The in vitro metabolism of mevalonate by sterol and non-sterol pathways.

The metabolism of mevalonic acid by both sterol and non-sterol pathways has been evaluated in nine tissues of the rat. An in vitro estimation of the non-sterol, or "shunt", pathway of mevalonate metabolism was made possible by determining the conversion of [2-14C]mevalonate or [5-14C]mevalonate to 14CO2 in tissue slices. In confirmation of our previous results, the kidney was found to play a major role in the metabolism of mevalonate to sterols and sterol precursors. The shunt pathway accounted for a significant percentage of the mevalonate metabolized in kidney, ileum, spleen, lung and testes, but was of minor importance or undetectable in liver, brain, skin, and adipose tissue. Kidney, however, proved to be by far the most active tissue site of mevalonate metabolism by the shunt mechanism in that, on an average, renal tissue metabolized (R)-[14C]mevalonate over the non-sterol pathway at a rate that was 21 times that of any other tissue examined. These results indicate that the kidneys are of major importance in the metabolism of mevalonate by each of the known pathways of metabolism of this sterol precursor.

The metabolism of mevalonic acid by both sterol and non-sterol pathways has been evaluated in nine tissues of the rat. An in vitro estimation of the non-sterol, or "shunt," pathway of mevalonate metabolism was made possible by determining the conversion of [2-"Clmevalonate or [&"C]mevalonate to "CO, in tissue slices. In confirmation of our previous results, the kidney was found to play a major role in the metabolism of mevalonate to sterols and sterol precursors. The shunt pathway accounted for a significant percentage of the mevalonate metabolized in kidney, ileum, spleen, lung, and testes, but was of minor importance or undetectable in liver, brain, skin, and adipose tissue. Kidney, however, proved to be by far the most active tissue site of mevalonate metabolism by the shunt mechanism in that, on an average, renal tissue metabolized (R)-["Cl mevalonate over the non-sterol pathway at a rate that was 21 times that of any other tissue examined.
These results indicate that the kidneys are of major importance in the metabolism of mevalonate by each of the known pathways of metabolism of this sterol precursor.
It has been previously demonstrated in this laboratory that, contrary to expectation, the kidneys rather than the liver represent the major site of uptake and metabolism of circulating mevalonic acid (1). This conclusion was based on the observation that, following the intravenous injection of labeled mevalonate, the kidneys converted approximately 4 times as much mevalonate to cholesterol and cholesterol precursors as did the liver (1). Moreover, in in vitro studies renal cortex was consistently more active in sterologenesis from mevalonate than was any other tissue of the body (2). Edmond  For this reason, all calculations of mevalonate conversion to its metabolites assume that one-half of the injected mevalonate is inactive.
In the studies involving the intrarenal localization of the mevalonate shunt, renal glomeruli and tubules were separated as previously described (2). Briefly, the entire rat kidney was forced through a loo-mesh screen, and the glomeruli were sedimented in 0.9% NaCl (saline) (6). Tubules were then isolated by centrifuging the supernatant fraction at 500 x g for 10 min to remove any remaining glomeruli; the 500 x g supernatant was then centrifuged at 10,000 x g to recover the tubular fractions. This procedure has been shown to yield approximately 95% pure preparations of glomeruli and tubules (2). At the end of the incubation period, 0.9 ml of 1 N sodium hydroxide was injected through the serum cap into the centerwell, and 1 ml of 1 N H,SO, was injected into the outer well. "CO, was then collected in the centerwell by reincubating the samples for 15 min in the Dubnoff shaker.
The incubation flasks were then opened; 0.1 ml of the centerwell contents was added to 10 ml of scintillation counting solution consisting of 1,000 ml of toluene, 300 ml (342 g) of Beckman Bio-Solv III, 100 ml of H,O, and 6.00 g of 2,5-diphenyloxazole; and the "C content was determined using a Beckman LS-230 liquid scintillation counter. All values were corrected for the small amounts of "CO, recovered from flasks incubated with boiled tissues. Nonsaponifiable Lipids-After addition of an internal standard of [3H]cholesterol, the contents of the outer well were transferred to a 125-ml Erlenmeyer flask. The mixture was then saponified by the addition of 1 ml of 90% potassium hydroxide and 15 ml of 70% ethanol. After refluxing for at least 4 hours on a hot plate, the nonsaponifiable material was extracted three times with 25 ml of petroleum ether; the petroleum ether extracts were dried and taken up in chloroform, and the nonsaponifiable sterols were separated by thin layer chromatography on Brinkmann Instruments Polygram Sil G plates by the method previously described (2). The radioactive bands were located by radioautography using Eastman Kodak RP-14 x-ray film. Strips, 1.5.cm, corresponding to standards of cholesterol, lanosterol, and squalene were cut from the plates, and each band was placed in a vial containing 10 ml of the following scintillation fluid: 0.3 g of 1,4-bis [2-(5-phenyloxazolyl)]benzene, 6.0 g of 2,5-diphenyloxazole, 133 ml of ethyl acetate, and 1,867 ml of toluene. The gain and discriminator window settings were adjusted so that less than 0.2% of the 'H counts were read in the "C window, while approximately 15% of the "C counts were recorded in the 3H window.
The amount of 3H cholesterol added as internal standard was adjusted so that 3H counts were approximately 5 times greater than the "C counts. Under these conditions the counting efficiency for 3H was 25% and for '"C, 75%. All calculations were corrected for spillover and background. Recovery of the cholesterol averaged 50 to 95%.
With the thin layer system employed, desmosterol and cholesterol as well as other 27.carbon sterols co-chromatograph; however, these 27.carbon sterols, like cholesterol, have lost three angular methyl groups, and as noted below, the same factor would apply in determining the "CO, derived from [  other tissues studied showing even less ability than liver to metabolize mevalonate by this route. This result was confirmed in Experiment 4 in which [5-'*C]mevalonate was employed as the substrate. In this experiment kidney m,etabolized mevalonate by the shunt mechanism at a rate 32 times that of liver.
The relative importance of these two pathways of mevalonate metabolism is indicated for each tissue in Table II, lines G  and H, and is summarized in Table III, columns 3 to 6. In liver almost all mevalonate was metabolized by way of the sterol pathway, with an average of only 1% of the metabolized mevalonate using the shunt mechanism. In the kidney, by contrast, a significant fraction (28%) of the mevalonate metabolized utilized the shunt pathway. Of the 72% of mevalonate utilizing the sterol pathway in the kidney, the majority (96%) was recovered as sterols and sterol precursors, and in the case of [2-"Clmevalonate a calculated 4% was released as "CO, derived by the demethylation of lanosterol. Although, as noted above, the absolute quantitative significance of the shunt pathway in the non-renal tissues studied is minor, in comparison with the sterol pathway the shunt mechanism does play a relatively significant role in the ileum, testes, spleen, skin, and lung (Table III, column 3).
If one compares the roles of each of these pathways of mevalonate metabolism in the production of the "CO, derived from [2-'YZ]mevalonate (Table III, columns 5 and 6), the difference between liver and other tissues is striking. In liver, an average of only 13% of the "CO, was derived by the shunt pathways, whereas in kidney over 90% of the "CO, produced from [2-'"Clmevalonate resulted from this route of mevalonate metabolism. Similarly, in spleen, lung, testes, adipose tissue, and ileum, this figure averaged 50% or more. Effect of Substrate Concentration on Routes of Mevalonate Metabolism-We have previously shown that the relative ability of various tissues to metabolize mevalonate by the sterol pathway is very dependent upon the concentration of mevalonate to which the tissue is exposed (1,2). To determine whether the role of the kidney in the mevalonate shunt pathway is similarly influenced by mevalonate concentration, a comparison of the two mechanisms of mevalonate metabolism was carried out at concentrations of (R)-mevalonate that were higher (0.22 and 1.95 mM) than those routinely used in the other phases of this study (0.10 mM). It is apparent from the data in Table IV that regardless of the concentration of mevalonate employed, the activity of the shunt pathway in kidney greatly exceeds that in liver. By contrast, the relative importance of these two tissues in the sterol pathway of mevalonate metabolism changes as the concentration of mevalonate increases. At lower mevalonate levels the kidney converts somewhat more mevalonate to sterols and squalene than does the liver; however, as noted previously (2), the V,,, for the conversion of mevalonate to squalene is higher in the liver than in the kidney. Consequently, at high and unphysiological concentrations of mevalonate the liver converts mevalonate to sterols at rates exceeding those of the kidney. These data therefore confirm a previous finding that the kidney is a major site of mevalonate incorporation into sterols when mevalonate is present in low concentrations.
On the other hand, at both    (3) have emphasized that, at least in vivo, the synthesis of long chain fatty acids from mevalonate can be used as an indication of the activity of the mevalonate shunt mechanism. Attempts were therefore made to determine whether [2-"Clmevalonate could be incorporated into fatty acids by any of the tissues that had been shown by the 14C0, technique to possess the shunt mechanism. No significant l*C was recovered in any of the individual long chain fatty acids isolated by gas liquid chromatography, nor could we detect labeled farnesoic acid in this preparation. The absence of "C in the long chain fatty acid fraction confirms the findings of Edmond and Popjak and suggests that quantification of the mevalonate shunt pathway by the determination of '"C incorporation into fatty acids represents an insensitive assay of these reactions.
Comparison of Renal Cortex, Medulla, Glomeruli, and Tubules as Sites of Mevalonate Metabolism-In view of the finding that the kidney represents the most active tissue site of mevalonate shunt activity, an attempt was next made to determine whether this metabolic pathway is localized primarily in the cortex, medulla, glomeruli, or tubules of the kidney. Slices of cortex and medulla and relatively pure glomeruli and tubules were isolated as previously described (2). They were then incubated separately with [2-"Clmevalonate, and their ability to metabolize mevalonate by way of the sterol and shunt pathways was compared. The results shown in Table V, line B, demonstrate that the sterol pathway of mevalonate metabolism is localized almost exclusively in the glomeruli. Moreover, the sterol synthetic activity in the cortex exceeded that of the medulla by approximately 3-fold. Similar findings were previously reported from this laboratory (2). Per unit weight of tissue, cortex and glomeruli possess 2 to 3 times the shunt activity found respectively in the renal medulla and tubules (Table V, line F). The difference between the glomeruli and tubules is therefore smaller than that noted for the sterol pathway. Both glomeruli and tubules apparently play significant roles in mevalonate metabolism via the shunt pathway.

DISCUSSION
It has been previously shown in this laboratory (8,9) and subsequently in several others (10)(11)(12) that the reaction responsible for the synthesis of mevalonate represents the primary biochemical site of cholesterol feedback control in the liver. This finding, coupled with the demonstration that this feedback reaction to dietary cholesterol is consistently and uniquely lost in all tumors (13)(14)(15)(16)(17)(18), has led to a re-examination of the metabolic fate of mevalonate by both in viuo and in vitro techniques. Unexpectedly, these studies demonstrated that circulating mevalonate is primarily metabolized not by the liver, as had been previously assumed, but by the kidneys (1,2). A similar result has been obtained in both mice and rabbits (19). In these initial studies, it could be shown that the kidneys convert mevalonate primarily to squalene and lanosterol and only to a lesser extent to cholesterol. The dominant role of the kidney in the metabolism of mevalonate to sterols and sterol precursors was subsequently confirmed by Edmond and Popjak (3) and by Edmond (20). These investigators, however, also reported that when injected in viuo into newborn rats, mevalonate may be metabolized in brain, spinal cord, and skin by way of a shunt pathway that bypasses sterol synthesis. This pathway of mevalonate metabolism did not appear to play a significant role in renal tissue as determined by either the in vivo or in vitro methods that were employed. However, Edmond and Popjak were unable to demonstrate mevalonate shunt activity in any tissue as assessed by in vitro incorporation of [2-'4C]mevalonate into fatty acids; hence there remained a possibility that redistribution of labeled fatty acids among various tissues might limit the validity of their in viuo approach to evaluating the tissue localization of the mevalonate shunt pathway. Since "CO, production from [2-"C]mevalonate or [5-'4C]mevalonate does not depend upon the ability of a tissue to synthesize fatty acids, this technique would appear to offer a more sensitive means of assessing the role of the mevalonate shunt in vitro.
As summarized in Fig. 1, the results of this approach to quantification of the pathways of mevalonate metabolism clearly demonstrate that the kidney is overwhelmingly the major tissue site of the mevalonate shunt pathway. In fact, each of the other tissues studied possesses, on an average, less than 5% of the "shunt" activity present in the kidney. The dominant role of the kidney in the metabolism of mevalonate by the shunt pathway is, moreover, observed at all concentrations of mevalonate examined (Table IV). This consistent renal localization of the shunt mechanism of mevalonate disposal (Fig. 1A)