www.elsevier.com / locate / bres
Research report
Lysosomal augmentation during aminoglycoside uptake in cochlear
hair cells
*
Eri Hashino , Marlene Shero, Richard J. Salvi
Center for Hearing and Deafness, State University of New York at Buffalo, Buffalo, NY 14214, USA Accepted 19 September 2000
Abstract
Aminoglycoside antibiotics, such as kanamycin, have ototoxic side effects, which often result in degeneration of cochlear and vestibular hair cells in the inner ear. Cytotoxic effects of aminoglycosides, however, do not appear immediately after cellular uptake of aminoglycosides. In order to understand the mechanisms responsible for the delayed emergence of aminoglycoside ototoxicity, changes in lysosomal activities in cochlear hair cells were evaluated during a repeated administration of kanamycin by two methods. Electron microscopic localization of acid phosphatase (AcPase) revealed that AcPase started to accumulate in vesicles 27 h after the start of kanamycin administration. In addition, the number and size of AcPase-filled vesicles increased with repeated kanamycin doses. Confocal microscopic localization of the LysoTracker probe, a vital lysosomal marker, showed an increase in the size of lysosomes in hair cells that were treated with kanamycin. The temporal changes in the augmentation of lysosomes paralleled those in intracellular kanamycin levels. These results suggest that the intralysosomal compartments can accumulate extensive amounts of aminoglycosides, which might lead to lysosomal swelling and subsequent rupture. 2000 Elsevier Science B.V. All rights reserved.
Theme: Sensory systems
Topic: Auditory, vestibular, and lateral line: periphery
Keywords: Hair cell; Lysosome; Aminoglycoside; Ototoxicity; Chicken
1. Introduction between the first sign of drug uptake and that of cell death [13,15]. The delayed cytotoxic effects suggest that hair Aminoglycoside antibiotics, such as kanamycin (KM) cells can live for a substantial period of time containing and gentamicin (GM), have long been used in antibacterial high levels of intracellular aminoglycosides.
therapy of severe gram-negative infection. Despite high Recent studies have uncovered several important aspects antibacterial efficacy, aminoglycosides have prominent of aminoglycoside cellular uptake mechanisms. Our previ-ototoxic side effects, causing the destruction of sensory ous studies, using immunogold electron microscopy, dem-hair cells in the inner ear when administered at high doses. onstrated that aminoglycosides are internalized into hair Although evidence clearly indicates that the intracellular cells by receptor-mediated endocytosis, primarily through accumulation of aminoglycosides is associated with hair the apical cell surface [11,24]. Mutant hair cells with cell damage [12,24,28], the cellular and molecular mecha- defects in apical endocytosis were resistant to amino-nisms responsible for the aminoglycoside-induced hair cell glycoside toxicity, strongly supporting our assumption of degeneration remain poorly elucidated. One intriguing endocytotic drug uptake [25]. The drugs are subsequently aspect of aminoglycoside ototoxicity is that the drugs do transported into lysosomes via a vesicle-mediated process not exert cytotoxic effects upon cellular uptake. A consid- [7,13]. These observations are consistent with evidence erable delay of several days to weeks has been reported from kidney epithelial cells, another preferential target of aminoglycoside toxicity. It has been shown that gentamicin is taken up into cultured kidney epithelial cell lines or
*Corresponding author. Tel.:11-716-829-2001 ext. 15; fax:1
1-716-renal proximal tubular cells by an energy-dependent
829-2980.
E-mail address: [email protected] (E. Hashino). endocytotic process and that the drug subsequently
cumulates in lysosomes [1,2,8,16]. In addition, the entry of largest vesicle showing AcPase reactivity in a hair cell was aminoglycosides into renal proximal tubular cells was measured from photomicrographs. A Student’s t-test or, in shown to be mediated by a cell surface receptor [22]. cases where a normality test or an equal variance test Collectively, the currently available data suggest that failed, a Mann–Whitney Rank Sum test was used to aminoglycosides are transported into lysosomes of target evaluate any statistically significant difference in the cells after their endocytotic uptake. Since little is known AcPase-positive area or the total area of a hair cell about the subsequent processes that lead to the slow between groups.
progressive degeneration of hair cells, the present study
investigated the temporal changes in lysosomal activities in 2.3. Immunogold electron microscopy hair cells and related them to the temporal changes in
intracellular KM levels [13]. Parts of this study have been Some of the inner ear tissues treated with KM for 5 days reported in an abstract form [14]. were processed for immunogold electron microscopy according to previously described procedures [13]. Briefly, the inner ear tissues were fixed with 2.5% glutaraldehyde
2. Materials and methods and 2% paraformaldehyde at 48C overnight. After being dissected out of the surrounding bone, specimens were 2.1. Kanamycin administration embedded in LR White resin (Ted Pella, Redding, CA). Ultrathin sections were cut at a position 10% of the A total of 45 White Leghorn chicks (7–15 day old) were distance from the basal end of the cochlea and collected on used as subjects. One group of animals was given a single thin bar nickel grids. Post-embedding immunostaining was intramuscular injection of kanamycin monosulfate (Sigma, carried out first by treating the sections with protein A 400 mg / kg) and sacrificed either 12 h (n510) or 27 h (Sigma, 100 mg / ml), 5% bovine serum albumin (BSA) (n510) post-injection. A second group of animals (n515) and 1% Tween 20 in tris buffered saline (TBS) for 30 min received daily injections of KM (400 mg / kg / day) for 5 at 378C. Then, the sections were incubated with anti-KM consecutive days and was sacrificed 24 h after the last serum (1:40, Western Chemical Research Co., Fort Collins, injection. A third group (n510) was not given KM and CO) for 60 min, followed with protein A-gold (10 nm, served as controls. 1:40, Ted Pella) for 30 min. Finally, the sections were stained with uranyl acetate and examined on a transmission 2.2. Acid phosphatase cytochemistry electron microscope.
Acid phosphatase cytochemistry was based on modi- 2.4. Vital staining of lysosomes fication of published protocols [3,20]. Briefly, the basilar
projected as a single image using Bio-Rad software. X–Z observed in hair cells 12 h after a single KM injection. The sections were also obtained from the frame to visualize the total area showing AcPase reactivity per hair cell
2
localization of LysoTracker staining and Phalloidin stain- (0.0560.008 mm ) was not significantly different from ing along the vertical axis of the hair cell. Quantitative that in control hair cells (t(22)50.799, P50.433, Fig. 2A). analysis of the LysoTracker staining was performed with Notable changes in the area and distribution of AcPase the Image-Pro Plus software. LysoTracker-positive puncta activity were first observed in hair cells 27 h after a single were detected and their areas were automatically calculated KM injection. An intense AcPase staining was observed in with the software. one or two smooth vesicles in each hair cell (Fig. 3). These vesicles were located either beneath the apical plasma membrane or in the subcuticular region of a hair cell. The
3. Results total area with positive AcPase reactivity per hair cell was 2
0.2460.05 mm , which was significantly greater than that 3.1. AcPase histochemistry in control hair cells (t(13,14)5253.5, P,0.001, Fig. 2A). In hair cells obtained from animals that were given KM AcPase histochemistry has long been used as a conveni- for 5 consecutive days, a notable increase in the AcPase-ent and reliable marker for lysosomal structures in many positive area was observed. The hair cells showed a 5-fold cell types [3,20]. Localization of acid hydrolases, the increase in the total AcPase-positive area per hair cell
2
major hydrolytic enzymes in lysosomes, can be visualized (1.3460.33 mm , Fig. 2A) compared to hair cells 27 h by precipitation with their specific substrate. In control post-injection. This increase in the AcPase-positive area animals that were not treated with KM, the AcPase activity can be attributable to an increase in the number of large was observed in a dispersed pattern over the cytoplasmic vesicles filled with AcPase. The majority of hair cells region of a hair cell (Fig. 1). Small puncta of AcPase contained more than two vesicles that showed AcPase activity were more often observed in the apical half rather reactivity (Fig. 4A, B). The average number of AcPase-than the basal half of a hair cell, because a large portion of positive vesicles per hair cell was 4.1 compared to 1.9 at the basal region is occupied by the nucleus. There was no 27 h post-injection. In addition, the size of the largest sign of accumulation of AcPase activity in any discrete vesicle in a hair cell was larger after 5 days of KM than region of a hair cell. The average AcPase positive area per after 1 day of KM (27 h) (t(28)524.43, P,0.001, Fig.
2
hair cell was 0.0660.01 mm . 2B).
A similar distribution of AcPase-positive areas was Immunogold electron microscopy was carried out to
3.2. LysoTracker staining
To verify the observations with AcPase histochemistry and to obtain data from the entire hair cell region, lysosomal structures in hair cells were labeled in vitro with LysoTracker probe and examined on a confocal micro-scope. LysoTracker, a newly developed acidotrophic probe, selectively accumulates in cellular compartments with low pH and can be used to label lysosomes in living cells [6,11]. In addition, acidic compartments of cells stained with the LysoTracker probe retain their staining pattern following fixation, which enables us to double-label the cells with another marker. LysoTracker staining was dispersed in small puncta throughout the cytoplasm of control hair cells (Fig. 5A), whereas significantly larger vesicles accumulating the LysoTracker probe were ob-served in hair cells that were treated with KM for 5 days (Fig. 5B).
The number of LysoTracker-positive puncta (.0.005 2
mm ) in a sampled region (30mm330 mm) was 608 and 226 in a control and KM-treated basilar papilla, respective-ly. Of all the stained puncta, vesicles equal to or larger
2
than 0.05mm were further analyzed. The number of these 2
vesicles (.0.05 mm ) in the control and KM-treated basilar papilla was 68 and 78, respectively. A histogram showing the size distribution of the vesicles for control and KM-treated hair cells is shown in Fig. 6. The vesicle size
2
of most control hair cells was less than 0.2mm , whereas 2
the number of larger vesicles (.0.2 mm ) increased dramatically after KM treatment (Fig. 6). The average size
2 of the vesicles in KM-treated hair cells (0.56 mm ) was significantly larger than that in control hair cells (0.14
2
Fig. 2. (A) The sum of AcPase-positive areas in a hair cell at various mm , t(68, 76)52950.5; P,0.01). The largest vesicle in
2 2
times after onset of KM treatment. Each time point indicates 12 h or 27 h KM-treated hair cells was 5.2mm , compared to 0.5 mm after a single injection of KM (400 mg / kg) or 120 h after the first
in control hair cells.
injection of 5 day doses of KM (524 h after the last injection). The AcPase activity appears to accumulate in hair cells over time. **: Significantly different from the control group, P,0.01. Also shown are the average and standard error of the total area of a transverse hair cell
section. There is no significant difference in the total area between any 4. Discussion two of the groups. (B) The size of largest AcPase-positive vesicle in a
hair cell. AcPase-positive vesicles become larger after repeated KM
Lysosomes in untreated control hair cells were seen as
doses. **: Significantly different from the 27 h group, P,0.01.
Fig. 3. AcPase histochemistry in hair cells at 27 h after a single injection of KM. AcPase activity is localized in a smooth vesicle located in the apical region of hair cells (big arrows). Dispersed staining (small arrows) is also seen in the hair cells. Cp: cuticular plate, St: stereocilia. Bar: 1mm.
Fig. 4. AcPase histochemistry (A, B) and immunogold labeling of KM (C) in hair cells at 120 h after the first injection of 5 day doses of KM. in KM-treated hair cells. Intense AcPase activity staining is filled in multiple vesicles (A, B). The shape, size and location of the vesicles coincide with vesicles accumulating gold particles (C). Bars: 1mm (A, B); 0.5mm (C).
Fig. 5. In vitro staining of lysosomes in control hair cells (A) or hair cells obtained from animals that were given 5 day KM injections (B). Each picture was constructed from a 15mm Z-series with 1mm step sizes. A dramatic increase in the LysoTracker Red accumulation is observed in hair cells treated with KM (B). Counterstained with BODIPY-phalloidin (green). Cp: cuticular plate, St: stereocilia. Bar: 10mm.
the cell death process, which may accompany lysosomal lysosomal markers (e.g. LysoTracker probe) can be used to disintegration. estimate the amount of internalized aminoglycosides in In summary, this study presents evidence that lysosomes hair cells. Further investigation is required to determine become larger in size in synchrony with the intracellular whether lysosomal rupture occurs as a consequence of accumulation of KM in hair cells. These findings support overaccumulation of aminoglycosides and whether the hypothesis that lysosomal targeting and accumulation lysosomal rupture triggers the process of subsequent hair of aminoglycosides are key cellular events that are in- cell degeneration.
volved in the aminoglycoside-induced progressive hair cell degeneration [26]. The capacity of lysosomes to hold an
extensive amount of aminoglycosides can explain, at least Acknowledgements partially, why aminoglycoside ototoxicity appears several
days to weeks after cellular uptake of the drug. In addition, This study was supported by National Institutes of the close correlation between increases in lysosomal sizes Health grants R01 DC03900 (E.H.) and R01 DC01685 and increases in intracellular KM levels suggests that (R.J.S). The authors would like to thank Drs. Mark Warchol and Jeffrey Corwin for their guidance at an initial stage of in vitro experiments.
References
[1] D. Beauchamp, P. Gourde, M.G. Bergeron, Subcellular distribution of gentamicin in proximal tubular cells, determined by immunogold labeling, Antimicrob. Agents Chemother. 35 (1991) 2173–2179. [2] D. Beauchamp, P. Gourde, M. Simard, M.G. Bergeron, Subcellular
localization of tobramycin and vanomycin given alone and in combination in proximal tubular cells, determined by immunogold labeling, Antimicrob. Agents Chemother. 36 (1992) 2204–2210. [3] T. Barka, Electron histochemical localization of acid phosphatase
activity in the small intestine of mouse, J. Histochem. Cytochem. 12 (1964) 229–238.
[4] S.J. Bird, J.B. Lloyd, Mechanisms of lysosomal rupture by di-peptides, Cell Biochem. Funct. 13 (1995) 79–83.
[6] P.J. Coopman, D.M. Thomas, K.R. Gehlsen, S.C. Mueller, Integrin [17] R.T. Hudson, R.K. Draper, Interaction of coatmer with
amino-a3b1 participates in the phagocytosis of extracellular matrix mole- glycoside antibiotics: Evidence that coatmer has at least two dilysine cules by human breast cancer cells, Mol. Biol. Cell 7 (1996) binding sites, Mol. Biol. Cell 8 (1997) 1901–1910.
1789–1804. [18] T. Kido, G.M. Cohen, Acid phosphatase activity in the chick’s inner [7] J.C.M.J. de Groot, F. Meeuwsen, W.E. Ruizendaal, J.E. Veldman, ear, Acta Oto-Laryngologica 116 (1996) 540–545.
Ultrastructural localization of gentamicin in the cochlea, Hear. Res. [19] L. Laszlo, J. Lowe, T. Self, N. Kenward, M. Landon, T. McBride, 50 (1990) 35–42. C. Farquhar, I. McConell, J. Brown, J. Hope et al., Lysosomes as [8] D.M. Ford, R.H. Dahl, C.A. Lamp, B.A. Moliyoris, Apically and key organelles in the pathogenesis of prion encephalcpathies, J.
basolaterally internalized aminoglycosides colocalize in LL-PK1 Pathol. 166 (1992) 333–341.
lysosomes and alter cell function, Am. J. Physiol. 266 (1994) [20] P.R. Lewis, D.P. Knight (Eds.), Cytochemical Staining Methods for
C52–C57. Electron Microscopy, Elsevier, Amsterdam, 1992.
[9] K.M. Hager, M.A. Pierce, D.R. Moore, E.M. Tytler, J.D. Esko, S.L. [21] W. Li, X. Yuan, G. Nordgren, H. Dalen, G.M. Dubowchik, R.A. Hajduk, Endocytosis of a cytotoxic human high density lipoprotein Firestone, U.T. Brunk, Induction of cell death by lysosomortropic results in disruption of acidic intracellular vesicles and subsequent detergent MSDH, FEBS Lett. 470 (2000) 35–39.
killing of African trypanosomes, J. Cell Biol. 126 (1994) 155–167. [22] S.K. Moestrup, S. Cui, H. Vorum, C. Bregengard, S.E. Bjorn, K. [10] T. Haller, P. Dietl, P. Deetjen, H. Volkl, The lysosomal compart- Norris, J. Gliemann, E.I. Christensen, Evidence that epithelial ments as intracellular calcium store in MDCK cells: a possible glycoprotein 330 / megalin mediates uptake of polybasic drugs, J.
21
involvement in InsP3-mediated Ca release, Cell Calcium 19 Clin. Invest. 96 (1995) 1404–1413.
(1996) 157–165. [23] S. Ohkuma, T. Takano, ATP-dependent lysis of isolated lysosomes [11] E. Hashino, M. Shero, Endocytosis of aminoglycoside antibiotics in by basic substances and acidic ionophores, Cell Struct. Funct. 22
sensory hair cells, Brain Res. 704 (1995) 135–140. (1997) 253–268.
[12] E. Hashino, R.J. Salvi, Regenerated hair cells exhibit a transient [24] G.P. Richardson, A. Forge, C.J. Kros, J. Fleming, S.D.M. Brown, resistance to aminoglycoside toxicity, Brain Res. 720 (1996) 172– K.P. Steel, Myosin VIIA is required for aminoglycoside
accumula-182. tion in cochlear hair cells, J. Neurosci. 17 (1997) 9506–9519.
[13] E. Hashino, M. Shero, R.J. Salvi, Lysosomal targeting and accumu- [25] C. Seiler, T. Nicolson, Defective calmodulin-dependent rapid apical lation of aminoglycoside antibiotics in sensory hair cells, Brain Res. endocytosis in zebrafish hair cell mutants, J. Neurbiol. 41 (1999)
777 (1997) 75–85. 424–434.
[14] E. Hashino, M. Shero, R.J. Salvi, In vivo and in vitro assessment of [26] E. Wang, M.D. Lee, K.W. Dunn, Lysosomal accumulation of drugs lysosomal activity during kanamycin uptake in hair cells, Abstr. in drug-sensitive MES-SA but not multidrug-resistant MES-SA / Dx5 Assoc. Res. Otolaryngol. 21 (1998) 60. uterine sarcoma cells, J. Cell. Physiol. 184 (2000) 263–274. [15] H. Hiel, J.-P. Erre, C. Aurousseau, R. Bouali, D. Dulon, J.-M. Aran, [27] S.E. Williams, H.P. Zenner, J. Schacht, Three molecular steps of
Gentamicin uptake by cochlear hair cell precedes hearing impair- aminoglycoside ototoxicity demonstrated in outer hair cells, Hear. ment during chronic treatment, Audiology 32 (1993) 78–87. Res. 30 (1987) 11–18.
[16] R. Hori, M. Okuda, Y. Ohishi, M. Yasuda, M. Takano, Surface [28] J.L. Zheng, W.Q. Gao, Concanavalin A protects hair cells against binding and intracellular uptake of gentamicin in the cultured kidney gentamicin ototoxicity in rat cochlear explant culture, J. Neurobiol. epithelial cell line (LLC-PK ), J. Pharmacol. Exp. Ther. 261 (1992)1 39 (1999) 29–40.
