Influence of Growth Characteristics of Induced Pluripotent Stem Cells

Influence of Growth Characteristics of Induced Pluripotent Stem Cells on Their Uptake Efficiency for Layer-by-Layer Microcarriers Uta Reibetanz,*,† Denise Hübner,‡ Matthias Jung,§ Uwe Gerd Liebert,‡ and Claudia Claus‡ †

Institute for Medical Physics and Biophysics, Faculty of Medicine, University of Leipzig, 04107 Leipzig, Germany Institute of Virology, University of Leipzig, 04103 Leipzig, Germany § Department of Psychiatry, University of Halle-Wittenberg, Halle, Germany ‡

S Supporting Information *

ABSTRACT: Induced pluripotent stem cells (iPSCs) have the ability to differentiate into any specialized somatic cell type, which makes them an attractive tool for a wide variety of scientific approaches, including regenerative medicine. However, their pluripotent state and their growth in compact colonies render them difficult to access and, therefore, restrict delivery of specific agents for cell manipulation. Thus, our investigation focus was set on the evaluation of the capability of layer-by-layer (LbL) designed microcarriers to serve as a potential drug delivery system to iPSCs, as they offer several appealing advantages. Most notably, these carriers allow for the transport of active agents in a protected environment and for a rather specific delivery through surface modifications. As we could show, charge and mode of LbL carrier application as well as the size of the iPSC colonies determine the interaction with and the uptake rate by iPSCs. None of the examined conditions had an influence on iPSC colony properties such as colony morphology and size or maintenance of pluripotent properties. An overall interaction rate of LbL carriers with iPSCs of up to 20% was achieved. Those data emphasize the applicability of LbL carriers for stem cell research. Additionally, the potential use of LbL carriers as a promising delivery tool for iPSCs was contrasted to viral particles and liposomes. The identified differences among those delivery tools have substantiated our major conclusion that LbL carrier uptake rate is influenced by characteristic features of the iPSC colonies (most notably colony size) in addition to their surface charges. KEYWORDS: LbL carrier, iPSC, carrier uptake, iPSC colony morphology, liposome, virus particle

T

issues associated with ESCs and can be used in a patient- or disease-specific manner. Currently, there are multiple approaches for the manipulation of pluripotent stem cells, including cell-penetrating peptides,5 the use of viral vectors or standard plasmid transfection,6 and the use of various polymer-based systems for DNA delivery7 up to biodegradable polymeric nanoparticles.8,9 Each method has its own advantages and drawbacks. On one hand, while lentiviral vectors represent one of the most efficient viral vectors,10 they generally carry the risk of reactivation and continuous expression of viral genes. Additionally, and most importantly, mutations can result from the integration of the viral vector into the host genome.11 On

he generation of human induced pluripotent stem cells (hiPSCs) through the introduction of Oct4 (octamerbinding transcription factor 4), Sox2 (SRY-related HMG-box gene 2), Klf4 (Kruppel-like factor 4), and c-Myc as four defined transcription factors (OSKM) in differentiated fibroblasts has revolutionized scientific research on human diseases and pathological outcomes in multiple ways.1 Like embryonic stem cells (ESCs) iPSCs have the ability to differentiate into endo-, meso-, and ectodermal lineages and thus into all somatic cell types. Very recently the question on how similar iPSCs and ESCs are has been resolved, highlighting that any previously reported differences are probably due to the genetic background of the respective donor of the somatic cells used for reprogramming.2,3 This makes iPSCs even more attractive for basic and medical research, including drug discovery, regenerative and transplantational (personalized) medicine, and modeling of human diseases,4 as they lack ethical © 2016 American Chemical Society

Received: February 8, 2016 Accepted: June 30, 2016 Published: June 30, 2016 6563

DOI: 10.1021/acsnano.6b00999 ACS Nano 2016, 10, 6563−6573

Article

www.acsnano.org

Article

ACS Nano

Figure 1. Assessment of pluripotent properties of the iPSC colonies. (A) Verification of the pluripotency markers SSEA4, Oct4 (shown in red), and Sox2 (shown in green) through immunofluorescence analysis. DNA was labeled with Hoechst (shown in blue). (B) Morphology of iPSC colonies 3 days after plating. (C) Analysis of the mitochondrial membrane potential of 4-day-old colonies with the dye JC-10. The aggregated form shows fluorescence in the TRITC channel and is indicative of a high membrane potential. Scale bar: 100 μm.

morphologically distinct from the center but also physiologically. The rather compact and rounded up structure of colonies does not leave much space for flexibility and restricts access to the cytoplasm.25 Moreover, the marginal cells put pressure on the cells in the center. This pressure increases during growth of the colonies and makes it difficult for exogenous materials to be internalized.26 Hence the aim of our study is a fundamental investigation of the ability of iPSCs to interact with LbL carriers. SiO2 particles were coated with PAH (poly(allylamine hydrochloride)) and PSS (poly(styrenesulfonate sodium salt)). The fluorescencelabeled component PAH-FITC served as a pH sensor. Thus, the intra- and extracellular stability of the PAH/PSS multilayer provides optimal conditions for carrier localization inside the cell. The pH-dependent fluorescence intensities can be directly correlated with the carrier-containing cell compartment without the consideration of intensity loss due to polymer disassembly. These unwanted effects are more likely to occur with biodegradable polymers as LbL building blocks.27,28 Despite the small size of the cells (roughly 10 μm in diameter), up to 20% of the iPSCs are able to interact with those LbL carriers, and subsequently a distinct portion is efficiently internalized. Nevertheless, the charge of the outer layer, carrier number, and co-incubation conditions have a major influence on uptake behavior. However, LbL carriers do not seem to alter colony properties such as the size and morphology of the colonies. We could substantially support with our data the potential application of LbL carriers as a highly promising delivery vehicle for iPSCs. Without affecting properties of the iPSC colonies, they provide a highly promising application platform for a wide range of cargo, including its time-restricted release.

the other hand, most of the nonviral delivery methods suffer from a low delivery efficiency.6,9,12 Another important issue to be solved is that some applications for iPSCs require a time-controlled as well as time-restricted expression of the heterologous protein instead of prolonged expression levels. We have addressed these prevailing obstacles through layer-by-layer (LbL) carriers that are so far unexplored for their applicability to pluripotent stem cells. The LbL technique is based on self-assembly of various materials onto spherical templates to build up a (multi)layered structure and was introduced by Decher.13 Many new assembly methods and materials have been investigated.14,15 In particular, their biomedical application as a drug delivery system strongly benefits from the strategy of a modular assembly of different biopolymers onto templates of various sizes and materials as well as the template dissolution to produce hollow shells (capsules), if required.16,17 Such a design facilitates not only the integration of one or multiple types of active agents but also the addition of functional components (e.g., surface antibody modification for specific uptake) as well as reporter or sensor molecules (e.g., carrier tracking), and even nanoparticles (e.g., induction of release, carrier tracking) within different positions of the multilayer and core.17,18 Polymers or a linker can be selected such that they are able to react to changes of intracellular pH or redox potential, by which a controlled release through step-by-step disassembly or burst within specific cellular subcompartments can be induced.19,20 Even multicompartment polymeric capsules can be designed to enhance the carrier’s functionality through a more efficient combination of sensor, drug delivery, and reactor functions.21 A wide variety of applications already show the potential of those carriers for a specific and individual transport and release of bioactive agents.22−24 iPSCs are substantially different from their differentiated counterparts. Especially their growth in rather tight colonies poses an obstacle, because the periphery of a colony is not only

RESULTS For assessment of LbL carriers as a delivery tool for pluripotent stem cells, we have used the Gibco iPSC cell line, which was generated through application of episomal factors. Figure 1A 6564

DOI: 10.1021/acsnano.6b00999 ACS Nano 2016, 10, 6563−6573

Article

ACS Nano

Figure 2. Distribution of LbL particles on iPSC colonies in the direct and reverse plating approach. The green signal of the LbL particles is a result of the FITC label within the LbL layer. Positively and negatively charged LbL particles (4 × 106 particles per well of a six-well plate) were added to Matrigel before plating of iPSCs or at 24 or 72 h of cultivation of iPSCs. Microscopic analysis of LbL particles on smaller size colonies (A) in comparison to bigger colonies (B). Scale bars: 100 μm.

Information), in the absence of iPSCs LbL carriers were evenly spread on the Matrigel coat as the growth substrate for iPSCs. LbL carriers stick to the Matrigel coat and remain there even after several washing steps. Afterward, iPSCs can be seeded on top of the carrier-covered Matrigel (reverse approach). In the second (direct) approach, LbL carriers were applied after seeding and cultivation of the iPSCs (direct approach24h after 24 h of cultivation and direct approach72h after 72 h). Again, after sedimentation carriers are evenly distributed over the Matrigel coat. In the direct approach, LbL carriers are dissolved in the medium and added to the cells, while in the reverse approach carriers are applied to the Matrigel coat. Thereafter cells are plated on an LbL-covered surface. Both approaches have different effects on the mode of interaction with iPSCs. In the direct approach, iPSCs are allowed to spread out as colonies before getting in contact with LbL carriers, while in the reverse approach they are in direct contact with LbL carriers before they settle down on the Matrigel coat. Thereafter these three experimental setups were used to investigate the carrier’s ability to interact with iPSCs over time of cultivation and to show their potential use as a drug delivery system through verification of several performance characteristics. They comprise analysis of the overall internalization rate and pattern in response to the number, surface charge, and the application approach of the carriers. a. Influence of the iPSC Colony Size on Their Interaction with LbL Carriers. In a first investigation, the lateral carrier distribution on cells was investigated in consideration of the charge of the carrier surface layer and the size of the interacting iPSC colony. In all approaches, iPSCs have been incubated with carriers for 72 h. This time frame was generally chosen to enable endosomal uptake, acidification of

highlights the expression of Oct4, SSEA4 (stage-specific embryonic antigen-4), and Sox2 as important pluripotency factors. These pluripotent properties were further emphasized through verification of the morphology of the colonies, which is represented by a three-dimensional structure with a defined rim region (Figure 1B). The cells in the center of an iPSC colony are distinct from the rim, in that the first type is epithelial-like, while the latter is rather mesenchymal-like.29 In comparison to their differentiated counterparts iPSCs rely on glycolysis instead of mitochondrial respiration.30 This results in a low mitochondrial membrane potential (ΔΨm). For further characterization of iPSC colonies, ΔΨm was assessed through the cationic dye JC-10 (Figure 1C), which forms aggregates in the presence of a high ΔΨm. Over the time of cultivation the cells in the rim become distinguishable from the cells in the center. They represent the only portion within an iPSC colony that develops an increase in ΔΨm. To illustrate the conditions of the iPSC/carrier interaction, SiO2 particles with a template size of 2.76 μm were used and coated with PAH and PSS as a well-studied polymer system. Equipped with a fluorescence-labeled component (PAHFITC), pH-dependent fluorescence properties can be employed for carrier characterization. PAH/PSS-coated LbL carriers are thereafter referred to as LbL carriers or carriers, if not otherwise stated. Interaction in our understanding applies to both a mere binding to the cell surface and uptake through internalization and subsequently presence in phagolysosomes and cytoplasm. For those experiments, the distribution of these fluorescent carriers (exemplarily shown for 4 × 106 carriers per well) on Matrigel-coated culture wells in the presence or absence of iPSCs was visualized according to the different experimental setup. As shown in Figure S1-A (Supporting 6565

DOI: 10.1021/acsnano.6b00999 ACS Nano 2016, 10, 6563−6573

Article

ACS Nano

Figure 3. Flow cytometric histograms of region M reflecting cell/carrier interaction. M was analyzed in the presence of 2 × 106 carriers (left column: negatively charged outer layer, right column: positively charged outer layer). Black lines represent measurements under physiological conditions; red lines illustrate the sample’s response to FITC quenching after trypan blue (TB) application. A, B, and C mark the different carrier approaches as reverse and direct (24 h, 72 h). The subdivision of region M into M1 and M2 focuses on carriers in phagolysosomes (M1) and cytoplasm/outer membrane (M2), depending on the pH of the environment. A partial shift of M2 toward lower intensities after TB application indicated the amount of TB-accessible carriers, and the external carrier rate can be approximated.

endolysosomes, and subsequently carrier release into the cytoplasm. As can be seen in Figure 2, independent from the abovementioned parameters and experimental approaches (number and surface charge of the LbL carriers in direct and reverse approaches), two effects can be observed: (1) The cells appear to attract the carriers located around the colonies, such that not only carriers in close proximity to cell membranes will be affected but also carriers at a distance of several micrometers. (2) The application of LbL carriers before (reverse approach) or after (direct approach) iPSC seeding revealed a different mode of lateral attraction, which was dependent on colony size. Carriers appear to be mostly homogeneously distributed over smaller colonies (up to 200 nm size, Figure 2A, upper row), whereas in the presence of bigger colonies (Figure 2B, lower row) the carriers will concentrate on the rim. As this observation occurred independently from the approach, an influence of colony growth (cell division) and a constant reduction of available carriers can be neglected. That is, in the case of iPSC cultivation for 72 h bigger colonies are dominantly present at the time point of addition of the LbL carriers (direct

approach72), where the same rim-like distribution was observable as for bigger colonies during carrier application before cell seeding (reverse approach). The influence of morphological properties of iPS colonies on the mere distribution pattern of LbL carriers was further highlighted through contrasting collagenase-passaged iPSCs to terminally differentiated cells and Accutase-passaged iPSCs. In contrast to collagenase treatment, which is based on passaging of iPSCs in clumps or clusters, Accutase propagation enables their dissociation into single cells and subsequently their cultivation as a monolayer without losing pluripotency.31 Both methods of enzymatic passaging of iPSCs were used for characterization of LbL distribution on iPSCs. LbL carriers were applied 24 h after plating and analyzed during an incubation period of up to 24 h. In contrast to the more rimcentered LbL carrier localization after collagenase-passaging, the distribution of LbL carriers on Accutase-passaged iPSCs more closely resembles the one found on HEK293T/17 cells as a representative terminally differentiated cell line (Figure S2, Supporting Information). b. Assessment of the Mode of Interaction. As a next step the uptake efficiency of FITC-labeled LbL carriers was quantitatively investigated by flow cytometry for all three approaches with regard to the distribution of fluorescence 6566

DOI: 10.1021/acsnano.6b00999 ACS Nano 2016, 10, 6563−6573

Article

ACS Nano

Figure 4. Relative rates of cells interacting with carriers based on regions M, M1, and M2. A and B represent the presence of a low amount of carriers (2 × 106 per well); C and D, a high amount of carriers (4 × 106 per well). The columns again show cell interaction with negatively charged carriers on the left and positively charged carriers on the right. Scale bars on the left and right of each graph were chosen according to presentation.

intensities of LbL carriers interacting with or internalized by iPSCs. The use of these fluorescence-labeled carriers enables analysis of their uptake as a function of selected carrier and iPSC colony properties. Figure 3 (black lines) exemplarily illustrates for 2 × 106 carriers per well the intensity distribution of the cell regions R1 (extracted from the dot plot in Figure S3, Supporting Information) for all approaches and carrier surface charges. Region M (cell/carrier interaction) of all approaches was subdivided into two populations, M1 and M2. Whereas M1 represents a narrow peak, M2 often forms a broad shoulder up to another distinguishable population toward higher intensities. In consideration of one carrier per cell, the detected fluorescence intensities can be assigned to specific pH values by carrier calibration measurements in defined buffer solutions (Figure S4, Supporting Information). In this given situation geometric mean (gmean) of M1 relates to pH 4−5 (typical for phagolysosomal environment) and the gmean of M2 to a pH value around 8 (typical for an extracellular and cytoplasmic environment). Detailed fluorescence intensity distributions are highlighted in Figure S5 (Supporting Information). To distinguish between internal and external carriers, trypan blue (TB) was used as a FITC quencher (Figure 3, red lines). The cell-impenetrable substance TB affects only carriers attached to the outside of the plasma membrane by reducing their fluorescence intensity. Intracellular carriers (in both phagolysosomes and cytoplasm) are not affected by TB exposure. The part of M2 that shifts after TB application toward a region of lower fluorescence intensity thus indicates TB-accessible carriers, leaving the cytoplasmic portion of the carriers within M2. With this design, the number of cells containing one carrier in phagolysosome can be easily

quantified (steady peak with lowest intensity). However, a number of cells will certainly interact with more than one carrier (visualized by the shoulder of M2 at higher fluorescence intensities). This renders peaks with higher intensities difficult to assign to specific pH values. In our experiments M1 (cells with one carrier in phagolysosome) could be assigned with certainty and was thus used representatively for quantification of the uptake efficiency. As can be seen in Figure 3 (now focusing on cell/carrier interaction and uptake rate related to the correspondent regions), after the reverse approach a more distinct region M1 can be observed and the fluorescence intensity distribution of neither positively nor negatively charged carriers was affected by TB. This indicates that besides a large amount of carriers in phagolysosomes, carriers are also already located in the cytoplasm and very few carriers are attached to the outer membrane. The same behavior was found for direct approach24h and direct approach72h, but only in the case of positively charged carriers. Negatively charged carriers, however, show a completely different behavior. The M2 peak is higher than M1 and shifts nearly completely toward lower intensities after TB application. Both facts indicate that under both direct approach conditions a large amount of carriers was adsorbed to the outside of the plasma membrane, where they are accessible to TB. Relative cell/carrier ratios of three independent measurements are shown in Figure 4 for the co-incubation of iPSC with 2 × 106 positively and negatively charged carriers (A and B, respectively) or with 4 × 106 positively and negatively charged carriers (C and D, respectively). The bars in the left part (I) of the diagrams in Figure 4 represent the total cell contacts with 6567

DOI: 10.1021/acsnano.6b00999 ACS Nano 2016, 10, 6563−6573

Article

ACS Nano

The highest rate of interaction of iPSCs with LbL carriers (I: M, 19.6 ± 1.1%) and the highest uptake rate as reflected by carriers contained in phagolysosomes (II: M1, 8 ± 0.4%) were found after application of negatively charged carriers before iPSC seeding (reverse approach), whereas both direct approaches were less successful (M: 12.0 ± 8%, 3.9 ± 2.0%, direct approach24h, M1: 3.6 ± 2.3%, 1.6 ± 0.9%, direct approach72h). Applying positively charged carriers to the cells, a maximum rate of interaction (M: 16.7 ± 3.8%) and phagolysosomal uptake (M1: 9 ± 1.4%) was found after direct approach24h as compared to reverse and direct approach72h (M: 17 ± 3.8%, 4.5 ± 2.1% and M1: 8.1 ± 2.4%, 2.8 ± 1.1%, respectively). In summary, negatively charged carriers appear to attach much better to the Matrigel coat than the positively charged counterparts. In turn, this increases not only the potential interaction events with iPSCs in the reverse approach but also the number of carriers attached to the plasma membrane. Therefore, while the interaction efficiency after either of the direct approaches is not influenced by the charge of the applied carriers, the efficiency of the reverse approach is dependently affected. It is important to note that the application of positively charged carriers shows in all approaches a negligible response to TB. In this case carriers can be predominantly found in phagolysosomes or were already released to the cytoplasm. Thus, in this case interaction rate is equal to uptake rate. d. Analysis of the Accessibility of Bigger Colonies to LbL Carriers. The co-incubation of carriers with iPSCs after 3 days of cultivation resulted in a lesser interaction and uptake rate as compared to 1 day of cultivation (Figure 4). Figure 2 highlights the inhomogeneous lateral carrier interaction with bigger iPSC colonies, which appears to be restricted to the rim, leaving the inner section of the colony almost carrier-free. This microscopic appearance was confirmed through flow cytometry results, revealing an overall low interaction and uptake rate.

carriers (region M, Figure S3, Supporting Information), while the bars in the right part (II) of the diagrams differentiate between cell/carrier behavior in the M1 and M2 region. c. Influence of LbL Characteristics: Carrier Number and Surface Charge. The values depicted in Figure 4 are listed in Table 1 as a mean value of at least three independent Table 1. Relative Cellular Interaction with LbL Carriers in Dependence on Charge and Number relative (% ± SD) cellular interaction intracellular localization phagolysosomal (M1 w/o TB)

LbL carrier type Reverse Approach positively 2× charged 4× negatively 2× charged 4× Direct Approach24h positively 2× charged 4× Nnegatively 2× charged 4× Direct Approach72h positively 2× charged 4× negatively 2× charged 4×

max. cytoplasmic (M2 w/ TB)

total interaction events (M)

106 106 106 106

5.1 8.1 6.9 8.0

± ± ± ±

1.4 2.4 1.0 0.4

3.0 6.4 5.6 7.7

± ± ± ±

1.0 2.4 1.4 1.4

8.4 17.0 14.0 19.6

± ± ± ±

2.0 3.8 1.0 1.0

106 106 106 106

7.1 8.9 2.7 3.9

± ± ± ±

1.7 1.4 1.7 2.0

4.8 6.0 2.2 4.5

± ± ± ±

3.4 2.1 1.8 3.6

13.5 16.7 6.7 12.0

± ± ± ±

5.6 3.8 4.8 8.0

106 106 106 106

2.3 2.8 1.3 1.6

± ± ± ±

2.0 1.1 0.9 0.9

1.7 1.4 0.5 0.9

± ± ± ±

1.5 0.8 0.5 1.0

3.6 ± 3.6 4.5 ± 2.1 2.9 ±1.6 3.6 ± 2.3

experiments. In general and as expected, for all application approaches a higher number of carriers leads to a more pronounced interaction rate with cells, resulting in an increased uptake rate.

Figure 5. Assessment of the effect of LbL carriers (4 × 106 negatively (A and B) and positively (C) charged carriers) on the pluripotency and viability of iPSC colonies. LbL carriers were added in the reverse (A and B) or direct (C) approach. Analysis was performed after three (A and B) and two (C) days of incubation. (A) Immunofluorescence analysis of the Oct4 pluripotency marker (shown in red) in LbL carrier (shown in green, indicated by arrows)-containing iPSCs. The pink color results from the overlay with the DNA stain. (B) Alkaline phosphatase staining of iPSC colonies. (C) Analysis of live and dead cell fluorescence through the MultiTox-Fluor Multiplex cytotoxicity assay after incubation of iPSCs with PAH/PSS carriers as compared to pARG/DXS carriers and the positive controls (lack of medium change and treatment with NP-40). 6568

DOI: 10.1021/acsnano.6b00999 ACS Nano 2016, 10, 6563−6573

Article

ACS Nano e. Effect of LbL Carriers on the Pluripotency of iPSCs and Assessment of General Cytotoxicity. LbL carriers do not affect colony morphology and shape in either one of the three LbL carrier application approaches. As a next step, maintenance of pluripotency as one of the outstanding hallmarks of iPSCs was verified after LbL carrier application to iPSCs. Immunofluorescence analysis was performed for Oct4 as one of the most important pluripotency markers (Figure 5A). For this assay the reverse approach as the most vulnerable period for LbL carrier application was chosen (4 × 106 carriers per well). Figure 5 shows that expression of the pluripotency marker Oct4 was not affected by the presence of LbL carriers. The expression level of Oct4 in cells interacting with an LbL carrier (indicated by an arrow) is comparable to the one found in cells lacking carriers. However, with this approach we cannot clearly differentiate between carriers with an intracellular location on one hand and carriers attached to the membrane on the other hand. The alkaline phosphatase assay (Figure 5B) not only verifies the pluripotent properties of iPSCs after incubation with carriers but also the unaltered distribution, morphology, and size of iPSC colonies after application of LbL in the reverse approach. For cytotoxicity assessment, the composition of the LbL carriers was taken into account. Thus, positively charged PAH/ PSS LbL carriers were compared to equally positively charged biocompatible pARG/DXS (poly-L-arginine/dextran sulfate) carriers. The positive charge was chosen as it could possibly induce a higher stress level on the negatively charged plasma membrane than LbL carriers with a negative outer charge. Figure 5C indicates that PAH/PSS LbL carriers not only are comparable to pARG/DXS carriers but also lack a notable induction of cytotoxicity. The results of the cell cytotoxicity assay were complemented by a lactate dehydrogenase (LDH) assay as a screen for possible loss of membrane integrity due to the relatively high LbL carrier diameter, which is roughly one-third of an iPS cell. No increase in extracellular LDH levels was detectable for the reverse uptake approach, neither after application of positively nor of negatively charged particles (data not shown). This indicates that no membrane disruption as a consequence of necrosis or apoptosis has occurred. f. Characterization of the Accessibility of iPSC Colonies to Cargo: Liposomes and Viral Particles. Given the size of LbL carriers we have further addressed the question of whether small-sized carriers (