Trypanosoma brucei brucei and high-density lipoproteins: old and new thoughts on the identity and mechanism of the trypanocidal factor in human serum

Nature has provided humans with a surprising means of protection against the African trypanosome Trypanosoma brucei brucei There is consensus, in that this singular trypanocidal factor is serum high-density lipoproteins (HDL). which the trypanosomes engulf through a physiological, receptor-mediated pathway for delivery to acidic intracellular vesicles. There is also controversy, however, in that the active particles and their essential cytotoxic elements are disputed, in part reflecting the ill-defined mechanism by which the parasites are finally killed. Here Patrick Lorenz, Bruno Betschart and Jim Owen discuss the possibilities for resolving these discrepancies and speculate on the prospects of exploiting this unexpected property of human HDL for protecting livestock.     

Purine requirements for the expression of Cape buffalo serum trypanocidal activity

Cape buffalo serum contains xanthine oxidase which generates trypanocidal H₂O₂ during the catabolism of hypoxanthine and xanthine. The present studies show that xanthine oxidase-dependent trypanocidal activity in Cape buffalo serum was also elicited by purine nucleotides, nucleosides, and bases even though xanthine oxidase did not catabolize those purines. The paradox was explained in part, by the presence in serum of purine nucleoside phosphorylase and adenosine deaminase, that, together with xanthine oxidase, catabolized adenosine, inosine, hypoxanthine, and xanthine to uric acid yielding trypanocidal H₂O₂. In addition, purine catabolism by trypanosomes provided substrates for serum xanthine oxidase and was implicated in the triggering of xanthine oxidase-dependent trypanocidal activity by purines that were not directly catabolized to uric acid in Cape buffalo serum, namely guanosine, guanine, adenine monophosphate, guanosine diphosphate, adenosine 3′:5-cyclic monophosphate, and 1-methylinosine. The concentrations of guanosine and guanine that elicited xanthine oxidase-dependent trypanocidal activity were 30–270-fold lower than those of other purines requiring trypanosome-processing which suggests differential processing by the parasites.     

A survey for a trypanocidal factor in primate sera

The sera of 21 different species of primates were surveyed for the presence of a trypanocidal factor to a monomorphic human serum-sensitive clone of Trypanosoma brucei gambiense (T.b.g.); human, gorilla, baboon (2 species), and the mandrill were found to contain this factor. The factor in all the sera is in the high density lipoprotein (HDL) fraction, and has similar modes of biological action. It has been shown that the human and gorilla trypanocidal factor share cross-reactive antigenic epitopes, but do not share similar cross-reactive epitopes with the baboon and mandrill factor. There was no relationship between the presence or absence of this factor and the primate's position on the phylogenetic tree. In addition, there was also no obvious correlation between the animals' preferred diet, and the presence or absence of trypanocidal activity. The evidence to date suggests that only African ground-dwelling primates that live in tsetse endemic areas contain the trypanocidal factor. It is assumed that this factor is involved in resistance of these primates to T.b.b. We believe that the host has developed trypanocidal substances as a result of selective evolutionary pressure by the African trypanosomes.     

The trypanolytic factor of human serum

African trypanosomes (the prototype of which is Trypanosoma brucei brucei) are protozoan parasites that infect a wide range of mammals. Human blood, unlike the blood of other mammals, has efficient trypanolytic activity, and this needs to be counteracted by these parasites. Resistance to this activity has arisen in two subspecies of Trypanosoma brucei - Trypanosoma brucei rhodesiense and Trypanosoma brucei gambiense - allowing these parasites to infect humans, and this results in sleeping sickness in East Africa and West Africa, respectively. Study of the mechanism by which T. b. rhodesiense escapes lysis by human serum led to the identification of an ionic-pore-forming apolipoprotein - known as apolipoprotein L1 - that is associated with high-density-lipoprotein particles in human blood. In this Opinion article, we argue that apolipoprotein L1 is the factor that is responsible for the trypanolytic activity of human serum.     

Nature of the serum heat-labile accessory factor involved in the metabolic inhibition of Mycoplasma pneumoniae

Using the metabolic-inhibition (MI) test as an assay, sera from rabbits immunized with Mycoplasma pneumoniae were separated on the basis of a requirement for a heat-labile (56 C) accessory factor (HLAF). As in previous studies, this requirement was found in sera collected early in the immune response, but disappeared as the immunization procedure progressed. Results obtained following exposure of HLAF-requiring sera to zymosan, hydrazine, and ethylene-diaminetetraacetic acid (EDTA) suggested that complement components C3, C8, and possible other components which follow C3 in the complement cascade, were required to demonstrate MI activity. Results obtained when HLAF-non-requiring sera were treated in the same manner, suggested the presence of two antibody populations. One population required component C3, and possibly C5, C6, and C9, while the other population required none of the complement components.     

Molecular feature of the nerve growth factor in human serum

High molecular weight binding components which bind [¹²⁵I] mouse β nerve growth factor exist in human serum. The binding of β nerve growth factor to the serum components was inhibited at alkaline condition. After gel filtration of human serum on a Sephadex G-150 column at neutral condition, the nerve growth factor-like immunoreactivity was observed in only one peak, differing from the high molecular weight serum components. However, at alkaline condition two peaks with nerve growth factor-like immunoreactivity appeared; one was almost at the position observed at neutral pH, and the other was a new peak eluted approximately to the column volume. these results suggest that there are at least two nerve growth factor-like molecules in human serum and most of the nerve growth factor in the serum exists in a complex form associated with serum components with high molecular weight.     

Heparin-binding properties of human serum spreading factor

Human serum spreading factor (SF) is a blood glycoprotein that promotes attachment and spreading and influences growth, migration, and differentiation of a variety of animal cells in culture. SF purified from human plasma or serum by chromatographic methods reported previously (Barnes, D. W., and Silnutzer, J. (1983) J. Biol. Chem. 258, 12548-12552) does not bind to heparin-Sepharose under conditions of physiological ionic strength and pH. In a further examination of the heparin-binding properties of human serum SF, we found that exposure of purified SF to 8 M urea altered several properties of the protein, including heparin affinity, and these alterations remained after removal of the urea from SF solutions. Urea-treated SF bound to heparin under physiological conditions, and salt concentrations of 0.4 M or higher were required for elution of urea-treated SF from heparin-Sepharose at pH 7.0. The alteration of heparin-binding properties of SF also was observed upon exposure of the protein to heat or acid. Treatment of SF with urea, heat, or acid resulted additionally in greatly decreased cell spreading-promoting activity of the molecule. The decreased biological activity was associated with a reduced ability of the treated SF to bind to the cell culture substratum, a prerequisite for the attachment-promoting activity of the molecule. Experiments examining the heparin-binding properties of native SF in unfractionated human plasma indicated that the major portion of SF in blood did not bind to heparin under conditions of physiological ionic strength and pH.     

Nature of the inhibitory serum factor in mice injected with isologous antibodies conjugated with cellulose

The formation of antigen-specific serum inhibitory factor was induced by injection of covalently bound to cellulose syngeneic antibodies to sheep red blood cells into (CBA X C57BL/6)F1 mice. This factor was absorbed by cellulose immunosorbents immobilized with antibodies against sheep red blood cells and with rabbit antibodies against mouse gamma-globulin and was not absorbed by immunosorbents immobilized with immunoglobulins of intact mice or immunoglobulin containing antibodies against rat red blood cells. These data, and evidence obtained by the authors previously, indicate that inhibitory factor of the serum is likely to be due to idiotypic antibodies.     

Immunological studies of the anticryptococcal factor of normal human serum

Preliminary studies have shown a very high inhibitory activity in the alpha₂ and gamma zone of human serum towards the growth of Cryptococcus neoformans. These findings are now corroborated by single radial immunodiffusion tests, which showed the some loss of IgA and IgM globulins and of the other three globulin fractions (ceruloplasmin, alpha₂ macroglobulin and alpha₂ HS glycoprotein) which migrate in the alpha₂ zone. The data was obtained by single radial immunodiffusion tests. The losses were not statistically significant however. No change in the immunoglobulin content of the sera kept for 6 days in contact with a heat-killed suspension of C. neoformans was noted. These findings suggest, that the inhibitory activity of the normal human serum on the in-vitro growth of C. neoformans is due to the above mentioned globulin fractions and not to a single specific factor.     

Characterization of the protein which binds insulin-like growth factor in human serum

The binding of the 125I-labelled insulin-like growth factors I and II (125I-IGF I and 125I-IGF II) to the high-molecular-mass binding protein of human serum was characterized. With diluted human serum both growth factors showed optimal specific binding at 4 degrees C and pH 5-6. When 0.1% Triton X-100 was present in the incubation buffer an increase in the affinity of the IGF-binding protein was induced, which produced an enhanced binding of IGF I and IGF II. Competition experiments with various peptide hormones revealed that the native IGF-binding protein complex binds both the IGF I and IGF II with high specificity. Analysis of binding data according to the method of Scatchard resulted in linear plots for IGF I and IGF II respectively, indicating that in human serum only a single class of non-interacting binding sites is present. At optimal binding conditions the dissociation constants were determined to be 0.28 x 10(-9) M for IGF I binding and 0.66 x 10(-9) M for IGF II. Human serum was gel-filtered on Sepharose CL-6B at neutral pH and the eluate was assayed for binding activity with both IGF I and IGF II. One peak with an apparent molecular mass of 175 kDa and a Stokes radius of 4.8 nm was determined for both growth factors. Thus, our data suggest that human serum contains one class of high-molecular-mass binding protein with comparable binding characteristics for IGF I and IGF II.     

Inhibitory factor of phospholipase A2 in normal human serum

Normal human serum and plasma were shown to contain a factor inhibiting phospholipase A2. This factor has been separated from human serum and plasma by chromatography on a Blue-Ultrogel column and was eluted by tris-HCl buffer (pH 7.2); the proteins eluted by 1 M NaCl-tris HCl buffer exhibited phospholipase A2 activity. This activity was abolished when the inhibitory factor was added to proteins possessing such activity. The inhibitory factor was not dialysable, sensitive to both heat and trypsin treatment, suggesting that it is a protein. In vitro, the same factor inhibited phospholipase A2 rat serum.     

Preparation of highly purified F-actin-depolymerizing factor of human serum

F-actin depolymerizing factor (ADF) of human serum was purified 250-400-fold to more than 98% purity with high reproducibility. The purification included (1) 30-50% (NH4)2SO4 precipitation, (2) ion-exchange chromatography on DEAE-Sepharose, (3) chromatofocusing on Polybuffer exchanger 94 and (4) affinity chromatography on ConA-Sepharose. The recovery of ADF was estimated to be 20-30% whereas the ADF activity yield was 5-17%. The lower activity yield was thought to be due-partly to proteolysis and partly to destabilization of highly purified ADF.     

Thymosin-induced human serum factor increasing cyclic AMP.

Thymosin has virtually no in vitro effect on cyclic AMP levels in thymocytes and seems not to react with either the beta-adrenergic receptor or with the putative human serum thymic factor (SF) receptor. However, when thymosin is injected into patients lacking SF activity, it induces the appearance of a serum factor which increases cellular cyclic AMP levels in thymocytes in vitro. This factor appears to act on a membrane site distinct from the beta-adrenergic receptors.     

Soluble inhibitory factor (SIF) in normal human serum

We have previously noted that dividing T cells release soluble inhibitory factor (SIF) into culture fluids. SIF was distinguished from most other suppressor factors by consisting of two components: a protein and a glycolipid, lipid suppressor substance (LSS). We had noted large quantities of LSS in serum of a patient with a cutaneous lymphoma. This prompted the present study in which SIF was found in normal human serum in a fraction derived by ethanol precipitation. The SIF complex reduced uptake of [³H]thymidine into mixed leukocyte culture (MLC) by 77%. SIF from serum (SIF-serum) resembled SIF in culture fluids (SIF-sup) by chromatography and function at all stages of purification. LSS was extracted from SIF-serum, as measured by an 88% reduction of [³H]thymidine incorporation into phytohemagglutinin-stimulated lymphocytes. The apparent MW of SIF was between 100, 000 and 150, 000. LSS from serum was purified by two-dimensional thin-layer chromatography to apparent homogeneity. The presence of SIF in normal human serum suggests that it may have an in vivo role in immune regulation.     

Characterization of human epidermal growth factor in human serum and urine under native conditions

The objective of this study was to investigate the molecular nature of the human epidermal growth factor (EGF) in serum and urine samples of normal subjects. Recombinant EGF emerged as a single peak and did not interact with human IgG1 and albumin up to the concentration of 12 microg/ml. Freshly separated human serum contained only trace amounts of EGF. However, EGF appeared and increased in serum separated from blood after spontaneous overnight clotting. The authentic 6 kDa form of EGF made up nearly 40% of the total EGF in serum and revealed relatively homogeneous feature. The remaining immunoreactive fractions corresponded to 160 kDa proEGF. Immunoreactive EGF in blood seemed to be associated with the EGF release from platelets. TSKgel G3000SW chromatography of freshly-voided morning and day urines revealed that urine samples mainly contained two major form of EGF; a high-molecular-weight (HMW) and low-molecular-weight (LMW) forms. In the sense of molecular nature of EGF contents, morning urine was more heterogeneous than day urine of the same individuals. The LMW form of EGF in morning urine, in which its proportion was more than 90% of the total EGF, revealed further heterogeneous feature generally containing three to four different components.     

Isolation of tumor disturbing factor on the proliferation of tumor cells in human serum

The classified sediment with ethanol from sera of nude mice and humans showed a disturbing effect on L1210 cells in vitro and a lifesaving effect on L1210 cell-bearing mice in vivo. This factor was purified more than 2300-fold to a specific activity of approximately 1 X 10(5) U/mg by ethanol classified precipitation, Sephadex G-200 gel filtration, DEAE cellulose ion exchange chromatography with Nacl and pH gradient aqueous solution, and preparative polyacrylamide gel electrophoresis.     

Levels of human serum granulocyte colony-stimulating factor and granulocyte-macrophage colony-stimulating factor under pathological conditions

Levels of serum granulocyte colony-stimulating factor (G-CSF) and granulocyte-macrophage colony stimulating factor (GM-CSF) in patients with various leukocyte disorders were estimated by enzyme linked immunosorbent assay (ELISA). Some cases of acute myelogenous leukemia and aplastic anemia showed elevated serum levels of G-CSF and/or GM-CSF, whereas almost all of 23 healthy controls showed G-CSF and GM-CSF levels lower than 100 pg/ml. High levels of both types of CSF were noted in patients with granulocytosis due to infection. These levels became lower after resolution of the infection. Daily changes in serum CSF levels were also examined in a patient with autoimmune neutropenia, and it was found that the peripheral neutrophilic granulocyte count changed almost in parallel with the serum G-CSF level but not with GM-CSF, following the pattern with a delay of about 4–5 h, suggesting the possibility that G-CSF mainly regulates peripheral neutrophil circulation.