|
|
||||||||
Articles |
1
Department of Radiobiology, Westfälische Wilhelms-Universität Münster, Robert-Koch-Strasse 43, 48149 Münster, Germany.
a Author for correspondence. Fax 49-251-8355303; e-mail wedemey{at}uni-muenster.de
| Abstract |
|---|
|
|
|---|
Methods: For flow cytometric quantification, PCR products were labeled with both digoxigenin and biotin during amplification. Subsequently, amplicons were simultaneously bound to anti-digoxigenin microparticles and fluorescently labeled with streptavidin-R-phycoerythrin. Fluorescence intensity per bead was determined by flow cytometry. To study this assay, we examined the expression of the p21WAF1/CIP1 gene and the proliferating cell nuclear antigen (PCNA) gene in ultraviolet irradiation-exposed human keratinocytes lacking functional p53.
Results: Fluorescence was linear with 6010 000 pg of PCR product. As little as 0.4 fmol (40 pg of a 163-bp amplicon) of PCR product could be distinguished from background. The between-run CV of the fluorescent signal for 10 ng of p21 cDNA was 12% (n = 10). The fluorescence-template curve was sigmoidal. p21WAF1/CIP1 mRNA was decreased after ultraviolet irradiation of keratinocytes, whereas PCNA mRNA was markedly increased.
Conclusion: The flow cytometric assay permits rapid (25 min) and reproducible identification of changes in mRNA abundance.
| Introduction |
|---|
|
|
|---|
Flow cytometry is a sensitive technique for quantifying fluorescence-labeled particles or cells. Fluorescence can be introduced by target-specific ligands, e.g., antibodies conjugated with a fluorophore. These samples are carried in a flow stream through a laser beam, which excites the fluorescent dyes associated with the cells or particles. Finally, emitted fluorescent light is multiplied and detected by a photomultiplier. Flow cytometry is capable of measuring thousands of particles per second and has an average detection limit of several hundred fluorescent molecules per particle. This technique is widely used to study cellular physiology, especially to analyze cell cycle progression or cells representing particular antigens. In recent years, bead-based flow cytometric methods have been developed to detect PCR products of viral DNA or RNA with high sensitivity (6)(7)(8)(9)(10). However, for routine clinical diagnostics these techniques are too labor- and time-intensive (at least 2 h). Usually, internally digoxigenin-labeled PCR products are hybridized after amplification to biotinylated probes to introduce a second label. Hybrid DNA is then captured using streptavidin-coated beads. Fluorescence is introduced separately by staining the bead-coupled products with anti-digoxigenin antibodies conjugated to a fluorescent dye. Finally, the beads are analyzed by flow cytometry.
In this report we describe a novel flow cytometric assay for more rapid and direct quantification of RT-PCR products in ~25 min without the need for additional hybridization steps and antibody reactions. In contrast to the above techniques, we offer a simple two-step method: both labels are introduced during PCR, using primers labeled with digoxigenin and biotin, respectively. These two labels enable simultaneous binding of PCR products to microparticles coated with anti-digoxigenin antibodies and staining with streptavidin-R-phycoerythrin utilizing the biotin label. Because unincorporated oligonucleotides can decrease fluorescence intensity (6), we also developed a rapid and easy technique to remove these primers before binding/staining.
Using this flow cytometric assay, we studied the expression of the p21WAF1/CIP1 and proliferating cell nuclear antigen (PCNA) genes in a human nontumorigenic keratinocyte cell line (HaCaT) exposed to a suberythemal dose of ultraviolet irradiation (UV-B). Both genes are part of the cellular response to irradiation and are involved in cell cycle arrest and nucleotide excision repair, respectively (11)(12).
| Materials and Methods |
|---|
|
|
|---|
After the medium was decanted, cells were exposed at 37 °C and 5% CO2 to a 0.16 minimal erythema dose (MED) of UV-B. A TL 4W/12 lamp was used, delivering 0.8 J/m2 s at the position of the sample (Philips; main emission, 280370 nm; 1 MED = 200 J/m2 erythema effective radiation for Caucasian skin type; for emission spectrum of the given source, 230 J/m2 corresponds to ~0.16 MED). Control cells were treated in the same manner but not exposed. After irradiation, medium kept at 37 °C was added.
RNA isolation and reverse transcription
Total RNA was isolated from 2 x 106
HaCaT cells using the RNeasy Mini kit (Qiagen). After photometric
quantification, 2 µg of total RNA was used for a 20-µL reverse
transcription reaction containing 50 pmol of
oligo-dT18 primer, 200 U of OMNISCRIPT reverse
transcriptase (Qiagen), 40 units of rRNasin RNase inhibitor (Promega),
dNTPs (final concentration, 500 µmol/L), and buffer as recommended by
the supplier. Samples were incubated at 37 °C for 1 h and
finally heated at 95 °C for 3 min. For PCR, 1 µL of the reverse
transcription sample was used.
PCR
PCRs (volume, 25 µL) were set up using 1 ng of plasmid DNA
(human p21 full-length cDNA clone; clone ID 309881) or 1 µL of
reverse transcription sample, respectively; 25 pmol of each primer
(listed in Table 1
); 1 U of Taq DNA polymerase (Qiagen); 100 µmol of each
nucleotide; and buffers as recommended by the supplier. Cycling
conditions were as follows: 1 cycle at 94 °C for 4 min, and up to 35
cycles at 60 °C for 40 s, 72 °C for 1 min, and 92 °C for
40 s, with final extension at 72 °C for 5 min. Amplifications
were performed in the Autogene II thermocycler (Grant Instruments).
|
Primer removal
To remove unincorporated primers, 1 µL of silica magnetic
particles (Merck) and 30 µL of 8 mol/L NaClO4
were added to 10 µL of PCR product. After incubation at 50 °C for
5 min, the mixture was placed in a magnetic concentrator and the
supernatant was discarded. The PCR products were eluted by resuspension
in 10 µL of H2O and incubation at 50 °C for
5 min, whereas primers remained attached to the particles. After
immobilization of the particles in a magnet holder, the supernatant was
transferred into a fresh tube.
Capturing and staining of PCR products
Before capturing the PCR products to anti-digoxigenin magnetic
particles (Roche Diagnostics), we washed the particles twice in 500
µL of 3x binding buffer (15 mmol/L Tris, pH 7.5, 1.5 mmol/L EDTA,
150 mmol/L NaCl), using a magnetic holder. After washing, 1 µL of
beads was incubated with 1 µL of a purified PCR reaction and 1 µL
of 90 g/L streptavidin-R-phycoerythrin (Sigma) by rotating at room
temperature for 10 min. Finally, the beads were washed with 500 µL of
phosphate-buffered saline containing 0.02 mL/L Tween 20, and
resuspended in 1.5 mL of Tris-EDTA.
Flow cytometric analysis
The labeled beads were analyzed on a PAS III flow cytometer
(Partec) with FloMax instrument software. The excitation light was
provided by a 25 mW argon laser operated at 488 nm. The orange emission
was measured with a 575605 nm bandpass filter. The sensitivity of the
flow cytometer was adjusted using control beads
(FluoroSphere; Dako). The data were displayed as fluorescence
histograms, and the mean value was used to quantify the fluorescence
signals.
| Results |
|---|
|
|
|---|
|
To test the feasibility and sensitivity of this assay, a purified
163-bp p21 PCR product was used. The concentration of the purified
amplification product was determined by measurement of the absorbance
at 260 nm. Two independent dilution series of the PCR product ranging
from 10 ng (0.1 pmol) to 10 pg (0.1 fmol) were generated. Simultaneous
capturing to the magnetic particles and staining with
streptavidin-R-phycoerythrin was performed twice for each dilution. As
negative control for every assay, beads were stained in parallel in the
absence of PCR products, and their mean fluorescence intensity was
defined as background. With regard to their relative fluorescence
signals, a representative population of 10 000 beads was analyzed by
flow cytometry (Fig. 2
). For all dilution series, 40 pg (0.4 fmol) of PCR
products produced a mean fluorescence higher than the fluorescence of
the beads alone, whereas the mean fluorescence of 20 pg (0.2 fmol) of
PCR product could not be clearly discriminated from the mean
fluorescence of the beads. The calculated detection limit was 38 pg
(the mean fluorescence of blank beads plus 2 SD was 1.36 arbitrary
units and the mean fluorescence of 40 pg of PCR product was 1.45
± 0.11). The good reproducibility of the assay is reflected in
small standard deviations of the mean fluorescence values (Fig. 2B
). In
10 independent PCR reactions of the same template (10 ng of p21 cDNA),
the CV was 12%.
|
To determine the PCR kinetics, different amounts of template DNA were
used. After each PCR cycle, products were examined by flow cytometry.
Fig. 3
shows that amplification of 1 ng of p21 cDNA plasmid template
(2.4 x 108 molecules) was detectable after
12 cycles and reached the stationary phase after 21 cycles, whereas the
PCR product of 1 pg of template DNA (2.4 x
105 molecules) was observed after 20 cycles. To
compare the sensitivity of the flow cytometric assay with conventional
methods, amplicons of a 163-bp p21 sequence were generated using
various amounts of p21 cDNA template each in two independent PCR
reactions (25 cycles). After purification, 10 µL of each PCR product
was separated by gel electrophoresis. The photograph of the ethidium
bromide-stained gel served for densitometric quantification. In
parallel, 1 µL of each PCR product was used for the flow cytometric
assay. The densitometric quantification of ethidium bromide-stained
agarose gels permitted the detection of PCR product from 1 pg (2.4
x 105 molecules) of p21 template cDNA, whereas
10-fold less product from 0.1 pg (2.4 x 104
molecules) template was still detectable by flow cytometry (Fig. 4
).
|
|
Because of the competitive binding properties of digoxigenin-labeled
primers and PCR products to the anti-digoxigenin magnetic particles, a
rapid method for removal of unincorporated primers by silica magnetic
particles was developed. Small single-stranded DNA fragments exhibit a
high affinity to these particles, whereas DNA fragments larger than 40
bp can be completely eluted by H2O (Fig. 5
A). The usefulness of this purification was demonstrated by flow
cytometric quantification of purified and nonpurified p21 PCR products
after various cycles. Nonpurified samples revealed substantially lower
fluorescence intensities (Fig. 5B
).
|
Expression of p21 and PCNA in UV-B-exposed
keratinocytes
It previously has been reported that p21 and PCNA are both
up-regulated in primary fibroblasts after UV irradiation
(14). However, UV-exposed cancer cell lines showed a
significant decrease of p21 expression (15). To determine
the time-dependent expression of p21 and PCNA in p53-deficient
nontumorigenic human keratinocytes (HaCaT cell line) after
UV-B exposure, we performed a quantitative RT-PCR together with the
flow cytometric assay. After exposure of HaCaT cells to 0.16 MED of
UV-B, an accumulation in the S-phase of the cell cycle occurred
(16), in contrast to a G1 block
observed in irradiated primary fibroblast cells (14). The
total RNA from cells harvested at various times after UV irradiation
was isolated to analyze the physiological alterations of the mRNA
abundance. The RNA integrity was verified by gel electrophoresis.
Because RNA extraction and reverse transcription are the most common
sources of variability in quantitative RT-PCR (3),
endogenous ß-actin mRNA was used as a
calibrator.
After reverse transcription, PCR was performed using 5'-modified
primers specific for ß-actin, p21, and PCNA, respectively. To
identify DNA contamination, primers localized in two different exons
were used, which enabled gel electrophoretic size distribution of
products derived from genomic DNA and mRNA (Table 1
). Gel
electrophoretic evaluation of amplification products of each gene was
performed after 15, 20, and 25 PCR cycles, respectively, to determine
the exponential amplification phase. Because the amount of
amplification product accumulating during the exponential phase is
proportional to the quantity of the initial target sequence
(17), RT-PCR of ß-actin was carried out for 15 cycles,
PCNA for 20 cycles, and p21 for 25 cycles. Two independent PCR
reactions were performed per gene and incubation time after
irradiation. After purification, two samples of each amplification
product were analyzed independently by flow cytometry.
Although p21 expression was slightly increased 1 h after
irradiation, a significant decrease of expression was observed between
3 and 7 h after treatment (Fig. 6
). In contrast, the expression of PCNA increased linearly and
the expression of ß-actin remained almost constant throughout
the observation period.
|
| Discussion |
|---|
|
|
|---|
Internal standards have been proposed to overcome the high tube-to-tube variability of the RT-PCR (2). This, however, requires multiple reverse transcription and amplification of each sample at several concentrations of a competitor RNA, which greatly increases costs and labor input and is therefore impractical for clinical studies on a large scale. Although competitive quantitative RT-PCR is the method of choice for quantification of mRNA, the measurement of relative changes of gene expression frequently is sufficient. The flow cytometric assay is also capable of competitive quantification, and several PCR products can be assessed in a single reaction, e.g., amplicons of target RNA and a heterologous RNA calibrator of known concentration. For this purpose, each amplification product has to carry a different modification for the simultaneous binding to certain bead populations differing in size and binding specificity. In the flow cytometer, these bead populations can be readily distinguished by size to enable separate quantification of each PCR product. The easy-to-use procedure of the presented flow cytometric assay in combination with magnetic particle separation facilitates automation for high-throughput analyses of mRNA abundance.
Flow cytometry is an important, widely used diagnostic tool for
sensitive quantification of cells, antigens, DNA, and molecular
interactions (18)(19). PCR product
quantification by flow cytometry previously has been shown for viral
sequences (6)(7)(8)(9)(10). Although these multiple-step procedures
permit high sensitivity, they are too time-consuming. In comparison,
the flow cytometric assay described here does not require hybridization
steps and antibody reactions but does offer high sensitivity. Although
the multiple incorporation of labeled nucleotides yielded a
high fluorescence intensity in former assays, quantification was more
difficult. Therefore, we used a biotinylated primer to bind exactly one
phycoerythrin molecule per amplicon. This provides a strong
proportionality of bound amplicons and fluorescence intensity. One
single fluorescent dye per PCR product is sufficient for high
sensitivity over a broad range of template DNA concentrations (
Figs. 24
).
In contrast to previously described methods, we used anti-digoxigenin magnetic particles, which provide binding capacities and kinetics similar to those of streptavidin-coated particles. However, the capture of PCR products using anti-digoxigenin magnetic particles allows simultaneous fluorescence labeling using streptavidin-R-phycoerythrin, which provides more rapid binding kinetics than antibodies.
The presence of unincorporated digoxigenin-labeled primers markedly
decreases fluorescence intensity and thus lowers sensitivity (Fig. 5B
).
Therefore, we developed a rapid and inexpensive procedure that uses
silica magnetic particles to remove unreacted primers before amplicon
binding to the beads.
The flow cytometric assay was applied for the analysis of p21 and PCNA expression in UV-B-irradiated immortalized keratinocytes lacking functional p53 (13)(20). A reduction of p21 mRNA in response to UV irradiation was observed. Wang et al. (15) also showed a down-regulation of p21 in various human cancer cell lines, regardless of the functional status of p53. They suggested that the failure of UV to induce p21 is linked to immortalization and transformation because primary mouse and human cells, including keratinocytes, revealed an up-regulation of p21 after UV exposure (14)(21).
HaCaT cells were recently demonstrated to be highly susceptible to UV-induced apoptosis, unlike normal keratinocytes (22). Moreover, p16 is absent in HaCaT cells because of hypermethylation of its promoter (22). Because cyclin-dependent kinase inhibitors, such as p21 and p16, play a key role in protecting cells from apoptosis, we suggest a link between immortalization, the altered regulation of apoptosis, and the down-regulation of cyclin-dependent kinase inhibitors in HaCaT cells.
In contrast to p21, the PCNA mRNA was increased after UV exposure of HaCaT cells. PCNA is required for both DNA replication and nucleotide excision repair. Moreover, it was suggested that the p53/p21 signal transduction pathway is important for the regulation of PCNA expression following irradiation (23). However, in agreement with Chang et al. (24), our results demonstrate that UV induction of PCNA is independent of p53 and p21.
With increasing knowledge of gene sequences, the determination of disease-related expression patterns of mRNA has become clinically relevant. Cancer, for example, arises from multiple changes in the cellular phenotype that are not restricted to mutated genes. A mutation in a gene may affect the expression of a series of genes that are below it in the regulatory cascade (25). To confirm expression changes detected with the new high-throughput technologies, such as hybridization of microarrays (26), rapid, sensitive, and reliable assays for quantification of mRNA abundance have to be developed. The flow cytometric method described here fulfils these criteria.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
The following articles in journals at HighWire Press have cited this article:
![]() |
T. Hata, H. Yamamoto, C. Y. Ngan, M. Koi, A. Takagi, B. Damdinsuren, M. Yasui, Y. Fujie, T. Matsuzaki, H. Hemmi, et al. Role of p21waf1/cip1 in effects of oxaliplatin in colorectal cancer cells Mol. Cancer Ther., October 1, 2005; 4(10): 1585 - 1594. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Wedemeyer, T. Potter, S. Wetzlich, and W. Gohde Flow Cytometric Quantification of Competitive Reverse Transcription-PCR Products Clin. Chem., September 1, 2002; 48(9): 1398 - 1405. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |