Hindawi Publishing Corporation

BioMed Research International

Volume 2013, Article ID 171573, 6 pages
http://dx.doi.org/10.1155/2013/171573

Research Article
Allicin Attenuates Inflammation and Suppresses HLA-B27
Protein Expression in Ankylosing Spondylitis Mice

Xin Gu,1,2 Haishan Wu,2 and Peiliang Fu2

1
Department of Orthopaedics, Shanghai Tenth People’s Hospital Affiliated to TongJi University, Shanghai 20072, China
2
Department of Orthopaedics, The Second Affiliated Hospital of the Second Military Medical University, No. 415 Fengyang Road,
Huangpu District, Shanghai 200003, China

Correspondence should be addressed to Haishan Wu; wuhaishan12121@gmail.com

Received 21 March 2013; Revised 18 September 2013; Accepted 18 September 2013

Academic Editor: Kazim Husain

Copyright © 2013 Xin Gu et al. This is an open access article distributed under the Creative Commons Attribution License, which
permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Here we aimed to determine the therapeutic effect of allicin on ankylosing spondylitis (AS) and explore the mechanism(s) of action.
AS mouse model was constructed by transferring the HLA-B2704 gene into Kunming mice and verified by RT-PCR and CT imaging.
Verified AS mice were randomly divided into model group (𝑛 = 6) and allicin-treated groups (50, 100, and 200 mg/kg, resp., 𝑛 = 6,
p.o., for 2 months). Wild type mice were used as control (𝑛 = 6). The levels of AS-related inflammatory factors were measured by
ELISA. mRNA and protein expressions of HLA-B27 were checked by RT-PCR and western blotting. As the results, the mouse model
of AS was successfully established, and high-dose allicin could markedly alleviate spine inflammatory injury possibly via reducing
the secretion of the inflammatory factors (IL-6, IL-8, and TNF-𝛼) sharply in AS mice. Moreover, allicin significantly inhibited HLA-
B27 protein translation but failed to suppress HLA-B27 gene transcription in AS mice, indicating a posttranscriptional mechanism
of this modulation. In conclusion, allicin has potential to be used for AS treatment as an anti-inflammatory nutraceutical.

1. Introduction metabolites of garlic, is responsible for most of the functions

of garlic [12]. Researches revealed that allicin could have
Ankylosing spondylitis (AS), a chronic inflammation dis- different biological activities such as antibacterial effects [13,
order, occurs in lumbar spine and sacroiliac peripheral 14]. Moreover, allicin can reduce the release of TNF-𝛼 and IL-
joints. The type and progression of AS are associated with 8, and has the function of immune regulation [7], indicating
prognostic factors such as race, gender, age at onset, and early that it may have potential to treat systemic autoimmune
axial skeleton involvement [1]. The incidence of AS among disease (e.g., AS).
xanthoderm is about 2‰ and males are more easily affected In this study, we aimed to investigate whether allicin has
than females [2]. Pathogenesis of AS is strongly related to a therapeutic effect on AS. We established AS model in HLA-
major histocompatibility complex (MHC) class I, and about B2704 (a common subtype of HLA-B27) transgenic mice,
95% of patients with AS are born to have the HLA-B27 gene, treated them with allicin, and measured relevant inflam-
which leads to a tendency for familial genetic association [3, matory factors as well as HLA-B2704 mRNA and protein
4]. However, the mechanism of AS onset is still unclear, and levels in order to find out a promising therapeutic approach
effective therapies are urgently needed for this disease. It was for AS.
reported that the levels of inflammatory factors increased in
AS patient, including TNF-𝛼, IL-6, and IL-10 [4, 5]. Inhibiting 2. Materials and Methods
the release of these inflammatory factors could partly ease
the symptoms of AS and appears to reduce disease severity 2.1. Materials and Chemicals. Allicin chloride was obtained
as well. from Changsha Yaying Bio-tech Co., Ltd. (Changsha, China).
Garlic (Allium sativum L.) was considered as a food with Trizol was purchased from Invitrogen Corporation (Carls-
many therapeutic activities including antioxidant, antibacte- bad, CA, USA). ECL kit and skimmed milk powder were
rial, and anticholesterol activities [6–11]. Allicin, as one of the purchased from Guangzhou Maygene Biotech Co., Ltd.
2 BioMed Research International

(Guangzhou, China). Protein extraction kit was purchased real-time PCR assay. HLA-B27 gene amplification was carried
from Shanghai Times Bio-tech Co., Ltd. (Shanghai, China). out as follows: reverse transcription at 48∘ C for 45 min; initial
activation of HotStar Taq DNA Polymerase at 96∘ C for 2 min;
2.2. Construction of AS Mouse Model and Drug Administra- 40 cycles in three steps: 95∘ C for 30 s, 60 ∘ C for 30 s, and
tion. All animal studies have been approved by China Ethics 72∘ C for 30 s. Data acquisition and analysis were carried out
Committee and performed in accordance with the ethical using CFX-96 real-time quantitative PCR instrument (Bio-
standards. Kunming mice were obtained from the Laboratory Rad) and SPSS 15.0 software.
Animal Institute of Chinese Academy of Medical Sciences.
Plasmid (pBR322-HLA-B2704) was provided by our labo- 2.6. Enzyme-Linked Immunosorbent Assay (ELISA). TNF-𝛼,
ratory. HLA-B2704 DNA fragment was prepared via ampli- IL-6, and IL-8 secretion was analyzed by sandwich ELISA.
fication, EcoR I digestion, electrophoresis, extraction, and Antibodies (rabbit anti-TNF-𝛼, anti-IL-6, and anti-IL-8 anti-
purification. AS model mice were constructed by transgene body) were added to 96-well plates and incubated overnight
method [15]. Three hundred mice were injected with hor- at 4∘ C. The plates were then washed and blocked with 2%
mone for super ovulation. Zygotes were collected and HLA- BSA. Samples and corresponding standard samples were
B2704 gene DNA fragment was injected into pronucleus by added to the wells and the plate was incubated at 4∘ C for
microinjection. Survived zygotes of second cell stage were 30 min. After washing, the wells were incubated with a
transferred into pseudocyesis mice for generation. biotinylated goat anti-rabbit secondary antibody and perox-
The AS mice were randomly divided into AS model group idase labelled avidin at 4∘ C for 30 min. Following multiple
(𝑛 = 6) and allicin-treated groups. Treatment dosages were washes, chromogenic agent was added into the wells and the
50, 100, and 200 mg/kg body weight [16] for the low-dose optical absorbance was detected at 405 nm. Concentration of
group (𝑛 = 6), the moderate dose group (𝑛 = 6), and the samples was calculated according to the standard curve.
high-dose group (𝑛 = 6), respectively. The drug was infused
to stomach of mice every morning, whereas the control group 2.7. Western Blotting Analysis. HLA-B27 and 𝛽-actin expres-
(wild type Kunming mice, 𝑛 = 6) and the model group sion were determined by western blotting analysis. Whole
received no addition. All the mice were killed after 2 months proteins were prepared by protein extraction kit. Equal
of treatment and tissue sample near foot joint as well as amounts of proteins (20 𝜇g) were separated by 12.5% SDS-
peripheral blood from individual animals was collected. PAGE and transferred to PVDF membranes, which were
blocked with 5% skimmed milk and incubated with the
2.3. Reverse Transcription-Polymerase Chain Reaction (RT- rabbit anti-HLA-B2704 antibody and anti-𝛽-actin antibody.
PCR) Assay. Total RNA (from peripheral blood) was pre- These were followed by incubation with the goat anti-
pared by the acid phenol method using Trizol reagent as rabbit antibody conjugated with horseradish peroxidase as
instructed by the manufacturer. RT-PCR was performed as the secondary antibody for 1 h. Immunoreactive bands were
described previously, and the reverse-transcribed cDNA was detected by X-ray film developer.
then amplified by PCR using the following primers: HLA-B27,
2.8. Statistical Analysis. The data were expressed as mean ±
5󸀠 -GGGTCTCACACCCTCCAGAAT-3󸀠 (sense), and 5󸀠 -
SEM and analyzed by one-way ANOVA or two-tailed
CGGCGGTCCAGGAGCT-3󸀠 (antisense). PCR conditions
unpaired Student’s 𝑡-test. A probability value of less than 0.05
were as follows: one cycle of 48∘ C for 45 min and 96∘ C for
was considered statistically significant. All results are repre-
2 min; 95∘ C for 30 s, 60∘ C for 45 s, and 72∘ C for 30 s, for 40
sentative of at least three independent experiments.
cycles; the last cycle for 7 min at 70∘ C. PCR products were
separated by electrophoresis on 2% agarose gels and visual-
3. Results
ized by staining with ethidium bromide. The sequence length
of target product is 135 bp. 3.1. Construction of AS Mouse Model. HLA-B2704 gene
expression in peripheral blood of the AS mice was measured
2.4. CT Imaging. Siemens 16 layers spiral CT scanner was by RT-PCR (Figure 1). A total of 26 positive AS model mice
used for imaging. The parameters were as follows: rota- were obtained, in which HLA-B2704 gene was expressed.
tional speed 0.4 r/s, precision diameter 0.75 mm, tube voltage It was found by CT imaging that the spine of control mice
120 kV, tube current “automatic mA control,” screw pitch was normal, as articular surface was smooth and intact, and
1 mm, and scanning layer thickness 0.75 mm for 16 layers. spine interval has no deformation (Figure 2(a)). However, the
spine interval of AS model mice was changed obviously, as the
2.5. Real-Time Quantitative PCR. Total RNA (from tissue continuity was lost (Figure 2(b)).
sample near foot joint) was prepared as mentioned above.
HLA-B27 primer sequences were designed by an online tool 3.2. Effects of Allicin on Symptoms of AS Mice. Mice of control
(http://www.idtdna.com/), and 𝛽-actin primer sequences group showed no indisposition phenomenon. Model group
were obtained from the reference [17]. The primer sequences showed severe red swelling in foot and toes and local
are as follows: HLA-B27, 5󸀠 -GAGAACGGGAAGGAC- depilation. After treated with 50 mg/kg or 100 mg/kg of
AAGC-3󸀠 (forward), 5󸀠 -GATCTCCGCAGGGTAGAAAC- allicin, red swelling and depilation were alleviated. The group
3󸀠 (reverse); 𝛽-actin, 5󸀠 -AGCGGGAAATCGTGCGTG- which received 200 mg/kg allicin showed no apparent red
AC-3󸀠 (forward), 5󸀠 -ACTCCTGCTTGCTGATCCACATC- swelling and depilation phenomenon in foot and skin, and
3󸀠 (reverse). mRNA level was measured by SYBR green the impaired spine was almost recovered (Figure 2(c)).
BioMed Research International 3

M 1 2

200
(a)
150

100

Figure 1: Verification of the expression of HLA-B2704 in peripheral

blood by RT-PCR. M: marker; 1: positive group; 2: control group.

3.3. Effects of Allicin on the Release of Inflammatory Factors.

As illustrated in Figure 3, the release of IL-6, IL-8, and TNF-
𝛼 was significantly increased in AS model mice compared
with control mice, indicating that there were inflammation
reactions in joints. Allicin could extremely inhibit these three (b)
inflammatory factors releasing in a dose-dependent manner.
Allicin at the dose of 100 mg/kg could reduce the level of
inflammatory factors to normal, and 200 mg/kg allicin even
reduced IL-6 and TNF-𝛼 to the levels a little lower than
normal (Figures 3(a) and 3(c)).

3.4. Effects of Allicin on mRNA Content of HLA-B27 in AS

Mice. HLA-B2704 mRNA level in joints was determined
by fluorescence quantitative PCR. As shown in Figure 4,
mRNA level of HLA-B27 in AS model mice was not altered
significantly by allicin treatment, indicating that allicin might
not affect the transcription of HLA-B2704 gene.

3.5. Effects of Allicin on HLA-B2704 Protein Expression in (c)

AS Mice. HLA-B2704 protein expression was investigated
by western blotting analysis. Results showed that HLA-B27 Figure 2: Effects of allicin on spine impairment of AS mice. Siemens
protein expression was significantly increased in AS model 16 layers spiral CT scanner was used for imaging. (a) Control mice;
group. Allicin showed an inhibiting effect on HLA-B2704 (b) AS model mice; (c) 200 mg/kg allicin-treated AS mice.
protein expression in all three doses, indicating that allicin
could inhibit HLA-B2704 translation (Figure 5).
suggesting a posttranscriptional mechanism of the modula-
4. Discussion tion of allicin on HLA-B2704 gene.
Researches have shown that allicin could affect immune
This study investigated the effects of allicin on AS in HLA- function in mice [18, 19]; that is, allicin increases the weight
B2704 transgenic mice. Our results show that allicin could of the spleen and thymus, activates T-lymphocyte, and stim-
ameliorate spine impairment and reduce the release of IL-6, ulates the secretion of mononuclear cell and lysozyme releas-
IL-8, and TNF-𝛼 in AS model mice. Specifically, we found ing. Thus, allicin could improve cellular immunity, humoral
out that allicin could inhibit HLA-B2704 protein expression immunity and nonspecific immune functions. It was reported
but had no effect on the expression of HLA-B2704 gene, that anti-TNF-𝛼 therapy was effective for several weeks in AS.
4 BioMed Research International

240 600
200 ∗ 500 ∗
IL-6 (ng/L)

IL-8 (ng/L)
160 ∗∗ 400 ∗∗ ∗∗ ∗∗
∗∗
120 ∗∗ 300
80 200
40 100
0 0
Control Model 50 mg/kg 100 mg/kg 200 mg/kg Control Model 50 mg/kg 100 mg/kg 200 mg/kg
(a) (b)
1800
1500
TNF-𝛼 (ng/L)
1200
∗∗ ∗∗
900 ∗∗ ∗∗
600
300
0
Control Model 50 mg/kg 100 mg/kg 200 mg/kg
(c)

Figure 3: Effects of allicin on the releasing of IL-6, IL-8, and TNF-𝛼 in HLA-B2704 transgenic mice. Mice were orally administered with
allicin (50 mg/kg, 100 mg/kg, and 200 mg/kg, resp.) every morning for 2 months. IL-6, IL-8, and TNF-𝛼 were tested by ELISA analysis. All data
presented are the mean ± SEM of 6 mice. (a) IL-6; (b) IL-8; (c) TNF-𝛼. ∗ 𝑃 < 0.05, ∗∗ 𝑃 < 0.01 versus model group.

6 HLA-B2704
Relative expression

5
𝛽-Actin
4
3 (a)
2 1.5
1
0
Control Model 50 mg/kg 100 mg/kg 200 mg/kg
Relative expression

1.0
Figure 4: Effects of allicin on HLA-B2704 gene expression. Mice
were orally administered with allicin (50 mg/kg, 100 mg/kg and
200 mg/kg, resp.) every morning for 2 months. HLA-B2704 mRNA
expression was analyzed by Real-time fluorescence quantitative PCR 0.5 ∗∗ ∗∗ ∗∗
analysis. All data presented are the mean ± SEM of 6 mice. ∗∗ 𝑃 <
0.01 versus Model group.
∗∗
0.0
Control Model 50 mg/kg 100 mg/kg 200 mg/kg
We have shown the elevated TNF-𝛼 in joints, the primary
source of which is the activated macrophages in chronic (b)
inflammatory disease [20]. The activated macrophages com-
bined with apparent enhancement of IL-6 and IL-8 suggest Figure 5: Effects of allicin on HLA-B2704 protein expression. Mice
were orally administered with allicin (50 mg/kg, 100 mg/kg, and
two stages in this inflammatory process: firstly, Th1 cells
200 mg/kg, resp.) every morning for 2 months. HLA-B2704 expres-
activate the secretion of IFN-𝛾, leading to low level secretion sion was analyzed by western blotting analysis. (a) Blots of HLA-
of TNF-𝛼 and secretion of other macrophage cytokines (e.g., B2704 and 𝛽-actin; (b) quantitative analysis of the blots. All data
IL-6, IL-8); in the second stage, TNF-𝛼, in combination with presented are the mean ± SEM of 6 mice. ∗∗ 𝑃 < 0.01 versus model
T-cell-derived IFN-𝛾, activates macrophages and polymor- group.
phonuclear neutrophils (PMN) which are attracted to the site
of inflammation (by IL-8, e.g.), thus leading to self-amplifying
inflammation [21, 22]. In our study, allicin significantly inhib- tents to be misfolded [25] but also forms heavy chain dimers
ited the secretion of TNF-𝛼, IL-6, and IL-8 in AS model mice, by disulfide interaction [26, 27]. Both events are related
indicating that allicin could reduce inflammatory reaction of to the triggering of inflammatory responses, including the
AS efficiently. release of inflammatory factors, such as TNF-𝛼 [25–27]. In
Although the pathogenesis of AS is still not fully clarified, our study allicin inhibited the release of three inflammatory
numerous studies have shown that HLA-B27 gene is closely factors, meanwhile it also inhibited the expression of HLA-
linked to the occurrence of AS [23, 24]. HLA-B27 not only B2704 protein expression. Thus, we speculate that allicin
BioMed Research International 5

might suppress inflammation via inhibiting HLA-B2704 [11] K. Prasad, V. A. Laxdal, M. Yu, and B. L. Raney, “Antioxidant
protein expression, although the exact mechanism deserves activity of allicin, an active principle in garlic,” Molecular and
further investigation. Moreover, whether allicin could affect Cellular Biochemistry, vol. 148, no. 2, pp. 183–189, 1995.
the activation of NF-𝜅B and/or other signal pathways related [12] C. Liu, F. Cao, Q.-Z. Tang et al., “Allicin protects against car-
to AS should be conducted further as well [28, 29]. diac hypertrophy and fibrosis via attenuating reactive oxygen
To sum up, our present research provides evidence species-dependent signaling pathways,” Journal of Nutritional
that allicin can significantly reduce TNF-𝛼, IL-6, and IL- Biochemistry, vol. 21, no. 12, pp. 1238–1250, 2010.
8 secretion and HLA-B2704 protein expression in HLA- [13] Z. M. Ross, E. A. O’Gara, D. J. Hill, H. V. Sleightholme, and D.
B2704 transgenic AS model mice. These results indicate J. Maslin, “Antimicrobial properties of garlic oil against human
that allicin would be a potential therapeutic nutraceutical enteric bacteria: evaluation of methodologies and comparisons
with garlic oil sulfides and garlic powder,” Applied and Environ-
for AS treatment. It is of great importance to identify the
mental Microbiology, vol. 67, no. 1, pp. 475–480, 2001.
direct protein target of allicin as well as further molecular
[14] S. Ankri and D. Mirelman, “Antimicrobial properties of allicin
mechanisms in the future.
from garlic,” Microbes and Infection, vol. 1, no. 2, pp. 125–129,
1999.
Conflict of Interests [15] M. J. Turner, M. L. DeLay, S. Bai, E. Klenk, and R. A. Col-
bert, “HLA-B27 up-regulation causes accumulation of mis-
No conflict of financial interests exists. folded heavy chains and correlates with the magnitude of the
unfolded protein response in transgenic rats: Implications for
the pathogenesis of spondylarthritis-like disease,” Arthritis &
References Rheumatism, vol. 56, no. 1, pp. 215–223, 2007.
[1] R. Burgos-Vargas, J. Vázquez-Mellado, N. Cassis et al., “Genuine [16] X. H. Li, C. -Y Li, Z.-G. Xiang et al., “Allicin ameliorates cardiac
ankylosing spondylitis in children: a case-control study of hypertrophy and fibrosis through enhancing of Nrf2 antioxi-
patients with early definite disease according to adult onset dant signaling pathways,” Cardiovascular Drugs and Therapy,
criteria,” The Journal of Rheumatology, vol. 23, no. 12, pp. 2140– vol. 26, no. 6, pp. 457–465, 2012.
2147, 1996. [17] Z. Li Yongsheng, “Effect of RNAi-mediated gene silencing of
[2] A. Calin, S. Brophy, and D. Blake, “Impact of sex on inheritance HER2 /neu on proliferation and apoptosis in bladder cancer cell
of ankylosing spondylitis: a cohort study,” The Lancet, vol. 354, line BIU-87,” Journal of Third Military Medical University, vol.
no. 9191, pp. 1687–1690, 1999. 34, no. 1, pp. 20–23, 2012.
[18] Y. Feng, X. Zhu, Q. Wang et al., “Allicin enhances host pro-
[3] D. A. Brewerton, F. D. Hart, A. Nicholls, M. Caffrey, D. C. James,
inflammatory immune responses and protects against acute
and R. D. Sturrock, “Ankylosing spondylitis and HL-A 27,” The
murine malaria infection,” Malaria Journal, vol. 11, no. 1, p. 268,
Lancet, vol. 301, no. 7809, pp. 904–907, 1973.
2012.
[4] J. Londono, M. C. Romero-Sanchez, V. G. Torres et al., “The [19] N. Ota, F. Takano, S. Muroga et al., “Garlic extract and its select-
association between serum levels of potential biomarkers with ed organosulphur constituents promote ileal immune responses
the presence of factors related to the clinical activity and poor ex vivo,” Journal of Functional Foods, vol. 4, no. 1, pp. 243–252,
prognosis in spondyloarthritis,” Revista Brasileira de Reuma- 2012.
tologia, vol. 52, no. 4, pp. 536–544, 2012.
[20] S. H. Murch, C. P. Braegger, J. A. Walker-Smith, and T. T. Mac-
[5] L. Zeng, M. J. Lindstrom, and J. A. Smith, “Ankylosing spondyli- Donald, “Location of tumour necrosis factor 𝛼 by immunohis-
tis macrophage production of higher levels of interleukin- tochemistry in chronic inflammatory bowel disease,” Gut, vol.
23 in response to lipopolysaccharide without induction of a 34, no. 12, pp. 1705–1709, 1993.
significant unfolded protein response,” Arthritis & Rheumatism,
[21] A. M. Williams, C. V. Whiting, K. Bonhagen et al., “Tumour
vol. 63, no. 12, pp. 3807–3817, 2011.
necrosis factor-alpha (TNF-𝛼) transcription and translation in
[6] C. J. Cavallito, J. S. Buck, and C. M. Suter, “Allicin, the antibac- the CD4+ T cell-transplanted scid mouse model of colitis,”
terial principle of Allium sativum. II. Determination of the Clinical and Experimental Immunology, vol. 116, no. 3, pp. 415–
chemical structure,” Journal of the American Chemical Society, 424, 1999.
vol. 66, no. 11, pp. 1952–1954, 1944. [22] O. Dienz and M. Rincon, “The effects of IL-6 on CD4 T cell
[7] A. Lang, M. Lahav, E. Sakhnini et al., “Allicin inhibits spon- responses,” Clinical Immunology, vol. 130, no. 1, pp. 27–33, 2009.
taneous and TNF-𝛼 induced secretion of proinflammatory [23] G. Zhang, J. Luo, J. Bruckel et al., “Genetic studies in familial
cytokines and chemokines from intestinal epithelial cells,” ankylosing spondylitis susceptibility,” Arthritis & Rheumatism,
Clinical Nutrition, vol. 23, no. 5, pp. 1199–1208, 2004. vol. 50, no. 7, pp. 2246–2254, 2004.
[8] K. T. Augusti and P. T. Mathew, “Lipid lowering effect ollicin [24] L.-S. Tam, J. Gu, and D. Yu, “Pathogenesis of ankylosing
(diallyl disulphide oxide) on long term feeding to normal rats,” spondylitis,” Nature Reviews Rheumatology, vol. 6, no. 7, pp.
Experientia, vol. 30, no. 5, pp. 468–470, 1974. 399–405, 2010.
[9] Y. Lu, Z. He, X. Shen et al., “Cholesterol-lowering effect of allicin [25] N. S. Dangoria, M. L. Delay, D. J. Kingsbury et al., “HLA-B27
on hypercholesterolemic ICR mice,” Oxidative Medicine and misfolding is associated with aberrant intermolecular disulfide
Cellular Longevity, vol. 2012, Article ID 489690, 6 pages, 2012. bond formation (dimerization) in the endoplasmic reticulum,”
[10] M. F. Melzig, E. Krause, and S. Franke, “Inhibition of adenosine Journal of Biological Chemistry, vol. 277, no. 26, pp. 23459–
deaminase activity of aortic endothelial cells by extracts of garlic 23468, 2002.
(Allium sativum L.),” Die Pharmazie, vol. 50, no. 5, pp. 359–361, [26] S. Kollnberger, L. A. Bird, M. Roddis et al., “HLA-B27 heavy
1995. chain homodimers are expressed in HLA-B27 transgenic rodent
6 BioMed Research International

models of spondyloarthritis and are ligands for paired Ig-like

receptors,” The Journal of Immunology, vol. 173, no. 3, pp. 1699–
1710, 2004.
[27] T. M. Tran, N. Satumtira, M. L. Dorris et al., “HLA-B27 in trans-
genic rats forms disulfide-linked heavy chain oligomers and
multimers that bind to the chaperone BiP,” The Journal of
Immunology, vol. 172, no. 8, pp. 5110–5119, 2004.
[28] A. Taylan, I. Sari, B. Akinci et al., “Biomarkers and cytokines of
bone turnover: extensive evaluation in a cohort of patients with
ankylosing spondylitis,” BMC Musculoskeletal Disorders, vol. 13,
no. 1, p. 191, 2012.
[29] J. C. C. Wei, K. S. Hung, Y. W. Hsu et al., “Genetic polymor-
phisms of stromal interaction molecule 1 associated with the
erythrocyte sedimentation rate and C-reactive protein in HLA-
B27 positive ankylosing spondylitis patients,” PloS ONE, vol. 7,
no. 12, Article ID e49698, 2012.