Proteomic analysis of the acquired enamel pellicle formed on human and bovine tooth: a study using the Bauru in situ pellicle model (BISPM)

Abstract The acquired enamel pellicle (AEP) is an organic film, bacteria-free, formed in vivo as a result of the selective adsorption of salivary proteins and glycoproteins to the solid surfaces exposed to the oral environment. Objective: This study aimed to compare the proteomic profile of AEP formed in situ on human and bovine enamel using a new intraoral device (Bauru in situ pellicle model - BISPM). Material and Methods: One hundred and eight samples of human and bovine enamel were prepared (4×4 mm). Nine subjects with good oral conditions wore a removable jaw appliance (BISPM) with 6 slabs of each substrate randomly allocated. The AEP was formed during the morning, for 120 minutes, and collected with an electrode filter paper soaked in 3% citric acid. This procedure was conducted in triplicate and the pellicle collected was processed for analysis by LC-ESI-MS/MS. The obtained mass spectrometry MS/MS spectra were searched against human protein database (SWISS-PROT). Results: The use of BISPM allowed the collection of enough proteins amount for proper analysis. A total of 51 proteins were found in the AEP collected from the substrates. Among them, 15 were common to both groups, 14 were exclusive of the bovine enamel, and 22 were exclusive of the human enamel. Proteins typically found in the AEP were identified, such as Histatin-1, Ig alpha-1, Ig alpha 2, Lysozyme C, Statherin and Submaxillary gland androgen-regulated protein 3B. Proteins not previously described in the AEP, such as metabolism, cell signaling, cell adhesion, cell division, transport, protein synthesis and degradation were also identified. Conclusion: These results demonstrate that the proteins typically found in the AEP appeared in both groups, regardless the substrate. The BISPM revealed to be a good device to be used in studies involving proteomic analysis of the AEP.


Introduction
Teeth are constantly bathed by constituents from the gingival fluid, bacterial products, and by saliva. These constituents are rich in proteins and glycoproteins. As a result of this exposure, a bacteria-free organic film, known as acquired enamel pellicle (AEP), is formed 1 . The AEP formation is quick.
Scanning electron microscopy showed that AEP can be detected even one minute after the enamel samples are exposed to the oral cavity 2 . Moreover, another in vivo study using proteomic approaches detected the presence of 89 proteins within the AEP, formed 5 minutes after dental prophylaxis 3 , while a recent study identified 190 proteins within the acquired pellicle, formed in situ for 3 minutes on ceramic specimens 4 .
The protection of the tooth surface by the AEP is well established in the literature and has been demonstrated in several studies. The AEP acts as a diffusion barrier or permeable membrane, diminishing the direct contact between the acids and the tooth surface, thus reducing the dissolution rate of hydroxyapatite [5][6][7][8][9][10] . In vitro studies revealed that the first proteins to electrostatically interact with the enamel surface are proline-rich proteins (PRPs), statherin and histatins 11,12 , while in vivo experiments revealed also the presence of mucins, amylase, cystatin, lysozyme and lactoferrin in the very initial stages of pellicle formation 3,13 .
The comprehension of the AEP protein profile was greatly increased with the advent of proteomic tools.
Most of the AEP proteomic studies available so far were conducted in vivo 3,10,[13][14][15][16] . While the in vivo model provides the most clinically relevant information, in some cases it cannot be used. One of them is when it is desirable to know the protein composition of the acquired pellicle formed on mixed surfaces constituted by teeth and restorative materials. Another situation is when it is necessary to evaluate the protein composition of the acquired pellicle formed on dentin surfaces 17 , considering it is quite difficult to obtain exposed dentin surfaces in an extent that allows the collection of material to be analyzed. In these situations, in situ models are desirable and to the best of our knowledge only two studies are available in the literature so far 4,17 . In the study by Delecrode,et al. 17 (2015), human root dentin specimens were used in a palatal appliance. One of the main limitations of the study was the fact that only a few typical proteins of the acquired pellicle were identified. As for the study by Delius, et al. 4 (2017), the authors employed ceramic specimens. Despite they were able to identify more than 100 proteins, some of which are typically found in the AEP 4 , the composition of the ceramic specimens is quite different of human enamel, which certainly impacts in the protein profile of the acquired pellicle.
In studies involving dental caries and erosion, bovine teeth, which are easier to obtain, are often used as surrogate for human teeth 18-20 . However, there are no studies comparing the protein profile of the AEP formed on human and bovine specimens.
Considering that structural differences between these two types of substrates exist 21 , with bovine crystallites being thicker 22 and bovine enamel presenting higher radiographic density 23 than the human counterparts, there could be differences in the protein profile of the AEPs formed on these two types of substrates.
In addition, one of the main limitations of studies involving proteomic analysis of the acquired pellicle is to obtain enough material to be analyzed. Thus, it is of great interest to develop devices to be used in in situ studies that make possible the collection of appropriate amounts of AEP to allow proper protein identification and quantification in proteomic studies. Therefore, this study aimed to compare the proteomic profile of

In situ experiment
The experiment was conducted during the morning to abstain from circadian effects on the composition of the pellicle 16 for 3 consecutive days, aiming to obtain enough material to be submitted to the proteomic analysis. Firstly, the volunteers inserted the intraoral device into their mouths and for 120 minutes they were instructed not to eat or drink to allow the AEP to form on the enamel surfaces. For the AEP collection, the intraoral device was removed from the mouth and the samples were washed with deionized water then dried by air. An electrode filter paper 5x10 mm (Electrode Wick, Bio-Rad ® , Hercules, Califórnia, USA) pre-soaked in 3% citric acid 10        Regarding the quantitative analysis, two proteins increased in human enamel, when compared to bovine enamel (Lysozyme C and Pancreatic alpha-amylase), while two isoforms of Actin cytoplasmic, Histatin 1 and Statherin decreased (Table 1).

Discussion
The main challenge in studies involving proteomic of the AEP is to obtain enough protein to allow proper analysis. To overcome this difficulty, we developed a new device, the BISPM that has a special design in order to optimize the collection of enough proteins from the AEP. The placement of number 0.8 orthodontics wire suspended above the specimens to avoid direct contact of the mucosa with them 24 was the main responsible for it. In addition, the experiment was carried out in 3 consecutive days, and the samples collected from the same treatments were pooled. Furthermore, we worked with a mandibular apparatus instead of a palatal one, because more saliva is expected to bath the specimens in the first condition due to the gravity force. These strategies were effective to allow enough proteins in the AEP to be analysed. However, it has been shown that the composition of the AEP changes according to its location in the dental arches 15 , which is a limitation of our model, despite a recent study revealed no difference in the protective ability against initial erosion of the AEP formed in situ using palatal and mandibular intraoral device 24 . It is worth mentioning that the removable apparatus developed was well tolerated by the volunteers, without any reported discomfort.
In this study, the total number of identified proteins, when both substrates are considered, was 51. This is quite similar to the number of proteins identified in an experiment where the pellicle was collected from dentin specimens in situ using a mandibular device 17 . were able to identify many typical proteins, such as isoforms of Cystatin, Actin, Alpha-amylase and Ig A, Histatin 1, Lysozyme C, Statherin, Submaxillary gland androgen-regulated protein 3B and Alpha-amylase.
These differences might be explained due to the distinct types of substrates (dentin X enamel).
In a recent study, the authors collected the acquired pellicle formed for 3 min on ceramic specimens in situ and a total of 190 proteins were identified 4 , among which 58% have been described in the pellicle before 4 and some of them were also identified in this study, This study aimed to compare the protein composition of the acquired pellicle formed on bovine and human enamel. Bovine enamel has been used as surrogate for human enamel dental research.
There are publications comparing both substrates regarding chemical composition, physical properties, dental caries, dental erosion/abrasion, bonding and microleakage studies 27 , however, the composition of the acquired pellicle formed on both substrates has never been compared. Bovine teeth are much easier to obtain than human teeth and are also bigger, thus providing specimens with a higher surface area, which is a desirable characteristic for studies involving collection of acquired pellicle. In this study, despite