Self-Driven “Microfiltration” Enabled by Porous Superabsorbent Polymer (PSAP) Beads for Biofluid Specimen Processing and Storage

A remote collection of biofluid specimens such as blood and urine remains a great challenge due to the requirement of continuous refrigeration. Without proper temperature regulation, the rapid degradation of analytical targets in the specimen may compromise the accuracy and reliability of the testing results. In this study, we develop porous superabsorbent polymer (PSAP) beads for fast and self-driven “microfiltration” of biofluid samples. This treatment effectively separates small analytical targets (e.g., glucose, catalase, and bacteriophage) and large undesired components (e.g., bacteria and blood cells) in the biofluids by capturing the former inside and excluding the latter outside the PSAP beads. We have successfully demonstrated that this treatment can reduce sample volume, self-aliquot the liquid sample, avoid microbial contamination, separate plasma from blood cells, stabilize target species inside the beads, and enable long-term storage at room temperature. Potential practical applications of this technology can provide an alternative sample collection and storage approach for medically underserved areas.


Additional
. Mercury intrusion test of porous PSAP beads prepared with different PEG contents.

Preparation and Characterization of Liquid Samples (1) Glucose Samples
Glucose was dissolved in saline or synthetic urine medium to reach a concentration of 5 mM. Bovine plasma already contained ~5 mM of glucose, and there would be no need for an extra dosage. The glucose concentration was detected by the colorimetric method using EnzyChrom glucose assay kit II (BioAssay Systems, Hayward, CA) 1 . In general, 20 µL sample was added to a 1 mL plastic cuvette containing 80 µL of working reagent and 900 µL of water, which was mixed well and incubated at room temperature for 20 min. The absorbance of the reaction mixture at 340 nm was proportional to the glucose concentration in the sample, and the coefficient was obtained S4 from the slope of the standard curve. The effective concentration range for this glucose assay was 0.1 to 3 mM, and any sample beyond this range required appropriate dilution.

(2) Catalase Samples
The catalase stock solution was prepared by dissolving catalase powder in saline medium (5 mg mL −1 ) and then filtered by 0.2 µm syringe filter (VWR International, Radnor, PA) to remove any undissolved particles or potential microorganisms. Then, the catalase stock was diluted 10 times by saline or biofluid media (synthetic urine or bovine plasma) to reach a final activity at ~800 U mL −1 . Since catalase exited in red blood cells, bovine blood didn't require a dosage for catalase. The activities of all catalase samples were determined by a standard spectrophotometric method 2 . Specifically, the catalase sample was first diluted to the desired detection range of ~50 to 100 U mL −1 . Next, 0.1 mL of the catalase sample was added to a 3 mL quartz cuvette containing 2.9 mL of hydrogen peroxide solution (0.04 wt% of H2O2 in the phosphate buffer) and mixed immediately by inversion. The decrease in absorbance of the mixture was monitored by a UV-Vis spectrophotometer (Cary 50, Santa Clara, CA) at 240 nm for 1 min and analyzed using a linear regression model. Three replicates were performed for each sample. The catalase activity in the 0.1 mL sample (c, U mL −1 ) was determined by the reduction rate of absorbance (k, min −1 ): in which 1.0 µmole min −1 U −1 was the decomposition rate of H2O2 by catalase, 43.6 M −1 cm −1 was the molar extinction coefficient for H2O2 at 240 nm, 1.0 cm was the light path, and 30 was the dilution ratio.

(3) Bacteriophage MS2 Samples
Bacteriophage MS2 was cultured with the host E. coli (ATCC 15597) in a TS broth at 35°C for ~6 hours. The TS broth was subsequently centrifuged at 4000 rpm for 5 min, and the supernatant was filtered through a 0.2 µm filter to remove the residual E. coli cells. The purified MS2 was diluted and dispersed in saline or biofluid media (synthetic urine or bovine plasma) to achieve a concentration of ~1000 PFU mL −1 . The concentration of MS2 was subsequently quantified by the double agar layer method with three replicates for each sample 3 .

(4) E. coli Samples
E. coli cells (ATCC 10798) were cultured in the LB broth to log phase (35°C overnight) and harvested by centrifugation at 4000 rpm. After washing with normal saline three times, the E.
coli suspension was diluted and dispersed in the saline medium to achieve the desired S5 concentration of ~500 CFU mL −1 . The concentration of live E. coli was measured using a standard spread plating method with three replicates for each measurement 4 .

Calculation of Recovery or Rejection Efficiency (1) Recovery Efficiency
The recovery efficiency of the target species (e.g., glucose, catalase, or bacteriophage MS2) was determined by comparing the concentration or activity of the target species released from the PSAP beads (ct) and in the original liquid sample (c0). In brief, 10 PSAP beads were applied to treat 1 mL of the liquid sample. The 10 PSAP beads were taken out until swelling equilibrium achieved and then weighted (m). Next, the hydrated beads in a clean tube were immersed by 2 mL of added water (mwater) and then broken by 5 s ultrasonication using a probe sonicator (Qsonica Q125, Newtown, CT) at 75% of amplitude. Thus, the dilution factor (DF) for the target species was calculated: The dilution factor could be adjusted by changing the amount of added water (mwater). After releasing the target species from the hydrated beads, the concentration or activity of the target species in the well-mixed suspension was measured (ct) to analyze the recovery efficiency for each target species: For the recovery efficiency calculation, the weight of the dried beads was negligible due to the high swelling ratio of the beads. Another assumption to simply the calculation was the density change during the dilution process was ignored, i.e., the density of the suspension was equal to that of the hydrated beads.

(2) Rejection Efficiency
The rejection efficiency of the target species (e.g., E. coli or red blood cell) was determined by comparing the concentration of the target species in the residual liquid sample after the treatment (cr) and in the original liquid sample (c0). The operation was similar to the process that determined the recovery efficiency. A total of 10 PSAP beads were applied to treat 1 mL of liquid sample (V0), which were subsequently taken out and weighted (m). The volume of the residual liquid sample (Vr) was calculated: in which the ρ was the density of the hydrated beads (approximately, the density of the biofluid media). Therefore, the rejection efficiency was determined as follows:

Pore Structure Control
Based on previous studies, we investigate the effects of the polymerization conditions (polymerization temperature, monomer composition, crosslinking degree, and porogen concentration) on the pore structure and swelling behaviors of the resultant PSAP beads. 5,6 The reaction mixtures containing 10% of AM, 4% MBA and 0.3% APS were added PEG as the porogen (5-30 wt%) and polymerized at three temperatures (70-90°C) to prepare porous polyacrylamide (PAM) beads. The results shown in Figure S8 indicate that the morphology of the beads is mainly dependent on the PEG content. When the polymerization temperature is 70°C, no obvious pores can be found in the PAM beads with 5 wt% of PEG. As the PEG content increases from 5 to 10 wt%, the resultant beads have a few pores, but those pores are small and disconnected. When the PEG content is further increased, both the porosity and the average pore size of the beads increase, and the as-synthesized beads show a uniform pore distribution.
On the other hand, the polymerization temperature is another important factor for phase separation together with the pore structure formation during the polymerization. At the same PEG content (e.g., 10 wt%), the porosity and the average pore size of the PAM beads increase as the polymerization temperature increases ( Figure S8). Nevertheless, although a higher temperature can accelerate and enhance the phase separation between the polymer network and the porogen, a relatively rapid separation rate causes more uneven distribution of pores and the formation of some big pores in the beads. Therefore, 70°C was selected as the optimal polymerization temperature for further optimization of the PSAP bead synthesis.

Swelling Capacity Control
The swelling capacity of the polymer depends on the crosslinking degree, charge density, and polymer concentration of the polymer network. 7 These three factors are related to the concentrations of the crosslinker, the ionic comonomer, and the total monomers in the initial reaction precursor, respectively. Based on the design criteria, the PSAP beads should have a high swelling ability in the water and aqueous solutions, so that the proposed microfiltration treatment of biofluid samples enabled by the PSAP beads is a spontaneous process and does not need additional driving forces. However, regardless of the PEG contents, the swelling ratio of the as-S8 prepared porous PAM beads is less than 10 g g −1 in DI water, which is not adequate for the demand of super absorbency.
To improve the swelling capacity, an ionic comonomer, SA, was introduced to the reaction system for the production of PSAP beads. The porous poly(SA-co-AM) beads were prepared by precursors containing 2-8 wt% of SA monomers while the total monomer concentration was fixed at 10 wt%. The addition of ionic monomers changed the precursor properties and polymerization mechanism thus affected the subsequent phase separation and pore generation. As illustrated in Figure S9, when the SA concentration is 2 wt%, the surface of the resulting bead is covered by a dense nonporous polymer layer and numerous extremely small pores distribute inside the bead. As the SA concentration is increased to 4 or 6 wt%, the beads show a uniform porous dendritic structure and the average pore size is between 1 to 2 µm. If the SA concentration is further increased to 8 wt%, the porosity significantly decreases as well as some particles are embedded on the polymer network. Therefore, a reaction mixture with the monomer ratio (SA/AM) of 4:6 or 6:4 was chosen in the following experiments for further improvement of the bead swelling capacity.
As mentioned, the swelling capacity of the polymer is highly affected by the crosslinking degree. A high crosslinking degree results in the difficulty of water diffusing into the polymer network due to a more compact polymer structure. 7 Hence porous poly(SA-co-AM) beads with the MBA concentration from 1 to 0.1 wt% were prepared and characterized. Figure S10 illustrates the weight swelling ratio of the beads at equilibrium in DI water. The results indicate that the crosslinking degree is the dominant factor for the swelling capacity of the beads. As expected, the swelling ratio of the beads increases as the MBA concentration decreases despite the monomer composition. At a high crosslinking degree (0.5-1 wt% of MBA), there is almost no difference in the swelling ratio for both the precursors. While at a low crosslinking degree (<0.5 wt% of MBA), the impact of the charge density in the swelling ratio becomes more distinct. For example, the swelling ratio of the beads prepared by 6 wt% of SA is ~60% higher than those prepared by 4 wt% of SA (117 versus 74 g g −1 ) when the MBA concentration is 0.1 wt%. This is because the increase of the ionic group number from the SA monomers results in the increase of counterion concertation inside the polymer network, which contributes to an additional osmotic pressure improving the polymer swelling. 8 The change of the crosslinking degree also has a slight impact on the pore structure of the resultant beads. As shown in Figure S11, a higher crosslinking degree (i.e., a high MBA S9 concentration) results in a more rigid polymer network with more small pores while a lower crosslinking degree (i.e., a low MBA concentration) results in a more relaxed network with a uniform pore distribution. Since the poly(SA-co-AM) beads containing 6 wt% of SA and 4 wt% of AM have both a high swelling capacity and a uniform pore distribution, we have selected this composition as the optimal monomer composition for preparing the prospective PSAP beads.