Feature Articles

Published: November 3, 2010
Find more content on:
Portable real-time PCR for rapid viral identification

The ability to perform reliable, low-cost PCR amplification with either intercalating dyes or FRET probes will lead to an increased use of such a tool in clinical IVDs.

By: George Maltezos, Wasun Chantratita, Alvaro Gomez, Frank Gomez, Emil Kartalov, and Axel Scherer

A robust, portable, battery-operated real-time polymerase chain reaction (PCR) system has been developed, which amplifies viral nucleic acid with equal accuracy and at a lower cost than currently available state-of-the-art devices. Such inexpensive PCR virus identification could serve remote regions of the world and lead to an early warning system for pandemic outbreaks. This article compares the portable real-time PCR system with a conventional desktop unit. The article also shows that the H5N1 avian influenza virus, HIV/AIDS, and other common viruses can be identified at low viral loads in a clinical setting.

Figure 1. Schematic view of the portable PCR device.

The miniaturized amplification system demonstrated precise temperature control, which results in amplification efficiency and specificity that is at least equivalent to larger commercial instruments. The precise temperature control also enables the use of less expensive intercalating dyes as fluorescent reporters, which further reduces the cost and increases the accessibility of PCR testing.

Since the mid-1990s, real-time PCR has had a revolutionary impact on biochemistry.1-4 Such reactions utilize selective primers to identify and amplify specific nucleic acid molecules simultaneously in a sample. By allowing the user to monitor the progress of PCR as it occurs, real-time PCR provides information on the concentration of DNA in a sample, which removes many of the limitations of earlier endpoint PCR.5,6

More recent advances include techniques for the precise and rapid thermal control of nanoliter and microliter analyte volumes using Peltier junctions, and amplification of RNA sequences using real-time reverse transcription PCR.7-12 Real-time PCR is unrivalled in sensitivity for virus identification and determination of viral load. However, such diagnostic tests are traditionally performed only in large, expensive thermal cyclers.

Figure 2. Lentiviral DNA samples were amplified in the prototype device. They were then melted with the commercial instrument (purple) and the portable PCR device with silicon PIN diode (green).

Viral epidemics are a growing public health concern and have proven to be very costly. For example, the estimated cost of the recent SARS outbreak ranges from $10 to 30 billion. Even the 1994 local outbreak of plague in Surat, India cost an estimated $2 billion, and the 1997 avian flu in Hong Kong cost hundreds of millions of dollars in lost poultry production, commerce, and tourism.13

The introduction of small, inexpensive instruments that can use lower-cost fluorophore reporters is expected to lead to widespread availability of PCR testing, with the goal of preventing pandemics through effective monitoring of viral outbreaks.14 Miniaturized systems may be easily and rapidly deployed to airports, rural areas, and anywhere a rapid, specific viral test is needed. Point-of-care PCR analysis may eventually lead to earlier diagnosis of viral diseases and more efficient patient monitoring.

Materials and Methods
The prototype portable real-time PCR thermal cycler consists of Peltier junction thermoelectric coolers by Marlow Industries (Dallas) that are soldered to a copper heat-exchange jacket into which a glass tube is inserted (see Figure 1). A programmable thermal controller by Oven Industries (Mechanicsburg, PA) was used to achieve temperature ramp sequences. A battery pack consisting of 12 NiMH 10,000-mAH D-cell batteries by Tenergy Corp. (Fremont, CA) was able to power the unit for more than 100 PCR runs, with the number of runs depending on the specific thermal cycling profile.

Figure 3. Real-time fluorescence monitoring system showing the amplification of Lentivirus DNA using SYBR green I as the reporter.

While analysis using this prototype device was limited to one sample run at a time, the unit could be expanded to accept multiple samples for only the incremental cost of additional Peltier elements. The dimensions of the thermal cycler unit consisting of a PCR Peltier head, controller, and batteries are approximately 200 mm × 200 mm × 140 mm. No additional electronic devices or computers were required to perform thermal cycling.

To perform dissociation curve and real-time amplification fluorescence measurements, a high-power 480-nm ±10 nm 700-mA Xlamp LED light source by Cree (Durham, NC) and a 520-nm ±10 nm filtered silicon PIN photodiode detector by Intor (Socorro, NM) were integrated into the thermal cycler. For comparison, fluorescence was also detected using a cooled CCD photomultiplier by Hamamatsu (Iwata City, Japan). The output of the PIN photodiode was monitored using a picoammeter by Keithley Instrument (Cleveland) and analyzed using a custom Labview program by National Instruments (Austin, TX).

To test the prototype thermal cycler, lentiviral DNA was amplified using both the prototype unit and a LightCycler Real-time PCR System by Roche Applied Science (Indianapolis). Clinical samples were obtained through a research program at Ramathibodi Hospital (Bangkok). The nucleic acid was isolated from 200-μl samples of plasma, whole blood, or cerebrospinal fluid (CSF) using the NucliSens easyMAG platform for total nucleic acid extraction by bioMérieux (Marcy l’Étoile, France). Elution of DNA, RNA, or total nucleic acid from the column was performed using 30-50 μl of elution buffer.

Figure 4. Negative derivative of the fluorescence melting curves of H5N1 (avian influenza). Samples from infected chickens were amplified with 45 PCR cycles in the Roche LightCycler (black line) and the portable PCR system (gray line).

The viral loads of the HIV plasma samples were determined using the Cobas AmpliPrep/Cobas Amplicor HIV-1 Monitor Test v1.5 by Roche Molecular Systems (Pleasanton, NJ). PCR reactions were prepared as mastermixes, and identical 20-µl samples were aliquoted for simultaneous analysis in the LightCycler desktop system and the portable PCR system. All amplifications and measurements were conducted in triplicate. Detailed PCR profiles are provided in electronic supplementary information.

Results
Testing of the Prototype Real-Time PCR Unit. To evaluate selectivity and efficiency of target amplification in the prototype thermal cycler, fluorescence intensity was measured during lentiviral DNA amplification with SYBR green I intercalating dye. Side-by-side reactions of aliquots from a single lentivirus DNA sample were performed on the portable PCR system and a Roche LightCycler. The threshold cycle (i.e., the cycle number at which fluorescence intensity was observed to pass the threshold) was approximately 14 for each system. The melting curves of the resulting amplicons are shown in Figure 2 to demonstrate specificity of the reactions. Figure 3 shows a typical amplification curve obtained from the prototype device.

Amplification of Clinical Samples. The portable PCR instrument was further tested using clinical plasma, whole blood, and CSF samples at Ramathibodi Hospital. The sensitivity and selectivity of the portable PCR system was evaluated by comparing amplifications with commercially available primers that were run on a LightCycler I desktop system and the PCR system. Resulting products from both instruments were analyzed by performing melting curves in the LightCycler instrument. The results demonstrated that the performance of the portable PCR system was comparable to or better than the LightCycler for a number of clinically significant viruses.

Figure 5. Negative derivative of the fluorescence melting curves of HIV/AIDS. Human blood samples were amplified through 65 PCR cycles in the Roche LightCycler (black line) and the portable PCR system (gray line).

H5N1 Avian Influenza. The potential for an international pandemic resulting from the H5N1 strain of avian influenza has recently caused great concern among policymakers in search of effective identification techniques. The portable real-time PCR instrument was therefore tested with samples from H5N1-infected chickens. Using the hospital’s previously optimized real-time PCR primers and thermal cycling profile, the in vitro–transcribed H5 gene RNA was first amplified to serve as a positive control, and then H5N1 virus RNA from infected chicken samples were analyzed. Several concentrations of the H5N1 equivalent were amplified with the PCR system to simulate the effects of different viral loads.

The amplification products were analyzed by running melting curves with the LightCycler instrument. Figure 4 shows hybridization probe melting curves (i.e., dissociation of the hybridization probe from the amplified product) of H5N1 samples, which plot the negative derivative of the fluorescence as a function of the temperature. The melting curves of samples produced by both the Roche LightCycler and the portable real-time PCR system do not differ greatly and indicate a hybridization probe melting temperature of 62.9o C (±2.5o C). Close examination also reveals a reduced shoulder peak in the melting curve of the sample that was produced by the portable instrument, which can be attributed to lower non-specific binding and reduced primer–dimer interactions during amplification.

HIV/AIDS. The effective treatment and prevention of HIV transmission relies on early detection, which could be offered by the introduction of inexpensive PCR testing in clinics throughout the world. PCR is capable of identifying HIV as early as two weeks after infection. Blood samples containing HIV were analyzed at several viral loads with both the LightCycler and the portable PCR instrument by using a primer set corresponding to the HIV-1 gag gene (GenBank accession no. EF680874). This method used a two-probe hybridization mix for detection of the complete subtype of HIV.

Figure 6. Negative derivative of the fluorescence melting curves of hepatitis B (HBV). Human blood samples were amplified in the Roche LightCycler (black line) and the portable PCR system (grey line).

Typical hybridization probe melting curves for very dilute viral loads after 65 thermal cycles are shown in Figure 5. The DNA melting temperature of approximately 56o C (±2.5o C) was similar for samples that were amplified using both the LightCycler and the portable PCR system. By exploring the lowest limits of viral load, the ultimate sensitivity of both instruments was found to be similar. Because dosage with fairly expensive drug cocktails often varies from patient to patient, it is beneficial to be able to monitor patient viral loads frequently during treatment and adjust dosage appropriately.

Hepatitis B. To test that the portable PCR instrument can amplify both DNA and RNA viruses, amplification experiments were performed on hepatitis B (HBV), a common DNA virus. Human blood samples containing HBV were amplified with both the LightCycler and the portable PCR instrument using primers corresponding to the HBV polymerase gene (GenBank accession no. DQ995858). Figure 6 shows the SYBR green I melting curves of samples that were amplified in the LifeCycler and the portable system after 45 cycles. The melting temperature was determined to be 79.6o C (±3.0o C). The data demonstrate that the portable PCR instrument can amplify both DNA and RNA viruses with results that are similar to a larger commercially available amplification system.

Herpes Simplex I and II. The portable PCR system can also amplify and genotype more than one virus strain in the same sample by testing human CSF that contains both herpes simplex I (HSV-I) and herpex simplex II (HSV-II) strains. SYBR green I melting curves of the amplicons show two peaks at 54o C (±2.5o C) and 67.2o C (±2.5o C), corresponding to HSV-I and HSV-II, respectively (see Figure 7). The ability to distinguish these two peaks indicates that the portable PCR system can amplify and differentiate dual or coinfection of viruses in the same sample.

Figure 7. Negative derivative of the fluorescence melting curves of HSV I and HSV II (herpes simplex) mixture. Human blood samples were amplified with 45 PCR cycles in the Roche LightCycler (black line) and the portable PCR system (gray line).

Cytomegalovirus. Cytomegalovirus (CMV), another clinically relevant virus, was also amplified from human blood samples by using primers corresponding to the human cytomegalovirus UL122 gene (GenBank accession no. X92746). Figure 8 shows SYBR green melting curves of the amplicons that were generated by 45 PCR cycles using the LightCycler and the portable PCR system. The DNA melting temperature was approximately 85.8o C (±3.0o C).
Figure 9 shows fluorescence images of thermal cycler capillary tubes that contained CMV samples (panels A and C) or negative controls (panels B and D), and had been processed in the prototype system (panels A and B) or in the Roche LightCycler (panels C and D). Comparing the negative controls reveals a stronger signal in the LightCycler, which can be explained by a modest primer–dimer peak at 80.7o C (see Figure 10). No such peak appears in the negative control samples that were amplified by the prototype PCR system (see Figure 10).

Discussion
The portable real-time PCR system was constructed for approximately $500 in parts and is simple, expandable, portable, and robust. Even though the battery pack is the heaviest component of this system, the system has a low weight and a small size, thus making it fully portable.

Figure 8. Negative derivative of the fluorescence melting curves of cytomegalovirus (CMV). Human blood samples were amplified through 45 PCR cycles in the Roche LightCycler (black line) and the portable PCR system (gray line).

PIN-type solid-state detectors provide an adequate solution for real-time measurement of typical fluorescent signals during PCR amplification, and the signal can be easily measured by normal electronics. The portable PCR device heats and cools glass capillary tubes that were also used in real-time LightCycler instruments via the Peltier effect.7 A typical temperature ramp curve that was measured in the PCR sample demonstrated thermal cycling with 0.1o-C control over temperature. Because the ramp rate typically determines PCR amplification time, it is possible to reduce reaction time to less than 10 minutes by increasing ramping to about 30o C/s. However, in this article, the standard ramping protocols that were applied were optimized for benchtop PCR systems.

Careful analysis of the negative control sample during CMV PCR revealed an unexpected advantage of the portable PCR system when using intercalating dye probes such as SYBR Green I rather than more expensive FRET probes. Figures 9 and 10 present evidence of primer–dimer amplification in the Light-Cycler system, but not in the prototype PCR system. The false peaks were also found in the melting-curve analysis of the HBV-negative controls that were amplified with the LightCycler, but not in the negative samples that were amplified with the portable unit.

Figure 9. Fluorescence images of CMV capillary tubes after amplification with SYBR green showing (a) positive CMV sample from the portable unit, (b) negative control from the portable unit, (c) positive CMV sample from the LightCycler, and (d) negative control from the LightCycler.

The most significant advantage of miniaturization is the reduction in the thermal response time as the thermal mass of a PCR reactor is reduced. Miniaturized systems can heat and cool 20-μl samples using approximately 30o C/s ramp rates, and smaller sample volumes allow further increases in temperature ramp rates. Rates greater than 100° C/s heating and almost 90° C/s cooling of a 400-nl sample in a Peltier cooling system were achieved. Fast ramping reduces the overall time required for the PCR amplification process from hours to minutes. However, the advantage of reducing analyte volumes must be carefully weighed against the difficulty of amplifying samples that contain dilute viral loads; larger sample volumes improve sensitivity. 10-20 μl constitutes a good compromise for current PCR systems, and 20-μl volumes were used in this study.

Conclusion
This article described a portable real-time PCR system with precise temperature control and demonstrated the detection of several clinically important viruses using this system. The ability to perform reliable, low-cost PCR amplification with either intercalating dyes or FRET probes will lead to an increased use of this sensitive tool in livestock monitoring, forensic testing, and medical diagnostics.

Acknowledgement
The authors would like to thank Chutatip Srichantaratsamee, Konstantin Taganov, John Zhong, William F. Anderson, and Koichi Okamoto for their help and contributions to this article. Funding support was provided by the Boeing Corp. under the SRDMA program and the National Institutes of Health under R01 HG002644 and R00 EB007151.

Figure 10. Negative derivative of fluorescence melting curve of the CMV negative control amplified with SYBR green. The negative control run in the LightCycler produces an amplification product fluorescence signal, while negative control run in the portable unit does not produce such a signal.

References
1. R Saiki, et al., “Enzymatic Amplification of b-Globin Genomic Sequences and Restriction Sight Analysis for Diagnosis of Sickle Cell Anemia,” Science 230 (1985): 1350-1354.
2. RK Saiki, et al., “Analysis of Enzymatically Amplified Beta-Globulin and HLA-DQ-Alpha DNA with Allele-Specific Oligonucleotide Probes,” Nature 324 (1986): 335–351.
3. HA Erlich, PCR Technology (New York: Stockton Press, 1989.)
4. K Edwards, J Logan, and N Saunders, Real-time PCR: An Essential Guide (Norfold, UK: Horizon Bioscience, 2004.)
5. R Higuchi, et al., “Simultaneous Amplification and Detection of Specific DNA Sequences,” Biotechnology 10 (1992): 413-417.
6. R Higuchi, et al., “Kinetic PCR Analysis Real-Time Monitoring of DNA Amplification Reactions,” Biotechnology 11 (1993): 1026-1030.
7. CT Wittwer, et al., “The LightCycler: a Microvolume Multisample Fluorimeter with Rapid Temperature Control,” Biotechniques 22 (1997): 176-181.
8. MA Northrup, et al., “A Miniature Analytical Instrument for Nucleic Acids Based on Micromachined Silicon Reaction Chamber,” Analytical Chemistry 70 (1998): 918–922.
9. P Belgrader, et al., “Rapid Pathogen Detection Using a Microchip PCR Array Instrument,” Clinical Chemistry 44 (1998): 2191–2194.
10. CT Wittwer, et al., “High-Resolution Genotyping by Amplicon Melting Analysis Using LC Green,” Clinical Chemistry 49 (2003): 853–860.
11. D Simpson, RM Crosby, and TR Skopek, “A Method for Specific Cloning and Sequencing of Human HPRT cDNA for Mutation Analysis,” Biochemical Biophysical Research Communications 151 (1988): 487–492.
12. H Vrieling, JW Simons, and AA van Zeeland, “Nucleotide Sequence Determination of Point Mutations at the Mouse HPRT Locus Using In Vitro Amplification of HPRT mRNA Sequences,” Mutation Research 198 (1988): 107–113.
13. National Intelligence Council, “The Global Infectious Disease Threat and its Implications for the United States” (Washington, DC: Office of the Director of National Intelligence, 2000 [cited 23 September 2010]); available from Internet: www.dni.gov/nic/PDF_GIF_otherprod/infectiousdisease/infectiousdiseases.pdf.
14. M Enserink, “Infectious Diseases: Veterinary Scientists Shore up Defenses Against Bird Flu,” Science 308 (2005): 341.

George Maltezos, PhD, is founder and chief engineer of Helixis Inc. He can be reached at george@helixis.com.

Wasun Chantratita, PhD, is chief of virology and the molecular microbiology unit at Ramathibodi Hospital, Mahidol University (Bangkok). He can be reached at rawct@mahidol.ac.th.

Alvaro Gomez is an undergraduate research assistant at the California Institute of Technology (Pasadena, CA). He can be reached at matadorsin@hotmail.com.

Frank A. Gomez, PhD, is a professor of chemistry in the department of chemistry and biochemistry at California State University, Los Angeles. He can be reached at fgomez2@calstatela.edu.

Emil Kartalov, PhD, is an assistant professor in the pathology department at Keck School of Medicine, University of Southern California (Los Angeles). He can be reached at kartalov@usc.edu.

Axel Scherer, PhD, is a professor of electrical engineering, physics, and applied physics in the department of electrical engineering at the California Institute of Technology. He can be reached at etcher@caltech.edu.
 


Your rating: None Average: 5 (1 vote)

Login or register to post comments