S. Ye, N. Patodi, K. Walker-Bone, I. Reading, C. Cooper & E.

Dennison
International Journal of Immunogenetics 2006: 34, 8185
Rheumatoid arthritis (RA) is a multifactorial,
polygenic disorder with an estimated
heritability of 60% (Jirholt et al., 2001;
Svendsen et al., 2002).
Susceptibility to and severity of the disease has
been associated with variations of human
leukocyte antigen genes on chromosome 6
The pathogenesis of RA is characterized by
chronic joint inflammation, and in most cases
the synovitis will lead to permanent damage of
the articular cartilage and bone
Matrix metalloproteinases (MMPs), a group
of proteases expressed in rheumatoid
synovium, play an important role in the
above pathological processes
The genes for MMP1, MMP3, MMP7,
MMP12 and MMP13 contain single
nucleotide polymorphisms that appear to
also have an influence on their expression
In this study, the authors examined whether in RA
patients, these MMP gene polymorphisms are associated
with the degree of physical impairment as assessed by the
Steinbrocker index (Steinbrocker et al., 1949) and the
Health Assessment Questionnaire (HAQ) score (Kirwan
& Reeback, 1986), both of which are indicators of the
degree and extent of joint inflammation and tissue
damage and have been shown to be associated with other
parameters including Ritchie articular index (a measure of
joint inflammation), circulating inflammatory markers
such as the erythrocyte sedimentation rate and C-reactive
protein, and the presence of erosive disease on hand
radiographs
Eighty two British Caucasian patients (27 men and 55
women) who fulfilled the American College of
Rheumatology 1987 criteria for RA and who were
seropositive for rheumatoid factor were recruited to the
study.
Each subject was approached by a trained
rheumatologist when they attended outpatients as part
of their normal rheumatology follow-up, and written
consent was obtained.
Review of case notes was performed to obtain
information on duration of disease, presence of erosive
disease, current erythrocyte sedimentation rate,
prednisolone therapy and second-line medication (with
duration).
The patient was then invited to complete the
Steinbrocker criteria (which divides patients
into four broad functional classes according to
their ability to perform usual self-care,
vocational and avocational activities)
(Steinbrocker et al., 1949) and
HAQ (a short self-report questionnaire of 20
questions in 8 categories that asks about the
difficulties patients have with certain activities)
(Kirwan & Reeback, 1986) and the Ritchie
articular index (Ritchie et al., 1968) was
calculated. The study conformed to the
Declaration of Helskinki guidelines.
The genotyping for the MMP1-1607 1G >
2G, MMP3-1612 5A > 6A, MMP7-181G > A,
MMP12- 82A > G and MMP13-77 A > G
polymorphisms were done using PCR-PFLP

The digests were fractionated by gel
electrophoresis, stained with Vistra Green
and visualized using a fluorimager.
The STATA statistical software package
was used for the analyses.
Univariate analysis was performed to
examine relationships between functional
outcome (Steinbrocker score or HAQ score)
and the following variables: age, sex,
erythrocyte sedimentation rate, presence
of erosive disease, Ritchie score and
prednisolone therapy.
Allele and genotype frequencies were calculated by the
gene counting method.
The HardyWeinberg equilibrium (HWE) programme
(ftp://linkage.rockefeller.edu/software/utilities) was used to
assess whether the genotype distribution of the
polymorphisms was in HWE.
Linear regression analyses were performed to explore the
associations between the MMP gene polymorphisms and
functional outcome (Steinbrocker index and HAQ score) of
RA, first without and then with, adjustment for age, sex,
erythrocyte sedimentation rate, presence of erosive
disease, Ritchie score, prednisolone therapy and years of
diagnosis.
The genotype distributions of all polymorphisms were in
agreement with HWE, and the allele frequencies were similar to
those previously reported in Caucasian samples
Steinbrocker and HAQ scores were highly correlated (r = 0.78, P
< 0.0001) and higher in women than men (P < 0.05).
There was no association of Steinbrocker score or HAQ score
with age. Similarly, patients with a longer duration of disease
had higher Steinbrocker (r = 0.27, P = 0.01) and HAQ scores (r =
0.29, P = 0.009), respectively.
There was no significant association of Steinbrocker or HAQ
score with erythrocyte sedimentation rate, presence of erosive
disease or prednisolone therapy.
However, both Steinbrocker and HAQ scores were higher in
patients with a higher Ritchie index (P < 0.001).
There was an association between the MMP13 - 77A > G
polymorphism and Steinbrocker index, with patients of the A/A
genotype having highest mean score, and the A allele appeared to
have a recessive effect (P = 0.005).
The association remained significant after adjusting for age, sex,
erythrocyte sedimentation rate, presence of erosive disease, Ritchie
score, prednisolone therapy and years of diagnosis (P = 0.003).

There was also an association between the MMP3-
1612 5A/6A polymorphism and Steinbrocker index,
with patients of the 5A/5A genotype having the
highest mean score. The 5A allele appeared to have
a dose effect (P = 0.021 under an additive genetic
model) or a recessive effect (5A/5A compared with
5A/6A and 6A/ 6A, P = 0.020). The statistical
significance of the association was reduced after
adjusting for age, sex, erythrocyte sedimentation
rate, presence of erosive disease, Ritchie score,
prednisolone therapy and years of diagnosis (P =
0.065 under the additive genetic model and P =
0.082 under the recessive model).
An association between the MMP7-181A >
G polymorphism and Steinbrocker score
was observed, with patients of the A/A
genotype having lower score than patients
of the A/G or G/G genotypes, after
adjusting for the covariates mentioned
previously (P = 0.037).

An association between the MMP12-82A > G
polymorphism and Steinbrocker score was also
detected, with patients of the A/A genotype
having higher score than patients of the A/G
genotype (none of the patients was of the G/G
genotype), after adjusting for the covariates
mentioned previously (P = 0.043).
The degree of disability expressed by a patient
with RA is known to be related to a number of
factors, such as the degree and extent of
tissue inflammation and the amount of
structural damage.

MMPs that can degrade various structural
proteins and whose expression is increased in
synovial cells and chondrocytes in arthritic
joints are known to play an important role in
recruitment of inflammatory cells and in
cartilage and bone degradation
These high-expression alleles were found in
this study to be associated with more severe
physical functional impairment indicated by
higher Steinbroker score in RA patients.

The finding of an association between MMP
gene promoter polymorphisms and RA
phenotypes suggest that genetic
polymorphisms could also be a factor that
influences MMP expression during the
pathogenesis of RA.
Six polymorphisms versus two phenotypes
(Steinbrocker and HAQ scores) were tested in this
study, and a Bonferroni correction for the number of
test (n = 12) would require that the P value is < 0.0041
for a test to be considered significant at the = 0.05
level. If this correction is applied, the MMP13-77A > G
with Steinbrocker score (P = 0.002) would be significant,
whereas the effect of the MMP3-1612 5A > 6A, MMP7-
181A > G and MMP12-82A > G polymorphisms would fail
to reach significance.

This correction, however, is likely to be conservative, as
the polymorphisms were correlated and so were the
two phenotypes.
The results of this hypothesis-generating study indicate
that the MMP3, MMP12 and MMP13 gene polymorphisms
are likely to be among genetic factors that underlie inter-
individual variability in the functional outcome of RA.

As identifying the underlying genetic factors could
enhance our understanding of the pathogenesis of this
complex disorder that could eventually lead to improved
prevention and treatment of this common disease,
further studies of MMP gene polymorphisms to include
the MMP3, MMP12 and MMP13 polymorphisms, in large
cohorts of RA patients would be warranted.