Adriamycin promotes neurite outgrowth in the “neurite-minus” N1A-103 mouse neuroblastoma cell line

EXPERIMENTAL

CELL

203,

RESEARCH

Adriamycin

‘72-79

Promotes Neurite Outgrowth in the “Neurite-Minus” Nl A-l 03 Mouse Neuroblastoma Cell Line

JEANCHRISTOPHELARCHER, *Laboratoire

(19%‘)

*.'LAURENCECORDEAU-LOSSOUARN,* GEORGESROMEY,~ BERNARDCROIZAT,*ANDJEAN Luc VAYSSIERE*

de Biochimie Cellulaire, URA 1115, Colltige Mol&ulaire et Cellulaire, C.N.R.S.,

de France, 75005 Sophia-Antipolis,

Press.

Inc.

INTRODUCTION Neuroblastoma cells, usually regarded as transformed derivatives of neuroblasts [l, 21, have been extensively used as experimental models to study neuronal differentiation. A large number of clonal cell lines have been derived from neuroblastoma tumors of mouse, rat, and human origin. These cells, although commited, can be maintained in culture as proliferating and undifferentiated neuroblasts, which are induced to differentiate into amitotic neurite-bearing cells by various treatments including serum deprivation [3]. Many mouse neuroblastoma cell lines with distinct characteristics have been isolated from a single tumor, C-1300 [2-41. Most of the cell lines, such as NlE-115,

’ To

whom

reprint

requests

should

0014.4827/X? $5.00 Copyright 0 1992 by Academic Press, All rights of reproduction in any form

be addressed.

72 Inc. reserved.

de Pharmacologic

differentiate upon serum withdrawal from the culture medium or upon exposure to some chemical inducers, such as dimethyl sulfoxide, DMSO [5], hexamethylene bisacetamide, HMBA [6], and l-methyl cyclohexane carboxylic acid, CCA [7]. Extensive work has been devoted to elucidating the mechanisms of action of neuroblastoma chemical inducers [5, 6, 8, 91 and the effects of serum deprivation [lo, 111. In particular, it has been shown that one likely molecular target of CCA is at the level of the mitochondrial ATP-synthetase [la] while the effects produced by serum deprivation may be due to the removal of serine proteases present in serum [lo, 111. However, some cell lines derived from the C-1300 mouse neuroblastoma clone are unable to extend neurites under the same conditions. Among them, the NlA103 cell line is probably the most thoroughly studied [3, 41. When incubated in the absence of serum or in the presence of a chemical inducer, cells stop dividing but fail to extend neurites [4,13,14]. However, recent work indicated that some morphogenesis can been obtained by a combination of serum deprivation and hydroxyurea addition [15], suggesting that the conditions under which blockade of the cell cycle is achieved might be of importance in triggering the initial morphogenesis. In the present study, the effect of adriamycin on the “neurite-minus” NlA-103 neuroblastoma cell line was examined. Adriamycin has been used as a routine anticancer agent for two decades [16-191. However, its antineoplastic utilization is limited by an extreme cardiotoxicity [20, 211. Its clinical importance has prompted people to try to elucidate its mode of action, with the objective of designing derivatives based on a logical understanding of the receptor sit.e. Alt,hough membrane binding [22-241 and impairment of topoisomerase II activity [25, 261 have been proposed as possible targets, there is evidence showing that the biological activity of these drugs correlates with DNA-associated events [27-301. Despite its antineoplastic action, adriamycin has rarely been tested on proliferative cells and more particularly on cells able to differentiate after they stop divid-

Adriamycin, an anticancer agent acting on topoisomerase II, promotes the arrest of cell division and neurite extension in a “neurite-minus” murine neuroblastoma cell line, NlA-103. This morphological differentiation is accompanied by a blockade in the S phase of the cell cycle, modification of the amount of peripherin, and appearance of the /37-tubulin isoform. Yet, adriamycin-induced NlA-103 cells fail to express other neuronal markers, such as long-lasting Ca’+ channels, synaptophysin, and the shift in the proportion of the 8’1 tubulin isoform to the /3’2 isoform, whose appearance parallels the terminal differentiation of the wild type neuroblastoma cell line NlE-115. Hence, a comparison of the behavior of these two cell lines leads to the proposal that there are two programs of neuroblastoma differentiation: one where expression is triggered by the arrest of cell division and which is observed in adriamytin-induced NlA-103 variant cells, and the other, presumably occurring further downstream, which would involve further changes in morphogenesis and acquisition of new electrophysiological properties. (6 1992 Academic

Paris, France; and tInstitut 06560 Valbonne, France

FRANCOISGROS,*

ADRIAMYCIN-INDUCED

NlA-103

NEUROBLASTOMA

ing. Nevertheless, adriamycin was described as an excellent inducer of a murine neuroblastoma cell line, S2OY [31]. We report here on a study which aims at comparing the effects of adriamycin and other chemical inducers (CCA and DMSO) on the neurite-minus NlA103 neuroblastoma cell line and on NlE-115, a neuroblastoma cell line which extends neurites upon induction. Three complementary approaches were used: (1) a morphological and cell cycle analysis, (2) an electrophysiological study, and (3) a biochemical analysis of some neuron-specific markers including special elements of the cytoskeleton. Our data indicate that when NlA-103 neuroblastoma cells are cultured in the presence of adriamycin, in a serum-supplemented medium, neurite extension takes place and the cells express certain neuronal features, characteristic of the differentiating NlE-115 cell line, while other markers of terminal differentiation fail to be expressed. MATERIALS

AND

METHODS

Cell cultures. We used the NlE-115 and NlA-103 clones deriving from mouse neuroblastoma Cl300 [l]. As previously described [13], cells were grown attached in Falcon tissue culture dishes in Dulbecco’s modified Eagle’s medium (DMEM) containing 7.5% fetal calf serum, supplemented with 0.2 mg/ml lincomycin, 0.2 mg/ml gentamycin, and 2.5 mg/ml fungizone, under 7% CO, at 37°C. Chemical differentiation was induced by addition of 10m7 M adriamycin (Adriblastine, Laboratoire Roger Bellon, Neuilly-stir-Seine, France), or 6 mM CCA (Aldrich), or 2% (250 mM) DMSO (Sigma) at the complete medium. Incorporation of [sH]thymidine. The cells, grown in 35mm tissue culture dishes, were labeled (after 23,47,71, or 95 h of culture) in 1 ml of medium containing 0.185 MBq of [3H]thymidine (5 &i/ml) diluted in 10-r M thymidine. After 1 h of labeling, cells were washed twice with PBS, trypsinized, pelleted, and kept at -20°C in 50 (.d PBS. TCA insoluble material present in the pellet was then quantified following classical procedure. Forty microliters of each sample was spotted onto a GFC filter and air dried. The filters were washed three times in TCA (5%, w/v) for 10 min at 4”C, then twice in 100% ethanol, and air dried and the radioactivity was counted. Analysis of cellular DNA content. The DNA content of the cells was determined by flow cytometry, after staining with ethidium bromide. Cells were maintained in tissue culture dishes for 4 days in the presence of adriamycin. They were then treated as described [32], with minor modifications, and analyzed using a Cytofluorograph 11s (Ortho Diagnostics Systems). Electrophysiological analysis. Cells were grown in 35.mm tissue culture dishes for 4 days in the presence of adriamycin, CCA, or DMSO. Electrophysiological analysis was previously described in [33], and in [34] for the Ca2+ channels study. Electrophoresis and immunoblotting. Cells were grown in 145mm tissue culture dishes for 4 days in the presence of adriamycin, CCA, or DMSO. Cells were washed with PBS and lysed by addition of 1% NP-40 and protein amounts were determined according to Bradford [35]. Proteins (80 pg) of each cell extract were analyzed by one- or twodimensional gel electrophoresis. The SDS-PAGE was performed according to Laemmli [36], in a gel containing 15% acrylamide and 0.2% bisacrylamide. The 2-D PAGE was performed according to O’Farrell [37] with modifications [38]. The SDS polyacrylamide gel for the second dimension contained 8% acrylamide and 0.11% bisacrylamide. Jmmunoblotting was performed according to Towbin [39], using rabbit anti-peripherin antiserum (a gift from Dr. M. M. Portier,

73

DIFFERENTIATION

Paris) or rabbit anti-y-enolase antiserum (a gift from Dr. N. Lamande, Paris), or mouse anti-@-tubulin antiserum (Amersham), and peroxidase-conjugated antirabbit or antimouse globulin serum (BioSys). Immunoreactivity was revealed by using “Luminol reaction” (ECL kit, Amersham). Autoradiographs corresponding to immunoblots, obtained after “Luminol” reaction, were quantified by densitometric scanning (Vernon integrating densitometer).

RESULTS

AND

DISCUSSION

We have studied the adriamycin-induced differentiation of NlA-103 neuroblastoma cells known as a “neurite minus” cell line [4]. At the same time, we have compared the effects of some chemical inducers on NlA-103 cells and NlE-115 cells, a cell line also derived from the Cl300 clone [l, 41, which can be morphologically differentiated with chemical inducers or by serum withdrawal [ 141. Previous studies had shown that NlA-103 cells, after treatment with chemical inducers or serum deprivation, enter into a postmitotic phase but do not exhibit morphological differentiation [4, 13, 141. Diaz-Nido et al. [ 151 have recently observed morphological differentiation of these cells following serum deprivation and treatment with hydroxyurea (an antineoplastic drug that interferes with DNA synthesis). Here, we have studied the effects of adriamycin, a molecule known for its action on DNA synthesis, its target being topoisomerase II [26]. Morphology and Rates of DNA Synthesis As previously described [4], the cell body diameter was smaller in NlA-103 growing cells (Fig. 1A) than in NlE-115 (Fig. 1E). In both clones, untreated cells (Figs. 1A and 1E) exhibit an undifferentiated morphology, most of them being round shaped and some displaying poorly developed neurites [5]. Addition of adriamycin to logarithmically growing cells promotes morphological differentiation in both clones (Figs. 1B and lF), whereas addition of CCA (Figs. 1C and 1G) or DMSO (Figs. 1D and 1H) induce neurite outgrowth only in the NlE-115 cells (Figs. 1G and 1H). In adriamycin-induced NlA-103 cells (Fig. IB), the cell body diameter rapidly (24 h after treatment) and dramatically (fivefold) increases, and, after 72 h of culture, the cells extend stable neurites of a length corresponding roughly to twice the cell body diameter. In NlE-115 cells, adriamycin also promotes morphological differentiation (Fig. 1F) with an increase of the cell body diameter and very long branched neurites. The cell diameter increases after 24 h of treatment and neurite formation starts after 48 h. With the formation of a dense neurite network, adriamycin-induced morphological differentiation is quicker than is obtained with inducers previously used, such as CCA (Fig. 1G) [7] or DMSO (Fig. 1H) [5].

74

LARCHER

ET

AL.

ADRIAMYCIN-INDUCED

NlA-103

NEUROBLASTOMA

crease of cellular proliferation occured, followed, after 3 days, by neuritogenesis. Thus, induction of neuroblastoma cell differentiation requires an effect on DNA synthesis, as also shown by Minana [40] in the N2a cell line using actinomycin D. In the case of NlA-103 cells, and, as shown by Diaz-Nido et al. [15], addition of hydroxyurea alone, although it also blocks the cell cycle, is not sufficient in itself to induce morphological differentiation, and must probably be complemented by another event(s) that would take place during adriamycin induction. Hence, it is likely that the type of blockade exerted by various agents on the neuroblastoma cell cycle is of importance in the triggering of neurite extension. Moreover, since CCA and DMSO are slower to block DNA synthesis (Fig. 2), a longer treatment of NlA-103 cells with these agents does not promote neurite outgrowth (data not shown).

NlA-103

a E 6

100

75

DIFFERENTIATION

NlE-115 a

Cell Cycle Analysis

0

24

48 Hours

after

72

96

induction

FIG. 2. Rates of DNA synthesis during the adriamycin (A), CCA (C), or DMSO (D) induction of NlA-103 and NlE-115 cells. Synthesis of DNA was measured by incorporation of [“Hlthymidine after a l-h pulse. The results are expressed as a percentage of the values obtained for growing cells in complete medium. Each point corresponds to the average of three independent experiments.

The stage of the cell cycle at which cell growth is arrested has been investigated by cytofluorometry. In both clones, a previous study had shown that DMSO treatment promoted arrest of the cell cycle in the GO/G1 stage, whereas CCA blocked most of them in GO/G1 phase and a few in G2 [14]. After 4 days of adriamycin treatment, NlA-103 cells were arrested in S stage while most of the NlE-115 cells were blocked in S phase and some in GO/G1 (Fig. 3). Since the adriamycin effect has been ascribed to an inhibition of topoisomerase II [26], an enzyme involved in the separation of the DNA strands during replication, blockade of adriamycin-induced cells in S phase is in good accordance with its postulated molecular target. Ionic Channel Expression NlE-115 Cells

At various times following induction, synthesis of cellular DNA was measured by incorporating [3H]thymidine during a l-h pulse (Fig. 2). Adriamycin promotes a cell division blockade after 24 h in both cell lines. With CCA or DMSO, NlA-103 cells are more resistant than NlE-115 to the effects of inducers; arrest of division occurs after 48 (CCA) or 92 h (DMSO) in the case of NlA-103 while taking place after only 24 h in NlE-115 cells. Nuclear [3H]thymidine incorporation decreases in all cells as observed by autoradiography (data not shown), indicating a homogeneous response to the chemical inducers. When NlA-103 cells were induced by adriamycin alone in serum supplemented medium, a very rapid de-

in Induced

NIA-103

and

Previous studies had shown that NlA-103 cells were less excitable than NlE-115, and that, after induction, membrane excitability was enhanced only in NlE-115 cells [4,8]. Noninduced or induced NlA-103 and undifferentiated NlE-115 cells have an identical sodium response to the depolarization (Fig. 4A), whereas differentiated NlE-115 cells exhibit an enhanced sodium response and, in addition, a calcium response (figure 4B). The nature of these responses was confirmed by using tetrodotoxin (TTX) and Co’+ ions, specific and reversible inhibitors of the sodium and calcium channels, respectively (data not shown). In both logarithmically growing and induced cells, we found Na+ and Kt channels (data not shown). Two dif-

FIG. 1. Effect of adriamycin, CCA, or DMSO induction on neuritogenesis in NIA-103 and NlE-115 neuroblastoma cells. NlA-103 and NlE-115 cells (E-H) were cultured as described in the text in complete medium (A and E) and in the presence of 10m7 M adriamycin F), or 6 mM CCA (C and G), or 2% DMSO (D and H). Cells were photographed after 4 days culture and/or treatment (X200).

(A-D) (B and

76

LARCHER

ET

AL.

Nl A-103

Nl E-115 DNA

content

FIG. 3. Adriamycin effect on cell cycle progression in NlA-103 and NlE-115 cells. NlA-103 (A and B) and NlE-115 cells (C and D) were cultived 4 days in the absence (A and C) or in the presence (B and D) of 10m7 M adriamycin. The cells were permeabilized with Triton X-100, stained with ethidium bromide, and analyzed by cytofluorometrv for their DNA content. Each curve corresponds to the analysis of 20,000 cells. 2n corresponds to the GO/G1 stade and 4n to G2.

ferent types of Ca2+ channel are expressed in neuroblastoma cells [41]. One, called the transient channel, is found in all NlA-103 and NlE-115 cells (Fig. 5), and the second, the so-called long-lasting channel, is found only in induced NlE-115 cells (Fig. 5B) and corresponds to calcium channels present in presynaptic terminals [41]. Long-lasting Ca2+ channels were not present in the adriamycin-differentiated NlA-103 cells. Therefore, their presence in NlE-115 cells is not directly associated, as previously supposed, with mere neurite extension; however, the type of neuritogenesis that is achieved after induction can be a relevant parameter in the expression of these Ca2+ channels.

lHS

IL.-h--@

200 ms FIG. 4. Adriamycin-treated NlA-103 and NlE-115 cells action potentials. (A) 4-day-adriamycin-treated NlA-103 cells. (B) 4-dayadriamycin-treated NlE-115 cells. Action potential was stimulated by a brief depolarizing pulse of current (see Materials and Methods). Membrane potentials have been previously adjusted to a steady level of about -80 mV. C and S, calcium and sodium response to the depolarization.

Peripherin NlA-103

and y-Enolase and NIE-115

Expressions Cells

in Induced

Cellular amounts of peripherin, a neurospecific intermediate type III filament [42], and of y-enolase (neuron-specific enolase or NSE) were analyzed by Western blotting (Fig. 6). Equal quantities of crude extract proteins (prepared from neuroblastoma cells induced for 4 days with adriamycin, CCA, or DMSO) were subjected to SDS-PAGE and antigens were detected, after transfer onto nitrocellulose membranes, using specific rabbit polyclonal antibodies [43, 441. In all cases analyzed (Figs. 6C and 6D), the amount of peripherin increased about twofold compared to control cells. This stimulation could be related to entry into the postmitotic phase and to the increase in cell body diameter, both involving modifications in cytoskeletal composition. No change was observed in the amount of y-enolase following chemical induction of NlA-103 cells (Fig. 6A), whereas, in all induced-NlE-115 cells, a twofold increase took place using growing cells as a reference (Fig. 6B). A previous study had shown an isoform switch from o(- to y-enolase during synaptogenesis, possibly related to alterations in intracellular chloride levels [45]. We can surmise that, in the case of the NIE-115 cells, this isoform switch is also related to a more complete neuronal maturation, i.e., the stimulation of membrane excitability, the presence of long-lasting Ca2+ channels, the expression of synaptophysin (a synaptic vesicle membrane protein, data not shown), etc. However, this switch did not occur in adriamycin-induced NlA-103 cells.

ADRIAMYCIN-INDUCED

NlA-103

NEUROBLASTOMA

% % l-

100 ms

DIFFERENTIATION

77

P ,I 200 ms

in adriamycin-induced NlA-103 (A) and NlE-115 cells (B and C). Inward FIG. 5. Transient and long-lasting components of Ca2+ current membrane Ca*+ currents, recorded with the whole-cell patch clamp technique, in response to pulses from a holding potential of -90 mV (A and B) or -50 mV (C) to the indicated membrane potential (in mV). Left scale corresponding to the A recording and right scale to B and C in B a transient and the appearance of a long-lasting component, and recordings. In A, we can observe a transient component of Ca*+ current, in C, with a change in the holding potential, only the long-lasting component.

p-Tub&in

Analysis

Using two-dimensional gel electrophoresis followed by immunoblotting, we attempted to characterize 01and /3-tubulin isoforms present in neuroblastoma cells. No change was observed in the a-tubulin isoforms expressed in induced or noninduced NlA-103 and NlE115 cells (data not shown here), whereas interesting modulations were observed in the distribution of some @-tubulins isoforms. Figure 7 illustrates the P-tubulin pattern from NlA103 and NlE-115 neuroblastoma cells grown in complete medium or induced for 4 days by adriamycin or DMSO. In growing and DMSO-treated NlA-103 cells (Fig. 7A and 7C), we observe four fl-tubulin isoforms (p’l, 0’2, p3, and p5), whereas in adriamycin-induced cells (Fig. 7B) an acidic isoform (p7) appears between 24 and 48 h after the beginning of induction (data not shown). This isoform, constitutively expressed in NlE-

115 cells (Figs. 7D-7F), has been reported to appear during the early postnatal development of mouse brain [46] and corresponds to a post-translationally modified isoform [47]. Surprisingly, we did not observe a shift in the proportion of the 0’1 isoform to p’2 during adriamycin-induced NlA-103 cell differentiation (Fig. 7B), as observed in chemically induced NlE-115 cells (Figs. 7E and 7F). This previously described isoform [48] corresponds to a phosphorylated derivative of p’l [49]. This situation in NlA-103 cells-the presence of four P-tubulin isoforms (p’l, p3, p5, and /37) and the absence of a shift to /3’2-is similar to that observed in induced teratocarcinoma cells that exhibit neuritogenesis [50] or in NlE-115 undifferentiated cells. Thus, distinct combinations of ptubulins are expressed in different neuronal cell lines during neurite extension. This observation suggests several possible explanations: (i) /3-tubulin isoforms are interchangeable, (ii) the absence of some /3-tubulin iso-

78

LARCHER

2.0 ii z 1.0

2.0

1.0

AL.

. . : :: :::: ’ ::::, ::::: : ::y:::: .I,” .I,an :: .. .. .. I,I,, ::: .. .. .. :::: ‘I :: ::: ;I,: : : .. . : :: : :::., . . . : : :: :::.: . . . . . .: ::’ :: ::: . . . ‘I”

D i-Q--L

.r cG .o . If

ET

:: 1:

::: ::: : : ::: : : .. .. .. ::::’ :::*:2, a: :::., :::: : : .. .. j :

dlI G

A NlA-103

C

0

G

A

c

0

1

NlE-115

neuritogenesis. In particular, they illustrate situations in which entry into a postmitotic phase is either accompanied (NlE-115 or adriamycin-induced NIA-103) or not (NlA-103) by morphological and electrophysiological differentiation. The comparison between these different culture conditions allows us to separate the effects due to the arrest of division (peripherin accumulation, modification in the resting membrane potential), from those related to morphological differentiation (inversion in the fl’l-tubulinlfi’2-tubulin ratio or P7-tubulin appearance) or to the electrophysiological differentiation (modifications of the action potential associated with the appearance of long-lasting Ca2+ channels, y-enolase, and synaptophysin accumulation). Thus, neuroblastoma cells constitute a good model to analyze, starting from commited cells, how the arrest of cellular division leads to a stepwise program of an almost complete neurogenesis, leading these cells to a state that is close to that of mature neurons, but without functional synapse formation.

FIG. 6. Expression of NSE and peripherin in NlA-103 and NIE115 cells after 4 days of treatment with adriamycin (A), CCA (C), or DMSO (D). NlA-103 and NlE-115 cells were cultived 4 days in the absence (G) or in the presence of 1O-7 M adriamycin, 6 mM CCA, or 2% DMSO. 80 pg of total proteins were separated on 1-D SDS-PAGE (see Material and Methods). After blotting, NSE or peripherin were detected by using monospecific rabbit polyclonal antibodies. NSE or peripherin quantifications were obtained by electrophoregram scanning. Results are expressed as a ratio of the values obtained from treated cells and growing cells (G).

forms is balanced by the expression of other proteins, i.e., acidic ol-tubulins in differentiated teratocarcinoma cells [50], or (iii) the nature or the properties of the neurites are not equivalent from one cell line to the other. We have analyzed changes in synaptophysin in the two neuroblastoma cell lines at different developmental stages. Synaptophysin is a synaptic vesicle membrane component [51], expressed during neurite outgrowth [52]. This protein was constitutively expressed in NlE115 cells, and its expression was stimulated after chemical induction (data not shown) whereas it was not detected in treated or noninduced NlA-103 cells. The absence of the long-lasting Ca2+ channel and of synaptophysin in NlA-103 cells, as well as the lack of stimulation of y-enolase after induction, could be related to the type of neuritogenesis that is induced in adriamycin-treated NlA-103 cells as compared to those found in NlE-115. In NlA-103 cells, adriamycin could drive the growth of neurite-like spikes. The presence of P7-tubulin could promote this growth or could be involved in the stabilization of the microtubules in these spikes. In conclusion, NlA-103 and NlE-115 neuroblastomas, which are committed neuronal cell lines, offer an opportunity to dissect in some detail the various steps of

IEF

OH-

H+

P A G E

NlA-103

NlE-115

FIG. 7. Comparative analysis of /+tubulin expression in NlA103 and NlE-115 cells growing in complete medium or 4-day-induced by adriamycin or DMSO. NlA-103 (A-C) and NlE-115 cells (D-F) were cultured 4 days in the absence (A and D) or in the presence of 1O-7 A4 adriamycin (B and E) or 2% DMSO (C and F). 80 rg of total proteins were separated on 2-D SDS-PAGE (see Material and Methods). After blotting, P-tubulins were detected by using monospecific mouse monoclonal antibody (Amersham). Only the relevant areas of the immunoblots are shown. Circles indicate that P7-tubulin is only expressed in adriamycin-induced NlA-103 cells (B) and in all NlE115 cells (D-F).

ADRIAMYCIN-INDUCED

NlA-103

NEUROBLASTOMA

We thank Dr. M. M. Portier for the gift of the anti-peripherin antibody, Dr. N. Lamande for the gift of the anti-NSE antibody, and Dr. P. Metezeau (Institut Pasteur, Paris) for cell cycle analysis. We acknowledge Dr. Ph. Denoulet and Dr. J. Cohlberg for critical reading of the manuscript. This work was supported by grants from the AFM and the LNFC. J.C.L. and L.C.-L. are recipient of an AFM fellowship.

REFERENCES 1. 2. 3. 4.

Augusti-Tocco, G., and Sato. G. (1969) Proc. Natl. Acad. Sci CJSA 64, 311-315. Pochedly, C. (1977) in Neuroblastoma (Pochedly, C., Ed.), pp. 273-306, Edward Arnolt Ltd., London. Seeds, N. W., Gilman, A. G., Amano, T., and Nirenberg, M. W. (1970) Proc. Natl. Acad. Sci. USA 66, 160-167. Amano, Acad.

5. 6.

T., Richelson, Sci.

USA

8.

M. (1972)

Proc.

258-263.

Rex

Commun.

76,937-942.

Croizat, B., Berthelot, F., Ferrandes, B., Eymard, P., and Sahu quillo, C. (1979) C. R. Acad. Sci. Paris 289, 1283-1286. Zisapel,

N., and Littauer,

IJ. Z. (1979)

Eur.

J. Biochem.

95,

51-

59.

9.

10.

Vayssiere, J. L.. Larcher, J. C., Berthelot, F., Benlot, C., Gros, F., and Croizat, B. (1986) Biochem. Biophys. Res. Commun. 140, 789-796. Monard, D., Niday, E., Limat, A., and Solomon, F. (1983) Pro&. Brain

11.

Res.

Gurwitz, IJSA

12.

14. 15.

359-364.

D., and Cunningham, 85,

Vayssiere, Biochem.

13.

58,

D. D. (1988)

Proc. Natl.

Acad.

J. L., Larcher, Bioph,y.s.

Res.

J. C., Gros, F., and Croizat, Commun. 145, 443452.

B. (1987)

Phillips, D. R., White, R. J., and Cullinane, C. (1989) FEBS Lett. 246, 123331240. Larcher, J. C., Vayssiere, ,J. L., Lossouarn, I,., Gros, F., and Croizat, B. (1991) Oncogene 6, 633-638. Diaz-Nido, J., Armas-Portela, R., and Avila, J. (1992) J. Neurothem.

58,

1820-1828.

16. 17.

Henry, D. W. (1979) Cancer Treat. Rep. 63, 845-854. Wiernik, P. H. (1980) in Anthracyclines: Current Status and New Developments (Crooke, S. T., and Reich, S. D., Eds.), pp. 2733294, Academic Press, New York.

18.

Arcamone, F. (1981) Doxorubicin: demic Press, New York.

19.

Schwartz, H. S. (1983) in Molecular Aspects of Anticancer Design (Neidle, S., and Waring, M. J., Eds.), pp. 93-125, Millan, New York.

20. 21.

Brown, J. R. (1978) Prog. Med. Chem. 15, 125-164. Gianni, L., Croden, B. J., and Myers, C. E. (1989) in Review in Biochemical Toxicology (Hodgson, E., Bend, J., and Philpot, R. A., Eds.), Vol. 5, pp. l-82, Elsevier, New York. Cheneval, D., Muller, M., Toni, R., Ruetz, S., and Carafoli, E. (1985) J. Biol. Chem. 260, 13003-13007. Myers, C. E., Muindi, J. R. F., Zweier, J., and Sinha, B. K. (1987) J. Biol. Chem. 262, 11671-11577.

22. 23.

Received Revised

March version

13, 1992 received July

21, 1992

Anticancer

Antibiotics,

Aca-

P., Praet,

E., Huart, J.-M. (1990)

Biophys.

M.,

Brasseur,

Chem.

35,

R.,

and

247-257.

Tewey, K. M., Rowe, T. C., Yang, L. F. (1984) Science 226, 4666468.

26.

Tarr, M., 141-146.

27.

Valentini, L., Nicolella, V., Vannini, E., Menozzi, M., Pence, and Arcamone, F. (1985) Farmaco Ed. Ski. 40, 377-390.

28.

Maehara, Tsuji-Tani, 327.

29. 30.

and

Helden,

L., Halligan,

P. D. V. (1990)

Y., Emi, Y., Anai, S., and Sugimachi,

Mot.

B. D., and Liu. Cell.

Biochem.

H., Sakagushi, Y., Kohnoe, K. (1989) J. Pathol. 159,

Portier, M. M., Croizat, B., and Gros, F. (1982) FEBS 2833288. Cullinane, C., and Phillips, D. R. (1990) Biochemistry 5646.

M. (1982)

Brain

Cognard,

G. (1986).

Proc.

C., Lazdunski, M., (ISA 83, 517-521. M. M. (1976) Anal. IT. K. (1970) Nature

Oncology Pathol.

and Romey,

S., S.,

32% 146, 563%

Vindelov, L. (1977) Virchoas Arch. B. Cell. Tourneur, Y., Romey, G., and Lazdunski, 245, 154-158. Sci.

B. A. (1982)

29,

32. 33.

Acad.

and Sheffler,

Lett.

Schengrund, 190.

34.

C.-L.,

93,

31.

39,185-

24,

227-242. Res. Natl.

35. 36.

Bradford, Laemmli,

37. 38.

O’Farrell, I’. H. (1975) J. Biol. Chem. 250, 400774021. Boyer, S., Maunoury, R., Gomes, D., Nechaud. B. D., Hill, A. M., and Dupouey, P. (1990) J. Neurosci. Res. 27, 55-64. Towbin, H., Staehelin, T., and Gordon, ,J. (1979) Proc. Natt.

39.

Acad.

Sci.

IJSA

76,

Biochem. 227,

72,

248-254.

680-685.

4350-4354.

40.

Minana, M. D., Felipo, 280, 245-246.

41.

Narahashi, 383,231-249.

42.

Portier, M. M., Croizat, B., Nbchaud, B. D., Gumpel, M., and Gras, F. (1983) C. R. Acad. Sci. Paris 297, 57-61. Lamande, N., Zeitoun, Y., Gras, F., and Legault, L. (1985) Neuro-

Sci.

3440-3444.

Goormaghtigh, Ruysschaert,

25.

Natl.

Kimhi, Y., Palfrey, C., Spector, I., Barak, Y., and Littauer, U. Z. (1976) Proc. Natl. Acad. Sci. lJSA 73, 462-466. Palfrey, C., Kimhi, Y., and Littauer, U. 2. (1977) B&hem Biaphys.

7.

69,

E., and Nirenberg,

24.

79

DIFFERENTIATION

43.

them.

44. 45.

46. 47.

Drug Mac 48.

Int.

T.,

Tsunoo,

A., and

Grisolia, Yoshii,

S. (1991) M.

(1987)

FEBS J.

Lett. Physiol.

7, 867-874.

Escurat, M., Djabali, K., Gumpel, M., Gras, F., and Portier, M. M. (1990) J. Neurosci. 10, 764-784. Marangos, P. J. (1988) in Neuronal and Glial Proteins (Marangas, P. .J., Campbell, I. C., and Cohen, R. M., Eds.), pp. 120-136, Academic Press, New York. Wolff, A., Denoulet, P., and Jeantet, C. (1982) Neurosci. Lrtt. 3 1,323-328. Denoulet, I’., Edde, B., Pinto-Hentique, D., Koulakoll’, A., Berwald-Netter, Y., and Gros, F. (1988) in Structure and Functions of the Cytoskeleton-Biological and Physiopathological Aspects (Rousset, B. A. F., Ed.), pp. 231-237, John Libbey Eurotext, London/Paris. Edde,

B., Jeantet,

49.

Gard, D. L., and 764-174.

50.

Edde, 1478.

51.

Jahn, Proc.

C., and Gras,

F. (1981)

Biochem.

Biophys.

Rcs.

Biol.

100,

103, 103551043.

Commun.

52.

V., and

B., Jakob,

Kirschner,

H., and Darmon,

R., Schiebler, Natl.

Wiedenmann,

Acad.

M.

W., Ouimet, Sci.

USA

B., and Franke,

82,

W. (1985)

J. Cell.

M. (1983)

EMBOJ.

2,1473-

C., and Greengard, P. (1985) 4137-4141. W. W. (1985) Cell 41,1017-1028.