Branched DNA (bDNA) Technology

Introduction

The branched DNA (bDNA) assay provides a unique and powerful tool for reliable quantification of nucleic acid molecules. Fundamentally different from target amplification methods such as PCR, the bDNA assay directly measures nucleic acid molecules at physiological levels by boosting the reporter signal, rather than replicating target sequences as the means of detection, and hence avoids the errors inherent in the extraction and amplification of target sequences. The bDNA assay employs linear signal amplification using synthetic oligonucleotide probes and bDNA molecules, and can accurately and precisely measure between approximately 500 and 10,000,000 molecules. New advances in bDNA technology include the addition of preamplifier molecules and the incorporation of novel nucleotides, isoC and isoG, into oligonucleotide probe sequences to further enhance signal and reduce noise caused by nonspecific hybridization of bDNA assay components. These improvements have extended the quantitative detection limit of the bDNA assay to as low as 50 molecules.

History of bDNA technology

Well suited to routine use in a clinical or research setting, the bDNA assay has been applied successfully in a number of areas, including the prognosis and monitoring of patients with viral diseases. Providing a reliable means for direct quantification of viral load in clinical specimens, bDNA assays have been developed to measure hepatitis B virus DNA, hepatitis C virus RNA), human immunodeficiency virus type 1 RNA, and cytomegalovirus DNA. With the custom design of oligonucleotide probes, the potential applications of the bDNA assay reach far beyond viral nucleic acid quantification. By creating oligonucleotide probes for specific sequences of target nucleic acid molecules, the bDNA assay has been adapted to a wide variety of applications. For example, researchers have designed probes for the bDNA assay to measure cellular mRNA levels. This has proven to be a fruitful approach for a number of research applications, including monitoring changes in cytokine mRNA levels in healthy and immunocompromised patient populations, investigation of insulin splicing patterns, and evaluation of stress gene induction for molecular toxicology applications. Given its versatility, ease of use and high level of performance, the bDNA assay rapidly is becoming the method of choice for nucleic acid quantification

Outline

The bDNA assay uses a 96-well microplate format, and is based on a series of specific hybridization reactions and chemiluminescent detection of hybridized probes. Attached to the surface of each microwell are “capture” probes that contain a specific nucleotide sequence. These capture probes bind to a subset of “target probes” which are bound to specific nucleotide sequences in the target nucleic acid molecule. This series of hybridizations anchors the target nucleic acid molecule to the microwell surface. Detection of the target nucleic acid and amplification of the signal is accomplished through another series of hybridizations.

A second subset of target probes links the target nucleic acid molecule to bDNA amplifier molecules. With the addition of preamplifier molecules to provide an additional layer of enhancement between the target probes and the bDNA amplifier molecules, even greater signal amplification can be achieved. Each bDNA amplifier molecule has been designed to contain 15 arms, each of which contains three binding sites for alkaline phosphatase-conjugated “label probes”. A chemiluminescent signal is generated upon introduction of a dioxetane substrate which is activated by the alkaline phosphatase. This signal is easily quantified by counting the number of photons emitted in a luminometer. The bDNA assay is inherently quantitative since the number of photons emitted is directly related to the amount of target nucleic acid in the specimen.

Materials

Equipment – Chiron plate heater equipped with 8×12 microwell holder – Chiron plate-reading luminometer

Reagents for day 1

  • Oligonucleotide-modified microwells
  • Lysis diluent
  • Lysis reagent (proteinase K)
  • Target probes for capture
  • Target probes for label
  • Specimens
  • Standards and controls (optional)

Reagents for day 2

  • bDNA amplifier concentrate
  • Label diluent
  • Label probe concentrate
  • Dioxetane substrate (LumiphosPlus)
  • 10% sodium dodecyl sulfate (SDS)
  • Wash A (0.015 M NaCl, 0.0015 M sodium citrate, 0.1% SDS, 0.05% sodium azide, 0.05% Proclin 300
  • Wash B (0.015 M NaCl, 0.0015 M sodium citrate, 0.05% sodium azide, 0.05% Proclin 300)

Procedure

Day 1

  1. Prepare Specimen Working Reagent by combining:
  • 6 mllysis diluent
  • 600 ml lysis reagent
  • 32 ml 500 fm/ml target probes for capture (20 fm/200 ml final)
  • 96 ml500 fm/ml target probes for label (60 fm/200 ml final)

Note: The actual concentrations of target probes will vary depending on the target nucleic acid molecule.

  1. For plasma or serum specimens, pipette 150 ml Specimen Working Reagent and 50ml serum or plasma into duplicate oligonucleotide-modified microwells. For cell or tissue specimens, prepare and extract RNA pellets in Specimen Working Reagent as described, and pipette 200 mlof each RNA extract into duplicate oligonucleotide-modified micro wells.
  2. If desired, positive and negative controls can be included for specificity purposes. Controls should be treated in the same manner as described for specimens in step 2 (above) and run in duplicate.
  3. If desired, standards can be included for absolute quantification. Pipette 150 ml of Specimen Working Reagent and 50 ml of the appropriate standard curve member into duplicate oligonucleotide-modified microwells.
  4. Seal the microwells with Mylar film and incubate the microwells overnight at 53°C in the Chiron plate heater. (Note: Temperature will depend upon the defined optimal temperature for the specific assay.)

Day 2

  1. Take the plate from the Chiron plate heater and let stand 10 min at room temperature. While the plate is cooling down, prepare the amplifier solution by diluting the amplifier concentrate at 200 fm/ml in label diluent. The final concentration should be 10 fm/50 ml.
  2. Wash the microwells twice with Wash A, then pipette 50 ml of diluted amplifier into each well. Incubate the microwells for 30 min at 53°C in the Chiron plate heater.
  3. Take the plate from the Chiron plate heater and let it stand 10 min at room temperature. While the plate is cooling down, prepare the Label Working Solution by diluting the label probe at 500 fm/ml into label diluent. The final concentration should be 20 fm/50 ml.
  4. Wash the microwells twice with Wash A, then pipette 50 ml of the Label Working Solution into each well. Incubate the microwells for 15 min at 53°C in the Chiron plate heater.
  5. Take the plate from the Chiron plate heater and let it stand 10 min at room temperature. During this time, prepare the Substrate Solution made of 99.7% v/v LumiphosPlus and 0.3% v/v 10% SDS (for example: 3 ml LumiphosPlus, 9 ml 10% SDS).
  6. Wash the microwells twice with Wash A and then three times with Wash B. Add 50 ml of the Substrate Solution into each well.
  7. Measure the chemiluminescence in the plate-reading luminometer. (This includes incubation of the microwells for 30 minutes at 37°C prior to measurement to reach steady-state kinetics).

Troubleshooting

  • Oligonucleotide-modified wells should be handled with care. Do not break apart well strips into segments. Seal the wells securely with a fresh plate sealer to prevent evaporation during incubations. When removing the plate sealer following incubations, use care not to pull the wells out of the microwell holder.
  • All specimens, standards and controls should be assayed in duplicate for optimal quantification. A void lipemic or turbid specimens as these may yield inconclusive results.
  • For optimal results, all reagents should be brought to the temperature indicated in the assay procedure before use. Add reagent to the microwells by touching the pipet tip to the wall near the midpoint of the well, above the surface of the fluid in the well.

Application

Viral quantification or viral load testing has become part of the routine management of patients infected with HIV-1 or hepatitis C virus (HCV). There are currently several molecular technologies that are available for use in the clinical laboratory setting. Of these, only the branched DNA (bDNA) assays are FDA-approved for HIV-1 and HCV viral load testing. This signal amplification technology is built on a series of hybridization reactions that are highly amenable to full automation and thus lessen the amount of labor required to perform this type of analysis.

Molecular diagnostic assays using bDNA technology for detection of nucleic acid target molecules are sensitive, specific, and reliable tools in the diagnosis of viral and bacterial infections and for monitoring disease progression during the course of therapy. bDNA tests have evolved from developmental stages in the research laboratory to US Food and Drug Administration–approved quantitative assays with valuable clinical applications. The bDNA assays are less labor-intensive than many molecular-based procedures because they are highly amenable to total automation. Using bDNA, amplification of a target sequence is not required, and, thus, cross-contamination between replicate samples due to excessive amplicons or carryover is less likely in bDNA assays. In addition, because bDNA is a signal amplification technology, the assay is able to quantify with less than a 0.5 log or 3-fold variability for its entire dynamic range.

bDNA technology has proved versatile because methods have been developed for the detection of infection by a wide range of microorganisms, including the parasite Trypanosoma

brucei, cytomegalovirus, antibiotic-sensitive and antibioticresistant Staphylococcus bacteria, human papillomavirus, and hepatitis B virus. However, more recent efforts have focused on the development of bDNA assays for the quantification of HIV-1 and hepatitis C virus (HCV) RNA, leading to the routine application of bDNA methods in the clinical molecular diagnostics laboratory. In describing the advancement of bDNA methods, this review emphasizes bDNA assays for HIV-1 and HCV.

First-Generation HIV-1 bDNA Assays

Accurate, reproducible first-generation bDNA assays were developed for the detection of HIV-1 RNA and HCV RNA in human plasma (Quantiplex HIV-1 or HCV RNA 1.0 assay, Bayer, Tarrytown, NY).9-11 In one of the first reports of the Quantiplex bDNA assay, no reactivity using plasma samples from seronegative donors was observed using the firstgeneration

bDNA assay, indicating excellent specificity.9 Positive results in the bDNA assay were observed in 83% of samples from 348 patients who were seropositive for HIV-1.9 The dynamic range for quantification using the Quantiplex HIV-1 RNA 1.0 assay extended from 104 (lower limit of detection) to more than 106 HIV RNA copies/mL.9,11 Changes in viral load of 2- to 3-fold were statistically significant using the bDNA test, indicating that the Quantiplex HIV-1 RNA 1.0 assay was highly accurate and reproducible.11 By comparison, changes in viral load of at least 3.7- to 5.8-fold were necessary before results were statistically significant using the earliest versions of the RT-PCR assay.11 Thus, bDNA became the method of choice for most clinical trials evaluating viral load testing and the clinical efficacy of new HIV-1 reverse transcriptase and protease inhibitors.

The sensitivity of the bDNA detection method was enhanced in the second-generation HIV-1 assay (Quantiplex HIV-1 RNA 2.0 assay, Bayer) by changing the design of the target and capture probes and by the addition of preamplifier oligonucleotides. The improved design of the target and capture probes allowed an increase in the stringency of hybridization, thereby decreasing the assay background. Preamplifiers dramatically increase signal intensity because each preamplifier molecule has multiple regions for hybridization to many bDNA molecules. In addition, each bDNA molecule has multiple, repeat sequences for hybridization of AP-labeled probes. The signal output using the Quantiplex HIV-1 RNA 2.0 assay showed linearity from approximately 500 copies/mL to 1.6 × 106 copies/mL (stated dynamic range was 500 to 8 × 105 copies/mL). The sensitivity of the second-generation HIV-1 bDNA assay was increased by 20-fold compared with the first-generation assay (lower limits of detection were 500 copies/mL and 10 × 104 copies/mL, respectively). In an extensive analysis of precision, initial test results and retest results in the Quantiplex HIV-1 RNA 2.0 assay were compared. The HIV-1 RNA 2.0 assay was found to be highly reproducible.14 Of 174 samples with viral loads of more than 5,000 copies/mL, 96% had differences of less than 0.3 log10 in copy number between initial results and retest results. Of 69 samples with viral loads between 500 and 5,000 copies/mL, 86% had less than 0.3 log10 differences between initial and retest results. However, among 5,339 patients who were testedin routine clinical testing during a 1-year interval, viral loads of fewer than 500 copies/mL were observed in 41.6% of samples.

Therefore, a bDNA assay with higher sensitivity (ie, lower limit of detection of <500 copies/mL) was needed for a substantial proportion of patients.

The performance characteristics of the Quantiplex HIV-1 RNA 2.0 assay and an RT-PCR test (Amplicor HIV-1 Monitor 1.0 assay, Roche Diagnostics, Indianapolis, IN) were compared using dilutions of standard samples that had known HIV-1 virus copy numbers. When dilutions of the same standards were tested in the 2 assays, HIV-1 copy numbers generally were higher from the RT-PCR assay than from the bDNA assay. For example, the ranges of RNA copy numbers were 900 to 7.68 × 105 copies/mL in the bDNA test and 3,360 to 1.88 × 106 copies/mL in the RT-PCR assay. Both assays were linear for the stated dynamic ranges.

Comparison of the slopes of HIV-1 copy number vs signal output regression lines suggested that the bDNA test had less proportional systematic error. Between-run variability, using a standard sample with 1,650 HIV-1 copies/mL, was lower in the bDNA test than in the RT-PCR assay; coefficients of variations for the bDNA and RT-PCR assays were 24.3% and 34.3%, respectively, indicating that the bDNA assay was slightly more precise at this HIV-1 RNA copy number. Compared with using the 1,650-copies/mL standard, the between-run coefficients of variation were higher for a sample containing 165 HIV-1 RNA copies/mL (44.0% and 42.7% for the bDNA and RT-PCR assays, respectively). Assay results were similar when equal HIV-1 copy numbers were compared across HIV subtypes A through F by using the bDNA test. However, by this version of RT-PCR, subtypes A, E, and F were detected less efficiently than the B, C, and D subtypes. The differences between the second-generation bDNA test and the Amplicor Monitor RT-PCR test for HIV-1 quantification indicated that consistency in testing method was required for each individual patient throughout diagnosis and treatment.

HCV bDNA Assays

While continuing to develop an improved third-generation HIV-1 bDNA assay, Bayer recognized the need to develop an HCV viral load assay with an emerging clinical usefulness similar to that for HIV-1 testing. For the detection of HCV, the first-generation bDNA assay (Quantiplex HCV RNA 1.0 assay, Bayer) had a dynamic quantification range in human plasma from 3.5 × 105 to 1.2 × 108 HCV RNA copies/mL. Genotypes 1 through 6 were detected by using this assay, although the sensitivity was lower for genotypes 2 and 3 (67% detection rate: positive signal for 60 of 89 serum samples known to contain HCV genotypes 2 or 3) compared with genotype 1 (97% detection rate: positive signal for 67 of 69 serum samples known to contain the HCV genotype 1). A comparison of the Quantiplex HCV RNA 1.0 assay and a research laboratory–developed RT-PCR test for HCV showed greater sensitivity in the RT-PCR test, which had a lower limit detection of 2.5 × 104 HCV RNA copies/mL. However, the HCV bDNA test had greater reproducibility and was less time-consuming than the laboratory-developed HCV RT-PCR test.

To improve the detection rate for HCV genotypes 2 and 3, a second-generation assay was developed (Quantiplex HCV RNA 2.0 assay, Bayer).15 The major design change in the second- generation HCV RNA assay was to use probes to sequences in the HCV genome that were more highly conserved across genotypes. These conserved regions were 5′ untranslated sequences and sequences in the core gene of the HCV genome. The result of changes to the target and capture probes was to dramatically reduce the variation in detection rate among HCV genotypes in the second-generation assay. Each of the 6 HCV genotypes had a high detection rate, and there was marked improvement in the detection of HCV genotypes 2 and 3 (detection of 93% vs 67% of samples known to contain HCV genotypes 2 or 3 in the HCV 2.0 and HCV 1.0 assays, respectively). Also, the sensitivity of the second-generation bDNA assay was slightly enhanced compared with the HCV 1.0 assay (lower limits of quantification were 2.0 × 105 [HCV 2.0] versus 3.5 × 105 [HCV 1.0]).

Third-Generation bDNA Assays for HIV-1 and HCV

In the first- and second-generation bDNA assays, nonspecific hybridization of oligonucleotide probes to nontarget sequences limited assay sensitivity. The critical technological improvement in the third-generation bDNA assay (Quantiplex [also referred to as VERSANT] HIV-1 RNA 3.0 assay, Bayer) was to use nonnatural bases, 5′-methyl-2′- deoxyisoguanosine (isoG) and 5′-methyl-2′-isodeoxycytidine (isoC), in the synthesis of all probes in the bDNA system, with the exception of capture extenders that mediate capture of the target viral nucleic acid to the plate surface. Because oligonucleotides containing isoG and isoC are not present in nature, nonspecific hybridization is reduced significantly. Thus, probes modified with the nonnatural bases do not form stable hybrids with the capture probe in the absence of target RNA. In the initial description of the thirdgeneration assay, the limit of detection of HIV-1 in plasma samples from 11 patients was 50 copies per mL.This represents a 10-fold improvement in the limit of detection compared with the second-generation assay. During treatment with highly active antiretroviral therapy, HIV-1 viral load decreased to below the limit of detection for all 11 patients.

The VERSANT HIV-1 RNA 3.0 assay has a dynamic range of 75 to 5 × 105 HIV RNA copies/mL. Version 3.0 also has been approved to a lower limit of detection equal to 50 copies/mL in several other countries. When matched samples were compared in the secondgeneration HIV-1 bDNA assay (Quantiplex version 2.0) and the Amplicor Monitor 1.0 RT-PCR assay, consistently lower HIV-1 copy numbers were obtained using the bDNA test. However, a close quantitative correlation in HIV-1 RNA copy numbers was observed between the third-generation (version 3.0) assay and the Amplicor Monitor 1.5 RT-PCR test.18 Viral load results in the version 3.0 bDNA assay and in the Amplicor RT-PCR assay were approximately 2-fold higher than in the version 2.0 assay. Quantitatively similar results in the version 3.0 bDNA test and the RT-PCR test are important in patient care because of the likely possibility that different methods will be used in testing samples from the same patient for the course of anti-HIV therapies. Recent data suggest that rebaselining for patients tested with an RT-PCR assay may not be necessary.

Similar to the HIV-1 version 3.0 bDNA assay, the thirdgeneration bDNA assay for HCV also used isoC- and isoG-substituted oligonucleotides to reduce nonspecific hybridization. The use of isoC- and isoG-substituted oligonucleotides increased assay sensitivity approximately 62-fold. The HCV RNA 3.0 assay lower limit of detection was 3.2 × 103 copies/mL compared with 2 × 105 copies/mL in the HCV RNA 2.0 assay. The dynamic linear quantification range of the HCV RNA 3.0 assay extended from 3.2 × 103 copies/mL (615 IU/mL) to 4 × 107 HCV RNA copies/mL (7.7 × 106 IU/mL). The HCV RNA 3.0 assay had a high specificity (98.2%) and, similar to the second-generation assay, was equally effective in the quantification of HCV RNA across all genotypes. Between-run and within-run standard deviations for replicate samples were 0.2 log10 and 0.14 log10, respectively, indicating that the third-generation assay was highly reproducible.

In addition to the eradication of HCV in serum, an important goal of anti-HCV therapies is to reduce HCV levels in the liver. The usefulness of the HCV RNA 3.0 assay for the detection and quantification of HCV RNA in liver biopsy specimens was studied in 25 patients coinfected with HCV and HIV.20 The reproducibility of the third-generation HCV bDNA assay was similar between liver biopsy specimens and serum samples. Also, detection of HCV RNA in liver specimens from patients infected with genotypes 1, 3, and 4 by the HCV RNA 3.0 assay was highly specific and sensitive. In this study of 25 patients, high pretreatment levels of intrahepatic

HCV correlated with a low frequency of response to anti- HCV therapy. Pretreatment intrahepatic HCV levels were highest in patients infected with HCV genotype 1. Results of this study demonstrated that important markers of HCV disease progression, HCV levels in liver and in serum, can be quantitated reliably by bDNA analysis during treatment. The methods for third-generation bDNA assays include sample preparation, hybridization, and signal detection forHIV-1 RNA and HCV RNA. All 3 s eps are performed in the microwells on the System 340 platform for the HCV RNA 3.0 assay that does not require a separate extraction step. In the version 3.0 HIV-1 RNA assay, sample preparation is different from the HCV RNA method and is performed outside the System 340 platform. A recent study evaluated an adaptation of the version 3.0 HIV-1 RNA method in which the HIV- 1 sample processing step was modified to accommodate simultaneous testing for HCV and HIV-1 on the System 340 platform. The HIV-1 method was modified by omitting the 2-hour incubation at 63°C for viral lysis. Instead, HIV-1 and HCV lysis was performed on the System 340 platform. The HCV bDNA test methods were unchanged in the combined assay. The specificity and quantification by the combined bDNA method were within specifications for the individual HIV-1 and HCV assays. Simultaneous testing for HIV-1 and for HCV improved workflow in the clinical laboratory and resulted in lower costs. Because HIV-1 RNA and HCV RNA detection and quantification are crucial for diagnosis and for the evaluation of responses to therapy, the ability to carry out simultaneous testing for both viruses represents a significant advance in molecular diagnostics.