Cell-cell interaction in erythropoiesis. Role of human monocytes

Serum supplement, inoculum cell density, and accessory cell effects are dependent on the cytokine combination selected to expand human HPCs ex vivo

BACKGROUND

The prolonged periods of pancytopenia associated with cord blood transplants suggest that in some cases cell numbers may be limiting. The possibility that limiting cell numbers may be overcome and prolonged periods of pancytopenia abrogated by the transplantation of human umbilical cord blood cells expanded ex vivo has led to efforts to define optimal culture conditions for these cells.

STUDY DESIGN AND METHODS

Cord blood CD34+ cells were cultured with three cytokine combinations: SCF+G-CSF+GM-CSF+MGDF (SGGM); IL-6+ SCF+MGDF+Flt3-ligand (6SMF); and IL-1+IL-3+IL-6+G-CSF+GM-CSF+SCF+Epo (GFmix). Serum effects, inoculum concentration (cells/mL) seeding density (cell/cm(2)) and accessory cell effects on the expansion of CD34+ cells were determined.

RESULTS

Cellular outputs were significantly higher with fetal calf serum (FCS) than with cord blood serum (CBS) or adult group AB serum (ABS) in the presence of 6SMF, however, CBS was as effective as FCS. The best seeding concentrations varied for each of the cytokine combinations, and inoculum densities exceeding 1000 cells per cm(2) proved detrimental for cultures containing GFmix and SGGM. Accessory cell studies indicated that populations expressing the CD33 antigen inhibited the expansion of purified CD34+ cells in the presence of GFmix or SGGM, but not in the presence of 6SMF.

CONCLUSION

Serum supplement, inoculum cell concentration, seeding densities, and accessory cell effects are dependent upon the cytokine combination selected to expand cord blood HPCs ex vivo. Thus, each of these measures should be assessed to establish reproducible and reliable conditions for the selection of different cytokine combinations to culture cord blood HPCs.

Serum from continuous ambulatory peritoneal dialysis patients with acute bacterial peritonitis inhibits in vitro erythroid colony formation

During episodes of acute infection there is a reduced response to epoetin therapy. It is well known that "endogenous pyrogens," such as interleukin-1 (IL-1) and tumor necrosis factor, inhibit erythropoiesis when administered exogenously. To determine whether there is a relationship between these observations, serum samples were obtained from nine patients with chronic renal failure maintained by continuous ambulatory peritoneal dialysis, during and after recovery from bacterial peritonitis, to study the effect of circulating factors on erythropoiesis. Normal human bone marrow-derived erythroid progenitors were cultured in vitro in 5% and 10% patient serum. Depression of the growth of late progenitors, colony-forming units-erythroid (at 10% serum, P = 0.005; 95% confidence intervals, 6.2 and 24.4, respectively), was observed but there was no effect on the earlier progenitors, burst-forming units-erythroid (at 10% serum, P = 0.7; 95% confidence intervals, -18.5 and 13, respectively). The effect was not prevented by antisera to IL-1. Similarly, when added to cultures, IL-1 inhibited the colony-forming units-erythroid and the effect was abrogated by IL-1 antisera. These findings suggest that a circulating soluble factor that is inhibitory to erythropoiesis and may contribute to loss of response to epoetin therapy, is present in cases of peritonitis in continuous ambulatory peritoneal dialysis patients.

Low lymphocyte interferon-gamma production and variable proliferative response in anorexia nervosa patients

Interferon-gamma (IFN-gamma) production by peripheral blood mononuclear cells (PBMC) in 14 patients with anorexia nervosa (AN) was significantly lower than in 14 age-matched healthy controls. Follow-up samples in four patients displayed low levels, except in two when they recovered the IFN-gamma production as the hormonal cycles were restored. A large interindividual variation for the lymphocyte proliferative response was observed in 30 AN patients. DNA synthesis of PBMC was normal in 8 patients (27%), significantly increased in 6 (20%) (P < 0.001), and significantly decreased in 16 (53%) (P < 0.001). IFN-gamma inhibition was reversed by culturing a control lymphocyte population with monocytes from patients with AN. This was not observed in cultures of control monocytes and AN lymphocytes. IL-2 receptor (TAC subunit) was assessed and no difference was found in the number of TAC-positive cells between patients and controls. These results point out impaired production of the immunomodulator cytokine IFN-gamma as a major functional defect of AN peripheral lymphocytes.

Inhomogeneity of the circulating BFU-E regulation in sickle cell anaemia: accessory cells properties and BFU-E growth factor response pattern

Sickle cell anaemia (SS) patients with low (< 9%) HbF levels (LFSS) are characterized by an increased number of circulating BFU-E in active DNA synthesis, and release of burst promoting activity (BPA) by unstimulated low density (LD) adherent cells. In contrast, circulating BFU-E from SS patients with high (> 9%) HbF levels (HFSS) are normal in number, largely in resting phase, and their LD cells do not release BPA-like activity. We report now that in LFSS patients, adherent cell depletion decreases BFU-E growth in culture and apparent BFU-E cycling. Furthermore, addition of conditioned media (CM) from LD cells of LFSS patients restored cycling BFU-E expression in culture. Neutralization analysis with anti-GM-CSF antibody demonstrated that GM-CSF is, at least, one factor responsible for BPA activity present in this CM. Thus, GM-CSF is constitutively produced by unstimulated monocytes in LFSS patients. In contrast, HFSS patients' adherent cell depletion increases cycling of BFU-E in culture. CM from HFSS patients inhibits BFU-E expression in culture. Hence, LD adherent cells from HFSS patients may release a yet unknown inhibitor factor(s). In addition, we report a distinct response pattern in SS patients' BFU-E to growth factor (GM-CSF, IL-3): (a) LFSS patients have a BFU-E population, equally responsive to GM-CSF and IL-3; (b) HFSS patients, have a subset of BFU-E exclusively dependent on IL-3 (20-40% of the circulating BFU-E). This pattern is very similar to that of normal BFU-E. In conclusion, BFU-E from LFSS patients represent an actively proliferating population, equally responsive to GM-CSF and IL-3, controlled by constitutively produced GM-CSF, suggesting a unique BFU-E behaviour in SS patients with low HbF levels and high haemopoietic stress. The heterogeneous regulation of BFU-E in SS disease seems to be the epiphenomenon of HbF levels, and not vice versa.

Interactions between purified human cord blood haemopoietic progenitor cells and accessory cells

The role of accessory cells in haemopoiesis remains confused. This appears in large part to reflect the use of impure populations of accessory cells and progenitor cells in previous studies. In this study, cell sorter purified populations of both accessory cells and haemopoietic progenitor cells were used to examine interactions between these cell types. We used a double culture protocol in which purified CD34+ cells were cultured with purified NK, T or monocytic cells in the first liquid culture phase after which the cells were transferred to secondary agar cultures to determine the number of colony forming cells (CFC). NK cells co-cultured with CD34+ cells resulted in an increased number of erythroid progenitors with no effect on the number of nonerythroid progenitors. In contrast, there were increased numbers of erythroid and non-erythroid CFC when the CD34+ cells were co-cultured with either purified T cells or monocytes. CD34+ cells cultured with cell-conditioned media derived from NK cells, T cells or monocytes in transwells where the CD34+ cells and the accessory cells were separated by a 0.2 micron membrane, showed no enhancement in CFC. These results suggest that intimate cell-cell contact is required for these events.

Early circulating erythroid progenitors (BFU-E) in sickle cell anemia

Sickle cell anemia (SS) patients can be divided into two sub-populations according to peripheral HbF levels. Patients with low (< 9%) HbF levels (LFSS) are characterized by an increased number of circulating BFU-E in active DNA synthesis, and release of burst promoting activity (BPA) by unstimulated low density (LD) adherent cells. In contrast, circulating BFU-E from SS patients with high (> 9%) HbF levels (HFSS) are normal in number, largely in resting phase, and their LD cells do not release BPA-like activity. More recently further heterogeneity has been found among these two groups. In LFSS patients GM-CSF is constitutively produced by unstimulated monocytes. In contrast, HFSS patients' adherent cell depletion increases cycling of BFU-E in culture. CM from HFSS patients inhibits BFU-E expression in culture. Hence, LD adherent cells from HFSS patients may release an inhibitory factor(s). The nature of this factor has to be determined. In addition, there are distinct subpopulations of BFU-E responsiveness to growth factor (GM-CSF, IL-3): a) LFSS patients have a homogeneous BFU-E population, equally responsive to GM-CSF and IL-3; b) HFSS patients, in addition to this subpopulation, have a subset of BFU-E dependent exclusively on IL-3 which is 20 to 40% of the total number of circulating BFU-E. This is similar to BFU-E from normal individuals. Hence, LFSS BFU-E represent an actively proliferating population, equally responsive to GM-CSF and IL-3, controlled by at least constitutively produced GM-CSF and possibly other factors. These observations suggest a significant modification in BFU-E behavior in the subset of SS patients with low HbF levels and high hemopoietic stress. The heterogenous regulation of BFU-E in SS disease seems to be an epiphenomenon of HbF levels, and not vice-versa.

Inhibition of interferon gamma production by peripheral blood mononuclear leukocytes of patients with sarcoidosis. Pathogenic implications

Interferon gamma (IFN-gamma) production by stimulated peripheral mononuclear leukocytes of 30 patients with sarcoidosis was studied. A significant inhibition (64 percent, 5 to 330 IU/ml vs normal individuals = 1,000 +/- 250 IU/ml) in the IFN-gamma synthesis was found. The inhibition is due to a defect in the circulating monocytes and not in the peripheral T lymphocytes of these patients. This defect in the peripheral IFN-gamma production could be involved in the pathogenesis of this disease.

Macrophages and cancer

The uncontrolled growth of metastases resistant to conventional therapeutic modalities is a major cause of death from cancer. Data from our laboratory and others indicate that metastases arise from the nonrandom spread of specialized malignant cells that preexist within a primary neoplasm. These metastases can be clonal in their origin, and different metastases can originate from different progenitor cells. In addition, metastatic cells can exhibit an increased rate of spontaneous mutation compared with benign nonmetastatic cells. These data provide an explanation for the clinical observation that multiple metastases can exhibit different sensitivities to the same therapeutic modalities. These findings suggest that the successful therapy of disseminated metastases will have to circumvent the problems of neoplastic heterogeneity and the development of resistance. Appropriately activated macrophages can fulfill these demanding criteria. Macrophages can be activated to become tumoricidal by interaction with phospholipid vesicles (liposomes) containing immunomodulators. Tumoricidal macrophages can recognize and destroy neoplastic cells in vitro and in vivo, leaving nonneoplastic cells uninjured. Although the exact mechanism(s) by which macrophages discriminate between tumorigenic and normal cells is unknown, it is independent of tumor cell characteristics such as immunogenicity, metastatic potential, and sensitivity to cytotoxic drugs. Moreover, macrophage destruction of tumor cells apparently is not associated with the development of tumor cell resistance. Macrophages are found in association with malignant tumors in a definable pattern, suggesting that the most direct way to achieve macrophage-mediated tumor regression is in situ macrophage activation. Intravenously administered liposomes are cleared from the circulation by phagocytic cells, including macrophages, so when liposomes containing immunomodulators are endocytosed, cytotoxic macrophages are generated in situ. The administration of such liposomes in certain protocols has been shown to bring about eradication of cancer metastases. Macrophage destruction of metastases in vivo is significant, provided that the total tumor burden at the start of treatment is minimal. For this reason, we have been investigating various methods to achieve maximal cytoreduction in metastases by modalities such as chemotherapy or radiotherapy prior to macrophage-directed therapy. It is important to note that even the destruction of 99.9% of cells in a metastasis measuring 1 cm2 would leave 10(6) cells to proliferate and kill the host. The ability of tumoricidal macrophages to distinguish neoplastic from bystander nonneoplastic cells presents an attractive possibility for treatment of the few tumor cells which escape destruction by conventional treatments. Macrophage-directed therapy has been studied in several human protocols, yielding important biological information about the use of liposome-encapsulated macrophage activators in cancer patients.(ABSTRACT TRUNCATED AT 400 WORDS)

Effects of hemin on erythropoiesis

It is clear that in vitro hemin increases the number of blood BFU-E derived colonies from normal donors. This occurs with sickle donors as well, despite the increased levels of hemin in vivo in these patients. The effect of hemin on relative gamma globin synthesis is inconsistent, however. In a few cases, delayed addition of hemin led to increased gamma globin synthesis. In time course studies of cultures from normal donors, hemin added on day 0 shifted the day of peak colony number from 13-14 to 16-20 days. The temporal decline in gamma globin synthesis was not altered. In cultures from sickle donors we found that the time for maximal colony number was later than in normals, occurring at 16-20 days even without hemin, and was not further delayed by hemin. The relative proportion of gamma globin synthesis was higher on day 14 in the sickle than the normal cultures, and the temporal decline was somewhat slowed in the sickle cultures by hemin. The elevated gamma synthesis and the later time for peak colony growth in the sickle cultures suggest that the erythroid progenitors in the blood of the sickle patients are less mature than those from normal individuals. There are several possible explanations for the detection of increased numbers of colonies in cultures containing hemin. Hemin may delay the final maturation of erythroblasts within erythroid colonies, thus shifting the time of maximal growth. It may also increase the extent of final maturation, leading to more complete hemoglobinization of the erythroblasts within the colonies, and thus increasing the number of colonies that are eventually recognized as erythroid.(ABSTRACT TRUNCATED AT 250 WORDS)

Serotonin and melatonin synthesis in peripheral blood mononuclear cells: stimulation by interferon-gamma as part of an immunomodulatory pathway

Serotonin and melatonin inhibit phytohemagglutinin- (PHA) induced interferon-gamma (IFN-gamma) production by lymphocytes. In this paper, it is shown that IFN-gamma-increased tryptophan uptake by lymphocytes and macrophages led to an enhanced production of serotonin. When IFN-gamma and serotonin were added together to a lymphocyte culture, N-acetyl serotonin and melatonin production was increased, whereas the path to 5-hydroxy-indoleacetic acid remained unchanged. Therefore, the stimulated IFN-gamma production of serotonin and melatonin by lymphocytes and macrophages and the inhibition of IFN-gamma synthesis by these indoleamines suggest a hypothesis for an immunoregulatory circuit.

Immunomodulation by indoleamines: serotonin and melatonin action on DNA and interferon-gamma synthesis by human peripheral blood mononuclear cells

Different concentrations of indoleamines, serotonin and melatonin, inhibited phytohemagglutinin stimulated DNA synthesis. Thus, 10(-3) to 10(-4) M of either indoleamine acted at the optimal phytohemagglutinin concentration, while 10(-3) to 10(-7) M acted at suboptimal phytohemagglutinin levels. The serotonin effect was reversed by the serotonergic S1-S2 receptor antagonist methysergide but not by the S2 antagonist ketanserin. This indicates that only the S1 receptor is involved in the inhibitory effect. Inhibition of lymphoproliferation by indoleamines was also exerted on pokeweed mitogen and protein A from Staphylococcus aureus stimulations. Serotonin and melatonin also inhibited phytohemagglutinin and protein A from Staphylococcus aureus induction of interferon-gamma synthesis. The initial uptake of Ca2+ was not affected by indoleamines, suggesting that it is not the mechanism of their inhibitory effects. As interferon-gamma induced tryptophan uptake by T lymphocyte- and macrophage-depleted populations, and tryptophan is the metabolic precursor of serotonin and melatonin, a new immunoregulatory circuit is postulated.

Anaemia in juvenile chronic arthritis: serum inhibition of normal erythropoiesis in vitro

Serum from patients with juvenile chronic arthritis (JCA) was shown to inhibit colony formation by normal erythropoietic progenitor cells cultured in vitro. The inhibition was proportional to the degree of anaemia and to certain indices of activity of the arthritis and systemic disease. It occurred in a dose dependent manner with increasing serum concentration and was independent of previous blood transfusion or administered drugs. Erythropoietic progenitor cells from the bone marrows of anaemic patients with JCA showed normal requirements for accessory cells (T lymphocytes and macrophages) in culture, and autologous accessory cells were not deficient in providing normal growth requirements.

Regulation of hematopoiesis by T lymphocytes and natural killer cells

T lymphocytes and natural killer (NK) cells exert both stimulatory and suppressive effects that regulate growth and differentiation of hematopoietic cells. Activated T and NK cells have been demonstrated in different pathological states of bone marrow failure and are proposed to play a role in the pathogenesis of the disease. T and NK cells have also been shown to be responsible for bone marrow graft rejection in both allogeneic and syngeneic donor/recipient combinations. Lymphocytes can regulate hematopoietic cell growth by direct cellular contact or by releasing soluble factors, such as colony-stimulating factors, immune interferon, lymphotoxin, and tumor necrosis factor, active on hematopoietic precursor cells.

T-lymphoblasts with erythropoietic helper function in acute T-cell leukemia

A patient with acute T-lymphoblastic leukemia was found to maintain a normal hemoglobin concentration both at presentation and preterminally several months later, despite a replaced bone marrow and over 80% circulating lymphoblasts on both occasions. Cell surface marker analysis demonstrated the T-lymphoblasts both at presentation and preterminally to belong to the T-helper subpopulation. In vitro culture studies demonstrated that the patient's T-lymphoblasts, as well as conditioned medium derived from these lymphoblasts, significantly stimulated normal bone marrow erythroid colony growth (CFU-E). These findings suggest that in this patient the preservation of erythropoiesis resulted from a helper effect exerted by his T-lymphoblasts.

Regulation of human peripheral blood BFU-E growth in vitro by leukaemic B-lymphocytes

We report a stimulating effect of leukaemic B-lymphocytes from anaemic and non-anaemic patients with CLL on the proliferation of normal peripheral blood BFU-E. Coculture of leukaemic B-cells at various concentrations (2.5 X 10(3)-10(6)) with 2.5 X 10(5) mononuclear cells from normal peripheral blood increased the number of BFU-E derived erythroid colonies. The same effect was observed when the number of target cells was varied in the presence of a fixed number of B-lymphocytes, with a clear linear relationship. B-cell conditioned medium gave a similar increase when added to the culture instead of B-cells. At high concentration of B-cells from anaemic patients, the size of the colonies was increased and a large number of macroscopic colonies was seen. The place of the B-cells in the regulation of erythroid progenitors in relation to monocytes and T-lymphocytes has still to be established.

A regulatory role of activated T-lymphocytes on human megakaryocytopoiesis in vitro

Cellular interactions responsible for regulating in vitro megakaryocytopoiesis were studied using a microagar culture system which permits the simultaneous proliferation of human megakaryocytic progenitor cells (CFU-M) and T-lymphocytic colonies (CFU-TL). The proliferation of these colony types depends mainly on two factors: phytohaemagglutinin (PHA) and erythropoietin (EP). The direct addition of increasing PHA concentrations to the liquid overlayer resulted in a parallel increase of CFU-M and CFU-TL. If the T-lymphocytes were removed by an E-rosetting technique a marked diminution of CFU-M and CFU-TL numbers was observed. However, monocyte depletion resulted in a marked augmentation of CFU-M proliferation compared to unfractionated mononuclear cells. In order to confirm that the reduction of CFU-M proliferation observed after T-depletion was primarily mediated by the absence of T-lymphocytes, we have co-cultured different concentrations of previously removed autologous T-lymphocytes with a constant number of T-depleted bone marrow cells. A parallel increase of CFU-TL and CFU-M was found if 0.75 - 7.5 X 10(4) T-lymphocytes were added to the culture. In conclusion, our results indicate that activated T-lymphocytes augment proliferation of human bone marrow CFU-M and that monocytes are less important for the growth of megakaryocytic colonies.

Effects of thymocytes on marrow cells from normal and busulfan-treated mice

To study the effects of thymocytes on hemopoiesis, we 1) cocultured marrow cells with and without thymocytes in an erythroid burst (BFU-E) culture system, 2) injected marrow cells with and without thymocytes into supralethally irradiated mice to assay CFU-S, and 3) assessed the survival of supralethally irradiated mice transplanted with marrow cells with or without thymocytes. The marrow cells used were either from mice given six injections of busulfan and then permitted to rest for 2, 5, or 10 weeks or from mice treated similarly with the busulfan vehicle alone. Thymocytes did not alter spleen surface colony counts or survivorship in any of the test groups. Thymocytes did effect an increase in BFU-E cultured from marrow obtained from the vehicle-treated mice but not from marrow of busulfan-treated mice. Thus, in addition to decreasing the population of hemopoietic precursors in the marrow, busulfan alters the nature of the remaining early erythroid precursor cell population rendering it unresponsive to thymocytes in vitro.

An analysis of prognostic factors in preleukemia: interest of bone marrow scintigraphy

Simultaneous bone marrow scintigraphy with 99m Technetium colloids and 111 Indium transferrin was performed on 34 cases of preleukemic anemias and was shown to be of good prognostic value. Groups of different outcome were defined: for a normal and parallel uptake of the two markers, 90% of the patients died of acute leukemia; for a low Indium and high Technetium uptake, only 1 patient out of 15 died of leukemia (P less than 0.001). Standard clinical and hematological data were of no predictive value. Iron kinetic data and CFU/GM colony growth were correlated to the scintigraphic results. Taken together, these three kinetic parameters have a good sensitivity and specificity for the prognosis of preleukemic states.

Clonal growth of human megakaryocytic progenitor cells in a micro-agar culture system: simultaneous proliferation of megakaryocytic, granulocytic, and erythroid progenitor cells (CFU-M, CFU-C, BFU-E) and T-lymphocytic colonies (CFU-TL)

A simple and reproducible micro-agar culture technique for cloning human CFU-M is described. Human bone marrow mononuclear cells were suspended in agar and incubated for 12 days. Stimulation was provided by the direct addition of phytohemagglutinin-P (PHA-P), erythropoietin (Epo) and 2-mercaptoethanol (2-ME) to the liquid overlayer. A shift from BFU-E and CFU-C proliferation to CFU-M and CFU-TL was observed with increasing PHA concentrations. Under optimal conditions (PHA 50 micrograms, Epo 1.2 IU, 2-ME 2 x 10(-4) M, 1% purified BSA, 0.04% human transferrin, saturated with Fe C13) a linear relationship between colonies formed and plated cell number were observed. For the routine morphological analysis, the whole agar layers were stained using the Pappenheim method. For further characterization of CFU-M, cytochemical stainings and immunofluorescence tests with rabbit-antihuman factor VIII-related antigen were performed on the whole agar layers.

Lymphocyte characteristics and function in paroxysmal nocturnal haemoglobinuria

In six patients with paroxysmal nocturnal haemoglobinuria (PNH) lymphocyte studies were performed to investigate whether the observed immunological dysfunction could be ascribed to a defect in lymphocytes as result of the PNH characteristics, or to an imbalance between T-cell subsets. The PNH characteristics were studied by means of the effect of serum and acidified serum on Indium111-oxine labelled lymphocytes. No increase in release of Indium111-oxine was found when lymphocytes were exposed to acidified sera. Thus a complement mediated lymphocyte lysis in PNH could not be demonstrated. T-cells were defined by monoclonal antibodies, directed at total T-cells (OKT3), helper T cells (OKT4) and suppressor/cytotoxic T-cells (OKT8). In two of the six patients a decreased proportion of OKT3 cells was found, while a significantly depressed ratio of OKT4/OKT8 cells was present in the whole group. No obvious correlation was found between a functional assay - the concanavalin-A suppressor cell activity - and the ratio of OKT4/OKT8 positive cells. It is concluded that the PNH characteristics could not be demonstrated in the lymphocytes; and that the immunological dysregulation in PNH may be ascribed to an imbalance of T-cell subsets, while a decreased number of monocytes, defined by the monoclonal antibody OKM1, may contribute to this defect.

Susceptibility to merocyanine 540-mediated photosensitization: a differentiation marker on murine hematopoietic progenitor cells

Merocyanine 540 (MC 540) is an impermeant fluorescent dye that binds preferentially to fluidlike domains of the cell membrane. Photoexcitation of membrane-bound dye causes a breakdown of the normal permeability properties of the membrane and, eventually, cell death. We have used in vitro and in vivo clonal assays to determine the relative sensitivities of different classes of normal murine hematopoietic progenitor cells to MC 540-mediated photosensitization. Late erythroid progenitors (CFU-E) were the most sensitive cells, followed in order of decreasing sensitivity by early erythroid progenitors (BFU-E), megakaryocyte progenitors (CFU-Meg), day 7-spleen colony forming cells (day 7-CFU-S), granulocyte/macrophage progenitors (CFU-GM), and day 11-spleen colony forming cells (day 11-CFU-S). Bipotent progenitors of the granulocyte/macrophage lineage were more sensitive than unipotent macrophage progenitors but less sensitive than unipotent granulocyte progenitors. Progenitors giving rise to large granulocyte/macrophage colonies were more sensitive than progenitors giving rise to small colonies ("clusters"). We conclude that sensitivity to MC 540-mediated photosensitization is develop-mentally regulated and that differences occur even between the most closely related classes of progenitor cells. Our findings indicate the usefulness of MC 540 as a plasma membrane probe. They also support the contention that early and late-appearing spleen colonies are the progeny of two distinct classes of progenitor cells.

Fetal hemoglobin accumulation in vitro. Effect of adherent mononuclear cells

In clonal cultures of erythroid burst-forming units (BFU-E) obtained from blood, the accumulation of fetal and adult hemoglobins (Hb F and Hb A) was measured by radioligand immunoassay. Inclusion of adherent mononuclear cells in the culture promoted a striking increase in the relative amount of Hb F in each of 44 experiments with 14 donors. In two-thirds of the instances, this was accounted for by a selective increase in the absolute amount of Hb F. The differential effect on Hb F and Hb A accumulation was achieved without altering the maturity of the erythroid cells, their mean hemoglobin content, or the asynchrony of the production of the two hemoglobins. Virtually all bursts produced Hb F, and the population of BFU-E as a whole, rather than a selected subset, appeared to be the target of adherent cell action. When the adherent cells were excluded from the culture input, the base-line value of Hb F was reproducible for each donor over a period of several months, and correlated with the number of in vivo circulating F cells.

Switch from fetal to adult hemoglobin is associated with a change in progenitor cell population

To examine the switch from fetal to adult hemoglobin at the cellular level, erythroid progenitor cells from newborn infants and adults were cultured in methyl cellulose with erythropoietin. Individual erythroid colonies were labeled with [3H]leucine at various times, and globin synthesis patterns examined by gel electrophoresis and fluorography. The percent gamma- or beta-globin synthesis was determined from the total of gamma + beta, and the percent G gamma from the total of G gamma + A gamma. The nonparametric correlation coefficients of percent G gamma with percent gamma or beta were obtained. Each group of colonies at each time point was examined separately. In colonies from adult blood, the proportion of G gamma-synthesis did not correlate with the proportion of gamma-synthesis. Colonies from newborn blood fell into two groups. Those that developed from relatively mature progenitor cells, and were seen on day 14, showed a strong negative correlation of G gamma with beta-globin synthesis. However, those newborn colonies that developed from immature progenitors, and were seen later in culture (days 17 and 21), showed no correlation of G gamma with beta-synthesis. These findings are compatible with a clonal model for hemoglobin switching. Fetal progenitors, in which G gamma- and beta-syntheses are negatively correlated, are gradually replaced during ontogeny by adult progenitors. The adult progenitors produce more beta (less gamma), and the proportions of G gamma- and gamma- or beta-synthesis are not correlated.

Anemia associated with rheumatoid disease. Inverse correlation between erythropoiesis and both IgM and rheumatoid factor levels

Cell culture techniques were used to evaluate the number of erythroid colonies formed by circulating progenitor cells from 24 patients with rheumatoid disease and controls. A highly significant inverse correlation was demonstrated between erythroid colony counts and serum IgM and rheumatoid factor concentrations in the rheumatoid patients. The potential role of these factors in the pathogenesis of the anemia of rheumatoid disease is discussed.

Two different in vitro growth patterns for erythroid precursors in 18 patients with pure erythrocytosis

Growth patterns of marrow and blood erythroid progenitors were studied in 18 cases of pure erythrocytosis using different doses of erythropoietin. 8 cases demonstrated 'spontaneous' growth of CFU-E and blood BFU-E as observed in myeloproliferative disorders, but without an excess of circulating CFU-GM. 3 of these patients also had other symptoms of a pan-myelopathy. All these cases showed good sensitivity to 32P myelo-suppression. 10 cases demonstrated growth patterns of erythroid progenitors similar to those observed in normal subjects, except for an excess of blood BFU-E, which suggests an abnormality of homeostatic regulation. In 5 of these cases, myelo-suppression was not effective. It is suggested that a stem cell study could differentiate patients with pure erythrocytosis due to 'autonomous' abnormal stem cell growth from cases due to abnormal regulation factors, and that such a discrimination might be useful for the choice of therapy.

Lymphocyte dysfunction in congenital hypoplastic anemia

Congenital hypoplastic anemia (Diamond-Blackfan syndrome) is thought to involve the erythropoietic cell line alone. In this study, the evaluation of lymphocyte function in five patients with this syndrome revealed a number of abnormalities. Peripheral blood T lymphocyte percentages as assessed by monoclonal antibodies were decreased in three patients. T-helper/T-suppressor cell (OKT4:OKT8) ratios were almost unity in four of the five patients. We usually find a ratio of 2:1 in normal populations. Studies of lymphocyte-mediated suppression of lymphoproliferation demonstrated an inability to generate concanavalin A-induced suppressor cells in the same four patients and impaired prostaglandin-mediated suppression in two patients. Co-culture studies revealed a T lymphocyte-mediated suppression of erythropoiesis in a single patient, who also showed suppression of the mixed lymphocyte reaction. The four remaining patients showed no excessive suppressor effects either upon erythropoiesis or lymphoproliferation. These studies demonstrate that in congenital hypoplastic anemia, the cellular defect is not restricted to the erythroid progenitor cells, but extends to the lymphocytes.

Natural history of smouldering leukaemia

The natural history of 45 cases of smouldering leukaemia has been studied. Males and females were equally represented, with a median age of 60.5. The median survival of the whole group was only 20 months, but rare cases lived 10 years or longer. 38% developed acute leukaemia; the remainder usually died of the results of marrow failure. Although it was possible to divide these marrow dysplasias morphologically into 3 major subgroups (refractory anaemia with excess of myeloblasts, chronic myelomonocytic leukaemia and chronic erythraemic myelosis), several displayed transitional features. Many showed refractory macrocytosis at diagnosis. The survival of the 3 groups was similar, though patients with high monocyte counts tended to present with less anaemia and fared rather better than the others. Statistical analysis suggests that increasing age, severe anaemia, thrombocytopenia and hepatomegaly are associated with a poor prognosis. Chemotherapy, when attempted, was usually unsuccessful.

DR antigen positive monocyte-macrophages control granulocyte-macrophage colony-stimulating activity and burst promoting activity elaboration in man

To investigate the mechanisms that modulate granulocyte-macrophage colony-stimulating activity (GM-CSA) and burst promoting activity (BPA) elaboration, we studied human peripheral blood-derived monocyte-macrophage (M0) and T-lymphocyte (TL) interaction. Coincubation of live M0 with autologous TL at a 1:3 ratio in the presence of 1% phytohemagglutinin synergistically increased GM-CSA (6 of 6 experiments) and BPA (4 of 6 experiments) (P less than 0.002). Prior treatment of M0 with cycloheximide or actinomycin D significantly (P less than 0.002) diminished this M0's capacity to collaborate with TL. Mitomycin C treatment did not. Live M0 also enhanced TL-derived GM-CSA (P less than 0.002) and BPA (P less than 0.001). This enhancement was again compromised by prior cycloheximide or actinomycin D treatment, but not by mitomycin C treatment. Further experiments in which we blocked DR antigen on M0 membrane with monoclonal anti-DR antibodies suggested that M0 required their membrane DR antigen to collaborate with TL in elaborating GM-CSA and BPA.

Inhibitory effects of T cells on in vitro granulopoiesis, erythropoiesis, and immunoglobulin production in patients with aplastic anaemia

Coculture with mononuclear cells from aplastic anaemia (AA) patients and normal bone marrow cells produced different responses among 10 patients studied. In 4 cases, CFU-E growth was inhibited while in another 2 cases CFU-C growth was inhibited. To determine the suppressive effects of mononuclear cells from 6 AA patients, mononuclear cells were separated into T cells, B cells, and monocytes. It was found that T cells from 6 AA patients exhibited suppressive effects on normal CFU-C or CFU-E growth, whereas T cells from normal subjects enhanced the colony formation of CFU-C and CFU-E. On the other hand, monocytes fom 6 AA patients and normal subjects strongly suppressed normal CFU-C and CFU-E growth, although no significant difference was observed in both groups. The immune system in 10 AA patients was evaluated, 3 AA patients (Cases No. 1, 6, 8) exhibited an increase in Tr cells, impairment of PHA responsiveness, and low level of immunoglobulin. T cells from these patients showed a significant suppressive effect on PWM-induced differentiation of allogenic normal B cells. Thus, T cells from 3 out of 10 AA patients exhibited suppressor activities on both haematopoietic cell differentiation and B cell differentiation into plasma cells in vitro.

In vitro suppression of erythropoiesis by bone marrow adherent cells from some patients with fungal infection

We present evidence that alterations in marrow adherent cell (M phi) function may play a role in the suppression of erythropoiesis in some patients with fungal infection. Bone marrow (BM) cells from 12 normals and 10 patients with histoplasmosis were cultured in plasma clots before and after removal of M phi. BM from five patients (Group A) produced normal numbers of erythroid colonies (EC). In the remaining patients (Group B), smaller numbers of EC were detected (P less than 0.01). Removal of M phi from BM of normals and Group A patients resulted in decreased growth of EC. In contrast, M phi depletion of BM from patients in Group B resulted in greater EC formation (P less than 0.01). When normal M phi were admixed with normal or patients' BM-M phi, enhanced EC formation resulted. Whereas, at similar concentrations, M phi from group B patients caused inhibition of EC formation (P less than 0.005). The erythro-regulatory function of M phi, including the inhibitory action of patients' M phi, was mediated via a soluble agent(s) since media conditioned by M phi mimicked the action of these cells. Three patients in Group B were restudied 14 months after treatment with amphotericin B, when blood parameters had returned to normal. At this time, normal patterns of EC formation and M phi activity were observed.

Differential sensitivity of mouse hematopoietic stem cells to merocyanine 540

In vivo and in vitro clonal assays of immature mouse blood cells showed that different populations of hematopoietic progenitor cells differ considerably with respect to their sensitivity to photodynamic damages caused by the fluorescent dye Merocyanine 540. Late erythroid progenitors were the most sensitive cells followed in order of decreasing sensitivity by pluripotent stem cells, early erythroid progenitors, and granulocyte/macrophage progenitors. Only about 2%-4% of all nucleated marrow cells were stained with Merocyanine 540 which correlated well with current frequency estimates of progenitor cells in mouse bone marrow. Our findings indicate that the expression of Merocyanine binding sites is developmentally regulated and might, therefore, provide a useful molecular marker for blood cell differentiation and a basis for an effective purification of hematopoietic progenitor cells.

Erythroid colony growth in vitro from human peripheral blood null cells: evidence for regulation by T-lymphocytes and monocytes

Colony assays in methylcellulose of primitive erythroid precursors (BFU-E) were carried out from the null cell fraction of normal human peripheral blood lymphoid cells. There was little proliferation or maturation of BFU-E as assessed by both the number and the size of colonies formed, when null cells alone were cultured. Culture of null cells with up to 8 X 10(5) autologous T-lymphocytes per ml led to considerable stimulation of colony growth and maturation. Culture of null cells with peripheral blood monocytes also resulted in the indication of BFU-E growth, although the response was inferior to the seen with T-cells. Co-culture of null cells together with both T-cells and monocytes resulted in a uniformly greater response than with either alone, and this was shown to be due to a positive interaction between these two cell types.

Human erythroid burst-forming units. Growth in vitro is dependent on monocytes, but not T lymphocytes

The roles of monocytes and T lymphocytes in the regulation of human peripheral blood erythroid burst-forming units (BFU-E) were studied in erythropoietin-containing plasma clot cultures of subpopulations of human blood mononuclear cells. BFU-E growth was decreased significantly after depletion of monocytes alone (mean 11% of expected, range 0 to 42% of expected) or depletion of both monocytes and T cells (mean 6.5% of expected, range 0.5 to 12% of expected) from mononuclear cells. T cell depletion did not impair BFU-E growth in vitro. Using 10(5) monocyte- and T lymphocyte-depleted mononuclear cells as target cells (less than 1% monocytes, less than 5% T cells), BFU-E growth was restored to 40% of expected by addition of 10(4) monocytes, and to 96% of expected by 10(5) monocytes alone. Addition of as many as 2 X 10(5) T cells but no monocytes resulted in stimulation to only 34% of expected BFU-E growth. Addition of 2 X 10(4) T cells, which alone did not affect BFU-E growth, could augment significantly the stimulatory effect of 5-20 X 10(3) monocytes on BFU-E growth. Thus, monocytes alone appear to be capable of stimulating BFU-E growth in vitro in the presence of erythropoietin. T cells also may make small quantities of BFU-E stimulators. However, it seems more likely that the most important role of T lymphocytes in BFU-E regulation in vitro is a result of interactions with monocytes and augmentation of monocyte production of stimulators of BFU-E growth.

Monoclonal antibody assessments of T cell interactions in erythropoietin studies

The contemporary application of clonal assay techniques has greatly expanded our knowledge of the regulation of hematopoiesis. Our efforts have been directed toward the investigation of non-erythropoietin-mediated regulation of human erythropoiesis in the form of cell-to-cell interaction between mature T cells and erythroid progenitors. Our data indicate that three such progenitors, the early marrow erythrocyte precursor BFU-E, the more mature marrow erythrocyte precursor, CFU-E and the peripheral blood BFU-E, each exhibit totally different requirements for their colony expression in culture, with respect to the absence of erythropoietin and the presence of mature T cells or their products. The capacity of erythroid progenitors to withstand incubation in the absence of erythropoietin appears to be a characteristic of immature rather than mature erythroid progenitors. Furthermore, use of OKT3 antibody depletion techniques shows that peripheral blood-derived BFU-E appear to depend upon mature lymphocytes or T cell-conditioned medium for erythropoietin-stimulated differentiation while bone marrow BFU-E and CFU-E have no requirement for mature T cells to produce erythropoietin-dependent maturation. Our results, plus a vast array of data provided by other investigators in the field, are integrated into a proposed framework for further investigation of T cell induction of erythropoietin-dependent erythroid differentiation aimed at more specifically identifying the inducer cell subset(s) in that system.

Autoimmune pancytopenia. Lymphocyte inhibition of autologous but not allogenic bone marrow growth in vitro

A patient with autoimmune renal failure, cavitary lung lesions and arthritis experienced pancytopenia while prednisone therapy was being tapered. Utilizing semisolid culture techniques, a population of nonadherent peripheral blood mononuclear cells was demonstrated, which inhibited autologous but not allogeneic bone marrow erythroid colony-forming units (CFU-E) and myeloid colony-forming units (CFU-c) in vitro. No inhibition of CFU-E or CFU-c colony formation was seen when patient's serum or immunoglobulin G (IgG) was added to cultures. Reinstitution of prednisone therapy resulted in normalization of peripheral blood counts, which was accompanied by the loss of the hemopoietic inhibitor cell activity in the patient's peripheral blood. These results demonstrate the need for testing autologous marrow samples when looking for possible immune-mediated inhibition of hematopoiesis.

Bone-marrow microenvironment defect in congenital hypoplastic anemia

An abnormal bone-marrow microenvironment has been thought to have an important role in the pathogenesis of aplastic and hypoplastic anemia in some patients, but direct evidence of such a defect has not been found. We have investigated the pathogenesis of chronic anemia in a young woman. Her bone-marrow cells, obtained by means of aspiration, showed exuberant erythroid growth in methylcellulose despite marked erythroid hypoplasia. The erythroid nature of the colony growth was further confirmed through measurement of heme synthesis and messenger RNA-globin accumulation in a liquid-culture system. In contrast, when whole bone fragments were similarly cultured, no appreciable hemoglobin synthesis was observed. The experimental evidence suggests that, in this patient, hypoplastic anemia resulted from an unfavorable microenvironment.

Synthesis and release of erythroid colony- and burst-potentiating activities by purified populations of murine peritoneal macrophages

We investigated the effects of murine resident peritoneal macrophages on the in vitro proliferation of erythropoietin (Ep)-sensitive committed precursors colony-forming unit-erythroid (CFU-E) and burst-forming unit-erythroid (BFU-E) with a two-layer cloning system of methylcellulose and semisolid agar. The addition of increasing numbers of macrophages to the agar underlayer resulted in a progressive increase in the numbers of both CFU-E and BFU-E that proliferated in the presence of Ep. CFU-E, but not BFU-E, proliferated to form colonies in the absence Qf exogenously added Ep, and this proliferation was enhanced in a dose-dependent fashion by the presence of macrophages in the underlayer. The enhancing effects of a given number of macrophages and a given concentration of Ep were greater than the sum of the individual effects of macrophages and Ep alone. This erythropoietic syngerism was more evident with BFU-E because burst formation was not seen in the absence of exogenously added Ep. Macrophage underlayers stimulated three to five times the number of erythroid bursts seen with Ep alone. Cell-free agar underlayers or agar underlayers prepared with nonadherent peritoneal cells or unseparated cells from thymus, lymph node, or spleen failed to augment Ep- dependent erythroid colony formation. No enhancement of CFU-E or BFU-E was demonstrable after depletion ofadherent cells from peritoneal cell suspensions by passage over columns of Sephadex G-10. Analysis by sedimentation velocity of peritoneal cells confirmed that the cells responsible for elaborating the erythroid-enhancing activity were macrophages on the basis of morphologic, histochemical, and functional criteria. Serum- free media conditioned by macrophages in the absence of Ep contained the erythroid-enhancing activities, which indicated that Ep was not required for the elaboration of these diffusible substances. These studies indicate that although macrophages are not obligate for the growth of erythroid precursors, they serve as an important source of diffusible factors that reduce the in vitro requirement for Ep.

Differentiation of subpopulations of human and murine hemopoietic stem cells by hypotonic lysis

Both human and mouse bone marrow contain subpopulations of hemopoietic stem cells that greatly vary in their resistance to water exposure: The cells forming erythroid colonies or bursts in methyl cellulose in vitro are most sensitive to hypotonic conditions and are destroyed within 60 s in the hypotonic milieu. The murine pluripotent stem cells assayed by the spleen colony technique, as well as both murine and human myeloid stem cells assayed by the plasma clot diffusion chamber technique, displayed intermediate sensitivity and were nearly completely eliminated by 120 s of exposure to water. Both human and mouse bone marrow stem cells producing myeloid colonies in agar are most resistant to hypotonic conditions. The addition of monocyte-macrophages and lymphoid cells to water-exposed mouse bone marrow cell populations to compensate for losses did not restore either erythroid or myeloid colony formation.

Monocytes do not inhibit peripheral blood erythroid burst forming unit colony formation

To determine a possible role of peripheral blood monocytes in erythroid differentiation, various fractions of peripheral blood mononuclear cells were prepared from normal volunteers. The fractions contained 3-95% monocytes. These freshly prepared monocytes did not inhibit erythroid burst forming unit expression in plasma clot erythroid colony culture. Null cell preparations contaminated by up to 84% monocytes expressed erythroid burst forming unit colony formation when either T lymphocytes or T-cell conditioned medium was added. These results indicate that certain peripheral blood null cells engage the program of erythroid differentiation in the presence of T cells and erythropoietin. Monocytes do not inhibit this engagement.