The effect of charge and albumin on cellular uptake of supramolecular polymer nanostructures

Abstract

Intracellular delivery of functional biomolecules by using supramolecular polymer nanostructures has gained significant interest. Here, various charged supramolecular ureido-pyrimidinone (UPy)-aggregates were designed and formulated via a simple "mix-and-match" method. The cellular internalization of these UPy-aggregates in the presence or absence of serum proteins by phagocytic and non-phagocytic cells, i.e., THP-1 derived macrophages and immortalized human kidney cells (HK-2 cells), was systematically investigated. In the presence of serum proteins the UPy-aggregates were taken up by both types of cells irrespective of the charge properties of the UPy-aggregates, and the UPy-aggregates co-localized with mitochondria of the cells. In the absence of serum proteins only cationic UPy-aggregates could be effectively internalized by THP-1 derived macrophages, and the internalized UPy-aggregates either co-localized with mitochondria or displayed as vesicular structures. While the cationic UPy-aggregates were hardly internalized by HK-2 cells and could only bind to the membrane of HK-2 cells. With adding and increasing the amount of serum albumin in the cell culture medium, the cationic UPy-aggregates were gradually taken up by HK-2 cells without anchoring on the cell membranes. It is proposed that the serum albumin regulates the cellular internalization of UPy-aggregates. These results provide fundamental insights for the fabrication of supramolecular polymer nanostructures for intracellular delivery of therapeutics.

Fig. 1. Chemical structures of the ureido-pyrimidinone (UPy)-monomers and characterizations of the various UPy-aggregates. (a) Chemical structures of the UPy-monomers with different functional groups. (b) Composition and illustration of the UPy-aggregates with different charge properties. (c) Nile red (NR) measurement for the confirmation of the assemblies of UPy-aggregates in PBS with the formation of lateral hydrophobic pockets, where the NR could be encapsulated and emitted intensive fluorescent signals. (d) Fluorescence resonance energy transfer (FRET) measurement with NR/Cy5 pair to confirm the encapsulation of fluorescent reporter UPy–Cy5 into the UPy-aggregates in PBS. (e) Total internal reflection fluorescence microscope (TIRF) of the various UPy-aggregates in PBS, the morphology of these aggregates were examined with Cy5 fluorophore. Scale bars represent 10 μm. (f) Zeta-potential measurements of the various UPy-aggregates (50 μM) in 5 mM HEPES buffer at pH 7.4, the transition of the charge properties of the UPy-aggregates from anionic to cationic has been confirmed.

Fig. 2. Cytocompatibility of the UPy-aggregates. The various UPy-aggregates all have good cytocompatibility with examined concentrations for both (a) THP-1 derived macrophages and (b) Human kidney cells (HK-2 cells). The tests were performed with an LDH method with the presence of fetal bovine serum, the cells without the addition of any UPy-aggregates were used as the control group. Error bars are the standard deviation of the mean value averaged over triplicates.

Fig. 3. Internalization of UPy-aggregates (10 μM) by THP-1 derived macrophages and human kidney cells (HK-2) in the presence of fetal bovine serum (FBS). (a) Confocal laser scanning micrographs of the internalization of the UPy-aggregates at 120 min. The UPy-aggregates, nuclei and membranes of the cells were labeled red, blue and green, respectively. All of the UPy-aggregates with different charge properties were internalized by both THP-1 and HK-2 cells. Scale bars represent 30 and 50 μm for THP-1 and HK-2 cells, respectively. (b) Flow cytometry analysis of the internalization of the UPy-aggregates by THP-1 derived macrophages and HK-2 cells. The results are illustrated as the signal distribution of the 10 000 gated cells. The experimental group without addition of UPy-aggregates was used as the control. Both the internalizations of pure UPy–Cy5 and Cy5–NH₂ were compared with the other four types of UPy-aggregates containing the same amount of Cy5, and much stronger fluorescence intensity for the cells of the four types of UPy-aggregates can be observed. (c) Illustration of the average fluorescence intensity of the 10 000 gated cells incubated with different materials. Error bars are the standard deviation of the mean cell fluorescence intensity averaged over triplicates. p < 0.0001 (****).

Fig. 4. Internalization of the UPy-aggregates (10 μM) by THP-1 derived macrophages and human kidney cells (HK-2) in the absence of fetal bovine serum (FBS). (a) Confocal laser scanning micrographs of the internalization of UPy-aggregates at 120 min. The UPy-aggregates, nuclei and membranes of the cells were labeled red, blue and green, respectively. Both neutral and anionic UPy-aggregates were much less internalized compared to cationic and neutral (+/−) counterparts, which indicates the internalization of UPy-aggregates without the presence of FBS is charge dominated. Scale bars represent 30 and 50 μm for THP-1 derived macrophages and HK-2 cells, respectively. (b) Flow cytometry analysis of the internalization of the UPy-aggregates by THP-1 and HK-2 cells. The results are illustrated as the signal distribution of the 10 000 gated cells. The experimental group without addition of UPy-aggregates was used as the control. Both the internalizations of pure UPy–Cy5 and Cy5–NH₂ were compared with the other four types of UPy-aggregates containing the same amount of Cy5, and much stronger fluorescence intensity for the cells of the cationic and neutral (+/−) groups of UPy-aggregates can be observed. (c) Illustration of the average fluorescence intensity of the 10 000 gated cells incubated with different materials. Error bars are the standard deviation of the mean cell fluorescence intensity averaged over triplicates. p < 0.0001 (****).

Fig. 5. Locations of the cationic UPy-aggregates (10 μM) after being incubated with mammalian cells for 120 min. The cationic UPy-aggregates, GFP-modified mitochondria and nuclei of the cells were labelled red, green and blue, respectively. (a) The cationic UPy-aggregates can be internalized by THP-1 derived macrophages under the conditions of presence or absence of fetal bovine serum (FBS), and the Cy5 signal overlapped with mitochondria. Scale bars represent 10 μm. (b) The cationic UPy-aggregates can be internalized by human kidney cells (HK-2 cells) under the condition of presence of FBS, and the Cy5 signal overlapped with mitochondria. Under the condition of absence of FBS, the cationic UPy-aggregates anchored on the membranes of the HK-2 cells with subtle internalization. Scale bars represent 20 μm.

Fig. 6. Bovine serum albumin (BSA) determined the internalization of cationic UPy-aggregates (10 μM) by human kidney cells. (a) Nile red (NR) test for the confirmation of the formation of UPy-aggregates in the DMEM medium with the formation of lateral hydrophobic pockets, where the NR could be encapsulated and emitted intensive fluorescent signals. The DMEM medium without addition of UPy-monomers was used as the control. (b) Zeta-potential of the cationic UPy-aggregates (50 μM) incubated with different amount of BSA in the medium of 5 mM HEPES buffer at pH 7.4. BSA strongly interacted with the cationic UPy-aggregates as shown by the zeta-potential shifting from positive to negative with different amount of BSA. (c) Confocal fluorescent micrographs of the internalization of the cationic UPy-aggregates with different amount of BSA. The cationic UPy-aggregates, GFP-modified mitochondria and nuclei of the cells were labeled red, green and blue, respectively. Scale bars represent 20 μm. With the addition of BSA into the cationic UPy-aggregates, the UPy-aggregates were gradually internalized by the cells and co-localized with mitochondria instead of anchoring on the membrane of the cells.