Inflammation modulates human HDL composition and function in vivo

Abstract

Objectives: Inflammation may directly impair HDL functions, in particular reverse cholesterol transport (RCT), but limited data support this concept in humans.

Methods and results: We employed low-dose human endotoxemia to assess the effects of inflammation on HDL and RCT-related parameters in vivo. Endotoxemia induced remodelling of HDL with depletion of pre-β1a HDL particles determined by 2-D gel electrophoresis (-32.2±9.3% at 24 h, p<0.05) as well as small (-23.0±5.1%, p<0.01, at 24 h) and medium (-57.6±8.0% at 16 h, p<0.001) HDL estimated by nuclear magnetic resonance (NMR). This was associated with induction of class II secretory phospholipase A2 (~36 fold increase) and suppression of lecithin:cholesterol acyltransferase activity (-20.8±3.4% at 24 h, p<0.01) and cholesterol ester transfer protein mass (-22.2±6.8% at 24 h, p<0.001). The HDL fraction, isolated following endotoxemia, had reduced capacity to efflux cholesterol in vitro from SR-BI and ABCA1, but not ABCG1 transporter cell models.

Conclusions: These data support the concept that "atherogenic-HDL dysfunction" and impaired RCT occur in human inflammatory syndromes, largely independent of changes in plasma HDL-C and ApoA-I levels.

Figure 1

Effects of endotoxin (3ng/kg, iv) on (A) HDL sub-populations by nuclear magnetic resonance spectroscopy (*p<0.05, **p<0.01, ***p<0.001 vs time 0, analyzed by repeated measures one-way ANOVA) and (B) selective ApoA-I containing subpopulations of HDL particles by 2D gel electrophoresis (**p<0.01, ***p<0.001 vs pre-LPS, analyzed by repeated measures one-way ANOVA). Data presented as mean ± SEM (n=20).

Figure 2

Plasma levels of (A) secretory phospholipase A₂ and endothelial lipase (***p<0.001 vs. time 0, assessed by repeated measures one-way ANOVA), (B) plasma cholesterol esterification rate (nmols/h/ml plasma), an index of lecithin:cholesterol acyltransferase (LCAT) activity, and (C) cholesterol ester transport protein levels (**p<0.01 vs pre-LPS; assessed by t-test). Data presented as mean ± SEM (n=20).

Figure 3

(A) ABCA1-dependent cholesterol efflux (over 4h) to HDL fraction during endotoxemia (**p<0.01 vs. pre-LPS efflux, n=12, assessed by repeated measures one-way ANOVA). (B) ABCA-1 efflux correlations (r²) with lipid parameters pre-LPS and 24h post LPS. Figure 3A data, published previously, is presented for comparative interpretation with Figure 3B and Figure 4.

Figure 4

(A) SR-BI-dependent cholesterol efflux (over 4h) to HDL fraction during endotoxemia (*p<0.05, **p<0.01, ***p<0.001 vs. pre-LPS, n=20, assessed by repeated measures one-way ANOVA). (B) SR-BI efflux correlations (r²) with lipid parameters pre-LPS and 24h post LPS.