The Characterization of Newly Secreted Dental Enamel by Electron Energy Loss Spectroscopy

Teeth are one of the hardest organs in the human body. Typically, mature enamel contains up to 95% of inorganic minerals and 5% of organic components, with hydroxyapatite (HA, Ca10(PO4)6(OH)2) constituting the majority of the inorganic material [1]. However, during the development of the tooth, the forming HA phase in enamel lacks a high degree of crystallization, leading to cheesy texture and weak mechanical properties until the maturation process is completed. The abundance of protein in developing enamel also makes it sensitive to the electron beam, posing a challenge in characterizing evolving changes in crystallization. Moreover, in 2016, X-ray diffraction (XRD) analysis revealed the presence of octacalcium phosphate (OCP, Ca8H2(PO4)6) in the enamel of the amelogenin knockout mice [2]. In 2018, Yamazaki et al. detected a strong signal of OCP in the matrix metalloproteinase-20 (MMP20) knockout teeth using Raman microspectroscopy [3]. Lately, our previous research observed a large portion of OCP fan-shaped enamel crystals in the MMP20 knockout teeth with the characterization of the selected area diffraction pattern (SADP) [4]. Due to the similar composition and crystal structure, distinguishing between HA and OCP is difficult. While XRD and SADP allow identification of the crystal structure, they lack elemental and composition information. Raman microspectroscopy offer bonding information but lacks spatial resolution. Hence, a more powerful technique is needed for the characterization of abnormal secretory teeth. In this research, the analysis of electron energy loss spectroscopy (EELS) was presented to provide the elemental and bonding information and coupled with excellent spatial resolution for elemental mapping. Pure HA and OCP particles were first accessed as reference for EELS analysis. When observed under the transmission electron microscope (TEM), HA and OCP particles showed distinguishable shapes (Fig. 1A). The HA particles appear as small spherical clusters, and the OCP particles appear as large plane crystals. Fig. 1B, 1C, and 1D are the phosphorus L2,3-edge, calcium L2,3-edge, and oxygen K-edge EEL spectra respectively. These spectra are high-loss spectra, and they are associated with electron transitions between the orbitals of each element. For the phosphorus L2,3-edge, both spectra exhibited similar shapes. This similarity suggests that the bonding of phosphorus in both HA and OCP is alike. This result is consistent with the phosphorus configurations in HA and OCP as they both originate from the phosphate groups (PO43-). Similarly, the calcium L2,3-edge of HA and OCP were alike because all the calcium signals originated from Ca2+. In contrast, the oxygen K-edge from HA and OCP showed a slight difference. Two distinct peaks could be observed at 537 and 540 eV in the HA spectrum but less noticeable in the OCP spectrum. Because this difference was not very significant, an EELS simulation was conducted with FEFF software (Fig. 1E). The HA simulation revealed a two-peak feature, whereas the OCP simulation displayed only one peak. This simulation is consistent with the experimental results. This minor variation can be attributed to the hydroxide group (OH-) in HA [5]. The elemental quantification was carried out with a spatial resolution of 2 nm per pixel (Fig. 2). Mapping of the phosphorus and calcium signals not only facilitates the outline of the crystal shape clearly but also helps to distinguish between HA and OCP. In the overlay image (Fig. 2C), the color of the plane crystal appears redder than that of the spherical particles, implying different Ca/P ratios of the two crystalline phases. In addition, the plot of intensity versus distance (Fig. 2D) revealed similar intensity of the phosphorus and calcium signal in the HA particles, but the phosphorus intensity was much higher than the calcium in the OCP crystal. This plot assists in classifying HA and OCP while also indicating their boundaries. Aside from the mapping, because of the known Ca/P ratios of HA and OC, an experimental factor could be calculated for the use of relative quantification in further analysis. Two sites of forming enamel in wild-type mouse incisors were observed with EELS in this research. Fig. 3A and 3B shows the TEM images and the elemental mapping from the region of the dentin-enamel junction (DEJ). Comparing the two images, it was observed that the dark regions in the TEM image display a higher carbon signal and lack of crystals which are made of calcium and phosphorus. The fiber structure in the TEM images were clearly delineated in the mapping image. With the experimental factor calculated from the reference samples, the Ca/P ratio of the fibers and dentin was around 1.68, which is close to the Ca/P ratio of HA. Surprisingly, besides the fibers, certain regions (labeled stars in Fig. 3B) also displayed high calcium and phosphorus signals, and those regions contained higher carbon signals than the fibers. Fig. 3C and 3D represent another EELS acquisition that includes enamel crystals and space of Weber (SW). SW is a distinctive area usually lacking mineral, emerging during the development of enamel in rodent incisors. Fig. 3D illustrates low calcium and phosphorus intensity in the SW area. With this overlaid image, we can visibly observe the microstructure of the enamel crystal and note that there were strong carbon signals between enamel crystals. Fig. 3E revealed the O K-edge spectrum from the reference samples and enamel crystals. A small peak around 536 eV was observed in the enamel spectrum, suggesting that the early enamel crystals are more similar to HA than OCP. EELS stands out as a potential technique for material characterization, offering not only bonding and elemental information but also excellent spatial resolution for elemental mapping and delineation of material boundaries. In this study, the HA and OCP reference samples were characterized and differentiated with EELS analysis. The elemental mapping outlined the shape of the crystals and revealed different Ca/P ratios. With the established analysis method, a wild-type tooth sample was examined. With the EELS elemental mapping, the enamel crystals and enamel matrix were distinctly identified. Relative quantification and the feature of O K-edge implied immature enamel crystals were more similar to HA instead of OCP. In the future, further investigation on the MMP20 knockout tooth will be observed and analyzed with the same method to figure out more information about tooth development. (A) Scanning transmission electron microscopy (STEM) image of HA and OCP crystal. (B) Phosphorus L2,3-edge, (C) Calcium L2,3-edge, and (D) Oxygen K-edge EEL spectra of HA and OCP. (E) EELS simulation for Oxygen K-edge of HA and OCP with FEFF software. Elemental mapping with (A) Phosphorus L2,3-edge and (B) Calcium L2,3-edge EELS signal. (C) The overlay image of phosphorus L2,3-edge and calcium L2,3-edge signal. (D) The x-axis projection of phosphorus and calcium signal. The STEM image and elemental mapping (A, B) in the region of DEJ, and (C, D) in the region enamel layer including the space of Weber from the wild-type mouse incisor. (E) The O K-edge spectra of HA, OCP, and enamel crystals. SW, space of Weber.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.