Discontinuous Bovine Serum Albumin Gradients Research Articles

Deletion of Bcl-X in the Megakaryocyte Compartment Results in Profound Thrombocytopenia Caused by Impaired Platelet Production and Survival.

Abstract 1495Poster Board I-518Recent studies have suggested a role for the intrinsic apoptosis pathway in megakaryocytic differentiation and platelet shedding. The intrinsic pathway is regulated by the Bcl-2 family of pro- and anti- cell death proteins. We recently demonstrated that platelet life span is controlled by an intrinsic cell death pathway, whereby the anti-apoptotic protein Bcl-xL constrains the pro-apoptotic activity of Bak to maintain platelet survival. As Bcl-xL is expressed in megakaryocytes, we investigated whether this protein is required for megakaryocyte survival, differentiation and/or platelet shedding. We specifically deleted the Bcl-x gene in the megakaryocyte lineage by crossing mice carrying a floxed allele of Bcl-x with mice carrying a platelet factor 4-regulated Cre transgene. Bcl-xfl/flCre+ mice were profoundly thrombocytopenic (26 ± 5 × 103/μl, n=14) compared with Bcl-xfl/flCre− animals (1157 ± 202 × 103/μl, n=13). Platelet life span in these mice was reduced to only 5 hours, as compared to 5 days in wild type littermates. This result confirmed that Bcl-xL is absolutely required for platelet survival. To determine whether Bcl-x deletion has an impact on platelet production, we analyzed the megakaryocyte compartment in Bcl-xfl/flCre+ and Bcl-xfl/flCre− mice. We observed that the number of megakaryocyte progenitors, and number of megakaryocytes in the bone marrow were increased in Bcl-xfl/flCre+ mice (23 ± 9 megakaryocyte progenitors vs 11 ± 5, and 51 ± 9 megakaryocytes vs 12 ± 1). This result suggested that Bcl-xL is not required for the survival of megakaryocytes or their progenitors. To determine whether Bcl-xL is required for the last step of megakaryocyte differentiation, i.e. platelet shedding, we cultured fetal liver cells with thrombopoietin. Large megakaryocytes were isolated after 3 days of differentiation on discontinuous bovine serum albumin gradient. They were cultured for 3 more days in the same media and the percentage of megakaryocytes displaying proplatelets was determined each day. Interestingly, Bcl-xfl/flCre+ megakaryocytes died much more quickly than Bcl-xfl/flCre− megakaryocytes, and almost none of those that survived were able to form proplatelets. Our study indicates that Bcl-xL is not only essential for platelet survival, but it is also required for the survival of mature megakaryocytes at the stage of platelet shedding. Disclosures:No relevant conflicts of interest to declare.

Bortezomib Induces Thrombocytopenia by Inhibiting Proplatelet Formation but Not Proliferation and Endomitosis in Human Primary Megakaryocytes

The proteasome inhibitor bortezomib has therapeutic activity in patients with multiple myeloma. The most common adverse event from its application is thrombocytopenia, which has kinetics that differ from those induced by other cytotoxic agents. After treatment with bortezomib, platelet counts usually decrease within a couple of days but rapidly recover toward baseline during the rest periods between each cycle. The lowest count of platelets in each cycle does not worsen during the 8 courses of bortezomib treatment. Furthermore, bortezomib does not induce any cytotoxic injury in megakaryocyte in the murine model. Therefore, we postulated that bortezomib-induced thrombocytopenia is caused by inhibition of the platelet releasing process without megakaryocyte toxicity. In vitro assays using human bone marrow-derived CD34-positive hematopoietic stem cells revealed that bortezomib did not inhibit colony formation and endomitosis of human primary megakaryocytes in the presence of recombinant human thrombopoietin (rhTPO). As proplatelet formation (PPF) is often used as the indicator of the platelet releasing process in vitro, we evaluated the inhibitory effects of bortezomib for PPF. Seven days after culture of human CD34-postive cells with 10 ng/ml rhTPO, mature megakaryocytes were enriched by discontinuous bovine serum albumin gradients (purity>90%). The enriched mature megakaryocytes were treated with various concentrations of bortezomib for a further 4 days and the percentage of megakaryocytes bearing PPF was calculated under a microscope. Bortezomib dose-dependently inhibited PPF from mature megakaryocytes. Other proteasome inhibitors such as lactacystin and MG132 also demonstrated inhibitory effects on PPF without inhibiting colony formation of megakaryocytes. Since the inhibition of transcriptional factor NF-kB activity is one of the major pathways of proteasome inhibitors, we evaluated the effects of NF-kB inhibitors such as (−)-DHMEQ and Bay11-7082. Both of these inhibitors also demonstrated inhibitory effects on PPF but did not inhibit the colony formation of megakaryocytes. To exclude the direct effects of bortezomib on human platelets, we analyzed the effects of bortezomib for the activation of caspase-3 and mitochondrial potential in human platelets. We found that bortezomib did not directly induce apoptosis in human platelets. Our results demonstrate that bortezomib induces thrombocytopenia by inhibiting PPF but does not affect proliferation of megakaryocytes, endomitosis and platelet apoptosis. We believe this is the first report using human primary megakaryocytes to clarify the pathogenesis of thrombocytopenia caused by bortezomib therapy.

Long-term liquid storage of porcine spermatozoa separated using a discontinuous bovine serum albumin gradient.

Two experiments were conducted to determine efficacy of a discontinuous bovine serum albumin (BSA) gradient for isolating viable porcine spermatozoa more tolerant to 5-d liquid storage in Beltsville Thawing Solution (BTS) at 15 degrees C. The gradient, contained in a 500-ml separatory funnel, consisted of 4% BSA (60 ml) over 10% BSA (60 ml). Spermatozoa were extended in 26 ml of BTS, layered on top of the gradient, and allowed to migrate through the BSA. The quality of spermatozoa separated by the gradient varied among boars. However, populations of spermatozoa isolated from the bottom 30 ml of the gradient (Fraction 4) consistently contained a high percentage of spermatozoa with acrosomes possessing normal apical ridges (NAR; 89.6%) and progressively motile spermatozoa (MOT; 84.0%), as well as spermatozoa with high velocity (VEL; 336.5 mu/s). Increasing sperm migration time, but not gradient temperature, increased the number of spermatozoa recovered in Fraction 4, but it did not reduce quality of the separated spermatozoa. Spermatozoa isolated in Fraction 4 had greater NAR, MOT and VEL after 5-d storage in BTS than did unseparated spermatozoa. Boar spermatozoa isolated on a discontinuous BSA gradient were more tolerant to storage at 15 degrees C than were unseparated spermatozoa. Such a population may be desirable for use in artificial insemination programs.

Isolation of a population of intact, highly-motile porcine spermatozoa using a discontinuous bovine serum albumin gradient

This study was conducted to determine the efficacy of isolating intact, highly-motile porcine spermatozoa using a discontinuous bovine serum albumin (BSA) gradient. Either 0.25, 0.5, 1, 2.5 or 5 × 10 9 spermatozoa extended to 26 ml with Kiev extender were layered on a discontinuous BSA gradient (4% BSA over 10% BSA) contained within a 500-ml separatory funnel. After 1 h of sperm migration at 23° C, four 30-ml aliquots designated Fractions 1, 2, 3, 4 and Fraction T were collected from the bottom of the funnel. Fraction 4 was the bottom fraction and Fraction T was the remnant of the applied sample. For all concentrations of applied spermatozoa, the percentage of progressively motile spermatozoa was greater in Fraction 4 than in Fractions T or 1 (P<0.05). Fraction 4 contained a greater (P < 0.05) proportion of spermatozoa with normal apical ridge acrosomes than Fraction T. Regardless of the concentration of spermatozoa applied, Fraction 4 contained more than 90% of progressively motile spermatozoa and spermatozoa possessing normal apical ridge acrosomes. The percentage of applied cells recovered in Fraction 4 decreased as the concentration of spermatozoa applied to the gradient increased.

Cell size, DNA, and cytokeratin analysis of human head and neck tumors by flow cytometry

Cell subsets have been discriminated in cell suspensions derived from 37 human head and neck tumors by means of light scatter, DNA, and cytokeratin flow cytometry (FCM). Cell dispersion was performed overnight at 4 degrees C in two different enzyme mixtures, i.e., trypsin/dithioerythritol and collagenase/DNase, under slight agitation of sliced tumor tissue. Cells were examined before and after fractionation on a discontinuous low-density bovine serum albumin (BSA) gradient. Forward and right-angle light scatter FCM of 23 tumor specimens revealed four main subpopulations with different size and structure. Fractionation of primary cell suspensions on a BSA gradient at unit gravity separated debris, small cells and large cells. DNA FCM of the enriched populations demonstrated a relation between large cells and DNA aneuploidy. Epithelial cells, as recognized by cytokeratin antibodies, were also related with large cells. The results demonstrated the usefulness of light scatter, DNA, and cytokeratin analysis of crude and fractionated tumor cell suspensions for assessment of the efficacy of a particular dispersion technique and to obtain information of the cell subsets dispersed.

Evaluation of accessory cell heterogeneity. II. Failure of dendritic cells to activate antigen‐specific T helper cells to soluble antigens

The activation of antigen-specific T cells requires Ia+, antigen-presenting accessory cells (AC). Dendritic cells (DC) and macrophages (M phi) isolated from spleen an peritoneal exudate were tested as AC for the activation of the activation of T helper cells and the induction of T cell proliferation. The cell separations to obtain DC and splenic M phi were performed by discontinuous bovine serum albumin gradients, adherence on petri dishes and rosetting with opsonized sheep erythrocytes. DC as well as the M phi were able to induce antigen-specific T cell proliferation, but only the M phi and not the DC activated antigen-specific T helper cells which help B cells for antibody production to soluble antigens. Keyhole limpet hemocyanin-specific T cells repeatedly stimulated with DC and antigen also did not express helper activity. The failure of DC to induce T helper cells was not due to the activation of a suppressor pathway. Thus, dendritic cells, although very efficient as AC in the induction of various T cell functions, are not able to activate T helper cells required for carrier-specific T-B cooperation and therefore cannot be the sole accessory cells. Based on these results and on previous data using Ia+ tumor cell lines as AC, we confirm the existence of functional AC heterogeneity.

Human monocyte heterogeneity: interleukin 1 and prostaglandin E2 production by separate subsets.

Human peripheral blood monocytes were separated into four different subpopulations by means of a discontinuous bovine serum albumin gradient. Of the least dense population, 7% were present in fraction A, 11% in fraction B, 28% in fraction C and of the most dense, 34% were in fraction D. The rest (17%) of the recovered cells sedimented as a pellet, of which 95% were dead. The monocytes of fraction D (= greater than or equal to 1.075 kg/l) were major interleukin 1 (IL 1) producers and their presence enhanced immunoglobulin synthesis in vitro. Fraction C (= greater than or equal to 1.070 kg/l) were the major prostaglandin E2 (PGE2) producers and demonstrated suppressor activity on in vitro IgG and IgM synthesis. Fractions A and B had minimal production of either IL 1 or PGE2 and lesser effects on the IgG and IgM synthesis. These data demonstrate functional heterogeneity of peripheral blood monocytes with respect to production of both IL 1 and PGE2 as well as accessory cells for immunoglobulin synthesis.

Rat Bone Marrow Precursors Develop into Dendritic Accessory Cells under the Influence of a Conditioned Medium

Primary in vitro immune responses require the interaction between T lymphocytes and Ia-positive antigen-presenting cells. Recently, techniques were established including discontinuous bovine serum albumin gradients, adherence, irradiation, and Fc-rosetting which led to a positive selection of rat dendritic cells and their identification as the major if not the only accessory cell for rat T lymphocytes treated with the mitogen, sodium periodate, or stimulator cells in mixed lymphocyte reactions. Rat dendritic cells which were found to be present in both lymphoid and non-lymphoid tissues have a low density, are non-adherent, radioresistant, nonspecific esterase-negative, Fc-receptor negative and Ia-positive. Dendritic accessory cells were detected in cultures of low-density bone marrow cells and developed from radiosensitive precursors especially in a medium containing 10-20% of a supernatant elaborated by mitogen-preactivated syngeneic spleen cells. When compared to purified lymph node dendritic cells they were found to be identical in potency and phenotype. Adherent cells from bone marrow possessed neither accessory activity nor influenced the development of dendritic accessory cells. The mature dendritic accessory cells which could be enriched 1000-fold were no longer radiosensitive. The production of dendritic accessory cells from bone marrow precursors was influenced by the culture medium. Serum components were found to suppress their development. However, semipurified factor(s) released by mitogen-preactivated spleen cells enhanced the number of dendritic cells considerably. Activity resided in substances with a molecular weight in the range of 40,000 Daltons. This material could be separated on a Sephadex G75 column from rat Interleukin 2 which had no effect on the formation of dendritic cells but did support proliferation of accessory cell-depleted, mitogen-treated lymph node cells.

Characteristics of bovine spermatozoa after migration through a bovine serum albumin gradient.

Experiments were conducted to determine the efficiency with which viable, morphologically normal bovine spermatozoa could be isolated using a discontinuous bovine serum albumin (BSA) gradient. In the first experiment, extended semen was layered on top of a BSA gradient (4% BSA over 10% BSA) contained in a 500-ml separatory funnel. When comparing 1, 3, 5, 7, 14 or 21 X 10(9) spermatozoa applied to the gradient, the percentage of spermatozoa recovered from the lower third of the 10% BSA ranged from 2.9 to 18.5%. The greatest recovery was achieved when 1 X 10(9) sperm cells were applied. Increasing the number of spermatozoa applied to the gradient increased the percentage of spermatozoa remaining in the upper portions of the gradient. Motility of spermatozoa immediately after collection from the 10% BSA layer of the gradient was greater than 90%, regardless of the number of spermatozoa applied. In a second experiment with freeze-thawed separated or unseparated spermatozoa, post-thaw motility (greater than 60%) and acrosomal integrity (greater than 85%) of separated spermatozoa (4 or 10% BSA layer) was greater (P less than .05) than that of unseparated spermatozoa (38 and 66%, respectively). The discontinuous gradient excluded decapitated spermatozoa and spermatozoa with mid-piece and principal piece abnormalities from entering the lower layers. Sperm cells with head abnormalities were not separated. These data indicate that a population of spermatozoa with a high frequency of viable, motile, morphologically-normal bovine spermatozoa can be isolated using a discontinuous BSA gradient.

Sex ratio after insemination of bovine spermatozoa isolated using a bovine serum albumin gradient.

This experiment was undertaken to determine if a method reported to successfully enrich the proportion of Y-chromosome-bearing spermatozoa in human semen could be adapted for separation of bovine spermatozoa. Semen was collected from four Angus bulls and aliquots were either separated on discontinuous gradients of bovine serum albumin (BSA) or untreated before processing for cryopreservation. Two hundred seventy-one cows or heifers were assigned randomly to be artificially inseminated (20 X 10(6) sperm/insemination) with separated or unseparated spermatozoa. The proportions of male offspring were 45 and 54% after inseminations with separated or unseparated spermatozoa, respectively. In a second phase of the experiment, pooled semen from three Holstein bulls was either extended and frozen without separation or frozen after separation using the discontinuous BSA gradient. Separated and unseparated spermatozoa were analyzed by flow cytometry to determine the ratio of X- and Y-chromosome-bearing spermatozoa based on differences in DNA content. The ratios of X- and Y-bearing spermatozoa in separated or unseparated samples were indistinguishable. We concluded that the separation method did not enrich the proportion of Y-bearing bovine spermatozoa.

Functional heterogeneity of human cord blood monocytes.

Human cord blood monocytes were separated into four different subpopulations by means of a discontinuous bovine serum albumin (BSA) gradient. Of the least dense, 31% were present in fraction A, 17% in fraction B and 13% in fraction C and of the most dense, 20% were in fraction D. The rest (17%) sedimented as a pellet, of which 93% were dead cells. The monocytes of fraction C (density greater than or equal to 1.070) demonstrated suppressor activity on in vitro antibody synthesis of maternal B cells. Fraction D (density greater than or equal to 1.075) monocytes enhanced antibody synthesis of maternal B cells compared with synthesis produced in a similar experiment with unfractionated monocytes. Addition of either fraction A or B monocytes to the mixed culture of T and B cells resulted in antibody production comparable to that produced by addition of unfractionated monocytes. The functional heterogeneity of the cord monocytes was assayed also by 2-deoxyglucose (2-DOG) uptake, a marker for immunological macrophage activation. Fractions A and D showed significantly higher 2-DOG transport than that of unfractionated monocytes; in contrast, fraction C showed a 50% reduction of 2-DOG uptake. Furthermore, in contrast to fraction D, fraction C possessed only minimal phagocytic activity (for antibody-coated sheep erythrocytes) and minimal hemoxygenase enzyme activity. These data demonstrate functional heterogeneity of cord blood monocytes.

Immunological properties of thymus cell subpopulations: rat thymic dendritic cells are potent accessory cells and stimulators in a mixed leukocyte culture

Rat thymus cells were fractionated by centrifugation on a discontinuous bovine serum albumin gradient into two subpopulations: one of high density that accounted for>90% of the recovered cells, and a minor low-density subpopulation containing 4 to 10% of the total cells. The high-density subpopulation consisted mainly of uniform small-sized thymocyteS, whereas the low-density subpopulation contained mostly larger-sized cells. High-density thymus cells did not function either as stimulators in a mixed leukocyte reaction or as accessory cells required for T-cell response to mitogens, Con A and sodium periodate, as determined by 3H-thymidine incorporation. Dense thymus cells also responded poorly to allogeneic and to mitogenic stimulation, even when accessory cells were added. In contrast, the low-density thymus cells responded well to allogeneic stimulation and to both mitogens. In addition, low-density thymus cells possessed stimulatory activity in mixed leukocyte cultures, as well as accessory activity for mitogenic responses. Both activities were found to reside in dendritic cells that were purified extensively (70-90% of the preparation) with good yield. When tested as accessory cells for T-cell responses to periodate, thymic dendritic cells were as potent as lymph node dendritic cells on a per cell basis. A small number of thymic dendritic cells was able to cause marked enhancement in T-lymphocyte proliferation in response to stimulation. By immunofluorescence thymic dendritic cells were shown to be Ia-positive, but Thy 1.1-negative.

Correlation between production of inflammatory substances and generation of effector T cells mediating delayed-type hypersensitivity in mice.

In vitro exposure to human serum albumin (HSA) of splenic lymphocytes from mice sensitized for delayed-type hypersensitivity (DTH) against HSA resulted in the release of substances that could induce a footpad inflammatory reaction with a maximum of 6 hr after injection into normal mice. The substances were fractionated mainly in a molecular weight range of 30,000 to 70,000 daltons on Sephadex G-200. The ability of sensitized lymphocytes to produce the substances was dependent on T cells, was antigen specific, and correlated well with the ability of the lymphocytes to mediate DTH reactions. Moreover, the substances were produced efficiently by the DTH effector cell population generated in the in vitro culture system and also by the effector cell-enriched fractions on discontinuous bovine serum albumin gradients. These results suggest that the substances are produced by DTH-effector cells.

Regulatory mechanism of delayed-type hypersensitivity in mice. III. In vitro analysis of memory T cells involved in augmentation of DTH responses.

The memory of delayed-type hypersensitivity (DTH), manifested by the augmented responsiveness upon challenge with alum-absorbed ovalbumin (OA), was induced in mice primed 7 days, 21 days, or 90 days previously with 1 microgram of reduced and alkylated OA. The memory cells involved in the augmentation of DTH responses were analyzed in the in vitro induction system of T cells which mediate DTH against OA. Spleen cells from the primed mice generated DTH-effector T cells (DTH-Te) in a significantly accelerated fashion, compared with unprimed spleen cells, when cultured with OA. The accelerated generation of DTH-Te in vitro was induced antigen specifically and was dependent on a certain T cell population in the primed spleen. The T cell population was found in the spleen of primed mice for at least 3 months after priming, corresponding to the persistence of DTH-memory in vivo. Moreover, it was fractionated in the high-density layer by discontinuous bovine serum albumin gradient centrifugation. The high-density cell population decreased in density with increase in the time of culture and developed into DTH-Te, which were separated in the low-density layer on day 4 of culture. These results indicate that the T cells involved in the accelerated generation of DTH-Te in vitro are long-lived DTH-memory T cells, which are probably precursor cells, capable of differentiating into DTH-Te upon challenge with the antigen.

Glucocorticoid sensitive and resistant cell populations in the mouse thymus.

We have studied in parallel the effects of dexamethasone on in vitro incorporation of [ 3 H]-uridine and [ 3 H]-thymidine and on cell viability in thymocyte subpopulations isolated by centrifugation on a discontinuous bovine serum albumin gradient from intact, adrenalectomized and hydrocortisone-treated mice. In cell populations isolated from intact and adrenalectomized animals, the steroid induces both inhibition of precursor incorporation and a decrease in cell viability, whereas cells isolated from hydrocortisone-treated mice show marked inhibition of precursor incorporation but no cell death after 24 h. Our present results suggest that differences in steroid sensitivity are not directly related to variations in cell receptor content but may be associated with cell differentiation.

Suppression of Humoral and Delayed Hypersensitivity Repsonses by Distinct T Cell Subpopulations

Abstract Immunization of mice with a supraoptimal dose of sheep red blood cells (SRBC) results in splenic T cell populations capable of specifically suppressing recipients' plaque forming cell (PFC) and delayed-type hypersensitivity (DTH) responses to SRBC when tested in an adoptive transfer system. By localization on discontinuous bovine serum albumin (BSA) gradients and relative sensitivity to Cytoxan, two distinct T cell subpopulations suppressing DTH reactivity were identified. One population could not be distinguished from T cells capable of inhibiting direct and indirect PFC responses. However, another population appeared quite distinct and capable of inhibiting DTH, but not PFC responses.

An improved syncytia infectivity assay for the bovine leukemia virus

Several factors that influence the sensitivity of the syncytia infectivity assay for the bovine leukemia virus (BLV) and BLV-infected lymphocytes have been examined. The use of early-passage indicator bovine embryonic spleen (BESP) cells and their pretreatment with diethylamino-ethyl-dextran (DEAE-D) was essential for optimal sensitivity. Polybrene was less effective than DEAE-D. The combination of DEAE-D and polybrene was more effective than DEAE-D alone when BLV-infected leukocytes were used as the inoculum, but not when the inoculum was a cell-free BLV preparation. Using BESP cell passages 4 to 11 as indicators, reproducible titers were obtained when aliquots of the same virus stock were assayed at different times after freezer storage. When assaying peripheral blood lymphocytes from infected cattle, optimal syncytia responses were observed consistently by inoculating 5 X 10(6) viable lymphocytes per 60-mm Falcon dish. Centrifugation of peripheral blood leukocytes from BLV-infected cattle in discontinuous bovine serum albumin gradients can be used to separate a subpopulation of infected lymphocytes. Use of this subpopulation as the inoculum, rather than unseparated buffy-coat leukocytes, greatly increases the sensitivity of the syncytia infectivity assay.

Suppressor T cells in low zone tolerance II. Characterization of suppressor T and amplifier cells by physical and serological methods

Suppressor cells involved in low zone tolerance to bacteriophage fd have been characterized by serological and physical methods. Suppressor T cells are peripheral T2 cells which are Ly-1-, Ly-2+ and Ia+. They are of high electrophoretic mobility, sediment 5 mm/h at 1 x g in a linear Ficoll 70 gradient, but are not restricted to a distinct band in a discontinuous bovine serum albumin gradient. Adherent phagocytic cells of low electrophoretic mobility can act as amplifier cells with a much shorter half-life than suppressor T cells.

Migratory properties of normal lymphoid cells with different concanavalin A-binding capacities

It has been proposed that Con A receptor sites on lymphoid cell surfaces control lymph node localization of normal T cells. To investigate this hypothesis, suspensions of whole spleen, splenic T cells, and thymocytes from normal BALB/c mice were fractionated on discontinuous bovine serum albumin (BSA) gradients and fractions were examined separately for their ability to bind 125I-labeled concanavalin A and to migrate to lymph nodes. Low density cells from whole spleen or splenic T-cell gradients bound more concanavalin A (Con A) per cell than high density cells. However, dense cells, of low Con A-binding capacity, migrated in higher numbers to recipient lymph nodes than less dense cells. Similarly, studies with thymocytes revealed no correlation between Con A-binding and lymph node-seeking propensity. Thus, although Con A significantly inhibits lymph node localization of lymphoid cells, migration to lymph nodes and Con A binding of normal lymphocytes are not directly correlated. The data do not support a hypothesis that normal T-cell migration to lymph nodes is promoted by Con A receptors on lymphocyte surfaces.

Distribution of Terminal Deoxynucleotidyl Transferase in Bovine Serum Albumin Gradient-Fractionated Thymocytes and Bone Marrow Cells of Normal and Leukemic Mice

The distribution of terminal deoxynucleotidyl transferase (TdT) peaks I and II, in single cell suspensions of thymuses, bone marrow, and peripheral lymphoid organs fractionated in discontinuous bovine serum albumin gradients, was examined in a variety of mouse strains and Fischer 344 rats to relate the normal patterns of thymocyte differentiation to the leukemic process. TdT peaks I and II were found in fractions A (10 to 23%), B (23 to 26%), and C (26 to 29%) of the thymus of both normal and leukemic C57BL/6 mice, whereas only peak I was found in the same fractions of AKR mice. TdT in bone marrow was found mainly in fraction A in both normal and leukemic mice. The specific activity of TdT in this fraction, which comprises only 1 to 5% of the total bone marrow cell population, was similar to that of the thymus. The cell population of fraction A of the bone marrow was found to increase (10 to 15-fold) in leukemic mice. Only low levels of TdT activity were found in either whole or fractionated bone marrow of athymic NIH Swiss (nu/nu) mice.