Microgels Sopping Up ToxinsâGM1a-Functionalized Microgels as

Research Article Cite This: ACS Appl. Mater. Interfaces 2019, 11, 25017−25023

www.acsami.org

Microgels Sopping Up ToxinsGM1a-Functionalized Microgels as Scavengers for Cholera Toxin Sarah Boesveld,†,⊥ Alexander Jans,‡,⊥ Dirk Rommel,‡ Matthias Bartneck,† Martin Möller,‡ Lothar Elling,§ Christian Trautwein,† Pavel Strnad,*,† and Alexander J. C. Kuehne*,‡,∥ †

Department of Internal Medicine III, University Hospital, RWTH Aachen University, Pauwelsstraße 30, 52074 Aachen, Germany DWILeibniz Institute for Interactive Materials, RWTH Aachen University, Forckenbeckstraße 50, 52076 Aachen, Germany § Laboratory for Biomaterials, Institute for Biotechnology and Helmholtz-Institute for Biomedical Engineering, RWTH Aachen University, Pauwelsstraße 20, 52074 Aachen, Germany ∥ Institute of Organic and Macromolecular Chemistry, Ulm University, Albert-Einstein-Allee 11, 89081 Ulm, Germany

Downloaded via BUFFALO STATE on July 21, 2019 at 10:39:18 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

‡

S Supporting Information *

ABSTRACT: Vibrio cholerae is a Gram-negative bacterium that causes secretory diarrhea and constitutes a major health threat in the industrialized world and even more in developing countries. Its main virulence factor is the cholera toxin, which is internalized by intestinal epithelial cells after binding to the glycosphingolipid receptor GM1a on their apical surface. A potential future solution to dampen complications of cholera infection is by scavenging the cholera toxin by presenting competitive binding motifs to diminish the in vivo toxicity of V. cholerae. Here, we generate GM1a-functionalized and biocompatible microgels with diameters of 20 μm using drop-based microfluidics. The microgels are designed to exhibit a mesoporous and widely meshed network structure, allowing diffusion of the toxin protein deep into the microgel scavengers. Flow cytometry demonstrates strong and multivalent binding at high capacity of these microgels to the binding domain of the cholera toxin. Cell culture-based assays reveal the ability of these microgels to scavenge and retain the cholera toxin in direct binding competition to colorectal cells. This ability is evidenced by suppressed cyclic adenosine monophosphate production as well as reduced vacuole formation in mucus-forming colorectal HT-29 cells. Therefore, glycan-functionalized microgels show great potential as a non-antibiotic treatment for toxinmediated infectious disorders. KEYWORDS: microfluidics, drug-delivery, toxin binding, polyethylene glycol, multivalency

■

INTRODUCTION

million infections and 100 thousand deaths worldwide each year.8,9 The cholera toxin (CT), the main virulence factor of V. cholerae, exerts its effect by binding to GM1a ganglioside receptors, presented at the apical surface of small intestinal epithelial cells.8 After its endocytosis and cleavage in the endoplasmic reticulum, CT is translocated into the cytosol, where it induces the characteristic massive secretion of electrolytes.10 Given its key importance for the entry of CT into the cell, the described GM1a binding constitutes an attractive therapeutic target. CT has a pentameric binding domain [cholera toxin B (CTB)] featuring five ganglioside binding sites.11,12 Compared to other carbohydrate−protein interactions, normally being in the millimolar range, the CTB/ GM1 interaction features high affinity with Kd values between 10 and 40 nM for the monovalent glycan ligand.13 To effectively bind CT and compete with GM1a gangliosides expressed on epithelial cells, multivalent interactions are

Toxin-mediated intestinal infections, such as those caused by Vibrio cholerae (cholera), Clostridium difficile (CD), Shigella dysenteriae (shigellosis), or pathogenic Escherichia coli strains, represent a major health threat and cause substantial mortality both in industrial and developing countries.1−4 Overgrowth of harmful bacteria in the gut results in accumulation of enterotoxins causing secretory diarrhea and intestinal inflammation. This may lead to rapid deterioration by shock or loss of fluid and electrolytes. For these diseases, fluid replenishment constitutes the mainstay of therapy, while antibiotic treatment might also be required.4,5 Antibiotics antagonize the underlying bacterial infection; however, they do not eliminate the accumulated toxins. Therefore, antibiotic treatment is not always efficient and bears the risk for relapse.6 To countervail this problem, a monoclonal antibody against CD toxin B has been developed, binding the toxin and preventing recurrent CD infections in clinical trials.7 By contrast, no toxinscavenging treatment is available for shigellosis or cholera, which are more common in developing countries. Here, cholera is the more dangerous disease, accounting for around 3 © 2019 American Chemical Society

Received: April 12, 2019 Accepted: June 21, 2019 Published: June 21, 2019 25017

DOI: 10.1021/acsami.9b06413 ACS Appl. Mater. Interfaces 2019, 11, 25017−25023

Research Article

ACS Applied Materials & Interfaces

Figure 1. Synthesis of microgels and attachment of GM1a. (a) Molecular structures of the prepolymer sPEG and GMA. (b) Emulsification process (aqueous phase in fluorinated oil) via droplet-based microfluidics. Photo-initiated free radical polymerization yields monodisperse microgels containing epoxide functional groups. (c) Image of monodisperse microgels in aqueous media. (d) CT-specific binding motif GM1a is bound to microgels via nucleophilic substitution-type epoxy-amine coupling at pH 9.2 for 24 h. (GM1a: Galβ1-3GalNAcβ1-4(Neu5Acα2-3)Galβ14GlcβCer.) Scale bar represents 25 μm.

competition with natively expressed binding motifs in the intestine. Because of their open network, the binding capacity per weight of microgels should be higher than for scavengers that bind toxins only on their surface. Accordingly, microgel systems may become useful for treating toxin-mediated intestinal infections and can be used as an alternative or in combination with antibiotics. Moreover, in contrast to antibiotics, microgels are unlikely to alter intestinal microbiota and might reduce the risk of recurring disease. Here, we develop such a microgel toxin scavenger system by incorporating GM1a gangliosides into polyethylene glycol (PEG)-based microgels. Our new microgel system exhibits high binding strength and high capacity compared to other CT scavengers.

needed for strong interactions of CT with a therapeutically effective decoy.14−17 Several oligomer or polymer decoys with multiple binding sites have been developed to bind to the CT binding domain and, therefore, inhibit its activity.14,15,18,19 With respect to cell culture and in vivo experiments, recent examples demonstrate that multivalent GM1a-modified glycodendrimers and nanoparticles can effectively bind CT and decrease toxic effects to the cells.18,19 For example, GM1a motifs can be coated onto polymer nanoparticles and if applied in excess (compared to GM1a gangliosides on epithelial cells), this competitive interaction is able to diminish the toxicity of V. cholerae both in cell culture and in a murine infection model.19 However, because of their small size and restricted steric orientations of surface-grafted recognition motifs, therapeutic nanoparticles do not reach multivalent binding and the binding capacity and strength are, therefore, limited. Furthermore, previous studies have not applied toxin scavengers or inhibitors in direct binding competition with colorectal cells but always pre-incubated the toxin with the scavengers before exposing cells.18,19 To overcome this concern, we hypothesize that microgels with a flexible and open polymer network structure are useful as novel CT-scavengers, which enable multivalent binding and entrapment of toxins allowing their clearance from the intestine. Microgels are biocompatible, soft entities that can deform and take up guest molecules and small nanoparticles inside of their widely meshed polymer network.20 Microgels functionalized with glycan recognition motifs have been shown to act as specific scavengers for the corresponding lectins.21 Because of their high degree of internal functionalization and the accessibility of their network, microgels represent interesting candidates for uptake and multivalent binding of toxins. This cooperative interaction of multiple glycans with the toxin would increase affinity and enable successful

■

RESULTS AND DISCUSSION To prevent cellular uptake of the microgel scavengers by intestinal epithelial cells, we prepare microgels in the range of tens of micrometers. This size regime is accessible using microfluidic emulsification technology. We prepare large microgels of ∼20 μm in diameter by emulsifying a solution of six-armed star-shaped PEG (sPEG, with ethylene oxide to propylene oxide ratio of 80:20) macromolecules and terminal acrylate groups22,23 using a microfluidic flow-focusing device followed by on-chip polymerization (see Figure 1a). The inner phase producing the microgels is composed of 10 wt % of sPEG, glycidyl methacrylate (GMA; 10 mol per mol sPEG) to yield microgels with functional epoxide groups, and 0.6 wt % of Irgacure 2959 photoinitiator in water. This formulation is emulsified on a microfluidic polydimethylsiloxane silicone chip produced by soft lithography. The continuous phase is composed of Novec HFE 7500 (a fluorinated ether) with 1.8 vol % of Krytox as a stabilizing agent, which prevents 25018


Research Article

ACS Applied Materials & Interfaces aggregation and fusion of the droplets (see Figure 1b). Microgels are obtained after illuminating the produced droplets with light of λmax = 365 nm at 6 W. On average, each droplet is exposed for 180 s, at fluid driving pressures of pdisp = 300 mbar for the dispersed phase and pcont = 350 mbar for the continuous phase. Each droplet represents a dedicated reaction container, in which exactly one microgel is produced. The polymerized microgels are collected in an excess of the continuous phase. The obtained microgels are purified followed by a solvent exchange to water to obtain highly monodisperse swollen sPEG-co-GMA microgels in aqueous phase (see Figure 1c and Supporting Information for more details). We determine the amount of incorporated and available epoxide groups by titration, confirming an incorporation rate of 90% (see Supporting Information for details).24 In view of the application of our microgels as CT scavengers, several conditions need to be fulfilled. First, the microgels need to be stable toward pH of the intestine to avoid colloidal aggregation or degradation on a molecular level, which would limit or impede the uptake of toxin. Second, the microgels need to have an open and porous network structure, so that toxins can diffuse deep into the microgel. Third, the microgel needs to be functionalized with specific binding motifs, which are distributed across the entire microgel so that the scavenged toxins coordinate to multiple binding motifs in a cooperative and multivalent way. To investigate the colloidal stability, we expose our microgels (18 kD sPEG) to pH ranging from 4 to 10 and incubate for 24 h at 37 °C. We inspect the microgel solutions with regard to their turbidity and sedimentation and check their structural integrity by bright-field microscopy. Neither degradation nor agglomeration could be observed (see Supporting Information Figure S2). Furthermore, we have conducted shelf life tests to investigate the long-term stability of our microgels. Microgels that have been stored at 4 °C for 12 months show the same structural integrity and colloidal stability as freshly prepared samples, confirming that the microgels are molecularly intact and stable over extended times (see Figure S3). For the uptake of biomoleculessuch as the CTinto the microgel, it is important to understand how far and how quickly molecules of different molecular weight can diffuse into the microgel. To study the effect of cross-link density, we vary the sPEG chain-length. We prepare three batches of microgels using sPEG precursors of 3, 12, and 18 kDa. As model compounds for uptake into microgels, we apply fluorescent fluorescein isothiocyanate (FITC)-labeled dextran with molecular weights of 10, 70, 250, and 500 kDa at concentrations of 0.5 mg/mL for 10 kDa dextran and 2 mg/mL for higher homologues. We investigate all possible permutations of the three microgel samples with the different dextrans by incubating them and sampling aliquots of 50 μL every 10 min. We purify these aliquots by sedimentation, exchange the supernatant, and redisperse the microgels before transfer to confocal laser-scanning microscopy (CLSM) for analysis. After 10 min of incubation, we see that the 3 kDa microgel is completely fluorescing, indicating that 10 kDa dextran can easily diffuse deep into the microgel (see Figure 2a). By contrast, dextrans of higher molecular weights (70 and 250 kDa) require much longer times to diffuse deep into the microgel. The 500 kDa dextran requires more than 45 min to only produce a fluorescent corona, substantiating that the network of the microgel produced from 3 kDa sPEG is too

Figure 2. Morphology of the functionalized microgels. (a−c) Diffusion study of FITC-functionalized dextran molecules with increasing molar mass (10, 70, 250, and 500 kDa) through differently sized microgel networks produced from prepolymers of (a) 3, (b) 12, and (c) 18 kDa. (d−f) Confocal microscopy study of epoxyfunctionalized microgels prepared from 18 kDa sPEG. The epoxides have been functionalized with 6-aminofluorescin: (d) confocal z-stack through a microgel. The stronger signal at z = 1 μm is an artifact from the microgel substrate interaction; otherwise the distribution of fluorescence is homogeneous across the entire microgel, indicating consistent epoxy-functionalization across the microgel volume. (e) Deconvolved confocal microscopy images of an ensemble of microgels and (f) close-up of an optical slice through the middle of a microgel. Scale bars represent 10 μm in (d), 20 μm in (e), and 2 μm in (f).

closely meshed to allow diffusion of such a large molecule into the microgel (see Figure 2a). Microgels produced from 12 and 18 kDa sPEGs do not exhibit this diffusion limitation and all dextrans fully infiltrate the microgel after 10 min (see Figure 2b,c). To gain further insight into the morphology of our microfluidically produced microgel networks, we apply those microgels prepared from 18 kDa sPEG and bind the fluorescent dye 6-aminofluorescein to the incorporated epoxide groups using well-established epoxide−amine coupling chemistry. We study the microgel using CLSM and performing a z-scan through one of the microgels (from the substrate z = 0 μm into the sample volume z = 20 μm). We observe homogeneous fluorescence throughout all slices, demonstrating a homogeneous distribution and high conversion of the functional epoxide groups throughout the microgel (see Figure 2d). The higher intensity of the first slice at z = 1 μm is an 25019


Research Article


Figure 3. Specific binding of FITC-labeled CTB subunit to GM1a-functionalized microgels. (a−d) Bright field and fluorescence microscopy images of non-functionalized (a,b) and GM1a-conjugated microgels (c,d) incubated with 3 μmol FITC-labeled CTB. Scale bars in (a) and (b) represent 10 μm in (c) and (d) 50 μm. (e,f) Flow cytometry experiments to quantify the amount of FITC-labeled CTB uptake in (e) non-functionalized vs (f) GM1a-functionalized microgels after their exposure to different concentrations of FITC-labeled CTB. Tested concentrations of FITC-labeled CTB are 0, 0.1, 0.3, 1.0, 3.0, 10, 30, and 60 μM. (g) Mean fluorescence intensity (MFI) of non-functionalized vs GM1a-functionalized microgels incubated with different concentrations of FITC-labeled CTB. (h) MFI of the supernatants from the above-described experiments to quantify the ability of non-functionalized and GM1a-functionalized microgels to absorb FITC-labeled CTB.

same range as the hydrodynamic diameter of the CT (3.7 nm).26 To incorporate the GM1a motif into the microgels for specific and strong binding of CT, we couple an aminefunctionalized GM1a via nucleophilic addition to the epoxide moieties. We monitor the conversion using 1H NMR and find that 60% of the epoxides are functionalized after the reaction (see Supporting Information). The extant epoxides are quenched using amino ethanol. Effectively, this degree of conversion translates into concentrations of order 10−8 to 10−7 mol of GM1a per mg sPEG. To verify covalent coupling and investigate the ability of these GM1a units in our microgels to bind CT, we employ a fluorescein (FITC)-labeled CTB subunitthe non-toxicbinding domain of CT (ca. 60 kDa in pentameric form). CTB is incubated with GM1a-functionalized and nonfunctionalized microgels. CLSM is performed after centrifugation of the microgels and redispersion in fresh supernatant (6 times), showing uniformly fluorescing microgels and therefore deep and homogeneous uptake of FITC-labeled CTB by GM1a-functionalized microgels. By contrast, no detectable fluorescence is observed for non-functionalized microgels (Figure 3a−d). This observation signifies that specific interactions with the toxin are necessary for efficient uptake into the microgels.

artifact, which originates from the microgel adhering to the glass substrate, resulting in a flattened microgel shape and increased signal at the microgel glass interface. Comparing the microgels within a sample, we observe uniform fluorescence among the microgels, indicating that there is little variation in the degree of epoxide functionalization and microgel morphology (see Figure 2e). When we improve the resolution by deconvolving a confocal slice through one of the microgels, we see that the microgel exhibits mesoporosity with uniformly distributed pores in the range of hundreds of nm (see Figure 2e). On a smaller hierarchical level, the lower limit for porosity is on the molecular scale and defined by the arm length of the 18 kDa sPEG. The observed hierarchical porosity of the microgels on the mesoscale and molecular scale is beneficial, as it allows fast diffusion of biomacromolecules through the mesopores into the microgel, where on the molecular level there is a dense presentation of the GM1a-binding motifs, which can retain the CT once inside the microgel. The CT has a molar mass of M = 84 kDa,25 and we therefore continue with the 18 kDa sPEG as a precursor for our toxin scavenging microgels. The smallest loop that can occur in our formulation is when two arms of the same 18 kDa sPEG molecule react with each other during polymerization. The resulting loop has a circumference of ca. 18 nm and a maximum diameter of ca. 6 nm,21 which is in the 25020


Research Article


In HT-29 cells, CT exposure induces formation of pleomorphic vacuoles as well as a release of cyclic adenosine monophosphate (cAMP) into the supernatant (Figure 4). Both vacuole formation and cAMP are indicators for toxininduced injury of HT-29 cells.

To quantify the observed uptake and binding, we perform flow cytometric analysis of these incubated microgels. Our data supports strong fluorescence of GM1a-conjugated microgels, confirming efficient binding of CTB. Non-functionalized microgels exhibit only low fluorescence because of minimal unspecific binding of CTB (Figure 3e,f). Forward and sideward scatters reveal a homogeneous microgel population for the non-functionalized microgels and a slightly increased sizedispersity for GM1a-functionalized microgels (see Figure S4). Increasing the CTB concentration results in an almost linear increase in MFI of the microgels, with only a small deviation from this linear trend at 60 μM of CTB (Figure 3g). This slight leveling at high CTB concentrations indicates that equilibrium of CTB uptake is reached at concentrations >60 μM (=720 μg/mL). To further assess the ability of microgels for CTB absorption and binding, we also measure the fluorescence of the supernatants after incubation with non-functionalized and GM1a-conjugated microgels. The supernatant from nonfunctionalized microgels exhibits a linear increase of the MFI with increasing CTB concentrations, indicating no specific uptake of CTB by the microgels. In GM1a-functionalized microgels, we see that CTB concentration of up to 30 μM do not exhibit significant fluorescence in the supernatants. This demonstrates that GM1a-functionalized microgels are able to scavenge CTB efficiently. However, concentrations higher than 30 μM of CTB leave behind fluorescent supernatants, indicating that not all CTB is taken up from solution at this concentration (Figure 3h). The latter observation might have two reasons: first, we might be reaching the capacity limit of the microgels, or second, the surface of the microgels might be passivated by bound CTB, thereby reducing further diffusion of FITC-labeled CTB molecules deeper into the microgel. Above we determined the degree of GM1a functionalization to be of order 10−8 to 10−7 mol/mg sPEG. Based on the recorded fluorescence data, we determine the binding capacity Bmax of the microgels to be of order 10−8 mol CTB per mg sPEG in the swollen microgel. This indicates that we are approaching the maximum capacity of the highly porous microgels (cf. Figure 2c). Considering that the binding study is performed at neutral pH, CTB will exist in its pentameric form.27,28 Therefore, the capacity of our microgels translates to ca. 1 mg of CTB per mg dry (nonswollen) microgel. This high capacity is in the same range as previously reported CT scavenger systems.19 However, when we determine the dissociation constant of CTB to our microgels using the Hill−Waud method,29 we find Kd of 6 × 10−6 M, representing strong binding and a cooperativity factor of 2.02 indicating that CTB is multivalently bound on average by two GM1a units in the microgel. This good binding performance indicates that our microgels can be applied in a scenario where they bind the full CT in competition to mucosal epithelial cells.30 Next, we investigate whether our microgels have sufficient uptake performance to scavenge the complete CT. We observe the response of HT-29 colorectal adenocarcinoma cells to a solution of toxin, which has been pre-incubated with GM1afunctionalized microgels. Alternatively, we expose the HT-29 cells to both, toxin and microgels at the same time. In the latter experiment, strong binding of the toxin to the microgels is essential, as otherwise CT will interact with the cells and induce its toxic effect. This biologically essential, competitive binding experiment is often neglected in toxin inhibition studies.

Figure 4. GM1a microgels protect intestinal cells from CT’s toxicity. (a−d) Bright field microscopy images (magnification 20×) showing HT-29 cells that are either (a) left untreated as a control or (b) exposed to 20 ng/mL of CT for 2 h. Alternatively, cells are subjected to a CT solution that has been pre-incubated (c) with 100 μg/mL of non-functionalized microgels or (d) with 100 μg/mL of GM1afunctionalized microgels. Red arrows indicate vacuoles. Scale bars represent 50 μm. (e) Morphometric quantification of pleomorphic vacuoles in the above-described experiments (a−d). (f−h) Supernatants of the above-described samples (a−d) are analyzed for their amount of released cAMP to further illustrate the response of HT-29 cells to the described treatment regimen. (f) Control experiment, where HT-29 cells have been exposed to non-functionalized and GM1a-functionalized microgels. No cAMP production is detected. (g) HT-29 cells are exposed to solutions of CT. In case of the indicated microgel experiments, the toxin solution has been preincubated with microgels. After removal of the microgels, the cells are exposed to the solution. (h) Cells are simultaneously exposed to toxin and microgel. Here, the microgels directly compete with the cells for the CT. Confidence intervals derived from p-values are indicated by stars. 25021


Research Article


Figure 5. Determination of the optimal amount of microgels and incubation time for CT scavenging. Impact of (a) different microgel concentrations and (b) incubation times on CT toxicity. The concentration of released cAMP is assessed in HT-29 cells that are either left untreated (control) or are exposed to 20 ng/mL CT for 2 h (20 ng/mL CT). Alternatively, HT-29 cells are subjected to a CT solution (20 ng/mL) that has been pre-incubated with non-functionalized or with GM1a-functionalized microgels. Pre-incubations are carried out (a) for 2 h with different microgel concentrations or (b) for the indicated periods with 100 μg/mL microgels. For additional control, we pre-incubate medium without CT with 100 μg/mL of non-functionalized or GM1a-functionalized microgels for 2 h.

microgel concentrations further protect HT-29 cells from cell injury showing maximal protection at a concentration of 100 μg/mL of GM1a-functionalized microgels. Finally, we investigated the impact of incubation time toward efficient toxin binding. The 30 min pre-incubation is sufficient to maximally suppress cAMP-release and vacuole formation. Longer pre-incubation times do not further improve this effect. By contrast, pre-incubating for only 15 min nearly doubles cAMP levels compared to the 30 min pre-incubation time (Figure 5b). This observation indicates that diffusion and multivalent binding requires time. Assuming that in the human body the intestinal mucosa is replaced approximately every 8 h, the microgels should have sufficient time to take up CT to the maximum of their capacity.

Interestingly, pre-incubation of CT solution with GM1afunctionalized microgels significantly reduces vacuole formation of HT-29 cells (see red arrows in Figure 4a−e). Additionally, CT solution with GM1a-functionalized microgels pre-incubation also significantly reduces cAMP concentrations in the supernatant of HT-29 cells. To better mimic the situation in vivo, we directly expose HT-29 cells to CT and GM1a-functionalized microgels. Here, toxin binding has to take place in competition with the cells. Co-incubation in contrast to CT exposure alone significantly reduces cAMP in the supernatant. These results suggest that CT has a higher affinity to GM1a-functionalized microgels compared to the surface receptors on epithelial cells (cf. Figure 4g,h). By contrast, pre-incubation or co-exposition with nonfunctionalized microgels shows no protective effect as evidenced by a strong vacuole formation and cAMP release (see Figure 3 and blue data in Figure 4f,h). In fact, recorded vacuole numbers and cAMP concentrations are higher than in the pure toxin reference experiments (cf. blue and red bars in Figure 4e,g,h). Notably, exposure of HT-29 cells to either microgel suspension (GM1a-functionalized or not) shows no obvious toxicity in both assays applied, indicating that microgels alone are non-cytotoxic (see Figure 4f). We hypothesize that this increased toxicity of non-functional microgels in conjunction with CT is an excluded volume effect of the microgels, effectively increasing the concentration of CT. Additional studies will be required to elucidate this observation further. It is noteworthy that storage of GM1a-modified microgels at 4 °C for 2 months does not change their ability to protect HT29 cells from CT-induced injury. This observation suggests that the microgel approach enables long-term storage before use in acute situations of cholera infection (see Figure S5). Next, we determine the effective microgel concentration needed to protect HT-29 cells from CT-induced injury. Hence, we pre-incubate CT with increasing microgel concentrations (Figure 5a). Already at a concentration of 10 μg/mL of GM1afunctionalized microgels, we find a significant reduction in cAMP-release compared to the cell response to pure CT and/ or CT in the presence of non-functional microgels. Increasing

■

CONCLUSIONS The glycan-functionalized microgel can efficiently bind CT, and this represents an attractive approach to ameliorate the consequences of V. cholerae infection. Our finding defines a novel non-antibiotic means to treat V. cholerae infection, which at present has not been implemented as a first-line treatment approach. Our concept seems of general relevance as it might also apply to other diseases associated with enterotoxininduced diarrhea such as CD, shigellosis, or enterohemorrhagic E. coli.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.9b06413. Applied models and statistics including 1H NMR spectra of non- and GM1a-functionalized microgels; pH and long-term stability tests; bright field microscopy images of microgels; fluorescence-activated cell sorting analysis; cAMP release experiments; and experimental details including materials and methods, master preparation, surface functionalization, microgel synthesis, characterizations, and cell culture studies (PDF) 25022


Research Article


■

(13) Turnbull, W. B.; Precious, B. L.; Homans, S. W. Dissecting the Cholera Toxin−Ganglioside GM1 Interaction by Isothermal Titration Calorimetry. J. Am. Chem. Soc. 2004, 126, 1047−1054. (14) Tran, H.-A.; Kitov, P. I.; Paszkiewicz, E.; Sadowska, J. M.; Bundle, D. R. Multifunctional Multivalency: A Focused Library of Polymeric Cholera Toxin Antagonists. Org. Biomol. Chem. 2011, 9, 3658−3671. (15) Branson, T. R.; McAllister, T. E.; Garcia-Hartjes, J.; Fascione, M. A.; Ross, J. F.; Warriner, S. L.; Wennekes, T.; Zuilhof, H.; Turnbull, W. B. A Protein-Based Pentavalent Inhibitor of the Cholera Toxin B-Subunit. Angew. Chem., Int. Ed. 2014, 53, 8323−8327. (16) Branson, T. R.; Turnbull, W. B. Bacterial Toxin Inhibitors Based on Multivalent Scaffolds. Chem. Soc. Rev. 2013, 42, 4613−4622. (17) Zuilhof, H. Fighting Cholera One-on-One: The Development and Efficacy of Multivalent Cholera-Toxin-Binding Molecules. Acc. Chem. Res. 2016, 49, 274−285. (18) Zomer-Van Ommen, D. D.; Pukin, A. V.; Fu, O.; Quarles Van Ufford, L. H. C.; Janssens, H. M.; Beekman, J. M.; Pieters, R. J. Functional Characterization of Cholera Toxin Inhibitors Using Human Intestinal Organoids. J. Med. Chem. 2016, 59, 6968−6972. (19) Das, S.; Angsantikul, P.; Le, C.; Bao, D.; Miyamoto, Y.; Gao, W.; Zhang, L.; Eckmann, L. Neutralization of Cholera Toxin with Nanoparticle Decoys for Treatment of Cholera. PLoS Neglected Trop. Dis. 2018, 12, No. e0006266. (20) Plamper, F. A.; Richtering, W. Functional Microgels and Microgel Systems. Acc. Chem. Res. 2017, 50, 131−140. (21) Jans, A.; Rosencrantz, R. R.; Mandić, A. D.; Anwar, N.; Boesveld, S.; Trautwein, C.; Moeller, M.; Sellge, G.; Elling, L.; Kuehne, A. J. C. Glycan-Functionalized Microgels for Scavenging and Specific Binding of Lectins. Biomacromolecules 2017, 18, 1460−1465. (22) Lensen, M. C.; Mela, P.; Mourran, A.; Groll, J.; Heuts, J.; Rong, H.; Möller, M. Micro- and Nanopatterned Star Poly(Ethylene Glycol) (PEG) Materials Prepared by UV-Based Imprint Lithography. Langmuir 2007, 23, 7841−7846. (23) Möller, M.; Mourran, C.; Rong, H.; Spatz, J. P. Sternförmige Präpolymere Für Die Herstellung Ultradünner Hydrogel-Bildender Beschichtungen. EP 1469893 A1, 2002. (24) Park, J.-G.; Kim, J.-W.; Suh, K.-D. Monodisperse GlycidylFunctional Polymer Particles in the Micron-Size Range by Seeded Polymerization. Colloid Polym. Sci. 2001, 279, 638−645. (25) Finkelstein, R. A.; LoSpalluto, J. J. Pathogenesis of Experimental Cholera: Preparation and Isolation of Choleragen and Choleragenoid. J. Exp. Med. 1969, 130, 185−202. (26) Dwyer, J. D.; Bloomfield, V. A. Subunit Arrangement of Cholera Toxin in Solution and Bound to Receptor-Containing Model Membranes. Biochemistry 1982, 21, 3227−3231. (27) Hardy, S. J.; Holmgren, J.; Johansson, S.; Sanchez, J.; Hirst, T. R. Coordinated Assembly of Multisubunit Proteins: Oligomerization of Bacterial Enterotoxins in Vivo and in Vitro. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 7109−7113. (28) Zrimi, J.; Ling, A. N.; Arifin, E. G.-R.; Feverati, G.; Lesieur, C. Cholera Toxin B Subunits Assemble into PentamersProposition of a Fly-Casting Mechanism. PLoS One 2010, 5, No. e15347. (29) Worstell, N. C.; Krishnan, P.; Weatherston, J. D.; Wu, H.-J. Binding Cooperativity Matters: A GM1-like Ganglioside-Cholera Toxin B Subunit Binding Study Using a Nanocube-Based Lipid Bilayer Array. PLoS One 2016, 11, No. e0153265. (30) Lian, T.; Ho, R. J. Y. Cholera Toxin B-Mediated Targeting of Lipid Vesicles Containing Ganglioside GM1 to Mucosal Epithelial Cells. Pharm. Res. 1997, 14, 1309−1315.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (P.S.). *E-mail: [email protected] (A.J.C.K.). ORCID

Matthias Bartneck: 0000-0003-1516-9610 Martin Möller: 0000-0002-5955-4185 Lothar Elling: 0000-0002-3654-0397 Alexander J. C. Kuehne: 0000-0003-0142-8001 Author Contributions ⊥

S.B. and A.J. contributed equally.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Collaborative Research Center (CRC) grant SFB 985 project C3 from DFG (Deutsche Forschungsgemeinschaft) and the European Commission (EUSMI, 731019). This work was performed in part at the Center for Chemical Polymer Technology CPT, which was supported by the EU and the federal state of North Rhine-Westphalia (grant EFRE 30 00 883 02). The authors would like to thank Silvia Roubrocks for excellent technical support.

■

REFERENCES

(1) Gugnani, H. C. Some Emerging Food and Water Borne Pathogens. J. Commun. Dis. 1999, 31, 65−72. (2) Pfeiffer, M. L.; Dupont, H. L.; Ochoa, T. J. The Patient Presenting with Acute DysenteryA Systematic Review. J. Infect. 2012, 64, 374−386. (3) Hodges, K.; Gill, R. Infectious DiarrheaCellular and Molecular Mechanisms. Gut Microb. 2010, 1, 4−21. (4) Dupont, H. L. Acute Infectious Diarrhea in Immunocompetent Adults. N. Engl. J. Med. 2014, 370, 1532−1540. (5) Leffler, D. A.; Lamont, J. T. Clostridium Difficile Infection. N. Engl. J. Med. 2015, 372, 1539−1548. (6) Oldfield, E. C.; Wallace, M. R. The Role of Antibiotics in the Treatment of Infectious Diarrhea. Gastroenterol. Clin. North Am. 2001, 30, 817−835. (7) Wilcox, M. H.; Gerding, D. N.; Poxton, I. R.; Kelly, C.; Nathan, R.; Birch, T.; Cornely, O. A.; Rahav, G.; Bouza, E.; Lee, C.; Jenkin, G.; Jensen, W.; Kim, Y.-S.; Yoshida, J.; Gabryelski, L.; Pedley, A.; Eves, K.; Tipping, R.; Guris, D.; Kartsonis, N.; Dorr, M.-B. Bezlotoxumab for Prevention of Recurrent Clostridium Difficile Infection. N. Engl. J. Med. 2017, 376, 305−317. (8) Ali, M.; Nelson, A. R.; Lopez, A. L.; Sack, D. A. Updated Global Burden of Cholera in Endemic Countries. PLoS Neglected Trop. Dis. 2015, 9, No. e0003832. (9) Clemens, J. D.; Nair, G. B.; Ahmed, T.; Qadri, F.; Holmgren, J. Cholera. Lancet 2017, 390, 1539−1549. (10) Ma, T.; Thiagarajah, J. R.; Yang, H.; Sonawane, N. D.; Folli, C.; Galietta, L. J. V.; Verkman, A. S. Thiazolidinone CFTR Inhibitor Identified by High-Throughput Screening Blocks Cholera ToxinInduced Intestinal Fluid Secretion. J. Clin. Invest. 2002, 110, 1651− 1658. (11) Merritt, E. A.; Sarfaty, S.; Martial, J. A.; Hol, W. I. M. G. J.; Hoir, C. L. Crystal Structure of Cholera Toxin B-Pentamer Bound to Receptor GM1 Pentasaccharide. Protein Sci. 1994, 3, 166−175. (12) Heggelund, J. E.; Mackenzie, A.; Martinsen, T.; Heim, J. B.; Cheshev, P.; Bernardi, A.; Krengel, U. Towards New Cholera Prophylactics and Treatment: Crystal Structures of Bacterial Enterotoxins in Complex with GM1 Mimics. Sci. Rep. 2017, 7, 2326. 25023


Microgels Sopping Up ToxinsâGM1a-Functionalized Microgels as

Recommend Documents

Microgels Sopping Up ToxinsâGM1a-Functionalized Microgels as