(Top panel) Two possible filtering strategies could be employed by mucus hydrogels. Size filtering (left) allows particles with dimensions smaller than the hydrogel mesh size to pass while larger particles are rejected. Interaction filtering (right) allows to distinguish particles according to their interaction strength with the hydrogel polymers as mediated by the particle surface properties: Strongly interacting particles (orange) are retained in the hydrogel, whereas weakly interacting particles (blue) are allowed to pass. (Lower panels) Two examples of the trajectory analysis as conducted for the single particle tracking data (see ) that was obtained for amine-terminated polystyrene particles in a 0.5% (w/v) mucin hydrogel at pH 7 (A and B) and at pH 3 (C and D). From the mean-square displacement (A and C) of the particle trajectories, apparent diffusion coefficients are determined (see A and for the ensemble average value). To illustrate the heterogeneity of the obtained diffusion coefficients, particles from a single region of interest are colored according to their mobility. At pH 7, all particles show a highly similar diffusion behavior (B). However, at pH 3, mobile and immobile particles can be found in close vicinity, demonstrating a high variance of particle mobility (D). The examples depicted in panels B and D show relative distribution widths of 38% and 178%, respectively.

(A) Microscopic mobility of particles of ∼1-μm size with different surface properties in a 0.5 (w/v) mucin hydrogel at pH 3. The background shading indicates the surface properties of the particles changing from negative to positive charge. PEG(a) and PEG(c) particles were created by PEGylation of amine-terminated and carboxy-terminated particles, respectively. Apparent diffusion coefficients D were determined by single particle tracking for ensembles of ∼50 particles per condition (see ). The error bars denote the error of the mean , where σ represents the standard deviation of the obtained distribution and N denotes the number of particles analyzed per condition. (B) Spatial heterogeneity in particle diffusion behavior as represented by the relative distribution width (see and ). Low particle mobilities correlate with higher variances in the diffusion behavior of the particle ensemble. (C) Zeta-potentials for the test particles investigated in panels A and B.

(A) Schematic representation of mobile (= neutral or neutralized, gray) and immobile (= charged, yellow) particles in mucin hydrogels. At high salt concentrations, ions should weaken the electrostatic interaction of the charged particles with the charged mucin biopolymers by Debye screening. (B) The mucin hydrogel permeability depends on the buffer salt concentration. The data shown was obtained at pH 3 for a 0.5% (w/v) mucin hydrogel and a 1% (w/v) mucin hydrogel, respectively. In both cases, the mobility of charged amine-terminated particles is considerably increased at high ionic strengths compared to low salt conditions. This increase in mobility is more pronounced at the higher mucin concentration. In contrast, the mobility of neutral PEG(c) particles in a 1% (w/v) mucin hydrogel at pH 3 is virtually unaffected by changes in the buffer salt conditions.

(A) Mucus hydrogels in the human body exist at different levels of pH as indicated by the color scheme. In the nose or the lung, the mucus layer is slightly alkaline, whereas gastric and intestinal mucus is exposed to rather harsh acidic environments. In the female genital tract, pH conditions can significantly change during the ovulatory cycle, ranging from neutral to acidic conditions. The concentration of mucin polymers in human mucus hydrogels covers a range of 1–5% (see main text). A similar range is explored in our in vitro system albeit at reduced absolute values. (B) Mean diffusion coefficient and (C) spatial heterogeneity of the particle diffusion behavior, σ_rel, for PEG(a) particles in mucin hydrogels at different pH levels and mucin concentrations. Small σ_rel values indicate homogeneous distributions of diffusion constants, while large values indicate strong heterogeneities. For 1% (w/v) at pH 3 (bar is labeled with an asterisk), σ_rel = 100% is chosen for clarity. Note that the calculated value is much larger, σ_rel = 600%. This is because, at this particular condition, 90% of the particles are completely immobile and show no detectable motion within the resolution of our tracking algorithm. A few particles show weak mobility.

(A) The human mucin MUC5AC protein contains numerous acidic and basic amino acids. The pK values of Asp and Glu are 4.7, whereas those of His, Lys, and Arg are considerably higher (6.5–12). Thus, the mucin MUC5AC will undergo a major transition in its charge state between high and low values of pH. To visualize the magnitude of this effect, we assume Asp and Glu to contribute one negative elementary charge each at high pH, whereas His, Lys, and Arg each contribute one positive elementary charge at low pH. The entire mucin MUC5AC polypeptide sequence is contracted by pooling subsequent groups of 50 amino acids into virtual domains, which are colored according to their total net charge. The numbers indicate the amino acid position in the mucin sequence. Please note that for this schematic representation the glycosylation of the mucin polymer was neglected, although glycosylated regions might considerably contribute to the barrier function of mucin hydrogels (see ). (B) Biophysical model for the filtering function of mucin hydrogels: The mucin hydrogel can be represented as a network consisting of either positively or negatively charged segments. Charged particles (yellow, purple) are trapped by the opposite charge on the mucin polymer (purple, yellow), although neutral particles (gray) can diffuse nearly unhindered. At neutral pH (left), the charge on the mucin polymer is relatively weak, whereas at acidic pH (right), its strength is significantly increased. The different sizes of the two hydrogel pieces illustrate gel shrinkage upon pH decrease, as also observed experimentally.