Aminoglycosides: Molecular Insights on the Recognition of RNA and Aminoglycoside Mimics

Abstract

RNA is increasingly recognized for its significant functions in biological systems and has recently become an important molecular target for therapeutics development. Aminoglycosides, a large class of clinically significant antibiotics, exert their biological functions by binding to prokaryotic ribosomal RNA (rRNA) and interfering with protein translation, resulting in bacterial cell death. They are also known to bind to viral mRNAs such as HIV-1 RRE and TAR. Consequently, aminoglycosides are accepted as the single most important model in understanding the principles that govern small molecule-RNA recognition, which is essential for the development of novel antibacterial, antiviral or even anti-oncogenic agents. This review outlines the chemical structures and mechanisms of molecular recognition and antibacterial activity of aminoglycosides and various aminoglycoside mimics that have recently been devised to improve biological efficacy, binding affinity and selectivity, or to circumvent bacterial resistance.

Keywords

aminoglycosides aminoglycoside mimics RNA recognition 2-deoxystreptamine (2-DOS) mimics aminoglycoside resistance aminoglycoside toxicity

Introduction

RNA performs pivotal biological functions when organisms, including bacterial and viral pathogens, replicate. Developing small molecules that can interact with RNA and interrupt undesired cellular activities is a promising new path for drug design.¹ To date, RNA remains largely unexplored as a drug target and may provide untapped opportunities to develop novel therapeutics.

Antisense and RNAi strategies have been widely recognized as a powerful tool to control RNA genes and regulate development of various organisms and cellular functions of interest.^2–4 However, these strategies are rather limited in practical therapeutic application due to complex and unusual RNA folding patterns and lack of efficient cellular delivery methods for antisense nucleotides.⁵ This makes development of small RNA binding molecules imperative.

RNAs can fold into complex three-dimensional structures such as loops, bulges and pseudo-knots, and consequently, they present a unique challenge for custom design of the RNA binding molecules. Although some progress has been made in recent years in understanding how small molecules recognize RNA,^6–8 there are still very few examples of small molecules that bind to natural RNA structures with high affinity and specificity. So far, there has been no general rule found that can be applied to design small molecules targeting specific RNA sequences or structures.

Since they have been shown to selectively bind to a variety of RNA targets, aminoglycosides⁹ have become a central focus in an effort to understand underlying principles of RNA recognition by small-molecules. Recent NMR and X-ray crystal structures of RNAs complexed with various aminoglycosides provide significant insights on the RNA recognition by small molecules at the molecular level.^10–12 Based on this structural information, many chemical modifications have been devised to improve the binding affinity and selectivity and to increase the antibacterial efficacy of existing aminoglycosides. Some modifications were aimed at synthetic aminoglycosides that are less prone to bacterial resistance while others targeted RNAs from sources other than bacteria such as viral or oncogenic RNAs. This review examines various chemical approaches taken to realize these goals.

Chemical Structures of Aminoglycosides

Clinically significant aminoglycosides include both naturally occurring drugs and some semi-synthetic derivatives, all of which have a highly-conserved aminocyclitol ring (ring II, Figure 1), a central scaffold that is linked to various aminosugar moieties.^13,14 The aminocyclitol ring is comprised primarily of 2-deoxystreptamine (2-DOS) and has 1,3-diamino functionality and three or four hydroxyl groups that provide anchoring points for aminosugars. Aminoglycosides can be divided into 3 subclasses depending on the substitution pattern: 4-monosubsituted, or 4,5- or 4,6-disubsituted (Fig. 1). Aminoglycosides in each subclass show close structural resemblance. Although 2-deoxystreptamine or 2-deoxy-myo-inosa-1,3-diamine is actually derived from D-glucose-6-phosphate biosynthetically, the numbering system is based on streptamine's biogenic precursor myo-inositol as described by Fletcher.¹⁵

Figure 1.

Chemical structures of the three representative classes of aminoglycosides and their substitution sites. The central scaffold 2-deoxystreptamine (2-DOS) is ring II.

Neamine, paromamine, and apramycin fall under the 4-monosubstituted aminoglycosides. Paromamine and neamine differ only in the R₁ substituent and are typically not used alone as drugs, leaving apramycin being the only 4-monosubstituted compound that is actually used pharmaceutically. Apramycin is unique in that its ring I is bicyclic.

Ribostamycin, butirosin B, neomycin B, paromomycin, and lividomycin A belong to 4,5-disubstituted aminoglycosides. These compounds can be further divided into three groups based on the additional ring III and its substituents. Lividomycin A is the only aminoglycoside in this category that contains a fifth ring.

4,6-disubstituted aminoglycosides includes kanamycin A, kanamycin B, kanamycin C, tobramycin, and amikacin in one category and gentamicin C1, gentamicin C2, gentamicin C1A, and geneticin in another. The major differences between these two subclasses is the variation of an extra R group attached to the 6’ position of ring I as well as the ring III connected to the 2-DOS.

Recognition of RNAs by Aminoglycosides

Aminoglycosides have been recognized to have binding capabilities with many different functional RNAs such as the prokaryotic ribosomal A-site,^16–18 HIV TAR,¹⁹ HIV RRE,²⁰ Group I intron,²¹ RNase P,²² tmRNA²³ and the eukaryotic A site²⁴ even in some cases with relatively low micromolar binding affinities.

The binding of the aminoglycosides to these target RNAs is mediated through two different interactions: (1) hydrogen bonding between amino and hydroxyl functional groups of aminoglycosides and RNA bases,²⁵ and (2) electrostatic interactions between the negatively charged phosphate backbone of the RNA and positively charged amino functional groups of the aminoglycosides.²⁶ The latter dominates the interactions between aminoglycosides and target RNAs due to the presence of multiple amines among the aminosugar rings (ring I, III, IV and V), which makes RNA-aminoglycoside binding highly promiscuous. For instance, neomycin binds to bulge regions of unrelated RNA sequences from the 16S ribosome,²⁷ HIV TAR,²⁸ HIV RRE,^20,29 and Group I intron^30,31 with affinities in the low micromolar range. The ambiguous binding characteristic of aminoglycosides originates not only from the electrostatically driven binding mode but also from inherent conformational adaptability of aminoglycosides.³² The glycosidic connection assumes a variety of conformations and permits aminoglycosides to structurally adjust to diverse RNA targets.

Experiments have shown that aminoglycoside binding is somewhat dependent on the size of an asymmetric interior loop of the target RNAs rather than its sequence.³³ These examples demonstrate that the design of small RNA-binding molecules is highly dependent on experimental trials and makes sequence- and/or site-specific recognition by small molecules a very difficult task.

Highly Conserved Interactions Observed by the Ring I and Ring II (2-DOS)

Despite the convincing facts that aminoglycosides recognize RNAs in a promiscuous manner, studies suggest that there are some sequence-dependent elements in the recognition of target RNAs. A wide variety of RNAs, such as A-site rRNA, ribozymes, and HIV-1 RRE and TAR, are known to bind to various aminoglycosides. These RNAs contain 5‘-GU-3’ or 5‘-GG-3’ base steps in common next to the nucleotides in bulges or in noncanonical stem structures. A similar binding pattern was observed when A-site RNA sequence was treated with aminoglycosides of various sizes and substitution patterns on ring II. Specifically, the crystal structures of six aminoglycoside antibiotics (neamine, gentamicin C1A, kanamycin A, ribostamycin, lividomycin A and neomycin B) and oligonucleotides containing the decoding A site of bacterial ribosome revealed that rings I and II (2-DOS) are essential for recognition of A-site RNA, and interactions between the rings I and II of all aminoglycosides and the target RNA are highly conserved and sequence-specific (Fig. 2).¹¹ The two amino groups of the ring II (2-DOS) unanimously recognizes the 5‘-GU-3’ base step (bold) in A-site through hydrogen bonding with N7 of G1494 and O4 of U1495 and electrostatic interactions with negatively-charged phosphate backbone that are conserved throughout all six aminoglycosides. The puckered sugar ring I is inserted into the A site helix by stacking against G1491 and forms a pseudo base pair with two hydrogen bonds to the Watson-Crick sites of the universally conserved A1408. This clearly suggests that contacts made by the ring I and ring II (2-DOS) of various aminoglycosides to the target RNA are somewhat independent of their structural context of aminosugar subunits and more dependent on helical sequence.

Figure 2.

Sequence-specific recognition of A-site RNA by rings I and II (2-DOS) of six aminoglycosides of different sizes and substitution patterns. Highly conserved hydrogen-bonding (,,,,,,) and electrostatic interactions (––) are indicated with dashed lines.

The sequence-specific recognition pattern of 2-DOS is further supported by the following mutation study. A mutation of G1494A in the target site was deleterious for paromomycin binding because it prevented specific hydrogen bonding of 2-DOS to the RNA.¹⁸ This result indicates that the 2-DOS ring of paromomycin strictly recognizes the 5‘-G(N7)-U(O4)-3’ sequence but not 5‘-A(N7)-U(O4)-3’, probably because of a steric clash between the bulky amino group of A and the 2-DOS moiety.

Strikingly, the same binding pattern of 2-DOS was demonstrated when Puglisi and coworkers treated isolated 2-DOS, without the aminosugar subunit attached, with wild type A-site of 16S rRNA (Fig. 3).¹² The binding activity between 2-DOS and A-site RNA was monitored by high-resolution NMR techniques. The 2-DOS specifically recognized the two 5‘-GU-3’ base steps (G1494-U1495 and G1405-U1406) even though their binding affinity is low (~1 mM). A titration experiment with the RNA whose A1492 bulge nucleotide was deleted confirmed that the bulge residue of A-site RNA was required to open the major groove for accommodation of the deoxystreptamine molecule.

Figure 3.

Recognition of 5‘-GU-3’ base sequence step by the 2-DOS monomer that does not have any aminosugar substituent.

2-DOS Mimics

Universal existence of the 2-DOS moiety at the central position of almost all clinically important aminoglycosides suggests its crucial role in RNA recognition and biological activity. Therefore, mimicking the 2-DOS structure has been of great interest in preparing novel aminoglycoside analogs. The 2-DOS mimics that have been reported so far can be divided into two different groups based on whether an aminosugar subunit was included in the mimic or not (Fig. 4).

Figure 4.

The 2-DOS mimics without or with an aminosugar subunit. The 2-DOS mimics with an aminosugar subunit use mostly neamine as the structural platform for modification.

Although most 2-DOS mimics are based on the neamine backbone as the building platform, some use 2-DOS as the sole component, as 2-DOS alone has demonstrated RNA-binding capability without the assistance of aminosugars. Wang et al. demonstrated this example (1) by removing aminosugar rings from the original aminoglycoside structures, leaving only the core 2-DOS as the main constituent of the compound.³⁴ This compound demonstrated strong affinity (K_d value of 88 µM) for the target RNA A-site and inhibited bacterial translation. This example supports the idea of generating small synthetic compounds that mimic the conformation and function of aminoglycosides without using aminosugar residues.

Following are examples that use 2-DOS as the central scaffold and replace the aminosugar ring I of neamine with other structural motifs. Ding and coworker prepared heterocyclic 2-DOS derivatives 2 based on modeling studies.³⁵ Most 2-DOS analogs demonstrated a modest increase in binding affinity upon conjugation with a variety of heterocycles, while the binding affinity of the compound containing an electron-withdrawing group (CF₃) did not show any increase.

Hermann et al prepared a small library of monomeric 2-DOS analogs 3 conjugated with non-sugar heterocycles through an amide bond (Fig. 7).³⁶ Although some compounds show slightly improved binding affinity to the target RNA A-site of bacterial 16S RNA, most did not, suggesting that the rigid amide linkage locked the non-sugar scaffold in an unfavorable conformation for the recognition of the target RNA.

3,5-Diamino-piperidinyl triazines (DAPT) 4 were reported as novel antibacterial translational inhibitors using cis-3,5-diaminopyridine (DAP) to mimic the 2-DOS scaffold.³⁷ Many of the DAPT compounds that behaved as potent inhibitors of bacterial protein synthesis did not show corresponding antibacterial activity. The potency of the DAPT compounds in the antibacterial activity was finally achieved when the R₂ substituent was replaced by aromatic structures that facilitated better cell membrane penetration. Recently, van Delft and coworkers reported a carbohydrate mimic of the 2-DOS that can be used to prepare conformationally constrained aminoglycosides.³⁸

Aminoglycosides like neomycin show binding affinity to various RNA targets such as the 16S ribosome,³⁹ HIV TAR,⁴⁰ and HIV RRE,²⁰ but do not typically bind to RNA hairpin loops, which are a major RNA secondary structural motif. However, when Hergenrother and coworkers prepared simple dimers of 2-DOS (5) by connecting two 2-DOS molecules with various aliphatic and aromatic linkers of different sizes, they were found to bind tightly to RNA hairpin loops of various sizes from 4 to 8 nucleotides, most of which contain 5‘-GU-3’ sequence.⁴¹ The dimers with the aromatic linkers exhibited slightly tighter binding over their aliphatic counterparts.

These examples show that novel RNA binding molecules may be built solely based on 2-DOS without the assistance of conjugated aminosugar(s), and the intrinsic RNA binding affinity of the 2-DOS may be useful targeting not only helical RNAs whose base pairing is disrupted, but also RNAs of other types of secondary structure such as loops that conventional aminoglycosides do not bind.

Many other 2-DOS mimics were prepared based on the neamine backbone. To have a better understanding of the effect of stereospecificity of aminoglycosides, Rando and co-workers synthesized and compared the activity of (+)-neamine 6, (-)-neamine ent-6, its 5-positional isomers 7, ent-7 as well as 6-positional isomers 8, ent-8.⁴² The enantiomeric series ent-8 was found to be the most effective inhibitor among this series, which allows 1-NH₂ and 3-NH₂ to face the active site. The isomer ent-6 with both amino groups pointing away from the active site exhibited the least potent activity, while 7 and ent-7 with only one of the amino groups away from active site exhibited less binding affinity.

In an effort to mimic the unique spatial arrangement of the two amino functional groups in 2-DOS, the Hermann group prepared two series of compounds by replacing the 2-DOS in neamine with azepane- and piperidine-based structures. Azepane-glycosides 9 and 10 exhibited moderate antibacterial activity against Gram-positive Staphylococcus aureus and retained this activity against aminoglycoside-resistant strains.⁴³ This result substantiates the importance of the bacterial decoding site as a valuable target for future development of novel antibiotics. Compounds 11 and 12, designed based on 3-(aminomethyl)-piperidine,^44,45 demonstrated higher activity and specificity towards the bacterial targets than eukaryotic RNAs.

While conformationally-restricted azepane- and 3-(aminomethyl)-piperidine-based compounds 9–12 showed moderate antibacterial activities, the acyclic 2-DOS mimics 13 that have increased flexibility have a detrimental effect on their RNA binding efficacy.⁴⁶

Since Wong and coworkers found that the 5-position is sensitive for aminoglycoside-RNA recognition, small libraries of neamine mimetics were reported with various substituents at the 5‘-OH position. Polyamine and 6-aminoquinoline analogs 14 demonstrated a high affinity for the oncogenic Bcr-Abl and PAX3-FKHR single-stranded mRNA breakpoint and a capability to regulate gene expression.⁴⁷ This provides evidence that aminoglycosides can be used to target not only bacterial and viral but also oncogenic mRNAs. Short semi-peptidic moieties were also introduced on the 5‘-OH position of the neamine scaffold by combinatorial synthesis to form analogs like 15 in an attempt to inhibit interactions between HIV-1 RRE and Rev protein. But, the inhibition activity was only slightly better than neomycin B.⁴⁸ Other neamine analogs including 16 were reported with modifications at 6‘-OH and 1-NH₂ positions.⁴⁹ The substituent on 1-NH₂ was adopted from another aminoglycoside butirosin. This compound 16 was found to be especially interesting because it appears to be uncompromised by aminoglycoside-resistant enzymes and show a considerable increase in antibacterial activity. The crystal structure of the complex between 16 and A-site of the ribosome suggested that existence the AHB substituent on the 1-NH₂ hinders proper complex formation with key aminoglycoside-modifying enzymes.

Aminosugar and Aminoglycoside Mimics

A number of synthetic aminosugars and aminoglycosides have been prepared in an effort to accomplish several goals including: mimicking the core aminosugar rings for enhanced antibacterial activity, limiting the overall structural flexibility of the aminoglycosides, and addressing recognition of the conventional aminoglycosides by the aminoglycoside-modifying enzymes in resistant strains (Fig. 5). Below are relevant examples of aminosugar and aminoglycoside modifications.

Figure 5.

Aminosugar and aminoglycoside modifications.

One of the aminosugar mimics was achieved by –NH₂ and –OH functional groups at various positions of ring I, II and IV of aminoglycosides to find optimal substitution patterns for aminosugars. Wong and coworkers prepared a series of compounds by placing –NH₂ and –OH functional groups at various positions of the ring I in neamine and found that 2’,6‘-diamino substitutions on ring I to be the most effective A-site binder.⁵⁰ The mono-amino derivative 17 showed activity against E. coli, P. aeruginosa, as well as S. aureus. However, its MIC was less than that of neamine, which can be explained by the loss of the highly conserved hydrogen bonds and electrostatic interactions observed in crystal structures. This suggests that neamine is the most promising pseudodisaccharide core that should be kept for further development.

Similar approaches have been performed on the ring III of tobramycin by replacing the ring with various mono- or diaminosugars.⁵¹ The tobramycin derivatives were tested for binding affinity with several conserved RNA sequences from bacterial and viral sources. Surprisingly, no direct correlation was observed between the dissociation constants and the number of charges on the derivatives, considering that the binding of aminoglycosides to RNA targets is driven by electrostatic interactions to a large extent. Chang and coworkers replaced the ring III of pyranmycin with various L-aminosugars and L-pyranoses.⁵²

Some aminosugar mimetics did not include the 2-DOS core in its modified structures. A library of compounds such as 18 have been developed for the discovery of small molecules using a 1,3-hydroxyamine motif.^53–55 Wong and coworkers prepared a library of compounds 18a by substituting R₁- and R₂-positions with highly basic amino acids and amine-containing linear chains, repectively.⁵⁴ This library of compounds recognized both wild type 16 S rRNA and several related mutant sequences, which proved that the hydroxylamine core motif is ideal for the design of high affinity RNA binders. However, when Wengel and coworkers introduced substituents on R₂- or R₃-positions to make compounds like 18b, they were only minimally active against various target RNAs.⁵³ Sugar diamino acids (SDAs) with unprotected amino groups represent a new class of potential aminoglycoside mimetics.⁵⁶ The synthesis of the first examples of oligosaccharide mimetics such as 19 suggest that a high degree of diversity can be achieved by conjugating only a small set of different SDAs through amide bond.

Carlomagno and coworkers reported a neooligoaminodeoxysaccharide 20 as a novel aminolgycoside analogue by connecting four aminosugar rings using linear alkyl linker.⁵⁷ Unlike other aminoglycosides, this compound was able to make simultaneous contacts with the bulge residues required for Tat binding and A35 residue of the hexanucleotide loop of HIV-2 TAR RNA.

Most aminoglycoside mimics have been prepared by adding extra functionality onto the original aminoglycoside structures without removing aminosugar substituents. A number of intercalating agents were included in the extra functionality with the anticipation that aromatic compounds may have an intercalating capability to the RNA targets that aminoglycosides typically bind. Hamasaki and coworkers conjugated acridine, anthracene and other similar aromatic structures to the ring I of neamine.⁵⁸ Pyreneconjugated neamine analogs 21 demonstrated up to 87 times stronger binding affinity to HIV RRE mRNA than neamine. Neomycin B was also conjugated with acridine at its 5“-position on the ring III via a short linker. Conjugate 22 was one of the strongest competitive inhibitors of HIV rev-RRE interaction, which came at the expense of losing binding selectivity to RRE.⁵⁹ Quinacridine conjugated with two aminoglycosides, used as a potential telomerase activity modulator targeting the P6.1 element of RNA telomerase, was another example of intercalator-conjugates.⁶⁰

Guadinylation of multiple amino groups of aminoglycosides shows similar effects as intercalating agents. Guanidylated neomycin B (23) was reported to inhibit the viral replication 100 times more potently than neomycin B by targeting HIV RRE.^61,62 A similar result was also observed with tobramycin.⁶⁰ It is notable that aminoglycosides generally exhibit poor uptake by eukaryotic cell lines but guadinylation of the aminoglycosides dramatically enhanced cellular uptake of aminoglycosides.⁶³ Lapidot and coworkers reported high inhibition activity against HIV-1 by conjugating arginine to introduce the guanidine functionality to neomycin B.⁶⁴

Multivalent aminoglycosides have been reported to target RNA structures such as HIV-1 TAR RNA due to their unique structure that allows for multiple binding interactions. Four symmetrical 4’,4‘- and 5,5-neamine dimers 24–25 were prepared and studied for their affinity toward TAR RNA using fluorimetric binding assays.⁶⁵ All these dimers inhibited TAR-Tat binding at submicromolar concentrations. Trimeric neamine was also prepared but not useful for activity studies as it resulted in precipitation of the TAR RNA. Other reported dimeric aminoglycosides⁶⁶ (neomycin-neomycin, neomycin-tobramycin, and neomycinkanamycin A) demonstrated much higher inhibitory activity than exhibited by doubling the concentration of each aminoglycoside due to the favorable entropic factors.

The lack of high RNA target selectivity of aminoglycosides is partially attributed to their conformational flexibility that allow adjusting to different shapes of RNAs. Tor and coworkers prepared conformationally-constrained neomycin and paromomycin derivative 26a and 26b by covalently connecting two aminosugars (ring I and ring III) to reduce the number of available conformations, thus changing target selectivity.³² However, the conformational constraint did not help in discriminating different RNA targets, which suggests that the structure of the target RNA and its flexibility also play an important role in the binding event. Mobashery and coworkers reported another kind of structurally constrained aminoglycoside 27 based on neamine.⁶⁷ However, no biological activity was reported for this compound.

Arya and coworkers reported an oligonucleotide-neomycin conjugate for sequence-specific targeting of RNA. The nucleotide-neomycin conjugate demonstrated enhanced duplex formation for its target RNA α-sarcin loop.⁶⁸

While most aminoglycoside mimetics employed single point modification on one of the rings of aminoglycosides, Houston and coworkers reported multisite modification on the structure of neomycin B.⁶⁹ Three rings (rings II, III, and IV) of the neomycin B were modified at the same time using Mitsunobu and Click chemistry.

Aminoglycoside Resistance

The increasing bacterial resistance to clinically important aminoglycosides catalyzed the search for novel aminoglycoside mimics. Among several different resistance mechanisms, bacterial inactivation of aminoglycosides by intracellular aminoglycoside-modifying enzymes is the most significant source of resistance development.⁷⁰ More than 50 different types of aminoglycoside-modifying enzymes have been found in resistant bacteria. These enzymes modify aminoglycosides through acetylation of amino groups (N-acetyltransferases, AAC), and adenylation (O-adenyltransferase, ANT) and phosphorylation (phosphotransferases, APH) of hydroxyl groups on aminoglycosides.

According to recent studies mainly focused on tobramycin, kanamycin A and B, amikacin, gentamicin, and geneticin, differences in ring III did not seem to alter the interaction of the drug with target rRNA, but subtle variations of ring I significantly influenced binding. Ring II is also affected by the enzymes. The 2‘-and 6‘-NH₂ groups of the ring I and 3-NH₂ of the ring II are prone targets for acetylation.¹⁶ These three amino groups make highly conserved interactions with A-site RNA. The two hydroxyl groups at 2‘- and 4‘-positions of ring I are targets for adenylation and phosphorylation. When these important functional groups are recognized by the enzymes and modified by one of the three reactions, their highly conserved interactions, both hydrogen bonding and electrostatic interactions, are prevented and ultimately cause aminoglycosides not to be able to bind to RNA targets.

One way to meet the challenge presented by aminoglycoside resistance is to discover new aminoglycoside analogs that are less prone to enzymatic modification. Following are some examples of the aminoglycoside mimics that are designed rationally to fight against aminoglycoside resistance or that are found to be effective against resistant bacterial strains (Fig. 6), many of which are already discussed above.

Figure 6.

The aminoglycoside mimics that are designed or found to be effective against resistant bacterial strains.

Various azepane mimics of the 2-DOS including 9 and 10 demonstrated moderate antibacterial activity against resistant strains as well as S. aureus.⁴³ This experiment indicates that the biological activity of the natural antibiotics can be maintained while the structural difference of the designed scaffold from the natural antibiotics help escape some resistance mechanisms.

The ring II mimic of neamine (16) with substitutions at 1-NH₂ and 6-OH revealed potential structural aspect that may be used to target resistant strains.⁵⁰ Its crystal structure suggests that the aminohydroxybutyryl (AHB) substituent on the 1-NH₂ position may cause steric clash with the aminoglycoside-binding loop of the aminoglycoside phosphotransferase APH(3’)-IIIa and prohibits proper complex formation.

Dimers of neamine and nebramine were reported to be active against several aminoglycoside resistant bacterial strains, especially for the treatment of P. aeruginosa infection.^71,72 Interestingly, the length of the linker of the dimers was found to be an important determinant of antibacterial activity of the bivalent aminoglycosides.⁷²

Asensio and coworkers also prepared conformationally-restricted derivatives of neomycin 26a and 26c and examined whether they would be susceptible to enzyme inactivation.⁷³ The modification provided an effective protection against aminoglycoside inactivation enzymes involved in aminoglycoside resistance, which represents a test case that a structure-based approach may be used to design ligands that maintain binding affinity to desired RNA targets but a poor substrate for enzymes that lead to inactivation.

Distinct from all of the above examples that are designed to avoid interactions with resistance enzymes, Mobashery and coworkers designed an aminoglycoside that binds to the resistance enzyme APH(3’) but not affected by its enzymatic activity.⁷⁴ The versatile aminoglycoside kanamycin A was made clinically obsolete by the widespread expression of resistance enzymes APH(3’)s in pathogens. These enzymes phosphorylate the 3‘-OH of amino-glycosides. In the design, the 3‘-OH was oxidized to ketone 28, which in turn goes through (1) hydration to give geminal-diol, (2) phosphorylation of the hydroxy group by APH(3’), then (3) self regeneration of the ketone by dephosphorylation. While the MIC of kanamycin A increased 500 to 1000 fold when treated with resistant E. coli, the self-regenerating analog 28 has shown only 4-fold increase of MIC toward resistant E. coli, which signifies that the self-regenerating strategy is a feasible approach in lowering MIC for resistant strains.

Medicinal Aspects: Mechanisms of Antibacterial Activity and Toxicity

In addition to being a useful model in understanding the RNA recognition process by small molecules, aminoglycosides have been used as versatile antibiotics for almost six decades due to their enormous therapeutic value. For instance, gentamicins and amikacin are used to treat meningitis, pneumonia, and sepsis. Paromomycin is used for amoebic dysentery⁷⁵ and neomycin is used for ulcers and dermatitis.⁷⁶

Aminoglycosides exert their antibacterial activities by two different mechanisms. The first is by inhibiting the translation of essential proteins for bacterial growth. The binding of aminoglycosides to bacterial 16S rRNA, as shown in Figure 2, stops the translocation of the peptidyl-tRNA from A-site to the P-site resulting in the misreading of the mRNA. Thereby, prevention of the production of the essential bacterial proteins leads to bacterial cell death.⁷⁷ However, this binding mechanism alone does not explain the significant bactericidal effect of aminoglycosides because other antibiotics like tetracycline that function in a similar manner are not as bactericidal. Secondly, aminoglycosides, being highly positively charged, interact with negatively charged outer membranes of Gram-negative bacterial cells through electrostatic interaction and disrupt membrane integrity by displacing Mg²⁺ and Ca²⁺ bridges that connect neighboring lipopolysaccharides.⁷⁸ This creates temporary openings in the membrane and results in leaking of intracellular contents and increased antibiotic uptake through the membrane. This transport across the cytoplasmic membrane is known to be oxygen-dependent, thus aminoglycosides are less effective in anaerobes.⁷⁹ As a result, aminoglycosides are active primarily for aerobic Gram-negative bacteria such Pseudomonas, Acinetobacter, and Enterobacter and some Gram-positive bacteria.

Despite the versatile antibacterial application against a broad-spectrum of bacteria, the high level of ototoxicity and nephrotoxicity of the aminoglycosides often limits their use in broader applications. While nephrotoxicity is known to originate from reversible accumulation of aminoglycosides in the renal cortex,⁸⁰ the ototoxicity has shown more complex mechanisms such as binding of aminoglycosides to mitochondrial 12S rRNA,^81,82 increase in nitric oxide synthase activity in the vestibular epithelium,⁸³ activation of N-methyl-D-aspartate (NMDA) receptors in cochlea,⁸⁴ and formation of free radicals from complex formation between aminoglycosides and iron⁸⁵ (Fig. 7). In cases of the last two examples, it was demonstrated that respective use of antagonist for the NMDA receptor⁸⁶ and free radical scavengers⁸⁷ or iron chelators⁸⁵ alleviated the toxicity.

Figure 7.

Major mechanisms of toxicity of aminoglycosides.

Several toxicity elements of aminoglycosides were identified in relation to structures of aminoglycosides. These were determined based on effect of the number of amino groups available in aminoglycosides and compared to the toxicity of clinically used aminoglycosides and their derivatives, Baasov and coworkers pointed out that decreases in the number of amino groups in a given aminoglycoside resulted in lower toxicity.⁸⁸ In addition to the number of amino groups, the basicity of a given amino group was an important factor in toxicity. Introducing an electron-withdrawing fluoro group in place of 5-OH of the 2-DOS ring of amikacin and other aminoglycosides effectively decreased toxicity as it decreased the basicity of the neighboring amino groups.⁸⁹ A similar result was observed with the 2‘-NH₂ group located on the ring I of neamine, gentamine, and other aminoglycoside analogs.⁹⁰ Removal of the 3‘-OH of kanamycin B, which is located next to the 2‘-NH₂, increased the toxicity dramatically, while removal of the relative distant 4‘-OH had a limited effect on toxicity. This is also understood by the basicity change of the neighboring amino group. Owada reported that N-acetylated amino-glycosides possess considerably lower toxicity than their free aminocompounds.⁹¹

Recently, a significant discovery was reported regarding the toxicity of gentamicin. Sandoval and coworkers isolated each of four congeners (C1, C1a, C2, and C2a) of gentamicin using HPLC from a native gentamicin sample and found that gentamicin C2 does not have nephrotoxicity but retains the native antimicrobial efficacy.⁹² It is notable that different substitution pattern on only one ring (ring I) can make the difference in the toxicity profile.

Conclusion

The unique and complex three-dimensional structures that RNA forms provide unmet opportunities to design novel therapeutics targeting various RNAs. However, it has been generally believed that the intrinsic flexibility of RNA structures makes the structure-based rational drug design approach, which resulted tremendous success in protein-targeting therapeutic development, less suitable unless the target RNA is locked into a rigid conformation, which is not the case for most mRNAs. Despite the unfavorable view, the examples described in this review demonstrate that the structure-based drug design approach has been instrumental to a great degree in understanding the fundamental principles that govern the interactions between various RNA targets and small organic molecules and to design aminoglycoside mimics with better target selectivity and/or binding affinity. Thus, structure-based drug design will continue to play a significant role in meeting the challenges to overcome bacterial resistance to natural aminoglycoside antibiotics and design novel RNA-binding ligand targeting RNAs other than bacterial sources. The highly conserved sequence-specific recognition pattern of rings I and II of neamine, a common component for many clinically important aminoglycosides, may be put to use through the rational drug design approach to help devise small molecules that recognize specific sequences or sites of RNA. High throughput screening approaches, combined with the rational design, may offer fresh insight into the design of novel ligands, especially in identifying target-specific ligands.

In addition to aminoglycosides, riboswitches may serve as another important model in understanding the principles that govern small molecule-RNA recognition. Riboswitches are embedded within 5‘-untranslated regions (5‘-UTR) of certain bacterial mRNAs and regulate gene expression in response to small cellular metabolites. Since their discovery in 2002,⁹³ more than a dozen classes of riboswitches have been discovered. Recent work suggests that certain antibacterial compounds discovered decades ago exert their antibacterial activity by targeting riboswitches,^94–96 which establishes the viability of this approach for novel antibiotics. The fact that riboswitches form structurally complex and highly selective receptors for a given metabolite should allow highly active compounds to be designed that target riboswitches without interfering with other cellular RNAs or proteins,⁹⁷ which may provide wealth of information in RNA targeting using small molecules.

Disclosure

The authors report no conflicts of interest.

Footnotes

Biography: Maruthi Chittapragada obtained his undergraduate degree from Osmania University, Hyderabad, India. Later he received his Master's degree in Organic Chemistry from University of Pune, Pune, India. He moved to Indian Institute of Chemical Technolgy, Hyderabad, India and worked for his PhD degree with Dr. J. S. Yadav, during which he worked on synthesis of Natural Products and development of Novel Methodologies and holds CSIR fellowship. After graduation, he joined the Chemistry and Biochemistry department at University of Denver with Prof. Andrei Kutateladze, where he focused on the synthesis of photolabile compounds. Currently he is a post doc at Department of Chemistry and Biochemistry, Brigham Young University, Provo with Prof. Young Wan Ham.

Biography: Sarah Roberts received her bachelor's degree from Idaho State University in chemistry. She is currently working on a master's degree at Brigham Young University under the direction of Dr. Young Wan Ham. Here, she is involved in synthesis of aminoglycoside mimics to target RNAs.

Biography: Young Wan Ham received his B.Sc. and M.S. degrees at Hanyang University, Seoul, Korea, and joined Prof. Jean Chmielewski group at Purdue University for PhD, where he worked on studies related to inhibition of HIV-1 protease dimerization. He continued his training at The Scripps Research Institute with Prof. Dale Boger and worked on DNA binding small molecules and proteins and related assay development. He then moved to Molecular Therapeutics, Inc. in Ann Arbor in 2004 as a senior research scientist. In 2006 he became an Assistant Professor at Brigham Young University and his research is focused on aminoglycoside mimetics, peptidomimetics, and macromolecular recognition.

References

Zaman

G.J.R.

, Michiels

P.J.A.

, van Boeckel

C.A.A.

Targeting RNA: new opportunities to address drugless targets.

Drug Discov. Today 2003; 8(7): 297–306.

Inouye

Antisense RNA: its functions and applications in gene regulation.

Gene. 1988; 72(1-2): 25–34.

Simons

R.W.

Naturally occurring antisense RNA control-a brief review.

Gene. 1988; 72(1-2): 25–34.

Mello Craig

C.C.D.

Jr . Revealing the world of RNA interference. Nature 2004; 431(7006): 338–42.

RFSP

Gilmore Ian

, Hollins Andrew

, Akhtar

Delivery strategies for siRNAmediated gene silencing.

Curr Drug Deliv. 2006; 3(2): 147–55.

Thomas

J.R.

, Liu

X.J.

, Hergenrother

P.J.

Biochemical and thermodynamic characterization of compounds that bind to RNA hairpin loops: Toward an understanding of selectivity.

Biochemistry. 2006; 45(36): 10928–38.

Busscher

G.F.

, Rutjes

, van Delft

F.L.

2-deoxystreptamine: Central scaffold of aminoglycoside antibiotics.

Chem Rev. 2005; 105(3): 775–91.

Ding

Y.L.

, Hofstadler

S.A.

, Swayze

E.E.

, Griffey

R.H.

An efficient synthesis of mimetics of neamine for RNA recognition.

Org Lett. 2001; 3(11): 1621–3.

Vicens

, Westhof

Crystal Structure of a Complex between the Aminoglycoside Tobramycin and an Oligonucleotide Containing the Ribosomal Decoding A Site.

Chem Biol. 2002; 9(6): 747–755.

10.

Vicens

, Westhof

RNA as a drug target: the case of aminoglycosides.

Chembiochem. 2003; 4(10): 1018–23.

11.

Francois

, Russell

R.J.M.

, Murray

J.B.

, Aboul-ela

, Masquida

, Vicens

, Westhof

Crystal structures of complexes between aminoglycosides and decoding A site oligonucleotides: role of the number of rings and positive charges in the specific binding leading to miscoding.

Nucleic Acids Res. 2005; 33(17): 5677–690.

12.

Yoshizawa

, Fourmy

, Eason

R.G.

, Puglisi

J.D.

Sequence-Specific Recognition of the Major Groove of RNA by Deoxystreptamine.

Biochemistry. 2002; 41(20): 6263–70.

13.

Zembower

T.R.

, Noskin

G.A.

, Postelnick

M.J.

, Nguyen

, Peterson

L.R.

The utility of aminoglycosides in an era of emerging drug resistance.

Int J Antimicrob Agents. 1998; 10(2): 95–105.

14.

Mingeot-Leclercq

M.P.

, Glupczynski

, Tulkens

P.M.

Aminoglycosides: activity and resistance.

Antimicrob Agents Chemother. 1999; 43(4): 727–37.

15.

Fletcher

H.G.

Jr , Anderson

, Lardy

H.A.

The nomenclature of the cyclohexitols and their derivatives.

J Org Chem. 1951; 16: 1238–46.

16.

Fourmy

, Recht

M.I.

, Blanchard

S.C.

, Puglisi

J.D.

Structure of the A site of Escherichia coli 16S ribosomal RNA complexed with an aminoglycoside antibiotic.

Science. 1996; 274(5291): 1367–71.

17.

Miyaguchi

, Narita

, Sakamoto

, Yokoyama

An antibiotic-binding motif of an RNA fragment derived from the A-site-related region of Escherichia coli 16S rRNA.

Nucleic Acids Res. 1996; 24(19): 3700–06.

18.

Recht

M.I.

, Fourmy

, Blanchard

S.C.

, Dahlquist

K.D.

, Puglisi

J.D.

RNA sequence determinants for aminoglycoside binding to an A-site rRNA model oligonucleotide.

J Mol Biol. 1996; 262(4): 421–36.

19.

Sannes-Lowery

K.A.

, Mei

H.Y.

, Loo

J.A.

Studying aminoglycoside antibiotic binding to HIV-1 TAR RNA by electrospray ionization mass spectrometry.

Int J Mass Spectrom. 1999; 193(2/3): 115–22.

20.

Wang

, Hamasaki

, Rando

R.R.

Specificity of Aminoglycoside Binding to RNA Constructs Derived from the 16S rRNA Decoding Region and the HIV-RRE Activator Region.

Biochemistry. 1997; 36(4): 768–79.

21.

Von Ahsen

, Davies

, Schroeder

Noncompetitive inhibition of group I intron RNA self-splicing by aminogylcoside antibiotics.

J Mol Biol. 1992; 226(4): 935–41.

22.

Mikkelsen

N.E.

, Brannvall

, Virtanen

, Kirsebom

L.A.

Inhibition of RNase P RNA cleavage by aminoglycosides.

PNAS. 1999; 96(11): 6155–60.

23.

Corvaisier

, Bordeau

, Felden

Inhibition of transfer messenger RNA aminoacylation and trans-translation by aminoglycoside antibiotics.

J Biol Chem. 2003; 278(17): 14788–97.

24.

Hyun Ryu

, Rando

R.R.

Aminoglycoside binding to human and bacterial A-Site rRNA decoding region constructs.

Bioorg Med Chem. 2001; 9(10): 2601–8.

25.

Jiang

, Patel

D.J.

Solution structure of the tobramycin-RNA aptamer complex.

Nature Structural Biology. 1998; 5(9): 769–74.

26.

Wang

, Tor

Electrostatic Interactions in RNA Aminoglycosides Binding.

J Am Chem Soc. 1997; 119(37): 8734–35.

27.

Hobbie

S.N.

, Bruell

, Kalapala

, Akshay

, Schmidt

, Pfister

, Bottger

E.C.

A genetic model to investigate drug-target interactions at the ribosomal decoding site.

Biochimie. 2006; 88(8): 1033–43.

28.

Wang

, Huber

P.W.

, Cui

, Czarnik

A.W.

, Mei

H Y.

Binding of Neomycin to the TAR Element of HIV-1 RNA Induces Dissociation of Tat Protein by an Allosteric Mechanism.

Biochemistry. 1998; 37(16): 5549–57.

29.

Kwon

Studies on the binding affinity of aminoglycoside antibiotics to the HIV-1 rev responsive element for designing potential antiviral agents.

J Microbiol Biotechnol. 2006; 16(1): 109–17.

30.

Waldsich

, Semrad

, Schroeder

Neomycin B inhibits splicing of the td intron indirectly by interfering with translation and enhances missplicing in vivo.

RNA. 1998; 4(12): 1653–63.

31.

Wallis

M.G.

, Vonahsen

, Schroeder

, Famulok

A Novel Rna Motif for Neomycin Recognition.

Chem Biol. 1995; 2(8): 543–52.

32.

Blount

K.F.

, Zhao

, Hermann

, Tor

Conformational Constraint as a Means for Understanding RNA-Aminoglycoside Specificity.

J Am Chem Soc. 2005; 127(27): 9818–829.

33.

Ryu Do

, Rando Robert

Decoding region bubble size and aminoglycoside antibiotic binding.

Bioorg Med Chem Lett. 2002; 12(16): 2241–4.

34.

Wang

, Migawa

M.T.

, Sannes-Lowery

K.A.

, Swayze

E.E.

The synthesis and 16S A-site rRNA recognition of carbohydrate-free amino-glycosides.

Bioorg Med Chem Lett. 2005; 15(22): 4919–22.

35.

Ding

, Hofstadler

S.A.

, Swayze

E.E.

, Griffey

R.H.

An Efficient Synthesis of Mimetics of Neamine for RNA Recognition.

Org Lett. 2001; 3(11): 1621–23.

36.

Vourloumis

, Takahashi

, Winters

G.C.

. Novel 2,5-dideoxystreptamine derivatives targeting the ribosomal decoding site RNA. Bioorg Med Chem Lett. 2002; 12(23): 3367–72.

37.

Zhou

, Sun

, Froelich

J.M.

, Hermann

, Wall

Structure-activity relationships of novel antibacterial translation inhibitors: 3,5-Diamino-piperidinyl triazines.

Bioorg Med Chem Lett. 2006; 16(20): 5451–6.

38.

Busscher

G.F.

, van den Broek

SAMW

, Rutjes

FPJT

, van Delft

F.L.

Carbohydrate mimic of 2-deoxystreptamine for the preparation of conformationally constrained amino-glycosides.

Tetrahedron. 2007; 63(15): 3183–8.

39.

Fourmy

, Recht

M.I.

, Puglisi

J.D.

Binding of neomycin-class aminoglycoside antibiotics to the A-site of 16 S rRNA.

J Mol Biol. 1998; 277(2): 347–62.

40.

Edwards

T.E.

, Sigurdsson

S.T.

Electron Paramagnetic Resonance Dynamic Signatures of TAR RNA-Small Molecule Complexes Provide Insight into RNA Structure and Recognition.

Biochemistry. 2002; 41(50): 14843–7.

41.

Liu

, Thomas

J.R.

, Hergenrother

P.J.

Deoxystreptamine Dimers Bind to RNA Hairpin Loops.

J Am Chem Soc. 2004; 126(30): 9196–97.

42.

Ryu Do

, Litovchick

, Rando Robert

Stereospecificity of aminoglycoside-ribosomal interactions.

Biochemistry. 2002; 41(33): 10499–509.

43.

Barluenga

, Simonsen

K.B.

, Littlefield

E.S.

. Rational design of azepane-glycoside antibiotics targeting the bacterial ribosome. Bioorg Med Chem Lett. 2004; 14(3): 713–18.

44.

Zhou

, Gregor

V.E.

, Ayida

B.K.

. Synthesis and SAR of 3,5-diamino-piperidine derivatives: novel antibacterial translation inhibitors as aminoglycoside mimetics. Bioorg Med Chem Lett. 2007; 17(5): 1206–10.

45.

Zhou

, Chow

, Murphy

D.E.

. Antibacterial activity in serum of the 3,5-diamino-piperidine translation inhibitors. Bioorg Med Chem Lett. 2008; 18(11): 3369–75.

46.

Vourloumis

, Winters

G.C.

, Takahashi

. Novel acyclic deoxystreptamine mimetics targeting the ribosomal decoding site. Chem Bio Chem. 2003; 4(9): 879–85.

47.

Sucheck

S.J.

, Greenberg

W.A.

, Tolbert

T.J.

, Wong

C-H

. Design of small molecules that recognize RNA: development of aminoglycosides as potential antitumor agents that target oncogenic RNA sequences. Angew Chem Int Ed. 2000; 39(6): 1080–84.

48.

Park

W.K.C.

, Auer

, Jaksche

, Wong

C-H

. Rapid Combinatorial Synthesis of Amino Glycoside Antibiotic Mimetics: Use of a Polyethylene Glycol-Linked Amine and a Neamine-Derived Aldehyde in Multiple Component Condensation as a Strategy for the Discovery of New Inhibitors of the HIV RNA Rev Responsive Element. J Am Chem Soc. 1996; 118(42): 10150–155.

49.

Russell

R.J.M.

, Murray

J.B.

, Lentzen

, Haddad

, Mobashery

The Complex of a Designer Antibiotic with a Model Aminoacyl Site of the 30S Ribosomal Subunit Revealed by X-ray Crystallography.

J Am Chem Soc. 2003; 125(12): 3410–11.

50.

Greenberg

W.A.

, Priestley

E.S.

, Sears

P.S.

. Design and Synthesis of New Amino Glycoside Antibiotics Containing Neamine as an Optimal Core Structure: Correlation of Antibiotic Activity with in Vitro Inhibition of Translation. J Am Chem Soc. 1999; 121(28): 6527–541.

51.

Liang

F-S

, Wang

S-K

, Nakatani

, Wong

C-H

. Targeting RNAs with tobramycin analogues. Angew Chem Int Ed. 2004; 43(47): 6496–6500.

52.

Wang

, Li

, Tuttle

, Takemoto

J.Y.

, Chang

C-WT

. The Synthesis of L-Aminosugar and the Studies of L-Pyranoses on the Ring III of Pyranmycins. Org Lett. 2002; 4(23): 3997–4000.

53.

Rosenbohm

, Vanden Berghe

, Vlietinck

, Wengel

Parallel synthesis of a small library of novel amino glycoside analogs based on 2-amino-2-deoxy-D-glucose and D-ribose scaffolds.

Tetrahedron. 2001; 57(29): 6277–87.

54.

Wong

C-H

, Hendrix

, Manning

D.D.

, Rosenbohm

, Greenberg

W.A.

A Library Approach to the Discovery of Small Molecules That Recognize RNA: Use of a 1,3-Hydroxyamine Motif as Core.

J Am Chem Soc. 1998; 120(33): 8319–27.

55.

Hendrix

, Alper

P.B.

, Priestley

E.S.

, Wong

C.H.

Hydroxyamines as a new motif for the molecular recognition of phosphodiesters: implications for aminoglycoside-RNA interactions.

Angew Chem Int Ed. 1997; 36(1/2): 95–98.

56.

Sicherl

, Wittmann

Orthogonally protected sugar diamino acids as building blocks for linear and branched oligosaccharide mimetics.

Angew Chem Int Ed. 2005; 44(14): 2096–2099.

57.

Raghunathan

, Sanchez-Pedregal

V.M.

, Junker

. TAR-RNA recognition by a novel cyclic aminoglycoside analogue. Nucleic Acids Res. 2006; 34(12): 3599–3608.

58.

Hamasaki

, Woo

M.C.

, Ueno

An aminoglycoside antibiotic, neamine, and its aromatic ring-substituted derivatives as potential inhibitors for HIV-1 RRE-Rev.

Tetrahedron Lett. 2000; 41(43): 8327–8332.

59.

Kirk

S.R.

, Luedtke

N.W.

, Tor

Neomycin-Acridine Conjugate: A Potent Inhibitor of Rev-RRE Binding.

J Am Chem Soc. 2000; 122(5); 980–981.

60.

Kaiser

, Sainlos

, Lehn

J-M

, Bombard

, Teulade-Fichou

M-P

. Aminoglycoside-quinacridine conjugates: towards recognition of the P6.1 element of telomerase RNA. Chem Bio Chem. 2006; 7(2): 321–329.

61.

Baker

T.J.

, Luedtke

N.W.

, Tor

, Goodman

Synthesis and anti-HIV activity of guanidinoglycosides.

J Org Chem. 2000; 65(26): 9054–8.

62.

Staple

D.W.

, Venditti

, Elson-Schwab

, Niccolai

, Tor

, Butcher

S.E.

Guanidinoneomycin B recognition of an HIV-1 RNA helix.

Chembiochem. 2008; 9: 93–102.

63.

Luedtke

N.W.

, Carmichael

, Tor

Cellular Uptake of Aminoglycosides, Guanidinoglycosides, and Poly-arginine.

J Am Chem Soc. 2003; 125(41): 12374–5.

64.

Lapidot

, Vijayabaskar

, Litovchick

, Yu

, James

T.L.

Structure-activity relationships of aminoglycoside-arginine conjugates that bind HIV-1 RNAs as determined by fluorescence and NMR spectroscopy.

FEBS Lett. 2004; 577(3): 415–21.

65.

Riguet

, Desire

, Boden

, Ludwig

, Goebel

, Bailly

, Decout

J-L

. Neamine dimers targeting the HIV-1 TAR RNA. Bioorg Med Chem Lett. 2005; 15(21): 4651–5.

66.

Tok

J.B.

, Dunn

L.J.

, Des Jean

R.C.

Binding of dimeric aminoglycosides to the HIV-1 rev responsive element (RRE) RNA construct.

Bioorg Med Chem Lett. 2001; 11(9): 1127–31.

67.

Kling

, Hesek

, Shi

, Mobashery

Design and Synthesis of a Structurally Constrained Aminoglycoside.

J Org Chem. 2007; 72(14): 5450–3.

68.

Charles

, Xi

, Arya

D.P.

Sequence-specific targeting of RNA with an oligonucetide-neomycin conjugate.

Bioconjug Chem. 2007; 18: 160–9.

69.

Quader

, Boyd

S.E.

, Jenkins

I.D.

, Houston

T.A.

Multisite Modification of Neomycin B: Combined Mitsunobu and Click Chemistry Approach.

J Org Chem. 2007; 72(6): 1962–1979.

70.

Smith Clyde

, Baker Edward

Aminoglycoside antibiotic resistance by enzymatic deactivation.

Curr Drug Targets Infect Disord. 2002; 2(2): 143–60.

71.

Agnelli

, Sucheck

S.J.

, Marby

K.A.

Dimeric aminoglycosides as antibiotics.

Angew Chem Int Ed. 2004; 43(12): 1562–66.

72.

Liang

C.H.

, Romero

, Rabuka

. Structure-activity relationships of bivalent aminoglycosides and evaluation of their microbiological activities. Bioorg Med Chem Lett. 2005; 15(8): 2123–28.

73.

Bastida

, Hidalgo

, Chiara

J.L.

. Exploring the Use of Conformationally Locked Amino-Glycosides as a New Strategy to Overcome Bacterial Resistance. J Am Chem Soc. 2006; 128(1): 100–116.

74.

Haddad

, Vakulenko

, Mobashery

An Antibiotic Cloaked by Its Own Resistance Enzyme.

J Am Chem Soc. 1999; 121(50): 11922–11923.

75.

Loebenberg

, Counelis

, Waitz

J.A.

Antibiotic G-418, a new Micromonospora-produced aminoglycoside with activity against protozoa and helminths. Antiparasitic activity.

Antimicrob Agents Chemother. 1975; 7(6): 811–15.

76.

Saap

, Fahim

, Arsenault

. Contact sensitivity in patients with leg ulcerations: a North American study. Arch Dermatol. 2004; 140(10): 1241–6.

77.

Magnet

, Blanchard

J.S.

Molecular Insights into Aminoglycoside Action and Resistance.

Chem Rev. 2005; 105(2): 477–497.

78.

Hancock

R.E.W.

, Farmer

S.W.

, Li

, Poole

Interaction of aminoglycosides with the outer membranes and purified lipopolysaccharide and OmpF porin of Escherichia coli.

Antimicrob Agents Chemother. 1991; 35(7): 1309–14.

79.

Bryan

L.E.

, Kwan

Roles of ribosomal binding, membrane potential, and electron transport in bacterial uptake of streptomycin and gentamicin.

Antimicrob Agents Chemother. 1983; 23(6): 835–45.

80.

Mingeot-Leclercq

M.P.

, Tulkens

P.M.

Aminoglycosides: nephrotoxicity.

Antimicrob Agents Chemother. 1999; 43(5): 1003–12.

81.

Chen

, Yang

. Maternally inherited aminoglycosideinduced and nonsyndromic hearing loss is associated with the 12S rRNA C1494T mutation in three Han Chinese pedigrees. Gene. 2007; 401(1-2): 4–11.

82.

Hobbie

S.N.

, Akshay

, Kalapala

S.K.

, Bruell

C.M.

, Shcherbakov

, Bottger

E.C.

Genetic analysis of interactions with eukaryotic rRNA identify the mitoribosome as target in aminoglycoside ototoxicity.

PNAS. 2008; 1–6.

83.

Hong

S.H.

, Park

S.K.

, Cho

Y.S.

. Gentamicin induced nitric oxide-related oxidative damages on vestibular afferents in the guinea pig. Hear Res. 2006; 211(1-2): 46–53.

84.

Harvey

S.C.

, Skolnick

Polyamine-like actions of aminoglycosides at recombinant N-methyl-D-aspartate receptors.

J Pharmacol Exp Ther. 1999; 291(1): 285–91.

85.

Song

B.B.

, Anderson

D.J.

, Schacht

Protection from gentamicin ototoxicity by iron chelators in guinea pig in vivo.

J Pharmacol Exp Ther. 1997; 282(1): 369–77.

86.

Basile

A.S.

, Brichta

A.M.

, Harris

B.D.

, Morse

, Coling

, Skolnick

Dizocilpine attenuates streptomycin-induced vestibulotoxicity in rats.

Neurosci Lett. 1999; 265(2): 71–74.

87.

Sinswat

, Wu

W.J.

, Sha

S.H.

, Schacht

Protection from ototoxicity of intraperitoneal gentamicin in guinea pig.

Kidney Int. 2000; 58(6): 2525–32.

88.

Hainrichson

, Nudelman

, Baasov

Designer aminoglycosides: the race to develop improved antibiotics and compounds for the treatment of human genetic diseases.

Org Biomol Chem. 2008; 6: 227–239.

89.

Shitara

, Umemura

, Tsuchiya

, Matsuno

Synthesis of 5-deoxy-5-epifluoro derivatives of arbekacin, amikacin, and 1-N-[(S)-4-amino-2-hydroxybutanoyl]tobramycin (study on structure-toxicity relationships).

Carbohydrate Res. 1995; 276: 75–89.

90.

Chen

, Hainrichson

, Bourdetsky

, Mor

, Yaron

, Baasov

Structure-toxicity relationship of aminoglycosides: Correlation of 2‘-amine basicity with acute toxicity in pseudo-disaccharide scaffolds.

Bioorg Med Chem. 2008; 16(19): 8940–951.

91.

Owada

Experimental studies on the toxicity of kanamycin, its hydrolyzed products and neomycin.

Chemotherapia. 1962; 5: 277–93.

92.

Sandoval

R.M.

, Reilly

J.P.

, Running

. A Non-Nephrotoxic Gentamicin Congener That Retains Antimicrobial Efficacy. J Am Soc Nephrol. 2006; 17: 2697–705.

93.

Winkler

, Nahvi

, Breaker Ronald

Thiamine derivatives bind messenger RNAs directly to regulate bacterial gene expression.

Nature. 2002; 419(6910): 952–6.

94.

Sudarsan

, Cohen-Chalamish

, Nakamura

, Emilsson Gail

, Breaker Ronald

Thiamine pyrophosphate riboswitches are targets for the antimicrobial compound pyrithiamine.

Chem Biol. 2005; 12(12): 1325–35.

95.

Sudarsan

, Wickiser

J.K.

, Nakamura

, Ebert Margaret

, Breaker Ronald

An mRNA structure in bacteria that controls gene expression by binding lysine.

Genes Dev. 2003; 17(21): 2688–97.

96.

Vogl

, Grill

, Schilling

, Stuelke

, Mack

, Stolz

Characterization of riboflavin (vitamin B2) transport proteins from Bacillus subtilis and Corynebacterium glutamicum.

J Bacteriol. 2007; 189(20): 7367–75.

97.

Blount

K.F.

, Wang

J.X.

, Lim

, Sudarsan

, Breaker

R.R.

Antibacterial lysine analogs that target lysine riboswitches.

Nat Chem Biol. 2006; 3(1): 44–9.