Acta OtolaryngolS7: 72-78, 1979

THE RELATIONSHIP BETWEEN THE CYTOTOXICITY OF KANAMYCIN AND ETHACRYNIC ACID FOR MAMMALIAN

CELLS IN VITRO AND THEIR OTOTOXICITY IN VIVO

K. E. Fox and R. E. Brummett

Depurtment of Pharmacology, School of Medicine, and Kresge Hearing Research Laboratory, University of Oregon Health Sciences Center, Portland, Oregon, USA

(Received April 7, 1978)

Abstruct. Dose-effect curves for inhibition of growth of P388/P mouse lymphoma cells by ethacrynic acid and kanamycin used alone and in combination were deter- mined in vitro. Ethacrynic acid was 600 times more potent than kanamycin and combinations of the drugs resulted in overall additive effects. These results were compared with known dose-effect data on the ototoxicity of these drugs in vivo. Kanamycin was highly selective in its toxicity for cochlear hair cells compared to cultured cells. The dose- effect data for ethacrynic acid was coincident with that re- ported for functional and biochemical effects on the coch- lea following perilymphatic perfusion with the drug. The potentiation observed following the ototoxic interaction of the two drugs in vivo was not observed following com- binations of the drugs in vitro.

Kanamycin, an aminoglycoside antibiotic, and ethacrynic acid, a "loop-inhibiting'' diuretic, produce dose-related toxic effects on mam- malian cochleae. Their ototoxic actions, how- ever, are qualitatively different. Ethacrynic acid after a single dose produces a rapid-onset reversible loss of hearing of several hours du- ration which correlates with a transient inter- stitial swelling of the stria vascularis (Schnei- der & Becker, 1966; Quick & Duvall, 1970; West et al., 1973). By contrast, kanamycin is ototoxic only after multiple doses and loss of hearing occurs with a latency of days to weeks following drug administration. In addition, ka- namycin produces permanent deafness which correlates with cochlear hair cell destruction (Benitez et al., 1962; Hawkins & Engstrom, 1964; Lundquist & Wersall, 1966). A dramatic interaction occurs between these two drugs when ethacrynic acid is administered shortly bsfore or several hours after a single nonoto- toxic dose of kanamycin (Johnson & Hamil-

4 < I" Otolu, 1 lZ,yiIl 87

ton, 1970; West et la., 1973). The result is a rapid-onset permanent loss of hearing as- sociated with hair cell destruction which ap- pears within a few hours. An intriguing ques- tion is how these drugs interact to damage the cochlea.

The purpose of this study was to determine the dose relationships for the cytotoxic effects of kanamycin and ethacrynic acid on mam- malian cells in vitro. We wanted to compare this data with that for the in vivo ototoxic ef- fects of these drugs obtained in other experi- ments. In particular, we were interested in whether or not the two drugs exhibited a po- tentiating interaction in the in vitro system as they do in vivo. There would be distinct ex- perimental advantages in having an in vitro model of the cytotoxic effects of these drugs, particularly if it paralleled the ototoxicity in vivo. The effect we chose for the quantita- tion of drug cytotoxicity in vitro was inhibition of mouse lymphoma cell proliferation during log phase growth in culture.

METHODS AND MATERIALS

P3WP mouse lymphoma cells were obtained from Dr Robert Roosa of the Wistar Institute, Philadelphia. The cells were grown in antibi- otic-free suspension culture medium (Gibco MEM no. F-14) supplemented with M L- serine and 10-3M sodium pyruvate and ster-

This research was supported by NIH grant 5 POI NS09889.

Act

a O

tola

ryng

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cgill

Uni

vers

ity o

n 11

/11/

14Fo

r pe

rson

al u

se o

nly.

Cytotoxicity of kanamycin and ethacrynic acid 73

Table I. Cytolytic effect of ethacrynic acid on P388/P cells in vitro

(9) under the various treatments by the follow- ing expression:

Cellslml x 10"

Drug conc. Inoculum 2-day culture (mgiml) (1) (El E/I X 100

0.03 1 1 1 5 2 105.OOi 12.00 95 0.5 1 1 1 2 2 3.05f 0.02 2.7 5.0 1 1 1 2 2 0.35k 0.04 0.3

12.5 111*2 0.06k 0.02 0.05

Mean + I S.D

ilized by filtration through 0.22 p pore size membrane filters (Millipore GSWP 047). The culture bottles contained a 5% CO, plus 95% air mixture at the time they were inoculated. Cultures were incubated at 37". Cell numbers in the cultures were determined with a model A Coulter Counter by making a minimum of two counts per culture. All counts were cor- rected for coincidence. A phosphate buffered Ringer's solution (pH 7.2) was used as a dilu- ent when counting cells.

For the drug experiments a 500 ml suspen- sion culture of P388/P cells was started from a small inoculum and grown in log phase to a concentration of 5 to 10X105 cellslml. The cells were then dispersed with gentle stirring and replicate 1.8 ml cultures were transferred to sterile 5 ml screw-cap tubes with the aid of a Filamatic vial filler. Each tube contained the C0,-air mixture and 0.2 ml of sterile drug or vehicle (control). At least 3 replicate cultures were prepared for each drug treatment and 10 replicate cultures for the vehicle. The cultures were incubated in a 37" incubator undisturbed for 2 days and then counted. In addition, 2 hours and 24 hours after setting up the cul- tures, triplicate sets of vehicle treated tubes were counted to determine the inoculum cell number and to check on log phase growth during the first day. The doubling time of the cells under these conditions ranged from 22 to 25 hours.

The cell counts made at day 2 were used to calculate the number of generations of growth

log N,-log N ,

.log 2 g =

where N 1 is the inoculum cell concentration and N , is the cell concentration at the end of the experiment. Means and standard devia- tions were computed from the .cell counts of replicate cultures.

Kanamycin sulfate (795 pg/mg potency) was a gift of Bristol Laboratories. Solutions were prepared in distilled water and the reported concentrations are expressed in terms of ka- namycin base. Ethacrynic acid was a gift of Merck, Sharp and Dohme Laboratories. Solu- tions were prepared fresh for each experiment by dissolving the ethacrynic acid in an equi- molar amount of 0.1 M NaOH and diluting with distilled water. Stock solutions of both drugs were sterilized by filtration through Mil- lipore membrane filters (0.22 p pore size) and dilutions were made in sterile distilled water.

RESULTS

Fig. 1 shows the dose relationships for inhibi- tion of proliferation of P388/P cells by etha- crynic acid and kanamycin. The estimated 50% inhibitory concentrations are 7 mg/ml (1.4X lop2 M) for kanamycin and 0.007 mg/ml (2.3X lop5 M) for ethacrynic acid. Ethacrynic acid is thus 600 times more potent than kana- mycin, on a molar basis, in inhibiting growth of the cell cultures.

Table I shows that ethacrynic acid at con- centrations above 0.03 mglml caused a dose related lysis of the P388/P cells as evidenced by the fact that the final cell number was less than the inoculum and decreased with increas- ing drug concentration. Ethacrynic acid may have caused lysis of cells at growth inhibitory concentrations below 0.03 mglml but this ef- fect would go undetected because replication of unaffected cells would offset the cells lost through lysis. Kanamycin at the highest con-

Actu Otoluryngol87

Act

a O

tola

ryng

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cgill

Uni

vers

ity o

n 11

/11/

14Fo

r pe

rson

al u

se o

nly.

74 K . E. Fox and R . E. Brummett

Table 11. Effect of KC1 on growth inhibition of P388/P cell cultures by ethacrynic acid

KCI Conc. W m l ) Without EA With EA

Percent of control growtha

4OOb 100 6 500 106 0 600 105 4 800 103 1

2 400 98 0 4 400 71 0

a EA=ethacrynic acid (0.02 mglml)

medium. This is the concentration of KC1 in the standard growth

centration tested did not demonstrate a similar lytic effect. In addition, light microscopic ob- servations were made of cells grown on glass slides in the presence of concentrations of each drug which inhibited cell proliferation by 50%. The ethacrynic acid treated slides had half the number of cells present in the control and the surviving cells appeared normal. By contrast, the kanamycin treated slides showed clones of normal cells and areas of debris containing an occasional outline of a single cell. The degenerative effects of kanamycin appeared to occur in an entire clone of cells.

Because ethacrynic acid has been reported to cause potassium loss from cultured mam- malian cells (Hammill et al., 1972), we at- tempted to prevent the toxic effects of the drug on P388/P cells by increasing the concentration of KCl in the culture medium. Table I1 shows the results of this experiment. Potassium chlo- ride itself inhibited cell proliferation at concen- trations greater than 800 pg/ml and failed to protect the cells from the toxic effects of 0.02 mg/ml ethacrynic acid at all concentrations tested.

Table I11 shows the results of experiments on the effect of various combinations of con- centrations of ethacrynic acid and kanamycin on proliferation of P388/P cells in culture. The nine combinations studied spanned a range of expected effects from 0 to 92% inhibition of cell culture growth. It can be seen from the

Ar IU O ~ o l r r t \ i ~ g o I 87

data that in all cases the effects of the two drugs were simply additive. Linear regression analysis (see Table 111, footnote) of the data confirmed this conclusion.

DISCUSSION

Aminoglycoside antibiotics have a long history of use in mammalian tissue culture medium to prevent bacterial contamination. For the most part investigators assumed that the drugs were nontoxic at the concentrations employed and this has been confirmed by a few quantitative studies of the effects of these drugs on mam- malian cells in vitro (Maskowitz & Kelker, 1963; Casemore, 1967; Schafer et al., 1972; Fischer, 1975). In the cited studies a number of mammalian and a few nonmammalian cell lines were shown to be unaffected by anti- bacterial concentrations of streptomycin (100 pg/ml) and gentamicin (50 to 100 pg/ml). At higher concentrations of the drugs, all cell lines tested showed toxic effects at essentially

Table 111. Cytotoxic effect of combinations of ethacrynic acid and kanamycin on P388/P cells in vitro The experiment was performed as described in Methods. In this case the percent inhibition was obtained by sub- tracting the percent of control growth from 100. The values for the Expected results were derived by adding the inhibition obtained with each drug used alone at the in- dicated concentrations and the Observed values were ob- tained with the drugs used together

Percent inhibition of cell Drug conc.(mg/ml)a multiplication

EA KAN Expected Observed

0.0005 0.01 0.001 0.01 0.001 0.10 0.001 1.00 0.005 1 .oo 0.01 1 .OO 0.01 10.00 0.03 1 .oo 0.10 1.00

1 0 5 0

15 10 8 3 34 28 48 46 63 59 99 93 92 89

a EA=ethacrynic acid; KAN=kanamycin. Linear regression analysis of Expected vs. Observed:

correlation coefficient= 1.00; slope= 1.00; intercept= 3.99%.

Act

a O

tola

ryng

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cgill

Uni

vers

ity o

n 11

/11/

14Fo

r pe

rson

al u

se o

nly.

Cytotoxicity of kanamycin and ethacrynic acid 75

100 1

i e d meo-

+ g60- 0

b $40- W

L L

20

-

-

P 388/P Mouse Lymphoma Cells

I M e a n t l S D

Ethacrynic Acid Kanamycin

Fig. 1. Dose-effect curves,for inhibition of growth of P388/P cell cultures by kanamy-

, I , I , , I . . I cin and ethacrynic acid. See Methods. 1 2 5 1 2 5 1 2 - 5 1 2 5 1 2 5 1

XD-3 XD* xIO-1 XI00 XIO' Kanamycin

x0-4 x10-3 xIO-2 x10-1 XI@ Eiimcrync Acid

MILLIGRAMS PER ML

the same concentration. Gentamicin inhibited growth of mammalian cell cultures progres- sively between 0.2 and 2.0 mg/ml while strep- tomycin required concentrations of 2.0 mg/ml and higher. When we compare our results (Fig. 1) with those reported for gentamicin, we find that kanamycin is roughly ten times less potent than gentamicin in inhibiting mammalian cell growth in vitro. Kanamycin is also less potent than gentamicin in other, in vivo, effects (e.g., acute lethality in mice (Nord et al., 1967), neuromuscular blockade (Adams, 1973), ther- apeutic daily dose recommended for humans (Weinstein, 1975)).

In studies on guinea pigs we have found that gentamicin in a daily dose of approximately 75 mg/kg for 4 weeks will produce a 50% loss of cochlear hair cells associated with a similar loss of cochlear function as measured by the AC cochlear potential (Brummett & Fox, 1978~). By interpolating the data, the con- centration of gentamicin in the perilymph of animals receiving this dose can be estimated to have been 12 pg/ml. From studies of kana- mycin in guinea pigs (Brummett & Fox, 1978b), we estimate that perilymph concentra- tions at maximally ototoxic doses are between 50 and 100 pg/ml. These perilymph concentra- tions of gentamicin and kanamycin are well below those shown to be cytotoxic for mam- malian cells in culture. Therefore, assuming

that the perilymph which surrounds the coch- lear hair cells is equivalent to the culture medium of the in vitro experiments, it can be concluded that the hair cells exhibit a marked sensitivity to the toxic effects of the amino- glycoside antibiotics. It remains to be shown whether this sensitivity is the result of pre- ferential accumulation of the aminoglycosides by the hair cells or whether it reflects a funda- mental difference between hair cells and cul- tured mammalian cells at the sites of action of the drugs. With regard to the latter, the morph- ological effects of the aminoglycoside anti- biotics on hair cells strongly suggest the cell membrane as a primary target of the drugs. The drugs are known to damage bacterial cell membranes (Dubin et al., 1963; Hancock, 1964). Furthermore, the appearance of cul- tured mammalian cells after treatment with toxic concentrations of the drugs is consist- ent with membrane breakdown. Recently, Schacht (1976) has reported effects of neo- mycin (another aminoglycoside antibiotic) on membrane phospholipid metabolism in the organ of Corti; however, the relevance of the results of these novel experiments to amino- glycoside ototoxicity remains to be estab- lished.

In reports of the effects of ethacrynic acid on cultured mammalian cells, Gordon (1968) and Gordon & Hartog (1969) have shown that

Acra Otolaryngol87

Act

a O

tola

ryng

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cgill

Uni

vers

ity o

n 11

/11/

14Fo

r pe

rson

al u

se o

nly.

76 K . E. Fox and R. E. Brummett

ethacrynic acid decreased oxygen uptake and glycolysis of intact Ehrlich ascites cells. How- ever, the concentration of ethacrynic acid re- quired to produce a 50% decrease in both pro- cesses was about M which is fifty-fold greater than that for 50% inhibition of P388/P cell proliferation (Fig. 1). The cytolytic effects we observed suggest that damage to mem- brane function may be the primary mechanism of the cytotoxic effects of this drug. The in- hibition of energy metabolism reported above would then be a secondary event following entrance of greater amounts of ethacrynic acid into the cells.

The stria vascularis of the cochlea is a well documented site of action of ethacrynic acid in the inner ear (Brummett et al., 1977). A dramatic interstitial edema occurs transiently in the stria vascularis following intravenous administration of ethacrynic acid and the time course of the strial swelling parallels the loss and recovery of electrophysiological function in the cochlea. Investigators have sought the mechanism of this effect through biochemical studies of the stria vascularis. Recently, Thal- mann et al. (1977) have reported data showing that the dose-effect curves for inhibition of the endocochlear potential and inhibition of strial adenyl cyclase by ethacrynic acid are coin- cident following perfusion of ethacrynic acid into the perilymphatic space. The 50% inhib- itory concentration was lop5 M. Thalmann’s dose-effect curves (Thalmann et al., 1977, fig. 1) coincide with ours (Fig. 1) for the cytotoxic effect of ethacrynic acid on P388/P cells when the data are expressed in equivalent units (Mw of ethacrynic acid=303). The coincidence be- tween these three dose-effect curves is even more striking in view of the fact that Palo- heimo & Thalmann (1976) have shown with 14C-ethacrynic acid that the concentration of drug in the stria vascularis following perilym- phatic infusion is essentially identical to that in the perfusion fluid indicating that the drug freely permeates any one or combination of the following structures: basilar membrane; Reissner’s membrane; spiral ligament. How-

A ( i ( i Ut ( i /~ i i I i rqol 87

ever, we have found (Fox & Brummett, 1978) that following an intravenous dose of 40 mg/kg of 14C-ethacrynic acid in the guinea pig, we were unable to detect ethacrynic acid in the perilymph at a concentration of lop5 M, the limit of our assay, at the time of maximal depression of the AC cochlear potential by the drug. This observation and the cytotoxic dose- effects of ethacrynic acid on cultured cells raise some questions about the relationship be- tween the effects of ethacrynic acid following intravenous administration and those seen fol- lowing perfusion of the drug into the peri- lymphatic space. First, the decrease in the endocochlear potential after perilymphatic perfusion could be a result of drug action not on the stria vascularis but on the cells over- laying the basilar membrane, in particular the hair cells, or those composing Reissner’s membrane since both serve to separate the perilymphatic space from the endolymphatic space and thereby maintain the endocochlear potential. Second, there is the fact that etha- crynic acid is cytolytic to cultured mammalian cells at the concentrations introduced into the cochlea in the perilymphatic perfusion experi- ments, yet following a single intravenous dose the drug causes a decrease in the endolym- phatic and AC cochlear potentials without per- manently damaging any of the cells of the stria or organ of Corti (Fox et al., 1978). Third, is the question of the amount of ethacrynic acid in the stria vascularis at the time the effects are seen on the endocochlear potential. It is possible that the concentrations of drug in the stria vascularis reported by Paloheimor & Thalmann (1976) following perilymphatic per- fusion may also occur following intravenous injection of effective doses without sufficient drug diffusing into the perilymph to be detect- able at lop5 M. It seems more probably, how- ever, that the drug should be able to permeate the membranes equally in either direction. Determination of the concentration of etha- crynic acid in the stria following an effective intravenous dose would settle this question.

We also tested combinations of ethacrynic

Act

a O

tola

ryng

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cgill

Uni

vers

ity o

n 11

/11/

14Fo

r pe

rson

al u

se o

nly.

Cytotoxicity of kanamycin and ethacrynic acid 77

acid and kanamycin for inhibition of prolifera- tion of P388/P cells in vitro. The results of that experiment (Table 111) showed that the effects of the two drugs were simply additive over a wide range of concentrations. Therefore, etha- crynic acid and kanamycin do not potentiate each other with regard to their toxic effects on the in vitro system. This is in contrast to their ototoxic interaction in vivo in which a single nonototoxic dose of kanamycin in combina- tion with a reversibly ototoxic dose of etha- crynic acid produces cochlear hair cell death and permanent deafness (West et al., 1973). Failure to observe an interaction in vitro may be due to the absence of metabolites of either kanamycin or ethacrynic acid which may be produced in the intact animal and act as the final mediators of the ototoxic interaction.

In summary, our experiments have accom- plished the following: (1) reemphasized the unique selectivity of the aminoglycosides to destroy cochlear hair cells; (2) shown that ethacrynic acid is a potent cytotoxic drug and thereby raised questions about the relationship of effects on the cochlea obtained by peri- lymphatic perfusion of this drug to those fol- lowing its intravenous injection; and ( 3 ) shown that the in vitro mammalian cell culture system does not demonstrate quantitatively a toxic in- teraction between ethacrynic acid and kana- mycin analogous to that found in the cochlea in vivo.

ZUSAMMENFASSUNG

Die Wachstumshemmung einer Saugetier-Zel- h i e in vitro durch Ethacryn-Saure und Kana- mycin, allein und kombiniert, wurde in Dosis- Wirkungskurven bestimmt und mit bekannten ototoxischen In-vivo-Konzentrationen dieser Substanzen verglichen. Kanamycin war vie1 giftiger fur cochleare Haarzellen als fur Kul- turzellen. Ethacryn-Saure war ein wirksame- rer Wachstumshemmer von Kulturzellen als Kanamycin. Es war wirksam in Konzentra- tionen, die, wie berichtet, nach penlympha- tischer Perfusion funktionelle und bioche-

mische Anderungen in der Cochlea verur- sachen. Bei kombinierter Anwendung in vitro verursachten Kanamycin und Ethacryn-Saure eine additive Wirkung. Potenzierung, die in vivo charakteristisch fur die ototoxische Wechselwirkung ist, wurde nicht beobachtet

REFERENCES Adams, H. 1973. Neuromuscular blocking effect of amino-

glycoside antibiotics in nonhuman primates. J A m Vet Med Assoc 163, 613.

Benitez, J., Schuknecht, H. & Brandenburg, J. 1962. Pathologic changes in human ear after kanamycin. Arch Otolaryngol75, 192.

Brummett, R. & Fox, K. 1978a. Comparative ototoxic liability of netilmicin and gentamicin. Arch Otolaryn- go1 (in press).

Brummett, R. & Fox, K. 19786. Unpublished observa- tion.

Brummett, R., Smith, C., Ueno, Y . , Cameron, S. & Rich- ter, R. 1977. The delayed effects of ethacrynic acid on the stria vascularis of the guinea pig. Acta Oto- laryngol (Stockh.) 83, 98.

Casemore, D. 1967. Gentamicin as a bactericidal agent in virological tissue culture. J Clin Pathol20, 298.

Dubin, D., Hancock, R. & Kavis, B. 1963. The sequence of some effects of streptomycin in Escherichia coli. Biochem Biophys Acta 74, 476.

Fischer, A. 1975. Gentamicin as a bactericidal antibiotic in tissue culture. Med Microbiol Immunol (Berl.) 161, 23.

FOX, K. & Brummett, R. 1978. Manuscript in preparation. Fox, K., Russell, N . & Brummett, R. 1978. Cochlear al-

terations produced by the ototoxic interaction of kana- mycin and ethacrynic acid. Fed Proc 37, 502.

Gordon, E. 1968. Site of ethacrynic acid action on Ehr- lich ascites tumor cells. Biochem Pharmucol 17, 1237.

Gordon, E. & Hartog, M. 1969. The relationship between cell membrane potassium ion transport and glycolysis: The effect of ethacrynic acid. J Gen Physiol54, 650.

Hammill, N., Lamb, G. & Lindsay, R. 1972. Transient ef- fects of ethacrynic acid on sodium and potassium movements in L-cells. J Physiol (Lond.) 222, 43P.

Hancock, R. 1964. Early effects of streptomycin on Ba- cillus megaterium. J Bacteriol88, 633.

Hawkins, J. & Engstrom, H. 1964. Effect of kanamycin on cochlear cytoarchitecture. Acta Otolaryngol (Suppl.) (Stockh.) 188, 100.

Johnson, A. & Hamilton, C. 1970. Kanamycin ototoxic- ity-possible potentiation by other drugs. South Med 363, 511.

Lundquist, P. & Wersall, J. 1966. Kanamycin induced changes in cochlear hair cells of the guinea pig. Z Zellforsch Mikrosk Anat 72, 543.

Maskowitz, M. & Kelker, N. 1963. Sensitivity of cultured mammalian cells to streptomycin and dihydrostrepto- mycin. Science 141, 647.

Nord, N., Watanabe, F. , Parker, R. & Hoeprich, P. 1967.

Act

a O

tola

ryng

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cgill

Uni

vers

ity o

n 11

/11/

14Fo

r pe

rson

al u

se o

nly.

78 K . E. Fox and R. E. Brummett

Comparative acute toxicity of four drugs. Arch Intern Med 119, 493.

Paloheimor, S. & Thalmann, R. 1976. Comments on “Tis- sue concentrations of ethacrynic acid and inhibition of enzymes and electrophysiological responses in the cochlea”. J Acoust Soc Am 59, 708.

Quick, C. & Duvall, A. 1970. Early changes in the coch- lear duct from ethacrynic acid: An electromicroscopic evaluation. Laryngoscope 80, 954.

Schacht, J. 1976. Biochemistry of neomycin toxicity. J Acoust Soc Am 59, 940.

Schafer, T., Pascale, A , , Shimonaski, G. & Came, P. 1972. Evaluation of gentamicin for use in virology and tissue culture. Appl Microbiol23, 565.

Schneider, W. & Becker, E. 1%6. Acute transient hearing loss after ethacrynic acid therapy. Arch Intern Med 117, 715.

Thalmann, R., Ise, I., Bohne, B, & Thalmann, I. 1977. Actions of “loop” diuretics and mercurials upon the cochlea. Acta Otolaryngol (Stockh.) 83, 221.

Weinstein, L. 1975. Streptomycin, gentamicin and other aminoglycosides. In The Pharmacological Basis of Therapeutics (ed. L. Goodman & A. Gilman), chap. 58. Macmillan, New York.

West, B., Brummett, R. & Himes, D. 1973. Interaction of kanamycin and ethacrynic acid. Arch Otolaryngol 87. 32.

Kaye E . Fox, P1.D. Health Sciences Center Portland Oregon 97201 USA

Act

a O

tola

ryng

ol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

cgill

Uni

vers

ity o

n 11

/11/

14Fo

r pe

rson

al u

se o

nly.