Enhanced Ki]]ing of Acanthamoeba Cysts in vitro Using Dimethylsulfoxide Patrick P. R. Saunders, MDCM,l Eileen M. Proctor, PhD, 2 David F. Rollins, MD, FRCSC,l John S. F. Richards, MD, FRCSC 1 Purpose: Acanthamoeba keratitis is difficult to treat and requires prolonged therapy despite the well-documented in vitro effectiveness of a variety of drugs. The authors propose that this may be due to the cysts formed by the organism in response to hostile conditions. Consequently, the study concentrates on increasing penetration of drugs effective against the parasite into the cysts using dimethylsulfoxide (DMSO). Methods: The organism is forced to encyst in vitro on solid media by nutrient deprivation. In the first set of experiments, serial dilutions of a standard treatment regimen are applied to the organisms, and these treated cysts are then subcultured onto nutrientrich material and observed for growth. The experiments are then repeated with DMSO added to the serially diluted standards. In a second set of experiments, the effects of retreatment on a larger concentration of organisms is examined. Results: When applied to a cyst-only population of Acanthamoeba, none of three standard drugs, propamidine isethionate 0.1 %, neomycin 1%, or miconazole 1%, was cysticidal. When combined with DMSO 30%, propamidine isethionate was clearly cysticidal even in low dilution. This was confirmed by the retreatment experiments using a larger, standardized cyst population. Conclusion: The authors propose that DMSO is acting as a "carrier" for the propamidine isethionate and increases its penetration into the normally drug-resistant cyst form of the organism. Because DMSO has been used topically in the past and shown to be quite safe, this may be a viable new therapy for this difficult condition. Ophthalmology 1992;99; 1197-1200
Acanthamoeba keratitis is a cause of major morbidity. I Despite reports of medical cures, this disease often requires long-term therapy with toxic agents combined with penetrating keratoplasty.2-5 Recurrence in the graft is comOriginally received: September 30, 1991. Revision accepted: March 5, 1992. I Department of Ophthalmology, University of British Columbia, Vancouver. 2 Division of Medical Microbiology, Department of Pathology, University of British Columbia, Vancouver. Funding for this study was obtained from general departmental funding, University of British Columbia, Department of Ophthalmology. Drugs used in this study were donated by their respective manufacturers. The authors have no proprietary interests in the drugs used in this study. Reprint requests to Patrick P. R. Saunders, MDCM, Department of Ophthalmology, Faculty of Medicine, 2550 Willow St, Vancouver, British Columbia V5Z 3N9, Canada.
mon, and subsequent procedures can produce unsatisfactory results. 6 Despite excellent in vitro demonstrations of the effectiveness of a variety of drugs,I,4,6.7 the organism is very resistant to therapy. One possible basis for resistance is the organism's ability to form very durable cysts for survival in hostile conditions. Thus, when treated, the organism encysts for survival; when treatment is reduced or stopped, the organism excysts and continues its growth as a motile amoeba, Dimethylsulfoxide (DMSO), a common laboratory solvent, has long been known for capacity to act as a "carrier" for other molecules. 8,9 It has been used in dermatologic practice to enhance penetration of antifungal agents. It also has been used in ophthalmologic practice, both topically and intravitreally, for fungal infections. 10,11 In this study, a simple model to test drug effects on encysted Acanthamoeba po/yphaga was designed. In this
1197
Ophthalmology
Volume 99, Number 8, August 1992
study, we compare the effectiveness of a standard treatment regimen and the results when DMSO is added to the regimen.
Materials and Methods The organism used in this study was A. polyphaga ATCC 30461 (an axenically cultured organism) obtained from the British Columbia Centre for Disease Control. To raise a cyst-only population, all testing was performed on organisms obtained from liquid medium and pipetted onto solid, non-nutrient agar covered with an Escherichia coli lawn. After a period of approximately 8 days, all the E. coli were consumed and the organisms encysted. This was verified using an inverted microscope. Drugs to be tested were then added to the plate, which was incubated overnight, and washed free of drug the next morning (16 hours after addition). A block of agar (with the treated cysts on the surface) was then cut from the plate and inverted onto
a fresh E. coli-Iawned non-nutrient agar plate. These plates were then incubated and observed daily for any known signs of growth or activity such as consumption of E. coli or tracks in the agar. Drugs were tested as follows: propamidine isethionate 0.1 %, neomycin 1%, miconazole 1%, and DMSO 100%, 50%, 30%, 25%, 10%, 1%. A control plate also was subcultured as per the protocol. The neomycin, miconazole, and propamidine runs were then repeated with the drugs in their standard concentrations combined with DMSO. Serial dilutions of each drug also were prepared with DMSO, and the experiment was repeated. A second series of experiments was performed using a standardized concentration of organisms of 5 X 104/ml. This concentration was obtained by counting the number of organisms per milliliter in liquid media using a hemocytometer and then performing appropriate dilutions. The organisms were again pipetted onto lawned non-nutrient agar and observed to encyst. Because propamidine proved to be the only cysticidal agent in the first experi-
Table 1. Multiple Drug Exposure Protocol Test Organisms Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Fill Hole
Fill Hole
Fill Hole
Fill Hole
Fill Hole
Fill Hole
Fill Hole
1198
Saunders et al . Acanthamoeba and DMSO ment this series with a larger, standardized concentration of or~anisms was performed using only propamidine as the test drug mixed with DMSO. Test plates were treated with 30% DMSO combined with serial dilutions of propamidine isethionate ranging from 0.1 % (standard clinical concentration) to 0.00 I %. These plates were then incubated overnight, washed, and subcultured as per the previous protocol. The holes in the agar where the subculture blocks were removed wer: th:n refilled with non-nutrient agar to prevent drug leakmg m under the agar or diffusing through the cut edge into the remaining agar. The plates were then retreated with the same test dilution of propamidine combined with 30% DMSO, incubated, "Y,9shed, and resubcultured. Thus, the same organisms could be given multiple exposures to the desired agents (Table I). A total of eight drug applications were administered to the original test plates, and, after each application, subcultured blocks were observed for signs of growth. A continuously subcultured control also was produced to rule out attenuation of the organism by age and repeated subculture.
Table 3. Effect of Standard Agents Combined with 30% DMSO [Propamidine Isethionate] + 30%DMSO 0.1 %
Control
0.05%
2
+ 30% DMSO
[Miconazole] Control
1.0%
0.5%
0.25%
0.1 %
2
2
2
2
2
Days to growth
+ 30% DMSO
[Neomycin] Control
5.0%
4.0%
3.0%
2.0%
1.0%
2
2
2
2
2
2
Days to growth =
0.001 %
2
Days to growth
DMSO
0.01 %
dimethylsulfoxide.
Days to growth: number of days from subculture of treated agar block until growth observed. • No growth demonstrated.
Results In the first series of experiments, as shown in Table 2, none of the standard regimen drugs were cysticidal. Dimethylsulfoxide alone also was not cysticidal, although it did delay growth of the organisms compared with the control in 100%, 50%, and 30% concentrations. The only drug of the three standard agents to perform similarly was propamidine isethionate. Because 30% has been shown to be the highest concentration ofDMSO that is least irritating to the external eye,12 it was chosen as the test dilution of DMSO to be used in subsequent combination experiments. Table 3 shows that, when combined with 30% DMSO, neither neomycin nor miconazole was cysticidal; growth was identical to that of the control. However, the combination of propamidine and DMSO proved to be cysticid~l (defined as no growth after 15 days on lawned media; observation was, however, continued and as of this writing no growth had occurred in more than 12 months in the original plates) to as low as one-tenth standard concentration. When the experiments were repeated using a larger, standardized concentration of organisms, propamidine and DMSO were not cysticidal (Table 4). However, with repeated treatment, even quite dilute concentrations of drugs were clearly cysticidal (Table 5).
Discussion Because no animal model exists for in vivo testing of drugs used in the treatment of Acanthamoeba keratitis, in vitro testing on organisms has been the only method of developing improved therapies. This has been an unsatisfactory but necessary compromise. Because of the resistant nature of the disease, prolonged therapy with multiple toxic agents, with or without surgery, has been nece~sa~. The results of this study have shown that the combmatIOn of propamidine isethionate and DMSO is effective ag~inst low concentrations of Acanthamoeba cysts m a smgle treatment in vitro. Further, even with larger numbers of organisms, multiple applications of this combination are also effective in vitro. The question remains as to how relevant these in vitro findings are to the disease in vivo. Topical application of drugs to maintain a stable concentration in the cornea for 16 hours could be difficult, even if half-hourly administration and/or a saturated collagen shield was used. Also, drug penetration into organisms in the cornea is considerably less than into those on agar. This may be offset by Table 4. Effect of Propamidine Isethionate with 30% DMSO Against Standardized Concentration of Organisms (5 X 1Q4/m l)
Table 2. Effect of Standard Agents
Days to growth
[Propamidine Isethionate]
Control
Propamidine 0.1%
Miconazole 1.0%
Neomycin 1.0%
2
6
2
2
Days to growth: number of days from subculture of treated agar block until growth observed.
Days to growth DMSO
=
+ 30% DMSO
Control
0.1 %
0.09%
0.08%
0.075%
0.05%
2
2
2
2
2
2
dimethylsulfoxide.
Days to growth: number of days from subculture of treated agar block until growth observed.
1199
Ophthalmology
Volume 99, Number 8, August 1992
Table 5. Effect of Retreatment on Standardized Concentrations of Organisms (5 X 104/ml) [Propamidine Isethionate] Number of treatments 1 2 3
4 5 6
7
8
+ 30% DMSO
Control
0.1%
0.09%
0.08%
0.075%
0.05%
0.01%
+ + + + + + + +
+
+ +
+ +
+
+
+ + + +
+ + + + + + +/+/-
+ = Growth observed. - = No growth observed.
DMSO's carrier function, as we know that it can facilitate the entry of other drugs into the cornea. Ocular toxicity of propamidine and DMSO in combination is unknown, but individually they are both well tolerated topically. This study offers a possible new method of treatment for a disabling disease that may be more effective than standard treatment protocols. If the in vitro findings are applicable in vivo, the time required for treatment and thereby some of the subsequent local toxic effects may be substantially reduced. To prove or refute these findings, clinical trials will be set up.
References 1. Moore MB. Parasitic infections. In: Kaufman HE, Barron BA, McDonald MB, Waltman SR, eds. The Cornea. New York: Churchill Livingstone Inc, 1988; 271-97. 2. Wright P, Warhurst D, Jones BR. Acanthamoeba keratitis successfully treated medically. Br J Ophthalmol 1985;69: 778-82. 3. Berger ST, Mondino BJ, Hoft RH, et aI. Successful medical management of Acanthamoeba keratitis. Am J Ophthalmol 1990; 110:395-403.
1200
4. Driebe WT Jr, Stern GA. Successful medical management of Acanthamoeba keratitis [letter]. Am J Ophthalmol 1991;111:256-7. 5. Ishibashi Y, Matsumoto Y, Kabata T, et aI. Oral intraconazole and topical miconazole with debridement for Acanthamoeba keratitis. Am J Ophthalmol 1990;109:121-6. 6. Driebe WT Jr, Stern GA, Epstein RJ, et aI. Acanthamoeba keratitis: potential role for topical clotrimazole in combination chemotherapy. Arch Ophthalmol 1988; 106: 1196201. 7. Silvany RE, Dougherty JM, McCulley JP. Effect of contact lens preservatives on Acanthamoeba. Ophthalmology 1991 ;98:854-7. 8. Kligman AM. Topical pharmacology and toxicology of dimethyl sulfoxide-part 1. JAMA 1965;193:796-804. 9. Kligman AM. Dimethyl sulfoxide-part 2. JAMA 1965;193: 923-8. 10. Hanna C, Fraunfelder FT, Meyer SM. Effects of dimethyl sulfoxide on ocular inflammation. Ann OphthalmoI1977;9: 61-5. II. Y oshizumi MO, Banihashemi AR. Experimental intravitreal ketoconazole in DMSO. Retina 1988;8:210-5. 12. Grant WM. Toxicology of the Eye. 3rd ed. Springfield: Charles C. Thomas, 1986;352-5.
Acanthamoeba keratitis is a cause of major morbidity. I Despite reports of medical cures, this disease often requires long-term therapy with toxic agents combined with penetrating keratoplasty.2-5 Recurrence in the graft is comOriginally received: September 30, 1991. Revision accepted: March 5, 1992. I Department of Ophthalmology, University of British Columbia, Vancouver. 2 Division of Medical Microbiology, Department of Pathology, University of British Columbia, Vancouver. Funding for this study was obtained from general departmental funding, University of British Columbia, Department of Ophthalmology. Drugs used in this study were donated by their respective manufacturers. The authors have no proprietary interests in the drugs used in this study. Reprint requests to Patrick P. R. Saunders, MDCM, Department of Ophthalmology, Faculty of Medicine, 2550 Willow St, Vancouver, British Columbia V5Z 3N9, Canada.
mon, and subsequent procedures can produce unsatisfactory results. 6 Despite excellent in vitro demonstrations of the effectiveness of a variety of drugs,I,4,6.7 the organism is very resistant to therapy. One possible basis for resistance is the organism's ability to form very durable cysts for survival in hostile conditions. Thus, when treated, the organism encysts for survival; when treatment is reduced or stopped, the organism excysts and continues its growth as a motile amoeba, Dimethylsulfoxide (DMSO), a common laboratory solvent, has long been known for capacity to act as a "carrier" for other molecules. 8,9 It has been used in dermatologic practice to enhance penetration of antifungal agents. It also has been used in ophthalmologic practice, both topically and intravitreally, for fungal infections. 10,11 In this study, a simple model to test drug effects on encysted Acanthamoeba po/yphaga was designed. In this
1197
Ophthalmology
Volume 99, Number 8, August 1992
study, we compare the effectiveness of a standard treatment regimen and the results when DMSO is added to the regimen.
Materials and Methods The organism used in this study was A. polyphaga ATCC 30461 (an axenically cultured organism) obtained from the British Columbia Centre for Disease Control. To raise a cyst-only population, all testing was performed on organisms obtained from liquid medium and pipetted onto solid, non-nutrient agar covered with an Escherichia coli lawn. After a period of approximately 8 days, all the E. coli were consumed and the organisms encysted. This was verified using an inverted microscope. Drugs to be tested were then added to the plate, which was incubated overnight, and washed free of drug the next morning (16 hours after addition). A block of agar (with the treated cysts on the surface) was then cut from the plate and inverted onto
a fresh E. coli-Iawned non-nutrient agar plate. These plates were then incubated and observed daily for any known signs of growth or activity such as consumption of E. coli or tracks in the agar. Drugs were tested as follows: propamidine isethionate 0.1 %, neomycin 1%, miconazole 1%, and DMSO 100%, 50%, 30%, 25%, 10%, 1%. A control plate also was subcultured as per the protocol. The neomycin, miconazole, and propamidine runs were then repeated with the drugs in their standard concentrations combined with DMSO. Serial dilutions of each drug also were prepared with DMSO, and the experiment was repeated. A second series of experiments was performed using a standardized concentration of organisms of 5 X 104/ml. This concentration was obtained by counting the number of organisms per milliliter in liquid media using a hemocytometer and then performing appropriate dilutions. The organisms were again pipetted onto lawned non-nutrient agar and observed to encyst. Because propamidine proved to be the only cysticidal agent in the first experi-
Table 1. Multiple Drug Exposure Protocol Test Organisms Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Drug Application
=>
Incubate
=>
Wash
=>
Subculture
=>
Observe
Fill Hole
Fill Hole
Fill Hole
Fill Hole
Fill Hole
Fill Hole
Fill Hole
1198
Saunders et al . Acanthamoeba and DMSO ment this series with a larger, standardized concentration of or~anisms was performed using only propamidine as the test drug mixed with DMSO. Test plates were treated with 30% DMSO combined with serial dilutions of propamidine isethionate ranging from 0.1 % (standard clinical concentration) to 0.00 I %. These plates were then incubated overnight, washed, and subcultured as per the previous protocol. The holes in the agar where the subculture blocks were removed wer: th:n refilled with non-nutrient agar to prevent drug leakmg m under the agar or diffusing through the cut edge into the remaining agar. The plates were then retreated with the same test dilution of propamidine combined with 30% DMSO, incubated, "Y,9shed, and resubcultured. Thus, the same organisms could be given multiple exposures to the desired agents (Table I). A total of eight drug applications were administered to the original test plates, and, after each application, subcultured blocks were observed for signs of growth. A continuously subcultured control also was produced to rule out attenuation of the organism by age and repeated subculture.
Table 3. Effect of Standard Agents Combined with 30% DMSO [Propamidine Isethionate] + 30%DMSO 0.1 %
Control
0.05%
2
+ 30% DMSO
[Miconazole] Control
1.0%
0.5%
0.25%
0.1 %
2
2
2
2
2
Days to growth
+ 30% DMSO
[Neomycin] Control
5.0%
4.0%
3.0%
2.0%
1.0%
2
2
2
2
2
2
Days to growth =
0.001 %
2
Days to growth
DMSO
0.01 %
dimethylsulfoxide.
Days to growth: number of days from subculture of treated agar block until growth observed. • No growth demonstrated.
Results In the first series of experiments, as shown in Table 2, none of the standard regimen drugs were cysticidal. Dimethylsulfoxide alone also was not cysticidal, although it did delay growth of the organisms compared with the control in 100%, 50%, and 30% concentrations. The only drug of the three standard agents to perform similarly was propamidine isethionate. Because 30% has been shown to be the highest concentration ofDMSO that is least irritating to the external eye,12 it was chosen as the test dilution of DMSO to be used in subsequent combination experiments. Table 3 shows that, when combined with 30% DMSO, neither neomycin nor miconazole was cysticidal; growth was identical to that of the control. However, the combination of propamidine and DMSO proved to be cysticid~l (defined as no growth after 15 days on lawned media; observation was, however, continued and as of this writing no growth had occurred in more than 12 months in the original plates) to as low as one-tenth standard concentration. When the experiments were repeated using a larger, standardized concentration of organisms, propamidine and DMSO were not cysticidal (Table 4). However, with repeated treatment, even quite dilute concentrations of drugs were clearly cysticidal (Table 5).
Discussion Because no animal model exists for in vivo testing of drugs used in the treatment of Acanthamoeba keratitis, in vitro testing on organisms has been the only method of developing improved therapies. This has been an unsatisfactory but necessary compromise. Because of the resistant nature of the disease, prolonged therapy with multiple toxic agents, with or without surgery, has been nece~sa~. The results of this study have shown that the combmatIOn of propamidine isethionate and DMSO is effective ag~inst low concentrations of Acanthamoeba cysts m a smgle treatment in vitro. Further, even with larger numbers of organisms, multiple applications of this combination are also effective in vitro. The question remains as to how relevant these in vitro findings are to the disease in vivo. Topical application of drugs to maintain a stable concentration in the cornea for 16 hours could be difficult, even if half-hourly administration and/or a saturated collagen shield was used. Also, drug penetration into organisms in the cornea is considerably less than into those on agar. This may be offset by Table 4. Effect of Propamidine Isethionate with 30% DMSO Against Standardized Concentration of Organisms (5 X 1Q4/m l)
Table 2. Effect of Standard Agents
Days to growth
[Propamidine Isethionate]
Control
Propamidine 0.1%
Miconazole 1.0%
Neomycin 1.0%
2
6
2
2
Days to growth: number of days from subculture of treated agar block until growth observed.
Days to growth DMSO
=
+ 30% DMSO
Control
0.1 %
0.09%
0.08%
0.075%
0.05%
2
2
2
2
2
2
dimethylsulfoxide.
Days to growth: number of days from subculture of treated agar block until growth observed.
1199
Ophthalmology
Volume 99, Number 8, August 1992
Table 5. Effect of Retreatment on Standardized Concentrations of Organisms (5 X 104/ml) [Propamidine Isethionate] Number of treatments 1 2 3
4 5 6
7
8
+ 30% DMSO
Control
0.1%
0.09%
0.08%
0.075%
0.05%
0.01%
+ + + + + + + +
+
+ +
+ +
+
+
+ + + +
+ + + + + + +/+/-
+ = Growth observed. - = No growth observed.
DMSO's carrier function, as we know that it can facilitate the entry of other drugs into the cornea. Ocular toxicity of propamidine and DMSO in combination is unknown, but individually they are both well tolerated topically. This study offers a possible new method of treatment for a disabling disease that may be more effective than standard treatment protocols. If the in vitro findings are applicable in vivo, the time required for treatment and thereby some of the subsequent local toxic effects may be substantially reduced. To prove or refute these findings, clinical trials will be set up.
References 1. Moore MB. Parasitic infections. In: Kaufman HE, Barron BA, McDonald MB, Waltman SR, eds. The Cornea. New York: Churchill Livingstone Inc, 1988; 271-97. 2. Wright P, Warhurst D, Jones BR. Acanthamoeba keratitis successfully treated medically. Br J Ophthalmol 1985;69: 778-82. 3. Berger ST, Mondino BJ, Hoft RH, et aI. Successful medical management of Acanthamoeba keratitis. Am J Ophthalmol 1990; 110:395-403.
1200
4. Driebe WT Jr, Stern GA. Successful medical management of Acanthamoeba keratitis [letter]. Am J Ophthalmol 1991;111:256-7. 5. Ishibashi Y, Matsumoto Y, Kabata T, et aI. Oral intraconazole and topical miconazole with debridement for Acanthamoeba keratitis. Am J Ophthalmol 1990;109:121-6. 6. Driebe WT Jr, Stern GA, Epstein RJ, et aI. Acanthamoeba keratitis: potential role for topical clotrimazole in combination chemotherapy. Arch Ophthalmol 1988; 106: 1196201. 7. Silvany RE, Dougherty JM, McCulley JP. Effect of contact lens preservatives on Acanthamoeba. Ophthalmology 1991 ;98:854-7. 8. Kligman AM. Topical pharmacology and toxicology of dimethyl sulfoxide-part 1. JAMA 1965;193:796-804. 9. Kligman AM. Dimethyl sulfoxide-part 2. JAMA 1965;193: 923-8. 10. Hanna C, Fraunfelder FT, Meyer SM. Effects of dimethyl sulfoxide on ocular inflammation. Ann OphthalmoI1977;9: 61-5. II. Y oshizumi MO, Banihashemi AR. Experimental intravitreal ketoconazole in DMSO. Retina 1988;8:210-5. 12. Grant WM. Toxicology of the Eye. 3rd ed. Springfield: Charles C. Thomas, 1986;352-5.
