HDL-associated vitamin D binding protein levels are inversely associated with necrotic plaque burden in psoriasis

Abstract

Background and aims: Vitamin D binding protein (DBP) serves a dual function as a vitamin D carrier and actin scavenger. Free DBP is present in high concentrations in serum, while a smaller pool is bound to lipoproteins like HDL and VLDL. The role of DBP's interaction with lipoproteins remains unclear. Given that HDL has been proposed to have both atheroprotective and anti-inflammatory properties, we sought to compare whether HDL-associated DBP and/or total serum DBP could serve as useful biomarkers for assessing disease severity in psoriasis and cardiovascular disease.

Methods: Psoriasis (PSO) patients (N = 83), which were part of a prospective, observational cohort and non-psoriasis (non-PSO) subjects (n = 35) underwent blood collection for HDL purification by liquid chromatography and CCTA scans to assess coronary plaque burden. Serum and HDL-bound DBP levels were measured by ELISA.

Results: The psoriasis cohort was middle-aged (mean ± IQR: 50 (38-59), predominantly male (n = 55, 66 %) and had moderate-to-severe skin disease [psoriasis area severity index score, PASI score, med (IQR): 9.6 (6-18.3)]. Consistent with our previous reports, PSO patients had significantly higher Framingham Risk Score (FRS), high sensitivity C-reactive protein (hs-CRP), Body Mass Index (BMI), insulin resistance (HOMA-IR) and total coronary plaque burden, driven by the rupture-prone non-calcified necrotic core. However, while the concentration of serum DBP (S-DBP) between PSO and non-PSO was unchanged (PSO: 177.80 (125.77-250.99) vs non-PSO: 177.74 (104.32-254.04), the concentration of DBP associated with HDL (HDL-DBP) was decreased in psoriatics (PSO μg/ml: 1.38 (0.64-2.75) vs non-PSO: 1.72 (1.18-3.90). Although both S-DBP and HDL-DBP levels showed inverse correlations with a measure of skin disease severity (PASI) (S-DBP, Rho = -0.022 vs HDL-DBP, Rho = -113), only HDL-DBP exhibited an inverse relationship with necrotic plaque burden [Rho -0.226, p = 0.085 vs S-DBP (0.041, p = 0.76)]. This relationship was strengthened after adjusting for traditional cardiovascular risk factors such as age and sex (β = -0.237, p = 0.045), FRS (β = -0.295, p = 0.033) and including biological treatment and HDL-cholesterol (β = -0.213, p = 0.048).

Conclusions: In conclusion, we found HDL-DBP levels may better capture the severity of psoriatic disease and association with cardiovascular risk factors than S-DBP.

Keywords: Coronary artery disease; High density lipoprotein; Imaging; Inflammation; Psoriasis; Vitamin D binding protein.

Graphical abstract

Fig. 1

Identification of DBP in isolated High Density Lipoproteins Elution profiles from two NGC size exclusion columns in series (a). Collected fractions (18–29) were separated by SDS-PAGE gel (4–12 % NuPAGE™ Bis-Tris, Thermo Fisher Scientific) and stained with Coomassie blue (SimplyBlue™ Safestain, Thermo Fisher Scientific) (b). A second SDS-PAGE gel using the fractions in (b) was probed with a monoclonal antibody to Apolipoprotein A1 (c). FPLC fractions (20–28) in (b) were treated with Cleanascite™ lipid removal agent, separated with by SDS-PAGE gel (4–12 % NuPAGE™ Bis-Tris, Thermo Fisher Scientific) and stained with Coomassie blue (d). A second SDS-PAGE gel using the Cleanascite™ treated fractions in (d) was probed with a monoclonal antibody to Apolipoprotein A1 (e). Phospholipid analysis of FPLC fractions (18–29) ± Cleanascite™ treatment (f). FPLC fractions (22–28) ± Cleanascite™ treatment assessed for DBP levels by ELISA (g). The indicated amounts of recombinant DBP were treated with Cleanascite™ or phosphate-buffered saline control. Supernatent was isolated and assessed for DBP levels by ELISA (h).