Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Jun 16.
Published in final edited form as: Nat Prod Commun. 2012 Mar;7(3):317–320.

Live Cell Imaging of a Fluorescent Gentamicin Conjugate

Jorge O Escobedo a, Yu-Hsuan Chu a, Qi Wang b, Peter S Steyger b, Robert M Strongin a,*
PMCID: PMC3376166  NIHMSID: NIHMS384080  PMID: 22545403

Abstract

Understanding cellular mechanisms of ototoxic and nephrotoxic drug uptake, intracellular distribution, and molecular trafficking across cellular barrier systems aids the study of potential uptake blockers that preserve sensory and renal function during critical life-saving therapy. Herein we report the design, synthesis characterization and evaluation of a fluorescent conjugate of the aminoglycoside antibiotic gentamicin. Live cell imaging results show the potential utility of this new material. Related gentamicin conjugates studied to date quench in live kindney cells, and have been largely restricted to use in fixed (delipidated) cells.

Keywords: Aminoglycosides, Gentamicin, Ototoxic drugs

Aminoglycoside antibiotics are essential natural products used in the treatment of tuberculosis and Gram-negative infections [1,2], as well as in the prophylactic protection of burns and bacterial sepsis in premature infants [3–5]. However, acute nephrotoxicity and permanent ototoxicity are serious side-effects that increase patient morbidity, or induce permanent auditory and vestibular deficits.

Several approaches have been used to study the intracellular mechanisms induced by ototoxic drugs. However, ototoxic drugs must first cross the blood-labyrinth barrier (BLB) before entering sensory hair cells to exert their cytotoxic effect that leads to hearing loss and deafness. The BLB is similar to the blood-brain barrier (BBB), and is composed of tight junction-coupled endothelial and epithelial barrier layers that transport nutrients and ions from one side of the barrier to the other.

Fluorophore-tagged drugs, such as gentamicin-Texas Red (GTTR) conjugates, have been used to identify drug trafficking routes across the BLB and into hair cells, and specifically in the cochlea - the sensory organ responsible for hearing [6–9]. Identifying the mechanisms involved in the intracochlear trafficking of ototoxic drugs to hair cells is of fundamental as well as clinical importance. Morphological studies show endocytotic uptake of aminoglycosides at the apical pole of hair cells, i.e., the membrane bathed in the positively-charged endolymph [10]. In vitro data indicate that aminoglycosides not only block the mechanosensory transduction channel located on apical hair cell stereociliary membranes bathed in endolymph, but that these drugs permeate through the mechanotransduction channel directly into the cytoplasm of hair cells [11]. If cationic aminoglycosides enter hair cells from endolymph in vivo, they must overcome the electrorepulsive force of the positively-charged endolymph (+80 mV), and the tight junction-coupled epithelial barrier lining the scala media prior to entering endolymph. Competitive inhibition of GTTR uptake into the cochlea in vivo suggests that aminoglycosides are actively transported across the stria vascularis, a major component of the BLB that regulates endolymph composition [6,8,9].

GTTR in vivo studies have largely been performed using fixed tissues, as live cells can quench GTTR fluorescence [12]. It is known that aminoglycosides preferentially bind to phosphatidylinositol 4,5-bisphosphate (PIP2) [13]. Previous studies have shown that GTTR binding to PIP2 and similar phospholipids (PLs) are likely to be responsible for quenching GTTR fluorescence in live cells, and that this quenching can be unmasked by fixing and delipidating cells using either Triton X-100 or similar delipidating agents [12]. The linker between gentamicin and Texas Red conjugates used until now has flexible properties (it can adopt many unrestricted conformations along each bond). Thus, we investigated whether the effect of a rigid linker between the fluorescent probe and gentamicin can overcome the observed fluorescence quenching. Using a commercially-available Texas Red derivative, we designed a new version of a Texas Red-type probe (compound 4, scheme 1). Molecular simulations (Figure 1) show that the biaryl linker between the fluorescent probe and gentamicin meets the requirements to prevent supramolecular interactions between the two moieties and potentially act as a molecular ruler (1.35 nm) of channel pore diameter in GTTR-permissive ion channels, such as TRPV4, and the mechano transduction channel [8,14,15]. Functional hair cell mechanotransducer channels are required for aminoglycoside ototoxicity.

Scheme 1.

Scheme 1

Open in a new tab

Synthesis of compound 4.

Figure 1.

Figure 1

Open in a new tab

Energy-minimized structure of an isomeric conjugate of the inflexible Texas Red (compound 4) with gentamicin. Note that potential direct intramolecular contacts between the drug and fluorophore are suppressed. Hydrogens are hidden for clarity.

As shown in scheme 1, compound 4 containing an isothiocyanate amine reactive group for conjugation to gentamicin was prepared in four steps with an overall yield of 13.4%. The initial step involves the condensation of 8-hydroxyjulolidine with p-bromobenzaldehyde under microwave irradiation to produce bromorhodamine (1) [16]. Suzuki coupling of 1 with 4-(N-Boc-amino) phenylboronic acid leads to the formation of compound 2 (procedure based on protocols [17]). Removal of the Boc group via acid hydrolysis yielded compound 3, which is reacted with 1,1′-thiocarbonyldi-2(1H)-pyridone to obtain compound 4 (using procedure [18]). Conjugation of compound 4 to gentamicin to give conjugate 5 (iGTTR, inflexible gentamicin-Texas Red) is carried out at pH 10 using a molar ratio of compound 4: gentamicin of 1:100 in order to ensure that only one molecule of the fluorophore is conjugated to one molecule of gentamicin.

Spectroscopic characterization of compound 4 (Figure 2) shows that its fluorescence emission maxima (600 nm) is blue shifted compared with that of Texas Red (608 nm).

Figure 2.

Figure 2

Open in a new tab

Excitation/Emision Fluorescence spectra of compound 4 in MeOH.

Cellular uptake of iGTTR was investigated in MDCK cells using confocal fluorescence microscopy. When incubated with iGTTR for 1 min, and imaged live, strong, distinctive fluorescence is observed in the live MDCK cells. (Figure 3A). The majority of this fluorescence is lost after fixation with 4% paraformaldehyde, although some puncta fluorescence remained visible, presumably in endosomes, as described previously for GTTR [12]. When live cells treated with iGTTR are fixed with 4% paraformaldehyde containing 0.5% Triton X-100 (FATX), or formaldehyde-fixed cells subsequently treated with Triton X-100 (0.5%), this reduced fluorescence is completely masked. This preliminary result shows that this newly-developed conjugate iGTTR will be extremely valuable for drug trafficking studies in live cells and across tight-junction-coupled cellular barriers. These fluorescence properties also provide further insight into the mechanism that explains why fluorescence occurs in live cells, but not in fixed (delipidated) cells. This knowledge will be valuable in generating a new range of fluorophores that retain their fluorescence properties in both live and fixed cell states.

Figure 3.

Figure 3

Open in a new tab

MDCK cells treated with 10 μg/mL purified iGTTR with an inflexible, rigid linker, and imaged as described in the text. All images acquired with identical confocal imaging parameters.

The relative quantum yield of 5 with respect to Texas Red is 0.7 and its extinction coefficient at 576 nm is 71,975 L mol−1 cm−1.

Drug-fluorophore conjugates that (i) fluoresce in live cells, (ii) demonstrate uptake modulation by cell regulatory drugs, (iii) are cytotoxic, and (iv) can cross the blood-labyrinth barrier into marginal cells and hair cells will be considered bioactive drug-fluorophore conjugates with superior imaging qualities. These superior qualities will allow investigators to use these compound-fluorophore combinations for (i) live cell imaging over time in the same cell types, or (ii) trafficking across epithelial barrier layers, and (iii) allow subsequent post-fixation imaging for high resolution imaging of the final disposition of compound-fluorophore distribution. Characterizing the cellular mechanisms of ototoxic and nephrotoxic drug uptake, their intracellular distribution, and molecular trafficking mechanisms across cellular barrier systems will enable identification of potential uptake blockers that preserve sensory and renal function during critical life-saving therapy.

Experimental

General

1H and 13C NMR spectra were obtained on a ARX-400 Advance Bruker spectrometer. HR ESI MS were carried out on a Accela LC system and a Finnigan LTQ-XL linear ion trap and LTQ Orbitrap discovery mass spectro-meter detectors. Microwave synthetic procedures were per-formed in an Initiator™ microwave synthesizer (Biotage). Simulations were performed on a Dell™ precision 490 loaded with Red Hat™ Enterprise 4 Linux 64-bit operating system using the molecular dynamics routines within the SYBYL™ modeling suite version X-1.3.

Compound 1

In a 20 mL microwave tube, 8-hydroxyjulolidine (380 mg, 2 mmol, 2 equiv) and 4-bromobenzaldehyde (185 mg, 1 mmol, 1 equiv) were dissolved in methanesulfonic acid (7.5 mL). A magnetic stirring bar was introduced and the tube was sealed and irradiated in the microwave reactor keeping the temperature at 150ºC under stirring for 5 min. The mixture was allowed to cool down to room temperature, and the tube was opened. Tetrachloro-1,4-benzoquinone (369 mg, 1.5 mmol, 1.5 equiv) was added to the solution, and stirred for an additional 15 min. The dark blue mixture was neutralized to pH 7 with 10 M KOH. The solution was poured into water and extracted with MeOH-CH2Cl2 [5:95, (3 × 100 mL)]. The combined organic layers were washed with brine, dried over Na2SO4, and the solvent removed under vacuum. The mixture was separated by flash chromatography on silica using solutions of 5–10% MeOH in CH2Cl2 for elution to give a dark violet solid (250 mg, 47%). Rf: 0.2 (MeOH-CH2Cl2 5:95). 1H and 13C NMR spectroscopic and MS data corresponded with published values [16].

Compound 2

Bromorhodamine 1 (200 mg, 0.378 mmol, 1 equiv) was dissolved in toluene (7 mL/mmol) and an aqueous 2 M Na2CO3 (3.2 mL/mmol) and an ethanolic solution (3.2 mL/mmol) of 4-(N-Boc-amino) phenylboronic acid (459 mg 1.938 mmol, 5.12 equiv). The mixture was deoxygenated under reduced pressure and flushed with Ar 3 times. Pd(PPh3)4 (48 mg, 11 mol%) was added, and the reaction mixture was heated at 60ºC overnight. The solution was poured into H2O, and washed with 10% MeOH in CH2Cl2. The organic layers were washed with brine and dried over Na2SO4, filtered over a short plug of celite and the solvent removed under vacuum. The mixture was separated by flash chromatography on silica using 10% MeOH in CH2Cl2 for elution to give a dark violet solid (109 mg, 45%). Rf: 0.4 (MeOH-CH2Cl2 10:90).

1H NMR (CDCl3): 1.48 (9H, s, CH3), 1.92 (4H, m, CH2), 2.05 (4H, m, CH2), 2.65 (4H, t, J = 6.0 Hz, CH2), 2.98 (4H, d, J = 6.3 Hz, CH2), 3.47 (4H, t, J = 5.7 Hz, CH2), 3.51 (4H, t, J = 5.6 Hz, CH2), 6.83 (2H, s, CH), 6.95 (1H, s, NH), 7.29 (2H, sJ = 8.2 Hz, CH,), 7.50 (2H, d, J = 8.6 Hz, CH), 7.56 (2H, d, J = 8.6 Hz, CH), 7.70 (2H, d, J = 8.2 Hz, CH).

13C NMR (CDCl3): 19.9 (CH2), 20.0 (CH2), 20.8 (CH2), 27.8 (CH2), 28.5 (CH3), 50.6 (CH2), 51.1 (CH2), 80.8 (C), 105.5 (C), 112.9 (C), 119.1 (CH), 123.8 (C), 126.8 (CH), 126.9 (CH), 127.7 (CH), 130.1 (CH), 131.1 (C), 134.1 (C), 138.9 (C), 142.2 (C), 151.2 (C), 152.3 (C), 153.0 (C), 154.7 (C).

HRMS (ESI): m/z [M + H+] calcd for C42H44N3O3: 638.3377; found: 638.3378.

Compound 3

To a stirred solution of compound 2 (97 mg, 0.151 mmol) in CH2Cl2 (3 mL) TFA (4.5 mL) was added dropwise at 0ºC. The reaction was stirred for 2 h at the same temperature, and then neutralized with saturated aqueous NaHCO3 to pH 7. The organic layers were washed with water, then brine, dried over Na2SO4, and the solvent removed under vacuum. The mixture was separated by flash chromatography on silica using 5%–10% MeOH in CH2Cl2 for elution to give a dark violet solid (yield: 72.4 mg, 89%). Rf: 0.37 (MeOH-CH2Cl2 10: 90).

1H NMR (CDCl3): 1.99 (4H, m, CH2), 2.12 (4H, m, CH2), 2.72 (4H, t, J = 6.0 Hz, CH2), 3.05 (4H, d, J = 6.3 Hz, CH2), 3.53 (4H, t, J = 5.7 Hz, CH2), 3.57 (4H, t, J = 5.7 Hz, CH2), 6.82 (2H, d, J = 8.5 Hz, CH), 6.92 (2H, s, CH), 7.33 (2H, d, J = 8.3 Hz, CH), 7.74 (2H, d, J = 8.5 Hz, CH), 7.79 (2H, d, J = 8.3 Hz, CH).

13C NMR (CDCl3): 19.9 (CH2), 20.0 (CH2), 20.8 (CH2), 27.8 (CH2), 50.6 (CH2), 51.1 (CH2), 105.6 (C), 113.0 (C), 115.7 (CH), 123.7 (C), 126.4 (CH), 126.9 (CH), 128.1 (CH), 129.6 (C), 130.1 (CH), 130.3 (C), 142.8 (C), 147.1 (C), 151.2 (C), 152.4 (C), 155.0 (C).

HRMS (ESI): m/z [M + H+] calcd for C37H36N3O: 538.2853; found: 538.2855.

Compound 4

To a stirred solution of compound 3 (18.1 mg, 0.0325 mmol, 1 equiv) in anhydrous CH2Cl2 (8 mL) was added 1,1′-thiocarbonyldi-2(1H)-pyridone (15.6 mg, 0.0672 mmol, 2 equiv). The resulting mixture was stirred at room temperature for 18 h under Ar. The mixture was poured into CH2Cl2, washed with H2O, then brine, dried over Na2SO4, and the solvent removed under vacuum. The residue was purified by flash chromatography on silica using MeOH-CH2Cl2, 5:95, for elution to give a dark violet solid (yield: 13.8 mg, 71%). Rf: 0.32.

UV/Vis λmax (MeOH) nm (log ε): 576 nm (4.86).

1H NMR (CDCl3): 1.98 (4H, m, CH2), 2.11 (4H, m, CH2), 2.72 (4H, t, J = 5.9 Hz, CH2), 3.04 (1H, t, J = 6.3 Hz, CH2), 3.53 (4H, dd, J = 8.3 Hz, J = 6.2 Hz, CH2), 3.56 (4H, t, J = 5.6 Hz, CH2), 6.85 (2H, s, CH), 7.35 (2H, d, J = 8.5 Hz, CH), 7.41 (2H, d, J = 8.3 Hz, CH), 7.70 (2H, d, J = 8.5 Hz, CH), 7.79 (2H, d, J = 8.3 Hz, CH).

13C NMR (CDCl3): 19.8 (CH2), 20.0 (CH2), 20.8 (CH2), 27.8 (CH2), 50.6 (CH2), 51.1 (CH2), 105.6 (C), 112.8 (C), 123.9 (C), 126.5 (CH), 126.6 (CH), 127.4 (CH), 128.4 (CH), 130.3 (CH), 131.2 (C), 132.2 (C), 134.9 (N=C=S), 138.9 (C), 141.1 (C), 151.3 (C), 152.3 (C), 154.1 (C).

HRMS (ESI): m/z [M + H+] calcd for C38H34N3OS: 580.2417; found: 580.2419.

Compound 5

Inflexible Texas Red-gentamicin (iGTTR) conjugate was prepared as described for GTTR) [12,19]. Briefly, gentamicin sulfate and compound 4 were mixed respectively in a 100:1 molar ratio in 2.4 mL of 100 mM K2CO3, pH 10, buffer. The mixture was stirred at room temperature for 48 h. The target compound was isolated by SPE reversed phase C18 (Grace/Alltech, IL) equilibrated with MeOH and using 5% aqueous glacial acetic acid; MeOH; and CHCl3-MeOH-NH4OH (conc) 30:41:30 for elution. The CHCl3-MeOH-NH4OH (conc) fraction containing the iGTTR conjugate as an isomeric mixture was evaporated under vacuum.

UV-Vis absorption and fluorescence spectroscopy

UV-Vis spectra were collected with a Cary 50 Bio UV-Vis spectrophotometer at rt using a 1 cm quartz cuvette. Fluorescence spectra were obtained using a Cary Eclipse fluorescence spectrophotometer with a 1 cm path length quartz cell. Emission wavelengths were scanned with 1 nm step sizes using a band pass of 5 nm. Integration time was set to 1.0 second per point and 900 V was applied to the PMT detector. Excitation spectra were monitored with an emission and excitation band pass of 5 nm. All spectra were fully corrected. Relative fluorescence quantum yield was obtained with respect to Texas Red™ succinimyl ester.

Cell culture

Madin-Darby canine kidney (MDCK) cells of collecting duct origin were cultured in antibiotic- and phenol red-free minimal essential medium (MEM, Invitrogen, Ca) with 10% fetal bovine serum at 37°C with 5% CO2, 95% air. For iGTTR experiments, cells were further seeded into 8-well coverglass chambers (ISC BioExpress) and grown to ~90% confluent overnight and had developed tight junctions [18]. Confluent MDCK cells were treated with 1 μg/mL of iGTTR in DPBS (Invitrogen, Ca) with 1.25 mM Ca2+ for 30 secs at 37 °C. Cells were then washed 3 times with DPBS with 1.25 mM Ca2+ and imaged live, or fixed with (i) 4% paraformaldehyde alone (PFA) for 45 mins at room temperature; (ii) fixed with 4% paraformaldehyde and 0.5% Triton X-100 (FATX) in 0.1 M phosphate buffered saline (PBS) for 45 mins; or (iii) fixed with 4% PFA for 45 mins at room temperature, and followed by delipidation with 0.5% Triton X-100 in PBS [12]. After washing, cells were observed using a Nikon TE 300 inverted microscope (Melville, NY), and confocal images collected on a Bio-Rad (Hercules, CA) MRC 1024 ES scanning laser system fitted with standard excitation and emission filters for Texas Red fluorophores (excitation: 568 ± 32 nm; emission: 620 ± 16 nm). Image files were exported as TIF files and prepared for publication using Adobe Photoshop.™

Acknowledgments

Support from the National Institutes of Health via award R01 EB002044 is gratefully acknowledged. PSS and QW were supported by NIDCD R01 DC04555.

References

  • 1.Begg EJ, Barclay ML. Aminoglycosides - 50 years on. British Journal of Clinical Pharmacology. 1995;39:597–603. [PMC free article] [PubMed] [Google Scholar]
  • 2.de Jager P, van Altena R. Hearing loss and nephrotoxicity in long-term aminoglycoside treatment in patients with tuberculosis. International Journal of Tuberculosis and Lung Disease. 2002;6:622–627. [PubMed] [Google Scholar]
  • 3.Bockenhauer D, Hug MJ, Kleta R. Cystic fibrosis, aminoglycoside treatment and acute renal failure: the not so gentle micin. Pediatric Nephrology. 2009;24:925–928. doi: 10.1007/s00467-008-1036-2. [DOI] [PubMed] [Google Scholar]
  • 4.Hall JW, Herndon DN, Gary LB, Winkler JB. Auditory brain-stem response in young burn-wound patients treated with ototoxic drugs. International Journal of Pediatric Otorhinolaryngology. 1986;12:187–203. doi: 10.1016/s0165-5876(86)80075-1. [DOI] [PubMed] [Google Scholar]
  • 5.Pillers DAM, Schleiss MR. Gentamicin in the clinical setting. Volta Review. 2005;105:205–210. [Google Scholar]
  • 6.Dai CF, Steyger PS. A systemic gentamicin pathway across the stria vascularis. Hearing Research. 2008;235:114–124. doi: 10.1016/j.heares.2007.10.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Dai CF, Mangiardi D, Cotanche DA, Steyger PS. Uptake of fluorescent gentamicin by vertebrate sensory cells in vivo. Hearing Research. 2006;213:64–78. doi: 10.1016/j.heares.2005.11.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Wang Q, Steyger PS. Trafficking of systemic fluorescent gentamicin into the cochlea and hair cells. Jaro-Journal of the Association for Research in Otolaryngology. 2009;10:205–219. doi: 10.1007/s10162-009-0160-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Wang Q, Kachelmeier A, Steyger PS. Competitive antagonism of fluorescent gentamicin uptake in the cochlea. Hearing Research. 2010;268:250–259. doi: 10.1016/j.heares.2010.06.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Hashino E, Shero M. Endocytosis of aminoglycoside antibiotics in sensory hair cells. Brain Research. 1995;704:135–140. doi: 10.1016/0006-8993(95)01198-6. [DOI] [PubMed] [Google Scholar]
  • 11.Marcotti W, van Netten SM, Kros CJ. The aminoglycoside antibiotic dihydrostreptomycin rapidly enters mouse outer hair cells through the mechano-electrical transducer channels. Journal of Physiology-London. 2005;567:505–521. doi: 10.1113/jphysiol.2005.085951. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Myrdal SE, Johnson KC, Steyger PS. Cytoplasmic and intra-nuclear binding of gentamicin does not require endocytosis. Hearing Research. 2005;204:156–169. doi: 10.1016/j.heares.2005.02.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Schacht J. Purification of polyphosphoinositides by chromatography on immobilized neomycin. Journal of Lipid Research. 1978;19:1063–1067. [PubMed] [Google Scholar]
  • 14.Karasawa T, Wang Q, Fu Y, Cohen DM, Steyger PS. TRPV4 enhances the cellular uptake of aminoglycoside antibiotics. Journal of Cell Science. 2008;121:2871–2879. doi: 10.1242/jcs.023705. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Alharazneh A, Luk L, Huth M, Monfared A, Steyger PS, Cheng AG, Ricci AJ. Functional hair cell mechanotransducer channels are required for aminoglycoside ototoxicity. PLoS One. 2011;6:e22347. doi: 10.1371/journal.pone.0022347. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Jiao GS, Castro JC, Thoresen LH, Burgess K. Microwave-assisted syntheses of regioisomerically pure bromorhodamine derivatives. Organic Letters. 2003;5:3675–3677. doi: 10.1021/ol035327u. [DOI] [PubMed] [Google Scholar]
  • 17.Read MW, Escobedo JO, Willis DM, Beck PA, Strongin RM. Convenient iterative synthesis of an octameric tetracarboxylate-functionalized oligophenylene rod with divergent end groups. Organic Letters. 2000;2:3201–3204. doi: 10.1021/ol000204k. [DOI] [PubMed] [Google Scholar]
  • 18.Sibrian-Vazquez M, Jensen TJ, Fronczek FR, Hammer RP, Vicente MGH. Synthesis and characterization of positively charged porphyrin-peptide conjugates. Bioconjugate Chemistry. 2005;16:852–863. doi: 10.1021/bc050057g. [DOI] [PubMed] [Google Scholar]
  • 19.Li H, Wang Q, Steyger PS. Acoustic trauma increases cochlear and hair cell uptake of gentamicin. PLoS One. 2011;6:e19130. doi: 10.1371/journal.pone.0019130. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES