J.M. MacSween

The Isolation of Migration Inhibition Factor

Migration inhibition factor (MIF) has been separated from the supernates of tuberculin stimulated lymphocytes and from fetal calf serum by an affinity column of fucosamine bound to agarose beads. A glycoprotein with the characteristics of biologically active MIF and a molecular weight of approximately 50,000 Daltons was obtained by this one stage purification procedure.

Adhesion and spreading behaviour of human peripheral blood mononuclear cells (PBMC) in vitro.

Abstract While the in vitro adhesion and spreading behaviour of fibroblastic cells have been extensively documented, similar studies on the cells of immunological importance are still lacking. We report here the in vitro adhesion and spreading behaviour of normal human peripheral blood mononuclear cells (PBMC) isolated by the Ficoll-Hypaque technique. Since the presence of serum retarded cell adhesion by 25%, serum-free RPMI-1640 medium was used in these studies. Based on the light and scanning electron microscope studies, glass-immobilized PBMC were classified on the basis of their spreading behaviour as follows: (A) Cells with non-deformable nucleus: I, Cytoplasm non-spread; II, Cytoplasm partially spread; III, Cytoplasm well spread. (B) Cells with deformable nucleus: IV, Nucleus partially spread; cytoplasm well spread; V, Nucleus and cytoplasm well spread. These different cell types occur in the normal human peripheral blood in the relative frequency of 24 ± 3: 27 ± 4: 13 ± 2: 20 ± 5: 16 ± 2 percent. Cell types I–III show the characteristics of lymphocytes and the cells of type V were identified as monocytes, while the nature of cell type IV is not known at present. The immobilized PBMC contained T and B cells as confirmed by sheep erythrocyte (SE) and mouse erythrocyte (ME) rosette formation and surface immunoglobulin. The spreading behaviour of these cells appears to be limited by factors such as available surface membrane area, number and density of adhesion sites, other membrane receptors and the deformability of cell and nuclear membranes. Further characterization of these cell types would aid as a behavioural marker for the subpopulations of lymphocytes and monocytes in health and disease.

Models of normal and transformed cell adhesion, and capping and locomotion in vitro.

Abstract It is proposed that patching, capping and endocytosis, and cell locomotion are manifestations of a single process whereby the cell discards foreign materials. Capping results from the binding to the cell surface of particulate (or molecular) objects which cannot function as immovable substratum. This might be described as unsuccessful or abortive cell adhesion in that the particles adhere to the cell rather than the cell adhering to the substratum. Lateral particle movements on the cell surface membrane are effected by the submembranous microfilament-microtubule system, resulting in capping without displacement of the cell. Successful adhesion of the cell to a substratum renders capping and endocytosis impossible and the cell attempts to discard the substratum by mechanisms analogous to capping. The cell achieves this by lateral movement and detachment of the trailing edge. The concept of abortive adhesion leading to capping has been amplified by the development of molecular models of normal and neoplastic cell adhesion in vitro in the presence and absence of serum. In these models, the normal cells have molecule A (adhesion sites) on their surface; they can spread on the substratum in the absence of serum. In the presence of serum, the A molecules on the normal cell surface bind with B molecules in serum, which may be substratum-bound or free in suspension. Binding of free B molecules with cell surface A molecules results in blockage of adhesion sites; these are cleared via capping. New adhesion sites ( A molecules) are produced at the active edges of the cell. Binding of cell surface A molecules with the substratum bound B molecules results in cell adhesion. Transformed cells do not have A molecules on their surface; they cannot spread in the absence of serum. The transformed cells may recruit A molecules from the serum to attain deformability and spreading. These models also relate to capping of gold or resin particles, cell locomotion and regulation of cell division, and lectin-induced agglutination of transformed cells.

Macrophage migration inhibition factor (MIF): reducing the variables.

While the phenomenon of macrophage migration inhibition is a useful indicator of lymphokine release, it may be caused by other substances, it is subject to considerable variability, and it may be masked by the concomitant presence of substances stimulating migration. We have investigated certain aspects of the lymphokine macrophage interaction in order to circumvent these problems. Cells from the murine macrophage cell line RAW 264-7 migrated with less variability than fresh guinea pig peritoneal macrophages and were more sensitive target cells for human macrophage migration inhibition factor (MIF). Assays were performed with serum-free and endotoxin-free medium and in all cases in the presence and absence of L-fucose. This added specificity to the assay in that biological MIF activity was invariably blocked by L-fucose whereas migration inhibition produced by antigen complexed antibody, endotoxin, and periodate was not affected by L-fucose. It was also possible to demonstrate MIF activity in mixtures of MIF and migration stimulation factor by using L-fucose. We suggest that MIF activity is determined with less variability by using a macrophage cell line as indicator cells and performing the assays in the presence and absence of L-fucose.

Lymphokine responses to concanavalin A stimulation: Association with HLA DR antigens

Stimulation of human lymphocytes with concanavalin A (Con A) resulted in variable lymphokine responses as indicated by factors inhibiting macrophage migration (MIF) or stimulating macrophage migration (MStF), or resulted in negligible responses. These responses were consistent for a given individual when repeated after several months. MIF responses were observed more frequently than MStF responses in patients with renal failure who had demonstrable alloantibodies. MStF responses were statistically associated with the presence of HLA DR1 antigens in patients with renal failure and two separate groups of healthy individuals, while MIF responses were associated with DR7 in the three groups studied. There was no correlation between immunoglobulin allotypes and lymphokine responses. These results suggest that lymphokine responses to Con A are indicators of nonspecific immunological responsiveness and are influenced by genes associated with the major histocompatibility complex.

Presence of mitomycin resistant T cells in peripheral blood of normal individuals

T cell colonies can be easily grown from peripheral blood, and are an index of cellular immunocompetence. Mitomycin-treated T cells are used as stimulator cells in mixed lymphocyte reactions and as feeder cells for growth of B cell colonies, the assumption being that mitomycin prevents proliferation of T cells. We tested this assumption by comparing the proliferation of mitomycin-treated T cells in response to stimulation with phytohemagglutinin (PHA) with that of untreated T cells in liquid cultures and in T cell colony assay. We found that incorporation of tritiated thymidine by cells from 11 healthy individuals pretreated with 25, 50 and 100 micrograms/ml mitomycin C was reduced to 13, 11 and 8%, respectively, of that of untreated cells when stimulated by an optimal concentration of PHA (10 micrograms/ml) in liquid cultures. However, parallel experiments with aliquots of the same cells showed that pretreatment with 25, 50 or 100 micrograms/ml mitomycin C merely reduced T cell colonies to 49, 45 and 45%, respectively, of untreated cells. In five additional experiments mitomycin 200 and 400 micrograms/ml reduced T cell colony numbers to 47 and 60%, respectively. Treatment of T cells with 9000 rad completely abolished T cell colony formation. Lower doses of radiation up to 6000 R did not abolish T cell colony formation, although it effectively blocked T cell proliferation to PHA in liquid cultures. 24 h preincubation of T cells with suboptimal doses of PHA and then treatment with mitomycin or radiation did not abolish T cell colony formation. T cells were recovered from the mitomycin-resistant T cell colonies and stimulated in liquid cultures with PHA, untreated or after exposure to 25 micrograms/ml mitomycin C. Incorporation of tritiated thymidine by the mitomycin-treated cells was reduced to 8% of the untreated controls. Our observations suggests: There may be inaccuracies in B cell colony assays using mitomycin-treated T cells because of significant T cell colony formation. There is a population of T cells in the peripheral blood of normal individuals which form colonies and are resistant to mitomycin.

The cellular basis for differential lymphokine responses to mitogen stimulation

Human mononuclear cells from some individuals produce macrophage migration inhibition factor (MIF) when stimulated with Con A while those of others produce migration stimulation factor (MStF). T cells were responsible for these different responses but T4 cells produced MIF and T8 cells produced MStF regardless of the global response which was not explained by the individual T4:T8 ratios. Admixing the T-cell subpopulations in vitro revealed that MIF responses switched to MStF responses between T4:T8 ratios of 75:25 and 50:50 with MStF responders switching at higher ratios than MIF responders. Pulse exposure to supernatants from Con A-stimulated T4-enriched cells significantly reduced migration indices resulting from stimulation of fresh cells, promoting MIF responses regardless of the responder status of the supernatant donor. In contrast, supernatants from T8-enriched cells, when obtained from MStF responders, significantly increased migration indices while there was no effect when the supernatants were obtained from MIF responders. These results suggest that soluble factors from T8 cells are primarily responsible for determining whether an individual mounts a MIF or MStF response to Con A stimulation.

E Rosette Formation by Human T Lymphocytes: A Spontaneous Cell Mediated Cytotoxic Phencmenon

The nature and significance of spontaneous association between unsensitized human T lymphocytes and sheep erythrocytes has been studied in relation to a possible cytotoxic reaction. Human lymphocytes and 51Cr labelled sheep erythrocytes when mixed in ratios of 5 or 10:1 released 70-100% of 51Cr into the supernatants. This suggests that E rosette formation may be the first step in a spontaneous T cell mediated cytotoxic reaction to sheep erythrocytes.

MODELS OF CELL SURFACE EVENTS DURING NORMAL AND TRANSFORMED HUMAN CELLULAR ADHESION IN VITRO

Coordination among cellular adhesion, detachment, locomotion, proliferation, and differentiation regulated by the microenvironment guides the destiny of the developing embryo through morphogenesis and the maintenance of the adult organism. A loss of coordination among these phenomena appears to characterize neoplastic conditions. Neoplastic cells do not respond to controls on cellular division and cell-cell interaction at the level of the cell surface. Studies on cellular adhesion have demonstrated that neoplastic cells are weakly adherent’ and defective in locomotion.* Recent studies on cellular adhesion indicate that several phenomena, such as capping, endocytosis, locomotion, and adhesion, may be interrelated.3-5 Based on these and other observations,&’ we propose that patching, capping, endocytosis, and cellular locomotion are manifestations of a single process: the cell discarding foreign and unwanted materials. Capping is caused by abortive adhesion and does not result in locomotion. Successful adhesion to the substratum renders capping and endocytosis impossible; the cell detaches itself from the substratum by spurning it in a manner analogous to capping. The interpretation of capping as a result of abortive adhesion permits construction of molecular models of normal and neoplastic cellular adhesion in the presence and absence of serum. In this model, normal cells have A molecules (adhesion sites) on their surface; they can spread on the substratum in the absence of serum. In the presence of serum, A molecules on the normal cell surface bind to B molecules in serum, which may be substratum bound or free in suspension. Binding of free B molecules with cell surface A molecules results in blockage of adhesion sites; these blocked sites are cleared via capping. New adhesion sites (A molecules) are produced at the active edges of the cell. Binding of cell surface A molecules with substratum-bound B molecules results in cellular adhesion, Transformed cells do not bear A molecules on their surfaces; they cannot spread in the absence of serum, Transformed cells may recruit A molecules from serum to attain deformability and spreading (FIGURE 1). The transformed cell may also be defective in capping surface-bound ligands. Thus, the degree of dependence on serum for deformation and spreading may be determined by the rate a t which the cell can synthesize endogenously or “recruit” exogenously “A” molecules or by a combination of both rates. The major differences between normal and transformed cellular adhesion and spreading appear to be the lack of adhesion sites and the lower rate of formation of adhesion sites a t the active edges in the latter. The nature of molecules A and B is not known at present. Several recent