MXene-based protective strategy for diabetic gingival wound healing: shielding fibroblasts from oxidative stress

Abstract

The bidirectional association between diabetes mellitus (DM) and periodontitis remains a major focus in oral and systemic health research. DM is a key risk factor influencing the onset, progression, and severity of periodontitis, yet the molecular mechanisms underlying this relationship are not fully understood. Periodontitis is characterized by bacterial biofilm formation and a destructive host immune-inflammatory response. In this study, we explored the therapeutic potential of Ti₃C₂T_x MXene, a two-dimensional nanomaterial, for enhancing gingival wound healing under diabetic conditions. Ti₃C₂T_x MXene treatment of fibroblast cells derived from diabetic rat gingival tissue modulated oxidative stress and restored glutathione balance. The material exhibited significant biocompatibility, preserved mitochondrial membrane potential, and reduced intracellular reactive oxygen species (ROS) levels. Moreover, Ti₃C₂T_x MXene influenced lipid peroxidation and cytochrome c release, contributing to controlled caspase activation and balanced apoptotic responses. These results suggest that Ti₃C₂T_x MXene supports cellular and mitochondrial homeostasis, promoting improved wound repair in diabetic gingiva. Collectively, this study presents the first evidence of Ti₃C₂T_x MXene as a promising nanotherapeutic platform for managing diabetic oral wounds and potentially other chronic wounds, paving the way for future applications in nano-enabled topical formulations and interdisciplinary oral healthcare strategies.

---

1. Introduction

Diabetes mellitus (DM) is a metabolic disorder marked by persistently high blood glucose levels. It is recognized as one of the major causes of both macrovascular and microvascular complications.^1,2 Individuals with diabetes tend to experience a higher incidence and faster progression of periodontal disease, largely due to their increased vulnerability to infections.³ Periodontal disease involves inflammation of the gums triggered by pathogenic bacteria, which results in gradual destruction of the alveolar bone surrounding affected teeth. Growing evidence supports a two-way relationship between DM and periodontal disease.^4–6 In diabetic patients, dental healing is often compromised. This impaired repair process involves reduced neutrophil function, limited fibroblast migration and proliferation, and hindered angiogenesis under diabetic conditions.^7,8 Consequently, diabetic individuals generally respond less favorably to periodontal therapy.^9,10 Elevated levels of advanced glycation end products (AGEs) are found in the gingival tissues of diabetic patients¹¹ contributing to oxidative stress and promoting more rapid tissue damage. These AGEs disrupt normal cell–matrix interactions by modifying extracellular matrix cross-linking, further delaying wound healing.¹² Traditionally, wound management has relied on medications combined with natural or synthetic materials that maintain a warm, moist environment to support healing and minimize microbial infection. Despite their widespread use, these conventional approaches often show limited efficacy because therapeutic agents can be absorbed systemically, diminishing their targeted action at the wound site.¹³ To overcome these limitations, researchers have investigated new materials and technologies for diabetic wound repair, including growth factors,¹⁴ engineered grafts,^15,16 hydrofibers,¹⁷ synthetic polymers,¹⁸ hydrocolloid dressings,¹⁹ and silver-based dressings.²⁰ Yet, these advanced therapies also face several obstacles—such as delayed epithelialization, allergic responses, irregular bioactive compound release, cytotoxicity, impaired angiogenesis, physiological rejection, and degradation of growth factors at diabetic ulcer sites.¹³ More recently, attention has shifted toward nanotechnology-driven solutions in clinical practice, primarily due to the unique characteristics of nanomaterials.²¹ With particle sizes typically between 1 and 100 nm, they possess a high surface-area-to-volume ratio and exhibit distinctive optical, electrical, and thermal properties.^22–25 These nanoscale materials show promise for improving treatment efficacy and reducing adverse effects in chronic diabetic wound care.²⁶

MXenes are a novel class of two-dimensional (2D) nanomaterials composed of transition metal carbides, nitrides, or carbonitrides. They follow the general formula M_n+1X_nT_x (n = 1–3), where M represents an early transition metal such as titanium (Ti), niobium (Nb), or molybdenum (Mo); X denotes carbon and/or nitrogen; and T_x corresponds to surface terminations like –O, –F, –Cl, or –OH.^27,28 Since their discovery,²⁹ MXenes have garnered increasing attention owing to their graphene-like features and superior physicochemical properties, including large specific surface area, high electrical conductivity, remarkable mechanical strength, hydrophilicity, and efficient photothermal conversion.^30,31 Owing to their unique properties, MXenes have emerged as promising candidates for energy conversion and storage, catalysis, antimicrobial membranes, and a wide range of biomedical applications.^{24,28,32–34}

To date, only a few types of MXenes have been explored for biological and biomedical applications,³⁵ such as Ti₂CT_x,³⁶ Ti₃C₂T_x,³⁷ and Nb₂CT_x.³⁸ Among these, Ti₃C₂T_x has been the most extensively investigated, primarily due to its relatively straightforward synthesis process. A notable attribute of MXenes is their strong near-infrared absorption, which shows promise in theranostic platforms and photothermal therapies, including combination cancer treatments.^39–44 Their hydrophilic nature and rich surface functionalities provide versatile sites for binding biologically active molecules, making them excellent candidates for biomedical interfacing.⁴⁵ However, a key concern surrounding the application of MXenes in biological systems is their potential cytotoxicity. Cytotoxicity refers to the capacity of a substance to damage or kill cells, typically through mechanisms involving oxidative stress, excessive generation of reactive oxygen species (ROS), and subsequent cellular apoptosis or necrosis.⁴⁶ While some toxicity stems from intrinsic biological variations such as genetic and metabolic differences, extrinsic factors related to the nanomaterial itself including chemical composition, solubility, size, surface chemistry, morphology, and aggregation play a significant role. Toxic substances may compromise cell membrane integrity, reduce cell viability, and hinder cellular proliferation.⁴⁷ Current evidence indicates that the cytotoxicity and biocompatibility of MXenes are highly dependent on multiple physicochemical and experimental factors, including synthesis method, oxidation state, surface functionalization, layer structure (single- versus multilayered), particle size, concentration, exposure duration, and route of administration. Both in vitro and in vivo studies consistently show that variations in these parameters determine whether MXenes elicit toxic or biocompatible responses.^46,48–51 Notably, pristine single-layer Ti₃C₂T_x nanosheets exhibit high compatibility with a variety of cell types and have shown promise as scaffolds for tissue culture applications.⁴⁸

Despite the growing scientific interest in MXenes for a wide range of technological applications, studies assessing their cytotoxicity remain relatively scarce. A MXene-based composite has been reported to detect hydrogen peroxide for the indication of periodontal disease⁵² and an injectable MXene-loaded hydrogel was investigated in the management of inflammation control and bone regeneration in severe periodontitis,⁵³ however, to the best of our knowledge, there is no investigation of the effect of MXene in diabetic gingival wound healing. In the present study, we prepared Ti₃C₂T_x MXene-based nanodrug formulations and investigated their therapeutic potential in promoting the healing of diabetic gingival wounds by regulating oxidative stress pathways and boosting glutathione activity in fibroblast cells isolated from diabetic rat models.

2. Materials and methods

2.1. Chemicals and materials

Ti₃AlC₂ (325 Mesh, purity: 99+%) was purchased from Nanografi (Ankara, Turkey). Lithium fluoride (LiF) and concentrated HCl were obtained from Thermo Fisher Scientific and Merck, respectively. 3-[4,5-Dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT), dimethyl sulfide (DMSO), 2′,7′-dichlorofluorescein diacetate (DCFH-DA) probes, malondialdehyde (MDA), thiobarbituric acid (TBA), N-butanol, tetramethoxypropane (TEP), O-phthalaldehyde, (OPA) probes, and N-ethylmaleimide (NEM) probes were purchased from Sigma-Aldrich. A Mouse/Rat Cytochrome c ELISA Kit (AB210575) was purchased from Abcam. An Annexin V-FITC/PI Apoptosis Kit (E-CK-A211) and Acridine Orange (A1301) were purchased from Elabscience and Invitrogen, respectively, and a blood glucose meter (GLUCOCARD G Black) was purchased from ARKRAY, Inc.

2.2. Instruments

Incubator (37 °C, Sensor CO₂ Sanyo MCO, Japan); refrigerated centrifuge (Harrier 18/80, Sanyo, Japan); spectrofluorometer (RF-5000, Shimadzu, Japan); digital scale (Japan); shaker (REAX2000, Iran); ELISA reader (Infinite 200 PRO, Tecan, Basel, Switzerland); and flowcytometer (FACS Calibure TM flow cytometer, BD Biosciences, USA) were used in this study. The structures of the MAX phase and MXene were analyzed using an X-ray diffractometer (XRD) (Bruker D2 advance, Germany) with monochromatized Cu-Kα radiation (λ = 1.54184 Å) generated at 30 kV and 10 mA, with a scan rate of 1° per min and a step size of 0.05°. The morphological structure of the materials was visualized by JSM-6010LA. X-ray photoelectron spectroscopy (XPS, the Thermo Scientific KAlpha) was used to analyze the compositions of the samples. The hydrodynamic diameters and zeta potential of aqueous samples in ultrapure water were determined at 25 °C using a Zetasizer Pro (Model MAL1256845, Malvern Panalytical Ltd, UK). Each test was repeated three times, and an averaged value was derived to ensure accuracy.

2.3. Synthesis of Ti₃C₂T_x MXenes

Ti₃C₂T_x MXenes were synthesized according to previous literature.^32,37 2 g of LiF was mixed with 40 ml of 6 M HCl and stirred for 10 minutes. Subsequently, 2 g of Ti₃AlC₂ powder was gradually introduced into the mixture and agitated at a temperature of 40 °C for 45 hours. Following the etching procedure, the sediments obtained were rinsed several times using deionized water and subjected to centrifugation at a speed of 3500 rpm for 5 minutes, until the pH approached nearly 6. Additional water was introduced to the sediment, which was then bath sonicated for 30 minutes under N₂ gas and centrifuged at a speed of 10 000 rpm for 30 minutes to collect the supernatant as few-layer MXenes. The final product was obtained by freeze-drying after 3 days.

2.4. Experimental design

2.4.1. Preparation of animals. We acquired Sprague-Dawley male rats (total number = 20, 5 animals per group) from Tehran Medical University in Tehran, Iran. The sample size was established based on a priori power analysis, utilizing an expected effect size derived from preliminary studies. This analysis indicated adequate power (≥80%) to identify statistically significant differences (α = 0.05) in the primary outcomes related to cell population metrics. This methodology ensures the collection of reliable data while strictly adhering to the 3Rs principle, which emphasizes the reduction of animal use in research. Furthermore, the subsequent cellular analyses involved the quantification of over 10 000 individual cells from these biological replicates, thereby increasing the statistical confidence in the resulting population data. These rats were all the same age, measuring around 8–10 weeks, and weighing approximately 150 g. They were housed under controlled conditions, including a temperature range of 20–12 °C, humidity levels between 50 and 60%, and a consistent 12-hour cycle of light and dark. Furthermore, the rats had unrestricted access to tap water and were given standard food.

Animals were divided into 4 groups which were treated as follows: Group 1: The control group received only sterile phosphate-buffered saline (PBS) and served as the healthy control group (CH). Group 2: the diabetic group received PBS and served as the disease control group (CD). Group 3: the diabetic group and gingival wound (CDD). Group 4: the diabetic group with gingival wound that received Ti₃C₂T_x MXene (CDDM). All experiments were approved by the research ethics committee of Iran University of Medical Sciences under the IR.IUMS.AEC.1403.009 Ethical code.

2.4.2. Preparation of animal models and surgical procedures. The animals were injected with streptozotocin (STZ) at a dose of 65 mg kg⁻¹ of body weight via intravenous administration. STZ induces diabetes within three days by destroying the beta cells. Diabetic and non-diabetic control groups were housed individually in metabolic cages, where their feeding and metabolism were monitored, and blood glucose levels were measured using a commercial glucometer. Rats with blood glucose levels exceeding 10 mmol L⁻¹ were classified as diabetic. Prior to anesthetic induction, the rats were fasted for 4 to 6 hours, while maintaining ad libitum access to water. Anesthesia was initiated by placing the animals in an induction chamber with 3.0–4.0% isoflurane vapor delivered in oxygen until all motor reflexes were abolished. For maintenance, the animals were transferred to a heated surgical surface set at 37 °C and were ventilated via a nose cone delivering 1.2–2.0% isoflurane.⁵⁴ Blood was collected from the tail vein of each animal without anesthesia between 9 : 00 and 10 : 00 AM using heparin-coated or EDTA-coated microhematocrit capillaries at baseline and on days 0, 7, 14, 21, and 28. Blood glucose levels were measured using a glucometer. The HA-8180T device assesses HbA1c levels using reversed-phase distribution exchange chromatography. A blood sample, diluted with a hemolysis and washing solution, is loaded onto the column, where it is fractionated and eluted into each Hb component using HPLC. The eluted components are quantified with an ELISA reader operating at wavelengths of 420 nm and 500 nm.⁵⁵

Operations were performed under sterile conditions. Rats were anesthetized by intracardiac injection of sodium pentobarbital. Anesthesia was maintained by inhalation of halothane (1.5–2.0 vol%). The root of the nose was shaved and disinfected using 70% ethanol. Local anesthesia was implemented using 2% xylocaine/epinephrine. A mucosal flap was made to expose the gum laterally. Dental holes were made at the eye level and 7 mm lateral toward the midline into the gum area, using a 1-mm diameter, 4-mm long slow-speed dental drill under sterile saline irrigation. The Ti₃C₂T_x MXene was inserted into the hole using a sterile spatula. Tetracycline hydrochloride paste was injected into the surgical site to prevent bacterial infection.⁵⁶

2.4.3. Preparation of the Ti₃C₂T_x MXene suspension for topical application. The dose per application was calculated based on the wound surface area and a target dose range of 330 µg mL⁻¹. This dose range was selected to achieve a local effective concentration (the IC50 of Ti₃C₂T_x MXene was 34.83 µg mL⁻¹, aiming for 175–350 µg mL⁻¹ in non-surgical tissue, after 72 h) while accounting for the small wound size and the potential for rapid clearance. A volume of 3 µL of the respective Ti₃C₂T_x MXene was carefully applied directly onto the wound surface using a micropipette.

2.5. Isolation of skin fibroblast cells

The tissue was extracted and washed in PBS and left for 10 minutes in an incubator at 37 °C with 5% CO₂ and 85% humidity, to enhance its adhesion to the plastic surface. DMEM culture medium was then carefully added. Tissue cultures were incubated at 37 °C in a humidified incubator with 5% CO₂ until confluent fibroblast layers were obtained.⁵⁷

2.6. Cellular toxicity assay test

2.6.1. Morphology of cells. Cells from the minced animal tissue were subjected to flow cytometry cell sorting. The crude cells were dissolved in 0.5 mL PBS. 100 µL of aliquots was redistributed into the BD flow cytometry tube. Results of the light scattering (forward/side scatter; FSC/SSC) were analyzed for at least 10 000 counts per sample in the flow cytometer (FSC indicated cell morphology and SSC showed cell granularity).⁵⁸

2.6.2. Cell viability. We used a rat model with a diabetic gingival wound to assess the in vivo toxicity of Ti₃C₂T_x MXene. We found that the IC50 of Ti₃C₂T_x MXene was 34.83 µg mL⁻¹, aiming for 330 µg mL⁻¹ in surgical tissue, and we used the IC50 for cellular toxicity factors (Fig. S1). Cells cultured from the minced rat diabetic gingival wound tissues were subjected to flow cytometry cell sorting based on cell morphology information (side scatter (SSC) and forward scatter (FSC)). The crude cells were dissolved in 0.5 mL of PBS. One hundred microliter aliquots were redistributed into BD flow cytometry tubes. The light scattering results were analyzed for at least 10 000 counts per sample using the flow cytometer.

Cell viability was evaluated using the MTT test. We prepared the cells (1 × 10⁴ cells per well) and incubated them in 96-well plates to a final volume of 50 ml. 20 mL of MTT was added to each well and then incubated for a supplementary 4 h at 37 °C. The purple-blue MTT formazan precipitate was dissolved in 100 ml of DMSO, and the absorbance at 570 nm was measured with an ELISA reader. Each test/group was examined with three replicates for each sample.⁵⁹

2.6.3. Reactive oxygen species assay. In this experiment, isolated cells (1 × 10⁶ cells) were placed in respiration buffer. Afterwards, 10 µM DCFH was added to the cellular suspension and then incubated for 15 min at 37 °C. In the next step, the fluorescence was measured using a Shimadzu RF-5000U fluorescence spectrophotometer with excitation and emission wavelengths of λ_excitation/λ_emission 488 nm/527 nm. Each group was examined with three replicates for each sample.⁶⁰

2.6.4. Lipid peroxidation (LPO) assay. The LPO was assayed by the determination of the amount of thiobarbituric acid reactive substances (TBARS) formed during the decomposition of lipid hydroperoxides on isolated cells (1 × 10⁶ cells) by following the absorbance at 532 nm in an ELISA reader analyzer by determining the MDA level following the manufacturer's instructions. Each group was examined with three replicates for each sample.⁶¹

2.6.5. Protein carbonyl content assay. Protein was precipitated by adding an equal volume of 20% TCA and centrifuged at 11 000 g for 5 min. The cells (1 × 10⁶ cells per well) were re-suspended in 10 mmol L⁻¹ 2,4-dinitrophenylhydrazine solution for 15–30 min at room temperature before 20% TCA was added. The samples were centrifuged at 11 000 g for 3 min. The carbonyl content was measured at 450 nm using an ELISA reader. Each group was examined with three replicates for each sample.⁶²

2.6.6. Glutathione assay. On isolated cells (1 × 10⁶ cells), the reduced form glutathione reductase (GSH) and glutathione peroxidase form (GSSG) are the most important scavengers of ROS that can be utilized as a biomarker of the redox balance. They were measured by OPA probe and NEM probe, respectively. Each sample was measured in quartz cuvettes using a Shimadzu RF-5000U fluorescence spectrophotometer set at λ_excitation/λ_emission 350/420 nm wavelengths. Each group was examined with three replicates for each sample.⁶³

2.6.7. Cytochrome c release assay. Cytochrome c release was determined at 450 nm according to the instructions provided by the manufacturer of the kit. All analysis stages were carried out using an ELISA reader at the desired concentrations in all groups. Each group was examined with three replicates for each sample.

2.6.8. Lysosome damage assay. The integrity of the lysosome membrane was established from the fluorescent dye (AO) redistribution; allocated cell (1 × 10⁶ cells) suspensions were stained with AO (5 µM) and the cells were separated from the incubation plate via centrifugation for 1 min at 1000 rpm. Subsequently, the cell pellet was placed in 2 mL of fresh DMEM medium. The wash process was performed twice to remove the shiny fluorescent dye from the media. AO redistribution in the cell's suspension was then determined using a Shimadzu RF-5000U fluorescence spectrophotometer set at λ_excitation/λ_emission 490 and 535 nm. Each group was examined with three replicates for each sample.⁶⁴

2.6.9. Apoptosis and necrosis assay. After the treatment, the cells (1 × 10⁶ cells) were stained with 5 ml of Annexin V and 5 ml of PI at room temperature for 20 min. The cells were diluted in the banding buffer (400 ml) and analyzed with flow cytometry. The fluorescence signals of Annexin V and PI were measured by flow cytometry in the FL1 and FL3 channels. Software version 1.2.5 was used and each determination was based on the mean fluorescence intensity of 10 000 counts and following the manufacturer's instructions.

2.7. Statistical analysis

All statistical analyses were conducted using GraphPad Prism Software, version 10.0 (San Diego, CA, USA). The data are presented as mean ± standard deviation (SD). The Shapiro–Wilk test was employed to assess the normality of the data distribution. For experiments comparing the mean values of three or more independent groups, a one-way analysis of variance (ANOVA) was performed initially. In cases where two independent factors were analyzed simultaneously (e.g., treatment and time), a two-way analysis of variance (ANOVA) was applied. To control for the family-wise error rate and to adjust P-values following the ANOVA, a Tukey's post-hoc test was conducted for all necessary pairwise comparisons. Statistical significance for all tests was established using a threshold of P < 0.05.

3. Results and discussion

Diabetes mellitus (DM) is a significant global health burden, currently affecting approximately 463 million adults globally.⁶⁵ It is a long-term systemic metabolic condition characterized by defective insulin secretion and/or activity, resulting in sustained high blood glucose levels and multiple microvascular complications.⁶⁶ People with DM frequently experience various oral health issues, such as tooth loss, slow wound healing, dry mouth (xerostomia), dental decay, burning mouth syndrome, oral lichen planus, and, in advanced cases, bacterial osteomyelitis of the jaw. These manifestations often complicate dental management and can adversely affect treatment outcomes.⁶⁷ Hyperglycemia is closely linked to both systemic and oral health deterioration. Poor glycemic control compromises the immune response, particularly the function of neutrophils, which are the body's primary defense against oral bacterial infections.⁶⁸ Evidence suggests that maintaining optimal blood glucose levels significantly reduces the risk of diabetes-associated complications, including those affecting oral health, as well as cardiovascular, neural, and ocular systems.

Individuals with diabetes often present a range of oral health disorders, the most common of which include bad breath (halitosis), impaired wound healing, dental cavities, dysfunction of the salivary glands, oral lichen planus, tongue lesions, various oral infections, and periodontal disease.⁶⁹ These complications are largely attributed to microvascular damage and sustained hyperglycemia. Epidemiological studies indicate that over 90% of individuals with DM develop oral health issues, and the prevalence of oral mucosal disorders is significantly higher among diabetic patients compared to non-diabetic populations.⁷⁰ With DM affecting roughly 8.5% of the global adult population, the burden of oral complications in this group warrants heightened clinical attention.⁷¹

2D nanomaterials exhibit exceptional properties, including high biocompatibility, potent antimicrobial effects, tunable phototherapeutic capabilities, and enhanced electrostimulation. These characteristics allow them to precisely modulate the wound microenvironment, leading to their widespread and successful application in tissue repair.⁷² Thus, in this research, we studied the effect of Ti₃C₂T_x MXene in gingival wound healing.

MXene was synthesized by selective etching of the aluminum (Al) layer from the Ti₃AlC₂ MAX phase. XRD analysis was used to investigate the structural properties of both the MAX phase and the resulting MXene (Fig. 1a). The diffraction pattern of the MAX phase exhibited sharp peaks at 2θ values of approximately 9.71° (002), 19.3° (004), and 39.10° (104), reflecting its well-ordered crystalline structure.³² In contrast, the disappearance of the (104) peak and the shift and broadening of the (002) peak in the Ti₃C₂T_x MXene confirmed the successful removal of the Al layer during the etching process. The morphology of MXene is shown in Fig. 1b. Delaminated few-layer MXene sheets were obtained after sonication. Similar to graphene oxide,⁷³ the MXene flakes are wrinkled and curved at the edges.³²

Fig. 1 (a) The XRD patterns of MAX phase and MXene, (b) SEM image of MXene, (c) the XPS survey scan of MXene, and the high-resolution XPS spectra of (d) Ti 2p, (e) C 1s, and (f) O1s of MXene.

The XPS analysis was carried out to identify the surface species and corresponding chemical states of the MXene. The survey spectrum in Fig. 1c confirms the presence of Ti, C, O, and F signals.³⁷ The deconvoluted high-resolution spectra for Ti2p, C1s, and O1s are presented in Fig. 1d–f. Analysis of the Ti2p_3/2 (2p_1/2) region shows four doublets assigned to Ti–C, Ti(II), Ti(III), and TiO₂, with binding energies of 455.78 eV (461.88 eV), 456.28 eV (462.38 eV), 457.38 eV (463.78 eV), and 459.48 eV (464.78 eV), respectively.³² The C1s spectrum is resolved into five components at 282.28 eV (C–Ti), 283.38 eV (C–Ti–O/F), 285.08 eV (C–C), 285.8 eV (C–O), and 286.88 eV (O–C=O). Similarly, the O1s envelope consists of peaks at 530.58 eV (Ti–O), 530.98 eV (C–Ti–O_x), 532.19 eV (C–Ti–(OH)_x), 533.28 eV (C–O), and 534.48 eV (adsorbed H₂O).

Dynamic light scattering (DLS) analysis showed that the hydrodynamic diameter of MXene was approximately 198 nm. The zeta potential measurements confirmed the high colloidal stability of the MXene which exhibited zeta potentials of −38.13 mV, consistent with their inherent negative surface charges.

In this study, we investigated the impact of Ti₃C₂T_x MXene on oxidative stress pathways in fibroblast cells. Previous studies have explored the biocompatibility and safety of Ti₃C₂T_x MXene in various vivo models. For instance, Nasrallah et al. employed a zebrafish embryo model to evaluate the toxicity of Ti₃C₂T_x nanosheets and reported that concentrations up to 50 µg mL⁻¹ did not impair neuronal or muscular activity.⁵¹ Zhang et al. implanted Ti₃C₂T_x MXene films into subcutaneous tissues and calvarial defect sites in rats, followed by micro-CT imaging and histological analysis, revealing favorable bone regeneration and osteoinductive properties without evidence of toxicity or inflammation.⁷⁴ Li et al. synthesized Bi₂S₃/Ti₃C₂T_x which exhibited excellent cytocompatibility and biocompatibility, promoted collagen fiber formation, and accelerated wound healing. Furthermore, the material's Schottky junction demonstrated outstanding biosafety in vivo.⁷⁵ Recently, Zahrabi et al. showed that a melt electrowritten (3-aminopropyl)triethoxysilane-modified Ti₃C₂T_x/polycaprolactone 3D scaffold enhanced the osteogenic differentiation of MC3T3-E1 preosteoblast cells.³⁷

A major obstacle in diabetic wound healing is the elevated susceptibility to bacterial and fungal infections around ulcerated regions, that can progress into severe complications such as foot amputation or even mortality. Multiple studies have shown that Ti₃C₂T_x MXene exhibits intrinsic antibacterial activity,⁷⁶ making it a promising candidate for diabetic ulcer management and the development of advanced wound dressings.

Hussein et al.⁷⁷ fabricated two Ti₃C₂T_x-based nanocomposites, Au/MXene and Au/Fe₃O₄/MXene, and assessed their photothermal therapeutic performance in MCF-7 human breast cancer cells. Both composites demonstrated comparable photothermal efficacy; however, the hybrid nanocomposites exhibited reduced in vivo toxicity relative to pristine MXene. Acute toxicity studies in zebrafish embryos further indicated lower embryonic mortality for the composite materials. These findings suggest that surface modification and hybridization can enhance the biocompatibility of MXenes. Despite these advances, there remains a scarcity of comprehensive in vivo investigations assessing both short-term and long-term biosafety, underscoring the need for further preclinical validation before clinical translation.

Wojciechowska et al. investigated the effects of Ti₃C₂T_x/poly L-lactide flakes on human malignant melanoma cells (A375, ATCC) and human immortal keratinocytes (HaCaT). Their results demonstrated that concentrations up to 375 mg L⁻¹ exhibited no cytotoxic effects.⁷⁸ Wang et al. coated Ti₃C₂T_x films with silken protein and assessed cytotoxicity using human skin fibroblast HSAS1 cells. The silk fibroin-coated MXene maintained approximately 99% cell viability after six days of incubation that confirmed no significant reduction in viability, indicating enhanced biocompatibility of the coated films.⁷⁹

As illustrated in the graph, HbA1c levels remained stable in the CH group (healthy control) throughout the 30-day period. In contrast, the CD group exhibited a progressive and statistically significant increase in HbA1c levels, peaking at approximately 12% by day 30. Statistical analysis indicates that the discrepancy between the groups reached a high level of significance (****P < 0.0001) from day 20 onwards, confirming the successful and sustained induction of the diabetic state in the study model. The elevated HbA1c levels in the CD group at the end of the 30 day period signify chronic dysregulation of glucose metabolism. Since HbA1c reflects the mean blood glucose concentration over a 2–3-month period, achieving a 12% threshold in this 30 day model not only confirms severe hyperglycemia but also establishes a rigorous environment for assessing wound healing interventions under high-oxidative stress conditions. These findings validate the robustness of the diabetic animal model for subsequent experimental phases (Fig. S1).

Blood glucose levels in the CH group remained relatively stable throughout the 28-day period (∼110–125 mg dL⁻¹). In contrast, the CD group developed a time-dependent and statistically significant hyperglycemic shift starting at day 7. Glucose levels increased to ∼250 mg dL⁻¹ by day 14 and then remained markedly elevated at ∼260–280 mg dL⁻¹ on days 21 and 28. The significance markers (from **P < 0.0001 to ****P < 0.0001) indicate robust statistical differences across the corresponding time points, supporting a clear divergence between CH and CD over time. The temporal pattern observed in the CD cohort suggests successful establishment of hyperglycemia and persistent impairment of glycemic control. The onset of glucose elevation at day 7, followed by attainment of high and sustained concentrations (∼250–280 mg dL⁻¹) through day 28, indicates that the diabetic phenotype is not merely transient but progresses toward a stable dysregulated state. From a model-validation perspective, these glucose measurements provide acute, high-resolution evidence of glycemic dysregulation, which when interpreted alongside long-term markers such as HbA1c strengthens the overall confirmation of the induced diabetic condition (Fig. S2).

To establish a biologically relevant concentration for subsequent studies, the cytotoxicity of Ti₃C₂T_x MXene on standard fibroblast cells was quantified. Cells were treated with a concentration gradient of Ti₃C₂T_x MXene ranging from 0 to 100 µg mL⁻¹ (0, 2.5, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100 µg mL⁻¹). Cell viability data were analyzed using nonlinear regression via software. This analysis determined the half-maximal inhibitory concentration (IC50) to be 34.83 ± µg mL⁻¹ (wound surface area and target dose range 330 µg mL⁻¹ after 72 h) (Fig. S3). Preliminary range-finding indicated that concentrations below 20 µg mL⁻¹ showed minimal cytotoxic effects, whereas doses exceeding 60 µg mL⁻¹ resulted in severe cell death. Therefore, the calculated IC50 of 34.83 µg mL⁻¹ was adopted as the standard concentration for assessing the biological effects of Ti₃C₂T_x MXene in this study.

Wound dimensions were quantitatively evaluated using a caliper on days 0, 5, 15, and 30 days, following wound induction. Caliper measurements were applied in the early stage because the wound surface was still open and dynamically changing. As observed, the CH and CD groups showed neither wound cavity formation nor Ti₃C₂T_x MXene exposure throughout the study period, resulting in negligible changes in macroscopic wound dimensions. In the CDD group, only wound cavity formation was evident. Since this group did not receive Ti₃C₂T_x MXene treatment, the wound exhibited limited improvement in surface dimensions during the observation window. In contrast, the CDDM group demonstrated a clear healing trend, as wound dimensions decreased progressively up to day 15, followed by complete recovery by day 30. Overall, these macroscopic results provide quantitative evidence that Ti₃C₂T_x MXene enhances wound healing under diabetic conditions (Table S4).

In our study in IC50, analysis of morphology (population∼10 000) using normal counts (Q₁), FSC (Q₂) and SSC (Q₁) for cells on days 0, 5, 15 and 30 was conducted. In Q₁ on day 0 in CH∼97.07%, CD∼80.69%, CDD∼57.93%, CDDM∼67.31%, on day 5 in CDDM∼72.56%, on day 15 in CDDM∼78.59%, and on day 30 in CDDM∼86.81%. In Q₂ on day 0 in CH∼0.13%, CD∼16.86%, CDD∼37.25%, CDDM∼30.89%, on day 5 in CDDM∼25.58%, on day 15 in CDDM∼19.64%, and on day 30 in CDDM∼11.71%. In Q₃ on day 0 in CH∼2.27%, CD∼2.09%, CDD∼0.22%, CDDM∼1.33% on day 5 in CDDM∼0.94%, on day 15 in CDDM∼1.46%, and on day 30 in CDDM∼1.11% (Fig. 2). The flow cytometric analysis of approximately 10 000 cells, assessing viability (Q₁), morphological changes (Q₂), and granularity changes (Q₃), revealed distinct temporal patterns within the study groups. Notably, the CDDM group exhibited a significant shift over 30 days. In Q₁ (viable cells area), CDDM showed a progressive increase from 67.31% on day 0 to 86.81% on day 30, indicating enhanced cell viability over time. Concurrently, CDDM demonstrated a decrease in Q₂ (changed shape cells area), dropping from 30.89% on day 0 to 11.71% on day 30, suggesting a reduction in cells undergoing morphological alteration. While Q₃ (changed granularity cells area) for CDDM fluctuated, it remained low, with a slight decrease from 1.33% on day 0 to 1.11% on day 30. These findings collectively suggest that the treatment promotes cell viability and a reduction in aberrant cellular morphology and granularity in the CDDM group by day 30.

Fig. 2 Morphological analysis of the population of gingival cells (FSC/SSC) using flow cytometry.

As depicted in Fig. 3a, cell viability showed a slight yet significant decrease on day 0 in the CDD group (*P < 0.05) and a more pronounced reduction in the CDDM group (****P < 0.0001). Subsequently, from day 5 onward, a notable decline in cell viability was observed in the CDD and CDDM groups (****P < 0.0001). This decreasing trend continued until day 15, with significant reductions noted in the CD, CDD and CDDM groups (****P < 0.0001). By day 30, cell viability had further decreased in the CD and CDD groups (****P < 0.0001). Interestingly, on the same day, an increase in cell viability was registered in the CDDM group (*P < 0.05) compared to the control healthy group (CH).

Fig. 3 (a) Detection of cell viability, (b) detection of reactive oxygen species (ROS), (c) detection of lipid peroxidation (LPO), (d) detection of carbonyl protein content in control health (CH), control diabetic (CD), diabetic and surgical gingivae (CDD), and diabetic, and surgical gingivae and treated by Ti₃C₂T_x MXene (CDDM) groups in fibroblast cells after 0, 5, 15 and 30 days. Data are represented as mean ± SD for data determined from three separate experiments. Values are represented as mean ± SD (n = 5). ns = no significant, *P < 0.05 and ****P < 0.0001 compared with the control health (CH) group.

Our results showed that Ti₃C₂T_x MXene treatment reduced ROS levels and improved fibroblast viability (CDDM) over a 30 day period (Fig. 3b). This finding aligns with the observations of Song et al.⁸⁰ who reported that diabetic wound microenvironments often produce excessive ROS due to activation of oxidative stress pathways, resulting in decreased cell survival. Biochemically, succinate dehydrogenase plays a crucial role in the mitochondrial electron transport chain, where it catalyzes the conversion of succinate to fumarate, accompanied by the reduction of flavin adenine dinucleotide (FAD) to its reduced form, flavin adenine dinucleotide dehydrogenated (FADH₂). The resulting electrons are transferred through the respiratory chain to oxygen, regenerating FAD. This enzymatic activity serves as an important indicator of mitochondrial function and oxidative metabolism. In our research, on days 0 and 5, significantly elevated levels of ROS were detected in the CDDM, CDD, and CD groups (****P < 0.0001). On day 15, an increase was observed in the CD and CDD groups (****P < 0.0001), whereas the CDDM group (**P < 0.01) showed a decrease. By day 30, substantial increases were observed in the CD and CDD groups (****P < 0.0001), while the CDDM group showed an opposite trend, suggesting a potential mitigating or stabilizing effect of the MXene treatment over the longer term (Fig. 3b). All these observed increases in ROS levels across days 0, 5, 15, and 30 were consistently compared against the control healthy group (CH). The results in the CDDM group indicate the role of Ti₃C₂T_x MXene in decreasing the level of ROS (Fig. 3b). By attenuating ROS production, a strong correlation was observed between Ti₃C₂T_x MXene exposure and alterations in key components of the cell death signaling pathway, suggesting a potential regulatory role in fibroblasts and support cell survival, even under prolonged exposure. Collectively, our results suggest that Ti₃C₂T_x MXene exerts protective effects against oxidative stress; thereby enhancing cell viability in wound-healing contexts, particularly within diabetic gingival wounds in rat models.

Pierce et al.⁸¹ reported that ROS promote lipid and protein peroxidation in diabetic wound tissues, thereby activating multiple inflammatory and oxidative stress pathways, including interleukin-1β (IL-1β), tumor necrosis factor-α (TNF-α), cytokines, nuclear factor erythroid 2–related factor 2 (Nrf-2), and nuclear factor kappa B (NF-κB). Tiwari et al.⁸² further demonstrated a significant clinical association between lipid peroxidation and hyperglycemia, with higher fasting blood glucose and elevated glycated hemoglobin/fasting plasma glucose (HbA1c/FPG) ratios correlating strongly with increased oxidative stress in diabetic patients. ROS-mediated lipid membrane oxidation is believed to disrupt the mitochondrial electron transport chain, initiating cell death signaling cascades.

As shown in Fig. 3c, the MDA levels, that indicate a significant increase in lipid peroxidation compared to the control healthy (CH) group, exhibited significant changes in the CD, CDD and CDDM groups. Specifically, on days 0 and 5, the CD, CDD, and CDDM groups showed alterations (****P < 0.0001) compared with the CH group. After 15 days, the CD and CDD groups continued to display a highly significant increase (****P < 0.0001), and the CDDM group also presented a dramatic change (*P < 0.05) when compared to the CH group. Furthermore, our findings indicated a slight but statistically significant decrease in lipid peroxidation levels within the CDDM group after 30 days.

Despite this reduction, it was not sufficient to prevent cell injuries. In our study, lipid peroxidation was assessed by measuring the formation of thiobarbituric acid reactive substances (TBARS) in fibroblasts exposed to Ti₃C₂T_x MXene (Fig. 3c). After 30 days, fibroblasts in the Ti₃C₂T_x MXene-treated group exhibited markedly reduced lipid peroxidation compared to CH, indicating a protective effect against oxidative membrane damage.

The data in Fig. 3d show that protein carbonyl levels significantly increased in the CD (**P < 0.01), CDD and CDDM (****P < 0.0001) groups, and marked cell injuries occurred in diseased cells compared to the CH group on day 0. On day 5, protein carbonyl levels increased in the CD (***P < 0.001), CDD, and CDDM (****P < 0.0001) groups compared to the CH group. On day 15, protein carbonyl levels significantly increased in the CD and CDD groups (****P < 0.0001), whereas levels decreased in the CDDM group (***P < 0.001) compared to the CH group. On day 30, protein carbonyl levels significantly increased in the CD and CDD groups (****P < 0.0001), while the CDDM group showed no significant change compared to the control group.

Their findings emphasized that therapeutic strategies targeting precise modulation and control of inflammation during the healing process may offer considerable benefits for managing diabetic and other chronic wounds. The observed malondialdehyde (MDA) levels were strongly associated with ROS generation and consequent lipid membrane damage. Likewise, the increase in protein carbonyl content likely reflects oxidative modification of cellular proteins by elevated ROS, a process considered a critical indicator of oxidative stress and frequently linked to loss of protein function.²⁶ In fibroblast cells exposed to Ti₃C₂T_x MXene for 30 days, carbonyl protein levels remained markedly lower compared with the CD group (Fig. 3d). As highlighted by Boniakowski et al. elevated ROS not only contributes to the onset and progression of diabetic wounds, but also intensifies the inflammatory response, thereby impairing tissue repair and reducing the efficacy.

A previous study⁸² showed that GSH serves as a key cellular antioxidant defense, capable of directly interacting with ROS in non-enzymatic reactions, resulting in its oxidation to GSSG. GSSG can negatively impact cells through two primary pathways: it may function as a toxic compound by forming conjugates with cellular proteins, or it can be converted back to its reduced form by glutathione reductase before being either exported via glutathione S-transferase or involved in nonenzymatic adduct formation.

Fig. 4a illustrates the content of reduced GSH. On days 0 and 5, significant differences were observed in the CD, CDD and CDDM groups (****P < 0.0001) compared to the control healthy (CH) group. On days 15 and 30, the CD and CDD groups exhibited a significant decrease in GSH content (****P < 0.0001), while the CDDM group also showed significant reductions (**P < 0.01 and *P < 0.05, respectively) relative to the CH group. A slight decrease was also noted in the CDDM group on these later days compared to the other diseased groups.

Fig. 4 (a) Detection of glutathione reductase (GSH), (b) detection of glutathione peroxidase (GSSG), (c) detection of release cytochrome c, and (d) detection of lysosomal membrane damage in control health (CH), control diabetic (CD), diabetic and surgical gingivae (CDD), and diabetic, surgical gingivae and treated by Ti₃C₂T_x MXene (CDDM) groups in fibroblast cells after 0, 5, 15 and 30 days. Data are represented as mean ± SD of data determined from three separate experiments. Values are represented as mean ± SD (n = 5). ns = no significant, *P < 0.05; **P < 0.01 and ****P < 0.0001 compared with the control health (CH) group.

On days 0 and 5, significant differences were observed in the CD, CDD, and CDDM groups (****P < 0.0001) compared to the control healthy (CH) group. On days 15 and 30, the CD and CDD groups exhibited a significant decrease in GSH content (****P < 0.0001), indicating a sustained state of oxidative stress, while the CDDM group also showed a significant reduction (**P < 0.01) after 15 days. However, the CDDM group did not show a significant change on day 30.

As illustrated in Fig. 4a and b, exposure to Ti₃C₂T_x MXene led to changes in the levels of both GSH and GSSG in fibroblasts from the gingival tissue of diabetic animals. Glutathione is recognized as a vital cellular antioxidant, with its GSH capable of directly neutralizing ROS through nonenzymatic reactions, resulting in its oxidation to GSSG. The GSSG can be detrimental to cells by acting as a toxic compound, forming conjugates with proteins, or participating in nonenzymatic adduct formation. Alternatively, it can be recycled back to GSH via GSSG before being exported from the cell through glutathione S-transferase.⁸³ Under several pathological conditions, including diabetes and various cancers, apoptosis serves as a fundamental mechanism of cell death, characterized by features such as cell shrinkage, nuclear fragmentation, chromatin condensation, and double-stranded DNA fragmentation, whereas necrosis is typically associated with cell membrane damage. Apoptosis is a tightly controlled cellular mechanism initiated by intracellular signaling pathways in response to stress conditions like elevated glucose levels, oxygen deprivation, or increased temperature, ultimately resulting in programmed cell death. According to Saelens et al.⁸⁴ mitochondrial swelling disrupts the configuration of the mitochondrial permeability transition (MPT) pore, leading to the release of pro-apoptotic molecules such as cytochrome c, Smac/DIABLO, and the serine protease HtrA2/Omi.⁸⁵

Fig. 4c provides details on the release of cytochrome c. On day 0, a slight increase in cytochrome c release was observed in the CDD and CDDM groups (**P < 0.01) relative to the CH group. On day 5, a slight increase in cytochrome c release was observed in the CD (**P < 0.01) and CDD (****P < 0.0001) groups relative to the CH group. However, by days 15 and 30, the pattern of release had evolved significantly. A dramatic increase in cytochrome c release was noted in the CD and CDD groups (****P < 0.0001). In contrast, the CDDM group showed no increase compared to the CH group on days 5, 15, and 30. Permeabilization of the mitochondrial outer membrane triggers the release of cytochrome c from the intermembrane space, a critical event in the initiation of cell death. Cytochrome c, carrying a strong positive charge, binds to negatively charged lipids on the outer surface of the inner mitochondrial membrane. This release is closely linked to ROS production. Our findings showed that exposure to Ti₃C₂T_x MXene reduced the release of cytochrome c into the cytoplasm of fibroblasts in the CDDM group. In contrast, the diabetic control group exhibited pronounced mitochondrial outer membrane permeabilization facilitating substantial cytochrome c release from the intermembrane space.⁸⁴

The release of cytochrome c into the cytoplasm is a key step in triggering the activation of caspases. Once there, with the help of Apoptotic Protease Activating Factor-1 (Apaf-1) and ATP, cytochrome c recruits pro-caspase-9. This forms a complex called the apoptosome which then activates caspase-9. The activated caspase-9 subsequently activates the downstream effectors caspase-3. Moreover, proteins like Smac/DIABLO and HtrA2/Omi promote apoptotic cell death by suppressing the activity of Inhibitor of Apoptosis Proteins (IAPs).⁸⁶

Exposure to ROS can compromise the integrity of lysosomes by causing lipid peroxidation of their membranes. The resulting proton leakage leads to lysosomal alkalinization, which is believed to contribute to many diseases because an appropriate acidic pH is crucial for lysosomal function. Consequently, strategies to restore lysosomal pH, either through compensatory biological responses or therapeutic interventions, are expected to offer significant benefits. Interestingly, it is worth noting that not all lysosomes respond to oxidative stress in the same way or even maintain the same luminal pH.⁸⁷

Fig. 4d illustrates the level of lysosomal damage. A statistical difference in lysosomal damage was observed between the control healthy (CH) group and the CD, CDD (****P < 0.0001), and CDDM (*P < 0.05) groups after 0 and 5 days. Furthermore, over the 15- and 30-days period, a dramatic increase in lysosomal damage was evident in both the CD and CDD groups (****P < 0.0001) compared to the CH group. No lysosomal damage was observed in the CDDM group compared to the CH group.

Apoptosis is an essential cellular mechanism involved in the progression of various diseases, including diabetes and cancer. It is defined by specific morphological changes such as cell contraction, chromatin condensation, nuclear fragmentation, and the cleavage of double-stranded DNA. Unlike necrosis, which is marked by membrane rupture and cell lysis, apoptosis follows a regulated pathway. This process is generally initiated by intracellular stressors such as elevated glucose levels, oxygen deprivation, or heat stress, ultimately leading to programmed cell death.⁸⁸

Apoptosis/necrosis pathways were analyzed using annexin V/PI double staining in which the apoptosis was quantified by the externalization of phosphatidylserine (PS). PI (stains the nuclear) was used as an indicator of membrane integrity at 0, 5, 15 and 30 days. The results showed cell death signaling. The cell viability counts (Q₁) on day 0 were CH∼99.67%, CD∼99.37%, CDD∼29.26%, and CDDM∼57.48%, on day 5 were CH∼99.50%, CD∼82.86%, CDD∼23.3%, and CDDM∼73.38%, on day 15 were CH∼99.13%, CD∼79.63%, CDD∼12.97%, and CDDM∼84.29%, and on day 30 were CH∼98.65%, CD∼ 75.52%, CDD∼8.42%, and CDDM∼92.13%. The necrotic cell counts (Q₂) on day 0 were CH∼0.13%, CD∼0.10%, CDD∼5.14%, and CDDM∼21.36%, on day 5 were CH∼0.11%, CD∼8.37%, CDD∼21.39%, and CDDM∼9.97%, on day 15 were CH∼0.12%, CD∼19.91%, CDD∼16.09%, and CDDM∼0.10%, and on day 30 were CH∼0.12%, CD∼23.72%, CDD∼15.82%, and CDDM∼2.65%. The early apoptotic cell counts (Q₃) on day 0 were CH∼0.12%, CD∼0.29%, CDD∼47.27%, and CDDM∼14.66%, on day 5 were CH∼0.19%, CD∼5.07%, CDD∼55.27%, and CDDM∼0.76%, on day 15 were CH∼0.34%, CD∼0.08%, CDD∼68.51%, and CDDM∼1.62%, and on day 30 were CH∼0.30%, CD∼0.50%, CDD∼75.28%, and CDDM∼4.62%. The late apoptotic cell counts (Q₄) on day 0 were CH∼0.08%, CD∼ 0.24%, CDD∼18.18%, and CDDM∼6.39%, on day 5 were CH∼0.20%, CD∼ 3.27%, CDD∼0.19%, and CDDM∼5.71%, on day 15 were CH∼0.4%, CD∼0.00%, CDD∼2.38%, and CDDM∼13.93%, and on day 30 were CH∼0.93%, CD∼0.00%, CDD∼0.47%, and CDDM∼0.58% (Fig. 5).

Fig. 5 Effect of Ti₃C₂T_x MXene on apoptosis and necrosis. Detection of apoptosis/necrosis pathways using annexin V/PI double staining, in which the apoptosis was quantified by the externalization of phosphatidylserine (PS). PI (stains the nuclear) was used as an indicator of membrane integrity on days 0, 5, 15 and 30 in the control health (CH), control diabetic disease (CD), diabetic diseases and surgical tooth gum (CDD), and diabetic diseases and surgical gingiva and treatment by Ti₃C₂T_x MXene (CDDM) fibroblast cell groups. Data are represented as mean ± SD of the data determined from three separate experiments. Values are represented as mean ± SD (n = 5). ns = no significant, *P < 0.05; **P < 0.01 and ***P < 0.001 compared with the control health (CH) group.

Apoptosis was quantified by phosphatidylserine (PS) externalization and assessed using annexin V/propidium iodide (PI) double staining on day 0, 5, 15, and 30. PI, which stains the nucleus, served as an indicator of membrane integrity, allowing differentiation between apoptotic and necrotic cells. The current findings reveal that exposure to Ti₃C₂T_x MXene showed a notable association with variations in apoptosis-related signaling rather than direct modulation of these pathways. Apoptosis is a fundamental biological process contributing to the resolution of inflammation and guiding the transition of granulation tissue toward mature scar formation during wound healing. In the context of diabetes, compromised wound healing continues to pose a major clinical concern, predominantly linked to disrupted apoptotic homeostasis resulting from inadequate glycemic regulation. The observed correlation between Ti₃C₂T_x MXene presence and apoptosis-associated responses highlights a promising avenue for further investigation into its potential implications for enhancing wound-healing dynamics under diabetic conditions.

4. Conclusion

The present study demonstrates that Ti₃C₂T_x MXene exhibits noteworthy biocompatibility and shows a clear association with well-regulated cellular and mitochondrial responses in fibroblast cells derived from abnormal gingival tissue of diabetic animal models. The experimental evidence revealed correlations between Ti₃C₂T_x MXene exposure and controlled oxidative stress, sustained cell viability, and lowered intracellular ROS levels. These relationships were accompanied by preservation of glutathione balance, stabilization of mitochondrial membrane potential, and moderation of cytochrome c-associated caspase activation, collectively aligning with a more regulated apoptotic profile in abnormal cells. Taken together, the findings support further investigation of Ti₃C₂T_x MXene as a potentially valuable candidate for developing advanced nano-enabled therapeutic formulations such as topical medicines, gels, creams, or powders aimed at improving wound healing under diabetic conditions. Furthermore, integration of such approaches into proactive oral healthcare frameworks, supported by interdisciplinary collaboration, may contribute to enhanced oral health outcomes among diabetic patients.

Author contributions

PN: conceptualization, investigation, methodology, characterization, data management, validation, visualization, original draft preparation, review and editing, and resources. MN: conceptualization, investigation, methodology, characterization, data management, validation, visualization, original draft preparation, review and editing, and resources. AR: original draft preparation, review and editing. SA: original draft preparation, characterization, data management, review and editing.

Conflicts of interest

The authors declare that they have no competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The authors declare that the data supporting the findings of this study are available within the paper and its supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d6tb00505e.

Acknowledgements

This work was financially supported by the FLAG-ERA grant [GRAPH-OCD], provided by the Scientific and Technological Research Council of Turkey (TUBITAK) [223N171].

References

1. S. A. J. Mossavi, J. F. Ramandi, F. Ghanbary, N. Hosseini, P. Naserzadeh, M. Akhshik, F. R. Tay and B. Ashtari, Nano, 2024, 2450047.
2. M. Brownlee, Nature, 2001, 414, 813–820.
3. G. L. King, J. Periodontol., 2008, 79, 1527–1534.
4. E. Lalla and P. N. Papapanou, Nat. Rev. Endocrinol., 2011, 7, 738–748.
5. W. S. Borgnakke, P. V. Yl€ostalo, G. W. Taylor and R. J. Genco, J. Periodontol., 2013, 84, S135–S152.
6. T. C. Simpson, J. C. Weldon, H. V. Worthington, L. MacDonald, J. C. Weldon, I. Needleman, Z. Iheozor-Ejiofor, S. H. Wild, A. Qureshi, A. Walker, V. A. Patel, D. Boyers and J. Twigg, Cochrane Database Syst. Rev., 2022, 4(4), CD004714.
7. O. Z. Lerman, R. D. Galiano, M. Armour, J. P. Levine and G. C. Gurtner, Am. J. Pathol., 2003, 162, 303–312.
8. V. Falanga, Lancet, 2005, 366, 1736–1743.
9. M. Christgau, K. D. Palitzsch, G. Schmalz, U. Kreiner and S. Frenzel, J. Clin. Periodontol., 1998, 25, 112–124.
10. L. Kardeşler, N. Buduneli, Ş. Çetinkalp, D. Lappin and D. F. Kinane, Inflammation Res., 2011, 60, 143–151.
11. J. Katz, I. Bhattacharyya, F. Farkhondeh-Kish, F. Perez, R. Caudle and M. Heft, J. Clin. Periodontol., 2005, 32, 40–44.
12. B. L. Mealey and T. W. Oates, J. Periodontol., 2006, 77, 1289–1303.
13. P. Palumbo and L. J. Melton III, Diabetes Am., 1995, 401.
14. J. P. Yadav, D. K. Patel, A. Verma and P. Pathak, Obesity Med., 2024, 47, 100535.
15. E. H. Espensen, B. P. Nixon, L. A. Lavery and D. G. Armstrong, J. Am. Podiatric Med. Assoc., 2002, 92, 395–397.
16. Y. Teng, Y. Li, J. Wang, K. Yang, Y. Zhang, Y. Wang, J. Tian, B. Ma, J. Wang and X. Yan, Diabetes, Obes. Metab., 2010, 12, 307–315.
17. L. Chen, P. Wu, Y. Zhu, H. Luo, Q. Tan, Y. Chen, D. Luo and Z. Chen, APL Bioeng., 2025, 9.
18. M. Mir, M. N. Ali, A. Barakullah, A. Gulzar, M. Arshad, S. Fatima and M. Asad, Prog. Biomater., 2018, 7, 1–21.
19. W. S. Tan, P. Arulselvan, S.-F. Ng, C. N. Mat Taib, M. N. Sarian and S. Fakurazi, BMC Complementary Altern. Med., 2019, 19, 20.
20. A. Montaser, A. Abdel-Mohsen, M. Ramadan, A. Sleem, N. Sahffie, J. Jancar and A. Hebeish, Int. J. Biol. Macromol., 2016, 92, 739–747.
21. L. Hou, X. Zhang and H. Du, Diabetes/Metab. Res. Rev., 2023, 39, e3638.
22. M. Namvari, C. S. Biswas, Q. Wang, W. L. Liang and F. J. Stadler, J. Colloid Interface Sci., 2017, 504, 731–740.
23. G. M. Choi, M. Park, Y. H. Shim, S. Y. Kim, M. Namvari and H. S. Lee, Adv. Funct. Mater., 2022, 32, 2202564.
24. A. Altan and M. Namvari, 2D Mater., 2023, 10, 042001.
25. A. Bahadoran, N. Ajinkya, M. Sharghi, F. Hasanvandian, Y. Wang, H. Chen, M. Namvari, B. Kakavandi, E. Marsili, M. Galluzzi and S. Ramakrishna, J. Mater. Chem. A, 2024, 12, 14619–14635.
26. V. Vijayakumar, S. K. Samal, S. Mohanty and S. K. Nayak, Int. J. Biol. Macromol., 2019, 122, 137–148.
27. M. Namvari and B. K. Chakrabarti, Adv. Colloid Interface Sci., 2024, 331, 103208.
28. R. Afshar Ghotli, M. Namvari and B. K. Chakrabarti, Adv. Mater. Technol., 2026, e01403.
29. M. Naguib, M. Kurtoglu, V. Presser, J. Lu, J. Niu, M. Heon, L. Hultman, Y. Gogotsi and M. W. Barsoum, in MXenes, Jenny Stanford Publishing, 2023, pp. 15–29.
30. M. Namvari, T. Inan and A. Altan, Graphene 2D Mater., 2023, 8, 5–26.
31. A. Ben Ayed and M. Namvari, Adv. Compos. Hybrid Mater., 2025, 8, 410.
32. A. Güngör, M. Namvari, A. B. Ayed and E. Erdem, Small, 2025, e05698.
33. R. Afshar Ghotli, Z. Kudaş, R. Zaidi, K. B. Dönmez, Z. Çobandede, A. Ben Ayed, M. Namvari, B. Alkhateab, S. Çelik, F. S. Mjalli, T. Y. İnan, M. K. Bayazıt, S. H. Soytaş, N. P. Brandon and B. K. Chakrabarti, ChemCatChem, 2026, e01280.
34. A. Ben Ayed, M. Masoudi and M. Namvari, FlatChem, 2026, 57, 101052.
35. M. Soleymaniha, M.-A. Shahbazi, A. R. Rafieerad, A. Maleki and A. Amiri, Adv. Healthcare Mater., 2019, 8, 1801137.
36. Z. Liu, C. Sun, M. Xu and W. Huang, Mater. Lett., 2024, 365, 136437.
37. M. Zahrabi, M. Altunbek, S. Celik, M. Namvari and B. Koc, Biofabrication, 2025, 17(4), 045011.
38. Y. He, H. Sun, Y. Wang, Y. N. Yu, C. Mu and L. Chen, Chem. Eng. J., 2024, 485, 150047.
39. Z. K. Li, Y. Wei, X. Gao, L. Ding, Z. Lu, J. Deng, X. Yang, J. Caro and H. Wang, Angew. Chem., Int. Ed., 2020, 59, 9751–9756.
40. H. Lin, X. Wang, L. Yu, Y. Chen and J. Shi, Nano Lett., 2017, 17, 384–391.
41. C. Dai, H. Lin, G. Xu, Z. Liu, R. Wu and Y. Chen, Chem. Mater., 2017, 29, 8637–8652.
42. G. Liu, J. Zou, Q. Tang, X. Yang, Y. Zhang, Q. Zhang, W. Huang, P. Chen, J. Shao and X. Dong, ACS Appl. Mater. Interfaces, 2017, 9, 40077–40086.
43. X. Han, J. Huang, H. Lin, Z. Wang, P. Li and Y. Chen, Adv. Healthcare Mater., 2018, 7, 1701394.
44. C. Yang, D. Xu, W. Peng, Y. Li, G. Zhang, F. Zhang and X. Fan, Nanoscale, 2018, 10, 15387–15392.
45. H. Vojoudi and M. Soroush, Adv. Healthcare Mater., 2025, 14, 2500359.
46. G. P. Lim, C. F. Soon, N. L. Ma, M. Morsin, N. Nayan, M. K. Ahmad and K. S. Tee, Environ. Res., 2021, 201, 111592.
47. J. Liu and C. N. Pope, in An Introduction to Interdisciplinary Toxicology, Elsevier, 2020, pp. 285–293.
48. B. E. Scheibe, J. K. Wychowaniec, M. Scheibe, B. Peplińska, M. Jarek, G. Nowaczyk and Ł. Przysiecka, ACS Biomater. Sci. Eng., 2019, 5, 6557–6569.
49. A. Jastrzębska, A. Szuplewska, T. Wojciechowski, M. Chudy, W. Ziemkowska, L. Chlubny, A. Rozmysłowska and A. Olszyna, J. Hazard. Mater., 2017, 339, 1–8.
50. A. Jastrzębska, A. Szuplewska, A. Rozmysłowska-Wojciechowska, M. Chudy, A. Olszyna, M. Birowska, M. Popielski, J. A. Majewski, B. Scheibe and V. Natu, 2D Mater., 2020, 7, 025018.
51. G. K. Nasrallah, M. Al-Asmakh, K. Rasool and K. A. Mahmoud, Environ. Sci.: Nano, 2018, 5, 1002–1011.
52. K. Wang, X. Zheng, M. Qi, W. Zhang, J. Du, Q. Han, C. Li, B. Dong, L. Wang and L. Xu, Sens. Actuators, B, 2023, 390, 133955.
53. Y. Zhou, M. Li, H. Zheng, B. Huang and G. Liu, J. Mater. Chem. B, 2025, 13, 10225–10238.
54. E. Asadpour, H. R. Sadeghnia, A. Ghorbani, M. Sedaghat and M. T. Boroushaki, J. Nanopart. Res., 2016, 18, 14.
55. A. Alshalani, N. AlAhmari, H. A. Amin, A. Aljedai and H. AlSudais, J. Clin. Med., 2025, 14.
56. F. Khodagholi, N. Ansari, M. Amini and S. K. Tusi, Cell Stress Chaperones, 2012, 17, 409–422.
57. A. Weir, P. Westerhoff, L. Fabricius, K. Hristovski and N. Von Goetz, Environ. Sci. Technol., 2012, 46, 2242–2250.
58. M. Duvvuri, S. Konkar, K. H. Hong, B. S. Blagg and J. P. Krise, ACS Chem. Biol., 2006, 1, 309–315.
59. E. Afjeh-Dana, P. Naserzadeh, H. Nazari, F. Mottaghitalab, R. Shabani, N. Aminii, B. Mehravi, F. T. Rostami, M. T. Joghataei and K. Mousavizadeh, Int. J. Biol. Macromol., 2019, 129, 1034–1039.
60. H. Ahmadi, Z. Nayeri, Z. Minuchehr, F. Sabouni and M. Mohammadi, PLoS One, 2020, 15, e0233088.
61. F. Shaki, M.-J. Hosseini, M. Ghazi-Khansari and J. Pourahmad, Biochim. Biophys. Acta, Gen. Subj., 2012, 1820, 1940–1950.
62. F. Dong, C. Fang, X. Yang, X. Zhang, F. Lopez and J. Ren, Diabetologia, 2006, 49, 1421–1433.
63. C. Sadegh and R. P. Schreck, Murj, 2003, 8, 39–43.
64. J. A. D. De Leon and C. R. Borges, J. Visualized Exp., 2020,(159), e61122.
65. P. Saeedi, I. Petersohn, P. Salpea, B. Malanda, S. Karuranga, N. Unwin, S. Colagiuri, L. Guariguata, A. A. Motala and K. Ogurtsova, Diabetes Res. Clin. Pract., 2019, 157, 107843.
66. T. Meldgaard, S. S. Olesen, A. D. Farmer, K. Krogh, A. A. Wendel, B. Brock, A. M. Drewes and C. Brock, J. Diabetes Res., 2018, 2018, 3827301.
67. J. J. Yoo, D. W. Kim, M. Y. Kim, Y. T. Kim and J. H. Yoon, J. Periodontol., 2019, 90, 576–583.
68. S.-Y. Yu, S.-K. Lee, B. Yang, H. Lee, H. J. Jeon and D.-H. Lee, J. Korean Med. Sci., 2024, 39.
69. B. Rohani, World J. Diabetes, 2019, 10, 485.
70. R. Ahmad and M. Haque, Diabetes, Metab. Syndr. Obes., 2021, 3001–3015.
71. M. A. Nazir, L. AlGhamdi, M. AlKadi, N. AlBeajan, L. AlRashoudi and M. AlHussan, Open access Maced. J. Med. Sci., 2018, 6, 1545.
72. M. Cui, J. Zhang, P. Han, L. Shi, X. Li, Z. Zhang, H. Bao, Y. Ma, Z. Tao and X. Dong, Mater. Today Bio, 2024, 28, 101186.
73. M. Namvari, C. S. Biswas, M. Galluzzi, Q. Wang, B. Du and F. J. Stadler, Sci. Rep., 2017, 7.
74. J. Zhang, Y. Fu and A. Mo, Int. J. Nanomed., 2019, 10091–10103.
75. J. Li, Z. Li, X. Liu, C. Li, Y. Zheng, K. W. K. Yeung, Z. Cui, Y. Liang, S. Zhu and W. Hu, Nat. Commun., 2021, 12, 1224.
76. F. Seidi, A. Arabi Shamsabadi, M. Dadashi Firouzjaei, M. Elliott, M. R. Saeb, Y. Huang, C. Li, H. Xiao and B. Anasori, Small, 2023, 19, 2206716.
77. E. A. Hussein, M. M. Zagho, B. R. Rizeq, N. N. Younes, G. Pintus, K. A. Mahmoud, G. K. Nasrallah and A. A. Elzatahry, Int. J. Nanomed., 2019, 4529–4539.
78. A. Rozmysłowska-Wojciechowska, J. Mitrzak, A. Szuplewska, M. Chudy, J. Woźniak, M. Petrus, T. Wojciechowski, A. S. Vasilchenko and A. M. Jastrzębska, Materials, 2020, 13, 2347.
79. D. Wang, L. Wang, Z. Lou, Y. Zheng, K. Wang, L. Zhao, W. Han, K. Jiang and G. Shen, Nano Energy, 2020, 78, 105252.
80. R. Song, L. Ren, H. Ma, R. Hu, H. Gao, L. Wang, X. Chen, Z. Zhao and J. Liu, Am. J. Transl. Res., 2016, 8, 4682.
81. G. F. Pierce, Am. J. Pathol., 2001, 159, 399.
82. B. K. Tiwari, K. B. Pandey, A. Abidi and S. I. Rizvi, J. Biomarkers, 2013, 2013, 378790.
83. M. Zhao, J. Zhou, L. Yuan, X.-Q. Zhang, Y. Hu and H. Yu, J. Nutr. Sci. Vitaminol., 2018, 64, 26–33.
84. X. Saelens, N. Festjens, L. V. Walle, M. V. Gurp, G. V. Loo and P. Vandenabeele, Oncogene, 2004, 23, 2861–2874.
85. F. Dong, C. X. Fang, X. Yang, X. Zhang, F. L. Lopez and J. Ren, Diabetologia, 2006, 49, 1421–1433.
86. A. D. Schimmer, Cancer Res., 2004, 64, 7183–7190.
87. D. A. Breustedt, D. L. Schönfeld and A. Skerra, Biochim. Biophys. Acta, Proteins Proteomics, 2006, 1764, 161–173.
88. A. Johnson, M. Francis and L. A. DiPietro, Adv. Wound Care, 2014, 3, 751–761.

---