Use of Time-Saving Flow Cytometry for Rapid Determination of Resistance of Human Cytomegalovirus to Ganciclovir
http://www.100md.com
微生物临床杂志 2005年第10期
The Clinical Research Institute, St. Mary's Hospital, The Catholic Hemopoietic Stem Cell Transplantation Center, The Catholic University of Korea, College of Medicine, Seoul 150-713, Korea
Water Quality Analysis Team I, International Water Analysis Center, Korea Water Resources Corporation, Daejeon 306-711, Korea
Department of Microbiology, Natural Sciences College, Chungbuk National University, Cheongju 361-763, Korea
ABSTRACT
There are two ways to assess the susceptibility of human cytomegalovirus (HCMV) to ganciclovir (GCV): one is a genotypic test that detects resistance-related mutations and the other is a phenotypic test that actually assesses susceptibility. The advantages of genotyping the UL97 gene are its rapidity and accuracy. However, to detect novel mutations or mutations affecting the UL54 DNA polymerase, a phenotypic test such as the plaque reduction assay (PRA) is also required. To avoid the shortcomings of PRA such as its time-consuming nature and labor-intensiveness, we developed a time-saving fluorescence-activated cell sorting (TS-FACS) technique. We obtained a GCV 50% inhibitory concentration (IC50) from five clinical isolates and an HCMV laboratory strain (AD169) and compared the results with those from the PRA. The laboratory strain and three clinical isolates were sensitive to GCV. Although there was a minor discrepancy in the case of one of the three isolates, the GCV IC50 values obtained by TS-FACS analysis correlated well with the results of the PRA. The remaining two isolates were resistant to GCV; one was GCV resistant due to the mutation M460V, and the GCV IC50 results obtained by TS-FACS analysis and by PRA were also comparable. The advantages of TS-FACS analysis are the shorter time required, the possibility of automation, and its comparability to PRA, considered the gold standard. Thus, TS-FACS analysis may be useful as an alternative to PRA in the clinic.
INTRODUCTION
Human cytomegalovirus (HCMV) is a betaherpesvirus that causes serious diseases in immunocompromised hosts, particularly transplantation recipients. Drugs used currently against HCMV include ganciclovir (GCV), foscarnet, and cidofovir, and of these, GCV is the initial drug of choice (11, 13). However, the advent of HCMV resistant to GCV has made control of HCMV difficult. Mutant HCMV strains strongly resistant to GCV are unable to phosphorylate GCV owing to mutations in key regions of the UL97 gene (domains VIII, VI, and IX) (1, 2, 5, 8, 9, 21, 26). Mutations in UL54, the viral DNA polymerase, also confer resistance to GCV (10, 15, 24).
There are two types of methods for assessing the susceptibility of HCMV to GCV: one is genotypic and detects the resistance-related mutant genes, and the other is phenotypic and assesses resistance directly (3, 13). The genotypic methods include DNA sequencing, detection of restriction fragment length polymorphisms (RFLP), and probe-specific and primer-specific hybridization. DNA sequencing is the reference method for detecting resistance-related GCV mutations, and RFLP has been widely used to detect GCV resistance due to mutations in the UL97 gene (4, 7). Although genotypic tests are fast, phenotypic tests are necessary to detect novel mutations and mutations in the UL54 DNA polymerase gene. Examples of phenotypic tests are the plaque reduction assay (PRA), in situ enzyme-linked immunosorbent assay, DNA reduction assay, and flow cytometry-based assay. The flow cytometric assay counts trypsinized fibroblasts expressing the HCMV antigen using a fluorochrome-labeled antibody and has the advantage that automation and quantitation are feasible (16, 18, 20).
PRA is the gold standard for evaluating the drug susceptibility of a virus. For HCMV, it consists of counting the number of viral plaques on fibroblasts. However, this conventional phenotypic assay is time-consuming and labor-intensive and requires highly skilled laboratory technicians (17, 25). This is because most clinical isolates examined grow slowly, so that it takes a long time to obtain the required virus titers (12). To overcome these shortcomings, Prix et al. and Landry et al. have reported a method without the titration step of classical PRA (17, 23).
We have modified and applied the method of mixed culture of cell-associated viruses to flow cytometric analysis and have compared this new time-saving fluorescence-activated cell sorting (TS-FACS) analysis with PRA. We have also assessed whether TS-FACS analysis could be used as a GCV susceptibility test for clinical HCMV isolates.
MATERIALS AND METHODS
Cells and virus. As the source of cells, human foreskin fibroblasts (HFF, ATCC CRL-2097; American Type Culture Collection, Manassas, VA) between passages 10 and 14 were used. Cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Cambrex Bio Science, Walkersville, MD) supplemented with 10% fetal bovine serum (FBS; U.S. Bio-Technologies Inc., Parkerford, PA) and penicillin (100 units/ml)-streptomycin (100 μg/ml) solution (JBI, Daegu, Korea), at 37°C, in a 5% CO2 incubator. HCMV strain AD169 was the source of virus, and the stock was obtained by plating fresh HFF inoculated with HCMV at a multiplicity of infection (MOI) of 0.01 to 0.03 in 100-mm cell culture plates (Nalge Nunc International, Rochester, NY); the infected cells were detached with a scraper when the cytopathic effect (CPE) had spread completely and collected in a new tube with culture medium. Cell debris was removed, the freeze-thaw cycle was repeated three times, and the suspension was centrifuged. The stock thus obtained was stored in 1 ml at –70°C, and the titer of the stock was 106 to 107 PFU/ml.
Clinical isolates. Three milliliters of blood was obtained from patients who were suspected of having a GCV-resistant HCMV infection and transferred to a 15-ml conical tube (Becton Dickinson Labware, Franklin Lakes, NJ). The blood was centrifuged at 800 x g for 15 min, the plasma layer was taken, 0.2 ml of this was used to infect fresh HFF prepared in advance, and fresh medium was substituted after 3 h. Plasma was collected separately, and to the remaining fraction containing blood cells was added phosphate-buffered saline (PBS; Sigma Diagnostics Inc., St. Louis, MO), adjusted to a final volume of 6 ml. This was loaded carefully, without disturbing the layers, into a 15-ml tube containing 3 ml Histopaque 1119 (Sigma Diagnostics Inc.). After centrifugation at 700 x g for 30 min, the white blood cell band formed above the Histopaque was transferred to a new 15-ml tube with a Pasteur pipette, washed with 10 ml PBS, and centrifuged at 800 x g for 10 min. The supernatant was removed, and the remaining cell pellet was washed with 5 ml PBS and centrifuged at 800 x g for 5 min. This was repeated twice, to yield the white blood cells. The cells were counted by the Trypan blue exclusion assay, and approximately 2 x 106 cells were inoculated into prepared HFF. The medium was changed every 4 to 5 days, and if CPE was detected, the cells were cultured with 2% FBS in DMEM until plaques formed. These plaques were then mixed with uninfected fresh HFF and cultured in T25 tissue culture flasks.
Plaque reduction assay. The classical plaque reduction assay was performed by culturing fresh HFF in 35-mm culture plates and infecting them with virus. The cells were washed with PBS and overlaid with the appropriate concentration of GCV (Roche Diagnostics, Basel, Switzerland) contained in semisolid medium consisting of DMEM plus 2% FBS, 0.25% SeaPlaque agarose (Cambrex Bio Science, Rockland, ME), 100 units/ml penicillin (JBI), 100 μg/ml streptomycin (JBI), and 250 ng/ml amphotericin B (JBI). Approximately 7 days later, a second overlay of the same semisolid medium containing various concentrations of GCV was added. When plaques had formed, the cells were fixed with 10% formalin in 0.85% saline solution. The fixed-cell monolayer was stained with 0.03% methylene blue, and the plaques were counted with a light microscope (BH-2; Olympus Corporation, Tokyo, Japan).
TS-FACS analysis. The overall process is illustrated in Fig. 1. When the CPE had spread over 50% of the T25 flask containing the mixed culture, the culture supernatant and infected cells were collected in a tube. Fresh HFF grown in 100-mm culture plates were suspended in 10 ml and divided into two tubes. HCMV-infected cells (0.2 or 1 ml) were mixed well with the uninfected cells and plated in six-well plates. Approximately 5 to 6 h later, when the cells had attached to the culture plates, GCV was added at concentrations of 0, 1, 3, 10, 30, or 100 μM. Three to four days later the cells were detached by trypsinization, and samples of 105 cells were added to 1-ml microcentrifuge tubes. The cells were washed once with PBS, fixed with 500 μl 0.2% paraformaldehyde (Sigma Diagnostics Inc.) at room temperature for 1 h, and exposed to 500 μl 0.2% Tween 20 (J. T. Baker, Phillipsburg, NJ) for 20 min for permeabilization. After being washed with 1 ml PBS, they were suspended in 100 μl PBS, and 1 drop of mouse anti-HCMV immediate-early (IE) protein fluorescein isothiocyanate-conjugated antibody (MAB 5090; Chemicon, Temecula, CA) was added according to the manufacturer's instructions, and incubation continued for 40 min in the dark at room temperature. The cells were then washed twice with 1 ml PBS and resuspended in 200 μl PBS, flow cytometric analysis was performed with a FACScalibur (BD Biosciences, San Jose, CA), and results were analyzed with Cell Quest (BD Biosciences). The 50% inhibitory concentration (IC50) was the median concentration that caused 50% inhibition of IE gene expression relative to that in the control HCMV-infected cells. All experiments were performed three times independently. GCV resistance was defined as cells with an IC50 exceeding 6 μM to 12 μM GCV (6, 7, 14).
DNA sequencing and data analysis. DNA was extracted from HCMV-infected cells using a QIAamp DNA minikit (QIAGEN, Hilden, Germany). Primers for the UL97 region (776 bp; nucleotides 142728 to 143503) were selected from HCMV AD169 (GenBank accession number NC_001347). The primers were 460F, 5'-GTTGGCCGACGCTATCAAAT-3', and 650R, 5'-GGTCCTCCTCGCAGATTATG-3' (22). After a hot-start step to activate i-star Taq (iNtRON Biotechnology, Sungnam, Korea), the samples underwent 40 cycles of denaturation at 94.5°C for 1 min, annealing at 55°C for 1 min, and extension at 72°C for 1 min, followed by a final extension at 72°C for 5 min. Amplification was carried out with a T1 Thermocycler (Whatman Biometra, Goettingen, Germany). Purified PCR-amplified fragments were sequenced (QIAquick PCR purification kit; QIAGEN) by automated DNA sequencing (ABI PRISM 377; Applied Biosystems, Foster City, CA), and sequences were assembled, aligned, and viewed with ClustalW (version 1.82 at the European Bioinformatics Institute, http://www.ebi.ac.uk) and BOXSHADE (version 3.21 in Swiss EMBnet, http://www.ch.embnet.org). Regression analysis was carried out with SigmaPlot Windows version 7.0 (SPSS Inc., Chicago, IL).
Nucleotide sequence accession numbers. GenBank accession numbers for the HCMV isolates are as follows: SM301, AY729049; SM302, AY729050; SM303, AY659927; SM304, AY659928; and SM305, AY727868.
RESULTS
TS-FACS analysis of the effect of MOI on HCMV infection and GCV susceptibility. To assess the change of expression of the immediate-early gene as a function of the concentration of infecting virus, AD169 was inoculated at various MOIs (0, 0.03, 0.1, 0.3, 1, and 3). As shown in Fig. 2a, as MOI increased, the number of cells expressing the HCMV gene increased. To see whether inhibition of the expression of HCMV genes in response to GCV can be detected by this method, the GCV-sensitive HCMV strain AD169 was infected and exposed to various concentrations of GCV, and TS-FACS analysis was performed. Albeit the IE gene expression inhibited by GCV was not that of the initially infected cells but of secondary infected cells 2 to 3 days after the inoculum was mixed with the uninfected cells, HCMV IE-positive cells decreased as a function of GCV concentration, and as little as 1 to 3 μM GCV reduced the number of cells expressing IE by over 50% (Fig. 2b).
Determination of the IC50 of GCV by PRA and TS-FACS analysis. HFF were infected with the GCV-sensitive strain AD169 or clinical isolates from patients and treated with various concentrations of GCV, the pattern of expression of HCMV genes was assessed by TS-FACS analysis, and GCV IC50 values were obtained. These were then compared with IC50 values from PRA. As shown in Fig. 3, concentrations of GCV below 3 μM reduced IE-positive cells by more than 50% in AD169 (Fig. 3a) as well as in clinical isolates SM301 (Fig. 3b) and SM303 (Fig. 3d). In SM302 (Fig. 3c), IC50 was approximately 16 μM, and in SM305 (Fig. 3f), GCV up to 30 μM failed to reduce positive cells by more than 50% (Fig. 3). Table 1 shows that the IC50 values for AD169 obtained by PRA and FACS analysis were very similar. For isolate SM304 (Fig. 3e), the IC50 obtained by TS-FACS was 4.92 μM and that by PRA was 2.73 μM, slightly higher than the IC50 for SM301 or HCMV SM303 but still within the sensitive range. In contrast, the IC50 obtained by PRA for HCMV SM302 was 9.51 μM. This differed somewhat from the 17.99 μM obtained by TS-FACS, and in HCMV SM305, the IC50 obtained by PRA was 44.5 μM and by TS-FACS it was over 30 μM. Regression analysis showed that the results of PRA and TS-FACS analysis were comparable statistically. r2 was 0.729 (P < 0.0001), and the regression equation was y = 0.89x + 0.62 (Fig. 4).
DNA sequences. We sequenced partial genes for mutations including the GCV resistance-related mutation M460V and detected a GCV resistance mutation only in UL97 of HCMV SM305 (Fig. 5).
DISCUSSION
PRA is the gold standard method for assessing resistance to antiviral agents. It tests directly whether the formation of viral plaques is affected by antiviral agents and thus can be considered more definitive than genotypic tests. However, for clinical strains and other slow-growing strains, especially viruses with long life cycles such as HCMV, PRA requires a minimum of 2 weeks to over a month. To reduce this time, Landry et al. developed a PRA method based on the plaque-forming cell concept that measures approximate numbers of plaques without titration, but this method also requires at least 2 weeks to obtain countable plaques (17). To shorten the time required for plaque formation, Prix et al. cultivated mixtures of HCMV-infected and uninfected cells in 96-well plates and used fluorescent antibody against immediate-early genes instead of conventional dyes to detect foci of infection (23); they defined a colony formed by more than 10 cells expressing IE genes as one plaque. A drawback of that method, however, is that it has not yet been automated. Though there have been several reports of GCV susceptibility tests using flow cytometry that can be automated (16, 18, 20), they were performed on cell-free virus and required a lot of time to obtain cell-free virus stocks. McSharry et al. used infected cells rather than cell-free virus to overcome this problem, but they commented that measuring the input of virus-infected cells was the major challenge if virus titer was not determined (19).
In our experiment, to reduce the time required for plaque formation, we used HCMV-infected cells as inoculum and mixed them with uninfected HFF. In addition, to decrease fluctuations due to variation in the ratio of HCMV-infected to uninfected cells, we established the required input volume of HCMV-infected cells for TS-FACS analysis from various experimental data and reference 23. We also used a fluorescent antibody against the HCMV IE gene product. The time required to obtain IC50 values by TS-FACS was 2 to 3 weeks, apart from the time for tube culture, while PRA requires at least twice as long. The TS-FACS analysis itself took only about a week. Also, since the data collection step of TS-FACS analysis is automated, if investigators are properly trained in its use the analysis is objective and hence more reproducible.
For isolate SM305, GCV IC50 values from PRA and TS-FACS analysis were high and similar (Table 1). In addition, in the genotypic test, although no known resistance mutation was detected in UL97, isolate SM302 was resistant to GCV by both methods. Therefore, phenotypic tests like TS-FACS analysis are needed to suggest the possibility of novel mutations and mutations in the UL54 DNA polymerase gene. For the sensitive isolate SM304, the IC50 from the PRA method was somewhat lower than that from the TS-FACS analysis.
Overall the results obtained by TS-FACS analysis were not substantially different from those obtained by PRA. There was an excellent statistical correlation between TS-FACS analysis and PRA by independent sample t test (P < 0.05) and multiple regression plot analysis (r2 = 0.729, P < 0.0001). The reason for the excellent r2 value may be that we defined the input of HCMV-infected cells in the mixtures with uninfected cells.
Clinical HCMV isolates are not like laboratory strains that form plaques readily. In assessing the resistance of HCMV to GCV or other antiviral agents in the clinic, fast and accurate methods are required. By performance of TS-FACS analysis as a phenotypic assay in parallel with RFLP and sequencing as genotypic assays, GCV resistance mutations that have been already identified as well as novel mutations can be identified rapidly, simply, and accurately (Fig. 6).
In summary, we developed TS-FACS analysis to overcome the shortcomings of PRA, and the results from the two methods were similar. TS-FACS analysis takes less time and can be automated and may well be a useful alternative to PRA in the clinic. Further study of GCV IC50 values obtained by TS-FACS analysis of more clinical isolates are required to define the cutoff value for sensitivity versus resistance and other clinically applicable criteria.
ACKNOWLEDGMENTS
This work was supported by a Korea Research Foundation grant (2003-005-E00010).
REFERENCES
Abraham, B., S. Lastere, J. Reynes, F. Bibollet-Ruche, N. Vidal, and M. Segondy. 1999. Ganciclovir resistance and UL97 gene mutations in cytomegalovirus blood isolates from patients with AIDS treated with ganciclovir. J. Clin. Virol. 13:141-148.
Alain, S., P. Honderlick, D. Grenet, M. Stern, C. Vadam, M. J. Sanson-Le Pors, and M. C. Mazeron. 1997. Failure of ganciclovir treatment associated with selection of a ganciclovir-resistant cytomegalovirus strain in a lung transplant recipient. Transplantation 63:1533-1536.
Baldanti, F., and G. Gerna. 2003. Human cytomegalovirus resistance to antiviral drugs: diagnosis, monitoring and clinical impact. J. Antimicrob. Chemother. 52:324-330.
Baldanti, F., L. Simoncini, A. Sarasini, M. Zavattoni, P. Grossi, M. G. Revello, and G. Gerna. 1998. Ganciclovir resistance as a result of oral ganciclovir in a heart transplant recipient with multiple human cytomegalovirus strains in blood. Transplantation 66:324-329.
Baldanti, F., M. R. Underwood, C. L. Talarico, L. Simoncini, A. Sarasini, K. K. Biron, and G. Gerna. 1998. The Cys607Tyr change in the UL97 phosphotransferase confers ganciclovir resistance to two human cytomegalovirus strains recovered from two immunocompromised patients. Antimicrob. Agents Chemother. 42:444-446.
Chou, S. 1999. Antiviral drug resistance in human cytomegalovirus. Transl. Infect. Dis. 1:105-114.
Chou, S., A. Erice, M. C. Jordan, G. M. Vercellotti, K. R. Michels, C. L. Talarico, S. C. Stanat, and K. K. Biron. 1995. Analysis of the UL97 phosphotransferase coding sequence in clinical cytomegalovirus isolates and identification of mutations conferring ganciclovir resistance. J. Infect. Dis. 171:576-583.
Chou, S., G. Marousek, S. Guentzel, S. E. Follansbee, M. E. Poscher, J. P. Lalezari, R. C. Miner, and W. L. Drew. 1997. Evolution of mutations conferring multidrug resistance during prophylaxis and therapy for cytomegalovirus disease. J. Infect. Dis. 176:786-789.
Chou, S., and C. L. Meichsner. 2000. A nine-codon deletion mutation in the cytomegalovirus UL97 phosphotransferase gene confers resistance to ganciclovir. Antimicrob. Agents Chemother. 44:183-185.
Cihlar, T., M. D. Fuller, and J. M. Cherrington. 1998. Characterization of drug resistance-associated mutations in the human cytomegalovirus DNA polymerase gene by using recombinant mutant viruses generated from overlapping DNA fragments. J. Virol. 72:5927-5936.
De Clercq, E. 2004. Antiviral drugs in current clinical use. J. Clin. Virol. 30:115-133.
Drew, W. L., R. Miner, and E. Saleh. 1993. Antiviral susceptibility testing of cytomegalovirus: criteria for detecting resistance to antivirals. Clin. Diagn. Virol. 1:179-185.
Erice, A. 1999. Resistance of human cytomegalovirus to antiviral drugs. Clin. Microbiol. Rev. 12:286-297.
Jabs, D. A., C. Enger, J. P. Dunn, and M. Forman. 1998. Cytomegalovirus retinitis and viral resistance: ganciclovir resistance. J. Infect. Dis. 177:770-773.
Jabs, D. A., C. Enger, M. Forman, and J. P. Dunn for The Cytomegalovirus Retinitis and Viral Resistance Study Group. 1998. Incidence of foscarnet resistance and cidofovir resistance in patients treated for cytomegalovirus retinitis. Antimicrob. Agents Chemother. 42:2240-2244.
Kesson, A. M., F. Zeng, A. L. Cunningham, and W. D. Rawlinson. 1998. The use of flow cytometry to detect antiviral resistance in human cytomegalovirus. J. Virol. Methods 71:177-186.
Landry, M. L., S. Stanat, K. Biron, D. Brambilla, W. Britt, J. Jokela, S. Chou, W. L. Drew, A. Erice, B. Gilliam, N. Lurain, J. Manischewitz, R. Miner, M. Nokta, P. Reichelderfer, S. Spector, A. Weinberg, B. Yen-Lieberman, and C. Crumpacker. 2000. A standardized plaque reduction assay for determination of drug susceptibilities of cytomegalovirus clinical isolates. Antimicrob. Agents Chemother. 44:688-692.
Lipson, S. M., M. Soni, F. X. Biondo, D. H. Shepp, M. H. Kaplan, and T. Sun. 1997. Antiviral susceptibility testing-flow cytometric analysis (AST-FCA) for the detection of cytomegalovirus drug resistance. Diagn. Microbiol. Infect. Dis. 28:123-129.
McSharry, J. M., N. S. Lurain, G. L. Drusano, A. Landay, M. Nokta, M. O'Gorman, A. Weinberg, H. M. Shapiro, P. Reichelderfer, and C. Crumpacker. 1998. Rapid ganciclovir susceptibility assay using flow cytometry for human cytomegalovirus clinical isolates. Antimicrob. Agents Chemother. 42:2326-2331.
McSharry, J. M., N. S. Lurain, G. L. Drusano, A. Landay, J. Manischewitz, M. Nokta, M. O'Gorman, H. M. Shapiro, A. Weinberg, P. Reichelderfer, and C. Crumpacker. 1998. Flow cytometric determination of ganciclovir susceptibilities of human cytomegalovirus clinical isolates. J. Clin. Microbiol. 36:958-964.
Mendez, J. C., I. G. Sia, K. R. Tau, M. J. Espy, T. F. Smith, S. Chou, and C. V. Paya. 1999. Novel mutation in the CMV UL97 gene associated with resistance to ganciclovir therapy. Transplantation 67:755-757.
Prix, L., K. Hamprecht, B. Holzhuter, R. Handgretinger, T. Klingebiel, and G. Jahn. 1999. Comprehensive restriction analysis of the UL97 region allows early detection of ganciclovir-resistant human cytomegalovirus in an immunocompromised child. J. Infect. Dis. 180:491-495.
Prix, L., J. Maierl, G. Jahn, and K. Hamprecht. 1998. A simplified assay for screening of drug resistance of cell-associated cytomegalovirus strains. J. Clin. Virol. 11:29-37.
Smith, I. L., J. M. Cherrington, R. E. Jiles, M. D. Fuller, W. R. Freeman, and S. A. Spector. 1997. High-level resistance of cytomegalovirus to ganciclovir is associated with alterations in both the UL97 and DNA polymerase genes. J. Infect. Dis. 176:69-77.
Wentworth, B. B., and L. French. 1970. Plaque assay of cytomegalovirus strains of human origin. Proc. Soc. Exp. Biol. Med. 135:253-258.
Wolf, D. G., I. L. Smith, D. J. Lee, W. R. Freeman, M. Flores-Aguilar, and S. A. Spector. 1995. Mutations in human cytomegalovirus UL97 gene confer clinical resistance to ganciclovir and can be detected directly in patient plasma. J. Clin. Investig. 95:257-263.(Gyu-Cheol Lee, Dong-Gun L)
Water Quality Analysis Team I, International Water Analysis Center, Korea Water Resources Corporation, Daejeon 306-711, Korea
Department of Microbiology, Natural Sciences College, Chungbuk National University, Cheongju 361-763, Korea
ABSTRACT
There are two ways to assess the susceptibility of human cytomegalovirus (HCMV) to ganciclovir (GCV): one is a genotypic test that detects resistance-related mutations and the other is a phenotypic test that actually assesses susceptibility. The advantages of genotyping the UL97 gene are its rapidity and accuracy. However, to detect novel mutations or mutations affecting the UL54 DNA polymerase, a phenotypic test such as the plaque reduction assay (PRA) is also required. To avoid the shortcomings of PRA such as its time-consuming nature and labor-intensiveness, we developed a time-saving fluorescence-activated cell sorting (TS-FACS) technique. We obtained a GCV 50% inhibitory concentration (IC50) from five clinical isolates and an HCMV laboratory strain (AD169) and compared the results with those from the PRA. The laboratory strain and three clinical isolates were sensitive to GCV. Although there was a minor discrepancy in the case of one of the three isolates, the GCV IC50 values obtained by TS-FACS analysis correlated well with the results of the PRA. The remaining two isolates were resistant to GCV; one was GCV resistant due to the mutation M460V, and the GCV IC50 results obtained by TS-FACS analysis and by PRA were also comparable. The advantages of TS-FACS analysis are the shorter time required, the possibility of automation, and its comparability to PRA, considered the gold standard. Thus, TS-FACS analysis may be useful as an alternative to PRA in the clinic.
INTRODUCTION
Human cytomegalovirus (HCMV) is a betaherpesvirus that causes serious diseases in immunocompromised hosts, particularly transplantation recipients. Drugs used currently against HCMV include ganciclovir (GCV), foscarnet, and cidofovir, and of these, GCV is the initial drug of choice (11, 13). However, the advent of HCMV resistant to GCV has made control of HCMV difficult. Mutant HCMV strains strongly resistant to GCV are unable to phosphorylate GCV owing to mutations in key regions of the UL97 gene (domains VIII, VI, and IX) (1, 2, 5, 8, 9, 21, 26). Mutations in UL54, the viral DNA polymerase, also confer resistance to GCV (10, 15, 24).
There are two types of methods for assessing the susceptibility of HCMV to GCV: one is genotypic and detects the resistance-related mutant genes, and the other is phenotypic and assesses resistance directly (3, 13). The genotypic methods include DNA sequencing, detection of restriction fragment length polymorphisms (RFLP), and probe-specific and primer-specific hybridization. DNA sequencing is the reference method for detecting resistance-related GCV mutations, and RFLP has been widely used to detect GCV resistance due to mutations in the UL97 gene (4, 7). Although genotypic tests are fast, phenotypic tests are necessary to detect novel mutations and mutations in the UL54 DNA polymerase gene. Examples of phenotypic tests are the plaque reduction assay (PRA), in situ enzyme-linked immunosorbent assay, DNA reduction assay, and flow cytometry-based assay. The flow cytometric assay counts trypsinized fibroblasts expressing the HCMV antigen using a fluorochrome-labeled antibody and has the advantage that automation and quantitation are feasible (16, 18, 20).
PRA is the gold standard for evaluating the drug susceptibility of a virus. For HCMV, it consists of counting the number of viral plaques on fibroblasts. However, this conventional phenotypic assay is time-consuming and labor-intensive and requires highly skilled laboratory technicians (17, 25). This is because most clinical isolates examined grow slowly, so that it takes a long time to obtain the required virus titers (12). To overcome these shortcomings, Prix et al. and Landry et al. have reported a method without the titration step of classical PRA (17, 23).
We have modified and applied the method of mixed culture of cell-associated viruses to flow cytometric analysis and have compared this new time-saving fluorescence-activated cell sorting (TS-FACS) analysis with PRA. We have also assessed whether TS-FACS analysis could be used as a GCV susceptibility test for clinical HCMV isolates.
MATERIALS AND METHODS
Cells and virus. As the source of cells, human foreskin fibroblasts (HFF, ATCC CRL-2097; American Type Culture Collection, Manassas, VA) between passages 10 and 14 were used. Cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Cambrex Bio Science, Walkersville, MD) supplemented with 10% fetal bovine serum (FBS; U.S. Bio-Technologies Inc., Parkerford, PA) and penicillin (100 units/ml)-streptomycin (100 μg/ml) solution (JBI, Daegu, Korea), at 37°C, in a 5% CO2 incubator. HCMV strain AD169 was the source of virus, and the stock was obtained by plating fresh HFF inoculated with HCMV at a multiplicity of infection (MOI) of 0.01 to 0.03 in 100-mm cell culture plates (Nalge Nunc International, Rochester, NY); the infected cells were detached with a scraper when the cytopathic effect (CPE) had spread completely and collected in a new tube with culture medium. Cell debris was removed, the freeze-thaw cycle was repeated three times, and the suspension was centrifuged. The stock thus obtained was stored in 1 ml at –70°C, and the titer of the stock was 106 to 107 PFU/ml.
Clinical isolates. Three milliliters of blood was obtained from patients who were suspected of having a GCV-resistant HCMV infection and transferred to a 15-ml conical tube (Becton Dickinson Labware, Franklin Lakes, NJ). The blood was centrifuged at 800 x g for 15 min, the plasma layer was taken, 0.2 ml of this was used to infect fresh HFF prepared in advance, and fresh medium was substituted after 3 h. Plasma was collected separately, and to the remaining fraction containing blood cells was added phosphate-buffered saline (PBS; Sigma Diagnostics Inc., St. Louis, MO), adjusted to a final volume of 6 ml. This was loaded carefully, without disturbing the layers, into a 15-ml tube containing 3 ml Histopaque 1119 (Sigma Diagnostics Inc.). After centrifugation at 700 x g for 30 min, the white blood cell band formed above the Histopaque was transferred to a new 15-ml tube with a Pasteur pipette, washed with 10 ml PBS, and centrifuged at 800 x g for 10 min. The supernatant was removed, and the remaining cell pellet was washed with 5 ml PBS and centrifuged at 800 x g for 5 min. This was repeated twice, to yield the white blood cells. The cells were counted by the Trypan blue exclusion assay, and approximately 2 x 106 cells were inoculated into prepared HFF. The medium was changed every 4 to 5 days, and if CPE was detected, the cells were cultured with 2% FBS in DMEM until plaques formed. These plaques were then mixed with uninfected fresh HFF and cultured in T25 tissue culture flasks.
Plaque reduction assay. The classical plaque reduction assay was performed by culturing fresh HFF in 35-mm culture plates and infecting them with virus. The cells were washed with PBS and overlaid with the appropriate concentration of GCV (Roche Diagnostics, Basel, Switzerland) contained in semisolid medium consisting of DMEM plus 2% FBS, 0.25% SeaPlaque agarose (Cambrex Bio Science, Rockland, ME), 100 units/ml penicillin (JBI), 100 μg/ml streptomycin (JBI), and 250 ng/ml amphotericin B (JBI). Approximately 7 days later, a second overlay of the same semisolid medium containing various concentrations of GCV was added. When plaques had formed, the cells were fixed with 10% formalin in 0.85% saline solution. The fixed-cell monolayer was stained with 0.03% methylene blue, and the plaques were counted with a light microscope (BH-2; Olympus Corporation, Tokyo, Japan).
TS-FACS analysis. The overall process is illustrated in Fig. 1. When the CPE had spread over 50% of the T25 flask containing the mixed culture, the culture supernatant and infected cells were collected in a tube. Fresh HFF grown in 100-mm culture plates were suspended in 10 ml and divided into two tubes. HCMV-infected cells (0.2 or 1 ml) were mixed well with the uninfected cells and plated in six-well plates. Approximately 5 to 6 h later, when the cells had attached to the culture plates, GCV was added at concentrations of 0, 1, 3, 10, 30, or 100 μM. Three to four days later the cells were detached by trypsinization, and samples of 105 cells were added to 1-ml microcentrifuge tubes. The cells were washed once with PBS, fixed with 500 μl 0.2% paraformaldehyde (Sigma Diagnostics Inc.) at room temperature for 1 h, and exposed to 500 μl 0.2% Tween 20 (J. T. Baker, Phillipsburg, NJ) for 20 min for permeabilization. After being washed with 1 ml PBS, they were suspended in 100 μl PBS, and 1 drop of mouse anti-HCMV immediate-early (IE) protein fluorescein isothiocyanate-conjugated antibody (MAB 5090; Chemicon, Temecula, CA) was added according to the manufacturer's instructions, and incubation continued for 40 min in the dark at room temperature. The cells were then washed twice with 1 ml PBS and resuspended in 200 μl PBS, flow cytometric analysis was performed with a FACScalibur (BD Biosciences, San Jose, CA), and results were analyzed with Cell Quest (BD Biosciences). The 50% inhibitory concentration (IC50) was the median concentration that caused 50% inhibition of IE gene expression relative to that in the control HCMV-infected cells. All experiments were performed three times independently. GCV resistance was defined as cells with an IC50 exceeding 6 μM to 12 μM GCV (6, 7, 14).
DNA sequencing and data analysis. DNA was extracted from HCMV-infected cells using a QIAamp DNA minikit (QIAGEN, Hilden, Germany). Primers for the UL97 region (776 bp; nucleotides 142728 to 143503) were selected from HCMV AD169 (GenBank accession number NC_001347). The primers were 460F, 5'-GTTGGCCGACGCTATCAAAT-3', and 650R, 5'-GGTCCTCCTCGCAGATTATG-3' (22). After a hot-start step to activate i-star Taq (iNtRON Biotechnology, Sungnam, Korea), the samples underwent 40 cycles of denaturation at 94.5°C for 1 min, annealing at 55°C for 1 min, and extension at 72°C for 1 min, followed by a final extension at 72°C for 5 min. Amplification was carried out with a T1 Thermocycler (Whatman Biometra, Goettingen, Germany). Purified PCR-amplified fragments were sequenced (QIAquick PCR purification kit; QIAGEN) by automated DNA sequencing (ABI PRISM 377; Applied Biosystems, Foster City, CA), and sequences were assembled, aligned, and viewed with ClustalW (version 1.82 at the European Bioinformatics Institute, http://www.ebi.ac.uk) and BOXSHADE (version 3.21 in Swiss EMBnet, http://www.ch.embnet.org). Regression analysis was carried out with SigmaPlot Windows version 7.0 (SPSS Inc., Chicago, IL).
Nucleotide sequence accession numbers. GenBank accession numbers for the HCMV isolates are as follows: SM301, AY729049; SM302, AY729050; SM303, AY659927; SM304, AY659928; and SM305, AY727868.
RESULTS
TS-FACS analysis of the effect of MOI on HCMV infection and GCV susceptibility. To assess the change of expression of the immediate-early gene as a function of the concentration of infecting virus, AD169 was inoculated at various MOIs (0, 0.03, 0.1, 0.3, 1, and 3). As shown in Fig. 2a, as MOI increased, the number of cells expressing the HCMV gene increased. To see whether inhibition of the expression of HCMV genes in response to GCV can be detected by this method, the GCV-sensitive HCMV strain AD169 was infected and exposed to various concentrations of GCV, and TS-FACS analysis was performed. Albeit the IE gene expression inhibited by GCV was not that of the initially infected cells but of secondary infected cells 2 to 3 days after the inoculum was mixed with the uninfected cells, HCMV IE-positive cells decreased as a function of GCV concentration, and as little as 1 to 3 μM GCV reduced the number of cells expressing IE by over 50% (Fig. 2b).
Determination of the IC50 of GCV by PRA and TS-FACS analysis. HFF were infected with the GCV-sensitive strain AD169 or clinical isolates from patients and treated with various concentrations of GCV, the pattern of expression of HCMV genes was assessed by TS-FACS analysis, and GCV IC50 values were obtained. These were then compared with IC50 values from PRA. As shown in Fig. 3, concentrations of GCV below 3 μM reduced IE-positive cells by more than 50% in AD169 (Fig. 3a) as well as in clinical isolates SM301 (Fig. 3b) and SM303 (Fig. 3d). In SM302 (Fig. 3c), IC50 was approximately 16 μM, and in SM305 (Fig. 3f), GCV up to 30 μM failed to reduce positive cells by more than 50% (Fig. 3). Table 1 shows that the IC50 values for AD169 obtained by PRA and FACS analysis were very similar. For isolate SM304 (Fig. 3e), the IC50 obtained by TS-FACS was 4.92 μM and that by PRA was 2.73 μM, slightly higher than the IC50 for SM301 or HCMV SM303 but still within the sensitive range. In contrast, the IC50 obtained by PRA for HCMV SM302 was 9.51 μM. This differed somewhat from the 17.99 μM obtained by TS-FACS, and in HCMV SM305, the IC50 obtained by PRA was 44.5 μM and by TS-FACS it was over 30 μM. Regression analysis showed that the results of PRA and TS-FACS analysis were comparable statistically. r2 was 0.729 (P < 0.0001), and the regression equation was y = 0.89x + 0.62 (Fig. 4).
DNA sequences. We sequenced partial genes for mutations including the GCV resistance-related mutation M460V and detected a GCV resistance mutation only in UL97 of HCMV SM305 (Fig. 5).
DISCUSSION
PRA is the gold standard method for assessing resistance to antiviral agents. It tests directly whether the formation of viral plaques is affected by antiviral agents and thus can be considered more definitive than genotypic tests. However, for clinical strains and other slow-growing strains, especially viruses with long life cycles such as HCMV, PRA requires a minimum of 2 weeks to over a month. To reduce this time, Landry et al. developed a PRA method based on the plaque-forming cell concept that measures approximate numbers of plaques without titration, but this method also requires at least 2 weeks to obtain countable plaques (17). To shorten the time required for plaque formation, Prix et al. cultivated mixtures of HCMV-infected and uninfected cells in 96-well plates and used fluorescent antibody against immediate-early genes instead of conventional dyes to detect foci of infection (23); they defined a colony formed by more than 10 cells expressing IE genes as one plaque. A drawback of that method, however, is that it has not yet been automated. Though there have been several reports of GCV susceptibility tests using flow cytometry that can be automated (16, 18, 20), they were performed on cell-free virus and required a lot of time to obtain cell-free virus stocks. McSharry et al. used infected cells rather than cell-free virus to overcome this problem, but they commented that measuring the input of virus-infected cells was the major challenge if virus titer was not determined (19).
In our experiment, to reduce the time required for plaque formation, we used HCMV-infected cells as inoculum and mixed them with uninfected HFF. In addition, to decrease fluctuations due to variation in the ratio of HCMV-infected to uninfected cells, we established the required input volume of HCMV-infected cells for TS-FACS analysis from various experimental data and reference 23. We also used a fluorescent antibody against the HCMV IE gene product. The time required to obtain IC50 values by TS-FACS was 2 to 3 weeks, apart from the time for tube culture, while PRA requires at least twice as long. The TS-FACS analysis itself took only about a week. Also, since the data collection step of TS-FACS analysis is automated, if investigators are properly trained in its use the analysis is objective and hence more reproducible.
For isolate SM305, GCV IC50 values from PRA and TS-FACS analysis were high and similar (Table 1). In addition, in the genotypic test, although no known resistance mutation was detected in UL97, isolate SM302 was resistant to GCV by both methods. Therefore, phenotypic tests like TS-FACS analysis are needed to suggest the possibility of novel mutations and mutations in the UL54 DNA polymerase gene. For the sensitive isolate SM304, the IC50 from the PRA method was somewhat lower than that from the TS-FACS analysis.
Overall the results obtained by TS-FACS analysis were not substantially different from those obtained by PRA. There was an excellent statistical correlation between TS-FACS analysis and PRA by independent sample t test (P < 0.05) and multiple regression plot analysis (r2 = 0.729, P < 0.0001). The reason for the excellent r2 value may be that we defined the input of HCMV-infected cells in the mixtures with uninfected cells.
Clinical HCMV isolates are not like laboratory strains that form plaques readily. In assessing the resistance of HCMV to GCV or other antiviral agents in the clinic, fast and accurate methods are required. By performance of TS-FACS analysis as a phenotypic assay in parallel with RFLP and sequencing as genotypic assays, GCV resistance mutations that have been already identified as well as novel mutations can be identified rapidly, simply, and accurately (Fig. 6).
In summary, we developed TS-FACS analysis to overcome the shortcomings of PRA, and the results from the two methods were similar. TS-FACS analysis takes less time and can be automated and may well be a useful alternative to PRA in the clinic. Further study of GCV IC50 values obtained by TS-FACS analysis of more clinical isolates are required to define the cutoff value for sensitivity versus resistance and other clinically applicable criteria.
ACKNOWLEDGMENTS
This work was supported by a Korea Research Foundation grant (2003-005-E00010).
REFERENCES
Abraham, B., S. Lastere, J. Reynes, F. Bibollet-Ruche, N. Vidal, and M. Segondy. 1999. Ganciclovir resistance and UL97 gene mutations in cytomegalovirus blood isolates from patients with AIDS treated with ganciclovir. J. Clin. Virol. 13:141-148.
Alain, S., P. Honderlick, D. Grenet, M. Stern, C. Vadam, M. J. Sanson-Le Pors, and M. C. Mazeron. 1997. Failure of ganciclovir treatment associated with selection of a ganciclovir-resistant cytomegalovirus strain in a lung transplant recipient. Transplantation 63:1533-1536.
Baldanti, F., and G. Gerna. 2003. Human cytomegalovirus resistance to antiviral drugs: diagnosis, monitoring and clinical impact. J. Antimicrob. Chemother. 52:324-330.
Baldanti, F., L. Simoncini, A. Sarasini, M. Zavattoni, P. Grossi, M. G. Revello, and G. Gerna. 1998. Ganciclovir resistance as a result of oral ganciclovir in a heart transplant recipient with multiple human cytomegalovirus strains in blood. Transplantation 66:324-329.
Baldanti, F., M. R. Underwood, C. L. Talarico, L. Simoncini, A. Sarasini, K. K. Biron, and G. Gerna. 1998. The Cys607Tyr change in the UL97 phosphotransferase confers ganciclovir resistance to two human cytomegalovirus strains recovered from two immunocompromised patients. Antimicrob. Agents Chemother. 42:444-446.
Chou, S. 1999. Antiviral drug resistance in human cytomegalovirus. Transl. Infect. Dis. 1:105-114.
Chou, S., A. Erice, M. C. Jordan, G. M. Vercellotti, K. R. Michels, C. L. Talarico, S. C. Stanat, and K. K. Biron. 1995. Analysis of the UL97 phosphotransferase coding sequence in clinical cytomegalovirus isolates and identification of mutations conferring ganciclovir resistance. J. Infect. Dis. 171:576-583.
Chou, S., G. Marousek, S. Guentzel, S. E. Follansbee, M. E. Poscher, J. P. Lalezari, R. C. Miner, and W. L. Drew. 1997. Evolution of mutations conferring multidrug resistance during prophylaxis and therapy for cytomegalovirus disease. J. Infect. Dis. 176:786-789.
Chou, S., and C. L. Meichsner. 2000. A nine-codon deletion mutation in the cytomegalovirus UL97 phosphotransferase gene confers resistance to ganciclovir. Antimicrob. Agents Chemother. 44:183-185.
Cihlar, T., M. D. Fuller, and J. M. Cherrington. 1998. Characterization of drug resistance-associated mutations in the human cytomegalovirus DNA polymerase gene by using recombinant mutant viruses generated from overlapping DNA fragments. J. Virol. 72:5927-5936.
De Clercq, E. 2004. Antiviral drugs in current clinical use. J. Clin. Virol. 30:115-133.
Drew, W. L., R. Miner, and E. Saleh. 1993. Antiviral susceptibility testing of cytomegalovirus: criteria for detecting resistance to antivirals. Clin. Diagn. Virol. 1:179-185.
Erice, A. 1999. Resistance of human cytomegalovirus to antiviral drugs. Clin. Microbiol. Rev. 12:286-297.
Jabs, D. A., C. Enger, J. P. Dunn, and M. Forman. 1998. Cytomegalovirus retinitis and viral resistance: ganciclovir resistance. J. Infect. Dis. 177:770-773.
Jabs, D. A., C. Enger, M. Forman, and J. P. Dunn for The Cytomegalovirus Retinitis and Viral Resistance Study Group. 1998. Incidence of foscarnet resistance and cidofovir resistance in patients treated for cytomegalovirus retinitis. Antimicrob. Agents Chemother. 42:2240-2244.
Kesson, A. M., F. Zeng, A. L. Cunningham, and W. D. Rawlinson. 1998. The use of flow cytometry to detect antiviral resistance in human cytomegalovirus. J. Virol. Methods 71:177-186.
Landry, M. L., S. Stanat, K. Biron, D. Brambilla, W. Britt, J. Jokela, S. Chou, W. L. Drew, A. Erice, B. Gilliam, N. Lurain, J. Manischewitz, R. Miner, M. Nokta, P. Reichelderfer, S. Spector, A. Weinberg, B. Yen-Lieberman, and C. Crumpacker. 2000. A standardized plaque reduction assay for determination of drug susceptibilities of cytomegalovirus clinical isolates. Antimicrob. Agents Chemother. 44:688-692.
Lipson, S. M., M. Soni, F. X. Biondo, D. H. Shepp, M. H. Kaplan, and T. Sun. 1997. Antiviral susceptibility testing-flow cytometric analysis (AST-FCA) for the detection of cytomegalovirus drug resistance. Diagn. Microbiol. Infect. Dis. 28:123-129.
McSharry, J. M., N. S. Lurain, G. L. Drusano, A. Landay, M. Nokta, M. O'Gorman, A. Weinberg, H. M. Shapiro, P. Reichelderfer, and C. Crumpacker. 1998. Rapid ganciclovir susceptibility assay using flow cytometry for human cytomegalovirus clinical isolates. Antimicrob. Agents Chemother. 42:2326-2331.
McSharry, J. M., N. S. Lurain, G. L. Drusano, A. Landay, J. Manischewitz, M. Nokta, M. O'Gorman, H. M. Shapiro, A. Weinberg, P. Reichelderfer, and C. Crumpacker. 1998. Flow cytometric determination of ganciclovir susceptibilities of human cytomegalovirus clinical isolates. J. Clin. Microbiol. 36:958-964.
Mendez, J. C., I. G. Sia, K. R. Tau, M. J. Espy, T. F. Smith, S. Chou, and C. V. Paya. 1999. Novel mutation in the CMV UL97 gene associated with resistance to ganciclovir therapy. Transplantation 67:755-757.
Prix, L., K. Hamprecht, B. Holzhuter, R. Handgretinger, T. Klingebiel, and G. Jahn. 1999. Comprehensive restriction analysis of the UL97 region allows early detection of ganciclovir-resistant human cytomegalovirus in an immunocompromised child. J. Infect. Dis. 180:491-495.
Prix, L., J. Maierl, G. Jahn, and K. Hamprecht. 1998. A simplified assay for screening of drug resistance of cell-associated cytomegalovirus strains. J. Clin. Virol. 11:29-37.
Smith, I. L., J. M. Cherrington, R. E. Jiles, M. D. Fuller, W. R. Freeman, and S. A. Spector. 1997. High-level resistance of cytomegalovirus to ganciclovir is associated with alterations in both the UL97 and DNA polymerase genes. J. Infect. Dis. 176:69-77.
Wentworth, B. B., and L. French. 1970. Plaque assay of cytomegalovirus strains of human origin. Proc. Soc. Exp. Biol. Med. 135:253-258.
Wolf, D. G., I. L. Smith, D. J. Lee, W. R. Freeman, M. Flores-Aguilar, and S. A. Spector. 1995. Mutations in human cytomegalovirus UL97 gene confer clinical resistance to ganciclovir and can be detected directly in patient plasma. J. Clin. Investig. 95:257-263.(Gyu-Cheol Lee, Dong-Gun L)