Expression, purification, and analysis of three recombinant ECD disintegrins (r-colombistatins) from P-III class snake venom metalloproteinases affecting platelet aggregation and SK-MEL-28 cell adhesion

Abstract

Crotalid venoms are rich sources of components that affect the hemostatic system. Snake venom metalloproteinases are zinc-dependent enzymes responsible for hemorrhage that also interfere with hemostasis. The disintegrin domain is a part of snake venom metalloproteinases, which involves the binding of integrin receptors. Integrins play an essential role in cancer survival and invasion, and they have been major targets for drug development and design. Both native and recombinant disintegrins have been widely investigated for their anti-cancer activities in biological systems as well as in vitro and in vivo systems. Here, three new cDNAs encoding ECD disintegrin-like domains of metalloproteinase precursor sequences obtained from a Venezuelan mapanare (Bothrops colombiensis) venom gland cDNA library have been cloned. Three different N- and C-terminal truncated ECD disintegrin-like domains of metalloproteinases named colombistatins 2, 3, and 4 were amplified by PCR, cloned into a pGEX-4T-1 vector, expressed in Escherichia coli BL21, and tested for inhibition of platelet aggregation and inhibition of adhesion of human skin melanoma (SK-Mel-28) cancer cell lines on collagen I. Purified recombinant colombistatins 2, 3, and 4 were able to inhibit ristocetin- and collagen-induced platelet aggregation. r-Colombistatins 2 showed the most potent inhibiting SK-Mel-28 cancer cells adhesion to collagen. These results suggest that colombistatins may have utility in the development of therapeutic tools in the treatment of melanoma cancers and also thrombotic diseases.

Keywords: Bothrops colombiensis; Colombistatins; ECD disintegrin-like domain; SK-Mel-28; Tumor.

Fig. 1. Multiple alignment of the partially deduced amino acid sequences of the ECD disintegrin-like domain of metalloproteinases predicted from partially sequenced clones from B. colombiensis with other homologous venom proteins

Colombistatin 2 (Col-2), colombistatin 3 (Col-3), and colombistatin 4 (Col-4) are aligned with closely related protein in the database, and the % identities are shown in the figure. The alignment was generated with the ClustalW multiple sequence alignment program with manual adjustment and displayed with box shaded. The numbers in parenthesis are the NCBI accession numbers. The tripeptide (ECD) binding motif is marked by the letter “X” above the sequences. All cysteine residues (letter “C” above the sequences) are conserved.

Fig. 2. Phylogenetic tree analysis of snake P-III disintegrin-like domains based on their predicted amino acid sequences

The colombistatins identified in the cDNA library of the Venezuelan mapanare (colombistatins 2, 3, and 4) are marked with an asterisk (*). The tree was constructed using the neighbor-joining methods with a bootstrap for 1000 replications. The number at the branches represents the bootstrap probability. Colombistatin, a RGD disintegrin reported by Sánchez et al. (2009) was used as outgroup.

Fig. 3. SDS-PAGE of r-colombistatins 2, 3, and 4

Samples (3 μg) were run on 4–12% (w/v) Bis-Tris Gel using an Xcell SureLock Mini-Cell at 200 V for 30 min. The gel was stained with RapidStain. Lane 1: SeeBlue Plus2 Markers; lane 2: purified r-colombistatin 2; lane 3: purified r-colombistatin 3; lane 4: purified r-colombistatin 4. An asterisk (*) represents the N-terminal amino acid sequences of purified r-colombistatins containing the five amino acids from the vector (italicized) before the disintegrin-like sequences, which are shown in bold letters.

Fig. 4. Effects of r-colombistatins on adhesion of SK-Mel-28 cancer cell lines on collagen I

SK-Mel-28 cells were seeded in 96-well plates, which was pre-coated with collagen I in the absence (PBS added), or presence of various concentrations of colombistatins 2, 3 and 4. Cell adhesion was measured by MTT technique and the results were expresses as percent of inhibition. The error bars represent the standard deviation from two independent experiments with n = 3.