|
Epicentre Forum 6 (2)
||
Rémi N. Charrel(1,2) and Laura J. Chandler(1)
1Center for Tropical Diseases, Department of Pathology, University
of Texas Medical Branch, Galveston, Texas
2Laboratoire de Virologie Moléculaire, Tropicale et Transfusionnelle,
Faculté de Médecine, Marseille, France
Since the earliest days of virology, typing of viruses has been an
important tool to characterize viral populations and to study their epidemiology.
Typing provides information on the relationship among isolates within
the same group, species, or genus. Historically, serological methods
have been used to identify antigenic differences among virus populations.
Increasingly, nucleotide or deduced amino acid data have largely replaced
serology to provide more sophisticated epidemiological information. However,
most molecular methods other than DNA sequencing, such as PCR-based techniques,
pulsed field gel electrophoresis, and ribotyping, are either not suitable
for viruses or provide data that are too weakly discriminative for typing
purposes. Although sequencing is still the preferred method, it is an
expensive and time consuming technique, especially when processing a
large number of samples. Thus, other molecular methods that are easier
and less expensive to perform would be useful for typing of virus isolates
for epidemiological studies.
St. Louis encephalitis (SLE) is a human disease caused by a mosquito-borne
virus, which is a member of the genus Flavivirus within the family Flaviviridae.
SLE virus is widely distributed throughout Canada, the United States,
Mexico, Central and South America, and the Caribbean, and is an important
cause of arboviral encephalitis in the U.S., with cases occuring in both
epidemic and endemic forms.1 SLE
virus consists of a positive-sense RNA genome of about 11 kb in length
containing 5' and 3' non-coding regions flanking a single open reading
frame of approximately 10,500 nucleotides.1
Epidemiological and experimental evidence has suggested the existence
of SLE strains harboring distinct biological differences correlated to
genetic characteristics, and may impact transmission and disease potential
including neurovirulence, ability to cause viremia in birds, and mosquito
infectivity. Molecular techniques that have been used for identification
of genetic variation among SLE virus isolates include oligonucleotide
fingerprinting, single-strand conformation polymorphism (SSCP), and nucleotide
sequencing methods.2,3
In this report, we demonstrate the use of the BESS-T™ Base
Reader Kit to identify genetic variation and perform phylogenetic analysis
of the envelope gene of 22 SLE virus isolates from various geographical
locations in the Americas, and compare the results with those obtained
by direct PCR sequencing in the envelope gene.4 The
BESS-T Kit is a rapid and powerful method for detecting and localizing
approximately 95% of all mutations.5 The
BESS procedure involves PCR amplification using one labeled primer in
the presence of a limiting amount of dUTP. The uracil-containing PCR
product is then enzymatically cleaved at the sites of deoxyuridine incorporation,
resulting in a set of nested fragments that are separated on a sequencing
gel.
Each of the 22 virus strains4 was
inoculated onto a confluent Vero cell monolayer and incubated for 3-5
days at 37°C until the cytopathogen effect was evident. RNA was extracted
with Trizol LS reagent (Gibco-BRL) under manufacturer's recommendations
and resuspended in 50 µl of sterile, RNase-free water. Reverse
transciption was performed with the reverse primer at 42°C for one hour.
Five microliters were used to amplify a 750-bp fragment of the envelope
gene with the forward primer F880 (GATTGGATGGATGCTAGGTAG) and the reverse
primer B1629 (GGTTCAAGTCGTGAAACCAGTC). The PCR reaction was performed
with the following program: denaturation at 92°C for 1 minute, primer
annealing at 56°C for 1 minute, and extension at 72°C for 2 minutes for
25 cycles, followed by a 7-minute final extension. PCR products were
purified using the Wizard™ PCR Preps System (Promega).
BESS-T analysis
Two microliters of a 1:50 dilution of the PCR product was used as target
DNA for amplification in a 25 µl reaction containing 2.5 µl
of 10X PCR buffer, 1.5 mM final concentration MgCl2, 2 µl
of BESS-T dNTP Mix (containing 2.5 mM of each dNTP and 200 µM dUTP,
Epicentre), 1.25 U Taq DNA polymerase, 3 pmoles of 6-FAM fluorescent-labeled
forward primer (Perkin-Elmer), and 3 pmoles of reverse primer. Amplification
cycles were identical to those aforementioned. Eight microliters of PCR
product containing dUTP were then mixed on ice with 1 µl of BESS-T
10X Excision Enzyme buffer, 0.5 µl of BESS-T Excision Enzyme Mix,
and 0.5 µl of sterile water, and incubated at 37°C for 30 minutes.
The reaction was stopped by heating at 95°C for 2 minutes.
Gel electrophoresis
Two microliters of the excision reaction were mixed with 2 µl
of formamide and 0.5 µl of fluorescent dye-labeled GS500 size marker
(Perkin-Elmer), heated at 95°C for 2 minutes, then quickly cooled on
ice. Two microliters were loaded on a standard sequencing gel (6 M urea,
4.8% PAGE-PLUS, Amresco) and run for 4 hours on an ABI Prism® 377XL
automated sequencer (Perkin-Elmer).
Data collection
The analysis of the resulting BESS profiles was performed with the Genotyper® 2.0
software (Perkin-Elmer). Analysis was performed using only the BESS-T
fragments between nucleotides 924 and 1105 (numbered after the strain
MSI-7, GenBank accession no. M16614). Results were provided in a spreadsheet
format where labeled fragments were ordered by size and converted into
a sequence-like format, which was amenable to analysis using phylogenetics
software. The sequence-like data was constructed by replacing each detected
fragment by a "T." Nucleotides other than "T," potentially "A," "C," or "G," were
identified as "V," the code for "non-T" nucleotide according to the nomenclature
to identify redundancies (Table 1). The data were analyzed with the MEGA
software program6 and the robustness of the resulting branching
patterns was tested by bootstrap analysis (a statistical method based
on repeated random sampling with replacement from an original data set
to provide a collection of new pseudoreplicate samples, from which sampling
variance can be estimated) with 500 replications.
 |
Table 1. BESS-T profile analysis. BESS-T analysis of fragments
detected with Genotyper 2.0 software (rows A and B) and the binary
sequence (row C) used for phylogenetic analysis. A, size of the marker
co-migrating with the sample; B, each value corresponds to the size
of a BESS-T fragment; C, the binary sequence deduced from the BESS-T
fragments, each detected fragment corresponds to a T base; D, DNA
sequence of SL-33 isolate; E, nucleotide positions are numbered relative
to the MSI strain (GenBank accession no. M16614); - corresponds to
the absence of detected fragment; V corresponds to nucleotide A or
C or G. |
Nucleotide sequence determination of the 22 SLE virus strains
PCR products were sequenced directly using the F880 and B1629 primers
on an ABI Prism 377XL automated sequencer with the BigDye™ Terminator
Cycle Sequencing Ready Reaction Kit using AmpliTaq™ FS (Perkin-Elmer).
Nucleotide sequences located between positions 924 to 1604 were used
to perform phylogenetic analysis.
The conversion of the primary data (BESS-T scan, Figure 1) into a binary
sequence suitable for phylogenetic analysis and virus isolates comparison
is presented in Table 1. Phylogenetic trees based on the 182-nucleotide
fragment analyzed by the BESS method and from 182- and 681-nucleotide
sequence analyses are presented in Figures 2, 3, and 4 respectively.
 |
Figure 1. Representative BESS-T scans of two different isolates. Differences
between the two strains are indicated by the shaded peaks. |
 |
Figure 2. Typing of SLE isolates into 4 groups (A1, A2, B1,
and B2) based on geographic origin using the BESS-T method. Phylogenetic
analysis using the BESS technique of SLE virus isolates was based
on a 182 nucleotide fragment encompassed between nucleotides 924
and 1105 (numbered after the strain MSI-7, GenBank accession no.
M16614) located in the envelope gene. Distances and groupings were
determined using the MEGA software program. Bootstrap values are
indicated and correspond to 500 replications. |
 |
Figure 3. Typing of SLE isolates into 4 groups (A1, A2, B1,
and B2) based on geographic origin by DNA sequencing. Phylogenetic
analysis of SLE virus isolates was based on a 182 nucleotide DNA
sequence encompassed between nucleotides 924 and 1105 (numbered
after the strain MSI-7, GenBank accession number M16614) located
in the envelope gene. Distances and groupings were determined using
the MEGA software program. Bootstrap values are indicated and correspond
to 500 replications. |
 |
Figure 4. Typing of SLE isolates into 4 groups (A1, A2, B1,
and B2) based on geographic origin by DNA sequencing. Phylogenetic
analysis of SLE virus isolates based on a 681 nucleotide DNA sequence
encompassed between positions 924 to 1604 (numbered after the strain
MSI-7, GenBank accession number M16614) located in the envelope
gene. Distances and groupings were determined using the MEGA software
program. Bootstrap values are indicated and correspond to 500 replications. |
Regardless of the method (BESS-T or nucleotide sequence), the 22 isolates
formed 4 groups based on geographic origin. Eight Panamanian isolates
form a clade designated A1 and five South American isolates form the
A2 group. North American isolates were divided into two groups: B1, including
four isolates recovered from Missouri, Maryland, and Florida; and B2,
containing five isolates from California and Texas. The robustness of
these groupings was confirmed by the high bootstrap values, ranging from
69% to 100%, obtained for each of the four groups in Figure 4.
Molecular techniques can provide a powerful approach to epidemiological
typing because of their ability to detect minor genetic changes. DNA
sequencing is currently the "gold standard" in molecular analysis of
viruses. However, DNA sequencing of RNA viruses, especially viruses that
are difficult to cultivate, remains expensive, time-consuming, and sometimes
technically challenging. A highly purified product is required for direct
sequencing, and in some cases, sequencing must be preceded by cloning
and subcloning procedures.
Viruses, because of their small genome size and in many cases their
RNA content, are not easily typed by such methods as pulsed field gel
electrophoresis of DNA macrorestriction fragments and many PCR-based
techniques. Recently, viruses have been typed by various techniques used
as alternatives to DNA sequencing. Most of these involve PCR amplification
followed by restriction fragment length polymorphism (RFLP), SSCP,7,8 Cleavase® fragment
length polymorphism (CFLP®),9 or
heteroduplex migration analysis (HMA).10
In RFLP, the choice of enzymes used for restriction digestion needs
to be based on the comparison of a large number of sequences to be reliable,
and thus is more frequently applied to highly studied viruses. However,
even in this favorable case, only relatively short genomic regions (few
to 20-base sequences) are investigated for mutations. Additionally, when
isolates are closely related, specific distinguishing restriction patterns
can rarely be determined. CFLP has recently been used for hepatitis C
virus genotype determination9 and
seems to be a promising technique, but as in SSCP and HMA, precise localization
of the mutations is not possible. Moreover, the pattern provided presents
many bands, and computerized analysis may be required when numerous isolates
are investigated. RFLP, SSCP, and HMA allow analysis of longer DNA sequences
than BESS, but because quantitative data are not collected, they are
not suitable for phylogenetic analysis. Mutation mapping cannot be achieved
by HMA and SSCP and both need to be optimized to provide reliable results.
CFLP and RFLP provide a more precise localization of the mutations, but
neither method indicates the nature or the exact positions of the mutations.
Finally, most of these methods cannot be used for phylogenetic analysis
because of the paucity of quantitative data they generate. Moreover,
in none of these techniques do the resulting patterns consistently correlate
with the nucleotide sequences of the isolates under study.
In this study, we evaluated the BESS-T Base Reader Kit as a tool to
study the molecular epidemiology of SLE isolates in the Americas.4 We
compared the ability of the BESS-T method to that of direct sequencing
to identify genetic variation among the isolates and to provide phylogenetic
data. The 22 isolates have been correctly assigned according to their
geographical origin.
Compared to nucleotide sequence data, results obtained by the BESS-T
method demonstrate its capability to discriminate between closely related
virus strains and thus its usefulness for epidemiological and phylogenetic
studies. Because of its simplicity and low-cost, it is applicable for
processing large numbers of samples. The sequence-like format of the
BESS-T results allows analyses based on computer programs, such as multiple
alignment procedures and phylogenetic inferences, which is mostly unlikely
with concurrent techniques aforementioned. Although not yet tested, BESS
has potential for genetic comparisons of viral strains involved in clinical
outbreaks of viral diseases to help investigate their epidemiology. In
conclusion, any viral population that can be PCR amplified is a candidate
for study of its phylogenetic relationships and its molecular epidemiology
by application of the BESS technology.
- Monath, T.P. et al. (1996). Flaviviruses In Fields
Virology, Lippincott-Raven, Philadelphia, PA.
- Chandler, L. J. and Nordoff, N. G. (1999) J. Virol. Methods, 80,
169.
- Kramer, L.D. et al. (1997) Am. J. Trop. Med. Hyg. 57,
222.
- Charrel, R.N. et al. (1999) J. Clin. Microbiol. 37 (6),
1935.
- Hawkins, G. A. et al. (1997) Nat. Biotechnol. 15,
803.
- Kumar, S. et al. (1993) MEGA: Molecular Evolutionary Genetics
Analysis, version 1.02, University Park, PA.
- Fujioka, S. et al. (1995) J. Virol. Methods 51,
253.
- Lareu, R.R. et al. (1997) J. Virol. Methods 64,
11.
- Marshall, D.J. et al. (1997) J. Clin. Microbiol. 35,
3156.
- Delwart, E.L. et al. (1993) Science 262, 1257.
Epicentre's BESS-G™ Base Reader Kit detects G sequence changes. The
BESS-T&G™ Base Reader Kit combines both the BESS-T and BESS-G technologies
allowing detection of 100% of all mutations.
|
