| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 280, Issue 27, 25396-25402, July 8, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
¶
From the
Department of Nutritional Sciences &
Toxicology, University of California-Berkeley, Berkeley, California 94720 and
the Department of Medicine, San Francisco
General Hospital, University of California-San Francisco, San Francisco,
California 94110
Received for publication, December 13, 2004 , and in revised form, May 11, 2005.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
-glycerol phosphate, which can have glycolytic and
glyceroneogenic origins. We present a technique for determining the relative
glyceroneogenic contribution to triacylglyceride glycerol by labeling the
glycerol moiety with 2H2O. The number of hydrogen atoms
(n) incorporated from H2O into C–H bonds reflects
the metabolic source of -glycerol phosphate and can be calculated by
combinatorial analysis of the distribution of mass isotopomers in
triacylglyceride glycerol. Three physiological settings with potential effects
on glyceroneogenesis and glycolysis were studied in rodents. Adipose tissue
acylglyceride glycerol in mice fed a low carbohydrate diet had significantly
higher values of n than in mice fed a high carbohydrate diet,
suggesting an increased contribution from glyceroneogenesis of from 17 to 50%
on the low carbohydrate diet. Similarly, mice administered rosiglitazone had a
significant relative increase in glyceroneogenesis (from 17 to 53%), indicated
by an increase in adipose acylglyceride glycerol n. Fructose infusion
in overnight fasted rats rapidly lowered plasma triacylglyceride glycerol
n, reflecting a decreased contribution from glyceroneogenesis (from
66 to 34%) presumably because of increased glycolytic input. In conclusion, we
demonstrate that the number of C–H atoms derived from cellular
H2O in triacylglyceride glycerol is an informative indicator of
-glycerol phosphate origin and, ultimately, triacylglycerol metabolism.
Under certain physiological conditions, glyceroneogenesis can be up-regulated
in adipose (e.g. low carbohydrate diet) or down-regulated in liver
(e.g. fructose infusion). Additionally, stimulation of
glyceroneogenesis by rosiglitazone in adipose tissue may be an important
factor in the antilipolytic actions of thiazolidinediones. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
-glycerol phosphate (-GP), the
direct intracellular precursor for TG synthesis. There are three primary
sources of -GP for TG synthesis
(Fig. 1): plasma glycerol, via
direct phosphorylation by glycerol kinase (traditionally thought not to be
present to a significant extent in adipose tissue); glucose, via glycolysis to
the level of triose phosphate; and glyceroneogenesis, which is defined as the
synthesis of -GP from gluconeogenic precursors (i.e.
precursors other than glucose or glycerol)
(2). A better understanding of
the relative contributions of these pathways to TG synthesis under different
conditions could aid our understanding of TG metabolism.
Recently, methods have been presented for the in vivo measurement
of TG synthesis using heavy water (2H2O)
(3,
4). We present here the
application of 2H2O labeling combined with mass
isotopomer distribution analysis (MIDA), to the discernment of sources of
-GP. MIDA involves quantifying, by mass spectrometry, the relative
abundances of molecular species of a polymer differing only in mass (mass
isotopomers), after introduction of a stable isotope-labeled precursor. The
degree and pattern of incorporation of 2H into C–H bonds of
-GP is dependent on two factors: the 2H enrichment of body
water and the number of C–H bonds in -GP that were derived from
body water. The latter is dependent on the biochemistry of hydrogen
incorporation from solvent H2O into glyceroneogenic and glycolytic
intermediates (Figs. 1 and
2)
(5). The maximal theoretical
number of hydrogen atoms (n) that can be incorporated from
H2O into C–H bonds is 5 (italicized in
Fig. 1). Any -GP
generated via glyceroneogenesis (i.e. through the tricarboxylic acid
cycle and "bottom" portion of the gluconeogenesis pathway, from
pyruvate or oxaloacetate) will have all five hydrogens incorporated from body
water (Figs. 1 and
2A,
italicized). The -GP generated via glycolysis from unlabeled
glucose, in contrast, has an n of at most 3.5 (i.e. at most
three and a half hydrogen atoms are incorporated from body water (Figs.
1 and
2B, bold)).
This non-integer value derives from the phosphoglucose isomerase step of
glycolysis, which adds label to C1
(Fig. 2B)
(6). This results in a labeled
hydrogen on the phosphate carbon of dihydroxyacetate phosphate but not
glyceraldehyde 3-phosphate. Of the two -GPs generated from one glucose,
one will therefore have an n of 3 and one will have an n of
4 (average = 3.5). -GP generated in the liver from unlabeled glycerol
via glycerol kinase could theoretically have an n of 0
(Fig. 1). However, it has been
shown previously that there is rapid exchange of -GP with glycolytic
intermediates (e.g. in the triose phosphate pool) in liver
(7,
8) resulting in an n
closer to 3.
|
4.0 and from liver TG 4.7
(4). These estimates were based
upon both plateau label incorporation after long term
2H2O administration and combinatorial labeling patterns
using MIDA. We hypothesized that experimentally observed values of n
less than 5 could be explained by contributions from non-glyceroneogenic
sources of -GP (Fig. 1).
By this interpretation, the resulting experimentally determined n
reflects the mixture of TG-glycerol from glycolysis (glyc3.5),
glyceroneogenesis (glyc5), and glycerol kinase (glyc3).
The different values of n observed in different tissues, with a
higher n observed in liver compared with adipose tissue, were
consistent with the known higher activity of glyceroneogenesis in liver
(8,
9). If this is, in fact, the
physiological explanation for a measured value of n less than 5, we
reasoned that n in TG-glycerol might be informative in its own right
as a parameter of cellular metabolism and glyceroneogenesis.
The goal of the present study was to test this hypothesis. Three
experimental conditions were imposed. First, the effect of a high fat, low
carbohydrate (LC) diet on adipose TG n was studied in mice. Reduced
dietary glucose input in animals fed a LC diet should result in a lesser
contribution to TG-glycerol from uptake and glycolytic metabolism of dietary
glucose (glyc3.5) and a greater contribution from the
glyceroneogenic pathway (glyc5), resulting in a higher measured
n. Second, we investigated the effect of fructose infusion on hepatic
TG n, sampled from plasma TG-glycerol in rats. Fructose floods the
hepatic triose phosphate pool from the glycolytic direction (glyc3)
and might result in a lowering of plasma TG-glycerol n. Finally,
having established the informative value of TG-glycerol n, we
evaluated the effect on glyceroneogenesis of rosiglitazone, a PPAR
agonist known to increase adipose TG deposition and alter adipose tissue
glycerol metabolism (10). The
relative importance of rosiglitazone induced increases in PEPCK
versus glycerol kinase expression and activity in adipose tissue has
been controversial (11). If an
increase in PEPCK was the predominant mechanism involved in
rosiglitazone-induced -GP synthesis (and hence TG synthesis), the
measured n should increase with increased contribution from
glyc5 (i.e. glyceroneogenesis). In contrast, a significant
increase in glycerol kinase activity would result in a decrease in measured
n (glyc3 or less). We show here that measurement of
n in acylglyceride glycerol can be informative as an index of
glyceroneogenesis, and, ultimately, TG metabolism under different
conditions.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
Mice—4-week-old male C57Bl/6J mice (16–18 g; Jackson
Laboratories, Bar Harbor, ME) were used. Mice were fed ad libitum a
high carbohydrate, low fat (HC, 70% carbohydrate, 10% fat) diet or a low
carbohydrate, high fat diet (LC, 35% carbohydrate, 45% fat) (Research Diets
Inc., New Brunswick, NJ). An additional group of mice were fed an HC diet
containing rosiglitazone (6.34 mg/kcal diet), which resulted in a dose of
3 mcg/kg mouse/day (Research Diets Inc., New Brunswick, NJ). After 11
days on the diet, all animals were given a priming dose of 99.8%
2H2O in saline via intraperitoneal injection to achieve
4.8% 2H2O enrichment in body water (30 µl/g
mouse) followed by administration of 8% 2H2O in drinking
water. Mice (n = 6 per group) were sacrificed after either 15 or 64
days on heavy water. Mesenteric, epididymal, retroperitoneal, and inguinal
adipose tissue depots were removed, and blood and urine samples were
obtained.
In an additional experiment to investigate potential changes in deuterium enrichment in glucose, mice, as above, were put on the LC diet, HC diet, or HC diet with rosiglitazone for 2 weeks with 2H2O administered as above during the final week (n = 10 per group). Half the animals in each group were sacrificed in the morning (7:30–9:30 AM) and half at the end of the day (4:30–6:15 PM), roughly corresponding to the start of the light and dark cycles (7 AM–7 PM). Animals were ad libitum fed throughout. Blood samples were obtained at the time of sacrifice.
Rats—Sprague-Dawley rats (400–500 g, Charles River Laboratories, Wilmington, MA) were purchased with indwelling jugular vein and carotid artery catheters in place. Rats (n = 4) were fed ad libitum a Purina chow diet. After an overnight fast, animals were anesthetized with isoflurane and a priming dose of 99.8% [2H2O]saline was given via intraperitoneal injection, as above. After a minimum of 60 min, to allow for 2H2O equilibration with body water, two baseline blood samples were collected (times –30 and –15 min). At time 0, fructose was infused intravenously (16–19 mg/kg/min for 4 h). Blood samples were taken after 3 and 4 h of fructose administration.
Isolation of Acylglyceride Glycerol from Adipose Tissue—Lipids from adipose tissue were extracted by a modified Folch extraction (12). The lipid fraction was transesterified by incubation with 3 N methanolic HCl (Sigma-Aldrich) at 55 °C for 60 min. Fatty acid methyl esters were separated from glycerol by the Folch technique. The aqueous phase containing glycerol was lyophilized, and glycerol was converted to glycerol triacetate by incubation with acetic anhydride-pyridine (2:1) as described elsewhere (13).
Isolation of TG-glycerol from Plasma—Plasma, obtained from fresh whole blood, was extracted by the Folch technique. TG was isolated by TLC as described previously (14). Glycerol isolation and derivatization were then performed as described above.
Isolation of Glucose from Plasma—Plasma glucose was isolated using the method of Van Dijk et al. (15). Briefly glucose was extracted from plasma spotted on filter paper using an ethanol/water extraction. Glucose was then derivatized to the aldonitrile tetra-acetate derivative as described previously (13).
Measurements of 2H2O Enrichment in Body Water—2H2O enrichment in body water (from plasma or urine) was measured by one of two methods. Briefly, 15–20 µl of plasma or urine were reacted in an evacuated GC vial with calcium carbide to produce acetylene. The acetylene gas was then removed with a syringe and injected into a GC vial containing 10% bromine in carbon tetrachloride and incubated at room temperature for 2 h to produce tetrabromoethane. Excess bromine was neutralized with 25 µl of 10% cyclohexene, and the sample was suspended in ethyl acetate (16). Alternatively, the acetylene gas was directly measured by a new mass spectrometric method (17). Briefly, 25 µl of sample were injected into a closed Exitainer vial containing calcium carbide in a dry helium atmosphere. A small amount (0.5 ml) of the acetylene gas generated from the reaction was removed and injected into another closed vial with a helium atmosphere for direct analysis. The two methods, used with standard curves, give identical results.2 However, the direct acetylene method is less time consuming and, thus, became the preferred method during the course of this study.
|
GC-MS Analyses—Glycerol triacetate and glucose aldonitrile were analyzed for isotope enrichment by GC-MS as described previously (13). Mass isotopomer abundances were analyzed by selected ion monitoring of mass-to-charge ratios (m/z) 159–161 (M0-M2) for glycerol and 328–331 (M0-M3) for glucose. Tetrabromoethane was analyzed for isotope enrichment by GC-MS method as described previously (18). The isotopic enrichment of acetylene (m/z 26 and 27) was measured by cycloidal mass spectrometry (Monitor Instruments, Pittsburgh, PA).
Calculations
Body 2H2O Enrichment—Isotopic
enrichment of body 2H2O was calculated by comparison to
a standard curve prepared gravimetrically from natural abundance water and
2H2O.
[2H]Glycerol Enrichment—Isotope enrichments of [2H]glycerol derived from acylglycerides were calculated by subtraction of mass isotopomer abundances in unlabeled glycerol standards (19). EM1 and EM2 were calculated as a fraction of the sum of mass isotopomers M0-M2, as previously described for MIDA calculations (4).
[2H]Glucose Enrichment—Glucose labeled with
2H via gluconeogenesis could result in a glycerol n of
greater than the theoretical glycolytic value of 3.5, thereby resulting in an
overestimation of the contribution of glyceroneogenesis to adipose TG.
Therefore we measured 2H glucose enrichments at two different times
of the day for each dietary group. 2H label was calculated using a
modification of Lee et al.
(20) as shown in
Equation 1,
 | (Eq. 1) |
MIDA Calculations of n and
A
1—MIDA is a technique based
on combinatorial analysis of the labeling patterns present in polymers
(4,
19,
20). The
EM2 to EM1 ratio (R) is one
embodiment of this labeling pattern. R is dependent on two factors:
the proportion (p) of labeled hydrogen atoms present in tissue water
(i.e. the enrichment of 2H2O in tissue water)
and the possible number of C–H bonds in glycerol that are derived from
this tissue water (n). If one assumes that the 2H isotopic
enrichment (p) of hydrogen in each actively incorporated C–H of
-GP is equal to the 2H enrichment of body water and that
2H2O enrichment is constant during the labeling period,
n can be calculated from the measured p in body water and
the measured R (4,
19,
20).
Using calculation algorithms based on combinatorial probabilities as previously described (4, 19), a "lookup" table can be generated, describing R over a given range of values of p for discrete values of n (n = 3, n = 4, n = 5) (e.g. Table I). By using the value of p from measured body water enrichment, the expected R across the range of values for n is compared with the measured value of R (based on the lookup table, e.g. p = 4.00% yields R = 0.121 for n = 3, R = 0.143 for n = 4, R = 0.166 for n = 5). Because physiologic samples contain a mixture of glyc3.5 and glyc5 (see below), we treat n as a non-integral value for this modeling. A linear regression equation is then generated, reflecting the relationship between R and n at the experimentally determined p (the latter based on measured 2H2O enrichment). The value for n is then calculated from the measured R (e.g. if p = 4.00%, the equation is n = 43.86R–2.29, so for R = 0.152, n = 4.38) (Fig. 3).
|
|
1or the theoretical asymptotic value
for fully labeled TG-glycerol. Accurate determination of this asymptotic or
plateau value requires knowledge of n and is necessary for the
determination of fractional synthesis (f) using the precursor product
or rise-to-plateau approach
(19) in
Equation 2,
 | (Eq. 2) |
1 represents
the asymptotic or plateau value possible for the isotopic enrichment of the
mass +1 species of glycerol, calculated from p and a non-integer
value of n.
Relative Contribution of Glyceroneogenesis to Acylglyceride
Glycerol— The term "n" is a probability
parameter; therefore, we also calculated a measure with more physiological
meaning, by translating n into a relative contribution from
glyceroneogenesis to acylglyceride glycerol (a percentage). Assuming no
contribution from glycerol kinase, n can be assumed to come from only
two sources (glyc5 and glyc3.5). Thus n = 5x +
3.5(1–x), and in Equation
3, solving for x.
 | (Eq. 3) |
Statistical Analyses—A one-way analysis of variance with planned pair-wise comparisons was used with p < 0.05 as the criterion for significance for comparison of adipose n. Glucose enrichment was analyzed using a Kruskal-Wallis non-parametric analysis of variance (p < 0.05 criteria) followed by Mann-Whitney pair-wise comparisons. An independent group Student's t test and a nonparametric Mann-Whitney test with p < 0.05 as the criterion for significance was used for the fructose infusion comparison.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
|
Effect of Fructose Infusion—Body water enrichment for rats 5 h after an intraperitoneal bolus of 2H2O was 4.67 ± 0.21%. In the fasting state, prior to fructose infusion, values of n from plasma TG-glycerol were 4.51 ± 0.20 at –30 min and 4.41 ± .15 at –15 min, reflecting a stable n (Fig. 6) and an average relative contribution of glyceroneogenesis of 66%. After 3 h of fructose infusion, n in plasma TG-glycerol dropped significantly to 3.84 ± 0.08 (p < 0.01). After 4 h of infusion, n remained significantly suppressed from baseline at 4.01 ± 0.08, corresponding to a relative contribution from glyceroneogenesis of 34%.
Plasma Glucose Enrichment—2H enrichments for glucose were determined in both the morning and evening. Results from the morning and evening groups were pooled resulting in 10 measures for each diet/treatment group. The LC group resulted in the highest net [2H]glucose enrichment (3.2 ± 0.5), which was significantly (p = 0.004) higher than the rosiglitazone treatment group (2.3 ± 0.6) and minimally significantly higher (p = 0.03) than the HC group (2.6 ± 0.6). Therefore increased glucose enrichment from increased gluconeogenesis on a low carbohydrate diet could contribute to the increased n seen with the LC diet. The decrease in glucose enrichment with rosiglitazone treatment suggests all the increase in n is caused by increased glyceroneogenesis, and in fact its contribution if anything would be underestimated in this group.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
What had been less well recognized is the role that glyceroneogenesis may play in adipose tissue TG synthesis under altered physiologic states or after pharmacologic treatment. Glyceroneogenesis, a process first described almost 40 years ago (21–23), has been shown to be up-regulated during periods of decreased glucose availability such as during starvation (2) or feeding of a carbohydrate-free diet (24). We show here, in vivo, that an LC diet increases adipose TG-glycerol n, suggestive of increased glyceroneogenesis. It has previously been shown that a carbohydrate-free diet results in an increased PEPCK activity in adipose tissue (25), which could directly account for the increase in glyc5 contribution seen. The LC diet we used has been shown to increase gluconeogenesis in liver (26), and we did see an increase in plasma glucose enrichment. If plasma glucose derived from gluconeogenesis were subsequently taken up by adipose tissue and metabolized through glycolysis, this could also contribute to increased values of n. However, LC diets also reduce glucose uptake in adipose tissue in C57Bl/6J mice (27) and as shown by our data and others, glucose is preferentially taken up in the postprandial period when unlabeled dietary glucose predominates. Nonetheless, increased gluconeogenesis on the LC diet is clearly a confounder, and therefore, the percent contribution from glyceroneogenesis should be seen as a maximum possible contribution.
We also asked whether it was possible to experimentally decrease n
in liver TG, by infusion of fructose. A fructose load directly contributes
glycolytically derived -GP (glyc3 as there is no
contribution of 0.5 H-atom from the phosphoglucose isomerase reaction) and
suppresses the gluconeogenic contribution to triose phosphate formation in
liver (28). The decreased
liver TG-glycerol n that we observed experimentally is consistent
with a simultaneous effect of increasing glyc3 contribution (both
from increased glycolysis and potentially increased contribution from recycled
unlabeled glycerol via glycerol kinase) and decreasing glyc5
contribution by suppression of glyceroneogenesis.
Having shown under two conditions that the triglyceride n reflects
expected physiology and can be altered in predictable ways, we explored the
role of glyceroneogenesis in adipose TG synthesis following PPAR
agonist treatment. PPAR agonists, such as rosiglitazone, are
insulin-sensitizing agents that induce a number of actions, including
stimulation of adipogenesis and increased fat storage. PPAR is required
for the transcription of the PEPCK gene in adipocytes
(29), and rosiglitazone has
been shown to increase PEPCK expression and activity in adipose tissue
(30–32).
Conversion of oxaloacetate to phosphoenolpyruvate via PEPCK is generally
accepted as the rate-limiting step in glyceroneogenesis
(11)
(Fig. 2A). Controversy
has recently emerged about the effect of the PPAR agonist
thiazolidinediones on glyceroneogenesis and their role in promoting insulin
sensitivity and reducing net lipolysis. Rosiglitazone has been reported to
increase glycerol kinase activity in isolated adipocytes
(33). However, despite this
increase in glycerol kinase activity, it was subsequently shown in
vitro by Tordjman et al.
(32) that flux through
glycerol kinase in rosiglitazone-treated adipocytes remains a quantitatively
minor contributor to -GP synthesis and therefore to suppression of
fatty acid release. A physiologically significant contribution to adipogenesis
from increased glycerol kinase activity would result in a decrease in
n, whereas an increased contribution from glyceroneogenesis should
increase n (Fig. 1).
The latter was clearly observed during rosiglitazone treatment
(Fig. 5). There was a
nonsignificant decrease in net [2H]glucose enrichment with
rosiglitazone treatment, and therefore the increase in n cannot be
accounted for and is not confounded by an increased contribution from labeled
glucose. If, rosiglitazone treatment does significantly increase glycerol
kinase activity there could be recycling of previously labeled glycerol.
However that recycled glycerol would simply reflect its prior origin
(e.g. glyceroneogenesis or glycolysis). Whereas we would not be able
to detect that it had gone through glycerol kinase, we would know that it had
already gone through glyceroneogenesis and, as a matter of definition, we
consider that as glyceroneogenesis. If unlabeled glycerol were recycled, that
would appear with primarily an n of 3 and would look more like
glycolysis glycerol, which was not what we observed. The observed increase in
n with rosiglitazone treatment provides in vivo evidence for
greater significance of up-regulation of glyceroneogenesis, presumably via
increased PEPCK expression, compared with up-regulation of glycerol
kinase.
In vitro studies have suggested a greater increase in rosiglitazone-induced glyceroneogenesis in visceral adipose (retroperitoneal and mesenteric) compared with subcutaneous depots (34). We observed no significant differences in n between different adipose depots in vivo. It is possible that this difference was because of the duration of our study, which was weeks as opposed to hours in the in vitro study.
The observed stimulation of glyceroneogenic activity in adipose tissue may
be important to the mechanism of action of PPAR agonists.
Glyceroneogenic-induced increases in -GP availability favor fatty acid
re-esterification in adipose tissue and inhibition of fatty acid release into
plasma. This is believed to play a key role in the insulin-sensitizing action
of PPAR agonists (10).
Measurement of the glyceroneogenic contribution to adipose tissue -GP,
based on a change in the value of n, may therefore provide a novel
measure of in vivo net antilipolytic activity for PPAR
agonists or other agents that reduce free fatty acid release into the
plasma.
Our interpretation of these results could be altered by several metabolic
processes. First, we assume that -GP derived from glycerol kinase is in
rapid exchange with glycolytic intermediates in the triose phosphate pool,
resulting in an n of 3 rather than 0. This is not entirely accurate.
It has been shown in humans
(7), that of the plasma
glycerol that is incorporated via glycerol kinase into very low density
lipoprotein TG (i.e. TG synthesized in liver), 5% does not have
any exchange into the triose phosphate pool (i.e. retained all 5
2H labels from [2H5]glycerol infused). In the
presence of 2H2O, this would result in a small
percentage of n = 0 (glyc0). However, hepatic -GP
generated via glycerol kinase represents <20% of the total contribution to
TG-glycerol under normal conditions
(28,
35), so that 5% of this
relatively small number should not alter our quantitative interpretation. In
any case, since TG-glycerol from this source would be unlabeled, the measured
EM2/EM1 ratio is not affected and
would not affect calculated n
(19). There is also
potentially labeled glycerol being recycled via glycerol kinase and
-GP. We will consider the original source of glycerol (e.g.
glyceroneogenesis or glycolysis) for our calculation.
Second, glycolytic cycling of glucose to the level of pyruvate or
oxaloacetate then back to -GP can contribute to TG
(7,
36). Cycling of this type
converts glyc3.5 to glyc5, and thus can complicate
estimates of the proportion of TG-glycerol derived from glycolysis. As a
matter of definition, therefore, glyceroneogenesis must be recognized to
include glycolytic cycling or any other prior metabolic source of pyruvate or
oxaloacetate. Last, contributions to the triose phosphate pool from other
pathways such as the non-oxidative arm of the pentose phosphate cycle can
introduce non-glyc5 species to TG-glycerol. Those pathways make a
relatively small contribution relative to glycolytic flux
(37); however these
complexities of intracellular metabolism make precise calculations of the
quantitative sources of -GP difficult.
The results presented here have implications for the measurement of TG
synthesis from 2H. We report here that n can range from
3.73 to 4.44 in adipose tissue. The calculation of the theoretical plateau or
asymptotic value (A1) for fully labeled
TG-glycerol depends on the value of n, and accurate determination of
this plateau is required to measure fractional synthesis (f) of TG
using the precursor product or rise-to-plateau approach
(4,
19). Calculations of
A1 based on the range of n
determined here could cause A1 to vary
from 12.95% to 14.90% for p = 4.8%, as an example. This range of
calculated A1 can lead to a difference
in calculated f of up to 13%. Thus, considering the mutability of
n demonstrated in this study, experimental measurement of n
by the technique described here is optimal when calculating TG-glycerol
fractional synthesis, particularly under new experimental conditions.
In summary, we have demonstrated that values of n in TG-glycerol
can be calculated by MIDA and are influenced by physiological and
pharmacologic conditions. The variations observed were consistent with
expected differences in the glycolytic versus glyceroneogenic origins
of -GP. Our findings also underscore the important role that
glyceroneogenesis serves in maintaining TG homeostasis in adipose tissue and
liver. In particular, stimulation of glyceroneogenesis may play a central role
in the antilipolytic actions of thiazolidinediones on adipose tissue. Thus,
measurements of n can serve as an informative indicator of -GP
and, ultimately, TG metabolism under different physiological conditions.
|
FOOTNOTES |
|---|
¶ To whom correspondence should be addressed: San Francisco General Hospital, 1001 Potrero Ave., Bldg. 30, Rm. 3501-K, San Francisco, CA 94110. E-mail: smurph{at}itsa.ucsf.edu.
1 The abbreviations used are: TG, triacylglycerol; GP, glycerol phosphate;
MIDA, mass isotopomer distribution analysis; LC, low carbohydrate; HC, high
carbohydrate; GC-MS, gas chromatography-mass spectrometry; PEPCK,
phosphoenolpyruvate carboxykinase; PPAR, peroxisome
proliferator-activated .
2 R. Neese, unpublished observations.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Minehira, S. G. Young, C. J. Villanueva, L. Yetukuri, M. Oresic, M. K. Hellerstein, R. V. Farese Jr., J. D. Horton, F. Preitner, B. Thorens, et al. Blocking VLDL secretion causes hepatic steatosis but does not affect peripheral lipid stores or insulin sensitivity in mice J. Lipid Res., September 1, 2008; 49(9): 2038 - 2044. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Mancuso, H. F. Sims, X. Han, C. M. Jenkins, S. P. Guan, K. Yang, S. H. Moon, T. Pietka, N. A. Abumrad, P. H. Schlesinger, et al. Genetic Ablation of Calcium-independent Phospholipase A2{gamma} Leads to Alterations in Mitochondrial Lipid Metabolism and Function Resulting in a Deficient Mitochondrial Bioenergetic Phenotype J. Biol. Chem., November 30, 2007; 282(48): 34611 - 34622. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A. Varady, D. J. Roohk, Y. C. Loe, B. K. McEvoy-Hein, and M. K. Hellerstein Effects of modified alternate-day fasting regimens on adipocyte size, triglyceride metabolism, and plasma adiponectin levels in mice J. Lipid Res., October 1, 2007; 48(10): 2212 - 2219. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Mazzucotelli, N. Viguerie, C. Tiraby, J.-S. Annicotte, A. Mairal, E. Klimcakova, E. Lepin, P. Delmar, S. Dejean, G. Tavernier, et al. The Transcriptional Coactivator Peroxisome Proliferator Activated Receptor (PPAR){gamma} Coactivator-1{alpha} and the Nuclear Receptor PPAR{alpha} Control the Expression of Glycerol Kinase and Metabolism Genes Independently of PPAR{gamma} Activation in Human White Adipocytes Diabetes, October 1, 2007; 56(10): 2467 - 2475. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Tordjman, S. Leroyer, G. Chauvet, J. Quette, C. Chauvet, C. Tomkiewicz, C. Chapron, R. Barouki, C. Forest, M. Aggerbeck, et al. Cytosolic Aspartate Aminotransferase, a New Partner in Adipocyte Glyceroneogenesis and an Atypical Target of Thiazolidinedione J. Biol. Chem., August 10, 2007; 282(32): 23591 - 23602. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Gardner, S. M. Turner, A. Bautista, G. Lindwall, M. Awada, and M. K. Hellerstein Measurement of liver collagen synthesis by heavy water labeling: effects of profibrotic toxicants and antifibrotic interventions Am J Physiol Gastrointest Liver Physiol, June 1, 2007; 292(6): G1695 - G1705. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. J. Parks and M. K. Hellerstein Thematic review series: Patient-Oriented Research. Recent advances in liver triacylglycerol and fatty acid metabolism using stable isotope labeling techniques J. Lipid Res., August 1, 2006; 47(8): 1651 - 1660. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. N. Leroyer, J. Tordjman, G. Chauvet, J. Quette, C. Chapron, C. Forest, and B. Antoine Rosiglitazone Controls Fatty Acid Cycling in Human Adipose Tissue by Means of Glyceroneogenesis and Glycerol Phosphorylation J. Biol. Chem., May 12, 2006; 281(19): 13141 - 13149. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. R. Bederman, D. A. Dufner, J. C. Alexander, and S. F. Previs Novel application of the "doubly labeled" water method: measuring CO2 production and the tissue-specific dynamics of lipid and protein in vivo Am J Physiol Endocrinol Metab, May 1, 2006; 290(5): E1048 - E1056. [Abstract] [Full Text] [PDF] |
