If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Address requests for reprints to Dr Franz E. Weber, University of Zurich, Center for Dental Medicine/Cranio-Maxillofacial and Oral Surgery, Oral Biotechnology and Bioengineering, Plattenstrasse 11, CH-8032, Zürich, Switzerland.
Under the conditions of the current study using human dentin disks, a 10% citric acid solution (pH = 2) was significantly (P < .05) better than 17% EDTA (pH = 8) regarding the following: -The release of transforming growth factor beta 1 (TGF-β1) into solution. -The exposure of TGF-β1 on the dentin surface. -The attraction of stem cells toward conditioned dentin. -Stem cell attachment on the conditioned surface. -Stem cell survival.
In carious teeth, transforming growth factor beta 1 (TGF-β1) is released from the dentin matrix and possibly activated in an acidic environment. Conversely, EDTA solutions with a neutral to slightly alkaline pH are used in clinics to promote cell homing in regenerative endodontic procedures. We hypothesized that citric acid (CA) might be more beneficial.
TGF-β1 release from human dentin disks conditioned with either 10% CA (pH = 2) or 17% EDTA (pH = 8) and the behavior of human stem cells toward such pretreated dentin were studied. The protein concentration in conditioning solutions after 10 minutes of dentin exposure was determined using a pH-independent slot blot technique.
There was a 5-fold higher concentration of the target protein in CA (382 ± 30 ng/disk) compared with EDTA (66 ± 3 ng/disk, P < .005). Using confocal laser scanning microscopy on immunofluorescent-labeled disks, we identified a high density of TGF-β1 in peritubular dentin after CA treatment. A migration assay showed that CA conditioning attracted significantly more stem cells toward the dentin after 24 hours compared with EDTA (P < .05) or phosphate-buffered saline (P < .005). To investigate whether the cell response to these dentin surfaces could be affected by different pretreatments, we cultured stem cells on conditioned dentin disks and found that CA had a significantly (P < .05) better effect than EDTA on cell attachment and cell survival.
CA conditioning could be useful and may have significant benefits over current treatments.
The results of the current study suggest the possibility that a 10% citric acid solution could be more beneficial than the gold standard 17% EDTA solution for cell homing in regenerative endodontic procedures.
A long-standing aspiration in dentistry has been to regenerate a pulp (
). The dentin matrix acts as a reservoir from which growth factors can be released on demand. In nature, this occurs (eg, in an acidic environment during development of caries to promote tertiary dentinogenesis underneath the site of attack) (
). Hence, the goals of the present study were 2-fold:
To check the release of TGF-β1 from human dentin treated with either CA or EDTA using a pH-insensitive method and to assess the induction of cell migration
To visualize TGF-β1 on dentin by immunofluorescence confocal laser scanning microscopy and to assess cell–pretreated dentin interaction
We speculated that CA should be more potent than EDTA to release and expose TGF-β1 from/on dentin and should thus have a better effect on cell attraction, attachment, and survival.
Materials and Methods
Preparation of Dentin Disks
For this study, we used extracted human molars from the department's collection. All patients had given informed consent. The teeth were stored in tap water and cut with a circular saw (Leica, Wetzlar, Germany) under water cooling. The disks (200-μm thick) were prepared from below the cementoenamel junction. Three adjacent disks were cut from the same tooth and randomly allocated to the treatment group. For each experiment, we used at least 3 different teeth. To check if the groups were well-balanced, the disk diameter was measured using a MarCal caliper (Mahr GmbH, Esslingen, Germany), and the dentin surface was calculated using ImageJ software (National Institutes of Health, Bethesda, MD).
The following solutions (300 μL) were used as conditioning agents: 10% CA (pH = 2, 476 mol/L), 17% Na2 EDTA (adjusted to a pH of 8), or phosphate-buffered saline (PBS; Gibco, Paisley, UK). The conditioning treatment was 10 minutes.
Slot Blot Protein Immunoassay for TGF-β1 Release
The protein concentration after conditioning was determined using a slot blot technique. The probes were directly bound to a nitrocellulose membrane using a Bio-Dot apparatus (Bio Rad, Hercules, CA). As a standard, recombinant human TGF-β1 (Sigma-Aldrich, St Louis, MO) in concentrations from 0.2–12.5 ng/μL was loaded. The membrane was blocked with 3% gelatin (Sigma-Aldrich) in Tris-buffered saline. Samples were incubated with anti–TGF-β1 produced in mouse (Sigma-Aldrich) and horse/antimouse immunoglobulin G and horseradish peroxidase–linked antibody (Cell Signaling Technology, Danvers, MA). Intensities of bands were analyzed using Image Lab software (BioRad). A precision balance (PJ3000; Mettler Toledo, Nänikon, Switzerland) was used to weigh disks before the experiment.
Spatial Distribution of TGF-β1 on CA Pretreated Dentin
In order to investigate the expression of TGF-β1 on the pretreated surface, an immunofluorescence study was performed. CA pretreated disks were washed and blocked with 0.2% gelatin and 0.5% bovine serum albumin (Albumin Fraction V; AppliChem, Darmstadt, Germany) in PBS. Samples were incubated with chicken anti–TGF-β1 (Invitrogen, Carlsbad, CA) 1:30 and goat antichicken Alexa Fluor 568 (Invitrogen) 1:200. Samples were observed with confocal laser scanning microscopy (SP8 Inverse STED 3X, Leica) and excited with a white light laser (561 nm). Images were acquired with a 63×/1.4 numerical aperture oil objective. A hybrid detector (564–651 nm) was used for fluorescence and a photomultiplier tube for transmission detection. For negative controls, disks were incubated only with secondary antibody or both antibodies were withdrawn.
Human bone marrow–derived mesenchymal stem cells (PCS-500-012; American Type Culture Collection, Manassas, VA) were cultured in MesenPro RS with 1% Pen Strep and 1% GlutaMAX-I (Gibco). Cells at passages 3–6 at 80% confluence were used in all experiments.
To test cell migration, polycarbonate Transwell inserts with a pore size of 8 μm (Corning, Corning, NY) were used. Cells were starved overnight and added to inserts. Dentin disks were disinfected with 5% NaOCl and randomly allocated to the groups. Disks were placed on the bottom of the well together with 650 μL starvation medium. Inserts preseeded with cells were immerged into the wells and maintained in the incubator for 24 hours. After fixation in 3.7% formaldehyde, cells were permeabilized by 100% methanol and stained with crystal violet (Acros Organics, Geel, Belgium). The inside of each insert was cleaned using cotton swabs. Cells, migrated from the inside of the insert to the other side of the membrane facing the dentin slices, were visualized using an inverted microscope (CKX53; Olympus, Tokyo, Japan). Ten fields were randomly selected from each insert, and cells were counted therein.
Dentin disks were disinfected with 5% NaOCl and randomly divided into conditioning groups. After washing, 6 disks from the same group were placed in a single layer in a petri dish. On the top of every disk, cells at a density of 5.5 × 105 cells/mL were seeded. The petri dish was kept in the incubator overnight. The next day disks were placed in a new petri dish, washed, detached, and counted using an EVE Automatic Cell Counter (NanoEnTek, Seoul, Korea).
Everything before the incubation step was performed in the same manner as in the cell attachment assay described previously. The petri dish was kept in the incubator for 48 hours. After trypsinization, cells were counted.
All experiments were repeated at least 3 times independently. Data were statistically analyzed using Prism 7 software (GraphPad, La Jolla, CA). Results are expressed as mean values ± standard deviation and were compared by 1-way analysis of variance and the Student t test. Bonferroni adjustment was applied for multiple comparisons. The level of statistical significance was set at P < .05.
Slot Blot Protein Immunoassay for TGF-β1 Release
The discrete band was visible when a standard at a concentration of 0.4 ng/μL was loaded, which corresponded to 117 ng recombinant TGF-β1, whereas saturation occurred at 3750 ng (Fig. 1A). PBS treatment showed no staining on the membrane, neither did pure CA or EDTA. A significantly (P < .05) higher release of the target protein was evident in the CA group (Fig. 1B), with 382 ± 30 ng TGF-β1 per dentin disk, compared with 66 ± 3 ng TGF-β1 released with EDTA.
Spatial Distribution of TGF-β1 on CA Pretreated Dentin
A high amount of TGF-β1 was exposed and detectable in the CA group (Fig. 2A–C). The exposed protein formed roundish structures. By overlaying with a bright-field view, we perceived that a high exposure of the protein was manifested in peritubular dentin. On the dentin disks without antibody incubation, the protein was not detected (Fig. 2D–F).
The modified Boyden chamber assay showed that both CA and EDTA conditioning induced migration of stem cells toward pretreated dentin (Fig. 3A–D) but CA treatment significantly more than EDTA (P < .05) or PBS (P < .01).
Cell Attachment and Survival
The cell-dentin interaction was investigated using cell attachment (Fig. 4A) and cell survival assays (Fig. 4B). The attachment was significantly more promoted by CA than EDTA (P < .05) or PBS (P < .01). Moreover, cell survival was also significantly (P < .05) higher after CA treatment compared with EDTA or PBS.
The humoral defense against caries and recruitment of pluripotent cells to replace damaged odontoblasts occurs in an acidic environment. This study showed that 10% CA released and exposed more TGF-β1 from human root dentin than 17% EDTA did. Consequently, more pluripotent cells were attracted and adhered to such biomimetically conditioned dentin surfaces.
The current data were obtained in the laboratory using human specimens. Consequently, no clinical conclusions should be drawn. However, there was a similarity to dental clinics. For disinfection, we used NaOCl, which is the most common disinfectant in regenerative endodontics (
), who found a higher density of apparently exposed TGF-β1 on specimens treated with EDTA compared with CA. However, they also observed better smear layer removal with EDTA compared with CA, which is in contrast to the literature that specifically dealt with this topic. Because dentin is a rather heterogeneous tissue, results obtained by mere 2-dimensional imaging should be interpreted with care (
), but no significant level of TGF-β1 was detected for CA treatment. This is in contrast to our results clearly showing a significant TGF-β1 release upon CA treatment and the presence of TGF-β1 protein around tubular openings. Moreover, our results of spatial distribution of TGF-β1 on dentin slices indicate that TGF-β1 is exposed in peritubular dentin, which is known to contain an organic matrix, including bioactive macromolecules (
). Indeed, protonation causes changes on the amino acids and dissociation of binding proteins that may compete for the same binding spots with ELISA antibodies and block the signal from the targeted protein (
). These reports are in line with our findings (Figs. 3 and 4); the increased roughness upon CA treatment could be 1 of the reasons for the superiority of CA conditioning in terms of stem cell attachment and survival. Moreover, acidification converts latent TGF-β to its active form (
). However, other growth may contribute to the observed regenerative potential. It has been shown that CA treatment was more potent to extract bone morphogenetic protein 2 and vascular endothelial growth factor from dentin slices than EDTA treatment (
). This was confirmed using human stem cells in this investigation. However, in our study, CA induced a 3-fold higher cell number than EDTA to migrate toward the conditioned dentin.
In conclusion, this study provided direct evidence of the release and exposure of TGF-β1 from and on chemically conditioned human dentin. CA treatment appeared superior to the gold standard EDTA in terms of TGF-β1 release as well as stem cell recruitment, attachment, and survival. Therefore, CA conditioning, which mimics natural conditions under caries attack, may improve pulp regeneration.
We thank Andres Kaech for scientific comments; Gery Barmettler for technical support with sample preparation of the immunofluorescence experiments; Claudia Cucuzza and Beatrice Sener for swift laboratory work; and Tse-Hsiang Chen, Nupur Khera, Indranil Bhattacharya, Kathrin Schumann, and Ana Perez for their helpful advice.
Supported by a Swiss Government Excellence Scholarship (A.I.).
The authors deny any conflicts of interest related to this study.
The role of the blood clot in endodontic therapy. An experimental histologic study.