Pulmonary phosphatidic acid phosphatase. A comparative study of the aqueously dispersed phosphatidate-dependent and membrane-bound phosphatidate-dependent phosphatidic acid phosphatase activities of rat lung

Pulmonary phosphatidic acid phosphatase. A comparative study of the aqueously dispersed phosphatidate-dependent and membrane-bound phosphatidate-dependent phosphatidic acid phosphatase activities of rat lung

226 Biochimica et Biophysics Acta, 574 (1979) 0 Elsevier/North-Holland Biomedical Press 226-239 BBA 57406 PULMONARY PHOSPHATIDIC ACID PHOSPHATAS...

998KB Sizes 0 Downloads 16 Views

226

Biochimica et Biophysics Acta, 574 (1979) 0 Elsevier/North-Holland Biomedical Press

226-239

BBA 57406

PULMONARY

PHOSPHATIDIC

ACID PHOSPHATASE

A COMPARATIVE STUDY OF THE AQUEOUSLY DISPERSED PHOSPHATIDATE-DEPENDENT AND MEMBRANE-BOUND PHOSPHATIDATE-DEPENDENT PHOSPHATIDIC ACID PHOSPHATASE ACTIVITIES OF RAT LUNG

ALEX YEUNG *, PAUL G. CASOLA FRED POSSMAYER a*b *

b, CHING WONG a, J. FRASER

FELLOWS

a Department of Obstetrics and Gvnaecoloav and b DeDartment of Biochemistry, of Western Ontorio, London, Ontario, N6AiA5 (Canaha) (Received November (Revised manuscript

30th, 1978) received February

Key words: Phosphatidate phosphatase;

16th,

a and University

1979)

(Rat lung, Membrane-bound,

Dispersed)

Summary 1. The properties of the aqueously dispersed phosphatidate-dependent phosphatidic acid phosphatase (EC 3.1.3.4) activities of rat lung have been studied in microsomal and cytosol preparations and compared with the properties of the membrane-bound phosphatidate-dependent activities. 2. The microsomal phosphatidic acid phosphatase displayed a prominent pH optimum at 6.5 with a minor peak which varied between 7.5-8 in different experiments. With the cytosol, the major activity was at the higher pH (7.58.0) but a distinct optimum was also observed at pH 6.0-6.5. With the membrane-bound substrate, a single broad optimum was observed between pH 7.4 and 8.0 with the cytosol and 6.5-7.5 with the microsomal fraction. 3. Subcellular fractionation studies revealed that the microsomal fraction possessed the greatest proportion of the total phosphatidic acid phosphatase activity and the highest relative specific activity. However, studies with marker enzymes indicated that the aqueously dispersed phosphatidatedependent activity could be present in plasma membrane, lysosomes and osmiophilic lamellar bodies as well as in the endoplasmic reticulum. 4. The aqueously dispersed phosphatidic acid-dependent activities present in the microsomal and supernatant fractions were inhibited by Ca’+, Mn”, F- and * To whom reprint requests should be directed. Abbreviations: The abbreviations used are dePined in the preceding paper [II.

227

phosphaby high concentrations of Mg*+. In contrast to the membrane-bound tidatedependent activities, there was little Mg*+ stimulation and only a very slight inhibitory effect was noted with EDTA. A small EDTA-dependent Mg*+ stimulation could be observed with the microsomal fraction but only at the lower pH optimum (6.5). 5. The presence of a number of phosphate esters tended to stimulate rather than inhibit the microsomal activity, indicating that the hydrolase is relatively specific for lipid substrates. Marked inhibitions were noted with lysophosphatidic acid and phosphatidylglycerol phosphate. Phosphatidylcholine produced a slight inhibition. 6. The results indicate that the bulk of the aqueously dispersed phosphatidatedependent phosphatidic acid phosphatase activities of rat lung microsomes and cytosol is not related to the activities observed with membranebound phosphatidate. The Mg*‘-dependent hydrolase activities may by synonymous. However, unequivocal conclusions will only be possible when the polypeptide or polypeptides responsible for these activities can be purified.

Introduction Lung tissue contains at least four operationally distinguishable phosphatidic acid phosphohydrolase (PAPase) activities: (i) an aqueously dispersed phosphatidic acid (PA,,)dependent membrane-associated activity; (ii) a PA,,-dependent activity associated with the high speed cytosol; (iii) a membrane-bound PA (PA,,)dependent membrane-associated activity; and (iv) a PAmbdependent activity associated with the cytosol [l--8]. It should be stressed that any one of these operationally defined activities (e.g., the PA,, membraneassociated activity) may be catalysed by more than one protein. The converse may also be true: more than one of these activities (e.g., the PA,,, activities associated with the cytosol and the microsomal fractions) could be catalysed by a single polypeptide whose properties are altered by its environment (e.g., release from the membrane) or the physical nature of the substrate. The PA,,-dependent PAPase of fetal lung increase as term is approached [2,6] and after the induction of pulmonary maturation by glucocorticoids [3,4]. These observations have led to the suggestion that by controlling the production of, 1,2diacyl-sn-glycerol, these activities could function in the regulation of the synthesis of phosphatidylcholine [2,4--81 for pulmonary surfactant. The significance of these investigations, conducted with aqueously dispersed substrate, is questioned in the accompanying paper [ 11 which describes the PA,,-dependent PAPase activities of rat lung and emphasizes their potential role in the synthesis of phosphatidylcholine. In the present report, the relation between the PA,, and PA,,,,, -dependent PAPase activities of rat lung was carefully investigated. While there was some similarity in the properties of these four activities, it is apparent that the PA,,-dependent PAPase activities of microsomes and cytosol resemble each other more closely than the PA,,dependent activities. The PAPase activities assayed with membrane-bound substrate also demonstrated common characteristics.

228

Materials and Methods Unless stated otherwise biochemicals were purchased from Sigma Chemical Co., St. Louis, MO, U.S.A., and radioactive compounds were obtained from New England Nuclear (Canada, Ltd.), Montreal, Canada. Other reagents, of the highest quality available, were normally obtained from Fisher Scientific, Toronto, Ont., Canada. PA (sodium salt) and 1,2diacyl-sn-glycerol were prepared from egg phosphatidylcholine as previously described [ 3,9]. [32P] PA was prepared by incubating mc-glycerol 3-[32P]phosphate with rat liver microsomes as described in the accompanying paper [ 13. The radioactive PA was isolated by TLC on silica gel G plates impreganted with 0.35 M oxalic acid using light petroleum/acetone/ formic acid (154:46 : 0.5, v/v), extracted 3 times with chloroform/methanol (2 : 1, v/v) washed by the method of Bligh and Dyer [lo] and converted to the sodium salt [ 31. Subcellular fractions were prepared from adult Sprague-Dawley rats (200300 g) essentially as previously described [ 111. The fractions were redispersed with 0.32 M sucrose/O.1 mM EDTA (pH 7.2-7.4) to produce a final protein concentration of 5-10 mg protein/ml. PA,,-dependent PAPase was routinely assayed with a reaction mixture containing in a final volume of 0.125 ml: 50 mM Tris-maleate buffer, pH 7.4 (unless otherwise stated), 5 mM sodium phosphatidate (converted to liposomes by incubating in 0.9% NaCl, at 37”C), and enzyme (microsomes, 30-50 E.cg protein; supernatant 150-300 pug protein). The vessels were incubated at 37’C for 60 min and the reaction terminated with 1.0 ml of ice-cold trichloroacetic acid (lo%, w/v). Pi released was measured as previously described [3]. PA,I,-dependent PAPase activity was determined as described in the accompanying paper [ 11. The remaining assay procedures have been published previously [ 3,9]. Results Apparent distribution of the PA,, and PA,b -dependent soluble and microsomal fractions of rat lung

PAPase activities in the

Although PA,,-dependent PAPase has been detected in all subcellular fractions of rat lung [6,7], only those activities present in the microsomal and cytosol fraction could logically function in the de novo synthesis of phosphatidylcholine. In Table I, the distributions of the PA,,, and PA,,-dependent activities of these fractions are compared on the basis of specific activity (per mg protein) and total activity (per g lung). The total activities were corrected for losses during the subcellular fractionation by using endoplasmic reticulum and cytosol markers as described in the legend. In agreement with studies in other tissues, the apparent distribution was dependent on the physical nature of the substrate. When PA,, was used, the microsomal fraction contained a far greater proportion of PAPase activity whether the results were expressed as total or specific activity. On the other hand, although the microsomes and cytosol possessed comparable specific activities with PA,,, the total activity of the cytosol was approx. 3-fold greater than that of the microsomal tissues. The

229

TABLE

I

DISTRIBUTION

OF

THE MICROSOMAL

THE

[32PlPAmb

AND CYTOSOL

and the PAaq- DEPENDENT FRACTIONS

PAPase

ACTIVITIES

BETWEEN

OF RAT LUNG

32P-labelled phosphatidate or aqueowly The subcellular fractions were assayed with membrane-bound dispersed phosphatidate. The total activities per g lung were calculated using the recovery of protein in the microsomal (11.1 mg per g lung) and the cytosol (52.5 per g lung) fractions. The microsomal value was further corrected for the recovery of microsomes (48.6%) from the whole homogenate using the average recovery of three endoplasmic reticulum markers (NADPH : cytochrome c reductase (EC 1.6.2.4). estrone sulphatase (EC 3.1.6.2) and cholinephosphotransferase (EC 3.1.6.2) and the cytosol value was also corrected for the recovery of cytosol (78.6%) by using the average recovery of the cytosol marker, lactate dehydrogenase (EC 1.1.1.27) from three separate [email protected] Fraction

Membrane-bound

phosphatidate

Aqueous

dispersions

of phosphatidate Total activity (nmol/min per

g lung)

Specific activity * * (nmol/min per mg protein)

? 0.004

0.96

? 0.08

16.9

386.0

? 26.2

f 0.008

2.76

r 0.51

58.4

f 12.0

Specific activity * (nmol/min per mg protein) Microsomal

0.042

CYtosol

0.041

* Mean f S.E. of four separate * * Mean + S.E. of four separate

Total activity (nmol/min per

determinations determinations

with fractions with fractions

0.88 obtained obtained

i 1.1 + 0.18 from lungs pooled from lungs pooled

g lung)

from 6-8 rats. from 6-12 rats.

PA,,dependent activities of the microsomal and cytosol fractions were 400and 20-fold greater, respectively, than the corresponding activities measured with membrane-bound substrate. Although part of the differences in specific activity could be attributed to the limited amounts of PA which can be bound by the microsomal membranes used as substrate, the paradoxical distribution of the PA,, and PA,,-dependent activities among these two fractions can only be explained by a difference in substrate susceptibility. Nevertheless, these data cannot be taken as proof that distinct polypeptides are involved. For example, the release of a membrane-associated protein due to the cleavage of a hydrophobic portion could result in an altered configuration ‘which favours the recognition of membrane-bound phosphatidate. The possibility thus exists that the PA,,dependent activities could represent a limited expression or a small part of the total activities observed with PA,,. General properties of PA,,-dependent PAPase in rat lung Under the conditions utilized in these studies, the release of Pi from water dispersed sodium phosphatidate with microsomal and cytosol preparations was linear for at least 60 min, and over a considerable range of protein (up to 0.3 mg with the microsomes and 1.0 mg with cytosol). The specific activities of the microsomal preparations were 15- to 20-fold greater than the specific activity of the cytosol. The Km for PA,, was 0.454 f 0.075 (S.E.) mM with the microsomal activity and 0.735 f 0.180 (S.E.) mM with the cytosol. These activities were quite stable when the fractions were kept frozen at -7O”C, little change being evident over several months. The subcellular localization of PA,,dependent PAPase was investigated by comparing the relative specific activities of various marker enzymes for subcellular organelles with the distribution of PAPase (Table II). The profile of the activities of the marker enzymes, 5’-nucleotidase (plasma membrane), succinate

TABLE

II

DISTRIBUTION

OF PAaq- DEPENDENT

28.7 + 1.2 0.62 r 0.10

33.2 r 0.2 0.69 r 0.04

40.0 + 6.9 1.06 f 0.42

29.3 + 3.0 0.76 f 0.20

21.6 t 4.1 0.53 ?r0.13

40.9 f. 6.9 1.23 t 0.22

Succinate dehydrogenase 8.12 * 0.08 * (3) % Total RSA

Monooxidase 39.5 f 0.2 * * (2) % Total RSA

NADPH: cytochrome c reductase 0.619 ? 0.030 * (3) % Total RSA

Cholinephosphotransfera!ze (endogenous 0.24 f 0.08 * (3) 96Total RSA

Cholinephosphotransfemse 4.75 f 1.12 * (3) 96 Total RSA

Electrone sulphate sulphatase 16.6 + 0.9 * (2) % Total RSA

activity)

33.1 r 7.8 0.61 f 0.21

f 3.8

43.0

5’-Nucleotidase 3.22 ? 0.86 * (3) 96 Total RSA

Nuclear plus debris

ENZYMES

f 1.1

15.8 ? 0.6 4.78 f 0.79

19.0 k 2.4 3.55 f 0.49

11.8 + 3.7 2.07 f 0.23

12.2 r 3.2 1.73 f 0.58

37.8 ? 4.1 8.11 f 2.30

39.0 t 3.0 7.29 * 1.55

7.5 * 2.8 1.10 r 0.42

5.7

Mitocbondrial

OF MARKER

Relative distribution

PAPase AND

Protein 4%Total

Specific activity (per mg protein)

The results are expressed as the mean i S.E.

SUBCELLULAR

+ 1.1

+ 6.3 f 0.75

37.4 + 3.8 6.00 ? 0.35

58.6 k 6.2 7.09 t 1.35

53.1 f 5.9 6.95 ? 2.14

38.5 3.9

21.4 f 3.4 2.41 ? 0.32

a.3 f 0.5 0.93 + 0.07

36.6 t 6.2 3.12 f 0.54

7.7

Microsomal

IN RAT LUNG

+ 1.4

5.9 ? 2.4 0.20 * 0.07

0.8 f 0.3 0.02 c 0.01

5.9 * 2.2 0.17 r 0.07

7.8 ? 3.4 0.19 f 0.09

10.1 + 4.6 0.26 t 0.13

26.0 + 1.5 0.64 ? 0.05

22.8 f 1.4 0.42 + 0.04

40.3

supematant

f 7.0

153.2 + 16.0

112.1 ? 18.6

121.8 t 32.9

91.6

106.5 + 5.3

106.0 f 6.2

81.1 * 9.5

96.5 + 1.0

Rec0vel-Y (% homogenate)

231

f

P

1

232

dehydrogenase (mitochondrial inner membrane), monoamine oxidase (mitochondrial outer membrane), NADPH:cytochrome c reductase (endoplasmic reticulum), cholinephosphotransferase (endoplasmic reticulum), estrone sulphatase (endoplasmic reticulum), acid phosphatase (lysosomes), and lactate dehydrogenase (cytosol) were consistent with the anticipated nature of the fractions and previous studies conducted with lung [6,7,11--141. Because of the two pH optima observed (Fig. l), and since the PAmb-dependent activities were assayed at pH 7.4, the PA,,-dependent PAPase activity was estimated at pH 6.5 and 7.4 (Table II). In both cases, the highest relative specific activities and the largest proportion of the total activity were associated with the microsomal fraction. However, in agreement with previous studies [ 6,7], the distribution of PA,,-dependent PAPase did not precisely parallel the distribution of the marker enzymes for the endoplasmic reticulum. Nor did the profile of the PAPase activities resemble the distribution of the other marker enzymes. The possible significance of the observed distributions are treated more fully in the Discussion. Caras and Shapiro [ 151 have demonstrated that the PA,,-dependent activity of rat liver microsomes can be fractionated to yield a specific and a non-specific PAPase activity. The substrate specificity of rat lung microsomal PAPase was (Table III). investigated with aqueously dispersed 32P-labelled phosphatidate In several experiments it was observed that the addition of a number of phosphate esters at various concentrations produced a stimulation in the release of 32Pi. A similar phenomenon was observed with the PA,,-dependent activity of the cytosol (see accompanying paper [l]). These results indicate the microsoma1 activity towards [32P] PA,, is rather specific. In contrast, the addition of phosphatidylglycerophosphate (kindly provided by Mr. Peter McDonald, Dept. of Biochemistry, University of Western Ontario, London, Canada) and 2-lysophosphatidic acid (Serdary Research Laboratories) markedly inhibited the

2.0

4.0

6.0

6.0

10.0

PH

Fig. 1. Effect of pH on the PAaq-dependent tion and cytosol.

PAPase activities of rat lung homogenate,

microsomal frac-

233

TABLE

III

EFFECT

OF THE

ACTIVITY PAPase substrate

ADDITION

OF VARIOUS

PHOSPHATE

ESTERS

ON PAaq-DEPENDENT

PAPase

IN RAT LUNG MICROSOMES

activity

was assayed

was reduced

using aqueoualy

dispersed

to 2.5 mM in order to increaese Relative Addition

(mM):

P-Glycerophosphate a-Glycerophosphate p-Nitrophenylphosphate Glucose B-phosphate NADPH AMP pi Phosphatidylglycerophosphate Lysophosphatidic acid Phosphatidylcholine (egg)

sodium

[ 32P]phosphatidate.

the competitor/substrate activity

The concentration

of

ratio.

(% control)

1

2.5

5

116.2 133.2 116.8 112.4 108.6 117.2 -

118.7 116.2 124.3 112.5 112.3 126.1 120.7 30.0 30.4 87.5

95.8 127.4 114.3 108.9 100.6 122.9 111.5 12.7 77.6

release of 32Pi. Previous studies have suggested that these lipids can act as substrates for microsomal PA,,-dependent PAPase [ 7,8]. A slight inhibition was observed with the addition of 3-sn-phosphatidylcholine from egg yolk. Comparative

properties

of PA,,

and PA,,-dependen

t PAPase in rat lung

The pH profile of the release of Pi from PA,, varied with the different fractions (Fig. 1). With the microsomal fraction and the whole homogenate, there was a broad optimum at pH 6.5 and a smaller optimum or a shoulder in the region from 7.5 to 8.0. The minor pH optimum was observed in several experiments with the exact optimum varying from 7.5 to 7.8. Although the cytosol activity displayed an optimum in the pH region 6.0-6.5, there was also a greater activity with a pH optimum which varied from pH 7.5-8.0 in different experiments. Sedgwick and Hiibscher [16] have previously noted that the PA,,-dependent activity of membranous fractions from rat liver demonstrated more than one pH optimum and the pH of the optimum varied somewhat from experiment to experiment and with repeated freezing and thawing of the samples. However, in the present studies samples were thawed only once and discarded. The PAmI,-dependent PAPase activities of the whole homogenate (Fig. 2) exhibited a broad pH optimum between pH 7.0-8.0. A slightly higher optimum (pH 7.4-8.0) was observed with the cytosol, although slightly lower optima were observed in some experiments. The pH optimum of the microsomal activity was somewhat lower (pH 6.5-7.5). In contrast to the PA,,dependent activities, no evidence for a secondary peak or a shift in the relative contribution at different pH values could be observed. The ionic requirements for PAPase activities appear to vary considerably with different subcellular fractions in different tissues. The effect of the addition of various ions and chelators on the major PA,, and PAmbdependent activities of rat lung were compared. The PA,,-dependent activity with the cytosol (Fig. 3) was more susceptible to inhibition by Ca2+ or Mn” but more

234

I

5

I

I

6

I

I

I

7 PH

I

I

6

9

Fig. 2. Effect of PH on tbe PA,b-dependent cytosol.

activities

of rat lung homogenate,

microsomal

fraction and

resistant to inhibition by F- than the PA,,-dependent activity in the microsomes (Fig. 4). In contrast to the marked inhibition of the PAmb-dependent activities by EDTA and EGTA (see Ref. 1, and Fig, 3), a slight but consistent inhibition was observed with the microsoma_l PACT-dependent activity (Fig. 4).

Concentration (mtvl) Fig. 3. Effect of various ions and

chelators

on

the hydrolysis

of [ 3 2f?] PA,b

by rat

hang

CYtOsoI.

235

-------oMn

I

I

1

I

5.0

IO .Concentration (mM)

I

I5

Fig. 4. The effect of various ions and chelators on the PA,@ependent PAP&w activity somal fraction. Ions and chelating agents were added in the chloride or sodium forms.

of rat lung micro-

Although variable Mg2+ effects have been reported, even for the same tissue, in general low concentrations of Mg2+ stimulate the PA,,-dependent activities, particularly if the endogenous Mg2+ level is reduced. The addition of Mg” produced a slight but definite stimulation of the PA,,-dependent microsomal activity at low concentrations at both pH 6.5 and 7.4 (Fig. 5a). With greater concentrations of Mg2+, both activities were inhibited. These effects were noted in a number of experiments. These observations contrast with the relative resistance to inhibition below 10 mM Mg2+ observed with the PA,,-dependent activities in cytosol and microsomes (Fig. 3 and Ref. 1). The effects of ions and chelators on the PA,,-dependent activity of the cytosol were examined at pH 6.4 and 8.0 (Table IV). Although the activity at

A 30-

0 pH 6.5 -e---__._--_---

_---

. pH 7.4 1

5.0

IO

30 Mg Cl,&

Fig. 5. The effect of the addition of various concentrations activity of rat lung microsomes. The assays were conducted with 2.5 mM EDTA

present

(B).

of MgCl2 on the PA,,+ependent PAPase either under the standard conditions (A) or

236

TABLE EFFECT RAT

OF

LUNG

PAPase presented. cytosol

IV VARIOUS

IONS

AND

CHELATORS

ON

THE

PAaq -DEPENDENT

PAPase

ACTIVITY

OF

CYTOSOL

activity The

was

assayed

control

as described

activities

at pH

in the 6.0

and

text. 8.0

The were

mean 1.02

? S.E.

t 0.08

of and

2 independent 1.51

? 0.05

experiments nmol/min

per

is mg

protein. Relative pH Addition

(mM):

activity

(%

control)

6.0

pH 8.0

5

10

5

10

Mg2+

98

? 19

71

+ 16

111+

MnZ+

61

+ 10

45

2 22

101

7

93*

2

? 18

81f

7

ca2+

58+

1

39?

8

79

F-

55r

2

41?

5

87?

EDTA

78?

7

81

? 25

86?

5

92+_

EGTA

87

89

f 20

94*

1

95

? 22

+ 10

60?

7

3

77?

4

pH .6.5 was more susceptible to inhibition by Mg2+, Mn”, activities resemble the microsomal associated PA,,-dependent of relatively small inhibitions observed with EDTA.

5 r 13

Ca2+ and F- both activity by virtue

Discussion PA,;-dependen t membrane-associated PAPase In agreement with a number of studies in various tissues [ 16 -201 including lung [2,6,7], the major proportion of the PA,,-dependent PAPase activity was associated with the membranous fractions. The microsomal activity demonstrated a major optimum at pH 6.5 with a minor optimum at pH 7.8. Although previous reports [6,7] have concluded that EDTA and Mg” did not affect the microsomal activity, a small decrease was consistently observed with EDTA and Mg2+ produced a slight stimulation at low concentrations followed by an inhibition. Furthermore, in the presence of EDTA, the activity assayed at pH 6.5 selectively demonstrated a small Mg2’-dependent stimulation reminiscent of the more striking stimulation observed with the PA,,-dependent activities described in the accompanying paper [ 11. The observation that the microsomal fraction contained the greatest proportion and the highest relative specific activity of the PA,,-dependent activity in rat lung is consistent with previous studies in adult rat lung [ 6,7]. These reports emphasized the potential role of the microsomal activity in the production of diacylglycerol for phosphatidylcholine synthesis. Therefore, it is important to note that both in these publications [6,7] and in Table II, the distribution of PA,,-dependent PAPase differed somewhat from the distribution of the endoplasmic reticulum markers. In tissues other than lung, PA,,-dependent PAPase activity is also present in lysosomes [ 16,21,22], plasma membrane [ 20,231 and perhaps mitochondria [16]. Of particular significance to this discussion is the conclusion that a,major proportion of the PAPase activity of rat lung may be localized in the lysosomes [16,21,22]. Histological studies with hamster lung

237

have suggested that the PA,,-dependent PAPase is principally associated with the interior aspect of the limiting membrane of the lamellar bodies [5]. These organelles function in the storage and secretion of surfactant [ 24,251: operations consistent with their lysosomal nature [25 -271. This overall scheme could explain the presence of PA,,-dependent PAPase activity in lung wash surfactant and the correlation between PAPase activity and the phosphatidylcholine/sphingomyelin ratio in amniotic fluid (Benson, B.J. and Clements, J.A., as cited in Ref. 29) [28,30]. It has recently been shown that the PAPase activity in amniotic fluid is associated with organelles identical in appearance to lamellar bodies (Clements, J.A., personal communication). Nevertheless this interpretation is still controversial. Although Spitzer et al. [ 311 have concluded that in pig lung the lamellar bodies contain over 40% of the PA,,-dependent activity, Mavis et al. [7] have reported that rat lung lamellar bodies do not contain a significant proportion of the total activity. Furthermore, when these bodies are extruded into the alveolus, the lamellar body outer membrane remains fused to the cell membrane [25]. Consequently, the association of PAPase with the limiting membrane of the lame&r body cannot explain its presence in lung wash or amniotic fluid but is compatible with its association with the plasma membrane [20,23]. Finally, since lamellar bodies lack a number of enzymes required for phosphatidylcholine synthesis [14], these organelles cannot function in surfactant production as has been previously suggested [ 5,28,31]. PA,,-dependent

cytosol-associated

PAPase

In agreement with studies in lung [2,6,7] and other tissues [16,17,19], only a small proportion of the total PA,,-dependent PAPase activity was present in the cytosol. The observation that the major activity in the cytosol demonstrated a pH optimum between 7.5 and 8.0 was of interest in view of its similarity to the pH optima reported for the soluble activity in other tissues [19,32] and the pH optimum of the PA,, -dependent activity of lung cytosol. However, in terms of their response to divalent cations and fluoride, the PAPase activities at both pH optima resembled the membrane-associated PA,,-dependent PAPase. Furthermore, the addition of EDTA, which abolished the cytosolassociated PA,,, -dependent activity only slightly depressed the activity with aqueously dispersed substrate at either pH. It has previously been reported that the PA,,-dependent activity assayed at pH 6.0 is not susceptible to EDTA and/ or Mg2+ [ 6,7]. The results presented here for the PA,,-dependent activity are surprising in view of the numerous reports that Mg*+ markedly stimulates the analogous cytosol activity in liver [33] and adipose tissue [17,34]. Nor was the activity in lung cytosol stimulated by phosphatidylcholine as was previously observed with these tissues [ 33,341. Jambdar and Fallon [ 191 have presented evidence indicating that in adipose tissue the Mg*‘-dependent PAPase activity observed with PA,, is equivalent to the activities observed with PAmb. Conditions producing an alteration in the specific activity of the PA,,,-dependent PAPase activity of rat liver cytosol are consistently reflected in the Mg2’-dependent activity with aqueously dispersed phosphatidate when assayed at pH 7.4 under the appropriate conditions (Brindley, J.A., personal communication; Ref. 35).

238

However, the Mg*‘-dependent PAPase activity of rat lung cytosol can only account for a small proportion of the total activity observed with aqueous phosphatidates (Fig. 6 Ref. 6). PA,,-dependent cytosol-associated PAPase The properties of the PAmi,-dependent hydrolase activity in lung cytosol correspond to the activities previously investigated in liver [36], intestine [ 171 and adipose tissue [ 191. Although the activity was not highly dependent on exogenous Mg*+, it was abolished by EDTA [l]. This distinction and the observation that the PA,,-dependent activity is more susceptible to inhibition by Mn*’ and Ca*’ but less susceptible to Mg*+ appear to distinguish this activity from the bulk of the PA,,-dependent activities. However, it appears plausible that the phosphatidic acid liposomes can sequester divalent cations. It is also possible that the cation requirements for the interaction of the hydrolase with its substrate may be affected by the presence of other lipids and/or proteins on the microsomal membranes containing the bound phosphatidate. Finally, it is known that the divalent cation requirements of membrane-associated enzymes such as glycerol 3-phosphate acyltransferase [ 371 and phosphatidylglycerophosphate phosphatase (McMurray, W.C., personal communication) are altered when these polypeptides are removed from their natural environment. Therefore, a definitive conclusion requires the purification of the polypeptides responsible for these activities. PA,,-dependent membrane-associated PAPase In liver and intestine, little PA,,-dependent PAPase activity is associated with the membraneous fractions [36,17]. In adipose tissue [ 191 the specific activity of the microsomal activity towards PAmI, was three times as active as the soluble activity. However, the soluble fraction contains 8 times as much protein and therefore the cytosol contains a larger proportion of the total activity [ 191. A similar situation is also present in lung where the specific activity of the microsomal fraction with endogenous phosphatidate is greater than that of the cytosol activity using PA,,,, but the cytosol contains a higher proportion of the total activity [ 11. On the other hand, lung cytosol demonstrates a limited activity (with or without Mg*+) towards aqueously dispersed phosphatidate, whereas in adipose tissue, the cytosol possesses the major proportion of the Mg*‘-dependent activity with PA,, [ 191. Thus, although both lung and adipose tissue contain microsomal and supernatant activity towards membrane-bound as well as water dispersed phosphatidate, the relative expressions of these activities differ. As stated above, it has been suggested that when assayed in the appropriate manner, the Mg*+-dependent activities observed with aqueous phosphatidate in microsomes and cytosol from liver and adipose tissue represent the activities of these fractions towards membrane-bound substrate. In lung, the PA,,,-dependent PAPase activities demonstrate a marked (microsomes, 50%; cytosol, 100%) inhibition with EDTA and an EDTA-dependent Mg*+-stimulation, indicating some factor is common to these activities. On the other hand, the PA,,dependent activities of rat lung are only slightly though consistently inhibited by EDTA, although a small EDTA-dependent Mg*+ stimulation was observed at pH 6.5 (Fig. 5). In view of the results observed with other tissue, it

239

is tempting to speculate that the proportion of the PA,,-dependent activity which is Mg”-dependent and inhibited by EDTA corresponds to the PA,bdependent activities. The function of the remaining bulk of the PA,,-dependent activity is not known. Nevertheless, it is clear that studies in which PA,, is assayed in lung do not necessarily provide results reflecting the PAmI,-dependent activities or glycerolipid synthesis. Acknowledgements These investigations were supported by grants from the Medical Research Council of Canada and the Ontario Ministry of Health, The expert technical assistance of Dr. J. Stewart-DeHaan and Ms. G. Duwe with the subcellular experiments is greatly appreciated. References 1 Casola, P.G. and Possmayer, F. (1979) Biochim. Biophys. Acta 574, 212-225 2 Schultz. F.M., Jimenez, J.M., MacDonald. P.C. and Johnston, J.M. (1974) 3 4 5 6 7 8

G~necol.

Invest.

5, 222 -

229 P.J.. Wang. C., Las Heras. J. and Harding, Possmayer, F., Duwe, G., MetcaIfe. R., Stewart-DeHaan. P.G.R. (1977) Biochem. J. 166.485-494 Brehier, A.. Benson, B.J., WilIiams, M.C., Mason, R.J. and Ballard. P.L. (1977) Biochem. Biophys. Res. Commun. 77.883-890 Meban. C. (1972) J. CeII Biol. 53, 249-252 Ravinuthala, H.R., Miller, J.C. and Weinhold. P.A. (1978) Biochim. Biophys. Acta 530, 347-356 Mavis, R.D., Finkelstein, J.N. and Hall, B.P. (1978) J. Lipid Res. 19.467-477 Johnston, J.M., Reynolds, G., Wylie, M.B. and MacDonald, P.C. (1978) Biochim. Biophys. Acta 531,

65-71 Possmayer, F., Duwe. G.. Hahn, M. and Buchenea. D. (1977) Can. J. Biochem. 55,609-617 BIigh, E.G. and Dyer, W.J. (1959) Can. J. Biochem. Physiol. 37,911-917 Hendry, A.T. and Possmayer, F. (1974) Biochim. Biophys. Acta 369.156-172 HaIlman, M. and Gluck. L. (1974) Biochem. Biophys. Res. Commun. 60, l-7 Rooney, S.A., Page-Roberts, B.A. and Motoyama, E.K. (1975) J. Lipid Res. 16,418-425 Baranska. J. and van Golde, L.M.G. (1977) Biochim. Biophys. Acta 488.285-293 Caras. I. and Shapiro, B. (1975) Biochim. Biophys. Acta 409. 201-211 Sedgwick. B. and Hiibscher, G. (1965) Biochim. Biophys. Acta 106,63-77 Johnston, J.M., Rao, G.A., Lowe. P.A. and Schwarz, B.E. (1967) Lipids 2, 14-20 Hokin, L.E.. Hokin, M.R. and Mathison. D. (1963) Biochim. Biophys. Acta 144.397-408 Jarndar. S.C. and FaIIon. H.J. (1973) J. Lipid Res. 14, 517-524 Kent, C. and Vagelos, P.R. (1976) Biochim. Biophys. Acta 436. 377-386 W&am, G.F. and Kennedy, E.P. (1963) J. Biol. Chem. 238,2615-2619 Stoffel, W. and Schiefer, H.-G. (1968) HoppeSeyler’s Z. Physiol. Chem. 349,1017-1026 Coleman, R. (1968) Biochim. Biophys. Acta 163. 111-113 Goerke, J. (1974) Biochim. Biophys. Acta 344. 241-261 Heath, M.F.. Gandy. G. and Jacobson, W. (1976) in Lysosomes in Biology and Pathology (Dingle, J.T. and Dean, R.T., eds.). Vol. 5. pp. 34-55. North-Holland Publ. Co., Amsterdam 26 DiAugustine, R.P. (1974) J. Biol. Chem. 249,584-593 27 Gordon, A.H. (1973) in Lysosomes in Biology and Pathology (DingIe, J.T., ed.). Vol. 3. PP. 102-137, North-Holland Publ. Co., Amsterdam 28 Jimenez, J.M. and Johnston. J.M. (1976) Paediat. Res. 10.767-769 29 Ballard. P.L., Benson. J. and Brehier, A. (1977) Am. Rev. Resp. Dis. 115, 529-536 30 BleasdaIe. J.E.. Davis, C.-S. and Agranoff, B.W. (1978) Biochim. Biophys. Acta 528, 331-343 31 Spitzer, H.L., Rice. J.M., MacDonald, P.C. and Johnston, J.M. (1975) Biochem. Biophys. Res. Commun. 66.17-23 32 Lamb, R.G. and FalIon, H.J. (1974) Biochhn. Biophys. Acta 348.166-178 33 Hosaka. K.. Yamashita. S. and Numa. S. (1975) J. Biochem. 77, 501-509 34 MoIIer, F., Green, P. and Harkness, E.J. (1977) Biochim. Biophys. Acta 486, 359-368 35 Sturton. R.G. and Brindley. D.N. (1978) Biochem. J. 171. 263-266 36 Mitchell, M.P., Brindley, D.N. and Hiibscher. G. (1971) Eur. J. Biochem. 18,214-220 37 Yamashita, S. and Numa, S. (1972) Eur. J. Biochem. 31.565-573 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25