The plasma membrane NADH-diaphorase is active during selective phases of the cell cycle in mouse neuroblastoma cell line NB41A3. Its relation to cell growth and differentiation

BiBttoch,~Pic~a et Biophysica AEta ELSEVIER

Biochimica et Biophysica Acta 1312 (1996) 215-222

The plasma membrane NADH-diaphorase is active during selective phases of the cell cycle in mouse neuroblastoma cell line NB41A3. Its relation to cell growth and differentiation Rinaldo Zurbriggen, Jean-Luc Dreyer

*

Department of Biochemistry, Universi~ of Fribourg, CH-1700 Fribourg, Switzerland

Received 1 March 1996; accepted 20 March 1996

Abstract Plasma membrane oxidoreductases have been described in all cells and use extracellular impermeant electron acceptors (DCIP, Ferricyanide) that are reduced by NADH. They appear to regulate the overall cell activity in response to oxidative stress from the cellular environment. An NADH-DCIP reductase has been described at the plasma membrane of NB41A3, a neuroblastoma cell line (Zurbriggen and Dreyer (1993) Biochim. Biophys. Acta 1183, 513-520) whose activation with extracellular impermeant substrates promotes cell growth. Elutriation was performed to separate cells and the various fractions were analysed for enzyme activity on intact cells combined with flow cytometry. These studies showed that the enzyme is mostly induced and activated during the G1 and during the G2/M-phases. These observations were further corroborated with specific inhibitors of the cell cycle. A three-fold increase in enzyme activity was observed in the presence of c~-amanitin, a specific cell cycle inhibitor of the Gl-phase. Taxol, a specific inhibitor of the M-phase, also induces a significant increase in enzyme activity. FACS analysis of taxol-treated and a-amanitin-treated cells corroborated these data. The cells have been synchronized and the enzyme activity was measured at different time intervals. An activity increase was observed after ca. 2-3 h, that corresponds to a raise in the M-phase, according to FACS data. Furthermore, NTera-2 cells - a human neuroblastoma cell line that differentiates into fully mature neurones in the presence of retinoic acid - exhibit a 50% decrease in the enzyme activity during the G0-phase upon differentiation, compared to undifferentiated cells. Together the data presented in this paper show that this plasma membrane NADH-diaphorase affects cell growth and differentiation and is strongly modulated at various phases of the cell cycle. Keywords: Plasma membrane redox; Cellular activation; Diaphorase; Cell cycle

1. Introduction Plasma membrane redox enzymes have been demonstrated in all cell types but their real function is not yet established [1-3]. In response to hormones and other growth factors, they transfer electrons from reducing agents in the cytoplasm (e.g. N A D H ) to external impermeable oxidants such as ferricyanide or other externally generated substrates (e.g. ascorbate free radicals, thiols) as an electron acceptor [4]. Plasma membranes contain b-cytochromes, flavin, iron, and quinones, therefore components for

Abbreviations:DC1P, dichlorophenol-indophenol;DMEM, Dulbecco's modifified Eagle's medium. * Corresponding author. Fax: +41 37 299735; e-mail: [email protected]. 0167-4889/96/$15.00 © 1996 Elsevier Science B.V. All fights reserved PII S0167-4889(96)00037-7

electron transport are present but their participation, except for quinone, has not been demonstrated [2,4]. Several systems are identified in animal and plant plasma membranes and include: NADH-diaphorase [5] and N A D H oxidase (controlled by growth hormones), iron reductases, N A D H - c y t o c h r o m e b 5 reductase, NADH-Ferricyanide reductase [6,7], peroxidase, superoxide-generating N A D P H oxidase, semi-dehydro-ascorbate reductase [8-11], xanthine oxidase [12], thioredoxin reductase [13], or redox enzymes involved in the photo-reduction in some phototropic cells [14], etc. In a different way, phagocytes possess a superoxide-generating oxidase that is essential for the efficient killing o f microorganisms [15] and is a special case of plasma membrane oxydoreductase. Reduced pyridine nucleotides are substrates for the plasma membrane dehydrogenases and natural ferric chelates or ferric transferrin can act as electron acceptors.

216

R. Zurbriggen, J.-L. Dreyer / Biochimica et Biophysica Acta 1312 (1996) 215-222

Stimulation of electron transport with impermeable oxidants and hormones activates proton release from cells [ 16]. Inhibitors of electron transport, such as certain antitumor drugs, inhibit proton release. The high ratio of protons released to electrons transferred, the stimulation of proton release by sodium ions and the inhibition by amiloride together suggest that electron transport activates the N a + / H ÷ antiport [16]. The nature and components of the electron transport system and the mechanism by which proton release is activated are unknown. Cell proliferation and a more oxidising state (e.g. a higher N A D ÷ / N A D H ratio) in the cytosol are associated in some mammalian cell types [17] and electron flow through the plasma membrane oxidoreductases also stimulates growth and cell proliferation [l 8]. Several studies have described the effects of cell transformation on plasma membrane oxidoreductases [19]. Transformation of 3T3 cells by SV40 virus changes the properties of a transplasma membrane electron transport activity. Bombesin, a mitogenic neuropeptide that stimulates proliferation of Swiss 3T3 cells, also stimulates transplasma membrane reduction of diferric transferrin or ferricyanide [20]. From these and other studies a function of transplasma membrane oxidoreductases in cell growth, proliferation, transformation and differentiation is implicated [3-5,9,18,21]. In addition activation of transplasma membrane oxidoreductases is also related to exocytosis of clathrin-coated vesicles as well as to receptor-mediated and adsorptive endocytosis [22]. In previous studies we have described a variety of isoforms of oxidoreductases in purified synaptic plasma membranes from rat or beef brain and their subunit composition has been characterised [23-25]. The presence of a NADH-DCIP reductase in the plasma membrane of a neuroblastoma cell line, NB41A3, and its purification have also been reported [5]. In NB41A3 the enzyme accounts for over one third of total cellular diaphorase and strongly affects cell growth. Low substrate concentrations promote a 75% increase in cell growth [5]. In the present report the function of this enzyme during cell proliferation and differentiation has been investigated by means of activity measurements on intact cells and by flow cytometry at different stages of development. From these studies, it clearly appears that the enzyme is highly activated at specific stages during the cell cycle and turned off after differentiation.

2. M n t e r i a l and m e t h o d s 2.1. Chemicals

et-Amanitin, taxol, dimethyl-sulfoxide, Triton X-100 and NAD(H) were purchased from FlukaTM. DCIP, DT-diaphorase (EC.1.8.1.4), and PMSF were from SigmaTM, growth media, DMEM, foetal calf serum, trypsin and PBS were from Seromed.

2.2. Growth o f neuroblastoma cells NB41A3 and NTera-2

NB41A3 cells (from ATCC) were grown according to Agustini and Sato [26]. Confluent monolayer cultures were trypsinized in the presence of 0.25% trypsin and 0.1% EDTA and the cell suspension was pelleted at 500 X g for 10 min. The pellet was washed with PBS, re-centrifuged and diluted in the same buffer to the appropriate final concentration. NTera-2 neuroblastoma cells (from ATCC) were handled under similar conditions. This cell line differentiates into neurones in the presence of 10 -5 M trans-retinoic acid. After incubation for 3 - 4 days in DMEM medium containing 10% FCS, cell differentiation is induced by incubation for 3 days in DMEM medium containing retinoic acid, followed by a medium change and a further incubation in the same medium containing retinoic acid for 4 more days. 2.3. Membrane N A D H - D C I P reductase activity measurements in cultured cells

NADH-DCIP reductase activity was tested according to [27]. In all cases, activity measurements were performed in triplicates. Ca. 10 6 intact cells in 800 Ixl PBS buffer were incubated for 1 min before addition of 100 Ixl of 1.3 mM NADH, followed after 30 s by the addition of 100 ixl of 0.44 mM DCIP. After incubation for 90 s at room temperature the absorption change at 600 nm was measured over 2 min. Where necessary, appropriate modifications were performed, as mentioned in the figures. 2.4. Elutriation

Elutriation was performed in a Beckmann JE-5.0 Elutriation system that has been cooled to 4°C and washed with cold PBS. At 2800 rpm the system was loaded with ca. 108 cells at a flow rate of 10.5 ml/min. The elutriation started at a flow rate of 31 m l / m i n and was terminated at 231 ml/min. 50 ml fractions were collected. The cells from each fractions were then centrifuged at 500 X g for 10 min and resuspended in 2 ml PBS. 2.5. FACS-analysis l 0 6 cells in 0.3 ml PBS (4°C) were mixed dropwise with 0.9 ml ice-cold ethanol. After incubation for 12 h at 4°C, cells were washed twice in PBS and suspended in 0.9 ml PBS. 0.1 ml propidium Iodide (50 m g / m l in 0.38 mM Na-citrate buffer, pH 7.0) was added, followed by 10 Ixl RNase (DNase-free, 1 m g / m l ) and the suspension was incubated for 30 min at 37°C. Thereafter cells were kept in the dark until use. RNase (10 m g / m l in 0.01 M Na-acetate, pH 5.2) had been freed of DNase by treatment for 15 min at 100°C and slow cooling up to RT, then neutralised with

R. Zurbri ggen, J.-L. Dreyer / Biochimica e: Biophysica Acta 1312 (1996)215-222 0260

addition of 0.1 vol 1 M Tris-HC1, pH 7.4 and kept at - 20°C. 2.6. Cell synchronisation This was performed according to Minana et al. [28]. To synchronise at G 0, 4 × 105 cells/ml per well were seeded in 24-well plates and cultured in MEM without glutamine and containing 0.5% inactivated FCS. After 48 h of incubation in a moist 5% CO2/95% air incubator at 37°C, most of the cells were in the G o phase. The G o synchronous culture was obtained by transferring the resting cells into complete medium (MEM containing 10% FCS). To synchronise cells at the G t / S boundary, 1 mM hydroxyurea was added to the above culture 8 h after the medium exchange. After 14 h of incubation, the cells were washed and transferred to fresh complete medium. After different incubation times, (as indicated in the figures) cells were then trypsinized, washed trice with PBS and intact cells were then tested for DCIP-reductase activity. 2.7. Other assays Protein concentrations were determined according to the micro-BCA method of Pierce TM or according to Bradford [29].

3. Results

NB41A3 cells are neuroblastoma cells that are easily cultivated and grown and display high amounts of plasma membrane NADH-DCIP reductase activity [5]. During our studies with NB41A3 cells, we observed that the enzyme activity fluctuates from probe to probe, depending upon the culture conditions. Enzymatic activity was lowest at confluent cell's concentration in culture dishes. On the basis of this observation we; investigated the fluctuation in activity during cell cycle.

217

i

0.160

~' 0.100 0.050

O.O(X)

~ 50

i 100

i 160

i 200

260

now (mienM)

Fig. 1. Elutriation of NB41A3. About 108 cells were submitted to elutriation (see Section 2). Intact NB41A3 ceils near confluency were applied in PBS at a flow rate of ca 10.5 m l / m i n and elutriation was initiated at an initial flow rate of ca. 30 ml/min. Each collected fraction (50 ml) was then centrifuged at 500× g for 10 min the pellet was resuspended in 2 ml PBS and the NADH-DC1P reductase activity was measured on intact cells together with protein determination, as described in Section 2. Activity is expressed in p+mol DCIP reduced per min per mg of protein. Each measurement has been performed at least in triplicates.

relationship between the plasma membrane NADH-DCIP reductase activity and cell division can be inferred. To further investigate this correlation cells were submitted to FACS analysis (Fig. 2). First a FACS analysis of the whole cell population before elutriation was undertaken as a control (Fig. 2A). These cells displayed a normal cell cycle pattern with cells mostly in the G1 phase, but also in the S phase and significantly in the G 2 / M phase. Thereafter various fractions from the elutriation experiments were also submitted to FACS analysis. Pooled fractions at 31 to 42 m l / m i n of the elutriation experiment, corresponding to the initial peak of NADH-DCIP reductase activity, were analyzed, showing that cells are almost completely in

I el

. . . . . . . .

t~a,i

. . . . . . . .

le~

PZ

®,~

'

'

;

'

" '

PI

'

3.1. Elutriation and FACS-analysis At near confluency cells were separated by means of elutriation (Fig. 1). Intact cells from the first fractions to be elutriated displayed highest plasma membrane NADHDCIP reductase activity, about 0.2 m U / m g protein. This early peak corresponds to the lightest cells that mainly had undergone cell division. In the following elutriation fractions the NADH-DCIP reductase activity measured on intact cells was only ca. 0.03 m U / m g , ca. 15% of the activity displayed in the initial peak On the other hand intact cells from the fractions elutriated last at flow rates from 130 to 250 m l / m i n , and corresponding to cells shortly before division, again displayed a very high activity of ca. 0.2-0.25 m U / m g protein. From these data a

~d

P!

PI

Fig. 2. FACS analysis of NB41A3 cells after elutriation. Fractions after elutriation were pooled, prepared for FACS analysis and submitted to FACS analysis according to Section 2. (A) Control analysis of NB41A3 cells before elutriation; (B) pooled fractions of intact NB41A3 cells elutriated at a flow rate of 31-42 ml/min; (C) pooled fractions of intact NB41A3 cells elutriated at a flow rate of 42-78 ml/min; (D) pooled fractions of intact NB41A3 cells elutriated at a flow rate of 78-231 ml/min. PI = fluorescence intensity.

R. Zurbriggen, J.-L. Dreyer / Biochimica et Biophysica Acta 1312 (1996) 215-222

218

the Gl-phase (Fig. 2B) in strong contrast to the control. These cells display little S-phase and virtually no G 2 / M phase. On the other hand, cells from fractions at 42 to 78 ml/min, i.e. fractions that displayed lowest plasma membrane NADH-DCIP reductase activity, are found mainly in the S-phase (Fig. 2C). Finally cells elutriated last (fractions at 78 to 231 ml/min), that displayed high plasma membrane NADH-DCIP reductase activity, were almost completely found in the G 2 / M phase (Fig. 2D).

i

,

,

J

|

i

,

i

,

i

i

i

l

PI

i

i

i

i

PI

"

D

3.2. Cell cycle inhibitors To further corroborate these data we investigated the effects of some cell cycle inhibitors on the plasma membrane NADH-DCIP reductase and on the cell cycle. The cell cycle has been blocked at various stages by means of specific inhibitors. Shortly before activity measurements, cells treated with the inhibitors were washed to remove the cell cycle inhibitor and to prevent unspecific interferences with the assay, e~-amanitin is a specific inhibitor of RNA polymerase II and therefore is a specific cell cycle blocker at the G1 phase [28]. In the presence of a-amanitin intact NB41A3 cells display a drastic increase in plasma membrane NADH-DCIP reductase activity (Fig. 3). The raise in activity is concentration dependent, with a 300% increase in enzyme activity being observed with 10 p~g/ml e~amanitin. FACS analysis of cells treated with increasing concentrations of o~-amanitin were performed (Fig. 4A-D). Untreated control cells display a normal FACS pattern (Fig. 4A) whereas e~-amanitin treated cells display a severe drop of both the S-phase and the G2/M-phase, concomitant with an increase of the Gl-phase (Fig. 4B-C). The effect is concentration dependent and parallels the observed ac-

i

i

u

i

i

i

i

i

s

PI

P][

Fig. 4. FACS analysis of ~x-amanitin-treated NB41A3. Cells were treated as described in Fig. 3. After 48 h in the presence of a-amanitin cells were collected, washed and submitted to FACS analysis as described in Section 2. (A) untreated cells as a control; (B) cells treated with 2,5 ~ g / m l cx-amanitin; (C) cells treated with 5 ~ g / m l a-amanitin; (D) cells treated with 10 Ixg/ml a-amanitin. PI = fluorescence intensity.

tivity increase of the plasma membrane NADH-DCIP reductase. Another possible cell cycle blocker at the Gl-phase is DMSO. This reagent has been used as cell cycle blocking agent in various cells [30-33]. NB41A3 cells incubated in the presence of 1.5% DMSO for 96 h also displayed a 46% increase in plasma membrane NADH-DCIP reductase activity over controls, confirming the results obtained with a-amanitin. The cell cycle can also be specifically blocked in the M phase by means of taxol [32] which blocks the mitosis in the transition from the metaphase to the anaphase. In the presence of 10 nM taxol a 45% increase in plasma membrane NADH-DCIP reductase was found with intact

÷

oit

U

-I--

I"

O.8

÷

0.4

0.4 0.1 O2

0.1.

0.1

0 0

1

5

0.0

:

:

I

Fig. 3. Plasma membrane NADH-DCIP reductase activity of NB41A3 cells treated with c~-amanitin. About 106 NB41A3 cells were grown for 48 h in the presence of cx-amanitin at the indicated concentrations. Then the cells were trypsinized, collected, washed trice in PBS to remove a-amanitin and the NADH-DCIP reductase activity was measured on treated, intact NB41A3 cells. Activity is expressed in ixmol DC1P reduced/min per mg of protein. Each measurement has been performed at least in triplicates and the mean values with SD are presented.

I

S

I

10

T ~ (hi)

Fig. 5. Plasma membrane NADH-DC1P reductase activity of NB41A3 cells treated with taxol. 106 ceils were grown for 48 h in the presence of taxol at the indicated concentrations. Then the cells were trypsinized, collected, wahsed trice in PBS to remove taxol and the NADH-DCIP reductase activity was measured on intact cells. Activity is expressed in p~mol DCIP reduced/min per mg of protein. Measurements have been performed at least in triplicates. Mean values with SD are presented.

219

R. Zurbriggen, J.-L. Dreyer / Biochimica et Biophysica Acta 1312 (1996) 215-222

NB41A3 cells (Fig. 5). The activity increase is concentration-dependent up to ca. 10 nM taxol. At higher taxol concentration an activity increase is observed over controls also, but of smaller intensity, probably due to the toxic effects of taxol (data not shown). FACS analysis of taxol treated cells (between 1 and 10 nM taxol) were then undertaken (Fig. 6). Compared to untreated cells (Fig. 6A) that display a normal FACS profile, taxol-treated cells clearly show a blockage in the G 2 / M phase in a concentration dependent manner (Fig. 6 B - D ) . This indicates that blocking the cells at the G 2 / M phase induces an increase in the plasma membrane NADH-DCIP reductase activity. 3.3. Cell synchronisation

Cultured cells normally consist of a mixture of various cell populations at various stages in the cell cycle. To get a homogenous population, NB41A3 cells were synchronised in the G 1 / S phase, as described under Section 2. Thereafter the enzyme activity has been measured on intact cells at various times (Fig. 7) and FACS analysis has been performed (Fig. 8). After synchronization a first raise in the plasma membrane NADH-DCIP reductase activity was observed ca. 2 h after the G 1 / S transition, followed by a second smaller peak after ca. 5.5 h. In the first activity peak the specific activity almost doubled over controls, whereas during the second peak the increase in activity was about 25% over resting cells. Immediately after synchronization cells were mainly in the G 1 phase as expected (Fig. 8A). After ca. 2 h a raise in the G 2 / M - p h a s e was observed (Fig. 8B) that correlates with the raise in NADH-DCIP reductase activity from the first peak (Fig. 7). The G 2 / M - p h a s e then declines (Fig. 8C) in correlation

A

0.0

0.4

O2

0.1.

0.0

0

*

i

i

i

i

i

i

i

1

2

3

4

S

6

7

8

free a l b r syaehremlmdlel .mGtlS (11)

Fig. 7. Plasma Membrane NADH-DCIP reductase activity of NB41A3 cells after synchronisation.NB41A3 cells were synchronisedas described in Section 2 by growing with 0.5% FBS in MEM without glutaminefor 48 h, followed by medium change for 8 h and by treatment with 1 mM hydroxyurea for 14 h. After synchronisation,cell aliquots were collected at various times as indicatedand the NADH-DCIPactivitywas measured. Activity is expressed in ~mol DCIP reduced/rain per mg of protein. Each measurementhas been performed at least in triplicates and the mean values with SD are presented.

with the drop in enzyme activity. About 4 hours after G 1 / S transition, i.e. when the NADH-DCIP reductase activity is minimal (Fig. 7), an increase in the S-phase is

. 1

A

,hlSmin

"E |

~

5hl5minl

fl

,

i . . . . . . . . . . . . PI

Q

e

i

PI

m i

i PI

~ •

"

w d

........

td24 PI

|

|

a

•

|

i

PI

pT

i

•

|

m

O

.

i

,

i

|

PI

Fig. 6. FACS analysis of taxol-treated NB41A3. Cells were treated as described in Fig. 3. After a 48-h incubationin the presence of taxol, cells were trypsinized,collected and submittedto FACS analysis as described in Section 2. (A) control, untreated cells; (B) cells treated with 1 nM taxol; (C) cells treated with 5 nM taxol; (D) cells treated with 10 nM taxol. PI = fluorescenceintensity.

.~,

6h15rain

........... pT

Fig. 8. FACS analysis of NB41A3 cells after synchronisation,NB41A3 Cells were synchronizedas in Fig. 7. At various times, as indicated, the cells were removed from the medium,trypsinized,washed and submitted to FACS analysis. (A) FACS analysis immediatelyafter synchronization (t = 0 rain) (B) FACS analysisof NB41A3 cells 2 h after synchronization (C) FACS analysis of NB41A3 cells 2.75 h after synchronization(D) FACS analysis of NB41A3 cells 4.25 h after synchronization(E) FACS analysis of NB41A3 cells 5.25 h after synchronization(F) FACS analysis of NB41A3 cells 6.25 h after synchronization.

220

R. Zurbriggen, J.-L. Dreyer/ Biochimica et BiophysicaActa 1312 (1996) 215-222

Table 1 NADH-DCIP reductase activity in differentiated and undifferentiated NTera-2 cells NTera-2 cells NADH-DCIPreductase activity (mU/mg) differentiated undifferentiated

4.4 8.9

NTera-2 cells were grown in DMEM in the presence of 10% FCS with or without 10-5 M retinoic acid for 3 days. After a medium change and a further sub-cultivation for 4 days in the same medium cells were trypsinized, washed thrice with PBS and assayed for plasma membrane NADH-DC1P reductase activity. Control, undifferentiated cells were grown under the same conditions without retinoic acid. Activitymeasurements were performedin triplicates

clearly observed by means of FACS analysis (Fig. 8D). The raise in the S-phase then progressively decreases as the cell cycle undergoes completion (Fig. 8E and F). 3.4. Differentiation o f NTera-2 cells

According to the above observations a change in the plasma membrane NADH-DCIP reductase correlates with specific cell cycle phases. Therefore it is expected that during differentiation, i.e. when cells are in the G1 and M-phases, a strong increase in activity should be observed over differentiated cells, where cells are mostly in the G0-phase. To check this hypothesis we have used NTera-2 cells, a human neuroblastoma cell line that can be differentiated in the presence of 10 IxM retinoic acid [34]. Undifferentiated cells display a basal activity for the plasma membrane NADH-DCIP reductase of ca. 4 - 5 m U / m g protein (Table 1). However when the cells are exposed to low amounts of retinoic acid for several days an increase in enzyme activity is observed that correlates well with the observed cell differentiation. After 7 days of exposure to 10 ixM retinoic acid the enzyme activity has increased more than 2-fold over the controls, as shown in Table 1. These observations corroborate the other data and further establishes a good correlation between enzyme activation and cell differentiation. It should also be noted that under normal cell culture conditions a significant amount (15-20%) of NB41A3 cells undergo spontaneous differentiation into morphologically mature neurons, even in the absence of growth factors or retinoic acid in the medium. This explains in part why FACS analysis usually displays significant amounts of cells at the G0-phase, in the various conditions described above.

4. Discussion

The NADH-DCIP reductase that we have characterized in this and in a previous paper [5] is apparently very much

related to the plasma membrane oxidoreductase described in astrocytes by Mersel et al. [35,36] and Malviya et al. [37] and may also be similar to the NADH-ferricyanide reductase described by Cherry [6] or by Grebing [7]. A similar enzyme activity has been isolated from erythrocyte by Wang et Alaupovic [36]. In earlier studies [5]. we have shown that extracellular, impermeant electron acceptors stimulate growth of NB41A2 neuroblastoma cells, with ferricyanide or DCIP stimulating growth by 75% at micromolar concentrations The data presented here further characterize this enzyme and give evidence for that phenomenon. Elutriation studies show that the plasma membrane NADH-DCIP reductase is mostly induced and activated during the GI and during the G2/M-phases. A 300% increase in activity is observed in the presence of a-amanitin and a 50% increase with taxol, specific cell cycle inhibitors of the G1 and M phases respectively. These data could be corroborated by FACS analysis. Furthermore NTera-2 cells, a human neuroblastoma cell line that differentiates into fully mature neurones in the presence of retinoic acid, exhibit a 50% decrease in plasma membrane NADH-DCIP reductase activity during the G0phase upon differentiation, compared to undifferentiated cells. These data together show that the enzyme activity varies at various stages of the cell cycle and affects cell growth and differentiation. These studies are in agreement with observations made on melanoma cells, HeLa cells, Ehrlich Ascites or on other studies [38-47]. Growth factors rapidly modulate the electron flow across the plasma membrane of living cells. A cyanide-insensitive NADH oxidase purified from rat liver plasma membranes is stimulated by growth factors and inhibited by agents that inhibit growth or induce differentiation (retinoic acid, calcitriol and the monosialo-ganglioside GM 3) [41]. The enzyme is not affected by common inhibitors of microsomal or mitochondrial oxidoreductases. The stimulation by growth factors is decreased or absent with hepatoma plasma membranes [42,43]. Growth factors, EGF and transferrin also stimulate the reduction of external ascorbate free radical, a substrate of transplasma membrane oxidoreductase. By contrast wheat germ agglutinin inhibits this reduction, demonstrating, under physiological conditions, the operation of a growth factor- and lectin-responsive electron transport system at the cell surface in a cultured human cell line [47]. Ferric-lactoferrin, another growth stimulant in isolated rat liver plasma membranes, stimulates both the reduction of external ferric iron by cells and NADH oxidase activity or the amiloride-sensitive proton release from K562 cells [48]. In addition antitumoral agents, such as adfiamycin, cis-dichlorodiaminePt(II), actinomycin D and bleomycin also inhibit both transplasma membrane oxidoreductases and cell growth [49-51]. Conjugates of adriamycin cross-linked to transferfin inhibit proliferation of K562 cells as well as oxidoreductase activity in the plasma membrane at concentrations ten times lower than concentrations of free adri-

R. Zurbriggen, J.-L. Dreyer / Biochimica et Biophysica Acta 1312 (1996) 215-222

amycin [44]. They are cytotoxic to human promyelocytic (HL-60) and erythroleukemic (K562) cells, growth inhibition being higher for conjugates than for free adriamycin [45]. Pt(II) complexes are antitumoral in their cis but not trans form (although both forms bind DNA with similar binding affinities) and 10 --7 M cis-Pt(II) completely inhibits transplasma membrane oxidoreductases and growth within 3 min, whereas the trans-Pt(II) complex is inactive [50]. Retinoic acid and calcitriol also are inhibitors of both growth and transplasma membrane NADH oxidase in normal and immortalised cells [44]. In normal HKc the plasma membrane NADH oxidase is stimulated by EGF whereas that of transformed cells, HKc/HPV16, is not. In all cases the enzyme is inhibited by calcitriol (lct-l,25-dihydroxy vitamin D-3) and retinoic acid [46]. Another clear indication of the role of transplasma membrane oxidoreductases in the control of growth and proliferation comes from their effects on oncogene activation. The Ha-ras oncoprotein increases the activity of the transplasma membrane electron transport system, since both externally added cytochrome c and ferricyanide are reduced at a faster rate when cells are expressing the Ha-ras oncogene [42]. Ha-ras transformed cells also extrude protons faster than their normal counterparts in response to stimulation by external oxidants, suggesting that transformation events, initiated by the expression of the Ha-ras oncoprotein p21, would control the transplasma membrane electron transport system [42]. On the other hand, extracellular ferricyanide reduction is also inhibited by amiloride, a specific Na+/H+-antiport inhibitor [52,53]. This and other evidences that the plasma membrane oxidoreductases lead to activation of N a + / H + exchange provide an alternative explanation for their proliferative effect. It has been suggested that the transplasma membrane NADH oxidoreductase could modulate the intracellular pH, by affecting the antiport leading to cell proliferation via a pathway probably involving MAPkinase. Gaillard [54] have shown in PC12 cells that external ferricyanide induces a persistent intracellular acidification of ca. 0.25 pH-units, whereas no intracellular drop is observed after stimulation and proliferation by growth factors in the absence of an external electron acceptor. From these and our works, a function of plasma membrane NADH-oxidoreductases in cell differentiation is well established. Our report clarifies further that the enzyme is turned on at specific stages of the cell cycle. The molecular mechanisms by which this regulation occurs, the transduction pathways and the implicataion of these effects on cellular fonctions remains to be further investigated.

Acknowledgements The authors are indebted to Mrs. C. Deforel-Poncet for skilled technical assistance and to Dr. F. Perriard, from the cantonal Hospital, for making available its FACS facilities.

221

The present work was supported by a grant No. 3136488.92 from the Swiss National Foundation.

References [1] Crane, F.L. and Barr, R. (1989) Crit. Rev. PI. Sciences 8, 273-297. [2] Crane, F.L., Sun, I.L., Sun E.E. and Crow, R.A. (1995) Protoplasma 184, 3-7. [3] Crane, F.L., L~Jw, H., Sun, I.L. and Isaksson, M. (1990) in Oxidoreduction at the Plasma Membrane Vol. 1 (Crane, Morr~ and LiSw, eds.), pp. 141-170, CRC Press, Boston. [4] Rodriguez-Aguilera, J.C., Navarro, F., Arroyo, A., Alcain, F.J., Villaba, J.M. and Navas, P. (1995) Protoplasma 184, 229-232. [5] Zurbriggen, R. and Dreyer, J.L. (1993) Biochim. Biophys. Acta 1183, 513-520. [6] Grebing, C., Crane, F.L., LSw, H. and Hall, K. (1984) J. Bioenerg. Biomemb. 16, 517-522. [7] Cherry, J.M., MacKeller, W., Mort6, D.J., Crane, F.L. Jacobson, L.B. and Schirrmacher, W. (1981) Biochim. Biophys. Acta 634, 11-25. [8] Navas, P. and Gomez-Diaz, C. (1995) Protoplasma 194, 8-13. [9] Rodriguez-Aguilera, J.C. and Navas, P. (1994) J. Bioenerg. Biomembr. 26, 379-384. [10] Alcain, F.J., Villalba, J.M. and Navas, P. (1991) Biochim. Biophys. Acta 1073, 380-385. [11] Morr6, D.J., Crane, F.L. and Navas, P. (1987) Ann. NY Acad, Sci. USA 498, 153-171. [12] Crane, F.L., Mort6, D.J., LiSw, H.E. and B&tger, M. (1991) in Oxidoreduction at the Plasma Membrane, Vol. 2 (Crane, Morr6 and L~iw, eds.), pp. 21-33, CRC Press, Boston. [13] Moiler, I.M., Askerlund, P. and Widell, S. (1991) in Oxidoreduction at the Plasma Membrane, Vol. 2 (Crane, Morr6 and L~Sw, eds.), pp. 36-59, CRC Press, Boston. [14] Asard, H.H. and B6rczi, A. (1995) Protoplasma 184, 140-144. [15] Abo, A., Boylan, A., West, I., Thrasher, AJ. and Segal, A.W. (1992) J. Biol. Mol. 267, 16767-16770. [16] Crane, F.L., Sun, I.L., Barr, R. and Low, H. (1991) J. Bioenerg. Biomembr. 23, 773-803,. [17] LiSw, H., Crane, F.L., Grebing, C., Isaksson, M., Lindgren, A. and Sun, I.L. (1991)J. Bioenerg. Biomembr. 23, 903-917. [18] Crane, F.L., L~iw, H.E. and Morr6, D.J. (1990) in Oxidoreduction at the Plasma Membrane: Relation to Growth and Transport, Vol. 1 (Crane, Morr~ and L~Sw, eds.), CRC Press, Boston, 1-28. [19] Crowe, R.A. and Crane, F.L. (1995) Protoplasma 184, 209-213. [20] Sun, I.L., Crane, F.L. and Low, H. (1994) Biochim. Biophys. Acta 1221, 206-210. [21] Morr6, D.J. and Morr~, D.M. (1995) J. Bioenerg. Biomembr. 26, 7-24. [22] Minnifield, N. and Morr6, D.J. (1984) Cell Biol. Int. Rpts. 8, 215-217. [23] Dreyer, J.L. (1990) J. Bioenerg. Biomemb. 22, 619-633. [24] Yue, Y. and Dreyer, J.-L. (1995) Dev. Brain. Res. 89, 235-252. [25] Yue, Y. and Dreyer, J.-L. (1995) Dev. Brain Res. 89, 253-263. [26] Agustini-Tacco, G. and Sato, G. (1980) Proc. Natl. Acad. Sci. USA 64, 311-315. [27] Crane, F.L. and L~Sw, H. (1976) FEBS Lett. 68, 153-156. [28] Minana, D., Felipo, V. and Grisolia, S. (1992) Brain Res. 596, 157-162. [29] Bradford, M.M.A. (1976) Anal. Biochem. 72, 248-254. [30] Adolf, S., Brusselbach, S. and Muller, R. (1993) J. Cell. Sci. 105, 113-122. [31] Sawai, M.D., Mattson, G., Desai, S., Nielander, G.W., Morgensen, S. and Conklin, E.J. (1982) Avidin-Biotin Chemistry: A Handbook. (Pierce, ed.).

222

R. Zurbriggen, J.-L. Dreyer / Biochimica et Biophysica Acta 1312 (1996) 215-222

[32] Brennan, J.K., Lee, K.S., Frazel, M.A., Keng, P.C. and Young, D.A. (1991) J. Cell. Physiol. 146, 425-435. [33] Prochownik, E.V., Kukowska, J. and Rodgers, C. (1988) Mol. Cell. Biol. 8, 3683-3695. [34] Jordan, M.A., Toso, R.J., Thrower, D. and Wilson, L. (1993) Proc. Natl. Acad. Sci. USA, 90, 9552-9556. [35] Mersel, M., Malviya, A.N., Hindelang, C. and Mandel, P. (1984) Biochim. Biophys. Acta 778, 144-158. [36] Mersel, M., Vitkovic, L., Vincendon, G. and Malviya, A.N. (1988) in Plasma Membrane Oxidoreductases in Control of Animal and Plant Growth (Crane, F.L. et al., eds.) Plenum Press, New York, 402. [37] Malviya, A.N., Mandel, P. and Mersel, M. Biochim. Biophys. Acta 849, 288-296. [38] Wang, C.S. and Alaupovic (1978) J. Supramol. Struct. 9, 1-12. [39] Alcain, F.L., Buron, M.L., Rodriguez-Aguilera, J.C., Vollalba, J.M. and Navas, P. (1990) Cancer Res. 50, 5887-5891. [40] Sun, I.L., Crane, F.L., Grebig, C. and L/Sw, H. (1985) Exp. Cell Res. 156, 528-535. [41] Brightman, A.O., Wang, J., Miu, R.K., Sun, I.L, Barr, R, Crane, F.L and Morre, D.J. (1992) Biochim. Biophys. Acta 1105, 109-I 17. [42] Crowe, R.A., Taparowsky, E.J. and Crane, F..L. (1993) Biochem. Biophys. Res. Commun. 196, 844-50,. [43] Bruno, M., Brightman, A.O., Lawrence, J., Werderitsh, D., Morre, D.M. and Morre, D.J. (1992) Biochem. J. 284, 625-628.

[44] Faulk, W.P., Barabas, K., Sun, I.L., Crane, F.L. (1991) Biochem. Int. 25, 815-22. [45] Ruthner, M., Berczi, A. and Goldenberg-H. (1994) Life Sci. 54, 35-40. [46] Morre, D,J,, Morre, D.M., Paulik, M., Batova, A., Broome, A.M., Pirisi, L. and Creek, K.E. (1992) Biochim. Biophys. Acta 1134, 217-222. [47] Navas, P., Alcain, F.J., Buron, I., Rodriquez-Aguilera, J.C., Villalba, J.M., Morre, D.M. and Morre, D.J. (1992) FEBS Lett. 299, 223-226. [48] Sun, I.L., Crane, F.L., Morre, D,J., Low, H and Faulk, W.P. (1991) Biochem. Biophys. Res. Commun. 176, 498-504. [49] Tritton, T.R. and Yee, G. (1982) Science 217, 248-252. [50] Sun, I.L. and Crane, F.L. (1988) in Oxidoreduction at the Plasma Membrane, Vol. 1 (Crane, Morr6 and Ltiw, eds.), pp. 257-280, CRC Press, Boston. [51] Pascoe, J.M. and Roberts, J.J. (1974) Biochem. Pharmacol. 23, 1345-1357. [52] Garcia-Canero. R., Diaz-Gill, J.J. and Guerra, M.A. (1987) in Redox functions in the eukaryotic Plasma Membrane (Ramirez, J.M. ed.),pp. 42-47, Graficas, Madrid. [53] Sun, I.L., Garcia-Canero, R., Liu, W., Toole-Simms, W., Crane, F.L., Mort6, D.J. and LiSw, H. (1987) Biochem. Biophys. Res. Commun. 145, 467-473. [54] Gaillard, S., Thomas, D. and Malviya, A.N. (1995) 2nd Colloquium of the French Society for Neurosciences, Lyon, F-10, 232.