70
BBALIP
Biochimica et Biophysics Acta, 1047 (1990) 70-76 Elsevier
53501
Retinyl ester hydrolase and vitamin A status in rats treated with 3,3’,4,4’-tetrachlorobiphenyl Muriel Mercier, Gbard Laboratoire
Pascal and VQonique
de Nutrition et Skuritk
Alimentaire,
(Received
Key words:
Retinyl ester hydrolase;
INCA-CRJ,
Azais-Braesco
Jouy-enJosas
(France)
17 May 1990)
Vitamin
A; Tetrachlorobiphenyl;
(Rat)
Previous studies have shown that rats exposed to 3,3’,4,4’-tetrachlorohiphenyl (TCB) exhibit decreased liver vitamin A stores. The activity of retinyl ester hydrolase (REH), the enzyme responsible for the hydrolysis of the storage form of vitamin A (retinyl esters) into free retinol, may therefore be altered by TCB. This study was carried out to investigate the effect of TCB on vitamin A distribution and on REH activity in the rat. REH activity was measured in liver homogenates and microsomes (650 pg protein), in Tris-maleate buffer 0.1 M at pH 7.2 in the presence of 150 mM CHAPS and 1.5 mM retinyl palmitate dispersed in Triton X-100 0.2%. Using these conditions, the kinetic parameters of the enzyme were determined and the inter-animal variation coefficient (10%) allowed statistical comparisons between experimental groups. Male Wistar rats of sufficient or deficient vitamin A status were treated IP with 340 pmol of TCB/kg. Vitamin A levels were significantly depressed in liver. REH activity was decreased about 20%, and serum retinol was decreased about 50%, independent of the initial vitamin A status of the animals. Vitamin A levels in lungs and testes were also decreased, suggesting that TCB could interfere with vitamin A delivery to target organs. The negative effect of TCB on REH activity in vivo was also observed when TCB was added in vitro to the incubation medium at concentrations near to those expected after in vivo treatment. TCB is a non-competitive inhibitor of retinyl palmitate hydrolase.
Introduction Vitamin A deficiency is one of the main public health problems in numerous developing countries, where xerophtalmia leads to irreversible blindness for several hundreds of thousands of children each year and is an important factor of mortality for thousands of others [1,2]. In industrialized areas, such acute problems do not appear, but several epidemiological studies show that important segments of the population (children, young women, elderly people) do not fulfill their needs by their dietary intake of vitamin A or precursors [3]. The health effects of the resulting sub-optimal levels of vitamin A are not precisely known, but one may think that such individuals are more vulnerable to external factors. For example, vitamin A is thought to have a
Abbreviations: BHT, butylated hydroxytoluene; CHAPS, 3[(3cholamidopropyl)-dimethylammoniol-l-propane sulfonate; IP, intraperitoneally; PCBs, polychlorinated biphenyls; RBP, retinol-binding protein; RE, retinol equivalent; TCB, 3,3’,4,4’-tetrachlorobiphenyl. Correspondence: V. Azais-Braesco, Laboratoire de Nutrition Skcurite Alimentaire INRA-CRJ, 78350 Jouy-en-Josas, France.
00052760/90/$03.50
0 1990 Elsevier Science Publishers
et
B.V. (Biomedical
protective effect against chemical-induced carcinogenesis, as has been suggested by epidemiological studies in humans [4,5], and experimental investigations in animals [6]. The recent discovery of several nuclear receptors for retinoic acid, a first step metabolite of retinol, seems to indicate that vitamin A could trigger the expression of various genes encoding for proteins involved in cellular differentiation [7]. Besides inadequate dietary supply, vitamin A deficiency could arise from other sources, such as infectious diseases [8], or exposure to foreign compounds. Numerous xenobiotics lead to diminution of the liver vitamin A stores, and among them, polychlorinated biphenyls (PCBs) which are ubiquitous and persistent pollutants of food and environment. PCBs have been shown to alter vitamin A metabolism and distribution [9-111 and lead to a reduction in vitamin A stores. Among the possible explanations for such interactions between vitamin A and xenobiotics, one could hypothesize that some step of vitamin A metabolism is modified by the xenobiotic. When the liver stores are decreased, the storage form of vitamin A (retinyl esters), which normally represent 95% of the hepatic content, must be diminished. Such mobilization should involve an enDivision)
71 hanced hydrolysis of retinyl esters into free retinol. The specificity of the reaction is not clear and several enzymes, namely carboxylesterases, are able to catalyse such a reaction in vitro [12]. However, until recently, the retinyl palmitate hydrolase (RPH, EC 3.1.1.21) was the only one shown to be active in vivo. According to early investigators, this enzyme seemed to exhibit unusual characteristics, in particular a very high variation coefficient and a puzzling subcellular distribution [13]. During the past two years, the same hydrolysis reaction was assayed by a different method by two authors [14,15] who assigned new characteristics to the enzyme, especially for the optimum pH and the variation coefficient. This enzyme was called Retinyl Ester Hydrolase (REH); however, it is not clear if its difference with RPH comes from the assay technique or if they really are different enzymes. Last year, Harrison and Gad [16] described the existence of a neutral, bile salt-independent retinyl ester hydrolase that seems to be different from REH or RPH. In the first part of this study, we have determined the assay conditions for measuring REH activity in rat liver homogenate, displaying higher activity and the lowest variation coefficient. We then investigated the effect of a pure polychlorinated biphenyl compound on REH activity in rat liver. We chose the PCB 3,3’,4,4’-tetrachlorobiphenyl (TCB), a compound whose effects on vitamin A distribution (decrease in liver and serum) have been previously described [17]. Materials and Methods Chemicals TCB was synthesized, purified and characterized as previously described [17] and was greater than 99% pure. Retinoids used for diet supplementation, as calibration standards for high-performance liquid chromatography (HPLC) or as substrate for REH were kind gifts from Hoffmann-La Roche (Neuilly, France). [11,12-3H]Retinol (40-60 Ci/mmol) was obtained from New England Nuclear (Paris, France). HPLC solvents, HPLC grade, were purchased from Prolabo (Paris, France). All other chemicals were obtained from Sigma and were of the highest grade available. Animals and treatments All the animals used in this study were male Wistar rats raised in our animal colony; they were kept at 20 k 2” C, with a 12 h light cycle and had free access to food and water; they were killed after they had been fasted overnight. For determination of the conditions for assaying REH activity, the rats were fed a standard commercial mixture (Pietrement, France), containing 10000 IU (international units) of vitamin A per kg; they weighed 150 g at the time of killing. Under ether anesthesia, the liver was perfused with ice-cold Tris-HCl
buffer 0.15 M (pH 7.4), then removed, homogenized in 9 volumes of ice-cold Tris-Maleate buffer 0.05 M (pH 7.4) filtered through gauze, then frozen in liquid nitrogen and stored at - 80 o C. For studying the effects of TCB exposure, a semisynthetic diet (48% corn starch, 24% sucrose, 18% delipidated-casein supplemented with D,L-methionine, 3% soybean oil, 2% cellulose, 4% salt mixture and 1% vitamin mixture without vitamin A), essentially free of vitamin A was used. The amounts of retinyl palmitate needed to obtain diets of differing vitamin A contents were added. Rats were born from mothers fed a diet containing only 800 IU vitamin A/kg, a concentration that allows normal pregnancy, delivery and milking but that leaves the pups with very little liver vitamin A. After weaning, rats were maintained for 1 week on a vitamin A-free diet to assure further depletion of liver vitamin A stores. Less than 8 RE/liver were detected in six rats killed for checking purposes. The animals were then separated into two groups: one was fed a diet containing 25000 IU vitamin A/kg (vitamin A-sufficient group) and the other received a diet containing 800 IU vitamin A/kg (vitamin A-deficient group). After 18 days, each group was again divided in two, one half receiving an IP injection of TCB dissolved in corn oil (340 pmol/kg) and the other half receiving corn oil alone via the same route. 6 days after treatment, blood was collected by ocular puncture and serum was obtained by centrifugation (15 min, 1400 g) after clotting at 4“ C in the dark. Liver was perfused as described above, then separated into three parts, one for vitamin A determination (homogenized in 9 volumes of sucrose 0.25 M, KC1 0.025 M, MgCl, 0.05 M, buffered with Tris-HCl 0.05 M (pH 7.4), one for preparation of microsomes (homogenized in 4 volumes of sucrose 0.25 M, EDTA 1 mM buffered with Tris 0.05 M, then processed according Omura and Sato [18]) and one for REH activity determination (homogenized in 9 volumes of Tris-maleate buffer 0.05 M, containing 400 mg/l ascorbic acid). Testes and lungs were excised and processed for vitamin A determination as for the liver samples, except that they were homogenized in 4 volumes of buffer. All samples were frozen in liquid nitrogen and stored at - 80 o C until analysis. Vitamin A determinations Proteins in 1 ml of the homogenate were precipitated by an equal volume of ethanol. A known amount of retinyl propionate as internal standard was added, and then extracted twice with 2 ml of hexane containing butylated hydroxytoluene (BHT); the hexane phases were evaporated under nitrogen and the extract was dissolved in 400 ~1 of methanol. Vitamin A was determined by reverse-phase HPLC on a Kontron apparatus equipped with a Nucleosil column (25 X 0.39
72 cm, pore diameter: 5 pm). Using pure methanol as eluant and a flow rate of 2 ml/mm, elution was achieved in 2.8 min for retinol, 3.9 min for retinyl propionate and 19.2 min for retinyl palmitate. Ultraviolet detection was done at 330 nm and results calculated using both internal and external standards on an Anacomp computer (Kontron).
retinol measured. When necessary, tritiated retinol (1 pCi/incubation) or TCB was added in 5 ~1 methanol. Proteins. Proteins were determined according to Lowry et al. [19]. Statistical analysis. All data are expressed as means + standard deviation. Statistical analysis were done by variance analysis and Student-Newman-Keuls’s test [20].
REH assays Conditions for measuring rat liver REH activity were adapted from Cooper et al. [14] and from Friedman et al. [15]. An aliquot of liver homogenate or microsomes (650 pg of protein) was incubated in Tris-maleate buffer 0.1 M (pH 7.2) at 37 o C in a shaking water bath (90 movements/mm) for 90 min in presence of 150 mM CHAPS and 1.5 mM retinyl palmitate dispersed in 0.2% Triton X-100. The final volume was 0.5 ml and the reaction was terminated by the addition of 0.5 ml ethanol containing BHT (50 mg/l). Retinol in the incubates was monitored following the procedure used for vitamin A determinations. Control incubations without homogenate allowed correction for non-enzymatic retino1 formation. Endogenous retinol from the homogenates was verified to represent less than 0.1% of the
Results
REH ectlvlty 20 - fnmol rellnol
REH
/Ch.O.S
ml lncubntlon
25
mcdlum)l
i/_ A
; j,_--
16-
__~--
600 Protelno
REH
activity
lnmol
retinoVmg
600
inmol
retlnol/(h.mq
prot)]
:
1000
1200
01’ 4
Ii
” 4.5
5
5.5
c
6.5
(ug)
proteid
““” 7
7.5
0
a.5
9
9.5
10
PH
REH
36r
40 lnoubatlon
actlvlty
_-
I
400
200
r
;
01
0
Determination of assay conditions for REH activity The optimal conditions were determined from the data shown in Fig 1. Protein concentration (650 pg/O.5 ml) and incubation time (90 min) were in the linear part of the curve. Optimal pH was shown to be 7.2. CHAPS and substrate concentrations were chosen after variance analysis to give simultaneously a maximal enzyme activity and a minimized variation coefficient. Under these conditions, kinetic parameters for REH activity were determined and we found an apparent K, of 1.34 mM and a maximal velocity of 28 nmol of retinol formed/h (Fig. 2). Although the rate increase of the specific activity of REH is diminishing, a plateau is still not reached for a substrate concentration of 8 mM (Fig. 3).
actlvlty
Inmol
retlnol/(hmg
prot)l
00
tlmc fmln)
Fig. 1. Determination of the assay conditions for measuring REH activity. Incubations were carried out at 37 o C, in a shaking water bath, in a final volume of 0.5 ml in Tris-maleate buffer 0.1 M. Upper left panel (A): effect of protein concentration. Upper right panel (B): effect of pH. Lower left panel (C): effect of the incubation duration. Lower right panel (D): effect of the CHAPS concentration. Each point represents the mean of six different incubations and has been corrected for non-enzymatic blank.
73 l
0.4
z
TCB
TABLE
P
r
I
Effect of substrate (retinyl palmitate) concentration on variation coefficient
0.3-
c
Incubations have been carried out as described in the Material and Methods section. Variation coefficient has been calculated as the ratio: (S.D./mean) X 100, n = 6.
2 - TCB
VLSI Fig. 2. Kinetic ting). Inhibition
REH
2
15
1
0.5
0
-0.5
(l/mM)
parameters of REH activity (Lineweaver-Burk plotof REH activity by TCB. Each point represents the mean of six different determinations.
activity lnmol
mg proleinl
retinal/h
50
Substrate concentration
Variation coefficient
(mM)
(W)
0.5 1 1.5 2 2.5 3 5 8
12 8 7 12 21 21 29 60
in the same homogenates on different days was found to be 4% (data not shown). The activity of the enzyme stored at - 80°C was stable for up to 6 months (less than 2% variation). Table I shows the increase of the variation coefficient with increased substrate concentration. I
OV
0
10
8
2
Fig. 3. Effect of retinyl pahnitate concentration on REH activity. Rat liver homogenates were incubated as described in the Material and Methods section, with the indicated retinyl palmitate concentration. Each point is the mean of REH activity on six different rats. Variation coefficients are given on Table I.
It was not possible to study the effect of higher substrate concentrations because, above a concentration of 10 mM, retinyl palmitate does not dissolve in 0.2% Triton X-100. From a concentration of 0.5 mM, the substrate was never rate-limiting, since more than 90% was found at the end of incubation. The reproducibility of the assay, checked by measuring the enzyme activity TABLE
Vitamin A content of the organs (Table II) As expected, the vitamin A status was very dependent on the amount of vitamin A in the diets. The rats fed 25000 IU vitamin A/kg exhibited a normal distribution of vitamin A: (1) high hepatic stores, with more than 98% in esterified form; (2) normal serum retinol (according to data of various authors [11,21]); and (3) small amounts of vitamin A in extra hepatic organs, with approx. 80% in esterified form in target organs (lungs and testes). On the other hand, the animals receiving the 800 IU vitamin A/kg showed a deficient status characterized by: (1) much lower liver vitamin A content, exclusively as free retinol; (2) decreased serum retinol (-SOW), a
II
Serum retinol concentration and vitamin A treatment
(retinol+ retinyl palmitate) content of organs of rats of different vitamin A status with and without TCB
Results are expressed in retinol equivalent (1 RE =l jtg of retinol or 0.55 pg of retinyl palm&ate). R is the ratio: palmitate) X 100. Values in the same column, with different superscripts differ significantly (P < 0.05) -(n = 6). Vitamin
A content
liver
(R.E.)
R
R
lungs
R
testes
(free retinol/retinol+retinyl
Retinol concentration (R.E./ml) serum
25 000 IU/kg
diet
Control TCB-treated 800 IU/kg
61-l 339
590” f90b
1.1 2.4
2.47+1.51 = 1.47 f 0.59 b
16 15
1.65 * 1.06 a 0.91+ 0.30 Gb
21 16
0.67 f 0.10 = 0.31+0.10 b
0.08 f 0.03 ’ 0.03 f 0.02 c
96 96
0.33 f 0.05 b 0.12 f 0.06 b
40 60
0.32 _+0.03 b 0.10*0.02 =
diet
Control TCB-treated
2.55+ 1.72 f
0.85 ’ 0.57 d
100 100
74 clear symptom of vitamin A deficiency [22]; and (3) drastic diminution in the target organs. Vitamin A had nearly disappeared from lungs (- 97%) and only remained as free retinol, whereas retinyl palmitate still represented 60% of the vitamin A in the testes although the total quantity decreased by 80%. TCB treatment led to significant modifications of vitamin A levels, which do not seem to depend on the nutritional status of the animals. Liver stores were diminished by the exposure to TCB ( - 50% in the rats fed the normal-vitamin A diet and -33% in the vitamin A-deficient animal). Important losses of vitamin A were observed in testes and lungs and the circulating retinol levels were significantly diminished. The relative proportion of retinol and retinyl esters was not altered. In the vitamin A-sufficient animals, vitamin A was determined in the microsomal and cytosolic fraction. A significant decrease of 52% in the microsomes (0.081 k 0.025 ER/mg protein in control rats vs. 0.039 & 0.036 ER/mg protein in TCB-treated animals) and 77% in the cytosols (0.347 k 0.28 ER/mg protein vs. 0.078 &-0.12 ER/mg protein) occurred following TCB administration. REH activity in liver Vitamin A status did not modify the activity of REH in liver homogenates (Table III). TCB treatment induced a diminution of REH activity by 18% in both dietary groups, which was significant in the vitamin A-deficient group only. In the microsomal fraction, where the enzyme activity is greatly enhanced on a mg of protein basis, TCB also diminished REH activity. Effect of TCB on REH activity in When added to the incubation solved in methanol inhibited the dose-dependent way, and behaved tive inhibitor for retinol towards 1.05 IJM (Fig. 2). Methanol alone
TABLE
vitro medium, TCB disenzyme activity in a like a non-competiREH, with a Ki of had no effect. When
III
REH activity in liver and hepatic microsomes A status with and without TCB treatment
of rats of different vitamin
In the same column, values with different cantly (P < 0.05 - n = 6). n.d., not done.
superscripts
REH activity (nmol retinal/h
differ
mg proteins)
liver
hepatic
25 000 IV/kg diet Control TCB-treated
microsomes
15.22 f 2.69 a,b.c 12.48 f 1.26 a,b
49.42 + 15.05 a 39.50+ 9.21 b
800 IU/kg diet Control TCB-treated
16.44 f 2.78 ’ 13.59& 2.58 b
n.d. n.d.
signifi-
incubations were carried out with tritiated retinol, all the radioactivity extracted after 90 min of incubation (80% of the initial activity) was found in the retinol peak, both in the presence or in the absence of TCB (data not shown). This indicates that the inhibition of REH activity by TCB is not an artefact due to a non-enzymic degradation of newly formed retinol, as suggested by other authors [23]. Discussion This study was designed to investigate the role of REH in the modifications of vitamin A distribution in animals treated with TCB. Such alterations in vitamin A metabolism have been reported previously [9-11,171, and occur after exposure to a great variety of xenobiotics, such as pesticides [24], alcohol [25], oral contraceptives [26] and drugs [27]. The common feature observed is a decrease in the liver vitamin A stores that was confirmed in the present study, but we found that extrahepatic organs were also involved. It is interesting to observe that vitamin A in target organs (lungs and testes) was drastically decreased in TCB-treated rats although the liver stores, in spite of a 50% diminution, are still abundant enough to theoretically ensure a normal vitamin A concentration. Alterations in the mechanisms of retinol transport and delivery may have occurred following TCB exposure. Alterations in the system of retinol delivery to target organs are more likely. The assay technique proposed by early investigators [13] to determine REH activity was based upon the detection of free radioactive palmitic acid liberated from radioactive retinyl palmitate. The reproducibility of the assay and especially the very high variation coefficients reported were not easily compatible with experimental investigations. Moreover, the authors did not give any detail on the substrate concentrations nor on kinetic parameters of the enzyme. Recently, Cooper et al. [28] reported a more convenient assay technique, displaying low variation coefficients when pig liver homogenates were used, but still a 42% variation when rat liver homogenates were assayed [14]. Friedman and co-workers [15] recently published a study in which they found a variation coefficient of only 13%. Our modifications to the initial technique are similar and we find a variation coefficient of 7%. In studies of REH activity the difficulty of determining saturating substrate concentration arises. Friedman and co-workers did not give any data for REH activity with a substrate concentration above that which they found to be saturating, i.e., 900 PM. Moreover, while using the same assay conditions as Cooper et al. [14] and ourselves, they found much lower kinetic parameters [apparent K, = 0.325 . mg protein)] than we or mM and V,,, = 7.7 nmol/(h Cooper et al. did (K, = 1.3 mM). The latter author investigated the effect of substrate concentration up to
75 5 mM, which gave a REH activity of 21 nmol/(h . mg protein), which was not saturating. This body of experimental data leads one to think that there might exist some peculiarity of REH that could prevent the attainment of a plateau of activity while increasing the substrate concentration. 95% of hepatic vitamin A is stored in lipid droplets of fat storing cells where it is unable to come into contact with the enzyme which is, as shown on Table III, concentrated in the microsomal membrane. Hendriks et al. [29] showed that RPH was present in fat-storing cells and preliminary results of our laboratory indicate that REH can also be detected as well in these cells by our assay technique (unpublished data). It is likely that in vivo some mechanism that favors the reaction occurs. Bile salts might be involved, or a substrate carrier, possibly a fatty acid binding protein. This could facilitate the reaction, as has been recently shown for cellular retinol-binding protein which enhances the esterification of retinol by acyl : CoA acrylretinol transferase [30]. Although the assay conditions for measuring REH activity are not completely satisfactory and need further improvements, one could estimate that the data obtained are representative of the activity of the enzyme in the liver of rats having received different treatments. TCB exposure leads to a depletion of liver retinyl ester stores. One may have thought that REH, the enzyme responsible for the necessary hydrolysis of retinyl esters into retinol, would then present an enhanced activity. On the contrary, our results show a diminution of REH activity in the liver homogenates or microsomes of TCB-treated rats. Investigating TCB-treated female Sprague-Dawley rats, Powers and co-workers [31] observed a similar dose-dependent decrease in REH activity. It must be pointed out that TCB leads to the same decrease in the liver of vitamin A-deficient rats, which are free of retinyl esters. Togehter with the fact that vitamin A status has no effect on REH activity, this indicates that this enzyme is not regulated by the amount of its substrate. In vitro studies were carried out with TCB added to the incubation medium at a concentration of 1.7 PM, which had been estimated to be in the range of magnitude of the rat liver concentration, 1 week following the injection of 340 pmol/kg [32]. This amount of TCB leads to a diminution of the V,,, of REH activity, while the K, is not modified: this indicates that TCB behaves like a non-competitive inhibitor, whose Ki was determined to be 1.05 FM. The addition of such an amount of TCB in the incubation mixture of REH leads to a decrease of REH activity of 50%. Besides the uncertainty in the determination of the TCB concentration added in vivo, this stronger TCB effect in vitro could be explained by the metabolisation and the distribution of TCB. Recent studies [33] indicate that TCB
undergoes metabolic detoxification by hydroxylation and the metabolites can account for the decrease in REH activity observed. Moreover, in vivo, TCB is carried by lipoproteins, like related chlorinated hydrocarbons [34], or by a cytosolic receptor [35]. It is then possible that the TCB molecule itself (or a metabolite) has inhibiting properties towards REH, binding a specific site of the protein. This could explain the observed decrease in serum retinol that does not occur in animals treated with other vitamin A-depleting compounds. For example, alcohol, which decreases liver vitamin A, was recently shown to increase REH activity in vitro [15], and we have obtained similar results with our experimental conditions (data not shown). Jensen and co-workers [36] described a decrease of REH activity in livers from hexabromobiphenyl-treated rats, not related to a diminution of circulating levels of vitamin A. However, this compound was inactive in vitro, possibly because this xenobiotic requires metabolism to exert its toxic properties and to bind REH. In vivo and in vitro studies on the effects of various xenobiotic on REH activity would be needed to determine the role of this enzyme in the relationships between vitamin A metabolism and the toxicity of foreign compounds. Acknowledgments The authors wish to thank Larry W. Robertson for very helpful readings of the manuscript and also Paul Bellenand who did part of the statistical analysis for determination of REH assay conditions. Jean-Paul Macaire and Madalena Ferreira provided useful technical assistance. The authors thank the North Atlantic Treaty Organisation Collaborative research grants Programme (CRG 890 526) for financial support. References 1 Sommer, A. (1982) in Nutritional Blindness: Xerophtalmia and Keratomalacia, Oxford University Press, New York. 2 Sommer, A., Tarwotjo, I., Hussaini, G., et al. (1983) Lancet 2, 585-588. 3 Kubler, W. (1988) in Malnutrition - a problem of industrial societies? (Somogui, J.C., ed.), Bibl. Nutr. Dieta Basel, Karger, 42, 88-100. 4 Das, N.P., Ma, C.W. and Salmon, Y.M. (1987) B&hem. Med. Metab. Biol. 37, 213-219. 5 Kolonel, L.N., Hinds, M.W., Nomura, M.Y., Hankin, J.N. and Lee, J. (1985) Natl. Cancer Inst. Monogr. 69, 137-142. 6 Spom, M.B. (1977) Nutr. Rev. 35 (4) 65-69. I Giguere, V., Ong, E.S., Segui, P. and Evans, R.M. (1987) Nature 330, 624-629. 8 Sommer, A., Tarwotjo, I. and Katz, J. (1987) Am. J. Clin. Nutr. 45, 977-980. 9 lnnami, S., Nakamura, A., Miyazaki, M., Nagayama, S. and Hishide, E. (1976) J. Nutr. Sci. Vitaminol. 22, 409-418. 10 Brouwer, A. and Van den Berg, K.J. (1984) Toxicol. Appl. Pharmacol. 730, 204-209.
76 11 Spear, P.A., Gamin, H. and Narbonne, J.F. (1988) Can. J. Physiol. Pharmacol. 66, 1181-1186. 12 Mentlein, R. and Heymann, E. (1987) Biochem. J. 245, 863-867. 13 Prystowsky, J.H., Spith, J.E. and Goodman, D.W.S. (1981) J. Biol. Chem. 256 (9) 4498-4503. 14 Cooper, D.A., Furr, H.C. and Olson, J.A. (1987) J. Nutr. 117, 2066-2071. 15 Friedman, H., Mobarhan, S., Hupert, J., Lucchesi, D., Henderson, C., Langenberg, P. and Layden, T.J. (1989) Arch. Biochem. Biophys. 269 (1) 69-74. 16 Harrison, E.H. and GAD, M.Z. (1989) J. Biol. Chem. 264, 17142-17147. 17 Azais, V., Arand, M., Rauch, P., Schramm, H., Bellenand, P., Narbonne, J.F., Oesch, F., Pascal, G. and Robertson, L.W. (1987) Toxicology 44, 341-354. 18 Omura, T. and Sato, R. (1964) J. Biol. Chem. 239, 2370-2378. 19 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. 20 Einot, 1. and Gabriel, K.R. (1975) J. Am. Statis. Assoc. 70, 351. 21 Willett, W.C., Stampfer, M.J., Underwood, B.A., Taylor, J.O. and Hennekens, C.H. (1983) Am. J. Clin. Nutr. 172, 275-290. 22 Underwood, B.A., Loerch, J.D. and Lewis, K.C. (1979) J. Nutr. 109.796-806. 23 Brouwer, A., Blaner, W.S., Kukler, A. and Van den Berg, K.J. (1988) Chem. Biol. Interact. 68, 203-217. 24 Phillips, W.E.J. and Hatina, G. (1972) Nutr. Rep. Internat. 5, 357-362.
25 Sato, M. and Lieber, C.S. (1982) J. Nutr. 112, 1189-1195. 26 Supopark, W. and Olson, J.A. (1975) Internat. J. Vit. Nutr. Res. 45, 113-123. 27 Azais, V., Rachman, F., Gras, S., Pascal, G. and Amedee-Manesme, 0. (1987) Drug. Nutr. Interact. 5, 81-88. 28 Cooper, D.A. and Olson, J.A. (1986) Biochim. Biophys. Acta 884, 251-258. 29 Hendriks, H.F.J., Blaner, W.S.. Wennekers, H.M., Piandetosi, R., Brower, A., De Leeuw, A.M., Goodman, D.W.S. and Knook, D.L. (1988) Eur. J. Biochem. 171, 237-244. 30 Yost, R.W., Harrison, E.H. and Ross, A.C. (1988) J. Biol. Chem. 263 (35), 18693-18701. 31 Powers, R.H., Gilbert, L.C. and Aust, SD. (1987) Toxicol. Appl. Pharmacol. 89, 370-377. 32 Gillette, D.M., Corey, R.D., Helferich, W.R., Macfarland, J.M., Lowenstine, L.J., Moody, D.E., Hammock, B.D. and Shull, L.R. (1987) Biochim. Biophys. Acta 929, 310-320. 33 Yoshimura, H., Yonemoto, Y.. Yamada, H., Koga, N., Oguri. K. and Saeki, S. (1987) Xenobiotica 17, 897-910. 34 Maliwal, B.P. and Gutrie, F.E. (1982) J. Lipid. Res. 23. 474-479. 35 Andres, J., Lambert. I., Robertson, L.. Bandiera, S., Sawyer, T.. Lovering, S. and Safe, S. (1983) Toxicol. Appl. Pharmacol. 70. 204-215. 36 Jensen, K.J., Cullum, M.E., Deyo, J. and Zile, M.H. (1987) Biochim. Biophys. Acta 926. 310-320.
BBALIP
Biochimica et Biophysics Acta, 1047 (1990) 70-76 Elsevier
53501
Retinyl ester hydrolase and vitamin A status in rats treated with 3,3’,4,4’-tetrachlorobiphenyl Muriel Mercier, Gbard Laboratoire
Pascal and VQonique
de Nutrition et Skuritk
Alimentaire,
(Received
Key words:
Retinyl ester hydrolase;
INCA-CRJ,
Azais-Braesco
Jouy-enJosas
(France)
17 May 1990)
Vitamin
A; Tetrachlorobiphenyl;
(Rat)
Previous studies have shown that rats exposed to 3,3’,4,4’-tetrachlorohiphenyl (TCB) exhibit decreased liver vitamin A stores. The activity of retinyl ester hydrolase (REH), the enzyme responsible for the hydrolysis of the storage form of vitamin A (retinyl esters) into free retinol, may therefore be altered by TCB. This study was carried out to investigate the effect of TCB on vitamin A distribution and on REH activity in the rat. REH activity was measured in liver homogenates and microsomes (650 pg protein), in Tris-maleate buffer 0.1 M at pH 7.2 in the presence of 150 mM CHAPS and 1.5 mM retinyl palmitate dispersed in Triton X-100 0.2%. Using these conditions, the kinetic parameters of the enzyme were determined and the inter-animal variation coefficient (10%) allowed statistical comparisons between experimental groups. Male Wistar rats of sufficient or deficient vitamin A status were treated IP with 340 pmol of TCB/kg. Vitamin A levels were significantly depressed in liver. REH activity was decreased about 20%, and serum retinol was decreased about 50%, independent of the initial vitamin A status of the animals. Vitamin A levels in lungs and testes were also decreased, suggesting that TCB could interfere with vitamin A delivery to target organs. The negative effect of TCB on REH activity in vivo was also observed when TCB was added in vitro to the incubation medium at concentrations near to those expected after in vivo treatment. TCB is a non-competitive inhibitor of retinyl palmitate hydrolase.
Introduction Vitamin A deficiency is one of the main public health problems in numerous developing countries, where xerophtalmia leads to irreversible blindness for several hundreds of thousands of children each year and is an important factor of mortality for thousands of others [1,2]. In industrialized areas, such acute problems do not appear, but several epidemiological studies show that important segments of the population (children, young women, elderly people) do not fulfill their needs by their dietary intake of vitamin A or precursors [3]. The health effects of the resulting sub-optimal levels of vitamin A are not precisely known, but one may think that such individuals are more vulnerable to external factors. For example, vitamin A is thought to have a
Abbreviations: BHT, butylated hydroxytoluene; CHAPS, 3[(3cholamidopropyl)-dimethylammoniol-l-propane sulfonate; IP, intraperitoneally; PCBs, polychlorinated biphenyls; RBP, retinol-binding protein; RE, retinol equivalent; TCB, 3,3’,4,4’-tetrachlorobiphenyl. Correspondence: V. Azais-Braesco, Laboratoire de Nutrition Skcurite Alimentaire INRA-CRJ, 78350 Jouy-en-Josas, France.
00052760/90/$03.50
0 1990 Elsevier Science Publishers
et
B.V. (Biomedical
protective effect against chemical-induced carcinogenesis, as has been suggested by epidemiological studies in humans [4,5], and experimental investigations in animals [6]. The recent discovery of several nuclear receptors for retinoic acid, a first step metabolite of retinol, seems to indicate that vitamin A could trigger the expression of various genes encoding for proteins involved in cellular differentiation [7]. Besides inadequate dietary supply, vitamin A deficiency could arise from other sources, such as infectious diseases [8], or exposure to foreign compounds. Numerous xenobiotics lead to diminution of the liver vitamin A stores, and among them, polychlorinated biphenyls (PCBs) which are ubiquitous and persistent pollutants of food and environment. PCBs have been shown to alter vitamin A metabolism and distribution [9-111 and lead to a reduction in vitamin A stores. Among the possible explanations for such interactions between vitamin A and xenobiotics, one could hypothesize that some step of vitamin A metabolism is modified by the xenobiotic. When the liver stores are decreased, the storage form of vitamin A (retinyl esters), which normally represent 95% of the hepatic content, must be diminished. Such mobilization should involve an enDivision)
71 hanced hydrolysis of retinyl esters into free retinol. The specificity of the reaction is not clear and several enzymes, namely carboxylesterases, are able to catalyse such a reaction in vitro [12]. However, until recently, the retinyl palmitate hydrolase (RPH, EC 3.1.1.21) was the only one shown to be active in vivo. According to early investigators, this enzyme seemed to exhibit unusual characteristics, in particular a very high variation coefficient and a puzzling subcellular distribution [13]. During the past two years, the same hydrolysis reaction was assayed by a different method by two authors [14,15] who assigned new characteristics to the enzyme, especially for the optimum pH and the variation coefficient. This enzyme was called Retinyl Ester Hydrolase (REH); however, it is not clear if its difference with RPH comes from the assay technique or if they really are different enzymes. Last year, Harrison and Gad [16] described the existence of a neutral, bile salt-independent retinyl ester hydrolase that seems to be different from REH or RPH. In the first part of this study, we have determined the assay conditions for measuring REH activity in rat liver homogenate, displaying higher activity and the lowest variation coefficient. We then investigated the effect of a pure polychlorinated biphenyl compound on REH activity in rat liver. We chose the PCB 3,3’,4,4’-tetrachlorobiphenyl (TCB), a compound whose effects on vitamin A distribution (decrease in liver and serum) have been previously described [17]. Materials and Methods Chemicals TCB was synthesized, purified and characterized as previously described [17] and was greater than 99% pure. Retinoids used for diet supplementation, as calibration standards for high-performance liquid chromatography (HPLC) or as substrate for REH were kind gifts from Hoffmann-La Roche (Neuilly, France). [11,12-3H]Retinol (40-60 Ci/mmol) was obtained from New England Nuclear (Paris, France). HPLC solvents, HPLC grade, were purchased from Prolabo (Paris, France). All other chemicals were obtained from Sigma and were of the highest grade available. Animals and treatments All the animals used in this study were male Wistar rats raised in our animal colony; they were kept at 20 k 2” C, with a 12 h light cycle and had free access to food and water; they were killed after they had been fasted overnight. For determination of the conditions for assaying REH activity, the rats were fed a standard commercial mixture (Pietrement, France), containing 10000 IU (international units) of vitamin A per kg; they weighed 150 g at the time of killing. Under ether anesthesia, the liver was perfused with ice-cold Tris-HCl
buffer 0.15 M (pH 7.4), then removed, homogenized in 9 volumes of ice-cold Tris-Maleate buffer 0.05 M (pH 7.4) filtered through gauze, then frozen in liquid nitrogen and stored at - 80 o C. For studying the effects of TCB exposure, a semisynthetic diet (48% corn starch, 24% sucrose, 18% delipidated-casein supplemented with D,L-methionine, 3% soybean oil, 2% cellulose, 4% salt mixture and 1% vitamin mixture without vitamin A), essentially free of vitamin A was used. The amounts of retinyl palmitate needed to obtain diets of differing vitamin A contents were added. Rats were born from mothers fed a diet containing only 800 IU vitamin A/kg, a concentration that allows normal pregnancy, delivery and milking but that leaves the pups with very little liver vitamin A. After weaning, rats were maintained for 1 week on a vitamin A-free diet to assure further depletion of liver vitamin A stores. Less than 8 RE/liver were detected in six rats killed for checking purposes. The animals were then separated into two groups: one was fed a diet containing 25000 IU vitamin A/kg (vitamin A-sufficient group) and the other received a diet containing 800 IU vitamin A/kg (vitamin A-deficient group). After 18 days, each group was again divided in two, one half receiving an IP injection of TCB dissolved in corn oil (340 pmol/kg) and the other half receiving corn oil alone via the same route. 6 days after treatment, blood was collected by ocular puncture and serum was obtained by centrifugation (15 min, 1400 g) after clotting at 4“ C in the dark. Liver was perfused as described above, then separated into three parts, one for vitamin A determination (homogenized in 9 volumes of sucrose 0.25 M, KC1 0.025 M, MgCl, 0.05 M, buffered with Tris-HCl 0.05 M (pH 7.4), one for preparation of microsomes (homogenized in 4 volumes of sucrose 0.25 M, EDTA 1 mM buffered with Tris 0.05 M, then processed according Omura and Sato [18]) and one for REH activity determination (homogenized in 9 volumes of Tris-maleate buffer 0.05 M, containing 400 mg/l ascorbic acid). Testes and lungs were excised and processed for vitamin A determination as for the liver samples, except that they were homogenized in 4 volumes of buffer. All samples were frozen in liquid nitrogen and stored at - 80 o C until analysis. Vitamin A determinations Proteins in 1 ml of the homogenate were precipitated by an equal volume of ethanol. A known amount of retinyl propionate as internal standard was added, and then extracted twice with 2 ml of hexane containing butylated hydroxytoluene (BHT); the hexane phases were evaporated under nitrogen and the extract was dissolved in 400 ~1 of methanol. Vitamin A was determined by reverse-phase HPLC on a Kontron apparatus equipped with a Nucleosil column (25 X 0.39
72 cm, pore diameter: 5 pm). Using pure methanol as eluant and a flow rate of 2 ml/mm, elution was achieved in 2.8 min for retinol, 3.9 min for retinyl propionate and 19.2 min for retinyl palmitate. Ultraviolet detection was done at 330 nm and results calculated using both internal and external standards on an Anacomp computer (Kontron).
retinol measured. When necessary, tritiated retinol (1 pCi/incubation) or TCB was added in 5 ~1 methanol. Proteins. Proteins were determined according to Lowry et al. [19]. Statistical analysis. All data are expressed as means + standard deviation. Statistical analysis were done by variance analysis and Student-Newman-Keuls’s test [20].
REH assays Conditions for measuring rat liver REH activity were adapted from Cooper et al. [14] and from Friedman et al. [15]. An aliquot of liver homogenate or microsomes (650 pg of protein) was incubated in Tris-maleate buffer 0.1 M (pH 7.2) at 37 o C in a shaking water bath (90 movements/mm) for 90 min in presence of 150 mM CHAPS and 1.5 mM retinyl palmitate dispersed in 0.2% Triton X-100. The final volume was 0.5 ml and the reaction was terminated by the addition of 0.5 ml ethanol containing BHT (50 mg/l). Retinol in the incubates was monitored following the procedure used for vitamin A determinations. Control incubations without homogenate allowed correction for non-enzymatic retino1 formation. Endogenous retinol from the homogenates was verified to represent less than 0.1% of the
Results
REH ectlvlty 20 - fnmol rellnol
REH
/Ch.O.S
ml lncubntlon
25
mcdlum)l
i/_ A
; j,_--
16-
__~--
600 Protelno
REH
activity
lnmol
retinoVmg
600
inmol
retlnol/(h.mq
prot)]
:
1000
1200
01’ 4
Ii
” 4.5
5
5.5
c
6.5
(ug)
proteid
““” 7
7.5
0
a.5
9
9.5
10
PH
REH
36r
40 lnoubatlon
actlvlty
_-
I
400
200
r
;
01
0
Determination of assay conditions for REH activity The optimal conditions were determined from the data shown in Fig 1. Protein concentration (650 pg/O.5 ml) and incubation time (90 min) were in the linear part of the curve. Optimal pH was shown to be 7.2. CHAPS and substrate concentrations were chosen after variance analysis to give simultaneously a maximal enzyme activity and a minimized variation coefficient. Under these conditions, kinetic parameters for REH activity were determined and we found an apparent K, of 1.34 mM and a maximal velocity of 28 nmol of retinol formed/h (Fig. 2). Although the rate increase of the specific activity of REH is diminishing, a plateau is still not reached for a substrate concentration of 8 mM (Fig. 3).
actlvlty
Inmol
retlnol/(hmg
prot)l
00
tlmc fmln)
Fig. 1. Determination of the assay conditions for measuring REH activity. Incubations were carried out at 37 o C, in a shaking water bath, in a final volume of 0.5 ml in Tris-maleate buffer 0.1 M. Upper left panel (A): effect of protein concentration. Upper right panel (B): effect of pH. Lower left panel (C): effect of the incubation duration. Lower right panel (D): effect of the CHAPS concentration. Each point represents the mean of six different incubations and has been corrected for non-enzymatic blank.
73 l
0.4
z
TCB
TABLE
P
r
I
Effect of substrate (retinyl palmitate) concentration on variation coefficient
0.3-
c
Incubations have been carried out as described in the Material and Methods section. Variation coefficient has been calculated as the ratio: (S.D./mean) X 100, n = 6.
2 - TCB
VLSI Fig. 2. Kinetic ting). Inhibition
REH
2
15
1
0.5
0
-0.5
(l/mM)
parameters of REH activity (Lineweaver-Burk plotof REH activity by TCB. Each point represents the mean of six different determinations.
activity lnmol
mg proleinl
retinal/h
50
Substrate concentration
Variation coefficient
(mM)
(W)
0.5 1 1.5 2 2.5 3 5 8
12 8 7 12 21 21 29 60
in the same homogenates on different days was found to be 4% (data not shown). The activity of the enzyme stored at - 80°C was stable for up to 6 months (less than 2% variation). Table I shows the increase of the variation coefficient with increased substrate concentration. I
OV
0
10
8
2
Fig. 3. Effect of retinyl pahnitate concentration on REH activity. Rat liver homogenates were incubated as described in the Material and Methods section, with the indicated retinyl palmitate concentration. Each point is the mean of REH activity on six different rats. Variation coefficients are given on Table I.
It was not possible to study the effect of higher substrate concentrations because, above a concentration of 10 mM, retinyl palmitate does not dissolve in 0.2% Triton X-100. From a concentration of 0.5 mM, the substrate was never rate-limiting, since more than 90% was found at the end of incubation. The reproducibility of the assay, checked by measuring the enzyme activity TABLE
Vitamin A content of the organs (Table II) As expected, the vitamin A status was very dependent on the amount of vitamin A in the diets. The rats fed 25000 IU vitamin A/kg exhibited a normal distribution of vitamin A: (1) high hepatic stores, with more than 98% in esterified form; (2) normal serum retinol (according to data of various authors [11,21]); and (3) small amounts of vitamin A in extra hepatic organs, with approx. 80% in esterified form in target organs (lungs and testes). On the other hand, the animals receiving the 800 IU vitamin A/kg showed a deficient status characterized by: (1) much lower liver vitamin A content, exclusively as free retinol; (2) decreased serum retinol (-SOW), a
II
Serum retinol concentration and vitamin A treatment
(retinol+ retinyl palmitate) content of organs of rats of different vitamin A status with and without TCB
Results are expressed in retinol equivalent (1 RE =l jtg of retinol or 0.55 pg of retinyl palm&ate). R is the ratio: palmitate) X 100. Values in the same column, with different superscripts differ significantly (P < 0.05) -(n = 6). Vitamin
A content
liver
(R.E.)
R
R
lungs
R
testes
(free retinol/retinol+retinyl
Retinol concentration (R.E./ml) serum
25 000 IU/kg
diet
Control TCB-treated 800 IU/kg
61-l 339
590” f90b
1.1 2.4
2.47+1.51 = 1.47 f 0.59 b
16 15
1.65 * 1.06 a 0.91+ 0.30 Gb
21 16
0.67 f 0.10 = 0.31+0.10 b
0.08 f 0.03 ’ 0.03 f 0.02 c
96 96
0.33 f 0.05 b 0.12 f 0.06 b
40 60
0.32 _+0.03 b 0.10*0.02 =
diet
Control TCB-treated
2.55+ 1.72 f
0.85 ’ 0.57 d
100 100
74 clear symptom of vitamin A deficiency [22]; and (3) drastic diminution in the target organs. Vitamin A had nearly disappeared from lungs (- 97%) and only remained as free retinol, whereas retinyl palmitate still represented 60% of the vitamin A in the testes although the total quantity decreased by 80%. TCB treatment led to significant modifications of vitamin A levels, which do not seem to depend on the nutritional status of the animals. Liver stores were diminished by the exposure to TCB ( - 50% in the rats fed the normal-vitamin A diet and -33% in the vitamin A-deficient animal). Important losses of vitamin A were observed in testes and lungs and the circulating retinol levels were significantly diminished. The relative proportion of retinol and retinyl esters was not altered. In the vitamin A-sufficient animals, vitamin A was determined in the microsomal and cytosolic fraction. A significant decrease of 52% in the microsomes (0.081 k 0.025 ER/mg protein in control rats vs. 0.039 & 0.036 ER/mg protein in TCB-treated animals) and 77% in the cytosols (0.347 k 0.28 ER/mg protein vs. 0.078 &-0.12 ER/mg protein) occurred following TCB administration. REH activity in liver Vitamin A status did not modify the activity of REH in liver homogenates (Table III). TCB treatment induced a diminution of REH activity by 18% in both dietary groups, which was significant in the vitamin A-deficient group only. In the microsomal fraction, where the enzyme activity is greatly enhanced on a mg of protein basis, TCB also diminished REH activity. Effect of TCB on REH activity in When added to the incubation solved in methanol inhibited the dose-dependent way, and behaved tive inhibitor for retinol towards 1.05 IJM (Fig. 2). Methanol alone
TABLE
vitro medium, TCB disenzyme activity in a like a non-competiREH, with a Ki of had no effect. When
III
REH activity in liver and hepatic microsomes A status with and without TCB treatment
of rats of different vitamin
In the same column, values with different cantly (P < 0.05 - n = 6). n.d., not done.
superscripts
REH activity (nmol retinal/h
differ
mg proteins)
liver
hepatic
25 000 IV/kg diet Control TCB-treated
microsomes
15.22 f 2.69 a,b.c 12.48 f 1.26 a,b
49.42 + 15.05 a 39.50+ 9.21 b
800 IU/kg diet Control TCB-treated
16.44 f 2.78 ’ 13.59& 2.58 b
n.d. n.d.
signifi-
incubations were carried out with tritiated retinol, all the radioactivity extracted after 90 min of incubation (80% of the initial activity) was found in the retinol peak, both in the presence or in the absence of TCB (data not shown). This indicates that the inhibition of REH activity by TCB is not an artefact due to a non-enzymic degradation of newly formed retinol, as suggested by other authors [23]. Discussion This study was designed to investigate the role of REH in the modifications of vitamin A distribution in animals treated with TCB. Such alterations in vitamin A metabolism have been reported previously [9-11,171, and occur after exposure to a great variety of xenobiotics, such as pesticides [24], alcohol [25], oral contraceptives [26] and drugs [27]. The common feature observed is a decrease in the liver vitamin A stores that was confirmed in the present study, but we found that extrahepatic organs were also involved. It is interesting to observe that vitamin A in target organs (lungs and testes) was drastically decreased in TCB-treated rats although the liver stores, in spite of a 50% diminution, are still abundant enough to theoretically ensure a normal vitamin A concentration. Alterations in the mechanisms of retinol transport and delivery may have occurred following TCB exposure. Alterations in the system of retinol delivery to target organs are more likely. The assay technique proposed by early investigators [13] to determine REH activity was based upon the detection of free radioactive palmitic acid liberated from radioactive retinyl palmitate. The reproducibility of the assay and especially the very high variation coefficients reported were not easily compatible with experimental investigations. Moreover, the authors did not give any detail on the substrate concentrations nor on kinetic parameters of the enzyme. Recently, Cooper et al. [28] reported a more convenient assay technique, displaying low variation coefficients when pig liver homogenates were used, but still a 42% variation when rat liver homogenates were assayed [14]. Friedman and co-workers [15] recently published a study in which they found a variation coefficient of only 13%. Our modifications to the initial technique are similar and we find a variation coefficient of 7%. In studies of REH activity the difficulty of determining saturating substrate concentration arises. Friedman and co-workers did not give any data for REH activity with a substrate concentration above that which they found to be saturating, i.e., 900 PM. Moreover, while using the same assay conditions as Cooper et al. [14] and ourselves, they found much lower kinetic parameters [apparent K, = 0.325 . mg protein)] than we or mM and V,,, = 7.7 nmol/(h Cooper et al. did (K, = 1.3 mM). The latter author investigated the effect of substrate concentration up to
75 5 mM, which gave a REH activity of 21 nmol/(h . mg protein), which was not saturating. This body of experimental data leads one to think that there might exist some peculiarity of REH that could prevent the attainment of a plateau of activity while increasing the substrate concentration. 95% of hepatic vitamin A is stored in lipid droplets of fat storing cells where it is unable to come into contact with the enzyme which is, as shown on Table III, concentrated in the microsomal membrane. Hendriks et al. [29] showed that RPH was present in fat-storing cells and preliminary results of our laboratory indicate that REH can also be detected as well in these cells by our assay technique (unpublished data). It is likely that in vivo some mechanism that favors the reaction occurs. Bile salts might be involved, or a substrate carrier, possibly a fatty acid binding protein. This could facilitate the reaction, as has been recently shown for cellular retinol-binding protein which enhances the esterification of retinol by acyl : CoA acrylretinol transferase [30]. Although the assay conditions for measuring REH activity are not completely satisfactory and need further improvements, one could estimate that the data obtained are representative of the activity of the enzyme in the liver of rats having received different treatments. TCB exposure leads to a depletion of liver retinyl ester stores. One may have thought that REH, the enzyme responsible for the necessary hydrolysis of retinyl esters into retinol, would then present an enhanced activity. On the contrary, our results show a diminution of REH activity in the liver homogenates or microsomes of TCB-treated rats. Investigating TCB-treated female Sprague-Dawley rats, Powers and co-workers [31] observed a similar dose-dependent decrease in REH activity. It must be pointed out that TCB leads to the same decrease in the liver of vitamin A-deficient rats, which are free of retinyl esters. Togehter with the fact that vitamin A status has no effect on REH activity, this indicates that this enzyme is not regulated by the amount of its substrate. In vitro studies were carried out with TCB added to the incubation medium at a concentration of 1.7 PM, which had been estimated to be in the range of magnitude of the rat liver concentration, 1 week following the injection of 340 pmol/kg [32]. This amount of TCB leads to a diminution of the V,,, of REH activity, while the K, is not modified: this indicates that TCB behaves like a non-competitive inhibitor, whose Ki was determined to be 1.05 FM. The addition of such an amount of TCB in the incubation mixture of REH leads to a decrease of REH activity of 50%. Besides the uncertainty in the determination of the TCB concentration added in vivo, this stronger TCB effect in vitro could be explained by the metabolisation and the distribution of TCB. Recent studies [33] indicate that TCB
undergoes metabolic detoxification by hydroxylation and the metabolites can account for the decrease in REH activity observed. Moreover, in vivo, TCB is carried by lipoproteins, like related chlorinated hydrocarbons [34], or by a cytosolic receptor [35]. It is then possible that the TCB molecule itself (or a metabolite) has inhibiting properties towards REH, binding a specific site of the protein. This could explain the observed decrease in serum retinol that does not occur in animals treated with other vitamin A-depleting compounds. For example, alcohol, which decreases liver vitamin A, was recently shown to increase REH activity in vitro [15], and we have obtained similar results with our experimental conditions (data not shown). Jensen and co-workers [36] described a decrease of REH activity in livers from hexabromobiphenyl-treated rats, not related to a diminution of circulating levels of vitamin A. However, this compound was inactive in vitro, possibly because this xenobiotic requires metabolism to exert its toxic properties and to bind REH. In vivo and in vitro studies on the effects of various xenobiotic on REH activity would be needed to determine the role of this enzyme in the relationships between vitamin A metabolism and the toxicity of foreign compounds. Acknowledgments The authors wish to thank Larry W. Robertson for very helpful readings of the manuscript and also Paul Bellenand who did part of the statistical analysis for determination of REH assay conditions. Jean-Paul Macaire and Madalena Ferreira provided useful technical assistance. The authors thank the North Atlantic Treaty Organisation Collaborative research grants Programme (CRG 890 526) for financial support. References 1 Sommer, A. (1982) in Nutritional Blindness: Xerophtalmia and Keratomalacia, Oxford University Press, New York. 2 Sommer, A., Tarwotjo, I., Hussaini, G., et al. (1983) Lancet 2, 585-588. 3 Kubler, W. (1988) in Malnutrition - a problem of industrial societies? (Somogui, J.C., ed.), Bibl. Nutr. Dieta Basel, Karger, 42, 88-100. 4 Das, N.P., Ma, C.W. and Salmon, Y.M. (1987) B&hem. Med. Metab. Biol. 37, 213-219. 5 Kolonel, L.N., Hinds, M.W., Nomura, M.Y., Hankin, J.N. and Lee, J. (1985) Natl. Cancer Inst. Monogr. 69, 137-142. 6 Spom, M.B. (1977) Nutr. Rev. 35 (4) 65-69. I Giguere, V., Ong, E.S., Segui, P. and Evans, R.M. (1987) Nature 330, 624-629. 8 Sommer, A., Tarwotjo, I. and Katz, J. (1987) Am. J. Clin. Nutr. 45, 977-980. 9 lnnami, S., Nakamura, A., Miyazaki, M., Nagayama, S. and Hishide, E. (1976) J. Nutr. Sci. Vitaminol. 22, 409-418. 10 Brouwer, A. and Van den Berg, K.J. (1984) Toxicol. Appl. Pharmacol. 730, 204-209.
76 11 Spear, P.A., Gamin, H. and Narbonne, J.F. (1988) Can. J. Physiol. Pharmacol. 66, 1181-1186. 12 Mentlein, R. and Heymann, E. (1987) Biochem. J. 245, 863-867. 13 Prystowsky, J.H., Spith, J.E. and Goodman, D.W.S. (1981) J. Biol. Chem. 256 (9) 4498-4503. 14 Cooper, D.A., Furr, H.C. and Olson, J.A. (1987) J. Nutr. 117, 2066-2071. 15 Friedman, H., Mobarhan, S., Hupert, J., Lucchesi, D., Henderson, C., Langenberg, P. and Layden, T.J. (1989) Arch. Biochem. Biophys. 269 (1) 69-74. 16 Harrison, E.H. and GAD, M.Z. (1989) J. Biol. Chem. 264, 17142-17147. 17 Azais, V., Arand, M., Rauch, P., Schramm, H., Bellenand, P., Narbonne, J.F., Oesch, F., Pascal, G. and Robertson, L.W. (1987) Toxicology 44, 341-354. 18 Omura, T. and Sato, R. (1964) J. Biol. Chem. 239, 2370-2378. 19 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. 20 Einot, 1. and Gabriel, K.R. (1975) J. Am. Statis. Assoc. 70, 351. 21 Willett, W.C., Stampfer, M.J., Underwood, B.A., Taylor, J.O. and Hennekens, C.H. (1983) Am. J. Clin. Nutr. 172, 275-290. 22 Underwood, B.A., Loerch, J.D. and Lewis, K.C. (1979) J. Nutr. 109.796-806. 23 Brouwer, A., Blaner, W.S., Kukler, A. and Van den Berg, K.J. (1988) Chem. Biol. Interact. 68, 203-217. 24 Phillips, W.E.J. and Hatina, G. (1972) Nutr. Rep. Internat. 5, 357-362.
25 Sato, M. and Lieber, C.S. (1982) J. Nutr. 112, 1189-1195. 26 Supopark, W. and Olson, J.A. (1975) Internat. J. Vit. Nutr. Res. 45, 113-123. 27 Azais, V., Rachman, F., Gras, S., Pascal, G. and Amedee-Manesme, 0. (1987) Drug. Nutr. Interact. 5, 81-88. 28 Cooper, D.A. and Olson, J.A. (1986) Biochim. Biophys. Acta 884, 251-258. 29 Hendriks, H.F.J., Blaner, W.S.. Wennekers, H.M., Piandetosi, R., Brower, A., De Leeuw, A.M., Goodman, D.W.S. and Knook, D.L. (1988) Eur. J. Biochem. 171, 237-244. 30 Yost, R.W., Harrison, E.H. and Ross, A.C. (1988) J. Biol. Chem. 263 (35), 18693-18701. 31 Powers, R.H., Gilbert, L.C. and Aust, SD. (1987) Toxicol. Appl. Pharmacol. 89, 370-377. 32 Gillette, D.M., Corey, R.D., Helferich, W.R., Macfarland, J.M., Lowenstine, L.J., Moody, D.E., Hammock, B.D. and Shull, L.R. (1987) Biochim. Biophys. Acta 929, 310-320. 33 Yoshimura, H., Yonemoto, Y.. Yamada, H., Koga, N., Oguri. K. and Saeki, S. (1987) Xenobiotica 17, 897-910. 34 Maliwal, B.P. and Gutrie, F.E. (1982) J. Lipid. Res. 23. 474-479. 35 Andres, J., Lambert. I., Robertson, L.. Bandiera, S., Sawyer, T.. Lovering, S. and Safe, S. (1983) Toxicol. Appl. Pharmacol. 70. 204-215. 36 Jensen, K.J., Cullum, M.E., Deyo, J. and Zile, M.H. (1987) Biochim. Biophys. Acta 926. 310-320.
