| |
Effect of atazanavir and ritonavir on the differentiation and adipokine secretion of human subcutaneous and omental preadipocytes
|
| |
| |
[BASIC SCIENCE: CONCISE COMMUNICATIONS]
AIDS:Volume 22(11)11 July 2008p 1293-1298
Jones, Simon Pa; Waitt, Catrionaa; Sutton, Robertb; Back, David Ja; Pirmohamed, Munira
From the aDepartment of Pharmacology and Therapeutics, UK
bDepartment of Surgery, The University of Liverpool, Liverpool, UK.
Abstract
Background: Treatment of HIV with some protease inhibitors has been associated with dyslipidemia, insulin resistance and fat redistribution. It has been hypothesized that some protease inhibitors may alter the differentiation of subcutaneous and visceral adipocytes in a disparate manner. The aim of this study was to investigate whether isolated human preadipocytes display regio-specific sensitivity to the effects of ritonavir and atazanavir by examining differentiation, as well as adipokine secretion, following a 10-day drug exposure.
Methods: Paired subcutaneous and omental human preadipocytes (n = 8) were induced to differentiate for 6 days, before being exposed to atazanavir or ritonavir (1-10 μmol/l) for 10 days. Lipid metabolism was assessed by Oil Red O staining and glycerol 3-phosphate dehydrogenase enzyme activity, whereas leptin and adiponectin secretion were assessed by enzyme-linked immunosorbent assay.
Results: There was no difference in differentiation between subcutaneous and omental adipocytes. Repeated exposure to ritonavir, but not to atazanavir, led to significant reductions in adipocyte differentiation. There were no differences in adiponectin secretion for any of the atazanavir treatments in both subcutaneous and omental adipocytes, whereas significant reductions were evident at 10 μmol/l for ritonavir exposed subcutaneous adipocytes. In contrast, both atazanavir and ritonavir were associated with altered leptin secretion.
Conclusions: Ritonavir, but not atazanavir exposure, can inhibit differentiation of subcutaneous and omental adipocytes to a similar extent. Regio-specific differences, however, were apparent for adiponectin and leptin secretion. The role of region-specific alterations in adipokine secretion and apoptosis in the pathogenesis of HIV-lipodystrophy requires further attention.
Introduction
Adipose tissue is well known to play a key pathological role in the development of metabolic disorders such as type 2 diabetes (T2D), dyslipidemia, polycystic ovary syndrome and more recently, HIV-associated lipodystrophy [1,2]. The anatomical location of adipose tissue, rather than fat mass per se, appears to be more predictive of the development and severity of these conditions [3,4]. These relationships have been largely attributed to the heterogeneity of adipose tissue [5], as fat cells isolated from the omental depot differ in size, their response to catecholamine and insulin stimulation, as well as in their susceptibility to apoptosis compared to subcutaneous adipocytes [6-8]. To date, however, very little work has been done to define how regio-specific differences impact on the pathogenesis of HIV-associated lipodystrophy, despite the fact that lipodystrophy is characterised by depletion of the subcutaneous depot and potential enlargement of the omental depot [2,9].
Regio-specific differences in the secretion of adipokines, such as leptin, tumour necrosis factor (TNF-alpha), resistin, interleukin-6 (IL-6) and adiponectin have also been identified [4,10]. Leptin plays a key role in the control of both appetite and energy control, with subcutaneous adipocytes secreting more leptin than omental adipocytes [11]. The same trend is also evident for adiponectin [12], an adipokine that appears to be strongly associated with the metabolic abnormalities of lipodystrophy [13,14]. Protease inhibitors have also been shown to alter adipokine secretion from murine cell lines [15] but it is not known whether protease inhibitors have species-specific or region-specific effects on subcutaneous and visceral fat depots.
To date, only a few studies have examined the effects of protease inhibitors on human adipocytes [16,17]. It has recently been shown that ritonavir (RTV) inhibits the differentiation of subcutaneous adipocytes and modulates adipokine expression [16], which is consistent with its known effects on lipid dysregulation. The effect of atazanavir (ATV), a relatively newer compound, which in vivo does not appear to be associated with lipid or glucose abnormalities [18], however currently remains unknown. Similarly, there have been no studies comparing the effects of any protease inhibitors on human adipocytes from different regions, namely the subcutaneous and omental depots. Theoretically, protease inhibitors could have disparate effects on these two depots, which could explain both the depletion of fat and potential enlargement of adipose tissue, typical manifestations of lipodystrophy. The mechanisms for this potential disparate effect remain unknown, although factors such as protein binding, blood flow and drug transports may all play a role in certain protease inhibitors preferentially changing the morphology of a given fat depot.
The aim of this study was to investigate whether isolated human preadipocytes display regio-specific sensitivity to the effects of the protease inhibitors, RTV and ATV, by examining differentiation as well as adipokine secretion following treatment for 10 days.
Discussion
In this study, we report that RTV, but not ATV, inhibits the differentiation of both subcutaneous and omental preadipocytes. The contrasting effects of the two protease inhibitors support recent in vitro data from murine 3T3-L1 preadipocyte line [21], as well as clinical studies, which have demonstrated that ATV has very little effect on lipid levels after 96 weeks of treatment [18]. Morphologically, cells exposed to RTV (5 and 10 μmol/l) appeared to undergo de-differentiation (data not shown) with an increased number of smaller, less lipid-laden adipocytes, consistent with previous studies in a murine cell line [22,23], as well as the morphological appearances of biopsy samples taken from patients with lipodystrophy [24].
No regio-specific differences in inhibition of differentiation were apparent between the two depots for any of the drug treatments. Whether antiretrovirals preferentially cause a loss of subcutaneous fat rather than omental fat is controversial [9]. A recent imaging study utilising both MRI and CT scans showed that both fat depots decreased in patients with lipodystrophy [9]. However, other investigations have suggested that the subcutaneous depot is preferentially depleted, whereas the omental depot remains largely unaffected [25]. While our results may suggest that both depots appear susceptible to de-differentiation, it is important to note that ex vivo, access of drugs to the omental preadipocytes is likely to be easier than witnessed in vivo, where factors such as protein binding, blood flow and drug transporters may play a key role in drug distribution.
An increasing number of studies have shown that the secretion and mRNA expression of various adipokines is altered in patients with lipodystrophy [14,26-29]. In our study, adiponectin secretion was significantly higher in subcutaneous than in omental preadipocytes, irrespective of drug exposure, supporting previous work in non-HIV infected obese patients [30]. Vernochet et al. [16] using differentiating subcutaneous preadipocytes reported that RTV, but not lopinavir or amprenavir, decreased adiponectin mRNA expression. More recently, Kim et al. [31] reported a differential effect of RTV, but not ATV on mature adipocyte cell viability, which appeared to be associated with reductions in adiponectin and increased IL-6 levels. Comparing RTV and ATV in this study indicated that the former but not the latter was associated with reduced adiponectin levels, supporting the earlier findings of Kim et al. [31]. Leptin was also found to be secreted to a greater extent in subcutaneous than in omental adipocytes, confirming earlier findings from both mature [32] and differentiating human preadipocytes [33]. Moreover, both ATV and RTV were found to alter leptin secretion, with a more marked effect evident in subcutaneous adipocytes, concentrations up to 10 μmol/l increasing leptin secretion. When examined in light of data suggesting that RTV, but not ATV, was influencing adipocyte differentiation, this concentration-related difference in leptin secretion is interesting. Potentially, if only RTV was found to increase leptin secretion, it could be suggested that this occurred as a result of a loss of cell viability leading to increased secretion from the damaged adipocyte. However, as both drugs were found to affect leptin secretion, an alternative aetiology is likely. The role of leptin in HIV-associated lipodystrophy remains complex in vivo [27,34], with discordant results being reported for both its secretion, as well as its relationship with fat redistribution and insulin sensitivity [35]. How certain protease inhibitors alter adipokine levels [36] as well as the physiological significance of such changes at a regio-specific level remains unknown [6]. It has been suggested that adiponectin may simply reflect the absolute amount of peripheral fat, as has been suggested previously for leptin [35,37]. This concept is supported by two studies indicating that in patients with HIV-associated lipodystrophy, peripheral subcutaneous rather than omental visceral fat acts as an independent contributor to adiponectin secretion and mRNA expression [14,28].
In conclusion, our data indicate that therapeutic concentrations of RTV but not ATV can inhibit differentiation to a similar extent in both subcutaneous and omental adipocytes. Regio-specific differences however were apparent for adiponectin and leptin secretion. The role of regio-specific alterations in adipokine secretion and apoptosis in the pathogenesis of HIV-lipodystrophy requires further attention.
Methods
Patients
Omental and subcutaneous adipose tissue samples were obtained from patients undergoing elective open abdominal surgery. Participants were eight men, presumably HIV-seronegative, with a mean age of 56 ± 5.6 years (mean ± SD; range 40-65 years), and a mean weight of 78 ± 5.2 kg (BMI 23 ± 1 kg/m2). All patients underwent general anaesthesia and had fasted for at least 12 h prior to surgery. No patients were taking diabetic medications or drugs known to impact on lipolysis or adiponectin secretion. The study was approved by the Liverpool Local Research Ethics Committee and all patients gave written informed consent.
Preadipocyte isolation
The isolation and differentiation of preadipocytes was performed according to the method described by Hauner et al. [19]. The adipose tissue was placed in saline and rapidly transported to the laboratory where it was dissected from skin and blood vessels and finely diced in a laminar air-flow cabinet. Following three washes with saline, the adipose tissue was digested with a freshly prepared collagenase solution (2 mg/ml, pH 7.4), containing 4% dialysed bovine serum albumin (BSA, Fraction V). Following an approximate 30 min incubation in a shaking water bath at 37°C, the disrupted tissue was filtered through a nylon filter to separate the adipose tissue from the connective fibrous tissue. The remaining cell pellet was resuspended in 10 ml of erythrocyte lysis buffer (0.154 mol/l NH4Cl, 5.7 mM K2HPO4, 0.1 mM EDTA, pH 7.3) prior to being resuspended in 10ml of DMEM/F12 medium (Invitrogen, Paisley, Scotland). After an additional centrifugation step, the preadipocyte cell fraction was resuspended in 1 ml of DMEM/F12 medium containing 10% nonheat inactivated fetal calf serum and 100 units of penicillin/streptomycin. Following 18 h of attachment, cells were washed twice with DMEM/F12 medium and re-fed a chemically defined serum-free medium (DMEM/F12) [20]. Following 6 days of culture, the cells were treated with RTV and ATV (1, 5 or 10 μmol/l) or vehicle control [methanol (0.1%)]. The media was replaced every 2-3 days until the end of the differentiation programme (day 16).
Assessment of differentiation
Differentiation was assessed on day 16 using glycerol 3-phosphate dehydrogenase (GPDH) enzyme activity as described previously [17]. In addition, the effect of the drugs on intracellular lipid accumulation was also assessed using Oil Red O staining [17].
Adipokine secretion by enzyme-linked immunosorbent assay
Leptin and adiponectin concentrations were determined in the cell culture media by enzyme-linked immunosorbent assay (ELISA) following the manufacturer's standardized protocol (R&D Systems, UK). Media was removed from cells on day 16 of adipokine differentiation (day 10 of drug exposure) and frozen at -80°C until it was processed in duplicate.
Statistical analysis
Data between treatments were examined using one-way ANOVA with Bonferoni correction on normally distributed data. Differences between vehicle controls for the subcutaneous and omental fat depots were analysed using a paired T-test (SPSS, Illinois, USA). Differences were considered significant when P < 0.05.
Results
Differentiation was quantified using the eluted Oil Red O stain (Fig. 1a). Triglyceride accumulation was not significantly different between the two fat depots (subcutaneous vs. omental: 0.10 ± 0.017 vs. 0.11 ± 0.01 optical density (OD) units). RTV (10 μmol/l) significantly decreased triglyceride accumulation in subcutaneous and omental adipocytes by 26 and 28%, respectively, whereas ATV was not associated with a decrease in either fat depot. Differentiation was also quantified by GPDH enzyme activity, which revealed a significant reduction with RTV. At a concentration of 5 and 10 μmol/l respectively (Fig. 1b), RTV inhibited GPDH activity in subcutaneous cells by 40 and 45%, respectively (P < 0.05), when compared with vehicle control. A similar effect was also obtained in the omental cells with RTV. ATV had no effect on GPDH activity in both depots, consistent with the results of the Oil Red O staining (Fig. 1a).
Adipokine secretion
Adiponectin secretion was significantly higher in subcutaneous than in omental adipocytes during differentiation. Specifically, in vehicle-treated control subcutaneous cells, adiponectin was 32% higher than in omental cells (P < 0.001), a trend that was apparent for all treatments (Fig. 2a). No differences in secretion were evident at any concentration of ATV in either subcutaneous or omental adipocytes. In contrast, RTV (10 μmol/l) led to a reduction in adiponectin secretion in subcutaneous adipocytes. There was no effect on adiponectin secretion for omental adipocytes following exposure to RTV. Leptin secretion was significantly higher in the vehicle control subcutaneous adipocytes than in the omental vehicle control (P < 0.01) (Fig. 2b) by 40%. Both RTV and ATV increased leptin secretion in subcutaneous cells. In omental cells, however, while RTV led to small decreases in leptin, ATV had little effect, apart from at 10 μmol/l where a significant increase was observed compared with vehicle control.
|
|
| |
| |
|
|
|
