Stressing the endothelium to assess localized inflammatory potential and the risk for atherosclerotic cardiovascular disease

Despite the well-established fact that inflammation is intrinsically linked to atherosclerosis, interrogating the effects of inflammation is likely more complex than the simple conclusion that more inflammation leads to more disease. Three specific concepts appear to be relevant to better understanding the role of inflammation in vascular disease: 1) timing of the inflammatory cascade; 2) tissue-specific impact of inflammation; and 3) bidirectional effects of lipids and proteins and the inflammatory cascade. Acute versus chronic inflammation can differentially affect the architecture and functionality of lipids and proteins as well as the magnitude of their effects on cardiometabolic phenotypes. Thus, understanding the role of lipids and proteins in inflammation requires ascertainment of timing of the inflammatory cascade to quantify these relationships more precisely.

In addition to the differential effects of timing, inflammation imparts heterogeneous effects on various tissues relevant to ASCVD. While hepatocytes may be the driver of global markers and mediators of inflammation (Interleukin-1beta, Interleukin-6, and C-reactive protein [CRP]), endothelial cells, macrophages, and adipocytes manifest tissue-specific effects of inflammation and, in concert, promote the atherosclerotic process. Measuring these tissue-specific effects of inflammation may provide a more precise assessment of the role of inflammation in ASCVD than non-specific global markers such as C-reactive protein. Lastly, while inflammation can certainly promote changes in lipids and proteins, rendering them “pro-inflammatory,” these molecular entities are intrinsically involved in promoting and preventing the initiation, maintenance, and resolution of inflammation as well. Understanding the bidirectional relationships between specific molecular species and the inflammatory cascade with respect to timing and tissue may better provide a more complete picture of the role of inflammation in ASCVD.

Considering these main points, focusing on high-density lipoproteins (HDLs) could reveal important insights into the complex role of inflammation regarding timing of the inflammatory cascade, tissue-specific impact, and bidirectional effects on ASCVD and other diseases.¹ HDLs are composed of more than 200 proteins as well as a multitude of phospholipids which differentiate HDLs from all other lipoproteins. This distinctive molecular signature suggests complex and dynamic, context-dependent functionality. Indeed, HDLs exert not only cholesterol transport functions but also anti-oxidative, anti-inflammatory, and anti-apoptotic functions, among others.^{1, 2} With respect to inflammation, HDLs can exert both anti-inflammatory and pro-inflammatory functions depending on the timing of the inflammatory cascade.¹ For example, in sepsis, HDLs become pro-inflammatory and facilitate clearance of pathogens,³ whereas in healthy states, HDLs serve to counter-regulate pro-inflammatory pathways. HDLs exert these effects on the inflammatory cascade on all relevant tissue types related to ASCVD, including endothelium, macrophages, and adipocytes. Specifically, HDLs interact with these different cell types via cell-surface transporters to accept and deliver cholesterol, phospholipids, and non-coding RNAs. However, the specificity of these interactions regarding specific transporters (ABCA1, ABCG1, SR-B1) varies by tissue type. Furthermore, beyond exchange of cargo, HDLs exert direct actions in these tissues such as activation of Toll-like receptors and inflammasomes. Lastly, the inflammatory cascade itself promotes a remodeling of HDL particles with a change in size distribution and gain as well as loss of certain proteins and phospholipids such as serum amyloid albumin (gain) and spingosine-1-phophate (loss).^{4, 5}

The study by Jia, Tietge and colleagues in this issue of Circulation⁶ provides important insights into the role of HDL, inflammation, and ASCVD in a large human cohort with specificity related to the above-mentioned concepts. They found that HDL anti-inflammatory capacity measured from baseline samples using a novel cell-based assay described below, was inversely associated with the odds of incident ASCVD, independent of traditional risk factors and HDL-C levels in 340 cases matched to 340 controls (OR per 1SD: 0.74, 95%CI 0.61-0.90). In this study, a nested case-control study design was undertaken in the PREVEND cohort among individuals free of existing CVD, diabetes, or other inflammatory illnesses such as recent or chronic infection, auto-immune disease, or cancer. Given this construct, the relationship between HDL anti-inflammatory capacity and incident ASCVD was specifically in the context of stable health and not overt acute or chronic inflammation. This is in contrast to some other studies assessing HDL function and outcomes in the setting of an acute myocardial infarction or other inflammatory illness such as psoriasis or sepsis, contexts within which the inflammation and disease process may have rendered HDL dysfunctional already. Thus, the interpretation of this study is that in the absence of self-reported acute or chronic inflammation, impaired HDL anti-inflammatory capacity is linked to incident ASCVD.

With regard to tissue-specific effects, this study assessed the ability of human umbilical venous endothelium to secret vascular adhesion molecule-1 (VCAM-1) in response to the pro-inflammatory stimulus tumor necrosis factor alpha (TNF-alpha). Endothelial activation is directly related to the initiation and promotion of a localized inflammatory cascade within the intima and subsequent atherosclerotic plaque development. VCAM-1 is one of several key adhesion molecules activated and shed into the circulation from the endothelium.⁷ HDLs are known to directly interact with the endothelium via ABCA1, ABCG1, and SR-BI, contributing to production of vasodilatory nitric oxide, cholesterol uptake and delivery, and delivery of non-coding RNAs.^8-10 The assay used in this study assessed the ability of HDL from frozen plasma of study participants to blunt the inflammatory response of standardized endothelial cells collected from blood donors. An impaired ability of a person’s HDL to blunt this endothelial inflammatory response was linked to increased risk of incident ASCVD.

This is the first report of an HDL anti-inflammatory function linked to incident ASCVD events in a large cohort. What does this marker represent? Intriguingly, no traditional risk factors, standard lipids, or other common circulating markers of atherosclerotic risk were correlated with HDL anti-inflammatory capacity. In addition, there was essentially no correlation with hsCRP. While hsCRP reflects systemic inflammation, this assay likely represents localized capacity for inflammation in endothelial cells, which may or may not reflect inflammatory potential at other tissue sites or more globally. Lastly, the only other HDL function that has been linked to incident ASCVD is cholesterol efflux capacity (CEC),¹ which is a similar cell-based assay assessing the first key step of reverse cholesterol transport from standardized macrophages to circulating acceptors from participant’s plasma or serum. CEC had been previously measured in the same PREVEND cohort and linked to incident ASCVD.¹¹ In the current study, the authors included both HDL functional markers and found that there was essentially no correlation between CEC and HDL anti-inflammatory capacity. Both remained significantly associated with incident ASCVD even when adjusted for the other, and neither attenuated the other’s effect estimates. Taken together, HDL anti-inflammatory capacity may be a truly unique cardiometabolic marker in that it appears to reflect a pathway completely independent of multiple established cardiometabolic risk factors, global inflammation, and reverse cholesterol transport. Few cardiovascular markers if any have displayed such characteristics.

What are the implications of this study? First, the findings require replication and validation in other cohorts. Notably, the PREVEND cohort was 100% White European and 70% men with an average LDL-C of 150 mg/dL. The cohort required the presence of urinary microalbumin but excluded prevalent diabetes and CVD. Future studies should test this marker in more diverse cohorts with more women, lower atherogenic lipid levels, and in those with diabetes. CEC has been shown to associate with coronary but not cerebrovascular atherosclerosis.¹² Therefore, it will be important to demonstrate whether HDL anti-inflammatory capacity is also associated with ischemic stroke and peripheral arterial disease. Lastly, HDL metabolism appears to be relevant to several non-CV disease pathways as well,¹³ including cancer, renal disease, diabetes, auto-immune disease, adverse pregnancy disorders,¹⁴ and infection. It will be of interest to determine the role of this novel marker in these non-CV pathways.

While cell-based assays will unlikely develop into clinical tests, they hold great translational potential into elucidating whether tissue-specific pathways are clinically relevant. Specifically, cell-based assays hold the potential to reveal druggable targets with precise tissue-specific mechanisms of action and limited to no systemic effects. This approach is warranted given that therapies targeting global inflammation, such as canakinumab, may result in increased serious infectious risk and toxicity (CANTOS: NCT01327846).¹⁵ Lastly, deep -omics investigations that identify the molecular species driving variation in this novel trait may lead to clinically relevant biomarkers for improved risk prediction and ability to modulate HDL’s anti-inflammatory potential to enhance both primary and secondary preventive efforts.