To determine if the biotinylated compounds are able to form complex with protein conjugated streptavidin, we first aimed to confirm the presence of the complex formed by means of polyacrylamide electrophoresis, followed by cytotoxicity study of compounds and formed complexes. The final experiments were designed to provide the information regarding efficacy of transport across the cell membrane of model enzyme.
Synthesis
Two DAPEG polymers consisting of six (Compound 1) or eight (Compound 2) monomers were synthesized and labeled with an N-terminal biotin. Synthesis were perfomed on solid support using combined synthesis system. Initially DAPEG polymers with 6 or 8 mer length were synthesized using microwave irradiation. After this stage the side chain protecting group were removed and appropriate oxa acid with guanidine group were attached to free amino group of polymer. The molecular weights of the compounds were nearly identical to the calculated weights and ranged from 1947.1 Da to 2719.3 Da (Table 1). It is generally known that biotin-containing compounds display exceptionally high affinity toward streptavidin or avidin, reaching a subpicomolar dissociation constant.
Table 1
Physiochemical characteristics of peptidomimetic compounds.
No | Sequence | Retention time* [min] | Molecular weight calculated/determined [Da]** | Ref. |
1 | Bt-O2Oc-[Dap(GO2)]6-O2Oc-NH2 | 7.46 | 2174.0/2173.1 | - |
2 | Bt-O2Oc-[Dap(GO2)]8-O2Oc-NH2 | 11.77 | 2720.0/2719.3 | - |
1a | O2Oc-[Dap(GO2)]6-O2Oc-NH2 | 3.40 | 1947.1/1947.1 | [12] |
2a | O2Oc-[Dap(GO2)]8-O2Oc-NH2 | 2.97 | 2493.7/2493.5 | [12] |
*UPLC analysis (Nexera X2 LC-30AD [Shimadzu, Kyoto, Japan] equipped with a Phenomenex column (150 × 2.1 mm) with grain size 1.7 µm (peptide XB-C18) equipped with a UV‒Vis detector and a fluorescence detector). A linear gradient from 2–80% B within 15 min was applied (A: 0.1% trifluoroacetic acid; B: 80% acetonitrile in A). ** HR MALDI analysis with 2,5-dihydroxybenzoic acid as a matrix.
Complex formation
Next, we aimed to verify the ability of the abovementioned compounds to form a complex with fluorescently labeled streptavidin (FITC-Strp). Electrophoretic separation of a mixture composed of Compound 1 or 1a and 2 or 2a with FITC-Strp at three different ratios (1:1, 1:2 or 1:4) resulted in the appearance of additional multiple bands (Fig. 3, boxes in lanes 3–5). In Fig. 4, the same extra bands were observed (boxes in lanes 2–7).
Cytotoxicity
Our results suggest that the complexes were successfully formed. Next we proceed to determine whether such complexes are able to translocate into the selected cell lines without significant cytotoxicity. Analysis of the cytotoxicity profiles of Compound 1 against a panel of breast cancer cell lines (MDA-MB-231, T47D and SKBR3 and the noncancerous control HB2) indicated that neither the formed complex nor its constituents significantly impacted cell viability at concentrations up to 20 µM. See Figs. 5 and 6.
Based on the above results, we incubated the abovementioned cell lines with FITC-Strp + Compound 1 or 2 in stoichiometric ratios of 1:2 or 1:4 at concentrations (referred to as protein amounts) equal to 10 µM (molar ratio 1:2) or 20 µM (molar ratio 1:4). As shown in Figs. 7–10, green fluorescence indicating the presence of FITC-Strp was observed in all cell lines.
No fluorescence was observed for any control system, including FITC alone and FITC-Strp alone. Localization of the signal differed among cell lines. For HB2, membrane and nuclear localization dominate, whereas for the cancerous cell lines, fluorescence is visible in the cytoplasmic/membrane compartment (Fig. 8). The in-cell fluorescence measured for all tested systems indicates that Bt-O2Oc-[Dap(GO2)]8-O2Oc-NH2 is a better mediator of labeled protein translocation than Bt-O2Oc-[Dap(GO2)]6-O2Oc-NH2 (Fig. 7–10). The obtained results correspond well with several reports describing the efficient implementation of several biotinylated CPPs complexed with fluorescently labeled streptavidin or avidin. Numerous CPPs have been used with different combinations of fluorescent labels [18, 19].
Next, we analyzed whether we were able to translocate model proteins with enzymatic activity using these systems, with streptavidin β-galactosidase (Strp-β-gal) conjugate as the selected model. β-Gal is a protein formed by four subunits, each with 1023 amino acid residues, and is frequently utilized as a reporter molecule. The enzyme hydrolyzes lactose and other β-D-galactosides into monosaccharides. Although very specific for the galactose moiety, it tolerates a variety of structures elsewhere in the substrate; thus, any number of organic dyes with a galactose moiety can be used as reporters. For the activity study, we utilized resorufin β-D-galactopyranoside, which produces free resorufin upon hydrolysis. For the cellular study, X-Gal was used.
To verify the efficacy of complex formation, we performed electrophoretic separation in native polyacrylamide gel with reverse polarization under acidic conditions. This analysis indicated the presence of a complex of Strp-β-gal and one of the biotinylated peptidomimetics Bt-O2Oc-[Dap(GO2)]6-O2Oc-NH2 or Bt-O2Oc-[Dap(GO2)]8-O2Oc-NH2 (red boxes) (Fig. 11 (lanes 3–5) and Fig. 12 (lanes 2–4), respectively).Above results are in parallel with those obtained for FIT-Strp conjugates. It indicates that variouof streptavidin did not altered the ability to biotin. With this proof of mutual and selective interactions of these two molecules, we proceeded to investigate the impact of the formed complexes on selected cell lines.
The analysis of cytotoxicity against the selected cell lines indicated that HB2 and MDA-MB-231 viability was not significantly influenced by the presence of the tested complexes at concentrations up to 20 µM. Above this concentration, a 20–30% reduction in cell number was observed. For T47D or SKBR3, the complexes significantly reduced the number of live cells at concentrations starting from 10 µM (Fig. 13). These data forced us to exclude T47D and SKBR3 cell lines from further experiments.
Delivery the active enzyme to the cells
Incubation of the formed Strp-β-gal:Bt-O2Oc-[Dap(GO2)]6/8-O2Oc-NH2 complexes with cell lines followed by β-galactosidase activity assays using a fluorescent β-gal substrate (resorufin β-D-galactopyranoside) indicated that after 24 hours of incubation, β-gal activity was observed in the majority of the systems, with different activity levels (see Fig. 14).
The greatest fluorescence increase was observed for systems including biotinylated polymers, with increases 2-6-fold greater than those for nonbiotinylated compounds. Culturing the cells for an additional 2 h, resulted in a significant reduction in the activity of β-gal in all systems other than those with complexes formed with biotinylated polymers. The level of recorded enzymatic activity was several times greater than that of the control systems.
These observations were confirmed by microscopy observations of intracellular β-gal substrate (X-gal) (Figs. 15–16); that is, blue color resulting from X-gal enzymatic breakdown was visible inside the tested cell lines at all time points, confirming the presence of active enzyme. No blue spots were observed for control – nontreated cell lines (Fig. 17). However, some background blue color was observed as the result of endogenous and transported β-gal, since X-gal is a universal substrate for any β-galactosidase activity.
To differentiate between endogenous β-gal activity, which can be found in every living cell at some background level, and the presence of the delivered enzyme, which is the active form of E. coli-expressed β-gal that is present as cargo in this experiment, we performed immunostaining experiments. As shown in Fig. 18, the control systems, cells treated with Strp-β-gal alone or without any additional component, display minimal fluorescence, which indicates the presence of a secondary antibody labeled with DyLight488. Significantly elevated fluorescence was observed in all systems in which peptidomimetic Strp-β-gal conjugates were present (Figs. 19 and 20). Antibody staining was observed mainly in the membrane compartment but also in vesicle-like structures (Figs. 19b and 19c). The octamer mimetic was a more effective transporter of the Strp-β-gal complex than the hexamer regardless of the ratio used. Minimal differences between the two cell lines in terms of the amount of delivered enzyme were observed, with slightly more in the MDA-MB-231 cell lane.
Analyzing the overall fluorescence observed for particular well, we observed a similar pattern for specific systems, as shown in Fig. 21. Again, the presence of exogenous β-gal after 24 hours of incubation with the complexes followed a similar pattern. For the HB2 cell lane, all systems, except the controls, displayed a greater amount of detected β-gal (Fig. 21). MDA-MB-231 cells were more sensitive to the composition and stoichiometry of the complexes used. Only one system, with Compound 2 in complex with Strp-β-gal in a 1:2 molar ratio, showed significantly greater fluorescence than any other system. After an additional 2 h, as visible in Figs. 19 (HB2 cells) and 20 (MDA-MB-231 cells), the β-gal was mainly visible in vesicle-like structures. The greatest fluorescence intensity was observed for the system incubated with the complex formed by compound Bt-O2Oc-[Dap(GO2)]8-O2Oc-NH2, followed by compound Bt-O2Oc-[Dap(GO2)]6-O2Oc-NH2.
In summary, the results presented above suggest the efficacy of a novel biotinylated cell-penetrating polymer able to mediate the cellular translocation of a model protein while retaining its activity. Such compounds could be utilized to transport functional exogenous or nonhost proteins into a variety of cells.