| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Molecular Diagnostics and Genetics |
1 Department of Internal Medicine, Liver Research Institute, Seoul National University College of Medicine, Seoul, Korea.
2 GeneMatrix, Seoul, Korea.
aAddress correspondence to this author at: GeneMatrix, 8F, Cancer Research Institute, Seoul National University College of Medicine, 28 Yongon-Dong, Chongno-Gu, Seoul 110-799, Korea. Fax 82-2-764-8076; e-mail sunphong{at}genematrix.net.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: The assay is based on PCR amplification and mass measurement of oligonucleotides containing genotype-specific motifs in the 5' untranslated region, into which a type IIS restriction endonuclease recognition was introduced by PCR amplification. Enzymatic cleavage of the products led to excision of multiple oligonucleotide fragments representing variable regions whose masses were determined by MALDI-TOF MS.
Results: The RFMP assay identified viral genotypes present at concentrations as low as 0.5% and reliably determined their relative abundance. When sera from 318 patients were analyzed, the RFMP assay exhibited 100% concordance with results obtained by clonal sequencing and identified mixed-genotype infections in 22% of the samples, in addition to several subtype variants.
Conclusions: The RFMP assay has practical advantages over existing methods, including better quantitative detection of mixed populations and detection of genotype variants without need for population-based cloning, enabling reliable viral genotyping in laboratories and efficient study of the relationship between viral genotypes and clinical outcome.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Many HCV genotyping methods have been developed, including restriction fragment length polymorphism analysis, genotype-specific PCR, heteroduplex mobility analysis, melting curve analysis with fluorescence resonance energy transfer probes, and line probe assay, a DNA hybridization method; however, nucleotide sequencing of an appropriate subgenomic region remains the most widely accepted method (13)(14)(15)(16)(17)(18). Most of these methods, however, are labor-intensive, complex to standardize, and/or ineffective for determining multiple HCV genotypes in an experimental setup. Because of a high rate of mutation, extensive HCV genetic heterogeneity may occur in an individual during the course of infection, and more than 1 genotype may occur in some patients, particularly in individuals who have received multiple transfusions or are intravenous drug users (19). The perceived rate of quasispecies and mixed-genotype infections is highly variable in patients tested by different assays (20), and it is difficult to assess the true prevalence of mixed infection by current methods because they are designed to identify only the predominant HCV genotype in the population. The presence or clinical significance of mixed-genotype infection thus may be underestimated (19)(21).
We have recently developed a matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS)–based genotyping assay that exploits differences between wild-type and variant sequences in hepatitis B virus (22). In the current study, we used restriction fragment mass polymorphism (RFMP) analysis, an advanced MALDI-TOF MS–based assay, for HCV genotyping capable of differentiating a cluster of multiple variations in the 5' untranslated region (UTR). We assessed the sensitivity and reliability of the assay by comparing the results with direct or clonal sequencing; we then used the RFMP assay to determine the genotype distribution and prevalence of mixed-genotype infections in 318 chronically infected patients.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
HCV RNA extraction and cDNA amplification
HCV RNAs were extracted from 140-µL serum samples by the QIAamp® Viral RNA method (Qiagen) according to the manufacturer’s instructions. Reverse transcription and PCR were performed in 1-step reactions. Known positive and negative sera and water controls were included in all reactions from the extraction step. The 5' UTR was amplified with first-round primer pairs described previously (23): KY80 (sense; 5'-GCAGAAAGCGTCTAGCCATGGCGT-3'; nucleotides –274 to –251) and KY78 (antisense; 5'-CTCGCAAGCACCCTATCAGGCAGT-3'; nucleotides –29 to –54). For the second round, we used 2 sets of nested PCR primer. A sense primer, CV160F, specific to bases –199 to –168 (5'-GTGGTCTGCGGAATCCAACGGTGAGTACACCGGAAT-3') and an antisense primer, CV100R, specific to bases –67 to –92 (5'-TTCGCRACCCAACRCTACTCCAACGGTCCGGCTAG-3') were used for detection of sequences from –160 to –100 (R = A or G). A sense primer, CV140F, specific to bases –158 to –140 (5'-GACIGGGTCCTGGATGTCTTGGA-3') and an antisense primer, CV120R, specific to bases –100 to –117 (5'-GCGGGGGCACGGATGCCCAAAT-3') were used for detection of sequences from –140 to –120. As illustrated in Fig. 1
, sequences underlined in the CV160F and CV100R primers were modified to insert MmeI and MmeI/AvaII restriction recognition sites, respectively, in the PCR products, and underlined sequences in the CV140F and CV120R primers were modified to introduce FokI sites in the products.
|
cDNA amplifications were performed with 20-µL reaction mixtures containing 10 µL of RNA, 20 mM Tris-HCl (pH 9.0), 50 mM KCl, 0.2 mM each deoxynucleotide triphosphate, 1.2 mM MgSO4, 0.4 µM each primer, 0.1 mM dithiothreitol, 40 units of RNase inhibitor (Invitrogen), 200 U of SuperScriptTM III Moloney murine leukemia virus reverse transcriptase (Invitrogen), and 2.5 U of Platinum® Taq polymerase (Invitrogen). After incubation at 50 °C for 45 min and at 94 °C for 2 min, PCR was performed for 35 cycles; each cycle consisted of denaturation at 94 °C for 15 s, annealing at 55 °C for 15 s, and extension at 72 °C for 20 s. To 17 µL of the second-round PCR mixture, 1 µL of the first-round PCR product was added. The nested PCR mixture was identical to the first-round mixture except for the omission of dithiothreitol, RNase inhibitor, and Moloney murine leukemia virus reverse transcriptase. PCR was conducted with the conditions described above, except that the annealing temperature was 60 °C. Nucleotide sequence positions were numbered according to the prototype HCV type 1 sequence (24).
RFMP assay for HCV genotyping
Restriction enzyme digestion of the nested PCR product obtained with the CV160F and CV100R primer pair was performed by mixing the PCR reaction mixture with 10 µL of reaction buffer containing 50 mmol/L potassium acetate, 20 mmol/L Tris-acetate, 10 mmol/L magnesium acetate, 1 mmol/L dithiothreitol, and 1 U of MmeI (New England Biolabs). The reaction mixture was incubated at 37 °C for 2 h and then incubated at 37 °C for 2 h with AvaII (New England Biolabs). Enzyme digestion of the nested PCR products obtained with the CV140F and CV120R primer pair was identical, except for the addition of 50 µmol/L S-adenosylmethionine in the mixture, the use of FokI (New England Biolabs), and the subsequent digestion with BstF5I (New England Biolabs). The resulting digest was purified by vacuum filtration through a custom-built 384-well sample preparation plate containing 5 mg of polymeric sorbent per well (Waters) or Oasis® µElution Plates (Waters). Each well was equilibrated with 90 µL of 1 mol/L triethylammonium acetate (TEAA; pH 7.6). Each cleavage reaction mixture was added to 70 µL of 1 mol/L TEAA (pH 7.6) and loaded in a well. After the wells were rinsed 5 times with 85 µL of 0.1 mol/L TEAA (pH 7.0), the plate was reassembled on a vacuum manifold and eluted with 60 µL of 600 mL/L isopropanol in water into a collection plate, which was placed on a heating block at 115 °C for 90 min. The desalted reaction mixtures were resuspended with matrix solution containing 15 g/L 3-hydroxypicolinic acid, 0.023 mol/L ammonium citrate, and 120 mL/L acetonitrile; they were then spotted in 3-µL volumes on a polished MTP AnchorChipTM plate (Bruker Daltonics). Mass spectra were acquired on a Biflex IV linear MALDI-TOF MS (Bruker Daltonics) workstation equipped with a 337 nm nitrogen laser and a nominal ion flight path length of 1.25 m as described previously, with slight modification (22). The samples were analyzed in negative-ion mode with a total acceleration voltage of 20 kV, extraction voltage of 18.25 kV, laser attenuation of 55, and delayed extraction of long time delay. Typically, time-of-flight data from 10 individual laser pulses were recorded and averaged on a transient digitizer with a time base of 2 ns and delay of 24 000 ns, after which the averaged spectra were automatically converted to mass by the accompanying data-processing software (Bruker Daltonics). With such settings, the instrument usually provides mass accuracy of 40–80 ppm (10–6), mass resolution of 1500–2000, and sensitivity of 10–50 fmol in the 2- to 6-kDa mass range for oligonucleotides.
validation of RFMP analysis for HCV genotyping
We compared the RFMP assay results with the results from both direct sequencing and the clonal sequencing assay. For confirmation of viral quasispecies or mixed-genotype infection, we cloned the PCR products into the PCR-Script Amp cloning vector (Stratagene) for sequence analysis of each clone (at least 10 clones per sample). The commercialized genotype controls, Armored RNA®-HCV genotype 1b and 2a/c (Ambion Diagnostics), were mixed in various ratios and used to evaluate the sensitivity and specificity of the RFMP assay. Sequence analysis was performed by the ABI PRISM 310 Genetic Analyzer (Applied Biosystems). To clarify the relationship among different HCV isolates, we constructed a phylogenetic tree as described elsewhere (25).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. The first step requires PCR amplification using primers flanking the genotype-specific motifs. From the alignment of the 5' UTR from published sequences, 3 variable regions, H1 to H3, surrounded by conserved regions were chosen as genotype-specific motifs of HCV. Two nested sets of universal primers flanking the variable regions, CV160F/CV100R and CV140F/CV120R, were selected and modified to introduce a type IIS restriction endonuclease recognition sequence ahead of the variable regions on amplification. For interrogating H1 and H2 motifs, CV160F and CV100R primers were designed to introduce the MmeI and combined MmeI/AvaII sites in the amplified product by substituting the restriction recognition sequence TCCAAC for CC (nucleotides –186 to –185) and TCCAACGGTCC for TC (nucleotides –49 to –50), respectively. To interpret the H3 motif, primers CV140F and CV120R were designed to introduce a FokI site (an isoschizomer of BstF5I) in the amplified product by substituting the restriction recognition sequence GGATG for T (nucleotide –147) and G (nucleotide –110), respectively. When the complementary strands are copied, the inserted sequences are incorporated into the product. Type IIS restriction enzymes such as MmeI, FokI, and BstF5I cleave DNA outside the recognition sequence: MmeI, 20 bases 3' to the recognition site on one strand and 18 bases from the recognition site on the other strand, leaving a 2-base overhang protruding 3' end; FokI, 9 bases 3' to the recognition site on one strand and 13 bases from the recognition site on the other strand, leaving a 4-base overhang protruding 5' end; BstF5I, 2 bases 3' to the recognition site on one strand and immediately 3' to the recognition site on the opposite strand, leaving a 2-base overhang. As outlined in Fig. 1
, cleavage of a 142-bp amplified segment with MmeI and AvaII leads to excision of the 12mer/17mer and 14mer/19mer fragments representing H1 and H3 motifs, respectively, whereas digestion of a 67-bp amplified segment with FokI and BstF5I leads to excision of 2 sets of 7mer/14mer/13mer fragments, reflecting the sequence variation of the H2 motif.
HCV 5' UTR sequences retrieved from the Entrez Nucleotide database of the National Center for Biotechnology Information were aligned against the HCV prototype 1 sequence (n = 871 as of May 2004; queried by "hcv and 5' UTR"). From some of the reported sequences covering all 3 variable motifs, we obtained 10 H1, 4 H2, and 28 H3 mass patterns, and a total of 41 combined patterns were assigned to the 6 major HCV genotypes without duplication (see Table 1
of the Data Supplement that accompanies the online version of this article athttp://www.clinchem.org/content/vol51/issue7/).
|
We found 10 different patterns for samples belonging to type 1, with subtype 1b accounting for the most diverse mass patterns (6 of 10). Genotype 2 could be divided into 8 patterns, with –138A, –132A, and –128T the most distinctive bases among this type. The 2a and 2c subtypes are not completely differentiated, but there is a tendency that TC at nucleotides –117 to –116 is abundant in subtype 2a, whereas CC is abundant in subtype 2c. Most of genotypes 3 and 4 were predicted to show specific patterns in the H1 motif indicative of the existence of –163G and –159T. The patterns for genotype 4 were more complicated and diverse, presenting 12 kinds, among which 4c and 4d could not be differentiated. Genotypes 5 and 6 showed 2 subtype patterns each. From these analyses, 6 major genotypes and 27 subtypes of HCV were predicted to be differentiated into 41 prototype mass patterns as summarized in Table 1
of the online Data Supplement.
detection of mixed HCV genotypes
To assess the sensitivity and validity of the RFMP assay for detecting mixed genotypes in a viral population, we performed the assay with samples composed of defined ratios of 5' UTR sequences of HCV genotype 1b and 2a/c in RNAs. The RNAs were mixed in the following proportions of genotype 1b to the mixture of 2a/c and 1b: 50%, 40%, 30%, 20%, 10%, 5%, 0.5%, and 0%. The RFMP assay detected 0.5% genotype 1b when assessed with 12mer from the H1 motif (Fig. 2A
). The DNA sequencing assay could not detect genotype 1b mixed with a 5-fold greater amount of genotype 2a/c, in agreement with previous reports (26).
|
To test for a relationship between estimated and real proportions, we calculated ratios using the mean peak areas generated from multiple mass spectra and compared them with the defined pooling ratios. We observed a linear correlation of the estimated to the expected proportions between genotypes in the range from 0.5% to 50% (R2 = 0.992; Fig. 2B
). The significant differences obtained in dynamic range did not depend on which spectral band was chosen for quantification among the 10 restriction fragments, except for a pair of 7mers coming from the H3 motif. The 7mers from the H3 motif showed a lower limit of minor genotype detection as low as 0.5% and a dynamic range of 0.5%–50% (R2 = 0.895). Although the 7mers from the H3 motif seem inferior to other fragments in representing the relative abundance of a mixed-genotype infection, they do not affect the lowest concentrations needed for distinguishing mixed genotypes because the sensitivity was comparable to that for other fragments. The RFMP assay thus could detect mixed genotypes in a viral population and determine their relative abundances.
HCV genotype distribution in patients determined by RFMP analysis
We determined the genotype distribution of the 318 samples from chronically infected patients by RFMP and its representative spectra (Table 1
and Fig. 1
of the online Data Supplement). All of the samples were successfully genotyped by RFMP analysis, and the results agreed with clonal sequence analysis. Consistent with previous reports, genotypes 1b and 2a/c were the most prevalent, constituting 83% of the samples (27). Of interest, 22% of the samples were determined to be infected with mixed genotypes. Similar to single-genotype infections, the 2 predominating genotypes in mixed-genotype infections were 1b and 2a/c, which were found in 77.3% of the samples. Of the mixed-genotype infections, 
30% were determined by direct sequencing to be a single genotype or ambiguous but were further confirmed by clonal sequencing (data not shown). Among samples with mixed-genotype infections, genotype 1b was found in 87.1% (61 of 70) of the individuals, and it was commonly mixed with 2a/c in 42.6% (26 of 61). Of 318 chronic patients with samples analyzed by the RFMP assay, 21 among 38 cases were observed with deviations that did not correspond to the predicted mass pattern from the reference sequences. For example, a motif code pattern (H1/H2/H3) of 2/2/7 was found in 5 cases diagnosed as single genotypes, and in 1 case it was mixed with 2/1/7 (genotype 2a/c). Although the sequences lacked an exact match in Table 1
of the online Data Supplement, they could be inferred by combining the motif sequences corresponding to codes 2, 2, and 7 in motif order, as shown in the first line of genotype 2 in Table 1
. The sequences extracted by the above-mentioned approaches fully agreed with the sequencing result. By direct comparison with or construction of a phylogenetic tree linked to the known HCV genotype sequences, the 21 types of mass pattern deviants were assigned to 6 and 15 variants falling into genotypes 1 and 2, respectively (Table 1
).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MS generates precise information on the molecular mass of the analyte, both DNA strands can be analyzed in parallel, and the procedure can be fully automated. MS directly assesses the nature of the PCR products, either as a whole or as fragmented oligonucleotides, whereas other technologies only indirectly measure PCR products, either through hybridization or by primer-extended minisequencing reactions, which use PCR products as templates. Genotyping by creating or abolishing recognition sites for restriction enzymes, similar to conventional restriction fragment length polymorphism analysis, has been used in combination with MALDI MS detection (28). Either naturally occurring restriction sites were used or base changes were incorporated in one of the PCR primers to create a recognition site with one of the alleles of the polymorphic site. Only a small number of polymorphisms, however, will alter known restriction sites, and the design of amplification primers to create restriction sites in connection with 1 allele is not straightforward in most cases, reducing the usefulness of this approach to very special circumstances.
In this study, we established an RFMP assay for HCV genotyping that exploits differences among HCV genotypes in the molecular masses of oligonucleotides comprising the viral 5' UTR. The assay is based on amplification and mass detection by MALDI-TOF MS of oligonucleotides excised from type IIS enzyme digestions (Fig. 1
). The use of a type IIS restriction enzyme makes this assay independent of the chance occurrence of restriction sites because these enzymes have cleavage sites distal to their recognition sites. Recognition sites are incorporated into the amplification primers, and short fragments that contain the polymorphisms can be generated for mass spectrometric analysis. In addition to its speed and high-throughput capacity, the RFMP assay is very sensitive. It distinguishes between mixed-genotype infections when the minor species constitutes as little as 0.5% of the population (Fig. 2
), revealing clear correlations between peak ratios and relative genotype concentrations in mixed populations. For samples from 318 chronically infected patients, the RFMP assay showed excellent concordance with clonal sequencing and detected diverse polymorphisms whose sequences deviated from reported HCV genotypes (Table 1
). The significant incidence of mixed-genotype infections was detectable because of the assay’s higher sensitivity, leading us to conclude that HCV mixed-genotype infections are more common than previously estimated.
Taken together with previous findings that mixed infections seem to be associated with unfavorable disease outcomes and that detection of minor genotypes before antiviral therapy may be important for choosing an adequate regimen (21)(29)(30)(31), our results indicate that further testing of mixed-genotype infections is needed to expand current treatment algorithms that rely on specific, single genotypes. This research should be facilitated by the high sensitivity of the RFMP assay for detecting mixed infections and the capacity of the RFMP assay to determine relative abundances among viral species. In terms of cost-effectiveness, we estimated the direct cost per test (reagents and labor) of the RFMP assay to be $30.00 (US), including viral nucleic acid extraction, PCR, restriction digestion, restriction fragment purification, and matrix, which is much less expensive than the line probe assay at $130.00 (US) per test. Assuming that the capital equipment costs for the Biflex IV in our laboratory, estimated at $50 000.00 including annual amortization and maintenance, are similarly applicable, clinical laboratories that expect to test >500 specimens referred for HCV genotyping per year could consider the RFMP assay as an alternative to the line probe assay. In laboratories already using MALDI-TOF MS for clinical genotyping or for clinical proteomics, these costs could be avoided or minimized because the same instrumentation is used for the RFMP assay.
Accurate determination of ion masses is achievable because of high mass resolution and low ppm precision in the peak centroid positions for the raw flight time data (32)(33), which are necessary features for resolving adjacent peaks, particularly with increasing oligonucleotide fragment size. Although the absolute error, on average, was less than +0.5 Da and the width of the mass peak at the half maximum was as low as 1–4 Da for 2- to 6-kDa mass ranges in our experimental setup, concerns could arise relating to distinctions between multiple fragments of extremely similar masses, such as codes 1 and 4 of motif H2, and consequently between genotypes 1a (codes 1/1/1) and 1b (1/4/1) if present in the same specimen. The resolution achieved by the RFMP assays reported here should be achievable with MALDI-TOF MS instruments with specifications and settings comparable to or beyond those used here. A limitation of the RFMP assay, however, is that it determines only molecular mass and, therefore, will fail to detect alterations in sequences that do not lead to a change in molecular mass. To provide certainty in such cases, anticipated from the reported sequences as exemplified in Table 1
of the online Data Supplement for codes 3 and 6 of motif H1 and code 1 of motif H2, we designed the assay so that mass patterns obtained from other motifs would enable exact genotype differentiation. The presence of a novel genotype variant with sequence alterations that do not cause a deviation from the molecular mass pattern predicted from up-to-date HCV databases is possible, but we believe it would be extremely rare, as corroborated by the fact that we found no such case in 318 patients by either RFMP or sequencing assays. Alternative mass spectrometric techniques to consider involve collision-induced dissociation of polyprotonated oligonucleotides produced by electrospray ionization introduced to allow de novo sequencing of DNA oligomers up to 19 bases in length (34). Even longer fragments could be handled if the sequences differ at not more than 2 nucleotide positions from a reference sequence with mass-to-charge values predicted by established fragmentation pathways (35). These techniques could be considered to investigate potential sequence variations leading to no change in molecular mass, although whether they are amenable to high-throughput genotyping of multiple variations for clinical applications is not yet established.
In conclusion, the RFMP genotyping method uses the mass difference of oligonucleotides, requires the simple steps of single PCR amplification and restriction enzyme digestion, and is amenable to high-throughput analyses. The RFMP method has practical advantages over existing methods, including more sensitive and specific detection of mixed populations, as well as primary screening of novel viral genotype variants leading to deviation from the predicted reference mass patterns without need for cloning or multiple PCR steps. This screening method allows reliable viral genotyping in laboratories and efficient analysis of large samples and viral quasispecies to address the dynamics of HCV infections and the relationship between HCV genotypes and clinical outcome. The RFMP genotyping assay is easily adaptable for the detection of other genetic variations, including polymorphisms, deletions, and additions, as well as motifs with multiple variations, because the range and sizes of the resulting restriction fragments analyzed in MALDI-TOF MS can be chosen according to the investigator’s needs. We believe, therefore, that the RMFP genotyping assay will be quite useful for developing molecular diagnostics needed to differentiate clusters of multiple variations, such as virus genotyping, and for conducting pharmacogenomic studies of multiple single-nucleotide polymorphisms.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 Nonstandard abbreviations: HCV, hepatitis C virus; MALDI-TOF MS, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry; RFMP, restriction fragment mass polymorphism; UTR, untranslated region; and TEAA, triethylammonium acetate.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
therapy and outcome. Transplantation 1997;64:422-426.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
: long-term follow-up and prognostic relevance of HCV genotypes. J Hepatol 1996;24:67-73.[ISI][Medline]
[Order article via Infotrieve]
therapy in chronic hepatitis C. Hepatology 1995;22:1050-1056.[CrossRef][Medline]
[Order article via Infotrieve]
2a therapy in Japanese patients with chronic hepatitis C. Liver 1996;16:23-27.[ISI][Medline]
[Order article via Infotrieve]
therapy in hepatitis C virus infection. Hepatology 1994;19:1088-1094.[CrossRef][ISI][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  M. I. L. Sjoholm, J. Dillner, and J. Carlson Multiplex Detection of Human Herpesviruses from Archival Specimens by Using Matrix-Assisted Laser Desorption Ionization-Time of Flight Mass Spectrometry J. Clin. Microbiol., February 1, 2008; 46(2): 540 - 545. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Cook and K. R. Jerome Mass Spectrometry for Genotyping Hepatitis C Virus: A Promising New Approach Clin. Chem., July 1, 2005; 51(7): 1091 - 1092. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
