Characterization of aminoglycoside-lipid interactions and development of a refined model for ototoxicity testing

Improving the Safety of Clinical Management of COVID-19 Patients Receiving Aminoglycosides and Parenteral Nutrition: Y-site Compatibility Studies

PURPOSE

Aminoglycosides (AMGs) are broad-spectrum bactericidal antibiotics that can resolve bacterial infections co-existing with COVID-19 or exploit their potential antiviral activities. Patients presenting the most severe forms of COVID-19 due to escalating catabolism and significant lean body mass loss often require the concomitant administration of parenteral nutrition (PN) and antibiotics. The Y-site administration is one of the approaches allowing the co-administration of two intravenous medications in patients with limited vascular access. Our study aimed to investigate the compatibility of AMGs and selected commercial PN admixtures enriched in omega-3 fatty acids.

METHODS

Gentamycin (GM), amikacin (AM), and tobramycin (TM) solutions for infusion were combined with Nutriflex Omega Special (NOS) and Smofkabiven (SFK). Three different volume ratios were investigated: 1:2, 1:1, and 2:1, simulating Y-site administration. Samples underwent visual examination and determination of the lipid emulsion particle size, zeta potential, and pH immediately after preparation and after four hours of storage at room temperature (22 ± 2 °C) with sunlight exposure.

RESULTS

GM and AM combined with NOS in all studied ratios met the set-up acceptance criteria. The addition of TM to NOS in a 2:1 volume ratio and all tested AMGs to SFK in all studied combinations significantly influenced the stability of the oil-water system leading to the appearance of globules larger than 5 µm exceeding the pharmacopeial limit of 0.05% immediately after preparation or after four hours of storage.

CONCLUSION

In conclusion, our study showed that NOS was less prone to destabilization of oil-in-water systems by AMGs than SFK. In justified clinical cases, due to the lack of appearance of precipitate or enlarged lipid droplets, the combined administration of GM and AM with the NOS could be considered, provided tested volume ratios of the drug and MCB in the infusion line are maintained. However, it should be noted that such an infusion may be associated with the risk of changes in the pharmacokinetics of the drug.

Phosphatidylinositol 4,5-bisphosphate is an HCV NS5A ligand and mediates replication of the viral genome

BACKGROUND & AIMS

Phosphoinositides (PIs) bind and regulate localization of proteins via a variety of structural motifs. PI 4,5-bisphosphate (PI[4,5]P2) interacts with and modulates the function of several proteins involved in intracellular vesicular membrane trafficking. We investigated interactions between PI(4,5)P2 and hepatitis C virus (HCV) nonstructural protein 5A (NS5A) and effects on the viral life cycle.

METHODS

We used a combination of quartz crystal microbalance, circular dichroism, molecular genetics, and immunofluorescence to study specific binding of PI(4,5)P2 by the HCV NS5A protein. We evaluated the effects of PI(4,5)P2 on the function of NS5A by expressing wild-type or mutant forms of Bart79I or FL-J6/JFH-5'C19Rluc2AUbi21 RNA in Huh7 cells. We also studied the effects of strategies designed to inhibit PI(4,5)P2 on HCV replication in these cells.

RESULTS

The N-terminal amphipathic helix of NS5A bound specifically to PI(4,5)P2, inducing a conformational change that stabilized the interaction between NS5A and TBC1D20, which is required for HCV replication. A pair of positively charged residues within the amphipathic helix (the basic amino acid PI(4,5)P2 pincer domain) was required for PI(4,5)P2 binding and replication of the HCV-RNA genome. A similar motif was found to be conserved across all HCV isolates, as well as amphipathic helices of many pathogens and apolipoproteins.

CONCLUSIONS

PI(4,5)P2 binds to HCV NS5A to promote replication of the viral RNA genome in hepatocytes. Strategies to disrupt this interaction might be developed to inhibit replication of HCV and other viruses.

Disruption of intracellular calcium regulation is integral to aminoglycoside-induced hair cell death

Intracellular Ca(2+) is a key regulator of life or death decisions in cultured neurons and sensory cells. The role of Ca(2+) in these processes is less clear in vivo, as the location of these cells often impedes visualization of intracellular Ca(2+) dynamics. We generated transgenic zebrafish lines that express the genetically encoded Ca(2+) indicator GCaMP in mechanosensory hair cells of the lateral line. These lines allow us to monitor intracellular Ca(2+) dynamics in real time during aminoglycoside-induced hair cell death. After exposure of live larvae to aminoglycosides, dying hair cells undergo a transient increase in intracellular Ca(2+) that occurs shortly after mitochondrial membrane potential collapse. Inhibition of intracellular Ca(2+) elevation through either caged chelators or pharmacological inhibitors of Ca(2+) effectors mitigates toxic effects of aminoglycoside exposure. Conversely, artificial elevation of intracellular Ca(2+) by caged Ca(2+) release agents sensitizes hair cells to the toxic effects of aminoglycosides. These data suggest that alterations in intracellular Ca(2+) homeostasis play an essential role in aminoglycoside-induced hair cell death, and indicate several potential therapeutic targets to stem ototoxicity.

Roles of phosphoinositides and phosphoinositides kinases in hepatitis C virus RNA replication

Phosphoinositides (PIs) play an essential role in mediating key signaling pathways on biological membranes. Hepatitis C virus (HCV) replicates its RNA genome by establishing a viral replication complex (RC) on host cell membranes. Recently, an increasing body of literature reported that not only PIs themselves but also several PIs-specific kinases are required for efficient replication of HCV RNA genome. Especially, PI 4-kinases type III alpha, beta as well as their enzymatic products including phosphatidylinositol 4-phosphate (PI(4)P) and phosphatidylinositol 4,5-bisphosphate (PI(4,5)P(2)) are consistently identified to be host factors essential for HCV replication. In this article, the current state of our knowledge of PIs and PIs-specific kinases together with their roles in modulating HCV replication is reviewed. The effects of various PIs-specific kinases inhibitors on HCV replication are also highlighted, proposing them as promising candidate targets to which a new class of anti-HCV therapeutics can be envisaged.

Response of mechanosensory hair cells of the zebrafish lateral line to aminoglycosides reveals distinct cell death pathways

We report a series of experiments investigating the kinetics of hair cell loss in lateral line neuromasts of zebrafish larvae following exposure to aminoglycoside antibiotics. Comparisons of the rate of hair cell loss and the differential effects of acute versus chronic exposure to gentamicin and neomycin revealed markedly different results. Neomycin induced rapid and dramatic concentration-dependent hair cell loss that is essentially complete within 90 min, regardless of concentration or exposure time. Gentamicin-induced loss of half of the hair cells within 90 min and substantial additional loss, which was prolonged and cumulative over exposure times up to at least 24h. Small molecules and genetic mutations that inhibit neomycin-induced hair cell loss were ineffective against prolonged gentamicin exposure supporting the hypothesis that these two drugs are revealing at least two cellular pathways. The mechanosensory channel blocker amiloride blocked both neomycin and gentamicin-induced hair cell death acutely and chronically indicating that these aminoglycosides share a common entry route. Further tests with additional aminoglycosides revealed a spectrum of differential responses to acute and chronic exposure. The distinctions between the times of action of these aminoglycosides indicate that these drugs induce multiple cell death pathways.

Lipid vesicles as membrane models for toxicological assessment of xenobiotics

Traditionally animals and cell cultures have been used to assess the toxic potential of xenobiotics on cell membranes. In search for more reproducible, quantitative, cost- and time-effective assays, toxicologists have recently become interested in biomimetic lipid vesicle-based test systems. Lipid vesicles (liposomes) have long been appreciated as simple cell membrane models in biochemical and biophysical studies providing a good understanding of the physicochemical properties of liposome systems. More recently a number of reports have been published on the interactions of toxic substances with vesicles. Literature reports on liposome assays have appeared for widely different classes of xenobiotics, such as dental materials, antibiotics, detergents, and peptides. In this review we focus on those reports that contain a quantitative and significant correlation with more established toxicological tests like cell culture assays. We provide an introduction to the structure and main characteristics of vesicles and related lipid aggregates. The two main assays presented are leakage of fluorescence dyes and differential scanning calorimetry (DSC) measurements of the solid-ordered/liquid-disordered main phase transition temperature (Tm).

Understanding drug ototoxicity: molecular insights for prevention and clinical management

Ototoxicity is a trait shared by aminoglycoside and macrolide antibiotics, loop diuretics, platinum-based chemotherapeutic agents, some NSAIDs and antimalarial medications. Because their benefits in combating certain life-threatening diseases often outweigh the risks, the use of these ototoxic drugs cannot simply be avoided. In this review, the authors discuss some of the most frequently used ototoxic drugs and what is currently known about the cell and molecular mechanisms underlying their noxious effects. The authors also provide suggestions for the clinical management of ototoxic medications, including ototoxic detection and drug monitoring. Understanding the mechanisms of drug ototoxicity may lead to new strategies for preventing and curing drug-induced hearing loss, as well as developing new pharmacological drugs with less toxic side effects.

A novel phosphatidylinositol 4,5-bisphosphate-binding domain targeting the Phg2 kinase to the membrane in Dictyostelium cells

Phg2 is a ser/thr kinase involved in adhesion, motility, actin cytoskeleton dynamics, and phagocytosis in Dictyostelium cells. In a search for Phg2 domains required for its localization to the plasma membrane, we identified a new domain interacting with phosphatidylinositol 4,5-bisphosphate (PI(4,5)P(2)) and phosphatidylinositol 4-phosphate (PI(4)P) membrane phosphoinositides. Deletion of this domain prevented membrane recruitment of Phg2 and proper function of the protein in the phagocytic process. Moreover, the overexpression of this PI(4,5)P(2)-binding domain specifically had a dominant-negative effect by inhibiting phagocytosis. Therefore, plasma membrane recruitment of Phg2 is essential for its function. The PI(4,5)P(2)-binding domain fused to GFP (green fluorescent protein) (GFP-Nt-Phg2) was also used to monitor the dynamics of PI(4,5)P(2) during macropinocytosis and phagocytosis. GFP-Nt-Phg2 disappeared from macropinosomes immediately after their closure. During phagocytosis, PI(4,5)P(2) disappeared even before the sealing of phagosomes as it was already observed in mammalian cells. Together these results demonstrate that PI(4,5)P(2) metabolism regulates the dynamics and the function of Phg2.

Hair cells require phosphatidylinositol 4,5-bisphosphate for mechanical transduction and adaptation

After opening in response to mechanical stimuli, hair cell transduction channels adapt with fast and slow mechanisms that each depend on Ca(2+). We demonstrate here that transduction and adaptation require phosphatidylinositol 4,5-bisphosphate (PIP(2)) for normal kinetics. PIP(2) has a striking distribution in hair cells, being excluded from the basal region of hair bundles and apical surfaces of frog saccular hair cells. Localization of a phosphatidylinositol lipid phosphatase, Ptprq, to these PIP(2)-free domains suggests that Ptprq maintains low PIP(2) levels there. Depletion of PIP(2) by inhibition of phosphatidylinositol 4-kinase or sequestration by aminoglycosides reduces the rates of fast and slow adaptation. PIP(2) and other anionic phospholipids bind directly to the IQ domains of myosin-1c, the motor that mediates slow adaptation, permitting a strong interaction with membranes and likely regulating the motor's activity. PIP(2) depletion also causes a loss in transduction current. PIP(2) therefore plays an essential role in hair cell adaptation and transduction.

Membrane targeting of protein tyrosine phosphatase PTPL1 through its FERM domain via binding to phosphatidylinositol 4,5-biphosphate

PTPL1 is the largest known cytoplasmic protein tyrosine phosphatase (PTP) containing a FERM (four point-1, ezrin, radixin and moesin) domain. Enzyme localization and PTP-substrate specificity are thought to play crucial roles in the regulation of PTP activity, which determines their functions. Here we report that PTPL1 is predominantly localized at the apical face of plasma membrane enriched in dorsal microvilli when expressed in HeLa cells. By comparing localization of the full-length enzyme with its FERM domain or FERM-deleted PTPL1 construct, we first concluded that PTPL1-FERM domain is necessary and sufficient to address the wild-type enzyme at the membrane. Two potential phosphatidylinositol 4,5-biphosphate [PtdIns(4,5)P2]-binding motifs were identified within the PTPL1-FERM sequence. We further showed that mutation of both sites altered PTPL1 localization similarly to FERM domain deletion, and impaired its subcellular distribution as confirmed biochemically by cell-fractionation experiments. Using protein-lipid overlays, we demonstrated an interaction of the FERM domain of PTPL1 with PtdIns(4,5)P2, which was lost after mutation of potential PtdIns(4,5)P2-binding motifs. Moreover, neomycin, which masks PtdIns(4,5)P2 polar heads, was shown to decrease by 50% the association of PTPL1 with the cytoskeletal fraction. These results identify the crucial role of the FERM domain in PTPL1 intracellular targeting and demonstrate that localization of PTPL1 is regulated by phosphoinositide metabolism.

Electrostatic interaction between stereocilia: II. Influence on the mechanical properties of the hair bundle

This paper is based on our model [Dolgobrodov et al., 2000. Hear. Res., submitted for publication] in which we examine the significance of the polyanionic surface layers of stereocilia for electrostatic interaction between them. We analyse how electrostatic forces modify the mechanical properties of the sensory hair bundle. Different charge distribution profiles within the glycocalyx are considered. When modelling a typical experiment on bundle stiffness measurements, applying an external force to the tallest row of stereocilia shows that the asymptotic stiffness of the hair bundle for negative displacements is always larger than the asymptotic stiffness for positive displacements. This increase in stiffness is monotonic for even charge distribution and shows local minima when the negative charge is concentrated in a thinner layer within the cell coat. The minima can also originate from the co-operative effect of electrostatic repulsion and inter-ciliary links with non-linear mechanical properties. Existing experimental observations are compared with the predictions of the model. We conclude that the forces of electrostatic interaction between stereocilia may influence the mechanical properties of the hair bundle and, being strongly non-linear, contribute to the non-linear phenomena, which have been recorded from the auditory periphery.

Aminoglycoside antibiotics

In the 50 years since their discovery, the aminoglycoside antibiotics have seen unprecedented use. Discovered in the 1940s, they were the long-sought remedy for tuberculosis and other serious bacterial infections. The side effects of renal and auditory toxicity, however, led to a decline of their use in most countries in the 1970s and 1980s. Nevertheless, today the aminoglycosides are still the most commonly used antibiotics worldwide thanks to the combination of their high efficacy with low cost. This review first summarizes the history, chemistry, antibacterial actions and acute side effects of the drugs. It then details the pathophysiology of aminoglycoside ototoxicity including experimental and clinical observations, risk factors and incidence. Pharmacokinetics, cellular actions and our current understanding of the underlying molecular mechanisms of ototoxicity are discussed at length. The review concludes with recent advances towards therapeutic intervention to prevent aminoglycoside ototoxicity.

Effects of growth factors on the hair cells after ototoxic treatment of the neonatal mammalian cochlea in vitro

The aim of this study was to test the possible regenerative potential of several molecules and growth factors such as retinoic acid (RA), insulin, epidermal growth factor (EGF) and transforming growth factors alpha (TGFalpha) and beta (TGFbeta) on the neonatal cochlea in vitro after neomycin intoxication. Our studies show that cochlear sensory epithelium behaves differently while maintained in various culture conditions, although we did not observe regeneration whatever the molecules or growth factors tested. The ototoxic action of neomycin in vitro produced a specific death of hair cells, except in the apical region. Organ of Corti of rats 3 days after birth always presented two regions that responded differently to the antibiotic: a widespread scar region extending from the basal cochlea up to the beginning of the apical turn, where most hair cells had disappeared, and a second region called the resistance region localized in the apex, and which was more or less developed depending on culture conditions. The length of the resistance region was modulated by molecules or growth factors added to the feeding solution suggesting that some of them could produce a protective action on hair cells against neomycin. Slight protection effects may be found with RA and insulin, however, the most definite protection results from the combination of insulin with TGFalpha as shown by the large increase in the length of the resistance region compared to organ of Corti treated with antibiotic alone. The tested molecules and growth factors did not promote cochlear hair cell regeneration in vitro after neomycin treatment, however some of them may offer a protective action against ototoxicity.

Formation of reactive oxygen species following bioactivation of gentamicin

The present study investigated the ability of gentamicin to catalyze free radical reactions and probed the underlying mechanisms by hydroethidine imaging, oxygen consumption, and reduction of cytochrome c. In Epstein-Barr virus-transformed lymphoblastoid cells, a respiratory burst was induced by phorbol ester and detected by hydroethidine, a fluorescent indicator of superoxide radical. The addition of gentamicin increased the fluorescence two-fold while gentamicin did not produce fluorescence in the absence of phorbol ester. In membrane preparations, gentamicin did not enhance NADPH consumption ruling out a direct activation of NADPH oxidase. The formation of reactive oxygen species by gentamicin was additionally supported by experiments that showed gentamicin increased oxygen consumption two-fold in intact cells and a cell-free system. In addition, generation of superoxide was indicated by the gentamicin-stimulated reduction of cytochrome c. The stimulation by gentamicin depended upon the presence of iron (FeII/FeIII) and of arachidonic acid as an electron donor. These results support the hypothesis that an iron-gentamicin complex can increase reactive oxygen species in nonenzymatic and in biological systems. The requirement for a reductive activation in intact cells (e.g., by a respiratory burst) is interpreted as the conversion of an inactive FeIII-gentamicin to a redox-active FeII-gentamicin complex.

Stimulation of free radical formation by aminoglycoside antibiotics

We have previously shown gentamicin to form a redox-active iron chelate. This study investigates whether other aminoglycosides can likewise stimulate the generation of reactive oxygen species (free radicals). Kanamycin, neomycin and streptomycin were compared to gentamicin in intact cells and in cell-free in vitro assays using luminescence detection with lucigenin or luminol. Neutrophils and Epstein-Barr virus-transformed lymphoblastoid cells served as cell models in which a respiratory burst of superoxide was induced by phorbol ester. The addition of millimolar amounts of any of the aminoglycosides increased the luminescence significantly. The drugs also increased the formation of free radicals in an enzymatic (hypoxanthine-xanthine oxidase) and a non-enzymatic (phenazine methosulfate-NADH) superoxide-generating system. Half-maximal stimulation was reached with (0.4 mM gentamicin, and there was an absolute requirement for an electron donor, arachidonic acid. In both intact cells and cell-free systems, gentamicin-enhanced luminosity was suppressed by iron chelators. These results demonstrate that different aminoglycoside antibiotics can stimulate the formation of free radicals in biological and in cell-free systems. Luminescence detection is a convenient assay method to investigate the redox properties of these drugs.

Role of rho proteins in agonist regulation of phospholipase D in HL-60 cells

Rho family GTP-binding proteins have been demonstrated to play a role in the regulation of phospholipase D (PLD) activity. In the present study, we examined the role of Rho proteins in PLD activation in differentiated HL-60 cells using C3 exoenzyme from Clostridium botulinum, which ADP-ribosylates and inactivates Rho proteins. Introduction of C3 exoenzyme into differentiated HL-60 cells by electroporation resulted in complete inhibition of PLD activity stimulated by formyl methionine-leucine-phenylalanine (fMLP) and ATP, two receptor agonists. Phorbol myristate acetate-induced PLD activation was also inhibited in C3 exoenzyme-treated cells, but the inhibition was only partial. GTPgammaS-dependent activation of PLD, measured in the absence or presence of ATP in permeabilized cells, was also partially affected by C3 exoenzyme treatment. Thus, these results indicate that Rho proteins play a key role in receptor-mediated PLD regulation in differentiated HL-60 cells, but play a partial role in the in vivo action of PMA and in vitro action of GTPgammaS on PLD. ATP produced a significant enhancement of the in vitro effect of GTPgammaS on PLD activity, but the effect of ATP was not altered by inhibitors of serine/threonine and tyrosine kinases. However, it was markedly reduced by neomycin and accompanied by an increase in phosphatidylinositol 4,5-bisphosphate (PtdInsP2) synthesis. These data indicate that in permeabilized HL-60 cells, the stimulatory effect of ATP on PLD does not involve protein phosphorylation but is due to an increase in PtdInsP2.

Regeneration of cochlear efferent nerve terminals after gentamycin damage

Chickens recover auditory function after hair cell loss caused by ototoxic drug damage or acoustic overstimulation, indicating that mechanisms exist to reestablish appropriate neuronal connections to regenerated hair cells. However, despite similar hair cell regeneration times, hearing recovery takes substantially longer after aminoglycoside than after sound damage. We have therefore begun examining damage and regeneration of efferent nerve terminals by immunolabeling whole-mount cochleae for differentially localized synaptic proteins and by visualizing the distribution of label with confocal microscopy. In undamaged cochleae, the synaptic proteins synapsin and syntaxin show similar distribution patterns corresponding to the large cup-like terminals on short hair cells. After gentamycin administration, these terminals are disrupted as hair cells are lost, leaving smaller, more numerous synapsin-reactive structures in the sensory epithelium. Syntaxin reactivity remains associated with the extruded hair cells, indicating that the presynaptic membrane is still attached to the postsynaptic site. In contrast, after sound damage, both synapsin and syntaxin reactivity are lost from the epithelium with extruded hair cells. As regenerated hair cells differentiate after gentamycin treatment, the synapsin labeling associated with cup-like efferent endings reappears but is not completely restored even after 60 d of recovery. Thus, efferent terminals are reestablished much more slowly than after sound damage (), consistent with the prolonged loss of hearing function. This in vivo model system allows comparison of axonal reconnection after either complete loss (sound damage) or partial disruption (gentamycin treatment) of axon terminals. Elucidating the differences in recovery between these injuries can provide insights into reinnervation mechanisms.

Cooperativity of phosphatidylinositol transfer protein and phospholipase D in secretory vesicle formation from the TGN--phosphoinositides as a common denominator?

Phosphatidylinositol transfer protein (PITP) and phospholipase D (PLD) stimulate the formation of constitutive secretory vesicles (CSVs) and immature secretory granules (ISGs) from the trans-Golgi network (TGN) in a cell-free system. The stimulatory effects of PITP and PLD are additive. Stimulation by either PITP or PLD is blocked by geneticin, a member of the aminoglycoside antibiotics known to bind to phosphoinositides. Since the PLD we used is insensitive to geneticin, our results suggest that phosphoinositides promote secretory vesicle formation as downstream effectors of both PITP and PLD, possibly via the recruitment of proteins mediating membrane budding and fission.

Transforming growth factor alpha treatment alters intracellular calcium levels in hair cells and protects them from ototoxic damage in vitro

To determine if transforming growth factor alpha (TGF alpha) pretreatment protects hair cells from aminoglycoside induced injury by modifying their intracellular calcium concentration, we assayed hair cell calcium levels in organ of Corti explants both before and after aminoglycoside (i.e. neomycin, 10(-3) M) exposure either with or without growth factor pretreatment. After TGF alpha (500 ng/ml) treatment, the intracellular calcium level of hair cells showed a five-fold increase as compared to the levels observed in the hair cells of control cultures. After ototoxin exposure, calcium levels in hair cells of control explants showed an increase relative to their baseline levels, while in the presence of growth factors pretreatment, hair cells showed a relative reduction in calcium levels. Pretreatment of organ of Corti explants afforded significant protection of hair cell stereocilia bundle morphology from ototoxic damage when compared to explants exposed to ototoxin alone. This study correlates a rise in hair cell calcium levels with the otoprotection of hair cells by TGF alpha in organ of Corti explants.

Formation of free radicals by gentamicin and iron and evidence for an iron/gentamicin complex

Participation of free radicals in the adverse renal and cochlear side effects of aminoglycoside antibiotics is controversial. We measured the production of free radicals by gentamicin in vitro through the oxidation of arachidonic acid. Gentamicin alone (0.05 to 10 mM) did not cause lipid peroxidation. However, it dramatically promoted radical formation in the presence of iron salts. Peroxidation was maximal at 1 mM gentamicin plus 0.1 mM Fe(II)/Fe(III) (0.05 mM FeSO4 and FeCl3 each). At these iron concentrations, peroxidation was not significant in the absence of gentamicin. Since chelators can enhance iron-catalyzed oxidations, this finding suggested that gentamicin-dependent radical formation was based upon iron chelation. This hypothesis was tested by measuring the influence of gentamicin on the oxidation of salicylate by Fe-EDTA complexes, a reaction that is inhibited by competing iron chelators. Gentamicin was a concentration-dependent inhibitor. In contrast, concentrations of gentamicin as high as 50 mM did not interfere with iron-independent salicylate oxidation. These results suggest that gentamicin acts as an iron chelator, and that the iron-gentamicin complex is a potent catalyst of free radical formation.

Phosphatidylinositol 4,5-bisphosphate synthesis is required for activation of phospholipase D in U937 cells

Phospholipase D (PLD) has been implicated in signal transduction and membrane traffic. We have previously shown that phosphatidylinositol 4,5-bisphosphate (PtdIns-4,5-P2) stimulates in vitro partially purified brain membrane PLD activity, defining a novel function of PtdIns-4,5-P2 as a PLD cofactor. In the present study we extend these observations to permeabilized U937 cells. In these cells, the activation of PLD by guanosine 5'-3-O-(thio)triphosphate (GTP gamma S) is greatly potentiated by MgATP. We have utilized this experimental system to test the hypothesis that MgATP potentiates PLD activation by G proteins because it is required for PtdIns-4,5-P2 synthesis by phosphoinositide kinases. As expected, MgATP was absolutely required for maintaining elevated phosphatidylinositol 4-phosphate (PtdIns-4-P) and PtdIns-4,5-P2 levels in the permeabilized cells. In the presence of MgATP, GTP gamma S further elevated the levels of the phosphoinositides. The importance of PtdIns-4,5-P2 for PLD activation was examined by utilizing a specific inhibitory antibody directed against phosphatidylinositol 4-kinase (PtdIns 4-kinase), the enzyme responsible for the first step in the synthesis of PtdIns-4,5-P2. Anti-PtdIns 4-kinase completely inhibited PtdIns 4-kinase activity in vitro and reduced by 75-80% PtdIns-4-P and PtdIns-4,5-P2 levels in the permeabilized cells. In parallel, the anti-PtdIns 4-kinase fully inhibited the activation of PLD by GTP gamma S and caused a 60% inhibition of PLD activation by the phorbol ester 12-O-tetradecanoylphorbol-13-acetate, indicating that elevated PtdIns-4,5-P2 levels are required for PLD activation. This conclusion is supported by the fact that neomycin, a high affinity ligand of PtdIns-4,5-P2, also blocked PLD activation. Furthermore, the activity of PLD in U937 cell lysate was stimulated by PtdIns-4,5-P2 in a dose-dependent manner. The current results indicate that PtdIns-4,5-P2 synthesis is required for PLD activation in permeabilized U937 cells and strongly support the proposed function of PtdIns-4,5-P2 as a cofactor for PLD. In addition, the results further establish PtdIns-4,5-P2 as a key component in the generation of second messengers via multiple pathways including phosphoinositide-phospholipase C, phosphoinositide 3-kinase and PLD.

Molecular parameters involved in aminoglycoside nephrotoxicity

Aminoglycoside antibiotics are hydrophilic molecules consisting of an animated cyclitol associated with amino sugar. They bind in vivo as well as in vitro to negatively charged membranes. Their use as chemotherapeutic agents is unfortunately accompanied by oto- and nephrotoxic reactions, and the purpose of this review is to examine the role of the molecular interactions between aminoglycosides and membranes in the development of nephrotoxicity. 31P Nuclear magnetic resonance (NMR) and fluorescence depolarization have been used to characterize the effect of aminoglycosides on phosphate heads and fatty acyl chains of phospholipids. 15N NMR has been used to obtain interesting information on regioselective interactions of amino groups of antibiotics with phospholipids. The binding of aminoglycosides with negatively charged membranes is associated with impairment of phospholipid catabolism, change in membrane permeability, and membrane aggregation. Biochemical analysis and 1H NMR spectroscopy have brought information on the molecular mechanism involved in the impairment of phospholipid catabolism. Nephrotoxic aminoglycosides could induce sequestration of phosphatidylinositol and therefore reduce the amount of negative charge available for optimal lysosomal phospholipase activity toward phosphatidylcholine included in liposomes that also contain cholesterol and sphingomyelin. Conformational analysis shows that aminoglycosides, which have a high potency to inhibit lysosomal phospholipase activity, adopt an orientation parallel to the lipid/water interface. This orientation of the aminoglycoside molecule at the interface is also critical to explain the marked increase of membrane permeability induced by less nephrotoxic aminoglycosides such as isepamicin and amikacin. This effect is indeed only observed with aminoglycosides oriented perpendicular to this interface, probably related to the creation of a local condition of disorder. The impairment of phospholipid catabolism, which is considered to be an early and significant step in the development of aminoglycoside toxicity, is therefore not related to the change in membrane permeability. However, the role of this latter phenomenon and of membrane aggregation for aminoglycoside nephrotoxicity could be further investigated.

Actions of (1S,3R)-1-aminocyclopentane-1,3-dicarboxylic acid (1S,3R-ACPD) in retinal ON bipolar cells indicate that it is an agonist at L-AP4 receptors

Metabotropic glutamate receptors (mGluRs) include receptors sensitive to L-2-amino-4-phosphonobutyrate (L-AP4) and 1S,3R-1-aminocyclopentane-1,3-dicarboxylic acid (1S,3R-ACPD). To determine whether 1S,3R-ACPD is an agonist at retinal L-AP4 receptors, whole cell voltage clamp recordings were obtained from mudpuppy ON bipolar cells in a superfused retinal slice and L-AP4 and 1S,3R-ACPD were bath applied. Both compounds evoked similar outward currents which reversed near 0 mV and were accompanied by an increased input resistance. Responses to both agonists washed out in parallel suggesting they act through the same second messenger pathway(s). Inhibitors of cGMP-PDE activity suppressed responses to both L-AP4 and 1SR,3RS-ACPD, suggesting that both compounds activate cGMP-PDE. Responses to 1S,3R-ACPD were occluded by prior activation of L-AP4 receptors, but not blocked by the non-AP4, mGluR antagonists, L-aminophosphonopropionic acid (L-AP3) or 4-carboxy-3-hydroxyphenylglycine (4C3H-PG). These results indicate that 1S,3R-ACPD is an agonist at L-AP4 receptors. 1S,3S-ACPD and 4C3H-PG evoked outward currents similar to L-AP4 suggesting they may also be L-AP4 receptor agonists. Using the b-wave of the ERG as an assay for ON bipolar cell responses, concentration/response curves were obtained for ACPD enantiomers. The rank-order potency of ACPD enantiomers at L-AP4 receptors in ON bipolar cells is similar to their rank-order potency at non-AP4, mGluRs in brain which suggests that the receptors possess similar binding sites and may be members of a common receptor family.

Ototoxicity in vitro: effects of neomycin, gentamicin, dihydrostreptomycin, amikacin, spectinomycin, neamine, spermine and poly-L-lysine

The effects that the aminoglycoside-aminocyclitol antibiotics amikacin, dihydrostreptomycin, gentamicin, neomycin, and spectinomycin, the neomycin fragment neamine, and the polybasic compounds spermine and poly-L-lysine, have on outer hair cells in cochlear cultures prepared from the early post-natal mouse have been assessed using both scanning and transmission electron microscopy (SEM and TEM). The antibiotics were used at concentrations ranging from 0.25-1.0 mM, spermine from 10 microM to 3.0 mM, and poly-L-lysine from 0.05-2 microM. Qualitative assessment of apical surface damage allows the antibiotics to be ranked in the following order: neomycin > gentamicin > dihydrostreptomycin > amikacin > neamine > spectinomycin. At a concentration of 1 mM spectinomycin is essentially non-toxic and the effects of neamine are marginal. Poly-L-lysine and spermine also cause surface damage, with poly-L-lysine being substantially more toxic than any of the antibiotics, and spermine ranking, on the basis of SEM observations, between dihydrostreptomycin and amikacin. TEM indicates that although all toxic compounds cause damage to the apical surface of the hair cell, only neomycin, poly-L-lysine and spermine induce the formation of whorls of tightly packed membrane resembling myelin within the apical surface lesions to any great extent. Apical-surface changes induced by dihydrostreptomycin and amikacin are simply large distensions of the cell filled with cytoplasmic organelles of normal appearance. Although the effects of the aminoglycoside antibiotics are largely limited to the apical surface of the cell, poly-L-lysine induces complete necrosis of the cell, and spermine causes a dramatic increase in cytoplasmic electron density and condensation of the nuclear chromatin.

Freeze fracture analysis of apical membranes in cochlear cultures: differences between basal and apical-coil outer hair cells and effects of neomycin

Previous studies have shown that exposure of cochlear cultures to the ototoxic aminoglycoside antibiotic neomycin leads to the rapid formation of numerous membrane filled protrusions on the apical surface of the hair cells but not on the surrounding supporting cells, and that hair cells in basal-coil cultures are more sensitive to these effects of neomycin than those in the distal end of apical-coil cultures. Freeze-fracture has been used to examine and compare the apical surfaces of hair cells and supporting cells in basal and apical-coil cultures in order to look for features that may explain the differential sensitivity of the various cell types to neomycin, and to characterize the membrane type that forms in response to neomycin and compare it with the normal apical membrane of the hair cell. The apical surface of the highly responsive basal-coil outer hair cells differs significantly from the apical surfaces of apical-coil outer hair cells and supporting cells in both regions of the cochlea in both surface area and the number and density of endocytotic vesicles associated with this surface. Basal-coil hair cells have an average of 120 +/- 39 vesicles per cell surface and a density of 3.5 +/- 0.89 vesicles per microns 2, whereas apical-coil hair cells have 14.8 +/- 15.8 vesicles/cell surface and density of 0.73 +/- 0.72 vesicles per microns 2. There are no significant differences in intramembrane particle (IMP) density on the apical surfaces of all the cell types examined, and qualitative observations of filipin-treated specimens indicate that cholesterol densities are also similar. The membrane that accumulates in response to neomycin treatment at the apical pole of the hair cell is IMP free, does not respond to filipin, and fractures in a manner that is indicative of a high content of unsaturated phospholipid in a fluid phase, and is therefore different in several respects from the normal apical surface of the hair cell. The results of this study suggest that apical surface associated endocytotic vesicle numbers may determine the differential sensitivity of apical and basal-coil hair cells to neomycin, and that neomycin may interfere with some aspect of phospholipid metabolism or membrane turnover in sensory hair cells.

Inhibition of the carbachol-evoked synthesis of inositol phosphates by ototoxic drugs in the rat cochlea

The ability of amikacin, neomycin, ethacrynate, mercuric chloride and cisplatin to alter the inositol phosphate (IP) signalling pathway was assessed in the 12-day-old rat cochlea, where the turnover of IPs is coupled to muscarinic receptors. This study was motivated by: (1) the demonstration of neomycin binding to phosphatidylinositol 4,5-biphosphate, the precursor of IPs, and (2) the fact that ototoxic drugs induce some common symptoms in outer hair cells. At concentrations below 1 mM, none of the compounds changed the control 3H-IP formation. Mercuric chloride, cisplatin and ethacrynate inhibited the carbachol-induced formation of IPs in a dose-dependent manner with IC50 values of 74,340 and 430 microM, respectively. The aminoglycosides were less efficient in reducing the carbachol-stimulated accumulation of IPs, since neither amikacin nor neomycin, both at 1 mM, had any significant effect. However, neomycin applied at 15 and 30 microM induced 29% and 43% of inhibition of the stimulated IP response. Finally, additive effects are obtained between some of the toxic drugs. The results suggest that a block of the IP transduction system, associated with the cholinergic efferent innervation of the organ of Corti, is a feature that may be involved in some types of ototoxicity. The inefficiency of aminoglycosides and the putative targets of the ototoxic agents are discussed.

Drug-phospholipid interactions: role in aminoglycoside nephrotoxicity

Aminoglycoside antibiotics are known to be transported and accumulated within lysosomes of renal proximal tubular cells and to cause proximal tubular cell injury and necrosis. The pathogenesis of aminoglycoside nephrotoxicity is postulated to be related to the capacity of these organic polycations to interact electrostatically with membrane anionic phospholipids and to disrupt membrane structure and function. Aminoglycoside antibiotics have been shown to bind to anionic phospholipids of model membranes and to alter membrane permeability and promote membrane aggregation. In vivo these drugs induce phospholipiduria and a renal cortical phospholipidosis. The latter reflects the accumulation of phospholipid-containing myeloid bodies within the lysosomal compartment consequent to aminoglycoside-induced inhibition of lysosomal phospholipases. The mechanism of drug-induced inhibition of phospholipases has been shown to be secondary to the binding of these cationic drugs to anionic phospholipids. As the lysosomes became progressively distended with myeloid bodies, they become unstable and eventually rupture, which results in the release of acid hydrolases as well as high concentrations of aminoglycosides into the cytoplasm where they interact with and disrupt the function of other membranes and organelles including mitochondria and microsomes. It is postulated that the redistribution of drug from the lysosomal compartment to organellar membranes is the critical event which triggers the irreversible injury cascade. Polyaspartic acid is a polyanionic peptide which when administered in vitro or in vivo forms electrostatic complexes with aminoglycoside antibiotics and prevents these drugs from interacting with anionic phospholipids, from perturbing phospholipid metabolism and from causing cell injury and necrosis.

Inhibition of neural phospholipase D activity by aminoglycoside antibiotics

The effects of aminoglycoside antibiotics on phospholipase D (PLD) activity were investigated in permeabilized NG108-15 cells and in isolated rat brain membranes. Neomycin inhibited guanosine 5'-[gamma-thio]triphosphate-stimulated PLD activity in digitonin-permeabilized NG108-15 cells in a concentration-dependent manner (50% inhibition at 100 microM). Neomycin similarly inhibited PLD activity present in rat brain membranes and assayed in vitro with [3H]phosphatidylcholine as substrate (50% inhibition at 65 microM). Other aminoglycosides tested (kanamycin, geneticin and streptomycin) were nearly equipotent inhibitors of rat brain PLD. These results indicate that aminoglycoside antibiotics inhibit phosphatidylcholine-PLD activity with comparable and sometimes greater potency than their well known inhibition of phosphoinositide-phospholipase C. The possibility that PLD inhibition could mediate some of the toxic side effects of aminoglycosides is suggested.

Neomycin does not interfere with the inositol phospholipid metabolism, but blocks binding of alpha-thrombin to intact human platelets

Neomycin was demonstrated to inhibit the binding of thrombin to intact human platelets. The effects of neomycin on both thrombin binding and thrombin-induced changes in inositol phospholipid metabolism could be reproduced by the thrombin antagonist hirudin. We propose that neomycin inhibits thrombin-induced platelet activation by interference with the cellular receptor.

Comparative study of the effect of gentamicin on the vestibulo-ocular and visual vestibulo-ocular reflexes in the cat

The ototoxic effect of an aminoglycoside, gentamicin, on the vestibular system was investigated in cats given daily doses of 40 mg/kg i.m. for 14 days. Periodically, measurements were made of the vestibulo-ocular reflex (VOR) and visual vestibulo-ocular reflex (ViVOR) responses induced by rotatory stimuli at various frequencies from 0.0125 Hz to 0.8 Hz. After the cessation of drug administration, a progressively declining response to VOR stimuli continued, manifested by gain (ratio of peak response to peak stimulus amplitude) and phase relationships. The ViVOR was affected only in the gain measurements. The changes in the response amplitude (gain) were greater for the VOR than for the ViVOR responses and also for the lower (0.0125 Hz) than for the higher frequencies (0.8 Hz). There was some indication that the responses improved about one month after treatment was terminated. All of these specific response changes in cats are comparable to the known effects of aminoglycides in humans, and the same theoretical interpretation of the data in the context of a model of vestibular function can be applied in both cases.

Polyaspartic acid inhibits gentamicin-induced perturbations of phospholipid metabolism

We investigated whether polyaspartic acid (PAA) can inhibit aminoglycoside-induced perturbations of phospholipid metabolism in cultured renal cells of opossum and rabbit and examined the mechanism involved. Cells incubated in medium containing gentamicin (10(-3) M) manifested a time-dependent increase in total phospholipid in association with the appearance of lysosomal myeloid bodies, impaired degradation of phospholipid, and disruption of the phosphatidylinositol (PI) cascade in response to bradykinin stimulation. These alterations of phospholipid metabolism were either completely or almost completely prevented in cells grown in medium containing gentamicin (10(-3) M) and PAA (3 x 10(-4) M, mol wt 11,000) even though PAA did not inhibit the cellular accumulation of gentamicin (40 +/- 1 vs. 42 +/- 1 micrograms/mg protein). In other in vitro studies, we demonstrated that gentamicin depressed the permeability of phosphatidylcholine (PC)/PI liposomes to glycerol and promoted liposomal aggregation. Both effects were blocked by prior addition of PAA. Methylene blue, a cationic dye, was shown to form an electrostatic complex with PAA; gentamicin competitively displaced methylene blue bound to PAA. Our results support the conclusion that the protective effect of PAA is related to its ability to serve as an anionic substrate that electrostatically binds aminoglycoside antibiotics and, thereby, prevents these polycationic drugs from interacting electrostatically with anionic phospholipid of cell membranes.

Stereocilia bundle stiffness: effects of neomycin sulphate, A23187 and concanavalin A

The effects that the aminoglycoside antibiotic neomycin sulphate, the ionophore A23187 and the lectin Concanavalin A (Con A) have on the steady state stiffness of the hair-cell stereocilia bundle have been studied using organotypic cultures of the early postnatal mouse cochlea. In normal saline, stereocilia bundle stiffness is increased 1.5-2.0 fold by neomycin sulphate, 1.3-1.7 fold by the ionophore A23187 and 3.0-5.0 fold by Con A. In low-calcium saline neither neomycin sulphate nor A23187 cause increases in stiffness, and the stiffness increases elicited by these two agents in normal saline are reversed on washout with low-calcium saline. In normal saline neomycin sulphate has two independent effects on hair cells; one effect is a reversible inhibition of transduction and the other effect is to cause an irreversible increase in bundle stiffness. Neither Con A nor the ionophore A23187 block transduction. No obvious changes in stereocilia bundle morphology are associated with the increases in stiffness caused by neomycin sulphate, A23187 and Con A. Succinylated Con A binds to stereocilia bundles but does not cause an increase in stiffness, suggesting that the stiffness increase caused by native Con A results from receptor cross-linking. The effects of Con A and neomycin are non-additive, saturating concentrations of neomycin sulphate block the effects of low doses of Con A, and pretreatment of cells with succinylated Con A prevents subsequent neomycin sulphate treatment from causing an increase in stiffness suggesting that neomycin sulphate and Con A are acting via a similar mechanism at the cell surface. Ionophore treatment prevents the subsequent application of neomycin sulphate from causing a further increase in stiffness, but when cells are treated with neomycin sulphate followed by ionophore the effects of the two drugs are additive indicating that they are operating via different mechanisms. The possible nature of these mechanisms and their role in the control of steady state stereocilia bundle stiffness are discussed.

R 56865 and flunarizine displace Ca2+ from phosphatidylserine monolayers in a stoichiometric manner

The monovalent cationic drugs, R 56865 and flunarizine added to the subphase, displaced Ca2+ from phosphatidylserine (PS) monolayers at pH 5 with EC50 values of 1.2 X 10(-6) and 1.3 X 10(-6) mol/l. The EC50 of flunarizine for the increase in surface pressure due to incorporation of the drug into the PS film was 1.8 X 10(-6) mol/l. The maximal binding of both drugs to PS monolayers was estimated from experiments in which the drug was added to the spreading solution for the monolayer. The maximal molar binding ratio of R 56865 was identical to twice the maximal Ca2+ binding (2 X 0.13 Ca/PS) to the monolayer. The uptake of [14C]R 56865 into PS monolayers could be measured directly and was well described by a Langmuir isotherm in which the EC50 for 45Ca displacement was taken as the concentration for half-maximal binding, and the molar ratio for twice the Ca2+ binding was taken as maximal molar binding ratio. These results show that two molecules of the univalently charged R 56865 displace one molecule of Ca2+. Maximal binding of the cationic drug flunarizine to PS monolayers showed a similar stoichiometry. 45Ca displacement data may thus be useful for predicting monolayer and perhaps also membrane concentrations of the charged species of cationic drugs.

Drug-induced ototoxicity. Pathogenesis and prevention

Ototoxicity is a disabling adverse effect of several widely used classes of drugs, such as diuretics, anti-inflammatory agents, antineoplastic agents and aminoglycoside antibiotics. High-dose therapy with either diuretics or anti-inflammatory agents is primarily associated with acute and transient impairment of hearing or tinnitus. In contrast, long term treatment with antineoplastic agents or aminoglycoside antibiotics is typically associated with delayed and irreversible loss of hearing; lesion in the organ of Corti include the destruction of auditory sensory cells. Vestibular function can also be compromised by ototoxic drugs. Occasional cases of ototoxicity have been reported for a variety of other therapeutic compounds and environmental toxins. In addition, the simultaneous administration of multiple agents which are potentially ototoxic can lead to synergistic loss of hearing. Exposure to loud noise may also potentiate the hearing loss due to cochleotoxic drugs. Ototoxic agents can impair the sensory processing of sound at many cellular or subcellular sites. However, the molecular mechanisms of ototoxicity have not been established for most of these drugs, and structure-toxicity relationships have not been determined. It has therefore been difficult to predict the ototoxic potential of new drugs, and rational approaches to the prevention of ototoxicity are still lacking. The clinical and experimental features of ototoxicity are reviewed for several classes of drugs, with an emphasis on current knowledge of the mechanism and the possibilities for the prevention of ototoxicity for each.

Binding of neomycin to phosphatidylinositol 4,5-bisphosphate (PIP2)

Schacht (Schacht, J. (1976) J. Neurochem. 27, 1119-1124) demonstrated that neomycin, an aminoglycoside antibiotic, binds with high affinity to phosphatidylinositol 4,5-bisphosphate (PIP2). We investigated the binding of neomycin to PIP2 by making electrophoretic mobility measurements with multilamellar bilayer vesicles and surface potential measurements with monolayers. The bilayers and monolayers were formed from mixtures of PIP2 and egg phosphatidylcholine (PC) in 0.1 M KCl at pH 7. Neomycin does not bind to PC; 10(-3) M neomycin affects neither the zeta potential of PC vesicles nor the surface potential of PC monolayers. In contrast, 10(-6) M neomycin reduces the magnitude of the zeta potential of PC/PIP2 vesicles (5, 9, and 17 mol% PIP2) and the surface potential of monolayers (17 mol% PIP2) to less than 50% of their initial values. The electrophoretic mobility results indicate that neomycin forms an electroneutral complex with PIP2; high concentrations (greater than 10(-4) M) of neomycin reduce the zeta potential of the PC/PIP2 vesicles to zero. We could describe our data with the Gouy-Chapman-Stern theory assuming the intrinsic association constant of the 1:1 neomycin-PIP2 complex is 10(5) M-1. Neomycin is widely used in cell biology to interfere with the generation of second messengers; we discuss the relevance of our results to these studies. Specifically, 10(-6) M neomycin binds greater than 50% of the PIP2 in a bilayer or monolayer but 10(-5)-10(-3) M neomycin is required to affect the turnover of PIP2 in permeabilized platelets, mast cells, and sea urchin eggs. This result is consistent with a hypothesis that most of the PIP2 in the inner leaflet of these plasma membranes is not accessible to neomycin because it is associated with proteins.

Separation of liposome populations differing in phosphoinositide content by chromatography on immobilized neomycin

A method is presented by which liposomal populations can be separated based on their content of phosphatidylinositol 4,5-bisphosphate (PIP2). Neomycin is reductively coupled to sepharose 2B and used as the stationary phase in column chromatography. Liposomes of egg lecithin (EL) plus acidic phospholipid containing as little as 10 mol% PIP2 are quantitatively retained in 0.2 M NaCl, while liposomes containing other anionic lipids are recovered to 70-97%. It is possible that lipid redistribution of the bilayers upon contact with the sepharose:neomycin adduct may result in the formatin of multiple populations, and the percent retention by the column for each of the EL:phosphoinositide liposomes would be determined by the new equilibrium established as a result of these interactions. When mixed liposomal populations were chromatographed, the recovery in the 0.2 M NaCl eluates was the sum of the individual recoveries, and there is no significant interchange of lipids between the two liposome populations. It is suggested that this procedure may be useful in the preparation of asymmetric liposomes by facilitating separation of populations devoid of PIP2 in the outer leaflet from unreacted liposomes after enzymatic treatment of symmetric PIP2-containing liposomes.

Transtympanic administration of aminoglycosides in patients with Menière's disease

Transtympanic administration of gentamicin was used to treat patients with unilateral Menière's disease and was stopped at the first signs of damage to cochlear and/or vestibular function. Twenty-eight patients were treated between 1974 and 1981 and 34 patients were managed between 1981 and 1987. In the two groups of patients, objective vertigo disappeared in 85%, hearing improved in 26%, aural fullness disappeared in 78%, while tinnitus ceased altogether in 51% and became less intense in 24%. Our findings show that this kind of treatment can be used after unsuccessful medical therapy and prior to any labyrinthine or endolymphatic sac surgery.

pH-dependent influence of membrane-incorporated flunarizine on Ca-binding to phosphatidylserine monolayer membranes

The pH-dependent 45Ca binding to phosphatidylserine monolayers was investigated. Ca binding increased with increasing pH. Between pH 10 and pH 11 a steep increase of Ca binding could be observed. This increase was interpreted to be due to complex Ca binding opposed to ionic binding at low pH. Flunarizine added to the spreading solution of the monolayer dose dependently displaced up to 100% Ca at pH 5 independently of phospholipid packing. At pH 11 less than 20% of Ca could be displaced by flunarizine. Intermediate results were found at pH 7. Flunarizine displaced less Ca from dense than from loosely packed monolayers at pH 7. The results suggest two binding states of flunarizine: ionic binding at low pH and apolar binding at high pH. The latter is much less effective in displacing Ca from phosphatidylserine monolayers. The Ca displacing properties of charged flunarizine may prevent a deleterious phospholipid reorientation within the membrane induced by the intracellular Ca rise during ischemia.

Pharmacokinetics of aminoglycoside antibiotics in blood, inner-ear fluids and tissues and their relationship to ototoxicity

This review critically evaluates the literature on aminoglycoside pharmacokinetics in order to answer the question how fluid and tissue levels of the drugs relate to the development of ototoxic and nephrotoxic side effects. We will summarize the evidence that: (1) aminoglycosides do not accumulate in inner-ear fluids; (2) aminoglycoside levels in fluids do not correlate with the ototoxic potential of a drug, and (3) selective toxicity cannot be explained by selective tissue penetration of the drugs. We suggest that studies of drug disposition at the cellular level after chronic aminoglycoside treatment be conducted to establish whether a cell-specific uptake contributes to the selective toxicity of the aminoglycoside antibiotics. A sequence of biochemical events that may lead to the development of toxicity at the molecular level is briefly described.

Three molecular steps of aminoglycoside ototoxicity demonstrated in outer hair cells

Previously postulated molecular mechanisms of aminoglycoside ototoxicity were investigated in outer hair cells in vitro. Cells were isolated by microdissection from the organ of Corti of the guinea pig and maintained in Hank's balanced salt solution. (1) Approx. 100 cells (the standard number per assay) bound 432 +/- 198 pmol calcium as determined with 45Ca2+ (1.2 mM). 1 mM neomycin or gentamicin lowered this value by 28% and 45%, respectively. (2) Binding of radiolabeled gentamicin (28 fmol per 100 cells at 0.1 microM gentamicin) was reduced by 55% by 1 mM neomycin or spermine. Washing with an excess of unlabeled gentamicin displaced only 60% of the drug, the remainder being tightly bound to a less accessible compartment. Incubation at low temperature essentially abolished gentamicin uptake. (3) Phospholipids were labeled with [32P]orthophosphate which was mostly incorporated into phosphatidylinositol 4,5-bisphosphate (PIP2), phosphatidylinositol 4-phosphate, phosphatidylinositol, and phosphatidic acid. When the lipids were chromatographed over immobilized gentamicin PIP2, in contrast to other lipids, was strongly retained. These results are compatible with the following actions of aminoglycosides in outer hair cells: (1) competition with calcium; (2) energy-dependent uptake competitive with polyamines; and (3) binding to the phospholipid PIP2.

Neomycin inhibits inositol phosphate formation in human platelets stimulated by thrombin but not other agonists

Neomycin (0.1-1 mM) added to human platelet-rich plasma or washed platelets prelabeled with [3H]inositol inhibits aggregation, ATP secretion (ID50 0.2 mM) and formation of [3H]inositol mono-, bis- and trisphosphate (ID50 0.6-0.8 mM) in response to thrombin (0.25 U/ml). The production of inositol phosphates in response to other platelet agonists (vasopressin, platelet activating factor, prostaglandin endoperoxide analogs and collagen) is not inhibited by neomycin, even at a concentration of 2 mM. At this concentration neomycin reduces the secretion of ATP stimulated by these agents (by up to 50%). The results indicate that neomycin has multiple effects on platelets that are unrelated to a specific inhibition of inositol phospholipid degradation by phospholipase C. Low concentrations (0.1-1 mM) of neomycin might selectively inhibit the interaction of thrombin with the platelet surface, and high concentrations (greater than 2 mM) might unspecifically reduce platelet secretion in response to various platelet agonists.

Membrane perturbation by aminoglycosides as a simple screen of their toxicity

A simple test to assess the toxicity of aminoglycosides was designed based on the interaction of these drugs with phosphatidylinositol bisphosphate. The fluorescent probe 1-anilino-8-naphthalene sulfonate was incorporated into liposomes composed of L-alpha-phosphatidylcholine and phosphatidylinositol bisphosphate (4:1 molar ratio), and changes in fluorescence induced by the aminoglycosides were measured. The magnitude of these changes correlated well with the intrinsic ototoxicity of the drugs previously established in cochlear perfusions: neomycin greater than gentamicin approximately equal to tobramycin greater than amikacin approximately equal to kanamycin approximately equal to netilmicin greater than neamine approximately equal to spectinomycin. Correlations were not accurate when other phospholipids were substituted for phosphatidylinositol bisphosphate. An alternative in vitro test, measurement of the zeta potential of liposomes, ranked the drugs according to their electrostatic charge but not their toxicity. These results suggest that: a simple screen for aminoglycoside toxicity can be established and perturbation of membranes containing phosphatidylinositol bisphosphate is a key step in aminoglycoside ototoxicity.

Comparative uptake of gentamicin, netilmicin, and amikacin in the guinea pig cochlea and vestibule

The kinetics of the entry of three aminoglycosides into inner-ear tissues of the guinea pig after acute and chronic administration were compared: gentamicin toxic to the cochlea and the vestibule, amikacin preferentially cochleotoxic, and netilmicin of low ototoxic liability. During constant intravenous infusion, levels of the three drugs in plasma tended to reach a plateau after 1 h, while levels in perilymph did not reach a plateau within 6 h. The drug concentrations in both vestibular and cochlear tissues quickly reached saturation. Amikacin and gentamicin concentrations were similar in vestibular and cochlear tissues, while netilmicin values were somewhat lower. After 1 week of chronic treatment (100 mg of drug per kg of body weight daily subcutaneously), levels of gentamicin and amikacin in tissue were similar to each other and were not significantly different between cochlear and vestibular tissues. Netilmicin concentrations again were somewhat lower in the tissues, but identical to those of the other drugs in the perilymph. After 3 weeks of treatment, all of the drugs were equally distributed in the inner-ear tissues. Release of the drug from the tissues after the 3-week treatment was faster for amikacin (83% decrease after 20 days) than for netilmicin and gentamicin (approximately 50% decrease). There was no correlation, under any of the experimental conditions, between the drug concentrations and their degrees of toxicity. These results demonstrate that selective aminoglycoside ototoxicity cannot be explained by a preferential uptake or accumulation of drugs in the afflicted tissues or in the perilymph.