Numbers of rosette formine cells in human peripheral blood.

Thymus-derived (T-cell) and “bursal” derived (B-cell) lymphocytes in human peripheral blood were quantitated by assaying percentages of cells forming erythrocyte rosettes. T-cell rosettes were formed with neuraminidase treated sheep erythrocytes. B-cell rosettes were formed with complement coated sheep erythrocytes. Large differences in the percentages of T-rosette forming cells were noted depending on the method used to assay these cells. When rosette forming cells (RFC) and non-RFC were counted concurrently the percentage of T-cell rosettes were 53–75% whereas methods involving the separate counting of RFC and total cells gave T-cell RFC percentages of 23–40%. These differences were due to the “co-rosetting” of non-RFC into the T-cell rosette clusters. This occurred because of the gentleness required to resuspend the fragile T-cell rosettes. “Co-rosetting” was demonstrated by forming stable complement receptor rosettes with complement-coated human erythrocytes and resuspending them either gently or vigorously. Significantly higher percentages of rosettes were noted with gentle cell suspension than with vigorous resuspension. The percentages of rosette forming T-cells in human peripheral blood are therefore lower than previously estimated.     

Human autologous rosette-forming cells. III. Binding of erythrocytes from different species to the T-cell receptors for autologous red blood cells

Spontaneous rosette formation in humans is restricted to a subpopulation of the circulating T cells. We have previously shown that the interaction between lymphocytes and autologous red blood cells (auto-RBC) is not mediated by a self-recognition mechanism, since allogeneic (allo-) RBC bind to T cells through the same receptors. In this work, we have extended these observations to thymocytes. Using a mixed-rosette assay in which one type of erythrocyte was identified by FITC labeling, we have shown that almost all the thymocytes which attached auto-RBC could also fix allo-RBC. However, as for the peripheral blood lymphocytes (PBL), binding of human RBC to thymocytes occurred with varying affinities according to the erythrocyte's origin. In order to further study the specificity of the erythrocyte to lymphocyte binding in rosette formation, PBL were mixed with auto-RBC and erythrocytes of xenogeneic (xeno-) origin. Although very disparate incidences of rosettes were found according to the species from which the RBC were derived, most of the autorosetting lymphocytes also had receptors for xeno-RBC. In addition, preincubation of PBL with monoclonal antibody OKT11A (directed against the sheep RBC receptors on T cells) completely abrogated rosette formation with all the erythrocytes tested (human auto- and allo-, sheep, pig, and rabbit) except mouse RBC. Taken together these data strongly suggest that human auto- or allo-, as well as sheep or some other xeno-RBC, bind to T lymphocytes by a single receptor and that the combining sites are expressed with different densities or varying affinities depending upon the RBC's origin. Therefore, spontaneous autorosettes may represent T lymphocytes having high-affinity receptors for sheep RBC.     

Human autologous rosette-forming cells: II. Specificity of the binding between lymphocytes and erythrocytes

The relationship between autorosettes and allorosettes was investigated using a mixed rosette assay in which the origin of the erythrocytes was assessed by the fluorescein isothiocyanate (FITC) labeling of one type of erythrocyte. The data show that auto- and allorosettes belong to the same T-cell subset: (1) in most of the subjects, the percentages of T cells binding autologous red blood cells (auto-RBC) are equivalent to those binding allogeneic RBC (allo-RBC); (2) the percentage of rosettes formed after the simultaneous addition of auto- and allo-RBC is similar to that of autorosettes alone or allorosettes alone; and (3) nearly 80% of the resetting cells bind both types of RBC as directly visualized in the mixed rosette assay. The experiments in which the lymphocytes are resetted first with one type of RBC, and then with the other type support the finding that auto- and allo-RBC may bind to the lymphocytes through a single receptor which exhibits a varying affinity for RBC according to their origin.     

On the mechanism of early recovery of specifically depleted lymphoid cell populations by nonspecific activation of T cells.

Specific depletion from normal CBA mouse spleen cells of those bound on pigeon erythrocyte (PRBC) immunoabsorbent columns before transfer of the depleted population into irradiated syngeneic recipients resulted in elimination of the anti-PRBC responsiveness as assessed by rosette (RFC) and hemolytic plaque (PFC) formation. The anti-sheep erythrocyte (SRBC) responses of cell populations treated in the same manner remained unimpaired. When, however, these populations were stimulated with both PRBC and muramyl dipeptide (MDP), an early recovery of specific anti-PRBC responsiveness was produced. PFC response in particular, suddenly increased between the fourth and fifth day after transfer and stimulation thus exhibiting a doubling time of only 4 to 6 hr. This effect of MDP was T-cell dependent since treatment of the depleted population with anti-θ antigen serum and complement hindered early recovery. Depleted populations stimulated with PRBC alone resumed their T-dependent RFC (but not PFC) responsiveness after the eighth day. In spite of the existence of these educated T cells, a second stimulation on the tenth day with PRBC was unable to elicit a specific PFC response. On the other hand stimulation with MDP alone on the day of cell transfer (Day 0) followed by stimulation with PRBC on Day 10 resulted in a specific PFC response on Day 15. Thus, MDP appeared to do more than simply promote education of T cells by antigen. In vitro cultures of depleted populations also recovered their specific reactivity when stimulated by antigen and MDP.     

Studies on mouse autoreactive cells: I. Role of H-2 antigens in mouse autologous rosette formation

A minority (1–2%) of normal mouse lymphoid cells bind autologous erythrocytes and form rosettes. In this study we examined the antigenic specificity involved in the formation of such rosettes. A significant difference in the incidence of rosettes formed, respectively, with autologous and allogeneic mouse erythrocytes is found. Moreover, preincubation of lymphoid cells with low concentrations of syngeneic erythrocytic ghosts causes significant competitive inhibition of subsequent rosette formation. Allogeneic ghosts obtained from nonrelated or from congenic resistant strains of mice do not display this inhibitory effect under the same conditions. It is thus suggested that mouse autologous rosette-forming cells bear receptors for syngeneic H-2 antigens that are involved in the binding of autologous erythrocytes. More precisely, compatibility between lymphocyte and erythrocyte restricted to K or D only is sufficient to ensure a level of rosettes similar to that obtained when complete identity occurs for K, I, and D regions.     

Histidine-rich glycoprotein blocks T cell rosette formation and modulates both T cell activation and immunoregulation

Histidine-rich glycoprotein (HRGP) is a plasma and platelet protein with undefined function in vivo. It has been reported to inhibit rosette formation between murine T cells and erythrocytes. We have shown that HRGP binds specifically to human T lymphocytes but not sheep erythrocytes and have demonstrated a 56-kDa HRGP-binding protein on the T cell surface which is distinct from the CD2 sheep erythrocyte receptor. We have now investigated whether HRGP can inhibit human T cell-sheep erythrocyte rosette formation and whether HRGP can modulate T cell activation. HRGP at physiologic concentrations specifically inhibited rosette formation between human T lymphocytes and sheep erythrocytes. HRGP suppressed proliferation of antigen receptor (CD3)-triggered T cells induced by interleukin 2; this suppression was specifically reversed by prior incubation of HRGP with affinity-purified anti-HRGP IgG. Addition of HRGP 12-24 h after CD3 triggering no longer suppressed T cell proliferation, suggesting HRGP suppressed T cell division by interfering with one or more early events in the process of T cell activation. Human serum (containing 100-150 micrograms/ml HRGP) was also capable of suppressing T cell proliferation; serum which had been immunodepleted of HRGP no longer inhibited T cell proliferation. Furthermore, HRGP inhibited interleukin 2 receptor expression on activated T cells, causing decreased T cell interferon-gamma release and altered T cell-dependent inhibition of erythropoiesis. HRGP is thus capable of modulating T cell activation and T cell immunoregulation; HRGP may function as a natural suppressive regulator of human T lymphocyte activation.     

Rosette formation of concanavalin A-treated erythrocytes around polymorphonuclear leucocytes and lymphocytes.

Concanavalin A (Con A) induces rosette formation of erythrocytes around polymorphonuclear leucocytes and lymphocytes in cell suspensions of autologous human blood cells. The effect which is most characteristic in a concentration between 25 and 50 microgram/ml is due to Con A bound on the erythrocyte membrane. A similar effect, although less pronounced, was observed with phytohaemagglutinin at concentrations of 10 and 25 microgram/ml. The treated erythrocytes showed a higher affinity to polymorphonuclears when compared with lymphocytes. At the contact area, the membrane of the erythrocyte became highly folded while its free surface was smooth and spherical. The effect of the local concentration and immunobilization of the lectin on the erythrocyte membrane and the similarity of the contact pattern to that of erythrophagocytosis are discussed.     

Human complement receptor type 1 (CR1) binds to a major malarial adhesin

Plasmodium falciparum erythrocyte membrane protein 1 (PfEMP1), a major adhesin molecule expressed on Plasmodium-falciparum-infected erythrocytes, interacts with several receptors on endothelial cells and uninfected erythrocytes. This 'stickiness', known as rosetting, is a strategy used by the parasite to remain sequestered in the microvasculature to avoid destruction in the spleen and liver. Erythrocyte rosetting causes obstruction of the blood flow in microcapillaries. Recent data suggest a direct interaction between PfEMP1 and a functional site of complement receptor type 1 (CR1; CD35) on uninfected erythrocytes. Consistent with the hypothesis that CR1 is important in malaria pathogenesis is a 40-70-fold increase in the frequency of two CR1 blood-group antigens (at least one of which might rosette less efficiently) in malaria-exposed African populations. Furthermore, structural differences in erythrocyte CR1 between human and non-human primates are probably explained by the selective pressure of malaria.     

Transport of neutral amino acids by human erythrocytes

The transport of several neutral amino acids by human erythrocytes in vitro was studied. The measurements made included steady-state distributions, kinetics of initial rates of uptake, effects of monovalent cations and anions, general mutual inhibitory interactions, kinetics of inhibitions, effluxes, ability to produce accelerative exchange diffusion, and the inhibitory action of the thiol reagent N-ethylmaleimide. The results are interpreted as showing that the human erythrocyte membrane possesses several distinct transport systems for these amino acids, including one Na⁺-dependent system and one dependent on both Na⁺ and a suitable anion, that are qualitatively similar to those systems previously described in pigeon erythrocytes and mammalian reticulocytes. Quantitatively, however, the systems differ among the different kinds of red cell and a major difference lies in their abilities to produce accelerative exchange diffusion.     

Appearance of B- or T-lymphocyte markers after diffusion chamber culture of acute lymphoblastic leukemia cells

Blast cells derived from blood and marrow samples of a patient with acute lymphoblastic leukemia (ALL), as well as from the Reh line originally established from an ALL patient, were cultured in diffusion chambers implanted i.p. into preirradiated CBA mice. At different intervals over a period of up to 20 days, surface immunoglobulins, ALL antigen, and T-cell antigen were investigated by using direct immunofluorescence. Rosette formation was tested with sheep and mouse erythrocytes. On day 0, the cells expressed only ALL antigen at the surface, and no rosette formation was observed. During culture the patient's lymphoblasts, which originally had cytoplasmic IgM in addition to ALL antigen, expressed surface immunoglobulins as well as mouse erythrocyte receptors. The Reh line cells were ambivalent in two experiments developing T-cell antigens and sheep erythrocyte receptors as well as mouse erythrocyte receptors. Our data suggest that the differentiation arrest in leukemic lymphoblasts can be overcome, thus entailing a surface pattern similar to mature T- or B-lymphocytes.     

Rheopathologic Consequence of Plasmodium vivax Rosette Formation

Malaria parasites dramatically alter the rheological properties of infected red blood cells. In the case of Plasmodium vivax, the parasite rapidly decreases the shear elastic modulus of the invaded RBC, enabling it to avoid splenic clearance. This study highlights correlation between rosette formation and altered membrane deformability of P. vivax-infected erythrocytes, where the rosette-forming infected erythrocytes are significantly more rigid than their non-rosetting counterparts. The adhesion of normocytes to the PvIRBC is strong (mean binding force of 440pN) resulting in stable rosette formation even under high physiological shear flow stress. Rosetting may contribute to the sequestration of PvIRBC schizonts in the host microvasculature or spleen.     

Studies on cell fusion between Babesia rodhaini-infected mouse erythrocytes and baby hamster kidney cells.

Heterokaryons were formed by fusion of B. rodhaini-infected mouse erythrocytes and baby hamster kidney (BHK) cells, using Sendai virus. The erythrocyte membrane rapidly lysed inside the BHK cell cytoplasm releasing free parasites. There was no evidence that parasite multiplication occurred inside the BHK cells, nor that parasitized BHK cells were infective for mice.Transient erythrocyte homokaryons were observed in some preparations.The approach indicates a possible method for the in vitro cultivation of Babesia.     

Duplicated nucleosome repeat generated in the erythrocyte chromatin by DNAse I. The role of lysine-rich histones

Dinucleosome periodicity of DNA fragmentation produced by DNAse I in nuclei of pigeon and trout erythrocytes differing in the content of histones H1 and H5 has been investigated. In spite of differences in the content of histone H5 (H1 to H5 ratio is approximately equal to 0.5 and 2 in pigeon and trout erythrocytes respectively) the double-nucleosome repeat was revealed clearly in pigeon and trout erythrocyte nuclei. To elucidate the role of lysine-rich histones we carried out the selective extraction of histone H1 from erythrocyte nuclei by a solution containing 0.3-0.35 M NaCl (pH 3.0) or cleavage of histones H1 and H5 by mild trypsinization in the presence of Mg2+ ions. It was shown that lysine-rich histones play a principal role in formation and maintenance of the so-called dinucleosomal chromatin structure.     

Lymphocyte populations of Callithrix jacchus marmosets.

A lymphocyte population of common marmosets (Callithrix jacchus) was identified by rosette formation with African green monkey erythrocytes; the rosette-forming cells appeared to be T lymphocytes, as approximately 62% of circulating lymphocytes and 85% of thymus cells formed rosettes with African green monkey erythrocytes. In addition, common marmoset lymphoid cells carrying T-lymphotropic Herpesvirus saimiri or Herpesvirus ateles formed rosettes with African green monkey erythrocytes and treatment of common marmoset circulating lymphocytes with an anti-T cell serum and complement (C′) eliminated rosette-forming cells. Common marmoset T lymphocytes apparently carry a surface receptor for African green monkey erythrocytes, but unlike humans and other closely related nonhuman primates, T lymphocytes of common marmosets fail to form rosettes with sheep erythrocytes.     

Demonstration of choleragen-dependent ADP-ribosylation in whole cells and correlation with the activation of adenylate cyclase

3T3C₂ mouse fibroblasts rendered permeable to (α^?32P)NAD⁺ show cholera toxin-dependent labeling of a 45,000 m.w. protein and of a doublet of polypeptides around 52,000 m.w. These same bands are ADP-ribosylated in broken cells. Membranes prepared from pigeon erythrocytes pretreated with choleragen show a decrease in subsequent cholera toxin-specific ADP-ribosylation of a 43,000 m.w. polypeptide. Both whole cell and broken cell adenylate cyclase activation and toxin-specific ADP-ribosylation are reversed specifically by low pH and high concentrations of toxin and nicotinamide in all systems. Thus ADP-ribosylation appears to be relevant to the molecular action of choleragen in whole cells as well as in broken cells.     

Guinea pig and gerbil erythrocytes rosette with different cells in the blood, bone marrow, and thymus of the cat

Cat thymocytes, bone marrow cells, and peripheral blood leukocytes (PBL) formed rosettes with guinea pig (GP) and gerbil (G) erythrocytes (E). In PBL from adult cats the frequency of rosettes was 27% with GPE and GE, while an average of 33% bone marrow cells formed rosettes with GPE and only 4% with GE. Thymocytes from kittens showed a high percentage of rosettes with both GPE and GE (35 to 81%), with the frequency of each type varying with the thymus tested. Fluorescein isothiocyanate labeling of one of the erythrocyte species revealed these cells to be rosetting with different nucleated cells; i.e., a low percentage (3-5%) of the rosettes formed with PBL and bone marrow had both labeled and unlabeled erythrocytes. In contrast, "mixed" rosettes were observed with a significant number of thymocytes, averaging 33% of thymocytes from six animals. A further distinction between the GE- and GPE-rosetting cells was revealed by a monoclonal antibody which blocked GE rosette formation without interfering with the binding of GPE to PBL and thymocytes. PBL could be depleted of either GPE- or GE-rosetting cells, with retention of IgG+ cells and cells capable of rosetting with the second erythrocyte species in the nonrosetting fractions. Stimulation of the latter nonrosetting fractions with pokeweed mitogen for induction of Ig synthesis revealed a T-lymphocyte specificity of the GE- and GPE-rosetting cells. PBL depleted of GE-rosetting cells yielded an increased Ig production, two- to threefold above the control; in contrast, depletion of GPE-rosetting cells from PBL resulted in a failure of the remaining cells to respond. These results suggest that T-suppressor cells of the cat are contained in the GE-rosetting fraction and T-helper cells are rosetted with GPE.     

Can cytological features differentiate reactive renal tubular cells from low‐grade urothelial carcinoma cells?

H. Ohsaki, E. Hirakawa, Y. Kushida, S. Yokoshita, M. Nakamura, H. Kiyomoto and R. Haba Can cytological features differentiate reactive renal tubular cells from low‐grade urothelial carcinoma cells? Objective: To compare the cytomorphological and immunocytochemical features of reactive renal tubular cells and low‐grade urothelial carcinoma cells (LG‐UCs). Methods: We examined 15 cytological parameters in 38 cases with reactive renal tubular cells in renal disease and 20 cases of LG‐UCs from bladder cancer that had been diagnosed by histological examination. Voided urine cytological parameters evaluated were as follows: (i) maximum cell numbers of clusters, (ii) cannibalism, (iii) rosette‐like arrangement, (iv) hobnail‐shaped cells, (v) vacuolated cytoplasm, (vi) intracytoplasmic haemosiderin, (vii) irregular nuclear contours, (viii) chromatin pattern, (ix) prominent nucleoli, (x) cast encasement, (xi) casts, (xii) dysmorphic erythrocytes, (xiii) isomorphic erythrocytes, (xiv) necrosis, and (xv) vimentin reactivity. The above parameters were determined using Mann–Whitney U‐test and chi‐square test, with differences considered significant at P < 0.05. Results: In reactive renal tubular cells, low to moderate cell numbers of clusters (fewer than 50 cells), rosette‐like arrangement, hobnail‐shaped cells, vacuolated cytoplasm, intracytoplasmic haemosiderin, euchromatin pattern, prominent nucleoli, dysmorphic erythrocytes and vimentin reactivity were present in significantly higher proportions compared with those in LG‐UCs. In LG‐UCs, high cell numbers of clusters (50 cells or more), cannibalism, heterochromatin pattern, isomorphic erythrocytes and necrosis were seen in significantly higher proportions. No significant differences were observed in irregular nuclear contours, cast encasement or casts. Conclusions: Based on results of the present study, maximum cell numbers of clusters, cannibalism, rosette‐like arrangement, hobnail‐shaped cells, vacuolated cytoplasm, intracytoplasmic haemosiderin, chromatin pattern, prominent nucleoli, dysmorphic erythrocytes, isomorphic erythrocytes, necrosis, and vimentin reactivity were capable of distinguishing reactive renal tubular cells from LG‐UCs.     

Macrophage recognition of periodate-treated erythrocytes: involvement of disulfide formation of the erythrocyte membrane proteins

Upon exposure to 2 mM periodate at 0 degrees C for 15 min, mouse erythrocytes underwent membrane lipid oxidation, oxidation of cell surface sialyl residues into aldehyde-bearing derivatives, and oxidation of SH groups of the membrane proteins into disulfides. The periodate-treated erythrocytes exhibited a remarkable increase in rosette attachment to resident mouse peritoneal macrophages in the absence of serum. The relationship between the oxidation of the membrane constituents and the macrophage recognition of these cells was investigated. Periodate treatment of erythrocytes in the presence of butylated hydroxytoluene, an inhibitor of lipid oxidation, did not affect the subsequent attachment of the erythrocytes to the macrophages. Reduction of the periodate-treated erythrocytes with borohydride or cyanoborohydride did not affect the erythrocyte attachment. Neuraminidase treatment of erythrocytes before periodate did not affect the attachment either. On reduction of the disulfides of the membrane proteins with dithiothreitol, the periodate-treated erythrocytes lost their ability to attach to the macrophages. Erythrocytes treated with an SH-oxidizing agent, diamide, were then examined for the macrophage recognition. The diamide-treated cells also showed rosette attachment to the macrophages in the absence of serum, but did not when reduced with dithiothreitol. These results indicate that oxidation of the SH groups of the membrane proteins to disulfides causes reversible membrane changes that macrophages recognize, and it is this mechanism that is responsible for the macrophage recognition of the periodate-treated erythrocytes.     

The specificity and significance of the inhibition of Fc receptor binding by anti H-2 sera

The possibility that Ia antigens are unique among H-2 antigens in their relationship to the Fc receptor was investigated in an EA rosette assay. Antibody specific for antigens in various regions of theH-2 complex was incubated with mouse cells, and the ability of the cells to form rosettes with antibody-coated chicken erythrocytes was tested. Antibody raised against the H-2 antigens of Ia-negative tumor cells was highly effective in inhibiting rosette formation. A variety of antisera againstK-, I-, andD-region antigens tested in recombinant mice inhibited EA rosette formation, suggesting that antigens in each of these regions could be detected in rosette inhibition. The F(ab′)₂ fragments of all antisera tested also produced specific EA rosette inhibition. Finally, antibody against Ia antigens failed to inhibit bone marrow RFCs, although antibody against H-2K and H-2D antigens did inhibit. Although H-2 serology is in a state of rapid change at present, it must be concluded that in this assay, antibody against antigens in theK andD regions as well as theI region can inhibit EA rosette formation. Inhibition of these rosettes by anti H-2 sera is therefore not due to a special association of Ia antigens with Fc receptors.     

Sodium-induced conformation changes in membrane transport proteins

In the presence of KCl, tryptic digestion of vesicles derived from pigeon erythrocyte membranes inactivates sodium-dependent uptake of alanine by the vesicles, whereas digestion in the presence of NaCl does not. Extensive degradation of vesicle proteins occurs under both conditions. Similarly, the extent of inhibition by N-ethylmaleimide of the sodium-dependent influxes of both glycine and alanine into human erythrocytes is greater when the cells are exposed to the thiol reagent in the presence of KCl than when NaCl is used. These observations are interpreted as providing evidence for sodium-induced conformation changes in these transport proteins.