Vol. 172, No. 3, 1990

November 15, 1990

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Pages 979-984

Evidence for Coenzyme Q Function in Transplasma Membrane Electron Transport

I. L. Sunl, E. E. Sunl, F. L. Crane' and D. J. Morre2

Departments of lBiologica1 Sciences and 'Medicinal Chemistry and Pharmacognosy

Purdue University, West Lafayette, Indiana 47907

Received September 23, 1990

SUMMARY: Transplasma membrane electron transport activity has been associated with stimulation of cell growth. Coenzyme Q is present in plasma membranes and because of its lipid solubility would be a logical carrier to transport electrons across the plasma membrane. Extraction of coenzyme Q from isolated rat liver plasma membranes decreases the NADH ferricyanide reductase and added coenzyme QlC restores the activity. Piericidin and other analogs of coenzyme Q inhibit transplasma membrane electron transport as measured by ferricyanide reduction by intact cells and NADH ferricyanide reduction by isolated plasma membranes. The inhibition by the analogs is reversed by added coenzyme QlO. Thus, coenzyme Q in plasma membranes may act as a transmembrane electron carrier for the redox system which has been shown to control cell growth. 0 1990 Ac*idemlc Press, 1°C.

A role for coenzyme Q in mitochondrial electron transport is well

established. It acts to shuttle electrons from the primary dehydrogenase to

the cytochrome bcl complex (1,2). It also can act as a proton carrier as part

of the mechanism for creating a chemiosmotic proton gradient required for ATP

synthesis (3).

Coenzyme Q is also found in highly purified liver plasma membranes at one

half the concentration found in mitochondria (4). It has been recognized for

some time to be present in endomembranes and is especially concentrated in

membranes of the Golgi apparatus (5-8). The question to consider is whether

the coenzyme Q in plasma membranes has a function in an electron transport

system present in these membranes. As an alternative it could serve as an

antioxidant or be part of a pool of quinone in transit from sites of synthesis

in the endoplasmic reticulum or Golgi membranes to the cell surface (4). Our

results now show that coenzyme Q can be a part of the transplasma membrane

Vol. 172, No. 3, 1990 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

electron transport system which is implicated in control of cellular

proliferation (9).

MATERIALS AND METHODS: Plasma membranes were purified from rat liver by the phase separation procedure using dextran and polyethylene glycol as described (10). Purity of the membrane fraction was determined by morphometry and assay for marker enzymes. The succinic dehydrogenase and cytochrome c oxidase activity showed less than 4% contamination by mitochondrial membranes.

HeLa cells were grown under a 5% C02, 95% air at 37°C in o! modified minimal essential media (aMEM) containing 10% fetal calf serum, 100 U of penicillin and 100 pg/ml streptomycin at pH 7.4. Confluent monolayer cultures were prepared for assay by pelleting the trypsin treated, suspended cells at 15,000 x g. The pellet was taken up in TD tris buffer (salts in g/l NaCl, 8, KC1 0.38, Na2HP04 0.1 and Trizma base 3 adjusted to pH 7.4). Aliquots of 0.01 g fresh weight of cells per ml of suspension media were used for assay of redox activity (11).

Heptane extraction of lyophilized rat liver plasma membrane was in the dark at room temperature with gentle shaking for 3 hours. The heptane was decanted and the residual heptane removed by evaporation. 10 volumes of heptane was used for each volume of dry membrane. Extracted membranes were resuspended in sucrose-Tris medium (0.25 M sucrose, 0.01 M Tris-HCl, pH 7.5) (12). Oxidation-reduction assay was carried out on a DW2a spectrophotometer at 420-500 nm with plasma membrane. Assays with whole cells were done in incubator tubes at 37" with aliquots removed at 2, 5 and 10 min intervals to centrifuge out the cells and measure ferricyanide in the supernatant at 420 nm (17).

Extraction of Coenzyme Q from Plasma Membranes with Loss of Activity RESULTS:

and Restoration with Added Coenzyme Q. Extraction of lyophilized rat liver

plasma membranes with heptane removed 80% of the coenzyme Q. The quinone was

Table I

Effect of Heptane Extraction on NADH Ferricyanide Reductase Ability of Rat Liver Plasma Membranes and Restoration of Activity

with Added Coenzyme QlO

Membrane Treatment NADH Ferricyanide Reductase

Fast Slow nmole min -1 (mg protein)-1

Untreated 217 k 49 (3) 106 t 13 (3) Extracted with Heptane 79 43 Extracted with Coenzyme QlO Added 162 76

Lyophilized membrane 40 mg extracted with 15 ml heptane for 3 hr in dark at room temperature. Heptane was removed by decantation and evaporation under "acullm Restoration with coenzyme 0.5 pmoles in 50 ml heptane evaporated after addition to 20 mg membranes. Membranes were resuspended in tris sucrose medium after removal of all heptane by evaporation. Procedure is based on Norling et al. as applied to mitochondria (12). Ferricyanide reduction assay by measuring decrease in absorbance at 420 run after incubation of NADH 25 PM, 0.1 mM ferricyanide and 0.05 mg membranes in a total of 3.0 ml 50 mM Tris Cl buffer pH 7.4 according to (11). The fast rate of ferricyanide reduction is in the first two minutes whereas the slow rate is the steady rate established after 2 min.

980

Vol. 172, No. 3, 1990 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

returned to the extracted membrane in a nonaqueous solution. The solvent was

then evaporated to leave the coenzyme Q closely associated with the membrane.

Coenzyme Q at a concentration equivalent to 10 PM in the final reaction

mixture restored more than 50% of the activity found in the unextracted

membrane (Table I).

Inhibition of Plasma Membrane Electron Transport by Coenzyme Q Analogs.

Analogs of coenzyme Q have been shown to be effective inhibitors of NADH

oxidation in mitochondria (13-15). The analog 5 N-pentadecyl-6-hydroxy-4,7-

dioxybenzthiazole (PHDBT) (16) is effective as an inhibitor of NADH

ferricyanide reductase activity in rat liver plasma membrane and the electron

transport across the plasma membrane which results in ferricyanide reduction

by HeLa cells (Table II)

Reversal of Inhibition by Coenzyme Q~,J. Two other analogs of coenzyme Q

piericidin A and 2,methoxy,3 ethoxy, 5 methyl, 6 n hexadecylmercapto-1,4

benzoquinone (monoethoxy coenzyme Q) also inhibit the ferricyanide reduction

by HeLa cells and the inhibition is reversed by added coenzyme Qlo (Table

III).

High concentrations of coenzyme QlO can be added to the extracted

membranes and this added coenzyme Qlo is reduced as measured at 410 nm when

Table II

Inhibition of NADH Ferricyanide Reductase Activity of Rat Liver Plasma Membranes and Ferricyanide Reduction by HeLa Cells

by 5n-Pentadecyl-6-hydroxy-4,7-dioxybenzthiazole, an Analog of Coenzyme Q

PHDBT added NADH Ferricyanide Reductase in Plasma Membrane

nmole min -' (mg protein)‘l

Ferricyanide Reduction by HeLa-yells

nmole min gww -1

None 199 * 17 (3) 805 + 104 (3) 35 PM 271 588 70 /JM 135 406

105 pM 56 231

Assays: NADH ferricyanide reductase was assayed in 2.8 ml 50 mM Tris Cl buffer pH 7.4 with 0.05 mg rat liver plasma membrane incubated with inhibitor for 3 min before adding NADH. Blank rate with NADH and ferricyanide and no membrane subtracted. Absorbance measured at 420-500 run. Reduction of ferricyanide by HeLa cells was measured in TD tris buffer pH 7.4 as described (11). Cells (0.01 g wet weight per 1 ml buffer) incubated 3 min with inhibitor before starting the assay with 0.1 mM ferricyanide.

981

Vol. 172, No. 3, 1990 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Table III

Effect of Coenzyme Q Analogs on Ferricyanide Reduction by HeLa Cells

and Reversal of the Inhibition by Coenzyme QIO

Addition to Cells Ferricyanide Reptio? Rate nmole min gww

None 750 Coenzyme Qlo 10 PM 600

Piericidin A 10m7M 400 Piericidin A 10m7M and Coenzyme QIO 10 PM 780

(1) Mono ethoxy coenzyme Q 90 PM 520 Mono ethoxy coenzyme Q + Coenzyme QIO 800

(2) 5-Chloro-6-naphthyl-mercapto coenzyme Q 134 PM 250 5-Chloro-6-naphthyl-mercapto coenzyme Q + Coenzyme Qlo 660

Assay with 0.01 gww cells per 1.0 ml TD tris buffer and 0.1 mM ferricyanide. Reaction stopped by cooling sample on ice followed by centrifugation to remove

cells. Decrease in ferricyanide measured in the supernatant at 420 nm as described in Clark et al. (17). (1) 2-methoxy-3-ethoxy-5-methyl-6,heptadecylmercapto-l,4-benzoquinone. (2) 2,3-dimethoxy-5-chloro-6-naphthylmercapto-l,4-benzoquinone,

NADH is added (not shown). Among other lipophilic quinones tested, a

tocopherol quinone gives restoration of activity at concentrations similar to

coenzyme QlG whereas vitamin Kl has no effect on the NADH ferricyanide

reductase activity (not shown).

DISCUSSION: Several laboratories have reported the presence of high levels of

coenzyme Q in the microsomal fraction from cells and especially in Golgi

apparatus membranes (4,8). Procedures have now been developed for preparation

of highly purified plasma membranes by the phase separation technique.

Analysis of these membranes demonstrates that rat liver plasma membranes have

a high concentration of coenzyme Q relative to endoplasmic reticulum as well

(4).

Examination of the redox state of coenzyme Q in microsomes (which would

include plasma and Golgi membranes) has shown that the coenzyme Q in the

membranes is reduced by NADH and reoxidized by ferricyanide. Since rotenone

did not inhibit the reduction of coenzyme Q in microsomes by NADH, there is

evidence for an electron transport system in these membranes which can reduce

coenzyme QlG that is not based on mitochondrial contamination (18).

982

Vol. 172, No. 3, 1990 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

The purification of liver plasma membranes by the phase separation

procedure has been thoroughly documented (10) and tests for mitochondrial

contamination show less than 4%. The response of the NADH ferricyanide

reductase activity in these membranes is different from the response of

mitochondrial membranes to solvent extraction. In mitochondria the NADH

oxidase is inhibited by heptane extraction but ferricyanide reductase is not

affected since ferricyanide reduction occurs at a site prior to the coenzyme Q

reduction site (12,19). Piericidin and other coenzyme Q analogs also inhibit

at a site after the primary NADH dehydrogenase, so that they do not affect

NADH ferricyanide reductase in mitochondria (13,20). The effect of extraction

and analog inhibition on NADH ferricyanide reductase of the plasma membrane is

clearly different from the lack of effects seen with the ferricyanide

reductase in mitochondria.

The inhibition of transplasma membrane ferricyanide reduction with HeLa

cells by coenzyme Q analogs and the reversal of the inhibition by coenzyme Qlo

indicates that coenzyme functions as a transplasma membrane electron carrier

for transfer of electrons from cytosolic NADH to external ferricyanide. Since

the transplasma membrane electron transport has been shown to stimulate cell

growth in serum free media (9,21-23), to promote the expression of c myc and

c fos protooncogenes (24) and to activate a tyrosine kinase (25), it is

possible that coenzyme Q is involved in growth control functions other than

those based on mitochondrial ATP production.

ACKNOWLEDGMENTS: Research supported by NIH career award K6 21837 (FLC) and CA 138801 (DJM). Provision of coenzyme Q analogs by Prof. K. Folkers, University of Texas, has made this study possible.

REFERENCES

1. G. Lenaz, ed., Coenzyme Q, Biochemistry, Bioenergetics and Clinical Applications, 517 pp. (1985), J. Wiley and Sons, Chichester.

2. B.L. Trunpower, ed., Function of Quinones in Energy Conserving Systems, 582 pp. (1982), Academic Press, New York.

3. P. Mitchell, Biol. Rev. 41, 445-502 (19661. 4. A. Kalin, Bl Norling, E.L.'Appelkvist, G.'Dullner, Biochim. Biophys. Acta

926, 70-78 (1987). 5. S. Leonhauser, K: Lebold, K. Krisch, H.S. Standinger, F.C. Gale, A.C.

Page, Jr., K. Folkers, Arch. Biochem. Biophys. 96, 580-82 (1962). 6. J. Jayaraman, T. Ramasarma, Arch. Biochem. Biophys 103, 258-266 (1963).

983

Vol. 172, No. 3, 1990 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

a

9

10 11

12

13

14

15

16 17

18

19 20 21 22

23

24

25

S.E. Nyquist, R. Barr, D.J. Morre, Biochim. Biophys. Acta 208, 532-34 (1970). F.L. Crane, D.J. Morre, in Biomedical and Clinical Aspects of Coenzyme Q (K. Folkers and Y. Yamamura, eds.) pp. 3-14 (1977), Elsevier, Amsterdam. F.L. Crane, D.J. Morre, H. Low, eds., Plasma Membrane Oxidoreductases in Control of Animal and Plant Growth, 443 pp. (1988) Plenum Press, New York. P. Navas, D.D. Nowack, D.J. Morre, Cancer Res. 49, 2147-56 (1989). I.L. Sun, F.L. Crane, C. Grebing, H. Low, J. Bioenerg. Biomemb. 16, 583- 95 (1984). B. Norling, E. Glazek, B.D. Nelson, L. Ernster, Eur. J. Biochem. 47, 475-82 (1974). A. Palmer, D.J. Horgan, H. Tisdale, T.P. Singer, H. Beinert, J. Biol. Chem. 243, 844-47 (1968). J. Catlin, R.S. Pardini, G. Doyle Daves, Jr., J.C. Heidker, K. Folkers, J. Amer. Chem. Sot. 90, 3572-74 (1968). A. Castelli, E. Bertli, G.P. Littarru, G. Lenaz, K. Folkers, Biochem. Biophys. Res. Communs. 42, 806-10 (1971). B.L. Trumpower, J.G. Haggerty, J. Bioenerg. Biomemb. 12, 151-64 (1980). M.G. Clark, E.J. Partick, G.S. Patten, F.L. Crane, H. Low, C. Grebing, Biochem. J. 200, 565-72 (1981). M. Takada, S. Ikenoya, T. Yuzwika, K. Katayama, Biochim. Biophys. Acta 679, 308-14 (1982). L. Sarkozka, Arch. Biochem. Biophys. 113, 519-528 (1966). D.J. Horgan, T.P. Singer, J.F. Casida, J. Biol. Chem. 243, 834-43 (1968). G.F. Kay, K.A.O. Ellem, J. Cell. Physiol. 126, 275-80 (1986). R. Garcia-Caiiero, J.J. Diaz-Gil, M.A. Guerra, in Redox Functions of the Eukaryotic Plasma Membrane (J. Ramirez, ed.) pp. 43-47 (1987), CSIC, Madrid. I.L. Sun, F.L. Crane, C. Grebing, H. Low, Exp. Cell. Res. 156, 528-36 (1985). C.E. Wenner, A. Cutry, A. Kinniburgh, L.D. Tomei, K.J. Leister, in Plasma Membrane Oxidoreductases in Control of Animal and Plant Growth (F.L. Crane, D.J. Morre, H. Low, eds.) pp. 7-16 (1988) Plenum Press, New York. P.S. Low, R.L. Geahlin, E. Mehler, M.L. Harrison, Biomed. Biochem. Acta 49, 135-140 (1990).

984