Utilization of L( +)-3=Hydroxybutyrate, D( -)-3-Hydroxybutyrate, Acetoacetate, and Glucose for Respiration and Lipid Synthesis in the 1%Day-old Rat*

A comparison has been made in vivo between IS+)-3- hydroxy[3-Wlbutyrate, o(-)-3-hydroxy[3-“‘Clbutyrate, [3-Wlacetoacetate, and n-[Z-Wlglucose for sterol and fatty acid synthesis and respiration in the B-day-old suckling rat. (a) Sterols and fatty acids in spinal cord, brain, and skin were preferentially labeled by these metabolites over sterols and fatty acids in the liver and kidneys. (b) More label was incorporated into sterols and fatty acids in spinal cord, brain, and kidneys from L( + )-3-hydroxy[3-14Clbutyrate than from o(-)-3-hydroxy[3-14CIbutyrate. (c) More label was incorporated into sterols and fatty acids in spinal cord, and skin from than from [3-14Clacetoacetate. (c-z’) In all organs

From the Department of Biological Chemistry and Mental Retardation Research Center, School of Medicine, University of California, Los Angeles, California 90024 A comparison has been made in vivo between IS+)-3hydroxy  o(-)-3-hydroxy[3-"'Clbutyrate,  and n-[Z-Wlglucose for sterol and fatty acid synthesis and respiration in the B-day-old suckling rat. (a) Sterols and fatty acids in spinal cord, brain, and skin were preferentially labeled by these metabolites over sterols and fatty acids in the liver and kidneys. (b) More label was incorporated into sterols and fatty acids in spinal cord, brain, and kidneys from L( + )-3-hydroxy  than from o(-)-3-hydroxy  (c) More label was incorporated into sterols and fatty acids in spinal cord, brain, and skin from DC-)-3-hydroxy  than from  (c-z') In all organs less label was incorporated into sterols and fatty acids from o- [2-14Clglucose than from the other metabolites; unexpectedly poor were the liver and kidneys which contained substantially less label. (e) The retention of label from D( -)-3-hydroxy[3-14Clbutyrate in the sterols and fatty acids from spinal cord and brain was investigated.
(0 The time course of evolution of W02 over 2% h from each of these metabolites revealed a more rapid utilization of [3-14C]acetoacetate maximum at 10 min than D(-))-3-hydro~y[3-~~Clbutyrate maximum at 30 min; by contrast, label from o-  and L(+)-3-hydroxy[3-14Clbutyrate was retained maximally in metabolic pools over a 2-h period, indicating a much slower utilization.
(g) The evidence that the L(+)-3-hydro~y  is a favored substrate for the synthesis of sterols and fatty acids but less favored for oxidation, while ~(-)-3-hydroxy[3-'~Clbutyrate is a favored substrate for oxidation but less favored for the synthesis of sterols and fatty acids, suggests that these isomers are preferentially metabolized in different compartments.
The ketone bodies, acetoacetate and 3-hydroxybutyrate, have been shown to be efficient precursors for the biosynthesis of sterols and fatty acids in the central nervous system at early stages in the development of the rat (l-3). It was shown that acetoacetate and 3-hydroxybutyrate were preferentially utilized by the organs of ectodermal origin, the brain, spinal cord, * This research was supported by United States Public Health Service Research Grants HD-06576, HD-04612, and GM-00364 from the National Institutes of Health.
$ To whom inquiries should be addressed. and skin, over the lung, kidney, and liver for sterol and fatty acid synthesis in developing rats at 9 to 11 days after birth (3).
The data indicated that m-3-hydroxy[3J4C]butyrate labeled fatty acids and sterols more efficiently than did [3-"C]acetoacetate, suggesting that the former substrate was the preferred precursor for the biosynthesis of lipids. Although the comparison was made by injection of equal amounts of mass and of label of each substrate, this was an unexpected finding (3). It was thought that only 50% of the m-3-hydroxybutyrate, the physiological DC-1 isomer, would be utilized by the DC-)specific P-hydroxybutyrate dehydrogenase (EC 1.1.1.30) (4). In addition, it has been shown that the concentration of 3-hydroxybutyrate in the circulation of the young rat was 2 to 5 times greater than the concentration of acetoacetate (5) and therefore the specific activity of the circulating acetoacetate should have been 4 to 10 times higher than that of 3-hydroxybutyrate after the injection. Contrary to expectation, the cholesterol labeled from [3J4C]acetoacetate had a lower specific activity than the cholesterol labeled from 3-hydroxy[3-'%]butyrate (3). Additionally, the rate of uptake of acetoacetate by the brain at a given concentration is twice the rate of uptake of 3-hydroxybutyrate (5). It was expected from these three observations that acetoacetate should be the favored substrate, but this was not found (3). Two of the possibilities that could explain this apparent anomaly were investigated in this study. (a) Acetoacetate, as compared to 3-hydroxybutyrate, may have been preferentially utilized for respiration, an alternative metabolic fate, and therefore less would have been available for lipid synthesis. (6) Since the racemic mixture of 3-hydroxybutyrate was used in the previous study (31, the L(+) isomer may have made an unanticipated contribution to the synthesis of lipids.
In this study we report on a comparison of L( +)-3-hydroxy [3-14C]butyrate, D(-)-3-hydroxy[3J4C]butyrate, [3J4C]acetoacetate, and n-[2J4C]glucose as a source of carbon for respiration and as precursors of sterols and fatty acids in various organs in the suckling rat at 18 days after birth. It is known that at this age the ketone body concentration in the circulation, as well as the level of the various enzymes needed for ketone body utilization in the brain, are at a maximum (5)(6)(7). We have found that although L(+)-3-hydroxybutyrate was oxidized to CO2 much less efficiently than the D( -) isomer, it was superior t0 the other metabolites as a substrate for sterol and fatty acid synthesis in the brain, spinal cord, and kidney. Long Term Studies -Ten 18-day-old rata from a single litter were injected between the scapulae with 10.0 &i of D(-)-3-hydroxy [3-14C]butyrate (26 Cilmol) in less than 0.025 ml of 0.9% NaCl. Pairs of rats were killed by decapitation at 4 h and at 30, 60, 90, and 120 days her the injection.
The '*C content in the sterol and fatty acid fractions in spinal cord, brain, skin, kidney, and liver was determined as described above.

AND DISCUSSION
Lipid Synthesis by Three Hours-The amount of label incorporated into each of the lipid fractions in organs after the administration of equivalent amounts of L( t )-3-hydroxy [3-14C]butyrate, D(-)-3-hydroxy[3-'4C]butyrate, [3-14C]acetoacetate, or D[2J4C]glucose is shown in Tables I and II. The sterol and fatty acid fractions in spinal cord, brain, and skin are labeled preferentially over the sterol and fatty acid fractions  in the lo-day-old rat (1,2), and results from the differences in the ability of these organs to produce or utilize acetoacetate and D( -)-3-hydroxybutyrate, a feature determined by their enzymic complement during the developmental period (1). To our surprise we found that L(+)-3-hydroxy[3J4C1butyrate not only labeled the sterols and fatty acids in the various organs in a pattern qualitatively similar to D( -)-3-hydroxy  but also that twice as much label was incorporated from it into the sterol fraction in spinal cord, brain, and kidney as from D(-)-3-hydroxy[3-"C]butyrate (Table I) and 55 to 70% more label was incqrporated from it than from the D(-) isomer into the fatty acid fraction in these organs (Table II). Significantly more label was also incorporated into the sterol and fatty acid fractions of the skin, spinal cord, and brain from D(-)-3-hydroxybutyram than from acetoacetate (Tables I and II).
These findings explain the observation made previously that label from DL-3-hydroxyl3-"C1butyrat.e was incorporated more efficiently into the sterols and fatty acids in the central nervous system in neonatal rats than label from [3-"C]acetoacetate.
Since D(-)-3-hydroxybutyram is converted to acetoacetate by the D( -)-specific 3-hydroxybutyrate dehydrogenase in the mitochondrion, our findings led us to believe there may exist an alternative route for the metabolism of 3hydroxybutyrate.
However, this seems unlikely since a comparison of Dr.,-3-hydroxy[3-Wlbutyrate and DL-3-hydroxyD-3H1butyrate showed that sterols and fatty acids in tissues of the developing rat were poorly labeled by the latter, less than 10% of that observed from an equivalent dose of DL-3-hydroxy[3J4C!]butyrate (14). Since the label from DL-3-hydroxy[3-3H]butyrate was not distributed preferentially to either fatty acids or sterols and 90% of the tritium was lost during metabolism, it was concluded that the utilization was mediated via acetoacetate or acetoacetyl-CoA (14). A comparison of the incorporation of label in sterols and fatty acids from L(f)-3-hydroxy[3-Wlbutyrate and from 14Clabeled ketone bodies and glucose shows that the incorporation of label from glucose is very poor (Tables I and II). In the organs of ectodermal origin 4 to 10 and 2 to 7 times less label was incorporated from D-[2J4Clglucose than from D(-)-3-hy-droxy13-'4C]butyrat.e in the sterols and fatty acids, respectively (Tables I and II). The incorporation of label from D-[2-'YJglucose into the lipids of the liver and kidney was unexpectedly poor. The sterols in liver and kidney contained 12 to 40 times less label from D [2-14C]glucose than from D(-)-3hydroxy[3-*4C]butyrate; fatty acid in liver and kidney contained 3 to 10 times less label. It is generally accepted that the ketone bodies are poorly utilized by the liver of the developing rat (15, 16), since it is known that the liver does not contain 3oxo-acid succinyl-CoA transferase (EC 2.8.3.5) (16). Our studies also show that glucose is not efficiently utilized by the liver or kidneys for lipid synthesis. This finding supports the contention that during development the liver and kidneys are primarily committed to the simultaneous production of ketone bodies (17,18) and glucose (19) with negligible synthesis and metabolism of glycogen (20,21).
Retention of Label in Lipids from o(-)-3-HydroxylS-'Clbutyrate-The retention of label in the sterol and fatty acid fractions in tissues after the injection of D( -)-3-hydroxy- [3-'4C]butyrate into 18-day-old rats was measured at intervals up to 120 days after the injection. Ey 30 days after the injection the 14C content in the lipids of skin, kidney, and liver was negligible.
A comparison of the amounts of label retained in the sterol and fatty acid fractions in spinal cord and brain is presented in Fig. 1. The 14C content in the fractions at 30, 60, 90, and 120 days after injection is plotted as a percentage of the 14C content in these fractions 4 h after the injection.
The amount of label retained in sterols in spinal cord over 120 days after the injection of D(-)-3-hydroxy  was significantly different than the amount retained in brain (Fig. 1). In spinal cord only 10% of the label in sterols at 4 h after the injection was lost by the first time point (30 days). After this small decrease the 14C content in the sterols in the spinal cord remained relatively stable. In brain 10% of the label present at 4 h was also lost by the first time point; however, by the 60th day an additional 30% was lost. Thereafter the level remained relatively constant. These findings are in agreement with the studies reported by Smith and Eng (22) and Hajra and Radin (23). In contrast to the small decrease in the 14C content of the sterols, 70 to 80% of the label in the fatty acids in spinal cord and brain disappeared by 30 days after the injection of D(-)-3-hydroxy[3J4Clbutyrate, but thereafter the 14C content of the fatty acids remained relatively constant for the next 90 days. This is also in agreement with the studies reported by Hajra and Radin (23).
In these previous studies (22, 23) [l-14Clacetate was used as the precursor for lipid synthesis. However, acetate is not a circulating metabolite in the rat. For this reason the retention of label by the sterols and fatty acids in the spinal cord and brain from a natural metabolite, D(-)-3-hydroxy[3-'*C]butyrate was investigated.
It has been shown that label from butyrate was preferentially incorporated into the sterols and fatty acids in spinal cord, brain, and skin in the lo-day-old rat (3). It has also been shown that within minutes after the injection of 12J4C1acetate the ketone bodies in the circulation of the 18-day-old rat contain 'C (14). This may explain the similarity in the preferential distribution of label from acetate and ketone bodies into sterols and fatty acids in the organs of ectodermal origin (3).
A comparison of the amount of 14C incorporated into the TIME (DAYS) FIG. 1. Ten 18-day-old rate from a single litter (mean body weight, 31.9 ? 0.8 g) were injected with 10 &i of u(-)-3-hydroxyl3-"Clbutyrate (26 Ci/mol) between the scapulae and two were killed at 4 h and at 30, 60, 90, and 120 days after injection. The 14C content in organs was determined ("Materials and Methods"). The amount of label in the fractions at the various times is shown as a percentage of the label in the equivalent fraction at 4 h aRer injection. The total 14C content in sterol and fatty acid fractions in brain was 36,700 and 44,900 dpm and in spinal cord was 25,200 and 18,800 dpm, respectively, at 4 h after the injection. Sterol fractions: brain, 0; spinal cord, 0. Fatty acid fractions: brain, 0; spinal cord, n . lipids in spinal cord and brain shows that spinal cord has a greater capacity per unit weight than the brain to utilize DC-)-3-hydroxy[3-Wlbutyrate for the synthesis of sterols and fatty acids, Tables I and II. This suggests that myelination in the spinal cord is more active than in the brain at 18 days after birth. Also, a larger fraction of the label initially incorporated is retained in the sterol fraction in spinal cord as compared to brain as a function of time after the injection of D(-)-3-h--droxyl3JClbutyrate in the 18-day-old rat ( Fig. 1). This should be expected since the spinal cord and its myelin lamellae are more mature than the brain and its myelin lamellae at this time (24).
Respiration -The time course of the oxidation of each substrate as monitored by the expiration of 14C02 from the intact rat gave an indication of how long the metabolic pools were maximally labeled from each substrate (Fig. 2). There was a very rapid evolution of 14C0, from [3J4C]acetoacetate which had a maximum 10 min after injection. However, the evolution of 14C0, from D( -)-3-hydroxy[3-'4C]butyrate did not reach a maximum until 30 min after the injection. This comparison suggests that during the longer period in which the metabolic pools were labeled by D(-)-3-hydroxy13J4Clbutyrate, labeled precursors were available for lipid synthesis and may have accounted for the heavier labeling of the sterols and fatty acids as compared to the labeling from acetoacetate (Tables I and  II). Since the profiles of 14C0, evolution from [3-"C]acetoacetate and D(-)-3-hydroxy13-14C]butyram by the 18day-old rat are quite distinct (Fig. 21, it would appear that once acetoacetate and D(-)-3-hydroxybutyrate are present in the circulation there is minimal interconversion. This observation is currently under investigation as it has been assumed that there was rapid equilibration between acetoacetate and 3hydroxybutyrate in the circulation of the developing rat (25).
H 'j&j  Table I. Expired "CO, was determined as described under "Materials and Methods." '*CO, was collected over 5-min intervals until 1 h, then at 15-min intervals until 2 h after the injection. A final collection was made at 2l/z h. The data are plotted in disintegrations per min x 1O-4 in "CO, expired per min at t'he end point of each interval after the injection. All the data points are not included since they would only crowd the curves without improving their definition. 14C0, from L(+)-3-hydroxy[3-"Clbutyrate, 0-O; n(-) A. . . A. In the preparation of the data in Fig. 2 corrections have not been made for the different specific activities of the substrates in the circulation; thus the amplitude of each curve is relative. However, the profile of each curve is an indicator of the time carbon from each labeled precursor persisted in the metabolic pools. In contrast to acetoacetate and D( -)-3-hydroxybutyrate which were oxidized rapidly, D[2J4C]glucose and L(+)-3-hydroxy[3J4C]butyrate must have been utilized relatively slowly since label from both of these metabolites persisted maximally in the metabolic pool longer than that from either DC-)-~hydroxybutyrate or acetoacetate. By 150 min 68% of the label injected as either  or D(-)-3-hydroxy [3-14C]butyrate was expired as 14C0,. However, by 150 min only 38% of the label injected as glucose and 42% of the label injected as L(+)-3-hydroxybutyrate was expired as 14C02. Although label from D[ 2J4C]glucose persisted maximally in the metabolic pool longer than that from D( -)-3-hydroxybutyrate or acetoacetate, the sterol and fatty acid fractions were not as extensively labeled from glucose as from the other substrates. Hawkins et al. (5) have shown that the concentrations of 3hydroxybutyrate and glucose in the circulation of 16-to 18day-old rats were about 2 and 6 mM, respectively. Even if one were to consider that the specific activity of circulating 3hydroxybutyrate must have been about 3 times greater than that of circulating glucose during the early parts of the experiment, the amount of label found in sterols and fatty acids from D(-)-3-hydroxy [3-14C]butyrate was much greater than expected (cf . Tables I and II). Current experiments seek an explanation for the apparent slow utilization of label from Dglucose and the reason why label from glucose appears to remain maximally in metabolic pools for extended periods.  (Tables I and II). The amount of label expired as 14C0, from L(+)-3-hydroxy[3-"C1butyrate over the 2Vz-h period was one-third less than the amount of label expired from either D(-)-3-hydroxyl3-"C1butyrate or 13-'4C]acetoacetate. However, by 3 h more label from L(+)-3-hydroxy[3J4C]butyrate was incorporated into sterols and fatty acids in all the organs studied than from either ~(-)-3-hydroxyl3-~~Clbutyrate or 13-'4Clacetoacetate. The poorer oxidation of the DC+)-3-hydroxy13J4Clbutyrate to CO, may reflect in part its more extensive incorporation into sterols and fatty acids. The evidence that the L(+) isomer of 3hydroxybutyrate is a favored substrate for lipid synthesis but less favored for respiration, while the D(-) isomer is the favored substrate for respiration but less favored for lipid synthesis, suggests that these isomers are preferentially metabolized in different compartments.
In the liver and kidney the amount of label incorporated by 3 h from L(+)-3-hydroxy [3-W]butyrate is 20 to 40 and 5 to 10 times greater in sterols and fatty acids, respectively, than the amount of label incorporated from [ 2J4C]glucose. Although label from both [2J4C]glucose and DC+)-3-hydroxy[3-14C1butyrate persisted in metabolic pools for over 2 h as suggested by the time course of evolution of expired 'YJO,, it was surprising that the sterols and fatty acids in liver and particularly kidney should be more efficiently labeled from L(+)-3hydroxy[3-'4C]butyrate than from 12-'4C]glucose. This is indeed striking since L(+)-3-hydroxybutyrate has yet to be reported as a circulating metabolite.
of L( +)-3-Hydroxybutyrate and Ketone Bodies Early studies by McKenzie (261 and Lehninger and Greville (27) suggested that mammalian tissue had the capacity to metabolize L( +)-3-hydroxybutyrate in uiuo and in uitro. It was thought, however, that the metabolism of the L(+) isomer was physiologically unimportant (4,27,28). McKenzie, as part of his studies on the resolution of the isomers of 3-hydroxybutyrate, showed that the urine of a dog, which had been fed gram quantities of nL-3-hydroxybutyrate, contained ketone bodies enriched in the levorotatory n( -l-3-hydroxybutyrate (26). He concluded that the dog had the capacity to utilize preferentially the L(+) isomer. Early in vitro studies on the metabolism of 3-hydroxybutyrate by Lehninger and Greville (27) showed that L( + )-3-hydroxybutyrate was metabolized but to a lesser extent than the D(-) isomer. However, it was established that the 3-hydroxybutyrate dehydrogenase associated with the mitochondrion had an absolute stereospecificity for n(-)-3-hydroxybutyrate (29, 301, and Klee and Sokoloff demonstrated that the mitochondria from the brain of developing rats did not metabolize L(+)-3-hydroxybutyrate to acetoacetate (4). It is worthy of note that 3-hydroxybutyrate is unique in that both isomers are found in intermediary metabolism; n(-)-3-hydroxybutyrate is found as a circulating metabolite and as D( -)-3-hydroxybutyryls-fatty acid synthetase during fatty acid synthesis, while L( +)-3-hydroxybutyrate is found as L(+)-3-hydroxybutyryl-coenzyme A in the p oxidation of fatty acids. The suckling rat, especially at 18 days after birth, has serum ketone body concentration approaching 2 mM (5) from very active ketogenesis which is fueled by the fl oxidation of fatty acids derived from dietary triglycerides. L(+)3-Hydroxybutyryl-coenzyme A may be an intermediate of greater importance during the suckling period than previously thought, should L(+)-3-hydroxybutyrate prove to be a circulating metabolite generated by active p oxidation. The physiological significance of the metabolism of L(+)-3-hydroxybutyrate during the suckling period in the rat remains to be determined.