Demonstration of intracellular trafficking, cytosolic bioavailability, and target manipulation of an antibody delivery platform

Abstract

Intracellular antibody delivery into live cells has significant implications for research and therapeutic applications. However, many delivery systems lack potency due to low uptake and/or endosomal entrapment and understanding of intracellular delivery processes is lacking. Herein, we studied the cellular uptake, intracellular trafficking and targeting of antibodies using our previously developed Hex antibody nanocarrier. We demonstrated Hex-antibodies were internalized through multiple endocytic routes into lysosomes and provide evidence of endo/lysosomal disruption and Hex-antibody release to the cytosol. Cytosolic antibodies retained their bioactivity for at least 24 h. Functional effect of Hex delivered anti-STAT3 antibodies was evidenced by inhibition of nuclear translocation of cytosolic transcription factor STAT3. This study has generated understanding of key steps in the Hex intracellular antibody delivery system and will facilitate the development of effective cytosolic antibody delivery and applications in both the therapeutic and research domains.

Graphical Abstract

The Hex carrier can bind three antibodies to form a protein complex and deliver antibody into cells. We have tracked Hex-antibody in living cells and demonstrated that the Hex-antibody employed multiple endocytic pathways to enter cells, and then sorted to the endo-lysosomal pathway. Some Hex-antibodies disrupted endosomes and released to the cytosol. Additionally, we delivered Hex-STAT3 antibodies into living cells, which blocked the translocation of STAT3, a transcription factor to the nucleus. This study contributes to the understanding the protein nanoparticle trafficking in living cells and enables the application of antibodies to intracellular targets.

Background

Antibodies have significant value as therapeutics and basic research and diagnostic tools due to their binding specificity ¹. However, poor cell membrane permeability and cellular recycling processes restrict their therapeutic and diagnostic use to extracellular targets and research use to fixed, permeabilized cells and tissues ². There is a significant list of intracellular targets for which research antibodies exist, including those targeting “undruggable” proteins, disease related protein-protein interactions and cytosolic pathogens ³. In order to realize the full potential of antibodies, various technologies have been developed for intracellular delivery of antibodies, antibody fragments, or antibody-like molecules ^4–9. However, two major challenges remain. First, a sufficient number of antibodies should be delivered across the cell membrane. Low efficiency of delivery can limit practical applications, whether therapeutic or research ^10,11. Second, due to high molecular weight and polarity of antibodies, both free antibodies and most carrier-antibody complexes are internalized by endocytosis processes. Internalized antibodies are often trapped in endosomes and lysosomes, which prevents their access to cytosolic targets and, thus, inhibits desired biological activity ³. For example, cell penetrating peptides (CPPs) are known to cross the cell membrane by direct membrane penetration ^8,12. However, it was experimentally proven that antibodies fused with cell penetrating peptides are primarily taken up by endocytosis and only a small amount of antibody could escape this degradative pathway ¹³. In addition, the effects of endosomal escape on the subsequent intracellular trafficking of the antibody and/or its carrier remain unclear. Therefore, with many current strategies exhibiting endosomal entrapment, insight into each of the intracellular delivery steps is essential to determine delivery efficacy, evaluate intracellular bioavailability, and identify methods for overcoming these challenges.

We recently reported a protein-based self-assembling antibody carrier that consists of a Hex domain and six Fc-binding Staphylococcal protein A domain B (SPAB) domains ¹⁴. The Hex nanocarrier assembles with 3 antibodies to form a small complex, ~25 nm, with a high antibody/carrier mass ratio, and stability under physiological conditions ¹⁵. Hex-antibody complexes exhibited internalization and cytosolic target localization in living cells. However, the mechanisms and kinetics of Hex-antibody complex cellular uptake, trafficking and cytosolic localization have not been investigated. In the present study, we address these aspects and, in addition, evaluate of the capacity of Hex delivered antibody to target and manipulate an important intracellular antigen, STAT3. These results provide insight into the mechanisms of Hex-antibody complex cellular uptake, intracellular trafficking and delivery to and influence on cytosolic targets. They facilitate the development of rational strategies for improving the cytosolic delivery and applications towards valuable intracellular therapeutic, diagnostic or research targets.

Methods

Materials

Rabbit immunoglobulin G (IgG) was purchased from Sigma-Aldrich. Polyclonal rabbit anti-nuclear pore complex protein antibody (aNPC) was obtained from Bioss antibodies Inc (bs-11678R). Monoclonal rabbit IgG_2B human phospho-STAT3 (pY705) antibody (aSTAT) (MAB46071) and monoclonal mouse IgG_2B human phospho-STAT3 (pS727) antibody (MAB4934) were kind gifts of R&D Systems, Inc. Degree of labeling (DOL) of 0.5 was used for all labelling with Alexa Fluor™ 488 5-SDP Ester (AF488, Thermo Fisher Scientific) or Alexa Fluor™ 647 NHS Ester (AF647, Thermo Fisher Scientific). All chemical reagents were obtained from Thermo Fisher Scientific or Sigma-Aldrich, and were used as received unless noted.

Cell Lines

SK-BR-3 (ATCC® HTB-30) and MDA-MB-231 (ATCC® HTB-26) cell lines were purchased from American Type Culture Collection (ATCC). Yellow fluorescent protein-Galectin-8 (YFP-Gal8) MDA-MB-231 cell line was generously provided by Professor Craig Duvall from Vanderbilt University. All cell lines were cultured in corresponding culture media (SK-BR-3: McCoy’s 5A Medium (Gibco); MDA-MB-231: Dulbecco’s Modified Eagle Medium (Gibco)), supplemented with 10% (v/v) fetal bovine serum (FBS) and 5% penicillin/streptomycin at 37 °C and 5% CO₂ according to ATCC protocols. Turkey red blood cells (RBCs) (washed 5% suspension) were obtained from LAMPIRE Biological Laboratories, Inc.

Preparation and Characterization of Hex-antibody Complex

The Hex nanocarrier was prepared as previously described ¹⁴. Hex-SPAB and SPAB-Hex were expressed in Top10 Escherichia coli. Protein was collected and purified using Ni-NTA (Qiagen) affinity chromatography under native conditions according to the manufacturer’s instructions. Protein purity was confirmed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Purified Hex-SPAB and SPAB-Hex assemblies were disassembled with 10% w/v solution of sodium dodecyl sulfate (SDS) and mixed in a 1:1 molar ratio. To assemble the Hex nanocarrier, the SDS was removed by passing through a PD-10 column (GE Healthcare) twice to a final volume of 3.5 mL Hex in phosphate buffered saline (PBS). Hex nanocarrier was syringe filtered (0.2 μm) and stored at 4 °C in PBS for no longer than 2 months. Hex nanocarrier was mixed with rabbit serum IgG at a molar ratio of 1:3 and incubated at 37 °C for 48 h to facilitate the formation of stable Hex-IgG complexes. The Hex nanocarrier and Hex-IgG were characterized with dynamic light scattering (DLS, Zetasizer Nano ZS, Malvern Instruments) and scanning transmission electron microscopy (TEM, FEI Tecnai F30) (Details in Supplementary Section 2).

Live Cell Imaging

Typically, 1 × 10⁴ cells (SKBR-3, MDA-MB-231, YFP-Gal8 MDA-MB-231) per well were seeded in a 35 mm 4 chamber glass bottom dishes (Thermo Fisher Scientific Inc.) at 37 °C for 24 h before treatment. Unless specified otherwise, all cellular assays were performed by washing with PBS and replacing with fresh, serum-supplemented culture media with Hex-IgG (or Hex-IgGAF488, Hex-aNPC, Hex-aNPCAF488) at a final antibody concentration of 0.6 μM. The sample was incubated for the indicated period of time, washed with cell culture media, and imaged by confocal microscopy (Carl Zeiss LSM 700, or spinning disk confocal microscopy system (PerkinElmer UltraVIEW VoX)) in a live cell imaging chamber (37 °C, 5% CO₂). Nuclei were stained with 16.23 μM Hoechst 33342. Whole cells were visualized using differential interference contrast (DIC) in the RBC assay and intracellular bioactivity and fate of Hex-IgG experiments. For Gal8 recruitment assays, images were analyzed using Fiji imageJ. Background subtraction was performed to suppress the diffuse cytosolic fluorescence and Gal8-positive puncta were identified by thresholding.

To evaluate the internalization mechanism of Hex-IgG, SKBR-3 cells were co-incubated with Hex-IgG and endocytosis inhibitors for 3 hrs, following pre-incubation with inhibitors alone for 1 hr. Macropinocytosis was inhibited using 2 mM amiloride (MP Biomedicals, LLC), caveolae-mediated endocytosis using 300 μM genistein (TCI America), and Clathrin-mediated endocytosis with 20 μg/ml chlorpromazine (Alfa Aesar). Cells were analyzed by a CytoFLEX flow cytometer (Beckman Coulter, Inc.) and endocytosis was quantified as the mean fluorescence of the samples. Hex-IgGAF488/Ethidium homodimer-1 (EH) incubated with RBCs to determine the cell membrane disruption (See supplement for more details).

For intracellular trafficking experiments, red fluorescent protein (RFP) tagged Ras-associated binding proteins 5 (Rab5-RFP) and RFP tagged lysosome associated membrane protein-1 (LAMP1-RFP) fusion constructs (CellLight™ Early Endosomes-RFP, BacMam 2.0 and CellLight™ Lysosomes-RFP, BacMam 2.0, Invitrogen) were transfected into SKBR-3 cells according to the manufacturer’s instructions. Cells were exposed to Hex-IgGAF488 for 1 h, washed and monitored using spinning disk confocal microscopy for 4 h. Three consecutive z-stacks (20 optical sections) were imaged at each time point. Colocalization was defined as the fraction of cell-associated Hex-IgGAF488 (green pixels) colocalized with Rab5-RFP or LAMP1-RFP structures (red pixels) at each time point. It was quantified by the Manders’ coefficient of colocalization using Fiji-ImageJ software ¹⁶. The result at each time point is the average of data from four individual cells.

Targeted Manipulation of STAT3 Antigen in Living Cells

MDA-MB-231 cells were incubated with PBS or 0.3 μM Hex-aSTAT3 (pS727, mouse) for 24 h. Cells were washed, fixed with 4% paraformaldehyde, and permeabilized with 0.1% Triton X-100. Hex-aSTAT3 (pS727) treated cells were incubated with rabbit aSTAT3 (pY705) antibody. PBS treated cells were incubated with mouse aSTAT3 (pS727, mouse) antibody and Rabbit aSTAT3 (pY705) antibody. After 1 h of incubation, goat anti-mouse IgG H&L (Alexa Fluor® 488, green) and goat anti-rabbit IgG H&L (Alexa Fluor® 647, red) were used as secondary antibodies to label STAT3 antibodies. Images were acquired by confocal microscopy and the colocalization was analyzed using Fiji-ImageJ software. The STAT3 distribution was quantified as the mean fluorescence of STAT3 proteins in the cytosol and Hoechst signal indicated nuclear area.

Statistical Analysis

All data collected from experiments were pooled together and expressed as mean ± SD with n =3, unless otherwise indicated. Statistical analyses in the internalization mechanism studies were performed using ANOVA followed by Tukey’s post hoc test for multiple comparisons. Statistical significance was determined at p < 0.05.

Results

Hex-antibody Complex Characterization

The Hex nanocarrier is a recombinant protein complex made from six α-helical domains that self-assemble into a hexameric coiled coil and are each fused to SPAB, which non-covalently binds the Fc region of antibodies (Figure 1A, S1) ¹⁴. The antibody binding SPAB domains are connected to the coiled-coil Hex bundle with flexible peptide linkers consisting of glycine and serine repeats (GGGS)₄ ¹⁷. The Hex nanocarrier was mixed with rabbit immunoglobulin G (IgG) with molar ratio of 1:3 and incubated at 37 °C for 48 h to facilitate the formation of stable Hex-IgG complex, as confirmed in prior work by size exclusion chromatography and analytical ultracentrifugation ¹⁵. The hydrodynamic diameters of Hex nanocarrier and Hex-IgG were confirmed by DLS to be 10.4 ± 2.6 nm (PDI: 0.22 ± 0.02) and 30.6 ± 5.7 nm (PDI: 0.29 ± 0.03), respectively. The shape of Hex nanocarrier and Hex-IgG were visualized by negative stain-TEM (NS-TEM). Figure 1C shows the elongated, non-spherical appearance of the Hex nanocarriers and larger Hex-IgG complexes.

Figure 1.

Schematic and characterization of Hex nanocarrier and Hex-antibody complex. (A) Schematic illustration of Hex nanocarrier consisting of a coiled coil hexamer domain (PDB ID: 3R47), (GGGS)₄ linkers, and six Fc-binding SPAB domains (PDB ID: 1BDC). The 6x His tag is omitted for clarity. Not drawn to scale. (B) Hydrodynamic size distribution of the Hex nanocarrier determined by DLS in the absence or presence of rabbit serum IgG antibody at 1:3 molar ratio. (C) NS-TEM images of Hex nanocarrier and Hex-antibody complex. The His tags are labeled by 1.8 nm gold nanoparticles. Individual Hex nanocarriers and Hex-antibody complexes are circled. Scale bar represents 50 nm.

Uptake Mechanism and Intracellular Trafficking of the Hex-IgG Complexes

We previously performed cellular uptake experiments at 4 °C and confirmed that the uptake of Hex-IgG is an energy-dependent process in HeLa cells ¹⁴. To investigate the uptake mechanism, three endocytosis inhibitors were used to determine the cell entry route of Hex-IgG. Amiloride is an inhibitor of Na⁺/H⁺ antiport used to inhibit micropinocytosis ¹⁸. Chlorpromazine can disrupt the formation and budding of clathrin-coated pits and has been used as a clathrin-mediated endocytosis inhibitor ¹⁹. Genistein is a tyrosine-kinase inhibitor and inhibits caveolae-mediated endocytosis by locally disrupting the actin network and impeding recruitment of dynamin II ²⁰. Using concentrations of each inhibitor determined to be non-toxic (Figure S2), SKBR-3 cells were pretreated with inhibitors for 1 h at 37 °C to inhibit the corresponding endocytosis pathways. Following 3 h of exposure with Hex-IgGAF488 and inhibitors, spinning disk confocal images showed low or undetectable levels of green fluorescence while untreated cells showed the punctate fluorescence spots typical of endocytic uptake (Figure 2A). Quantitative assessment by flow cytometry demonstrated that all inhibitors significantly reduced Hex-IgGAF488 uptake (Figure 2B). Particularly, uptake of Hex-IgGAF488 in the presence of chlorpromazine was completely inhibited, indicating that cellular uptake is preferentially mediated by clathrin-mediated endocytosis. Amiloride and genistein did not completely block the cellular uptake of Hex-IgGAF488, but reduced it to approximately 60 % and 70 % of non-inhibitor treated cells, respectively. These data suggest that macropinocytosis and caveolae-mediated endocytosis are also routes of cellular entry of Hex-IgGAF488.

Figure 2.

Uptake mechanism of Hex-IgG in SKBR-3 cells. (A) Confocal images of SKBR-3 cells treated with endocytosis inhibitors and Hex-IgGAF488. Control cells were not incubated with Hex-IgGAF488 or inhibitors. Scale bars represent 20 μm. (B) Quantification of cellular uptake of Hex-IgG complex with/without inhibitors. * indicates statistical significance compared to Hex-IgGAF488 group (no inhibitors) by one-way ANOVA with post-hoc Tukey’s HSD test, p<0.05.

Red blood cells (RBCs) are known to lack fluid phase or receptor mediated endocytosis and can be used to identify internalization by non-endosomal pathways such as direct penetration or membrane fusion ²¹. Confocal images in Figure 3 illustrate that Hex-IgGAF488 treated RBCs maintained normal morphology and neither the membrane integrity indicator, ethidium homodimer-1 (EH), nor Hex-IgGAF488 entered the RBCs. In contrast, both EH and Hex-IgGAF488 entered fixed and permeabilized control RBCs. These results proved that at physiological conditions, Hex-IgG does not disrupt the cell membrane and suggest it is unlikely that intracellular delivery of Hex-IgG to SKBR-3 cells was achieved by direct penetration or fusion with the cell membrane. This is consistent with our previous observations and endocytosis inhibitor results, that altogether demonstrate that intracellular delivery of Hex-IgG is an energy-dependent, endocytic process ¹⁴.

Figure 3.

Red blood cell membrane disruption assay. Turkey RBCs were incubated with PBS, Hex-IgGAF488 and IgGAF488. Fixed and permeabilized RBCs incubated with Hex-IgGAF488 served as control. Cell membrane-impermeable fluorescent dye ethidium homodimer (EH) was added in all groups. Scale bar represents 20 μm.

Following endocytic uptake, Hex-IgG intracellular trafficking was observed over time using confocal microscopy. To visualize the intracellular pathway taken by Hex-IgG and avoid artifacts caused by cell fixation and organelle staining ^22–25, we transiently transfected SKBR-3 cells with plasmids encoding Rab5-RFP and LAMP1-RFP, which are fluorescent fusions of proteins associated with endosomal-lysosomal trafficking ^26–28. Rab5 is an early endosome marker found on early endosomes, phagosomes, and caveosomes ^29,30. Uptake of Hex-IgGAF488 and colocalization with RFP labeled endosomes were detected in some cells as early as 5 min after treatment (Figure 4A). To better visualize intracellular trafficking of Hex-IgG without confounding signal of newly internalized protein, 60 min pulse-chase experiments were performed. Confocal images and the time-lapse colocalization profile showed an increase in colocalization of Hex-IgGAF488 and Rab5-RFP labeled endosomes up to 90 min after incubation, which reached a plateau at about 120 min and declined at 210 min (Figure 4B, D, S3). This profile clearly suggested that the majority of Hex-IgGAF488 accumulated in Rab5-positive structures after internalization. The decline phase indicated that Hex-IgGAF488s escaped from Rab5-positive structures, either to the cytosol or to the next structures in the degradative pathway. However, a considerably high level of Hex-IgGAF488 remained in Rab5-positive structures after 4 h, revealing that a large amount of Hex-IgGAF488 was still in endosomes. Using the same approach, LAMP1-RFP expressing SKBR-3 cells were used to investigate the lysosome trafficking of Hex-IgG after a 60 min pulse of Hex-IgGAF488. LAMP1 is a major protein component of the lysosomal membrane and is used as a lysosome biomarker ³¹. Colocalization of Hex-IgGAF488 and LAMP1-RFP labeled lysosomes remained constant at low levels for the initial 120 min following incubation. Afterward, a rapid and steady increase in colocalization was observed up to 210 min, that began to plateau at 240 min (Figure 4C, D, S4). The increase in colocalization of Hex-IgGAF488 with lysosomes coincident with the plateau and decline phases in endosome colocalization at 120 min suggests that Hex-IgGAF488 is sequentially trafficked from Rab5-RFP labeled endosomes to LAMP1-RFP labeled lysosomes. We did not observe a decline in colocalization between Hex-IgGAF488 and lysosomes up to 4 h, indicating that Hex-IgGAF488 may be trapped in LAMP1-positive structures or the protein complexes were degraded but the fluorescent dye AF488 remained in lysosomes. It is worth noting that this is the trafficking and final fate of only a portion of intracellular Hex-IgG in the degradative pathway, since at 4 h significant Hex-IgG is still present in endosomes and also, as shown below, nuclear membrane accumulation of nuclear pore targeted Hex-IgG. The Hex-IgG complexes escaped from the endosome-lysosome pathway may be too diffuse to be readily detectable by microscopy or may have been cleared from the cytosol. It is also possible that endosomal disruption by Hex-IgG may trigger autophagy and localization in autophagosomes that fuse with lysosomes or trafficked to the extracellular space by exocytosis ³².

Figure 4.

Hex-IgGAF488 uptake and intracellular transport to endosomes/lysosomes over time. (A) Representative confocal time-lapse images of Rab5-RFP expressing SKBR-3 cells treated with Hex-IgGAF488 for 5 min. (B,C) Center slice images and maximum intensity projections (MIPs) of representative 3D stack images of Rab5-RFP expressing (B) and LAMP1-RFP expressing (C) SKBR-3 cells after incubation with Hex-IgGAF488 for 1h and washing. Nuclei were labeled with Hoechst 33342 (blue). All scale bars represent 20 μm. Images from additional timepoints are in Figures S2 and S3. (D) Time-dependent colocalization profiles where 0 minutes indicates the end of the Hex-IgGAF488 pulse. A value of 0 represents not colocalized, a value of 1.0 represents complete colocalization. Each time point is the average of four individual cells from three frames. The error bars represent the standard deviation of the averaged frames.

Endosomal Disruption of Hex-IgG Complexes

To determine the fate of lysosomal Hex-IgG, as well as its intracellular bioavailability, we utilized a YFP-Gal8 reporter MDA-MB-231 breast cancer cell line, which was developed for direct visualization of endosomal disruption events in vitro ^33,34. Gal8 is a cytosolic lectin that selectively binds to glycans located on the inner side of the endosomal membrane when the endosomal membrane is disrupted, forming a punctate spot at the endosome ³⁵. We investigated endosomal disruption events over 25 min in YFP-Gal8 MDA-MB-231 cells incubated with IgG, Hex nanocarrier, and Hex-IgG. Bright yellow punctate Gal8 spots were observed in Hex-IgG treated cells (Figure 5A, B) after 5 min, indicating these vesicles lost integrity and YFP-Gal8 proteins were recruited into the vesicles. Furthermore, over time Hex-IgG treated cells exhibited an increasing trend of YFP-Gal8 recruitment events. In contrast, IgG and Hex treated cells showed dramatically less, but non-zero, YFP-Gal8 recruitment over 25 min (Figure 5C). The results show that endosomal disruption by Hex-IgG coincided with endosomal trafficking and endosomal disruption events increase with increasing Hex-IgG internalization over time. We also evaluated Hex-IgGAF488, Hex-AF488, and IgGAF488 endosomal disruption events over 24 h of incubation. Only Hex-IgGAF488 and, to a lesser extent, Hex-AF488 exhibited significant Gal8 recruitment (Figure 5D, E).

Figure 5.

Endosomal escape of Hex, IgG, and Hex-IgG complexes in YFP-Gal8 expressing MDA-MB-231 cells. (A) Time-lapse confocal images exhibiting Gal8 recruitment, a signal of endosomal escape, as well as quantification of Gal8 recruitment events for 25 min. White triangles indicate Gal8 recruitment events (bright, punctate fluorescent spots). Scale bar represents 20 μm. Blue bar indicates time of treatment. (B) Digitally magnified images of Gal8 recruitment at 25 min. White triangles indicate Gal8 recruitment events. Scale bar represents 20 μm. (C) Quantification of Gal8 recruitment events from time-lapse confocal images. (D) Confocal images and quantification of Gal8 recruitment events after treatment with Hex-IgGAF488 for 24 h. Nuclei were labeled with Hoechst 33342 (blue). Scale bar represents 5 μm. (E) Quantification of Gal8 recruitment events of 24 hrs images. * p<0.05 (F) Confocal images taken every 4 s of cells incubated with Hex-IgGAF488. Red rectangle shows real-time YFP-Gal8 recruiting events. Scale bar represents 5 μm.

The details of individual disruption events were assessed by taking images of YFP-Gal8 expressing MDA-MB-231 cells incubated with Hex-IgGAF488 every 4 s. At this time scale, we observed endosomal disruption events that originated from the sudden increase in fluorescent intensity of YFP-Gal8 recruitment and subsequent dispersal of the punctate YFP-Gal8 signal (Figure 5F). These results established that any Hex-IgG released during endosome disruption likely occurred within 4–8 s of Gal8 recruitment detection.

Intracellular Bioactivity and Fate of Hex Nanocarrier with Functional Antibody

To correlate the endosomal disruption events with intracellular bioactivity of Hex-IgG that requires endosomal escape, we first compared the intracellular biodistribution of AF488 labeled aNPC antibodies and non-specific IgG that does not have a binding target. Confocal images revealed that internalized Hex-IgGAF488 displayed both punctate and dispersed distributions in cells at all time points (Figure 6A). At 30 min the Hex-aNPCAF488 intracellular distribution looks fairly similar to that of Hex-IgGAF488. However, by 4 hours Hex-aNPCAF488 was found to localize around the outer nuclear envelope, forming bright ring-like structures, which demonstrated that escaped antibodies conserved their antigen-binding capability. Furthermore, by 24 hrs most of the aNPCAF488 antibody signal was visible around the nuclei and very little was seen elsewhere in the cells. This shows that over time, a significant amount of escaped aNPC antibody accumulated around the nuclei, and that non-escaped aNPC antibodies were eliminated or expelled by the endosome/lysosome system.

Figure 6.

Intracellular biodistribution and fate of Hex-antibody complexes. (A) Intracellular distribution of Hex-aNPCAF488 and Hex-IgGAF488 in SKBR-3 cells after 30 min, 4 h and 24 h treatment. (B) Intracellular distribution of HexAF647 and aNPCAF488 in SKBR-3 cells after incubation with HexAF647-aNPCAF488 for 30 min, 4 h and 24 h. Nuclei were labeled with Hoechst 33342 (blue). Scale bars represent 5 μm.

Our previous work showed that antibodies retained nanomolar affinity to the Hex nanocarrier even at acidic pH ¹⁴. However, it was unknown whether antibodies remained bound to the Hex nanocarrier throughout trafficking to their final target. To address this, we labeled Hex nanocarrier and aNPC with AF647 and AF488, respectively, and incubated with SKBR-3 cells. At 30 min and 4 hr, the signals were largely colocalized, as indicated by yellow punctate spots as well as some diffuse yellow signal (Figure 6B). There were also some diffuse red spots and green spots that appeared not to be colocalized. This observation implied that either free IgG, rather than intact Hex-IgG, escaped into the cytosol and/or that Hex-IgG escaped but Hex and IgG disassociated in the cytosol. Surprisingly, by 24 hrs no HexAF647 signal was observed in cells, only aNPCAF488 signal was visible around the nuclei, showing that the complexes do disassociate, and Hex was cleared from cells.

Targeted Manipulation of Antigen in Living Cells

With evidence that bioactive Hex-aNPC escapes the endosomal pathway and binds to its target, we next asked if antibody binding could alter the intracellular distribution of mobile target proteins within living cells. For this purpose, we chose signal transducer and activator of transcription 3 (STAT3) as the target intracellular functional protein and breast cancer cell line MDA-MB-231, which exhibits constitutively activated (phosphorylated) STAT3 ^36,37. STAT3 is a transcription factor that plays a crucial role in cell proliferation, survival, and other biological processes ³⁸. It is activated by tyrosine phosphorylation at position 705 (Y705) and serine phosphorylation within the transactivation domain (S727). STAT3 shuttles between the nucleus and cytoplasm, and activation of STAT3 is required for STAT3 dimerization, nuclear translocation, and DNA binding ^39,40. As unactivated STAT3 is predominantly present in the cytoplasm and translocated to the nucleus upon activation, we hypothesized that after cellular delivery of Hex-aSTAT3, cytosolic anti-STAT3 (aSTAT3) antibodies would bind to cytosolic STAT3 and block its nuclear translocation. We tested this by using two aSTAT3 antibodies, mouse aSTAT3 (pS727) and rabbit aSTAT3 (pY705), which bind to the different STAT3 phosphorylation sites. Anti-STAT3 (pS727) was delivered into live cells using Hex nanocarrier. The cells were then fixed, permeabilized, and the aSTAT3 (pY705) antibody, which bound to a site distinct from the S727 site, was used to indicate the cellular location of STAT3, and secondary antibodies were used to visualize both antibodies p(S727 green, Y705 red). Immunofluorescence revealed that cells treated with Hex-aSTAT3 exhibited green and red staining mainly colocalized in the cytoplasm with a high Mander’s coefficient of 0.932 ± 0.031. Neither red nor green signal was detected in the nucleus, revealing that the majority of STAT3 stayed in the cytoplasm and nuclear translocation was blocked (Figure 7A). In contrast, the PBS treated cells exhibited colocalization of green and red staining mainly in the nucleus with Mander’s coefficient of 0.657 ± 0.082, indicating translocated of activated STAT3 from the cytoplasm into the nucleus in the native, activated state (Figure 7B). Quantitatively, significantly higher levels of STAT3 were detected in the cytosol compared to the nucleus of cells incubated with Hex-aSTAT3, whereas for PBS treated cells most of the STAT3 proteins were in the nucleus (Figure 7C). Hence, we show that functional antibodies delivered to the cytosol by the Hex nanocarrier manipulated their target antigen, raising the possibility to study cellular physiology as well as intracellular antibody therapies.

Figure 7.

Manipulation of STAT localization by intracellular antibodies. MDA-MB-231 cells incubated live for 24 hrs with (A) Hex-aSTAT3 (pS727, mouse, 0.3 μM) and (B) PBS. Cells were washed, fixed, and permeabilized. Hex-aSTAT3 cells (A) were then incubated with aSTAT3 (pY705, rabbit) antibody. PBS cells (B) were incubated with mouse aSTAT3 (pS727) antibody and Rabbit aSTAT3 (pY705) antibody. Both were labeled with secondary antibodies, goat anti-mouse IgG (Alexa Fluor® 488, green) and goat anti-rabbit IgG (Alexa Fluor® 647, red). Hoechst 33342 (blue) was used as the nuclear counterstain. Scale bar represents 10 μm. (C) STAT3 distribution profile after Hex-aSTAT3 or PBS treatment. * p<0.05

Discussion

Effective and safe intracellular delivery of bioactive macromolecules into living cells is a challenge ^41,42. Full-size IgG cannot cross cell membranes, and even if some enters cells via endocytosis, it is usually not released into the cytosol, and is either recycled back to the cell surface or degraded in lysosomes ^5,13. Many attempts have been made to facilitate the intracellular delivery of antibodies, however, most of CPPs and nanoparticle-based delivery approaches result in endosomal entrapment of antibody and degraded in lysosomes ^43–46. Because interfering with the function of intracellular targets requires at least a 1:1 molar ratio between antibody and target, a small number of antibodies released into cytosol may be detectable but may not provide significant biological outcomes ^5,43,45. A successful case is the re-engineered antibody, based on a naturally internalizing antibody, that targets intracellular oncogenic Ras mutants and showed anti-tumor effects in vivo ⁴⁷. However, we note that other studies have relied on reporting only total cellular uptake and/or identifying endosomal-lysosomal entrapment without further understanding the intracellular bioavailability and the final fate of the delivered antibodies in live cells ^5,48,49. This systematic study of the internalization, intracellular trafficking, intracellular bioavailability and the fate of Hex-IgG in live cells connects each step and demonstrates biological impact of intracellular antibody delivery.

The time-dependent endosomal and lysosomal colocalization profiles revealed detail of Hex-IgG trafficking from cell surface to lysosomes. It has been suggested that an efficacious intracellular delivery system requires not only high dose, but also fast delivery into diseased cells to achieve high bioavailability per unit time ⁵⁰. The internalization of Hex-IgG occurred in just 5 min after treatment. Although endocytosis rates vary in different cell types ^51,52, this important cellular parameter can dramatically affect endocytic kinetics, which should be taken into consideration in the design of intracellular delivery. The colocalization profiles also suggest a time frame of endosomal escape about 120–180 min after internalization, because the antibody complexes could be digested after translocated in the lysosomes ⁵³. By comparing the time frame of endosome colocalization and YFP-Gal8 recruitment assay, we found that the number of endosomal escape events increases after 15 min, far away from the plateau phase of endosomal colocalization and increasing phase of lysosomal colocalization. This implies that many endosomal escape events occur in the early endosome. Early endosomal escape can offer significant advantages, especially for protein cargos such antibodies or therapeutic peptides or proteins, because it minimizes damage by low pH in late endosomes and protease degradation in lysosomes ⁵⁴. Due to the hexameric nature of the Hex nanocarrier, it contains 36 histidine residues. These are expected to be protonated by the acidic pH in endosomes and subsequently may cause osmotic swelling and disruption of the endosome (the ‘proton sponge’ effect), which could result in the cytoplasmic release of Hex-IgG and escape from lysosomal enzymes ⁵⁵. Additionally, amphipathic α-helical peptides have been shown to disrupt the membrane barrier by a variety of mechanisms⁵⁶ and have been used as endosomolytic components for intracellular delivery of recombinant transcription factors and CRISPR-Cas9/Cpf1 ribonucleoproteins ⁵⁷. The Hex coiled coil α-helices are also amphipathic⁵⁸ and may also contribute to the disruption of endosomal membrane. Although the underlying molecular mechanism by which Hex-IgG escapes from early endosomes into the cytosol is not clear, further work to investigate the factors that influence the escape efficiency could help apply the Hex system to delivery of other types of macromolecules.

It is important to note that our study demonstrated the process of Hex-IgG trafficking from cell surface to its final target, not only the IgG cargo. During endocytosis, antibodies remain bound to Hex with high affinity after cellular internalization ¹⁴, which likely prevents antibodies from binding to the neonatal Fc receptor (FcRn) located in the lumen of endosomes and being recycled back to the cell surface ^59,60. The FcRn binding site on IgG overlaps with the SPAB binding site ⁶¹. The Hex nanocarrier should fully occupy the two SPAB-binding sites in the Fc region of IgG since it is saturated with 3 antibodies ⁶², and therefore block FcRn binding. In our previous study, we measured the binding affinity of the Hex nanocarrier towards IgG to be very strong at neutral pH, with K_D below picomolar concentration ¹⁴. At pH 5.0, the binding was weaker with K_D of 1.39 ± 0.03 nM, but was still stronger than the interaction between FcRn and IgG (K_D: 0.2 μM at pH 6) ^63,64. As demonstrated by the HexAF647-aNPCAF488 co-localization assay, antibodies disassociate from the Hex nanocarrier but retain their bioactivity in the cytosol. It is reported that cytosolic IgG can engage the Fc receptor TRIM21, which targets antibody-bound proteins for degradation by the proteasome ⁶⁵. The TRIM21 system could lead to a decrease of measurable cytosolic IgG. While this is a challenge in general for intracellular IgG delivery, if it is active on Hex delivered IgG following disassociation, this endogenous protein degradation system may be harnessed to specifically degrade desired target proteins ^65,66. It is also likely that when IgG is bounding by the SPAB domains in Hex, engagement of the TRIM21 machinery is prevented, so further stabilizing the SPAB-IgG interaction may inhibit the TRIM21 degradation mechanism.

While this work was performed with breast cancer cell lines, cellular uptake efficacy of Hex-IgG complex also depends on cell type. Hex-IgG complexes can be internalized by other cell types, including human umbilical vein endothelial cell line (HUVEC) and macrophage J774 cell line (Figure S6). As is seen for most types of nanomedicines, uptake varied with cell type and the macrophages exhibited relatively higher levels of uptake. Uptake by HUVEC cells was lower than both SKBR-3 and J774 cells. Since there is no inherent cell or receptor specificity of Hex-IgG, some applications will require targeting of particular cell types. In this case, extracellular targeting antibodies may be used in place of one of the intracellular antibodies, achieved simply by mixing the 2 antibodies in the desired ratio prior to mixing with Hex.

Taking advantage of Hex-IgG intracellular delivery, we confirmed that the cytoplasmic concentration of escaped aSTAT3 antibodies was sufficient to bind and block the nuclear translocation of the majority of STAT3 proteins within living cells. From a therapeutic point of view, the role of STAT3 in breast cancer oncogenesis is still controversial ⁶⁷. Other studies with MDA-MB-231 cells showed that STAT3 inhibition with small molecule inhibitors reduced cell viability, but STAT3 knockdown in the same cells using short hairpin RNA induced tumor formation in a xenograft experiment ^68,69. Because anti-STAT3 delivery was capable of blocking nuclear translocation of STAT3, we anticipate that the Hex-IgG system can be easily adopted to inhibit translocation of other transcription factors or nuclear proteins. Future work will investigate the cellular outcomes and imaging capabilities of other antibodies with different intracellular targets to demonstrate therapeutic or research value for Hex-IgG.

Conclusions

In this study, we performed systematic intracellular trafficking studies that assessed the delivery performance of Hex-IgG, intracellular bioavailability, and bioactivity of Hex-IgG in the cytosol. We demonstrated Hex-IgG rapidly triggers endocytosis into cells via multiple routes, including clatherin-mediated endocytosis, and colocalizes with lysosomes. However, Hex-IgG was able to disrupt the endosomal membrane and escape to the cytosol. During or after the escape process, antibodies disassociate from Hex nanocarrier, and retained antigen-binding bioactivity for at least 24 h in the cytoplasm. As a demonstration of functional effect, Hex-IgG was used to block an important intracellular process, translocation of STAT3 transcription factor to the nucleus. These results can guide future work toward the development of Hex-IgG variations and applications that enable therapeutic and research tool functionality.