Influence of Cholesterol and Unsaturated Fatty Acids

ENDOCRINOLOGIA EXPERIMENTALIS, Vol. 19, 1985

Influence of Cholesterol and Unsaturated Fatty Acids on (1 25 l]hCG Binding to Rat Testicular Membranes J. KOLENA, S. HORKOVICS-KOVATS, E. SEBOKOVA, P. BLAZfcEK Institute of Experimental Endocrinology, Centre of Physiological Sciences, Slovak Academy of Sciences, 833 06 Bratislava, Czechoslovakia

Kolena J., Horkovics-Kovats S., Sebokova E., Blazicek P.: Influence of Cholesterol and Unsaturated Fatty Acids on [125I]hCG Binding to Rat Testicular Membranes. Endocrinologia Experimentalis 19, 195-203, 1985.

Incubation of testicular membranes with cholesteryl-hemisuccinate resulted in an increase of both the membrane microviscosity and [125I]hCG specific binding. 14C-labelled cholesterol as well as fatty acids were effectively incorporated into membrane preparations. Insertion of unsaturated fatty acids (cis-vaccenic, linoleic, oleic, linolenic and arachidonic) into the membrane decreased its microviscosity under simultaneous disappearance of [ 125I]hCG specific binding. The latter phenomenon seems to be caused by an increase of nonspecific binding which was not due to the uptake of radioactivity from labelled gonadotropin by fatty acids or double bonds of fatty acids incorporated into testicular membranes or taken up by a resin Amberlite IRA-400. The disappearance of specific binding of [125I]hCG by the, fluidizing action of unsaturated fatty acids can be reversed by the subsequent action of cholesterol.

Gonadotropin receptors are a heterogenous population of molecules which interact with membrane components e.g. phospholipids and protein [Catt and Dufau 1977]. Several studies suggested that some reactions can alter the comp03ition and the physical state of membrane lipids and thereby affect also the function of membraneproteins which are embedded in the lipid core of the membrane. These membrane proteins are normally free to migrate laterally in the plane of the membrane, the rate of their diffusion being restricted by several factors, one of which is the fluidity of membrane lipids [Shinitzky and Henkart 1979]. Growing body of evidence indicates that the fluidity of the lipid environment may influence the activity of membrane enzymes such as hormone-responsive adenylate cyclase [Orly and Schramm 1975] and the accessibility of protein hormone receptors [Ginsberg et al. 1981; Daveand Witorsch 1983; Kolena et al. 1984].

195

Previous study in this laboratory referred to positive correlation between testicular membrane fluidity and LH/hCG receptors during development of rats [Kolena and Ondrias 1984]. In this study the modulation of the accessibility of LH/hCG receptors in testicular membranes was investigated.

Materials and Methods Materials: Purified hCG (CR 123, 12 780 U mg- 1 ) and ovine LH (NIH-LH 13, 0.093 x NIH-LH-S 1 U mg- 1 ) were generously supplied by NIAMDD, NIH (Bethesda, Md.). Na 125I, 4- 14 C-cholesterol, l- 14 C-palmitic acid and l- 14 C-arachidonic acid were purchased from the RADIOCHEMICAL CENTRE (Amersham). Cholesteryl-hemisuccinate, fatty acids, polyvinylpyrrolidine (PVP) and l,6-diphenyl-1,3,5-hexatriene (DPH) were products of SIGMA. Amberlite IRA-400 was obtained from SERVA. Other reagents used were of analytical reagent grade. Crude membrane preparation: Male Wistar rats aged 55--65 days were killed by , decapitation. The homogenates of decapsulated testes in ice-cold 50 mmol 1- 1 Tris-HCl buffer (pH 7.4) was filtered through 6 layers of surgical gauze, centrifuged at 100 x g for 15 min and the supernatant was further centrifuged at 20 000 x g for 30 min [Kolena 1976]. The final pellet was resuspended in the same buffer(200 mg of tissue per ml). Similar method was used for a preparation of hepatic membrane fractions. Incubation of membranes with lipids: 10 mg of lipid was dissolved in 0.2 ml of warm glacial acetic acid. The solution was diluted under stirring to 50 volumes with 50 mmol 1- 1 Tris-HCl, pH 7.4, containing 3.5% PVP. The pH was readjusted to 7.4 with solid Tris base [Heron et al. 1980]. 1.5 ml of crude testicular membrane fractions in Tris-HCI buffer were incubated 90 min with varying concentrations of lipid suspensions at 25 °C. The control samples were incubated with the same concentration of Tris-acetate buffer with PVP. The testicular membranes were then centrifuged at 20 000 X g for 30 min and pellets were washed once with Tris-HCl buffer and once with 0.05 mol 1- 1 phosphate buffer (pH 7.4) with 0.15 mol 1- 1 sodium chloride (PBS buffer). The final pellet was suspended in PBS buffer (200 mg ml- 1). Before the incubation with lipids Amberlite IRA-400 was treated for 5 min with 1 mol 1- 1 sodium hydroxide. The resin was allowed to settle and then successively washed with several portions of water. Ten mg of the resin were incubated with the same lipid suspensions as used for incubation of testicular membranes. hCG binding assay: 0.1 ml aliquots of testicular membrane fraction were incubated 16 hat room temperature with 0.1 ml PBS containing 1 mg ml- 1 BSA with or without 100-fold excess of unlabelled hCG and 0.1 ml [125l]hCG (1-1.5 ng, specific activity about 2.3 TBq g- 1 ). After incubation and centrifugation the pellets were washed twice with 2 ml of cold PBS (Kolena and Sebokova 1983; Kolena et al. 1983]. The results were expressed as specific binding per mg protein. Protein concentration was determined according to Lowry et al. [1951]. [125I]insulin binding was carried out in a similar manner by the method described previously [Fickova and Macho 1981]. Fluorescence polarization measurement: Fluorescence polarization was measured with an AMINCO BOWMAN SPF spectrofluorometer [Shinitzky and Inbar 1976]. Cmde testicular membranes (100 µg protein) were incubated with 2 ml of DPH (2 µmol 1- 1 ) in PBS buffer for 1 hat 24 °C. The fluorescence polarization was computed by the equation:

196

I where lvv and Ivh are the fluorescence intensities detected through a polarizer oriented parallelly and perpendicularly to the direction of vertically polarized light, Ihv/Ihh represents the ratio when the excitation is polarized horizontally and the emission observed through the analyzer oriented perpendicularly and parallelly, respectively. Lipid microviscosity was estimated by the empirical relation 2P/(0.46 - P) [Heron et al. 1980]. Statistical evaluation: Student's t-test was used and results were expressed as means± S. E.

Results

1

Fig. I. shows that the incubation of crude testicular membranes with varying concentrations of cholesteryl-hemisuccinate resulted in a dose-dependent increase of apparent membrane microviscosity and [125l]hCG specific binding. 14C-labelled cholesterol is effectively incorporated into rat testicular membrane preparations. Testicular membranes took up 16.6 % of cholesterol added under in vitro conditions (Fig. 2). 14C-labelled palmitic and arachidonic acids were incorporated into the membranes at nearly the same level (9.5 % and 9.1 %, respectively). To determine whether the lipids with fluidizing action on plasmatic membrane can change the accessibility of LH/hCG receptors, the effect of unsaturated fatty acids on [125l]hCG binding to testicular membranes was investigated. The insertion of unsaturated fatty acids into the membrane caused a disappearance of [125I]hCG specific binding. The incubation of the membranes with cis-vaccenic (18 : 1), oleic (18: 1), linoleic (18 : 2), linolenic (18 : 3) and arachidonic (20 : 4) acids resulted in a decrease of total [125l]hCG binding and increase of nonspecific binding (Fig. 3 and 4).

Fig. 1. Effect of increasing doses of cholesteryl-hemisuccinate on microviscosity (open circles) and [125I]hCG binding (close circles). Aliquots of particulate preparations were incubated with cholesterol in Tris-acetate buffer (pH 7.4) containing 3,5% PVP for 90 min at 25 °C. After washing, the membranes were assayed for lipid microviscosity using DPH as a probe, and for [125I]hCG specific binding. Each point is the mean± S. E. of 5 estimations.

.

,e

-..,

.,.

vt

E 12

6

0

o Q.

!

0

0,2

o., CHOLESTEROL

o.&

.. o.a (111g ml I

197

Fig. 2. Incorporation of 14C-cholesterol, 14 C-palmitic acid and 14 C-arachidonic acid into testicular membrane. The membranes were incubated with lipids in Tris-acetate buffer (pH 7.4) with 3.5% PVP for 90 min at 25 °C. After washing, the membranes were solubilized with 0.5 ml Soluene-350 and counted in a PACKARD liguid scintilation counter. The results are means± S. E. of 3 estimations.

9 CHOLESlEROL

0

3 CPM

6 9 ADDED ( 105)

The rise of nonspecific binding in testicular membranes by increasing incorporation of unsatureated fatty acids into the membranes could be due to the uptake of radioactivity from labelled hCG into inserted fatty acids. Therefore, the radioactivity uptake from [125l]hCG, [125l]insulin and Na[ 125l] by the membranes with incorporated linoleic acid was measured (Fig. 5). Increasing concentration of linoleic acid in the membranes enhanced nonspecific binding in both testicular and hepatic membranes. 125 [ l]hCG was not specifically bound to crude membrane preparations of liver, which is not a target tissue for this hormone. Although [125l]insulin was specifically bound to hepatic as well as testicular membranes, incorporated linoleic acid did not increase nonspecific binding. Further, inserted linoleic acid did not change the radioactivity of any type of membrane incubated with Na[ 125I]. A further series of experiments was performed to determine whether the radioactivity from labelled gonadotropin can bind to the double bonds of unsaturated fatty acid incorporated into membranes. The results presented in Fig. 3 and 5 demonstrate that the rise of nonspecific binding in crude testicular membranes with inserted unsaturated fatty acids was not caused by the binding of radioactivity on the double bonds of fatty acids. The membranes with incorporated arachidonic (4 double bonds), linolenic (3 double bonds) and linoleic acid (2 double bonds) failed to take up more radioactivity than the fatty acids with one double bond (oleic and cis-vaccenic acids). In addition, similar results were obtained with fatty acids coupled to the resin Amberlite IRA-400. Fig. 6 shows that the resin bound both arachidonic and palmitic acids in a considerable extent. Moreover, incubation of (1 20I]hCG with resin-coupled fatty acids resulted in a lower binding of radioactivity on the resin with coupled unsaturated (oleic, petroselinic and linoleic acid) than with saturated fatty acids (plamitic and stearic acids). The experiments indicated that is possible to change the disappearance of specific 125 [ l]hCG binding caused by unsaturated fatty acids. Preicubation of testicular membranes with increasing concentrations of linoleic acid resulted in a decrease of membrane microviscosity with simultaneously disappearance of specific binding

198

~

100

C 0 I.I IO

~O"I

~

0

~ IO ::, 0 ID 40

E ...... Cl

z

::::)

I!)

u

,=...

18

E

- - Total binding - --- Nonspecific binang

20

0,2

0,4

FA TTY

-1

0

co

------------a

l!J

~ 6

o.,

LINOLENIC A.

.F

ACID (mg ml )

~

Fig. 3. Effect of increasing dose of ~ns~turated fatty acids on [125l]hCG bmdmg. Incubation of testicular membrane with cis-vaccenic (full circles), linoleic (open circles) and arachidonic acid (triangles) was done as described in the legend to Fig. 1. Each value is the mean S. E. of 5 estimations.

±

0

0,2

0,4 0,6 FATTY ACID (mg mr1) Fig. 4. Total (full line) and nonspecifi_c (dashed line) binding of [125l]hCG to testicular membranes incubated with increasing amounts of unsaturated fatty acids. Each point is a mean S. E. values of 5 estimations.

±

----r-----.----.----,

16

12

8 (Tl

S2 4 Q

z

::::)

0

co :E

a.. u

(

Fig. 5. Total (full line) and nonspecific (dashed line) binding of labelled hCG, insulin and Nal to crude testicular (closed circles) and hepatic (open circles) membranes. Incubation was performed similarly as described in a legend to Fig. 1. Each point is a mean S. E. of 4 estimations.

±

02

125

0,4

11 INSULIN

0,6

LINOLEIC ACID ( mg m(·1l

199

B

A

0

z

~ :::i:

f;

2

1

CPM ADDED

2 FATTY ACID (mgml- 1)

Fig. 6. Binding of [ 125 I]hCG to the fatty acids taken up on resin Amberlite IRA-400. A. 14O-labelled arachidonic or palmitic acids were incubated with 10 mg of resin in Tris-acetate buffer containing 3. 5 % PVP for 90 min at 25 °0. B. Increasing amounts of fatty acids were incubated with resin as indicated in the legend to A. After 3-fold washing, fatty acids-coupled resin was assayed for [ 125 I]hCG binding. Each point is the mean of 3 (A) or 6 (B) estimations.

.f.. ,/ ······f..... I

,

.. _!

, *

4

•!

•

.,. f

.,.

• ., ,

,,

ff,

0

0,19

0,37

0,75

0

0

0

0

L"~~L~l1V·

CHOLESTEROL

0,19

0,19

0,37

0,75

0

0,75

0,75

0,75

Fig. 7. Effect of cholesterol on testicular membrane preincubated with linoleic acid. The membranes were preincubated with increasing concentrations of linoleic acid in Tris-acetate buffer (pH 7.4) containing 3.5% PVP for 30 min at 25 °0. After washing the membranes were incubated for next 60 min without or with 0. 7 5 mg ml-1, cholesteryl-hemisuccinate and then assayed for lipid microviscosity and for [ 125I]hCG binding. · The asterics indicate statistically significant differences (P < 0.01) of total from nonspecific binding and the bars show the means S. E. of 5 estimations.

±

(Fig. 7, left part), but further incubation of the membranes with cholesteryl-hemisuccinate increased the microviscosity of membrane lipids and caused reappearence of specific [125l]hCG binding at concentrations of linoleic acid of 0.19 and 0.37 mg ml- 1 (Fig. 7, right part).

Discussion The observation that cholesterol increased the accessibility of LH/hCG receptors in testicular membranes is consistent with another report from this laboratory [Kolena et al., submitted]. These results showed that the affinity of binding sites

200

for hCG was not appreciably affected by cholesterol, but that the changes of [125l]hCG binding to testicular membranes apparently were rather associated with alterations in receptor number. This is consistent with the concept that the accessibility of cryptic receptors in membranes is regulated by membrane fluidity. Depending on the system used and the type of receptor the rise of membrane fluidity may exert various effects on the accessibility ofreceptor. Thus, the exposition of beta-adrenergic receptors in rabbit reticulocytes [Strittmatter et al. 1979] and of serotonin receptors in mouse brain membranes [Heron et al. 1980] was enhanced by the increase of microviscosity. On the other hand, Dave and Witorsch [1983] reported that aliphatic alcohols which increase the membrane fluidity may also increase the binding capacity of rat prostatic membranes for prolactin. Nevertheless, the reason for the rise of gonadotropin specific binding by the action of cholesterol on testicular membranes still remains speculative. After the insertion of cholesterol into the membrane its rigid sterol ring with 3 ~-hydroxyl group is oriented toward the aqueous phase and its flexible hydrocarbon chain toward the centre of the bilayer [Huang 1977]. Previous experiments with spin probes I(l,14), 1(12,3) and CAT 16 indicated that the motion in both the hydrophobic and the polar testicular membrane parts with the highest ordering effect at the C16 carbon depth was considerably affected by cholesterol [Kolena et al., submitted]. Heron et al. [1980] explained the increased binding of serotonin to brain membrane incubated with cholesterol by the vertical displacement of membrane protein. According to this concept the bulk of membrane proteins becomes more exposed to the aqueous medium by increasing lipid microviscosity [Borochov and Shinitzky 1976]. However, such explanation of the increased accessibility of LH/hCG receptors by a rigidifying action of cholesterol in testicular membranes is unlikely, because a comparable rigidifying effect of saturated fatty acids did not result in any increase [125l]hCG specific binding [Kolena et al., submitted]. Therefore, it is likely that the effect of cholesterol on testicular receptors may be related rather to its chemical action on biological membranes then to its ordering influence. Among other factors affecting the fluidity of membrane lipid regions the most prominent are unsaturated fatty acids. It has often been assumed that membrane fluidity and fatty acids unsaturation are related in a direct manner. Fatty acids are gradually incorporated into membrane phospholipids, probably as a result of action of the phospholipase and acyltransferase systems [Lands and Hart 1965], but their simple insertion into the lipid bilayer also cannot be ruled out. Fluidizing effect of unsaturated fatty acids on testicular membranes was related to a disappearance of [125l]hCG specific binding, perhaps as a consequence of an increase of nonspecific binding which may result in an abolishment of stimulatory effect of LH on cAMP and testosterone formation by rat Leydig cells [Kolena et al., submitted]. The reason of the rise of nonspecific binding of hormone by the action of different unsaturated fatty acids remains to be explained. This binding may be partly explained by the fact that fatty acids may change the chemical structure and the order of the membrane which is then able to bind labelled gonadotropin in a nonspecific manner. However, this is not a sole attribute of unsaturated fatty acids. We have also observed a similar effect on nonspecific binding of [125I]hCG when gonadal membranes were incubated with chlorpromazine, concavalin-A, diethylpyrocarbonate and acetanhydride (data not shown). The action of some of these compounds probably consisted of the alterations in the protein and lipid milieu in which the membrane receptor is embedded rather than of these in the receptor itself. Even the

201

incubation of water-soluble LH/hCG receptors occurring in porcine follicular fluid with these chemicals did not result in any increase of nonspecific binding [Kolena et al. 1984]. It is obvious that substantially higher number of reagents or treatments may decrease a specific binding of hormone via a rise of nonspecific binding, but the reports on these findings are rare. In the case of testicular membrane an additional effect of unsaturated fatty acids on hormone receptors cannot be ruled out. In addition, such an effect may be related only to gonadal membranes and [125I]hCG binding. Ginsberg et al. [1981] observed that Fried erythroleukemia cells enriched in unsaturated fatty acids exhibit increased numbers of insulin receptors without any concomitant changes of nonspecific binding. In this context, it is interesting to note that the change occurring in rat testicular membranes under the influence of unsaturated fatty acids are not irreversible. The disappearance of specific binding of [125I]hCG by fluidizing action of oleic acid can be restored by a subsequent influence of cholesterol.

References Borochov H., Shinitzky M.: Vertical displacement of membrane proteins mediated by changes in microviscosity. Proc. Natl. Acad. Sci. USA, 73, 4526-4530, 1976. Catt K. J., Dufau M. L.: Peptide hormone receptors. Ann. Rev. Physiol. 39, 529-557, 1977. Dave J. R., Witorsch R. J.: Modulation of prolactin binding sites in vitro by membrane fluidizers. I. Effect on adult rat ventral prostatic membranes. Biochem. Biophys. Res. Commun. 113, 220-228, 1983. Fickova M., Macho L.: Insulin receptors in isolated adypocytes from rats with different neonatal nutrition. Endocr. Exper. 15, 259-268, 1981. Ginsberg B. H., Brown T. J., Simon I., Spector A. A.: Effect of the membrane lipid environment on the properties of insulin receptors. Diabetes 30, 773-780, 1981. Heron D.S., Shinitzky M., Hershkowitz M., Samuel D.: Lipid fluidity markedly modulates the binding of serotonin to mouse brain membranes. Proc. Natl. Sci., USA, 77, 7563-7467, 1980. Huang C.-H.: A structural model for the cholesterol-phosphatidylcholine complexes in bilayer membranes. Lipids 12, 348-356, 1977. Kolena J.: Binding of 1251-hCG by rat testis homogenates in early neonatal period. Endocr. Exper. 10, 113- 118, 1976. Kolena J., Kiss A., Channing C. P.: Purification of porcine granulosa cells by continuous Percoll gradient. Experientia 39, 908-909, 1983. Kolena J., Sebokova E.: Hormonal regulation of testicular LH/hCG receptors in rat. Exp. Clin. Endocr. 82, 1-7, 1983. Kolena J., Ondrias K.: Age-dependent changes in rat testicular LH/hCG receptors in relation to the membrane fluidity. Gen. Physiol. Biophys. 3, 89-92, 1984. Kolena J., Sebokova E., Horkovics-Kovats S.: Characterization of the soluble LH/hCG receptor in porcine follicular fluid. Physiol. Bohemoslov. 33, 540, 1984. Kolena J., Blazicek P., Horkovics-Kovats S., Ondrias K., Sebokova E.: Modulation of rat testicular LH/hCG receptors by membrane lipids fluidity, sub· mitted. Lands W. E. M., Hart P.: Metabolism of glycerolipids. J. Biol. Chem. 240, l! 05-1911, 1965.

202

Lowry 0. H., Rose bro ugh N. J., Farr A. L., Randall R. J.: Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265-275, 1951. Orly J., Schramm M.: Fatty acids as modulators of membrane functions: Catecholamine-activated adenylate cyclase of the turkey erythrocyte. Proc. Natl. Acad. Sci. USA, 72, 3433-3437, 1975. Shini tzky M., In bar M.: Microviscosity parameters and protein mobility in biological membranes. Biochim. Biophys. Acta 433, 133-149, 1976. Shinitzky M., Henkart P.: Fluidity of cell membrane!'! - current concepts and trends. Internat. Rev. Cytol. 60, 121-147, 1979. Strittmatter W. J., Hirata F., Axelrod J.: Phospholipid methylation unmasks cryptic beta-adrenergic receptors in rat reticulocytes. Science 204, 1204-1207, 1979.

203