- Email: [email protected]
The solubilisation of some steroids by phosphatidylcholine and phosphatidylcholine-cholesterol vesicles
Chemistry and Physics of Lipids, 24 (1979) 257-263 © Elsevier/North-Holland Scientific Publishers Ltd.
THE SOLUBILISATION OF SOME STEROIDS BY PHOSPHATIDYLCHOLINE AND PHOSPHATIDYLCHOLINE-CHOLESTEROL VESICLES
B. LUNDBERG
Department of Biochernistry and Pharmacy, .~bo AkademL SF-20500 ,~bo (Finland] Received December 20th, 1978
accepted March 8th, 1979
The solubility of the three steroid hormones, progesterone, testosterone, and estradiol-17~ in water and phosphatidylcholine vesicles was measured after shaking and ultrasonication. All three steroids have low water solubility, which increases considerably at sonication for testosterone and estradiol-17/L The phosphatidylcholine vesicles have a very small solubilising capacity for the steroids; about 20 ~tmol/mol. This increases at sonication for estradiol-17~ and.decreases for testosterone. The capacity for progesterone is almost unaltered. The incorporation of cholesterol in the vesicles decreased the solubilisation capacity for testosterone and estradiol-17# but increased that for progesterone of shaked preparations. For the sonicated systems the cholesterol decreased the solubilising capacity for estradiol-17~ but increased that for testosterone. The solubilisation experiments indicate that the steroid hormones are solubilised in the hydrocarbon part of the phosphatidylcholine bilayer and also 13C NMR results support this conclusion.
Introduction The mechanism o f steroid hormone action involves permeation through the plasma membrane and the binding to specific receptor proteins in the cytoplasm. It is usually assumed that steroids, like other lipophilic compounds, diffuse freely through cell membranes. In favour of this opinion are results which show that steroids enter target and nontarget tissues with equal facility at a rate directly proportional to steroid concentration [ 1,2]. On the other hand it has been shown that the steroid uptake is dependent on temperature and with increasing concentration o f steroid, saturation can be observed [3,4]. These results suggest that the translocation o f steroid through the plasma membrane is mediated by carrier proteins. It has been reported that pharmacologically-active steroids affect the membranes o f lysosomes, mitochondria, and erythrocytes [5,6]. In these systems steroids, which do not possess oxygen functions in the 11- or 17-position, appear to disrupt membranes while steroids which have such oxygen functions protect against disruption. This property o f steroids might be o f importance for their pharmacological 257
258
B. Lundberg, Steroid solubilisation by phosphatidylcholine
and physiological activity. Thus it is of interest to study the interaction of steroids with membrane lipids such as phospholipids and cholesterol. Ultrasonicated phospholipid-water systems have been much used as membrane models. The phospholipid aggregates in such systems, the so-called 'vesicles', can solubilise cholesterol in a 1 : 1, and under certain conditions 1 : 2, tool ratio between phospholipid and cholesterol [7,8]. Several authors have pointed out that a 3/~-hydroxy group and an intact side chain at the 17-position of the steroid nucleus are essential for maximal steroid-phospholipid interaction [9,10]. In accordance with this the steroid hormones which lack a C-17 side chain are solubilised to a much less extent than cholesterol [I 1,12]. Because of the higher water solubility [13] and lower solubisation capacity for the steroid hormones compared to cholesterol there is a great deal of scatter in the results obtained. Also the solubilisation mechanism of phospholipids for steroid hormones remains to be confirmed. It has been proposed that steroid molecules pack into a lipid leaflet so that one polar group is in the interface and the other in the hydrocarbon region [14]. By the surface balance technique it was not, however, possible to assess the orientation of the progesterone molecules in a lecithin film owing to the small quantities present in the film [15]. A proton NMR study indicates that the relative condensing activity for different steroids is inversely proportional to the molar volume of the incorporated steroid. This effect may be interpreted so that the steroids penetrate the phospholipid bilayer in a way similar to that of cholesterol [16,17].
Materials and methods The egg phosphatidylcholine used was a highly-purified product prepared in this laboratory [18]. The mean molecular weight of the phosphatidylcholines was 768, calculated from the fatty acid composition. The cholesterol used (Merck p.a.) was recrystallised three times from 1,2-dichlorethane and found to be chromatographically pure by GLC and TLC. The steroid hormones; progesterone, testosterone and estradiol-17/$, were purchased from Fluka and their purity were checked by TLC. [14C]phosphatidylcholine was purchased from New England Nuclear, Boston, MA and [3H]cholesterol and ~H-labelled progesterone, testosterone, and estradiol17/~ from the Radiochemical Centre, Amersham. They had a radiochemical purity of 98% by TLC on silica gel and were used without further purification. Deuterium oxide (approx. 99.8%) was obtained from Sigma. The buffer used was 0.1 M NaC1 in 0.01 M barbital sodium/HC1, (pH 8.0). The lipid mixtures were prepared by lyophilisation from benzene solutions in thin-walled glass ampoules. Then the buffer was added to give ritual phosphatidylcholine concentration of 1 mg/ml. The ampoules were riffled with N2 gas and tightly stoppered. The sonication was performed with a bath-type Branson Sonifier for 30 min at 25°C. The ultrasonicated preparations were clarified by
B. Lundberg, Steroid solubilisation by phosphatidylcholine
259
centrifugation at 50 000 g for 2 h at 25°C. After completed centrifugation aliquots of the supernatant solutions were carefully removed and the concentrations of phosphatidylcholine, cholesterol, and steroid hormones were determined by double-label liquid scintillation counting. The mechanical shaking was performed by rotation of the stoppered glass ampoules in a thermostated shaker at 25°C for 48 h. The solubilisation process was performed by two alternative methods. Either the phosphatidylcholine (-cholesterol) was sonicated first and then the resulting vesicles shaken with an excess of steroid hormone or the sonication was performed directly in the presence of an excess steroid hormone. The water solubility of the steroids after shaking and sonication of an excess of them in pure water was measured by liquid scintillation counting after centrifugation to remove undissolved material. 13C-NMR spectra were obtained with a Jeol FX-60 spectrometer operating in the Fourier transform mode. The sample contained 30 mg phosphatidylcholine/ml and were thoroughly outgassed, whereafter the tubes were filled with nitrogen and tightly stoppered.
Results The water solubilities of the steroid hormones; progesterone, testosterone, and estradiol-17/3, obtained at 25°C after shaking and sonication are presented in Table I. The val~es are mean values of 4 separate measurements. Testosterone has the largest water solubility and estradiol-17/3 the lowest. Sonication increases the water solubility of testosterone and estradiol-17/3 considerably. The results of the solubilisation of the steroid hormones (by shaking with egg phosphatidylcholine vesicles and by sonication of the phospholipid together with an excess of steroid) are shown in Table II. All solubilisation experiments in this study are mean values of 4 separate measurements, The lipid solubility of the TABLEI WATER SOLUBILITYOF STEROID HORMONES Solubility after shaking and sonication at 25°C in ~tmol/1.The standard deviations are given within brackets. Steroid
Progesterone Testosterone Estradiol-17#
Water solubility (#mol/1) Shaking
Sonication
33.0 (0.8) 86.7 (2.5) 11.7 (0.1)
36.7 (2.6) 156.0 (t.8) 22.5 (0.2)
B. Lundberg, Steroid solubilisation by phosphatidylcholine
260 TABLE II
SOLUBILISATION CAPACITY OF EGG PHOSPHATIDYLCHOLINE FOR STEROID HORMONES Solubilisation in #mol steroid per tool phospholipid. The standard deviations are given within brackets. Solubilisation capacity (#mol/mol)
Steroid
Progesterone Testosterone Estradiol-17~3
Shaking
Sonication
21.5 (0.9) 20.6 (0.4) 17.6 (0.7)
26.8 (0.6) 9.8 (1.0) 42.2 (2.5)
steroids was obtained by subtraction of the solubility in water from the solubility in the vesicle preparation. Also the simultaneous solubilisation of two steroids was tested with all possible combinations of the three steroids, but no significant change in solubilisation capacity could be detected.
40 lr'-'
~ ' ~ 30 u o c
o
E
m
~20 ~0
1
I
I
f
0.25
0.50 0.75 1.00 Mole ratio ch o L e s t e r o i / p h o s p h a t i d yt c h o l i n e Fig. 1. Effect of cholesterol on the solubilisation in phosphatidylcholine vesicles of steroid hormones by shaking; progesterone (o . -~), testosterone (a z), and estradiol-17# (t~~) or by sonication; progesterone (¢ --), testosterone ( A - -'), and estradiol- 17fl (u u).
B. Lundberg, Steroid solubilisation by phosphatidylcholine
261
The effect of increasing cholesterol concentration in the phosphatidylcholine vesicles on the solubilisation of steroids by shaking and by sonication are shown in Fig. 1. The solubilisation capacities are based on phosphatidylcholine concentration only. An incorporation of cholesterol in the vesicles has no drastic effect on the solubilisation of steroid hormones. At solubilisation by shaking there is a small increase in the progesterone values and a decrease in those of testosterone and estradiol-17/L The sonicated preparations show an increase for testosterone and a decrease for estradiol-17/3, while progesterone is constant. The solubilisation capacity of egg phosphatidylcholine for cholesterol in the presence of the steroid hormones was also tested. No significant difference from that found for systems without the steroid hormones was found [8]. In order to elucidate the solubilisation mechanism of phosphatidylcholine tor the steroid hormones a 13C NMR study of vesicles with and without steroids was performed. The ratio of the peak heights can be used to elucidate the relative condensing effect of steroids [16]. The values obtained for the ratios of the -(CH2)n-/+N(Me)3 peak heights were 1.10, 1.47, 1.16, and 1.18 for vesicles with no steroid, progesterone, testosterone, and estradiol-17/3, respectively. The results show a small increase in peak heights of the acyl methylenes compared to those of the choline N-methyls. The largest effect is shown by progesterone while that of the two other steroids is almost insignificant.
Discussion
The water solubility of the steroids obtained by centrifugation in this study is in good agreement with that found by fftitration [19]. The higher aqueous solubility of testosterone compared to that of progesterone is predictable from chemical considerations because the latter is rendered less hydrophilic by the extra two-carbon chain at C-17. The low aqueous solubility of estradiol-17/3 obviously must depend on the phenolic A ring. It has been proposed that a high lattice energy of estradiol may explain the low solubility [20]. The relative indifference of the water solubility of progesterone to sonication is in accordance with the slight effect of temperature on this [ 19]. The solubilisation capacity of egg phosphatidylcholine (by shaking) is almost the same for all three steroids. This fact is surprising if one considers the large difference obtained at solubilisation by surfactants [21,22]. Rather large differences in phospholipid solubility of progesterone and estradiol has been found by dialysis experiments [23]. By this technique also a reduction in lipid solubility of progesterone and an increase in that of estradiol in the presence of equimolar mixtures of both steroids were found. A similar behaviour was shown at solubilisation of steroids in surfactants and explained by differences in solubilisation mechanisms [22]. In this study no significant discrepancy between simultaneous and individual solubilisation could be detected.
262
B. Lundberg, Steroid solubilisation by phosphatidylcholine
The increased solubilisation at sonicatiQn, which is observed for estradiol-17/L has also been noted for cholesterol [8] and is obviously due to an alteration in the bilayer structure. The lower solubilisation o f testosterone at sonication is not easily explained but may depend on the increased water solubility of the steroid which alter the water-lipid partition. The sonication procedure is a common method for solubilisation in phospholipid vesicles [7,16]. It is, however, important to note that the ultrasonic irradiation may alter the solubilisation capacity o f vesicles for both uni- and bipolar steroids. If the phosphatidylcholine-steroid vesicles are in a metastable state from the thermodynamic point o f view [8] the solubilisation by shaking is the more appropriate method. The slight effect of cholesterol on the solubilisation of the steroid hormones indicates a different solubilisation mechanism for these. It is evident that cholesterol penetrates the phosphatidylcholine molecules in the bilayer [ 17] and it can thus be proposed that the steroid hormones are solubilised in the hydrocarbon region without penetration o f the fatty acid chains. This conclusion is supported by the laC NMR study which shows no condensing effect but rather an 'expanding' one based on the increased - ( C H 2 ) n - signal compared to that of+N(Me)3. This result is however, in conflict with that of a proton NMR study which showed a small condensing effect o f the steroid hormones [ 16]. It has also been proposed that steroid hormones which resemble cholesterol in that they contain a rigid polycyclic ring system should pack into a lipid leaflet in the same way as this and have specific structural effects on the bilayer system [14]. It is not justified to draw far-reaching conclusions from this study in regard to the biological action of steroid hormones. However, the low solubilising capacity o f phosphatidylcholine for the steroids does not indicate any appreciable accumulation of these in biological membranes. On the other hand the solubility of the steroids in the lipid bilayer may be large enough to admit passive diffusion through a membrane.
References 1 2 3 4 5 6 7 8 9 10
E. Gurpide and M. Welch, J. Biol. Chem., 244 (1969) 5159. G.G. Rousseau, J.D. Baxter, S.J. Higgins and G.M. Tomkins, J. Mol. Biol., 79 (1973) 539. H. Breuer, M.L. Rao and G.S. Rao, J. Steroid Biochem., 5 (1974) 359. M.L. Rao, G.S. Rao, M. H611erand H. Breuer, Hoppe Seyler's Z. Biol. Chem., 355 (1974) 1239. C. de Duve, R. Wattiaux and M. Wibo, Biochem. Pharmacol., 9 (1962) 97. G. Weissmann, Biochem. Phaxmacol., 14 (1965) 525. S.D. Hoyes and L. Saunders, Biochim. Biophys. Acta, 116 (1966) 184. B. Lundberg, Chem. Phys. Lipids, 18 (1977) 212. R.A. Demel, K.R. Bruckdorfer and L.L.M. Van Deenen, Biochim. Biophys. Acta, 255 (1972) 311,321. A. Darke, E.G. Finer, A.G. Flook and M.C. Phillips, FEBS Lett., 18 (1971) 326.
B. Lundberg, Steroid solubtlisation by phosphatidylcholine
263
11 J.W. Kellaway and L. Saunders, Biochim. Biophys. Acta, 144 (1967) 145. 12 M.M. Gale and L. Saunders, Bioehim. Biophys. Acta, 248 (1971) 466. 13 D.K. Madan and D.E. Cadwallader, J. Pharm. Sci., 62 (1973) 1567. 14 E.N. WiUmer, Biol. Rev. Cambridge Phil. Soc., 36 (1961) 368. 15 J.L. Taylor and D.A. Haydon, Biochirn. Biophys. Acta, 94 (1965) 488. 16 P. Ahmad and A. MeUors, J. Membrane Biol., 41 (1978) 235. 17 C. Huang, Lipids, 12 (1977) 348. 18 B. Lundberg, Acta Chem. Scand., 27 (1973) 3545. 19 B. Lundberg, Acta Pharm. Suec. in press. 20 K. Eik-Nes, J.A. ScheUman, R. Lumry and L.T. Samueles, J. Biol. Chem., 206 (1954) 411. 21 T. L6vgren, B. Heikius, B. Lundberg and L.S. SjO'blom, J. Pharm. Sci., 67 (1978) 1419. 22 B. Lundberg, T. L6vgren and B. Heikius, J. Pharm. Sci. in press. 23 R.B. Heap, A.M. Symons and J.C. Watkins, Biochim. Biophys. Acta, 233 (1971) 307.
THE SOLUBILISATION OF SOME STEROIDS BY PHOSPHATIDYLCHOLINE AND PHOSPHATIDYLCHOLINE-CHOLESTEROL VESICLES
B. LUNDBERG
Department of Biochernistry and Pharmacy, .~bo AkademL SF-20500 ,~bo (Finland] Received December 20th, 1978
accepted March 8th, 1979
The solubility of the three steroid hormones, progesterone, testosterone, and estradiol-17~ in water and phosphatidylcholine vesicles was measured after shaking and ultrasonication. All three steroids have low water solubility, which increases considerably at sonication for testosterone and estradiol-17/L The phosphatidylcholine vesicles have a very small solubilising capacity for the steroids; about 20 ~tmol/mol. This increases at sonication for estradiol-17~ and.decreases for testosterone. The capacity for progesterone is almost unaltered. The incorporation of cholesterol in the vesicles decreased the solubilisation capacity for testosterone and estradiol-17# but increased that for progesterone of shaked preparations. For the sonicated systems the cholesterol decreased the solubilising capacity for estradiol-17~ but increased that for testosterone. The solubilisation experiments indicate that the steroid hormones are solubilised in the hydrocarbon part of the phosphatidylcholine bilayer and also 13C NMR results support this conclusion.
Introduction The mechanism o f steroid hormone action involves permeation through the plasma membrane and the binding to specific receptor proteins in the cytoplasm. It is usually assumed that steroids, like other lipophilic compounds, diffuse freely through cell membranes. In favour of this opinion are results which show that steroids enter target and nontarget tissues with equal facility at a rate directly proportional to steroid concentration [ 1,2]. On the other hand it has been shown that the steroid uptake is dependent on temperature and with increasing concentration o f steroid, saturation can be observed [3,4]. These results suggest that the translocation o f steroid through the plasma membrane is mediated by carrier proteins. It has been reported that pharmacologically-active steroids affect the membranes o f lysosomes, mitochondria, and erythrocytes [5,6]. In these systems steroids, which do not possess oxygen functions in the 11- or 17-position, appear to disrupt membranes while steroids which have such oxygen functions protect against disruption. This property o f steroids might be o f importance for their pharmacological 257
258
B. Lundberg, Steroid solubilisation by phosphatidylcholine
and physiological activity. Thus it is of interest to study the interaction of steroids with membrane lipids such as phospholipids and cholesterol. Ultrasonicated phospholipid-water systems have been much used as membrane models. The phospholipid aggregates in such systems, the so-called 'vesicles', can solubilise cholesterol in a 1 : 1, and under certain conditions 1 : 2, tool ratio between phospholipid and cholesterol [7,8]. Several authors have pointed out that a 3/~-hydroxy group and an intact side chain at the 17-position of the steroid nucleus are essential for maximal steroid-phospholipid interaction [9,10]. In accordance with this the steroid hormones which lack a C-17 side chain are solubilised to a much less extent than cholesterol [I 1,12]. Because of the higher water solubility [13] and lower solubisation capacity for the steroid hormones compared to cholesterol there is a great deal of scatter in the results obtained. Also the solubilisation mechanism of phospholipids for steroid hormones remains to be confirmed. It has been proposed that steroid molecules pack into a lipid leaflet so that one polar group is in the interface and the other in the hydrocarbon region [14]. By the surface balance technique it was not, however, possible to assess the orientation of the progesterone molecules in a lecithin film owing to the small quantities present in the film [15]. A proton NMR study indicates that the relative condensing activity for different steroids is inversely proportional to the molar volume of the incorporated steroid. This effect may be interpreted so that the steroids penetrate the phospholipid bilayer in a way similar to that of cholesterol [16,17].
Materials and methods The egg phosphatidylcholine used was a highly-purified product prepared in this laboratory [18]. The mean molecular weight of the phosphatidylcholines was 768, calculated from the fatty acid composition. The cholesterol used (Merck p.a.) was recrystallised three times from 1,2-dichlorethane and found to be chromatographically pure by GLC and TLC. The steroid hormones; progesterone, testosterone and estradiol-17/$, were purchased from Fluka and their purity were checked by TLC. [14C]phosphatidylcholine was purchased from New England Nuclear, Boston, MA and [3H]cholesterol and ~H-labelled progesterone, testosterone, and estradiol17/~ from the Radiochemical Centre, Amersham. They had a radiochemical purity of 98% by TLC on silica gel and were used without further purification. Deuterium oxide (approx. 99.8%) was obtained from Sigma. The buffer used was 0.1 M NaC1 in 0.01 M barbital sodium/HC1, (pH 8.0). The lipid mixtures were prepared by lyophilisation from benzene solutions in thin-walled glass ampoules. Then the buffer was added to give ritual phosphatidylcholine concentration of 1 mg/ml. The ampoules were riffled with N2 gas and tightly stoppered. The sonication was performed with a bath-type Branson Sonifier for 30 min at 25°C. The ultrasonicated preparations were clarified by
B. Lundberg, Steroid solubilisation by phosphatidylcholine
259
centrifugation at 50 000 g for 2 h at 25°C. After completed centrifugation aliquots of the supernatant solutions were carefully removed and the concentrations of phosphatidylcholine, cholesterol, and steroid hormones were determined by double-label liquid scintillation counting. The mechanical shaking was performed by rotation of the stoppered glass ampoules in a thermostated shaker at 25°C for 48 h. The solubilisation process was performed by two alternative methods. Either the phosphatidylcholine (-cholesterol) was sonicated first and then the resulting vesicles shaken with an excess of steroid hormone or the sonication was performed directly in the presence of an excess steroid hormone. The water solubility of the steroids after shaking and sonication of an excess of them in pure water was measured by liquid scintillation counting after centrifugation to remove undissolved material. 13C-NMR spectra were obtained with a Jeol FX-60 spectrometer operating in the Fourier transform mode. The sample contained 30 mg phosphatidylcholine/ml and were thoroughly outgassed, whereafter the tubes were filled with nitrogen and tightly stoppered.
Results The water solubilities of the steroid hormones; progesterone, testosterone, and estradiol-17/3, obtained at 25°C after shaking and sonication are presented in Table I. The val~es are mean values of 4 separate measurements. Testosterone has the largest water solubility and estradiol-17/3 the lowest. Sonication increases the water solubility of testosterone and estradiol-17/3 considerably. The results of the solubilisation of the steroid hormones (by shaking with egg phosphatidylcholine vesicles and by sonication of the phospholipid together with an excess of steroid) are shown in Table II. All solubilisation experiments in this study are mean values of 4 separate measurements, The lipid solubility of the TABLEI WATER SOLUBILITYOF STEROID HORMONES Solubility after shaking and sonication at 25°C in ~tmol/1.The standard deviations are given within brackets. Steroid
Progesterone Testosterone Estradiol-17#
Water solubility (#mol/1) Shaking
Sonication
33.0 (0.8) 86.7 (2.5) 11.7 (0.1)
36.7 (2.6) 156.0 (t.8) 22.5 (0.2)
B. Lundberg, Steroid solubilisation by phosphatidylcholine
260 TABLE II
SOLUBILISATION CAPACITY OF EGG PHOSPHATIDYLCHOLINE FOR STEROID HORMONES Solubilisation in #mol steroid per tool phospholipid. The standard deviations are given within brackets. Solubilisation capacity (#mol/mol)
Steroid
Progesterone Testosterone Estradiol-17~3
Shaking
Sonication
21.5 (0.9) 20.6 (0.4) 17.6 (0.7)
26.8 (0.6) 9.8 (1.0) 42.2 (2.5)
steroids was obtained by subtraction of the solubility in water from the solubility in the vesicle preparation. Also the simultaneous solubilisation of two steroids was tested with all possible combinations of the three steroids, but no significant change in solubilisation capacity could be detected.
40 lr'-'
~ ' ~ 30 u o c
o
E
m
~20 ~0
1
I
I
f
0.25
0.50 0.75 1.00 Mole ratio ch o L e s t e r o i / p h o s p h a t i d yt c h o l i n e Fig. 1. Effect of cholesterol on the solubilisation in phosphatidylcholine vesicles of steroid hormones by shaking; progesterone (o . -~), testosterone (a z), and estradiol-17# (t~~) or by sonication; progesterone (¢ --), testosterone ( A - -'), and estradiol- 17fl (u u).
B. Lundberg, Steroid solubilisation by phosphatidylcholine
261
The effect of increasing cholesterol concentration in the phosphatidylcholine vesicles on the solubilisation of steroids by shaking and by sonication are shown in Fig. 1. The solubilisation capacities are based on phosphatidylcholine concentration only. An incorporation of cholesterol in the vesicles has no drastic effect on the solubilisation of steroid hormones. At solubilisation by shaking there is a small increase in the progesterone values and a decrease in those of testosterone and estradiol-17/L The sonicated preparations show an increase for testosterone and a decrease for estradiol-17/3, while progesterone is constant. The solubilisation capacity of egg phosphatidylcholine for cholesterol in the presence of the steroid hormones was also tested. No significant difference from that found for systems without the steroid hormones was found [8]. In order to elucidate the solubilisation mechanism of phosphatidylcholine tor the steroid hormones a 13C NMR study of vesicles with and without steroids was performed. The ratio of the peak heights can be used to elucidate the relative condensing effect of steroids [16]. The values obtained for the ratios of the -(CH2)n-/+N(Me)3 peak heights were 1.10, 1.47, 1.16, and 1.18 for vesicles with no steroid, progesterone, testosterone, and estradiol-17/3, respectively. The results show a small increase in peak heights of the acyl methylenes compared to those of the choline N-methyls. The largest effect is shown by progesterone while that of the two other steroids is almost insignificant.
Discussion
The water solubility of the steroids obtained by centrifugation in this study is in good agreement with that found by fftitration [19]. The higher aqueous solubility of testosterone compared to that of progesterone is predictable from chemical considerations because the latter is rendered less hydrophilic by the extra two-carbon chain at C-17. The low aqueous solubility of estradiol-17/3 obviously must depend on the phenolic A ring. It has been proposed that a high lattice energy of estradiol may explain the low solubility [20]. The relative indifference of the water solubility of progesterone to sonication is in accordance with the slight effect of temperature on this [ 19]. The solubilisation capacity of egg phosphatidylcholine (by shaking) is almost the same for all three steroids. This fact is surprising if one considers the large difference obtained at solubilisation by surfactants [21,22]. Rather large differences in phospholipid solubility of progesterone and estradiol has been found by dialysis experiments [23]. By this technique also a reduction in lipid solubility of progesterone and an increase in that of estradiol in the presence of equimolar mixtures of both steroids were found. A similar behaviour was shown at solubilisation of steroids in surfactants and explained by differences in solubilisation mechanisms [22]. In this study no significant discrepancy between simultaneous and individual solubilisation could be detected.
262
B. Lundberg, Steroid solubilisation by phosphatidylcholine
The increased solubilisation at sonicatiQn, which is observed for estradiol-17/L has also been noted for cholesterol [8] and is obviously due to an alteration in the bilayer structure. The lower solubilisation o f testosterone at sonication is not easily explained but may depend on the increased water solubility of the steroid which alter the water-lipid partition. The sonication procedure is a common method for solubilisation in phospholipid vesicles [7,16]. It is, however, important to note that the ultrasonic irradiation may alter the solubilisation capacity o f vesicles for both uni- and bipolar steroids. If the phosphatidylcholine-steroid vesicles are in a metastable state from the thermodynamic point o f view [8] the solubilisation by shaking is the more appropriate method. The slight effect of cholesterol on the solubilisation of the steroid hormones indicates a different solubilisation mechanism for these. It is evident that cholesterol penetrates the phosphatidylcholine molecules in the bilayer [ 17] and it can thus be proposed that the steroid hormones are solubilised in the hydrocarbon region without penetration o f the fatty acid chains. This conclusion is supported by the laC NMR study which shows no condensing effect but rather an 'expanding' one based on the increased - ( C H 2 ) n - signal compared to that of+N(Me)3. This result is however, in conflict with that of a proton NMR study which showed a small condensing effect o f the steroid hormones [ 16]. It has also been proposed that steroid hormones which resemble cholesterol in that they contain a rigid polycyclic ring system should pack into a lipid leaflet in the same way as this and have specific structural effects on the bilayer system [14]. It is not justified to draw far-reaching conclusions from this study in regard to the biological action of steroid hormones. However, the low solubilising capacity o f phosphatidylcholine for the steroids does not indicate any appreciable accumulation of these in biological membranes. On the other hand the solubility of the steroids in the lipid bilayer may be large enough to admit passive diffusion through a membrane.
References 1 2 3 4 5 6 7 8 9 10
E. Gurpide and M. Welch, J. Biol. Chem., 244 (1969) 5159. G.G. Rousseau, J.D. Baxter, S.J. Higgins and G.M. Tomkins, J. Mol. Biol., 79 (1973) 539. H. Breuer, M.L. Rao and G.S. Rao, J. Steroid Biochem., 5 (1974) 359. M.L. Rao, G.S. Rao, M. H611erand H. Breuer, Hoppe Seyler's Z. Biol. Chem., 355 (1974) 1239. C. de Duve, R. Wattiaux and M. Wibo, Biochem. Pharmacol., 9 (1962) 97. G. Weissmann, Biochem. Phaxmacol., 14 (1965) 525. S.D. Hoyes and L. Saunders, Biochim. Biophys. Acta, 116 (1966) 184. B. Lundberg, Chem. Phys. Lipids, 18 (1977) 212. R.A. Demel, K.R. Bruckdorfer and L.L.M. Van Deenen, Biochim. Biophys. Acta, 255 (1972) 311,321. A. Darke, E.G. Finer, A.G. Flook and M.C. Phillips, FEBS Lett., 18 (1971) 326.
B. Lundberg, Steroid solubtlisation by phosphatidylcholine
263
11 J.W. Kellaway and L. Saunders, Biochim. Biophys. Acta, 144 (1967) 145. 12 M.M. Gale and L. Saunders, Bioehim. Biophys. Acta, 248 (1971) 466. 13 D.K. Madan and D.E. Cadwallader, J. Pharm. Sci., 62 (1973) 1567. 14 E.N. WiUmer, Biol. Rev. Cambridge Phil. Soc., 36 (1961) 368. 15 J.L. Taylor and D.A. Haydon, Biochirn. Biophys. Acta, 94 (1965) 488. 16 P. Ahmad and A. MeUors, J. Membrane Biol., 41 (1978) 235. 17 C. Huang, Lipids, 12 (1977) 348. 18 B. Lundberg, Acta Chem. Scand., 27 (1973) 3545. 19 B. Lundberg, Acta Pharm. Suec. in press. 20 K. Eik-Nes, J.A. ScheUman, R. Lumry and L.T. Samueles, J. Biol. Chem., 206 (1954) 411. 21 T. L6vgren, B. Heikius, B. Lundberg and L.S. SjO'blom, J. Pharm. Sci., 67 (1978) 1419. 22 B. Lundberg, T. L6vgren and B. Heikius, J. Pharm. Sci. in press. 23 R.B. Heap, A.M. Symons and J.C. Watkins, Biochim. Biophys. Acta, 233 (1971) 307.