Formation of methyl sterols in brain cholesterol biosynthesis. Sterol formation in vitro in actively myelinating rat brain.

Abstract A study was made of the methyl sterols involved in sterol biosynthesis in developing rat brain. In addition to the determination of the endogenous sterols present in 15-day-old rat brain, experiments were conducted to establish which of these sterols could be labeled by a minced tissue preparation in the presence of [2-14C]mevalonic acid. The turnover of [14C]lanosterol, [4-3H]squalene oxide, and [14C]squalene by minced preparations was also shown. In the experiments involving incubation of labeled mevalonic acid with minced tissue preparations, the diunsaturated methyl sterols had a larger degree of labeling than the monounsaturated methyl sterols. Labeled squalene was also shown to be present in the mevalonate incubations. Under these in vitro conditions cholesterol was not found to be significantly labeled. Endogenous sterols of two types were examined, those occurring as steryl esters and those that were unesterified. A comparison of the distribution of sterols in these two fractions indicated a relatively small percentage of cholesterol (54.3%) in the steryl ester fraction. Based on gas chromatographic retention data and gas chromatography-mass spectrometry, the following sterols were identified: lanosterol, 4,4-dimethyl-5α-cholesta-8,24-dien-3β-ol, 4,4-dimethyl-5α-cholest-8-en-3β-ol, 4α-methyl-5α-cholest-8-en-3β-ol, 4α-methyl-5α-cholesta-8,24-dien-3β-ol, 4α-methyl-5α-cholesta-7,24-dien-3β-ol, 4α-methyl-5α-cholest-7-en-3β-ol, 5α-cholesta-7,24-dien-3β-ol, desmosterol, and cholesterol.


SUMMARY
A study was made of the methyl sterols involved in sterol biosynthesis in developing rat brain.
In addition to the determination of the endogenous sterols present in 15-day-old rat brain, experiments were conducted to establish which of these sterols could be labeled by a minced tissue preparation in the presence of [Z-W]mevalonic acid. The turnover of [14C]lanostero1, [4-3H]squalene oxide, and [14C]squalene by minced preparations was also shown.
In the experiments involving incubation of labeled mevalonic acid with minced tissue preparations, the diunsaturated methyl sterols had a larger degree of labeling than the monounsaturated methyl sterols.
Labeled squalene was also shown to be present in the mevalonate incubations. Under these in vitro conditions cholesterol was not found to be significantly labeled. Endogenous sterols of two types were examined, those occurring as steryl esters and those that were unesterified. A comparison of the distribution of sterols in these two fractions indicated a relatively small percentage of cholesterol (54.3%) in the steryl ester fraction.
The gas chromatography-mass spectral det,erminations were supported by Grant AiV-09992 from the National Institutes of Health.
Cholesteryl esters have been known to be present in developing brain for a number of years. They have been shown to be present in the central nervous system of the chick embryo (7-9), the developing rabbit (lo), and the 'i-day-old rat (11). In humans also the presence of steryl esters has been shown in fetal (12) and newborn (13) brain.
As myelination progresses, it has been found that steryl esters decrease (3,12,14). It has been proposed that this elevated level of steryl ester is related to the onset of myelination.
When the above esters have been characterized, the emphasis has been on fatty acid content.
It has been generally assumed that the bulk of the esterified sterol was cholesterol.
We have found that when developing brain was incubated with sodium[2-14]acetate or [2-14C]mevalonic acid (15), a portion of the steryl ester formed contained cholesterol and other 4-desmethyl sterols and methyl sterols.
It was felt that the examination of the sterol portion of the steryl ester might shed more light on the function of steryl esters in developing brain.
Since we have found that considerable steryl ester is formed by minced preparations of 15.day-old rat brain, and does form reasonable amounts of labeled nonsaponifiable material from [2J4C]mevalonic acid (15), the 15-day-old minced tissue system was chosen for the following study on developing brain sterol biosynthesis.  Vol. 247,No. 11 Both instruments were equipped with hydrogen flame detectors. 2 min) on a steam bath to digest the tissue. The alkaline The Barber-Colman gas chromatograph was also equipped mixture was extracted five times with triple volumes of pewith a radioactive monitoring system, model 5190 which was L( troleum ether. We have found that treatment with alkali in used for examination of the i4C-labeled compounds. The above system was equipped with a stream splitter which allowed 10 parts of the effluent to flow to the proportional counter and 1 part to the mass detector. Before entering the counter, the carrier gas was diluted 10% with quench gas (propane) at a flow rate of 6 ml per min. The mass of the sample and its radioactivity were recorded simultaneously on separate recorders. The columns, 6 foot by 4 mm, were packed with 3% OV-17 on Gas-chrom Q (100 to 120 mesh). The column temperature was 265". Other conditions, flash heater 270" and detector 275". Column packings were purchased from Applied Science Laboratories, Inc., State College, Pennsylvania. Bcu-Cholestane was used as a reference standard.
Where possible unknowns were compared to known compounds. Where a standard compound was not available, the retention times of the proposed compounds were calculated by determining the contribution of various substituents to the retention times, as has been described by Clayton (20).
Radioactive determinations were made with an Ansitron liquid scintillation spectrometer as previously described (21). Thin layer chromatography was carried, out on Silica Gel H (0.50-mm thickness).
The labeled compounds were located by chromatographing known reference compounds. Bands corresponding to the reference substances were then scraped from the plate and eluted with ethyl acetate.
Gas-liquid chromatography-mass spectroscopy was conducted on an LKB 9000 gas chromatograph-mass spectrometer. The column, a 3a/, OV-17 on Gas-Chrom Q, was operated at 260". The carrier gas was helium at 30 ml per min. Other conditions were flash heater 280" and molecular separator 270". The ionizing voltage was 70 e.v.
Incubations-Minced preparations of 15-day-old developing rat brain were prepared in 0.15 M phosphate buffer, pH 7.4 and incubations were carried out as previously described (15).
Minced tissue derived from 1.5 g of brain tissue, was incubated for 20 hours in a shaking incubator at 37" with 2.5 PC of m,-[2-i4C]mevalonic acid (4.60 mCi per mmole, the penicillin G salt of NJ'-dibenzylethylenediamine (DBED) ; New England Nuclear) per flask. Each incubation flask had a total volume of 3 ml. A total of 15 g of minced brain was incubated with a total of 25 &i of [2-14C]mevalonic acid. Previous examination (21) has shown that the metabolism of the labeled precursor was not due to bacterial or other contamination over this lengthy incubation period, which is required for sterol formation. Because we found that incubations of labeled squalene, squalene oxide, and lanosterol of the length used for mevalonic acid (20 hours) did not demonstrate increased turnover with time and did show significant auto-oxidation, incubation of these labeled compounds was carried out for only 6 hours. It was felt that an incubation of at least this length was necessary to see any significant metabolism, for we have shown (19) that while squalene formation is essentially complete under subcellular conditions at 2 hours an additional 6 hours are necessary to accomplish maximal sterol synthesis.
Tissue content and total volume per flask was the same as for the mevalonic acid incubations above.
To the incubations were added 30 ml of ethanolic potassium hydroxide (15%); the mixtures were then warmed briefly (1 to this manner reduces the percentage of 14C-esters content by less than 1% when compared to direct extraction of the incubation mixture with chloroform-methanol (2: 1). The petroleum ether extract is referred to in the text as the neutral lipid fraction.
The neutral lipid from the incubated tissue was placed on an alumina column (60 g), which was eluted first with petroleum ether (500 ml) to remove squalene, benzene (250 ml) to remove steryl esters, and then stripped with ethanol (200 ml) to remove free sterols. The free sterols were then further fractionated by thin layer chromatography.
The steryl ester fraction was saponified by refluxing 1 hour in the presence of ethanol potassium hydroxide (15%).
The free sterols were then extracted with petroleum ether and fractionated on thin layer as described above.
In certain cases, digitonides were prepared (22) and subsequently cholesterol was isolated from these digitonides and purified by the dibromide method (23).
The material that was eluted with petroleum ether was concentrated and further fractionated on thin layer chromatography in a solvent system of petroleum ether only. Squalene had an RF of 0.40 when the plate was developed at 4". This region was scraped and eluted for further examination by gas-liquid chromatography.
The above squalene and sterol regions were examined by means of gas-liquid chromatography, radioactive monitored liquid chromatography, and gas-liquid -chromatography-mass spectroscopy.
Amounts of sterol present were quantitated from the gas-liquid chromatograms by triangulation.

Identijkation of the Endogenous Free Sterols Present in Developing Rat Brain-The
individual regions from the above described sterol thin layer chromatography were subjected to gas-liquid chromatography and gas-chromatography-mass spectrometry. The relative retention times of reference compounds and the various unknown sterols found are indicated in Table I.
The following 4,4-dimethyl sterols were isolated from the 4,4-dimethyl sterol thin layer fraction: lanosterol, 4,4-dimethyl-5cu-cholesta-8,24-dien-3/I-01, and 4,4-dimethyl-5oL-cholest-g-en-3fi-01. The first two compounds have previously been identified in adult brain (16). The mass spectrum of the third is shown in Table II. The quantity of each of these free sterols and the free sterols to be described later in this section is shown in Table III.
The 4ol-methyl sterol region yielded four sterols.  I&nti$cation of Endogenous Esteri$ed Sterols Present in Developing Rat Brain-Gas-liquid chromatographic analysis of the sterols recovered from the steryl ester fraction resulted in a sterol mass distribution quite unlike the free sterols (Table  IV).
Whereas in the free sterol fraction the 4cY-methyl sterols were least in abundance and cholesterol the most abundant, in the esteritied sterols, cholesterol, while still the major sterol, constituted only about one-half of the steryl ester. The 4~ methyl sterols as a group were the most abundant component of the ester fraction.
Neither lanosterol nor desmosterol was detected.

Biosynthesis of Isoprenoid Material by Developing Rat Brain
In Vitro-The incorporation of [2-14C]mevalonic acid into brain neutral lipid fractions gave the following distribution after alumina column chromatography, isoprenoid hydrocarbon, 1.54 x lo5 dpm; steryl esters, 1.41 X lo6 dpm; and 9.69 x lo6 dpm, free sterols.
The isoprenoid hydrocarbon fraction was further fractionated on thin layer chromatography as described.
The mass spectra of the first two have been described in adult rat brain (16). The mass spectra of the second two are shown in Table II. Little free 4a-methyl sterol was present in developing brain (Table  III), for these sterols appear in the lowest concentration of any of the free sterols detected in the brain.
The 4-desmethyl sterol region yielded only cholesterol and desmosterol.
When the individual thin layer regions were analyzed by gas-liquid chromatography for the presence of endogenous sterols, the distribution previously described in Table  III was indicated. When this gas-liquid chromatographic analysis was accompanied by radioactive monitoring of the effluent with the proportional counter that has been described, it was found that few of these endogenous sterols were labeled by incubation with labeled mevalonic acid. Those sterols that were determined to be highly labeled gave radioactive peaks in the proportions indicated in Table V. While other 4-desmethyl sterols are no doubt labeled in addition to 5acholesta-7,24-dien-3/3-01, no other radioactive peaks were found, suggesting labeled sterols of much lower specific activity. Examination of this same fraction by the cholesterol dibromide method indicated no significant labeling of cholesterol.
When lipophilic precursors of cholesterol were incubated with minced 15-day-old brain, a direct relationship was indicated between the proximity of the precursor to the sought end product, cholesterol, and the degree to which the precursor Analysis both by digitonide preparation and thin layer chromatography (Table VI) indicated greater turnover of lanosterol as compared to squalene oxide and in turn greater turnover of squalene oxide than of squalene.
Sufficient quantities of highly labeled sterols were present in several instances to allow gas-liquid chromatography-radioactive monitor determinations (Table VII). Although a large portion of the [14C]nonsaponiflable material from the lanosterol incubation was 4,4-dimethyl in nature, the radioactive peaks that this region yielded indicated that no lanosterol was present. Instead a mixture of mono-and diunsaturated Czs sterols was found.
The same pattern was present in the 4Lu-methyl region. This labeling of monounsaturated methyl sterols is quite different than the mevalonate experiments where only diunsaturated sterols were significantly labeled. Lanosterol and 4,4-dimethyl-5Lu-cholesta-8, 24-dien-3fl-o1 were the major labeled sterols when [14C]squalene was used as the precursor.
In none of the above lipophilic precursor incubations could significant labeling of cholesterol be shown, when the 4-desmethyl fractions were examined by formation of cholesterol dibromides.
When the 4-desmethyl regions from mevalonate and lanosterol incubations were fractionated by means of AgN03 impregnated thin layer chromatography, 79oj, of the radioactivity from the [2-14C]mevalonic acid incubation was found in the diunsaturated fraction while only 57% of the 4-desmethyl region from the lanosterol incubation was found in the diunsaturated fraction, another indication that the side chain double bond was saturated much earlier in the lanosterdl incubation than in the mevalonate incubations.

DISCUSSION
The literature on the presence and metabolism of methyl sterols in developing brain is limited.
As stated previously, the only extensive study of the area has dealt with methyl sterols and labeling of these by means of mevalonic acid administration in chick embryo (6). One might assume that there are general similarities between sterol biosynthesis in developing chick brain and developing rat brain.
Indeed, this has proved to be the case. Several marked differences were found, however. The present study has been able to establish the presence of several AT-double bond methyl sterols in developing rat brain in addition to the A*-double bond methyl sterols isolated from developing chick brain.
Methyl sterols containing a A7-double bond are not unique to the developing brain, for we have found that several methyl sterols with A7-double bonds are also present in adult rat brain (16).
To better assess the role(s) of these various methyl sterols in developing brain, incubations were carried out using [2-14C]mevalonic acid and several lipophilic cholesterol precursors. Again, similarities were found to the chick embryo studies. Mevalonate was found to label primarily the diunsaturated sterols and, in addition, only those methyl sterols having a A*, as well as a Az4, double bond. In variance to the chick, where lanosterol was not labeled or found, this precursor was present in the developing rat brain and was heavily labeled by mevalonic acid. 5cr-Cholesta-7,24-dien-3fl-ol was identified in both tissues but was found to be highly labeled only in the rat brain, whereas, in the chick brain, 5a-cholesta-8,24-dien-3fl-01 was found to be a highly labeled sterol.
We have found in our own studies that sterol biosynthesis in developing brain is quite different from adult brain with respect to the structures of the intermediates formed.
Incubation of cell-free preparations of adult rat brain with mevalonic acid indicated the formation of two unusual C-29 and C-28 diunsaturated sterols (16). Intracerebral administration of mevalonic acid and subsequent sacrifice after a short period of time produced labeled methyl sterols much like those reported herein in in vitro developing brain.
There was one important exception.
The significance of these differences in isomerization cannot be assessed at this time.
The incubation of the labeled squalene, squalene oxide, and lanosterol with the developing brain also suggests some differences in metabolism, depending on the position of the incubated cholesterol precursor in the general pathway of cholesterol formation.
Generally, no extensive turnover was indicated, particularly in the case of squalene. The very lipophilic nature of these precursors and the difficulty in solubilizing them may be the cause of this limited turnover.
In the case of lanosterol, one of the methyls seems to be removed readily, for analysis of the 4,4-dimethyl fraction showed only the presence of C-29 sterols, suggesting that the 14cY-methyl group of lanosterol is easily removed by brain.
Another interesting point concerning the [14C]lanosterol studies is the seemingly easy reduction of the A24-double bond, for both mono-and diunsaturated C-29 and C-28 were found to be labeled. This is in direct opposition to the results of the mevalonate incubations where only diunsaturated methyl sterols were labeled.
The above observation suggests that the lanosterol formed from [2-l%]mevalonate was not exposed to the A24 reductase as was the incubated [14C]lanosterol.
The presence of sterols other than cholesterol in the steryl ester fraction could have considerable significance with regard to sterol biosynthesis in developing brain.
Most of the emphasis with regard to steryl esters in developing brain has been on the content of the fatty acid moiety (9,11,14). Presumably the researchers dealing with these esters felt that the sterol esterified was cholesterol.
The role of the ester may take on added meaning once it is established that the ester may be biosynthesis, and not cholesterol itself. That cholesterol precursors, primarily methyl sterols, were esterified in the skin was suggested a number of years ago (24). It is also known that in animal tissues other than brain, methyl sterols are generally esterified to a greater degree than is cholesterol and other 4-desmethyl sterols (25, 26). Brady and Gaylor (27) have demonstrated that esterification of and demethylation of methyl sterols in the liver are in direct competition; methyl sterols that are esterified are not easily demethylated. Brady and Gaylor have also shown that in the skin the rate of esterification of methyl sterols is greater than the rate of demethylation.
What has not been established in the case of skin or brain is the metabolic role of steryl esters. In brain it could be a means of regulation of sterol formation by impeding methyl sterol demethylation.