|Home | About | Journals | Submit | Contact Us | Français|
We previously reported a high aspect ratio peptide nanofiber that could be effectively delivered to tumors with minimal nonspecific uptake by other organs. The peptidic nature offers the design flexibility of smart formulation with unique responsiveness. Two new formulations that behave congruously as nanotransformers (NTFs) are reported herein. NTF1 and NTF2 could biomechanically remodel upon enzyme activation to generate a degradable and an aggregable effect, respectively, within the lysosomal compartment. These NTFs were further evaluated as carriers of mertansine (DM1), a microtubule inhibitor. DM1-loaded NTF1 could be degraded by cathepsin B (CathB) to release the same active metabolite, as previously described in the lysosomal degradation of antibody-DM1 conjugate. In contrast, CathB only partially digested DM1-loaded NTF2 and induced aggregate formation to become a storage reservoir with slow payload release property. The DM1-loaded NTF1 exhibited a comparable cytotoxicity to the free drug and was more effective than the NTF2 formulation in eradicating triple negative breast cancer. Our data suggested that biological transformers with distinct enzyme-induced structural changes and payload release profiles could be designed for the intracellular delivery of cytotoxic and imaging agents.
Increasing numbers of peptide-based nanofibers (PNFs) are being used for biomedical applications, including tissue engineering,1 promoting angiogenesis,2 vaccination,3 and cell labeling.4 Particularly in drug delivery, PNFs have been used to promote the delivery of hydrophobic drug molecules to treat diseases such as myocardial infarction,5 fracture healing,6 cervical spinal cord injury,7 and cancer.8–10 An advantage of PNFs is their flexibility to design by incorporating a functional domain that responds to the specific environment at the target site to release a therapeutic payload. For example, smart PNFs that can be activated by endogenous stimuli such as tumor-associated metalloproteinase-2 enzyme to release cytotoxic drugs, like cisplatin, have been designed.11 A taxol-peptide conjugate triggered by alkaline phosphatase to assemble into PNFs and hydrogels was synthesized.12 Peptide filamentous gels incorporated with 6-propionyl-2-dimethylaminonaphtha-lene via acid-liable hydrazone linkers have been used to investigate their controlled release property.13 There are other PNF systems, such as a diphenylalanine peptide, that can reversibly assemble into different fibril-like structures in response to external stimuli, such as temperature changes.14
Recently, we reported a unique PNF composed of multiple β-sheet peptides that stack parallel to each other in a tapelike structure.15–17 Each peptide consisted of alternate hydrophobic and hydrophilic residues (KLDLKLDLKLDL) and was conjugated to a 2k Dalton methoxypolyethylene (mPEG) chain to prevent aggregation after self-assembling into nanofibers. Unlike other PNFs that are cylindrical,18–20 our PNF platform displayed a unique single-layered structure with a high aspect ratio (0.5 × 5 × 100 nm3) that translates into effective uptake by tumors via the enhanced permeability and retention effect. Any undelivered vehicle could be eliminated faster through renal clearance.21 This PNF could also be surface-functionalized with specific ligands such as antibodies, polymers, and peptide moieties for active targeting.22
In the present studies, we further designed two new formulations of PNF that behave congruously as “nanotransformers” (NTFs). NTF1 and NTF2 could respond differently to activation by cathepsin B (CathB), a proteolytic enzyme commonly found to be upregulated in cancer cells.23–25 Inside the targeted cells, the lysosomal CathB degraded NTF1 to rapidly release the payload. On the other hand, CathB was only able to remove the hydrophilic mPEG component of NTF2 and resulted in the formation of aggregate (storage reservoir),26,27 delaying the payload release (Figure 1). Here, we compared the two NTFs as carriers of mertansine (DM1), a microtubule inhibitor for the treatment of breast cancer. The cytotoxicities of the drug-loaded nanofibers toward human epidermal growth factor receptor 2 (HER2)-positive breast cell line (BT474) and triple negative breast cancer (TNBC) cell lines (MDA-MB-231 and MDA-MB-468) were investigated.
All of the protected amino acids, resins, and solvents for peptide synthesis were purchased from Protein Technologies Inc. (Tucson, AZ). Trifluoroacetic acid (TFA), thioanisole, anisole, methyl-tert-butyl ether, N,N-diisopropylethylamine (DIPEA), and hydrazine were obtained from Sigma-Aldrich (St. Louis, MO). N2′-deacetyl-N2′-(3-mercapto-1-oxopropyl)-maytansine (DM1) was supplied by Carbosynth Ltd. (Compton, WB). CathB (purified from bovine spleen) and its inhibitor II were obtained from EMD Millipore (Billerica, MA). Uranyl formate was purchased from Electron Microscopy Sciences (Hatfield, PA). Sulfosuccinimidyl 4-[N-maleimidomethyl]cyclohexane-1-carboxylate (sulfo-SMCC), 4′6-dia-minidino-2-pheylindole (DAPI), and LysoTracker Red were obtained from Life Technologies Inc. (Norwalk, CT).
Peptide synthesis and methoxypolyethylene glycol (mPEG) conjugation were performed on rink-amide resin employing the traditional N-α-Fmoc methodology, as previously described.17 After peptide elongation and mPEG coupling, the 1-(4,4-dimethyl-2,6-dioxocyclohexylidene)3-methylbutyl (ivDde) side-chain protection group of C-terminal lysine was selectively removed with hydrazine (2% w/w) in dimethylformamide (DMF). For fluorophore conjugation, FITC (97.3 mg, 5 equiv) in DMSO (4 mL) or Cy5.5 NHS ester (50 mg, 1.4 equiv) in DMF (4 mL) was added to resin (0.05 mmoL, 1 equiv) in the presence of DIEPA (1 mL) overnight. The peptides were cleaved from the resin using a cocktail (5 mL) composed of TFA/thioanisole/water/anisole (90:5:3:2) and then precipitated in methyl-tert-butyl ether and purified by reverse-phase high-performance liquid chromatography (rp-HPLC).
After the removal of the ivDde side-chain protection group of the C-terminal lysine, as described above, sulfo-SMCC (25 mg, 1 equiv) in N-methylpyrrolidone (NMP; 4 mL) was then added to the resin (0.05 mmoL, 1 equiv) and allowed to react overnight. The peptides were separated from the resin using the cleaving cocktail (5 mL) and then precipitated and purified by rp-HPLC. DM1 (1 mg, 1 equiv) was added to the resulting MCC-peptides (10 mg, 3 equiv) in a cosolvent of NMP (100 μL) and phosphate-buffered saline (PBS; 10 mM, pH 7, 100 μL) and allowed to react for 2 days at room temperature. The final DM1-MCC-peptide constructs were purified by rp-HPLC. All the peptides and their intermediates were characterized according to their average molecular weights using MALDI-TOF analysis (Tufts Medical School, Core Facility, Boston, MA).
NTFs were assembled using the solvent evaporation method.28 Briefly, peptide constructs (0.5 mg) in dimethyl sulfoxide (DMSO; 10 μL) were added to a cosolvent of acetonitrile and water (1.5 mL). The assembled NTFs were purified by size exclusion chromatography (Sephadex G-25) to remove the free peptides and then homogenized into 100 nm lengths using a mini-extruder (Avanti Polar Lipids, Alabaster, AL) and polycarbonate membrane of the appropriate pore size (Whatman, Maidstone, U.K.). The concentration of NTFs was determined by ultraviolet (UV) absorbance according to the extinction coefficient of Cy5.5 (209 000 cm−1 M−1) or FITC (60 000 cm−1 M−1) in 5% (v/v) PBS in methanol.
NTFs (50 μM of peptide content) were incubated with CathB (0.3 U) in sodium acetate buffer (50 μM; pH 4). NTFs incubated in buffer were used as the negative controls. After 24 h, the samples were taken out for transmission electron microscopy (TEM), rp-HPLC, and MALDI-TOF analysis. For the inhibition study, CathB inhibitor II (100 μM) was added to the samples during incubation.
Intact or digested NTFs (10 μM of peptide content) in buffer (20 μL) were transferred onto Formvar/carbon-coated 400 mesh copper grids (Electron Microscopy Sciences, Hatfield, PA). Excess NTFs were blotted off using filter paper and the samples were stained with 0.5% (v/v) uranyl formate solution (20 μL). After drying at room temperature, the grids were examined under TEM (JEOL JEM-1400 LaB6 TEM operating at 120 Kv).
All supplies for cell cultures were purchased from Corning Cellgro Inc. (Tewksbury, MA). The cell lines (ATCC, Manassas, VA) were cultured in an RPMI 1640 medium supplemented with 10% (v/v) fetal bovine serum (FBS), penicillin (50 000 units/L), and streptomycin (50 mg/L), and then maintained in 5% of CO2 at 37°C.
Cells (3 × 103/well) were seeded on an 8-well chamber slide until they reached 60% confluency. To study the cellular distribution, NTFs (10 μM of peptide content) were added to the cells for 6 h at 37 °C. Prior to microscopic imaging, DAPI (9 μM) and LysoTracker Red (1 μM) were added to the culture medium to stain the nucleus and lysosomes, respectively. Cells were then washed with PBS. Images were acquired with an EVOS FL Auto Fluorescence Microscope (Life Technologies) using the appropriate excitation and emission filters. To investigate the release of fluorophores intracellularly, the cells were first incubated with NTFs for 6 h. The culture medium was then replaced with a fresh medium. Images were acquired at 0 and 24 h. For the control experiments, lysosomal catabolic functions were inhibited by adding ammonium chloride (100 mM) to the culture medium 2 h prior to incubation with NTFs.
The cytotoxicity of drug-loaded NTFs was assessed using the CellTiter-Glo Luminescent Cell Viability Assay (Promega, Madison, WI) as previously described.29 Briefly, cells were seeded on a 96-well plate (2 × 103/well for MDA-MB-231 and MDA-MB-468 and 2 × 104/well for BT474) and allowed to stabilize overnight. The culture medium was then replaced with fresh media containing different concentrations of NTFs, T-DM1 (Genentech, South San Francisco, CA), or free DM1 according to the DM1 content (100 μM to 10 μM). After 5 days of incubation, CellTiter-Glo reagent (100 μL) was added to each well. The generated oxyluciferin was quantified for emitted luminescence using a microplate reader (Tecan US Inc., Morrisville, NC). All experiments were performed in triplicate, and the results were presented as mean ± standard deviation. Statistical analyses were performed using Graph Pad Prism 6.0 software. All data were normalized to the values obtained with the untreated control cells, and half maximal inhibitory concentrations (IC50) were calculated by fitting the obtained data to a sigmoidal curve.
To monitor the active DM1 metabolites release by ultraviolet (UV) absorbance, DM1-MCC-NTFs (50 μM of peptide content) were incubated with CathB (0.3 U) in sodium acetate buffer (50 μM, pH 4) for different time points (1 to 24 h) at 37 °C. The samples were then freeze-dried and resuspended in acetonitrile to extract the DM1 metabolites. After centrifugation (5 min, 15 000 rpm), the absorbance, representing DM1, was measured at 288 nm. Finally, the concentration of the active DM1 metabolites released upon enzyme activation was calculated according to our determined extinction coefficient of DM1 in acetonitrile (4800 cm−1 M−1). The study was performed in triplicate.
Our NTFs are composed of multiple self-assembling peptides (KLDLKLDKL-DKL). Each peptide is conjugated to a 2 kDa mPEG at the N-terminal.30–32 In aqueous media, multiple peptide constructs can self-assemble into a single-layer structure, which can be further homogenized into 100 nm lengths using a mini-extruder.15 NTF is stabilized by strong hydrophobic and electrostatic interactions, allowing the incorporation of fluorophores, antibodies, and peptide substrates without affecting the final morphology.33–35
In the present studies, we designed and synthesized two enzyme-activatable NTFs, NTF1, and NTF2. These NTFs were assembled from peptides in an L-configuration. The multiple lysine residues of the self-assembling domain provided the intrinsic digestion sites for CathB.36,37 The difference between NTF1 and NTF2 is that the peptide construct of the former was supplemented with a functional domain, a hexalysine substrate to aid in improving the sensitivity to CathB digestion (Figure 2a). For the ease of characterizing the enzyme digestion profile, each peptide was conjugated with a fluorescein isothiocyanate (FITC) fluorophore (as the model drug). The peptide constructs could self-assemble into NTFs in PBS. The resulting FITC-labeled NTF1 (FITC-NTF1) and NTF2 (FITC-NTF2) were characterized by TEM analysis of the morphology.
Given the difference in the peptide components, we expected that NTF1 and NTF2 would respond differently to CathB activation. HPLC analysis showed that CathB digested FITC-NTF1 into multiple peptide fragments after 24 h (Figure 2b). To determine the enzyme cleavage sites of the NTFs, we collected the HPLC elution peaks for further MALDI-TOF analysis. On the basis of the molecular weight differences between the peptide fragments and the undigested peptide construct, we identified multiple enzyme cleavage sites located mainly at the lysine residues of the functional (hexalysine substrate) and self-assembling (KLDLKLDLKLDL) domains. Interestingly, one cleavage site was found in between a leucine and an aspartic acid (Table 1). On the other hand, we were only able to identify one fragment with the enzyme digestion of FITC-NTF2. This fragment corresponded to a loss of the hydrophilic N-terminal mPEG chain of the peptide construct (Table 1) and was more hydrophobic, as shown by the broader and red-shifted HPLC peak (Figure 2b). Using the commercially available CathB inhibitor, we confirmed that the activations of both FITC-NTF1 and FITC-NTF2 were enzyme specific (Figure S1a).
The unique activation profiles of NTF1 and NTF2 prompted us to further investigate their structural changes in response to CathB digestion. TEM studies revealed that 24 h after incubation with the enzyme, FITC-NTF1 disintegrated and became shorter in length. On the other hand, multiple FITC-NTF2 assembled into an interfibril network (Figure 2c). The two contrasting observations could be explained by the difference in the aggregation states. To compensate for the loss of hydrophilicity after the removal of mPEG by CathB (Figure 2b), multiple FITC-NTF2 assembled into an aggregate and thereby hindered other enzyme cleavage sites from further digestion by the enzyme. In the case of FITC-NTF1, the hydrophilic lysine residues from the hexalysine substrate could prevent the transformer from aggregation, even in the absence of the mPEG chain (Figure S1b), and thus CathB could digest FITC-NTF1 until completion.
The unique digestion profile of NTF1 and NTF2 suggested that they might exhibit distinctive behavior in cells where there is abundant amount of CathB, particularly inside the lysosomes. To investigate the cellular distribution, we planned to incorporate Cy5.5 fluorophore into NTF1 and NTF2 to perform imaging studies by fluorescence microscopy (see Investigation of Intracellular Delivery). However, incorporating substantial amounts of fluorophores into NTFs caused fluorescence quenching.28 To control the Cy5.5 loading, we used a comixture of Cy5.5-conjugated and FITC-conjugated peptide constructs (serving as the spacer) to assemble the NTFs (Figure 3a). To identify the formulations with a minimal quenching effect, we first prepared fluorescent NTF1 (F-NTF1) and NTF2 (F-NTF2) containing the same total amount but different ratios of Cy5.5 and FITC (Figure S2a,b). We then compared the optical properties of the NTFs. Using this approach, we determined the optimal fluorophore ratio (1:19) in F-NTF1 and F-NTF2 that exhibited fluorescence emission (at 710 nm) comparable to the free Cy5.5 (Figure 3b).
To investigate the lysosomal degradation of NTFs, we also prepared quenched formulations of NTF1 (Q-NTF1) and NTF2 (Q-NTF2) that were optically silent in their native states. The NTFs were assembled from Cy5.5-conjugated peptide constructs (Figure 3c and Figure S2a,c). As expected, CathB digested Q-NTF1 to release Cy5.5-peptide fragments and resulted in recovery of the fluorescence signal (Figure 3d). In contrast, the fluorescence of Q-NTF2 was only partially recovered upon enzyme activation since the nanotransformers could evolve into aggregates to slowly release the payload. This minimal increase of fluorescence was unlikely due to the induction of fluorescence quenching because of aggregation, as the same experiment performed on the fluorescent F-NTF2 did not show fluorescence quenching upon CathB digestion (Figure S2d). Both Q-NTFs were further used to investigate the lysosomal degradations of NTF1 and NTF2 using fluorescence microscopy (see Investigation of Intracellular Delivery).
Using F-NTF1 and F-NTF2 (with matching fluorescence intensity) would allow us to compare the intracellular delivery. Here, we first confirmed that both F-NTFs could internalize into various breast, ovarian, and prostate cancer lines (Figure S3). Using the human triple negative breast cancer (TNBC) cell line MDA-MB-468 as the cell model, we observed that both F-NTF1 and F-NTF2 predominately accumulated in lysosomes after cellular uptake (Figure 4a and Figure S4a). Cells treated with F-NTF1 also had a Cy5.5 fluorescence signal throughout the cytoplasm; presumably some of the NTFs were digested by lysosomal CathB to release Cy5.5-peptide fragments. To compare the intracellular degradation of NTF1 and NTF2, we performed cellular imaging studies using the Q-NTF formulations (Figure 3c). Cells preincubated with Q-NTF1 showed a fluorescence increase over time, indicating that the NTFs were degraded (Figure 4b). The NTF1 degradation was lysosome-specific, as shown by the addition of ammonium chloride (an inhibitor of lysosomal function)38 inhibiting the fluorescence changes (Figure S4b). On the other hand, we could only detect a fluorescence signal in cells 5 days after Q-NTF2 treatment (Figure 4b), suggesting that the NTF2 was degrading slowly. This result could be explained by the NTFs evolved into aggregates within the lysosomes and prevented them from further enzymatic degradation. TEM analysis of different organelle fractions isolated by differential centrifugation39,40 confirmed the aggregation of Q-NTF2 occurred only in the lysosomal compartment (Figure S4c) but not in any other cellular fractions.
The distinct cellular behaviors and drug release dynamics of NTF1 and NTF2 prompted us to further investigate their drug delivery efficiencies. Because NTF1 displayed an antibody-like lysosomal degradation and drug release mechanism, we incorporated mertansine (DM1), an inhibitor of microtubule polymerization41,42 used as the drug component of an antibody-drug conjugate (T-DM1), into the NTFs to study cytotoxicity. Using the same sulfosuccinimidyl 4-[N-maleimidomethyl]cyclohexane-1-carboxylate (sulfo-SMCC) linker technology to attach DM1 to the antibody, we prepared DM1-conjugated NTF1 and NTF2 peptide constructs by reacting the N-hydroxysuccinimide (NHS) ester of SMCC to the peptides in the solid phase (Figure 5a). This reaction was performed in the absence of an organic base to avoid the ring opening of the maleimide group,43,44 and the introduced SMCC linker was stable under strongly acidic peptide cleavage conditions. Subsequently, DM1 was conjugated to the peptides using the standard thiol-maleimide reaction (Figure 5a) leading to a synthesis yield close to 25%. The synthesis of DM1-MCC-peptide constructs and their intermediates were confirmed by MALDI-TOF analysis (Figure 5b). To assemble the DM1-loaded NTF1 (DM1-MCC-NTF1) and NTF2 (DM1-MCC-NTF2), we used a comixture of the DM1-MCC-peptide and Cy5.5-conjugated peptide constructs at a ratio of 9:1 and confirmed the formation of NTFs by TEM analysis (Figure 5c). On the basis of the absorbance, we determined that each NTF carried approximately 180 DM1 molecules, a significantly larger number than T-DM1 could afford (3.5 DM1 molecules/antibody).45
Prior to investigating the cytotoxicity of DM1-MCC-NTF1 and DM1-MCC-NTF2, we confirmed that CathB cleaved both NTFs. After enzyme digestion, we identified Lys-MCC-DM1, the same metabolite previously observed as a result of the intracellular degradation of T-DM146,47 in DM1-MCC-NTF1 (Figure 6 and Table 2). We also found a second metabolite, Leu-Lys-MCC-DM1, due to the enzymatic cleavage occurring between the leucine and the aspartic acid. Each drug metabolite had two isomers, as two HPLC peaks were characterized with the exact molecular mass (Figure 6b). These isomers were attributed to the R and S configuration at the carbon of the thioether bond formed during the DM1 conjugation.46 In the case of DM1-MCC-NTF2, no drug metabolite was identified during enzyme digestion due to the slow release property. However, the HPLC spectrum showed a notable aggregation pattern (P3) corresponding to a loss of PEG (Figure 6a), and this result was confirmed by MALDI-TOF analysis (Table 2). To confirm the difference in the drug release profile of NTF1 and NTF2, we performed an in vitro drug release experiment (Figure 6c). At 24 h after CathB activation, more than 85% of active DM1 metabolites were released due to the enzymatic degradation of NTF1 whereas the majority of the drug was still trapped in the enzyme-induced NTF2 aggregate, leading to a slow drug release profile.
Finally, we evaluated the cytotoxicity of the DM1-loaded NTFs using various human HER2-postive breast cancer (BT474) and triple-negative breast cancer (MDA-MB-231 and MDA-MB-468) cell lines as ex vivo models. A luminescent cell viability assay was employed to determine the cell viability through quantification of the adenosine triphosphate (ATP) expressed by metabolically active cells. This assay offered more accurate and reproducible results compared to the widely used MTS assay (data not shown).48 Because the cell killing of DM1 is slow, the assays were performed following the incubation of the tested samples for 5 days, as previously described.29 We first evaluated the toxicity of the naked NTFs using MDA-MB-231 and MDA-MB-468 cell lines (Figure 7a). Below a concentration of 1 μM, both NTF1 and NTF2 were nontoxic. However, NTF1 showed significant cell growth inhibition at a high concentration (10 μM); presumably the multiple positive charges of hexalysine substrate contributed to the toxicity.49 Subsequently, we compared the cytotoxicity of the DM1-loaded NTFs with the antibody-drug conjugate (T-DM1) and the free drug (DM1). T-DM1 was most effective in eradicating HER2-positive BT474 cells (IC50 = 2.3 nM), as it specifically targeted the HER2 receptors for enhanced cellular uptake (Figure 7b and Table 3). In contrast, DM1-MCC-NTF1 exhibited a competitive cytotoxic effect compared to free DM1 toward MDA-MB-231 (IC50 = 65.5 nM) and MDA-MB-468 cells (IC50 = 3.4 nM), which suggested that the NTFs could be used to deliver cytotoxic agents to TNBC cells, as they were able to completely release the drug loaded. As the naked NTF1 only displayed a minimal cell-killing activity at a low concentration, the enhanced cytotoxicity of DM1-MCC-NTF1 was unlikely due to the synergistic effect of the transformers and DM1 but rather a result of the increased cellular uptake to subsequently release the active DM1 metabolites via an antibody-like lysosomal degradation mechanism. Even though DM1-MCC-NTF2 exhibited a higher cytotoxic effect than T-DM1 in TNBC cell lines, the cell growth inhibition induced by the NTF was weaker than that of the free DM1 due to the slow and controlled release of the active drug metabolites inside the cancer cells (Figure 7b and Table 3). Overall, our data suggested that both NTFs could be potentially used for the intracellular delivery of cytotoxic and imaging agents, even though their particular drug release dynamics induce different cell-killing activities.
The design flexibility of self-assembling PNFs is a key parameter in generating different drug release profile. In the present study, we designed two new formulations (nanotransformers), namely NTF1 and NTF2, which are responsive to the lysosomal environment. We found that in the case of NTF1, an incorporation of an additional hexalysine between the PEG and the self-assembling domain (KLDLKLDLKLDL) resulted in a significant difference in structural changes upon CathB activation. NTF1 was digested to evolve into multiple degraded fragments, whereas NTF2 evolved into large aggregates to form an interfibril network. When used as a drug carrier, NTF1 showed enhanced cytotoxicity against TNBC and competitive cell-killing activity compared to T-DM1 and free DM1, respectively. Although NTF1 was less effective than T-DM1 for eradicating HER2-positive breast cancer due to its lack of targeting capability, it has a higher drug loading capacity. Previously, we immobilized multiple trastuzumab antibodies onto the PNF surface and showed enhanced cellular uptake and cytotoxicity due to the multivalent effect.50 In the future, the same approach can be applied to NTF1 to deliver more drug molecules to achieve an enhanced cytotoxic effect. On the other hand, NTF2 stayed inside the lysosomes after cellular uptake and, thus, displayed lower cytotoxicity compared to a free drug. This controlled release could have potential biomedical applications for cell labeling such as for long-term imaging of cell based therapy and tissue engineering. In the future, the drug delivery profile of NTF2 could be improved by incorporating a pH-sensitive linker, such as hydrazone group, so that the drug could be effectively released within the lysosome and tumor acidic environment. Overall, we think that the easy modification and the associated drug release profile of NTF1 would allow the incorporation of many other bioactive molecules, offering strong potential for delivering combination therapy.51
Research reported in this publication was supported by the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Number UL1TR000457. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors would also like to thank Dr. Sarah Horvath for her technical support.
Author ContributionsThe manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.
The authors declare no competing financial interest.
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.bio-mac.6b00227.
HPLC spectra of NTF1 and NTF2 coincubated with CathB and inhibitor; TEM image of nonpegylated NTF1; characterizations of NTF1 and NTF2 incorporated with different fluorophore ratios; fluorescence images to confirm the cellular uptakes of NTF1 and NTF2 by different human cancer cell lines; fluorescence images to show the intracellular distribution of NTF1 and NTF2; and TEM image of the lysosomal fraction of MDA-MB-468 cells after incubation with NTF2. (PDF)