC(CH3)3
CH = NOCH 2CH2NH2
CH = NOH
OH
OCH3
H
1l
1m
1n
H

S. Dallavalle et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2781–2787
2783
plex with DNA. Thus we compared the intermolecular
interaction energy of all the couples of 7- and 9-substi-
tuted compounds with the model of topoisomerase I
cleavage site. The results indicated intermolecular inter-
action energies of the same order of magnitude for the 7-
and 9-substituted corresponding compounds, suggesting
not only that there is space for substitution in position 9
of CPT, in accordance with Staker’s crystallographic
data and modelling,8 but also that their binding activity
should be quite strong, being comparable to that of 7-
substituted CPT.
At this stage removal of the 10-hydroxy group was nec-
essary to obtain the desired 9-substituted camptothecins.
Attempts to obtain 9-formylcamptothecin by Pd-cata-
lyzed deoxygenation of 9-formyl-10-hydroxycamptothe-
cin triflate or tosylate17 were hampered by the instability
of the sulfonyl derivatives which partially reverted to the
starting compound during the workup of their synthesis.
The subsequent reaction with Et3SiH catalyzed by
Pd(OAc)2 led to the desired 9-formylcamptothecin only
in an unacceptably low yield (Scheme 1).
Our synthetic plan required CPT-9-aldehyde as a key
intermediate. As a first approach to the synthesis, we
decided to exploit the activating effect of the hydroxy
group, starting from 10-hydroxy-CPT (2). The formyla-
tion of 10-hydroxy-CPT based upon Duff reaction has
already been reported.15 However, in our hands, the
formyl derivative was obtained with erratic yields that
never exceeded 26%, despite numerous attempts of
optimization. Several other approaches to introduce
the aldehyde group at position 9 of 10-hydroxy-CPT,
including reaction with NaOH in chloroform, formal-
dehyde and SnCl4 in toluene and with a,a-dichlorom-
ethylmethylether and TiCl4 in dichloromethane,
failed. Thus, in spite of the low yield, 9-formyl-10-
hydroxycamptothecin (3) was synthesized via Duff
reaction and then converted into 9-formyl-10-methoxy-
camptothecin (5) by methylation with diazomethane.
Both compounds 3 and 5 were easily converted into
oximes 4a and 8a and cyano derivatives 4b and 8b
(Scheme 1).
These difficulties led us to devise an alternative proce-
dure based upon a total synthesis of 9-formylcamptothe-
cin via Friedla¨nder condensation of the chiral tricyclic
ketone 1418 with an appropriately substituted o-amino-
benzaldehyde. The structure of the final product
required that this intermediate contained a protected
or a potential second aldehyde group. After several
attempts, the synthetic sequence of Scheme 2 gave
satisfactory results. Commercially available (2-methyl-
3-nitrophenyl)methanol 9 was protected by treatment
with MEMCl in basic conditions to give 10,19 which
was converted into enamine 11 by treatment with
dimethylformamide dimethylacetal and pyrrolidine in
DMF. Oxidation with sodium periodate followed by
catalytic hydrogenation gave o-aminoaldehyde 13 in
quantitative yield (Scheme 2).
The Friedla¨nder condensation of 13 with ketone 14 cat-
alyzed by acetic acid or PTSA led only to degradation
products, whereas treatment with ZnCl2 in ethanol20
R
CHO
HO
O
O
N
N
N
N
N
N
4a R = CH=N-OC(CH3)3
O
O
4
7
4b R = CN
e
b
OH
O
OH
O
O
O
CHO
CHO
HO
HO
O
TsO
a
N
N
d
N
N
O
O
O
6
3
2
OH
O
OH
O
O
OH
O
c
R
CHO
H3CO
H3CO
f
O
N
N
N
N
O
O
5
8
OH
OH
O
O
8a R = CH=N-OC(CH3)3
8b R = CN
Scheme 1. Reagents and conditions: (a) HMTA, TFA, N2, 10 h, 26%; (b) for 4a: t-BuONH2ÆHCl, NaOH, EtOH, rt, 48 h, 57% for 4b: NH2OHÆHCl,
HCOONa, HCOOH, reflux, 11 h, 77%; (c) CH2N2, MeOH, rt, 3 h, 90%; (d) tosyl chloride, TEA, DMAP, CH2Cl2, N2, 0 °C, then rt 40 min, 34%; (e)
Pd(OAc)2, dppp, Et3SiH, DMF, N2, 60 °C, 1 h, 20%; (f) for 8a: t-BuONH2ÆHCl, NaOH, EtOH, rt, 72 h, 62%; for 8b: NH2OHÆHCl, HCOONa,
HCOOH, reflux, 9 h, 65%.

2784
S. Dallavalle et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2781–2787
OMEM
OMEM
OH
CH3
OMEM
CHO
NMe2
CH3
a
b
c
NO2
NO2
NO2
NO2
12
11
10
OMEM
9
OMEM
f
CHO
e
d
O
N
O
N
N
O
O
NH2
O
14
13
15
HO
O
OH
O
CN
OH
N
CHO
O
g
O
O
h
N
N
N
N
N
O
O
O
18
7
16
OH
O
OH
OH
O
O
i
R
O
N
17a R= CH=NOH
N
17b R= CH=NOCH2CH2NH2
17c R= CH=NOC(CH3)3
O
17
OH
O
Scheme 2. Reagents and conditions: (a) MEMCl, DIEA, CH2Cl2, 0 °C 10 min, then rt 12 h, 88%; (b) DMFDMA, pyrrolidine, DMF, N2, 135 °C,
12 h, 95%; (c) NaIO4, THF/H2O 1:1, rt, 3 h, quantitative; (d) H2, Pd/C 10%, EtOH, 2 h, quantitative; (e) ZnCl2, molecular sieves, EtOH, N2, 70 °C,
5 h, 40%; (f) TFA, CH2Cl2, rt, 24 h, 70%; (g) MnO2, CHCl3, rt, 12 h, 64%; (h) NH2OHÆHCl, HCOONa, HCOOH, reflux, 6 h, 47%; (i) RONH2,
EtOH, Py, reflux, 2–4 h.
afforded 9-(2-methoxyethoxymethoxymethyl)-campto-
thecin 15 in good yield. The desired CPT-9-aldehyde 7
was obtained by the deprotection of 15 and subsequent
oxidation of alcohol 16. The availability of 7 allowed us
to synthesize the three derivatives 17a–c , and nitrile 18.
most cases higher than that of the reference compound
topotecan. However, all the 9-oximinomethyl deriva-
tives (17a–c) are of an order of magnitude less potent
in growth inhibition than the corresponding 7-counter-
parts (1n, 1m, 1h). On the contrary, the presence of a
single, small polar group, such as CH2OH (16), CHO
(7) and CN (18), in 9-position resulted in an enhance-
ment of the activity with respect to the corresponding
7-analogues (1a, 1b, 1e), aldehyde 7 reaching the level
of the most potent camptothecins so far prepared. This
could be related to more strict steric requirements for
the 9-substitution, or to a favourable effect of a H-
accepting group in this position. Modelling experiments
on compound 16 support this second hypothesis, show-
ing the formation of two hydrogen bonds between the
OH group in position 9 and the carbonyl group of the
side chain of Asn 352 and the base C + 1 (Fig. 3).
Among the compounds tested in the DNA cleavage as-
say (Fig. 1), no precise correlations could be found be-
tween Topo I inhibition and antiproliferative activity.
This observation is not surprising, because the cytotoxic
effect reflects not only target inhibition, but a number of
events, including cellular pharmacokinetic behaviour.
Indeed the most hydrophobic derivatives (e.g., 1h and
17c), which are expected to have favourable cellular
accumulation, were more potent than the hydrophilic
derivatives 1m and 17b. However the substantially re-
duced cytotoxic potency of 17b reflected also a reduced
stabilization (Fig. 1) and persistence (Fig. 2) of the
cleavable complex. The contribution to the cytotoxic po-
tency of factors other than Topo I inhibition is clearly
The prepared compounds were tested for their growth
inhibitory activity against the human non-small cell lung
carcinoma cell line H46021 (Table 1), a cell model selected
for its sensitivity to Topo I inhibitors.22 Topoisomerase
I-mediated DNA cleavage assays with purified human
topoisomerase I were used to investigate the ability of
some derivatives to stimulate the DNA damage.23 All
the tested CPT revealed an intensity of DNA damage
comparable to that of the reference compound SN-38
(Fig. 1). Moreover, DNA cleavage patterns were found
identical to those observed in the presence of SN-38. Since
the drug interaction with the DNA–enzyme complex is ex-
pected to be reversible, the persistence of the cleavable
complex was evaluated after the addition of high salt con-
centration (0.6 M NaCl), which favours the dissociation
of the ternary drug–enzyme–DNA complex. As already
observed in our previous work,12 such a stability was par-
ticularly evident for 1h, which exhibited a DNA damage
persistence around 90% after 10 min (Fig. 2). All the three
9-substituted derivatives tested (7, 17b, 17c) appreciably
reduced the persistence of DNA damage as compared to
the corresponding 7-substituted analogues.
From the data reported in Table 1, it appears that the 9-
substituted CPT showed potent cytotoxic activity, in

S. Dallavalle et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2781–2787
2785
C
T M
Figure 1. Topoisomerase I-mediated DNA cleavage assays. Samples were reacted with 1, 10 and 50 lM drug at 37 °C for 30 min. Reaction was then
stopped by adding 1% SDS, 0.3 mg/ml of proteinase K and incubating for 45 min at 42 °C before loading on a denaturing 8% polyacrylamide gel. C,
control DNA; T, reaction without drug; M, purine markers. The experiment was repeated three times and the results of a representative value are
reported.
100
90
80
70
60
50
40
30
20
10
0
1m
1h
1b
7
17c
17b
SN38
0
2
4
6
8
10
Time (min)
Figure 2. Cleavage persistence. The samples were reacted for 30 min with 10 lM drug. DNA cleavage was then reversed by adding 0.6 M NaCl. The
100% value is referred to the extent of DNA cleavage at 30 min of incubation. Each value was obtained by densitometric analysis. The experiment
was repeated three times and the results of a representative value are reported.
documented by the quite different cytotoxicity of 1h and
1m, in spite of a comparable ability of these compounds
in stabilizing DNA cleavage.
effect on the cytotoxicity, as observed in the 7-series,12
most likely due to the increased hydrophilicity of the
10-OH derivatives.7 This decrease of activity was stron-
ger in the 9-series than in the 7-series in the presence of
small substituents (cfr. 3/7, 4b/18, vs 1c/1b, 1f/1e ). Meth-
ylation of the 10-hydroxy group restored the cytotoxic
The introduction of an OH substituent in position 10 of
9-substituted CPT was expected to have a detrimental

2786
S. Dallavalle et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2781–2787
Figure 3. Structure of compound 16 in the ternary complex model. For the sake of clarity all the base pairs and only some amino acids are shown.
Tagliatatela, O., Eds.; Wiley-Vch: Weinheim, 2008; pp
503–520.
2. Wall, M. E.; Wani, M. C.; Cooke, C. E.; Palmer, K. T.;
McPhail, A. T.; Sim, G. A. J. Am. Chem. Soc. 1966, 88,
3888.
3. Hsiang, Y.-H.; Hertzberg, R.; Hecht, S. M.; Liu, L. F.
J. Biol. Chem. 1985, 260, 14873.
4. Liu, L. F.; Desai, S. D.; Li, T. K.; Mao, Y.; Sun, M.; Sim,
S. P.; Liehr, J. G.; Giovanella, B. C.; Verschraegen, C. F.
(Eds.) Ann. N.Y. Acad. Sci. 2000, 922, 1.
potency in the 7-substituted series (cfr. 1d vs 1c, 1g vs 1f
and 1l vs 1i).12 In the 9-substituted series a restoring ef-
fect of the methylation was observed for the 9-cyano
derivative (8b vs 4b), whereas in the case of 9-formyl-
CPT 5 and oxyiminoderivative 8a the drop of activity re-
mained. These results suggest that the presence of sub-
stituents in position 10 is detrimental for the activity
of 9-substituted CPT derivatives.
In conclusion, we have developed a synthetic method
suitable for a convenient preparation of 9-substituted
CPT derivatives using 9-formyl CPT as a versatile inter-
mediate. By means of this synthetic pathway we have
synthesized a series of novel camptothecins endowed
with potent cytotoxic activity. The results show that
the highest potency is obtained when a small polar, pos-
sibly H-acceptor group, is present in position 9.
5. Pommier, Y. Nat. Rev. Cancer 2006, 6, 789.
6. Thomas, C. J.; Rahier, N. J.; Hecht, S. M. Bioorg. Med.
Chem. 2004, 12, 1585.
7. Burke, T. G.; Mishra, A. K.; Wani, M. C.; Wall, M. E.
Biochemistry 1993, 32, 5352.
8. Staker, B. L.; Hjerrild, K.; Feese, M. D.; Behnke, C. A.;
Burgin, A. B., Jr.; Stewart, L. Proc. Natl. Acad. Sci.
U.S.A. 2002, 99, 15387.
9. Zunino, F.; Pratesi, G. Expert Opin. Invest. Drugs 2004,
13, 69.
10. Dallavalle, S.; Delsoldato, T.; Ferrari, A.; Merlini, L.;
Penco, S.; Carenini, N.; Perego, P.; De Cesare, M.; Pratesi,
G.; Zunino, F. J. Med. Chem. 2000, 43, 3963.
11. Dallavalle, S.; Ferrari, A.; Merlini, L.; Penco, S.; Carenini,
N.; Perego, P.; De Cesare, M.; Pratesi, G.; Zunino, F.
Bioorg. Med. Chem. Lett. 2001, 11, 291.
12. Dallavalle, S.; Ferrari, A.; Biasotti, B.; Merlini, L.; Penco,
S.; Gallo, G.; Marzi, M.; Tinti, M. O.; Martinelli, R.;
Pisano, C.; Carminati, P.; Carenini, N.; Beretta, G.;
Perego, P.; De Cesare, M.; Pratesi, G.; Zunino, F. J. Med.
Chem. 2001, 44, 3264.
Acknowledgments
We are indebted to Sigma-Tau, Pomezia and to Associ-
azione Italiana per la Ricerca sul Cancro, Milan, for
financial support.
Supplementary data
13. Sorbera, L. A.; Serradell, N.; Bolos, J.; Rosa, E.; Bozzo, J.
Drugs Future 2007, 32, 859.
Experimental: Synthesis and analytical data of com-
pounds and docking procedures. Supplementary data
associated with this article can be found, in the online
version, at doi:10.1016/j.bmcl.2008.04.016.
14. Huang, M.; Gao, H.; Chen, Y.; Zhu, H.; Cai, Y.; Zhang,
X.; Miao, Z.; Jiang, H.; Zhang, J.; Shen, H.; Lin, L.; Lu,
W.; Ding, J. Clin. Cancer Res. 2007, 13, 1298.
15. Kingsbury, W. D.; Boehm, J. C.; Jakas, D. R.; Holden, K.
G.; Hecht, S. M.; Gallagher, G.; Caranfa, M. J.; McCabe,
F. L.; Faucette, L. F.; Johnson, R. K.; Hertzberg, R. P.
J. Med. Chem. 1991, 34, 98.
References and notes
16. Gao, H.; Zhang, X.; Chen, Y.; Shen, H.; Sun, J.; Huang,
M.; Ding, J.; Li, C.; Lu, W. Bioorg. Med. Chem. Lett.
2005, 15, 2003.
1. (a) Du, W. Tetrahedron 2003, 59, 8649; (b) Dallavalle,
S.; Merlini, L. In Modern Alkaloids; Fattorusso, E.,

S. Dallavalle et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2781–2787
2787
17. Cabri, W.; Candiani, I.; Zarini, F.; Penco, S.; Tedeschi, A.
Tetrahedron Lett. 1995, 36, 9197.
20. McNaughton, B. R.; Miller, B. L. Org. Lett. 2003, 5, 4257.
21. For experimental details, see Ref. 12.
18. Ejima, A.; Terasawa, H.; Sugimori, M.; Tagawa, H.
J. Chem. Soc., Perkin Trans. I 1990, 27.
22. Giaccone, G.; Gazdar, A. F.; Beck, H.; Zunino, F.;
Capranico, G. Cancer Res. 1992, 52, 1666.
19. Shin, C.; Yamada, Y.; Hayashi, K.; Yonezawa, Y.;
Umemura, K.; Tanji, T.; Yoshimura, J. Heterocycles
1996, 43, 891.
23. For experimental details, see Refs. 12 and 24.
24. Beretta, G. L.; Binaschi, M.; Zagni, E.; Capuani, L.;
Capranico, G. Cancer Res. 1999, 59, 3689.