Assay , Kinetics , and Lysosomal Localization of an Acid Cholesteryl Esterase in Rabbit Aortic Smooth Muscle Cells

A sensitive radioisotope microassay has been developed for the estimation of cholesteryl ester hydrolase (EC 3.1.1.13) in preparations of isolated rabbit aortic smooth muscle cells. Kinetic studies served to establish optimal assay conditions, which involve incubation in a total volume of 0.2 ml containing 5 pM cholesteryl oleate tritiated in the cholesterol moiety, 0.4 mM egg yolk lecithin, 0.3 mu taurocholate, 25 pg of bovine serum albumin (defatted) per ml, and 0.05 M sodium acetate bufIer, pH 4.25. The labeled cholesterol was separated from the ester by thin layer chromatography with double development. Hydrolysis of as little as 1 pmole of substrate could be detected by this method. Under these conditions no measurable activity was found at neutral or alkaline pH in either phosphate or Tris buffer. The enzyme is several times more active on cholesteryl esters of unsaturated fatty acids than on those of saturated fatty acids in the Cl8 and CZO series. Among esters of even numbered saturated fatty acids (Cl2 to &Jr cholesteryl myristate (C,,) was hydrolyzed fastest. Fractionation studies indicate that cholesteryl esterase is associated with lysosomes in aortic smooth muscle cells.


SUMMARY
A sensitive radioisotope microassay has been developed for the estimation of cholesteryl ester hydrolase (EC 3.1.1.13) in preparations of isolated rabbit aortic smooth muscle cells. Kinetic studies served to establish optimal assay conditions, which involve incubation in a total volume of 0.2 ml containing 5 pM cholesteryl oleate tritiated in the cholesterol moiety, 0.4 mM egg yolk lecithin, 0.3 mu taurocholate, 25 pg of bovine serum albumin (defatted) per ml, and 0.05 M sodium acetate bufIer, pH 4.25. The labeled cholesterol was separated from the ester by thin layer chromatography with double development.
Hydrolysis of as little as 1 pmole of substrate could be detected by this method.
Under these conditions no measurable activity was found at neutral or alkaline pH in either phosphate or Tris buffer.
The enzyme is several times more active on cholesteryl esters of unsaturated fatty acids than on those of saturated fatty acids in the Cl8 and CZO series.
Fractionation studies indicate that cholesteryl esterase is associated with lysosomes in aortic smooth muscle cells.
It is well known that there is a marked accumulation of cholesterol and of cholesteryl esters in the arterial wall during the development of atherosclerosis (1). These lipids accumulate both within the cells and in the extracellular region of-the arterial wall (2)(3)(4).
Earlier investigations from this laboratory have shown that, in the aortic smooth muscle cells of rabbits rendered * This work was supported by United States Public Health Grant HD-05065. A report of it was read at the 3rd meeting of the European Group for the Study of Lysosomes, Louvain, Belgium, September 15, 1972. This is the third paper of a series entitled "Lysosomes of the Arterial Wall." Refs. 9 and 5 are Parts I and II of this series; Ref. 6 is Part IV.
$ Present address, Department of Physiological Chemistry, Hiroshima University, School of Medicine, Kasumi l-2-3, Hiroshima, Japan. Address reprint requests to The Rockefeller University, New York, N.Y. 10021.
5 Present address, Royal Postgraduate Medical School, Du Cane Road, London W12 OHS, England. severely atheromatous by a cholesterol-rich diet, the lysosomes are an important site of intracellular cholesterol and cholesteryl ester accumulation (4)(5)(6). Marked elevations in lysosomal acid hydrolase activities (up to more than 12-fold for certain enzymes) and an increase in lysosomal stability accompany this accumulation. Many workers subscribe to the view, recently reiterated by Kothari et al. (7), that cholesterol enters the artery in free form and is then esterified locally.
Such a mechanism does not, however, readily account for the lysosomal localization of the deposits.
Lysosomes are digestive organellcs, and their overloading is most often the consequence of inadequate breakdown of materials taken in by endocytosis or sequestered by autophagy (8).
In view of these facts, it seemed of importance to measure the ability of aortic smooth muscle cells to hydrolyze cholesteryl esters, and to determine the intracellular localization of the enzyme or enzymes responsible for this ability.
Our technique of cell isolation yields only about 1 to 2 mg of cell protein from a single aorta, and this is then further subdivided into some 15 subcellular fract,ions (9). A highly sensitive assay method was therefore required, especially since the cholesteryl ester hydrolase (EC 3.1.1.13) activity of rabbit aortic cells was found to be low. This paper describes the development of such an assay, together with the results of relevant kinetic studies carried out on homogenates of isolated aortic smooth muscle cells. Data on the subcellular localization of cholcsteryl esterase in normal aortic cells will also bc presented. cholesterol to react with 0.02 ml of fatty acyl chloride as described by Deykin and Goodman (13). The plates were developed by ascending chromatography in cyclohexane-benzene-acetic acid (240:240:1, by volume). After development, the plate was allowed to dry at room temperature and the chromatogram was redeveloped in the same solvent system.
The chromatogram was viewed under ultraviolet light, and the plate was then cut into five portions: (a) l-cm strip including origin, InJluence of Bovine Serum Albumin-With some preparations, addition of bovine serum albumin had no effect on the assay, with others it increased the observed activity.
Its inclusion in the incubation medium was decided as a precautionary measure. Its mechanism of action was not investigated.
Influence of pH-The influence of pH on the enzyme activity is shown in Fig. 4. Under our assay conditions, the highest rate of cholesteryl oleate hydrolysis is observed at pH 4.25, and there is no measurable activity around neutrality, in either phosphate or Tris buffer.

Influence of Enzyme Concentration-As
shown in Fig. 5, good proportionality with enzyme concentration was observed up to an activity of about 0.1 microunit, which in the present case corresponded to hydrolysis of 0.67e of the substrate.
As little as No of this activity could be detected.
These limits define the range of optimal applicability of our assay. Standard assay conditions. Homogenate, 60-min incubation.

Influence of Substrate
substrate was found to follow normal Michaelis-Menten kinetics (Fig. 6). The apparent K, for the micellar substrate calculated from the linear double reciprocal plot (Fig. 6, insel) is 40 FM.
InJEuence of E$ectors- Table  II shows the effect of various activators and inhibitors on the hydrolysis of cholesteryl oleate. There was slight activation of the enzyme by 10 mM CaC12, whereas CuClz was inhibitory at this concentration.
EDTA and eserine had no effect on enzyme activity.
The lack of inhibition by eserine is significant since this compound is a potent inhibitor of choline&erase activity. Among the sulfhydryl group inhibitors, p-chloromercuribenzoate was strongly inhibitory at 0.1 mM concentration, whereas N-ethylmaleimide had little effect on enzyme activity and iodoacetamide became inhibitory only at 10 rnM concentration.
Spec$&y-The fatty acid specificity of cholesteryl esterase on a series of substrates with different fatty acid moieties is illustrated in Table III.
All assays were carried out under standard conditions at identical substrate concentration (5 PM). It is apparent that the shorter chain fatty acid esters of cholesterol were hydrolyzed more rapidly than the longer chain esters.    Localization-The subcellular localization of cholesteryl oleate hydrolase was studied in isolated aortic smooth muscle cells from normal rabbits with the "inner layering" technique previously described (5,9). Fig. 7 shows the distribution profiles of cholesteryl oleate hydrolase, cytochrome oxidase, N-acetyl-/3-glucosaminidase, and a-glucosidase in preparations from normal rabbits.
The distribution of cholesteryl esterase resembles most closely that of the lysosomal marker N-acetyl$glucosaminidase.
It differs greatly from that of the mitochondrial cytochrome oxidase, and significantly from that of the presumedly microsomal cr-glucosidase, which shows distinctly more activity in the region of density higher than 1.20.
Earlier studies have shown that pretreatment of the rabbits with repeated injections of Triton WR-1339 causes a selective decrease in the density of aortic cell lysosomes (9). As shown in Fig. 8, this treatment altered the distributions of N-acetyl-P- The most striking differences were observed when fatty acid esters of the same chain length but different degrees of unsaturatlon were compared. Hydrolytic activity was much greater against the unsaturated than against the corresponding saturated fatty acid esters.
Since the differences shown in Table III could have resulted from differences in the physical state of the substrate, experiments were also performed on identical mixtures of four or five esters with a different ester labeled in each sample. As shown in Table IV glucosaminidase and cholesteryl e&erase in a very similar fashion. The main activity peak is shifted toward a lower density, and at the same time there is an unexplained, but common to both enzymes, increased smearing of activity in the lower half of the gradient.
In contrast, the distribution of a-glucosidase, which in earlier experiments was unaltered by Triton WR-1339 treatment of the animals, actually shows a shift in the opposite direction. There is less activity in the starting layer, which contains both soluble enzyme and low density particles, and correspondingly more associated with the main peak. The modal density of ar-glucosidase is essentially unaltered.
These results provide a strong indication that at least the major part of the cholesteryl esterase activity assayed under our conditions is localized in lysosomes.
we find an activity of the order of 12.5 microunits per mg of cell protein, corresponding to about 0.25 milliunit per g (wet weight) of aortic tissue. Almost 10 times as much would be found at saturating substrate concentration.
Closest to our results are those of Kothari and co-workers, who, working with human aorta (16), and later with acetone powders of rat and rabbit aortas (7), have investigated a cholesteryl esterase apparently localized in lysosomes (16), but showing a pN optimum of 6.6 when tested a&nst a micellar substrate, and of 7.4 when acting on an emulsion of cholesteryl oleate stabilized by bovine serum albumin.
These results are in striking contrast with our finding of an optimum at pH 4.25 and total lack of activity above pH 6.5, even in phosphate buffer, used by Kothari el al. (Fig. 4). Another difference between the two sets of results lies in the K, value, which Kothari et al. (7) estimate at 200 PM for the rabbit enzyme, as against 40 PM determined in the present work. On the other hand, the maximum activities measured in the two laboratories are of the same order of magnitude. For their rabbit aorta preparation, Kothari et al. (7) report values of 8 to 10 milliunits per g of acetone powder, or about 3 milliunits per g (wet weight) of tissue, which is essentially what we would find in our preparations at saturating substrate concentration.
Properties of Enzyme-Cholesteryl esterase activity with a pH optimum in the neighborhood of neutrality has been demonstrated in the aorta of several animal species (7,(16)(17)(18)(19)(20)(21)(22)'. But this is the first report of the existence of a distinctly acid cholesteryl esterase in aortic tissue. Such an enzyme has, however, been detected in human, rat, and bovine liver (23)(24)(25), in mouse peritoneal macrophages (26), and in rabbit ovary (27). For all three tissues, there is evidence of a lysosomal localization of the enzyme, as in the rabbit aorta. The macrophage cholesteryl e&erase resembles the aortic enzyme also in having a strong preference for esters of unsaturated fatty acids (26).
It seems likely that differences in assay conditions account for the differences between our results and those of other workers. In particular, the physical state of the substrate presented to the enzyme has been shown to be of critical importance (7,16,25). The conditions worked out in the present study seem well adapted to the measurement of the acid cholesteryl esterase.
With a substrate concentration equal to fg the K, of the enzyme, a At first sight it would appear that two distinct enzymes are involved, were it not for the difficulty of explaining how each group manages to miss entirely the enzyme detected by the other. It is conceivable that, in the work of Kothari et al. (7), the enzyme may be altered by the procedure used to prepare an acetone powder, but this does not seem very probable since these authors also find a pH optimum of 6.6 for the activity in fresh human aortic homogenates (16). It would seem more plausible that the differences in enzyme properties are due to differences in the composition of the assay medium, which are considerable: Kothari et al. use I20 times more cholesteryl oleate and about 4 times more taurocholate and lecithin per unit volume than we do. However, the results of Fig. 6 render this explanation also improbable since we can go up to 35 the substrate concentration used by Kothari et al., with substrate to emulsifier ratios very similar to theirs, without detecting any anomalous behavior of the enzyme.
In their early work (16) Kothari et al. for some unexplained reason included oleic acid in their incubation mixture at 3 times the concentration of cholesteryl oleate. But they did not do so in their later work (7). Presumably, therefore, this factor is not responsible for the observed differences.
These cannot be explained either, as we have verified, by the presence of serum albumin in our test. There remains the possibility that the method followed for preparing the substrate is the critical factor. Kothari et al. add their lipid components in ether solution to the aqueous buffer, homogenize, remove the ether by evaporation at 37", and then sonicate at 15 to 20 kc per s for 15 to 30 min or longer.
What precautions are taken against thermal damage during thii treatment is not stated.
In our procedure, the mixture of cholesteryl oleate and lecithin is introduced first in benzene solution, taken up to dryness, suspended in a taurocholate solution by two 10-s sonications with rigid precautions against overheating, and finally mixed with the buffer and albumin.
Assay Method-As an analytical tool, the procedure described here appears reasonably optimized, and it obeys satisfactorily the requirements for linearity with respect to time (Fig. 3) and enzyme concentration (Fig. 5). It is highly sensitive and can be applied to subcellular fractions isolated from a single aorta preparation (Figs. 7 and 8).
The choice of a very low substrate concentration, not generally advocated in enzyme work, has several advantages in the case of a radioassay of this type. It is economical, not only in terms of cost, but also in terms of time and effort since very pure substrate is more easily prepared on a small than on a large scale. It also increases the sensitivity of the assay, which for a given amount of labeled cholesteryl oleate added decreases linearly with increasing (K, + [S]). From this we calculate, for instance, that about 3 times more enzyme is needed to release the same amount of radioactivity in the free cholesterol fraction at 100 C(M than at 5 pM cholesteryl oleate concentration, This ratio exceeds 14 if the substrate concentration is increased up to 600 PM, as in the method of Kothari et al. (16). Increasing the specific radioactivity of the substrate to offset the requirement for more enzyme is of little help since it causes blank and test values to increase in the same proportion.
Substrate depletion, a possible hazard of working at low substrate concentration, is hardly a risk in our case since only a very small fraction of the substrate is hydrolyzed and the reaction follows essentially pseudo-zero order kinetics.
On the other hand, the real kinetics being close to first order, dilution of the labeled substrate by endogenous cholesteryl ester should not greatly affect the result of the assay. For instance, addition of 2 nmoles or 1.3 pg of cold cholesteryl oleate (i.e. twice the amount of labeled substrate) together with the tissue preparation reduces the percentage of hydrolysis by only 18%. Since the assay is performed with microgram amounts of total protein, such large amounts of endogenous cholesteryl esters are not likely to be added, except, however, in the case of severely atheromatous preparations, in which the cholesterol content approaches the protein content (4,5). It remains to be seen whether the cholesteryl esters of these preparations would adopt in the incubation mixture the physical state necessary for optimal enzymic attack.
Preliminary experiments indicate that they do not.
Biological Signijicance-The presence of an active acid cholesteryl esterase in the lysosomes of aortic smooth muscle cells is of considerable interest in view of the massive intralysosomal accumulation of cholesteryl esters in the aortic cells of cholesterolfed rabbits (5,6).
As we have mentioned in a preliminary report (4) and will elaborate in more detail in a subsequent paper in this series, there is evidence that the amount of cholesteryl ester stored in the lysosomes is inversely related to their cholesteryl esterase activity, suggesting that storage may be the consequence of a relative enzyme deficiency, as in many congenital storage diseases (28).