The dissertation is devoted to a study on mechanisms of interaction of dephosphorylated 2'-5'-triadenylates with a human calcium-binding protein S100A1 in vitro. For the first time, it was shown that upon binding, a protein secondary structure is changing, i.e. a proportion of alpha-helical elements decreases and the percentage of disordered elements increases. It was demonstrated that besides the secondary structure alterations, 2'-5'-triadenylates changed (not significantly, though) the S100A1-Ca2+ binding constant. We have determined the amino acid residues which are directly involved in the 2'-5'-triadenylate-S100A1 interaction. Considering the earlier described ability of 2'-5'-triadenylate to stimulate the muscular contraction in vivo, we may suggest, that 2'-5'-triadenylate interacts with S100A1 directly, causing structural and functional alterations within the S100A1, and, therefore, influencing of S100A1 interaction with a ryanodine receptor.
The possibility of complex formation between 2′-5′-А3/2′-5′-А3-еро and S100A1 was studied as well. Using the circular dichroism spectroscopy, we showed that a CD spectrum of S100A1 is altered upon interaction with 2′-5′-А3/2′-5′-А3-еро. Noteworthy, complex formation was shown in both, presence and absence of Ca2+ ions.
It was shown that 2′-5′-А3/2′-5′-А3-еро binding causes changes in the secondary structure of protein. The complex formation initiates a decrease of alpha-helical content for 6% and 5% of the apo-S100A1, respectively. Contrary to the apo-form, the alpha-helical content of the protein holo-form was decreased less significantly – for 3% and 4%, respectively. We managed to identify the increase of disordered secondary structure elements for both, apo- and holo-S100A1, what allowed us to suggest, that a proportion of alpha-helices transform into disordered elements upon 2′-5′-А3 or 2′-5′-А3-еро binding.
Further, we used Fourier infrared spectroscopy, which allowed to identify the Amide I and II bands shifts. The values of the shifts equaled 1cm-1 and 3 cm-1, which allows us to assume that the percentage of disordered secondary structure elements increased.
We identified also the amino-acid residues within the S100A1 protein, that demonstrated the highest CSP values upon binding to 2′-5′-А3. They turned out to be located within the Ca2+-binding loops, which constitute a central part of EF-hands. The bulk of the signals originated from the N-terminal region of the Ca2+-binding domain, where His18, Lys21, Asp24, Lis25 and Lis30, were localized. These amino-acid residues are known to be strongly depended on the experimental conditions, such as temperature and pH.
Other amino-acid residues, Val69 and Gln72, which are located within the C-terminal domain of S100A1, showed significantly lower CSP values upon binding to 2'-5'-A3.
Considering the low solvent accessibility of the aforementioned amino-acid residues, we assume that 2'-5'-A3 does not have a possibility to bind those residues directly. It is more likely that 2'-5'-A3 binds S100A1 elsewhere, probably, in Ca2+-binding domain and/or linker region; the conformational changes, caused by the binding, are transmitted to the sensitive monomer interface.
We identified existence of 3 bonds within the 5Å radius, using computer modeling. One of those, a hydrogen bond, is formed between the NH2- group and the adenine I and AMP and CO group of Ala80. The second and third bonds are electrostatic. They are formed between the PO2- group of the II AMP residue and the CO group of Val69 and between the PO2- group of the III AMP residue and CO group of Asn64. Computer modeling of the binding between 2'-5'-А3 and S100A1 supports the NMR data, obtained earlier – the PO2- group of the second AMP and the CO group of Val69 interact electrostatically; this might explain the significant CSP value of Val69. The latter is a part of C-terminal domain of S100A1.
We demonstrated the insignificant impact of 2′-5′-A3 on the Ca2+-affinity of S100А1. Interestingly, 2′-5′-A3 caused the decrease of Са2+-affinity to S100A1 by 0.2·10-4, while the epoxy modified analogue caused the increase of the binding constant by 0.8·10-3, compared to the control experiment.
We also found that 2′-5′-oligoadenylates influence significantly the protein kinase activity. Of note, they could act either as activators or inhibitors, depending on the nature of a 2′-5′-oligoadenylate. This effect depended also on the concentration of both, the 2′-5′-oligoadenylate and ATP. One of the protein kinases, Aurora, was shown to be inhibited by 2′-5′-A3 і 2′-5′-A3-еро by 35% and 40%, respectively.
