Nucleic acid amplification tests (NAATs) are powerful tools for infectious disease diagnostics. While NAATs are
routinely used in the clinic, their use in point-of-care (POC) contexts is constrained by complex procedures
needed to extract and purify nucleic acids from patient samples. This is especially troublesome for RNA
targets, due to the fragility of RNA and abundance of RNA-degrading ribonucleases in samples and the
environment. Nevertheless, RNA viruses are key targets for POC diagnostics, as they are abundant in low-
resource settings (e.g. low-income countries, where roughly 20 million HIV patients live) or in settings where
fast turnaround and patient self-testing are useful (e.g. outbreaks of airborne viruses such as influenza and
coronavirus). Thus, simpler approaches to RNA sample preparation are needed in POC-NAAT contexts.
Much of the complexity of NAAT sample preparation stems from the paradoxical need to add chaotropes to
extract nucleic acids, then remove chaotropes before amplification. Chaotropes, such as guanidinium
thiocyanate (GuSCN), are chemical denaturants that disrupt the structure of biological macromolecules. They
are used in NAAT sample preparation to lyse target virions/cells and denature inhibitors, such as ribonucleases
and proteases. Chaotropes are effective, but also inhibit polymerase activity, so they must be removed before
amplification. Many POC adaptations of NAAT workflows involve robotic or microfluidic automation of
chaotrope addition and removal, but still require specialized equipment and/or laboratory resources.
We will address the POC-NAAT sample preparation bottleneck in a different way: instead of automating
chaotrope removal, we will eliminate the need for it. We propose to engineer a polymerase to be
“chaostable”, or active in a chaotropic amplification buffer, enabling simultaneous extraction,
amplification, and detection of viral RNA targets in a single tube. To realize this vision, we propose three
specific aims, using a starting polymerase previously developed by our lab and HIV-1 as a model RNA target.
Aim 1: Develop a chaostable polymerase via compartmentalized self-replication in chaotropic
conditions. We will use a high-throughput directed evolution approach to develop a polymerase that retains
activity in 3M GuSCN (the minimum recommended concentration for RNA extraction buffers).
Aim 2: Investigate mechanisms of polymerase chaotrope resistance via deep mutational scanning and
molecular dynamics simulations. We will use deep mutational scanning and molecular dynamics simulations
to study interactions of GuSCN on our polymerase and identify rational design methods for GuSCN resistance.
Aim 3: Incorporate chaostable polymerase into a proof-of-concept HIV diagnostic that performs sample
lysis, RT-LAMP amplification, and colorimetric detection in a single tube. Using the best-performing
chaostable polymerases developed in Aims 1 and 2, we will design a single-tube RT-LAMP assay for HIV, and
test it against pure HIV RNA and HIV patient plasma samples.