Enhanced transferrin modified targeting delivery system via AFP promoter MR imaging reporter gene transfection

Background: Effective treatment and early diagnosis is essential for patients with hepatocellular carcinoma. Nowadays, there is an increasing interest in the utilization of AFP/Fth as an endogenous contrast agent for the early diagnosis of liver cancers. The transfection of AFP/Fth also leads to a considerable upregulation of transferrin receptors (TfR), which might be a potential target for an enhanced delivery of nanomedicines. However, there is no information regarding the utilization of overexpressed TfR for the treatment of liver cancers. Thus, the objective of our study was to investigate whether the transfection of AFP/Fth could be used for the early diagnosis, but also for an enhanced treatment of live cancers. Results: It was found that both of ferritin and TfR were upregulated after the transfection of AFP/Fth plasmid. The transfected cells showed a higher uptake of ferric ion than the un-transfected ones, resulting in a lower T2WI intensity. This result was due to the upregulation of ferritin in transfected cells, suggesting that transfection of AFP/Fth plasmid could be a potential strategy for early diagnosis of liver cancer. As compared with the un-transfected cells, the transfected cells showed a higher uptake of transferrin-modified liposomes, which was due to the specific interaction between transferrin with TfR overexpressed on the transfected cells. This is also the reason why the transferrin modified doxorubicin loaded liposomes (Tf-LP/DOX) showed better in vitro and in vivo anticancer ability than the LP/DOX. This results also suggested that transfection of AFP/Fth could result in an enhanced therapy of liver cancer. Conclusions: Transfection of AFP/Fth could be used for the early diagnosis, but also for an enhanced treatment of live cancers. fluorescence signal of Tf-LP/ICG in tumor collected in transfected mice was significant stronger than that of transfected ones. Nevertheless, the tumor of mice treated with LP/ICG showed significant differences in fluorescence intensity, further verifying the enhanced targeting ability of transferrin-modified liposomes to transfected AFP positive hepatoma carcinoma cell. These results were nearly consistent with those obtained in cellular uptake. pathological difference between the four preparation groups, the free drug group and the physiological group. The microscopic images were observed under Leica fluorescence microscope.

Then, the drug loading (DL) and encapsulation efficacy (EE) of liposomes were quantified by a fluorescence spectrophotometer via a centrifugal ultrafiltration method. As displayed in Table 1, both of LP and Tf-LP had a high DOX encapsulation efficiency, which was 88.11±3.39% and 93.82±1.28%, respectively. The drug loading of LP and Tf-LP were calculated to be 9.85±0.69% and 10.26±0.72%, respectively. The drug release profiles suggested that free DOX exhibited a rapid release behavior, with approximately 100% of DOX was released within 10 h ( Figure 1F). In comparison, Tf-LP/DOX had a proper sustained release profile, with 89.8% of DOX was released within 24h. We also found that the drug release behavior of Tf-LP/DOX was quite similar to that of LP/DOX, suggesting that the transferrin modification slightly influenced the drug release performance of liposomes.

AFP specific expression and AFP promoter drives Fth expression in AFP-positive cells
The expression levels of AFP protein in HepG2 and LO2 cells were verified by western blot analysis.
The results showed that LO2 cells had a lower amount of APF protein as compared with HepG2 cells (Figures 2A and 2C). As the expression level of AFP protein is a good indicator of AFP activity [25], HepG2 cells were considered AFP positive, while LO2 cells were considered as AFP-negative in this study. The above result demonstrated the feasibility of AFP as the promoter in AFP/Fth plasmid and provide the basis for the subsequent verification test.
To examine whether transfected by the plasmid which contains ferritin heavy chain at the same time with AFP as a promoter caused ferritin overexpression, western blot was performed. As presented in Figures 2B and 2D, the expression level of ferritin in HepG2 with transfection was significantly higher than those without transfection. On the other hand, transfection for 48 h resulted in a higher amount of ferritin expressed in HepG2 cells than those for 24 h. These results have supported that ferritin could be overexpressed after the liver cancer cells were transfected with AFP/Fth. Transfection for 48h as a key time point was then chosen for the following studies.

Specific upregulation of TfR
To examine the effects of AFP/Fth transfection time (24 and 48h) on the TfR expression level, hepatoma carcinoma cells (HepG2) were transfected with or without ferritin reporter gene and observed through a laser confocal microscope via an immunofluorescence method. Apparently, TfR was specifically up-regulated in transfected hepatoma carcinoma cells, as evidence showing that the green fluorescence intensity of transfected HepG2 was much stronger than that of untransfected one ( Figure 2E). It was also found that the HepG2 transfected with ferritin reporter for 48 h had a higher fluorescence intensity than those transfected for 24 h, indicating that the expression level of TfR could be increased after transfection and showed a more satisfied expression with a proper prolonged transfection time. Collectively, these growing evidences demonstrated that the AFP / Fth plasmid induced the upregulation of TfR on its cell surface after transfection, providing a basis for subsequent targeted therapy.

The signal decreased effect in MRI
The collected cells were subjected to MR imaging to study the contrast imaging effect caused by changed iron uptake ability via AFP/Fth transfection. It was observed in all three groups (nontransfection HepG2 cells group, 24h-transfection HepG2 cells group, 48h-transfection HepG2 cells group) that cells supplemented with an extra FAC concentration resulted in a significant decrease in the signal intensity. It can be explained that on the basis of regulation of transferrin receptors on cancer cell surface, the extra FAC supplement leaded to more cellular iron accumulation, which caused the T2WI intensity signal decline. The growing results of T2-/T2*-WI also suggested that the signal intensity of 24h/48h-transfection HepG2 cells were significantly lower than those without transfection, whether FAC or not ( Figure 2F). These phenomena maybe induced by the ferritin overexpression and transferrin receptor upregulation in hepatoma cells after AFP/Fth plasmid transfected. On the other hand, 48h-transfection HepG2 cells group further showed lower signal intensity than those 24h-transfection. It was indicated that 48-hour transfection give rise to a more receptor overexpression possibility so that more iron could be transported into the cell, compared to 24-hour transfection. The quantitative analysis according to the region of interest (ROI=8px) of the T2*WI were analyzed and showed in Figure 2G and 2H. This serial of analysis confirmed that as shown in previous transfection experiments, ferritin overexpression, transferrin receptor up-regulation could lead to the intracellular iron accumulation in the condition of FAC provided, which then bring about the lower T2WI intensity.

Cellular uptake of transferrin modified particles in HepG2 cells
The cellular uptake of liposomes was observed by a fluorescence inverted microscope. A fluorescence dye of FITC was used to label the blank and target-modified liposomes, which was abbreviated as LP FITC and Tf-LP FITC , respectively. As shown in Figures 3A and 3B, both of transfected and untransfected HepG2 cells showed a gradual increase in fluorescence intensity as the incubation time was increased, suggesting that the cellular uptake of liposomes was time dependent. Looking deeply, transfected HepG2 cells treated with Tf-LP FITC had a stronger fluorescence intensity than those treated with LP FITC, indicating that transferrin-modification could remarkably increase the uptake of liposomes. However, this behavior was not observed in the untransfected HepG2 cells, suggesting that transferrin-modification could specifically target the liposomes to the transfected cells, rather than the untransfected ones. As we mentioned previously, transferrin receptors were largely upregulated on the surface of HepG2 cells after the cells were transfected. The transferrin on the surface of liposomes could specifically target the transferrin receptors, thereby leading to the observation that a higher number of Tf-LP FITC were uptake by the transfected HepG2 cells. Previous research also reported that the transferrin modified nanoparticles have a similar specific homing function for tumor cells that owned abundant TfR to mediate the cellular uptake of drugs [28,44,45].

The cytotoxicity of various liposomes
The toxicity of liposomes loaded with or without Dox was investigated via an MTT assay, in which HepG2 cells with or without transfection were used. As revealed in Figure   Cell viability was measured for HepG2 cells incubated for 24 h with LP/DOX and Tf-LP/DOX at various doses for 24h; (E) Cell viability was measured for HepG2 cells that transfected for 48h and then incubated for with LP/DOX and Tf-LP/DOX at various doses for 24h; **p<0.01, (***p<0.001, vs free DOX group, n=6).

Western blotting for tumor transfection
To further verify the targeting expression of Fth in AFP positive hepatocellular carcinoma, the expression level of ferritin was studied via the western blotting method. As shown in Figure 4A, the transfection induced a higher expression level of Fth observed in tumor as compared to liver tissue, which should be mainly attributed to the active targeting ability of Tf-LP/DOX. The quantitative data in Figure 4B also showed that the expression level of Fth protein in transfected tumor was much higher than that in liver and non-transfected tissues, which was in accordance with the easier results obtained in western blotting analysis of cells.

Targeting ability of transferrin modified particles in vivo.
In vivo targeting ability of transferrin-modified liposomes was investigated by intravenously injected the mice with ICG-labeled Tf-LP (Tf-LP/ICG) or LP/ICG. After 24h and 48h of injection, the fluorescence images of major organs and tumors were captured, followed by quantifying the fluorescence intensity of tumors ( Figures 4C and 4D). It was shown that the liposomes were mainly accumulated in the liver and tumor. The fluorescence signal of Tf-LP/ICG in tumor collected in transfected mice was significant stronger than that of transfected ones. Nevertheless, the tumor of mice treated with LP/ICG showed significant differences in fluorescence intensity, further verifying the enhanced targeting ability of transferrin-modified liposomes to transfected AFP positive hepatoma carcinoma cell. These results were nearly consistent with those obtained in cellular uptake.

The anti-tumor effects
To confirm the therapeutic efficacy of Tf-LP/DOX to the HepG2 cell subcutaneous hepatoma model, we randomly divided the model mice for 6 groups ( Figure 5A), and administrated for every 5 days ( Figure   5B) with a single dose of 4mg/Kg according to the mean weight of each group via tail vein injection. It should be noted that intratumor injection transfection was given two days prior to treatment for two of the group ( Figure 5A). As shown in Figure Figure 5D, the body weights of mice treated with free DOX significantly decreased, which was on account of its systemic toxicity. Besides the similar results as cytotoxicity experiments, Tf-LP/DOX showed surprising antitumor effects after transfection, and its efficacy was even beyond free doxorubicin, which could be explained as the tumor microenvironment was more suitable for Tf-LP / DOX to accumulate and free doxorubicin had no targeting property in vivo in the meantime.

Histological analysis
Anti-tumor effect and side effects of different drugs on important organs during drug administration were observed through H&E staining. The anti-tumor extent was presented in Figure 5E, a multitude of tumor cells were destroyed in Tf-LP/DOX of transfection group. No obviously serious myocardial damage was observed in all groups, which may cause by the short administration period [34].

Western blotting for AFP and Ferritin expression
The ferritin expression was confirmed by Western Blot assay. HepG2 cell was incubated for 48 h after transfection, using 80 ul RIPA buffer that containing protease inhibitor (Sigma, USA) per well to lyse the cells in 6-well plate on ice for 30 minutes. After centrifuging and collecting the cells in EP tubes, the protein concentration was evaluated by the bicinchoninic acid assay. The protein solution was added to 5* loading buffer in a ratio of 4:1 and denaturated in boiling water bath for 15 minutes. Then the samples were loaded onto 12% gels for SDS-PAGE. The cut gels which including the molecular weight between 21KD (ferritin heavy chain) and 42KD (β-actin) were transferred onto PVDF membranes for blotting. The membranes were blocked in 5% skim milk powder solution for 2 h at 37 °C and then incubated overnight at 4 °C with rabbit anti-Ferritin monoclonal antibody (1:2000, Abcam), the membranes were incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG

Transfection in vivo
The equal amount of plasmid that have mixed with PEI for 30 minutes were given to the nude mice of transfected group two days before the drug was administered, and then intratumorally injected for transfection, which was set for 24 h and 48 h. Finally, the tumors were ground into a slurry for WB detection to observe the expression of ferritin, and the liver tissues were also used as a control [47].
Targeting ability of transferrin modified particles in vivo.
HepG2 cells (1 × 10 6 cells/0.2 ml) were injected subcutaneously into in the right axillary of the female BALB/C nude mice (Zhejiang University, Hangzhou, China) and 0.2 ml Tf-LP/ICG and LP/ICG were given to the different groups with the same concentration by tail vein injection when tumors' volume reached 100mm 3 approximately. The distribution in vivo will be observed through whole-body fluorescent imaging system at 24 h, 48 h respectively.
The anti-tumor effect in mice After successful tumor modeling, mice bearing tumor about 100 mm 3 were used in this study, the nude mice were randomly divided into 6 groups of five to six (Fig. 5A). Two of the groups will be transfected, and Tf-LP/DOX and LP/DOX were administrated separately. Tf-LP/DOX and LP/DOX were also given to another two groups, but no transfection pretreatment was performed. The remaining two groups were the free doxorubicin group and the saline group. Each group of mice were administered every 5 days at a single dose of 4 mg/Kg according to the mean weight of each group via tail vein injection (Fig. 5B). The volume of the tumor and the body weight of the mice were monitored on the day of the next administration to observe the anti-tumor effect among transfection and non-transfection, preparation and free drug. At indicated times after injection, mice were sacrificed and the heart, liver, spleen, lung, kidney, tumor were excised instantaneously. After gently washing away the excess blood, the tissue sections were used for HE staining observation. The tumor volume was determined by the two bisecting diameters of each tumor, which represented the largest (a) and smallest (b) diameter(mm) of the tumors, respectively, and the volume was calculated using the formula 0.5X (a*b 2 ).

The histology
The tumor tissues were submerged into 10% formalin for 48 h, then embedded in paraffin. Then HE staining was used to analyze the pathological changes of important organs between different groups of nude mice, and to observe the pathological difference between the four preparation groups, the free drug group and the physiological group. The microscopic images were observed under Leica fluorescence microscope.

Statistical analysis
All the quantitative data were shown as means ± standard deviation (SD) of three separate experiments. Discrepancies between two groups were conducted by Student's t test and one-way analysis of variance (ANOVA) was used for discerning discrepancies between multiple groups. A Pvalue < 0.05 was considered statistically significant. All statistical analyses were done using STATA software.

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download.