Development of Biodegradable GQDs-hMSNs for Fluorescence Imaging and Dual Cancer Treatment via Photodynamic Therapy and Drug Delivery

Recently, nano-based cancer therapeutics have been researched and developed, with some nanomaterials showing anticancer properties. When it comes to cancer treatment, graphene quantum dots (GQDs) contain the ability to generate 1O2, a reactive oxidative species (ROS), allowing for the synergistic imaging and photodynamic therapy (PDT) of cancer. However, due to their small particle size, GQDs struggle to remain in the target area for long periods of time in addition to being poor drug carriers. To address this limitation of GQDs, hollow mesoporous silica nanoparticles (hMSNs) have been extensively researched for drug delivery applications. This project investigates the utilization and combination of biomass-derived GQDs and Stöber silica hMSNs to make graphene quantum dots-hollow mesoporous silica nanoparticles (GQDs-hMSNs) for fluorescent imaging and dual treatment of cancer via drug delivery and photodynamic therapy (PDT). Although the addition of hMSNs made the newly synthesized nanoparticles slightly more toxic at higher concentrations, the GQDs-hMSNs displayed excellent drug delivery using fluorescein (FITC) as a mock drug, and PDT treatment by using the GQDs as a photosensitizer (PS). Additionally, the GQDs retained their fluorescence through the surface binding to hMSNs, allowing them to still be used for cell-labeling applications.


Introduction
Cancer is one of the leading causes of death around the world [1] and since its discovery researchers have tried multiple means of treatment to find a cure for the deadly disease. Current methods of cancer treatment such as chemotherapy and radiology, although common and effective, are aggressive to the host and as a result cause damage and stress to the patient under treatment. Physical symptoms of fatigue, hair loss, neutropenia, lymphedema, deep vein thrombosis, and psychological symptoms such as depression and memory loss are a short list of side effects brought on by traditional cancer treatment methods [2][3][4][5]. Therefore, several disciplines such as medicine, biology, and chemistry have begun to research alternative means of cancer treatment with less severe side effects for a better quality of life [6][7][8]. Nanoscience is one such discipline that has extensively researched the use of nanoparticles for cancer treatment [9]. Nanoparticles contain several unique properties that are being exploited in the hope of finding new treatment methods that are not as taxing on the patient.
One class of nanoparticles being explored are graphene quantum dots (GQDs), which are a subset of nanoparticles defined by their size range of 2-10 nm. These particles contain several key characteristics including biocompatibility, low toxicity, resistance to photobleaching, high photostability, and photoluminescence that make them applicable in biological applications [10,11]. One vital feature of GQDs is their fluorescence capabilities, A variety of analytical instrumentation were used to characterize the new GQDs-hMSNs. Firstly, their size distribution was examined using transmission electron microscopy (TEM) and dynamic light scattering (DLS) due to nanoparticles being defined by their size range of one dimension being 1-100 nm [47] and previous studies showing that nanoparticle size range greatly impacts their cell interactions [48,49]. Elemental composition analysis was determined using X-ray photoelectron spectroscopy (XPS) to confirm the presence of silicon, oxygen, and carbon along with Fourier transform infrared spectroscopy (FT-IR) to confirm bond formations. Zeta-potential was determined due to solution stability and surface charge playing vital roles in cellular interactions, as charged particles typically are better ingested by cells than noncharged particles [50,51]. Then, their ultraviolet-visible (UV-Vis) absorption and fluorescence features were collected to facilitate fluorescence imaging. Their biodegradability was also tested using various pH solutions [32] to mimic biological environments. Lastly, utility was assessed for in vitro fluorescence cell imaging using confocal microscopy, ROS formation using cell viability tests for type II PDT treatment due to 1 O 2 formation via GQDs [52,53] and drug delivery by examining loading and releasing capabilities into the targeted cancer cells using fluorescein (FITC) dye as a mock drug [54].

Design of GQDs-hMSNs and Synthesis
The primary focus of this study is to derive nanoparticles from biomass that help support their biocompatibility and biodegradation. Using cis-3,4-di(furan-2-yl)cyclobutene-1,2-dicarboxylic acid (CBDA-2), a biomass-derived molecule, for synthesizing the GQDs yields synthetically green and environmentally friendly graphene-based particles. Additionally, with CBDA-2 being synthesized from agricultural waste provides the means to recycle waste products into potentially vital nanoparticles for cancer imaging and treatment. Furthermore, the nanoparticles need to be biodegradable as to avoid accumulation in the patient body that could eventually lead to toxicity. Stöber method-based silica nanoparticles have previously been shown to possess excellent biocompatibility [55][56][57] and, with their pH sensitivity, are prone to biodegradation [57][58][59] in a matter of days or weeks. These key features make them valuable to avoid host accumulation and toxicity over time as the human body will naturally break down the silica for use in biological pathways.
To start, cetrimonium chloride (CTAC) and triethylamine (TEA) were mixed together before a Stöber silica suspension was spiked in. The solution was allowed to react before tetraethyl orthosilicate (TEOS) was added. After heating for 12 h on reflux at 60 • C, (3-aminopropyl)triethoxysilane (APTES) was added to introduce amine functional groups to the particle surface to help in biocompatibility, low pH stability, and agglomeration, and the solution was allowed to react for another 3 h while stirring at 60 • C. This resulted in mesoporous silica nanoparticles (MSNs). Na 2 CO 3 was added, and the solution was stirred for another 3 h at 50 • C before being washed with a solution of hydrochloric acid/ethanol (HCl/EtOH) three times. EtOH was added to resuspend the particles and allowed to boil for 2-3 h at 80 • C. These steps etched out the silica core of the MSNs to make hMSNs. Once the EtOH evaporated, the hMSNs were dried to a solid. 1.0 mg was dissolved in 1.0 mL of deionized-water (DI-water) and 0.2 mg GQDs were added before the solution was diluted to 5.0 mL total volume with 20 mM 2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid (HEPES) pH 7.0 buffer. The solution was vortexed for 3 h at 1000 rpm before being centrifuged and dried to purify the final product of GQDs-hMSNs (Scheme 1).

Characterization of GQDs-hMSNs 2.2.1. Size Distribution and Surface Morphology
Particle size is essential to identify for nanoparticles as it is the determining factor as to whether or not they can be classified as nanoparticles. It is also important to determine particle size for biological applications as larger nanoparticles can struggle to diffuse across the cell membrane for imaging or drug delivery. Smaller particles tend to diffuse quicker and easier across the cell membrane; however, they are more common to cause cell toxicity [60]. Therefore, two different analyses were conducted to determine the size distribution of the GQDs-hMSNs. TEM images showed particles ranging from 20-50 nm ( Figure 1A). The larger particles were more difficult to differentiate between single particles and smaller clusters. Thus, DLS was also used to determine particle size, averaging 47.0 ± 19.0 nm ( Figure 1B). Scheme 1. Synthesis of GQDs-hMSNs using Stöber silica suspension and previously synthesized GQDs from biomass-derived material. Surface charge is also essential for nanoparticle determination as the charge will impact cellular uptake. Since uncharged particles are more difficult for cells to ingest, it is important to ensure and analyze the surface charge of the GQDs-hMSNs. Therefore, zeta potential measurements were taken at various pHs (1.0-11.0) and depict a decreasing potential from +5.71 to −21.8 mV (Figure 2). The most agglomeration and solution instability occurred in the more acidic pHs of maleate and citrate (1.0-5.0) but increased in stability Surface charge is also essential for nanoparticle determination as the charge will impact cellular uptake. Since uncharged particles are more difficult for cells to ingest, it is important to ensure and analyze the surface charge of the GQDs-hMSNs. Therefore, zeta potential measurements were taken at various pHs (1.0-11.0) and depict a decreasing potential from +5.71 to −21.8 mV (Figure 2). The most agglomeration and solution instability occurred in the more acidic pHs of maleate and citrate (1.0-5.0) but increased in stability from pH 7.0-11.0 (HEPES, 2-(cyclohexylamino)ethanesulfonic acid (CHES), and 3-(cyclohexylamine)-1-propanesulfonic acid (CAPS), indicating an overall negative surface charge. Lower agglomeration and greater solution stability in neutral pHs benefit biological applications. Additionally, higher agglomeration was observed for prolonged sonication times, greater than 30 min, and at concentrations higher than 50 µg/mL. Surface charge is also essential for nanoparticle determination as the charge wi pact cellular uptake. Since uncharged particles are more difficult for cells to ingest important to ensure and analyze the surface charge of the GQDs-hMSNs. Therefore potential measurements were taken at various pHs (1.0-11.0) and depict a decreasin tential from +5.71 to −21.8 mV (Figure 2). The most agglomeration and solution insta occurred in the more acidic pHs of maleate and citrate (1.0-5.0) but increased in sta from pH 7.0-11.0 (HEPES, 2-(cyclohexylamino)ethanesulfonic acid (CHES), and 3-(c hexylamine)-1-propanesulfonic acid (CAPS), indicating an overall negative su charge. Lower agglomeration and greater solution stability in neutral pHs benefit bi ical applications. Additionally, higher agglomeration was observed for prolonged cation times, greater than 30 min, and at concentrations higher than 50 µg/mL.

Functional Group Formation and Elemental Composition
Functional groups are essential for biological applications and additionally provide insight into the surface composition of the newly synthesized GQDs-hMSNs. FT-IR spectra were collected for hMSNs, GQDs, and GQDs-hMSNs to compare the differences of functional groups and additional bond formation after the synthesis of the nanoparticles and quantum dots together ( Figure 3). For hMSNs without additional GQDs, functional groups were determined to be Si-O-H stretching (3222 cm −1 , Figure 3A(a)), Si-O-Si bending (1058 cm −1 , Figure 3A(b)), Si-OH vibration (944 cm −1 , Figure 3A(c)), and Si-O vibration (796 cm −1 , Figure 3A(d)) [61][62][63]. GQDs functional groups have been identified as -OH (3114 cm −1 , Figure 3B(e)), C=O (1668 cm −1 , Figure 3B(f)), and C-O/C-N (1384 cm −1 , Figure 3B (g)). The spectrum of GQDs-hMSNs contained both bonds found in hMSNs and GQDs, indicating the compilation of the two nanoparticle species, along with an additional C-H bond formation (2821 cm −1 , Figure 3C(h)). Furthermore, the -OH and C-O peaks are more distinct than what is observed in the GQDs spectrum, along with a decrease in C=O. This indicates the breaking of C=O bonds and the formation of Si-O-C bonds and potential C−Si bonds between the GQDs and hMSNs; which signifies that the GQDs are chemically bonding to the surface of the hMSNs rather than being absorbed into the pores [64,65]. This is beneficial as the GQDs are less likely to escape the hMSNs before being delivered into the cancer cells for imaging and PDT treatment in addition to not interfering or interacting with the mock loading drug for delivery. tinct than what is observed in the GQDs spectrum, along with a decrease in C=O. This indicates the breaking of C=O bonds and the formation of Si-O-C bonds and potential C−Si bonds between the GQDs and hMSNs; which signifies that the GQDs are chemically bonding to the surface of the hMSNs rather than being absorbed into the pores [64,65]. This is beneficial as the GQDs are less likely to escape the hMSNs before being delivered into the cancer cells for imaging and PDT treatment in addition to not interfering or interacting with the mock loading drug for delivery. XPS analysis was also conducted to confirm the elemental composition and bond formation of the GQDs-hMSNs seen in the FT-IR spectra and ensure that silicon, oxygen, and carbon were all present in the new synthesized particles. These three primary elements were identified in addition to the presence of nitrogen. This nitrogen content was attributed to APTES bonded to the hMSNs surface [66]. The elements of silicon and oxygen were attributed to be from the hMSNs, carbon and oxygen from the GQDs, and nitrogen from APTES. High-resolution scans of the carbon, nitrogen, and silicon peaks of the GQDs-hMSNs showed C-O/C-N and C-C/C-Si bonds within the carbon peak ( Figure 4A), NSi 2 O and NSi 2 O x in the nitrogen peak ( Figure 4B), and the silicon showed SiO and SiO x bonds ( Figure 4C). The presence of the C-Si bond supported the spectrum seen in FT-IR and confirmed the chemical binding of GQDs to the hMSNs surface, and further explained the slight peak shift in the excitation wavelength. from APTES. High-resolution scans of the carbon, nitrogen, and silicon peaks of the GQDs-hMSNs showed C-O/C-N and C-C/C-Si bonds within the carbon peak ( Figure 4A), NSi2O and NSi2Ox in the nitrogen peak ( Figure 4B), and the silicon showed SiO and SiOx bonds ( Figure 4C). The presence of the C-Si bond supported the spectrum seen in FT-IR and confirmed the chemical binding of GQDs to the hMSNs surface, and further explained the slight peak shift in the excitation wavelength.

Optical Properties and pH Effects
Cell labeling for confocal microscopy imaging require fluorescent characteristics of nanoparticles in order to be applicable. Thus, the absorption, excitation and emission spectra were collected of the GQDs-hMSNs. Previously, the GQDs contained an absorbance of 300 nm which was attributed to the C-C bonds π-π* transition as well as the n-π* of the

Optical Properties and pH Effects
Cell labeling for confocal microscopy imaging require fluorescent characteristics of nanoparticles in order to be applicable. Thus, the absorption, excitation and emission spectra were collected of the GQDs-hMSNs. Previously, the GQDs contained an absorbance of 300 nm which was attributed to the C-C bonds π-π* transition as well as the n-π* of the oxygen-containing groups [67]. However, after conversion with hMSNs, the new GQDs-hMSNs spectral features slightly shifted, with an absorbance at 310 nm ( Figure 5A) and an excitation peak at 330 nm ( Figure 5B) due to chemical binding on the silica surface. The emission peak at 440 nm remained the same ( Figure 5B). Additionally, the GQDs-hMSNs showed concentration dependence ( Figure 5C) as the GQDs have shown previously. However, the radiation dependency and red-shifting of the emission peak with increasing excitation wavelength (270-370 nm) did not occur as drastically as analyzed before ( Figure 5D), and thus the GQDs now displayed a better rapid internal conversion from higher initial excited states to lower states that prevented the red-shift since being attached to hMSNs.
The human body contains a range of pHs, therefore the GQDs-hMSNs are required to show fluorescence stability across different pHs and avoid quenching to optimize in vitro cell imaging. Therefore, the GQDs-hMSNs fluorescence was analyzed at pHs 1.0-11.0 for any quenching tendencies in addition to test their stability and agglomeration characteristics in this wide range of pH. Due to the negative charge of the GQDs-hMSNs as seen in Section 2.2.1, more acidic environments could potentially cause the particles to destabilize and lose their surface charge. When this happens to particles, they will agglomerate and fall out of suspension. Additionally, they will lose their fluorescent signature and forfeit their use in cell labeling applications. For these GQDs-hMSNs, at pHs 5.0-9.0, the nanoparticles retained their fluorescence and were only slightly quenched at pH 11.0. Furthermore, pHs 1.0 and 3.0 showed greater quenching, with pH 1.0 nearly losing fluorescence altogether ( Figure 6). However, this is typical of GQDs as extreme acidic conditions lead to agglomeration and solution instability, as previously mentioned, primarily due to excessive positive charge of the solution caused the negatively charged particles to destabilize. This was also seen in the zeta potential analysis (Section 2.2.1). Nonetheless, the highest fluorescence observed in the neutral pHs was encouraging for cell labeling applications as the fluorescence will be maintained in the biological pHs. oxygen-containing groups [67]. However, after conversion with hMSNs, the new GQDs-hMSNs spectral features slightly shifted, with an absorbance at 310 nm ( Figure 5A) and an excitation peak at 330 nm ( Figure 5B) due to chemical binding on the silica surface. The emission peak at 440 nm remained the same ( Figure 5B). Additionally, the GQDs-hMSNs showed concentration dependence ( Figure 5C) as the GQDs have shown previously. However, the radiation dependency and red-shifting of the emission peak with increasing excitation wavelength (270-370 nm) did not occur as drastically as analyzed before ( Figure  5D), and thus the GQDs now displayed a better rapid internal conversion from higher initial excited states to lower states that prevented the red-shift since being attached to hMSNs. The human body contains a range of pHs, therefore the GQDs-hMSNs are required to show fluorescence stability across different pHs and avoid quenching to optimize in vitro cell imaging. Therefore, the GQDs-hMSNs fluorescence was analyzed at pHs 1.0-11.0 for any quenching tendencies in addition to test their stability and agglomeration characteristics in this wide range of pH. Due to the negative charge of the GQDs-hMSNs as seen in Section 2.2.1, more acidic environments could potentially cause the particles to destabilize and lose their surface charge. When this happens to particles, they will agglomerate and fall out of suspension. Additionally, they will lose their fluorescent signature and forfeit their use in cell labeling applications. For these GQDs-hMSNs, at pHs 5.0-9.0, the nanoparticles retained their fluorescence and were only slightly quenched at pH 11.0. Furthermore, pHs 1.0 and 3.0 showed greater quenching, with pH 1.0 nearly losing

GQDs-hMSNs Biodegradation
A goal of this project is to lessen immune system taxation via traditional cancer treatments and therefore it is important to ensure that the GQDs-hMSNs are not retained long-term in the patient body and are broken down after a couple of weeks; or at least after they have fulfilled their purpose for drug delivery and as a PS agent for PDT treatment. Solutions of 10 mM phosphate-buffered saline (PBS) buffer, pH 7.0 and 6.5, and 10 mM citrate buffer, pH 5.0 and 4.0, were used to analyze the biodegradation of the GQDs-hMSNs to mimic biological systems. The GQDs-hMSNs were added to the buffer pHs (0.2 mg/mL) and vortexed for 6 days at 37 • C, 1000 rpm. Samples were taken every 24 h and analyzed via DLS and TEM to observe particle degradation over time.
fluorescence altogether ( Figure 6). However, this is typical of GQDs as extreme acidic conditions lead to agglomeration and solution instability, as previously mentioned, primarily due to excessive positive charge of the solution caused the negatively charged particles to destabilize. This was also seen in the zeta potential analysis (Section 2.2.1). Nonetheless, the highest fluorescence observed in the neutral pHs was encouraging for cell labeling applications as the fluorescence will be maintained in the biological pHs.

GQDs-hMSNs Biodegradation
A goal of this project is to lessen immune system taxation via traditional cancer treatments and therefore it is important to ensure that the GQDs-hMSNs are not retained longterm in the patient body and are broken down after a couple of weeks; or at least after they have fulfilled their purpose for drug delivery and as a PS agent for PDT treatment. Solutions of 10 mM phosphate-buffered saline (PBS) buffer, pH 7.0 and 6.5, and 10 mM citrate buffer, pH 5.0 and 4.0, were used to analyze the biodegradation of the GQDs-hMSNs to mimic biological systems. The GQDs-hMSNs were added to the buffer pHs (0.2 mg/mL) and vortexed for 6 days at 37 °C, 1000 rpm. Samples were taken every 24 h and analyzed via DLS and TEM to observe particle degradation over time.
DLS of pH 4.0 revealed particle agglomeration after the first 24 h in the buffer ( Figure  7A). The TEM image of day 1 depicts large, swollen particles with some already rupturing under the acidic conditions ( Figure 7B). For day 3, the particles were approximately the same size as on day 1, as observed in the DLS. However, more debris particles emerged in lower quantities ( Figure 7C). Lastly, day 6 TEM depicted a large increase in debris particles along with breakage and misshaped larger particles as they continued to degrade ( Figure 7D). For pH 5.0, a similar trend was observed for the first day, with increased particle agglomeration observed in the DLS ( Figure 7E). Under TEM, day 1 GQDs-hMSNs at pH 5.0 depicted the same enlarged particles. However, they appeared to aggregate together instead of swelling like at pH 4.0 ( Figure 7F). By day 3, the particles at pH 5.0 TEM ( Figure 7G) were similar to day 3 particles at pH 4.0 while the DLS also confirmed slight degradation of the particles although they were degrading more slowly. Day 6 TEM (Figure 7H) showed rupturing particles that clumped together to yield larger size distributions observed in DLS. DLS of pH 4.0 revealed particle agglomeration after the first 24 h in the buffer ( Figure 7A). The TEM image of day 1 depicts large, swollen particles with some already rupturing under the acidic conditions ( Figure 7B). For day 3, the particles were approximately the same size as on day 1, as observed in the DLS. However, more debris particles emerged in lower quantities ( Figure 7C). Lastly, day 6 TEM depicted a large increase in debris particles along with breakage and misshaped larger particles as they continued to degrade ( Figure 7D). For pH 5.0, a similar trend was observed for the first day, with increased particle agglomeration observed in the DLS ( Figure 7E). Under TEM, day 1 GQDs-hMSNs at pH 5.0 depicted the same enlarged particles. However, they appeared to aggregate together instead of swelling like at pH 4.0 ( Figure 7F). By day 3, the particles at pH 5.0 TEM ( Figure 7G) were similar to day 3 particles at pH 4.0 while the DLS also confirmed slight degradation of the particles although they were degrading more slowly. Day 6 TEM ( Figure 7H) showed rupturing particles that clumped together to yield larger size distributions observed in DLS.
PBS buffers showed slightly different particle formations. For pH 6.5, the particles immediately aggregated and swelled after one day, as observed in DLS and TEM ( Figure 7I,J). However, the swelling was more severe than in the more acidic buffers. Day 3 TEM images showed a small decrease in particle size ( Figure 7K), with less swollen particles and the appearance of smaller degradation debris particles. Day 6 TEM ( Figure 7L) showed even more debris particles, as seen in the DLS.
In comparison, pH 7.0 showed a similar trend as pH 6.5. After day 1, the DLS results ( Figure 7M) showed the same particle size distribution as observed previously (Section 2.2.1). However, the TEM images for day 1 ( Figure 7N) showed burst or rippled particles, indicating that even after a short period of time, the particles started to degrade and lose their surface integrity. Day 3 for pH 7.0 ( Figure 7O) showed increased particle agglomeration and swelling with smaller debris particles, much like pH 6.5 on day 6. Lastly, for day 6 at pH 7.0, there was an increase in debris particles with some swollen and agglomerated particles ( Figure 7P). Therefore, the GQDs-hMSNs have a prolonged degradation at pH 7.0 compared to 4.0. Thus, they will function longer at a natural, healthy pH, allowing the particles more time to accumulate in cancerous cells without releasing the chemotherapy drug or GQDs. Furthermore, as observed in the zeta potential measurements (Section 2.2.1) the GQDs-hMSNs show agglomeration and solution instability in acidic pHs. This will contribute to the increased agglomeration seen at pH 5.0 and 4.0 after one day of exposure. From there, the particles degraded faster at pH 4.0 based on the particle size in TEM. However, retained agglomeration at pH 5.0 for day 6 yielded a larger particle size in DLS than what was observed in TEM. Furthermore, pH 7.0 and 6.5 showed less severe particle agglomeration after one day, as expected due to the zeta potential measurements previously discussed. Nonetheless, the slightly more acidic pH of 6.5 showed larger particle size and agglomeration, initially, compared to pH 7.0 and showed a faster rate of degradation. Therefore, the GQDs-hMSNs will most likely degrade faster in cancerous environments that are more acidic, releasing the chemotherapy drug and GQDs into the cancer cells for treatment and imaging.

Cell Viability, In Vitro Cell Imaging, PDT Treatment, and Mock Drug Delivery of GQDs-hMSNs
In order to determine the biocompatibility of the GQDs-hMSNs, cytotoxicity tests were also analyzed using LDH assays. Effects of GQDs-hMSNs on cell viability were tested using RAW 264.7 cells, incubating the cells for 24 h with different concentrations of GQDs-hMSNs. Thermos Fisher CyQUANT™ LDH Cytotoxicity Assay Kit was performed to measure cell viability. The cells showed no signs of toxicity up to 100 µg/mL. However, at 200 µg/mL the cell counts decreased to 70-60% of the original amount (Figure 8). This is attributed to the hMSNs as the GQDs showed no cytotoxicity up to 200 µg/mL in previous studies [67]. Furthermore, to determine GQDs-hMSNs applicability in cell imaging, in vitro confocal microscopy images were collected using RAW 264.7 cells. The cells were incubated for 4 h with 100 and 200 µg/mL GQDs-hMSNs ( Figure 9A). Additionally, TO-PRO-3 dye was used to label the cell nucleus. Upon imaging, weak fluorescence signal was found using the 488 nm laser for the GQDs-hMSNs ( Figure 9B). This is due to the 488 nm laser being just outside the GQDs optimal excitation wavelength. However, the concentration of GQDs added in the synthesis can be increased to bind more GQDs to the surface of the hMSNs to provide a stronger fluorescent signal in the cells. Furthermore, to determine GQDs-hMSNs applicability in cell imaging, in vitro confocal microscopy images were collected using RAW 264.7 cells. The cells were incubated for 4 h with 100 and 200 µg/mL GQDs-hMSNs ( Figure 9A). Additionally, TO-PRO-3 dye was used to label the cell nucleus. Upon imaging, weak fluorescence signal was found using the 488 nm laser for the GQDs-hMSNs ( Figure 9B). This is due to the 488 nm laser being just outside the GQDs optimal excitation wavelength. However, the concentration of GQDs added in the synthesis can be increased to bind more GQDs to the surface of the hMSNs to provide a stronger fluorescent signal in the cells.
The GQDs-hMSNs were also examined for their use in PDT treatment of RAW 264.7 cells using LysoTracker Red in confocal microscopy and LDH assay for cytotoxicity analysis. A concentration range of 0-200 µg/mL GQDs-hMSNs was tested and after 30 min of laser irradiation at 11.81 mW/cm 2 , the cells were incubated for another 24 h. The cytotoxicity showed a linear decrease in cell viability with increasing concentrations of GQDs-hMSNs until 20 µg/mL, signifying that the GQDs-hMSNs were generating 1 O 2 and induced toxicity in the RAW 264.7 cells ( Figure 10A). Concentrations higher than 20 µg/mL decreased cell viability compared to the control, however, the decrease was similar to the 20 µg/mL concentration. This decrease in 1 O 2 generation is attributed to the induced toxicity of the hMSNs observed in the cytotoxicity analysis previously discussed. The cells were subjected to a greater toxicity due to the presence of hMSNs before the GQDs could generate 1 O 2 . Therefore, fewer viable cells would be available for PDT treatment and display a decrease in effective PDT treatment, but not necessarily due to a lack of 1 O 2 generation. Additionally, LysoTracker Red was used to track toxicity due to the ROS generation within the RAW 264.7 cells. Confocal microscope images were taken using the 488 nm laser ( Figure 10B) for GQDs-hMSNs and the 561 nm laser ( Figure 10C) for LysoTracker Red. The TD ( Figure 10D) and merged ( Figure 10E) images showed that the GQDs-hMSNs and LysoTracker Red are in the RAW 264.7 cells' cytoplasm. This signified that the ROS generation came from the incubated GQDs-hMSNs, which were the main contributor to the cells' decrease in viability.
Furthermore, to determine GQDs-hMSNs applicability in cell imaging, in vitro confocal microscopy images were collected using RAW 264.7 cells. The cells were incubated for 4 h with 100 and 200 µg/mL GQDs-hMSNs ( Figure 9A). Additionally, TO-PRO-3 dye was used to label the cell nucleus. Upon imaging, weak fluorescence signal was found using the 488 nm laser for the GQDs-hMSNs ( Figure 9B). This is due to the 488 nm laser being just outside the GQDs optimal excitation wavelength. However, the concentration of GQDs added in the synthesis can be increased to bind more GQDs to the surface of the hMSNs to provide a stronger fluorescent signal in the cells.  The GQDs-hMSNs were also examined for their use in PDT treatment of RAW 264.7 cells using LysoTracker Red in confocal microscopy and LDH assay for cytotoxicity analysis. A concentration range of 0-200 µg/mL GQDs-hMSNs was tested and after 30 min of laser irradiation at 11.81 mW/cm 2 , the cells were incubated for another 24 h. The cytotoxicity showed a linear decrease in cell viability with increasing concentrations of GQDs-hMSNs until 20 µg/mL, signifying that the GQDs-hMSNs were generating 1 O2 and induced toxicity in the RAW 264.7 cells ( Figure 10A). Concentrations higher than 20 µg/mL decreased cell viability compared to the control, however, the decrease was similar to the 20 µg/mL concentration. This decrease in 1 O2 generation is attributed to the induced toxicity of the hMSNs observed in the cytotoxicity analysis previously discussed. The cells were subjected to a greater toxicity due to the presence of hMSNs before the GQDs could generate 1 O2. Therefore, fewer viable cells would be available for PDT treatment and display a decrease in effective PDT treatment, but not necessarily due to a lack of 1 O2 generation. Additionally, LysoTracker Red was used to track toxicity due to the ROS generation within the RAW 264.7 cells. Confocal microscope images were taken using the 488 nm laser ( Figure 10B) for GQDs-hMSNs and the 561 nm laser ( Figure 10C) for LysoTracker Red. The TD ( Figure 10D) and merged ( Figure 10E) images showed that the GQDs-hMSNs and LysoTracker Red are in the RAW 264.7 cells' cytoplasm. This signified that the ROS generation came from the incubated GQDs-hMSNs, which were the main contributor to the cells' decrease in viability.  Drug delivery via GQDs-hMSNs using FITC dye as a mock drug was also analyzed by confocal microscopy. Concentrations of 0, 50, 100, and 200 µg/mL GQDs-hMSNs@FITC were cultured with RAW 264.7 cells for 4 h. The results showed an increasing fluorescence signal with increasing concentration of GQDs-hMSNs@FITC ( Figure 11A), confirming that FITC is being delivered by the GQDs-hMSNs. Lower concentrations, such as 100 µg/mL of the GQDs-hMSNs@FITC still adequately delivered the mock drug and yielded exceptional cell images ( Figure 11B). Due to the increased toxicity of the GQDs-hMSNs above this concentration, lower amounts of the nanoparticles can still be effectively used as a drug carrier. Drug delivery via GQDs-hMSNs using FITC dye as a mock drug was also analyzed by confocal microscopy. Concentrations of 0, 50, 100, and 200 µg/mL GQDs-hMSNs@FITC were cultured with RAW 264.7 cells for 4 h. The results showed an increasing fluorescence signal with increasing concentration of GQDs-hMSNs@FITC ( Figure 11A), confirming that FITC is being delivered by the GQDs-hMSNs. Lower concentrations, such as 100 µg/mL of the GQDs-hMSNs@FITC still adequately delivered the mock drug and yielded exceptional cell images ( Figure 11B). Due to the increased toxicity of the GQDs-hMSNs above this concentration, lower amounts of the nanoparticles can still be effectively used as a drug carrier.

Materials and Sample Preparations
A Millipore water purification system (18.3 Ω cm) was used to produce DI-water.

Materials and Sample Preparations
A Millipore water purification system (18.3 Ω cm) was used to produce DI-water. HEPES buffer (pH 7.0) and mixed at 1000 rpm at 37 • C for 3 h before being dried at 60 • C to a solid.

Cell Toxicity
A total of 1.0 mL of 5.3 µM FITC (2.06 µg) was added to 1.0 mg GQDs-hMSNs in 4.0 mL of 20 mM HEPES buffer (pH 7.0) and vortexed at 1000 rpm for 3 h at 37 • C. The solution was then centrifuged at 10,000 rpm to separate any FITC not absorbed by the GQDs-hMSNs. Then, the particles were resuspended in 20 mM HEPES buffer pH 7.0 for absorption spectroscopy to quantify how much FITC was absorbed. Compared to the control, on average, 75% of FITC added was absorbed by the GQDs-hMSNs (GQDs-hMSNs@FITC).
Cell viability of GQDs-hMSNs was tested with CyQUANTTM LDH Cytotoxicity Assay kit. Briefly, the cells were plated into a 96-well plate overnight. A gradient concentration of GQDs-hMSNs (0, 1, 2, 5, 10, 20, 50, 100 and 200 µg/mL) was added into the wells. After 24 h incubation in a CO 2 incubator at 37 • C, 10 µL of 10X Lysis buffer was added into 3 wells of 0 µg/mL group to serve as the maximum LDH controls. After 45 min, 50 µL aliquots of all groups were transferred to a new 96-well plate, mixed with 50 µL of the Reaction Mixture solution. After 30 min incubation at room temperature protected from light, 50 µL of Stop Solution was added to the wells, and the absorbance read using an EXL800 microplate reader to quantify cell viability.

Cell Imaging, PDT, and Drug Delivery
Cell imaging of GQDs-hMSNs or GQDs-hMSNs was investigated by adding 0, 100, and 200 µg/mL GQDs-hMSNs in to an 8-well slide chamber with cells inside. After a 4 h of incubation, the wells were rinsed with 1× PBS for 15 min 3 times, then fixed with 4% PFA for 15 min. After additional washes with PBS, To-Pro-3 was added to stain the nuclei for 15 min. After washing with PBS, the chamber was removed and the slide was covered with a cover slide by mounting medium. When dry, the slides were imaged using an Olympus FV3000 Laser Scanning Confocal Microscope.
PDT analysis of GQDs-hMSNs was conducted using a LysoTracker Red lysosome kit and LDH assay. For cell viability assays, cells were seeded into a 96-well plate and incubated overnight. GQDs-hMSNs were then added in concentrations of 0, 1, 2, 5, 10, 20, 50, 100 and 200 µg/mL and incubated for 4 h before being washed with 1× PBS buffer. The cells were then subjected to 785 nm laser irradiation for 30 min at 11.81 mW/cm 2 and then incubated for another 24 h before analysis of the LDH assay kits. For confocal imaging, RAW 264.7 cells were seeded in an 8-well slide chamber and incubated overnight. After incubation with 200 µg/mL GQDs-hMSNs for 4 h, the cells were washed with 1× PBS buffer and after a 30 min incubation, fresh medium was added before for confocal imaging.

Conclusions
hMSNs were successfully synthesized using the Stöber method that allowed GQDs to be incorporated onto their surface via sonication to form 47.0 ± 19.0 nm GQDs-hMSNs for cell imaging and dual cancer treatment. The fluorescence characteristics of the GQDs-hMSNs were comparable to GQDs, showing pH sensitivity in acidic conditions, and zeta potential measurements showing particle agglomeration at pH 1.0-3.0, and concentration dependency. Particle degradation was investigated at pH 7.0, 6.5, 5.0, and 4.0 after being vortexed for 6 days at 1000 rpm, 37 • C. After 6 days, it was observed that the GQDs-hMSNs at pH 7.0 degrade at a slower rate compared to pH 4.0, indicating that the particles would degrade faster in cancer cells and would not linger in the patient to induce toxic effects.
The GQDs-hMSNs showed slight cytotoxic effects at 200 µg/mL, attributed to the hM-SNs. With the fluorescence characteristics of the GQDs being maintained, the GQDs-hMSNs were successful absorbed by RAW 264.7 cells after 4 h of incubation to yield fluorescent in vitro images. Furthermore, RAW 264.7 cells were successfully imaged using the GQDs-hMSNs. However, higher concentrations of GQDs were needed as the 488 nm laser was out of range for optimal excitation of the GQDs-hMSNs. This issue can be addressed the last step of the synthesis by adding greater concentrations of GQDs to the hMSNs. Adding more GQDs to the GQDs-hMSNs will also aid in the toxicity analysis, as less GQDs-hMSNs need to be added to increase the fluorescence signal if a greater concentration of GQDs are attached. Furthermore, the hMSNs have shown great adaptability to the GQDs, and thus a higher concentration of GQDs can be added during the synthesis to increase their concentration for cell imaging without increasing the concentration of hMSNs to help reduce toxicity. PDT treatment was also successful using the GQDs-hMSNs, showing induced toxicity with increasing concentration as more 1 O 2 was generated. Additionally, the hMSNs also displayed excellent drug loading and release characteristics, utilizing FITC dye as the mock drug. RAW cells were incubated with the GQDs-hMSNs doped with FITC, and the resulting confocal images displayed excellent delivery of FITC into the cells.
Author Contributions: Conceptualization, data curation, methodology, writing-original draft preparation, writing-review and editing, S.R.; data curation, methodology, writing-review and editing, Y.W.; data curation, methodology, writing-review and editing, D.S.; data curation, methodology, writing-review and editing, C.M.; data curation, methodology, writing-review and editing, N.O.; resources, writing-review and editing, C.C.; writing-review and editing, resources, supervision, J.X.Z. All authors have read and agreed to the published version of the manuscript.
Funding: This work was supported by the NSF grant CHE 1709160 and NSF EPSCoR RII Track I Cooperative Agreement 1946202. Imaging studies were conducted in the UND Imaging Core facility supported by NIH grant P20GM113123, DaCCoTA CTR NIH grant U54GM128729, and UNDSMHS funds.

Data Availability Statement:
The authors confirm that the data supporting the findings of this study are available within the article.