Half-Sandwich Type Platinum-Group Metal Complexes of C-Glucosaminyl Azines: Synthesis and Antineoplastic and Antimicrobial Activities

While platinum-based compounds such as cisplatin form the backbone of chemotherapy, the use of these compounds is limited by resistance and toxicity, driving the development of novel complexes with cytostatic properties. In this study, we synthesized a set of half-sandwich complexes of platinum-group metal ions (Ru(II), Os(II), Ir(III) and Rh(III)) with an N,N-bidentate ligand comprising a C-glucosaminyl group and a heterocycle, such as pyridine, pyridazine, pyrimidine, pyrazine or quinoline. The sugar-containing ligands themselves are unknown compounds and were obtained by nucleophilic additions of lithiated heterocycles to O-perbenzylated 2-nitro-glucal. Reduction of the adducts and, where necessary, subsequent protecting group manipulations furnished the above C-glucosaminyl heterocycles in their O-perbenzylated, O-perbenzoylated and O-unprotected forms. The derived complexes were tested on A2780 ovarian cancer cells. Pyridine, pyrazine and pyridazine-containing complexes proved to be cytostatic and cytotoxic on A2780 cells, while pyrimidine and quinoline derivatives were inactive. The best complexes contained pyridine as the heterocycle. The metal ion with polyhapto arene/arenyl moiety also impacted on the biological activity of the complexes. Ruthenium complexes with p-cymene and iridium complexes with Cp* had the best performance in ovarian cancer cells, followed by osmium complexes with p-cymene and rhodium complexes with Cp*. Finally, the chemical nature of the protective groups on the hydroxyl groups of the carbohydrate moiety were also key determinants of bioactivity; in particular, O-benzyl groups were superior to O-benzoyl groups. The IC50 values of the complexes were in the low micromolar range, and, importantly, the complexes were less active against primary, untransformed human dermal fibroblasts; however, the anticipated therapeutic window is narrow. The bioactive complexes exerted cytostasis on a set of carcinomas such as cell models of glioblastoma, as well as breast and pancreatic cancers. Furthermore, the same complexes exhibited bacteriostatic properties against multiresistant Gram-positive Staphylococcus aureus and Enterococcus clinical isolates in the low micromolar range.


Introduction
Registered platinum complexes (cisplatin, oxaliplatin and carboplatin) constitute the backbone of modern oncological chemotherapy in multiple solid tumors with poor prognosis, including a large set of carcinomas, such as ovarian cancer, sarcomas and hematological malignancies [1,2]. The applicability of platinum compounds is limited by platinum resistance and toxicity [3][4][5][6][7][8].
We recently reported a series of half-sandwich complexes with five-membered chelate rings constructed with the use of Nand C-glycopyranosyl heterocyclic N,N-bidentate ligands (Figure 1, I) [32,45,46]. Several representatives of I displayed low micromolar or, in certain cases, submicromolar (e.g., Ia) cytostatic activities against cancer cells, in addition to proving to be selective for such cells. The antiproliferative potency of these complexes is thought to be related to reactive oxygen species production [32,45,46]. It is worth mentioning that the complexes with antineoplastic activities (e.g., Ia) were also shown to be effective against Gram-positive multiresistant bacteria [31,32]. A short summary of the structure-activity relationships (SARs) of these complexes is presented in Figure 1, while for a more detailed explanation of the SARs, the reader is referred to our previous publications [31,32,45,46]. One of the most important structural motifs related to biological efficacy is the presence of the sugar moiety O-protected with large hydrophobic acyl, preferably with benzoyl groups. This feature contributes, to a large extent, to the favorably increased lipophilic character of the biologically active complexes [32,45,46]. Apart from the above complexes, there are only two more literature examples of halfsandwich complexes with sugar-based N,N-chelators. 1,4-Bis(β-D-glycopyranosyl)tetrazenes [47] (e.g., II) and methyl 2,3-diamino-2,3-dideoxy-hexopyranosides [48] (e.g., III) were incorporated into the coordination sphere of the reported complexes. The antineoplastic effects of these organometallics were also studied, some of which were found to be cytotoxic at low micromolar concentrations against various cancer cells (II and III represent the most efficient compounds of the respective series) [47,48].
Based on the structures of the sugar-derived ligands of complexes I and III, we considered that C-glucosaminyl N-heterocycles, with an N-donor atom in the glycon and another in the heterocyclic aglycon part, are capable of forming half-sandwich type complexes with a six-membered chelate ring. In the present study, the preparation of a series of C-glucosaminyl azines and their incorporation as N,N-bidentate ligands into type IV complexes were envisaged. Biological studies were also conducted to reveal the anticancer and antibacterial potencies of the new organometallic compounds. With the exception of the glycosyl heterocyclic ligand, the main structural elements of the new complexes as depicted in formula IV were designed to be identical to those of type I complexes with biological activities. In addition, the replacement of the O-benzoyl groups of the monosaccharide unit by O-benzyl groups was envisaged in order to examine the effect of ether-type protection on the biological efficiency of the complexes.
Molecules 2023, 28, x FOR PEER REVIEW 3 of 52 the glycosyl heterocyclic ligand, the main structural elements of the new complexes as depicted in formula IV were designed to be identical to those of type I complexes with biological activities. In addition, the replacement of the O-benzoyl groups of the monosaccharide unit by O-benzyl groups was envisaged in order to examine the effect of ethertype protection on the biological efficiency of the complexes.
In the next step, the reduction of the nitro group of compounds 2 showed much less uniform behavior. To achieve O-perbenzylated C-glucosaminyl azines, reduction of the nitro group of 2a-e by Zn-HCl (ii) was investigated first. This transformation of 2a to the expected 2-glucosaminyl pyridine (3a) was smoothly accomplished in good yield. However, similar reactions of nitro derivatives 3b-e led to multicomponent reaction mixtures, from which only the 2-glucosaminyl pyrazine 3d could be isolated in low yield. Unfortunately, our further attempts to obtain the glucosamine derivatives 3d,e from 2d,e under different reductive conditions also failed; either no reaction took place (Fe, ccHCl, THF-H2O 1:1, 0 °C; SnCl2, dry EtOH, reflux; SiCl3H, DIPEA, dry CH3CN, 0 °C; B2(OH)4, THF-H2O 1:1, 80 °C) or formation of inseparable product mixtures (Sn, ccHCl, THF-H2O 1:1, 0 °C) was observed. In the next step, the reduction of the nitro group of compounds 2 showed much less uniform behavior. To achieve O-perbenzylated C-glucosaminyl azines, reduction of the nitro group of 2a-e by Zn-HCl (ii) was investigated first. This transformation of 2a to the expected 2-glucosaminyl pyridine (3a) was smoothly accomplished in good yield. However, similar reactions of nitro derivatives 3b-e led to multicomponent reaction mixtures, from which only the 2-glucosaminyl pyrazine 3d could be isolated in low yield. Unfortunately, our further attempts to obtain the glucosamine derivatives 3d,e from 2d,e under different reductive conditions also failed; either no reaction took place (Fe, ccHCl, THF-H2O 1:1, 0 °C; SnCl2, dry EtOH, reflux; SiCl3H, DIPEA, dry CH3CN, 0 °C; B2(OH)4, THF-H2O 1:1, 80 °C) or formation of inseparable product mixtures (Sn, ccHCl, THF-H2O 1:1, 0 °C) was observed.  In the next step, the reduction of the nitro group of compounds 2 showed much less uniform behavior. To achieve O-perbenzylated C-glucosaminyl azines, reduction of the nitro group of 2a-e by Zn-HCl (ii) was investigated first. This transformation of 2a to the expected 2-glucosaminyl pyridine (3a) was smoothly accomplished in good yield. However, similar reactions of nitro derivatives 3b-e led to multicomponent reaction mixtures, from which only the 2-glucosaminyl pyrazine 3d could be isolated in low yield. Unfortunately, our further attempts to obtain the glucosamine derivatives 3d,e from 2d,e under different reductive conditions also failed; either no reaction took place (Fe, ccHCl, THF-H2O 1:1, 0 °C; SnCl2, dry EtOH, reflux; SiCl3H, DIPEA, dry CH3CN, 0 °C; B2(OH)4, THF-H2O 1:1, 80 °C) or formation of inseparable product mixtures (Sn, ccHCl, THF-H2O 1:1, 0 °C) was observed.  In the next step, the reduction of the nitro group of compounds 2 showed much less uniform behavior. To achieve O-perbenzylated C-glucosaminyl azines, reduction of the nitro group of 2a-e by Zn-HCl (ii) was investigated first. This transformation of 2a to the expected 2-glucosaminyl pyridine (3a) was smoothly accomplished in good yield. However, similar reactions of nitro derivatives 3b-e led to multicomponent reaction mixtures, from which only the 2-glucosaminyl pyrazine 3d could be isolated in low yield. Unfortunately, our further attempts to obtain the glucosamine derivatives 3d,e from 2d,e under different reductive conditions also failed; either no reaction took place (Fe, ccHCl, THF-H2O 1:1, 0 °C; SnCl2, dry EtOH, reflux; SiCl3H, DIPEA, dry CH3CN, 0 °C; B2(OH)4, THF-H2O 1:1, 80 °C) or formation of inseparable product mixtures (Sn, ccHCl, THF-H2O 1:1, 0 °C) was observed.

Het
Conditions and Yields (%) 4 5 a plex product mixtures, in which the desired C-glucosaminyl pyridine 5a, pyridazine 5b and quinoline 5e were detected by TLC analysis; however, they could not be separated in a pure state. Treatment of 4e with Sn powder in the presence of ccHCl (iv) was also carried out to afford the target 5e in acceptable yield. In order to obtain O-perbenzoylated C-glucosaminyl azines, a direct exchange of the O-benzyl protecting groups with benzoyl groups by a Zn(OTf)2-mediated reaction [55] of compounds 2a-d with benzoyl chloride (Table 3, i) was performed to afford C-(2′-deoxy-2′-nitro-3′,4′,6′-tri-O-benzoyl-β-D-glucopyranosyl)azines 6a-d in good to excellent yields. Subsequent reduction of the nitro group of the pyridine derivative 6a by Zn-HCl (ii) afforded the O-perbenzoylated glucosamine derivative 7a in moderate yield. Analogous reactions (ii) carried out with compounds 6b-d led to complex reaction mixtures, from which the desired C-glucosaminyl heterocycles 7b-d could not be isolated. For the transformation of 6b-d into 7b-d, further experiments were conducted under various reductive conditions (e.g., H2, Pd(C) or Pd(OH)2, dry EtOH, reflux; SnCl2, dry EtOH, reflux; Sn, ccHCl, THF-H2O 1:1, 0 °C; B2(OH)4, THF-H2O 1:1, 80 °C); however, none of these experiments was successful.
Due to the above difficulties, another three-step procedure starting from 5c-e was applied to obtain the planned 7c-e (Table 3). Thus, the NH2 group of 5c-e was protected first as a carbamate using Boc2O (iv), and the resulting 8c-e were O-perbenzoylated upon treatment with benzoyl chloride (v) to afford the O-and N-protected glucosaminyl derivatives 9c-e. Finally, acid-mediated liberation of the NH2 group in 9c-e (vi) was carried out, providing the final products 7c-e in high yields. In order to obtain O-perbenzoylated C-glucosaminyl azines, a direct exchange of the O-benzyl protecting groups with benzoyl groups by a Zn(OTf)2-mediated reaction [55] of compounds 2a-d with benzoyl chloride (Table 3, i) was performed to afford C-(2′-deoxy-2′-nitro-3′,4′,6′-tri-O-benzoyl-β-D-glucopyranosyl)azines 6a-d in good to excellent yields. Subsequent reduction of the nitro group of the pyridine derivative 6a by Zn-HCl (ii) afforded the O-perbenzoylated glucosamine derivative 7a in moderate yield. Analogous reactions (ii) carried out with compounds 6b-d led to complex reaction mixtures, from which the desired C-glucosaminyl heterocycles 7b-d could not be isolated. For the transformation of 6b-d into 7b-d, further experiments were conducted under various reductive conditions (e.g., H2, Pd(C) or Pd(OH)2, dry EtOH, reflux; SnCl2, dry EtOH, reflux; Sn, ccHCl, THF-H2O 1:1, 0 °C; B2(OH)4, THF-H2O 1:1, 80 °C); however, none of these experiments was successful.
Due to the above difficulties, another three-step procedure starting from 5c-e was applied to obtain the planned 7c-e (Table 3). Thus, the NH2 group of 5c-e was protected first as a carbamate using Boc2O (iv), and the resulting 8c-e were O-perbenzoylated upon treatment with benzoyl chloride (v) to afford the O-and N-protected glucosaminyl derivatives 9c-e. Finally, acid-mediated liberation of the NH2 group in 9c-e (vi) was carried out, providing the final products 7c-e in high yields.  In order to obtain O-perbenzoylated C-glucosaminyl azines, a direct exchange of the O-benzyl protecting groups with benzoyl groups by a Zn(OTf)2-mediated reaction [55] of compounds 2a-d with benzoyl chloride (Table 3, i) was performed to afford C-(2′-deoxy-2′-nitro-3′,4′,6′-tri-O-benzoyl-β-D-glucopyranosyl)azines 6a-d in good to excellent yields. Subsequent reduction of the nitro group of the pyridine derivative 6a by Zn-HCl (ii) afforded the O-perbenzoylated glucosamine derivative 7a in moderate yield. Analogous reactions (ii) carried out with compounds 6b-d led to complex reaction mixtures, from which the desired C-glucosaminyl heterocycles 7b-d could not be isolated. For the transformation of 6b-d into 7b-d, further experiments were conducted under various reductive conditions (e.g., H2, Pd(C) or Pd(OH)2, dry EtOH, reflux; SnCl2, dry EtOH, reflux; Sn, ccHCl, THF-H2O 1:1, 0 °C; B2(OH)4, THF-H2O 1:1, 80 °C); however, none of these experiments was successful.
Due to the above difficulties, another three-step procedure starting from 5c-e was applied to obtain the planned 7c-e (Table 3). Thus, the NH2 group of 5c-e was protected first as a carbamate using Boc2O (iv), and the resulting 8c-e were O-perbenzoylated upon treatment with benzoyl chloride (v) to afford the O-and N-protected glucosaminyl derivatives 9c-e. Finally, acid-mediated liberation of the NH2 group in 9c-e (vi) was carried out, providing the final products 7c-e in high yields.  In order to obtain O-perbenzoylated C-glucosaminyl azines, a direct exchange of the O-benzyl protecting groups with benzoyl groups by a Zn(OTf)2-mediated reaction [55] of compounds 2a-d with benzoyl chloride (Table 3, i) was performed to afford C-(2′-deoxy-2′-nitro-3′,4′,6′-tri-O-benzoyl-β-D-glucopyranosyl)azines 6a-d in good to excellent yields. Subsequent reduction of the nitro group of the pyridine derivative 6a by Zn-HCl (ii) afforded the O-perbenzoylated glucosamine derivative 7a in moderate yield. Analogous reactions (ii) carried out with compounds 6b-d led to complex reaction mixtures, from which the desired C-glucosaminyl heterocycles 7b-d could not be isolated. For the transformation of 6b-d into 7b-d, further experiments were conducted under various reductive conditions (e.g., H2, Pd(C) or Pd(OH)2, dry EtOH, reflux; SnCl2, dry EtOH, reflux; Sn, ccHCl, THF-H2O 1:1, 0 °C; B2(OH)4, THF-H2O 1:1, 80 °C); however, none of these experiments was successful.
Due to the above difficulties, another three-step procedure starting from 5c-e was applied to obtain the planned 7c-e (Table 3). Thus, the NH2 group of 5c-e was protected first as a carbamate using Boc2O (iv), and the resulting 8c-e were O-perbenzoylated upon treatment with benzoyl chloride (v) to afford the O-and N-protected glucosaminyl derivatives 9c-e. Finally, acid-mediated liberation of the NH2 group in 9c-e (vi) was carried i 76 iii NI * (from 4e) iv 29 (from 4e) * Formation of the expected product was detected, but the compound could not be isolated in a pure state.
Attempted transformation of compounds 2a-e into 5a-e in one step via simultaneous O-debenzylation and reduction of the nitro group by catalytic hydrogenation (Pd(C) or Pd(OH) 2 , cat. ccHCl, EtOH, reflux,) was unsuccessful; in each case, decomposition of the starting materials (2a-e) was observed. Therefore, the synthesis of the glucosamine derivatives 5a-e from 2a-e was performed in a two-step procedure. First, the O-benzylprotecting groups of 2a-e were removed by BCl 3 (i) to afford C-(2 -deoxy-2 -nitro-β-Dglucopyranosyl)azines 4a-e in high yields. Reduction of the nitro group of 4a-e was then examined under two conditions. Catalytic hydrogenation of 4c and 4d (iii) afforded the desired pyrimidine-and pyrazine-containing C-glucosaminyl derivatives 5c and 5d, respectively, in acceptable yields. The same conditions (iii) applied to compounds 4a,b,e resulted in complex product mixtures, in which the desired C-glucosaminyl pyridine 5a, pyridazine 5b and quinoline 5e were detected by TLC analysis; however, they could not be separated in a pure state. Treatment of 4e with Sn powder in the presence of ccHCl (iv) was also carried out to afford the target 5e in acceptable yield.
Conditions: (i) 6 equiv. of benzoyl chloride, 2 equiv. of Zn(OTf) 2   Next, the newly prepared heterocyclic glucosamine derivatives were used as N,Nbidentate ligands in the formation of platinum-group metal half-sandwich complexes.
Treatment of O-perbenzylated 2-glucosaminyl pyridine with dichloro(η 6 -p-cymene)ruthenium(II) and -osmium(II) and dichloro(pentamethylcyclopentadienyl)iridium(III) and -rhodium(III) dimers (Ru-/Os-/Ir-/Rh-dimer) in the presence of the halide abstractor TlPF6 afforded the expected cationic complexes Ru-3a, Os-3a, Ir-3a and Rh-3a, Due to the above difficulties, another three-step procedure starting from 5c-e was applied to obtain the planned 7c-e (Table 3). Thus, the NH 2 group of 5c-e was protected first as a carbamate using Boc 2 O (iv), and the resulting 8c-e were O-perbenzoylated upon treatment with benzoyl chloride (v) to afford the Oand N-protected glucosaminyl derivatives 9c-e. Finally, acid-mediated liberation of the NH 2 group in 9c-e (vi) was carried out, providing the final products 7c-e in high yields.
As mentioned earlier, the 3-(2 -amino-2 -deoxy-β-D-glucopyranosyl)pyridazine 5b could not be obtained in a pure state from 4b ( Table 2). In order to obtain the O-perbenzoylated 9b, a consecutive three-step procedure starting from 4b was conducted to avoid the need to use pure intermediate 5b (Table 3). Thus, the NO 2 →NH 2 transformation was carried out by catalytic hydrogenation of 4b (Table 3, iii), followed by the Boc protection of the amino group of intermediate 5b (iv); subsequent O-perbenzoylation of the resulting 8b (v) furnished the desired C-glucosaminyl pyridazine 9b in acceptable overall yields (27% for three steps). Then, standard N-Boc deprotection (vi) afforded the desired 7b in high yield.
Next, the newly prepared heterocyclic glucosamine derivatives were used as N,Nbidentate ligands in the formation of platinum-group metal half-sandwich complexes.
Our previous studies [45,46] on other series of half-sandwich complexes constructed with glycopyranosyl azole ligands revealed that the O-protection of the hydroxyl groups of the sugar moiety by large, apolar protecting groups played a pivotal role in achieving significant biological effects. Complexes containing O-unprotected monosaccharide-based ligands proved to be biologically inactive. Nevertheless, for a comparative study of the new set of platinum-group metal complexes presented here, compound Ru-5a incorporating O-deprotected 2-glucosaminyl pyridine 5a was also synthesized ( Table 4, entry 7).
In most of the complexations presented in Tables 4 and 5, a single diastereoisomer of the complexes was formed. As an exception, the reactions of 7b yielded complexes Ru-7b, Os-7b, Ir-7b and Rh-7b as mixtures of two diastereoisomers ( Table 5, entries 5-8).
A single crystal of complex Ru-3a was obtained by slow evaporation of a CHCl 3 -MeOH solvent mixture. A search of the Cambridge Structural Database (Ver 5.43, Update November 2021) [56] resulted in 98 hits for similar Ru·Cl·η 6 ··NH 2 ··N coordination. However, our structure is unique, as the Ru-Cl distance is the shortest in this family of compounds by 2.374(5) Å (average: 2.416(15) Å), while the angle of the arene ring and the N-N-Ru plane is rather high, at 60.7 • (average: 57(2) • ). Moreover, none of the hits contains a pyranose ring attached to one coordinated amino nitrogen atom in any position. A more detailed search for Ru·Cl·η 6 ··NH 2 coordination revealed more than 200 hits; the Ru-Cl distance was also in the very short region, with an average of 2.42 Å. Ring puckering analysis [57] indicates that the C1-O5 ring has a chair conformation (Θ = 19.6(18) • , Φ = 275(5) • ), which is in agreement with the NMR data related to the coupling constants of the proton resonances of the sugar skeleton.
X-ray crystallography analysis of Ru-3a provided unequivocal evidence of the existence of a six-membered chelate ring and revealed the spatial arrangement of structural elements in the coordination sphere of the Ru(II) ion ( Figure 2). Thus, following the general convention [58], the absolute configuration of the stereogenic Ru(II) was assigned as R. The absolute configuration is confirmed by the analysis of the anomalous dispersion data as the Flack parameter [59] (Table S3).
For structural elucidation of the prepared compounds, 1 H and 13 C NMR measurements were also performed. Comparison of the 1 H NMR spectroscopic data of the complexes to those of the starting Ru/Os/Ir/Rh dimers and the C-glucosaminyl heterocyclic ligands revealed several significant changes in the chemical shifts, some of which are representatively highlighted by the superposition of the 1 H NMR spectra of Ru-dimer, 3a and Ru-3a ( Figure 3). the new set of platinum-group metal complexes presented here, compound Ru-5a incorporating O-deprotected 2-glucosaminyl pyridine 5a was also synthesized ( Table 4, entry 7). Complexations of the O-perbenzoylated C-glucosaminyl azines 7a-e with the dimeric chloro-bridged platinum-group metal complexes Ru-dimer, Os-dimer, Ir-dimer and Rh-dimer were also performed under the same conditions as described above to afford the expected half-sandwich type complexes Ru-7a-Ru-7e, Os-7a-Os-7e, Ir-7a-Ir-7e and Rh-7a-Rh-7e, respectively, in good to excellent yields ( As a consequence of the complexation, the H-2' signal of the sugar skeleton of 3a shifted upfield by 1.  Table S1). It should be noted that the formation of the minor stereoisomers of Ru/Os/Ir/Rh-7b from ligand 7b resulted in practically no (for 7b → Ir/Rh-7b) or less significant changes (+0.15 ppm for 7b → Ru/Os-7b) in the chemical shift of the H-5 signal (Table S1).  In most of the complexations presented in Tables 4 and 5, a single diastereoisomer of the complexes was formed. As an exception, the reactions of 7b yielded complexes Ru-7b, Os-7b, Ir-7b and Rh-7b as mixtures of two diastereoisomers (   In most of the complexations presented in Tables 4 and 5, a single diastereoisomer o the complexes was formed. As an exception, the reactions of 7b yielded complexes Ru-7b Os-7b, Ir-7b and Rh-7b as mixtures of two diastereoisomers ( ence of a six-membered chelate ring and revealed the spatial arrangement of stru elements in the coordination sphere of the Ru(II) ion ( Figure 2). Thus, following th eral convention [58], the absolute configuration of the stereogenic Ru(II) was assig R. The absolute configuration is confirmed by the analysis of the anomalous disp data as the Flack parameter [59] (Table S3). For structural elucidation of the prepared compounds, 1 H and 13 C NMR me ments were also performed. Comparison of the 1 H NMR spectroscopic data of the plexes to those of the starting Ru/Os/Ir/Rh dimers and the C-glucosaminyl hetero ligands revealed several significant changes in the chemical shifts, some of which ar resentatively highlighted by the superposition of the 1 H NMR spectra of Ru-dimer, Ru-3a ( Figure 3).
As a consequence of the complexation, the H-2' signal of the sugar skeleton shifted upfield by 1.  Table S1). It should be noted that the formation of the minor s somers of Ru/Os/Ir/Rh-7b from ligand 7b resulted in practically no (for 7b → Ir/R   (Table S1). The transformation of Ru-dimer into Ru-3a significantly affected the proton resonances of the p-cymene moiety. For example, the signal of CH3 attached to the benzene ring displayed remarkable upfield shifts (0.46 ppm) as a result of the complexation (E). A similar trend in the appearance of this CH3 signal was observed for all single isomeric Ru(II) and Os(II) complexes (Os/Ir/Rh-3a, Ru/Os-3d, Ru-5a, Ru/Os/Ir/Rh-7a and Ru/Os/Ir/Rh-7c-d) and for the main components of complexes Ru-7b and Os-7b (Δ = δcom- Table S1). In the case of the minor isomers of Ru-7b and Os-7b, the same signal indicated a slight downfield shift (Δ = δcomplex−δdimer = ~+0.1 ppm, Table S1) relative to that of the corresponding Ru-dimer and Os-dimer, respectively.
These data strongly suggest that in each complex obtained as a single isomer (Os/Ir/Rh-3a, Ru/Os-3d, Ru-5a, Ru/Os/Ir/Rh-7a and Ru/Os/Ir/Rh-7c-d) and in the major component of Ru/Os/Ir/Rh-7a, the absolute configuration of the metal center is identical to that of the reference complex, Ru-3a.
A more detailed collection of the comparative spectroscopic data are presented in Table S1 in the Supplementary Materials.

Cell Biology
The complexes described above are intended to replace registered platinum complexes. Platinum complexes constitute the core of the chemotherapy regimen used in ovar- The transformation of Ru-dimer into Ru-3a significantly affected the proton resonances of the p-cymene moiety. For example, the signal of CH 3 attached to the benzene ring displayed remarkable upfield shifts (0.46 ppm) as a result of the complexation (E). A similar trend in the appearance of this CH 3 signal was observed for all single isomeric Ru(II) and Os(II) complexes (Os/Ir/Rh-3a, Ru/Os-3d, Ru-5a, Ru/Os/Ir/Rh-7a and Ru/Os/Ir/Rh-7c-d) and for the main components of complexes Ru-7b and Os-7b Table S1). In the case of the minor isomers of Ru-7b and Os-7b, the same signal indicated a slight downfield shift (∆ = δ complex − δ dimer =~+0.1 ppm, Table S1) relative to that of the corresponding Rudimer and Os-dimer, respectively.
These data strongly suggest that in each complex obtained as a single isomer (Os/Ir/Rh-3a, Ru/Os-3d, Ru-5a, Ru/Os/Ir/Rh-7a and Ru/Os/Ir/Rh-7c-d) and in the major component of Ru/Os/Ir/Rh-7a, the absolute configuration of the metal center is identical to that of the reference complex, Ru-3a.
A more detailed collection of the comparative spectroscopic data are presented in Table S1 in the Supplementary Materials.

C-Glucosaminyl Azines Exert Cytostatic Activity
The complexes described above are intended to replace registered platinum complexes. Platinum complexes constitute the core of the chemotherapy regimen used in ovarian cancer [8]; therefore, we used a cellular model of ovarian cancer, A2780, and primary human dermal fibroblasts as models of non-transformed cells (controls). For the characterization of the complexes, we used an MTT assay after 4 h of treatment for the detection of early toxicity and an SRB assay 48 h after treatment for the detection of cytostasis [60][61][62].
First, we assessed the complexes of ligands 7a-e. The complexes of the pyridineand pyridazine-containing ligands 7a and 7b had superior bioactivity relative to that of the complexes of ligands 7c, 7d and 7e, with pyrimidine, pyrazine and quinoline aglycon parts, respectively (Figures 4 and 5, Table 6). Complexes of 7a-e induced early toxicity, as evidenced by the MTT assays. The complexes of the ligand 7a were the most effective in inducing early toxicity, with IC 50 values ranging between 9 and 14 µM and achieving more than 90% inhibition ( Figure 4, Table 6). Other complexes did not reach over 90% inhibition, although they induced early toxicity (Figures 4 and 5, Table 6). In terms of long-term cytostatic activity, the complexes of 7a and 7b exerted complete inhibition of cell growth in SRB assays, with IC 50 values between 4 and 9 µM ( Figure 4, Table 6). Complexes of ligands 7c-e did not inhibit cell proliferation fully up to 100 µM ( Figure 5, Table 6). The best IC 50 values fell into the low micromolar range; Ru-3a and Ir-3a had IC 50 values of 1.86 and 1.69 µM, respectively.
In general, the IC 50 values of the Ru(II)-containing complexes were lower than those of the Os(II), Ir(III) and Rh(III) analogs constructed with the same ligand (e.g., Ru-7a vs. Ir-7a, Os-7a and Rh-7a; Table 6). In terms of effectiveness, the Ru-containing complexes were followed by the corresponding iridium complexes, then by the osmium and, finally, by the rhodium complexes (Figures 4 and 5, Table 6). Of note, the difference between Ru, Os, Ir and Rh complexes was not as pronounced as we observed in our prior studies with glycosyl azole-type ligands [32,45,46]. Furthermore, the free ligands 7a, 7c, 7d and 7e effectively induced cytostasis (Figures 4 and 5, Table 6).
Next, we assessed 3d and its ruthenium and osmium complexes, Ru-3d and Os-3d. Ru-3d and Os-3d exerted early toxicity on A2780 and human primary dermal fibroblast cells in low micromolar concentrations ( Figure 6B, Table 6). The ligand 3d had no activity in MTT assays on A2780 and human primary dermal fibroblast cells in low micromolar concentrations ( Figure 6, Table 6). Similar to their activity in the MTT assay, Ru-3d and Os-3d exerted early toxicity in SRB assays on A2780 cells in low micromolar concentrations, while 3d did not exert considerable activity in SRB assays on A2780 cells ( Figure 6B, Table 6). In contrast to their activity in MTT assays, 3d and Os-3d did not inhibit cell proliferation in SRB assays, and Ru-3d had only a minor effect on primary human dermal fibroblasts ( Figure 6B, Table 6).
We assessed the effect of the O-deprotection of the carbohydrate moiety, which was shown to abrogate the bioactivity, similar to our previous observations [45][46][47]. We used the free ligand, 5a, which is the deprotected equivalent of 3a and 7a, and its ruthenium complex, Ru-5a. Ligand 5a and its ruthenium complex, Ru-5a, did not exhibit any biological activity on A2780 cells either in MTT or SRB assays (Figure 7, Table 6).

Compound Ru-3a Is Cytostatic in Multiple Carcinoma Cell Lines
Carbohydrate-containing ruthenium, osmium and iridium complexes with similar structures were shown to be effective in a large set of carcinoma, sarcoma and lymphoma cell lines [32,[45][46][47][63][64][65][66][67]; therefore, we assessed the bioactivity of these complexes in other carcinoma cell lines. For this assay, we chose Ru-3a and the ligand, 3a, as this complex had one of the best IC 50 values in A2780 cells (IC 50 = 1.86 µM) and had the best-performing protective group attached.
In agreement with the data presented in Figure 6, Ru-3a inhibited the proliferation of another ovarian cancer cell line (ID8), a glioblastoma cell line (U251), a breast carcinoma cell line (MCF7) and a pancreatic adenocarcinoma cell line (Capan2), with IC 50 values in the low micromolar range falling between 2 and 4 µM (Figure 8, Table 7). Importantly, 3a was also active in these cell lines, with IC 50 values falling between 10 and 30 µM (Figure 8, Table 7).

Complexes with Cytostatic Properties Are Bacteriostatic on Gram-Positive Multiresistant Staphylococcus aureus and Enterococcus isolates
Prior investigations by us [31,32] and others [30][31][32][33][34][35][36][37][38][39][68][69][70][71][72][73][74][75] showed that complexes of the platinum-group metals (platinum, palladium, ruthenium, osmium, iridium and rhodium) can exert bacteriostatic activity. Furthermore, we showed that those compounds were bacteriostatic and cytostatic in neoplasia models [31,32,45,46]. Therefore, we assessed Ru/Os/Ir/Rh-3a, Ru/Os-3d, Ru/Os/Ir/Rh-7a and Ru/Os/Ir/Rh-7b complexes and the corresponding free ligands. Free ligands and rhodium complexes did not exert bacteriostatic activity on any of the investigated strains or isolates. In contrast, the remaining ruthenium, osmium and iridium complexes (Ru/Os/Ir-3a, Ru/Os-3d, Ru/Os/Ir-7a and Ru/Os/Ir-7b) exhibited bacteriostatic activity on the reference strain of Enterococcus faecalis and Staphylococcus aureus, VRE and MRSA, with the exception of Os-3d on the reference strain of Enterococcus faecalis and Ir-7b on the reference strain of Staphylococcus aureus (Figure 9, Table 8). Ruthenium and osmium complexes were characterized by the lowest MIC values, followed by iridium complexes. The ruthenium complexes of 3a and 7a had the lowest MIC values on the reference strain and clinical isolates of Staphylococcus aureus and Enterococcus faecalis, highlighting the superior performance of pyridine-containing complexes in terms of their bacteriostatic activity, similar to their higher performance as cytostatic agents.  Table 6. Color code: black-free ligand (3a or 3d), khaki-ruthenium complex (Ru-3a or Ru-3d), blue-osmium complex (Os-3a or Os-3d), green-iridium complex (Ir-3a), red-rhodium complex (Rh-3a).   Table 6. Color code: black-free ligand (3a or 3d), khaki-ruthenium complex (Ru-3a or Ru-3d), blue-osmium complex (Os-3a or Os-3d), green-iridium complex (Ir-3a), red-rhodium complex (Rh-3a). We assessed the effect of the O-deprotection of the carbohydrate moiety, which was shown to abrogate the bioactivity, similar to our previous observations [45][46][47]. We used the free ligand, 5a, which is the deprotected equivalent of 3a and 7a, and its ruthenium complex, Ru-5a. Ligand 5a and its ruthenium complex, Ru-5a, did not exhibit any biological activity on A2780 cells either in MTT or SRB assays (Figure 7, Table 6). Figure 7. Assessment of the free ligand 5a and its ruthenium complex, Ru-5a, for cytotoxic and cytostatic activity. For MTT and SRB assays, 3 × 10 3 A2780 cells and 1.5 × 10 3 A2780 cells were plated in 96-well plates, respectively. Cells were treated with the compounds in the concentrations indicated for either 4 h for the MTT assay or for 48 h for the SRB assay. Data are represented as average ± SD from three biological replicates; individual assays were performed in duplicate. Values were normalized for vehicle-treated cells; absorbance for vehicle-treated cells equals 1. Statistical significance was assessed using a one-way ANOVA test comparing all values to the lowest concentration of a compound. Before the test, normality was assessed using the Shapiro-Wilk test, and the post hoc test was chosen accordingly. For better visibility, the p values and the distributions are presented in an Excel sheet available at https://figshare.com/s/9ec2a005e6b9e5874c07. Nonlinear regression was performed on the datasets indicated in Table 6. Color code: black-free ligand (5a), khaki-ruthenium complex (Ru-5a).

Compound Ru-3a Is Cytostatic in Multiple Carcinoma Cell Lines
Carbohydrate-containing ruthenium, osmium and iridium complexes with similar structures were shown to be effective in a large set of carcinoma, sarcoma and lymphoma cell lines [32,[45][46][47][63][64][65][66][67]; therefore, we assessed the bioactivity of these complexes in other carcinoma cell lines. For this assay, we chose Ru-3a and the ligand, 3a, as this complex had one of the best IC50 values in A2780 cells (IC50 = 1.86 µM) and had the best-performing protective group attached.
In agreement with the data presented in Figure 6, Ru-3a inhibited the proliferation of another ovarian cancer cell line (ID8), a glioblastoma cell line (U251), a breast carcinoma cell line (MCF7) and a pancreatic adenocarcinoma cell line (Capan2), with IC50 values in the low micromolar range falling between 2 and 4 µM ( Figure 8, Table 7). Importantly, 3a was also active in these cell lines, with IC50 values falling between 10 and 30 µM ( Figure  8, Table 7). Figure 7. Assessment of the free ligand 5a and its ruthenium complex, Ru-5a, for cytotoxic and cytostatic activity. For MTT and SRB assays, 3 × 10 3 A2780 cells and 1.5 × 10 3 A2780 cells were plated in 96-well plates, respectively. Cells were treated with the compounds in the concentrations indicated for either 4 h for the MTT assay or for 48 h for the SRB assay. Data are represented as average ± SD from three biological replicates; individual assays were performed in duplicate. Values were normalized for vehicle-treated cells; absorbance for vehicle-treated cells equals 1. Statistical significance was assessed using a one-way ANOVA test comparing all values to the lowest concentration of a compound. Before the test, normality was assessed using the Shapiro-Wilk test, and the post hoc test was chosen accordingly. For better visibility, the p values and the distributions are presented in an Excel sheet available at https://figshare.com/s/9ec2a005e6b9e5874c07. Nonlinear regression was performed on the datasets indicated in Table 6. Color code: black-free ligand (5a), khaki-ruthenium complex (Ru-5a).  Table 7.  Table 7.  Table 8). Ruthenium and osmium complexes were characterized by the lowest MIC values, followed by iridium complexes. The ruthenium complexes of 3a and 7a had the lowest MIC values on the reference strain and clinical isolates of Staphylococcus aureus and Enterococcus faecalis, highlighting the superior performance of pyridine-containing complexes in terms of their bacteriostatic activity, similar to their higher performance as cytostatic agents.  Color code: khaki-ruthenium complex, blue-osmium complex, green-iridium complex.

Discussion
In this study, we described a set of half-sandwich type platinum-group metal complexes with O-protected C-glucosaminyl heterocyclic N,N-bidentate ligands. Among these, the pyridine-containing complexes had the lowest IC 50 and MIC values, followed by the pyrazine-and pyridazine-containing complexes, while their pyrimidine and quinoline counterparts proved to be inactive ( Figure 10A).

Discussion
In this study, we described a set of half-sandwich type platinum-group metal complexes with O-protected C-glucosaminyl heterocyclic N,N-bidentate ligands. Among these, the pyridine-containing complexes had the lowest IC50 and MIC values, followed by the pyrazine-and pyridazine-containing complexes, while their pyrimidine and quinoline counterparts proved to be inactive ( Figure 10A). Another important structural feature of the complexes is the metal ion with the polyhapto arene/arenyl moiety. In this study, ruthenium complexes with p-cymene and iridium complexes with Cp* achieved the best performance in ovarian cancer cells, followed by osmium complexes with p-cymene and rhodium complexes with Cp* (Figure 10A). This is at odds with our previous observations, as we identified osmium complexes with the most potent cytostatic properties, followed by ruthenium complexes and, with a large gap, iridium complexes [32,45,46]. Furthermore, rhodium complexes were inactive in cellular models of carcinomas [32,45,46] in contrast to the presently reported results. However, the effect of the metal ions with polyhapto arene/arenyl moiety proved to be similar in terms of the bacteriostatic effects as in our previous experience [31,32], where the ruthenium complexes had the lowest MIC values, followed by osmium and iridium complexes ( Figure 10A). Rhodium complexes did not exhibit bacteriostatic properties. Another important structural feature of the complexes is the metal ion with the polyhapto arene/arenyl moiety. In this study, ruthenium complexes with p-cymene and iridium complexes with Cp* achieved the best performance in ovarian cancer cells, followed by osmium complexes with p-cymene and rhodium complexes with Cp* (Figure 10A). This is at odds with our previous observations, as we identified osmium complexes with the most potent cytostatic properties, followed by ruthenium complexes and, with a large gap, iridium complexes [32,45,46]. Furthermore, rhodium complexes were inactive in cellular models of carcinomas [32,45,46] in contrast to the presently reported results. However, the effect of the metal ions with polyhapto arene/arenyl moiety proved to be similar in terms of the bacteriostatic effects as in our previous experience [31,32], where the ruthenium complexes had the lowest MIC values, followed by osmium and iridium complexes ( Figure 10A). Rhodium complexes did not exhibit bacteriostatic properties.
Prior studies have shown that the chemical composition of the protective groups of the sugar part plays a key role in the bioactivity of monosaccharide-containing half-sandwich complexes, showing O-benzoyl-protected complexes to be the most effective [32,47]. In this study, besides the O-benzoyl-protected C-glucosaminyl pyridine 7a and pyrazine 7d complexes, their O-benzylated counterparts (complexes 3a and 3d, respectively) were also tested. Importantly, the benzyl-protected compounds had better IC 50 values in cancer cell models than the benzoyl-protected compounds ( Figure 10B). Interestingly, the bacteriostatic properties of the benzyl/benzoyl-protected compounds did not differ drastically (Figures 9 and 10A).
Based on the positive logD values (Table 6), all complexes with O-benzyl and Obenzoyl protective groups are lipophilic. The lipophilic character of the complexes is a prerequisite of their biological activity [31,32,[45][46][47]63]. In fact, among the currently identified complexes, the readout of apolarity (logD) and the IC 50 value correlate ( Figure 10C). Apparently, increasing the apolar character of the complexes improves the biological effectiveness, which is further strengthened by the fact that when the protective groups are absent, as in Ru-5a or in comparable members of the previously reported series [45,46], the bioactivity of the complexes is lost.
An unexpected observation was that the C-glucosaminyl heterocycles used as ligands, namely pyridines 3a and 7a, pyridazine 7b, pyrimidine 7c, pyrazines 3d and 7d and quinoline 7e, showed cytostatic effects, which proved comparable to those of the respective complexes in A2780 cells. In addition, 3a induced cytostasis in primary human fibroblasts. To the best of our knowledge, such effects have not yet been described with C-glycosyl heterocycles; therefore, this finding deserves further investigation.
Importantly, the complexes with a cytostatic property were less active on primary, nontransformed human dermal fibroblasts. While this property of the complexes suggests a selectivity for transformed neoplastic cells over non-transformed cells, the anticipated therapeutic window is relatively narrow as compared to previous observations of complexes with ligands of similar [32,45,46] or different structure [76,77].

General Procedure I for the Preparation of C-(2 -Deoxy-2 -nitro-3 ,4 ,6 -tri-O-benzyl-β-D-glucopyranosyl)heterocycles 2a-e
Method A: In a dry, round-bottom flask, the corresponding halogenated heterocycle (4.33 mmol, 2 eq.) was dissolved in freshly distilled dry THF (10 mL). The stirred solution was cooled down to −78 • C, and a 2.5 M solution of n-butyllithium in n-hexane (1.74 mL, 4.33 mmol, 2 eq.) was added dropwise over 10 min, with stirring continued for 5 min to form the corresponding lithiated heterocycle. In another dry, round-bottom flask containing activated 4 Å molecular sieves (powder, 0.2 g), 2-nitroglucal 1 (1.0 g, 2.17 mmol) was dissolved in freshly distilled dry THF (10 mL). After cooling this solution to −78 • C, the solution of freshly prepared lithiated heterocycle was added. The reaction mixture was then stirred at −78 • C, and the transformation was monitored by TLC (1:4 EtOAc-hexane). When the TLC indicated complete disappearance of 1, the reaction was quenched by the addition of sat. aq. NH 4 Cl solution (100 mL) and allowed to warm to rt. The molecular sieves were then filtered off. The filtrate was diluted with EtOAc (200 mL) and extracted with water (100 mL) and brine (100 mL). The separated organic phase was dried over MgSO 4 and filtered, and the solvents were removed under reduced pressure. The residue was purified by column chromatography.
Method B: In a dry, round-bottom flask, 2-nitroglucal 1 (1.0 g, 2.17 mmol) and the corresponding halogenated heterocycle (2.60 mmol, 1.2 eq.) were dissolved in freshly distilled dry THF (20 mL). The stirred solution was cooled down to −78 • C, and a 2.5 M solution of n-butyllithium in n-hexane (1.04 mL, 2.60 mmol, 1.2 eq.) was added over 15 min by means of a syringe pump. Stirring was continued for an additional 15 min at the same temperature. After completion of the reaction (~0.5 h), as judged by TLC (1:2 EtOAc-hexane), the reaction was quenched by the addition of sat. aq. NH 4 Cl solution (100 mL) and allowed to warm to rt. The mixture was diluted with EtOAc (200 mL) and extracted with water (100 mL) and brine (100 mL). The separated organic phase was dried over MgSO 4 and filtered, and the solvents were removed under reduced pressure. The residue was purified by column chromatography. Heterocycle 2a-e, 3a to obtain compounds 4a-e and 5a A solution of the corresponding O-perbenzylated C-glycopyranosyl heterocycle 2a-e or 3a in dry CH 2 Cl 2 (10 mL/100 mg substrate) was cooled down to −78 • C, and a 1 M solution of BCl 3 in CH 2 Cl 2 (5 eq.) was added dropwise over 5 min. The reaction mixture was stirred at this temperature until the TLC (9:1 CHCl 3 -MeOH) showed the completion of the reaction. Then, the reaction was quenched by the addition of MeOH (10 mL) and allowed to warm to rt. The solvents were then removed under reduced pressure, and the residue was purified by column chromatography.

General Procedure III for the Preparation of
To a solution of the corresponding C-(2 -deoxy-2 -nitro-3 ,4 ,6 -tri-O-benzyl-β-Dglucopyranosyl)heterocycle 2a-d in dry dichloroethane (10 mL/100 mg substrate), Zn(OTf) 2 (2 eq.) and benzoyl chloride (6 eq.) were added, and the reaction mixture was stirred at rt. After completion of the reaction, as judged by TLC (1:2 EtOAc-hexane), the reaction mixture was diluted with CH 2 Cl 2 (40 mL) and extracted with sat. aq. NaHCO 3 solution (50 mL), then with water (50 mL). The separated organic phase was dried over MgSO 4 and filtered, and the solvents were removed under diminished pressure. The crude product was purified by column chromatography.

General Procedure IV for the Preparation of C-(2 -(tert-Butoxycarbonyl)amino-2 -deoxy-β-D-glucopyranosyl)heterocycles 8c-e
To a solution of the corresponding C-(2 -amino-2 -deoxy-β-D-glucopyranosyl)azine 5c-e in a 1:1 mixture of water and 1,4-dioxane (5 mL/50 mg substrate), Boc 2 O (2 eq.) was added, and the reaction mixture was stirred at rt. When the TLC (9:1 CHCl 3 -MeOH) showed complete transformation of the staring material (~1 day), the solvents were removed under reduced pressure. The residue was purified by column chromatography. To a solution of the corresponding C-(2 -(tert-butoxycarbonyl)amino-2 -deoxy-β-Dglucopyranosyl)azine 8c-e in dry pyridine (5 mL/100 mg substrate), benzoyl chloride (1.2 eq./OH group) was added at rt. The reaction mixture was stirred at 60 • C for 1 h. Since the TLC (1:1 EtOAc-hexane) showed the incompleteness of the reaction, an additional portion of benzoyl chloride (1.2 eq./OH group) was added to the reaction mixture, and heating was continued for 1 h. The reaction mixture was allowed to cool to rt and further stirred overnight. The reaction mixture was then diluted with CH 2 Cl 2 (50 mL) and extracted with sat. aq. solution of NaHCO 3 (25 mL), then with water (25 mL). The separated organic phase was dried over MgSO 4 and filtered, and the solvents were removed under diminished pressure. The residue was purified by column chromatography. 5.1.7. General Procedure VI for the Preparation of C-(2 -Amino-2 -deoxy-3 ,4 ,6 -tri-O-benzoyl-β-D-glucopyranosyl)heterocycles 7b-e from Compounds 9b-e The corresponding C-(2 -(tert-butoxycarbonyl)amino-2 -deoxy-3 ,4 ,6 -tri-O-benzoylβ-D-glucopyranosyl)azine 9b-e was dissolved in dry CH 2 Cl 2 (5 mL/100 mg substrate), and trifluoroacetic acid (2 eq.) was added. The reaction mixture was stirred at rt until the TLC (95:5 CHCl 3 -MeOH or 1:1 EtOAc-hexane) indicated complete disappearance of the starting material (~1 h). The solvent and the excess CF 3 COOH were then removed under reduced pressure. The residue was dissolved in CH 2 Cl 2 (50 mL) and extracted with sat. aq. solution of NaHCO 3 (25 mL) and with water (25 mL). The separated organic phase was dried over MgSO 4 and filtered, and the solvent was removed under diminished pressure. The residue was purified by column chromatography.

General Procedure VII for the Synthesis of Half-Sandwich Platinum-Group Metal Complexes
To a solution of the corresponding complex dimer (Ru-dimer, Os-dimer ([(η 6 -pcym)M II Cl 2 ] 2 (M = Ru, Os) or Ir-dimer, Rh-dimer [(η 5 -Cp*)M III Cl 2 ] 2 (M = Ir, Rh)) in CH 2 Cl 2 (1 mL/10 mg dimer), the appropriate C-glucosaminyl azine (1.9-2.3 eq.) and TlPF 6 (2 eq.) were added. To this stirred reaction mixture, MeOH (1 mL/10 mg dimer) was added at rt in order to accelerate the precipitation of the TlCl. The heterogeneous mixture was then continued further stirred at rt, and the completion of the reaction was monitored by TLC (95:5 CHCl 3 -MeOH). When TLC showed total disappearance of the starting dimer (~1 h), the precipitated TlCl was filtered off. The resulting solution was evaporated under diminished pressure. The remaining crude complex was purified by column chromatography and/or crystallization.   Compound 2a (0.19 g, 0.35 mmol) and Zn powder (0.69 g, 10.55 mmol, 30 eq.) were suspended in a solvent mixture of THF (10 mL) and water (5 mL). This suspension was cooled down in an ice bath, and ccHCl solution was added (0.7 mL, 8.14 mmol, 23 eq.). The reaction mixture was stirred at rt until the TLC (1:1 EtOAc-hexane) showed total consumption of the starting material (1 h). The reaction was quenched by the addition of sat. aq. NaHCO 3 solution (50 mL). The insoluble inorganic salts and the rest of the Zn were filtered off, and the remaining solution was extracted with CH 2 Cl 2 (2 × 50 mL). The combined organic phase was extracted with water (50 mL), then with brine (50 mL), dried over MgSO 4       2-(2 -Amino-2 -deoxy-3 ,4 ,6 -tri-O-benzoyl-β-D-glucopyranosyl)pyridine (7a) Compound 6a (0.10 g, 0.17 mmol) and Zn powder (0.11 g, 1.71 mmol, 10 eq.) were suspended in a solvent mixture of THF (10 mL) and water (5 mL). To this stirred mixture, a 2 M aq. solution of HCl was added (2.6 mL, 5.14 mmol, 30 eq.). The reaction mixture was further stirred at rt until the TLC (95:5 CHCl 3 -MeOH) showed total consumption of the starting material (5 h). The reaction was quenched by the addition of sat. aq. NaHCO 3 solution (50 mL). The insoluble inorganic salts and the rest of the Zn were filtered off, and the remaining solution was extracted with CH 2 Cl 2 (2 × 50 mL). The combined organic phase was extracted with water (50 mL), then with brine (50 mL), dried over MgSO 4 and filtered. The solvent was removed under reduced pressure. The residue was purified by column chromatography (100: 13  A degassed, vigorously stirred suspension of 10% Pd(C) (65 mg) in dry EtOH (13 mL) was saturated with H 2 , and compound 4b (0.13 g, 0.48 mmol) was added. The reaction mixture was heated at reflux temperature until the TLC (3:2 CHCl 3 -MeOH) indicated complete conversion of the starting material. After completion of the reaction (3 h), the catalyst was filtered off through a pad of celite and washed with EtOH. The resulting solution was then evaporated under reduced pressure. Purification of the residue by column chromatography (3:2 CHCl 3 -MeOH) yielded 100 mg of a white, amorphous solid containing the desired 3-(2 -amino-2 -deoxy-β-D-glucopyranosyl)pyridazine 5b, along with unidentified impurities. This mixture was dissolved in a solvent mixture of water (5 mL) and 1,4-dioxane (5 mL), and Boc 2 O (0.21 g, 0.96 mmol) was added. The reaction mixture was stirred at rt until the TLC (9:1 CHCl 3 -MeOH) showed complete transformation of 5b (1 day). Then, the solvents were removed under reduced pressure. Column chromatographic purification of the residue (9:1 CHCl 3 -MeOH) resulted in 70 mg of 3-(2 -(tert-butoxycarbonyl)amino-2deoxy-β-D-glucopyranosyl)pyridazine (8b) contaminated with inseparable impurities. To a solution of the resulting 8b in dry pyridine (5 mL), benzoyl chloride (0.2 mL, 1.72 mmol) was added at rt. The reaction mixture was stirred at 60 • C for 1 h. Since the TLC (1:1 EtOAc-hexane) showed incompleteness of the reaction, an additional portion of benzoyl chloride (0.2 mL, 1.72 mmol) was added to the reaction mixture, and heating was continued for 1 h. The reaction mixture was allowed to cool to rt and further stirred overnight. The reaction mixture was then diluted with CH 2 Cl 2 (50 mL) and extracted with sat. aq. solution of NaHCO 3 (25 mL), then with water (25 mL). The separated organic phase was dried over MgSO 4 and filtered, and the solvents were removed under diminished pressure. The residue was purified by column chromatography (1:1 EtOAc-hexane) to afford the title compound 9b (83 mg, 27% for 3 steps) as a white, amorphous solid. R f = 0.28 (1:1 EtOAc-hexane. 1  Structural parameters such as bond length and angle data were in the expected range (Table S5), except the short Ru-Cl distance. The solid-state structure was stabilized by strong N-H..Cl and N-H..F, as well as weak C-H..Cl hydrogen bonds ( Figure S2 and Table S4).

Determination of the Distribution Coefficients (logD)
The logD values of the newly synthesized complexes were determined according to a procedure described in our previous publications [45].

Chemicals for Biology Experiments
All chemicals used in the cell biology and biochemistry assays were obtained from Sigma-Aldrich unless otherwise stated. The free ligands and complexes investigated in this study were dissolved in dimethyl-sulfoxide for biology experiments, and 0.1% dimethylsulfoxide was used as a vehicle control.

Cell Lines
Cells were cultured under standard cell culture conditions: 37 • C, 5% CO 2 , humidified atmosphere.

Bacterial Reference Strains
The reference strains of Staphylococcus aureus (ATCC29213) and Enterococcus faecalis (ATCC29212) were purchased from the ATCC (Manassas, VA, USA).

Clinical Isolates of S. aureus and E. faecium
We used a set of clinical isolates of S. aureus and E. faecium that were collected at the Medical Center of the University of Debrecen (Hungary) between 1 January 2018. and 31 December 2020. The isolates were reported in [31] and are presented in Table 9. The clinical isolates were identified using a Microflex MALDI-TOF mass spectrometer (Bruker, Billerica, MA, USA). The antibiotic susceptibility of the isolates was tested following the European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines, which were valid at the time of collection.

Methylthiazolyldiphenyl-Tetrazolium Bromide (MTT) Reduction Assay
An MTT reduction assay was measures the activity of mitochondrial complex I and can be used to detect toxicity [60,61]. The assay was performed in a manner similar to that described in [84]. Briefly, cells were plated in 96-well plates the day before the assay. Cells were treated with the compounds for 4 h; then, MTT was added to a 0.5 mg/mL final concentration, and cells were incubated at 37 • C in a cell incubator for 40-60 min as a function of the cell line being assessed. Culture medium was removed, the reduced MTT dye was dissolved in dimethyl-sulfoxide, and plates were measured in a plate photometer (Thermo Scientific Multiscan GO spectrophotometer, Waltham, MA, USA) at 540 nm. On each plate, wells were designed to contain vehicle-treated cells. In calculations, the readings for these wells were considered to 1, and all readings were expressed relative to these values. fit" utility of GraphPad, which yielded IC 50 and Hill slope values if the sigmoid curves reached a plateau of inhibition and there was no decrease between two subsequent data points or when inhibition was over 90%. In other cases, the percentage of inhibition was taken for the maximum concentration (100 µM).