Deformation of swollen erythrocytes provides a model of sickling- induced leak pathways, including a novel bromide-sensitive component

Deoxygenation-induced red blood cell (RBC) sickling probably activates multiple cation leak pathways. In an attempt to model this, we examined the net passive K efflux ("K leak") from normal and sickle RBCs undergoing elliptical deformation in hypotonic media (200 mOsmol/L). This hypotonic deformation activates two deformation-dependent K leak pathways that are not detectable during the balanced leak (Kefflux = Nainflux) resulting from deformation of RBCs in isotonic medium. These are (1) a calcium-dependent leak component and (2) a novel leak pathway that is inhibited by substitution of bromide (but not sulfamate) for chloride, which converts the unbalanced K leak (Kefflux > Nainflux) of hypotonic deformation to a residual balanced leak. This dramatic effect of hypotonic deformation is reversible, is detected in both normal and sickle RBCs, and is inhibited significantly by 4,4'-diisothiocyano-2,2'- stilbene disulfonate. Remarkably, bromide also inhibits by 55% the K leak resulting from authentic deoxygenation-induced RBC sickling and, thereby, blunts the imbalance of accompanying monovalent cation leaks. The unique effect of bromide is not readily explainable on the basis of known behaviors of known ion leak/transport pathways. The mechanical threshold for triggering K leak during hypotonic deformation is at applied shear stress of 164 dyne/cm2, a value similar to the abnormal susceptibility we previously found for oxygenated sickle RBCs during isotonic deformation. These data suggest that membrane stretch accompanying hypotonic deformation activates the same multiple leak pathways that contribute to net K leak during authentic RBC sickling, including a previously unknown bromide-sensitive leak.     

Murine thymocytes proliferate in direct response to interleukin-7

The ability of interleukin-7 (IL-7) to stimulate murine thymocyte proliferation was investigated. IL-7, either alone or in concert with lectin, induced proliferation of adult thymocytes as well as day 13 fetal and adult CD4-/CD8-thymocytes. The IL-7-induced proliferative response of unfractionated thymocytes could not be inhibited by antibodies to IL-2, or IL-4, IL-6, or the IL-2 receptor. In addition, IL-2, IL-4, and IL-6 were not produced by thymocytes activated with IL- 7, as judged by the absence of biologically active cytokine in IL-7- stimulated culture supernatants. IL-7 could act in concert with IL-2 and IL-4 or with IL-4 to enhance the proliferative response of thymocyte cultures. Thus, IL-7 may cause proliferation of thymocytes directly, not indirectly, through production of IL-2, IL-4, or IL-6. IL- 7 may then play a significant role in differentiation of T lymphocytes.     

Growth and development following marrow transplantation for leukemia

One hundred forty-two patients between the ages of 1 and 17 years who survived disease-free more than 1 year after marrow transplantation for hematologic malignancy had growth and development evaluations from one to 14 years posttransplant (median 4 years). Prior to transplant all children received multiagent chemotherapy and 55 also received central nervous system irradiation, but none had growth and development evaluations. Marrow transplant preparation included high-dose chemotherapy and total body irradiation (TBI) given as a single dose of 9.2 to 10.0 Gy (79 patients) or as fractionated doses of 2.0 to 2.25 Gy/d for six to seven days (63 patients). After transplant abnormal thyroid function was present in 39%. Stimulated 11-desoxycortisol levels were subnormal in 24% of patients evaluated. Growth hormone (GH) deficiency was present in 17 of 25 children who received previous cranial irradiation. Partial GH deficiency was present in 4 of 25 who received previous cranial irradiation and in 6 of 18 who had not received cranial irradiation. Height velocity was decreased in all patients. After transplant, height was significantly influenced by chronic graft-v-host disease and single-dose TBI. Sixty-eight percent had delayed development of secondary sexual characteristics. Gonadal failure occurred in nearly all who were postpubertal at transplant. While it is not possible to determine how many of these endocrine abnormalities occurred as a result of treatment administered prior to transplantation, these data do demonstrate that children who become long-term survivors after marrow transplantation for hematologic malignancy have endocrine abnormalities that adversely affect growth and development.     

Immunohistochemical localization of membrane and alpha-granule proteins in plastic-embedded mouse bone marrow megakaryocytes and murine megakaryocyte colonies

Using an immunoperoxidase technique that permits optimal antigen localization at the light microscope level, we have detected two platelet alpha-granule constituents and three platelet membrane glycoproteins in mouse bone marrow megakaryocytes and in murine megakaryocyte colonies grown in soft agar culture for three to seven days. Using polyclonal antibodies prepared against human platelet proteins, we have demonstrated labeling for von Willebrand factor, fibrinogen, and the membrane glycoproteins IIIa and GMP-140 in both bone marrow megakaryocytes and megakaryocyte colonies after seven days of culture. Using monoclonal antibodies to membrane glycoproteins IIb and GMP-140, we have demonstrated label in mouse bone marrow megakaryocytes. Granulocyte and macrophage colonies were negative for each of these markers. Murine bone marrow megakaryocytes and megakaryocyte colonies demonstrated a similar enzyme histochemical pattern: weakly positive for alpha-naphthyl acetate esterase and negative for chloroacetate esterase. These data indicate that megakaryocytes grown in soft agar culture express many of the same glycoproteins as bone marrow megakaryocytes. Furthermore, the ability of antibodies directed against human platelet membrane glycoproteins to identify murine megakaryocyte glycoproteins indicates that these constituents have been highly conserved during evolution.     

Methotrexate, cyclosporine, or both to prevent graft-versus-host disease after HLA-identical sibling bone marrow transplants for early leukemia?

Optimal prophylaxis of graft-versus-host disease (GVHD) is controversial. We compared efficacy of three posttransplant immune suppressive regimens in 2,286 recipients of HLA-identical sibling bone marrow transplants for acute lymphoblastic leukemia (ALL) in first remission, acute myelogenous leukemia (AML) in first remission, or chronic myelogenous leukemia (CML) in first chronic phase. Six hundred forty received methotrexate, 977 received cyclosporine, and 669 received combined cyclosporine and methotrexate. In children, the three regimens resulted in similar outcomes. In adults, cyclosporine and methotrexate had comparable risks of acute and chronic GVHD. Compared with methotrexate, cyclosporine was associated with less interstitial pneumonia (relative risk [RR] = 0.6; P < .001), less treatment-related mortality (RR = 0.6; P < .001), more relapses (RR = 1.6; P < .05), and less treatment failure (relapse or death from any cause; RR = 0.7; P < .001). Different effects were observed in different leukemias. In ALL, the rate of leukemia relapse was increased with cyclosporine versus methotrexate, with no effect on other outcomes. In AML and CML, interstitial pneumonia, treatment-related mortality, and treatment failures were decreased with cyclosporine, with no increase in relapse. Similar analyses comparing cyclosporine plus methotrexate with cyclosporine alone showed that adults receiving the combination had less acute GVHD (RR = 0.5; P < .001), less chronic GVHD (RR = 0.7; P < .01), and less interstitial pneumonia (RR = 0.7; P < .001). Treatment failure (RR = 0.8; P < .05) was marginally reduced. Separate analyses in ALL and AML showed less acute GVHD with combined therapy, but no significant effect on other outcomes. In CML, acute GVHD, interstitial pneumonia, treatment-related mortality, and treatment failure were decreased with combined therapy.     

P-selectin glycoprotein ligand-1 (PSGL-1) is a ligand for L-selectin in neutrophil aggregation

In inflammation, activated neutrophils adhere to endothelial cells and aggregate with one another. While beta 2-integrin and L-selectin are essential for aggregation, their ligands remain to be identified. We have previously shown that L-selectin mediates a carbohydrate-dependent interaction in aggregation (Simon et al: J Immunol 149:2765, 1992; Rochon et al: J Immunol 152:1385, 1994). We have suggested that the L- selectin counter-structure is a mucinlike protein and proposed that aggregation occurs through a two-step process involving L-selectin, beta 2-integrin, and their distinct counter-structures (Bennett et al: J Leuk Biol 58:510, 1995). A candidate ligand for L-selectin is P- selectin glycoprotein ligand-1 (PSGL-1), a mucinlike protein on neutrophils that binds P-and E-selectin. Using flow cytometry we show that the number and size of neutrophil aggregates is reduced with Fab fragments of PL1, an anti-PSGL-1 monoclonal antibody that blocks the interaction between P-selectin and PSGL-1 (Moore et al: J Cell Biol 128:661, 1995). In addition, monoclonal antibodies to L-selectin and PSGL-1 were used simultaneously to modulate the availability of these adhesion molecules on individual cell populations. The inhibition of aggregation by these antibodies is consistent with L-selectin and PSGL- 1 being counter-structures. We suggest that L-selectin and PSGL-1 support a collisional cell-cell interaction that represents the first step in neutrophil aggregation.     

Lipid hydroperoxides permit deformation-dependent leak of monovalent cation from erythrocytes

Subtle peroxidative perturbation of normal red blood cells (RBC) using t-butylhydroperoxide creates a leak pathway for monovalent cations that is reversibly activated by cell deformation. To determine what factor promotes expression of this unique membrane defect, we have dissected "peroxidation" into components that can be evaluated separately by comparing K leak from suitably modified RBC during elliptical deformation and parallel control incubation. Selective introduction of phospholipid hydroperoxides into normal RBC membranes successfully induces a deformation-dependent leak pathway having the same phenomenology as that previously documented for cells treated with t- butylhydroperoxide itself (fully recoverable; calcium-independent; inhibited at lower pH; K efflux balanced by Na influx). This leak pathway occurs in the absence of detectable secondary peroxidative change and appears to reflect a direct influence of lipid hydroperoxide. Using micropipette examination of vesicular bilayers reconstituted from RBC lipid extracts, we find that lipid from peroxidized RBC exhibits only a slight tendency to be less cohesive than normal lipid, apparently precluding isolated lipid properties as an explanation for altered permeability barrier function. However, addition of a hydrophobic membrane-spanning peptide to these same lipids significantly diminishes bilayer cohesion, an effect that is exacerbated further by the presence of peroxidized lipid. These observations suggest that lipid hydroperoxide is a necessary, but perhaps not sufficient, factor for induction of this unique leak pathway. Our results may be relevant to the abnormal cation homeostasis of sickle RBC in which deformation of an oxidatively perturbed membrane occurs during the sickling phenomenon.     

Anti-Jka, -C, and -E in a single patient, initially demonstrable only by the manual hexadimethrine bromide (Polybrene) test, with incompatibilities confirmed by 51Cr-labeled red cell studies

Published reports have confirmed the superior sensitivity of the manual hexadimethrine bromide (Polybrene) test (MPT) for demonstrating many alloantibodies in vitro; however, the clinical significance of alloantibodies demonstrable exclusively by MPT has not been shown conclusively. A patient with macroglobulinemia experienced chills, fever, hemoglobinemia, and hemoglobinuria following the transfusion of 1 unit of red cells (RBCs) shown to be compatible by the low-ionic-strength antiglobulin (LIS-AG) method. Serologic investigation was negative. Intravascular hemolysis occurred with a second "compatible" unit. Serologic studies were again negative by LIS-AG and ficin-AG methods, but revealed anti-Jka by MPT. Both donors were Jk(a+b-), and 51Cr studies of the second donor's RBCs revealed a t1/2 of less than 30 minutes, with marked intravascular hemolysis. A LIS-AG-compatible Jk(a-) unit was transfused uneventfully, but with no rise in hematocrit. MPT next revealed anti-C; subsequent 51Cr studies with the Jk(a-), Cc donor's RBCs showed a 51Cr t1/2 of 100 minutes with slight intravascular lysis. Four transfusions of Jk(a-), C- blood were uneventful, but 5 days later the patient's hemoglobin declined. The following day, anti-E was demonstrable exclusively by MPT. 51Cr-labeled Jk(a-), C-, E- RBCs had normal 24-hour survival. The patient's hemoglobin rose to 11 g per dl following transfusions of Jk(a-), C-, E- RBCs, and he was discharged. In vitro studies employing the patient's purified IgM paraprotein revealed no interference with alloantibody binding or detection.