Cryptobiosis-inspired assembly of "AND" logic gate platform for potential tumor-specific drug delivery.

Developing tumor-specific drug delivery systems with minimized off-target cargo leakage remains an enduring challenge. In this study, inspired from the natural cryptobiosis explored by certain organisms and stimuli-responsive polyphenol‒metal coordination chemistry, doxorubicin (DOX)-conjugated gelatin nanoparticles with protective shells formed by complex of tannic acid and FeIII (DG@TA-FeIII NPs) were successfully developed as an "AND" logic gate platform for tumor-targeted DOX delivery. Moreover, benefiting from the well-reported photothermal conversion ability of TA-FeIII complex, a synergistic tumor inhibition effect was confirmed by treating 4T1 tumor-bearing mice with DG@TA-FeIII NPs and localized near-infrared (NIR) laser irradiation. As a proof of concept study, this work present a simple strategy for developing "AND" logic gate platforms by coating enzyme-degradable drug conjugates with detachable polyphenol‒metal shells.

Introduction

In the past decades, innovations in the development of targeted and controlled drug delivery vehicles could always exhilarate scientists for overcoming the lack of specificity of conventional chemotherapeutic agents and combating life-threatening diseases. Although promising in preclinical animal models, the stochastic nature of ligand–receptor interactions in vivo and non-specific Fickian diffusion-governed drug leakage before arriving at pathological regions still remain the main barriers for clinical translation of these reported drug carriers2, 3, 4, 5. Recently, due to the altered metabolic pathways and/or pathological microenvironment, exploiting drug vehicles that are sensitive to disease-associated biochemical markers (such as dysregulated pH, enzymes and redox balance) has been preferred for liberating cargos in a particular area with efficient spatial, temporal and dosage control6, 7, 8. Frustratingly, it is still a tremendous challenge to realize accurate drug delivery triggered by a single biomarker which is rarely unique in the diseased area. Despite dual/multi-stimuli responsive systems have been further designed for accelerating the offloading of therapeutic drugs, the commonly utilized “OR” logic gates always lead to adverse off-target drug leakage on the way activated by each stimulus in complex biological milieu10, 11, 12, 13, 14, 15. For instance, Amir et al. reported the application of the logic gate concept to engineer a two enzymatic reaction-triggered drug release system. Although the presence of either of the enzymes would be sufficient to open the switch of the “drug releasing gate”, but the “OR” logic would also increase the risk of premature drug release before arriving at disease regions. Inspired by the Boolean logic idea, “AND” logic gate-based systems are favored for sequence-activated drug release in pathological sites with desired specificity17, 18, 19, 20. Wei et al. developed an “AND” logic gate drug delivery system with two orthogonal molecular triggers (acidic pH and reduction) based on dithiodiethanoic acid crosslinked PEO-b-P(MAA-g-Hyd) block copolymer micelles conjugated with adriamycin drug through the hydrazone bonds. Despite promising and effective, subjected to the inherent constraints on material composition and vehicle geometry, the uniqueness of each previously reported “AND” logic gate-based platform invariably necessitates a brand new material design that is generally not synthetically tractable. Therefore, a versatile strategy for the construction of “AND” logic gate-based platform was highly pursued for programmed disease-specific drug delivery with minimized off-target effect.

Cryptobiosis, also known as “secret life”, has been developed in nature by certain bacteria, ciliates and even higher organisms (e.g., tardigrades) as well as seeds of some plants to withstand stressful and often lethal environmental conditions23, 24, 25. This cytoprotective strategy commonly involves the formation of tough biomolecular sheaths on cellular structures to enter a dormant state for survival from outside threats and subsequent revival to proliferate by breaking the protective shells apart when the environment becomes favorable,. Inspired from this fascinating cryptobiotic mechanism, chemically forming a thin (<100 nm) but durable shell on living cells (cell-in-shell structures) have been demonstrated to be effective for enhancing tolerance against harmful threats as well as controlled cellular metabolism which are anticipated to find various applications in biomedical areas including biocatalysis, cytotherapeutics and single-cell biology28, 29, 30. For instance, individual labile mammalian/microbial yeast cell has been evolved into a micrometer-sized “Iron Man” by coating with polyphenol‒metal complexes comprised of natural polyphenol tannic acid (TA) and FeIII for customizable cellular adherence and proliferation as a consequence of programmed shell formation and degradation,. Yet, in contrast to the flourishing research situation for cell-in-shell structures, developing cryptobiotic systems with “AND” logic gate behaviors for sequence-activated drug delivery have been rarely reported.

As a proof-of-concept study, a cryptobiosis inspired assembly of pH/ATP “AND” enzyme logic gate platform was successfully developed here to realize tumor-specific drug delivery. As shown in Scheme 1A, after conjugating chemotherapeutic doxorubicin hydrochloride (DOX∙HCl) molecules on the surface of gelatin nanoparticles (DG NPs) through typical aldimine condensation, conformal TA-FeIII shell was subsequently deposited on the surface of DG NPs (DG@TA-FeIII NPs) for realizing “AND” logic gate-controlled drug delivery in tumor regions. By preventing the contact between DG NPs and enzyme with durable TA-FeIII shells in healthy tissues, the obtained DG@TA-FeIII NPs were expected to exhibit a cryptobiotic behavior with negligible premature drug leakage. Once arrived at tumor regions, the mild acidic pH and/or upregulated adenosine triphosphate (ATP) in the tumor microenvironment would firstly unlock the TA-FeIII shells and then the subsequently exposed inner DG cores could be hydrolysed by overexpressed matrix metalloproteinase 2/9 (MMP 2/9, also known as gelatinase) in solid tumor regions to release cytotoxic DOX (Scheme 1B). In addition to this “AND” logic gate-controlled DOX delivery in tumor, the good photothermal conversion activity of TA-FeIII complex would endow DG@TA-FeIII NPs with synergistic photothermal‒chemo cell-killing ability under near-infrared (NIR) light irradiation. Benefiting from the dynamic nature of polyphenolic-metal coordinative complex, our study was anticipated to present a versatile strategy for developing “AND” logic gate-based systems for disease-specific drug delivery.

Scheme 1

Schematic illustration for (A) the preparation of cryptobiotic DG@TA-FeIII NPs and (B) the underlying mechanism for “AND” logic gate controlled DOX release performance in tumor region.

Materials and methods

Reagents and chemicals

Gelatin type A (from porcine skin, powder, ∼300 bloom) were purchased from Sigma–Aldrich (9000-70-8, St. Louis, MO, USA). Tannin acid (TA, >99%, 1401-55-4) and iron chloride hexahydrate (FeCl3·6H2O, >99%, 10025-77-1) were obtained from Aladdin Industrial Corporation (Shanghai, China). Acetone (67-64-1), absolute ethanol (64-17-5) and glutaraldehyde solution (GTA, 25%, 111-30-8) were received from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Doxorubicin hydrochloride (DOX·HCl, 480453-73-4) was bought from Beijing Huafeng United Technology Co., Ltd. (Beijing, China). All chemicals were used as received without additional purification.

Synthesis of DG NPs

Monodispersed gelatin NPs were firstly developed from a well-reported two-step desolvation process and subsequently crosslinked by GTA molecules. After mixing the DOX·HCl aqueous solution (1 mg/mL, 10 mL) with GTA solution of various feeding mass ratios (1:1, 1:5, 1:10, 1:50, and 1:100) for 2 h at room temperature, the gelatin NPs (10 mg/mL, 10 mL) were then quickly added into the aldehydated DOX suspension and stirred vigorously for further 2 h. The DG NPs could be successfully collected through typical centrifugation/washing cycles.

Fabrication of cryptobiotic DG@TA-FeIII NPs

The cryptobiotic DG@TA-FeIII NPs were developed by subsequently depositing TA and FeIII molecules on the surface of DG NPs to form conformal TA-FeIII shell via polyphenol‒metal coordination chemistry. Typically, 1 mL of TA aqueous solution (10 mg/mL) was quickly added into DG NPs suspension (100 mg, 99 mL) under vigorous stirring for 30 min. After several centrifugation/washing cycles, the collected DG@TA NPs was then resuspended in DI water (99 mL) and followed addition of FeCl3·6H2O solution (1 mL, 10 mg/mL). With magnetic stirring for 30 min at room temperature, the cryptobiotic DG@TA-FeIII NPs could be obtained by standard purification process.

Characterization of DG@TA-FeIII NPs

Hydrodynamic diameters and zeta potential of as-prepared NPs were measured at 25 °C by using a Zetasizer® Nano-Series instrument (Nano ZS90, Malvern Instruments Ltd., Worcestershire, UK). The ultraviolet–visible–near-infrared light (UV–Vis–NIR) absorption spectra of NPs were measured by a U-5100 UV spectrophotometer (Hitachi, Tokyo, Japan) at room temperature. The morphologies of NPs were observed by scanning electron microscope (SEM, Hitachi SU8020, Tokyo, Japan) and transmission electron microscope (TEM, JEM-2100F, JEOL, Tokyo, Japan).

In vitro iron ion and DOX release performance

A typical 1,10-phenanthroline chromogenic method was utilized to test the ATP/pH sensitive disassembly of TA-FeIII shell. In brief, 3 mL of DG@TA-FeIII NPs (1 mg/mL) was packaged in a dialysis bag with a molecular weight cut-off of 14 kDa. After immersed in 27 mL PBS with various concentrations of ATP and pH values, 3 mL of the releasing buffer was taken out and supplemented with the same volume of fresh PBS at predetermined time intervals. After adding hydroxylamine hydrochloride to the taken solution and reacting with 1,10-phenanthroline (1 mg/mL) for 15 min, the released iron ion could be calculated by measuring the absorbance at 510 nm. The released amount of DOX was also determined by incubating the sealed DG@TA-FeIII NPs (1 mg/mL, 3 mL) with different stimuli (ATP, acidic pH, gelatinase).

Photothermal conversion effect of DG@TA-FeIII NPs

Gradient concentrations of DG@TA-FeIII NPs (200 μL) were irradiated for 3 min by a continuous NIR laser (MDL–III–808-2, Changchun New Industries Optoelectronics Technology Co., Ltd., Changchun, China). The solution temperature was monitored and recorded by an IR thermal camera (Ti125, Fluke, Everett, WA, USA). For the photothermal conversion efficiency test of as-prepared DG@TA-FeIII NPs, 2 mL of DG@TA-FeIII NPs were irradiated to reach the plateau of maximum temperature (58 °C), and finally the photothermal conversion efficiency was calculated to be around 36.1% according to the method reported elsewhere.

In vitro biocompatibility evaluation of DG@TA-FeIII NPs

Human Umbilical Vein Endothelial Cells (HUVECs) were used here as a normal cell line to investigate the cytotoxicity of DG@TA-FeIII NPs by a standard MTT assay. For evaluating the hemocompatibility, gradient concentrations of DG@TA-FeIII NPs were incubated with diluted healthy rabbit blood at 37 °C for 3 h. After centrifugation, the absorbance of supernatant at 450 nm was measured to calculate the relative percent hemolysis. The blood mixed with isosmotic normal saline and DI water was used as the negative and positive controls, respectively.

Enhanced cellular uptake of DOX upon NIR laser irradiation

Exponentially growing 4T1 cancer cells were seeded in 24-well culture plates with a density of 1 × 105 cells per well. After proliferation for 24 h, the culture medium was refreshed with 1 mL of RPMI 1640 containing DG@TA-FeIII NPs (100 μg/mL) or free DOX with equivalent concentration of 1.1 μg/mL. The cells were then irradiated by an NIR laser (808 nm, 2 W/cm2) immediately for 0, 3 and 5 min. Following a further incubation for 10 min, the cells were fixed with 4% of paraformaldehyde for 30 min and treated with 0.1% of Triton X-100/PBS (10 min) for enhanced permeabilization. After that, a standard nucleus staining assay was performed by incubating 4T1 cancer cells with 10 μg/mL of 4′,6-diamidino-2-phenylindole (DAPI). Finally, the intracellular uptake of DOX with various treatments after typical rinsing with PBS buffer was imaged by inverted fluorescence microscope.

In vivo tumor inhibition study

All animal experimental protocols were performed in accordance with the Guidelines for Care and Use of Laboratory Animals of Hefei University of Technology and approved by the Institutional Animal Care and Use Committee of Hefei University of Technology, Hefei, China.

Female BALB/c mice were subcutaneously injected with murine 4T1 cancer cells (1 × 106 cells per mouse, 100 μL) and randomly grouped when the tumor volume achieved to ∼80 mm3. The dosage of intravenously administered free DOX·HCl was 2 mg/kg. For DG@TA-FeIII NPs-treated mice, the dosage of DG@TA-FeIII NPs was 50 mg/kg with an equivalent DOX dosage of 0.5 mg/kg (10 mg/mL, 100 μL). Then, tumors were selectively illuminated with/without NIR laser (808 nm, 1 W/cm2) for 10 min. After various treatments, the tumor volume and body weight of mice were recorded every other day as well as the characterizations of typical H&E histological and TUNEL immunofluorescent staining of tumor slices.

Statistical analysis

The data statistical analysis was carried out by using typical Student's t-test model. Data were expressed as mean ± SD, n = 3. Differences at P < 0.01 and P < 0.001 were considered statistically significant and highly significant, respectively.

Results and discussion

After mixing DOX and glutaraldehyde (GTA) molecules with various feeding ratio at room temperature for 2 h, the monodispersed gelatin NPs which developed from a well-reported two-step desolvation method were then quickly added into the aldehydated DOX solution,. Following 2-h intense agitation, the DG NPs could be collected by a typical centrifugation/rinsing process. As shown in Supporting Information Table S1 and Fig. S1, the content of covalently-linked DOX and diameter of obtained DG NPs were elevated as the feeding amount of GTA molecules increased, further confirmed by the gradually deepened pink solution color of DG NPs (inset of Fig. S1). Considering the delivery advantage of NPs with small size, optimized DG NPs with a feeding weight ratio of 1:50 of DOX versus GTA were thus chosen for further study.

Benefitting from the adhesive catechol chemistry,, the polyphenol TA and FeIII ions could be deposited successfully on the surface of DG NPs evidenced by the corresponding solution color change of obtained DG@TA and DG@TA-FeIII NPs (Fig. 1A). In addition, the characteristic absorption peak at 276 nm of TA (assigned to the π-system of the benzene ring) was shifted and splited into two absorption peaks at 235 and 315 nm, further suggesting the formation of TA-FeIII complex. As seen from Fig. 1B, the zeta potential of NPs was reversed from positive DG NPs (+15.9 mV) to negative DG@TA NPs (−13.7 mV) and DG@TA-FeIII NPs (−24.8 mV), demonstrating the good packaging effect of TA-FeIII shell with a slight increase in hydrodynamic diameter distribution (Fig. 1C). As seen from TEM and SEM images (Fig. 1D), the obtained DG@TA-FeIII NPs exhibited monodispersed spherical morphologies with rough surfaces consisted of TA-FeIII complex which were confirmed by the element mapping results (Fig. 1E).

Figure 1

Characterization of as-prepared DG@TA-FeIII NPs. (A) UV–Vis–NIR absorbance spectra (inset: photographs of corresponding NPs aqueous solution); (B) zeta potential and (C) average hydrodynamic diameter of freshly prepared NPs, data are expressed as mean ± SD (n = 3); (D) typical cartoon analogues, TEM and SEM images; (E) element mapping of DG@TA-FeIII NPs.

Due to the protonated hydroxyl groups of TA molecules in acidic conditions, the TA and FeIII-coordinated complex would dynamically transit from tris-, bis-to mono-analogues as evidenced by the stability constants of TA-FeIII are 2.8 × 1017, 3.4 × 109 and 1.5 × 105 at pH 8.0, 5.0 and 2.0, respectively. This pH-dependent and dynamic binding performance urged us to investigate the storage stability of as-prepared DG@TA-FeIII NPs. Due to the hydrophilic and negative charged surface, no significant change in hydrodynamic diameter and absorbance (808 nm) of DG@TA-FeIII NPs could be observed in various mediums including DI water, PBS and RPMI-1640 cell culture medium with a pH value of 7.4 (Fig. 2A and B). In contrast, the absorbance of DG@TA-FeIII NPs solution could be drastically decreased after incubating in acidic mediums (pH 5.0 and 6.0). The underlying mechanism was that the hydroxyl groups of TA molecules would be protonated at acidic conditions and partial of TA-FeIII shells were then destabilized from the NPs’ surface to not only expose the inner DG cores but also crosslink the neighboring NPs which consequently led to rapid aggregation of DG@TA-FeIII NPs (Fig. 2C). Actually, although the increased diameter of DG@TA-FeIII NPs may not be useful for promoting intracellular uptake by tumor cells, this acidic pH-induced unlock and aggregation are supposed to be beneficial for the long-retention and DOX release in mild acidic tumor microenvironment.

Figure 2

Stability and biocompatibility evaluation of DG@TA-FeIII NPs. (A) Average hydrodynamic diameter and (B) relative absorbance at 808 nm of NPs in various medium for different days; (C) photographs and typical TEM images of aqueous NPs solutions with different pH; (D) hemolysis assay. Data are expressed as mean ± SD (n = 3).

Next, the long-term storage stability of DG@TA-FeIII NPs at pH 7.4 motivated us to test their cytotoxicity by standard methylthiazolyldiphenyl-tetrazolium bromide (MTT) and hemolytic procedures. Despite containing chemotherapeutic DOX molecules, the as-prepared DG@TA-FeIII NPs still occupied good biocompatibility to HUVECs (used here as model normal cells) as the relative viability of HUVECs could reach 83.7 ± 3.1% after 24 h incubation with 600 μg/mL of DG@TA-FeIII NPs, suggesting negligible DOX was released during the incubation (Supporting Information Fig. S2). In addition, after adding gradient concentrations of DG@TA-FeIII NPs into freshly diluted rabbit blood for 3 h, all NPs incubated groups exhibited negligible percent hemolysis (less than 3%) which were comparable to negative control (isosmotic normal saline, Fig. 2D). Thus, all of the above data indicated the good biocompatibility of as-prepared DG@TA-FeIII NPs which was essential for further biomedical application.

“AND” logic gate-controlled DOX release

According to our hypothesis, the cryptobiotic DG@TA-FeIII NPs was expected to show negligible premature DOX leakage before arriving at tumor sites and “AND” logic gate-controlled DOX delivery by sequential disassembly of TA-FeIII protective shell and enzymolysis of inner DG cores (Fig. 3A). To approve this “AND” logic gate performance, the pH and ATP-responsive disassembly behaviors of TA-FeIII shell were firstly verified in vitro. By packaging DG@TA-FeIII NPs which were dispersed in PBS with various pH or gradient concentrations of ATP in dialysis bags (MW = 14,000 Da), the concentrations of released iron were determined by a typical 1,10-phenanthroline chromogenic method. As expected, as the stability constants of TA-FeIII decreased when the solution became more acidic, nearly 47.8% and 84.3% of iron could be disassembled from protective TA-FeIII shells for 12 h when DG@TA-FeIII NPs were incubated in acidic PBS with a pH value of 6.0 and 5.0, respectively. In sharp contrast, only 9.8% of iron was released at pH 7.4 (Fig. 3B). Moreover, inspired by the upregulated level of ATP in cancerous tissues and stronger binding affinity of ATP-FeIII than polyphenol-FeIII coordination39, 40, 41, the ATP-triggered disassembly of TA-FeIII shell from DG@TA-FeIII NPs was also investigated. As shown in Fig. 3C, the amount of iron released from DG@TA-FeIII NPs was almost negligible in the absence of ATP, while it increased as not only the prolonged time but also elevated ATP dosage. Particularly, nearly 68.4% and 75.6% of iron could be destabilized from TA-FeIII shells when mixing with 1.0 mg/mL of ATP for 4 and 12 h, respectively. Therefore, the specific and sensitive pH/ATP dependent TA-FeIII shells disassembly was attractive for realizing minimized premature DOX leakage at circulated body fluids and potential burst DOX release at acidic and ATP-upregulated tumor microenvironment.

Figure 3

In vitro “AND” logic gate controlled DOX delivery performance. (A) Schematic illustration of sequential disassembly and enzymolysis process of DG@TA-FeIII NPs; time-dependent release of FeIII from DG@TA-FeIII NPs at various (B) pH and (C) ATP concentration; (D) DOX release performance of DG NPs (3.0 mg/mL, 3.0 mL) with the enzymolysis of gelatinase; cumulative DOX release profiles from DG@TA-FeIII NPs (1 mg/mL, 3.0 mL) under the assistance of various (E) pH and (F) ATP concentration; (G) cumulative DOX release profiles under various conditions. Data are expressed as mean ± SD (n = 3), ∗∗P < 0.01, ∗∗∗P < 0.001.

Motivated by the sensitive pH/ATP responsive disassembly of TA-FeIII shells, DOX release profiles of totally naked DG NPs with or without gelatinase was evaluated subsequently in vitro (Fig. 3D). Upon addition of gelatinase (mimic MMP 2/9 in tumor region) for 12 h, around 82.1% of conjugated DOX could be quickly liberated from DG NPs which is over 6-fold increase versus the system without enzyme, suggesting the efficient enzymolysis of DG NPs could be realized by gelatinase. Hence, the sequential “AND” logic gate-controlled DOX release from cryptobiotic DG@TA-FeIII NPs was then further tested by synergistic TA-FeIII shell disassembly (triggered by acidic pH or ATP) and DG core enzymolysis (sensitive to MMP 2/9). From Fig. 3E and F, although either acidic pH or ATP could unlock the TA-FeIII shells, nearly no conjugated DOX molecules would be released without the assistance of gelatinase. Equally, it seems also impossible for gelatinase to directly cut off DOX molecules from DG@TA-FeIII NPs due to the hinder effect of intact TA-FeIII shells, indicating a significant sequential “AND” logic gate-based drug release profile. Moreover, to mimic the coexistence of acidic pH and upregulated ATP level, the simultaneous disassembly of TA-FeIII shells by ATP and acidic pH would provide more chances for gelatinase to hydrolyze inner DG cores (Fig. 3G), consequently resulting in facilitated cytotoxic DOX release which is beneficial for tumor-targeted drug delivery with minimized adverse effects.

Enhanced cellular uptake of DOX under NIR light irradiation

In light of the well-reported phenomenon that hyperthermia would promote cellular uptake of chemotherapeutic drug by accelerated cell metabolism and membrane fluidity, the concentration-dependent broad UV–Vis–NIR absorption spectra of DG@TA-FeIII NPs solution which dispersed either in DI water or RPMI-1640 cell culture medium not only suggested its good dispersity but also encouraged us to evaluate the photothermal conversion efficiency (Fig. 4A and Supporting Information S3). As expected, DG@TA-FeIII NPs possessed good photothermal conversion capacity derived from their photosensitive TA-FeIII shells with both NPs’ dosage and power density of incident NIR light dependent manners (Fig. 4B and C). For instance, exposing DG@TA-FeIII NPs solution (200 μL) with gradient concentrations of 50, 100, 150, 200, 300 and 400 μg/mL to NIR light (808 nm, 2.0 W/cm2) for 3 min would elevated the solution temperature from 25 to 31.4, 36.2, 39.6, 46.1, 51.4 and 58 °C, respectively, while only 0.8 °C temperature rise was observed for DI water. In addition, the considerable photothermal stability and conversion efficiency (around 36.1%) suggested that DG@TA-FeIII NPs could act as an efficient photothermal agent (Supporting Information Fig. S4).

Figure 4

(A) UV–Vis–NIR absorption spectra and photograph (inset) of gradient concentrations of DG@TA-FeIII NPs aqueous solution; photothermal conversion effect which depended on (B) NPs dosage and (C) power density of incident NIR light; (D) fluorescent images of 4T1 cancer cells with various treatments.

Chemocytotoxic DOX molecules were covalently conjugated to the gelatin NPs through typical aldimine condensation without forming any thermo-sensitive bonds which led to the non-thermo responsive DOX release performance from DG NPs (Supporting Information Fig. S5). Nevertheless, the well-known heat-promoted cellular uptake of drug inspired us to investigate the possibility of accelerated DOX internalization induced by DG@TA-FeIII NPs under NIR laser illumination. As shown in Fig. 4D and Supporting Information Fig. S6, no significant difference on cellular uptake of free DOX (1.1 μg/mL) could be observed with NIR laser irradiation (808 nm, 2.0 W/cm2) for 0, 3 and 5 min. In contrast, a steady increase of DOX fluorescence was found in 4T1 cells treated with 100 μg/mL of DG@TA-FeIII NPs and NIR light illumination. Thus, benefiting from the controlled tumor-targeted NIR laser irradiation, DG@TA-FeIII NPs would be anticipated to exhibit a noteworthy synergistic cancer cell killing effect.

In vivo synergistic tumor inhibition

After evaluating the localized photothermal ablative capacity of DG@TA-FeIII NPs by a well-reported Live/Dead cell staining assay (Supporting Information Fig. S7), the synergistic PTT-chemo cancer cell killing efficiency resulted from DG@TA-FeIII NPs plus NIR light illumination was further evaluated in vivo by using 4T1 tumor xenograft in BALB/c female mice. Randomly grouped mice (n = 8) with average tumor volume of ∼80 mm3 were treated with PBS + NIR laser, Gelatin@TA-FeIII NPs, free DOX, DG@TA-FeIII NPs, Gelatin@TA-FeIII NPs + NIR laser and DG@TA-FeIII NPs + NIR laser, respectively. Upon 10 min of tumor-targeted NIR laser irradiation (808 nm, 1.0 W/cm2), the localized tumor temperature of PBS treated group was only slightly increased to 38.9 °C, demonstrating the safety of incident light. In contrast, owing to the good photothermal conversion ability of TA-FeIII shells, the temperature of tumor treated with Gelatin@TA-FeIII NPs and DG@TA-FeIII NPs could locally achieved to around 50.2 and 50.4 °C, respectively (Fig. 5A and B).

Figure 5

Characterization of synergistic tumor inhibition in vivo from DG@TA-FeIII NPs under the assistance of NIR laser irradiation. (A) IR thermal images and (B) quantized tumor temperature changes; (C) relative tumor volume and (D) photograph of dissected tumor samples as well as tumor weight at 16th day after various treatments as indicated, data are expressed as mean ± SD (n = 5); (E) H&E histological analysis and (F) TUNEL immunofluorescence characterizations of dissected tumor samples with various treatments; ∗P < 0.05, ∗∗P < 0.01.

The tumor volume of mice was monitored every other day to evaluate the therapeutic efficiency. As shown in Fig. 5C, the tumors treated with PBS + NIR laser and Gelatin@TA-FeIII NPs showed fast proliferating behaviors within 16 days, while quadruple free DOX (2.0 mg/kg) was also shown slight tumor inhibition effect due to poor tumor accumulation and the low dosage used here. For the mice treated with Gelatin@TA-FeIII NPs + NIR laser and DG@TA-FeIII NPs, notable tumor inhibition but subsequent recurrence occurred due to insufficient generated heat and dosage of released DOX in this study. Satisfactorily, nearly completely tumor inhibition and no recurrence occurred in the following 16 days were achieved in DG@TA-FeIII NPs + NIR laser group, indicating a synergistic cancer cell killing effect by combined photothermal‒chemo therapy (Fig. 5D and Supporting Information Fig. S8).

Typical H&E histological (Fig. 5E) and TUNEL immunofluorescent staining (Fig. 5F) of tumor slices in various treated groups were further carried out. No significant morphological changes and TUNEL signals (green fluorescence) were observed in PBS + NIR laser and Gelatin@TA-FeIII NPs groups. In comparison to free DOX-treated tumors, more cytoplasm shrinking and TUNEL signals could be found in DG@TA-FeIII NPs group due to the tumor targeted burst release of cytotoxic DOX molecules. As expected, DG@TA-FeIII NPs + NIR laser group present the most serious cell destruction and TUNEL signals in tumor cells owing to the synergistic photothermal‒chemo effect. Moreover, the characterizations of body weight (Supporting Information Fig. S9) and major organ histological analysis (Supporting Information Fig. S10) suggested that no significant side effects would be induced by DG@TA-FeIII NPs as well as NIR laser illumination. Therefore, cryptobiotic DG@TA-FeIII NPs would occupy promising tumor-preferred cytotoxicity due to “AND” logic gate-controlled DOX delivery as well as synergistic photothermal‒chemo effect with the assistance of NIR laser irradiation.

Conclusions

Inspired by the natural cryptobiosis with environment sensitive shells and polyphenol‒metal coordination chemistry, DG@TA-FeIII NPs were successfully developed here as a “AND” logic gate nanoplatform for controlled DOX delivery. Benefiting from the altered tumor microenvironment, DG@TA-FeIII NPs would go through a TA-FeIII shells disassembly by acidic pH/ATP and subsequent enzymolysis of inner DOX-conjugated DG cores to realize tumor-specific DOX release with minimized premature drug leakage in surrounding healthy tissues. Moreover, the photothermal conversion ability of TA-FeIII complex also endowed DG@TA-FeIII NPs good potential for synergistic photothermal‒chemo tumor inhibition. Therefore, our study presented a simple strategy for developing crytobiotic “AND” logic gate nanoplatform with tumor-specific drug delivery by forming polyphenol‒metal complexes on enzyme degradable drug conjugates.