Current and future therapies for addressing the effects of inflammation on HDL cholesterol metabolism

Abstract

Cardiovascular disease (CVD) is a major cause of morbidity and mortality worldwide. Inflammatory processes arising from metabolic abnormalities are known to precipitate the development of CVD. Several metabolic and inflammatory markers have been proposed for predicting the progression of CVD, including high density lipoprotein cholesterol (HDL‐C). For ~50 years, HDL‐C has been considered as the atheroprotective ‘good’ cholesterol because of its strong inverse association with the progression of CVD. Thus, interventions to increase the concentration of HDL‐C have been successfully tested in animals; however, clinical trials were unable to confirm the cardiovascular benefits of pharmaceutical interventions aimed at increasing HDL‐C levels. Based on these data, the significance of HDL‐C in the prevention of CVD has been called into question. Fundamental in vitro and animal studies suggest that HDL‐C functionality, rather than HDL‐C concentration, is important for the CVD‐preventive qualities of HDL‐C. Our current review of the literature positively demonstrates the negative impact of systemic and tissue (i.e. adipose tissue) inflammation in the healthy metabolism and function of HDL‐C. Our survey indicates that HDL‐C may be a good marker of adipose tissue health, independently of its atheroprotective associations. We summarize the current findings on the use of anti‐inflammatory drugs to either prevent HDL‐C clearance or improve the function and production of HDL‐C particles. It is evident that the therapeutic agents currently available may not provide the optimal strategy for altering HDL‐C metabolism and function, and thus, further research is required to supplement this mechanistic approach for preventing the progression of CVD.

Linked Articles

This article is part of a themed section on Targeting Inflammation to Reduce Cardiovascular Disease Risk. To view the other articles in this section visit http://onlinelibrary.wiley.com/doi/10.1111/bph.v174.22/issuetoc and http://onlinelibrary.wiley.com/doi/10.1111/bcp.v82.4/issuetoc

Abbreviations

Apo‐A1

apolipoprotein A1

Apo‐A2

apolipoprotein A2

AT

adipose tissue

CETP

cholesterol ester transfer protein

CRP

C‐reactive protein

CVD

cardiovascular disease

FA

fatty acid

LCAT

lecithin cholesterol acyltransferase

LPL

lipoprotein lipase

PLTP

phospholipid transfer protein

RA

Rheumatoid arthritis

RCT

reverse cholesterol transport

sC5b‐9

serum complement membrane attack complex generated by the assembly of C5 through C9 complements

SR‐B1

scavenger receptor class B type 1

SREBP‐1c

sterol regulatory element binding protein‐1c

TG

triglyceride

VLDL‐TG

very LDL triglyceride

Tables of Links

TARGETS
Other protein targets a	Enzymes e
IL‐1β	5‐LOX
TNF‐α	Caspase 1
Nuclear hormone receptors b	Cathepsin B
PPAR‐α	COX‐1
PPAR‐β/δ	COX‐2
PPAR‐γ	JNK
LXR‐α	MMP3
LXR‐β	MMP7
RAR‐α	MMP9
Catalytic receptors c	MMP12
NLRP3	MPO
Transporters d	P38 MAPK
ABCA1	PCSK9
ABCG1	sPLA2
ABCB11	tPA

LIGANDS
Adiponectin	Methotrexate
Aspirin	Metformin
C3	MCP‐1 (CCL2)
Canakinumab	Montelukast
Etanercept	Niacin
Fibrinogen	Pioglitazone
FLAP	Rosiglitazone
ICAM‐1	SAA (serum amyloid A)
IFN‐γ	Sildenafil
IL‐6	Tocilizumab
IL‐10	Tofacitinib
IL‐18	Theophylline
Leptin	Thromoboxane A2
LTB4	VCAM‐1

These Tables list key protein targets and ligands in this article which are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Southan et al., 2016), and are permanently archived in the Concise Guide to PHARMACOLOGY 2015/16 (^a,b,c,d,eAlexander et al., 2015a,b,c,d,e).

Introduction

There have been several sharp turns in the evidence trail marking the atheroprotective role of high density lipoproteins (HDLs). Very early studies on dyslipidaemia highlighted the positive correlation between total triglycerides (TGs) with the risk of developing cardiovascular disease (CVD). Since the mid‐1970s, numerous epidemiological and animal studies, including the Framingham Study (Gordon et al., 1977), reported that HDL cholesterol (HDL‐C) has the strongest inverse relationship with the development of CVD of the known serum lipid factors (Badimon et al., 1990; Rubin et al., 1991; Liu et al., 1994; Plump et al., 1994). This association is underpinned by reverse cholesterol transport (RCT) – a process by which HDL‐C transfers the cholesterol from peripheral cells, for example, lipid‐laden foam cells, to the liver for secretion into bile and faeces. The promotion of RCT is considered a major anti‐atherogenic function of HDL‐C (Gofman et al., 1966; Miller and Miller, 1975; Rhoads et al., 1976). Based on these findings, interventions to increase the levels of HDL‐C were developed but found to be ineffective for preventing cardiovascular outcomes in several large clinical trials (Brousseau et al., 2004; McKenney et al., 2006; Barter et al., 2007; Boden et al., 2011; Lüscher et al., 2012; Schwartz et al., 2012). Given these results, the importance of HDL‐C for preventing CVD has been questioned and revisited. An expert panel from the National Lipid Association concluded that although ‘HDL‐C is not a therapeutic target at the present time’, ‘rigorous research into the biology and clinical significance of low HDL‐C should continue’ and that ‘the development of novel drugs designed to modulate the serum levels and functionality of HDL particles should also continue’ (Toth et al., 2013), a recommendation which has been echoed by other experts in the field (Brown et al., 2014; Toth et al., 2014). In our opinion, the key issue in this matter is the functionality of HDL‐C, which can be affected by adipose tissue (AT) and low‐grade systemic inflammation (Brewer, 2007; Rader and Daugherty, 2008; Rosenson, 2010; Zhang et al., 2010; Chung et al., 2011). Here, we discuss the data available illustrating the effect of inflammation on the functionality of HDL particles and potential therapeutic interventions that can help reverse these effects and, thus, prevent the development of metabolic abnormalities leading to the progression of CVD and associated co‐morbidities.

Mechanism of the synthesis of mature or functional HDL‐C

Currently, the most important known function of HDL‐C is to provide the successful transfer of cholesterol from peripheral tissues to the liver for extraction (i.e. RCT). The HDL‐C particles that can effectively accomplish this task are the functionally mature ones, which are rich in apolipoprotein A1 (Apo‐A1) and cholesterol (Rader and Daugherty, 2008). Several apolipoproteins, enzymes and transfer proteins participate in formation and function of these mature HDL‐C particles. The first step in the formation of HDL‐C requires Apo‐A1 and ATP‐binding cassette transporter A1 (ABCA1) (Lee and Parks, 2005; Zannis et al., 2006). ABCA1 mediates the efflux of phospholipids and free cholesterol from AT to Apo‐A1, a step that is necessary for the initial lipidation of Apo‐A1 and formation of nascent HDL‐C particles (Verghese et al., 2007; Phillips, 2014). The next step is the maturation of HDL‐C particles, which involves several enzymes – lipoprotein lipase (LPL), phospholipid transfer protein (PLTP) and lecithin cholesterol acyltransferase (LCAT). LPL hydrolyzes circulating very LDL triglyceride (VLDL‐TGs), while PLTP transfers phospholipids and free cholesterol from the surface of VLDL‐TG to HDL‐C (Tall et al., 1985; Rinninger et al., 1998; 2001; Ji et al., 2014). Thereafter, LCAT esterifies cholesterol, rendering it more hydrophobic and amenable for efficient packaging and transport by HDL‐C to the liver (Rader, 2009; Dobiásová and Frohlich, 1999; Asztalos et al., 2007). Finally, the transporter ABCG1 mediates cholesterol efflux from the surface of cells and macrophages to mature HDL‐C particles (Kennedy et al., 2005). The mature particles are subject to cholesterol ester transfer protein (CETP)‐mediated exchange of cholesteryl esters with TGs from VLDL or LDL, which subsequently binds to LDL receptors in the liver (Bruce et al., 1998). Successful completion of this exchange process and binding of HDL‐C to scavenger receptor class B type I (SR‐B1) receptors in the liver allows for elimination of cholesterol in the liver, thereby preventing the deposition of cholesterol in the endothelium and the development of atherosclerosis.

In addition to playing a major role in RCT, the HDL‐C particles have been shown to (a) have anti‐inflammatory, anti‐oxidative and anti‐apoptotic properties; (b) contribute to innate immunity, the modulation of glucose metabolism and platelet function; and (c) influence stem cells and embryogenesis (Gordon et al., 2011). The changes in the functionality of HDL‐C particles are discussed throughout this review article; however, our review is mainly focused on the effect of inflammatory processes on the functionality and atheroprotective properties of HDL‐C particles. The functional diversity of HDL‐C particles is related to their compositional complexity and heterogeneity. As an example, mature cholesterol and Apo‐A1‐rich HDL‐C particles have been shown to be successful at RCT, while the smaller, cholesterol‐poor, TG‐ and Apo‐A2‐rich HDL‐C particles degrade easily and are unable to contribute to RCT. The published data suggest that the differences in HDL‐C functionality depend on the composition of the HDL‐C particles (Asztalos et al., 2011).

Several assays have been proposed to assess the functionality of HDL‐C. Some assays are designed to measure the anti‐inflammatory and anti‐oxidative properties of HDL‐C, while others evaluate HDL‐C RCT efflux (Navab et al., 1991; 2001; Zhang et al., 2003; Annema et al., 2010; Suzuki et al., 2010; Khera et al., 2011). Furthermore, electrophoretic and NMR methods have been developed to estimate HDL‐C particle size and composition. Thus far, there has been no consensus regarding the superiority of one method versus another for HDL‐C characterization, and attempts to standardize the various nomenclature systems of HDL‐C are a work in progress (Asztalos et al., 2011). Further research is needed to elucidate the relationship between HDL‐C particle heterogeneity and function (Gordon et al., 2011).

Effect of inflammatory processes on HDL‐C metabolism

Several factors and conditions, including genetic (i.e. familial disorders) and acquired (e.g. decreased cholesterol efflux, inflammation, hypertriglyceridaemia and AT dysfunction), affect the concentration and functionality of HDL‐C. Here, we will focus on mechanisms related to inflammation, which have been depicted in detail in Figure 1.

Figure 1.

Schematic presentation of the mechanisms causing decreased HDL‐C concentration or function due to inflammation. (A) The majority of mechanisms that affect HDL‐C metabolism are associated with AT inflammation and function. Increased levels of FFA and VLDL‐TG can enhance the activity of CETP resulting in TG‐enriched HDL‐C particles. These particles are prone to higher liver clearance rates. Decreased adiponectin levels affect the expression of the Apo‐A1 gene. Inflammatory markers have been shown to affect LPL and ABCA1 gene expression impeding maturation of HDL‐C. Other mechanisms which affect HDL‐C concentration include (a) enhanced sPLA₂ activity causing increased hydrolysis of HDL‐C; (b) increased secretion of complement sC5b‐9 from liver; and (c) decreased expression of RAR‐α in macrophages. The later mechanism also affects HDL‐C function and results in (a) increased production of SAA in AT which displaces Apo‐A1 from HDL‐C and decreases PON‐1 activity; (b) interferon‐γ secretion which decreases the expression of SR‐B1 and CD36; (c) MPO expression which modifies Apo‐A1 structure; and (d) MMP and macrophage cathepsin B‐induced cleavage of Apo‐A1. (B) Mechanisms associated with the modification of HDL‐C particles. Inflammatory environments induce HDL oxidation activating NLRP3 inflammasome pathways and the secretion of cytokines while decreasing ABCA1 and ABCG1 expression and the cholesterol carrying capacity of HDL‐C via RCT. FFA, free fatty acid; RAR‐α, retinoic acid receptor‐α.

Adipose tissue lipid kinetics

Cholesterol efflux occurs in several tissues, including the liver, intestine and AT (Basso et al., 2003; Sahoo et al., 2004; Lee and Parks, 2005; Timmins et al., 2005; Singaraja et al., 2006; Zannis et al., 2006; Verghese et al., 2007; Chung et al., 2011; Ji et al., 2012; Phillips, 2014). Regarding the role of AT, in vitro and animal studies demonstrated that cholesterol efflux from adipocytes plays a significant role in the initial lipidation of Apo‐A1 and the formation of mature and functional HDL‐C particles (Zhang et al., 2010; Chung et al., 2011). These results are supported by studies demonstrating that hepatic cholesterol efflux was essential but insufficient to correct HDL‐C deficiency in hepatic ABCA1^−/− knockout mice, illustrating that extrahepatic ABCA1 expression and cholesterol metabolism are critical for the formation of mature HDL‐C particles (Singaraja et al., 2006). Others showed that individuals with compensated liver cirrhosis have higher levels of IL‐6 and NF‐κB but lower levels of HDL‐C and Apo‐A1 (Trieb et al., 2016). Thus, it appears that although the liver plays a significant role, it is not isolated in its contribution to the formation of mature HDL‐C particles.

With the development of AT inflammation, several pathways are activated leading to the impairment of HDL‐C metabolism. AT inflammation has been shown to suppress the expression and function of cholesterol transfer proteins (e.g. ABCA1) leading to decreased efflux of cholesterol from AT (De Haan et al., 2014; Figure 1). This results in the formation of immature rather than mature HDL‐C particles, which fail to successfully transfer cholesterol to the liver (Rashid and Genest, 2007). In our recent clinical study, we used a deuterium labelling approach to estimate the fractional synthesis of triglycerides (f _TG) in AT in humans with differing degrees of obesity. Our results demonstrated that f _TG is inversely associated with the markers of insulin sensitivity (Tuvdendorj et al., 2013). Furthermore, f _TG is associated with the total concentration of HDL‐C and the fractional contribution of large HDL‐C particles (Tuvdendorj et al., 2016). Based on the principles of the stable isotope tracer labelling approach (synthesis‐breakdown/lipolysis = net balance; Turner et al., 2003; Wolfe and Chinkes, 2005; Tuvdendorj et al., 2013, 2016) and the reports that AT TG efflux directly correlates with the efflux of cholesterol (Le Lay et al., 2003; Verghese et al., 2007), we assumed that the f _TG represented AT cholesterol efflux in these individuals. Taken together, these data suggest that inflammation in AT is one of the principal factors affecting HDL‐C functionality (Figure 1A). Thus, these data suggest that metabolically healthier people have high cholesterol efflux and higher levels of circulating total and functional HDL‐C particles that are able to fulfil their atheroprotective role. Notably, our data describing the association between AT lipid flux and HDL‐C metabolism were true for women but not for men, supporting the sex‐dependent nature of lipid metabolism (Hazzard and Applebaum‐Bowden, 1990; Williams, 1997).

An additional factor in reduced HDL‐C functionality arises from AT dysfunction and chronic inflammation hindering the ability of adipocytes to take up excess dietary calories. As a result, the concentration of circulating fatty acids (FAs) is increased. Because FAs are used in the liver for the synthesis of triglycerides and VLDL‐TG, a concomitant increase in the overall concentration of circulating lipids occurs (Figure 1A). As a result, CETP modulates the increased exchange of TGs for cholesteryl esters in HDL‐C particles. These TG‐enriched HDL‐C particles are vulnerable to clearance resulting in a decreased concentration of HDL‐C (Rashid et al., 2002; Figure 1A).

Inflammatory markers

As depicted in Figure 1A, from a cell signalling perspective, AT inflammation is associated with increased activation of pro‐inflammatory pathways resulting in enhanced secretion of inflammatory cytokines [i.e. TNF‐α, IL‐6, IL‐1β, C‐reactive protein (CRP) and serum amyloid A (SAA; Mortensen, 2001; Ryden et al., 2002; Berg and Scherer, 2005; Stienstra et al., 2012; Rodríguez‐Hernández et al., 2013)]. These cytokines increase the activity of downstream factors (e.g. transcription factor activator protein 1, NF‐κB and INF regulatory factor), thus up‐regulating the gene expression of inflammatory mediators (Figure 1A). Notably, TNF‐α plays a central role in this inflammatory process. It promotes the secretion of other pro‐inflammatory cytokines and decreases the production of the anti‐inflammatory cytokine adiponectin (Hotamisligil et al., 1993; Kern et al., 2001; Ryden et al., 2002; Xu et al., 2003). Animal, but not human, studies have shown that decreased production of adiponectin results in decreased expression of mRNA for Apo‐A1 and ABCA1, resulting in decreased levels of Apo‐A1 and HDL‐C (Arita et al., 1999; Hotta et al., 2000, 2001; Kondo et al., 2002; Oku et al., 2007). Additionally, all these cytokines suppress the activity of LPL, which also leads to decreased levels of both the Apo‐A1 and HDL‐C (Hotamisligil et al., 1993; Kern et al., 2001; Ryden et al., 2002; Dusanov et al., 2016; Jung et al., 2016; O'Reilly et al., 2016; Ottobelli et al., 2016) (Figure 1A).

Impaired HDL‐C metabolism is also associated with inflammation‐induced macrophage migration to AT (Figure 1A). Macrophages cause cell‐mediated modifications of Apo‐A1, such as chlorination, nitration, oxidation and proteolysis. In vitro studies demonstrated that macrophages limit the ability of Apo‐A1 to solubilize lipids and promote ABCA1‐dependent cholesterol efflux. The proteolytic mechanism identified is C‐terminal cleavage of Apo‐A1 at Ser ²²⁸ by cathepsin B, which diminishes the functionality of Apo‐A1 and HDL‐C (Figure 1A). Cathepsins are proteases that are secreted by inflammatory macrophages (Brehm et al., 2014; Abd‐Elrahman et al., 2016; Yan et al., 2016). In point of fact, this cathepsin B‐promoted cleavage process is inhibited by the lipidation of Apo‐A1, which causes the C‐terminal region of Apo‐A1 to become more α‐helical, thereby providing cleavage protection (Dinnes et al., 2016).

In chronic inflammation, adipocytes secrete SAA (Poitou et al., 2005; Sjöholm et al., 2005), which has a high affinity for HDL‐C and is known to displace apolipoproteins from HDL‐C (Figure 1A). When SAA is attached to HDL‐C, LCAT‐mediated esterification of HDL‐C is reduced affecting the formation of mature HDL‐C particles. SAA also inhibits the activity of HDL‐associated antioxidant enzyme paraoxonase (PON‐1) resulting in HDL‐C being unable to prevent the oxidation of LDL. Thus, SAA affects both the synthesis of mature HDL‐C and its anti‐oxidative properties (Clifton et al., 1985; Malle et al., 1993; Kappelle et al., 2011). Interestingly, patients with rheumatoid arthritis (RA) have been shown to have increased levels of SAA and low levels of HDL‐C (van Eijk et al., 2009) along with higher rates of cardiovascular complications (Lehtinen, 1993; Lautermann and Braun, 2002; Peters et al., 2004). Interventions to ameliorate inflammation in patients with RA have been shown to decrease SAA (van Eijk et al., 2009; McInnes et al., 2013) and improve HDL‐C metabolism (van Eijk et al., 2009; Charles‐Schoeman et al., 2016), which is discussed below (see subsection ‘Methotrexate and Etanercept’).

Other factors affecting HDL‐C metabolism

Complement systems 3 and 4 (C3 and C4 respectively) play a significant role in inflammation, dyslipidaemia and metabolic syndrome. They have been shown to increase the levels of CRP, TG and fibrinogen. Increased plasma levels of circulating sC5b‐9 complex have been shown to be inversely correlated with the concentration of HDL‐C (Pasqui et al., 2000; 2002; Onat et al., 2005; Liu et al., 2016); Figure 1A). Inflammation also results in increased secretion of secretory PLA₂ (sPLA2) from various tissues. sPLA₂ increases the hydrolysis of HDL‐C particles, breaking down the phospholipids from HDL‐C and thus decreasing the HDL‐C concentration (Tietge et al., 1999; Rye and Duong, 2000). The above factors affect the concentration and, consequently, the functionality of HDL‐C (Figure 1A).

Several other agents have also been shown to directly affect the functionality of HDL‐C. Low‐grade systemic inflammation causes a down‐regulation in the expression of retinoic acid receptor‐α and alters its binding to the promoters of SR‐B1 and ABCA1 (Maitra and Li, 2013). This leads to decreased expression of SR‐B1, ABCA1 and ABCG1 in macrophages and thus decreased RCT (Figure 1A). Matrix metalloproteinases (MMPs) are expressed during inflammation and modulate the function of inflammatory cytokines. MMP3, MMP7 and MMP12 have been shown to cleave Apo‐A1 at its carboxyl terminus (Lindstedt et al., 1999). Furthermore, INF‐γ, another inflammatory cytokine, has been shown to induce activation of macrophages and to decrease the expression of SR‐B1 and messaging associated with CD36 (Zuckerman et al., 2000; Imachi et al., 2001; Bujold et al., 2009). Additionally, the conditional acute phase reactant myeloperoxidase (MPO) uses chloride ions and cell‐generated hydrogen peroxide to create hypochlorous acid that damages Apo‐A1 (Smith, 2010; Figure 1A), thus altering the structure and function of HDL‐C. Plasma Apo‐A1 is a selective target for MPO‐mediated protein modification which results in high levels of covalent modifications. All of these mechanisms have been shown to reduce the capacity of HDL‐C to take up cholesterol and thus inhibit HDL‐C function (i.e. RCT). HDL‐C particles become oxidized during acute inflammation and present an additional source of impaired transport (Figure 1B). The oxidized HDL‐C particles exacerbate inflammation by activating the NLRP3 inflammasome and thereby impacting HDL‐C and the inflammatory environment by (1) activating downstream cytokines and caspase 1; (2) inducing the secretion of IL‐18 and IL‐1β; and (3) decreasing ABCA1 and ABCG1 activity (Figure 1B), thus further impairing HDL‐C (van Lenten et al., 1995; Nakajima et al., 2000; He et al., 2013; Li et al., 2016).

Anti‐inflammatory interventions and HDL‐C metabolism

The essential question at this time is whether interventions to suppress inflammation and inflammatory markers can influence HDL‐C levels and improve HDL‐C functionality. The CETP inhibitors and other established pharmaceuticals are known to affect HDL‐C metabolism; however, these drugs do not affect inflammatory processes; therefore, we will not be discussing them in this review. In this section, we will discuss interventions that alter HDL‐C metabolism via inflammatory pathways, including several pharmaceutical and non‐pharmaceutical interventions that have been shown to ameliorate inflammation and potentially improve HDL‐C metabolism, as well as others which are currently being evaluated. A summary of these interventions, including positive and negative treatment outcomes, is presented in Table 1.

Table 1.

Effects of pharmaceutical agents on inflammation and HDL‐C metabolism

Pharmaceutical agents	Effect on		References
	Inflammation	HDL‐C metabolism
Aspirin	Inhibits NF‐κB and MAPK	Increases expression of ABCB11 in the liver leading to increased bile acid excretion and RCT	Amann and Peskar, 2002 Demetz et al., 2014
	Decreases inflammatory cytokines	Increases SR‐B1 expression leading to increased cholesterol transport.	Tancevski et al., 2006 Herová et al., 2014
	–	Increases LCAT and PON‐1 activity	Jafarnejad et al., 2008
Salsalates	Inhibits NF‐κB	Increases adiponectin and expression of ABCA1 and AMPK leading to increased cholesterol efflux to Apo‐A1 and HDL‐C	Fullerton et al., 2015 Fakhri et al., 2014 Goldfine et al., 2013a Goldfine et al., 2013b
Metformin	Decreases NF‐κB, CRP and IL‐6	Increases PON‐1 enzyme activity and the anti‐oxidative function of HDL‐C	Camps et al., 2016 Yoshifumi, 2016 Goldberg et al., 2016
Statins	Decreases NF‐κB(at high dose). Also decrease MMP9 and CRP	Increases the expression of Apo‐A1 gene by activating PPAR‐α Decreases VLDL‐TG, TGs and CETP activity and raises HDL‐C concentration	Kim et al., 2007; Singh et al., 2008; Bonnet et al., 2008; van de Ree et al., 2003 Martin et al., 2001 Schaefer and Asztalos, 2006
Extended release niacin	Decreases NF‐κB, TNF‐α and IL‐6	Decreases clearance of HDL‐C by decreasing CETP activity. Increases Apo‐A1 and adiponectin mRNA expression raising HDL‐C concentration	Si et al., 2014 Yadav et al., 2015 Digby et al., 2010
–	–	May increase anti‐ApoA1 antibody formation that reduces HDL‐C anti‐oxidative function	Batuca et al., 2016
Methotrexate	Decreases MMP‐1, TNF‐α, IL‐6, IL‐1 and IFN‐γ	Increases expression of ABCA1 leading to enhanced RCT	Coomes et al., 2011 Cutolo et al., 2001 Reiss et al., 2008 Chan and Cronstein, 2010
PPAR‐α agonists	Decrease NF‐κB, CRP and IL‐6	Increase Apo‐A1 and adiponectin leading to increase HDL‐C concentration. Increase ABCA1, ABCG1 and SR‐B1 activity in macrophages leading to enhanced RCT	Ogata et al., 2009 Mahdy et al., 2012 Colin et al., 2015 Wagner et al., 2010
–	PPAR‐α agonists did not decrease inflammation in rodents with renal crystal formation. Fenofibrate did not decrease inflammatory markers in one study	–	Taguchi et al., 2016 Hogue et al., 2008
PPAR‐γ agonists	Suppresses IFN‐γ and increases M2 macrophages	Increase adiponectin and ABCA1 leading to higher HDL‐C concentration. Enhance cholesterol efflux and increase expression of SR‐B1	Colin et al., 2015 Zuckerman et al., 2000 Bujold et al., 2009
Biological agents (IL‐6 inhibitor, JAK inhibitor and TNF‐α inhibitor)	Modulate immune response Decrease inflammatory cytokines	Improve HDL‐C concentration and metabolism	Souto et al., 2015; Genovese et al., 2008 Mathieu et al., 2010 Kawashiri et al., 2011 Ghoreschi et al., 2011 van Eijk et al., 2009 McInnes et al., 2013 Charles‐Schoeman et al., 2016
–	–	Etanercept may cause hypertriglyceridaemia	Haroon and Devlin, 2009
LXR agonists	Decrease TNF‐α, IL‐6, IL‐1β, MMP‐9 and IFN‐γ	Increase HDL concentration and function by increasing ABCA1, ABCG1 and RCT	Joseph et al., 2003 Wang et al., 2006 Jamroz‐Wisniewska et al., 2007 Jiang et al., 2003 Miao et al., 2004 Repa et al., 2000b Naik et al., 2006
–	–	May cause increase in TGs, VLDL and LDL by inducing SREBP‐1c	Repa et al., 2000a Schultz et al., 2000 Grefhorst et al., 2002
Sildenafil	Decrease NF‐κB, TNF‐α and IL‐1	Increase HDL‐C function by unknown mechanism	Nunes et al., 2015
HDL‐C reconstituted therapy Apo‐A1 Milano	Decrease MMP‐9	Increase HDL‐C function by increasing ABCA1 activity	Uehara et al., 2015 Nasr et al., 2015
4F mimetic peptide	Decrease inflammatory cytokines	Increase RCT	Bloedon et al., 2008 Smythies et al., 2010
5‐Lipoxygenase inhibitors	Decrease adipose tissue inflammation, SAA, CRP and MPO levels	Improve HDL‐C metabolism	Horrillo et al., 2010 Bäck et al., 2007 Allayee et al., 2007

Pharmaceutical approaches

Current therapies

Aspirin and salsalates

Aspirin is one of the oldest drugs in use and its effects in treating inflammation are widely known. It inhibits COX‐1 and COX‐2 and thus the synthesis of prostaglandins and thromboxane (Spite and Serhan, 2010). In terms of CVD treatment, aspirin has been shown to reduce the risk of first myocardial infarction in men in direct correlation with initial CRP levels ( Ridker et al., 1997). Aspirin does not, however, appear to reduce inflammation via pathways which impact CRP levels, as aspirin has been shown to be minimally effective in reducing CRP levels in at‐risk diabetic populations (Hovens et al., 2008) and aspirin‐treated healthy volunteers (Feng et al., 2000; Feldman et al., 2001). Currently, meta‐analyses indicate that aspirin therapy should only be recommended for the most at‐risk patients for the prevention of cardiovascular events in diabetics due to its pro‐haemorrhagic effects (Pignone et al., 2010). Mechanistically, in addition to its effects on COX‐1 and COX‐2, aspirin inhibits the activation of NF‐κB, activator protein 1 and MAPK (Amann and Peskar, 2002). In terms of its direct impact on HDL‐C metabolism, studies in isolated human macrophages showed that aspirin increased SR‐B1 expression and labelled HDL‐associated cholesteryl oleate uptake, as well as enhancing SR‐B1 expression in mice in vivo (Tancevski et al., 2006). Additional studies in M1 macrophages showed reduced inflammatory cytokine secretion upon exposure to aspirin, which was associated with a reduction in chemerin secretion by adipose tissue (Herová et al., 2014). Increased levels of circulating chemerin have been shown to inversely associate with HDL‐C levels (Herová et al., 2014). Rodent studies demonstrated that aspirin increases the expression of ATP binding cassette subfamily B member 11 (ABCB11) in the liver leading to increased bile acid excretion and enhanced RCT, indicating an improvement in cholesterol transfer to HDL‐C (Demetz et al., 2014). Further studies in diabetic rats showed that long‐term aspirin therapy reduces HbA1c and advanced glycated end product formation while improving HDL functionality (Jafarnejad et al., 2008). Additionally, HDL‐C may play a role in increasing the oxidative capacity of aspirin via the HDL‐associated enzyme PON‐1, which actively hydrolyzes aspirin to salicylate, whose free radical scavenging properties may protect against atherosclerosis (Santanam and Parthasarathy, 2007). Interestingly, aspirin‐resistant patients were revealed to have reduced HDL‐C levels, indicating that aspirin efficacy may in some way be associated with HDL‐C function (Azmin et al., 2013).

Salsalate, which is chemically related to aspirin, has been shown to increase the levels of adiponectin, HDL‐C and Apo‐A1 (Goldfine et al., 2013a,b; Fakhri et al., 2014; Fullerton et al., 2015) while also suppressing the activation of NF‐κB. Salsalate increases the expression of ABCA1 gene and activates AMPK, leading to increased cholesterol efflux to HDL‐C and Apo‐A1 (Goldfine et al., 2013a,b; Fakhri et al., 2014; Fullerton et al., 2015). Additionally, salsalate decreases the levels of HbA1c, fasting blood sugar and C‐peptide.

Metformin

Metformin is an anti‐diabetic drug, which also suppresses inflammation, decreases the levels of LDL‐C and leptin and reduces body weight (Camps et al., 2016; Yoshifumi, 2016). It activates AMPK, which suppresses pathways associated with NF‐κB. It also reduces the levels of other biomarkers of inflammation, including CRP, IL‐6, E‐selectins, intracellular adhesion molecule 1 (ICAM‐1), fibrinogen and tissue plasminogen activator, but with less effect than lifestyle modifications (Goldberg et al., 2016). Furthermore, metformin up‐regulates the activity of PON‐1 and thus increases the anti‐oxidative capacity of HDL‐C to prevent LDL oxidation. Goldberg et al. (2016) report that metformin increases HDL‐C independently of changes in adiponectin. In spite of these effects, recent reports suggest that metformin‐sensitive AMPK could be a key player in the development of Alzheimer's disease pathology (Domise et al., 2016). Thus, careful consideration in using metformin, as well as salsalate (Goldfine et al., 2013a,b; Fakhri et al., 2014; Fullerton et al., 2015), may be required. Metformin has been reported to cause metabolic acidosis and renal failure in some patients, which also needs to be taken into consideration.

Statins

Statins have been shown to have anti‐inflammatory effects, which are most evident at higher dosages. High‐dose atorvastatin (80 mg) suppresses NF‐κB‐associated inflammation, reduces the levels of MMP9 (Kim et al., 2007; Singh et al., 2008) and markedly decreases the concentration of CRP (van de Ree et al., 2003; Bonnet et al., 2008). Statin therapy is known for its ability to not only decrease LDL‐C levels but also concurrently increase HDL‐C. Multiple pathways are involved in the elevation of HDL‐C levels. Firstly, statins inhibit Rho factor, which results in increased activation of PPAR‐α and thus in an increased expression of the Apo‐A1 gene (Martin et al., 2001; Schaefer and Asztalos, 2006). Secondly, statins decrease VLDL‐TG and TG levels as well as CETP activity. The synergetic effect is an increased concentration of HDL‐C (Martin et al., 2001; Schaefer and Asztalos, 2006).

Extended release niacin

Niacin exerts its atheroprotective effects by acting on AT through GPCRs, thereby influencing both pro‐ and anti‐inflammatory markers. In vivo studies show that niacin decreases the levels of inflammatory markers TNF‐α and IL‐6 and suppresses the activation of the NF‐κB pathway (Si et al., 2014). The use of extended release niacin increases Apo‐A1 and HDL‐C levels while decreasing levels of total cholesterol, TG, LDL, monocyte chemoattractant protein (MCP‐1) and SAA (Si et al., 2014; Yadav et al., 2015). While some studies showed that niacin administration had no effect on the anti‐oxidative capacity of HDL, others demonstrated a decrease in oxidized‐LDL‐induced cell apoptosis and a reduction in blood vessel wall inflammation (Si et al., 2014; Yadav et al., 2015). Niacin also up‐regulates the expression of factors involved in RCT (Si et al., 2014), decreases CETP activity and reduces HDL‐C clearance. Additionally, niacin increases adiponectin mRNA expression (Digby et al., 2010).

Methotrexate and etanercept

Methotrexate is a disease modifying anti‐rheumatic drug, which is used as a first line treatment for RA. It exerts its anti‐inflammatory effect by promoting the accumulation of adenosine, which subsequently binds to the A_2A receptor. As a result, the expression of ABCA1 and 27‐hydroxylase is promoted and RCT is increased. This action reverses the atherosclerotic effect of other COX‐2 inhibitors used to treat RA, as has been demonstrated in RA patients (Reiss et al., 2008; Chan and Cronstein, 2010). Additionally, methotrexate decreases MMP1, LTB₄, inflammatory cytokine (e.g. TNF‐α, IL‐6 and IL‐1β) and INF‐γ expression while increasing the expression of anti‐inflammatory IL‐10 (Coomes et al., 2011; Cutolo et al., 2001). The ongoing cardiovascular inflammation reduction trial will probe the efficacy of methotrexate therapy on patients who have suffered from prior myocardial infarction combined with diabetes or metabolic syndrome (Everett et al., 2013). Although a negative impact on the ratio of total cholesterol to HDL‐C has been observed following methotrexate administration, the impact of methotrexate on HDL‐C function is unknown (Navarro‐Millán et al., 2013). The impact of targeting methotrexate‐sensitive pathways on CVD progression will provide useful insights into the inflammation processes at work in atherosclerosis.

Etanercept is a TNF‐α inhibitor that is used to treat RA. Several clinical trials have investigated its effect on inflammation and HDL‐C parameters in patients with RA (van Eijk et al., 2009; Charles‐Schoeman et al., 2016). These and other studies have demonstrated improvements in HDL‐C metabolism. Interestingly, the majority, but not all (Rodriguez‐Jimenez et al., 2014), reports showed that treatment with etanercept also decreased the concentration of TNF‐α (van Eijk et al., 2009; Charles‐Schoeman et al., 2016). It is possible that the effect of etanercept on HDL‐C metabolism was exerted via the SAA mechanism (van Eijk et al., 2009).

Peroxisome proliferator‐activated receptors agonists

PPARs are a nuclear receptor subfamily with three members, PPAR‐α, PPAR‐γ and PPAR‐β/δ. PPAR agonists have been shown to have anti‐inflammatory properties. PPAR‐γ agonists increase the expression of adiponectin and display anti‐inflammatory activity by promoting the polarization of monocytes towards alternative M2 macrophages (Colin et al., 2015). Rosiglitazone decreases the levels of inducible NO synthase (iNOS), ICAM‐1 and COX‐2 (Cuzzocrea et al., 2004). The PPAR‐γ agonist 15‐deoxy‐delta12, 4‐PGJ₂ (15d–PGJ2) reverses the INF‐γ‐related effects on HDL‐C metabolism (Zuckerman et al., 2000; Bujold et al., 2009). This agonist enhances CD36 messaging, leading to increased cholesterol efflux to HDL‐C, and increases the expression of SR‐B1 and its binding to HDL‐C (Zuckerman et al., 2000; Bujold et al., 2009). PPAR‐α agonists, for example, fibrates, ameliorate inflammation by decreasing the levels of CRP and IL‐6 while inhibiting the activation of the NF‐κB pathway (Ogata et al., 2009; Mahdy et al., 2012). PPAR‐α agonists additionally decrease VLDL‐TG levels by increasing β‐oxidation of free FAs in the liver, thus decreasing the availability of FAs for VLDL‐TG synthesis (Mahdy et al., 2012). Reduced VLDL‐TG levels are also facilitated by PPAR‐γ agonists, which increase the expression of ABCA1. However, the reports on the effect of PPAR‐γ agonists on the concentration of HDL‐C are not consistent. Carreón‐Torres et al. (2009) demonstrated that in rabbits, rosiglitazone increases the production rate of Apo‐A1, resulting in increased concentration of HDL‐C and increased activity of PON‐1. Mizoguchi et al. (2011), who studied insulin tolerant and diabetic patients, reported that treatment with pioglitazone increased HDL‐C levels while decreasing CRP and the size of atherosclerotic plaques. In this particular study, patients receiving aspirin, renin angiotensin system inhibitors and statins were included, which may have biased the final results. In contrast, Millar et al. (2010) demonstrated that in subjects with metabolic syndrome, rosiglitazone increased the production rate of Apo‐A2 with no effect on Apo‐A1 metabolism. PPAR‐α agonists have been shown to stimulate the synthesis of both Apo‐A1 and Apo‐A2 and thus increase plasma HDL‐C levels (Colin et al., 2015). They have also been shown to stimulate the activity of ABCA1, ABCG1 and SR‐B1 in macrophages and thus increase RCT (Mahdy et al., 2012; Colin et al., 2015).

PPAR agonists are already being used in clinical practice; however, they are primarily used to treat dyslipidaemia and insulin resistance. PPAR‐γ agonists are primarily used to treat type 2 diabetes and they play a significant role in enhancing FA oxidation in the liver (Colin et al., 2015). PPAR‐β/δ agonists are used to improve lipid metabolism, as they by reduce TG and LDL‐C and increase HDL‐C levels. Moreover, PPAR‐β/δ activation increases the expression of genes promoting insulin sensitivity (Colin et al., 2015). Unfortunately, many of the currently available medications are either ineffective or have adverse effects that may outweigh their benefits for treating inflammation‐related impairments in HDL‐C function. For example, fibrates have a weak impact on PPAR‐α activity, although a newer agent, K‐877, binds strongly to PPAR‐α and is in phase III clinical trials for atherosclerotic dyslipidaemia in Japan (Colin et al., 2015). Wagner et al. (2010) demonstrated that in monkeys, the new PPAR‐α agonist CP‐900691 increases the levels of adiponectin and HDL‐C while decreasing the levels of CRP, TG, VLDL and LDL‐C. Other selective PPAR agonists, for example, CER‐002 (PPAR‐δ), DSP‐8658 (PPAR‐α/ϒ), INT131 (PPAR‐α/ϒ) and GFT505 (PPAR‐α/δ), are undergoing clinical trials and exhibit promise for treating the cardiovascular risks associated with metabolic syndrome and type 2 diabetes (Colin et al., 2015). To date, there is no information on the effect of these new agonists on inflammation‐related abnormalities.

Future therapies

HDL‐C reconstituted therapy

The use of artificial components of HDL‐C, as a reconstituted therapy (i.e. rHDL), has also been investigated. Apo‐A1 Milano and Fukuoka Apo‐A1 mimetic peptides have proved effective in animal models (Uehara et al., 2015). This approach enhanced the biological function of HDL‐C without elevating its concentration. Both therapeutics act as anti‐atherosclerotic agents and remove cholesterol via the ABCA1 transporter. Notably, when these agents were used in patients with symptomatic carotid plaque, no significant differences were noted in expression of genes involved in formation of thrombus (Nasr et al., 2015). However, the use of reconstituted peptides prevented the significant postoperative surge in plasma IL‐6, which was seen in the placebo group. Surgical intervention reduced systemic levels of tissue factor, MMP9 and MCP‐1 in the rHDL group, although the effects on MMP9 and MCP‐1 were abolished in the immediate postoperative period (Nasr et al., 2015). The 4F mimetic peptide was studied and found to have anti‐inflammatory properties in vitro and in humans (Bloedon et al., 2008; Smythies et al., 2010). The mimetic peptide decreased the levels of pro‐inflammatory cytokines and the adhesion of monocytes to human endothelial cells while increasing RCT by enhancing cholesterol efflux in macrophages (Smythies et al., 2010).

Liver X receptors agonists

The results from studies in mice have demonstrated that agonists of liver X receptors ‐α and ‐β (LXR‐α and LXR‐β respectively) increase the total concentration and the size of HDL‐C particles by up‐regulating the expression of ABCA1 and ABCG1 (Repa et al., 2000b; Jiang et al., 2003; Miao et al., 2004). Furthermore, LXR agonists have been shown to promote RCT resulting in the reduced deposition of cholesterol in atherosclerotic plaques (Naik et al., 2006). Conflictingly, LXR agonists have also been shown to up‐regulate the expression of CETP in CETP‐expressing transgenic mice, which completely abolished the beneficial effect of LXR on HDL‐C metabolism and increased the levels of LDL‐C and VLDL‐TGs (Jiang et al., 2003; Masson et al., 2004; Beltowski, 2008). LXR agonists also induced the sterol regulatory element binding protein‐1c in the liver, which has been shown to associate with the increased concentration of TGs (Repa et al., 2000a; Schultz et al., 2000; Grefhorst et al., 2002). In contrast, it has been hypothesized that LXR‐β selective agonists would decrease the levels of TG and LDL‐C. Several LXR‐β selective pyrazole and imidazole biaryl sulfones have been prepared. In particular, imidazole 18 (EXEL‐04286652, BMS‐779788) is an LXR‐β partial agonist that induces ABCA1, making it a reagent of interest for further study (Kick et al., 2015; Matsuda et al., 2015). Additionally, a novel synthetic, steroidal LXR ligand, ATI‐111, has been developed. This molecule exhibits a strong effect on LXR‐α with a modest effect on LXR‐β. Encouragingly, animal and in vitro studies indicate that ATI‐111 has beneficial anti‐atherosclerotic and anti‐inflammatory effects ranging from reduced hypertriglyceridaemia to decreased atherosclerotic lesions. To assess the full potential of ATI‐111, clinical trials will be necessary. Presently, a phase I clinical trial with XL‐652 (XL‐014), a novel LXR ligand, is underway (Colin et al., 2015). Furthermore, the LXR agonists have been shown to have anti‐inflammatory properties. Mouse studies with LXR agonists T0901317 and GW3965 demonstrated decreased levels of inflammatory cytokines TNF‐α, IL‐6, IL‐1β, IFN‐γ, MMP‐9 and ICAM‐1 (Joseph et al., 2003; Wang et al., 2006; Jamroz‐Wisniewska et al., 2007), indicating another beneficial side of LXR agonists warranting further investigation.

Phosphodiesterase‐5 inhibitors

Sildenafil inhibits cGMP‐specific PDE5, an enzyme that promotes degradation of cGMP. Inhibition of PDE5 results in smooth muscle relaxation, which alleviates erectile dysfunction and pulmonary arterial hypertension (Balhara et al., 2015; Igarashi et al., 2016). Sildenafil therapy increased HDL‐C and decreased P‐selectin and LDL‐C levels through mechanisms which are not yet understood (Mandosi et al., 2015). Nunes et al. (2015) reported that administration of sildenafil reduces the expression of pro‐inflammatory cytokines IL‐1β and TNF‐α while increasing the levels of anti‐inflammatory cytokine IL‐10. In addition, sildenafil has been shown to reduce the expression of GFAP, NF‐κB, inactive AMPK and iNOS and to increase IKβα levels (Nunes et al., 2015). Thus, sildenafil may potentially be used to treat inflammation and the associated decrease in HDL‐C function.

Monoclonal antibodies

A monoclonal antibody, canakinumab, is an IL‐1β inhibitor which has been shown to be effective in treating juvenile RA (Gencer et al., 2015). The canakinumab anti‐inflammatory thrombosis outcomes study trial is currently underway to determine the effects of canakinumab on stable CVD patients who exhibit high levels of inflammation (hsCRP >2 mg. L^‐1; Ridker et al., 2011). While existing data on canakinumab would indicate that overall HDL‐C levels will not be affected (Ridker et al., 2012), the impact of targeting inflammation via the IL‐1β pathway for reducing cardiovascular events will provide insight into the role of inflammation in CVD and the potential benefits of a more thorough investigation of the effect of canakinumab on HDL‐C. A similar but more direct approach for treating dyslipidaemia may reside in the use of monoclonal antibody inhibitors of proprotein convertase subtilisin/kexin type 9 (PCSK9), such as alirocumab and evolocumab, which were approved by the FDA in 2015. PCSK9, which contributes to the development of atherosclerosis, is believed to be expressed in macrophages, smooth muscles and endothelium. Functionally, PCSK9 down‐regulates the expression of the stress response genes and reduces inflammation in liver cells, indicating that PCSK9 affects metabolic pathways beyond cholesterol metabolism. Inhibition of PCSK9 improves the removal of LDL‐C from blood by the liver (Lan et al., 2010). Furthermore, Walley et al. (2014) demonstrated that reducing PCSK9 function increases pathogen lipid clearance via the LDL receptor, thereby decreasing the inflammatory response and improving sepsis outcomes in both mice and humans. This finding is in contrast to that of Sahebkar et al. (2016), who conducted a meta‐analysis evaluating the effects of several PCSK9 inhibitors on the levels of high‐sensitivity CRP (hs‐CRP) and demonstrated that PCSK9 inhibitors do not affect the hs‐CRP levels. The PCKS9 story is relatively new and will be revisited in coming years. Nevertheless, monoclonal antibodies show promise for ameliorating inflammation and related dyslipidaemia.

Biological agents

According to several recent studies and meta‐analyses, biological agents for the treatment of inflammatory arthritis induced changes in several lipid profiles. Specifically, tocilizumab, an IL‐6 inhibitor, and tofacitinib, a JAK inhibitor, are recombinant proteins, which have been shown to decrease the levels of inflammatory cytokines and increase HDL‐C concentration. The mechanisms whereby these agents exert their beneficial effect involve modulation of the immune response. JAK inhibitors block intracellular signalling of several cytokines. This effect was confounded by the inability of the TNF‐α antagonists to show any marked improvement in HDL‐C (Dahlqvist et al., 2006; Genovese et al., 2008; Soubrier et al., 2008; Mathieu et al., 2010; Ghoreschi et al., 2011; Kawashiri et al., 2011; van Vollenhoven et al., 2012; McInnes et al., 2013; Souto et al., 2015).

5‐LOX inhibitors (theophylline and montelukast)

Theophylline and montelukast have been investigated for the treatment of inflammation‐related asthma (Allayee et al., 2007). Both of these drugs inhibit the pathway controlled by 5‐lipoxygenase (5‐LOX), a critical agent in the leukotriene pathway, which is expressed in AT and plays a significant role in obesity‐related AT inflammation (Horrillo et al., 2010). Modulation of the leukotriene pathway, using 5‐LOX activated protein (i.e. FLAP) inhibitors, has been shown to decrease the levels of systemic pro‐inflammatory cytokines, AT macrophage content and systemic insulin resistance (Bäck et al., 2007; Horrillo et al., 2010). Hakonarson et al. (2005) demonstrated that interventions using 5‐LOX inhibitors decreased the levels of inflammatory markers such as SAA, CRP and MPO. Similarly, Allayee et al. (2007) showed that treatment of asthmatics with theophylline and montelukast decreased the levels of CRP, IL‐6, VLDL‐TG and LDL‐C. Unfortunately, in this report, the HDL‐C levels were reduced in the treatment group compared with placebo, indicating a potential detrimental effect of montelukast and theophylline therapy on HDL‐C metabolism. Nevertheless, because of the reported significance of 5‐LOX‐associated pathways in obesity‐related inflammation, further studies that target this strategy for reducing inflammation are of interest.

Non‐pharmaceutical approaches

Diet and dietary components

A hypocaloric, high‐fat, low‐carb diet has been found to decrease CRP and increase adiponectin. This regime has also been shown to lower hepatic VLDL‐TG and TG secretion and decrease their hydrolysis by hepatic lipase, thereby increasing HDL‐C (Ruth et al., 2013). Increased intake of tree nuts causes a decrease in total cholesterol, LDL‐C, TG and Apo‐B, while exerting no effect on HDL‐C, Apo‐A1, CRP and hypertension (Ros et al., 2004; Demonty et al., 2009; Sabate and Wien, 2013; Del Gobbo et al., 2015). Although numerous isolated reports show the beneficial effect of red wine and resveratrol supplementation on inflammatory markers, the meta‐analyses demonstrated no positive impact from resveratrol supplementation on cardiovascular risk factors (Sahebkar et al., 2015). In fact, a slight decrease in HDL‐C levels has been reported (Sahebkar et al., 2015). Krill oil consumption, which has an anti‐oxidative effect, is associated with increased HDL‐C and Apo‐A1 and decreased TG and inflammation levels in healthy young adults (Berge et al., 2015; Cicero et al., 2016). Caffeic acid is a naturally occurring phenolic compound found in many fruits, vegetables and herbs (Moon et al., 2009). It decreases TNF‐α‐induced induction of adhesion molecules including ICAM‐1, vascular adhesion molecule (VCAM‐1) and P selectin. It also decreases TNF‐α‐induced activation of IL‐8 and NF‐κB (Moon et al., 2009). Aloe vera exhibits several beneficial effects. It decreases the activation of inflammasome NLRP3, IL‐8, IL‐6, IL‐1β and TNF‐α, as well as the activation of inflammatory NF‐κB, p38 and JNK pathways, and thus decreases inflammation and raises HDL‐C levels (Budai et al., 2013; Kumar et al., 2013).

Exercise and life style interventions; bariatric surgery; and electro acupuncture therapy

Exercise and life style interventions show promising effects on inflammatory markers and lipid profiles. These strategies increase the levels of adiponectin and HDL‐C and decrease inflammatory cytokines such as INF‐γ (Nishida et al., 2015; Davidson et al., 2017, Wefers et al., 2016). The improvement in HDL‐C metabolism with weight loss can occur via improvements in several of the mechanisms discussed above. For example, increased levels of adiponectin activate ceramidase and the formation of sphingosine‐1‐phosphate, thus altering HDL‐C sphingolipid content, thereby improving HDL‐C function (Belalcazar et al., 2012). Goldberg et al. (2016) showed that lifestyle intervention increases the levels of adiponectin and HDL‐C while decreasing the levels of inflammatory markers CRP, IL‐6, E selectin, ICAM‐1 and fibrinogen. Moreover, weight loss results in decreased levels of TG, which diminishes the activity of CETP and thus results in higher levels of functional HDL‐C. Animal studies demonstrated that weight loss results in increased expression of ABCG1 protein in AT leading to increased cholesterol efflux (Edgel et al., 2012), which is associated with enhanced levels of functional HDL‐C particles (Wesnigk et al., 2016). Taken together, these and other reports suggest that weight loss beneficially affects HDL‐C metabolism.

Roux en‐Y gastric bypass surgery boosts HDL‐C levels and endothelial function. This results in decreased apoptosis of endothelial cells and increased production of nitric oxide and enhanced PON‐1 activity. Additionally, there is an increase in macrophage‐induced cholesterol efflux. Furthermore, anti‐inflammatory and anti‐oxidative effects are enhanced because of a decrease in TNF‐α‐mediated VCAM‐1 expression and NADPH oxidase activity (Adams et al., 2012; Osto et al., 2015).

Electro‐acupuncture shows promise in treating obesity and controlling inflammation. It decreases BMI and the concentrations of IL‐6, TNF‐α, TG and LDL‐C, while increasing the levels of adiponectin and HDL‐C (Firouzjaei et al., 2016).

Conclusions

While the association of HDL‐C metabolism with the progression of CVD is still being investigated, the evidence supporting a link between tissue and systemic inflammation, lipid kinetics and CVD progression continues to grow. Our current review details the manner in which tissue and systemic inflammation modulates HDL‐C metabolism via several pathways, for example, cholesterol efflux, hyperlipidaemia and apolipoprotein modification. The data available suggest that regardless of its correlation with the progression of CVD, HDL‐C metabolism may provide a window into systemic or tissue (i.e. adipose) health. Our survey of available anti‐inflammatory interventions indicates that increasing Apo‐A1, ABCA1, SR‐B1 and adiponectin levels may improve the production and functioning of HDL‐C particles. The potential health benefits and indirect improvements in HDL‐C metabolism resulting from anti‐inflammatory interventions must, however, be balanced with their potential side effects (Table 2). Additionally, it should be noted that not all the agents discussed are purely anti‐inflammatory and thus may affect HDL‐C metabolism via other factors and mechanisms. Although a healthy lifestyle is the best approach to prevent the development of inflammation, the challenges of implementing lifestyle modification in the general population will require consistent social and medical support. Thus, pharmaceutical interventions to ameliorate inflammation and improve the functionality of HDL‐C and dyslipidaemia are of significant interest. A summary of the common anti‐inflammatory agents of interest for enhancing HDL‐C function, with a list of associated side effects, is presented in Table 2. The question of whether the side effects outweigh the benefits of these agents for ameliorating inflammation and HDL‐C metabolism will need to be addressed by future clinical trials. In conclusion, further research is needed to elucidate and target the mechanisms linking HDL‐C metabolism to both inflammation and the progression of CVD. Some of the studies should address the key mechanisms underlying the complexity and heterogeneity of HDL‐C particles, which should provide a more detailed understanding of the specific functions of these particles.

Table 2.

Side effects of pharmaceutical agents that have anti‐inflammatory properties and can contribute to improving the HDL‐C function

Pharmaceutical agents	Side effects	References
Aspirin	Drug resistance can occur in some patients.	Azmin et al., 2013; Zhang et al., 2017
	Polyp formation and exacerbation of respiratory disease has been observed.	Eskandarian et al., 2012; Cook and Stevenson, 2016
	Chronic salicylate intoxication can cause SIRS.	Chalasani et al., 1996
	Aspirin can cause gastritis and increased risk of gastrointestinal bleeding.	Gartner, 1976
Salsalates	Poor tolerance in HIV patients.	Hileman et al., 2010
	Activate AMPK and may cause Alzheimer's disease.	Domise et al., 2016
Metformin	Effect may depend on race and ethnicity.	Zhang et al., 2015
	It may cause lactic acidosis especially if given in renal diseases.	Lalau, 2010
	It causes hepatotoxicity in PON‐1‐ deficient mice.	García‐Heredia et al., 2016
	Activates AMPK and may cause Alzheimer's disease.	Domise et al., 2016
Statins	Down‐regulate ABCA1 and ABCG1 activity in macrophages.	Sone et al., 2004; Wang et al., 2013; Wong et al., 2008
	Cause myopathy in some patients.	Lahaye et al., 2014; Brinton et al., 2016; Jacobson, 2009; Rosenson, 2004
	Rosuvastatin did not reduce inflammation in sepsis associated acute respiratory distress syndrome.	Truwit et al., 2014
	Unsafe in pregnancy.	Hosokawa et al., 2003
	They may cause hepatotoxicity.	Russo et al., 2014
Niacin	There is increased risk of flushing with niacin use.	Maccubbin et al., 2009
	May cause macular oedema.	Domanico et al., 2015
	May lead to the development of hepatitis.	Etchason et al., 1991
PPAR‐α agonists	Did not decrease inflammation in rodents with renal crystal formation.	Taguchi et al., 2016
	Fenofibrate did not decrease inflammatory markers in one study.	Hogue et al., 2008
	Fibrates may cause increased risk of renal problems.	Zhao et al., 2012
PPAR‐γ agonists	Thiazolidinediones may increase risk of myocardial infarction and heart failure especially rosiglitazone.	Singh et al., 2007 Nissen and Wolski, 2007
	Increased risks of fractures in women.	Loke et al., 2009
	There is increased risk of bladder cancer with pioglitazone.	Ferwana et al., 2013
Biological agents (IL‐6 inhibitor, JAK inhibitor and TNF‐α inhibitor)	Several toxic side effects.	Pichler, 2006
	Increase chance of fungal infections with anti‐TNF‐α fusion inhibitors.	Tragiannidis et al., 2016
LXR agonists	May cause increase in TGs, VLDL and LDL by inducing SREBP‐1c	Repa et al., 2000a; Schultz et al., 2000; Grefhorst et al., 2002
Sildenafil	May cause hypotension if given with nitrates.	Webb et al., 1999
Monoclonal antibodies	Adverse effects including acute anaphylaxis, serum sickness, cardiotoxicity etc.	Hansel et al., 2010; Kizhedath et al., 2016
	Increase chance of fungal infections with anti‐TNFα monoclonal antibodies.	Tragiannidis et al., 2016
5‐Lipoxygenase inhibitors
Theophylline	Can cause adverse effects due to theophylline toxicity.	Eason and Markowe, 1989
Montelukast	Increased risk of ecchymosis.	Aypak et al., 2013
Methotrexate	May cause elevation of liver enzymes and hepatotoxicity.	Curtis et al., 2010
	May cause bone marrow suppression.	Sosin and Handa, 2003

Author contributions

F.I., W.S.B. and D.T. did the literature search; F.I., W.S.B. and D.T. wrote the manuscript; and F.I., W.S.B., M.I.K., S.T., K.H.M., N.A. and D.T discussed and edited the manuscript.

Conflict of interest

The authors declare no conflicts of interest.

Iqbal, F. , Baker, W. S. , Khan, M. I. , Thukuntla, S. , McKinney, K. H. , Abate, N. , and Tuvdendorj, D. (2017) Current and future therapies for addressing the effects of inflammation on HDL cholesterol metabolism. British Journal of Pharmacology, 174: 3986–4006. doi: 10.1111/bph.13743.