Short Antimicrobial Peptide Derived from the Venom Gland Transcriptome of Pamphobeteus verdolaga Increases Gentamicin Susceptibility of Multidrug-Resistant Klebsiella pneumoniae.

Infectious diseases account for nine percent of annual human deaths, and the widespread emergence of antimicrobial resistances threatens to significantly increase this number in the coming decades. The prospect of antimicrobial peptides (AMPs) derived from venomous animals presents an interesting alternative for developing novel active pharmaceutical ingredients (APIs). Small, cationic and amphiphilic peptides were predicted from the venom gland transcriptome of Pamphobeteus verdolaga using a custom database of the arthropod's AMPs. Ninety-four candidates were chemically synthesized and screened against ATCC® strains of Escherichia coli and Staphylococcus aureus. Among them, one AMP, named PvAMP66, showed broad-spectrum antimicrobial properties with selectivity towards Gram-negative bacteria. It also exhibited activity against Pseudomonas aeruginosa, as well as both an ATCC® and a clinically isolated multidrug-resistant (MDR) strain of K. pneumoniae. The scanning electron microscopy analysis revealed that PvAMP66 induced morphological changes of the MDR K. pneumoniae strain suggesting a potential "carpet model" mechanism of action. The isobologram analysis showed an additive interaction between PvAMP66 and gentamicin in inhibiting the growth of MDR K. pneumoniae, leading to a ten-fold reduction in gentamicin's effective concentration. A cytotoxicity against erythrocytes or peripheral blood mononuclear cells was observed at concentrations three to thirteen-fold higher than those exhibited against the evaluated bacterial strains. This evidence suggests that PvAMP66 can serve as a template for the development of AMPs with enhanced activity and deserves further pre-clinical studies as an API in combination therapy.

Improving the Annotation of the Venom Gland Transcriptome of Pamphobeteus verdolaga, Prospecting Novel Bioactive Peptides.

Spider venoms constitute a trove of novel peptides with biotechnological interest. Paucity of next-generation-sequencing (NGS) data generation has led to a description of less than 1% of these peptides. Increasing evidence supports the underestimation of the assembled genes a single transcriptome assembler can predict. Here, the transcriptome of the venom gland of the spider Pamphobeteus verdolaga was re-assembled, using three free access algorithms, Trinity, SOAPdenovo-Trans, and SPAdes, to obtain a more complete annotation. Assembler's performance was evaluated by contig number, N50, read representation on the assembly, and BUSCO's terms retrieval against the arthropod dataset. Out of all the assembled sequences with all software, 39.26% were common between the three assemblers, and 27.88% were uniquely assembled by Trinity, while 27.65% were uniquely assembled by SPAdes. The non-redundant merging of all three assemblies' output permitted the annotation of 9232 sequences, which was 23% more when compared to each software and 28% more when compared to the previous P. verdolaga annotation; moreover, the description of 65 novel theraphotoxins was possible. In the generation of data for non-model organisms, as well as in the search for novel peptides with biotechnological interest, it is highly recommended to employ at least two different transcriptome assemblers.