MINIREVIEW
Molecular aspects of rheumatoid arthritis: role of
environmental factors
Shu Kobayashi, Shigeki Momohara, Naoyuki Kamatani and Hiroshi Okamoto
Institute of Rheumatology, Tokyo Women’s Medical University, Japan
Introduction
Rheumatoid arthritis (RA) is a systemic, chronic
inflammatory disease that affects 0.5–1% of the popu-
lation and causes progressive joint destruction that
leads to the restriction of activities of daily living and
deterioration of quality of life. Although the patho-
genesis of RA has not yet been fully elucidated, it is
considered to be a complex, multifarious disease that
is influenced by both genetic and environmental fac-
tors. Genetic influences that contribute to RA suscepti-
bility have been demonstrated in both studies of twins
[1] and families [2], as well as in genome-wide linkage
scans [3]. These studies estimated that genetic factors
are responsible for 50–60% of the risk of developing
RA and that environmental factors may explain the
remaining risk. This quantification was made by a
classical approach to separating the quantitative
influence of genetic factors in RA with nationwide
studies of twins with RA, as described previously [1].
This minireview focuses on the contribution of envi-
ronmental risk factors to the development of RA.
Infection
Several infectious agents have been reported to be risk
factors for RA (Table 1), including human parvovirus
B19 (B19), Epstein–Barr virus (EBV), retroviruses,
alphaviruses, hepatitis B virus, Mycobacterium tuberculo-

oping RA. Thus, environmental factors may explain the remaining risk of
developing RA. A large variety of environmental factors such as infectious
agents, smoking, sex hormones, pregnancy etc. have been extensively
studied previously. Understanding of how these factors contribute to the
development of RA may lead to the better understanding of pathogenesis
of RA.
Abbreviations
AhR, aryl hydrocarbone receptor; B19, human parvovirus B19; CCP, cyclic citrullinated peptide; EBV, Epstein–Barr virus; HLA, human
leukocyte antigen; IL, interleukin; JCP, Japanese cedar pollinosis; RA, rheumatoid arthritis; RF, rheumatoid factor; TCDD, 2,3,7,8-
tetrachlorodibenzo-p-dioxin; Th, T helper; TNF-a, tumor necrosis factor-a.
4456 FEBS Journal 275 (2008) 4456–4462 ª 2008 The Authors Journal compilation ª 2008 FEBS
clinical spectrum, 25–68% of infections are asymptom-
atic. The most common clinical manifestation of acute
infection is erythema infectiosum, characterized by a
‘slapped cheek’ rash. Interestingly, 50% of adults and
10% of children with erythema infectiosum have joint
manifestations, which tend to occur 2 weeks after
infection. Arthritis caused by B19 infection is usually
symmetrical, primarily affecting the small joints of the
hands, wrists and knees, and it is more commonly
observed in female subjects than in male subjects (as is
RA). Certain evidence suggests that B19 may play a
role in the pathogenesis of RA. For example, B19
arthritis often meets the clinical diagnostic criteria for
RA and is sometimes accompanied by bone erosion
and – at least transiently – by rheumatoid factor (RF)
positivity. Human parvovirus B19 DNA may be
detected in the synovial fluid, synovial cells and ⁄ or
synovial tissue of affected joints. Furthermore, a posi-
tive association with the human leukocyte antigen

in synovial membranes has revealed that the levels of
B19 DNA are similar in RA and control synovial
membranes [11]. Thus, the contribution of B19 infec-
tion in RA remains controversial.
Epstein–Barr virus is a 172-kb, double-stranded
DNA virus. It causes infectious mononucleosis and has
been linked to the development of several malignant
tumors, including Burkitt’s lymphoma, Hodgkin’s
disease and nasopharyngeal carcinoma. In most indivi-
duals, primary infection is asymptomatic. Epstein–Barr
virus first infects B lymphocytes or epithelial cells of the
oropharynx and then activates resting B lymphocytes,
which is followed by proliferation; the virus then estab-
lishes a so-called ‘latent infection’ in memory B lympho-
cytes. A potential role of EBV in RA pathogenesis has
been proposed for many years [12]. For example, it has
been reported that sera from RA patients contains high-
titer antibodies to latent and replicative EBV antigens,
such as Epstein–Barr nuclear antigen, viral capsid anti-
gen and early antigen [13,14]. Furthermore, RA patients
tend to have more EBV-infected B lymphocytes in their
peripheral blood than normal control individuals, and
the mean frequency of spontaneously transforming B
lymphocytes is five times higher in RA patients than in
controls. The QKRAA amino acid sequence of the
HLA-DRB1*0401 allele, which is associated with RA
disease susceptibility, is also found on the EBV glyco-
protein gp110 [15], a major replicative antigen impli-
cated in the control of infection and that is expressed
on the surface of lymphoblastoid cell lines and on the

Healthy individuals with prior EBV infection have
serum antibodies to gp110, and their T lymphocytes
recognize peptides containing the QKRAA motif. gp110
is a target for both T lymphocytes and certain anti-
bodies, recognition of which allows control of infection.
Conversely, HLA-DRB1*0404, an RA predisposing
allele, is associated with low T-lymphocyte responses to
gp110. In RA patients, poor control of EBV replication
may result in a higher EBV load. In situ hybridization
experiments have demonstrated the presence of EBV-
encoded RNAs in synovial tissue B lymphocytes from
RA patients, but not in tissue from osteoarthritis
patients. Furthermore, semiquantitative PCR tech-
niques have been used to show that EBV DNA is
detected more frequently in peripheral blood mononu-
clear cells, synovial fluid and saliva from RA patients
than from controls and patients with other arthritides
[16]. An EBV DNA quantification assay using real-time
PCR revealed that the EBV DNA load is significantly
higher in peripheral blood mononuclear cells from RA
patients than in peripheral blood mononuclear cells
from patients with non-RA inflammatory diseases and
normal control individuals. In addition, the EBV load
in RA patients expressing HLA-DRB1*0404 is slightly
higher than that observed in patients expressing
HLA-DRB1*07 [17].
Although human retroviruses have long been
suspected as triggering factors in autoimmune diseases,
no conclusive evidence confirming these suspicions has
been reported thus far. One group reported that

agents [27].
These associations between RA and infectious agents
have been supported by increased antibody titers or
DNA to the infectious organism in RA patients com-
pared with other individuals. However, some epidemi-
ological studies oppose these possible links, and there
has been no consistent evidence that a single infectious
agent or other environmental factor is responsible for
the effect of the environment on RA. Therefore, the
pathological role of these and other infectious agents
needs to be explored further.
Smoking
A large number of epidemiological studies have dem-
onstrated an association between cigarette smoking
and the development of RA [28–31]. In particular,
smoking is known to be associated with RF-positive
RA and anti-CCP-positive RA [32], and to interact
with the HLA-DRB1 shared epitope alleles [33]. In
addition, some studies have suggested that smoking
also influences RA disease severity [34,35], although
this remains controversial. While the mechanisms
responsible for the influence of smoking in RA are not
clear, some studies have shown an association between
RA and the toxic compounds found in cigarette
smoke, such as nicotine, 2,3,7,8-tetrachlorodibenzo-
p-dioxin (TCDD) and reactive oxygen species. We
found that the mRNA and protein levels of aryl
hydrocarbone receptor (AhR) were higher in RA syno-
vial tissue than in osteoarthritis tissue, and that TCDD
upregulated the expression of IL-1b, IL-6 and IL-8

Exogenous hormones have also been reported to
influence RA disease risk. For example, several studies
have shown that women who use oral contraceptive pills
have a reduced risk of developing RA [39]. On the other
hand, Walitt et al. reported that there were no statisti-
cally significant differences in the risk of developing RA
or the severity of RA between postmenopausal hormone
therapy groups and placebo groups [40]. The molecular
mechanisms underlying hormone involvement in RA
pathogenesis require further elucidation.
Pregnancy
It is widely accepted that RA frequently remits during
pregnancy. Although the mechanism for this is
unclear, Nelson et al. [41] reported that the amelio-
ration of RA during pregnancy is associated with a
disparity in HLA class II antigens between mother and
fetus. These investigators suggested that the maternal
immune response to paternal HLA antigens may play
a role in pregnancy-induced RA remission. By con-
trast, several studies have suggested that nulliparous
women have an increased risk of developing RA,
although there is no increased risk in women who are
single [42]. These reports support the hypothesis that
pregnancy is related to the development of RA. Recent
studies have also suggested that the risk of disease
development is increased during the postpartum
period, particularly after the first pregnancy [43,44].
Subsequent investigations demonstrated that much of
this increased risk is associated with breastfeeding, and
that women who breastfeed after their first pregnancy

–1
)
None
TCDD (nM)
0.01 0.1 1 10 100
0
TCDD (smoking, etc.)
(RA joint space)
AhR
AhR
TNF-α
AhR
(RA synoviocyte)
ERK
NF-κΒ
IL-
β
, IL-6, IL-8
Target gene
(nucleus)
Fig. 1. (A) TCDD increases IL-1b, IL-6 and IL-8 production from
fibroblast-like synoviocytes obtained from patients with RA. Fibro-
blast-like synoviocytes were stimulated with various concentrations
of TCDD (0–100 n
M) for 12 h and the cytokine concentrations in
fibroblast-like synoviocyte culture supernatants were determined
using cytokine-specific ELISA kits [36]. (B) Schematic representa-
tion of the role of AhR in RA. TCDD, a major component of ciga-
rette smoke, enhances the RA inflammatory process. TCDD
induces inflammatory cytokines via its association with AhR, result-

ity tends to be negatively associated with the presence of
JCP and speculated that patients with JCP have a
genetic predisposition to develop milder RA [49].
Diet
Few reports have addressed the influence of diet on RA
development and progression. The addition of omega-3
fatty acids to the diets of RA patients has been associ-
ated with improvement in RA [50]. Furthermore, diets
high in eicosapentaenoic acid have a favourable effect
on the clinical outcome of RA [51]. Although the asso-
ciation between diet and RA onset is unclear, it is
accepted that such fatty acids compete with arachidonic
acids, the latter of which are involved in inflammation.
Some reports have demonstrated the influence of vita-
mins on RA. For example, greater intake of vitamin D,
primarily from fish and fish products, has been associ-
ated with a lower risk of RA [52]. In addition, we
recently reported that vitamin K, which is primarily
derived from vegetables and legumes [particularly fer-
mented soybeans (natto)], could inhibit the proliferation
of fibroblast-like synoviocytes and the development of
collagen-induced arthritis [53].
Weather conditions
Although no good biologic hypotheses have been
proposed that might explain the association between
RA and weather conditions, Strusberg et al. reported
that low temperature, high atmospheric pressure and
high humidity were significantly correlated with pain
in RA patients [54]. Verges et al. suggested that some
meteorological variables affect the occurrence of pain

Proc Natl Acad Sci U S A 95, 10746–10750.
4 Franssila R & Hedman K (2006) Infection and musculo-
skeletal conditions: viral causes of arthritis. Best Pract
Res Clin Rheumatol 20, 1139–1157.
5 Ray NB, Nieva DR, Seftor EA, Khalkhali-Ellis Z &
Naides SJ (2001) Induction of an invasive phenotype by
human parvovirus B19 in normal human synovial fibro-
blasts. Arthritis Rheum 44, 1582–1586.
6 Lu J, Zhi N, Wong S & Brown KE (2006) Activation
of synoviocytes by the secreted phospholipase A2 motif
in the VP1-unique region of parvovirus B19 minor
capsid protein. J Infect Dis 193, 582–590.
7 Takahashi Y, Murai C, Shibata S, Munakata Y, Ishii
T, Ishii K, Saitoh T, Sawai T, Sugamura K & Sasaki T
(1998) Human parvovirus B19 as a causative agent for
rheumatoid arthritis. Proc Natl Acad Sci U S A 95,
8227–8232.
8 Kozireva SV, Zestkova JV, Mikazane HJ, Kadisa AL,
Kakurina NA, Lejnieks AA, Danilane IN & Murovska
Role of environmental factors in RA S. Kobayashi et al.
4460 FEBS Journal 275 (2008) 4456–4462 ª 2008 The Authors Journal compilation ª 2008 FEBS
MF (2008) Incidence and clinical significance of
parvovirus B19 infection in patients with rheumatoid
arthritis. J Rheumatol 35, 1265–1270.
9 Speyer I, Breedveld FC & Dijkmans BA (1998) Human
parvovirus B19 infection is not followed by inflammatory
joint disease during long term follow-up. A retrospective
study of 54 patients. Clin Exp Rheumatol 16, 576–578.
10 So
¨

patients with rheumatoid arthritis: accurate quantifica-
tion using real-time polymerase chain reaction. Arthritis
Rheum 48, 1223–1228.
18 Griffiths DJ, Cooke SP, Herve
´
C, Rigby SP, Mallon E,
Hajeer A, Lock M, Emery V, Taylor P, Pantelidis P
et al. (1999) Detection of human retrovirus 5 in patients
with arthritis and systemic lupus erythematosus. Arthri-
tis Rheum 42, 448–454.
19 Piper KE, Hanssen AD, Lewallen DG, Matteson EL,
Osmon DR, Duffy MC, Hagan RA, Steckelberg JM &
Patel R (2006) Lack of detection of human retrovirus-5
proviral DNA in synovial tissue and blood specimens
from individuals with rheumatoid arthritis or osteo-
arthritis. Arthritis Rheum 55, 123–125.
20 Gaudin P, Moutet F, Tuke PW & Garson JA (1999)
Absence of human retrovirus 5 in French patients with
rheumatoid arthritis: comment on the article by
Griffiths et al Arthritis Rheum 42, 2492–2494.
21 Forsman A, Uzameckis D, Ro
¨
nnblom L, Baecklund E,
Aleskog A, Bindra A, Pipkorn R, Lejniece S, Kozireva
S, Murovska M et al. (2003) Single-tube nested quanti-
tative PCR: a rational and sensitive technique for detec-
tion of retroviral DNA. Application to RERV-
H ⁄ HRV-5 and confirmation of its rabbit origin.
J Virol Methods 111 , 1–11.
22 Lidbury BA, Rulli NE, Suhrbier A, Smith PN, McColl

tion. Ann Rheum Dis 49, 980–982.
29 Heliovaara M, Aho K, Aromaa A, Knekt P & Reuna-
nen A (1993) Smoking and risk of rheumatoid arthritis.
J Rheumatol 20, 1830–1835.
30 Stolt P, Bengtsson C, Nordmark B, Lindblad S, Lund-
berg I, Klareskog L, Alfredsson L & EIRA study group
(2003) Quantification of the influence of cigarette smok-
ing on rheumatoid arthritis: results from a population
based case-control study, using incident cases. Ann
Rheum Dis 62, 835–841.
31 Voigt LF, Koepsell TD, Nelson JL, Dugowson CE &
Daling JR (1994) Smoking, obesity, alcohol consump-
tion, and the risk of rheumatoid arthritis. Epidemiology
5, 525–532.
32 Klareskog L, Stolt P, Lundberg K, Ka
¨
llberg H, Bengts-
son C, Grunewald J, Ro
¨
nnelid J, Harris HE, Ulfgren
AK, Rantapa
¨
a
¨
-Dahlqvist S et al. (2006) A new model
for an etiology of rheumatoid arthritis: smoking may
trigger HLA-DR (shared epitope)-restricted immune
S. Kobayashi et al. Role of environmental factors in RA
FEBS Journal 275 (2008) 4456–4462 ª 2008 The Authors Journal compilation ª 2008 FEBS 4461
reactions to autoantigens modified by citrullination.

(1998) The influence of sex on the phenotype of
rheumatoid arthritis. Arthritis Rheum 41, 817–
822.
39 Brennan P, Bankhead C, Silman A & Symmons D
(1997) Oral contraceptives and rheumatoid arthri-
tis: results from a primary care-based incident
case-control study. Semin Arthritis Rheum 26, 817–
823.
40 Walitt B, Pettinger M, Weinstein A, Katz J, Torner J,
Wasko MC, Howard BV & Women’s Health Initiative
Investigators (2008) Effects of postmenopausal
hormone therapy on rheumatoid arthritis: the women’s
health initiative randomized controlled trials. Arthritis
Rheum 59, 302–310.
41 Nelson JL, Hughes KA, Smith AG, Nisperos BB,
Branchaud AM & Hansen JA (1993) Maternal-fetal
disparity in HLA class II alloantigens and the
pregnancy-induced amelioration of rheumatoid arthritis.
N Engl J Med 329, 466–471.
42 Silman AJ (1994) Epidemiology of rheumatoid arthritis.
Apmis 102, 721–728.
43 Nelson JL & Ostensen M (1997) Pregnancy and
rheumatoid arthritis. Rheum Dis Clin North Am 23,
195–212.
44 Silman A, Kay A & Brennan P (1992) Timing of
pregnancy in relation to the onset of rheumatoid
arthritis. Arthritis Rheum 35, 152–155.
45 Brennan P & Silman A (1994) Breast-feeding and the
onset of rheumatoid arthritis. Arthritis Rheum 37,
808–813.

quinone-4) – a new potential therapeutic strategy for
rheumatoid arthritis. FEBS J 274, 4588–4594.
54 Strusberg I, Mendelberg RC, Serra HA & Strusberg
AM (2002) Influence of weather conditions on
rheumatic pain. J Rheumatol 29, 335–338.
55 Verge
´
s J, Montell E, Toma
`
s E, Cumelles G, Castan
˜
eda
G, Marti N & Mo
¨
ller I (2004) Weather conditions can
influence rheumatic diseases. Proc West Pharmacol Soc
47, 134–136.
Role of environmental factors in RA S. Kobayashi et al.
4462 FEBS Journal 275 (2008) 4456–4462 ª 2008 The Authors Journal compilation ª 2008 FEBS