Polyzwitterionic Coating of Porous Adsorbents for Therapeutic Apheresis

Abstract

Adsorbents for whole blood apheresis need to be highly blood compatible to minimize the activation of blood cells on the biomaterial surface. Here, we developed blood-compatible matrices by surface modification with polyzwitterionic polysulfobetainic and polycarboxybetainic coatings. Photoreactive zwitterionic terpolymers were synthesized by free-radical polymerization of zwitterionic, photoreactive, and fluorescent monomers. Upon UV irradiation, the terpolymers were photodeposited and mutually crosslinked on the surface of hydrophobic polystyrene-co-divinylbenzene and hydrophilic polyacrylamide-co-polyacrylate (DALI) beads. Fluorescent microscopy revealed coatings with an average thickness of 5 µm, which were limited to the bead surface. Blood compatibility was assessed based on polymer-induced hemolysis, coagulation parameters, and in vitro tests. The maintenance of the adsorption capacity after coating was studied in human whole blood with cytokines for polystyrene beads (remained capacity 25-67%) and with low-density lipoprotein (remained capacity 80%) for polyacrylate beads. Coating enhanced the blood compatibility of hydrophobic, but not of hydrophilic adsorbents. The most prominent effect was observed on coagulation parameters (e.g., PT, aPTT, TT, and protein C) and neutrophil count. Polycarboxybetaine with a charge spacer of five carbons was the most promising polyzwitterion for the coating of adsorbents for whole blood apheresis.

Keywords: adsorbents; blood compatibility; coat; extracorporeal therapies; polyzwitterions.

Figure 1

Chemical structures of polyzwitterions (zwitterionic moiety-blue, crosslinking unit-red, fluorescent label-green), and scanning electron micrographs of the DALI (“Direct Adsorption of Lipoproteins”, Fresenius Medical Care, Bad Homburg, Germany) and CG161c (Amberchrom CG161c, Dow Chemical, Philadelphia, PA, USA) beads (inserts bead cross-section). Scale bars represent 50 µm, and 25 µm for inserts.

Figure 2

Coating protocol with physical pre-coating and UV photocrosslinking of the coating.

Scheme 1

Synthesis of the zwitterionic pSBE, pCBMA-C1, and pCBMA-C5 copolymers used in this study.

Figure 3

Confocal laser scanning microscopy images of a coated DALI-pSBE bead (A), its 3D appearance (B), a coated CG161c-pSBE bead (C), and its 3D appearance (D). Scale bars represent 50 µm.

Figure 4

Micrographs of thin sections of CG161c beads at bright field (upper row) and fluorescence (lower row). From left to right, unmodified CG161c, CG161c-pSBE, CG161c-pCBMA-C1, and CG161c-pCBMA-C5. Scale bars represent 25 µm.

Figure 5

Assessment of coagulation parameters upon incubation of citrated plasma (37 °C, 60 min) with coated and uncoated adsorbent polymers: (A) prothrombin time (PT), reflecting alterations of the extrinsic coagulation pathway; (B) activated partial thromboplastin time (aPTT), reflecting alterations of the intrinsic pathway; (C) thrombin time (TT); (D) fibrinogen concentration; (E) antithrombin III (AT-III); (F) protein C. Dotted lines indicate the reference range for the individual coagulation parameters. Data represent the mean of 3 experiments ± standard deviation; ctrl, plasma incubated without adsorbent. Variables of significance (* p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001) were calculated by one-way ANOVA. AT-III and protein C were calibrated to 100% using standardized control plasma. The reference range comprises 75% to 125% of this standardized control plasma, and therefore, levels of >100% of protein C can be observed.

Figure 6

Assessment of blood cell counts upon incubation of citrated whole blood (37 °C, 60 min) with coated and uncoated CG161c: (A) red blood cells, (B) platelets, (C) leukocytes, (D) neutrophils, (E) lymphocytes, (F) monocytes. Horizontal lines indicate the individual counts at 0 min for whole blood anticoagulated with EDTA (full line) and sodium citrate (dotted line). Data represent the mean of 3 experiments ± standard deviation; ctrl, blood incubated without adsorbent. Variables of significance (* p ≤ 0.05, *** p ≤ 0.001) were calculated by one-way ANOVA.

Figure 7

Depletion of (A) IL-6 and (B) TNF-α from plasma derived from LPS-stimulated human whole blood after treatment with 10% vol of CytoSorb, CG161c, CG161c-pSBE, CG161c-pCBMA-C1, and CG161c-pCBMA-C5. Data are given as mean of 3 experiments ± standard deviation, except for CytoSorb and CG161c (n = 6); ctrl, plasma incubated without adsorbent. Variables of significance (* p ≤ 0.05: ns, not significant) were calculated by one-way ANOVA.