Expression and production of the long pentraxin PTX3 in rheumatoid arthritis (RA)

Abstract

PTX3 is a secreted molecule which consists of a C-terminal domain similar to classical pentraxins (e.g. C-reactive protein (CRP)) and of an unrelated N-terminal domain. Unlike the classical pentraxins, the long pentraxin PTX3 is expressed in response to IL-1beta and tumour necrosis factor-alpha (TNF-alpha), but not to IL-6, in various cell types. The present study was designed to investigate the expression of PTX3 in RA. Dissociated RA and osteoarthritis (OA) type B synoviocytes were cultured in the presence and in the absence of inflammatory cytokines. PTX3 mRNA expression in synoviocytes was evaluated by Northern analysis. PTX3 protein levels in synovial cell cultures and synovial fluid were estimated by ELISA, and PTX3 distribution in synovial tissues by immunohistochemical techniques. OA synoviocytes were induced to express high levels of PTX3 mRNA by TNF-alpha, but not by other cytokines including IL-1beta and IL-6. RA synoviocytes, unlike OA synoviocytes, constitutively expressed high levels of PTX3 in the absence of deliberate stimulation. The constitutive expression of PTX3 in RA synoviocytes was not modified by anti-TNF-alpha antibodies, IL-1 receptor antagonist or a combination of the two agents. In contrast, interferon-gamma and transforming growth factor-beta inhibited PTX3 constitutive expression in RA synoviocytes. The joint fluid from RA patients contained higher levels of immunoreactive PTX3 than controls and the synovial tissue contained endothelial cells and synoviocytes positive for PTX3 by immunohistochemistry. In conclusion, PTX3 may play a role in inflammatory circuits of RA, and its relevance as a marker of disease activity deserves further study.

Fig. 1

Northern blot analysis showing PTX3 expression in type B synoviocytes from six patients with RA and six patients with osteoarthritis (OA). Total RNA was prepared and analysed by Northern blot analysis as described in Patients and Methods. Expression of β-actin mRNA is shown for normalization of the quantities of RNA loaded in gel lanes. (a,b) Distinct Northern analysis with different patients and at different times.

Fig. 3

Northern blots showing the effects of different cytokines on PTX3 expression in RA synoviocytes (a), and in osteoarthritis (OA) synoviocytes (b). Cytokine concentrations and times of stimulation are detailed in Patients and Methods. Expression of GAPDH mRNA is shown for normalization of the quantities of RNA loaded in gel lanes.

Fig. 2

Northern blots showing the effects of different cytokines on PTX3 expression by osteoarthritis (OA) synoviocytes (a) and normal skin fibroblasts (b). (c) Induction of IL-8 mRNA in OA synoviocytes by the preparation of the IL-1β used in all the PTX3 experiments. Cytokine concentrations and times of stimulation are detailed in Patients and Methods. Expression of GAPDH mRNA is shown for normalization of the quantities of RNA loaded in gel lanes.

Fig. 4

PTX3 protein levels in supernatants from RA (a) and osteoarthritis (OA) (b) synoviocyte cultures after stimulation with different cytokines and growth factors. For details see Patients and Methods. PTX3 protein level was assessed by ELISA.

Fig. 5

PTX3 protein levels in synovial fluids (11 RA patients and nine controls). PTX3 level was determined by ELISA. Data are reported as mean ± s.d. The horizontal line represents the upper limit of the normal.

Fig. 6

(a) Synovial tissue from a patient with RA. A strong immunostaining for PTX3 is detected in the cytoplasm of hyperplastic synovial lining cells. Stromal and endothelial cells are also labelled. (Immunoperoxidase–haematoxylin, × 250.) (b) Tissue from a patient with osteoarthritis. Synovial stromal and endothelial cells (arrowheads) show a faint staining for PTX3 (Immunoperoxidase–haematoxylin, × 250).