Rapid detection by reverse hybridization of mutations in the UL97 gene of human cytomegalovirus conferring resistance to ganciclovir

Journal of Clinical Virology 13 (1999) 53 – 59

Short Communication

Rapid detection by reverse hybridization of mutations in the UL97 gene of human cytomegalovirus conferring resistance to ganciclovir Lin Zhou a, T.C. Harder b,*, U. Ullmann b, P. Rautenberg b b

a Department of Microbiology, Zhejiang Medical Uni6ersity, Hangzhou, Zhejiang, People’s Republic of China Institute of Medical Microbiology and Virology, Christian-Albrechts-Uni6ersity, Brunswiker Str. 4, D-24105 Kiel, Germany

Received 13 November 1998; accepted 12 December 1998

Abstract Background of study: Diseases due to human cytomegalovirus (HCMV) infection constitute a major threat in marrow and solid organ transplant recipients. Ganciclovir (GCV) is widely used in prophylaxis and pre-emptive therapy of active HCMV infection. Resistance to ganciclovir (GCV) may arise at variable frequency under GCV therapy and is conferred by mutations (i) in the UL97 gene (codons 460, 520, and 591 – 607) encoding a phosphotransferase which is essential for monophosphorylation of GCV and, to a lesser extent, (ii) in the UL54 gene coding for the DNA polymerase of HCMV. Objective: The purpose was to develop a rapid assay to screen for emerging GCV resistance mutations in the UL97 gene of HCMV whereby avoiding virus isolation and nucleotide sequencing procedures. Study design: A nested PCR (nPCR) amplifying UL97 codons 450 – 672 was developed. Nested amplicons were subsequently sequenced directly. Oligonucleotides for use in a reverse hybridization assay were designed to detect relevant non-synonymous mutations at codons UL97 460, 520, 603 and 607. Strain AD169 served as a wild-type control. Results: UL97-specific nPCR amplicons were obtained from 18 EDTA blood samples of ten transplant recipients receiving GCV for more than 30 days. In three consecutive samples from a single patient a GCV resistance mutation at codon 603 (C “W) was detected. In addition, two out of four cell culture-adapted HCMV isolates known to exhibit GCV resistance in vitro revealed mutations at codons 460 (M “ V) and 607 (C “ Y), respectively. By reverse hybridization a discrimination of single nucleotide changes at codons 460, 520, 603 and 607 was possible whereby matching exactly the results of the nucleotide sequence analysis for all 23 amplicons examined.

* Corresponding author. Tel.: +49-431-597-3331; fax: + 49-431-597-3296. E-mail address: [email protected] (T.C. Harder) 1386-6532/99/$ - see front matter © 1999 Elsevier Science B.V. All rights reserved. PII: S 1 3 8 6 - 6 5 3 2 ( 9 9 ) 0 0 0 0 4 - 9

54

L. Zhou et al. / Journal of Clinical Virology 13 (1999) 53–59

Conclusions: Reverse hybridization appeared to be a rapid and convenient alternative to nucleotide sequencing when screening the UL97 gene of HCMV for selected markers of GCV resistance. © 1999 Elsevier Science B.V. All rights reserved. Keywords: Human cytomegalovirus; Transplantation medicine; Ganciclovir resistance; Reverse hybridization

Clinical disorders associated with infections by the human cytomegalovirus (HCMV), a member of the beta-Herpes6irinae subfamily, are a major cause of morbidity and mortality in immuno-compromised individuals including recipients of bone marrow and solid-organ transplants (Griffiths and Emery, 1997). Ganciclovir (9-[1,3-dihydroxy-propoxymethyl]guanine, GCV) is the accepted firstline option for antiviral management (prophylaxis, pre-emptive therapy and treatment) of patients at risk of HCMV-associated disease (Noble and Faulds, 1998). Under pressure of GCV HCMV may evolve mutants with reduced sensitivity to this drug. The frequency and time frames at which resistant clones arise appear to vary between 5 and 30% of cases treated for 10 – 100 days (Chou et al., 1995a; Jabs and Jacobson, 1997; Erice et al., 1998). Resistance of HCMV to GCV is acquired by two separate mechanisms: (i) mutations in the UL97 phosphotransferase gene and/or (ii) mutations in the UL54 DNA polymerase gene (Sullivan et al., 1993; Lurain et al., 1994). The product of UL97 is essentially required for the monophosphorylation of GCV in infected cells (Littler et al., 1992; Sullivan et al., 1992; Michel et al., 1998). About 90% of GCV-resistant clinical HCMV isolates were shown to carry non-synonymous mutations at one or more of codons located at positions 460, 520, or 591 through 607 of UL97 (Chou et al., 1995a,b; Hanson et al., 1995; Baldanti et al., 1998). Marker transfer experiments using the fully GCV-sensitive AD169 strain of HCMV as background revealed an association of these mutations with an up to 5-fold increase of the ID50 (Boivin et al., 1996). Upon further phosphorylation by cellular kinases GCV-TP acts as a chain terminator competing for GTP at the viral DNA polymerase (UL54). Several loci in the conserved domaines of UL54 have been described to

be mutated in GCV-resistant HCMV isolates as well (Chou et al., 1997; Smith et al., 1997). Phenotypically, GCV resistance is verified by measuring the ID50 of cell culture-adapted isolates, e.g. in plaque reduction assays (Lurain et al., 1992). This requires advanced cell culture facilities and the availability of infectious virus. Genotypic resistance assays have mainly concentrated on the UL97 gene and are based on PCR amplication followed by verification of specific mutations by nucleotide sequence analysis, restriction fragment length polymorphism (RFLP), or by a point mutation hybridization assay (Hanson et al., 1995; Spector et al., 1995; Wolf et al., 1995; Bowen et al., 1997). Among these, nucleotide sequence analysis proved to be the best standardized and most accurate, but also most cost-intensive, method. RFLP analysis potentially allows a rapid identification of resistance-conferring mutations in clinical samples, but due to the complexity of the fragment patterns some experience is required for the interpretation of results. Reverse hybridization, an oligonucleotide membrane-based method, potentially allows the clearcut analysis of single nucleotide exchanges even in heterogeneous amplicon populations. This technique has been used extensively e.g. for genotyping of hepatitis C viruses and for detection of resistance-conferring mutations in the reverse transcriptase locus of human immunodeficiency virus-1 (Stuyver et al., 1996, 1997). In this study we attempted to develop oligos capable of screening, by reverse hybridization, for emerging GCV resistance mutations in the UL97 gene of HCMV derived from transplant recipients whereby avoiding both cultural virus isolation and nucleotide sequence analysis. In our study, ten patients from cohorts of heart/lung transplant recipients at Kiel University Clinics were included. All of them were diagnosed with active HCMV infection by monitoring pp65

L. Zhou et al. / Journal of Clinical Virology 13 (1999) 53–59

55

Table 1 Amplification by nested PCR of a UL97 gene fragment of HCMV encompassing codons 450–672a Primer

Sequence (5%–3%)

Orientation

Amplicon size (bp)

UL97-1up UL97-1low UL97-2up UL97-2low

(1247)b CAGACATGTTTCATCACGAC-3% (2175) TACCTTCTCTGTTGYCYTTC-3 (1347) GTGTCGTGTATGCCAYTTYG-3% (2016) CAGACGCTCCACGTTCTTTC-3%

Sense Antisense Sense Antisense

929 670

A 50 ml PCR reaction contained: 5 ml of template DNA, 1.5 mM MgC12, 200 mM of each dNTP, 100 pmol of primers and 1.5 U of AmpliTaq Gold™ (Perkin Elmer) in the appropriate buffer. Thermocycling on a Perkin Elmer 9600 series cycler, following an initial 10 min incubation at 94°C required for the activation of the polymerase, consisted of 35 cycles of: (i) UL97 1st PCR: denaturation at 94°C for 30 s, annealing at 58.5°C for 35 s, extension at 72°C for 60 s; and (ii) UL97 2nd PCR: 94°C for 30 s, 59°C for 30 s, and 72°C for 40 s. A final extension step at 72°C for 5 min was run following each PCR. b Position of the most 5% nucleotide. Position 1 is at the first nucleotide of the UL97 translation initiation codon. a

antigenaemia (mab C10, Cll, Biotest) and HCMV DNA load by quantitative hybridization (Murex) in EDTA blood samples. All patients had received GCV intravenously at doses of 5 mg kg − 1 for at least 30 days when obtaining samples for this study. Total DNA was extracted from 200 ml of EDTA-anticoagulated blood according to the manufacturer’s instructions (Qiagen, blood kit). By nested polymerase chain reaction (nPCR) a 670-bp fragment of the UL97 gene spanning codons 450–672 was amplified. Primer sequences and amplification conditions are specified in Table 1. The nPCR products were detected by agarose gel electrophoresis. Despite an increased risk of contamination, the nested PCR format was chosen for the amplification of the UL97 gene fragment in order to generate template amounts sufficient for direct sequencing. Amplicon sequencing would enable the detection of mixed amplificate populations and avoids time- and cost-intensive plasmid cloning. Specifically positive nPCR products were purified (QIAquick PCR Purification Kit, Qiagen) and subjected to cycle sequencing using the ABI PRISM™ BigDye™ cycle sequencing technology (PE Applied Biosystem). Sense and antisense primers used for nPCR were also employed in cycle sequencing. Products were analysed on a 310 Genetic Analyzer (PE Applied Biosystems). The HCMV strain AD169 was used as a wild-type control throughout this study.

Oligonucleotide probes for use in reverse hybridization were selected for codons UL97 460, 603, 607 and 520 with the aid of the MacMollyTetra software (SoftGene; Scho¨neberg et al., 1994). Properties of these probes are presented in Table 2. The oligonucleotides were poly(dT)-tailed using terminal transferase (Promega). Briefly, 200 pmol of primer were incubated for 1 h at 37°C in 20 ml buffer containing 3.2 mM dTTP, 25 mM Tris– HCl (pH7.5), 0.1 M sodium cacodylate, 1 mM CoCl2, 0.1 M dithiothreitol, and 48 U of terminal deoxynucleotidyl transferase. The reaction was stopped by adding 2.0 ml of 0.5 M EDTA (pH 8.0) and further diluted with 20 × SSC until a final concentration of 6× SSC and 4.0 pmol oligonucleotide per ml was reached. In order to achieve optimal signal/noise ratios, up to two pmol of each probe was dotted onto nylon membrane strips (Hybond + , Amersham) and immobilized by UV cross linking at 240 mJoules (UV Stratalinker, Stratagene). Nested PCR for the UL97 amplicon, this time using a mixture of dNTPs containing digoxigenin-11-dUTP (DIG, Boehringer Mannheim), was performed on all 23 samples previously sequenced. Ten microliters of the nested, DlG-labeled PCR amplicon were heatdenatured and diluted in 2 ml hybridization buffer (3×SSC in ECL blocking buffer, Amersham). Hybridization was carried out in a shaking water bath at 42°C for 2 h. The strips were rinsed twice at room temperature for 5 min with 2 ml washing buffer I (3 ×SSC, 0.1% SDS) followed by a stringent wash at 45°C for 30 min with washing buffer

L. Zhou et al. / Journal of Clinical Virology 13 (1999) 53–59

56

Table 2 Properties of probes used in reverse hybridization assaya Probes

Probe Probe Probe Probe Probe Probe Probe Probe Probe

Sequence (5%–3%)

1 2 3 4 5 7 6 8 9

CGTTCATGGGTGTAAT-3% GCACGTTCACGGGTGTA-3% CACGTTCAATGGGTGT-3% AGCAGGGTGGTAACATT-3% TCGGAGCAGTGCGTGA-3% TCGGAACAGTGCGTGA-3% TCGGACCAGTGCGTGA-3% TGAGCAGACAGGCGTC-3% ATGAGCAGATAGGCGTCG-3%

Specificity

460 wt (Met) M460V M4601 520 wt (His) 603 wt1 (Cys) 603 wt2 (Cys) C603W 607 wt (Cys) C607Y

Melting temperature (Tm) Match

Mismatch

57.2 65.9 59.8 61.1 64.9 62.4 64.9 64.9 66.8

51.0 60.0 53.6 55.2 58.7 56.2 58.7 58.7 61.3

a All oligos are presented in anti-sense orientation. The respective codon is shown in bold letters. Melting temperatures were calculated according to the %GC method against the homologous (perfect match) and the mutant (one mismatch) sequences.

II (1 ×SSC, 0.1% SDS), and two brief washing steps (buffer II) at room temperature. The strips were then placed in conjugate diluent (0.1 M NaCl, 0.1 M Tris–HCl pH 7.5) and further incubated with anti-digoxigenin alkaline phosphatase conjugate (approximately 0.4 U per strip, Boehringer Mannheim) for 30 min at room temperature. Subsequent washing was carried out with substrate diluent (0.1 M NaCl, 0.1 M Tris – HCl, 5 mM MgCl2, pH 9.5) three times each for 3 min. Colour development was achieved by incubating with BCIP/NBT (Pierce) at room temperature for 30 min. The colour development was stopped by rinsing with distilled water for at least 5 min. Nucleotide sequence analysis was performed on 18 UL97 amplicons obtained from ten solid organ transplant recipients. In three subsequent samples of a single patient (Patient A) nucleotide sequence analysis revealed a C“ G change, resulting in a cystein (TGC) to tryptophane (TGG) substitution at codon 603 (Table 3). This mutation has previously been shown by marker transfer to confer resistance to GCV (Chou et al., 1997). Synonymous nucleotide substitutions (C“ T) at the same codon were found in two additional samples of another patient (Patient B; Table 3). Further nonsynonymous mutations, associated with GCV resistance, were not found in the remaining eight patients (Patients C – J). Although only few reports on GCV resistance of HCMV from recipi-

ents of solid organ transplants exist, the frequency of genotypically detectable GCV resistance associated with mutations in the UL97 gene in our cohorts is in line with these studies (Chou et al., 1995b, 1997; Erice et al., 1998). In addition to the clinical samples, we were able to examine DNA from four cell culture-adapted HCMV isolates which were previously shown to exhibit GCV resistance in vitro1 (Metzger et al., 1994; Baldanti et al., 1998). In the UL97 fragment of two of these isolates, one non-synonymous mutation each, an A“ G change at codon 460 (isolate Ulm 3: methionine [ATG]“valine [GTG]), and an exchange of G“ A at codon 607 (isolate VR 5474: cystein[TGT] “ tyrosine [TAT]), were detected (Table 3). Previous marker transfer experiments have shown that these mutations confer GCV resistance (Chou et al., 1995a; Baldanti et al., 1998). Nucleotide sequence analysis of the UL54 (polynerase-encoding) gene of these isolates is currently in progress to further elucidate the molecular basis of their in vitro GCV resistance. In order to evaluate the suitability of reverse hybridization, we used nine oligonucleotide probes to screen for mutations at four codons of the UL97 gene (see Table 2). All 23 amplicons typed by reverse hybridization assay gave results fully concordant with nucleotide sequencing as 1 Kindly provided by Prof. Dr Mertens, University of Ulm, Germany, and by Prof. Dr Gerna, University of Pavia, Italy.

L. Zhou et al. / Journal of Clinical Virology 13 (1999) 53–59

57

Table 3 Comparison of nucleotide sequence analysis and reverse hybridization for the detection of GCV-conferring mutations in the UL97 gene of HCMVa Samples

Patient

In vitro resistance

Sequence analysis

Reverse hybridizationb

AD169

–

–

Wild-typec

Wild-type

2230/Kiel97 3493/Kiel97 4028/Kiel97

A A A

n.t. n.t. n.t.

C603W C603W C603W

C603W C603W C603W

929/Kiel97 1474/Kiel97

B B

n.t. n.t.

Wild-type (603: TGT) Wild-type (603: TGT)

Wild-type (603: TGT) Wild-type (603: TGT)

Kiel cohort (n =13)

n=8 (C-J)

n.t.

Wild-type

Wild-type

Ulm1 Ulm2 Ulm3 VR5474

C D E F

+ + + +

Wild-type Wild-type M460V C607Y

Wild-type Wild-type M460V C607Y

a

In vitro resistance was tested in the laboratories of Dr Mertens, University of Ulm, Germany, and Dr Gerna, University of Pavia, respectively who kindly made DNA of these isolates available for this study (Metzger et al., 1994; Baldanti et al., 1998). n.t., not tested. b Mutations at codons 460, 520, 603 and 607 were tested by reverse hybridization. c ‘Wild-type’ indicates identity of the deduced amino acid sequences with that of AD169.

shown in Table 3. Eighteen amplicons showed reactivity, similar to AD169, with wild-type probes exclusively. Examples of amplificates revealing mutations by reverse hybridization are presented in Fig. 1. Although the staining intensity appeared to be variable, discrimination of positive versus negative reactions was unambiguous with each probe. Therefore, each of the amplificates appeared to be homogenous by showing a uniform reactivity at the indicated codons. In order to confirm this assumption and to further investigate the sensitivity of the reverse hybridization assay, mixtures of amplificates from AD169 (wild-type) and the mutant M460V were examined. As indicated in Fig. 2 heterogenous amplificate populations can be detected at frequencies at least as low as 10%. For nucleotide sequencing, in contrast, it is generally accepted that frequencies ] 25% are required to be detected with certainty. No reactivity was evident with another probe, specific for a M460i mutation, showing that the oligos chosen for this codon were highly discriminatory. In conclusion, point mutations at codons 460, 603, and 607 of the HCMV UL97 gene were readily detected by reverse hybridization within

ten working hours. Reverse hybridization is simple to implement in laboratories experienced in standard PCR technology and might prove an useful alternative to nucleotide sequencing in genotypic GCV resistance analysis of HCMV. Problems with reverse hybridization, however, might be expected when designing specifically discriminating oligonucleotides in GC-rich regions and for codons where multiple mutations, also at neighbouring sites, do occur. These features are combined at codons 591–597, another GCV-relevant hot spot of the UL97 gene (Erice et al., 1996). Designing probes for this area as well as including relevant codons of the UL54 DNA polymerase gene are challenges for continuing studies in our laboratory.

Acknowledgements We are indebted to Prof. Dr Mertens, Institute of Virology, University of Ulm, Germany, and Prof. Dr. Gerna, Serv. Virologica, IRCC Policlinica, Pavia, Italy, who made DNA of cell cultureproven GCV-resistant HCMV isolates available to us. For technical assistance we are grateful to

58

L. Zhou et al. / Journal of Clinical Virology 13 (1999) 53–59

This project was partly financed by the scientistexchange programme in the frame of a partnership contract between Zhejiang University, Hangzhou, and Christian-Albrechts-University (C.A.U.), Kiel, and by an institutional fund through Prof. Dr U. Ullmann, Head, Institute of Medical Microbiology and Virology, C.A.U.

References

Fig. 1. Detection of GCV resistance-conferring mutations in UL97 amplicons by reverse hybridization. Wt, HCMV strain AD169 (wild-type control); 1, isolate Ulm 3 (M460V); 2, 2230/Kiel97 (C603W); 3, 929/Kiel97 (603 TGT); 4, Pavia VR5474 (C607Y). The boxed dots indicate non-synonymous mutations which were previously shown by marker rescue to confer resistance to GCV.

Mrs. B. Grafelmann and Mrs. P. Grossmann, Inst. Med. Microbiol. Virol., University of Kiel.

Fig. 2. Detection of heterogenous amplificate populations by reverse hybridization. UL97 nested amplicons of HCMV strain AD169 (wild-type) and mutant Ulm3 (codon 460 M“ V) were mixed at various percentages. The mixtures were subsequently examined for polymorphisms at codon 460: Wild-type (ATG), mutation M “V (GTG), and mutation M“ I (ATT). Populations present at frequencies of 10% can be safely detected.

Baldanti F, Underwood MR, Talarico CL, Simoncin L, Sarasini A, Biron KK, Gerna G. The Cys607-Tyr change in the UL97 phosphotransferase confers ganciclovir resistance to two human cytomegalovirus strains recovered from two immunocompromised patients. Antimicrob Agents Chemother 1998;42:444 – 6. Boivin G, Chou S, Quirk MR, Erice A, Jordan MC. Detection of ganciclovir resistance mutations and quantitation of cytomegalovirus (CMV) DNA in leukocytes of patients with fatal disseminated CMV disease. J Infect Dis 1996;173:523 – 8. Bowen EF, Johnson MA, Griffiths PD, Emery VC. Development of point mutation assay for the detection of human cytomegalovirus UL97 mutations associated with ganciclovir resistance. J Virol Methods 1997;68:225 – 34. Chou S, Guentzel S, Michels KR, Miner RC, Drew WL. Frequency of UL97 phosphotransferase mutations related to ganciclovir resistance in clinical cytomegalovirus isolates. J Infect Dis 1995a;172:239– 42. Chou S, Erice A, Jordan MC, Vercellotti GM, Michel KR, Talarico CL, Stanat SC, Biron KK. Analysis of the UL97 phosphotransferase coding sequence in clinical cytomegalovirus isolates and identification of mutations conferring ganciclovir resistance. J Infect Dis 1995b;171:576– 83. Chou S, Marousek G, Guentzel S, Follansbe E, Poscher ME, Lalezari JP, Miner RC, Drew WL. Evolution of mutations conferring multidrug resistance during prophylaxis and therapy for cytomegalovirus disease. J Infect Dis 1997;176:786 – 9. Erice A, GilRoda C, Pe´rez JL, Balfour HH, Sannerud KJ, Hanson MN, Boivin G, Chou S. Antiviral susceptibilities and analysis of UL97 and DNA polymerase sequences of clinical cytomegalovirus isolates from immunocompromised patients. J Infect Dis 1996;175:1087– 92. Erice A, Borrell N, Li W, Miller WJ, Balfour HH. Ganciclovir susceptibilities and analysis of the UL97 region in cytomegalovirus (CMV) isolates from bone marrow recipients with CMV disease after antiviral prophylaxis. J Infect Dis 1998;178:531 – 4. Griffiths PD, Emery VC. Cytomegalovirus. In: Richman DD:, Whitley RJ, Hayden FG, editors. Clinical Virology. New York: Churchill Livingstone, 1997:445 – 70.

L. Zhou et al. / Journal of Clinical Virology 13 (1999) 53–59 Hanson MN, Preheim LC, Chou S, Talarico CL, Biron KK, Erice A. Novel mutation in the UL97 gene of a clinical cytomegalovirus strain conferring resistance to ganciclovir. Antimicrob Agents Chemother 1995;39:1204–5. Jabs DA, Jacobson M. Cytomegalovirus (CMV) culture results, drug resistance, and clinical outcome in patients with AIDS and CMV retinitis treated with foscarnet or ganciclovir. J Infect Dis 1997;176:50–8. Littler E, Stuart AD, Chee MS. Human cytomegalovirus UL97 open reading frame encodes a protein that phosphorylates the antiviral nucleoside analogue ganciclovir. Nature 1992;358:160 – 2. Lurain NS, Thompson KD, Holmes EWand, Read GS. Point mutation in the DNA polymerase gene of human cytomegalovirus that result in resistance to antiviral agents. J Virol 1992;66:7146 – 52. Lurain NS, Spafford LE, Thompson KD. Mutation in the UL97 open reading frame of human cytomegalovirus strains resistant to ganciclovir. J Virol 1994;68:4427–31. Metzger C, Michel D, Schneider K, Luske A, Schlicht H-J, Mertens T. Human cytomegalovirus UL97 kinase confers gancicloir susceptibility to recombinant vaccinia virus. J Virol 1994;68:8423 – 7. Michel D, Schaarschmidt P, Wunderlich K, Heuschmid M, Simoncini L, Muhlberger D, Zimmermann A, Pavic L, Mertens T. Functional regions of the human cytomegalovirus protein pUL97 involved in nuclear localization and phosphorylation of ganciclovir and pUL97 itself. J Gen Virol 1998;79:2105–12. Noble S, Faulds D. Ganciclovir: An update of its use in the prevention of cytomegalovirus infection and disease in transplant recipients. Drug 1998;56:115–46. Scho¨neberg U, Vahrson W, Priedemuth U, Wittig B. Analysis

.

59

and interpretation of DNA and protein sequences using MacMollyTetra. Karaoi, Bielefeld, 1994. Smith HL, Cherrington JM, Jiles RE, Fuller MD, Freeman WR, Specctor SA. High-level resistance of cytomegalovirus to ganciclovir is associated with alteration in both the UL97 and DNA polymerase gene. J Infect Dis 1997;176:69 – 77. Spector SA, Hsia K, Wolf DG, Shinkai M, Smith IL. Molecular detection of human cytomegalovirus and determination of genotypic ganciclovir resistance in clinical specimens., Clin. Infect. Dis., 21. Suppl 1995;2:170 – 3. Stuyver L, Wyseur A, Arnhem WV, Hernandez F, Maertens G. Second-generation line pobe assay for hepatitis C virus genotyping. J Clin Microbiol 1996;34:2259 – 66. Stuyver L, Wyseur A, Rombout A, Louwagie J, Scarcez T, Verhofstede C, Rimland D, Schinazi RF, Rossau R. Line probe assay for rapid detection of drug-selected mutations in the human immunodeficiency virus type 1 reverse transcriptase gene. Antimicrob Agents Chemother 1997;41:284 – 91. Sullivan V, Talarico CL, Stanat SC, Davis M, Coen DM, Biron KK. A protein kinase homologue controls phosphorylation of ganciclovir in human cytomegalovirus-infected cells. Nature 1992;358:162 – 4. Sullivan V, Biron KK, Talarico CL, Stanat SC, Davis M, Pozi LM, Coen DM. A point mutation in the human cytomegalovirus DNA polymerase gene confers resistance to ganciclovir and phosphonylmethoxyalkyl derivatives. Antimicrob Agents Chemother 1993;37:19 – 25. Wolf DG, Smith IL, Lee DJ, Freeman WR, Flores-Aguilar M, Spector SA. Mutations in human cytomegalovirs UL97 gene confer clinical resistance to ganciclovir and can be detected directly in patient plasma. J Clin Invest 1995;95:257 – 63.