Cerium and Its Oxidant-Based Nanomaterials for Antibacterial Applications: A State-of-the-Art Review

Infectious diseases caused by bacteria remain a serious and global problem in human health. Herein, the discovery and application of antibiotics, which has attracted much attention in the past years, has helped to cure numerous infectious diseases and made huge contributions in the biomedical field. However, the widespread use of the broad-spectrum antibiotics has led to serious drug resistance for many human bacterial pathogens. In particular, bacterial drug resistance decreases therapeutic effects and leads to high mortality. Inspiringly, the introduction of nanomaterials in the biomedical field makes it possible to overcome the difficulty in bacterial drug resistance attributed to their unique antibacterial mechanism. Recently, a variety of metal- and metal oxide-based nanomaterials have been fully integrated in antibacterial applications and achieved excellent performances. Among them, cerium- and cerium oxide-based nanomaterials have received much attention worldwide. Remarkably, cerium oxide nanoparticles with lower toxicity act as effective antibacterial agents owing to their unique functional mechanism against pathogens through the reversible conversion of oxidation state between Ce(III) and Ce(IV). This article provides an overview of the state-of-the-art design and antibacterial applications of cerium- and cerium oxide-based materials. The first part discusses the underlying antibacterial mechanisms for cerium- and cerium oxide-based materials currently applied in biomedicine. The second part focuses on various antibacterial-related materials and their applications. In addition, the existing problems and future perspectives of the cerium- and cerium oxide-based materials make up the third part of this review. This paper could provide the possible mechanisms for antibacterial activities, various designs of cerium- and cerium oxide-based materials, and related antibacterial properties.


INTRODUCTION
During the past few hundred years, infectious diseases have been a persistent global and critical health problem, which has attracted considerable public attention worldwide as a human health threat (Jones et al., 2008;Fauci and Morens, 2012). Due to the high morbidity and mortality burden of infectious diseases, lots of investigations have focused on the aspects of understanding, controlling, treating, and preventing them for centuries (Casadevall et al., 2004;Matsuzaki et al., 2005;Fauci and Morens, 2012;Caliendo et al., 2013). Since the discovery of penicillin, which could be regarded as a significant contribution in the medical field (Gaynes, 2017), the therapeutic method of antibiotics has played an important role in addressing the issue of human infectious diseases (Nigam et al., 2014). However, the widespread use and abuse of antibiotics have caused the emergence of drug resistance, accompanied with limited treatment efficacy (D'Costa et al., 2011;Frieri et al., 2017). Therefore, it is urgent to develop novel antimicrobial agents to tackle the current challenges in treating infectious diseases, which can kill bacteria efficiently and with biosafety.
Recently, the development of nanomaterials and nanotechnology, which have already been used in a wide range of applications such as in agriculture (Khot et al., 2012), the environment , analytical chemistry (Scida et al., 2011), and electronics and sensors (Chaudhary and Mehta, 2014), has attracted great interest. Owing to the specific structure and tailorable physicochemical properties of nanomaterials, they are regarded as promising materials utilized in the fields of biomedical research (Mitragotri et al., 2015). There are plenty of investigations on metal and metal oxide nanoparticles (NPs), such as silver (Franci et al., 2015), copper (Rodhe et al., 2015), copper oxide (CuO) (Ivask et al., 2014), titanium dioxide (TiO 2 ) (Jayaseelan et al., 2013), and zinc oxide (ZnO) (Sirelkhatim et al., 2015), which all showed outstanding antibacterial activities. The application of antibacterial agents against infectious diseases needs not only excellent bacterial killing efficacy but also lower or even no toxicity to the human normal cells. However, the biocompatibility of most nanomaterials is not ideal. For example, silver NPs with good antibacterial effect have been incorporated into some materials serving as an antibacterial component, but they have serious biosafety problems and easily causes cytotoxicity, which have limited their applications for human health care (Ávalos Fúnez et al., 2013).
Cerium dioxide (CeO 2 ), as a rare-earth material, has a variety of properties at the nanoscale and can be widely used as polishing agents , solid oxide fuel cells Sun et al., 2018), catalysts , solar cells (Hu et al., 2018), and pharmacological agents (Charbgoo et al., 2017). Due to their lower toxicity to mammalian cells and unique antibacterial mechanism, CeO 2 NPs have also been widely applied in biomedical sciences such as in antitumor (Corsi et al., 2018), anti-inflammation , and antibacterial activities , in combating neurodegenerative diseases (Naz et al., 2017), and as immunosensors (Naz et al., 2017). Remarkably, CeO 2 have shown promising approaches to circumvent the existing problems of drug-resistant bacteria and served as excellent antibacterial agents in biology and medical sciences in comparison with other metal oxides . However, it was found that CeO 2 NPs at low concentrations (1 mg/L) could not destroy the bacterial biofilms (Hou et al., 2015). Therefore, various cerium-and cerium oxide-based nanomaterials including doping metal ions (Atif et al., 2019), antibacterial element Au/Ag decorations (Babu et al., 2014;Wang et al., 2014), nanocontainers (Gagnon et al., 2015), dopants (Bomila et al., 2018), and a component of blend (Pandey et al., 2018a) were developed to optimize antimicrobial performance as well as physical and biological properties (Anastasiou et al., 2019;Bharathi and Stalin, 2019;Liu et al., 2019).
In recent years, numerous valuable reviews on cerium and its oxides have been published, focusing on the synthesis (Nyoka et al., 2020), catalysis (Walkey et al., 2015), defect engineering (Seal et al., 2020), pharmaceutical properties (Xu and Qu, 2014), and biological and biomedical effects (Dhall and Self, 2018;Kargozar et al., 2018;Zhang et al., 2019;Casals et al., 2020). Nevertheless, the antibacterial effects of cerium and its oxides with different material designs, particularly on infectious diseases, were rarely reviewed. Therefore, this review began with a brief introduction of the antibacterial mechanism of CeO 2 NPs, followed by summarizing the various designs of ceriumand cerium oxide-based nanomaterials. In addition, this article presented a comprehensive overview of new developments in the antibacterial applications of cerium-and cerium oxide-based nanomaterials against infectious diseases and would provide guidance to develop new cerium-related antibiotics in the future.

ANTIBACTERIAL MECHANISM OF CERIUM-BASED MATERIALS
Nowadays, cerium has been verified to possess the potential to serve as an effective and long-lasting biocide for preventing bacterial infections, due to its higher safety to human cells and unique antibacterial mechanism compared with other metal ions. Cerium-and cerium oxide-based antimicrobials have attracted tremendous attention, and many studies have reported various kinds of designs about the cerium-related nanomaterials against microorganisms, which will be detailed in the latter part. The main antibacterial mechanism for CeO 2 can be attributed to the direct contact between CeO 2 and bacterial membranes. The schematic representation of antibacterial mechanism of cerium-based materials is shown in Figure 1.
First, the positively charged NPs were adsorbed onto the membranes of negatively charged Gram-positive and Gramnegative bacteria through electrostatic interaction. Due to the electrostatic interaction and bacterial membrane blocking, the NPs persisted on the surface of bacteria for a long time rather than penetrating the membrane. Afterward, NPs may change the viscosity of the membrane, impair the specific ionic pumps, and eventually strongly change the transport exchanges between the bacterial cell and the solution to disturb bacterial growth (Thill et al., 2006).
Second, CeO 2 could attack proteins after adsorbing on the outer membrane of the bacterial cell. The released cerium ions could alter electron flow and respiration of bacteria (Pelletier et al., 2010) and react with the thiol groups (-SH) or be absorbed onto transporters and/or porins to hamper nutrient transportation (Zeyons et al., 2009;Li et al., 2019b). In addition, the irregular shapes and rough edges of CeO 2 per se contribute to the physical damage of bacterial membranes, especially for Grampositive bacteria Zhang et al., 2019).
Third, oxidative stress is also an important factor for CeO 2 during the antibacterial process. Generally, oxidative stress was induced by the generation of reactive oxygen species (ROS) in vivo, whereas the ROS generated was induced by the reversible conversion between Ce(III) and Ce(IV) on the surface of bacterial membranes . The ROS can attack the nucleic acids, proteins, polysaccharides, lipids, and other biological molecules to cause the loss of their function, eventually killing and decomposing bacteria (Li et al., 2012). Although CeO 2 can be excited to produce ROS by ultraviolet (UV) irradiation, there were very few researches on bacterial activity by using CeO 2 alone. Usually, CeO 2 combined with other photocatalysts like TiO 2 . In the presence of CeO 2 , the band gap can be changed in the host lattices of photocatalysts, which improves the photocatalytic activity of TiO 2 (Kasinathan et al., 2016).
Finally, Ce(IV) ion could induce hydrolysis of a DNA oligomer into the fragments, which then could be successfully manipulated by natural enzymes (Sumaoka et al., 1998). Extracellular DNA (eDNA) is one of the important components in the process of biofilm formation, making the biofilm hard to eliminate. Therefore, taking advantage of Ce-based nanozymes with deoxyribonuclease (DNase) mimetic activity could lead to high cleavage ability toward eDNA and disrupt established biofilms (Chen et al., 2016;Liu et al., 2019).

Cerium Oxide for Antibacterial Application
This part mainly reviewed the antibacterial effect of pure CeO 2 with different morphologies in recent years. The details are summarized in Table 1, and parts of TEM or SEM images of CeO 2 with different morphologies are shown in Figure 2. Sphericalshaped CeO 2 NPs showed antibacterial activity against Gramnegative bacteria Escherichia coli (E. coli) and Pseudomonas aeruginosa (P. aeruginosa) and less or no antibacterial activity against Gram-positive bacteria Staphylococcus aureus (S. aureus) (Ravishankar et al., 2015;Selvaraj et al., 2015;Surendra and Roopan, 2016;Senthilkumar et al., 2019), because of the direct contact onto the membrane surface of E. coli that blocks the synthesis of the cell wall and membrane and disturbs other cellular processes (Senthilkumar et al., 2019). Similarly, it was found that the CeO 2 NPs showed superior activity against Gramnegative E. coli than Gram-positive Bacillus subtilis (B. subtilis), which has a slight reduction in survival (Patil et al., 2016). However, in other studies, the spherical-shaped CeO 2 NPs with a smaller size showed contradictory antibacterial effects Arunachalam et al., 2018). The CeO 2 NPs showed mild and moderate antibacterial behavior against Gram-negative P. aeruginosa and Proteus vulgaris (P. vulgaris) with a zone of inhibition (ZOI) of 4.09 ± 0.22 and 4.38 ± 0.44 mm, respectively, whereas they had highly potent antibacterial behavior against Gram-positive S. aureus and Streptococcus pneumoniae (S. pneumoniae) with a ZOI of 12.43 ± 0.36 and 14.56 ± 0.23 mm, respectively. The authors thought various factors such as smaller-sized particles, rigid morphology, surface interaction, and variation in the band gap energy of CeO 2 NPs contributed to the antimicrobial activity (Arunachalam et al., 2018). Arumugam et al. (2015) thought that the uneven ridges and oxygen defects of CeO 2 NPs led to these inconsistent results. A similar antibacterial activity was found in CeO 2 NPs with irregular morphologies and poor dispersity (Yadav et al., 2016).
CeO 2 microspheres by green synthesis showed significant antibacterial activity against E. coli and S. aureus with ZOI values of 4.67 and 3.33 mm, respectively. The ZOI depends on the type of bacteria, the concentration, surface area, shape, and size of NPs. Besides, the structural and configurational differences of the cell membrane lead to the difference of antibacterial effect against different types of bacteria (Malleshappa et al., 2015). The inhibition effect of CeO 2 nanocube is B. subtilis > Enterococcus faecalis (E. faecalis) = Salmonella typhimurium > E. coli. Moreover, they found that interaction between CeO 2 nanocubes and E. coli led to bacterial cell wall damage or disruption by detected β-D-galactosidase which is present in E. coli (Krishnamoorthy et al., 2014). Interestingly, it was reported that CeO 2 nanosheets exhibited stronger antibacterial activity compared with the CeO 2 NPs. In the synthesis process, the former route is the same, and the CeO 2 nanosheets were formed by hydrothermal treatment, while the CeO 2 NPs were synthesized by washing and annealing. The CeO 2 nanosheets showed a higher surface area, leading to a higher concentration of oxygen vacancies on the surface, which caused enhanced ROS generation. Therefore, CeO 2 nanosheets exhibited stronger antibacterial activity (Abbas et al., 2016).
Living conditions such as solution pH and phosphate solution could also affect the bacteriostatic effect. Dextran-coated CeO 2 NPs were much more effective against P. aeruginosa and Staphylococcus epidermidis (S. epidermidis) at basic pH values (pH = 9) compared with acidic pH values (pH = 6). The size of CeO 2 NPs is smaller and have a positive charge at pH = 9, whereas CeO 2 NPs show a negative charge at pH = 6. These properties at an alkaline pH make CeO 2 NPs easier to absorb on the bacteria and achieve a favorable antibacterial effect (Alpaslan et al., 2017). It was believed that CeO 2 NPs with a lower Ce 3+ /Ce 4+ ratio show antibacterial activity (Gupta et al., 2016), but in a phosphate solution, it showed that CeO 2 NPs in the (III) oxidation state had a significant antibacterial activity other than those in the (IV) oxidation state. Exposure of E. coli to the equimolar mixture of CeO 2 NPs in the (III) oxidation state and phosphate for 3 h resulted in ∼90% reduction. However, the equimolar mixture of CeO 2 NPs in the (IV) oxidation state and phosphate for a long time did not show significant bacterial reduction (Sargia et al., 2017).
Apart from pure CeO 2 , cerium oxide containing some elements also showed antibiotic activity. CeVO 4 NPs showed excellent antibacterial activity against Streptococcus mutans and Streptococcus pyogenes with minimum inhibitory concentration (MIC) values at 200 µg/ml and against Vibrio cholera, Salmonella typhi, and Shigella flexneri with MIC values at 350 µg/ml (Mishra et al., 2020).

Metal Element Doping Into Cerium Oxide NPs for Antibacterial Enhancement
Most transition metals have unfilled d-orbitals and thus are redox active. Their ability to easily cycle between oxidation states contributes to both their catalytic properties and their toxicity   (Palmer and Skaar, 2016). In a recent study, different transition metal ions (Fe 2+ , Mn 2+ , and Co 2+ ) doped with nanosized hydroxyapatite (HA) were prepared for the increase of loadbearing capacity and strength; meanwhile, higher antibacterial activities were achieved against most tested bacteria compared with erythromycin (Panneerselvam et al., 2020). These transition metal elements (Fe, Mn, and Co) were also separately doped into CeO 2 NPs and proven to have enhanced antimicrobial activities Atif et al., 2019;Rahdar et al., 2019). The metal ions were placed into the atomic site of the Ce ions and substituted in the matrix of the host CeO 2 . The bonding of metal ions with CeO 2 NPs was formed Atif et al., 2019;Rahdar et al., 2019). The antibacterial activity of various transition metal-doped metal oxide nanostructures comes from an electrostatic induction nature between NPs and the bacterial cell membrane . The synergetic effects of transition metal doped into CeO 2 NPs increase the antibacterial activity caused by ROS. Doping a certain amount of trivalent rare-earth metals could decrease the lattice parameter to enhance the antibacterial efficiency of CeO 2 NPs. Among the lanthanide series, samarium ion (Sm 3+ ) could increase active oxygen vacancies in the CeO 2 NPs and had properties of highintensity optical, electrical, and antimicrobial activities (Artini et al., 2015). The antibacterial activities were found increased as the amount of metal dopants increased (Artini et al., 2015;Atif et al., 2019;Khadar et al., 2019). However, all of these experiments had a common problem: they did not evaluate or only evaluated the toxicity of cancer cells instead of normal cells.

Antibacterial Element Incorporation/Modification for Antibacterial Amplification
Recently, silver NPs and silver compounds have attracted more and more researchers' attention because of their efficient antimicrobial activities toward both Gram-positive (Li et al., 2020) and Gram-negative (Ramalingam et al., 2016) bacteria as well as fungi (Ballottin et al., 2017). The multifaceted mode of action of Ag + against bacteria makes it impossible to cause a high-level, single-step, target-based mutation to silver resistance (Chopra, 2007). Instead, this property makes sliver-based drugs a potential alternative to antibiotics. CeO 2 nanorods, nanocubes, and NPs were synthesized, and different proportions of Ag + were incorporated in CeO 2 , forming Ag/CeO 2 . CeO 2 nanocubes and nanorods achieved much higher bactericidal activities due to exposed crystal planes and oxidation ability. After loading with a small amount of Ag, the bactericidal activities were largely increased by both extracellular and intracellular ROS. E. coli treated with Ag/CeO 2 nanorods showed the highest relative ROS level, while those treated with Ag/CeO 2 NPs exhibited the lowest relative ROS level. The redox cycle of Ag + /Ag 0 and Ce 3+ /Ce 4+ co-maintained the catalytic process of extracellular ROS formation, and Ag + also promoted the generation of intracellular ROS through respiratory enzymes . Negi et al. (2019) synthesized Ag/CeO 2 nanostructured materials for inhibiting the growth of S. aureus and P. aeruginosa.
The MIC values of Ag/CeO 2 against the aforementioned bacteria were 3.125 and 6.25 µg/ml, respectively. The observation results showed the significantly improved catalytic and antibacterial properties compared with pure CeO 2 , probably since Ag weakened the Ce-O bond situated nearby and accelerated ROS generation to enhance the catalytic reactions (Negi et al., 2019).
Various semiconductors like TiO 2 and CeO 2 could be coupled with compounds derived from silver to enhance charge separation and induce photocatalytic activity under visible light. Eswar et al. (2015) synthesized CeO 2 by two different methods. The low-band-gap CeO 2 synthesized using the polyethylene glycol (PEG)-assisted sonochemical method was named P-CeO 2 . The results of the photocatalytic degradation of E. coli using CeO 2 , CeO 2 /AgBr, CeO 2 /Ag 3 PO 4 , CeO 2 /AgBr/Ag 3 PO 4 , and P-CeO 2 /AgBr/Ag 3 PO 4 composites showed differently in the dark and under visible light (Figure 3). All the composites had at least 1 log colony-forming unit (CFU) reduction in the dark, and after eliminating the photolysis factor, all the composites had at least 3 log CFU extra reduction under visible light. Among them, the CeO 2 /AgBr/Ag 3 PO 4 composite showed a superior bactericidal efficiency compared to other composites  under visible light due to ROS and elution of Ag + into the reaction system (Eswar et al., 2015).
However, most silver-based drugs have the disadvantages of a short-term antimicrobial activity and fast Ag + release and may increase the risk of impairing the surrounding tissues (Gagnon et al., 2016). Gagnon et al. designed CeO 2 nanocontainers that were used to encapsulate silver nitrate and silver NPs (AgNO 3 /CeO 2 and AgNP/CeO 2 nanocontainers). AgNO 3 /CeO 2 nanocontainers had a lower minimum bactericidal concentration (MBC) than AgNP/CeO 2 nanocontainers against E. coli, indicating a higher antibacterial activity. Based on this, another nanocontainer, AgNP/CeO 2 /TiO 2 , was designed to control the Ag + release. After further encapsulation of TiO 2 , only about 7% of total Ag load was released in 3 months, and the remaining Ag release was triggered by the oxidation of silver using nitric acid. Decrease of the pH caused by the bacterial metabolism in the environment could promote the oxidation of Ag and increase the Ag + release. This design could function as antimicrobial coatings to prevent implant infection. However, compared to AgNP/CeO 2 nanocontainers, AgNP/CeO 2 /TiO 2 nanocontainers gained increased cytotoxicity toward a model epithelial barrier cell type (A549 cells) (Gagnon et al., 2016).
Au NPs were regarded as an antibacterial agent with unique biocompatibility. Decorating Au NPs on CeO 2 NPs (Au/CeO 2 ) would achieve a stronger bacterial inhibitory activity than CeO 2 NPs and pure Au NPs. Furthermore, after coculture bacteria with Lactobacillus plantarum (L. plantarum), L. plantarum and Au/CeO 2 NPs possibly act synergistically in inhibiting bacteria (Babu et al., 2014). In monoculture systems, the results showed that Au/CeO 2 NPs exhibited a broad-spectrum effect against four tested bacteria: E. coli, Salmonella enteritidis (S. enteritidis), B. subtilis, and S. aureus. Although ampicillin had a higher inhibiting effect against most bacteria, it failed to show an obvious antibacterial effect against B. subtilis. After coculture bacteria with L. plantarum, Au/CeO 2 NPs showed enhanced antibacterial activities against S. aureus, B. subtilis, and S. enteritidis. With different NPs or antibiotic doses increased, the bacterial concentration decreased at different degrees after incubation in a monoculture or coculture system. However, with the doses of Au NPs increased, it showed greater toxicity toward RAW 264.7 cells, whereas pure CeO 2 did not show any significant toxicity after 48 h (Babu et al., 2014).

Cerium Complexes
Cerium complexes with various ligands possessed different properties, such as antitumor and antimicrobial activities (Sang et al., 2017). Schiff base cerium complexes were reported to have interesting antibacterial activity (Sang et al., 2015(Sang et al., , 2017. The MIC values of Schiff base cerium(IV) complex against B. subtilis, S. aureus, E. coli, and Pseudomonas fluorescens (P. fluorescens) ranged from 1.56 to 12.5 µg/ml. The inhibition activity of this cerium complex is weaker against B. subtilis, while stronger against other tested bacteria compared with penicillin (Sang et al., 2015). The Schiff base cerium(III) complex had stronger antibacterial activities against B. subtilis, S. aureus, E. coli, and P. fluorescens than the free Schiff base, and corresponding MIC values were in the range of 1.56 to 25 µg/ml (Sang et al., 2017). In another study, the Schiff base cerium(III) complex was reported to exhibit antibacterial activity with all MIC values of > 50 µg/ml against S. aureus, K. pneumonia, E. coli, P. vulgaris, and Candida albicans (C. albicans) (El-Shafiy and Shebl, 2018). The lipophilicity of the complex made it easier to penetrate into the lipid membrane and block the metal binding sites on enzymes of bacteria (Sang et al., 2015). In addition, the complex NaCe(MoO 4 ) 2 could act as antibiotic activity "modulators, " increasing the activity of a specific antibiotic for a certain strain. When using sodium-cerium molybdate [NaCe(MoO 4 ) 2 ] alone, the MIC values were over 1,024 µg/ml, showing no antibacterial activity (Moura et al., 2019). NaCe(MoO 4 ) 2 synergistically modulated the activity of gentamicin against S. aureus, reducing the MIC values from 16 to 4 µg/ml, and norfloxacin against E. coli, reducing the MIC values from 8 to 3.17 µg/ml. NaCe(MoO 4 ) 2 showed antagonism when using norfloxacin against S. aureus and gentamicin and imipenem against E. coli together. This resulted from a chelation of the antibiotic by NaCe(MoO 4 ) 2 or a reverse reaction or binding to specific sites within the antibiotics, causing antibiotic spectrum reduction (Moura et al., 2019). The antimicrobial activities of elementdoped CeO 2 , antibacterial element incorporation/modification, and cerium complexes are summarized in Table 2.

CERIUM-AND CERIUM OXIDE-DOPED MATERIALS
Currently, semiconductor materials in nanoscale have drawn much attention because of their unique physical and chemical properties (Suresh, 2013). As such, metallic oxides, such as ZnO and TiO 2 , are used as biocides or disinfecting agents due to their ability to induce the generation of ROS under UV light (Khan et al., 2015). When cerium and cerium oxide were doped into semiconducting nanomaterials, they functioned as inorganic disinfectants, with a higher antibacterial activity than pure semiconducting nanomaterials alone (Bomila et al., 2018). When cerium was doped into some scaffold materials such as HA, bioactive glass (BG), and metal organic frameworks (MOFs), these materials not only showed enhanced antibacterial properties but also achieved improved mechanical properties and biological properties (Gopi et al., 2014;Deliormanlı et al., 2016;Li et al., 2019a). The antimicrobial activities of cerium-and cerium oxide-doped materials are summarized in Table 3.

Cerium and Cerium Oxide Doping Into Metal Oxide
The antibacterial activity of ZnO resulted from the process of oxidative stress. Zn 2+ ions could be absorbed by bacterial cells and inhibit the action of respiratory enzymes in cell membrane by interacting with them. Then they led to ROS formation, causing oxidative stress which led to irreversible damage to the bacterial cell (Mishra et al., 2017). The increased particle surface area, reduced band gap energy, and improved adsorption ability of the particle surfaces and dopant-ZnO interfaces contributed to better optical properties, high luminescence properties, and high photocatalytic activity in the doped semiconducting catalysts (Lee et al., 2016). Bomila et al. (2018) prepared pure and Cedoped ZnO NPs and applied them as an antibacterial agent. With the increase in Ce dopants, the antibacterial effectiveness increased. The maximum Ce-doped ZnO NPs possessed higher antibacterial activity, which was closer to the antibacterial activity of positive control ampicillin. It was found that Ce-doped ZnO NPs had more effective antibacterial property than pure ZnO NPs, and both NPs had a selectively antibacterial activity toward Gram-negative bacteria. Flower-like ZnO had better photocatalytic activity than ZnO with other morphologies, which  is closely related to antimicrobial activity (Chelouche et al., 2014).
With the adding of cerium ion doping, the band gap of ZnO crystallites decreased, which could take advantage of visible-light irradiation and hence produce more ROS (Hui et al., 2016). The 0.8% Ce-doped flower-shaped ZnO crystallites enhanced antimicrobial activity against C. albicans (75%) and Aspergillus flavus (80%) under visible-light sources than pure ZnO crystals (Hui et al., 2016). Reduced graphene oxide (rGO) could improve the adsorption of water and oxygen molecules and react with electron and hole pairs. This process is beneficial to production of excess ROS and direct contact with bacteria, which promote ZnO to form cavities on the bacterial cell. Ce could suppress the electron trapping effect and promote ROS generation. The Ce-doped ZnO/rGO (Ce/ZnO/rGO) with impaired flower-like morphology showed enhanced an antibacterial effect against both Gram-positive B. subtilis and Gram-negative Vibrio harveyi (Vanitha et al., 2018). TiO 2 with low toxicity, good biocompatibility, and high bactericidal and sterilizing effects has become one of the most promising antimicrobial materials (Qi et al., 2019a). However, its wide band gap made it function only in UV irradiation which might cause skin damage and disorders and restricted tissue penetration, seriously influencing the application of TiO 2 in the antibacterial field (Qi et al., 2019a). Doping non-metal ions or metal ions could broaden the visible-light response, which would make it possible for TiO 2 nanomaterials to have a photocatalytic effect under visible-light irradiation. The calcination temperature during the synthesis may affect the antibacterial effect. Wang et al. exchanged ions into the materials and prepared single-Zn-ion-doped TiO 2 (Zn-TiO 2 ) and doubleion-doped TiO 2 (Zn/Ce-TiO 2 , Zn/Y-TiO 2 , and Zn/B-TiO 2 ) at different calcination temperatures. The results showed that only Zn-TiO 2 nanomaterials calcinated at 500 • C had a bactericidal effect due to a large surface area. After doping with Ce, Y, or B ions, the bactericidal activities of the Zn-TiO 2 materials are significantly improved at different calcination temperatures . In their next experiment, TiO 2 materials codoped with B and Ce ions (B/Ce-TiO 2 ) with different calcination temperatures were prepared, and all materials had a large ZOI, suggesting that the materials have excellent antimicrobial activities . Besides the antibacterial effect of Ce, B 2 O 3 was found in the materials and is dissolved in water to produce metabolic acid and boric acid. Boric acid had an inhibitory effect on all kinds of bacteria and fungi. CeO 2 -doped TiO 2 nanocomposites were found to have no inhibitory effect on bacterial growth without UV light. When exposed to UV light, nanocomposites showed inhibition in different degrees. Gram-positive strains were more susceptible than Gram-negative strains. As doping concentrations of TiO 2 increased, the ZOI of both pathogens increased (Kasinathan et al., 2016). Similarly, Moongraksathum and Chen (2018) synthesized anatase TiO 2 codoped with Ag and CeO 2 (Ag/CeO 2 -TiO 2 ). Remarkably, they took full advantage of the photocatalytic properties of TiO 2 and CeO 2 , and the addition of silver could broaden the bactericidal activity of photocatalytic composite materials. After 30 min of illumination with UVA (λ = 365 nm) radiation, the bactericidal efficiency of Ag/CeO 2 -TiO 2 coating could reach more than 99.99% against both E. coli and S. aureus (Moongraksathum and Chen, 2018). The order of antibacterial effectiveness of tested materials against both E. coli and S. aureus was Ag/CeO 2 -TiO 2 > Ag/TiO 2 > CeO 2 -TiO 2 > TiO 2 .
Similar to ZnO and TiO 2 , ZrO 2 is also a significant semiconductor material in the biomedical field. ZrO 2 is an n-type semiconductor with a broad strap gap, and when it is doped with rare-earth metals, it exhibits an outstanding structural stability, high thermal conductivity, and irradiation stability. Ce doping could enhance the ROS generation capability of ZrO 2 NPs. The antibacterial property of Ce-doped ZrO 2 NPs was found  more efficient in Gram-positive bacteria than in Gram-negative bacteria. The ZOI values of Ce-doped ZrO 2 against B. subtilis and S. aureus were 15 and 16 mm, respectively, while those against Klebsiella pneumoniae (K. pneumoniae) and P. aeruginosa were 15 and 11 mm, respectively (Mekala et al., 2018). The Ce-doped nickel oxide (Ce-doped NiO) nanomaterial also achieved excellent antibacterial activity against K. pneumoniae, S. typhi, P. aeruginosa, B. cereus, B. subtilis, and S. aureus (Muthukumaran et al., 2016). Among the samples of Ce-doped NiO with different weight percentages (pure and 1, 3, 5, 7, and 9 wt.%), 5 wt.% Ce-doped NiO, which showed the maximal catalytic peak current during the anodic scan, got the best antibacterial activity against all tested bacteria with the MIC and MBC (Figure 4). Elayakumar et al. (2019a,b)  Besides, MOFs, as effective and promising therapeutic nanomaterials constructed by bridging metal ions with organic linkers, have sparked increasing interest in the field of antibacterial applications, as they have the ability of reserving metal ions and keeping continuous ion release in order to attain bacteriostatic and bactericidal effects (Wyszogrodzka et al., 2016;Zhang et al., 2018;Wang et al., 2019). Recently, the novel multifunctional NPs (ZIF-8 : Ce) were synthesized by doping Ce into the zeolitic imidazole framework-8 (ZIF-8). The materials possessed antibacterial capabilities against two periodontal pathogens Porphyromonas gingivalis (P. gingivalis) and Fusobacterium nucleatum (F. nucleatum) as well as antiinflammatory capabilities (Li et al., 2019a). As shown in Figure 5, ZIF-8 and ZIF-8 : Ce1% NPs obtained about 2 log CFU reduction against F. nucleatum and P. gingivalis. The antibacterial effect was decreased with the increase of Ce substituting quantity. The antimicrobial activity of ZIF-8 was attributed to Zn, Ce, and the disruption of liposome from imidazole. In addition, Ce doping endowed ZIF-8 NPs with superoxide dismutase (SOD) and catalase (CAT) enzyme mimic activities for ROS scavenging, which could be beneficial for eliminating the inevitable inflammatory response during antibacterial application (Li et al., 2019a).

Cerium and Cerium Oxide Doping Into Grafting Scaffold
Hydroxyapatite, an essential macroporous inorganic material, has the same component as dentine, enamel, and bones and has several excellent properties. Therefore, HA has been widely studied in tissue engineering as a grafting scaffold and applied in the biomedical field such as dentistry, maxillofacial reconstruction, and implant coatings (Phatai et al., 2018). However, admittedly, pure HA has several disadvantages such as being brittle and weak in intensity and adhesion, dissolution rate, and antibacterial capacity (Phatai et al., 2018). The introduction of Ce 3+ could improve mechanical properties, antibacterial activity, and cell adhesion and proliferation of HAbased composites (Gopi et al., 2014;Morais et al., 2015). Because the electronegativity and radius of Ce 3+ are very close to those of Ca 2+ , Ce 3+ can replace Ca 2+ within the HA lattice. Cedoped HA can stimulate metabolic activity in organisms and improve its antibacterial property (Yuan et al., 2016). Ce-doped glass-reinforced HA composite (GR-HA_Ce) was reported to have the ability to induce osteoblast adhesion and proliferation and enhance the expression of osteoblastic-related genes (Morais et al., 2015). Compared to the pure HA and GR-HA_control, the GR-HA_Ce surface had a significant decrease of the bacterial adhesion for S. aureus and S. epidermidis at 30, 60, and 90 min of incubation. However, for P. aeruginosa, it showed the opposite results in that the reduction of bacterial adhesion in the GR-HA control performs better than in the GR-HA_Ce. Ca/Sr/Ce-HA NPs were found to have an enhanced antibacterial effect after Sr 2+ and Ce 3+ co-substitution. The crystallinity decreased, and the surface area of the as-synthesized NPs increased through forming bonds with the microorganisms, causing cell death (Gopi et al., 2014). Ce ions acted as an antibacterial agent, and Si 4+ ions played a crucial role in the development of apatite nuclei (Priyadarshini and Vijayalakshmi, 2018). Priyadarshini and Vijayalakshmi synthesized 1.25% of Ce along with 1, 3, and 5% of Si co-doped HA. The CFU result showed that the developed 1-5% of Si-co-doped Ce-HA had excellent bacterial inhibition and that the 5% of Si-co-doped Ce-HA got the maximum antibacterial activity. The order of inhibition efficiency against pathogens was E. coli > S. aureus > P. aeruginosa > B. subtilis (Priyadarshini and Vijayalakshmi, 2018). In addition, Ciobanu et al. (2015) prepared new Ce 4+ -substituted HA nanopowders with a more effective killing efficiency against E. coli than against S. aureus. As the Ce concentration in HA increased to 25%, the microbial inhibition against E. coli and S. aureus came to the maximum, which was 93.75% and 60.61%, respectively, since more cerium ions were released to inhibit bacteria by the improved solubility of Ce 4+ -HA. HA coatings were also widely studied and used since they could promote the osseointegration between metal implants and bone, causing no immune rejection. It was reported that cerium-doped HA/collagen coatings could kill 92.61% of E. coli and 73.59% of S. aureus after 24 h of incubation, whereas a pure HA layer did not show antibacterial properties. As mentioned above, the decreased crystallinity caused increased solubility, resulting in more cerium ions released, which make contact with the bacterial membrane and eventually cause bacterial death (Ciobanu and Harja, 2019). In addition, the chitosan scaffold embedded with Ce 3+ -doped fluorapatite was developed as a scaffold material for preventing bacterial infections in orthopedics and regenerative dentistry. The results have shown that both undoped and Ce 3+ -doped fluorapatites present better antibacterial effect than the Sr 2+ -doped ones. Importantly, high osteoconductivity leading to the differentiation of the dental pulp stem cells into osteoblasts was detected in Ce 3+ -and Sr 2+ -doped fluorapatites (Anastasiou et al., 2019).
Bioactive glass is also one of the promising materials applied in bone-related biomedical fields for bone regeneration, vascularization stimulation, wound healing, monolithic medical devices, oral care for treatment of hypersensitivity, and implant coatings (Jones, 2015). It was reported that high concentration of BG could kill various microorganisms because the process of its dissolution could release cations and lead to pH rise (Stoor et al., 1998). However, the low concentration of BG showed no antibacterial effect, and when it is doped with various antibacterial ions, such as Ce 3+ , its antibacterial property could be improved (Goh et al., 2014). In a previous study, the antibacterial activity of BG doping with 1, 5, and 10 mol% Ce was investigated by the quantitative viable count method. As a result, the BG containing 5 and 10 mol% Ce showed a significant antibacterial effect on E. coli, but there were no significant differences between the antibacterial property of BG containing 5 and 10 mol% Ce (Goh et al., 2014). Deliormanlı et al. (2016) prepared cerium-, gallium-, and vanadium-doped porous scaffolds of borate BG and studied the soft tissue ingrowth and angiogenesis in the materials. After subcutaneous implantation in rats for 4 weeks, an increase was observed in angiogenesis for cerium-doped scaffolds (CeBG), whereas a decrease was obtained in angiogenesis for the other two types of scaffolds (Deliormanlı et al., 2016). However, all of the scaffolds prepared in this study did not exhibit any antibacterial activity toward E. coli and S. aureus by a zone inhibition method. Electrospun polylactic acid (PLA)/chitosan nanofibers coated with cerium-, copper-, or silver-doped bioactive glasses (CeBG/CuBG/AgBG) were prepared, and antibacterial activities against E. coli were measured by the disk diffusion method (Goh et al., 2016). However, in this study, CeBG-and CuBG-decorated PLA/chitosan nanofibers showed no ZOI against the bacteria. The authors indicated that the disk diffusion method is affected by the diffusion rate, and therefore, it was suggested to use other methods such as the quantitative viable count method to investigate their antibacterial activities.
In addition to doping, Ce and CeO 2 could also blend with other materials without any chemical bond which would take the advantages of each component and modify the original material. For example, HA with 5 wt.% CeO 2 NPs and 2.5 wt.% Ag NPs (HA-5C-2.5Ag) had enhanced mechanical properties and antibacterial properties. In addition, the materials showed advantages of ROS scavenging, rapid healing promotion, and cell growth improvement (Pandey et al., 2018a).

Polymer Matrixes Containing Cerium and Cerium Oxide
Chitosan is a natural non-toxic biopolymer with many advantages such as low toxicity, high susceptibility to biodegradation, mucoadhesive nature, capacity to enhance drug permeability and absorption, and antimicrobial and antifungal activity (Qi et al., 2004). Therefore, Ce 3+ ions and chitosan with alginate films (Ce-Chi films) could act as wound dressing materials, showing an increasing reduction in CFU counts of E. coli and S. aureus during the 3-h exposure period (Kaygusuz et al., 2017). It was also reported that NPs with Ce and CeO 2 blended with chitosan (hybrid Chi-CeO 2 ) had enhanced antimicrobial activity. The maximum ZOI values were around 11 and 8 mm against E. coli and B. subtilis, respectively. In addition, significant morphological changes against E. coli and B. subtilis could be observed from the SEM images (Senthilkumar et al., 2017).
In addition, some materials made up of polymer matrixes containing cerium and cerium oxide were found to have antibacterial efficiency. Ureasil polyethylene oxide (U-PEO) was loaded with a combination of Ce and Ag salts. Among various materials, the U-PEO : Ce0Ag1 hybrid showed the best antibacterial efficiency, and the U-PEO : Ce0.90Ag0.10 hybrid had the best compromise between antibacterial efficiency, transparency, and photostability (Truffault et al., 2016). A mesoporous hybrid nanocomposite (CeO 2 @AlOOH/PEI) prepared by CeO 2 NPs, boehmite (AlOOH), and polyethylene imine (PEI) showed excellent antibacterial efficiency toward E. coli, K. pneumoniae, and S. aureus (Shuhailath et al., 2016). However, these polymers had no biodegradation, which might limit their clinical applications.

CERIUM-AND CERIUM OXIDE-DECORATED MATERIALS
Titanium and its alloys have been widely used for orthopedic and dental implants because of their high rate of success. However, complications such as biomaterial-related infection often occurred and led to long-term hospital stay, higher morbidity, and mortality (Zhao et al., 2015). Therefore, people began to pay attention to the study of antibacterial TiO 2 coating. TiO 2 coatings doped with different percentages of CeO 2 (5%, 10%, and 20%) were deposited on titanium by an atmospheric plasma spraying technique (Zhao et al., 2015). The results showed that all coatings had micro-sized rough surfaces and a better interfacial bonding of TiO 2 and CeO 2 /TiO 2 coatings. The 10% CeO 2 /TiO 2 coating displayed the highest effectiveness in inhibiting S. aureus with 98% sterilization rates. Recently, CeO 2 with different nanostructures (nanorod, nanocube, and nanooctahedra) were synthesized via hydrothermal procedures and coated onto the surface of titanium to mimic dental implant coatings for antibacterial and anti-inflammatory purposes (Li et al., 2019b). In this paper, among the three shapes of nano-CeO 2 , the octahedra-CeO 2 exhibited the highest Ce 3+ value and the strongest ROS scavenging capacity and therefore had the strongest inhibition against early adhesion of peri-implantitisrelated pathogens, which are shown in Figure 6, and the best anti-inflammatory effect. The better properties of octahedra-CeO 2 mainly attributed to its small size and the unique octahedral structure exposing more crystalline planes.
Bacterial biofilm is one of the primary causes of antibiotic resistance and immune response resistance, which led to persistent infections and increase in the difficulties for clinical treatment (Wu et al., 2015;Liu et al., 2019). Cerium-or cerium oxide-decorated materials could effectively inhibit the formation of biofilm, disturb the established biofilm, and eliminate the biofilm (Chen et al., 2016;Liu et al., 2019;Qiu et al., 2019).
Herein, CeO 2 -decorated porphyrin-based MOFs were designed to inhibit biofilm formation (Qiu et al., 2019). A 40% reduction in biomass was found in 50 µg/ml of the MOF@CeO 2 NP group, and after light irradiation for 5 min, the inhibition of the formation of biofilm is over 70%. When the concentration of MOF@CeO 2 increased to 200 µg/ml, the inhibition rate was over 90%. The results were attributed to the synergic effect of extracellular adenosine triphosphate (eATP) deprivation and ROS generation. In the subcutaneous abscess model, only the group treated with MOF@CeO 2 NPs and light irradiation was found to have no evident abscess after 5 days, and there were almost no viable counts of bacteria observed by using the spread plate method. Besides, hematoxylin and eosin (H&E) staining sections of the group treated with MOF@CeO 2 NPs and light irradiation showed a small number of inflammatory cells, while a large number of aggregated inflammatory cells were observed in the control group. In another study, MOF −2.5Au−Ce , a series of MOF/Ce-based nanozymes with dual enzyme-mimetic activities including deoxyribonuclease (DNase) and peroxidase mimetic activities, was designed and, in the presence of H 2 O 2 , had ability to penetrate the biofilms, intensively inhibit bacterial biofilm formation in vitro, and treat subcutaneous abscess in vivo ; Figure 7). The cerium(IV) complexes could hydrolyze eDNA and disrupt established biofilms; meanwhile, in the presence of H 2 O 2 , the MOF could kill bacteria exposed in dispersed biofilms. MOF −2.5Au−Ce applied alone could disperse biofilms moderately without killing bacteria. When treating with MOF −2.5Au−Ce and H 2 O 2 , most of the biofilms were destroyed. In the subcutaneous abscess mice model, the group treated with MOF −2.5Au−Ce + H 2 O 2 showed good wound healing, and a scab was forming after 7 days, while the group treated with MOF −2.5Au−Ce alone and other groups still had suppuration and inflammation in the wound sites. Sections of histological evaluation of the abscess showed a dramatic reduction of inflammatory cells and intact epidermal layer in the MOF −2.5Au−Ce group. In addition, Chen et al. (2016) confined AuNPs with multiple cerium(IV) complexes on the surface of colloidal magnetic Fe 3 O 4 /SiO 2 core/shell particles to combat bacterial biofilms. Bacterial adhesion was strongly reduced, and biofilm formation was prevented over 120 h by the DNasemimetic artificial enzyme (DMAE) via degrading the eDNA. Preformed biofilms of varying ages were effectively dispersed; meanwhile, the DMAE enhanced the antibacterial activity against the biofilm of antibiotics.
ZrP is a zirconium bis-(monohydrogen orthophosphate) monohydrate [Zr(HPO 4 ) 2 ·H 2 O], serving as ion exchangers, catalysts, and carriers of intercalation (Cai et al., 2012). Layered ZrP-based antimicrobials were synthesized and modified with a series of Zn 2+ or/and Ce 3+ , denoted as Zn/ZrP, Ce/ZrP, and Zn-Ce/ZrP (Cai et al., 2012). Zn 2+ intercalated into the interlayer of ZrP, while Ce 3+ was adsorbed on the surface of ZrP through hydrogen bonds. In this study, Zn-Ce/ZrP achieved the strongest sterilization effect, due to the synergistic antibacterial effect of Zn 2+ and Ce 3+ . The two cations could target the cell membranes, causing huge damage to bacteria, and interact with each other to produce more hydroxylic free radicals to induce oxidative stress. Shu et al. loaded CeO 2 and ZnO NPs onto the surface of halloysite nanotubes (HNTs), forming CeO 2 -ZnO/HNTs ternary nanocomposites (Shu et al., 2017). The cell viability of E. coli treated with pure ZnO, ZnO/HNTs, and CeO 2 -ZnO/HNTs was 18%, 12%, and 8%, respectively. The modification of CeO 2 could slow down the recombination of electron-hole pairs and narrow the energy gap of ZnO NPs. The addition of HNTs could hinder the agglomeration of ZnO NPs and promote nanocomposites to react with the bacterial membranes. In another study, N,N,N-trimethyl chitosan (TMC) with CeO 2 NPs (TMC-CeO 2 ), a biopolymer with a positive charge, exhibited antibacterial behavior and protected normal cells from oxidation (Mohammad et al., 2017). From the results, the ZOI values of TMC-CeO 2 against S. aureus and E. coli at an MIC of 100 mg/ml were similar to ZOI values of pure CeO 2 at an MIC of 200 mg/ml. The contribution of effects was due to enhanced interactions between the composite and the bacterial cell.

CeO 2 BLEND WITH OTHER NANOMATERIALS
The CeO 2 -GO hybrid nanocomposites showed an excellent synergic effect in increasing photocatalytic activity due to interface staggered band alignments existing between the CeO 2 surfaces and GO (Kashinath et al., 2019). It was reported that enhanced photocatalytic activity showed a fivefold better performance, and therefore, it showed increasing antibacterial activity. The CeO 2 -GO hybrid nanocomposites showed highly controlled growth of bacteria in both S. aureus and P. aeruginosa. When compared with pure CeO 2 and GO, CeO 2 -GO hybrid nanocomposites had smaller values of MICs and MBCs. Based on this, CeO 2 -and peppermint oil (PM oil)-embedded polyethylene oxide/GO (PEO/GO) nanofibrous mats were developed as antibacterial wound dressings (Bharathi and Stalin, 2019). CeO 2 -PM oil-PEO/GO nanofibrous mats  were observed with the maximum ZOI against E. coli and S. aureus and showed a lower MIC value than other groups. In in vivo wound healing assay, CeO 2 -PM oil-PEO/GO nanofibrous mats exhibited a rapid healing process in terms of promoting wound contraction, enhanced collagen deposition, and reepithelialization.
In addition, mixed oxide coating containing 90.7% ZnO NPs and 9.3% CeO 2 possessed the highest bactericidal properties against S. aureus under UV irradiation (Evstropiev et al., 2017). The bimodal, ZnO : CeO 2 : nanocellulose : polyaniline bionanocomposite, as potential coating agents, had a noticeable antibacterial activity with MIC 50 values of 10.6 g/ml against B. subtilis and 10.3 g/ml against E. coli (Nath et al., 2016). Carbon nanotube (CNT) reinforcement blending with CeO 2 and Ag in HA (HA-CNT-CeO 2 -Ag) had 46% bacterial adhesion reduction and 4.8 times elevated cell density in comparison with pure HA coating. Meanwhile, the physical properties such as toughness and wear resistance were also improved (Pandey et al., 2018b). The cerium oxide-incorporated calcium silicate coatings (CeO 2 -CS) exhibited strong antimicrobial activity against E. faecalis with 93% CFU reduction (Qi et al., 2019b). The antibacterial activities of cerium-and cerium oxide-decorated/blended materials are summarized in Table 4.

PROBLEMS AND FUTURE PERSPECTIVES
The cytotoxicity of CeO 2 NPs can be affected by the size, shape, and surface charge of CeO 2 NPs. The smaller-sized CeO 2 NPs exhibited higher toxicity due to larger specific surface areas, higher Ce 3+ level, and higher cellular uptake Chen and Stephen Inbaraj, 2018). When the CeO 2 NPs achieved the same size, toxicity depended on the degree of agglomeration. Due to the various shapes of CeO 2 NPs, they showed different chemical, electrical, magnetic, and optical properties (Chen and Stephen Inbaraj, 2018). In addition, it was shown that the lower the energy barrier between CeO 2 NPs and the cell surface, the easier the adhesion to the cell surface, and the higher the cytotoxicity, the easier it is for cells to uptake CeO 2 NPs with negative zeta (Chen and Stephen Inbaraj, 2018). Yet in a phosphate condition, surface chemistry is the main reason for the antibacterial activity of CeO 2 NPs, instead of size or morphology-dependent toxicity (Sargia et al., 2017).
It was reported that CeO 2 NPs and PEGylated CeO 2 NPs could protect human dermal fibroblasts, hepatic cells, and keratinocyte HaCaT cells against cytotoxicity, genotoxicity, and oxidative stress (von Montfort et al., 2015;Singh et al., 2016;Singh and Singh, 2019). Endothelial cells could regulate the amount of intracellular CeO 2 NPs by exocytosis processes, preventing adverse effects on cells due to NP accumulation (Strobel et al., 2015). It is noteworthy that it is necessary to prevent inhalation of CeO 2 NPs. The CeO 2 NPs could penetrate the alveolar space and induce both acute and chronic inflammation in the lungs, leading to irreversible lesions (Morimoto et al., 2015;Schwotzer et al., 2017).
Whether CeO 2 NPs could promote the antibacterial properties of antibiotics is a controversial issue. Bellio et al. (2018) thought that CeO 2 NPs could act as antibiotic adjuvants to increase the effectiveness of antimicrobials. CeO 2 NPs increased bacterial outer membrane permeability, allowing the entrance of the antibiotics to increase their antibacterial activity against MDR pathogens. However, in another study, it was found that when using CeO 2 NPs and ciprofloxacin together, the antibacterial effect of ciprofloxacin could be dramatically reduced by CeO 2 NPs which may prevent ciprofloxacin from being absorbed onto the bacterial cell or interfere with ciprofloxacin activity on bacterial DNA inside the bacterial cell (Masadeh et al., 2014). When NaCe(MoO 4 ) 2 is combined with different antibiotics against different bacteria, NaCe(MoO 4 ) 2 showed a synergistical or antagonistic effect (Moura et al., 2019). Therefore, more studies associated with the combination of CeO 2 NPs and antibiotics are encouraged to be done.
It is generally believed that one of the antibacterial mechanisms is the reversible conversion between Ce(III) and Ce(IV). CeO 2 NPs with a lower Ce 3+ /Ce 4+ ratio show a higher catalase mimetic activity and have anticancer or antibacterial activity (Gupta et al., 2016). On the one hand, when CeO 2 NPs enter the intracellular environment, the oxidation state of CeO 2 NPs converts to a higher Ce 3+ , indicating a net decrease of CeO 2 NPs. This is a potentially beneficial redox reaction in which harmful ROS are oxidized to a less harmful or non-harmful species. The toxicity assessment of CeO 2 NPs showed no cell death. On the other hand, significant membrane damage was observed, suggesting CeO 2 NPs may have adverse effects on the health of the cells (Szymanski et al., 2015). DNase-mimetic activity of Ce 4+ is also one of the antibacterial mechanisms. Ce 4+ could induce hydrolysis of a DNA oligomer into the fragments. It was reported that Ce 4+ complexes had moderate inhibitory activity toward cancer cells by hydrolytic DNA-cleaving activities (Yang et al., 2014), but whether they could have inhibitory activity toward normal cells was still unknown. In addition, some cerium-and cerium oxide-based materials doped with Ce 3+ showed antibacterial activity, and as the concentration of Ce 3+ increased, the antibacterial activity increased (Bomila et al., 2018;Elayakumar et al., 2019a,b). In a phosphate condition, the surface chemistry of CeO 2 NPs with a high Ce 3+ /Ce 4+ ratio is altered, resulting in the loss of their intrinsic SOD activity, and this causes nutrient starvation, leading to oxidative stress in microbes. However, in this case, CeO 2 NPs with a higher Ce 4+ /Ce 3+ ratio did not show any antibacterial activity (Sargia et al., 2017).
To sum up, this article reviewed recent researches on ceriumand cerium oxide-based nanomaterials in the biomedical field at the present stage. This included the design of cerium-and cerium oxide-related antibacterial materials and the summary for their antibacterial effects on various species of bacteria. Cerium and cerium oxide themselves are antimicrobial agents, and the compounds containing cerium and cerium oxide received improved antimicrobial activities as well as other beneficial properties such as promoting angiogenesis, osteogenesis, and wound healing. Admittedly, researchers have made some progress in the development of various antibacterial agents and biomedical materials but far from enough. From this review, it was found that most researches were still at the level of in vitro experiments. In order to make ceriumand cerium oxide-related antibacterial materials obtain better application, more animal experiments and observation of longterm effects are expected. The cytotoxicity and mechanisms of new cerium-and cerium oxide-based materials need to be further investigated.

AUTHOR CONTRIBUTIONS
LW and CL designed the review article. MQ and WL wrote the manuscript. XZ, XL, YS, and YW participated and helped with the final revision of the article. All the authors read and approved the final version of the manuscript prior to submission.