Enhanced Cellular Uptake of Albumin-Based Lyophilisomes when Functionalized with Cell-Penetrating Peptide TAT in HeLa Cells

Lyophilisomes are a novel class of biodegradable proteinaceous nano/micrometer capsules with potential use as drug delivery carrier. Cell-penetrating peptides (CPPs) including the TAT peptide have been successfully implemented for intracellular delivery of a broad variety of cargos including various nanoparticulate pharmaceutical carriers. In the present study, lyophilisomes were modified using CPPs in order to achieve enhanced cellular uptake. Lyophilisomes were prepared by a freezing, annealing, and lyophilization method and a cystein-elongated TAT peptide was conjugated to the lyophilisomes using a heterobifunctional linker. Fluorescent-activated cell sorting (FACS) was utilized to acquire a lyophilisome population with a particle diameter smaller than 1000 nm. Cultured HeLa, OVCAR-3, Caco-2 and SKOV-3 cells were exposed to unmodified lyophilisomes and TAT-conjugated lyophilisomes and examined with FACS. HeLa cells were investigated in more detail using a trypan blue quenching assay, confocal microscopy, and transmission electron microscopy. TAT-conjugation strongly increased binding and cellular uptake of lyophilisomes in a time-dependent manner in vitro, as assessed by FACS. These results were confirmed by confocal microscopy. Transmission electron microscopy indicated rapid cellular uptake of TAT-conjugated lyophilisomes via phagocytosis and/or macropinocytosis. In conclusion, TAT-peptides conjugated to albumin-based lyophilisomes are able to enhance cellular uptake of lyophilisomes in HeLa cells.


Introduction
An innovative strategy in cancer therapy utilizes drug delivery carriers to increase the therapeutic effect of anti-tumor drugs. The role of drug delivery carriers in this context is to improve pharmacokinetics and dynamics by protecting the drug from degradation [1][2][3]. Contemporary drug delivery systems include nanoparticulate systems loaded with anti-tumor drug and conjugates directly coupled to the drug. The unique property of nanoparticle carriers is their ability to encapsulate and deliver a high dose of anti-tumor drugs, including poorly soluble drugs, and to exploit the enhanced permeability and retention (EPR) effect for tumor targeting [4]. Because of these favorable characteristics, there has been intense interest in the development of nanoparticulate drug delivery systems. Nanoparticles currently investigated for cancer therapeutic applications include liposomes, polymersomes, dendrimers, micelles, carbon nanotubes, nanoconjugates and (protein-based) nanospheres or capsules [5][6][7]. Among the available potential drug carrier systems, protein-based nanoparticles are particularly interesting as they hold certain advantages such as good stability during storage, non-toxicity, biocompatibility, and biodegradability in vivo [8,9]. Recently we showed that lyophilisomes, a novel class of proteinaceous biodegradable hollow nano/micrometer capsules, show potential as a drug delivery capsule [10,11]. Lyophilisomes can be prepared from a large variety of watersoluble macromolecules including proteins (e.g. albumin and elastin) but also polysaccharides (e.g. heparin). In fact, virtually any biomolecule can be incorporated into the wall/lumen of the capsule, resulting in a highly flexible carrier system with multiple applications. We previously demonstrated that enzymes intro-duced in the capsule's wall and in the lumen are bioactive and able to convert a substrate [10]. Furthermore, lyophilisomes have been efficiently loaded with doxorubicin resulting in tumor cell elimination in vitro [11]. In order to obtain a selective drug delivery system, lyophilisomes have also been modified with antibodies, resulting in specific targeting of the cell of interest in vitro [12]. Due to these properties, lyophilisomes can be deployed for the design of multifunctional targeting systems.
Lyophilisomes were prepared from albumin. Albumin is an attractive macromolecular carrier that has been shown to be nontoxic, non-immunogenic, biodegradable to produce innocuous degradation products, and easy to purify [13]. It is thus a suitable candidate for nanoparticle preparation, as demonstrated by FDAapproved products such as Abraxane [14,15] and Albunex [16,17].
Since delivery of nanocarriers is generally based on passive accumulation in pathological tissues, they do not efficiently deliver their cargo to specific cells. When drug delivery carriers arrive at the tumor site, they have to cross the plasma membrane in order to deliver the drug into the cell. The plasma membrane, however, prevent proteins, peptides, and nanoparticulate drug carriers from entering the cell in the absence of an active transport mechanism [18]. Cell-penetrating peptides (CPPs) have been successfully used to deliver a large variety of cargos to the cell interior including proteins [19], peptides [20], nucleic acids [21] and pharmaceutical nanocarriers [22][23][24]. CPPs are short peptides consisting of up to 30 amino acids that are able to translocate across the cellular membrane [25]. When CPPs are conjugated to drug delivery carriers, efficient cellular uptake of the carriers can be achieved [26]. A representative CPP is the TAT peptide, derived from the TAT protein (trans-activation transcriptional activator) of the human immunodeficiency virus type 1 (HIV-1) [27,28]. The TAT peptide consists of 11 amino acids with the sequence YGRKKRRQRRR. The abundance of lysine and arginine residues makes it highly positively charged, important for the interaction with the plasma membrane. In this study, the TAT peptide was conjugated to lyophilisomes to investigate whether CPPs are able to enhance cellular uptake of lyophilisomes.

Preparation of lyophilisomes
Lyophilisomes were prepared from albumin as described previously [11]. Briefly, droplets of a solution of 0.25% (w/v) bovine serum albumin (BSA) in 0.01 M acetic acid were frozen in liquid nitrogen (2196uC). The frozen albumin preparation was incubated at 210 to 220uC for 3 h (annealing step), and subsequently lyophilized. This procedure results in hollow nano/ micro spheres (''lyophilisomes''). In order to visualize the lyophilisomes, FITC-conjugated albumin was added to nonlabeled albumin (1:10) in the starting solution. To obtain stabilized lyophilisomes, they were vapor crosslinked with glutaraldehyde and formaldehyde. Generally we prepare 40 ml of a 0.25% BSA solution, which corresponds to 100 mg albumin (2.5 mg/ml). The final lyophilisome population was centrifuged three or four times at low speed (606g; Thermo, Heraeus Fresco 17; Newport Pagnell, Great Britain) to remove large lyophilisomes and sheetlike structures, until no pellet was observed. After this procedure, about 30% of the original weight of lyophilisomes remained. Lyophilisomes (1 mg/ml) were stored in 0.1% (v/v) Tween-20 (Sigma Aldrich, Steinheim, Germany) in phosphate buffered saline (PBS-T, pH 7.4).

Conjugation of cell-penetrating peptide to lyophilisomes
A schematic representation of the conjugation reactions is depicted in figure 1.

Sorting of lyophilisomes by fluorescence-activated cell sorting
Fluorescence-activated cell sorting (FACS) was applied to select for small lyophilisomes (,1,000 nm) using a Coulter Epics Elite flow cytometer (BeckmanCoulter, Miami, FL, USA). Only small FITC-positive lyophilisomes were sorted (for settings, see section ''Lyophilisomes sorted by fluorescence-activated cell sorting''). To achieve high sensitivity, gain was set at 20.

Particle size measurements by qNano
The qNano (Izon, Science Ltd., Burnside, New Zealand) was used to measure particle size distribution of lyophilisomes [11,29]. To ensure a continuous flow of particles, a pore size of 600-2000 nm was used. Data were analyzed with Izon Control Suite 2.1 software.
For FACS, cells were stained with the plasma membrane dye PKH26 (Sigma Aldrich, Missouri, USA) [30,31]. One million cells were incubated in 2 mM PKH26 dye in 500 ml buffer (according to Enhanced Cellular Uptake of TAT-Functionalized Lyophilisomes PLOS ONE | www.plosone.org manufacturer's protocol) for 5 min. To stop the staining reaction, 500 ml FCS was added and incubated for 5 min. Subsequently, cells were washed three times with culture medium.

Cellular binding and internalization of lyophilisomes with and without TAT peptide
To determine whether the TAT peptide can promote the binding and internalization of lyophilisomes with HeLa, OVCAR-3, Caco-2 and SKOV-3 cells were seeded in a 24-well plate (30,000 cells/well in 1 ml medium). Cells were left to adhere overnight and subsequently incubated for 1 h with unsorted lyophilisomes with and without TAT peptide conjugated to them (25 mg/ml). After incubation, cells were washed three times with PBS to remove unbound lyophilisomes and harvested with enzyme-free EDTA solution (PAA Laboratories). Finally, cells were resuspended in 0.2% BSA in PBS and analyzed by FACS (FACSCalibur Becton Dickinson, Breda, Netherlands). Using the appropriate positive and negative controls FACS settings were adjusted. When lyophilisomes did not bind to cells, lyophilisomes were depicted at a fluorescent signal of 10 1 or below and cells were regarded negative. When lyophilisomes did bind to cells, they showed a fluorescent signal higher than 10 1 . Data were analyzed by FlowJo software (Version 9.4, Treestar, Ashland, OR, USA).

Internalization studies of lyophilisomes by cells
FACS. In order to discriminate between attached and internalized lyophilisomes, a FITC quenching trypan blue assay was used [32][33][34]. Quenching of FITC signal occurs because trypan blue absorbs the light emitted by FITC-labeled lyophilisomes after excitation. The FITC signal of internalized lyophilisomes however, is not quenched since trypan blue cannot pass the plasma membrane. The fluorescence remaining after trypan blue quenching must therefore result from internalized lyophilisomes, as only extracellular fluorescence of FITC-lyophilisomes is quenched.
To investigate the cellular uptake of lyophilisomes, PKH26 stained HeLa cells were seeded in a 24-well plate (30,000 cells/ well) and left to adhere overnight. Cells were incubated with 500,000 FACS-sorted lyophilisomes with and without TAT peptide for 1 and 4 h. After incubation, cells were washed three times with 0.2% (w/v) BSA in PBS, dissociated with enzyme-free EDTA dissociation buffer, resuspended in 1 ml culture medium and transferred to an eppendorf tube. Subsequently, cells were washed three times with 0.2% BSA-PBS by centrifugation (3 min, room temperature, 1006g) and incubated with 0.5% (w/v) trypan blue for 10 min and washed three times with 0.2% BSA-PBS. Cells were analyzed by FACSCalibur flowcytometry. Data were analyzed by FlowJo Software.
Confocal microscopy. To visualize cellular uptake of lyophilisomes, confocal microscopy was performed on living cells. HeLa cells were seeded in an 8-well microscopy chamber (Nunc; 30,000 cells/well) and left to adhere overnight. Cells were incubated with sorted lyophilisomes with and without TAT peptide using 0.8 and 3.5 million lyophilisomes in 200 ml per sample for 4 h in RPMI medium containing 10% FCS at 37uC. As a control, medium without lyophilisomes was used. After incubation, cells were washed three times, incubated for 5 min with CellMask Orange (5 mg/ml) to visualize the plasma membranes and then washed again, all with the same medium. Cells were kept at 37uC on a temperature controlled microscope stage and living cells were imaged immediately with a Leica SP5 confocal microscope (Leica Microsystems, Mannheim, Germany). FITC was excited at 488 nm and emission was collected between 500-550 nm. CellMask orange was excited at 561 nm and emission was collected between 570-650 nm. Images were recorded sequentially using Leica Application Suite Software (Advanced Fluorescence Lite, 2.3.0. build 5131).
Transmission electron microscopy. Cells incubated with lyophilisomes with and without TAT peptide as described in the Materials and Methods section ''FACS'' were embedded in 1.5% (w/v) agarose, fixed in 2% (v/v) glutaraldehyde in 0.1 M phosphate buffer (pH 7.4), post-fixed with 1% (w/v) osmium tetroxide, dehydrated in an ascending series of ethanol, and embedded in Epon 812. Ultrathin sections (60 nm) were cut and picked up on Formvar-coated grids, post-stained with lead citrate and uranyl acetate, and examined with a JEOL 1010 transmission electron microscope (Tokyo, Japan).

Statistical Analysis
Data are presented as mean with standard deviation. Data of Results section ''Cellular binding and internalization of lyophilisomes with and without TAT peptide'' were analyzed using twotailed Student's t-tests. Data of Results section ''Trypan blue assay and FACS'' were analyzed using two-way Anova Bonferroni posthoc tests. All statistical analyses were performed in Graphpad

Conjugation of TAT peptide to lyophilisomes
To probe the possibility of using CPP for enhanced intracellular delivery, lyophilisomes were modified with a cysteine-functionalized TAT peptide using the heterobifunctional linker sulfo-GMBS (Fig. 1). The succinimidyl ester functionality is conjugated to primary amine groups on the lyophilisome while the maleimide functionality is used for conjugation to the free thiol of the cysteine residue coupled to the TAT peptide.

Cellular binding and internalization of lyophilisomes with and without TAT peptide
To address the presence of the TAT peptide on TATconjugated lyophilisomes, unmodified lyophilisomes and TATconjugated lyophilisomes were administered to HeLa, OVCAR-3, Caco-2 and OVCAR-3 cells. Using standard FACS as a functional assay, it is not possible to discriminate between cellular attachment and internalization. Instead, the total of cell binding and internalization is measured (Fig. 2). For HeLa cells, TATconjugated lyophilisomes showed an about 8-fold increase in lyophilisome-positive cells compared to lyophilisomes without the TAT peptide (8663% and 1264% lyophilisome-positive cells for TAT-conjugated and unmodified lyophilisomes, respectively. OVCAR-3 and Caco-2 cells showed about a 5-fold increase in lyophilisome-positive cells compared to lyophilisomes without the TAT peptide (lyophilisome-positive cells: 9763% and 1963% for OVCAR-3; 8763% and 1668% for Caco-2) for TAT-conjugated and unmodified lyophilisomes, respectively. SKOV-3 cells gave a high background value when incubated with lyophilisomes without TAT (67620%), but still showed a 1.6 fold statistically significant increase with the presence of TAT peptide (95610%).

Lyophilisomes sorted by fluorescence-activated cell sorting
When using lyophilisomes for tumor targeting, lyophilisomal size is an important parameter. The initial lyophilisome population included sizes up to 2.8 mm (Fig. 3a). To obtain a more monodisperse capsule population, lyophilisomes were sorted by FACS. Lyophilisomes were separated based on forward scatter and FITC fluorescence (FL1 channel; Fig. 3c). To verify the procedure, a rerun of sorted lyophilisomes was performed (Fig. 3d). Using this methodology, larger lyophilisomes as well as sheet-like structures (confirmed by scanning electron microscope) were separated from small lyophilisomes. These results were substantiated by qNano size analysis using a lyophilisome preparation before (Fig. 3a) and after (Fig. 3b) sorting (approximately 90% below 1 mm). Measurements using the qNano consisted of 200-250 particles. The size distribution contained multiple peaks that can be explained by the low number of particles. Due to the limitations of the qNano instrument particles smaller than 600 nm could not be reliably detected which overestimates the average size of the lyophilisomes. Pilot experiments with a smaller qNano pore (200-800 nm) revealed the presence of lyophilisomes below 600 nm (results not shown), but larger lyophilisomes in the preparation frequently blocked this pore.

Cellular uptake of TAT-conjugated lyophilisomes
Trypan blue assay and FACS. To determine the internalization efficiency of sorted lyophilisomes with and without TAT peptide, a trypan blue quenching assay was used in order to distinguish between internalized and non-internalized (but plasma membrane-associated) lyophilisomes. This assay is based on the quenching of fluorescence of FITC-labeled lyophilisomes by the vital stain trypan blue (which does not penetrate plasma membranes). To validate that trypan blue also quenches fluorescence of FITC labeled lyophilisomes, lyophilisomes were incubated with trypan blue in the absence of cells and evaluated by FACS (Fig. 4). Lyophilisomes that were not incubated with trypan blue showed a mean fluorescence intensity of 13396252. Lyophilisomes incubated in a 0.5% trypan blue solution gave a mean fluorescent intensity of 85631, corresponding to a quenching efficacy of 9462% (Fig. 4b). After one, two and three washings after the trypan blue incubation, the measured mean fluorescence was 161625, 176622, and 192619 or 8861%, 8761% and 8561% quenching efficacy, respectively. This indicates that trypan blue was not easily washed out of the lyophilisomes and fluorescence remained quenched. This is important as three washings steps were used prior to FACS analysis.
Confocal microscopy. Internalization of TAT-conjugated lyophilisomes in HeLa cells was visualized by confocal microscopy (Fig. 6). After 4 h of incubation, TAT-conjugated lyophilisomes showed extensive uptake for both low (0.8 million/200 ml) and high (3.5 million/200 ml) numbers of lyophilisomes ( Fig. 6c/g). Almost no internalization was observed when HeLa cells were incubated with unmodified lyophilisomes (Fig. 6a/e). The corresponding bright field images showed that lyophilisomes did not lead to detectable morphological changes of the cells (Fig. 6b/ Transmission electron microscopy. In order to investigate the binding and uptake of TAT-conjugated lyophilisomes in HeLa cells in detail, TEM was used (Fig. 7). When unmodified lyophilisomes were added to HeLa cells, no binding or uptake was observed and the plasma membrane appeared largely unruffled (Fig. 7a). However, when TAT-conjugated lyophilisomes were added, multiple stages of internalization could be distinguished, including attachment and internalization (Fig. 7b/  c). Initially, TAT-conjugated lyophilisomes bound to HeLa cells and initiated membrane ruffling (Fig. 7b). Subsequently, capsules were internalized (Fig. 7c). Furthermore, initial signs of degradation of lyophilisomes could be observed, as degradation products were visible (black arrows; Fig. 7d).

Discussion
Our laboratory previously demonstrated that lyophilisomes show potential as a drug delivery system [11,12]. To further optimize these biocapsules, we studied the effect of TATfunctionalization on the in vitro internalization of lyophilisomes. Intracellular delivery of therapeutic molecules is one of the key problems in drug delivery. Many pharmaceutical compounds have to be delivered intracellularly to exert their therapeutic action [22]. CPPs have been shown to act as a powerful transport vector for inducing the cellular uptake of a large variety of cargos [18]. At present, pharmaceutical nanocarriers are much in focus for their capacity to increase the stability of administered drugs, improve their concentration at their site-of-action and decrease undesired side effects. Various studies report increased uptake and specific delivery to intracellular organelles when conjugating CPPs to drug delivery systems, thereby increasing the efficiency of nanocarriers as drug delivery systems [35][36][37].
As previously reported, lyophilisomes range in size from 100 up to 3,000 nm in diameter [11]. In this study, we demonstrated that using FACS, small lyophilisomes could be sorted out of the initial population, narrowing the size distribution.
In the present study, FACS was used to investigate the total cell binding and internalization of (TAT-conjugated) lyophilisomes in HeLa, OVCAR-3, Caco-2, and SKOV-3 cells. To investigate the cellular uptake and subcellular distribution of (TAT-conjugated) lyophilisomes in HeLa cells in more detail, confocal microscopy, and TEM were utilized. FACS demonstrated a high number of lyophilisome-positive cells when incubated with TAT-conjugated lyophilisomes compared to unmodified lyophilisomes. This may be explained by interaction of the TAT peptide with negatively charged sulfated glycans at the cell surface, such as heparan sulphate [38,39]. Only SKOV-3 cells showed a high background when incubated with unmodified lyophilisomes, which has been shown for other drug carriers added to this cell line [40].
To discriminate between internalization and attachment, a trypan blue quenching assay was applied on HeLa cells incubated with lyophilisomes. This assay showed that cellular uptake was enhanced when TAT peptides were conjugated to lyophilisomes. The degree of internalization may even have been underestimated as our capsules are likely to enter the acidic lysosomes in the cell [41] and since FITC fluorescein is virtually non-fluorescent below  Enhanced Cellular Uptake of TAT-Functionalized Lyophilisomes PLOS ONE | www.plosone.org pH 5 [42]. The trypan blue assay also demonstrated that cellular uptake of TAT-conjugated lyophilisomes increased over time. After 1 h, 2561% of the cells internalized TAT-conjugated lyophilisomes, whereas 59614% of cells had done so after an incubation period of 4 h. We most likely added too little lyophilisomes in the trypan blue internalization study to achieve lyophilisome internalization of all available cells. This is supported by confocal microscopy, which revealed that almost all cells internalized at least one lyophilisome when administering more lyophilisomes. TEM images showed different stages of cellular uptake (attachment and internalization) of TAT-conjugated lyophilisomes. The results strongly suggest that TAT-conjugated lyophilisomes are internalized by phagocytosis and/or macropinocytosis, as intensive plasma membrane ruffling was observed during uptake. This suggestion would be in line with the particle size of lyophilisomes [43]. However, lyophilisomes smaller than 200 nm, could still be internalized by one of the mechanisms of endocytosis.  If we compare our findings to other studies, the increased internalization efficiency compared to unmodified particles is within the range of reported values for TAT peptides conjugated to cargos such as liposomes [44,45]. However, it is difficult to compare internalization for different kinds of nanoparticles, since different cell types and different amounts and sizes of nanoparticles are used for in vitro experiments, e.g. the larger the particle the more time it takes to establish internalization [46].
The ability of CPPs to enhance cellular uptake non-specifically and receptor-independently provides the opportunity to target diverse cell types with a variety of carriers. In literature, it has been observed that next to HeLa, OVCAR-3, Caco-2 and SKOV-3 cells, the TAT peptide can enter other tumor cells, for instance bladder cancer (HTB-9, MBT2), breast cancer (SK-BR-3, MCF7), and other colon cancer (C26) cells [45,47]. However, their nonspecificity presents a significant challenge in systemically administered applications for targeted delivery, as it requires precise control of CPP presentation only at the target site to prevent toxicity [48]. To overcome this problem, several approaches are being investigated to activate CPPs only at the target site. Stimulus-responsive materials may be used for this purpose, as they may provide triggered changes in material properties that allow spatially focused presentation of CPPs in response to intrinsic disease characteristics (e.g. abundantly present extracellular matrix proteases) or locally applied extrinsic cues (e.g. apply heat or light at a specific location) [49].
As demonstrated in this study, TAT peptides can enhance cellular uptake when conjugated to lyophilisomes. In previous in vitro studies, lyophilisomes were loaded with anti-tumor drugs, e.g. doxorubicin and curcumin, which could eliminate tumor cells. Antibodies were conjugated to lyophilisomes resulting in specific binding to target cells [11,12]. To probe the possibility of using lyophilisomes for the treatment of cancer, active targeting with specific antibodies and enhanced cellular uptake with CPPs can be combined with the drug delivery properties of lyophilisomes, thereby creating a potential powerful tool for drug delivery.

Conclusion
In the present study, albumin-based lyophilisomes were functionalized with TAT peptides to obtain a drug delivery system with enhanced cellular uptake. Lyophilisomes modified with TAT peptides efficiently bound to HeLa, OVCAR-3, Caco-2 and SKOV-3 cells. Additional cellular uptake studies were performed to verify that TAT-conjugated lyophilisomes were internalized in HeLa cells after binding. TAT-conjugated lyophilisomes may present a novel delivery system to ensure faster and higher cellular uptake of anti-tumor drugs to cancer cells.