Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Uridine Metabolism in HIV-1-Infected Patients: Effect of Infection, of Antiretroviral Therapy and of HIV-1/ART-Associated Lipodystrophy Syndrome

  • Pere Domingo ,

    pdomingo@santpau.cat

    Affiliation Infectious Diseases Unit, Hospital de la Santa Creu I Sant Pau, Barcelona, Spain

  • Javier Torres-Torronteras,

    Affiliation Laboratory of Mitochondrial Disorders, Institut de Recerca Hospital Vall d'Hebron and CIBERER, Barcelona, Spain

  • Virginia Pomar,

    Affiliation Infectious Diseases Unit, Hospital de la Santa Creu I Sant Pau, Barcelona, Spain

  • Marta Giralt,

    Affiliation Departament de Bioquimica i Biologia Molecular and Institut de Biomedicina de la Universitat de Barcelona (IBUB), Barcelona, Spain

  • Joan Carles Domingo,

    Affiliation Departament de Bioquimica i Biologia Molecular and Institut de Biomedicina de la Universitat de Barcelona (IBUB), Barcelona, Spain

  • Maria del Mar Gutierrez,

    Affiliation Infectious Diseases Unit, Hospital de la Santa Creu I Sant Pau, Barcelona, Spain

  • José M. Gallego-Escuredo,

    Affiliation Departament de Bioquimica i Biologia Molecular and Institut de Biomedicina de la Universitat de Barcelona (IBUB), Barcelona, Spain

  • Maria Gracia Mateo,

    Affiliation Infectious Diseases Unit, Hospital de la Santa Creu I Sant Pau, Barcelona, Spain

  • Pedro Cano-Soldado,

    Affiliation Departament de Bioquimica i Biologia Molecular and Institut de Biomedicina de la Universitat de Barcelona (IBUB), Barcelona, Spain

  • Irene Fernandez,

    Affiliation Infectious Diseases Unit, Hospital de la Santa Creu I Sant Pau, Barcelona, Spain

  • Marçal Pastor-Anglada,

    Affiliation Departament de Bioquimica i Biologia Molecular and Institut de Biomedicina de la Universitat de Barcelona (IBUB), Barcelona, Spain

  • Francesc Vidal,

    Affiliation Infectious Diseases Unit, Hospital Universitari Joan XXIII, Tarragona, Spain

  • Francesc Villarroya,

    Affiliation Departament de Bioquimica i Biologia Molecular and Institut de Biomedicina de la Universitat de Barcelona (IBUB), Barcelona, Spain

  • Antoni Andreu,

    Affiliation Laboratory of Mitochondrial Disorders, Institut de Recerca Hospital Vall d'Hebron and CIBERER, Barcelona, Spain

  • Ramon Marti

    Affiliation Laboratory of Mitochondrial Disorders, Institut de Recerca Hospital Vall d'Hebron and CIBERER, Barcelona, Spain

Uridine Metabolism in HIV-1-Infected Patients: Effect of Infection, of Antiretroviral Therapy and of HIV-1/ART-Associated Lipodystrophy Syndrome

  • Pere Domingo, 
  • Javier Torres-Torronteras, 
  • Virginia Pomar, 
  • Marta Giralt, 
  • Joan Carles Domingo, 
  • Maria del Mar Gutierrez, 
  • José M. Gallego-Escuredo, 
  • Maria Gracia Mateo, 
  • Pedro Cano-Soldado, 
  • Irene Fernandez
PLOS
x

Abstract

Background

Uridine has been advocated for the treatment of HIV-1/HAART-associated lipodystrophy (HALS), although its metabolism in HIV-1-infected patients is poorly understood.

Methods

Plasma uridine concentrations were measured in 35 controls and 221 HIV-1-infected patients and fat uridine in 15 controls and 19 patients. The diagnosis of HALS was performed following the criteria of the Lipodystrophy Severity Grading Scale. Uridine was measured by a binary gradient-elution HPLC method. Analysis of genes encoding uridine metabolizing enzymes in fat was performed with TaqMan RT-PCR.

Results

Median plasma uridine concentrations for HIV-1-infected patients were 3.80 µmol/l (interquartile range: 1.60), and for controls 4.60 µmol/l (IQR: 1.8) (P = 0.0009). In fat, they were of 6.0 (3.67), and 2.8 (4.65) nmol/mg of protein, respectively (P = 0.0118). Patients with a mixed HALS form had a median plasma uridine level of 4.0 (IC95%: 3.40–4.80) whereas in those with isolated lipoatrophy it was 3.25 (2.55–4.15) µmol/l/l (P = 0.0066). The expression of uridine cytidine kinase and uridine phosphorylase genes was significantly decreased in all groups of patients with respect to controls. A higher expression of the mRNAs for concentrative nucleoside transporters was found in HIV-1-infected patients with respect to healthy controls.

Conclusions

HIV-1 infection is associated with a decrease in plasma uridine and a shift of uridine to the adipose tissue compartment. Antiretroviral therapy was not associated with plasma uridine concentrations, but pure lipoatrophic HALS was associated with significantly lower plasma uridine concentrations.

Introduction

Pyrimidines are synthesized de novo through a multistep process starting from glutamine and carbon dioxide to form the pyrimidine ring, orotic acid [1]. The synthesis of orotic acid is catalyzed by dihydroorotate dehydrogenase (DHODH), an enzyme located in the inner mitochondrial membrane, and functional connection to the respiratory chain via ubiquinone ensures efficient oxidation of dihydroorotate [2]. Then, orotate is converted to its nucleotide form in the presence of 5-phosphorylribose-pyrophosphate. Orotate monophosphate is converted by a multifunctional enzyme, uridine monophosphate (UMP) synthase to the nucleotide UMP (Figure 1). UMP is the pivotal nucleotide from which uridine nucleotides di-and triphosphates are formed by ATP-dependent kinases [3]. A large portion of the pyrimidines are salvaged from the degradation of the nucleic acids and nucleotides [4]. The concentration of circulating plasma uridine is tightly regulated throughout different species and individuals [5], [6]. The liver appears to have this homeostatic control on uridine degradation and formation [7]. Uridine is essentially cleared in a single pass through the liver and is replaced by “new uridine” formed by de novo synthesis [7].

thumbnail
Figure 1. Schematic representation of uridine synthesis and its transport into the cell.

MP  =  monophosphate.

https://doi.org/10.1371/journal.pone.0013896.g001

Besides the critical role of uridine in the synthesis of RNA and bio-membranes, there is a role for uridine in regulating a series of biological functions such as complex effects in the regulation of vascular resistance, a role in spermatogenesis, a modulatory effect on the peripheral nervous system, and maintenance of normal CNS activity [8]. In animal models, uridine reduces anxiety, is a sleep-inducing factor, and may regulate body temperature [8]. Aside from its physiological effects, uridine appears to have remarkable functions in tissues under stress or pathological situations such as restoration of myocardial ATP concentrations [9], maintenance of brain metabolism during ischemia or severe hypoglycemia [10], recovery from neuronal degeneration produced by diabetic neuropathy [11], and as a rescue agent in 5-fluorouracil therapy to decrease bone marrow and gastrointestinal toxicity [12].

The triphosphates are the active forms of the nucleoside analogue reverse transcriptase inhibitors, but they also inhibit polymerase gamma, resulting in mitochondrial DNA depletion [13], although reduction of mitochondrial DNA copy number is also associated to HIV infection in itself [14]. Because the de novo pyrimidine synthesis needs the correct function of the mitochondrial respiratory chain, pyrimidine nucleotide depletion has been hypothesized to contribute to this toxicity by exacerbating the competitive effect of the analogues [15]. However, the effect of these drugs on patients' uridine concentrations has been studied in few cases [16]. Uridine has been able to abrogate the adverse effects of antiretroviral pyrimidine analogues in adipocytes [17], and the benefitial effects of uridine supplementation in cell culture models of stavudine toxicity seems to involve the recovery from nucleoside analogue-induced pyrimidine depletion, because uridine supplementation also prevented the effects of redoxal, a specific DHODH inhibitor which, in turn, exacerbated the mitochondrial toxicity of the stavudine [15]. In a clinical trial, supplementation of uridine to patients on thymidine analogues was associated with fat recovery in limbs [18].

Our working hypothesis was that variations in plasma and fat concentrations of uridine may be involved in the modulation of the toxic effects of thymidine analogues. Specifically, we postulated that the toxic effects of thymidine analogues might result in pyrimidine depletion, detectable as a reduction of systemic and/or fat uridine concentrations. Therefore, the aim of the present study was to look for a relationship between plasma and fat uridine concentrations and HIV-1/highly active antiretroviral (HAART)-associated lipodystrophy syndrome (HALS).

Materials and Methods

Subjects

All patients were recruited through the same clinic at the Hospital de la Santa Creu I Sant Pau, which attends a population of 1455 HIV-1-infected patients on active follow-up, and were consecutive patients with an established diagnosis of HIV-1 infection. Patients were eligible whether or not they had HALS, whether or not they were on antiretroviral therapy, and if they were not taking uridine or derivatives. Subjects who were hospitalized or had a frank cognitive impairment such as delirium or dementia on enrolment were not eligible. Patients with opportunistic infections, acute hepatitis, liver insufficiency, neoplasms or fever of undetermined origin were excluded from the study. At the time of study entry no patient used any other drug known to influence glucose metabolism or fat distribution. Informed consent was obtained from the patients at study entry. The diagnosis of AIDS was based on the 1993 revised case definition of the Centers for Disease Control and Prevention (CDC) [19]. Controls were recruited among Hospital personnel and had to be negative for HIV-1 infection and be between 35 and 45 years of age with a proportion of males of about 70%. To be eligible they did not have to meet any of the exclusion criteria. Written informed consent was obtained from patients and controls at study entry. The study was approved by the Ethics Committee of the Hospital de la Santa Creu i Sant Pau.

Body composition measurements, definitions of HALS and metabolic syndrome, and biochemistry laboratory measurements

Body composition measurements, definitions of HALS and metabolic syndrome have been described elsewhere [20]. All laboratory investigations were performed as previously described [21]. All laboratory samples, including those for uridine determination, were obtained after a 12 hr overnight fasting. (Supporting information file S1).

Fat tissue samples

These were obtained from subcutaneous adipose tissue (SAT) depots through a small surgical biopsy performed by an 8 mm punch under local anesthesia with mepivacaine. One half of the SAT obtained was immediately frozen in liquid nitrogen and stored at −80°C until RNA extraction (see below). The remaining was used in fresh for determining fat uridine concentrations.

Measurement of plasma and fat uridine concentrations

For uridine determination, anticoagulated blood was centrifuged at 1,500× g and plasma was separated and kept at −20 C until analysis. Fat tissue was homogenated in the presence of 10 volumes (e.g., 50 mg of tissue +500 µl) of homogenization buffer (10 mmol/l HEPES pH 7.5; 5 mmol/l EDTA; 5 mmol/l DTT; 5 mmol/l MgCl2; plus Complete-Mini™ protease inhibitor cocktail, Roche), in an ice-cold bath. After centrifugation, the supernatant was saved and protein concentration was determined by the Bradford method [22].

Uridine was measured by a binary gradient-elution HPLC method. 200 µl of plasma or tissue homogenate were deproteinized by the addition of 9 µl of cold perchloric acid 11.7 mol/l (final concentration 0.5 mol/l) and kept 5 min on ice. Precipitated proteins were eliminated by centrifugation and 5 µl of the supernatant were injected into an Acquity UPLC apparatus (Waters, Milford, MA) and eluted at 0.5 ml/min with a saline buffer eluent A (20 mM ammonium acetate, pH 5.6) and an organic eluent B (methanol gradient grade) according to the following gradient: 0 to 1.1 min, 100% eluent A; 1.1 to 6.1 min, 100% to 82.6% eluent A; 6.1 to 6.2 min, 82.6 to 100% eluent A; 6.2 to 7.2 min, 100% eluent A. The column used was an Acquity UPLC BEH C18 column 100×2.1 mm, 130 Å pore size, 1.7 µm particle size (Waters). Optical absorbance of the eluate was monitored at 267 nm and definitive identification of the uridine peak was based upon retention time and treatment of a second aliquot of the sample with a large excess of purified E. coli thymidine phosphorylase (Sigma-Aldrich, St. Louis, MO) to eliminate enzymatically the uridine peak. The area from residual peaks coeluting with uridine and still present after treatment with E. coli thymidine phosphorylase was substracted from the area of the uridine peak obtained in the untreated aliquot, to ensure no overestimation of uridine concentration. The quantitation of the nucleoside was based on peak areas using external aqueous standards. The method had a between-day imprecision of 2.3% (coefficient of variation) for a uridine concentration of 4.30 µmol/l. Uridine concentrations in plasma of ∼3 to 5 µmol/l are found in different species and individuals [5], [6], [12].

Expression of genes encoding uridine metabolism-related enzymes and nucleoside transporters

RNA extraction was performed using a column-affinity based methodology (Rneasy, Qiagen, Hilden. Germany). On-column DNA digestion was performed during RNA purification (Rnase-Free Dnase set, Qiagen). TaqMan Reverse Transcription and RT-PCR reagents were used for mRNA analysis (Applied Biosystems, Foster City, USA). One microgram RNA was transcribed into cDNA random-hexamer primers and the real-time reverse transcriptase-polymerase chain reaction was performed on the ABI PRISM 7700HT sequence detection system (Applied Biosystems). The TaqMan RT-PCR reaction was performed in a final volume of 25 µl using TaqMan Universal PCR Master Mix, NoAmpErase UNG reagent and the specific gene expression primer probes. The TaqManGene Expression assays (Applied Biosystems) used were: uridine-monophosphate synthase, Hs00923516; uridine cytidine kinase-1, Hs00258815; uridine phosphorylase, Hs00427695; uridine 5′-monophosphate hydrolase (also called cytosolic 5′-nucleotidase III), Hs00826433; concentrative nucleoside transport (CNT-1), Hs00984402; CNT-2, Hs01035852; CNT-3, Hs00223220. Controls with no RNA, primers, or reverse transcriptase were included in each set of experiments. Each sample was run in duplicate and the amount of the mRNA for the gene of interest in each sample was normalized to that of the reference housekeeping control (HPRT1, Hs999999009), as already reported [23] using the comparative (2-ΔCT) method. Calculations based on a second, independent, housekeeping gene (18S rRNA) led to similar results. Data are expressed as means ± SEM.

Measurement of plasma fatty acid concentrations

The serum composition of fatty acids was determined using the method by Lepage and Roy [24]. Aliquots of 300 µL of plasma were transferred into glass tubes for direct transesterification. 2 mL of methanol-benzene (4∶1, v/v) with internal standard (heptadecanoic acid, C17:0) and 0.01% butylhydroxytoluene, as antioxidant. Samples were vortexed at low speed while slowly adding 200 µL of acetyl chloride, little by little, over a period of 2 minutes. The tubes were tightly closed with teflon-lined caps and vortexed 30 seconds.

Samples were then heated for 60 minutes at 100°C in a heating block and shaking continuously at 600 rpm. After the tubes had been cooled to room temperature, five millilitres of 6% (w/v) potassium carbonate were then added. The samples were vortexed for 30 seconds and centrifuged at 2500 rpm for 20 minutes at 15°C. The fatty acid methyl esters contained in the upper benzene phase were transferred to gas chromatography vials and stored at 4°C until injection into the chromatograph.

The analysis was performed on a Varian CP-3900 gas chromatograph equipped with a flame ionization detector, using a capillary column model CP9205-VF-WAXms (Varian), 30 m length ×0.25 mm internal diameter ×0.25 µm film thickness. Individual fatty acids were identified by order of elution and upon comparison with known commercially prepared fatty acid standards (GLC 566-C, Nu-Chek Prep Inc.). Fatty acid methyl ester peaks were identified by comparison of retention times of standards and quantified in comparison to known commercially prepared reference standards. The percentage of each fatty acid class was expressed as percentage of total fatty acids.

Statistical analyses

All analyses were performed with the Statistical Package for Social Sciences version 17.0 (SPSS, Chicago, IL). Data are expressed as mean ± SD or median with interquartile range (IQR). Data that were not normally distributed, as determined using the Kolmogorov-Smirnov test, were logarithmically transformed before analysis. Student's t test was used for comparison between two groups, Pearson's correlations or one-way ANOVA and multiple testing were corrected using Bonferroni correction. Stepwise logistic regression analysis was used to examine the association of plasma uridine concentrations and other parameters with HALS. The variables selected to enter into stepwise regression were those that correlated significantly with plasma uridine concentrations (after Bonferroni correction for multiple testing). In all statistical tests, P values <0.05 were considered significant.

Results

Population studied

Two-hundred and twenty one HIV-1-infected patients and 35 healthy controls were studied. The demographics, HIV-1 infection, and antiretroviral exposure parameters from patients and controls are shown in table 1. Mean duration of HIV-1 infection was 7.8±6.8 years (median: 7.0 [IQR: 12.2 years]), and 63 patients (28.5%) had had a prior AIDS-defining condition. Fifty-nine patients were co-infected with hepatitis C virus (26.7%), whereas 10 (4.5%) had chronic hepatitis B virus infection. Anthropometric, metabolic data and plasma uridine concentrations are shown in Table 2.

thumbnail
Table 1. Demographics, HIV-1 infection and antiretroviral exposure parameters.

https://doi.org/10.1371/journal.pone.0013896.t001

Antiretroviral drug exposure and immuno-virological situation

Among treated patients, 59 (41.3%) had undetectable viral load at the moment of the study. The other 84 treated patients presented a median detectable viral load of 2.3 log10 copies/ml (IQR: 1.1 log10 copies/ml). The mean CD4 count was 589±320 cells/mm3 (median: 531 [IQR. 425] cells/mm3). Nadir CD4 cell count was <100 cells/mm3 in 57 patients (25.8%). The cumulated exposure to antiretroviral drugs is shown in table 1. Twenty patients (13.9%) of the treated group were on an antiretroviral regime that included AZT whereas 25 (17.5%) were on a d4T-based regime at the moment plasma and fat uridine concentrations were measured. There were not statistically significant differences in plasma and fat uridine concentrations between patients currently on AZT (P = 0.9930) or d4T (P = 0.4944), either considered individually or as a combined group (P = 0.5809).

Diagnosis of HALS and metabolic syndrome

HALS, based on 7 or more points on the Lipodystrophy Scale Grading Score (LSGS), was diagnosed in 84 treated patients (58.7%) whereas absence of HALS was found in 59 treated patients (41.3%), and in all the naives. All 84 patients presented with lipoatrophy, but in addition 44 (52.4%) had a mixed lipoatrophic-lipohypertrophic form. Differences between HALS and non-HALS patients are shown in tables 1 and 2. Patients with a mixed form had a median plasma uridine level of 4.0 (IC95%: 3.40–4.80) whereas in those with isolated lipoatrophy it was 3.25 (2.55–4.15) µmol/l/l (P = 0.0066).

Plasma uridine concentrations and its correlation with cardio-metabolic, HIV-1 infection and antiretroviral therapy factors

Plasma uridine concentrations for HIV-1-infected patients ranged from 1.60 to 7.4 µmol/l/l, with a mean value of 3.85±1.18 µmol/l (median: 3.80; IQR: 1.60). The corresponding values for controls were: range: 2.0–9.4 µmol/l, mean: 4.36±1.28 µmol/l, median: 4.60, IQR: 1.80 µmol/l (P = 0.0009). Plasma uridine concentrations were not significantly different between HALS and non-HALS patients (P = 0.3377). Patients with plasma uridine concentrations <3 µmol/l were 8.9% among controls, 16.7% among naïves, 19.0% among treated patients without HALS, and 29.6% among patients with HALS (P = 0.0460). There was no sex difference in plasma uridine concentrations for HIV-1-infected patients, but there were statistically significant differences between controls (men: median 5.0 [IQR: 1.5] vs. women: 3.45 [1.4], P = 0.0262). In naïve patients, there was a negative correlation between plasma concentrations of uridine and viral load (R = −0.2580, P = 0.0228). Patients who were currently receiving treatment with AZT or d4T had a median plasma uridine level of 3.90 [IQR: 1.67] whereas in those who did not it was 3.70 [IQR: 1.80) µmol/l (P = 0.4278). There was no correlation between cumulated d4T use either in time (months) (R = 0.094, P = 0.3935) or grams (R = 0.078, P = 0.4815). There were not statistically significant differences between patients who did or did not have AIDS, CD4 nadir <100 cells/mm3, did or did not have not clinical toxicity (peripheral neuropathy or pancreatitis), HALS, HCV infection, and metabolic syndrome with respect to plasma uridine concentrations. Lean subjects had median plasma uridine concentrations of 3.7 [IQR: 1.5], overweight subjects 4.1 [IQR: 1.8], and obese subjects 3.9 [IQR: 1.67] µmol/l (P = 0.0152 for lean vs. obese). The correlations of plasma uridine concentrations with metabolic, infectious and treatment factors are shown in table 3. Plasma uridine concentrations were not found to be independently associated either with the metabolic syndrome or with HALS in HIV-1-infected patients.

thumbnail
Table 3. Correlations of serum uridine levels with metabolic, infectious and treatment factors.

https://doi.org/10.1371/journal.pone.0013896.t003

Fat uridine concentrations

Uridine content was measured in fat biopsy samples from 19 patients (naïve: 10, no HALS: 5, HALS: 4) and 15 controls. Median uridine level in fat from HIV-1-infected patients was 6.0 [IQR: 3.67], and in controls, 2.8 [4.65] nmol/mg protein (P = 0.0118). Median fat uridine level was 6.1 [0.50], 2.6 [1.85], and 6.3 [3.4] nmol/mg protein, for naïve, non-HALS patients, and HALS patients, respectively (P = 0.3767). There was a negative correlation between plasma and fat uridine concentrations in HIV-1-infected patients (R = −0.5210, P = 0.0223), whereas such a correlation was not found for controls (R = 0.0680, P = 0.8720). Median fat uridine level was 5.5 [IQR: 4.0] nmol/mg protein for those who were currently treated with AZT or d4T, whereas it was 2.6 [IQR: 1.27] nmol/mg protein for those not treated (P = 0.5369). There were no statistically significant differences between patients with mixed HALS and pure lipoatrophic HALS with respect to fat uridine concentrations 6.80 [IQR: 0] vs. 6.55 [IQR: 0.50] nmol/mg protein, respectively, P = 0.4028). The only correlations were for EFV exposure (R = −0.4890, P = 0.0336), and for PUFAs level (R = −0.5740, P = 0.0252).

Expression of genes encoding uridine metabolism-related enzymes and concentrative uridine transporters

The expression of genes involved in uridine metabolism is depicted in figure 2. The expression of the transcript for UMP synthase was not significantly modified in adipose tissue from all HIV-1 infected patient groups with respect to controls, regardless of having been treated or having developed HALS or not. The expression of the gene for UMP hydrolase (also called cytosolic 5′-nucleotidase III) was slightly decreased in naïve patients but unaltered in the other groups of HIV-infected patients (Figure 2). The pattern of changes observed for uridine cytidine kinase and uridine phosphorylase (UPase) indicated a significant decrease in all groups of patients with respect to controls. There were no changes in comparisons between naïve and treated patients or in patients with or without HALS. The expression of concentrative nucleoside transporters (CNTs), potentially involved in concentrating intracellular uridine [25], showed a similar pattern of changes in the three subtypes of transporters in naïve and treated patients, in agreement with previous findings [23]. Naïve patients showed a higher expression of the mRNAs for CNTs with respect to healthy controls, which was statistically significant for CNT1 and CNT3. HAART-treated patients, either having HALS or not, showed significantly higher concentrations of expression of the three CNT mRNA subtypes with respect to controls and naïve, with distinct concentrations of statistical significance depending on the CNT subtype (Figure 2). No differences were observed for the expression of CNTs mRNAs between patients with or without HALS.

thumbnail
Figure 2. Expression of genes encoding enzymes involved in uridine metabolism and concentrative nucleoside transporters (CNT).

Data from subcutaneous adipose tissue of uninfected controls and HIV-1 infected patients. Figure shows means + SEM for each specific mRNA concentration (expressed as ratio relative to HPRT mRNA) from 6-8 patients/group. P values for statistical comparisons between groups are shown when <0.05. MP  =  monophosphate. HALS-  =  No HIV-1/HAART-associated lipodystrophy syndrome. HALS+  =  HIV-1/HAART-associated lipodystrophy syndrome. HAART  =  Highly Active AntiRetroviral Therapy.

https://doi.org/10.1371/journal.pone.0013896.g002

Discussion

Our study does not suggest an association of the development of features of HALS or metabolic syndrome with plasma and fat concentrations of uridine, except for lipohypertrophy, but suggests that HIV-1 infection is associated with a significant decrease in uridine plasma concentrations. However, the number of patients with uridine <3 µmol/l was higher in patients with than in non-HALS patients. Antiretroviral therapy that included thymidine analogues did not influence uridine plasma concentrations. On the other hand, fat uridine concentrations showed an inverse correlation with plasma concentrations. This finding may be explained by an over-expression of those genes that encode concentrative nucleoside transporters, and particularly CNT3 [23], [25], as well as by under-expression of genes encoding enzymes leading to anabolic (UCK) or catabolic (UPase) uridine consumption.

However, the present results have to be viewed in the light of their inherent limitations. The first one comes from the nature of the study; this is a cross-sectional study and therefore, no causal relationships can or should be drawn. Second, for the majority of patients, uridine was measured in plasma whereas the anatomical and biological processes giving rise to HALS occur in the adipose depots. We have some measurements of uridine in fat, and although such measurements are highly reproducible and with tight variability coefficients, their reliability is difficult to ascertain [26]. In addition, the number of patients included in this substudy was low, and this may hamper the validity of our results.

The plasma uridine results are trustworthy since we were especially careful to avoid overestimations of the uridine peak, as we have observed that some unknown compounds sometimes coelute with uridine in our chromatographic separation. This fact could explain, at least in part, the values slightly higher for uridine plasma concentration reported by other groups [18], and could account for the difference between our results, which do not detect reduced uridine concentrations in treated patients, an the work by Langmann et al that, in contrast, reports significat reductions in dideoxy-NRTI treated patients [16]. We double tested enzymatically the peak of uridine by treating a second aliquot of the samples with an excess of thymidine phosphorylase from E. coli, which ensures that we only measured true uridine, with no interfering compounds eluting at the same retention time in the chromatogram. This may explain why plasma uridine values obtained in our patients with HALS are lower than those reported by Sutinen et al. [18] for patients with clinically evident lipoatrophy. On the other hand, the results obtained in our study compared well with previously reported uridine plasma concentrations [5], [6], [12].

We have found that HIV-1 infection is associated with significantly lower concentrations of uridine in plasma, whereas there were not differences between HIV-1-infected patients irrespective of their treatment status, but it was by the presence of lipoatrophic HALS. This may be explained by the fact that, unlike what happens in non-infected patients, there is a shift of uridine towards adipose depots and maybe other tissues. Additionally, among naïve patients, a negative correlation was found between uridine plasma concentrations and the amount of plasma HIV-1 RNA. This may suggest its consumption during the HIV-1 replicative cycle. On the other hand, alterations due to HIV-1-infection in adipose tissue from patients, i.e. reduction in the amounts of mitochondrial DNA [27], or impaired expression of genes implicated in adipocyte anabolic functions [28], have been previously reported.

Uridine supplementation has been advocated as a successful treatment for thymidine analogues-related lipoatrophy based on a short-term randomized, double-blind, placebo-controlled trial [18]. However, recent data from a randomized, double blind, placebo-controlled trial, failed to show any benefit of uridine supplementation in terms of fat gain after 48 weeks of treatment [29]. In addition, when compared with a strategy of switching from thymidine analogues to tenofovir, supplementation with uridine induced a pro-inflammatory state without showing benefit in terms of fat gain [30]. Therefore, it seems that uridine administration cannot reverse the adipose toxic effects of thymidine analogues, and may even represent an additional risk because of its potential inflammatory induction. This is not at odds with our findings concerning the positive correlation between plasma uridine concentrations and polyunsaturated fatty acid (PUFAs) concentrations. The beneficial effects of PUFAs on the cardiovascular system and on lipid profile are well known [31].

The concentrations of uridine in animal experimentation tissues are usually far in excess of the concentration of uridine in plasma [1]. Both, plasma and intracellular concentrations of uridine are regulated by the catabolic activity of uridine phosphorylase (UPase) and by transport mechanisms [9][11], [32]. We have found an underexpression of UPase in adipocytes which may partly justify the greater concentration of fat uridine compared to that of plasma (Figure 2). In addition, the increased expression of CNTs in HIV-1-infected patients' adipose tissue contribute to the shift of uridine into the adipose cells. The overall shift in the expression of these genes leading to induce an increase in the fat tissue pyrimidine pool fits well with a previous report suggesting a role of intracellular pyrimidine depletion associated to nucleoside analogues toxicity [15]. Although we did not see any effects in the systemic uridine concentrations, the increase of intracellular uridine in treated patients might be, in fact, a reflex of a compensatory mechanism to avoid intracellular pyrimidine depletion.

In summary, our results suggest that uridine homeostasis is profoundly disturbed by HIV-1 infection. Although plasma concentrations in HIV-1-infected patients were lower than in controls, the only additional uridine disturbance that could be documented was unrelated to antiretroviral therapy but it was to the development of pure lipoatrophic syndrome. These results are in agreement with the lack of efficacy of uridine supplements in HALS reversal as shown by ACTG 5229.

Author Contributions

Conceived and designed the experiments: PD ALA RM. Performed the experiments: JTT MG JCD JMGE PCS IF MPA F. Villarroya ALA RM. Analyzed the data: PD JTT VP MG JCD MdMG MGM IF MPA F. Vidal F. Villarroya ALA RM. Contributed reagents/materials/analysis tools: JTT MG JCD MdMG JMGE MGM PCS MPA F. Villarroya ALA. Wrote the paper: PD F. Vidal F. Villarroya ALA RM. Selected and recruited patients: VP MdMG MGM IF.

References

  1. 1. Pizzorno G, Cao D, Leffert JJ, Russell RL, Zhang D, et al. (2002) Homeostatic control of uridine and the role of uridine phosphorylase: a biological and clinical update. Biochim Biophys Acta 1587: 133–144.G. PizzornoD. CaoJJ LeffertRL RussellD. Zhang2002Homeostatic control of uridine and the role of uridine phosphorylase: a biological and clinical update.Biochim Biophys Acta1587133144
  2. 2. Löffler M, Fairbanks LD, Zameitat E, Marinaki AM, Simmonds HA (2005) Pyrimidine pathways in health and disease. Trends Mol Med 11: 430–437.M. LöfflerLD FairbanksE. ZameitatAM MarinakiHA Simmonds2005Pyrimidine pathways in health and disease.Trends Mol Med11430437
  3. 3. Simmonds HA (1995) Enzymes of nucleotide biosynthesis: differences between intact and lysed cells as well as between species and tissues can be important. Biochem Soc Trans 23: 877–879.HA Simmonds1995Enzymes of nucleotide biosynthesis: differences between intact and lysed cells as well as between species and tissues can be important.Biochem Soc Trans23877879
  4. 4. Moyer JD, Oliver RE, Handschumacher RE (1981) Salvage of circulating nucleosides in the rat. Cancer Res 41: 3010–3017.JD MoyerRE OliverRE Handschumacher1981Salvage of circulating nucleosides in the rat.Cancer Res4130103017
  5. 5. Pizzorno G, Yee L, Burtness BA, Marsh JC, Darnowski JW, et al. (1998) Phase I clinical and pharmacological studies of benzylacyclouridine, a uridine phosphorylase inhibitors. Clin Cancer Res 4: 1165–1175.G. PizzornoL. YeeBA BurtnessJC MarshJW Darnowski1998Phase I clinical and pharmacological studies of benzylacyclouridine, a uridine phosphorylase inhibitors.Clin Cancer Res411651175
  6. 6. Traut TW (1994) Physiological concentrations of purines and pyrimidines. Mol cell Biochem 140: 1–22.TW Traut1994Physiological concentrations of purines and pyrimidines.Mol cell Biochem140122
  7. 7. Gasser T, Moyer JD, Handschumacher RE (1981) Novel single pass exchange of circulating uridine in rat liver. Science 213: 777–778.T. GasserJD MoyerRE Handschumacher1981Novel single pass exchange of circulating uridine in rat liver.Science213777778
  8. 8. Connolly GP, Duley JA (1999) Uridine and its nucleotides : biological actions, therapeutic potentials. Trends Pharmacol Sci 20: 218–225.GP ConnollyJA Duley1999Uridine and its nucleotides : biological actions, therapeutic potentials.Trends Pharmacol Sci20218225
  9. 9. Aussedat J (1983) Effect of uridine supply on glycogen resynthesis after ischemia in the isolated perfused rat heart. Cardiovasc Res 17: 145–151.J. Aussedat1983Effect of uridine supply on glycogen resynthesis after ischemia in the isolated perfused rat heart.Cardiovasc Res17145151
  10. 10. Benzi G, Villa RF, Dossena M, Vercesi L, Gorini A, et al. (1984) Cerebral endogenous substrate utiization during the recovery period after profound hypoglycemia. J Neurosci Res 11: 437–450.G. BenziRF VillaM. DossenaL. VercesiA. Gorini1984Cerebral endogenous substrate utiization during the recovery period after profound hypoglycemia.J Neurosci Res11437450
  11. 11. Gallai V, Mazzotta G, Montesi S, Sarchielli P, Del Gatto F (1992) Effects of uridine in the treatment of diabetic neuropathy: an electrophysiological study. Acta Neurol Scand 86: 3–7.V. GallaiG. MazzottaS. MontesiP. SarchielliF. Del Gatto1992Effects of uridine in the treatment of diabetic neuropathy: an electrophysiological study.Acta Neurol Scand8637
  12. 12. Van Groningen CJ, Leyva A, Kraal I, Peters GJ, Pinedo HM (1986) Clinical and pharmacokinetics studies of prolonged administration of high-dose uridine intended fro rescue from 5-FU yoxicity. Cancer Treat Rep 70: 745–750.CJ Van GroningenA. LeyvaI. KraalGJ PetersHM Pinedo1986Clinical and pharmacokinetics studies of prolonged administration of high-dose uridine intended fro rescue from 5-FU yoxicity.Cancer Treat Rep70745750
  13. 13. Lewis W, Dalakas MC (1995) Mitochondrial toxicity of antiviral drugs. Nat Med 1: 417–422.W. LewisMC Dalakas1995Mitochondrial toxicity of antiviral drugs.Nat Med1417422
  14. 14. Maagaard A, Holberg-Petersen M, Kvittingen EA, Sandvik L, Bruun JN (2006) Depletion of mitochondrial DNA copies/cell in peripheral blood mononuclear cells in HIV-1-infected treatment-naive patients. HIV Med 7: 53–58.A. MaagaardM. Holberg-PetersenEA KvittingenL. SandvikJN Bruun2006Depletion of mitochondrial DNA copies/cell in peripheral blood mononuclear cells in HIV-1-infected treatment-naive patients.HIV Med75358
  15. 15. Setzer B, Lebrecht D, Walker UA (2008) Pyrimidine nucleoside depletion sensitizes to the mitochondrial hepatotoxicity of the reverse transcriptase inhibitor stavudine. Am J Pathol 172: 681–690.B. SetzerD. LebrechtUA Walker2008Pyrimidine nucleoside depletion sensitizes to the mitochondrial hepatotoxicity of the reverse transcriptase inhibitor stavudine.Am J Pathol172681690
  16. 16. Langmann P, Trein A, Schnaitmann E, Eckert R, Klinker H, et al. (2008) Uridine levels in plasma correlate with clinical and laboratory signs of mitochondrial toxicity: AIDS 2008 - XVII International AIDS Conference: Abstract no. CDB0457.P. LangmannA. TreinE. SchnaitmannR. EckertH. Klinker2008Uridine levels in plasma correlate with clinical and laboratory signs of mitochondrial toxicity: AIDS 2008 - XVII International AIDS Conference: Abstract no. CDB0457.
  17. 17. Walker UA, Auclair M, Lebrecht D, Kornprobst M, Capeau J, et al. (2006) Uridine abrogates the adverse effects of antiretroviral pyrimidine analogues on adipose cell functions. Antivir Ther 11: 25–34.UA WalkerM. AuclairD. LebrechtM. KornprobstJ. Capeau2006Uridine abrogates the adverse effects of antiretroviral pyrimidine analogues on adipose cell functions.Antivir Ther112534
  18. 18. Sutinen J, Walker JA, Sevastianova K, Klinker H, Häkkinen AM, et al. (2007) Uridine supplementation for the treatment of antiretroviral therapy-associated lipoatrophy: a randomized, double-blind, placebo-controlled trial. Antivir Ther 12: 97–105.J. SutinenJA WalkerK. SevastianovaH. KlinkerAM Häkkinen2007Uridine supplementation for the treatment of antiretroviral therapy-associated lipoatrophy: a randomized, double-blind, placebo-controlled trial.Antivir Ther1297105
  19. 19. Centers for Disease Control. (1993) 1993 revised classification system for HIV infection and expanded surveillance for case definition for AIDS among adolescents and adults. MMWR 41(RR-17): 1–13.Centers for Disease Control.19931993 revised classification system for HIV infection and expanded surveillance for case definition for AIDS among adolescents and adults.MMWR41RR-17113
  20. 20. Domingo P, Cabeza MC, Pruvost A, Salazar J, Gutierrez MM, et al. (2010) Relationship between HIV/highly active antiretroviral therapy (HAART)-associated lipodystrophy syndrome and stavudine triphosphate intracellular levels in patients with stavudine-based antiretroviral regimes. Clin Infect Dis 50: 1033–1040.P. DomingoMC CabezaA. PruvostJ. SalazarMM Gutierrez2010Relationship between HIV/highly active antiretroviral therapy (HAART)-associated lipodystrophy syndrome and stavudine triphosphate intracellular levels in patients with stavudine-based antiretroviral regimes.Clin Infect Dis5010331040
  21. 21. Moreno-Torres A, Domingo P, Pujol J, Blanco-Vaca F, Arroyo JA, et al. (2007) Liver triglyceride content in HIV-1-infected patients on combination antiretroviral therapy studied with 1H-MR spectroscopy. Antivir Ther 12: 195–203.A. Moreno-TorresP. DomingoJ. PujolF. Blanco-VacaJA Arroyo2007Liver triglyceride content in HIV-1-infected patients on combination antiretroviral therapy studied with 1H-MR spectroscopy.Antivir Ther12195203
  22. 22. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254.MM Bradford1976A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding.Anal Biochem72248254
  23. 23. Guallar JP, Cano-Soldado P, Aymerich I, Domingo JC, Alegre M, et al. (2007) Altered expresión of nucleoside trasnsporter genes (SLC28 and SLC29) in adipose tissue from HIV-1-infected patients. Antivir Ther 12: 853–863.JP GuallarP. Cano-SoldadoI. AymerichJC DomingoM. Alegre2007Altered expresión of nucleoside trasnsporter genes (SLC28 and SLC29) in adipose tissue from HIV-1-infected patients.Antivir Ther12853863
  24. 24. Lepage G, Roy CC (1986) Direct transesterification of all classes of lipids in a one-step reaction. J Lipid Res 27: 114–120.G. LepageCC Roy1986Direct transesterification of all classes of lipids in a one-step reaction.J Lipid Res27114120
  25. 25. Pastor-Anglada M, Molina-Arcas M, Casado FJ, Bellosillo B, Colomer D, et al. (2004) Nucleoside transporters in chronic lymphocytic leucemia. Leukemia 18: 385–393.M. Pastor-AngladaM. Molina-ArcasFJ CasadoB. BellosilloD. Colomer2004Nucleoside transporters in chronic lymphocytic leucemia.Leukemia 18385393
  26. 26. Mouton J Theuretzbacher U, Craig WA, Tulkens PM, Derendorf H, et al. (2008) Tissue concentrations: do we ever learn?. J Antimicrob Chemother 61: 235–237.U. Mouton J TheuretzbacherWA CraigPM TulkensH. Derendorf2008Tissue concentrations: do we ever learn?.J Antimicrob Chemother61235237
  27. 27. Coté HC, Brumme ZL, Craib KJ, Alexander CS, Wynhoven B, et al. (2002) Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV-infected patients. N Engl J Med 346: 811–820.HC CotéZL BrummeKJ CraibCS AlexanderB. Wynhoven2002Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV-infected patients.N Engl J Med346811820
  28. 28. Giralt M, Domingo P, Guallar JP, Rodriguez de la Concepción ML, Alegre M, et al. (2006) HIV-1 infection alters gene expression in adipose tissue, which contributes to HIV-1/HAART-associated lipodystrophy. Antivir Ther 11: 729–740.M. GiraltP. DomingoJP GuallarML Rodriguez de la ConcepciónM. Alegre2006HIV-1 infection alters gene expression in adipose tissue, which contributes to HIV-1/HAART-associated lipodystrophy.Antivir Ther11729740
  29. 29. McComsey G, Walker U, C Budhathoki C, Su Z, Currier JS, et al. (2010) Uridine Supplementation in the treatment of HIV Lipoatrophy. AIDS 24: 2507–2515.G. McComseyU. WalkerC. C BudhathokiZ. SuJS Currier2010Uridine Supplementation in the treatment of HIV Lipoatrophy.AIDS2425072515
  30. 30. McComsey G, O'Riordan MA, Choi J, Libutti D, Rowe D, et al. (2010) A 48-week Randomized Study of Uridine Supplementation vs Switch to TDF on Limb Fat, Mitochondrial Function, Inflammation, and Bone Mineral Density in HIV Lipoatrophy 17th Conference on retroviruses and Opportunistic Infections. San Francisco, Abstract # 723. G. McComseyMA O'RiordanJ. ChoiD. LibuttiD. Rowe2010A 48-week Randomized Study of Uridine Supplementation vs Switch to TDF on Limb Fat, Mitochondrial Function, Inflammation, and Bone Mineral Density in HIV Lipoatrophy 17th Conference on retroviruses and Opportunistic Infections.San Francisco, Abstract # 723
  31. 31. Calzolari I, Fumagalli S, Marchionni N, Di Bari M (2009) Polyunsaturated fatty acids and cardiovascular disease. Curr Pharm Des 15: 4094–4102.I. CalzolariS. FumagalliN. MarchionniM. Di Bari2009Polyunsaturated fatty acids and cardiovascular disease.Curr Pharm Des1540944102
  32. 32. Geiger A, Yamasaki S (1956) Cytidine and uridine requirement of the brain. J Neurochem 1: 93–100.A. GeigerS. Yamasaki1956Cytidine and uridine requirement of the brain.J Neurochem193100