| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Divisions of Endocrinology (A.C., L.J.H., J.E.S.) and Internal Medicine (M.B., L.J.H.), Department of Medicine, Jewish General Hospital and Lady Davis Institute, and Department of Anesthesia (F.C.), McGill University, Montreal, Québec, Canada
Address all correspondence and requests for reprints to: J. Enrique Silva, M.D., Division of Endocrinology, Jewish General Hospital, Room E-163, 3755 Chemin de la Côte-Ste-Catherine, Montreal, Québec, Canada H3T 1E2. E-mail: mdsi{at}musica.mcgill.ca
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The mechanisms by which the cells of homeothermic animals generate heat are not fully understood (3). A substantial fraction of thermogenesis simply represents the thermodynamic inefficiency of the biological machine and is closely related to ATP turnover. In addition, homeothermic animals can produce heat in a facultative manner by varying the thermodynamic efficiency with which ATP is generated. Regulated uncoupling of phosphorylation is a well accepted mechanism of thermogenesis in brown adipose tissue, wherein an uncoupling protein (UCP1) or thermogenin can dissipate, in a regulated manner, the proton gradient generated across the inner mitochondrial membrane by respiration. The collapse of the proton gradient reduces ATP synthesis and accelerates respiration, with the attendant dissipation of energy as heat (see Refs. 4, 5 for reviews). However, brown fat UCP1 could hardly explain the variability in resting energy expenditure (REE) in humans, as UCP1 is expressed only in brown adipose tissue, the amount of this tissue becomes limiting in adult humans, and the resting metabolic rate is by definition measured under conditions of thermoneutrality and in the fasting state, whereas UCP1 is activated by cold and food intake.
In recent years, Brand and colleagues suggested that a regulated proton leak could also take place in the mitochondria of other tissues (6, 7). They further suggested that this is a physiological mechanism that contributes to maintain body temperature and could be regulated by thyroid hormone, among other factors, accounting for a significant fraction of the thermogenic effect of this hormone (8, 9, 10). However, the molecular mechanism mediating such a leak has remained elusive. Recently cloned UCPs, of broader tissue distribution (11, 12, 13, 14, 15), have the potential to fill this gap in the hypothesis. Thus, UCP2 and UCP3 seem reasonable candidates to explain a proton leak in the mitochondria of several tissues, notably skeletal muscle, a tissue accounting for a significant fraction of the variability in REE (16) and where both UCPs are expressed.
Studies aimed at defining a role for UCP2 and UCP3 in the control of REE and energy balance have investigated the effect of spontaneous or experimental perturbations of energy balance on the expression of the corresponding genes. Although these studies do not negate a role for these proteins in energy balance, the results are hard to explain and even to reconcile with a role for these proteins in energy dissipation (see Ref. 17 and references therein). We have here taken the approach of investigating whether the variability in UCP2 and UCP3 gene expression could explain the variability in REE in the absence, to the extent possible, of the confounding influence of factors known to affect REE. A tight correlation between REE and measures of thermogenesis and the levels of UCP2 and UCP3 gene expression would establish a role for the regulation of these genes in the control of REE, thermogenesis, or both, whereas the lack of such a correlation would redirect attention to other levels of control, possibly the activities of these proteins. We thus examined the expression of these two genes in healthy, lean young males and correlated it with measures of resting metabolic rate and levels of hormones as well as fuels involved in energy balance. Another objective was to investigate the effect of the thiazolidinedione troglitazone on UCP2 and UCP3 messenger ribonucleic acid (mRNA) levels to follow up our recent observation that these antidiabetic drugs stimulate the expression of UCP2 in cell lines representing white and brown adipose tissues and skeletal muscle (18).
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The study, approved by the research and ethics committee of the
Jewish General, was performed in 18 healthy male volunteers recruited
from advertisements placed in local newspapers. Subjects responding to
the ad were screened after they were given full information on the
study and consented in writing to participate. The study was limited to
males to avoid the potentially confounding variability introduced by
cyclic ovarian function. Exclusion criteria included any disease,
condition, or drug that could affect either energy balance or the
measurement of REE (e.g. anxiety or claustrophobia) as well
as ages below 20 yr or greater than 45 yr or body mass index below 21
or above 26. Alcohol and caffeine usage were accepted if intake was
equal to or less than the equivalent of two glasses of wine or two cups
of coffee per day, respectively. Smokers and subjects involved in
strenuous exercise regimens were also excluded. Qualifying subjects
were submitted to a complete physical exam and laboratory screening for
liver disease, diabetes, thyroid dysfunction, bleeding, and other
hematological disorders. They were invited to participate if all tests
results were normal. Pertinent information is presented in Table 1
.
|
Subjects were urged not to change their life style during the study. They reported after a normal night’s sleep and in the fasting condition to a special study room in the Jewish General Hospital, between 0700–0900 h. Blood and urine samples were obtained for basal studies. They then underwent training for the measurement of REE (see below) and returned within a few days for the formal study. This time, they brought a 12-h urine collection from the preceding night for the measurement of catecholamines and creatinine, to document the completeness of the urine collection.
Energy expenditure was measured using the Deltatrac ventilated hood indirect calorimeter (Summit Technologies, Oakville, Canada), calibrated, and manipulated as described previously (19). Only the subject and the tester (M.B.) were present in an ad hoc pleasant, semidarkened, well ventilated, and thermoregulated (22–24 C) room. After the subject rested quietly for 30 min, a clear plastic hood was placed over the head and shoulders and ventilated with room air at 38 L/min. At an appropriate signal from the subject, oxygen and carbon dioxide sampling was begun. Data from the first 5 min of these measurements were discarded. CO2 and O2 volumes for each of the subsequent 15 min were averaged, converted to kilocalories using the Weir equation (20), and expressed on a per day basis. Average REE measured during the training session did not differ from the first formal measurement. The absolute variation from the training to the formal session was 1.83 ± 0.8% (±SEM). Body temperature was monitored at the end of the REE measurement at several sites to calculate mean skin and mean body temperature. Core body temperature was monitored by a sensor inserted in the ear canal, placed in contact with the tympanic membrane (21). Mean skin temperature was obtained from measurements on the chest, arm, thigh, and calf, calculated as described previously (22). Mean body temperature, a weighted mean between skin and core temperature and a reflection of body heat production, was derived as reported previously (21, 23).
After measurement of REE, each subject underwent a biopsy of quadriceps muscle (vastus lateralis) and the overlying sc adipose tissue. Fifty to 100 mg of each tissue were collected, immediately frozen over dry ice wrapped in a sheet of aluminum foil, transported to the laboratory within 30 min, and frozen at -80 C until processing. Biopsy was performed as follows. Skin and muscle fascia were infiltrated with 1% lidocaine; a small, 0.5-cm or smaller incision was made, and a 6-mm Bergstrom needle was introduced to collect a sample of sc adipose tissue; in a second pass, the needle was forced through the muscle fascia to collect a fragment of muscle. Each of two consecutive subjects was blindly assigned to take either troglitazone, two 300-mg tablets daily, or identical placebo tablets for 7 days. Both the drug and placebo were provided by Parke-Davis (Ann Arbor, MI). On day 6, the subjects returned for repeating REE measurements, and on day 7, they came for the last blood and urine measurements. At this time, they took the last dose of drug or placebo and, approximately 4 h later, underwent the second biopsy in the thigh not previously biopsied.
Biochemical analysis
Serum or plasma was analyzed for the various hormones indicated below, glucose, and fatty acids using standard, commercially available kits or automated methods routinely employed at the clinical chemistry laboratories of the Jewish General Hospital or the Nutrition and Food Science Center Laboratory of the Royal Victoria Hospital (insulin and glucagon). TSH, free T4, and free T3 were measured by automated methods using the Elcsys system (Roche Molecular Biochemicals marketed by Roche Diagnostics, Laval, Canada); urinary norepinephrine (NE) was measured by high performance liquid chromatography using reagents from Bio-Rad Laboratories, Inc. (Hercules, CA); serum leptin was measured with a kit from Linco Research, Inc. (St Charles, MO); insulin and glucagon were measured using in-house produced primary antibodies and radiolabeled hormones as well as standards from Linco Research, Inc.. Plasma free fatty acids (FFA) were measured with an enzymatic colorimetric assay using a commercial kit from Roche Molecular Biochemicals. To avoid interassay variability, all specimens for a given substance were run in a single assay.
RNA analysis
RNA from the tissue samples was extracted with guanidinium isothiocyanate, as previously described (24), and quantified by absorbance at 260 nm. It was then stored in small aliquots in diethylpyrocarbonate-treated water containing 0.4 U/µl RNasin at -80 C until used. UCP2 and UCP3 mRNAs were quantified by a competitive RT-PCR assay established in our laboratory. The method uses a competitor with identical ends to the target RNA but with different internal sequence and cloned in a transcribable vector (PCR MIMIC kit, CLONTECH Laboratories, Inc., Palo Alto, CA). Using the same primers as the corresponding target RNA, this competitor generates in the PCR a product 25–30% different in size from target RNA or complementary DNA, clearly resolved from the product of the test RNA in agarose gel electrophoresis. The competitor can be added as RNA to the RT reaction or as DNA to the PCR. RT was routinely performed using random DNA primers. PCR was performed with the following sequence-specific primers: UCP2: sense, 5'-TCAGGCCCGGGCTGGAGGTGGTCGGA, corresponding to bases 526–551 of mRNA; and antisense, 5'-CCTCTCGGGAAGTGCAGGCAGCCATG, corresponding to bases 982-1007 of the mRNA (25), located respectively in exons 5 and 8 of the gene (26); UCP3: sense, 5'-CCCAAAGGCGCGGACAACTCCAGCC; and antisense, 5'-GGGCCACCATCTTTATCATACAGTCGAGGGG, corresponding, respectively, to bases 475–499 and 919–949 of the human UCP3 mRNA (12). The sense primer sequence spans the end of exon 3 and the beginning of exon 4, whereas the sequence complementary to the antisense primer is located in exon 6, between nucleotides 223 and 253 of this exon, upstream of the stop codon that generates the short UCP3 mRNA (UCP3S) splice variant (27); that is, our assay detects both splice variants, UCP3S and UCP3L.
Even though we optimized the RNA extraction procedure to eliminate significant traces of genomic DNA, we took additional precautions to avoid artifactual, confounding signals in the RT-PCR. Thus, the design of primers spanning separate exons caused any product derived from contaminating genomic DNA to be substantially larger than the products of the mRNAs. Besides, in each assay we ran sham RT reactions (reverse transcriptase omitted) to ensure the absence of products other than those from the reversed transcribed mRNAs. The assay was further validated and progressively simplified using RNA extracted from surgical specimens. Initially, graded amounts of the competitor RNA were added to constant size aliquots of RNA samples, before the RT, to test the efficiency and reproducibility of the RT reaction. This proved to be quite reproducible from assay to assay and had minimal intraassay variability. Consequently, we performed additional measurements on previously analyzed RT products, now adding the competitor as DNA in graded amounts. The results were indistinguishable from those obtained by adding RNA competitor, as has been reported by others (28). This avoided the need for performing multiple RT reactions, with various amounts of the corresponding competitors, for each RNA sample. RNA integrity and homogeneity in sampling were systematically confirmed by ribosomal RNA quantity and appearance as well as ß-actin mRNA abundance. Lastly, to avoid interassay variations, all RNA samples from a tissue were assayed for a given mRNA in the same assay. Results are expressed as attomoles of the corresponding mRNA per µg total RNA.
Statistical analysis
Results are expressed as the mean ± SEM. Paired Student’s t test was used for comparison of values obtained from the same individual in separate occasions. To investigate factors potentially affecting the mRNA levels of the UCPs, we performed linear regression analysis using the initial set of values or the changes occurred between paired measurements. We used Prism 3.0 software (GraphPad Software, Inc., San Diego, CA) for statistical analysis. Data were tested for normal distribution, and Spearman rank test correlation was used in the rare event of a nonnormally distributed set of data. Results were considered statistically significant if the appropriate two-tailed P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
. The mean age was 29.7 yr (range, 21–43 yr), and subjects
were within the target body mass index, with an average of 23.3 (range,
21.5–25.6). All variables investigated were within the corresponding
reference ranges.
The UCP2 and UCP3 mRNA levels are shown in Fig. 1 and Table 2 along with their descriptive
statistics. Our muscle UCP2 mRNA value is almost identical to that
reported by Millet et al. (29), whereas the UCP3 and adipose
tissue UCP2 mRNA values were somewhat lower. However, Bao et
al. reported lower values for UCP3 than UCP2 mRNA in muscle (30).
These differences may be due to bias in the assessment of the
competitor in the RT-PCR assays and are of no consequence for internal
comparisons. The levels of three mRNAs, namely, muscle UCP3 and
UCP2 as well as adipose tissue UCP2, where highly variable among
individuals (Table 2), with SE between 10–20%
of the mean, a range of values of about 5- to 10-fold, and coefficient
of variations of 50–70%. This large variability among individuals can
be visually appreciated in Fig. 1. It is also evident that
troglitazone treatment had no consistent effect on any of
the three mRNAs (nor did it affect any of the other variables
measured). In the paired analysis, the mean differences between the
pre- and posttreatment measurements were not significantly different
from zero (Table 3). However, values
changed markedly and randomly between the two measurements in the
majority of the subjects, whether they were given placebo or
troglitazone, indicating that levels of these messenger
RNAs are also highly variable within individuals. Accordingly, there
was no correlation between pre- and posttreatment values, whether
segregated by treatment or pooled, except for adipose tissue UCP2 mRNA
in the control group (r = 0.50; P < 0.05; Table 3). In contrast, despite ranges of interindividual variation as large
as 18-fold for some of the other variables measured, there was a
significant correlation between basal and posttreatment values of TSH,
free T4, free T3, insulin,
and leptin as well as glucose (not shown). Such a correlation was not
found for cortisol, nocturnal urinary excretion of NE, or FFA (not
shown).
|
|
|
as well as glucose and FFA
concentrations are all related to energy balance and could potentially
affect the expression of UCPs. It has also been reported that in obese
subjects UCP2 expression is correlated with BMI and fat mass (29, 31),
and that fasting (29) and FFA (32) can affect the expression of this
UCP as well as UCP3 in humans and experimental models. Of the
measurements made in our healthy study subjects (Table 1), only
nocturnal NE excretion was significantly correlated to the abundance of
both UCP2 and UCP3 mRNA in skeletal muscle, but not in sc white adipose
tissue (Fig. 2).
|
. REE was 1444
± 31.8 Cal/day, with a coefficient of variation of 9.3%. Respiratory
quotient was 0.86 ± 0.06, reflecting a mixed diet. Mean body
temperature, which is a weighted mean of core body temperature and body
surface temperature and a function of heat production (22), was
34.7 ± 0.15 C. The individual data are presented in Fig. 3 to show the intra- and interindividual
variabilities and the lack of effect of troglitazone. Of
all three values, the most constant within individuals was REE, as
reflected by a significant correlation between the basal and
posttreatment results (r = 0.61; P < 0.008),
whereas there was no significant correlation between the two other
values in the paired comparison. None of these measures of metabolic
rate and thermogenesis correlated to the level of expression of UCP2 or
UCP3 (data not shown).
|
|
, only nocturnal urinary excretion of NE correlated
significantly with REE. Because of the homogeneity of the sample, this
correlation was not affected when NE was normalized to creatinine
excretion. Furthermore, to avoid the influence of anthropometric
variables, such as lean body mass or body surface area, on REE and
focus only in the intraindividual variation, we examined the percent
change in REE between the first and second individual measurements and
related it to simultaneous changes in possibly causative factors. As
troglitazone had no effect on any of these measurements
analyzed, the data for all 18 individuals were pooled. As shown in Fig. 4, the intraindividual changes in REE
between the first and second measurements correlated significantly with
the corresponding changes in free T4 and NE.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
REE represents the basic energetic cost of living at rest (33). It has been demonstrated to be quite variable, if not in absolute terms, in its meaning for energy balance. If one considers that the basic, minimum energetic cost of the most vital functions is a large fraction of REE, the variability of the excess, that is, the difference between this minimum energetic requirement and the actual REE, is even more striking (33). The variability of this component seems to be genetically determined, as demonstrated in comparative studies within and between families and in subjects with variable shares of genetic material (1, 34). Such variability is not irrelevant, as individuals with lower REE gain more weight during their lives (1, 35). One can calculate that a sustained positive balance of as little as 50 Cal/day, which is well within the variability of REE of normal individuals, could result in the accumulation of more than 20 kg adipose tissue in 10 yr if stored.
As potential determinants of energetic cost of living, the novel UCPs are reasonable candidates to explain the variability in energy expenditure, but we found that the level of expression of the corresponding genes did not fulfill this expectation. An obvious question emerging from these findings is the extent to which the variability could be methodological, as opposed to biological. As the methodology to measure these mRNAs was carefully standardized; the absence of degradation and the uniformity of a housekeeping gene mRNA, ß-actin, in the samples were confirmed; the accuracy of the analytical techniques was optimized; and interassay variation was avoided, we are quite confident that the variability observed is truly biological. Our results show that the mRNA levels are highly variable, but that variability did not correlate with measures of energy expenditure or factors known to affect it, which suggest that if these two genes are indeed important in determining the variability of energy expenditure and thermogenesis, the relation takes place at another level, the protein abundance or its activity. A question that we did not intend to address, because the tools are not available, is whether the variability in mRNA levels was reflected in the abundance of the corresponding proteins. Except for indirect evidence that there may be a correlation between UCP2 and its mRNA (36), there are no other data. We return to this issue later.
Nocturnal NE excretion accounted for a significant fraction of the variability of UCP mRNAs in muscle. Urinary NE excretion is more a reflection of sympathetic nervous activity than adrenal medulla secretion, and nocturnal NE excretion is less influenced by daily hazards, reflecting better the sympathetic tone of the individual. The data suggest, therefore, that sympathetic tone on skeletal muscle could be an important factor in the regulation of expression of these UCP genes. It has been shown in animals that sympathetic activity stimulates the expression of UCP3 and UCP2 in muscle (37, 38), and we have found the NE increases severalfold the level of UCP2 mRNA in rat L6 muscle cells (unpublished observations). On the other hand, NE excretion correlated with REE and the differences between the two measurements in NE excretion correlated with the changes in REE, underscoring the importance of the sympathetic nervous system to the regulation of metabolic rate. Therefore, this system controls both the synthesis of these UCPs and REE, but the data do not support the conclusion that the levels of expression of the UCP genes in muscle and REE are related. An inclusive hypothesis is that sympathetic activity controls both the expression of the genes and the activity of the UCPs, as it does with UCP1 in brown adipose tissue. In the latter, changes in the protein follow the mRNA with substantial delay (39, 40, 41), but this delay is compensated by stimulation of the activity of the protein through the NE-induced release of fatty acids (5). A constraint from the experimental design could have obscured a correlation between changes in UCP expression and REE. Thus, to avoid the effect of stress of the biopsies on REE and NE excretion and to maximize the possibility of seeing a short term effect of troglitazone on UCPs mRNAs, the second biopsy and REE measurement were performed more than 24 h apart. Nonetheless, the alternative that sympathetic activity influences REE predominantly via other mechanisms cannot be excluded.
Another major factor affecting energy expenditure is thyroid hormone, a role acquired by this hormone with the advent of homeothermy (42). We have previously shown in humans that REE is very sensitive to minor changes in the availability of T4 (43). Here, we report that the basal level of UCP2 and UCP3 expression does not correlate to TSH, free T4, or free T3, but found instead that the changes in REE observed between two consecutive measurements correlate well to changes in the free T4 concentration. TSH, free T3, and free T4 varied widely among study subjects, but varied much less within individuals, as reflected by the high level of correlation between consecutive measurements noted above. Of the three, the range of differences between the two measurements for TSH and free T3 were respectively 1–2 mU/L and 1–2 pmol/L, whereas in the case of free T4 the differences spanned 7–8 pmol/L. The changes in TSH and free T3 are probably below the sensitivity of the assays, whereas in the case of free T4 they were wider. It is possible that the differences in free T4 were indeed due to minor variations in TSH or other factors perturbing the exchange between the large pool of bound hormone and the small pool of free T4. Regardless of the mechanism, these results underscore the role of thyroid hormone in regulating REE (43). Although neither the levels of free T4 nor its changes within an individual correlated to the mRNA levels of UCP2 or UCP3, we do not believe that our findings contradict experimental studies showing that thyroid hormone affects the expression of muscle UCP3 and heart UCP2 in rats (37, 44, 45, 46), because such studies were performed in models with profound alterations of the thyroid status, whereas the data reported here are from euthyroid, healthy subjects. As with NE, these observations suggest that wide fluctuations in thyroid hormone, such as those occurring in the dysthyroid states, can affect the level of expression of the UCP genes, whereas smaller, physiological fluctuations could affect the activities of these proteins, but also, as with NE, thyroid hormone could, alternatively, affect REE via unrelated mechanisms.
An additional objective of the present studies was to investigate whether the acute administration of the thiazolidinedione, troglitazone, increased the mRNA levels of the novel UCPs, particularly UCP2. This part of the study was motivated by our observation that thiazolidinediones rapidly and vigorously stimulate UCP2 expression in cells representing muscle as well as brown and white adipose tissues (18), and it was important to pursue, because the documentation of such an effect in humans could point to potential relations between the UCPs and insulin sensitivity. Because of the rapidity of the effect in cells (2–4 h), we administered the drug for only a few days, and the last dose was taken within 4 h of the biopsy of white adipose tissue and muscle in case there could be a rapid, short-lived effect. Despite all precautions, we found no acute effect of troglitazone on the mRNA of either UCP, nor did we detect any effect on any of the measurements performed. These negative findings, however, do not exclude the possibility of an effect of thiazolidinediones on UCPs during longer administration or an effect in diabetic patients with insulin resistance. Indeed, the lack of influence of troglitazone on any of the variables studied suggests that in the absence of insulin resistance or hyperglycemia, these drugs have few, if any, detectable metabolic effects in humans, at least acutely.
In conclusion, we found that the level of expression of the UCP2 and UCP3 genes is quite variable, particularly in muscle, but this variability in gene expression does not explain the variability in surrogates of energy expenditure and thermogenesis in a homogeneous population of healthy, lean males. However, the expression of these two genes in muscle is significantly influenced by sympathetic activity, and the latter correlates, in turn, with energy expenditure. Fluctuations of thyroid activity also contribute to the variability in REE, but they do not affect the level of expression of the UCPs. These results suggest more complexity than anticipated in the regulation of the function of these proteins, the expression of the corresponding genes being just one level of control. It is possible that the activities of these proteins are under complex regulation, among other factors, by the sympathetic nervous system and thyroid hormones. On the other hand, the possibility that the function of UCP2 and UCP3 is not related to energy expenditure regulation and that sympathetic activity and thyroid hormone control REE by unrelated mechanisms cannot be excluded.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received September 20, 1999.
Revised December 29, 1999.
Accepted January 18, 2000.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. E. Silva Thermogenic Mechanisms and Their Hormonal Regulation Physiol Rev, April 1, 2006; 86(2): 435 - 464. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Knudsen, P. Laurberg, L. B. Rasmussen, I. Bulow, H. Perrild, L. Ovesen, and T. Jorgensen Small Differences in Thyroid Function May Be Important for Body Mass Index and the Occurrence of Obesity in the Population J. Clin. Endocrinol. Metab., July 1, 2005; 90(7): 4019 - 4024. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. T. Murphy, R. J. Snow, A. C. Petersen, R. M. Murphy, J. Mollica, J. S. Lee, A. P. Garnham, R. J. Aughey, J. A. Leppik, I. Medved, et al. Intense exercise up-regulates Na+,K+-ATPase isoform mRNA, but not protein expression in human skeletal muscle J. Physiol., April 15, 2004; 556(2): 507 - 519. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. DosSantos, A. Alfadda, K. Eto, T. Kadowaki, and J. E. Silva Evidence for a Compensated Thermogenic Defect in Transgenic Mice Lacking the Mitochondrial Glycerol-3-Phosphate Dehydrogenase Gene Endocrinology, December 1, 2003; 144(12): 5469 - 5479. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Silva The Thermogenic Effect of Thyroid Hormone and Its Clinical Implications Ann Intern Med, August 5, 2003; 139(3): 205 - 213. [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Murphy, K. K. O. Watt, D. Cameron-Smith, C. J. Gibbons, and R. J. Snow Effects of creatine supplementation on housekeeping genes in human skeletal muscle using real-time RT-PCR Physiol Genomics, January 15, 2003; 12(2): 163 - 174. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J. Calsbeek, T. L. Thompson, J. A. Dahl, N. R. Stob, J. T. Brozinick Jr., J. O. Hill, and M. S. Hickey Metabolic and anthropometric factors related to skeletal muscle UCP3 gene expression in healthy human adults Am J Physiol Endocrinol Metab, October 1, 2002; 283(4): E631 - E637. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
