Killing of trypanosomatid parasites by a modified bovine host defense peptide, BMAP-18.

BACKGROUND: Tropical diseases caused by parasites continue to cause socioeconomic devastation that reverberates worldwide. There is a growing need for new control measures for many of these diseases due to increasing drug resistance exhibited by the parasites and problems with drug toxicity. One new approach is to apply host defense peptides (HDP; formerly called antimicrobial peptides) to disease control, either to treat infected hosts, or to prevent disease transmission by interfering with parasites in their insect vectors. A potent anti-parasite effector is bovine myeloid antimicrobial peptide-27 (BMAP-27), a member of the cathelicidin family. Although BMAP-27 is a potent inhibitor of microbial growth, at higher concentrations it also exhibits cytotoxicity to mammalian cells. We tested the anti-parasite activity of BMAP-18, a truncated peptide that lacks the hydrophobic C-terminal sequence of the BMAP-27 parent molecule, an alteration that confers reduced toxicity to mammalian cells. METHODOLOGY/PRINCIPAL FINDINGS: BMAP-18 showed strong growth inhibitory activity against several species and life cycle stages of African trypanosomes, fish trypanosomes and Leishmania parasites in vitro. When compared to native BMAP-27, the truncated BMAP-18 peptide showed reduced cytotoxicity on a wide variety of mammalian and insect cells and on Sodalis glossindius, a bacterial symbiont of the tsetse vector. The fluorescent stain rhodamine 123 was used in immunofluorescence microscopy and flow cytometry experiments to show that BMAP-18 at low concentrations rapidly disrupted mitochondrial potential without obvious alteration of parasite plasma membranes, thus inducing death by apoptosis. Scanning electron microscopy revealed that higher concentrations of BMAP-18 induced membrane lesions in the parasites as early as 15 minutes after exposure, thus killing them by necrosis. In addition to direct killing of parasites, BMAP-18 was shown to inhibit LPS-induced secretion of tumour necrosis factor alpha (TNF-alpha), a cytokine that is associated with inflammation and cachexia (wasting) in sleeping sickness patients. As a prelude to in vivo applications, high affinity antibodies to BMAP-18 were produced in rabbits and used in immuno-mass spectrometry assays to detect the intact peptide in human blood and plasma. CONCLUSIONS/SIGNIFICANCE: BMAP-18, a truncated form of the potent antimicrobial BMAP-27, showed low toxicity to mammalian cells, insect cells and the tsetse bacterial symbiont Sodalis glossinidius while retaining an ability to kill a variety of species and life cycle stages of pathogenic kinetoplastid parasites in vitro. BMAP-18 also inhibited secretion of TNF-alpha, an inflammatory cytokine that plays a role in the cachexia associated with African sleeping sickness. These findings support the idea that BMAP-18 should be explored as a candidate for therapy of economically important trypanosome-infected hosts, such as cattle, fish and humans, and for paratransgenic expression in Sodalis glossinidius, a bacterial symbiont in the tsetse vector, as a strategy for interference with trypanosome transmission.

Predicted function of the vaccinia virus G5R protein.

MOTIVATION: Of the approximately 200 proteins that have been identified for the vaccinia virus (VACV) genome, many are currently listed as having an unknown function, and seven of these are also found in all other poxvirus genomes that have been sequenced. The G5R protein of VACV is included in this list, and to date, very little is known about this essential and highly conserved protein. Conventional similarity searches of protein databases do not identify significantly similar proteins, and experimental approaches have been unsuccessful at determining protein function. RESULTS: Using HHsearch, a hidden Markov model (HMM) comparison search tool, the G5R protein was found to be similar to both human and archaeal flap endonucleases (FEN-1) with 96% probability. The G5R protein structure was subsequently successfully modeled using the Robetta protein structure prediction server with an archaeal FEN-1 as the template. The G5R model was then compared to the human FEN-1 crystal structure and was found to be structurally similar to human FEN-1 in both active site residues and DNA substrate binding regions.