Comparative Analysis of the Proteomic Profile of the Dental Pulp in Different Conditions. A Pilot Study.

This study aimed to quantitatively compare the difference in protein expression in the progression of pulp pathogenesis, as well as to describe the biological functions of proteins identified in pulp tissue. Samples were obtained from six patients treated at the Araçatuba School of Dentistry and were divided into three groups: normal pulp - from teeth extracted for orthodontic indication; inflamed pulp and necrotic pulp - from patients diagnosed with irreversible pulpitis and chronic apical periodontitis, respectively. After previous proteomic preparation, dental pulp samples were processed for label-free quantitative proteomic analysis in a nanoACQUITY UPLC-Xevo QTof MS system. The difference in expression between the groups was calculated using the Protein Lynx Global Service software using the Monte Carlo algorithm. A total of 465 human proteins were identified in all groups. The most expressed proteins in the inflamed pulp group in relation to the normal pulp group were hemoglobin, peroxiredoxins and immunoglobulins, whereas the less expressed were the tubulins. Expression levels of albumins, immunoglobulins and alpha-2-macroglobulin were higher in the necrotic pulp group than in the inflamed pulp group. As for the qualitative analysis, the most prevalent protein functions in the normal pulp group were metabolic and energetic pathways; in the inflamed pulp group: cellular communication and signal transduction; and regulation and repair of DNA/RNA, while in the necrotic pulp group proteins were associated with the immune response. Thus, proteomic analysis showed quantitative and qualitative differences in protein expression in different types of pulp conditions.


Introduction
Dental pulp is a complex specialized tissue with nourishing, restorative, sensorial and defensive functions (inflammatory or degenerative defenses, depending on the characteristics of the irritant). Although several etiological factors can cause pulp damage, the biological factor, represented by polymicrobial communities usually organized as biofilm, is the most relevant (1). The characteristic events of the inflammatory process, such as the recruitment of defense cells from the innate and adaptive immune response, and the presence of chemical mediators alter the physiology of the dental pulp in the attempt to remove the aggressive agent. Damage to the pulp tissue can occur in the persistence of the irritant agent, destructuring the tissue and leading to a process of necrosis. In this context, bacteria colonize the root canal system and start releasing potential antigens into the periapical tissues, thus triggering the development of apical diseases (2,3).
Proteomic analysis is the study of the whole set of proteins expressed by an organism in a particular environment during a specific stage of the cell cycle. It also covers their relative abundance, distribution, functions and interaction with other macromolecules (4). This approach has also been widely applied in medical microbiology, collecting information related to bacterial resistance and virulence, which has been used in the development of new diagnostic and therapeutic applications for the treatment of infectious diseases (5). Therefore, proteomic techniques are important tools for investigating the progression of both pulp alterations and study of the host / pathogen response to infections (6).
Characterization and expression of proteins using liquid chromatography (LC) / mass spectrometry (MS / MS) has gained prominence in proteomic analysis of pulp tissue, since this method provides the necessary technology for the study of small amounts of samples from complex biological systems (7). This method involves the proteolytic digestion of all proteins in the samples and their subsequent identification using a database of individual peptides, reducing sample handling time and eliminating the need for individual protein processing (8). The direct identification of proteins expressed in pulp and periapical diseases, involving descriptive analysis of pulp pathogenesis, focusing on: primary and persistent infections (9); endodontic abscesses (10); and cases of failure of endodontic treatment (11) has allowed the identification of several human proteins, which were mainly related to cellular processes, metabolism and immune defense.
Although the above-mentioned studies provided significant contribution to the comprehension of endodontic infections, they employed qualitative identification approaches, which do not allow a direct quantitative comparison of the expression of proteins of human origin. Furthermore, to date no study has assessed the proteomic profile of the human pulp tissue with different degrees of microbial injury, especially determining quantitatively the down-and up-regulated proteins. Therefore, considering the scarcity of proteomic studies of pulp diseases, this pilot study aimed to compare the protein profile at different clinical stages during the progression of pulp diseases, as well as to describe the biological functions of each protein detected in the samples within the different pulp conditions (normal, inflamed or necrotic pulp tissue).

Patient Selection
All patients signed an informed consent form prepared in accordance with the rules of the Research Ethics Committee of the Araçatuba School of Dentistry -UNESP (Nº 91331518.7.0000.5420). Samples were taken from patients with no history of systemic diseases, who attended the Endodontic Clinic of the Araçatuba School of Dentistry -UNESP for root canal treatment, and were divided into three groups: normal pulp group with pulp tissue samples obtained from teeth extracted for orthodontic indication (n=2); inflamed pulp group -samples obtained from patients diagnosed with irreversible pulpitis (n=2), and necrotic pulp group -samples obtained from patients diagnosed with chronic apical periodontitis (n=2). Clinical and radiographic characteristics and a detailed anamnesis of the patient's health conditions were recorded (Table 1).

Sample Collection
The collection was done aseptically. Firstly, the crown of the tooth to be sampled was cleaned with pumice paste and water, followed by removal of the restoration and/ or carious tissue without exposing the root canals. Then, the tooth was individually isolated from the oral cavity with a rubber dam, except for the normal pulp sample. The tooth and the surrounding field were cleaned with 30% hydrogen peroxide and decontaminated with 2.5% sodium hypochlorite solution for 30 s each, followed by neutralization of the solution with 5% sodium thiosulfate (3). The access to the pulp cavity was performed with sterile carbide bur without water spray. Irrigation during the access phase was done with sterile saline solution.
Normal pulp -To obtain the normal pulp sample, teeth with healthy pulp and without periodontal disease, indicated for orthodontic extraction, were selected. Immediately after the extraction, the pulp tissue was carefully removed with the aid of sterile manual files (Hedstroem file size #20, Dentsply Sirona, Ballaigues, Switzerland), avoiding contamination and complete disruption of the pulp tissue.
Inflamed pulp -In cases of irreversible pulpitis, consistent pulp tissue was collected from the palatal or distal canal with Hedstroem file, complemented by three sterile paper points introduced into the apparent length of the canal determined on diagnostic radiographs, and held in place for 60 s each, without any irrigation.
Necrotic pulp -Samples of the teeth with pulp necrosis, with radiographic lesion (chronic apical periodontitis), were obtained immediately after exposure of the pulp chamber. A sterile K-type file was introduced with minimal instrumentation, without the use of any irrigant to disrupt biofilms of the canal wall; then three sterile paper points were introduced into the apparent length of the canal determined on diagnostic radiographs and held in place for 60 s each. If the canal was completely dry, a drop of sterile saline was placed before removing the paper point. In cases of teeth with more than one canal, the sample was collected only from the widest canal, since it was associated with the apical lesion (otherwise the tooth would not be included in the study), to confine the analysis to a single environment.
After the collection, the paper points and tissue samples were placed in sterile, DNA-free and RNA-free cryotubes, which were frozen at -80° C until use for proteomic analysis.

Proteomic Analysis -Preparation of Pulp Samples
The paper points were cut and samples corresponding to the same groups were pooled. In the tubes containing the paper points an extraction solution containing 6 M urea, 2 M thiourea in 50 mM NH 4 HCO 3 pH 7.8 was added until the papers were covered. The samples were then vortexed for 10 min at 4 °C, followed by sonication for 5 min and centrifugation at 20,817 × g for 10 min at 4 °C. The supernatant was collected, and this procedure was repeated once more. The papers were placed in filter tubes (Corning® Costar® Spin-X® Plastic Centrifuge Tube Filters Sigma-Aldrich, New York, USA) and centrifuged at 20,817 × g for 10 min at 4 °C. The supernatant was collected and added to the previously collected supernatant. Soon after, 1.5 mL of 50 mM NH 4 HCO 3 was added to the samples. The samples were then placed in Falcon Amicon Ultra-4 10k tubes (Merck Millipore, Ireland) and centrifuged at 4,500 × g at 4°C to approximately 150 μL.
Then, 5 mM dithiothreitol (DTT) was added to the samples and they were incubated at 37°C for 40 min. After this time, 10 mM iodoacetamide (IAA) was added and the samples were incubated for 30 min in the dark. After the incubations, 100 μL of 50 mM NH 4 HCO 3 were added and shortly thereafter the tryptic digestion was performed for 14 h at 37°C by the addition of 2% (w/w) trypsin (Promega, Madison, USA). After the digestion, 5% formic acid was added to stop the action of trypsin and the procedures were performed with the C18 spin column (Thermo Scientific, United States) for desalting and purifying the samples. Thus, an aliquot of each sample (1 μL) was removed and protein quantification was performed by the Bradford method (Bio-Rad Bradford Assays). The remnants were dried to approximately 1 μL in SpeedVac (Thermo Scientific, United States). After drying the samples were resuspended in 3% acetonitrile and 0.1% formic acid for the application to the nano Liquid Chromatography Electron Spray Ionization Tandem Mass Spectrometer (nLC-ESI-MS / MS) (12).

Shotgun Label-Free Quantitative Proteomic Analysis
Peptides identification was performed on a nanoACQUITY UPLC-Xevo QTof MS system (Waters, Manchester, New Hampshire, UK). The nanoACQUITY UPLC was equipped with nanoACQUITY HSS T3, analytical reverse phase column (75 μm X 150 mm, 1.8 μm particle size (Waters, Manchester, New Hampshire, UK). The column was equilibrated with mobile phase A (0.1% formic acid in water). Then, the peptides were separated with a linear gradient of 7-85% mobile phase B (0.1% formic acid in ACN) for 70 min at a flow rate of 0.35 μL/min. The column temperature was maintained at 55 °C. The Xevo G2 Q-TOF mass spectrometer was operated in positive nanoelectrospray ion mode and data were collected using the MSE method in elevated energy (19-45 V), which allows data acquisition of both precursor and fragment ions, in one injection. Source conditions used included capillary voltage, 2.5 kV; sample cone, 30 V; extraction cone, 5.0 V and source temperature, 80 °C. Data acquisition occurred over 70 min and the scan range was 50-2000 Da. The lock spray, used to ensure accuracy and reproducibility, was run with a [Glu1] fibrinopeptide solution (1 pmol/μL) at a flow rate of 1 μL/min, as a reference ion in positive mode at m/z 785.8427. ProteinLynx Global Server (PLGS) version 3.0 was used to process and search the LC-MSE continuum data. Proteins were identified with the embedded ion accounting algorithm in the software and a search of the Homo sapiens database (UniProtKB/Swiss-Prot) downloaded on April 2017 from UniProtKB (http://www.uniprot.org/).
For label-free quantitative proteome, three MS raw files from normal, inflamed and necrotic pulp groups were analyzed using the Protein Lynx Global Service (PLGS, v 2.2.5, Waters Co., Manchester, UK) software. All the proteins identified with a score with confidence greater than that 95% were included in the quantitative statistical analysis embedded in the PLGS software. Identical peptides from each triplicate by sample were grouped based on mass accuracy (<10 ppm) and on time of retention tolerance <0.25 min, using the clustering software embedded in the PLGS. Difference in expression among the normal and inflamed pulp groups and inflamed and necrotic groups was calculated using Monte-Carlo algorithm and expressed as p< 0.05 for proteins present in lower abundance and 1-p>0.95 for proteins present in higher abundance.
In the quantitative analysis, two comparisons were made among the groups: The first comparison was between the normal and inflamed pulp groups, and the second was between the inflamed and necrotic pulp groups. Proteins expressed at a ratio >2.0 in relation to the group of comparison were regarded as up-regulated, while proteins expressed at a ratio <0.5 were regarded as down-regulated proteins. Proteins expressed with ratio between 0.5 and 2 were disregarded. The identified proteins were classified according to their biological functions using Homo sapiens database (UniProtKB/Swiss-Prot).

Results
Overall, 465 proteins were identified from the samples in all groups. Among them, 30 were common to all groups, including six isoforms of Actin, Albumin, Alpha-1_4 glucan phosphorylase, three isoforms of Glycogen and six isoforms of Hemoglobin, Serum albumin, among other proteins such as Apolipoprotein A-II, Haptoglobin and three isoforms of Immunoglobulin (Fig. 1).
The proteins expression were divided into 12 categories: metabolism and energy pathways, immune response, transport, structure, DNA/RNA regulation and repair, cell communication and signal transduction, cell growth and/ or maintenance, differentiation of neural cells, apoptosis, stress response, ions regulation and binding and proteins of unknown function (8) (Tables 2 and 3).
When comparing the inflamed pulp group with the normal pulp group, 39 proteins were found at higher levels in the first, among which 18 had more than a 2-fold increase, e.g. 4 subunits of Hemoglobin (100-fold higher), 2 isoforms of Peroxiredoxin, (20-fold higher), 3 isoforms of Immunoglobulin (10-fold higher), Glyceraldehyde-3phosphate dehydrogenase (2-fold higher). On the other hand, 41 proteins were found at lower levels in inflamed pulp group in comparison with normal pulp group, among which 17 were isoforms of Tubulin, besides other cytoskeletal proteins, such as Desmin. Serum albumin and neurofilament proteins, such as Alpha-internexin, Neurofilament medium polypeptide were also reduced ( Table 2).
When necrotic pulp group was compared with inflamed pulp group, 8 proteins were found at higher levels and 26 at lower levels in the necrotic pulp group. Among the high level proteins, 8 of them were more than 2-fold higher (2 isoforms of Serum albumin, Immunoglobulin heavy constant gamma 1 and Alpha-2-macroglobulin). As for the low level proteins, 13 were more than 2-fold lower (various isoforms of Hemoglobin, various isoforms of POTE ankyrin domain, various isoforms of Actin and Bromodomain-containing protein 3) ( Table 2). Table 3 shows the proteins exclusively identified in each one of the groups. Most of the proteins identified in the normal pulp group were involved in metabolic and energy pathways (20.4%) and in cell communication and signal transduction (20.4%). While in the inflamed pulp group, the cellular communication and signal transduction function (19.4%) also presented a higher percentage, followed by regulation and repair of DNA / RNA (17.9%). Finally, in the necrotic pulp group, most proteins were involved in the immune response (24.3%). (Fig. 2).
Alpha-2-macroglobulin, Transthyretin and Apolipoprotein A-I were not identified in the normal pulp group, while Beta-actin-like protein 2 was not found in necrotic pulp group. Isoforms of Neutrophil defensin were not found in the groups of normal and inflamed pulp, while Serotransferrin was identified in the groups of normal and necrotic pulp.

Discussion
Proteomic techniques have helped to improve the knowledge of the biology, function and pathology of the pulpal tissues. Tissue formation, diagnosis, identification of risk factors, tissue engineering and pathogenesis of endodontic infections represent some of the most important themes investigated in proteomic studies of the pulp. The understanding of proteins and their functions provide insights into the complex host-pathogen relationship and host antimicrobial strategies to combat infections (13). To our knowledge, this pilot study is the first to describe and quantify the proteome of the pulp tissue in relation to the progression of pulp diseases, comparing the protein profile of different pulp diagnoses and their relationship with the characteristic events of each diagnosis. The nLC-ESI-MS/ MS method allowed the identification of 465 different proteins. The high sensitivity of the method makes it possible to identify and comparatively quantify proteins in tiny amounts of samples, such as samples for root canals, allowing the study of its pathological processes (14).
In order comprehend the mechanisms related to the host in relation to the progression of the pulp disease, two comparisons were made between the groups. The first comparison (normal vs inflamed tissue) showed a significant up-regulation of 4 subunits of Hemoglobin. Approximately one third of the mass of a human red blood cell (RBCs) is hemoglobin (15). RBCs participate in the vascular system and its increase is related to the diagnosis of the pulp. The inflammation of the dental pulp causes an immediate increase in blood flow, along with vasodilation, increased blood supply and microcirculation (16).     The immunoinflammatory response is intended to restore the structural and functional integrity of the injured tissue by eliminating irritants as quickly as possible (17). Indicating the activity of the immune response in inflamed pulp, three immunoglobulin isoforms were present in greater amounts in the inflamed tissue than in the healthy tissue samples. The increase of Nuclear mitotic apparatus protein indicates cell division, probably related to the proliferation of immunoinflammatory cells.

Proteomic analysis of human dental pulp
The release of reactive oxygen species (ROS) by disintegrated neutrophils and macrophages within the pulp tissue might damage these tissues. ROS are mandatory byproducts of the metabolic activities of living aerobic organisms (18). Removal of ROS is done by peroxiredoxins -a family of antioxidant proteins that catalyze these substances. Two isoforms of Peroxiredoxins were upregulated in the inflamed pulp group when compared to the normal tissue, which are responsible for the protection of cellular components against oxidative damage. They are involved in processes such as cell proliferation and differentiation, protection of free radical-sensitive proteins, hemoglobin metabolism and intracellular signaling (19). Throughout the inflammatory process maintained by the aggressive agent, damage to the pulp tissue occurs, with consequent cell death and destruction of the extracellular matrix. The down-regulation of 17 isoforms of Tubulin in the inflamed pulp group shows the disorganization and destruction of the structural portion of the cell in front of this process (20). Neurofilament proteins, such as Alphainternexin and Neurofilament medium polypeptide were also down-regulated, which could be predicted due to the intense inflammation present in irreversible pulpitis. Some other proteins were also found with a high level expression in inflamed pulpal tissue, most of them participating in biological processes related to transport, and metabolism and energy pathways.
As a result of the evolution of the inflammatory process of the pulp, the vital functions of the pulp are compromised, followed by hypoxia and tissue necrosis. In this context, there are also changes in blood microcirculation that led to reduced pulp blood flow, explaining the down-regulation of hemoglobin in the necrotic pulp group, when compared with the inflamed one (21). The down-regulation of 4 isoforms of actins show that the mortification process of the pulp tissue leads to destruction of the cytoskeleton and rupture of actin microfilaments. Actin is a protein involved in structuring the cytoskeleton. The actin microfilaments participate in the generation of forces and cell adhesion, stabilizing the cell and determining the shape of the plasma membrane (22).
Among the proteins up-regulated in the necrotic group, serum albumin, albumin, immunoglobulin, and alpha-2macroglobulin were found. Proteins derived from albumin and serum albumin are constituents of fluids and exudates that infiltrate the apical and lateral foramen of the root canal. Albumin and immunoglobulins may be related to the immune response as they participate in reducing the diffusion of antigens when they adhere to the dentinal tubules (23).
Immunity-related proteins, such as immunoglobulins and protease inhibitors, involved in antigen presentation, defense cell activation and stress response can be identified in necrotic pulps, suggesting that host cells react to root canal system infections (24). One of the proteins found in high level in this group, Alpha-2-macroglobulin (α2M), protects the body against bacterial endotoxins, regulating apoptosis and inhibiting the generation of hydrogen peroxide. In addition, α2M can be used as a biomarker for the diagnosis and prognosis of various diseases (25).
The most recurrent biological processes found in normal pulp tissue provide tissue balance, including maintenance, renewal and energy supply for cellular interaction. In the inflamed pulp group, there was an increase in the percentage of proteins involved in the regulation and repair of DNA / RNA in order to allow cell viability. This may have occurred due to the damage suffered by the cells of the pulp tissue during the inflammatory process. Meanwhile, the increase of proteins associated with metabolism and energetic pathways is directly related to the greater cellular activity for the elimination of the aggressive agent. Moreover, samples representing the infected pulp had a higher percentage of proteins with biological function related to the immune response, which was also described by Provenzano et al. (24), revealing the presence of viable host cells at the site of infection. These results contribute to the understanding of the complex pathogen-host relationship, the host's antimicrobial strategies to fight the infections and shed light into the pathogenesis of the disease.
The present study was the first report to analyze both qualitative and quantitatively proteins differently expressed in normal, inflamed and necrotic pulp, thus providing important data that might not only contribute to the understanding of the complex pathogen-host relationship involved in the progression of pulp diseases, but also to direct future researches. Moreover, the present study reported higher levels of proteins that are considered constitutively produced and ideally should be constant such as beta actin, Glyceraldehyde-3 phosphate dehydrogenase, macroglobulin and tubulin. Therefore, future studies using quantitative methods such as Real time-PCR and Western Blotting analysis should avoid targeting such genes/proteins as reference when analyzing samples related to pathosis of the pulp.
In conclusion, this proteomic analysis showed quantitative differences in protein expression in different types of pulp conditions and revealed that pulp inflammation induced a high-level expression of proteins related to cellular communication and signal transduction. Nevertheless, with the progression to pulp necrosis, the proteins were associated with immune response.