The shunt pathway of mevalonate metabolism in the isolated perfused rat liver.

The shunt pathway of mevalonate metabolism (Edmond, J., and Popják, G. (1974) J. Biol. Chem. 249, 66-71) has been studied in isolated livers from fed rats perfused with physiological concentrations of variously labeled [14C]mevalonates. The measured rates of 14CO2 production were converted to rates of mitochondrial acetyl-CoA production from mevalonate by methods which take into account underestimations of metabolic rates derived from 14CO2 production. Our data confirm that the shunt pathway leads to mitochondrial acetyl-CoA. The apparent negligible rate of mevalonate shunting in liver, previously reported by others, stems from the very low contribution (congruent to 0.1%) of plasma mevalonate to total mevalonate metabolism in the liver. This contribution was assessed from the relative incorporations of 3H2O and [5-14C]mevalonate into sterols. In livers from fed rats, the shunt diverts about 5% of the production of mevalonate. The total rate of mevalonate shunting in the liver is about 200 times greater than in two kidneys. The liver is therefore the main site of mevalonate shunting in the rat.

The shunt pathway of mevalonatel metabolism hypothesized by Popjak in 1970 (1) is a series of reactions which links the pathways of cholesterol synthesis and leucine catabolism. Branching between the sterol and the shunt pathways occurs at the level of dimethylallyl pyrophosphate which is converted sequentialIy to dimethylallyl alcohol, P-methylcrotonate, and /j"methylcrotonyl-CoA. The latter substrate is an intennediate in the pathway of leucine catabolism to HMG-CoA2 which i s cleaved to acetyl-coA and acetoacetate. According to this scheme, carbon 1 of mevalonate is lost as COz at the level of pyrophosphomevalonate decarboxylase; carbons 4 and 5 of mevalonate become carbons 2 and 1 of acetyl-coA; carbons 2, 3, and 6 of mevalonate become carbons 2, 3, and 4 of acetoacetate (carbon 1 of acetoacetate derives from CO, fix-ation via methylcrotonyl-CoA carboxylase).
Evidence supporting this pathway came from Edmond and PopjAk (2) and Fogelman et al. (3) who injected [Z-14C]-or 15-14C]mevalonate into rats and humans. Label was recovered in ketone bodies and in products of ketone body and acetyl-coA metabolism (Con, n-fatty acids). Further support for the shunt pathway was reported by Brady et at. (4), based on the distribution of "C in p-hydroxybutyrate excreted by rats following injection of mevalonate specifically labeled on different carbons.
In uiuo (2,10) and in vitro (9) studies using [2-14C]or [5-"CJmevalonate have suggested that metabolism of mevalonate via the shunt is negligible in the liver. We suspected that the apparent inability of the liver to shunt mevalonate stems from poor permeation of exogenous ["C]mevalonate into the liver cell. Comparison between rates of ["Clmevalonate (5,9, 10) and of 3H20 (16,17) incorporation into liver sterols shows that the contribution of extracellular mevalonate to total mevalonate metabolism in the liver is indeed very small. We addressed this problem by perfusing livers with 3H20 and physiological concentrations of variously labeled ['4C]mevalonates. The fractional contribution of exogenous mevalonate to total metabolism of mevalonate was assessed by the incorporation of 3H and "C into liver sterols. The rate of mevalonate shunting was assessed by the incorporation of 14C into CO, and n-fatty acids. Rates of %On production were converted to rates of acetyl-coA production from mevalonate using methods that estimate under-recovery of label via exchange processes. Briefly, the extent of I4CO2 reincorporation into compounds nonvolatile in acid was assessed by the recovery of 14C02 from livers perfused with a tracer of ["C] bicarbonate (18). The yield of label from mitochondrial [14C1 acetyl-coA to 14C02 was obtained from the differential yield

Mevalonate Shunt Pathway in Liver
Supelco. Lanosterol (70-80%) was purchased from Sigma; the major impurity was dihydrolanosterol. Squalene (>99%) was obtained from Sigma. Farnesoate methyl ester was synthesized as described by Cornforth and Popjak (21 [4-"C]cholesterol, ["CI-and [3H]toluene counting standards and Oxifluor-COZ scintillation fluid were purchased from New England Nuclear. All samples of labeled RS-mevalonate were purified and resolved by specific phosphorylation of the R-enantiomer (6, 14) with mevalonate kinase purified from pig liver to a specific activity of 5 units/mg of protein as described by Popjak et al. (6).
Liver Perfusions-Adult male and female Sprague-Dawley rats (Charles River) were maintained for 2 weeks on Purina chow made available from 9 a.m. to 12 noon each day. The animals used were 8-10 weeks old and weighed between 250 and 300 g (males) or 200 and 250 g (females). Surgery for removal of the liver was started around 1 p.m. The surgical technique and the perfusion apparatus have been described previously (16). Livers were perfused with 150 ml of Krebs-Ringer bicarbonate buffer containing 4% dialyzed bovine serum albumin (fraction V, fatty acid poor, Miles Biochemicals) and 15 mM glucose.
After a 30-min equilibration period, labeled R-mevalonate was added to the perfusate in amounts calculated to achieve an initial concentration of 250-400 nM, except where indicated. These concentrations are in the physiological range of plasma mevalonate concentration in the rat (80-500 nM, Ref. 6). The perfusion was continued for an additional 90 min throughout which perfusate was sampled every 10 min. Aliquots of perfusate were incubated for 30 min with acetic acid to eliminate "CO, and were counted in a liquid scintillation spectrometer. The effluent "CO, from the oxygenator and the 14C02 present in the perfusate bicarbonate pool at 120 min were trapped in Oxifluor-COz as described previously (22).
Analytical Techniques-The frozen tissue was divided into 2 aliquots for determination of the dry weight to wet weight ratio and chloroform/methanol (21, v/v) extraction of lipids which were analyzed as described previously (12,16), with the following exception. The fatty acids were methylated with diazomethane and separated by preparative gas chromatography. A 2.4-m stainless steel column (4 mm, inner diameter) was packed with 3% SE-30 on Chrom G 80, ' 100. We used a Varian Aerograph model 1200 gas liquid chromatograph equipped with a flame ionization detector. The column was operated with temperature programming from 190 to 350 "C. Injection port and detector were set at 300 "C; outlet tube was set at 350 "C. Carrier gas was nitrogen (70 ml/min). Retention times of methyl farnesoate, methyl palmitate and methyl stearate were 4.0, 7.5, and 15 min, respectively. Individual methyl esters were trapped in Utubes (21) filled with silanized glass wool and cooled in liquid Nz. Mevalonate was assayed in ultrafiltrates from the perfusate by the radioenzymatic assay of Popjik et al. (6).
Calculatinns-All the countings of radioactivity were converted to disintegrations/min by internal standardization with [3H]-or ["C] toluene or both. Rates of mevalonate metabolism were calculated by dividing disintegrations/min incorporated by the integrated specific activity of the tracer in the perfusate.
The measured rates of 14C0, production were converted to rates of mevalonate shunting as follows. First, the recovery of standards of NaH"C03 infused into the perfusate to simulate metabolic production of "CO2 is 63 and 73% in livers from male and female rats, respectively (see "Results" and Ref. 18). Therefore, all rates of 14COz production were first divided by the "COP recovery factor. Second, an additional correction was applied to the production of 14C0, from [2-14C]mevalonate to compensate for the production of "CO, in the cholesterol synthesis pathway at the demethylation of lanosterol (23). For every six molecules of [2-'4CC]mevalonate converted to 27-carbon sterols, one atom of "C is lost as 14C0,. The following formula was used to calculate the rate of shunting from [2-"C]mevalonate. Rate of shunting = (total disintegrations/min in 14C02 -(disintegrations/ min in CZ7 ~terols)/5)/('~CO~ recovery factor X integrated specific activity of mevalonate).
Third, the rates of "COS: production were converted to rates of acetyl-coA production from ["C]mevalonate. In livers from male and female fed rats perfused with tracers of [l-"C]-and [2-'4C]a-ketoiso-caproate (4-methyl-2-oxovalerate), we have determined (19) that 25% of the label from acetyl-coA is transferred to l4COZ. We have assumed that this yield is applicable to the transfer to COz of label from acetyl-CoA derived from ["C]mevalonate (see Footnote a of Table I). RESULTS Throughout the 2-h recirculating perfusion, livers released mevalonate linearly ( r = 0.92) at a rate of 200 pmol/min or 5.4 f 0.43 (&S.E.; n = 6) nmol/g, dry weight x h. The rate of accumulation was not affected in experiments where a bolus of 250-400 nM of labeled mevalonate was added to the perfusate at 30 min. Production of endogenous unlabeled mevalonate was used to assess the dilution (less than 45%) of the specific activity of the tracer between 30 and 120 min of the experiment.
The uptake of [14C]mevalonate was calculated from the linear decrease over 90 min of the acid-stable radioactivity in the perfusate. This decrease amounted to about 20% of the initial concentration. That practically all of the acid-stable radioactivity in the final perfusate was ['4C]mevalonate was ascertained as follows. First, ultrafiltration of the perfusate did not affect its acid-stable radioactivity. Second, the '"Clabeled component present in the ultrafiltrate co-chromatographed with a standard of R[5-3H]mevalonate before treatment and with 5-pho~pho-R[5-~H]rnevalonate after treatment with mevalonate kinase and Mg-ATP (11). Practically equal fractions of the 14C (98%) and 3H (96%) label were converted to the phosphorylated derivative.
Total rates of sterol synthesis were assessed in separate perfusion experiments by the incorporation of 3H20. There was no significant difference in the rates of sterol synthesis between livers from males (3.28 f 1.14 pmol of mevalonate equivalent/g, dry weight X h; n = 5) and from females (4.14 & 1.19; n = 6). Similarly, rates of hepatic sterol synthesis in vivo (16) were not significantly different in males (5.71 k 0.68; n = 8) and females (6.76 * 0.94; n = 8).

Experinents with [l-'4C]Mevalonate-In the third reaction
of mevalonate metabolism, carbon 1 of the substrate is released as C02 by pyrophosphomevalonate decarboxylase. Since the intracellular accumulation of [ l-"Clmevalonate and of its phospho-and pyrophospho-derivatives is negligible, the uptake of [l-'4C]mevalonate should be balanced by an equivalent production of I4CO2. We have recently shown (18) that when metabolic production of 14C02 by the perfused liver is simulated by a constant infusion of NaH"C03 into the perfusate, the recovery of the infused label in ( i ) the effluent gas of the oxygenator, plus (ii) the pool of bicarbonate in the final perfusate, is considerably less than 100%. In livers from male and female fed rats, the recovery of NaH"C03 was 63 & 4% (n = 6 ) and 73 f 5% (n = 6), respectively. In livers from male and female rats perfused with 650 nM of [l-'4C]mevalonate, the production of 14C02 amounted to 58 & 10% (n = 5) and 67 f 5.6% (n = 11) of the uptake of [l-"C]mevalonate. We therefore concluded that the recovery factors measured in experiments with NaH'"C03 are applicable to the production of 14C02 from ['4C]mevalonate.
If all the metabolism of [l-14C]mevalonate goes through pyrophosphomevalonate decarboxylase, none of the end products of either the sterol or the shunt pathway should be labeled. An alternate series of reactions for the conversion of mevalonate to HMG-CoA was considered (24): In this scheme, mevalonate is first dehydrated to A' -or A3-anhydromevalonate which is converted to its CoA derivative. Oxidation of the primary alcohol function followed by hydration yields HMG-CoA. This series of reaction does not involve the G. Popjak ]acetyl-CoA whose label is expected to be found in COZ and in n-fatty acids. The uptake of mevalonate and the measured production of 14C02 were roughly proportional to the concentration of substrate and did not show any significant sex difference (Table I). Production of 14COa accounted for about 1% of the uptake of mevalonate. Note that this is a minimal estimate of the rate of the shunt pathway since a sizable fraction of the label of ['"Cl acetyl-CoA is lost from the tricarboxylic acid cycle via exchange reactions with amino acids and glycolytic intermediates (25). In addition, a fraction of 14C0, generated in the cycle is reincorporated via exchange processes. The measured rates of 14C02 production were converted to rates of ['"Cl acetyl-CoA production by methods described previously (18, 19), using the recovery coefficients listed under "Experimental Procedures." The total rate of operation of the shunt pathway thus calculated accounted for 4-6% of the uptake of mevalonate.
In livers perfused with [2-"Clmevalonate, the measured rate of 14C02 production (11.4-15.8% of mevalonate uptake; not shown in Table I) corresponds to the sum of 14C02 generated mostly in the sterol pathway (demethyiation of lanosterol) and to a small extent via the shunt pathway (following oxidation of [2-'4C]acetoacetate). The amount of 14C02 generated in the sterol pathway can be calculated from the rate of [2-14C]mevalonate incorporation into squalene + sterols (see "Experimental Procedures"). By difference, one calculates the rate of 14C02 production via the shunt pathway (reported in Table I). This rate, which is equivalent to 7-8% of the total rate of 14C02 production, represents a small difference between two large numbers and is therefore inherently imprecise.
In the saponifiable extracts of livers from rats injected with large doses of [P-"Clmevalonate, Edmond and Popjik (2) had identified a fraction corresponding to [*4C]farnesoate. This prenoate (26) is generated from the oxidation of the corresponding prenol, farnesol, which in turn derives from the hydrolysis of farnesyl pyrophosphate (27). We did not find any detectable [14C]farnesoate in the fatty acids extracted from livers perfused with physiological concentrations of ['"Cl mevalonate. When the methylated fatty acids were separated by gas-liquid chromatography, no peak was detected, and no label was collected at the retention time of methyl farnesoate. This was ascertained using reference standards of unlabeled and labeled methyl farnesoate synthesized as described by Cornforth and Popjak (21).
In addition, we attempted to identify labeled prenols by chromatographing the nonsaponifiable fraction on a column of aluminum oxide (21) followed by digitonin treatment of the squalene + sterol fraction. Less than 1.5% of the radioactivity of the nonsaponifiable fraction was found in the supernatant of the digitonides. It is likely that this small amount of radioactivity corresponds to sterols that are not precipitated by digitonin (28). We, therefore, conclude that under physiological concentrations of mevalonate, there is no significant production of long chain prenols and prenoates by the liver. These compounds are presumably generated when the sterol pathway is overloaded by large doses of mevalonate.
Incorporation In order to test whether HMG-CoA (and thus acetyl-CoA) generated by the shunt pathway is mitochondrial or cytosolic, livers were perfused with either [4,5-'4C]mevalonate or 100 mCi of 3H20 in the presence or absence of 2 mM (-)-hydroxycitrate. This inhibitor of ATP-citrate lyase (29) blocks the transfer via citrate of acetyl groups from the mitochondria to the cytosol, site of fatty acid synthesis. As we had shown previously (20), (-) -hydroxycitrate inhibited total fatty acid and squalene + sterol synthesis (measured by 3H incorporation) by 60%. It also inhibited the incorporation of [4,5-14C] mevalonate into n-fatty acids and squalene + sterols by 43% (from 16.4 + 2.2 (n = 5) to 8.9 f 1.3 (n = 5) pmol/g, dry weight X h; p < 0.01) and 24% (Table I; p = 0.06), respectively. On the other hand, (-)-hydroxycitrate did not affect the uptake of [4,5-14C]mevalonate, its conversion to 14C02, and the total rate of mevalonate shunting.
One group of experiments was conducted to test the involvement of alcohol dehydrogenase, dimethylallyl alcohol, and dimethylacrylate in the mevalonate shunt pathway. Livers from male rats were perfused with 400 nM of [5-14C]mevalonate and either 20 mM ethanol, 0.05 mM 4-methylpyrazole, 5 mM dimethylallyl alcohol or 5 mM dimethylacrylate. The rates of mevalonate uptake and of squalene + sterol synthesis from exogenous mevalonate were not affected by any of the tested compounds, but the rate of [5-14C]mevalonate incorporation into 14C0, was significantly decreased.
All the experiments described above were conducted in livers from fed rats perfused with a sterol-free medium containing 15 mM glucose to stimulate mevalonate synthesis. In order to test the influence of a decreased supply of mevalonate on the distribution of the substrate between the shunt and the sterol pathways, one group of livers was perfused with [5-14C]mevalonate and 75 pM Mevinolin, a potent inhibitor of HMG-CoA reductase (30). As expected, livers perfused with Mevinolin did not release mevalonate into the perfusate. Contrary to expectations, addition of Mevinolin led to a 39% decrease in the rate of mevalonate uptake by the liver. This is in contrast to what was observed in the perfused kidney (12) where Mevinolin increased the uptake of mevalonate by 30%. The rate of operation of the sterol pathway, measured by the incorporation of 3Hz0, was 2730 and 61 nmol of mevalonate equivalent/g, dry weight X h in control and Mevinolin-treated livers, respectively. The corresponding rates of exogenous mevalonate incorporation were 3.5 and 1.2 nmol/  'tr by correcting the measured production of 14C02 for (i) the recovery of a tracer of NaH"C0, (63 and 73% in livers luction of acetyl-coA from exogenous mevalonate, is calculated from males and females, respectively), and ( i i ) the yield of label from mitochondrial acetyl-coA to 14C02 in liver, determined in separate experiments from the differential yield in "CO2 from tracer of [l-"C]-and [2-l4C]aketoisocaproate (4-methyl-2-oxovalerate) (25% for both males and females (19)). These factors, which are not affected by (-)-hydroxycitrate, were not determined for perfusions with Mevinolin, ethanol, 4-methylpyrazole, dimethylallyl alcohol, and dimethylacrylate.
' Significantly different from the corresponding control ( p C-0.05 using two-sided t test).
In experiments with [2-"C]mevalonate, a large amount of "Con is generated in the sterol pathway. Production of "CO, in the shunt, which accounts for a small fraction of the total production, is inherently imprecise. This is indicated by the presence of brackets around the rates of "COz production and the derived rates. g, dry weight x h. Therefore, the fractional contribution of exogenous mevalonate to total squalene i -sterol synthesis, and presumably to total mevalonate metabolism, was 0.13 and 2% in control and Mevinolin-treated livers, respectively.

DISCUSSION
In perfusions with [5-14C]-or [4,5-'4C]mevalonate, the rate of 14C02 production amounted to the same fraction (1%) of the rate of mevalonate uptake by the liver. Since [5-14C]-and [4,5-14C]mevalonate yield 14C02 via [1-14C]-and [1,2-14C]acetyl-CoA, respectively, the yield of label from C-1 and C-2 acetyl-coA must be practically the same. In these livers from fed rats perfused with 15 mM glucose, the tricarboxylic acid cycle does not therefore operate as a synthetic pathway (25). In other words, there is no significant influx into the cycle of unlabeled carbon which leaves the cycle carrying a fraction of the label of acetyl-coA.
(-)-Hydroxycitrate inhibits about equally the incorporation into n-fatty acids of [4,5-"C]mevalonate and of labeled leucine (31, 32). This strongly suggests that acetyl-coA derived from mevalonate is mitochondrial. Acetyl-coA is transferred to the cytosol mostly (75%) via citrate and ATP-citrate lyase (29) and, to a minor extent (15%), via acetoacetate and cytosolic acetoacetyl-CoA synthetase (22). In the presence of (-)-hydroxycitrate, the inhibition of the citrate cleavage pathway is in part compensated by an activation of the transfer of acetyl groups via acetoacetate (22). If acetyl-coA derived from [4,5-14C]mevalonate was cytosolic, the incorporation of label into fatty acid would have been increased by (-)-hydroxycitrate. Inhibition of the citrate cleavage pathway would have decreased the dilution of the specific activity of cytosolic acetyl-coA by unlabeled acetyl groups transferred from the mitochondria.
It appears that the shunt pathway of mevalonate metabo-

Mevalonate Shunt Pathway in Liver 8943
lism begins in the extramitochondrial space and ends in the mitochondria. Alcohol dehydrogenase which catalyzes the oxidation of dimethylallyl alcohol to 8-methylcrotonaldehyde is a cytosolic enzyme. On the other hand, the steps p-methylcrotonyl-CoA to HMG-CoA, common with the leucine catabolism pathway, are mitochondrial. Therefore, the shunt pathway crosses the mitochondrial membrane at the level of either P-methylcrotonaldehyde or P-methylcrotonate. It has been shown (33) that ethanol is converted to acetaldehyde in liver cytosol, while acetaldehyde is oxidized to acetate mostly in the mitochondria. It is, therefore, likely that P-methylcrotonaldehyde is the intermediate of the shunt pathway that crosses the mitochondrial membrane. The production of "COZ from [5-'4C]mevalonate was significantly decreased by compounds that ( i ) inhibit alcohol dehydrogenase (4-methylpyrazole), (ii) increase the cytosolic [NADH]/[NAD+] ratio (ethanol and dimethylallyl alcohol), and (iii) dilute the specific activity of the shunt products (dimethylallyl alcohol and dimethylacrylate). Conversion factors needed to convert the production of 14C02 to production of [I-"Clacetyl-CoA were not determined for perfusions in the presence of these substrates. Nevertheless, taken as a whole, the data of these experiments are compatible with the involvement of alcohol dehydrogenase, dimethylallyl alcohol, and dimethylacrylate in the shunt pathway.
The 68% inhibition by Mevinolin of the incorporation of [5-'4C]mevalonate into sterols, also observed in the perfused kidney (12), stems probably from an inhibition of squalene synthesis by the fairly high concentration of inhibitor used (75 p~) .
This secondary site of action of Mevinolin is inferred by analogy with the effect of its analog Compactin (34). The actual inhibition of the incorporation of mevalonate into squalene + sterols is actually much greater than 68% since Mevinolin, by inhibiting the production of endogenous mevalonate, increased the specific activity of intracellular mevalonate. The fractional contribution of exogenous mevalonate to total mevalonate metabolism is 0.13 and 2.0% in control and Mevinolin-treated livers. Therefore, the ratio of these percentages, i.e. 2.0 t 0.13 = 15, represents the increase in the specific activity of intracellular mevalonate induced by Mevinolin. This increase in specific activity probably accounts in part for the 2.3-fold increase in 14C02 production induced by Mevinolin.
When a liver is perfused with physiological concentrations of labeled mevalonate, the contribution of the latter to total hepatic metabolism of the substrate is very low (less than 0.13%). Even when the rate of operation of the sterol pathway is 98% inhibited by Mevinolin, exogenous mevalonate contributes only 2% to the residual 2% of sterol synthesis. From the total rate of shunting of exogenous [4,5-14C]mevalonate (154 pmol/g, dry weight x h) and the fractional contribution of exogenous mevalonate to total mevalonate metabolism in the liver (0.13%), one calculates that the total rate of shunting of endogenous mevalonate is 118 nmol/g, dry weight X h. In kidney (12), the corresponding rate is 40 times lower (2.89 nmol/g, dry weight x h). Since the weight of the liver is about five times that of two kidneys, the amount of mevalonate shunted in the liver is about 200 times greater than in the kidneys. Since the liver accounts for at least half of the total production of mevalonate in the rat, one concludes that the liver is the main site of the shunt pathway of mevalonate metabolism.
In order to derive meaningful rates of mevalonate metabolism in the liver from the use of labeled mevalonate, one must determine the fractional contribution of the exogenous substrate to total mevalonate metabolism in the liver. This con-tribution can be assessed from the incorporation into sterols of tritium from 3Hn0 added to the perfusate. One should recognize that any error in the determination of this very small contribution can lead to a large error in the estimate of the absolute rates of mevalonate metabolism through pathways other than the sterol pathway.
The very small fractional contribution of perfusate ["C] mevalonate to total mevalonate metabolism in the liver implies that, inside the liver cell, the specific activities of mevalonate are three orders of magnitude lower than outside the cell. It follows that one should not be able to derive the turnover rate of whole body mevalonate from the turnover rate of a tracer of labeled mevalonate injected into the plasma.
In 250-g rats, the whole body rate of squalene + sterol synthesis (measured by 3H20 incorporation) is equivalent to about 2400 nmol of mevalonate equivalent/kg x min. The pool of extracellular mevalonate (7) is 0.15 nmol/ml X 200 ml/kg = 30 nmol/kg. Since the half-life of plasma mevalonate in the rat is about 9 min (7), the turnover rate of extracellular mevalonate is (In 2 + 9) X 30 = 2.31 nmol/kg X min. Therefore, in the rat, the turnover rate of plasma mevalonate accounts for (2.31 + 2400) x 100 = 0.1% of the whole body mevalonate turnover. Parker et al. (35) have recently shown that in humans given a constant infusion of R-[5-3H]mevalonate, the turnover of plasma mevalonate also accounts for about 0.1% of the turnover of whole body mevalonate measured by sterol balance.
Lakshmanan and Veech (36) have attempted to shut off the endogenous rate of mevalonate production by injecting large amounts of ["C]mevalonate + 3H20 intravenously into rats. From the evolution of the 3H/14C ratio in liver sterols, as a function of the dose of mevalonate injected, they concluded that one has to inject 1.0-5 mmol of R-mevalonatelkg of rat to block endogenous mevalonate production. In other words, the pool of extracellular mevalonate had to be increased more than 104-fold in order to equilibrate the specific activities of intra-and extracellular mevalonate in the liver.
From the above considerations, one can conclude that it is impossible to quantitate under physiological in uiuo conditions the rate of operation of the shunt or of the sterol pathway using tracers of radioactive mevalonate. When such tracers are injected in uiuo, they essentially bypass the liver which is the main site of operation of both the shunt and the sterol pathways. These in vivo protocols amount to little more than an intrarenal infusion of the tracer. Since the kidney accounts for less than 0.1% of the metabolism of mevalonate in the body, these experiments cannot yield any valid quantitative data on the metabolism of mevalonate in uiuo. In isolated cell or organ preparations, tracers of ['"Clmevalonate can be used in conjunction with 3Hz0 to measure the rate of operation of the shunt pathway, since the ratio of incorporation of 3H and 14C into sterols yields the fractional contribution of extracellular mevalonate to total mevalonate metabolism in this organ. This approach cannot be used in uiuo since the production of 14C02 reflects mostly the metabolism of plasma mevalonate in the kidney.

Meualonate Shunt
Pathway in Liver