1. Introduction
Cardiovascular disease (CVD) represents the primary cause of death globally [
1], with its incidence continuing to rise in parallel with the upsurge of obesity and type 2 diabetes mellitus (T2DM) [
2,
3]. In this context, the metabolic aberrations driven by obesity and T2DM have a pivotal role in disrupting the homeostasis of the circulating lipid profile which, in turn, is strongly associated with CVD. In particular, a rise in circulating triglycerides along with LDL-cholesterol, particularly small-dense-LDL-cholesterol, and a decrease in HDL-cholesterol increases CVD risk [
4]. While HDL-cholesterol has been traditionally deemed to be cardio protective, the relationship between HDL-cholesterol and cardiovascular mortality is U-shaped, with both low and high circulating HDL-cholesterol levels being associated with CVD and mortality [
5,
6,
7]. Thus, the impact of HDL-cholesterol on cardiovascular health is not only dictated by its circulating levels, but also by the functionality of HDL particles, namely, their antioxidant, anti-inflammatory, and cholesterol efflux capacity, which are crucial for HDL lipoproteins to elicit their cardioprotective effects [
8,
9]. In line with this, an impairment in HDL-antioxidant, anti-inflammatory, and cholesterol efflux capacity is intimately linked with CVD [
10,
11,
12,
13,
14,
15]. In addition to HDL functionality, HDL-cholesterol distribution is also emerging as an important player in dictating the effect of HDL lipoproteins on cardiovascular health [
16,
17]. In particular, while an increase in cholesterol distribution in large HDL subfractions has been associated with improved cardiovascular health, the opposite is true for small HDL subfractions [
18,
19,
20]. Indeed, small HDL subfractions have also been shown to be augmented in obese, type 2 diabetic, and insulin resistant individuals who typically harbor an increased CVD risk [
21,
22,
23].
With regard to the modifiable lifestyle factors, diet is central in shaping cardio-metabolic health, with the consumption of highly processed foods promoting the development of obesity and its comorbidities [
24,
25], which, in turn, increase CVD risk [
26]. In this regard, maintaining a positive energy balance, in which energy intake is higher than energy expenditure, is pivotal in fostering these metabolic aberrations by promoting adiposity gain, particularly at the central level [
27,
28]. However, the relationship between diet and cardiometabolic health is not only dependent upon the amount of energy introduced through the diet, but also on the quality of the nutrients consumed [
29]. In particular, the overconsumption of long-chain saturated fatty acids, particularly as part of highly processed foods [
30], has a key role in this context by promoting inflammatory responses [
31] and insulin resistance [
32,
33], as well as increasing circulating LDL-cholesterol [
34], despite not affecting or even increasing HDL-cholesterol levels [
35,
36]. On the contrary, the monounsaturated fatty acids and omega-3 fatty acids have been shown to elicit beneficial effects on cardiometabolic health. Indeed, oleic acid, a monounsaturated fatty acid, has been reported to induce the production and release of large chylomicrons and increase postprandial triglyceride clearance, promoting a shift from small dense LDL to large buoyant LDL lipoproteins [
37]. At the same time, the consumption of the omega-3 fatty acids, eicosatetraenoic acid (EPA) and docosahexaenoic acid (DHA), are known to promote a cardioprotective circulating lipid profile characterized by a decrease in circulating triglycerides [
38], LDL, and an increase in HDL-cholesterol [
39].
Nevertheless, despite anthropometric and metabolic factors, as well as diet quality and energy density, all playing a role in the development and progression of CVD, their relationship with HDL-cholesterol subfraction dimensional distribution remains to be fully elucidated. Thus, this study aims to identify the relationship between the HDL-cholesterol subfraction dimensional distribution and anthropometric, metabolic, as well as nutritional risk factors for CVD in a population of fertile and postmenopausal women.
4. Discussion
The present study provides further support to the positive relationship between cholesterol distribution in small HDL subfractions and CVD risk, in line with previous reports [
18,
19,
20]. In keeping with this, cholesterol distribution in small HDL subfractions displayed a positive relationship with known cardiovascular risk factors for CVD, whereas the opposite occurred for cholesterol levels in large HDL subfractions. Additionally, the data presented herein indicate that dietary lipid intake and energy density may impact upon cholesterol distribution in HDL subfractions. Particularly, an increase in energy intake emerged as being predictive for a decrease in cholesterol distributed within large HDL subfractions, whereas an increase in lipid intake was identified as a predictor of a rise in cholesterol within small HDL subfractions.
Despite HDL-cholesterol being deemed as cardioprotective, its circulating levels are not sufficient to explain its effects on cardiovascular health, especially considering that both low and high levels of plasma HDL-cholesterol are associated with CVD risk and mortality [
5,
6,
7]. In this regard, cholesterol distribution in HDL subfractions may also be pivotal in shaping CVD risk. The data reported herein support the possibility that an increase in cholesterol distribution in large, as opposed to small, HDL-subfractions is associated with a decrease in CVD risk. This notion is corroborated by the fact that the distribution of cholesterol in HDL subfraction 1 is lower in individuals who are overweight, affected by hypertension and insulin resistance, and who have a moderate to high CVD risk score; however, an increase in cholesterol levels in small HDL subfractions are a peculiar characteristic of individuals affected by hypertension and who have increased CVD risk estimated using SCORE2, particularly when considering cholesterol distribution in HDL subfraction 10. Additionally, while cholesterol distribution in large HDL subfractions correlated negatively with known risk factors for CVD disease, the opposite was true for the amount of cholesterol distributed in small HDL subfractions. This finding is in agreement with previous reports, indicating that an increase in cholesterol distribution in small, as opposed to large, HDL subfractions is associated with increased CVD risk [
18,
49], and is observed in individuals affected by Familial Hypercholesterolemia characterized by early CVD [
45]. Despite the observed relationship between cholesterol distribution in HDL subfractions and established CVD risk factors, it is not possible to infer from this or prior studies [
49,
50] whether an increase in cholesterol distribution within large HDL and a decrease in its levels within small HDL subfractions provide cardioprotective effects. Indeed, it remains to be elucidated whether the shift in cholesterol distribution in small HDL subfractions represents a compensatory mechanism driven by an upregulation of nascent HDL synthesis in order to mitigate the CVD risk or if, instead, this may be due to an increase in small HDL at the end of their life cycle.
Therefore, the decrease in cholesterol distribution in large HDL subfractions may represent an adaptive response to the metabolic aberrations associated with body weight gain, central adiposity, and high LDL-cholesterol levels. In particular, it may be a consequence of an increased activity of cholesterol ester transfer protein (CETP), as observed in obese individuals [
51], which mediates the exchange of triglycerides and cholesterol esters between triglyceride rich and LDL, as well as HDL lipoproteins [
52]. This, in turn, results in an enrichment of HDL lipoproteins with triglycerides, which are readily catabolized and cleared with a consequent decrease in large and a concomitant increase in small HDL subfractions, as suggested by the data reported herein. Furthermore, it remains to elucidate the characteristics of these small HDL subfractions, and particularly if they lose their antioxidant and anti-inflammatory properties due to changes in the amount and activity of proteins like Paraoxonase or Myeloperoxidase in their proteome [
9]. These metabolic irregularities may be a direct consequence of increased visceral adiposity which, in turn, is a key driver of insulin resistance [
53,
54,
55]. In this regard, increased free fatty acid release from dysfunctional insulin-resistant visceral adipose tissue depots provide the substrates in order to increase triglyceride synthesis within the liver leading to a subsequent increase in very low-density lipoproteins (VLDL) export. These lipoproteins, in turn, supply triglycerides to HDL lipoproteins in exchange for cholesterol esters, with a consequent decrease in cholesterol-rich HDL2 [
56,
57]. This possibility is also supported by the fact that HDLs, enriched in triglycerides, undergo hydrolysis of their lipid component and lose APO A1. Indeed, as part of this study, it was found that APO A1 correlated positively only with large HDL subfractions. Furthermore, in agreement with this hypothesis, cholesterol distribution in large HDL subfractions correlated negatively with circulating triglycerides and APO B100, even though this apolipoprotein also characterized LDL-C apart from VLDL. Instead, the opposite is true for cholesterol distribution in small HDL subfractions. Thus, visceral adiposity and insulin resistance may be the responsible for the decrease in cholesterol distribution in large HDL subfractions in overweight individuals and those affected by hypertension, insulin resistance, and high LDL cholesterol levels. In support to this possibility, not only HOMA-IR and VAT correlated negatively with cholesterol distribution in large and positively in small HDL subfractions, but VAT was also able to predict cholesterol distribution in HDL subfractions. However, despite the fact that a decrease in cholesterol distribution in large HDL subfractions may potentially indicate a parallel decrease in large HDL subfractions, this cannot be inferred using the Lipoprint system. Indeed, this analytical technique does not directly quantify the amount of HDL subfractions; instead, it directly assesses the level of cholesterol in each on the ten HDL subfractions. In light of this, the data reported herein and in other studies [
16,
18,
50] do not provide a direct readout of the amount of HDL subfractions, but rather the amount of cholesterol they carry. However, changes in cholesterol distribution in HDL subfractions may still represent an additional biomarker of cardiovascular risk.
Diet is a key player is shaping cardiometabolic health [
37,
41,
48], as confirmed by the relationship between dietary parameters and cholesterol distribution in HDL subfractions reported herein. In particular, individuals in the highest tertile for energy intake also displayed an increase in cholesterol distribution in small, and a decrease in large, HDL subfractions, respectively. Additionally, an increase in energy intake was able to predict a decrease in cholesterol distribution in large HDL subfractions. These finding are in agreement with the fact that an energy-restricted low-carbohydrate dietary intervention aimed at eliciting body weight loss was able to increase large HDL subfractions [
58], whereas an increase in adiposity, particularly central adiposity, led to a decrease in cholesterol distribution in large HDL subfractions [
50]. An increase in energy intake, in the absence of a concomitant increase in energy expenditure, shifted the energy balance toward the positive, promoting body weight gain. This is in line with the data reported herein, with overweight individuals displaying a decrease in cholesterol abundance in large HDL subfractions.
Despite energy intake emerging as a key driver of HDL-cholesterol subfraction dimensional distribution, this effect appears to be nutrient specific. Indeed, while the intake of carbohydrates and proteins did not correlate with the distribution of cholesterol in HDL subfractions, total lipid intake correlated positively with cholesterol distribution in small, and negatively in large, HDL subfractions. However, this effect, rather than being dependent upon cholesterol intake, appears to be driven by the intake of dietary fatty acids. Indeed, saturated and monounsaturated fatty acids correlated positively with cholesterol distribution in small HDL subfractions. The fact that both monounsaturated and saturated fatty acid intake are related to cholesterol distribution suggests that this the relationship between lipid intake and the levels of cholesterol in HDL subfractions may not be fatty acid specific, and may depend upon the overall lipid intake. This possibility is supported by the fact that lipid intake, after adjusting for individual fatty acid groups, was able to predict cholesterol distribution in small HDL subfractions independently from energy intake. Nevertheless, polyunsaturated fatty acids did not show any correlation with cholesterol distribution in HDL subfractions. This suggests that the impact of lipid intake on cholesterol distribution may be driven by the sum of monounsaturated and saturated fatty acids, rather than polyunsaturated fatty acids themselves. Saturated fatty acids, and particularly long-chain saturated fatty acids, are considered to be deleterious for cardiovascular health [
59], whereas the opposite is true for monounsaturated fatty acids [
37]. However, the fact that these fatty acids have the same relationship with cholesterol distribution in HDL subfractions is in contrast with the notion that a decrease in cholesterol distribution in large, and an increase in small, HDL subfractions is associated with CVD risk [
18]. The reason for this apparent discrepancy remains to be elucidated. Indeed, while there are reports describing the impact of dietary fatty acids on LDL cholesterol subfractions [
17,
37,
60], to the best of our knowledge, this is the first report describing the relationship between fatty acid quality and HDL-cholesterol subfraction distribution. Nevertheless, these fatty acids may differently modulate HDL lipoprotein functionality [
61], independently of cholesterol distribution, and therefore still exert opposite effects on cardiovascular health.
This study is not without limitations. First, its retrospective nature and the lack of information about pharmacological or dietary intervention prevented us from directly evaluating how the modulation of CVD risk factors reflects upon HDL-cholesterol subfraction dimensional distribution. For this, a longitudinal study setup would be more appropriate. Second, dietary data are derived from 24 h recalls which, despite being a validated method, do not provide information to directly infer a cause–effect relationship between energy, as well as nutrient, intake and cholesterol distribution in HDL subfractions. Finally, the sample size of the cohort represents an additional limitation of the study which, however, was mitigated by the fact that the study’s participants underwent an extensive metabolic characterization. Another aspect that should not be overlooked is related to the fact that the present study only includes female participants. Despite this, the data presented herein are in line with previous reports that included both males and females [
18,
19,
20]. Nevertheless, this study, to the best of our knowledge, is the first to provide evidence of the relationship between dietary intake and cholesterol distribution in HDL subfractions which, instead, represents a strength of the present report.