Homotypic Cancer Cell Membranes Camouflaged Nanoparticles for Targeting Drug Delivery and Enhanced Chemo-Photothermal Therapy of Glioma

Glioma is among the deadliest types of brain cancer, for which there currently is no effective treatment. Chemotherapy is mainstay in the treatment of glioma. However, drug tolerance, non-targeting, and poor blood–brain barrier penetrance severely inhibits the efficacy of chemotherapeutics. An improved treatment method is thus urgently needed. Herein, a multifunctional biomimetic nanoplatform was developed by encapsulating graphene quantum dots (GQDs) and doxorubicin (DOX) inside a homotypic cancer cell membrane (CCM) for targeted chemo-photothermal therapy of glioma. The GQDs with stable fluorescence and a superior light-to-heat conversion property were synthesized as photothermal therapeutic agents and co-encapsulated with DOX in CCM. The as-prepared nanoplatform exhibited a high DOX loading efficiency. The cell membrane coating protected drugs from leakage. Upon an external laser stimuli, the membrane could be destroyed, resulting in rapid DOX release. By taking advantage of the homologous targeting of the cancer cell membrane, the GQDs/DOX@CCM were found to actively target tumor cells, resulting in significantly enhanced cellular uptake. Moreover, a superior suppression efficiency of GQDs/DOX@CCM to cancer cells through chemo-photothermal treatment was also observed. The results suggest that this biomimetic nanoplatform holds potential for efficient targeting of drug delivery and synergistic chemo-photothermal therapy of glioma.


Introduction
Glioma is a common primary tumor of the central nervous system [1], accounting for about 45% of intracranial tumors. Glioma has a fast growth rate and a strong invasive ability, and it could infiltrate to the healthy brain tissues by proliferation and invasion, which makes it the most challenging intracranial tumor. Conventional therapy for glioma usually includes surgery, radiotherapy, and chemotherapy [2]. However, the median survival time of glioma patients has so far been only 12 months [3,4]. The current treatment of glioma faces several main obstacles, such as relapse, radiation tolerance, drug tolerance, and the blood-brain barrier [5]. Glioma poses a serious threat to human health and safety. Therefore, it is pivotal to further explore new treatment methods for glioma.
With the development of precision medicine, the nano-sized drug delivery system (NDDS) has become a research hotspot in the treatment of tumors. NDDS can accumulate in tumors via the enhanced permeability and retention (EPR) effects to achieve precise drug release [6][7][8]. However, the passive aggregation effect of nanoparticles is still limited, drug release [6][7][8]. However, the passive aggregation effect of nanoparticles is still limited, depending on the varying degrees of tumor vascularization and the permeability associated with tumor types and development stages [9][10][11]. To exert pharmacological activities, nano-therapeutics must overcome a cascade of physiological challenges, including blood circulation, cell toxicity, immune evasion, cell internalization, tumor targeting, etc. [12][13][14][15]. In recent years, graphene quantum dots (GQDs) nanomaterials have aroused great interest in the scientific community due to their unique physicochemical properties. As a zero micro-nano material, GQDs share some desirable properties of graphene oxide, including water solubility, stable fluorescence, and a variety of attributes, such as size, biocompatibility, excellent photothermal capacity, and the ability to cross the blood-brain barrier [16,17]. These properties make GQDs an excellent drug carrier and could be used in emerging tumor treatments, such as photothermal therapy (PTT) and biological imaging.
Inspired by biological systems in nature, cell membranes are increasingly being used to camouflage nanoparticles, which endow them with properties of natural cells, such as high biocompatibility, low immunogenicity, long circulation, and active targeting [18][19][20]. The camouflage of the red blood cell membrane for nanoparticles can prolong their circulation time in physical environments and help to escape the phagocytosis of the reticuloendothelial system so that drugs could reach tumor site at higher concentrations [21][22][23][24]. Benefiting from the homing properties of stem cells [25,26], the mesenchymal stem cell membrane was wrapped on nanoparticles to bestow them with a targeting ability, enabling NDDS to escape from phagocytosis by the immune system to play a role upon reaching the tumor site [27][28][29]. It was reported that the tumor cell could aggregate to form metastatic cells, which was highly dependent on multimolecular-mediated cell-surface interactions, including tumor-specific binding proteins, such as the Thomsen-Friedenreich antigen and E-cadherin presenting on the membrane of the cancer cells [30,31]. Given the ability of homologous adhesion between tumor cells, it was speculated that the cancer cell membrane (CCM) coating could be applied to target the source cancer cells for anti-cancer drug delivery [32][33][34].
In this study, a novel cancer cell membrane camouflaged NDDS for targeted tumor treatment was developed (Scheme 1). The glioma cell membrane was utilized to camouflage GQDs with anti-cancer drug doxorubicin (DOX). The fabricated GQDs/DOX@CCM were fully characterized. Afterwards, cellular uptake and tumor targeting ability to homotypic glioma cells were investigated. Finally, the chemo-photothermal combination anti-cancer effect of GQDs/DOX@CCM was studied.

Characterization of GQDs
GQDs were synthesized according to the previous report [35]. As illustrated in Figure 1a, 1,3,6-trinitropyrene (TNP) and BPEI were dispersed in deionized water for ultrasonic treatment, and then they were reacted at 230 • C for 5 min. Finally, GQDs were obtained after a series of purification and dialysis. After GQDs were prepared, the morphology and structure were characterized. As shown in Figure 1b, GQDs have a broad peak at 23.9 • , which corresponds to a wider layer (002) spacing of the graphite structure (3.71 Å), indicating the crystallinity of GQDs. The Raman spectrum shown in Figure 1c suggested that the ordered G band at 1617 cm −1 was stronger than the disordered D band at 1376 cm −1 , with a large G to D intensity ratio of 1.2, which was consistent with a previous report [35]. The FTIR spectrum in Figure 1d showed a strong vibration at 1646 cm −1 , ascribed to the C=C bonds, and a strong, rather broad vibration at 3471 cm −1 ascribed to the O-H bonds. The O-H signal was mainly ascribed to the hydroxyl functionalization of the GQDs, which was further confirmed by the XPS. As shown in Figure 1e, GQDs contained three obvious characteristics: C1s at 284.85 eV, N1s at 398.84 eV, and O1s at 531.06 eV, respectively. The C1s spectrum (Figure 1f) exhibited three characteristic peaks corresponding to C=C (284.43 eV), C-N (287.08 eV), and C-OH (287.83 eV). The N 1s spectrum (Figure 1g) showed the binding energy at 398.28 eV and 400.05 eV, corresponding to pyrrolic N and graphitic N, respectively. This result revealed that the fabricated GQDs were nitrogen-doped quantum dots. Besides this, the oxygen-containing species referred to hydroxyl (-OH) (Figure 1h), as confirmed by the C1s and O1s spectra, which further accounted for the good water solubility of GQDs.

Characterization of GQDs
GQDs were synthesized according to the previous report [35]. As illustrated in Figure  1a, 1,3,6-trinitropyrene (TNP) and BPEI were dispersed in deionized water for ultrasonic treatment, and then they were reacted at 230 °C for 5 min. Finally, GQDs were obtained after a series of purification and dialysis. After GQDs were prepared, the morphology and structure were characterized. As shown in Figure 1b, GQDs have a broad peak at 23.9°, which corresponds to a wider layer (002) spacing of the graphite structure (3.71 Å), indicating the crystallinity of GQDs. The Raman spectrum shown in Figure 1c suggested that the ordered G band at 1617 cm −1 was stronger than the disordered D band at 1376 cm −1 , with a large G to D intensity ratio of 1.2, which was consistent with a previous report [35]. The FTIR spectrum in Figure 1d showed a strong vibration at 1646 cm −1 , ascribed to the C=C bonds, and a strong, rather broad vibration at 3471 cm −1 ascribed to the O-H bonds. The O-H signal was mainly ascribed to the hydroxyl functionalization of the GQDs, which was further confirmed by the XPS. As shown in Figure 1e, GQDs contained three obvious characteristics: C1s at 284.85 eV, N1s at 398.84 eV, and O1s at 531.06 eV, respectively. The C1s spectrum (Figure 1f) exhibited three characteristic peaks corresponding to C=C (284.43 eV), C-N (287.08 eV), and C-OH (287.83 eV). The N 1s spectrum (Figure 1g) showed the binding energy at 398.28 eV and 400.05 eV, corresponding to pyrrolic N and graphitic N, respectively. This result revealed that the fabricated GQDs were nitrogendoped quantum dots. Besides this, the oxygen-containing species referred to hydroxyl (-OH) (Figure 1h), as confirmed by the C1s and O1s spectra, which further accounted for the good water solubility of GQDs. As a novel fluorescent probe, the photoluminescence properties of GQDs were investigated. As shown in Figure 2a, the maximum excitation and emission wavelength of GQDs were 490 nm and 520 nm, respectively. The emission peak of GQDs hardly moved, while the excitation wavelength changed from 390 to 490 nm (Figure 2b). The distinguished luminous performance endowed GQDs with a potential application in tumor As a novel fluorescent probe, the photoluminescence properties of GQDs were investigated. As shown in Figure 2a, the maximum excitation and emission wavelength of GQDs were 490 nm and 520 nm, respectively. The emission peak of GQDs hardly moved, while the excitation wavelength changed from 390 to 490 nm ( Figure 2b). The distinguished luminous performance endowed GQDs with a potential application in tumor tracing and diagnosis. Thereafter, the photothermal performance of GQDs was verified. The GQDs solution with a concentration of 500 µg/mL was exposed to the near-infrared (NIR) laser (808 nm, 1.44 W/cm 2 ), and the photothermal curves were recorded. As shown in Figure 2c, the temperature of the GQDs solution rose to 84 • C after 8 min of laser irradiation, while the aqueous solution did not change significantly. This result was further verified and visualized in Figure 2d. Results in Figure 2e,f suggested that the photothermal performance of GQDs exhibited concentration and laser power-dependent manner. After six cycles of laser irradiation, the photothermal conversion capability of GQDs remained stable (Figure 2g). After the photothermal sensitivity of GQDs was characterized, we further explored the photothermal conversion efficiency of GQDs. It was seen from Figure S1a that GQDs have a clear absorption peak in the near-infrared region, which was due to the doping of nitrogen and the presence of oxygen-containing functional groups. The absorbance of GQDs at 808 nm was 0.160. Then, the GQDs solution was exposed to a NIR laser (808 nm, 1.44 W/cm 2 ) for 300 s and cooled naturally. The temperature change in Figure S1b showed that under the irradiation of the 808 nm laser, the temperature of the GQDs solution rose from 25 to 66 • C, and after 600 s of natural cooling, it dropped to 30 • C. The time constant (τs) of GQDs was calculated to 204.66 s ( Figure S1c), and the photothermal conversion efficiency of GQDs was further calculated to be 49.97%. After cell membrane encapsulation, the photothermal ability of GQDs@CCM and GQDs/DOX@CCM showed only slight changes compared with GQDs ( Figure 2h). These results demonstrated the excellent photothermal sensitivity of GQDs and their promising potential for application in the photothermal therapy of tumors.
tracing and diagnosis. Thereafter, the photothermal performance of GQDs was verified. The GQDs solution with a concentration of 500 μg/mL was exposed to the near-infrared (NIR) laser (808 nm, 1.44 W/cm 2 ), and the photothermal curves were recorded. As shown in Figure 2c, the temperature of the GQDs solution rose to 84 °C after 8 min of laser irradiation, while the aqueous solution did not change significantly. This result was further verified and visualized in Figure 2d. Results in Figure 2e,f suggested that the photothermal performance of GQDs exhibited concentration and laser power-dependent manner. After six cycles of laser irradiation, the photothermal conversion capability of GQDs remained stable (Figure 2g). After the photothermal sensitivity of GQDs was characterized, we further explored the photothermal conversion efficiency of GQDs. It was seen from Figure S1a that GQDs have a clear absorption peak in the near-infrared region, which was due to the doping of nitrogen and the presence of oxygen-containing functional groups. The absorbance of GQDs at 808 nm was 0.160. Then, the GQDs solution was exposed to a NIR laser (808 nm, 1.44 W/cm 2 ) for 300 s and cooled naturally. The temperature change in Figure S1b showed that under the irradiation of the 808 nm laser, the temperature of the GQDs solution rose from 25 to 66 °C, and after 600 s of natural cooling, it dropped to 30 °C. The time constant (τs) of GQDs was calculated to 204.66 s ( Figure S1c), and the photothermal conversion efficiency of GQDs was further calculated to be 49.97%. After cell membrane encapsulation, the photothermal ability of GQDs@CCM and GQDs/DOX@CCM showed only slight changes compared with GQDs ( Figure 2h). These results demonstrated the excellent photothermal sensitivity of GQDs and their promising potential for application in the photothermal therapy of tumors.

Preparation and Characterization of GQDs/DOX@CCM
The biomimetic drug delivery nanoplatform GQDs/DOX@CCM was synthesized as follows: GQDs were mixed with the anticancer drug hydrochloride doxorubicin (DOX) for π-π stacking and electrostatic interaction at room temperature for 24 h. Then, the mixture was co-extruded with CCM, which was gained from freshly harvested BV2 cells. After ultracentrifugation, the GQDs/DOX@CCM was obtained and characterized.
The morphology and structure of GQDs and GQDs/DOX@CCM were visualized by transmission electron microscopy (TEM). As a zero-dimensional nanomaterial, GQDs exhibited a clear lattice structure with an ultra-small particle size of 5.8 ± 2 nm (Figure 3a). After being encapsulated in CCM with DOX, GQDs/DOX@CCM showed an increased particle size of approximately 130 ± 10 nm, which was further confirmed by the hydrodynamic particle size measured by DLS (Figure 3b,c). As shown in Figure 3d, the surface charge of the CCM was determined to be −18.76 ± 1.41 mV, which was consistent with previous reports [36,37]. The GQDs/DOX@CCM exhibited a surface potential (−19.8 ± 1.34 mV) similar to that of CCM, suggesting a successful membrane coating. Subsequently, the protein contents of GQDs/DOX@CCM were assessed by SDS-PAGE. Figure 3e indicated that the compositions of membrane proteins were mostly retained on GQDs/DOX@CCM. protein contents of GQDs/DOX@CCM were assessed by SDS-PAGE. Figure 3 that the compositions of membrane proteins were mostly retained on GQDs/D To further verify the integrity of the core-shell particle structure, fluoresc was used to stain the CCM of GQDs@CCM. Figure 3f showed that the emiss GQDs was around 480 nm when excited at 405 nm, and the emission peak o around 600 nm. The characteristic peaks of GQDs and DOX were both GQDs/DOX@CCM, which indicated the successful loading of GQDs and DO ing co-cultured with BV2 cells for 1 h, GQDs@CCM was visualized by CLSM in Figure 3g, the green fluorescent signals from the GQDs and red fluorescen membrane exhibited a high degree of colocalization, indicating the structural GQDs@CCM. Moreover, the long-term stability of GQDs/DOX@CCM was m DLS, which revealed that the GQDs/DOX@CCM was stable at around 33-36 d (Figure 3h). These results demonstrated the successful synthesis of GQDs/D with an intact natural surface membrane protein and long-term stability.  To further verify the integrity of the core-shell particle structure, fluorescent dye DII was used to stain the CCM of GQDs@CCM. Figure 3f showed that the emission peak of GQDs was around 480 nm when excited at 405 nm, and the emission peak of DOX was around 600 nm. The characteristic peaks of GQDs and DOX were both shown in GQDs/DOX@CCM, which indicated the successful loading of GQDs and DOX. After being co-cultured with BV2 cells for 1 h, GQDs@CCM was visualized by CLSM. As shown in Figure 3g, the green fluorescent signals from the GQDs and red fluorescence from the membrane exhibited a high degree of colocalization, indicating the structural integrity of GQDs@CCM. Moreover, the long-term stability of GQDs/DOX@CCM was measured by DLS, which revealed that the GQDs/DOX@CCM was stable at around 33-36 • C within 7 d (Figure 3h). These results demonstrated the successful synthesis of GQDs/DOX@CCM with an intact natural surface membrane protein and long-term stability.

Biocompatibility In Vitro
Considering the importance of the biocompatibility of nanomaterials for clinical applications, we investigated the biosafety of GQDs and GQDs-based membrane camouflaged nanoparticles through blood compatibility and cell cytotoxicity. It was reported that incompatible materials can break the red blood cells, activate and aggregate platelets, and then induce the formation of thrombosis when entering the blood [38]. Therefore, a hemolysis assay is an important index to explore the biocompatibility of materials. The results of blood compatibility (Figure 4a,b) showed that the hemolysis rate was less than 5% after incubating with GQDs@CCM or GQDs/DOX@CCM of different concentrations for 4 h. To further verify the cytotoxicity of GQDs@CCM, cells from different sources were co-cultured with GQDs or GQDs@CCM for 24 h. The CCK-8 result (Figure 4c,d) showed that cell viability in all groups remained above 95%, even at a high concentration of 200 µg/mL. These results suggested the high biosafety and bio-compatibility of GQDs and GQDs@CCM.

Biocompatibility In Vitro
Considering the importance of the biocompatibility of nanomaterials for clinical a plications, we investigated the biosafety of GQDs and GQDs-based membrane camo flaged nanoparticles through blood compatibility and cell cytotoxicity. It was reporte that incompatible materials can break the red blood cells, activate and aggregate platele and then induce the formation of thrombosis when entering the blood [38]. Therefore hemolysis assay is an important index to explore the biocompatibility of materials. T results of blood compatibility (Figure 4a,b) showed that the hemolysis rate was less tha 5% after incubating with GQDs@CCM or GQDs/DOX@CCM of different concentratio for 4 h. To further verify the cytotoxicity of GQDs@CCM, cells from different sources we co-cultured with GQDs or GQDs@CCM for 24 h. The CCK-8 result (Figure 4c,d) showe that cell viability in all groups remained above 95%, even at a high concentration of 2 μg/mL. These results suggested the high biosafety and bio-compatibility of GQDs an GQDs@CCM.

DOX Encapsulation and Releasing
Drug loading efficiency was investigated under the presence of DOX with differe concentrations. It was shown that the DOX encapsulation efficiency was calculated to 89% ± 6.37, 88% ± 6.19, and 74% ± 6.54 when the initial added DOX concentration was 3 μg/mL, 400 μg/mL, and 500 μg/mL, respectively ( Figure 5a). Next, the drug release b havior of GQDs/DOX@CCM was detected at about 33-36 °C. As shown in Figure 5b, t drug release rate in a pH 5.0 solution was about 46% ± 0.4 at 72 h, which was significant higher than that in pH 7.4 (22% ± 2.1), suggesting pH-dependent drug release behavior GQDs/DOX@CCM. Moreover, the release rate of DOX from GQDs/DOX@CCM with N laser irradiation (808 nm, 1.44 W/cm 2 ) was significantly improved compared with the no irradiated group (48% ± 1.2 vs. 78% ± 2.0) (Figure 5c). The laser irradiation induced am high temperature, which helped boosting DOX release. These results suggest the favor ble controlled release behavior of GQDs/DOX@CCM for reducing the side effects of DO to normal tissues.

DOX Encapsulation and Releasing
Drug loading efficiency was investigated under the presence of DOX with different concentrations. It was shown that the DOX encapsulation efficiency was calculated to be 89% ± 6.37, 88% ± 6.19, and 74% ± 6.54 when the initial added DOX concentration was 300 µg/mL, 400 µg/mL, and 500 µg/mL, respectively (Figure 5a). Next, the drug release behavior of GQDs/DOX@CCM was detected at about 33-36 • C. As shown in Figure 5b, the drug release rate in a pH 5.0 solution was about 46% ± 0.4 at 72 h, which was significantly higher than that in pH 7.4 (22% ± 2.1), suggesting pH-dependent drug release behavior of GQDs/DOX@CCM. Moreover, the release rate of DOX from GQDs/DOX@CCM with NIR laser irradiation (808 nm, 1.44 W/cm 2 ) was significantly improved compared with the non-irradiated group (48% ± 1.2 vs. 78% ± 2.0) (Figure 5c). The laser irradiation induced amhigh temperature, which helped boosting DOX release. These results suggest the favorable controlled release behavior of GQDs/DOX@CCM for reducing the side effects of DOX to normal tissues.

In Vitro Homologous Targeting
When cultured with BV2 cells, GQDs tended to locate at the cytoplasm and exhibit stable green fluorescence (Figure 6a). The cellular internalization of GQDs and GQDs@CCM in BV2 cells were investigated by CLSM and FCM. As displayed in Figure  6a, GQDs could be internalized and located in the cytoplasm of cancer cells. However, after being wrapped by the BV2 cell membrane, a stronger green fluorescence was observed in BV2 cells at the same incubation time. This result was further verified by the quantitative analysis of FCM ( Figure S2a-c). After 60 min of co-culturing, the fluorescence intensity of intracellular GQDs@CCM was 1.6 times that of GQDs at the same GQDs concentration of 200 μg/mL (Figure 6b). The enhanced cellular uptake of GQDs@CCM could be attributed to the surface adhesion molecules in the cell membrane, which endowed the nanocomposites with the homotypic tumor self-recognition ability. To further demonstrate the self-recognition ability of cancer cell membrane camouflaged GQDs with homotypic cancer cells, cellular internalization of BV2 cell membrane coated GQDs@CCM by BV2 cells and MCF-7 cells was studied. It was found that the fluorescence intensity originating from BV2 cell membrane coated NPs was far superior in the source BV2 cells than in heterotypic cells MCF-7 after co-culturing for 1 h, which approximated to 2.0-2.3-fold in terms of the mean fluorescence intensity inside cells (Figure  7a,b). These results were also validated by FCM ( Figure S2d-f). Besides this, a more rapid

In Vitro Homologous Targeting
When cultured with BV2 cells, GQDs tended to locate at the cytoplasm and exhibit stable green fluorescence (Figure 6a). The cellular internalization of GQDs and GQDs@CCM in BV2 cells were investigated by CLSM and FCM. As displayed in Figure 6a, GQDs could be internalized and located in the cytoplasm of cancer cells. However, after being wrapped by the BV2 cell membrane, a stronger green fluorescence was observed in BV2 cells at the same incubation time. This result was further verified by the quantitative analysis of FCM ( Figure S2a-c). After 60 min of co-culturing, the fluorescence intensity of intracellular GQDs@CCM was 1.6 times that of GQDs at the same GQDs concentration of 200 µg/mL (Figure 6b). The enhanced cellular uptake of GQDs@CCM could be attributed to the surface adhesion molecules in the cell membrane, which endowed the nanocomposites with the homotypic tumor self-recognition ability.

In Vitro Homologous Targeting
When cultured with BV2 cells, GQDs tended to locate at the cytoplasm and exhibit stable green fluorescence (Figure 6a). The cellular internalization of GQDs and GQDs@CCM in BV2 cells were investigated by CLSM and FCM. As displayed in Figure  6a, GQDs could be internalized and located in the cytoplasm of cancer cells. However, after being wrapped by the BV2 cell membrane, a stronger green fluorescence was observed in BV2 cells at the same incubation time. This result was further verified by the quantitative analysis of FCM ( Figure S2a-c). After 60 min of co-culturing, the fluorescence intensity of intracellular GQDs@CCM was 1.6 times that of GQDs at the same GQDs concentration of 200 μg/mL (Figure 6b). The enhanced cellular uptake of GQDs@CCM could be attributed to the surface adhesion molecules in the cell membrane, which endowed the nanocomposites with the homotypic tumor self-recognition ability. To further demonstrate the self-recognition ability of cancer cell membrane camouflaged GQDs with homotypic cancer cells, cellular internalization of BV2 cell membrane coated GQDs@CCM by BV2 cells and MCF-7 cells was studied. It was found that the fluorescence intensity originating from BV2 cell membrane coated NPs was far superior in the source BV2 cells than in heterotypic cells MCF-7 after co-culturing for 1 h, which approximated to 2.0-2.3-fold in terms of the mean fluorescence intensity inside cells (Figure  7a,b). These results were also validated by FCM ( Figure S2d-f). Besides this, a more rapid To further demonstrate the self-recognition ability of cancer cell membrane camouflaged GQDs with homotypic cancer cells, cellular internalization of BV2 cell membrane coated GQDs@CCM by BV2 cells and MCF-7 cells was studied. It was found that the fluorescence intensity originating from BV2 cell membrane coated NPs was far superior in the source BV2 cells than in heterotypic cells MCF-7 after co-culturing for 1 h, which approximated to 2.0-2.3-fold in terms of the mean fluorescence intensity inside cells (Figure 7a,b). These results were also validated by FCM ( Figure S2d-f). Besides this, a more rapid cell internalization by the source cell membrane coated GQDs@CCM was found when the incu- cell internalization by the source cell membrane coated GQDs@CCM was found when the incubation period was further prolonged. These results proved the high specific selfrecognition affinity of GQDs@CCM to the source tumor cells.

Antitumor Efficacy In Vitro
Encouraged by the aforementioned results, we further investigated the anticancer effect of GQDs/DOX@CCM in vitro. Then GQDs were co-cultured with BV2 cells and subjected to different laser treatments. The cell survival situation was studied by using Calcein-AM and PI as probes to indicate the live (green) and dead (red) cells, respectively (Figure 8a). The result showed that GQDs treatment did not cause obvious cell death. However, after laser treatment, the red fluorescence of dead cells increased with the increase of the GQDs concentration and irradiation time, indicating their excellent photothermal performance in cancer treatment. Subsequently, the antitumor activity of GQDs/DOX@CCM by the combination of chemo-phototherapy in vitro was investigated. BV2 cells were treated with free DOX and GQDs/DOX@CCM for 24 h, and cell viability was evaluated through a CCK-8 assay (Figure 8b). The reduction in BV2 cell viability was dose-dependent in all groups. Both GQDs (under NIR) and free DOX could effectively inhibit cell proliferation. After being co-encapsulated in CCM, there was a slight reduction in the cytotoxicity of GQDs/DOX@CCM, which was mainly caused by the relatively long drug release time in cancer cells. However, after being irradiated by an external laser, less than 10% of cells survived. The significantly enhanced toxicity to cancer cells could be attributed to the cooperation of the homologous targeted delivery, superior photothermal effect, and controlled drug release responding to the temperature increment. These results were further confirmed by Calcein-AM/PI staining, which was consist with the viability assay (Figure 8c). The above results demonstrated that GQDs/DOX@CCM possess superior therapeutic performance in homogeneous cancer cells. However, further in vivo experiments are still needed before making a final statement on their antitumor effect.

Antitumor Efficacy In Vitro
Encouraged by the aforementioned results, we further investigated the anticancer effect of GQDs/DOX@CCM in vitro. Then GQDs were co-cultured with BV2 cells and subjected to different laser treatments. The cell survival situation was studied by using Calcein-AM and PI as probes to indicate the live (green) and dead (red) cells, respectively (Figure 8a). The result showed that GQDs treatment did not cause obvious cell death. However, after laser treatment, the red fluorescence of dead cells increased with the increase of the GQDs concentration and irradiation time, indicating their excellent photothermal performance in cancer treatment. Subsequently, the antitumor activity of GQDs/DOX@CCM by the combination of chemo-phototherapy in vitro was investigated. BV2 cells were treated with free DOX and GQDs/DOX@CCM for 24 h, and cell viability was evaluated through a CCK-8 assay (Figure 8b). The reduction in BV2 cell viability was dose-dependent in all groups. Both GQDs (under NIR) and free DOX could effectively inhibit cell proliferation. After being co-encapsulated in CCM, there was a slight reduction in the cytotoxicity of GQDs/DOX@CCM, which was mainly caused by the relatively long drug release time in cancer cells. However, after being irradiated by an external laser, less than 10% of cells survived. The significantly enhanced toxicity to cancer cells could be attributed to the cooperation of the homologous targeted delivery, superior photothermal effect, and controlled drug release responding to the temperature increment. These results were further confirmed by Calcein-AM/PI staining, which was consist with the viability assay (Figure 8c). The above results demonstrated that GQDs/DOX@CCM possess superior therapeutic performance in homogeneous cancer cells. However, further in vivo experiments are still needed before making a final statement on their antitumor effect.

Cell Culture
Mouse glioma cells (BV2), human cervical carcinoma cells (HeLa), human b cer cells (MCF-7), human normal liver cells (LO2), human embryonic kidney ce rat microglia (GMI-R1), and rat astrocytes cells were obtained from the Chinese of sciences. All cells were cultured in a DMEM medium supplemented with 1% p streptomycin and 10% fetal bovine serum at 37°C in a humidified atmosphere o midity and containing 5% CO2.

Isolation of Cancer Cell Membrane
When BV2 cells grew to 90% confluence in culture flasks, cells were coll suspended in a hypotonic cell lysis buffer, placed in 4 °C for 12 h, then disrupt 130 homogenizer (Beijing, China) for 20 passes. The cell suspensions were cent 3000 rpm for 20 min at 4 °C; the resulting supernatant was collected and centrifu at 100,000× g for 30 min. The plasma membrane pellet was resuspended in PBS w and stored in −80 °C for further experiments.

Cell Culture
Mouse glioma cells (BV2), human cervical carcinoma cells (HeLa), human breast cancer cells (MCF-7), human normal liver cells (LO2), human embryonic kidney cells (293T), rat microglia (GMI-R1), and rat astrocytes cells were obtained from the Chinese academy of sciences. All cells were cultured in a DMEM medium supplemented with 1% penicillin-streptomycin and 10% fetal bovine serum at 37 • C in a humidified atmosphere of 90% humidity and containing 5% CO 2 .

Isolation of Cancer Cell Membrane
When BV2 cells grew to 90% confluence in culture flasks, cells were collected and suspended in a hypotonic cell lysis buffer, placed in 4 • C for 12 h, then disrupted by a D-130 homogenizer (Beijing, China) for 20 passes. The cell suspensions were centrifuged at 3000 rpm for 20 min at 4 • C; the resulting supernatant was collected and centrifuged again at 100,000× g for 30 min. The plasma membrane pellet was resuspended in PBS with PMSF and stored in −80 • C for further experiments.

Preparation of GQDs and GQDs/DOX@CCM
According to the previous report [39], the 1,3,6-trinitropyrene (TNP) was firstly prepared by the nitrification reaction of pyrene. Then, TNP and BPEI were dispersed in deionized water for ultrasonic treatment and placed in a dedicated reactor at 230 • C for 5 min. Finally, the pure GQDs were obtained after filtration and dialysis.
A total of 1 mL of DOX (800 µg/mL) was added to 1 mL of GQDs (400 µg/mL) at room temperature in the dark for 24 h. The mixture was co-extruded with 2 mL of BV2 cancer cell membrane vesicles through a 200 nm polycarbonate membrane for 20 passes. The unloaded DOX and GQDs were removed by centrifugation at 13,000× g for 10 min.

Characterization
Transmission electron microscopy (TEM) images were carried out by a G2F20 microscope (Thermo Fisher Scientific, Wyman Street, Waltham, MA, USA). Raman spectra were recorded on a horiba evolution laser Raman spectrometer (Horiba Scientific, Piscataway, NJ, USA). X-ray photoelectron spectroscopy (XPS) measurements were performed using a K-Alpha + X-ray photoelectron spectrometer (Thermo Fisher Scientific, Wyman Street, Waltham, MA, USA). X-ray diffraction (XRD) patterns were obtained with a D8 Advance X-ray diffractometer (Showa, Tokyo, Japan). The fluorescence spectrum was recorded with an LS55 fluorophotometer (PerkinElmer Inc., Waltham, MA, USA). UV-Vis-NIR spectroscopy was recorded on a U-4150 Ultraviolet visible near infrared spectrophotometer (Hitachi, Tokyo, Japan). The FTIR spectrum was recorded on the VERTEX70 Fourier infrared spectrometer (Bruker, Karlsruhe, Germany). Particle size and surface zeta potential were measured by dynamic light scattering (DLS) using Zetasizer equipment (Zetasizer Nano ZN, Malvern Analytical Ltd., Malvern, UK). The protein components were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).

Photothermal Performance Measurements
GQDs with different concentrations were exposed to a near-infrared laser (808 nm) for 5 min with different power densities; the temperature of the solution was measured and recorded with an E53 infrared thermal imager (FLIR Systems Inc, Portland, OR, USA). In the photothermal stability experiment, the solution of GQDs was radiated with the near-infrared laser (808 nm, 1.44 W/cm 2 ) for 5 min, and then the solution was cooled to room temperature naturally. After being repeated 6 times, the temperature was recorded, and the temperature curve was drawn. Different concentrations of GQDs were co-cultured with BV2 cells for 4 h and then irradiated by a near-infrared laser (808 nm, 1.44 W/cm 2 ) for 3 or 5 min. Cell mortality was measured through the Calcein-AM/PI Double Staining Kit.

Photothermal Conversion Efficiency Measurement
The photothermal conversion efficiency of GQDs was investigated by irradiating the centrifuge tube containing a dispersion of GQDs at 300 µg/mL (808 nm, 1.44 W/cm 2 ). The temperature of the GQDs solution was measured every 25 s by an infrared thermal imager (FLIR Systems Inc, Portland, OR, USA). Q Dis can be determined by the following Equation [40,41]: where Q Dis represents the baseline energy input of water, m D is the mass of aqueous solution, c D is the specific heat capacity of aqueous solution, ∆ T is the temperature change of aqueous solution during the 5 min laser irradiation (808 nm, 1.44 W/cm 2 ), and t is the irradiation time (5 min). h represents heat-transfer coefficient, A represents the surface area of the container, and hA can be determined by the following equation: Here, θ is a dimensionless parameter, and τ s is the time constant of the sample system. c D and m D represent the mass of GQDs (0.5 g) and heat capacity (4.2 J/g), T max represents the maximum temperature of GQDs solution, and T surr represents the ambient temperature.
Photothermal conversion efficiency (η) can be determined by the following equation: I is the laser power, and Aλ is the absorption value at 808 nm.

Biocompatibility In Vitro
The blood of healthy Kunming mice was collected from the eyeball and then centrifuged several times to obtain red blood cells and suspended in PBS. A total of 0.2 mL of erythrocytes were mixed with I 0.2 mL PBS (negative control), II 0.2 mL of deionized water (positive control), and III 0.2 mL of GQDs, GQDs@CCM, or GQDs/DOX@CCM solutions of different concentrations. The mixtures were placed at room temperature (33-36 • C) for 4 h. Then, the supernatants were collected and measured at 540 nm to calculate the hemolysis rate after centrifugation at 10,000× g for 5 min.
The cell viability was investigated by Cell Counting Kit (CCK)-8 assays. BV2, MCF-7, HeLa, LO2, 293T, GMI-R1, and rat astrocyte cells were seeded into 96-well plates (n = 5) at a density of 4000 cells per well. After adhering, the medium was taken out, and 100 µL of fresh medium containing GQDs or GQDs@CCM with various GQDs concentrations (10, 20, 50, 100, and 200 µg/mL) were added and co-cultured for another 24 h. Then, 10 µL of the CCK-8 solution was added to each well and incubated for 2 h. Finally, the absorbance at 450 nm was detected by a VICTOR X3 Multifunctional microplate reader (PerkinElmer Inc., Waltham, MA, USA).

Drug Encapsulation Efficiency and Drug Release In Vitro
To assess the drug loading capacity, GQDs/DOX@CCM was centrifuged to remove the unloaded DOX. The unloaded DOX was determined by measuring absorption at 480 nm using a Multifunctional microplate reader (Perkin Elmer, Waltham, MA, USA). The drug encapsulation efficiency was calculated using the following formula: Drug encapsulation efficiency (%) = Initial mass of drug − Mass of unload drug Initial mass of drug × 100% To measure the profile of drug release, 1.5 mL of GQDs/DOX@CCM (DOX concentration of 300 µg/mL) were dialyzed in 8 mL PBS with a different pH (7.4 or 5.0) at about 33-36 • C in the dark (MWCO = 3.5KD). At different time intervals, 500 µL of dialysis buffer was withdrawn to detect the amount of DOX released. After the measurement, the solution was poured back to maintain the constant total amount.

Cellular Uptake
BV2 cells were seeded into a 35 mm confocal dish (n = 3) at a density of 4000 cells per well and cultured for 24 h. Then, cells were cultured with GQDs or GQDs@CCM for 10 min to 60 min. Cells were washed three times and treated with paraformaldehyde for 20 min. After that, DAPI was added to label the nuclei for 5 min. Then, cells were washed with PBS and visualized by confocal laser scanning microscopy (CLSM) (Occult International Ltd., DE). For quantitative analysis, BV2 cells were seeded into 12-well plates (n = 3) at a density of 40,000 cells per well and cultured for 24 h; then, GQDs or GQDs@CCM were added with a final GQDs concentration of 200 µg/mL for 10 to 60 min; after the cells were washed and digested, the fluorescence of GQDs and GQDs@CCM were observed with a Flow cytometer (Beckman Coulter, Inc., S. Kraemer Boulevard Brea, CA, USA) at the PB450A channel.

Homologous Recognition In Vitro
BV2 cells and MCF-7 cells were seeded into a 35 mm confocal dish (n = 3) at a density of 4000 cells per well and cultured for 24 h. Then, cells were incubated with GQDs@CCM at a final GQDs concentration of 200 µg/mL from 10 to 60 min. After the cells were washed and fixed with paraformaldehyde for 20 min, the cells were stained with DAPI for 5 min. The internalization of GQDs or GQDs@CCM by cells was observed via CLSM. Simultaneously, flow cytometry (FCM) was also used to evaluate the specific targeting of cells. BV2 cells and MCF-7 cells were seeded into 12-well plates (n = 3) at a density of 40,000 cells per well and cultured for 24 h. Then, cells were incubated with GQDs@CCM from 10 to 60 min. After the cells were washed and digested, the fluorescence of BV2 and MCF-7 cells were observed with a Flow cytometer (Beckman Coulter, Inc., S.Kraemer Boulevard Brea, CA, USA) at the PB450A channel.

Anti-Tumor Effect In Vitro
BV2 cells were seeded into 96-well plates (n = 5) at a density of 4000 cells per well and cultured for 24 h. The cell culture medium was replaced with 100 µL of fresh medium containing an increased concentration of DOX and GQDs/DOX@CCM complexes, respectively. The treatment concentration of GQDs was 100 µg/mL in the GQDs/DOX@CCM group. After co-cultivation for 24 h, the cells were washed twice with PBS, and then they were incubated with 10 µL of CCK-8 solution for another 2 h. The absorbance at 450 nm was detected using a Multifunctional plate reader ((PerkinElmer Inc., Waltham, MA, USA). The PTT groups were exposed to an 808 nm laser with a power density of 1.44 W/cm 2 for 3 min after co-cultivation for 4 h, with other conditions kept the same. The formula for calculating the cell viability (%) was listed as follows: Note: A is the OD 450 of the treated BV2 cells; A 0 is the OD 450 of the untreated BV2 cells.
For the Calcein-AM/PI double-staining assay, BV2 cells were seeded in 35 mm confocal dishes (n = 3) at a density of 40,000 cells per well and co-cultivated with free DOX, GQDs, GQDs@CCM, and GQDs/DOX@CCM for 4 h. The concentration of GQDs was kept consistent at 100 µg/mL. Then, cells were washed three times with PBS and subjected to the light irradiation (808 nm, 1.44 W/cm 2 ) for 3 min. After being washed by PBS, cells were stained with Calcein-AM and PI for 15 min at 37 • C. Finally, cells were visualized by CLSM.

Conclusions
In summary, a cancer cell membrane biomimetic drug delivery system for chemophotothermal combination therapy of homogeneous cancer cells was developed. The synthesized GQDs with favorable properties of excellent dispersibility, controllable size, stable fluorescence, and superior photothermal performance was utilized as a therapeutic agent and co-encapsulated with DOX in a BV2 cell membrane. The GQDs were not only capable of converting near-infrared light to heat for photothermal therapy, but they also improved the release of DOX for chemotherapy. Besides this, benefiting from the homologous targeting of the cancer cell membrane, the GQDs/DOX@CCM can actively target BV2 cells in vitro, resulting in a higher cellular uptake. The antitumor results demonstrated the superior killing efficiency of GQDs/DOX@CCM to cancer cells through chemo-photothermal treatment. Based on this study, the fabricated GQDs/DOX@CCM are capable of providing an effective combination strategy for precision oncology therapy.
Supplementary Materials: The following supporting information can be downloaded at https:// www.mdpi.com/article/10.3390/ph15020157/s1: Figure S1: characterization of GQDs. (a) UV-Vis-NIR spectrum of GQDs. (b) Photothermal effect of the GQDs solution (300 µg/mL) exposed to the NIR laser (808 nm, 1.44 W/cm 2 ). The lasers were shut off after 300 s irradiation. (c) Plot of cooling time versus negative natural logarithm of the temperature driving force obtained from the cooling period after the NIR irradiation (808 nm, 1.44 W/cm 2 ). Figure   Institutional Review Board Statement: The animal study protocol was approved by the Ethics Committee of Shanghai University (protocol code ECSHU-2020-030).