Background

Radiotherapy (RT), or the precise application of high energy ionizing radiation at the tumor site, can directly induce DNA breakages in tumor cells and/or indirectly damage tumor cells by generating reactive oxygen species (ROS) via water radiolysis and is an important tumor treatment strategy [1, 2]. More than 60% of malignant tumor patients receive RT at some stage during their illness, and 40% of the treatments are successful [3, 4]. However, the tumor microenvironment (TME) is complex and can contribute to the acquisition of RT resistance. Hypoxic conditions in the solid TME can be ascribed to an imbalance between the supply and consumption of O2 in rapidly proliferating tumor cells, as well as dysfunctional tumor vasculature [5]. The hypoxic TME prevents ROS generation and formation of DNA double-strand breaks (DSBs), which are usually induced by ionizing radiation [2, 5], ultimately leading to RT resistance. Furthermore, tumor hypoxic conditions can bring about upregulation of hypoxia-inducible factor 1α (HIF-1α), which promotes endothelial cell survival following RT and thus further promotes RT resistance [6].

Traditional medical methods have made use of hyperbaric oxygen inhalation to alleviate hypoxic conditions and improve the tumor concentration of O2, which acts a reservoir of radiation-induced ROS [6,7,8]. However, the dysfunctional tumor vascular system hinders the delivery of inhaled O2 to the tumor, and the potential for oxygen poisoning, barometric injury, and decompression disease seriously limits the clinical utility of this method [9]. For decades, research has focused on the enhancement of RT efficacy by increasing radiation doses, but the severe side effects caused by excessive high-energy radiation to normal tissue and organs are unavoidable [5, Full size image

The affinity of the dimeric form affibody for its target is usually higher than that of the monomeric form [47]. So, the ZHer2 affibody was dimerized upon the addition of a cysteine residue at the C-terminus. As expected, the molecular weight of ZHer2 under natural conditions (Fig. 2b, in the absence of 2-ME) was approximately double than that of ZHer2 under reductive conditions (Fig. 2b, in the presence of 2-ME), indicating that ZHer2 forms disulfide bonds and dimers under natural conditions.

A slight increase in molecular weight was observed after ZHer2 was labelled with the fluorescent agent 6-Carboxfluorescein (6-FAM) (Fig. 2c), with a labelling efficiency was approximately 100%. This label allowed for visual evaluation of the binding capacity between FAM-ZHer2 and Her2-positive cells.

Finally, the ZHer2 affibody was coupled with Pt@mPDA/MnO2/PDA NPs via a Michael addition/Schiff base reaction by conjugating the amino group to the oxidized quinone form of the catechol groups under weak alkaline conditions. The mean hydrodynamic size of NPs increased slightly from 185 to 201 nm after ZHer2 affibody conjugation (Additional file 1: Fig. S5a). Both NPs exhibited high colloidal stability and could be dispersed in PBS without aggregation over 24 h (Additional file 1: Fig. S5b). The elemental map** images (Fig. 2d–k) of mPDA/MnO2/PDA-ZHer2 NPs showed a homogeneous distribution of C, N, O, S, Mn, and Pt, further confirming the successful loading of cisplatin and coupling of ZHer2. In addition, the ZHer2 content was approximately 0.8 mg/g according to ICP-AES quantitation of the ZHer2-specific S element.

Her2-positive cell-specific binding and cytotoxicity assay

Flow cytometry (FCM) and confocal laser scanning microscopy (CLSM) (Additional file 1: Fig. S6a–c) showed a concentration-dependent increase in fluorescence for a human ovarian cancer cell line (SKOV-3) after incubation with FITC-labelled anti-Her2 antibodies, while negligible FITC fluorescence was visible in a breast cancer cell line (MCF-7). This confirmed that SKOV-3 cells are Her2-positive, while MCF-7 cells are Her2-negative, which is consistent with previous results [48]. The binding activity of FAM-ZHer2 with Her2-negative MCF-7 cells and Her2-positive SKOV-3 cells was examined by CLSM. As shown in Additional file 1: Fig. S7a, FAM-ZHer2 specifically binds to SKOV-3 cells, while negligible FAM-ZHer2 binding to MCF-7 cells was seen, indicating that these affibodies bind specifically to Her2-positive cancer cells. This Her2-specific binding activity was further confirmed by FCM, as strong FAM fluorescence was visible in SKOV-3 cells, while the binding was significantly reduced after pre-incubation with free ZHer2 (Additional file 1: Fig. S7b).

FCM analysis was performed to evaluate the affinity of affibody monomers and dimers to Her2 receptors by measuring the FAM fluorescence intensity. As shown in Additional file 1: Fig. S8a, the binding rates of the dimeric Her2 affibody were 58.4%, compared to 33.8% for the monomer Her2 affibody at the same molar concentration. This confirms that the affinity of the dimeric Her2 affibody was higher than that of the monomeric Her2 affibody for Her2 receptors.

To determine whether ZHer2 could enhance the internalization of NPs into Her2-overexpressing tumor cells, the uptake of Cy5.5-labelled mPDA/MnO2/PDA (Cy5.5@mPDA/MnO2/PDA) and mPDA/MnO2/PDA-ZHer2 (Cy5.5@mPDA/MnO2/PDA-ZHer2) NPs by SKOV-3 cells was measured using CLSM and quantified via a FCM assay. Minimal intracellular red fluorescence was observed in SKOV-3 cells after treatment with Cy5.5@mPDA/MnO2/PDA NPs (Fig. 3a). In striking contrast, strong Cy5.5 (red) fluorescence was visible in SKOV-3 cells after incubation with Cy5.5@mPDA/MnO2/PDA-ZHer2 NPs. A blocking experiment was carried out by pre-incubating SKOV-3 cells with free ZHer2 (20 μg/mL) to further explore the mechanism of endocytosis. CLSM images showed an attenuated fluorescence signal inside pre-incubated SKOV-3 cells, indicating that the occupation of Her2 receptors resulted in reduced endocytosis and therefore failed to bind the NP-conjugated ZHer2. Similarly, the FCM data showed that the uptake of cellular NPs significantly increased after conjugating ZHer2 to Cy5.5@mPDA/MnO2/PDA NPs; however, the cellular uptake was significantly reduced after pre-incubation of SKOV-3 cells with free ZHer2 (Fig. 3b). Collectively, these results confirm that the ZHer2 affibody can enhance the internalization of NPs into Her2-positive cancer cells due to the specific affinity between the ZHer2 affibody and the Her2 receptor.

Fig. 3
figure 3

a CLSM images of SKOV-3 cells after 2 h incubation with Cy5.5@mPDA/MnO2/PDA NPs and Cy5.5@mPDA/MnO2/PDA-ZHer2 NPs and after a 1 h pre-treatment with or without ZHer2. b Flow cytometry data for untreated SKOV-3 cells, SKOV-3 cells incubated for 4 h with Cy5.5@mPDA/MnO2/PDA NPs or Cy5.5@mPDA/MnO2/PDA-ZHer2 NPs, and cells pre-incubated with ZHer2 for 1 h before being exposed to Cy5.5@mPDA/MnO2/PDA-ZHer2 NPs for 4 h, and fluorescence intensity quantified. c MTT viability results for SKOV-3 cells after incubation with free cisplatin (Pt), Pt@mPDA/MnO2/PDA NPs, and Pt@mPDA/MnO2/PDA-ZHer2 NPs for 24 h. d Fluorescence images of calcein-AM/PI co-stained SKOV-3 cells after different treatments (dose of cisplatin: 48 μg/mL). e Intracellular ROS levels in SKOV-3 cells after treatment with different formulations

The cytocompatibility of drug-free mPDA/MnO2/PDA-ZHer2 NPs with human umbilical vein endothelial cells (HUVECs) was assessed using the MTT assay. As shown in Additional file 1: Fig. S8b, negligible toxicity to non-cancerous cells was observed, even when the concentration of mPDA/MnO2/PDA-ZHer2 NPs reached 250 μg/mL (> 90% viability), thus demonstrating the good cytocompatibility of the carrier materials.

All cisplatin-containing formulations exhibited dose-dependent cytotoxicity to SKOV-3 cells (Fig. 3c). The half maximal inhibitory concentrations (IC50) of free cisplatin, Pt@mPDA/MnO2/PDA NPs, and Pt@mPDA/MnO2/PDA-ZHer2 NPs for SKOV-3 cells were 9.79 ± 0.8 μg/mL, 3.18 ± 0.3 μg/mL, and 2.56 ± 0.1 μg/mL, respectively. In contrast to free cisplatin, the Pt@mPDA/MnO2/PDA NPs exhibited more potent toxicity to cells over the entire cisplatin dose range (0.6–48 μg/mL), which can be ascribed to more efficient cellular uptake of NPs. More encouragingly, the conjugation of ZHer2 to Pt@mPDA/MnO2/PDA NPs further improved the cytotoxicity of the NPs because of more efficient internalization mediated by the targeted affinity for ZHer2 on the NPs to Her2 receptors on SKOV-3 cells.

Calcein-AM/PI double staining was performed to evaluate the degree of apoptosis of the cells. The optimal degree of cell apoptosis was observed upon treatment with Pt@mPDA/MnO2/PDA-ZHer2 NPs (Fig. 3d), which is consistent with the MTT results.

Since O2 is the ROS-generating resource induced by X-ray treatment, we next evaluated the effect of MnO2-containing NPs on the intracellular oxidative stress levels. As shown in Fig. 3e, compared to X-ray treatment alone, treatment with MnO2-free mPDA/PDA-ZHer2 NPs combined with X-ray did not increase the intracellular ROS levels, as low DCF fluorescence was observed in cells receiving X-ray treatment with or without mPDA/PDA-ZHer2 NPs. The intracellular oxidative stress levels of SKOV-3 cells improved after treatment with mPDA/MnO2/PDA-ZHer2 NPs alone, which can be ascribed to their ability of MnO2 to produce HO• (one ROS species) [14, 15]. In stark contrast, cells receiving mPDA/MnO2/PDA-ZHer2 NPs and X-ray combined treatment largely increased the intracellular ROS levels compared with cells receiving MnO2-free mPDA/PDA-ZHer2 NPs + X-ray treatment. This finding indicates that treatment with MnO2-containing NPs can facilitate ROS generation induced by ionizing radiation, confirming that these NPs have the capacity to increase radiosensitivity.

Hemolysis assay in vitro

A hemolysis assay was performed to determine the pharmacological safety of the NPs. No morphological changes and no significant lysis were observed after RBCs were incubated with Pt@mPDA/MnO2/PDA NPs or Pt@mPDA/MnO2/PDA-ZHer2 NPs (Additional file 1: Fig. S9a, b). Both Pt@mPDA/MnO2/PDA NPs and Pt@mPDA/MnO2/PDA-ZHer2 NPs had low hemolytic activity with only ~ 4.4% and ~ 4.7% RBC lysis, respectively (Additional file 1: Fig. S9c). The results indicate that both Pt@mPDA/MnO2/PDA NPs and Pt@mPDA/MnO2/PDA-ZHer2 NPs are blood compatible biomaterials.

Tumor targeting profiles in vivo

Based on the in vitro results, ZHer2-containing NPs were expected to target Her2-overexpressing tumor in vivo. The TME-activated MRI ability of NPs was first investigated in the SKOV-3 tumor-bearing mouse model. The T1-weighted MRI signal in the tumors increased gradually over time, while negligible MRI signal enhancement was observed in the muscles (Fig. 4a and b). This can be attributed to the reduction of the MnO2 layer to the MRI agent Mn2+ by the high levels of GSH in the TME [15, 49], which makes the MnO2-containing NPs particularly attractive for tumor-specific imaging applications. Tumor accumulation of Pt@mPDA/MnO2/PDA and Pt@mPDA/MnO2/PDA-ZHer2 NPs in SKOV-3 tumor-bearing mice at different times after intravenous injection was evaluated by tracking the MRI signal. The MRI signal was observed at the tumor site at 1 h post-injection of Pt@mPDA/MnO2/PDA-ZHer2 NPs, and the intensity increased gradually over time, indicating tumor-specific accumulation of these NPs (Fig. 4c). In contrast, decreased tumor accumulation of Pt@mPDA/MnO2/PDA NPs was observed as the signal of the Pt@mPDA/MnO2/PDA-ZHer2 NPs was significantly stronger than that of Pt@mPDA/MnO2/PDA NPs (p < 0.05) at 6 h (Fig. 4d). The biodistribution of free cisplatin, Pt@mPDA/MnO2/PDA, and Pt@mPDA/MnO2/PDA-ZHer2 NPs in SKOV-3 tumors was determined by quantifying the Pt content using ICP-AES at 12 h post-injection (Additional file 1: Fig. S10). The Pt content in the tumors of mice treated with Pt@mPDA/MnO2/PDA-ZHer2 NPs was approximately 3.5-fold and two fold higher than that in mice treated with free cisplatin and Pt@mPDA/MnO2/PDA NPs, respectively. The results reflect the more tumor retention of cisplatin-loaded NPs and tumor-targeting ability of the ZHer2 affibody.

Fig. 4
figure 4

a T1-MR images of SKOV-3 tumor-bearing mice before and after direct injection of Pt@mPDA/MnO2/PDA NPs (50 μL in PBS; [Mn] = 50 mM) into tumor (right, circle indicated) and muscle (left, circle indicated) tissues; b The quantified T1-MR signals corresponding to (a); c T1-weighted images of mice bearing SKOV-3 tumor grafts (arrow indicated) intravenously injection with Pt@mPDA/MnO2/PDA NPs or Pt@mPDA/MnO2/PDA-ZHer2 NPs ([Mn] = 3 mM, 100 μL in PBS per mouse) at 1 h, 3 h and 6 h, respectively; d The quantified T1-MR signals corresponding to (c)

Immunofluorescence and bio-TEM assays were performed to further evaluate the tumor-targeting ability of Cy5.5@mPDA/MnO2/PDA-ZHer2 NPs at the histological level. Immunofluorescence staining showed that Cy5.5 fluorescence (representing Cy5.5@mPDA/MnO2/PDA NPs) was mainly restricted to the tumor peripheral tissue (Fig. 5a and c (I)), suggesting poor penetration of the NPs into the TME. It was very encouraging to see that in the presence of ZHer2, the NPs overcame these biological barriers and penetrated deeply into the tumor tissue. As shown in Fig. 5b and c (II), strong red fluorescence was observed throughout the tumor, which co-localized with green fluorescence (representing Her2). These results confirm the efficient tumor homing and penetration of Cy5.5@mPDA/MnO2/PDA-ZHer2 NPs, which is afforded by ZHer2.

Fig. 5
figure 5

Immunofluorescence assay of Cy5.5@mPDA/MnO2/PDA NPs and Cy5.5@mPDA/MnO2/PDA-ZHer2 NPs in SKOV-3 tumor tissues. Mice bearing SKOV-3 tumors were intravenously injected with Cy5.5@mPDA/MnO2/PDA NPs or Cy5.5@mPDA/MnO2/PDA-ZHer2 NPs (100 μL; 1 mg/mL in PBS). Tumors were collected and further studied by immunofluorescence assay after 12 h post-injection. a The immunofluorescence images of Cy5.5@mPDA/MnO2/PDA NPs (red) in tumor tissues; b The immunofluorescence images of Cy5.5@mPDA/MnO2/PDA-ZHer2 NPs (red) in tumor tissues; c Representative magnified immunofluorescence images corresponding to (a) and (b). Scale bar = 100 μm. The nuclei and Her2 were stained with DAPI (blue), anti-Her2 antibody (green), respectively

This effect was also confirmed by bio-TEM imaging (Fig. 6) of tumor tissue. Greater amounts of Cy5.5@mPDA/MnO2/PDA-ZHer2 NPs were present within the cytoplasm, with little Cy5.5@mPDA/MnO2/PDA NPs being observed in the tumor tissue. This superior tumor penetration and targeting capabilities are crucial for enhancing therapeutic efficacy. All these findings suggest that ZHer2 modification combined with a TME-triggered off-to-on diagnostic agent make Pt@mPDA/MnO2/PDA-ZHer2 NPs particularly attractive for MRI-guided tumor-targeting treatment.

Fig. 6
figure 6

Bio-TEM images of Cy5.5@mPDA/MnO2/PDA NPs and Cy5.5@mPDA/MnO2/PDA-ZHer2 NPs in SKOV-3 tumor tissues. The red arrows indicate the NPs. Mice bearing SKOV-3 tumors were intravenously injected with Cy5.5@mPDA/MnO2/PDA NPs or Cy5.5@mPDA/MnO2/PDA-ZHer2 NPs (100 μL; 1 mg/mL in PBS). Tumors were collected and further studied by bio-TEM after 12 h post-injection

In vivo chemo-sensitized radiotherapy

Encouraged by the tumor-targeting and penetration effects of ZHer2-functionalized NPs, the in vivo chemotherapeutic activities in SKOV-3 tumor-bearing mice were then assessed. In mice treated with PBS, the tumor volume increased steadily during the treatment period (Fig. 7a). Mice that were administered free cisplatin did not exhibit appreciable tumor suppression, probably due to insufficient accumulation of cisplatin in the tumors. In sharp contrast, mice that received Pt@mPDA/MnO2/PDA NPs showed inhibited tumor progression as a result of the improved tumor retention of cisplatin-containing NPs. As expected, with the guidance of the ZHer2 affibody, Pt@mPDA/MnO2/PDA-ZHer2 NPs achieved the most potent inhibition of tumor growth, and by the end of the treatment, these extracted tumors had the smallest volume and mass (Fig. 7b and c), persuasively demonstrating the excellent targeted antitumor activity of Pt@mPDA/MnO2/PDA-ZHer2 NPs. Body weight was monitored every day, and no obvious difference was observed among all the treated groups (Fig. 7d).

Fig. 7
figure 7

In vivo chemotherapy efficacy of different formulations in mouse bearing SKOV-3 tumor grafts. When the volume of tumors reached about 50 mm3, mice were intravenously injected with cisplatin, Pt@mPDA/MnO2/PDA NPs or Pt@mPDA/MnO2/PDA-ZHer2 NPs (dose of cisplatin = 2 mg/kg) every 2 days. Tumor-bearing mice intravenously injected with the same volume of PBS were used as control group. The tumor volume and body weight were measured every day. On the 14th day after inoculation, all the tumor grafts were removed, weighed and analyzed by TUNEL staining. a SKOV-3 tumor growth curves of different groups after intravenously injection of the formulations. b The images of tumors isolated after treatment for 8 days. c Mean weights of the tumors isolated on day 14. d Body weight changes over the 8 days of the experiments. e TUNEL-stained images of tumor slices excised from each treatment group on day 14. The nuclei of cells were visualized using DAPI (blue)

A TUNEL assay was conducted to determine the levels of apoptosis in the tumor tissue. The largest number of green-colored cells (indicating the highest levels of apoptosis) were observed in tumors taken from mice receiving Pt@mPDA/MnO2/PDA-ZHer2 NPs (Fig. 7e), indicating that these NPs had the most potent antitumor effects.

Based on the remarkable tumor inhibition effects of Pt@mPDA/MnO2/PDA-ZHer2 NPs, the combined chemo-radiation therapeutic efficacy with these NPs, especially the enhanced RT effect, by adopting MnO2 as a radiosensitizer for hypoxic tumors was next assessed. The treatment schedule for chemo-radiotherapy is illustrated in Fig. 8a. RT alone did not have any significant antitumor effects, possibly because of RT resistance caused by the hypoxic TME and a sharp growth in tumor size (Fig. 8b–d) was observed during the treatment period. Mice receiving Pt@mPDA/MnO2/PDA-ZHer2 NPs and X-ray irradiation exhibited the most profound inhibition of tumor growth, which was significantly more noticeable higher than that of the mice receiving MnO2-free Pt@mPDA/PDA-ZHer2 NPs + X-ray treatment (p < 0.001). These results indicated successful sensitization to RT induced by MnO2. No obvious body weight changes were observed during any of the treatments for the duration of the experiment (Fig. 8e).

Fig. 8
figure 8

In vivo chemo-radiation combined therapy efficacy of different formulations in mice bearing SKOV-3 tumor grafts. When the volume of tumors reached about 50 mm3, mice were intravenously injected with MnO2-free Pt@mPDA/PDA-ZHer2 NPs, Pt@mPDA/MnO2/PDA NPs or Pt@mPDA/MnO2/PDA-ZHer2 NPs every 2 days. Mice in control group were intravenously injected with same volume of PBS. Mice received an X-Ray radiation at a dose of 6 Gy for 24 h post-injection every 4 days except for PBS group. Tumor sizes and body weights were recorded every day. On the 14th day after inoculation, all the tumor grafts were removed, weighed. a Schematic illustration of process of the chemo-radiation combined therapy; b SKOV-3 tumor growth curves of different groups after intravenously injection of the formulations; c The images of tumors isolated after treatment for 8 days; d Mean weights of the tumors isolated on day 14; e Body weight changes over the 8 days of the experiments

To better understand the mechanisms by which MnO2 sensitizes cells to RT, the HIF-1α expression level was measured in tumor tissue extracted from each treatment group by an immunofluorescence assay. Compared with that in all MnO2-free formulation treatment groups, the tumor tissues from the mice treated with Pt@mPDA/MnO2/PDA NPs showed a remarkable reduction in red fluorescence signal (HIF-1α) (Fig. 9). It indicated that TME in these mice was less hypoxic, which can be attributed to the decomposition of endogenous H2O2 to O2 by MnO2. Furthermore, with the targeted guiding of ZHer2 affibody, HIF-1α expression was further reduced by treatment with Pt@mPDA/MnO2/PDA-ZHer2 NPs, confirming that the tumor-targeting ability of the ZHer2 affibody contributed to the sensitization to RT.

Fig. 9
figure 9

Immunofluorescence images of HIF-1α (red)-stained tumor slices excised from the various treatment groups. The nuclei of cells were visualized using DAPI (blue)

Safety evaluation

Evaluation of off-target toxicity is a prerequisite for in vivo use or clinical translation of novel therapeutic agents. Thus, the biocompatibility and biosafety of Pt@mPDA/MnO2/PDA-ZHer2 NPs were systematically evaluated by histology and serum biochemistry assays. H&E-stained images of major organs from all treatment mice showed no obvious tissue damage, compared with that from the PBS group (Additional file 1: Fig. S11), suggesting that the drug, materials, or both combined did not cause any significant systemic toxicity. Similarly, the levels of serum biomarkers (ALT, AST, UREA, CREA, and UA) have no significant difference from those in the control group (Additional file 1: Fig. S12), possibly due to the relatively low side effects of clinical cisplatin and biomimetic PDA, as well as the short treatment period. Therefore, this biocompatible nano-theranostic agent exhibits promising potential for clinical translation.

Her2 is overexpressed in a variety of human cancers and is closely related to cell proliferation, differentiation, adhesion, migration, and anti-apoptosis, as well as poor prognosis and rapid recurrence of tumors [48]. Antibody–drug conjugates (ADCs, such as FDA-approved trastuzumab, pertuzumab, and T-DM1), which integrate Her2-specific targeting antibodies with highly cytotoxic small-molecule chemotherapeutic agents, possess the ability to selectively deliver highly potent cytotoxic drugs to tumor sites and have become a powerful cancer-targeting treatment approach. However, these ADCs can have a poor response in the TME and do not offer any diagnostic abilities, as well as being high cost, which limits their wider usage. In the present study, a Her2-targeting ZHer2 affibody was generated using a genetic engineering approach and successfully coupled to Pt@mPDA/MnO2/PDA NPs. Compared to the widely used intact antibody, this engineered affibody has the advantage of higher specific affinity, stronger tissue penetrability, higher synthetic yield at a lower price, and low immunogenicity [48, 50]. Such affibodies can be employed to guide nano-agents for enhanced therapeutic efficacy. In addition, in marked contrast to monofunctional antibody–drug conjugates, the ZHer2 affibody with a nano-agent, which features specific targeting, RT sensitization, and diagnostic abilities, has been prepared and provide a new concept for clinical translation in tumor theranostics.