67 (4588 0 Da), Bg 34 22 (4684 1 Da) On the other hand U-AITX-Bg

67 (4588.0 Da), Bg 34.22 (4684.1 Da). On the other hand U-AITX-Bg1a was not completely sequenced at the N-terminus; nonetheless the multiple sequence alignment (Fig. 4A) suggested

that the missing fragment is GT. Accordingly, the molecular mass of U-AITX-Bg1a should be 4593.3 Da which is in good agreement with the molecular mass of Bg 30.66b (4592.5 Da). For more clarity, refer to Fig. 2 and Fig. 3 to observe the peaks from RPC18 chromatography corresponding to the mentioned peptides. We should also stress that on sequence similarity search procedure, a translated nucleotide sequence from Anthopleura elegantissima encoding a putative neurotoxin (GenBank ID: gi|193259782) similar Vincristine to these U-AITX-Bg1a–e peptides was identified [68]. We named it as U-AITX-Ael1a, following OSI-744 the nomenclature

proposed by King et al. [44]. Even though its initial Met amino acid in the precursor is not determined, we may assume that its full CDS is as shown in Fig. 4B, based on the similarity in the alignment with the U-AITX-Bg1a–e peptides here reported. Interestingly, the precursors of U-AITX-Ael1a, U-AITX-Bg1b–d are closely related and present the KR cleavage site, as usual for most of the sea anemone neurotoxins. On the other hand, U-AITX-Bg1e precursor is more variable, being nine amino acids longer than the others and presenting the RR cleavage site. This is the first report of full CDS and precursors for this family of sea anemone toxins, and curiously, species from different genera (Bunodosoma vs. Anthopleura) present similar precursors, an unusual characteristic of sea anemone genes [58], [59] and [60]. On the contrary, the similarity search against the EST database of A. viridis (39,939 ESTs) provided no match to these toxins, revealing that such a category of peptides is not expressed in that species, in agreement to Kozlov and Grishin [45]. Molecular models of U-AITX-Bg1

(a–e), U-AITX-Ael1a and BcIV obtained by the I-Tasser server are represented in Fig. 5. The C-score for each model, as predicted by I-TASSER server were 0.861, 0.814, 0.769, 0.882, 0.570, MRIP 0.953 and 0.395 (typically in the range from −5 to 2, higher values signifies a model with a high confidence), respectively. Also, their QMEAN scores and other parameters showed adequate values (data not shown), confirming a good agreement of structures based on APETx1 template and validating our models. Similarly to APETx1 [15] and APETx2 [16], the new APETx-like peptides U-AITX-Bg1 (a, b, d, and e) are composed of a compact core comprising four-stranded β sheets, from which the loop (16–27) and the N- and C-termini emerge. The β sheets sequence obeys (with slight differences) the APETx pattern: residues 3–6 (strand I), 9–14 (strand II), 28–32 (strand III) and 35–39 (strand IV), are connected by a type II β-turn (between strands I and II), a loop (between strands II and III) and a type I β-turn (between strands III and IV).

Comments are closed.