Development of simple, rapid, and sensitive methods for detection of hepatitis C virus RNA from whole blood using reverse transcription loop-mediated isothermal amplification

Abstract

Hepatitis C virus (HCV) infection is an underdiagnosed global health problem. Diagnosis of current HCV infections typically requires testing for HCV RNA using high-complexity laboratory tests. Methods for the detection of HCV RNA that are simple, inexpensive, rapid, and compatible with use outside of a laboratory setting are very important in order to improve access to hepatitis C diagnostic testing and facilitate accelerated linkage to care. We developed and evaluated three simple workflows for extracting HCV RNA from small volumes of whole blood for use in a sensitive, pan-genotypic RT-LAMP assay. The water workflow uses osmotic stress to release HCV RNA and has a limit of detection of 4.3 log₁₀(IU/mL) (95% CI 4.0-4.9). The heat workflow uses a heating step to release HCV RNA and has a limit of detection of 4.2 log₁₀(IU/mL) (95% CI 3.8-5.1). The bead workflow, which uses chemical lysis of the sample and a streamlined paramagnetic solid phase reversible immobilization bead procedure for nucleic acid purification, has a limit of detection of 2.8 log₁₀(IU/mL) (95% CI 2.5-3.4). When used to test whole blood spiked with HCV RNA-positive plasma samples in which most HCV levels were below 5.0 log₁₀(IU/mL), the water, heat, and bead workflows detected HCV RNA in 69%, 75%, and 94% of samples, respectively. These workflows are compatible with visual lateral flow dipsticks, and each takes less than 60 min from sample to result. Each workflow can be performed with minimal and inexpensive equipment. With further procedural simplifications, these workflows may form the basis of assays for the point-of-care diagnosis of HCV infections.

Keywords: RNA detection; assay development; hepatitis C virus; point of care.

Fig 1

Workflows for extraction of HCV RNA from whole blood samples. We optimized and evaluated three workflows for preparing HCV RNA from whole blood samples for amplification by RT-LAMP. (A) In the water workflow, whole blood is diluted fourfold in water. After 5 min at room temperature, the diluted sample is added to the RT-LAMP mix at 10% of the final volume. (B) In the heat workflow, whole blood is spun in a mini-centrifuge for 20 s. The red-cell free portion is diluted fourfold in 66% QuickExtract DNA extraction solution and then heated at 98°C for 5 min. The sample is then added to the RT-LAMP mix at 10% of the final volume. (C) In the bead workflow, whole blood is spun in a mini-centrifuge for 20 s. Twenty microliters of the red-cell free portion is mixed with 170 µL of buffer AVL/SPRI bead solution. After 5 min, the sample tube is placed on a magnetic rack, and the liquid portion is removed and discarded. Without removing the tube from the magnetic rack, the beads are washed once with 195 µL of 70% ethanol. The beads are resuspended in 20 µL of water and placed back on the magnetic rack. The bead-free elution is added to the RT-LAMP mix at 40% of the final reaction volume.

Fig 2

Time of RT-LAMP reactions with HCV RNA-positive whole blood samples. Eighty plasma samples positive for HCV RNA were diluted 10-fold in EDTA whole blood and tested using the (A) water, (B) heat, or (C) bead workflows with RT-LAMP. The time from the start of the RT-LAMP reaction at which an increase in SYTO9 fluorescence was detected is indicated on the y-axis. Samples that did not have an increase in fluorescence by 50 min (dashed line) were classified as target not detected and assigned a time of 52 min for inclusion on these graphs. Samples are color-coded by HCV genotype/subtype.

Fig 3

Lateral flow visual detection of low-level HCV RNA from whole blood. Three replicate whole blood samples containing the lowest HCV RNA level detected 10/10 times in the limit of detection analysis were evaluated for use with lateral flow dipsticks using each of the three workflows with RT-LAMP. These HCV RNA levels were 4.7 log₁₀(IU/mL) for the water workflow (A), 4.2 log₁₀(IU/mL) for the heat workflow (B), and 2.8 log₁₀(IU/mL) for the bead workflow (C). The bi-thermal RT-LAMP reactions were run for 40 min for the water and heat workflows and 30 min for the bead workflow. These times were selected because they were the shortest that allowed for reliable detection of low-titer samples. Lateral flow was stopped 5 min after the insertion of the dipstick into the sample tube. Due to the nature of the lateral flow detection, the intensity of the control line is inversely proportional to the intensity of the test line. neg, HCV RNA-negative plasma diluted in whole blood.

Fig 4

Bead workflow with magnetic wand processing. (A) A magnetic wand for purifying HCV RNA using paramagnetic SPRI beads was created using a 0.2-mL PCR tube, a 2-mL pipette tip, and a spherical magnet. (B) The magnetic wand was used to remove the beads from the lysed sample solution, dip the beads into the wash solution, transfer the beads to the elution solution, and remove the beads from the nucleic acid eluate. (C) Twenty HCV RNA-positive plasma were diluted 10-fold in EDTA whole blood and tested using the bead workflow with a magnetic wand replacing the magnetic rack and the pipetting steps for the wash and elution. The time from the start of the RT-LAMP reaction at which an increase in SYTO9 fluorescence was detected is indicated on the y-axis. Samples that did not have an increase in fluorescence by 50 min (dashed line) were classified as target not detected and assigned a time of 52 min for inclusion on these graphs. Samples are color-coded by HCV genotype.

Fig 5

Bead workflow with magnetic wand processing. (A) A magnetic wand for purifying HCV RNA using paramagnetic SPRI beads was created using a 0.2-mL PCR tube, a 2-mL pipette tip, and a spherical magnet. (B) The magnetic wand was used to remove the beads from the lysed sample solution, dip the beads into the wash solution, transfer the beads to the elution solution, and remove the beads from the nucleic acid eluate. (C) Twenty HCV RNA-positive plasma samples diluted 10-fold in EDTA whole blood and tested using the bead workflow with a magnetic wand replacing the magnetic rack and the pipetting steps for the wash and elution. The time from the start of the RT-LAMP reaction at which an increase in SYTO9 fluorescence was detected is indicated on the y-axis. Samples that did not have an increase in fluorescence by 50 min (dashed line) were classified as target not detected and assigned a time of 52 min for inclusion on these graphs. Samples are color-coded by HCV genotype.