icon-folder.gif   Conference Reports for NATAP  
  6th International Workshop on Adverse Drug Reactions and Lipodystrophy in HIV
Washington, D.C., USA
25-28 October 200
Back grey_arrow_rt.gif
Lipids--the Liver & Heart Disease in HIV
  Report on the 6th International Workshop on Adverse Drug Reactions and Lipodystrophy in HIV
25-28 October 2004, Washington DC, USA
Written for NATAP by Jacqueline Capeau
Medicine Saint Antione, INSERM, Paris, France
Following a report below by Capeau on a presentation at the Workshop you will see her comments noted.
Session on lipid metabolism and liver alterations
The first part of the session examined the role of the transcription factors of the SREBP family in lipodystrophy and lipid alterations. (see Background on SREBPs at end of this report).
As explained by T Osborne (University of California, Irvine), SREBPs are key sensors of nutrient level and metabolism and play an important role in regulation of lipid metabolism. The SREBP transcription factors family consists of 3 proteins. SREBP2 is involved in cholesterol synthesis and activated when the cellular level of cholesterol is low through a sterol-detection mechanism. SREBP-1a and 1c are maturated from the same gene by alternative splicing of the first exon : the 1a isoform is a strong induced of expression of SREBP-target genes while the 1c isoform is a bad one. All SREBPs are expressed as precursors inserted in the endoplasmic reticulum membrane and activated by a double proteolytic step to be released and enter the nucleus where they act as dimers. The SREBP-1 target genes are mainly involved in the lipogenesis pathway leading to the synthesis of fatty acid from glucose or acetyl-CoA at the level of hepatocytes and adipocytes but also in muscle cells. Insulin is able to induce SREBP-1c expression in the different tissues. The group of Osborne is working on the ability of the different dimers (1a/1a, 1a/1c, 1c/1c) to activate target gene transcription. Their results indicate that, in hepatocytes, which have a 10-fold greater SREBP-1c than 1a level, insulin will increase total SREBP-1 activity resulting in increased lipid synthesis and possibly steatosis (see below). By contrast, at the adipocyte level where the 1c to 1a ratio is 3, insulin, by inducing the 1c isoform, will impede 1a activity through dimer formation, resulting in decreased lipid synthesis. These data are interesting since they support a model, consistently observed in different transgenic animal models, where the increase in SREBP-1c protein expression that occurs in response to insulin would increase SREBP activity in the liver, resulting in steatosis (fatty liver) and increased VLDL production, and decrease SREBP activity in adipose tissue, resulting in a phenotype resembling lipodystrophy with insulin resistance and diabetes.
Comments by Capeau: What is the relevance of these findings for ART- related lipodystrophy ?
Some PIs, and particularly ritonavir, have been shown to increase the expression of SREBP-1 in the liver together with increased liver lipid synthesis and steatosis in experimental animal and cell models (Riddle J Biol Chem, 2001, Liang Nat Med 2001, Lennhard ATVB 2000). By contrast, in vitro models generally reported that PIs decreased lipid levels and the expression of SREBP in adipocytes (Dowell J Biol Chem, 2000, Caron Diabetes, 2001, Caron AIDS, 2003). In accordance SREBP-1c expression is markedly decreased in subcutaneous adipose tissue of lipodystrophic patients under HAART (Bastard Lancet 2002, Kannisto AIDS 2003). Therefore, we have to face opposite actions of ART on SREBP in the liver and adipose tissue which could play a role in metabolic alterations and lipodystrophy and the work of Osborne could help us to understand this discrepancy.
The group of J Capeau (M Lemoine, Paris, France) studied the hepatic lesions observed in HIV-infected patients with lipodystrophy and increased transaminases but without other reason for liver alterations. Liver pathology and the expression of the main transcription factors involved in lipid metabolism were evaluated.
In 9 HIV-infected patients compared to 9 non-HIV infected patients with fatty liver disease and to 10 controls, they found that HIV-infected patients present less steatosis but an increased prevalence of fibrosis and inflammation indicating a more severe liver disease. Steatosis in HIV patients was related to an increased liver expression of SREBP-1. Since an effect of some PIs on the hepatic expression of SREBP-1 has been shown in experimental models, SREBP-1 overexpression could result from the treatment with PIs. SREBP-1 overexpression could be involved in steatosis and liver insulin resistance. The pathophysiology of the liver inflammatory lesions does not seem to be directly linked to the transcription factors of the PPAR family and could result from an altered mitochondrial function and cytokine expression.
Note from Jules Levin. The program book also reported: HIV-lipodystrophy and NAFLD (HIV-negative with fatty liver) patients had higher insulin (median 6.9 and 15.7 vs 2.8 mU/l, p<0.05) and lower adiponectin levels (median 5.2 and 6.4 vs 12 ug/ml, p<0.05). Liver histology in HIV-LD patients showed steatohepatitis in 4 out of 9 cases, associated with portal fibrosis in all cases. Hepatic expression of SREBP-1 was higher in HIV-LD patients than in NAFLD or control patients (median 90 vs 50 and 40 104 copies/ug RNA, respectively, p<0.05). HIV-LD patients with steatosis had significantly higher SREBP-1 expression than those without steatosis (median 175 vs 85 104 copies/ug RNA, p<0.05). The authors conclude in ART-related lipodystrophy steatosis is frequent.
JPH van Wiljk (Utrecht, The Netherlands) evaluated the mechanisms of post-prandial fatty acid trapping in 26 patient with HIV-related lipodystrophy compared with 12 non-lipodystrophic patients and to 35 controls. They observe that, after a meal, the amount of FFA released from lipoproteins is increased in patients with lipodystrophy probably due to a defective incorporation into TG in adipocytes. This can result, at the liver level, in increased fatty acid oxydation but, in lipodystrophic subjects, increased FFA would rather result in increased resynthesis of TG and exportation as VLDL thereby increasing the cardio-vascular risk.
Comments by Capeau: This study confirms the decreased clearance of TG-rich lipoproteins at the level of adipose tissue observed in lipodystrophic patients. It suggests that, in addition to decreased hydrolysis of these TG, there could be a decreased entrance and/or reesterification of the hydrolysed TG resulting in post-prandial increased FFA.
The effect of ritonavir was evaluated in an animal mice model that present a metabolism of lipoproteins close to the human one by M den Boer (Leiden, The Netherlands). They observed that RTV but not IDV increased circulating TG levels in 2 weeks. In their model, this increase resulted from decreased lipoproteins clearance at the peripheral level (adipose tissue and muscles) while hepatic production of VLDL was not modified.
Comments by Capeau: This latter point is not in accordance with previous animal studies using different models (see above). From most studies it can be concluded that RTV increases TG levels through both increased production by the liver and decreased clearance.
The metabolic effects on lipid and glucose metabolism of the intermittent ART with and without IL-2 (ACTG A5102) was presented by P Tebas (Pennsylvania, on behalf of several centers from USA). Forty seven patients were randomized to receive (23) or not (24) IL-2 for 18 weeks. After 48 weeks of TI, a significant decrease in lipid parameters (chol, HDL, LDL, TG) was observed but no change in glucose and insulin. Treatment with IL-2 had no effect on these parameters. IT can decrease the cardiovascular risk.
Session on the metabolic syndrome and the cardio-vascular risk
E Bonara (Verona, Italy) reported the result of the longitudinal study of a large sample from the general population of the Bruneck Italian city aged from 40 to 80 years. According to the criteria used for its definition (WHO or NCEP-ATPIII) 34 to 18 % of this population could be diagnosed with a metabolic syndrome since they cluster several metabolic alterations (impaired glucose tolerance, insulin resistance, dyslipidemia, obesity, elevated blood pressure). These patients presented with an increased prevalence of signs of chronic mild inflammation, procoagulant state, increased oxidative stress, endothelial dysfunction and early atherosclerosis. Most of these alterations were associated with insulin resistance and predict future cardio-vascular disease (CAD). In addition liver enzyme alterations predict liver pathology and NASH. In the general population, preventive treatment with thiazolidinediones or metformin are evaluated in patients with a metabolic syndrome and give indirect evidence of decreased insulin resistance and cardiovascular risk.
To further study the potential deleterious impact of PIs on the vascular risk, M Dubé (SS Shankar, Indiana, USA) evaluated the effect of indinavir on the endothelial function in 16 HIV-negative controls treated for 4 weeks with IDV. While IDV did not modify lipid parameters or insulin sensitivity, the treatment resulted in a marked blunting of endothelium-dependant and insulin-dependant vasodilatation. The authors conclude that IDV can induce endothelium-dysfunction independent of an effect of the molecule on insulin sensitivity.
Comments by Capeau: These data suggest that the increased cardio-vascular risk observed in lipodystrophic HIV-infected patients could result from a direct effect of some ART molecules in addition to that resulting from the « metabolic syndrome » and insulin resistance.
In an animal model of atherosclerosis, NM Shildmeyer from the group of DY Hui (Cincinnati) investigated the effect of IDV and RTV on early (foam cell deposition in the arterial wall) and late (smooth muscle cells, SMC, proliferation) events of CAD. While IDV and RTV increased the formation of foam cells, they inhibited SMC migration and proliferation after endothelial injury which are involved in neovascularisation. Therefore, PIs acted on initial phases of CAD through their effect on metabolic parameters, but not on the late phase of SMC activation. They propose that treatment of metabolic disorders could alleviate the accelerated CAD events associated with PI therapy.
Comments by Capeau: However, in patients with lipodystrophy and insulin resistance, an activation of SMC could be present as a result from the insulin resistant state.
In a study evaluating endothelial dysfunction in 23 HIV infected patients treated with pravastatin, PA Sklar (Philadelphia) observed that an active treatment with pravastatin improved cholesterol, LDL and TG values but yielded no significant improvement in endothelial dysfunction nor in CRP values. The authors suggest that the presence of a chronic inflammation may be an ongoing stimulus towards CV risk which precludes the beneficial effect of statins observed in the general population.
Comments by Capeau: In a poster, F Boccara (Paris, France) similarly observed in a cross-sectional study that there was no improvement of the IMT (intima-media thickness) value in dyslipidaemic HIV-infected patients under HAART treated with pravastatin as compared to non-pravastatin treated patients even if the lipid parameters were improved.
--These studies suggest that in patients with lipodystrophy and metabolic alterations, treating lipid alterations is not sufficient to improve the CV risk.--
M Larrousse (Barcelona, Spain) and the Barcelona group have further documented the data previously presented on the increase prevalence of pre-eclampsia in pregnant women under HAART. They observe that these women present an increased insulin resistance and endothelial dysfunction as compared to pregnant women under HAART without pre-eclampsia.
Comments by Capeau: In conclusion, in the HIV population, patients with lipodystrophy and metabolic alterations could probably be considered as presenting caricatured forms of the metabolic syndrome with a high risk for cardio-vascular and liver complications. Even if some studies on these complications have been presented, more works are required in this population to correctly analyze the risk of complications and to try to prevent it. The data presented in this workshop suggest a direct deleterious effect of some ART molecules, in particular certain PIs, mainly through their dysmetabolic action. In addition, the altered lipodystrophic adipose tissue, which is involved in further metabolic alterations and insulin resistance, could play an important role in the CV risk. From a therapeutical point of view, interventions on lipid parameters are probably not sufficient to reduce this risk. Even if TZD have not shown a major impact on the correction of lipodystrophy, these molecules are working at the level of adipose tissue and the vascular wall and are studied in the general population for their ability to reduce insulin resistance and the progression of atherosclerosis and to reverse steatosis. Therefore, it would be important to consider this class of molecules for their potential beneficial impact on insulin resistance, atherosclerosis and NASH in HIV-infected patients with high CV and liver risks.
SREBP-1: A target for insulin resistance

In both the US and the UK the incidence of obesity is increasing by about 1% per year, a tendency related to the rise in other serious co-morbidities. As a general guide, obesity increases the likelihood of death from coronary heart disease by 25%. In addition obese individuals are at twice the risk of developing diabetes. Many people are obese, suffer cardiovascular problems and are also diabetic, leading to the hypothesis that the coexistence of these diseases is not a coincidence, but that a common underlying abnormality allows them to develop. In 1988 it was suggested that co-morbid obesity and diabetes was related to insulin resistance, and hence the insulin resistance syndrome was first described. It is estimated that this syndrome affects 70 to 80 million Americans and is characterized by a failure of insulin to stimulate glucose utilization and uptake into tissues. Considerable attention has been paid to the development of molecules able to reduce insulin resistance. Alterations in the abundance and activity of transcription factors lead to complex dysregulation of gene expression, which might be a key to understand insulin resistance-associated clinical clustering of coronary risk factors at the cellular or gene regulatory level. Recent examples are members of the nuclear hormone receptor superfamily-for example, peroxisome proliferator-activated receptors (PPARs) and sterol regulatory element-binding proteins (SREBPs). Besides their regulation by metabolites and nutrients, these transcription factors are also targets of hormones (like insulin and leptin), growth factors, inflammatory signals, and drugs. MAP kinase cascades represent major signaling pathways that couple transcription factors to extracellular stimuli. German researchers have recently suggested that SREBPs are substrates of MAP kinases and propose that SREBP-1 might play a role in the development of cellular features belonging to lipid toxicity and possibly insulin resistance. Thus, the metabolic syndrome appears to be not only a disease or state of altered glucose tolerance, plasma lipid levels, blood pressure, and body fat distribution, but rather a complex clinical phenomenon of dysregulated gene expression. Targeting insulin resistance with PPAR agonists has already received considerable pharmaceutical attention. The activation of SREBP has however lags behind. The therapeutic potential of such molecules has recently been exploited by GSK researchers who have developed a new class of compounds that directly binds to the SREBP cleavage-activating protein (SCAP). This series is currently being developed as new class of hypolipidemic drugs however the potential of this and similar classes of molecule is expected to eventually cover the treatment of insulin resistance in general. (www.Bioportfolio.com; Ann N Y Acad Sci. 2002 Jun;967:19-27, SREBP-1: gene regulatory key to syndrome X?, Muller-Wieland D, Kotzka J.).