Diagnostic and prognostic importance of T-cell receptor gene analysis in patients with Sézary syndrome.

BACKGROUND: Sézary syndrome (SS) is characterized by erythroderma, peripheral lymphadenopathy, and circulating Sézary cells and is clinically heterogeneous. METHODS: T-cell receptor (TCR) gene analysis was performed using DNA extracted from peripheral blood mononuclear cells from 74 patients, and the results were correlated with a variety of other diagnostic parameters and patient outcomes. RESULTS: Two groups were identified: 66 patients with clonal TCR gene rearrangement (clonal patients) detected with Southern blot analysis and/or polymerase chain reaction/single-strand conformational polymorphism analysis and 8 patients with no clonal rearrangement detected (nonclonal patients) using either technique. Clonal patients were compared with nonclonal patients. The following median blood parameters were significantly greater in the clonal group: total white cell count (13.7 10(9)/L vs. 9.6 10(9)/L), lymphocyte count (4.9 10(9)/L vs. 2.2 10(9)/L), absolute Sézary count (3.22 10(9)/L vs. 0.99 10(9)/L), CD4 count (3.17 10(9)/L vs. 1.36 10(9)/L), and CD4:CD8 ratio (15.86 vs. 3.21). An expanded population of T-cells of a specific TCR variable beta subset was detected in 7 of 36 clonal patients and in 1 of 4 nonclonal patients. Cytogenetic analysis of peripheral blood from 1 nonclonal patient and 6 clonal patients was normal. The median survival from the time of diagnosis was 45 months in the clonal group, and 40 of 49 deaths were cutaneous T-cell lymphoma (CTCL)-related, whereas 3 deaths in the nonclonal group were unrelated to CTCL (P < 0.01; log-rank test). Multivariate proportional hazards analysis showed that the absolute Sézary count and lymph node status were independent prognostic variables (P = 0.016 and P = 0.036, respectively). CONCLUSIONS: TCR gene analysis defines a distinct clinicopathologic group of patients with SS. Clonal patients have a poor prognosis and are likely to die from leukemia/lymphoma, whereas nonclonal patients may have a reactive, inflammatory T-cell disorder. The authors suggest that the definitive diagnostic criteria for patients with SS should include the presence of a clonal TCR gene rearrangement.

Highly active antiretroviral therapy normalizes the potential for MIP-1alpha production in HIV infection.

DESIGN: The CC chemokines RANTES, MIP-1alpha and MIP-1beta are ligands for CCR5, which has been identified as the principal co-receptor for macrophage tropic strains of HIV-1. This study investigated whether the inducible levels of RANTES, MIP-1alpha and MIP-1beta produced by cultured whole blood samples related to different rates of progression of HIV infection and to the introduction of Nelfinavir-based highly active anti-retroviral therapy (HAART). METHODS: Study subjects were HIV-positive and categorized as "slow progressors" (n= 8) or as "fast progressors" (n= 7); the latter group were treated with HAART. MIP-1alpha, MIP-1beta and RANTES production was determined using commercial ELISA kits. RESULTS: The inducible production of MIP-1alpha by whole blood cells in culture was significantly depressed in patients starting therapy compared with "slow progressors" and "normal donors". The levels of MIP-1alpha significantly increased with therapy at 12 weeks compared with pre-HAART levels (P= O.05) and became comparable to that of "normals" and "slow progressors". Differences in the inducible levels of MIP-1beta and RANTES for the separate subject groups were not significant. CONCLUSIONS: The increase in inducible MIP-1alpha production following HAART might suggest a role for the chemokines in HIV disease, either for monitoring the outcome of therapy of HIV disease, or as a direct therapeutic intervention.

Rapid cytokine up-regulation of integrins, complement receptor 1 and HLA-DR on monocytes but not on lymphocytes.

Short-term (3 hr) incubation of whole blood with human recombinant cytokines induced rapid changes in the expression of monocyte but not of lymphocyte surface molecules. The percentage of monocytes bearing CD11b molecules was enhanced by tumour necrosis factor-beta (TNF-beta), whilst that of CD11c was increased by both TNF-alpha and TNF-beta. The mean fluorescence intensity (MFI) of monocyte CD11a was enhanced by interleukin-2 (IL-2), TNF-alpha and TNF-beta, and that of CD11b, CD11c and CD18 was increased by IL-2, IL-4, TNF-alpha and TNF-beta. The proportion of monocytes expressing HLA-DR antigens was not modified by the cytokines investigated, but its MFI was increased by IL-2, IL-4, TNF-alpha and TNF-beta. In contrast, the percentage of monocytes bearing complement receptor 1 (CD35) was enhanced by IL-2, TNF-alpha and TNF-beta but the MFI of this molecule was not modified by these cytokines. The highest up-regulation of CD18, HLA-DR and CD35 was observed with 100 U/ml of either IL-2, IL-4, TNF-alpha or TNF-beta. Decreasing the concentration of all four cytokines from 100 to 10 and 1 U/ml diminished the levels of expression of all molecules, with the exception of CD35, which reached its maximum upon incubation with 1 U/ml of TNF-alpha. IL-1 beta, IL-6 or interferon-gamma (IFN-gamma) did not modify the expression of any of the above monocyte surface determinants. Moreover, none of the lymphocyte surface molecules investigated was modified by 3-hr incubation of blood with cytokines. The demonstration that cytokines selectively and rapidly up-regulate integrins, complement receptor 1 and HLA-DR molecules, on monocytes but not on lymphocytes, suggests that similar mechanisms of mononuclear cell activation by cytokines may control the development and duration of the inflammatory process.

Cytokines in human intraocular inflammation.

The presence of interleukin 6 (IL-6), interleukin 1 (IL-1), interleukin 2 (IL-2) and tumour necrosis factor (TNF) was investigated in vitreous and aqueous aspirates from eyes undergoing vitrectomy for the treatment of different inflammatory conditions. Cadaveric vitreous from 10 normal subjects were used as controls. IL-6 was observed in 5 specimens from eyes with idiopathic uveitis (range = 26-264 pg/ml), in 2 specimens from eyes with uveitis complicated with retinal detachment (28 and 279 pg/ml, respectively), in 6 samples from eyes with diabetic retinopathy (range = 5-480 pg/ml), in one sample from an eye with phacolytic glaucoma (1190 pg/ml) and in one specimen from an eye with Behçet's disease (366 pg/ml). Although IL-1 was detected in 80% of all the samples investigated, concentrations of this cytokine greater than 3 pg/ml were only observed in 2 specimens from eyes with uveitis (5 and 20 pg/ml, respectively) and 2 samples from eyes with diabetic retinopathy (3 and 31 pg/ml, respectively). TNF was present in 3 specimens from eyes with uveitis (range = 2-24 pg/ml) and 1 sample from eyes with diabetic retinopathy (4 pg/ml), but was not detected in the eyes with phacolytic glaucoma or Behçet's disease. IL-2 (less than 0.1 U/ml) was detected in one sample from an eye with uveitis, one specimen from an eye with uveitis complicated with retinal detachment and 2 samples from eyes with diabetic retinopathy. None of the cytokines measured were detected in any of the control vitreous. The present observations suggest that cytokines, particularly IL-6 and IL-1, may act as local amplification signals in pathological processes associated with chronic eye inflammation.

Selective up-regulation of human granulocyte integrins and complement receptor 1 by cytokines.

The percentage of human granulocytes expressing the integrins CD11b and CD11c as well as complement receptor 1 (CD35) was increased by short-term incubation of whole blood with interleukin-2 (IL-2), interleukin-4 (IL-4) and tumour necrosis factors alpha and beta (TNF-alpha and TNF-beta). The mean fluorescence intensity of granulocyte CD18 was also increased by the above cytokines, whilst that of CD11b was only increased by TNF-alpha. Up-regulation of granulocyte CD18 expression was seen with 1 U/ml of IL-2, TNF-alpha or TNF-beta, in contrast to the effect of IL-4 which was only observed with 100 U/ml. Similarly, enhanced expression of CD35 was induced by 1 U/ml of IL-2 or TNF-alpha but not by concentrations of IL-4 or TNF-beta lower than 100 U/ml. Cytokine effects on the CD11/CD18 complex and CD35 molecules were not modified by cycloheximide, suggesting that their increased expression was not due simply to synthesis de novo. None of the granulocyte surface determinants investigated was altered upon short-term incubation of blood with either IL-1, IL-6 or interferon-gamma (IFN-gamma). The demonstration in vitro that cytokines selectively up-regulate granulocyte integrins and complement receptor 1, suggests that similar mechanisms may be operating during the control of granulocyte-mediated inflammatory processes.

The influence of zinc status and malnutrition on immunological function in Crohn's disease.

Cellular immunity is likely to be important in the pathogenesis of Crohn's disease; whether it is abnormal is not clear. The heterogeneity of patients with Crohn's disease probably underlies the disparity of reports, but attempts to determine which clinical features influence cellular immunity have been largely unsuccessful. This is probably caused by the omission of nutritional status as a potential factor, even though zinc deficiency has frequently been linked with abnormal immunity. Therefore, a detailed study of nutritional and tissue zinc status, nonspecific cellular immunity, and a measure of phagocytic function was performed in 32 patients with Crohn's disease and in a control group of 18 normal subjects and 12 patients with anorexia nervosa. Fourteen patients with Crohn's disease, all patients with anorexia nervosa, but none of the normal controls were malnourished. Peripheral blood lymphocyte population levels were normal in patients with Crohn's disease and in normal controls, but there was a small decrease in the levels of patients with anorexia nervosa. In vivo delayed hypersensitivity skin test responses were profoundly depressed in patients with anorexia nervosa and decreased in patients with Crohn's disease who were malnourished or receiving systemic glucocorticoids. In vitro lymphocyte transformation was reduced in malnourished patients with Crohn's disease, but there were only minor changes in patients with anorexia nervosa. There were alterations of in vitro immunoregulation in Crohn's disease, but they were not responsible for the abnormal lymphocyte transformation responses in malnourished patients. In vitro phagocytic function was reduced in patients with active Crohn's disease. These findings suggest that depressed in vivo and in vitro cellular immunity in malnourished patients with Crohn's disease is caused by a qualitative lymphocyte defect and that depressed in vivo but normal in vitro cellular immunity in anorexia nervosa is caused by a quantitative defect. Thus, malnutrition in Crohn's disease resembles kwashiorkor; in anorexia nervosa, it resembles marasmus. Tissue zinc status was mostly normal in Crohn's disease and anorexia nervosa, and zinc deficiency was not responsible for depressed nonspecific cellular immunity in either condition.

Cytokines in proliferative vitreoretinopathy.

This study determined the presence of interleukin 1 (IL-1), interleukin 6 (IL-6), tumour necrosis factor alpha (TNF alpha), tumour necrosis factor beta (TNF beta), interferon gamma (IFN gamma), transforming growth factor beta 2 (TGF beta 2) and fibroblast proliferation activity (FPA) in vitreous aspirates from eyes undergoing vitrectomy for the treatment of retinal detachment complicated by proliferative vitreoretinopathy (PVR) or uncomplicated retinal detachment (RD). Cadaveric vitreous from normal subjects were used as controls. The results showed that IL-1 and IL-6 predominated in vitreous from eyes with PVR or RD, and that concentrations of IL-6 greater than 20 pg/ml were more frequently found in PVR than in RD (p = 0.031) or control specimens (p = 0.006). Low levels of TNF alpha were observed in 4/18 eyes with PVR, 1/15 eyes with RD and 1/15 control vitreous, and small concentrations of TNF alpha were seen in 3/18 eyes with PVR, 1/15 eyes with RD and 2/15 control vitreous. IFN gamma was detected in 12/18 eyes with PVR, but only in 5/15 eyes with RD (p = 0.048) and 6/15 control specimens. TGF beta 2 was present in all vitreous samples at concentrations ranging from 100 to 4,500 pg/ml with no significant differences among the three groups. Control vitreous possessed the greatest FPA when compared with vitreous from eyes with PVR (p = 0.031) or RD (p = 0.048). These observations provide further evidence that cytokine-mediated pathways of inflammation are involved in the pathogenesis of PVR and point to the possible involvement of IL-1, IL-6 and IFN gamma in cellular interactions leading to chronicity.

Release of cytokines during generation of lymphokine-activated killer (LAK) cells by IL-2.

Supernatants of IL-2-activated mononuclear cells (MNC) that displayed an optimal lymphokine-activated killer (LAK) cell activity at 48-72 hr in culture were found to contain increased levels of tumour necrosis factor alpha (TNF alpha), interleukin-1 alpha (IL-1 alpha) and interferon-gamma (IFN-gamma) when compared with supernatants from mononuclear cells cultured in the absence of IL-2. The concentration of TNF alpha and IL-1 alpha produced by MNC at 24 hr was either increased or maintained by extending the cultures to 96 hr. In contrast, TNF beta was only detected at very low levels after 72-96 hr culture, irrespective of whether IL-2 was present or absent. Optimal concentrations of IL-2 needed to induce maximum release of TNF alpha, IL-1 alpha and IFN-gamma by MNC varied among different individuals. Enriched populations of lymphocytes secreted higher levels of all measured cytokines upon activation with IL-2 in contrast to untreated cells. Supernatants from purified monocyte preparations contained high concentrations of TNF alpha and IL-1 alpha regardless of the presence of IL-2 in the cell cultures. This work suggests that in addition to the generation of LAK cell activity, by promoting the release of other cytokines with potential anti-tumoricidal activity, IL-2 may be amplifying cell-mediated cytotoxicity, which is associated with protection against neoplastic disease.

The production of arthritis in the guinea-pig by intra-articular reaction between lymphokines and inflammatory leucocytes.

A single intra-articular injection of lymphokine into the guinea-pig knee joint resulted in a sequence of changes in joint architecture whose histopathological features resembled that of an acute inflammatory reaction progressing to a chronic state. At 24 h there was a mild hyperplasia and hypertrophy of the synovium with intense polymorphonuclear leucocyte infiltration. At 72 h, the synovium was heavily infiltrated with diffuse and focal aggregations of mononuclear cells; erosion of cartilage and bone by synovial pannus was accompanied by a subsynovial fibrosis. By 1 week, leucocytic infiltration of the synovium had decreased markedly although the erosion and fibrosis persisted. However, when lymphokine was injected together with oil-elicited peritoneal exudate cells a more intense arthritis ensued: at 72 h synovial pannus was prominently eroding bone and this was accompanied by the appearance of multinucleate cells resembling osteoclasts in the zone of erosion. These features were shown to resemble closely the histopathology of experimental allergic arthritis in the guinea-pig, in contrast to the lesser severity of synovitis resulting from the adoptive cellular transfer of delayed hypersensitivity into the joint. The results indicate that lymphokines may play a role in the induction of experimental allergic arthritis by recruiting and activating cells involved in chronic inflammation.

Modulation of Fc and C3b receptor expression on guinea-pig macrophages by lymphokines.

The decrease in Fc-receptor-positive cells that occurred during a 6 h incubation of resident and elicited guinea-pig macrophages was partly abrogated when lymphokines were present in the culture. When the same lymphokine preparations were tested on C3b receptor-expression they preferentially sustained the percentage of C3b rosettes formed by resident rather than elicited macrophages. This lymphokine-induced maintenance of Fc and C3b rosettes by cultured macrophages may have been due to an inhibition of receptor release or an increase in receptor synthesis. Supernatants from cultured macrophages contain shed Fc and C3b receptors which inhibit rosette formation by other macrophages. From the demonstration that culture supernatants from both lymphokine-treated and untreated macrophages significantly inhibited Fc and C3b rosette formation by freshly obtained macrophages it seems that the shedding of Fc and C3b receptors is not modified by lymphokines. The maintenance of Fc and C3b rosettes by lymphokines was inhibited by treatment of the macrophages with cycloheximide, suggesting that the lymphokine effect was due to an increase in synthesis de novo of the Fc and C3b receptors. The lymphokine-inducing antigens, BGG and PPD, and control lymphokine preparations were devoid of receptor modifying activity. The reduction in the percentage of Fc rosettes after 6 h culture appears to be due to a loss of Fc receptors for IgG1. Although lymphokines partly inhibited this effect they could not prevent the loss of these receptors following 24 h culture, unlike their action in augmenting the expression of Fc receptors for IgG2. These findings suggest that a selective enhancement of Fc receptor synthesis by lymphokines may modify the functional activities of macrophages.

Interleukin-1 and interleukin-2 activity in chronic hepatitis B virus infection.

Abnormalities of lymphocyte proliferation in chronic hepatitis B virus infection are well documented, although the underlying mechanisms are poorly understood. To determine whether these defects may be secondary to disordered lymphokine production, we have simultaneously assayed interleukin-1 and interleukin-2 production in 31 chronic carriers of the hepatitis B virus. Supernatants from mononuclear cells cultured both in the presence and absence of lipopolysaccharide contained significantly increased quantities of interleukin-1 activity in patients compared with normal controls (p less than 0.01). Lysates of monocytes from patients also contained more interleukin-1 than those of controls (p less than 0.05) in the presence of lipopolysaccharide or silica, or both. These results indicate that interleukin-1 production is markedly elevated in patients with chronic hepatitis B virus infection, whereas in contrast, interleukin-2 production was found to be reduced in these patients (p less than 0.01). As one of the biological properties of interleukin-1 is to stimulate fibroblasts to produce collagen, the relationship between fibrosis in the liver biopsy specimen and interleukin production was examined. There was a highly significant correlation (p less than 0.001) between interleukin-1 production and the severity of fibrosis, suggesting that this lymphokine may be closely related to the development of cirrhosis in such patients.

Polymorphonuclear leucocytes in Crohn's disease and ulcerative proctocolitis: association between enhanced adherence to nylon fibre and disease variables.

The adherence of polymorphonuclear leucocytes (PMN) to nylon fibre was investigated in patients with Crohn's disease, ulcerative proctocolitis, and anorexia nervosa, and compared with changes of circulating PMNs, C reactive protein concentrations, erythrocyte sedimentation rates, and clinical assessment of disease activity. PMN adherence was in excess of the maximum value detected for healthy subjects in 14 of 25 patients with Crohn's disease and two of 10 with proctocolitis, but it was within the normal range for all eight with anorexia nervosa. High adherence in Crohn's disease, however, was not associated with quantitative or qualitative changes of PMN populations, absolute concentrations of C reactive protein, erythrocyte sedimentation rates, disease severity, drug regimens, malnutrition, or zinc deficiency. High PMN adherence in Crohn's disease may therefore reflect the activation in vivo of normal PMN by humoral factors.

Shedding and synthesis de novo of Fc and C3b receptors by cultured guinea-pig macrophages.

Resident macrophages freshly obtained from the peritoneal cavity of guinea-pigs were demonstrated to form a higher percentage of Fc and C3b rosettes than elicited macrophages when low concentrations of IgG and IgM-C3b were used to sensitize ox red blood cells (ORBC) in rosette assays. Culture of the total resident and elicited macrophages for 6 h at 37 degrees C resulted in a decrease of Fc and C3b rosette-forming cells, the loss of Fc receptor-bearing cells by resident macrophages only being apparent when using a sub-optimal concentration of sensitizing IgG. After 24 h incubation the percentages of Fc and C3b rosettes returned to their initial values. In contrast, there was no decline in the percentage of Fc and C3b rosettes formed by the adherent population of resident and elicited macrophages cultured for 6 h. However, extending the incubation of the adherent macrophage to 24 h produced an increase of Fc receptor-positive cells and a dramatic decrease of C3b receptor-positive cells. Culture supernatants of the total macrophage population that had been incubated for 6 h inhibited Fc and C3b rosette formation by freshly obtained elicited macrophages. These results, together with the demonstration that treatment of the total macrophage population with cycloheximide led to an inhibition of Fc and C3b receptor expression after 24 h culture, suggest that the Fc and C3b receptors of guinea-pig macrophages are shed and synthesized de novo during short-term culture. This system could be applied to the study in vitro of soluble immunoregulatory mediators on macrophage functions which are dependent on the expression of Fc and C3b receptors.

Analysis of human lymphocyte transformation responses to graded doses of T cell mitogens by curve fitting.

The between-group comparison of complete lymphocyte transformation dose-response curves is complex. We have therefore derived a mathematical model of the dose-response characteristics of human mononuclear cells to stimulation by concanavalin A (ConA) and purified phytohaemagglutinin (PHA), in order to simplify such analyses. This model describes dose-response curves in terms of the magnitude of the peak response, the dose of mitogen that elicits the peak and an estimate of the range of mitogen doses which induce a response. Responses to ConA were described by the model more precisely than those to PHA. Furthermore, use of the model revealed differences between anorexia nervosa patients and healthy subjects in terms of the dose of mitogen necessary to elicit a peak response and the range of mitogen concentrations producing a response. It is proposed that this form of mathematical treatment may be of use for the comparison of lymphocyte transformation dose-response curves and for the valid rejection of suspect results.

Failure of exogenous interleukin 1 and interleukin 2 to correct decreased lymphocyte transformation in chronic hepatitis B virus carriers.

To test the hypothesis that reduced lymphocyte transformation in response to PHA in chronic hepatitis B virus infection might be due to deficient lymphokine production, lymphocyte transformation was measured in the presence or absence of exogenous interleukin 1, interleukin 2 or both, or, as a source of mixed lymphokines, supernatants from mixed lymphocyte reactions. The response to PHA was significantly impaired in patients compared to controls, but was not corrected by interleukin 1, interleukin 2 or supernatant from mixed lymphocyte reactions over a wide range of concentrations. Variation of the proportion of monocytes in culture or the addition of indomethacin had no effect on lymphocyte transformation. Thus, reduced lymphocyte proliferation in response to PHA in patients with chronic hepatitis B virus infection cannot be attributed to deficient lymphokine production or to active suppression by monocytes or prostaglandins and a direct role for the hepatitis B virus or a viral product is under investigation.

Interleukin 1 in Crohn's disease.

A number of the activities currently ascribed to the mediator interleukin 1 (IL-1) are relevant to chronic inflammatory bowel disease. Using the mouse thymocyte stimulation assay, lymphocyte-activating factor (LAF) activity was measured in plasma samples and supernatants from cultures of peripheral blood mononuclear cells from 16 patients with Crohn's disease, six with ulcerative colitis, and 10 healthy subjects. Results were compared with disease activity, drug therapy, granulocyte count, and plasma levels of zinc and C-reactive protein (CRP). Very low levels of LAF were detected in a few plasma samples from each of the subject groups. Mononuclear cells from healthy subjects produced LAF only when cultured with lipopolysaccharide, but stimulated cells from patients produced greater amounts. Moreover, cells from six patients with Crohn's disease, not receiving steroids, produced LAF spontaneously. Crohn's disease patients also had low plasma zinc but elevated levels of CRP and granulocytes. This enhanced production of LAF in vitro may reflect a primary cellular defect in Crohn's disease, or a secondary consequence of monocyte activation.

Cell-mediated immunity in anorexia nervosa.

Twelve patients with anorexia nervosa were studied for cell-mediated immunity in terms of delayed hypersensitivity reactions to recall antigens, lymphocyte transformation responses to T-cell mitogens, and numbers of circulating leucocytes and T-cell subpopulations. Compared to controls, all patients had reduced cutaneous reactions and four were anergic. There was a mild leucopenia in patients and both T4+ and T3+ numbers were slightly reduced. Mean peak transformation responses for patients were slightly lower than controls for phytohaemagglutinin, but not for concanavalin A; however, patients required greater doses of mitogens to elicit peak transformation responses. Plasmas from patients did not contain inhibitors of transformation responses. We conclude that there are functional cellular abnormalities associated with the under-nutrition of anorexia nervosa.

Plans and reagents for a collaborative study to establish an international reference preparation for human interleukin-2.

Different preparations of interleukin 2 have been lyophilised and aliquotted for use in a collaborative study designed to establish a reference preparation. Laboratories which would like to take part in the collaborative study should contact the authors.

A reappraisal of the macrophage electrophoretic mobility (MEM) test for the measurement of lymphokine activity.

Using a cell electrophoretic apparatus, which was sensitive in detecting small changes in electrophoretic mobility (EPM), the macrophage electrophoretic mobility (MEM) test was investigated as a routine method for detecting lymphokine activity. Electrophoretic analysis of guinea-pig macrophages revealed 2 main subpopulations, one with an EPM of 0.90 micron cm s-1 V-1 (fast) and the other, an EPM of 0.83 micron cm s-1 V-1 (slow). From 23 experiments the fast and slow populations were found to consist of 90% and 10% cells, respectively. When macrophages were incubated with standard guinea-pig lymphokine preparations there was a significant decrease in the fast population with a corresponding increase in the slow population. This lymphokine induced 'slowing' of the macrophages was shown to be very reproducible. Since only 50% of macrophages of high EPM were observed to respond to lymphokine activity, it is not surprising that the MEM test has failed in the past when investigators have accepted as significant a 10-15% reduction in EPM, estimated from measurements made on only 10 macrophages. Parallel bioassays indicated that there were appreciable potency differences for macrophage slowing factor (MSF) and macrophage migration inhibition factor (MIF) activities in the lymphokine preparations used which suggest that these activities may be due to different molecular entities.

Localization of macrophage agglutination factor activity to the gelatin-binding domain of fibronectin.

We previously proposed that macrophage agglutination factor (MAggF, a T cell-derived guinea pig lymphokine) is a fibronectin (FN). We now show MAggF binding to gelatin and to peritoneal macrophages is mediated by domains similar to corresponding domains of plasma FN. MAggF activity in lymphokine concentrates prepared by two different methods differed nearly 10-fold in m.w. on gel filtration chromatography. Despite this difference, MAggF dose-activity curves of both preparations were parallel, and MAggF in both preparations bound reversibly to gelatin and to monoclonal anti-guinea pig FN immunoadsorbents. MAggF activity in one preparation was inhibited by the addition of soluble monoclonal antibody specific for the gelatin-binding domain of human FN; inhibitory activity of this antibody was blocked by purified guinea pig plasma FN or partially purified MAggF from the other preparation. Measured MAggF activity of both preparations was reduced in a dose-dependent manner by pretreatment of indicator macrophages with monoclonal anti-human monocyte FN receptor antibody or F(ab')2 fragments or with guinea pig plasma FN. Neither anti-FN receptor antibody nor plasma FN interacted directly with MAggF. Indirect immunofluorescence studies confirmed the presence of uncomplexed plasma membrane receptors for FN on indicator macrophages in MAggF-responsive populations that were able to bind added FN. Our identification of MAggF as lymphokine FN provides a basis for future biochemical analysis of delayed hypersensitivity inflammatory reactions.