Targeted SPION siderophore conjugate loaded with doxorubicin as a theranostic agent for imaging and treatment of colon carcinoma

Recently, the siderophores have opened new horizons in nanomedicine. The current study aimed to design a theranostic platform based on superparamagnetic iron oxide nanoparticles-pyoverdine (SPION/PVD) conjugates bound to MUC1 aptamer (MUC1Apt) and loaded with doxorubicin (DOX) as an anti-cancer agent. The SPION/PVD complex was covalently conjugated to MUC1Apt and loaded with DOX to prepare a targeted drug delivery system (SPION/PVD/MUC1Apt/DOX). The investigation of cellular cytotoxicity and uptake of formulations by MTT and flow cytometry in both MUC1 positive (C26) and MUC1 negative (CHO) cell lines revealed that MUC1Apt could improve both cellular uptake and toxicity in the C26 cell line. The evaluation of tumor-targeting activity by in vivo bio-distribution showed that the targeted formulation could enhance tumor inhibitory growth effect and survival rate in C26 tumor-bearing mice. Furthermore, the potential of synthesized SPION/PVD/MUC1Apt/DOX complex as diagnostic agents was investigated by magnetic resonance imaging (MRI) which improved the contrast of tumor site in MRI. Our findings confirm that aptamer-targeted PVD chelated the SPION as a diagnostic agent and loaded with DOX as a chemotherapeutic drug, would be beneficial as a novel theranostic platform.

FTIR analysis of PVD and SPION/PVD conjugates. FTIR analysis showed the presence of the main functional groups in PVD (see Supplementary Fig. S2a online). The broad absorption at ~ 3225.4 cm −1 has contributions from O-H and N-H stretching while the peak at 2404.1 cm −1 corresponds to NH 3 overtone. The peaks at 1664 and 1268 cm −1 are attributed to C=O and C-O stretching modes in C=O moiety of amide group (O=C-NH 2 ) and phenolic group, respectively. The peak at 1082.6 cm −1 has possibly some contributions from C-O-C bending. The absorption at 868.74 cm −1 is attributed to the CH plane binding in the aromatic ring. In the spectrum of SPION/PVD, the C=O stretch band at 1664 cm −1 disappeared whereas two new bands at 1402 and 1630 cm −1 observed, confirming the prominent absorptions related to the amide I and II modes of the PVD (see Supplementary Fig. S2b online). Also, the CH 2 bands at 2924 (asymmetric stretching) and 2851 cm −1 (symmetric stretching) have shifted to a lower frequency wavelength suggesting a rearrangement of the hydrocarbon molecules surrounding the NPs. Peak corresponding to the O-H group on the catechol-like group of PVD-chromophore changed to 3414 cm −1 indicating a bidentate covalent bond with Fe. The bands at 1402 and 3124 cm −1 show the major contributions of bending vibrations of amides and N-H stretching on the peptide chain bonded to SPION. The C-O stretching frequency for PVD at 1088 cm −1 is shifted to 1025 cm −1 , which further confirmed that the 2, 3-diamino 6, 7-dihydroxyquinoline group of the PVD binds covalently to the Fe 3   www.nature.com/scientificreports/ lower than that of SPION/PVD ( Table 1). Results of gel retardation assay indicated that SPION/PVD/MUC1 Apt complex (Lane 2) was unable to penetrate through the pores in the gel, demonstrating the high molecular weight of SPION/PVD/MUC1 Apt complex compared to free MUC1 Apt (Lane 3) which was significantly shifted (see Supplementary Fig. S3 online). Also, no band was observed for SPION/PVD (lane 1). The covalent conjugation of MUC1 aptamer to the SPION/PVD complex was confirmed by treating the SPION/PVD/MUC1 Apt with DTT as the reducing agent (Fig. S3b). A band corresponding to free aptamer appeared indicating the cleavage of the aptamer from the SPION/PVD/MUC1 Apt conjugate.
In vitro drug loading and release. To    www.nature.com/scientificreports/ the fluorescent intensity of DOX was observed (Fig. 4a). The entrapment efficiency (EE) was ~ 92% with loading capacity (LC) of 6.18%. The size analysis of SPION/PVD/MUC1 Apt by DLS revealed that an increase from 119.8 ± 4.6 to 127.6 ± 5.8 nm verifying the loading of DOX (Table 1). Moreover, the change of zeta potential of SPION/PVD/MUC1 Apt from − 10.2 to − 4.33 mV displayed the loading of DOX as a cationic molecule ( Table 1). The SEM images also confirmed that the SPION/PVD/MUC1 Apt /DOX complex has < 200 nm diameter with spherical structure which can be sutible in intravenous delivery (see Supplementary Fig. S1 online). Figure 4b shows the release rate of DOX in citrate buffer (pH 5.5) and PBS (pH 7.4). When the pH value reduced from 7.4 to 5.5, the content of DOX released from SPION/PVD/MUC1 Apt /DOX increased significantly (*p ≤ 0.05). DOX represented a constant release at pH 7.4 and only 23% of DOX could be released after 96 h,  Cellular uptake analysis. The cellular uptake ability of free DOX, SPION/PVD/MUC1 Apt /DOX and SPION/PVD/Scr Apt /DOX in CHO and C26 cell lines was evaluated by means of flow cytometry. As revealed in Fig. 5a,b, cellular uptake of MUC1 Apt targeted-complex was higher than that of scramble aptamer-modified  Fig. 7c,d). ROI analysis also exhibited that the mean intensity of the targeted formulation in the tumor was significantly higher than that of non-targeted formulations (*p ≤ 0.05). The liver accumulation after 6 h postinjection was high for all groups while SPION/PVD/MUC1 Apt /DOX injected group proved rapid liver clearance 24 h post-injection. Beside, due to the attachment of targeted formulation to specific receptor in tumor, their leakage from tumor to blood was reduced, leading to low accumulation and low intensity in liver after 24 h in comparison with SPION/PVD/Scr Apt /DOX-treated group. Moreover, as it is shown in Fig. 7b,d, in heart and lung tissues of mice treated with SPION/PVD/MUC1 Apt /DOX at 24 h post-administration, lower intensity of fluorescence was detected in comparison with free DOX and non-targeted formulation.
In vivo therapeutic potency evaluation. The therapeutic capability of the prepared nanoformulations was determined by assessing survival time and tumor growth rate of C26 tumor-bearing BALB/c mice after single-dose intravenous injection of free DOX, SPION/PVD/MUC1 Apt /DOX and SPION/PVD/Scr Apt /DOX (equivalent DOX concentration 0.2 mg/kg). The obtained date revealed that the inhibitory effect of SPION/PVD/ MUC1 Apt /DOX, as a targeted-formulation, was remarkably higher than those of PBS, free DOX, and SPION/ PVD/Scr Apt /DOX (Fig. 8a). No significant difference was also obtained in tumor size between groups receiving free DOX and SPION/PVD/Scr Apt /DOX (*p > 0.05). The survival rate results confirmed high survival rate for SPION/PVD/MUC1 Apt /DOX receiving group compared to other treated groups (Fig. 8b). In addition, the body weight loss in free DOX-receiving mice was greater than SPION/PVD/MUC1 Apt /DOX-received mice (Fig. 8c).

Discussion
In summary, we provided a MUC1 Apt -based targeted system for the delivery of DOX-loaded SPION/PVD (SPION/PVD/MUC1 Apt /DOX), capable of providing MRI images and preventing cancer cell growth in vitro and in vivo. As far as we know, this is the first study to provide a SPION/PVD conjugate as a MRI diagnostic agent which is targeted with an aptamer and loaded with an anti-cancer drug, providing a theranostic platform. Due to the aggregation and the instability of bare Fe 3 O 4 NPs at physiological pH, numerous surface coatings have been used to modify their surface properties and improve the stability 25 . Our results also indicated that while bare SPION is not stable, but PVD/SPIONs/MUC1 Apt /DOX is stable enough to be tested in vivo (see Supplementary Table S1 online). Besides, the bare SPIONs can be degraded via the body metabolism causing an overload of iron ions in tissues. Inspired by reports that used Fe 3 O 4 NPs to selectively adsorb siderophores as a microbial chelator with high affinity 26,27 , we demonstrated the capability of magnetic NPs chelated with purified bacterial PVD. Previous studies revealed that SPION can be conjugated with any drugs and/or natural compounds containing sulfhydryl (-SH), amine (-NH 2 ), phosphate (PO 4 3-), hydroxyl (-OH), and carboxyl (-COOH) groups or their combinations 28,29 . Therefore, hydroxamate and catecholate functional groups of PVD can contribute to the SPION chelation with PVD. It is well-known that the PVD can efficiently bind to metal ions such as Fe, Cu, Zn via the cooperation of dihydroxyquinoline and peptide chain 30 . As PVD chelates iron ion with a 1:1 (PVD:Fe 3+ ) stoichiometry 1 , so PVD bound to SPION with a stoichiometry of 1:1. The formation of the Fe(III)-PVD gives rise to a shift in the λ max of the free PVD absorption spectrum and further quench the fluorescence of PVD through electron-transfer pathways (Fig. 2a,b) 31 . In our research, the SPION could be also captured by the hydroxyl and carbonyl groups on PVD and the electron transfer occurred between PVD and the www.nature.com/scientificreports/ SPION, which further resulted in the λ max shifting and the fluorescence quenching of PVD (Fig. 3). An insight into the possibility of SPION/PVD formation was also confirmed by the FTIR (Fig. S2) which was in agreement with previous reports 27,30,32 . However, SPION/PVD conjugates were previously used as analytical platforms. The binding ability of SPION to PVD indicates that it has more potential for efficient isolation of siderophores from microorganisms media based on magnetic properties of the SPION. Conjugation of SPION with tumor targeting moieties such as aptamers and loaded with anti-cancer drug represent a promising platform for the efficient capture of cancer biomarker and specific delivery of MRI agents and drugs. Herein, the DOX release pattern from SPION/PVD/MUC1 Apt /DOX formulation was compared at pH 5.5 and 7.4 to mimicking the tumor cell endosome and physiological conditions, respectively, and the accelerated release rate of DOX was observed at pH 5.5 in comparison with pH 7.4 (Fig. 4b). The accelerated release in the acidic state might be owing to the superior dissolution of the protonated DOX at low pH which also enhanced its water solubility 33 . It is notable to mention that the faster release of anti-cancer drug from our complexes under slightly acidic conditions, similar to pH of the tumor cells endosome, facilitates the therapeutic performance of the formulation and thus improving cellular uptake and cytotoxicity at the site of action 34 . Thus the synthesized www.nature.com/scientificreports/ SPION/PVD/MUC1 Apt /DOX complex provides an ideal drug carrier with a pH-responsive trait which can control the release of intercalated DOX. The evaluation of cytotoxicity of formulations on C26 and CHO cell lines (MTT assay) indicated the validity of MUC1 Apt on the uptake and cell killing effect of the targeted formulation. This is in agreement with former studies in which delivery of DOX with aptamers like MUC1 Apt , increased the cytotoxicity and cellular internalization of DOX in vitro 33,35 . The mechanism of aptamer-based drug uptake might be receptor-mediated endocytosis (RME) leading to effective internalization of formulation into the cells with expressed receptor 36,37 . The high dose infusion, adverse effects, low bioavailability, low therapeutic index, and non-specific targeting are some of the major drawbacks of chemotherapy 38,39 . In this line, the nonselective distribution of DOX in the normal organs leads to severe systemic toxicity especially cardiotoxicity, which limits localization of free DOX in target tumor site 40,41 . In previous attempts to overcome these limitations, the nano-sized aptamer-targeted systems loaded with DOX could improve the biodistribution and antitumor efficacy compared with free DOX 42,43 . It has been also demonstrated that the GC base pairs of the aptamer afford proper places for DOX loading 21,23 . The structure of MUC1 Apt was demonstrated to intercalate 2-3 molecules of DOX into the GC sequence of MUC1 Apt . Using this approach, we intercalated DOX into a MUC1 aptamer conjugated to SPION/PVD (Fig. 4a). In this regard, the intercalation phenomenon of DOX into aptamer structure did not affected the delivery of other components of the formulation (SPION and PVD). In vivo experiments indicated that targeted DOX-loaded complex could simultaneously deliver DOX and SPION to tumor site and increased the accumulation of DOX and SPION in tumor tissues and obviously decreased the their accumulation in normal tissues particularly heart and lungs (Fig. 7). The targeted formulation could also be accumulated in liver 6 h post-injection, probably due to the colloidal nature of non-sized formulation. However, after 24 h, the SPION/PVD/MUC1 Apt /DOX formulation was less accumulated in liver compared to other treatments because of its targeting characteristics, leading to its high accumulation in tumor tissue. Meanwhile, the greater loss of body weight in free DOX-administrated mice in comparison with mice receiving SPION/PVD/MUC1 Apt /DOX (Fig. 8c) could be related to the easily circulation of DOX within the body and DOX-induced systemic toxicity as well as modified pharmacokinetic pattern of the prepared targeted formulation in comparison with free DOX. It should be noted that the 5 mg/kg of DOX was found as the best-acceptable dose which injected intravenously in mice 33 . However, here we used 0.2 mg/kg of DOX. The aforementioned proper biodistribution (Fig. 7), the tumor inhibitory effects (Fig. 8a), and survival rate of SPION/PVD/MUC1 Apt /DOX (Fig. 8b) suggested that this low dose of DOX can be the appropriate therapeutic dosage with slight toxicity when DOX directly intercalated within the aptamer sequence. Regarding the similarity between the tumor growth inhibition behavior of free DOX and non-targeted formulation, it could be ascribed to the partial accumulation of SPION/PVD/Scr Apt /DOX in tumor due to the EPR effect and thus showing tumor inhibitory effect similar to that of free DOX. In in vitro condition, free DOX enters the cell freely through passive diffusion, thereby inducing high toxicity in both target (C26) and non-target cells (CHO). On the contrary, in in vivo study, free DOX before reaching the tumor and entering the cell through diffusion, would be eliminated through renal clearance and liver metabolism thereby showing less anti-tumor activity in comparison with targeted formulation.
The biodistribution outcomes can be also ascribed to the decreasing of corona shielding around the Fe 3 O 4 after PVD and MUC1 Apt conjugation on the surface of the targeted complex. It is well-accepted that when Fe 3 O 4 NPs enter into relevant biological systems, their interaction with the plasma proteins results in the protein corona formation 25,44 . Protein coronas on nanomaterials surfaces can critically influence their target in recognition and internalization into the target cell. Another factor influencing the target cell uptake and elimination of nanomaterial-based complexes is emanated from the size of synthesized complexes 45 . The DOX-loaded SPIONbased nanocarriers with size smaller than 200 nm escape from capturing by the reticuloendothelial system (RES), leading to the prolonged circulation time of formulations in blood-stream 46 and better accumulation in tumors via the enhanced permeability and retention (EPR) effect due to increased tumor angiogenesis 47 . The obtained data verified that the prepared complex through the chelation of Fe 3 O 4 NPs with PVD, despite its high PDI and heterogeneity, with size less than 200 nm, can safely be considered as an intravenous delivery system.
SPIONs are one of the US food and drug administration (FDA)-approved NPs that are successfully used as contrast agents in MRI. They can be easily functionalized for drug delivery, demonstrating great potential for theranostic applications 48,49 . The coating of SPION with organic materials can improve the colloidal stability which can facilitate the implementation of SPION as contrast agents for MRI. The MRI experiment revealed a noteworthy higher tumor accumulation of targeted SPION/PVD which can still be detected even 24 h postinjection (Fig. 6). These results are in agreement with enhanced cellular toxicity, the beneficial effects on inhibiting tumor, and higher survival rate of SPION/PVD/MUC1 Apt /DOX versus SPION/PVD/Scr Apt /DOX because of its receptor-mediated endocytosis. All of these findings might be related to specific binding of MUC1 Apt to its overexpressed receptors (specific ligand-receptor interaction) in cancer cells, helping possibly a delay in extravasations from tumor tissues leading to the remain of the targeted formulation in the tumor site.
The efficiency of a contrast agent arise from its relaxivity, which is the proportionality constant of the measured rate of relaxation over a range of contrast agent concentrations 50 . The relaxivity relates to the magnetic properties of the contrast agent (including particle size, composition, and crystallinity), the molecular structure and kinetic of the prepared complex, as well as experimental conditions such as temperature, field strength, and the measurement media 50 . For example, Resovist® as an organ-specific MRI contrast agent, consists of carboxydextran-coated USPIO with predominantly results in a negative enhancement of normal liver parenchyma on both T2 and T1 weighted images 51 . Considering our previous studies 52,53 , in this study, a negative contrast (dark signal) was also obtained in both T1 and T2 weighted images of liver and tumor tissues after injection of formulation containing SPION. In general, T1 images of fat tissue (i.e. tumor tissue) exhibited a greater difference between tumor and SPIONs on T1 weighted images. www.nature.com/scientificreports/ In conclusion, we provided a PVD-based theranostic nanocarrier for the targeted co-delivery of DOX as an anticancer drug and SPION as an imaging agent. MRI and flow cytometry assays confirmed the accumulation of designed nanoformulation in the tumor cells. Furthermore, the accumulation of MUC1 Apt -targeted complex (SPION/PVD/MUC1 Apt /DOX) was detected in tumor site even 24 h post-injection using ex vivo fluorescence imaging. The in vivo analysis revealed that the significant tumor inhibition and survival rate in the mice receiving a single-dose targeted-NPs in comparison with the non-targeted NPs. It could be concluded that SPION/ PVD/MUC1 Apt /DOX formulation provide an efficient dual-modality targeted NPs which could be employed as theranostic platform for the clinical cancer diagnosis and therapy. The dried PVD was resuspended in HEPES buffer (pH 7.0) and was passed through copper-chelated SP-Sepharose fast flow column (2.5 × 1.6 cm, 5 mL; GE Healthcare) to desalt and preliminary purify. After washing with HEPES buffer, the column was eluted with 20 mM acetate buffer (pH 4.0) and PVD-Cu containing fractions with the highest A 400 absorbance were collected and lyophilized. Subsequently, the lyophilized residue was dissolved in 10 mM EDTA and was further purified by Sephadex G-15 column (1.5 × 80 cm; GE Healthcare), followed by ultrapure water elution and fractions with the highest fluorescence intensity (excitation/emission at 400/460 nm) were collected and lyophilized to obtain pure PVD 31,55 . The concentration of purified PVD was estimated based on A = εBC formula 56 and characterized by using Arnow's and Czsaky's methods for catecholate and hydroxamate groups assay, respectively 57 . The purified PVD was also further analyzed by determination of λ max , fluorescence excitation/emission (400/460 nm), iron-chelation properties (fluorescence quenching of PVD at 460 nm by FeCl 3 solution), FTIR in the range of 4000-400 cm −1 , thin layer chromatography (TLC) and LC/ MS/MS spectroscopy (AB SCIEX, Darmstadt, Germany) 57-59 . Synthesis of SPIONs and conjugation to PVD. The synthesis of SPIONs was performed using a coprecipitation method as previously described 60 . In deionized/deoxygenated water with nitrogen (N 2 ), a mixture of Fe (III) chloride (0.1 M) and Fe (II) chloride (0.1 M) with the molar ratio 1:2 was prepared for 15 min. 3 mL NH 4 OH solution (5 M) was added slowly to this solution while stirred under N 2 atmosphere at 25 °C until a deep black color appeared (~ 15-20 min). Another half-hour, the suspension was constantly stirred at 85 °C. Lastly, the ammonia was vaporized and an external magnetic field was applied to separate the black color precipitate, and freeze-dried after washing with double-distilled water (ddH 2 O). In the next step, a solution containing 2 mg/ mL of both SPIONs and PVD was sonicated in an ultrasonic bath for 20 min at room temperature. Then, the mixture was centrifuged at 14,000×g for 10 min at 4 °C to further remove very large aggregates.

Materials
The particle size distribution, polydispersity index (PDI) and zeta potential of SPIONs and SPION/PVD conjugates were assayed by dynamic light scattering (DLS) ZetaSizer (NANO-ZS, Malvern, UK). Field emission scanning electron microscope (FE-SEM, TESCAN MIRA3, Brno, Czech Republic) was applied to evaluate the morphology, particle size and EDX pattern of complexes. The magnetic properties of SPION and SPION/PVD conjugates were assessed by the vibrating sample magnetometer (VSM) to evaluate the Magnetic field dependence at room temperature under circulate magnetic field in the range of − 20,000 up to 20,000 Oe. The thermal stability and the content of conjugated PVD to SPION, was investigated by the thermogravimetric analysis (TGA) method under nitrogen in the temperature range 50-600 °C with a heating rate of 15 °C/min. The FT-IR spectra of PVD and SPION-PVD were also performed by Nicolet Avatar 360 FTIR spectrometer (Thermo Nicolet Corp, USA) in the range of 400-4000 cm −1 . The SPION-PVD conjugates were also further analyzed by fluorescence quenching at 460 nm. www.nature.com/scientificreports/ Conjugation of MUC1 Apt to SPION/PVD. The NH 2 -MUC1 Apt molecules were covalently linked to the exposed carboxyl group (-COOH) of PVD in SPION/PVD complex using the EDC/NHS reaction 61 . Briefly, a suspension of SPION/PVD (2 mg/mL) in 980 µL phosphate-buffered saline (10 mM, pH 7.4) was adjacent with an excess of EDC (5 mg/mL) and NHS (3 mg/mL) for 1 h at room temperature to activate the terminal carboxyl group on SPION/PVD. After incubation, 20 µL of NH 2 -MUC1 Apt (10 μM) was supplemented to the reaction medium and stirred for 12 h at 25 °C. Then, the mixture was concentrated to 100 µL by centrifugation (12,000×g, 10 min, 4 °C). The formed conjugate was magnetically collected from the system and washed twice with DNase/ RNase-free water to remove the reactants and finally was re-suspended in 100 µL PBS. To confirm the formation of conjugates, 2.5% agarose gel electrophoresis in TBE buffer for 25 min at 90 mV was applied. Besides, in order to investigate the attachment of MUC1 Apt to SPION/PVD, the formed conjugate was treated with Dithiothreitol (DTT) as a reducing solution and then run on the gel. The SPION/PVD/MUC1 Apt size, ζ potential, PDI, and morphology were also investigated using Zetasizer and FE-SE. The size, ζ potential, PDI and morphology of SPION/PVD/MUC1 Apt /DOX (final formulation) were also investigated using Zetasizer and FE-SE.

DOX loading in SPION
In vitro DOX release. The in vitro release pattern of DOX was accomplished in either PBS at pH 7.4 and citrate buffer at pH 5.5. SPION/PVD/MUC1 Apt /DOX complex was resuspended in 400 µL PBS and citrate buffer and incubated at 37 °C with shaking at a speed of 80 rpm. For estimation of the released DOX, the complex was magnetically collected and the supernatant was taken at 0, 3, 6, 12, 24, 36, 48, 72 and 96 h to measure the released DOX spectrofluorimetrically (excitation/emission at 480/595 nm). At predetermined time intervals, the drawn media was replaced by the same amount of fresh buffer. The following equation was applied to estimate the DOX accumulative release percentage (AR%) 62 .
Cell viability assay of the synthesized formulations. According to the results of a dose-escalating experiment, IC 50 of DOX was obtained 0.15 and 0.5 μM for C26 and CHO cells, respectively. The cytotoxicity of free DOX, SPION/PVD/MUC1 Apt and SPION/PVD/MUC1 Apt /DOX with equivalent concentration of 0.15 μM DOX for C26 and 0.5 μM DOX for CHO cells, was investigated using MTT assay. A DOX-loaded scrambled aptamer (Scr Apt ) which conjugated to SPION/PVD was also used as a non-targeted complex (SPION/PVD/ Scr Apt /DOX). 5 × 10 3 of C26 and CHO cells were seeded onto 96-well plates for 24 h and then treated with the above-mentioned formulations for 4 h. Then, the culture media were exchanged with fresh medium and incubated at 37 °C and CO 2 5% for 48 h. Then, a further incubation (~ 4 h) was carried out after adding 20 μL of MTT solution (5 mg/mL in PBS) to each well. Then, MTT was replaced with 100 μL of DMSO to dissolve formazan crystals at room temperature. The optical density was detected at 570 and 630 nm using a microplate reader (BioTeK, USA) 33 . Furthermore, a competitive assay was also done to verify the selective targeting property of MUC1 by the addition of an excessive amount of free MUC1 Apt 30 min before the adding of SPION/ PVD/MUC1 Apt /DOX to the wells 33 . Viability (%) was compared with untreated cells according to data of three individual assessments. Cellular uptake level of synthesized formulations. C26 and CHO cells, as respective MUC1 positive and MUC1 negative control, were seeded in 12-well plates at density of 1 × 10 5 cells/well and cultured in RPMI 1640 containing 10% FBS and 1% penicillin/streptomycin and kept 37 °C in a humidified incubator with 5% CO 2 . After 24 h, the cells were treated with free DOX, SPION/PVD/MUC1 Apt /DOX and SPION/PVD/Scr Apt / DOX (at DOX equivalent concentration 1 μM for C26 and CHO) for 2.5 h 33,35,63 . To survey the targeting efficiency of the SPION/PVD/MUC1 Apt /DOX, a competitive assay was also designed, in which C26 and CHO cells were incubated with an excess amount of free MUC1 Apt 30 min before the addition of SPION/PVD/MUC1 Apt / DOX. To perform fluorescence microscopy imaging, the media were discarded and replaced with fresh media and the appropriate pictures were accordingly taken. For flow cytometry study, after removing of media and washing twice, the cells were trypsinized and the cells suspension in non-FBS media was centrifuged at 1600 rpm www.nature.com/scientificreports/ for 5 min. Afterward, the pellet was dissolved in cold PBS (pH 7.4) and the fluorescence intensity was measured using a BD FACSCalibur™ in the FL2 channel and analyzed by FlowJo 10.6 software.
In vivo antitumor efficacy of the synthesized formulations. BALB/c mice were obtained from Animal Resources Center (Pasteur Institute, Iran). To obtain tumor-bearing mice, C26 cells (3 × 10 5 cells/100 μl PBS) were inoculated into the right side subcutaneous region of male 4-5 week old BALB/c (~ 20 g) mice. When the size of the tumor reached ~ 50 mm 3 , the mice were randomly divided into five groups (n = 5). Group I (control group) was injected with 200 μL PBS, groups II, III, and IV were intravenously injected with 200 μL of free DOX, SPION/PVD/MUC1 Apt /DOX and SPION/PVD/Scr Apt /DOX (DOX equivalent concentration 0.2 mg/kg) via a single tail-vein injection. The mice body weight and survival rates of the tumor-bearing mice were investigated and the tumor volume was calculated by the formula length × width × height × 0.5 33,35 . The mice were checked 26 days after tumor induction until they reached endpoint (body weight loss of > 20%).
Biodistribution study of the synthesized formulations. When the size of the tumor of BALB/c mice reached 200-300 mm 3 , the mice were intravenously injected with free DOX, SPION/PVD/MUC1 Apt /DOX and SPION/PVD/Scr Apt /DOX (DOX equivalent concentration 0.2 mg/kg). After 6 and 24 h of injection, the animals were sacrificed and tumor tissues, heart, lungs, liver, kidneys, and spleen were excised and biodistribution pattern was measured according to the fluorescent intensity of these organs at λ ex = 450 and λ em = 580 by a Kodak FX Pro in vivo imaging system. Statistical analysis. The statistical analysis was performed by one-way analysis of variance (ANOVA) to determine the significant difference between groups (a P-value less than 0.05) by using GraphPad Prism Version 8.0 (GraphPad Software Inc, USA, https:// www. graph pad. com). Quantitative results are indicated as the mean ± standard deviation (SD).