Function of lymphocyte subpopulations in chronic lymphocytic leukemia. Activity in the allogeneic and autologous mixed lymphocyte reaction

Lymphocyte subpopulations from patients with chronic lymphocytic leukemia (CLL) and from normal age-matched controls were evaluated for their ability to participate in allogeneic and autologous mixed lymphocyte reactions (MLR). Unfractionated and T enriched lymphocyte populations from normal age-matched controls responded well to allogeneic stimulation. T enriched lymphocytes from patients will CLL also responded to allogeneic lymphocytes. B enriched lymphocytes from normal age-matched individuals produced a stronger stimulus in the allogeneic MLR than did unfractionated mononuclear cell populations. Unfractionated and B enriched lymphocyte subpopulations from patients with CLL were poor stimulators in the allogeneic MLR. In normal age-matched controls T enriched lymphocyte subpopulations were able to respond to autologous B enriched lymphocytes. Autologous mixed lymphocyte cultures from patients with CLL failed to demonstrate any activity.     

Expression of an accessory cell phenotype by hairy cells during lymphocyte colony formation in agar culture

Human T lymphocytes require the cooperation of accessory cells to generate lymphocyte colonies in agar culture under PHA stimulation. Various hairy cell enriched fractions, as well as normal monocytes, have been found to be able to initiate colony formation by normal lymphocytes. Leukemic monocytes from CMML patients were also effective, but not the leukemic lymphocytes from CLL patients. The phenotype expressed by HC in agar colonies was further studied using cell surface and enzymatic markers. We have concluded that HC in agar culture in the presence of both normal T lymphocytes and PHA lose the B phenotype that they express in vivo and function like an accessory cell in contrast to normal or leukemic B lymphocytes.     

Lymphocyte response to pokeweed mitogen in chronic lymphocytic leukemia

The response of lymphocyte subpopulations to pokeweed mitogen (PWM) was studied in normal volunteers and patients with B-cell chronic lymphocytic leukemia (CLL). Since unfractionated peripheral blood mononuclear (PBM) cells from CLL patients consist of a markedly increased proportion of B-lymphocytes and a decreased proportion of T-lymphocytes, enriched fractions of CLL B-cells and CLL T-cells were cultured in 1:1 proportions in autologous and allogeneic combinations with normal B-cell and T-cell-enriched fractions. Cultures containing normal B-cells with either autologous or allogeneic normal T-cells responded well to PWM. CLL T-cells were capable of providing a helper function for both proliferation and differentiation of normal B-cells, which was not significantly different from that provided by allogeneic normal T-cells. CLL B-lymphocytes were unresponsive to PWM when cultured in the presence of either autologous CLL T-lymphocytes or allogeneic normal T-lymphocytes. The responsiveness of CLL B-cells was not restored by the addition of normal peripheral blood monocytes to the cultures. These experiments indicate that there is an intrinsic B-cell defect which prevents CLL B-lymphocytes from responding to PWM.     

Non-T, non-B acute lymphocytic leukemia cells enhance the mitogen induced proliferation of normal lymphocytes

Peripheral blood and bone marrow mononuclear cells (MNC) from patients with untreated non-T, non-B acute lymphocytic leukemia (ALL) were assessed for capacity to respond to the mitogens phytohemagglutinin (PHA), pokeweed (PWM) and staphylococcal enterotoxin A (SEA). At low mitogen concentrations considerable reactivity was observed to PHA and SEA, but not to PWM by patient MNC including specimens comprised of at least 90% leukemic cells. The responding cells were identified in cell separation experiments and by surface marker analysis as residual normal T-cells. The non-T, non-B leukemic cells enhanced the mitogen induced proliferation of responding lymphocytes as determined by co-culture experiments in which 5 × 10³ or 10⁴ MNC from normal donors were cultured with feeder layers of 3 × 10⁵ mitomycin C treated MNC from autologous or unrelated normal donors or from ALL patients. In contrast to MNC feeder layers from normal donors, feeder layers of non-T, non-B leukemic cells enhanced normal lymphocyte reactivity to PHA and SEA but not to PWM as determined by ³H-thymidine uptake and cell survival. The potentiating effect of non-T, non-B leukemic cells was not mediated entirely by an allogeneic interaction as shown by the demonstration that frozen-thawed non-T, non-B leukemic cells enhanced the mitogen response of autologous MNC obtained during remission. Mitomycin treated feeder layers of leukemia cells from patients with acute myelogenous leukemia as well as bone marrow MNC from a normal donor or from lymphoma patients without bone marrow involvement failed to enhance PHA and SEA reactivity of normal lymphocytes. In addition, leukemia cells from patients with T-cell ALL did not potentiate normal lymphocyte reactivity in contrast to non-T, non-B ALL cells, providing a functional distinction between these two subtypes of ALL.     

Cultured human leukemic non-T/non-B lymphoblasts and their stimulating capacity in "one-way" mixed lymphocyte reaction: suggestive evidence for early T-cell or B-cell precursors.

It is now well recognized that a large proportion of cases with acute lymphoblastic leukemia are classified as non-T/non-B neoplastic disease. The origin of leukemic non-T/non-B cells is at present not known. It has been shown that fresh or cultured leukemic T lymphoblasts exert no stimulating capacity while leukemic B lymphoblasts exert a strong stimulation in "one-way" mixed lymphocyte reaction. It has also been shown that fresh leukemic cells from some patients with non-T/non-B acute lymphoblastic leukemia possess a strong stimulation while leukemic cells from other patients with this disease possess no stimulation on allogeneic lymphocytes. The present study shows that cultured leukemic lymphoblasts from 3 non-T/non-B cell lines (NALL-1, NALM-6 and NALM-16) consistently exert a strong stimulation on allogeneic lymphocytes. On the other hand, cultured leukemic lymphoblasts from 2 non-T/non-B cell lines (REH and KM-3) consistently fail to stimulate in "one-way" mixed lymphocyte reaction. Our data clearly support the speculation that leukemic non-T/non-B cells which possess the stimulating capacity may represent less differentiated leukemic B lymphoid cells (pre-B cells) and leukemic non-T/non-B cells which possess no stimulating capacity may represent less differentiated leukemic T lymphoid cells (per-T cells).     

Functional and phenotypic analysis of a T cell prolymphocytic leukemia

Peripheral blood mononuclear cells obtained from a patient with prolymphocytic leukemia expressed the surface membrane markers characteristic of resting mature T helper lymphocytes. These cells responded to the T cell mitogens PHA and Con A in a blast transformation assay but not the anti-T cell monoclonal antibody Leu 4 and the B cell mitogen, PWM. The concentration of PHA or Con A eliciting maximum blast transformation was less than that required by normal mononuclear cells. The leukemic cells recognised and responded to allogeneic pooled mononuclear cells in a mixed lymphocyte culture. In addition, although they did not express Ia antigens, they served as effective stimulators in the mixed lymphocyte reaction. Consistent with the helper phenotype, the leukemic cells did not produce suppressor factors, but provided help for normal B-enriched lymphocytes to respond to PWM as assessed by both blast transformation and IgG production. T lymphocyte colonies developed when the leukemic cells were treated with PHA during a 20 h liquid culture prior to being seeded into semisolid agar medium containing either PHA or an IL2-containing lymphokine. There was no growth when untreated cells were seeded directly into IL2-containing agar. Analysis of colony formation indicated that, as with normal resting T lymphocytes, proliferation occurred in two distinct steps; activation in response to PHA and replication in response to IL2-like growth factors. These findings demonstrate that in this case the helper T prolymphocytes have the functional capabilities of normal mature T lymphocytes as predicted from their helper phenotype.     

alpha-L-Fucosidase activity and isoenzyme characteristics analyzed by chromatofocusing in normal and leukemic lymphocytes

alpha-L-Fucosidase (EC 3.2.1.51) activity and isoenzyme characteristics were analyzed in normal lymphocyte subpopulations, chronic lymphocytic leukemia subpopulations, and acute lymphoblastic leukemia blasts. Similar pH activity profiles revealed that pH 5.0 was optimal in normal and leukemic cells. Unfractionated CLL lymphocytes had a lower specific activity than normal unfractionated lymphocytes (2.5 +/- 1.0 u/10(6) cells v. 4.0 +/- 1.1). CLL B cells and T-cells had lower specific activity than their respective normal counterparts (1.8 +/- 0.2 v. 5.9 +/- 2.0) (B-cells); (2.2 +/- 0.5 v. 3.7 +/- 1.0) (T-cells) suggesting T and B cells in CLL are abnormal. ALL blasts had a higher specific activity compared to unfractionated normal lymphocytes (9.7 +/- 3.0 v. 4.0 +/- 1.1; p less than 0.001). The isoenzyme pattern of normal, CLL and ALL lymphocytes were obtained by automated chromatofocusing on PBE 94 microcolumns using 0.025 M histidine and polybuffer 74. Two major isoenzyme components (B and A) were isolated. The activity ratio of B/A was different in normal, ALL, and CLL cells.     

Immunoglobulin synthesis and secretion by leukemic B cells from patients with chronic lymphocytic leukemia

We investigated the ability of purified E-rosette negative largely leukemic B cells from 15 patients with B-cell chronic lymphocytic leukemia (CLL) to synthesize and secrete IgM, IgA and IgG spontaneously or in the presence of purified autologous or allogeneic T4 cells from normal donors, in PWM-induced differentiation system. We observed moderate but significant IgM synthesis and secretion (19.7 +/- 8.9 micrograms/dl, n = 5) by leukemic B cells alone in 5 of 15 patients examined. These IgM concentrations were significantly higher (p less than 0.005) than those produced by purified E-rosette negative cells from normal donors (4.3 +/- 4.5 micrograms/dl; n = 6) in the absence of T cells. Purified E-rosette negative leukemic B cells alone from patients with CLL did not produce IgA or IgG. Addition of purified autologous or allogeneic T4 cells from normal donors resulted in significant increase of IgM production by leukemic B cells from certain patients or initiated IgM secretion in others. However, these IgM levels (73.9 +/- 56.6 micrograms/dl) were significantly lower (p less than 0.003) to those produced by mixtures of T4 cells and B cells form normal donors (211.6 +/- 58.0 micrograms/dl, n = 6). Addition of purified autologous or allogeneic T4 cells from normal donors to purified largely leukemic B cells from patients with CLL resulted in production of very small amounts of IgA in 4 of 15 patients (10.6 +/- 6.3 micrograms/dl vs 154.7 +/- 35.8 micrograms/dl produced by T4 and B cells from normal donors; n = 6), but did not support IgG synthesis and secretion. Purified T4 cells from certain patients with CLL exhibit defective helper function to immunoglobulin production by E-rosette negative cells from normal donors.     

Enumeration and identification of human leukemic lymphocytes by their natural binding of bacteria.

The recently described property of bacteria to bind to human lymphocytes was used to distinguish between normal and chronic leukemic lymphocyte (CLL) populations. Strains of the following bacteria were used in this study: Arizona hinshawii, Escherichia coli strains 1 and 2, Bacillus globigii, Brucella melitensis, Corynebacterium diphtheriae strains 1 and 2, Corynebacterium xerosis, Sarcina lutea, Staphylococcus aureus, and Staphylococcus epidermidis. For identification of immunoglobulin-bearing lymphocytes, a strain of E. coli that did not bind to human lymphocytes was coated with anti-human light-chain antibody. Labeling of lymphocytes with bacteria was promoted by centrifugation. In the eight CLL patients studied, in which greater than 90% of the lymphocytes were leukemic cells, 52 to 77% were labeled by anti-human light-chain antibody-E. coli, 80 to 93% were labeled by Br. melitensis, and 78 to 95% were labeled by E. coli 1 compared to 11 to 24, 11 to 22, and 30 to 44%, respectively, in normal individuals, Thus, Br. melitensis, E. coli 1, and the anti-human light-chain antibody-E. coli may have diagnostic value for CLL. The percentage of the lymphocyte population that bound each of the other bacteria varied from patient to patient. Preliminary results obtained by studying the pattern of binding of E. coli 2, B. globigii, Sa. lutea, or S. aureus by leukemic lymphocytes suggest that categories of CLL patients may be distinguished by this method.     

Immunologic characterization of chronic lymphocytic leukemia cells.

Immunologic characterization of the neoplastic cells in the circulation of patients with CLL suggests these cells show significant differences in membrane characteristics from normal B lymphocytes. Although the leukemic cells bear a homogenous membrane-associated immunoglobulin, they also react with an anti-human T cell serum. In all patients studied, 60-90% of the cells, were stained by this antiserum. This suggests that the leukemic cells share antigenic determinants with T lymphocytes. CLL cells, unlike normal B cells, showed a marked increase in mouse-complement receptors. No increase in receptors for guinea pig complement was observed in the leukemic cells. The population of SIg-bearing lymphocytes was significantly greater than that of complement-receptor bearing lymphocytes. The total number of E-rosetting cells was increased in all CLL patients. Mitogenic responses of the leukemic cells were depressed and delayed. These results suggest that neoplastic lymphocytes cannot be classified as T- or B-derived on the basis of criteria used to define normal lymphocytes.     

Effects of preactivated autologous T lymphocytes on CD80, CD86 and CD95 expression by chronic lymphocytic leukemia B cells

Profound immune dysfunction is a constant feature in B-cell chronic lymphocytic leukemia (B-CLL) patients. Immunological abnormalities include hypogammaglobulinemia, impaired immunoglobulin class switching and diminished germinal center formation. This state of immune suppression renders B-CLL patients highly susceptible to infections, which contribute greatly to morbidity and mortality in this disease. Impaired T cell function in B-CLL is well-documented and has been suggested to result from inhibitory effects exerted by malignant B lymphocytes. Because the presence of leukemic cells may represent a major obstacle to efficient T cell activation, T lymphocytes were separated from CLL B cells, stimulated with phorbol 12-myristate 13-acetate (PMA) and ionomycin for 4h, and then cocultured with autologous leukemic B cells both at a 1:1 ratio or at the same ratio as in vivo for 24-40 h. CLL B cell expression of CD86 and CD95 was markedly upregulated using this approach, whereas CD80 expression was augmented only in a minority of patients; these effects were partially preserved even when preactivated T cells were rechallenged with CLL B cells at the same low T/B cell ratio as that observed in vivo. Finally, CD80 upregulation on CLL B cells appeared to be mainly dependent on CD40L-mediated stimulation, whereas CD86 and CD95 expression was efficiently augmented by soluble factors released by preactivated T lymphocytes. In conclusion, efficient activation of T lymphocytes in B-CLL may be achieved which, in turn, may result in enhanced antigen-presenting capacity and susceptibility to apoptosis of leukemic cells via CD86 and CD95 upregulation, respectively.     

Cell surface changes in human normal and abnormal lymphoid cells detected by sialic acid content and sialyltransferase activity

The sialic acid content and the sialyltransferase activity of human lymphoid cells (T, B, CLL lymphocytes, monocytes) were determined. Purified B lymphocytes isolated from tonsils had lower amounts of total and neuraminidase-susceptible sialic acid than T lymphocytes, monocytes and CLL lymphocytes. The highest activities of sialyltransferase were found in CLL lymphocytes and paralleled the sialic acid content of these cells. These results suggest that CLL lymphocytes which are in most cases mature cells, may survive longer and recirculate differently from normal cells because of their high sialic acid content and sialyltransferase activity.     

Interferon induces proliferation in leukemic and normal B-cell subsets

3H-thymidine incorporation following stimulation with interferon (IFN) in vitro was investigated in cell cultures from peripheral blood of patients with chronic lymphocytic leukemia (CLL), spleens from necro-kidney transplants and healthy blood donors. It was demonstrated, that IFN can induce a proliferative response in some normal as well as leukemic B lymphocyte subsets. The responses were not T-cell dependent. The results indicate, that B-cell subsets that proliferate in the presence of IFN, are present in higher proportions in spleen than in peripheral blood, and that they constitute a portion of the leukemic blood lymphocyte pool in some patients with CLL. We have previously demonstrated, that IFN induces varying degrees of transformation and differentiation in blood lymphocytes from a majority of CLL patients. The functional characteristics of different B-cell subsets, and their heterogeneous distribution in leukemia, may be important for the results of IFN treatment in various malignant B-cell disorders.     

Comparison of histone variant synthesis in human lymphocytic leukemia cells and in normal lymphocytes

The synthesis of core histone variants and of histone H1 variants was determined in fresh leukemic cells of eight patients with leukemia [seven acute lymphoblastic (ALL) and one chronic lymphocytic (CLL)], in normal lymphocytes from healthy donors or from ALL patients in complete remission. Histone variant synthesis was evaluated by incubating cells with [14C]Lys and [3H]Arg in medium without Lys and Arg and then by two-dimensional polyacrylamide gel electrophoretic separations (acetic acid-urea-Triton x-100 acetic acid-urea-hexadecyltrimethylammonium bromide for core histone variants; sodium dodecyl sulfate/acetic acid-urea-hexadecyltrimethyl ammonium bromide for H1 variants). As previously reported, quiescent lymphocytes and lymphocytes stimulated with phytohaemagglutinin (PHA) showed clearcut changes in the proportions of synthesis of core histone variants and H1 variants. Leukemic lymphocytes freshly obtained from blood showed a pattern of core histone synthesis and H1 synthesis intermediate between that of quiescent and PHA-stimulated lymphocytes; this is probably due to the presence of a mixture of resting and growing cells. When leukemic cells were stimulated to grow by mitogens, the pattern of core histone and H1 variant synthesis was similar to that in mitogen-stimulated normal lymphocytes. Histone variants whose synthesis is associated with the S-phase were not synthesized in leukemic cells treated with the DNA synthesis inhibitors hydroxyurea and 1-beta-D-arabinofuranosylcytosine (Ara-C). The pattern of acetylation of histone H4 was also apparently similar in leukemic cells and normal lymphocytes. The radioactivity associated with the ubiquitinated forms of H2A increased in nongrowing lymphocytes and in leukemic cells treated with DNA synthesis inhibitors whereas they decreased after mitogenic stimulation. Variability was wide in the synthesis of ubiquitinated H2A in different cases of leukemia. The only clear-cut difference between leukemic cells and normal lymphocytes was that leukemic cells from ALL patients, but not lymphocytes from normal donors or from ALL patients in complete remission, synthesized appreciable amounts of H1 degrees, increasing after hydroxyurea/Ara-C treatment and decreasing after PHA-stimulation. In leukemic cells from a CLL patient H1 degrees synthesis was undetectable.     

Role of myeloid cell factor-1 (Mcl-1) in chronic lymphocytic leukemia

The primary abnormality in chronic lymphocytic leukemia (CLL) is a defect in apoptosis, probably related to alterations in the expressions of Bcl-2 family members. In transgenic mice over expressing the anti-apoptotic Bcl-2 family member, myeloid cell factor-1 (Mcl-1), B cell lymphomas occur. Moreover, mice conditional for the loss of Mcl-1 display a profound reduction in B and T lymphocytes. This suggests that Mcl-1 is an essential survival factor in lymphocytes. In the present study, we have evaluated the role of Mcl-1 in CLL. Mcl-1 protein expression was measured by Western blot analysis in the CLL cells of 45 patients and correlated with clinical variables and survival. Mcl-1 levels were similar in 29 patients to normal B and T lymphocytes, were decreased in 8 patients and increased in 12 patients. An inverse correlation was found between Mcl-1 expression and Rai stage (P = 0.001). When assessed by flow cytometry, Mcl-1 expressions were normally distributed among CLL cells in individual patients and the mean levels correlated with those obtained by Western blotting. To evaluate the role of Mcl-1 in drug resistance, Mcl-1 levels were sequentially measured in the leukemic cells of 4 CLL patients during therapy with fludarabine (Flu). The Mcl-1 levels were found to increase in 2 patients while the peripheral blood lymphocyte counts dropped, suggesting that the residual drug-resistant cells had the highest Mcl-1 levels. Primary CLL cells were also treated with chlorambucil (CLB) or Flu in vitro and the Mcl-1 levels decreased correlating with the sensitivity of these cells to undergo apoptosis. Drug sensitivities of the CLL cells to CLB and Flu were also measured by MTT assay and the concentrations of drug required to decrease cell viability by 50% (IC50) varied from 1.9 to 9.27 microM for Flu (median, 9.4 microM) and 10 to 32.5 microM (median, 5.5 microM) for CLB. The sensitivities of the leukemic cells to CLB correlated inversely with Mcl-1 levels (P < 0.05). These results suggest that Mcl-1 may contribute to cell survival in CLL.     

Cloning of human lymphocytes reactive with autologous leukemia cells

The primed lymphocyte typing test has been used to detect leukemia-associated antigens, but interpretation has been difficult because of significant levels of reactivity with normal cells. Elimination of unwanted reactivities could be accomplished by (a) use of the patient's own lymphocytes as responders to the leukemia cells and (b) cloning of the responding cells. Cloning of antigen-activated human lymphocytes can be accomplished through the use of T-lymphocyte growth factor, which permits the long-term growth of antigen-activated lymphocytes. In the study reported here, the remission lymphocytes of a patient with acute myelogenous leukemia were sensitized in culture to the patient's own leukemic myeloblasts and then grown from wells containing one or a few replicating units. Sufficient cells of three clones were growth for further testing of specificity: one responded only to the sensitizing myeloblast but not to normal cells tested; one responded to the sensitizing myeloblasts and one allogeneic myeloblast but not to normal cells; and one responded to none of the cells tested, although it proliferated vigorously with growth factor alone. These results demonstrate the feasibility of cloning human lymphocytes putatively responsive to leukemia-associated antigens in order to improve their discriminatory capacity in the primed lymphocyte typing test. The response pattern observed was that expected of a clone responding to a leukemia-associated antigen.     

Stimulation by CLL cells in mixed lymphocyte culture (MLC)

Cells from patients with chronic lymphocytic leukaemia (CLL) do not respond in the mixed lymphocyte culture (MLC) but are able to stimulate the response of normal lymphocytes. Mixed lymphocyte cultures were performed using cells from 24 patients with CLL and cells from 16 normal donors. The stimulatory capacity of 8 of these CLL cells was reduced when a common DR antigen was shared with the normal responding cell. We suggest that cells from certain selected CLL patients may be used in the mixed-lymphocyte reaction for determining the D-locus specificity of normal donors. CLL cells when expressed 1 DR antigen only, induced more clearly defined typing responses than cells with 2 DR antigens. There was no correlation between the ability of a CLL cell to induce a typing response and the T-cell status of the patient. However, a correlation with clinical course was suggested, because most cells which induced a typing response were obtained from patients who had received intensive treatment for the disease.     

Response of remission lymphocytes to autochthonous leukaemic myeloblasts.

Thymidine incorporation in vitro by remission lymphocytes from a total of 6 patients with acute myeloid leukaemia (AML) was measured following stimulation by autochthonous and allogeneic AML blasts and cell lines. The early peak response to autochthonous blasts in 2 of these patients (48-72 h) is consistent with the concept of a population of lymphocytes pre-immunized to antigens carried by the blasts. Although stimulation in one patient was increased in the presence of more stimulating (S) blasts than responding (R) lymphocytes, positive responses in other tests were obtained at an S : R ratio of 1 : 1-5. When different methods of treatment of the stimulating autochthonous blasts were compared with untreated cells, mitomycin C gave the highest stimulation indices 2 out of 3 tests. Tissue culture medium in which autochthonous blasts had been incubated for 3-5 days failed to stimulate either remission lymphocytes alone, or combined cultures of lymphocytes with autochthonous or allogeneic blasts, suggesting that mitogenic factors released from autochthonous blasts are not responsible for lymphocyte stimulation. Treatment of autochthonous or allogeneic AML blasts with glycine-HC1(pH 3-0) to remove putative "blocking" factors failed to increase the stimulatory capacity of the leukaemic blasts.     

Lymphocytotoxic T lymphocytes in a patient with B-chronic lymphocytic leukemia and pure red cell aplasia

The peripheral blood T cells of a hypertransfused patient with B-chronic lymphocytic leukemia and pure red cell aplasic were found to exhibit unusual spontaneous cytotoxic activity in vitro. The patient's E-rosette positive cells were cytotoxic for K562 (cultured human crythroleukemia cells) and allogeneic B and T lymphocytes freshly isolated from the peripheral blood of normal and CLL donors. They failed to kill autologous B cells, erythroid progenitors present in allogeneic bone marrow, and a number of cultured human tumor cells (Malme, CAKI) even after prolonged (36 h) co-culture. Peripheral blood T cells isolated from normal controls, other CLL patients, and hypertransfused individuals (n=13) (lid not exhibit spontaneous lymphocytotoxic activity. Circulating cytotoxic T cells having the ability to kill freshly isolated allogeneic lymphocytes have, heretofore, not been reported in humans. Our findings suggest that among this patient's peripheral blood T cells, there exists a subpopulation of lymphocytotoxic cells that closely resemble cytotoxic T cells generated in vitro after priming with allogeneic target cells. Although the lymphocytotoxic cells could have been induced in this patient by previous HLA-mismatched transfusions, it is possible they may have arisen spontaneously and underly the patient's erythroblastopenic state.     

Cyclic adenosine 3':5'-monophosphate levels and activities of related enzymes in normal and leukemic lymphocytes.

The role of cyclic adenosine 3':5'-monophosphate (cyclic 3':5'-AMP) in the regulation of cell division in lymphocytes obtained from healthy donors and from patients with chronic lymphocytic leukemia (CLL) was investigated by determining the levels of cyclic 3':5'-AMP and glycogen and also the activities of several enzymes that are closely associated with the metabolism of these cellular components. Intracellular levels of cyclic 3':5'-AMP were measured in normal and CLL lymphocytes in nondividing, dividing, and quiescent [after phytohemagglutinin (PHA) addition]states. In normal lymphocytes the levels of cyclic 3':5'-AMP fluctuated throughout the cell cycle after PHA addition, whereas in CLL lymphocytes the levels were approximately 3-fold lower than in normal cells and remained relatively constant before, during, and after mitogenic stimulation. Normal cells contained approximately 3-fold lower levels of glycogen than CLL cells, whereas glycogen phosphorylase activities were increased 2- to 4-fold above those in nondividing cells in normal but not in CLL lymphocytes after stimulation with PHA. Furthermore, cyclic 3':5'-AMP phosphodiesterase activities were higher in CLL lymphocytes than in normal ones. Collectively, these studies demonstrated that (a) the intracellular levels of cyclic 3':5'-AMP differ in these two cell types; (b) the levels of cyclic 3':5'-AMP and glycogen qualitatively correlate with the activities of the enzymes that are related to these components; and (c) an inverse relationship between the levels of cyclic 3':5'-AMP and cell growth exists in mitogen-stimulated lymphocytes from healthy donors but not from patients with CLL. These biochemical differences are presumed to exist between normal and "leukemic" lymphocytes, but alternatively they may reflect normal populations of immunologically distinct lymphocytes.