Human cytomegalovirus is a common opportunistic pathogen in individuals with compromised or immature immune systems, such as transplant recipients, patients with acquired immunodeficiency syndrome, or congenitally or perinatally infected children. Three classes of systemic drugs are currently licensed for treatment of HCMV infections and their associated pathology. GCV and ValGCV are phosphorylated by viral protein kinase (UL97) and inhibit DNA synthesis after incorporation by viral DNA polymerase (UL54). CDV does not require phosphorylation by UL97 and inhibits DNA synthesis. PFA inhibits viral DNA synthesis by acting as a pyrophosphate analogue. Mutations associated with GCV or PFA resistance have been detected in the UL97 and UL54 genes of HCMV identified in clinical specimens. Thus, the targets of current anti-HCMV drugs are UL97 and UL54 [1-5].
We report a rapid and sensitive assay for UL97 and UL54 mutations by direct amplification of patient specimens and sequencing for GCV and PFA resistance that can be performed within a few days and does not require growing HCMV in cell culture. In this study, we directly amplified HCMV in clinical specimens and determined the sequences of the mutation sites of UL97 and UL54 genes that confer resistance to GCV, CDV and PFA. Furthermore, sequencing results for virus isolated by cell culture and for virus from patient specimens directly amplified by PCR were in agreement.
We obtained 13 clinical specimens from 7 bone marrow and/or renal transplant recipients and examined the susceptibilities of the CMV isolates to anti-CMV drugs and the UL97 and UL54 sequences by plaque reduction assay and DNA sequencing [6-9].
DNA was extracted from plasma, urine and peripheral blood mononuclear cell samples using a DNA extraction kit (QIAmp DNA blood mini kit; Qiagen, Tokyo, Japan). The UL97 gene was amplified by the following primer sets: forward primer U2 (5′-TGCCCAAAGAGGACGATTTT-3′) and reverse primer UL97-R (5′-GTAGTCCAAACTCGAGACGC-3′). For amplification of the UL54 region, single-step PCR using a KOD FX Neo (Toyobo Life Science, Osaka, Japan) DNA polymerase was performed and optimized using Touchdown and Stepdown PCR programming, with forward primer 155 sense (5′-ACGGTCAGACGGGGTTGATCAAGCA-3′) and reverse primer 3602 antisense (5′-AGCACGTTGGTTACAGCCTTGAGAACCT-3′). First, one cycle of denaturing for 2 mins at 94°C, a second step of denaturing for 10 s at 98°C, annealing for 30 s at 70°C and primer extension for 5 mins at 68°C was performed. Then, the annealing temperature was decreased by 1°C/cycle beginning at 70°C, followed by 45 additional cycles of denaturing for 10 s at 98°C, annealing for 30 s at 60°C and primer extension for 5 mins at 68°C. This PCR program yields a single strong amplicon from a small number of copies of HCMV genomic DNA. The amplified regions were then sequenced with multiple forward and reverse primers using standard fluorescent dideoxy sequencing (BigDye Terminator ver. 3.1 kit and 3130 Analyzer, Applied Biosystems, Foster, CA, USA). All amino acid differences from the Towne strain (accession numbers GU980198) and AD169 strain (X17403) were tabulated, including mixtures of sequence variants from the wild type as recognized by multiple peaks at single base locations in the context of a clean baseline.
To measure HCMV genome copy number, 2 μL of extracted DNA solution in reaction mixture was used for real-time PCR with a HCMV UL54 DNA polymerase gene-specific primer set, forward primer (5′-GCGCGTACCGTTGAAAGAAAAGCATAA-3′) and reverse primer (5′-TGGGCACTCGGGTCTTCATCTCTTTAC-3′) (SYBR Premix Ex Taq; Takara Bio, Otsu, Shiga, Japan). Before the quantitative PCR, the mixture was incubated at 95°C for 10 s. The quantitative PCR reaction was performed at 95°C for 5 s and 60°C for 30 s for 45 cycles and the amounts of PCR products monitored with a Takara Dice thermal cycler for the real-time PCR system and analyzed with Takara Real-time PCR software (Takara Bio). This study was approved by the Ethics Committee of the University of Toyama.
Table 1 presents a summary of the clinical specimens and drug resistances. Blood and urine were collected and sent to our laboratory by courier service at room temperature, 4°C or frozen. On receipt, blood was separated into PBMCs and plasma. All specimens, namely PBMCs, serum, plasma and urine, were successfully amplified for CMV DNA and mutations related to drug resistance identified as described after determination of their UL54 and UL97 sequences. The DNA sequences identified in clinical samples and in simultaneously isolated CMV from the same specimens were identical and drug resistance was confirmed by susceptibility assays of the isolated viruses. The copy numbers of HCMV genomic DNA in the samples ranged from 6 copies/µg of DNA in PBMCs to 609,000 copies/mL in plasma, indicating high sensitivity for the resistance genotype.
Table 1. Clinical specimens and drug resistances
|Hospital of origin of clinical isolate||Transplanted organ||Age (yrs)||Therapy||Nature of sample||Date||Antigenemia||HCMV DNA||Amino acid substitutions||IC50 (µg/mL)|
|Kanazawa||Bone marrow||11|| ||PBMC||5 April 2011|| ||Not tested||D605E*1||V355A*4, A688V*5||5.00 + 0.22||>100|
|Nagoya-1||Bone marrow||59||GCV, PFA||PBMC||27 April 2009||(20, 29) 27 April 2009||Not tested||A594P*2, D605E*1||V787A*6|| || |
|Nagoya-2a|| || || ||PBMC||9 December 2011|| ||1,863|| || ||1.58 + 0.38||>100|
|Nagoya-2b||Bone marrow||9||GCV, PFA, CTL, CDV||Urine||(271, 238) 1 December 2011||27,475|| || || || |
|Nagoya-2c|| || || ||PBMC||13 December 2011|| ||156||A594V*3||V355A*4, A688V*5||1.58 + 0.52||>100|
| || || || ||Urine|| ||1,260|| || || || |
| || || || ||PBMC||14 December 2011|| ||276|| || ||1.78 + 0.65||>100|
| || || || ||Urine|| ||987|| || || || |
|Nagoya-3a|| || || ||PBMC||11 January 2012||(147, 102) 4 January 2012||491|| || || || |
|Nagoya-3b||Bone marrow||8||GCV, PFA||PBMC||12 January 2012|| ||262||A594V*3||V355A*4, A688V*5|| || |
|Nagoya-3c|| || || ||PBMC||13 January 2012|| ||259|| || || || |
|Wakayama||Bone marrow||41||GCV, PFA, CDV||PBMC||16 April 2012||283 / 89,400 cells||438,814||No substitition||R71G*7, V355A*4, A688V*5, V715M*8|| || |
| || || || ||Plasma|| || ||609,000|| || || || |
|Niigata-1||Renal and bone marrow||53||GCV||PBMC||11 July 2012||(0, 1) / 15 × 104 cells||6||A594V*3||V355A*4, A688V*5|| || |
| || || || ||Plasma|| || ||1,112|| || || || |
|Niigata-2a|| || || ||PBMC|| || ||252|| || || || |
|Niigata-2b||Renal (2003, 2012) and bone marrow||42||GCV, ValGCV, IVIG||Plasma||11 July 2012||(3, 18) / 15 × 104 cells||177,650|| || || || |
| || || || ||PBMC||17 July 2012||(2, 6) / 15 × 104 cells||295||A594V*3||V355A*4|| || |
| || || || ||Plasma|| || ||10,535|| || || || |
|Niigata-2c|| || || ||PBMC||24 July 2012||(3, 2) / 15 × 104 cells||36|| || || || |
| || || || ||Plasma|| || ||2,218|| || || || |
Genotypic analyses covered the drug resistance mutation regions, namely 320–707 of 707 aa in UL97 and 53–1200 of 1242 aa in UL54 [1, 2, 4, 5, 10-17]. UL97 substitutions A594V and A594P were frequently detected and the UL97 substitution D605E was detected in two patients. UL54 substitutions V355A and A688V were detected in most samples. Substitutions V787A and V715M seem to confer resistance to PFA . The sample from Wakayama had no substitution in UL97 but had four substitutions related to resistance in UL54, these being identified by UL54 sequencing. Thus, simultaneous identification of UL97 and UL54 is important for genotypic sequencing of drug resistance. In some clinical isolates of HCMV, genotypic mutations were confirmed to be GCV resistant phenotypes by plaque reduction assay (Table 1). Identification from clinical specimens of the UL97 mutation alone has been reported, the 95% detection limit for obtaining usable sequence information having been <750 copies/mL for both amplicons . Rapid detection of drug resistant mutations of UL97 by novel procedures has been reported [16, 19-25]. However, using the present assay, drug-resistant mutations in clinical specimens were successfully identified from samples with as few as 6 and 36 copies/µg of DNA.
The sensitivity of this assay was quite good for amplifying the UL97 and UL54 genes in urine, blood, PBMCs, serum and plasma and the drug resistant mutations identified directly from the clinical specimens were confirmed in the isolated drug-resistant viruses. Thus, this rapid and sensitive assay system for genotyping drug resistance associated with UL97 and UL54 sequences may be useful for determination of anti-CMV resistance in clinical specimens, thus facilitating appropriate choices of antiviral agents for treating HCMV-related pathology. In summary, this rapid assay based on direct PCR and UL97 and UL54 sequencing analysis of clinical specimens can identify HCMV-drug resistant mutants quickly, is a highly sensitive method compared with previous methods and allows prompt, optimal treatment.