Diet and C-Reactive Protein
Peter M. Clifton, PhD, FRACP
Address
CSIRO Health Sciences and Nutrition, PO Box 10041 BC,
Adelaide, SA 5000, Australia.
E-mail: peter.clifton@csiro.au
Current Atherosclerosis Reports 2003, 5:431–436
Current Science Inc. ISSN 1523–3804
Copyright © 2003 by Current Science Inc.
Introduction
The acute phase reactant C-reactive protein (CRP) has
gained increasing strength as a predictor of cardiovascular
and cerebrovascular events both in healthy subjects and in
those with known coronary disease [1]. There is also a
growing body of evidence that CRP itself may contribute to
the disease process and that some of the risk factors for
cardiovascular disease may be mediated partially by
increased concentrations of CRP. Thus, environmental
factors that contribute to increased concentrations of CRP
will be important to identify.
C-Reactive Protein and Promotion of Disease
As noted in a previous review by Hielbronn and Clifton
[2], there are several potential mechanisms by which CRP
may enhance atherosclerosis. Since that review, the
evidence has increased considerably.
C-reactive protein in vitro has been found to upregu-
late the angiotensin-1 receptor on vascular smooth muscle
cells and promote migration and proliferation of these
cells, as well as enhance reactive oxygen species formation
[3], albeit at much higher levels than seen in vivo with
atherosclerosis alone (up to 100 mg/L). CRP has been
shown to stimulate the production of plasminogen-
activator inhibitor-1 (PAI-1) antigen and activity in aortic
endothelial cells in culture [4]. Verma et al. [5] showed that
CRP at a relatively high level (25 mg/L) caused a marked
and sustained increase in native low-density lipoprotein
(LDL) uptake by macrophages. These proatherosclerotic
effects of CRP were mediated, in part, via increased secre-
tion of endothelin-1 (ET-1) and interleukin-6 (IL-6)
(P<0.01), and were attenuated by both bosentan (an ET-1
receptor antagonist) and IL-6 antagonism with anti–IL-6
antibodies. CRP also caused a marked increase in the
expression of vascular cell adhesion molecule (VCAM-1),
intracellular adhesion molecule (ICAM-1), and monocyte
chemoattractant protein-1 (MCP-1) in human saphenous
vein endothelial cells. Pasceri et al. [6] showed that normal
levels of CRP (5 mg/L) induce MCP-1 in human umbilical
vein endothelial cells, and that this was blocked by sim-
vastatin and fenofibrate. Thus, it appears to be fairly clear
that CRP may promote atherosclerotic disease and endo-
thelial dysfunction via many mechanisms, and that strate-
gies to lower CRP itself via diet, drugs, and exercise may be
important in disease prevention.
C-reactive protein, interleukin-6, and obesity
There is a substantial body of evidence that CRP levels are
increased in obesity and that this may be partially medi-
ated by increased production of IL-6 from the increased
mass of adipose tissue [2]. Connelly et al. [7] found that
body mass index (BMI) and IL-6 levels predicted CRP
levels in Cree women, whereas waist circumference and IL-
6 levels were predictive in Cree men. The proportion of
subjects with high levels (>3.8 mg/L) was higher in women
than in men (51% vs 32%). Fernandez-Real et al. [8] found
that IL-6 and CRP were correlated in 288 healthy men and
women ( r =0.39 overall), but the association was not
present in smokers (r=0.16) or in women (r=0.11). The
former observation occurs probably because smoking per
se elevates IL-6 [9]. The levels of CRP (3 mg/L) and IL-6
(5.8 to 6.4 pg/mL) were similar in these normal-weight
men and women. If IL-6 from adipose issue is a major
inducer of CRP, then weight loss should lower both IL-6
and CRP in parallel, but the evidence is not entirely consis-
tent, although weight loss invariably lowers CRP [10].
Ziccardi et al. [11] found that IL-6 fell in obese women with
weight loss. However, Laimer et al. [12] found that CRP fell
markedly from 13 mg/L to 4 mg/L in morbidly obese
patients after gastric banding, which induced a fat loss of
C-reactive protein (CRP) is an independent predictor of
cardiovascular events in healthy individuals and those with
pre-existing disease. It also probably contributes to the
disease process. CRP levels are higher in obese subjects
and this link is almost certainly because of increased insulin
resistance. Interventions that alter insulin resistance, such
as weight loss, exercise, and conjugated linoleic acid, also
alter CRP. Glycemic load is associated with CRP, but there
have been no interventions with altered macronutrient
composition. In the context of weight loss, macronutrient
composition is probably not important. Alcohol lowers
CRP, but the mechanism is unknown. The interaction
between gender and obesity needs further work, but
it appears that obesity has a greater effect on CRP levels
in women.