Acellular biomaterial strategies for endodontic regeneration

Ignacio Medina-Fernandez and Adam D. Celiz *
Department of Bioengineering, Imperial College, London, SW7 2AZ, UK. E-mail:

Received 14th October 2018 , Accepted 10th December 2018

First published on 11th December 2018

Dental decay is treated by removing infected dental tissues such as dentine and restoring the tooth with a material. However, the vast majority of these materials have been designed to be mechanically robust and bioinert, whereas the potential regenerative properties of a biomaterial have not been considered. In endodontics for example, materials are used to seal the pulp cavity to avoid bacterial colonisation of the tooth and prevent further infection. While these treatments are effective in the short term, many of these materials have not been designed to interface with the pulp tissue in a biocompatible manner and are often cytotoxic. This can lead to less favourable long-term outcomes such as devitalisation of the tooth via root-canal therapy or extraction of the tooth. Clinical outcomes could be improved if regenerative approaches were followed whereby the biology of the tooth is engineered for repair and regeneration often with the support of a biomaterial. Within these, acellular or cell homing approaches are particularly interesting, as some regulatory hurdles associated with cellular therapies could be circumvented which may aid their clinical translation. In this review, we highlight progress in regenerative dentistry and focus on exciting developments using acellular biomaterials for regenerating dental tissues.

image file: c8bm01296b-p1.tif

Ignacio Medina-Fernandez

Ignacio is currently a PhD student at Imperial College London, working in the Celiz Lab to develop regenerative scaffolds modified with bioactive ceramics for dentine–pulp regeneration. He joined Imperial College to study an MSc in Biomedical Engineering with a specialisation in Biomaterials and Tissue Engineering, where he worked in the Georgiou Polymer Group in the synthesis and characterisation of polymers for miRNA delivery. He has a bachelor's degree in Biomedical Engineering in the Polytechnic University of Valencia in Spain.

image file: c8bm01296b-p2.tif

Adam D. Celiz

Adam graduated with a first-class bachelor's degree from the University of Brighton in Pharmaceutical and Chemical Sciences in 2006 and gained a PhD in Chemistry from the University of Cambridge in the Melville Laboratory for Polymer Synthesis in 2010.

Following this, Adam performed postdoctoral research at the University of Nottingham and the Wyss Institute for Biologically Inspired Engineering at Harvard University with a Marie Curie International Outgoing fellowship. Adam's research in biomaterials has led to an Early Career Researcher Award from the American Vacuum Society's Biomaterials Interfaces Division and the Larry Hench Young Investigators Prize from the UK Society for Biomaterials. In 2017, Adam joined the Department of Bioengineering at Imperial College London as Lecturer (Assistant Professor) to develop biomaterials for tissue repair and regeneration.


Tooth decay, according to the World Health Organisation, is the most prevalent non-communicable disease, affecting half of the worldwide population.1 It has a significantly detrimental effect on quality of life and great economic impact; e.g. tooth extractions for children cost the UK National Health Service >50 million pounds between 2014–2015.2 When tooth extraction can be avoided, an alternative treatment is the removal of the decayed tissues and substitution with a biomaterial. These treatments, depending on the severity of the injury, range from pulp capping, in which only dentine and enamel are removed, to pulpotomies and pulpectomies, in which partial and complete removal of the pulp tissue is performed, respectively.

A great variety of materials of different nature (polymers, ceramics, alloys…) have been used for these therapies. If they are classified according to their interaction with the surrounding tissues, two kinds may be distinguished: bioinert and bioactive materials. Bioinert materials are those which are designed to interact with the surrounding tissues as little as possible, while bioactive materials are designed to integrate and even regenerate these tissues. Common bioinert materials used in dentistry include amalgam, zinc oxide eugenol, formocresol, ferric sulphate, composite resins, gutta percha and glass ionomer cements.

The search and development of inert materials for endodontics is very active, but the goal of this research is to replace tissues instead of inducing repair and regeneration. They are, therefore, beyond the scope of this review. On the other hand, certain bioactive materials currently used in endodontics have been proved to promote tissue repair after endodontic procedures. The extensive use of these materials can provide some insight for the design of new biomaterials for dental regeneration.

Bioactive materials in endodontics

Bioactive materials used in endodontics have a long history, with some dating back over a century e.g. calcium hydroxide (CaOH – a full list of abbreviations are found in Table 1). They are primarily restorative materials used in contact with the pulp and are commonly employed in vital pulp therapies. These materials are usually cements that can form a paste with water, which is cured in situ after implantation to seal cavities and impede the infiltration of bacteria. A few common and representative examples are CaOH, Mineral Trioxide Aggregate (MTA) and Biodentine™ (Septodont). Of these, CaOH is the oldest and was introduced in endodontics in 1920. It consists of a white, alkaline powder with very low aqueous solubility.3 It is antimicrobial, due to the creation of a very alkaline environment, and induces the formation of hard tissue through localised necrosis of surrounding tissue and the release of Ca2+ ions.4 Although there are now alternatives, CaOH is one of the cheapest and easiest materials to handle and thus it is still frequently used.
Table 1 Abbreviations
AFM Atomic force microscopy
ALP Alkaline phosphatase
APC Autologous plasma concentrate
ARS Alizarin-red staining
Atgx Autophagy-related protein x
BDNF Brain-derived neurotrophic factor
bFGF Basic fibroblast growth factor
BMP Bone morphogenetic protein
(B)MSC (Bone) mesenchymal stem cell
BSP Bone sialoprotein
CAM Chick chorioallantoic membrane
CD Circular dichroism
CNC Cellulose nanocrystals
COL1A1 Collagen type I, alpha 1 (gene)
COL-III Collagen type III (gene)
(d)ECM (Dentine) extracellular matrix
DMOG Dimethyloxalylglycine
DMP-1 Dentine matrix acidic phosphoprotein 1
DSPP Dentine sialophosphoprotein
FBS Foetal bovine serum
FDA US Food and drugs administration
FOXQ1 Forkhead box Q1 (gene)
FS Fibroin silk
FTIR Fourier transform infrared
GF Growth factor
HA Hyaluronic acid
HE Hematoxylin and eosin
HUVEC Human umbilical vein endothelial cell
LC3-II Microtubule-associated proteins 1A/1B light chain 3B type II
MAPK Mitogen-activated protein kinase
MEM Minimum essential medium
MNP Magnetite nanoparticles
MTA Mineral trioxide aggregate
MTS 3-(4,5-Dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium
NF-κB Nuclear factor κ-light-chain-enhancer of activated B cells
NFL Neurofilament light chain polypeptide
NGF Nerve growth factor
OCN Osteocalcin
OPN Osteopontin
PCL(F) (Nanofibrous) poly-ε-caprolactone
PDGF Platelet-derived growth factor
PDLSC Periodontal ligament stem cell
PECAM-1 Platelet endothelial cell adhesion molecule
PDL Periodontal ligament
PL Platelet lysate
PRF Platelet rich fibrin
PRP Platelet rich plasma
RGD Arginylglycylaspartic acid
RT-PCR Reverse transcription polymerase chain reaction
RUNX3 Runt-related transcription factor 3
SCF Stem cell factor
SDF1 Stromal-derived factor 1
SEM Scanning electron imicroscopy
SOST Sclerostin (gene)
TDM Treated dentine matrix
TGF-β Transforming growth factor beta
TSS Tooth fragment/silk fibroin scaffold complex
VEGF Vascular endothelial growth factor
SCN1 Stanniocalcin (gene)

MTA is composed of several phases, including dicalcium and tricalcium silicates, bismuth oxide (to increase radiopacity), sulphates and aluminates. The material is named after the trioxide aggregate found in its composition made up of calcium, aluminium and selenium.5 It was adapted from Portland Cement, to create a biocompatible material that shared the cement's ability to set in wet environments such as the one in the tooth.6,7 Its biocompatibility, bioactivity and sealing ability explain its prevalence as one of the most common materials in endodontics for apical plugs and vital pulp therapy. There are some drawbacks to the use of this material, namely discoloration of the tooth, difficulty of handling, high cost and long setting time (ca. 2–3 hours),8 but many studies report a greater sealing ability which makes MTA treatments more long-lasting and cost-effective than CaOH.9–11 That said, it is not clear whether MTA is a robust bioactive material mostly due to the lack of reports of long-term outcomes.8,12

Different formulations of MTAs and other calcium silicate-based materials have been designed and commercialised, claiming shorter setting times and, essentially, more effective treatments. Among these, Biodentine™ is generally regarded as one of the most promising bioactive materials for endodontic therapy. Based on tricalcium silicate and thus not so different in composition from MTA, several studies have shown it to be easier to handle and to have a shorter setting time than MTA.13–15 Since its commercialisation in 2009, Biodentine has been the focus of many studies in which it has proved to be a valid alternative to MTA.14–17 Again, lack of standardisation in the methodology of studies and reports of long-term outcomes hinder any definitive judgements on the superiority claims of Biodentine over other bioactive ceramics used in endodontics.

The use of these materials has allowed the development of vital pulp therapies, in which part of the pulp is removed via a pulpotomy, but the vitality of the tooth is maintained. Vital pulp therapies are of great importance, particularly in permanent immature teeth where pulp vitality is key for the apices of the teeth to reach maturity. These therapies, however, cannot be regarded as regenerative, as the original tissues of the tooth are not regrown. Therefore, there has been a great effort in the search for regenerative biomaterials that are able to restore dental tissues after injury or infection.

To date, studies on the regeneration of dental tissues using biomaterials have focused on cell transplantation strategies. Nevertheless, searching in several clinical trial databases (using the terms “pulp”, “dentin”, “dental” and “regeneration”), only 5 trials for cell transplantation techniques were found against 12 for acellular techniques.18–20 This highlights the great interest there is in cell homing strategies, as it is generally accepted that there are lower technological and regulatory barriers for clinical translation. Due to the potential of these therapies to revolutionise endodontics, this review focuses on acellular approaches for the regeneration of dental tissues.

Design parameters for acellular scaffolds

Acellular approaches for regeneration in endodontics have been explored in the clinic since the middle of the last century. The great majority of clinical trials on dental regeneration studied the efficacy of different variants of a traditional acellular therapeutic approach: apical revascularisation. In this procedure, a pulpectomy (full removal of the pulp) is followed by disinfection, the application of an antibiotic paste and evoking bleeding from the periapical tissues. The immediate result is the formation of fibrous connective tissue in the root canal.21 Later studies involving several animal models and few clinical cases found ingrowth of mineralised tissues, mainly cementum and bone-like tissue.22,23 Nevertheless, there is a lack of consistent results regarding mineralisation and reports of full regeneration of the dentine–pulp complex have so far been absent.22,24 A recent meta-analysis by He et al. on 22 studies on apical revascularisation for necrotic immature permanent teeth concluded that the procedure did promote development of the root, but results were not consistent regarding root widening or apical closure.25 Another clinical approach related to acellular regeneration is the use of autologous platelet concentrates (APCs), mainly platelet rich plasma (PRP) and platelet rich fibrin (PRF). It essentially consists in the preparation of concentrates from plasma of the patients and the implantation after apical revascularisation. Its main advantage against evoked bleeding alone is the high concentration of growth factors (GFs) and cytokines in APCs.26 It has been reported to promote tissue repair to a greater extent than apical revascularisation, but with similar limitations, namely lack of consistent results and methodology, interpatient variability and unknown long-term outcomes.22,26,27

The non-satisfactory results obtained with traditional acellular regenerative endodontic therapies highlight how cell recruitment and GFs are insufficient to achieve regeneration: the use of an appropriate biomaterial is essential. Biomaterial scaffolds not only serve as a support for the growing tissue, but they may also act as drivers of regeneration, especially in acellular approaches. A biomaterial for cell-free tissue regeneration must have the following basic properties:

- Biodegradability at a similar rate as the growth of the tissue.

- Biocompatibility, as the tissue must not be damaged by the material or by its degradation products.

- Interconnected macroporosity, so revascularisation and innervation of the tissue occurs.

A biomaterial for regeneration should also be bioactive, promoting cell recruitment, attachment, proliferation and differentiation. These features may not be intrinsic to the material itself and can be incorporated via certain moieties. These moieties may either modify the properties of the material (e.g. RGD peptide modification, which improves cell attachment) or may be released (e.g. signalling molecules like GFs and cytokines). For the latter scenario, spatiotemporal control of delivery may be advantageous, depending on the application. Moreover, the scaffold should utilise the endogenous bioactive molecules, for instance boosting the production and/or release of cytokines and GFs in the host tissue. A few examples of regenerative strategies are illustrated in Fig. 1.

image file: c8bm01296b-f1.tif
Fig. 1 Different regenerative strategies for dental tissue regeneration via cell homing. An apical revascularisation via evoked bleeding (A) may be combined with the implantation of a bioactive scaffold for regeneration of the tissues in the pulp chamber and root canals (B). Scaffolds must promote cell recruitment, proliferation and differentiation, as well as encouraging angiogenesis and cell attachment and for that several moieties may be included (C). Depending on the severity of the lesion, less invasive methods may be performed (D). The insertion of a cut-to-size scaffold or the injection of a biomaterial with self-assembly or shear-thinning properties are also possible approaches for dentine–pulp regeneration (E). In F, the typical, physiological distribution in palisade of the odontoblasts is shown. The newly-differentiated odontoblasts must produce a dentine bridge that permanently seals the pulp chamber of the tooth. Figure adapted from image obtained from ©Elena Baryshkina –

In our search of clinical trials in dental regeneration, only one of the twelve acellular studies involved the use of a biomaterial scaffold (other than commercial collagen sponges).28 This draws attention to the necessity of researching new biomaterial strategies, given the importance of scaffolds in acellular tissue regeneration.

Scaffolds in acellular dental tissue regeneration

Due to the shortcomings of the materials described above, there has been great effort recently to design new materials for dentine–pulp regeneration. Most of this work has been focused on the search for cytokines and GFs that act as signalling molecules for mesenchymal stem cells (MSC) recruitment.

In one of the first instances to assess the endogenous regenerative potential of the dentine–pulp complex, Kim et al. reported the growth of connective tissue on the surface of dentine of endodontically treated human teeth implanted in the dorsum of mice.29 No additional cells were transplanted into the teeth, which suggests that new tissue was formed via the recruitment and proliferation of MSCs, presumably from bone marrow niches. Regeneration was achieved thanks to the injection of a collagen gel solution loaded with a cocktail of cytokines, namely bFGF, PDGF or VEGF, for recruitment and proliferation, along with a basal set of NGF and BMP7 for neural growth and odontogenic differentiation, respectively.

Since then, many other studies have reported cell migration and differentiation of BMSCs and DPSCs, both in vitro and in vivo, in response to different cytokines, such as SCF,30,31 SDF132–34 and BDNF,35 to name a few. Most of these studies focused on studying the potential of the signalling molecules by themselves or in combination with others, using unmodified commercial biomaterials, such as gels and sponges based on collagen. Therefore, these will not be discussed here, but for a comprehensive review dealing with the role of cytokines for cell homing in dental pulp regeneration, the reader is directed to the work by Eramo et al.36 In this review, we will address research studies that focus on the use of materials to create scaffolds and the assessment of its capabilities in acellular dental tissue regeneration. These studies include the quantification of the ability of the material to induce proliferation, differentiation and migration or recruitment of progenitor cells, the latter being crucial in cell homing.

In vitro assays

Six in vitro studies in dentine–pulp regeneration reporting chemotactic properties for biomaterial scaffolds have been found in the literature (Table 2).
Table 2 In vitro studies
Authors (year) Material of the scaffold Other bioactive moieties Cells Methods Results
Ji et al. (2015)56 dECM (named as TDM in the study) Canine PRF Canine BMSCs and PDLSCs Cell migration with PRF extracts via 8 μm pore size Transwell assay. Migration of BMSCs and PDLSCs to PRF-conditioned medium matched that to 10% FBS medium (basal medium α-MEM).
Odontogenic effects of PRF and TDM using RT-PCR. Greatly increased expression of BSP and OPN (p < 0.005) and OCN only for PDLSC (p < 0.05) when influenced by TDM and PRF.
Yun et al. (2015)44 Macroporous PCL Magnetite nanoparticles (MNP) at 5 and 10 wt% Human DPSCs Cell viability measured via MTS. No effects on cell viability were observed. PCL scaffolds increased migration and adhesion rate (measured up to 12 hours) but had no effect on odontogenesis (measured up to 14 days). Migration, odontogenesis, mineralisation and adhesion were all improved by MNP addition.
Yun et al. (2016)50 Nanofibrous PCL MNPs at 10 and 20 wt% Human DPSCs Migration induction measured by the speed of a cell monolayer to reform after mechanical disruption. PCL nanofibrous scaffolds with no MNP addition promoted higher ALP activity, odontogenesis and mineralisation (measured up to 14 d) vascularisation (after 12 h) and migration (after 24 h) than control (p < 0.05). Addition of MNP improved these (no significant difference found between 10 wt% and 20 wt% groups).
Optical density of fixed cultures gave measure of adhesion speed.
Odontogenic differentiation assessed by ALP activity and RT-PCR and mineralisation via ARS.
Angiogenic induction measured via RT-PCR and counting of tube formation in culture (only 2016).
Almeida et al. (2018)53 Photocroslinked HA hydrogel Platelet lysate (100% v/v) Human DPSCs Viability was measured through ABA (metabolic activity) and dsDNA quantification (proliferation). More stable metabolic activity of HA/PL (after 21 d) and higher dsDNA amounts (7, 14 and 21 d) indicated higher cell viability for PL-loaded gels.
Mineralisation measured via ALPase activity and calcium deposition. ALPase activity was higher at all timepoints for test group and calcium deposition at 21 d.
Odontogenesis measured via RT-PCR. After 21 d, ALP and COL1A1 exhibited higher expression levels, indicating odontogenesis.
Migration assessed by curing gel on top of cell monolayer and culture for up to 14 d. Cells infiltrated the first 20 μm of both the test and control hydrogels.
Test group was PL-loaded HA. Control was plain HA.
Silva et al. (2018)51 Injectable HA hydrogel Platelet lysate (50% v/v), Celluose Nanocrystals (CNC) Human DPSCs and HUVEC HDPSC viability in hydrogel encapsulated cells measured via ABA and dsDNA quantification. PL increased stability for metabolic activity and greatly induced proliferation, while CNC appeared to decrease it up to 7 d (not at 14).
HDPSC adhesion and morphology studied via confocal microscopy. Cells cultured in PL-loaded hydrogels exhibited more elongated shapes. Well-developed 3D networks were observed for loaded and unloaded gels.
Mineralisation via ALPase activity. Sprout length greatly increased with PL loading and porosity (greater at 0.25% CNC).
Migration and angiogenesis assessed via analysis of sprouting of encapsulated hDPSCs and HUVEC. Only hDPSCs were involved in sprouting.
Nguyen et al. (2018)55 Self-assembling Dentonin peptide hydrogel Mouse 3T3 fibroblasts and hDPSCs FTIR and CD spectroscopies used to determine secondary structure. β-Sheet structure in agreement with predicted self-assembly scenario.
Oscillatory rheometry used to determine viscoelastic properties and storage modulus Hydrogel possesses thear-shinning abilities.
SEM and AFM used for assessing ultrascrutcture. Nanofibrous structure achieved and recovered after liquefaction (post-injection).
3T3 fibroblast viability via LIVE/DEAD kit. Cytocompatible on flibroblasts.
hDPSC proliferation via CCK8 assay. Increased proliferation of hDPSCs against only media control (p < 0.01).
Calcium phosphate deposition induction on hDPSCs via ARS Hydrogel encouraged calcium deposition of hDPSCs against absence in only media control.

Ji et al. described the first instance of an in vitro assessment of cell migration for dental tissue regeneration with a biomaterial-based strategy. Canine decellularised dentine ECM or treated dentine matrix (TDM) was used to create an in situ model for acellular root regeneration. ECM contains a great variety of bioactive molecules that provide an information-rich network which promotes and directs tissue growth. Thanks to decellularisation techniques, these structures have been widely used as bioactive materials to induce regeneration in various tissues.37,38 The use of TDM as a scaffold is promising due to the presence in the dentine of non-collagenous proteins and GFs that are believed to mediate repair and regeneration upon damage of the dental tissues. A recent in vivo study in mice by Neves et al. demonstrated, however, that odontogenic dentine-sequestered BMPs and TGF-β had a more minor role in regeneration than previously believed, as they appeared to modify tertiary dentine organisation without affecting its production.39 Despite this, the dentine matrix serves as a reservoir for many other bioactive molecules that may promote cell recruitment, vascularisation, innervation and odontogenic differentiation.40 In fact, the degradation of pulpal and dentinal ECM appears to be a key player in the reparative process following disease.41 The ability of TDM scaffolds for promoting PDL, cementum and pulp – like tissue growth had been previously studied by the same group in cell transplantation approaches.42,43 In this study, TDMs were combined with PRF so the scaffolds released a cocktail of GFs and bioactive molecules to promote recruitment, dental tissue growth and vascularisation. The capabilities of PRF-conditioned media on canine BMSCs and PDLSCs migration and proliferation were found to match those of 10% FBS media in in vitro assays. Both types of cells were also found to have undergone odontogenic differentiation in a transwell assay. PRF and TDM were laid on the upper compartment while cells were cultured on the lower unit for 7 days. BSP and OPN gene expression was greatly upregulated (p < 0.005) and so was OCN for PDLSCs. COL-III expression was not affected compared to control. The capacity of the scaffolds for cell recruitment and odontogenesis appeared promising, and an in vivo assay followed as described in a following section.

Yun et al. (2015) reported the use of a poly-ε-caprolactone (PCL) scaffold with embedded magnetite nanoparticles (MNPs) to study migration and odontogenesis of hDPSC.44 PCL is an FDA-approved polyester with a very high degradation time due to its hydrophobicity. This long degradation time, along with the fact that it is not naturally cell adhesive, can be liabilities in tissue engineering. However, these issues can be overcome by modification of the material, and its mechanical strength and non-immunogenicity have allowed it to be a widely used material in tissue regeneration, particularly in dental regeneration.45 This macroporous PCL scaffold was prepared using 200–500 μm NaCl particles as porogens and cell migration, viability, adhesion, and odontogenic differentiation were assessed. All groups were grown on odontogenic differentiation media, except a negative control, and the three scaffolds had varying amounts of embedded MNPs (0 wt%, 5 wt% and 10 wt%). Paramagnetic MNPs have been previously reported to induce osteogenesis,46–49 and the main goal of this study was to study their influence and mechanism of action in DPSCs. Migration, measured as the ability for a cell monolayer to reform after 12 hours, was significantly increased when MNP was added, being significantly higher than control for all PCL groups and highest at 10 wt% (statistical significance was determined at p < 0.05). Optical density, measured at different timepoints up to 12 hours, suggested a more rapid adhesion for the MNP/PCL scaffolds. MNP concentration was also found to be relevant for odontogenic differentiation with 5 wt% and 10 wt% MNP scaffolds promoting higher ALP activity compared to controls. Expression of odontogenic markers (OPN, OCN, DSPP, DMP-1) also appeared to be dependent on MNP concentration, while PCL alone did not seem to influence these values. In a follow-up study, Yun et al. (2016) prepared a fibrous scaffold by dissolving PCL (10 wt%) and MNP (10 wt% and 20 wt%) in dichloromethane and ethanol followed by electrospinning.50 The nanofibrous structure was the likely responsible in the improvement of cell viability against controls in this study, with MNP presence inducing significant growth after 14 days. ALP activity, as well as most of the odontogenic markers measured, evolved in a similar way as in the previous study. The expression of these markers increased significantly up to concentrations of 10 wt% after 7 and 14 days. Additionally, angiogenesis and cell migration capabilities were assessed. HDPSCs were seeded on the materials for three days and levels of VEGF and angiopoetin-1 were measured to find an increase of these factors proportional to MNP concentration. The media from these assays was used to induce angiogenesis and migration on HUVECs for 12 hours. Capillary-like structures were formed more often in PCL groups, peaking for 10 wt% MNP. Migration rate had a similar behaviour, with cell monolayer recovering between approximately 1.3 (0 wt% MNP) and 1.7 (10 wt% MNP and 20 wt% MNP) times quicker than the control group at 24 hours. The authors hypothesised and proved that a higher expression of integrins α1, α2, β1 and β3 was related with the improved regenerative capabilities of the magnetic scaffolds.44 This upregulation and improved odontogenic properties were found to be mediated through MAPK, NF-κB and Wnt signalling pathways.44,50

In a recent study, Silva et al. prepared an injectable Hyaluronic acid (HA) hydrogel, a naturally occurring polysaccharide widely used as a biomaterial in tissue engineering.52 HA is a glycosaminoglycan commonly found in the ECM of animal tissues, which explains its adequate degradation profile and biocompatibility.52 Injectable materials are a very promising tool due to their ease of application and subsequent potential for translation into the clinic. This bionanocomposite incorporated HA with cellulose nanocrystals (CNCs), to increase hydrophobicity and degradation time and platelet lysate (PL) for GFs release. Gels were formed in situ through the simultaneous injection of two solutions containing modified HA at 2% w/v (with either hydrazide or aldehyde pendant groups) that cross-link with one another upon mixing. PL and amine-modified CNC were diluted in these solutions prior to crosslinking. A thorough material characterisation of gels with different concentrations of CNC (up to 0.05 wt%) and loaded with 50% (v/v) PL was performed. Essentially, increased CNC concentrations were associated with higher stiffness and longer degradation and drug release times. PL was related to a decrease in pore size and stiffness. DNA concentration metabolic activity assays after up to 14 days of culture with hDPSCs indicated that PL caused a great increase in cell proliferation (p < 0.001), while CNC-loading appeared to have a mild toxic effect. For the in vitro assessment of migration and angiogenesis, hDPSC or cocultured hDPSC and HUVEC pellets were placed on the surface of transwells and hydrogels were injected over them. The aim of the cocultures is to favour the formation of vascular-like structures. For both groups, cell migration was greatly enhanced when PL was present compared to CNC-only gels (p < 0.001), peaking at any of the three days of culture for the 0.25% CNC/PL hydrogels. Interestingly, this peak coincides with the peak in pore size of the hydrogels. In the hDPSC/HUVEC cocultures, the sprouts were essentially formed by hDPSCs, while HUVECs appeared to promote the creation of tubular-like structures and enhance migration. An ex ovo CAM assay was carried out, using filer paper as a positive control. Four groups, 0% CNC and 0.25% CNC/PL with and without encapsulated hDPSCs were compared to control to assess chick tissue infiltration and vascularisation. Migration of endogenous cells was only achieved when encapsulated hDPSCs were delivered in the 0.25% CNC/PL hydrogel. This formulation, with and without hDPSCs also appeared to promote the creation of concentric vessels around it, presumably thanks to PL. The authors, however, did not provide any data regarding odontogenic differentiation. Furthermore, angiogenic properties were only proven when cells were transplanted with the hydrogel. Therefore, the composite needs to be modified and further assessed in regards of odontogenic effects and endogenous cell recruitment to be considered a completely valid approach for acellular dentine–pulp regeneration.

In similar studies, a photocrosslinkable HA/PL hydrogel system was developed which could be cured in situ to obtain versatile scaffolds to regenerate a variety of tissues. Almeida et al. modified HA with methacrylic anhydride, which was dissolved in either PBS (control) or PL at 1.5 wt% and cured by UV light exposure to create a hydrogel for dentine–pulp repair. After odontogenic induction, calcium deposits were most abundant for the HA/PL group (p < 0.05). Similar results, as in the previously described HA hydrogel, were obtained for cell proliferation and metabolic activity, with more stable metabolic activity and higher proliferation for PL-loaded gels at all time points (7, 14 and 21 days). PL also promoted the expression of odontogenic/osteogenic markers (ALPL, COL1A1, RUNX3) (p < 0.05 for all described assays). Although the focus of this study was on viability, proliferation and odontogenic/osteogenic effects over hDPSCs, chemotactic properties of the composites were assessed in vitro. Gels were cured directly over cell monolayers and after 14 days, cells invaded up to 20 μm into the hydrogel, regardless of the presence of PL. The authors argued that less dense networks could aid in promoting cell invasion, however, in a previous study, HA/PL gels of smaller porosity enabled human periodontal ligament fibroblasts (HPLF) migration up to 70 μm after 21 days from seeding on the material.54 A longer time on culture, different cell types and variations in methodologies can certainly account for this variation, but a characterisation of the microstructure and porosity of this material is essential to analyse cell recruitment abilities of the material. Babo et al. reported similar results regarding cytocompatibility and antimicrobial effects against Staphylococcus aureus in an agar well diffusion assay.54 PL was found responsible for this antimicrobial effect, as only cross-linked hydrogels in undiluted PL inhibited bacterial growth. These studies show the promising applications of these composites to promote recruitment and proliferation of endogenous progenitor cells. A more thorough characterisation regarding odontogenic differentiation properties should be carried out in the future, followed by in vivo assays to prove endogenous cell recruitment and mineralisation capabilities.

Among injectable materials, self-assembling peptide hydrogels stand out for their tailorability and bioactivity. Nguyen et al. developed a hydrogel bearing the dentonin peptide motif (TDLQERGDNDISPFSGDGQPFKD).55 Dentonin is a portion of Matrix Extracellular Phosphoglycoprotein (MEPE), and it has been proposed to be responsible for its bioactive nature (which favours recruitment and proliferation) due to the presence of both integrin-binding and glycosaminoglycan-binding motifs (RGD and SGDG, respectively).55,57,58 The shear-thinning abilities of the hydrogel made it injectable, with a stable elastic modulus of 400 Pa. Its nanofibrous structure, determined through AFM and SEM is purportedly recovered after shearing. The gel was shown to have good cytocompatibility for murine 3T3 fibroblasts and encouraged proliferation of hDPSCs, compared to only media control (p < 0.01). Despite the potential of dentonin to recruit progenitor cells thanks to its cell adhesion domains, no in vitro migration assays were carried out to prove cell homing potential in this study. The high hydrophilicity of the material made it unsuitable as a working scaffold for in vivo cell homing, due to a very rapid degradation after implantation in rats (3 days). Even with these limitations, its self-assembly and odontogenic properties make dentonin a promising material for regenerative therapies.

In vivo dentine–pulp regeneration

Studies for dentine–pulp regeneration in vivo are listed in Table 3. The use of a biomaterial scaffold for dentine–pulp regeneration via cell homing was first attempted by K. Kim et al. This in vivo study consisted in the implantation in mice of tooth-shaped 3D-printed scaffolds both ectopically in the dorsum (described later in this section) and orthotopically in the place of incisor teeth (described in next section). The preparation of the 3D-printed scaffolds consisted in the co-melting at 120 °C of PCL (80 wt%) and hydroxyapatite (20 wt%) followed by extrusion through a 27-gauge metal nozzle. This resulted in a structure with interconnected microchannels each with a diameter of 200 μm. Microchannels were infused with a neutralised type I collagen solution loaded with SDF1 and BMP7 to assess the effects of these cytokines in cell recruitment, mineralisation and vascularisation of the microchannels by comparing them to those of the unloaded scaffolds nine weeks after implantation. Histology showed cell infiltration and vascularisation in the microchannels, even for the scaffolds not loaded with SDF1 and BMP7. Loaded scaffolds, however, presented a significantly higher cell and vessel density (p < 0.01 and p < 0.05, respectively). Von Kossa staining indicated the presence of mineralised tissue in isolated perivascular areas in the ectopic implants. There was no presence of true dentine and pulp-like tissues and no reports on the phenotype of the recruited cells, so strictly speaking, this study cannot be considered an example of cell homing for dental tissue regeneration.
Table 3 Dentine–pulp regeneration
Authors (year) Material of the scaffold Other bioactive moieties Animal model Methods Results
K. Kim et al. (2010)59 Macroporous PCL Hydroxyapatite Sprague-Dawley rats Scaffold microchannels were injected with GF-loaded collagen and implanted either ectopically (dorsum). All scaffolds presented cell infiltration, tissue growth and angiogenesis.
SDF1 & BMP7 – loaded neutralised collagen type I Loaded and unloaded scaffolds were implanted. Cell and vessels more abundant in GF-loaded scaffolds.
Histological cuts were taken from the implant and mandible. Mineralisation present inside canals in areas of high cell and vessel density.
Lee et al. (2012)60 Nanofibrous PCL MTA Beagle dogs Exposed pulp from the test group was capped by implanting a PCLF scaffold in contact with the pulp that was covered with MTA. Dentine bridge thickness in test group was ∼4 times higher than for control group.
Pulps in control group were capped with only MTA. Recruited cells underwent odontoblastic differentiation and organised forming columns.
Histological cuts were taken from the teeth.
Yang et al. (2015a)61 Silk Fibroin (SF) scaffold bFGF Nude mice SF scaffolds were loaded with the GF. Formation of connective tissue for complexes loaded with cells.
Mineralisation only found on bFGF-loaded complexes. Staining with DSP indicated odontogenic differentiation.
Yang et al. (2015b)32 Silk Fibroin (SF) scaffold SDF-1α Nude mice, Beagle dogs Scaffolds were inserted in treated human tooth fragments and the complexes (TSS) were implanted on the dorsum on nude mice. Vascularisation more abundant and cells were more uniformly spread.
Groups: Tooth fragment, TSS, TSS + hDPSCs, GF-TSS + hDPSC, GF-TSS (only 2015b). Presence of recruited cells: 62% for bFGF-loaded complexes and 52% for unloaded ones
Histology analysis and immunostaining to assess dentine–pulp formation and autophagy (Atg5 and LC3). Formation of connective tissue and for complexes loaded with cells and/or SDF-1α. Mineralisation only in SDF-1α-loaded complexes.
(Only 2015b) western blot analysis for expression of markers for proliferation (cyclinD1, PCNA), angiogenesis (CD31, CD34), odontogenesis (DSP, DMP1, OSX and RUNX2), neurogenesis (neurofilament) and autophagy Markers for proliferation, angiogenesis, odontogenesis and neurogenesis upregulated in ectopic model when SDF-1α was loaded.
(Only 2015b) In situ revascularisation with acellular SDF-1α loaded scaffold (experimental group) or blood clot (control) in dogs. Revascularisation procedure resulted in connective tissue formation with higher cell density and vascularisation in experimental group. Thin mineralised layer over dentinal wall in experimental against thick in control.
Presence of Atg5 and LC3-positive cells especially for SDF-1α-loaded complexes.
Yoo et al. (2018)63 Nanofibrous PCL DMOG Nude mice Scaffolds were inserted into canine dental slices, and then implanted in the dorsum of mice. DMOG-loaded scaffolds presented a much higher number of cells and increase expression of all markers.
HDPSCs were seeded prior to implantation. Transplanted cells were differentiated to odontoblast-like cells and acquired a palisade structure.
RT-PCR of the transplanted cells was used to compare odontogenic, angiogenic and neurogenic markers between cells cultured in DMOG/PCL and only PCL. Recruited cells were abundant but did not express odontogenic markers.
Histological cuts were stained with DSP, Runx2 to find location of odontoblast-like cells. Origin (transplanted human or murine) was determined by staining as well.

PCL was also used by Lee et al. alongside MTA in the proposal of a regenerative pulp capping procedure.60 While the goal of this study was to establish the performance of an electrospun nanofibrous PCL material (PCLF) as a buffer between the pulp and MTA in a canine model (to decrease the cytotoxicity of the cement and improve its dentine formation ability), the scaffold was found to be able to recruit endogenous progenitor cells. In fact, these were reported to have differentiated into odontoblast-like cells, disposed in a columnar way, with processes embedded into de novo dentine-like tissue. This characteristic structure from the dentine–pulp complex was not found when only MTA was implanted. The thickness of the tertiary dentine bridge for the PCLF/MTA group was around four times that of the MTA one, which highlights the importance of achieving these structures. An in vitro assay was also performed to further study viability and odontogenic differentiation of murine foetal dental cells. Results showed significantly improved cell viability, calcium deposition and expression of odontogenic markers (namely ALP and DSPP, measured via RT-PCR) when cells were seeded on top of PCLF-covered MTA compared to MTA alone. The nanofibrous PCL scaffold successfully shielded cells, both in vitro and in vivo, from the highly caustic and cytotoxic environment created immediately after hydration of MTA. More importantly, it enabled infiltration by endogenous progenitor cells and promoted their differentiation into odontogenic types via the slow release of the degradation products of the MTA.

In 2015, Yang et al. used fibroin silk (FS) scaffolds in combination with bFGF61 and SDF1α32 to study dentine–pulp regeneration in a canine model. Scaffolds were prepared via solid–liquid phase separation of silk solutions through freezing. They were then lyophilised and treated with 70% ethanol for crystallisation. The scaffolds displayed an interconnected porosity of more than 80%, with pore sizes between 150 nm and 300 nm.62 Ectopic in vivo dentine–pulp regeneration was assessed following a similar process in both studies. The FS were inserted into the canal space of treated horizontal sections of radicular portions of human teeth. These complexes (TSS), were loaded with or without hDPSCs and GF which were implanted subcutaneously into the dorsum of immunocompromised nude mice. In the bFGF study, histology revealed tissue growth for the TSS + hDPSC and bFGF-TSS + hDPSC (not for TSS or the tooth section alone). DSP-positive cells and mineralised tissue deposits along dentinal walls was only observed in the bFGF-loaded complexes, which also induced the growth of a highly vascularised, pulp-like tissue. No acellular bFGF-loaded complexes were implanted, so the assessment of recruitment capabilities is not ideal, but immunostaining indicated that between 52% and 62% (for unloaded and bFGF-loaded complexes, respectively) of the cells present in the regenerated tissues were host-derived.

A similar scenario was presented for the SDF-1α-TSS complexes. Regenerated dentine along with highly vascularised pulp-like tissue were only found for the SDF-1α-TSS + hDPSC complexes. The expression of proliferation, odontogenic, angiogenic and neurogenic markers, measured via western blot analysis, was also upregulated for this group compared to the TSS + hDPSC group. Like bFGF, SDF-1α also appeared to be responsible for a higher recruitment of host mouse cells in the regenerated tissues (52% against 34% for the unloaded complexes). This study did assess regeneration on implanted acellular SDF-1α-TSS and histology showed the formation of a vascularised connective tissue with no signs of mineralisation or odontogenic differentiation. To further study the cell homing potential of the SDF-1α-SF scaffolds, apical revascularisation was performed on two beagle dogs. In the experimental group, the scaffolds were mixed with induced blood from the apex, while in the control bleeding was used to create a clot and no scaffold was implanted. Root canals were sealed using MTA for both groups. Histologic analysis showed the formation of mineralised tissue and connective tissue in both groups. Mineralised tissue was abundant in the control group, while only a thin layer formed over the dentinal walls of the experimental group. Furthermore, connective tissue in the implanted scaffold group had a higher cell density and a fibrous matrix more similar to that of healthy pulp, although there was no mention of nerve tissue growth. The authors hypothesised that SDF-1 could promote odontogenesis through the induction of autophagy in cells, so to assess this, expression of autophagy-related proteins Atg5–Atg12 conjugate, Beclin1 and LC3-II was quantified on the regenerated tissues. For both the ectopic and the in situ vascularisation assays, expression of these markers was higher when SDF-loaded scaffolds had been implanted. As it is argued by the authors, autophagy may have played a role in the formation of a highly vascularised tissue, but in this study, it did not appear to influence the formation of dentine-like tissue. Mineralisation and odontogenic markers were present in regenerated tissues when SDF was loaded into the implanted scaffolds, but odontoblast-like cells in their characteristic palisade configuration were not present. In any case, dentine–pulp-like tissue was regenerated in these studies, thanks to the combination of the porous and biocompatible FS scaffold with SDF-1α and bFGF (Table 3).

Yoo et al. used a similar approach with a PCLF scaffold, which was loaded with dimethyloxalylglycine (DMOG), transplanted into canine dentine slices and implanted ectopically in the dorsum of mice.63 Human DPSCs were seeded on the material prior to implantation, but immunostaining indicated that the scaffold promoted the migration of endogenous stem cells. Observation of the histological cuts suggested that transplanted cells had undergone odontoblastic differentiation, as they were highly polarised, acquiring a columnar configuration, typical of odontoblasts. Histological staining for antibodies against mineralisation markers DSP and Runx2 also indicated odontogenic differentiation. Recruited cells, on the other hand, were observed in high quantities surrounding the implant and were found to express high amounts of VEGF, endothelial marker PECAM-1 and neural marker NFL compared to cells surrounding PCLF-only scaffolds. This in vivo assay indicates good odontogenic properties for this scaffold, but it is not possible to assume that the reported endogenous cell recruitment would hold if an acellular approach was followed. Moreover, endogenous cell expression suggests that recruited cells did not infiltrate the material, and that they were oriented mostly towards vascularisation and neurogenesis, and not differentiated down an odontogenic lineage.

In vivo root regeneration

The root of the tooth is a complex soft–hard tissue interface that connects the tooth with the alveolar bone. The regeneration of the tissues associated with it, cementum and periodontal ligament (PDL), is currently being addressed in the clinic in root coverage procedures for gingival recession treatments via the use of tissue flaps. In fact, acellular regeneration through the implantation of acellular dermal matrix in these defects has been performed both academically64,65 and clinically (usually in combination with tissue flaps).66,67

Acellular whole root regeneration, however, has been far less studied. The study by Kim et al., consisting of a 3D-printed tooth-shaped PCL scaffold (with a content of 20 wt% hydroxyapatite), studied the orthotopic implantation under the same conditions as its ectopic model, described in the previous section.59 While the aim of this study was whole tooth regeneration, the lack of mineralisation and tissue growth in the upper areas of the implant rather make it an example of root regeneration. Bone tissue and periodontal-ligament-like tissue were observed histologically in the proximal end of the implants. Tissue growth, cell infiltration and angiogenesis were observed at the proximal end of the scaffolds. An assessment of the PDL-like tissue and its fibres was lacking, and cementum was not observed. Nevertheless, it serves as an early example of root tissue regeneration.

Finally, Ji et al.56 performed an in vivo assay to assess the chemotactic and regenerative capabilities of PRF/TDM scaffolds. These were implanted orthotopically in dogs, which were divided in four groups. Groups I and II received PRF-wrapped TDM scaffolds, with those for group II having been autoclaved prior to implantation. Group III received only a TDM scaffold while animals from group IV were implanted PRF/TDM scaffolds three months after extraction. Histological analysis with HE-staining of the mandibles 3 months after implantation revealed that de novo PDL-like tissue formed between TDM and bone for groups I and II. The surfaces of the scaffolds were clearly distinct from surrounding tissues and they were covered by a layer of cementum-like tissue. This was not observed in groups III and IV, for which TDMs were completely osseointegrated, suggesting that the complex root tissue interface was not being restored, as the scaffold was replaced by bone. Masson staining permitted to observe PDL fibres in groups I and II. A perpendicular disposition of the fibres with insertion in the cementum, as it is normally found in mature PDL, was only observed by tissue samples in group I (mean ± std = 47.3 ± 12.0°). In group II, fibres ran parallel to the surface of the root. The presence of cementoblast-like cells between the fibres of the material (for group I) also points to a more complete restoration of the root tissues. This difference is probably caused by the premature degradation of the TDM scaffolds from group II during autoclaving. The release of bioactive molecules by the TDM during degradation is in great part responsible for its capacity to induce stem cell migration and recruitment (Table 4).69

Table 4 Root regeneration
Authors (year) Material of the scaffold Other bioactive moieties Animal model Methods Results
K. Kim et al. (2010)59 Macroporous PCL Hydroxyapatite Sprague-Dawley rats Scaffold microchannels were injected with GF-loaded collagen and implanted either ectopically (dorsum). Implants presented no intra-root mineralisation but did promote PDL-like tissue growth and good integration with bone.
SDF1 & BMP7 – loaded neutralised collagen type I Loaded and unloaded scaffolds were implanted. No presence of cementum was reported.
Histological cuts were taken from the implant and mandible.
Ji et al. (2015)56 dECM (named as TDM in the study) Canine PRF Beagle dogs PRF was wrapped around TDM and implanted orthotopically. Formation of PDL-like and cementum-like tissue for groups I and II.
Group I: PRF/TDM; group II: PRF/autoclaved TDM; group III: TDM; group IV: PRF/TDM implanted 3 months after extraction. Scaffolds in groups III and IV fused with bone.
Histological analysis was carried out (HE and Masson staining) Quasi-perpendicular disposition against cementum of PDL fibres with cementoblast-like cells only for Group I.

Future outlooks

There is a growing interest in regeneration strategies that harness endogenous elements and structures to support the reparative properties and normal function of tissues.70 One way to achieve this in the tooth is to devise strategies to repair the enamel after decay and maintain a protective barrier which prevents decay and infection of the dentine–pulp complex. In a recent review, Elsharkawy et al. addressed the complex hierarchical structure of biomineralised tissues, with an emphasis on enamel and bone, and the different approaches designed to replicate and repair them.71 This approach was demonstrated by the creation of tunable, hierarchical polycrystalline structures exploiting elastin-like recombinamers and its self-assembly properties.72 The resulting material exhibited high hardness and stiffness and, when grown onto etched dental tissues, showed acid-resistant properties comparable to those of enamel and biointegration with dentine. Alternatively, harvesting the GFs sequestered in the dentine matrix can utilise the intrinsic regenerative potential of dental tissues. In current treatments (e.g. apical revascularisation) both irrigants used for disinfection and sealing materials such as MTA or CaOH have been proven to release these GFs.68 Yet, in research, the ability of the materials to release these key bioactive moieties is rarely addressed. The study of a materials ability for dentine dissolution is particularly relevant in acellular regeneration, given the necessity of stem cell recruitment.

The release of the sequestered molecules in the dentine can also have a great impact in regeneration due to the modulation of inflammation in the pulp.73 In fact, in endodontics, the adequate regulation of inflammation is one the most important requirements to achieve healing. The immunomodulatory properties of MSCs are key in this process and dental tissue-derived stem cells have been shown to possess immunosuppressing properties.74 For regeneration to occur, the initial acute inflammatory process, characterised by the release of pro-inflammatory cytokines, must progress into a healing stage. The shift to this stage involves the release of anti-inflammatory cytokines and growth factors, the recruitment of MSCs and changes in macrophage function and activity, generally known as the macrophage polarity switch. In essence, there is a delicate balance, influenced by many interacting elements, that controls the resolution of inflammation. In stem cell transplantation approaches, the immunosuppressing properties of MSCs may aid in this process towards regeneration. In acellular strategies, other ways must be devised to compensate for this, as inflammation might compromise immunomodulatory properties of the host stem cells.75 Modulation may be achieved via the spatiotemporally controlled release of pro and anti-inflammatory cytokines, although the structure and geometry of the material used may also play a role.76 Naturally, the chemistry of the material also has a great influence on the inflammatory response. This is especially critical when ECM-based materials are used, as, depending on the treatments performed, different moieties can remain in the material provoking unexpected, sometimes undesired, effects. For instance, incomplete decellularisation has been proved to cause macrophage polarisation to pro-inflammatory phenotypes (M1 to M2).77

There are alternatives to the traditional delivery of GFs and cytokines to boost migration, proliferation and differentiation of endogenous stem cells. Small molecules can direct signalling pathways to encourage regeneration. Wnt signalling, a metabolic pathway involved in tissue repair,78 has been recently proved by Neves et al. to play a role in dentine–pulp regeneration.79 Several Wnt agonists were loaded in separate collagen sponges and implanted in murine molars with previously prepared narrow injuries. Molars implanted with the agonist-loaded sponges remained completely vital and presented a significant formation of tertiary dentine (compared to MTA and unloaded sponge control). Shh signalling pathway may also be relevant in dental tissue regeneration, as it appears to be involved in odontoblast differentiation and function.80 Other alternative strategies include RNA interference. Recent reports of FOXQ1 and SOST affecting proliferation and mineralisation ability of DPSCs (respectively) open the door to delivery of micro or small interfering RNA molecules to regulate the expression of these genes.81,82 Successful delivery of these molecules can be achieved incorporating appropriate carriers, cleavable links or via electrostatic interactions.

The scaffolds and materials included in this review have potential to support the growth of dental tissues, but in endodontic regeneration, sealing materials are still required to isolate the pulp chamber from the oral cavity while regeneration occurs. As alternatives to traditional materials (e.g. composites and bioactive cements), whose design is not based on cytocompatibility, new materials that support dental stem cell adhesion and proliferation may be used. Recently, high-throughput methods to discover new dental materials supportive of hDPSCs using polymer microarrays have been explored.79 A library of commercially-available light-curable materials were preliminarily screened by studying adhesion of hDPSCs after 24 h and 48 h of culture on a support hDPSCs adhesion. Interestingly, based on an adhesion index (defined as the product between cell count and cell area), two triacrylates monomers trimethylolpropane triacrylate (TMPTA) and pentaerythritol triacrylate (PETA) were identified as the most-promising materials from a chemically-diverse library of materials. The materials were engineered into a scalable and rapid light-curing formulation to assess proliferation and differentiation of hDPSCs in common cultureware. Polymers were shown to support hDPSC adhesion with high viability in serum-free media whereas bisphenol A glycidyl methacrylate (BisGMA), a common dental restorative material, could not support robust hDPSC adhesion with cells ultimately de-adhering and failing to reach confluency. Similar negative results were observed on, trimethylolpropane trimethacrylate (TMPTMA), a methacrylate structurally analogous to TMPTA, appearing to suggest that a hDPSC supportive niche was specific to triacrylate polymers. Adhesion was promoted via Collagen I expression and integrin-β1 receptor binding for TMPTA and PETA, while for TMPTMA and BisGMA this was not the case and other matrix genes were upregulated as a compensatory mechanism. The triacrylate polymers were found to allow odontogenic differentiation of hDPSCs when induced by addition of osteogenic media, as measured via the increased expression of several odontogenic markers (ALPL, OCN, OSP, SCN1 and RUNX2, upon the addition of TFGβ1, DSPP). However, further testing using an in vivo model of reversible pulpitis to confirm their efficacy is required, but their cytocompatibility makes them promising as sealing materials to be used in vital pulp therapy or in conjunction with scaffolds for dentine–pulp regeneration.

Finally, anisotropy provides an interesting and largely unexplored design parameter for biomaterials for dental tissue regeneration. It may be utilised for tissue regeneration, as it can aid in cellular trafficking and recruitment, mimicking the natural structure of dental tissues such as dentine. Furthermore, biphasic gradient scaffolds, both for physical properties, such as stiffness83 and porosity,84 and chemically can induce differentiation, facilitate adhesion and migration. This may be particularly useful at the dentine–pulp interface to orchestrate multiple cell types to perform functions such as matrix production (ECM and tertiary dentine) and organise into complex structures mimicking the dentine–pulp complex.

Taking inspiration from the fields of tissue engineering and regenerative medicine, dental biomaterials have progressed from inert materials aimed at replacing and restoring dental structures to bioactive biomaterials, which are designed to regenerate tissues and restore the natural functions of the tooth. For the next generation of regenerative dental procedures to be successful, bioinstructive materials will be required. We believe acellular biomaterials such as those described in this review will facilitate the translation of such regenerative dental therapies.

Conflicts of interest

There are no conflicts to declare.


The authors would like to acknowledge the Department of Bioengineering, Imperial College for funding.


  1. WHO Department of Nutrition for Health and Development, 2017.
  2. Public Health England, Child oral health: applying All Our Health - GOV.UK,, (accessed 7 August 2018).
  3. Z. Mohammadi and P. M. H. Dummer, Int. Endod. J., 2011, 44, 697–730 CrossRef CAS PubMed.
  4. U. Schröder, J. Dent. Res., 1985, 64, 541–548 CrossRef PubMed.
  5. L. Ajay Sharma, A. Sharma and G. J. Dias, J. Investig. Clin. Dent., 2015, 6, 85–98 CrossRef PubMed.
  6. P. Z. Tawil, D. J. Duggan and J. C. Galicia, Compend. Contin. Educ. Dent., 2015, 36, 247–252 Search PubMed ; quiz 254, 264.
  7. T. Dammaschke, H. U. V. Gerth, H. Züchner and E. Schäfer, Dent. Mater., 2005, 21, 731–738 CrossRef CAS PubMed.
  8. M. Parirokh, M. Torabinejad and P. M. H. Dummer, Int. Endod. J., 2018, 51, 177–205 CrossRef CAS PubMed.
  9. F. Schwendicke, F. Brouwer and M. Stolpe, J. Endod., 2015, 41, 1969–1974 CrossRef PubMed.
  10. C. Zhu, B. Ju and R. Ni, Int. J. Clin. Exp. Med., 2015, 8, 17055–17060 CAS.
  11. Z. Li, L. Cao, M. Fan and Q. Xu, J. Endod., 2015, 41, 1412–1417 CrossRef PubMed.
  12. M. Torabinejad, M. Parirokh and P. M. H. Dummer, Int. Endod. J., 2018, 51, 284–317 CrossRef CAS PubMed.
  13. T. Dammaschke, L. Goupy, M. T. Firla and F. Bronnec, Focus on Biodentine, 2012 Search PubMed.
  14. B. N. Çelik, M. S. Mutluay, V. Arıkan and Ş. Sarı, Clin. Oral Investig., 2018, 1–6 Search PubMed.
  15. S. Rajasekharan, L. C. Martens, R. G. E. C. Cauwels and R. P. Anthonappa, Eur. Arch. Paediatr. Dent., 2018, 19, 1–22 CrossRef CAS PubMed.
  16. N. P. Solanki, K. K. Venkappa and N. C. Shah, J. Conservative Dent., 2018, 21, 10–15 Search PubMed.
  17. Ö. Malkondu, M. Karapinar Kazandağ and E. Kazazoğlu, BioMed Res. Int., 2014, 2014, 160951 Search PubMed.
  18. World Health Organization, International Clinical Trials Registry Platform,, (accessed 30 September 2018).
  19. European Medicines Agency and Heads of Medicine Agencies, EU Clinical Trials Register,, (accessed 30 September 2018).
  20. U.S. National Library of Medicine,,, (accessed 30 September 2018).
  21. B. N. Östby, Acta Odontol. Scand., 1961, 19, 323–353 CrossRef.
  22. L. He, J. Zhong, Q. Gong, B. Cheng, S. G. Kim, J. Ling and J. J. Mao, Dent. Clin. North Am., 2017, 61, 143–159 CrossRef PubMed.
  23. P. Becerra, D. Ricucci, S. Loghin, J. L. Gibbs and L. M. Lin, J. Endod., 2014, 40, 133–139 CrossRef PubMed.
  24. L. M. Lin, D. Ricucci and G. T.-J. Huang, Int. Endod. J., 2014, 47, 713–724 CrossRef CAS PubMed.
  25. L. He, J. Zhong, Q. Gong, S. G. Kim, S. J. Zeichner, L. Xiang, L. Ye, X. Zhou, J. Zheng, Y. Liu, C. Guan, B. Cheng, J. Ling and J. J. Mao, Sci. Rep., 2017, 7, 13941 CrossRef PubMed.
  26. A. Lolato, C. Bucchi, S. Taschieri, A. El Kabbaney and M. Del Fabbro, Platelets, 2016, 27, 383–392 CrossRef CAS PubMed.
  27. V. Y. Shivashankar, D. A. Johns, R. Kumar Maroli, M. Sekar, R. Chandrasekaran, S. Karthikeyan and S. K. Renganathan, J. Clin. Diagn. Res., 2017, 11, 34–39 Search PubMed.
  28. D. David Clanton and D. Paul D Eleazer, Novel Collagen Scaffold vs Conventional Scaffold in Regeneration of Human Dental Pulp Tissue - Full Text View -,, (accessed 30 September 2018).
  29. J. Y. Kim, X. Xin, E. K. Moioli, J. Chung, C. H. Lee, M. Chen, S. Y. Fu, P. D. Koch and J. J. Mao, Tissue Eng., Part A, 2010, 16, 3023–3031 CrossRef CAS PubMed.
  30. N. Ruangsawasdi, M. Zehnder, R. Patcas, C. Ghayor, B. Siegenthaler, B. Gjoksi and F. E. Weber, Tissue Eng., Part A, 2017, 23, 115–123 CrossRef CAS PubMed.
  31. S. Pan, S. Dangaria, G. Gopinathan, X. Yan, X. Lu, A. Kolokythas, Y. Niu and X. Luan, Stem Cell Rev. Rep., 2013, 9, 655–667 CrossRef CAS PubMed.
  32. J. W. Yang, Y. F. Zhang, C. Y. Wan, Z. Y. Sun, S. Nie, S. J. Jian, L. Zhang, G. T. Song and Z. Chen, Biomaterials, 2015, 44, 11–23 CrossRef CAS PubMed.
  33. Y. Akazawa, T. Hasegawa, Y. Yoshimura, N. Chosa, T. Asakawa, K. Ueda, A. Sugimoto, T. Kitamura, H. Nakagawa, A. Ishisaki and T. Iwamoto, Recruitment of mesenchymal stem cells by stromal cell-derived factor 1α in pulp cells from deciduous teeth YUKI, University of Crete, Faculty of Medicine, 2015, vol. 36 Search PubMed.
  34. T. Suzuki, C. H. Lee, M. Chen, W. Zhao, S. Y. Fu, J. J. Qi, G. Chotkowski, S. B. Eisig, A. Wong and J. J. Mao, J. Dent. Res., 2011, 90, 1013–1018 CrossRef CAS PubMed.
  35. L. Li and Z. Wang, RSC Adv., 2016, 6, 109519–109527 RSC.
  36. S. Eramo, A. Natali, R. Pinna and E. Milia, Int. Endod. J., 2018, 51, 405–419 CrossRef CAS PubMed.
  37. T. W. Gilbert, T. L. Sellaro and S. F. Badylak, Biomaterials, 2006, 27, 3675–3683 CAS.
  38. J. Hodde, Aust. N. Z. J. Surg., 2006, 76, 1096–1100 CrossRef PubMed.
  39. V. C. M. Neves and P. T. Sharpe, J. Dent. Res., 2018, 97, 416–422 CrossRef CAS PubMed.
  40. P. L. Tomson, P. J. Lumley, A. J. Smith and P. R. Cooper, Int. Endod. J., 2017, 50, 281–292 CrossRef CAS PubMed.
  41. A. J. Smith, B. A. Scheven, Y. Takahashi, J. L. Ferracane, R. M. Shelton and P. R. Cooper, Arch. Oral Biol., 2012, 57, 109–121 CrossRef CAS PubMed.
  42. B. Yang, G. Chen, J. Li, Q. Zou, D. Xie, Y. Chen, H. Wang, X. Zheng, J. Long, W. Tang, W. Guo and W. Tian, Biomaterials, 2012, 33, 2449–2461 CrossRef CAS PubMed.
  43. W. Guo, K. Gong, H. Shi, G. Zhu, Y. He, B. Ding, L. Wen and Y. Jin, Biomaterials, 2012, 33, 1291–1302 CrossRef CAS PubMed.
  44. H.-M. Yun, E.-S. Lee, M. Kim, J.-J. Kim, J.-H. Lee, H.-H. Lee, K.-R. Park, J.-K. Yi, H.-W. Kim and E. Kim, PLoS One, 2015, 10, e0138614 CrossRef PubMed.
  45. N. Siddiqui, S. Asawa, B. Birru, R. Baadhe and S. Rao, Mol. Biotechnol., 2018, 60, 506–532 CrossRef CAS PubMed.
  46. Y. Wu, W. Jiang, X. Wen, B. He, X. Zeng, G. Wang and Z. Gu, Biomed. Mater., 2010, 5, 015001 CrossRef PubMed.
  47. J. Meng, Y. Zhang, X. Qi, H. Kong, C. Wang, Z. Xu, S. Xie, N. Gu and H. Xu, Nanoscale, 2010, 2, 2565 RSC.
  48. D. Shan, Y. Shi, S. Duan, Y. Wei, Q. Cai and X. Yang, Mater. Sci. Eng., C, 2013, 33, 3498–3505 CrossRef CAS PubMed.
  49. W.-Y. Lai, S.-W. Feng, Y.-H. Chan, W.-J. Chang, H.-T. Wang, H.-M. Huang, W.-Y. Lai, S.-W. Feng, Y.-H. Chan, W.-J. Chang, H.-T. Wang and H.-M. Huang, Polymers, 2018, 10, 804 CrossRef.
  50. H.-M. Yun, S.-K. Kang, R. K. Singh, J.-H. Lee, H.-H. Lee, K.-R. Park, J.-K. Yi, D.-W. Lee, H.-W. Kim and E.-C. Kim, Dent. Mater., 2016, 32, 1301–1311 CrossRef CAS PubMed.
  51. C. R. Silva, P. S. Babo, M. Gulino, L. Costa, J. M. Oliveira, J. Silva-Correia, R. M. A. Domingues, R. L. Reis and M. E. Gomes, Acta Biomater., 2018, 77, 155–171 CrossRef CAS PubMed.
  52. A. Fallacara, E. Baldini, S. Manfredini, S. Vertuani, A. Fallacara, E. Baldini, S. Manfredini and S. Vertuani, Polymers, 2018, 10, 701 CrossRef.
  53. L. D. F. Almeida, P. S. Babo, C. R. Silva, M. T. Rodrigues, J. Hebling, R. L. Reis and M. E. Gomes, J. Mater. Sci. Mater. Med., 2018, 29, 88 CrossRef PubMed.
  54. P. S. Babo, R. L. Pires, L. Santos, A. Franco, F. Rodrigues, I. Leonor, R. L. Reis and M. E. Gomes, ACS Biomater. Sci. Eng., 2017, 3, 1359–1369 CrossRef CAS.
  55. P. K. Nguyen, W. Gao, S. D. Patel, Z. Siddiqui, S. Weiner, E. Shimizu, B. Sarkar and V. A. Kumar, ACS Omega, 2018, 3, 5980–5987 CrossRef CAS PubMed.
  56. B. Ji, L. Sheng, G. Chen, S. Guo, L. Xie, B. Yang, W. Guo and W. Tian, Tissue Eng., Part A, 2015, 21, 26–34 CrossRef CAS PubMed.
  57. H. Liu, W. Li, C. Gao, Y. Kumagai, R. W. Blacher and P. K. DenBesten, J. Dent. Res., 2004, 83, 496–499 CrossRef CAS.
  58. N. Six, D. Septier, C. Chaussain-Miller, R. Blacher, P. DenBesten and M. Goldberg, J. Dent. Res., 2007, 86, 780–785 CrossRef CAS.
  59. K. Kim, C. H. Lee, B. K. Kim and J. J. Mao, J. Dent. Res., 2010, 89, 842–847 CrossRef CAS PubMed.
  60. W. Lee, J.-H. Oh, J.-C. Park, H.-I. Shin, J.-H. Baek, H.-M. Ryoo and K. M. Woo, Acta Biomater., 2012, 8, 2986–2995 CrossRef CAS PubMed.
  61. J. Yang, Y. Zhang, Z. Sun, G. Song and Z. Chen, J. Biomater. Appl., 2015, 30, 221–229 CrossRef CAS PubMed.
  62. Y. Zhang, W. Fan, Z. Ma, C. Wu, W. Fang, G. Liu and Y. Xiao, Acta Biomater., 2010, 6, 3021–3028 CrossRef CAS PubMed.
  63. Y.-J. Yoo, J.-H. Oh, Q. Zhang, W. Lee and K. M. Woo, J. Endod., 2018, 44, 98–103.e1 CrossRef PubMed.
  64. Y. Shirakata, A. Sculean, Y. Shinohara, K. Sena, N. Takeuchi, D. D. Bosshardt and K. Noguchi, Clin. Oral Investig., 2016, 20, 1791–1800 CrossRef CAS PubMed.
  65. N. Okubo, T. Fujita, Y. Ishii, M. Ota, Y. Shibukawa and S. Yamada, J. Biomater. Appl., 2011, 27, 627–637 CrossRef PubMed.
  66. L. Chambrone, F. Sukekava, M. G. Araújo, F. E. Pustiglioni, L. A. Chambrone and L. A. Lima, J. Periodontol., 2010, 81, 452–478 CrossRef CAS.
  67. F. Cairo, M. Nieri and U. Pagliaro, J. Clin. Periodontol., 2014, 41, S44–S62 CrossRef.
  68. A. J. Smith, B. A. Scheven, Y. Takahashi, J. L. Ferracane, R. M. Shelton and P. R. Cooper, Arch. Oral Biol., 2012, 57, 109–121 CrossRef CAS PubMed.
  69. J. G. Smith, A. J. Smith, R. M. Shelton and P. R. Cooper, Exp. Cell Res., 2012, 318, 2397–2406 CrossRef CAS PubMed.
  70. R. X. Wu, X. Y. Xu, J. Wang, X. T. He, H. H. Sun and F. M. Chen, Appl. Mater. Today, 2018, 11, 144–165 CrossRef.
  71. S. Elsharkawy and A. Mata, Adv. Healthcare Mater., 2018, 7, 1800178 CrossRef PubMed.
  72. S. Elsharkawy, M. Al-Jawad, M. F. Pantano, E. Tejeda-Montes, K. Mehta, H. Jamal, S. Agarwal, K. Shuturminska, A. Rice, N. V. Tarakina, R. M. Wilson, A. J. Bushby, M. Alonso, J. C. Rodriguez-Cabello, E. Barbieri, A. del Río Hernández, M. M. Stevens, N. M. Pugno, P. Anderson and A. Mata, Nat. Commun., 2018, 9, 2145 CrossRef.
  73. J. G. Leprince, B. D. Zeitlin, M. Tolar and O. A. Peters, Int. Endod. J., 2012, 45, 689–701 CrossRef CAS.
  74. Z. Li, C.-M. Jiang, S. An, Q. Cheng, Y.-F. Huang, Y.-T. Wang, Y.-C. Gou, L. Xiao, W.-J. Yu and J. Wang, Oral Dis., 2014, 20, 25–34 CrossRef CAS PubMed.
  75. F. B. Yazid, N. Gnanasegaran, W. Kunasekaran, V. Govindasamy and S. Musa, Clin. Oral Investig., 2014, 18, 2103–2112 CrossRef PubMed.
  76. O. Veiseh, J. C. Doloff, M. Ma, A. J. Vegas, H. H. Tam, A. R. Bader, J. Li, E. Langan, J. Wyckoff, W. S. Loo, S. Jhunjhunwala, A. Chiu, S. Siebert, K. Tang, J. Hollister-Lock, S. Aresta-Dasilva, M. Bochenek, J. Mendoza-Elias, Y. Wang, M. Qi, D. M. Lavin, M. Chen, N. Dholakia, R. Thakrar, I. Lacík, G. C. Weir, J. Oberholzer, D. L. Greiner, R. Langer and D. G. Anderson, Nat. Mater., 2015, 14, 643–651 CrossRef CAS PubMed.
  77. T. J. Keane, R. Londono, N. J. Turner and S. F. Badylak, Biomaterials, 2012, 33, 1771–1781 CrossRef CAS.
  78. J. L. Whyte, A. A. Smith and J. A. Helms, Cold Spring Harbor Perspect. Biol., 2012, 4, a008078–a008078 Search PubMed.
  79. V. C. M. Neves, R. Babb, D. Chandrasekaran and P. T. Sharpe, Sci. Rep., 2017, 7, 39654 CrossRef CAS PubMed.
  80. Y. Ishikawa, M. Nakatomi, H. Ida-Yonemochi and H. Ohshima, Cell Tissue Res., 2017, 369, 497–512 CrossRef CAS PubMed.
  81. S. Tu, J. Zheng, X. Gao, C. Guan, B. Cai and L. Xiang, Biochem. Biophys. Res. Commun., 2018, 497, 543–549 CrossRef CAS PubMed.
  82. A.-M. Collignon, N. Amri, J. Lesieur, J. Sadoine, S. Ribes, S. Menashi, S. Simon, A. Berdal, G. Y. Rochefort, C. Chaussain and C. Gaucher, J. Dent. Res., 2017, 96, 815–821 CrossRef CAS PubMed.
  83. T. Qu, J. Jing, Y. Ren, C. Ma, J. Q. Feng, Q. Yu and X. Liu, Acta Biomater., 2015, 16, 60–70 CrossRef CAS PubMed.
  84. S. Panseri, M. Montesi, S. M. Dozio, E. Savini, A. Tampieri and M. Sandri, Front. Bioeng. Biotechnol., 2016, 4, 48 Search PubMed.

This journal is © The Royal Society of Chemistry 2019