Understanding the heterogeneity and dysfunction of HDL in chronic kidney disease: insights from recent reviews

Abstract

Chronic kidney disease (CKD) is a complex disease that affects the global population’s health, with dyslipidemia being one of its major complications. High density lipoprotein (HDL) is regarded as the “hero” in the bloodstream due to its role in reverse cholesterol transport, which lowers cholesterol levels in the blood and prevents atherosclerosis. However, in the complex internal environment of CKD, even this “hero” may struggle to perform its beneficial functions and could potentially become harmful. This article reviews HDL heterogeneity, HDL subclasses, functional changes in HDL during the progression of CKD, and the application of HDL in CKD treatment. This review aims to deepen understanding of lipid metabolism abnormalities in CKD patients and provide a basis for new therapeutic strategies.

Introduction

Chronic kidney disease (CKD) poses a significant global health burden and is characterized by either an estimated glomerular filtration rate (eGFR) below 60 mL/min/1.73 m² or the presence of other indicators of kidney damage, such as albuminuria [1]. According to a study conducted by the Global Burden of Disease group in 2020, CKD was ranked among the top 10 leading contributors to adverse outcomes on a global scale [2]. The morbidity of CKD is continuously rising, with contributing factors including demographic shifts (aging of the population), an increase in type 2 diabetes and hypertension cases, as well as low early detection rates for CKD [3,4,5].

In recent years, numerous studies have indicated that individuals with CKD exhibit a significantly increased mortality rate from cardiovascular diseases (CVD). The factors leading to this outcome can be broadly categorized into two types: traditional risk factors and specific risk factors (Fig. 1). Traditional factors include dyslipidemia, hypertension, hyperglycemia, smoking, and old age, all of which not only promote atherosclerosis but also damage the renal vasculature [6,7,8,9]. Specific risk factors include inflammation, endothelial dysfunction, vascular calcification, genetic factors, and proteinuria [10,11,12,13]. Both types of factors significantly increase the cardiovascular burden.

Fig. 1

Traditional and specific risk factors for cardiovascular disease in patients with chronic kidney disease. (Adapted from “Risk factors for sarcopenia”, by BioRender.com (2024). Retrieved from https://app.biorender.com/biorender-templates)

Among these risk factors, dyslipidemia is particularly noteworthy. One common manifestation of dyslipidemia in patients with CKD is a reduction in high-density lipoprotein cholesterol (HDL-C) levels [14]. In both CKD and end-stage renal disease (ESRD) patients, the serum HDL is not only quantitatively altered but also qualitatively compromised, further impairing its functionality.

HDL is a complex lipoprotein that exhibits significant heterogeneity in the bloodstream. This heterogeneity arises from HDL’s structural characteristics and its involvement in complex metabolic and interconversion processes within the body. HDL is synthesized in the liver and intestines and performs several critical functions in the blood, including reverse cholesterol transport and antioxidant activities [15]. The morphology and size of HDL vary according to its function and are influenced by cholesterol content, lipid components, and protein composition [16]. Additionally, HDL is modulated by various factors such as lifestyle, genetic factors, and medications, all of which contribute to changes in HDL heterogeneity [17].

In this review, we will examine four key aspects: (1) the structure and composition of HDL; (2) the primary functions of HDL; (3) metabolic alterations of HDL in chronic kidney disease (CKD); and (4) the application of HDL in the treatment of CKD.

Structure and composition of HDL

As early as 1955, Havel et al. succeeded in isolating and characterizing lipoproteins with densities ranging from 1.063 to 1.21 g/mL using ultracentrifugation [18]. Based on their findings, this lipoprotein was subsequently defined as HDL. HDL is a complex of lipids and proteins, exhibiting a sophisticated discoidal or spherical structure. The hydrophilic envelope on the surface of HDL particles consists mainly of phospholipids (PL), free cholesterol (FC), and apolipoproteins, whereas the hydrophobic core consists primarily of cholesteryl esters (CE) and triglycerides (TG) [19]. Compared to other lipoproteins, HDL contains a higher proportion of proteins and relatively lower levels of lipids, which is why it has the highest density.

Heterogeneity of HDL structure

Discoidal HDL

In 1971, Trudy Forte and her colleagues made the first observation of HDL particles using electron microscopy, marking a significant milestone in the understanding of HDL structure [20]. In normal individuals, HDL particles typically have a diameter ranging from 70 to 100 Å and a thickness of 35 to 50 Å, with a spherical shape [20]. As illustrated in Fig. 2A, these normal HDL particles are uniformly sized and aggregate into a monolayer structure. However, in patients with lecithin-cholesterol acyltransferase (LCAT) gene defects, HDL exhibits notable heterogeneity, forming disc-like shapes that stack with a repeating spacing of approximately 50 to 55 Å, as depicted in Fig. 2B [20].

Interestingly, this disc-like structure of HDL can also be observed in the plasma of healthy individuals. In 1975, Alan R. Tall reported that incubating isolated apoproteins and phospholipids from healthy HDL in vitro leads to the formation of disc-like particles similar to those depicted in Fig. 2B, which do not stack [21]. In 1977, Jere Segrest proposed that apolipoprotein A-I (ApoA-I) stabilizes the disc-like structure of HDL by encircling its edges, akin to a “bicycle tire” [22, 23]. Subsequently, John Kane’s laboratory identified a subclass in human plasma that migrates more slowly than the major HDL class in agarose gel electrophoresis, positioning itself between HDL and low-density lipoprotein (LDL) [24]. This subclass was therefore named pre-β HDL, also known as nascent HDL or discoidal HDL. Other research teams further confirmed the presence of discoidal HDL, suggesting that free cholesterol initially tends to associate with discoidal HDL [25, 26]. During the 1990s, multiple studies delved into the mechanism of discoidal HDL formation, proposing that lipid-free ApoA-I interacts with macrophage membranes via ATP-binding cassette transporter A1 (ABCA1), binding phospholipids and free cholesterol to form discoidal HDL [27,28,29,30]. Subsequent studies gradually confirmed this formation process [31, 32].

Fig. 2

Electron microscopy image showing the morphology of HDL. A: HDL particles in normal individuals are uniform in size, spherical, and aggregate into a single-layer structure. B: HDL particles in patients with LCAT gene deficiency are disc-like and arranged in a stacked formation. (These two images are extracted from the study by Trudy Forte et al. [20])

Spherical HDL

The free cholesterol within discoidal HDL is converted into cholesterol esters under the action of LCAT. Cholesterol esters are highly hydrophobic, and once formed, they aggregate into the core of the HDL particle. This process drives the transformation of discoidal HDL particles into spherical HDL particles [33,34,35]. Consequently, the levels of discoidal HDL in the plasma of healthy individuals are relatively low. Spherical HDL particles exhibit considerable heterogeneity in diameter, ranging from 7 nanometers (nm) to 14 nm [36]. This heterogeneity is also evident in the significant differences among particles in lipid proportions, protein ratios, and types of apolipoproteins they carry.

The structure of spherical HDL particles can be described using the “droplet” model, where lipids and apolipoproteins are primarily held together by non-covalent forces. The widely accepted model of human HDL particles depicts them as spherical entity with a diameter of approximately 10 nm, where the surface layer is composed of proteins and amphipathic lipid molecules surrounding a nonpolar core of CE and TG (Fig. 3). Many understandings of HDL structure stem from the preparation and study of reconstituted HDL (rHDL). rHDL is a self-assembled mixture of ApoA-I and dimyristoylphosphatidylcholine (DMPC) in aqueous solution, with diameters ranging from 9 to 20 nm, depending on the initial lipid-to-ApoA-I ratio [37, 38]. Research has revealed that discoidal HDL typically contains only two ApoA-I molecules, while spherical HDL contains three ApoA-I molecules. These three molecules are arranged in a “three-leaf clover model,” with each helical domain providing mutual support [39]. Together with lipid molecules, they have formed the spherical structure of HDL particles.

Fig. 3

Schematic diagram of the HDL structure. (Adapted from “Icon Pack - Metabolism”, by BioRender.com (2024). Retrieved from https://app.biorender.com/biorender-templates)

Complexity of HDL composition

Protein components

Compared to other lipoproteins, HDL contains a richer proportion of protein components, including apolipoproteins, enzymes, and lipid transfer proteins, as shown in Table 1. Among these components, apolipoproteins and enzymes play pivotal roles in the metabolic functions of HDL.

Table 1 The protein components of HDL [40]

Major apolipoproteins

ApoA-I is the predominant protein in HDL particles, accounting for approximately 70% of the total HDL protein. Synthesized in the liver and small intestine, ApoA-I features eight α-helical domains, each containing 22 amino acids. These domains’ amphipathic properties facilitate lipid binding [41]. Additionally, ApoA-I contains two repeated 11-amino-acid sequences, which enhance the molecule’s stability and solubility [41]. ApoA-I plays a crucial role in maintaining the structural integrity of HDL particles and promotes reverse cholesterol transport by activating LCAT, thus contributing significantly to the prevention of atherosclerosis [42].

Apolipoprotein A-II (ApoA-II), synthesized primarily in the liver, constitutes about 20% of the total HDL protein. ApoA-II exhibits strong lipophilicity, which enhances the structural stability of HDL particles and prevents the dissociation of lipid-free ApoA-I from HDL [43]. It is hypothesized that ApoA-II may influence the rate of HDL synthesis; however, its association with the reduced serum HDL levels observed in cardiovascular disease patients requires further investigation.

Apolipoprotein E (ApoE), although present in smaller amounts in HDL, plays a crucial functional role. In nascent discoidal HDL, ApoE adopts a “double belt” model, with two ApoE molecules arranged in a reverse parallel conformation around the nano-disk [44]. This conformation facilitates the binding of HDL to cell surface glycosaminoglycans, thereby modulating interactions between lipoprotein particles and cells [45,46,47].

Auxiliary proteins and enzymes

CETP is a highly hydrophobic glycoprotein that primarily facilitates the transfer of CE and TG between HDL, LDL, and VLDL [48]. CETP’s structure includes an N-terminal spherical domain and a long C-terminal region, which together enable CETP to transfer lipids between different lipoprotein types through a “tunneling mechanism.” Specifically, the N-terminal domain of CETP binds to HDL and transfers cholesterol esters to LDL or VLDL via the C-terminal domain [49].

LCAT is the only enzyme capable of maintaining the balance of free cholesterol between peripheral tissues and HDL [50]. Mature LCAT is a glycoprotein composed of an α/β-hydrolase domain, a membrane-binding domain, and a cap domain [51,52,53,54]. LCAT catalyzes the conversion of free cholesterol in plasma to cholesteryl esters, thereby facilitating reverse cholesterol transport. Additionally, LCAT hydrolyzes the oxidized products of phosphatidylcholine and platelet-activating factor, thereby protecting platelet function and maintaining the antioxidant capacity of HDL [55,56,57].

Paraoxonase 1 (PON1) is an esterase and lactonase synthesized by the liver and primarily carried by HDL [58,59,60]. Structural studies reveal that PON1 is an interfacial activated enzyme, anchored to HDL by an N-terminal helix and another amphipathic helix at the active site [62, 63]. HDL particles carrying ApoA-I have a high affinity for PON1, forming a stable complex [62, 63]. However, the mechanisms of how PON1 integrates into HDL are still relatively limited and require further exploration. PON1 reduces lipid peroxides, thereby inhibiting the production of monocyte chemotactic protein 1 (MCP-1) and potentially delaying or reversing the progression of atherosclerosis [64,65,66]. Several animal and cohort studies have confirmed that PON1 stimulates HDL-mediated endothelial nitric oxide production and enhances macrophage cholesterol efflux [67,68,69,70,71,72].

Phospholipases, such as platelet-activating factor-acetylhydrolase (PAF-AH), hydrolyze short-chain pro-inflammatory oxidized phospholipids, helping to reduce inflammatory responses [73]. Additionally, glutathione peroxidase 3 (GSPx-3) exerts antioxidant effects on HDL, protecting biomolecules from oxidative damage [74].

Lipids composition

With the continuous advancement of mass spectrometry (MS) technology, the lipidomics of HDL is also evolving. Currently, over 200 lipid molecules can be analyzed and identified on HDL particles [75, 76]. The main lipid categories of HDL, as shown in Table 2, include phospholipids, sphingolipids, and neutral lipids.

Table 2 The lipid components of HDL [40]

Phospholipids are the most abundant lipids in HDL, constituting approximately 35–50% of the total lipid mass of HDL. Phosphatidylcholine (PC), an essential component of HDL and cell membranes, contains a high proportion of polyunsaturated fatty acid residues—a feature less common in other lipoproteins [77, 78]. The main molecular species of PC in HDL include 16:0/18:0, 18:0/18:2, 16:0/20:4, and 16:0/18:1 [79,80,81]. Lysophosphatidylcholine (LPC), a product of LCAT-mediated hydrolysis of PC, accounts for 15% of the phospholipid mass [77, 79, 82]. Additionally, phosphatidylethanolamine (PE) and phosphatidylinositol (PI) are also significant phospholipids in HDL, with PE being notably important in plasma due to its antioxidant properties [77, 83,84,85,86,87,88]

Sphingolipids also play a significant role in HDL. Sphingomyelin (SM) is the most abundant sphingolipid in HDL, and its content affects lipid membrane properties and the surface pressure of lipoproteins, thereby regulating the activity of associated proteins [89,90,91]. Ceramide, although present in smaller amounts, also contributes to HDL structure, with 16:0 ceramide being particularly common [77, 79]. Sphingosine-1-phosphate (S1P) is another important sphingolipid in HDL, and its presence is crucial for maintaining the stability of HDL’s biological functions by modulating cell proliferation, migration, and apoptosis [92,93,94]

Neutral lipids, such as cholesterol esters and triglycerides, also play a crucial role in HDL. Cholesterol esters account for 30–40% of the lipid content in HDL, with the majority existing as cholesteryl linoleate, which is vital for HDL’s structural and functional roles [77, 89]. Triglycerides, derived from very low-density lipoprotein (VLDL) and LDL, exhibit similar strong hydrophobic properties as cholesterol esters. The principal types of triglycerides include oleic acid, palmitic acid, and linoleic acid, which play essential roles in lipid metabolism and transport [77]

Complexity of HDL subclasses and analytical methods

In 1954, Gofman and his colleagues first described the differences among HDL subclasses using ultracentrifugal analysis, revealing the diversity of HDL in lipid and protein composition [95]. This complexity of HDL is not only reflected in the classification of its subclasses but also in the identification and measurement of HDL by various analytical techniques

Traditional classification

Based on density, HDL is divided into HDL2 and HDL3: HDL2 has a density of 1.063–1.125 g/mL and is rich in lipids, while HDL3 has a density of 1.125–1.21 g/mL and is higher in protein content [95]. Using non-denaturing gradient gel electrophoresis (GGE), HDL2 and HDL3 can be further subdivided into multiple subclasses, including HDL2a, HDL2b, HDL3a, HDL3b, and HDL3c, with diameters of 8.8–9.7 nm, 9.7–12.0 nm, 8.2–8.8 nm, 7.8–8.2 nm, and 7.2–7.8 nm, respectively [96]

Electrophoresis-based classification

HDL carries surface charges and can be classified into α-HDL and pre-β HDL through agarose gel electrophoresis (AGE) [97, 98]. Most circulating HDL is α-HDL, while pre-β HDL is primarily discoidal. By combining agarose gel electrophoresis with GGE, two-dimensional electrophoresis can be performed, further subdividing HDL into α1, α2, α3, α4, preα1, preα2, preα3, preβ1, and preβ2 [19, 99]

Protein composition-based classification

HDL can also be classified according to the protein composition of its particles into HDL containing only ApoA-I and HDL containing both ApoA-I and ApoA-II [100]. HDL particles containing both ApoA-I and ApoA-II are generally more compact than those containing only ApoA-I [101]

Nuclear magnetic resonance (NMR)-based classification

NMR is a technique based on the physical phenomenon of atomic nuclei absorbing and emitting electromagnetic radiation in an external magnetic field. In the analysis of HDL, particles of different sizes exhibit distinct NMR signal characteristics, enabling the differentiation of various HDL subclasses based on these signals. By analyzing NMR spectra, the diameter range of HDL particles can be determined, classifying HDL into large, medium, and small subclasses [102]

Despite the increasing diversity of methods for studying HDL subclasses and the deepening understanding of HDL, significant challenges remain. Differences in the nomenclature, number, composition, and measurement parameters of HDL subclasses exist across various techniques. Some studies focus on measuring the concentration of HDL subclasses, while others emphasize the percentage of proteins or cholesterol. These discrepancies make it difficult to compare results across different studies. To promote consistency and comparability in HDL research, there is an urgent need to standardize the classification and measurement methods of HDL subclasses

HDL and reverse cholesterol transport (RCT)

HDL plays a crucial role in various physiological processes, with its most significant function being RCT. RCT is a mechanism that transports cholesterol from peripheral tissues back to the liver and is considered the primary anti-atherosclerotic function of HDL

ABCA1 and HDL formation

ABCA1 is an integral membrane protein that utilizes ATP as energy to transport lipids from the intracellular environment to the cell membrane [103]. ABCA1 is found in several cellular locations, including lysosomes, the Golgi apparatus, and the plasma membrane [104]. Under the influence of ABCA1, cholesterol is transported from lysosomes to the Golgi apparatus and then to the plasma membrane. This process helps remove excess cholesterol from cells, influencing various cellular metabolic processes such as endogenous cholesterol synthesis, LDL receptor expression, and cholesterol ester transport [104]. ABCA1 transfers cholesterol and phospholipids from the inner leaflet of the plasma membrane to the outer leaflet, where they bind with lipid-free ApoA-I from the liver and intestine, forming discoidal HDL (Fig. 4) [105,106,107]. This process marks the initiation of RCT, with ABCA1 being pivotal in the formation of discoidal HDL

Fig. 4

The Process of HDL-Mediated Reverse Cholesterol Transport. The liver and intestines synthesize lipid-free ApoA-I, which initially acquires cholesterol and phospholipids from macrophages via ABCA1, forming nascent discoidal HDL. Subsequently, LCAT esterifies the cholesterol, rendering it fully hydrophobic and causing its migration to the core of the discoidal HDL, promoting its maturation into spherical HDL. Meanwhile, membrane proteins such as ABCG1 and SR-BI continue to transfer intracellular cholesterol to HDL, further aiding in HDL maturation. HDL2 and HDL3, the two subclasses of HDL, can interconvert. Next, HDL exchanges cholesteryl esters and triglycerides with VLDL and LDL in equimolar proportions. The final step of reverse cholesterol transport involves the selective uptake of cholesteryl esters from HDL particles by the hepatic SR-BI or the clearance of cholesteryl esters from VLDL and LDL via the LDL receptor (LDLR). (Adapted from “Icon Pack - Lipid”, by BioRender.com (2024). Retrieved from https://app.biorender.com/biorender-templates)

HDL maturation and cholesterol esterification

After the formation of discoidal HDL, LCAT, activated by ApoA-I, esterifies the cholesterol [108]. The newly synthesized cholesterol esters are completely hydrophobic and gradually migrate to the core of the discoidal HDL, facilitating its transformation into mature spherical HDL. Meanwhile, membrane proteins such as ATP-binding cassette transporter G1 (ABCG1) and scavenger receptor class B type I (SR-BI) continue to transport intracellular cholesterol to HDL, further promoting HDL maturation (Fig. 4). ABCG1 is also an integral membrane protein with similar cellular localization to ABCA1, but it primarily targets mature HDL [109]

Cholesterol transport and clearance

Subsequently, HDL exchanges cholesterol esters and triglycerides with VLDL and LDL in an equimolar ratio [110]. The final step of RCT is mediated by SR-BI, a membrane protein of the CD36 family responsible for the selective uptake of cholesterol esters from HDL (Fig. 4) [111]. Through SR-BI, the liver absorbs cholesterol esters from HDL and hydrolyzes them into free cholesterol. The absorbed cholesterol can be converted into bile acids and stored in the gallbladder. During digestion, bile acids are released into the small intestine to aid in the digestion and absorption of fats. Additionally, the liver can utilize cholesterol for the synthesis of cell membranes and steroid hormones

How does CKD affect HDL metabolism?

In the previous sections, we have thoroughly discussed the physiological functions of HDL and its critical role in RCT. However, HDL metabolism is not always stable; its function and structure can undergo significant changes, particularly in certain disease states. CKD is one such pathological condition that can markedly impact HDL metabolism. In the following discussion, we will explore how CKD interferes with HDL metabolic processes through various mechanisms, thereby altering its protective role against atherosclerosis.

Structural interference

In HDL metabolism research, traditional perspectives have primarily focused on the liver, as it is the central site for HDL catabolism. However, recent studies suggest that the kidneys also play a significant role in HDL metabolism—a role that is not yet fully understood. Specifically, research has shown that the kidneys do not directly participate in the clearance of intact HDL particles but instead selectively remove specific HDL components, such as ApoA-I and ApoE [112]. This finding implies that the reduced plasma HDL levels observed in CKD patients may be related to this selective clearance mechanism of the kidneys. In patients with CKD, impairment of the glomerular filtration barrier allows apolipoproteins to be lost in the urine, which results in proteinuria. Therefore, the presence of proteinuria is considered a key indicator of apolipoprotein loss [113,114,115]. Additionally, CKD significantly affects the activity of enzymes involved in HDL metabolism, with changes in enzyme activity influencing the maturation and remodeling of HDL particles, ultimately altering HDL composition and function [116,117,118]. Therefore, the role of the kidneys in HDL metabolism extends beyond filtration, involving selective component removal and regulation of enzyme activity. This discovery offers a new perspective for understanding dysregulated plasma lipid metabolism in CKD patients.

During the progression of CKD, patients often experience complex changes in lipoprotein metabolism, profoundly affecting overall lipid metabolism. Notably, the lipid composition of HDL particles undergoes significant alterations, impacting not only their structure but also their function and metabolic pathways. The most prominent changes include an increase in TG content and a decrease in PL content within HDL particles. These alterations are closely associated with reduced hepatic lipase (HL) activity and increased phospholipid transfer protein (PLTP) activity [116, 119]. HL is a 65-kDa glycoprotein primarily synthesized and secreted by hepatocytes [120]. As a member of the triglyceride lipase family, HL acts as an important negative regulator of HDL levels by reducing the triglyceride and phospholipid content of HDL [121, 122]. Its decreased activity may hinder TG metabolism within HDL particles, leading to the accumulation of TG. PLTP is a member of the lipid transfer protein family. The N-terminal and C-terminal domains of PLTP are structurally connected, forming a hydrophobic channel that facilitates lipid molecule transport [123, 124]. PLTP is predominantly released by the liver, adipose tissue, and macrophages in plasma, playing a crucial role in mediating the exchange of phospholipids between lipoproteins [125, 126]. Simultaneously, upregulation of PLTP activity may promote the transfer of PL from HDL particles, resulting in reduced PL content, which negatively affects HDL functionality and stability. Furthermore, the protein composition of HDL particles also undergoes significant changes. Levels of ApoA-I, ApoA-II, and ApoM tend to decrease, while levels of serum amyloid A (SAA), ApoC-II, ApoC-III, and ApoA-IV increase markedly [127]. These changes not only reflect substantial adjustments in the composition of HDL particles but may also have far-reaching effects on HDL’s physiological functions, including its antioxidant, anti-inflammatory, and anti-atherosclerotic properties.

Impact of HDL subclasses

The changes in HDL subclasses in patients with kidney disease represent a complex and nuanced area of research. In a study by Alabakovska et al., it was observed that the level of the HDL2b subclass decreased in ESRD patients, while the level of the HDL3c subclass increased [128]. This finding suggests that kidney disease affects HDL subclasses not only in terms of quantity but also potentially through the selective regulation of specific subclasses. In addition, multiple studies have demonstrated a significant reduction in the levels of the HDL2 subclass in ESRD patients, highlighting the adverse impact of ESRD on HDL subclass distribution [129, 130]. These effects include an increase in PL, unesterified cholesterol, and TG within HDL2, as well as a decrease in apolipoproteins in HDL3, along with reductions in both the concentration and activity of PON1 [129, 130].

However, research on patients with proteinuria shows different results compared to ESRD patients. Soto-Miranda et al. found that patients with proteinuria had a higher proportion of large HDL2b subclasses and a lower proportion of small HDL3c subclasses [131]. This trend differs from the one observed in ESRD patients, suggesting that different types or severities of kidney injury may have varying effects on HDL subclass distribution. However, the specific mechanisms by which kidney injury leads to these divergent changes remain unclear.

On the other hand, some studies also support the observation of a relative increase in large HDL subclasses in kidney disease patients [132,133,134]. This observation indicates that despite a decrease in total HDL levels, there may be compensatory mechanisms, particularly involving large HDL subclasses, to maintain HDL functionality and integrity even in the face of lipid metabolism disorders associated with kidney disease.

Overall, understanding HDL subclass changes in kidney disease patients is still in its early stages. Future research needs to further explore the impact of different types of kidney lesions on HDL subclasses and elucidate the underlying biological mechanisms to better understand and manage lipid metabolism abnormalities in these patients.

Functional alterations

Several studies have demonstrated that the RCT capacity of HDL is significantly reduced in uremic patients undergoing dialysis [135,136,137]. This phenomenon not only reveals severe lipid metabolism abnormalities in these patients but also suggests a substantially increased risk of cardiovascular disease. The RCT capacity of HDL depends on the binding of cholesterol to HDL particles, a process closely associated with phospholipid levels, particularly the content of PC [138,139,140,141,142]. Research has shown that co-incubation of HDL with PC emulsions can significantly enhance HDL lipidation, thereby promoting cholesterol uptake from macrophages [138]. Furthermore, it has been reported that the infusion of PC emulsions containing ApoA-I into patients with severe atherosclerosis can improve angiographic parameters [143]. These findings suggest that alterations in phospholipid levels may have a critical impact on the cholesterol efflux capacity of HDL. However, clinical studies on phospholipid level changes in HDL among patients with kidney disease remain relatively scarce.

Impairment of HDL RCT capacity may be associated with decreased levels of specific apolipoproteins, such as ApoA-I, ApoC-III, and ApoD [144,145,146]. In ApoE-deficient mouse models, studies have shown a negative correlation between ApoE levels and HDL RCT capacity [144]. Additionally, elevated SAA levels may also be related to the reduced RCT capacity of HDL. Tsun et al. demonstrated that in patients with type 2 diabetes and early or severe kidney disease, increased SAA levels were associated with impaired cholesterol efflux capacity mediated by SR-BI [147]. Similarly, Beer et al. reported that SAA could adversely affect certain steps in the RCT pathway during acute inflammation [148]. In summary, these studies suggest that the decline in HDL RCT capacity in patients with kidney disease may be closely related to changes in phospholipid levels and the reduction of specific apolipoproteins. Further research in this area will help to better understand and manage cardiovascular risk in CKD patients.

Further research has revealed that patients with CKD exhibit significantly reduced expression levels of ABCA1, leading to a decrease in ABCA1-mediated cholesterol efflux capacity [149]. As a critical transport protein, the reduced expression of ABCA1 not only affects intracellular cholesterol efflux but may also have profound implications for overall cholesterol metabolism. Investigating the specific mechanisms underlying the reduced expression of ABCA1, scientists have discovered that this phenomenon may be associated with post-translational modifications induced by the binding of myeloperoxidase (MPO) to ApoA-I [150]. Specifically, MPO can generate cyanate (OCN−), which is closely linked to protein carbamylation [151]. Additionally, OCN − is a decomposition product of urea, which is particularly significant in kidney disease, as plasma levels of both urea and carbamylated proteins are significantly elevated in kidney disease patients [152]. Research indicates that the presence of just one carbamylated lysine residue on ApoA-I associated with HDL particles is sufficient to influence pathways related to ABCA1 and SR-BI, which induces cholesterol accumulation in macrophages [150]. This finding offers a new perspective on the lipid metabolism disorders observed in CKD patients.

HDL typically exerts protective effects on blood vessels, but this protection is significantly diminished in the context of CKD. Symmetric dimethylarginine (SDMA) has been identified in the HDL of CKD patients, and this compound can transform normal HDL into dysfunctional HDL [153]. As an endogenous inhibitor of endothelial nitric oxide synthase (eNOS), SDMA impairs the protective effects of HDL on the vascular endothelium by inhibiting nitric oxide production in endothelial cells [153]. Endothelial damage can be observed even in the early stages of CKD, and this damage progressively worsens as renal function declines [154]. Endothelial injury is a key precursor to atherosclerosis, and early endothelial dysfunction may lead to more severe cardiovascular disease.

Moreover, in CKD patients, HDL reduces nitric oxide availability in endothelial cells via Toll-like receptor 2 (TLR-2) [153]. This mechanism not only impairs the repair capacity of the endothelium but also increases the risk of pro-inflammatory activation and elevates arterial blood pressure. These physiological changes suggest that in CKD patients, HDL not only loses its protective effects but may even have detrimental impacts on vascular health. This phenomenon is particularly pronounced in patients undergoing dialysis, as the dialysis process may further impair HDL function, minimizing its protective effects on the vascular endothelium. Studies have also shown that the protective function of HDL on the vascular endothelium is weakest in dialysis patients but partially recovers after kidney transplantation [154]. Kidney transplantation significantly improves renal function, thereby partially restoring the normal function of HDL. This recovery may be related to the improvement in post-transplant inflammatory status and changes in HDL composition. However, kidney transplantation does not fully restore HDL’s protective function, indicating a continued need for research into therapeutic strategies targeting HDL dysfunction.

Oxidative stress is closely associated with the progression of CKD. Under normal physiological conditions, HDL reduces the formation of oxidized low-density lipoprotein (ox-LDL) through antioxidant enzymes such as PON1 and glutathione peroxidase, thereby decreasing oxidative stress and inflammatory responses [155,156,157]. These antioxidant enzymes enable HDL to effectively protect the vascular endothelium and inhibit atherosclerosis. However, in CKD patients, particularly those undergoing dialysis, the antioxidant activity of HDL is significantly reduced. For example, Moradi et al. found that the antioxidant capacity of HDL in dialysis patients is markedly lower compared to that in healthy controls [155].

Other studies have shown that the activity of PON1 in HDL2 and HDL3 in the plasma of CKD patients is significantly reduced, indicating a substantial decrease in their antioxidant protective capacity [129]. Additionally, it has been found that the decline in PON1 antioxidant activity is associated with the carbamylation of lysine 290 (K290) [158]. The reduction in antioxidant capacity not only signifies a diminished ability of HDL to clear ox-LDL but also suggests increased levels of free radicals and peroxides in these patients, further exacerbating oxidative stress and inflammatory responses. These changes highlight the severe dysregulation of lipid metabolism and antioxidant protection in CKD patients.

These findings underscore a critical point: under conditions of inflammation and oxidative stress in CKD patients, HDL does not always fulfill its atheroprotective role. Instead, the decline in HDL function may make CKD patients more susceptible to cardiovascular events such as heart disease and stroke. In this context, HDL not only loses its protective effects but may also have a detrimental impact on vascular health.

Application and challenges of HDL in the treatment of CKD

Changes in serum HDL levels and impaired HDL function are closely associated with CKD. This phenomenon suggests that pharmacological modulation or improvement of HDL metabolism could have potential benefits for the treatment of CKD.

ApoA-I mimetics

ApoA-I, as the primary protein component of HDL particles, has long been considered a promising therapeutic target. Research from the 1990s demonstrated that overexpression of ApoA-I and its infusion had protective effects in animal models, effectively preventing the formation of atherosclerotic lesions [159,160,161]. This finding was later applied to the renal field, where it was discovered that overexpression of ApoA-I significantly reduced serum creatinine levels in a lipopolysaccharide-induced renal injury mouse model [162]. Other studies have confirmed that infusion of ApoA-I mimetic peptides not only improved glomerular filtration rate but also reduced tubular damage and fibrosis [163, 164]. These results indicate that modulating ApoA-I levels can effectively alleviate renal injury.

Among ApoA-I mimetic peptides, D-4 F has been shown to reduce oxidized phospholipids and alleviate renal inflammation in atherosclerosis mouse models [165]. Peterson et al. found that D-4 F increases the activity of heme oxygenase-1 in the kidneys, thereby enhancing its antioxidant capacity [166]. Additionally, recent studies on ETC-642, a 22-amino acid ApoA-I mimetic peptide complexed with phospholipids, have demonstrated significant efficacy in sepsis treatment by markedly increasing serum HDL-C levels, reducing renal inflammation, and substantially improving renal function [167].

Enzyme therapy

In patients with CKD at different stages, the concentration or activity of LCAT is typically reduced [14]. Additionally, patients with familial LCAT deficiency (FLD) exhibit significantly lower serum HDL-C levels and often progress to renal failure in their forties [168]. Animal studies also indicate a close association between chronic renal failure and downregulation of the hepatic LCAT gene [169]. These observations suggest that a decline in LCAT activity may be closely related to the occurrence and progression of CKD.

Research by Baragetti et al. found that reduced serum LCAT levels can predict the early progression of CKD and confirmed that changes in HDL subtypes contribute to the worsening of renal damage [170]. Based on these findings, therapeutic targeting of LCAT is considered potentially beneficial for the kidneys. For example, recombinant human LCAT (rhLCAT) therapy has been shown to normalize lipid levels and significantly reduce proteinuria [171]. In vitro studies further confirmed that rhLCAT can reduce the formation of intracellular reactive oxygen species (ROS), thereby restoring the antioxidant capacity of HDL [170].

Additionally, clinical data suggest that decreased PON1 activity in circulation is associated with poor outcomes in CKD, indicating that PON1 may have potential renal protective effects [172,173,174,175,176,177]. However, the exact role of PON1 in CKD pathophysiology remains unclear. No specific drugs targeting PON1 have been reported to date, highlighting the need for further research and development.

rHDL

Infusion of rHDL has shown therapeutic potential for atherosclerosis in CKD patients. CSL-111, an rHDL composed of human ApoA-I and soybean phosphatidylcholine, was demonstrated in 2007 by Tardif et al. through the “Effects of CSL-111 on Atherosclerosis: Safety and Efficacy” trial to improve plaque characteristics and coronary artery scores [178]. However, the trial also revealed that CSL-111 treatment induced liver dysfunction, leading to the termination of subsequent studies [179].

The second-generation recombinant HDL, CSL-112, shows promising prospects and does not induce hepatotoxicity [180]. Phase I clinical trials indicated that CSL-112 significantly increased pre-β HDL levels (by 36-fold) and cholesterol efflux (by 270%), without causing liver damage [180, 181]. In the CSL-112-2001 Phase II trial, CSL-112 also significantly elevated ApoA-I and cholesterol efflux levels in patients with acute myocardial infarction (AMI), including those with moderate CKD [182]. Additionally, a 6-gram dose of CSL-112 demonstrated good safety, but further research is needed to explore its renal protective capabilities [182].

Another rHDL, CER-001, was designed to mimic nascent pre-β HDL [179]. Recent studies have shown that CER-001 normalizes lipoprotein profiles and reduces lipoprotein X (LpX) in patients with rapidly progressive kidney disease and FLD [183, 184]. CER-001 appears to improve or at least slow the decline in renal function by reducing toxic LpX deposition in the kidneys and clearing LpX-induced lipid deposits [183, 184]. Research also indicates that CER-001 significantly improves conditions in patients with acute coronary syndrome (ACS) with extensive atherosclerotic lesions [185].

SGLT-2 inhibitors

Sodium-glucose cotransporter 2 inhibitors (SGLT-2i) were initially developed as treatments for type 2 diabetes [186]. However, subsequent studies have revealed significant renal protective benefits of SGLT-2i [187, 188]. In two large prospective studies focused on renal outcomes—the “Dapagliflozin and Prevention of Adverse Outcomes in Chronic Kidney Disease (DAPA-CKD)” and the “Canagliflozin and Renal Disease in Diabetes (CREDENCE)” trials—SGLT-2i treatment was associated with a 30–40% reduction in the risk of all renal endpoints. These endpoints include the occurrence and/or progression of albuminuria, decline in eGFR, end-stage renal disease (dialysis or transplantation), and renal death [187, 189].

Systematic reviews and meta-analyses have demonstrated that SGLT-2i provide substantial cardiovascular and renal outcome protection in CKD patients, supporting their continued use even with declining renal function [190]. Based on these findings, the Kidney Disease: Improving Global Outcomes (KDIGO) organization recommends SGLT-2i as the only drug class with Level 1 A evidence for diabetic CKD patients [191]. Despite the observed renal protective effects of SGLT-2i, their underlying mechanisms remain not fully understood.

A study by Li et al. explored the relationship between SGLT-2i, atrial fibrillation, and circulating metabolites, finding that SGLT-2i significantly impacts HDL subtype metabolism [192]. Specifically, SGLT-2i significantly increased HDL concentrations, particularly small HDL subtypes, while decreasing very large HDL levels [192]. Research by Fadini also supports this finding, showing an increase in small HDL subtype levels and a decrease in large HDL subtype levels after three months of dapagliflozin treatment [193]. Further studies have confirmed that SGLT-2i increases serum HDL-C concentrations [194,195,196,197,198]. These findings suggest that SGLT-2i have a regulatory effect on HDL, potentially providing new insights into the renal protective mechanisms of SGLT-2i for future research.

GLP-1 RAs

In the “2021 Standards of Medical Care in Diabetes,” the American Diabetes Association (ADA) recommends considering glucagon-like peptide-1 receptor agonists (GLP-1 RAs) in addition to SGLT2i for patients with high risk or established CKD [199]. GLP-1, an incretin hormone, improves glucose control through various mechanisms, including stimulating insulin secretion, enhancing pancreatic β-cell mass, inhibiting glucagon secretion, and slowing gastric emptying [200]. As GLP-1 mimetics, GLP-1 RAs have been shown to have significant effects on glycemic control, reduction of cardiovascular events, and renal protection [201,202,203,204,205,206,207].

Clinical trials have demonstrated the positive impact of GLP-1 RAs on renal health. For example, in a large randomized, double-blind, controlled trial conducted by Muskiet et al., liraglutide was found to help slow the increase in the urine albumin-to-creatinine ratio in patients with significant albuminuria [208]. Additionally, a prospective study in Japan revealed that liraglutide significantly improved eGFR, particularly in patients with eGFR ranging from 30 to 60 mL/min/1.73 m² [209]. Two other studies further confirmed the renal protective effects of liraglutide, showing significant reductions in proteinuria and 24-hour urinary protein excretion rate [210, 211]. These results indicate that liraglutide not only improves renal function but also effectively reduces markers of renal damage.

Despite the widespread recognition of the renal protective effects of GLP-1 RAs, research on their impact on serum HDL-C levels remains limited. One study found that while GLP-1 RAs combined with metformin resulted in a slight increase in serum HDL-C levels, this change was not statistically significant [202]. This suggests that the potential effects of GLP-1 RAs on HDL-C are not yet fully understood. Therefore, further research is needed to explore the mechanisms by which GLP-1 RAs affect HDL-C and to evaluate their comprehensive role in the prevention and treatment of CKD, to better understand the overall benefits of these medications in CKD management.

Conclusions and outlook

This review explores the heterogeneity of HDL, the role of HDL in RCT, and the metabolic changes of HDL in CKD.

The kidneys affect HDL levels and its maturation and function by selectively removing HDL components and altering enzyme activity. These changes result in significant alterations in HDL structure and function, such as increased TG content, decreased PL content, and changes in HDL subtype distribution. These impacts not only weaken HDL’s anti-atherosclerotic effects but may also increase cardiovascular risk in CKD patients.

Furthermore, HDL function in CKD patients significantly declines, leading to impaired cholesterol reverse transport capability, which is associated with reduced PL levels and decreased apolipoproteins (e.g., ApoA-I) in HDL. The loss of HDL’s protective function for endothelial cells may exacerbate endothelial damage and oxidative stress, further increasing cardiovascular disease risk.

In terms of treatment, ApoA-I mimetics and rHDL have shown potential in improving HDL function and reducing renal inflammation, but their long-term safety and efficacy need further validation. SGLT-2i and GLP-1 RAs also show promise in improving cardiovascular and renal outcomes, but the specific mechanisms of their effects on HDL metabolism remain to be clarified.

Future research should focus on the mechanisms by which CKD affects HDL function and structural changes to better understand its role in CKD. Additionally, novel therapeutic approaches, such as targeted therapies for HDL function, should be explored, along with further investigation into the impact of SGLT-2i and GLP-1 RAs on HDL metabolism, to optimize treatment strategies. These studies hold the potential to provide more effective treatment options for CKD patients, improving their quality of life and prognosis.