Lymphoid cells with an affinity for the epidermis (epidermotropic lymphocytes) have been proposed to play a role in the immune functions of the epidermis. However, antigen-presenting Langerhans cells (LC) and indeterminate cells are presently the only cells in the human epidermis which have been demonstrated to originate in the bone marrow. Recent studies of murine epidermis have identified a population of bone marrow-derived cells which express Thy-1 antigen and which are present in a similar density to, but distinct from, LC. We therefore sought to identify the potential human analogue of the murine Thy-1+ epidermal cell utilizing a battery of antileukocyte reagents in immunohistochemical, flow cytometric, and cell sorting studies. A panel of antibodies failed to detect significant numbers of human Thy-1 antigen-bearing cells, T cells, B cells, monocytes/macrophages (other than LC), and natural killer cells in tissue sections, epidermal sheets, and epidermal cell (EC) suspensions. This was the case using EC suspensions either unfractionated or fractionated on Ficoll-Hypaque to enrich for leukocyte subpopulations. Since the nature of the murine Thy-1+ EC is uncertain, it is possible that antibodies directed against well-defined leukocyte subpopulations may not be of value in the detection of a potential human analogue. We therefore utilized double fluorescence staining with anti-HLe-1, an antibody which identifies all human leukocytes, and anti-HLA-Dr (Dr), which identifies epidermal LC, in order to demonstrate a potential population of HLe-1+ Dr- non-LC, bone marrow-derived cells. The vast majority of HLe-1+ cells were HLA-Dr+ LC; these were present at a density of 608 cells/mm2 in epidermal sheets. A minor population of HLe-1+ cells which did not express HLA-Dr (HLe-1+ Dr-) was observed in tissue sections, epidermal sheets, and EC suspensions. The nondendritic morphology and low density of these HLe-1+ Dr- EC in epidermal sheets (mean density of 4.2 +/- 1.6 cells/mm2) precluded their representing a strict human analogue of the murine Thy-1+ EC, since murine Thy-1+ EC are dendritic and are present in a density similar to that of LC. Purified preparations of the minor HLe-1+ Dr- EC population obtained by electronic cell sorting or panning and examined ultrastructurally were not enriched for any bone marrow-derived cell population. Thus, using currently available markers and sorting technology, we have been unable to identify a human analogue of the murine dendritic Thy-1+ epidermal cell.

Binding of peanut agglutinin (PNA) to normal human peripheral blood mononuclear cells was analyzed on a cell sorter, and compared to the binding of the monocyte specific monoclanal antibodies Mac-1 and Leu-M3. Each of the reagents labeled 9–11% of the mononuclear cells and similar binding patterns were observed. Of the PNA+ cells, 67% adhered to plastic petri dishes, whereas 76% of Mac-1+ cells were adherent. No competition for binding was observed between PNA and Mac-1 on the one hand, or PNA and Leu-M3 on the other. In double staining experiments, about 10% of the cells, comprising 80% of the monocytes, were PNA+ Leu-M3+. Our results show that PNA can serve for the identification and enumeration of monocytes in human peripheral blood.

The present study examines the role of monocytes in the in-vitro activation of human T cells and B cells by pokeweed mitogen (PWM). The T cell-dependent PWM-induced B-cell activation process was found to be monocyte dependent. Fluorescence-activated cell sorter (FACS) analysis revealed that upon addition to peripheral blood mononuclear cells, fluoresceinated PWM, at concentrations that provided optimal B-cell and T-cell activation, bound predominantly to human monocytes. The binding of PWM to monocytes was reversible and could be displaced within the first few hours of binding by oligomers of N-acetylglucosamine (GlcNAc). As a functional correlate of the binding studies, it was shown that PWM-pulsed monocytes could induce B lymphocytes to become plaque-forming cells (PFC) and T lymphocytes to undergo proliferation. In contrast, markedly reduced PFC and blastogenic responses were observed when monocyte-depleted B lymphocytes and T lymphocytes were respectively pulsed with PWM and washed, followed by the addition of non-PWM-pulsed monocytes to the cultures. Thus, the initial event in the PWM-induced activation of human lymphocytes, for both in-vitro T-lymphocyte blastogenic responses and B-lymphocyte Ig secretion, appears to be binding of the mitogen to sugar residues on the surface membrane of the monocyte, followed by subsequent interaction with the appropriate lymphocytes. The process of PWM binding to monocytes did not appear to affect the baseline production of interleukin-1 (IL-1) by human monocytes, nor could soluble factors from PWM-pulsed monocytes substitute for intact cells in the initiation of the lymphocyte-activation process.

The binding of the mitogenic lectins concanavalin A, Wistaria floribunda mitogen, Pisum sativum hemagglutinin, Lens culinaris lectin, phytohemagglutinin, and two of the pokeweed mitogens (Pa-1 and Pa-2), in mitogenic concentrations to the cell surface acceptors of murine splenic lymphocytes caused a rapid, transient increase in the methylation of lymphocyte phospholipid. The extent of this increase was dependent upon the concentration of the lectin, and paralleled the dose-response curve for subsequent mitosis. Higher or lower doses of lectin did not cause the observed changes in methylation nor induce mitosis. The increase of phospholipid methylation reached a maximum about 10 min after the addition of the mitogenic lectins and then decreased to control levels within about 45 min. Accordingly, mitogenic lectins not only stimulate phospholipid methylation in lymphocytes, but also stimulate degradation of the phospholipids, probably by phospholipase A2. The non-mitogenic lectins, Sophora japonica hemagglutinins (Sj-I and Sj-II), W. floribunda hemagglutinin, and Bauhinea purpurea hemagglutinin did not significantly affect the phospholipid methylation.

Native Con A and two chemical derivatives, divalent dimeric Con A and monovalent dimeric Con A. induced a transient increase of phospholipid methylation, Ca2+ influx, and also increased DNA synthesis in murine lymphocytes. For each of the individual mitogens, the dose-response curves for these three activities were very similar. However, there were major differences between the dose-response curves for Con A and each of its two chemical derivatives. On the other hand, the time course of phospholipid methylation for each lectin reached a maximum at about 10 min after the addition of lectin, and then gradually decreased to control levels. In like manner, Ca2+ influx reached its maximum at approximately 5 min. The lectin-stimulated increase in phospholipid methylation occurred in calcium-free medium, while the inhibitor of phospholipid methylation, 3-deaza-SIBA, also suppressed the increased calcium influx. This suggests that the Ca2+ influx might be regulated by early phospholipid methylation. Further, in the absence of calcium, the methylated phospholipids do not undergo Con A-accelerated breakdown by phospholipase A2. This suggests that the increased influx of calcium is necessary for the activation of phospholipase A2, an enzyme that hydrolyses methylated phospholipids to yield arachidonic acid and lysolecithin. Blocking any of these biochemical steps also blocked subsequent DNA synthesis, suggesting that the pathway may be required for the activation of lymphocytes.

Monocytes play essential roles in the activation of both T and B lymphocytes (1,2,3). To study the interactions of monocytes, as well as other cells, with the responding lymphocytes, methods of fractionating the different cell types are necessary. During our studies on cell surface carbohydrates and carbohydrate receptors on lymphoid cells, it became apparent that human monocytes strongly bound both a complex carbohydrate, and rather specifically bound the lectin Pa-4. These observations led to two new methods for the fractionation of human monocytes from peripheral blood.

Cell-free supernatant fluid, from cultures of Phytolacca americana (pokeweed) lectin 2 (Pa-2)-pulsed murine spleen or thymus cells, contains factors which induce cultured lymphocytes to differentiate into IgM-secreting cells (assayed by a reverse plaque technique) and to proliferate (measured by the incorporation of tritiated thymidine) without the addition of mitogen. The factors in this supernatant fluid responsible for these activities have been designated as lymphocyte stimulating factors (LSF). LSF showed no genetic restrictions related to the major histocompatability complex; LSF made in one strain of mice worked in other strains. Indeed, LSF is not restricted by species barriers; human peripheral blood mononuclear cells were also stimulated by murine LSF to proliferate and differentiate into immunoglobulin-secreting cells without further addition of antigen or mitogen. Maximum production of LSF was achieved within 12 hr of culture and was independent of cell division. In contrast to TRF, no further production of LSF was detectable after 24 hr of culture. Unlike T-cell growth factor, this material stimulated increased mitosis of thymic, T, and B lymphocytes without the addition of mitogen or antigen. LSF also stimulated polyclonal B-cell differentiation into IgM-secreting cells. Maximal numbers of immunoglobulin-secreting cells were generated when LSF was added at the initiation of the culture. Indeed, unlike TRF, LSF needed to be present only during the first 6 hr of culture to achieve maximum stimulation, and did not require the presence of antigen. The production of LSF by a T-cell population in the spleen was shown by two independent methods. Spleen cells treated with anti-Thy 1 plus complement failed to produce detectable levels of LSF. On the other hand, purified populations of surface immunoglobulin-negative spleen cells produced LSF. Furthermore, the subset of thymocytes responsible for LSF production was the small population (approximately 10%) of cells in the thymus, which are not agglutinated by peanut agglutinin.

Fluorescein-labeled lectins bound to mouse thymocytes were analyzed by flow microfluorometry. This technique has identified several lectins that bind differentially to thymocyte subsets. The most complex fluorescence distributions were obtained using lectins with nominal specificities for galactose or N-acetylglucosamine. Inhibition of binding by sugars confirmed that the fluoresceinated lectins were bound to cells at their carbohydrate binding site. Simultaneous analyses of lectin fluorescence and forward light scatter intensity showed that cell subpopulations of different sizes can exhibit marked differences in the level of binding such that the amount of lectin bound per cell is often independent of cell size. A minor population of dull or unstained cells, delineated by several of these lectins, correlates with the subpopulation of medium-sized thymocytes resistant to in vivo cortisone treatment.