pH-responsive nanocarriers are the most widely used drug delivery systems in the oral cavity. To enable pH-responsive release, nanocarriers are generally fabricated with pH responsive groups such as amines or acid-labile bonds. pH changes induce the protonation/deprotonation or lysis of the chemical bonds, resulting in drug release [27]. Various polymers are used for the preparation of nanocarriers, and the following sections categorise nanocarriers by their polymer composition.

DMAEMA (dimethylaminoethyl methacrylate) is a cationic monomer and its polymer (p(DMAEMA)) has often been used as a drug delivery vehicle due to its pH-responsive properties. The chemical structure of DMAEMA contains a tertiary amine group which has a pKa of 7.5. It undergoes structural changes (swelling) in acidic conditions due to the protonation of the tertiary amino moity, and becomes hydrophilic. When the environmental pH is above its pKa, the amine group will be deprotonated and becomes hydrophobic [28,29], which is an important property in loading hydrophobic drugs.

Horev et al. (2015) and Zhou et al. (2016) used a two- step reversible addition–fragmentation chain transfer (RAFT) polymerization method to synthesise a delivery vehicle which allowed the modulation of the polymer molecule weights and polydispersity indices (PDI) [30,31]. In the first step, poly (dimethylaminoethyl methacrylate) p(DMAEMA) was synthesised by introducing DMAEMA and dimethylformamide (DMF) into a reaction vessel with a chain transfer agent (CTA). The initiator (2,2-azobisisobutyronitrile) was then added, and the polymerization reaction occurred at 60 °C for 6 h. In the next step, p(DMAEMA) was used as a macro CTA and crosslinked with DMAEMA, 2-propylacrylic acid (PAA), and butyl methacrylate (BMA) monomers (25:25:50%, respectively) to form p(DMAEMA)-b-p(DMAEMA-co-BMA-co-PAA) diblock copolymers with 2,2-Azobisisobutyronitrile (AIBN) as the initiator. The copolymers were then self-assembled into micelles due to the hydrophobic interactions among BMA residues in the p(DMAEMA-co-BMAco-PAA) core blocks.

The loading of farnesol inside the micelles was achieved via sonication, with a loading ratio up to 22 wt% and a loading efficiency of 100%. Farnesol is a hydrophobic antibacterial agent, but has limited antibiofilm effects after topical applications. The pH-responsive delivery vehicle was therefore used to enhance drug efficacy through high affinity binding and pH-responsive drug release. The farnesol release half-life was 7 h and 15 h at pH 4.5 and pH 7.2, respectively, indicating pH-responsive release behaviour. Moreover, nearly all farnesol (75%) was released within 12 h at pH 4.5, in contrast to the 30 h required for complete release at pH 7.2. At physiological pH (pH 7.2), DMAEMA was protonated and PAA was deprotonated, creating an amphiphilic core for farnesol, while at an acidic pH (pH 4.5) DMAEMA was completely protonated and PAA was neutralized. Electrostatic repulsion occurs due to protonation between the micelle corona and the core, leading to the disruption of micelle structures and drug release. Significant antibiofilm activity of farnesol-loaded nanoparticles was observed, with an 80% decrease in the S. mutans biofilms’ viability compared to a modest 20% reduction in the free farnesol treated group. In addition, nanocarriers by themselves did not show any antibacterial/antibiofilm activity. The enhanced antibiofilm effect was likely due to the greater binding between the positively charged nanoparticles at acidic conditions and the negatively charged microbial surfaces, as well as the pH-triggered drug release, which prolonged the farnesol retention and increased drug bioavailability at the pellicle–biofilm interface as well as inside the biofilm [30,31].

Peng et al. (2022) also used DMAEMA to synthesise poly(DMAEMA-co-HEMA) to encapsulate chlorhexidine (CHX); the overall structure was known as p(DH)@CHX, and the loading capacity and encapsulation efficiencies of CHX in poly(DMAEMA-co-HEMA) were 16.03% and 80.15%, respectively. When placed in an acidic environment, the amine groups in the DMAEMA polymer become protonated, resulting in structural changes (swelling) and causing the subsequent release of CHX. Results from the release study confirmed that p(DH)@CHX released CHX significantly faster in the acidic environment than in the neutral environment, with 60% of CHX released in the acidic condition compared to 30% in pH 7.4 after one hour [29]. CHX is considered a gold standard antiplaque agent (mouth rinse) in dental clinical practice and has been widely used for its broad-spectrum antibacterial effects [32]. Nevertheless, the long term application of CHX is not recommended due to several disadvantages, including cytotoxicity, and few extenuating effects on oral biofilm [33,34,35]. In the study, the p(DH)@CHX retained the same antibacterial/antibiofilm effect as free CHX, and both reduced lactic acid production and biofilm viability to 80%, respectively, compared to the control, yet no significant improvement was credited for using the delivery vehicle, except for a lower cytotoxicity. Therefore, more efforts are required to optimize the current delivery system and to enhance adhesion to the biofilm [36].

Poly (ethylene glycol) (PEG) is the one of the most frequently used hydrophilic polymers used in drug delivery and nanotechnology. Considered a gold standard with “stealth” properties, it has fewer interactions with blood components, thus avoiding the rapid blood clearance seen with other drug carriers, and has excellent biocompatibility [37,38]. Furthermore, PEG can be used to stabilize polymeric nanocarriers as it decreases the chance of agglomeration of the particles via steric stabilization, increasing the stability of the produced formulations during storage and application [39,40]. In addition, PEG is highly soluble in organic solvents, making modification in the end-group comparatively simple, and it is also suitable for biological applications because of its water solubility and low intrinsic toxicity [37]. Due to these superior properties, self-assembled micelles using poly (ethylene glycol) (PEG) as the corona-forming block have been investigated further [41], as they can be modified with pH-responsive groups or acid-labile bonds which respond to low pH, causing structural changes that release the drug.

Zhao et al. [42] synthesised a cationic poly (ethylene glycol)-block-poly(2-(((2-aminoethyl) carbamoyl) oxy) ethyl methacrylate) (PEG-b-PAECOEMA), then modified with citraconic anhydride (CA) to form negatively charged PEG-b-PAECOEMA/CA as the pH-responsive delivery vehicle for CHX. PEG-b-PAECOEMA/CA could encapsulate cationic CHX via electrostatic interactions and self-assemble into core–shell polyionic complex micelles (PICMs). In this structure, the neutral PEG block was used for the stabilization of the polymer and the PAECOEMA/CA block as the pH sensitive group. In neutral conditions, the PAECOEMA/CA block was negatively charged because of the carboxylate groups at the end, facilitating the encapsulation of the cationic CHX inside the core via electrostatic interactions, and exposing PEG as the shell. The loading efficiency and encapsulation efficiency of CHX in CA-PICMs were detected at 16.48% and 75.02%, respectively. In a healthy oral microenvironment, the synthesised PICMs were relatively stable, but in acidic environments, citraconic amide was degraded and transformed to a positively charged primary amine, facilitating the rapid release of CHX from the micelles via electrostatic repulsion. Nearly 69% of CHX was released from CA-PICMs in the first 3 h, compared to 39% at pH 7.4 during the same interval, confirming the pH responsive release profile of CHX from PICMs. Furthermore, they also demonstrated the pH-responsive bacterial killing of CA-PICMs on S. mutans biofilms grown on hydroxyapatite (HA) discs, which was represented by an abundance of dead bacteria (dyed red) after the treatment with CA-PICMs via live/dead bacterial staining, with a similar dead/live bacteria ratio to the CHX group. Although no antibacterial effect was found in the delivery vehicle (PEG-b-PAECOEMA/CA), CA-PICMs were less cytotoxic to human oral keratinocyte (HOK) cells compared to the corresponding free CHX, which greatly reduced the negative impact of CHX on oral tissues.

Zhang et al. (2021) used mPEG-b-PDPA to produce a pH-responsive core–shell nano micelle [44]. 2-(Diisopropylamino) ethyl methacrylate (DPA) copolymer was used to form the hydrophobic core for loading bedaquiline (a hydrophobic antibacterial agent). The loading ratio and encapsulation efficiency of bedaquiline in mPEG-b-PDPA were 37 and 92.5%, respectively. The micelles acted as a pH-responsive agent that shifted the core from hydrophobic to hydrophilic via protonation as the pH dropped below 6. This caused the swelling and disassembly of the micelles and subsequent drug release. At pH 5, the cumulative release of bedaquiline from micelles was 92.2% in 3 h, but no more than 35% was detected at pH 7 in the first 12 h. This nanocarrier not only enabled the targeted release of the therapeutic agent under acidic environments, it also improved the working concentration in the local acidic area to exert a greater antibacterial/antibiofilm effect. In the antibacterial model at pH 5, free micelles did not show an inhibitory effect against planktonic S. mutans, while the 1% bedaquiline-loaded micelles group (equivalent to ~25 μg mL^–1 bedaquiline) significantly reduced the growth of S. mutans to less than 50% of the control. In addition, the live/dead staining images of the S. mutans biofilm revealed that most bacteria in the biofilm were dead (dyed red) in the 1% bedaquiline-loaded micelles group, as opposed to the massive green cells observed in the control and free micelle groups. Cytotoxicity studies using periodontal ligament stem cells showed that the micelles were not toxic, suggesting a good potential for clinical application for the treatment of dental caries.

PEG-based polymers can also be modified with acid liable bonds, which are cleaved upon changes in pH and release therapeutic agents. For example, Xu et al. (2023) [45] developed MAL-PEG-b-PLL/PBA-based polymeric micelles (PM) and loaded them with sodium fluoride (NaF) and tannic acid (TA) for the treatment of dental caries. In this nano-construct, the phenylboronic acid group in PBA forms a pH-sensitive boric acid ester bond with tannic acid via phenylboronic acid-catechol interactions, which can be cleaved under acidic (cariogenic) conditions. Poly(l-lysine) (PLL) is positively charged at physiological pH and interacts with TA (negatively charged) to form core–shell structure micelles. NaF was co-loaded during the micelles’ assembly. The loading content (LC) and loading efficiency (LE) of NaF in the nanocarrier were 5.5% and 20.9%, respectively. For TA, the LC and LE were 8.3% and 31.7%, respectively. The pH-responsive cleavage of the boronate ester enabled a controlled release of TA and NaF, the release rate for which were 70% and 80%, respectively, within 24 h at pH 5.0, much faster than those at pH 7.4 (45% and 50%, respectively). The constructed nanoparticles (PMs@NaF-SAP) were also coated with the salivary-acquired peptide DpSpSEEK (SAP) to enable selective binding to the tooth enamel surface. The pH-responsive antibacterial/antibiofilm effect of PMs@NaF-SAP was comparable to CHX (ca. 25% of reduction in biofilm formation compared to control), but with less cytotoxicity, and produced minimal changes in the oral microbiota diversity.

Yi et al. (2020) exploited acid-labile bonds as a pH-responsive release mechanism for PPi-PEGhyd-Far polymeric micelles [47]. Farnesal (Far) is a hydrophobic derivative of farnesol and has proven anti-caries efficacy [48,49]. To increase the therapeutic bioavailability, Farnesal was linked to PEG via an acid-labile hydrazone bond to form PEG-hyd-Far for enhanced solubility, which was then further conjugated to pyrophosphate (PPi) and self-assembly to form PPi-Far-PMs for enamel-targeting delivery. The HPLC method was utilised for determining the Far loading and encapsulation efficiencies in PPi-Far-PMs, which were 9.51 ± 0.40% and 78.30 ± 1.40%, respectively. The pH responsive release profile of PPi-Far-PMs was confirmed, with 90% of Far being released within 24 h at pH 4.5 in contrast to 40.6% at pH 7.4. Additionally, the in vivo anti-caries study showed that blank PMs had no antibacterial effect, while PPi-Far-PMs significantly reduced the amount of S. mutans compared to CHX (less than 30%). The collective evidence suggested that this delivery system could be used for the targeting delivery of antibacterial agents in the oral environment, and also as a potential treatment or prevention tool for dental caries.

Chitosan (CS) is a copolymer that is produced by deacetylating chitin in the presence of alkaline chemicals [50]. It is the only known naturally occurring polycationic polysaccharide that can form complexes with anionic molecules [51,52,53] due to its biodegradability and biocompatibility. These features have inspired innovative nanotechnology and glycol-chemistry to produce CS-based nanoparticles as a promising drug delivery vehicle in biomedical and pharmaceutical industries [54,55,56].

Nguyen et al. (2017) reported the preparation of stable, spherical, and monodisperse CS nanoparticles for loading with NaF using sodium tripolyphosphate (TPP) as a crosslinker. The loading capacities of fluoride for chitosan nanoparticles prepared in 0.2% NaF and 0.4% NaF were determined at 33 and 113 ppm (μg/g), respectively, and the corresponding entrapment efficiencies (%EE) were 3.6 and 6.2%, respectively. A high level (55% at pH 5 and 43% at pH 7 at 24 h) of fluoride release was observed at pH 5 compared to pH 7 [57]. The same method was applied by Zhu’s group for the preparation of histatin (HTNs)-loaded CS nanoparticles, which was based on their transition from liquid to gel via ionic interactions with a polyanion [58]. Briefly, histatin (HTN3), which has demonstrated properties in tooth homeostasis and dental caries prevention, was first mixed with CS solution. The mixture then went through ionic gelation with TPP to form HTN3-loaded CS nanoparticles. Results showed that the loading ratio of HTN3 in the CS nanoparticles was tunable, without altering the particle size or dispersity. And the pH responsive release profile of CS nanoparticles was confirmed, as they selectively swelled under acidic conditions, which accelerated the release of HTN3 (58% ± 9% at pH 4 and 2% ± 2% at pH 6.8). However, CS nanoparticles loaded with or without HTN3 showed a significant reduction in bacterial viability (by half) compared to the control and almost the same antibiofilm effects against S. mutans with a lower biofilm mass. Specifically, the average wet biofilm mass was detected at a descending trend, i.e., 15 ± 2 mg for the control group, 12 ± 1 mg for HTN3, 8 ± 2 mg for fluoride, 7 ± 1 mg for unloaded CNs, and 6 ± 1 mg for HTN3-loaded CNs. This result indicated the bacterial inhibitory effects of CNs by themselves, and no significant contribution of HTN3 was observed for the antibiofilm/antibacterial activity. Even though CNs can be considered a potential nanocarrier, future studies could consider optimizing the mass ratio between CNs and HTN3 to increase the loading amount of HTN3 in CNs for a better synergistic antibacterial/antibiofilm effect.

Mesoporous silica nanoparticles (MSNs) are a group of inorganic porous materials that have been extensively studied as a drug delivery platform since 2001 [59,60,61,62]. This is due to their favourable properties, including tuneable pore size, large pore volumes that allow for high cargo loading, and high specific surface area, which facilitates surface functionalization [63,64,65,66,67]. Through surface modification, MSNs can be engineered to be stimuli-responsive controlled release systems, with the release of medications triggered by intracellular stimuli or the alterations in the microenvironment of the diseased sites [68,69,70,71]. For the treatment of dental caries, MSN-based pH-responsive delivery is of special interest as a stimuli-responsive system.

Akram et al. (2021) synthesized poly-L-glycolic acid (PLGA)-grafted MSNs and incorporated chlorhexidine (CHX) [72]. A typical method using cetyltrimethylammonium bromide (CTAB) as a structure-directing agent/template and tetraethylorthosilicate (TEOS) as the silica source was used for synthesis. This was followed by functionalization with L-glutamic acid γ-benzyl ester (BLG) on the surface to form PLGA-grafted MSNs (MSN-PLGA). CHX was then incorporated into MSN-PLGA via simple immersion, sonication, and centrifugation. The loading and encapsulation efficiency were 24% and 96%, respectively. A pH-dependent release of CHX from the CHX-loaded/MSN-PLGA was observed, and the cumulative release of CHX reached up to 70% at pH 5.0 after 24 h, substantially higher than that at pH 7.4 (49%). The difference in release profiles was attributed to the tendency of both MSN and PLGA to fracture at low pH [73]. The antibiofilm effect was evaluated using an MTT assay, and the best performance was demonstrated by the 50:50:50 CHX-loaded/MSN-PGA-treated group after 24 h, for which the S. mutans biofilm viability was decreased to less than 20% compared to 90% in the MSN-PLGA treated group. In addition, the biofilm viability remained lower than 30% even after 30 days, indicating a continuous release of CHX at a low pH. All experimental nanoparticles demonstrated low cytotoxicity profiles, with the viability levels of the treated dental pulp stem cells (DPSCs) all above 80%. Although the current results suggested that the MSN-PLGA was promising as a pH-responsive nanocarrier delivery system for CHX, further research is recommended to include multi-species bacterial biofilms to confirm antimicrobial activity. In addition, future research will also investigate whether resin infiltration has an impact on the release profile of the nanoparticles after mixing with commercial dentin adhesive systems.

Fullriede et al. (2016) developed poly(4-vinylpyridine) (PVP)-modified nanoporous silica nanoparticles (NPSNPs) for the delivery of CHX [74] and tested their antibacterial activity against S. mutans and Staphylococcus aureus (S. aureus). The methodology for loading CHX into a nanocarrier was via the simple incubation of the CHX solution with nanoparticles for 3 days, and D-gluconic acid was used to adjust the pH to 3 to open the pores. The PVP-modified NPSNPs were able to incorporate 24 wt% CHX. PVP also served as a pH-responsive gatekeeper, as it blocked the pore openings under physiological pH and prevented the release of CHX. However, in acidic conditions (e.g., bacterial infection), the PVP polymer chains became protonated and repelled each other due to electrostatic repulsion, leading to opened pores and CHX being released. This assumption was later confirmed by the cumulative release of CHX at pH 4 compared to pH 7.4. A pH-responsive burst release of CHX occurred in the first 12 h at pH 4 and reached 260 μg mg⁻¹; this was higher than the 220 μg mg⁻¹ released at pH 7.4. Following the burst release, a constant release of small amounts of CHX was observed at both pH 4 and 7. Antibacterial efficacy was performed on S. mutans and S. aureus planktonic suspensions as well as biofilms, and the results showed a strong antibacterial effect against both planktonic suspensions at higher concentrations than 5 μg mL⁻¹, reducing the bacterial viability to less than 20%, while no obvious antibacterial effect was observed for the delivery vehicle. However, no antibiofilm effect was found against the corresponding mature biofilms across all experimental concentrations of CHX-loaded NPSNPs. The cytotoxicity study of gingival fibroblasts revealed that the cytocompatibility of NPSNPs loaded with CHX was up to 25 μg mL⁻¹. Although the biological evaluation of CHX-loaded NPSNPs provided a therapeutic window where fibroblasts were still viable at a bacterial inhibition concentration, more optimization work is still required. For example, an increased loaded amount of CHX in combination with a better chemical crosslinker to compact the polymer corona may explain the poor antibacterial effect against biofilms and the unwanted release of CHX at physiological pH.

To achieve maximum antibacterial efficacy and given that the oral biofilm was characterized for locally acidic and reducing microenvironments, Lu’s group (2018) incorporated redox-active disulfide bonds into the mesoporous silica nanoparticle (MSNs) framework for the dual delivery of both silver nanoparticles and CHX [75]. The developed Ag-MSNs had a large pore size and therefore were able to load higher levels of CHX (21.5% ± 2.2%). The release of CHX and Ag ions happened in a GSH- and pH-responsive manner, where 55% of CHX and 0.5 ppm of Ag ion were released at pH 5.5 with 5 mM of GSH after 24 h compared to 15%, and 0.15 ppm at pH 7.4 without GSH. This supported the assumption that the disintegration of the Ag-MSN matrix consisting of disulfide bridges was accelerated under acidic and reducing conditions. S. mutans biofilm was used as a model for the investigation of the antibacterial/antibiofilm activity using Ag-MSNs@CHX, the minimum inhibitory concentration (MIC) was 12.5 µg/mL, and the minimum bactericidal concentration (MBC) was 25 µg/mL. For the antibiofilm test, the Ag-MSNs@CHX group demonstrated the highest biofilm inhibition, with biofilm viability reduced to around 20% at the minimal biofilm inhibitory concentration (MBIC) (50 µg/mL) compared to the control (100% biofilm viability) and Ag-MSNs groups (50%), and the inhibition occurred over 72 h. This indicated a strong synergistic antibiofilm effect due to the continuous release of CHX and silver ion over the long term. Cytotoxicity studies with human immortalized oral epithelial cells (HIOECs) showed that the encapsulation of CHX in the Ag-MSNs nanocarrier could significantly reduce the toxicity of CHX, although the Ag-MSNs@CHX-treated group only retained 60% cell viability after 6 h incubation.

Secondary caries has frequently been observed in resin composite restorations [76,77], characterized by resin surface degradation which increased surface roughness and decreased hardness; at the same time, the unpolymerized monomer and dentin binding agent were eluted from composites, promoting the growth of cariogenic microorganisms [78,79,80,81].

To address this problem, Liang et al. (2020) synthesised two monomers based on tertiary amine (TA), DMAEM (dodecylmethylaminoethyl methacrylate) and HMAEM (hexadecylmethylaminoethyl methacrylate), and incorporated them into an adhesive resin to make TA-modified resins (TA@RAs). TA@RAs showed antibacterial effects only in an acidic environment, as the nitrogen atoms of TA were protonated in low pH and formed quaternary ammonium monomers (QAMs) for bacterial killing [82,83]. The pH-responsive antibiofilm effect of TA@RAs was observed against S. mutans biofilms, with bacterial viability significantly decreased to less than 20% at a low pH compared to that at a pH above 5.5 (more than 70%). 16S rRNA gene sequencing results showed significantly higher microbial diversity in the DMAEM/HMAEM group than in the control group, which further confirmed the pH-responsive antibacterial effect of TA@RAs, indicating the potential of TA@RAs in preventing secondary caries and shifting the oral microbial community towards a healthy and balanced condition [84].