J. clin. Path., 1970, 23, 649-652 J Clin Pathol: first published as 10.1136/jcp.23.8.649 on 1 November 1970. Downloaded from Sensitivity of Hartmannella (Acanthamoeba) to 5-fluorocytosine, hydroxystilbamidine, and other substances

DAVID P. CASEMORE From the Department of Virology, Public Health Laboratory, Middlesbrough, Teesside

SYNOPSIS The effect of 24 anti-amoebic and other chemotherapeutic compounds on six strains of hartmannellid amoebae was studied by tissue culture, agar diffusion, and liquid axenic culture. Ofthe recognized anti-amoebic compounds only one inhibited growth of the amoebae, which were remarkable for their resistance.

Two compounds, hydroxystilbamidine isethionate and 5-fluorocytosine, showed somecopyright. amoebicidal activity while some others were inhibitory. http://jcp.bmj.com/ Hartmannella and Naegleria have both been Materials and Methods reported as causes of primary amoebic meningo- encephalitis but recent reports on the role of normally saprophytic amoebae in human disease AMOEBAE have tended to emphasize Naegleria as the chief Six strains ofamoebae were used. They conformed potential pathogen (eg, Carter, 1968; Apley, to Warhurst and Armstrong's description (1968) Clarke, Roome, Sandry, Saygi, Silk, and War- of Hartmannella castellanii. Their sources are on September 28, 2021 by guest. Protected hurst, 1970). However, the ability of some shown in Table I. hartmannellae to grow at 33-37'C, to produce degenerative changes in tissue culture, and to produce a lethal meningo-encephalitis in small Strain Number and Origin Source laboratory mammals suggests that they should as as pro- 29 Manchester Postmortem bronchial swab be regarded facultatively pathogenic, 324 Cirencester Throat swab posed by Griffin and others (Armed Forces B147 London Throat swab Institute of Pathology, 1968). 565 Leeds Throat swab 524 Coventry Throat swab The taxonomy of the hartmannellid group has M67 Middlesbrough Tissue culture contaminant not yet been resolved. Page (1967a and b, and personal communication) would give the amoebae Table I Sources ofamoebae and geographical used in this study the generic status Acantha- origins moeba: I do not feel competent to judge this point and have retained the nomenclature previously used. The potentially pathogenic COMPOUNDS UNDER TEST nature of some hartmannellid amoebae suggested Compounds were chosen because they had re- that an examination oftheir drug sensitivity would puted antiprotozoal activity, because they were be worthwhile. chemically related to known antiprotozoal agents, Received for publication 30 April 1970. because they were constituents of tissue-culture 650 David P. Casemore J Clin Pathol: first published as 10.1136/jcp.23.8.649 on 1 November 1970. Downloaded from

media that might be used for attempts at iso- EXAMINATION lation of amoebae, or because their mode of Although growth could often be detected by action against other organisms suggested possible naked-eye examination, especially in PPG activity against amoebae. medium, all cultures were examined micro- Some compounds used for the treatment of scopically after 24 hours and then at irregular entamoebiasis were not tested in this study in intervals during the week after inoculation. Tests vitro because they depend for their activity on a were controlled by inoculating each strain, in chemical reaction within the intestinal tract, parallel with the test, into the same medium free resulting in the local release of iodine, arsenic, from the compound under test. Results were etc. recorded as heavy, moderate, or scanty growth, Compounds were obtained as pure powders growth not detected, and, when marked differen- from the makers, or as pharmaceutical prepa- ces occurred,approximate proportions of cysts to rations for injection. Stock solutions were pre- trophozoites. Equivocal results, especially those pared, usually in distilled water, at a concentration involving few survivors after exposure to in- of 1,000 ,jg/ml. Each was tested for anti-amoebic hibitory compounds, were checked by recovering activity at an arbitrary concentration of 100 pkg/ the amoebae and subculturing to assess their ml. Compounds showing some effect at this viability. concentration were tested by titration in at least two of the test systems used; those showing no effect or only slight inhibition at 100 ,ug/ml by all INTERPRETATION three methods were not tested further. The relative amounts of growth recorded for tests and controls were interpreted as follows:

TISSUE CULTURE 1 No effect HeLa cells were grown and maintained using When test and control showed no difference in conventional media and methods. Sensitivity rate of multiplication, yield of trophozoites, or tests were performed in media without any of the rate of encystment. antibiotics normally used to control bacterial and mycotic growth. 2 Inhibition without amoebicidal effect copyright. When any of the following effects were noted in test cultures: (a) slower rate of multiplication, AGAR DIFFUSION (b) lower yield of trophozoites, or (c) more rapid Klebsiella-water-agar medium (Warhurst and encystment. Armstrong, 1968) was used; 1 ml of the tenfold concentration of the compound being tested was 3 Amoebicidal effect mixed with 9 ml of cooled, molten agar. Killed When (a) cysts recovered from (2) failed to Klebsiella was used to avoid inconsistencies that vegetate in medium free from the compoundhttp://jcp.bmj.com/ might arise when the Klebsiella was itself in- being tested, or (b) when the addition of a test hibited by the compound under test. compound to a dense, actively growing axenic culture of trophozoites caused loss of viability as indicated by subculture of trophozoites to fresh AXENIC CULTURE medium free from the compound being tested. Several peptone media, with and without glucose, were tried (Neff, 1957; Adam, 1964). Best growth on September 28, 2021 by guest. Protected was obtained in a thin layerof medium containing 4% proteose-peptone (Difco) and 1 % glucose, Results dissolved in tap water and sterilized by auto- claving (PPG medium). Table II shows the results ofpreliminary screening In most cases a tenfold concentration of the of the test substances or combinations at as compound was diluted with the medium and concentration of 100 ,ug/ml of each compound. then inoculated with the amoebae. In some cases, Only four substances, amphotericin B, paromo- however, the compound was added to a dense, mycin, hydroxystilbamidine isethionate, and actively growing culture. 5-fluorocytosine, were sufficiently active in the Tests for combined drug action were performed screening tests to justify further studies. The using chess-board titrations to give varying purpose of the study was to investigate the sensi- concentration ratios in PPG medium in a divided tivity pattern rather than to determine minimum Petri dish (Sneath and Stevens, 1967). inhibitory concentrations (MICs). Although the strains of amoebae showed slight differences in their growth characteristics and MICs there was INCUBATION no difference in their sensitivity patterns, so drug! All cultures were incubated at 37°C. Proteose- effects on the six strains are not shown separately. peptone glucose and agar-culture containers were Antimetabolites acting on the folic acid cycle- enclosed in polythene bags to reduce evaporation. sulphonamides, trimethoprim, pyrimethamine- to andothersubstances 651 SensitivityofHartmannella(Acanthamoeba) 5-fluorocytosine, hydroxystilbamidine, J Clin Pathol: first published as 10.1136/jcp.23.8.649 on 1 November 1970. Downloaded from tend to act synergistically. Different sulphon- each at concentrations of 100, 80, 60, 40, and 20 amides may vary in their efficacy to potentiate ,g/ml, failed to show any potentiation of the other substances acting on the folic acid cycle or inhibition indicated by the screen tests (Table II). in their suitability for use in this way. In any such Sulphadiazine by itself was only feebly in- combination the degree of potentiation may be hibitory against all the strains. However, Dr C. affected by the ratio of compounds used. Chess- G. Cuthbertson (personal communication) found board titrations of trimethoprim/sulphamethox- it to be effective in mouse protection tests with azole and pyrimethamine/sulphormethoxine, Hartmannella strains that he tested. Amphotericin B was shown by Carter (1969), in laboratory studies, to be effective against Naegleria at a concentration of 0-075 ug/ml. In Compound Effect on Amoebae at mytests(TableIl)it was inhibitorytoHartmannel- 100 ,ug/ml' la at 10 ,ug/ml, but during prolonged incubation, Acetarsol No effect which causes inactivation of the drug, survivors Diloxanide furoate No effect multiplied vigorously. The effect was greater than Oxytetracycline No effect Benzyl penicillin No effect could be explained by the effect of the bile salt Streptomycin sulphate No effect incorporated with the drug to effect solution. Neomycin No effect Kanamycin No effect Paromomycin, an aminoglycoside of the Paromomycin I (marked) neomycin group, is effective against Entamoeba Chloroquin Morphological changes only had a marked Nystatin No effect histolytica. Although it inhibitory Metronidazole VI effect on Hartmannella, the surviving amoebae Polymyxin B I (feeble) were viable and multiplied rapidly when sub- Sulphonamides' VI Trimethoprim lactate No effect cultured to medium free from paromomycin. Pyrimethamine sulphate No effect Hydroxystilbamidine isethionate is a diamidine Trimethoprim + sulphamethoxazole I Trimethoprim + polymyxin B I with properties similar to stilbamidine but with Pyrimethamine + sulphormethoxine I considerably less human toxicity. It is used in the Trimethoprim + pyrimethamine I treatment of various protozoal and mycotic Amrphotericin B I (marked) Hydroxystilbamidine AC infections and may be used for a prolonged

5-Fluorocytosine AC period (Snapper, Schneid, McVay, and copyright. Lieben, 1952). It may be incorporated into cells Table II Preliminary screening tests of24 compounds and remain active for long periods (Snapper, alone or in combination Schneid, Greenspan, and Lieben, 1950). Titration 'In tests oftwo drugs together, each was used at a concentration of in all three media showed marked inhibition by as 100 gg/ml. little as 0'125 for some strains (Table IV). A I = inhibition without amoebicidal effect ,ug/ml AC = amoebicidal concentration of 100 ,ug/ml was amoebicidal but VI = variable inhibition reactions was toxic to the HeLa cells. -= not tested 5-Fluorocytosine is a nucleotide analogue used http://jcp.bmj.com/ 'Separate tests on sulphanilamide, sulphadiazine, sulphasomizole, in the treatment of infections due to cryptococci sulphafurazole, sulphathiazole, sulphamethoxazole, and sulphor- and candida, including cryptococcal meningitis methoxine (Watkins, Campbell, Gardner-Medwin, Ingham, and Murray, 1969). It was inhibitory at 12-5 ,ug/ml and amoebicidal at 100 Hig/ml (Table IV). Compound Effect on Amoebae at Concentration

(tLg/ml) on September 28, 2021 by guest. Protected 100 10 1 Discussion Amphotericin B Marked Inhibition Feeble inhibition inhibition Sensitivity testing of amoebae in vitro is much tests on Paromomycin Marked Feeble No effect more difficult than bacteria. The longer inhibition inhibition period of growth, the varying ability of different strains of the same species to grow in the same Table III Further tests on the two inhibitory medium, and the ability of amoebae to encyst, all substances contribute to the difficulty. The correlation of sensitivity patterns shown by the three test systems used suggest that the techniques, as Effect on Amoebae at Concentration (jsgIml) outlined here, probably give a fair indication of 100 50 25 12 5 6 3 1t5 0-7 035 0 125 the sensitivity of these organisms and should be Hydroxystilbamidine of use in testing other compounds. isethionate AC I II I VI VI VI VI VI The results of the tests in vitro reported here 5-Fluorocytosine AC I I VI NE NE NE - - - cannot be extrapolated to indicate therapeutic action; for example, amphotericin had no Table IV Further tests on the two amoebicidal compounds amoebicidal activity in my tests, but Apley et al "Meaning of abbreviations as indicated at foot of Table II. (1970) have reported its use, with encouraging 652 David P. Casemore J Clin Pathol: first published as 10.1136/jcp.23.8.649 on 1 November 1970. Downloaded from effect, for one proven and two probable cases of Armed Forces Institute of Pathology (1968). Facultatively Patho- genic Amoebae: A Survey on Nomenclature with a Biblio- Naeglerial meningo-encephalitis. graphy, edited by J. L. Griffin. Armed Forces Institute of Animal studies are in progress to determine Pathology, Washington, DC. Carter, R. F. (1968). Primary amoebic meningo-encephalitis: whether the more active substances can protect clinical, pathological and epidemiological features of six mice against challenge infection of amoebae. fatal cases. J. Path. Bact., 96, 1, 1-25. Carter, R. F. (1969). Sensitivity to amphotericin B of a Naegleria sp. isolated from a case of primary amoebic meningo- encephalitis. J. clin. Path., 22, 470-474. Neff, R. 1. (1957). Purification, axenic cultivation and description of a soil amoeba, Acanthamoeba sp. J. Protozool., 4, I wish to thank Dr M. S. Pereira for supply- 176-182. ing five of the strains of amoebae, Dr J. Garrod of Page, F. C. (1967a). Taxanomic criteria for Limax amoebae, with descriptions of 3 new species of Hartmannella and 3 of Roche Products Ltd, and Mr S. L. Squires of Vahlkampfia. J. Protozool., 14, 499-521. May & Baker Ltd for the supply of drugs and Page, F. C. (1967b). Re-definition of the genus Acanthamoeba with much useful information, Boots Pure Drug Co. descriptions of three species. J. Protozool., 14, 709-724. Snapper, I., Schneid, B., Greenspan, E., and Lieben, F. (1950). On and Parke-Davis & Co. for the supply of drugs, the deposition ofstilbamidine and 2-hydroxystilbamidine in and Dr R. Blowers for help in the preparation of cytoplasm and nuclei of different organs and tumours. Bull. N. Y. Acad. Med., 26, 269-270. the typescript. Snapper, I., Schneid, B., McVay, L., and Lieben, F. (1952). Pharmacology and therapeutic value of diamidine deriva- tives-particularly of 2-hydroxystilbamidine. Trans. N. Y. Acad. Sci, 14, 269-271. Sneath, P. H. A., and Stevens, M. (1967). A divided petri dish for References use with multipoint inoculators. J. appl. Bact., 30, 495-497. Warhurst, D. C., and Armstrong, J. A. (1968). Study of a small Adam, K. M. G. (1964). A comparative study of Hartmannellid amoeba from mammalian cell cultures infarcted with 'Ryan amoebae. J. Protozool., 11, 423-30. Virus'. J. gen. Microbiol., 50, 207-215. Apley, J., Clarke, S. K. R., Roome, A. P. C. H., Sandry, S. A., Watkins, J. S., Campbell, M. J., Gardner-Medwin, D., Ingham, Saygi, G., Silk, B., and Warhurst, D. C. (1970). Primary H. R., and Murray, I. G. (1969). Two cases of crypto- amoebic meningoencephalitis in Britain. Brit. med. J., 1, coccal meningitis, one treated with 5-fluorocytosine. Brit. 596-599. med. J., 3, 29-3 1. copyright. http://jcp.bmj.com/ on September 28, 2021 by guest. Protected