Some Nanocarrier’s Properties and Chemical Interaction Mechanisms with Flavones

Flavones such as 7,8-dihydroxyflavone (tropoflavin), 5,6,7-trihydroxyflavone (baicalein), 3′,4′,5,6-tetrahydroxyflavone (luteolin), 3,3′,4′,5,5′,7-hexahydroxyflavone (myricetin), 4′,5,7-trihydroxyflavone (apigenin), and 5,7-dihydroxyflavone (chrysin) are important both for their presence in natural products and for their pharmacological applications. However, due to their chemical characteristics and their metabolic processes, they have low solubility and low bioavailability. Knowledge about the physicochemical properties of nanocarriers and the possible mechanisms of covalent and non-covalent interaction between nanoparticles (NPs) and drugs is essential for the design of nanocarriers to improve the bioavailability of molecules with pharmacological potential, such as tropoflavin, baicalein, luteolin, myricetin, apigenin, and chrysin. The parameters of characterization of some NPs of these flavones, such as size, polydispersity index (PDI), zeta potential, encapsulation efficiency (EE), and % release/time, utilized in biomedical applications and the covalent and non-covalent interactions existing between the polymeric NPs and the drug were analyzed. Similarly, the presence of functional groups in the functionalized carbon nanotubes (CNTs), as well as the effect of pH on the % adsorption of flavonoids on functionalized multi-walled carbon nanotubes (MWCNT-COOH), were analyzed. Non-covalent interaction mechanisms between polymeric NPs and flavones, and covalent interaction mechanisms that could exist between the NPs and the amino and hydroxyl functional groups, are proposed.


Introduction
The application of advanced research in nanotechnology together with advances in biomedical and pharmacological sciences has allowed nanomedicine to evolve rapidly over recent decades. This has been achieved primarily through advances in nanocarrier design and research. According to their nature, nanocarriers are classified as organic, inorganic, and hybrid [1,2].
Several mechanisms exist for the cellular internalization of nanocarriers, depending on their physicochemical properties. When the pharmaceutical target is located inside the cell, its pharmacological action takes place through mechanisms that involve its passage through the cell membrane in cell-cell or cell-specialized tissue interaction, for example, its passage through mucous membranes, epithelia, or endothelium; diffusing through the plasma membrane; or accessing the specific organelle where the biological target is located [6].
Nanocarriers' functionality is influenced by their physicochemical properties, such as shape, size, loading surface, hydrophobicity, material, stiffness, and elasticity [7]. The Flavones such as tropoflavin, baicalein, luteolin, myricetin, apigenin ( Figure 1) are studied mainly for their properties as pharmacological age droxyflavone (tropoflavin 1) is a flavone found in species such as Godman Tridax procumbens, Primula farinosa L., and Chrysanthemum morifolium [33]. A tective agent, the presence of 7,8-OH groups in its structure makes it a chem emulates the biochemical and physiological action of brain-derived neu (BDNF) and serves as a selective agonist of the tropomyosin-related kina (TrkB) [34,35]. The anticancer activity of tropoflavin has been demonstrate (2020) [33], as an inhibitor of human ornithine decarboxylase (ODC), in in Similarly, it has been shown that tropoflavin had anti-enterovirus (EV71) ac centration of 50 Μm. It inhibits 40% of viral IRES-internal ribosome entry tivity by interfering with virus replication [36,37]. The presence of tropoflavin in plasma is detectable after 8 h (5 ng/mL) o tion. In vivo metabolism study shows that tropoflavin undergoes glucuro fation, and methylation. Among these modifications, glucuronidation and mainly responsible for the in vivo elimination of flavonoids [34] and their l bility.
Baicalein at a dose of 20 mg/kg, 5 d/week for 21 days can inhibit M cancer by 40%, which is comparable to the positive effect of the drug cispla [55]. Baicalein has also shown anticancer effects on hepatocellular carcinom by modulating associated molecules and signaling pathways resulting in inh   The presence of tropoflavin in plasma is detectable after 8 h (5 ng/mL) of administration. In vivo metabolism study shows that tropoflavin undergoes glucuronidation, sulfation, and methylation. Among these modifications, glucuronidation and sulfation are mainly responsible for the in vivo elimination of flavonoids [34] and their low bioavailability.
Baicalein at a dose of 20 mg/kg, 5 d/week for 21 days can inhibit MDA468 breast cancer by 40%, which is comparable to the positive effect of the drug cisplatin (5 mg/kg) [55]. Baicalein has also shown anticancer effects on hepatocellular carcinoma. It does this by modulating associated molecules and signaling pathways resulting in inhibition of cell proliferation, induction of apoptosis, cell cycle arrest, and induction of autophagy [42]. Xiaoling et al. (2018) [56] evaluated the effect of baicalein on cervical cancer cells and other cancer types, they concluded that baicalein arrests the cell cycle in the G 0 /G 1 phase and cyclin D1 decrease through the AKT-GSK3β signaling pathway [43]. Baicalein was able to reduce endometriosis by suppressing the viability of human endometrial stroma in cells in vitro [57].
In studies carried out by Chen et al. (2017) [53], it was found that baicalein affects the osteogenic differentiation of human periodontal ligament cells (hPDLCs). These cells are important in periodontal tissue regeneration, decreasing the growth of hPDLCs, increasing alkaline phosphatase and calcium deposition in a dose-dependent manner.
Błach-Olszewska et al. (2008) [58], demonstrated that baicalein regulates natural innate immunity by modulating cytokine production and stimulating resistance in human leukocytes. Sithisarn et al. (2013) [47] evaluated the effect of baicalein on human lung epithelial cells (A549) infected with avian influenza strain A/Thailand/Kan-1/04 and found that baicalein inhibited the production of virus proteins and the viral replication cycle. Similarly, baicalein has been shown to have antiviral effects against replication of the mosquito-borne Chikungunya virus (CHIKV), which causes disabling arthritis in infected individuals. It is considered that baicalein exerts its antiviral action in the first hours of treatment and in the early stages of infection, decreasing the production of CHIKV proteins at concentrations of 100 µg/mL [45].
In addition to the above pharmacological applications, baicalein also a strong synergistic effect with penicillin G/amoxicillin against 20 penicillin-producing clinical strains of S. aureus [46].
Baicalein has a low bioavailability due to its poor water solubility, being a class IV compound according to the Biopharmaceuticals Classification System (BCS) (solubility: 0.052 mg/Ml; lipophilicity: Papp = 0.037 × 10 −6 cm/s) [59]. The low availability of baicalein may be due to the transformations it undergoes following the processes of glucuronidation and sulfation through the small intestine. For this reason, baicalein is considered a drug with a presystemic or first-pass metabolism, since when the drug is administered orally, the active ingredient of the drug is considerably reduced before it reaches the circulatory system [60].
Epidemiological studies have found that the consumption of high levels of luteolin decreases the risk of developing some chronic diseases [61]. Several studies have demonstrated luteolin's activities as an antioxidant [63], anti-inflammatory [64], anti-cancer [65], and chemotherapy agent [66]. Castellino et al. (2019) [67] evaluated the effect of luteolin supplementation on cardiometabolic risk factors in human patients and found an improvement in vascular function through dilation of the brachial artery. Luteolin also has other biological activities [68].
Tropoflavin, baicalein, luteolin, myricetin, apigenin, and chrysin are flavones that have low bioavailability due to their low solubility in water. The enhancement of the drug-like properties of natural compounds, such as their bioavailability, targeting, and controlled release, has been achieved through the incorporation of nanoparticles. However, increasing the bioavailability of a drug also depends on the chemical interactions that occur between the functional groups of the chemical compound itself with cell surface molecules either through cell-cell, cell-tissue, and cell-surrounding environment interactions. Properties of some NPs of these flavones, the covalent and non-covalent interactions mechanism between the NPs and drugs, as well as the effect of factors such as pH on the functionalized MWCNT-COOH, have been reviewed and analyzed. Some mechanisms of non-covalent and covalent interaction with the functional groups have also been proposed. Regarding the effect of pH on the stability of tropoflavin liposomes, it was observed that at pH values of 2-12, the size of liposomes presented relative stability, due to the protection provided by the structure of phospholipids and cholesterol to the liposomal membrane from the effects of acidic pH. At a tropoflavin concentration range of 1-100 µM, the tropoflavin liposomes do not exhibit cytotoxicity and were used in transmembrane transport assays. However, the same authors in another work [92] obtained an EE of 98.31% for tropoflavin-zein-lactoferrin NPs, with a NP size of 82.3 ± 1.01 nm. On the other hand, Prasanna et al. (2021) [93] encapsulated tropoflavin in gold NPs, which presented a size of 35 nm and a zeta potential of −34.1 mV.

NPs-Flavones
Since flavones are non-polar, they should be confined to the hydrocarbon region of the lipid membrane; however, in a study by Scheidt et al. (2004) [94], chrysin, luteolin, and myricetin molecules presented a rapid reorientation in the membrane, detecting a maximum distribution at the lipid/water interface. Thus, the hydroxyl groups of the flavones participate in the bond networks of the lipid-water interface. Therefore, the interaction between flavone and lipid occurs via the H-bond between the OH groups of the flavone with both the C=O groups of the lipid and the phosphate group. A π-cation interaction occurs between the NH 3 + group of the phospholipid and the A-ring of the flavone ( Figure 2). In addition to the above, other favorable interactions between the flavone and the lipid molecule involve charge-dipole and dipole-dipole interactions.

5,6,7-Trihydroxyflavone
With respect to the characterization of the 5,6,7-trihydroxyflavone (baicalein) NPs, it has been found that the size of mesoporous silica NPs (MSNs) is considerably larger, 367 ± 94 nm [95], than the size of liposomes 135-154 nm [96] or NPs 82.5 ± 1.7 nm [97]. Regarding the variation of the size of NPs in the NP population that is PDI, it is observed that nanoemulsions (NE) and NPs are monodisperse, while liposomes vary in size. The surface charge of baicalein encapsulated in NE and represented by the zeta potential is very low −22.4 ± 3.1 mV [98], indicating nanoaggregates formation, as in mesoporous silica NPs (MSNs), in contrast to liposomes and NPs, where high potential values, such as −1.89 -−2.11 mV and −1.5 ± 0.4 mV respectively, indicate suspension stability. Baicalein liposomes' stability was homogeneous for more than two weeks, and its in vivo stability was affected by the interaction with other proteins and between lipoproteins present in the circulation. Encapsulated baicalein increased the cell viability of Hs68 fibroblasts by reducing their cytotoxicity. Regarding EE, a decrease from 33.65 to 25.40% occurred when baicalein concentration was increased from 30 to 80 µg/mL and was much lower than the EE that occurred for NE, 98%, and NPs, 86.2% (Table 1).

5,6,7-Trihydroxyflavone
With respect to the characterization of the 5,6,7-trihydroxyflavone (baicalein) NPs, i has been found that the size of mesoporous silica NPs (MSNs) is considerably larger, 36 ± 94 nm [95], than the size of liposomes 135-154 nm [96] or NPs 82.5 ± 1.7 nm [97]. Regard ing the variation of the size of NPs in the NP population that is PDI, it is observed tha nanoemulsions (NE) and NPs are monodisperse, while liposomes vary in size. The surfac charge of baicalein encapsulated in NE and represented by the zeta potential is very low −22.4 ± 3.1 mV [98], indicating nanoaggregates formation, as in mesoporous silica NP (MSNs), in contrast to liposomes and NPs, where high potential values, such as −1.89 −2.11 mV and −1.5 ± 0.4 mV respectively, indicate suspension stability. Baicalein lipo somes' stability was homogeneous for more than two weeks, and its in vivo stability wa affected by the interaction with other proteins and between lipoproteins present in th circulation. Encapsulated baicalein increased the cell viability of Hs68 fibroblasts by re ducing their cytotoxicity. Regarding EE, a decrease from 33.65 to 25.40% occurred whe baicalein concentration was increased from 30 to 80 µg/mL and was much lower than th EE that occurred for NE, 98%, and NPs, 86.2% (Table 1).   Baicalein liposomes containing mainly phosphatidylcholine have a neutral surface, and the zeta potential was around 0. The presence of baicalein does not alter the electrophoretic mobility of the liposomes. Fang et al. (2018) [96] concluded that the reduction in particle size may be mainly due to strong interactions with baicalein via the H-bond but has no effect on the zeta potential.
Baicalein has low solubility in aqueous solution and low bioavailability in vivo, which limits its application; liposomal nanoencapsulation is recommended due to its high compatibility and easy incorporation efficiency [96]. Therefore, drugs encapsulated in NPs present a reduced dispersion of the drug, longer permanence time on the skin, and facilitate the internalization of the drug into the cells [99]. Similarly, because of their ease of preparation, handling, biocompatibility, and biodegradability, NEs are a model for drug delivery [100].
In cancer chemotherapy and other diseases, several strategies have been developed to improve the transport, delivery, solubility, and bioavailability of drugs from bioactive or synthetic compounds. To improve these properties, mainly solubility and bioavailability, conventional nanocarriers such as liposomes, micelles, nanocrystals, and polymeric NPs have been designed. In the case of baicalein, several nanocarriers have been developed to improve the low solubility and bioavailability ( Table 2). It is important to highlight the studies carried out on the development of delivery systems based on conjugated nano drugs or self-assembled NPs. In a study carried out by   [97], they developed self-assembled hyaluronic acid-lysine-baicalein (HA-L-BCL) NPs and evaluated their anticancer effect on A549 human lung cancer cells. Firstly, the L-BCL complex was synthesized by the Mannich reaction ( Figure 3). Subsequently, the N-hydroxysuccinimide-lysine-baicalein (NHS-L-BCL) system was synthesized in the presence of 1,4-diamino butane as reactant and sodium cyanoborohydride (NaCNBH 3 ) as reagent and then HA-L-BCL NPs were prepared in the presence of poly-(lactic-co-glycolic acid) (PLGA). limits its application; liposomal nanoencapsulation is recommended due to its high c bility and easy incorporation efficiency [96]. Therefore, drugs encapsulated in NPs p reduced dispersion of the drug, longer permanence time on the skin, and facilitate t nalization of the drug into the cells [99]. Similarly, because of their ease of preparati dling, biocompatibility, and biodegradability, NEs are a model for drug delivery [100 In cancer chemotherapy and other diseases, several strategies have been de to improve the transport, delivery, solubility, and bioavailability of drugs from b or synthetic compounds. To improve these properties, mainly solubility and bioa ity, conventional nanocarriers such as liposomes, micelles, nanocrystals, and po NPs have been designed. In the case of baicalein, several nanocarriers have bee oped to improve the low solubility and bioavailability ( Table 2).  [97], they developed self-assembled hyaluronic acid-lysine-b (HA-L-BCL) NPs and evaluated their anticancer effect on A549 human lung canc Firstly, the L-BCL complex was synthesized by the Mannich reaction ( Figure 3) quently, the N-hydroxysuccinimide-lysine-baicalein (NHS-L-BCL) system was sized in the presence of 1,4-diamino butane as reactant and sodium cyanoboro (NaCNBH3) as reagent and then HA-L-BCL NPs were prepared in the presence (lactic-co-glycolic acid) (PLGA  The non-covalent interaction for HA-L-BCL-PLGA complex formation can be presented as the following: HA-Lysine interaction is H-bond type interaction between the H-C=O carbonyl group of HA and the NH 2 group of lysine ( Figure 4).  The non-covalent interaction for HA-L-BCL-PLGA complex formation can be presented as the following: HA-Lysine interaction is H-bond type interaction between the H-C=O carbonyl group of HA and the NH2 group of lysine ( Figure 4). The interaction between baicalein and lysine could be by the H-bond type between the NH2 group of lysine and the B-ring 7-OH group of baicalein or between the C=O group of lysine and the same 7-OH group of baicalein. An H-bond interaction can also occur between the alkylammonium ion (-CH2-NH3 + ) of lysine and the 6-OH group of baicalein as in the scheme above.
In the HA-L-BCL-PLGA polymeric complex, an H-bond interaction can occur between the carbonyl group of the C-ring of baicalein and the carboxyl group of PLGA or between the 5-OH and 6-OH groups of baicalein and PLGA as depicted Figure 4.
According to Pool et al. (2012) [101], a negative ζ potential value at pH 7.0 of the polymeric NPs of some flavonoids may be due to the presence of the ionized carboxyl groups of the PLGA matrix. In the same way, the presence of each specific flavonoid changes the charge of the polymeric NPs, without altering the stability of the NPs. Values of ζ between ≈+30 and ≈−30 mV exert enough electrostatic repulsive force to prevent aggregation of the nanoparticles. The interactions that exist between the flavonoid and the PLGA matrix are non-covalent such that each flavonoid is dispersed in a non-crystalline state within the polymeric PLGA matrix.
The utilization of polymers with PLGA has advantages such as chemical and mechanical stability, ease of modification and adjustment of properties, as well as low non-  The interaction between baicalein and lysine could be by the H-bond type between the NH 2 group of lysine and the B-ring 7-OH group of baicalein or between the C=O group of lysine and the same 7-OH group of baicalein. An H-bond interaction can also occur between the alkylammonium ion (-CH 2 -NH 3 + ) of lysine and the 6-OH group of baicalein as in the scheme above.
In the HA-L-BCL-PLGA polymeric complex, an H-bond interaction can occur between the carbonyl group of the C-ring of baicalein and the carboxyl group of PLGA or between the 5-OH and 6-OH groups of baicalein and PLGA as depicted Figure 4.
According to Pool et al. (2012) [101], a negative ζ potential value at pH 7.0 of the polymeric NPs of some flavonoids may be due to the presence of the ionized carboxyl groups of the PLGA matrix. In the same way, the presence of each specific flavonoid changes the charge of the polymeric NPs, without altering the stability of the NPs. Values of ζ between ≈+30 and ≈−30 mV exert enough electrostatic repulsive force to prevent aggregation of the nanoparticles. The interactions that exist between the flavonoid and the PLGA matrix are non-covalent such that each flavonoid is dispersed in a non-crystalline state within the polymeric PLGA matrix.
The utilization of polymers with PLGA has advantages such as chemical and mechanical stability, ease of modification and adjustment of properties, as well as low non-specific protein binding. However, it has disadvantages such as the possibility of polymer toxicity and non-biodegradability [102].
HA-L-BCL NPs were prepared by nanoprecipitation. BCL NPs were characterized and the EE for BCL NPs was 86.2 ± 2.7% (Table 1). Since the bonds formed with amino acids are weak, the use of amino acids such as lysine as ligands is recommended in the production of prodrugs as it facilitates drug release. Similarly, although PEG is a widely used connector for drug development, its efficiency has been demonstrated to be lower in relation to the individual drug. In HA-L-BCL NPs, its amphiphilic character allows it to self-assemble in such a way that the hydrophobic center is BCL and the hydrophilic shell is HA. At 50-200 µM concentration (p < 0.05), the cytotoxic effect of HA-L-BCL NPs was higher than that of the other formulations tested and at this same concentration, the IC 50 of the BCL NPs was 0.517 [97].  [104] obtained EE values of 74.80 and 76.4% when encapsulating luteolin in methoxy poly(ethylene glycol)-poly(lactic-co-glycolic acid) (mPEG-PLGA)-luteolin coated with a pH-dependent copolymer Eudragit S100 (Table 3). Although the percentage of EE is very similar, the size difference of 47 ± 0.51 and 197.45 ± 20.09, respectively, is notable, in addition to the difference in zeta potential of −9.62 and −23.5 ± 1.16. Ding et al. (2020) [30] encapsulated luteolin in poly(lactic-co-glycolic acid) PLGA NPs and modified them with Her-2 antibody to produce immunolabeled microspheres. They worked with SGC-7901 gastric cancer cells. The size and ES for PLGA-luteolin were 184 nm and 91.80 ± 6.2% and for Her-2-PLGA-luteolin, 203 and 90.4 ± 6.1%, respectively. The modification with the antibody altered the surface of PLGA-luteolin. However, although PLGA-luteolin has strong inhibitory activity on cancer cell proliferation, this effect was enhanced by Her-2-PLG-luteolin.  [106] for luteolin micelles encapsulated in the copolymer monomethoxy poly(ethylene glycol)-poly(ε-caprolactone) (MPEG-PCL)-luteolin. Poly(ε-caprolactone)/poly(ethylene glycol) (PCL/PEG) can selfassemble into NPs with a hydrophobic PLC center and hydrophilic PEG outside. In such a way, when a hydrophobic drug is encapsulated, the PLC and the drug will constitute the hydrophobic center, while the outside will be constituted by the hydrophilic PEG, thus originating an injectable intravenous drug. However, an EE of 51.6% was obtained by   [107] when they functionalized luteolin micelles with various copolymers (methoxy polyethylene glycol-poly(lactic-co-glycolic acid) (mPEG 5K -PLGA 10K )-luteolin). Tawornchat et al. (2021) [108] obtained an EE of 89.3% in luteolin NPs by enzymatic polymerization using H 2 O 2 as the reagent, polyphenol oxidase (PPO) as catalyst, and PEG as the matrix. They found that the presence of the oxidative enzyme-horseradish peroxidase (HRP) was necessary for the chemical transformation. Although the antiinflammatory activity of luteolin NPs is dose-dependent, there is no cytotoxicity at high doses unlike the cytotoxicity exhibited by the drug luteolin.
When the relationship between the type of NPs and the release time of these flavones was analyzed ( Figure 6) it was found that for the flavones tropoflavin, baicalein, and luteolin, the release time of the NPs was 24 h to pH 7.4 at 37 • C. However, it is important to highlight that the release percentage of baicalein from nanoemulsions was 83% [98] with respect to the 47% release of luteolin from mPEG 5K -PLGA 10K [107] and the 34% release of tropoflavin from liposomes [91]. The drug release depends on the size of the NPs, the trapping efficiency, the composition, and the biodegradation of the NPs [110].
When the relationship between the type of NPs and the release time of these flavones was analyzed ( Figure 6) it was found that for the flavones tropoflavin, baicalein, and luteolin, the release time of the NPs was 24 h to pH 7.4 at 37 °C. However, it is important to highlight that the release percentage of baicalein from nanoemulsions was 83% [98] with respect to the 47% release of luteolin from mPEG5K-PLGA10K [107] and the 34% release of tropoflavin from liposomes [91]. The highest release time of 72 h was presented by luteolin release from MPEG-PCL with 82.5% [106], while tropoflavin was released at 2 h from zein-lactoferrin NPs with 63% [92]. The same percentage of 88% release of both SLNs [103] and mPEG-PLGA [104] was presented at 48 and 4 h, respectively. NPs' pegylation with PEG provides a convenient method to provide high colloidal stability to NPs in physiological media and to avoid undesirable interactions of NPs with proteins or other blood components [111]. Luteolin release from MPEG-PCL was 49 [112] evaluated the cationic interactions between myricetin, a flavone with hydroxyl groups at the 3, 3′, 4′, 5, 5′, 7 positions, and polymeric nanoparticle carriers (NPCs). They found that flavone-NPC interaction was influenced by H-bond, ππ interactions, and van der Waals forces. Electrostatic forces between the tertiary amines in the NPC corona and the myricetin-specific hydroxyl groups stimulate π-π interactions between the unsaturated rings and enhance the conjugation between the aromatic compounds enabling π-π* transitions. The highest release time of 72 h was presented by luteolin release from MPEG-PCL with 82.5% [106], while tropoflavin was released at 2 h from zein-lactoferrin NPs with 63% [92]. The same percentage of 88% release of both SLNs [103] and mPEG-PLGA [104] was presented at 48 and 4 h, respectively. NPs' pegylation with PEG provides a convenient method to provide high colloidal stability to NPs in physiological media and to avoid undesirable interactions of NPs with proteins or other blood components [111]. Luteolin release from MPEG-PCL was 49 [112] evaluated the cationic interactions between myricetin, a flavone with hydroxyl groups at the 3, 3 , 4 , 5, 5 , 7 positions, and polymeric nanoparticle carriers (NPCs). They found that flavone-NPC interaction was influenced by H-bond, π-π interactions, and van der Waals forces. Electrostatic forces between the tertiary amines in the NPC corona and the myricetin-specific hydroxyl groups stimulate π-π interactions between the unsaturated rings and enhance the conjugation between the aromatic compounds enabling π-π* transitions.

4 ,5,7-Trihydroxyflavone
Zhai et al. (2013) [113] elaborated apigenin-polymeric micelles by thin-film dispersion method to evaluate their effect on human liver hepatocellular carcinoma (HepG2) cells and human breast cancer cell line MCF-7 (Table 5). H-bonding can occur between the hydroxyl groups of apigenin and the carboxyl of the PEG chains in the apigenin-conducting micelles, leading to a decrease in micelle size compared to the non-apigenin micelles which was 18.9 nm. Such interactions are also responsible for the negative value of the zeta potential of the micelles. Apigenin release from the polymeric micelles was about 85% at 50 h and at 37 • C. They concluded that the polymeric micelles have a greater cytotoxic effect on MCF-7 cells than on HepG2 cells. Ganguly et al. (2021) [114] elaborated apigenin NPs such as apigenin-poly(Lactic-coglycolic acid) nanoparticles (API-PLGA-NPs) and apigenin-galactose-nanoparticles (API-GAL NPs) and evaluated their effect on human liver hepatocellular carcinoma (HepG2) cells. Their internalization, apoptotic, and cytotoxic potential of both free apigenin and apigenin NPs were measured. The release of apigenin from API-NPs was 88% and from API-GAL-NPs was 86% after 8 days at pH 7.4. NPs' stability was confirmed for 90 days at 4 • C. They concluded that galactosylation of apigenin NPs significantly enhances the internalization of apigenin NPs into HepG2 cells, possibly due to the presence of asialoglycoprotein receptors on the surface of HepG2 cells, thus improving the apoptotic and cytotoxic effects of API-GAL-NPs on both API-NPs and API-PLGA-NPs.  [116], evaluated the effect of chrysin chitosan CCNPs nanoparticles on the activity of the enzyme succinate: ubiquinone oxidoreductase of mitochondrial complex II in human fibroblasts. The size of CCNPs was 49.7 ± 3.02 nm, with a positive zeta potential between +35.5 and +77.02 mV, an EE of 92.63%, and a release percentage of 90% at 18 h, pH 7.4, and at 37 • C. They found that the inhibitory effect on the activity of the enzyme succinate: ubiquinone oxidoreductase of mitochondrial complex II was greater with chrysin than with CCNPs in human fibroblasts.  Table 6.  AgNPs have been synthesized by different chemical, photochemical, electrochemical, and biological methods, as well as through organic synthesis or green synthesis of nanoparticles [120].   [121] prepared AgNPs using methoxy polyethylene glycol mPEG luteolin with a reducing agent and stabilizer without additives. They evaluated the antibacterial action of the mPEG-luteolin-AgNPs complex on Staphylococcus aureus, -Lactamases Staphylococcus aureus, Escherichia coli, and y-Lactamases Escherichia coli. Furthermore, they evaluated the cytotoxicity of the mPEG-luteolin-AgNPs complex on human neuroblastoma SK-N-SH and normal HVEC cells. The mPEG-luteolin-AgNPs presented a size of 25 nm and a zeta potential of −25.5 mV. The Ag + ions were reduced to metallic Ag 0 by the adjacent hydroxyl of the mPEG-luteolin, which is a reducing agent. So, the multiple Ag atoms collide with each other and form a crystalline Ag core that adsorbs the Ag ions and forms a colloidal [Ag]mnAg + core. Furthermore, mPEG-luteolin is strongly adsorbed on the surface of AgNPs, thus reducing their surface activity, preventing flocculation, and remaining monodisperse due to the presence of mPEG-luteolin as a stabilizing agent. They concluded that mPEG-luteolin-AgNPs exhibit a better inhibitory effect on large negative bacteria at low concentrations and exhibit toxicity on human neuroblastoma SK-N-SH cells in a dose-dependent action.

Other Flavonoids
Organic synthesis of AgNPs has been performed from extracts of numerous plants that have been utilized as reducing and stabilizing agents for nanoparticles. Kobylinska et al. (2020) [120] developed a methodology to elaborate green synthesis of AgNPs from extracts of adventitious roots of Artemisia spp. and evaluated their antimicrobial activity.  Table 6. Table 6. Some characteristics of luteolin, apigenin, and hesperetin PLGA-NPs.  AgNPs have been synthesized by different chemical, photochemical, electrochemical, and biological methods, as well as through organic synthesis or green synthesis of nanoparticles [120].   [121] prepared AgNPs using methoxy polyethylene glycol mPEG luteolin with a reducing agent and stabilizer without additives. They evaluated the antibacterial action of the mPEG-luteolin-AgNPs complex on Staphylococcus aureus, -Lactamases Staphylococcus aureus, Escherichia coli, and y-Lactamases Escherichia coli. Furthermore, they evaluated the cytotoxicity of the mPEG-luteolin-AgNPs complex on human neuroblastoma SK-N-SH and normal HVEC cells. The mPEG-luteolin-AgNPs presented a size of 25 nm and a zeta potential of −25.5 mV. The Ag + ions were reduced to metallic Ag 0 by the adjacent hydroxyl of the mPEG-luteolin, which is a reducing agent. So, the multiple Ag atoms collide with each other and form a crystalline Ag core that adsorbs the Ag ions and forms a colloidal [Ag] m nAg + core. Furthermore, mPEG-luteolin is strongly adsorbed on the surface of AgNPs, thus reducing their surface activity, preventing flocculation, and remaining monodisperse due to the presence of mPEG-luteolin as a stabilizing agent. They concluded that mPEG-luteolin-AgNPs exhibit a better inhibitory effect on large negative bacteria at low concentrations and exhibit toxicity on human neuroblastoma SK-N-SH cells in a dose-dependent action.

PLGA-NPs Size (nm) PDI ζ (mV) EE (%) Reference
Organic synthesis of AgNPs has been performed from extracts of numerous plants that have been utilized as reducing and stabilizing agents for nanoparticles. Kobylinska et al. (2020) [120] developed a methodology to elaborate green synthesis of AgNPs from extracts of adventitious roots of Artemisia spp. and evaluated their antimicrobial activity. They concluded that the solvent, composition, and concentration of reducing agents present in the plant extracts directly influence the characterization parameters of the NPs and their antimicrobial activity.
In the interaction of AgNPs with flavonoids, the presence of hydroxyl groups with high reducing activity can reduce Ag + ions through the formation of an intermediate complex followed by oxidation through the abstraction of an H atom (see Figure 8). In this process, two states of the formation of AgNPs have been recognized: the nucleation state and the growth state that finally gives rise to AgNPs. Similarly, in the sequence of reactions that initiate the Ag + ions, the formation of radical oxygen species -ROS is involved, which, when reacting with antioxidants such as flavonoids, originate oxidation or decomposition compounds. On the other hand, the presence of biomolecules allows the formation of a protective cover over the AgNPs that acts as a stabilizing agent for the nanoparticles [120]. They concluded that the solvent, composition, and concentration of reducing agents present in the plant extracts directly influence the characterization parameters of the NPs and their antimicrobial activity.
In the interaction of AgNPs with flavonoids, the presence of hydroxyl groups with high reducing activity can reduce Ag + ions through the formation of an intermediate complex followed by oxidation through the abstraction of an H atom (see Figure 8). In this process, two states of the formation of AgNPs have been recognized: the nucleation state and the growth state that finally gives rise to AgNPs. Similarly, in the sequence of reactions that initiate the Ag + ions, the formation of radical oxygen species -ROS is involved, which, when reacting with antioxidants such as flavonoids, originate oxidation or decomposition compounds. On the other hand, the presence of biomolecules allows the formation of a protective cover over the AgNPs that acts as a stabilizing agent for the nanoparticles [120].

CNTs
The functionalization processes of CNTs generally involve the breaking of C=C bonds by oxidation. Among the techniques utilized for oxidation are wet chemistry, photooxidation, oxygen plasma, and gaseous treatment methods. The wet chemistry technique consists of the functionalization of CNTs with carboxyl -COOH and hydroxyl -OH groups [32]. Table 7 reports the band shifts corresponding to the -OH, C=O, C=C, and C=O functional groups obtained from FT-IR studies performed on functionalized MWCNT-COOH.

CNTs
The functionalization processes of CNTs generally involve the breaking of C=C bonds by oxidation. Among the techniques utilized for oxidation are wet chemistry, photooxidation, oxygen plasma, and gaseous treatment methods. The wet chemistry technique consists of the functionalization of CNTs with carboxyl -COOH and hydroxyl -OH groups [32]. Table 7 reports the band shifts corresponding to the -OH, C=O, C=C, and C=O functional groups obtained from FT-IR studies performed on functionalized MWCNT-COOH.  [123] In the functionalization of MWCNTs with HNO 3 /H2SO 4 acids, agglomeration was observed after 30 min of functionalization. This agglomeration was attributed to the hydrophobicity of the graphene sidewalls and π-π interactions between the individual CNTs [32]. However, the dispersed solution was formed due to the positively and negatively charged functional groups on the walls of the CNTs, which causes repulsion between them [19]. The treatment with excess acid causes the cleavage of C=C bonds in graphene CNTs, generating functional groups at the open ends of nanotubes and cut nanotubes. Similar results were observed by Yudanti et al. (2011) [124] when functionalizing MWCNTs with HNO 3 /H2SO 4 acids. Possibly such an effect is caused by the acidic environment that produces a higher number of oxidation sites on the carbon atom by the exfoliation of graphite [125]. In addition, the increase of -COOH groups on MWCNTs caused by prolonged sonication leads to the disintegration of CNTs, making them shorter and thinner and converting them into amorphous carbon. This is possibly due to a disruption of the π electronic system of the CNT, which originates a degradation of the charge mobility and its mechanical properties [32]. Osorio et al. (2008) [19] evaluated several acid functionalization procedures with HNO 3 /H 2 SO 4 and HCl on SWCNTs and MWCNTs. In the FT-IR studies, they found a band at 1600 cm −1 corresponding to C=C, and a band between 2800 and 3500 cm −1 that corresponds to C-H and O-H related to carboxyl and hydroxyl groups. The band at 1450 cm −1 corresponding to C-O indicates the presence of carboxyl groups on the oxidation surface and the band at 620 cm −1 indicates the presence of the H-bond. In turn, bands at 1640 cm −1 corresponding to the C=O group and 1560 cm −1 corresponding to C-O-C-were found by Park et al. (2015) [27] when they functionalized MWCNT with HNO 3 /H 2 SO 4 acids.
Furthermore, Dong et al. (2013) [125] by functionalizing SWCNTs and MWCNTs with HNO 3 /H 2 SO 4 , and evaluated their effect on biocompatibility on cells and enzymes. They concluded that controlled CNT oxidation allows removal of the metal catalyst, increases the number of functional groups on the CNTs with the ability to accept electrons, originates cleaved CNTs, and improves the solubility in aqueous environments. The biocompatibility with human epithelial cells and enzymes is also improved. Figure 9 illustrates the H-bond interactions in SWCNTs after acid functionalization and possible non-covalent interactions with the bovine serum albumin protein frequently utilized in drug delivery.

CNTs-Flavonoids
In studies in silico carried out by the author between 5,6,7-trihydroxyflavone (baicalein) and SWCNTs with the Autodock 4.0 program [127], a non-covalent interaction at ΔG = −11.77 kcal/mol was found ( Figure 10). Salam and Burk (2017) [23] functionalized MWCNTs with HNO 3 and their subsequent modification with ODA-octadecyl amine (CH 3 -(CH 2 ) 17 -NH 2 ) and PEG-polyethylene glycol, HO-(CH 2 -CH 2 -O) n -OH). The solubility of MWCNTs in different solvents was evaluated and they found that MWCNTs can be soluble in non-polar solvents such as dichloromethane due to the hydrophobicity of CNTs, but insoluble in water and methanol. However, this condition changed when functionalized with COOH, which caused the increase of MWCNTs hydrophilic character, resulting in them being soluble in water and methanol, but poorly soluble in hexane and dichloromethane. This same solubility was present in MWCNT-PEG. However, in MWCNT-ODA the hydrophilic character decreased so that the solubility in polar solvents decreased while it was increased in non-polar solvents. The functional groups carboxylate -COOH and amine -NH 2 can be utilized with PEG without affecting the colloidal stability of the PEG nanoparticles in blood and plasma [126].

CNTs-Flavonoids
In studies in silico carried out by the author between 5,6,7-trihydroxyflavone (bai calein) and SWCNTs with the Autodock 4.0 program [127], a non-covalent interaction a ΔG = −11.77 kcal/mol was found ( Figure 10). The high capacity of CNTs to adsorb organic and inorganic molecules is mainly due to electrostatic interactions between the oxygen functional of the molecule and the adsor bent, but van der Waals forces and π-π stacking interactions between the aromatic ring of the molecule and the graphene mesh also contribute. Thus, the functionalization o CNTs with -OH, -COOH, and C=O groups is important [27]. However, in the following illustrative scheme (Figure 11), the interaction between 3′,4′,5,7-tetrahydroxyflavone and MWCNTs without functionalization is proposed: The high capacity of CNTs to adsorb organic and inorganic molecules is mainly due to electrostatic interactions between the oxygen functional of the molecule and the adsorbent, but van der Waals forces and π-π stacking interactions between the aromatic rings of the molecule and the graphene mesh also contribute. Thus, the functionalization of CNTs with -OH, -COOH, and C=O groups is important [27]. However, in the following illustrative scheme (Figure 11), the interaction between 3 ,4 ,5,7-tetrahydroxyflavone and MWCNTs without functionalization is proposed: Morais et al. (2020) [25] functionalized CNTs with the flavonoid naringenin-Ngn (8) (Figure 12) for targeted delivery into lung cancer cells. They found that in flavonoid functionalization, acid treatments modify the surface of CNTs, leading to the formation of oxygenated groups such as hydroxyl, ketones, carboxyl, and quinones on the surface of carbon nanotubes. This also originates the displacement of the bands corresponding to -OH and C=O at high frequencies, which indicates the formation of an intermolecular H-bond [128], and the presence of the band at 1470 cm −1 indicates the C=C elongation of the aromatic ring of the flavonoid [25]. They concluded that Ngn was adsorbed on the CNT surface by non-covalent interactions through H-bond, weak electrostatic interactions, and π-π stacking.
For their part, Gholizadeh et al. (2019) [29], evaluated the property of CNTs as adsorbents, a property related to their extensive surface area, electrostatic π-π interactions, and short equilibrium time. They utilized MWCNT-COOH as a method for the adsorption of flavonoids from orange peel. It was found that the presence of OH groups on the flavonoid molecule gives rise to H-bond interaction with the MWCNT-COOH surface.  [25] functionalized CNTs with the flavonoid naringenin-Ngn (8) (Figure 12) for targeted delivery into lung cancer cells. They found that in flavonoid functionalization, acid treatments modify the surface of CNTs, leading to the formation of oxygenated groups such as hydroxyl, ketones, carboxyl, and quinones on the surface of carbon nanotubes. This also originates the displacement of the bands corresponding to -OH and C=O at high frequencies, which indicates the formation of an intermolecular H-bond [128], and the presence of the band at 1470 cm −1 indicates the C=C elongation of the aromatic ring of the flavonoid [25]. They concluded that Ngn was adsorbed on the CNT surface by non-covalent interactions through H-bond, weak electrostatic interactions, and ππ stacking.  [29], evaluated the property of CNTs as adsorbents, a property related to their extensive surface area, electrostatic π-π interactions, and short equilibrium time. They utilized MWCNT-COOH as a method for the adsorption of   [25] functionalized CNTs with the flavonoid naringenin-Ngn (8) (Figure 12) for targeted delivery into lung cancer cells. They found that in flavonoid functionalization, acid treatments modify the surface of CNTs, leading to the formation of oxygenated groups such as hydroxyl, ketones, carboxyl, and quinones on the surface of carbon nanotubes. This also originates the displacement of the bands corresponding to -OH and C=O at high frequencies, which indicates the formation of an intermolecular H-bond [128], and the presence of the band at 1470 cm −1 indicates the C=C elongation of the aromatic ring of the flavonoid [25]. They concluded that Ngn was adsorbed on the CNT surface by non-covalent interactions through H-bond, weak electrostatic interactions, and ππ stacking.  [29], evaluated the property of CNTs as adsorbents, a property related to their extensive surface area, electrostatic π-π interactions, and short equilibrium time. They utilized MWCNT-COOH as a method for the adsorption of flavonoids from orange peel. It was found that the presence of OH groups on the flavonoid molecule gives rise to H-bond interaction with the MWCNT-COOH surface.
It is to highlight that although Gholizadeh (Table 7), the bands of the FT-IR spectra present different bands for the C=O and C=C groups. The peak corresponding to C=O is at 1639 cm −1 and that of C=C is at 1470 cm −1 for naringenin [25], whereas for the same flavonoid, these bands are at 1721 cm −1 and 1559 cm −1 , respectively [29]. Flavonoids at high pH values dissociate into their anions where their functional groups are negatively charged or neutral; therefore, at high pH values the adsorption of  (Table 7), the bands of the FT-IR spectra present different bands for the C=O and C=C groups. The peak corresponding to C=O is at 1639 cm −1 and that of C=C is at 1470 cm −1 for naringenin [25], whereas for the same flavonoid, these bands are at 1721 cm −1 and 1559 cm −1 , respectively [29]. Flavonoids at high pH values dissociate into their anions where their functional groups are negatively charged or neutral; therefore, at high pH values the adsorption of flavonoids decreases due to the electrostatic repulsion of identical charges. Thus, Gholizadeth et al. (2019) [29] utilized MWCNT-COOH as an adsorption method for flavonoids such as rutin (9) (Figure 12) and naringenin. The adsorption percentage was determined using the following equation where A 0 and A e are the initial and final adsorptions. They found that the percentage adsorption of flavonoids was pH-dependent (see Figure 13). So, at pH 2 the adsorption was 45% while at pH 7 it was 25%. However, Yang et al. (2019) [28] found that the adsorption of rutin on MWCNT-COOH was 48% at pH 6.0.
using the following equation % 100 where A0 and Ae are the initial and final adsorptions. They found that the percen sorption of flavonoids was pH-dependent (see Figure 13). So, at pH 2 the adsorp 45% while at pH 7 it was 25%. However, Yang et al. (2019) [28] found that the ad of rutin on MWCNT-COOH was 48% at pH 6.0.  [29] also ob high 96.2% release of this same flavonoid from MWCNTs at pH 9. An 83% rutin from MWCNT-COOH functionalized with polyamidoamine was obtained by Ya (2019) [28].
Compounds such as polydopamine, cyclodextrin, chitosan, and polyacrylic a also been used to modify CNTs, as well as to change the functional groups and/or the adsorbent properties. The utilization of polyamidoamine dendrimers is freq cause of the amide units and the large number of amino groups in their structur make them suitable for bonding to other compounds and improving their adsorp pacity. Thus, the introduction of amino groups to an adsorbent could improve its tion capacity specifically with flavonoids.   [28].
Compounds such as polydopamine, cyclodextrin, chitosan, and polyacrylic acid have also been used to modify CNTs, as well as to change the functional groups and/or modify the adsorbent properties. The utilization of polyamidoamine dendrimers is frequent because of the amide units and the large number of amino groups in their structure, which make them suitable for bonding to other compounds and improving their adsorption capacity. Thus, the introduction of amino groups to an adsorbent could improve its adsorption capacity specifically with flavonoids. The non-covalent interactions are based on van der Waals forces, π-π stacking, and are thermodynamically controlled [128]. The frequent presence of the six-membered ring of the flavonoid on the CNTs gives rise to a moderate π-π interaction with the aromatic rings of the rutin. In addition, polyamidoamine demonstrated that some amino and especially H-bond groups can form between the -NH2 of CNTs and the -C=O and -OH groups of rutin (see Figure 14). Therefore, π-π conjugation and H-bond interactions are important in the adsorption processes of flavonoid to CNTs functionalized with polyamidoamine; however, adsorption capabilities can be improved with NHS and EDC [28].
of the flavonoid on the CNTs gives rise to a moderate π-π interaction with the aromatic rings of the rutin. In addition, polyamidoamine demonstrated that some amino and especially H-bond groups can form between the -NH2 of CNTs and the -C=O and -OH groups of rutin (see Figure 14). Therefore, π-π conjugation and H-bond interactions are important in the adsorption processes of flavonoid to CNTs functionalized with polyamidoamine; however, adsorption capabilities can be improved with NHS and EDC [28]. In the process of functionalization of NPs with amine groups by the addition of NHS, the carboxyl groups are activated and by reaction with the exposed amines of the proteins that are present on the surface, the molecule binds to the cell surface. For bioconjugated molecules, as well as for the conjugation of small molecules and synthetic macromolecules, this type of NHS ester-based reaction (reaction 1) is frequently utilized. (see Figure  15).  In the process of functionalization of NPs with amine groups by the addition of NHS, the carboxyl groups are activated and by reaction with the exposed amines of the proteins that are present on the surface, the molecule binds to the cell surface. For bioconjugated molecules, as well as for the conjugation of small molecules and synthetic macromolecules, this type of NHS ester-based reaction (reaction 1) is frequently utilized. (see Figure 15).
in the adsorption processes of flavonoid to CNTs functionalized with polyamidoamine; however, adsorption capabilities can be improved with NHS and EDC [28]. In the process of functionalization of NPs with amine groups by the addition of NHS, the carboxyl groups are activated and by reaction with the exposed amines of the proteins that are present on the surface, the molecule binds to the cell surface. For bioconjugated molecules, as well as for the conjugation of small molecules and synthetic macromolecules, this type of NHS ester-based reaction (reaction 1) is frequently utilized. (see Figure  15).  An amide bond is usually formed with a protein or other molecule through aqueous two-phase coupling using EDC and NHS or sulfo-NHS. This involves the formation of a sulfo-NHS ester intermediate that has a higher reactivity than the EDC of the starting reagent for amine coupling. The utilization of NHS esters or reactive acyl groups of imidazole in organic solvent activation processes can be employed when there are stable particles in the solvent to obtain the same product with an amine-containing molecule [91].
Amide or amine groups can be produced by several covalent reactions, as depicted in Figure 16. The acyl azides are activated carboxylates that react with primary amines to form amide bonds. Nucleophilic action on the electron-deficient carbonyl group occurs in the reaction between the acyl azide and an amino group (reaction 2). The higher the pH of the reaction medium, the higher the reactivity for both amine reactivity and hydrolysis, which leads to competition between acyl azide reactions and hydrolysis. An N-hydroxysuccinimide (NHS) ester is the main activating chemical needed to produce reactive acylating agents, since, for example, it forms an acylated product as in reaction 3, with compounds containing the ester-NHS.
in the solvent to obtain the same product with an amine-containing molecule [91].
Amide or amine groups can be produced by several covalent reactions, as de in Figure 16. The acyl azides are activated carboxylates that react with primary am form amide bonds. Nucleophilic action on the electron-deficient carbonyl group oc the reaction between the acyl azide and an amino group (reaction 2). The higher of the reaction medium, the higher the reactivity for both amine reactivity and hyd which leads to competition between acyl azide reactions and hydrolysis. An N-hy succinimide (NHS) ester is the main activating chemical needed to produce reactiv ating agents, since, for example, it forms an acylated product as in reaction 3, wit pounds containing the ester-NHS. When sodium borohydride or sodium cyanoborohydride is added to a reactio taining an aldehyde and an amine as illustrated in Figure 16, a reduction of the Sch intermediate will occur, giving rise to a secondary amine bond (reaction 4) and a s ary amine, thioether, or ether bond formation occurs when an epoxy or oxirane reacts with primary amines, sulfhydryls, or hydroxyl groups, respectively, in a reac which the opening of the β-ring gives rise to a hydroxyl group in the epoxy com (reaction 5). In the reaction of succinic anhydride with a nucleophile, the anhydro opened and an acylated product containing a carboxylate group is formed (reactio most cases the fluorophenyl ester is more stable in an aqueous solution against hydr When sodium borohydride or sodium cyanoborohydride is added to a reaction containing an aldehyde and an amine as illustrated in Figure 16, a reduction of the Schiff base intermediate will occur, giving rise to a secondary amine bond (reaction 4) and a secondary amine, thioether, or ether bond formation occurs when an epoxy or oxirane group reacts with primary amines, sulfhydryls, or hydroxyl groups, respectively, in a reaction in which the opening of the β-ring gives rise to a hydroxyl group in the epoxy compound (reaction 5). In the reaction of succinic anhydride with a nucleophile, the anhydro ring is opened and an acylated product containing a carboxylate group is formed (reaction 6). In most cases the fluorophenyl ester is more stable in an aqueous solution against hydrolysis; however, upon reaction with amines at slightly alkaline pH conditions, amide-bonded NHS esters are produced (reaction 7).
In the case of histidine, the nitrogens of the imidazole ring side chain are acylated with the NHS ester, but they hydrolyze rapidly in aqueous media because the presence of imidazole in the reaction buffer increases the rate of hydrolysis. In contrast, stable bonds are produced in the reaction of these esters with primary and secondary amines. This occurs in the coupling of NHS-ester bonds with N-terminal α-amines and lysine chain amines in proteins [91].
Under alkaline conditions (pH 7.2-8.5), the ester-NHS-activated compound reacts with primary amines to form an amide bond and release NHS removed by dialysis or desalting. This reaction is also utilized at room temperature or at 4 • C for 0.5-4 h in a phosphate buffer at pH 7.2-8.0 to modify the primary amines on the cell surface.
Yang et al. (2019) [28] concluded that the ability of polyamidoamine-functionalized MWCNT-COOHs to adsorb flavonoids depends on the number of hydroxyl groups present on the flavonoid molecule.
Molecules that react with the -OH hydroxyl allow this group to be activated and facilitate its coupling to another functional group, forming a stable bond. Carbohydrates, polysaccharides, and glycoproteins can be coupled by reacting with hydroxyl groups. Similarly, organic molecules such as PEG, which contains hydroxyl, can also be conjugated into another compound. This allows these types of reactions to be generally employed in functionalization processes. Soddu et al. (2020) [129] evaluated the effect of the size and surface area of silica NPs and carbon NPs on the processes presented in human plasma. They evaluated the presence of functional groups on the surface of silica and carbon NPs on platelet-dependent and platelet-independent aggregation, platelet activation, and platelet adhesion. Both types of NPs had hydrophilic and negatively charged surfaces. Hydroxyl functional groups were present on the surface of silica NPs while carboxylic acid and phenolic groups were present on the surface of carbon NPs. Both NPs presented a zeta potential between −40 and 70 mV at pH 7.4. They observed that both NPs interacted with plasma proteins forming the corona protein. However, in carbon NPs this protein induces platelet-independent aggregation, but not in silica NPs. This difference is possibly due to the distribution of the protein molecules on the surface.
For their part, Li et al. (2017) [95] functionalized MSNs with amino groups provided by 3-aminopropyltriethoxy silane, for the elaboration of baicalein NPs. They found that after the formation of MSNs, the surface of the NPs was changed from hydroxyl to amino groups, which caused a change in the zeta potential from negative −7.5 mV to positive +7.16 mV (in water). The hydroxyl groups present on the baicalein molecule can form H-bonds with the nitrogen of the amino groups. Such molecular interactions are the basis for the conduction and delivery of drugs from MSNs. Thus, the presence of functional groups on these nanoparticles can affect interactions with the cell.
Hydroxyl groups can be activated by several reaction mechanisms to form covalent bonds (see Figure 17). For interaction with ligands containing hydroxyl, amine, and thiol groups, activation processes with epoxy or vinyl sulfone are required, as well as cyanogen bromide, CDI, and DSC for coupling with amine molecules [91].
In the covalent interactions of hydroxyl groups, these may react with N',N-carbonyl diimidazole (CDI), and the reactive intermediate imidazole carbamate is formed (reaction 8). Hydroxylated molecules react with amines to form stable urethane (N-alkyl carbamate) bonds, in which the amine action releases the imidazole but not the carbonyl (reaction 9). This type of interaction has been utilized both for the activation of chromatographic supports and for the activation of polyethylene glycol utilized for ligand immobilization and modification of amine molecules, respectively [92,93].
In nonaqueous environments, the N',N-disuccinimidyl carbonate (DSC), containing two NHS esters, activates a hydroxyl group forming a succinimidyl carbonate derivative (reaction 10). However, in aqueous environments, DSC forms by hydrolysis of two N-hydroxysuccinimide (NHS) molecules with the release of CO 2 . The DSC-activated hydroxylated molecules can react with amine compounds, forming urethane-derived bonds or carbamate bonds (reaction 11) in stable cross-linked compounds. In non-aqueous environments, N-hydroxysuccinimidyl chloroformate can also activate the hydroxyl groups (reaction 12), and the reaction proceeds in the same manner as reaction 9. hydroxysuccinimide (NHS) molecules with the release of CO2. The DSC-activated ylated molecules can react with amine compounds, forming urethane-derived b carbamate bonds (reaction 11) in stable cross-linked compounds. In non-aqueous e ments, N-hydroxysuccinimidyl chloroformate can also activate the hydroxyl gro action 12), and the reaction proceeds in the same manner as reaction 9. The formation of a urethane bond (carbamate) occurs by the rearrangement azides to form isocyanate and the reaction of this with a hydroxylated compound (r 13).
Guzman-Mendoza et al. (2022) [123] functionalized MWCNT-COOH insul MWCNT purification with HNO3/H2SO4. They utilized insulin since the threonine present at the C-terminus of the protein binds through the hydroxyl groups of th to the COOH groups on the MWCNT surface. The FT-IR results presented severa (Table 7), in addition peaks at 2875-2950 cm −1 correspond to C-H and a peak at 7 corresponds to N-H of the protein residues.

Conclusions
Size variation, zeta potential, and PDI depend on the composition of the NPs. flavones 7,8-dihydroxyflavone (tropoflavin), 5,6,7-trihydroxyflavone (baicalein), tetrahydroxyflavone (luteolin), and 4′,5,7-trihydroxyflavone (apigenin), the inclu poly(lactic-co-glycolic acid)] (PLGA) in both liposome and NP processing increa size and entrapment efficiency. According to the analysis of flavone NPs, it was o that when the composition of the NPs includes several compounds, a low potenti tained, which indicates the formation of aggregates. The NP release percentage d on the composition of the NPs and the pH of the medium, being more efficient functionalized with proteins and at high pH. A non-covalent H-bond, π-cation a stacking-type molecular interactions, as well as covalent interactions with -OH C=C, and C-H functional groups are characteristic of flavone NPs. The functiona of MWCNTs with acids alters the structure of MWCNTs, unlike MWCNTs functio The formation of a urethane bond (carbamate) occurs by the rearrangement of acyl azides to form isocyanate and the reaction of this with a hydroxylated compound (reaction 13).
Guzman-Mendoza et al. (2022) [123] functionalized MWCNT-COOH insulin from MWCNT purification with HNO 3 /H 2 SO 4 . They utilized insulin since the threonine residue present at the C-terminus of the protein binds through the hydroxyl groups of the lipase to the COOH groups on the MWCNT surface. The FT-IR results presented several peaks (Table 7), in addition peaks at 2875-2950 cm −1 correspond to C-H and a peak at 700 cm −1 corresponds to N-H of the protein residues.

Conclusions
Size variation, zeta potential, and PDI depend on the composition of the NPs. For the flavones 7,8-dihydroxyflavone (tropoflavin), 5,6,7-trihydroxyflavone (baicalein), 3 ,4 ,5,7tetrahydroxyflavone (luteolin), and 4 ,5,7-trihydroxyflavone (apigenin), the inclusion of poly(lactic-co-glycolic acid)] (PLGA) in both liposome and NP processing increases the size and entrapment efficiency. According to the analysis of flavone NPs, it was observed that when the composition of the NPs includes several compounds, a low potential is obtained, which indicates the formation of aggregates. The NP release percentage depends on the composition of the NPs and the pH of the medium, being more efficient in NPs functionalized with proteins and at high pH. A non-covalent H-bond, π-cation and π-π stacking-type molecular interactions, as well as covalent interactions with -OH, -NH 2 , C=C, and C-H functional groups are characteristic of flavone NPs. The functionalization of MWCNTs with acids alters the structure of MWCNTs, unlike MWCNTs functionalized with flavonoids and proteins. The percentage adsorption of flavonoids on MWCNT-COOH generally varies with pH; a higher percentage of flavonoid adsorption is obtained at low pH. The presence of an OH group on the surface of the NPs enhances properties for adsorption or bioconjugation processes with flavones. Some mechanisms of non-covalent and covalent interaction of NPs with amino -NH 2 and hydroxyl -OH functional groups have been proposed.
Funding: This research received no external funding.