Ti–O based nanomaterials ameliorate experimental autoimmune encephalomyelitis and collagen-induced arthritis

Abstract

Multiple sclerosis (MS) and rheumatoid arthritis (RA) are the most common chronic autoimmune inflammatory diseases that affect the central nervous system and joints respectively. Treatment of autoimmune diseases usually concentrates on alleviating symptoms. High-mobility group box 1 protein (HMGB1) cytokine had been reported to play a key role in autoimmune disorders as HMGB1 levels correlate with active inflammation and neutralizing HMGB1 can rescue various autoimmune diseases. Nano-size titania (TiO₂) is an exceptional multi-functional material that showed several practical applications ranging from pigments in paints, UV light absorbent in sunscreen lotion to coatings on non-fogging surfaces, biomedicine and agriculture. However, the in vivo role of Ti–O based nanomaterials in autoimmune disease models has not been examined. This study was designed to investigate the role of Ti–O based nanomaterials such as H₂Ti₃O₇ nanotubes (TNT) and anatase TiO₂ fine particles (TFP) in well established animal models experimental autoimmune encephalomyelitis (EAE) and collagen induced arthritis (CIA). We showed for the first time that the administration of Ti–O based nanomaterials attenuated clinical signs of pathophysiology and correlated with the reduction of the pro-inflammatory cytokine HMGB1. The clinical signs, histology and HMGB1 secretion data showed the therapeutic role of TNT and TFP in EAE and TNT in CIA. Thus, TNT and TFP have potential applications in specific treatment of MS/RA and this may provide an effective novel therapeutic approach for other autoimmune diseases.

---

Introduction

Autoimmune diseases are caused by a failure of peripheral T-cell tolerance, resulting in the imbalance of immunoregulatory and inflammatory processes. The most common diseases attributed to autoimmune disorders are multiple sclerosis (MS), rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), myasthenia gravis, pernicious anemia, and scleroderma. MS is an immune-mediated, demyelinating and neurodegenerative disease affecting the central nervous system (CNS).¹ MS approximately affecting 2.5 million people are preferentially young adult women worldwide.² The widely accepted animal model of MS is experimental autoimmune encephalomyelitis (EAE) used to study the pathophysiology of the disease induced in rodents using self-antigenic epitope peptides from myelin oligodendrocyte glycoprotein (MOG), myelin basic protein (MBP) and proteolipid protein (PLP).³ In the current study MOG_35–55 peptide is used, which is a powerful antigen inducing EAE in C57BL/6 mice.⁴ Subsequently Sprague-Dawley (SD) rats were used to induce arthritis by collagen-induced arthritis (CIA) model which shares many clinical, pathological and immunological similarities with rheumatoid arthritis (RA) in human.⁵ RA is a systemic, chronic autoimmune inflammatory disease characterized by synovial hyperplasia that affects the joints and other tissues in the body.⁶ Though the exact causes of RA and MS are not known, but role of various inflammatory immune cells and network of cytokines were evidenced to be involved in disease progression.⁷ High-mobility group box 1 protein (HMGB1) is a ubiquitous DNA-binding protein, released from activated immune cells or damaged, dying cells during necrosis and during the late phase of cellular apoptosis [reviewed in ref. 8 and 9]. Extracellular HMGB1 binds to receptors such as RAGE (receptor for advanced glycation end-products), toll-like receptor (TLR)-2, TLR-4 and intracellular receptor TLR-9 (ref. 10 and 11) and results in production of a spectrum of pro-inflammatory cytokines, such as IL-1β, TNF-α, IL-6, IL-8 (ref. 12) and chemokines.¹³ It has been reported that HMGB1 plays a key role in autoimmune disorders including multiple sclerosis (MS)¹⁴ and rheumatoid arthritis (RA) by mediating the proliferation of T cells in response to anti-CD3 antibody and RAGE.¹⁵ Reynolds et al. found that TLR4 expression by T cells is essential for the development of EAE and TLR4^−/− animals efficiently abrogated the EAE disease symptoms. Further, it has been suggested that TLR4 dependent pathways are very essential for induction of EAE, which were involved in development and recruitment of leucocytes in the autoimmune CNS disease.¹⁶ Furthermore, VGX-1027 [(S,R)-3-phenyl-4,5-dihydro-5-isoxasole acetic acid] acts as antagonist for TLR4 and significantly slower progression of the arthritic disease with lower clinical and histological arthritic score by inhibiting the cytokines such as IL-1β, TNF-α and IL-10, which play an important immunopharmacological role.¹⁷ In active lesions of MS/EAE and synovium of RA/CIA, HMGB1 levels correlate with active inflammation^18,19 and neutralizing HMGB1 antibody can rescue mice from EAE^20,21 as well as rats from CIA.²² Malhotra and co-workers found that MS patients showed increased mRNA and protein levels of HMGB1, particularly in patients with relapsing-remitting MS and secondary progressive MS as compared to healthy controls.²³ These facts suggest that HMGB1 plays a critical role in MS/EAE and RA/CIA and it is the target for therapeutic treatment of autoimmune disorders.

Non-steroidal anti-inflammatory drugs (NSAIDs), glucocorticoids and immunosuppressants are usually used as autoimmune disease treatment.²⁴ However, these drugs have side effects and their toxicity leads to other diseases. Recently use of nanomedicines increased enormously and nanomaterials were shown to offer promising strategies to optimize and improve the treatment of autoimmune disorder. Moreover nanomedicine based therapy has the ability to overcome the limitations of current immunosuppressive and biological therapies.^25–27 The restoration of immune tolerance and using nanoparticles (NPs) is a crucial for autoimmune therapy. More recently, Maldonado et al. showed that, pegylated PLGA rapamycin and OVA_323–334 NPs significantly reduced the production of OVA-specific IgG.²⁸ Yeste et al. using the MOG_35–55 or PLP_39–151 EAE model, pegylated gold NPs loaded with the aryl hydrocarbon receptor 2-(1′H-indole-3′-carbonyl)-thiazole-4-carboxylic acid methyl ester (ITE) and MOG_35–55 or PLP_39–151 were tested for their ability to ameliorate disease, these NPs successfully suppress the EAE.²⁹ Dexamethasone (Dex) has been used for evaluating the effects of both prophylactic and therapeutic treatments in different forms of rodent EAE.³⁰ Dexamethasone ameliorate the development of EAE by increased frequency of autoantigen-specific IFNγ secreting lymph node mononuclear cells. Administration of dexamethasone to the CIA in rodents, suppress the foot swelling and decrease of bone mineral density by inhibiting the overproduction of inflammatory cytokines.^31,32 Cyclophosphamide has been reported to treat not only cancer, but also autoimmune diseases. However cyclophosphamide was unable to influence the clinical course of EAE in either MOG induced EAE in C57Bl/6 mice or PLP-induce EAE in SJL mice suggesting that these models may be refractory to immunopharmacological manipulation by cyclophosphamide.³³ Nanoscience and technology has been witnessing an exponential growth in research and development on material synthesis, properties and its applications. Nanomaterials composed of particle size ≤100 nm are exciting due to their extraordinary physico-chemical properties such as high specific surface area and surface-to-volume ratio resulting unique properties than that of their bulk counterpart. Often these materials showed enhanced biocompatibility of biological cells. Previous studies have reported that fullerene nanoparticles and it derivatives accumulate in the joints of murine and effectively inhibit the inflammatory cascade in CIA.^34–36 Furthermore, suppression of CIA in rats without toxic effects on the internal organs by intra-articular administration of 13 nm gold nanoparticles (AuNPs) with a concentration of 180 μg ml⁻¹.^37,38 Nagai et al. stated that, adjuvant induced arthritis (AA) in rats can be treated with gel ointment containing tranilast nanoparticles.³⁹ A biodegradable polymer poly (lactic-co-glycolic acid) (PLGA) nanoparticles entrapping type II collagen (CII) when administration of administered 3 mg PLGA-containing 40 μg CII, result in the significantly lower mean arthritis score and severity CIA in mice.⁴⁰

Titania is widely used in a number of industrial applications ranging from pigments in paints, UV light absorbent in sunscreen lotion to coatings on non-fogging surfaces. It has been recognized that properties of nano structured titania are different from the bulk form, which could lead to new applications or provide better materials for existing ones. Nano-sized titania based materials have showed excellent photo catalytic properties, anticorrosion, high stability and good biocompatibility.^41,42 Titanium and its alloys are widely used as orthopedic implant materials include hip and dental implants^43,44 as well as jaw fractures.⁴⁵ The combination of doxorubicin–TiO₂ effectively enhancing the anticancer efficacy in human SMMC-7721 hepatocarcinoma cells.⁴⁶ An improvement in cancer cells killing was demonstrated using photocatalytic action of antibody–TiO₂ bioconjugates.⁴⁷ Schanen et al. reported the immunomodulatory properties of TiO₂ in human peripheral blood mononuclear cells.⁴⁸ However, the toxic effect of nano-TiO₂ remains debatable, as conflicting reports have showed that, after initial absorption of nano-TiO₂ can be distributed to other organs and tissues in the body. Thus, nano-TiO₂ interact with plasma membrane exerts genotoxicity via reactive oxygen species induction.⁴⁹ J. Xu et al. demonstrated that intragastric and intravenous injection of TiO₂ nanoparticles at high doses in mice, because acute toxicity effects in the brain, lung, spleen, liver and kidney.⁵⁰ Recent studies reported that TiO₂ NPs are more toxic than TiO₂ fine particles (FPs).⁵¹ Oberdorster et al. demonstrated that TiO₂ NPs caused a greater pulmonary inflammatory response than TiO₂ FPs at same mass burden.⁵² Although the roles of TiO₂ nanomaterials have been shown in several biological applications, no literature exists on the role of these materials in autoimmune disease.

Here we report for the first-time, H₂Ti₃O₇ nanotubes have therapeutic role in well established autoimmune disease animal models EAE and CIA. The results are compared with commercially available anatase TiO₂ fine particles as standard material to repeat the experiments in future and to explain the morphology effect. The clinical signs, histology and HMGB1 secretion data show that therapeutic role of TNT and TFP in EAE and TNT in CIA.

Materials and methods

Materials

RPMI-1640, Dulbecco's modified Eagles Medium (DMEM), phosphate buffered saline (PBS), antibiotic solution (ABS) and fetal bovine serum (FBS) were purchased from Invitrogen. Lipopolysaccharide (LPS), 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyl tetrazolium bromide (MTT), tryphan blue and mitomycin C were purchased from Sigma; Dulbecco's phosphate buffered saline (DPBS), 0.05% trypsin–EDTA and EZcount™ MTT cell assay kit from Himedia Laboratories, India; MOG_35–55 peptide, complete Freund's adjuvant (CFA), pertussis toxin from Hooke laboratories Inc, MA, USA; bovine type II collagen and HMGB1 Detection Kit from Chondrex Inc., WA, USA.

Synthesis of H₂Ti₃O₇ nanotubes

The protonic trititanate (H₂Ti₃O₇) nanotubes were synthesized by alkaline hydrothermal method as reported earlier.⁵³ In a typical synthesis, TiO₂ fine particles denoted as TFP (TiO₂ LAB, Merck, India) dispersed into 10 M NaOH aqueous solution was transferred in Teflon-lined stainless steel autoclave and heated at 130 °C for 20 h. The white precipitate was washed twice with distilled H₂O, dil. HCl and C₂H₅OH, finally dried at 80 °C for 12 h, the bright white powder denoted as TNT. Endotoxin content analysis of the NPs was performed using Limulus amebocyte assay to determine the level of endotoxin in the TiO₂ nanomaterials. The amount of endotoxin detected in 1 μg of the TFP and TNT injected into mice was 0.4 and 0.35 pg respectively, which did not stimulate production of any cytokines in the mouse ligated ileal loops.

Characterization techniques

Powder X-ray diffraction (PXRD) data were recorded using a D8 ADVANCE X-ray diffractometer (Bruker), with λ_CuKα = 1.54056 Å. Transmission electron microscopy (TEM) measurements were carried out by using a FEI Tecnai F20ST electron microscope operated at 200 keV, equipped with high angle annular dark field (HAADF) detector and energy dispersive X-ray (EDX) spectrometer. For the TEM measurements, all samples were sonicated in ethanol and the resulting dispersions were transferred on to holey carbon coated copper grids (200 mesh). The particle size and surface charge (zeta potential) of TFP and TNT was measured by using zeta analyzer (SZ-100-Horiba, Japan) to find out the possibilities of any charge based interaction.

Cell culture and HMGB1 quantification

The murine macrophage RAW 264.7 cell line was obtained from National Centre for Cell Science (NCCS), Pune, India and cultured in DMEM medium supplemented with 10% FBS, 1% penicillin–streptomycin incubated at 37 °C in 5% CO₂ incubator. Cells were plated at a density of 1 × 10⁶ cells per well in a 6 well plate and treated with LPS in the presence or absence of TNT or TFP at a concentration of 50 μg ml⁻¹ for 24 h. The supernatants were collected and stored at −80 °C until use. The level of HMGB1 in supernatants was detected using HMGB1 ELISA detection kit (Chondrex Inc., WA, USA) according to the manufacturer's instructions.

Animals

Female C57BL/6 mice at 8–10 weeks of age (20–22 g) and female Sprague-Dawley (SD) rats at 6–8 weeks of age (180–200 g) were housed in free of murine specific pathogens under optimal conditions of hygiene, temperature, humidity, light (cycles of 12 h dark/light) and fed with standard rodent chow and water ad libitum. Experimental animal protocols were approved by the Institutional Animal Ethics Committee (IAEC) and all procedures were conducted in accordance with the “Guide for the Care and Use of Laboratory”.

Toxicity studies

Mixed lymphocyte reaction (MLR) was used to measure in vitro toxicity of nanomaterials by measuring the proliferation of splenocytes. MLR was carried out as previously described, with some modifications.⁵⁴ Splenocytes were isolated from 6–8 week-old C57BL/6 (H-2^b) mice and incubated (1 × 10⁶ cells per ml) with different concentrations of TNT and TFP for 24 h at 37 °C in a humidified 5% CO₂ incubator. Cells were inactivated with mitomycin C for 30 min, washed and used as stimulators. Splenocytes of BALB/c (H-2^d) mice were used as responders. Stimulators (0.01 × 10⁶ cells per ml) and responders (0.1 × 10⁶ cells per ml) were co-cultured in a flat bottom 96 well plate for three days. Proliferation response was measured by MTT assay using EZcount™ MTT Cell Assay Kit (Himedia Laboratories, India) according to the manufacturer's instructions.

EAE induction, treatment and assessment

Mice were randomly divided into four groups (n = 5), namely control, EAE, EAE-TNT and EAE-TFP. Mice were immunized for EAE induction with Hooke kits (Hooke laboratories Inc, MA, USA) according to the manufacturer's instructions. Briefly, a volume of 0.1 ml emulsion of MOG_35–55 peptide in complete Freund's adjuvant was injected on either side of the back subcutaneously for each mouse (0.2 ml per animal). Additionally, at days 0 and 1, mice were administered 200 ng pertussis toxin via intraperitoneally. Among them, EAE-TNT group received 15 mg kg⁻¹ (in PBS) of TNT and to EAE-TFP group TFP through intraperitoneally on days 7 and 14 from the day of EAE induction. Clinical signs of EAE were assessed according to following score: 0, no signs of disease; 1, loss of tone in the tail; 2, hind limb weakness or partial paralysis; 3, complete hind limb paralysis; 4, front and hind limb paralysis; 5, moribund state.

Arthritis induction, treatment and assessment

Rats were randomly divided into three groups (n = 5), namely control, CIA and CIA-TNT. The animals were anesthetized with ketamine and then injected intradermally with 100 μl of the bovine type II collagen (2 mg ml⁻¹ in 0.05 M acetic acid) emulsified in Freund's complete adjuvant and Freund's incomplete adjuvant on days 0 and 7, respectively at the base of the tail. Subsequently, on the day of immunization rats were given subcutaneously 100 μl of PBS alone or with TNT (15 mg kg⁻¹ body weight) at the base of the tail. The levels of arthritis were evaluated according to the arthritis score every two days by two independent observer's up to the day of sacrifice (21 day). Score condition; 0 = normal, 1 = mild swelling and redness, 2 = moderate redness and swelling of ankle of wrist, 3 = severe redness and swelling of the entire paw including digits, 4 = maximally inflamed limb with involvement of multiple joints.

Histology

For histological analysis brain and spinal cord were fixed in 10% neutral buffered formalin, dehydrated in 70% ethanol and processed for paraffin embedding. 4 μm sections were cut on a microtome and placed on a glass slides, deparaffinised and stained with hematoxylin/eosin (H & E) and Luxol fast blue to evaluate inflammatory infiltrates and degree of demyelination.

For histopathology assessment, paws and knees were removed and fixed in 10% buffered formalin. Sections of paraffin-embedded ankle joints were prepared, stained with H & E and histophathological scoring was done based on density of resident stromal cells and inflammatory infiltrates. Score 0–1 was graded as normal or no synovitis, score of 2–4 as low grade synovitis and score 5–9 as high grade synovitis.⁵⁵

T cell proliferation assay

The effect of TNT and TFP on neural MOG_35–55 antigen-induced T cell proliferation was measured by MTT assay. To determine the ex vivo response, the mouse spleen cells were isolated on day 23 of control, EAE, EAE-TNT, EAE-TFP mice and cultured in RPMI medium supplemented with 10% FBS in 96-well plate (2 × 10⁵ per 200 μl per well) with 20 μg ml⁻¹ MOG_35–55 peptide. After 24 h, 48 h and 72 h incubation MTT assay was performed.

Detection of HBGB1 levels

Mice of EAE and rats of CIA were sacrificed at onset or peak of disease and blood was collected into a fresh tube by cardiac puncture. Blood samples were centrifuged for 15 min at 5000g, and serum was transferred to new tubes and stored at −80° until use.

Splenocytes (2 × 10⁶ cells per well) were prepared from different groups of mice and plated in 24 well plate. Cells were re-stimulated with MOG_35–55 peptide (20 μg ml⁻¹), culture supernatants were collected after 48 h and stored at −80 °C until use. HMGB1 cytokine concentration in serum and culture supernatants was measured using HMGB1 Detection Kit (Chondrex Inc., WA, USA) according to the manufacturer's instructions.

Statistical analysis

Each experiment was repeated three times with n = 5 animals per group. Values were expressed as means ± SD. Data were analyzed with the unpaired t tests and two-way analysis of variance (ANOVA), using Prism 5 software (GraphPad Software, CA). Statistical significance was defined as P < 0.05.

Results

Characterization of titania based materials

Transmission Electron Microscopy (TEM) was used to characterize the titania based materials such as TiO₂ fine particles (TFP) and titanate nanotubes (TNT) and images are displayed in Fig. 1. The TFP image shows flakes-like fine particles with high agglomeration. The particle size ranges from 60–160 nm, random dark spots in the images are due to alignment of particles one on the other (Fig. 1A). The TEM images of TNT shows nanotubular morphology having 3–5 layers of wrapped nanosheets, having cylindrical in shape with hollow inside and open at both ends. The inner diameter of tube is 3 to 4 nm and outer diameter having various sizes from 8 to 10 nm and length is between 100 to 300 nm as shown in Fig. 1B and C.

Fig. 1 TEM image of TFP and TNT nanomaterials.

The X-ray diffraction pattern of TFP exhibited characteristic peak at 2θ = 25.4° confirms the tetragonal structure with anatase phase (JCPDS no. 21–1272) as shown in Fig. 2A. The characteristic diffraction peak of TNT exhibited at 2θ = 10.2° indicates the presence of typical layered crystal structure (Fig. 2B). All other peaks centred at 2θ = 24.1, 28.3, and 48.2° can be well indexed as the monoclinic structure of H₂Ti₃O₇ (JCPDS no.47–0561). During hydrothermal synthesis, anatase phase of TFP particles undergoes dissolution in alkaline solution and crystallized the new material having layered H₂Ti₃O₇ phase with monoclinic structure. The BET surface area analysis of TFP and TNT showed interesting results such as 5.1 and 286 m² g⁻¹ respectively. The higher surface area value of TNT is ascribed to one dimensional hollow structure having adsorption sites both exterior and interior of nanotubes. These results are in tune with our earlier reports.⁵³

Fig. 2 XRD pattern of TFP and TNT nanomaterials.

Particle size and surface charge analysis

To understand the nature of surface charge and size of TFP or TNT, DLS and zeta potential analysis were carried out and results are displayed in Table 1. DLS data showed the particle size of TFP is 575 nm, which are about four folds higher than TEM analysis of the same particle, the higher size in suspension revealed the strong agglomeration behavior. On the other hand, TNT showed 295.3 nm, it is almost similar to TEM analysis, the size is explained through well dispersion in experimental medium. The zeta potential analysis of TFP and TNT showed −1.2 and −1.1 mV respectively.

Table 1 Zeta potential evaluation and DLS analysis for TFP and TNT

Sl. no.	Sample ID	Particle size analysis (nm)	Zeta potential (mV)
1	TFP	575	−1.2
2	TNT	295.3	−1.1

---

Nano-TiO₂ decreased HMGB1 secretion from LPS induced RAW cells

HMGB1 is a ubiquitous nuclear protein, recently recognized as a pro-inflammatory mediator and an actively secreted cytokine by macrophages and apoptotic/necrotic cells upon cell injury and infection.⁸ In this study we have investigated whether nano-TiO₂ inhibit the secretion of HMGB1 in LPS stimulated RAW 264.7 cells. As shown in the Fig. 3 enhanced HMGB1 secretion from LPS induced RAW 264.7 cells was observed and is co-related with previous studies.^56,57 Interestingly, cells treated with LPS and TNT or TFP at a concentration of 50 μg ml⁻¹ for 24 h, we observed decreased HMGB1 secretion into culture supernatants (Fig. 3). Recently, Neacsu et al. stated that TiO₂ nanotubes involved in the attenuation of inflammation via inhibition of mitogen-activated protein kinase (MAPK) nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) pathways in RAW 264.7 cells.⁵⁸ Previous studies reported that HMGB1 activates the MAPK-NF-κB pathway by interacting with RAGE, and that it plays an important role in inflammation.^59–61 Probably, TNT and TFP reduce the HMGB1 levels by inhibiting the MAPK-NF-κB pathways.

Fig. 3 Nanomaterials TNT and TFP decreased the release of HMGB1 in LPS-induced RAW cells. RAW 264.7 cells were treated with LPS in the presence or absence of TNT or TFP at a concentration of 50 μg ml⁻¹ for 24 h. After incubation, culture supernatant were collected and subjected to ELISA for quantification of HMGB1. Data are presented as mean ± SD of three independent experiments. Statistical significance was defined as *p < 0.05.

Cytotoxicity of TNT and TFP on murine splenocytes

MLR is a model of T-cell response to alloantigenic peptide complex with major histocompatibility (MHC) proteins on antigen presenting complex (APC). Splenocytes from C57BL/6 mice were cultured in the presence of 10, 25, 50, 75, 100, 125 and 150 μg ml⁻¹ concentration of TNT and TFP for 24 h. After 24 h treatment, splenocytes of C57BL/6 mice (H-2^b) were inactivated with mitomycin C and used as stimulators. In MLR, these stimulators were co-cultured with responder splenocytes of BALB/c mice (H-2^d) for 72 h. TNT (Fig. 4A) and TFP (Fig. 4B) inhibited MLR in a dose dependent manner, with an IC₅₀ value of 36.595 μg ml⁻¹ and 117.809 μg ml⁻¹ respectively for a 72 h co-culture. These results suggest that TNT and TFP probably block T cell mediated responses in vitro.

Fig. 4 Cytotoxicity of TNT and TFP in mixed lymphocyte reaction (MLR) of mice splenocytes using the MTT assay on a 72 h culture. TNT (A) and TFP (B) inhibited MLR in a dose dependent manner, with an IC₅₀ value of 36.5 μg ml⁻¹ and 117.8 μg ml⁻¹ respectively for a 72 h co-culture.

TNT and TFP ameliorates the EAE

C57BL/6 mice were used to determine the effect of TNT (protinic trititanate nanotubes) and TFP (TiO₂ fine particles) on evolution of EAE. Upon immunization with MOG_35–55 peptide, mice developed clinical signs of EAE around day 8, reaching the peak of severity at about 18 days post induction. TNT or TFP (15 mg kg⁻¹) administration on days 7 and 14 after EAE induction resulted a significant decrease in the severity of the disease according to the EAE score. TNT and TFP treated mice showed a mean clinical score 1.1 and 1.6 respectively at the peak of disease (day 18) compared with 3.5 clinical score of untreated mice as shown in Fig. 5A. Therefore compared with TFP treated mice, TNT treated mice effectively barred the EAE development.

Fig. 5 EAE-TNT, EAE-TFP mice developed an attenuated and delayed course of EAE. C57BL/6 mice were induced to develop EAE by immunization with MOG_35–55 peptide and treated one group with 15 mg kg⁻¹ of TNT and other group with TFP on the days 7 and 14 via intraperitoneally (A). The clinical scores were evaluated daily and were plotted as the mean ± SD (n = 5 per group). Maximum clinical scores as well as scores on day 18 and 22 evidence marked attenuation of disease severity after TNT and TFP administration. (B) Spleen cells were isolated on day 23 from each group and stimulated with MOG_35–55 antigen (20 μg ml⁻¹) ex vivo for 24 h, 48 h and 72 h. Proliferation response was measured by MTT assay. The data are presented as mean ± SD (*p < 0.05).

TNT and TFP inhibits infiltration of inflammatory cells and degree of demyelination in the spinal cord

The effect of TNT and TFP treatment on central nervous system (CNS) infiltration was determined by H & E staining in the cross-section of the spinal cord. White matter of EAE mice spinal cord shows multiple foci of chronic inflammatory infiltrate with perivascular round cell collection and focal vacuolar degeneration. In contrast, TNT and TFP treated mice exhibited markedly decreased infiltration of inflammatory cells and focal vacuolar degeneration in the white matter of spinal cord (Fig. 7A).

To determine the degree of demyelination we stained sections of spinal cord with Luxol fast blue/cresyl echt violet and observed wide spread demyelination zones in the white matter of spinal cord of EAE mice. In contrast, mice received TNT had minimal evidence of demyelination. Whereas occasional demyelination seen in grey matter of TFP treated spinal cord which is lesser than EAE group, is indicated by a markedly attenuated course of disease (Fig. 7B).

TNT and TFP inhibit ex vivo spleen cell proliferation

To investigate the mechanism in the regulation of EAE by TNT and TFP, we examined neural antigen-induced T cell proliferation in 24 h, 48 h and 72 h culture. Splenocytes were isolated from each group of mice on day 23 and re-stimulated with MOG_35–55 (20 μg ml⁻¹) in vitro. When compared to EAE cells, spleen cells from EAE-TNT, EAE-TFP mice showed significant decrease in the T cells proliferation in response to ex vivo re-stimulation (Fig. 5B). These results suggest that TNT and TFP ameliorate EAE by inhibiting the expansion of neural antigen specific T cells in C57BL/6 mice.

TNT and TFP suppress the HMGB1 cytokine production in EAE

HMGB1 released from activated immune cells or damaged, dying cells during necrosis and during the late phase of cellular apoptosis, is now recognized as a serum biomarker for EAE.²⁰ The expression and release of HMGB1 are significantly increased in various stages of EAE.⁶² We aimed to test whether the administration of TNT/TFP on days 7 and 14 from the day of immunization suppress the HMGB1 production, as HMGB1 levels correlate with disease progression. C57BL/6 mice immunized with MOG_35–55/CFA were sacrificed at onset or peak of clinical disease and serum HMGB1 was quantified by ELISA. Compared to TNT/TFP treated EAE mice, control untreated EAE mice had significantly elevated levels of HMGB1 in serum (Fig. 6A, p < 0.05). We further tested whether HMGB1 in peripheral blood correlated with ex vivo stimulation of splenocytes with MOG_35–55 (20 μg ml⁻¹). We observed higher concentrations of HMGB1 in control untreated EAE splenocytes alone and in the presence of MOG_35–55 (20 μg ml⁻¹) when compared with TNT/TFP treated EAE mice (Fig. 6B, p < 0.05). These results indicate that TNT/TFP inhibit the HMGB1 secretion result in the reduction of disease pathogenesis.

Fig. 6 TNT and TFP suppress the HMGB1 cytokine production in EAE. (A) The immunized mice were euthanized on day 23, the serum was collected from each group of mice and amount of HMGB1 was analyzed by ELISA. The concentration of HMGB1 was calculated using the standard plot and shown as mean ± SD (p < 0.05). (B) Spleen cells were cultured with MOG_35–55 peptide (20 μg ml⁻¹) ex vivo for 48 h, culture supernatants were collected and concentration of HMGB1 was determined by ELISA. The data are presented as mean ± SD (*p < 0.05).

Ameliorating function of TNT on arthritis model

We tested whether in addition to EAE model, TNT would also be effective for other autoimmune disease model. We used CIA model to evaluate the attenuation effect of TNT. Groups of 5 rats were immunized with collagen on days 0 and 7, subsequently on the same day of immunization injection of TNT (15 mg kg⁻¹) via subcutaneously. Fig. 8A arthritis score shows untreated CIA group rats developed arthritis beginning from day 8 onwards and severe ankle swelling reaches on the day 18. However, TNT-treated group were significantly attenuated the incidence of arthritis (∼50%) and ankle swelling. The mean maximum arthritis score of CIA and CIA-TNT are 4 and 2 respectively.

Fig. 7 Attenuation of inflammation progression and demyelination in the CNS region of mice that received EAE-TNT and EAE-TFP, spinal cords from each group of mice were removed on day 23. In EAE-TNT mice, the number of immune-cell infiltrates (H & E, (A-c)) and demyelination (Luxol fast blue, (B-c)) were both significantly reduced. (A) Hematoxylin and eosin staining. (B) Luxol fast blue staining. (a) Control, (b) EAE, (c) EAE-TNT, (d) EAE-TFP.

Fig. 8 Amelioration functions of TNT on CIA model. Rats were immunized with collagen on days 0 and 7, followed by administration of TNT (15 mg kg⁻¹) subcutaneously on day 0 and 7. Arthritis score, the levels of arthritis measurements were taken every two days (A). At the day 21 all rats were sacrificed and blood was collected by cardiac puncture, serum was separated and quantified the HMGB1 cytokine (B). TNT inhibit the proinflammatory HMGB1 levels in CIA-TNT model and protects from the inflammatory arthritis. The data are presented as mean ± SD (*p < 0.05).

The effect of TNT treatment on the histological changes in the ankle joints of rats and synovial tissue of the knee with CIA after the animals had been sacrificed on day 21 as shown in Fig. 9A and B. Histological evaluation showed that TNT inhibited synovial hyperplasia, inflammatory cell infiltration, cartilage erosion, and bone destruction, which were observed in CIA rats. Taken together, these results indicate that TNT administration results in a significant reduction of joint-tissue inflammation.

Fig. 9 Representative hematoxylin and eosin micrographs of joint tissue in arthritis model compared with TNT treated rats with CIA. Ankle joint tissue (A), synovium tissue of knee (B), showed synovial hyperplasia and infiltration of inflammatory cells in untreated CIA and relatively less or no inflammatory cells, damage to the synovial membrane of TNT-CIA.

TNT inhibit the HMGB1 production in CIA

It has been demonstrated that HMGB-1 is the key proinflammatory cytokine that plays a crucial role in experimental arthritis models as well as in patients with arthritis.^63,64 Increased concentration of HMGB1 in CIA may serve as a biomarker for arthritis.⁶⁵ Serum was collected from 3 groups of rats on day 21. As shown in Fig. 8B, compared with control rat, rat with CIA showed increased circulating levels of HMGB1 in serum. In contrast CIA-TNT treated rat shows significantly reduced level of HMGB1. These results indicate that TNT treatment inhibit the HMGB1 secretion result in the reduction of disease pathogenesis of CIA.

Discussion

EAE is the widely used animal model for MS. MS is driven by myelin-specific auto-reactive T cells that infiltrate the CNS and mediate an inflammatory response that result in demyelination and axon degradation.⁶⁶ EAE can be induced by immunization with a variety of myelin antigens. Among those MOG_35–55 is an important candidate and MOG-reactive T cells also play significant roles in the pathogenesis of MS.⁶⁷ C57BL/6 mice develop chronic disease following immunization with MOG_35–55 peptide.

In this study, we determined the role of TNT and TFP in the regulation of EAE model of MS. The administration of TNT and TFP on the day 7 and 14, appearance of clinical signs of EAE could control the evolution of the disease. To study the suppression of EAE by TNT and TFP, we analyzed the in vitro effect on the T cell recall response to MOG_35–55 peptide. Treatment with TNT and TFP resulted in a significantly decreased proliferation by MOG_35–55 in vitro experiments; this decreased proliferation was significantly stronger in TNT treated group. HMGB1 is a DNA-binding protein with proinflammatory properties, contributes to neuroinflammatory responses that drive EAE pathogenesis and that HMGB1 blockade may be a novel means to selectively disrupt the proinflammatory loop that drives MS autoimmunity. In the present study, extracellular HMGB1 was found to be increased in the sera and culture supernatant of ex vivo re-stimulation with MOG_35–55 (20 μg ml⁻¹) of EAE. HMGB1 levels co-relate with the disease severity of the EAE score, implicating a dynamic systemic inflammatory response. Previous studies have shown that, anti-HMGB1 antibody ameliorates EAE.²¹ This is consistent with our HMGB1 cytokine levels in sera and in vitro re-stimulated culture. Histopathology results indicated that mice were rescued from EAE with less or no inflammatory cells as well as demyelination lesions in TNT and TFP treated EAE that are observed in EAE mice. TFP treated EAE found that relatively high inflammatory cells as well as demyelination lesions compared to TNT treated EAE.

We tested whether in addition to EAE model, TNT would also be effective for other autoimmune disease model. We demonstrated that the effect of TNT on CIA model, which has been the most widely used model of RA. This model has shortest duration between immunization and disease manifestation and shares several clinical, pathophysiological features with RA. Control SD rats were immunized with collagen on days 0 and 7 and CIA-TNT rats immunized with collagen along with TNT (15 mg kg⁻¹) on day 0 and 7. We observed that late onset with low severity of disease signs in CIA-TNT model compared with CIA model. Representative histological images of ankle joint tissue and knee synovial tissue showed that the administration of TNT into CIA model inhibit the cartilage degeneration, synovial hyperplasia with infiltration of inflammatory cells into the synovial tissue, which were observed in CIA model images. These results co-relate with the HMGB1 levels in the serum separated from CIA-TNT group rat blood. HMGB1 is a novel proinflammatory cytokine, involved in the pathogenesis of RA. Extracellular HMGB1 induces the secretion of proinflammatory cytokine TNF, IL-1, and IL-6 from macrophages/damaged cells.¹² Taniguchi et al. reported that high level of HMGB1 expression in the synovium of RA patients as well as adjuvant-induced arthritis and CIA.⁶³ Administration of HMGB1 into mice joints, itself induce joint inflammation by activating monocytes/macrophages and inducing proinflammatory cytokines leads to arthritis changes.⁶⁸ Experimental arthritis could be effectively treated by the administration of polyclonal and monoclonal anti-HMGB1 antibodies which are specific for the HMGB1 cytokine.^22,69 HMGB1 play a key role in the development and disease progression of the arthritis. Interestingly, our results indicated that CIA serum shows high levels of HMGB1 cytokine compared with TNT treated group of rats. With this we found that TNT effectively reduced the clinical signs and pathophysiology of arthritis in CIA-TNT model. To the best of our knowledge, this is the first study to show that administration of TNT/TFP can reduce the EAE and CIA pathogenesis.

In conclusion, our findings suggest that Ti–O based nanomaterials administration ameliorated the clinical severity of EAE (TNT/TFP), CIA (TNT) significantly by ameliorating pathology, and presumably attenuating the immune response via HMGB1 cytokine release. Finally we suggest that, TNT and TFP may have therapeutic potential not only for MS, RA but also for other autoimmune disorders.

Acknowledgements

This work was supported by grants from Science and Engineering Research Board (SERB), India (Grant #: SR/FT/LS-149/2010) and Council of scientific and industrial research (CSIR), India, (Grant# 37(1517)/11/EMR-II) to Dakshayani Lomada and University Grants Commission, New Delhi (UGC F. No: 42-176/2013(SR)) to Madhava C Reddy.

References

1. A. Compston and A. Coles, Lancet, 2008, 372, 1502–1517.
2. V. Browning, M. Joseph and M. Sedrak, JAAPA: Journal of the American Academy of Physician Assistants, 2012, 25, 24–29.
3. T. M. Rivers, D. H. Sprunt and G. P. Berry, J. Exp. Med., 1933, 58, 39–53.
4. I. Mendel, N. Kerlero de Rosbo and A. Ben-Nun, Eur. J. Immunol., 1995, 25, 1951–1959.
5. B. Joe and R. L. Wilder, Mol. Med. Today, 1999, 5, 367–369.
6. P. Brooks and P. Kubler, Ther. Clin. Risk Manage., 2006, 2, 45–57.
7. P. E. Lipsky, L. S. Davis, J. J. Cush and N. Oppenheimer-Marks, Springer Semin. Immunopathol., 1989, 11, 123–162.
8. R. Kang, R. Chen, Q. Zhang, W. Hou, S. Wu, L. Cao, J. Huang, Y. Yu, X. G. Fan, Z. Yan, X. Sun, H. Wang, Q. Wang, A. Tsung, T. R. Billiar, H. J. Zeh III, M. T. Lotze and D. Tang, Mol. Aspects Med., 2014, 40, 1–116.
9. M. C. Reddy and K. M. Vasquez, Radiat. Res., 2005, 164, 345–356.
10. P. Scaffidi, T. Misteli and M. E. Bianchi, Nature, 2002, 418, 191–195.
11. J. Tian, A. M. Avalos, S. Y. Mao, B. Chen, K. Senthil, H. Wu, P. Parroche, S. Drabic, D. Golenbock, C. Sirois, J. Hua, L. L. An, L. Audoly, G. La Rosa, A. Bierhaus, P. Naworth, A. Marshak-Rothstein, M. K. Crow, K. A. Fitzgerald, E. Latz, P. A. Kiener and A. J. Coyle, Nat. Immunol., 2007, 8, 487–496.
12. U. Andersson, H. Wang, K. Palmblad, A. C. Aveberger, O. Bloom, H. Erlandsson-Harris, A. Janson, R. Kokkola, M. Zhang, H. Yang and K. J. Tracey, J. Exp. Med., 2000, 192, 565–570.
13. A. Rouhiainen, J. Kuja-Panula, E. Wilkman, J. Pakkanen, J. Stenfors, R. K. Tuominen, M. Lepantalo, O. Carpen, J. Parkkinen and H. Rauvala, Blood, 2004, 104, 1174–1182.
14. M. Magna and D. S. Pisetsky, Mol. Med., 2014, 20, 138–146.
15. E. Sundberg, A. E. Fasth, K. Palmblad, H. E. Harris and U. Andersson, Immunobiology, 2009, 214, 303–309.
16. J. M. Reynolds, G. J. Martinez, Y. Chung and C. Dong, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 13064–13069.
17. I. Stojanovic, S. Cuzzocrea, K. Mangano, E. Mazzon, D. Miljkovic, M. Wang, M. Donia, Y. Al Abed, J. Kim, F. Nicoletti, S. Stosic-Grujicic and M. Claesson, Clin. Immunol., 2007, 123, 311–323.
18. A. Andersson, R. Covacu, D. Sunnemark, A. I. Danilov, A. Dal Bianco, M. Khademi, E. Wallstrom, A. Lobell, L. Brundin, H. Lassmann and R. A. Harris, J. Leukocyte Biol., 2008, 84, 1248–1255.
19. R. S. Goldstein, A. Bruchfeld, L. Yang, A. R. Qureshi, M. Gallowitsch-Puerta, N. B. Patel, B. J. Huston, S. Chavan, M. Rosas-Ballina, P. K. Gregersen, C. J. Czura, R. P. Sloan, A. E. Sama and K. J. Tracey, Mol. Med., 2007, 13, 210–215.
20. A. P. Robinson, M. W. Caldis, C. T. Harp, G. E. Goings and S. D. Miller, J. Autoimmun., 2013, 43, 32–43.
21. A. Uzawa, M. Mori, J. Taniguchi, S. Masuda, M. Muto and S. Kuwabara, Clin. Exp. Immunol., 2013, 172, 37–43.
22. R. Kokkola, J. Li, E. Sundberg, A. C. Aveberger, K. Palmblad, H. Yang, K. J. Tracey, U. Andersson and H. E. Harris, Arthritis Rheum., 2003, 48, 2052–2058.
23. S. Malhotra, N. Fissolo, M. Tintore, A. C. Wing, J. Castillo, A. Vidal-Jordana, X. Montalban and M. Comabella, J. Neuroinflammation, 2015, 12, 48.
24. J. M. Escandell, M. C. Recio, S. Manez, R. M. Giner, M. Cerda-Nicolas and J. L. Rios, J. Pharmacol. Exp. Ther., 2007, 320, 581–590.
25. T. Minko, L. Rodriguez-Rodriguez and V. Pozharov, Adv. Drug Delivery Rev., 2013, 65, 1880–1895.
26. Y. Diebold, M. Jarrin, V. Saez, E. L. Carvalho, M. Orea, M. Calonge, B. Seijo and M. J. Alonso, Biomaterials, 2007, 28, 1553–1564.
27. P. Debbage, Curr. Pharm. Des., 2009, 15, 153–172.
28. R. A. Maldonado, R. A. LaMothe, J. D. Ferrari, A. H. Zhang, R. J. Rossi, P. N. Kolte, A. P. Griset, C. O'Neil, D. H. Altreuter, E. Browning, L. Johnston, O. C. Farokhzad, R. Langer, D. W. Scott, U. H. von Andrian and T. K. Kishimoto, Proc. Natl. Acad. Sci. U. S. A., 2015, 112, E156–E165.
29. A. Yeste, M. Nadeau, E. J. Burns, H. L. Weiner and F. J. Quintana, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 11270–11275.
30. M. Donia, K. Mangano, C. Quattrocchi, P. Fagone, S. Signorelli, G. Magro, A. Sfacteria, K. Bendtzen and F. Nicoletti, Scand. J. Immunol., 2010, 72, 396–407.
31. T. Takagi, P. W. Tsao, R. Totsuka, T. Suzuki, T. Murata and I. Takata, Jpn. J. Pharmacol., 1998, 78, 225–228.
32. H. W. Minne, J. Pfeilschifter, S. Scharla, S. Mutschelknauss, A. Schwarz, B. Krempien and R. Ziegler, Endocrinology, 1984, 115, 50–54.
33. K. Mangano, A. Nicoletti, F. Patti, M. Donia, L. Malaguarnera, S. Signorelli, G. Magro, V. Muzio, B. Greco, P. Zaratin, P. Meroni, M. Zappia and F. Nicoletti, Clin. Exp. Immunol., 2010, 159, 159–168.
34. J. H. Duarte, Nat. Rev. Rheumatol., 2015, 11, 319.
35. A. L. Dellinger, P. Cunin, D. Lee, A. L. Kung, D. B. Brooks, Z. Zhou, P. A. Nigrovic and C. L. Kepley, PLoS One, 2015, 10, e0126290.
36. K. Yudoh, R. Karasawa, K. Masuko and T. Kato, Int. J. Nanomed., 2009, 4, 217–225.
37. C. Y. Tsai, A. L. Shiau, S. Y. Chen, Y. H. Chen, P. C. Cheng, M. Y. Chang, D. H. Chen, C. H. Chou, C. R. Wang and C. L. Wu, Arthritis Rheum., 2007, 56, 544–554.
38. L. Leonaviciene, G. Kirdaite, R. Bradunaite, D. Vaitkiene, A. Vasiliauskas, D. Zabulyte, A. Ramanaviciene, A. Ramanavicius, T. Asmenavicius and Z. Mackiewicz, Medicina, 2012, 48, 91–101.
39. N. Nagai and Y. Ito, Biol. Pharm. Bull., 2014, 37, 96–104.
40. W. U. Kim, W. K. Lee, J. W. Ryoo, S. H. Kim, J. Kim, J. Youn, S. Y. Min, E. Y. Bae, S. Y. Hwang, S. H. Park, C. S. Cho, J. S. Park and H. Y. Kim, Arthritis Rheum., 2002, 46, 1109–1120.
41. S. M. Moghimi, A. C. Hunter and J. C. Murray, FASEB J., 2005, 19, 311–330.
42. N. Sanvicens and M. P. Marco, Trends Biotechnol., 2008, 26, 425–433.
43. W. W. Brien, E. A. Salvati, F. Betts, P. Bullough, T. Wright, C. Rimnac, R. Buly and K. Garvin, Clin. Orthop. Relat. Res., 1992, 66–74.
44. R. S. Flatebo, A. C. Johannessen, A. G. Gronningsaeter, O. E. Boe, N. R. Gjerdet, B. Grung and K. N. Leknes, J. Periodontol., 2006, 77, 1201–1210.
45. H. Schliephake, H. Lehmann, U. Kunz and R. Schmelzeisen, Int. J. Oral Surg., 1993, 22, 20–25.
46. Y. Chen, Y. Wan, Y. Wang, H. Zhang and Z. Jiao, Int. J. Nanomed., 2011, 6, 2321–2326.
47. J. Xu, Y. Sun, J. Huang, C. Chen, G. Liu, Y. Jiang, Y. Zhao and Z. Jiang, Bioelectrochemistry, 2007, 71, 217–222.
48. B. C. Schanen, A. S. Karakoti, S. Seal, D. R. Drake III, W. L. Warren and W. T. Self, ACS Nano, 2009, 3, 2523–2532.
49. R. K. Shukla, A. Kumar, D. Gurbani, A. K. Pandey, S. Singh and A. Dhawan, Nanotoxicology, 2013, 7, 48–60.
50. J. Xu, H. Shi, M. Ruth, H. Yu, L. Lazar, B. Zou, C. Yang, A. Wu and J. Zhao, PLoS One, 2013, 8, e70618.
51. H. Shi, R. Magaye, V. Castranova and J. Zhao, Part. Fibre Toxicol., 2013, 10, 15.
52. G. Oberdorster, J. Ferin and B. E. Lehnert, Environ. Health Perspect., 1994, 102(suppl. 5), 173–179.
53. D. Praveen Kumar, M. V. Shankar, M. M. Kumari, G. Sadanandam, B. Srinivas and V. Durgakumari, Chem. Commun., 2013, 49, 9443–9445.
54. T. Itoh, K. Ishii, T. Irikura, Y. Ueno, A. Kojima and Y. Horie, J. Antibiot., 1993, 46, 1575–1581.
55. V. Krenn, L. Morawietz, G. R. Burmester, R. W. Kinne, U. Mueller-Ladner, B. Muller and T. Haupl, Histopathology, 2006, 49, 358–364.
56. D. Tang, Y. Shi, R. Kang, T. Li, W. Xiao, H. Wang and X. Xiao, J. Leukocyte Biol., 2007, 81, 741–747.
57. S. Cao, S. Li, H. Li, L. Xiong, Y. Zhou, J. Fan, X. Yu and H. Mao, PLoS One, 2013, 8, e54647.
58. P. Neacsu, A. Mazare, P. Schmuki and A. Cimpean, Int. J. Nanomed., 2015, 10, 6455–6467.
59. J. A. Nogueira-Machado, C. M. Volpe, C. A. Veloso and M. M. Chaves, Expert Opin. Ther. Targets, 2011, 15, 1023–1035.
60. U. Andersson and H. Rauvala, J. Intern. Med., 2011, 270, 296–300.
61. R. Vitali, L. Stronati, A. Negroni, G. Di Nardo, M. Pierdomenico, E. del Giudice, P. Rossi and S. Cucchiara, Am. J. Gastroenterol., 2011, 106, 2029–2040.
62. Y. Sun, H. Chen, J. Dai, H. Zou, M. Gao, H. Wu, B. Ming, L. Lai, Y. Xiao, P. Xiong, Y. Xu, F. Gong and F. Zheng, J. Neuroimmunol., 2015, 280, 29–35.
63. N. Taniguchi, K. Kawahara, K. Yone, T. Hashiguchi, M. Yamakuchi, M. Goto, K. Inoue, S. Yamada, K. Ijiri, S. Matsunaga, T. Nakajima, S. Komiya and I. Maruyama, Arthritis Rheum., 2003, 48, 971–981.
64. R. Kokkola, E. Sundberg, A. K. Ulfgren, K. Palmblad, J. Li, H. Wang, L. Ulloa, H. Yang, X. J. Yan, R. Furie, N. Chiorazzi, K. J. Tracey, U. Andersson and H. E. Harris, Arthritis Rheum., 2002, 46, 2598–2603.
65. U. Andersson and H. E. Harris, Biochim. Biophys. Acta, 2010, 1799, 141–148.
66. C. S. Constantinescu, N. Farooqi, K. O'Brien and B. Gran, Br. J. Pharmacol., 2011, 164, 1079–1106.
67. H. Shao, Z. Huang, S. L. Sun, H. J. Kaplan and D. Sun, Invest. Ophthalmol. Visual Sci., 2004, 45, 4060–4065.
68. R. Pullerits, I. M. Jonsson, M. Verdrengh, M. Bokarewa, U. Andersson, H. Erlandsson-Harris and A. Tarkowski, Arthritis Rheum., 2003, 48, 1693–1700.
69. H. Schierbeck, P. Lundback, K. Palmblad, L. Klevenvall, H. Erlandsson-Harris, U. Andersson and L. Ottosson, Mol. Med., 2011, 17, 1039–1044.

---