Clinical Chemistry 45: 1938-1943, 1999;
(Clinical Chemistry. 1999;45:1938-1943.)
© 1999 American Association for Clinical Chemistry, Inc.
Plastic Microchip Electrophoresis for Analysis of PCR Products of Hepatitis C Virus
Yu-Hung Chen1,
Wei-Chang Wang1,
Kung-Chia Young2,
Ting-Tsung Chang3 and
Shu-Hui Chen1,a
1
Department of Chemistry, National Cheng Kung University, Tainan 701, Taiwan.
Departments of
2
Medical Technology and
3
Internal Medicine, National Cheng Kung University
Hospital, Tainan 701, Taiwan.
a Author for correspondence. Fax 886-6-2740552; e-mail shchen{at}mail.ncku.edu.tw
 |
Abstract
|
|---|
Background: Electrophoresis on polymeric rather than glass
microstructures is a promising separation method for analytical
chemistry. Assays on such devices need to be explored to allow
assessment of their utility for the clinical laboratory.
Methods: We compared capillary and plastic microchip
electrophoresis for clinical post-PCR analysis of hepatitis C virus
(HCV). For capillary electrophoresis (CE), we used a separation medium
composed of 10 g/L hydroxypropyl methyl cellulose in
Tris-borate-EDTA buffer and 10 µmol/L intercalating dye. For
microchip electrophoresis, the HCV assay established on the fused
silica tubing was transferred to the untreated polymethylmethacrylate
microchip with minimum modifications.
Results: CE resolved the 145-bp amplicon of HCV in 15 min. The
confidence interval of the migration time was <3.2%. The same HCV
amplicon was resolved by microchip electrophoresis in <1.5 min with
the confidence interval of the migration time <1.3%.
Conclusion: The polymer microchip, with advantages that include
fast processing time, simple operation, and disposable use, holds great
potential for clinical analysis.
|
Introduction
|
|---|
Microchip electrophoresis is emerging as a highly promising method
for rapid and sensitive analysis that is potentially well suited for
clinical analysis. Whereas most of the reports on microfabricated
electrophoretic devices have utilized glass or silica as substrates
(1)(2)(3), recent attention has been given to the use of
polymeric microstructures. The polymer substrates are viewed as
promising alternatives for the production of microfluidic systems
(4)(5) because they are less expensive and
easier to manipulate than silica-based substrates. Moreover, the
inherent neutral hydrophilic nature of the polymer substrate allows
direct use of the channel for clinical analysis of biomolecules without
the need of surface modifications to reduce wall adsorption
(6). However, in the past 15–20 years, the use of glass
tubing such as fused silica has become a well-established technique in
capillary electrophoresis
(CE).1
The deprotonated silanol groups on the glass surface provide
satisfactory electroosmotic flow for the fluid delivery and separation.
Although polymer-based capillary tubing, such as polypropylene
(7)(8)(9), polytetrafluoroethylene (10), and
hydrophilic polymethylmethacrylate (PMMA) (6), has been
investigated for CE separations, experience with polymer-based
microchip CE is limited. Analytical assays on these devices need to be
explored before their usefulness can be further addressed.
One of the major clinical applications that uses electrophoresis is
associated with PCR (11). PCR enzymatically generates
millions or billions of exact copies, thereby making genetic analysis
of small samples a relatively simple process. It is widely used in
molecular biology, with direct applications in the field of medical
diagnosis. Conventionally, slab gel electrophoresis has been used for
the analysis of PCR products for sizing, mutations, or polymorphisms,
but the technique is time-consuming, labor-intensive, and
nonquantitative. Because of the superior separation efficiency and
speed in an automated format, CE is rapidly becoming an important tool
for PCR analysis. However, as the electrophoresis technique advances to
microchip devices, the use of microchip electrophoresis for DNA
analysis is emerging as a promising method. The PCR-CE chip has also
been microfabricated and investigated for some model reactions
(12)(13)(14).
In this study, DNA sizing of the hepatitis C virus (HCV) amplicon was
attempted by CE with fused-silica tubing as well as by microchip
electrophoresis on PMMA substrate. Initial results of the
CE/laser-induced fluorescence method for the detection of
HCV-specific reverse transcription followed by one-stage PCR
amplification have been described briefly in the literature
(15). This method has also been validated and applied to
diagnostic detection of 152 HCV amplicons (16).
Interinstrument and intercapillary reproducibility (CVs) of the
migration time has ranged from 0.319% to 1.833%. Moreover, a
complete agreement between the CE/laser-induced fluorescence and
agarose gel electrophoresis/Southern blotting methods for product
detection has been reported (16). A two-stage PCR
amplification that involves two pairs of primers deduced from the 5'
noncoding region of the HCV genome (17) was adopted in the
present study. This two-stage amplification is reported to be 10-fold
more sensitive than the one-stage assay (17) and has also
been adopted for routine clinical diagnoses (18).
|
Materials and Methods
|
|---|
chemicals and reagents
Tris was purchased from Fluka, EDTA from Sigma, and
boric acid from Janssen Chimica. Hydroxypropyl methyl cellulose (HPMC)
with a viscosity range of 0.08-0.12 Pa · s (20 g/L in
H2O) was from Aldrich. DNA fragment
calibrator 
X-174-RF DNA digested by HaeIII was
from Amersham-Pharmacia Biotech. The intercalating dyes were ethidium
bromide (EtBr) from Sigma and TOPRO-3 from Molecular Probes. All
reagents were of the highest grade available. CE water was deionized
distilled water filtered through a Barnstead E-pure system. The
resistance of the water was >18.0 M
/cm3.
pcr amplification
The serum of patients with HCV-related disease and hepatitis B
surface antigen negative was collected from the hospital of
National Cheng Kung University. The nucleic acid was then extracted
from the serum, and cDNA was synthesized by reverse transcription. DNA
amplification of HCV cDNA was performed by a two-stage PCR with two
pairs of primers deduced from the 5' noncoding region
(17)(18). The PCR products were analyzed by
agarose gel electrophoresis as well as capillary and microchip
electrophoresis. The protocols for the PCR and agarose gel
electrophoresis methods have been described by Liou et al.
(18).
ce instrumentation
The experiments were performed using Beckman P/ACE System 5500
equipped with a ultraviolet absorbance detector (Beckman
Instruments). The inner walls of fused-silica capillaries (i.d., 50
µm; o.d., 375 µm; effective length, 40 cm; total length, 47 cm;
Polymicro Technologies) were covalently bound with a non-cross-linked
polyacrylamide according to procedures described elsewhere
(19). Unless specified, samples were injected by a stream of
nitrogen gas at 3447.38 Pa (0.5 psi) for 20 and 10 s for the DNA
marker and the HCV amplicon, respectively. Separations were carried out
in the reversed polarity mode (-12 kV at the injector end; -255
V/cm). The detection wavelength was set at 254 nm throughout the
experiment.
microchip system
A schematic diagram of the device configuration is shown in Fig. 1
. The channels were fabricated on PMMA plexiglass pieces ~2 cm
in width, 10 cm in length, and 2.0 mm thick by a wire-imprinting
method. The imprinting method was adopted from Martynova et al.
(20) with some modifications (21). The resulting
channels had a rounded shape and were 75 µm in depth (21).
Two identical power suppliers (CZE 1000R; Spellman) were utilized to
furnish the loading and separation voltages, respectively, and the
power switching was controlled by a program written in LabView
(National Instruments) running on a Pentium 75 MHz computer
(22). The sample loading was performed by applying -750 V
(-150 V/cm) to the buffer channel (between reservoirs III and IV) for
0.15, with no voltage applied to the separation channel (between
reservoirs I and II). The amplicons were diluted (threefold) with
deionized water, and no desalting step was performed before the
injection. [Note: the mismatch in ionic strengths between the sample
and the buffer is likely to cause an inconsistency in the actual amount
injected (23)]. Both the dilution and the desalting steps
serve to reduce the ionic strength of the sample, and the desalting
step may provide better sensitivity. For the separation, -1.0 kV
(-200 V/cm) was applied to the separation channel with no voltage
applied to the buffer channel. The fabricated device was used
directly for chip electrophoresis without further modifications.
View larger version (9K):
[in this window]
[in a new window]
|
Figure 1. Channel configuration of the microchip.
The buffer, analyte, and two waste reservoirs were indicated as
I, III, II, and
IV, respectively.
|
|
Signals were detected on the microchip via laser-induced fluorescence.
The detection system was constructed through modifications of a
commercial reflection microscope (Model BX40; Olympus). Briefly, a
helium-neon laser with a wavelength of 632.8 nm (10 mW, model
LHR-991; Melles Griot) was focused at a position 3 cm downstream from
the cross-section within the channel by use of a x50 (numerical
aperture = 0.5) working distance objective. Fluorescence was
collected by the objective and passed through a dichroic cube with a
band-pass filter, followed by spatial filtering before photomultiplier
detection operated at -725 V (model R928; Hamamatsu). Amplified
photoelectron pulses were converted to an analog signal and acquired by
a commercial interface (model 9524; SISC) running on the same computer
as the voltage switching power supplies.
|
Results
|
|---|
The HCV fragment (145 bp) amplified by the two-stage method was
shorter than that (308 bp) produced by the one-stage method
(15)(16). Therefore, the CE conditions published
in the literature (15)(16) would not be directly
transferable and would need to be modified to achieve an effective
resolution. A polymer buffer solution is usually prepared to achieve an
effective sizing, and its concentration can be varied to fit a
particular separation. However, the solution viscosity increases as the
polymer concentration increases. Ideally, one would like to choose a
polymer solution with a low viscosity when going to a smaller mesh
size. A few methylcellulose polymers, such as HPMC and hydroxyethyl
cellulose, were investigated at several concentrations for the
separation of the HCV amplicon. We found that HPMC at a concentration
of 10 g/L provided the best resolution with an adequate
viscosity. Moreover, intercalating dyes were added to the buffer
solution in an attempt to increase the resolution
(15)(16). For the HCV amplicon, when EtBr was
added, the peak height increased substantially and was accompanied by a
slight increase in elution time. In addition, the separation was also
attempted at 50 °C to reduce the elution time because the solute
mobility would increase at a higher temperature. The elution time
decreased with a concomitant increase of peak height. However, the
resolution became poorer, and a longer equilibrium time before each
electrophoretic run was required.
As seen in Fig. 2
, the HCV amplicon was easily detected at a migration time of
~13 min when a physical gel solution composed of 10 g/L HPMC and 10
µmol/L EtBr in Tris-borate-EDTA buffer at 23 °C was used.
There were no peaks during this time period in the negative control
(Fig. 2
). Identification of the amplified HCV fragment was further
assessed by a DNA digest marker. Fig. 3
shows that the HCV amplicon 145 bp in size was eluted at a
migration time between those of the 118- and 194-bp marker fragments.
Moreover, analysis of >10 patient samples by both CE and agarose gel
electrophoresis showed 100% correlation. The confidence interval of
the migration time was 0.45% for within-day tests (n = 3) and
1.38% and 3.15% for five samples pooled from different days with the
same and different columns, respectively.
View larger version (14K):
[in this window]
[in a new window]
|
Figure 2. Capillary electropherograms of the amplified HCV products
from serum samples with negative (top electropherogram)
and positive (bottom electropherogram) response.
The separation buffer was composed of 10 g/L HPMC and 10
µmol/L EtBr in Tris-borate-EDTA buffer (100 mmol/L Tris-borate, 5
mmol/L EDTA, pH 8.2). UV, ultraviolet.
|
|
View larger version (15K):
[in this window]
[in a new window]
|
Figure 3. Size confirmation of the amplified HCV products (145 bp)
using CE.
Bottom electropherogram, X-174-RF DNA (total
concentration, 5 x 104 µg/L) digested with
HaeIII; top
electropherogram, HCV amplified products from serum
samples. The CE separation buffer was the same as in Fig. 2
.
UV, ultraviolet.
|
|
Analysis of the same HCV amplicon was further attempted using microchip
electrophoresis fabricated on PMMA substrate. As shown in Fig. 4
, using the same separation medium except that 10 µmol/L EtBr
was replaced by 1 µmol/L TOPRO-3 for fluorescence detection, 10 of 11
DNA fragments could be resolved in <2.5 min on the untreated PMMA
chip. The electropherogram was much simplified because of the lack of
fluorescence signal of some unwanted species. The only unresolved
fragments (271 and 281 bp) could be resolved by increasing the HPMC
concentration at the expense of analysis time. Despite that, the 145-bp
HCV amplicon could be easily identified in <1.5 min under the
separation conditions described for the DNA marker (Fig. 4
). It was
also noticed that the sensitivity of the described method using
microchip electrophoresis (Fig. 4
, HCV amplicon diluted threefold
before injection) was increased compared with that of CE (Figs. 2
and 3
). Tentative comparisons showed that the detection limit of the
microchip method was approximately two orders of magnitude lower than
that of CE method. Fig. 5
shows that consecutive injections and separations were
completed by voltage switching alone, with a relative SD <1.3% for
the migration time. The chip-to-chip variations were also small: One
PMMA chip could perform >100 analyses of the HCV amplicon
without significant changes of the obtained electropherograms.
View larger version (14K):
[in this window]
[in a new window]
|
Figure 4. Microchip electropherograms of the amplified HCV products
(145 bp, diluted threefold; top electropherograms)
together with X-174-RF DNA (total concentration, 5 x
104 µg/L; bottom electropherogram)
digested with HaeIII.
The separation buffer was composed of 10 g/L HPMC and 1 µmol/L
TOPRO-3 in Tris-borate-EDTA buffer (100 mmol/L Tris-borate, 5 mmol/L
EDTA, pH 8.2).
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Figure 5. Consecutive injections and separations of the HCV amplicon
(diluted threefold) by voltage switching.
The separation buffer was the same as in the legend for Fig. 4
.
|
|
|
Discussion
|
|---|
The improved sensitivity with the use of microchip electrophoresis
compared with CE could be partly attributed to the use of fluorescence
detection and the much shorter separation time. The use of
near-infrared dye (TOPRO-3) as the intercalator coupled with a He-Ne
laser provided several advantages compared with other visible
intercalators coupled with an argon ion laser: (a) the
near-infrared region is inherently low in biological interferences and
thus allows reduced background noise and a lower detection limit
compared with those using visible dyes (21)(24);
and (b) the near-infrared-emitting semiconductor laser
diodes, which are inexpensive, small, reliable and compact in size, may
be used for the miniaturization of chip-based devices.
As demonstrated above, the use of the PMMA chip increased the
separation speed by one order of magnitude. No column coating is
needed, which would otherwise require tedious derivatization procedures
or the purchase of a coated fused-silica column. The major concern in
using the wall-derivatized column is that the silyl-oxygen bond is
likely to be washed off because of the reversible hydrolysis, which
leads to a limited column lifetime, especially at high pH values. For
the PMMA microchip, >100 repetitive injections appear to be
acceptable. In addition, mass production of disposable devices should
also be possible because of the low cost of PMMA material. These
features are well suited for clinical analysis, which requires the
handling of a large amount of biological fluids. However, the
disadvantages of the use of PMMA material include low resistance to
organic solvents and high temperatures (>90 °C). Furthermore, the
optical transmittance of PMMA at wavelengths below 250 nm is low. These
disadvantages might hinder the fabrication of a monolithic PCR-CE
device on PMMA substrate because of the requirement for PCR thermal
cycling up to 90 °C. In addition, certain applications, such as the
use of low wavelength optical measurements and nonaqueous buffer
systems, for the separation may also be prohibited. However, creative
fabrication methods or the use of other polymer materials may overcome
these problems. Moreover, many electrophoresis methods have been
developed that incorporate aqueous phases at room temperature together
with optical measurements within an acceptable wavelength range of
PMMA. This study did demonstrate that clinical assays, such as the
sizing and detection of the HCV amplicon, established on silica tubing
using CE could be transferred directly to PMMA chips with tremendous
gains in speed of processing, ease of operation, and use of
disposables.
|
Acknowledgments
|
|---|
We thank the National Science Council of the Republic of China, the
Industrial Technology Research Institute (ITRI) in Taiwan, and the
Engineering College of National Cheng Kung University for financial
support. We also thank Mei-Hwei Young for technical assistance.
|
Footnotes
|
|---|
1 Nonstandard abbreviations: CE, capillary electrophoresis; PMMA, polymethylmethacrylate; HCV, hepatitis C virus; HPMC, hydroxypropyl methyl cellulose; and EtBr, ethidium bromide. 
|
References
|
|---|
-
Harrison DJ, Manz A, Fan ZH, Ludi H, Widmer HM. Capillary electrophoresis and sample injection systems integrated on a planar glass chip. Anal Chem 1992;64:1926-1932.
-
Fan ZH, Harrison DJ. Micromachining of capillary electrophoresis injectors and separators on glass chips and evaluation of flow at capillary intersection. Anal Chem 1994;66:177-184.
-
Effenhauser CS, Bruin GJM, Paulus A, Ehrat M. Integrated capillary electrophoresis on flexible silicone microdevices: analysis of DNA restriction fragments and detection of single DNA molecules on microchips. Anal Chem 1997;69:3451-3457.
-
McCormick RM, Nelson RJ, Alonso-Amigo MG, Benvegnu DJ, Hooper HH. Microchannel electrophoretic separations of DNA in injection-molded plastic substrates. Anal Chem 1997;69:2626-2630.
[Medline]
[Order article via Infotrieve]
-
Ogura M, Agata Y, Watanabe K, McCormick RM, Hamaguchi Y, Aso Y, Mitsuhashi M. RNA chip. quality assessment of RNA by microchannel linear gel electrophoresis in injection-molded plastic chips. Clin Chem 1998;44:2249-2255.
[Abstract/Free Full Text]
-
Lee ML, Chen S. Hydrophilic polymethylmethacrylate hollow fibers for capillary electrophoresis of biomolecules. J Micro Sep 1997;9:57-62.
-
Nielen MWF. Capillary zone electrophoresis using a hollow polypropylene Fiber. J High Res Chromatogr 1993;16:62-64.
-
Liu PZ, Malik A, Kuchar MCJ, Vorkink WP, Lee ML. Polymeric hollow fibers for capillary electrophoresis. J Micro Sep 1993;5:245-253.
-
Liu PZ, Malik A, Kuchar MCJ, Lee ML. Polyacrylamide-modified polypropylene hollow fibers for capillary electrophoresis. J Micro Sep 1994;6:581-589.
-
Mikkers FEP, Everaerts FM, Verheggen TPEM. High-performance zone electrophoresis. J Chromatogr 1979;169:11-20.
[ISI]
-
Baba Y. Analysis of disease-causing genes and DNA-based drugs by capillary electrophoresis towards DNA diagnosis and gene therapy for human diseases. J Chromatogr A 1996;687:271-302.
-
Wilding P, Kricka LJ, Cheng J, Hvichia G, Shoffner MA, Fortina P. Integrated cell isolation and polymerase-chain-reaction analysis using silicon microfilter chambers. Anal Biochem 1998;257:95-100.
[ISI][Medline]
[Order article via Infotrieve]
-
Waters LC, Jacobson SC, Kroutchinina N, Khandurina J, Foote RS, Ramsey J. Microchip device for cell-lysis, multiplex PCR amplification, and electrophoretic sizing. Anal Chem 1998;70:158-162.
[Medline]
[Order article via Infotrieve]
-
Cheng J, Waters LC, Fortina P, Hvichia G, Jacobson SC, Ramsey JM, et al. Degenerate oligonucleotide primed polymerase-chain-reaction and capillary electrophoretic analysis of human DNA on microchip-based devices. Anal Biochem 1998;257:101-106.
[ISI][Medline]
[Order article via Infotrieve]
-
Felmlee TA, Oda RP, Persing DA, Landers JP. Capillary electrophoresis of DNA potential utility for clinical diagnoses. J Chromatogr A 1995;717:127-137.
[Medline]
[Order article via Infotrieve]
-
Pancholi P, Oda RP, Mitchell PS, Landers JP, Persing DH. Diagnostic detection of herpes simplex and hepatitis C viral amplicons by capillary electrophoresis: comparison with southern blot detection. Mol Diagn 1997;2:27-37.
[ISI][Medline]
[Order article via Infotrieve]
-
Okamoto H, Okada S, Sugiyama Y, Tanaka T, Sugai Y, Akahane Y, et al. Detection of hepatitis C virus RNA by a two-stage polymerase chain reaction with two pairs of primers deduced from the 5'-noncoding region. Jpn J Exp Med 1990;60:215-222.
[Medline]
[Order article via Infotrieve]
-
Liou TC, Chang TT, Young KC, Lin XZ, Lin CY, Wu HL. Detection of HCV RNA in saliva, urine, seminal fluid, and ascites. J Med Virol 1992;37:197-202.
[ISI][Medline]
[Order article via Infotrieve]
-
Hjerten S. High-performance electrophoresis elimination of electroendosmosis and solute adsorption. J Chromatogr 1985;347:191-198.
-
Martynova L, Locascio LE, Gaitan M, Kramer GW, Christensen RG, MacCrehan WA. Fabrication of plastic microfluid channels by imprinting methods. Anal Chem 1997;69:4783-4789.
[Medline]
[Order article via Infotrieve]
-
Chen YH, Chen SH, Analysis of DNA fragments by microchip
electrophoresis fabricated on PMMA substrates using wire-imprinting
method. Electrophoresis;in press..
-
Chen SH, Lin YC, Chen YH, Liao HS, Wu CY, Wang SH. Electrophoretic separation of antisense DNA using polymer-solution filled capillary by cross injection. Harrison DJ van den Berg A eds. Micro total analysis systems `98 1998:153-156 Kluwer Academic Publishers Norwell, MA. .
-
Shultz-Lockyear LL, Colyer CL, Fan ZH, Roy KI, Harrison DJ. Effects of injector geometry and sample matrix on injection and sample loading in integrated capillary electrophoresis devices. Electrophoresis 1999;20:529-538.
[ISI][Medline]
[Order article via Infotrieve]
-
Shealy DB, Lohrmann R, Ruth JR, Narayanan N, Sutter SL, Casay GA, et al. Spectral characterization and evaluation of modified near-infrared laser dyes for DNA sequencing. Anal Chem 1995;49:1815-1820.
The following articles in journals at HighWire Press have cited this article:
|
|
|
|
 
S. F.Y. Li and L. J. Kricka
Clinical Analysis by Microchip Capillary Electrophoresis
Clin. Chem.,
January 1, 2006;
52(1):
37 - 45.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. Ferrance, K. Snow, and J. P. Landers
Evaluation of Microchip Electrophoresis as a Molecular Diagnostic Method for Duchenne Muscular Dystrophy
Clin. Chem.,
February 1, 2002;
48(2):
380 - 383.
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
