Bh~,himit'o et Biophysica Acta. 11185( 19911315-321

c131991 ElsevierScience PublishersB.V. All rights reserved 00!15-27611/91/S113.511

315

ADONIS 000527609100256U

BBALIP 53720

Compartmentalization of cholesterol metabolism and cellular growth in cultured intestinal crypt cells F r a n k M. R e i m a n n , G e r h a r d H e r o l d , A l b r e c h t S c h n e i d e r a n d E d u a r d F. S t a n g e D(,partm(.t.t z>fbtternal .~,h'dicine II. Unil ersity Of Ubn. ~ "ha ( F. R. G. )

(Received 5 March It)t)l)

Key words: Cholesterol;HydroxymethylglutaryI-CoAreductase: Acyl-('oAcholeste~'lacyltransferase:LDL: }tDL:(.'ell growth: (Rat intestine1;(IEC-6 cell) Growth of rat intestinal crypt derived cells IEC-6 ceased when the key enzyme of cholesterol synthesis, hydroxymethylglutaryI-CoA reductase, was blocked by the competitive inhibitor mevinolin. This effect was reversed by the addition of mevalonolactone. LDL suppressed reductase activity as well as cholesterol synthesis from [ t4C]octanoate and stimulated acyI-CoA cholesterol acyltransferase, but failed to support cell growth despite rapid receptor mediated degradation even in the presence of low mevalonolactone concentrations. Inhibition of cholesterol esterification by Sandoz.Compound 58.035 enhanced cell growth in the presence of mevinolin, but did not promote proliferation in the additional presence of low-density lipoproteins. HDL 3 but not HDL z or tetranitromethanemodified HDL s totally reversed the mevinolin induced inhibition of cell growth. This rescue by HDL 3 was overcome by an increased dose of mevinolin. HDL s derepressed reductase, stimulated cholesterol synthesis and reduced cholesterol esterification, but did not reverse the cholesterol synthesis inhibition by mevinoli~. It is concluded that IEC-6 cells preferentially use endogenously synthesized cholesterol for membrane formation re.ther than low-density lipoprotc-"r, cholesterol. High-density lipoproieins appear to normalize cell growth in the presence of mevinolin by inhibition of cholesterol esterification and probably by inducing the formation of non sterol products of mevalonate.

Introduction Cholesterol metabolism in the small intestine is particularly complex because the enterocyte absorbs luminal cholesterol from dietary and biliary origin, synthesizes cholesterol de novo and also takes up circulating low-density lipoprotein cholesterol, both at a rate second only to the liver [1]. Most of the absorbed cholesterol is esterified by acyI-CoA cholesteryl acyltransferase [2] a n d / o r a pancreatic cholesterol esterase [3] and is finally secreted as ehylomierons into the lacteals. In addition to a full complement of LDL receptors [4], the gut mueosal cell also displays receptors for HDL3 [5,6]. The morphological and functional heterogeneity of

Abbreviations: FCS. fetal calf serum; LPD-FCS. lipoprotein-deficient fetal calf serum: LDL low-densitylipoprotein: HDL. high-density lipoprotein; TNM, tetranitromethanc. Correspondence: E.F. Stange, Deplrtment of Internal Medicine II. Universityof UIm,Oberer Eselsberg. Robert Koch Str. 8, 7~}01IUIm, F.R.G.

the gut further complicates studies on intestinal cholesterol metabolism. The fate of locally formed cholesterol appears to be dichotomized into membrane synthesis in the crypt cell compartment with its high mitotic activity and lipoprotein surface material in the villus epithelium, the latter depending on the presence of triacylglycerol absorption [7]. The fate of LDL cholesterol which is taken up at a constant rate mostly in the crypts and lower villus [4] remains unknown. Similarly, the binding of HDL 3 also appears to bc unregulated [8]. Its function in gut cholesterol metabolism is unclear because internalization and degradation are followed by a paradoxical increase of stcrol synthesis in both rat [9] and human enterocytes [6]. To further elucidate the interplay of growth, Ioca! cholesterol synthesis and lipoprotein metabolism we studied the intestinal cell line IEC-6 [10]. Morphological and immunological studies have shown that IEC-6 cells display features of undifferentiated small intestinal crypt cells [10], but partial differentiation may be induced tinder certain conditions [1 1,12]. The present study shows that cholesterol metabolism in cultured

316 crypt derived IEC-6 cells is strictly compartmentalized because these cells mainly use endogenously synthesized cholesterol rather than LDL-cholesterol for membrane formation. Materials and Methods

Cell culture Rat intestinal epithelial cells IEC-6 were obtained from American Type Culture Collection (1592-CRL) and were used at passage 20-22. Cells were grown as monolayers on 100 mm plastic Petri dishes (NUNC). The cultures were maintained in 10% Dulbecco's Minimal Essential Medium supplemented with L-Glutamine (2 mM), L-Arginine (0.4 mM), 0.1% non-essential amino acids, sodium lactate (2 m ~ l , NaCO 3 (40 mM), glucose (10 mM), 0.3% phenol red, 400 l . U . / m l insulin, 1% Pen Strep (Gibco), and 5% fetal calf serum. Cells were grown in a Cytoperm incubator (Heraeus) at 37°C. The atmosphere was humified and maintained at 8% CO2. Subcultures for experiments were derived from stock cultures by splitting one confluent dish (usually grown for 7 days) at a ratio of I : 10. Incubation medium was changed twice a week. Cell counts were determined at day 4, 7, 11 and 14 after seeding using a Coulter Counter.

Preparation of lipoproteins Lipoprotein fractions were separated by ultracentrifugatlon of human plasma as described previously [13]. The fractions isolated were low-density lipoproteins (LDL, d = 1.019-1.063 g / m l ) as well as high-density lipoproteins (HDL2, d = 1.063-1.125; and H D L 3, d = 1.125-1.21(I g/ml). Lipoprotein concentrations are given in terms of their total cholesterol content. The tetranitromethane modification of high-density lipo~rotcins was performed according to the method of Chacko [14]. Lipoprotein-deficient serum was prepared as described [13]. Cellular LDL binding and degradation was determined as detailed in [15].

Enzyme assays The rate of conversion of [14C]hydrox3,methylglutaryl-CoA to [14C]mevalonate as a measure for hydroxymcthylglutaryI-CoA reductase was determined in homogenates of detergent-solubilized cells as described by Goldstein, Basu and Brown [15]. For the assay of cholesteryl ester formation cells cultured for 7 days received 40 gl of sodium-[14C]oleate complex (10 mM [14C]olcate in 1.2 m g / m l albumin, 10000 d p m / n m o l of [~4C]oleate-albumin) per dish containing 4 ml of medium. The sodium-[ 14C]oleate-albumin complex was prvpare:i ~:; described in [15]. Cholesteryl ester formation was linear for at least 5 h under these conditions and measured as detailed in [15]. The rate of C2-flux from [~C]octanoate into free cellular cholesterol was

determined essentially as described in [16] with minor modifications. Briefly, measurement was started by the addition of 0.1 mM ['4C]octanoate (5300 d p m / n m o l ) of subconfluent cells. After 48 h the incubation medium was discarded, cells were ha,vested using a trypsin-containing buffer (150 mM NaCI, 10 mM Tris-HCI, 20 mM EDTA, 4 1 U . / m l Trypsin, pH 7.4). Cells were washed twice using the same buffer without trypsin. Lipids were extracted and separated from cell pellet and medium as described by Herold et al. [16].

Other methods The cholesterol content in lipoprotein fractions was determined with the cholesterol oxidase method using a commercial kit from Boehringer Mannheim. The protein content of homogenate extracts was determined according to the Lowry method [17].

Materials D,L-Hydroxym ethyl-[3- ~4C]glutaryI-CoA (47.8 mCi/mmol), O,L-[5-3H]mevalonolaetone (24.0 m C i / mmol), [1-14C]oleic acid (57 mCi/mmol), [1-14C]octanoie acid, sodium salt (53.5 mCi/mmol), [7-3H]cholesterol (21 Ci/mmol), eholesteryi-[l,2,6,7-3H]oleate (82 m C i / m m o l ) and Nat251 (17 C i / m g ) were obtained from Du Pont-New England Nuclear. Kyro EOB was from Proctor & Gamble. Cell culture supplies and fetal calf serum were supplied by Gibco. Mevinolin was a kind gift of Dr. K. Eckardt (MSD, Miinchen) and Sandoz-Compound 58-035 was kindly provided by Mrs. Carolyn E. Pickens (Sandoz Research Institute, East Hanover). All other chemicals were from Sigma Chemicals, Merck or Boehringer Mannheim. Results

Regulation of cholesterol synthesis a~d esterification To define the basic patterns of lipoprotein mediated regulation of cholesterol metabolism in these crypt derived cells the response of hydroxymethylglutarylCoA reductase, of [ 14C]acetate flux from [14C]octanoate into cholesterol and of cholesterol esterification to lowand high-density lipoprotein fractions was tested. Addition of low-density lipoprotein cholesterol suppressed reductase in a dose-dependent fashion (Table I). The suppressive effect of LDL on cholesterol synthesis was also evident from a reduction of [~4C]acetate flux into cholesterol (Table 11). The Sandoz-Compound 58-035 effectively suppressed cholesterol ¢sterification (Table i), but did not enhance the inhibitory effect of LDL on cholesterol synthesis (Table I1). In the presence of low-density lipoproteins cholesteryl ester formation was stimulated in a dose-dependent way (Table 1). These results suggested that low-density lipoproteins were taken up by the 1EC-6 cells and lipoprotein derived

317 TABLE 1

cholesterol entered a regulatory subpool for reductase a n d t h e s u b s t r a t e pool f o r e s t e r i f i c a t i o n . T h e r e s p o n s e to H D L , w a s s i m i l a r to t h a t o f lowdensity lipoprotein, but the extent of reductase supp r e s s i o n a n d s t i m u l a t i o n o f cholestery.I e s t e r f o r m a t i o n w a s s m a l l e r w h e n n o r m a l i z e d to l i p o p r o t e i n c h o l e s t e r o l c o n t e n t ( T a b l e I). By c o n t r a s t , H D L 3 s t i m u l a t e d r e d u c t a s e ( T a b l e I) as well a s C_,-flux i n t o c h o l e s t e r o l ( T a b l e I!). In a b s o l u t e t e r m s t h e l a b e l l e d c h o l e s t e r o l assqcia t e d w i t h t h e cells w a s n o t d i f f e r e n t f r o m a c o n t r o l g r o u p ( s e e T a b l e 11), w h e r e a s t h e a m o u n t o f newly s y n t h e s i z e d c h o l e s t e r o l relea:;ed to t h e m e d i u m w a s increased. in m e v i n o l i n t r e a t e d cells r e d u c t a s e activity w a s i n d u c e d m o r e t h a n 4 - f o l d a n d t h i s i n c r e a s e in e n z y m e activity w a s r e v e r s e d by t h e l a c t o n e f o r m o f m e v a l o n i c a c i d at a c o n c e n t r a t i o n o f 10 m M ( T a b l e i). A d d i t i o n o f 50 p , g / m l H D L 3 to m e v i n o l i n - t r e a t e d cells p r o m p t e d a f u r t h e r 1.7-fold s t i m u l a t i o n o f r e d u c t a s e activity. In c o n t r a s t to r e d u c t a s e activity, c e l l u l a r c h o l e s t e r o l synt h e s i s d r o p p e d by 8 4 % at 0.50 p , g / m l m e v i n o l i n a n d this i n h i b i t i o n w a s n o t r e v e r s e d by H D L 3 ( s e e T a b l e 11). T c t r a n i t r o m e t h a n m o d i f i c a t i o n o f H D L 3 to r e d u c e i n t e r a c t i o n w i t h t h e p u t a t i v e H D L r e c e p t o r [14] d i m i n i s h e d t h e i n d u c t i o n o f r e d u c t a s e activity ( T a b l e I).

Regulation of hydro.U,'methylglutatyI-CoA reduetase and atTI-CoA cholesteryl acyltransferase acticity Cells were cultured as described in the Mat'-rials md Methods section. The various effectors were added at da~, ,+ k,r 24 b. Lipoprotein concentrations refer to their cholestero! content. Enzyme activities are expressed as mean % of the control values (range). control values of hydroxymethylglutaryI-CoA reductase activity ranged from 420-720 pmol/mg per h, that of acyI-CoA cholesteryl acyltransferase activity ranged from 1310-2570 pmol/mg per h. The data represent the mean values of 2-4 experiments (n.d.. not determined). Abbreviations: Mev. mevinolin; MVL. mevalonolaetone; TNM, tetranitromethane: S.C., Sandoz-Compound 58-035.

LDL 25 p.g/ml LDL 50 tzg/ml LDL 100 p.g/ml LDL200/zg/ml HDL 2 25 p,g/ml H D L : 50p, g / m l HDL~ 25#,g/ml HDL 3 50/.tg/ml TNM-HDL 3 50/zg/ml Mev. 0.5/,~g/ml MVL 10 mM Mev. 0.5 g,g/ml + MVL 10 mM HDL 3 50/xg/ml + Mev. 0.5/,~g/ml S,C. I /~g/ml

Hydroxymet hylglutaryI-CoA reductase (% of control)

AeyI-CoA cholesteryl acyltransferase (% of control)

82 (78- 93) 59 (57- 72) 48 (33- 57) 44 (32- 54) 97 (74-110) 75 (60- 8 1 1 ) 130(1~-140) 178 (160-185)

1211(I 10-136) 135 1131-1451 144 (120-148) n.d. I I I ( 1115-121)) 124OI8-1361 85 (83- 86) 77 (75- 81))

128 ( 115-1361 406(300-5131 72 (68- 77)

n.d. 66 frO,- 671 n.d.

Low-density lipoproteins and celhdar growth 70 (66- 73)

n.d.

702 (635-767) 202(173-233)

n.d. 12

Suppression of endogenous cholesterol biosynthesis o f I E C - 6 cells by t h e c o m p e t i t i v e h y d r o x y m e t h y l g l u t a r y I - C o A r c d u c t a s e i n h i b i t o r m e v i n o l i n at 0 . 1 - 0 . 3 / . t g / m l r e d u c e d cell p r o l i f e r a t i o n in a d o s e - d e p e n d e n t

(9- 141

TABLE It

Regulation of C2-fllorfrom [ t4CIocmnoate into free cholesterolof cells and t'ultun. ,,wai:on The cells were grown for 7 days in the presence of the various cffeetors. Between day 7 and 90.I mM [14C]-Iab~:led octanoate tspec, act. :,3ull dpm/nmoi) as well as the effector were added. Lipoprotein concentrations refer to Iheir cholesterol content. At day 9 cells were chilled ap.d free cholesterol was isolated from the cell pellets as well as from the corresponding medium samples. De mwo cholesterol synthesis was measured as C2-flux from octanoate into free cholesterol of cells and medium (means and range of 2-4 experiments). Abbreviations: Mev., mevinolin: MVL, mevalonolactone; Chol., cholesterol dissolved in 1'7~ ethanol: S.C., Sandoz-Compound 58-1135.

Control Mev. 0.05 p.g/ml Mev. 0.10 p,g/ml Mev. 0.30 ~ug/ml Mev. 0.50/~g/ml HDL 3 50 p,g/ml Mev. 0.50 izg/ml + HDL 3 50 ~g/ml S.C. I ~ g / m l Chol. 50 p,g/ml LDL 50,u.g/ml LDL 200 p,g/ml LDL 200/.zg/ml + S.C. 1.0 ~ g / m l

C2-flux (nmol/mg)

Range

% in

15 86 6.32 2.57 2.51 2.59 28.52

14.5-17.3 2.8- 9.8 2.1- 3.1 1.9- 3.1 2.4- 2.8 27.2-29.9

60.3 "77:t 42.5 28.5 40.8 37.2

39.7 22.1 57.5 71.4 59.2 62.8

1.88 23.60 3.16 7.13 4.{H)

1.2- 2.(I 23.4-23.9 2.7- 3.6 6.6- 7.5 3.9- 4.1

311.5 48.6 43.7 47.9 37.2

69.5 51.4 56.3 52.1 62.8

7.56

6.0- q.I

36.9

63. !

318 co,t,o, 4

~

o

c - - - ~

~

0 2

,

6

5

,o

,

,,

,2

Fig. I. Growth curves of IEC-6 cells under control conditions (means of 30 experiments, bars denote I S.E.) or in the presence of mevinolin (tl = 6). Abbreviation: Mev. mevinolin. m a n n e r (Fig. 1). At a dose of 0 . 5 / z g / m l mevinolin in the culture m e d i u m growth ceased completely and cell n u m b e r per dish even slightly declined during 14 days of culture. Addition of the lactone form of mevalonic acid, the product of the reductase reaction, totally reversed the antiproliferative effect of mevinolin at an optimal dose of 1 mM (Fig. 2). This indicated that the effect of mevinolin was indeed caused by a block at the reductase level and was not simply due to unspecific toxicity. However, addition of low-density lipoproteins at 200 / x g / m l did not prevent growth inhibition. Since this was possibly due to the need for non-sterol products of mevalonate [18], an i n t e r m e d i a t e dose of mevalonolactone (0.025 mM) was supplied to the cells e i t h e r alone or in combination with a high concentration of low-density lipoprotein (20(I # g / m l ) . Again, the lipoprotein fraction did not restore growth and there was clearly no additive effect c o m p a r e d to mevalonolactone alone (Fig. 2). Blocking the esterification of endogenously synthe5I~evlnolm 0 , 5 , ~ g / m l

+

IAVL 1 , 0 0

mU

4

o

2

(

e e 10 Days after Plat~n~

12

6

tl

10

12

1(

Days after p)at~ng

Days after Piot;n0

14

Fig. 2. Growth curves of 1FC-6 cells under control conditions and in the prescllce oJ mcvinolin plus mevalonolactone or LDL (tl = 4). Abbreviations: Mev. mcvinolin; MVL, mcvalonolactonc.

Fig. 3. Growth curves of IEC-6 cells under control conditions or in the presence of mevinolin plus Sandoz-compound 58-035 with and without LDL (n = 8). Abbreviations: Mev. mevinolim S.C.: SandozCompound 58-035. sized cholesterol by the addition of S a n d o z - C o m p o a n d 58-035 at ! ~,~g/ml did not affect growth u n d e r control condition~, t'i~;. 3), but partially prevented growth inhibition by mevinolin. The further addition of low-density lipoproteins even at very high levels resulted in little e n h a n c e m e n t of cellular proliferation. H i g h e r levels of the Sandoz c o m p o u n d ( 1 0 / . t g / m l ) were apparently toxic and diminished cell growth (data not shown). T h e failure of L D L to restore growth of mevinolin inhibited cells occurred despite a high capacity to specifically bind and d e g r a d e this lipoprotein fraction (Fig. 4). Saturation of specific binding was observed at an L D L concentration of 12.5 # g / m l and after an incubation period of 6 h at 4 ° C . After a time lag degradation was linear between 2 and 8 h.

High-density lipoproteins and cellular growth Studies on the effect of H D L 2 and H D L 3 in IEC-6 cells showed that n e i t h e r fraction affected cellular proliferation directly if a d d e d to the m e d i u m at concentrations up to 100 p g / m l (data not shown). In contrast, H D L 3 but not H D L 2 were able to reverse the growth inhibition affected by mevinolin (Fig. 5). This growth support was striking since the H D L 3 subfraction is generally thought tc cause an efflux of cholesterol in most of the cultured cells studied so far [19,20]. It was conceivable, however, that this fraction d e p l e t e d a regulatory pool of sterols which m o d u l a t e d hydroxymethylglutaryl-CoA reductase and actually e n h a n c e d the synthesis of mevalonate metabolites essential for proliferation. Indeed, such an enzyme induction was observed (Table 1) and increasing the mevinolin concentration up to 2 / z g / m l to block the catalytic activity of this surplus reductase interfered with the H D L 3 mediated normalization of cell growth (Fig. 6). To elucidate if the H D L 3 effect was m e d i a t e d by a receptor-dependent mechanism tetranitromethane-

319 o~ 3so

A.

s

BINDING

= ........

o s .q/ ....

0L-~ so

,~/~,

~" soo 4

. /

2So =00 -J Cn 150 c

~.

tOO

I ~..~_____.I~

0

....

B.

DEGRADATION

j

o

.

.

.

.

.

.

....... .

.

-

Oeys after Plating

Fig. h. Growth curves of IE('-6 cells under contrttl ctmditittns or mevinolin plus tlDL 3 (n = 3). Abbreviations: Mev. mcvinolin.

~

10oo 500

0

o

=

4

e

a

lO

hours

Fig. 4. Binding at 4 ° C and degradation at 37°C of human LDL in IEC-6 cells at day 6 after preincubation for 36 h in lipoprotein free FCS-containing medium. modified H D L 3 particles were tested. In fact, this modification reduced the growth restoring effect of H D L 3 (Fig. 5) and, concomitantly, its stimulatory effect on reduetase (Table !). Discussion

Cells grown in culture can principally derive their m e m b r a n e cholesterol from two sources: E n d o g e n o u s

Me¥inolin 0.5 ~,tJ/ml + HDL-J SO

.

l,.q/ml

.: !iE!i ii!ii;~! :"~:: ............

+

o

z

4

8 a ~o Doym after Plating

12

14

Fig. 5. Growth curves of IEC-6 cells under control conditious or in the presence of mevinolin plus HDL,, HDL~ or tetranitromcthane (TNM) modified HDL~ (n= 4). Abbreviations: Mev. mevinolin: TNM, tetranitromethane.

synthesis or uptake of exogenous cholesterol. For example, growth of chinese hamster ovary and lymphoma cells has been d e m o n s t r a t e d to d e p e n d on receptor m e d i a t e d LDL uptake if hydroxmethylglutaryI-CoA reductase is partially inactivated by a specific inhibitor [21]. If the inhibitor concentration is further increased the total block in reductase activity requires the additional supply of mevalonate to support the formation of certain n o n - s t e r o l m e t a b o l i t e s like dolichol, ubiquinone or isopentenyl adenine in addition to cholesterol [18,22]. Thus, if reductase activity is virtually eliminated, LDL s u p p l e m e n t a t i o n alone will not support cellular proliferation [18]. The present work with a cell line derived from rat intestinal crypts suggests that this concept may not apply to the enterocyte, because low-density lipoprot o n did not support cell growth even in the presence of low levels of mevalonate. First, it is important to note that the lack of a cell rescue was not simply due to ineffective lipoprotein uptake. The suppression of hydrox'ymeth~ ~glutaryI-CoA reductase and of C_,-flux into cholesterol as well as the stimulation of acyI-CoA cholesterol acyl transferase activity indicate a positive sterol balance and argue strongly in fav,3ur of active LDL u p t a k e and degradation, because it is the free, liberated cholesterol exerting these metabolic effects [23,24] possibly through some other mediators [25]. Furthermore, the studies detailed above have confirmed high affinity binding sites and active degradation of human L D L in IEC-6 cells [26], similar to those observed in freshly isolated [27] or cultured [9] rat intestinal cells. This is remarkable because in rat fibroblasts homologous LDL is bound more avidly than heterologous human LDL [28], but the finding of more potent reductase suppression by human as compared to rat lipoproteins in IEC-6 cells [29] is compatible with our findings. However, the degree of reductase suppression is much less in the present study and in previous work on intestinal cultures [9,13,25] than in

320 fibroblasts [23]. Quantitatively, the rate of LDL degradation wa~ comparable to previous work [26]. Even whet: the factors of dilution of substrate specific activity [30] and partial suppression of cholesterol synthesis by FCS are taken into account, cholesterol supply from LDL uptake was 5-10-times in excess of the amount derived from synthesis. Thus, principally enough cholesterol to support cell growth should be delivered through LDL. An alternative reason for the absence of growth promotion by LDL in cells with blocked reductase may be the interference of mevinolin with the formation of non-sterol metabolites of mevalonate which are not supplied by LDL. This explanation is unlikely, however, because LDL had no additive effect on cell growth even when enough mevalonate was provided to support cell proliferation to approximately half tbe normal rate. This is reminiscent of a recent report on a lymphoblastic leukemia T cell line which was also refractory to LDL during growth inhibition by glucocorticoids [31], but in these cells tl~e most likely defect was a low activity of acid cholesterol ester hydrolase. However, as discussed above, our data show good evidence for an effective lysosomal hydrolysis of LDL cholesteryl esters in IEC-6 cells. Apparently. once the LDL cholesterol is taken up it is compartmentalized in a cholesterol pool separate from the substrate pool for membrane synthesis. Possibly, the cholesteryl ester stores derived from LDL cholesterol are resistant to rehydrolysis due to a deficiency in neutral cholesterol ester hydrolase and represent such a pool. On the other hand, growth was also not enhanced by LDL when cholesterol ester formatio,l from LDL cholesterol was suppressed by the Sandoz compound 58-035. Thus, these cells derive their cholesterol requirements mostly from endogenous synthesis rather than lipoproteins despite the presence of an excess of LDL and even when rapid esterification of the free cholesterol is prevented by an acyl CoA cholesterol acyl-transferase inhibitor. In contrast to LDL, H D L 3 restored growth of mevinolin treated cells. To the best of our knowledge only vascular endothelial cells [32], but not chinese hamster ovary or other cells [21], exhibit a similar response to HDL in the state of a blocked reductase. A striking difference to 1EC-6 cells is the fact that the cholesterol-rich HDL 2 subfraction supported growth of vascular endothelial cells more effectively than the H D L 3 subfraction. In addition, the H D L fractions are direct mitogens in endothelial ceils, whereas in IEC-6 cells no such response was observed. It seems unlikely that the growth restoration was caused by a delivery of cholesterol to the cells, particularly since there is ample evidence that HDL 3 particles cause an efflux of cholesterol in cultured cells [19,33]. The efflux of cholesterol in IEC-6 cells was evidenced by the in-

creased amount of [t4C]-Iabeled cholesterol released to the medium when HDL 3 was added to the culture medium. The negative cholesterol balance was also reflected by the increase of reduetase and decrease in esterifying activity. The stimulation of reductase by low concentrations of H D L [13,6] as well as the reduced rate of cholesterol esterification [6] has previously been observed in intestinal cells, although higher levels may be suppressive for reductase [9,13]. Most importantly, when H D L 3 was combined with mevinolin the rate of C2-flux into sterols remained essentially unchanged, whereas reduetase activity was further increased. Thus, it seems that the increase in reductase activity provides non sterol products of mevalonate, that are vital for cell growth, rather than cholesterol itself. The fact that the H D L 3 rescue was blocked by increasing levels of mevinolin suggests that in this cell type the induction of reductase is indeed causally related to the normalization of cell growth. This mechanism is similar to that proposed for the rescue by H D L from compaetin toxicity in vascular endothelial cells [32]. Furthermore, the inhibiting effect of the H D L 3 particles on acyl-CoA cholesterol acyi transferase might in part explain their growth restoring effect in cells deprived of endogenous cholesterol by mevinolin. This hypothesis is supported by the fact that a specific inhibitor of cholesterol esterification (Sandoz-Compound 58-035) reversed partially the growth suppressing effects of mevinolin. The fact that tetranitromethane-treatment of H D L s reduced both the growth restoring effect and its stimulatory effect on reductase suggests a receptor-dependent mechanism [14,33]. Interestingly, high numbers of specific H D L 3 binding sites have been described at the basolateral membrane both in isolated rat and human intestinal cells [5,6,9,27] and may mediate cholesterol efflux through retroendocytosis of H D L 3 [34]. It may be concluded that proliferating IEC-6 cells depend on endogenous cholesterol formation even in the presence of LDL in the medium. This is compatible with previous findings in the live rat that newly synthesized cholesterol is not readily exported from the gut and subserves mostly local functions [7]. HDL3 rescues the cells from the inhibitory effect of mevinolin on reductase and growth possibly through the induction of the synthesis of mevalonate metabolites other than cholesterol as well as a suppression of cholesterol esterification.

Acknowledgments We gratefully appreciate the technical assistance by C. Lindauer, I. Gastrock and R. Barnikel.

321

References 1 Stange, E.F. and Dietschy, J.M. 119851 in Slerols and Bile Acids (Danielsson, H. and Sjfvall, J.. eds.), pp. 121--149, Elsevier. Amsterdam. 2 Suckling. K.E. and Stange, E.F. 119851 J. Lipid Res. 26, 647-671. 3 Treadwell, C.R. and Vahouny, G.V. 119681 in Handbook of Physiology. See. 6, Vol Ill (C.F. Code and W. Heidel, eds.), pp. 1407-1438, American Physiological Society. Washington, D.C. 4 Stange, E.F. and Dietschy, J.M. 119831 Proc. Natl. Acad. Sci. USA 80, 5739-5743. 5 Kagami, A.. Fidge, N., Suzuki, N. and Nestel, P. 119841 Biochim. Biophys. Acta 795, 179-190 6 Sviridov, D.D., Safonova, I.G.. Gusev, A.G.. Talalev, V.P.. Tsibnlsky, V.P., Ivanov, V.O., Preobrazensky, S.N., Repin. V.S. and Smirnov. V.N. 119861 Metabolism 35, 588-595. 7 grange, E.F. and Dietschy, J.M. (1985) J. Lipid Res. 26. 175-184. 8 Kagami, A., Fidge, N.H. and Nestel, P.J. (19851J. l,ipia Res. 26, 705-712. 9 Nano, J.L., Barbaras, R., Negrel, R. and RampaL P. (19861 Biochim. Biophys. Acta 876. 72-79. 10 Quaroni. A., Wands, J., Trelstad, R.T. and Isselbacher, K.J. (19791 J. Cell Biol. 80. 248-265. I1 CarolL K.M., Wong, T.T., Drabik, D.L. and Chang, E.B. (19881 Am. J. Physiol. G 355-360. 12 Kurokowa, M., Lynch, K. and Podolsky, D.K. (19871 Biochem. Biophys. Res. Commun. 142, 775-782. 13 Stange, E.F., Alavi, M., Schneider, A., Preclik. G. and Di'"chuneit, H. 11980) Biochim. Biophys. Aeta 620. 520-527. 14 Chacko, G.K. (19851 J. Lipid Res. 26, 7,15 754. 15 Goldstein, J.L., Basu, S.K. and Brown, M.S. (19831 Methods Enzymol. 98, 241-260. 16 Herold, G., Schneider, A.. Preclik. G., Ditschuneit, H. and Stange, E.F. (1984) Biochim. Biophys. Acta 796, 27-33. 17 Lowry, O.H., Rosebrough, N.J.. Farr. A.L. and Randall. R.J. (1951) J. Biol. Chem. 193, 265-275.

18 Habenicht, A.J.R., Glomset. J.A. and Ross. R. (198O1 J. B;~,I Chem. 255, 5134-5140. 19 Gram, J.F., Albers, J.J., Cheung, M.C. and Bierman, E.L. (19811 J. Biol. Chem. ~6, 8348-8356. 20 Barbaras, R.. Puchois, P., Grimaldi, P. Barkia, A., Fruchart, J.C. and Ailhaud, G. (19871 Biochem. Biophys. Res. Commun. 149, 545-554. 21 Goldstein, J.L., Helgeson, J.A.S. and Brown, M.S. (1979) J. Biol. Chem. 254, 54113-5409. 22 Ouesney-Huneeus, V., Galick, H.A., Siperstein, M.D., Erickson, S.K., Spencer. T.A. and Nelson, J.A, (1983) L Biol. Chem. 258, 378 385. 23 Brown. M,S,, Dana, S.E. and Goldstein, J.k. (',9"74) J. Biol. Chem. 249, 789-7%. 24 Goldstein, J.L., Dana, S.E. and Brown, M.S. 119741 Prec. Nat. Acad. Sci. USA 71, 4288-4292, Z5 Panini, S.R., Sexton, R.C. and R~dney, H. 119841 J. Biol. Chem. ~59. 7767-7771. 26 Gupta, A.. Sexton, R.C. and Rudney, H. (19861 J. Biol. Chem_ 261, 83,18-8356. 27 Fidge, N.H., Nestel, P.J. and Suzuki, N. (19831 Biochim, Biophys. Acta 753, 14-21. 28 Innerarity, T.L., Pitas, R.E. and Mahley, R.W. (19801 J. Biol. Chem. !~5, 11163-11172. 29 Panini, S.R.. Sexton, R.C. and Rudney. H. (19821 Fed. Proc. Fed. Am. Soc. Exp. Biol. 41, 1215. 30 Stange, E.F. and Dietschy, J.M. (19831J. Lipid Res. 24, 72. 31 Cutts, J.L. and Mclnykowitch, (3. (19881 Biochim. Biophys. Acla 961, 65-72. 32 Cohen, D C., Massoglia, S.L and Gospodarowicz, D. (19821 J. Biol, Chem, 257, 9429-9437. 33 Brinton, E.A., Gram, J.F., Chert, C.-H., Albers, J.J. and Bierman, E.L. (19861 J. Biol. Chem. 261,495. 34 Herold, G., Rogler, G., Reimann, F.M. and Stange, E.F. (199111 Kiln. Wochenschrift 68 (Suppl XXII), 1-6,