| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Translational Metabolism Unit, Division of Diabetes, Endocrinology, and Metabolism, Department of Medicine; 2United States Department of Agriculture/Agricultural Research Service Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine; 3Endocrine Service, Ben Taub General Hospital, Houston, Texas; 4Division of Engineering and Physical Sciences, Office of Research Sciences, 5Kidney Disease Section, National Institute of Diabetes and Digestive and Kidney Diseases and 6Reproductive Biology and Medicine Branch, National Institute of Child Health and Human Development, National Institutes of Health, Department of Health and Human Services, Bethesda, Maryland; 7University of Colorado Health Sciences Center, Denver, Colorado; 8Institute of Clinical and Molecular Virology, University of Erlangen-Nürnberg, Bavaria; and 9Institute of Biochemistry, Humboldt University, Berlin, Germany
Submitted 5 April 2006 ; accepted in final form 11 July 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSDISCLOSURESREFERENCES |
|---|
and affect mitochondrial function in vitro. To test the hypothesis that expression of Vpr in liver and adipocytes can alter lipid metabolism in vivo, we engineered mice to express Vpr under control of the phosphoenolpyruvate carboxykinase promoter in a tissue-specific and inducible manner and investigated the effects of dietary fat, indinavir, and dexamethasone on energy metabolism and body composition. The transgenic mice expressed Vpr mRNA in white and brown adipose tissues and liver and immunoaffinity capillary electrophoresis revealed that they had free Vpr protein in the plasma. Compared with wild-type (WT) animals, Vpr mice had lower plasma triglyceride levels after 6 wk (P < 0.05) but not after 10 wk of a high-fat diet and lower plasma cholesterol levels after 10 wk of high-fat diet (P < 0.05). Treatment with dexamethasone obviated group differences, whereas indinavir had no significant independent effect on lipids. In the fasted state, Vpr mice had a higher respiratory quotient than WT mice (P < 0.05). These data provide the first in vivo evidence that HIV-1 Vpr expressed at low levels in adipose tissues and liver can 1) circulate in the blood, 2) regulate lipid and fatty acid metabolism, and 3) alter fuel selection for oxidation in the fasted state. human immunodeficiency virus lipodystrophy; glucose; triglycerides; cholesterol; energy expenditure; fat oxidation
or TNF- (5). Several antiretroviral drugs have been implicated, especially protease inhibitors and nucleoside reverse transcriptase inhibitors. The evidence for and against models that relate to the effects of the protease inhibitor class of drugs has been well summarized (8). In brief, some protease inhibitors can directly affect adipocyte metabolism and insulin-mediated glucose uptake in ways that might cause peripheral lipoatrophy (4, 17, 39) and insulin resistance (22). Fat distribution abnormalities could result from the use of some nucleoside reverse transcriptase inhibitor drugs (31, 20), potentially due to their toxic effects on mitochondrial structure and function. However, the pathophysiology could also involve factors intrinsic to HIV-1 itself or complex interactions between the antiretroviral drugs, nutritional factors (such as a high-fat diet), and HIV-1 derived proteins.
HIV-1 expresses several proteins that enhance its ability to replicate, propagate infection, and induce cellular apoptosis in host cells. One of these, the 96-amino acid viral protein R (Vpr), facilitates entry of the viral preintegration complex and inhibits the host cell cycle in the G2/M phase (11, 18, 29, 33). Vpr can also affect mitochondrial function through specific interactions with the adenine nucleotide translocator in the inner membrane, leading to dissipation of the transmembrane potential (10) and inducing apoptosis (11). Vpr is present in the sera of HIV-infected individuals in concentrations that correlate with the degree of viremia and disease progression (18, 36). Cells chronically infected by HIV-1 may continue to produce Vpr even after viral replication has been effectively suppressed by antiretroviral drugs (18, 29). The arginine-rich carboxyl terminus of Vpr resembles a typical protein transduction domain, and full-length Vpr can traverse plasma membranes of a variety of cells in a concentration-dependent but energy- and receptor-independent manner and localize in the nucleus (33) and mitochondria (11). Vpr contains at least two nuclear localization signals and can shuttle between nucleus and cytoplasm (33). Within the nucleus, it can modulate the transcriptional activity of host nuclear receptors. In vitro studies suggest that Vpr, through its LXXLL motif, is a coactivator of glucocorticoid receptor (GR)-mediated gene transcription (12, 33) and could act as a corepressor of peroxisome proliferator-activated receptor- (PPAR)-mediated gene transcription (34).
Signaling through GR or PPAR can regulate the expression of genes involved in lipid metabolism (35), and interference with mitochondrial membrane potential (10) could affect energy expenditure or substrate oxidation. Vpr-induced apoptosis (11) in adipocytes could alter adipocyte mass or function. Hence, a demonstration that in vivo expression of Vpr is associated with alteration of lipid, energy, or oxidative metabolism could suggest novel mechanisms underlying the complex metabolic changes that occur in HIV-infected patients. We hypothesized that Vpr, either per se or through interactions with dietary fat and a protease inhibitor drug, would alter lipid and glucose metabolism, circulating lipid levels, and energy balance in mice. Several transgenic mouse models expressing Vpr have been developed by other investigators. These include animals that express the protein in the heart (19), T lymphocytes (37), and kidneys (3). Although each model produced a phenotype consistent with known complications of HIV infection, lipid metabolic data were not reported, and the tissue specificities of Vpr expression would not directly and specifically test our hypothesis. We therefore engineered transgenic mice that express Vpr in the liver, white adipose tissue, and brown adipose tissue and also release Vpr into the bloodstream.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSDISCLOSURESREFERENCES |
|---|
Plasmid HUD 15.1, containing a 1.02-kb fragment encoding the tetracycline transactivator protein (tTA) was generously provided by Dr. Hermann Bujard, University of Heidelberg, Germany. The tTA gene was released by restriction digestion with EcoR1 and BamH1, and then Bgl II and EcoRV sites were created at the 5' and 3' ends, respectively. The tTA fragment was ligated into a plasmid containing nt –2,086 to +69 of the murine phosphoenolpyruvate carboxykinase (PEPCK) promoter, generously provided by Dr. Robert Hammer, Texas Tech University (1). A linearized PEPCK/tTA DNA construct free of vector sequences was obtained by restriction digestion at the Hpa1 and Xho1 sites.
A plasmid encoding the Vpr gene derived from the HIV-1 proviral clone pNL4–3 was cloned into the pTRE vector (Clontech, BD Biosciences, Franklin Lakes, NJ) at the SacII and BamH1 sites, which are located 3' to a tetracycline response operon promoter element (tet-op) and initiates transcription from the cytomegalovirus (CMV) promoter. A linearized tet-op/Vpr DNA construct free of vector sequences was obtained by restriction digestion at the Xho1 and HindIII sites.
Generation of Tetracycline-Regulated Transgenic Mice with Tissue-Specific Vpr Expression
The protocol was approved by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Animal Care and Use Committee. The PEPCK/tTA and tet-op/Vpr constructs were injected into the oocytes of FVB/N mice. Southern blotting identified 14 PEPCK/tTA founder mice, from which six lines were established, and 4 tet-op/Vpr founders, from which which lines were established (designated L2, L3, and L4). PEPCK/tTA founder mice were crossed with a reporter mouse bearing a tet-op/LacZ transgene (a gift of Dr. Bruce Conklin, University of California, San Francisco, CA) to test the system for tetracycline-regulated gene expression, and tissue cryosections were stained for 
-galactosidase activity (data not shown). Several PEPCK/tTA lines demonstrated induction of LacZ expression in liver, kidneys, and adipose tissue, and one, designated T8, was selected for further study.
PEPCK/tTA T8 mice were crossed with tet-op/Vpr L2, L3, and L4 mice to generate bi-transgenic mouse lines, designated T8L2, T8L3, and T8L4. The diet for pregnant and lactating dams contained tetracycline to inhibit Vpr transgene expression in fetuses and suckling mice. After weaning, mice were not exposed to tetracycline. Since the PEPCK promoter is activated by amino acids, transgene expression was upregulated by placing the mice on a high-protein diet for 4 wk.
Genotyping
Genomic DNA was isolated and purified from mouse tail tissue obtained at 3 wk of age using the DNeasy tissue kit (Qiagen, Valencia, CA) according to the manufacturer's recommendations. PCR was used to amplify a portion (444 bp) of the PEPCK/tTA transgene and a portion (558 bp) of the tet-op/Vpr transgene. The primers were: PEPCK/tTA forward 5'-ATTAGAGCTGCT TAATGAGGTCGG-3', reverse 5'-GGTGTTT CCCTTTCTTCTTTAGCG-3'; tet-op/Vpr forward 5'-CGCCTGGAGACGCCATCC-3', reverse 5'-CCACACCTCCCCCTGAAC-3'. The PCR products were resolved by agarose gel electrophoresis.
Mouse Breeding and Maintenance
The study was approved by the Baylor College of Medicine Institutional Animal Care and Use Committee. Pregnant dams were generated by cross-breeding hemizygotes for PEPCK/tTA and tet-op/Vpr as described above. This subcolony of mice was expanded by randomly mating one PEPCK/tTA transgenic mouse with one tet-op/Vpr transgenic mouse. Males aged 9–10 wk and females aged 7–8 wk were used for the breeding pairs. Three days before the animals were paired, the females were placed on a diet (catalog no. TD98186; Harlan Teklad, Madison, WI) containing 200 ppm doxycycline to suppress Vpr expression in utero. Offspring were weaned at 3 wk and fed a standard mouse diet [Purina, PicoLab Rodent Diet 20 (no. 5053, irradiated)] with ad libitum access to water and food.
For experiments, bitransgenic Vpr mice and wild-type (WT) mice of both sexes were randomly selected at 10–14 wk of age and housed individually at 23°C, 64% humidity, with a 12:12-h day-night cycle. They had ad libitum access to water and food. If a littermate was not available, then a mouse closest in age (±2 wk) was paired with that mouse (Vpr-Vpr and WT-WT). (Both male and female mice were used because of the limited supply of the transgenic animals and the large number required for all the experiments. Sex was included as a covariate in the statistical model.)
Experimental Design and Diets
The experiment was designed to measure the effects of varying durations (4 wk, 6 wk, 10 wk) of high-fat (41% of calories from fat) or low-fat (13% of calories from fat) diets on lipid, glucose, and energy metabolism in Vpr transgenic and WT mice. The protein content of the diet was controlled to regulate the promoter of the PEPCK/tTA transgene. The compositions of the prepared diets were as follows: in the high-protein diets, 52.2% of calories from protein; in the low-protein diets, 17.7% of calories from protein; in the high-fat/high-protein diet, 10.8 from carbohydrate and 20.6% from fat; in the low-fat/high-protein diet, 26.5% from carbohydrate and 5.6% from fat; in the high-fat/low-protein diet, 49.7% from carbohydrate and 20.2% from fat; in the low-fat/low-protein diet, 65.4% from carbohydrate and 5.2% from fat.
Four-week nutrition study. Vpr (n = 10, 5 male) and WT (n = 10, 4 male) mice were randomly assigned to either a high-fat/high-protein or a low-fat/high-protein diet for 4 wk. At the end of 4 wk, they were fasted for 12 h overnight and then killed under isoflurane anesthesia. Blood was collected from the heart in 20 g/l EDTA, and the plasma was separated by centrifugation and stored at –70°C. The liver, kidneys, heart, and samples of perigonadal fat and upper leg muscle were harvested, weighed, snap-frozen in liquid nitrogen, and stored at –70°C.
Six-week nutrition study. Vpr (n = 11, 5 male) and WT (n = 9, 3 male) mice were randomly assigned to either a high-fat/high-protein or low-fat/high-protein diet for 4 wk. At the end of 4 wk, their feed was changed to high fat/low protein or low fat/low protein, respectively, and this diet was maintained for 2 wk. The rationale for this maneuver was the possibility that PEPCK-driven Vpr expression might wane if the high-protein diet was continued beyond 4 wk and to see whether there were persistent metabolic effects despite diminished Vpr expression. At the end of the total 6-wk period, the animals were killed and plasma and organs harvested as described above.
Ten-week nutrition study, with or without indinavir. Vpr and WT mice of the group were randomly assigned to either a high-fat/high-protein or low-fat/high-protein diet for 4 wk. At the end of 4 wk their feed was changed to high-fat/low-protein or low-fat/low-protein, respectively, for 2 wk and then switched back to high-fat/high-protein or low-fat/high-protein for 4 more wk. (Thus animals received either a high-fat or a low-fat diet for 10 wk continuously, but the accompanying high-protein intake to induce Vpr expression was interrupted for 2 wk in the middle to limit the possibility that PEPCK-driven Vpr expression might wane with prolonged, continuous feeding of a high-protein diet). At the beginning of the 8th week of the experiment, Vpr and WT mice in both the high-fat and low-fat diet groups were randomly assigned to treatment with either indinavir or placebo, administered by subcutaneously implanted pellets (Innovative Research of America, Sarasota, FL). The implants were designed to release a dose of 20 mg indinavir/placebo over a 21-day period. At the end of the total 10-wk period, the animals were killed and plasma and organs harvested as described above. The numbers and sex distributions in the different groups were: Vpr + placebo (n = 19, 9 male); Vpr + indinavir (n = 21, 11 male); WT + placebo (n = 20, 11 male); WT + indinavir (n = 24, 13 male).
Dexamethasone study. The 4-wk nutritional study as described above was repeated in groups of mice randomly assigned to receive a dexamethasone pellet implanted through a subcutaneous incision or a sham incision at the beginning of the second week. The implant was designed to release dexamethasone at a dose of 12 µg/day for 21 days, a low dose previously used to stimulate biological effects in mice without producing pharmacological adverse effects (25). The total numbers of animals and sex distributions in the different groups were: Vpr + no implant (n = 22, 12 male); Vpr + dexamethasone (n = 10, 4 male); WT + no implant (n = 22, 9 male); WT + dexamethasone (n = 10, 4 male).
Energy expenditure.
Vpr (n = 25, 12 male) and WT (n = 19, 9 male) mice were randomly assigned to either a high-fat/high-protein or a low-fat/high-protein diet for 4 wk. At the end of 4 wk the mice were placed in a closed calorimetry chamber to monitor oxygen and carbon dioxide at the inlet and outlet ports using the Oxymax system (0100-005L; Columbus Instruments International, Columbus, OH). Animals were placed in the chambers for 24 h from 0800 with ad libitum feeding. Oxygen consumption (
O2) was calculated by the difference between the input and output oxygen flows. Carbon dioxide production (CO2) was calculated by the difference between the output and input carbon dioxide flows. The input ventilation rate (Vi) was calculated from the nitrogen fractions in the input and output samples, using the standard Haldane transform. Thus O2 = ViXi – VoXo, where Vi and Vo are the input and output ventilation rates and Xi and Xo are oxygen fractions at the inlet and outlet; CO2 = VoYo – ViXi, where Vi and Vo are input and output ventilation rates and Yi and Yo are carbon dioxide fractions at the inlet and outlet; Vi = Vo x No/Ni, where Ni and No are nitrogen fractions at the inlet and outlet; and Ns = 1 – Xs – Ys, where Ns is the nitrogen fraction of the sample, Xs is the observed oxygen fraction of the sample, and Ys is the observed carbon dioxide fraction of the sample.
At the end of the 24-h calorimetry period, body composition was measured by dual-energy X-ray absorptiometry, using a GE Lunar machine (PixiMus, Madison, WI). The animals were anesthetized with an intraperitoneal injection of Avertin (tribromoethanol/amylene hydrate, 240 mg/kg body wt) prior to the scan.
In other experiments to evaluate energy expenditure and substrate oxidation under fasting conditions, mice were restrained within the calorimetry chamber in a specially designed plastic sleeve to prevent voluntary movement and studied with access to water only but not food for 5 h, either from 6 AM to 11 AM (12 Vpr, 8 male; 10 WT, 4 male) or from 6 PM to 11 PM (9 Vpr, 6 male; 5 WT, 3 male).
Plasma Metabolites
Plasma glucose was measured by the Infinity Glucose Hexokinase Reagent, total cholesterol by the Infinity Cholesterol Reagent, and triglycerides by the Triglyceride Kit, all obtained from ThermoTrace (Victoria, Australia).
Tissue Vpr mRNA Expression
Total RNA was extracted from the livers, kidneys, and perigonadal white adipose tissues of three Vpr transgenic mice and three WT mice used in the 4-wk nutrition experiment and from brown adipose tissues of three Vpr and three WT mice used in the calorimetry experiment, using the TRIzol reagent (Invitrogen, Carlsbad, CA). The RNA was treated with RNAse-free DNAse and subjected to RT-PCR using the Superscript First Strand Synthesis System (Invitrogen) according to the manufacturer's instructions. The following primers were used to amplify a 201 bp Vpr product: forward 5'-CAGAAGACC AAGGGCCACAG-3' and reverse 5'-AATGAATAAACAGCAGTTGTTGCAG-3'. The PCR conditions were melting at 94°C for 45 s, annealing at 56°C for 60 s, and extension at 72°C for 60 s (42 cycles).
Serum Vpr Measurement by Immunoaffinity Capillary Electrophoresis
Monoclonal antibodies were raised against full-length synthetic Vpr. One antibody, termed 9F12, was selected for further use due to its strong reactivity with the NH2-terminal domain of Vpr. This antibody was digested by pepsin, and FAb fragments were produced by splitting the disulfide bridge with mercaptoethylene HCl (Pierce Biotechnology, Rockford, IL) (27, 28). The FAbs were directly immobilized to the internal surface of the first 5 cm of a 100-cm-long sulfhydryl-modified 100-µm (ID) fused silica capillary (Polymicro Technologies, Phoenix, AZ) via their free thiol groups. Five-microliter samples of mouse plasma were labeled with an equal volume of 1 ng/ml Alexa fluor 633 laser dye (Molecular Probes, Eugene, OR) in 0.5 M carbonate buffer, pH 9.5, for 30 min at room temperature. Fifty nanoliters of each labeled sample were introduced into the capillary by vacuum injection and allowed to remain there for 
10 min to ensure that any Vpr antigen was bound by the immobilized anti-Vpr antibody. Unbound material was purged by a phosphate buffer (pH 7.0) wash. The bound material was then recovered by electroelution at 100 µA constant current following a change in buffer pH to 1.5. Vpr was detected online with a laboratory-built, laser-induced fluorescence detector (26). Quantification was carried out by comparison with standards composed of WT mouse serum containing synthetic, full-length Vpr (7), and the lower limit of detection was 1.5 pg/ml.
Measurement of Plasma Indinavir Levels
Indinavir was extracted from 200 µl of mouse plasma using a liquid-liquid extraction assay. Extracted samples were dried under nitrogen and resuspended in 100 µl of acetonitrile-water (50:50). Samples were run in duplicate on an HPLC single quadrupole mass spectrometer. The HPLC mobile phase consisted of 53% acetonitrile and 47% ammonium acetate buffer (pH 4.9) and was run at a flow rate of 0.200 ml/min. The standard curve was linear in the range of 0.5–10 ng/ml; methyl-indinavir was added as an internal standard. Ions were detected using atmospheric pressure chemical ionization/positive ion mode with selected ion monitoring for indinavir and methyl-indinavir. Ion detection was at 614.80 for indinavir and 628.80 for methyl-indinavir (M + 1). Concentrations >10 ng/ml were analyzed with a UV detector at 266 nm.
Statistical Methods
General linear modeling techniques were used to assess the following outcomes: weight, body composition, and fasting glucose and lipids. The model included genotype (Vpr or WT), diet (high fat or low fat), implant (indinavir or placebo, dexamethasone or not), and time as factors. Interactions among these factors were also assessed. When interactions were detected, the analysis was split on the associated factor levels and reanalyzed. In addition, for analysis of the 10-wk experiment and the 4-wk experiment with dexamethasone, eight groups were formed on the basis of the combination of three factors: genotype, diet, and implant. These groups were compared using one-way ANOVA followed by post hoc pairwise testing procedures when indicated. To compare the curves for serial measurements of energy expenditure or respiratory quotient (RQ) in the calorimetry experiments, mean values per hour for each animal were analyzed by repeated-measures ANOVA, using the Statistical Package for Social Sciences software (SPSS, Chicago, IL). Data are presented as means (SD). Statistical significance was defined as P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSDISCLOSURESREFERENCES |
|---|
Vpr messenger RNA was expressed in the livers, kidneys, perigonadal white (WAT) and brown adipose tissue (BAT) of Vpr mice, but not in tissues from WT animals (Fig. 1). The presence of a Vpr product after treatment of tissue extracts with RNAse-free DNAase confirmed that Vpr mRNA, rather than transgenic DNA, was amplified in the RT-PCR reaction. (Control tissue extracts that were treated in this manner but did not undergo the reverse transcriptase reaction did not show a Vpr mRNA signal; data not shown). Since mRNA expression was not quantified, it was not possible to correlate tissue expression levels with circulating levels of Vpr protein in the blood.
|
|
The 10-wk experiment was designed to determine the effects of combinations of Vpr, dietary fat content, and protease inhibitor treatment on lipid metabolism. Plasma Vpr levels in the transgenic mice were similar to those in the 4- and 6-wk experiments (Table 2). Hence, the 10-wk experiment represented an even longer exposure to the dietary treatment (± indinavir) while maintaining similar plasma levels of Vpr as in the shorter protocols. Plasma indinavir levels were in the 1–40 ng/ml range. Plasma cholesterol levels were now lower in the Vpr than in the WT mice. In pairwise comparisons, this was true for all pairs except for the comparison between Vpr-high fat-indinavir and WT-high fat-indinavir groups, in which a possible interaction between indinavir and high-fat feeding might have overridden the effect of Vpr. Indinavir had no significant effect in WT or Vpr mice. The group difference in fasting triglyceride levels, seen in the 6-wk diet intervention, was lost in the 10-wk experiment. This was entirely due to a decrease in the triglyceride concentrations in the WT mice under all conditions, suggesting an adaptive process on the part of the WT mice to the experimental diets.
|
|
These experiments were designed to measure the whole body effects of Vpr on energy expenditure, substrate selection for oxidation and body composition, under basal conditions with dietary fat modification. In experiments where the mice were permitted to move freely within the calorimetry chamber and permitted to feed ad libitum, there was no measurable effect of Vpr on O2, CO2, or total energy expenditure (Table 4). However, the Vpr mice preserved body fat mass and percent body fat while on a low-fat diet, whereas the WT mice (appropriately) experienced a decline in body fat mass on the low-fat diet. To explore the possibility that the aberrant behavior of the Vpr mice might be due to a defect in whole body fat oxidation, we measured RQ every 10 min for 5 h while the mice were restrained in a comfortable plastic sleeve within the calorimetry chamber with access to water but not food (Fig. 2). This "fasting" experiment was performed by removing food from the cages either from 6 AM to 11 AM or from 6 PM to 11 PM (to account for different antecedent feeding behaviors of the mice during the dark or light cycle). Under these conditions, the WT mice had a steeper and progressively greater decline in RQ than the Vpr mice as the fast proceeded (P < 0.05), suggesting an impairment in the ability of the latter to utilize fat as a substrate for oxidation in the fasted state. This phenomenon occurred during both morning (early in the dark cycle) and evening (early in the light cycle) experimental conditions.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSDISCLOSURESREFERENCES |
|---|
These data are consistent with the hypothesized mechanism of Vpr action involving regulation of GR and PPAR; however, the in vivo effects are more complex than those suggested by in vitro cellular transcription studies (which would point to increased fatty acid release from adipocytes and perhaps increased VLDL-triglyceride synthesis in the liver leading to hypertriglyceridemia). In vitro, Vpr functions as a coactivator of GR (12) and a corepressor of PPAR (34). Glucocorticoid effects on appetite and lipid metabolism depend on the cellular site of action and the sensitivity of the specific pathways to their natural ligands. Zakrzewska et al. (38) have shown distinct effects on appetite and lipid metabolism with intracerebroventricular compared with intraperitoneal administration of dexamethasone in rodents. Central administration increased neuropeptide Y (NPY) expression in the hypothalamus, resulting in increased appetite, body weight, and plasma triglyceride levels, whereas peripheral administration of the same dose resulted in decreased hypothalamic NPY, with diminished appetite and body weight but increased leptin levels. The authors postulated that direct glucocorticoid action in the hypothalamus markedly increases NPY expression, overriding any opposing effect of leptin, whereas glucocorticoid action in the periphery does not change central NPY expression, so that leptin signaling is predominant in the hypothalamus. It is possible that, in our model, Vpr-enhanced GR activation via endogenous ligands could have produced an effect similar to short-term peripheral administration of glucocorticoid; conversely, persistent activation of GR by a potent exogenous ligand (the 4-wk dexamethasone implant experiment) could have produced an effect similar to central administration of glucocorticoid.
The PEPCK promoter used in our transgene contains a PPAR/retinoid X receptor binding element, the adipocyte-specific enhancer PCK2 (1), explaining why Vpr mRNA was expressed not only in the liver and kidneys but also in adipose tissues. Hence, in a negative feedback fashion, Vpr might inhibit PPAR-mediated activation of the PEPCK promoter, thus auto-limiting its own production and attenuating its GR-coactivating effects on lipolysis and triglyceride synthesis.
Vpr mRNA expression within liver, kidneys, WAT, and BAT could be responsible for its tissue-specific biochemical effects. This does not preclude the possibility that Vpr released from these tissues has distant effects on other tissues. Vpr was clearly detected in the plasma of the transgenic animals, and we have shown previously that Vpr readily transduces a variety of cell types and localizes within the nucleus (7, 33). Hence, the Vpr effects could result from complex mechanisms involving tissues other than those in which Vpr was initially expressed and could in part explain the differences between the effects of Vpr on GR- and PPAR-mediated transcription in vitro (12, 34) and the phenotypic effects in vivo.
This is the first study to report the quantitative measurement of Vpr in blood. Vpr has previously been detected, but not absolutely quantified, in plasma and cerebrospinal fluid (18, 36). A novel method, immunoaffinity capillary electrophoresis (ICE), was developed for this purpose. ICE is an electrokinetic assay employing immobilized antibodies to isolate specific analytes before separation and online detection by laser-induced fluorescence. Biological samples as small as 1 µl can be analyzed by comparing the area under the resolved peak to standard curves of known materials run under identical conditions. Thus, ICE combines the resolving power of capillary electrophoresis with the selectivity of immunoaffinity isolation and can measure protein analytes at concentrations as low as 400 fg/ml (27, 28). Employing a specific, immobilized, monoclonal antibody directed against the amino-terminal domain of Vpr enabled us to isolate Vpr from the plasma of the transgenic mice and measure its concentration in absolute units. Levy et al. (18), using polyclonal antibodies directed against partially purified recombinant Vpr or Vpr peptides, previously expressed Vpr levels in optical density units and compared them to p24 Gag protein.
Plasma levels of Vpr averaged 10–30 pg/ml in the transgenic animals. Vpr has some biological effects in cell culture at similar concentrations; these include neuronal cell death (9) and abrogation of Bcl-2 activity (23). Other in vitro effects require higher concentrations; these include promotion of viral replication (33), suppression of monocyte IL-12 production (21), and induction of TNF- secretion (24). It is possible that low levels of circulating Vpr were responsible for the mild phenotype of the transgenic mice in respect of the lipid parameters or its deviation from a typical "lipodystrophic" pattern.
The transgenic animals may have been "selected" to express Vpr protein at low levels. Vpr is toxic to the cells in which it is produced; our attempts to produce a constitutively Vpr-expressing transgenic mouse failed due to embryonic lethality. Only suppression of Vpr expression in utero with the "tet-off" system permitted the generation of viable transgenic mice. For the same reason, perhaps only mice with low levels of Vpr expression could survive to constitute the experimental lines. This might also explain why only mRNA expression by RT-PCR, but not protein expression, was readily detectable in the tissues. ICE is sensitive enough to detect and quantify Vpr protein in the plasma; it is not yet efficient in detecting Vpr in tissue lysates, probably because of the complexity and heterogeneity of proteins in solid tissues compared with plasma.
HIV-infected patients, both antiretroviral naïve (24) and those with antiretroviral-associated lipodystrophy (13), have elevated rates of energy expenditure, largely due to elevated resting energy expenditure, adjusted for fat-free mass (14). During a short-term "fast," Vpr mice manifest a distinct difference from WT animals in selection of fuels for energy, tending to spare fats as the fast progresses. Although careful kinetic studies are needed to ascertain whether this difference is due to altered fatty acid mobilization from adipose stores or in their oxidative or nonoxidative disposal, the known effects of Vpr suggest an alteration in mitochondrial oxidative phosphorylation. Vpr is known to localize to mitochondria and affect their function (10). Synthetic Vpr can penetrate the outer mitochondrial membrane through the voltage-dependent anion channel, interact with the adenine nucleotide translocator in the inner membrane to form a composite ion channel, and dissipate the transmembrane potential (11). This has been described in the context of Vpr's propensity to promote cellular apoptosis (10), but it is possible that, at low levels of expression, in the absence of other elements of HIV-1 infection, the predominant "mitochondriotoxic" effect of Vpr is dysregulation of energy production and impairment of fatty acid oxidation while fasting. It is possible that apoptotic effects resulting in adipocyte loss ("lipoatrophy") could occur with higher expression levels of Vpr. Further investigations are clearly required to validate these suggestive findings, and experiments are under way to obtain detailed measurements of caloric intakes and core body temperature and activity, in addition to calorimetry.
There was a trend toward an effect of indinavir on lipid metabolism in mice treated for 10 wk in the context of a low-fat diet. Lenhard et al. (16) previously found that mice of a different strain, treated with the same dose and delivery system of indinavir or placebo as well as proportions of dietary calories from fat, had higher triglyceride concentrations than control mice when fed a high-fat diet. Together, the data suggest that indinavir may tend to elevate plasma triglyceride levels in mice. Data from clinical studies suggest that nucleoside reverse transcriptase inhibitor drugs may also induce aspects of HIV lipodystrophy/dyslipidemia (2), and other protease inhibitors may cause more profound lipid changes in humans than indinavir. It is possible that a combination of antiretroviral drugs, as is commonly administered to humans with HIV infection, might have caused more marked metabolic effects than indinavir alone.
It is unclear why plasma triglyceride levels were not elevated in Vpr compared with WT mice despite blunted fat oxidation while fasting and why these levels decreased in both Vpr mice and WT mice over time so that there was no group difference at the end of 10 wk of dietary manipulation. These trends raise the possibility that Vpr may have little impact on the development of the adipocyte and lipid defects characteristic of human HIV infection or that the metabolic effects in mice are too different from those in humans to extrapolate from one species to the other. However, we believe that the present results point to clinically relevant but more complex effects of Vpr that require additional methods of investigation to uncover. We have recently found in detailed stable isotope/mass spectrometry studies of lipid turnover and oxidation in these animals that lipid kinetics are definitely deranged in the Vpr mice, in a manner consistent with the metabolic defects in humans with HIV lipodystrophy. Those studies must be completed and the data presented in a separate report, but they include altered lipolysis and fat mass. It is also important to keep in mind that the physiology of the transgenic animal would reflect only the effect of Vpr and not effects of intact HIV-1 or the inflammatory response, which could provide additional factors that might perturb lipid metabolism. In this regard, the biochemical effects of Vpr per se might more closely parallel the metabolic defects in HIV patients who are naïve to antiretroviral treatment. In such patients, changes in circulating lipids, though present, are generally less striking and lipodystrophic changes more subtle than in patients treated with a combination of antiretroviral drugs.
In conclusion, Vpr expression in the liver and adipose tissues of mice is associated with alterations in lipid metabolism and fuel selection for oxidation in the fasted state. Vpr is also released into the blood, where it could potentially transduce other tissues to affect their function. The cellular and molecular mechanisms underlying these effects, and their relevance to the syndromes of lipodystrophy, dyslipidemia, and insulin resistance in humans with HIV-1 infection, remain to be specified.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSDISCLOSURESREFERENCES |
|---|
|
DISCLOSURES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSDISCLOSURESREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
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.
* These two authors contributed equally to this work. 
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSDISCLOSURESREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
