Dendritic cells and scavenger macrophages in chronic inflammatory bowel disease.

We used enzyme (acid phosphatase [AP]) and immunohistochemical techniques and a set of monoclonal antibodies (CD11, CD5, CD4, CD19, CD8, OKIa), including two recently developed antibodies--for example, HECA-452 (specific for an adhesion molecule on high endothelial venules) and RFD1 (specific for 'active' human dendritic cells) to analyse the composition of the gut wall infiltrate of 10 well defined cases of chronic inflammatory bowel disease (CIBD) (six Crohn's disease (CD), four ulcerative colitis (UC]. Two polar forms in a spectrum of gut mononuclear phagocyte types (CD11+) were identified: at the one extreme scavenger macrophages with blunted projections (AP+, Heca-452-, RFD1-) and at the other extreme, dendritic cells with long dendritic cytoplasmic projections (AP-, Heca-452+, RFD1+). Dendritic cells were mainly found in highly organised lymphoid tissue present at the deeper layers in the gut wall (normal gut: underneath the muscularis mucosae and T-cell areas of lymph follicles [25-30 per follicle]; surrounding the broad zone of scavenger macrophages at the bottom of ulcers (CIBD) and fissures (CD) and in the lymphoid aggregates [25-30 dendritic cells per aggregate] adjacent to granulomas (CD]. These observations can be taken as evidence that exaggerated antigen handling and presentation and stimulation of the immune response takes place at these foci. The observation that scavenger macrophages were localised more superficial, as band like zones (normal gut: subepithelial; mainly surrounding ulcers (CIBD) and fissures (CD] can be taken as evidence that at these spots the ingestion and degradation of foreign material takes place.     

Spirochetal antigens and lymphoid cell surface markers in lyme synovitis

Using monoclonal antibodies to spirochetal antigens and lymphoid cell surface markers, we examined the synovial lesions of 12 patients with Lyme disease, and compared them with rheumatoid synovium and tonsillar lymphoid tissue. The synovial lesions of Lyme disease patients and rheumatoid arthritis patients were similar and often consisted of the elements found in normal organized lymphoid tissue. In both diseases, T cells, predominantly of the helper/inducer subset, were distributed diffusely in subsynovial lining areas, often with nodular aggregates of tightly intermixed T and B cells. IgD-bearing B cells were scattered within the aggregates, and a few follicular dendritic cells and activated germinal center B cells were sometimes present. Outside the aggregates, many plasma cells, high endothelial venules, scattered macrophages, and a few dendritic macrophages were found. HLA–DR and DQ expression was intense throughout the lesions. In 6 of the 12 patients with Lyme arthritis, but in none of those with rheumatoid arthritis, a few spirochetes and globular antigen deposits were seen in and around blood vessels in areas of lymphocytic infiltration. Thus, in Lyme arthritis, a small number of spirochetes are probably the antigenic stimulus for chronic synovial inflammation.     

Spirochetal antigens and lymphoid cell surface markers in Lyme synovitis. Comparison with rheumatoid synovium and tonsillar lymphoid tissue

Using monoclonal antibodies to spirochetal antigenes and lymphoid cell surface markers, we examined the synovial lesions of 12 patients with Lyme disease, and compared them with rheumatoid synovium and tonsillar lymphoid tissue. The synovial lesions of Lyme disease patients and rheumatoid arthritis patients were similar and often consisted of the elements found in normal organized lymphoid tissue. In both diseases, T cells, predominantly of the helper/inducer subset, were distributed diffusely in subsynovial lining areas, often with nodular aggregates of tightly intermixed T and B cells. IgD-bearing B cells were scattered within the aggregates, and a few follicular dendritic cells and activated germinal center B cells were sometimes present. Outside the aggregates, many plasma cells, high endothelial venules, scattered macrophages, and a few dendritic macrophages were found. HLA-DR and DQ expression was intense throughout the lesions. In 6 of the 12 patients with Lyme arthritis, but in none of those with rheumatoid arthritis, a few spirochetes and globular antigen deposits were seen in and around blood vessels in areas of lymphocytic infiltration. Thus, in Lyme arthritis, a small number of spirochetes are probably the antigenic stimulus for chronic synovial inflammation.     

Role of adhesion molecules in the lymphoid cell distribution in rheumatoid synovial membrane

Summary We studied the staining pattern of a group of adhesion molecules in the lining layer and lymphocytic infiltrates of the rheumatoid synovial membrane, using monoclonal antibodies against lymphocyte function associated antigen-1 (LFA-1), very late activation antigen-4 (VLA-4), VLA-5, endothelial leucocyte adhesion molecule-1 (ELAM-1) and intercellular adhesion molecule-1 (ICAM-1). The cells of the the lining layer were strongly ICMA-1 positive and VLA-5 positive, suggesting (1) that ICAM-1 may function to facilitate the adhesion of ICAM-1-bearing type A cells to type B lining cells and (2) that the lining cells may utilize VLA-5 for anchorage to fibronectin at the surface of the synovial membrane. In the lymphocyte-rich and transitional areas, the endothelial cells of the postcapillary venules were both ELAM-1 positive and ICAM-1 positive. ICAM-1 staining was weak in lymphoid aggregates, but strong in the transitional areas, indicating a paucity of ICAM-1-bearing cells in the lymphocyte-rich areas. On the other hand, LFA-1 staining was very strong in the lymphoid aggregates and only moderate in transitional areas. This suggested that the large numbers of T4 cells present in the lymphocyte-rich areas are sufficiently activated to express substantial levels of LFA-1, and also that the LFA-1 molecule is an important receptor for emigration from postcapillary venules. In germinal centre like areas in lymphoid aggregates, most of the cells stained strongly for ICAM-1 and VLA-4, suggesting that the proliferation of B lymphocytes may be facilitated by LFA-1- and VLA4-dependent T and B cell interaction. The VLA molecules stained in the transitional areas may provide appropriate adhesion and anchorage of lymphocytes for the achievement of the variety of immune reactions in which these cells are engaged in these areas.     

Dendritic cells in rheumatoid synovial membrane after total removal of the hyaline articular cartilage

OBJECTIVE: To investigate the effect of total removal of the hyaline articular cartilage on dendritic cells in synovial membrane in rheumatoid arthritis (RA) or ankylosing spondylitis (AS). PATIENTS AND METHODS: Immunohistochemical staining for two dendritic cell markers, CD35 and RFD1, was carried out on synovial membrane specimens from arthritis patients undergoing primary (n=10) or revision (n=8) total hip replacement (THR). The results are expressed as the number (mean+/-standard deviation) of positive cells per 1000 total cells. RESULTS: CD35-(112+/-9) and RFD1-(27+/-5) positive cells were found in all primary RA synovial membrane, while only two out of eight synovial membrane samples from revision THR contained CD35-positive follicular dendritic cells (nine and 12 cells), and no revision samples contained any RFD1-positive interdigitating dendritic cells. CONCLUSION: Removal of the hyaline articular cartilage reduces the infiltration and functional differentiation of dendritic cells in synovial membrane. Our findings suggest that the antigen driving chronic arthritis/synovitis is contained in the hyaline articular cartilage.     

Macrophage heterogeneity in normal colonic mucosa and in inflammatory bowel disease

Immunohistological techniques using monoclonal antibodies were employed to study the morphology and phenotypic expression of macrophage like cells in ulcerative colitis, Crohn's colitis and histologically normal colonic mucosa. The antibody RFD1 identifies interdigitating (antigen presenting) cells whereas RFD7 binds to mature tissue macrophages. In normal colonic mucosa, the majority of cells recognised by these reagents were positive for Class II antigen expression and a median 87% (range 80-95%) were positive for both RFD1 and RFD7, with 6.5% (ranges 1-14%) positive for either antibody alone. There was much greater macrophage heterogeneity in the ulcerative colitis and Crohn's colitis biopsies than in normal mucosa. Clusters of RFD9+ cells (epithelioid cells) were found in Crohn's colitis and, to a lesser extent, in ulcerative colitis. Some Crohn's colitis sections showed replacement of the normal colonic macrophage phenotype with RFD1-RFD7+ cells (classical scavenger macrophages). The degree of this replacement correlated with the histological severity of the disease. By contrast, large numbers of RFD1+ RFD7- cells, with long dendritic processes, were found in intimate association with the lymphoid infiltrates in the lamina propria of the ulcerative colitis sections. Future studies of the factors controlling macrophage differentiation in tissues may help to explain the greater macrophage heterogeneity in inflammatory bowel disease and the differences between ulcerative colitis and Crohn's colitis observed in this study.     

Gold treatment of rheumatoid arthritis decreases synovial expression of the endothelial leukocyte adhesion receptor ELAM-1.

Leukocyte adhesion receptors on endothelial cells play an important role in the evolution of synovitis. We studied sequential synovial biopsies at Weeks 0, 2 and 12 in 11 patients with rheumatoid arthritis beginning parenteral gold therapy either alone or combined with 120 mg intramuscular methylprednisolone acetate at Weeks 0, 4 and 8 of treatment. Expression of endothelial leukocyte adhesion molecule 1 (ELAM-1) decreased on synovial blood vessels after both 2 and 12 weeks treatment (p less than 0.05), while the overall vascularity of the synovium did not change. Neutrophil numbers within the synovial membrane also decreased although this did not reach statistical significance. In contrast, there was no significant change in numbers or subset distribution of T cells or in Class II MHC expression by synovial lining cells, mononuclear cells or endothelial cells. Our results suggest that one of the early effects of intramuscular gold and glucocorticoid therapy may be a downregulation of the acute inflammatory process associated with the endothelial expression of a neutrophil adhesion receptor and the subsequent recruitment of neutrophils into the joint.     

Immunohistologic characterization of synovial membrane lymphocytes in rheumatoid arthritis

Synovial membrane biopsy specimens from 15 rheumatoid arthritis patients were examined using routine histologic stains and monoclonal antibodies directed against cell surface antigens. Three patterns of lymphoid cell infiltrates were recognized: 1) diffuse infiltration of T cells that surrounded clusters of germinal center B cells (3 patients); 2) diffuse T cell infiltration, lacking germinal centers (8 patients); and 3) proliferation of subsynovial fibroblasts, with relatively few lymphoid cells (4 patients). The synovial, subsynovial, and perivascular tissues in each of the patterns exhibited a high frequency of HLA-DR antigen, HLA-DS antigen, transferrin receptor, and/or epidermal growth factor receptor. In contrast, normal or osteoarthritic synovial tissues did not display a marked increase of these antigens or receptors. Cells bearing natural killer antigen were infrequent in each of these patterns. Active synovitis, synovial effusions, anemia, and elevated sedimentation rate were present in rheumatoid arthritis patients with each of the three histologic patterns. Immunohistologic characterization of synovial membrane infiltrates by these monoclonal antibodies provides additional information about pathogenesis of rheumatoid arthritis and may help in predicting responses to different therapeutic modalities.     

Binding of normal human mononuclear cells to blood vessels in rheumatoid arthritis synovial membrane

We examined the binding of mononuclear cells to blood vessels in the rheumatoid arthritis synovial membrane. Two of 4 synovia that had lymphocyte-rich areas contained blood vessels adhesive for mononuclear cells. These reactive vessels showed striking similarities to high endothelial venules of the paracortical region of lymph node, where normal lymphocyte-endothelial cell adhesion and lymphocyte emigration into tissue occurs.     

Expression of the cutaneous lymphocyte antigen and its counter-receptor E-selectin in the skin and joints of patients with psoriatic arthritis

We have investigated whether the skin-homing T lymphocytes identified by the cutaneous lymphocyte antigen (CLA) are increased in the synovial membrane of patients with psoriatic arthritis. Twenty-six synovial samples (13 psoriatic arthritis, seven rheumatoid arthritis, six osteoarthritis) were obtained from involved knees. Lesional skin biopsies were taken from nine of the patients with psoriatic arthritis and six patients with psoriasis alone. All samples were single- and dual-stained for CLA and CD3 (to identify T lymphocytes) using HECA-452 (anti-CLA) and anti-CD3 monoclonal antibodies. E-selectin expression was also determined. The percentage of dual-stained lymphocytes was significantly greater in psoriatic skin than in synovium (P < 0.001) and similar between psoriatic and rheumatoid synovium. There was no significant difference in the percentages of CLA-positive cells in psoriatic skin in patients with psoriatic arthritis compared with psoriasis alone. The intensity of endothelial E-selectin expression was significantly greater in skin psoriasis than in synovium (P < 2 x 10(- 5)), and rheumatoid synovium had significantly greater expression than psoriatic synovium (P < 0.05). However, there was no significant correlation between E-selectin expression and the percentages of CLA- positive lymphocytes. This study provides further evidence that the CLA antigen is enriched on skin-homing lymphocytes. Conversely, the link between skin and joint inflammation in psoriatic arthritis does not seem to be explained by increased trafficking of CLA T cells to psoriatic synovium.      

Vascular endothelial growth factors C and D and their VEGFR-2 and 3 receptors in blood and lymphatic vessels in healthy and arthritic synovium

OBJECTIVE: To localize vascular endothelial growth factor C (VEGF-C) and VEGF-D in synovial specimens in relation to their VEGFR-2 and VEGFR-3 receptors in blood and lymphatic vessels. METHODS: Immunohistochemical staining and messenger RNA analysis from control and arthritic synovial membrane specimens. RESULTS: Quantitative RT-PCR disclosed that VEGF-C mRNA copy numbers were higher than VEGF-D mRNA copy numbers in the rheumatoid arthritis (RA), osteoarthritis, and control patient groups studied (p < 0.01). Immunohistochemical staining localized VEGF-C to synovial lining cell layer, pericytes, and smooth muscle cells of blood vessels. The number of VEGF-C positive cells was increased in the synovial lining of ankylosing spondylitis (AS) and RA compared to control synovium. However, in contrast to control synovial lining, little if any VEGF-D was detected in AS or RA synovial lining. VEGFR-2 expressing stromal blood vessels, also positive for the vascular endothelial marker PAL-E and the basement membrane marker laminin, were more abundant in RA and AS than in controls. Interestingly, the lymphatic endothelial receptor VEGFR-3 was also expressed in most synovial vessels, especially in the sublining capillaries and venules. CONCLUSION: VEGF-C is strongly expressed in the hypertrophic synovial lining of arthritic joints, whereas VEGF-D expression is very low in AS and RA. The expression of VEGF-C and VEGF-D in pericytes and smooth muscle cells suggests that these factors may have a role in maintaining vascular homeostasis. The VEGF receptors VEGFR-2 and VEGFR-3 are present in most of the sublining blood vessels. The expression of the lymphatic marker VEGFR-3 in the sublining blood vessels may relate to fluid filtration and/or fenestrations. The relatively few lymphatic vessels along with increased vascular permeability in RA may contribute to the development of tissue edema and joint stiffness.     

Involvement of subchondral bone marrow in rheumatoid arthritis: lymphoid neogenesis and in situ relationship to subchondral bone marrow osteoclast recruitment

OBJECTIVE: To evaluate the presence and immunohistochemical characteristics of subchondral bone marrow inflammatory infiltrate in rheumatoid arthritis (RA) and to determine the in situ relationship between marrow inflammation and osteoclast recruitment. METHODS: Bone samples and paired synovia from 8 RA patients undergoing joint surgery were analyzed by immunohistochemistry and in situ hybridization for specific lymphoid neogenetic features, such as T and B cell composition, follicular dendritic cell (FDC) networks, peripheral lymph node addressin (PNAd)-positive high endothelial venules, and lymphoid chemokine expression. Osteoclasts were identified as multinucleated tartrate-resistant acid phosphatase (TRAP)-positive and cathepsin K-positive cells adherent to the bone surface. RESULTS: An inflammatory infiltrate with perivascular aggregates of variable size was detected in 7 (87.5%) of 8 synovial samples and in paired bone samples. Lymphoid neogenetic features typical of rheumatoid synovium were also recognized in the bone marrow. PNAd+ blood vessels were found in 4 of 8 patients, CD21+ FDC networks in 2 patients, CXCL13+ cells in 5 patients, and CCL21+ cells in 6 patients. TRAP-positive and cathepsin K-positive osteoclasts were identified on both the synovial and marrow sides of the bone surface. Bone marrow samples showing a higher degree of inflammation were characterized by a significantly increased number of osteoclasts adherent to the subchondral bone. CONCLUSION: Our data demonstrate that lymphoid aggregates with lymphoid neogenetic features are detectable on the subchondral side of the joint in established RA. Moreover, the local inflammation/aggregation process appears to be related to osteoclast differentiation on the marrow side of subchondral bone, supporting a functional role of the bone compartment in local damage.     

Evidence in support of a self-perpetuating HLA-DR-dependent delayed-type cell reaction in rheumatoid arthritis.

Originating from observations on similarities between the rheumatoid synovial tissue and skin lesions in delayed-type hypersensitivity reactions--similarities as to massive infiltrates of "helper" T lymphocytes close to HLA-DR-expressing macrophage/dendritic cells--a notion is formed on the importance of local macrophage-dependent helper T-cell activation in the rheumatoid joint similar to that in a delayed-type skin reaction. In vitro studies on suspended synovial cells have been used to test and qualify these ideas. It is shown that (i) HLA-DR-expressing cells in normal synovial intima can, like epidermal Langerhans cells, mediate T-cell activation; (ii) the large numbers of rheumatoid synovial HLA-DR-expressing macrophage-like/dendritic cells are heterogeneous and mediate either efficient activation or suppression of T-lymphocyte proliferation, and (iii) specificity of rheumatoid T cells can be analyzed with the help of autologous synovial antigen-presenting cells; a specific anti-collagen type II response is reported in three patients.     

Glycosylation-dependent inhibition of cutaneous lymphocyte-associated antigen expression: implications in modulating lymphocyte migration to skin

Constitutive E-selectin expression on dermal microvascular endothelial cells plays a critical role in mediating rolling adhesive interactions of human skin-homing T cells and in pathologic accumulation of lymphocytes in skin. The major E-selectin ligand on human skin-homing T cells is cutaneous lymphocyte-associated antigen (CLA), a specialized glycoform of P-selectin glycoprotein ligand-1 (PSGL-1) defined by monoclonal antibody HECA-452. Since HECA-452 reactivity, and not PSGL-1 polypeptide itself, confers the specificity of human T cells to enter dermal tissue, inhibition of HECA-452 expression is a potential strategy for modulating lymphocyte migration to skin. In this study, we examined the efficacy of several well-characterized metabolic inhibitors of glycosylation and of a novel fluorinated analog of N-acetylglucosamine (2-acetamido-1,3,6-tri-O-acetyl-4-deoxy-4-fluoro-D-glucopyranose [4-F-GlcNAc]) to alter HECA-452 expression on human CLA(+) T cells and prevent cell tethering and rolling on selectins under shear stress. At concentrations that did not affect PSGL-1 expression, we found that swainsonine (inhibitor of complex-type N-glycan synthesis) had no effect on HECA-452 expression or selectin ligand activity, whereas benzyl-O-N-acetylgalactosamide (BAG; inhibitor of O-glycan biosynthesis) ablated HECA-452 expression on PSGL-1 and significantly lowered selectin ligand activity. We found that 4-F-GlcNAc (putative inhibitor of poly-N-acetyllactosamine biosynthesis) was more potent than BAG at lowering HECA-452 expression and selectin binding. In addition, we show that 4-F-GlcNAc was directly incorporated into native CLA expressed on T cells, indicating direct inhibition on poly-N-acetyllactosamine elongation and selectin-binding determinants on PSGL-1 O-glycans. These observations establish a potential treatment approach for targeting pathologic lymphocyte trafficking to skin and indicate that 4-F-GlcNAc may be a promising agent for treatment of dermal tropism associated with malignancies and inflammatory disorders.     

Behaviour of synovial blood vessels in semi--thin sections in progressive chronic polyarthritis]

The synovial blood vessels of 17 patients with clinically definite rheumatoid arthritis (RA) were studied histologically and ultramicroscopically. No differences between seronegative and seropositive RA subjects were found. The abnormalities observed concerned mainly the endothelial cells, which showed changes varying in degree from swelling to fibroblastic transformation. The destruction of endothelial cells seemed to be linked with a positive test for C-reactive protein as a consequence of an acute relapse. Areas of cellular infiltration and partial homogenization were found in the tunica media of the vessel wall. There was a striking hyperemia of the capillaries and venules.     

Cells expressing dendritic cell markers are present in the rheumatoid nodule

OBJECTIVE: To determine if dendritic antigen-presenting cells (DC) are present in rheumatoid nodules, as has been reported in the synovial lesions of rheumatoid arthritis. METHODS: Nodules (n = 14) were examined with monoclonal antibodies (Mab) recognizing the DC differentiation/activation markers CD83, CMRF44, and CMRF56 and an antibody recognizing the CD1a antigen present on epithelial tissue associated DC. Results. Cells expressing CMRF44 were common in rheumatoid nodules, comprising 22% of nucleated cells versus 13% in synovial membranes (n = 10). Cells positive for CD1a (5%) and CD83 (2%) were less common. A majority (86%) of CMRF44 positive cells were also positive for the macrophage marker CD14. This left a significant minority of putative DC that were single stained with CMRF44. CONCLUSION: Cells bearing DC markers are as frequent in the rheumatoid nodule as in the synovial lesions. A majority are "indeterminate" cells that are CD14 positive but a proportion are single stained putative DC. The lack of lymphoid collections containing DC and T and B lymphocytes in the nodule suggests that local presentation of antigen may not occur in the rheumatoid nodule, as is thought to be the case in synovial membranes containing lymphoid follicles. This difference could potentially be explained by different states of activation, and differentiation of DC within the 2 lesions.     

Immunohistochemical localization of somatostatin receptor sst2A in human rheumatoid synovium

OBJECTIVE: To identify the somatostatin receptor-expressing cells in rheumatoid synovium using a recently developed antiserum directed against the somatostatin receptor subtype 2A (sst2A). METHODS: We carried out immunohistochemical studies of synovial biopsies from 7 patients with rheumatoid arthritis (RA) and one non-RA patient, using a rabbit polyclonal antiserum directed against sst2A and monoclonal antibodies directed against phenotypic markers. RESULTS: SSt2A was expressed by the endothelial cells of the synovial venules but also by a subset of synovial macrophages. CONCLUSION: The identification of somatostatin receptors on macrophages, which are thought to be important effector cells in RA, may offer mechanistic insights into the potential therapeutic effect of somatostatin (analogs) in RA.     

Chondrocyte-derived cells and matrix at the rheumatoid cartilage-pannus junction identified with monoclonal antibodies

Summary In the cartilage-pannus junction of 14 patients with rheumatoid arthritis (RA) and seven patients with osteoarthritis (OA), monoclonal antibodies to keratan sulphate (KS) and chondroitin sulphate (CS) stained a transitional fibroblastic zone (TFZ) within the pannus in nine RA patients and one OA patient. In three patients this was clearly localised to the cytoplasm of cells in this zone, but in all remaining cases KS and CS could be demonstrated in the surrounding matrix. This area was distinguished from adjacent pannus which contained many blood vessels and cells positive for MHC Class II antigen. Specific markers for glycosaminoglycans have been employed to demonstrate that chondrocyte-derived cells and matrix contribute to the changes seen at the cartilage-pannus junction in RA-affected joints.     

Identification of cutaneous lymphocyte-associated antigen as sialyl 6-sulfo Lewis X, a selectin ligand expressed on a subset of skin-homing helper memory T cells

We previously identified the carbohydrate determinant sialyl 6-sulfo Lewis X (Le(x)) as the major L-selectin ligand on high endothelial venules of peripheral lymph nodes. In this study, we examined the distribution of the sialyl 6-sulfo Le(x) determinant among peripheral lymphocytes. The determinant was expressed on a subset of helper memory T and NK cells. The helper memory T cells expressing sialyl 6-sulfo Le(x) were CD45RO(bright+) PSGL-1(high+) CCR4+ L-selectin+ CCR7+ but did not express alpha4beta7 integrin or CCR9, indicating that they were the skin-homing population of central memory T cells. The T-cell subset significantly expressed mRNA for 6-sulfotransferase HEC-GlcNAc6ST and fucosyltransferase Fuc-T VII, responsible for the synthesis of sialyl 6-sulfo Le(x). Characteristics of the T-cell population were similar to those previously described for cutaneous lymphocyte-associated antigen (CLA)-positive T cells defined by the HECA-452 or 2F3 antibody. The binding of the T-cell subset with the specific anti-sialyl 6-sulfo Le(x) antibody G152 was almost completely abrogated by HECA-452 or 2F3. Binding of recombinant E-, P-, and L-selectins to the T-cell subset was significantly inhibited by G152 and by HECA-452 antibodies. We propose that CLA, which is expressed without any activation stimuli on peripheral skin-homing helper memory T cells in healthy persons, is at least partly the sialyl 6-sulfo Le(x) determinant.     

Demonstration of lymphatics in human synovial tissue

Summary Using a cocktail of monoclonal antibodies PAL-E and DE-U-10 (anti-desmin), combined in double labelling techniques with the lectin Ulex europaeus agglutinin I (UEAI), vessels consistent with lymphatics were demonstrated in normal human synovial tissue. These vessels were negative for the monoclonal cocktail and positive for UEAI, were thin-walled and were located close to deep arterioles and venules as expected. Elastin was not found to assist identification of lymphatics in synovium. In rheumatoid arthritic synovium no vessels staining in the manner of normal lymphatics were found. This may indicate absence or change of phenotype of this type of endothelium in disease.