P/~uvz/,r (IYYZ), 13, 463-473

Subcellular Localization and Properties of Adenosine Diphosphatase in Human Placenta

F. MARTiNEZ”, R. MONCADA*, F. J. BARCENAS* & T. ESPINOSAGARCfA Departamnto de Bioquimica, Facultad de .Uedicinn, Ckiz~ersidod NacionalAutonoma de.Wxico, .ApartadoPostal 7O-179, Co~aa& 04510, .zIPsiro,D. F., .Ilexico ” To mhom correspondenceshould be addressed

Paper accepted2. I2.1991

SUMh4ARY It was jound that naitochondria from human placenta exhibited an ‘4DPase activity with the .following characteristics. The enzyme responsible for this actizity was associated with the inner mitochondrial membrane. It was not released by treatment of the submitochondrialparticles with solutions of high ionic strength. Maximal ADP hydrolysis was reached at pH 8. Specific inhibitors jar alkaline myokinase (Pt,P”-di(adenosine-S)pentaphosphatase (L-phenylalanine), phosphate), nr 5’-nucleotidase (concanacalin A) did not decrease ADP hydrolysis. /1TP synthesis from ADP by myokinase was about 13 nmol/mg/min, whereas ADP hydrolysis reached values around 500 to 550 nmol/nag/min, indicating that a myokinase-H+ATPase combination could not account jar the obsened rates ofADP hydrolysis. The acti@ was stimulated by M2’, but high concentrations of this cation produced inhibition. High ADP concentrations did not inhibit ADPase acti@. Kinetic measurements ofthe acticity in thesubmitochondrialparticles showed that thr true substrate was ADP-Mg. The kinetic studies showed V&t, values of 476 and 270 nmol/mg/min, and Km,,, calues of 416 and 8.7,~tus~

INTRODUCTION ADP is used by the cell for ATP synthesis and its metabolic fate is determined either by phosphorylation through oxidative phosphorylation or by kinases such as adenylate kinase. However, ADP hydrolysis by enzymes from different cells has been reported in lungs (Crutchley, Eling and Anderson, 1978), rat heart (DeVente, Velema and Zaagsma, 1984), ’ Students

from ITIEM (Integration

0143-+004/92/050463

Temprana

+ 11 $08.00/O

a la Investigation

de Estudiantes

de Xledicina). 0

1992 BailliPre Tindall

I,td

Placenta (I 992), Vol. 13

464

rabbit muscle (Lanzara, Magri and Grazi, 1986), rat liver (Smith, Smith and Peters, 1980), human polymorphonuclear leukocytes (Smith and Peters, 1981a), smooth muscle from vessels (Pearsen, Carleton and Gordon, 1980; Smith et al, 1981b), endothelial cells and fibroblasts (Dosne et al, 1978). In some of these tissues, the ADPase activity has been shown to exist in mitochondria (Smith, Smith and Peters, 1980), or associated with mitochondria and microsomal fractions (Smith et al, 1981b). Although a role for some ADPase has been suggested (Lanzara, Magri and Grazi, 1986; Pearsen, Carleton and Gordon, 1980), its function still remains largely unknown. In human placenta, adenosine nucleotidases were described (Cerletti, Fronticelli and Zichella, 1960a; Cerletti and Zichella, 1960b), but no characterization ofthe enzymes was made. The present study relates to the subcellular localization and characterization of the ADPase activity of human term placental tissue.

MATERIAL

AND METHODS

Isolation of mitochondxia, mitoplasts, and submitochondrial particles Placentae were processed within 30 min of delivery. The tissues were washed three times with 0.25 ~4 sucrose, 1 mu EDTA, pH 7.3. Mitochondria were obtained as reported by Martinez et al (1987), and suspended in the same medium. Protein was determined as described by Lowry et al (1951), using bovine serum albumin as standard. Respiratory control (RC) of mitochondria was assessed as reported earlier (Martinez et al, 1987); only mitochondria with RC higher than three were used to prepare the submitochondrial particles for kinetic determinations of ADPase activity. Mitoplasts were prepared with digitonin, using the method reported by Schnaitman and Greenawalt (1968). Submitochondrial particles were obtained by suspending mitochondria in 0.25 I\/I sucrose, 1 mu EDTA, pH 7.3, at a concentration of 20 mg/ml; the pH of the suspension was adjusted to 8.3 with Tris-base. This suspension was exposed by sonic treatment in aliquots of 50 ml on ice, using an MSE Soniprep model 150 at maximal input for three periods of 45 set with 60-set intervals, The sonicate material was centrifuged at 12 000 g for 10 min, and the supernatant was centrifuged at 120 000 g for 45 min. The pellet containing the submitochondrial particles was suspended in 0.25 14 sucrose, 1 rn_MEDTA, pH 7.3. Determination of enzyme activities Lactate dehydrogenase (LDH, EC 1.1 .1.27) was assessed as described by Bergmeyer and Bernt (1974). Cytochrome oxidase (EC 1.9.3.1.) was assayed polarographically in a YSI apparatus model 53 in a medium containing 0.5 rnx{ TMPD (N,N,N’,N’-tetramethyl pphenylene diamine); 7 rnhi ascorbate; 5 pM cytochrome r, 100 nivl antimycin, 0.25 hl sucroseTris, adjusted to pH 7.3, and 2 mg of protein. The reaction was initiated by the addition of TMPD. Oxidation of NADH-dichloro phenol indophenol (DCPIP) by the electron transport chain was performed in a medium containing 50 mM KzHP04, 100 mhl KCN, 200 ,UUM NADH, and 80 ,uM DCPIP, pH 7.6. The final volume totalled 2 ml and contained 1 mg of protein. The reaction was initiated by the addition of DCPIP and its reduction was followed in a spectrophotometer Aminco DW-2a by wavelength difference between 590 and 600 nm. An extinction coefficient of 10.1 mM/cm was used to calculate the activity. Oxidation of 20 rnlu succinate was performed in a similar way, but 5 ,ug rotenone and 1.1 rnM phenazine methosulphate were added. Reduction of cytochrome c by submitochondrial particles was

Ilartimz et ai: /1DPaseActi@

in Human Term Placenta

-MC

assayed in a medium with 50 rn31 KzHP04, pH 7.6,5 ph’ NADH, and 0.5 mg cytochrome L’. Reduction was followed spectrophotometrically by recording the difference between 540 and 550 nm. An extinction coefficient of 9.5 m&‘/cm was used for the purposes of all calculations.

Assay of adenine nucleotides hydrolysis Hydrolysis of ADP, ATP, AMP, or p-nitrophenyl phosphate (pNPP) was performed in an incubation medium containing 30 rnhl Tris-HCI, pH 8, 1 m&l MgC12, and 1 or 2 mhl of the substrate. The final volume was 1 ml and the temperature of the reaction was 30°C. ADPase activity at a different pH was measured with the following buffers: citrate for a pH range from 3 to 5, MES from 6 to 6.5; Tris-HCl from 7 to 9, and carbonate/bicarbonate from pH 10 to 11. The reaction was stopped by trichloroacetic acid (6 per cent final concentration). An aliquot of clear supernatant was used for phosphate determination using the methods described by Sumner (1954). .ADPase activity was also assayed in the presence of an ATP-trapping system in a medium containing 30 mhl Tris-HCl, pH 8, 1 m&l MgC12, 1 m\’ ADP, 20 ml{ glucose, and 10 U hexokinase (HK, EC 2.7.1.1). ADP hydrolysis was also determined in the presence of SO!L\’ P’,P”-di(adenosine-5’)pentaphosphate (Ap5), a specific adenylate kinase inhibitor (Lienhard, 1973). Experiments with Concanavalin A (3.41, 6.82, and 13.64 ,~,l) were performed using 250 /rg of protein from submitochondrial particles. After incubation at 30°C for 10 min, the reaction was initiated by the addition of 3 rnM ADP, and stopped after 4 min by adding 6 per cent trichloroacetic acid, with the phosphate released being quantified.

Determination of ATP synthesis in submitochondrial particles ATP synthesis by myokinase in submitochondrial particles was measured spectrophotometrically, following the reduction of NADP+ in a medium containing 50 rr~ Tris-HCI, 3 mh’ Mg-acetate, 20 rn41 glucose, 8 rnhl phosphate, pH 7.3, 170 rn%i sucrose, 10 U HK, and 10 U glucose-6-phosphate dehydrogenase (G6P-DH, EC 1.1.1.49). After 8-10 min of recording, 8 ma’ AMP was added to estimate its inhibitor effect on myokinase activity.

Quantification of ADP after hydrolysis The stoichiometry of ADP hydrolysis and the released phosphate was assessed by quantitying the ADP concentration at different times as described by Jawore, Gruber and Bergmeyer (1974); phosphate was determined as mentioned before.

Kinetic determinations For kinetic determinations, the submitochondrial particles were washed with 0.5 ~1 KC1 and passed through a discontinuous sucrose gradient (15,25 and 40 per cent) at 120 000 g for 2 h to eliminate any possible contamination. The submitochondrial particles in the interface between 25 and 40 per cent of sucrose were used. Calculations of free ADP, free Mg2+, or the ADP-Mg complex were carried out as reported by Fabiato (1988). Vapp and Kmapp were calculated using the simplex method (Caeci and Cacheris, 1984). The reagents were purchased from Sigma Chemical Co., St Louis, MO, USA. Other reagents were analytical grade products. Data represent the average of six independent determinations. In all cases, the standard deviation was under 15 per cent.

466

Hacetlta (I 992), Vol. 13 Table 1. ADPase and ATPase activities in different submitochondrial Total ADPase activity

Total homogenate Mitochondria Mitoplasts Sonicated

8.63 6.36

mitochondria

Submitochondrial

Yield (1%)

particles

100 73

particles from human term placenta .ATPase/ADPase ratio

-1DPase

ATPase

28 130 250

46 197 343

1.64 1.M 1.37

2.13

22.5

400

595

1.48

6.49

68.6

433

661

1.52

Mitochondria and mitoplasts (1 mg) were added to a solution containing 30 m\t Tris-HCl, pH 7.3; SO+1 DNP, and 100 mht KCI. After 5 min preincubation, the reaction u-as initiated b!- the addition of 2.5 rn\i ADP or ATP. The ADPase and ATPase activities in submitochondrial particles were determined in a similar way, but replacing DNP by 1 rnxt MgCla. Final volume 1 ml. Temperature 30°C; n = 3. The results are {cmol Pi released/mg/min for total ADPase activity, and as nmol Pi released/mg/min for total ADPase and ATPase activities.

RESULTS Distribution of ADPase activity in various subcellular fractions Table 1 shows the subcellular ADPase activity distribution in human term placenta. The mitochondrial fraction showed the highest rate of ADP hydrolysis. Preparations of mitoplasts and inner mitochondrial membrane showed that the activity was principally localized near to the latter fraction. The H+ATPase (EC 3.6.1.4) is localized exclusively to the inner membrane of mitochondria; this activity was determined in the various preparations and, it was found that its distribution paralleled that of the ADPase. Of the total ADPase activity, 75 per cent was found in mitochondria and, in this organelle, 70 per cent was in the submitochondrial particles. To estimate the degree of contamination of mitochondria with other cellular fractions, the activity ofvarious marker enzymes was determined. The results show that: submitochondrial particles had 6 per cent LDH activity as compared to that of the total homogenate; Cytochrome oxidase activity in submitochondrial particles increased by about 50 per cent with respect to the activity in whole mitochondria; Succinate dehydrogenase, NADH dehydrogenase activities, and the reduction of cytochrome c in mitochondria were in the particles the activities were range of nmol/mg/min (NegriC et al, 1979); in submitochondrial two or three orders of magnitude higher (Table 2). The specific ATPase and ADPase activities of submitochondrial particles were 20 times higher than those of the total homogenate. These results indicate that the enrichment of ADPase activity occurs in parallel to that of the enzymes that localize near the inner mitochondrial membrane (Tables 1 and 2).

Table 2. Determination

of some inner membrane activities of submitochondrial particles from human placental tissue

NADH-cptochrome c NADH-DH DCPIP Succinate DCPIP Cytochrome oxidase

8.55 4.08 6.48 46.30

+ f k +

0.50 pmol 0.26 ymol 0.36pmol 4.50 ng at

Cp c reduced/mg/min (n = 4) DCPIP/mg/min (n = 5) DCPIP/mg/min (n = 5) oxygen/mg/min (?r = 3)

The enzymatic activities were performed as described under Material and Methods. The enzymatic activities in whole mitochondria (succinate oxidase and cytochrome oxidase) were reported in the range nmol by NCgriC et al (27).

I lortj,,~~.. etal: -lDPa.w

Lrbll,.i.

ictiri~~,in Hwmztz Term Placenta Hydrolysis

465

of .4DP and ATP in submitochondrial particles sonicated with different KC1 concentrations .4DPase

-

576 600 610 640 645

and

_iTPase

nmol Pi released/mg/min Control R-ashed with 100 mv KC1 Sonicated with 0.1 \I KC1 Sonicated with 0.5 11 KC1 Sonicated with 1.0 \I KC1

washed

792 810 875 942 910

ATPaxe/-\DPase ratio

1.37 1.35 I.43 l.-+i I.-i1

Submitochondrial particles were obtained as described under Alaterial and Llethods. Once obtained, KC1 was added to the concentration shown, and then disrupted by sonic treatment, in periods of 45 set and intervals of 60 set, three cycles. 4fter this treatment, the submitochondrial particles were recovered in 0.25 \I sucrose. 1 m\t EDT-\, pH 7.3. Hydrolysis conditions were as in Table 1. N = 3.

Possible sources of an apparent ADPase activity ADPase could be a protein associated unspecifically with mitochondria and submitochondrial particles. To explore this possibility, the submitochondrial particles were washed with 100 mht KCl. In another series of experiments, the submitochondrial particles were suspended in 0.1, 0.5,or 1 41 KCl, sonicated and subsequently recovered by ultracentrifugation. Table 3 shows that the ADPase activity was not released from submitochondrial particles by these treatments, but a small increase in activity was observed under the same conditions. This suggests that this enzyme is tightly bound to the mitochondrial membrane. In fact, the ratio of H+ATPase/ADPase activity remained fairly constant regardless of the v-arious treatments. AL>Pase activity, assayed at different pH values, exhibited a high activity between pH 8 and 10 [Figure l(a)]. These results could suggest the presence of two enzymes. Indeed, it has been described that human placenta has a high concentration of alkaline phosphatase (EC 3.1.3.2) (Farnlay, 1971; Doellgast et al, 1977), with an optimal pH around 10 and which is strongly inhibited by EDTA (Harkness, 1968). n:;‘e found that ADP hydrolysis was significantly lower in the absence of added Mg2+ at pH 8 [Figure l(b)], but not at pH 10; at this latter pH, EDTA inhibited the activity [compare data of Figures 1 (a) and (b)]. These results suggest the presence of a Mg-independent alkaline phosphatase; with respect to the effect of EDTA, it is to be noted that there are various reports that indicate that alkaline phosphatase is inhibited by the removal of zinc ions by EDTA (Farnley, 1971; Harkness, 1968). The time courses for ADP and p-nitro phenyl phosphate (pNPP) hydrolysis at pH 8 were determined. Figure 2 shows that the highest activity observed was with ADP. In addition, when the effect of L-phenyl alanine, a potent inhibitor of alkaline phosphatase activity in human placenta (Fishman and Ghosh, 1967), was tested on the pNPPase and ADPase activity, it was found that it produced an important decrease of the phosphate released from pNPP (about 72 per cent at 20 min), whereas the ADPase was inhibited 6 per cent. The data suggest that the phosphate released from ADP at pH 8 occurs mainly through the action of an ADPase and not by an unspecific phosphatase. Figure 2 also shows that ADP hydrolysis is nearly linear with time at least in the first minutes of incubation. Thusthe overall data indicate that hydrolysis at pH 10 occurs through an unspecific phosphatase, whereas the activity at pH corresponds to that of the ADPase. It was also considered that the ADPase activity could be due to a contamination of a myokinase activity coupled to ATP hydrolysis, i.e. myokinase would form ATP from ADP

468

Placenta (1992), lid. 13

5oc

400

5oc

I_

lb)

E \ r

300

2 w B TiJ L c 0 s

200

-

E

E =

100

200

,

0

1 3

I

I

I

I

II

7 PH

100

L 7

8

9

IO

PH

F&we 1. Effect ofpH and Mg*’ on the ADPase activity of submitochondrial particles from human placenta. .4DP hydrolysis was performed in a medium containing 5 rnht MgCIL, 100 mar KCI, and 2.5 m\t ADP. The experiment depicted in (a) was carried out as described under IMaterial and IMethods. The buffer concentration was 30 m\t. (b), ADPase activity was assessed in absence of added Mg*+ (O),and in absence of added Mg*+ plus 10 mxt EDTA (0). The buffer conditions to maintain the indicated pH value were as in (a). Final volume 1 ml. Temperature 30°C. n = 7.

followed by hydrolysis of ATP through the mitochondrial H+ ATPase, giving a stoichiometry of 2:l. Experiments carried out to explore this possibility showed that the presence in the incubation medium of P’,P’-di(adenosine-5’)pentaphosphate (Ap5), an inhibitor of myokinase activity (Lienhard, 1973), did not change the rate of ADP hydrolysis. In addition, to ascertain the contribution of myokinase activity to the presently studied ADPase activity, the capacity of submitochondrial particles to synthesize ATP by myokinase activity was determined. Maximal ATP production by myokinase reached values around 13 nmol/mg/min with 3 rnM ADP added, and this synthesis was inhibited 80 to 90 per cent by the addition of 8 rnM AMP. Since, on average, the rate of ADP hydrolysis with 1 rntil ADP was 550 nmol/ mg/min, these experiments indicated that myokinase activity could not account for the amount of phosphate formed from ADP. Furthermore, it was also found that 8 rnkr AMP, which inhibited the formation of ATP from ADP by more than 80 per cent, did not affect ADP hydrolysis, even at a concentration of 20 rnM AMP (data not shown). By the same token, ADPase activity was measured in the presence of an excess of ATPtrapping system, i.e. glucose plus hexokinase. The result showed that ADPase activity was around 550 nmol/mg/min, indicating that in the presence of this ATP-trapping system, the ADPase activity was not modified. Therefore, myokinase activity does not contribute to a significant extent to the ADPase activity of submitochondrial particles. Furthermore, the stoichiometry ofADP hydrolysis and phosphate released was investigated in the same system

2.5

-

2.0

-

Y 0

2 2

I.!?-

%

P m L

h 0E

I.0

-

0.5

-

a

0.0

0

4

8 Time

F