- Email: [email protected]
Transforming growth factor alpha treatment alters intracellular calcium levels in hair cells and protects them from ototoxic damage in vitro
~ Pergamon
Int. J. Devl Neuroscience, Vol. 15, No. 4/5, pp. 553 562, 1997
Copyright ~7. 1997 1SDN. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0736~5748/97 $17.00+0.00 P l h S0736-5748(96)00110-4
TRANSFORMING GROWTH FACTOR ALPHA TREATMENT ALTERS INTRACELLULAR CALCIUM LEVELS IN HAIR CELLS AND PROTECTS THEM FROM OTOTOXIC DAMAGE IN VITRO
HINRICH STAECKER,* t STEFAN DAZERT,~§ BRIGITTE MALGRANGE,t PHILIPPE P. LEFEBVRE,~" ALLEN F. RYAN~ and THOMAS R. VAN DE WATER*t¶ ~ *Department of Otolaryngology, Albert Einstein College of Medicine, Bronx, NY, U.S.A.; tDepartment of Human Physiology and Pathophysiology, University of Li6ge, Li6ge, Belgium; :~Department of Otolaryngology, University of California, San Diego, CA, U.S.A.; §Department of Otolaryngology, University of W~irzburg, Wiirzburg, Germany; ¶Department of Neuroscience, Albert Einstein College of Medicine, Bronx, NY, U.S.A.
Abstract--To determine if transforming growth factor alpha (TGFct) 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 TGFct (500ng/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 TGFct in organ of Corti explants. © 1997 ISDN Key words: organ of Corti explant; aminoglycoside ototoxicity; growth factor protection; hair cells; calcium
levels.
Gram negative sepsis is a life threatening condition that is frequently treated with aminoglycoside antibiotics. Unfortunately, the usefulness of this class of antibiotics is limited by the development of nephrotoxicity and ototoxicity? j Despite many years of research and a clear understanding of the bacteriostatic effect of the aminoglycosides, the exact mechanism of their ototoxicity is not known. Recent studies suggest that the acute and chronic effects of aminoglycosides are mediated by different mechanisms. 31 Acute effects are thought to be related to the ability of aminoglycosides to bind to phosphatidyl 4,5-biphosphate, whereas chronic toxicity may be related to the production of (a) toxic metabolite(s). TM A number of studies has sought to counter the toxic effects of aminoglycosides. Calcium has been shown to reverse the neuromuscular blockade produced by aminoglycosides, and has been shown to prevent the loss of cochlear microphonics in guinea pigs treated with aminoglycosides. 34 Excess calcium has also been shown to prevent aminoglycoside induced toxicity in vitro 27 (Malgrange, personal communication). Recently, several groups have reported the use of growth factors that are important in wound healing to prevent neomycin induced hair cell toxicity in postnatal rat organ of Corti explants. ~°'19'2°To understand the effect of transforming growth factor alpha (TGF~) on auditory hair cells, we have used calcium green dye to determine relative calcium levels in TGFc~ pretreated 3-day-old (P-3) rat organ of Corti explants and subsequently to determine the effect of exposure to an aminoglycoside (i.e. neomycin) on these growth factor protected and on unprotected control explants. The epithelial growth factor receptor (EGFR) is the receptor for both TGFe and epithelial growth factor (EGF) and functions through induction of a wide variety of cell signalling pathways 2'15,1834
~To whom all correspondence should be addressed at: 1410 Pelham Parkway South, Kennedy Center Room 302, Bronx, NY 10461, U.S.A. Abbreviations: TGFct, transforming growth factor alpha; Ca, z+ calcium; EGF, epidermal growth factor; EGFR, epidermal growth factor receptor; DAG, di-acetyl glycerol. 553
554
H. Staecker et al.
including tyrosine kinase activity.26'32'33 Activation of the EGFR results in increased intracellular calcium (Ca 2+) levels.7'24 In addition, some studies have linked the activity of EGFR to several oncogenes that have been associated with prevention of apoptosis. 6 EXPERIMENTAL PROCEDURES
Mater&& The following materials were used: Dulbecco's minimal essential medium, and Dulbecco's phosphate buffered saline (GIBCO, Grand Island, NY); Triton-X-100, and phalloidin-fluorescein (FITC) conjugate (Sigma, St. Louis, MO); Cytotak® (Collaborative Biomedical Products, Bedford, MA); TGF~ (R and D Systems, Minneapolis, MN); Vectashield® (Vector Labs, Burlingane, CA); antiEGFR antibody (UBI, Lake Placid, NY); cholera toxin j~ subunit-rhodamine (TRITC) conjugate (List Biologicals Labs, Campbell, CA); calcium green dye (Molecular Probes, Eugene, OR). Cultures P-3 rat pups were euthanized. Their cochleae were quickly removed and placed in Dulbecco's phosphate buffered saline (PBS). Corti's organ and the adjacent spiral ganglion were removed as a unit from individual cochlea as previously described 19 and placed in Dulbecco's minimal essential medium supplemented with a N1 supplement 1 and glucose to 6g/1 (DMEM +N1 +G). Explants were divided into a control and a TGF~ (500ng/ml) pretreatment group. After 48hr in vitro, neomycin (i.e. 5 x 10 -4 M , 10-3M or 10 .2 M) was added to both the control and TGFe pretreated explants for a second 48 hr culture period. TGFc~ (500 ng/ml) supplementation of the growth factor pretreated explants was continued during the 48 hr period of ototoxin-exposure. All explants were incubated at 37°C and media were changed after every 24 hr. Immunochemistry P-3 cochleae were fixed in methacarn. 8 After paraffin embedding, 7 #m sections were cut and mounted on poly-l-lysine coated slides. Immunostaining for EGFR was carried out with an antiEGFR sheep polyclonal antibody (UBI; against recombinant EGFR fusion protein). The immunohistochemical procedure was our standardized protocol for paraffin sections. 8
FITC-phalloidin staining After culture, all explants were fixed in PBS buffered 4% paraformaldehyde for 1 hr. Fixed explants were washed in PBS, incubated in PBS + 10% fetal calf serum + 0.3% Triton-X-100 for 30 rain followed by a brief rinse in PBS, and subsequently incubated at 25°C in PBS + phalloidinFITC conjugate (3#g/ml) for 45rain. The explants were washed with PBS 3x for 10min and mounted on glass slides with Vectashield® mounting medium. Analysis of stereocilia bundles integrity was carried out with a confocal microscope. Hair cell survival in treated and untreated explants was determined by creating cochleograms of the entire organ of Corti explant by measuring the percentage of the explant bearing four rows of hair cells with normal appearing stereocilia bundles with each explant sampled three times. Damaged or fused hair cells were not included because counts were too subject to bias.
Calcium imaging After 24 hr & vitro, cholera toxin fl subunit-TR1TC conjugate was added to the medium at a concentration of 0.1 pg/ml. This technique specifically labeled neuronal elements and allowed for localization of the auditory hair cells of Corti's organ by identifying the area of termination of the peripheral axons in the living explant. A unit area, often a length of hair cells containing all four rows of these cells (i.e. 40 hair cells per unit area), was defined with the confocal microscope for sequential sampling with the caveat that the area sampled also contained portions of other cell types in the area (e.g. supporting cells). The volume of the area sampled was kept constant for all explants. The changes in intracellular Ca 2+ levels in response to growth factor pretreatment and exposure to neomycin occurred predominantly in the hair cells of the area sampled with the confocal microscope. At the end of the 48 hr culture period, calcium green dye was added at a concentration of 10 mM.
TGF~ alters hair cell C a 2+ and protects from ototoxin
555
Calcium green was used to determine intracellular Ca 2+ levels because of its resistance to bleaching and adaptability to confocal microscopic analysis. The samples were protected from light and incubated at room temperature for 1 hr with calcium green, then placed on glass coverslips that had been pretreated with C y t o t a k ® and washed twice with 100/d of D M E M + N 1 + G. After placing the specimens on the microscope stage the medium was withdrawn and replaced with D M E M + N 1 + G + n e o m y c i n at a concentration of either 5 x 10 -4, 10 -3, or 10-2 M. Fluorescence of calcium green was visualized with an M R C 600 Confocal microscope at an excitation wavelength of 488 nm (argon-krypton laser). Scans were collected as a time series at 25 sec intervals with a 40 x objective lens using a Kallman scanning pattern. All scans for hair cell Ca -,+ levels were made on the mid-turn region of the explants. Specimens were imaged as pairs (i.e. control and TGFc~ pretreated) to insure that exposure time to calcium green dye was identical for both groups. Data was collected as fluorescent intensity per unit area at 25 sec intervals (x-axis) for a total sampling period of 500 sec. Fluorescence per unit area values for each time point (AF) were divided by baseline fluorescence values established at 0 sec. (F0) for each condition to determine the fluorescence ratio (Fr) measurement for each time point of each condition, i.e. AF/Fo = Ft. The establishment of a ratio value (Fr) based on variation from the baseline allows for a comparison between specimen groups. An unpaired Student's t-test was used to make statistical comparisons between two groups with small means, and A N O V A was used to determine statistical significance when more than two groups were compared.
RESULTS Immunostainin 9
A n t i - E G F R staining of P-3 rat temporal bones was localized over the inner and outer hair cells of the organ of Corti but not to the surrounding supporting cells or the auditory neurons (see Fig. 1).
Fig. 1. A 3-day-old (P-3) rat temporal bone immunostained (fluorescein label) for EGFR. A dense accumulation of anti-EGFR-FITC stain localizedover the inner (open arrow) and outer (arrows) hair cells of Corti's organ with almost no immunostainingof either the supporting cells (arrowheads) or the auditory neurons (now shown in this confocal image). The arrows point to the apical portions of the hair cells. Observations were made from five different specimens, N = 5, Bar = 10/~m.
556
H. Staecker et al.
Phalloidin-FITC staining of stereocilia bundles
The protective effect of TGF~ on auditory hair cells in the organ of Corti explants is clearly shown in Figs 2 and 3. Pretreatment of explants with TGF~ resulted in the preservation of both inner and outer hair cells in all but the most basal segment of the organ of Corti explants. Phalloidin FITC staining clearly demonstrated the presence of normal stereocilia bundle morphology in both control (Fig. 2a) and TGF~ protected, ototoxin-exposed explants (Fig. 2c). The use of a low concentration of neomycin (i.e. 5 x 10 -4 M) resulted in incomplete loss of hair cell stereocilia bundles (i.e. a 60% loss) and in the highest level of hair cells with normal stereocilia bundle morphology (i.e. 60% normal stereocilia bundles) as a consequence of TGF~ pretreatment (Fig. 3). Exposure of control explants to neomycin (10 -2 M) resulted in a loss of almost all phalloidinFITC staining of stereocilia bundles (Fig. 2b). Preservation of auditory hair cell stereocilia bundle morphology was quantified and is presented in Fig. 3. After pretreatment with TGF~ (500 ng/ml), exposure to 10 -3 M neomycin resulted in preservation of ca 50% of the auditory hair cell stereocilia bundles, whereas the control, ototoxin-exposed cultures showed less than a 5% preservation rate of hair cell stereocilia bundles (Fig. 3). Increasing the neomycin concentration ten-fold (i.e. 10 -2 M) resulted in complete loss of normal hair cell stereocilia bundles and in a 9% survival rate of normal stereocilia bundle morphology in the TGF~ pretreated explants (Fig. 3). Calcium imaging
Pretreatment of organ of Corti explants with TGF~ (500 ng/ml) for 48 hr resulted in a significant increase in the baseline intracellular calcium levels of hair cells with their fluorescence per unit area measuring five-fold higher than those of control explants cultured without the growth factor (Fig. 4). The levels of fluorescence remained constant during the 500 sec observation period in the TGF~ pretreated explants (Fig. 4), demonstrating that calcium levels were stable and that there was no degradation, bleaching or loss of the calcium green dye from the hair cells. Addition of neomycin (10 --3 M) to control explants resulted in an increase in the intracellular Ca 2+ levels of hair cells that began at 50 sec post-neomycin exposure and increased to its peak level by 150 sec post-exposure (Fig. 5). Subsequently, C a 2+ levels of the hair cells exposed to neomycin return to a baseline level
Fig. 2a.
T G F 7 alters hair cell Ca 2+ a n d protects from ototoxin
Fig. 2. P-3 rat organ of Corti explants, after 4 days in vitro. Phalloidin-FlTC staining of stereocilia bundles of midturn segments of organ of Corti explants: (a) a control explant showing a normal pattern of staining for inner (open arrow) and outer (arrow) hair cell stereocilia bundles; (b) a control explant exposed to 10 3M neomycin for 48 hr showing the complete destruction of hair cell stereocilia bundles with only epithelial scar tissue present (arrowheads); (c) a TGF~ (500 ng/ml) pretreated (48 hr) explant exposed to 10 3M neomycin+TGF~ (500ng/ml) for 48hr which shows good preservation of stereocilia bundle morphology of both inner (open arrow) and outer (arrow) hair cells. Bar in (c) equals 10 #m for micrographs in (a~z).
557
H. Staecker et al.
558 100
-
90.--I .,.J
uJ 0 n, < "r" "r I-
z < L9 n¢ 0 03 Iel,' 0 0
8070605040302010-
m 0
,
CONTROL
5xlO 4
-]10 3
0 10 2
~ 5xlO"4
1 10 3
1(~2
neomycin (M)
neomycin (M)
lr UNTREATED
T G F a (5(30 ng/ml)
/
Fig. 3. The protective effect of TGF~ on auditory hair cells is represented as a percentage of the organ of Corti explant length covered with normal hair cell stereocilia bundles. Protection of auditory hair cells from aminoglycoside toxicity by TGFc¢ (500 ng/ml) pretreatment was evident for all concentrations of neomycin tested, i. e. 5 × 10 -4, 10 -3, and 10-2 M. The most dramatic effect of TGF~ pretreatment on the protection of auditory hair cell stereocilia bundles from ototoxic damage was seen at a neomycin concentration of 10-3 M. (Each determination represents values derived from ten specimens, N= 10.) Error bars represent standard deviation from the mean (_+S.D.).
by 250 sec post-exposure in the control explants. Calcium levels began to rise again at 300 sec and plotted at 350sec p o s t - n e o m y c i n exposure (Fig. 5). A d d i t i o n o f n e o m y c i n ( 1 0 - 3 M ) to T G F ~ pretreated explants had a dramatic effect. Within 100 sec o f n e o m y c i n exposure, hair cell Ca 2+ levels had decreased to approximately one half that o f their baseline Ca 2+ value. Hair cell Ca 2+ levels reached a new steady state by 150 sec post-neomycin exposure and remained stable at this lower Ca 2+ level for the remainder o f the observation period (Fig. 5). Overall Ca 2+ levels in the T G F ~ protected, neomycin exposed (10 -3 M) explants remained at a level that was 60% less that o f the hair cell C a 2+ levels observed in the unexposed, T G F ~ pretreated explants (Fig. 5). DISCUSSION
Effects of aminoglycoside on Ca 2+ levels Aminoglycosides have a complex, but p o o r l y u n d e r s t o o d effect on auditory hair cells. Schacht and colleagues 3'3°'31 have divided the development o f ototoxicity into acute and chronic phases o f toxicity based on clinically observed effects. Acute treatment with aminoglycosides has been shown to directly effect the cochlear m i c r o p h o n i c in guinea pigs. 34 The mechanism o f this acute effect is not known, but aminoglycosides appear to block the hair cell transduction channel. 13'~4 Acute ototoxicity m a y be related to the acute effects o f aminoglycosides in other systems. A m o n g the other well-documented acute effects o f aminoglycoside toxicity is neuromuscular blockade, which can be reversed t h r o u g h treatment with c a l c i u m ? 9 However, it is not clear whether similar mechanisms are active in cochlea. Acute effects o f aminoglycosides on synaptic neurotransmission have also been
TGFc~ alters hair cell Ca2+ and protects from ototoxin
559
CONTROL
TGFa
0
I
I
[
I
I
2000
4000
6000
8000
10000
12000
FLUORESCENCE PER UNIT AREA Fig. 4. P-3 rat organ of Corti explants, after 48 hr in vitro. Fluorescence intensity levels for calcium green in the hair cells of TGFc~ (500 ng/ml) pretreated and control explants at 1 hr post dye addition, representing the 0 sec. (baseline) observation point, N = 10. Pretreatment of explants with T G F e resulted in five-fold higher intracellular Ca 2+ levels in hair cells when compared to the Ca 2+ values of control explant hair cells. Both of these hair cell Ca 2+ levels were stable in the explants, as shown in Fig. 5.
observed. 529 Intracellularly, aminoglycosides have well-documented inhibitory effects on phospholipase C, 11~2 which, through activation of the phosphoinositol cascade, can produce major changes in intracellular calcium. Studies on artificial membranes have demonstrated that aminoglycosides can also have direct effects on the permeability of cell membranes] 5'37The initial spike in hair cell calcium seen after neomycin exposure (i.e. at 50 sec) may be explained by such acute effects of neomycin on the cell membrane. 35'37 The subsequent second rise in hair cell intracellular calcium may reflect perturbation of intracellular second messengers. 3~ Chronic ototoxicity and destruction of auditory hair cells may function through a somewhat different pathway. As Schacht has pointed out, whereas the acute toxicity of aminoglycosides affects many tissues and cell types, chronic aminoglycoside toxicity selectively targets the kidney and the inner ear. 3~ He and his colleagues have presented extensive evidence that ototoxicity involves phosphoinositol metabolism. Phosphoinositol-4,5-biphosphate (PIP2) has been shown to be an important phospholipid in many cell signalling pathways.42j'25 Studies have shown that aminoglycosides to bind specifically to the PIP2 of hair cell membranes and that there is a strong correlation between the degree of this binding and the degree of ototoxic damage caused by individual aminoglycosides.3~'37 In addition to their effects on hair cell phospholipid metabolism, Schacht has documented that aminoglycosides compromise essential cellular repair mechanisms by inhibiting ornithine decarboxylase.3°3~ Interestingly, chronic ototoxicity appears to require prior modification of aminoglycosides into toxic metabolites. TM There are numerous ways in which activation of the EGFR by TGF~ might influence both intracellular calcium and ototoxicity. Binding of TGF~ to the EGFR initiates receptor dimerization with both autophosphorylation and activation of tyrosine kinase activity in the intracellular signalling domains. 17This in turn leads to phosphorylation of intracellular signalling molecules, many of which contain src homology 2 (SH2) domains. One important such molecule is phospholipase C,
560
H. Staecker et al. .5
-
-
1.41.31.2
01.1 ~_I.0 nl 0.9
\ ~
Z 0.8 LU 0.7 09 UJ 0 . 6 n"
-O-- CONTROL •-[5]--,'~ -V . . .
J\,7-
oOoO,
~_\
NEOMYCIN TGFoc TGFa + NEO . .
0.2 0.1 0.0
--~1
I
]
I
I
1
I
I
7-
I
I
I
I
1
I
I
I
I
I
I
I
I
100 200 300 400 500 Fig. 5. P-3 rat organ of Corti explants, after 48 hr in vitro followed by 1 hr of incubation with calcium green dye. Fluorescence intensity of hair cells was recorded at 25 sec intervals for a total of 500 sec (x-axis). Fluorescence intensity is expressed as a fluorescence ratio (F,) (y-axis) that was derived by dividing individual time point fluorescence values (AF) by the baseline fluorescence value (i.e. the initial 0 sec value, F0). Neomycin (10-3M) was added to the explants at the 0sec time point. Both control and TGFc~ pretreated explants had stable hair cell Ca2+ levels for the entire period of observation. Control and neomycin exposed explants had an early (i.e. 50-250 sec) and a later (i.e. 300-500 sec) increase in hair cell Ca2+ levels. TGFc~pretreated, ototoxin exposed explants had an almost immediate decline (i.e. after 25 sec) in hair cell Ca2+ levels that stabilized at 150 sec post-exposure and remained depressed for the rest of the observation period. Each time point represents measurements from ten specimens (N= 10). Error bars represent standard deviation from the mean (+ S.D.).
which cleaves PIP2 into inositol triphosphate (IP3) and diacylglycerol ( D A G ) . 4 Inositol triphosphate can directly open calcium channels or mobilize intracellular calcium stores, whereas D A G can activate protein kinase C and initiate the p r o d u c t i o n o f arachidonic acid. 3° Other second messenger systems can also be recruited by E G F R activation.15'18,22'24'26'36These signalling cascades can directly affect the function o f calcium channels, is'21'32 Activation o f E G F R has been shown to increase intracellular calcium levels significantly. 7 In this system, there are two gradations o f effect. A rapid influx o f extracellular calcium effected by IP3 causes a fast increase o f intracellular calcium, but p r o d u c t i o n o f D A G and arachidonic acid leads to a slower and more stable increase in calcium levels. 4'21 This i n f o r m a t i o n m a y explain our observation that pretreatment o f P-3 rat organ o f Corti explants with TGFc~ results in a very large (i.e. five-fold) increase in the baseline intracellular calcium levels o f hair cells (Fig. 4). Protective effects o f TGFc~ A d d i t i o n o f n e o m y c i n to cultures pretreated with T G F ~ resulted in protection o f auditory hair cells, as gauged by the integrity o f stereocilia bundle m o r p h o l o g y observed in this study (Fig. 2 and Fig. 3) and as previously reported. 19'2°The rapid decrease in hair cell calcium levels observed in the T G F ~ protected cultures after n e o m y c i n exposure (i.e. beginning 25 sec post-ototoxin exposure) again suggests an acute effect o f the aminoglycoside molecule on either a cell m e m b r a n e c o m p o n e n t (e.g. PIP2) or a cell m e m b r a n e calcium channel, z3'28 However, it should be noted that even after n e o m y c i n exposure, calcium levels in these TGFc~ pretreated explants remained elevated c o m p a r e d to the control explants and did n o t show a secondary change in hair cell calcium levels during the 300-500 sec post-exposure period as observed in the untreated, neomycin-exposed explants (Figs 4 and 5). Because previous studies have linked calcium with otoprotection f r o m aminoglycoside
TGFe alters hair cell Ca2+ and protects from ototoxin
561
damage both in vitro, 27 and in vivo, 34 the possibility of a direct effect of elevating intracellular calcium levels in hair cells by TGFe pretreatment (Fig. 4) should be considered as a potential mechanism for the otoprotective effect of this growth factor. It is important to remember that although hair cell calcium levels rise in response to ototoxin in control explants and fall in ototoxin exposed, TGFc~ pretreated cultures, the intracellular calcium levels are still 65% higher in the hair cells of the TGFct explants because these explants have such a high baseline level of calcium at the onset of ototoxin exposure (see Figs 4 and 5). Alternatively, both EGF and TGFa have been linked to the activation of oncogenes that have been shown to rescue cells from apoptosis. These oncogenes may also play a role in rescuing hair cells from aminoglycoside induced damage. 6 It is also well known that growth factors such as TGFc~ can exert a generalized trophic effect on many types of cells by activation of the ras-MAP kinase pathway via its tyrosine kinase receptor (i.e. EGFR)? 2'33 Future studies will examine the effect of TGFa on the production of free radical scavengers, 9'~6 as well as on cellular signal pathways that control apoptosis 6 and general trophic action. 3233 Hopefully, identifying the underlying biochemical and molecular pathophysiology of hair cell injury, protection, and repair will lead to increasingly effective ways to treat and prevent aminoglycoside induced hearing loss. Acknowledgements--kThis work was supported by grants from the: N1DCD (DC00088) and Shulsky Hearing Research Fund of the Montefiore Medical Center to TRV; NIDCD (DC00139) and Research Service of the Veterans Administration to AFR; Deutsche Forschungsgemeinshaft (DA344/1-2) to SD; and National Fund for Scientific Research of Belgium to BM and PPL. The authors thank Rose Imperati for word processing this paper.
REFERENCES 1. Bottenstein J. and Sato G. (1979) Growth of a rat neuroblastoma cell line in serum free medium. Proc. Natl Sci. Acad. U.S.A. 76, 514-517. 2. Clapham D. E. (1994) Direct G protein activation of ion channels? A. Rev. Neurosci, 17, 441-461. 3. Crann S., Huang M., McLaren J. D. and Schacht J. (1992) Formation of a toxic metabolite from gentamicin by a hepatic cytosolic fraction. Biochem. Pharmacol. 43, 1835-1839. 4. Exton J. (1990) Signalling through phosphatidylcholine breakdown. J. Biol. Chem. 265, 1-4. 5. Frank C., Longo R. and Sagratella S. (1994) Pertussis toxin prevents neomycin-induced calcium-dependent eletrophysiological effects in rat hippocampal slices. Gen. Pharmac. 25, 957 961. 6. Franklin C. and Kraft A. (1992) Protein kinase C-independent activation ofc-jun and c-fos transcription by epidermal growth factor. Biochem. Biophys. Acta 1134, 137-142. 7. Fearn J. and King A. (1985) EGF receptor affinity is regulated by intracellular calcium and protein kinase C. Cell 40, 991-1000. 8. Galinovic-Schwartz V., Peng D., Chiu F. C. and Van De Water T. R. (1991) Temporal pattern of innervation in the developing mouse embryo: An immunohistochemical study of a 66kD subunit of mammalian neurofilament. J. Neurosci. Res. 30, 124-130. 9. Garetz S. L., Rhee D. J. and Schacht J. (1994) Sulfhydryl compounds and antioxidants inhibit cytotoxicity to outer hair cells o f a gentamicin metabolite in vitro. Hearing Res. 77, 75-80. 10. Gray E. M. G., Warchol M. E. and Corwin J. T. (1996) IGF-1 protects hair cells from aminoglycoside-induced apoptotic cell death. Assoc. Res. Otolaryngol. 19th Midwinter Meetin 9, 4-8 February 1996, St. Petersburg, FL, Abstract No. 792, p. 198. 11. Hatmi M., Randon J., Faili A. and Vargaftig B. B. (1993) Carrageenan-induced activation of human platelets is dependent on the phospholipase C pathway. Br. J. Haernatol. 83, 270-275. 12. Hedberg K. M., Bengtsson T., Safiejko-Mroczka B., Bell P. B. and Lindroth M. (1993) PDGF and neomycin induce similar changes in the actin cytoskeleton in human fibroblasts. Cell Motil. Cvtoskel. 24, 139-149. 13. Jaramillo F. and Hudspeth A. J. (1991) Localization of the hair cell's transduction channels at the hair bundle's top by ionophoretic application of a channel blocker. Neuron 7, 409-420. 14. Kimitsuki T. and Ohmori H. (1993) Dihydrostreptomycin modifies adaptation and blocks the mechano-electric transducer in chick cochlear hair cells. Brain Res. 624, 143-150. 15. Klein S., Fisher G., Jensen T., Mendelson J., Voorhees J. and Elder T. (1992) Regulation o f T G F J, alpha expression in human keratinocytes: PKC dependent and independent pathways. J. Cell Physiol. 151,326-336. 16. Kopke R., Liu W., Garcia P., Feghali J., Malgrange B., Gabaizadeh R., Spray D., Staecker H. and Van De Water T. R. (1996) Protection of auditory hair cells from cisplatin induced damage by N-acetyl-cysteine in an in vitro model. Assoc. Res. Otolaryngol. 19th Midwinter Meeting, 4-8 February 1996, St. Petersburg, FL, Abstract No. 818, p. 205. 17. Luetteke N. and Lee D. (1990) Transforming growth factor alpha: Expression, regulation and biological action of its integral membrane precursor. Seminars Ca z+ Biol. 1, 265-275. 18. Liu P., Wen M., Sun L. and Hayash J. (1993) Activation of phospholipase A2 and stimulation of prostaglandin E2 production by transforming growth factor alpha in rat thymic epithelial cells requires influx of calcium. Biochem. J. 293, 109 113. 19. Low W., Dazert S., Baird A. and Ryan A. F. (1996) Basic fibroblast growth factor (FGF-2) protects rat cochlear hair cells in organotypical culture from aminoglycoside injury, J. Cell Physiol. 167, 443-450.
562
H. Staecker et al.
20. Malgrange B., Lefebvre P. P., Baron H., Liu W., Staecker H., Van De Water T. R. and Moonen G. (1994) Transforming growth factor alpha (TGFc0 acts with retinoic acid to protect auditory hair cells from aminoglycoside induce ototoxicity. Assoc. Res. Otolaryngol. 17th Midwinter Meeting, 6--10 February 1994, St. Petersburg, FL, Abstract No. 551, p. 138. 21. Mohammadi M., Djonne C., Li W., Spivak T., Honneger A., Jaye M. and Schlessinger J. (1992) Point mutation in FGF receptor eliminates PI hydrolysis without affecting mitogenesis. Nature 358, 681-684. 22. Nishizuka Y. (1984) The role of protein kinase C in cell surface signal transduction and tumor promotion. Nature 308, 693~598. 23. Parsons T. D., Obaid A. L. and Salzberg B. M. (1992) Aminoglycoside antibiotics block voltage-dependent calcium channels in intact vertebrate nerve terminals. J. Gen. Physiol. 99, 491 504. 24. Peppelenbosch M., Tertoolen L. and de Laat S. (1991) Epidermal growth factor-activated calcium and potassium channels. J. Biol. Chem. 266, 1938-1944. 25. Peters K. G., Marie J., Wilson E., Ives E., Escobedo J., Delrosario M., Mirda D. and Williams L. T. (1992) Point mutation of an FGF receptor abolishes phosphatidylinositol turnover and Ca 2÷ influx but not mitogenesis. Nature 358, 678 684. 26. Reynolds N., Talmar H., Baldassari J., Henderson P., Voorhees J. and Fisher G. (1993) Differential induction of phosphatidylcholine hydrolysis, diacylglycerol formation and protein kinase C activation by EGF and TGF alpha on normal skin fibroblasts and keratinocytes. Biochem. J. 294, 535 544. 27. Richardson G. P. and Russell I. J. (1991) Cochlear cultures as a model system for studying aminoglycoside induced ototoxicity. Hearing Res. 53, 293 311. 28. Robles I., Barrios M. and Baeyens J. M. (1992) Centrally administered aminoglycoside antibiotics antagonize naloxoneprecipitated withdrawal in mice acutely dependent on morphine. Neurosci. Lett. 145, 189-192. 29. Said A., Matsuk N. and Kasuya Y. (1995) Effects of aminoglycoside antibiotics on cholinergic autonomic nervous transmission. Pharmacol. Toxicol. 76, 128--132. 30. Schacht J. (1979) Molecular mechanisms of drug induced hearing loss. Arch. O tolaryng. 224, 129-134. 31. Schacht J. (1993) Biochemical basis of aminoglycoside ototxicity. Otolaryngol. Clinics North Am. 26, 845-856. 32. Schlessinger J. (1995) Cellular signalling by receptor tyrosine kinases. The Harvey Lectures Series 89, 105-123. 33. Schlessinger J. and Ullrich A. (1992) Growth factor signaling by receptor tyrosine kinases. Neuron 9, 383-391. 34. Takada A. and Schacht J. (1982) Calcium antagonism and reversibility of gentamicin induced loss of cochlear microphonics in the guinea pig. Hearing Res. 8, 279-286. 35. Van Bambeke F., Mingiot-Leclercg M., Schank A., Brasseur R. and Tulkens P. (1993) Alterations in membrane permeability induced by aminoglycoside antibiotics: Studies on liposomes and cultured cells. Eur. J. Pharmacol. 247, 155 168. 36. Wahl M. and Carpenter G. (1988) Regulation of epidermal growth factor stimulated formation of inositol phosphates in A-431 cells by calcium and protein kinase C. J. Biochem. 263, 7581-7590. 37. Wang B., Weiner N. and Takada A. (1984) Characterization of amonoglycoside-lipid interactions and development of a refined model for ototoxicity testing. Biochem. Pharmacol. 33, 3257-3262.
Int. J. Devl Neuroscience, Vol. 15, No. 4/5, pp. 553 562, 1997
Copyright ~7. 1997 1SDN. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0736~5748/97 $17.00+0.00 P l h S0736-5748(96)00110-4
TRANSFORMING GROWTH FACTOR ALPHA TREATMENT ALTERS INTRACELLULAR CALCIUM LEVELS IN HAIR CELLS AND PROTECTS THEM FROM OTOTOXIC DAMAGE IN VITRO
HINRICH STAECKER,* t STEFAN DAZERT,~§ BRIGITTE MALGRANGE,t PHILIPPE P. LEFEBVRE,~" ALLEN F. RYAN~ and THOMAS R. VAN DE WATER*t¶ ~ *Department of Otolaryngology, Albert Einstein College of Medicine, Bronx, NY, U.S.A.; tDepartment of Human Physiology and Pathophysiology, University of Li6ge, Li6ge, Belgium; :~Department of Otolaryngology, University of California, San Diego, CA, U.S.A.; §Department of Otolaryngology, University of W~irzburg, Wiirzburg, Germany; ¶Department of Neuroscience, Albert Einstein College of Medicine, Bronx, NY, U.S.A.
Abstract--To determine if transforming growth factor alpha (TGFct) 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 TGFct (500ng/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 TGFct in organ of Corti explants. © 1997 ISDN Key words: organ of Corti explant; aminoglycoside ototoxicity; growth factor protection; hair cells; calcium
levels.
Gram negative sepsis is a life threatening condition that is frequently treated with aminoglycoside antibiotics. Unfortunately, the usefulness of this class of antibiotics is limited by the development of nephrotoxicity and ototoxicity? j Despite many years of research and a clear understanding of the bacteriostatic effect of the aminoglycosides, the exact mechanism of their ototoxicity is not known. Recent studies suggest that the acute and chronic effects of aminoglycosides are mediated by different mechanisms. 31 Acute effects are thought to be related to the ability of aminoglycosides to bind to phosphatidyl 4,5-biphosphate, whereas chronic toxicity may be related to the production of (a) toxic metabolite(s). TM A number of studies has sought to counter the toxic effects of aminoglycosides. Calcium has been shown to reverse the neuromuscular blockade produced by aminoglycosides, and has been shown to prevent the loss of cochlear microphonics in guinea pigs treated with aminoglycosides. 34 Excess calcium has also been shown to prevent aminoglycoside induced toxicity in vitro 27 (Malgrange, personal communication). Recently, several groups have reported the use of growth factors that are important in wound healing to prevent neomycin induced hair cell toxicity in postnatal rat organ of Corti explants. ~°'19'2°To understand the effect of transforming growth factor alpha (TGF~) on auditory hair cells, we have used calcium green dye to determine relative calcium levels in TGFc~ pretreated 3-day-old (P-3) rat organ of Corti explants and subsequently to determine the effect of exposure to an aminoglycoside (i.e. neomycin) on these growth factor protected and on unprotected control explants. The epithelial growth factor receptor (EGFR) is the receptor for both TGFe and epithelial growth factor (EGF) and functions through induction of a wide variety of cell signalling pathways 2'15,1834
~To whom all correspondence should be addressed at: 1410 Pelham Parkway South, Kennedy Center Room 302, Bronx, NY 10461, U.S.A. Abbreviations: TGFct, transforming growth factor alpha; Ca, z+ calcium; EGF, epidermal growth factor; EGFR, epidermal growth factor receptor; DAG, di-acetyl glycerol. 553
554
H. Staecker et al.
including tyrosine kinase activity.26'32'33 Activation of the EGFR results in increased intracellular calcium (Ca 2+) levels.7'24 In addition, some studies have linked the activity of EGFR to several oncogenes that have been associated with prevention of apoptosis. 6 EXPERIMENTAL PROCEDURES
Mater&& The following materials were used: Dulbecco's minimal essential medium, and Dulbecco's phosphate buffered saline (GIBCO, Grand Island, NY); Triton-X-100, and phalloidin-fluorescein (FITC) conjugate (Sigma, St. Louis, MO); Cytotak® (Collaborative Biomedical Products, Bedford, MA); TGF~ (R and D Systems, Minneapolis, MN); Vectashield® (Vector Labs, Burlingane, CA); antiEGFR antibody (UBI, Lake Placid, NY); cholera toxin j~ subunit-rhodamine (TRITC) conjugate (List Biologicals Labs, Campbell, CA); calcium green dye (Molecular Probes, Eugene, OR). Cultures P-3 rat pups were euthanized. Their cochleae were quickly removed and placed in Dulbecco's phosphate buffered saline (PBS). Corti's organ and the adjacent spiral ganglion were removed as a unit from individual cochlea as previously described 19 and placed in Dulbecco's minimal essential medium supplemented with a N1 supplement 1 and glucose to 6g/1 (DMEM +N1 +G). Explants were divided into a control and a TGF~ (500ng/ml) pretreatment group. After 48hr in vitro, neomycin (i.e. 5 x 10 -4 M , 10-3M or 10 .2 M) was added to both the control and TGFe pretreated explants for a second 48 hr culture period. TGFc~ (500 ng/ml) supplementation of the growth factor pretreated explants was continued during the 48 hr period of ototoxin-exposure. All explants were incubated at 37°C and media were changed after every 24 hr. Immunochemistry P-3 cochleae were fixed in methacarn. 8 After paraffin embedding, 7 #m sections were cut and mounted on poly-l-lysine coated slides. Immunostaining for EGFR was carried out with an antiEGFR sheep polyclonal antibody (UBI; against recombinant EGFR fusion protein). The immunohistochemical procedure was our standardized protocol for paraffin sections. 8
FITC-phalloidin staining After culture, all explants were fixed in PBS buffered 4% paraformaldehyde for 1 hr. Fixed explants were washed in PBS, incubated in PBS + 10% fetal calf serum + 0.3% Triton-X-100 for 30 rain followed by a brief rinse in PBS, and subsequently incubated at 25°C in PBS + phalloidinFITC conjugate (3#g/ml) for 45rain. The explants were washed with PBS 3x for 10min and mounted on glass slides with Vectashield® mounting medium. Analysis of stereocilia bundles integrity was carried out with a confocal microscope. Hair cell survival in treated and untreated explants was determined by creating cochleograms of the entire organ of Corti explant by measuring the percentage of the explant bearing four rows of hair cells with normal appearing stereocilia bundles with each explant sampled three times. Damaged or fused hair cells were not included because counts were too subject to bias.
Calcium imaging After 24 hr & vitro, cholera toxin fl subunit-TR1TC conjugate was added to the medium at a concentration of 0.1 pg/ml. This technique specifically labeled neuronal elements and allowed for localization of the auditory hair cells of Corti's organ by identifying the area of termination of the peripheral axons in the living explant. A unit area, often a length of hair cells containing all four rows of these cells (i.e. 40 hair cells per unit area), was defined with the confocal microscope for sequential sampling with the caveat that the area sampled also contained portions of other cell types in the area (e.g. supporting cells). The volume of the area sampled was kept constant for all explants. The changes in intracellular Ca 2+ levels in response to growth factor pretreatment and exposure to neomycin occurred predominantly in the hair cells of the area sampled with the confocal microscope. At the end of the 48 hr culture period, calcium green dye was added at a concentration of 10 mM.
TGF~ alters hair cell C a 2+ and protects from ototoxin
555
Calcium green was used to determine intracellular Ca 2+ levels because of its resistance to bleaching and adaptability to confocal microscopic analysis. The samples were protected from light and incubated at room temperature for 1 hr with calcium green, then placed on glass coverslips that had been pretreated with C y t o t a k ® and washed twice with 100/d of D M E M + N 1 + G. After placing the specimens on the microscope stage the medium was withdrawn and replaced with D M E M + N 1 + G + n e o m y c i n at a concentration of either 5 x 10 -4, 10 -3, or 10-2 M. Fluorescence of calcium green was visualized with an M R C 600 Confocal microscope at an excitation wavelength of 488 nm (argon-krypton laser). Scans were collected as a time series at 25 sec intervals with a 40 x objective lens using a Kallman scanning pattern. All scans for hair cell Ca -,+ levels were made on the mid-turn region of the explants. Specimens were imaged as pairs (i.e. control and TGFc~ pretreated) to insure that exposure time to calcium green dye was identical for both groups. Data was collected as fluorescent intensity per unit area at 25 sec intervals (x-axis) for a total sampling period of 500 sec. Fluorescence per unit area values for each time point (AF) were divided by baseline fluorescence values established at 0 sec. (F0) for each condition to determine the fluorescence ratio (Fr) measurement for each time point of each condition, i.e. AF/Fo = Ft. The establishment of a ratio value (Fr) based on variation from the baseline allows for a comparison between specimen groups. An unpaired Student's t-test was used to make statistical comparisons between two groups with small means, and A N O V A was used to determine statistical significance when more than two groups were compared.
RESULTS Immunostainin 9
A n t i - E G F R staining of P-3 rat temporal bones was localized over the inner and outer hair cells of the organ of Corti but not to the surrounding supporting cells or the auditory neurons (see Fig. 1).
Fig. 1. A 3-day-old (P-3) rat temporal bone immunostained (fluorescein label) for EGFR. A dense accumulation of anti-EGFR-FITC stain localizedover the inner (open arrow) and outer (arrows) hair cells of Corti's organ with almost no immunostainingof either the supporting cells (arrowheads) or the auditory neurons (now shown in this confocal image). The arrows point to the apical portions of the hair cells. Observations were made from five different specimens, N = 5, Bar = 10/~m.
556
H. Staecker et al.
Phalloidin-FITC staining of stereocilia bundles
The protective effect of TGF~ on auditory hair cells in the organ of Corti explants is clearly shown in Figs 2 and 3. Pretreatment of explants with TGF~ resulted in the preservation of both inner and outer hair cells in all but the most basal segment of the organ of Corti explants. Phalloidin FITC staining clearly demonstrated the presence of normal stereocilia bundle morphology in both control (Fig. 2a) and TGF~ protected, ototoxin-exposed explants (Fig. 2c). The use of a low concentration of neomycin (i.e. 5 x 10 -4 M) resulted in incomplete loss of hair cell stereocilia bundles (i.e. a 60% loss) and in the highest level of hair cells with normal stereocilia bundle morphology (i.e. 60% normal stereocilia bundles) as a consequence of TGF~ pretreatment (Fig. 3). Exposure of control explants to neomycin (10 -2 M) resulted in a loss of almost all phalloidinFITC staining of stereocilia bundles (Fig. 2b). Preservation of auditory hair cell stereocilia bundle morphology was quantified and is presented in Fig. 3. After pretreatment with TGF~ (500 ng/ml), exposure to 10 -3 M neomycin resulted in preservation of ca 50% of the auditory hair cell stereocilia bundles, whereas the control, ototoxin-exposed cultures showed less than a 5% preservation rate of hair cell stereocilia bundles (Fig. 3). Increasing the neomycin concentration ten-fold (i.e. 10 -2 M) resulted in complete loss of normal hair cell stereocilia bundles and in a 9% survival rate of normal stereocilia bundle morphology in the TGF~ pretreated explants (Fig. 3). Calcium imaging
Pretreatment of organ of Corti explants with TGF~ (500 ng/ml) for 48 hr resulted in a significant increase in the baseline intracellular calcium levels of hair cells with their fluorescence per unit area measuring five-fold higher than those of control explants cultured without the growth factor (Fig. 4). The levels of fluorescence remained constant during the 500 sec observation period in the TGF~ pretreated explants (Fig. 4), demonstrating that calcium levels were stable and that there was no degradation, bleaching or loss of the calcium green dye from the hair cells. Addition of neomycin (10 --3 M) to control explants resulted in an increase in the intracellular Ca 2+ levels of hair cells that began at 50 sec post-neomycin exposure and increased to its peak level by 150 sec post-exposure (Fig. 5). Subsequently, C a 2+ levels of the hair cells exposed to neomycin return to a baseline level
Fig. 2a.
T G F 7 alters hair cell Ca 2+ a n d protects from ototoxin
Fig. 2. P-3 rat organ of Corti explants, after 4 days in vitro. Phalloidin-FlTC staining of stereocilia bundles of midturn segments of organ of Corti explants: (a) a control explant showing a normal pattern of staining for inner (open arrow) and outer (arrow) hair cell stereocilia bundles; (b) a control explant exposed to 10 3M neomycin for 48 hr showing the complete destruction of hair cell stereocilia bundles with only epithelial scar tissue present (arrowheads); (c) a TGF~ (500 ng/ml) pretreated (48 hr) explant exposed to 10 3M neomycin+TGF~ (500ng/ml) for 48hr which shows good preservation of stereocilia bundle morphology of both inner (open arrow) and outer (arrow) hair cells. Bar in (c) equals 10 #m for micrographs in (a~z).
557
H. Staecker et al.
558 100
-
90.--I .,.J
uJ 0 n, < "r" "r I-
z < L9 n¢ 0 03 Iel,' 0 0
8070605040302010-
m 0
,
CONTROL
5xlO 4
-]10 3
0 10 2
~ 5xlO"4
1 10 3
1(~2
neomycin (M)
neomycin (M)
lr UNTREATED
T G F a (5(30 ng/ml)
/
Fig. 3. The protective effect of TGF~ on auditory hair cells is represented as a percentage of the organ of Corti explant length covered with normal hair cell stereocilia bundles. Protection of auditory hair cells from aminoglycoside toxicity by TGFc¢ (500 ng/ml) pretreatment was evident for all concentrations of neomycin tested, i. e. 5 × 10 -4, 10 -3, and 10-2 M. The most dramatic effect of TGF~ pretreatment on the protection of auditory hair cell stereocilia bundles from ototoxic damage was seen at a neomycin concentration of 10-3 M. (Each determination represents values derived from ten specimens, N= 10.) Error bars represent standard deviation from the mean (_+S.D.).
by 250 sec post-exposure in the control explants. Calcium levels began to rise again at 300 sec and plotted at 350sec p o s t - n e o m y c i n exposure (Fig. 5). A d d i t i o n o f n e o m y c i n ( 1 0 - 3 M ) to T G F ~ pretreated explants had a dramatic effect. Within 100 sec o f n e o m y c i n exposure, hair cell Ca 2+ levels had decreased to approximately one half that o f their baseline Ca 2+ value. Hair cell Ca 2+ levels reached a new steady state by 150 sec post-neomycin exposure and remained stable at this lower Ca 2+ level for the remainder o f the observation period (Fig. 5). Overall Ca 2+ levels in the T G F ~ protected, neomycin exposed (10 -3 M) explants remained at a level that was 60% less that o f the hair cell C a 2+ levels observed in the unexposed, T G F ~ pretreated explants (Fig. 5). DISCUSSION
Effects of aminoglycoside on Ca 2+ levels Aminoglycosides have a complex, but p o o r l y u n d e r s t o o d effect on auditory hair cells. Schacht and colleagues 3'3°'31 have divided the development o f ototoxicity into acute and chronic phases o f toxicity based on clinically observed effects. Acute treatment with aminoglycosides has been shown to directly effect the cochlear m i c r o p h o n i c in guinea pigs. 34 The mechanism o f this acute effect is not known, but aminoglycosides appear to block the hair cell transduction channel. 13'~4 Acute ototoxicity m a y be related to the acute effects o f aminoglycosides in other systems. A m o n g the other well-documented acute effects o f aminoglycoside toxicity is neuromuscular blockade, which can be reversed t h r o u g h treatment with c a l c i u m ? 9 However, it is not clear whether similar mechanisms are active in cochlea. Acute effects o f aminoglycosides on synaptic neurotransmission have also been
TGFc~ alters hair cell Ca2+ and protects from ototoxin
559
CONTROL
TGFa
0
I
I
[
I
I
2000
4000
6000
8000
10000
12000
FLUORESCENCE PER UNIT AREA Fig. 4. P-3 rat organ of Corti explants, after 48 hr in vitro. Fluorescence intensity levels for calcium green in the hair cells of TGFc~ (500 ng/ml) pretreated and control explants at 1 hr post dye addition, representing the 0 sec. (baseline) observation point, N = 10. Pretreatment of explants with T G F e resulted in five-fold higher intracellular Ca 2+ levels in hair cells when compared to the Ca 2+ values of control explant hair cells. Both of these hair cell Ca 2+ levels were stable in the explants, as shown in Fig. 5.
observed. 529 Intracellularly, aminoglycosides have well-documented inhibitory effects on phospholipase C, 11~2 which, through activation of the phosphoinositol cascade, can produce major changes in intracellular calcium. Studies on artificial membranes have demonstrated that aminoglycosides can also have direct effects on the permeability of cell membranes] 5'37The initial spike in hair cell calcium seen after neomycin exposure (i.e. at 50 sec) may be explained by such acute effects of neomycin on the cell membrane. 35'37 The subsequent second rise in hair cell intracellular calcium may reflect perturbation of intracellular second messengers. 3~ Chronic ototoxicity and destruction of auditory hair cells may function through a somewhat different pathway. As Schacht has pointed out, whereas the acute toxicity of aminoglycosides affects many tissues and cell types, chronic aminoglycoside toxicity selectively targets the kidney and the inner ear. 3~ He and his colleagues have presented extensive evidence that ototoxicity involves phosphoinositol metabolism. Phosphoinositol-4,5-biphosphate (PIP2) has been shown to be an important phospholipid in many cell signalling pathways.42j'25 Studies have shown that aminoglycosides to bind specifically to the PIP2 of hair cell membranes and that there is a strong correlation between the degree of this binding and the degree of ototoxic damage caused by individual aminoglycosides.3~'37 In addition to their effects on hair cell phospholipid metabolism, Schacht has documented that aminoglycosides compromise essential cellular repair mechanisms by inhibiting ornithine decarboxylase.3°3~ Interestingly, chronic ototoxicity appears to require prior modification of aminoglycosides into toxic metabolites. TM There are numerous ways in which activation of the EGFR by TGF~ might influence both intracellular calcium and ototoxicity. Binding of TGF~ to the EGFR initiates receptor dimerization with both autophosphorylation and activation of tyrosine kinase activity in the intracellular signalling domains. 17This in turn leads to phosphorylation of intracellular signalling molecules, many of which contain src homology 2 (SH2) domains. One important such molecule is phospholipase C,
560
H. Staecker et al. .5
-
-
1.41.31.2
01.1 ~_I.0 nl 0.9
\ ~
Z 0.8 LU 0.7 09 UJ 0 . 6 n"
-O-- CONTROL •-[5]--,'~ -V . . .
J\,7-
oOoO,
~_\
NEOMYCIN TGFoc TGFa + NEO . .
0.2 0.1 0.0
--~1
I
]
I
I
1
I
I
7-
I
I
I
I
1
I
I
I
I
I
I
I
I
100 200 300 400 500 Fig. 5. P-3 rat organ of Corti explants, after 48 hr in vitro followed by 1 hr of incubation with calcium green dye. Fluorescence intensity of hair cells was recorded at 25 sec intervals for a total of 500 sec (x-axis). Fluorescence intensity is expressed as a fluorescence ratio (F,) (y-axis) that was derived by dividing individual time point fluorescence values (AF) by the baseline fluorescence value (i.e. the initial 0 sec value, F0). Neomycin (10-3M) was added to the explants at the 0sec time point. Both control and TGFc~ pretreated explants had stable hair cell Ca2+ levels for the entire period of observation. Control and neomycin exposed explants had an early (i.e. 50-250 sec) and a later (i.e. 300-500 sec) increase in hair cell Ca2+ levels. TGFc~pretreated, ototoxin exposed explants had an almost immediate decline (i.e. after 25 sec) in hair cell Ca2+ levels that stabilized at 150 sec post-exposure and remained depressed for the rest of the observation period. Each time point represents measurements from ten specimens (N= 10). Error bars represent standard deviation from the mean (+ S.D.).
which cleaves PIP2 into inositol triphosphate (IP3) and diacylglycerol ( D A G ) . 4 Inositol triphosphate can directly open calcium channels or mobilize intracellular calcium stores, whereas D A G can activate protein kinase C and initiate the p r o d u c t i o n o f arachidonic acid. 3° Other second messenger systems can also be recruited by E G F R activation.15'18,22'24'26'36These signalling cascades can directly affect the function o f calcium channels, is'21'32 Activation o f E G F R has been shown to increase intracellular calcium levels significantly. 7 In this system, there are two gradations o f effect. A rapid influx o f extracellular calcium effected by IP3 causes a fast increase o f intracellular calcium, but p r o d u c t i o n o f D A G and arachidonic acid leads to a slower and more stable increase in calcium levels. 4'21 This i n f o r m a t i o n m a y explain our observation that pretreatment o f P-3 rat organ o f Corti explants with TGFc~ results in a very large (i.e. five-fold) increase in the baseline intracellular calcium levels o f hair cells (Fig. 4). Protective effects o f TGFc~ A d d i t i o n o f n e o m y c i n to cultures pretreated with T G F ~ resulted in protection o f auditory hair cells, as gauged by the integrity o f stereocilia bundle m o r p h o l o g y observed in this study (Fig. 2 and Fig. 3) and as previously reported. 19'2°The rapid decrease in hair cell calcium levels observed in the T G F ~ protected cultures after n e o m y c i n exposure (i.e. beginning 25 sec post-ototoxin exposure) again suggests an acute effect o f the aminoglycoside molecule on either a cell m e m b r a n e c o m p o n e n t (e.g. PIP2) or a cell m e m b r a n e calcium channel, z3'28 However, it should be noted that even after n e o m y c i n exposure, calcium levels in these TGFc~ pretreated explants remained elevated c o m p a r e d to the control explants and did n o t show a secondary change in hair cell calcium levels during the 300-500 sec post-exposure period as observed in the untreated, neomycin-exposed explants (Figs 4 and 5). Because previous studies have linked calcium with otoprotection f r o m aminoglycoside
TGFe alters hair cell Ca2+ and protects from ototoxin
561
damage both in vitro, 27 and in vivo, 34 the possibility of a direct effect of elevating intracellular calcium levels in hair cells by TGFe pretreatment (Fig. 4) should be considered as a potential mechanism for the otoprotective effect of this growth factor. It is important to remember that although hair cell calcium levels rise in response to ototoxin in control explants and fall in ototoxin exposed, TGFc~ pretreated cultures, the intracellular calcium levels are still 65% higher in the hair cells of the TGFct explants because these explants have such a high baseline level of calcium at the onset of ototoxin exposure (see Figs 4 and 5). Alternatively, both EGF and TGFa have been linked to the activation of oncogenes that have been shown to rescue cells from apoptosis. These oncogenes may also play a role in rescuing hair cells from aminoglycoside induced damage. 6 It is also well known that growth factors such as TGFc~ can exert a generalized trophic effect on many types of cells by activation of the ras-MAP kinase pathway via its tyrosine kinase receptor (i.e. EGFR)? 2'33 Future studies will examine the effect of TGFa on the production of free radical scavengers, 9'~6 as well as on cellular signal pathways that control apoptosis 6 and general trophic action. 3233 Hopefully, identifying the underlying biochemical and molecular pathophysiology of hair cell injury, protection, and repair will lead to increasingly effective ways to treat and prevent aminoglycoside induced hearing loss. Acknowledgements--kThis work was supported by grants from the: N1DCD (DC00088) and Shulsky Hearing Research Fund of the Montefiore Medical Center to TRV; NIDCD (DC00139) and Research Service of the Veterans Administration to AFR; Deutsche Forschungsgemeinshaft (DA344/1-2) to SD; and National Fund for Scientific Research of Belgium to BM and PPL. The authors thank Rose Imperati for word processing this paper.
REFERENCES 1. Bottenstein J. and Sato G. (1979) Growth of a rat neuroblastoma cell line in serum free medium. Proc. Natl Sci. Acad. U.S.A. 76, 514-517. 2. Clapham D. E. (1994) Direct G protein activation of ion channels? A. Rev. Neurosci, 17, 441-461. 3. Crann S., Huang M., McLaren J. D. and Schacht J. (1992) Formation of a toxic metabolite from gentamicin by a hepatic cytosolic fraction. Biochem. Pharmacol. 43, 1835-1839. 4. Exton J. (1990) Signalling through phosphatidylcholine breakdown. J. Biol. Chem. 265, 1-4. 5. Frank C., Longo R. and Sagratella S. (1994) Pertussis toxin prevents neomycin-induced calcium-dependent eletrophysiological effects in rat hippocampal slices. Gen. Pharmac. 25, 957 961. 6. Franklin C. and Kraft A. (1992) Protein kinase C-independent activation ofc-jun and c-fos transcription by epidermal growth factor. Biochem. Biophys. Acta 1134, 137-142. 7. Fearn J. and King A. (1985) EGF receptor affinity is regulated by intracellular calcium and protein kinase C. Cell 40, 991-1000. 8. Galinovic-Schwartz V., Peng D., Chiu F. C. and Van De Water T. R. (1991) Temporal pattern of innervation in the developing mouse embryo: An immunohistochemical study of a 66kD subunit of mammalian neurofilament. J. Neurosci. Res. 30, 124-130. 9. Garetz S. L., Rhee D. J. and Schacht J. (1994) Sulfhydryl compounds and antioxidants inhibit cytotoxicity to outer hair cells o f a gentamicin metabolite in vitro. Hearing Res. 77, 75-80. 10. Gray E. M. G., Warchol M. E. and Corwin J. T. (1996) IGF-1 protects hair cells from aminoglycoside-induced apoptotic cell death. Assoc. Res. Otolaryngol. 19th Midwinter Meetin 9, 4-8 February 1996, St. Petersburg, FL, Abstract No. 792, p. 198. 11. Hatmi M., Randon J., Faili A. and Vargaftig B. B. (1993) Carrageenan-induced activation of human platelets is dependent on the phospholipase C pathway. Br. J. Haernatol. 83, 270-275. 12. Hedberg K. M., Bengtsson T., Safiejko-Mroczka B., Bell P. B. and Lindroth M. (1993) PDGF and neomycin induce similar changes in the actin cytoskeleton in human fibroblasts. Cell Motil. Cvtoskel. 24, 139-149. 13. Jaramillo F. and Hudspeth A. J. (1991) Localization of the hair cell's transduction channels at the hair bundle's top by ionophoretic application of a channel blocker. Neuron 7, 409-420. 14. Kimitsuki T. and Ohmori H. (1993) Dihydrostreptomycin modifies adaptation and blocks the mechano-electric transducer in chick cochlear hair cells. Brain Res. 624, 143-150. 15. Klein S., Fisher G., Jensen T., Mendelson J., Voorhees J. and Elder T. (1992) Regulation o f T G F J, alpha expression in human keratinocytes: PKC dependent and independent pathways. J. Cell Physiol. 151,326-336. 16. Kopke R., Liu W., Garcia P., Feghali J., Malgrange B., Gabaizadeh R., Spray D., Staecker H. and Van De Water T. R. (1996) Protection of auditory hair cells from cisplatin induced damage by N-acetyl-cysteine in an in vitro model. Assoc. Res. Otolaryngol. 19th Midwinter Meeting, 4-8 February 1996, St. Petersburg, FL, Abstract No. 818, p. 205. 17. Luetteke N. and Lee D. (1990) Transforming growth factor alpha: Expression, regulation and biological action of its integral membrane precursor. Seminars Ca z+ Biol. 1, 265-275. 18. Liu P., Wen M., Sun L. and Hayash J. (1993) Activation of phospholipase A2 and stimulation of prostaglandin E2 production by transforming growth factor alpha in rat thymic epithelial cells requires influx of calcium. Biochem. J. 293, 109 113. 19. Low W., Dazert S., Baird A. and Ryan A. F. (1996) Basic fibroblast growth factor (FGF-2) protects rat cochlear hair cells in organotypical culture from aminoglycoside injury, J. Cell Physiol. 167, 443-450.
562
H. Staecker et al.
20. Malgrange B., Lefebvre P. P., Baron H., Liu W., Staecker H., Van De Water T. R. and Moonen G. (1994) Transforming growth factor alpha (TGFc0 acts with retinoic acid to protect auditory hair cells from aminoglycoside induce ototoxicity. Assoc. Res. Otolaryngol. 17th Midwinter Meeting, 6--10 February 1994, St. Petersburg, FL, Abstract No. 551, p. 138. 21. Mohammadi M., Djonne C., Li W., Spivak T., Honneger A., Jaye M. and Schlessinger J. (1992) Point mutation in FGF receptor eliminates PI hydrolysis without affecting mitogenesis. Nature 358, 681-684. 22. Nishizuka Y. (1984) The role of protein kinase C in cell surface signal transduction and tumor promotion. Nature 308, 693~598. 23. Parsons T. D., Obaid A. L. and Salzberg B. M. (1992) Aminoglycoside antibiotics block voltage-dependent calcium channels in intact vertebrate nerve terminals. J. Gen. Physiol. 99, 491 504. 24. Peppelenbosch M., Tertoolen L. and de Laat S. (1991) Epidermal growth factor-activated calcium and potassium channels. J. Biol. Chem. 266, 1938-1944. 25. Peters K. G., Marie J., Wilson E., Ives E., Escobedo J., Delrosario M., Mirda D. and Williams L. T. (1992) Point mutation of an FGF receptor abolishes phosphatidylinositol turnover and Ca 2÷ influx but not mitogenesis. Nature 358, 678 684. 26. Reynolds N., Talmar H., Baldassari J., Henderson P., Voorhees J. and Fisher G. (1993) Differential induction of phosphatidylcholine hydrolysis, diacylglycerol formation and protein kinase C activation by EGF and TGF alpha on normal skin fibroblasts and keratinocytes. Biochem. J. 294, 535 544. 27. Richardson G. P. and Russell I. J. (1991) Cochlear cultures as a model system for studying aminoglycoside induced ototoxicity. Hearing Res. 53, 293 311. 28. Robles I., Barrios M. and Baeyens J. M. (1992) Centrally administered aminoglycoside antibiotics antagonize naloxoneprecipitated withdrawal in mice acutely dependent on morphine. Neurosci. Lett. 145, 189-192. 29. Said A., Matsuk N. and Kasuya Y. (1995) Effects of aminoglycoside antibiotics on cholinergic autonomic nervous transmission. Pharmacol. Toxicol. 76, 128--132. 30. Schacht J. (1979) Molecular mechanisms of drug induced hearing loss. Arch. O tolaryng. 224, 129-134. 31. Schacht J. (1993) Biochemical basis of aminoglycoside ototxicity. Otolaryngol. Clinics North Am. 26, 845-856. 32. Schlessinger J. (1995) Cellular signalling by receptor tyrosine kinases. The Harvey Lectures Series 89, 105-123. 33. Schlessinger J. and Ullrich A. (1992) Growth factor signaling by receptor tyrosine kinases. Neuron 9, 383-391. 34. Takada A. and Schacht J. (1982) Calcium antagonism and reversibility of gentamicin induced loss of cochlear microphonics in the guinea pig. Hearing Res. 8, 279-286. 35. Van Bambeke F., Mingiot-Leclercg M., Schank A., Brasseur R. and Tulkens P. (1993) Alterations in membrane permeability induced by aminoglycoside antibiotics: Studies on liposomes and cultured cells. Eur. J. Pharmacol. 247, 155 168. 36. Wahl M. and Carpenter G. (1988) Regulation of epidermal growth factor stimulated formation of inositol phosphates in A-431 cells by calcium and protein kinase C. J. Biochem. 263, 7581-7590. 37. Wang B., Weiner N. and Takada A. (1984) Characterization of amonoglycoside-lipid interactions and development of a refined model for ototoxicity testing. Biochem. Pharmacol. 33, 3257-3262.