Abstract
Frequent, accurate, and highly sensitive HIV-1 viral load monitoring is a critical component of AIDS antiretroviral
therapy, a tool for reducing the incidence of mother-to-child HIV transmission, and a required element of routine
diagnostic testing to make people aware of their HIV status. Although enormous research and product
development effort has been applied to point-of-care viral load testing, the current paradigm of nucleic acid tests
and antigen assays continues to demonstrate fundamental limitations that derive from their inherent complexity
and lack of robustness, which in turn impact their costs and practicality for adoption in resource-limited settings.
We seek to address an important gap in the capabilities of existing technologies through a combination of three
innovations to yield an integrated, rapid, simple, ultrasensitive, highly selective, robust, and inexpensive system
for quantitative viral load measurement. First, we utilize microfluidic separation of virions from whole blood,
yielding a 10-50 µl plasma sample from 20-100 µl of whole blood in <10 min, with >95% virus extraction
efficiency. Second, we will achieve ultraselective recognition of intact HIV virions from the resulting serum using
designer DNA nanostructures that take the form of a macromolecular “net” whose vertices are a precise
mechanical match to the spacing and positioning of the spike gp120 protein matrix displayed on the HIV outer
surface. The DNA net vertices incorporate nucleic acid aptamer probes that have been selected for selectively
targeting the HIV gp120, resulting in multiple sites of high affinity attachment, and thus the “net” can be used as
an effective capture probe when covalently attached to a photonic crystal biosensor surface. Finally, we will
utilize a newly-invented form of biosensor microscopy called Photonic Resonator Interference Scattering
Microscopy (PRISM) in which the photonic crystal surface amplifies laser light scattering from captured intact
virions, enabling each one to be counted with high signal-to-noise ratio. Because PRISM does not require labels
or enzymatic amplification, our approach enables dynamic, real-time counting of captured virus with digital
precision and ultrasensitivity. In the proposed project, we will integrate viral separation and the photonic crystal
biosensor into a plastic cartridge and develop a rapid workflow that will be simple and rapid for compatibility with
point-of-care settings, with the goal of yielding a result in <30 minutes sample-to-answer. Our Aims include
development of a point-of-care version of the PRISM instrument, and statistically robust characterization of
detection limits, repeatability, and robustness. Our study will conclude with validation of the system using clinical
specimens and direct comparison against gold-standard laboratory RT-PCR analysis.
