We describe here a simple and rapid paper-based DNA extraction method of HIV proviral DNA from whole blood detected by quantitative PCR. This protocol can be extended for use in detecting other genetic markers or using alternative amplification methods.
FINA, filtration isolation of nucleic acids, is a novel extraction method which utilizes vertical filtration via a separation membrane and absorbent pad to extract cellular DNA from whole blood in less than 2 min. The blood specimen is treated with detergent, mixed briefly and applied by pipet to the separation membrane. The lysate wicks into the blotting pad due to capillary action, capturing the genomic DNA on the surface of the separation membrane. The extracted DNA is retained on the membrane during a simple wash step wherein PCR inhibitors are wicked into the absorbent blotting pad. The membrane containing the entrapped DNA is then added to the PCR reaction without further purification. This simple method does not require laboratory equipment and can be easily implemented with inexpensive laboratory supplies. Here we describe a protocol for highly sensitive detection and quantitation of HIV-1 proviral DNA from 100 µl whole blood as a model for early infant diagnosis of HIV that could readily be adapted to other genetic targets.
Several reports have discussed the development of paper- or membrane-based extraction methods for use in point-of-care (POC) devices 1-5 with the aim of making the exquisite sensitivity and specificity of molecular diagnostics available to all. The World Health Organization (WHO) Sexually Transmitted Diseases Diagnostics Initiative coined the term ASSURED (Affordable, Sensitive, Specific, User-friendly, Rapid and Robust, Equipment-free and Delivered to those who need it) to describe the ideal characteristics of a POC test 6. Of these guidelines, the equipment-free characteristic is particularly challenging to achieve for molecular diagnostics. However, every innovation in the field will advance the goal of reaching those in most need, and there is hope for near-term improvements in test performance by adapting existing technology 7.
Here we describe a simple protocol for extracting DNA from whole blood that does not require complex chemistry or laboratory instrumentation. The FINA (filtration isolation of nucleic acids) sample preparation method was originally developed to extract leukocyte DNA from whole blood in order to detect the HIV-1 provirus as part of a sample-to-answer point-of-care (POC) quantitative PCR (qPCR) early infant diagnostic (EID) platform for use in limited resource settings 8-11. FINA extraction differs from conventional purification methods which use silica membranes or silica-coated paramagnetic particles to reversibly bind DNA in the presence of chaotropic agents 12. Instead, FINA uses vertical filtration via a separation membrane to extract cellular DNA from whole blood directly. The membrane containing the entrapped DNA can be placed directly in a PCR tube and either immediately used in a PCR reaction or air dried for later amplification 9. No chaotropic agents, phenol, or alcohols are used in the sample extraction, eliminating the extensive washing steps needed to remove potent qPCR inhibitors derived from the extraction process 13,14.
The FINA membrane can capture either cells 9 or genomic DNA liberated by cell lysis 11 before the specimen is added to the membrane. For the cell capture, whole blood is added directly to the membrane. The cells are subsequently lysed in the membrane by adding 10 mM NaOH. The advantage to this method is that it involves only 3 steps: 1) sample addition; 2) cell lysis/wash and 3) filter disk placement in qPCR tube. The drawback to this method is that the membrane can only hold a defined number of cells directly proportional to the diameter of the membrane disk. To reach the limit of detection required for EID, 100 µl of whole blood is required for sample input which entails a filter that is too large to be placed in a qPCR tube. Lysing the blood cells with detergent before adding the sample to the collection membrane adds a processing step, but allows the use of a smaller filter for the same sample size. We were able to demonstrate high reproducibility, single copy detection, and quantification of as little as 10 copies of HIV-1 proviral DNA from 100 µl of blood using this test configuration 11.
In this report, we describe the FINA protocol as originally developed for laboratory use. The membrane/filter sandwich known as the FINA sample preparation module can be prepared in large batches and stockpiled for later use. When specimens are to be extracted this process takes 2 min and can be performed in varying size batches. The qPCR can be run immediately or the filters containing the embedded DNA can be stored until it is convenient to perform qPCR. This method is very convenient for routine analysis of specimens in both low and high resource settings.
Ethics statement: The whole blood specimens used in this study are not considered to be research involving human subjects. The specimens were obtained for clinical diagnostic purposes, which were satisfied, and the remaining portion of these specimens was provided for the FINA research assay. The specimens were coded such that the investigators were not able to readily ascertain the identity of individuals.
1. Preparation of FINA Sample Preparation Module
2. Preparation of qPCR Tube
3. Contrive Blood Specimen
Note: If a genuine test specimen is being prepared, proceed to step 4.
4. Perform FINA Extraction
5. qPCR Reaction
The workflow for extracting proviral DNA from whole blood spiked with 8e5-LAV cells is shown in Figure 1. Figure 2 shows the FINA sample module and prepared qPCR tube. This method allows for efficient amplification of HIV-1 provirus from 8e5-LAV cells at different copy numbers, as shown in the standard curve of contrived specimens (Figure 3). PCR had an efficiency of 103%, as calculated from Efficiency = -1+10(-1/slope). Equation of the line was y= -3.25x + 27.95, with a correlation of R² = 0.996. The HIV proviral DNA standard curve replicates also have highly reproducible amplification, as evidenced in Table 1. Additionally, an internal control of 500 copies Hydroxypyruvate Reductase is present to show that there is no PCR inhibition resulting from extracting blood with FINA, independent of the number of copies of HIV-1 provirus present (Figure 4). IC amplification was efficiently obtained in all 18 samples tested, with an average Cq of 21.92 ± 0.25 and a CV of 1%.
Figure 1: Workflow for detecting proviral DNA from whole blood spiked with 8e5-LAV cells to qPCR. (A) From left to right, the prepared FINA module, after blood is added to the capture membrane and after wash buffer was added. (B) qPCR tubes with capture disks stuck to double-coated tape with qPCR master mix added. Please click here to view a larger version of this figure.
Figure 2: FINA in-house module and qPCR tube preparation. (A) An 8.35-mm diameter capture membrane disk was sandwiched between a square 707 blotting pad and a thin sheet of paraffin tape containing a 7.14-mm-diameter hole in the center such that the hole of the paraffin tape was centered on the capture disk for the direct application of lysed blood by pipet. (B) A 5.1 mm double-coated polyester diagnostic tape is stuck on side of 200 µl qPCR tube. The second liner of tape is removed to expose sticky surface for applying capture disk when ready. Please click here to view a larger version of this figure.
Figure 3: Standard curve of HIV-1 proviral DNA contrived specimens. (A) Amplification plots obtained from 100 µl of 4 replicates HIV-1 negative whole blood spiked with 4,000, 400, 40 and 10 8e5-LAV cells with 500 copies/reaction of the internal control Solid lines = amplification plots; Dotted line = threshold. Y-axis is fluorescence units and X-axis is qPCR cycle number. (B) Standard curves of Cq values calculated from amplification plot versus log copy number of 8e5-LAV cells per 100 µl whole blood. Equation of the line: y = -3.25x + 27.95; R² = 0.996; PCR efficiency = 103.23% calculated via Efficiency = -1+10(-1/slope) Please click here to view a larger version of this figure.
Figure 4: Internal control indicates no qPCR inhibition. 500 copies Hydroxypyruvate Reductase Amplification control plasmid added per reaction. Solid lines = amplification plots; Dotted line = threshold. Average Cq of IC = 21.92 ± 0.25. Cqs ranged from 21.21 to 22.32. Coefficient of variation (CV %) = 1%; N = 18. Please click here to view a larger version of this figure.
8e5-LAV cells/ | |||
100 µl WB | Ave. Cq | SD | CV (%) |
4,000 | 16.27 | 0.14 | 1 |
400 | 19.54 | 0.15 | 1 |
40 | 22.56 | 0.3 | 1 |
10 | 24.83 | 0.15 | 1 |
Table 1: Average Cq, standard deviation (SD) and coefficient of variation (CV %) of 4,000, 400, 40 and 10 copies of 8e5-LAV standard curve with FINA extraction and HIV-1 specific primers and probes. N = 4. WB = whole blood.
EID linked to rapid treatment access has been demonstrated to reduce infant mortality due to HIV infection 16. Because of the persistence of maternal antibodies in an infant's blood, rapid HIV antibody tests have limited utility in determining the status of HIV-exposed infants. The WHO recommends that all infants born to HIV-1 positive mothers should be tested at 4-6 weeks of age, using a virological test 17. We have reported the development of an assay for the detection and quantitation of HIV-1 proviral DNA in whole blood specimens that is capable of single copy detection with demonstrated 100% sensitivity and specificity in a small field evaluation of 61 South African infants 11. One hundred microliters or more of whole blood can be collected via finger or heel stick and processed via the FINA modules 18,19. These blood samples can then be stored for at least one month before amplification making this method ideal for testing infants that are far from laboratories.
In the study presented in Figure 3, a standard curve of contrived specimens from HIV-1 negative whole blood spiked with thawed cells harboring HIV-1 proviral DNA is depicted. A caveat to this study is that some of the cells added to the sample could have already lysed due to the freeze/thaw, which could have potentially improved the detection level of the target. A more stringent experimental design would have been to have contrived the specimen using freshly cultured cells which would have more accurately mimicked whole blood.
The FINA sample prep method was designed for incorporation into an EID POC qPCR device (under development) for use in limited resource settings 8; however, the manual format described here can also be a useful addition to a laboratory's sample prep portfolio. FINA does not use the bind, wash and elute strategy used in dried blood spots and other paper-based extraction systems that dilutes the extracted nucleic acids, possibly leading to sub-optimal detection sensitivity 5. The FINA system is highly flexible and can be readily adapted for other genetic tests besides proviral HIV DNA detection. Lower volumes of blood or other biological specimens can be processed for abundant targets and higher volumes for less abundant ones. In our original embodiment of the FINA system, we capture the cells from whole blood using a blood separation membrane and then lyse them by the addition of 10 mM NaOH 9. This version has the advantage of one fewer user steps, but the cell number/blood volume that can be processed is smaller. In addition to the bound glass membrane used in this report, other blood separation or blood collection filter papers have been used successfully with this method, and several different quantitative PCR instruments have also been shown to amplify template embedded in the filter disk 9. The only stipulation is that the filter does not block the fluorescence reading of the real-time PCR instrument.
A significant limitation of FINA extraction is that there is a size constraint on the diameter of the capture membrane. A disk with a membrane larger than 9 mm cannot be placed in a 200 µl qPCR tube without the membrane overlapping in the tube which limits the surface area exposed to the reaction mix. Additionally, the larger the disk diameter, the more qPCR reaction volume is required to cover the filter which increases the cost and turn-around-time of the assay. If the filter is too small to efficiently capture the DNA in the specimen, it may clog preventing efficient washing.
Using the bound glass capture membrane, we were able to capture only HIV-1 proviral DNA and not viral RNA from whole blood samples (data not shown). We do not know if alternative membranes or buffers could be used to promote isolation of RNA instead of or in addition to DNA. If the reader would like to only isolate DNA from the specimen this protocol does not require the enzymatic removal of RNA required with some protocols. Detecting both HIV-1 RNA and DNA in clinical samples would improve the sensitivity of EID and this lack of RNA detection is a limitation of our diagnostic test.
The most critical step in adapting the FINA process to a new assay is to validate the size of the filter to accommodate the amount of DNA, i.e., the number of cells, in the specimen to be tested. In general FINA extraction works best with specimens with smaller numbers of cells so that smaller capture disks can be used. To determine the binding capacity of the filter, simply add a second filter disk to the FINA module stacked below the collection membrane. Add the sample and wash the membranes as in steps 4.1-4.5. Place each of the DNA capture disks into separate qPCR tubes and amplify. Using a standard curve, determine the percent of input DNA captured on each filter. The size of the filter can be optimized to the needs of the assay. For the FINA proviral DNA assay described here, more than 99% of human genomic DNA is captured when cell lysate is added to the filter (unpublished observations) enabling superior detection capabilities.
In addition to EID, future applications of FINA could include the detection and quantitation of proviral DNA as a biomarker for HIV-1 disease monitoring with patients who have achieved viral suppression due to antiretroviral therapy. Treatment monitoring aimed at HIV eradication could be performed by screening other inputs besides peripheral blood for reservoirs of HIV-1 infected cells. HIV proviral testing could also be used to screen vaccine trial subjects for true infection. The blood samples are stable during storage on FINA modules and can be collected under field conditions for later shipment to central laboratories for PCR analysis making this method ideal for epidemiological studies. In addition to HIV-1 detection, this highly flexible method could also be used to amplify precision medicine targets found in biological specimens such as pharmacogenetics screening or other SNP detection or for rapid genotyping of animal models. Isothermal amplification methods such as helicase dependent amplification (HDA) or loop-mediated isothermal amplification (LAMP) could also be used with FINA extracted template or the DNA-embedded on the filter could be used as template for conventional PCR amplification.
The authors have nothing to disclose.
This protocol development was supported by the Bill and Melinda Gates Foundation Grand Challenges in Global Health grant 37774. Real-time PCR reagents and advice were provided by Abbott Molecular Inc. Des Plaines, IL. 8E5-LAV cells were provided by the Virology Quality Assurance Laboratory, Rush Presbyterian; St. Luke’s Medical Center. HIV-1 negative blood was provided by Core Lab, NorthShore University HealthSystems, Evanston, IL. Thanks to Mark Fisher for photography help.
Fusion 5 | GE Healthcare Life Sciences | 8151-9915 | |
707 blotting pad | VWR International | 28298-014 | |
PARAFILM M | VWR International | 52858-000 | |
3M Double-Coated Polyester Diagnostic Tape | 3M Medical Specialties | 9965 | |
200 µl qPCR strip tubes | Agilent | 401428 | |
optical strip caps | Agilent | 401425 | |
Mx3005p qPCR System | Agilent | 401456 | |
sodium hydroxide | Sigma-Aldrich | 221465 | A.C.S. Reagent |
Triton X-100 | Sigma-Aldrich | T9284 | BioXtra |
dimethyl sulfoxide | Sigma-Aldrich | D8418 | for molecular biology |
fetal bovine serum, certified, U.S. origin | Thermo Fisher Scientific | 16000-044 | |
Hammer driven small hole punch 3/16" hole diameter (5.1 mm) | McMaster Carr | 3424A16 | |
Hammer driven small hole punch 1/4" hole diameter (7.14 mm) | McMaster Carr | 3424A19 | |
Hammer driven small hole punch 5/16" hole diameter (8.35 mm) | McMaster Carr | 3424A23 |