Abstract
Most drugs used for prevention and treatment of Pneumocystis jirovecii pneumonia target enzymes involved in the biosynthesis of folic acid, i.e., dihydropteroate synthase (DHPS) and dihydrofolate reductase (DHFR). Emergence of P. jirovecii drug resistance has been suggested by the association between failure of prophylaxis with sulfa drugs and mutations in DHPS. However, data on the occurrence of mutations in DHFR, the target of trimethoprim and pyrimethamine, are scarce. We examined polymorphisms in P. jirovecii DHFR from 33 patients diagnosed with P. jirovecii pneumonia who were receiving prophylaxis with a DHFR inhibitor (n = 15), prophylaxis without a DHFR inhibitor (n = 11), or no prophylaxis (n = 7). Compared to the wild-type sequence present in GenBank, 19 DHFR nucleotide substitution sites were found in 18 patients with 3 synonymous and 16 nonsynonymous mutations. Of 16 amino acid changes, 6 were located in positions conserved among distant organisms, and five of these six positions are probably involved in the putative active sites of the enzyme. Patients with failure of prophylaxis, including a DHFR inhibitor, were more likely to harbor nonsynonymous DHFR mutations than those who did not receive such prophylaxis (9 of 15 patients versus 2 of 18; P = 0.008). Analysis of the rate of nonsynonymous versus synonymous mutations was consistent with selection of amino acid substitutions in patients with failure of prophylaxis including a DHFR inhibitor. The results suggest that P. jirovecii populations may evolve under selective pressure from DHFR inhibitors, in particular pyrimethamine, and that DHFR mutations may contribute to P. jirovecii drug resistance.
Pneumocystis jirovecii (human-derived Pneumocystis) is a common opportunistic pathogen in immunocompromised patients that causes severe life-threatening pneumonia. Most drugs used for prevention and treatment of P. jirovecii pneumonia (PcP) target enzymes involved in the biosynthesis of folic acid. The sulfa drugs sulfamethoxazole (SMZ), sulfadoxine (SD), and dapsone (D) inhibit the dihydropteroate synthase (DHPS), whereas the diaminopyrimidines trimethoprim (TMP) and pyrimethamine (PM) are inhibitors of the dihydrofolate reductase (DHFR). DHPS is involved in the condensation of p-aminobenzoic acid and 6-hydroxymethyl-dihydropterin pyrophosphate to produce dihydropteroate. Dihydropteroate is subsequently converted into dihydrofolate by the dihydrofolate synthase. DHFR catalyzes the NADP-dependent reduction of dihydrofolate to tetrahydrofolate. Two antifolates are most often administrated together. Cotrimoxazole, the antifolate combination of TMP and SMZ, is the first-choice drug, whereas fansidar, the combination of PM and SD, is rarely used.
The emergence of P. jirovecii drug resistance has been suggested recently by the association between failure of sulfa prophylaxis and mutations in the gene encoding DHPS (5). The most frequent DHPS mutations are at nucleotide positions 165 and 171, leading to an amino acid change at positions 55 (Thr to Ala; mutation 1 [M1]) and 57 (Pro to Ser; M2). They are observed either as a single or a double mutation (M3). According to the three-dimensional structure of Escherichia coli DHPS, these mutations are located in the putative sulfa binding site of P. jirovecii DHPS. Moreover, similar mutations in other microbial pathogens are known to confer sulfa resistance (18, 19).
Alteration of DHFR enzyme is a common resistance mechanism in clinically important microbial pathogens, such as Plasmodium falciparum (15) and Streptococcus pneumoniae (10). Two studies have investigated the possibility of mutations in P. jirovecii DHFR gene. Ma et al. (7) found only one synonymous DHFR mutation in clinical specimens from 32 patients, 22 of them having received TMP-SMZ as prophylaxis (7 patients) or treatment of a previous PcP episode (15 patients). Takahashi et al. (17) reported four mutations in P. jirovecii DHFR from 27 patients, only three of them having been previously exposed to TMP/SMZ for treatment of a prior PcP episode. Two of the mutations were nonsynonymous but were not associated with prior exposure to TMP-SMZ. Thus, thus far there is no evidence that there was a change in enzyme protein sequence due to treatment with TMP and that TMP has affinity for P. jirovecii DHFR. This is consistent with experiments in animal models that suggested that the antipneumocystis activity of TMP-SMZ is due only to SMZ (20). However, we hypothesized that the use of PM may be effective on P. jirovecii DHFR and that accumulation of DHFR mutations may have occurred in patients who developed PcP infection while receiving this drug. To investigate this possibility, we analyzed clinical specimens from PcP patients who experienced failure of various types of prophylaxis, including PM.
(Preliminary results of this study were presented in a conference report [12].)
MATERIALS AND METHODS
Specimens and patients.
Bronchoalveolar lavage samples were obtained from 33 patients with confirmed PcP who were hospitalized between 1993 and 1996 in Lausanne University Hospital in Lausanne, Switzerland (3 patients), and in five different hospitals in Lyon, France (30 patients). Two patients had a subsequent PcP episode which was excluded from the present study. The 30 patients from Lyon were also included in one of our previous studies (13).
Specific information on demographic, clinical characteristics, and chemoprophylaxis were obtained from patients' medical charts. Patients were considered as having received anti-Pneumocystis prophylaxis if they have received TMP-SMZ, PM-SD, D, pentamidine (P), atovaquone (A), or PM combined with A during the 3 months preceding the date of diagnosis of PcP. The duration of the prophylaxis ranged from 7 days to the entire 3-month period. PM-SD was the first-choice regimen for prophylaxis in three hospitals of Lyon and, as described and discussed elsewhere (13), its dosage was suboptimal. A failure of prophylaxis was defined as the development of PcP in patients who received anti-Pneumocystis prophylaxis.
Amplification of P. jirovecii DHPS and DHFR.
P. jirovecii DNA was extracted from bronchoalveolar lavage samples with Qiamp blood kit (Qiagen, Hilden, Germany). A region of 318 bp spanning the putative drug binding site of the DHPS, in which P. jirovecii mutations were observed, was amplified by using the primers and thermal cycling conditions described elsewhere (13). The full length of the coding region of DHFR (663 bp, including a 42-bp intron) was amplified by using primers FR208 and FR1038 described elsewhere (7). The PCR was carried out with reagents of the HotStar Taq DNA polymerase kit (Qiagen). PCR conditions included a hot start for 10 min at 95°C, followed by 35 cycles consisting of 30 s at 94°C, 30 s at 52°C, and 1 min 72°C. The reaction ended with 5 min of extension at 72°C. The PCR yielded a 858-bp fragment, which was analyzed by electrophoresis in a 1.5% agarose gel containing ethidium bromide (0.5 μg/ml).
DHPS genotyping.
DHPS genotype was determined by using PCR-single strand conformation polymorphism, as previously described (13).
Cloning and sequencing of PCR products from DHFR amplification.
PCR products were purified by using a QiaQuick PCR purification kit (Qiagen) and cloned by using a TOPO TA cloning kit (Invitrogen, The Netherlands) according to the manufacturer's instructions. Five recombinant plasmids for each PCR product were purified by using Qiamp plasmid extraction kit (Qiagen), and both strands of each clone were sequenced by using an ABI Prism 310 automated sequencer and a Big Dye terminator DNA sequencing kit (both from Perkin-Elmer Biosystems). The absence of PCR-induced errors was excluded by a control experiment consisting of amplifying four known DHFR sequences present in recombinant clones, cloning the PCR products, and sequencing five subclones. All sequences obtained were identical to the initial sequence.
Molecular evolution of DHFR.
The Nei Gojobori model implemented in the MEGA software version 2.1 (6) was used to analyze the molecular evolution of DHFR gene by comparing dN to dS (dN is the number of observed substitutions at nonsynonymous sites divided by the total number of nonsynonymous sites in DHFR gene, and dS is the same ratio for synonymous substitutions).
Statistical analysis.
The Fisher exact test was used to assess the association between P. jirovecii DHFR mutations and failure of prophylaxis. A P value of <0.05 was considered statistically significant.
Nucleotide sequence accession numbers.
The accession numbers of the new DHFR sequences obtained in this study are AY685184 through AY685198.
RESULTS
Patients.
Of 33 patients included in the study, 25 (75.8%) were infected with human immunodeficiency virus (HIV) and 8 (24.2%) were HIV negative with various disorders causing immunodeficiency. The median age was 36 years (range, 20 to 60 years). The CD4-cell count at PcP episode was documented only for the HIV-infected patients, with a median of 35.5 cells/μl (range, 0 to 99). Eighteen patients (54.6%) were not receiving a DHFR inhibitor at the PcP occurrence (seven had no prophylaxis at all, and two had received A, two had received D, seven had received P; patients 1 to 18, Table 1). Fifteen (45.4%) patients were receiving prophylaxis, including a DHFR inhibitor (two received PM-A, six received TMP-SMZ, and seven received PM-SD; patients 19 to 33).
TABLE 1.
P. jirovecii DHFR mutations and DHPS genotype in 33 patients with PcP according to the prophylaxis regimen
| Patient no.a | Prophylaxis | DHPS genotypeb | DHFR allele | No. of recombinant plasmids | DHFR mutationa
|
|
|---|---|---|---|---|---|---|
| Nonsynonymous | Synonymous | |||||
| 1 | No | WT | 1 | 5 | - | - |
| 2 | No | WT | 1 | 5 | - | - |
| 3 | No | WT | 1 | 5 | - | - |
| 4 | No | WT | 1 | 5 | - | - |
| 5 | No | M2 | 1 | 5 | - | 277 (T to C) |
| 312 (T to C) | ||||||
| 6 | No | M2 | 1 | 5 | - | 312 (T to C) |
| 7* | No | M2 | 1 | 3 | 107 (T to A)↔F36C | - |
| 2 | 2 | 194 (T to C)↔L65P | - | |||
| 8 | A | M3 | 1 | 5 | - | - |
| 9 | A | M3 | 1 | 5 | - | - |
| 10 | D | M1 | 1 | 5 | - | 312 (T to C) |
| 11 | D | M3 | 1 | 5 | - | 312 (T to C) |
| 12 | P | WT | 1 | 5 | - | - |
| 13 | P | WT | 1 | 5 | - | - |
| 14 | P | WT | 1 | 5 | - | - |
| 15 | P | WT | 1 | 5 | - | - |
| 16 | P | M3 | 1 | 5 | - | - |
| 17 | P | M3 | 1 | 5 | - | - |
| 18 | P | M2 | 1 | 5 | 500 (A to T)↔D153V | 312 (T to C) |
| 19 | TMP-SMZ | WT | 1 | 5 | - | - |
| 20 | TMP-SMZ | WT | 1 | 5 | - | - |
| 21 | TMP-SMZ | WT | 1 | 5 | - | - |
| 22 | TMP-SMZ | WT | 1 | 5 | - | 277 (T to C) |
| 23 | TMP-SMZ | WT | 1 | 5 | 110 (G to C)↔S37T | - |
| 24 | TMP-SMZ | WT | 1 | 5 | 110 (G to C)↔S37T | - |
| 25 | PM-A | WT | 1 | 5 | 110 (G to C)↔S37T | - |
| 26 | PM-A | M3 | 1 | 5 | 110 (G to C)↔S37T | - |
| 27 | PM/SD | M2 | 1 | 5 | - | 312 (T to C) |
| 28 | PM/SD | M2 | 1 | 5 | - | 312 (T to C) |
| 29* | PM-SD | M2 | 1 | 2 | 514 (A to G)↔I158V | 312 (T to C) |
| 2 | 1 | 235 (G to A)↔V79I | 312 (T to C) | |||
| 3 | 2 | - | 312 (T to C) | |||
| 30* | PM-SD | M2 | 1 | 4 | 632 (T to C)↔Y197L | 36 (A to G) |
| 312 (T to C) | ||||||
| 2 | 1 | - | 312 (T to C) | |||
| 31* | PM-SD | M2 | 1 | 2 | 40 (A to G)↔T14A | 312 (T to C) |
| 77 (C to A)↔P26Q | ||||||
| 2 | 3 | - | 312 (T to C) | |||
| 32* | PM-SD | M2 | 1 | 2 | 156 (G to A)↔M52I | 312 (T to C) |
| 188 (A to G)↔E63G | ||||||
| 2 | 1 | 472 (A to G)↔T144A | 312 (T to C) | |||
| 553 (A to G)↔K171E | ||||||
| 3 | 2 | - | 312 (T to C) | |||
| 33* | PM-SD | M2 | 1 | 1 | 358 (T to C)↔S106P | 277 (T to C) |
| 422 (A to G)↔E127G | 312 (T to C) | |||||
| 2 | 1 | 550 (A to C)↔R170G | 312 (T to C) | |||
| 3 | 3 | - | 312 (T to C) | |||
Patients that harbored two or three DHFR alleles, which suggests coinfection with several P. jirovecii genotypes, are indicated by an asterisk.
M1, A to G at nucleotide position 165, leading to the replacement of Thr by Ala at position 55; M2, C to T at position 171, leading to the replacement of Pro by Ser at position 57; M3, both M1 and M2; WT, wild type.
Relative to the wild-type sequence present in GenBank (accession no. AF090368). -, No mutation.
Mutations in P. jirovecii DHFR.
Sequences of DHFR coding region obtained were compared to the wild-type sequence present in GenBank (accession number AF090368). No P. jirovecii DHFR mutations were observed among 15 patients (no. 1 to 4, 8, 9, 12 to 17, and 19 to 21). We found 19 substitution sites among the remaining 18 patients with three synonymous (silent) and 16 nonsynonymous mutations. The three synonymous mutations were at nucleotide positions 36, 277 (in the unique intron), and 312 and were identified, respectively, in one patient (patient 30), three patients (patients 5, 22, and 33), and 12 patients (patients 5, 6, 10, 11, 18, and 27 to 33). The 16 nonsynonymous mutations were identified in 11 patients (patients 7, 18, 23 to 26, and 29 to 33). Only 1 of these 16 mutations, located at position 110, was observed in several patients (patients 23 to 26). Each of the 15 remaining nonsynonymous substitution sites was observed only once in a single patient. Six patients harbored both synonymous and nonsynonymous mutations (patients 18 and 29 to 33). The positions of the 16 amino acid changes in P. jirovecii DHFR are shown in Fig. 1, together with an alignment of DHFRs from other distantly related organisms, including Homo sapiens. Ten of the sixteen positions (positions 14, 26, 36, 52, 63, 79, 144, 153, 171, and 197) were fully conserved (identical) in Pneumocystis isolated from different hosts. Compared to DHFR of the other organisms, three of these ten positions were fully conserved (positions 26, 144, and 153), two were strongly conserved (positions 65 and 79), and one was weakly conserved (position no. 36). Figure 1 also shows the positions of the amino acid changes that confer resistance to DHFR inhibitors in P. falciparum (15), Streptococcus pneumoniae (10), and Staphylococcus aureus (4). The amino acid changes in P. jirovecii at positions 37 and 106 matched with changes involved in drug resistance of P. falciparum and S. pneumoniae.
FIG. 1.
Alignment of amino acid sequence of P. jirovecii DHFR with those from distantly related organisms. Positions of amino acid changes that confer resistance to PM or TMP in P. falciparum (15), S. pneumoniae (10), and S. aureus (4) are shown in boldface. Amino acid changes we found in P. jirovecii are also shown in boldface. The scale at the top of alignment corresponds to P. jirovecii protein. Identical, strongly or weakly conserved residues are indicated, respectively, by asterisks, double or single points at the bottom of the sequence. Dashes indicate missing residues. Pc, P. carinii. GenBank accession numbers for the wild-type sequences: P. jirovecii, AF090368; rat PC, M26495; H. sapiens, J00140; S. aureus, S32014; S. pneumoniae, AF055727; and P. falciparum, J03772. Partial DHFR amino acid sequences from mouse and rabbit-derived Pneumocystis were provided by J. Kovacs.
Association of DHFR mutations with failure of prophylaxis.
Patients who developed PcP while receiving prophylaxis including a DHFR inhibitor were more likely to be infected with P. jirovecii strain harboring a DHFR amino acid change than those who did not receive such prophylaxis (nine of fifteen patients versus two of eighteen; P = 0.008 [two-tailed the Fisher exact test]). Seven of the nine patients with failure of prophylaxis, including a DHFR inhibitor and nonsynonymous DHFR mutations, received PM (combined with A or as part of PM-SD [patients 25 to 33]); only two received TMP (as part of TMP-SMZ [patients 23 and 24]).
P. jirovecii DHPS genotyping and association of DHPS mutations with failure of prophylaxis.
The P. jirovecii DHPS genotype was determined for the 33 specimens of the 33 patients by using PCR-single strand conformation polymorphism. Of these 33 patients, 18 (55%) harbored a mutated P. jirovecii DHPS (M1, one patient; M2, 11 patients; M3, six patients [Table 1]). Mutation M2 was associated with failure of PM-SD prophylaxis (7 of 7 patients versus 4 of 26 [P < 0.0001]).
Analysis of molecular evolution of DHFR.
The DHFR alleles observed among the 33 patients were classified into three groups: alleles from patients without prophylaxis (8 alleles, patients 1 to 7, Table 1), alleles from patients with prophylaxis not including a DHFR inhibitor (11 alleles, patients 8 to 18), and alleles from patients with prophylaxis including a DHFR inhibitor (23 alleles, patients 19 to 33). The alleles from patients without prophylaxis or from patients with a prophylaxis not including a DHFR inhibitor had a dS greater than the dN (Table 2), indicating that substitutions were mostly neutral. In contrast, sequences from patients with prophylaxis including a DHFR inhibitor had a dS equal to the dN, indicating that an elevated number of substitutions led to an amino acid change. Thus, the evolution of DHFR seems accelerated in the group of patients receiving prophylaxis including a DHFR inhibitor.
TABLE 2.
Estimation of dS and dN among P. jirovecii DHFR alleles from patients with different prophylaxis regimens by using a Nei Gojobori model implemented in MEGA software
| Group of DHFR alleles | No. of patients | No. of alleles | dS | dN | dN/dS |
|---|---|---|---|---|---|
| No prophylaxis | 7 | 8 | 0.0031 | 0.001 | 0.32 |
| Prophylaxis without a DHFR inhibitor | 11 | 11 | 0.0032 | 0.0004 | 0.125 |
| Prophylaxis with a DHFR inhibitor | 15 | 23 | 0.004 | 0.004 | 1.0 |
DISCUSSION
In the present study involving 33 patients diagnosed with PcP, 19 DHFR nucleotide substitution sites were found in P. jirovecii from 18 patients with 3 synonymous and 16 nonsynonymous mutations. The observed P. jirovecii DHFR mutations are reported here for the first time, with the exception of the synonymous mutation at position 312 (7, 17). According to the alignment of DHFRs from distantly related microorganisms, 6 of the 16 nonsynonymous mutations were at positions that were conserved, and two matched with changes involved in drug resistance of other organisms. Based on the crystal structure of DHFR of Pneumocystis carinii infecting rats (2, 3), five of the six mutations conserved among distantly related organisms are located at positions that are predicted to be part of the active sites of the enzyme (positions 26, 36, 37, 65, and 144). The amino acid change at position 26 is located in the putative signature for the folate substrate binding, which was identified at residues 19 to 27 (7). These observations suggest that the distribution of the DHFR mutations is not random and thus that these mutations may have been selected by the pressure of a DHFR inhibitor and might confer some level of resistance. Such selection is also supported by the association between these mutations and the failure of prophylaxis, including a DHFR inhibitor, mainly PM. These mutations are predicted to alter the structure of the enzyme and reduce its affinity for DHFR inhibitors. Our comparison of rates of substitutions at synonymous and nonsynonymous sites provides additional support for an “accelerated evolution” of the DHFR protein in the presence of the DHFR inhibitor PM. The classes of alleles from patients without prophylaxis or without DHFR inhibitors both showed substitutions that were mostly silent, indicating a pressure to maintain the protein function, as expected for most functional genes. In contrast, alleles observed in patients with failure of prophylaxis including a DHFR inhibitor showed a number of nonsynonymous substitutions equal to that of synonymous substitutions, indicating a potential change of structure and/or function of the protein. Such change may be involved in drug resistance.
The fact that 15 of the 16 nonsynonymous mutations were observed each in a single patient suggests that the mutations may have accumulated independently within each patient due to the drug pressure rather than in patients who would have then transmitted their mutated P. jirovecii strains to several other patients. Our previous analysis of recurrent episodes of PcP suggested that DHPS mutations may also be selected in each patient under the drug pressure (14).
Ma and coworkers (7) have found nonsynonymous mutations in P. jirovecii DHPS but not in P. jirovecii DHFR genes from 32 patients, seven of which were receiving prophylaxis with TMP/SMZ. Takahashi et al. (17) reported four DHFR mutations, two synonymous and two nonsynonymous, in specimens of 27 patients, but the mutations were not associated with prior exposure to TMP-SMZ. The results of the present study differ from those of these studies. The reason for this difference probably lies on the prophylactic regimens used. Indeed, in the present study, DHFR mutations were found mostly in patients receiving the DHFR inhibitor PM (in combination with A or SD), whereas in the two other studies TMP (as part of TMP-SMZ) was the drug of first choice for prophylaxis and treatment. Our results are consistent with a pressure of PM on P. jirovecii DHFR higher than that exerted by TMP. This conclusion is supported by in vitro kinetics studies that showed that PM is an inhibitor of P. jirovecii DHFR four times stronger than TMP (8). An in vivo study using expression in yeast reported that the effect of TMP was stronger than that of PM, but this was suggested to be due to a difference in the ability of the two drugs to penetrate the yeast cell (9).
All seven patients with failure of prophylaxis with PM-SD were infected with a P. jirovecii strain harboring DHPS with mutation M2. We already reported this association in a previous study (13). This suggests that this mutation confers resistance specifically to SD. In support of this latter hypothesis, introduction of the M2 mutation within Saccharomyces cerevisiae DHPS conferred resistance toward SD in this organism (11). DHFR amino acid changes were also associated with failure of PM-SD prophylaxis, and five of the seven patients with PM-SD failure harbored P. jirovecii strains with mutations in both DHPS and DHFR. This observation suggests that a synergism between these mutations might be involved in resistance to PM-SD. DHFR mutations may confer resistance toward PM, but it is also possible that they reduce the affinity of the enzyme for the compound formed by DHPS with SD as a substrate. Indeed, such compounds were recently shown in S. cerevisiae to inhibit growth by competing with dihydrofolate, the substrate of DHFR (16). The level of resistance conferred by these mutations or combination of them toward PM and SD remains to be determined. Mutations in both DHPS and DHFR are also involved in PM-SD resistance of P. falciparum (15).
PM-SD has been widely used as malaria treatment in Africa since many years. Despite the absence of any evidence, our results suggest that this use of PM-SD may have resulted in P. jirovecii strains with mutations in both DHFR and DHPS. One can speculate that some of these strains may exhibit some level of resistance to TMP-SMZ. If this is true, this may reduce the benefit of the recent recommendation of the World Health Organization and United Nations Program on HIV/AIDS to use TMP-SMZ for prophylaxis against opportunistic infections in adults and children living with HIV/AIDS (1).
In conclusion, we report here a correlation between P. jirovecii DHFR mutations and the use of anti-Pneumocystis DHFR inhibitors, in particular PM, and suggest that amino acid changes in DHFR may contribute to P. jirovecii emerging drug resistance.
Acknowledgments
We are indebted to Gerrit Kuhn for performing analysis of molecular evolution of DHFR gene. We thank the physicians responsible for the involved wards of Lyon for access to patients' charts, particularly J. L. Touraine, D. Peyramond, and C. Trepo. We also thank S. Picot for storage of the specimens, A. Cruchon for DNA extraction, and J. Kovacs, National Institutes of Health, Bethesda, Md., for kindly providing DHFR amino acid sequences of rabbit- and mouse-derived Pneumocystis.
This study was supported by the Swiss National Program on AIDS Research grant 3345-65407; the Swiss Federal Office for Education and Science for participation in EUROCARINII project, Framework V Program, European Commission (contract QLK2-CT-2000-01369); a North-South fellowship from the University of Lausanne (to A.N.); and the Centre de Coordination de la Lutte contre les Infections Nosocomiales Sud-Est et Hospices Civils de Lyon. Meja Rabodonirina is a member of the Equipe d’Accueil-3609 scientific project (EA3609) of the French Ministry of Research.
REFERENCES
- 1.AIDS/WHO. 2000. Provisional WHO/UNAIDS secretariat recommendations on the use of co-trimoxazole prophylaxis in adults and children living with HIV/AIDS in Africa. UNAIDS, Geneva, Switzerland.
- 2.Champness, J. N., A. Achari, S. P. Ballantine, P. K. Bryant, C. J. Delves, and D. K. Stammers. 1994. The structure of Pneumocystis carinii dihydrofolate reductase to 1.9 Å resolution. Structure 15:915-924. [DOI] [PubMed] [Google Scholar]
- 3.Cody, V., N. Galitsky, D. Rak, J. R. Luft, W. Pangborn, and S. F. Queener. 1999. Ligand-induced conformational changes in the crystal structures of Pneumocystis carinii dihydrofolate reductase complexes with folate and NADP+. Biochemistry 38:4303-4312. [DOI] [PubMed] [Google Scholar]
- 4.Dale, G. E., C. Broger, A. D'Arcy, et al. 1997. A single amino acid substitution in Staphylococcus aureus dihydrofolate reductase determines trimethoprim resistance. J. Mol. Biol. 266:23-30. [DOI] [PubMed] [Google Scholar]
- 5.Kovacs, J. A., V. J. Gill, S. Meshnick, and H. Masur. 2001. New insights into transmission, diagnosis, and drug treatment of Pneumocystis carinii pneumonia. JAMA 286:2450-2460. [DOI] [PubMed] [Google Scholar]
- 6.Kumar, S., K. Tamura, I. B. Jakobsen, and M. Nei. 2001. MEGA2: molecular evolutionary genetics analysis software. Bioinformatics 17:1244-1245. [DOI] [PubMed] [Google Scholar]
- 7.Ma, L., L. Borio, H. Masur, and J. A. Kovacs. 1999. Pneumocystis carinii dihydropteroate synthase but not dihydrofolate reductase gene mutations correlate with prior trimethoprim-sulfamethoxazole or dapsone use. J. Infect. Dis. 180:1969-1978. [DOI] [PubMed] [Google Scholar]
- 8.Ma, L., and J. A. Kovacs. 2000. Expression and characterization of recombinant human-derived Pneumocystis carinii dihydrofolate reeducate. Antimicrob. Agents Chemother. 44:3092-3096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ma, L., Q. Jia, and J. A. Kovacs. 2002. Development of a yeast assay for rapid screening of inhibitors of human-derived Pneumocystis carinii dihydrofolate reductase. Antimicrob. Agents Chemother. 46:3101-3103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Maskell, J. P., A. M. Sefton, and L. M. C Hall. 2001. Multiple mutations modulate the function of dihydrofolate reductase in trimethoprim-resistant Streptococcus pneumoniae. Antimicrob. Agents Chemother. 45:1104-1108. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Meneau, I., D. Sanglard, J. Bille, and P. M. Hauser. 2004. Pneumocystis jirovecii dihydropteroate synthase polymorphisms confer resistance to sulfadoxine and sulfanilamide in Saccharomyces cerevisiae. Antimicrob. Agents Chemother. 48:2610-2616. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Nahimana, A., M. Rabodonirina, P. Francioli, J. Bille, and P. M. Hauser. 2003. Pneumocystis jirovecii dihydrofolate reductase polymorphisms associated with failure of prophylaxis. J. Eukaryot. Microbiol. 50:656-657S. [DOI] [PubMed] [Google Scholar]
- 13.Nahimana, A., M. Rabodonirina, G. Zanetti, I. Meneau, P. Francioli, J. Bille, and P. M. Hauser. 2003. Association between a specific Pneumocystis jiroveci dihydropteroate synthase mutation and failure of pyrimethamine/sulfadoxine prophylaxis in HIV-positive and -negative patients. J. Infect. Dis. 188:1017-1023. [DOI] [PubMed] [Google Scholar]
- 14.Nahimana, A., M. Rabodonirina, J. Helweg-Larsen, I. Meneau, P. Francioli, J. Bille, and P. M. Hauser. 2003. Sulfa resistance and dihydropteroate synthase mutants in recurrent Pneumocystis carinii pneumonia. Emerg. Infect. Dis. 9:864-867. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Olliaro, P. 2001. Mode of action and mechanisms of resistance for antimalarial drugs. Pharmacol. Ther. 89:207-219. [DOI] [PubMed] [Google Scholar]
- 16.Patel, O., J. Satchell, J. Baell, R. Fernley, P. Coloe, and I. Macreadie. 2003. Inhibition studies of sulfonamide-containing folate analogs in yeast. Microb. Drug Resist. 9:139-146. [DOI] [PubMed] [Google Scholar]
- 17.Takahashi, T. T., Endo, T. Nakamura, H. Sakashitat, K. Kimurat, K. Ohnishit, Y. Kitamura, and A. Iwamoto. 2002. Dihydrofolate reductase gene polymorphisms in Pneumocystis carinii f. sp. hominis in Japan. J. Med. Microbiol. 51:510-515. [DOI] [PubMed] [Google Scholar]
- 18.Triglia, T., and A. F. Cowman. 1999. The mechanism of resistance to sulfa drugs in Plasmodium falciparum. Drug Resist. Update 2:15-19. [DOI] [PubMed] [Google Scholar]
- 19.Vedantam, G., and B. P. Nichols. 1998. Characterization of a mutationally altered dihydropteroate synthase contributing to sulfathiazole resistance in Escherichia coli. Microb. Drug Resist. 4:91-97. [DOI] [PubMed] [Google Scholar]
- 20.Walzer, P. D., J. Foy, P. Steele, C. K. Kim, M. White, R. S. Klein, B. A. Otter, and C. Allegra. 1992. Activity of antifolate, antiviral, and other drugs in an immunosuppressed rat model of Pneumocystis carinii pneumonia. Antimicrob. Agents Chemother. 36:1935-1942. [DOI] [PMC free article] [PubMed] [Google Scholar]