OP0097 PROOF-OF-CONCEPT OF FIBROBLAST-TARGETED PHOTODYNAMIC THERAPY IN HUMAN 3D SCLERODERMA SKIN MODEL AND BLEOMYCIN-INDUCED SKIN FIBROSIS IN MICE

Background

Systemic sclerosis (SSc) is a rare, severe auto-immune disease characterized by inflammation, vasculopathy and fibrosis. Activated (myo)fibroblasts are crucial drivers of fibrosis. By exploiting their expression of fibroblast activation protein (FAP) to perform targeted photodynamic therapy (tPDT), we can locoregionally deplete these pathogenic cells.

Objectives

We explored the use of FAP-tPDT to selectively target primary skin fibroblasts from SSc patients, both in 2D and 3D cultures, as well as in biopsies from fibrotic skin lesions in the murine bleomycin-induced skin fibrosis model.

Methods

The FAP targeting monoclonal antibody (clone 28H1) was conjugated with the photosensitizer IRDye700DX. Primary skin fibroblasts were obtained from lesional skin biopsies of SSc patients via spontaneous out-growth and subsequently cultured on plastic or collagen type I. For 2D FAP-tPDT, cells were incubated in buffer with or without the antibody-photosensitizer (Ab-PS) construct, washed after 4 h and exposed to λ = 689 nm light (50 J/cm2 at 280 mW/cm2). Cell viability was measured using CellTiter Glo. For 3D FAP-tPDT, cells were seeded in collagen plugs and underwent the same treatment procedure. Contraction of the plugs was followed over time to determine myofibroblast activity. Skin fibrosis was induced in mice by 3x/week injections of 1.5 IU bleomycin intradermally on the back of the mouse. After 5 weeks the mice were sacrificed and biopsies were taken from affected and not affected skin. These were incubated with or without the FAP targeting Ab-PS construct or a control Ab-PS. After washing, the biopsies were exposed to light, incubated for 1h and subsequently formalin fixed and paraffin embedded. Sections were immuno stained for the presence of cleaved caspase-3.

Results

FAP-tPDT resulted in antibody-dose dependent cytotoxicity in primary skin fibroblasts upon light exposure. Cells not exposed to light or incubated with a control Ab-PS construct did not show this response. FAP-tPDT fully prevented contraction of collagen plugs seeded with primary SSc fibroblasts. Even incubation with a very low dose of antibody (0.4 nM) inhibited contraction in 2 out of 3 donors. In fibrotic skin biopsies from mice with bleomycin-induced skin fibrosis, an upregulation of apoptosis, as evidenced by increased caspase-3 staining (Figure 1), was observed in response to FAP-tPDT. The same treatment on biopsies of control skin did not increase caspase-3 staining, nor did incubation with a control Ab-PS construct or light alone.

Figure 1.

Cleaved caspase-3 staining of biopsies from the skin of mice with bleomycin-induced dermal fibrosis upon treatment with FAP-tPDT. The red arrows indicate the presence of positive fibroblasts.

Conclusion

Here we have shown, for the first time, the potential of FAP-tPDT for the selective targeting of pathogenic fibroblasts and treatment of fibrosis in SSc skin both in relevant patient-derived culture models as well as an in vivo model of fibrosis.

Disclosure of Interests

Daphne Dorst: None declared, Arjan van Caam: None declared, Elly Vitters: None declared, Birgitte Walgreen: None declared, Monique Helsen: None declared, Christian Klein Employee of: CK is an employee of Roche pharmaceutics, Shreya Gudi: None declared, Tirza Wubs: None declared, Jyoti Kumari: None declared, Przemysław Błyszczuk: None declared, Madelon Vonk: None declared, Peter van der Kraan: None declared, Marije Koenders: None declared.

AB0139 DEEP IMMUNE PHENOTYPING OF T LYMPHOCYTE SUBSETS IN SYSTEMIC SCLEROSIS PATHOPHYSIOLOGY AND FOLLOWING RESPONSE TO TARGETED CYTOTOXIC TREATMENT

Background

Abnormalities in T lymphocyte populations are associated with the pathogenesis of many autoimmune diseases, such as systemic sclerosis (SSc). Various studies report on the aberrations of different T cell cytotoxic (CTL) and helper (Th) subsets that appear to be linked with inflammatory and/or fibrotic manifestations of patients with SSc. Since T cells seem to play a pivotal role in the pathophysiology of SSc, targeting the pathogenic T cell subsets might be a promising therapeutic option.

Objectives

Here we set out to comprehensively compare T lymphocyte phenotypes between SSc patients and healthy donors. We further test the in-vitro efficacy of a combination of anti-CD3/CD7 immunotoxins (CD3/CD7-IT), that have been developed to eliminate activated CD4+ and CD8+ T cells, to study specific sensitivity of T-cell subpopulations to CD3/CD7-IT.

Methods

30 SSc patients and 15 age and sex matched healthy donors were included. Of these patients, lymphocyte populations were quantified by 17-parameter flow cytometry of peripheral blood mononuclear cells (PBMCs) to identify CD4+ T helper cells (Th1, Th2, Th17, T peripheral helper), CD8+ naïve, memory, effector CTLs and senescent/exhausted subsets. We next developed a cell killing assay to evaluate the effect of T cell depletion. To address this, patients’ (N=6) PBMCs were first activated for 24 hours in the presence of phytohemagglutinin, followed by CD3/CD7-IT addition for 48 hours. Subset-specific T cell depletion was assessed by using a combination of CellTiter-Glow luminescent cell viability assay and multi-parameter flow cytometric (FCM) quantification of CD3/CD7-IT-induced cell death.

Results

Frequencies of effector CD8+ CTLs,Th2 and T peripheral helper cells were elevated in SSc patients compared to healthy controls. Furthermore, SSc patients exhibited lower percentages of the anti-fibrotic Th1 subset. A striking expansion of the senescent CD4+CD28- and CD8+CD28- populations was noted in patients, while these subsets were barely detectable in healthy controls. In-vitro adittion of anti-T cell immunotoxins effectively depleted 50 % of patients’ CD8+ T cells (including the CD8 effector subset) and 62% of CD4+ T cells (including Th2 and T peripheral helper cells). No difference in cytolytic sensitivity between different T cell subsets was observed.

Conclusion

Our findings demonstrate that deep FCM immunephenotyping reveals pathophysiological differences in peripheral T cell subsets of SSc patients. Strikingly, the developed cytolytic assays show that CD3/CD7-IT is able to target the potential disease-associated T cell subsets in an in-vitro setting.

References

Not applicable

Disclosure of Interests

None declared

Osteoarthritis-Related Inflammation Blocks TGF-β's Protective Effect on Chondrocyte Hypertrophy via (de)Phosphorylation of the SMAD2/3 Linker Region

Osteoarthritis (OA) is a degenerative joint disease characterized by irreversible cartilage damage, inflammation and altered chondrocyte phenotype. Transforming growth factor-β (TGF-β) signaling via SMAD2/3 is crucial for blocking hypertrophy. The post-translational modifications of these SMAD proteins in the linker domain regulate their function and these can be triggered by inflammation through the activation of kinases or phosphatases. Therefore, we investigated if OA-related inflammation affects TGF-β signaling via SMAD2/3 linker-modifications in chondrocytes. We found that both Interleukin (IL)-1β and OA-synovium conditioned medium negated SMAD2/3 transcriptional activity in chondrocytes. This inhibition of TGF-β signaling was enhanced if SMAD3 could not be phosphorylated on Ser213 in the linker region and the inhibition by IL-1β was less if the SMAD3 linker could not be phosphorylated at Ser204. Our study shows evidence that inflammation inhibits SMAD2/3 signaling in chondrocytes via SMAD linker (de)-phosphorylation. The involvement of linker region modifications may represent a new therapeutic target for OA.

AB0082 INHIBITION OF TGFβ SIGNALING USING SB-505124 BLOCKS TH17 DIFFERENTIATION AND RESTORES THE TH17/TREG BALANCE IN VIVO, BUT DOES NOT SUPPRESS EXPERIMENTAL ARTHRITIS

Background:

TGFβ is an important growth factor that promotes the differentiation of T helper 17 (Th17) as well as regulatory T-cells (Treg). Due to its dual role, the potential of TGFβ as therapeutic target is unclear.

Objectives:

In this study we aimed to investigate the effect of inhibition of TGFβ signaling with the ALK5 inhibitor SB-505124 on human Th17 differentiationin vitro, on cytokine production by human rheumatoid arthritis (RA) synovial explants, and study the effects of local SB-505124 treatmentin vivoduring innate immune and Th17-driven experimental arthritis models.

Methods:

Magnetic sorted naïve human T cells were differentiated into Th17 cells with CD3/CD28 activation beads, IL-2, TGFβ, IL-1β, IL-23, αIFNƳ and αIL-4 for 6 days. Human RA synovial biopsies were cultured for 24h w/o 5µM SB-505124, and supernatant was analyzed by Luminex. T cell-independent SCW arthritis and Th17-driven IL-1/mBSA arthritis were induced in C57Bl6, and mice were treated with SB-505124 by daily intra-articular injections from day 0-4.

Results:

SB-505124 potently reduced human Th17 differentiationin vitroby decreasing IL-17 and RORƳt gene expression and IL-17 protein production. SB-505124 significantly suppressed IL-6 and TNFα protein production by human RA synovial explants. In addition, SB-505124 did not affect acute joint inflammation during SCW-arthritis (T-cell independent model). Interestingly, SB-505124 reduced Th17 levels in draining lymph nodes (dLN) during IL-1/mBSA arthritis while increased levels of Tregs were observed. Surprisingly, despite this skewed Th17/Treg balance, this did not result in suppression of joint inflammation and destruction in this Th17-driven arthritis model, whereas anti-IL-17 antibody treatment showed significant therapeutic effects.

Conclusion:

We revealed suppressive effects of SB-505124 on human Th17 differentiation and the Th17/Treg balance in arthritic mice. However, SB-505124 did not suppress joint inflammation and destruction. This indicates that despite the importance of TGFβ in Th17 differentiation, targeting TGFβ signaling is not enough to suppress experimental arthritis.

Disclosure of Interests:

None declared

TGFβ-mediated expression of TGFβ-activating integrins in SSc monocytes: disturbed activation of latent TGFβ?

Introduction

The pathophysiology of systemic sclerosis (SSc) is closely linked to overactive TGFβ signaling. TGFβ is produced and circulates in latent form, making its activation crucial for signaling. This activation can be mediated via integrins. We investigated the balance between active and latent TGFβ in serum of SSc patients and investigated if this correlates with integrin expression on monocytes.

Methods

A TGFβ/SMAD3- or BMP/SMAD1/5-luciferase reporter construct was expressed in primary human skin fibroblasts. Both acidified and non-acidified sera of ten SSc patients and ten healthy controls were tested on these cells to determine total and active TGFβ and BMP levels respectively. A pan-specific TGFβ1/2/3 neutralizing antibody was used to confirm TGFβ signaling. Monocytes of 20 SSc patients were isolated using CD14+ positive selection, and integrin gene expression was measured using qPCR. Integrin expression was modulated using rhTGFβ1 or a small molecule inhibitor of TGFBR1: SB-505124.

Results

SSc sera induced 50% less SMAD3-reporter activity than control sera. Serum acidification increased reporter activity, but a difference between healthy control and SSc serum was no longer observed, indicating that total TGFβ levels were not different. Addition of a pan-specific TGFβ1/2/3 neutralizing antibody fully inhibited SMAD3-reporter activity of both acidified and not-acidified control and SSc sera. Both HC and SSc sera induced similar SMAD1/5-reporter activity, and acidification increased this, but not differently between groups. Interestingly, expression of two integrin alpha subunits ITGA5 and ITGAV was significantly reduced in monocytes obtained from SSc patients. Furthermore, ITGB3, ITGB5, and ITGB8 expression was also reduced in SSc monocytes. Stimulation of monocytes with TGFβ1 induced ITGA5 and ITGAV but lowered ITGB8 expression, whereas the use of the TGFβ receptor inhibitor SB-505124 had the opposite effect.

Conclusion

Total TGFβ serum levels are not different between SSc patients and controls, but TGFβ activity is. This coincides with a reduced expression of TGFβ-activating integrins in monocytes of SSc patients. Because TGFβ regulates expression of these integrins in monocytes, a negative feedback mechanism possibly underlies these observations.

Crucial role of synovial lining macrophages in the promotion of transforming growth factor beta-mediated osteophyte formation

OBJECTIVE

To investigate in vivo and in vitro whether macrophages have an intermediate role in transforming growth factor beta (TGFbeta)-induced osteophyte formation.

METHODS

In vivo, synovial lining macrophages were selectively depleted by injection of clodronate-laden liposomes 7 days prior to injection of 20 ng or 200 ng of TGFbeta into murine knee joints 3 times, on alternate days. Total knee joint sections were obtained on day 7 after the last injection and stained with Safranin O. Production of bone morphogenetic protein 2 (BMP-2) and BMP-4 was determined by immunolocalization. The interaction between murine macrophages and mesenchymal cells (precursors with chondrogenic potential) was studied in vitro using a Transwell system in which RAW macrophages were cocultured with C3H10T1/2 mesenchymal cells. Spheroid neocartilage formation was quantified microscopically after staining with May-Grünwald-Giemsa.

RESULTS

Triple injections of 20 ng or 200 ng of TGFbeta into normal murine knee joints induced significant osteophyte formation at the lateral and medial sites of the patella and femur on day 7 after the last injection. Strikingly, removal of synovial lining macrophages prior to TGFbeta injection resulted in a drastic reduction of osteophyte formation (by 70% and 64% after injection of 20 ng and 200 ng of TGFbeta, respectively). Synovial lining cells produced BMP-2 and BMP-4 after TGFbeta stimulation, whereas BMP-2 and BMP-4 were absent in the synovial tissue after macrophage depletion. In vitro, clustering and spheroid formation of C3H10T1/2 was induced by TGFbeta concentrations of >1 ng/ml. However, in the Transwell system, in the presence of murine macrophages, 0.5 ng/ml of TGFbeta was very effective in generating large spheroids, suggestive of macrophage-derived (co)factors. In coculture supernatants, TGFbeta concentrations were not elevated in the presence of macrophages, indicating generation of other growth factors involved in spheroid formation.

CONCLUSION

These findings indicate that macrophages are crucial intermediate factors in osteophyte formation induced by TGFbeta, probably by inducing other chondrogenic signals.

Adenoviral overexpression of Smad-7 and Smad-6 differentially regulates TGF-beta-mediated chondrocyte proliferation and proteoglycan synthesis

OBJECTIVE

To assess if various biological responses to transforming growth factor-beta (TGF-beta) in chondrocytes are differentially regulated by Smad-6 and Smad-7.

DESIGN

Adenoviral overexpression of Smad-6 or -7 mRNA in a chondrocyte cell line was determined via semi-quantitative RT-PCR and protein overexpression was studied by immunocytochemistry. Furthermore, the effect of Smad-6 and -7 overexpression on TGF-beta-induced PAI-1 and aggrecan mRNA upregulation was studied via quantitative RT-PCR. The effect of Smad-6 and -7 overexpression on TGF-beta-induced chondrocyte proliferation was studied via DNA quantification, whereas TGF-beta-induced proteoglycan (PG) synthesis was studied by 35S-sulfate incorporation.

RESULTS

Adenoviral transfection of chondrocytes with Smad-6 and -7 resulted in strong upregulation of Smad-6 and -7 mRNA levels, respectively. Immunocytochemistry showed overexpression of Smad-6 and -7 proteins in both the nucleus and cytoplasm. Smad-6 overexpression significantly inhibited TGF-beta-stimulated chondrocyte proliferation, although proliferation was not completely abolished. Smad-7 overexpression, however, completely antagonized the TGF-beta effect on proliferation. Smad-6 overexpression had no effect on TGF-beta-induced PAI-1 expression, while overexpression of Smad-7 completely blocked this TGF-beta effect. Additionally, overexpression of Smad-7, but not Smad-6, totally prevented TGF-beta-induced PG synthesis on the mRNA and protein levels.

CONCLUSIONS

Adenoviral transfection of chondrocytes with Smad-6 and -7 resulted in strong upregulation of Smad-6 and -7 mRNA and protein levels. Furthermore, overexpression of Smad-7 in chondrocytes totally inhibited important TGF-beta-mediated biological responses such as proliferation and PG synthesis, while overexpressed Smad-6 had no or only a partial inhibitory effect on TGF-beta activity. We conclude that in chondrocytes distinct TGF-beta activities are differentially regulated by Smad-6 and Smad-7.

Differential effect of transforming growth factor beta on freshly isolated and cultured articular chondrocytes

Transforming growth factor beta (TGF beta) is a multipotent regulator of cell proliferation and extracellular matrix synthesis. In our study we show that the effect of TGF beta on proliferation and proteoglycan synthesis of articular cartilage chondrocytes appears to be dependent on the period in culture of these chondrocytes. TGF beta inhibits DNA and proteoglycan synthesis of freshly isolated chondrocytes while the DNA and proteoglycan synthesis of chondrocytes cultured in monolayer was stimulated by TGF beta. Since TGF beta is present in high concentrations in synovial fluid of patients with osteoarthritis, TGF beta might play a role in the elevated proteoglycan synthesis and cell proliferation in osteoarthritic cartilage.

Effects of NSAIDs on the metabolism of sulphated glycosaminoglycans in healthy and (post) arthritic murine articular cartilage

Several non-steroidal anti-inflammatory drugs (NSAIDs) were studied for their effects on normal and damaged murine articular cartilage, both in vitro and in vivo. In vitro, in the absence of serum, sodium salicylate caused significant suppression of 35S-glycosaminoglycan (GAG) synthesis, whereas tiaprofenic acid, piroxicam, prednisolone sodium phosphate and several other NSAIDs were without effect. Trypsin-mediated proteoglycan depletion did not change the susceptibility of the articular chondrocyte to these drugs. Similarly, no enhancement of drug effect was seen when arthritic cartilage was taken from an acutely inflamed joint, and prenisolone sodium phosphate even seemed to diminish inflammation-mediated suppression of 35S-GAG synthesis. The short term in vivo effects of some of the drugs were tested in mice with unilateral zymosan-induced arthritis. At day 1 after arthritis induction, in vivo 35S-GAG synthesis by the cartilage of the arthritic joint was decreased to 63%. Only sodium salicylate suppressed in vivo 35S-GAG synthesis in the healthy and arthritic joint to the same extent in both. At day 28, GAG synthesis in the postarthritic joint was enhanced to 160%. This type of cartilage appeared to be more susceptible to drug effects, since all drugs tested showed clear suppression of the augmented GAG production in vivo. Finally, in vivo drug effects were tested on normal and enhanced 35S-GAG degradation, the latter in the zymosan-induced arthritic joint. Both tiaprofenic acid and prednisolone sodium phosphate appeared to suppress degradation in healthy and, to some extent, in arthritic cartilage.(ABSTRACT TRUNCATED AT 250 WORDS)

Effect of nonsteroidal antiinflammatory drugs on cartilage destruction in antigen induced arthritis in mice

The nonsteroidal antiinflammatory drugs, salicylate, piroxicam and tiaprofenic acid, and the steroid prednisolone were investigated in a long-term study for their potential detrimental or beneficial effects on joint cartilage in mice with antigen induced monoarthritis. Daily drug treatment over a period of 4-7.5 weeks did not affect the histological characteristics of normal joints at all. Articular chondrocyte synthetic activity was even stimulated after salicylate and tiaprofenic acid treatment, but the significance of this finding is not yet clear. Cartilage damage, caused by inflammation in the knee joint, was neither markedly deteriorated nor attenuated by these drugs. Minor antiinflammatory properties as measured by decrease in edema using 99mTc-uptake and in the change of inflammatory cells were only evident with prednisolone, piroxicam and salicylate.

Determination of small quantities of sulfate (0-12 nmol) in serum, urine, and cartilage of the mouse

The colorimetric benzidine method of K. S. Dodgson and B. Spencer (1953, Biochem. J. 55, 436-440) for the measurement of inorganic sulfate can be scaled down about 100 times by using disposable 96-well microplates instead of individual cuvettes. Ten-microliter samples of serum and urine, derived from mice, can be analyzed in a simple, rapid, and reliable way without sacrificing the animals. Without prior isolation of sulfated glycosaminoglycans, ester sulfate in mouse patellar cartilage is liberated quantitatively as inorganic sulfate upon acid hydrolysis in 3 M HCl for 16 h at 80 degrees C. To this end the articular cartilage layer of the patella must be separated in toto from the underlying bone. Subsequent hydrolysis in polypropylene tubes gives accurate results. In contrast, hydrolysis in borosilicate glass vials is useless, since nanomoles of sulfate added cannot be recovered adequately. The thin patellar cartilage layer obtained from 10-week-old male mice contains about 5 nmol of sulfate, an amount easily measured with the developed microplate benzidine method.

Quantitation of glycosaminoglycan metabolism in anatomically intact articular cartilage of the mouse patella: in vitro and in vivo studies with 35S-sulfate, 3H-glucosamine, and 3H-acetate

We investigated the usefulness of the whole mouse patella to quantitate the synthesis of the glycosaminoglycan (GAG) backbone and its sulfation by intact murine articular cartilage, both in vitro and in vivo. Using 35S-sulfate, 3H-glucosamine, or 3H-acetate as precursors of GAG synthesis, it was found that more than 90% of the incorporated radioactivity was confined to the patellar cartilage layer compared to the whole patella. Overnight papain digestion was enough to liberate more than 95% of the incorporated radiolabels, except for 3H-acetate for which 15-25% was not digestible. Comparison of radioactivity in the patella and that in quantitatively isolated GAGs revealed that for 35S-sulfate incorporation studies the whole patella can be used as a reliable measure for sulfated GAG synthesis. The situation was different for the GAG backbone precursors 3H-glucosamine and 3H-acetate; more than 50% of the 3H labels were incorporated into compounds other than GAGs or non-covalently associated with matrix components. Hence, in studying GAG-backbone metabolism, polysaccharides must be isolated quantitatively from cartilage. In vivo studies made it clear that both 35S-sulfate and 3H-glucosamine are incorporated into patellar GAGs in amounts high enough to enable proper quantitation and that the route of administration (intraperitoneally or intravenously) is of minor importance. Due to its low specificity for cartilage GAGs, 3H-acetate is not suitable for such studies.

The effect of salicylate on anatomically intact articular cartilage is influenced by sulfate and serum in the culture medium

The established, suppressive effect of salicylate on sulfated glycosaminoglycan (sGAG) synthesis by normal articular cartilage was reinvestigated using anatomically intact articular cartilage of the whole mouse patella. Employing the physiological murine sulfate concentration (1.0 mM) sodium salicylate (1-5 X 10(-3) M) caused a dose dependent inhibition of 35S-sGAG synthesis (10-35%). At a lower sulfate concentration (0.4 mM) this inhibition was increased (15-45%) and the suppression was even more pronounced in sulfate deprived medium. This observation stresses the need of using physiological sulfate concentrations in cartilage culture studies. In the presence of 100% serum the therapeutic drug concentration (1-2 X 10(-3) M) had no longer any suppressive effect, either at 1.0 mM or at any lower sulfate concentration. Our data suggest that salicylate has no direct effect on sGAG metabolism in normal articular cartilage in vivo and that adverse effects may be due to the observed salicylate induced lowering of the endogenous sulfate level.