High-fidelity and iterative affinity extraction of hyaluronan

Abstract

The glycosaminoglycan hyaluronan (HA) serves a variety of crucial physiological functions in vertebrates. Synthesized at the plasma membrane and secreted into the extracellular environment, HA polymers span a wide range of molecular weights (MW) that define their activity through a notable size-function relationship. Analytical technologies for determining HA MW distributions typically require selective extraction from complex biofluids or tissues. A common method for achieving this is immunoprecipitation-like pull-down using specific HA-binding proteins bound to magnetic beads. Here, we present a systematic investigation of experimental variables involved in this process, leading to an affinity extraction protocol that enables iterative bead reuse and reagent lifetime maximization, thereby enhancing the efficiency of the HA extraction process. Our methods provide a framework for general optimization of immunoprecipitation in other contexts with heterogenous analyte sizes.

Keywords: biomarker; hyaluronan; microfluidics; nanopore; nanosensing.

Figure 1

(A) Schematic of the HA extraction protocol. VG1 beads and sample of HA (i) are mixed to capture a subset of HA (ii). HA‐bound beads are magnetically separated from unbound HA (iii) and are then incubated with high‐molarity LiCl to elute captured HA (iv) for subsequent analysis. (B) SSNP MW distribution boxplot and histogram for polydisperse HA before (Control: N = 1472 events; median 270 kDa, IQR 169–432 kDa) and after extraction (Salt elution: N = 474 events; median 296 kDa, IQR 175–477 kDa). HA, hyaluronan; IQR, interquartile range; SSNP, solid‐state nanopore.

Figure 2

(A) SSNP event rates measured for iterative extractions of polydisperse HA including extracted HA (Cycles 1–3; 2.96 ± 0.39 s⁻¹, 2.90 ± 0.49 s⁻¹, 3.07 ± 0.34 s⁻¹, respectively) and no‐HA blanks (Blanks 1–3; 0.15 ± 0.04 s⁻¹, 0.08 ± 0.04 s⁻¹, 0.02 ± 0.01 s⁻¹, respectively). (B) SSNP MW distribution boxplots and histograms for extracted HA samples. Cycle 1: N = 1985 events; median 354 kDa, IQR 221–618 kDa. Cycle 2: N = 728 events; median 335 kDa, IQR 206–543 kDa. Cycle 3: N = 1285 events; median 366 kDa, IQR 214–672 kDa. HA, hyaluronan; IQR, interquartile range; SSNP, solid‐state nanopore.

Figure 3

(A) Comparison of SSNP event rates measured for two sets of polydisperse HA extractions performed iteratively across 23 days: HA extracted two times (light) and five times (dark). Left to right: 1.11 ± 0.13 s⁻¹, 1.94 ± 0.43 s⁻¹, 2.15 ± 0.15 s⁻¹, 1.12 ± 0.11 s⁻¹, 1.70 ± 0.16 s⁻¹, 1.69 ± 0.13 s⁻¹, and 1.67 ± 0.32 s⁻¹. (B) SSNP MW distribution boxplots and histograms for HA extracted twice in the timespan, on Day 0 (N = 1518 events; median 371 kDa, IQR 158–847 kDa) and Day 23 (N = 1326 events; median 413 kDa, IQR 218–788 kDa). (C) SSNP MW distribution boxplots and histograms for HA extracted five times in the timespan, on Day 0 (N = 474 events; median 296 kDa, IQR 175–477 kDa), Day 2 (N = 1370 events; median 280 kDa, IQR 158–457 kDa), Day 9 (N = 1079 events; median 299 kDa, IQR 147–528 kDa), Day 16 (N = 1413 events; median 295 kDa, IQR 150–564 kDa), and Day 23 (N = 1307 events; median 256 kDa, IQR 128–529 kDa. HA, hyaluronan; IQR, interquartile range; SSNP, solid‐state nanopore.

Figure 4

(A) SSNP event rates for HA extracted at 2–8°C with 15‐min incubation steps. Technical replicates A–C: 1.33 ± 0.24 s⁻¹, 1.24 ± 0.18 s⁻¹, and 1.66 ± 0.19 s⁻¹. (B) SSNP MW distribution boxplots and histograms for the control polydisperse HA (N = 1472 events; median 270 kDa, IQR 169–432 kDa) and three replicates of the extracted HA: A (N = 1040; median 267 kDa, IQR 153–448 kDa), B (N = 1214 events; median 258 kDa, IQR 153–429 kDa), and C (N = 1184 events; median 251 kDa, IQR 155–428 kDa). HA, hyaluronan; IQR, interquartile range; SSNP, solid‐state nanopore.

Figure 5

(A) Exploded view of the microfluidic HA device layers. (B) Photograph of the assembled microfluidic device with tubing connected and magnets positioned above and below the center chamber. (C) Schematic representation of the assembled microfluidic device. Side view (above) shows magnetic trap surrounding a channel. Insets (top) show zooms of bead field trapped in the magnetic field (left), the porous network they form (center), and the exposed VG1 elements for capturing HA (right). Top view (below) shows main channel for analyte capture/release as well as inlets/outlet and magnet footprint. (D) SSNP MW distribution boxplot and histogram for control polydisperse HA (N = 1472 events; median 270 kDa, IQR 169–432 kDa) and polydisperse HA extracted with the microfluidic device (N = 1245 events; median 188 kDa, IQR 109–355 kDa). HA, hyaluronan; IQR, interquartile range; SSNP, solid‐state nanopore.