Skip to main content

Table 4 Proportion of the association between diet and lipoprotein concentrations mediated by diet-related plasma proteins in the OmniHeart triala

From: Associations of circulating proteins with lipoprotein profiles: proteomic analyses from the OmniHeart randomized trial and the Atherosclerosis Risk in Communities (ARIC) Study

   

HDL-C

(mmol/L)

Triglyceride (mmol/L)

non-HDL-C (mmol/L)

TC:HDL-C ratio

Name of the protein

Uniprot ID

Entrezgene

symbol

% mediated

P

% mediated

P

% mediated

P

% mediated

P

Protein-rich vs. carbohydrate-rich dietary pattern (reference) (n = 6)

Apolipoprotein M

O95445

APOM

97.6

0.01

—

—

—

—

—

—

Afamin

P43652

AFM

56.9

0.01

—

—

—

—

—

—

Collagen alpha-3(VI) chain

P12111

COL6A3

—

—

25.2

0.002

25.1

< 0.001

—

—

Chitinase-3-like protein 1

P36222

CHI3L1

—

—

20.6

0.004

—

—

—

—

Inhibin beta A chain

P08476

INHBA

—

—

36.0

< 0.001

—

—

—

—

Palmitoleoyl-protein carboxylesterase NOTUM

Q6P988

NOTUM

—

—

41.8

< 0.001

—

—

—

—

Unsaturated fat-rich vs. carbohydrate-rich dietary pattern (reference) (n = 4)

Cathelicidin antimicrobial peptide

P49913

CAMP

—

—

68.9

0.02

43.6

< 0.001

—

—

Guanylate-binding protein 2

P32456

GBP2

—

—

—

—

—

—

70.1

0.002

COP9 signalosome complex subunit 7b

Q9H9Q2

COPS7B

—

—

—

—

33.4

< 0.001

—

—

Sodium-coupled monocarboxylate transporter 1

Q8N695

SLC5A8

—

—

—

—

—

—

96.3

0.004

Protein-rich vs. unsaturated fat-rich dietary pattern (reference) (n = 1)

Sodium-coupled monocarboxylate transporter 1

Q8N695

SLC5A8

73.6

< 0.001

—

—

—

—

—

—

  1. a All diet comparisons lowered high-density lipoprotein (HDL)-cholesterol (C), triglycerides, non-HDL-C and the ratio of total cholesterol and HDL-C. The unsaturated fat-rich vs. carbohydrate-rich dietary patterns increased HDL-C concentrations, and the protein-rich vs. unsaturated fat-rich dietary patterns found no difference in non-HDL-C (Additional file 2: Table S1). Proportion mediated and P-values were calculated using causal mediation analysis. We fit two linear mixed models with random intercepts. The first model evaluated the association between each of the diet comparisons and proteins (mediator). The second model evaluated the association between each of the diet comparisons and lipoprotein outcomes, after controlling for proteins. Diet-related plasma proteins significantly associated with lipoprotein outcomes that had statistically significant mediation are presented. No diet-related protein was significantly associated low-density lipoprotein-C
  2. — indicates that there was no significant association between the diet-related protein and lipoprotein outcome