Inorganic phosphate is the major component of the thermostable cytoplasmic fraction which stimulates mitochondrial anion uniport

180

Biochimica et BiophysicaActa, 1183(1993) 180-184 Elsevier Science Publishers B.V.

BBABIO 43920

Inorganic phosphate is the major component of the thermostable cytoplasmic fraction which stimulates mitochondrial anion uniport Lay Tin Ng, Michael J. Selwyn * and Hui Lim Choo Department of Biochemistry, National Universityof Singapore, 10 Kent Ridge Cresent, Singapore 0511 (Singapore) (Received 8 March 1993) (Revised manuscript received 2 June 1993)

Key words: Mitochondrion;Inner membrane; Anion channel; Inorganic phosphate; Thyroid; Phosphate; (Liver) A low molecular weight thermostable cytoplasmic fraction isolated from rat liver homogenate when pre-incubated with mitochondria increases the rate at which anions enter mitochondria via the pH-dependent anion-conducting channel in the inner membrane. The crude fraction obtained by centrifuging and heating the liver homogenate was purified by gel filtration and chromatography on DEAE-cellulose. The resulting factor is stable to heating at 100°C, freeze-drying and extremes of pH. Inorganic phosphate co-purified with activity and activity was lost when the phosphate was removed by barium salt precipitation. A pure sample of KH2PO 4 produced stimulation of anion conductivity. These results show that the major portion of the activity which stimulates anion uniport can be accounted for by the presence of phosphate in the crude and purified fractions. Mersalyl blocks stimulation when added before, but not when added after, incubation with phosphate which shows that the stimulation is produced by phosphate in the mitochondrial matrix. The proposed role of this factor in thyroid hormone action is discussed in the light of its identification as inorganic phosphate.

Introduction

Materials and Methods

Gainutdinov et al. reported a thermostable factor in the soluble fraction of rat liver homogenate which, when pre-incubated with mitochondria, increased the mitochondrial permeability to phosphate and K ÷ ions [1-4]. The factor was partially purified by gel-filtration and anion-exchange chromatography, tentatively identified as a glycopeptide or an oligosaccharide with an aromatic moiety attached [2,4] and thought to be similar to the carbohydrate-phosphate factor released by insulin [5]. This report describes the extraction and partial purification, by methods similar to those of Gainutdinov et al., of a liver cytoplasmic fraction which, when pre-incubated with mitochondria, stimulates mitochondrial anion-uniport via the inner membrane pH-dependent anion-conducting channel [6-8] and the chemical characterisation of the major active component as phosphate(V) (inorganic ortho-phosphate).

Materials. Rotenone, Antimycin A, Hepes, Tris and mersalyl (free acid) were obtained from Sigma, FCCP from Aldrich, NH4C1 KC1 and K H 2 P O 4 from Merck, sucrose from BDH, Sephadex G-25 and G-10 from Pharmacia and DEAE-cellulose from Whatman. Mersalyl was dissolved in and neutralised with N a O H solution. All chemicals were obtained as analytical reagent grade when available. Preparation o f liver extract. Crude liver extracts were obtained by centrifuging homogenates of frozen or fresh rat livers, 1 g wet l i v e r / m l of medium containing 5 mM Hepes (pH 7.4), at 30000 × g for 30 min. The supernatant fluid was heated to near 100*C for 10 min, cooled and centrifuged at 1000 x g for 20 min to remove denatured proteins. This thermostable fraction was concentrated to 10 m l / 2 0 0 g wet weight liver by evaporation on a boiling water-bath under a stream of nitrogen. Preparation o f liver mitochondria. Liver mitochondria were prepared from adult Wistar rats, starved overnight, as described previously [9]. The mitochondria were suspended at 50 m g / m l in 0.25 M sucrose containing 5 mM H e p e s - K O H (pH 7.4). The

* Corresponding author. Fax: +65 7791453. Abbreviations: Hepes, N-2-hydroxyethylpiperazine-N'-ethanesulphonic acid; FCCP, carbonyleyanide4-(trifluoromethoxy)phenylhydrazone; mersalyl, 2-[N-(3-hydroxymercuri-2-methoxypropyl)-carbamoyl]phenoxyaceticacid.

181 mitochondria used in these experiments had a respiratory control ratio of not less than 4 and an A D P / O ratio of 1.5-2.0 with succinate as substrate. T h e mitochondrial suspension was stored on ice with continuous stirring to maintain high and constant anion conductivity [10,11] Assays. Anion uniport was assayed by passive osmotic swelling of mitochondria in 4.0 ml of a medium containing 100 m M NH4CI (adjusted to p H 8.0 with N H 4 O H ) , 0 . 6 / ~ M rotenone and 0 . 5 / z M antimycin A; 7 . 5 / z M FCCP was added prior to the mitochondria to allow rapid entry of H + ions to balance charge and p H changes produced by anion-uniport and N H 3 entry. The assay was started by adding the mitochondria and the rate of mitochondrial swelling was measured by recording the change in % light transmission of the suspension in a cylindrical glass cuvette thermostatted at 30°C [9]. Controls without FCCP showed that this technique measures electrophoretic uniport of C l ions. T o measure the stimulatory activity, 100/zl of sample was pre-incubated for 2 min at 25°C with 36 /zl mitochondrial suspension at 50 mg p r o t e i n / m l , and 30 /zl of 0.3 M KCI and 0.25 M Tris-HCl (pH 7.4) to maintain the ionic strength and pH. The pre-incubated mitochondria, 1.5 mg protein, were then added to the NH4CI assay medium as described above. The fractions from the purification steps were tested against controls consisting of their corresponding buffers or eluting media. Rates are expressed as % of the rate in the appropriate control assay. Phosphate was measured by the method of Fiske and SubbaRow [12], carbohydrate by the phenolsulphuric acid reaction [13], protein by the biuret assay [14] and c r e a t i n e / c r e a t i n i n e by reaction with picrate [15]. Osmolality was determined using a Wescor 5500 V a p o r Pressure Osmometer. Results

300 "

I N

200.

1.2

¢m ¢=

1.0 0.8

[]

100 0.6 0.4

_S d."

0.2

0

10

20

30

Fraction

5'0

4O

0

number

Fig. 1. Gel filtration of the thermostable cytoplasmicfactor prepared from rat liver cytoplasm. 10 ml of concentrated liver extract was applied to a column of Sephadex G-25 fine grade, 25 mm diameter × 500 mm long, equilibrated with 5 mM Hepes-KOH (pH 7.4) and eluted with the same buffer. 5-ml fractions were collected and assayed for phosphate content (bar diagram) and stimulation of mitochondrial anion uniport in 100 mM NH4CI plus FCCP as described in Materials and Methods (e). Rates of swelling are expressed relative to the rate after pre-incubation with the eluting buffer for 2 min at 25°C.

Sephadex G-10 column (Fig. 3). The highly purified fraction is very soluble in water and stable to freezedrying. The elution patterns and levels of stimulation of mitochondrial swelling in the anion uniport assay were very reproducible from one preparation to another. Recovery of activity is apparently more than 100%, but valid estimates cannot be made because the isolation procedure removes inhibitory factors as shown by the lower than control rates of swelling in fractions eluting prior to activity in Figs. 1 and 2.

300

i

250

I

200

The concentrated heat-treated liver extract, prepared as described under methods, was chromatographed on a column of Sephadex G-25 fine grade. Activity eluted at about 1 column volume in a light yellow fraction and increased the rate of mitochondrial swelling by almost 100% (Fig. 1). The active fractions were pooled and then c h r o m a t o g r a p h e d on a D E A E cellulose column equilibrated in the same buffer. Under these conditions, the active fractions together with coloured components were retained in the column but some of the materials which absorb at 260 nm were eluted by the same buffer. The active components were then eluted by a high ionic strength buffer of 95 m M KCI with 5 m M H e p e s at p H 7.4 (Fig. 2), leaving the coloured components bound to the column. T h e high activity fractions were pooled and desalted on. a

8.0

co' 150

6.0 ~a ,~

el er

100

5O

" -

-

i 5

-

-

J

-i . . . . . 10

Fraction

15

[]

- 4.0

s

~"

- 2.0

a.7

-0

2O

number

Fig. 2. Anion-exchange chromatography of the active fraction from Sephadex G-25 gel filtration. A 5-ml sample of pooled active fractions was applied to a column of DEAE-cellulose, 15 nun diameter x 200 mm long, equilibrated with 5 mM Hepes-KOH (pH 7.4). The column was eluted with 5 mM Hepes-KOH buffer and 5-ml fractions were collected; after fraction 10 the elution medium was changed to 95 mM KCI in 5 mM Hepes-KOH (pH 7.4). The fractions were assayed for activity (e) against controls using the appropriate eluting buffer and phosphate (bar diagram).

182 KCI

300

1.0

200

Q > 0.8 (J

250 150

i

150

100

E

lOO so

so

~

N" 200 ==,

-~

/

~

~a

¢o e~

0.6

o

0.4

x~

"~

0.2 d 0

.

200 0

0

Ial/\ l\ \ \

t,-

~ 10

15

i

J

20

25

.

.

.

,

250

.

.

.

.

~

•

•

300

Wavelength (nm)

Fraction number Fig. 3. Gel filtration of pooled active fractions from DEAE-cellulose chromatography. A 3-ml sample was applied to a Sephadex G-10 column, l0 mm diameter x300 mm long, equilibrated with 5 mM Hepes-KOH buffer (pH 7.4). 1-ml fractions were eluted with the same buffer, assayed for activity (e) and their osmolality measured (bar diagram). Rates of swelling are expressed relative to rate with the eluting buffer. The arrow indicates the elution peak of KCI.

Fig. 4 shows the UV spectra of preparations at various stages of purification: at the final stage the preparation is colourless and UV absorption is apparent only below 220 nm, indicating the absence of nucleosides, aromatic rings, etc. In preliminary trials the activity was retained on DEAE-cellulose at low ionic strength over the pH range from 3 to 9, indicating that the factor is negatively charged over this range. Activity was retained when the preparations were adjusted to pH 2 or 11 and then neutralised before assay. The purified fractions appeared to contain little or no carbohydrate or protein, by the phenol-sulphuric acid and biuret reactions respectively but there was a con-

Fig. 4. UV spectra of the active fractions at different stages of preparation. (a) crude extract diluted 1 : 400; (b) after Sephadex G,25 chromatography, 1 mg dry weight/ml; (c) after elution from DEAEcellulose, I mg/ml; (d) after Sephadex G-10 chromatography, 1 mg/ml. Absorbance was measured in 1-cm pathlength silica cuvettes against a water blank using a Hewlett-Packard 8452A Diode Array spectrophotometer.

sistent association of inorganic ortho-phosphate with the activity. An average of 10% ( w / w ) phosphate was found in the final fraction and much of the remaining mass and UV-absorbing material in this fraction is K ÷ ions and Hepes buffer from the eluting medium. The presence of phosphate also accounts for the yellow precipitate observed when active fractions were tested for chloride by adding silver nitrate. Because creatine phosphate hydrolyses in the Fiske and SubbaRow assay we performed a barium salt fractionation to separate inorganic phosphate from creatine phosphate [16] and obtained a water-insoluble barium salt which contained phosphate and was separated by centrifugation from a phosphate-free and inactive supernatant. This distin-

30

¢3 C

E 20

_J

10

a

b

c

in

d e

Time (sec) Fig. 5. Recordings of light scattering showing mitochondrial swelling in 100 mM NH4Cl in the presence of rotenone, antimycin and FCCP. 1.8 mg of mitoehondriai protein was pre-ineubated as described in Materials and Methods with 100 ~! of: (a) 9 mM KH2PO 4 (pH 7.4); (b) 9 mM potassium acetate (pH 7.4); (c) water; (d) the cytoplasmic factor at the first stage of purification, i.e., the eluate from Sephadex G-25, at 20 mg dry weight/ml; (e) as (d), but the mit~hondria were treated with mersalyl (10 nmol/mg protein) for 15 s prior to incubation with the factor. Swelling was initiated by transferring the pre-incubated mitochondrial suspensions to the NH4CI medium as shown at the arrows.

183 guishes the component from creatine phosphate which has a water-soluble barium salt. Furthermore, this fraction contained less than 0.65 tzg creatine/creatinine per mg dry weight. We conclude that the phosphate is present as inorganic ortho-phosphate. Fig. 5 shows recordings of light scattering by mitochondria suspended in 100 mM NH4CI in the presence of FCCP. After pre-incubation with the partially purified fraction (Fig. 5, recording d) or 9 mM KH2PO 4 (Fig, 5, recording a) the rate of mitochondrial swelling was much faster than the control rate (Fig. 5, recording c). Pre-incubation with 9 mM potassium acetate (Fig. 5, recording b) also stimulated swelling but to a lesser degree. The stimulatory activity of the final stage fraction was completely lost when inorganic phosphate was removed by precipitation as the barium salt (recording not shown). No effect on the rate of swelling was observed when the cytoplasmic fraction, phosphate or acetate were added to the NH4CI medium prior to mitochondria but without the 2-min pre-incubation (recordings not shown). Our initial hypothesis for the mechanism of stimulation was that phosphate was accumulated during pre-incubation and then, when the mitochondria are transferred to the assay medium, released via the phosphate/hydroxide antiporter causing alkalinisation of the matrix and activation of the pH-dependent anion-conducting channel. A similar mechanism involving uptake of acetate and release as acetic acid would account for stimulation by acetate. However when this hypothesis was tested by blocking the phosphate/hydroxide antiporter with mersalyl it was found that mersalyl treatment prior to incubation with phosphate blocked stimulation by phosphate but treatment with mersalyl after incubation with phosphate had no effect (Table I). Assays by passive osmotic swelling in ammonium phosphate medium at pH 7.4 showed that under these conditions of pre-incubation the phosphate/hydroxide antiporter was 99.5% inhibited by mersalyl (recordings not shown). As shown in Table I, at this concentration and under these conditions mersalyl had little or no effect either in control experiments in the absence of phosphate or on the stimulation by acetate. The large standard error for the experiments on phosphate stimulation was produced by variation in stimulation between batches of mitochondria: note the much decreased variation when stimulation was abolished by pre-treatment with mersalyl and that if the values for post-treatment with mersalyl are compared pairwise with their controls (with phosphate but without mersalyl) the variation in the relative value expressed as a percentage is much decreased, 98_+ 4.7%. The effects of mersalyl were tested on the extracted cytoplasmic factor and, as shown in recording e in Fig. 5 and Table I, stimulation by the cytoplasmic factor is blocked by mersalyl added before the factor. The similarity of the effects of mersalyl on

TABLE I

Effect of mersalyl on the stimulation of chloride uniport by pre-incubation in potassium phosphate, potassium acetate or the cytoplasmic factor Mitochondria were pre-incubated for 2 min with 100 /zl of 9 mM potassium phosphate (pH 7.4), 9 mM potassium acetate (pH 7.4), cytoplasmic factor (eluate from Sephadex G-25 column at 20 mg dry weight/ml) or water (control row) as described in Materials and Methods and in the legend to Fig. 5. In addition the mitochondria were treated for 15 s with 4.5/zl of water (control column) or 4 mM mersalyl (10 nmol/mg protein) either before or after the 2-min pre-incubation with stimulant. Data are the rates of light scattering decrease expressed as a percentage of the control value in the absence of stimulant and mersalyl. Data are given as means_+ S.E. (number of observations), using three different preparations of mitochondria for phosphate and acetate and one preparation for the cytoplasmic factor. Stimulant

Control

Control Potassium phosphate Potassium acetate Cytoplasmic factor

100 3855:60

Merslayl added before stimulant 995:11

(12)

Mersalyl added after stimulant 9 7 5 : 5 . 9 (12)

(12)

1065:4.9 (6)

379+69

(6)

1555: 9.0(12)

1635:7.9 (6)

1545:17

(6)

4125:20

1245:6.6 (3)

418_+16

(3)

(5)

the factor to those on potassium phosphate are added confirmation that phosphate is the major stimulatory component in this factor. Discussion

Our data confirm the consistent and reproducible presence of a thermostable factor in the cytoplasmic fraction of rat liver homogenate which, when pre-incubated with mitochondria, increases the rate of anion uniport. Phosphate co-purifies with activity, activity is lost when phosphate is precipitated from a purified fraction and a sample of K H E P O 4 produced stimulation of anion-uniport. The properties of the factor; stability to boiling and freeze-drying, solubility in water, net negative charge over the pH range 3 to 9, an apparent M r less than 1000 and absence of UV absorption above 220 nm are consistent with our identification of the factor as inorganic phosphate. From this we conclude that inorganic ortho-phosphate is the major active component of the heat-stable cytoplasmic fraction. In terms of biochemical activity, source and method of isolation our preparation is similar to the low molecular weight cytoplasmic regulators studied by Gainutdinov and co-workers [1-4], who considered their preparation to be a glycopeptide or an oligosaccharide. Our data show that phosphate is the major component of the active fraction but do not preclude the existence of other factors with similar activity in the crude extract.

184 However, in gel filtration phosphate will co-elute with cationic molecules and this could account for the distribution of activity among several fractions reported by Gainutdinov et al. [4]. The effects of mersalyl show that uptake of phosphate during the pre-incubation is required for stimulation but release via the phosphate/hydroxide antiporter during the assay is not necessary. Although the stimulation by acetate indicates that alkalinisation of the matrix may play some role in the stimulation, the effects of mersalyl show that some other mechanism is responsible for the much greater stimulation produced by phosphate. The situation is complicated because organo-mercurial compounds not only block the phosp h a t e / hydroxide antiporter but also react with regulatory sites in the mitochondrial inner membrane anionconducting channel [17,18]. Thus, hypotheses about the mechanism of phosphate stimulation would be speculative at present but it is clear that stimulation of anion uniport is produced by phosphate in the mitochondrial matrix. The amount of phosphate recovered in the final stage fraction corresponds to about 10 mmol Pi/kg wet liver which is in accord with the published value of 10 mM Pi liver cytosolic phosphate [19]. Gainutdinov et al. [4,20] observed changes in thermostable cytoplasmic factor activity in response to thyroid hormone and suggested that it has a role in the regulation of mitochondrial metabolism. In the light of our identification of this factor as inorganic phosphate these observations by Gainutdinov's group are seen to be in accord with the effects of thyroid hormone on Pi levels in muscle [21-23] and on the mitochondrial phosphate transporter [24,25]. Hafner et al. [26] found that thyroid hormones affected mitochondrial respiration via the phosphorylation rather than the redox reactions but concluded that these effects were on the adenine nucleotide translocator or the FIF0-ATPase because the phosphate transporter was reported by Kunz et al. [27] to have a very low flux control coefficient in isolated mitoehondria. Nevertheless, this work by Kunz et al. showed that decreasing the phosphate concentration from 10 mM to 3 mM increased the flux control coefficient for the ATPase from 0.17 to 0.41 and that below 3 mM Pi the rate of respiration declined sharply with decreasing phosphate concentration. Although it is not generally considered that phosphate has a regulatory role in mitochondrial respiration in vivo, measurements on perfused liver by Tanaka, Chance and Quistorff [28] using 31p-NMR showed that the level of free phosphate is about 60% of the total Pi as measured by chemical analysis and that under some conditions Pi could regulate the rate of respiration. It appears, therefore, that while there are relationships between thyroid hormone level, intracellular Pi concentration and mitochondrial respiration, the causal con-

nections and the role, if any, of phosphate in mediating thyroid action are as yet unknown.

Acknowledgement This research was supported by a grant from the National University of Singapore (RP890313).

References 1 Gainutdinov, M. Kh., Luchenko, M.B., Mirmakhmudova, S.I. and Turakulov, Ya. Kh. (1984) Dokl. Akad. Nauk USSR 2, 47-49. 2 Gainutdinov, M.Kh., Konov, V.V., Safarov, K.S. and Kasyov, A.K. (1986) Biokhimiya 51, 287-294. 3 Gainutdinov, M.Kh., Ishmukhamedov, R.N., Konov, V., Luchenko, M.B. and Mirmakhmudova, S. (1988) Biokhimiya 53, 196-204. 4 Gainutdinov, M.Kh., Zakharova, T.N., Ishmukhamedov, R.N., Konov, V.V. and Safarov, K.S. (1990) Biokhimiya 55, 1257-1265. 5 Saltiel, A.R. and Cuatrecasas, P. (1986) Proc. Natl. Acad. Sci. USA 83, 5793-5797. 6 Selwyn, M.J., Dawson, A.P. and Fulton, D.V. (1979) Biochem. Soc. Trans. 7, 216-219. 7 Garlid, K.D. and Beavis, A.D. (1986) Biochim. Biophys. Acta 853, 187-204. 8 Beavis, A.D. (1992) J. Bioenerg. Biomembr. 24, 77-90. 9 Selwyn, M.J., Ng, L.T. and Choo, H.L. (1990) FEBS Lett. 269, 205-208. 10 Selwyn, M.J., Comerford, J.G., Dawson, A.P. and Fulton, D.V. (1986) Biochem. Soc. Trans. 14, 1043-1044. 11 Halle-Smith, S.C. and Selwyn, M.J. (1990) Biochem. J. 266, 689-692. 12 Fiske, C.H. and SubbaRow, Y. (1925) J. Biol. Chem. 66, 375-400. 13 Ashwell, G. (1966) Methods Enzymol. 8, 185-98. 14 Gornall, A.G., Bardawill, C.I. and David, M. (1949) J. Biol. Chem. 177, 751-766. 15 Taussky, H.H. (1961) in Standard Methods in Clinical Chemistry 3, 99-113. 16 Cardini, C.E. and Leioir, L.F. (1957) Methods Enzymol. 3, 114115 and 835-840. 17 Beavis, A.D. (1989) Eur. J. Biochem. 185, 511-519. 18 Beavis, A.D. (1991) Biochim. Biophys. Acta 1063, 111-119. 19 Soboll, S. and Sies, H. (1983) in Isolation, Characterization and use of Hepatocytes (Harris, R.A. and Cornell, N.W., eds.), pp. 295-301, Elsevier, Amsterdam. 20 Gainutdinov, M. Kh., Konov, V.V., Ishmukhamedov, R.N., Zakharova, T.N., Khalilova, M.A., Mamatova, Z.A., Asrarov, M.I. and Mirmakhmudova, S.L. (1990) Biokhimiya 55, 2239-2246. 21 Keogh, J.M., Matthews, P.M., Seymour, A.M. and Radda, G.K. (1985) Adv. Myocardiol. 6, 299-309. 22 Kalderon, B., Hertz, R. and Bar-Tana, J. (1992) Endocrinology 131,400-407. 23 Kaminsky, P., Robin-Lherbier, B., Brunotte, F., Escanye, J.M., Walker, P., Klein, M., Robert, J. and Duc, M. (1992) J. Clin. Endocrinol. Metab. 74, 124-129. 24 Paradies, G. and Ruggiero, F.M. (1990) Biochim. Biophys. Acta 1019, 133-136. 25 Paradies, G., Ruggiero, F.M. and Dinoi, P. (1991) Biochim. Biophys. Acta 1070, 180-186. 26 Hafner, R.F., Brown, G.C. and Brand, M.D. (1990) Biochem. J. 265, 731-734. 27 Kunz, W., Gellerich, F.N., Schild, L. and Sch6nfield, P. (1988) FEBS Lett. 233, 17-21. 28 Tanaka, A., Chance, B. and Quistorff, B. (1989) J. Biol. Chem. 264, 10034-10040.