Rapid diagnosis of periodontitis, a feasibility study using MALDI-TOF mass spectrometry.

AIM: The aim of the present study was to assess the feasibility and diagnostic contribution of protein profiling using MALDI-TOF mass spectrometry applied to saliva, gingival crevicular fluid (GCF) and dental plaque from periodontitis and healthy subjects. We hypothesized that rapid routine and blinded MALDI-TOF analysis could accurately classify these three types of samples according to periodontal state.
MATERIALS AND METHODS: Unstimulated saliva, GCF and dental plaque, collected from periodontitis subjects and healthy controls, were analyzed by MALDI-TOF MS. Based on the differentially expressed peaks between the two groups, diagnostic decision trees were built for each sample.
RESULTS: Among 141 patients (67 periodontitis and 74 healthy controls), the decision trees diagnosed periodontitis with a sensitivity = 70.3% (± 0.211) and a specificity = 77.8% (± 0.165) for saliva, a sensitivity = 79.6% (± 0.188) and a specificity = 75.7% (± 0.195) for GCF, and a sensitivity = 72.1% (± 0.202) and a specificity = 72.2% (± 0.195) for dental plaque. The sensitivity and specificity of the tests were improved to 100% (CI 95% = [0.91;1]) and 100% (CI 95% = [0.92;1]), respectively, when two samples were tested.
CONCLUSION: We developed, for the first time, diagnostic tests based on protein profiles of saliva, GCF and dental plaque between periodontitis patients and healthy subjects. When at least 2 of these samples were tested, the best results were obtained.

Introduction

Periodontitis is a chronic multifactorial inflammatory disease associated with dysbiotic plaque biofilms that results in the progressive destruction of the supporting structures of the teeth [1]. The 2009–2012 NHANES estimated that 46% of adults aged 30 years or older in the United States had periodontitis with 8.9% having two or more interproximal sites with ≥ 6 mm clinical attachment loss (not on the same tooth) and one or more interproximal site(s) with ≥ 5 mm periodontal probing depth [2]. Due to its high prevalence and its consequences on quality of life, oral health and associations with systemic diseases, periodontitis is an important public health problem [3,4]. The diagnosis of periodontitis is currently based on clinical measurements of probing pocket depth (PD), bleeding on probing (BOP), plaque index, and clinical attachment level and is associated with a radiographic examination [1]. The major drawbacks of this clinical diagnosis are that it is time-consuming and that it requires professional skills and training, making it difficult to use for large-scale screening. Furthermore, because at the beginning of periodontal disease the patient does not experience any symptoms, he will not consult and will not be diagnosed until an advanced stage of periodontitis is reached. These assessments highlight the necessity to have a simple minimally invasive screening and diagnosis tool for periodontal health and periodontal disease at any stage.

Recently, the development of diagnostic tests based in either oral fluid or blood analysis has seen considerable growth for population screening in many medical disciplines [5-7]. Furthermore, the use of oral fluid as an alternative to venous blood has been intensively explored in various fields of research such as therapeutic drug monitoring or diseases biomarker discovery purpose [8,9]. Compared to blood and its derivatives, saliva carries many advantages including non-invasiveness, no need for highly trained personnel, safer to handle, ease to collect, ship and store [10]. Recently, Campanella et al. have found that analysis of volatile salivary metabolites could be promising for the indirect assessment of gut microbiota [11].

In the periodontal field, numerous studies have attempted to investigate potential changes in saliva, gingival crevicular fluid (GCF) and dental plaque that could be correlated to periodontitis and be used to develop new alternative diagnostic method [12-14]. However, to date, no unique biomarker has been identified as sensitive and specific enough to be used for diagnosing periodontitis [15,16]. Biological tests are still dramatically missing in the periodontal diagnosis field while they already exist or will exist tomorrow for other diseases [17,18]. Faced with this observation, the evaluation and comparison of global protein profiles seem more relevant than the search for and identification of biomarkers in the periodontitis diagnosis. The existence of specific protein profiles could represent the signature of a periodontal disease.

Matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) is a rapid and precise proteome profiling method that generates a characteristic spectrum for analytes in the sample. Profiling is a conventional approach that is widely used in bacterial identification by comparing spectra of unknown bacteria to databases that contain spectra of reference bacteria [19-20]. In this non-quantitative technique, intact cells can directly be mixed with MALDI matrix [20]. La Scola & Raoult have shown that this method was efficient for direct routine identification of bacterial isolates in blood culture and may surpass the conventional diagnostic methods in speed and accuracy [21]. Beyond the realms of microbial world, recent studies demonstrated that MALDI-TOF MS technique can be used to identify fleas [22], ticks [23] and mosquitoes [24] and also appears promising in the rapid diagnosis of cancerous lung nodule [25].

In the periodontology field, Ngo et al. were the first to use mass spectrometric techniques to identify 33 peptides and 66 proteins in GCF from inflammatory sites in periodontal maintenance subjects [26]. In another study, the same authors found that peptide profiles of GCF from periodontal maintenance subjects could be used to predict sites with attachment loss [27]. Zhang et al. compared the protein profiles of orthodontics patients with and without periodontitis using MALDI-TOF MS and found significantly different intensities of nine peaks, seven of which were higher in healthy subjects [28]. Chaiyarit et al. analyzed salivary protein profiles in oral cancer, oral lichen planus, and chronic periodontitis and found that mass signals at 5,835.73 and 9,801.83 Da were significantly decreased in periodontitis patients compared to the two others oral diseases [29]. More recently, in a small number of patients, changes in protein profiles in chronic periodontitis were reported in saliva, GCF and serum [30].

The few studies identifying protein profiles changes were case-control studies, which described differences between periodontitis and healthy subjects, but there are no data available from blinded experiments. The aim of the present study was to assess the feasibility and diagnostic contribution of protein profiling using MALDI-TOF mass spectrometry applied to saliva, GCF and dental plaque from periodontitis and healthy subjects. We hypothesized that rapid routine and blinded MALDI-TOF analysis could accurately classify these three types of samples according to periodontal state.

Materials and methods

This study was approved by the Institutional Review Board (IRB) and Independent Ethics Committee (IEC) of Institut Hospitalo-Universitaire (IHU), Microbes Evolution Phylogeny and Infections—(MEPHI) under N° 2019–008. Informed written consent was obtained from each patient.

Study population

The subject population was recruited from January 2019 until June 2019. Periodontitis subjects were recruited among patients referred to the periodontal department of the Pavillon Odontologique de la Timone, Marseille. Control subjects were recruited among young adults. A medical questionnaire (epidemiological data) including the main general risk factors for periodontal diseases was completed (S1 Table) and a periodontal examination was performed, including an intra-oral examination and full-mouth periodontal probing. A periodontal diagnosis was made, and subjects were divided into two groups: the periodontitis group, in which the periodontitis has been classified into stages and grades according the Chicago classification [1] and the healthy periodontium (control) group.

According to the Chicago classification [1], the inclusion criteria for:

the periodontitis group were interdental clinical attachment loss (CAL) detectable at ≥ 2 non-adjacent teeth or buccal or oral CAL ≥ 3 mm with pocketing ≥ 3 mm detectable at ≥ 2 teeth,

the healthy periodontium control group were bleeding score on probing < 10%, pocket depth < 3 mm and no clinical attachment loss.

All the patients were selected based upon periodontal status, regardless of the other criteria (e.g., systemic disease or disease/infection that may affect the periodontal health status, use of antibiotics or immunosuppressant medication within 3 months, current or former smokers), in order to reduce selection bias.

Non-inclusion criteria for both groups were history of periodontal therapy within the previous 6 months, pregnant/lactating women and orthodontic patients.

Sample collections

Saliva

All subjects were asked to not eat, drink or brush their teeth one hour before sample collection. Each subject was asked to expectorate whole saliva into a 50-mL centrifuge tube until a minimum saliva volume of 2 mL was collected.

GCF and subgingival dental plaque

GCF and subgingival dental plaque samples were collected from one site showing PD <3mm without CAL or BOP of one tooth in each quadrant (4 sites from each subject in total) in control group. In periodontitis group, samples were collected from one site with PD ≥ 5mm and CAL ≥ 3mm of one tooth in each quadrant of each subject (4 sites from each subject in total). For both GCF and subgingival dental plaque collections, all supra-gingival plaque facing the sampling area was removed with a sterile curette to avoid contaminating the samples. The site was then isolated by cotton rolls and gently air dried for 5 s to remove any saliva present. GCF was collected using sterile absorbent paper points (Paper points N°20, VDW-Zipperer®) carefully positioned into the periodontal pocket from periodontitis patients and into the gingival sulcus from control subjects and left for 30 s. GCF within the periodontal pocket or the sulcus was absorbed by the paper points through capillary action. A total of 4 paper points were obtained from each patient and placed into a 1.5-mL Eppendorf tube containing 100 μL HPLC-grade water (HPLC: high-performance liquid chromatography). Subgingival dental plaque was collected using a sterile curette and placed into a 1.5 mL Eppendorf tube containing 100 μL HPLC-grade water. Samples were immediately stored at 4°C and analyzed within 24 to 48 hours. Only samples not visually contaminated with blood were selected for the study.

Mass spectrometry

A volume of 0.5 μL of each sample was directly spotted six times onto a 96 polished steel MALDI target and then allowed to dry at room temperature. Deposits were then coated with 1 μL of a matrix solution containing α-cyano-4-hydroxycinnamic acid diluted into 500 μL of acetonitrile, 250 μL of 10% trifluoroacetic acid and 250 μL of HPLC-grade water. All manipulations were carried out under class II biological safety cabinets MSC-AdvantageTM (Thermo Fischer Scientific, Villebon sur Yvette, France). After drying for a few minutes at room temperature, the target was introduced into a Microflex LT MALDI-TOF mass spectrometer laser (Bruker Daltonics, Bremen, Germany; external mass spectrometer calibration accuracy ± 300ppm). Each sample generated 6 spectra from 6 deposits. Spectra were recorded in the positive linear mode at a laser frequency of 50 Hz within a mass range of 2–20 kDa. The acceleration voltage was 20 kV, and the extraction delay time was 200 ns. Each spectrum was obtained from 240 laser shots performed in 6 regions of the same spot and then automatically acquired using the AutoXecute acquisition control in FlexControl software 3.0 (Bruker Daltonics). The spectra of the six spots for each sample were imported into the BioTyper-RTCTM version 3.0 software (Bruker Daltonics GmbH). The calibration of the MS was fully automated and performed with a commercial solution (BTS: Bacterial standard test) and the procedure was completely automatic (BiotyperRTC user manual). For each analysis, the MALDI target was simultaneously tested with an inactivated strain of Escherichia coli as the positive control (objective score > 2,1) and with matrix solution alone as the negative control (objective score < 1,5). All spectra were controlled using the Flexanalysis® v3.4 software (Bruker Daltonics, Bremen, Germany). Quality criteria of the spectrum for global aspect and intensity were checked: intensity above 104 arbitrary units (AU), horizontal baseline curve and presence of visually identifiable peaks.

Bioinformatics analysis

Spectra were analyzed using a homemade R program [31] including the supplementary libraries MALDIQuant [32], seriation [33] and binDA [34].

A first step of spectra pre-processing included noisy spectra discarding using spectrum signal to noise ratio, smoothing (moving average with half window size 8), baseline correction (Statistics-sensitive Non-linear Iterative Peak-clipping algorithm, 100 iterations), intensity recalibration (total ion current), peak selection (MAD with half window size 8 and signal-noise ratio threshold 3), spectra alignment (quadratic warping function with 0.002 tolerance), averaging of technical replicates in main spectrum profiles (MSP), peak binning, and intensity matrix building, as recommended by Gibb & Strimmer [35].

Differences in the two groups analyzed were assessed on the basis of a discriminant peak identification list using binary predictors. To create the list of discriminant peaks for each sample, we performed a discriminant analysis between groups using the Binary Discriminant Analysis method [34]. Then, we searched for models able to correctly discriminate the two groups, periodontitis and healthy periodontal subjects, from each sample. We generated binary decision trees using Quinlan’s C5.0 algorithm, an extension of C4.5 [36]. In a second step, an internal validation was processed with a 10-fold cross-validation in order to verify the classification ability of the generated models. Each group of samples was randomly partitioned into 10 equal-sized subsamples. One subsample was selected as the validation data for testing the diagnostic decision tree, while the other 9 subsamples were used as training data. The process was repeated 10 times with a rotation of the subsamples tested. The performance of the models of each sample was evaluated by sensitivity and specificity. Finally, a principal component analysis was carried out between the top 10 ranking peaks and the epidemiological data from the medical questionnaire (age, gender, current smokers, former smokers, diabetes, cardiovascular disease, hypothyroidism, arthritis, respiratory disease, anti-diabetic medication, antibiotics, anti-inflammatory, antihypertensive, anticoagulants, thyroid hormone thyroxine, hormonal contraception and stress).

Results

Characteristics of the subjects in the study

A total of 141 subjects, specifically 39 males and 102 females aged from 20 to 77 years, were enrolled in the study. The periodontitis group included 67 subjects (mean age 50.18 ± 13.85 years, 14 males/53 females). The control group included 74 subjects (mean age 24.50 ± 3.28 years, 25 males/49 females). Patient details are listed in Tables 1 and 2. The same proportions of current smokers were present in our two groups. A large proportion of our periodontitis group was classified into stages III (46.3%) and IV (41.8%). A total of 119 samples of saliva (51 periodontitis; 68 controls), 104 samples of GCF (54 periodontitis; 50 controls) and 110 samples of dental plaque (51 periodontitis; 59 controls) were collected.

Table 1

Demographic characteristics contrasting the subjects from the periodontitis group and control group, with the p-value resulting from the between-group comparison of all subjects.

	Periodontitis group			Control group			p value
Variable	Male	Female	All	Male	Female	All
Number of patients	14	53	67	25	49	74
Age (years)	49.14 ± 15.35	50.45 ± 13.58	50.18 ± 13.85	24.12 ± 2.35	24.63 ± 3.67	24.50 ± 3.28	<0.0001*
Number of current smokers	4	14	18	8	9	17	0.593
Number of former smokers	5	11	16	2	2	4	0.0017*
Diabetes (HbA1c < 7)	0	4	4	0	0	0	0.0330*
Cardiovascular disease	2	6	8	1	2	3	0.0812
Hypothyroidism (TSH<4 mUI)	2	8	10	0	1	1	0.0027*
Arthritis	0	2	2	0	1	1	0.502
Respiratory disease	2	2	4	1	2	3	0.600
Anti-diabetic medication†	0	4	4	0	0	0	0.0330*
Antibiotics†	0	2	2	0	3	3	0.731
Anti-inflammatory†	0	2	2	2	9	11	0.0149*
Antihypertensive†	2	4	6	0	1	1	0.037*
Anticoagulants1	1	5	6	0	1	1	0.0379*
Thyroid hormone thyroxine (T4) †	2	6	8	0	41	1	0.0102*
Hormonal Contraception†	0	11	11	0	21	21	0.0904
Stress	4	25	29	10	31	41	0.150

†Medication within a month

*Significant difference (p<0.05)

Table 2

Stages and grades in the periodontitis group according to the Chicago classification [1].

The percentage in each box refers to the relative frequency in relation to the total of 67 periodontitis subjects.

StageGrade	1	2	3	4	Total
A	0 (0,0%)	3 (4,5%)	0 (0,0%)	0 (0,0%)	3 (4,5%)
B	0 (0,0%)	5 (7,5%)	15 (22,4%)	6 (8,9%)	26 (38,8%)
C	0 (0,0%)	0 (0,0%)	16 (23,9%)	22 (32,8%)	38 (56,7%)
Total	0 (0,0%)	8 (11,9%)	31 (46,3%)	28 (41,8%)	67 (100%)

Differentially expressed peptide peaks

Saliva, GCF and dental plaque samples were analyzed by MALDI-TOF MS, and protein profiles were obtained from each sample in the range of 2–20 kDa.

A total of 217 peaks were detected in saliva, among which 114 were significantly different between the two groups (p < 0.05) (S2 Table). Among the top 10 ranking peaks, 8 (m/z values: 3372 Da, 3443 Da, 3519 Da, 3550 Da, 6352 Da, 6735 Da, 12692 Da, and 13461 Da) had higher levels of intensities in the periodontitis group, and 2 (m/z values: 2620 Da and 7746 Da) had higher levels of intensity in the control group (Fig 1A).

Fig 1

Box plots of the 10 top ranking peaks differentially expressed between the periodontitis and the control groups (a. saliva, b. GCF, c. dental plaque). Most of the differentially expressed peaks had higher levels of intensities in the periodontitis group in saliva and GCF, while all 10 top ranking peaks were found to be decreased in dental plaque from the periodontitis group.

A total of 176 peaks were detected in GCF, among which 110 were significantly different between the two groups (p < 0.05) (S3 Table). Among the top 10 ranking peaks, 9 (m/z values: 3775 Da, 4235 Da, 5296 Da, 5728 Da, 5893 Da, 10586 Da, 11324 Da, 11359 Da and 11447 Da) had higher levels of intensity in the periodontitis group, and only 1 (m/z values: 4944 Da) had a higher level of intensity in the control group (Fig 1B).

A total of 124 peaks were detected in dental plaque, among which 54 were significantly different between the two groups (p < 0,05) (S4 Table). All top 10 ranking peaks (m/z values: 2407 Da, 2627 Da, 2783 Da, 2818 Da, 3038 Da, 3077 Da, 3194 Da, 4065 Da, 4931 Da and 14693 Da) had lower levels of intensity in the periodontitis group (Fig 1C).

Construction of the informatics decision tree for periodontal diagnosis

On the basis of protein profiles obtained by MALDI-TOF MS, a diagnostic decision tree for periodontitis based on the differentially expressed peaks was made for each type of sample (Fig 2). For saliva, 7 peaks were selected to build a diagnostic decision tree with a sensitivity = 98% and a specificity = 91%. For GCF and dental plaque, 9 peaks were selected to build a diagnostic decision tree with a sensitivity = 96% and 94% and a specificity = 98% and 96%, respectively.

Fig 2

Diagnostic decision trees based on differentially expressed peaks between periodontitis and control groups (a. saliva, b. GCF, c. dental plaque).

In the blind experiment with a 10-fold cross-validation, diagnostic decision trees were confirmed with a sensitivity = 70.3% (± 0.211) and a specificity = 77.8% (± 0.165) for saliva, a sensitivity = 79.6% (± 0.188) and a specificity = 75.7% (± 0.195) for GCF, and a sensitivity = 72.1% (± 0.202) and a specificity = 72.2% (± 0.195) for dental plaque.

When a concatenation of the 3 decision trees was performed with a minimum of two samples tested, we found a sensitivity = 1 (CI 95% = [0.91;1]), a specificity = 1 (CI 95% = [0.92;1]), a positive predictive value = 1 (CI 95% = [0.91;1]) and a negative predictive value = 1 (CI 95% = [0.92;1]) for our study population.

Relation between the peaks and the epidemiological data

A principal component analysis was carried out between the 10 top ranking peaks and the epidemiological data from the medical questionnaire (age, gender, current smokers, former smokers, diabetes, cardiovascular disease, hypothyroidism, arthritis, respiratory disease, anti-diabetic medication, antibiotics, anti-inflammatory, antihypertensive, anticoagulants, thyroid hormone thyroxine, hormonal contraception and stress). The 10 top ranking peaks were found independent of the general risk factors for periodontal diseases, which means that these peaks are specific of the periodontal status.

Discussion

Due to high and constantly increasing prevalence of periodontitis and its consequences on quality of life, oral and general health, a rapid, minimally invasive and large-scale periodontal diagnosis test is of a particular interest, especially in patients with periodontal risk factors. In this study, direct samples analysis of dental plaque, GCF and saliva by a MALDI-TOF mass spectrometer used in routine diagnostic (Microflex LT, Bruker) combined with the use of an algorithm for sample classification resulted in an acceptable performance to correctly classify periodontitis and healthy periodontium subjects. Our study was the first, to our knowledge, to demonstrate that MALDI-TOF MS differentiates periodontitis from healthy periodontium by blind identification of specific patterns in mass signals from protein profiles in saliva, GCF and dental plaque. Previous encouraging results were obtained using MALDI-TOF MS analysis to investigate differences in protein profiles according to the periodontal state, but might be limited by the small sample sizes of the studies [28-30]. In our study, we obtained samples from 141 patients, 67 periodontitis patients and 74 control subjects, and found a large ratio of peaks significantly different between the periodontitis group and the control group in saliva (114 over 217), GCF (110 over 176) and dental plaque (54 over 124). These results reflect that composition of saliva, GCF and dental plaque present important differences in correlation with periodontitis and emphasize that periodontal diagnosis does not depend on a unique biomarker [16]. In comparison, Tang et al. detected 91 salivary peptide peaks, 7 of which were significantly different between the periodontitis and the control group, and 48 in the GCF among 4 were significantly different between the two groups [30]. These different results might be explained by a greater numbers of subjects (119 for saliva and 104 for GCF versus 33) and a larger range of detection (2-20kDa versus 1-10kDa) in our study and also variations in the statistical analysis.

In this study, GCF showed a strong ability to distinguish, in blind experiment, periodontitis patients from control subjects with a sensitivity = 79.6% (± 0.188) and a specificity = 75,7% (± 0,195). This result is not surprising because GCF reflects site-specific periodontal status, while saliva expresses the global status of the oral cavity. Periodontitis has been associated with changes in GCF composition and more than 90 different components have been investigated so far [37-39]. Some of these components have been found to be reduced from healthy subjects to patients with periodontitis, while others have been found to have concentrations positively correlated with periodontal inflammation [40,41].

Saliva profiles allowed periodontitis diagnosis, in blind experiment, with a sensitivity = 70.3% (± 0.211) and a specificity = 77.8% (± 0.165). This result supports the promising interest of saliva as a diagnostic fluid for screening patients with periodontal diseases. Numerous salivary components including locally produced proteins and salivary microbiota have been investigated as potential reflect of periodontal status [42,43]. For example, Prevotella has been found to be overabundant in the saliva of healthy subjects, while Porphyromonas, Tannerella, Desulfobulbus, Eubacterium, Phocaeicola and Mogibacterium were associated with the salivary microbiota of periodontitis patients [44]. We can state that these shifts in the saliva composition associated with periodontitis have a direct impact on saliva’s protein profiles, which could be used as a signature of the periodontal status. Moreover, the potentiality of saliva in screening patients with periodontitis needs to be investigated further as saliva collection is non-invasive, simple, fast and dispensable of professional skills and training [16].

The decision tree obtained from dental plaque samples showed a sensitivity = 72.1% (± 0.202) and a specificity = 72.2% (± 0.195) for periodontitis diagnosis in blind experiment. We found large variations in peptide profiles (43.5% peaks differentially expressed) in subgingival plaque samples from healthy subjects and periodontitis patients in our study. It was interesting to note that the 10 most differentially expressed peaks in dental plaque had lower levels of intensity in the periodontitis group. The genes Porphyromonas, Treponema, Tannerella, Filifactor, and Aggregatibacter were found to be more abundant in periodontitis patients, whereas Streptococcus, Haemophilus, Capnocytophaga, Gemella, Campylobacter and Granulicatella were observed at higher levels in healthy subjects [45]. We hypothesize that these peaks could be associated with non-periodontal pathogens as initial colonizers that were less abundant in patients with periodontal diseases. If the average richness of microbes in subgingival plaque from patients with periodontitis was higher than in samples from healthy subjects, the microbial communities associated with periodontal health were correlated with higher Shannon indexes [45]. Subgingival microbiota in periodontal diseases seemed to be characterized by more microbes with low relative abundance compared to healthy microbiota [45]. These findings can explain the lower level of intensity of the top 10 ranking peaks in the periodontitis group.

The concatenation of the decision trees has improved the diagnosis of periodontitis with a sensitivity = 1 (CI 95% = [0.91;1]), a specificity = 1 (CI 95% = [0.92;1]), a positive predictive value = 1 (CI 95% = [0.91;1]) and a negative predictive value = 1 (CI 95% = [0.92;1]). As a consequence, our test provides the best results when at least 2 different samples are tested. Saliva and dental plaque should be preferred because their collection is easier and faster than GCF. In comparison, Tang et al. have evaluated the differentially expressed peaks to distinguish subjects with periodontal diseases from healthy controls with Area Under the Curve’s (AUC) values in saliva and in GCF ranged from 0.688 to 0.860 and from 0.926 to 1.000 respectively [30].

To our knowledge, our study was the first to demonstrate that the protein profiles are exclusively specific to periodontal status independently of the periodontal risk factors. These findings make the prospect of a large-scale diagnostic test that could be used without consideration for subjects’ conditions other than periodontal status. Previous study exploring protein profiles in periodontitis diagnostic [30] was conducted in very limited population, free of periodontal risk factors and the results cannot be applied in a large-scale population as periodontitis is strongly linked to several periodontal risk factors [1,46]. By enlarging the selected population and ensuring the independence of the discriminating peaks, we wanted to get closer to the clinical reality.

A large proportion of our periodontitis group was classified into stages III (46,3%) and IV (41,8%), which meant that significant damage to the attachment apparatus has occurred, supporting delays of patients’ consultations and professional care. To date, no data related to protein profiles at each stage of periodontitis based upon the new classification (2018) are available. Future studies are needed to investigate whether the early stages (I or II) of periodontitis present the same protein profiles as advanced stages (III or IV).

MALDI-TOF MS is a simple, inexpensive and fast technique that analyses protein profiles with a high reliability rate, and could be used as a rapid screening method in a large population [47]. The observed changes in protein profiles could reflect specific microbiota, as well as the inflammation process with host immune response and periodontal tissue breakdown. The proteomic profile of complex samples like saliva, GCF and dental plaque should be considered as a phenotypic expression resulting from a vast molecular network including bacterial colonization and host immunity. The identification of specific biomarkers responsible for each peak or group of peaks represents a difficult and demanding task that requires further specific studies. The spectra comparison is based on protein fingerprinting. As Maldi-TOF MS is semiquantitative [20,48], it allows to work on the relative frequency of peaks within the same spectrum, the comparison of these relative frequencies between spectra being ensured by normalization (TIC, total ion current) [20, 49, 50]. This mechanism is the basis of the Maldi-tof bacterial identification technique currently used in routine [20, 51, 52]. In this pilot study, our methodology based on direct sample analysis and profiling obtained good diagnostic performance without identification of specific proteins and has the advantage to not require technical expertise and to be learned by any paramedical personnel. As our strategy was to assess if Microflex LT (Bruker Daltonics, Bremen, Germany) mass spectrometer laser could meet the need of a fast routine diagnostic test in periodontology that could be easily be applied in real conditions, we choose to not freeze the samples compared to the previously-mentioned studies and to carry out the analysis in a short period of time (within 24 to 48 hours). In a study evaluating the right sampling conditions to obtain the best correlation between oral fluid and plasma concentrations of unbound WAR, RR/SS-warfarin and RS/SR-warfarin alcohols, Lomonaco et al. have shown that no significant degradation occurred in the extracted oral fluid samples for at least two months of storage at 4°C [53]. Others substances like cortisol or uric acid remained stable when saliva was stored for at least 4 weeks at 4°C whereas salivary α-amylase activity has been found decreased of about 15% in the same storage conditions [54,55]. For GCF sample, a recent study showed that 10 peaks, over the 20 top signals analyzed, resulted significantly changed after 3 months of storage at -20°C compared to 1 month at -80°C but the authors did not evaluate storage at 4°C and shorter durations [56]. Despite the lack of data available concerning saliva, GCF and dental plaque storage conditions before direct MALDI-TOF MS analysis, the same conditions applied to all of our samples thus ensuring the correct comparability of our results. However, further studies would be necessary to assess the optimal time frame for a routine diagnostic test.

In this pilot study, in order to analyze the performance of the MALDI-TOF test we have randomly chosen a "healthy" population and a periodontitis population according to the criteria of the new classification [1]. We will need to refine the performance with patients from intermediate groups (children, mixed teeth, adolescents, implants, seniors) in further studies.

Conclusion

In the present study, we developed, for the first time, diagnostic tests based on protein profiles in saliva, GCF and dental plaque between periodontitis patients and healthy subjects. When at least 2 of these samples were tested, the best results were obtained. Because identifying patients at risk of periodontitis remains a challenge in periodontology, it is urgent to develop fast routine biological diagnostic tests, non-invasive and easy to perform by the practitioner.

List of epidemiological data recording during the medical questionnaire.

(PDF)

Click here for additional data file.

List of peaks detected in saliva.

(PDF)

Click here for additional data file.

List of peaks detected in GCF.

(PDF)

Click here for additional data file.

List of peaks detected in dental plaque.

(PDF)

Click here for additional data file.

4 Dec 2019

PONE-D-19-31089

Comparative peptide profiles of saliva, gingival crevicular fluid and dental plaque using MALDI-TOF MS on samples obtained from periodontitis patients and healthy control subjects

PLOS ONE

Dear Virgine Monnet-Corti,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by 3th January 2020. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Tommaso Lomonaco, Ph.D

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

Additional Editor Comments:

Dear Authors,

as suggested by the reviewers, the paper requires a major revision before to be accepted in PlosOne journal. In particular all the questions must be addressed properly. I suggest to extend the introduction by discussing the main advantages of saliva compared to other biological fluids (e.g. blood and its derivatives) as well as the possible application of saliva analysis in the field of therapeutic drug monitoring, biomarkers for diseases and etc.

The following articles can be useful for the authors and should be included in the references:

doi.org/10.1371/journal.pone.0028182

doi.org/10.1016/j.microc.2017.02.010

doi.org/10.1007/s00216-019-02158-6

As stated by the reviewer n° 2, the discussion must be implemented considering the fact that saliva collection protocols may alter the chemical composition of sample leading to not reliable data. The following articles can be useful for the authors and used for the discussion:

doi.org/10.1016/j.microc.2017.02.032

doi.org/10.1016/j.microc.2017.04.033

doi.org/10.1371/journal.pone.0114430

Best regards,

Tommaso Lomonaco

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: No

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The current article, titled “Comparative peptide profile of saliva, gingival crevicular fluid and dental plaque using MALDI-TOF MS on samples obtained from periodontitis patients ad healthy control subjects”, discusses the differentially expressed protein peaks between two groups of subjects using multiple minimally invasive sample types (saliva, gingival crevicular fluid and dental plaque). Some important questions remain unanswered and require addressing. A major revision is required.

Major comments:

1) How were proteins extracted form each sample? This experimental part in the materials and methods section is lacking. Saliva is an aqueous fluid containing high amount of salts, cells and debris that should be removed in order to investigate the proteomic content. Similarly, is not clear how the proteins were extracted from GCF and subgingival dental plaque samples.

2) From the mass spectrometry paragraph in the materials and methods section is stated that each sample was spotted onto the MALDI target with a volume of 0.5 uL. Was the total protein concentration of each sample evaluated? In order to compare the proteomic profiles of the two groups (Parodontitis and healthy subjects) the amount of proteins from each sample (saliva, GCF and subgingival dental plaque) spotted onto the MALDI target has to be the same. Before mass spectrometry analysis the total protein amount has to be estimated in each sample in order to spot onto the target the same amount of proteins (ug) from each sample. There are different protein assays available on the market to evaluate the total protein concentration.

My concern, is that the differences in the proteomic profiles, observed by the authors, could be indeed affected by the differences in the protein amount of each samples and not related with the disease status of the subjects included in the study.

3) Line 155: “In addition, the usual negative and positive controls were employed for each target”. What is the usual negative and positive control? Please clarify this.

4) Although low resolving power mass spectrometers are used, it should be standard practice to mention external mass spectrometer calibration accuracy.

5) How was the mass spectra pre-processing performed (i.e. mass spectra normalization, peak picking)?

Minor comments:

1) Line 24 – please explain the abbreviation GCF and add: gingival crevicular fluid (GCF).

2) Line 39 – Change “in” with “of”: “peptide profiles in saliva, GCF and dental plaque”.

3) Line 351 – add the word “first”: “In the present study, we developed, for the…time […]”.

4) Lines 394 and 397 – Please remove the “a” and “b” superscript.

5) Please check the order of the Reference List. The first paper cited in the Introduction, Papapanou et al, 2018 correspond to the reference number 20 in the Reference list. Please order the cited papers in order of appearance in the text.

Reviewer #2: The manuscript entitled: “Comparative peptide profiles of saliva, gingival crevicular fluid and dental plaque using MALDI-TOF MS on samples obtained from periodontitis patients and healthy control subjects” aims at determining if peptide profiles of saliva, GCF and dental plaque could be used for discrimination between patients with periodontal diseases and healthy periodontal subjects in a blind test.

Although the issue is interesting, in order to perform comparative peptide profile analysis using MALDI-TOF mass spectrometry the standardization of sample collection and preparation is a fundamental prerequisite. In fact, one of the requirements for a platform to be used in clinical proteomic studies aiming at biomarker discovery is the standardization of the preanalytical and analytical phase, which is essential for the generation of reproducible and robust MS-data. This is completely missing in the present manuscript!

Indeed, the MALDI-TOF MS analysis on samples obtained from 141 patients (67 periodontitis and 74 healthy controls) indicate that this was a mammoth work. However, considering that the standardization of proteomics procedures represents the fundamental prerequisite for diagnostic goals fulfillment, the reviewer recommendation is to reject the work at this stage due to major pitfalls described below in the experimental plan:

• The paper is lacking information about the assays for protein concentration of saliva, GCF and dental plaque and consequently, any kind of normalization for protein concentration has not been performed. The total amount of biological samples was not determined and without this important information it is uncertain whether the statistical difference among disease groups and healthy controls is due to a real difference in protein expression levels or purely reflects different amount of collected samples. These are the fundamental evaluations in order to standardize the protocol and to proceed with MS analysis for comparative studies. The authors should start from this stage to make a quantitative analysis comparable;

• Another important step in the standardization protocol is the storage conditions adopted for the biological samples. The authors wrote that: “Samples were immediately stored at 4°C and analyzed within 24 to 48 hours”. A number of studies demonstrated that the proteins can undergo degradation processes and that collection/handling and storage conditions may influence the stability of endogenous peptidome of biological fluids (Del Boccio et al, Ann Neurol. 2007; del Campo et al., Biomark Med. 2012; Preianò el al., Proteomics. 2016) therefore the authors should have preliminary performed experiments to identify the storage conditions which enable optimal preservation of biological specimen in order to ensure that the variation of protein expression levels in spectra reflects real biological differences rather than experimental artefacts. In this respect, the claims are NOT properly placed in the context of the previous literature.

• The authors should provide more information about MALDI sample preparation protocol, instruments settings adopted for spectra acquisition, and about the parameters for data processing (such as signal to noise and spectra normalization). The authors should investigate the impact of more strictly analytical variables on the generation of reliable MALDI-TOF spectra. It is well known that before starting MALDI based protein profiling study, it is necessary to assess the reliability of profiles with exploratory experiments in order to increase the analytical performances and the robustness of the results; in particular, they should have preliminary analyzed the influence of MALDI sample preparation by modulating the matrix composition and the analyte/matrix ratio in order to optimize the reproducibility of the MALDI peptidome profiles.

• The authors have not shown how reproducible the system is. They have not indicated the number of replicate analyses and the resulting coefficient of variations (CVs) considering that for diagnostic test CVs must fall in a range of 1.5-10% (Albrethsen J. Reproducibility in protein profiling by MALDI-TOF mass spectrometry. Clin Chem 2007;53:852–8).

• The authors must provide more information about the parameters for data processing (such as signal to noise and spectra normalization) and criteria of statistical analysis.

Additionally, it must be considered that for quantitative analysis, it is necessary to use an internal standard, for example an endogenous control protein, that can be observed within the same MS run for all the sample analysis. Alternatively, at least spectra normalization is required.

• The authors wrote: “All the patients were selected based upon periodontal status, regardless of the other criteria (e.g., systemic disease or disease/infection that may affect the periodontal health status, use of antibiotics or immunosuppressant medication within 3 months, current or former smokers), in order to reduce selection bias”. Are the investigators sure that the above mentioned conditions, such as use of antibiotics or anti-inflammatory do not interfere with the analysis of the molecular profile precluding the eligibility of the subjects for the present study?

• Why did the authors decide to consider only the top 10 ranking peaks for the discriminant analysis? I think that this choice could lead to a loss of information.

Finally, some minor points are listed below

• Figure 1 shows spectra obtained from saliva, GCF and dental plaque samples. Do these spectra derive from representative individuals? In order to make data clearer to the reader, the spectra should be shown in the appropriate m/z range for the best detection of molecular features of samples. Moreover, the top 10 ranking peaks found differentially expressed between the periodontitis and the control groups for each type of sample should be highlighted in the figure. Considering that the study focuses on MALDI-MS analysis, the choice of the representative spectra is very important. Are the authors sure that the selected spectra in Fig 1 are the most representative? Particularly, for GCF, the spectra appears poorly resolved and few peaks are characterized by a good S/N ratio.

• Figure 3 should be also revised, in order to make it clearer.

• The figure legends do not have all the information required to easily follow the discussion, so the authors must provide more details to the reader.

• In the discussion, experimental data are not appropriately commented and some link between literature studies and authors findings should be revised. The authors have not treated the literature fairly.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

30 Jan 2020

Response to Reviewer #1:

We thank you for your review of our manuscript. We have carefully taken into consideration your comments and answered each of your points below.

Major comments:

1) How were proteins extracted form each sample? This experimental part in the materials and methods section is lacking. Saliva is an aqueous fluid containing high amount of salts, cells and debris that should be removed in order to investigate the proteomic content. Similarly, is not clear how the proteins were extracted from GCF and subgingival dental plaque samples.

Answer: We corrected the paragraph “Mass spectrometry” (line 175) in the manuscript in order to better detail our methodology. Proteins were not extracted and samples were directly spotted on steel target and analysed by MALDI-TOS MS. We used the conventional technique described for bacterial identification called profiling (or fingerprinting) which is based on the comparison of spectra of unknown bacteria to database containing spectra of known reference bacteria. In this approach, intact cells can directly be mixed with MALDI matrix (Sandrin TR et al. Mass Spectrom Rev. 2013). Therefore, we chose to use whole samples to generate spectra that were compared to each other in order to assess if MALDI-TOF profiling applied to saliva, gingival crevicular fluid and dental plaque could be used as a diagnostic tool in periodontology.

2) From the mass spectrometry paragraph in the materials and methods section is stated that each sample was spotted onto the MALDI target with a volume of 0.5 uL. Was the total protein concentration of each sample evaluated? In order to compare the proteomic profiles of the two groups (Parodontitis and healthy subjects) the amount of proteins from each sample (saliva, GCF and subgingival dental plaque) spotted onto the MALDI target has to be the same. Before mass spectrometry analysis the total protein amount has to be estimated in each sample in order to spot onto the target the same amount of proteins (ug) from each sample. There are different protein assays available on the market to evaluate the total protein concentration. My concern, is that the differences in the proteomic profiles, observed by the authors, could be indeed affected by the differences in the protein amount of each samples and not related with the disease status of the subjects included in the study.

Answer: We used MALDI-TOF MS to perform profiling of saliva, gingival crevicular fluid and dental plaque from periodontitis and healthy subjects, which is not a quantitative analysis. (Sandrin TR et al, Mass Spectrom Rev. 2013). This approach was performed in our laboratory for arthropods identification (El Hamzaoui B et al, PLoS Negl Trop Dis. 2018; Tandina F et al, Parasitology. 2018).

3) Line 155: “In addition, the usual negative and positive controls were employed for each target”. What is the usual negative and positive control? Please clarify this.

Answer: We clarified this in the paragraph “Mass spectrometry” (line 175). The calibration of the mass spectrometer is performed with a commercial solution (BTS: Bacterial standard test) (line 190). In addition, in each MALDI target, 4 spots were made with an inactivated strain of Escherichia coli as the positive control (objective score > 2, 1) and 4 spots were made with matrix solution alone as the negative control (objective score <1, 5) (lines 192-194).

4) Although low resolving power mass spectrometers are used, it should be standard practice to mention external mass spectrometer calibration accuracy.

Answer: External mass spectrometer calibration accuracy was +/-300ppm.

5) How was the mass spectra pre-processing performed (i.e. mass spectra normalization, peak picking)?

Answer: We precised the mass spectra pre-processing in the paragraph “Mass spectrometry” (line 175). Quality criteria of the spectrum for global aspect and intensity were checked: intensity above 104 arbitrary units (AU), horizontal baseline curve and presence of visually identifiable peaks (lines 195-197).

Minor comments:

1) Line 24 – please explain the abbreviation GCF and add: gingival crevicular fluid (GCF).

Answer: We clarified it.

2) Line 39 – Change “in” with “of”: “peptide profiles in saliva, GCF and dental plaque”.

Answer: We changed it.

3) Line 351 – add the word “first”: “In the present study, we developed, for the…time […]”.

Answer: We added it.

4) Lines 394 and 397 – Please remove the “a” and “b” superscript.

Answer: We removed it.

5) Please check the order of the Reference List. The first paper cited in the Introduction, Papapanou et al, 2018 correspond to the reference number 20 in the Reference list. Please order the cited papers in order of appearance in the text.

Answer: We modified the order of the reference list.

Response to Reviewer #2

We thank you for your review of our manuscript. We totally agreed with all your comments and answered each of your points below.

Major comments:

1) The paper is lacking information about the assays for protein concentration of saliva, GCF and dental plaque and consequently, any kind of normalization for protein concentration has not been performed. The total amount of biological samples was not determined and without this important information it is uncertain whether the statistical difference among disease groups and healthy controls is due to a real difference in protein expression levels or purely reflects different amount of collected samples. These are the fundamental evaluations in order to standardize the protocol and to proceed with MS analysis for comparative studies. The authors should start from this stage to make a quantitative analysis comparable.

Answer: The aim of this pilot study was not to identify biomarker but to assess the feasibility and diagnostic contribution of profiling using MALDI-TOF applied to saliva, GCF and dental plaque from periodontitis and healthy subjects. We clarified it in the introduction of our manuscript. Profiling was described as a conventional approach for bacterial identification by comparing spectra of unknown bacteria to libraries that contain spectra of known reference bacteria. In this technique, intact cells can directly be mixed with MALDI matrix (Sandrin TR et al, Mass Spectrom Rev. 2013). We choose to adapt this non quantitative approach to oral fluids and dental plaque samples to generate spectra that were compared to each other and to evaluate if they could classify subjects according to their periodontal status. This approach has already been performed in our laboratory for arthropods identification (El Hamzaoui B et al, PLoS Negl Trop Dis. 2018; Tandina F et al, Parasitology. 2018).

2) Another important step in the standardization protocol is the storage conditions adopted for the biological samples. The authors wrote that: “Samples were immediately stored at 4°C and analyzed within 24 to 48 hours”. A number of studies demonstrated that the proteins can undergo degradation processes and that collection/handling and storage conditions may influence the stability of endogenous peptidome of biological fluids (Del Boccio et al, Ann Neurol. 2007; del Campo et al., Biomark Med. 2012; Preianò el al., Proteomics. 2016) therefore the authors should have preliminary performed experiments to identify the storage conditions which enable optimal preservation of biological specimen in order to ensure that the variation of protein expression levels in spectra reflects real biological differences rather than experimental artefacts. In this respect, the claims are NOT properly placed in the context of the previous literature.

Answer: We thank you for this pertinent remark and we argued our choice of methodology in the discussion. Del Boccio et al. (2007) and Del Campo et al. (2012) analysed cerebrospinal fluid among different storage conditions. As we did not use the same samples, we couldn’t use the results of these two studies, even though Del Campo et al. (2012) “recommended to store the samples at 4°C, since it can be easily done and it does not modify the biochemical results”. Preianò et al. (2016) found that best preserved signatures were obtained when GCF samples were stored at -80°C for 1 month compared to samples stored at -20°C during 3 months but the authors did not evaluate storage at 4°C and shorter durations. As Bellagambi et al. (Microchem J, 2018) and Lomonaco et al. (PLoS One, 2014; Microchem. J, 2018) found, oral fluids storage at 4°C during 4 weeks can preserve substances like cortisol, uric acid or WAR, RR/SS-warfarin and RS/SR-warfarin alcohols. Despite the lack of data available concerning saliva, GCF and dental plaque storage conditions before direct MALDITOF MS analysis, the same conditions applied to all of our samples thus ensuring the correct comparability of our results.

3) The authors should provide more information about MALDI sample preparation protocol, instruments settings adopted for spectra acquisition, and about the parameters for data processing (such as signal to noise and spectra normalization). The authors should investigate the impact of more strictly analytical variables on the generation of reliable MALDI-TOF spectra. It is well known that before starting MALDI based protein profiling study, it is necessary to assess the reliability of profiles with exploratory experiments in order to increase the analytical performances and the robustness of the results; in particular, they should have preliminary analyzed the influence of MALDI sample preparation by modulating the matrix composition and the analyte/matrix ratio in order to optimize the reproducibility of the MALDI peptidome profiles.

Answer: Our study was a pilot study and we clarified it in the introduction of our manuscript. The aim was to assess the feasibility and diagnostic contribution of protein profiling using MALDI-TOF applied to saliva, GCF and dental plaque from periodontitis and healthy subjects. We hypothesized that rapid routine and blinded MALDI-TOF analysis could accurately classify these three types of samples according to the periodontal state. We chose a methodology following as closely as possible the conventional MALDI-TOF protocol (used in routine diagnostic, i.e Microflex LT, Brucker and a mass range of 2-20 kDa) and we did not look for biomarkers identification.

4) The authors have not shown how reproducible the system is. They have not indicated the number of replicate analyses and the resulting coefficient of variations (CVs) considering that for diagnostic test CVs must fall in a range of 1.5-10% (Albrethsen J. Reproducibility in protein profiling by MALDI-TOF mass spectrometry. Clin Chem 2007; 53:852–8).

Answer: The number of replicate analyses was 6 (line 176 and line 184). The spectra generated for each sample of each subject were merged in order to develop a main spectra (MSP) (lines 202-205).

5) The authors must provide more information about the parameters for data processing (such as signal to noise and spectra normalization) and criteria of statistical analysis. Additionally, it must be considered that for quantitative analysis, it is necessary to use an internal standard, for example an endogenous control protein that can be observed within the same MS run for all the sample analysis. Alternatively, at least spectra normalization is required.

Answer: We provided more informations about the parameters for date processing in paragraphs “Mass spectrometry” (line 175) and “Bioinformatics analysis” (line 199). The signal-to-noise applied was = 3 (line 204). The calibration of the MS was fully automated and performed with a commercial solution (BTS: Bacterial standard test) and the procedure was completely automatic (BiotyperRTC user manual) (lines 189192).

6) The authors wrote: “All the patients were selected based upon periodontal status, regardless of the other criteria (e.g., systemic disease or disease/infection that may affect the periodontal health status, use of antibiotics or immunosuppressant medication within 3 months, current or former smokers), in order to reduce selection bias”. Are the investigators sure that the above mentioned conditions, such as use of antibiotics or anti-inflammatory do not interfere with the analysis of the molecular profile precluding the eligibility of the subjects for the present study?

Answer: The aim of our pilot study was to assess the feasibility and diagnostic contribution of profiling using MALDI-TOF applied to saliva, GCF and dental plaque from periodontitis and healthy subjects. This idea has already been investigated by Tand et al. (Clin Chim Acta. 2019) using saliva, GCF and serum samples in a very limited population, free of periodontal risk factors. In such conditions, their results deviated from target population and could not be applied to a routine large-scale diagnostic test. That’s why we decided to randomly choose a large healthy and periodontitis population in order to be closer to the clinical reality and to assess whether there were co-factors that could impact on the diagnosis.

7) Why did the authors decide to consider only the top 10 ranking peaks for the discriminant analysis? I think that this choice could lead to a loss of information.

Answer: We have chosen the 10 most discriminating peaks because only 7 were needed to build the diagnostic decision tree for saliva and only 9 for GCF and dental plaque. The objective was to select the minimum number of peaks to obtain the best sensitivity and specificity (lines 271-272).

Minor comments:

• Figure 1 shows spectra obtained from saliva, GCF and dental plaque samples. Do these spectra derive from representative individuals? In order to make data clearer to the reader, the spectra should be shown in the appropriate m/z range for the best detection of molecular features of samples. Moreover, the top 10 ranking peaks found differentially expressed between the periodontitis and the control groups for each type of sample should be highlighted in the figure. Considering that the study focuses on MALDI-MS analysis, the choice of the representative spectra is very important. Are the authors sure that the selected spectra in Fig 1 are the most representative? Particularly, for GCF, the spectra appears poorly resolved and few peaks are characterized by a good S/N ratio.

Answer : You are right, we have chosen an example that we are not sure is representative. We removed the figure.

• Figure 3 should be also revised, in order to make it clearer.

Answer : We totally agree with your remark and we removed the figure.

• The figure legends do not have all the information required to easily follow the discussion, so the authors must provide more details to the reader.

Answer : We removed figure 2 and figure 3 that were not clear.

• In the discussion, experimental data are not appropriately commented and some link between literature studies and authors findings should be revised. The authors have not treated the literature fairly.

Answer : You are right and we clarified our manuscript.

Response to Academic Editor, Tommaso Lomonaco, Ph.D

We thank you for considering our manuscript and for allowing us the opportunity to respond to the peer review comments. We found all the comments very helpful to improve our manuscript. We thank you for the six articles that you suggested and we added them, as you advised us, in the introduction and in the discussion.

We hope now that the revised manuscript will be suitable for publication.

Submitted filename: Response to Reviewers.pdf

Click here for additional data file.

19 Feb 2020

PONE-D-19-31089R1

Rapid diagnosis of periodontitis, a feasibility study using MALDI-TOF mass spectrometry

PLOS ONE

Dear Virginie Monnet-Corti,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by 25 February. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Tommaso Lomonaco, Ph.D

Academic Editor

PLOS ONE

Additional Editor Comments (if provided):

Dear Authors,

the paper requires additional revisions before to be accepted in Plos One. Generally, I suggest to emphasize that MALDI-TOF approach was used to preliminary investigate the composition of saliva collected from patients suffering from periodontitis. Regarding the title,

I prefer the new version.

Please clarify the following aspects:

1. I suggest to move the table 1 in the supplementary information.

2. Please explain at which variable the p-value was related (table 2).

3. please explain the meaning of the percentage in the brackets (table 3).

4. please explain the meaning of the number in the brackets (L374).

5. please include the RSD regarding replicate analysis on the same sample.

Regards,

Tommaso Lomonaco

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: I Don't Know

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The paper was not properly improved and I can not suggest its publication. The experimental design is not technically sound. In order to determine differentially expressed proteins (as stated even in the abstract, lines 33-34) he authors should have determined the total protein concentration of each sample and to spot the same amount of proteins, as a consequence all of the results are not consistent.

Reviewer #2: In the revised version of the manuscript PONE-D-19-31089R1 entitled: “Rapid diagnosis of periodontitis, a feasibility study using MALDI-TOF mass spectrometry” the Authors have improved the overall quality of the study however the manuscript still remains scientifically and methodologically not accurate because of

1) lack of experiments concerning the short-term storage at 4° of saliva, GCF and dental plaque. For comparative profiling studies, storage time must be the same for all the samples analyzed: the protein profile of a sample stored for 48 h at 4° may not be compared to that stored for 24 h at 4°. Storage time must be standardized to ensure the comparability of the results.

2) lack of precise quantification (for example the lack in the use of internal standards for MALDI-TOF analysis; moreover none of the various kinds of normalization for protein concentration has been performed), especially for peptide and protein profile obtained by MALDI-TOF MS, might be very dangerous because MALDI-TOF is not inherently quantitative.

These limitations could affect the results of the statistical analysis and raise doubts whether the statistical difference among disease groups and healthy controls is due to a real difference in protein expression levels or purely reflects different amounts of collected samples.

Therefore this reviewer strongly suggests that the limitations of this study should be highlighted and openly and critically discussed by the Authors.

All these issues are much more imperative in a “feasibility” study as the Authors claim in the new proposed title of the revised manuscript “Rapid diagnosis of periodontitis, a feasibility study using MALDI-TOF mass spectrometry”

It would be more correct, for the benefit of the readers, to replace the new proposed title “Rapid diagnosis of periodontitis, a feasibility study using MALDI-TOF mass spectrometry”

with a new one for example “Rapid diagnosis of periodontitis, a feasibility study using routine and blinded MALDI-TOF analysis”. In the opinion of this reviewer it is better to clarify the non-conventional use of MALDI-TOF mass spectrometry in this study starting from the title.

Considering that the data presented here are very weak because not well supported by the requested experiments, systematic bias in the design of this study may cause erroneous results causing false expectations in the wide readership of PLos ONE. Nevertheless, in a reasonable scenario it could be also possible that such study could lead to controversial debate on a possible misuse of MALDI as already happened for SELDI pioneering studies.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

25 Feb 2020

Dear Editor,

We would like to thank you and the Reviewers for your comments on our manuscript. We have edited our manuscript and our point-by-point responses to the comments are outlined below.

Sincerely,

Professor MONNET-CORTI Virginie, DDS-Ph.D

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

27 Feb 2020

Rapid diagnosis of periodontitis, a feasibility study using MALDI-TOF mass spectrometry

PONE-D-19-31089R2

Dear Dr. Virginie Monnet-Corti,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Tommaso Lomonaco, Ph.D

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Dear Authors, all the points were discussed in the revised versione of the paper and thus the manuscript can be published in PlosOne.

Regards,

Tommaso Lomonaco

2 Mar 2020

PONE-D-19-31089R2

Rapid diagnosis of periodontitis, a feasibility study using MALDI-TOF mass spectrometry

Dear Dr. Monnet-Corti:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Tommaso Lomonaco

Academic Editor

PLOS ONE