ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Jan. 1997, p. 162–168 0066-4804/97/$04.0010 Copyright q 1997, American Society for Microbiology

Vol. 41, No. 1

Effects of Steroidal Allenic Phosphonic Acid Derivatives on the Parasitic Protists Leishmania donovani, Leishmania mexicana mexicana, and Pneumocystis carinii carinii DAVID H. BEACH,1 FRANKLIN CHEN,2 MELANIE T. CUSHION,2 ROGER S. MACOMBER,3 GEORGE A. KRUDY,3 MICHAEL A. WYDER,4 AND EDNA S. KANESHIRO4* Department of Microbiology and Immunology, State University of New York Health Science Center at Syracuse, Syracuse, New York, 132101; Department of Internal Medicine, University of Cincinnati College of Medicine, Cincinnati, Ohio 452672; and Department of Chemistry3 and Department of Biological Sciences,4 University of Cincinnati, Cincinnati, Ohio 45221 Received 10 September 1996/Returned for modification 14 October 1996/Accepted 25 October 1996

Several pathogenic fungi and protozoa are known to have sterols distinct from those of their mammalian hosts. Of particular interest as targets for drug development are the biosyntheses of the sterols of important parasites such as the kinetoplastid flagellates and the AIDS-associated opportunistic protist Pneumocystis carinii. These pathogens synthesize sterols with an alkyl group at C-24, and some have a double bond at C-22 of the side chain. Humans and other mammalian hosts are incapable of C-24 alkylation and C-22 desaturation. In the present study, three steroidal compounds with side chains substituted by phosphonyl-linked groups were synthesized and tested for their effects on Leishmania donovani and L. mexicana mexicana culture growth. The compounds inhibited organism proliferation at concentrations in micrograms per milliliter. The most potent inhibitors of this group of compounds were characterized by two ethyl groups at the phosphate function. Leishmania organisms treated with 17-[2-(diethylphosphonato)ethylidienyl]3-methoxy-19-norpregna-1,3,5triene exhibited reduced growth after transfer into inhibitor-free medium. Because there are currently no axenic methods available for the continuous subcultivation of P. carinii, the effects of these drugs on this organism were evaluated by two alternative screening methods. The same two diethyl phosphonosteroid compounds that inhibited Leishmania proliferation were also the most active against P. carinii as determined by the potent effect they had on reducing cellular ATP content. Cystic as well as trophic forms responded to the drug treatments, as evaluated by a dual fluorescent staining live-dead assay. Other modifications of steroidal phosphonates may lead to the development of related drugs with increased activity and specificity for the pathogens.

sterol inhibitors that exhibit synergistic activity, have been shown to inhibit kinetoplastid flagellate growth (8). Recently, it was demonstrated that the AIDS-associated opportunistic pathogen Pneumocystis carinii also synthesizes C28 and C29 sterols (5, 6, 12, 13). Although a method for the continuous subculturing of P. carinii has not yet been developed, procedures for obtaining organisms from rat lungs with consistent viabilities under short-term incubations in cell-free tissue culture media have been developed. Preparations for which purity has been well characterized have permitted direct biochemical analyses and metabolic studies with this organism (11). Limited subculturing of P. carinii on tissue culture cell lines has produced relatively pure populations of trophozoites that are currently used for a number of studies, including drug screening (15). Other approaches have been used to evaluate candidate drugs and to predict metabolic pathways. The ATP content of these organism preparations has been quantified and a protocol using ATP levels in organism populations has been used for in vitro drug screening (2). Phosphonates are organic compounds with stable phosphorus-carbon bonds resistant to oxidation and enzymatic hydrolysis (4, 9). A variety of phosphonates have been synthesized and developed as antibacterial and antiviral agents, e.g., against a simian immunodeficiency virus (21). Some phosphonate compounds are antimetabolites, including some that inhibit sterol biosynthesis (4). Thus, it seemed worthwhile to

The protozoan kinetoplastid flagellate parasites that cause human diseases such as leishmaniasis, Chagas’ disease, and African trypanosomiasis biosynthesize C28 and C29 sterols (8). These sterols are characterized by a methyl or ethyl group at C-24 of the sterol side chain. Some trypanosomatids also synthesize sterols with a double bond at C-22. Because mammals, including humans, are not capable of sterol C-24 alkylation or C-22 desaturation, the biosynthetic reactions present in the pathogen that lead to these side-chain modifications are attractive targets for chemotherapeutic intervention. It has been suggested that some endoparasites scavenge host or culture medium sterols such as cholesterol (a C27 sterol) and utilize these lipids for the formation of the bulk phase of cellular membrane bilayers (8, 12). However, organisms such as Leishmania species appear to synthesize their own “metabolic sterols” (8) for membrane functions (e.g., enzymes, pumps, and signal transduction components) that require the precise stereochemical features of the parasite-specific sterols at those special sites. Individual sterol inhibitors, or combinations of

* Corresponding author. Mailing address: Department of Biological Sciences, University of Cincinnati, Cincinnati, OH 45221-0006. Phone: (513) 556-9712. Fax: (513) 556-5280. E-mail: [email protected] .UC.EDU. 162

VOL. 41, 1997

EFFECTS OF STEROIDAL PHOSPHONATES

FIG. 1. Structures of compounds tested. (A) Mestranol, (17a)-3-methoxy-19norpregna-1,3,5(10)-triene-20-yn-17-ol (compound A); (B) 17-(2-phosphonatoethylidienyl)-3-methoxy-1,3,5(10)-estratriene (compound B), a steroidal phosphinate derived from mestranol; (C) 17-[2-(diethylphosphonato)ethylidienyl]3methoxy-19-norpregna-1,3,5-triene (compound C), a diethyl phosphono steroid derived from mestranol; (D) norethindrone, 17-hydroxy-19-norpregn-5(10)-en20-yn-3-one (compound D); (E) 17-[2-(diethylphosphonato)ethylidienyl]-19-norpreg-4-en-3-one (compound E), a diethylphosphono analog of norethindrone.

examine whether some steroidal phosphonates have antiLeishmania and/or anti-P. carinii activity. MATERIALS AND METHODS Syntheses of steroidal phosphonates. The steroidal phosphonic acid derivatives 17-[2-(phosphonato)ethylidienyl]-3-methoxy-1,3,5(10)-estratriene (molecular weight [MW], 374; compound B; Fig. 1B) and 17-[2-(diethylphosphonato)ethylidienyl]3-methoxy-19-norpregna-1,3,5-triene (MW, 430; compound C; Fig. 1C) were synthesized from (17a)-3-methoxy-19-norpregna-1,3,5(10)-triene-20yn-17-ol (mestranol, Searle, Skokie, Ill.; compound A; Fig. 1A) (14). Nitrogen was bubbled (100 ml/min) through a solution of 5 g (36.5 mmol) of phosphorus trichloride in 5 ml of dry dichloromethane while a solution of 507 mg (1.63 mmol) of mestranol in 10 ml of dichloromethane was added dropwise over 53 min at 238C. The resultant solution was stirred for 8 h at 258C; additional solvent was periodically added to replenish that which evaporated into the N2 stream. Rotary evaporation left 710 mg of a semisolid product. The crude product was redissolved in 5 ml of dioxane and added dropwise to 5 ml of 50% aqueous dioxane at 08C along with 175 mg (1.65 mmol) of sodium carbonate. After being stirred at 258C for 5 h, the pale yellow solution was evaporated to dryness (0.05 mm Hg pressure). The residue was taken up in 30 ml of hot dioxane, filtered to remove NaCl, and evaporated. Trituration of the crude phosphonic acid with acetone followed by two recrystallizations from dioxane gave 372 mg (61% of compound B, mp 193.5–194.58C; 1H nuclear magnetic resonance [NMR] spectrum [dimethyl sulfoxide (DMSO)-d6]; d 0.92 [s, 3H], 1.0–3.0 [envelope, 15H], 3.70 [s, 3H], 5.42 [d of t, JPH 5 5, JHH 5 4 Hz, 1H], 6.53 [br s, 1H], 6.60 [d, J 5 9 Hz, 1H], 7.10 [d, J 5 9 Hz, 1H], 7.22 [s, 2H]; infrared [IR] spectrum [KBr]: 3600–2000 [br], 1970 [m], 1616 [s], 1505 [s], 1150 [br vs], 1025 [br vs], 960 [vs]; mass spectrum [MS] [20 eV]: m/e 374 [mi, 3%], 294 [m-PO3H], 100%). Calculated composition for C21H27O4P was C, 67.34%; H, 7.27%. Found were C, 67.47%; H, 7.12%. Diethylchlorophosphite (0.68 g, 4.3 mmol) in 20 ml of dichloromethane was added dropwise to a stirred solution of pyridine (0.35 g, 4.4 mmol) and mestranol (979.3 mg, 3.15 mmol) in 40 ml of dichloromethane under nitrogen at 08C. The resulting mixture was stirred for 18 h at room temperature; washed successively with 1 N HCl, saturated aqueous sodium bicarbonate, and brine; dried over sodium sulfate; and evaporated, leaving 1.20 g (1.35 theoretical yield) of the crude allenic ester. The crude product was subjected to column chromatography (silica gel, eluted with ethyl acetate-benzene; 4/1, [vol/vol]) and recrystallized from methylene chloride-methylcyclohexane yielding 766.7 mg (56.8%) of compound C in two crops. Properties of the product were determined: Rf 5 0.29 (ethyl acetate, mp 96–97.58C; 1H NMR spectrum [CDCl3]: d 1.0–3.18 [envelope

163

including broad singlet at 1.00 and triplet {J 5 7 Hz} at 1.33, 24H], 3.78 [s, 3H], 4.13 [pentet, J 5 7, 4H], 5.42 [multiplet, 1H], 6.67 [broad singlet, 1.5H], 6.8 [singlet, 0.5H], 7.25 [d, J 5 8 Hz, 1H]; IR spectrum [CHCl3]: 3000 [s], 2940 [s], 1965 [m], 1610 [m], 1505 [s], 1240 [s], 1215 [sh], 1050 [vs], 1030 [vs], 970 [s] cm21; MS [70 eV]: 430 [mi], 83 [base]). Calculated composition for C25H35O4P was C, 69.74%; H, 8.19%. Found were C, 69.83%; H, 8.25% (14). 17-[2-(Diethylphosphonato)ethylidienyl]-19-norpreg-4-en-3-one (MW, 418; compound E; Fig. 1E) was synthesized from 17-hydroxy-19-norpregn-5(10)-en20-yn-3-one (norethindrone, Searle; compound D; Fig. 1D) in the presence of base (1, 14). Leishmania promastigote cultures. L. mexicana mexicana (LV732) and L. donovani (MHOM/SD/62/1S, clone 2D) cells were grown axenically at 258C in 5 ml of the Steiger and Steiger RE-III medium (20). Cultures were initiated by inoculation of 5 3 106 cells per culture tube. Test steroidal compounds (compounds A to E) were prepared individually in stock solutions of sterile DMSO at a concentration of 1 mg/ml. The final concentration of DMSO in all experimental cultures was ,0.1%; DMSO up to 0.5% was previously shown to have no effect on Leishmania culture growth (7). Initially, final drug concentrations of 0.5 to 10 mg/ml were tested for their effect on organism proliferation. The higher concentrations which killed the organisms and which required the largest amount of compounds were not reevaluated in subsequent experiments. The cultures were microscopically examined at day 3 or day 4 of culture growth to determine whether live cells were present, and cell density was roughly estimated (e.g., 1111, high-density live organisms observed, versus 1, low density, and 2, no live organisms observed). Actual cell numbers were determined by counting cells in a hemocytometer at day 5 and later time points. Individual values presented for culture density were from different experiments using separate organism cultures (not replicates of aliquots taken from the same culture). Organisms from growth-inhibited cultures were subcultured in fresh drug-free media to test the long-term effects of the compounds. Concentrations of drugs that inhibited growth by 40 to 60% of controls were used to grow L. donovani and L. mexicana mexicana for 7 days. Cell densities were analyzed on an aliquot of cultures to verify the level of inhibition, then 0.1 ml of each was transferred into fresh RE-III medium without inhibitor. After 7 days of culture growth, cell densities were analyzed to determine whether the growth-inhibited organisms recovered from the effects of the drugs. P. carinii carinii organisms. Organisms used for the ATP assays were obtained either from (i) male viral antibody-positive Brown Norway or Long Evans rats (originally from Charles River, Madison, Wis.) in which the infection was induced by immunosuppression with injections of methylprednisolone acetate (4 mg/week, subcutaneous) or (ii) viral antibody-negative CD rats (Charles River, Hollister, Calif.) inoculated with form 1 of P. carinii carinii (3). Aqueous ammonium chloride was used to lyse host erythrocyte contaminants, then organisms were purified using supplemented RPMI 1640 media, gravity sedimentation, and centrifugation (2). These preparations contained approximately 5% cystic forms. The ATP values represent the averages of populations of organisms composed mainly of trophic forms. Thus, the values for ATP content obtained in response to these compounds reflect the effects of these compounds mainly on trophozoite metabolism. For fluorescent probe-based viability assays, P. carinii carinii was obtained using the Boylan & Current infected rat model (1). Female Lewis viral antibodynegative rats (Harlan Sprague Dawley, Indianapolis, Ind.) were immunosuppressed with methylprednisolone acetate (Depo-Medrol, Upjohn Co., Kalamazoo, Mich.) and were twice intratracheally inoculated with previously frozen organism preparations containing 106 to 107 mixed life cycle stages according to previously described methods (1, 11). After 8 to 10 weeks of immunosuppression, moribund rats were sacrificed and their lungs were perfused, excised, and cut into small pieces. The P. carinii carinii organisms were isolated by homogenization (Stomacher, Tekmar, Cincinnati, Ohio) according to the procedures described earlier in which the mucolytic agent glutathione was included to disaggregate organisms from host cells and other P. carinii carinii organisms. Purification involved sieving, a series of centrifugation steps at different speeds, followed by membrane filtration (11). These preparations are .94 to 100% pure, as demonstrated by several microscopic and biochemical criteria (11). Routinely, these preparations contained 10 to 30% cystic forms. Greater percentages of cysts were recovered from lungs with greater parasite burdens. ATP assay. Isolated organisms were suspended in a supplemented RPMI 1640 medium containing 20% fetal bovine serum (FBS) (Sigma), and other additives, pH 7.5 to 8.0, 380 mosM (2). Drugs were added to the culture medium in DMSO (final concentration of DMSO was ,0.2%, vol/vol) and 108 organisms (as total nuclei) per ml were added to 2 ml of the culture medium in multiwell plates. Each drug concentration was assayed in triplicate. The final ATP content was expressed as the average of 9 values (three readings per well). After 24 and 48 h incubations at 378C in a 10% CO2 atmosphere, the ATP content was determined by the luciferin-luciferase assay using an AutoLumat LB 953 luminometer (Wallac, Inc., Gaithersburg, Md.) as detailed previously (2). The nuclei of all organisms were counted, and the ATP contents were expressed as nanomoles per 108 P. carinii carinii nuclei. The effects of the compounds on P. carinii carinii ATP content were compared with the ATP contents of vehicle controls (with primary solvent, without drug) and expressed as percent of control. Quench controls were also included to evaluate the effects of the highest drug concentrations used, and vehicle controls were tested on the enzymatic luciferase-luciferin reaction.

164

BEACH ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

Viability assessment by fluorescence microscopy. The Calcein acetoxy methyl ester (CAM) plus propidium iodide (PI) (Molecular Probes, Inc., Eugene, Oreg.) double-stain live-dead technique (10) was used to determine the percent of intact organisms with nonspecific esterase activity (assumed to be viable) versus those organisms in which the membrane integrity was sufficiently compromised so as not to exclude the nucleic acid intercalator PI (assumed to be nonviable). These viability assessments were made after 24-h incubation of the mixed life cycle population with various concentrations of the drugs tested. For these analyses, immediately after organisms were isolated and purified, they were resuspended into RPMI 1640 medium (Gibco BRL, Grand Island, N.Y.) supplemented with 10% FBS and 200 U of penicillin G per ml and 200 mg of streptomycin per ml (Pen-Strep) (Sigma Chemical Co., St. Louis, Mo.) at a density of 5 3 106 to 1 3 107 organisms per ml. The organisms were also tested in media lacking the FBS supplement. The compounds tested were dissolved in DMSO as the primary solvent. Stock solutions in DMSO were prepared and added to P. carinii carinii in incubation medium (RPMI 1640 supplemented with Pen-Strep). The final concentration of DMSO was kept constant in all incubations at 0.5% (vol/vol), including the vehicle controls. Incubation of organisms with drugs was carried out at 378C for 24 h in a 10% CO2 atmosphere. Cells were analyzed on a Nikon Optiphot microscope equipped with quartz epifluorescence optics (Nikon, Melville, N.Y.) and a 1003 objective. Excitation wavelength was 450 to 490 nm (Nikon B-2A filter). For each sample, .25 organisms in .10 different oil immersion fields were scored. The numbers of organisms (trophozoites and cysts) present in each sample before and after drug treatment were not determined. Organisms exhibiting only green fluorescence were scored as live; those exhibiting only red and moribund organisms exhibiting both green and red fluorescence were scored as dead. Data were expressed as percent of intact organisms that exhibited only green fluorescence (viable) compared to untreated controls in the absence of a drug. The terms “live,” “dead,” “viable,” and “viability,” used below are assumed and here defined as values obtained by this dual fluorescence live-dead analysis.

TABLE 1. Effects of steroidal allenic phosphonic acid derivatives on L. donovani promastigote culture growth

A

3b

4b

5

7

9

17.9 (100)

32.4 (100) 22.6 (100) 13.4 (41) 10.6 (33) 10.9 (48) 9.6 (30) 6.7 (21)

34.6 (100)

0

111

0.5 1.0

1 1

5.8 (32) 7.7 (43)

5.0 10.0

1 1

5.8 (32) 4.8 (27)

25.8 (75) 22.4 (65) 12.3 (36) 10.0 (29)

B 0

8.3 (59)

1.0

9.6 (68)

0

111

0.5

2

3.8 (35) 3.2 (23)

1.0

2

5.0 10.0

2 2

0.0 (0) 1.3 (9) 0.0 (0) 0.0 (0)

0

111

0.5 1.0

1 1

8.6 (48) 8.6 (48)

5.0 10.0

1 1

6.7 (38) 3.8 (21)

D

E

a b

14.1 (100)

0.5

C

RESULTS Effects of steroidal phosphonates on Leishmania promastigote proliferation in vitro. All steroids increased lag phase to some degree; in some cases, culture density eventually reached maximal densities equivalent to those of controls. At the higher concentrations, all five steroidal compounds (Fig. 1) decreased final cell densities of both L. donovani (Table 1) and L. mexicana mexicana cultures (Table 2). The most potent growth inhibitors were compounds C and E. At the higher concentrations tested, these two compounds killed all organisms by day 5. The recovery of organisms inhibited by approximately 50% of control growth following transfer into drug-free medium was observed for all compounds except compound C. The long-term deleterious effect of compound C on both L. donovani and L. mexicana mexicana organisms was exhibited by the reduced growth of these subcultures (Table 3). Effects on P. carinii carinii cellular ATP content. The steroid mestranol (compound A) and its phosphonic acid derivative, compound B, had no inhibitory effects on cellular ATP levels (Table 4). In marked contrast, the diethylphosphonate analog (compound C) clearly had anti-P. carinii carinii activity, as indicated by the dramatic reduction of cellular ATP levels. Thus, P. carinii carinii ATP content and Leishmania proliferation declined in a similar manner in response to this series of steroidal compounds. Norethindrone, compound D, had little effect on P. carinii carinii ATP content, whereas its diethylphosphono analog, compound E, caused a depletion of cellular ATP. This was consistent with the observations on the effects of mestranol (compound A) and its diethylphosphono analog (compound C) on P. carinii carinii cellular ATP levels and Leishmania growth. Viability assay. The dual fluorescent staining assay and ATP assays evaluated two different parameters: one, the effectiveness in reducing viability as detected by the staining properties of organisms (live-dead assay, which determines the final effects of an agent), and the other, cellular metabolism (ATP content, which measures dynamic changes occurring within live

Cell densitya (106 cells/ml [% drug-free control]) at the following culture age (days)

Compound (mg/ml)

10.9 (100) 14.1 (100)

17.9 (100)

0

111

0.5

11

1.0

11

5.0 10.0

1 2

12.9 (100) 7.7 (100) 9.6 (75) 7.7 (55) 5.8 (45) 7.0 (50) 0.0 (0) 0.0 (0)

30.2 (100) 22.6 (100) 20.2 (67) 11.9 (53) 19.0 (63) 25.6 (100) 30.2 (100) 22.6 (100) 4.3 (17) 5.6 (19) 6.3 (28) 1.4 (6) 0.0 (0) 0.0 (0) 0.0 (0) 32.4 (100) 22.6 (100) 27.8 (86) 21.1 (65) 12.0 (53) 20.1 (62) 7.7 (24)

34.6 (100) 41.4 (120) 37.0 (107) 21.2 (61) 12.3 (36)

24.8 (100) 30.2 (100) 22.6 (100) 19.2 (77) 19.0 (63) 12.1 (49) 20.0 (66) 12.0 (53) 1.4 (6) 0.0 (0)

2 to 111, no organisms to high densities of organisms observed. Gross estimates made by microscopic observations of cultures.

organisms). Although these two assays provide very different information, in some cases (e.g., a homologous series that includes a potent lethal agent) they can provide correlative data. Thus, in the homologous series of compounds A, B, and C, compound C had greater detrimental effects than compounds A and B, as evaluated by both assays (Fig. 2). Compound C was found to result in the eventual killing of P. carinii carinii which was preceded by a rapid depletion of cellular ATP. Cystic and trophic forms of P. carinii carinii tested in serumfree media appeared to respond similarly to the steroidal phosphonates and both stages exhibited greater sensitivity to the

VOL. 41, 1997

EFFECTS OF STEROIDAL PHOSPHONATES

TABLE 2. Effects of steroidal allenic phosphonic acid derivatives on L. mexicana mexicana promastigote culture growth Cell densitya (106 cells/ml [% drug-free control]) at the following culture age (days)

Compound (mg/ml)

A

3b

4b

7

9

11.2 (100)

20.4 (100) 20.8 (100) 12.3 (60) 16.8 (82) 14.6 (70) 16.8 (82) 10.0 (49)

38.4 (100)

0

111

0.5 1.0

11 1

4.7 (42) 5.8 (51)

5.0 10.0

1 1

5.8 (51) 5.8 (51)

19.0 (50) 22.4 (58) 19.0 (50) 15.7 (41)

B 0

17.0 (100)

0.5

8.8 (52)

1.0 5.0

8.0 (47) 7.2 (42)

C 0

1111

17.6 (100) 17.0 (100)

0.5

1111

1.0

11

18.4 (105) 5.6 (33) 5.1 (29) 3.2 (19)

5.0 10.0

1 1

D 0

111

0.5 1.0

11 1

7.6 (68) 7.7 (69)

5.0 10.0

1 1

8.6 (77) 9.6 (86)

E

a b

0.0 (0) 0.0 (0) 11.2 (100)

0

1111

18.4 (100) 17.0 (100)

0.5

1111

1.0

111

16.0 (87) 8.8 (52) 11.2 (61) 8.0 (47)

5.0

11

10.0

1

5.8 (31) 4.8 (28) 1.0 (0)

29.8 (100) 20.8 (100) 21.8 (74) 14.8 (71) 26.9 (91) 24.3 (83)

Compound Concn of druga

Control A B C D E

0 1.0 0.5 0.5 1.0 0.5

106 Cells/mlb

21.9 20.1 19.3 9.8 19.9 18.9

L. mexicana mexicana % Concn of 106 % Cells/mlb Controlb Controlb drug

100 92 88 45 91 86

0 1.0 0.5 1.0 1.0 1.0

19.0 18.5 18.4 8.0 17.9 18.0

100 97 97 42 99 95

a Concentration (mg/ml) of drug in cultures causing growth between 40 and 60% of controls prior to transfer of organisms into drug-free RE-III medium. b Values are from a single experiment.

To obtain some indication of whether compound C would be toxic to humans, it was tested on .50 human cancer cell lines. The concentration to obtain 50% inhibition of culture growth was between 1 and 10 mg/ml, and the 50% lethal dose values were between 15 and 45 mg/ml for most of the human cell lines treated with compound C (not shown).

45.9 (100) 29.8 (100) 20.8 (100) 42.6 (93) 22.0 (74) 6.4 (14) 7.7 (26) 5.6 (27) 0.0 (0) 0.0 (0) 20.4 (100) 20.8 (100) 13.4 (66) 15.7 (76) 13.2 (63) 15.6 (76) 17.9 (88)

TABLE 3. Growth of drug-inhibited Leishmania organisms in drug-free medium L. donovani

5

165

DISCUSSION

38.4 (100) 28.0 (73) 29.1 (76) 22.4 (58) 21.2 (55)

48.2 (100) 29.8 (100) 20.8 (100) 41.4 (86) 24.3 (83) 34.7 (72) 21.8 (74) 10.4 (50) 4.8 (10) 9.0 (30) 1.0 (0)

1 to 1111, low to high densities of organisms observed. Gross estimates made by microscopic observations of cultures.

diethyl derivatives. The 50% lethal concentration values obtained in serum-free media were as follows: compound A, .3.22 3 1024 M; compound B, .1.60 3 1023 M; compound C, 1.5 3 1024 M; compound D, .2.94 3 1024 M; and compound E, 2.6 3 1024 M. However, the viability assay itself gave relatively inconsistent results for any given compound. This was particularly evident when tests were performed on these lipidic drugs in media containing FBS, which contains a variety of lipids, including cholesterol (Table 5).

The steroidal phosphonates synthesized in this study were shown to have inhibitory activity on two species of Leishmania and on rat-derived P. carinii carinii. The most potent compounds against all three organisms were compounds C and E. Compound C, but not compound E, was shown to have lethal effects on Leishmania organisms; prolonged disruption of cellular functions of Leishmania species was demonstrated by the inhibition of growth after drug-exposed organisms were inoculated into drug-free medium. The relatively nontoxic nature of the parent steroidal compounds on Leishmania species and P. carinii carinii (compared to the activity seen with their diethylphosphonate analogs) demonstrated that the inhibitory activity is mainly in the phosphono group attached to the D ring of the steroid nucleus. Because compound B lacks the two ethyl groups and it is not toxic, the toxicity of compounds C and E appears to be associated with the presence of the ethyl groups on the phosphonyl phosphorus. The prolonged inhibitory and lethal effects observed with compound C compared to those of compound E on both L. donovani and L. mexicana mexicana culture growth suggest that the bulkier methoxy group at C-3 of compound C also has activity. This function may interfere with normal lipid bilayer packing and hence disrupt normal membrane function. The mechanism of action of compound C or E on Leishmania species and P. carinii carinii is not known; however, it is conceivable that reactions in the pathogens that involve normal sterol side chain metabolism were affected. More likely, these compounds may have inserted into the organism’s membranes and interfered with functions in which sterols are required. Inhibitors (16–18) targeted at the biosynthesis of C-24-alkylated sterols of parasites have been examined in other systems (8). Certain heteroatom-substituted sterols, e.g., 24(25)-epiminolanosterol, that can inhibit S-adenosylmethionine-dependent sterol D24 methyltransferase do not appear promising as antiparasite drugs because these compounds insert into mammalian cell membranes and perturb host membrane function. Other sterol C-24 alkylation inhibitors may be less toxic to mammals, e.g., Zeneca Pharmaceuticals’ triazole D0870 (22). Although compound C was a potent inhibitor of the test organisms in this study, it also inhibited growth and viability of

166

BEACH ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

TABLE 4. Effects of steroidal allenic phosphonic acid derivatives on Pneumocystis ATP contenta Drug concn (mg/ml) Compound and incubation time (h)

0

1

10

nmol of ATP/108 nuclei

% Control

nmol of ATP/108 nuclei

% Control

nmol of ATP/108 nuclei

24 48

78.5 91.1

100.0 100.0

109.0 70.3

138.9 77.2

24 48 72

67.2 60.3 54.5

100.0 100.0 100.0

72.3 64.2 67.4

24

87.2

100.0

24 48

78.5 91.1

24

87.2

100 % Control

nmol of ATP/108 nuclei

% Control

107.5 69.6

136.9 76.4

94.2 101.9

120.0 111.9

107.6 106.5 123.7

71.4 63.2 57.3

106.3 104.8 105.1

68.0 54.0 69.8

101.2 89.6 128.1

27.9

32.0

19.2

22.0

0.2

0.2

100.0 100.0

106.7 76.8

135.9 84.3

105.8 88.1

134.8 96.7

103.6 71.4

132.0 78.4

100.0

29.8

34.2

24.9

28.6

1.8

2.1

A

B

C D

E a

See text or Fig. 1 for names and structures of compounds. Data represent the results of single experiments performed in triplicate in which a compound was tested on the same organism preparation at different concentrations.

human cancer cells and viability of a diploid human fibroblast culture (unpublished), suggesting that it may have unacceptable toxicity levels in humans. Whether there is differential sensitivity of rapidly dividing cells to compound C in vivo in whole animals remains to be determined. Further modifications of these steroidal phosphonate molecules may result in safer compounds with greater antiparasite activities. In addition to these steroids, other nonsteroidal compounds with phosphorus-carbon bonds were tested on P. carinii carinii. Three oxaphospholene phosphonates (2-ethoxy-5,5-dimethyl4-[2,4-nitrophenylthio]-1,2-oxaphosphol-3-ene-2-oxide,

2-ethoxy-5,5-pentamethylene-4-[phenylseleno]-1,2-oxaphosphol-3-ene-2-oxide and 3,5-di-t-butyl-2,5-dihydroxy-1,2oxaphosphol-3-ene-2-oxide) and an oxaphospholene phosphinate (4-bromo-3,5-di-t-butyl-5-methoxy-2-phenyl-1,2oxaphosphol-3-ene-2-oxide) reduced P. carinii carinii ATP content. The levels to which these oxaphospholenes reduced ATP were comparable to those observed with steroidal compounds C and E (not shown). Hence phosphonates such as these oxaphospholenes, which are transition state analogs of butenolides (a family of biologically active compounds), also have potent anti-P. carinii carinii activity.

FIG. 2. Comparison of the effects of the homologous series, compounds A, B, and C, on P. carinii carinii viability as determined by cellular ATP level and the microscopic assay employing the CAM plus PI (tested in serum-free media). Organisms were isolated from infected rat lungs and incubated with a drug for 24 h. Broken lines, ATP content; solid lines, CAM-PI assay; squares, compound A; circles, compound B; triangles, compound C.

VOL. 41, 1997

EFFECTS OF STEROIDAL PHOSPHONATES

167

TABLE 5. In vitro effects of steroidal allenic phosphonic acid derivatives on Pneumocystis viability evaluated by the CAM-PI dual fluorescence assay Compound and developmental form

Viabilitya at the following drug concn (mg/ml) 0

1

5

10

25

50

100

200

Trophic Cystic

100 (144) 100 (22)

116 (153) 34 (13)

102 (115) 79 (13)

111 (152) 68 (13)

98 (158) 78 (19)

93 (128) 57 (18)

77 (64) 25 (76)

NDb ND

Trophic Cystic

100 (113) 100 (25)

120 (200) 166 (51)

160 (139) 168 (23)

163 (160) 168 (27)

123 (190) 100 (16)

129 (115) 148 (20)

131 (71) 86 (13)

169 (157) 114 (12)

Trophic Cystic

100 (271) 100 (35)

90 (175) 101 (24)

73 (169) 83 (19)

48 (168) 37 (19)

57 (181) 36 (20)

69 (203) 56 (18)

70 (131) 50 (17)

54 (176) 29 (15)

Trophic Cystic

100 (158) 100 (17)

97 (156) 103 (14)

89 (97) 86 (10)

82 (140) 109 (8)

71 (115) 77 (11)

87 (140) 123 (7)

84 (154) 143 (6)

ND ND

Trophic Cystic

100 (129) 100 (26)

85 (192) 90 (29)

70 (97) 43 (11)

75 (103) 119 (12)

55 (103) 55 (16)

50 (88) 57 (17)

65 (124) 55 (31)

48 (93) 57 (17)

A

B

C

D

E

a Viability is expressed as percent of controls (number of organisms analyzed are in parentheses). Values obtained for compound F were from experiments using two separate organism preparations; the rest were tested at different concentrations on single organism preparations. See Fig. 1 for names and structures of compounds. b ND, not determined.

Although growth inhibition studies similar to those performed on L. donovani and L. mexicana mexicana are not yet possible for P. carinii carinii, the effects of drugs on this pathogen can be assessed by alternate approaches. The effects of drugs on primary cultures of P. carinii carinii grown with mammalian cell lines (15) have identified some effective anti-P. carinii carinii agents. In the present study the ATP content assay indicated that the steroidal compounds (compounds C and E) effective against L. donovani and L. mexicana mexicana were also effective against P. carinii carinii. The ATP assay is exquisitely sensitive in detecting physiological and metabolic responses to a drug. However, primary cultures of P. carinii carinii and the ATP assay are not able to provide as reliable information on long-lasting lethal effects, as demonstrated in this study by subculturing compound C-treated Leishmania species in drug-free medium. The observations on P. carinii carinii viability made in this study suggest that care should be taken in using this assay, not only with drugs that emit interfering fluorescence wavelengths, but also with lipidic drugs, for which analyses may also be problematic. CAM is supposed to indicate live cells by diffusing through cell surface membranes where they become cleaved by intracellular nonspecific esterases. However, CAM esterases can remain active even in some dead cells and thus give a false live score. The exclusion of DNA intercalating dyes such as PI or ethidium homodimer supposedly guarantees against false live scores because these dyes are expected to penetrate only dead cells whose cell surface membranes have become permanent to these dyes. Hence the double-stain approach is superior to single dye-exclusion tests. However, testing of lipidic drugs by this assay is complicated by the fact that exogenous lipids such as sterols readily insert into the cell surface membranes (19). This could cause greater membrane stability and decreased permeability to the DNA stains even in dead cells. This notion is consistent with the observation that the addition of FBS (which contains lipids) to the test medium gave a higher percentage of viable cells than media lacking FBS. This is also consistent with the observation that many cells did not stain

with PI, even at the high steroid drug concentrations that killed Leishmania cells and depleted P. carinii carinii ATP. ACKNOWLEDGMENTS We thank the National Cancer Institute’s Developmental Therapeutics Program for the assays on human cancer cell lines. This study was supported by NIH grants UO1 AI31702 and RO1 AI38758. REFERENCES 1. Benn, W. R. December 1971. U.S. patent 3,631,075. 1a.Boylan, C. J., and W. L. Current. 1992. Improved model of Pneumocystis carinii carinii pneumonia: induced laboratory infections in Pneumocystis-free animals. Infect. Immun. 60:1589–1597. 2. Chen, F., and M. T. Cushion. 1994. Use of an ATP bioluminescent assay to evaluate viability of Pneumocystis carinii from rats. J. Clin. Microbiol. 32: 2791–2800. 3. Cushion, M. T., M. Kaselis, S. L. Stringer, and J. R. Stringer. 1993. Genetic stability and diversity of Pneumocystis carinii infecting rat colonies. Infect. Immun. 61:4801–4813. 4. Engel, R. 1983. Phosphonic acids and phosphonates as antimetabolites, p. 97–138. In R. L. Hilderbrand (ed.), The role of phosphonates in living systems. CRC Press, Inc., Boca Raton, Fla. 5. Florin-Christensen, M., J. Florin-Christensen, Y.-P. Wu, L. Zhou, A. Gupta, H. Rudney, and E. S. Kaneshiro. 1994. Occurrence of specific sterols in Pneumocystis carinii carinii. Biochem. Biophys. Res. Commun. 198:236–242. 6. Furlong, S. T., J. A. Samia, R. M. Rogers, and J. A. Fishman. 1994. Phytosterols are present in Pneumocystis carinii. Antimicrob. Agents Chemother. 38:2534–2540. 7. Goad, L. J., G. G. Holz, Jr., and D. H. Beach. 1985. Effect of the allylamine antifungal drug SF 86-327 on the growth and sterol synthesis of Leishmania mexicana mexicana promastigotes. Biochem. Pharmacol. 34:3785–3788. 8. Haughan, P. A., and L. J. Goad. 1991. Lipid biochemistry of trypanosomatids, p. 312–328. In G. Coombs (ed.), Biochemical protozoology. Taylor & Francis, London, U.K. 9. Hilderbrand, R. L. 1983. The role of phosphonates in living systems. CRC Press, Boca Raton, Fla. 10. Kaneshiro, E. S., M. A. Wyder, Y.-P. Wu, and M. T. Cushion. 1993. Reliability of calcein acetoxy methyl ester and ethidium homodimer or propidium iodide for viability assessment of microbes. J. Microbiol. Methods 17:1–16. 11. Kaneshiro, E. S., M. A. Wyder, L. H. Zhou, J. E. Ellis, D. R. Voelker, and S. G. Langreth. 1993. Characterization of Pneumocystis carinii preparations developed for lipid analysis. J. Eukaryot. Microbiol. 40:805–815. 12. Kaneshiro, E. S., J. E. Ellis, K. Jayasimhulu, and D. H. Beach. 1994. Evi-

168

13. 14. 15.

16. 17.

BEACH ET AL. dence for the presence of “metabolic sterols” in Pneumocystis: identification and initial characterization of Pneumocystis carinii sterols. J. Eukaryot. Microbiol. 41:78–85. Kaneshiro, E. S., J. E. Ellis, L. H. Zhou, H. Rudney, A. Gupta, K. Jayasimhulu, K. D. R. Setchell, and D. H. Beach. 1994. Isoprenoid metabolism in Pneumocystis carinii carinii. J. Eukaryot. Microbiol. 41:93S. Krudy, G. A. 1982. Preparation and properties of some oxaphospholenes. Ph.D. dissertation. University of Cincinnati, Cincinnati, Ohio. Lee, C. H., N. L. Bauer, M. M. Shaw, M. M. Durkin, M. S. Bartlett, S. F. Queener, and J. W. Smith. 1993. Proliferation of rat Pneumocystis carinii on cells sheeted on microcarrier beads in spinner flasks. J. Clin. Microbiol. 31:1659–1662. Mercer, E. I. 1991. Sterol synthesis inhibitors: their current status and modes of action. Lipids 26:584–597. Nes, W. D., P. K. Hanners, and E. L. Parish. 1986. Control of fungal sterol C-24 transalkylation: importance to developmental regulation. Biochem. Biophys. Res. Commun. 139:410–415.

ANTIMICROB. AGENTS CHEMOTHER. 18. Parish, E. J., P. K. Hanners, and W. D. Nes. 1987. Synthesis and biological evaluation of fungal bioregulators of sterol biosynthesis, p. 103–105. In P. K. Stumpf, J. B. Mudd, and W. D. Nes (ed.), The metabolism, structure and function of plant lipids. Plenum Publishing Inc., New York, N.Y. 19. Sleight, R. G., M. A. Mehta, and E. S. Kaneshiro. 1994. Uptake and metabolism of fluorescent lipid analogs by Pneumocystis carinii. J. Eukaryot. Microbiol. 41:111S. 20. Steiger, R. F., and E. Steiger. 1977. Cultivation of Leishmania donovani and Leishmania braziliensis in defined media: nutritional requirements. J. Protozool. 24:437–441. 21. Tsai, C. C., K. E. Follis, A. Sabo, T. W. Beck, R. F. Grant, N. Bischofberger, R. E. Benveniste, and R. Black. 1995. Prevention of SIV infection in macaques by (R)-9-(2-phosphonylmethoxypropyl)adenine. Science 270:1197– 1199. 22. Yamada, H., T. Tsuda, T. Watanabe, M. Ohashi, K. Murakami, and H. Mochizuki. 1993. In vitro and in vivo antifungal activities of D0870, a new triazole agent. Antimicrob. Agents Chemother. 37:2412–2417.