Low salivary thioredoxin-1 levels in periodontitis

Abstract

BACKGROUND

The protein thioredoxin-1 (TRX-1) has anti-oxidative, anti-inflammatory and anti-apoptotic effects, and it has been related to the regulation of ageing and different diseases. TRX-1 in periodontitis has been scarcely studied, and the results in humans and in mice are contradictory. In three small studies with humans (the highest with 96 subjects), it was found that patients with periodontitis, compared to periodontally healthy subjects, showed a higher expression of TRX-1 or of the TRX-1 gene in saliva. However, in one study with mice, it was found that mice with periodontitis, compared to periodontally healthy mice, showed a lower expression of TRX-1 in the periodontal tissues.

AIM

To explore the possible association of salivary TRX-1 concentrations with periodontitis and its severity, and of determining the capability of salivary TRX-1 concentrations to predict the diagnosis of periodontitis in a larger sample size study.

METHODS

Salivary TRX-1 concentrations were measured in subjects with and without periodontitis in this observational and prospective study. Criteria to establish periodontal health were the nonexistence, or existence in less than 10% of sites, of bleeding on probing, and the nonexistence of interproximal attachment and bone loss. Criteria to establish localized gingivitis were the existence of bleeding between 10%-30% of sites, and the nonexistence of interproximal attachment and bone loss. Criteria to establish periodontitis were the existence of interproximal attachment or bone loss. We carried out a multivariate logistic regression analysis to determine the variables associated with periodontitis, a correlation analysis to determine the possible association between salivary TRX-1 concentrations and periodontitis severity, and a receiver operating characteristic (ROC) analysis to determine the capability of salivary TRX-1 concentrations to predict the diagnosis of periodontitis.

RESULTS

A total of 144 subjects (58 with periodontitis and 86 without periodontitis) were included. Low salivary TRX-1 concentrations showed an association with periodontitis (P = 0.04) according to regression analysis, an association with periodontitis severity (rho = -0.47; P < 0.001) according to correlation analysis, and an area under curve of 75% (95%CI: 67%-82%; P < 0.001) for periodontitis diagnosis according to ROC analysis.

CONCLUSION

Novel findings of this study were the association of low salivary TRX-1 concentrations with periodontitis and its severity, and the capacity of salivary TRX-1 concentrations to help in the periodontitis diagnosis.

Key Words: Oxidation; Periodontitis; Salivary; Thioredoxin-1; Inflammation

Core Tip: The protein thioredoxin-1 (TRX-1) has anti-oxidative effects and is involved in senescence, aging and different diseases. Salivary levels of TRX-1 in patients with periodontitis have not been explored. Novel findings of this study were the association of low salivary TRX-1 concentrations with periodontitis and its severity, and the capacity of salivary TRX-1 concentrations to help in the diagnosis of periodontitis.

INTRODUCTION

Periodontitis consists of a chronic inflammatory disease of periodontal tissue that could motivate tooth loss. It has a high prevalence, affecting up to 90% of the subjects[1], and severely up to 15% of humans[2]. Different mechanisms are involved in periodontitis, such as inflammation[3], oxidation[4], and programmed cell death[5].

The thioredoxin system has an important role in oxidation, inflammation, apoptosis, senescence and aging; in addition, the thioredoxin system has been recently studied in different diseases such as inflammatory illnesses, neurodegenerative diseases, ischemic stroke, cancer, diabetes and chronic kidney disease, cardiovascular diseases[6-11]. Besides, the thioredoxin system could also play a role in periodontitis[12,13].

The thioredoxin system involves the thioredoxin protein (TRX), nicotinamide adenine dinucleotide phosphate (NADPH), the thioredoxin reductase (TrxR), and thioredoxin-interacting protein (TXNIP).

In mammalian cells, three TRX isoforms have been identified: Cytosolic TRX (TRX-1), mitochondrial TRX (TRX-2), and spermatozoa Trx (SpTRX). TRX-1 is the most studied TRX. TRX-1 has two forms (oxidized inactive form and reduced active form). The reduced active form of TRX-1 has anti-oxidative, anti-inflammatory and anti-apoptotic effects.

NADPH oxidase is involved in the production of reactive oxidative species (ROS) because it catalyzes the oxidation of NADPH to NADP⁺ by transferring one electron from NADPH to oxygen, and subsequently NADP⁺ and ROS appear (NADPH + O2 = NADP⁺ + O2^-). In addition, NADPH could also donate electrons to TrxR; afterwards TrxR transfers these electrons to oxidized inactive TRX-1, resulting in reduced active TRX-1 and NADP⁺ (NADPH + TRX-1 oxidized = NADP⁺ + TRX-1 reduced). Thus, TrxR facilitates this reaction and the appearance of the active and reduced TRX-1 form by the action of TrxR and NADPH axis.

TXNIP blocks active TRX-1 because it binds with the active reduced TRX-1 form and blocks the reactivation of TRX-1 by blocking the TrxR and NADPH axis. Thus, TXNIP contributes to oxidation, apoptosis, inflammation and senescence.

The antioxidant effect of TRX-1 consists of reversing the oxidation of oxidized proteins, neutralizing ROS via peroxiredoxin (Prx), and activating different endogenous antioxidant enzymes.

One antioxidant effect of TRX-1 consists of the reduced active form of TRX-1 transferring electrons to oxidized proteins. Therefore, those proteins are reduced, and TRX-1 appears in its oxidized inactive form. Afterwards, oxidized TRX-1 is reduced by TrxR using electrons from NADPH, restoring the reduced active TRX-1 form.

Another antioxidant effect of TRX-1 consists of transferring electrons to peroxiredoxins (Prx). Therefore, Prx is reduced (appearing as the reduced active Prx form) and TRX-1 appears in its oxidized inactive form. Then, reduced Prx transfers those electrons to neutralize ROS such as hydrogen peroxide (H₂O₂). Reduced Prx transfers electrons to H2O2, resulting in the formation of water (H₂O) through different reactions (H₂O₂ + 2H⁺ + 2e^- = 2 H₂O or 2 H₂O₂ = 2 H₂O + O₂). Again, oxidized inactive TRX-1 is transformed into the reduced active TRX-1 form by the action of TrxR and NADPH.

Another antioxidant effect of TRX-1 consists of producing redox signals to activate different endogenous antioxidant enzymes such as superoxide dismutase (SOD), glutathione peroxidase (GPX) and catalase (CAT)[14-16].

The anti-apoptotic effects of TRX-1 consist of inhibiting caspase-3, caspase-9, pro-apoptotic Bcl-2 proteins, and pro-apoptotic p53 protein.

The anti-inflammatory effects of TRX-1 consists of inhibiting nuclear factor kappa B and therefore inhibiting interleukin-6 (IL-6), and tumor necrosis factor α (TNF-α).

All these effects of reduced active TRX-1 form, including anti-oxidative, anti-inflammatory and anti-apoptotic actions make TRX1 play an important role in senescence, aging and different diseases[6-11]. Downregulation of active TRX-1 increases oxidative stress, inflammation and apoptosis. Thus, the administration of TRX-1 agonists, TrxR agonists and TXNIP inhibitors may be beneficial in age-associated diseases. However, the administration of TRX-1 antagonists, TrxR antagonists and TXNIP agonists might help in anticancer treatments to hinder tumor progression[6-11].

TRX-1 in periodontitis has been scarcely studied, and the results in humans[17-19] and in mice are contradictory. In three small studies with humans (the highest with 96 subjects in the study by Toczewska et al[19]), it was found that patients with periodontitis, compared to periodontally healthy subjects, showed a higher expression of TRX-1 or the TRX-1 gene in saliva[17-19]. However, in one study with mice, it was found that mice with periodontitis compared to periodontally healthy mice showed a lower expression of TRX-1 in periodontal tissues[16]. In addition, in the study by Wu et al[16], it was found that the administration of recombinant human TRX-1 (rhTRX-1) restored reduced TRX-1 expression, decreased ROS generation and promoted the alveolar bone repair in mice with periodontitis.

Thus, take into account that oxidation, inflammation, and tissue degradation are involved in periodontitis, that TRX-1 could modulate all these pathways[12,13], and the findings in the study by Wu et al[16] in mice with periodontitis (mice with periodontitis showed a lower expression of TRX-1, and that the administration TRX-1 decreased ROS generation and promoted alveolar bone repair), we hypothesize that TRX-1 could play a protective role in human periodontitis.

We conducted this study with the aim of comparing salivary TRX-1 concentrations in subjects with and without periodontitis in a larger sample size study, exploring the possible association of salivary TRX-1 concentrations with periodontitis and its severity, and determining the capability of salivary TRX-1 concentrations to predict the diagnosis of periodontitis.

MATERIALS AND METHODS

Design and subjects

This prospective and observational study was carried out at the Clínica Dental Cándido (La Laguna, Tenerife, Canary Islands, Spain). The study was previously approved by the clinical research ethics committee of the Hospital Universitario de Canarias (CHUC_2023_138; November 30, 2023). Each subject signed an informed consent form prior to participation in the study.

We included subjects without signs of periodontitis (with periodontal health or only localized gingivitis in < 30% of sites) and subjects with periodontitis (presence of periodontal tissue loss). We used current internationally accepted criteria for the diagnosis of periodontitis and for the severity classification of periodontitis[20]. We excluded subjects younger than 18 years.

Definitions

Criteria to establish periodontal health were the nonexistence, or existence in less than 10% of sites, of bleeding on probing, and the nonexistence of interproximal attachment and bone loss.

Criteria to establish localized gingivitis were the existence of bleeding between 10%-30% of sites, and the nonexistence of interproximal attachment and bone loss.

Criteria to establish periodontitis were the existence of interproximal attachment or bone loss. Criteria to establish periodontitis severity were the following: (1) Interproximal clinical attachment loss < 3 mm (stage I), between 3-4 mm (stage II) or loss ≥ 5 mm (stages III or IV); (2) Radiographic bone loss in the coronal third < 15% (stage I), in the coronal third between 15%-33% (stage II) or in the middle or apical third (stages III or IV); and (3) Tooth loss of none (stage I), between 1-4 (stage II) or ≥ 5 teeth (stage III).

Variables recorded

Consumption of coffee, tea, tobacco, drugs and alcohol was recorded. Also, sex, age, body mass index (BMI) (kg/m²) and the existence of obesity (defined as BMI ≥ 30 kg/m²). In addition, the personal history of diabetes mellitus, hypercholesterolemia, arterial hypertension, cardiovascular disease, rheumatoid arthritis, systemic lupus erythematosus, treatment with methotrexate (at a dose for rheumatoid arthritis), oral cancer, immunosuppressive therapy, radiotherapy was also registered, as well as dental hygiene.

Salivary samples

The technique of Navazesh was used to obtain samples of whole and non-stimulated saliva[21]. We took saliva samples in the morning (8 am to 10 am) to reduce potential alterations due to circadian rhythm on salivary biomarker concentrations. The subjects did not brush their teeth, smoke or drink during the 2 hours before sample collection. They rinsed the mouth with 10 mL of deionized water three times. Afterwards, they remained seated for 30 minutes, avoiding orofacial movements, with eyes open, with the head tilted forward, avoiding swallowing saliva, and allowing saliva collection in a bottle. Then, the collected saliva was centrifuged for 10 minutes at 3000 rpm at 24 °C to remove possible debris and cells. Finally, the supernatant was pipetted and introduced into small Eppendorf tubes that remained at -80 ºC stored for biomarker determinations

Some individuals included in this study were already included in other publications from our group. Salivary concentrations of 3-nitrotyrosine[22], uric acid[23], malondialdehyde[24], and nitrites[25] were analyzed in those publications; but the objective of our current research was the determination of salivary TRX-1 levels.

Salivary TRX-1 concentrations analysis

THR-1 concentration in saliva was calculated using a solid-phase competitive enzyme-linked immunosorbent assay (Invitrogen, Thermo Fisher Scientific Inc., Whaltam, MA, United States), according to the manufacturer’s protocol. Briefly, in each appropriate well: 100 μL standards or 100 μL saliva (diluted 4 times) were incubated overnight at 4-6 ºC (the plate was covered with a plate sealer for incubation). After washing each well with wash buffer, 100 μL biotinylated detection Ab working solution were incubated for 1 hour at room temperature. Again, after washing each well with wash buffer, 100 μL of streptavidin linked to Horseradish Peroxidase conjugate working solution was added into each well and incubated for 45 minutes at room temperature. After washing each well with wash buffer, 100 μL of Substrate Reagent (3,3′,5,5′-Tetramethylbenzidine as chromogenic substrate) were added to each well and incubated for 35 minutes at RT avoiding, at all times, direct light on the plate (reagent light-sensitive). Finally, 50 μL of stop solution (1 M H₂SO₄) was addeded to each well (yellow colour was developed) at room temperature. The calibrated curve was ranged between 0-25 ng/mL. Samples and standards absorbance units were read in a microplate spectrophotometer at 450 nm (Spectra MAX-190, Molecular Devices, Sunnyvale, CA, United States). A four-parameter algorithm provides the best standard curve fit (the correlation coefficients of different kits ranged between 0.9971 and 0.9977). The limit of detection of the different assays was established in a range between 0.010 ng/mL and 0.017 ng/mL. Intra-assay and inter-assay coefficients of variation were calculated at < 10% and < 12%, respectively.

Statistical analysis

We reported categorical variables as number and percentage, and they were compared between groups by the χ² test. We reported continuous variables as median and percentiles 25-75, and they were compared between groups by the Mann-Whitney U-test. We analysed the correlation between salivary TRX-1 concentrations and periodontitis severity by Spearman’s rho coefficient.

A multivariate logistic regression analysis was carried out to determine the variables that were associated with periodontitis. We included in the regression analysis those variables showing P value ≤ 0.05 when they were compared between subject groups and showing a reasonable number of events.

A receiver operating characteristic analysis was carried out to determine the ability of salivary TRX-1 concentrations to predict the diagnosis of periodontitis, and the area under curve (AUC) was reported. Negative and positive predictive values, sensitivity, negative and positive likelihood ratios and specificity for the cut-off (< 41.55 ng/mL) of salivary TRX-1 concentrations were reported. The value of cut-off was selected on the basis of the Youden J index[26].

MedCalc statistical software version 22.016 (MedCalc Software Ltd, Ostend, Belgium) and SPSS version 17.0 (SPSS Inc., Chicago, IL, United States) were the programs used to carried out the statistical analyses.

RESULTS

A total of 144 subjects (58 with periodontitis and 86 without periodontitis) were included. We report the number of subjects in each periodontal stage and its salivary TRX-1 concentrations in Table 1, and we found significant differences in salivary TRX-1 concentrations according to periodontal stage (P < 0.001). Besides, we found a negative correlation between periodontitis severity and salivary TRX-1 concentrations (rho = -0.47; P < 0.001).

Table 1 Number of subjects and salivary thioredoxin-1 concentration in each periodontal stage, n (%).

	Total (n = 144)	Salivary TRX-1 (ng/mL) median (P25%-75%)	P value
			< 0.001
Periodontal health	86 (59.7)	55.6 (30.4-67.6)
Periodontitis stage I	16 (11.1)	37.1 (27.8-54.8)
Periodontitis stage II	23 (16.0)	29.7 (26.2-33.6)
Periodontitis stage III	13 (9.0)	26.0 (22.6-31.2)
Periodontitis stage IV	6 (4.2)	23.2 (18.4-26.5)

TRX-1: Thioredoxin-1.

Subjects with periodontitis, compared to subjects without periodontitis, had lower salivary TRX-1 concentrations (P < 0.001), higher rate of arterial hypertension (P < 0.001), a higher rate of cardiovascular disease (P = 0.03), a higher rate of diabetes mellitus (P = 0.03), lower rate of never smoker history (P < 0.001), a lower rate of tea consumption (P = 0.005), and were older (P < 0.001). No significant differences were found between both groups of subjects (without and with periodontitis) in immunosuppressive therapy, rheumatoid arthritis, radiotherapy, methotrexate for rheumatoid arthritis, body mass index, hypercholesterolemia, obesity, coffee, alcohol and sex (Table 2). No subject (without or with periodontitis) had oral cancer, drug consumption or systemic lupus erythematosus.

Table 2 Comparisons between subjects without and with periodontitis, n (%).

	Subjects without periodontitis (n = 86)	Subjects with periodontitis (n = 58)	P value
Gender female	63 (73.3)	36 (62.1)	0.2
Age (years) - median (P25%-75%)	36 (20-44)	59 (51-68)	< 0.001
Arterial hypertension	4 (4.7)	17 (29.3)	< 0.001
Cardiovascular disease	0	4 (6.9)	0.03
Hyperocholesterolemia	2 (2.3)	2 (3.4)	0.99
Diabetes mellitus	0	4 (6.9)	0.03
Rheumatoid arthritis	2 (2.3)	3 (5.2)	0.39
Metrotexate for rheumatoid arthritis	2 (2.3)	0	0.52
Immunosupressive therapy	2 (2.3)	1 (1.7)	0.99
Radiotherapy	0	1 (1.7)	0.4
Body mass index (kg/m²) - median (P25%-75%)	24.2 (22.4-26.6)	24.8 (22.5-27.9)	0.38
Obesity	11 (12.8)	8 (13.8)	0.99
Never smoker	66 (76.7)	25 (43.1)	< 0.001
Coffee	68 (79.1)	51 (87.9)	0.19
Tea	20 (23.3)	3 (5.2)	0.005
Alcohol	33 (38.4)	31 (53.4)	0.09
Salivary TRX-1 levels (ng/mL) - median (P25%-75%)	55.6 (30.4-67.6)	29.1 (25.0-34.5)	< 0.001

TRX-1: Thioredoxin-1.

We found in multiple logistic regression analysis that age (years) (OR: 1.11; 95%CI: 1.064-1.165; P < 0.001) and salivary TRX-1 concentrations (ng/mL) (OR: 0.97; 95%CI: 0.946-0.998; P = 0.04) were the variables associated independently with periodontitis (Table 3).

Table 3 Multiple logistic regression analysis to determine that factors are associated with periodontitis.

	Odds ratio	95%CI	P value
Age (years)	1.11	1.064-1.165	< 0.001
Salivary thioredoxin-1 levels (ng/mL)	0.97	0.946-0.998	0.04
Never smoker (yes vs non)	0.54	0.195-1.469	0.23
Arterial hypertension (yes vs non)	1.17	0.263-5.223	0.84
Tea consumption (yes vs non)	0.30	0.044-2.027	0.22

CI: Confidence interval.

We found an AUC of salivary TRX-1 for the diagnosis of periodontitis of 75% (95%CI: 67%-82%; P < 0.001) (Figure 1). The selected cut-off (< 41.55 ng/mL) of salivary TRX-1 concentrations to the diagnosis of periodontitis had a negative predictive value of 84% (75%-91%), a positive predictive value of 60% (53%-67%), a sensitivity of 83% (71%-91%), a negative likelihood ratio of 0.3 (0.2-0.5), a positive likelihood ratio of 2.2 (1.7-3.0), and a specificity of 63% (52%-73%).

Figure 1 Receiver operating characteristic analysis to determine the capability of salivary thioredoxin-1 concentrations to periodontitis diagnosis. AUC: Area under the curve; TRX-1: Thioredoxin-1.

DISCUSSION

Novel findings of this study were the association of low salivary TRX-1 concentrations with periodontitis and its severity, and the capacity of salivary TRX-1 concentrations to help in the periodontitis diagnosis.

TRX-1 in patients with periodontitis have been studied scarcely[17-19]. In a study by Foratori-Junior et al[17] with 18 periodontitis patients and 20 periodontally healthy subjects, a higher expression of TRX-1 in saliva was found in periodontitis patients than in healthy subjects. In a study by Hartenbach et al[18] with 30 periodontitis patients and 10 periodontally healthy subjects, a higher expression of TRX-1 in saliva was found in periodontitis patients than in healthy subjects. In a study by Toczewska et al[19] with 65 periodontitis patients and 31 periodontally healthy subjects a higher expression of TRX-1 gene in saliva was found in periodontitis patients than in healthy subjects, and a lower expression of TRX-1 gene in gingival tissue was found in periodontitis patients than in healthy subjects; however, the levels of TRX-1 in saliva and in gingival crevicular fluid were not determined.

Therefore, our study is the largest to date evaluating TRX-1 concentrations in patients with periodontitis (144 subjects). Our findings of lower salivary TRX-1 concentrations in subjects with, compared to without, periodontitis are contrary to those previous studies reporting that patients with periodontitis compared to periodontally healthy subjects showed a higher expression of TRX-1 or of TRX-1 gene in saliva[17-19]. However, our findings could be in consonance with those from one study with mice showing that mice with periodontitis compared to periodontally healthy mice showed a lower expression of TRX-1 in the periodontal tissues[16]. Curiously, in the study by Toczewska et al[19], it was found that the expression of TRX-1 gene in periodontitis patients, compared to healthy subjects, was lower in gingival tissue and higher in saliva. In addition, novel findings of our study were the association between low salivary TRX-1 concentrations and periodontitis independently of other factors according to the results of regression analysis.

Our observed findings about lower salivary TRX-1 concentrations in subjects with periodontitis and the association of low salivary TRX-1 concentrations with periodontitis could be explained by the fact that TRX-1 has effects on reducing oxidation and inflammation[12,13], and also on reducing tissue degradation in periodontitis[16]. Therefore the low salivary TRX-1 concentrations observed in patients with periodontitis in our series could be related to higher oxidation (due to higher ROS concentrations because TRX-1 does not neutralize ROS and does not activate other antioxidant enzymes such as SOD, GPX and CAT), higher inflammation (with higher production of pro-inflammatory citokines such as IL-6 and TNF-α), and greater tissue degradation (due to collagen degradation) in periodontitis.

We found that salivary TRX-1 concentrations have the capacity to help in the diagnosis of periodontitis. The area under curve that we found was not high (75%)[27]; therefore, salivary TRX-1 concentrations could help in the periodontitis diagnosis, but they could not be used alone as the sole diagnostic parameter. As the cut-off of salivary TRX-1 concentrations lower 41.55 ng/mL showed a sensitivity of 83%, there is a probability of 83% that a subject with TRX-1 lower 41.55 ng/mL has periodontitis. As the cut-off of salivary TRX-1 concentrations lower 41.55 ng/mL showed a specificity of 63%, there is a probability of 63% that a subject with TRX-1 higher 41.55 ng/mL does not have periodontitis. Further investigations are necessary to confirm those preliminary findings.

In some previous studies, a positive association between the expression of TRX-1 or of TRX-1 gene in saliva and periodontitis severity was found[18,19]. Thus, our findings of a negative association between salivary TRX-1 concentrations and periodontitis severity are contrary to those previous studies. Thus, the lower the salivary TRX-1 concentrations, the greater the periodontitis severity. The correlation that we found was not strong (rho = -0.47; P < 0.001)[28]; therefore, salivary TRX-1 concentrations could help predict the severity of periodontitis, but they could not be used alone as the sole predictive parameter. Further investigations are necessary to confirm those preliminary findings.

We found that higher age was another risk factor for periodontitis, as was previously described[29]. Other factors with a higher risk of periodontitis were previously documented, such as consumption of drugs, oral cancer, dental hygiene, alcohol, immunosuppression, coffee, tea and tobacco[30], obesity[31], systemic lupus erythematosus[32], sex[33], rheumatoid arthritis[34], diabetes mellitus[35] and arterial hypertension[36]; however, these were not found in our series possibly due to sample size.

A point of interest is that thioredoxin system activity could be modulated by different antagonist and agonist agents of each member of the thioredoxin system[6-11]. On the one hand, the administration of TRX-1 agonists, TrxR agonists and TXNIP antagonists may be beneficials in age-associated diseases and in preventing cancer (reducing mutation accumulation) due to its anti-oxidative, anti-inflammatory and anti-apoptotic effects. On the other hand, the administration of TRX-1 antagonists, TrxR antagonists and TXNIP agonists might help in anticancer treatments to hinder tumor progression[6-11]. TrxR agonists and TXNIP antagonists facilitate the reduction of oxidized inactive TRX-1 form by TrxR using electrons from NADPH, restoring the reduced active TRX-1 form. On the contrary, TrxR antagonists and TXNIP agonists block this reaction and block the appearance of the active reduced TRX-1 form. Among the TRX-1 antagonists are 1-methyl propyl 2-imidazolyl disulfide, dimethyl fumarate, 4-(benzothiazol-2-yl)-4-hydroxycyclohexa-2,5-dienone (PMX464), suberoylanilide hydroxamic acid, sodium butyrate, isoforretin A, diallyl trisulfide, and eprenetapopt. The TrxR antagonists include ethaselen, auranofin, curcumin, piperlongumine and sodeoxyelephantopin.

With respect to periodontitis, in a study performed by Wu et al[16] including mice with diabetes mellitus and periodontitis, an increased ROS production and reduced TRX-1 expression were found in the periodontal tissues of mice with periodontitis compared to mice with periodontal health. In addition, it was found that the administration of recombinant human TRX-1 restored this reduced TRX-1 expression, decreased ROS generation and promoted the alveolar bone repair in periodontitis.

We recognize that our study has some limitations. One issue was that TRX-1 concentrations in other samples (gingival crevicular fluid, gingival tissue, or blood) was not determined, although we designed this study with the aim of being as non-invasive as possible. In addition, it is possible that more factors associated with periodontitis could have been found by including more subjects in the study. Besides, the observational nature of the study could imply the presence of some potential bias and that some unmeasured variables could influence the observed findings. In addition, sample size was not calculated; however, it was sufficient to find an association between low salivary TRX-1 levels and periodontitis. Therefore, further investigations are necessary to confirm these preliminary findings.

On the contrary, we think that our study also showed some strengths. This study has a larger sample and used a minimally invasive biologic sample (saliva) compared to previously published studies on TRX-1 in periodontitis, and it is the first one reporting the association of periodontitis and its severity with low salivary TRX-1 levels.

Finally, with respect to future research directions, it could be interesting to determine the role of salivary TRX-1 concentrations in the development of periodontitis, periodontitis severity and periodontitis diagnosis in periodontitis in larger studies controlling for more variables, and to determine the effects of the administration of agents that modulate thioredoxin system activity in randomized clinical trials on the clinical evolution of periodontitis and on the concentrations of oxidative and inflammatory buccal biomarkers. The diagnosis and follow-up of periodontitis have traditionally been based on radiographic and probing assessment. However, some factors related to the patient (such as limitation in mouth opening) and to the examiner (such as limited experience of the examiner or differences between examiners in the pressure applied during periodontal probing) may influence these assessments[37-42]. Therefore, the determination of biomarker levels in saliva, such as TRX-1, could help in the diagnosis and follow-up of periodontitis.

CONCLUSION

Novel findings of this study were the association of low salivary TRX-1 concentrations with periodontitis and its severity, and the capacity of salivary TRX-1 concentrations to help in the periodontitis diagnosis. Further investigations are necessary to confirm those preliminary findings.

ACKNOWLEDGEMENTS

Centro Integrado de Formacion Profesional (CIFP) Los Gladiolos (Santa Cruz de Tenerife, Spain) has facilitated the inclusion of participants in the study.