The role of glutathione in carboplatin ototoxicity was investigated in the chinchilla. Chinchillas hearing was tested with both distortion product otoacoustic ...
The role of glutathione in carboplatin ototoxicity in the chinchilla D. Henderson, B.H. Hu, S.L. McFadden, X.Y. Zheng and D. Ding Center jol' Hearing and Deafness, State University of New York at Buffa lo, Buffalo, USA
The rol e of glutathione in carboplatin ototoxicity was investigated in the chinchilla. Chinchillas hearing was tested with both distortion product otoacoustic emissions (DPOAE) and evoked potentials recorded from a chronic electrode in the inferior coHiculus (IC). All subjects had an osmotic pump fitted to their right ear and it received buthionine sulfoximine (BSO) at a dose of 15 mM delivered at 5 ml per hour for 14 days. A group (N=4) was given a double dose of carboplatin (25 mg/kg i.p, for 2 days). The pump was implanted three days before the carboplatin dose. The BSO treated ears showed a greater loss in both ,evoked potential and DPOAE measures, as well as substantially fewer missinghair cells. The results implicate ' reactive oxygen species' ,tROS) as a common factor in ototoxic reactions becau se suppression of glutathione antioxidant leads to greater ototoxic reactions. Keywords : Ca rbop lation ototoxi cit y, distortion product e m iss ions, glutathi one, chin c hilla
introduction Ototoxicity: Ca rbopla tin is a che m o therapeutic
mea sured by the di stortion product o toa co ustic emission (DP OAE) (Fi gure 2) .
drug oft en used as an a ltern ative to cisp latin because its side effec ts, nephrot oxicity, nause a and he aring loss are typically less severe and less prevalent than treatment with cisp latin . When hearing loss does occur w ith carbop latin, the loss is very similar to the effects of cisp latin, i.e., the loss starts at the highest frequencies o f the audiogra m an d coc hlea r lesi on s begin wi th outer hair cell s (OHC) in the base of the cochlea. H owever, carboplatin in tlie chincliil la produces a strang e and unique lesion that in vol ves a selective loss of inner hair cells (IHC). In studies of the carbo p la tin do se/response in the chinchill a, Hofstetter e t al. (1 997 a and b) showed that up to 70 percent of all IHC could be missing (Figure I ) wi thou t any o bserva ble
The literature on carboplatin is not as we ll devel oped but severallaboratorics (Kopke et a I., 1997 and Ryb ack, 1999) ha ve s hown tha t the ototoxic effects of cisplatin can be av o ide d or dramatically de creased by the ad m in is tra tion of an anti oxidant drug. Kopke e t al. ( 1997) in a systema tic study of the ototoxicity of cispl atin in cochlear organ cultures found that hair cell durna ge could be dram atically redu ced i f [he culture was treated wi th certain ant ioxidant drugs. Ryback and Somani (19 99) using Wistar r ats , treated them with c is p la tin and 4 meth ylthiobenzoic acid (MTBA), diethyl dithiocarbamate (DDTC) , alpha- lipoi c ac id or ebselen ; all the drugs were given LP. and the
change in either OHC morphologyor functionas antioxidant treated rats developed significantly N oise & Health 2000,3 ;9, 1-1 0
Selective inner hair cell damage
Normal control "r' "
Figure I. Micrographs of chinchilla cochlea stained with succinia dehydrogenase (SDH). Top: 4 weeks after single dose of 100 mg/kg carbopla tin, Bottom: 4 weeks after double dose of 100 mg/kg carboplatin. Note, the almost complete loss of IHC and normal OHC in the middle photo. (Hofstett.er et al., 1997b)
Carboplatin 100mg/kg 2X
less threshold shifts as measured by ABR. Biochemically, the cisplatin treated control rats showed a marked depression of glutathione and a decrease in other antioxidant enzymes along with a marked increase in malandialdehyde. Ryback and Somani (1999) concluded that "these findings suggest that cisplatin ototoxicity is related to lipid peroxidation and ... protective agents prevent hearing loss and lipid pcroxidation by sparing the antioxidant defence system in the cochlea". Returning to the question of carboplatin, it shows a similar molecular structure to cisplatin; it is less ototoxic but the hearing loss it creates is
2
similar in audiograms and pathology to that of cisplatin so it is reasonable to ask whether there are common pathological processes between carboplatin and cisplatin; i.e., suppression of glutathione (U SH ), USH peroxid ase and USH redu ctase, and elevation of superox ide dismutase (SOD), catalase and malandialdehyde activity. 1f depression of gluatathione is a factor 1D carboplatin ototoxicity, then inhibition of glutathione synthesis should lead to additio na l ototoxic reactions . L-buthionine-[S,R]-sulfoximine (BSO) is a specific and irr eversible inhibitor of y glutamylcysteine synthetase (y-GCS) , the rate limiting enzyme in the sy nthe sis of a SH . BSO
1.2 kHz 38 mgiKQ (double dose)
't
AVQ= 5 0.1
l00 f ·...,.
I
'r
" " "1
50 " 0> C
60
~
.tn
.:;: ~
11)
' I' " " "
~
5O~
40 ' ....J 0... (f)
cg
-IHe
· · · -· OHC1,2,3 ~~
?O
lil:5Gl Pre "-"- Post 4 Wk
30
w
..,.
20
10
-c o
a
•••"
Q... - 10
20
40
&0
80
100
":t'. Distance trorn ape x
c
:~~ ~
I It . lI l .... It " " " " " . " ' " " I I , , , ' .. ,, I •.
o 10
i
1
20 30 40 50 60 70 80
Input Level (dB SPL)
2.4 kHz
4.8 kHz
50
:::'~BJ Pre
40
....J
a. (J)
• •••• _-
i
Post 4 Wk
30
1Il 20
-0
-0
W
10
~' 1~~=-!.~._ ·20
·3 0
~,o ,, , 10" , ,20, ' , ,,30,' , ,,40" , ,, 50" ""60" "70" " '''80 "
I
In put Level (dB SPL)
W
:~ f ' ''~;~_~;j' ~~~ ''' '' ''''' '''' '1 ~ '' '~; I ''] 30 20 10
I~
",J!Z\ *' ~f.
.
C§ 0 ~ ;7i ~ - 10 E'~'.~
30
1Il
20
-0
W
< a
a. 0
*
. ""
~
o
10 20 30 40 50 60 70 80
Input Leve l (dB SPL)
9.6 kHz 50~
Pre ........ POSl4 Wk
~:-, wTl
40
a.
(J)
~
~ ~ l:~" "" " " " " " I " "~
7.2 kHz 50
....J
t
. .#.:'.~
p.OS." .Wk. .
~ ;~ r == ~;;;l
4 Wk
: ~~ I
10
~-1~~
0 - 10 ·2 0 ·30
0
10 20 30 40 50 60 70 80
Input Level (dB SPL)
~~~ t,,,,,,,,,,,,,,,,,,,,",,,,,,,,,,,,, o
J
10 20 30 40 50 60 70 80
Input Level (dB SPL)
Fi gure 2. Aver age DPO AE I/O factor of chinchillas in the 38 mg/kg doubl e d o s e group. N o tice that th e
average 1/0 function is normal.
3
has been widely used as a tool to deplete intracellular and extrace llu lar aSH ill v ari ous cells and organs (Lee et al ., 19R7; Lee et a l., 1992 ; Luthen et a1., 1994; Mi stry et aI., 1993, Mitchell et al., 198Y; Mizui et al., 1YY2; Morales et a1., 1994; Pileblad and Magnusson, 1989; Thanislass cr a1. , 1995), includ ing the coc hlea (Hoffma n et al., 1988; Laz enb y et al ., 1988). The purp ose of the present study was to determine if BS O enhances carbo plati n oto tox ici ty in the chinch illa coch lea . Th e severity of carboplatin ototox icity was assessed physiologically by measuring distortion pro duct otoacous tic emiss ions (DPOAEs) and evo ke d potent ials from a inferior colliculus (l CPs), and anatomica lly by the cytocochleograms showi ng !HC and OHC loss. Since previous studies have shown that carboplatin pre fere ntially dam ages the IR Cs in chinchilla, it was of interest to determine if BSO treatment would incr ease the amount of damage to the OH Cs.
ssa
treatment , and the second do se w as days after the firs t. Ch inch illas in the drug control gro up (n=5) only rece ived B SO in the right ear wi thout any ca rboplatin.
aLlUL iu i~lereLl4
ssa was infused into the right cochl ea of each anima l usin g an osmo tic pum p (2ML4, Alza Co rporation) (Bro wn et aI., 1993; Sch indler et al., 1995). The pump was filled wi th 15 mM BSO (Sigma Chemicals) in Hank s balanced sa lt solut ion (GIB Ca ). The pu mp infusion rate was 5 ul/hr and the infusion pcr iod last ed 14 days. Ou r selection of a 15 mM co ncentra tion of BSO was based on previous in vitro stud ies, most of which used BSO concen trations ranging from 0.05 to 5 mM to deplete ce llu lar GSH. We used a conc entrat ion of BSO exceeding the levels used in in vitro studies to all ow for intracochl ear dilu tion of the BSO solution hy the pe rilymph. A l 4-day pump was used to ensure that GSH levels were redu ced ove r the entire time per iod that carboplatin was active .
Materials and Methods Subject..
anrl
"lIrgery
for
implanrlng
recording electrodes. Nine adult chinchillas (450-600 g) with normal heuring served us subj ects. Tungsten record ing electrodes were stereotaxically implanted into the left and right IC and a grou nd electrode was implanted in the rostra l cranium (McFa dde n et a l., 1998). Follow ing surgery, the animals we re allowed to recover for at least two weeks prior to testing. Th e care and use of an imals in this study were approved by the State Un iversity of New York at Buffalo Inst itutional Animal Care and Use Comm ittee. In all subjec ts, BSO was applied to the right ears
and the left ea rs ser ved as controls. The anim als were randomly divid ed int o two group s, ca rbo p lat in-trcatcd and dru g co nt rol. The double-d ose carbopla tin group (n=4) receiv ed BSO in the right ear followed by two doses of carboplatin (2 5 mg/kg i.p. X 2). The first dose was adminis tere d 3 days afte r the beginnin g of 4
Audiometry
Tone burst stimuli were digita lly generated (10 ms duration, 5 ms cos ine rise/fall , constant starting phase, 10 s t irnu l i/s r-at 0.5, 1,2, 4, S a nd 16 kHz as describ ed prev iously (Hu ct al., 1997). On e hundr ed sa mples were average d at eac h level. Thres holds were determi ned by visual inspection of raw waveform s at each frequency. Th reshold was defined as the mid- point betwe en the lowest level at which a clear res ponse was seen and the next lower leve l where 110 response co uld be discerned. Mean thre sholds wer e determined by ave raging thresho ld estima tes for all ani mals in a grou p. IJO fun ctions show mean response amplitudes as a fu nctio n of input level. The am plitude of the evok ed resp onse was mea sured from the first positive peak to the foll ow ing nega ti VI:; tl ull ~l1 . T lu cshu lds dilL! response amplitudes were measured before and 2, 7 and 14 days after the beginning o f BSO treatm ent.
60
----
--
-0
60
--0
40
L..
o
20
'E
'-'"
a
Q,)
"0 ::l ~
a. E «
80
o
40
Normal 2 days 7 days t-toays
O.5k
~ 4k
60
a,
20
a 0
10 20 30 40 50 60 70
o
10 20 30 40 50 60 70
o
10 20 30 40 50 60 70
Stimulus Level (dB SPL)
Figure 3. Mean Iep amplitudes (mV, peak-to-peak) before BSO infusion (filled circles), and after 2 d>lY~
(open drdE'S), 7 OllyS (till(>o rriangles) llnd 14 days (open triangles) of BSO infusion in the dru g
control grou·p. Each panel shows I/O functions for a different frequency (0.5-16 kHz). Inpu t/output (l /O) fun ctions o f the cuhic (2 f,-f2 ) DPOAEs were record ed during the pre sent at ion of tw o primary tones (f and f 2) , with an f2/f l ratio of J .2. The levels of the primaries, L , and
L z, were equa l. DPOAE I/O func tions wer e recorded in 5 dB steps from a to 80 dB 8PL. DPOA E 1/0 func tio ns we re co llected in rand om ord er for f = I, 2, 4, and 8 kHz. Cochlear Histology Fourteen day s after BSO treatment, the animal s we re an aesthetised with so d iu m pentobarbital (50-100 mg/k g) and decapitated . Each bulla w as quickly removed and opened to expose th e cochlea. The round w ind ow and oval wi ndo w were opened and 0.2M so dium suc c inate in 0.1M phosphate buffer (p H 7.4) was slowl y perfused through the round window; then th e coc h lea was
immersed in the same so luti on for I hr at 37 °C. The coc hlea was p ost-fixed with 10% form ald eh yde for 24 Ill'S. Hair cell counts as
reflected in Sll CCtD18 deh ydr ogenas e ac t rvr ty (Figure 1) for example, we re o bta ined over 0. 24 mm interval s along the entire length of the cochlea an d the percentage of m iss ing or IC s and IHCs were plotted as a functio n of per cent distance a lo ng the len gth of the coc h lea as des cribed p reviously (H ofstetter et al., 1997 a and b; Hu et al., 1997).
Results Effects of Chronic BSO Infusion Data from the four animals that recei ved 13S0 (15 mM ) for 14 days w ithout carboplatin were used to ass ess the effects of BSO infusion o n cochlear function and morphology Threshol ds: and response amp litudes of rcps were exam ined before and at va rious tim es (2, 7, and 14 da ys) during th e B SO infu sion. Figure 3 show s average l CP amplitudes a t frequencies from 0.5 to 16 kHz , before and after 2, 7, and 14 days of BSO tre atmen t. N o te th at 2
5
50 . 40
::J'
30
CL Cf)
20 10
en ~
ill
0
Nonnal
I
----.- 7 days ~
' : G:: .. I
