Faculty Bibliography

The natural immunity of humans to the cattle pathogen Trypanosoma brucei brucei, but not to the morphologically indistinguishable human pathogens T. brucei gambiense and T. brucei rhodesiense, is due to the selective killing of the parasite by normal human serum. The factor in human serum that mediates lysis of T. brucei brucei has long been attributed to a minor subclass of high density lipoprotein (HDL). Evidence indicates that the trypanolytic activity of isolated human HDL is due to peroxidase activity of an associated haptoglobin-related protein-hemoglobin complex. However, recent data suggest that the trypanolytic activity of HDL may be completely inhibited in whole human serum, and that trypanolytic activity of norman human serum is due to a second, less well-defined factor of high molecular weight. Current research aimed at understanding the mechanisms of cytotoxicity and the affected metabolic pathways may open new approaches for the development of specific drugs and vaccines against trypanosomiasis

Natural immunity of humans to the cattle pathogen Trypanosoma brucei brucei has been attributed to the presence in normal human serum (NHS) of lytic factors for the parasites. We and others have shown that NHS contains two trypanolytic factors (herein termed TLF1 and TLF2) that can be separated by gel filtration. TLF1 copurifies with a subclass of high density lipoprotein (HDL), whereas TLF2 has a much higher molecular weight and does not appear to be a lipoprotein. We find that the trypanolytic activity of purified TLF1 is totally inhibited by exogenous haptoglobin (Hp) at concentrations (0.1 mg/ml) lower than those present in NHS (0.2-2 mg/ml). In contrast, exogenous Hp (up to 2.5 mg/ml) has no effect on the lytic activity of either NHS or isolated TLF2. Hp-depleted sera from patients with intravascular hemolysis is severalfold more trypanolytic than NHS. These sera contain only TLF1, and their lytic activity is totally abolished upon the addition of Hp (0.1 mg/ml). When NHS containing different Hp allotypes is fractionated by gel filtration, TLF1 activity is either revealed or remains masked, depending on whether it coelutes with Hp. Masked TLF1 activity in the column fractions is revealed if Hp is removed by density gradient ultracentrifugation. We conclude that endogenous Hp inhibits TLF1 activity, and that TLF2 is the main trypanolytic factor in NHS

In view of the importance of the low-density lipoprotein (LDL)-receptor in Trypanosoma brucei, we have examined whether other bloodstream trypanosomes of medical and veterinary importance (T.b. rhodesiense, T. equiperdum, T. vivax, T. congolense), but also related parasites developing in mammalian (Leishmania donovani) and non-mammalian hosts (Crithidia luciliae and Phytomonas sp. isolated from Euphorbia), would possess an LDL-receptor of their own. (1) All these parasites specifically accumulate human 125I-LDL with a relatively 2.5-fold higher rate for bloodstream trypanosomes. (2) A mixture of monoclonal antibodies raised against T.b. brucei LDL-receptor inhibit binding of LDL to all species but with different efficiency. (3) A single glycoprotein of similar M(r) (gp145) is isolated by LDL-affinity chromatography from all the above species, as well as from both human serum-resistant and sensitive strain of T.b. rhodesiense, and from the bodonid member of the Kinetoplastida Trypanoplasma borelli. (4) Several control experiments including 35S-metabolic labeling of procyclic T.b. brucei and of C. luciliae followed by LDL-affinity chromatography or immunoprecipitation demonstrate that gp145 is indeed synthesised by the parasites and is not a contaminant of the experimental system. (5) In immunoblots and ELISA, these gp145 cross-react with the polyclonal and monoclonal antibodies raised against the LDL-receptor of T.b. brucei, the highest degree of cross-reactivity being found among the members of the Trypanozoon subgroup. (6) Finally, immunisation of mice with the purified LDL-receptor from one strain of T.b. brucei is not sufficient to confer durable protection against another strain of this parasite

The variant surface glycoprotein of African trypanosomes has a glycosyl phosphatidylinositol (GPI) anchor that is unusual in that its fatty acids are exclusively myristate. We showed previously that the myristate is added to a free GPI in a fatty acid remodeling reaction involving deacylation and reacylation, forming glycolipid A, the anchor precursor. We now demonstrate that trypanosomes have a second pathway for GPI anchor myristoylation distinct from the fatty acid remodeling pathway, which we call 'myristate exchange.' This reaction involves exchange of myristate into both the sn-1 and sn-2 positions of glycolipid A, which already contain myristate. Myristoyl-CoA, the probable myristate donor in the exchange reaction, has an apparent Km of about 6 nM. We have now identified a lyso-GPI, named theta', which has myristate as its sole fatty acid; the kinetics of formation and utilization of theta' are consistent with it being an intermediate in exchange. Myristate exchange and fatty acid remodeling appear to occur in different subcellular compartments, and the two reactions have different sensitivities to inhibitors. The myristate exchange reaction may be a proofreading system to ensure that the fatty acids on variant surface glycoproteins are exclusively myristate

The glycosylphosphatidylinositol membrane anchor of variant surface glycoprotein of the African trypanosome Trypanosoma brucei contains several mannosyl residues for which dolichol phosphoryl mannose is supposed to be the precursor; this itself is probably synthesised by a dolichol-dependent mannosyltransferase. We have characterised and localised a mannosyltransferase activity of T. brucei which transfers mannose from GDP-[14C]mannose to exogenously added dolichyl phosphate. The enzyme was saturable for both its substrates and had a Km of 7.8 microM and 3.3 microM, respectively, for dolichyl phosphate and GDP-mannose. Mannosyltransferase was labile at 37 degrees C in the presence of Triton X-100, but its activity remained constant for at least 60 min at temperatures between 10-15 degrees C. The enzyme was inhibited by amphomycin and this inhibition was potentiated by the presence of 10 mM CaCl2. After subcellular fractionation of cell homogenates by differential centrifugation, mannosyltransferase was recovered mainly in the microsomal fraction and its distribution was very similar to that of RNA, a marker for the rough endoplasmic reticulum. After isopycnic centrifugation in a linear sucrose gradient the distribution of mannosyltransferase also resembled that of RNA. Both constituents exhibited a shift towards lower densities after pre-treatment of microsomal membranes with inorganic pyrophosphate, while other membrane markers such as acid phosphatase and nucleoside diphosphatase did not. It is concluded that the formation of dolichol phosphoryl mannose from GDP-mannose and dolichyl phosphate in T. brucei occurs mainly in the rough endoplasmic reticulum