90
BBALIP
Biochrmlca et Bi&wca
Acta. 1047 (1990) 90-98 Elsevier
53508
Neutral cholesteryl ester hydrolase in the rat lactating mammary gland: regulation by phosphorylation-dephosphorylation Maria Jo& Martinez
‘.* and Kathleen
M. Botham ’
’ Department of Veterinary Basic Scrences, The Royal Veterinary College, London (U.K.) and ’ Department of Physiology University of the Basque Country Medical School, Btlbao (Spain)
(Revised
Key words:
Neutral
(Received 15 March 1990) manuscript received 7 June 1990)
cholesteryl ester hydrolase; Hormonal Phosphorylation-dephosphorylation;
control; Enzyme regulation; (Rat mammary gland)
Cholesterol
metabolism:
The characteristics of neutral cholesteryl ester hydrolase activities found in the microsomal and cytosolic subcellular fractions of rat lactating mammary tissue were investigated. The enzymes were assayed using cholesteryl oleate dispersed as a mixed micelle with phosphatidylcholine and sodium taurocholate (molar ratio 1: 4 : 2) as substrate. This method gave activities approx. 20-fold higher than those seen when cholesteryl oleate was added in ethanol. Addition of phosphatidylcholine and sodium taurocholate to the assays using the ethanol-dissolved substrate did not increase the activities observed. When the cholesteryl oleate was dispersed with phosphatidylcholine only (molar ratio, 1:4) the activity of the two neutral cholesteryl ester hydrolases was also decreased considerably compared to that found with mixed micelles. In this case, however, approx. 60% of the cytosolic, but only 10% of the microsomal activity, was restored by separate addition of sodium taurocholate. The activities of both the microsomal and the cytosolic neutral cholesteryl ester hydrolases were inhibited by MgCl 2, and this inhibition was almost completely reversed by the addition of an equimolar concentration of ATP. At a fixed concentration of MgCl, increasing concentrations of ATP increased the enzyme activities in a dose-dependent way. The activity of the microsomal, but not the cytosolic enzyme was enhanced by a cyclic AMP-dependent protein kinase and both activities were inhibited by alkaline phosphatase (bovine milk). These results provide evidence for the regulation of neutral cholesteryl ester hydrolases in the rat lactating mammary gland by mechanisms involving phosphorylation-dephosphorylation and therefore suggest that these enzymes may be under hormonal control.
Introduction The mammary gland secretes both esterified and unesterified cholesterol into milk during lactation. Some of this cholesterol arises from de novo synthesis within the gland (20-40%) [1,2] while the remainder is derived from the plasma lipoproteins (60-80%) [1,3]. In tissues other than the mammary gland, cholesteryl esters taken up from the plasma are hydrolyzed by an acid cholesteryl ester hydrolase situated in the lysosomes [4]. Intracellular stores of esterified cholesterol, however, are hydrolysed by cholesteryl ester hydrolases acting at or
Abbreviations: Mes, 4-morpholine ethanesulphonic [bis(2-hydroxyethyl)amino]-2-(hydroxymethyl)propane-l,3-diol. Correspondence: K.M. Botham, Department ences, The Royal Veterinary College. Royal NW1 OTU, U.K.
OOOS-2760/90/$03.50
acid;
Bistris.
2-
of Veterinary Basic SciCollege Street. London,
0 1990 Elsevier Science Publishers
B.V. (Biomedical
near neutral pH which have been found associated with the cytosolic and microsomal subcellular fractions [5-71. Mammary tissue is known to contain cholesterol stored as cholesteryl esters [8], and recent experiments in our laboratory have demonstrated that neutral cholesteryl ester hydrolase activity is present in microsomal and cytosolic fractions prepared from rat lactating mammary glands [9]. Tawil et al. [lo] have also found neutral cholesteryl ester hydrolase activity associated with microsomes from similar tissue. Neutral cholesteryl ester hydrolase in tissues such as the adrenal cortex [ll], the liver [12] and adipose tissue [13], where the enzyme is believed to have an important role in cholesterol metabolism, has been shown to be hormonally regulated by a mechanism involving phosphorylation-dephosphorylation. In addition, hormoneactivable neutral cholesteryl ester hydrolase activity has been demonstrated in arterial tissue and in macrophages, which are known to accumulate cholesteryl ester in certain pathological conditions [14,15]. Division)
91 In the present work, we have studied the characteristics of the neutral cholesteryl ester hydrolase activities associated with the microsomal and cytosolic subcellular fractions of rat lactating m~ary tissue [9], and have investigated the possibility that one or both of the enzymes is regulated by a phosphorylation-dephosphorylation mechanism. Materials and Methods Bovine serum albumin (essentially fatty acid free), phosphatidylcholine (type Xl-E), oleic acid, cholesteryl oleate, sodium taurocholate, 3’5’-cyclic AMP, 3’5’-cyclic AMP-dependent protein kinase (from rabbit muscle), ATP and alkaline phosphatase (type XXI from bovine milk) were obtained from Sigma {Poole, U.K.). [l“C]Oleic acid and cholesteryl [1-14C]oleate were supplied by Amersham International (Aylesbury, U.K.).
Animak Abdo~nal and inguinal mammary glands were excised from lactating Wistar rats 2-9 days post-partum. The glands were cleaned of fat and connective tissue, finely minced with scissors and homogenised in 5 vol. of Tris-HCl buffer (pH 7.2) (50 mM), containing 0.25 M sucrose, using a glass-Teflon homogeniser.
Preparation of subcellular fractions Microsomal and cytosolic subcellular fractions were prepared from the homogenate by differential centrifugation (800 X g, 10 min; 12 000 X g, 10 min; 105 000 X g, 60 min). The floating fat layer was carefully removed and discarded after each centrifugation. The 105 000 X g pellet containing the microsomes was washed once with Tris-HCl buffer (pH 7.2) (50 mM) containing 0.25 M sucrose and finally resuspended in Tris-HCl buffer (pH 7.0) (100 mM). The 105000 X g supernatant was used without further treatment and is referred to as the cytosolic fraction.
fragments from the sonicator horn. Cholesteryl oleate: phosphatidylcholine vesicles were prepared in a similar way except that sodium taurocholate was omitted. In experiments where substrate was added in ethanol, the method of Gorban and Boyd [16] was used. The activity of cholesteryl ester hydrolase in the cytosolic fraction was routinely assayed as follows: 150 yl Mes buffer (pH 6.6, 100 mM) containing bovine serum albumin (essentially fatty acid free, 7.5 mg) was added to 200 ~1 cytosolic fraction (0.2-0.4 mg cytosolic protein) and incubated in a shaking water bath at 37°C for 10 min. 50 ~1 of the micellar cholesteryl [l-‘4C]oleate (144 nmol cholesteryl oleate, 12.95 KBq/pmol) was then added to start the reaction. After 15 min incubation at 37°C the reaction was stopped with 1.5 ml chloroform/ methanol/ toluene (2 : 2.4 : 1, v/v) containing 0.29 mM oleic acid as a carrier. Sodium hydroxide (50 ~1, 1 M) was then added. The mixture was vortexed for 20 s and finally centrifuged at 1500 x g for 25 min to separate the phases. The amount of [l-‘4C]oleate released into the upper phase was determined by liquid scintillation counting. Blank tubes containing substrate but no enzyme protein were included in each experiment. The efficiency of extraction of [l-‘4C]oleate was also estimated in each experiment, using a mixed micelle prepared in the same way as the substrate, except that [l-‘4C]oleic acid was substituted for cholesteryl [l“C]oleate. Cholesteryl ester hydrolase in the microsomal fraction was assayed in a similar way, except that Tris-HCl (100 mM, pH 7.0) and microsomal protein (0.2-0.4 mg) was used. A different buffer was required, as Mes does not buffer efficiently above 6.7. Experiments using Bistris-HCl buffer for both enzymes gave similar results to those found with Mes and Tris-HCl. Mixed micelles were used as substrate in all experiments except where indicated otherwise. Proteins were determined by the method of Lowry et al. [17]. Significance limits were calculated using Student’s t-test.
Assay of cholesteryl ester hydrolase Cholesteryl ester hydrolase activity was determined by measuring the release of [l-‘4C]oleate from cholesteryl [l-14C]oleate. The substrate was prepared as a mixed micelle with phosphatidylcholine and sodium taurocholate in a molar ratio of 1 : 4 : 2 essentially as described by Hajjar et al. [14]. Cholesteryl [l-14C]oleate (185 kBq, 1.92 GBq/mmol) was added to a toluene solution containing phosphatidylcholine (57.6 pmol) and non-radioactive cholesteryl oleate (14.4 pmol) and the solvent was removed under a stream of nitrogen. The lipids were then resuspended in 5.0 ml Tris-HCl buffer (pH 7.0) (50 mM) containing sodium taurocholate (28.8 pmol) and sonicated using a 0.5 inch horn for a total time of 30 min at 46°C. The resulting solution was centrifuged at 1500 x g for 10 min to remove any metal
Results
Actioiry of m~crosoma~ and cyiosolic ~el~traI cho~este~~ ester hydrolases The effect of time of incubation on the hydrolysis of cholesteryl oleate by microsomal and cytosolic subcellular fractions from rat lactating mammary tissue is shown in Fig. 1 (A and B). For both enzymes the rate of reaction was linear for about 20 min, and reached a plateau after approx. 60 min. The rate of hydrolysis was also linear with the amount of protein added below 0.5 mg for the microsomes and 0.7 mg for the cytosol (Fig. 1C). Fig. 2 shows the effect of pH on the two enzymes. Although there was considerable variation in the values
92 The effect of mode of presentation salt on microsomal
of the substrate and bile
and cytosolic neutral cholesteryl ester
hydrolase activities
0
40
20
60
time (min)
protein (ma)
In our experiments maximum activity of both the microsomal and the cytosolic neutral cholesteryl ester hydrolase was obtained when the substrate was added as a mixed micelle with phosphatidylcholine and sodium taurocholate in a molar ratio of 1 : 4 : 2 [14]. When a similar concentration of substrate was added dissolved in ethanol as described by Gorban and Boyd [16] the observed activity of both enzymes was decreased by 95-98% (Table I). Furthermore, addition of phosphatidylcholine (1.44 mM) and sodium taurocholate (720 PM) did not increase these values significantly. When the substrate was presented as cholesteryl oleate : phosphatidylcholine vesicles (molar ratio 1 : 4) the activity of both the cytosolic and the microsomal enzyme was also decreased considerably. Separate addition of 720 PM sodium taurocholate to this type of assay led to an increase in the cytosolic activity to approx. 60% that found using mixed micelles, but the microsomal activity was only increased to about 11% of that value (Table I). The effect of sodium taurocholate on the enzyme activities was further investigated by separate addition of varying concentrations of the bile salt to assays using the cholesteryl oleate : phosphatidylcholine vesicles as substrate. The results are shown in Fig. 3. The activity of the cytosolic neutral cholesteryl ester hydrolase increased with sodium taurocholate concentration up to a maximum of 59% of the activity seen using mixed micelles at 720 PM, while the microsomal activity reached a maximum of only 12% of that value at a level of 1 mM sodium taurocholate.
Fig. 1. The effect of time and enzyme protein concentration on microsomal and cytosolic neutral cholesteryl ester hydrolase activities. The effect of time of incubation (A and B) and concentration of microsomal or cytosolic protein (C) on neutral cholesteryl ester hydrolase activity in microsomes (A) and cytosol (0) isolated from rat lactating mammary glands. Each point is the mean of duplicate determinations and the experiments shown are typical of several performed.
obtained the pH profiles were similar in all experiments with the highest activity occurring at 6.6 for the cytosolic and 7.0 for the microsomal enzyme in every case. Preincubation of the assay mixture for more than 10 min before addition of the substrate led to a progressive decrease in the activity of neutral cholesteryl ester hydrolase in the microsomes, so that after 30 min preincubation activity was approx. 55% that seen using a 10 min preincubation period. Activity of the cytosolic enzyme showed a similar decrease with time of preincubation. The activity of both enzymes was markedly inhibited by 1 M NaCl (W control value, microsomes 8.5 k 3.7 [3], cytosol 19.6 + 10.9 [4]).
Of 5.5
75
65
PH
Fig. 2. The effect of pH on microsomal and cytosolic neutral cholesteryl ester hydrolase activities. The effect of pH on neutral cholesteryl ester hydrolase (CEH) activity in microsomes (A) and cytosol (0) prepared from rat lactating mammary glands. The activities of the enzymes were assayed in Mes buffer (pH 5.5-6.8) or Tris-HCI buffer (pH 7.0-8.0) (100 mM for both buffers). Each point is the mean from three separate experiments. Error bars show the standard deviation.
93
cholesteryl oleate @M) 0 0.0
OS
1s
10
20
taurocholate (FM) Fig. 3. The effects of sodium taurocholate on microsomal and cytosolic neutral cholesteryl ester hydrolase activities. Neutral cholesteryl ester hydrolase (CEH) activity in microsomal (A) and cytosolic (0) subcellular fractions prepared from rat lactating mammary glands was assayed using cholesteryl [l-t4C]oleate : phophatidylcho~ne vesicles (molar ratio, 1: 4) as substrate. Sodium taurocholate was added to the assays separately prior to the preincubation period. Data are expressed as a percentage of the values obtained using mixed micelies as substrate. Each point is the mean from three (microsomes) or four (cytosol) experiments. Error bars show the standard deviation. Absolute control values were as given for Table I.
These results show that the cytosolic neutral cholesteryl ester hydrolase is activated by bile salt, but the increased activity found when the substrate is presented as a mixed micelle is due only partially to the presence of sodium taurocholate, with an additional 40% increase being caused by the physical dispersion of the substrate. The microsomal activity is activated by bile salt to a lesser extent, but in this case approx. 90% of the activity found with mixed micelles can be attributed to the physical dispersion of the cholesteryl oleate. The effect of varying the concentration of cholesteryl oleate in the assay mixture while rn~nt~~ng the molar
Fig. 4. The effect of substrate concentration on microsomal and cytosolic neutral cholesteryl ester hydrolase activities. The effect of varying concentrations of substrate (mixed micelles) on neutral cholesteryl ester hydrolase (CEH) activity in microsomal (A) and cytosolic (0) subcellular fractions isolated from rat lactating mammary glands. Each point is the mean from three (microsomes) or four (cytosol) experiments. Error bars show the standard deviation.
ratio of cholesteryl oleate : phosphatidylcholine : sodium taurocholate at 1 : 4: 2 in the mixed micelles on the activity of the microsomal and cytosolic neutral cholesteryl ester hydrolases is shown in Fig. 4. The cytosolic enzyme did not show typical Michaelis-Menton kinetics, possibly because of the effect of the bile salt present in the mixed micelle substrate. Alternatively, as it is known that increasing the amount of substrate dispersion in a heterogeneous reaction system increases the surface area containing substrate, not its surface concentration, the atypical kinetics may reflect a weak affinity of the enzyme for the mixed micelles. The effect of MgCl and A TP on microsomal and cytosolic neutral cholesteryl ester hydrolase activities
The activity of neutral cholesteryl ester hydrolase in the microsomes was decreased in the presence of MgCl, in a dose-dependent way (Fig. 5). When ATP was
TABLE I l%e effect of mode of presentation of the substrate on microsomal and cytosolic neutral cholesteryl ester hydrolme activities Microsomal and cytosolic fractions were prepared from rat lactating mammary tissue as described in Materials and Methods. Neutral cholesteryl ester hydrolase (CELL) activities were determined using substrate prepared as mixed micelles (cholesteryl oleate : phosphatidylcho~ne : sodium taurochotate, molar ratio 1:4: 2) or as cholesteryl oleate:phosphatidylcho~ne vesicles (molar ratio 1: 4) or added in ethanol [16]. Data are expressed as a percentage of the value obtained with mixed micelles (microsomes, 1216 i 287, cytosol 897 f 360 pmol/min per mg protein) and are given as the mean f S.D. (more than three experiments) or the range of values obtained (two experiments). Figures in brackets represent the number of animals used. Substrate
Separate additions
CEH activity (W mixed micelle value) microsomes
cytosol
1. Mixed micelles
none
100
100
2. Cholesteryl oleate: phosphatidylchohne 3. Cholesteryl oleate in ethanol
none sodium taurocholate (720 PM) none phosphatidylcholine (1.44 mM) phosphatidylchohne (1.44 mM) + sodium taurocholate (720 PM)
5.8 4 1.2 (3) 11.3+0.8 (3) 6.5 f 2.6 (2) 4.3 f 2.5 (2) 10.1* 5.2 (2)
15.9* 7.0 (4) 59.1 f 15.6 (4) 2.3 f 2.0 f
0.8 (2) 0.6 (2)
4.6zbO.l (2)
94 included in the incubation in a 1 : 1 molar ratio with MgC12, however, activity was restored almost to control levels. The neutral cholesteryl ester hydrolase activity in the cytosol was also decreased by MgCl,, but the effect was less marked than that seen with the microsomal enzyme (Fig. 5). Activity of this enzyme was also restored to values near control levels in the presence of equimolar concentrations of MgCl, and ATP. In the absence of MgCl,, ATP had no effect on the activity of either enzyme. When the ATP concentration was increased while maintaining a constant MgCl, (1 or 2 mM) level the activity of the microsomal neutral cholesteryl ester hydrolase increased linearly as a function of ATP concentration (Table II). The cytosolic activity was also increased with increasing ATP concentrations (with 2 or 5 mM MgCl,), but the effect again was less marked than that seen with the microsomal activity (Table II).
40 00
Fig. 5. The effect of MgCI, and ATP on microsomal and cytosolic neutral cholesteryl ester hydrolase activities. The effect of MgCl, (0, A), or ATP and MgCl, (0, A) in a 1 : 1 molar ratio on neutral cholesteryl hydrolase activity in cytosol (0, 0) and microsomes (A, A) prepared from rat lactating mammary glands. Data are given as a percentage of the value found in the absence of ATP or MgCl, (absolute values (pmol/min per mg protein); microsomes 869 it 228; cytosol 784i 167). Each point is the mean from four experiments. Error bars show the standard deviation.
TABLE The
II
effect
cholesteryi
of MgCl,
and A TP on microsomal
ester hydrolase
and ~viosohc
neutrul
uctiuirm
Microsomal and cytosolic fractions from rat lactating mammary glands were prepared as described in Materials and Methods. Neutral cholesteryl ester hydrolase activity was determined in the presence of the MgCl, and ATP concentrations shown. Data are expressed as a percentage of the values obtained in the presence of the each concentration of MgCl, without ATP, and are given as the meankS.D. Figures in brackets show the number of animals used. nd., not determined. Absolute values in the absence of ATP (pmol/min per mg protein): microsomes. 1 mM MgCl, 550+ 80, 2 mM MgCl z 352+113; cytosol, 2 mM MgCl, 731k163, 5 mM MgCl, 574+103.
W&l,
ATP
CEH activity
(mM)
(mM)
microsomes
1 1 1 2 2 2 2 2 5 5 5 5
0 5 0 1 2 5 10 0 2 5 10
100 118.0*11.5 182.1 If- 4.9 100 120.4+ 5.1 135.0* 13.7 183.8 f 8.4 238.5 f 19.X n.d. n.d. n.d. nd.
R cytosol
(4) (4) (3) (4) (3) (3)
n.d. nd. 100 n.d. 118.4+3.0 129.4k3.5 130.7k8.2 100 110.3 + 3.5 126.2 f 7.1 122.3+4.2
(4) (4) (3) (4) (4) (4)
The effect of isolation of subcellular fractions in the presence of NaF and/or EDTA on microsomal neutral cholesteryl ester hydrolase activity If the neutral cholesteryl ester hydrolases under study were regulated by a mechanism involving phosphorylation-dephosphorylation, then their activities might be modified during the preparation of the subcellular fractions by endogenous protein kinases and/or phosphatases. This possibility was tested for the microsomal enzyme by preparing the fractions in the presence of 50 mM NaF and/or 10 mM EDTA, which are known to inhibit phosphatases and protein kinases, respectively [18,19]. Preparation of the microsomal fractions in the presence of 10 mM EDTA led to decrease of about 50% (49.5 f 17.5%) in the activity of the microsomal neutral cholesteryl ester hydrolase, while inclusion of 50 mM NaF during the isolation procedure did not cause any significant change (87.0 _t 12.0%). The results obtained when both NaF and EDTA were present were similar to those found with EDTA (51.2 f 18.7%). The absolute value found when the microsomes were prepared in the absence of NaF or EDTA was 938 f 114 pmol/min per mg protein. The effects of MgCl, and ATP with MgCl, on the activity of neutral cholesteryl ester hydrolase in microsomes prepared in the presence of NaF and/or EDTA and resuspended in buffer without these inhibitors are shown in Table III. Data are shown as a percentage of the value obtained when the subcellular fractions were
95 TABLE
III
The effect of MgCl,
and ATP on neutral cholestetyl ester hydrolase activity in microsomal fractions isolated in the presence of EDTA and/or
NaF
Microsomal fractions from rat lactating mammary tissue were prepared for assay as described in Materials and Methods. Neutral cholesteryl hydrolase (CEH) activity was determined in the presence of MgCl, and MgCl 2 with ATP. Data are expressed as a percentage of the value obtained with microsomes isolated in the absence of EDTA or NaF and assayed without MgCl, or ATP (absolute value, 892+133 pmol/min per mg protein). Each value is the mean from four separate experiments+ S.D. Additions
None MgCl, MgCl,
Neutral
to assays
(1 mM) (1 mM) & (ATP 5 mM)
Significance
limits: MgCl,
vs. no additions;
CEH activity
(% control
value)
control
+ EDTA
+ NAF
+ EDTA and NaF
100 77.5 f 4.4 146.8 +43.3 b
51.3* 9.2 49.8 + 10.0 101.7 + 31.8 b
82.3i 8.0 67.7i 9.0 a 150.9 * 41.5 c
61.7+ 5.8 59.1+ 1.1 105.1+ 11.4 c
a P < O.O5;MgCI,
+ATP
vs. MgCI,,
prepared in the absence of inhibitors and assayed without M&l, or ATP (control value). The activity of the microsomal enzyme was significantly inhibited by MgCl, in microsomes prepared without protein kinase or phosphatase inhibitors or with NaF, but not in those isolated in the presence of EDTA or EDTA and NaF, suggesting that inactivation has occurred during the preparation under the latter conditions. When ATP was present in addition to MgCl, neutral cholesteryl ester hydrolase activity was significantly increased compared to that found with MgCl, only in all cases. This type of experiment was not carried out on the cytosolic neutral cholesteryl ester hydrolase, as it was not possible to remove the EDTA or NaF completely from the cytosol before assay of the enzyme.
TABLE
b P < 0.025; ‘P < 0.01.
Reversible inactivation-activation of microsomal neutral cholesteryl ester hydrolase In order to test whether the observed effects of MgCl, and ATP on the microsomal neutral cholesteryl ester hydrolase were reversible microsomes were incubated with MgCl, (5 mM) for 10 min (stage l), ATP (5 mM) was then added and the mixture incubated for a further 10 min (stage 2). Aliquots were assayed for neutral cholesteryl ester hydrolase activity after stages 1 and 2. ATP and MgCl, were then removed by re-isolation of the microsomes by centrifugation at 105 000 x g (60 min). The pellet was washed once with Tris-HCl buffer (pH 7.0) (100 mM) and resuspended in similar buffer. The incubations and assays with MgCl, and ATP were then repeated (stages 3 and 4). The activity of
IV
The effect of protein kinase and alkaline phosphatase on neutral cholestetyl ester hydrolase activities Microsomal and cytosolic fractions from rat lactating ester hydrolase (CEH) activity was determined in the 1 mM for the microsomes and 2 mM for the cytosol. (pmol/min per mg protein): microsomes, 912 f 108; animals used. n.s., not significant. Additions
Neutral
1 none 2 MgCl, (1 or 2 mM) 3 MgCl, (1 or 2 mM) + ATP (5 mM) 4 CAMP (50 PM) + protein kinase (50 pg) 5 2+4 63+4 7 Alkaline phosphatase: 10 pg 20 Pg 50 pg 100 I-18 Significance 3 vs. 2 6 vs. 3 6 vs. 5
mammary glands were presence or absence of Data are expressed as cytosol 1121+ 274) and
prepared as described in Materials and Methods. Neutral cholesteryl the factors shown. The concentration of MgCl, used in each case was a percentage of the value obtained without additions (absolute values are given as the mean + S.D. Figures in brackets show the number of
CEH activity
(W control
value)
microsomes
cytosol
100 78.1 f 96.1+ 105.9 f 84.3 + 125.6+ 67.3 + 43.8 f 13.2 f 4.4 f
100 79.6 f 89.7 f 96.1 f 79.2 f 93.6 + 86.8 f 49.8 * 27.0* 17.7 f
5.8(5) 10.7 (5) 4.5 (5) 2.6 (4) 11.6 (5) 35.4 (3) 30.0 (3) 3.4 (3) 3.9 (4)
4.2(4) 6.0 (4) 7.8 (4) 15.0 (3) 3.9 (4) 3.1 (4) 7.9 (4) 17.2 (3) 14.2 (3)
limits P i 0.01 P i 0.005 P < 0.0005
P < 0.025 n.s. ns.
96 the enzyme in control microsomes incubated for a sirnilar time period in the absence of ATP and/or MgCl, was determined at each stage. Data are given as percentages of the control values found at each stage (absolute value at stage 1, 765 k 96 pmol/min per mg protein). The control value at stage 4 was 80% that found for the control at stage 1. Neutral cholesteryl ester hydrolase activity was decreased at stage 1 by MgCl, (77.6 f 12.1%), but this was reversed at stage 2 after the addition of ATP (112.4 + 4.7%). After removal of ATP and MgCl, by re-isolation of the microsomes the activity of the enzyme in control assays was similar to that found in stage 1 controls. Activity was decreased by MgCl, at stage 3 (70.6 k 22.8%) and this inactivation was reversed again by ATP at stage 4 (153.8 + 39.4%). The effect of protein kinase and alkaline phosphatase on neutral cholesteryl ester hydrolase activities The effect of a cyclic AMP-dependent protein kinase on microsomal and cytosolic neutral cholesteryl ester hydrolase activity is shown in Table IV. When microsomes were incubated with cyclic AMP and cyclic AMP-dependent protein kinase the activity of the enzyme was not significantly different from that found in control assays. In addition, the inhibition of neutral cholesteryl ester hydrolase activity by MgCl, was not prevented by cyclic AMP and the protein kinase. In the presence of ATP, MgCl,, cyclic AMP and the protein kinase, however, the activity was significantly higher than that found with MgCl, and ATP alone (P < 0.005). In contrast, the activity of the cytosolic neutral cholesteryl ester hydrolase found in the presence of cyclic AMP-dependent protein kinase, cyclic AMP, ATP and MgCl, was not significantly higher than that seen with ATP and MgCl, alone. Alkaline phosphatase (bovine milk) inhibited both the microsomal and the cytosolic neutral cholesteryl ester hydrolase activities in a dose-dependent way (Table IV). Discussion The results reported here, together with our previous work [9] demonstrate that neutral cholesteryl ester hydrolase activity is associated with the microsomal and cytosolic subcellular fractions isolated from rat lactating mammary tissue. The activities in the two fractions appear to have different pH optima and also differ in their response to bile salt (Table I, Fig. 3) and to certain conditions favouring phosphorylation or dephosphorylation (Table IV), indicating that they are distinct enzymes. Tawil et al. [lo] have also reported the presence of neutral cholesteryl ester hydrolase activity in rat lactating mammary glands. In their experiments, however, the greater proportion of the total activity was
associated with the microsomal fraction (60%) while we found the highest percentage in the cytosol (60%) [9]. In addition, the pH optimum of the main microsomal activity was 7.5-8.5 as opposed to 7.0 in our work. It is possible that some of these differences may be explained by the mode of presentation of substrate used in our laboratory. It has been reported [20] that the pH optimum of cholesteryl ester hydrolase in homogenates of aorta tissue differs depending on the physical dispersion of the substrate. In our experiments addition of the cholesteryl oleate as a mixed micelle with phosphatidylcholine and sodium taurocholate gave activities 10-20fold higher than those obtained when the substrate was added in ethanol as in the experiments of Tawil et al. [lo], and furthermore, the increase in activity was not reproduced by adding the appropriate amounts of phosphatidylcholine and sodium taurocholate to the assays using the latter method (Table I), suggesting that the physical dispersion of the substrate plays a large part in determining the observed activity of the neutral cholesteryl ester hydrolases. Omission of sodium taurocholate from the dispersion, so that the substrate consisted of cholesteryl oleate : phosphatidylcholine vesicles led to a very marked decrease in the activity of both enzymes, but in this case a substantial part of the cytosolic activity could be restored by separate addition of the bile salt. In contrast, sodium taurocholate was not able to restore the microsomal activity above 12% of that found with the micellar substrate (Table I, Fig. 3). Bile salt-dependent cholesteryl ester hydrolases have been described in rat liver cytosol [21,22] and pancreas [23,24], and there is some evidence to suggest that the bile salt acts to promote changes between monomeric and polymeric forms of the enzyme [24,25]. However, as the mammary gland activities are unlikely to be exposed to millimolar quantities of bile salt in vivo it seems more likely that the stimulatory effect on the cytosolic enzyme is caused by the detergent properties of taurocholate stabilising the enzyme at the lipid/water interface, as suggested by Tsujitsa et al. [26]. Both the microsomal and the cytosolic enzymes were inhibited by about 80-90% by 1 M sodium chloride, in agreement with the results of Tawil et al. [lo]. Neutral cholesteryl ester hydrolase activity in a number of other tissues, including the liver, adipose tissue, adrenal cortex, arterial tissue and macrophages [6,11- 151 has been shown to be activated by conditions favouring phosphorylation of the enzyme, and inactivated by those favouring dephosphorylation. In our experiments, the microsomal neutral cholesteryl ester hydrolase activity from rat lactating mammary glands also seemed to follow this pattern. The enzyme was deactivated in the presence of MgCl, (Fig. 5) which is known to activate phosphatases, and therefore favour dephosphorylation, but activity was almost fully restored to control levels
97 when an equimolar concentration of ATP was included. This effect could be explained by the chelation effect of ATP on the Mg2+ ions, however, at fixed concentrations of MgCl, the enzyme activity was increased by ATP in a dose-dependent way (Table II) and at the highest concentration used (10 mM) activities considerably above those found in the absence of MgCl, (150200%) were seen. As neither ATP (l-10 mM) alone, nor EDTA (2-5 mM) had any effect on the microsomal cholesteryl ester hydrolase, these increases cannot be accounted for by the chelation of endogenous Mg2+. In addition, the activity of the enzyme found with MgCl, and ATP was enhanced when cyclic AMP and a cyclic AMP-dependent protein kinase were included in the assay. The activity was also inhibited by alkaline phosphatase from bovine milk (Table IV), but a similar enzyme from calf intestine had no effect, suggesting that there is some specificity for the type of alkaline phosphate required for inhibition. The loss of activity (56687%, Table Iv) seen with 20-50 pg bovine milk alkaline phosphatase (equivalent to 0.06-0.15 units) in our experiments is comparable to that found in studies demonstrating the regulation of rat liver cholesteryl ester hydrolase by phosphorylation-dephosphorylation, in which 0.1 units of bovine liver alkaline phosphatase was shown to cause 69% loss of activity [12]. We cannot rule out the possibility, however, that the effect of bovine milk alkaline phosphase is caused by binding to the microsomes or the cholesteryl ester hydrolase enzyme. Taken together our results suggest that the microsomal neutral cholesteryl ester hydrolase is activated when phosphorylated and inactivated in the dephosphorylated state. The decrease in the activity of this enzyme found when microsomes were isolated under conditions where phosphorylation by endogenous protein kinases was inhibited with 10 mM EDTA during preparation [19] also supports this conclusion. The reversible nature of the phosphorylation-dephosphorylation is demonstrated by the restoration of activity of enzyme inactivated by MgCl, by subsequent addition of ATP, and by the repetition of this cycle after re-isolation of the microsomes. Our results, therefore, provide evidence that the microsomal neutral cholesteryl ester hydrolase in rat lactating mammary tissue may be regulated by hormone action involving a cyclic AMP-dependent protein kinase. The hormone-sensitive neutral cholesteryl ester hydrolases described in other tissues [6,11-151 are all associated with the soluble fraction of the cell. Our experiments provide some evidence that the mammary gland cytosolic neutral cholesteryl ester hydrolase may also be regulated by a phosphorylation-dephosphorylation mechanism. The activity was inhibited to some extent by MgCl, and this was reversed by the addition of ATP (Fig. 5). Increasing concentrations of ATP at fixed MgCl, concentrations also led to increased activ-
ity (Table II). The enzyme was also inhibited by alkaline phosphatase from bovine milk (Table IV). These results suggest that the cytosolic neutral cholesteryl ester hydrolase in the mammary gland is active when phosphorylated and inactive when dephosphorylated. Involvement of a cyclic AMP-dependent protein kinase, however, could not be demonstrated (Table IV). The neutral cholesteryl ester hydrolase activities that have been shown previously to be regulated by mechanisms involving phosphorylation-dephosphorylation are present in tissues which have a significant role in cholesterol metabolism, such as the liver [12], adrenal cortex [ll] and adipose tissue [6] or in those which can accumulate cholesteryl ester under pathological conditions, such as arterial tissue [14] and macrophages [15]. Our finding that rat lactating mammary tissue contains microsomal and cytosolic neutral cholesteryl ester hydrolases which may be regulated by this type of mechanism raises the possibility that these enzymes play an important part in the regulation of cholesterol metabolism in the mammary gland, and in the provision of cholesterol for secretion into milk. Acknowledgements This work was supported by a grant from The Wellcome Trust. Dr Maria Jose Martinez received a grant from the University of the Basque Country. We should like to thank Miss Adrienne Carroll for expert technical assistance. References 1 Clarenburg, R. and Chaikoff, I.L. (1966) J. Lipid Res. 7, 27-37. 2 Gibbons, G.F., Pullinger, C.R., Munday, M.R. and Williamson, D.H. (1983) Biochem J. 212, 843-848. 3 Raphael, B.C., Patten, S. and McCarthy, R.D. (1975) FEBS Lett. 58, 47-49. 4 Goldstein, J.L., Dana, S.E., Faust, J.R., Beaudet, A.L. and Brown, M.S. (1975) J. Biol. Chem. 250, 8487-8495. 5 Boyd, G.S. and Gorban, A.M.S. (1980) in Newly Discovered Systems of Enzyme Regulation by Reversible Phosphorylation (Cohen, P., ed.), Vol. 1, pp. 95-134, Elsevier/North-Holland, Amsterdam. 6 Steinberg, D. (1976) Adv. Cyclic Nucleotide Res. 7, 157-198. 7 Femandez, C.J., Lacort, M., Gandarias, J.M. and Ochoa, B. (1987) Biochem. Biophys. Res. Commun. 146, 1212-1217. 8 Keenan, T.W. and Patten, S. (1970) Lipids 5, 42-48. 9 Martinez, M.J. and Botham, K.M. (1990) B&hem. Sot. Trans. 18, 619-620. 10 Tawil, S.G., Shand, J.H. and West, D.W. (1989) B&hem. Sot. Trans. 17, 653-654. 11 Trzeciak, W.H. and Boyd, G.S. (1974) Eur. J. Biochem. 46, 201207. 12 Ghosh, S. and Grogan, W.M. (1989) Lipids 24, 733-736. 13 Pittman, R.C., Khoo, J.C. and Steinberg, D. (1975) J. Biol. Chem. 250, 4505-4511. 14 Hajjar, D.P., Minick, R. and Fowler, S. (1983) J. Biol. Chem. 258, 192-198. 15 Khoo, J.C., Mahoney, E.M. and Steinberg, D. (1981) J. Biol. Chem. 256, 12659-12661.
98 16 Gorban, A.M.S. and Boyd, G.S. (1977) FEBS Lett. 17 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and (1951) J. Biol. Chem. 193, 265-275. 18 Erickson, S.K., Shrewsbury, M.A., Gould, R.G. and (1980) Biochim. Biophys. Acta 620, 70-79. 19 Nordstrom, J.L., Rodwell, V.W. and Mitschelen, J.J.
79, 54-58. Randall, R.J. Cooper,
A.O.
(1977) J. Biol.
Chem. 252, 8924-8928. 20 Kothri, H.V., Bonner, M.J. and Miller, B.F. (1970) Biochim. Biophys. Acta 202, 325-331. 21 Harrison, E.H. (1988) Biochim. Biophys. Acta. 963, 28-34.
22 Camulli, E.D., Linke, M.J., Brockman, H.L. and Hui, D.Y. (1989) Biochim. Biophys. Acta. 1005, 177-182. 23 Vahouny, G.V., Weersing, S. and Treadwell, C.R. (1965) Biochim. Biophys. Acta. 98, 607-616. 24 Hyun. J., Steinberg, M., Treadwell, CR. and Vahouny. G.V. (1971) Biochem. Biophys. Res. Commun. 44. 819-825. 25 Tuhackova. I., Kriz, 0. and Hradec, J. (1980) Biochim. Biophys. Acta. 617, 439-445. 26 Tsujitsa, T., Mizuno, N.K. and Brockman, H.L. (1987) J. Lipid Res. 28, 1434-1443.
BBALIP
Biochrmlca et Bi&wca
Acta. 1047 (1990) 90-98 Elsevier
53508
Neutral cholesteryl ester hydrolase in the rat lactating mammary gland: regulation by phosphorylation-dephosphorylation Maria Jo& Martinez
‘.* and Kathleen
M. Botham ’
’ Department of Veterinary Basic Scrences, The Royal Veterinary College, London (U.K.) and ’ Department of Physiology University of the Basque Country Medical School, Btlbao (Spain)
(Revised
Key words:
Neutral
(Received 15 March 1990) manuscript received 7 June 1990)
cholesteryl ester hydrolase; Hormonal Phosphorylation-dephosphorylation;
control; Enzyme regulation; (Rat mammary gland)
Cholesterol
metabolism:
The characteristics of neutral cholesteryl ester hydrolase activities found in the microsomal and cytosolic subcellular fractions of rat lactating mammary tissue were investigated. The enzymes were assayed using cholesteryl oleate dispersed as a mixed micelle with phosphatidylcholine and sodium taurocholate (molar ratio 1: 4 : 2) as substrate. This method gave activities approx. 20-fold higher than those seen when cholesteryl oleate was added in ethanol. Addition of phosphatidylcholine and sodium taurocholate to the assays using the ethanol-dissolved substrate did not increase the activities observed. When the cholesteryl oleate was dispersed with phosphatidylcholine only (molar ratio, 1:4) the activity of the two neutral cholesteryl ester hydrolases was also decreased considerably compared to that found with mixed micelles. In this case, however, approx. 60% of the cytosolic, but only 10% of the microsomal activity, was restored by separate addition of sodium taurocholate. The activities of both the microsomal and the cytosolic neutral cholesteryl ester hydrolases were inhibited by MgCl 2, and this inhibition was almost completely reversed by the addition of an equimolar concentration of ATP. At a fixed concentration of MgCl, increasing concentrations of ATP increased the enzyme activities in a dose-dependent way. The activity of the microsomal, but not the cytosolic enzyme was enhanced by a cyclic AMP-dependent protein kinase and both activities were inhibited by alkaline phosphatase (bovine milk). These results provide evidence for the regulation of neutral cholesteryl ester hydrolases in the rat lactating mammary gland by mechanisms involving phosphorylation-dephosphorylation and therefore suggest that these enzymes may be under hormonal control.
Introduction The mammary gland secretes both esterified and unesterified cholesterol into milk during lactation. Some of this cholesterol arises from de novo synthesis within the gland (20-40%) [1,2] while the remainder is derived from the plasma lipoproteins (60-80%) [1,3]. In tissues other than the mammary gland, cholesteryl esters taken up from the plasma are hydrolyzed by an acid cholesteryl ester hydrolase situated in the lysosomes [4]. Intracellular stores of esterified cholesterol, however, are hydrolysed by cholesteryl ester hydrolases acting at or
Abbreviations: Mes, 4-morpholine ethanesulphonic [bis(2-hydroxyethyl)amino]-2-(hydroxymethyl)propane-l,3-diol. Correspondence: K.M. Botham, Department ences, The Royal Veterinary College. Royal NW1 OTU, U.K.
OOOS-2760/90/$03.50
acid;
Bistris.
2-
of Veterinary Basic SciCollege Street. London,
0 1990 Elsevier Science Publishers
B.V. (Biomedical
near neutral pH which have been found associated with the cytosolic and microsomal subcellular fractions [5-71. Mammary tissue is known to contain cholesterol stored as cholesteryl esters [8], and recent experiments in our laboratory have demonstrated that neutral cholesteryl ester hydrolase activity is present in microsomal and cytosolic fractions prepared from rat lactating mammary glands [9]. Tawil et al. [lo] have also found neutral cholesteryl ester hydrolase activity associated with microsomes from similar tissue. Neutral cholesteryl ester hydrolase in tissues such as the adrenal cortex [ll], the liver [12] and adipose tissue [13], where the enzyme is believed to have an important role in cholesterol metabolism, has been shown to be hormonally regulated by a mechanism involving phosphorylation-dephosphorylation. In addition, hormoneactivable neutral cholesteryl ester hydrolase activity has been demonstrated in arterial tissue and in macrophages, which are known to accumulate cholesteryl ester in certain pathological conditions [14,15]. Division)
91 In the present work, we have studied the characteristics of the neutral cholesteryl ester hydrolase activities associated with the microsomal and cytosolic subcellular fractions of rat lactating m~ary tissue [9], and have investigated the possibility that one or both of the enzymes is regulated by a phosphorylation-dephosphorylation mechanism. Materials and Methods Bovine serum albumin (essentially fatty acid free), phosphatidylcholine (type Xl-E), oleic acid, cholesteryl oleate, sodium taurocholate, 3’5’-cyclic AMP, 3’5’-cyclic AMP-dependent protein kinase (from rabbit muscle), ATP and alkaline phosphatase (type XXI from bovine milk) were obtained from Sigma {Poole, U.K.). [l“C]Oleic acid and cholesteryl [1-14C]oleate were supplied by Amersham International (Aylesbury, U.K.).
Animak Abdo~nal and inguinal mammary glands were excised from lactating Wistar rats 2-9 days post-partum. The glands were cleaned of fat and connective tissue, finely minced with scissors and homogenised in 5 vol. of Tris-HCl buffer (pH 7.2) (50 mM), containing 0.25 M sucrose, using a glass-Teflon homogeniser.
Preparation of subcellular fractions Microsomal and cytosolic subcellular fractions were prepared from the homogenate by differential centrifugation (800 X g, 10 min; 12 000 X g, 10 min; 105 000 X g, 60 min). The floating fat layer was carefully removed and discarded after each centrifugation. The 105 000 X g pellet containing the microsomes was washed once with Tris-HCl buffer (pH 7.2) (50 mM) containing 0.25 M sucrose and finally resuspended in Tris-HCl buffer (pH 7.0) (100 mM). The 105000 X g supernatant was used without further treatment and is referred to as the cytosolic fraction.
fragments from the sonicator horn. Cholesteryl oleate: phosphatidylcholine vesicles were prepared in a similar way except that sodium taurocholate was omitted. In experiments where substrate was added in ethanol, the method of Gorban and Boyd [16] was used. The activity of cholesteryl ester hydrolase in the cytosolic fraction was routinely assayed as follows: 150 yl Mes buffer (pH 6.6, 100 mM) containing bovine serum albumin (essentially fatty acid free, 7.5 mg) was added to 200 ~1 cytosolic fraction (0.2-0.4 mg cytosolic protein) and incubated in a shaking water bath at 37°C for 10 min. 50 ~1 of the micellar cholesteryl [l-‘4C]oleate (144 nmol cholesteryl oleate, 12.95 KBq/pmol) was then added to start the reaction. After 15 min incubation at 37°C the reaction was stopped with 1.5 ml chloroform/ methanol/ toluene (2 : 2.4 : 1, v/v) containing 0.29 mM oleic acid as a carrier. Sodium hydroxide (50 ~1, 1 M) was then added. The mixture was vortexed for 20 s and finally centrifuged at 1500 x g for 25 min to separate the phases. The amount of [l-‘4C]oleate released into the upper phase was determined by liquid scintillation counting. Blank tubes containing substrate but no enzyme protein were included in each experiment. The efficiency of extraction of [l-‘4C]oleate was also estimated in each experiment, using a mixed micelle prepared in the same way as the substrate, except that [l-‘4C]oleic acid was substituted for cholesteryl [l“C]oleate. Cholesteryl ester hydrolase in the microsomal fraction was assayed in a similar way, except that Tris-HCl (100 mM, pH 7.0) and microsomal protein (0.2-0.4 mg) was used. A different buffer was required, as Mes does not buffer efficiently above 6.7. Experiments using Bistris-HCl buffer for both enzymes gave similar results to those found with Mes and Tris-HCl. Mixed micelles were used as substrate in all experiments except where indicated otherwise. Proteins were determined by the method of Lowry et al. [17]. Significance limits were calculated using Student’s t-test.
Assay of cholesteryl ester hydrolase Cholesteryl ester hydrolase activity was determined by measuring the release of [l-‘4C]oleate from cholesteryl [l-14C]oleate. The substrate was prepared as a mixed micelle with phosphatidylcholine and sodium taurocholate in a molar ratio of 1 : 4 : 2 essentially as described by Hajjar et al. [14]. Cholesteryl [l-14C]oleate (185 kBq, 1.92 GBq/mmol) was added to a toluene solution containing phosphatidylcholine (57.6 pmol) and non-radioactive cholesteryl oleate (14.4 pmol) and the solvent was removed under a stream of nitrogen. The lipids were then resuspended in 5.0 ml Tris-HCl buffer (pH 7.0) (50 mM) containing sodium taurocholate (28.8 pmol) and sonicated using a 0.5 inch horn for a total time of 30 min at 46°C. The resulting solution was centrifuged at 1500 x g for 10 min to remove any metal
Results
Actioiry of m~crosoma~ and cyiosolic ~el~traI cho~este~~ ester hydrolases The effect of time of incubation on the hydrolysis of cholesteryl oleate by microsomal and cytosolic subcellular fractions from rat lactating mammary tissue is shown in Fig. 1 (A and B). For both enzymes the rate of reaction was linear for about 20 min, and reached a plateau after approx. 60 min. The rate of hydrolysis was also linear with the amount of protein added below 0.5 mg for the microsomes and 0.7 mg for the cytosol (Fig. 1C). Fig. 2 shows the effect of pH on the two enzymes. Although there was considerable variation in the values
92 The effect of mode of presentation salt on microsomal
of the substrate and bile
and cytosolic neutral cholesteryl ester
hydrolase activities
0
40
20
60
time (min)
protein (ma)
In our experiments maximum activity of both the microsomal and the cytosolic neutral cholesteryl ester hydrolase was obtained when the substrate was added as a mixed micelle with phosphatidylcholine and sodium taurocholate in a molar ratio of 1 : 4 : 2 [14]. When a similar concentration of substrate was added dissolved in ethanol as described by Gorban and Boyd [16] the observed activity of both enzymes was decreased by 95-98% (Table I). Furthermore, addition of phosphatidylcholine (1.44 mM) and sodium taurocholate (720 PM) did not increase these values significantly. When the substrate was presented as cholesteryl oleate : phosphatidylcholine vesicles (molar ratio 1 : 4) the activity of both the cytosolic and the microsomal enzyme was also decreased considerably. Separate addition of 720 PM sodium taurocholate to this type of assay led to an increase in the cytosolic activity to approx. 60% that found using mixed micelles, but the microsomal activity was only increased to about 11% of that value (Table I). The effect of sodium taurocholate on the enzyme activities was further investigated by separate addition of varying concentrations of the bile salt to assays using the cholesteryl oleate : phosphatidylcholine vesicles as substrate. The results are shown in Fig. 3. The activity of the cytosolic neutral cholesteryl ester hydrolase increased with sodium taurocholate concentration up to a maximum of 59% of the activity seen using mixed micelles at 720 PM, while the microsomal activity reached a maximum of only 12% of that value at a level of 1 mM sodium taurocholate.
Fig. 1. The effect of time and enzyme protein concentration on microsomal and cytosolic neutral cholesteryl ester hydrolase activities. The effect of time of incubation (A and B) and concentration of microsomal or cytosolic protein (C) on neutral cholesteryl ester hydrolase activity in microsomes (A) and cytosol (0) isolated from rat lactating mammary glands. Each point is the mean of duplicate determinations and the experiments shown are typical of several performed.
obtained the pH profiles were similar in all experiments with the highest activity occurring at 6.6 for the cytosolic and 7.0 for the microsomal enzyme in every case. Preincubation of the assay mixture for more than 10 min before addition of the substrate led to a progressive decrease in the activity of neutral cholesteryl ester hydrolase in the microsomes, so that after 30 min preincubation activity was approx. 55% that seen using a 10 min preincubation period. Activity of the cytosolic enzyme showed a similar decrease with time of preincubation. The activity of both enzymes was markedly inhibited by 1 M NaCl (W control value, microsomes 8.5 k 3.7 [3], cytosol 19.6 + 10.9 [4]).
Of 5.5
75
65
PH
Fig. 2. The effect of pH on microsomal and cytosolic neutral cholesteryl ester hydrolase activities. The effect of pH on neutral cholesteryl ester hydrolase (CEH) activity in microsomes (A) and cytosol (0) prepared from rat lactating mammary glands. The activities of the enzymes were assayed in Mes buffer (pH 5.5-6.8) or Tris-HCI buffer (pH 7.0-8.0) (100 mM for both buffers). Each point is the mean from three separate experiments. Error bars show the standard deviation.
93
cholesteryl oleate @M) 0 0.0
OS
1s
10
20
taurocholate (FM) Fig. 3. The effects of sodium taurocholate on microsomal and cytosolic neutral cholesteryl ester hydrolase activities. Neutral cholesteryl ester hydrolase (CEH) activity in microsomal (A) and cytosolic (0) subcellular fractions prepared from rat lactating mammary glands was assayed using cholesteryl [l-t4C]oleate : phophatidylcho~ne vesicles (molar ratio, 1: 4) as substrate. Sodium taurocholate was added to the assays separately prior to the preincubation period. Data are expressed as a percentage of the values obtained using mixed micelies as substrate. Each point is the mean from three (microsomes) or four (cytosol) experiments. Error bars show the standard deviation. Absolute control values were as given for Table I.
These results show that the cytosolic neutral cholesteryl ester hydrolase is activated by bile salt, but the increased activity found when the substrate is presented as a mixed micelle is due only partially to the presence of sodium taurocholate, with an additional 40% increase being caused by the physical dispersion of the substrate. The microsomal activity is activated by bile salt to a lesser extent, but in this case approx. 90% of the activity found with mixed micelles can be attributed to the physical dispersion of the cholesteryl oleate. The effect of varying the concentration of cholesteryl oleate in the assay mixture while rn~nt~~ng the molar
Fig. 4. The effect of substrate concentration on microsomal and cytosolic neutral cholesteryl ester hydrolase activities. The effect of varying concentrations of substrate (mixed micelles) on neutral cholesteryl ester hydrolase (CEH) activity in microsomal (A) and cytosolic (0) subcellular fractions isolated from rat lactating mammary glands. Each point is the mean from three (microsomes) or four (cytosol) experiments. Error bars show the standard deviation.
ratio of cholesteryl oleate : phosphatidylcholine : sodium taurocholate at 1 : 4: 2 in the mixed micelles on the activity of the microsomal and cytosolic neutral cholesteryl ester hydrolases is shown in Fig. 4. The cytosolic enzyme did not show typical Michaelis-Menton kinetics, possibly because of the effect of the bile salt present in the mixed micelle substrate. Alternatively, as it is known that increasing the amount of substrate dispersion in a heterogeneous reaction system increases the surface area containing substrate, not its surface concentration, the atypical kinetics may reflect a weak affinity of the enzyme for the mixed micelles. The effect of MgCl and A TP on microsomal and cytosolic neutral cholesteryl ester hydrolase activities
The activity of neutral cholesteryl ester hydrolase in the microsomes was decreased in the presence of MgCl, in a dose-dependent way (Fig. 5). When ATP was
TABLE I l%e effect of mode of presentation of the substrate on microsomal and cytosolic neutral cholesteryl ester hydrolme activities Microsomal and cytosolic fractions were prepared from rat lactating mammary tissue as described in Materials and Methods. Neutral cholesteryl ester hydrolase (CELL) activities were determined using substrate prepared as mixed micelles (cholesteryl oleate : phosphatidylcho~ne : sodium taurochotate, molar ratio 1:4: 2) or as cholesteryl oleate:phosphatidylcho~ne vesicles (molar ratio 1: 4) or added in ethanol [16]. Data are expressed as a percentage of the value obtained with mixed micelles (microsomes, 1216 i 287, cytosol 897 f 360 pmol/min per mg protein) and are given as the mean f S.D. (more than three experiments) or the range of values obtained (two experiments). Figures in brackets represent the number of animals used. Substrate
Separate additions
CEH activity (W mixed micelle value) microsomes
cytosol
1. Mixed micelles
none
100
100
2. Cholesteryl oleate: phosphatidylchohne 3. Cholesteryl oleate in ethanol
none sodium taurocholate (720 PM) none phosphatidylcholine (1.44 mM) phosphatidylchohne (1.44 mM) + sodium taurocholate (720 PM)
5.8 4 1.2 (3) 11.3+0.8 (3) 6.5 f 2.6 (2) 4.3 f 2.5 (2) 10.1* 5.2 (2)
15.9* 7.0 (4) 59.1 f 15.6 (4) 2.3 f 2.0 f
0.8 (2) 0.6 (2)
4.6zbO.l (2)
94 included in the incubation in a 1 : 1 molar ratio with MgC12, however, activity was restored almost to control levels. The neutral cholesteryl ester hydrolase activity in the cytosol was also decreased by MgCl,, but the effect was less marked than that seen with the microsomal enzyme (Fig. 5). Activity of this enzyme was also restored to values near control levels in the presence of equimolar concentrations of MgCl, and ATP. In the absence of MgCl,, ATP had no effect on the activity of either enzyme. When the ATP concentration was increased while maintaining a constant MgCl, (1 or 2 mM) level the activity of the microsomal neutral cholesteryl ester hydrolase increased linearly as a function of ATP concentration (Table II). The cytosolic activity was also increased with increasing ATP concentrations (with 2 or 5 mM MgCl,), but the effect again was less marked than that seen with the microsomal activity (Table II).
40 00
Fig. 5. The effect of MgCI, and ATP on microsomal and cytosolic neutral cholesteryl ester hydrolase activities. The effect of MgCl, (0, A), or ATP and MgCl, (0, A) in a 1 : 1 molar ratio on neutral cholesteryl hydrolase activity in cytosol (0, 0) and microsomes (A, A) prepared from rat lactating mammary glands. Data are given as a percentage of the value found in the absence of ATP or MgCl, (absolute values (pmol/min per mg protein); microsomes 869 it 228; cytosol 784i 167). Each point is the mean from four experiments. Error bars show the standard deviation.
TABLE The
II
effect
cholesteryi
of MgCl,
and A TP on microsomal
ester hydrolase
and ~viosohc
neutrul
uctiuirm
Microsomal and cytosolic fractions from rat lactating mammary glands were prepared as described in Materials and Methods. Neutral cholesteryl ester hydrolase activity was determined in the presence of the MgCl, and ATP concentrations shown. Data are expressed as a percentage of the values obtained in the presence of the each concentration of MgCl, without ATP, and are given as the meankS.D. Figures in brackets show the number of animals used. nd., not determined. Absolute values in the absence of ATP (pmol/min per mg protein): microsomes. 1 mM MgCl, 550+ 80, 2 mM MgCl z 352+113; cytosol, 2 mM MgCl, 731k163, 5 mM MgCl, 574+103.
W&l,
ATP
CEH activity
(mM)
(mM)
microsomes
1 1 1 2 2 2 2 2 5 5 5 5
0 5 0 1 2 5 10 0 2 5 10
100 118.0*11.5 182.1 If- 4.9 100 120.4+ 5.1 135.0* 13.7 183.8 f 8.4 238.5 f 19.X n.d. n.d. n.d. nd.
R cytosol
(4) (4) (3) (4) (3) (3)
n.d. nd. 100 n.d. 118.4+3.0 129.4k3.5 130.7k8.2 100 110.3 + 3.5 126.2 f 7.1 122.3+4.2
(4) (4) (3) (4) (4) (4)
The effect of isolation of subcellular fractions in the presence of NaF and/or EDTA on microsomal neutral cholesteryl ester hydrolase activity If the neutral cholesteryl ester hydrolases under study were regulated by a mechanism involving phosphorylation-dephosphorylation, then their activities might be modified during the preparation of the subcellular fractions by endogenous protein kinases and/or phosphatases. This possibility was tested for the microsomal enzyme by preparing the fractions in the presence of 50 mM NaF and/or 10 mM EDTA, which are known to inhibit phosphatases and protein kinases, respectively [18,19]. Preparation of the microsomal fractions in the presence of 10 mM EDTA led to decrease of about 50% (49.5 f 17.5%) in the activity of the microsomal neutral cholesteryl ester hydrolase, while inclusion of 50 mM NaF during the isolation procedure did not cause any significant change (87.0 _t 12.0%). The results obtained when both NaF and EDTA were present were similar to those found with EDTA (51.2 f 18.7%). The absolute value found when the microsomes were prepared in the absence of NaF or EDTA was 938 f 114 pmol/min per mg protein. The effects of MgCl, and ATP with MgCl, on the activity of neutral cholesteryl ester hydrolase in microsomes prepared in the presence of NaF and/or EDTA and resuspended in buffer without these inhibitors are shown in Table III. Data are shown as a percentage of the value obtained when the subcellular fractions were
95 TABLE
III
The effect of MgCl,
and ATP on neutral cholestetyl ester hydrolase activity in microsomal fractions isolated in the presence of EDTA and/or
NaF
Microsomal fractions from rat lactating mammary tissue were prepared for assay as described in Materials and Methods. Neutral cholesteryl hydrolase (CEH) activity was determined in the presence of MgCl, and MgCl 2 with ATP. Data are expressed as a percentage of the value obtained with microsomes isolated in the absence of EDTA or NaF and assayed without MgCl, or ATP (absolute value, 892+133 pmol/min per mg protein). Each value is the mean from four separate experiments+ S.D. Additions
None MgCl, MgCl,
Neutral
to assays
(1 mM) (1 mM) & (ATP 5 mM)
Significance
limits: MgCl,
vs. no additions;
CEH activity
(% control
value)
control
+ EDTA
+ NAF
+ EDTA and NaF
100 77.5 f 4.4 146.8 +43.3 b
51.3* 9.2 49.8 + 10.0 101.7 + 31.8 b
82.3i 8.0 67.7i 9.0 a 150.9 * 41.5 c
61.7+ 5.8 59.1+ 1.1 105.1+ 11.4 c
a P < O.O5;MgCI,
+ATP
vs. MgCI,,
prepared in the absence of inhibitors and assayed without M&l, or ATP (control value). The activity of the microsomal enzyme was significantly inhibited by MgCl, in microsomes prepared without protein kinase or phosphatase inhibitors or with NaF, but not in those isolated in the presence of EDTA or EDTA and NaF, suggesting that inactivation has occurred during the preparation under the latter conditions. When ATP was present in addition to MgCl, neutral cholesteryl ester hydrolase activity was significantly increased compared to that found with MgCl, only in all cases. This type of experiment was not carried out on the cytosolic neutral cholesteryl ester hydrolase, as it was not possible to remove the EDTA or NaF completely from the cytosol before assay of the enzyme.
TABLE
b P < 0.025; ‘P < 0.01.
Reversible inactivation-activation of microsomal neutral cholesteryl ester hydrolase In order to test whether the observed effects of MgCl, and ATP on the microsomal neutral cholesteryl ester hydrolase were reversible microsomes were incubated with MgCl, (5 mM) for 10 min (stage l), ATP (5 mM) was then added and the mixture incubated for a further 10 min (stage 2). Aliquots were assayed for neutral cholesteryl ester hydrolase activity after stages 1 and 2. ATP and MgCl, were then removed by re-isolation of the microsomes by centrifugation at 105 000 x g (60 min). The pellet was washed once with Tris-HCl buffer (pH 7.0) (100 mM) and resuspended in similar buffer. The incubations and assays with MgCl, and ATP were then repeated (stages 3 and 4). The activity of
IV
The effect of protein kinase and alkaline phosphatase on neutral cholestetyl ester hydrolase activities Microsomal and cytosolic fractions from rat lactating ester hydrolase (CEH) activity was determined in the 1 mM for the microsomes and 2 mM for the cytosol. (pmol/min per mg protein): microsomes, 912 f 108; animals used. n.s., not significant. Additions
Neutral
1 none 2 MgCl, (1 or 2 mM) 3 MgCl, (1 or 2 mM) + ATP (5 mM) 4 CAMP (50 PM) + protein kinase (50 pg) 5 2+4 63+4 7 Alkaline phosphatase: 10 pg 20 Pg 50 pg 100 I-18 Significance 3 vs. 2 6 vs. 3 6 vs. 5
mammary glands were presence or absence of Data are expressed as cytosol 1121+ 274) and
prepared as described in Materials and Methods. Neutral cholesteryl the factors shown. The concentration of MgCl, used in each case was a percentage of the value obtained without additions (absolute values are given as the mean + S.D. Figures in brackets show the number of
CEH activity
(W control
value)
microsomes
cytosol
100 78.1 f 96.1+ 105.9 f 84.3 + 125.6+ 67.3 + 43.8 f 13.2 f 4.4 f
100 79.6 f 89.7 f 96.1 f 79.2 f 93.6 + 86.8 f 49.8 * 27.0* 17.7 f
5.8(5) 10.7 (5) 4.5 (5) 2.6 (4) 11.6 (5) 35.4 (3) 30.0 (3) 3.4 (3) 3.9 (4)
4.2(4) 6.0 (4) 7.8 (4) 15.0 (3) 3.9 (4) 3.1 (4) 7.9 (4) 17.2 (3) 14.2 (3)
limits P i 0.01 P i 0.005 P < 0.0005
P < 0.025 n.s. ns.
96 the enzyme in control microsomes incubated for a sirnilar time period in the absence of ATP and/or MgCl, was determined at each stage. Data are given as percentages of the control values found at each stage (absolute value at stage 1, 765 k 96 pmol/min per mg protein). The control value at stage 4 was 80% that found for the control at stage 1. Neutral cholesteryl ester hydrolase activity was decreased at stage 1 by MgCl, (77.6 f 12.1%), but this was reversed at stage 2 after the addition of ATP (112.4 + 4.7%). After removal of ATP and MgCl, by re-isolation of the microsomes the activity of the enzyme in control assays was similar to that found in stage 1 controls. Activity was decreased by MgCl, at stage 3 (70.6 k 22.8%) and this inactivation was reversed again by ATP at stage 4 (153.8 + 39.4%). The effect of protein kinase and alkaline phosphatase on neutral cholesteryl ester hydrolase activities The effect of a cyclic AMP-dependent protein kinase on microsomal and cytosolic neutral cholesteryl ester hydrolase activity is shown in Table IV. When microsomes were incubated with cyclic AMP and cyclic AMP-dependent protein kinase the activity of the enzyme was not significantly different from that found in control assays. In addition, the inhibition of neutral cholesteryl ester hydrolase activity by MgCl, was not prevented by cyclic AMP and the protein kinase. In the presence of ATP, MgCl,, cyclic AMP and the protein kinase, however, the activity was significantly higher than that found with MgCl, and ATP alone (P < 0.005). In contrast, the activity of the cytosolic neutral cholesteryl ester hydrolase found in the presence of cyclic AMP-dependent protein kinase, cyclic AMP, ATP and MgCl, was not significantly higher than that seen with ATP and MgCl, alone. Alkaline phosphatase (bovine milk) inhibited both the microsomal and the cytosolic neutral cholesteryl ester hydrolase activities in a dose-dependent way (Table IV). Discussion The results reported here, together with our previous work [9] demonstrate that neutral cholesteryl ester hydrolase activity is associated with the microsomal and cytosolic subcellular fractions isolated from rat lactating mammary tissue. The activities in the two fractions appear to have different pH optima and also differ in their response to bile salt (Table I, Fig. 3) and to certain conditions favouring phosphorylation or dephosphorylation (Table IV), indicating that they are distinct enzymes. Tawil et al. [lo] have also reported the presence of neutral cholesteryl ester hydrolase activity in rat lactating mammary glands. In their experiments, however, the greater proportion of the total activity was
associated with the microsomal fraction (60%) while we found the highest percentage in the cytosol (60%) [9]. In addition, the pH optimum of the main microsomal activity was 7.5-8.5 as opposed to 7.0 in our work. It is possible that some of these differences may be explained by the mode of presentation of substrate used in our laboratory. It has been reported [20] that the pH optimum of cholesteryl ester hydrolase in homogenates of aorta tissue differs depending on the physical dispersion of the substrate. In our experiments addition of the cholesteryl oleate as a mixed micelle with phosphatidylcholine and sodium taurocholate gave activities 10-20fold higher than those obtained when the substrate was added in ethanol as in the experiments of Tawil et al. [lo], and furthermore, the increase in activity was not reproduced by adding the appropriate amounts of phosphatidylcholine and sodium taurocholate to the assays using the latter method (Table I), suggesting that the physical dispersion of the substrate plays a large part in determining the observed activity of the neutral cholesteryl ester hydrolases. Omission of sodium taurocholate from the dispersion, so that the substrate consisted of cholesteryl oleate : phosphatidylcholine vesicles led to a very marked decrease in the activity of both enzymes, but in this case a substantial part of the cytosolic activity could be restored by separate addition of the bile salt. In contrast, sodium taurocholate was not able to restore the microsomal activity above 12% of that found with the micellar substrate (Table I, Fig. 3). Bile salt-dependent cholesteryl ester hydrolases have been described in rat liver cytosol [21,22] and pancreas [23,24], and there is some evidence to suggest that the bile salt acts to promote changes between monomeric and polymeric forms of the enzyme [24,25]. However, as the mammary gland activities are unlikely to be exposed to millimolar quantities of bile salt in vivo it seems more likely that the stimulatory effect on the cytosolic enzyme is caused by the detergent properties of taurocholate stabilising the enzyme at the lipid/water interface, as suggested by Tsujitsa et al. [26]. Both the microsomal and the cytosolic enzymes were inhibited by about 80-90% by 1 M sodium chloride, in agreement with the results of Tawil et al. [lo]. Neutral cholesteryl ester hydrolase activity in a number of other tissues, including the liver, adipose tissue, adrenal cortex, arterial tissue and macrophages [6,11- 151 has been shown to be activated by conditions favouring phosphorylation of the enzyme, and inactivated by those favouring dephosphorylation. In our experiments, the microsomal neutral cholesteryl ester hydrolase activity from rat lactating mammary glands also seemed to follow this pattern. The enzyme was deactivated in the presence of MgCl, (Fig. 5) which is known to activate phosphatases, and therefore favour dephosphorylation, but activity was almost fully restored to control levels
97 when an equimolar concentration of ATP was included. This effect could be explained by the chelation effect of ATP on the Mg2+ ions, however, at fixed concentrations of MgCl, the enzyme activity was increased by ATP in a dose-dependent way (Table II) and at the highest concentration used (10 mM) activities considerably above those found in the absence of MgCl, (150200%) were seen. As neither ATP (l-10 mM) alone, nor EDTA (2-5 mM) had any effect on the microsomal cholesteryl ester hydrolase, these increases cannot be accounted for by the chelation of endogenous Mg2+. In addition, the activity of the enzyme found with MgCl, and ATP was enhanced when cyclic AMP and a cyclic AMP-dependent protein kinase were included in the assay. The activity was also inhibited by alkaline phosphatase from bovine milk (Table IV), but a similar enzyme from calf intestine had no effect, suggesting that there is some specificity for the type of alkaline phosphate required for inhibition. The loss of activity (56687%, Table Iv) seen with 20-50 pg bovine milk alkaline phosphatase (equivalent to 0.06-0.15 units) in our experiments is comparable to that found in studies demonstrating the regulation of rat liver cholesteryl ester hydrolase by phosphorylation-dephosphorylation, in which 0.1 units of bovine liver alkaline phosphatase was shown to cause 69% loss of activity [12]. We cannot rule out the possibility, however, that the effect of bovine milk alkaline phosphase is caused by binding to the microsomes or the cholesteryl ester hydrolase enzyme. Taken together our results suggest that the microsomal neutral cholesteryl ester hydrolase is activated when phosphorylated and inactivated in the dephosphorylated state. The decrease in the activity of this enzyme found when microsomes were isolated under conditions where phosphorylation by endogenous protein kinases was inhibited with 10 mM EDTA during preparation [19] also supports this conclusion. The reversible nature of the phosphorylation-dephosphorylation is demonstrated by the restoration of activity of enzyme inactivated by MgCl, by subsequent addition of ATP, and by the repetition of this cycle after re-isolation of the microsomes. Our results, therefore, provide evidence that the microsomal neutral cholesteryl ester hydrolase in rat lactating mammary tissue may be regulated by hormone action involving a cyclic AMP-dependent protein kinase. The hormone-sensitive neutral cholesteryl ester hydrolases described in other tissues [6,11-151 are all associated with the soluble fraction of the cell. Our experiments provide some evidence that the mammary gland cytosolic neutral cholesteryl ester hydrolase may also be regulated by a phosphorylation-dephosphorylation mechanism. The activity was inhibited to some extent by MgCl, and this was reversed by the addition of ATP (Fig. 5). Increasing concentrations of ATP at fixed MgCl, concentrations also led to increased activ-
ity (Table II). The enzyme was also inhibited by alkaline phosphatase from bovine milk (Table IV). These results suggest that the cytosolic neutral cholesteryl ester hydrolase in the mammary gland is active when phosphorylated and inactive when dephosphorylated. Involvement of a cyclic AMP-dependent protein kinase, however, could not be demonstrated (Table IV). The neutral cholesteryl ester hydrolase activities that have been shown previously to be regulated by mechanisms involving phosphorylation-dephosphorylation are present in tissues which have a significant role in cholesterol metabolism, such as the liver [12], adrenal cortex [ll] and adipose tissue [6] or in those which can accumulate cholesteryl ester under pathological conditions, such as arterial tissue [14] and macrophages [15]. Our finding that rat lactating mammary tissue contains microsomal and cytosolic neutral cholesteryl ester hydrolases which may be regulated by this type of mechanism raises the possibility that these enzymes play an important part in the regulation of cholesterol metabolism in the mammary gland, and in the provision of cholesterol for secretion into milk. Acknowledgements This work was supported by a grant from The Wellcome Trust. Dr Maria Jose Martinez received a grant from the University of the Basque Country. We should like to thank Miss Adrienne Carroll for expert technical assistance. References 1 Clarenburg, R. and Chaikoff, I.L. (1966) J. Lipid Res. 7, 27-37. 2 Gibbons, G.F., Pullinger, C.R., Munday, M.R. and Williamson, D.H. (1983) Biochem J. 212, 843-848. 3 Raphael, B.C., Patten, S. and McCarthy, R.D. (1975) FEBS Lett. 58, 47-49. 4 Goldstein, J.L., Dana, S.E., Faust, J.R., Beaudet, A.L. and Brown, M.S. (1975) J. Biol. Chem. 250, 8487-8495. 5 Boyd, G.S. and Gorban, A.M.S. (1980) in Newly Discovered Systems of Enzyme Regulation by Reversible Phosphorylation (Cohen, P., ed.), Vol. 1, pp. 95-134, Elsevier/North-Holland, Amsterdam. 6 Steinberg, D. (1976) Adv. Cyclic Nucleotide Res. 7, 157-198. 7 Femandez, C.J., Lacort, M., Gandarias, J.M. and Ochoa, B. (1987) Biochem. Biophys. Res. Commun. 146, 1212-1217. 8 Keenan, T.W. and Patten, S. (1970) Lipids 5, 42-48. 9 Martinez, M.J. and Botham, K.M. (1990) B&hem. Sot. Trans. 18, 619-620. 10 Tawil, S.G., Shand, J.H. and West, D.W. (1989) B&hem. Sot. Trans. 17, 653-654. 11 Trzeciak, W.H. and Boyd, G.S. (1974) Eur. J. Biochem. 46, 201207. 12 Ghosh, S. and Grogan, W.M. (1989) Lipids 24, 733-736. 13 Pittman, R.C., Khoo, J.C. and Steinberg, D. (1975) J. Biol. Chem. 250, 4505-4511. 14 Hajjar, D.P., Minick, R. and Fowler, S. (1983) J. Biol. Chem. 258, 192-198. 15 Khoo, J.C., Mahoney, E.M. and Steinberg, D. (1981) J. Biol. Chem. 256, 12659-12661.
98 16 Gorban, A.M.S. and Boyd, G.S. (1977) FEBS Lett. 17 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and (1951) J. Biol. Chem. 193, 265-275. 18 Erickson, S.K., Shrewsbury, M.A., Gould, R.G. and (1980) Biochim. Biophys. Acta 620, 70-79. 19 Nordstrom, J.L., Rodwell, V.W. and Mitschelen, J.J.
79, 54-58. Randall, R.J. Cooper,
A.O.
(1977) J. Biol.
Chem. 252, 8924-8928. 20 Kothri, H.V., Bonner, M.J. and Miller, B.F. (1970) Biochim. Biophys. Acta 202, 325-331. 21 Harrison, E.H. (1988) Biochim. Biophys. Acta. 963, 28-34.
22 Camulli, E.D., Linke, M.J., Brockman, H.L. and Hui, D.Y. (1989) Biochim. Biophys. Acta. 1005, 177-182. 23 Vahouny, G.V., Weersing, S. and Treadwell, C.R. (1965) Biochim. Biophys. Acta. 98, 607-616. 24 Hyun. J., Steinberg, M., Treadwell, CR. and Vahouny. G.V. (1971) Biochem. Biophys. Res. Commun. 44. 819-825. 25 Tuhackova. I., Kriz, 0. and Hradec, J. (1980) Biochim. Biophys. Acta. 617, 439-445. 26 Tsujitsa, T., Mizuno, N.K. and Brockman, H.L. (1987) J. Lipid Res. 28, 1434-1443.
