Synergistic cytotoxic effect of azidothymidine and recombinant interferon alpha on normal human bone marrow progenitor cells

Azidothymidine (AZT) and interferon alpha (IFN-alpha) are among the drugs showing strong in vitro activity against the human immunodeficiency virus type-1 (HIV-1). Each drug, however, has significant toxicity against normal marrow progenitor cells that frequently proves dose-limiting in patients. In this study, AZT and recombinant IFN-alpha 2a (rIFN-alpha 2a) were tested as single agents and in combination against normal myeloid (CFU-GM) and erythroid (BFU- E) colony forming cells in a standard methylcellulose culture assay. The data were analyzed using a quantitative computerized analysis based on the median-effect principle and the isobologram equation as described by Chou and Talalay (Adv Enz Regul 22:27, 1984). The ED90 for BFU-E and CFU-GM inhibition was then compared with previously measured in vivo plasma levels of each drug and the ED90 for the anti-HIV-1 effect in vitro. We demonstrate that (a) the drugs are strongly synergistic in inhibiting marrow progenitor cell growth and that this synergism occurs at drug levels that are within the range of measured plasma levels in phase I clinical trials, (b) BFU-E are more sensitive than CFU-GM to the inhibiting effects of AZT, rIFN-alpha 2a or both drugs in combination, (c) the drug concentrations in combination that synergistically inhibit bone marrow progenitors are much higher than those required to inhibit HIV-1 replication in vitro, and (d) the anti- HIV-1 effect for the combination of AZT and rIFN-alpha 2a was clearly superior to the effect of AZT or rIFN-alpha 2a alone as indicated by the combination index and the dose-reduction index. These data suggest that substantially lower doses of AZT and rIFN-alpha than those currently being tested in clinical trials might not only maintain a strong synergistic anti-HIV-1 effect but might also avoid significant hematologic toxicity.     

Differential effect of Serratia protease on platelet surface glycoproteins Ib and V

The effect of a zinc metalloprotease from Serratia marcescens on platelet surface glycoproteins (GP) Ib and V was analyzed. Increasing protease treatments caused progressive loss of GP Ib with appearance of the major fragment, glycocalicin, in the supernatant solution. No GP V was detected in the supernatant solution, and protease-pretreated platelets had the same capacity as control platelets to release fragment 1 of GP V in response to thrombin. The Serratia protease- pretreated platelets did show the lag before thrombin-induced dense granule secretion, characteristic of platelets modified by pretreatment with other nonstimulating proteases. Treatment with Serratia protease gives the only demonstrated selective loss of GP Ib without apparent effect on GP V. It suggests that GP V (1) does not depend on GP Ib for its association with platelets and (2) is not the substrate for protease modification of platelet function.     

Purification and properties of heparin-releasable lipoprotein- associated coagulation inhibitor

The lipoprotein-associated coagulation inhibitor (LACI) is present in vivo in at least three different pools: sequestered in platelets, associated with plasma lipoproteins, and released into plasma by intravenous heparin, possibly from vascular endothelium. In this study we have purified the heparin-relesable form of LACI from post-heparin plasma and show that it is structurally different from lipoprotein LACI. The purification scheme uses heparin-agarose chromatography, immunoaffinity chromatography, and size-exclusion chromatography and results in a 185,000-fold purification with a 33% yield. Heparin- releasable LACI (HRL), as analyzed by sodium dodecyl sulfate- polyacrylamide gel electrophoresis, under reducing conditions, appears as a major band at 40 Kd and a minor band at 36 Kd. Immunoblot analysis suggests that the 36-Kd band arises from carboxyl-terminus proteolysis that occurs during the purification. HRL has a specific activity similar to that of HepG2 or lipoprotein LACI. HRL and lipoprotein LACI combine with lipoproteins in vitro while purified HepG2 LACI does not. I125-labeled HRL, injected into a rabbit, is cleared more slowly than I125-labeled HepG2 LACI, which may be due to attachment to lipoproteins in vivo. Preliminary evidence suggests that HRL is associated with vascular endothelium, possibly by attachment to glycosaminoglycans.     

HIV type 1 subtype E-infected patients with broadened, dual (B/E) V3 loop serology have increased cross-neutralizing antibodies

The two prevalent subtypes of HIV-1 circulating in Thailand are subtypes E and B. While the most prevalent subtype continues to be E using molecular typing assays, immunologically, a subset of subtype E-infected patients (3.4% in 1997) have binding antibodies to both the E and B V3 loops in a peptide ELISA. To assess the potential function of this dual (B/E) V3 reactivity, plasmas from patients with genetically defined HIV-1 subtype E infection and either E or B/E V3 serotypes were compared for magnitude and breadth of neutralization of seven primary and laboratory-adapted subtype B and E viruses. Dually reactive (B/E) plasmas showed significantly increased cross-neutralizing activity against subtype B viruses (p < 0.001), and increased neutralization of the panel of viruses overall (p < 0.02), as compared to monoreactive E serotype plasmas. While the total envelope binding antibody titers to both subtype B and E envelopes did not differ significantly between the E and B/E plasmas, 67% of B/E plasmas neutralized >50% of the viruses in the panel, and only 14% of E plasmas showed this broadened neutralizing activity. These data suggest that dual (B/E) V3 loop reactivity may be a marker of broader immune recognition of HIV envelope epitopes in subtype E-infected patients. V3 loop antibody, perhaps in conjunction with antibodies to additional epitopes, may play a role in neutralization of virus isolates from Thailand.     

Lymphohematopoietic progenitors of normal mice

We have identified and characterized the lymphohematopoietic progenitors in the bone marrow of normal mice using a single-step methylcellulose culture assay. Lineage-negative Ly-6A/E (Sca-1)+ progenitors isolated from normal mice were plated in methylcellulose culture containing steel factor (SF), interleukin-7 (IL-7), erythropoietin (Ep), and IL-11. After 16 to 17 days of culture, pre-B- cell-containing multilineage myeloid colonies can be microscopically identified; however, flow-cytometric analysis of individual colonies for B220-positive cells proved superior to in situ microscopic identification of lymphomyeloid colonies. Approximately 10% (6/66) of the mixed colonies without a conspicuous B-cell component had B220- positive cells. The single cell origin of the lymphomyeloid colonies was confirmed by micromanipulation. Although the combination of SF, IL- 7, and Ep was sufficient to support formation of lymphomyeloid colonies, addition of IL-11, granulocyte colony-stimulating factor or IL-12 to the combination of SF, IL-7, and Ep increased the number of lymphomyeloid colonies. IL-1 alpha and IL-3 independently inhibited the expression of the B-lymphoid lineage when added to the combination of SF, IL-7, Ep, and IL-11. Approximately four times more lymphohematopoietic progenitors are present in normal mice than in mice treated with 5-fluorouracil.     

Effects of recombinant canine stem cell factor, a c-kit ligand, and recombinant granulocyte colony-stimulating factor on hematopoietic recovery after otherwise lethal total body irradiation

The effects of recombinant canine stem cell factor (rcSCF) on hematopoiesis were studied in normal dogs and in dogs given otherwise lethal total body irradiation (TBI) without marrow transplant. Results were compared with previous and concurrent data with recombinant granulocyte colony-stimulating factor (rG-CSF). Four normal dogs received 200 micrograms rcSCF per kilogram body weight daily either by continuous intravenous infusion for 28 days (n = 2) or by subcutaneous (SC) injection in two divided doses for 20 days (n = 2). All dogs showed at least a twofold increase in peripheral blood neutrophil counts starting approximately 7 days after the initiation of treatment. Hematocrit level and monocyte, lymphocyte, eosinophil, reticulocyte, and platelet counts were not elevated. Marrow sections after rcSCF treatment showed panhyperplasia. The only toxicity was facial edema during the first few days of rcSCF administration, presumably caused by mast cell stimulation. Ten dogs were given 400 cGy TBI at 10 cGy/min from two opposing 60Co sources. They were given no marrow infusion and received 200 micrograms/kg/d rcSCF SC in two divided doses for 21 days starting within 2 hours of TBI. Five of the 10 dogs showed complete and sustained hematopoietic recovery and survived as compared with 1 of 28 control dogs not receiving growth factor (P < .005). RcSCF treatment allowed for hematopoietic recovery in two of seven dogs administered 500 cGy of TBI but in none of five dogs given 600 cGy of TBI. Results with rcSCF are similar to those obtained with rG-CSF. The rate of neutrophil recovery in rcSCF-treated dogs after 400 cGy TBI was not different from that of rG-CSF-treated dogs (P = .65), but the rate of platelet recovery was faster (P = .06) in the rcSCF-treated animals. Combined treatment with rcSCF and rcG-CSF after 500 cGy TBI did not result in strongly improved survival as compared with results obtained with either factor alone.     

The Heartsink Patient: A Preliminary Study

Eight GPs identified 78 heartsink patients; in an open-endedinterview they were asked to explain why they regarded themin this way. A GP's definition of a heartsink patient was influencedby GP sex, practice location, and time of surgery, althoughthe number of participating GPs was too low to make any definiteassertions. Practitioners' anticipations of heartsink consultationswere generally over-exaggerated, with most of the encountersgoing better than expected. GPs expressed the view that thesepatients raised serious professional issues for them, whilstthere was also a dislike for these patients' personalities andbehaviour. Two levels of the heartsink state are hypothesized:one, a state of inertia, is when the heartsink patient has beena chronic high user of the primary health care system, and aGP has exhausted all avenues. The other is an acute situationwith those heartsink patients who have been low users of thesystem in the past. Recent, new events in these patients' liveshave raised an issue that is just as much to do with patientand doctor reaction to these events, as it is about findinga diagnosis or solution to the problem. We present the results and hypotheses to provoke further discussionand research.