Conditional Cell-Penetrating Peptide Exposure as Selective Nanoparticle Uptake Signal

Abstract

A major bottleneck diminishing the therapeutic efficacy of various drugs is that only small proportions of the administered dose reach the site of action. One promising approach to increase the drug amount in the target tissue is the delivery via nanoparticles (NPs) modified with ligands of cell surface receptors for the selective identification of target cells. However, since receptor binding can unintentionally trigger intracellular signaling cascades, our objective was to develop a receptor-independent way of NP uptake. Cell-penetrating peptides (CPPs) are an attractive tool since they allow efficient cell membrane crossing. So far, their applicability is severely limited as their uptake-promoting ability is nonspecific. Therefore, we aimed to achieve a conditional CPP-mediated NP internalization exclusively into target cells. We synthesized different CPP candidates and investigated their influence on nanoparticle stability, ζ-potential, and uptake characteristics in a core-shell nanoparticle system consisting of poly(lactid-co-glycolid) (PLGA) and poly(lactic acid)-poly(ethylene glycol) (PLA_10kPEG_2k) block copolymers with CPPs attached to the PEG part. We identified TAT47-57 (TAT) as the most promising candidate and subsequently combined the TAT-modified PLA_10kPEG_2k polymer with longer PLA_10kPEG_5k polymer chains, modified with the potent angiotensin-converting enzyme 2 (ACE2) inhibitor MLN-4760. While MLN-4760 enables selective target cell identification, the additional PEG length hides the CPP during a first unspecific cell contact. Only after the previous selective binding of MLN-4760 to ACE2, the established spatial proximity exposes the CPP, triggering cell uptake. We found an 18-fold uptake improvement in ACE2-positive cells compared to unmodified particles. In summary, our work paves the way for a conditional and thus highly selective receptor-independent nanoparticle uptake, which is beneficial in terms of avoiding side effects.

Keywords: TAT; charge-mediated uptake; nanoparticle surface charge; nanoparticle targeting; polyarginine; polycationic; polymer nanoparticles; sequential uptake.

Figure 1

Overview of uptake mechanisms of cell-penetrating peptides. The uptake mechanism can be categorized into endocytic (left) and direct (right) uptake ways. While the compound is instantly available in the cytoplasm after direct uptake, it initially ends up in early endosomes after endocytic uptake.^, In this case, a further hurdle must be circumvented by means of endosomal escape in order to enter the cytoplasm. If this is not possible, the early endosomes undergo a maturing process, ending up in lysosomes, where acidic pH values and lysosomal enzymes lead to the degradation of the cargo.

Figure 2

Nanoparticle structure and targeting approach. (A) Nanoparticle design. Different CPP modifications of polymeric nanoparticles with a core–shell structure were analyzed in a simplified particle design to select a suitable candidate for a conditional, sequential particle uptake (left). The nanoparticles consisted of PLGA and PLA_10kPEG_2k block copolymers (PLA drawn in red, PEG in blue). Due to its higher hydrophilicity, PEG formed the NP shell (blue halo). Since PLGA and PLA are generally miscible, a PLGA-rich inner core layer (shown in orange/red) and a PLA-rich outer core layer (shown in red/blue) were obtained. The CPP was tethered to the PEG part and thus directly visible on the nanoparticle surface and able to mediate nanoparticle uptake. Methoxy-terminated polymers (PLA_10kPEG_2kMeO) served as space-filling polymers between the CPP-modified chains (PLA_10kPEG_2kCPP). The most promising candidate was established in a more complex particle design which promoted conditional, sequential nanoparticle uptake (right). Therefore, the CPP tethered to PLA_10kPEG_2k was combined with PLA_10kPEG_5k polymers, which had a longer PEG chain. The additional PEG length shielded the CPP during a first cell contact. Additionally, MLN-4760, a selective ACE2 inhibitor was attached to these longer polymer chains for selective target cell recognition (PLA_10kPEG_5kMLN). The proportion of long polymers was set to 25% according to Walter et al. For control nanoparticles, the MLN-modified polymer was replaced by uncharged methoxy-terminated polymer (PLA_10kPEG_5kMeO). Different amounts of CPP modifications were evaluated (0–75%) (PLA_10kPEG_2kCPP) and the proportion of short polymer, which should not be modified, was accordingly replaced by short methoxy-terminated polymers as placeholders (PLA_10kPEG_2kMeO). (B) Concept of a conditional, sequential nanoparticle uptake. Since the CPP is shielded by longer polymers during a first cell contact, the uptake-enhancing abilities of the ligand do not directly promote cell uptake. Only after a previous, selective binding of the nanoparticle to ACE2 via MLN-4760, the established spatial proximity exposes the uptake signal previously hidden inside the polymer shell. This leads to CPP-mediated uptake exclusively into ACE2-positive target cells.

Figure 3

Characterization of the influence of nanoparticle modification with CPPs on their ζ-potential. The ζ-potential of nanoparticles modified with various CPPs (see diagram title) in different ratios of CPP-modified and unmodified methoxy-terminated polymer were analyzed. For all polycationic CPPs, increasing proportions of CPP-modified polymer led to increasing ζ-potentials. For the same DOM, a higher number of positive charges per ligand also led to higher absolute ζ-potentials (shown in the bottom right panel for a DOM of 50%) (n = 3 technical replicates).

Figure 4

CLSM image of the qualitative evaluation of nanoparticle uptake. R7-modified nanoparticle binding and uptake were tracked after two different time points: the cells were incubated with NPs for 1 h. Afterward, the particles were removed, followed by further incubation for 2 h (1 h + 2 h) or 24 h (1 h + 24 h), respectively. Colocalization of LysoTracker deep red (LTDR), staining late endosomes and lysosomes (red), and R7-modified nanoparticles with a DOM of 50% labeled with TAMRA (R7-NP-TAMRA) (yellow) is shown in turquoise. The simultaneous employment of direct and endosomal uptake ways is presumed, since after 3 h, most particles still accumulate at the cell surface, but there is already background fluorescence visible inside the cell showing nanoparticle distribution in the cytoplasm. At this time point, almost no colocalization between particles and LTDR was visible, which indicates direct nanoparticle uptake. After 25 h, high endosomal entrapment was shown, which proves additional endosomal uptake. The increasing background fluorescence demonstrated that the CPP-modified particles reached the cytosol and therefore their destination inside the cell. Scale bar: 10 μm.

Figure 5

Comparison of nanoparticle binding/uptake improvement for different CPP modifications and different ratios of modified to unmodified polymer. The respective CPPs are indicated in the diagram headings. The panel at the bottom right shows the correlation between particle binding and uptake and the number of positive charges per ligand (DOM = 50%). Results represent mean ± standard deviation (SD) (n = 3, levels of statistical significance are indicated as *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001).

Figure 6

CLSM images of the uptake of Cy5-labeled R7-modified NPs with different DOMs in HEK293 cells. Cell nucleus labeling with DAPI (blue) served for cell localization. Increasing ratios of CPP-modified polymer mediated enhanced uptake of the Cy5-labeled nanoparticles (Cy5-NP) (red) into HEK293 cells. Scale bar: 10 μm.

Figure 7

Correlation ζ-potential and uptake improvement of CPP-modified nanoparticles in HEK293 cells. The relationship between ζ-potential and uptake enhancement was verified using both an experimental (A, B) and a theoretical (C) approach. (A, B) NPs with identical CPP modifications (R7) and differently charged space-filling polymers (uncharged methoxy-terminated block copolymer (MeO) (left) and negatively charged carboxy-terminated block copolymer (COOH) (right)) were prepared. The APC-A median indicates the fluorescence signal induced by the binding or uptake of Cy5-labeled NPs in HEK293 cells. ζ-Potential measurements (A) and flow cytometric experiments (B) demonstrated that a positive ζ-potential was directly associated with a significant enhancement of particle uptake. (C) To quantitatively describe the relationship between surface charge and increased cellular uptake, a model was developed to deduct the maximum improvement of cellular uptake achievable (k_max) by CPP functionalization and the ζ-potential required to obtain half-maximum uptake improvement ζ_1/2. Results represent mean ± SD (n = 3, levels of statistical significance are indicated as *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001); (AFU, arbitrary fluorescence units).

Figure 8

Characterization of nanoparticles with shielded CPP modification. The nanoparticles with TAT modification on short polymer chains (PEG2k) and MLN-4760 attached to long polymer chains (PEG5k) were characterized via DLS and NTA. Control NP contained methoxy-terminated longer polymers instead of the MLN-4760-modified polymers. The proportion of long polymers was set to 25%. The proportion of TAT-modified polymer varies and is indicated for every bar in the diagram (n = 3 technical replicates).

Figure 9

Flow cytometric and CLSM analysis of steric shielded NP binding and uptake. (A) Flow cytometric measurements of NP binding/uptake to ACE2-positive stable transfected HEK293T target cells. Different degrees of TAT47–57 functionalization (percentage of the shell polymer) were investigated and are stated on the x-axis to find a suitable CPP amount for the sequential targeting concept. The APC-A median shown on the y-axis represents the fluorescence of the cells corresponding to the binding/uptake of Cy5-labeled particles. A DOM of 30% was considered for the further experiments (red arrow). (B) Negative control with ACE2-negative HEK293 cells. (C) Confirmation of the results of (A) via CLSM analysis for a DOM of 30%. NPs were labeled with Cy5, cell nuclei were stained with DAPI. In the right row, the Cy5 and DAPI channels were merged and supplemented by transmitted light (TL) for cell localization. (D) CLSM evaluation of MLN-TAT-NP from (D) with further 24 h of incubation. (E) Coculture experiments with HEK293T-ACE2 stable cells and untransfected HEK293 cells. Scale bar: 20 μm. Results represent mean ± SD (n ≥ 3, levels of statistical significance are indicated as *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001); (AFU, arbitrary fluorescence units).