Rapid, B lymphoid-restricted engraftment mediated by a primitive bone marrow subpopulation

Utilizing multiparameter flow cytometry, we have defined a subset of bone marrow cells containing lymphoid-restricted differentiation potential after i.v. transplantation. Bone marrow cells characterized by expression of the Sca-1 and c-kit Ags and lacking Ags of differentiating lineages were segregated into subsets based on allele-specific Thy-1.1 Ag expression. Although hematopoietic stem cells were recovered in the Thy-1.1low subset as previously described, the Thy-1.1neg subset consisted of progenitor cells that preferentially reconstituted the B lymphocyte lineage after i.v. transplantation. Recipients of Thy-1.1neg cells did not survive beyond 30 days, presumably due to the failure of erythroid and platelet lineages to recover after transplants. Thy-1.1neg cells predominantly reconstituted the bone marrow and peripheral blood of lethally irradiated recipients with B lineage cells within 2 weeks, although a low frequency of myeloid lineage cells was also detected. In contrast, myeloid progenitors outnumbered lymphoid progenitors when the Thy-1.1neg population was assayed in culture. When Thy-1. 1low stem cells were rigorously excluded from the Thy-1.1neg subset, reconstitution of T lymphocytes was rarely observed in peripheral blood after i.v. transplantation. Competitive repopulation studies showed that the B lymphoid reconstitution derived from Thy-1.1neg cells was not sustained over a 20-wk period. Therefore, the Thy-1. 1neg population defined in these studies includes transplantable, non-self-renewing B lymphocyte progenitor cells.     

Deceptive multilineage reconstitution analysis of mice transplanted with hemopoietic stem cells, and implications for assessment of stem cell numbers and lineage potentials

Hemopoietic stem cells (HSC) are identified through their unique ability, at the single cell level, to long-term reconstitute all blood cell lineages. Sustained myeloid reconstitution is considered the hallmark of HSC, because myeloid progenitors and their progeny have very short half-lives. Here we demonstrate that the established practice of relying on RB6-8C5 as a myeloid specific Ab can result in overestimation of HSC frequencies because the RB6-8C5 Ab also detects Ags expressed on a sizeable population of CD3(+)CD8(+) T cells, constitutively as well as following transplantation. Likewise, a high fraction of mice transplanted with limiting numbers of ex vivo expanded Lin(-)Sca(+)kit(+)CD34(-) HSC show long-term RB6-8C5(+)CD3(+) (lymphoid) but no RB6-8C5(+)CD3(-) (myeloid) reconstitution. Most noteworthy, the use of RB6-8C5 as a myeloid specific Ab can be deceptive by implicating the existence of lineage-restricted HSC capable of long-term reconstituting the myeloid and T, but not B, cell lineage. Because cross-lineage expression of "lineage-specific" markers is unlikely to be unique to the blood system, claims of unexpected cell fates should be substantiated not only by acquisition of lineage-specific markers, but also absence of markers of other lineages normally derived from the investigated stem cells.     

IL-4 regulates c-kit proto-oncogene product expression in human mast and myeloid progenitor cells.

The c-kit proto-oncogene encodes the receptor for a novel hemopoietic cytokine, termed stem cell factor (SCF) or mast cell growth factor (MGF) according to its stimulating spectrum. The human receptor for SCF/MGF is expressed in a subset of normal bone marrow progenitor cells, in leukemic myeloid cells, and in mast cells. In the present study, the effects of recombinant human growth regulators (IL-1 through -9, granulocyte-macrophage/granulocyte/macrophage-CSF, IFN, and TNF) on c-kit proto-oncogene product expression were analyzed by indirect immunofluorescence, by using the anti-SCF/MGFR mAb YB5.B8, and Northern blot analyses, by using a c-kit oligonucleotide probe. Of all cytokines tested, IL-4 was found to down-regulate expression of YB5.B8 Ag in the human mast cell line HMC-1 (maximum inhibition, 51.05 +/- 16.36% mean fluorescence intensity of control; p less than 0.02), as well as in primary leukemic myeloid cells. IL-4 was also found to down-regulate expression of YB5.B8 Ag in normal enriched bone marrow progenitor cells. The effects of IL-4 on expression of YB8.B8 Ag in myeloid/mast cell progenitors was dose and time dependent (maximum effects observed on days 2 and/or 4, by using 50 U/ml of rIL-4) and could be neutralized by using anti-IL-4 mAb. Moreover, IL-4 was found to down-regulate expression of c-kit mRNA in leukemic myeloid cells as well as in HMC-1 cells. Together, these observations identify IL-4 as a regulator of c-kit proto-oncogene product expression in the human system. The effects of IL-4 on human hemopoietic progenitor cells and mast cells may be mediated in part through regulation of SCF/MGFR expression.     

Population dynamics of "null" and Thy1lo lymphocytes in mouse bone marrow: genesis of cells with natural killer cell lineage characteristics.

The population dynamics of "null" small lymphocytes lacking B and T lineage markers in mouse bone marrow have been examined using a combination of immunolabeling and hydroxyurea (HU) deletion techniques. The binding of the B lineage-associated mAb, 14.8, and anti-Thy1.2 to bone marrow cells has been detected radioautographically. Null cells lacking 14.8 and Thy1.2 determinants (14.8- Thy1-) formed a substantial subset (12-14%) of bone marrow small lymphocytes, representing 0.5 x 10(6) cells per femur (2-3% of nucleated cells). HU treatment revealed an exceptionally rapid turnover of the null small lymphocyte population (T1/2, 7.5 hr) compared with 14.8+ cells (T1/2, 20.5 hr) and Thy1+ cells (T1/2, 53 hr). Small lymphocytes bearing low intensities of Thy1 (Thy1lo) were also rapidly renewed (T1/2, 28 hr) whereas those with high intensities of Thy1 (Thy1hi) were renewed only slowly (T1/2, 123 hr). During ontogeny, null small lymphocytes first appeared in the fetal liver by Day 11 and the fetal spleen by Day 16, but increased rapidly in the bone marrow in early postnatal life. Double immunolabeling techniques demonstrated that 10% of null small lymphocytes in the bone marrow expressed NK1.1 antigen, while larger proportions bound to tumor (YAC.1) cells in vitro and displayed Fc receptors. The NK1.1-bearing fraction of null small lymphocytes in bone marrow was depleted by HU treatment only after an initial delay. NK1.1 was also expressed on subsets of Thy1lo cells and Thy1hi cells. The results have revealed the continuous production in mouse bone marrow of null and Thy1lo small lymphocytes, totaling 1-3 x 10(7) cells/day and 1.2 x 10(6) cells/day, respectively. The findings suggest that the large-scale production of null lymphocytes in mouse bone marrow includes the genesis of NK lineage cells which express NK1.1 and Thy1lo during a period of terminal maturation.     

BP-3 alloantigen. A cell surface glycoprotein that marks early B lineage cells and mature myeloid lineage cells in mice

To explore the cell surface molecules expressed on pre-B cells we have produced a panel of alloantibodies against transformed pre-B cells from BALB/c mice by immunizing a wild mouse, Mus spretus. One of these antibodies, BP-3, recognized glycoproteins of Mr 38,000 to 48,000 on pre-B cells transformed either by the Abelson murine leukemia virus or an erb B oncogene construct. Removal of N-linked oligosaccharides from the BP-3 Ag revealed a single core protein of Mr 32,000. The Ag was expressed by bone marrow cells in all but one (A/J) of the inbred mouse strains tested and in wild mice of biochemical groups Mus-1 and Mus-2. Analysis of the tissue distribution revealed expression of the BP-3 reactive molecule on normal pre-B and B cells in the bone marrow, 35% of B cells in the circulation, 30% of the B cells in the spleen, and less than or equal to 20% of B cells in lymph nodes, peritoneal cavity, and Peyer's patches. The subpopulation of BP-3+ B cells in bone marrow and peripheral tissues displayed an immature phenotype (IgM IgD +/- ). Examination of a panel of transformed B lineage cells confirmed the early stage-specific expression of the BP-3 alloantigen. In addition, a myeloid cell line and normal myeloid cells were found to express the BP-3 alloantigen. In contrast to B lineage cells, the level of BP-3 expression increased as a function of myeloid cell differentiation. Myeloid cells in the bone marrow expressed relatively little Ag, whereas circulating neutrophils and peritoneal macrophages expressed relatively high levels of the BP-3 alloantigen with Mr 38,000, 41,000, and 46,000. The data suggest that this variably glycosylated cell surface protein could play different roles in the differentiation of B lineage and myeloid lineage cells. The BP-3 alloantigen appears to be a useful marker for virgin B cells that have recently migrated from the bone marrow to the periphery.     

Differential expression of a 70 kDa O-glycoprotein on T cells: a possible marker for naive and early activated murine T cells

We purified a 70 kDa O-glycoprotein that binds to the GalNAc specific lectin from Amaranthus leucocarpus (ALLr) and determined its expression pattern on T lymphocytes from different murine lymphoid organs. High level of ALLr expression was demonstrated in 95-98% of both CD4(+)8(+) and CD4(-)8(+) thymocytes, and in 80-95% of CD8(+) T cells from peripheral blood, lymph nodes, and spleen, whereas a minor fraction of CD4(+)8(-) thymocytes (46-67%) and peripheral CD4(+) T cells (9-40%) showed low ALLr expression. Peripheral CD19(+) B cells were ALLr negative and most of the peripheral ALL(+) T cells showed a CD62L(hi)CD45RB(hi)CD44(lo/-) phenotype, indicating features of naive cells. Mitogenic activation of peripheral T cells increased 3-fold the number of ALL(+)CD4(+) T cells 24 h after stimulation, as opposed to a >80% decrease in CD8(+) T cells 72 h after stimulation. Our results suggest that ALL detects a non-described surface O-glycoprotein selectively expressed by naive CD8(+) T cells and by early activated CD4(+) T cells.     

Absence of cross-reactivity between murine Ly-6C and Ly-6G.

BACKGROUND: The Ly-6 family has many members, including Ly-6C and Ly-6G. A previous study suggested that the anti-Ly-6G antibody, RB6-8C5, may react with Ly-6Chi murine bone marrow (BM) cells. This finding has been interpreted as cross-reactivity of RB6-8C5 with the Ly-6C antigen, and has been generalized to many hematopoietic cell types, using the terminology Ly-6G/C. The present study was undertaken to determine whether anti-Ly-6G antibodies truly cross-react with the Ly-6C antigen on multiple hematopoietic cell types. METHODS: Splenocytes, thymocytes, and BM cells obtained from Ly-6.1 and Ly-6.2 strains of mice were stained with a variety of antibodies to Ly-6C and Ly-6G. Flow cytometric analysis was performed on these populations. RESULTS: Evaluation of anti-Ly-6C and anti-Ly-6G staining showed only Ly-6C expression and no Ly-6G expression on subsets of splenic T and B cells and thymocytes from Ly-6.1 and Ly-6.2 mice. Bone marrow cells were identified that express both Ly-6G and Ly-6C; no Ly-6G+Ly-6C- populations were seen. CONCLUSIONS: Multiple Ly-6C+ hematopoietic cell populations were identified that do not stain with anti-Ly-6G antibodies. This calls into question the use of the Ly-6G/C nomenclature and suggests that epitopes recognized by anti-Ly-6G antibodies should simply be designated Ly-6G.     

Identification and analysis of a novel human surface CD5- B lymphocyte subset producing natural antibodies.

The production of "natural" autoantibodies or antibodies, i.e., Ig that bind a variety of self- and/or exogenous Ag and arise independently of known immunization, is though to be a feature of CD5+ B lymphocytes. To determine whether other lymphocyte subsets exist that might be committed to the production of natural antibodies, human peripheral blood B cells were sorted on the basis of surface CD5 expression and differential expression of surface CD45RA (CD5+CD45RAintermediate(int), CD5-CD45RAlow(lo), and CD5-CD45RAhigh(hi)), and analyzed for the type of Ig produced after EBV infection and culture. Like their CD5+ counterparts, most CD5-CD45RAlo B lymphocytes were precursors of cells producing IgM, a major proportion of which displayed the Ag-binding features of natural antibodies. In contrast, CD5-CD45RAhi B cells comprised a high frequency of IgG-producing cell precursors, possibly including memory B lymphocytes. Six of seven IgM mAb generated from sorted CD5-CD45RAlo B cells and three of four IgM mAb from sorted CD5+ B cells were polyreactive, binding with different affinities (Kd, 10(-5) to 10(-8) M) to two or more Ag; the remaining mAb from CD5-CD45RAlo and the mAb from CD5+ B cells each bound to a single Ag (Kd, 10(-7) to 10(-8) M), beta-galactosidase and ssDNA, respectively. CD5-CD45RAlo B cells account for 4.1 +/- 1.2% (mean +/- SD in 11 healthy subjects; CD5+ B cells, 23.3 +/- 6.9%) of total B lymphocytes and display the features of quiescent cells. In a given individual, the number of CD5-CD45RAlo B cells remains constant over time. CD5-CD45RAlo and CD5+ B cells bear surface CD11b and CD14, at densities and/or frequencies apparently higher than those of CD5-CD45RAhi B lymphocytes. Despite their surface CD5- phenotype, CD45RAlo B cells express CD5+ mRNA at levels comparable with those of CD5+ B lymphocytes, whereas CD5-CD45RAhi B cells express only trace amounts of CD5 mRNA. The commitment to natural antibody production and the degree of CD5 mRNA expression suggest that the newly defined CD5-CD45RAlo B cell subset is related to CD5+ B lymphocytes, and may constitute the human homologue of the mouse Ly-1-"sister" B cell population.     

Beta 1 integrin (CD29) expression on human postnatal T cell subsets defined by selective CD45 isoform expression

The integrin beta 1 (CD29) is a marker for total very late activation Ag integrins on cells, and exhibits considerable fluctuation in cell surface density at various stages of T cell development. We have analyzed beta 1 integrin expression on subsets of human thymus, and on T cells from healthy babies and children, in comparison to healthy adults aged 26 to 75. T cells from adult peripheral blood include a CD29-, a CD29lo, and a CD29hi set. Compared with adults, PBMC T cells from children have reduced numbers of both CD29lo and CD29hi subsets but equivalent numbers of CD29- T cells. The number of CD29hi T cells increases gradually with age, achieving adult levels only at about 26 yr of age; in aged adults (69 to 75 yr), nearly all T cells have a CD29hi phenotype. Most thymocytes and cord blood T cells, in contrast, have a single peak of CD29 staining that is intermediate to the two peaks seen in adults. Multi-negative progenitor and CD45RO- thymocytes (presumptive thymic generative line-age) are 98% CD29hi. Progenitor thymocytes and adult PBMC T cells express equivalent amounts of beta 1 and alpha 4, but progenitors are alpha 5hi, whereas PBMC T cells are alpha 5lo. T cells from children have reduced beta 1hi and alpha 5lo, but nearly comparable numbers of alpha 4hi. This suggests that the major very late activation Ag integrins during childhood may be alpha 5 beta 1 and alpha 4 complexed with an alternate beta chain. In children, the majority of CD29hi cells are also CD45RAhi, in contrast to the pattern in adults, in whom the majority of CD29hi T cells are CD45RA-. This suggests that in children, the main defense against infection may reside in the CD29hi45RAhi T cells, which have not yet made the transition to CD45RO and to bona fide memory status. The proliferative response to tetanus toxoid of 4- to 6-mo-old babies correlates with the number of CD29hi45RAhi T cells, suggesting that it derives at least in part from cells that do not express a "memory" phenotype. These observations show a pattern of alternating high and low density CD29 during T cell development, which is consistent with the idea that CD29 is a marker for functionally defined T cell sets. Analysis of the CD29 expression of CD29hi thymocytes developing in vitro supports this view. We suggest that the intensity of CD29 expression on a T cell varies, dependent upon the microenvironmental interactions required by a differentiating T cell.     

IL-15-dependent upregulation of GITR on CD8 memory phenotype T cells in the bone marrow relative to spleen and lymph node suggests the bone marrow as a site of superior bioavailability of IL-15

CD8 memory T cells are enriched in the bone marrow, a site where these cells are thought to receive homeostatic signals. However, the primary site where CD8 memory T cells receive their cytokine-induced homeostatic signals has recently come under debate. In this study, we demonstrate that the bone marrow contains a fraction of CD8 memory phenotype T cells with elevated expression of glucocorticoid-induced TNFR-related protein (GITR). In contrast, splenic and lymph node memory phenotype T cells have GITR levels similar to those on naive T cells. The bone marrow GITR(hi) memory T cells have a phenotype indicative of cytokine activation, with higher CD122 and lower CD127 than do the GITR(basal) memory T cells. Remarkably, these bone marrow-specific GITR(hi) cells are almost completely ablated in the absence of IL-15, whereas TNFR2 and 4-1BB expression on the CD8 memory T cells are IL-15 independent. Furthermore, adoptively transferred splenic CD8 memory phenotype T cells show IL-15-dependent GITR upregulation upon entry into the bone marrow. This result implies that the selective appearance of GITR(hi) memory phenotype T cells in the bone marrow reflects the local microenvironment rather than a different subset of memory T cells. GITR(-/-) mice have a lower frequency of CD8 memory phenotype cells in the bone marrow, yet the GITR(-/-) cells hyperproliferate compared with those in wild-type mice. Taken together, these data suggest that GITR plays a role in the survival of CD8 memory phenotype T cells and that GITR upregulation represents a precise marker of cells that have responded to IL-15.     

Functional expression of human CD8 in fully reconstituted mice after retroviral-mediated gene transfer of hemopoietic stem cells.

Retroviral-mediated gene transfer has been used in an attempt to efficiently and stably express functional cell-surface molecules in lymphoid and myeloid cells. The human CD8 molecule is a T cell-specific surface receptor that is intimately involved in class I MHC-restricted Ag recognition and subsequent T cell activation. After infection with a recombinant, replication-defective retrovirus containing the human CD8 alpha cDNA, bone marrow cells were transplanted into lethally irradiated recipients. The majority of lymphoid and myeloid cells of reconstituted animals expressed high levels of human CD8 for at least 8 months after transplantation. Transfer of bone marrow and spleen cells from these recipients 100 days after transplantation into secondary recipients also resulted in long term expression of CD8 in lymphoid and myeloid cells. CD8 expressed in splenic T cells associated with the lymphoid-specific tyrosine protein kinase p56lck, participated in T cell activation and conferred an increased xenogeneic response to human MHC class I Ag. Thus, retroviral-mediated gene transfer allows the long term, functional expression of cell-surface molecules in normal murine lymphoid and myeloid cells.     

Accelerated dendritic cell differentiation from migrating Ly6C(lo) bone marrow monocytes in early dermal West Nile virus infection

No study has investigated the participation of Ly6C(+) monocytes in the earliest phase of skin infection with the mosquito-borne West Nile virus. In a novel murine model mimicking natural dermal infection, CCL2-dependent bone marrow (BM)-derived monocyte migration, differentiation into Ly6C(+) dendritic cells (DC), and accumulation around dermal deposits of infected fibroblasts by day 1 postinfection were associated with increasing numbers of monocyte-derived TNF/inducible NO synthase-producing DC by day 2 postinfection in draining auricular lymph nodes (ALN). Adoptive transfer demonstrated simultaneous migration of bone marrow-derived Ly6C(lo) monocytes to virus-infected dermis and ALN, where they first become Ly6C(hi) DC within 24 h and then Ly6C(lo) DC by 72 h. DC migration from the infected dermis to the ALN derived exclusively from Ly6C(lo) BM monocytes. This demonstrates that Ly6C(hi) and Ly6C(lo) BM-derived monocytes have different fates in vivo and suggests that BM may be a reservoir of preinflammatory monocytes for rapid deployment as inflammatory DC during virus infection.     

Analysis of human megakaryocytic cells using dual-color immunofluorescence labeling

BACKGROUND: Megakaryocytes are classically identified by their cellular morphology and expression of platelet glycoproteins. METHODS: In this study, the expression of GPIIIa (CD61) on hemopoietic cells was analyzed by dual-fluorescence flow cytometry. RESULTS: All monocytic cells (CD14+) were shown to coexpress CD61. As the expression of platelet protein on these monocytic cells cannot be reduced by treating the cells with anticoagulant (ethlyenediaminetetraacetic acid [EDTA]), this observation is not simply due to platelet adhesion. When sorted CD61(lo)CD14+ cells were studied by light and electron microscopy, platelets or platelet fragments could not be detected on the cell surface. These cells were found to have typical monocytic morphology but no megakaryocytic characteristics. CONCLUSIONS: This finding demonstrates that without careful definition, the quantitation of megakaryocytic cells will be inappropriately high. A clear and unambiguous criteria for the identification of megakaryocytic cells is described based on the high expression of platelet glycoprotein (e.g., CD61(hi) or CD41(hi)) but not the monocyte marker (CD14(neg)).     

Chimeric antigen receptor T cells shape myeloid cell function within the tumor microenvironment through IFN-γ and GM-CSF

The infiltration of suppressive myeloid cells into the tumor microenvironment restrains anti-tumor immunity. However, cytokines may alter the function of myeloid lineage cells to support tumor rejection, regulating the balance between pro- and anti-tumor immunity. In this study, it is shown that effector cytokines secreted by adoptively transferred T cells expressing a chimeric Ag receptor (CAR) shape the function of myeloid cells to promote endogenous immunity and tumor destruction. Mice bearing the ovarian ID8 tumor were treated with T cells transduced with a chimeric NKG2D receptor. GM-CSF secreted by the adoptively transferred T cells recruited peripheral F4/80(lo)Ly-6C(+) myeloid cells to the tumor microenvironment in a CCR2-dependent fashion. T cell IFN-γ and GM-CSF activated local, tumor-associated macrophages, decreased expression of regulatory factors, increased IL-12p40 production, and augmented Ag processing and presentation by host macrophages to Ag-specific T cells. In addition, T cell-derived IFN-γ, but not GM-CSF, induced the production of NO by F4/80(hi) macrophages and enhanced their lysis of tumor cells. The ability of CAR T cell therapy to eliminate tumor was moderately impaired when inducible NO synthase was inhibited and greatly impaired in the absence of peritoneal macrophages after depletion with clodronate encapsulated liposomes. This study demonstrates that the activation of host macrophages by CAR T cell-derived cytokines transformed the tumor microenvironment from immunosuppressive to immunostimulatory and contributed to inhibition of ovarian tumor growth.     

T cells bearing the CD44hi "memory" phenotype display characteristics of activated cells in G1 stage of cell cycle.

T cells capable of anamnestic proliferative responses to antigen in vitro (i.e., "memory" cells) have been shown to display the CD44hi CD45RBlo surface phenotype. To assess the state of activation of these T cells, CD4+ T cells expressing the CD44hi or CD45RBlo phenotype were compared to CD4+ T cells expressing the CD44lo or CD45RBhi phenotype in the context of expression of the "activated" (asialo-GM1hi) vs "resting" (asialo-GM1lo) phenotype and in the context of cell size, total protein content, and total RNA content. Dual fluorescence analysis demonstrated that all CD4+ T cells expressing the CD44hi phenotype also expressed the asialo-GM1hi phenotype associated with cell activation. In vitro proliferative assays confirmed that the CD4+ asialo-GM1hi, the CD4+ CD45RBlo, and the CD4+ CD44hi FACS-sorted populations displayed stronger in vitro responsiveness to stimulation with immobilized anti-CD3 mAB than the CD4+ asialo-GM1lo, CD45RBhi, or CD44lo populations. Acridine orange analysis of sorted CD44hi/lo fractions revealed that the diploid (G1) population of the CD44hi T cells displayed a higher mean RNA content than the CD44lo T cells. Similarly, the CD44hi T cells displayed a higher mean cell size and a higher mean total protein content than the CD44lo CD4+ T cells. Similar results were obtained with asialo-GM1 and CD45RB subsets of CD4+ T cells. The basal rate of protein synthesis, as determined by [3H]leucine incorporation, was approximately 50% higher in the CD44hi small CD4+ T cells than in the CD44lo CD4+ T cells. Based on the knowledge that cell size, total protein and RNA content, and responsiveness to signals inducing proliferation are lowest in G0 stage of cycle and increase through G1 stage of cycle, it appears that the CD44hi CD45RBlo T cells exist in a higher activation state than CD44lo CD45RBhi T cells. The previously demonstrated association of CD44hi CD45RBlo phenotype with memory T cells suggests that the CD44hi memory T cells are maintained in G1 (not necessarily cycling) rather than resting "out of cycle" in G0.     

Differentiation of hematopoietic stem cells in irradiated mouse thymic lobes. Kinetics and phenotype of progeny

To define cell populations which participate in the very early stages of T cell development in the mouse thymus, we enriched hematopoietic stem cells from mouse bone marrow and injected them into thymic lobes of irradiated Ly-5 congenic recipients. The progeny of the stem cells were identified and their phenotypes were determined by two-color flow cytometry for the expression of various cell surface differentiation Ag during the course of their subsequent intrathymic development. The majority of the differentiation which occurred in the first 10 days after intrathymic cell transfer was myeloid in nature; hence, this study demonstrates that the irradiated thymus is not strictly selective for T cell development. Further, the maximum rate of T cell development was observed after intrathymic injection of 200 stem cells. Donor-derived cells which did not express Ag characteristic of the myeloid lineage could be detected and their phenotypes could be determined by flow cytometry as early as 7 days after intrathymic injection. At this time, the cells were still very similar phenotypically to the bone marrow hematopoietic stem cells. Exceptions to this were the expression of stem cell Ag 2 and a decrease in the level of MHC class I Ag expression. After 9 days, the donor-derived cells expressed high levels of the Thy-1 Ag and proceeded to change in cell surface phenotype as differentiation continued. These cell phenotypes are described for the time frame ending 18 days after injection, when most donor-derived cells were phenotypically small CD4+ CD8+ (double-positive) thymocytes.