Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, April 2005, p. 1326-1330, Vol. 49, No. 4
0066-4804/05/$08.00+0 doi:10.1128/AAC.49.4.1326-1330.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Department of Chemistry,1 Center for Human Genetics and Molecular Pediatric Disease,2 Department of Pediatrics,3 Department of Microbiology, School of Medicine and Dentistry, University of Rochester, Rochester, New York4
Received 29 September 2004/ Returned for modification 22 October 2004/ Accepted 22 December 2004
|
|
|---|
|
|
|---|
Hoechst 33258 is a compound that binds both DNA (2, 30, 33) and RNA (1, 6, 8). Its chemical structure (Fig. 1) is relatively simple, so derivatives are readily synthesized (21, 31, 32, 37). Hoechst 33258 is an effective inhibitor (13) of in vitro self splicing of the group I intron isolated from the human pathogen Candida albicans (25). The group I self-splicing intron in the large-subunit (LSU) rRNA precursor is a potential drug target, because self splicing is necessary for the maturation of ribosomes (29). Group I introns are also found in the fungal pathogens Pneumocystis and Aspergillus (28, 35) but have not been found in mammalian genomes. Compounds such as pentamidine and fluorocytosine (25, 26), which have antimicrobial activity against Pneumocystis carinii and Candida spp., have also been shown to interfere with the self splicing of group I introns (23, 26). These results suggest that Hoechst 33258 may also inhibit growth of C. albicans and C. dubliniensis in vitro and P. carinii in vivo.
![]() View larger version (13K): [in a new window] |
FIG. 1. Chemical structures of pentamidine (top) and Hoechst 33258 (bottom). R represents OH for Hoechst 33258.
|
|
|
|---|
Instruments and general protocols. All radioactivity was quantified on a Molecular Dynamics PhosphorImager with ImageQuaNT version 4.1 software. The P. carinii precursor was transcribed and purified as described previously (9, 36).
Growth inhibition of Candida species. C. albicans isolates were obtained from clinical samples from the University of Rochester Medical Center, N.Y. (11). C. dubliniensis isolates were obtained from Centraalbureau voor Schimmelcultures (CBS) in The Netherlands.
Candida strains were grown overnight in YPD (yeast extract peptone dextrose) media; all media were prepared as described previously (34). Cells were harvested by centrifugation, washed with an equal volume of sterile water, and placed into YNB (yeast nitrogen base) medium at pH 4.5 to give an optical density at 540 nm (OD540) of 0.01. Serial dilutions of inhibitors were added to cultures, and they were incubated at 37°C with vigorous aeration for 8 to 12 h until the OD540 in the absence of inhibitor reached about 0.6 (range, 0.6 to 0.9). The OD540 of all cultures was measured, and 50% inhibitory concentrations (IC50s) for Candida spp. were determined from plots of the ratio of OD540 with and without inhibitor versus inhibitor concentration. The plots were fit with the SIGMAPLOT 2001s Logistic four-parameter curve fit program.
Splicing assays. Splicing assays on purified P. carinii-truncated precursor were completed as described previously (10). In a typical experiment, 2 nM precursor was refolded in buffer by incubation at 50°C for 3 min. The sample was placed at 37°C for at least 2 min to allow the temperature to equilibrate, and then 3 µl of this solution was added to 3 µl of a solution containing inhibitor and 2 mM pG in buffer at 37°C. Samples were incubated at 37°C for 1 h, and a 2/3 volume of stop buffer was added to quench the reactions. Products were separated on a denaturing 5% polyacrylamide gel and were quantified with a PhosphorImager (10). The IC50s were determined as described previously (5).
Mouse modeling of P. carinii pneumonia. A mouse model of P. carinii pneumonia (Pcp) was used to test the activity of Hoechst 33258 against P. carinii. For this model, 20-g CB-17 SCID mice (Jackson Laboratories, Bar Harbor, Maine) were inoculated intranasally with 80 µl of physiologic saline containing 5 x 105 P. carinii cysts that had been partially purified from freshly isolated infected mouse lungs. Twenty-four and 48 h after inoculation, the mice were given a 100-µg intraperitoneal (i.p.) dose of gentamicin (Gibco, Rockville, Md.) to prevent bacterial pneumonias which sometimes occur when SCID mice are infected by inoculation rather than by cohousing. Two weeks after inoculation, when infection was well established, the mice were treated with either Hoechst 33258, trimethoprim-sulfamethoxazole (TMP-SMX) (Gensiasicor Pharmaceuticals, Irvine, Calif.), or physiologic saline. Hoechst 33258 was diluted in saline and given daily at a dose of approximately 7.5, 22.5, or 37.5 mg/kg of body weight by i.p. injection. Dosages were calculated based on the average body weight of the experimental mice. TMP-SMX was also diluted in saline and given at a single daily dose of 15 to 20 mg of the TMP component/kg. Control mice received saline alone. These treatments were continued for 14 days, at which time the mice were sacrificed and their lungs removed for quantitation of P. carinii by real-time PCR using primers specific for the single-copy kex1 gene of P. carinii sp. f. muris (38). The statistical significance in the number of P. carinii in the various treatment groups was determined by two-tailed t test. Results were graphed with Sigma Plot.
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. Inhibition of growth of C. albicans and C. dubliniensis by small molecules
|
Pentamidine was tested for inhibition of growth because part of its antifungal activity is due to inhibition of group I intron self splicing (26). For example, pentamidine is usually more effective at slowing growth of C. albicans strains that harbor a group I intron than ones that do not. For strains tested here, intron-containing strains have IC50s of 0.6 and 6 µM and the intronless strains have IC50s of 0.9 and 100 µM (Table 1).
Hoechst 33258 slows growth of all C. albicans strains tested (Table 1). The two strains that do not harbor a group I intron had IC50s of 32 and 12 µM. The two strains that contain a group I intron had IC50s for Hoechst 33258 of 9 and 7 µM.
Anti-Candida compounds were also tested for inhibition of C. dubliniensis growth (Table 1). All reported isolates of C. dubliniensis have a group I intron in their LSU rRNA precursor (3). As shown in Table 1, the three C. dubliniensis strains tested had IC50s of
1.5 µM for Hoechst 33258, which is lower than the IC50 for any C. albicans strain.
Antimicrobial activity of Hoechst 33258 on in vivo infection with P. carinii. Preliminary toxicity studies were done in normal BALB/c mice given 5 and 25 mg of Hoechst 33258 i.p./kg once daily for 10 days. Mice displayed decreased movement about the cages after the first two doses, but thereafter they displayed normal activity and appearance over the remainder of the 10-day trial. Lung tissue was obtained from mice after 2 days of treatment with Hoechst 33258 and examined by fluorescent microscopy. Cell nuclei were fluorescent, indicating that the compound was reaching the lung.
Based on preliminary experience, four groups of five SCID mice each were treated once daily with saline (Group A), TMP-SMX (Group B), 7.5 mg Hoechst 33258/kg (Group C), or 37.5 mg Hoechst 33258/kg (Group D), beginning 14 days after the mice were inoculated with P. carinii. Treatments were given for 14 days, and the mice were sacrificed one day after the last dose and assayed for the presence of P. carinii. As shown in Fig. 2A, there was an almost 2 log10 drop in P. carinii in mice receiving 37.5 mg of Hoechst 33258/kg compared to saline-treated animals as determined by PCR. (Group A, 3.9 x 106 ± 2.7 x 106 kex1 copies/ml; Group D, 5.9 x 104 ± 9.6 x 104 kex1 copies/ml; P = 0.01). Mice receiving the low dose of Hoechst 33258 had organism counts similar to those of the saline-treated mice. PCR results were cross-checked by silver staining the lung homogenates to identify P. carinii cysts. Cyst counts were 2.4 x 106 in Group A, 8.9 x 104 in Group B, 2.0 x 106 in group C, and undetectable in Group D.
![]() View larger version (15K): [in a new window] |
FIG. 2. Copies of P. carinii as determined by real-time PCR from harvested mouse lung after treatment. Low, intermediate (Mid), and high doses of Hoechst 33258 are 7.5, 22.5, and 37.5 mg/kg, respectively. TMP-SMX dose is 15 to 20 mg of TMP/kg and 75 to 100 mg of the SMX component/kg.
|
To test whether Hoechst 33258 might affect group I intron splicing in P. carinii, its IC50 was measured (Fig. 3) for inhibition of self splicing of the purified truncated precursor rRNA developed by Testa et al. (36). As shown in Table 2, the IC50 is about 30 µM at 2 and 4 mM Mg2+, which is similar to expected Mg2+ concentrations in vivo (14, 19, 39). The IC50 at 4 mM Mg2+ is not very sensitive to addition of up to 29 mM nucleotides of Torula yeast bulk RNA, suggesting some specificity in binding to the group I intron (Table 2). Similar results have been reported (13) for the truncated precursor rRNA of C. albicans (Table 2).
![]() View larger version (37K): [in a new window] |
FIG. 3. Autoradiogram of a gel showing inhibition of self splicing of the P. carinii group I intron at 2 mM Mg2+ by Hoechst 33258. From top to bottom, the bands are 5' exon-intron-3' exon (i.e., truncated precursor), 5' exon-intron, intron-3' exon intron, ligated exons, and 5' exon. The plot shows percentages of intron versus the concentration of Hoechst 33258 (in micromolars) on a logarithmic scale.
|
|
View this table: [in a new window] |
TABLE 2. IC50 (µM) for inhibition of truncated precursor self splicing by Hoechst 33258
|
|
|
|---|
In vivo, Hoechst 33258 at a single daily dose of 37.5 mg/kg given for 14 days to SCID mice infected with P. carinii resulted in a 99% decrease in organisms compared to levels for saline-treated control mice. A lower dose of 22.5 mg/kg still resulted in a statistically significant drop in organisms. The activity of Hoechst 33258 in this model of Pcp was as effective, or more so, over a 14-day period than was TMP-SMX, which was given at a similar dosing range to validate the ability of this model to demonstrate an anti-pneumocystis effect for a given compound. Because no pharmacokinetic studies were employed, however, direct comparison of the potency of Hoechst 33258 to TMP-SMX is not appropriate. However, Yasuoka et al. showed that a single daily i.p. injection of a similar dose of TMP-SMX for 21 days was effective treatment for Pcp in a nude mouse model (40). Thus, we can conclude from our experimental design that Hoechst 33258 brought about a more rapid drop in P. carinii than did TMP-SMX.
Biochemical experiments show that Hoechst 33258 inhibits self splicing of the group I intron from C. albicans (13) and P. carinii (Fig. 3), suggesting that this could be an antifungal target for Hoechst 33258. In cell culture, however, C. albicans strains that harbor a group I intron, as well as those that do not, were both inhibited by Hoechst 33258, suggesting that the compound has targets in addition to the group I intron demonstrated in vitro (13). One possible mechanism for this activity is through binding to DNA topoisomerase I (4).
Uptake and toxicity have been measured for Hoechst derivatives in mammalian cells. For example, Hoechst 33258 is internalized in mammalian cells at 10-fold lower concentrations than Hoechst 33342 (15). The only structural difference between these compounds is the substitution of a hydroxyl group for an ethoxy group. While toxicity issues are observed with Hoechst 33342 at concentrations greater than 5 and 10 µM with HeLa cells (16) and myeloid cells (24), respectively, Hoechst 33258 will likely be less toxic. The effects of intravenous administration of Hoechst 33342 in mice have also been studied (24). High doses (0.3 M) of Hoechst 33342 were administered in these studies and resulted in morphological changes to the spleen and lungs. The IC50 data here suggest that much lower concentrations of Hoechst 33258 will be required to slow fungal growth. These considerations, and the fact that Candida infections often involve skin or mucous membranes that allow drug to be applied topically, may reduce toxicity. Likewise, Pcp can potentially be treated with aerosolized drug as a means to reduce toxicity. Such an approach has been utilized to reduce systemic hematologic toxicity from ribavirin in the treatment of respiratory syncytial virus pneumonia. Finally, the relatively simple structure of Hoechst dyes allows the synthesis of many derivatives that can be tested for efficacy and toxicity (21, 31, 32, 37). For example, oligonucleotides can be attached covalently, which might allow rational optimization toward RNA and DNA targets (12, 31, 37).
In summary, Hoechst 33258 has clearly demonstrable antimicrobial activity against P. carinii in vivo and against C. albicans and C. dubliniensis in vitro. The effect of Hoechst 33258 against these fungi may be due to interaction with DNA, RNA, or both. Determination of the cellular target should allow derivatization of Hoechst 33258 to enhance its effectiveness and specificity.
This work was supported by NIH grants GM22939 (D.H.T.) and HL071659 (F.G.).
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»