Structural Characterization, Antimicrobial Activity and BSA/DNA Binding Affinity of New Silver(I) Complexes with Thianthrene and 1,8-Naphthyridine

Three new silver(I) complexes [Ag(NO3)(tia)(H2O)]n (Ag1), [Ag(CF3SO3)(1,8-naph)]n (Ag2) and [Ag2(1,8-naph)2(H2O)1.2](PF6)2 (Ag3), where tia is thianthrene and 1,8-naph is 1,8-naphthyridine, were synthesized and structurally characterized by different spectroscopic and electrochemical methods and their crystal structures were determined by single-crystal X-ray diffraction analysis. Their antimicrobial potential was evaluated against four bacterial and three Candida species, and the obtained results revealed that these complexes showed significant activity toward the Gram-positive Staphylococcus aureus, Gram-negative Pseudomonas aeruginosa and the investigated Candida species with minimal inhibitory concentration (MIC) values in the range 1.56–7.81 μg/mL. On the other hand, tia and 1,8-naph ligands were not active against the investigated strains, suggesting that their complexation with Ag(I) ion results in the formation of antimicrobial compounds. Moreover, low toxicity of the complexes was detected by in vivo model Caenorhabditis elegans. The interaction of the complexes with calf thymus DNA (ct-DNA) and bovine serum albumin (BSA) was studied to evaluate their binding affinity towards these biomolecules for possible insights into the mode of antimicrobial activity. The binding affinity of Ag1–3 to BSA was higher than that for DNA, indicating that proteins could be more favorable binding sites for these complexes in comparison to the nucleic acids.


Introduction
Infections caused by bacteria and fungi threaten human health on a daily basis because many microbes have developed resistance to the commercial antimicrobial agents over time [1]. It is estimated that antimicrobial resistance causes at least 50 thousand deaths each year in Europe and the United States and that it will cause 10 million deaths worldwide per year by 2050 [2]. Considering this, the problems related to the microbial infections and resistance occurrence demand urgent and effective solution, which is based on the development of novel antimicrobial agents. It is well known that silver(I) compounds have shown significant antimicrobial activity, such as silver(I) nitrate which was used for the AgCF3SO3 and AgPF6 in ethanol, polynuclear [Ag(CF3SO3) (1,8-naph)]n (Ag2) and d clear [Ag2 (1,8-naph)2(H2O)1.2](PF6)2 (Ag3) complexes, respectively, were formed. NMR and 13 C), ultraviolet-visible (UV-Vis) and IR spectroscopy, mass spectrometry, and c voltammetry were applied for characterization of the obtained complexes, while a sin crystal X-ray diffraction analysis was used for the determination of their structures. Scheme 1. Schematic presentation of the reaction route for the synthesis of silver(I) complexes with tia (thianthrene) (a) and 1,8-naph (1,8-naphthyridine) (b). Numeration of atoms in the lig is in accordance with IUPAC recommendations for fused ring systems and used for NMR cha terization of complexes Ag1-3.

Solid-State Characterization
Silver(I) complexes Ag1 and Ag2 are polynuclear species with infinite chains complex Ag3 is dinuclear species (Figure 1). Complex Ag1 has a polynuclear struc with tetrahedrally coordinated Ag(I) ion via two thianthrene sulfur atoms (Ag-S dista 2.6198(7) and 2.6747(8) Å), one nitrate oxygen atom (Ag-O distance 2.328(2) Å) and water oxygen atom (Ag-O distance 2.360(2) Å) ( Figure 2, Table 1). Thianthrene acts bridging ligand. Hydrogen bonding between coordinated water molecules of one p nuclear chain with the nitrato ligand of the adjacent polynuclear chain enables the mation of an infinite belt. The polynuclear belt is supported also by C3-H3···O1 inte tions (Table 2). Belts are connected into a supramolecular structure through H5···O2/O4 interactions. Scheme 1. Schematic presentation of the reaction route for the synthesis of silver(I) complexes with tia (thianthrene) (a) and 1,8-naph (1,8-naphthyridine) (b). Numeration of atoms in the ligands is in accordance with IUPAC recommendations for fused ring systems and used for NMR characterization of complexes Ag1-3.
Complex Ag2 has a polynuclear structure with Ag2(1,8-naph)2 units connected into a chain via bridging triflato ligands ( Figure 3). Within the Ag2(1,8-naph)2 unit, the 1,8-naph is a bridging ligand with Ag-N distances of 2.200(2) and 2.218(2) Å, and this unit is supported also by argentophilic interaction with Ag1···Ag1 i distance of 2.7871(5) Å representing the 81% of the sum of van der Waals radii (Table 1). Two triflato ligands connect two adjacent Ag2(1,8-naph)2 units into a chain through an Ag1-O1 bond distance of 2.555(2) Å and through a somewhat longer Ag1-O2 contact of 2.645(2) Å. The polynuclear chain is supported also by C1-H1···O2 and C7-H7···F2 interactions ( Table 2). The crystal structure is stabilized by π···π interactions between 1,8-naph rings of adjacent chains with a centroid-to-centroid distance of 3.6330(17) Å and ring slippage of 1.061 Å as well as by C8-H8···O3 interactions.  and, also, this unit is supported by argentophilic interaction with Ag1···Ag1 i distance of 2.7235(6) Å representing the 79% of the sum of van der Waals radii ( Table 1). The Ag(I) metal center is coordinated also by a water molecule with Ag-O distance of 2.449(4) Å. However, the water molecule was found to possess a 0.60 occupancy ratio. The crystal structure is stabilized by π···π interactions between 1,8-naph rings of adjacent cations with a centroid-to-  and, also, this unit is supported by argentophilic interaction with Ag1···Ag1 i distance of 2.7235(6) Å representing the 79% of the sum of van der Waals radii ( Table 1). The Ag(I) metal center is coordinated also by a water molecule with Ag-O distance of 2.449(4) Å. However, the water molecule was found to possess a 0.60 occupancy ratio. The crystal structure is stabilized by π···π interactions between 1,8-naph rings of adjacent cations with a centroid-to-centroid distance of 3.602(3) Å and ring slippage of 1.035 Å as well as by various O-H···F and C-H···F interactions ( Table 2). The IR spectroscopic data of the complexes Ag1-3 are following their structures determined by a single-crystal X-ray diffraction analysis. The IR spectra of complexes Ag1 and Ag3 exhibited a broad band at~3440 cm −1 due to the stretching vibration of the -OH group, confirming the presence of coordinated water to Ag(I) ion in these complexes. The nitrate coordination in the Ag1 complex is confirmed by a very strong band at 1384 cm −1 with a tendency to split into two bands [35], while in the case of Ag2, strong bands in the 1300-1000 cm −1 region originate from a coordinated triflate in this complex [36,37]. More precisely, the bands at 1163 and 1134 cm −1 can be attributed to the symmetric and asymmetric stretching vibrations of the -CF 3 group, while the bands at 1254, and 1044 and 1032 cm −1 are due to the asymmetric and symmetric stretching of the -SO 3 group in triflate, respectively [36]. Regarding the Ag3 complex, the presence of a strong band at 831 cm −1 assigned to the stretching vibrations of the PF 6 − is a consequence of its existence as a counter-anion in the crystal lattice of this complex [38,39].

Solution Behavior
The NMR spectra of the synthesized silver(I) complexes and corresponding ligands (Supplementary Materials) were assigned based on the previously reported NMR spectroscopic data for tia [40] and 1,8-naph [41]. 1 H-and 13 C-NMR spectra of the presently investigated silver(I) complexes contain the same number of signals as those for the uncoordinated tia (Ag1) and 1,8-naph (Ag2 and Ag3) ligands, being in accordance with the bidentate bridging coordination mode of these ligands to Ag(I) ion. While in the case of Ag1 complex, the resonances in both 1 H-and 13 C-NMR spectra are almost unshifted in respect to those for the uncoordinated tia ligand (vide infra in Materials and Methods section), the chemical shifts of 1 H-and 13 C-NMR resonances for Ag2 and Ag3 differ significantly from those of 1,8-naph. All 1 H resonances for coordinated 1,8-naph in Ag2 and Ag3 are downfield shifted with respect to those for the free ligand. The chemical shifts of 1 H-NMR resonances in these complexes are strongly dependent on the proton position with respect to the nitrogen donor atoms. As can be expected, the largest coordination shift (∆δ = δ complexδ 1,8-naph ) of + 0.32 (for both complexes) were observed for H2/H7 protons, which are adjacent to the nitrogen atoms. On the other hand, there is no strict rule in the shifting of 13 C resonances of the silver(I) complexes Ag2 and Ag3 with respect to the carbon position relative to the nitrogen atoms.
In the mass spectra of the investigated silver(I) complexes, the major doublet peaks are those centered at m/z = 540.9176 (Ag1) and 367.0107 (Ag2 and Ag3), which are consistent with the presence of mononuclear [Ag(tia) 2 ] + and [Ag(1,8-naph) 2 ] + cations, respectively. As it was found previously, the mass spectra of silver(I) complexes show doublet peaks with almost equal intensity for a mononuclear species or triplet peaks with approximately 1:2:1 intensity ratios for dinuclear silver(I) species, as a consequence of the nearly equal abundance of two silver isotopes, 107 Ag (51.84%) and 109 Ag (48.16%) [29].
The UV-Vis spectra of the complexes Ag1-3, recorded in dimethyl sulfoxide (DMSO), are similar to those of the uncoordinated tia and 1,8-naph ligands, suggesting that the absorption maxima (λ max ) at 259 nm for Ag1 and 307 nm for Ag2 and Ag3 are due to the characteristic π→π* transitions in the corresponding heterocyclic ligand [42,43]. In order to check the stability of complexes Ag1-3, their UV-Vis spectra were also recorded 24 h and 48 h after dissolution. As an illustration, time-dependent UV-Vis spectra of the Ag3 complex were shown in Figure S1. For Ag2 and Ag3, the observed transitions remained unmodified over 48 h at room temperature, implying their stability in the solution during this time. Nevertheless, a slight decrease in the intensity of the absorption maximum of 12% was noticed for Ag1, although without significant modifications of the spectrum shape and position of the absorption maxima, indicating that this complex is the least stable amongst the synthesized silver(I) complexes.
With the aim to investigate the air/light stability of complexes Ag1-3, which is important for their possible external application in the form of ointments, gels, and coating materials of dressings [44], sterile cellulose discs impregnated with their DMSO solutions were exposed to air and light for 48 h. The obtained results were compared with those for silver(I) salts used for the synthesis of the complexes (AgNO 3 , AgCF 3 SO 3, and AgPF 6 ) ( Figure S2). As can be seen, all investigated silver(I) complexes were less darkened than the salts during 48 h, indicating the slightly higher air/light stability of the complexes Ag1-3. Among the complexes, Ag1 turned out to be the least stable under the investigated experimental conditions.
The electrochemical behavior of the metal complexes is highly relevant for a better understanding of their stability and biological activity, as well as the way of interaction with biomolecules [45,46]. In accordance with this, the cyclic voltammograms (CV) of complexes Ag1-3 were recorded at the glassy carbon (GC) electrode in DMSO and 0.1 M tetrabutylammonium hexafluorophosphate (TBAHP) as a supporting electrolyte under the following conditions, E begin = −2.0 V and E end = 2.0 V ( Figure 4). As can be seen from this figure, a broad anodic peak, which can be attributed to Ag(I)→Ag(II) oxidation process, was observed [47,48]. When the cyclic voltammogram was recorded in the cathodic direction, two distinctive reduction peaks were detected for all complexes (Table 3), which were assigned to Ag(II)→Ag(I) and Ag(I)→Ag(0) reduction processes, in accordance with the other silver(I) complexes where electrochemical behavior was previously investigated [47,49]. With the aim to investigate the air/light stability of complexes Ag1-3, which is important for their possible external application in the form of ointments, gels, and coating materials of dressings [44], sterile cellulose discs impregnated with their DMSO solutions were exposed to air and light for 48 h. The obtained results were compared with those for silver(I) salts used for the synthesis of the complexes (AgNO3, AgCF3SO3, and AgPF6) (Figure S2). As can be seen, all investigated silver(I) complexes were less darkened than the salts during 48 h, indicating the slightly higher air/light stability of the complexes Ag1-3. Among the complexes, Ag1 turned out to be the least stable under the investigated experimental conditions.
The electrochemical behavior of the metal complexes is highly relevant for a better understanding of their stability and biological activity, as well as the way of interaction with biomolecules [45,46]. In accordance with this, the cyclic voltammograms (CV) of complexes Ag1-3 were recorded at the glassy carbon (GC) electrode in DMSO and 0.1 M tetrabutylammonium hexafluorophosphate (TBAHP) as a supporting electrolyte under the following conditions, Ebegin = −2.0 V and Eend = 2.0 V ( Figure 4). As can be seen from this figure, a broad anodic peak, which can be attributed to Ag(I)→Ag(II) oxidation process, was observed [47,48]. When the cyclic voltammogram was recorded in the cathodic direction, two distinctive reduction peaks were detected for all complexes (Table 3), which were assigned to Ag(II)→Ag(I) and Ag(I)→Ag(0) reduction processes, in accordance with the other silver(I) complexes where electrochemical behavior was previously investigated [47,49].  Table 3. Positions and redox reactions associated with observed cathodic and anodic peaks in cyclic voltammograms of complexes Ag1-3 recorded at GC electrode in DMSO and 0.1 M TBAHP as a supporting electrolyte with a scan rate of 50 mV/s.

Biological Evaluation of the Silver(I) Complexes Ag1-3 2.2.1. Antimicrobial and Antiproliferative Effect of Silver(I) Complexes Ag1-3
Given the traditional antimicrobial properties shown by silver(I) complexes [23][24][25][26][27][28][29][30][31][32][33], the activity of silver(I) complexes Ag1-3 and the ligands, tia and 1,8-naph used for their synthesis, was determined against two Gram-positive (S. aureus and L. monocytogenes), two Gram-negative (E. coli and P. aeruginosa ATCC 10332 and BK25H [50]) bacteria, and three Candida species (C. albicans, C. krusei, and C. parapsilosis; Table 4). The antimicrobial activity of different AgX salts against these bacterial and fungal strains was previously evaluated [25]. Although these salts have shown significant antimicrobial activity, their use has been limited due to their precipitation in the form of AgCl under physiological conditions. As a consequence of this process, Ag(I) ions could not reach the infected sites and manifest antimicrobial activity [10]. Table 4. Minimum inhibitory concentrations (MIC, µg/mL and µM) and cytotoxicity against healthy human fibroblasts MRC-5 (IC 50 , µg/mL and µM) of Ag1-3 complexes and ligands against fungal and bacterial strains. Silver(I) complexes with 1,5-naphthyridine (1,5-naph) and silver(I) sulfadiazine (AgSD) were given for comparative purposes [30,31]. NT = Non tested. Reproduced with permissions from refs. [30,31]. Copyright (2019) Elsevier. Overall, presently investigated silver(I) complexes Ag1-3 showed significant antimicrobial activity toward the Gram-positive S. aureus, two Gram-negative P. aeruginosa species, and tested Candida species, while the lower sensitivity was detected in the case of L. monocytogenes and E. coli (Table 4). The most sensitive species was C. krusei, with MIC values being 1.56 µg/mL for all tested silver(I) complexes (3.6, 4.0, and 1.0 µM for Ag1-3, respectively). On the other hand, both tia and 1,8-naph ligands did not affect the microbial growth at 200 µg/mL (925 and 1537 µM, respectively), suggesting that the observed activity of the complexes is due to the presence of Ag(I) ions. As suggested previously, the antimicrobial activity of silver(I) complexes are connected to the complex ability to release free Ag(I) ions that may exert their antimicrobial action through different pathways [11]. On the other hand, the antiproliferative effect of these complexes was considerably higher but comparable to the observed MIC values (Table 4). On some level, this may limit the antimicrobial application of these complexes, but it is worth mentioning that silver(I) complexes with different classes of ligands have been studied for their antitumor potential and some of them showed antiproliferative activities higher than cisplatin [51].
These results are in line with our previous work in which we showed that the silver(I) complexes with a 1,5-naphthyridine ligand, which is a structural isomer of 1,8-naph, have remarkable activity against a range of Candida species, while for the tested bacterial species, MIC values obtained in this study are 2-3 times lower than in our previously published work (Table 4) [31]. Considerable anti-Candida activity has been also reported for the silver(I) complexes with 1,7-and 4,7-phenanthroline [29,30], while, on the other hand, silver(I) complexes with metronidazole (mtz), [Ag(mtz) 2 (NO 3 )] and [Ag 2 (mtz) 4 ](BF 4 ) 2 , showed better activity against the tested bacterial species and very low activity against C. albicans [44].
The different antibacterial behavior of silver(I) complexes obtained in this study might be related to their mode of action [52]. Gram-positive and Gram-negative bacteria have a different peptidoglycan cell wall and therefore, it can be assumed that the different bactericidal behavior of the silver(I) complexes can be explained by their ability to interact with the cell wall and pass through the membrane causing bacterial growth inhibition [52][53][54].

C. elegans Toxicity
A genetically tractable, multicellular nematode C. elegans, is a widely used model organism as a high throughput platform for toxicity assessment and more recently for the discovery of new antimicrobial compounds [55,56]. The tested complexes Ag1-3 and the corresponding ligands showed low toxicity against C. elegans in concentrations relevant to the observed MICs and cytotoxicity on MRC-5 cells ( Figure 5 and Table 4). The highest percentage of dead worms was observed upon treatment with the highest concentrations of Ag2 complex and 1,8-naph, being 17 and 18%, respectively ( Figure 5). The moderate toxicity against C. elegans was also previously observed for silver(I) complexes with different pyridine-4,5-dicarboxylate esters as ligands, which showed significant activity on the cow mastitis-associated pathogens [49]. The differences in the two toxicity models are possibly due to the more complex uptake system present in nematode in comparison to the direct application to cells in in vitro model. have a different peptidoglycan cell wall and therefore, it can be assumed that the different bactericidal behavior of the silver(I) complexes can be explained by their ability to interact with the cell wall and pass through the membrane causing bacterial growth inhibition [52][53][54].

C. elegans Toxicity
A genetically tractable, multicellular nematode C. elegans, is a widely used model organism as a high throughput platform for toxicity assessment and more recently for the discovery of new antimicrobial compounds [55,56]. The tested complexes Ag1-3 and the corresponding ligands showed low toxicity against C. elegans in concentrations relevant to the observed MICs and cytotoxicity on MRC-5 cells ( Figure 5 and Table 4). The highest percentage of dead worms was observed upon treatment with the highest concentrations of Ag2 complex and 1,8-naph, being 17 and 18%, respectively ( Figure 5). The moderate toxicity against C. elegans was also previously observed for silver(I) complexes with different pyridine-4,5-dicarboxylate esters as ligands, which showed significant activity on the cow mastitis-associated pathogens [49]. The differences in the two toxicity models are possibly due to the more complex uptake system present in nematode in comparison to the direct application to cells in in vitro model.

BSA Binding Study
Serum albumin (SA) is one of the most abundant proteins present in the blood plasma and plays a crucial function in the transport of various endo-and exogenous compounds, including pharmacologically used agents [57]. Therefore, the distribution, free concentration, and metabolism of these agents can be changed as a result of their binding to serum albumin. Considering this, the study of the binding interaction of this protein with potential therapeutic agents could contribute to an understanding of the mechanism of their transport and distribution in the human body and identification of the dynamics of their action [57]. Bovine serum albumin (BSA) has been widely used as a model protein for this purpose due to its structural similarity with human serum albumin (HSA) of approximately 76%, low cost, and some unusual binding properties [57,58]. Fluorescence spectroscopy is one of the most convenient methods for the investigation of BSA binding interactions [59]. The intrinsic fluorescence of this protein is mainly due to tryptophan (Trp), tyrosine (Tyr), and phenylalanine (Phe) residues, whereas the binding interaction between the investigated compound and BSA will cause fluorescence quenching [59].
The emission spectra of BSA were recorded in the absence and presence of an increasing amount of the investigated silver(I) complexes Ag1-3. In all cases, a remarkable quenching of a fluorophore was observed with the gradual addition of the complex, indicating that its interaction with BSA occurred [57][58][59][60][61]. As can be seen from Figure 6, upon

BSA Binding Study
Serum albumin (SA) is one of the most abundant proteins present in the blood plasma and plays a crucial function in the transport of various endo-and exogenous compounds, including pharmacologically used agents [57]. Therefore, the distribution, free concentration, and metabolism of these agents can be changed as a result of their binding to serum albumin. Considering this, the study of the binding interaction of this protein with potential therapeutic agents could contribute to an understanding of the mechanism of their transport and distribution in the human body and identification of the dynamics of their action [57]. Bovine serum albumin (BSA) has been widely used as a model protein for this purpose due to its structural similarity with human serum albumin (HSA) of approximately 76%, low cost, and some unusual binding properties [57,58]. Fluorescence spectroscopy is one of the most convenient methods for the investigation of BSA binding interactions [59]. The intrinsic fluorescence of this protein is mainly due to tryptophan (Trp), tyrosine (Tyr), and phenylalanine (Phe) residues, whereas the binding interaction between the investigated compound and BSA will cause fluorescence quenching [59].
The emission spectra of BSA were recorded in the absence and presence of an increasing amount of the investigated silver(I) complexes Ag1-3. In all cases, a remarkable quenching of a fluorophore was observed with the gradual addition of the complex, indicating that its interaction with BSA occurred [57][58][59][60][61]. As can be seen from Figure 6, upon addition of an increasing concentration of Ag1 complex to BSA solution of a constant concentration, a significant decrease of fluorescence intensity at 367 nm was observed, leading to the formation of a BSA-complex system. addition of an increasing concentration of Ag1 complex to BSA solution of a constant concentration, a significant decrease of fluorescence intensity at 367 nm was observed, leading to the formation of a BSA-complex system. In order to study the quenching mechanism, Stern-Volmer and Scatchard equations were applied for the analysis of the fluorescence quenching data and the obtained data (Stern-Volmer constants (Ksv), quenching rate constants (Kq), binding constants (KA), and the number of binding sites per BSA (n)) are presented in Table 5. The values of Ksv constants indicate that the Ag1 complex containing tia ligand has a higher affinity for BSA compared to the remaining two complexes Ag2 and Ag3 with 1,8-naph ligand. The Ksv values for Ag2 and Ag3 are similar to those calculated for silver(I) complex with 1,2-bis(4pyridyl)ethene [62], as well as for those with N-methyl-1,3,5-triaza-7-phosphaadamantane and tris(pyrazol-1-yl)methanesulfonate [63]. The Kq constant depends on the probability of a collision between fluorophore and quencher and is a measure of the tryptophan residues exposure to the investigated compound [64]. As can be seen from Table 5, the Kq values for the complexes Ag1-3 are in accordance with their good quenching ability of the BSA fluorescence. The Kq values for all investigated complexes are significantly higher than 2 × 10 10 M −1 ·s −1 (the value of maximum diffusion collision quenching rate constant), indicating the static quenching mechanism [63,65]. The KA values for Ag1-3 are in an optimal range, i.e., these constants are high enough to suggest that the complexes can be efficiently carried in the blood (which is an important property of a drug [57]), but not so high to prevent their release from the BSA upon arrival to the target cell [66]. Finally, the calculated n values for complexes suggest their binding to only one binding site per protein molecule.

Lipophilicity Assay
The lipophilicity of a compound is in accordance with its cellular uptake efficiency and is of great importance for the new drug candidate design [67,68]. The octanol-water partition coefficient (logP) is a measure of lipophilicity and can be determined by the flaskshaking method [69]. Thus, the logP values for complexes Ag1-3 are 0.72, −0.14, and 0.37, respectively, being in accordance with the logP values range of −0.4 to 5.6 for new pharmacophores in the comprehensive medicinal chemistry [70]. The greater logP values of In order to study the quenching mechanism, Stern-Volmer and Scatchard equations were applied for the analysis of the fluorescence quenching data and the obtained data (Stern-Volmer constants (K sv ), quenching rate constants (K q ), binding constants (K A ), and the number of binding sites per BSA (n)) are presented in Table 5. The values of K sv constants indicate that the Ag1 complex containing tia ligand has a higher affinity for BSA compared to the remaining two complexes Ag2 and Ag3 with 1,8-naph ligand. The K sv values for Ag2 and Ag3 are similar to those calculated for silver(I) complex with 1,2-bis(4pyridyl)ethene [62], as well as for those with N-methyl-1,3,5-triaza-7-phosphaadamantane and tris(pyrazol-1-yl)methanesulfonate [63]. The K q constant depends on the probability of a collision between fluorophore and quencher and is a measure of the tryptophan residues exposure to the investigated compound [64]. As can be seen from Table 5, the K q values for the complexes Ag1-3 are in accordance with their good quenching ability of the BSA fluorescence. The K q values for all investigated complexes are significantly higher than 2 × 10 10 M −1 ·s −1 (the value of maximum diffusion collision quenching rate constant), indicating the static quenching mechanism [63,65]. The K A values for Ag1-3 are in an optimal range, i.e., these constants are high enough to suggest that the complexes can be efficiently carried in the blood (which is an important property of a drug [57]), but not so high to prevent their release from the BSA upon arrival to the target cell [66]. Finally, the calculated n values for complexes suggest their binding to only one binding site per protein molecule.

Lipophilicity Assay
The lipophilicity of a compound is in accordance with its cellular uptake efficiency and is of great importance for the new drug candidate design [67,68]. The octanol-water partition coefficient (logP) is a measure of lipophilicity and can be determined by the flask-shaking method [69]. Thus, the logP values for complexes Ag1-3 are 0.72, −0.14, and 0.37, respectively, being in accordance with the logP values range of −0.4 to 5.6 for new pharmacophores in the comprehensive medicinal chemistry [70]. The greater logP values of complexes Ag1 and Ag3 in respect of Ag2 indicate that these two complexes are mainly distributed in the octanol phase, further implying their higher cellular uptake efficiency.

DNA Interaction
Interactions of metal ions and their complexes with DNA are of great importance for the design of novel metal-based therapeutic agents [71]. Moreover, it is known that the mechanism of antimicrobial activity of silver(I) complexes can be related to their interaction with different cellular biomolecules, including DNA [10]. Firstly, the binding affinity of Ag1-3 complexes toward DNA was investigated by fluorescence spectroscopy by performing competitive binding experiments based on the displacement of ethidium bromide (EthBr) from DNA. It is well known that EthBr intercalates between adjacent base pairs in the DNA double helix resulting in the enhancement of its fluorescence [63]. After the addition of the investigated complex, a decrease in the fluorescence intensity of the EthBr-DNA system will occur if this complex intercalates into DNA or after its binding to EthBr-DNA, which leads to the formation of a new nonfluorescent EthBr-DNA-complex system [63,65]. In the case of complexes Ag1-3, their addition to the EthBr-DNA solution causes the reduction in its emission intensity, indicating the interaction of complexes with DNA (as an illustration, the fluorescence emission spectra of EthBr-DNA system in the presence of an increasing concentration of Ag2 complex were shown in Figure 7a). However, the calculated binding constants for all complexes (K A , Table 6) are significantly lower than that for EthBr (K A = 2 × 10 6 M −1 ) [65], implying that the reason for the reduction in the emission intensity of the EthBr-DNA system in the presence of silver(I) complexes Ag1-3 could be their binding to the EthBr-DNA and the formation of a new nonfluorescent EthBr-DNA-complex system. complexes Ag1 and Ag3 in respect of Ag2 indicate that these two complexes are mainly distributed in the octanol phase, further implying their higher cellular uptake efficiency.

DNA Interaction
Interactions of metal ions and their complexes with DNA are of great importance for the design of novel metal-based therapeutic agents [71]. Moreover, it is known that the mechanism of antimicrobial activity of silver(I) complexes can be related to their interaction with different cellular biomolecules, including DNA [10]. Firstly, the binding affinity of Ag1-3 complexes toward DNA was investigated by fluorescence spectroscopy by performing competitive binding experiments based on the displacement of ethidium bromide (EthBr) from DNA. It is well known that EthBr intercalates between adjacent base pairs in the DNA double helix resulting in the enhancement of its fluorescence [63]. After the addition of the investigated complex, a decrease in the fluorescence intensity of the EthBr-DNA system will occur if this complex intercalates into DNA or after its binding to EthBr-DNA, which leads to the formation of a new nonfluorescent EthBr-DNA-complex system [63,65]. In the case of complexes Ag1-3, their addition to the EthBr-DNA solution causes the reduction in its emission intensity, indicating the interaction of complexes with DNA (as an illustration, the fluorescence emission spectra of EthBr-DNA system in the presence of an increasing concentration of Ag2 complex were shown in Figure 7a). However, the calculated binding constants for all complexes (KA, Table 6) are significantly lower than that for EthBr (KA = 2 × 10 6 M −1 ) [65], implying that the reason for the reduction in the emission intensity of the EthBr-DNA system in the presence of silver(I) complexes Ag1-3 could be their binding to the EthBr-DNA and the formation of a new nonfluorescent EthBr-DNA-complex system.   The K sv values for the silver(I) complexes Ag1-3 are low and suggest that they bind to ct-DNA through the non-intercalative (electrostatic) mode, which can be also concluded from the percentage of hypochromism up to 15%. For comparison, the percentage of hypochromism of 50% was previously obtained for lucigenin, which is proven as a DNA intercalator [72]. The values of K sv constants for the presently investigated complexes are slightly lower than those obtained for silver(I) complexes {[Ag(bpa)]NO 3 } n and {[Ag(bpa) 2 ]CF 3 SO 3 ·H 2 O} n , where bpa stands for 1,2-bis(4-pyridyl)ethane [62]. On the other hand, silver(I) complex [Ag(daf)(1,10-phen)]NO 3 (daf = 4,5-diazafluoren-9-one and 1,10-phen = 1,10-phenanthroline) had high affinity for the intercalation between DNA base pairs, what can be concluded from the value of its K sv constant of 0.97 × 10 5 M −1 at 37 • C [73]. From the values of K q constants higher than 2 × 10 10 M −1 ·s −1 , it can be seen that the mechanism of interaction between silver(I) complexes Ag1-3 and DNA is a static quenching [65].
To further corroborate the possibility of complexes Ag1-3 toward DNA interaction, gel electrophoresis methodology was also applied (Figure 7b). The slight reduction of the emission of the EthBr-DNA system was observable upon incubation with the tested complexes (Figure 7b), suggesting their weak interaction. The significant effect was only noticed for the Ag3 complex at concentrations of 400 and 200 µM, as determined by Image J analysis ( Figure S3).

Measurements
Elemental microanalyses of complexes Ag1-3 for carbon, hydrogen, and nitrogen were done using a Perkin-Elmer 2400 Series II instrument (CHN) (Waltham, MA, USA). The ESI-HRMS spectra in the positive mode were recorded after dissolving 0.1 mg of silver(I) complexes Ag1-3 in 1.0 mL acetonitrile with an Agilent 62224 accurate mass spectrometer (Santa Clara, CA, USA). The NMR ( 1 H and 13 C) spectra were recorded at room temperature on a Varian Gemini 2000 spectrometer ( 1 H at 200 MHz, 13 C at 50 MHz). Five milligrams of a compound were dissolved in 0.6 mL of CDCl 3 (Ag1) and CD 3 CN/D 2 O (v/v 1:9) (Ag2 and Ag3) and transferred into a 5 mm NMR tube. Chemical shifts were calibrated relative to those of the solvent. The abbreviations for the peak multiplicities are the follows dd (doublet of doublets) and m (multiplet). In order to investigate the stability of silver(I) complexes in solution, the 1 H-NMR spectra were recorded immediately after their dissolution in DMSO-d 6 /D 2 O (v/v 1:9), as well as after 24 h and 48 h standing in the dark at room temperature. The IR spectra were recorded as KBr pellets on a Perkin-Elmer Spectrum 100 spectrometer (Shelton, CT, USA)) over the wavenumber range of 4000-450 cm −1 . The UV-Vis spectra were recorded over the wavelength range of 900-200 nm on a Shimadzu UV-1800 spectrophotometer (Duisburg, Germany) after dissolving the silver(I) complexes in DMSO at room temperature, immediately and 24 h and 48 h after dissolution. The cyclic voltammetry (CV) measurements were performed using a potentiostat/galvanostat AutoLab PGSTAT204 (Utrecht, Netherlands). The cell (5.0 mL) consisted of a three-electrode system, a glassy carbon (GC) electrode as a working electrode, Ag/AgCl (saturated KCl) as a reference electrode, and a platinum wire as a counter electrode. All reported potentials are referred versus the Ag/AgCl (saturated KCl) reference electrode. The electrode surface was renewed before every measurement by polishing with Al 2 O 3 micro-powder and with a piece of cotton due to the strong adsorption of the complexes. The concentration of the solutions of complexes Ag1-3 in DMSO used for these measurements was 1 × 10 −3 M. The emission spectra for DNA and BSA interactions of the complexes were recorded using Jasco FP-6600 spectrophotometer (Pfungstadt, Germany).

Air/Light Stability
The air/light stability of complexes Ag1-3 and the silver(I) salts used for their synthesis (AgNO 3 , AgCF 3 SO 3, and AgPF 6 ) were studied in direct light in an air atmosphere at room temperature [44]. For this purpose, sterile cellulose discs were impregnated with the corresponding silver(I) complex (5.0 µL of 50 mg/mL DMSO stock solution) and exposed to air and light. The stability was monitored visually over 48 h.

Crystallographic Data Collection and Refinement of the Structures
Single-crystal X-ray diffraction data were collected on an Agilent Technologies Su-perNova Dual diffractometer (Yarnton, Oxfordshire, UK) with an Atlas detector using monochromated Mo-Kα radiation (λ = 0.71073 Å) at 150 K. The data were processed using CrysAlis Pro [76]. The structure was solved by the SHELXT program [77] and refined by a full-matrix least-squares procedure based on F 2 with SHELXL [78] using the Olex2 program suite [79]. All non-hydrogen atoms were refined anisotropically. All the hydrogen atoms were readily located in difference Fourier maps and were subsequently treated as riding atoms in geometrically idealized positions with U iso (H) = kU eq (C,O), where k = 1.5 for OH groups and 1.2 for all other H atoms unless otherwise noted. In Ag1, the hydrogen atom H1A bonded to water molecule O1 was refined restraining the O-H bond length, while the hydrogen atom H1B was refined freely. In Ag3, the water molecule O1 was refined with a fixed 0.60 occupation factor. The crystallographic data are listed in Table S1.

Antimicrobial Studies
The minimum inhibitory concentration (MIC) values of silver(I) complexes Ag1-3 alongside corresponding ligands, tia, and 1,8-naph, were determined according to the standard broth microdilution assays, recommended by the Standards of European Committee on Antimicrobial Susceptibility Testing (v 7.3.1: the method for the determination of broth dilution minimum inhibitory concentrations of antifungal agents for yeasts [80]) for Candida spp. (C. albicans ATCC 10,231, C. parapsilosis ATCC 22,019 and C. krusei ATCC 6258), and the National Committee for Clinical Laboratory Standards (M07-A8) for bacteria (Pseudomonas aeruginosa ATCC 10,332, P. aeruginosa BK25H [50], Staphylococcus aureus ATCC 25,923, Listeria monocytogenes NCTC 11,994 and Escherichia coli NCTC 9001). Stock solutions of silver(I) complexes were prepared in DMSO at a final concentration of 50 µg/mL and the highest tested concentration was 250 µg/mL (making sure that the final concentration of the DMSO in the assay was lower than 0.05%, v/v). The inoculums were 1 × 10 5 colony forming units (cfu)/mL for Candida species, and 5 × 10 5 cfu/mL for bacteria. The MIC value was recorded as the lowest concentration that inhibited the growth after 24 h at 37 • C, using the Tecan Infinite 200 Pro multiplate reader (Tecan Group Ltd., Männedorf, Switzerland).

MTT Assay
Antiproliferative activity of silver(I) complexes Ag1-3 was determined by 3-(4,5 -dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay [81] on human lung fibroblasts cells (MRC-5, obtained from American Type Culture Collection (ATCC)). 1 × 10 4 cells per well were cultured in the complete RPMI 1640 medium (Sigma-Aldrich, Munich, Germany) as a monolayer, and incubated with the silver(I) complexes, at a concentration from a maximum of 100 µg/mL, in a humidified atmosphere of 95% air and 5% CO 2 at 37 • C for 48 h. The extent of MTT reduction was measured spectrophotometrically at 540 nm using Tecan Infinite 200 Pro multiplate reader (Tecan Group Ltd., Männedorf, Switzerland). Cytotoxicity was expressed as the concentration of the compound inhibiting cell growth by 50% (IC 50 ) in comparison with the control (DMSO-treated cells).

C. elegans Survival Assay
Caenorhabditis elegans N2 (glp-4; sek-1) was propagated under standard conditions, synchronized by hypochlorite bleaching, and cultured on nematode growth medium using E. coli OP50 as a food source, as described previously [82]. The C. elegans survival assay was carried out as described previously with some modifications [83,84]. In brief, synchronized worms (L4 stage) were suspended in a medium containing 95% M9 buffer (3.0 g of KH 2 PO 4 , 6.0 g of Na 2 HPO 4 , 5.0 g of NaCl, and 1 mL of 1 M MgSO 4 × 7H 2 O in 1 L of water), 5% LB (Luria-Bertani) broth (Oxoid, Basingstoke, UK), and 10 µg of cholesterol (Sigma-Aldrich, Munich, Germany) per mL. The experiment was carried out in 96-well flat bottomed microtiter plates (Sarstedt, Nümbrecht, Germany) in the final volume of 100 µL per well. 25 µL of this suspension of nematodes (25-35 nematodes) were transferred to the wells of a 96-well microtiter plate, where 50 µL of the medium was previously added. Next, 25 µL of a solvent control (DMSO) or 25 µL of a concentrated solution was added to the test wells. Final concentrations of the complexes were 10, 5, and 1 µg/mL and 200, 100, and 50 µg/mL for ligands, made out from the stock solutions (50 mg/mL in DMSO) of each compound. Subsequently, the plates were incubated at 25 • C for 2 days. The fraction of dead worms was determined after 48 h by counting the number of dead worms and the total number of worms in each well, using a stereomicroscope (SMZ143-N2GG, Motic, Germany). The compounds were tested at least three times in each assay and each assay was repeated at least two times (n ≥ 6). As a negative control experiment, nematodes were exposed to the medium containing 1% DMSO.

BSA Binding Study
The BSA (16 µM) binding study was performed by carrying out tryptophan fluorescence quenching experiments in PBS solution (pH = 7.4). The quenching of the emission intensity of BSA at 366 nm was monitored using the increasing concentration of the complexes Ag1-3 (0-310 µM). Fluorescence spectra were recorded in the range 280-500 nm with an excitation wavelength of 275 nm. The binding constants of the complexes (K A ) and binding sites (n) were calculated using the following equation [63,85]: where K A is the binding constant of the silver(I) complex with BSA, while n represents the number of binding sites of the complex per BSA molecule.

Lipophilicity Assay
The lipophilicity of the silver(I) complexes Ag1-3 was determined by the flask-shaking method [69]. The complexes were dissolved in DMSO, added to the water/n-octanol system, and vortexed for 1 h at room temperature. Thereafter, the solutions were allowed to stand for 24 h until the separation of the two phases was achieved. The concentration of complexes in both phases was determined by measuring absorbance values using previously determined calibration curves. logP values were calculated according to the equation: where c 0 and c w are the concentrations of the complex in the n-octanol and water phase, respectively. The silver(I) complexes Ag1-3 were dissolved in DMSO (10 mM). A stock solution of ct-DNA was prepared in PBS, and the concentration of this solution was determined from UV absorbance at 260 nm using the molar extinction coefficient ε = 6.6 × 10 3 M −1 ·cm −1 [86]. A stock solution of ethidium bromide (EthBr) was prepared in DMSO (1.01 × 10 −2 M) and kept at 4 • C prior to use. The competitive binding studies were carried out in PBS (pH = 7.4) by keeping [DNA]/[EthBr] = 10, while increasing the concentration of the complexes. Measurements were performed in the wavelength range of 525-800 nm with an excitation wavelength of 520 nm. The Stern-Volmer constants (K sv ) were calculated using the following equation [63]: where F 0 and F stand for the fluorescence intensity in the absence and presence of the silver(I) complex, respectively, K q represents the bimolecular quenching constant, and τ 0 (10 −8 s) is the average fluorescence lifetime of the fluorophore in the absence of the quencher. The binding constants (K A ) and apparent binding sites (n) were calculated as described above for BSA binding study [63,85].

Gel Electrophoresis Assay
DNA interaction assay using gel electrophoresis was performed according to the previously published procedure [29] with commercial lambda bacteriophage DNA (300 ng, Thermo Scientific™). DNA solution of the final concentration 20 ng/µL was incubated with 400, 200 and 100 µM of silver(I) complexes Ag1-3 and corresponding ligands, tia, and 1,8-naph, in 15 µL of reaction volume, for 1 h at 37 • C. 300 ng per lane of samples were run on 0.8% agarose gel with EthBr against a HyperLadder™ 1 kb DNA Ladder plus (FastGene) at 60 V for 1 h. Gels were visualized and analyzed using the Gel Doc EZ system (Bio-Rad, Life Sciences, Hercules, CA, USA), equipped with the Image Lab™ Software.

Conclusions
In this study, three new silver(I) complexes [Ag(NO 3 )(tia)(H 2 O)] n (Ag1), [Ag(CF 3 SO 3 ) (1,8-naph)] n (Ag2) and [Ag 2 (1,8-naph) 2 (H 2 O) 1.2 ](PF 6 ) 2 (Ag3), were synthesized, structurally characterized and biologically evaluated as potential antimicrobial agents. The present study confirms that the sulfur-and nitrogen-containing heterocycles, thianthrene and 1,8-naphthyridine, respectively, act as effective bridging ligands between two Ag(I) ions forming poly-or dinuclear complexes. The investigated complexes showed considerable activity against the Gram-positive S. aureus, two Gram-negative P. aeruginosa species and the tested Candida species, with low toxicity in vivo on the C. elegans nematode model. The type of bridging ligand in the investigated complexes plays an important role in determining their affinity toward DNA and BSA biomolecules. Thus, the Ag1 complex with sulfur-containing heterocycle as bridging ligand has a higher binding affinity toward BSA in respect to Ag2 and Ag3 complexes having nitrogen-containing bridging ligand. On the other hand, the DNA interaction of the latter two complexes is more significant than with the Ag1 complex. Nevertheless, it can be concluded that proteins could be more favorable binding sites for all three complexes in comparison to the nucleic acids. The obtained results from this study also suggest that the synthesized silver(I) complexes with thianthrene and 1,8-naphthyridine as ligands could be further evaluated as agents for the treatment of the mixed Candida-Pseudomonas aeruginosa and Candida-Staphylococcus aureus infections.
Supplementary Materials: The following are available online. CCDC 2065519-2065521 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif or by emailing data_request@ccdc.cam.ac.uk or by contacting The Cambridge Crystallography Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44-1223-336033. 1 H and 13 C-NMR spectra for Ag1-3, thianthrene (tia) and 1,8-naphthyridine (1,8-naph). Experimental data for thianthrene (tia) and 1,8-naphthyridine (1,8-naph). Figure S1: Time stability of Ag3 complex followed by UV-Vis spectrophotometry at room temperature in DMSO, Figure S2: Air/light stability of silver(I) complexes Ag1-3 and corresponding silver(I) salts used for their synthesis, Figure S3: Quantification of the interaction of silver(I) complexes Ag1-3 with commercial lambda bacteriophage DNA by gel electrophoresis done in the Excel program. ImageJ program was used for figure analysis, Table S1: Details of the crystal structure determination for complexes Ag1-3.

Data Availability Statement:
The experimental data used to support the findings of this study are available on request from the corresponding author.