Automation of Fluorescence-Based PCR for Confirmation of Meningococcal Disease

Abstract

A fluoresence-based PCR method was developed, fully automated, and used to confirm infection with Neisseria meningitidis by detection of the meningococcus-specific ctrA gene. The method provided a highly sensitive, high-throughput assay that was reproducible and less labor-intensive than manual methods.

There are a number of methods available for the detection or amplification of specific DNA sequences (2, 7, 8, 9). Some of these have disadvantages, including low sensitivity, lack of specificity, high cost, or laborious methodology (10). During the past decade, PCR has provided an invaluable tool and has been partly responsible for the explosion in molecular biology (4). The method has found its place in many areas that utilize molecular techniques in research and nonresearch environments, including microbiology, animal and human genetics, and clinical diagnostics. Traditionally, PCR is performed in a commercial thermocycler and the products are visualized with a gel-based system (3). However, various technologies are now available to further exploit the PCR method. New chemistries, such as Taqman and Molecular Beacons, have been developed commercially to provide real-time PCR methods that are more sensitive than the equivalent gel-based system because they are fluorescence based (2, 8). These chemistries have also allowed the further expansion of the applications of PCR into areas such as single-nucleotide polymorphism analysis (5), while standard PCR has been developed into providing amplicons for microarray analysis (6). Automation has also recently become more affordable and is therefore accessible to more laboratories. It is now used heavily in the pharmaceutical industry and more recently has found use in academic research and clinical diagnostics.

Presented here is a novel application of one type of PCR chemistry with the capacity for high throughput and full automation. We have termed this method dual-labeled end-point fluorescence PCR (DEF-PCR), and it is based on a previously described chemistry (2) whereby oligonucleotide primers are dual labeled with a reporter dye, carboxyfluorescein, covalently linked to the 5′ end and the quencher dye, carboxy-tetramethylrhodamine, linked to the 3′ end. A probe hybridizes to a specific DNA sequence upon PCR product formation but is subsequently digested by 5′ exonuclease activity of Taq DNA polymerase during primer extension, thus releasing reporter dye and increasing fluorescence emissions. The procedure is fully automated on a liquid handling robot, and the formation of PCR products is analyzed via the alteration and subsequent increase in fluorescence emissions using an integrated 96-well-format fluorimeter.

Nonculture diagnosis of certain infectious diseases is becoming increasingly important as antibiotics are given prior to hospital admission. One such example is meningococcal infection whereby a rapid confirmation of disease is required for both patient treatment and case contact prophylaxis. We therefore used this method to demonstrate and confirm disease in patients clinically suspected to have meningococcal infection. To do this we used ctrA primers for the detection of Neisseria meningitidis DNA, and the PCR conditions were those used previously (1). This gene target has previously been shown to be sensitive and specific for this purpose and does not amplify DNA from other neisseriae or other species which may cause septicaemia or meningitis (1). Due to its fluorescence-based chemistry, the method is highly sensitive and here we have also fully automated the method.

DNA was extracted using the Nucleospin Blood protocol (ABgene, Surrey, United Kingdom) for the isolation of genomic DNA from whole blood, serum, and plasma. The reproducibility of the method was tested on 96 individual samples comprising 48 N. meningitidis DNA controls of various serogroups (Table 1) and 48 negative controls (sterile distilled water). After DNA extraction, samples were transferred into 1.8-ml non-cross-contamination tubes and placed in the sample rack of the Roboseq 4204 SE robotic liquid handling system, possessing an integrated thermocycler and fluorescence reader (MWG Biotech, Milton Keynes, United Kingdom). All components of the system were in a 96-well microtiter plate format, allowing standardization and high-throughput methodology. Programming of the liquid handling robot was performed according to the manufacturer's instructions. All PCR reagents were maintained at 4°C on the robotic platform. Each reaction was performed in a final volume of 50 μl consisting of 48 μl of 1.1× Reddymix PCR Master Mix (ABgene) containing 1.25 U of Taq DNA polymerase; 75 mM Tris-HCl (pH 8.8 at 25°C); 20 mM (NH₄)₂; 1.5 mM MgCl₂; 0.01% (vol/vol) Tween 20; a 0.2 mM concentration (each) of dATP, dCTP, dGTP, and DTTP; 1 μl of each primer (1 pmol) (MWG Biotech); 1 μl of dual-labeled probe (0.5 pmol) (MWG Biotech); and 2 μl of extracted DNA. Each reaction was set up automatically by the robot within a refrigerated 96-well microtiter plate using disposable tips. Cross-contamination was avoided by the use of these tips, which were discarded automatically into a waste container. After PCR setup, optically clear disposable strips (ABgene) were manually placed over the wells to seal the contents. The microtitre plate was then automatically placed into the integrated thermocycler for the period of thermocycling. After amplification, the microtiter plate was automatically removed from the thermocycler into the integrated Bio-Tek FL600 fluorescence plate reader. Selected wavelengths of 485 to 420 nm and 530 to 525 nm for excitation and emissions respectively were used to detect the fluorescence emissions caused by carboxyfluorescein. A total of 100 endpoint readings were taken from each well, and the average was calculated by using the KC4 software (MWG Biotech). The KC4 software was programmed to calculate a cutoff value based on the subtraction of the average of the three controls from the positive control.

TABLE 1.

Fluorescence emission cutoff values for the detection of the ctrA gene from Neisseria meningitidis

Control sample	No. of samples	Detection of ctrA by DEF-PCR
		No. (%) positive	No. (%) negative
N. meningitidis serogroup B	8	8 (100)	0 (0)
N. meningitidis serogroup C	27	24 (89)	3 (11)
N. meningitidis serogroup Y	2	2 (100)	0 (0)
N. meningitidis serogroup W135	4	4 (100)	0 (0)
N. meningitidis serogroup X	5	4 (80)	1 (20)
N. meningitidis serogroup NG	2	2 (100)	0 (0)
Total	48	44 (92)	4 (8)
Negative control	48	0 (0)	100 (100)

A threshold value was determined as 0.5 standard deviation above the mean of the background fluorescence emission for all wells after endpoint calculations. This standard deviation and subsequent cutoff value was calculated using the 48 positive and 48 negative controls. Of the positive controls, 44 were positive by the DEF-PCR assay, providing a sensitivity of 92%. Of the negative controls, all 48 (100%) were negative. The reproducibility and sensitivity of the method was therefore demonstrated, with all positive control samples being detected, with a sensitivity of 92% and with all negative control samples being negative (Table 1).

The method is relatively cheap compared with similar methods, as (i) the chemistry is widely available without the need for specialist equipment and, as such, can be performed manually using conventional setup techniques and (ii) results can be read with a manually operated fluorescent plate reader possessing the appropriate filter set. Although we developed the method using N. meningitidis control DNA, the method is now under ongoing evaluation in the laboratory for the nonculture confirmation of meningococcal and pneumococcal infection. The assays are now set up to include one positive control, three negative controls, and up to 92 test samples in a 96-well microtiter plate format.

Despite the fact that the method can be performed manually, we have described full automation here, although additional automation can be achieved with robotic systems by the inclusion of an integrated vacuum manifold for DNA extraction from body fluids. Although we used the 96-well microtiter plate format, many robotic systems can be designed with the 384-well format to achieve even higher throughput. In conclusion, DEF-PCR has been demonstrated as an effective tool for the high throughput and sensitive detection of meningococcal DNA for the nonculture confirmation of meningococcal infection. However, the method could be applied to other bacterial and viral infections but also has the potential for application in many areas including microbiological research, clinical diagnostics, and the pharmaceutical industry.