Therapeutic effects of antibiotics loaded cellulose nanofiber and κ-carrageenan oligosaccharide composite hydrogels for periodontitis treatment

Periodontitis is an inflammatory disease that can lead to the periodontal pocket formation and tooth loss. This study was aimed to develop antimicrobials loaded hydrogels composed of cellulose nanofibers (CNF) and κ-carrageenan oligosaccharides (CO) nanoparticles for the treatment of periodontitis. Two antimicrobial agents such as surfactin and Herbmedotcin were selected as the therapeutic agents and the hydrogels were formulated based on the increasing concentration of surfactin. The proposed material has high thermal stability, controlled release, and water absorption capacity. This study was proceeded by investigating the in vitro antibacterial and anti-inflammatory properties of the hydrogels. This material has strong antibacterial activity against periodontal pathogens such as Streptococcus mutans, Porphyromonas gingivalis, Fusobacterium nucleatum, and Pseudomonas aeruginosa. Moreover, a significant increase in malondialdehyde (MDA) production and a decrease in biofilm formation and metabolic activity of the bacteria was observed in the presence of hydrogel. Besides, it reduced the reactive oxygen species (ROS) generation, transcription factor, and cytokines production in human gingival fibroblast cells (HGF) under inflammatory conditions. In conclusion, the hydrogels were successfully developed and proven to have antibacterial and anti-inflammatory properties for the treatment of periodontitis. Thus, it can be used as an excellent candidate for periodontitis treatment.

Fourier-transform infrared spectroscopic (FTIR) analysis. FTIR analysis of hydrogels was given in Fig. 2. Herbmedotcin is a complex mixture consisting of various bioactive components. The peak at 3428 cm −1 represents the presence of -OH and -NH stretching vibrations 32 . A small peak at 2944 cm −1 indicates the presence of alkenes and a peak at 2891 cm −1 corresponding to C-H stretching 33,34 . The peak at 1655 cm −1 represents N-H bending vibrations 32 . A sharp peak at 1401 cm −1 is corresponding to the -C=C-stretching vibration of carboxylic acids 35 . Another compound present in the hydrogel was surfactin. The peak at 3312 cm −1 , 2960 cm −1 , and 2927 cm −1 correspond to N-H stretching mode and lipopeptide portion of the surfactin 36 . A small peak at 2856 cm −1 represents the presence of an aliphatic chain and a deep peak at 1540 cm −1 indicates the deformation mode of the N-H bond combines with C-N stretching mode 37 . A detailed FTIR spectrum of CO-CNF was elucidated in our previously reported work 15 . CCH hydrogels were formed in the presence of an MBAA as a cross-linker. The band observed at 1639 cm −1 in CCH assigned to C=C stretching and presence of carboxylate ions indicated by two peaks, 1431 cm −1 , and 1376 cm −1 38 . The asymmetric stretch of O=S=O (1266 cm −1 ) and -O-SO 3 (848 cm −1 ) were also observed. It was noted that there was no big difference between the FTIR spectra www.nature.com/scientificreports/ of 100SH, 200SH, and 400SH. A small sharp peak formed at 3307 cm −1 was an amide bond A supposed to be coming from surfactin 39 . A small speak shifting at 2856 cm −1 in surfactin was also noted and a peak coming from Herbmedotcin was seen at 1657 cm −1 . A short peak at 848 cm −1 was supposed to be originated from CCH. So, it was understood that the antimicrobial agents were successfully entrapped in the hydrogel matrix.
Thermogravimetric analysis (TGA). The thermal degradation of hydrogels is given in Fig. 3. A slight reduction in weight was observed between 50 and 150 °C and this may due to the evaporation of moisture. In CCH, this weight loss was continued to 230 °C, then decreased suddenly, and finally reached to a degradation temperature at 432 °C. It was almost similar to 100H, 200SH, and 400SH with slight changes. A sharp decline in the weight loss was observed from about 166 °C in 100SH, 200SH, and 400SH and the degradation temperature of the above said hydrogels was around 510 °C.

Entrapment efficiency (EE), loading capacity (LC) and in-vitro drug release. The EE and LC of
surfactin loaded-CCH was 64.32 ± 3.85% and 41.87 ± 4.57%, respectively (Fig. 4). A rapid drug release of surfactin was observed from 0 to 2 h. It might be due to the release of physically entrapped drugs. After 4 h, a controlled release of drugs was observed and about 75% of the drug was released after 24 h. The in-vitro release profile of hydrogel showed a slower controlled release of surfactin. The slower release of surfactin may due to the high cross-linking of CO-CNF with surfactin that evident from the TGA and FTIR analysis. Release of a drug from a carrier depending upon various factors. Literature mentioned that the drug release can be divided into delayed-release, site-specific targeting, receptor targeting, and sustained release. And the sustained release can be divided into controlled-release and extended-release. Among them, the extended type of drug release system has slow drug release 40 . According to our results, the material has a slow release of drug and about 50% of drugs were released after 8 h. www.nature.com/scientificreports/ Swelling ratio of hydrogels. The water absorption by hydrogels at different time intervals is given in Fig. 5. It was noted that hydrogels absorbed a large amount of water initially and later slightly increased with time. The maximum swelling ratio was observed up to 6 h. After, the swelling ratio was decreased suddenly in all hydrogels due to the breakdown of the dry hydrogel system in the presence of water. This may due to the brittle nature of the hydrogel. Literature mentioned that the bond-breaking was observed in highly swollen hydrogels was due to their brittle nature 41 .
Antioxidant activity of hydrogels. The antioxidant activity of the samples was evaluated by the DPPH assay ( Fig. 6). Trolox was taken as the standard. It was noted that more than 60% of scavenging activity was observed in the presence of Herbmedotcin and 400SH. The antioxidant activity of doxycycline was almost similar to 200SH and the activity of surfactin was slightly higher than 100SH. The maximum scavenging activity  www.nature.com/scientificreports/  www.nature.com/scientificreports/ was observed in CCH was about 32%. It was assumed that the antioxidant activity of hydrogels may due to the synergetic effect of CO, surfactin, and Herbmedotcin.
Well (ditch) plate method. Figure 7 and Table 1 showed the antimicrobial activity of hydrogels against four periodontal pathogens such as S. mutans, P. gingivalis, F. nucleatum, and P. aeruginosa. As shown in Fig. 7, drug-loaded hydrogels possessed antimicrobial activity against S. mutans in a concentration-dependent manner, at the same time the bacteria were resistant to CCH. The zone of inhibition (ZI) of doxycycline was slightly higher than 400SH whereas free Herbmedotcin and surfactin showed 11.66 mm and 13.33 mm ZI, respectively. This situation was almost similar to P. gingivalis but, P. gingivalis shown little bit sensitivity towards CCH. Only 400SH has higher ZI than Herbmedotcin and doxycycline. Hydrogels such as 200SH and 400SH showed anti-  Minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) of hydrogels. The MIC and MBC of hydrogels against microorganisms were given in Table 2 and it was expressed as the percentage of its initial concentration (40 mg/ml). As shown in the Biofilm inhibition by crystal violet assay. Biofilm is the association of microbial species that are encased in a matrix 42 . The inhibition of biofilm by different samples towards periodontal pathogens were given in Fig. 8 and Supplementary Fig. S1. Doxycycline and 400SH shown more inhibitory effect on S. mutans and its activities were significantly higher than 100SH and 200SH. Like, in P. gingivalis, the inhibitory effect of 100SH, 200SH, and 400SH was increased. The highest percentage of biofilm inhibition was exhibited by 400SH and the next was doxycycline. There was no significant difference in the activity between doxycycline and 200SH. As compared to previous bacteria, the CCH treatment has some beneficial effects on F. nucleatum. There was no significant biofilm inhibition was observed between doxycycline, 200SH, and 400SH but it was higher than 100SH. The highest www.nature.com/scientificreports/ percentage of biofilm inhibition by 400SH on P. aeruginosa was observed and it was significantly higher than doxycycline. About 60% of biofilm inhibition was shown by 100SH and 200SH. CCH possessed more than 20% of inhibition activity in P. aeruginosa and F. nucleatum than S. mutans and P. gingivalis. Drug-loaded hydrogels produced biofilm inhibition after 24 h of treatment but 400SH has the highest biofilm inhibitory effect on all periodontal pathogens when compared to other hydrogels.
Metabolic activity of bacteria. The metabolic activity of the bacteria was estimated by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay ( Fig. 9). It is the cheapest and simplest technique that helps to understand the physiological state of the bacteria 43 . The viability of S. mutans was highly reduced after treatment with doxycycline, 100SH, 200SH, and 400SH. 400SH showed the highest reduction viability and doxycycline came to the second position. As shown in the Fig. 9, the viability of P. gingivalis was dramatically reduced after 400SH treatment and the viability was reduced in an order of CCH > 100SH > 200SH > 400SH.
There was no significant difference between the viability of doxycycline and 200SH groups and CCH and 100SH groups. 400SH did not possess any noteworthy difference in the viability of F. nucleatum when compared to doxycycline but it was much lesser than doxycycline. There were no significant variations in the metabolic activity of the F. nucleatum between CCH, 100SH, and 200SH groups. The lowest viability of P. aeruginosa was seen in the doxycycline-treated group and 400SH were come under the second position. 100SH and 200SH did not possess any significant difference in their activity towards P. aeruginosa. www.nature.com/scientificreports/ Malondialdehyde (MDA) assay. As shown in Fig. 10, the MDA level was highly reduced in the control group for all bacteria. In S. mutans, the highest level of MDA was observed in 400SH but it was not significantly different from other groups such as doxycycline, 100SH, and 200SH. This situation was slightly similar in the case of P. gingivalis, in which the highest MDA level was observed in 400SH. Doxycycline, 100SH, and 200SH have no significant difference in the MDA level but it was higher than CCH. In F. nucleatum, a low level of MDA was observed in positive control when compared to 100SH, 200SH, and 400SH. The highest MDA level was seen in the 400SH group but it was not significantly differed from 200SH. As shown in the Fig. 10, the level of MDA was increased in the CCH group than doxycycline in the case of P. aeruginosa. The highest MDA level was observed in 400SH followed by 200SH.
Acridine orange (AO) assay. The bacteria were treated with samples for 24 h and then stained with AO.
Images of AO assay was elucidated in Fig. 11. It was observed that S. mutans were highly viable in the control and CCH group. Very few numbers of viable cells were seen in doxycycline, 100SH, and 200SH whereas 400SH almost promoted 100% bacterial death. However, the situation was different in P. gingivalis. Along with control and CCH groups, 100SH also showed more viable cells and 200SH has an approximately higher number of viable cells as compared to doxycycline and 400SH. The viability of the F. nucleatum was extremely increased in all groups except doxycycline. Like S. mutans, viable P. aeruginosa was highly observed in control and CCH groups. 100SH showed an almost equal number of viable and non-viable cells whereas the number of dead cells more seen in doxycycline 200SH, and 400SH. It was confirmed that these hydrogels can kill 4 kinds of bacteria in its higher form (200SH and 400SH) together with doxycycline. www.nature.com/scientificreports/

Cytotoxicity of hydrogels in human gingival fibroblast (HGF) cells.
MTT assay was used to determine the viability of the cells. The cell viability is linearly proportional to its mitochondrial activity 44 . Figure 12 showed the viability of the HGF cell line after treated with different samples. It was noted that the cell viability was decreased with an increase in the concentration of each sample. As compared to other groups, CCH showed higher viability. It was also noted that the cell viability of all groups significantly differed from the control group from 50 µg/ml concentration. The cell viability of all samples was dropped below 80% from 100 µg/ml concentration. There was no big difference in cell viability between 100SH, 200SH, and 400SH but the viability was slightly reduced in the 400SH group when compared to others. All samples-treated cells exhibited more than 80% cell viability up to 50 µg/ml concentration. So, this concentration was selected for performing further analysis.
Oxidative stress. An inflammatory condition in HGF cells was evaluated by performing MDA, NO, and NBT analysis (Fig. 13). MDA production was evaluated as a marker to determine the oxidative stress in LPS stimulated HGF cells. The cells were subjected to samples treatment after LPS induction. The MDA level was higher in the control group and lower in the normal group. There was no significant difference in the MDA level between doxycycline, CCH, 100SH, and 200SH but it was significantly higher than 400SH. So, it was confirmed that the hydrogels were able to reduce the MDA level under inflammatory conditions. Excessive production of  www.nature.com/scientificreports/ ROS is a sign of oxidative stress and it causes damage to DNA, protein, and lipids 45 . Previous literature indicated that the systemic inflammatory responses can be evaluated from NO production 46 . It was observed that nitric oxide production was highly increased in the control group. As compared to doxycycline, drug-loaded hydrogel groups showed better results in terms of NO production. There was no significant difference between 100SH, 200SH, and 400SH but lower NO generation was observed in 400SH when compared to others. The ROS production in hydrogels treated inflammatory cells were again confirmed by NBT assay. It is used to determine the production of superoxide anion (O 2 − ) by cell 47 . A high reduction of superoxide anion production was observed in normal, doxycycline, and 400SH treated groups. CCH, 100SH, and 200SH also reduced the production of superoxide anion when compared to the control group. So, it was understood that hydrogels can reduce the production of ROS, and thereby it can protect the cells from oxidative stress.
Cytokines production. Literature showed that in cultural periodontal cells, the activation of NF-κB by IL-1β and TNF-α was observed and this may lead to the production of prostaglandin and matrix metalloproteinase (MMP) 48 . The production of NF-κB, PGE2, and IL-6 after treatment were given in Fig. 14. The production of NF-κB was higher in the control group and lower in the normal group. It was noted that there was no significant difference between doxycycline and hydrogels but CCH has higher NF-κB expression than others. The level of www.nature.com/scientificreports/ PGE2 and IL-6 was also higher in the control group and it was significantly lower in the normal group. Like NF-κB, the expression PGE2 was not significantly differed between other groups, and comparatively lower production was seen in 400SH. But this situation was quite different in IL-6. CCH group also showed a higher level of IL-6 like the control group and the level was lowered in the following order CCH > 100SH > 200SH > 400SH > doxycycline. 400SH and doxycycline have reduced IL-6 production below 60 pg/ml and it was significantly lower than 100SH and 200SH.

Discussion
Periodontitis is an inflammatory disease primarily resulted from the microbial accumulation and it is estimated that about 47.2% of the U.S. adult population was affected by periodontitis 49 . Periodontitis is developed due to the formation of biofilms composed by enzymes, protein epithelial cells, and food residues and the teeth provide a suitable surface for bacterial attachment and progression 50 . Literature mentioned that there are five different ecological niches such as teeth, saliva, periodontal pockets, gingival sacculus, and tongue existed for the initiation and progression of periodontitis 51 . The current study is focusing on four major bacteria such as P. gingivalis, S. mutans, F. nucleatum, and P. aeruginosa, which cause periodontitis directly or indirectly. P. gingivalis is a gramnegative anaerobic bacterium that initiates the inflammation by invading the epithelial cells. Besides, the major component of gingival connective tissue called gingival fibroblast may directly interact with P. gingivalis and its bacterial products called LPS 52 . S. mutans is another bacterium that indirectly leads to the progression of periodontitis. It is known as the major etiological agent for dental caries. The dental caries is formed due to the www.nature.com/scientificreports/ sensitivity of the teeth towards acidic pH and resulted in the dissolution of enamel. The literature mentioned that both dental caries and periodontitis linked directly or indirectly due to some common contributory factors and the pockets formed as a result of dental caries may facilitate the progression of periodontitis 53 . Like P. gingivalis, F. nucleatum is also an anaerobic gram-negative bacterium responsible for the progression of periodontitis. It increases pocket depth, inflammation, and plaque formation. It encodes several adhesins and binds to several mammalian cells. The virulent mechanism of F. nucleatum has been classified into the induction of host immune response, colonization, and dissemination 54 . A previous study reported the high prevalence of P. aeruginosa in the subgingival biofilm under different periodontal conditions and they isolated Pseudomonas species from dental biofilms and mucosa 55 . However, it is commonly associated with the generation of pulmonary disease. However, literature stated that the oral cavity is known as the important reservoir for P. aeruginosa. So, the eradication of P. aeruginosa is also crucial to maintaining normal health. Due to the host-bacterial interaction, an acute inflammatory response will occur and neutrophils invade the connective tissue and later interact with other immune cells. As a result, cytokines such as IL-6, IL-17, TNF-α, IL-1α, and IL-1β are produced 56 . IL-17 promotes the production of chemokines, matrix metalloproteinases (MMPs), osteoblast expression of receptor activator of nuclear factor-kB ligand (RANKL), and other tissue deductive molecules 57 . Recent literature mentioned that LPS can act as a mediator for inflammation in the gingival area by enhancing the production of cytokines such as IL-6, TNF-α, and IL-1β. In addition to this, ROS overproduction and the activity of certain cells such as neutrophils, leucocytes, T lymphocytes, and plasma cells may also attribute to LPS-induced periodontitis 58 . Also, receptors such as TLR-4 and TLR-2 are connected to the pathology of periodontitis. The activation of these receptors will lead to the activation of T helper cell 2 and subsequent production of pro-inflammatory cytokines 59 .
The current study is focusing on the in vitro antibacterial and anti-inflammatory properties of surfactin and Herbmedotcin-loaded CO-CNF hydrogels. A local drug delivery system provides the attachment and preventing the side effects of free-drugs. The main aim of the drug delivery system is to deliver antimicrobial agents in those areas where mechanical scaling instruments cannot access. Besides, this material should have cytocompatibility and biocompatibility. The hydrogels were prepared from a composite nanoparticle (CO-CNF) composed of CNF and CO 15 . CNF is widely used for hydrogel preparation due to its flexibility and entanglement formation 14 . Lohrasbi et al. identified that the addition CNF improves the gelation time and stability of collagen hydrogels 60 . Like cellulose, CO is also a natural polymer and it mimics the glycosaminoglycan structure 61 . So, the combined use of CNF and CO provides a suitable platform for drug delivery. Herbmedotcin is a positively charged, safe, and effective antibacterial agent widely used in healthcare, agriculture, and aquaculture 20,62 . Studies were done by Park et al. supported the usage of surfactin as a candidate to prevent inflammatory disease, especially periodontitis 63 .
Recent literature mentioned that nanostructured hydrogels can facilitate the biological functions and tissue growth by stimulating interactions between cells and their environment 64 . The morphology of the hydrogels was evaluated by SEM analysis (Fig. 1). This technique helps to understand the porosity and nature of the hydrogels 65 . Because of the highly reactive carbon-carbon double bonds, MBAA has been used as a cross-linking agent and it can form a three-dimensional network structure by interacting with certain groups such as -NH 2 , -OH, and -COOH 66 . The smooth surface observed in the hydrogels was due to the cross-linking nanofibers and the blocks were obtained due to the cross-linking of CO. Previous study also reported the smooth surface nature of freeze-dried NFC hydrogel in presence of calcium dinitrate (CaN 2 O 6 ) 67 .
The FTIR analysis confirms the interactions between the CO-CNF, surfactin, and Herbmedotcin (Fig. 2). Moreover, this material has high thermal stability (Fig. 3) and it might be due to the strong bonding between the polymers and drugs 68 . In addition to high thermal stability, the high swelling ratio was also observed in hydrogels. The previous study stated that the network structure of the hydrogels are formed by either physical interaction or chemical interaction and the integrity and swelling of the hydrogels are depending upon the degree of cross-linking 69 . As shown in Fig. 5, the swelling ratio was increased suddenly and continued up to 6 h. After, it was dropped due to the breakdown of the hydrogel system. The collapse of the hydrogel system may result from the brittleness nature of the dried hydrogel system. Previous study mentioned the brittleness nature of highly cross-linked hydrogels 69 . The swelling ability of hydrogels was due to the presence of hydrophilic groups present in the CNF and CO. The high swelling ability of nanocellulose hydrogel is due to its osmotic effects because the ions present in the external solution are diffused to the center of the gel. As a result, the presence of counterions around the charged polymers in aqueous media causes electrostatic neutrality 70 . Hezaveh and Muhamad reported the hydrogen bond breaking and subsequent diffusion of water in κ-carrageenan and hydroxyethyl cellulose-based hydrogels 71 .
It was known that excess production of ROS leads to the destruction of periodontal tissues by lipid peroxidation, release of cytokines, protein damage, DNA damage, and oxidation of important enzymes 72 . DPPH radical scavenging assay was conducted to evaluate the antioxidant activity of the hydrogels (Fig. 6). The mechanism behind the DPPH assay is based on the antioxidant ability of the samples to quench the DPPH and the dark purple color of the DPPH is changed to colorless 73 . It was understood that 400SH and Herbmedotcin have high scavenging activity than others. It may be due to the synergetic effect of Herbmedotcin, surfactin, and CO. Literature stated that the DPPH radical scavenging activity of surfactin is due to the presence of peptide ring and fatty acid chain 74 . Herbmedotcin has also a major role in antioxidant activities because it contains effective pure compounds from herbs 20 . The ROS scavenging activity of low molecular weight κ-carrageenan is closely related to the degree and position of sulfation, composition of monosaccharides, and molecular weight 75,76 .
Periodontitis is an inflammatory disease associated with biofilm formation by the dysbiotic bacterial community. Antimicrobial activity of hydrogels was primarily evaluated by agar ditch plate method against four periodontal pathogens such as P. gingivalis, S. mutans, F. nucleatum, and P. aeruginosa (Fig. 7). It was observed that P. gingivalis and P. aeruginosa were a little bit sensitive to CCH and better zone of inhibition of 200SH and 400SH may due to the high concentration of the antibacterial agents. Literature mentioned that MIC value is not accurate to determine whether the antibiotic is bacteriostatic or bactericidal. However, MBC value helps to Scientific Reports | (2020) 10:18037 | https://doi.org/10.1038/s41598-020-74845-9 www.nature.com/scientificreports/ choose the lowest concentration of antibiotics that kill (99.9%) the bacteria 77 . According to our study (Table 2), the MBC values of hydrogels were ranges from 50 to 100%. The MBC values of 100SH were 100% for two bacteria, 90% for another one. It was also similar to 200SH, except there were 90% and 70% MBC values for S. mutans and P. aeruginosa, respectively. The situation was different in the case of 400SH, in which the MBC values were ranging from 80 to 50%. Thus, to maintain the constituency, 100% concentration was selected for further studies while considering all bacteria. The antimicrobial activity of hydrogels was potentially due to the synergetic effect of all components present in the hydrogel. It was also noted that the activities of hydrogels were varied according to each bacterium. The previous study reported that the antimicrobial activity of surfactin involves the disruption of bacterial membrane followed by the leakage of the cellular component while Herbmedotcin uses its positive charge to attach the lipid membrane and cause bacterial lysis 20,78 . It was also noted that the antibacterial activity of oxidized form of κ-carrageenan was reported against P. aeruginosa. Literature mentioned that κ-carrageenan causes bacterial death by damaging the bacterial cell wall and cytoplasmic membrane 79 . During the adverse condition, different types of bacteria can colonize in the oral cavity and facilitate various interactions to initiate biofilm formation. Streptococcus and Actinomyces species lay the foundation for the development of bacterial biofilm 80 . Later, species such as Fusobacterium and Veillonella, etc. aggregate to colonizers and finally, it organized to form a complex matter formed by Corynebacterium, Streptococcus, Porphyromonas, Haemophilus/Aggregatibacter, Neisseriaceae, Fusobacterium, Leptotrichia, Capnocytophaga, and Actinomyces 80 . The Coaggregation of these species is happening due to intergeneric specific cell-to-cell recognition via surface adhesins and receptors 81 . The inhibition of biofilm by hydrogels were given in Fig. 8. It was noted that 400SH shown higher percentages of biofilm inhibition similar to doxycycline. Lowest biofilm inhibition (50%) by 400SH was observed in P. gingivalis but more than 80% of biofilm inhibition was seen in P. aeruginosa.
The production of ROS was drastically increased due to the innate immune system under inflammatory conditions and leads to oxidative stress and cell damage 82 82 . It will cause various adverse damages, specifically lipid peroxidation. MDA is known as the end product of lipid peroxidation. Figure 10 showed that hydrogels were able to increase MDA production in four bacteria, especially by 200SH and 400SH. It was also understood that the metabolic activity of each bacteria was dropped after the treatment with hydrogels. This may reveal that the hydrogels cause oxidative stress in bacteria and leads to the reduction of their metabolic activity and finally resulted in cell death (Figs. 9, 11).
The safety of the proposed drug is mostly evaluated based on cell viability and cytotoxicity assays. A cytotoxic drug is defined as a compound that affects cell growth, interferes with cellular attachment, causes morphological changes, and finally reduce overall viability 83 . HGF cells were selected to evaluate the anti-inflammatory properties of the hydrogels. It is known as the most abundant cells in the gingival connective tissue and expresses cell surface CD14, TLR4, and MD-2 and produce pro-inflammatory cytokines, such as IL-6 and IL-8, upon LPS stimulation 84 . MTT assay is a colorimetric assay widely used for determining the viability of cells. The principle behind this assay is based on the dehydrogenases present in the mitochondria of viable cells and it converts the tetrazolium compound into formazan crystals 85 . The viability of HGF cells in the presence of each hydrogel was given in Fig. 12. It was noted that up to 50 µg/ml concentration, all the hydrogels maintained more than 80% of cell viability. So, this concentration was selected for further studies. It was known that the major mechanism behind periodontitis is the oxidative stress-mediated inflammatory pathway. The interaction between the bacterial biofilms and resident cells are leading to the inflammatory response. The bacterial outer membrane is consists of LPS and it can trigger the release of cytokines such as IL-6, IL-1β, and TNF-α 86,87 . In addition to this, excessive production of ROS can damage lipid, protein, and DNA and this will lead to oxidative stress and further tissue damage 88 . So, the ROS production in the presence of hydrogels was evaluated by NO, NBT and MDA assays. Usually, the formation of NO was measured by using Griess assay and it measures the conversion of nitrite to a purple colored azo dye 89 . Like, NBT assay is used to detect the superoxide (O 2 − ) 90 . The literature stated that the level of MDA was higher under periodontitis conditions because of lipid peroxidation associated with oxidative stress 88 . As shown in Fig. 13, hydrogels reduced MDA, NO, and superoxide production. Local microbiota and host immune response have a crucial role in the pathogenesis of periodontitis. It has been understood that gingival fibroblasts can activate NF-κB and produce inflammatory cytokines such as IL-1β, and TNF-α 91 . IL-6 and PGE2 are known as the biomarkers for periodontitis and largely leads to bone resorption 92 . As shown in Fig. 14, the level of both of NF-κB, IL-6, and PGE2 were elivated under inflammatory condition. At the same time, the level of them were decreased after treatment with hydrogels.
For several years the hydrogels emerged as a potential drug delivery system due to its ability to mimic ECM. Pakzad and Ganji et al. developed a drug loaded thermoresponsive hydrogel composed of chitosan/gelatin/-glycerolphosphate for periodontal application. The material was biocomtible and has antimicrobial activity against Clostridium sporogenes 93 . An antimicrobial dental light curable bioadhesive hydrogel composed of visible-lightactivated naturally derived polymer (gelatin) and an antimicrobial peptide developed by sani et al. has higher adhesion to physiological tissues and showed antimicrobial actvity against P. gingivalis 94 . Chen et al. developed an thermosensitive nanoparticle hydrogel for the delivery of ibuprofen and basic fibroblast growth factor. This material has significant anti-inflammatory properties and promoted the proliferation and adhesion of human gingival fibroblasts cells 95 . However, most of the studies are focused on to either material characteristics or antimicrobial/anti-inflamamtory properties. Our present study investigated the antibacterial and anti-inflammatory properties of antimicrobials loaded cellulose nanofibers and κ-carrageenan oligosaccharide composite hydrogels under periodontitis conditions. We investigated the biofilm formation, oxidative stress, and metabolic activity of bacteria along with checking the antimicrobial activity. Moreover, we investigated the anti-inflammatory properties of hydrogels in HGF cells. According to our study, the sensitivity of periodontal pathogens to the hydrogel suggesting a new alternative to conventional treatments. In addition to this, the hydrogels can reduce Scientific Reports | (2020) 10:18037 | https://doi.org/10.1038/s41598-020-74845-9 www.nature.com/scientificreports/ the inflammatory condition and possesses strong antioxidant activity. These elements are very crucial in the management of periodontitis.

Methods
Preparation of antimicrobials loaded hydrogels. Hydrogels were prepared according to a previously reported work with some modifications 96 . CO-CNF nanoparticles were selected for preparing drugs-loaded hydrogels. Detailed information on the preparation of CO-CNF was described in our previously published work 15 . 50 ml of CO-CNF solution was prepared by dissolving 5 g of CO-CNF in 50 ml distilled water. Each quantity of surfactin (100,200, and 400 mg) (Sigma-Aldrich, Louis, Missouri, USA) was added to the above solution under stirring conditions (40 °C, 800 rpm) to prepare different concentrations of surfactin loaded hydrogel. The stirring was continued for 30 min. Then, 1 ml of Herbmedotcin (Giant Bio Tech, New Taipei City, Taiwan) was added to each above said system and stirred for 15 min. After, 5 ml (0.1 g/ml) of the MBAA solution was added and stirred for 2 h to obtain the complete gelation.
Scanning electron microscopic (SEM) analysis. Morphological analysis of the hydrogels was carried out by using SEM (S-4800 Scanning Electron Microscope, HITACHI, Tokyo, Japan). Dried samples were transferred to the metal stud using double-sided tape and coated with gold using a sputter gold coater (Model-E1010 Ion sputter). Images were observed at different magnifications at an accelerating voltage of 15 kV.

Fourier-transform infrared (FTIR) spectroscopy analysis.
The FTIR spectra of the samples were studied using an FTIR spectrometer (BRUKER, TENSOR II, Massachusetts, USA) by potassium bromide (KBr) disc method. The samples were dried, ground and pelletized using KBr (1:100, w/w) to form the thin films of samples. 64 scans were performed in the range of 400-4000 cm −1 and the resolution was taken as 4 cm −1 .
Thermogravimetric analysis (TGA). TGA instrument (NETZSCH TG 209F3, Germany) was used to evaluate the thermal degradation behavior of samples (5 mg). The temperature range was fixed at 40-600 °C with a constant heating rate of 20 °C/min under nitrogen atmosphere.
Evaluation of swelling ratio (SR). The swelling ratio of hydrogels was determined by immersing the dried samples in water from 0 to 12 h at room temperature. At particular points, samples were taken, remove the surface water, and subsequently transferred to measure the weight. All experiment was performed in triplicates 97 . The Swelling ratio was calculated by the following equation: www.nature.com/scientificreports/ W dry is the weight of hydrogel in dry condition, W total is the weight of swelled hydrogel.
DPPH radical scavenging activity. Antioxidant activities of the samples were evaluated by using DPPH radical scavenging assay according to a previously reported method 98 . 100 µl of DPPH (0.004 g in 100 ml methanol) was mixed with an equal volume of samples in a 96-well plate. The concentration of doxycycline (Swiss Pharmaceuticals Co., Ltd., Tainan, Taiwan), Herbmedotcin, and surfactin was 10 mg/ml, and the concentration of hydrogels was 40 mg/ml. Trolox (0.5 mg/ml) (Sigma-Aldrich, Louis, Missouri, USA). and methanol was selected as standard and blank, respectively. The absorbance was measured at 517 nm after 30 min of incubation period under dark. The DPPH radical scavenging activity was calculated by the following equation: (1) Swelling Ratio (SR) (%) = W total − W dry W dry × 100, www.nature.com/scientificreports/ umn Phenomenex (250 mm × 4.6 mm, 5 µm) with a temperature of 30 °C was selected and the flow rate was set at 1.2 ml/min 99,100 Inhibition of biofilm (%) = 1 − Absorbance of cells treated with drugs/absorbance of non treated control cells ×100.  www.nature.com/scientificreports/ incubator for 4 h. Then, the MTT solution has removed and 200 µl of DMSO solution was added under dark condition. The absorbance was measured at 570 nm. The experiment was performed in triplicates.
Nitroblue tetrazolium (NBT) reduction assay. The cells (2 × 10 5 cells/well) were pre-treated with 200 µl of (50 µg/ml) samples and 100 µl of LPS (1 µg/ml) (from Escherichia coli O26:B6, impurities < 5% Protein) in a 12-well plate and incubated for 24 h in a CO 2 incubator. Later, the cells were centrifuged at 800 × g for 15 min and the supernatant was removed. 300 µl of NBT [(1.5 mg NBT, 13.8 ml medium, 450 µl DMSO, and 750 µl Triss-buffered saline (TBS)] solution was added to the cells and incubated for 1 h in the absence of light. Centrifuged (1500 × g, 15 min) the obtained solution after 1 h and the supernatant was removed. 200 µl DMSO was added and the optical density was measured at 570 nm. The percentage of inhibition of NBT reduction was calculated according to the following equation: Evaluation of nitric oxide and MDA production. The cell concentration was adjusted to 1 × 10 4 cells/ well in a 96-well plate. The cells were pre-treated with 20 µl of samples (50 µg/ml) and 10 µl of LPS (1 µg/ml) and incubated for 24 h in a CO 2 incubator. After transferred 50 µl of the cultured medium into another 96-well plate and 50 µl of Griess reagent (mixture of 0.01 g sulphanilamide in 10 ml 5% phosphoric acid and 0.01 N-1-napthylethylenediamine dihydrochloride in 10 ml dd water) was added (kept for 10 min). Sodium nitrate (50 µl) was taken as the standard. 0.068 g sodium nitrate was dissolved in 100 ml dd water and serially diluted. The absorbance was measured at 540 nm. The MDA production was evaluated in LPS and hydrogels treated cells. Cells were treated with samples/LPS as above and 200 µl of MDA reagent (a mixture of 47 ml water, 1 ml HCl, 7.2 g trichloroacetic acid, and 0.18 g thiobarbituric acid) was added and placed in a water bath for 10 min at 100 °C. 1,1,3,3-tetramethoxypropane was taken as the standard. Later, 300 µl of 1-butanol was added and centrifuged (1500 × g, 10 min) after mixing. The absorbance was measured at 532 nm.
Cytokines production. HGF cells were cultured in a fibroblast medium containing 10 ml FBS, 5 ml fibroblast growth supplement, and 5 ml of penicillin/streptomycin. Later, cells (1 × 10 4 cells/well) were pre-treated with 20 µl of the sample (50 µg/ml) and 10 µl of LPS (1 µg/ml) and incubated for 24 h in a CO 2 incubator. Later, samples were centrifuged (1000 × g, 20 min) and the supernatant was collected for evaluating transcription factor and cytokines [NF-κB (Asia Bioscience Co. Ltd., Taipei, Taiwan), IL-6, and PGE2 (Taiclone Biotech Corp., Taipei, Taiwan)] productions. All procedures were conducted according to the manufacturer's protocol. For NF-κB determination, 100 µl standard/samples were added to respective wells and incubated at 37 °C for 1 h. After, aspirated the solution and 100 µl of detection reagent A was added, and again incubated for 1 h at 37 °C.

Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.