Functional characterisation of a nematode secreted GM2-activator protein.

We have identified a GM2-activator protein (GM2AP) with highly unusual properties secreted by the nematode parasite Trichinella spiralis. Expression in Pichia pastoris resulted in a hyperglycosylated protein of 28 kDa, but the 18 kDa native protein was not glycosylated. The parasite GM2AP does not facilitate degradation of GM2 ganglioside by N-acetyl-beta-hexosaminidase A, although it does inhibit phospholipase D activity. Lack of the former activity might be explained by the absence of a domain implicated in binding to hexosaminidase. In addition, and contrary to data on the human GM2AP, the nematode homologue does not inhibit platelet activating factor-induced calcium mobilisation in neutrophils, but actually enhances mediator-induced chemotaxis.

Trichinella spiralis secretes a homologue of prosaposin.

Infective larvae and adult stage Trichinella spiralis secrete a protein homologous to prosaposin, the precursor of sphingolipid activator proteins (saposins) A-D originally defined in vertebrates. The protein contains four saposin domains, with the six cysteine residues which form the three intramolecular disulphide bonds in close register in each case. It differs substantially from vertebrate prosaposins in the N-terminal prodomain, the region separating saposins A and B, and completely lacks the C-terminal domain which has been demonstrated to be essential for lysosomal targetting in these organisms. The protein is secreted in unprocessed form with an estimated mass of 56 kDa, and contains a single N-linked glycan which is bound by the monoclonal antibody NIM-M1, characteristic of the TSL-1 antigens which are capped by tyvelose (3,6-dideoxy-D-arabinohexose). Immuno-electron microscopy localised the protein to membrane-bound vesicles and more complex multi-lamellar organelles in diverse tissues including the hypodermis, intestine and stichosomes, although it was absent from the dense-core secretory granules typical of the latter. Possible functions of a secreted prosaposin are discussed.

An ex vivo culture model for screening drug activity against in vivo phenotypes of Mycobacterium tuberculosis.

Since the activity of drugs against Mycobacterium tuberculosis grown in microbiological culture can differ from their activity against bacteria present in infected tissues, compounds with optimal activity against in vivo phenotypes may be overlooked in drug-discovery programmes that rely on in vitro screens. The authors have investigated the use of an ex vivo cell-culture model to assess the action of drugs on M. tuberculosis in an environment resembling that encountered during infection. Mycobacterial viability in the ex vivo model was shown to be regulated by the cell-mediated immune system, with growth inhibited by CD4(+) T cells at an early stage of infection in BCG-vaccinated mice, and at a later stage after infection in naive mice. Screening of drugs in the ex vivo model demonstrated a window of pyrazinamide susceptibility that coincides with the onset of the T-cell-mediated immune response in naive or vaccinated mice. It is proposed that pyrazinamide acts on a population of bacteria that are exposed to an acidic environment as a result of immune activation. Clinically, administration of pyrazinamide during the initial phase of treatment reduces the risk of relapse after 6 months, suggesting that the early pyrazinamide-susceptible population may contribute to the later pool of mycobacteria that persist during prolonged chemotherapy.