Am J Physiol Endocrinol Metab
279: E221–E227, 2000.
Fatty acid cycling in the fasting rat
BELLA KALDERON, NINA MAYOREK, ELLIOT BERRY,
NOAM ZEVIT, AND JACOB BAR-TANA
Department of Human Nutrition and Metabolism, Faculty of Medicine,
Hebrew University, Jerusalem 91120, Israel
Received 1 November 1999; accepted in final form 5 January 2000
THE TRANSITION to the fasting state, in which free fatty
acids (FFA) derived by adipose tissue lipolysis become
a major source of oxidizable metabolic fuel for heart,
skeletal muscle, and liver, is accompanied by major
changes in overall lipid and fatty acid metabolism. The
FFA cycle is initiated by lipolysis of adipose triglycerides (TG) catalyzed by hormone-sensitive lipase (HSL).
Fat lipolysis may either yield plasma FFA or result in
intracellular reesterification of lipolyzed FFA back into
adipose fat before their release (primary reesterification). Plasma FFA that escape direct oxidation are
thought to be secondarily reesterified in liver into very
low density lipoprotein-triglycerides (VLDL-TG) (17,
41). These are hydrolyzed by intravascular lipoprotein
lipase, followed by reuptake of the released FFA into
adipose tissue, muscle, or liver. However, secondary
reesterification of FFA into VLDL-TG has recently
been shown in humans to account for only a minor
fraction of total FFA released from adipose tissue, thus
leaving unsolved the fate of plasma FFA that escape
oxidation or secondary reesterification into VLDL-TG
(12).
Long-chain fatty acid metabolism is closely associated with calorigenesis induced by hormones and
drugs and is therefore of particular interest under
conditions of treatment with thyroid hormones (TH).
However, despite studies concerned with the effects
exerted by TH on discrete steps of fatty acid metabolism, a quantitative and integrative in vivo view of
long-chain fatty acid metabolism in TH-treated animals is still lacking. Thus TH was reported to induce
lipolysis in isolated adipose tissues (1) or adipocytes
(18, 26) as a result of sensitization of adipocytes to the
lipolytic effect of catecholamines (5, 18). Increased lipolysis was corroborated by increased total body lipid
oxidation as well as a decrease in adipose tissue weight
(31). However, an integrative view consisting of quantitative evaluation of the in vivo production rates of
FFA and glycerol as well as the FFA fraction that is
either oxidized or secondarily reesterified in rats
treated with TH is still lacking. Similarly, despite in
vitro evidence suggesting that hepatic VLDL release
(11, 39) and lipoprotein lipase activity (10, 34) are
affected by TH treatment, the in vivo production rate of
hepatic VLDL-TG as a fraction of the overall fatty acid
flux still remains to be investigated.
Calorigenesis induced by ␤,␤⬘-tetramethylhexadecanedioic acid (Medica 16) (2) is apparently similar to
that induced by TH. Thus, similarly to TH, Medica 16
treatment results in sensitization of adipose tissue to
catecholamines, leading to lipolysis, ketogenesis, and
weight reduction (38). Furthermore, similarly to TH,
Medica 16 treatment induces a pronounced decrease in
liver phosphate and redox potentials (24, 25) with a
concomitant decrease in mitochondrial membrane potential observed in both isolated mitochondria and
liver cells (19). The thyromimetic activity of Medica 16
is further reflected by its capacity to induce liver genes
classically considered to be TH dependent, e.g., malic
enzyme, mitochondrial glycerol-3-phosphate dehydro-
Address for reprint requests and other correspondence: J. BarTana, Dept. of Human Nutrition and Metabolism, Faculty of Medicine, The Hebrew Univ., PO Box 12272, Jerusalem 91120, Israel
(E-mail: [email protected]).
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
adipose tissue; lipoproteins; stable isotopes; thyroid hormone; Medica 16
http://www.ajpendo.org
0193-1849/00 $5.00 Copyright © 2000 the American Physiological Society
E221
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.6 on June 18, 2017
Kalderon, Bella, Nina Mayorek, Elliot Berry, Noam
Zevit, and Jacob Bar-Tana. Fatty acid cycling in the fasting rat. Am J Physiol Endocrinol Metab 279: E221–E227,
2000.—Adipose tissue lipolysis and fatty acid reesterification
by liver and adipose tissue were investigated in rats fasted
for 15 h under basal and calorigenic conditions. The fatty acid
flux initiated by adipose fat lipolysis in the fasted rat is
mostly futile and is characterized by reesterification of 57% of
lipolyzed free fatty acid (FFA) back into adipose triglycerides
(TG). About two-thirds of FFA reesterification are carried out
before FFA release into plasma, whereas the rest consists of
plasma FFA extracted by adipose tissue. Thirty-six percent of
the fasting lipolytic flux is accounted for by oxidation of
plasma FFA, whereas only a minor fraction is channeled into
hepatic very low density lipoprotein-triglycerides (VLDLTG). Total body calorigenesis induced by thyroid hormone
treatment and liver-specific calorigenesis induced by treatment with ␤,␤⬘-tetramethylhexadecanedioic acid (Medica 16)
are characterized by a 1.7- and 1.3-fold increase in FFA
oxidation, respectively, maintained by a 1.5-fold increase in
adipose fat lipolysis. Hepatic reesterification of plasma FFA
into VLDL-TG is negligible under both calorigenic conditions. Hence, total body fatty acid metabolism is regulated by
adipose tissue as both source and sink. The futile nature of
fatty acid cycling allows for its fine tuning in response to
metabolic demands.
E222
FATTY ACID CYCLING IN THE FASTING RAT
METHODS
Animals. Male Albino rats of the Hebrew University strain
weighing 300–400 g were fed a standard laboratory chow diet
containing 4.5% fat. Rats were treated for 4–6 wk with 390
mg 䡠 kg⫺1 䡠 day⫺1 Medica 16 by mixing the drug in the
powdered diet. Hyperthyroidism was induced by six daily
intraperitoneal injections of 10 ␮g T3 䡠 100 g body wt⫺1 䡠
day⫺1 in 0.01 M NaOH in saline. Control rats were injected
with the vehicle only. All animals had free access to food and
drinking water. Animal care and experimental procedures
were in accordance with guidelines of the Animal Care Committee of Hebrew University.
Cannulation procedures. Rats fasted for 15 h were restrained in restriction cages (Harvard Apparatus) and cannulated under local anesthesia with lignocaine through the
tail artery and vein for blood sampling and isotope infusion,
respectively (28). After catheter placement, animals were
released to their cages, where they could move freely, and
allowed to recover for 90 min. Biting off of tail catheters was
prevented by wrapping the tail with a plastic sheath coated
with castor oil repellent. Catheter patency was maintained
with saline to avoid heparin during measurement of lipolytic
rates.
Indirect calorimetry. Whole body O2 consumption and CO2
production rates were measured several times before and
during the infusion study by open-circuit indirect calorimetry
using a NAGA O2/CO2 analyzer (Franztec, Haifa, Israel). The
time of urination and urine volumes during the infusion
period were recorded using an intracage moisture-sensitive
alarm. Urinary nitrogen was determined by Kjeldahl analysis in urine acidified with 1 N H2SO4.
FFA, glycerol, and VLDL-TG production rates. Cannulated
recovered animals were infused constantly for 150 min (20–
30 ␮l/min) through the tail vein with 98%-enriched [2,22
H2]palmitate (Cambridge Isotope Laboratories, Andover,
MA) (bound to albumin at a ratio of 6:1) at a rate of 0.16
␮mol 䡠 min⫺1 䡠 kg body wt⫺1 and with 98%-enriched
[2H5]glycerol (ISOTEC, Miamisburg, OH) in saline at a rate
of 0.6 ␮mol 䡠 min⫺1 䡠 kg body wt⫺1 using a Harvard Apparatus syringe pump. For measurement of FFA and glycerol
production rates, blood samples were withdrawn at intervals
of 20–30 min during the last 50- to 150-min period of con-
stant infusion for measurements of steady-state enrichment
of plasma palmitate and glycerol. For measurement of
VLDL-TG production rate, the infusion of label was stopped,
allowing for a decay in the enrichment of palmitate in plasma
VLDL-TG. The enrichment of plasma TG-palmitate was followed for 60–80 min in blood samples collected at intervals of
5–15 min. Sampled blood amounted to 1.5–3.0 ml.
Plasma free and esterified palmitates were extracted with
5 ml of isopropanol-heptane-2 N H2SO4 at a ratio of 40:10:1.
The upper heptane phase was separated and further extracted with 1 ml of alkaline 70% methanol, followed by
extraction of the alkaline methanol with heptane. The heptane phases consisting of plasma TG were combined. The
alkaline methanol containing the free fatty salt was then
acidified and extracted twice with 2 ml of heptane to recover
plasma FFA. The plasma TG and FFA extracts were subjected to further purification by TLC (heptane-diethyletherglacial acetic acid at a ratio of 157:39:3.9). The purified fatty
acids and TG were derivatized to their methyl esters as
described previously (6) and analyzed by gas chromatography-mass spectrometry (GC-MS) as described below.
Plasma glycerol was extracted from 100 ␮l of plasma by
barium hydroxide-zinc sulfate precipitation, and the supernatant was then passed through a mixed cation/anion exchange resin. The eluate was collected in reaction vials and
evaporated to dryness. Glycerol was derivatized to its tbutyldimethylsilyl derivative essentially as described previously (8). Briefly, 50–60 ␮l of anhydrous acetonitrile were
added to dried eluates, followed by the addition of 20 ␮l of
N-methyl-N-(tertbutyldimethylsilyl)trifluoroacetamide. The
reaction vials were sealed with Teflon-lined capes, heated in
a 110°C heating block for 15 min, and left to cool for another
hour at room temperature. The derivatized glycerol was
analyzed by GC-MS as described below.
Isotopic enrichment of t-butyldimethylsilyl glycerol and
methylpalmitate was determined by GC-MS analysis using a
Quatro II Fissons quadrupole mass spectrometer coupled to a
gas chromatographer (15m DB-1 GC capillary column; J & W
Scientific, Folsom, CA). The mass spectrometer was operated
in the electron impact mode at ionization energy of 70 eV and
source temperature of 180°C. The mass spectrometer was
daily tuned to the 219 and 264 m/e ions of heptacosa. Methylpalmitate enrichment was determined by selectively monitoring the 270 (M) and 272 m/e (M ⫹ 2) ions. The t-butyldimethylsilyl derivative of glycerol was determined by selectively monitoring the 217 (M) and 220 m/e (M ⫹ 3) ions. The
217 m/e ion appeared to be the most intense and stable
fragment derived from t-butyldimethylsilyl glycerol. Selectively monitoring the ions of 217 and 220 m/e provided an
accurate measurement of glycerol enrichment as verified by
analysis of glycerol standards of known glycerol enrichments.
Secondary reesterification in adipose tissue. Cannulated
recovered animals were primed with 12 ␮Ci [3H]palmitate in
saline, followed by constant infusion for 100–280 min with
0.6 ␮Ci [3H]palmitate 䡠 min⫺1 䡠 kg body wt⫺1 accompanied by
0.03 ␮mol 䡠 min⫺1 䡠 kg⫺1 fatty acid free albumin. An isotopic
steady state of plasma [3H]palmitate was reached within the
first few minutes of infusion. Three blood samples of 0.5 ml
each were then taken for measurement of specific activity of
plasma free palmitate. Animals were then killed by decapitation, and samples of the epididymal, subcutaneous, and
perirenal adipose tissues were quickly removed and stored in
liquid air.
Plasma free palmitate (100 ␮l) was extracted with isopropanol-heptane-2 N H2SO4 at a ratio of 40:10:1, as described
above, and derivatized with ␣-bromoacetophenone (29). The
bromophenyl acyl ester was purified by HPLC using a 150 ⫻
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.6 on June 18, 2017
genase, S14, and others (20, 21). However, in contrast
to TH, the calorigenic activity of Medica 16 is liver
specific. Thus, whereas TH induces a decrease in phosphate potential in both liver and heart, the decrease in
phosphate potential induced by Medica 16 is observed
in liver only (24, 25). Similarly, transcriptional activation by TH of TH-dependent genes is observed in both
liver and heart, whereas the respective nuclear activity
of Medica 16 is, again, liver specific. Also, the thyromimetic activity of Medica 16 has been shown to be
independent of TH levels and is not transduced by the
TH nuclear receptor (20, 21). The apparent liver specificity of Medica 16 has been shown elsewhere to be
accounted for by its failure to enter cardiac or skeletal
muscle (H. Bdeir, unpublished observations).
Long-chain fatty acid metabolism was evaluated
here in the fasting rat in terms of the in vivo fluxes of
adipose tissue lipolysis, fatty acid oxidation, and the
recycling of fatty acids in liver and adipose tissue as a
function of whole body calorigenesis induced by TH or
hepatic calorigenesis induced by Medica 16.
FATTY ACID CYCLING IN THE FASTING RAT
Table 1. Plasma insulin and metabolites in
postabsorptive nontreated and TH- and Medica
16-treated rats
Treatment
FFA, mM
TG,
mg %
Glucose,
mg %
Insulin,
ng/ml
Nontreated
TH
Medica 16
1.0 ⫾ 0.1
1.0 ⫾ 0.1
0.8 ⫾ 0.1
93 ⫾ 6
84 ⫾ 6
39 ⫾ 1*
72 ⫾ 2
89 ⫾ 4*
86 ⫾ 4*
0.7 ⫾ 0.2
1.4 ⫾ 0.2*
1.8 ⫾ 0.3*
Plasma parameters were determined in rats treated as described
in METHODS. Values are means ⫾ SE (n ⫽ 4–6 rats). FFA, free fatty
acids; TG, triglycerides; TH, thyroid hormones. * Significantly different from respective nontreated animals (P ⬍ 0.05).
R a ⫽ [(IE in/IE p) ⫺ 1] ⫻ F
where F is the isotope infusion rate (in ␮mol 䡠 min⫺1 䡠 kg⫺1),
IEin is the isotopic enrichment of the infusate, and IEp is the
isotopic enrichment of the tracer in plasma at isotopic steady
state, expressed in mole percent excess. The rate of appear-
ance of FFA (Ra FFA) was calculated by dividing Ra palmitate by the ratio of palmitate to total FFA in plasma.
Primary intra-adipose reesterification rates (in ␮mol 䡠
min⫺1 䡠 kg⫺1) were calculated as described by Wolfe et al. (41)
and Campbell et al. (7)
primary reesterification ⫽ (3 ⫻ Ra glycerol) ⫺ Ra FFA
Secondary reesterification rate was calculated as the difference between Ra FFA and total body lipid oxidation (41).
Hepatic VLDL-TG production rate was determined by fitting the decay curve of plasma TG-palmitate enrichment to a
single exponential equation: IEp ⫽ IEoe⫺kt, where IEp and
IEo are isotopic enrichments of plasma TG-palmitate at t and
t ⫽ 0, respectively, and k is the fractional turnover rate (12).
Secondary hepatic reesterification (in ␮mol 䡠 min⫺1 䡠 kg⫺1)
was calculated as k ⫻ (plasma TG pool) ⫻ 3. Plasma TG pool
was determined by multiplying plasma TG by plasma volume, estimated to be 55.6 ml/kg (32).
Secondary adipose reesterification was calculated by dividing the average label incorporated into triglycerides of the
three sampled adipose tissues by the steady state-specific
activity of plasma palmitate and by the ratio of palmitate to
total plasma FFA and then multiplying by total body fat
mass.
Statistics. All values are presented as means ⫾ SE. Statistical analysis was performed by one-way analysis of variance (ANOVA). When significant values were obtained, differences between individual means were analyzed by
pairwise multiple comparison analysis (Student-NewmanKeuls method).
RESULTS
Basal characteristics. The effect of TH and Medica 16
treatment on plasma concentrations of FFA, triacylglycerols, glucose, and insulin in rats in the fasting state
are presented in Table 1. Plasma FFA were not significantly affected by either treatment. Plasma triglycerides were decreased 2.4-fold in Medica 16-treated rats
and were only slightly but not significantly decreased
in hyperthyroid rats. Plasma glucose and insulin concentrations were slightly increased in both Medica 16and TH-treated rats.
Adipose tissue lipolysis and primary reesterification.
FFA production and release into plasma were determined by measuring the enrichment of labeled palmitate and glycerol in plasma under conditions of isotopic
steady state (Table 2). Glycerol production rates (Ra
glycerol), which reflect adipose tissue lipolysis, were
increased 1.5- and 1.4-fold by TH and Medica 16, respectively. These increases in lipolytic flux were accompanied by a 1.9-fold increase in primary (intra-adipose)
reesterification of FFA. Consequently, the increased
Table 2. Adipose tissue lipolysis and primary reesterification rates
Treatment
Ra Glycerol,
␮mol 䡠 min⫺1 䡠 kg⫺1
Ra FFA,
␮mol 䡠 min⫺1 䡠 kg⫺1
Ra FFA/Ra Glycerol
Lipolytic Flux (3Ra Glycerol),
␮mol 䡠 min⫺1 䡠 kg⫺1
Primary Reesterification,
␮mol 䡠 min⫺1 䡠 kg⫺1
Nontreated
TH
Medica 16
16.4 ⫾ 1.0
24.8 ⫾ 1.2*
23.2 ⫾ 1.1*
30.2 ⫾ 0.8
39.2 ⫾ 0.9*
33.5 ⫾ 2.7
1.84 ⫾ 0.10
1.58 ⫾ 0.06*
1.44 ⫾ 0.06*
49.2 ⫾ 3.0
74.4 ⫾ 3.6*
69.6 ⫾ 3.3*
19.0 ⫾ 2.4
35.2 ⫾ 3.2*
36.1 ⫾ 2.5*
Total body rates of appearance of glycerol (Ra glycerol) and FFA (Ra FFA) were measured in nontreated and TH- and Medica 16-treated
rats as described in METHODS. Primary reesterification was calculated (3Ra glycerol ⫺ Ra FFA) as described in METHODS. Values are means ⫾
SE (n ⫽ 4–6 rats). * Significantly different from respective nontreated animals (P ⬍ 0.05).
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.6 on June 18, 2017
4.6-mm Shendon ODS (5 ␮m) column eluted with 85% methanol in water at a flow of 2 ml/min. The specific activity of the
eluted palmitate was determined by relating the radioactivity to the respective concentration of palmitate as measured
using a palmitate standard. Incorporation of labeled palmitate into adipose TG was determined in adipose samples of
100 mg sonicated for 20 min in 5 ml of isopropanol-heptane-2
N H2SO4 at a ratio of 40:10:1. The heptane phase was
extracted with alkaline methanol and counted for radioactivity. The incorporation of [3H]palmitate into adipose TG was
linear for a period of 280 min.
Total body adipose mass was determined in nontreated
and Medica 16-treated rats as described previously (16) and
amounted to 11 ⫾ 1% of total body weight (mean ⫾ SE for 5
animals in each group).
Analytic procedures. Tissue triacylglycerols were determined in samples extracted with 20 vols of chloroform-methanol at a ratio of 2:1. The dried lipid extract was solubilized
in warm 0.4% SDS. Plasma and tissue triacylglycerols were
measured using a commercial kit (Boehringer Mannheim).
FFA composition of plasma was determined by GC analysis
of methyl ester derivatives. Plasma glucose was measured by
glucose test strips using an Elite glucometer (Bayer) based on
glucose oxidase. Plasma insulin was measured by a rat insulin RIA kit (Linco).
Calculations. Total body lipid oxidation was calculated, as
described previously (13), on the basis of O2 consumption and
CO2 production rates derived from indirect calorimetry and
was corrected for protein oxidation (0.13 mg/min) as deduced
from urinary nitrogen.
Rates of appearance of palmitate (Ra palmitate) and glycerol (Ra glycerol) (in ␮mol 䡠 min⫺1 䡠 kg⫺1) were calculated
using Steele’s equation for steady-state conditions (37) as
modified by Bier et al. (4) for stable isotopes
E223
E224
FATTY ACID CYCLING IN THE FASTING RAT
Table 3. FFA oxidation and recycling rates
Treatment
Nontreated
TH
Medica 16
Ra FFA,
FFA Oxidation,
Secondary Reesterification, Secondary Hepatic Reesterification, Secondary Adipose Reesterification,
␮mol 䡠 min⫺1 䡠 kg⫺1
␮mol 䡠 min⫺1 䡠 kg⫺1
␮mol 䡠 min⫺1 䡠 kg⫺1
␮mol 䡠 min⫺1 䡠 kg⫺1 ␮mol 䡠 min⫺1 䡠 kg⫺1
30.2 ⫾ 0.8
39.2 ⫾ 0.9*
33.5 ⫾ 2.7
17.9 ⫾ 0.8
30.6 ⫾ 1.4*
22.7 ⫾ 0.6*
12.3 ⫾ 0.9
8.6 ⫾ 1.5*
10.8 ⫾ 0.5
1.6 ⫾ 0.2
1.1 ⫾ 0.2*
0.5 ⫾ 0.2*
9.0 ⫾ 0.9
ND†
6.5 ⫾ 0.3*
Ra FFA, total body FFA oxidation, total body reesterification of plasma FFA (secondary reesterification), very low density lipoprotein-TG
production rate (secondary hepatic reesterification), and rate of reesterification of plasma FFA into adipose fat (secondary adipose
reesterification) were determined in nontreated and TH- and Medica 16-treated rats as described in METHODS. Values are means ⫾ SE (n ⫽
3–6 rats). ND, not determined. * Significantly different from respective nontreated animals (P ⬍ 0.05). † FFA incorporation into adipose
tissue TG of hyperthyroid animals amounted to 0.145 ⫾ 0.020 ␮mol 䡠 min⫺1 䡠 g adipose tissue⫺1 (compared with 0.082 ⫾ 0.008 and 0.059 ⫾
0.03 for nontreated and Medica 16-treated animals, respectively). Assuming total body adipose mass for hyperthyroid animals to be 6.8% of
body weight (based on Ref. 31) would result in secondary adipose reesterification of 9.8 ⫾ 1.4 ␮mol 䡠 min⫺1 䡠 kg⫺1.
state amounted to 70 ⫾ 5.5% of plasma free palmitate
(Fig. 1), indicating that only ⬃70% of liver fatty acids
incorporated into VLDL-TG were derived in nontreated animals from plasma FFA, whereas 30% were
derived from an unlabeled hepatic pool of fatty acids.
This nonplasmatic liver pool was significantly decreased in animals treated with TH or Medica 16 and
in which plasma TG-palmitate enrichment during isotopic steady state amounted to 82 ⫾ 2.9 and 83 ⫾ 1.8%
of plasma free palmitate enrichment, respectively. Hepatic VLDL-TG production rates as calculated from the
respective decay of the enrichment curves of the three
groups are shown in Table 3. VLDL-TG production
accounted for 13% of total body secondary reesterification in nontreated animals, and because only 70% of
VLDL-TG were synthesized from plasma FFA, hepatic
reesterification accounted for only 9% of secondary
Fig. 1. Plasma free fatty acid (FFA) and triglyceride (TG) enrichment. Plasma free palmitate (■) and TG palmitate (●) enrichment
were determined during a period of 145 min of labeled palmitate
infusion followed by 80 min of saline infusion. Interruption of
[2H2]palmitate infusion is marked by arrow. Enrichment is expressed in mole percent excess (MPE).
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.6 on June 18, 2017
lipolytic flux resulted in compromised FFA release into
plasma (Ra FFA) with a concomitant decrease in the
ratio of Ra FFA to Ra glycerol. Primary reesterification
accounted for 39, 47, and 52% of total lipolytic flux in
the fasting state in nontreated and TH- and Medica
16-treated rats, respectively.
Fatty acid oxidation. Total body O2 consumption
increased by 55% in TH-treated animals compared
with age-matched controls (34.2 ⫾ 0.9 vs. 22.1 ⫾ 0.5 ml
O2 䡠 min⫺1 䡠 kg⫺1, P ⬍ 0.05). Because total body O2
consumption decreased with age (when normalized to
body weight), and in light of the relatively long period
of treatment with Medica 16 compared with the short
treatment period with TH, O2 consumption of Medica
16-treated animals was measured twice, following 1
and 6 wk of treatment. O2 consumption was significantly increased by 15% in Medica 16-treated rats
following 1 wk of treatment (25.4 ⫾ 0.6 vs. 22.1 ⫾ 0.5
ml O2 䡠 min⫺1 䡠 kg⫺1, P ⬍ 0.05) and persisted through
the 6 wk of the treatment period (19.7 ⫾ 0.3 vs. 17.2 ⫾
0.2 ml O2 䡠 min⫺1 䡠 kg⫺1, P ⬍ 0.05).
Total body lipid oxidation rates measured by indirect
calorimetry several times during the isotope infusion
period were increased 1.7- and 1.3-fold in TH- and Medica 16-treated rats, respectively. The fraction of oxidized
FFA out of FFA released from adipose tissue (Ra FFA)
amounted to 59, 78, and 68% in nontreated, TH-, and
Medica 16-treated rats, respectively (Table 3).
Secondary reesterification. Secondary reesterification rates were evaluated by subtracting total body
FFA oxidation rates from Ra FFA. As shown in Table 3,
secondary reesterification was significantly decreased
in hyperthyroid rats in light of the substantial increase
in lipid oxidation.
Hepatic VLDL-TG production was evaluated by analyzing the decay curve of plasma TG-palmitate enrichment when labeled palmitate infusion was interrupted. As shown in Fig. 1, plasma TG-palmitate
enrichment reached isotopic steady state after 120–
140 min of labeled palmitate infusion compared with
50–60 min for plasma free palmitate. A slow turnover
rate for plasma TG-palmitate compared with that for
free palmitate was also reflected by the slow decay of
plasma TG-palmitate compared with that of free
palmitate (Fig. 1). Also, plasma TG-palmitate enrichment in nontreated animals during isotopic steady
FATTY ACID CYCLING IN THE FASTING RAT
Fig. 2. Incorporation of plasma FFA into adipose TG. Incorporation
of plasma FFA into adipose TG was measured during isotopic steady
state of plasma [3H]palmitate as described in METHODS. Each data
point represents the value for a single rat.
DISCUSSION
Fatty acid metabolism in fasted rats: an integrative
view. Long-chain fatty acid metabolism in the fasting
state is initiated by HSL-catalyzed lipolysis of adipose
fat, resulting in plasma FFA as a source of metabolic
fuel for the heart, skeletal muscle, and liver. Total body
FFA oxidation in the nontreated fasted rat amounts to
only 36% of total body lipolytic flux, indicating that
adipose tissue lipolysis is activated much beyond the
requirement for total body fat oxidation. Moreover,
plasma FFA channeled into VLDL-TG production is
only 5% of Ra FFA, indicating that the oxidizable fatty
acid pool in the postabsorptive state consists mainly of
plasma FFA released from adipose fat and directly
oxidized by respective tissues, rather than fatty acids
derived from intravascular lipolysis of VLDL-TG. The
higher production rates of VLDL-TG previously reported by Wolfe and Durkot (40) could perhaps result
from the use of labeled dog VLDL instead of rat VLDL
or from plasma VLDL enrichments lower than isotopic
steady-state levels.
Most FFA resulting from fat lipolysis in adipose
tissue in the postabsorptive state are reesterified back
into TG. Because hepatic reesterification of FFA into
VLDL-TG accounts for only a minor fraction of total
body FFA flux, most of the reesterified FFA would have
to be accounted for by extrahepatic tissues. Indeed,
39% of total body lipolytic FFA was found here to be
primarily reesterified into adipose fat even before its
release into plasma, in line with a previous report (40)
pointing to a ratio of 1:1 for Ra FFA/Ra glycerol in the
postabsorptive rat. Another 18% of total body lipolytic
FFA was found here to be secondarily reesterified in
adipose tissue, making this tissue the most important
site of reesterification of recycled FFA in general and of
secondary reesterification in particular. Hence, primary and secondary reesterification of lipolytic FFA in
adipose tissue accounts for almost all FFA produced by
HSL and that escape oxidation in oxidizing tissues.
Total body fatty acid metabolism and steady-state levels of plasma FFA appear, therefore, to be controlled by
adipose tissue, as both a source and sink.
Similarities and differences between the FFA cycle
reported here in fasted rats and previously in fasted
humans are worth noting. The overall FFA flux in rats
(49.2 ⫾ 3.0 ␮mol 䡠 min⫺1 䡠 kg⫺1) is ⬃5- to 15-fold higher
than in humans (3–9 ␮mol 䡠 min⫺1 䡠 kg⫺1) (12, 17, 33).
However, primary reesterification in fasted humans is
negligible (3, 9, 12, 35; see, however, Refs. 7, 17, 41,
and 42), whereas that in fasted rats accounts for 40–
50% of the total flux (Table 2; Ref. 39). Because human
primary reesterification combined with VLDL-TG production rate may account for a fraction of only 5% of
total body FFA turnover in the fasting state (12), and
because lipid oxidation accounts for 40% of FFA turnover under these conditions, extrahepatic secondary
reesterification must account for at least 50% of human
total body FFA turnover. The site(s) for secondary
reesterification of FFA in humans remained unresolved (7, 12). However, in light of the general similar-
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.6 on June 18, 2017
reesterification or only 3% of total body lipolytic flux.
Moreover, Medica 16 and TH treatments resulted in a
3- and 1.5-fold decrease in VLDL-TG production rate,
respectively, indicating that hepatic reesterification of
FFA under calorigenic conditions accounted for an
even lower fraction of total body secondary reesterification. The calorigenesis-induced decrease in hepatic
VLDL-TG production was not compromised by changes
in liver TG content (data not shown).
The contribution made by adipose tissue to total
body secondary reesterification was evaluated by the
recovery of labeled palmitate in adipose TG under
conditions of isotopic steady state in plasma free palmitate accompanied by linear accumulation of plasma
FFA in adipose tissue (Fig. 2). Palmitate incorporation
into adipose TG was calculated as the average of adipose tissue labeling in samples taken from the epididymal, perirenal, and subcutaneous fat. No significant
differences in palmitate incorporation were observed
between different fat pads (data not shown). As shown
in Table 3, FFA reesterification in adipose tissue accounted for 60–100% of total body secondary reesterification, indicating that adipose tissue is the primary
site for secondary reesterification in rats. Secondary
reesterification of FFA in adipose tissue essentially
accounted for plasma FFA that escaped oxidation or
hepatic reesterification into VLDL-TG under euthyroid
or induced calorigenesis. No significant differences
were observed in muscle TG content among the various
conditions (data not shown).
E225
E226
FATTY ACID CYCLING IN THE FASTING RAT
pose tissue, in line with that recently reported in obese
fa/fa rats, under either basal or hyperinsulinemic-euglycemic clamp conditions (28). Secondary reesterification in
liver was, however, significantly inhibited by Medica 16
treatment, leaving most secondary reesterification to adipose tissue. Hence, under conditions of increased total
body FFA turnover induced by Medica 16 treatment,
FFA reesterification (primary ⫹ secondary) in adipose
tissue again accounted for 61% of lipolytic flux in the
fasting state.
Medica 16 treatment resulted in a 15% increase in
total body O2 consumption with a concomitant 1.3-fold
increase in lipid oxidation. Calorigenesis induced by
Medica 16 was in line with that in a previous study from
this laboratory reporting uncoupling of mitochondrial
oxidative phosphorylation by Medica 16 (19). The lower
increase in total body O2 consumption induced by Medica
16 compared with that induced by TH reflects the liverspecific thyromimetic activity of Medica 16 compared
with the nonselective calorigenic activity of TH (25).
The pronounced inhibition of hepatic VLDL-TG production by Medica 16 may account for the hypolipidemic
effect of Medica 16 in the fasting state. This hypolipidemic effect complements the previously reported hypolipidemic activity of Medica 16 in the absorptive state due
to activated clearance of plasma VLDL and chylomicrons
as a result of transcriptional suppression of the hepatocyte nuclear factor-4 (HNF-4)-controlled apolipoprotein
CIII (14, 15, 22). The postabsorptive hypolipidemic effect
of Medica 16 could perhaps reflect transcriptional suppression of other HNF-4-controlled genes involved in
liver VLDL assembly and, in particular, of apolipoprotein
B and microsomal triglyceride transfer protein. Liver
VLDL-TG production could perhaps be further limited by
the limited availability of liver FFA for reesterification
under conditions of pronounced hepatic calorigenesis induced by Medica 16 treatment, in line with recent reports
(27, 36) pointing to the reciprocal interplay between fatty
acid oxidation and its hepatic reesterification into
VLDL-TG.
In summary, adipose tissue serves as a main source
as well as a sink for fatty acid metabolism in the fasted
rat. The fatty acid flux initiated in rat adipose tissue is
mostly futile and results in reesterification of 57% of
lipolyzed FFA back into adipose TG. Two-thirds of FFA
reesterification into adipose fat are carried out before
FFA release into plasma, whereas the rest results from
extraction of plasma FFA. Thirty-six percent of total
lipolytic FFA flux is accounted for by net FFA oxidation, whereas only a minor fraction of FFA flux is
channeled into hepatic VLDL-TG. Fatty acid cycling in
excess of oxidative demands allows for rapid response
to sudden changes in energy requirements (41). Moreover, the futile nature of the cycle allows for its fine
tuning in response to even mild changes in metabolic
demands (30). This high responsiveness of the fatty
acid cycle is preserved under conditions of calorigenesis induced by thyroid hormones or Medica 16 due to
an increase in lipolytic flux, maintaining essentially
constant the oxidative flux-to-cycling flux ratio.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.6 on June 18, 2017
ities between FFA cycling in human subjects and rats,
secondary reesterification is proposed to be essentially
carried out by adipose tissue in both species. The implied extensive flux of adipose secondary reesterification of FFA in humans under conditions in which
primary reesterification is almost nil may indicate that
the primary and secondary reesterification pathways
must be limited by distinct steps. Although epididymal, subcutaneous, and perirenal rat adipose tissues
were found here to account for similar secondary reesterification rates (based on tissue weight), the specificity of FFA uptake by individual human adipose tissues
still remains to be investigated. Lack of FFA uptake by
human abdominal subcutaneous adipose tissue in the
fasting state as recently claimed by Coppack et al. (9)
may indeed indicate that human individual adipose
tissues may differ in their FFA extraction capacities.
The fatty acid cycle in TH-treated fasted rats. TH was
found here to activate lipolysis of adipose fat 1.5-fold,
in line with 3-fold activation previously reported in
humans (3). This increase in lipolytic rate may be
ascribed to adipose tissue sensitization to catecholamines as a result of the increased expression of
␤-adrenoreceptors (5, 18). The increase in lipolytic rate
by TH was, however, compromised by a dramatic increase in primary reesterification. The increase in lipolytic flux induced by TH was accompanied by a 55%
increase in total body O2 consumption and a 1.7-fold
increase in lipid oxidation, in line with a similar increase previously reported by Oppenheimer et al. (31).
The increase in oxidation may be ascribed to mitochondrial uncoupling due to transition from a low to a high
mitochondrial conductance (23). TH-induced oxidation
of FFA limits the extent of secondary reesterification of
plasma FFA. The decrease in hepatic VLDL-TG production could perhaps reflect the lower availability of
hepatic fatty acids for reesterification under conditions
of TH-induced calorigenesis and may account for the
somewhat lower plasma TG observed in postabsorptive
TH-treated rats. Secondary reesterification in adipose
tissue completes the FFA cycle and again accounts for
most recycled plasma FFA. In summary, the calorigenic effect of TH is dominated by high turnover of the
FFA cycle with a concomitant absolute and relative
increase in primary reesterification and FFA oxidation.
The thyromimetic-calorigenic effect of Medica 16.
Similarly to TH, Medica 16 was found here to increase
adipose fat lipolysis by 41%, in line with activation of
lipolysis by Medica 16 in obese, insulin-resistant fa/fa
rats (28). Activation of the lipolytic flux under conditions
of sensitization to insulin (28) may imply that Medica
16-induced lipolysis of adipose fat is transduced by an
insulin-independent pathway. Activation of adipose lipolysis by Medica 16 is indeed accompanied by increased
sensitization of adipocytes to variable lipolytic activators
(e.g., catecholamines, forskolin, ACTH) (38). The 41%
increase in lipolytic flux induced by Medica 16 is only
partially compromised by the 27% increase in lipid oxidation, leaving a substantial fraction to primary and
secondary reesterification. Similarly to TH, Medica 16
indeed increased primary reesterification of FFA in adi-
FATTY ACID CYCLING IN THE FASTING RAT
This work was supported by the Israeli Ministry of Science and
Technology.
REFERENCES
21. Hertz R, Ben-Ishai A, Nicodem V, Berman I, and Bar-Tana
J. Thyromimetic mode of action of peroxisome proliferators:
transcriptional activation of liver malic enzyme. Biochem J 319:
241–248, 1996.
22. Hertz R, Bishara-Shieban J, and Bar-Tana J. Mode of action
of peroxisome proliferators as hypolipidemic drugs: suppression
of apolipoprotein C-III. J Biol Chem 270: 13470–13475, 1995.
23. Kalderon B, Hermesh O, and Bar-Tana J. Mitochondrial
permeability transition is induced by in vivo thyroid hormone
treatment. Endocrinology 136: 3552–3556, 1995.
24. Kalderon B, Hertz R, and Bar-Tana J. Effect of thyroid
hormone treatment on redox and phosphate potentials in rat
liver. Endocrinology 131: 400–407, 1992.
25. Kalderon B, Hertz R, and Bar-Tana J. Tissue selective modulation of redox and phosphate potentials by ␤,␤⬘-methyl-substituted hexadecanedioic acid. Endocrinology 131: 1629–1635,
1992.
26. Krähenbühl L and Bürgi U. Evidence that the effect of triiodothyronine on epinephrine-stimulated lipolysis in white adipose tissue is mediated via nuclear receptors. Horm Metab Res
Suppl 17: 20–22, 1987.
27. Lewis GF. Fatty acid regulation of very low density lipoprotein
production. Curr Opin Lipidol 8:146–153, 1997.
28. Mayorek N, Kalderon B, Itah E, and Bar-Tana J. Sensitization to insulin induced by ␤,␤⬘-methyl-substituted hexadecanedioic acid (Medica 16) in obese Zucker rats in vivo. Diabetes
46: 1958–1964, 1997.
29. Miles JM, Ellman MG, McClean KL, and Jensen MD. Validation of new methods for determination of free fatty acid
turnover. Am J Physiol Endocrinol Metab 252: E431–E438,
1987.
30. Newsholme EA and Crabtree B. Substrate cycles in metabolic
regulation and in heat generation. Biochem Soc Symp 41: 61–
109, 1976.
31. Oppenheimer JH, Schwartz HL, Lane JT, and Thompson
MP. Functional relationship of thyroid hormone-induced lipogenesis, lipolysis, and thermogenesis in rats. J Clin Invest 87:
125–132, 1991.
32. Petty C. Research Techniques in the Rat. Springfield, IL:
Thomas, 1982, p. 66–70.
33. Raguso CA, Coggan AR, Gastaldelli A, Sidossis LS, Bastyr
EJ, and Wolfe RR. Lipid and carbohydrate metabolism in
IDDM during moderate and intense exercise. Diabetes 44: 1066–
1074, 1995.
34. Saffari B, Ong JM, and Kern PA. Regulation of adipose tissue
lipoprotein lipase gene expression by thyroid hormone. J Lipid
Res 33: 241–249, 1992.
35. Samra JS, Clark ML, Humphreys SM, Macdonald IA, and
Frayn KN. Regulation of lipid metabolism in adipose tissue
during early starvation. Am J Physiol Endocrinol Metab 271:
E541–E546, 1996.
36. Sidossis LS, Mittendorfer B, Walser E, Chinken D, and
Wolfe RR. Hyperglycemia-induced inhibition of splanchnic fatty
acid oxidation increase hepatic triacylglycerol secretion. Am J
Physiol Endocrinol Metab 275: E798–E805, 1998.
37. Steele R. Influences of glucose loading and of injected insulin on
hepatic glucose output. Ann NY Acad Sci 82: 420–430, 1959.
38. Tzur R, Smith E, and Bar-Tana J. Adipose reduction by
␤,␤⬘-tetramethyl-substituted hexadecanedioic acid (Medica 16).
Int J Obes 13: 313–326, 1989.
39. Wilcox HG and Heimberg M. Effect of hyperthyroidism on
synthesis, secretion and metabolism of VLDL apoproteins by the
perfused rat liver. Biochim Biophys Acta 1081: 246–252, 1991.
40. Wolfe RR and Durkot MJ. Role of very low density lipoproteins in the energy metabolism of the rat. J Lipid Res 26:
210–217, 1985.
41. Wolfe RR, Klein S, Carraro F, and Weber JM. Role of
triglyceride-fatty acid cycle in controlling fat metabolism in
humans during and after exercise. Am J Physiol Endocrinol
Metab 258: E382–E389, 1990.
42. Wolfe RR, Peters EJ, Klein S, Holland OB, Rosenblatt J,
and Gary H. Effect of short term fasting on lipolytic responsiveness in normal and obese human subjects. Am J Physiol Endocrinol Metab 252: E189–E196, 1987.
Downloaded from http://ajpendo.physiology.org/ by 10.220.33.6 on June 18, 2017
1. Arner P, Wennlund A, and Östman J. Regulation of lipolysis
by human adipose tissue in hyperthyroidism. J Clin Endocrinol
Metab 48: 415–419, 1979.
2. Bar-Tana J, Kahn-Rose G, and Srebnik B. Inhibition of lipid
synthesis by ␤,␤⬘-tetramethyl-substituted C14–C22 ␣,␻ dicarboxylic acids in the rat in vivo. J Biol Chem 260: 8404–8410,
1985.
3. Beylot M, Martin C, Laville M, Riou JP, Cohen R, and
Mornex R. Lipolytic and ketogenic fluxes in human hyperthyroidism. J Clin Endocrinol Metab 73: 42–49, 1991.
4. Bier DM, Arnold KJ, Sherman WR, Holland WH, Holmes
WF, and Kipnis DM. In vivo measurement of glucose and
alanine metabolism with stable isotopic tracers. Diabetes 26:
1005–1015, 1977.
5. Bilezikian JP and Loeb JN. The influence of hyperthyroidism
and hypothyroidism on ␣-and ␤-adrenergic receptor systems and
adrenergic responsiveness. Endocr Rev 4: 378–388, 1983.
6. Bougneres PF and Bier DM. Stable isotope dilution method
for measurement of palmitate content and labeled palmitate
tracer enrichment in microliter plasma samples. J Lipid Res 23:
502–507, 1982.
7. Campbell PJ, Carlson MG, Hill JO, and Nurjhan N. Regulation of free fatty acid metabolism by insulin in humans: role of
lipolysis and reesterification. Am J Physiol Endocrinol Metab
263: E1063–E1069, 1992.
8. Clouette R, Jacob M, Koteel P, and Spain M. Confirmation of
11-nor9⌬-tetrahydrocannabinol in urine as its t-butyldimethylsilyl derivative using GC/MS. J Anal Toxicol 17: 1–4, 1993.
9. Coppack SW, Persson M, Judd RL, and Miles JM. Glycerol
and nonesterified fatty acid metabolism in human muscle and
adipose tissue in vivo. Am J Physiol Endocrinol Metab 276:
E233–E240, 1999.
10. Cruz ML and Williamson DH. Effect of triiodothyronine administration on dietary [14C]triolein partitioning between deposition in adipose tissue and oxidation to [14C]CO2 in ad libitumfed or food restricted rats. Biochim Biophys Acta 1168: 205–212,
1993.
11. Davidson NO, Carlos RC, and Lukaszewicz AM. Apolipoprotein B mRNA editing is modulated by thyroid hormone analogs
but not growth hormone administration in the rat. Mol Endocrinol 4: 779–785, 1990.
12. Diraison F and Beylot M. Role of human liver lipogenesis and
reesterification in triglycerides secretion and in FFA reesterification. Am J Physiol Endocrinol Metab 274: E321–E327, 1998.
13. Frayn KN. Calculation of substrate oxidation rates in vivo from
gaseous exchange. J Appl Physiol 55: 628–634, 1983.
14. Frenkel B, Bishara-Shieban J, and Bar-Tana J. The effect of
␤,␤⬘-methyl-substituted hexadecanedioic acid on plasma VLDL
metabolism in rats: role of apolipoprotein C-III. Biochem J 298:
409–414, 1994.
15. Frenkel B, Mayorek N, Hertz R, and Bar-Tana J. The
hypochylo-micronemic effect of ␤,␤⬘-methyl-substituted hexadecanedioic acid is mediated by a decrease in apolipoprotein C-III.
J Biol Chem 263: 8491–8497, 1988.
16. Hartsook EW and Hershberger TV. A simplified method for
sampling small animal carcasses for analysis. Proc Soc Exp Biol
Med 113: 973–977, 1963.
17. Hellerstein MK, Benowitz NL, Neese RA, Schwartz JM,
Hoh R, Jacob P, Hsieh J, and Faix D. Effect of cigarette
smoking and its cessation on lipid metabolism and energy expenditure in heavy smokers. J Clin Invest 93: 265–272, 1994.
18. Hellsström L, Wahrenberg H, Reynisdottir S, and Arner P.
Catecholamine-induced adipocyte lipolysis in human hyperthyroidism. J Clin Endocrinol Metab 82: 159–166, 1997.
19. Hermesh O, Kalderon B, and Bar-Tana J. Mitochondria
uncoupling by a long chain fatty acid analogue. J Biol Chem 273:
4237–4244, 1998.
20. Hertz R, Aurbach R, Hashimoto T, and Bar-Tana J. Thyromimetic effect of peroxisomal proliferators in rat liver. Biochem
J 274: 745–751, 1991.
E227