Time Course of Microbiologic Outcome and Gene Expression in Candida albicans during and following In Vitro and In Vivo Exposure to Fluconazo
http://www.100md.com
《抗菌试剂及化学方法》
Department of Medicine, Section of Infectious Diseases and Medical Microbiology and Immunology, University of Wisconsin, Madison, Wisconsin
ABSTRACT
Pharmacodynamics (PD) considers the relationship between drug exposure and effect. The two factors that have been used to distinguish the PD behaviors of antimicrobials are the impact of concentration on the extent of organism killing and the duration of persistent microbiologic suppression (postantibiotic effect). The goals of these studies were (i) to examine the relationship between antimicrobial PD and gene expression and (ii) to gain insight into the mechanism of fluconazole effects persisting following exposure. Microarrays were used to estimate the transcriptional response of Candida albicans to a supra-MIC F exposure over time in vitro. Fluconazole at four times the MIC was added to a log-phase C. albicans culture, and cells were collected to determine viable growth and for microarray analyses. We identified differential expression of 18% of all genes for at least one of the time points. More genes were upregulated (n = 1,053 [16%]) than downregulated (174 [3%]). Of genes with known function that were upregulated during exposure, most were related to plasma membrane/cell wall synthesis (18%), stress responses (7%), and metabolism (6%). The categories of downregulated genes during exposure included protein synthesis (15%), DNA synthesis/repair (7%), and transport (7%) genes. The majority of genes identified at the postexposure time points were from the protein (17%) and DNA (7%) synthesis categories. In subsequent studies, three genes (CDR1, CDR2, and ERG11) were examined in greater detail (more concentration and time points) following fluconazole exposure in vitro and in vivo. Expression levels from the in vitro and in vivo studies were congruent. CDR1 and CDR2 transcripts were reduced during in vitro fluconazole exposure and during supra-MIC exposure in vivo. However, in the postexposure period, the mRNA abundance of both pumps increased. ERG11 expression increased during exposure and fell in the postexposure period. The expression of the three genes responded in a dose-dependent manner. In sum, the microarray data obtained during and following fluconazole exposure identified genes both known and unknown to be affected by this drug class. The expanded in vitro and in vivo expression data set underscores the importance of considering the time course of exposure in pharmacogenomic investigations.
INTRODUCTION
The time course of antimicrobial activity is dependent on a drug's pharmacokinetics and two major pharmacodynamic characteristics (2, 13). The first is the rate of organism killing and whether increasing drug concentrations enhance the rate and extent of killing. The second is the presence or absence of inhibitory effects on organism growth which persist after drug levels have fallen below the MIC. These persistent effects are termed postantibiotic effects (PAE) (3, 4, 5, 9, 14, 16, 29, 38, 39, 40). The PAE pharmacodynamic phenomenon underscores the importance of investigating the effects of antimicrobial exposure over time. Drugs from the triazole class demonstrate time-dependent killing but prolonged persistent effects in vivo.
Several investigators have undertaken studies to discern the biologic basis for the PAE phenomenon with several antibacterial compounds (7, 11, 17-20, 43). For example, the duration of the PAE for quinolones and ?-lactams correlates closely with the time it takes for bacterial DNA synthesis to resume. Similar study with the aminoglycosides has shown that the duration of the PAE best correlates with the time it takes for protein synthesis to resume.
A number of investigations have examined the global transcriptional profile of Candida albicans following in vitro triazole exposure (1, 6, 15, 23). These important studies have provided insight into the mechanism of action of these compounds. However, these studies have not considered the impact of the pharmacodynamics or time course of exposure on the response of the organism. The goal of the current experiments was to describe the global response of C. albicans transcription during the time course of fluconazole exposure. The information obtained provides a mechanistic model of the molecular events during the time course of exposure. Furthermore, the results from these studies underscore the importance of considering antimicrobial pharmacodynamics in studies investigating the drug-organism interaction. Finally, these investigations point to the relevance of considering molecular end points in antimicrobial pharmacokinetic/pharmacodynamic studies.
MATERIALS AND METHODS
Microorganism and in vitro susceptibility testing. Candida albicans K1 was used for all studies (3). Susceptibility testing was performed in duplicate on two occasions, using the CLSI (formerly NCCLS) M-27A method (30).
In vitro fluconazole exposure. (i) Microarray study of microbiologic effect and gene expression. C. albicans K1 was grown in vitro in the presence or absence of fluconazole for comparison of the time courses of gene expression using cDNA microarrays. An overnight culture was diluted to an initial inoculum of 105 CFU/ml and grown in RPMI-morpholinepropanesulfonic acid (RPMI-MOPS) buffered with MOPS at 37°C with shaking at 250 rpm. The treatment group was exposed to fluconazole at a concentration of four times the MIC, which was added after 3 h of growth and continued for 3 h. Following drug exposure, cells were pelleted, washed, and resuspended in fresh medium, and incubation continued for an additional 12 h. Samples for CFU enumeration and RNA isolation were collected at two time points during drug exposure (1 and 3 h) and two time points after drug removal (6 and 9 h). The growth and drug exposure experiment described above was repeated on three occasions. The experiment design was consistent with proposed microarray study standards (21).
(ii) Confirmation of microbiologic effect and gene expression and expanded time course. C. albicans K1 was similarly grown in vitro in the presence or absence of fluconazole for comparison of gene expression levels, using both Northern blotting and quantitative reverse transcription-PCR (RT-PCR), to confirm the results from the microarray study (10, 35). In addition, these studies were used to expand the evaluation of the relationship between the concentration/time of exposure and the transcriptional profile of the cells. The treatment groups were exposed to one of five increasing (fourfold) fluconazole concentrations. The exposures included concentrations at the MIC (1x), above the MIC (4x and 16x), and below the MIC (0.06x and 0.25x). Fluconazole was added after 6 h of growth (to cells in log-phase growth) and continued for 3 h. The cells were then pelleted, washed, resuspended in fresh medium, and incubated for an additional 6 h. Samples for CFU enumeration were collected at 3-h intervals. Samples for RNA isolation were collected at one time point during drug exposure (after 3 h of fluconazole exposure) and one time point after drug removal (3 h after fluconazole removal). Experiments were repeated on three occasions. Candida cells were harvested for RNA isolation as previously described (37).
Pharmacokinetics. The pharmacokinetic values used in the current study were previously determined for the animal model used here (3). Briefly, the time courses of fluconazole concentrations in mouse serum were determined for three fourfold increasing doses (6.25, 25, and 100 mg/kg of body weight) after subcutaneous administration to neutropenic infected mice, using a gas chromatography method. Pharmacokinetic constants, including the elimination half-life, area under the serum concentration curve (AUC), and peak level (Cmax), were calculated using a noncompartmental model.
Animal infection model and in vivo fluconazole exposure. A neutropenic murine model of disseminated candidiasis was used to determine if the drug exposure-gene expression time course observed in vitro was similar to that observed in vivo (2, 31). The animal model has been described in detail in prior publications (2, 3).
Following infection, groups of mice received one of three fourfold increasing fluconazole dose levels (0.78, 3.125, and 12.5 mg/kg) administered subcutaneously during the log phase of growth (18 h after infection). The dose levels were chosen to include a sub-MIC exposure only (0.78 mg/kg) and two supra-MIC exposures (3.125 and 12.5 mg/kg). The serum pharmacokinetics of the last two doses were estimated to achieve concentrations exceeding the MIC of the organism for 9.6 and 15.4 h, respectively.
Groups of two animals from untreated control and fluconazole treatment groups were euthanized at 3- to 6-h intervals over a period of 45 h for CFU enumeration as previously described (3). Results were expressed as the mean CFU/2 kidneys from two mice. The serum pharmacokinetic results were used to estimate the time that the total drug level of fluconazole remained above the MIC for the organism. The postantifungal effect (PAFE) was calculated by determining the time that it took the burden of organisms in the untreated control mice to increase 1 log10 CFU/kidney (C) and subtracting this from the amount of time that it took the organisms from the treated animals to grow 1 log10 CFU/kidney (T) after levels had fallen below the MIC, as follows: PAFE = T – C (14).
Groups of 10 mice per time point were euthanized at 3- to 12-h intervals for Candida RNA isolation from mouse kidneys as recently described (2). Fungal cells were collected from animals receiving the highest fluconazole dose level at five time points (3, 6, 18, 21, and 27 h) to include two supra-MIC and three postexposure (serum levels estimated to have fallen below the C. albicans MIC) sample values. Following the middle dose, organisms were harvested for RNA isolation at two points which corresponded to postexposure times (after serum levels fell below the MIC [12 and 15 h]). Two sampling time points were used for the lowest time point (3 and 6 h), both corresponding to estimated sub-MIC exposures.
Transcriptional profiles of cell populations. Studies were undertaken to compare the global responses of C. albicans during and following fluconazole exposure in vitro over time, using cDNA microarrays (version 6.31) produced at the Biotechnology Research Institute, National Research Council, Montreal, Canada (http://www.irb-bri.cnrc.gc.ca/services/microarray/scanning_e.html) (2, 12). The microarray used is based on C. albicans SC5314. The in vitro transcriptional profiles of a population of C. albicans cells during and following fluconazole exposure were examined relative to those of nonexposed cells collected at the same time periods of growth, as described above. At each of the specified time points, cells were collected by centrifugation at 3,000 x g for 2 min at ambient temperature, and the cell pellets were flash frozen in an ethanol-dry ice bath. Candida RNAs were isolated by the hot phenol method (37). mRNAs were purified using a Fast Track mRNA isolation kit (Invitrogen). Each set of samples was used for separate array hybridizations with cDNAs from the drug-exposed and control populations. In the first hybridization, the drug-exposed sample was labeled with Cy3 and the control population was labeled with Cy5. In the second hybridization, the exposed sample was labeled with Cy5 and the control population was labeled with Cy3. A total of six hybridizations were performed per time point, as previously described (2, 12).
Microarrays were scanned using a ScanArray 5000 instrument (Packard BioScience, Billerica, MA) at a 10-μm resolution. The intensities of the spots were quantified with QuantArray (GSI Lumonics, Billerica, MA). QuantArray files were analyzed in Excel (Microsoft, Redmond, WA). To be included in normalization and analysis, each spot had to satisfy three quality control criteria (2, 12). The signal intensity minus half of the standard deviation had to be greater than the local background plus half of the standard deviation; the signal intensity had to be within the dynamic range of the photomultiplier tube; and the raw intensities of the duplicate spots for each gene had to be within 50% of each other. For spots that met these criteria, we examined the log ratio of the intensity of expression from the fluconazole-exposed cell population to that from the unexposed cells. The results presented consist of the averages of three independent experiments with dye swaps. When either channel value was below 100.0, the data point was considered unacceptable. A normalization factor was applied to account for systematic differences in probe label intensities. A two-sided Student t test was used to assess the statistical significance of log2 ratios.
Gene ontology identification was undertaken for differentially expressed genes, using Candida (CandidaDB [http://genolist.pasteur.fr/CandidaDB/] and Candida Genome Database [http://www.candidagenome.org/]) and Saccharomyces (Saccharomyces Genome Database [http://genome-www.Stanford.edu/Saccharomyces/]) (8) databases.
Confirmation of microarray expression data. The mRNA abundance of a subset of genes identified in the microarray analysis was estimated using either Northern blotting or RT-PCR (10, 35). The specific C. albicans genes examined included CDR1, CDR2, and ERG11 (34, 36, 41, 42). CDR1 and CDR2 encode efflux pumps for which fluconazole is a substrate. ERG11 encodes the fluconazole drug target, lanosterol 14-demethylase. Prior investigations have demonstrated that the expression of these genes can be induced by triazole exposure (22, 28). The current studies were intended to expand the examination of the relationship between the pharmacodynamics of the fluconazole exposure (effects of concentration and time) and the expression of these genes of interest. The initial cell concentrations, growth conditions, drug exposures, and sampling time points are described above. Growth experiments and mRNA abundance assays were performed in triplicate.
For both Northern blotting and RT-PCR, cells were harvested and RNAs were isolated as described above. Northern blots were performed using standard methodology. RT-PCR studies utilized TaqMan methodology (24). Probes for Northern blots and RT-PCR (primer and probe sets) have been previously described and were generated for CDR1, ERG11, and ACT1 by PCR amplification from Candida genomic DNA (see Table S5 in the supplemental material) (26, 34, 36, 41). For both methods, actin was utilized as a constitutively expressed control.
Band intensities in Northern blot studies were estimated using Image J software (NIH). Gene intensities were expressed as the background normalized ratios for the conditions of interest described above. RT-PCR quantitative data analysis was completed by using a modification of the CT method of Livak and Schmittgen (24, 25). For each amplification run, the threshold cycle, CT, of the target was normalized to the CT of the ACT1 gene (CT = CT of target – CT of normalizer). We compared the trends in CT values over time for cells grown in the presence of fluconazole and those grown in the absence of fluconazole. Real-time quantitative PCR was performed on three 10-fold cDNA dilutions to ensure similar primer efficiencies among the various amplified targets (2).
RESULTS
In vitro susceptibility testing. The fluconazole MIC for C. albicans K1 is 0.5 μg/ml.
Pharmacokinetics. The pharmacokinetic profile was linear, with the peak level and AUC increasing proportionally for each dose escalation. Peak levels occurred within 1 h of administration and ranged from 5.2 to 100 μg/ml (Table 1). The elimination half-lives ranged from 2.8 to 3.5 h. The 24-h AUCs ranged from 24 to 460 mg · h/liter.
In vitro fluconazole exposure. (i) Microarray study of microbiologic effect and gene expression. The starting inoculum of C. albicans in the RPMI-MOPS medium was 5.0 log10 CFU/ml. After 3 hours of growth, the viable counts had increased to 6.2 ± 0.14 log10 CFU/ml, at which time fluconazole was added to one of the flasks (Fig. 1). Over the 3-h drug exposure time at a fourfold excess of the MIC, the burden of viable cells did not increase. During the same time period, the untreated control cells grew slightly (0.45 log10 CFU/ml). After fluconazole removal, the growth curve assumed the trajectory of the untreated control. Persistent growth suppression following in vitro drug exposure (PAFE) was not observed. The change in viable counts over the next 3 hours increased 1.4 and 1.1 log10 CFU/ml in the previously exposed and untreated cell populations, respectively. The slope of the growth curve began to plateau for both cell populations between the 6- and 9-h time points, with increases in burden ranging from 0.2 to 0.6 log10 CFU/ml. In general, the growth phases were similar for both cell populations at the various sampling time points.
Microarray analysis over the time course of exposure identified more than 1,000 differentially expressed genes (see Tables S1, S2, S3, and S4 in the supplemental material). Over the entire experiment, more genes were upregulated (n = 1053 [16%]) than downregulated (174 [3%]). However, expression data from the time points during F exposure identified equal numbers of both up (2%)- and downregulated (2%) genes. The functions of the majority of the genes in both groups are unknown (42% of upregulated genes and 32% of downregulated). Of the upregulated genes (Fig. 2) with known ontology during exposure, many were related to plasma membrane/cell wall synthesis/maintenance (18%), stress responses (7%), and metabolism (8%). Nine genes from the ergosterol synthesis pathway were upregulated during fluconazole exposure. Only two genes from this group precede the fluconazole target step in the ergosterol synthesis pathway. The categories of downregulated genes during exposure (Fig. 2) included protein synthesis (15%), DNA synthesis/repair (7%), and transport (7%). The postexposure data time points (Fig. 3) included mainly upregulation (16%) (<1% downregulated). The majority of upregulated genes were from the protein (17%) and DNA (7%) synthesis or modification categories. Several genes from the ergosterol pathway remained elevated; however, the degree of upregulation was lower than that during fluconazole exposure.
(ii) Confirmation of microbiologic effect and gene expression and expanded time course. The starting inoculum of C. albicans in the RPMI-MOPS medium was near 5.0 log10 CFU/ml (Fig. 4). During the first 3 hours of growth, the viable counts had increased slightly. The cells shifted to log-phase growth over the subsequent 3 h (6-h time point), increasing >2 log10 CFU/ml, at which point the various fluconazole concentrations were introduced to the specified cell populations. Over the 3-hour drug exposure period, the sub-MIC concentrations exhibited minimal effects on the continued growth of cells. The larger fluconazole concentrations did inhibit growth in a dose-dependent fashion. However, after the removal of fluconazole, the slopes of the growth curves rapidly returned to that of the untreated control. Similar to our other in vitro experiment, there was no measurable microbiologic postantifungal effect with any of the fluconazole exposures (16).
The abundance of ERG11 (Fig. 5) was increased during fluconazole exposure (after 3 h of exposure) relative to that in untreated controls for each of the drug concentrations examined. The abundance increased relative to the fluconazole concentration. The smallest change in ERG11 expression in treated relative to untreated control cells was 1.2-fold at a concentration of one-eighth the MIC. The largest change was 2.3-fold, which was associated with exposure to fluconazole at a concentration 16-fold higher than the MIC. The differences in abundance for the two highest concentrations were statistically different from those for the three lowest concentrations (P < 0.001). The increase in mRNA abundance associated with the various drug exposures declined following 3 h of growth in the absence of fluconazole. The decline in expression during the postexposure period was similar to that observed with other genes in the ergosterol synthesis pathway from the microarray study.
Both Northern blots and quantitative RT-PCR were utilized to examine the relative mRNA abundances of the CDR1 gene associated with the same fluconazole concentrations (Fig. 5). Following a 3-hour fluconazole exposure, the mRNA abundance of CDR1 did not increase relative to that in untreated control cells for any of the five drug concentrations. However, during the regrowth period following fluconazole exposure, the expression of CDR1 increased for four of the five drug concentrations. The degree of upregulation was dependent on the fluconazole concentration, increasing >2.5-fold in association with exposure to the two highest fluconazole concentrations. The differences in expression for the two highest fluconazole concentrations were statistically different from those for the three lowest concentrations (P < 0.01).
Quantitative RT-PCR was used for estimations of the expression of CDR2 (Fig. 5). The relationship between the fluconazole exposure and mRNA abundance for CDR2 was remarkably similar to that for CDR1. During fluconazole exposure (3 h), the relative expression of CDR2 was no different in exposed and untreated cells. However, at the postexposure time point, the abundance of CDR2 increased markedly in exposed cells in a concentration-dependent manner. The 16x and 4x MIC exposures resulted in 10- and 6-fold increases in CDR2 expression relative to that in untreated controls (P < 0.001). The expression values for the lower drug concentrations were slightly higher than those for control cells and for the prior time point; however, these differences were insignificant.
Microbiologic effect and gene expression after in vivo fluconazole exposure. Two hours after intravenous inoculation, the viable burden of C. albicans in mouse kidneys was 3.23 ± 0.05 log10 CFU/2 kidneys (Fig. 6). At the time of fluconazole administration, the cells were in the late log phase of growth, with a viable kidney burden of 6.84 ± 0.49 log10 CFU/2 kidneys 18 h after the initiation of infection. The two lowest fluconazole drug concentrations had no detectable impact on viable counts in neutropenic mouse kidneys. The highest fluconazole dose level did inhibit growth for at least 6 h after drug administration. Within 18 h after the largest fluconazole dose, the viable organism burden began to increase. In contrast to prior in vivo fluconazole investigations with this model, a measurable PAFE was not apparent (3). We speculate that the lack of a measurable PAFE is related to the much later timing of drug administration (18 h compared to 2 h after infection).
The mRNA abundances of the two CDR pumps in relation to drug exposure over time are shown in Fig. 7 (CDR1 and CDR2 data are on the left and right, respectively). At each time point, the mRNA abundance is presented as a ratio for drug-treated cells to untreated controls. The CDR pump abundance was not elevated in response to the two supra-MIC concentrations or the single sub-MIC concentration. However, the mRNA abundance of both pumps was remarkably increased during the postantifungal period of time. This response was not related to the rate or phase of growth, as the control strain's growth was quite similar. The abundances of both transcripts increased from 12 to 21 h after the initial drug exposure and began to decline at the final time point (27 h). The maximal change in treated relative to untreated cells in vivo was 18- and 3.5-fold for CDR2 and CDR1, respectively.
DISCUSSION
The field of antimicrobial pharmacodynamics investigates the relationships between drug concentrations relative to the MIC over time and outcomes (13). Understanding these relationships has had an enormous impact on the manner in which these compounds are administered. The majority of pharmacodynamic investigations examine treatment efficacy by using a microbiologic end point (3, 9, 13, 14, 16, 39, 40). Few other drug effect measures have been considered in these types of studies. Studies with several antibacterial compounds have examined cell morphology and protein, DNA, and RNA synthesis end points by measuring the incorporation of DNA, RNA, and protein radiolabeled precursors during the postantibiotic effect in vitro in an attempt to understand the biological basis for the phenomenon (9, 17). These investigations have not yet taken advantage of newer genomic technologies to further understand these processes.
Pharmacodynamic evaluations of triazole antifungals have shown that these drugs inhibit the growth of fungi in a time-dependent manner (3, 4, 16, 27). Exposure to increasingly large concentrations does not enhance the inhibitory activities of these compounds. However, following exposure to supra-MIC concentrations in vivo, regrowth of fungi is delayed (3). This phenomenon of persistent growth inhibition following antifungal drug exposure has not been investigated mechanistically. Several investigators have, however, utilized microarray technology to examine the global transcriptional responses of C. albicans and Saccharomyces cerevisiae to several triazole compounds (1, 6, 15, 23). These studies have been important for furthering our understanding of the mechanism of action of these drugs. The triazole exposures used in these in vitro investigations have varied from study to study. First, the duration of drug exposure prior to RNA sampling has varied among studies. In addition, the studies have all used supra-MIC exposures and a single sampling time point. Thus, there has been minimal consideration of the pharmacokinetics of drug exposure. The current investigation examined genomic end points during the time course of triazole-C. albicans interaction. The goals of the study were (i) to determine if gene expression in this fungal pathogen is impacted by the pharmacokinetics of drug exposure and (ii) to further our understanding of the molecular basis for the time course effects of fluconazole.
The results from the current microarray studies underscore the importance of examining the impact of drug exposure over time. There were numerous genes identified at only one of the four time points. However, overall, the differential expression of cells collected from the two supra-MIC time points was similar. The same transcriptional pattern was also observed in the two postexposure cell populations. The ontology categories most commonly identified from genes upregulated during fluconazole exposure are similar to those identified in previous triazole-yeast exposure microarray studies (1, 6, 15, 23). The predominant categories of known function included those related to plasma membrane and cell wall synthesis, maintenance, and function. The impact of fluconazole was not surprisingly marked with regard to the ergosterol synthesis and sterol modification pathway. Among the genes affected in this pathway, six of seven were at or downstream of the fluconazole drug target. This pattern of expression in this pathway is similar to that previously reported for triazole exposure of S. cerevisiae and C. albicans. Another common subset of expressed genes included those previously shown to be involved in the cell stress response. The third most common functional group of genes with increased mRNA abundance during fluconazole exposure was the group involved in metabolism. Among this group, genes related to carbohydrate metabolism were present most frequently. The number of downregulated genes was small compared to the number of upregulated genes. The two most common ontology categories for this set of genes included genes related to DNA and protein synthesis. It is interesting that prior antibacterial PAE studies similarly demonstrated reduced DNA and protein synthesis during the period of microbiologic growth suppression. Conversely, genes expressed in our postexposure period, after which microbiologic recovery had begun, were marked primarily by the activation of nucleic acid and protein synthesis machineries. Perhaps not surprisingly, these genes were not identified in prior triazole-yeast interaction array investigations, which included only a single early exposure time point. This pattern of expression during the microbiologic effect and in the postexposure period can be used to propose a triazole time course activity model. The expression of genes associated with drug exposure and microbiologic activity represents the direct effect of the drug or additional organism perturbations in response to the sterol effects. The later set of upregulated genes suggested how the organism responds during the period of microbiologic recovery from the effects of the drug. Both groups of genes are interesting to consider for future study. The first group includes those upregulated during exposure, including, of particular interest, those associated with perturbation of the cell wall and plasma membrane. Any number of these gene products may provide useful drug targets, whose inhibition may potentiate the microbiologic effects of the triazole. Similarly, the set of genes identified during organism recovery may also provide interesting drug targets which, if inhibited, may reduce the ability of the organism to recover from the effects of the triazole and prolong the growth inhibition.
The impact of the time course of fluconazole pharmacokinetics was further examined for a subset of three genes. The expanded time course experiments included both a larger range of fluconazole concentrations and other time points during and following exposure. In addition, we examined the transcriptional responses of C. albicans following drug exposure both in vitro and in vivo. Importantly, the results of both in vitro and in vivo studies were congruent. The three genes chosen were CDR1, CDR2, and ERG11. The choice of these three genes was based on their identification in the current microarray data set as well as the fact that prior studies have demonstrated differential expression of each of the genes in response to triazole exposure (22, 28, 36, 42). The time course expression data for these genes confirmed the microarray results. The mRNA abundance of ERG11 was highest during fluconazole exposure and declined in the post-exposure period. Conversely, the abundance of each of the CDR efflux pumps was greatest in the postexposure time period and may result in protection from additional fluconazole exposure during the postexposure period. These data again underscore the importance of considering the time course of events in studying the expression profile of the antimicrobial/organism interaction, as demonstrated previously from the microbiologic standpoint. The study of an expanded concentration range in these expression studies also demonstrated a strong pharmacodynamic relationship. The mRNA abundance was clearly impacted by the concentration of fluconazole over five exposure levels, which varied 256-fold. In addition, the two lowest concentrations were below the MIC for the organism and demonstrated an impact on gene expression for each of the three genes. Previous studies with both triazoles and polyenes have demonstrated sub-MIC effects using different biologic end points (32, 33, 39). One explanation for the discrepancy between in vitro and in vivo investigations of the triazole postantifungal effect has been unavoidable sub-MIC exposures which are present in the latter study design (3, 16). Additional investigation of these sub-MIC effects at the genomic level may offer further insight into the molecular basis for in vivo triazole postantifungal effects.
In sum, these studies demonstrate a relationship between the concentration and time course of exposure and the transcriptional response of an organism. The change in expression profile over time and the change in level of expression relative to concentration underscore the importance of considering the pharmacokinetics of drug exposure in microarray experiments investigating antimicrobial-pathogen interactions. The global transcriptional profile of the fluconazole-Candida albicans interactions suggests a damage response model in which the plasma membrane and cell wall are structurally and functionally damaged, followed by a period of recovery manifested by enhanced nucleic acid and protein synthesis to repair the cell.
Supplemental material for this article may be found at http://aac.asm.org/.
REFERENCES
Agarwal, A. K., P. D. Rogers, S. R. Baerson, M. R. Jacob, K. S. Barker, J. D. Cleary, L. A. Walker, D. G. Nagle, and A. M. Clark. 2003. Genome-wide expression profiling of the response to polyene, pyrimidine, azole, and echinocandin antifungal agents in Saccharomyces cerevisiae. J. Biol. Chem. 278:34998-35015.
Andes, D., A Lepak, A. Pitula, K. Marchillo, and J. Clark. 2005. A simple approach for estimating gene expression in Candida albicans directly from a systemic infection site. J. Infect. Dis. 192:893-900.
Andes, D., and M. van Ogtrop. 1999. Characterization and quantitation of the pharmacodynamics of fluconazole in a neutropenic murine disseminated candidiasis model. Antimicrob. Agents Chemother. 43:2116-2120.
Andes, D. 2003. In vivo pharmacodynamics of antifungal drugs in treatment of candidiasis. Antimicrob. Agents Chemother. 47:1179-1186.
Andes, D., T. Stamstad, and R. Conklin. 2001. Pharmacodynamics of amphotericin B in a neutropenic mouse disseminated candidiasis model. Antimicrob. Agents Chemother. 45:922-926.
Bammert, G. F., and J. M. Fostel. 2000. Genome-wide expression patterns in Saccharomyces cerevisiae: comparison of drug treatments and genetic alterations affecting biosynthesis of ergosterol. Antimicrob. Agents Chemother. 44:1255-1265.
Barmada, S., S. Kohlhepp, J. Leggett, R. Dworkin, and D. Gilbert. 1993. Correlation of tobramycin-induced inhibition of protein synthesis with postantibiotic effect in Escherichia coli. Antimicrob. Agents Chemother. 37:2678-2683.
Braun, B. R., M. van het Hoog, C. d'Enfert, C. M. Martchenko, J. Dungan, A. Kuo, D. O. Inglis, M. A. Uhl, H. Hogues, M. Berriman, M. Lorenz, A. Levitin, U. Oberholzer, C. Bachewich, D. Harcus, A. Marcil, D. Dignard, T. Iouk, R. Zito, L. Frangeul, F. Tekaia, K. Rutherford, E. Wang, C. A. Munro, S. Bates, N. A. Gow, L. L. Hoyer, G. K?hler, J. Morschh?user, G. Newport, S. Znaidi, M. Raymond, B. Turcotte, G. Sherlock, M. Costanzo, J. Ihmels, J. Berman, D. Sanglard, N. Agabian, A. P. Mitchell, A. D. Johnson, M. Whiteway, and A. Nantel. 2005. A human-curated annotation of the Candida albicans genome. PLOS Genet. 1:e1.
Bundtzen, R. W., A. U. Gerber, D. L. Cohn, and W. A. Craig. 1981. Postantibiotic suppression of bacterial growth. Rev. Infect. Dis. 3:28-37.
Bustin, S. A. 2000. Absolute quantification of mRNA using real-time reverse transcription polymerase chain reaction assays. J. Mol. Endocrinol. 25:169-193.
Champney, W. S., and C. L. Tober. 1999. Molecular investigation of the postantibiotic effects of clarithromycin and erythromycin on Staphylococcus aureus cells. Antimicrob. Agents Chemother. 43:1324-1328.
Cowen, L. E., A. Nantel, M. S. Whiteway, D. Y. Thomas, D. C. Tessier, L. M. Kohn, and J. B. Anderson. 2002. Population genomics of drug resistance in Candida albicans. Proc. Natl. Acad. Sci. USA 99:9284-9289.
Craig, W. A. 1998. Pharmacokinetic/pharmacodynamic parameters: rationale for antibacterial dosing of mice and men. Clin. Infect. Dis. 26:1-12.
Craig, W. A., and S. Gudmundsson. 1996. Postantibiotic effect, p. 296-329. In V. Lorian (ed.), Antibiotics in laboratory medicine, 4th ed. Williams and Wilkins, Baltimore, Md.
DeBacker, M. D., T. Ilyina, X. J. Ma, S. Vandoninck, W. H. Luyten, and H. Vanden Bossche. 2001. Genomic profiling of the response of Candida albicans to itraconazole treatment using a DNA microarray. Antimicrob. Agents Chemother. 45:1660-1670.
Ernst, E., M. E. Klepser, and M. A. Pfaller. 2000. Postantifungal effects of echinocandin, azole, and polyene antifungal agents against Candida albicans and Cryptococcus neoformans. Antimicrob. Agents Chemother. 44:1108-1111.
Gottfredsson, M., H. Erlendsdottir, A. Gudmundsson, and S. Gudmundsson. 1995. Different patterns of bacterial DNA synthesis during postantibiotic effect. Antimicrob. Agents Chemother. 39:1314-1319.
Gottfredsson, M., H. Erlendsdottir, R. Kolka, A. Gudmundsson, and S. Gudmundsson. 1993. Ultrastructural alterations of bacteria during the postantibiotic effect. Chemotherapy 39:153-162.
Gottfredsson, M., H. Erlendsdottir, A. Sigfusson, and S. Gudmundsson. 1998. Characteristics and dynamics of bacterial populations during postantibiotic effect determined by flow cytometry. Antimicrob. Agents Chemother. 42:1005-1011.
Guan, L., R. M. Blumenthal, and J. C. Burnham. 1992. Analysis of macromolecular biosynthesis to define the quinolone-induced postantibiotic effect in Escherichia coli. Antimicrob. Agents Chemother. 36:2118-2124.
Hingamp, J., J. Quackenbush, G. Sherlock, P. Spellman, C. Stoeckert, J. Aach, W. Ansorge, C. A. Ball, H. C. Causton, T. Gaasterland, P. Glenisson, F. C. P. Holstege, I. F. Kim, V. Markowitz, J. C. Matese, H. Parkinson, A. Robinson, U. Sarkans, S. Schulze-Kremer, J. Stewart, R. Taylor, J. Vilo, and M. Vingron. 2001. Minimum information about a microarray experiment (MIAME)—toward standards for microarray data. Nat. Genet. 29:365-371.
Karababa, M., A. T. Coste, B. Rognon, J. Bille, and D. Sanglard. 2004. Comparison of gene expression profiles of Candida albicans azole-resistant clinical isolates and laboratory strains exposed to drugs inducing multidrug transporters. Antimicrob. Agents Chemother. 48:3064-3079.
Liu, T. T., R. E. Lee, K. S. Barker, R. E. Lee, L. Wei, R. Homayouni, and P. D. Rogers. 2005. Genome-wide expression profiling of the response to azole, polyene, echinocandin, and pyrimidine antifungal agents in Candida albicans. Antimicrob. Agents Chemother. 49:2226-2236.
Livak, K. J., S. J. Flood, J. Marmaro, W. Giusti, and K. Deetz. 1995. Oligonucleotides with fluorescent dyes at opposite ends provide a quenched probe system useful for detecting PCR product and nucleic acid hybridization. PCR Methods Appl. 4:357-362.
Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2CT method. Methods 25:402-408.
Losberger, C., and J. F. Ernst. 1989. Sequence of Candida albicans gene encoding actin. Nucleic Acids Res. 17:9488.
Louie, A., G. L. Drusano, P. Banerjee, Q. F. Liu, P. Kaw, M. Shayegani, H. Taber, and M. H. Miller. 1998. Pharmacodynamics of fluconazole in a murine model of systemic candidiasis. Antimicrob. Agents Chemother. 42:1105-1109.
Lyons, C. N., and T. C. White. 2000. Transcriptional analyses of antifungal drug resistance in Candida albicans. Antimicrob. Agents Chemother. 44:2296-2303.
Moore, R. D., P. S. Lietman, and C. R. Smith. 1987. Clinical response to aminoglycoside therapy: importance of the ratio of peak concentration to minimal inhibitory concentration. J. Infect. Dis. 155:93-99.
National Committee for Clinical Laboratory Standards. 1997. Reference method for broth dilution antifungal susceptibility testing for yeast. Accepted standard M27-A. National Committee for Clinical Laboratory Standards, Wayne, Pa.
National Research Council Committee on the Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, Commission on Life Sciences. 1996. Guide for the care and use of laboratory animals. National Academy Press, Washington, D.C.
Nosanchuk, J. D., W. Cleare, S. P. Franzot, and A. Casadevall. 1999. Amphotericin B and fluconazole affect cellular charge, macrophage phagocytosis, and cellular morphology of Cryptococcus neoformans at subinhibitory concentrations. Antimicrob. Agents Chemother. 43:233-239.
Nugent, K. M., and K. R. Couchot. 1986. Effects of sublethal concentrations of amphotericin B on Candida albicans. J. Infect. Dis. 154:665-669.
Prasad, R., P. De Wergifosse, A. Goffeau, and E. Balzi. 1995. Molecular cloning and characterization of a novel gene in Candida albicans, CDR1, conferring multiple drug resistance to drugs and antifungals. Curr. Genet. 27:320-329.
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Sanglard, D., F. Ischer, M. Monod, and J. Bille. 1997. Cloning of Candida albicans genes conferring resistance to azole antifungal agents: characterization of CDR2, a new multidrug ABC transporter gene. Microbiology 143:405-416.
Schmitt, M. E., T. A. Brown, and B. L. Trumpower. 1990. A rapid and simple method for preparation of RNA from Saccharomyces cerevisiae. Nucleic Acids Res. 18:3091-3092.
Tornqvist, I. O., S. E. Holm, and O. Cars. 1990. Pharmacodynamic effects of subinhibitory antibiotic concentrations. Scand. J. Infect. Dis. 74(Suppl.):94-101.
Turnidge, J. D., S. Gudmundsson, B. Vogelman, and W. A. Craig. 1994. The postantibiotic effect of antifungal agents against common pathogenic yeasts. J. Antimicrob. Chemother. 34:83-92.
Vogelman, B., S. Gudmundsson, J. Turnidge, J. Leggett, and W. A. Craig. 1988. In vivo postantibiotic effect in a thigh infection in neutropenic mice. J. Infect. Dis. 157:287-298.
White, T. C., K. A. Marr, and R. A. Bowden. 1998. Clinical, cellular, and molecular factors that contribute to antifungal drug resistance. Clin. Microbiol. Rev. 11:382-402.
White, T. C. 1997. Increased mRNA levels of ERG16, CDR, and MDR1 correlate with increases in azole resistance in Candida albicans isolates from a patient infected with human immunodeficiency virus. Antimicrob. Agents Chemother. 41:1482-1487.
Wickens, H. J., R. J. Pinney, D. J. Mason, and V. A. Gant. 2000. Flow cytometric investigation of filamentation, membrane patency, and membrane potential in Escherichia coli following ciprofloxacin exposure. Antimicrob. Agents Chemother. 44:682-687.(A. Lepak, J. Nett, L. Lin)
ABSTRACT
Pharmacodynamics (PD) considers the relationship between drug exposure and effect. The two factors that have been used to distinguish the PD behaviors of antimicrobials are the impact of concentration on the extent of organism killing and the duration of persistent microbiologic suppression (postantibiotic effect). The goals of these studies were (i) to examine the relationship between antimicrobial PD and gene expression and (ii) to gain insight into the mechanism of fluconazole effects persisting following exposure. Microarrays were used to estimate the transcriptional response of Candida albicans to a supra-MIC F exposure over time in vitro. Fluconazole at four times the MIC was added to a log-phase C. albicans culture, and cells were collected to determine viable growth and for microarray analyses. We identified differential expression of 18% of all genes for at least one of the time points. More genes were upregulated (n = 1,053 [16%]) than downregulated (174 [3%]). Of genes with known function that were upregulated during exposure, most were related to plasma membrane/cell wall synthesis (18%), stress responses (7%), and metabolism (6%). The categories of downregulated genes during exposure included protein synthesis (15%), DNA synthesis/repair (7%), and transport (7%) genes. The majority of genes identified at the postexposure time points were from the protein (17%) and DNA (7%) synthesis categories. In subsequent studies, three genes (CDR1, CDR2, and ERG11) were examined in greater detail (more concentration and time points) following fluconazole exposure in vitro and in vivo. Expression levels from the in vitro and in vivo studies were congruent. CDR1 and CDR2 transcripts were reduced during in vitro fluconazole exposure and during supra-MIC exposure in vivo. However, in the postexposure period, the mRNA abundance of both pumps increased. ERG11 expression increased during exposure and fell in the postexposure period. The expression of the three genes responded in a dose-dependent manner. In sum, the microarray data obtained during and following fluconazole exposure identified genes both known and unknown to be affected by this drug class. The expanded in vitro and in vivo expression data set underscores the importance of considering the time course of exposure in pharmacogenomic investigations.
INTRODUCTION
The time course of antimicrobial activity is dependent on a drug's pharmacokinetics and two major pharmacodynamic characteristics (2, 13). The first is the rate of organism killing and whether increasing drug concentrations enhance the rate and extent of killing. The second is the presence or absence of inhibitory effects on organism growth which persist after drug levels have fallen below the MIC. These persistent effects are termed postantibiotic effects (PAE) (3, 4, 5, 9, 14, 16, 29, 38, 39, 40). The PAE pharmacodynamic phenomenon underscores the importance of investigating the effects of antimicrobial exposure over time. Drugs from the triazole class demonstrate time-dependent killing but prolonged persistent effects in vivo.
Several investigators have undertaken studies to discern the biologic basis for the PAE phenomenon with several antibacterial compounds (7, 11, 17-20, 43). For example, the duration of the PAE for quinolones and ?-lactams correlates closely with the time it takes for bacterial DNA synthesis to resume. Similar study with the aminoglycosides has shown that the duration of the PAE best correlates with the time it takes for protein synthesis to resume.
A number of investigations have examined the global transcriptional profile of Candida albicans following in vitro triazole exposure (1, 6, 15, 23). These important studies have provided insight into the mechanism of action of these compounds. However, these studies have not considered the impact of the pharmacodynamics or time course of exposure on the response of the organism. The goal of the current experiments was to describe the global response of C. albicans transcription during the time course of fluconazole exposure. The information obtained provides a mechanistic model of the molecular events during the time course of exposure. Furthermore, the results from these studies underscore the importance of considering antimicrobial pharmacodynamics in studies investigating the drug-organism interaction. Finally, these investigations point to the relevance of considering molecular end points in antimicrobial pharmacokinetic/pharmacodynamic studies.
MATERIALS AND METHODS
Microorganism and in vitro susceptibility testing. Candida albicans K1 was used for all studies (3). Susceptibility testing was performed in duplicate on two occasions, using the CLSI (formerly NCCLS) M-27A method (30).
In vitro fluconazole exposure. (i) Microarray study of microbiologic effect and gene expression. C. albicans K1 was grown in vitro in the presence or absence of fluconazole for comparison of the time courses of gene expression using cDNA microarrays. An overnight culture was diluted to an initial inoculum of 105 CFU/ml and grown in RPMI-morpholinepropanesulfonic acid (RPMI-MOPS) buffered with MOPS at 37°C with shaking at 250 rpm. The treatment group was exposed to fluconazole at a concentration of four times the MIC, which was added after 3 h of growth and continued for 3 h. Following drug exposure, cells were pelleted, washed, and resuspended in fresh medium, and incubation continued for an additional 12 h. Samples for CFU enumeration and RNA isolation were collected at two time points during drug exposure (1 and 3 h) and two time points after drug removal (6 and 9 h). The growth and drug exposure experiment described above was repeated on three occasions. The experiment design was consistent with proposed microarray study standards (21).
(ii) Confirmation of microbiologic effect and gene expression and expanded time course. C. albicans K1 was similarly grown in vitro in the presence or absence of fluconazole for comparison of gene expression levels, using both Northern blotting and quantitative reverse transcription-PCR (RT-PCR), to confirm the results from the microarray study (10, 35). In addition, these studies were used to expand the evaluation of the relationship between the concentration/time of exposure and the transcriptional profile of the cells. The treatment groups were exposed to one of five increasing (fourfold) fluconazole concentrations. The exposures included concentrations at the MIC (1x), above the MIC (4x and 16x), and below the MIC (0.06x and 0.25x). Fluconazole was added after 6 h of growth (to cells in log-phase growth) and continued for 3 h. The cells were then pelleted, washed, resuspended in fresh medium, and incubated for an additional 6 h. Samples for CFU enumeration were collected at 3-h intervals. Samples for RNA isolation were collected at one time point during drug exposure (after 3 h of fluconazole exposure) and one time point after drug removal (3 h after fluconazole removal). Experiments were repeated on three occasions. Candida cells were harvested for RNA isolation as previously described (37).
Pharmacokinetics. The pharmacokinetic values used in the current study were previously determined for the animal model used here (3). Briefly, the time courses of fluconazole concentrations in mouse serum were determined for three fourfold increasing doses (6.25, 25, and 100 mg/kg of body weight) after subcutaneous administration to neutropenic infected mice, using a gas chromatography method. Pharmacokinetic constants, including the elimination half-life, area under the serum concentration curve (AUC), and peak level (Cmax), were calculated using a noncompartmental model.
Animal infection model and in vivo fluconazole exposure. A neutropenic murine model of disseminated candidiasis was used to determine if the drug exposure-gene expression time course observed in vitro was similar to that observed in vivo (2, 31). The animal model has been described in detail in prior publications (2, 3).
Following infection, groups of mice received one of three fourfold increasing fluconazole dose levels (0.78, 3.125, and 12.5 mg/kg) administered subcutaneously during the log phase of growth (18 h after infection). The dose levels were chosen to include a sub-MIC exposure only (0.78 mg/kg) and two supra-MIC exposures (3.125 and 12.5 mg/kg). The serum pharmacokinetics of the last two doses were estimated to achieve concentrations exceeding the MIC of the organism for 9.6 and 15.4 h, respectively.
Groups of two animals from untreated control and fluconazole treatment groups were euthanized at 3- to 6-h intervals over a period of 45 h for CFU enumeration as previously described (3). Results were expressed as the mean CFU/2 kidneys from two mice. The serum pharmacokinetic results were used to estimate the time that the total drug level of fluconazole remained above the MIC for the organism. The postantifungal effect (PAFE) was calculated by determining the time that it took the burden of organisms in the untreated control mice to increase 1 log10 CFU/kidney (C) and subtracting this from the amount of time that it took the organisms from the treated animals to grow 1 log10 CFU/kidney (T) after levels had fallen below the MIC, as follows: PAFE = T – C (14).
Groups of 10 mice per time point were euthanized at 3- to 12-h intervals for Candida RNA isolation from mouse kidneys as recently described (2). Fungal cells were collected from animals receiving the highest fluconazole dose level at five time points (3, 6, 18, 21, and 27 h) to include two supra-MIC and three postexposure (serum levels estimated to have fallen below the C. albicans MIC) sample values. Following the middle dose, organisms were harvested for RNA isolation at two points which corresponded to postexposure times (after serum levels fell below the MIC [12 and 15 h]). Two sampling time points were used for the lowest time point (3 and 6 h), both corresponding to estimated sub-MIC exposures.
Transcriptional profiles of cell populations. Studies were undertaken to compare the global responses of C. albicans during and following fluconazole exposure in vitro over time, using cDNA microarrays (version 6.31) produced at the Biotechnology Research Institute, National Research Council, Montreal, Canada (http://www.irb-bri.cnrc.gc.ca/services/microarray/scanning_e.html) (2, 12). The microarray used is based on C. albicans SC5314. The in vitro transcriptional profiles of a population of C. albicans cells during and following fluconazole exposure were examined relative to those of nonexposed cells collected at the same time periods of growth, as described above. At each of the specified time points, cells were collected by centrifugation at 3,000 x g for 2 min at ambient temperature, and the cell pellets were flash frozen in an ethanol-dry ice bath. Candida RNAs were isolated by the hot phenol method (37). mRNAs were purified using a Fast Track mRNA isolation kit (Invitrogen). Each set of samples was used for separate array hybridizations with cDNAs from the drug-exposed and control populations. In the first hybridization, the drug-exposed sample was labeled with Cy3 and the control population was labeled with Cy5. In the second hybridization, the exposed sample was labeled with Cy5 and the control population was labeled with Cy3. A total of six hybridizations were performed per time point, as previously described (2, 12).
Microarrays were scanned using a ScanArray 5000 instrument (Packard BioScience, Billerica, MA) at a 10-μm resolution. The intensities of the spots were quantified with QuantArray (GSI Lumonics, Billerica, MA). QuantArray files were analyzed in Excel (Microsoft, Redmond, WA). To be included in normalization and analysis, each spot had to satisfy three quality control criteria (2, 12). The signal intensity minus half of the standard deviation had to be greater than the local background plus half of the standard deviation; the signal intensity had to be within the dynamic range of the photomultiplier tube; and the raw intensities of the duplicate spots for each gene had to be within 50% of each other. For spots that met these criteria, we examined the log ratio of the intensity of expression from the fluconazole-exposed cell population to that from the unexposed cells. The results presented consist of the averages of three independent experiments with dye swaps. When either channel value was below 100.0, the data point was considered unacceptable. A normalization factor was applied to account for systematic differences in probe label intensities. A two-sided Student t test was used to assess the statistical significance of log2 ratios.
Gene ontology identification was undertaken for differentially expressed genes, using Candida (CandidaDB [http://genolist.pasteur.fr/CandidaDB/] and Candida Genome Database [http://www.candidagenome.org/]) and Saccharomyces (Saccharomyces Genome Database [http://genome-www.Stanford.edu/Saccharomyces/]) (8) databases.
Confirmation of microarray expression data. The mRNA abundance of a subset of genes identified in the microarray analysis was estimated using either Northern blotting or RT-PCR (10, 35). The specific C. albicans genes examined included CDR1, CDR2, and ERG11 (34, 36, 41, 42). CDR1 and CDR2 encode efflux pumps for which fluconazole is a substrate. ERG11 encodes the fluconazole drug target, lanosterol 14-demethylase. Prior investigations have demonstrated that the expression of these genes can be induced by triazole exposure (22, 28). The current studies were intended to expand the examination of the relationship between the pharmacodynamics of the fluconazole exposure (effects of concentration and time) and the expression of these genes of interest. The initial cell concentrations, growth conditions, drug exposures, and sampling time points are described above. Growth experiments and mRNA abundance assays were performed in triplicate.
For both Northern blotting and RT-PCR, cells were harvested and RNAs were isolated as described above. Northern blots were performed using standard methodology. RT-PCR studies utilized TaqMan methodology (24). Probes for Northern blots and RT-PCR (primer and probe sets) have been previously described and were generated for CDR1, ERG11, and ACT1 by PCR amplification from Candida genomic DNA (see Table S5 in the supplemental material) (26, 34, 36, 41). For both methods, actin was utilized as a constitutively expressed control.
Band intensities in Northern blot studies were estimated using Image J software (NIH). Gene intensities were expressed as the background normalized ratios for the conditions of interest described above. RT-PCR quantitative data analysis was completed by using a modification of the CT method of Livak and Schmittgen (24, 25). For each amplification run, the threshold cycle, CT, of the target was normalized to the CT of the ACT1 gene (CT = CT of target – CT of normalizer). We compared the trends in CT values over time for cells grown in the presence of fluconazole and those grown in the absence of fluconazole. Real-time quantitative PCR was performed on three 10-fold cDNA dilutions to ensure similar primer efficiencies among the various amplified targets (2).
RESULTS
In vitro susceptibility testing. The fluconazole MIC for C. albicans K1 is 0.5 μg/ml.
Pharmacokinetics. The pharmacokinetic profile was linear, with the peak level and AUC increasing proportionally for each dose escalation. Peak levels occurred within 1 h of administration and ranged from 5.2 to 100 μg/ml (Table 1). The elimination half-lives ranged from 2.8 to 3.5 h. The 24-h AUCs ranged from 24 to 460 mg · h/liter.
In vitro fluconazole exposure. (i) Microarray study of microbiologic effect and gene expression. The starting inoculum of C. albicans in the RPMI-MOPS medium was 5.0 log10 CFU/ml. After 3 hours of growth, the viable counts had increased to 6.2 ± 0.14 log10 CFU/ml, at which time fluconazole was added to one of the flasks (Fig. 1). Over the 3-h drug exposure time at a fourfold excess of the MIC, the burden of viable cells did not increase. During the same time period, the untreated control cells grew slightly (0.45 log10 CFU/ml). After fluconazole removal, the growth curve assumed the trajectory of the untreated control. Persistent growth suppression following in vitro drug exposure (PAFE) was not observed. The change in viable counts over the next 3 hours increased 1.4 and 1.1 log10 CFU/ml in the previously exposed and untreated cell populations, respectively. The slope of the growth curve began to plateau for both cell populations between the 6- and 9-h time points, with increases in burden ranging from 0.2 to 0.6 log10 CFU/ml. In general, the growth phases were similar for both cell populations at the various sampling time points.
Microarray analysis over the time course of exposure identified more than 1,000 differentially expressed genes (see Tables S1, S2, S3, and S4 in the supplemental material). Over the entire experiment, more genes were upregulated (n = 1053 [16%]) than downregulated (174 [3%]). However, expression data from the time points during F exposure identified equal numbers of both up (2%)- and downregulated (2%) genes. The functions of the majority of the genes in both groups are unknown (42% of upregulated genes and 32% of downregulated). Of the upregulated genes (Fig. 2) with known ontology during exposure, many were related to plasma membrane/cell wall synthesis/maintenance (18%), stress responses (7%), and metabolism (8%). Nine genes from the ergosterol synthesis pathway were upregulated during fluconazole exposure. Only two genes from this group precede the fluconazole target step in the ergosterol synthesis pathway. The categories of downregulated genes during exposure (Fig. 2) included protein synthesis (15%), DNA synthesis/repair (7%), and transport (7%). The postexposure data time points (Fig. 3) included mainly upregulation (16%) (<1% downregulated). The majority of upregulated genes were from the protein (17%) and DNA (7%) synthesis or modification categories. Several genes from the ergosterol pathway remained elevated; however, the degree of upregulation was lower than that during fluconazole exposure.
(ii) Confirmation of microbiologic effect and gene expression and expanded time course. The starting inoculum of C. albicans in the RPMI-MOPS medium was near 5.0 log10 CFU/ml (Fig. 4). During the first 3 hours of growth, the viable counts had increased slightly. The cells shifted to log-phase growth over the subsequent 3 h (6-h time point), increasing >2 log10 CFU/ml, at which point the various fluconazole concentrations were introduced to the specified cell populations. Over the 3-hour drug exposure period, the sub-MIC concentrations exhibited minimal effects on the continued growth of cells. The larger fluconazole concentrations did inhibit growth in a dose-dependent fashion. However, after the removal of fluconazole, the slopes of the growth curves rapidly returned to that of the untreated control. Similar to our other in vitro experiment, there was no measurable microbiologic postantifungal effect with any of the fluconazole exposures (16).
The abundance of ERG11 (Fig. 5) was increased during fluconazole exposure (after 3 h of exposure) relative to that in untreated controls for each of the drug concentrations examined. The abundance increased relative to the fluconazole concentration. The smallest change in ERG11 expression in treated relative to untreated control cells was 1.2-fold at a concentration of one-eighth the MIC. The largest change was 2.3-fold, which was associated with exposure to fluconazole at a concentration 16-fold higher than the MIC. The differences in abundance for the two highest concentrations were statistically different from those for the three lowest concentrations (P < 0.001). The increase in mRNA abundance associated with the various drug exposures declined following 3 h of growth in the absence of fluconazole. The decline in expression during the postexposure period was similar to that observed with other genes in the ergosterol synthesis pathway from the microarray study.
Both Northern blots and quantitative RT-PCR were utilized to examine the relative mRNA abundances of the CDR1 gene associated with the same fluconazole concentrations (Fig. 5). Following a 3-hour fluconazole exposure, the mRNA abundance of CDR1 did not increase relative to that in untreated control cells for any of the five drug concentrations. However, during the regrowth period following fluconazole exposure, the expression of CDR1 increased for four of the five drug concentrations. The degree of upregulation was dependent on the fluconazole concentration, increasing >2.5-fold in association with exposure to the two highest fluconazole concentrations. The differences in expression for the two highest fluconazole concentrations were statistically different from those for the three lowest concentrations (P < 0.01).
Quantitative RT-PCR was used for estimations of the expression of CDR2 (Fig. 5). The relationship between the fluconazole exposure and mRNA abundance for CDR2 was remarkably similar to that for CDR1. During fluconazole exposure (3 h), the relative expression of CDR2 was no different in exposed and untreated cells. However, at the postexposure time point, the abundance of CDR2 increased markedly in exposed cells in a concentration-dependent manner. The 16x and 4x MIC exposures resulted in 10- and 6-fold increases in CDR2 expression relative to that in untreated controls (P < 0.001). The expression values for the lower drug concentrations were slightly higher than those for control cells and for the prior time point; however, these differences were insignificant.
Microbiologic effect and gene expression after in vivo fluconazole exposure. Two hours after intravenous inoculation, the viable burden of C. albicans in mouse kidneys was 3.23 ± 0.05 log10 CFU/2 kidneys (Fig. 6). At the time of fluconazole administration, the cells were in the late log phase of growth, with a viable kidney burden of 6.84 ± 0.49 log10 CFU/2 kidneys 18 h after the initiation of infection. The two lowest fluconazole drug concentrations had no detectable impact on viable counts in neutropenic mouse kidneys. The highest fluconazole dose level did inhibit growth for at least 6 h after drug administration. Within 18 h after the largest fluconazole dose, the viable organism burden began to increase. In contrast to prior in vivo fluconazole investigations with this model, a measurable PAFE was not apparent (3). We speculate that the lack of a measurable PAFE is related to the much later timing of drug administration (18 h compared to 2 h after infection).
The mRNA abundances of the two CDR pumps in relation to drug exposure over time are shown in Fig. 7 (CDR1 and CDR2 data are on the left and right, respectively). At each time point, the mRNA abundance is presented as a ratio for drug-treated cells to untreated controls. The CDR pump abundance was not elevated in response to the two supra-MIC concentrations or the single sub-MIC concentration. However, the mRNA abundance of both pumps was remarkably increased during the postantifungal period of time. This response was not related to the rate or phase of growth, as the control strain's growth was quite similar. The abundances of both transcripts increased from 12 to 21 h after the initial drug exposure and began to decline at the final time point (27 h). The maximal change in treated relative to untreated cells in vivo was 18- and 3.5-fold for CDR2 and CDR1, respectively.
DISCUSSION
The field of antimicrobial pharmacodynamics investigates the relationships between drug concentrations relative to the MIC over time and outcomes (13). Understanding these relationships has had an enormous impact on the manner in which these compounds are administered. The majority of pharmacodynamic investigations examine treatment efficacy by using a microbiologic end point (3, 9, 13, 14, 16, 39, 40). Few other drug effect measures have been considered in these types of studies. Studies with several antibacterial compounds have examined cell morphology and protein, DNA, and RNA synthesis end points by measuring the incorporation of DNA, RNA, and protein radiolabeled precursors during the postantibiotic effect in vitro in an attempt to understand the biological basis for the phenomenon (9, 17). These investigations have not yet taken advantage of newer genomic technologies to further understand these processes.
Pharmacodynamic evaluations of triazole antifungals have shown that these drugs inhibit the growth of fungi in a time-dependent manner (3, 4, 16, 27). Exposure to increasingly large concentrations does not enhance the inhibitory activities of these compounds. However, following exposure to supra-MIC concentrations in vivo, regrowth of fungi is delayed (3). This phenomenon of persistent growth inhibition following antifungal drug exposure has not been investigated mechanistically. Several investigators have, however, utilized microarray technology to examine the global transcriptional responses of C. albicans and Saccharomyces cerevisiae to several triazole compounds (1, 6, 15, 23). These studies have been important for furthering our understanding of the mechanism of action of these drugs. The triazole exposures used in these in vitro investigations have varied from study to study. First, the duration of drug exposure prior to RNA sampling has varied among studies. In addition, the studies have all used supra-MIC exposures and a single sampling time point. Thus, there has been minimal consideration of the pharmacokinetics of drug exposure. The current investigation examined genomic end points during the time course of triazole-C. albicans interaction. The goals of the study were (i) to determine if gene expression in this fungal pathogen is impacted by the pharmacokinetics of drug exposure and (ii) to further our understanding of the molecular basis for the time course effects of fluconazole.
The results from the current microarray studies underscore the importance of examining the impact of drug exposure over time. There were numerous genes identified at only one of the four time points. However, overall, the differential expression of cells collected from the two supra-MIC time points was similar. The same transcriptional pattern was also observed in the two postexposure cell populations. The ontology categories most commonly identified from genes upregulated during fluconazole exposure are similar to those identified in previous triazole-yeast exposure microarray studies (1, 6, 15, 23). The predominant categories of known function included those related to plasma membrane and cell wall synthesis, maintenance, and function. The impact of fluconazole was not surprisingly marked with regard to the ergosterol synthesis and sterol modification pathway. Among the genes affected in this pathway, six of seven were at or downstream of the fluconazole drug target. This pattern of expression in this pathway is similar to that previously reported for triazole exposure of S. cerevisiae and C. albicans. Another common subset of expressed genes included those previously shown to be involved in the cell stress response. The third most common functional group of genes with increased mRNA abundance during fluconazole exposure was the group involved in metabolism. Among this group, genes related to carbohydrate metabolism were present most frequently. The number of downregulated genes was small compared to the number of upregulated genes. The two most common ontology categories for this set of genes included genes related to DNA and protein synthesis. It is interesting that prior antibacterial PAE studies similarly demonstrated reduced DNA and protein synthesis during the period of microbiologic growth suppression. Conversely, genes expressed in our postexposure period, after which microbiologic recovery had begun, were marked primarily by the activation of nucleic acid and protein synthesis machineries. Perhaps not surprisingly, these genes were not identified in prior triazole-yeast interaction array investigations, which included only a single early exposure time point. This pattern of expression during the microbiologic effect and in the postexposure period can be used to propose a triazole time course activity model. The expression of genes associated with drug exposure and microbiologic activity represents the direct effect of the drug or additional organism perturbations in response to the sterol effects. The later set of upregulated genes suggested how the organism responds during the period of microbiologic recovery from the effects of the drug. Both groups of genes are interesting to consider for future study. The first group includes those upregulated during exposure, including, of particular interest, those associated with perturbation of the cell wall and plasma membrane. Any number of these gene products may provide useful drug targets, whose inhibition may potentiate the microbiologic effects of the triazole. Similarly, the set of genes identified during organism recovery may also provide interesting drug targets which, if inhibited, may reduce the ability of the organism to recover from the effects of the triazole and prolong the growth inhibition.
The impact of the time course of fluconazole pharmacokinetics was further examined for a subset of three genes. The expanded time course experiments included both a larger range of fluconazole concentrations and other time points during and following exposure. In addition, we examined the transcriptional responses of C. albicans following drug exposure both in vitro and in vivo. Importantly, the results of both in vitro and in vivo studies were congruent. The three genes chosen were CDR1, CDR2, and ERG11. The choice of these three genes was based on their identification in the current microarray data set as well as the fact that prior studies have demonstrated differential expression of each of the genes in response to triazole exposure (22, 28, 36, 42). The time course expression data for these genes confirmed the microarray results. The mRNA abundance of ERG11 was highest during fluconazole exposure and declined in the post-exposure period. Conversely, the abundance of each of the CDR efflux pumps was greatest in the postexposure time period and may result in protection from additional fluconazole exposure during the postexposure period. These data again underscore the importance of considering the time course of events in studying the expression profile of the antimicrobial/organism interaction, as demonstrated previously from the microbiologic standpoint. The study of an expanded concentration range in these expression studies also demonstrated a strong pharmacodynamic relationship. The mRNA abundance was clearly impacted by the concentration of fluconazole over five exposure levels, which varied 256-fold. In addition, the two lowest concentrations were below the MIC for the organism and demonstrated an impact on gene expression for each of the three genes. Previous studies with both triazoles and polyenes have demonstrated sub-MIC effects using different biologic end points (32, 33, 39). One explanation for the discrepancy between in vitro and in vivo investigations of the triazole postantifungal effect has been unavoidable sub-MIC exposures which are present in the latter study design (3, 16). Additional investigation of these sub-MIC effects at the genomic level may offer further insight into the molecular basis for in vivo triazole postantifungal effects.
In sum, these studies demonstrate a relationship between the concentration and time course of exposure and the transcriptional response of an organism. The change in expression profile over time and the change in level of expression relative to concentration underscore the importance of considering the pharmacokinetics of drug exposure in microarray experiments investigating antimicrobial-pathogen interactions. The global transcriptional profile of the fluconazole-Candida albicans interactions suggests a damage response model in which the plasma membrane and cell wall are structurally and functionally damaged, followed by a period of recovery manifested by enhanced nucleic acid and protein synthesis to repair the cell.
Supplemental material for this article may be found at http://aac.asm.org/.
REFERENCES
Agarwal, A. K., P. D. Rogers, S. R. Baerson, M. R. Jacob, K. S. Barker, J. D. Cleary, L. A. Walker, D. G. Nagle, and A. M. Clark. 2003. Genome-wide expression profiling of the response to polyene, pyrimidine, azole, and echinocandin antifungal agents in Saccharomyces cerevisiae. J. Biol. Chem. 278:34998-35015.
Andes, D., A Lepak, A. Pitula, K. Marchillo, and J. Clark. 2005. A simple approach for estimating gene expression in Candida albicans directly from a systemic infection site. J. Infect. Dis. 192:893-900.
Andes, D., and M. van Ogtrop. 1999. Characterization and quantitation of the pharmacodynamics of fluconazole in a neutropenic murine disseminated candidiasis model. Antimicrob. Agents Chemother. 43:2116-2120.
Andes, D. 2003. In vivo pharmacodynamics of antifungal drugs in treatment of candidiasis. Antimicrob. Agents Chemother. 47:1179-1186.
Andes, D., T. Stamstad, and R. Conklin. 2001. Pharmacodynamics of amphotericin B in a neutropenic mouse disseminated candidiasis model. Antimicrob. Agents Chemother. 45:922-926.
Bammert, G. F., and J. M. Fostel. 2000. Genome-wide expression patterns in Saccharomyces cerevisiae: comparison of drug treatments and genetic alterations affecting biosynthesis of ergosterol. Antimicrob. Agents Chemother. 44:1255-1265.
Barmada, S., S. Kohlhepp, J. Leggett, R. Dworkin, and D. Gilbert. 1993. Correlation of tobramycin-induced inhibition of protein synthesis with postantibiotic effect in Escherichia coli. Antimicrob. Agents Chemother. 37:2678-2683.
Braun, B. R., M. van het Hoog, C. d'Enfert, C. M. Martchenko, J. Dungan, A. Kuo, D. O. Inglis, M. A. Uhl, H. Hogues, M. Berriman, M. Lorenz, A. Levitin, U. Oberholzer, C. Bachewich, D. Harcus, A. Marcil, D. Dignard, T. Iouk, R. Zito, L. Frangeul, F. Tekaia, K. Rutherford, E. Wang, C. A. Munro, S. Bates, N. A. Gow, L. L. Hoyer, G. K?hler, J. Morschh?user, G. Newport, S. Znaidi, M. Raymond, B. Turcotte, G. Sherlock, M. Costanzo, J. Ihmels, J. Berman, D. Sanglard, N. Agabian, A. P. Mitchell, A. D. Johnson, M. Whiteway, and A. Nantel. 2005. A human-curated annotation of the Candida albicans genome. PLOS Genet. 1:e1.
Bundtzen, R. W., A. U. Gerber, D. L. Cohn, and W. A. Craig. 1981. Postantibiotic suppression of bacterial growth. Rev. Infect. Dis. 3:28-37.
Bustin, S. A. 2000. Absolute quantification of mRNA using real-time reverse transcription polymerase chain reaction assays. J. Mol. Endocrinol. 25:169-193.
Champney, W. S., and C. L. Tober. 1999. Molecular investigation of the postantibiotic effects of clarithromycin and erythromycin on Staphylococcus aureus cells. Antimicrob. Agents Chemother. 43:1324-1328.
Cowen, L. E., A. Nantel, M. S. Whiteway, D. Y. Thomas, D. C. Tessier, L. M. Kohn, and J. B. Anderson. 2002. Population genomics of drug resistance in Candida albicans. Proc. Natl. Acad. Sci. USA 99:9284-9289.
Craig, W. A. 1998. Pharmacokinetic/pharmacodynamic parameters: rationale for antibacterial dosing of mice and men. Clin. Infect. Dis. 26:1-12.
Craig, W. A., and S. Gudmundsson. 1996. Postantibiotic effect, p. 296-329. In V. Lorian (ed.), Antibiotics in laboratory medicine, 4th ed. Williams and Wilkins, Baltimore, Md.
DeBacker, M. D., T. Ilyina, X. J. Ma, S. Vandoninck, W. H. Luyten, and H. Vanden Bossche. 2001. Genomic profiling of the response of Candida albicans to itraconazole treatment using a DNA microarray. Antimicrob. Agents Chemother. 45:1660-1670.
Ernst, E., M. E. Klepser, and M. A. Pfaller. 2000. Postantifungal effects of echinocandin, azole, and polyene antifungal agents against Candida albicans and Cryptococcus neoformans. Antimicrob. Agents Chemother. 44:1108-1111.
Gottfredsson, M., H. Erlendsdottir, A. Gudmundsson, and S. Gudmundsson. 1995. Different patterns of bacterial DNA synthesis during postantibiotic effect. Antimicrob. Agents Chemother. 39:1314-1319.
Gottfredsson, M., H. Erlendsdottir, R. Kolka, A. Gudmundsson, and S. Gudmundsson. 1993. Ultrastructural alterations of bacteria during the postantibiotic effect. Chemotherapy 39:153-162.
Gottfredsson, M., H. Erlendsdottir, A. Sigfusson, and S. Gudmundsson. 1998. Characteristics and dynamics of bacterial populations during postantibiotic effect determined by flow cytometry. Antimicrob. Agents Chemother. 42:1005-1011.
Guan, L., R. M. Blumenthal, and J. C. Burnham. 1992. Analysis of macromolecular biosynthesis to define the quinolone-induced postantibiotic effect in Escherichia coli. Antimicrob. Agents Chemother. 36:2118-2124.
Hingamp, J., J. Quackenbush, G. Sherlock, P. Spellman, C. Stoeckert, J. Aach, W. Ansorge, C. A. Ball, H. C. Causton, T. Gaasterland, P. Glenisson, F. C. P. Holstege, I. F. Kim, V. Markowitz, J. C. Matese, H. Parkinson, A. Robinson, U. Sarkans, S. Schulze-Kremer, J. Stewart, R. Taylor, J. Vilo, and M. Vingron. 2001. Minimum information about a microarray experiment (MIAME)—toward standards for microarray data. Nat. Genet. 29:365-371.
Karababa, M., A. T. Coste, B. Rognon, J. Bille, and D. Sanglard. 2004. Comparison of gene expression profiles of Candida albicans azole-resistant clinical isolates and laboratory strains exposed to drugs inducing multidrug transporters. Antimicrob. Agents Chemother. 48:3064-3079.
Liu, T. T., R. E. Lee, K. S. Barker, R. E. Lee, L. Wei, R. Homayouni, and P. D. Rogers. 2005. Genome-wide expression profiling of the response to azole, polyene, echinocandin, and pyrimidine antifungal agents in Candida albicans. Antimicrob. Agents Chemother. 49:2226-2236.
Livak, K. J., S. J. Flood, J. Marmaro, W. Giusti, and K. Deetz. 1995. Oligonucleotides with fluorescent dyes at opposite ends provide a quenched probe system useful for detecting PCR product and nucleic acid hybridization. PCR Methods Appl. 4:357-362.
Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2CT method. Methods 25:402-408.
Losberger, C., and J. F. Ernst. 1989. Sequence of Candida albicans gene encoding actin. Nucleic Acids Res. 17:9488.
Louie, A., G. L. Drusano, P. Banerjee, Q. F. Liu, P. Kaw, M. Shayegani, H. Taber, and M. H. Miller. 1998. Pharmacodynamics of fluconazole in a murine model of systemic candidiasis. Antimicrob. Agents Chemother. 42:1105-1109.
Lyons, C. N., and T. C. White. 2000. Transcriptional analyses of antifungal drug resistance in Candida albicans. Antimicrob. Agents Chemother. 44:2296-2303.
Moore, R. D., P. S. Lietman, and C. R. Smith. 1987. Clinical response to aminoglycoside therapy: importance of the ratio of peak concentration to minimal inhibitory concentration. J. Infect. Dis. 155:93-99.
National Committee for Clinical Laboratory Standards. 1997. Reference method for broth dilution antifungal susceptibility testing for yeast. Accepted standard M27-A. National Committee for Clinical Laboratory Standards, Wayne, Pa.
National Research Council Committee on the Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, Commission on Life Sciences. 1996. Guide for the care and use of laboratory animals. National Academy Press, Washington, D.C.
Nosanchuk, J. D., W. Cleare, S. P. Franzot, and A. Casadevall. 1999. Amphotericin B and fluconazole affect cellular charge, macrophage phagocytosis, and cellular morphology of Cryptococcus neoformans at subinhibitory concentrations. Antimicrob. Agents Chemother. 43:233-239.
Nugent, K. M., and K. R. Couchot. 1986. Effects of sublethal concentrations of amphotericin B on Candida albicans. J. Infect. Dis. 154:665-669.
Prasad, R., P. De Wergifosse, A. Goffeau, and E. Balzi. 1995. Molecular cloning and characterization of a novel gene in Candida albicans, CDR1, conferring multiple drug resistance to drugs and antifungals. Curr. Genet. 27:320-329.
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Sanglard, D., F. Ischer, M. Monod, and J. Bille. 1997. Cloning of Candida albicans genes conferring resistance to azole antifungal agents: characterization of CDR2, a new multidrug ABC transporter gene. Microbiology 143:405-416.
Schmitt, M. E., T. A. Brown, and B. L. Trumpower. 1990. A rapid and simple method for preparation of RNA from Saccharomyces cerevisiae. Nucleic Acids Res. 18:3091-3092.
Tornqvist, I. O., S. E. Holm, and O. Cars. 1990. Pharmacodynamic effects of subinhibitory antibiotic concentrations. Scand. J. Infect. Dis. 74(Suppl.):94-101.
Turnidge, J. D., S. Gudmundsson, B. Vogelman, and W. A. Craig. 1994. The postantibiotic effect of antifungal agents against common pathogenic yeasts. J. Antimicrob. Chemother. 34:83-92.
Vogelman, B., S. Gudmundsson, J. Turnidge, J. Leggett, and W. A. Craig. 1988. In vivo postantibiotic effect in a thigh infection in neutropenic mice. J. Infect. Dis. 157:287-298.
White, T. C., K. A. Marr, and R. A. Bowden. 1998. Clinical, cellular, and molecular factors that contribute to antifungal drug resistance. Clin. Microbiol. Rev. 11:382-402.
White, T. C. 1997. Increased mRNA levels of ERG16, CDR, and MDR1 correlate with increases in azole resistance in Candida albicans isolates from a patient infected with human immunodeficiency virus. Antimicrob. Agents Chemother. 41:1482-1487.
Wickens, H. J., R. J. Pinney, D. J. Mason, and V. A. Gant. 2000. Flow cytometric investigation of filamentation, membrane patency, and membrane potential in Escherichia coli following ciprofloxacin exposure. Antimicrob. Agents Chemother. 44:682-687.(A. Lepak, J. Nett, L. Lin)