**4. Discussion**

Currently, 5-FU is one of the first-line chemotherapeutic agents for the treatment of advanced CRC. It is a typical antimetabolite with a strong time-dependent mode of action [27]. Despite its therapeutic efficacy, 5-FU has some limitations, mainly tumor cell resistance and a short biological half-life. Consequently, multiple administrations of high doses, of which only a very low percentage reaches the tumor area, are required, and severe systemic (gastrointestinal, hematological, cardiac, and dermatological) toxicities occur. Indeed, 5-FU has been demonstrated to induce cardiotoxicity, followed by thrombosis, and alterations of the antioxidant defense capacities in myocardial tissues, with increases in the cardiac enzymes superoxide dismutase and glutathione peroxidase [28,29]. Therefore, the development of new delivery methods that selectively target the abnormal colonic mucosa could offer several benefits to patients affected by advanced CRC [30–32]. In this perspective, several nanovectors, such as lipid-based nanoformulations, nanogels, polymer-based micro/nanoparticles, carbon nanotubes and polysaccharides-based nanosystems have been designed to achieve an effective delivery of cytotoxic drugs and a selective targeting of CRC sites [33,34] Among them, targeted LPs have been demonstrated to exhibit an enhanced antitumor activity on CRC cells as compared to nontargeted counterparts. Folate, mannose and integrin receptors, overexpressed on CRC cells, are the main targets that have been explored for the development of CRC-targeted LPs [33]. Specifically for 5-FU, immune-LPs conjugated with proper folate or integrinβ6 antibodies have been proposed for CRC treatment [11,35,36]. Furthermore, O. Udofot et al. reported a study of the cytotoxic effect of 5-FU loaded pH-sensitive LPs on CRC cell lines; in this case, the LPs were conjugated with a selected anti-EGFR antibody [37]. In this study, PEG-functionalized 5-FU loaded LPs were covalently conjugated with a specific antibody able to recognize and bind the FZD10 surface cell receptor, to create innovative immuno-LPs that could potentially be useful for targeted therapy of CRC. In particular, FZD10 targeted and nontargeted LPs were prepared and characterized by complementary optical and morphological techniques. The average hydrodynamic diameter values obtained by DLS resulted lower than 200 nm for both formulated LPs. It is well known in literature that LPs of sizes between 100 and 300 mn are able to extravasate and localize in the tumor tissue, exploiting the EPR effect [38]. ζ-potential measurements proved that LPs exhibited a high negative surface charge, that further increased after LPs conjugation with FZD10 antibody, together with values of less than−30 mV for both formulated LPs, thus indicating their high colloidal stability in aqueous media and suggesting a good cell internalization capacity of the vesicles [39]. The encapsulation efficiency values achieved for 5-FU/LPs and anti-FZD10/5-FU/LPs are in agreement with the data already reported in the literature concerning other 5-FU encapsulating LPs [40].

The in vitro studies were performed on two selected CRC cell lines, namely CaCo-2 and CoLo-205. Preliminarily, quantitative and qualitative evaluation of the cellular FZD10 expression in CaCo-2 and CoLo-205 cells was made by immunoblotting and immunofluorescence, respectively, to validate the use of the FZD10 engineered LPs for targeted treatment of CRC (see Supplementary Materials). Furthermore, previous data had demonstrated the abundance of expression of FZD10 protein on CRC tissues, especially at late stages and in metastatic tissues [18]. Cell viability evaluation was conducted by testing the drug concentrations in the range of clinically relevant concentrations of 5-FU, previously reported in the literature and identified by performing clinical studies on pharmacokinetic modulating chemotherapy. E, Ojima et al. reported the range between 0.1 and 10 µM, while Beumer, J.H. et al. the range between 0.6 and 13 µM [41,42]. The two ranges slightly differ from each other, owing to different techniques used for the drug detection, modality of exposure to drug and substantial interindividual pharmacokinetic variability [42,43]. The results indicated that the cytotoxic activity of 5-FU is significantly enhanced when delivered by anti-FZD10/5-FU/LPs at the lowest tested drug concentrations (1 and 2 µM), as compared not only to the free drug but also to the nontargeted LPs. Therefore, at the lowest tested drug concentrations, the data obtained suggest that a more controlled drug release and higher cell accumulation of the targeted LPs occurred, likely due to their improved cell-internalization ability, exploiting not only passive transport but also antibody-receptor binding,

allowing the use of lower 5-FU concentrations, and achieving a higher expected performance in terms of reduced toxicity effects. Therefore, the 5-FU concentration of 2 µM was further considered for the in vitro studies.

The response induced on cell morphology after cell treatment with the two formulated LPs was investigated by means of the FE-SEM technique, that produces detailed images of a whole cell and ensures the visualization of several morphological details of a single cell (namely size, shape, nuclear/cytoplasmic ratio, number of pseudopodia and cell adherence), thus enabling any cell morphological changes induced by cell treatment with a specific pharmacological agent to be evaluated [27]. FE-SEM investigation visualized the several, severe time-dependent effects induced on the morphology and consequent viability of CaCo-2 and CoLo-205 cells, after treatment with 5-FU/LPs or anti-FZD10/5-FU/LPs. CoLo-205 cells appeared dead at 96 h treatment with both the formulated LPs, while, in the case of CaCo-2 cells, the anti-FZD10/5-FU/LPs exhibited a more efficacious cytotoxic activity than the 5-FU/LPs at 96 h, thus corroborating the data shown by MTS assay.

We also investigated the effect of the two formulated LPs on cell migration, a key factor in the malignant spread of cancer. For the metastatic CoLo-205 cells, the anti-FZD10/5-FU/LPs were found to induce a more significant inhibition of cell migration in the test area than the free 5-FU and 5-FU/LPs, the wound regions being only partially reduced at both 24 and, remarkably, at 48 h. The inhibition effect on migration of nonmetastatic CaCo-2 cells after treatment with 5-FU, free or embedded in the LPs, resulted more evident compared to CoLo-205 cells. Once again, the anti-FZD10/5-FU/LPs more consistently inhibited the cell migration compared to the free 5-FU and nontargeted LPs at 24 and 48 h treatment. Furthermore, the death of most of the cells was observed at 48 h.

Immunofluorescence analysis, performed on both fixed cell lines after incubation with free 5-FU, 5-FU/LPs or FZD10-anti/5-FU/LPs, was used to monitor the expression level of two proteins, namely vimentin and phospho-paxillin. The role of vimentin and phospho-paxillin as active players in cell focal adhesion and migration, as well as in pathological conditions, including cancer development and metastasis, is widely discussed in literature. The two proteins are found to be overexpressed in several cancer tissues and metastatic cancer cells, including also CRC, especially during adhesion and distant metastases formation [44–46]. High expression levels of both the vimentin and paxillin have been found to significantly contribute both to the malignancy and drug-resistance of CRC [47–49]. Consequently, the downregulation of the expression levels of the two proteins was found to be related to a reduction of metastatic development [50–53]. The investigation proved the time-dependent reduction in the fluorescence intensity of the vimentin and phospho-paxillin, confirming the effective activity of 5-FU on the inhibition of CRC cell proliferation [54]. The immunofluorescence analysis also indicated an enhanced toxic effect of 5-FU, when incorporated in the formulated LPs. In particular, the anti-FZD10/5-FU/LPs more consistently lowered the expression of the two proteins at 48 h in the case of CoLo-205, and already at 24 h treatment in CaCo-2 cells.

The overall results suggest that, at the lowest tested drug concentrations, the cytotoxic activity of targeted LPs was enhanced, as compared to free 5-Fu and nontargeted LPs. Although further studies are planned to investigate the specific mechanisms involved in the cell uptake, our preliminary findings are encouraging to validate the FZD10 protein as a novel, effective target for CRC. We expect that in future clinical applications, the use of anti-FZD10/5-FU/LPs could allow a reduction of the 5-FU doses administered, while maximizing the therapeutic efficacy. A new, promising scenario for targeted CRC therapy may be envisaged: administration of the anti-FZD10/5-FU/LPs here proposed, encapsulated in softgel capsules or suppositories, could provide an effective route for oral or rectal delivery, as employed for other compounds that need to reach the digestive canal, resisting gastric acids [55].

**Supplementary Materials:** The following are available online at http://www.mdpi.com/1999-4923/12/7/650/s1, Figure S1: UV-Vis absorption spectrum of FZD10-anti/5-FU/LPs before (A) and after (B) incubation with Alexa Fluor 555-conjugated secondary antibody, Figure S2: In vitro release of 5-FU from anti-FZD10/5-FU/LPs as a function of time, at 37 ◦C. Data are reported as mean value (±standard deviation) calculated on three replicates, Figure S3: (A) Representative Western blotting of FZD10 and GAPDH housekeeping protein and (A1) semi-quantitative estimation, by densitometry analysis of protein bands, of relative FZD10 expression levels in untreated HCEC-1CT, CaCo-2 and CoLo-205 cells, after loading the same total protein content (20 µg). For the semiquantitative analysis, FZD10 bands are evaluated after normalization with the corresponding housekeeping GAPDH protein band, for each sample. (\*) *p* < 0.001 versus negative control (HCEC-1CT cells). (B) Detection of FZD10 by immunofluorescence confocal microscopy in fixed HCEC-1CT, CaCo-2 and CoLo-205 cells. Blue channel: nuclei; green or red channel: labeled FZD10 and corresponding overlay. Scale bar 50 µm, Figure S4: Representative FE-SEM micrographs (EHT = 3.00 kV) of untreated CoLo-205 (A) and CaCo-2 (B) cells.

**Author Contributions:** Conceptualization, M.P.S.; methodology, M.P.S., A.C., E.F.; validation, E.F., N.D. (Nicoletta Depalo), and V.L.; formal analysis, N.D. (Nicoletta Depalo), A.C., E.F., M.P.S.; investigation, M.P.S., A.C., E.F., N.D. (Nicoletta Depalo); resources, G.G. (Giampietro Gasparini), G.G. (Gianluigi Giannelli) and N.D. (Nunzio Denora); data curation, M.P.S., N.D. (Nicoletta Depalo), and E.F.; writing—original draft preparation, M.P.S. and N.D. (Nicoletta Depalo); writing—review and editing, M.P.S., N.D.(Nicoletta Depalo), G.G. (Giampietro Gasparini) and G.G. (Gianluigi Giannelli); visualization, all authors; supervision, G.G. (Giampietro Gasparini) and G.G. (Gianluigi Giannelli); project administration, G.G. (Giampietro Gasparini) and G.G. (Gianluigi Giannelli); funding acquisition, G.G. (Giampietro Gasparini) and G.G. (Gianluigi Giannelli). All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** We are thankful to Babette Pragnell (Pragnell M.V., B.A) for English revision.

**Conflicts of Interest:** The authors declare no conflict of interest.
