In order to bring you the best possible user experience, this site uses Javascript. If you are seeing this message, it is likely that the Javascript option in your browser is disabled. For optimal viewing of this site, please ensure that Javascript is enabled for your browser.
Did you know that your browser is out of date? To get the best experience using our website we recommend that you upgrade to a newer version. Learn more.

We use cookies to optimise the design of this website and make continuous improvement. By continuing your visit, you consent to the use of cookies. Learn more

Renal dysfunction potentiates foam cell formation by repressing ABCA1

Interview with Dr Valentina Kon

Basic Sciences, Pharmacology, Genomics and Cardiovascular Pathology

Q: Dr. Kon, ATVB highlighted September 2009 as the National Cholesterol Awareness Month and in this context published a special work of yours on cholesterol trafficking in renal dysfunction that sheds new light on the increased risk of atherosclerosis in patients with chronic kidney disease. Would you mind recapturing the set-up and key findings of your experimental study?
A: The key finding of our study is that even very mild renal perturbation that results from uninephrectomy (UNx) disrupts macrophage cholesterol homeostasis and potentiates foam cell formation in the apolipoprotein E-deficient mice of atherosclerosis. The underlying mechanisms include impairment in cholesterol efflux through nuclear factor-kappa B (NF-kB)-mediated downregulation of the ATP-binding cassette transporter A1 (ABCA1). Treatment with the angiotensin receptor blocker (ARB), losartan, downregulates NF-kB subunits, upregulates ABCA1, and lessens the cellular cholesterol burden. We therefore conclude that accelerated atherosclerosis observed in the setting of renal damage is, at least in part, due to repression of ABCA1. We further suggest that the beneficial effects of ARBs to lessen this cholesterol handling abnormality may be of particular benefit in decreasing the heightened cardiovascular risk in patients with renal damage.

Q: It might be also worthwhile to briefly reflect on your previous work on this experimental model, which was published in JASN in 2006. In that particular article you showed that uninephrectomized apoE-deficient mice had more extensive atherosclerotic changes of the aortic wall including more extensive macrophage accumulation. Did this prior work of yours also show more extensive lipid accumulation in the vascular wall and may we therefore understand your current work therefore as an extension?

A: Our previous study showed increased lipid deposition and macrophage content in the vessel wall of UNx, suggesting that macrophages play a central role in renal-damage acceleration of atherosclerosis. The current studies were designed to define the underlying mechanisms of disturbed macrophage function. The current results represent a new breakthrough in the understanding of cellular mechanisms of augmented atherosclerotic lesions in the setting of renal impairment showing deranged macrophage lipid homeostasis, a key modulator of atherosclerosis.

Q:  With these introductory explanations, some readers might wonder less about the peritoneal macrophages. How representative are these for “plaque macrophages”?
A: Although foam cells within the atherosclerotic plaque constitute the hallmark of atherosclerotic lesion, these cells are relatively inaccessible to functional assessments such as cellular lipid homeostasis. Such functional studies are usually performed in vitro using established culture cells lines (THP1, J774) or mouse peritoneal macrophages that are exposed ex vivo to metabolites of interest. However, unlike other atherosclerosis-inducing conditions such as diabetes, where specific metabolites (i.e., glucose or advanced glycation end products) can be added to the culture system and asses the impact on macrophage lipid homeostasis, renal impairment does not have a specific metabolite that lends itself to such in vitro studies. We therefore used an in vivo model for foam cell formation in peritoneal macrophages in UNx mice fed a high fat Western diet. The cells were studied upon harvesting and not cultured or passaged. This methodology has been utilized to determine in vivo macrophage lipid homeostasis in LDLR-deficient mice and most recently applied to in vivo studies of macrophage lipid metabolism in a model of type-2 diabetes. While these macrophages are not from the plaque itself, their ability to traffic lipids should be similar to all macrophages exposed to the in vivo environment of reduced renal parenchyma.

Q:  It is interesting to notice that cholesterol serum concentrations increase in the uninephrectomized apoE-deficient mice. What is your explanation?

A: Removal of renal parenchyma in apoE-deficient mice is associated with variable degrees of hypercholesterolemia, reported to reflect increased VLDL and IDL/LDL cholesterol. Although metabolism of serum lipoproteins has not been specifically studied in this renal injury model, dyslipidemia in humans with CKD results both from overproduction and impaired clearance of lipoproteins and reflects a host of abnormalities in lipid substrates, enzymes, lipid transfer proteins and lipoprotein receptor activity as well as abnormalities in the lipoprotein composition.  Importantly, both the current and previous JASN studies emphasize that the level of serum cholesterol does not parallel the extent of foam cell formation. This also holds true in humans with renal impairment, especially as the renal damage becomes more severe. This dissociation between serum lipids and vascular lesions also pertain to effects of ARB. ARB significantly lessened atherosclerosis despite increasing serum cholesterol in UNx, while treatment with a non-specific vasodilator, hydralazine, which decreased the serum cholesterol, did not lessen atherosclerotic lesions. 

Q: The central focus though remains the cellular cholesterol content. Did you observe a close correlation between cholesterol serum concentrations and intracellular cholesterol content in your animals?

A: One of the interesting findings in our studies is the non-quantitative linearity between the levels of serum and cellular cholesterol. This observation emphasizes that local dynamics within the vascular wall particularly altered macrophage lipid handling, are the driving force of lesion formation.  Thus, while high fat diet feeding of UNx increased serum cholesterol by 32%; the cellular cholesterol content was 250% higher than in mice with intact kidneys. Further, while UNx mice fed normal chow tended to have higher serum cholesterol than mice with intact kidneys, the cellular cholesterol was more than 300% higher than the cholesterol content in macrophages from mice with intact kidneys. As discussed above, in vivo treatment with ARB dramatically lessened the cellular cholesterol in UNx, despite increasing serum cholesterol. Thus, while hyperlipidemia is well established as a primary driver for experimental and human atherogenesis, increased serum cholesterol per se may not be the best marker of atherosclerotic disease. There is increased recognition that not everyone with hyperlipidemia develops atherosclerotic cardiovascular disease and that CVD can occur without hyperlipidemia, an observation especially pertinent in individuals with renal damage. It is also now established that treatment with statins (atorvastatin, rosuvastatin) and achievement of very low LDL goals in the setting of end-stage renal disease does not protect against heart attacks or stroke.

Q: You found a marked increase in macrophage cholesterol content in the uninephrectomized apoE-deficient mice, which is due to alternations in cholesterol efflux rather than uptake, e.g. by alteration of scavenger receptor dynamics - is this correct? 

A: We did not specifically evaluate uptake of cholesterol by macrophages from UNx mice. However, foam cell formation is critically dependent on efflux, even in the face of increased uptake. Therefore, our data support a key role of blunted cholesterol efflux in the dramatic expansion of macrophage lipid content in UNx mice.  Nonetheless, previous reports have shown increased expression of the CD36 scavenger receptor in monocytes of patients on dialysis. It is therefore possible that renal damage induced perturbation in efflux is compounded by enhanced cholesterol uptake. Functional studies to examine this hypothesis are currently underway in our laboratory.

Q: ATP-binding cassette transporter A1 seems to be central for the reduced cholesterol efflux in macrophages of the uninephrectomized apoE-deficient mice. Did this match up your initial hypothesis or was this an unexpected finding as you envisioned other pathways?  

A: The first observation in these studies of increased lipid content in macrophages harvested from mice with reduced renal parenchymal mass implied an imbalance between uptake and an inadequate efflux of cholesterol. Our functional studies confirmed blunted efflux in cells of UNx mice. We postulated that ABCA1 and ABCG1, which are the primary regulators of cholesterol efflux in macrophages, were altered. In addition we also examined  SR-B1, which has also been shown to be capable of driving cholesterol from cells. We found significant alteration only in the ABCA1 protein, although ABCG1 tended to be decreased in cells of UNx mice.

Q: Ultimately, you identified activation of nuclear factor kappa B (NF-kB) as the regulating molecular pathway for the study findings. This transcription factor regulates the expression of a number of factors. How important is ATP-binding cassette transporter A1 for the atherosclerotic disease process relative to these other factors?

A: Nuclear factor kappaB transcription factors are major regulators of inflammatory responses and immune responses. Our study shows that macrophages from UNx have increased NF-kB activity together with supressed ABCA1 and that specific antagonism of the NF-kB activation pathways in these macrophages lessens the repression of ABCA1. ABCA1 is a major regulator of lipid homeostasis in macrophages. Therefore, downregulation of ABCA1 by NF-kB in renal dysfunction is expected to have major implications for foam cell formation and atherogenesis. Importantly, however, in addition to modulating cholesterol homeostasis, ABCA1 has major anti-inflammatory-functions. Since atherosclerosis is regarded as a chronic inflammatory response to cholesterol deposition in the vascular intima, repressed ABCA1 in macrophages of UNx may be especially detrimental. Indeed, macrophages from ABCA1-deficient mice have increased cholesterol accumulation together with increased signaling of Toll-like receptors enhancing the inflammatory response to LPS and other TLR ligands.

Q: In the current study you pointed out the inhibitory effect of losartan on NFkB. Even more, angiotensin receptor blockade led to improvement of all parameters including the lipid serum concentrations. To which extent does this translate into an attenuation of atherosclerosis and is this independent of any blood pressure-lowering effects?

A: Epidemiological studies have documented and firmly established fewer cardiovascular events and increased survival of individuals with intact kidney function on angiotensin II (AII) antagonists, with benefits that are independent of hemodynamic and lipid-lowering effects.  Our previous study found that losartan, but not a non-specific vasodilator, hydralazine, lessened UNx-induced acceleration in atherosclerosis, decreased the number of new plaques, and diminished the macrophage content within the plaque lesion, although both losartan and hydralazine decreased BP equally. In a separate study we also found that AII inhibition with ARB, but not hydralazine, not only decreased further expansion of advanced lesions but also modified the atherosclerotic plaque toward a more stable lesion, including a greater proportion of collagen, fewer elastin breaks and greater elastin content together with less proteolysis. These beneficial consequences complement the recognized effects of AII antagonism including decreased cellular migration, inflammation and proteolysis that are distinct from any effect on systemic blood pressure. Of note, ARB’s benefits are also not dependent on serum lipid levels. Indeed, the novelty of these findings relate to the increasingly recognized dissociation between serum lipids and atherosclerotic complications. Notably, a recent study of uremic patients found that treatment with cholesterol-lowering statins was not beneficial, instead, the findings emphasize local dynamics within the vascular wall as key mechanism for atherosclerosis such as macrophage lipid homeostasis.

Q: As you stated in the paper, there was no complete normalization of the current study parameters with ARB and hence the thought that other mechanisms might be involved? Would you mind speculating on some mechanisms?

A: There are several possibilities for the observation that losartan-induced increase in macrophage ABCA1 expression was not accompanied by complete normalization of the macrophage cholesterol efflux. It is possible that the dose of losartan given in vivo may have been insufficient to maximally activate ABCA1 either through more complete inhibition of NF-kB or through activation of PPAR. It is also possible that reduction in the renal parenchyma influences cellular cholesterol trafficking in pathways at multiple levels, including some not susceptible to the losartan intervention, (i.e., accumulation of unhydrolyzed cholesteryl esters in late endosomes and lipid droplets.  Indeed, lipid droplet-associated cholesteryl ester hydrolysis has been proposed to be a limiting factor in cholesterol efflux from macrophages.  There are currently no studies that describing whether and how renal dysfunction alters the intracellular mobilization of cholesterol and whether losartan modulates this pathway. 

Q: Considering inhibition of NFkB as a therapeutic goal, one class of drug that comes to mind is the HMG-CoA reductase inhibitors (“statins”). Should they be tested in an experimental model just like yours, possibly even in addition to ARBs?

A: The central finding in our study is that reduction in renal parenchyma represses ABCA1 and decreases cholesterol efflux in macrophages. The expression and function of ABCA1 is regulated by multiple processes including upregulation by cholesterol through the nuclear receptors liver X receptors (LXRa and/or LXRb) and retinoid X receptor (RXR) and downregulation by angiotensin II and inflammatory stimuli through the NF-kB signaling pathway. Thus, interventions that upregulate ABCA1, including stimulation of LXRs and or blunting of NF-kB activity predict a potentially beneficial response in disturbed macrophage lipid homeostasis observed in animal models or in patients with renal dysfunction. Both HMG-CoA reductase inhibitors (statins) and ARBs directly and indirectly modulate these transcription factors.  Their combined use may be especially efficacious in animal models and clinical settings of renal dysfunction. Although more specific modulators of these transcription factors would be most welcome, LXR agonists have so far shown both anti-atherogenic effects and undesirable effects on hepatic lipogenesis. 

Q: For patients with chronic kidney disease (CKD), do the current findings apply to all disease stages? What are the clinical implications – ARBs and possibly statins to every patient regardless of blood pressure and lipid serum concentrations? 

A: First of all we should be clear that our results were obtained in the mouse system and do not directly apply to humans. With that in mind, one of the interesting findings of our studies is that mice with only one kidney, although showing apparently normal renal function and blood pressure, are prone to accelerated atherosclerosis due to disrupted macrophage lipid homeostasis. These results complement observations in humans that even a modest decrease in renal function (GFR <60ml/min/1.72m2) increases the risk of cardiovascular disease.  We speculate that repressed ABCA1 and decreased cholesterol efflux may contribute to cardiovascular risk in such individuals. These patients may therefore benefit from ARB-mediated upregulation in ABCA1 that could unload cholesterol-laden macrophages. The benefits of such intervention do not depend on blood pressure and, similar to current concepts in treatment of proteinuria, could be instituted even in the absence of hypertension. Studies now clearly show safety and efficacy of statin therapy in patients with CKD stage 1-3, and ARB treatment may provide additional benefits. It is more difficult to predict the implications of our findings for individuals with more advanced CKD, in whom the benefit of therapy, including statins, remains more controversial.

Q: What other key research needs do you see with regards to atherosclerotic cardiovascular disease in patients with CKD?

A: First of all we should be clear that our results were obtained in the mouse system and do not directly apply to humans. With that in mind, one of the interesting findings of our studies is that mice with only one kidney, although showing apparently normal renal function and blood pressure, are prone to accelerated atherosclerosis due to disrupted macrophage lipid homeostasis. We are in the early stages of research into the underlying mechanisms of accelerated atherosclerosis and treatment in patients with CKD. Recognition of the magnitude of the problem and the subtlety in renal dysfunction that potentiates the risk of CVD has already brought about calls to address previous limitations such as exclusion of CKD patients from clinical trials, limited assessments of cardiovascular dysfunctions, and restrained treatment of CVD in CKD. Recent studies indicate benefit of established interventions in those with mild to modest renal impairment, whereas individuals with ESRD appear especially resistant to conventional interventions. However, it remains unknown how CKD imparts the heightened risk for lipid deposition in the vascular intima and the specific characteristics of the atherosclerotic plaques in CKD lesions, including the role of calcium, inflammation, apoptosis and necrosis that may render the lesion more vulnerable to rupture. Further understanding of the contribution of systemic and local inflammation to development and progression of vascular lesions in CKD, macrophage trafficking of cholesterol in CKD, and potential reversal of lipid deposition and plaque regression are but some of the areas that need to be explored. 



Q: Thank you so much for this interview, Dr. Kon, and the privilege to discuss your outstanding work in greater detail. In case someone has further discussion points, may we forward them via my E-mail ( ) or yours?

A: I would be happy to answer any questions sent directly to my e-mail:


Journal: Arterioscler Thromb Vasc Biol. 2009 Sep;29(9):1277-82.

The content of this article reflects the personal opinion of the author/s and is not necessarily the official position of the European Society of Cardiology.

Contact us


European Society of Cardiology

Les Templiers
2035 Route des Colles
CS 80179 BIOT

06903Sophia Antipolis, FR