University of Oklahoma, Health Sciences Center, College of Pharmacy, 1110 N. Stonewall, Oklahoma City, Oklahoma 73190
Received August 13, 1999; accepted April 3, 2000
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key Words: difluoromethylornithine (DFMO) ototoxicity; hearing loss; DFMO enantiomer ototoxicity; rat; guinea pig.
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The two enantiomers of DFMO differ in their ability to inhibit ODC with the L form being more potent then the D enantiomer (Danzin et al., 1987). Although the physiologic functions of polyamines are not completely understood, it is clear that their intracellular concentration is highly regulated and that normal cell growth, replication, differentiation, and secretory and repair functions require polyamines (Bachrach, 1973
; Luc and Baylin, 1984
; Oka et al., 1981
; Pegg, 1986
; Pegg and McCann, 1982
; Thet et al., 1984
; Williams-Ashman and Canellakis, 1979
). Polyamines have been found in high levels in many tumor cells (Pasic et al., 1997
) and support sustained cell growth that is essential for the multi-step process of cancer development. In animal models of colon carcinogenesis, inhibition of ODC by DFMO reduces the number and size of colon adenomas and carcinomas (Meyskens and Gerner, 1995
).
Polyamines have also been found in the cochlea, but their role in hearing is unknown (Schweitzer et al., 1986). Because of their amphipathic nature, a potential role of polyamines in the cochlea might be to interact with hydrophobic environments, thus altering the permeability of membranes that regulate the flux of inorganic cations (Canellakis et al., 1979
). Cochlear function is intimately dependent on the unique electrolytic concentration and positive polarization of the endolymph. Therefore, changes in the levels of polyamines could alter the kinetics of electrolytes that underlie the endocochlear potential and consequently affect cochlear function (Schweitzer et al., 1986
).
Animal models of DFMO ototoxicity produce conflicting data. The development of chemotherapeutic agents with less ototoxic potential is complicated by a lack of dosage data and by uncertainty over which species are susceptible to DFMO ototoxicity. The route of DFMO administration most commonly employed has been via ingestion of adulterated drinking water (e.g., Marks et al., 1991; Salzer et al., 1990
). However, accurate quantitation of water intake (and thereby drug dose) has not been carried out. Weight loss is a serious consequence of such DFMO exposure (Meyskens and Gerner, 1995
) though it is not clear whether the weight loss results from subjects limiting their intake of contaminated water (possibly due to factors such as taste) or from the toxic effect of the drug itself. In the guinea pig, DFMO inhibits ODC activity in cochlear tissue and a significant depletion of cochlear polyamines results (Marks et al., 1991
). Along with this, brainstem audiometry shows that treatment by water adulterated with 1% DFMO produces hearing loss accompanied by damage in the hook and first turn with a loss of hair cells in all rows. Inner hair cells are lost at a greater rate than outer hair cells (OHCs) (Salzer et al., 1990
). When the rat is used as the model, auditory thresholds evaluated by brainstem evoked potentials remain unchanged in the DFMO treated rats (Schweitzer et al., 1986
), although only a limited dose range was investigated. The suggestion has been made (Schweitzer et al., 1986
) that in the rat cochlear polyamines are not depleted by DFMO to a level considered critical for disrupting polyamine-dependent processes in other systems.
Due to the effectiveness of DFMO in treating certain types of cancer, the goal of this study is to determine whether the enantiomers of this compound differ in their ototoxicity in a susceptible species. To accomplish this, it was essential to establish a dose-effect function for D,L-DFMO on auditory function in both the rat and the guinea pig. The markers of auditory function are the compound action potential (CAP) threshold and cochlear microphonic (CM) amplitude. The CAP represents the output of the auditory nerve in response to sound stimuli. The CM is an AC electrical potential generated mainly in outer hair cells of the organ of Corti in response to acoustic stimulation. If inner hair cells are preferentially damaged by DFMO, then CAP threshold would be disrupted, but CM amplitude would not be reduced. If outer hair cells especially near the base of the cochlea are selectively disrupted by DFMO, then CM amplitude would be reduced and CAP threshold would be impaired secondary to the loss of outer hair cell function (modeled by CM amplitude). In most previous studies, DFMO was given in the drinking water, making the accurate determination of total drug intake very difficult. Therefore, in the current study, drug administration to the rats is by gavage, modeling the normal route of human exposure. Because of the volume of drug required in the guinea pigs over a 40-day exposure period, the use of daily gavage administration was not feasible. Because the primary goal in working with this DFMO- sensitive species was to compare ototoxic potential between the racemic and enantiomeric forms of the drug, the drugs were administered via injection so that accurate calculations could be generated concerning the amount of actual drug received.
![]() |
MATERIALS AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Procedures
Two initial studies were undertaken in order to establish a minimal ototoxic dose for D,L-DFMO in rats and guinea pigs that did not also compromise the health of subjects as reflected by failure to gain weight or by death. Once these dose regimens were defined, the principal study entailed a comparison of the ototoxicity of this dose of the racemate compared to the same dose of each of the enantiomers.
D,L-DFMO dose response in rats.
In order to establish a minimal ototoxic dose of D,L-DFMO that did not also compromise the general health of subjects as reflected by failure to gain weight, 24 animals were randomly placed into 8 groups of 3 subjects each. Treatment groups were as follows: 0, 200, 400, and 600 mg/kg/day D,L-DFMO by gavage daily (5 days/week) for 4 and 8 weeks. Four additional animals were initially assigned to 2 groups that received 400 and 600 mg/kg/day by gavage twice a day (5 days/week). These two extreme doses were discontinued because the initial subjects failed to gain weight normallly during exposure. Therefore, only preliminary data concerning weight gain and auditory function are available at these high doses in the rat. For the weekend, the appropriate dose of D,L-DFMO for each subject was diluted in 30 ml of drinking water per day. Weight measurements were taken daily (5 days/week) with assessment of auditory function performed at the conclusion of dosing.
D,L-DFMO dose response in guinea pigs.
Twenty-five juvenile pigmented guinea pigs were randomly assigned to 5 treatment groups of 5 subjects each. Treatment groups received 0 (saline), 0.5, 1, 2, and 3 g/kg/day D,L-DFMO daily (7 days/week) by ip injection. Only two subjects survived the 2 g/kg/day treatment while no subjects survived the 3 g/kg/day exposure. Additional guinea pigs were not added to these high-exposure groups.
Auditory thresholds were assessed in the 0.5-g/kg and 1-g/kg/day subjects at day 45 of dosing while the 2-g/kg/day subjects were evaluated at 22 days of dosing. Thus, subjects in the 1-g/kg/day and 2-g/kg/day dose groups of D,L-DFMO received approximately identical total D,L-DFMO doses per kg of body weight. Daily weight measurements determined the injection volume. Assessment of auditory function occurred at the conclusion of dosing.
DFMO enantiomers versus racemate in guinea pigs.
Twenty subjects were randomly assigned to treatment groups (n = 5) that received 1 g/kg/day D-DFMO, 1 g/kg/day L-DFMO, 1 g/kg/day D,L-DFMO, or saline ip daily for 45 days. Assessment of auditory function occurred at the conclusion of dosing.
CAP and CM recording and analysis.
At the end of drug treatment (24 h after the last drug dosage), the rats were anesthetized with xylazine (13 mg/kg, im) and ketamine (87 mg/kg, im) and guinea pigs were anesthetized with xylazine (5 mg/kg, im), ketamine (30 mg/kg, im), and urethane (160 mg/kg, ip). The round window was surgically exposed using a ventro-lateral approach and a silver wire electrode was carefully placed on the round window under a surgical microscope for recording compound action potential (CAP) thresholds and the 1 µV RMS cochlear microphonic (CM). A silver chloride electrode was placed in the neck muscle as the reference. The CAP and CM signals were amplified with a Grass A. C. preamplifier (Model P15). The gain was 1000. The band-pass frequency for CAP was 0.11.0 kHz and 0.150 kHz for CM. The CAP signals were observed using a digital oscilloscope (Nicolet Instrument Co., 2090-IIIA). The CM signals were sent to a SR530 Lock-in amplifier (Stanford Research Systems, Inc.) and then to a PC computer for automatic determination of the 1 µV RMS amplitude. The sound level of test frequencies that evoked a just detectable CAP was determined and this value was used to estimate the threshold at the frequency. The sound levels that evoked 1 µV RMS CM at each test frequency was determined by the computer program and the iso-amplitude curve was evaluated. CAP-threshold shifts were calculated as the difference between each CAP-threshold for each treated subject and the mean control value (from subjects exposed to vehicle alone) at the same frequency. CM iso-amplitude curve elevation was obtained in a similar way.
Pure tones for eliciting CAP and CM were generated with the SR530 Lock-in Amplifier (Stanford Research Systems, Inc.). The signals were attenuated by a programmable attenuator and then amplified by a high voltage amplifier and delivered to a high-frequency-sound source (made from an ACO " microphone, 7013) placed within a speculum that fit into the exposed external auditory meatus. CAP and CM were evaluated for pure-tone frequencies of 2, 4, 6, 8, 12, 16, 20, 24, 30, 35, and 40 kHz. Continuous tones were used for eliciting CM. Tone bursts were used to elicit the CAP. The duration of the tone bursts was 10 ms with a rise/fall time of 1.0 ms, and a repetition rate of 9.7/s. Sound levels at all testing frequencies were calibrated with a probe microphone located near the eardrum.
Separate repeated-measures analyses of variance (NCSS software) were used to evaluate the effects of drug treatment in each species and on each of the 2 outcome measures of auditory function. Frequency was evaluated as a within-subject variable and treatment group was a between-subject variable. Post hoc analysis was performed by Fisher's least squares difference.
Analysis of growth curves was initially evaluated qualitatively. Because of the significant systemic toxicity (loss of body weight) and subject mortality that occurred in the initial dose-response studies undertaken using the racemic form of DFMO, the power of these tests to identify a significant loss due to drug treatment was reduced.
Histology.
Surface preparations of the Organ of Corti were evaluated in one guinea pig that showed a profound loss in CAP threshold, but normal CM amplitude. These findings permit a detailed comparison of impairment in auditory function with precise characterization of sensory cell damage. The subject was decapitated while deeply anesthetized after CAP and CM recording. Cochleae were removed immediately. Both the round and oval windows were opened, and the apex of the cochlea was drilled open to facilitate perfusion. The cochleae were perfused with SDH (succinate dehydrogenase) incubative solution (0.05 M sodium succinate, 0.05 M phosphate buffer, and 0.05% tetranitro blue tetrazolium) and immersed in the solution for one hour (37°C). Then the cochleae were fixed with 10% formalin for 2 days. After fixation, the cochleae were decalcified in 10% EDTA solution (ethylenediamine tetraacetic acid) for 3 days. Cochlea microdissection was accomplished under a light microscope.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In contrast to the data observed in rats, substantial general toxicity and ototoxicity were observed when guinea pigs were treated with doses of D,L-DFMO between 500 mg/kg/day and 3 g/kg/day. Figure 1 shows the rate of weight gain in guinea pigs for which auditory assessments were made. The 500-mg/kg/day D,L-DFMO animals showed no obvious signs of systemic toxicity; their growth rate actually surpassed that of the controls. Weight gain for the 1-g/kg/day subjects appears to diverge from the controls during the last 10 days of the dosing period. Two of the 5 1-g/kg/day subjects did not survive the full 6-week dosing regimen. Weight gain was effectively blocked for the 2-g/kg/day subjects. By day 22, the 2-g/kg/day animals had gained virtually no weight so the animals' auditory functions were evaluated at this time. All 3-g/kg/day subjects died within the first 6 days of treatment. Repeated-measures ANOVA for effect of treatment on weight gain was restricted to those groups dosed for the full 42-day exposure period and excluded the 2-g/kg/day subjects. A significant effect of treatment was not observed for subjects receiving the DFMO racemate at doses of 0, 0.5, and 1.0 g/kg (F[2,8] = 1.60, p = 0.26). However, a significant reduction in weight gain was observed between the controls and 2-g/kg/day subjects at day 20 (F[1,3] = 65.90, p = 0.004).
|
|
|
|
|
|
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Previous studies have demonstrated that D,L-DFMO treatment in the guinea pig generates significant auditory dysfunction (Marks et al., 1991) and reduces cochlear polyamines to levels 3 times lower than that seen in rats (Henley et al., 1987
). However, such studies relied upon dosing animals via ingestion precluding the accurate determination of a dose-response function relating hearing loss and drug dose (Salzer et al., 1990
). Moreover, there was no means of determining from previous studies whether the loss in body weight shown in the guinea pigs resulted from D,L-DFMO toxicity or from a reduction in water (and subsequently food intake) due to taste aversion (from access only to water contaminated with drug). The current study shows substantial D,L-DFMO toxicity occurs at doses of 3 g/kg/day with 100% of the animals dying within a 6-day exposure period. In addition, the 2-g/kg/day D,L-DFMO subjects failed to show normal weight gain over the entire 22-day exposure period Overall, the remaining treatment groups did not produce significant toxicity (mortality or weight loss) within the trial period. Necropsies were carried out on one animal that died during the course of treatment and on the subjects that survived until the end of the experiment. The reports support the conclusion that trauma from repeated daily injection was not responsible for the subject's death or for depression in weight gain.
Even though D,L-DFMO did not result in significant weight loss at exposure levels of 1 g/kg/day and less, D,L-DFMO produces a significant impairment in auditory threshold sensitivity in the guinea pig. The loss of CAP sensitivity is consistent with the greater loss of inner hair cells seen in the study by Salzer and colleagues (1990). CM is generated mainly by the outer hair cells in the basal turn of the cochlea and is a sensitive measurement for detecting impairment following many ototoxic drugs, including the aminoglycoside and cisplatin (e.g., Riggs et al., 1996; Stengs et al., 1998
), but D,L-DFMO did not generate a change in the CM iso-amplitude curve. Again, this is consistent with published data suggesting that IHCs are more sensitive to D,L-DFMO than OHCs (Jansen et al., 1989
; Salzer et al., 1990
). It is supported also by the limited histology completed in one subject, which revealed that the majority of outer hair cells at the very basal turn were intact. The basal inner hair cells, however, showed extensive damage, with a majority missing. The middle of the cochlea showed both damage and loss of outer and inner hair cells. At the apex, neither outer nor inner hair cells were damaged or missing. Therefore, D,L-DFMO may have some preference for hair cells located at the basal turns of the cochlea. As expected, the average threshold shift across test frequencies is dose-dependent.
The current finding that rats are insensitive to the ototoxic effects of D,L-DFMO at doses below those which yield obvious signs of toxicity extends previously published data. Previous studies have also failed to show evidence of ototoxicity in the rat following D,L- DFMO. However, those studies have administered the drug in the subject's drinking water and have not clearly controlled for possible reduction in water intake due to unpalatability of the solution. Moreover, the range of exposure doses used in those studies has been far more limited than in this work, leaving open the possibility that the rat might show an ototoxic response, but only at very high exposure levels. For example, Schweitzer et al. (1986) found that treating rats with 1% DFMO in the drinking water (approximately 300 mg/kg, assuming a daily intake of 30 ml water) not only resulted in normal brainstem-evoked potentials, but found that the levels of polyamines were not depleted to a level considered disrupting in other systems. In the current study, rats were exposed to D,L-DFMO by intubation to much greater doses, up to 4 times those given previously. Even with these extreme dosages, some of which appear to impair weight gain, the rat model generated no significant threshold shifts (ototoxicity) over the audiometric range. The results imply that D,L-DFMO treatment in rats does not produce the ototoxic side effects previously reported in human cancer patients and in other laboratory models such as the guinea pig. If the levels of polyamines do play a role in auditory function, the metabolism of D,L-DFMO in the rat may not inhibit polyamine synthesis to a critical level, and therefore the rat cannot be used as a reliable model for determination of D,L-DFMO ototoxicity.
Determination of susceptibility to D,L-DFMOinduced ototoxicity in rats and guinea pigs was essential prior to assessing the ototoxic potential of its enantiomers. Comparisons were made among compounds using a dosage, 1 g/kg/day, which produced significant ototoxicity but little general toxicity to the racemate. By comparison, at this same dose level, the D form of DFMO was found to have no significant effects on either the compound action potential or cochlear microphonic. Evaluation of auditory function following administration of the L form of DFMO produced significant disruption of normal cochlear potentials. Since Danzin et al. (1987) reported that the L form of DFMO is able to inhibit ODC more effectively than the D form, these findings would seem to suggest a relationship between the extent of ODC inhibition and susceptibility to ototoxicity. Such an interpretation underscores the importance of assessing how polyamines influence auditory function.
![]() |
ACKNOWLEDGMENTS |
---|
![]() |
NOTES |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Canellakis, E. S., Viceps-Madore, D., Kyriakidis, D. A., and Heller, J. S. (1979). The regulation and function of ornithine decarboxylase and of the polyamines. In Current Topics in Cellular Regulation, Vol. 15, pp. 155202. Academic Press, New York.
Croghan, M. K., Aickin, M. G., and Meyskens, F. L. (1991) Dose-related -difluoromethylornithine ototoxicity. Am. J. Clin. Oncol. 14, 331335.[ISI][Medline]
Danzin, C., Ducep, J. B., Schirlin, D., and Wagner, J. (1987). Absence of stereospecificity in the suicide inhibition of ornithine decarboxylase. In Biochemistry of Vitamin B6, pp. 333336. Birkhauser Verlag, Basel-Boston.
Henley, C. M., III, Gerhardt, H. J., and Schacht, J. (1987). Inhibition of inner ear ornithine decarboxylase by neomycin in vitro. Brain Res. Bull. 19, 695698.[ISI][Medline]
Jansen, C., Mattox, D. E., Miller, K. D., and Brownell, W. E. (1989). An animal model of hearing loss for alpha-difluoromethylornithine. Arch. Otolaryngol. Head Neck Surg. 115, 12341237.[Medline]
Luc, G. D., and Baylin, S. B. (1984). Inhibition of intestinal epithelial DNA synthesis and adaptive hyperplasia after jejunectomy in the rat by suppression of polyamine biosynthesis. J. Clin. Invest. 74, 698704.[ISI][Medline]
Marks, S. C., Mattox, D. E., and Casero, R. A. (1991). The effects of DFMO on polyamine metabolism in the inner ear. Hear. Res. 53, 230236.[ISI][Medline]
Meyskens, F. L., Jr., Emerson, S. S., Pelot, D., Meshkinpour, H., Shassetz, L. R., Einspahr, J., Alberts, D. S., and Gerner, E. W. (1994). Dose de-escalation chemoprevention trial of -difluoromethylornithine in patients with colon polyps. J. Natl. Cancer Inst. 86, 11221130.[Abstract]
Meyskens, F. L., Jr., and Gerner, E. W. (1995). Development of difluoromethylornithine as a chemoprevention agent for the management of colon cancer. J. Cell. Biochem. Suppl. 22, 126131.[Medline]
Meyskens, F. L., Jr., and Gerner, E. W. (1999). Development of difluoromethylornithine (DFMO) as a chemoprevention agent. Clin. Cancer Res. 5, 945951.
Oka, T., Perry, J. W., Takemoto, T., Sakai, T., Terada, N., and Inoue, H. (1981). The multiple regulatory roles of polyamine in the hormonal induction of mammary gland development. In Advances Polyamine Research (C. M. Caldarera, V. Zappia, and U. Bachrach, Eds.), Vol. 3, pp. 309320. New York, Raven Press.
Pasic, T. R., Heisey, D., and Love, R. R. (1997). -Difluoromethylornithine ototoxicity. Chemoprevention clinical trial results. Arch. Otolaryngol. Head Neck Surg. 123, 12811286.[Medline]
Pegg, A. E. (1986). Recent advances in the biochemistry of polyamines in eukaryotes. Biochem. J. 234, 249262.[ISI][Medline]
Pegg, A. E., and McCann, P. P. (1982). Polyamine metabolism and function. Am. J. Physiol. 243, C212C221.[Abstract]
Riggs, L. C., Brummett, R. E., Guitjens, S. K., and Matz, G. (1996). Ototoxicity resulting from combined administration of cisplatin and gentamicin. Laryngoscope 106, 401406.[ISI][Medline]
Salzer, S. J., Mattox, D. E., and Brownell, W. E. (1990). Cochlear damage and increased threshold in -difluoromethylornithine (DFMO)-treated guinea pigs. Hear. Res. 46, 101112.[ISI][Medline]
Schweitzer, L., Casseday, J. H., Sjoerdsma, A., McCann, P. P., and Bartolome, J. V. (1986). Identification of polyamines in the cochlea of the rat and their potential role in hearing. Brain Res. Bull. 16, 215218.[ISI][Medline]
Stengs, C. H., Klis, S. F., Huizing, E. H., and Smoorenburg, G. F. (1998). Cisplatin ototoxicity. An electrophysiological dose-effect study in albino guinea pigs. Hear. Res. 124, 99107.[ISI][Medline]
Thet, L. A., Parra, S. C., and Shelburne, J. D. (1984). Repair of oxygen-induced lung injury in adult rats. The role of ornithine decarboxylase and polyamines. Am. Rev. Respir. Dis. 129, 174181.[ISI][Medline]
Wada, H. Sugawara, M., Kobayashi, T., Hozawa, K., and Takasaka, T. (1998). Measurement of guinea pig basilar membrane using a computer-aided 3-dimensional reconstruction system. Hear Res. 120, 16.[ISI][Medline]
Williams-Ashman, H. G., and Canellakis, Z. N. (1979). Polyamines in mammalian biology and medicine. Perspect. Biol. Med. 22, 421453.[ISI][Medline]