Clinical Chemistry AACC Online Job Center
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 49: 415-424, 2003; 10.1373/49.3.415
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gale, J. M.
Right arrow Articles by Conia, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gale, J. M.
Right arrow Articles by Conia, J.
Related Collections
Right arrow Molecular Diagnostics and Genetics
Right arrow Cancer Diagnostics (since 2002)
(Clinical Chemistry. 2003;49:415-424.)
© 2003 American Association for Clinical Chemistry, Inc.

Application of Optical Trapping for Cells Grown on Plates: Optimization of PCR and Fidelity of DNA Sequencing of p53 Gene from a Single Cell

James M. Gale1a, Christopher P. Romero1, Gregory B. Tafoya1 and Jérôme Conia2

1 Department of Cell Biology and Physiology, Steve Schiff Center for Skin Cancer University of New Mexico Health Sciences Center, Albuquerque, NM 87131.

2 Cell Robotics International, Inc., 2715 Broadbent Pkwy NE, Albuquerque, NM 87107.

aAddress correspondence to this author at: Department of Cell Biology and Physiology, Steve Schiff Center for Skin Cancer, University of New Mexico Health Sciences Center, Room 149 BMSB, 915 Camino de Salud NE, Albuquerque, NM 87131. Fax 505-272-9105; e-mail jgale{at}salud.unm.edu.


   Abstract
Top
 Abstract
Introduction
 Material and Methods
 Results and Discussion
 References
 
Background: Optical trapping has traditionally been used to visually select and isolate nonadherent cells grown in suspension because cells grown in monolayers will rapidly reattach to surfaces if suspended in solution. We explored methods to slow cell reattachment that are also compatible with high-fidelity PCR.

Methods: Using HeLa cells grown on plates and suspended after trypsinization, we measured the efficiency of capture by retention and movement of the cell by the laser. Success for removing a captured cell by pipette was determined by PCR amplification of the 5S rRNA gene. After optimizing PCR amplification of a 2049-bp region of the p53 gene, we determined PCR fidelity by DNA sequencing.

Results: Addition of bovine serum albumin to suspended cells slowed reattachment from seconds to minutes and allowed efficient trapping. The success rate of removing a cell from the trap by pipette to a PCR tube was 91.5%. The 5S PCR assay also revealed that DNA and RNA that copurify with polymerases could give false-positive results. Sequence analysis of four clones derived from a single cell showed only three polymerase errors in 7200 bp of sequence read and revealed difficulties in reading the correct number in a run of 16 A:T. Comparison of the HeLa and wild-type human sequences revealed several previously unreported base differences and an (A:T)n length polymorphism in p53 introns.

Conclusions: These results represent the first use of optical trapping on adherent cells and demonstrate the high accuracy of DNA sequencing that can be achieved from a single cell.


   Introduction
Top
 Abstract
 Introduction
Material and Methods
 Results and Discussion
 References
 
The manipulation and molecular analysis of single cells have many applications in research and medicine. Numerous reports in the literature describe the selective genetic analysis of a very small number of cells, including single cells. Types of analysis currently performed with single cells include PCR (1)(2)(3), reverse transcription-PCR (4)(5), gene expression and expression profiling (6)(7), gene amplification (8), cDNA libraries(9), DNA damage and repair (10)(11)(12), mRNA differential display(13), apoptosis (14), enzymatic assays(15), mutation and loss of heterozygosity (16)(17), comparative genomic hybridization (18)(19), cell signaling (20), and peptide profiling by matrix-assisted laser desorption/ionization mass spectrometry (21). DNA typing studies have been reported for diploid cells, single hairs, and sperm cells (22)(23)(24). Single-cell genotyping has been used for several diseases, including ß-thalassemia (25), retinoblastoma (26), cystic fibrosis(27), and multiple genetic diagnoses using multiplex PCR (28). These studies have generated advances in various aspects of cell biology, such as in the analysis of cell-specific gene expression and genetic variations, the characterization of infectious agents for disease mapping, and the forensic identification of individuals (29).

With the advent of scientific investigations using improved detection capabilities down to the molecular level, focus has been on elaborating methods for the handling and preparation of single cells. However, this task is not easy, and it is often addressed empirically. A routine approach to single cell isolation is the preparation of serial dilutions of a cell culture suspension. By lowering the initial cell density, an investigator can recover single cells in the culture medium per unit of volume. This method, however, lacks singular accuracy for the identification and selection of specific cells from a heterogeneous suspension. The use of immunomagnetic microparticles or flow cytometry has also been useful in cell-sorting procedures. With careful selection of sorting parameters or immunotargets, these methods have the ability to select a specific cell type from a mixed population of cells. However, flow sorting requires the availability of a large number of cells and is often blamed for damaging cells during the sorting process. Furthermore, none of these methods allows for the continuous observation of the specimen during the purification procedure, nor are they adequate for routine recovery at the single cell level. It is also possible to isolate a single cell from a heterogeneous suspension by use of mechanical micromanipulators. However, mechanical manipulations are tedious to perform and typically have a low success rate.

In the specific case where thin tissue sections are available, laser-capture microdissection under direct microscopic visualization is a powerful approach for isolating cells. In this technique, a transparent thermoplastic film is applied to the surface of a tissue section. A laser beam is then used to activate the adhesion property of the film directly above the cells of interest (30). The target cells are thus collected by attachment to the plastic film. This technique is capable of isolating single cells without contamination when the sections are very thin (~ 1 cell thickness) (31).

Another approach for the visualization, selection, and sorting of single cells is optical trapping. In the same manner that microscopic particles in suspension can be displaced by the forces exerted by a beam of photons (32)(33)(34), a highly focused beam of near-infrared light can be used for the in vitro manipulation of cells. The principle of operation for optical trapping has been well described; an extensive bibliography in PDF format can be found on the Cell Robotics International web site (http://www.cellrobotics.com/prod.html). In practice, a laser beam introduced into the optical path of the microscope and focused through a large numerical aperture objective lens converges to form an optical trap at the focal point for the laser. The cell, which is transparent to the incident laser beam and whose refractive index exceeds that of the surrounding medium, is drawn toward the brightest point (highest photon density). The laser photons that are refracted by the cell produce miniature pressures on it, forming a gradient force in three dimensions sufficient to make the cell levitate. Displacement of the captured cell relative to the rest of the sample is then achieved by either moving the microscope stage or by displacing the laser beam itself (35). Cell viability is maintained because at wavelengths from 830 to 1064 nm few chromophores extensively absorb the photon energy and the penetration depth is maximal (36). Because the laser light is absorbed only slightly or not at all by biological samples, optical trapping remains mostly nondamaging for living specimens (37). Moderate temperature increases within a few degrees centigrade have been observed within the confinement of the trap (38)(39).

Although optical trapping of single cells may have negligible biological effects, it is highly dependent on the wavelength of laser light and the dosage of irradiation to which cells are exposed (40)(41)(42). It has also been suggested that dual-photon absorption and ultraviolet-like energy deposition may have damaging effects on genetic material and present potential for mutation (43).

The advantages of optical trapping are that cells are isolated alive and intact, cells are visualized under the microscope before trapping (ensuring that only the cell type of interest is captured), isolation can be done quickly and efficiently, and the procedure requires very few cells in the sample. Because cells are visualized before trapping, this method allows for selection of specific cell types when a mixed population of cell types are present in the culture (44). In addition to selection of specific cell types from blood (the earliest use of optical trapping), the method has also been used to select specific vertebrate retinal cell types (45) and to select for cells interacting with natural killer cells (46).

In the present study, we evaluated the selection and isolation efficiency of optical trapping for cells grown on solid supports. Because optical trapping requires that a cell stay in suspension several minutes to allow capture and movement to a cell-free area of the capture chamber, the utility of this technique for cells grown in culture on plates was uncertain because of the typical rapid reattachment kinetics after cell suspension. Using a simple procedure to slow reattachment that is compatible with subsequent molecular analysis, we measured the success of single cell sorting by optical trapping by the effective collection and PCR amplification of genetic sequences. In addition, we demonstrated high-fidelity amplification of the single-copy gene p53 (in one round of PCR) for mutation analysis in exons and introns 5–9 from a single cell. Our results indicate that single cell sorting by optical trapping may be applied to cell lines that are usually adherent to solid supports (cell culture plates), dramatically expanding the scope of scientific investigations oriented toward the recovery of rare or abnormal cells for molecular analysis and further genetic identification.


   Material and Methods
Top
 Abstract
 Introduction
 Material and Methods
Results and Discussion
 References
 
cell lines and culture
HeLa S3 cells (cell line CCL-2.2; American Tissue Cell Culture Collection) were grown with complete DMEM supplemented with 100 mL/L fetal bovine serum (Life Technologies) on 100-mm cell culture plates (Falcon®; Becton Dickinson Labware) at 37 °C in 5% CO2. The complete medium contained 1x essential and nonessential amino acids and vitamins, 1x L-glutamine, and 1x penicillin-streptomycin (Sigma). For optical trapping experiments, one plate of confluent cells was harvested by first removing the medium and rinsing briefly with 5 mL of phosphate-buffered saline (PBS).1 Cells were then treated with 2 mL of 1.25 g/L trypsin (Sigma) until the cells began to detach from culture plate. Cells were then scrapped from the plate and washed three times with 10 mL of PBS in 15-mL conical tubes (Falcon; Becton Dickinson Labware). After the final wash, cells were suspended in 2 mL of PBS or PBS containing bovine serum albumin (BSA) at concentrations between 0.1 and 1.0 g/L and either used the same day or stored at 4 °C up to 2 days for optical trapping experiments.

laser-based microscope
Optical trapping was performed using a LaserTweezers® Workstation 1000/980 (Cell Robotics, Inc.) mounted on a Zeiss Axiovert 100 inverted microscope (Carl Zeiss Inc.). The incident infrared laser beam was focused using a x100/1.25N.A. Achroplan objective lens (Carl Zeiss). The suspension of cells was held on a glass coverslip placed on the microscope stage, above the objective lens. The coverslip was ~170 µm thick (no. 1 1/2), making the chamber compatible for use with the large numerical aperture objective lens necessary for optical trapping applications. A droplet of immersion oil was used as an optical link between the lens and the glass coverslip. The LaserTweezers Workstation included an xy, automated, motorized stage for accurate positioning of the target specimen relative to the laser beam and for displacement. In addition, although the laser module was firmly attached to the microscope body, the laser beam could be precisely aligned relative to the optical path of the microscope by use of a beam steering mechanism that works with a pair of quasi-parallel mirrors, gold-coated for optimized infrared reflectance (>98%). z-Axis motion was controlled through a motorized focus adjuster. Parfocality adjustment was through a Keplerian beam expander. The laser source was a semiconductor, continuous wave, 1 W, single longitudinal and single transversal mode, MOPA laser diode (collimated beam, 3 mm in diameter) with peak emission centered at the nondamaging wavelength of 985 ± 10 nm, with an emission bandwidth of 2 nm for ~80% relative intensity (SDL). A two-section monolithic laser structure was used. The first section (Master Oscillator) generated the single-mode and single-frequency characteristics; the second section (Power Amplifier) increased the output power. The output power of the laser source could be adjusted in 1% increments from 0 to 1000 mW. All system controls, including laser power, motorized stage, and focus adjuster, were controlled by custom LabVIEW®-based software (Cell Robotics) on a Pentium® 166 MHz computer (Dimension XPS M166S; DELL). The computer mouse was used for system operation. Command functions were with on-screen software menus with dedicated virtual tools.

isolation of single cells
Cell observation and identification were performed before and simultaneously with optical trapping. Cells were monitored with a standard monochrome PAL format camera with pixel resolution of 768 x 494 (Sony XC-75CE) that was mounted on the microscope camera port and connected to a video board and frame grabber (Bandit; Coreco Inc.). A live video display of the specimen was shown on the computer’s color monitor (multiscan 17-inch, high resolution; DELL).

To capture a single cell, 5–10 µL of HeLa cells suspended in PBS (or PBS containing BSA) were placed in a cell capture chamber (Fig. 1 ). The cell capture chambers used were made from double-sided adhesive elastomer (Cell Robotics) that was placed on standard microscope cover glasses (24 x 50 mm; 170-µm thick; Fisher Scientific). The compartment of the cell chamber in which the HeLa cells was placed was screened under the microscope for single cells that were still free floating (not attached to the cover glass or aggregating to other cells). When a single cell was found, the laser trap was activated and the cell was moved away from the bulk of the cells into a channel of the chamber that did not contain any cells. Because the cell was held stationary above the objective lens, movement of the cell was accomplished by a motorized microscope stage. While the cell was being moved, it was monitored through the microscope on the computer screen to make sure the cell stayed in the trap. The microscope stage was capable of speeds sufficient to knock cells out of the trap, so it was important to monitor the cells during movement. On average, microscope stage speeds >100 µm/s would displace the cell from the trap. In cases where a cell fell out of the trap during transit, it was often possible to recapture the cell in the trap and continue moving the cell down the channel. Once a trapped cell was moved away from the bulk of cells to an area in the chamber where no cells were seen on the computer screen (at least 5 mm from cell compartment), the cell was immediately removed by aiming the tip of a standard 20-µL pipetter at the active optical trap and withdrawing 5–15 µL. Cells were placed directly in 0.65-mL thin-walled PCR tubes (Sorenson BioScience) and stored at -20 °C for up to 4 months before PCR analysis.



View larger version (123K):
[in this window]
[in a new window]
  		Figure 1. Cell chambers used for optical trapping.

Two different cell chambers (A and B) were made from double-sided adhesive elastomer and placed on a microscope cover glass. Cells in suspension were placed in the chambers at the locations designated by the arrows.

pcr amplification of the 5S rRNA gene
Initial PCR amplification of the human 5S rRNA gene focused on a 190-bp region containing the 5S rRNA gene previously used in the laboratory (-75 to +113 of the gene; GenBank accession no. X12811). The oligonucleotide primer sequences used for amplification of the 5S rRNA gene were as follows: primer A, 5'-TGGCGGTGTCGGCTGCAATCC-3', and primer B, 5'-CAGCACCCGGTATTCCCAGG-3'. All primers used in the study were designed using a molecular biology software program, GeneRunner Ver. 3.05 (Hastings Software). Primers were synthesized by the Center for Genetics in Medicine of the University of New Mexico’s Department of Biochemistry and Molecular Biology. Subsequent amplification of a genomic region 3' to the 5S rRNA gene used primer C (5'-GCCCAGGCGATTCAATTCAC-3'), primer D (5'-TCGTCGCACCCTTCCAAACC-3'), primer E (5'-TTCTTGGATGAATTGCTTGC-3'), and primer F (5'-TCTTGAGCAGGCCGGGATAG-3'). The schematic locations of these primers on the genomic 5S repeat are shown in Fig. 1AUp . A typical 75-µL PCR reaction contained 200 µM each deoxynucleotide triphosphate, 1.0 U of polymerase, 20 pmol of each primer, reaction buffer (containing MgCl2) supplied with the polymerase, and either HeLa genomic DNA (100 ng) or a single HeLa cell isolated from an optical trap. PCR conditions were 95 °C for 45 s (2 min on first cycle), 57–62 °C for 45 s, and 72 °C for 1.5 min for 25–50 cycles on a PTC-100 thermal cycler (MJ Research). The polymerases tested were AmpliTaq® and AmpliTaqLD® (Perkin-Elmer), Z-TaqTM (TaKaRa/PanVera), Pfu (Stratagene), KlenTaq1TM (Ab Peptides), Hot Tub (Amersham Pharmacia), DyNAzymeTM (Finnzymes), Vent® and Deep Vent® (New England Biolabs), and PWO (Roche Molecular Biochemicals).

Before PCR amplification of single cells, the cell was first treated with proteinase K (PK; Worthington Biochemical) directly in the PCR tube into which the cell was first placed, as described in the Results and Discussion section. In addition, after PK digestion, the volume was adjusted to 25 µL with PBS, and 50 µL of a 1.5x PCR reaction mixture was added that contained all other PCR components. The final PCR reaction conditions for 5S rRNA amplification to determine the cell-capture efficiency from the optical trap used KlenTaq, 40 cycles, and an annealing temperature of 58 °C.

PCR products were separated on 1.5% agarose gels (SeaKem® LE; FMC BioProducts,) in Tris-borate-EDTA buffer containing 1x GelStar® stain (FMC BioProducts) added before the gel was poured. Images of the gels were captured with an AlphaImagerTM 2000 CCD camera system (AlphaInnotech) using a SYBR Green photographic filter (FMC BioProducts).

pcr amplification, cloning, and sequencing of the p53 gene from a single cell
A 2049-bp segment of the human p53 gene, containing exons 5–9, was amplified with use of primer G (5'-GTTTCTTTGCTGCCGTGTTCC-3') and primer H (5'-TGTATCAGGCAAAGTCATAGAACC-3'), corresponding to positions 12972–12992 and 14998–15021 of the p53 genomic sequence, respectively (GenBank accession no. U94788). A schematic of the locations of these primers with respect to exons 5–9 is shown in Fig. 4A . High-fidelity polymerases tested for p53 PCR were Pfu and PfuTurboTM (Stratagene), Pfu (Promega), and PlatinumTM Pfx (Life Technologies). PCR reaction conditions were 95 °C for 45 s (2 min for the first cycle), 58–60 °C for 45 s, and 72 °C for 4.5 min for 25–40 cycles in a PTC-100 thermal cycler. A typical 75-µL PCR reaction contained 1.5 U of polymerase, 200 µM each deoxynucleotide triphosphate, 20 pmol of each primer, reaction buffer (containing MgCl2) supplied with the polymerase, and HeLa genomic DNA (50 ng), human genomic DNA (~50 ng, isolated from skin biopsies), or a single HeLa cell isolated from an optical trap. For Pfx, MgSO4 was added (final concentration was 1 mM, as recommended by the manufacturer), and Enhancer Solution concentrations of 0x, 0.5x, and 1x were tested.



View larger version (71K):
[in this window]
[in a new window]
  		Figure 4. PCR amplification of a p53 segment containing exons 5–9.

(A), schematic representation of the genomic structure of human p53, showing the locations of exons 5–9, primers G and H (used to amplify the region), and primer J (used to sequence exon 7). (B), agarose gel images of p53 amplification from single cells for three different polymerases, performed in duplicate. Pfu turbo and cloned Pfu (cPfu) are from Stratagene, and pPfu denotes the cloned polymerase from Promega. Aliquots of the PCR reaction were sampled every 5 cycles from 25 to 40, shown at the top of the gel. For each polymerase shown, PCR reactions for two different single cells were sampled. The size of the expected p53 band is shown on the right, next to the {lambda}/HindIII + {Phi}X174/HaeIII size markers.

For cloning and sequencing from single cells, PfuTurbo was used with an annealing temperature of 58.5 °C for 35 cycles. In the same manner as the single-cell 5S rRNA gene PCR, after PK digestion, 50 µL of a 1.5x PCR reaction mixture was added that contained all other PCR components. PCR products were isolated from 1.0% agarose gels with use of DEAE membranes (Schleicher & Schuell) as described by the manufacturer. Pfu PCR products were phosphorylated on the 5' end with T4 polynucleotide kinase (New England BioLabs) in the presence of ATP (Sigma), and then ligated to 50 ng of blunt-end EcoRV-digested pBluescript SK+ vector (Stratagene) with T4 DNA Ligase H.C. (Life Technologies) at 16 °C overnight. Ligation reactions were terminated with a 10-min incubation at 65 °C and diluted 1:1 with sterile water. Each ligation product (2 µL) was transformed into XL-II Ultracompetent Sure Cells (Stratagene). Transformed cells were spread on ampicillin plates treated with 5-bromo-4-chloro-3-indolyl ß-D-galactopyranoside/isopropyl ß-D-thiogalactopyranoside (X-gal/IPTG) and incubated overnight at 37 °C. White colonies were selected and incubated overnight (37 °C; 225 rpm shaking air bath) in 3 mL of LB broth containing 50 mg/L ampicillin and 20 g/L glucose.

Plasmid DNA was harvested from bacteria by a lithium chloride-based miniprep method (47). To verify inserts, plasmid DNA was digested with BssHII (Stratagene) and electrophoresed on a 1.2% agarose gel stained with ethidium bromide. Plasmids containing the p53 insert were sequenced from each end with use of primers complementary to the plasmid T3 and T7 sites using the BigDyeTM terminator cycle sequencing reagent set (PE Applied Biosystems). Because p53 exons 5 and 6 are within 400 bp of the 5' end of the clone and exons 8 and 9 are within 500 bp of the 3' end (easily sequenced by the T3 and T7 primers), another primer, J (5'-GGTGGATGGGTAGTAGTATGG-3'), was made to sequence exon 7 (see Fig. 4AUp ). The following conditions were used for cycle sequencing of the plasmid DNA containing the p53 PCR product, which were modified from the conditions recommended by PerkinElmer. For a 10-µL total reaction volume: 1 µL (10 pmol) of primer, 1 µL of Terminator Ready Reaction Mix, 3 µL of reaction buffer (stock reaction buffer: 200 mM Tris-HCl, 5 mM MgCl, pH 8.8), 2 µL (5 µg) of plasmid DNA, and 3 µL of ACS-grade water. Thermocycling conditions were also modified slightly from manufacturer recommendations as follows: 95 °C for 20 s, 50 °C for 15 s, 60 °C for 4 min, 28 cycles total. Cycle sequencing was done on a MJ Research PTC-100 Thermocycler with the hot-top enabled so that mineral oil was not needed to prevent evaporation of the reaction. Analysis of the DNA sequence and chromatograms produced by automated sequencing were done with the GCG software (Wisconsin Package, Ver. 10; Genetics Computer Group).


   Results and Discussion
Top
 Abstract
 Introduction
 Material and Methods
 Results and Discussion
References
 
optimization of cell capture efficiency using optical trapping
The initial goal was to determine whether cells grown in monolayers could be resuspended and efficiently captured by the laser and moved far enough away from the bulk of the cells to be isolated as a single cell for further analysis by PCR. Because suspensions of cells grown in monolayers rapidly reattach to solid supports (as when they are split in culture), we tested several approaches to slow reattachment to the microscope cover glass that supported the cell-capture chamber. Using HeLa cells suspended in PBS, which had been isolated from cell culture by trypsinization, we observed that reattachment occurred within 30 s. Once a cell attached, the optical trapping force was not sufficient to capture and move it. Even when an unattached cell could be quickly captured by the laser, >90% of the time when the cell was being moved it broke free of the trap and attached to the cover glass (despite the ability of the optical trap to lift the cell ~20 µm above the surface after capture using the x100 objective lens). After failed attempts involving siliconizing the cover glass and adjusting the ionic strength of the PBS, we discovered that adding BSA to the PBS (used in both the cell suspension and in the cell-capture chamber) dramatically slowed reattachment. Using the BSA solutions that come with some restriction enzymes, we found that a final concentration between 0.2 and 1.0 g/L allowed 6–8 min before reattachment. This was enough time to capture, move, and isolate two to three cells before a fresh aliquot of cells was needed. Acetylated BSA gave us the same results. A BSA concentration of 0.5 g/L was chosen for all subsequent studies using isolated cells for PCR. One benefit of using BSA is that it would probably not interfere with subsequent PCR reactions because BSA has been shown to prevent PCR inhibition by melanin, hemin, iron, chloride, tannic acids, fulvic acids, feces extracts, and marine water (48)(49).

To isolate a cell from the optical trap, the cell had to be first moved away from the bulk of the cells before being removed from the trap with a pipetter. To move the captured cell to an area free of other cells, two cell-capture chambers designs were tested (Fig. 1Up ). The logic behind chamber B (Fig. 1Up ) was that the cells would be placed in the upper corner of the smaller compartment (10-µL aliquot) and moved to the larger compartment that was filled with PBS containing BSA (no cells) for isolation. Although this design effectively isolated the cell from the rest, it took too long to move the cells that far, allowing cells to evade the trap ~25% of the time during the long transit. With the laser at full power, the maximum speed we could move the microscope stage without washing the cell out of the trap was ~100 µm/s. When no BSA was present, the maximum speed useable was 25–30 µm/s. To reduce the time required to isolate a single cell, we redesigned our cell-capture chamber (Fig. 1Up , chamber A). In this configuration, 10 µL of suspended cells (with BSA) was placed in the upper, round portion of the chamber and the lower channel was filled with PBS containing BSA at the same concentration as the cell suspension. Cells were captured in the reservoir and moved approximately halfway down the lower channel for isolation. While moving the cell, we watched the microscope monitor on the computer to observe whether other cells had started to diffuse down the channel. Using this design, we could pick and isolate two to three cells before the cells in the upper reservoir started to attach. After two to three cells were picked, the entire cell-capture chamber was rinsed with water and refilled with cells and PBS for another round of selection. With this design and a BSA concentration of 0.5 g/L, only 6 cells of 89 evaded the trap during movement (6.7%). With cells smaller that HeLa or a more powerful laser (now available), the stronger optical trapping force would allow higher speeds and better retention of the cell in transit.

Once trapped, the single cell was removed from the chamber by pipette. With practice, the cell could be consistently withdrawn in a 5 µL volume. The cell removed from the trap was placed directly in a 0.65-mL thin-walled PCR tube and stored at -20 °C for future analysis. Aiming of the pipetter was accomplished by holding the pipetter steady with two hands while directly observing the location where the isolated cell was held in the chamber on the microscope stage. During the isolation procedure, the optical trap was still active (laser power on) to minimize the chance that the cell would settle and start attaching to the cover glass if the power was turned off before the cell was pipetted. Although no detectable laser radiation was present at ~20 cm from the frontal of the highly divergent objective lens, safety glasses were worn to prevent exposing an operator’s eyes to potentially harmful laser radiation. Although we did not test turning off the laser power just before capture (as recommended by the manufacture), this procedure may also be just as effective.

pcr assay to determine percentage of cells successfully removed from the optical trap
To determine the percentage of cells that were successfully removed to the PCR tubes, a PCR assay was used to detect cells based on amplification of the 5S rRNA gene. Because the human 5S rRNA gene is present at 1000–2000 copies/cell (50)(51), we hoped that amplification of this target would be robust, even from a single cell. Using primers A and B, schematically shown in Fig. 2A , we began to optimize PCR conditions, using isolated genomic DNA from HeLa cells (not captured single cells), but we quickly discovered that our no-DNA controls gave a PCR product of the expected size along with several other bands with Taq polymerase. After ruling out contamination of the PCR solutions, we concluded that because of the high degree of conservation of ribosomal sequences between species, our 5S ribosomal PCR primers were amplifying product from the low amount of DNA and RNA that copurifies with the polymerase preparations. Similar problems have also been reported during attempts to amplify regions of the 16S ribosomal gene (52)(53). To combat contamination of Taq preparations, methods have been published to reduce this background PCR by pretreating the Taq polymerase solution with DNase I, ultraviolet irradiation, 8-methoxypsoralen, or biphasic extractions (54)(55)(56)(57)(58). In addition, PerkinElmer offers a version of AmpliTaq called AmpliTaq-LD (low DNA), which is further purified to reduce the DNA content in the enzyme preparation. To determine whether Taq-LD or other polymerases showed similar background PCR with no added DNA template, PCR reactions containing 5S primers A and B were tested with Taq, Taq-LD, and eight other polymerases (Fig. 2B ). Aliquots of the PCR reactions were removed at 25, 30, 35, 40, 45, and 50 cycles to determine how efficient the background synthesis was. Interestingly, Taq-LD did not significantly reduce the background 5S synthesis or the number of cycles at which the first 5S band appeared (35 cycles). Six other polymerases also showed 5S synthesis between 30 and 35 cycles of PCR (Z-Taq, Pfu, KlenTaq, Hot Tub, DyNAzyme, and PWO). Only two polymerases, Vent and Deep Vent, showed no background synthesis of a 5S band or other nonspecific bands. However, these polymerases contain a very active 3'->5' exonuclease "proofreading" activity, much higher than those in Pfu or PWO, the only other enzymes shown in Fig. 2B that have this activity. In a previous study in which we compared the kinetics of primer digestion by proofreading polymerases under similar PCR conditions with no added DNA template, we found that both Vent and Deep Vent completely digested the PCR primer after only 8 PCR cycles. In contrast, 60–75% of full-length PCR primer could still be detected after 24 cycles when we used Pfu and PWO (J.M. Gale and G.B. Tafoya, manuscript submitted for publication). Because we were concerned that Vent or Deep Vent would digest the primers faster than they would amplify PCR product under conditions of low amounts of target DNA (expected from single cells), they were not tested in subsequent assay development.



View larger version (93K):
[in this window]
[in a new window]
  		Figure 2. PCR amplification of the 5S rRNA gene.

(A), schematic representation of the genomic structure of the repeat unit containing the 5S rRNA gene, showing the location of the 5S gene, an Alu repeat, and a purine/pyrimidine repeat (Pu/Py). The approximate locations of the different primers used (A–F) are shown above the sequence. (B), no-DNA control PCR reactions. Agarose gel images of PCR reactions that did not contain any input DNA. An aliquot of the PCR reaction was sampled every 5 cycles from 25 to 50 for 10 different polymerases, shown at the top of each gel image. The primers used were A and B, and the size of the expected 190-bp 5S product is indicated by the arrow on the left.

It was clear from these results that new primers needed to be tested. Four primers, C–F, were designed to the 3' region of the human 5S tandem repeat (51), shown schematically in Fig. 2AUp . In designing these primers, we first screened candidates against the Escherichia coli sequence to minimize homology. After testing the four possible primer combinations with several polymerases (data not shown), we found that primers C and E with KlenTaq gave the most robust, reproducible amplification of the 5S gene from low amounts of HeLa genomic DNA with no amplification detected from the no-DNA controls.

Before the 5S PCR assay could be tested with isolated single cells, the conditions for digesting the cell with PK needed to be determined. Conditions for the PK treatment before single-cell PCR vary widely in the literature, with PK at concentrations between 100 and 200 mg/L for 0.5–10 h at 40–50 °C, followed by 10–30 min at 80–99 °C to inactivate the PK (17)(19)(59). To directly test the amount of PK needed, we tested three concentrations of PK with the 5S PCR assay. The digestion conditions chosen were 50 °C for 1 h and 98 °C for 30 min. The results, shown in Fig. 3A , indicate that all three PK concentrations worked equally well and that faint bands of the 5S PCR product could be seen when no PK was added. Reassuringly, control PCR reactions with no cells (PBS + BSA only) did not synthesize a 5S product (Fig. 3 ). The lowest concentration of PK used (1.7 mg/L; 2µL of PK at 0.01 g/L added to ~10 µL of PBS + BSA containing the cell) was ~100-fold less than the concentration commonly used, although the highest concentration (~170 mg/L) did not show any inhibition of the 5S PCR.



View larger version (54K):
[in this window]
[in a new window]
  		Figure 3. PCR assay for single cells.

(A), effect of different concentrations of PK digest on subsequent amplification of a region of the 5S rRNA repeat. Numbers above each section of the agarose gel image indicate the final concentration of PK used (g/L), including no PK added (No). + indicates the PCR reaction from a different single cell; n, controls with no cells. The primers used in this PCR reaction were C and E (Fig. 2AUp ), and the expected size of the 5S band is shown at the left. (B), example of one PCR assay for 10 cell-captures (+) from the optical trap showing PCR amplification of the 5S band for 7 of the captures. Two no-cell controls (n) are also included. The DNA size markers used were {Phi}X174/HaeIII.

Using our PK protocol and the 5S PCR assay, we determined the percentage of successful cell captures from the optical trap. An example of one test, shown in Fig. 3BUp , showed that 7 of 10 captures tested actually did contain a cell. Overall, we found that 91.5% (75 of 82) of captures contained a cell when BSA was used. These results show that with the addition of BSA, optical trapping is a practical and efficient method to capture and isolate single cells grown on solid supports.

We are confident that only one cell is being isolated after trapping because (a) only a single cell can be seen caught in the trap under the microscope, (b) the cell is moved a relatively large distance away from the pool of cells in the upper part of the chamber, and (c) the region of the chamber to which the cell is moved can be viewed under lower power to verify that no other cells are in the vicinity before isolation by pipette. Furthermore, although no broken or lysed cells were seen in the chamber in any of our experiments, there is a potential for free DNA diffusion if any cell lysis has occurred. To control for this possibility (and cells in the vicinity of the trap), we took a second pipette isolation from the trap after the cell had been removed. No PCR product of either 5S or p53 was seen for any of these controls (data not shown).

amplification of exons 5–9 of the p53 gene from single cells for mutation analysis
Despite the initial problems in trying to amplify a ribosomal gene target, it was apparent that the multicopy 5S rRNA gene could be easily amplified from a single cell in one round of PCR. However, one of the more difficult challenges of single-cell PCR is the ability to amplify single-copy genes, with high fidelity, for subsequent analysis. As a result of our interest in quantitative mutation assays for ultraviolet irradiation-induced mutations in the p53 gene, we had already established PCR conditions to amplify a 2049-bp genomic fragment containing exons 5–9 and adjoining introns. A schematic of the p53 PCR is shown in Fig. 4AUp . Using primers G and H and our PK protocol for the single cells, we tested the high-fidelity polymerases, Pfu and Pfx. To determine the minimum number of PCR cycles necessary to amplify the p53 fragment (and keep the polymerase errors to a minimum), we sampled the PCR reactions every 5 cycles from 25 to 40. Shown in Fig. 4BUp are the results using cloned Pfu (Stratagene and Promega) and PfuTurbo (Stratagene). The results show that the p53 PCR product could be visualized in as few as 30 cycles with the highly sensitive GelStar stain (BioWhittaker Molecular Applications). Both PfuTurbo and Promega Pfu gave good products. No PCR products were formed in the no-cell controls for any polymerase tested (data not shown). The polymerase Pfx was also able to amplify the p53 fragment, but this PCR was not reproducible in our hands despite several attempts using various annealing temperatures and concentrations of the enhancer solution supplied with the polymerase.

The ability to amplify a single-copy gene from a single cell in as few as 30–35 PCR cycles suggested that polymerase errors might be low enough to obtain accurate sequencing for mutation analysis. The error rate for Pfu is reported to be ~1.3 x 10-6 mutations/bp/duplication (60). For 35 cycles, one Pfu error would occur every 22 000 bp synthesized, or for the 2049-bp p53 clone, 1 clone of every 11 should contain an error. However, this error rate is based on typical PCR conditions containing hundreds of nanograms of genomic DNA (100 ng of human DNA represents ~13 000 cells, based on ~7.2 pg DNA/diploid cell). A recent report indicated that when 50 copies or less of template were amplified with Taq polymerase, a significantly higher polymerase error rate was seen (61). Furthermore, the authors cautioned against using direct sequencing to determine mutations with small amounts of template. Although the mechanism for these sequence-specific PCR errors is unknown, the authors speculated that if Taq polymerase stalled because of secondary structure in the DNA template, the ability of Taq to incorporate a base on the end of a fragment, in a nontemplate manner, might produce an apparent mutation. Taq commonly adds a base to the ends of PCR fragments (usually adenine), a property that has been exploited in the T-A cloning approach using t-tailed vectors (62)(63). This hypothesis was further supported by parallel analysis with Pfu polymerase, which does not have nontemplate "extendase" activity (leaving blunt-ended PCR products). Their results with Pfu showed none of the increased polymerase errors at low template concentrations that were seen with Taq (61), suggesting that Pfu may still have high fidelity for single cells.

To directly test the accuracy of sequencing from a single cell and to look for Pfu polymerase errors, we amplified and cloned the 2049-bp p53 fragment from a single cell using 35 PCR cycles and PfuTurbo. Using automated sequencing and BigDye terminator chemistry, we sequenced four clones from both ends, using plasmid primers T3 and T7. Because exon 7 is not located close enough to either end of the clone to be sequenced from the T3 or T7 reactions, an additional sequencing primer, J, was used to analyze exon 7. For each of these three p53 regions, 600 bp were read for each of the four clones. After visually inspecting each sequence chromatogram from miscalls by the sequencer, we were left with an average of ~4 ambiguous calls per 600 bp read (44 ambiguous calls in a total of 7200 bp read). Calls were considered ambiguous when peaks from two or more different bases had similar intensities. In all cases, the ambiguous calls from one clone were not found on the other three and, when aligned to the published p53 sequence, gave the correct sequence. In contrast, there were only three calls that differed from the p53 sequence and appeared genuine on the chromatograms. Because these base changes did not occur in more than one clone, they are considered genuine polymerase errors. Although three polymerase errors in 7200 bp read is substantially higher than predicted (i.e., 1 in 22 000), it is still quite low and can be easily compensated for by comparing the sequence of several clones.

Additionally, sequencing through a run of 16 A:Ts (5' to exon 7, starting at base 13914; GenBank accession no. U94788) led to three of the four clones showing different lengths of the T run. Two clones sequenced the region as 17 A:Ts, and one clone showed 14 A:Ts. Interestingly, all four clones sequenced correctly through a run of 10 A:Ts 150 bp away from the 16 A:T run. This error in a long run of A:Ts was the only error that could not be compensated for by comparing the sequences of four clones. Caution is therefore recommended when sequencing long base runs from single cells. Overall, these results indicate that direct sequencing of several clones amplified from a single cell (using PfuTurbo) is an accurate and reliable method for detection of DNA mutations.

comparison of HeLa cell p53 sequences with the wild-type sequence
Comparison of our consensus sequence from four clones showed no mutations in exons 5–9 of the p53 gene, confirming previous observations that HeLa cells do not contain a mutated p53 (64). However, comparison of our p53 intron sequences with the most recent published sequence revealed three genuine sequence discrepancies (seen in all four clones). The first was a base difference in the intron between exons 7 and 8 (at base 14168; GenBank accession no. U94788). We were reassured to discover that this was actually an error in the original sequence, which was recently corrected by a direct-submission update to GenBank (accession no. AF209155). The second discrepancy was an additional base in the intron between exons 6 and 7 at base 13607 [TACGAG->TACAGAG (the added base is in bold italics)]. The third discrepancy involved another base difference we detected in the intron between exons 9 and 10 [base 14977; AATAGTT->AATTGTT (the base difference is in bold italics)]. A Blast search of this region discovered two p53 sequences that showed this base change (GenBank accessions nos. S166666 and S81486). These sequences were isolated from two different lymphoblastic leukemia cell lines (65)(66). To test whether these last two discrepancies were either sequence errors in the GenBank p53 sequence or mutations specific to HeLa cells, we amplified and isolated the p53 gene from normal human genomic DNA from five to six people, using the same protocol as for single cells. Sequence analysis of every individual showed the same sequence found in HeLa cells, indicating that both of these base changes were indeed more errors in the GenBank sequence and not specific to HeLa cells or the lymphoblastic leukemia cell lines.

During these comparisons of HeLa sequences with normal human clones from many individuals, we also decided to examine whether the (A:T)16 run was also difficult to sequence when adequate amounts of genomic DNA (100 ng) were used for the PCR. Under these conditions, we did not see any inconsistencies in the run length even when seven separate sequencing reactions were performed on the same clone. However, to our surprise, different individuals showed different lengths of this A:T run. To confirm this result, we cloned the p53 gene from 23 individuals and sequenced each clone two to seven times. Again, multiple sequencing of every individual showed the same length with the distribution of lengths between different individuals as follows: length (L) = 13 (1 person), L = 14 (1 person), L = 15 (2 persons), L = 16 (8 persons), L = 17 (9 persons), L = 18 (1 person), and L = 19 (1 person). We also confirmed the run length in HeLa cells as L = 16 in 10 genomic clones, consistent with the HeLa cell line being isolated from a single individual (with cervical cancer). Because this is the first report of this length polymorphism in the intron between exons 6 and 7, it will be of interest to determine whether this polymorphism correlates with any splicing variations in p53. Interestingly, this length polymorphism lies in an intron open reading frame of 84 amino acids. Changes in the length (except L = 13 and L = 19, which would maintain the reading frame) lead to significant truncation of this hypothetical protein. Whether this hypothetical protein is actually transcribed or whether the length polymorphism has any functional significance is unknown.

With the confirmation of one recently discovered GenBank error and the identification of two previously unreported errors in the GenBank p53 intron sequence, it is apparent that our approach of sequencing the entire genomic region of p53 containing exons 5–9 from single cells is both practical and highly accurate when four clones are examined. Additionally, the inclusion of intron sequences may yield potentially important information about splice junctions and other intron alterations frequently overlooked by the common approach of analyzing only coding sequences.


   Acknowledgments
 
This work was made possible with the support of a Small Business Innovation Research grant (SBIR) from the National Institutes of Health, National Cancer Institute (NIH-NCI), Grant CA68913. Christopher Romero was supported by a Research Supplement for Under-Represented Minorities to NIH-NCI Grant R29 CA70421.


   Footnotes
 
1 Nonstandard abbreviations: PBS, phosphate-buffered saline; BSA, bovine serum albumin; and PK, proteinase K.


   References
Top
 Abstract
 Introduction
 Material and Methods
 Results and Discussion
 References
 

  1. Findlay I. Single cell PCR—theory, practice, and clinical applications. Clin Appl PCR 1998;16:233-263.[CrossRef]
  2. Girardet A, Pellestor F, Claustres M. Development of DNA amplification techniques from single cells. MS Med Sci 1999;15:86-89.
  3. Hahn S, Zhong XY, Troeger C, Burgemeister R, Gloning K, Holzgreve W. Current applications of single-cell PCR. Cell Mol Life Sci 2000;57:96-105.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  4. Kumazaki T, Hamada K, Mitsui Y. Detection of mRNA expression in a single cell by direct RT-PCR. Biotechniques 1994;16:1017-1019.[ISI][Medline] [Order article via Infotrieve]
  5. Freeman TC, Lee K, Richardson PJ. Analysis of gene expression in single cells. Curr Opin Biotechnol 1999;10:579-582.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  6. Dixon AK, Richardson PJ, Lee K, Carter NP, Freeman TC. Expression profiling of single cells using 3 prime end amplification (TPEA) PCR. Nucleic Acids Res 1998;26:4426-4431.[Abstract/Free Full Text]
  7. Dixon AK, Richardsen PJ, Pinnock RD, Lee K. Gene-expression analysis at the single-cell level. Trends Pharmacol Sci 2000;21:65-70.[CrossRef][Medline] [Order article via Infotrieve]
  8. Noe V, Alemany C, Ciudad CJ. Determination of dihydrofolate reductase gene amplification from single cell colonies by quantitative polymerase chain reaction. Anal Biochem 1995;224:600-603.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  9. Brady G, Iscove NN. Construction of cDNA libraries from single cells. Methods Enzymol 1993;225:611-622.[ISI][Medline] [Order article via Infotrieve]
  10. Angelis KJ, Dusinskaa M, Collins AR. Single cell gel electrophoresis: detection of DNA damage at different levels of sensitivity. Electrophoresis 1999;20:2133-2138.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  11. Thomale J, Seiler F, Muller MR, Seebe S, Rajewsky MF. Repair of O6-alkylguanines in the nuclear DNA of human lymphocytes and leukaemic cells—analysis at the single-cell level. Br J Cancer 1994;69:698-705.[ISI][Medline] [Order article via Infotrieve]
  12. Buschfort C, Muller MR, Seeber S, Rajewsky MF, Thomale J. DNA excision repair profiles of normal and leukemic human lymphocytes: functional analysis at the single-cell level. Cancer Res 1997;57:651-658.[Abstract/Free Full Text]
  13. Renner C, Trumper L, Pfitzenmeier JP, Loftin U, Gerlach K, Stehle I, et al. Differential mRNA display at the single-cell level. Biotechniques 1998;24:720-724.[ISI][Medline] [Order article via Infotrieve]
  14. Piqueras B, Autran B, Debre P, Gorochov G. Detection of apoptosis at the single-cell level by direct incorporation of fluorescein-dUTP in DNA strand breaks. Biotechniques 1996;20:634-640.[ISI][Medline] [Order article via Infotrieve]
  15. Dyhrman ST, Palenik B. Phosphate stress in cultures and field populations of the dinoflagellate Prorocentrum minimum detected by a single-cell alkaline phosphatase assay. Appl Environ Microbiol 1999;65:3205-3212.[Abstract/Free Full Text]
  16. Becker I, Becker KF, Rohrl MH, Minkus G, Schutze K, Hofler H. Single-cell mutation analysis of tumors from stained histologic slides. Lab Invest 1996;75:801-807.[ISI][Medline] [Order article via Infotrieve]
  17. Garvin AM, Holzgreve W, Hahn S. Highly accurate analysis of heterozygous loci by single cell PCR. Nucleic Acids Res 1998;26:3468-3472.[Abstract/Free Full Text]
  18. Voullaire L, Wilton L, Slater H, Williamson R. Detection of aneuploidy in single cells using comparative genomic hybridization. Prenat Diagn 1999;19:846-851.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  19. Klein CA, Schmidt-Kittler O, Schardt JA, Pantel K, Speicher MR, Riethmuller G. Comparative genomic hybridization, loss of heterozygosity, and DNA sequence analysis of single cells. Proc Natl Acad Sci U S A 1999;96:4494-4499.[Abstract/Free Full Text]
  20. Schlessinger J. Novel fluorescent approaches for studying cell signaling in single cells. Nat Biotechnol 2000;18:262-263.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  21. Li LJ, Garden RW, Sweedler JV. Single-cell MALDI: a new tool for direct peptide profiling. Trends Biotechnol 2000;18:151-160.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  22. Arnheim N, Li H, Cui X. PCR analysis of DNA sequences in single cells: single sperm gene mapping and genetic disease diagnosis. Genomics 1990;8:415-419.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  23. Higuchi R, von Beroldingen CH, Sensabaugh GF, Erlich HA. DNA typing from single hairs. Nature 1988;332:543-546.[CrossRef][Medline] [Order article via Infotrieve]
  24. Li H, Gyllensten U, Cui X, Saiki R, Erlich H, Arnheim N. Amplification and analysis of DNA sequences in single human sperm and diploid cells. Nature 1988;335:414-417.[CrossRef][Medline] [Order article via Infotrieve]
  25. Vrettou C, Palmer G, Kanavakis E, Tzetis M, Antoniadi T, Mastrominas M, et al. A widely applicable strategy for single cell genotyping of ß-thalassaemia mutations using DGGE analysis: application to preimplantation genetic diagnosis. Prenat Diagn 1999;19:1209-1216.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  26. Sutterlin M, AbouSleiman P, Onadim Z, Delhanty J. Single cell detection of inherited retinoblastoma predisposition. Prenat Diagn 1999;19:1231-1236.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  27. Liu J, Lissens W, Devroey P, Van Steirteghem A, Liebaers I. Efficiency and accuracy of polymerase-chain-reaction assay for cystic fibrosis allele {Delta}F508 in single cell. Lancet 1992;339:1190-1192.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  28. Findlay I, Matthews P, Quirke P. Multiple genetic diagnoses from single cells using multiplex PCR: reliability and allele dropout. Prenat Diagn 1998;18:1413-1421.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  29. Erlich HA, Gelfand AD, Sninsky JJ. Recent advances in the polymerase chain reaction. Science 1991;252:1643-1651.[Abstract/Free Full Text]
  30. Curran S, McKay JA, McLeod HL, Murray GL. Laser capture microscopy. Mol Pathol 2000;53:64-68.[Abstract/Free Full Text]
  31. Schutze K, Lahr G. Identification of expressed genes by laser-mediated manipulation of single cells. Nat Biotechnol 1998;16:737-742.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  32. Ashkin A, Dziedzic JM, Bjorkholm JE, Chu S. Observation of a single-beam gradient force optical trap for dielectric particles. Optics Lett 1986;11:288-290.[ISI]
  33. Ashkin A, Dziedzic JM, Yamane T. Optical trapping and manipulation of single cells using infrared laser beams. Nature 1987;330:769-771.[CrossRef][Medline] [Order article via Infotrieve]
  34. Block SM. Making light work with optical tweezers. Nature 1992;360:493-495.[CrossRef][Medline] [Order article via Infotrieve]
  35. Conia J, Edwards BS, Voelkel S. The micro-robotic laboratory: optical rapping and scissing for the biologist. J Clin Lab Anal 1997;11:28-38.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  36. Welch AJ, Motamedi M, Rastegar S, LeCarpentier GL, Jansen D. Laser thermal ablation. Photochem Photobiol 1991;53:815-823.[ISI][Medline] [Order article via Infotrieve]
  37. Aufderheide KJ, Du Q, Fry ES. Directed positioning of micronuclei in Paramecium tetraurelia with laser tweezers: absence of detectable damage after manipulation. J Eukaryot Microbiol 1993;40:793-796.[ISI][Medline] [Order article via Infotrieve]
  38. Celliers P, Conia J. Measurement of localized heating in the focus of an optical trap. Appl Optics 2000;39:3396-3407.
  39. Liu Y, Cheng DK, Sonek GJ, Berns MW, Chapman CF, Tromberg BJ. Evidence for localized cell heating induced by infrared optical tweezers. Biophys J 1995;68:2137-2144.[Abstract/Free Full Text]
  40. Liang H, Vu KT, Krishnan P, Trang TC, Shin D, Kimel S, et al. Wavelength dependence of cell cloning efficiency after optical trapping. Biophys J 1996;70:1529-1533.[Abstract/Free Full Text]
  41. Vorobjev IA, Liang H, Wright WH, Berns MW. Optical trapping for chromosome manipulation: a wavelength dependence of induced chromosome bridges. Biophys J 1993;64:533-538.[Abstract/Free Full Text]
  42. Schneckenburger H, Hendinger A, Sailer R, Gschwend MH, Strauss WSL, Bauer M, et al. Cell viability in optical tweezers: high power red laser diode versus Nd:YAG laser. J Biomed Opt 2000;5:40-44.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  43. Konig K, Liang H, Berns MW, Tromberg BJ. Cell damage by near-IR microbeams. Nature 1995;377:20-21.[CrossRef][Medline] [Order article via Infotrieve]
  44. Greulich KO, Weber G. The light microscope on its way from an analytical to a preparative tool. J Microsc 1992;167:127-151.
  45. Townes-Anderson E, St. Jules RS, Sherry DM, Lictenberg J, Hassanain M. Micromanipulation of retinal neurons by optical tweezers. Mol Vis 1998;4:12-17.[Medline] [Order article via Infotrieve]
  46. Seeger S, Monajembashi S, Hutter K-J, Futterman G, Wolfrum J, Greulich KO. Application of optical tweezers in immunology and molecular genetics. Cytometry 1991;12:497-504.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  47. Goode BL, Feinstein SC. Speedprep purification of template for double-stranded DNA sequencing. Biotechniques 1992;12:374-375.[ISI][Medline] [Order article via Infotrieve]
  48. Giambernardi TA, Rodeck U, Klebe RJ. Bovine serum albumin reverses inhibition of RT-PCR by melanin. Biotechniques 1998;25:564.[ISI][Medline] [Order article via Infotrieve]
  49. Kreader CA. Relief of amplification inhibition in PCR with bovine serum albumin or T4 gene 32 protein. Appl Environ Microbiol 1996;62:1102-1106.[Abstract]
  50. Sorensen PD, Frederiksen S. Characterization of human 5S rRNA genes. Nucleic Acids Res 1991;19:4147-4151.[Abstract/Free Full Text]
  51. Little RD, Braaten DC. Genomic organization of human 5S rDNA and sequence of one tandem repeat. Genomics 1989;4:376-383.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  52. Schmidt TM, Pace B, Pace NR. Detection of DNA contamination in Taq polymerase. Biotechniques 1991;11:176-177.[ISI][Medline] [Order article via Infotrieve]
  53. Corless CE, Guiver M, Borrow R, Edwards-Jones V, Kaczmarski EB, Fox AJ. Contamination and sensitivity issues with a real-time universal 16S rRNA PCR. J Clin Microbiol 2000;38:1747-1752.[Abstract/Free Full Text]
  54. Louwrier A. Nucleic acid removal from Taq polymerase preparations using an aqueous/organic biphasic system. Biotechniques 1999;27:444-445.[ISI][Medline] [Order article via Infotrieve]
  55. Meier A, Persing DH, Finken M, Bottger EC. Elimination of contaminating DNA within polymerase chain reaction reagents: implications for a general approach to detection of uncultured pathogens. J Clin Microbiol 1993;31:646-652.[Abstract/Free Full Text]
  56. Hughes MS, Beck LA, Skuce RA. Identification and elimination of DNA sequences in Taq DNA polymerase. J Clin Microbiol 1994;32:2007-2008.[Abstract/Free Full Text]
  57. Sarkar G, Sommer SS. Removal of DNA contamination in polymerase chain reaction reagents by ultraviolet irradiation. Methods Enzymol 1993;218:381-388.[ISI][Medline] [Order article via Infotrieve]
  58. Rochelle PA, Weightman AJ, Fry JC. DNase I treatment of Taq DNA polymerase for complete PCR decontamination. Biotechniques 1992;13:520.[ISI][Medline] [Order article via Infotrieve]
  59. Miyagi T, Murakami K, Sawada T, Taguchi H, Miyoshi I. A novel single cell PCR assay: detection of human T lymphotropic virus type I DNA in lymphocytes of patients with adult T cell leukemia. Leukemia 1998;12:1645-1650.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  60. Cline J, Braman JC, Hogrefe HH. PCR fidelity of pfu DNA polymerase and other thermostable DNA polymerases. Nucleic Acids Res 1996;24:3546-3551.[Abstract/Free Full Text]
  61. Odeberg J, Ahmadian A, Williams C, Uhlen M, Ponten F, Lundeberg J. Context-dependent Taq-polymerase-mediated nucleotide alterations, as revealed by direct sequencing of the ZNF189 gene: implications for mutation detections. Gene 1999;235:103-109.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  62. Marchuk D, Drumm M, Saulino A, Collins FS. Construction of T-vectors, a rapid and general system for direct cloning of unmodified PCR products. Nucleic Acids Res 1991;19:1154.[Free Full Text]
  63. Holton TA, Graham MW. A simple and efficient method for direct cloning of PCR products using ddT-tailed vectors. Nucleic Acids Res 1991;19:1156.[Free Full Text]
  64. Scheffner M, Munger K, Byrne JC, Howley PM. The state of the p53 and retinoblastoma genes in human cervical carcinoma cell lines. Proc Natl Acad Sci U S A 1991;88:5523-5527.[Abstract/Free Full Text]
  65. Chow VTK, Quek HH, Tock EPC. Alternative splicing of the p53 tumor suppressor gene in the Molt-4 T-lymphoblastic leukemia cell line. Cancer Lett 1993;73:141-148.[CrossRef][ISI][Medline] [Order article via Infotrieve]
  66. Felix CA, Nau MM, Takahashi T, Mitsudomi T, Chiba I, Poplack DG, et al. Hereditary and acquired p53 gene mutations in childhood acute lymphoblastic leukemia. J Clin Invest 1992;89:640-647.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited