Biodegradable mesoporous manganese carbonate nanocomposites for LED light-driven cancer therapy via enhancing photodynamic therapy and attenuating survivin expression

Triple-negative breast cancer (TNBC) is one of the most daunting diseases, low toxicity and efficient approaches are in urgent demand. Herein, we developed degradable mesoporous manganese carbonate nanocubes (MnCO3 NCs), incorporated with survivin shRNA-expressing plasmid DNA (iSur-pDNA) and riboflavin (Rf), namely MRp NCs, for synergistic TNBC therapy. The MnCO3, itself, could generate O2 and CO2 under H2O2 and thus relieve the hypoxia and acidic tumor microenvironment (TME). Furthermore, the MnCO3 NCs exhibited high Rf loading capacity and iSur-pDNA delivery ability after polyethyleneimine modification. Specifically, MRp NCs decompose in TME, meanwhile they deprived the endogenous expression of survivin gene and significantly amplified the generation of reactive oxygen species after exposure to LED light, resulting in serious tumor destruction. The multifunctional MRp NCs with LED light-driven characters are able to provide a high efficiency, low toxicity and promising strategy for TNBC therapy. Supplementary Information The online version contains supplementary material available at 10.1186/s12951-021-01057-2.


Introduction
Triple-negative breast cancer (TNBC) is an important and intractable subtype of breast cancer due to the lack of biomarkers and its high metastasis [1]. There was no significant progress in TNBC treatment during the past decades [2]. Traditional chemotherapies are still the main approaches for TNBC therapy, but they exhibit high toxicity and low efficiency, resulting in poor life quality and low survival rates. It is urgent to develop new approaches with high safety, low toxicity, and high efficiency to deal with TNBC [3].
Photodynamic therapy (PDT) [4] and chemodynamic therapy (CDT) based on reactive oxygen species (ROS) provide new alternative opportunities for cancer therapy [5]. They exhibit high selectivity in cancer theranostic [6] and could be activated by inner (e.g., low pH, abundant glutathione, and H 2 O 2 ) [7,8] or external stimulus (light, magnetic field, and heat) [9][10][11] compared with chemotherapy. Manganese-based nanoparticles have been widely reported for cancer theranostic because of their excellent tumor microenvironment (TME) responsive characters and potential CDT effect [12]. These features selectively damage the tumor cells while protecting the normal cells because they are restrained in the specific tumor regions. We previously found Bi@MnO x nanoparticles could respond to both inner and external stimuli, exhibiting a mutual reinforcement for cancer therapy [13]. Recently, researchers have focused on the catalytic reaction of MnOx [14], and various ROS-based nanozymes have been developed for cancer therapy (e.g., MnOx-SPNs [15], Au-MnO [16], and MnOx [17]). However, it is still difficult to tackle with TNBC only using CDT/PDT approaches.
To address these issues, we developed MnCO 3 /Rf/ pDNA nanocomposites (denoted as MRp NCs) which consisted of mesoporous MnCO 3 nanocubes (NCs) loading with riboflavin (Rf ) and survivin shRNA expressing plasmid (iSur-pDNA) for combined TNBC therapy. Rf, as a necessary nutrient for the human body, could also work as a photosensitizer. Its ROS production was significantly amplified in the presence of MnCO 3 NCs. Moreover, the polyethyleneimine (PEI) modified MnCO 3 NCs could efficiently deliver iSur-pDNA to 4T1 cells for survivin gene silencing. The MRp NCs illustrate multiple roles in TNBC therapy: (i) as TME ameliorative agents for improving tumor acidity and hypoxia; (ii) as a biodegradable drug and pDNA carrier; The high surface potential enables the PEI-MnCO 3 NCs with high pDNA transfection efficiency; (iii) as an assistant for combined TNBC therapy. The MRp NCs can be decomposed under simulated TME solution, resulting in the release of Mn 2+ , O 2 and CO 2 for enhancing PDT and CDT, moreover, the generated O 2 and CO 2 could also destroy the tumor tissue, and the delivered pDNA could deprive the survivin gene, thus enhancing tumor cell destruction (Scheme 1).

Preparation of MnCO 3 NCs
The MnCO 3 synthetic process was according to our previous work [19]. CTAB (2 g), MnCl 2 ·4H 2 O (10 mmol) were mixed in 2.0 mL water, 3.0 mL 1-butanol and 60 mL cyclohexane, and then the mixture was vigorously stirred at room temperature, named as A solution. CTAB (8 g), of KHCO 3 (19 mmol), NH 4 HCO 3 (1 mmol), 8.0 mL water, 3.0 mL 1-butanol and 240 mL cyclohexane were mixed and vigorously stirred in container B. After magnetic stirring for 1 h, solution A was added to container B under continuous stirring. After reacted for another 0.5 h, the solution was centrifuged at 8000 rpm for 10 min to remove the supernatant. The final MnCO 3 was washed with ethanol and dd H 2 O three times, and then the precipitates were extracted several times using methanol with 1% NaCl to remove the redundant CTAB.

Modification of MnCO 3 NCs
Surface modification of MnCO 3 NCs with amine-containing PEI was followed below, 0.1 g MnCO 3 NCs were dispersed in 100 mL ddH 2 O with vigorously stirring, then 0.1 g PEI was added to the solution. The mixture was stirred at room temperature for another 2 h. The PEI-MnCO 3 NCs were collected by centrifugation (10,000 rpm, 10 min) and washed with water 3 times to remove the redundant PEI.

Characterization
The powder X-ray diffraction (XRD) patterns were collected with a Siemens Kristalloflex 810 D-500X-ray diffractometer using Cu Kα irradiation (λ = 1.5406 Å). High-resolution transmission electron microscopy was taken on a field emission scanning electron microscope (JEOL JEM-2100F, Japan). Zeta potential and hydrophilic size were measured using a zetasizer (Zetasizer Nano ZS, Malvern, UK). UV-vis-NIR absorption spectra and absorbance were examined using a multifunctional microplate reader (TECAN, infinite M200 PRO, Swiss).

Drug loading
PEI-MnCO 3 NCs (500 μg) were suspended in 5 mL PBS solution, and then Rf was dispersed in the solution at a concentration of 100 μg mL −1 . After stirring at 4 °C for 24 h, the solution was centrifuged. And the supernatant and precipitations were collected respectively. PEI-MnCO 3 /Rf NCs was named as MRf. The drug loading efficiency was calculated as below (Eq. 1):

Extracellular O 2 measurement
The O 2 production from PEI-MnCO 3 NCs in H 2 O 2 solution was monitored by a portable dissolved oxygen meter (HANNA HI 2400). Briefly, different concentrations of PEI-MnCO 3 NCs (0, 100, 200 μg mL −1 ) was added to 10 mM H 2 O 2 PBS solution. The data was recorded every 5 s for 10 min using the portable dissolved oxygen meter.

Biodegradation of MnCO 3 in TME simulation solution
PEI-MnCO 3 NCs were incubated in a solution of PBS (pH = 6.5) containing 2 mM H 2 O 2 for 3, 12 and 24 h, respectively. The morphologic changes of the PEI-MnCO 3 NCs were observed using TEM. In addition, PEI-MnCO 3 NCs were incubated in PBS (pH = 6.5) for 10, 30, 60 and 120 min, the CO 2 contents were assessed by meteorological chromatograph. After 2 h, the solution was centrifuged, the precipitates were analyzed using XPS.

Cell lines
The mouse TNBC cell line 4T1 and L929 cells were obtained from American Type Culture Collection. 4T1-Luc cells were maintained in RPMI 1640 medium (Sigma) with 10% FBS and penicillin (100 U/mL) and streptomycin (100 μg/mL) (Invitrogen). L929 cells were maintained in Dulbecco's modified Eagle's medium (Sigma) with 10% fetal bovine serum (FBS, Gibco) and penicillin (100 U/mL) and streptomycin (100 μg/mL) (Invitrogen). The cells were cultured at 37 °C under a humidified atmosphere of 95% air and 5% CO 2 and the medium was changed every 2 days.

Transfection of pDNA
4T1 cells were seeded in 6-well plates with a density of 5 × 10 4 cells per well. The medium was removed with fresh 1640 medium without FBS. All NCs were prepared by MRf/pDNA with a weight ratio of 15/1. Then 15 μL MnCO 3 /pDNA mixture was added to the 6-well plate co-cultured for 6 h. The medium was changed with fresh 1640 containing 10% FBS and 1% PS.

Intracellular ROS detection
Intracellular ROS production was detected by DCFH-DA. In brief, 4T1 cells were seeded in 24-well plate (1 × 10 5 cells per mL) and cultured overnight. Then cells were treated similarly as above (as live/dead staining). Finally, the cells were incubated with DCFH-DA probe (1 μM) for 15 min, washed with PBS and observed by fluorescence microscopy. Moreover, their quantitative analysis was using a multifunctional microplate reader (Ex/Em: 488/525 nm).

Intracellular pH detection
The changes of intracellular pH were using an intracellular pH fluorescence probe (BCECF AM). Briefly, the 4T1 cells were treated with PEI-MnCO 3 NCs (50 μg mL −1 ) as experiment group and medium as control group, then the cells were cultured with BCECF AM (5 μM) for 20 min, and their images were observed under fluorescence microscopy. And their quantitative analysis was using a multifunctional microplate reader (Ex/Em: 488/535 nm).

Animals
Balb/c nude mice (6-week-old, female) were purchased from Guangdong Medical Lab. Animal Center. The protocol was approved by the Institutional Animal Care and Use Committee of General Hospital of Southern Theater Command of PLA.
In vivo tumor therapy 5 × 10 6 4T1 cells were injected to the second breast nodule of the nude mice. After the tumors grew to a size of 50-70 mm 2 , the mice were divided into 5 random groups (n = 4) undergoing different treatments: (1)

Characterization of MRp NCs
Firstly, the monodisperse MnCO 3 NCs were prepared by a microemulsion method according to our previous method [19]. Transmission electron microscope (TEM) image (Fig. 1a) revealed that the MnCO 3 NCs had cubic-like morphology with the particle size of ca. 120 nm. As shown in Fig. 1b, high resolution TEM image with a typical individual nanocube inset revealed its highly porous nature and the marked lattice spacings of 0.285 nm which could be indexed to the (104) planes of MnCO 3 . X-ray diffraction (XRD) pattern was employed to detect the crystalline phase and purity of the samples. The results (Fig. 1c) revealed that the samples were pure rhombohedral MnCO 3 (JCPDS Card No. 44-1472). In addition, the porous structure of the MnCO 3 NCs was investigated by Brunauer-Emmett-Teller (BET) analysis. As depicted in Fig. 1d and e, the PEI-MnCO 3 NCs exhibited high Brunauer-Emmett-Teller surface area (49.97 m 2 g −1 ) and pore volume (0.293 cm 3 g −1 ), respectively. The average pore size is about 3.41 nm according to the N 2 adsorption-desorption isotherms. Zeta potential of the CTAB-MnCO 3 , fine MnCO 3 (removal of CTAB) and PEI-MnCO 3 were shown in Fig. 1f, illustrating the successful modification of PEI. The porous structure of the PEI-MnCO 3 NCs endows them with excellent Rf loading capacity. As can be seen in Fig. 1g, the loading efficiency (w.t%) of Rf in PEI-MnCO 3 NCs was calculated as high as 90%, confirmed by the absorption spectra.
More importantly, the PEI-MnCO 3 /Rf (MRf ) NCs presented a high binding ability to pDNA because of their high zeta potential. The binding ability was investigated by gel retardation assays (as shown in Fig. 1h), the results illustrated that the iSur-pDNA could be completely  Fig. S1, ESI †). The changes indicated the successfully loading of Rf and pDNA. Moreover, we have assessed the ROS production ability of PEI-MnCO 3 . Rf and MRf under the same condition using 5, 5-dimethyl-1-pyrroline-N-oxide (DMPO) as trapping agent. Inspiringly, MRf NCs exhibited significantly enhanced ROS production than PEI-MnCO 3 and Rf, respectively (Fig. 1i).

TME responsive characters
To verify the TME responsive characters of the PEI-MnCO 3 NCs, we systematically analyzed their degradation characters, catalytic ability, and gas production under simulated TME (pH = 6.5, 2 mM H 2 O 2 , 8 mM glutathione). As can be seen in Fig. 2a (Fig. 2a). Meanwhile, they decomposed slowly (Additional file 1: Fig. S2) in TME and produced CO 2 (Additional file 1: Fig. S3) simultaneously. Their morphology changes were observed by TEM, illustrating a dynamic change of morphology, i.e., cubic-roundcircle-dots, and PEI-MnCO 3 NCs finally could be degraded into tiny round nanodots (Additional file 1: Fig.  S2). Notably, the O 2 production soared to 18.1 μg mL −1 while the CO 2 production was in a relatively low speed (Fig. 2a, Additional file 1: Fig. S3). By contrast, there was negligible O 2 production in commercial MnCO 3 under the same condition (Additional file 1: Fig. S4). During the catalytic process, PEI-MnCO 3 NCs degraded into small pieces (Additional file 1: Fig. S2), release Mn 2+ and OH − (Eqs. 3, 4), the OH − ion is beneficial to improve the acidic TME while Mn 2+ facilitate the Fenton reaction [20] (Eqs. 4, 5) in tumors. To prove this process, we carried out XRD to evaluate the NCs in simulated TME, the results indicated the partially degraded NCs were still MnCO 3 without any impurities (Additional file 1: Fig. S5). The multivalence Mn in XPS spectra further confirmed the release of OH − and the redox reaction in this process (Additional file 1: Fig. S6). As expected, we discovered 1 O 2 generation in PEI-MnCO 3 NCs during degradation, which is beneficial for CDT ( Fig. 2b and Fig. 2b). Subsequently, we investigated whether the MRf produced more ROS compared with single Rf group under simulated TME solution when illuminated by white-light LED light. Both MRf and Rf generated ROS (Fig. 2c). Notably, the MRf could enhance the ROS production during the observing time (10 min) and this phenomenon could be repeated 5 times, indicating that PEI-MnCO 3 NCs as drug loading carriers could significantly increase ROS production as well as protect Rf from photobleaching and photodamage [21] (Fig. 2d).
Note: Rf ' is the excited Rf.

Cellular ROS production and pH-responsive ability
Thanks to the excellent performance of MnCO 3 -based NCs in TME, we then investigated their cancer-killing efficiency. Firstly, the intracellular ROS production was monitored by using the green probe, 2' ,7'-dichlorofluorescein diacetate (DCFH-DA). As shown in Fig. 3a and b, the cells treated with MRf + LED exhibited much stronger fluorescent intensity than Rf + LED and PEI-MnCO 3 + LED groups. By contrast, the groups (control group, PEI-MnCO 3 , Rf and MRf ) without LED illumination showed weak fluorescence. The results suggested that the ROS production capability of Rf could be significantly improved by PEI-MnCO 3 in vitro. Next, we tested the changes of intracellular acidity because of the OH − release and pH-sensitive characters of PEI-MnCO 3 NCs. Noteworthy, the synthesized MnCO 3 NCs exhibited better pH stability than commercial MnCO 3 (Fig. 3c), which may attribute to their high surface area and porous structure. The green pH probe (2' ,7'-bis(carboxyethyl)-5(6)carboxyfluorescein, BCECF) was employed to investigate the intracellular pH changes of 4T1 cells after different treatments. The PEI-MnCO 3 treated cells exhibited much stronger fluorescence intensity than the control group, suggesting they could improve the acidic environment in tumor cells (Fig. 3 d, e). Such an interesting pH improvement will help to destroy lysosomes and ameliorate TME ability, thus accelerating the death of cancer cells. (3)

The killing effect of MRp NCs on TNBC cells in vitro
To verify the biocompatibility and anticancer effect of the synthesized NCs, MnCO 3 and PEI-MnCO 3 NCs with different concentrations were co-cultured with 4T1 cells and L929 cells, respectively. As shown in Fig. 4a and b, 4T1 cells were significantly destroyed by MnCO 3 and PEI-MnCO 3 NCs (p < 0.05) compared to L929 cells under the same condition, which illustrated the TME-responsive characters and selective toxicity of PEI-MnCO 3 NCs on cancer cells. Moreover, we investigated the cancer-killing effects of MnCO 3 -based NCs, i.e. PEI-MnCO 3 , Rf, MRf, and MRp in dark or under LED light. As expected, MRp + LED light exhibited the highest toxicity to 4T1 cells, suggesting CDT, PDT and pDNA comprised an enhanced tumor therapeutic efficacy (Fig. 4c). Furthermore, the live/dead staining was employed to investigate the cell status with different treatments. As compared to normal cells, 4T1 cells suffered from different levels of damage in the MnCO 3 -related groups. Specifically, all the cancer cells in MRf + light group became red (death) and round, the results further confirmed their excellent tumor-killing effect (Fig. 4d).

Intracellular distribution and characters in 4T1 cells
Because of the high killing efficiency of MnCO 3 -based NCs, we next investigated the behaviors of NCs in 4T1 cells. As shown in Fig. 5a, the FITC-MnCO 3 NCs was distributed in the cytoplasm, some of them around the lysosomes, with an overlap coefficient of 50% with the lysosome at the first 6 h; With the increase of time, the FITC-MnCO 3 NCs fluorescence area covers 81% of lysosomes at 24 h, illustrating a lysosome targeted effect (Additional file 1: Fig. S7). The acidic environment in lysosomes is beneficial to the degradation of MnCO 3 -based NCs, and the alkaline environment provided by MnCO 3 NCs will destroy lysosomes, thus accelerating the death of cancer cells.
Hypoxia is an important character of solid tumors, which contribute greatly to tumor metastasis and the resistance to radio/chemotherapy [22]. Moreover, hypoxia also limited the efficiency of PDT and CDT [23]. Hypoxia-inducible factor-1α (HIF-1α) is highly active under hypoxic conditions, resulting in the changes of caspase-3 and Bax expression in tumors [24]. Here, the relief of hypoxia in PEI-MnCO 3 was evaluated both by hypoxia probe and Western blot. As shown in Fig. 5b and c, the red fluorescence in hypoxia treatment group was significantly enhanced compared with the untreated group. In contrast, the hypoxia cells co-cultured with PEI-MnCO 3 NCs showed weak red fluorescence, illustrating the relief of hypoxia by intracellular O 2 generation of MnCO 3 NCs.
Survivin is overexpressed in TNBC membranes [25], moreover, active Survivin induces the abnormal expression of several genes, including Bcl-2, Bax, and caspase-3 [26]. We evaluated the transfection properties of PEI-MnCO 3 . As shown in Fig. 5d, e, the FLUC-pDNA combined PEI-MnCO 3 NCs exhibited stronger red fluorescence intensity than the free FLUC-pDNA group, confirming the effective transfection efficiency of . Data are represented as mean ± SD; n = 4; Statistical significance was analyzed by the two-tailed Student's t-test. *p < 0.05, **p < 0.01. All light treated groups were compared to the dark treated group in c. And all groups are compared to control group in e. e BCECF fluorescence (pH changes) images of 4T1 cells after different treatment PEI-MnCO 3 NCs. Meanwhile, we examined the related protein expressions after different treatments (Fig. 5f ). The results showed MnCO 3 -based NCs treatment significantly decreased the HIF-α expression, thus relieved the hypoxia status of tumors. Moreover, MRp knockdown the survivin gene in 4T1 cells. Together with the downregulation of HIF-α and survivin genes, the related pro-apoptotic proteins, caspase-3 and Bax were upregulated (Fig. 5f ). These genes work together to accelerate the progress of cell apoptosis and death [27].

Therapeutic effect in vivo
The above results clearly confirm the anticancer ability of the MRp in vitro, we further evaluated their anticancer   Fig. 6a, b, the tumor volume was significantly inhibited in MRp, MRf and PEI-MnCO 3 groups during the observed period. However, the tumor in Rf + light treatment group exhibited first restrained effect and subsequently promoted dynamic changes. This is probably because that the ROS released by Rf + LED light inhibited the tumor growth at first, then the hypoxia caused by PDT promoted tumor growth. While MnCO 3 -based groups exhibited better therapeutic effects than the control group because of the sustained TME amelioration (pH, hypoxia) and 1 O 2 generation. In addition, there were no notable differences in body changes among all treatment groups, and the mice in PEI-MnCO 3 based treatment groups were active, indicating their potential biosafety. Furthermore, the hematoxylin and eosin (H&E) staining of the tumors illustrated the high complex and rich vessels in control tumor, suggesting the vigorous proliferation ability of TNBC tumors. In contrast, the tumor tissues in MRp + Light group suffered from great damage compared to other groups. Noteworthy, there were lots of bubbles-like destruction in tumor sites after being treated with MnCO 3 based nanomaterials, suggesting the sustained CO 2 and O 2 generation could result in serious tumor destruction.
H&E staining of main organs (heart, liver, spleen, lung, and kidney) was performed after different treatments. As shown in Fig. 7, the myocardial cells and the glomeruli were intact and clear in the treatment groups. The glomerulus the hepatocytes and splenocytes were normal, and no damage or inflammatory was observed in the examined organs relative to the control group.

Conclusions
In summary, the mesoporous PEI-MnCO 3 NCs serve as drug loading (Rf ) and transfection system (pDNA) for efficient TNBC therapy have been established because of their porous structure and positive zeta potential. Importantly, the PEI-MnCO 3 NCs possessed TMEresponsive characters, O 2 generation ability and Mn 2+ mediated CDT. Significantly, the ROS production ability could be amplified and the suvivin gene was silenced by MRp, which efficiently inhibited TNBC growth both in vitro and in vivo. Interestingly, the bubble (O 2 and CO 2 ) produced in the therapeutic process also destroyed the tumor tissue severely, which may provide a new idea for tumor therapy.
Additional file 1: Fig. S1. The mean hydrodynamic diameter of MnCO 3 and MRp measured by DLS.