A 29 000 molecular weight fraction from fetal rat liver adhering cells that cooperates with erythropoietin in stimulating the growth of erythroid progenitors

Summary The erythroid-potentiating effects of a protein fraction produced by 20-day rat fetal liver-adhering cells are studied. Partial purification by gel filtration gave an active fraction (apparent molecular weight = 29×10³) that significantly increased the erythroid colony counts (CFUe and late BFUe) in cultures of liver cell fractions depleted of adhering cells at both limiting and saturating concentration of recombinant human erythropoietin. The sensitivity of CFUe and BFUe to erythropoietin was increased by the activator.     

A novel mechanism of soluble guanylate cyclase stimulation: time-dependent activation by bacterial lipopolysaccharide in rat fetal spleen cells

Small amounts of bacterial lipopolysaccharide (LPS) greatly increase cGMP levels in short term cultures of rat fetal liver and spleen cells in a dose and time dependent manner. To determine the role of guanylate cyclase in this response, a series of experiments was undertaken using either intact or broken fetal spleen cells, the most sensitive tissue evaluated to date. The phosphodiesterase inhibitor, 1-methyl-3-isobutyl-xanthine, potentiated the LPS-cGMP effect in cultures of these cells even at maximal doses of LPS. Moreover, after incubation of intact cells with LPS for 4 h, soluble guanylate cyclase (EC 4.6.1.2) activity was increased 2-fold, whereas particulate activity was unchanged. This increase in soluble activity was proportional to the dose of LPS, was synchronous with the elevation of cGMP levels, and was not associated with any change in cGMP-phosphodiesterase (EC 3.1.4.17) activity. In contrast to intact cells, neither total nor soluble guanylate cyclase activity was increased by the addition of LPS to spleen cell whole sonicate or cytosol for various times from 10 min to 3.5 h. These results suggest that the LPS-cGMP response is due to a persistent indirect stimulation of soluble guanylate cyclase activity that is both dose and time dependent.     

The production of factors regulating the proliferation of haemopoietic spleen colony-forming cells by bone marrow macrophages

Media conditioned by normal murine bone marrow cells contain an inhibitor of haemopoietic spleen colony-forming cell proliferation that is concentrated in a nominal 50-100K fraction. Media conditioned by regenerating marrow cells contain a proliferation-stimulatory activity that is concentrated in a nominal 30-50K fraction. Cell separation experiments demonstrated that the activities are produced by adherent, phagocytic, radioresistant, Thy 1.2- Fc+, F4/80+ cells. Cultured macrophages, obtained from long-term marrow cultures or derived from progenitor cells in methyl cellulose cultures are also capable of producing inhibitory and stimulatory activities. The results are consistent with macrophages being an important source of stem cell proliferation regulators in the bone marrow.     

The production of factors regulating the proliferation of haemopoietic spleen colony-forming cells by bone marrow macrophages

Abstract. Media conditioned by normal murine bone marrow cells contain an inhibitor of haemopoietic spleen colony-forming cell proliferation that is concentrated in a nominal 50-100K fraction. Media conditioned by regenerating marrow cells contain a proliferation-stimulatory activity that is concentrated in a nominal 30-50K fraction. Cell separation experiments demonstrated that the activities are produced by adherent, phagocytic, radioresistant, Thy 1.2^- Fc⁺, F4/80⁺ cells. Cultured macrophages, obtained from long-term marrow cultures or derived from progenitor cells in methyl cellulose cultures are also capable of producing inhibitory and stimulatory activities. The results are consistent with macrophages being an important source of stem cell proliferation regulators in the bone marrow.     

A novel mechanism of soluble guanylate cyclase stimulation: time-dependent activation by bacterial lipopolysaccharide in rat fetal spleen cells

Small amounts of bacterial lipopolysaccharide (LPS) greatly increase cGMP levels in short term cultures of rat fetal liver and spleen cells in a dose and time dependent manner. To determine the role of guanylate cyclase in this response, a series of experiments was undertaken using either intact or broken fetal spleen cells, the most sensitive tissue evaluated to date. The phosphodiesterase inhibitor, 1-methyl-3-isobutylxanthine, potentiated the LPS-cGMP effect in cultures of these cells even at maximal doses of LPS. Moreover, after incubation of intact cells with LPS for 4 h, soluble guanylate cyclase (EC 4.6.1.2) activity was increased 2-fold, whereas particulate activity was unchanged. This increase in soluble activity was proportional to the dose of LPS, was synchronous with the elevation of cGMP levels, and was not associated with any change in cGMP-phosphodiesterase (EC 3.1.4.17) activity. In contrast to intact cells, neither total nor soluble guanylate cyclase activity was increased by the addition of LPS to spleen cells, neither total cytosol for various times from 10 min to 3.5 h. These results suggest that the LPS-cGMP response is due to a persistent indirect stimulation of soluble guanylate cyclase activity that is both dose and time dependent.     

Further studies on generation of macrophages in in vitro cultures of mouse spleen cells and its inhibition by the capsular polysaccharide of Klebsiella pneumoniae.

Although the number of macrophages detected in cultures of mouse spleen cells at the start of the culture was very small, it markedly increased during further incubation. Macrophages were generated not only from the glass-adherent cell fraction of spleen cells, but also from the nonadherent cell fraction obtained after removal of adherent cells either by incubating in glass petri dishes or by passing through a glass bead column. The generation of macrophages from the nonadherent cell fraction occurred even when it was separated as late as 48 hr after the start of the culture. The phagocytic activity of macrophages newly generated from the nonadherent cell fraction was relatively weak, but it was activated during further incubation. Based on these results, the maturation process of macrophages can be divided into at least the following four stages; glass-nonadherent nonphagocytic precursor cells, glass-adherent nonphagocytic precursor cells, immature macrophages with low phagocytic activity, and mature macrophages with full phagocytic activity. The addition of the capsular polysaccharide of Klebsiella pneumoniae (CPS-K) to cultures of spleen cells markedly suppressed the generation of macrophages. The suppressive effect of CPS-K depended on its dosage, and the minimum concentration of CPS-K showing a definite effect was 0.05 μg/ml. CPS-K inhibited further generation of macrophages in either the nonadherent or adherent cell fraction at any time after the start of the culture. The suppressive effect of CPS-K on the generation of macrophages could not be reversed by simple washing of spleen cells which had been kept in contact with CPS-K for 3 hr. There was no evidence which showed that CPS-K exhibited direct cytotoxic effects on spleen cells in the culture.     

Development of erythroid colony-forming cells in rat fetal spleen: apparent lack of sensitivity to an in vivo corticosteroid excess as compared to fetal liver

The development of splenic erythroid colony-forming cells from rat embryos in the last 4 days of intrauterine life was examined after 2 and 7 days in a methylcellulose culture system. The number of 2- and 7-day erythroid colonies decreased sharply between, respectively, days 20 and 21 of gestation and days 19 and 20. Concomitantly, a maturation of proerythroblasts and basophilic erythroblasts to mature erythroblasts was detected on smears of splenic cellular suspensions. The effect of a corticosteroid excess induced by a maternal laparotomy was tested on spleen and liver cultures from the same control or experimental fetuses. The ratio of the number of 2-day to the number of 7-day erythroid colonies did not differ in experimental and control splenic cultures, but in liver cultures was significantly lower at days 19 and 20 in experimental than in control cultures.     

Nature of cells forming erythroid colonies in agar after stimulation by spleen conditioned medium

Erythroid colony formation in agar cultures of CBA cells was stimulated by the addition of pokeweed mitogen-stimulated C57BL spleen conditioned medium. Both 48-hour colonies ("48-hour benzidine-positive aggregates") and day 7 large burst or unicentric erythroid colonies ("erythroid colonies") developed, together with many neutrophil and/or macrophage colonies. In CBA mice, the cells forming erythroid colonies occurred with maximum frequency (650/10(5) cells) in 10- to 11-day-old yolk sac and fetal liver but were present also in fetal blood, spleen and bone marrow. The frequency of these cells fell sharply with increasing age and only occasional cells (2/10(5) cells) were present in adult marrow. A marked strain variation was noted, CBA mice having the highest levels of erythroid colony-forming cells. The erythroid colony-forming cells in 12-day CBA fetal liver were radiosensitive (DO 110-125 rads), mainly in cycle and were non-adherent, light density, cells sedimenting with a peak velocity of 6-9 mm/hr. These properties are similar to those of other hemopoietic progenitor cells in fetal tissues. The relationship of these apparently erythropoietin-independent erythroid colony-forming cells to those forming similar colonies after stimulation by erythropoietin remains to be determined.     

Suppression of endotoxin mitogenicity of spleen cells by lipoxygenase inhibitors and its reversal by 13-hydroxyoctadecadienoic acid

Abstract The effects of several inhibitors of lipoxygenases were investigated in murine spleen cell cultures activated with endotoxin (lipopolysaccharide) It was found that these inhibitors interfere with the proliferative response of the cultures. Indomethacin, a specific cyclooxygenase inhibitor, had no such effect. Endotoxin induced the synthesis of tumour necrosis factor α in spleen cells which was prevented by treatment with a lipoxygenase inhibitor. The inhibition of the mitogenic effect of endotoxin could be reversed by addition of 13-hydroxyoctadecadienoic acid. This was not the case with leukotriene B₄ and C₄ or 15-hydroxyeicosatetraenoic acid. In contrast, these substances had inhibitory effects on the mitogenicity of spleen cells. It is suggested that 13-hydroxyoctadecadienoic acid is involved in the development of the mitogenic reaction, possibly on the level of tumour necrosis factor α production of macrophages present in the cultures.     

Expression of congenital defects in the haemopoietic micro-environment. Erythroid and granulocyte-macrophage progenitor cells in pre-natal 'steel' (Slj/Slj) anaemic mice.

The erythropietin sensitivities of dissociated cell cultures and explanted fragments of fetal livers of congenitally anaemic Slj/Slj mice, and their normal littermates, have been compared. The erythropoietin responsiveness of Slj/Slj foetal liver cells is deficient in both types of culture. The maximum liver complement of erythroid colony forming cells (CFUe) occurs on the 16th day of development when 'normal' livers contain approximately 6 X 10(5) erythroid colony forming cells/liver. In Slj/Slj fetuses the maximum reached is only 1 X 10(5). Granulocyte-macrophage colony forming cells (CFUc) in Slj/Slj fetal livers are also reduced to approximately 60% of normal numbers. Erythroid colony forming cells are also reduced in the spleen and femoral bone marrow of Slj/Slj mice in the 2-3 days preceding birth. Granulocyte-macrophage colony forming cells are rare in the femoral marrow of pre-natal Slj/Slj mice, but their production in the Slj/Slj pre-natal spleen appears unaffected.     

Immunosuppression by spleen cells from Moloney leukemia. Comparison of the suppressive effect on antibody response and on mitogen-induced response.

The inhibitory effect of cells from leukemic spleens on the immune functions of normal lymphocytes was studied. Suppressor cells were obtained as the nonadherent fraction (NA) from splenic tumors of mice infected with MuLV-Moloney. This fraction (NA MuLV- M) contained less than 10% membrane Ig-positive (Ig+) cells, 45 to 60% theta-positive cells (theta+) and 40 to 50% naught cells (theta-, Ig-). Similarly prepared fractions from normal control spleens (NAc) containing 75 to 90% theta+cells and less than 10% Ig+ and naught cells were utilized in control cultures. Addition of the NA MuLV- M cells into cultures (Marbrook system) of normal spleen cells with sheep red blood cells suppressed the specific antibody response determined by the number of hemolytic plaque forming cells (PFC). The PFC response was significantly suppressed at a suppressor cell to responder cell ratio of 1:100, and was completely abolished at a ratio of 1:10 or higher. The control NAc fraction showed some inhibitory effect only at high suppressor to responder ratios (1:2 or 1:1). In contrast, the suppressive effect of NAMuLV-M on mitogen-induced 3H-thymidine incorporation in normal B and T cells was much weaker. Very little, if any, suppression occurred at the ratio of 1:100 or 1:10, however, about 50% decrease in DNA synthesis was observed at the ratio 1:2 or 1:1. On the basis of this differential suppressive effect, it is suggested that leukemic spleen cells can suppress the function of immunocompetent cells by more than one mechanism.     

Spontaneous proliferation in unfractionated spleen cell cultures: autologous mixed-lymphocyte reactions (AMLR) which can be differentially regulated by prostaglandins and lymphokines

The present studies were undertaken to define the contribution of the autologous or syngeneic mixed-leukocyte reactions (AMLR/SMLR) to the cellular proliferation observed in unfractionated spleen cell cultures. Proliferation was studied in whole, untreated 6-day murine spleen cell cultures supplemented with syngeneic serum. These cultures exhibited relatively low but significant levels of cellular proliferation as measured by uptake of radioactive thymidine ([3H]TdR). Treatment of spleen cells with monoclonal anti-Thy 1.2 antibody and complement before culture, the addition of specific anti-I-A monoclonal antibodies to the cultures or removal of Ia+ adherent cells before initiation of culture all inhibited the proliferative response significantly. Thus, the autologous proliferation of untreated and unfractionated spleen cells manifests the main characteristics of the AMLR/SMLR, namely, its dependence on T (responder) and Ia+ (stimulator) cells and specific inhibition by anti-I-A antibodies. A marked augmentation in cellular proliferation was observed in unfractionated spleen cell cultures treated for the initial 24 hr of culture with 5 X 10(-6) M indomethacin, an inhibitor of prostaglandin synthesis. Conversely, the addition of 7 X 10(-9) M prostaglandin E1 (PGE1) to these cultures depressed cellular proliferation. This suppression of autologous splenic cell proliferation induced by PGE1 could be partially reversed by the addition of concanavalin A-induced lymphokine (LK) preparations early in the culture. These findings indicate that (a) the proliferation of unfractionated spleen cell cultures occurring in the absence of exogenous stimulatory signals is due largely to an ongoing AMLR, and (b) biologically active mediators with opposing influences, namely, prostaglandins and immunostimulatory LK, participate in the regulation of the AMLR.     

Preferential distribution of group-II-like phospholipase A2 in mononuclear phagocytic cells in rat spleen and liver

The localization of calcium-dependent phospholipase A2, (PLA2) immunochemically closely related to the enzyme of the viperid and crotalid type (group II), in cells isolated from rat spleen and liver was examined using a polyclonal antibody directed against rat spleen group II, PLA2 (PLA2M). In isolated spleen cells, the monocyte/macrophage fraction had the highest PLA2 activity (1.28 +/- 0.35.min-1.10(6) cells-1) which was almost completely inhibited by the anti-PLA2M antibody. An immunoblot analysis confirmed the presence of the enzyme in this fraction. An immunocytochemical study revealed that the PLA2 was present in spleen macrophages. In the isolated liver cells, Kupffer cells (0.92 +/- 0.22 nmol.min-1.10(6) cells-1) contained higher anti-PLA2M-antibody-inhibitable PLA2 activity than parenchymal cells (0.26 +/- 0.06.min-1.10(6) cells-1). The immunocytochemical study showed that cells immunopositive with anti PLA2M antibody were Kupffer cells. These results suggest that the mononuclear phagocytic cells in rat spleen and liver have relatively high activity of group-II-like PLA2. Subcellular distribution patterns of the anti-PLA2M-antibody-inhibitable phospholipase A2 activity in different cell populations from spleen and liver were compared. A mode of the distribution of the enzyme in the spleen macrophages was essentially similar to that in the spleen lymphocytes. The distribution in Kupffer cells was similar to that in parenchymal cells.     

Induction of nitrite production in mouse spleen cells by immunization.

Changes of nitrite production in mouse spleen cells of in vitro secondary antibody response were investigated. Mouse spleen cells immunized with gamma globulin fraction of rat serum produced nitrite 3 days after in vitro challenging with the same antigen. Nitrite production of rabbit IgG-challenged spleen cells was found to be about 2.9-times higher than that of spleen cells primed with the gamma globulin fraction of rat serum. Nitrite production in this system was completely suppressed by T cell depletion (99.7% inhibition). Furthermore, nitrite production in these cells significantly decreased by addition of anti-interferon gamma antibody (62.9% inhibition). These data indicate that nitrite production in antigen-immunized spleen cells is affected with the immunogenicity of an antigen and regulated by T cells, especially interferon (IFN) gamma.     

Functional hypersplenism in mice induced by adoptive transfer of syngeneic spleen cells

Investigation of peripheral blood cell count alterations in cases with hypersplenism, and an understanding of the relationship between splenic function and hematopoietic cell production require suitable experimental animal models. Previously described methods are either traumatic or require surgical intervention. We suggest a relatively simple method for achievement of a state mimicking hypersplenism in mice by intraperitoneal inoculation of syngeneic spleen cells. Mice were inoculated intraperitoneally with 3 x 10(7) splenocytes suspended in 0.3 ml phosphate buffered saline (PBS). After 2 months, the inoculated animals showed a progressive decrease in the peripheral white blood cell (WBC) counts and hyperplastic bone marrow that persisted until the experimental end point (7 months). Five days after inoculation of splenocytes stained with carboxy-fluorescein diacetate succinimidyl ester (CFSE), the majority of the stained cells was present in the peritoneal cavity (33%) and in the liver (13%), whereas the percentage of stained cells in the peripheral blood and the spleen cell suspension was negligible. The mitogen response of the peripheral blood mononuclear cells (PBMC) from treated mice to concanavalin A (Con A) remained unaltered. Splenocyte-inoculated mice that were further splenectomized did not show leukocytosis after splenectomy, as was observed in animals in which the spleen was removed without any pretreatment. The lack of any signs of discomfort in animals from the study group, in comparison with the visibly ill appearance and even death of mice in which hypersplenism was achieved by repeated injections of methylcellulose (MC), which served as controls, favors the convenience of the method.     

Formation of benzo[a]pyrene-DNA adducts by microsomal enzymes: comparison of maternal and fetal liver, fetal hematopoietic cells and placenta

The formation of benzo[a]pyrene (BP)-DNA adducts was studied in vitro in the presence of microsomes prepared from the isolated labyrinth zone of the rat placenta, the hematopoietic erythroblast cells of the fetal liver, the fetal liver, as well as the maternal liver. Pregnant rats received beta-naphthoflavone (beta NF; 15 mg/kg, i.p.) on day 17 gestation. One day later, placentae, fetal and maternal livers were obtained and hematopoietic erythroblast cells were separated from hepatocytes in the fetal livers. The respective microsomal fractions were incubated in the presence of calf thymus DNA, NADPH-regenerating system and [3H]BP (300 microCi) at 37 degrees C for 30 min. Following beta NF pretreatment, the levels of covalent binding (pmol/mg DNA/mg microsomal protein) for maternal liver, fetal liver, placenta and erythroblast cells were: 28.4, 2.4, 0.31 and 3.9, respectively, with the hematopoietic erythroblast cells being the most active among fetal tissue preparations. The extent of transplacental induction compared to control was greatest in the hematopoietic cells (18-fold) followed by fetal liver (16-fold) and labyrinth zone (5-fold). Further experiments characterized the BP-DNA adducts formed by induced microsomes. DNA was isolated, purified and digested sequentially with DNase I, snake venom phosphodiesterase type II and alkaline phosphatase type III. The deoxynucleoside-BP adducts were purified on a Sephadex LH-20 column and then separated on HPLC and the adducts were quantitated radiometrically. Seven distinct adducts were separated on HPLC and named A-G in order of elution. Adduct B was prominent in all preparations (22-55% total radioactivity). The adduct profile and retention time for peak B is similar to that reported for the adduct formed by microsomal activation of 9-hydroxy BP. Peak D constituted a major fraction (19%) in maternal liver profiles in comparison with the three fetal tissue preparations (8%). In subsequent experiments, peak D was shown to be derived from reaction of (+/-)7 beta,8 alpha-dihydroxy-9 alpha,10 alpha-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (BPDE) with DNA. Peak C was unique to erythroblast cell and labyrinth profiles, while peak G was specific for maternal liver and fetal liver profiles. These results demonstrate that fetal liver and its hematopoietic cells are significant sites of BP bioactivation which may contribute to the fetal toxicity of polyaromatic hydrocarbons.     

Analysis of the sinusoidal endothelial cell organization during the developmental stage of the fetal rat liver using a sinusoidal endothelial cell specific antibody, SE-1

The sinusoid organization during the development of fetal rat livers was studied using a SE-1 antibody, which we have previously established as a specific monoclonal antibody against rat sinusoidal endothelial cell (SEC). Expression and localization of the SE-1 antigen in the liver tissues of 13- to 21-day-old fetuses were immunofluorescently and immunoelectron microscopically examined. The first positive fluorescence was observed in the immature liver of 15-day-old fetuses. The initial positive staining was randomly distributed in the liver parenchyma and showed no direct relation to the large vessels which may be derived from the fetal vitelline veins. The positive linear staining increased in number and connected with each other during the course of development. The SE-1 staining pattern and the sinusoidal arrangement became similar to those of the adult liver after 20th day of gestation. Immunoelectron microscopically, the immature SEC showed a weak positive reaction for the SE-1 antigen at their membrane and was observed together with immature hepatocytes and hematopoietic cells in the 15-day-old fetal liver. Along with the liver development, SEC formed a sinusoid structure closely associated with hepatocytes and came to strongly express the SE-1 antigen. These results indicate that the organization of the hepatic sinusoid may start at around 15th day of the gestation and occurs randomly in the fetal liver parenchyma. It is also suggested that the expression of SE-1 antigen is possibly regulated by the intimate association with hepatocytes.     

Alternative induction of alpha/beta interferons and gamma interferon by listeria monocytogenes in mouse spleen cell cultures

Production of interferon (IFN) by Listeria monocytogenes (LM) in nonimmunized mouse spleen cell cultures was studied. IFN-gamma defined by virtue of its acid stability and antigenicity was produced in spleen cell cultures obtained from ddY mice, C57BL/6 mice, and BALB/c mice in response to heat-killed (HK) LM within 24 hr. On the other hand, production of IFN-alpha/beta was demonstrated in spleen cell cultures obtained from one of four nude mice (BALB/c, nu/nu). Therefore, it is important to know the reason why the spleen cells of mice other than nude mice did produce only IFN-gamma, but did not produce IFN-alpha/beta in response to HK-LM. Spleen cells obtained from ddY mice were fractionated, and the cellular source for IFN production of either IFN-alpha/beta or IFN-gamma induced by HK-LM was investigated. IFN-gamma was produced only by a mixture of T lymphocytes (nylon wool-nonadherent, Thy-1-positive cells) and macrophages by HK-LM. Neither T lymphocytes nor macrophages alone produced IFN by HK-LM. Macrophage-depleted spleen cells produced neither IFN-gamma nor IFN-alpha/beta, but these cells acquired the ability to produce IFN-alpha/beta, not IFN-gamma, only when they had been treated with IFN-alpha/beta. A possible mechanism of both IFN-gamma and IFN-alpha/beta induction by Listeria in mouse spleen cell cultures is discussed.     

An in vitro model for induction of immunologic unresponsiveness to turkey gamma-globulin in primed mouse spleen cells, II. Antigen-specific suppressor cells.

Unresponsiveness induced to turkey γ-globulin (TGG) in cultures of TGG-primed spleen cells by incubation with high concentrations of soluble TGG (sTGG) was shown to involve a state of active suppression. Upon transfer to secondary cultures of primed spleen cells stimulated with an optimal dose of TGG-conjugated erythrocytes, such tolerant spleen cells were able to actively inhibit a secondary plaque-forming cell response to TGG in these cultures. Almost complete inhibition was observed with a tolerant cell to primed cell ratio of as low as 0.1. The suppression was antigen specific in that tolerant spleen cells which were suppressive for the secondary TGG response were unable to inhibit a primary response to sheep erythrocytes. T cells were shown to be required for the suppressor effect, in that (i) suppressor activity could be removed by complement-mediated lysis with an anti-Thy 1.2 antiserum and (ii) suppressor activity was retained in the effluent fraction after passage of suppressor spleen cells over a nylon wool column. Induction of the T-cell suppressor activity was found to be associated with a loss of T-cell helper activity within the TGG-pulsed cell population. The presence of adherent cells was not required for induction of suppressor activity. Furthermore, the suppressor effect was found to be resistant to 1000 R of γ irradiation.     

Effect of an anti-murine transferrin receptor-ricin A conjugate on bone marrow stem and progenitor cells treated in vitro

A monoclonal antibody with specificity for murine transferrin receptor was conjugated with the toxic A subunit of ricin. The dose range, specificity, and kinetics of inhibition of protein synthesis of the conjugate were determined on the murine T-lymphoma cell line, BW5147. When toxin was present throughout the period of culture, in vitro myeloid (CFUc) and erythroid (CFUe and BFUe) bone marrow colonies were inhibited by doses of conjugate comparable to those that inhibit protein synthesis in murine cell lines (IC50 of 5 X 10(-11)M). Bone marrow exposed briefly (30 min to 6 h) to anti-transferrin receptor antibody-ricin A conjugate was assayed for myeloid (CFUc) and erythroid (CFUe and BFUe) progenitors in vitro and for in vivo spleen colony formation (CFUs). Only CFUe were depleted by this pulse exposure, consistent with the higher frequency of proliferating cells and transferrin receptor expression in the CFUe population relative to other progenitors.