Search

Early and accurate detection of bacterial infections can help save lives, prevent the spread of disease, and decrease the overuse of antibiotics. Our team at the Los Alamos National Laboratory has developed novel assays to detect bacterial biomarkers from patient blood at the point-of-care in order to facilitate a universal diagnostic platform. However, these biomarkers are amphiphilic in nature, and this biochemical property causes them to be sequestered by high-density and low-density lipoproteins (HDL and LDL) in the host’s blood. Extraction of the bacterial biomarkers from the lipoprotein complexes is thereby required for the development and deployment of a diagnostic platform. Accordingly, our team has developed a sample processing protocol to extract the biomarkers of interest; however, this procedure requires multiple pipetting, mixing, and centrifugation steps that must be performed by hand in a well-equipped laboratory. It also utilizes several chemical reagents, including chloroform and methanol, as well as potentially-infectious human blood. For use in resource-poor settings with minimally-trained personnel, sample processing should be automated in order to guarantee proper treatment of the sample and safety of the staff involved. Presented in this work is the development of a semi-automated microfluidic sample processing platform for the extraction of amphiphilic bacterial biomarkers. A fully automated system would require two phase separation steps and a re-suspension step to be integrated into a microfluidic platform. Herein, a novel cross-flow filtration scheme was designed to achieve phase separations on a microfluidic device. The devices were fabricated using a combination of laser-based microfabrication and lamination methods. We have taken a stepwise approach to determine the optimal combination of membrane material, membrane pore size, fabrication methods, and geometric design parameters that result in consistent performance. To demonstrate proof of principle, two major phase separations were performed: blood/serum separation, and biomarker extraction. Device materials and surface chemistry were determined to be suitable for this application, as measured by biomarker retention experiments. This study provides the groundwork for a potentially fully-automated sample processing platform for amphiphilic biomarker extraction from whole blood. The cross-flow filtration platform is a promising design for the complete automation of sample processing because it requires only 90µL of whole blood, is modular, and does not interfere with amphiphile detection. It is simple to manufacture, disposable, pump-free, and does not require the dilution of blood.