Share this article on:

Nuclear factor-κB and basic fibroblast growth factor augment pannus formation in rheumatoid synovitis

Ibrahim, Soha E.a; Salman, Manal I.b; Ibrahim, Eman A.b; Kamel, Nadia S.a; Zamzam, Mona L.a; Elshamy, Hebatallah A.a

Egyptian Journal of Pathology: July 2013 - Volume 33 - Issue 1 - p 1–7
doi: 10.1097/01.XEJ.0000429912.50802.db
Original Articles

Background: Rheumatoid arthritis (RA) is a multisystem autoimmune disease that causes marked deformity in the small joints.

Aim of the work: The aim of this work is to determine the role of nuclear factor-κB (NF-κB) and basic fibroblast growth factor (bFGF) in the histological changes of RA synovial specimens.

Patients and methods: This study was carried out on 25 RA patients who were subjected to surgery and 10 control participants with post-traumatic knee joint injury. The paraffin blocks were sectioned and the slides were stained conventionally with hematoxylin and eosin (H&E) and immunohistochemical markers for NF-κB and bFGF. Interpretation, scoring, and statistical correlation of the histological findings were performed.

Results: Histological examination and scoring of the specimens from both groups indicated a highly significant statistical increase in the number of synovial cell layers, vascular density (5.2±2.5 vs. 1.1±1.1 and 5.7±1.1 vs. 1.5±0.7 respectively; P<0.01), fibrosis (5.9±3.3 vs. 0.5±0.4), perivascular lymphocytic infiltrate (2±1.5 vs. 0), lymphoid aggregates (3.9±0.4 vs. 0), free lymphocytes (2.8±1.3 vs. 0.45±0.4), and immunohistochemical staining for NF-κB and bFGF (P<0.001) in RA patients in comparison with the control group. The number of synovial cell layers and lymphocytic aggregates showed a positive correlation with NF-κB and bFGF staining. Similarly, perivascular lymphocytic infiltrate correlated positively with bFGF staining (P<0.001).

Conclusion: NF-κB and bFGF played a crucial role in augmenting and maintaining pannus formation in the synovium of RA patients.

Departments of aPhysical Medicine and Rehabilitation

bPathology, Faculty of Medicine, Ain Shams University, Cairo, Egypt

Correspondence to Eman A. Ibrahim, Department of Pathology, Faculty of Medicine, Ain Shams University, Alabbassiya Square, PO Box 11841, Cairo, Egypt Tel: +0020226825344; fax: +0020226837722; e-mail:

Received November 20, 2012

Accepted December 5, 2012

Back to Top | Article Outline


Rheumatoid arthritis (RA) is a systemic chronic inflammatory disease that mainly causes erosive synovitis. Synovial neovascularization, synovial intimal lining hyperplasia, inflammatory cell infiltration, and pannus formation are the hallmarks of the disease (Mor et al., 2005).

Fibroblast-like synoviocytes play a crucial role in the pathogenesis of pannus formation. They show some characteristics of tumor-like cells, the actual mechanism of which has not been clarified completely. They proliferate actively in response to various types of cytokines secreted partly by fibroblast-like synoviocytes themselves, leading to a kind of autonomy, which leads to the characteristics of tumor-like growth (Guo et al., 2012)

Nuclear factor-κB (NF-κB) is a family of transcription factors central to immunity and inflammation. NF-κB is a heterodimeric DNA-binding protein and a critical element in the regulation of inflammatory cytokine production by inducing transactivation of genes including tumor necrosis factor-α, interleukin (IL)-1, IL-6, IL-8, granulocyte-macrophage colony-stimulating factor (GM-CSF), and intercellular adhesion molecule-1 (ICAM-1). These cytokines can activate the transcription factor of NF-κB, which induces inhibitors of apoptosis and promotes the production of proinflammatory factors (Brown et al., 2008).

Fibroblast growth factor-2 (FGF-2, bFGF) is a member of the family of heparin-binding growth factors that contain 10 members with potent mitogenic and angiogenic effects on a variety of cells of mesodermal and ectodermal origin and is released extracellularly. Thus, the proliferation of synovial cells in the inflamed joints may be the result of stimulation by basic fibroblast growth factor (bFGF) in an autocrine manner. FGF-2 decreases the sensitivity of RA synoviocytes to Fas-mediated apoptosis, leading to intractable synovial hyperplasia (Malemud, 2007).

The aim of this work is to determine and show the role of NF-κB and bFGF in synovial hyperplasia in RA patients to focus on possible new therapeutic strategies in the management of RA.

Back to Top | Article Outline

Patients and methods

This study was carried out on 25 patients with RA who were subjected to knee joint replacement, arthroscopic knee joint surgery, or diagnostic arthroscopy, and 10 control participants with post-traumatic knee injury who were scheduled for surgery (Table 1).

Back to Top | Article Outline

Control group

The control group included individuals matching in age and sex to the patient group and included individuals with traumatic knee injury. These control participants did not have any rheumatic diseases.

Back to Top | Article Outline

Patient group

Patients with RA were recruited from the Rheumatology and Rehabilitation Department of Ain Shams University and were diagnosed according to the American college of Rheumatology (ACR) criteria (Arnett et al., 1988). All RA patients were scheduled for knee joint replacement, arthroscopic knee joint surgery, or diagnostic arthroscopy. Both the patient and the control group were subjected to full assessment of medical history, thorough clinical examination that included general and articular examination, and full laboratory investigations (Table 2).

Synovial biopsies were obtained through knee joint replacement, arthroscopic knee joint surgery, or diagnostic arthroscopy by specialized biopsy forceps.

Back to Top | Article Outline

Histopathological processing and evaluation of synovial tissue biopsies

Paraffin-embedded tissue sections of 4–5 μm were prepared from 10% formalin-fixed synovial biopsies. The sections were subjected to routine H&E staining.

Back to Top | Article Outline

Interpretation of the histopathological results

Evaluation of histologic features by H&E-stained sections was carried out according to the scoring system proposed by Fonseca et al. (2000), who described six histologic features scored on a scale of 0–10:

The mean number of synoviocytes lining layers was counted, with a possible number of layers ranging from 1 (score 0) to more than 10 (score 10).

The mean percentage of fibrosis beneath the synovial layer was measured: less than 10% (score 0), 10–15% (score 1), 15–20% (score 2), 20–25% (score 3), 25–30% (score 4), 30–40% (score 5), 40–50% (score 6), 50–60% (score 7), 60–70% (score 8), 70–80% (score 9), greater than 80% (score 10).

The mean density of blood vessels per high-power field was evaluated (under ×400 magnification): less than 3 (score 0), 4–5 (score 1), 6–7 (score 2), 8–9 (score 3),10–11 (score 4), 12–13 (score 5), 14–15 (score 6), 16–17 (score 7), 18–19 (score 8), 20–22 (score 9), more than 22 (score 10).

The mean percentage of vessels with perivascular infiltrates of lymphocytes was estimated. Perivascular infiltrates were characterized as aggregates of lymphocytes that were contiguous with the vessel wall and were no more than 10 cells in number: less than 5% (score 0), 5–10% (score 1), 10–20% (score 2), 20–30% (score 3), 30–40% (score 4), 40–50% (score 5), 50–60% (score 6), 60–70% (score 7), 70–80% (score 8), 80–90% (score 9), 100% (score 10).

The mean number of lymphocytes in the diameter of the focal aggregates of lymphocytes was counted: less than 11 (score 0), 11–15 (score 1), 15–20 (score 2), 20–25 (score 3), 25–30 (score 4), 30–35 (score 5), 35–40 (score 6), 40–45 (score 7), 45–50 (score 8), 50–55 (score 9), more than 55 (score 10).

The mean percentage of (free lymphocytes) per high-power field was assessed by counting only the lymphocytes that did not fall in either one of the above categories of perivascular or focal aggregates: 0 (score 0), 1–10% (score 1), 10–20% (score 2), 20–30% (score 3), 30–40% (score 4), 40–50% (score 5), 50–60% (score 6), 60–70% (score 7), 70–80% (score 8), 80–90% (score 9), 90–100% (score 10).

Back to Top | Article Outline

Immunohistochemical staining and evaluation of the synovial tissue biopsies

Sections of 4 μm were cut and mounted on charged slides for the standard immunoperoxidase staining technique according to Hsu and Raine (1981) using the primary NFkB and bFGF rabbit polyclonal antibodies (Santa Cruz Biotechnology Inc., California, USA) in 1 : 150 dilution overnight at 4°C, and then the slides were rinsed with PBS three times, 5 min each, and were further incubated with secondary antibody biotinylated anti-mouse immunoglobulin G (Zymed Lab, San Francisco, California, USA); the slides were again rinsed with PBS and further incubated for one hour with avidin biotin–peroxidase complex (LSAB kit; Dako) at room temperature and again rinsed with PBS. The peroxidase-binding sites were detected by staining with 3,3′-diaminobenzidine (DAB) in 0.05 mol/l Tris-EDTA buffer, pH 7.6. Finally, counterstaining was performed using Mayer’s hematoxylin.

Back to Top | Article Outline

Interpretation of the histopathological results

NF-κB will be detected as a brown nuclear staining of the synoviocytes, sublining, and endothelial cells. The percentage of positive-stained cells was assessed and scored according to Handel et al. (1995): (−)=<5%, (+)=5–25%, (++)=25–50%, (+++)=50–75%, and (+++ +)=75–100%. bFGF was assessed according to the intensity of brown staining in both the cytoplasm and the nucleus of synoviocytes, fibroblasts, and endothelial cells. The intensity of stain was evaluated and divided into four grades according to Nakashima et al. (1994): no staining (−), mild staining (+), moderate staining (++), and strong staining (+++).

Back to Top | Article Outline

Statistical analysis

Analysis of data was carried out by an IBM computer using statistical program for social science (SPSS Inc.) as follows:

Description of quantitative variables as mean, SD, and range.

Description of qualitative variables as number and %.

The χ2-test was used to compare qualitative variables.

Unpaired t-test was used to compare two groups in terms of the qualitative and quantitative variables.

Correlation analysis was carried out for the assessment of the strength of association between two variables. The correlation coefficient, denoted symbolically as r, defines the strength and direction of the linear relationship between two variables.

Differences were considered statistically significant at P value less than 0.05 and highly significant at P value less than 0.01.

Back to Top | Article Outline


There were 25 RA patients whose age ranged from 32 to 58 years, mean 43.5±6.8 years. The disease duration ranged from 2 to 16 years, mean 8.9±4.4 years. The control group included 10 individuals ranging in age from 30 to 45 years, mean 30.7±1.1 years. Female predominance was remarkable in both groups. The correlation between the two groups was not significant (Table 1).

There was a significant statistical correlation between both groups in ESR and anemia, with the microcytic hypovolemic type being more commonly encountered in RA patients (Table 2).

There was a highly significant correlation between both groups in the number of synovial cells, fibrosis, vascular density, perivascular lymphocytic infiltrate, lymphocytic aggregates, and free lymphocytes. They were highly increased in RA patients compared with the individuals in the control group (Table 3 and Fig. 1).

There was a significant correlation between both groups in NF-κB immunohistochemical staining of synoviocytes, sublining, and endothelial cells, in which patients with RA had a much higher staining score (Table 4 and Fig. 2).

There was a significant correlation between both groups in bFGF immunohistochemical staining of synoviocytes, fibroblasts, and endothelial cells, in which patients with RA had a much higher staining score (Table 5 and Fig. 3).

There was a significant positive statistical correlation between the panel of (NF-κB and bFGF) staining and number of layers of synovial cells and lymphocytic aggregates. Furthermore, bFGF correlated significantly with perivascular lymphocytic infiltrate (Table 6).

Back to Top | Article Outline


Findings and interpretation

In this work, we observed middle aged female predominance among RA patients. There was a significant statistical correlation between the patient and the control group in elevated ESR and the presence of microcytic hypochromic anemia. Histological examination showed pannus formation together with fibrosis, increased vascular density, and dense lymphocytic infiltration in the patient group. Immunohistochemical staining panel for NF-κB and bFGF showed a strong statistically positive correlation with the RA group. These findings might focus our interest to these strong associations as NF-κB and bFGF might play a very active role in the initiation, augmentation, and maintenance of pannus formation.

The histological findings of fibrosis and pannus formation could be attributed to the synovial fibroblast-like synoviocytes (FLS). These FLS are known as pannocytes. They are prominent at the interface between the synovium and the cartilage, leading to invasion and destruction of the adjacent cartilage, even in the absence of inflammatory infiltrate (Longato et al., 2005).

The abnormal behavior of FLS in inflammatory arthritis might be the result of spontaneous mutations leading to dysregulation of FLS function as NF-κB, which increases the aggressiveness of proliferating FLS. Another driving force for autonomously activated and the transformed nature of FLS results from continuous stimulation by inflammatory mediators and growth factors, such as bFGF, leading to expansive and tumor-like synovial pannus that invades cartilage (Guo et al., 2012).

This pannus requires additional nutrients and oxygen. In the face of this demand, there is likely a shift in the balance between angiogenic and antiangiogenic factors, with a predominance of the angiogenic mediators, such as bFGF, leading to neovascularization (Rudolph and Woods, 2005).

Although the RA specimens showed increased vascular density in comparison with the control group (P<0.001), they lacked a correlation with bFGF staining; this might be because of excessive specimen fibrosis and nonusage of specific vascular endothelial immunohistochemical markers in our research.

Back to Top | Article Outline

Differences in results and conclusion in comparison with other studies

In terms of histopathological findings, our results showed that there was a statistically highly significant increase in the mean number of synovial cells and vascular density in RA patients (5.2±2.5 vs. 1.1±1.1 and 5.7±1.1 vs. 1.5±0.7, respectively; P<0.001).

This was in agreement with Roccaro et al. (2005), who added that neovascularization in RA may be caused by hypoxia of the hyperplastic synovium and the cytokine milieu, which stimulate the release of angiogenic factors such as bFGF, VEGF, angiopontin, and fractalkine.

In our study, we found a statistically highly significant increase in the mean number of free lymphocytes between RA patients (2.8±1.3) and control individuals (0.45±0.4) as P value less than 0.001.

Our results were in agreement with those of Fonseca et al. (2000) who found that the presence of lymphocytes was statistically significant in RA synovial samples. Other authors have reported that T lymphocytes play an important role in the pathogenesis of RA (Remans et al., 2005).

In our study, we detected a highly significant immunohistochemical expression of nuclear NF-κB in the synovial biopsy of RA patients compared with the control group in synovial lining, sublining, and endothelial cells (84 vs. 10%, 60 vs. 0%, and 76 vs. 0%,with P<0.001, P<0.05, and P<0.001), respectively.

Our results were in agreement with those of Handel et al. (1995), who identified p50 and p65 subunits of NF-κB in the nuclei of cells within 100% of the lining layer, 69.2% of sublining region, and 84.6% of endothelial cells of RA synovium.

NF-κB is activated and increased in the tissue as a result of local hypoxia, oxygen-free radicals, and many inflammatory cytokines. Many trials are ongoing for the treatment of many diseases, such as RA, by inhibiting NF-κB (Kannaiyan et al., 2011a, 2011b; Li et al., 2012).

Moreover, Kubota et al. (2007) reported the importance of inhibiting NF-κB to relieve the inflammation and bone destruction in an animal model of arthritis as it acted as a transcription factor implicated in diverse receptor-mediated signaling pathways including the differentiation and activation of osteoclasts.

In this study, we found a highly significant correlation between the panel expression of NF-κB in the lining synovium, sublining cells, endothelial cells with the mean number of synovial cell layers, and lymphocytic aggregates as P value less than 0.01.

This was in agreement with Benito et al. (2004) who reported that the NF-κB signaling pathway was a driving factor for the hyperplastic changes in the synovium; also, it acted as a key mediator of inflammation, which is of critical value in the survival of synovial fibroblasts in RA joints.

In our study, there were topographic differences in the distribution of bFGF staining; it was detected in the nuclei and cytoplasm of the synoviocytes of RA synovial specimens, whereas it showed only cytoplasmic stain in the control group. Moreover, bFGF was positively stained in fibroblasts, and endothelial cells in RA patients, and totally negative in the control group.

These results were in agreement with those of Nakashima et al. (1994), who detected bFGF immunohistochemically in the cytoplasm and/or nucleus of the synovial lining cells of all RA synovial specimens, 86% of fibroblasts, and 73.3% of endothelial cells, and they found no expression of bFGF in the synovial tissue sections of five normal control individuals with joint trauma.

Yamashita et al. (2002) and Nakano et al. (2004) found that bFGF not only augments the proliferation of rheumatoid synovial fibroblasts but is also involved in osteoclast maturation, which leads to bone destruction in RA.

In our study, we found that there was a statistically highly significant positive correlation between staining of bFGF in synoviocytes, fibroblasts, and endothelial cells with the mean number of synovial cell layers, perivascular lymphocytic infiltration, and lymphoid aggregates.

These findings were in agreement with Roccaro et al. (2005), who found in their study that although perivascular mononuclear cell infiltration and thickness of synovium were increased in both inflamed and noninflamed joints, vascular proliferation occurred only in tissues from inflamed joints. The endothelial cells of these proliferating vessels were shown to express cell-cycle-associated antigens, such as PCNA and Ki-67, which are usually associated with vascular proliferation.

Back to Top | Article Outline

Relevance of the findings: implications for clinicians

NF-κB and bFGF inhibitors may be a potentially important therapeutic approach for RA by correcting the imbalance between apoptosis and proliferation of synovial cells, on the one hand, and angiogenesis and antiangiogenesis of the synovial tissue, on the other, in RA patients.

Back to Top | Article Outline


From this study, we can conclude that NF-κB and bFGF play a crucial role in the pathogenesis of rheumatoid pannus by inducing synovial hyperplasia and angiogenesis. Inhibition of NF-κB and bFGF may represent a new tool for future therapeutic strategies.

Back to Top | Article Outline


Conflicts of interest

There are no conflicts of interest.

Back to Top | Article Outline


Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al..The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis.Arthritis Rheum1988;31:315–324.
Benito MJ, Murphy E, Murphy EP, Van Den Berg WB, FitzGerald O, Bresnihan B.Increased synovial tissue NF-κB1 expression at sites adjacent to the cartilage-pannus junction in rheumatoid arthritis.Arthritis Rheum2004;50:1781–1787.
Brown KD, Claudio E, Siebenlist U.The roles of the classical and alternative nuclear factor-κB pathways: Potential implications for autoimmunity and rheumatoid arthritis.Arthritis Res Ther2008;10:212.
Fonseca JE, Canhao H, Resende C, Saraiva F, Teixeira da Costa JC, Bravo Pimentao J, et al..Histology of the synovial tissue: value of semiquantitative analysis for the prediction of joint erosions in rheumatoid arthritis.Clin Exp Rheumatol2000;18:559–564.
Guo X, Pan Y, Xiao C, Wu Y, Cai D, Gu J.Fractalkine stimulates cell growth and increases its expression via NF-κB pathway in RA-FLS.Int J Rheum Dis2012;15:322–329.
Handel ML, McMorrow LB, Gravallese EM.Nuclear factor-kB in rheumatoid synovium. Localization of P50 and P65.Arthritis Rheum1995;38:1762–1770.
Hsu SM, Raine L.Protein A, avidin, and biotin in immunohistochemistry.J Histochem Cytochem1981;29:1349–1353.
Kannaiyan R, Manu KA, Chen L, Li F, Rajendran P, Subramaniam A, et al..Celastrol inhibits tumor cell proliferation and promotes apoptosis through the activation of c-Jun N-terminal kinase and suppression of PI3 K/Akt signaling pathways.Apoptosis2011a;16:1028–1041.
Kannaiyan R, Shanmugam MK, Sethi G.Molecular targets of celastrol derived from Thunder of God Vine: potential role in the treatment of inflammatory disorders and cancer.Cancer Lett2011b;303:9–20.
Kubota T, Hoshino M, Aoki K, Ohya K, Komano Y, Nanki T, et al..NF-kappaB inhibitor dehydroxymethylepoxyquinomicin suppresses osteoclastogenesis and expression of NFATc1 in mouse arthritis without affecting expression of RANKL, osteoprotegerin or macrophage colony-stimulating factor.Arthritis Res Ther2007;9:R97.
Li Y, He D, Zhang X, Liu Z, Zhang X, Dong L, et al..Protective effect of celastrol in rat cerebral ischemia model: Down-regulating p-JNK, p-c-Jun and NF-κB.Brain Res2012;1464:8–13.
Longato L, Anania A, Calosso L, Corino D, Tarocco RP.Histogenic characterization of the cells forming RA pannus.Recenti Prog Med2005;96:16–22.
Malemud CJ.Growth hormone, VEGF and FGF: involvement in rheumatoid arthritis.Clin Chim Acta2007;3751–210–19.
Mor A, Abramson SB, Pillinger MH.The fibroblast-like synovial cell in rheumatoid arthritis: a key player in inflammation and joint destruction.Clin Immunol2005;115:118–128.
Nakano K, Okada Y, Saito K, Tanaka Y.Induction of RANKL expression and osteoclast maturation by the binding of fibroblast growth factor 2 to heparan sulfate proteoglycan on rheumatoid synovial fibroblasts.Arthritis Rheum2004;50:2450–2458.
Nakashima M, Eguchi K, Aoyagi T, Yamashita I, Ida H, Sakai M, et al..Expression of basic fibroblast growth factor in synovial tissues from patients with rheumatoid arthritis: detection by immunohistological staining and in situ hybridization.Ann Rheum Dis1994;53:45–50.
Remans PHJ, Van Oosterhout M, Smeets TJM, Sanders M, Frederiks WM, Reedquist KA, et al..Intracellular free radical production in synovial T lymphocytes from patients with rheumatoid arthritis.Arthritis Rheum2005;52:2003–2009.
Roccaro AM, Russo F, Cirulli T, Di Pietro G, Vacca A, Dammacco F.Antiangiogenesis for rheumatoid arthritis.Curr Drug Targets Inflamm Allergy2005;4:27–30.
Rudolph EH, Woods JM.Chemokine expression and regulation of angiogenesis in rheumatoid arthritis.Curr Pharm Des2005;11:613–631.
Yamashita A, Yonemitsu Y, Okano S, Nakagawa K, Nakashima Y, Irisa T, et al..Fibroblast growth factor-2 determines severity of joint disease in adjuvant-induced arthritis in rats.J Immunol2002;168:450–457.
©2013Egyptian Journal of Pathology