Abstract
Excessive oxidative response, unbalanced immunomodulation, and impaired
mesenchymal stem cell function in periodontitis in diabetes makes it a
great challenge to achieve integrated periodontal tissue regeneration.
Here, a polyphenol-mediated redox-active algin/gelatin hydrogel
encapsulating a conductive poly(3,4-ethylenedioxythiopene)-assembled
polydopamine-mediated silk microfiber network and a hydrogen sulfide
sustained-release system utilizing bovine serum albumin nanoparticles
is developed. This hydrogel is found to reverse the hyperglycemic
inflammatory microenvironment and enhance functional tissue
regeneration in diabetic periodontitis. Polydopamine confers the
hydrogel with anti-oxidative and anti-inflammatory activity. The slow,
sustained release of hydrogen sulfide from the bovine serum albumin
nanoparticles recruits mesenchymal stem cells and promotes subsequent
angiogenesis and osteogenesis. Moreover,
poly(3,4-ethylenedioxythiopene)-assembled polydopamine-mediated silk
microfiber confers the hydrogel with good conductivity, which enables
it to transmit endogenous bioelectricity, promote cell arrangement, and
increase the inflow of calcium ion. In addition, the synergistic
effects of hydrogen sulfide gaseous-bioelectric coupling promotes bone
formation by amplifying autophagy in periodontal ligament stem cells
and modulating macrophage polarization via lipid metabolism regulation.
This study provides innovative insights into the synergistic effects of
conductivity, reactive oxygen species scavenging, and hydrogen sulfide
on the periodontium in a hyperglycemic inflammatory microenvironment,
offering a strategy for the design of gaseous-bioelectric biomaterials
to promote functional tissue regeneration in immune-related diseases.
Subject terms: Gels and hydrogels, Polymers, Biomedical materials
__________________________________________________________________
Biomaterials have potential in treating periodontitis in diabetes, but
addressing the range of issues with one material is challenging. Here,
the authors report the development of a redox-active hydrogel for
reducing inflammation and promoting tissue regeneration in diabetic
periodontitis.
Introduction
Periodontitis is an infection-driven inflammatory disease which can
cause irreversible damage to the periodontal tissue and subsequently
tooth loss^[46]1,[47]2. The severity of periodontitis is increased by
diabetes mellitus, even with little dental plaque accumulation^[48]3.
Both the innate and adaptive immune systems, which contribute to
periodontium damage, are affected by diabetes^[49]3. In addition, as
periodontitis-derived inflammatory factors released into the
bloodstream, glycemic control is adversely affected and the severity of
diabetic complications could be increased^[50]4,[51]5. Hence, the
treatment of periodontitis in diabetes patient is important not only
for the preservation of dentition, but also to minimize the effects of
the periodontitis-induced systemic inflammatory burden on glycemic
control and diabetic complications. Currently, general strategies for
the treatment of periodontitis and alveolar bone regeneration include
subgingival instrumentation accompanied by regenerative periodontal
surgery as well as local administration of antibiotics and/or bioactive
molecules^[52]6–[53]10. However, the beneficial effects of these
operations are limited by the overactivated oxidative stress response,
excessive inflammation, unbalanced immunomodulation, and impaired
mesenchymal stem cell (MSC) function present under diabetic
conditions^[54]11,[55]12. Insufficient new alveolar bone formation,
secondary bone resorption, and recurrence of infection are frequently
seen during the treatment of diabetic periodontitis^[56]13. Guided
bone/tissue regeneration (GBR/GTR), such as collagen based matrix,
pericardium membranes and functionally gradient membranes, are widely
used to reconstruct alveolar bone in periodontitis. However,
traditional GBR/GTR membranes are used as the physical barrier.
Although recent GBR/GTR membranes have been endowed with biological
properties to promote periodontal bone regeneration^[57]14,[58]15, the
therapeutic effects of these membranes are limited by the
diabetes-aggravated inflammatory environment.
Endogenous signaling such as bioelectric, mechanical and gaseous
signalings mediate dramatic changes in the composition of cells,
tissues and their microenvironments, and subsequently orchestrate the
healing process^[59]16–[60]19. In particular, hydrogen sulfide (H[2]S),
an endogenous gasotransmitter, influences several physiological and
pathophysiological processes^[61]20. It is reported to alleviate
alveolar bone resorption in periodontitis through attenuating
inflammation, apoptosis, and reactive oxygen species
(ROS)^[62]21,[63]22. Moreover, the osteogenic function of H[2]S in
periodontal tissue has been established^[64]23. However, the
biosynthesis of H[2]S is impaired in diabetes patients, which has an
adverse effect on MSC migration, angiogenesis, inflammation control and
subsequently tissue regeneration^[65]24–[66]28. Recently, although many
efforts have attempted to produce exogenous H[2]S in wounds using H[2]S
donors such as inorganic sulfide salt, GYY4137, and DTTs, their
uncontrollably fast rate of H[2]S generation and poor water solubility
have limited their therapeutic effects and clinical
applications^[67]29,[68]30.
Bioelectricity is an important physiological activity in the human
body, mostly realized through endogenous electric fields such as
membrane potential and nerve potential^[69]17,[70]31. Endogenous
electrical signals can modulate myriad biological processes, from the
cell cycle, migration, proliferation and differentiation to nerve
conduction, muscle contraction, embryogenesis and tissue
regeneration^[71]17,[72]32–[73]35. Notably, electrical signals
generated by mechanical stress act as stimuli for bone growth and
remodeling^[74]36. Moreover, electrical signals improve cell
arrangement, which is essential for functional periodontium
regeneration^[75]17,[76]32,[77]37. Unfortunately, it is difficult for
traditional GBR/GTR membranes to transmit endogenous electrical signals
due to their lack of conductivity. Thus, functional and conductive
GBR/GTR membranes to couple endogenous bioelectric signals have great
potential for diabetic periodontal tissue regeneration^[78]31.
Hydrogels have gained increasing attention in the field of periodontal
tissue regeneration due to their structure and composition, which are
similar to natural extracellular matrix^[79]38,[80]39. In the diabetic
periodontal environment, locally aggravated inflammatory responses can
be attenuated by hydrogels with immunomodulatory ability to reverse the
inflammatory conditions. However, most of these hydrogels have been
designed with a focus on the delivery of growth factors or stem
cells^[81]38–[82]40, with a lack of attention to the biological effects
of the main components of the hydrogels. In addition, conductive
hydrogel designs usually use conductive fillers such as ionic liquids,
conductive polymers, carbon and metal nanoparticles. Biosafety concerns
about the potential leakage of ionic liquid, the aggregation caused by
the hydrophobicity of conductive polymers and nanoparticles, and the
lack of biofunctionality have limited these conductive fillers used in
the construction of conductive immunomodulatory hydrogels. Polyphenols
such as tannic acid, pyrogallol, gallic acid and dopamine have been
widely applied to functionalize hydrogels due to their abundant
catechol and pyrogallol groups. Moreover, polyphenols also exhibit
excellent biocompatibility, antioxidant properties, and
immunomodulatory activity^[83]41. Our previous studies have
demonstrated that polyphenol-mediated strategies can promote the
hydrophilicity of conductive fillers, ensuring their uniform dispersion
throughout hydrogel networks. At the same time, polyphenol-mediated
conductive fillers endow these hydrogels with desirable biological
functions such as transmitting bioelectrical signals to promote tissue
regeneration.
In this study, we developed a polyphenol-mediated redox-active
algin/gelatin (AG) hydrogel possessing a H[2]S sustained-release system
as well as a conductive poly(3,4-ethylenedioxythiopene)-assembled
polydopamine-mediated silk microfiber (PEDOT-PSF) network to couple
endogenous electrical and gaseous signals. The H[2]S sustained-release
system was constructed using NaHS-encapsulated bovine serum albumin
nanoparticles (BNPs). The PEDOT-PSF networks were fabricated using a
polyphenol-mediated protection-extraction strategy. The PEDOT-PSF
uniformly distributed throughout the hydrogel, enhancing its mechanical
and conductive properties. With its H[2]S sustained-release and
conductivity, the hydrogel can couple endogenous electrical and gaseous
signals, promoting MSC recruitment, cell arrangement, angiogenesis and
osteogenesis as well as M2 macrophage polarization. These properties
enable the hydrogel to reverse the inflammatory microenvironment,
creating a pro-regenerative microenvironment in a hyperglycemic
inflammatory microenvironment (Fig. [84]1).
Fig. 1. Schematic illustration of the synthesis and properties of the
polyphenol-mediated redox-active hydrogel with H[2]S gaseous-bioelectric
coupling.
[85]Fig. 1
[86]Open in a new tab
a Schematic diagram of PEDOT-PSF synthesis. b Schematic diagram of BNPs
synthesis. c The BNP-PEDOT-PSF hydrogel. d PDA endows the hydrogel with
the ability of antioxidative and immunomodulatory. e The conductive
hydrogel transmits endogenous bioelectricity and consequently improves
cell alignment and increases Ca^2+ influx. f H[2]S released from BNPs
promotes MSCs migration, angiopoiesis, osteogenesis, and M2 macrophage
polarization. MSC mesenchymal stem cell.
Results
Design strategies
To couple endogenous electrical signals, the conductive pathway in a
hydrogel is essential. PEDOT-PSF was used to construct the conductive
pathway in our AG hydrogel network. First, the PSF was extracted from
raw silk via a polydopamine (PDA)-mediated protection-extraction
process. After extraction, the catechol groups on the surface of the
PSF serve as anchor sites to induce EDOT self-assembly in situ
(Fig. [87]1a). To rescue downregulated H[2]S levels in periodontitis in
diabetic environment (Supplementary Fig. [88]1), NaHS was encapsulated
in BNPs using a desolvation method to release H[2]S persistently
(Fig. [89]1b). To avoid NaHS exposure to water in the process of
fabrication, NaHS was resolved in ethanol phase. AG hydrogel was
crosslinked with Poly (ethylene glycol) diglycidyl ether (PEGDE) and
Ca^2+, and enhanced by PEDOT-PSF (Fig. [90]1c). In the hydrogel matrix,
the electron transfer from PEDOT to PDA induces the catechol-quinone
redox balance of PDA, such that there remain abundant catechol groups
for the hydrogel. The catechol-quinone redox balance plays an important
role in scavenging ROS, relieving oxidative stress, and modulating
local immune responses (Fig. [91]1d). Moreover, the conductive hydrogel
couples endogenous electrical signals to ameliorate cell arrangement,
increase Ca^2+ internal flow, and subsequently potentiate osteogenesis
(Fig. [92]1e). Meanwhile, the low concentration of H[2]S released by
BNPs replenishes the impaired endogenous gas signals to promote MSC
recruitment, angiogenesis and osteogenesis as well as M2 macrophage
polarization (Fig. [93]1f). Therefore, the catechol-quinone redox
balance, electroactivity, and H[2]S release of the hydrogel
synergistically promote diabetic periodontal tissue regeneration.
Characterization of the hydrogel
The successful assembly of PDA and PEDOT on the surface of silk
microfiber (mSF) was verified using scanning electron microscopy (SEM)
(Fig. [94]2a–c). The cross-section image showed that after assembly of
PEDOT on the surface of PSF, the surface of PEDOT-PSF is rougher than
that of mSF and PSF (Supplementary Fig. [95]2a). The element mapping of
the PEDOT-PSF showed the present of S element in the PEDOT-PSF,
confirming the successful assembly of PEDOT on the surface of PSF
(Fig. [96]2d). XRD and FTIR showed in Supplementary Fig. [97]2b, c also
proved the successful assembly of PEDOT on mSF. The BNPs appeared
spherical, with a diameter of approximately 300 nm (Fig. [98]2e).
Furthermore, to investigate the reduction of quinone groups into
catechol groups by conductivity PEDOT-PSF and reductive H[2]S, PSF,
PEDOT-PSF, and H[2]S treated PEDOT-PSF were examined using X-ray
photoelectron spectroscopy (XPS). After assembly of conductive PEDOT on
the surface of PSF, the content of C-O groups increased from
31.71 ± 8.28 to 57.60 ± 3.88%, and the content of C = O groups
decreased from 68.29 ± 8.28 to 42.40 ± 3.88%, which implies that
quinone groups were reduced to catechol groups through the electron
transfer from PEDOT to PDA (Fig. [99]2f, g, Supplementary
Table [100]3). Reductive H[2]S further increased the content of C-O
groups from 57.60 ± 3.88 to 64.39 ± 1.24% and deceased the content of
C = O groups from 42.40 ± 3.88% to 35.61 ± 1.24%, protecting the phenol
groups from oxidation (Fig. [101]2g, h, Supplementary Table [102]3).
The results demonstrated that this system has redox activity. In
addition, SEM images showed that AG hydrogel incorporating the BNPs and
PEDOT-PSF (BNP-PEDOT-PSF-AG) had a uniform porous structure
(Fig. [103]2i). BNPs and PEDOT-PSF were embedded in the hydrogel
(Fig. [104]2i). Notably, PSF-AG hydrogel had denser and more uniform
microstructures than the hydrogel without PSF (Supplementary
Fig. [105]4a). Moreover, incorporation of BNPs did not change the
microstructure of the hydrogel. After incorporation of BNPs, the
hydrogel could sustain release of H[2]S for 21 days (Fig. [106]2j).
Fig. 2. Characterization of the BNP-PEDOT-PSF hydrogel.
[107]Fig. 2
[108]Open in a new tab
SEM micrographs of mSF (a), PSF (b), and PEDOT-PSF (c). d Element
mapping of PEDOT-PSF. e SEM micrograph of NaHS@BNP. XPS analysis of PSF
(f), PEDOT-PSF (g), and H[2]S treated PEDOT-PSF (h). i SEM micrographs
of the BNP-PEDOT-PSF-AG hydrogel. Arrows indicate PEDOT-PSF and
NaHS@BNP embedded in the hydrogel. j The amount of H[2]S released from
the BNP-PEDOT-PSF hydrogel in 25 days (mean ± SEM, n = 4 independent
samples at each time point). k Cyclic compression test curve of the
BNP-PEDOT-PSF-AG hydrogel. l Conductivities of hydrogels with different
contents of PEDOT-PSF (mean ± SEM, n = 5 independent samples, one-way
ANOVA). m A light-emitting diode (LED) was illuminated in a
BNP-PEDOT-PSF-AG hydrogel-connected circuit. n The BNP-PEDOT-PSF-AG
hydrogel with an illuminated LED was compressed and recovered. o
Conductivity of the BNP-PEDOT-PSF-AG hydrogel after compression
(mean ± SEM, n = 4 independent samples, one-way ANOVA).
The ability to withstand compression is important for a periodontal
regenerative hydrogel, since it must bear masticatory force upon
implantation. Owing to the ultralong length and uniform dispersion of
PEDOT-PSF in the hydrogel, stress can be transferred along the
PEDOT-PSF in the system. Thus, PSF and PEDOT-PSF confer the AG
hydrogels with good mechanical properties. With the addition of PSF and
PEDOT-PSF, the compressive strength of the BNP-PEDOT-PSF-AG hydrogels
increased from about 82 to 100 kPa (Supplementary Fig. [109]4b).
Moreover, the BNP-PEDOT-PSF-AG hydrogels recovered quickly upon
withdrawal of compression force after 10 cycles, indicating excellent
recovery properties of the hydrogel (Fig. [110]2k). In addition, the
introduce of PSF and PEDOT-PSF delayed the degradation of the
hydrogels, which provided a stable place for cells in the early stage
of alveolar bone repair (Supplementary Fig. [111]4d).
PEDOT-PSF also endows the hydrogels with good electrical conductivity
(Fig. [112]2l–o). With the increase of PEDOT-PSF content, the
conductivity of the BNP-PEDOT-PSF-AG hydrogel increased (Fig. [113]2l).
The conductivity reached 13.12 ± 1.66 S/m with 2 wt ‰ PEDOT-PSF in the
hydrogel, which was selected for the subsequent experiments
(Fig. [114]2l). In a BNP-PEDOT-PSF-AG hydrogel-connected circuit, a
light-emitting diode (LED) was illuminated upon connection
(Fig. [115]2m). Notably, the LED maintained lighting during the whole
process of compression and release, indicating excellent dynamic
conductivity of the hydrogel (Fig. [116]2n). Moreover, the conductivity
of the hydrogel was stable during the deformation (Fig. [117]2o) and in
vivo 14 days after implantation (Supplementary Fig. [118]5).
Osteoinductive ability of the BNP-PEDOT-PSF-AG hydrogel in vitro
The BNP-PEDOT-PSF-AG hydrogel exhibited excellent potential to serve as
a conductive platform for modulating cell proliferation, elongation,
and osteogenesis under high-throughput electrical stimulation (ES).
Human-derived periodontal ligament stem cells (PDLSCs) obtained from
extracted premolars were cultured on the PEDOT-PSF-AG and
BNP-PEDOT-PSF-AG hydrogel performed better in cell proliferation
(Supplementary Fig. [119]6), demonstrating the proliferation of PDLSCs
is not affected by the hydrogel. Moreover, as the immunofluorescence
staining images in Supplementary Fig. [120]7 show, PDLSCs on all of the
hydrogels exhibited good ability to spread and adhere. Notably, cells
on the PEDOT-PSF-AG hydrogel performed better in cell proliferation and
elongation than those on hydrogels without conductivity. Specifically,
the aspect ratio of the cells increased to 4.93 ± 0.45 at 600 mV from a
baseline of 3.04 ± 0.16 without ES, indicating that ES improved the
elongation of PDLSCs cultured on BNP-PEDOT-PSF-AG hydrogel
(Supplementary Fig. [121]7b). According to previous research, the yield
of osteogenesis increases with aspect ratio^[122]42. We also analyzed
the expression of osteogenesis-related genes and ALP activity by the
PDLSCs. Compared to the PSF-AG and AG groups, significantly higher
expression of OCN (Fig. [123]3a) and ALP (Fig. [124]3b) and ALP
activity (Fig. [125]3c) were found in the PEDOT-PSF-AG group under ES,
confirming the osteoinductivity of PEDOT-PSF. Moreover, expression
levels of OCN and ALP, and ALP activity in PDLSCs on BNP-PEDOT-PSF-AG
hydrogel were significantly higher than on the PEDOT-PSF-AG hydrogels,
indicating the osteogenic effects of H[2]S. These results demonstrated
that the conductive PEDOT-PSF and osteoinductive H[2]S gas in the
hydrogel synergistically improved osteogenic differentiation of PDLSCs.
In addition, BNP-PEDOT-PSF-AG group showed the highest migration rate
of PDLSCs, confirming the ability of H[2]S in promoting PDLSC migration
(Supplementary Fig. [126]8).
Fig. 3. Osteogenesis of PDLSCs on the hydrogels under high throughput
electrical stimulation (ES).
[127]Fig. 3
[128]Open in a new tab
Osteogenesis-related gene expressions of OCN (a) and ALP (b) in PDLSCs
on the various hydrogels under various ES potentials (n = 3 independent
biological samples).ES electrical stimulation. c ALP activity of PDLSCs
on the various hydrogels under various ES potentials (n = 3 independent
biological samples). d GO Enriched pathways of all genes in three
groups. BP biological processes, CC cellular component, MF molecular
functions. e Heatmap analysis of differentially expressed genes
involved in autophagy, rection to ROS, calcium ion transport, bone
formation and angiogenesis. ROS reactive oxygen species. f Flow
cytometry of Fluo-8 AM-labeled cells in the fluorescein isothiocyanate
(FITC)-A channel, indicating intracellular Ca^2+ concentration of cells
on the BNP-PEDOT-PSF-AG hydrogel with or without ES. g Quantification
analysis for flow cytometric analysis of Ca^2+ concentration in PDLSCs
(mean ± SEM, n = 3 independent biological samples, two-tailed t-test).
h Schematic illustration of the mechanism of osteoinductive ability of
BNP-PEDOT-PSF-AG hydrogel on PDLSCs. The blue arrow showed the effect
of the BNP-PEDOT-PSF-AG hydrogel. LPS lipopolysaccharides.
To explore the osteogenesis mechanism of PDLSCs on BNP-PEDOT-PSF-AG
hydrogel, we performed transcriptome analyses of PDLSCs on PSF-AG,
PEDOT-PSF-AG, and BNP-PEDOT-PSF-AG hydrogels after 3 days of culturing
under high glucose and inflammatory conditions with 600 mV ES
potentials. Pathway enrichment analysis of three groups include
positive regulation of cell migration, angiogenesis, reactive oxygen
species metabolic process, and positive regulation of osteoblast
differentiation (Fig. [129]3d), which correlated with osteoinductive
properties. Moreover, GSEA analysis showed downregulation of
inflammation and upregulation of autophagy (Supplementary
Fig. [130]9b). Based on existing evidence, upregulation of autophagy in
PDLSCs promotes alveolar bone healing^[131]43. However, autophagy is
impaired in periodontitis. Furthermore, the heatmap showed that ATG2B,
ATG12P1 were upregulated in the BNP-PEDOT-PSF-AG group. While SQSTM1,
which negative regulates autophagy, was downregulated in the
BNP-PEDOT-PSF-AG group, indicating augmentation of autophagy in
BNP-PEDOT-PSF-AG group (Fig. [132]3e). Hence, we speculated that H[2]S
and bioelectricity promote osteogenesis through upregulation of
autophagy.
Furthermore, enrichment analysis of 552 genes (excluded 90 genes shared
by BNP-PEDOT-PSF-AG versus PSF-AG) between PEDOT-PSF-AG and PSF-AG,
which representing the effect of bioelectricity included positive
regulation of cell migration, cellular response to metal ion, cell
adhesion, positive regulation of cell motility, regulation of actin
cytoskeleton reorganization, cytokine-cytokine receptor interaction,
and calcium signaling pathway, which correlated with the migration and
adhesion of PDLSCs (Supplementary Fig. [133]9c). Interestingly, a
growing body of scientific literature has revealed that ES induced
Ca^2+ influx plays a key role in osteogenic differentiation^[134]44. In
this study, significantly upregulated intracellular Ca^2+ were observed
in PDLSCs on BNP-PEDOT-PSF-AG hydrogel under ES (Fig. [135]3f, g).
In summary, the osteoinductive ability of the hydrogels can be
attributed to the following mechanisms (Fig. [136]3h). First, H[2]S
released from the BNPs and PEDOT-PSF synergistically promotes the
osteogenic differentiation of the PDLSCs through augmentation of
autophagy^[137]43,[138]45. Second, PEDOT-PSF increases the
intracellular Ca^2+ derived from microenvironment and hydrogels, which
promotes osteogenic differentiation^[139]44. Moreover, PEDOT-PSF
enhances cell elongation, which also contributes to the enhanced
osteogenic differentiation of the PDLSCs^[140]42.
Antioxidative activities of the BNP-PEDOT-PSF-AG hydrogel in vitro
In this hydrogel, PDA plays an important role for scavenging the ROS
via its catechol groups. However, during the process, the catechol
groups were oxidized to the quinone groups, which would reduce its
ROS-scavenging ability. Note that, mussel could remain its long-term
adhesion in seawater through controlling the catechol-quinone redox
balance. Inspired by mussel adhesion, we fabricated a PDA-mediated
conductive silk fiber. The conductive fibers endow the hydrogel with
good conductivity. Moreover, the electron could transfer from PEDOT to
PDA, which cooperated with H[2]S to facilitated the dynamic redox
between catechol and quinone groups. In short, the PEDOT-PSF/H[2]S
provided a redox-activity for the hydrogel (Fig. [141]4a).
Fig. 4. Antioxidative property of the BNP-PEDOT-PSF-AG hydrogel in vitro.
[142]Fig. 4
[143]Open in a new tab
a Antioxidative mechanism of the BNP-PEDOT-PSF-AG hydrogel. ROS,
reactive oxygen species. b DPPH-scavenging efficiency of AG, PSF-AG,
BNP-PSF-AG, PEDOT-PSF-AG, and BNP-PEDOT-PSF-AG hydrogels (mean ± SEM,
n = 3 independent samples at each time point). c Quantification
analysis for flow cytometric analysis of DCFH-DA-labeled cells in the
FITC-A channel on the various hydrogels (mean ± SEM, n = 4 independent
biological samples, one-way ANOVA). d Intracellular ROS-scavenging
performance of THP-1 on the various hydrogels. e MTT analysis of THP-1
(n = 4 independent biological samples). f Live/dead staining of THP-1.
A 1,1-diphenyl-2-picrylhydrazyl (DPPH) free radical scavenging assay
was employed to evaluate the antioxidative properties of the hydrogels.
The results showed that the DPPH-scavenging ratios of PSF-AG,
BNP-PSF-AG, PEDOT-PSF-AG, and BNP-PEDOT-PSF-AG hydrogels increased over
time (Fig. [144]4b). The DPPH-scavenging ratio of PSF-AG hydrogel was
higher than that of AG hydrogel owing to the antioxidative property of
PDA on the surface of PSF. Notably, better antioxidative ability was
observed in PEDOT-PSF-AG and BNP-PSF-AG hydrogel than PSF-AG hydrogel,
due to enhancement of the catechol-quinone dynamic redox derived from
the electron transfer from PEDOT to PDA or the release of H[2]S from
BNPs.
H[2]S has been recognized as a reducing agent, as it regulates
oxidative stress through direct reaction with ROS and reactive nitrogen
species (RNS) and indirect effects on related enzymes or other targets
in signaling pathways^[145]46. Specifically, hydrosulfide anions,
dissociated from H[2]S as powerful one electron chemical reductants,
could transfer hydrogen atom or single electron at or near
diffusion-controlled rates to quench free radicals^[146]47. Moreover,
H[2]S could attenuate cellular oxidative stress by improving the
activities of related enzymes such as catalase and glutathione
peroxidase (Gpx) or upregulating the levels of nonenzymatic
antioxidants such as classic thioredoxin and reduced
glutathione^[147]48–[148]50. Owing to the synergetic effect of H[2]S
and PEDOT-PSF, the BNP-PEDOT-PSF-AG hydrogel achieved the highest
DPPH-scavenging ratio among all tested hydrogels (Fig. [149]4b). This
ROS-scavenging property endows the hydrogel with ability to protect
cells from ROS damage.
Then, to measure the intracellular ROS levels in THP-1 cells cultured
on different hydrogels, a 2′,7′-dichlorofluorescein diacetate (DCFH-DA)
probe was applied. Similar to the tendency in the DPPH free radical
scavenging assay, PSF-AG, BNP-PSF-AG, PEDOT-PSF-AG, and
BNP-PEDOT-PSF-AG groups exhibited lower fluorescence intensity,
representing lower ROS levels in cells (Fig. [150]4c, d). Among these
four groups, the BNP-PEDOT-PSF-AG group showed the lowest fluorescence
intensity, indicating excellent ROS-scavenging properties that we
attribute to the synergistic effects of PEDOT-PSF and H[2]S. Moreover,
THP-1 cells cultured on the PEDOT-PSF-AG and BNP-PEDOT-PSF-AG hydrogels
performed better in cell proliferation (Fig. [151]4e, f), demonstrating
the proliferation of THP-1 is not affected by the hydrogels.
Collectively, these results demonstrated that the BNP-PEDOT-PSF-AG
hydrogel possesses excellent antioxidative ability.
Immunomodulatory activities of the BNP-PEDOT-PSF-AG hydrogel in vitro
The BNP-PEDOT-PSF-AG hydrogel also has immunomodulatory properties
through shifting macrophage polarization from M1 to M2. The
polarization of macrophages plays an important role in tissue
regeneration. Specifically, M1 macrophages play a pro-inflammatory role
and induce chronic inflammation, while M2 macrophages play an
anti-inflammatory role and promote tissue remodeling^[152]51. However,
M2 polarization is impacted under diabetic conditions^[153]52. To
evaluate the immunomodulatory properties of the hydrogels, the number
of M1 macrophages and M2 macrophages were detected on various
hydrogels. Compared to the AG group, the number of M1 macrophages in
the PSF-AG, BNP-PSF-AG, PEDOT-PSF-AG, and BNP-PEDOT-PSF-AG groups
significantly decreased, which indicated the ability of PSF to inhibit
M1 macrophage polarization (Fig. [154]5a, b, Supplementary
Fig. [155]10a, b). In addition, thanks to the excellent
immunomodulation ability of H[2]S, the BNP-PEDOT-PSF-AG group exhibited
a significant increase in the number of M2 macrophages compared to the
other groups. Moreover, the expression levels of macrophage
polarization-related genes (IL-1β, TNF-α, NOS2, DECTIN, IL-10, and
ARG-1) exhibited a similar trend (Supplementary Fig. [156]10c, d).
Furthermore, expression of osteoclastogenic-related genes (TRAP,
NFATC1, CTSK, and C-FOS) were detected in RAW 246.7 cells cultured on
hydrogels for 3 days under high glucose and inflammatory conditions
with the addition of receptor activator of nuclear factor kappa-B
ligand (RANKL). Significantly decreased osteoclast activity in both the
PEDOT-PSF-AG and BNP-PEDOT-PSF-AG groups indicated the excellent
ability of PEDOT-PSF to inhibit osteoclastogenesis (Supplementary
Fig. [157]11). In short, these results demonstrated that the
BNP-PEDOT-PSF-AG hydrogel possesses excellent immunomodulatory and
osteoclast-inhibiting activity.
Fig. 5. Immunomodulatory property and mechanism of the BNP-PEDOT-PSF-AG
hydrogel in vitro.
[158]Fig. 5
[159]Open in a new tab
a Immunofluorescence staining of THP-1 cells on the different
hydrogels. iNOS (red) stained M1 macrophages. CD163 (green) stained M2
macrophages. b Polarization of macrophages on the different hydrogels
were evaluated by expression of CD86 (M1) and CD206 (M2) using flow
cytometry. GSEA analysis of lipid catabolic process and lipid
biosynthetic process (c) and glutathione peroxidase activity (d). e
Heatmap of genes involved in macrophage polarization, lipid catabolic
process, lipid biosynthetic process, and glutathione peroxidase
activity. f Schematic illustration of the mechanism of immunomodulatory
ability of BNP-PEDOT-PSF-AG hydrogel on THP-1. The blue arrow showed
the effect of the BNP-PEDOT-PSF-AG hydrogel. Gpx glutathione
peroxidase, FA fatty acid, PUFA polyunsaturated fatty acids.
To elucidate the immunomodulatory mechanism of the BNP-PEDOT-PSF-AG
hydrogel, we performed transcriptome analyses of THP-1 cells on AG,
PEDOT-PSF-AG, and BNP-PEDOT-PSF-AG hydrogels after 3 days of culturing
under high glucose and lipopolysaccharides (LPS) stimulation. The
enriched terms between three groups included the inflammatory response,
innate immune response, and lipid metabolism related pathway
(Supplementary Fig. [160]12). Further GSEA analysis showed that
downregulation of lipid biosynthesis and regulation of lipid catabolism
were presented in the BNP-PEDOT-PSF-AG group (Fig. [161]5c). Moreover,
Gpx activity was also upregulated in the BNP-PEDOT-PSF-AG group
(Fig. [162]5d). Based on existing evidence, lipid synthesis is
important in the regulation of macrophage functions. Specifically,
lipid biosynthesis is upregulated in M1 macrophages to regulate
membrane remodeling and synthesis of inflammatory
mediators^[163]53,[164]54. While, lipid catabolism is upregulated in M2
macrophages to maintain oxidative phosphorylation (OXPHOS)^[165]55. In
this study, the heatmap showed that lipid catabolic process and Gpx
activity related genes, which prevented fatty acid (FA) accumulation in
cells, were upregulated in the BNP-PEDOT-PSF-AG group. While, genes
related to lipid biosynthesis process, which caused FA accumulation in
BNP-PEDOT-PSF-AG group, were downregulated in the BNP-PEDOT-PSF-AG
group (Fig. [166]5e). In addition, M1 related genes were downregulated
in PEDOT-PSF-AG and BNP-PEDOT-PSF-AG groups (Fig. [167]5e). Notably, M2
related genes were upregulated in BNP-PEDOT-PSF-AG group, which was
accordance with our in vitro results. Hence, we speculated that H[2]S
and bioelectricity modulate macrophage polarization through regulation
of lipid metabolism (Fig. [168]5f).
Antioxidative and immunomodulatory activities of the BNP-PEDOT-PSF-AG
hydrogel in vivo
To further explore the in vivo antioxidative and immunomodulatory
activities of the hydrogels in the context of periodontitis,
periodontitis rats with diabetes were generated through intraperitoneal
injection of streptozotocin (STZ) and establishment of ligature-induced
periodontitis for 4 weeks (Fig. [169]6a, Supplementary Fig. [170]13b).
An incision was made in the gingiva on the mesial palatal side of the
first molar, and hydrogel was implanted under the gingival flap
(Supplementary Fig. [171]13c). Under diabetic conditions, oxidative
stress is elevated, which exacerbates periodontal disease and
negatively impacts periodontal regeneration^[172]56. Moreover, the
undermined innate antioxidative defense system is unable to reverse the
increase in oxidative stress^[173]57. Hence, eliminating excessive ROS
production and correcting the imbalanced oxidative stress is crucial
for treatment of diabetes-associated periodontal complications^[174]58.
In order to investigate the antioxidative effect of hydrogels in
periodontitis in diabetic environment, the extent of lipid peroxidation
and the activity of antioxidative enzymes were detected. The level of
oxidative stress was significantly increased in diabetic periodontitis
compared to normal periodontium (Supplementary Fig. [175]13f and g).
After hydrogel implantation, the ratio of GSH to GSSG, which represents
the antioxidative ability, increased in all groups, revealing the
therapeutic effect of scaling before hydrogel implantation
(Fig. [176]6b). Specifically, the ratios of GSH to GSSG were higher in
the PSF-AG, BNP-PSF-AG, PEDOT-PSF-AG, and BNP-PEDOT-PSF-AG groups
compared to the blank and pure AG groups, demonstrating the
antioxidative ability of PSF. The BNP-PEDOT-PSF-AG (2.82 ± 0.26) group
exhibited highest GSH to GSSG ratio of all, indicating a synergistic
effect of PEDOT-PSF and H[2]S in alleviating oxidative stress.
Furthermore, the examination of Gpx, a member of the innate antioxidant
defense system, also showed the same trend. Notably, the level of Gpx
was nearly twice as high in the BNP-PEDOT-PSF-AG group than in the AG
group (Fig. [177]6c), indicating the excellent antioxidant effect of
PEDOT-PSF and H[2]S. Taken together, these results established that
BNP-PEDOT-PSF-AG hydrogel facilitates the metabolism of excess
oxidation products in the microenvironment of diabetic periodontitis,
which can be attributed to the synergistic effects of PEDOT-PSF and
H[2]S. In addition, the hydrogels did not cause systemic toxicity
(Supplementary Fig. [178]14).
Fig. 6. Antioxidative and immunomodulatory activities of the BNP-PEDOT-PSF-AG
hydrogel in vivo.
[179]Fig. 6
[180]Open in a new tab
a The schedule of construction of periodontic rat with diabetes and
hydrogel implantation. b GSH/GSSG ratio in defect area (mean ± SEM,
n = 3 independent biological samples, one-way ANOVA). c Glutathione
peroxidase level in defect area (mean ± SEM, n = 3 independent
biological samples, one-way ANOVA). d Immunofluorescence staining of
iNOS (red) and CD68 (green) double-stained M1 macrophages on the defect
area. e Immunofluorescence staining of CD163 (red) and CD68 (green)
double-stained M2 macrophages on the defect area. f TRAP staining of
the defect area at 4 weeks after hydrogels implantation.
To further explore the in vivo immunomodulatory activity of the
hydrogels, the phenotypes of macrophages in periodontal tissue were
detected after implantation of hydrogels. As shown in Fig. [181]6d, e
and Supplementary Fig. [182]15a, the numbers of M1 macrophages in the
PEDOT-PSF-AG (4.88 ± 0.76%), BNP-PSF-AG (5.69 ± 1.26%), and
BNP-PEDOT-PSF-AG (1.98 ± 0.26%) groups were significantly decreased
compared to the blank (18.22 ± 1.62%) and pure AG (13.24 ± 1.21%)
groups, representing the ability of PSF to inhibit M1 macrophage
polarization. In our analysis of M2 macrophages, there were no
significant differences among the blank (1.54 ± 0.36%), AG
(2.40 ± 0.54%), and PSF-AG (2.47 ± 0.28%) groups. Increased M2
macrophages were observed in the PEDOT-PSF-AG (5.92 ± 1.16%),
BNP-PSF-AG (6.56 ± 0.67%), and BNP-PEDOT-PSF-AG (8.76 ± 0.90%) groups.
Specifically, the BNP-PEDOT-PSF-AG group exhibit the highest proportion
of M2 macrophages (Fig. [183]6e and Supplementary Fig. [184]15a). These
trends were in accordance with the in vitro results, which verified the
synergistic effects of PEDOT-PSF and H[2]S on immunomodulatory
activity.
In the process of bone remodeling, the balance between osteoblasts and
osteoclasts maintains the functional structure of bone. Osteoclasts are
active in the early stage of bone repair. However, due to the excessive
inflammation in the diabetic environment, a high level of osteoclast
activity remains, which negatively impacts functional bone
repair^[185]11,[186]59. Osteoclast activity in different groups was
evaluated using TRAP staining (Fig. [187]6f). The numbers of TRAP^+
cells were obviously decreased in the BNP-PEDOT-PSF-AG, PEDOT-PSF-AG,
BNP-PSF-AG, and PSF-AG groups compared to the blank and AG groups; the
BNP-PEDOT-PSF-AG group had the lowest number, indicating synergistic
effects of PEDOT-PSF and H[2]S strongly inhibited osteoclast activity
(Fig. [188]6f and Supplementary Fig. [189]15b).
The BNP-PEDOT-PSF-AG hydrogel promotes osteogenesis in diabetic periodontitis
PDLSCs possess multilineage differentiation potential and can
differentiate into various tissues such as osteoid tissue, periodontal
ligament-like connective tissue, and cementoid tissue, all of which
play important roles in functional periodontal regeneration^[190]60.
Therefore, we analyzed the number of PDLSCs (CD90^+ cells) in
conjunction with six types of hydrogels to investigate the response of
host stem cells to hydrogels during periodontium regeneration. At day
7, the numbers of CD90^+ cells in the BNP-PSF-AG, PEDOT-PSF-AG, and
BNP-PEDOT-PSF-AG groups were significantly higher than in the blank and
AG groups. Notably, significantly more CD90^+ cells were observed in
the BNP-PSF-AG (5.43 ± 0.46%) group compared to the PSF-AG group
(2.83 ± 0.46 %) (Fig. [191]7a). Similarly, the number of CD90+ cells in
the BNP-PEDOT-PSF-AG group (11.11 ± 0.97%) were significantly higher
than that in the PEDOT-PSF-AG group (4.61 ± 0.53%), which was in
accordance with the finding that the former group had the highest
migration rate of PDLSCs in an in vitro wound scratch assay
(Supplementary Fig. [192]8). These results indicated that H[2]S is
excellent at recruiting PDLSCs, which is essential in the early stage
of periodontal regeneration.
Fig. 7. Immunofluorescence staining of defect areas.
[193]Fig. 7
[194]Open in a new tab
a Immunofluorescence staining of CD90 at 7 days after operation in
defect area in various groups and quantification analysis (mean ± SEM,
n = 6 independent biological samples, one-way ANOVA).
Immunofluorescence staining of angiopoiesis-related protein CD31 (b)
and α-SMA (c) at 7 days after operation in defect area and
quantification analysis (mean ± SEM, n = 6 independent biological
samples, one-way ANOVA). Immunofluorescence staining of
osteogenesis-related protein RUNX2 (d) and OCN (e) at 4 weeks after
operation in defect area and quantification analysis (mean ± SEM, n = 6
independent biological samples, one-way ANOVA).
It is well known that blood vessels and bone exhibit a close
relationship in their spatial position as well as in their function.
Cells, nutrients, oxygen, and growth factors needed for periodontal
healing are transported to damaged periodontium through the
blood^[195]61. Hence, angiogenesis is essential for periodontal
regeneration. The ability of the hydrogels to promote revascularization
was evaluated by the immunofluorescence of angiogenesis-related markers
CD31 (Fig. [196]7b) and α-SMA (Fig. [197]7c) after 7 days of hydrogels
implantation. The results showed that CD31 and α-SMA were highly
expressed in the BNP-PEDOT-PSF-AG, PEDOT-PSF-AG, BNP -PSF-AG, and
PSF-AG groups, representing positive effects of the anti-inflammatory
microenvironment on revascularization. Moreover, the expression levels
of CD31 in the BNP-PEDOT-PSF-AG group were significantly higher than in
the other four groups without BNP, indicating that H[2]S has excellent
ability to promote angiogenesis (Fig. [198]6b). Furthermore, at 4 weeks
post-surgery, the expression of osteogenesis-related factors (OCN and
RUNX2) in the PEDOT-PSF-AG and BNP-PEDOT-PSF-AG groups were
significantly higher than in the AG group, which is in accordance with
the in vitro results, representing good osteoinductive ability of
PEDOT-PSF (Fig. [199]7d, e). BNP-PSF-AG group also showed better higher
expression of osteogenesis-related factors than AG group, indicating
the ability of H[2]S in bone formation (Fig. [200]7d, e). Specifically,
the expression levels of OCN and RUNX2 in the BNP-PEDOT-PSF-AG group
were highest among all five groups (Fig. [201]7d, e). These results
indicated excellent revascularization and osteoinductive abilities of
H[2]S gas and PEDOT-PSF in the PSF-induced anti-inflammatory
microenvironment.
The BNP-PEDOT-PSF-AG hydrogel promotes integrated periodontal regeneration in
diabetic periodontitis
Micro-CT and histomorphological analysis were used to confirm the
periodontal tissue repair promoted by the hydrogels at 4 weeks after
surgical implantation of the hydrogels in the periodontitis rats with
diabetes described above. The Micro-CT images showed that the ratio of
bone volume to total volume in the BNP-PSF-AG (77.50 ± 1.56%),
PEDOT-PSF-AG (84.25 ± 4.52%) and BNP-PEDOT-PSF-AG (90.35 ± 3.18%)
groups was significantly increased compared to the blank (57.54 ± 1.71
%) and AG (60.17 ± 3.28%) groups at 4 weeks post-surgery (Fig. [202]8a,
b). Specifically, bone volume in BNP-PEDOT-PSF-AG was the highest among
all six groups (Fig. [203]8a, b, Supplementary Fig. [204]13e).
Furthermore, the distance between the root furcation and alveolar bone
crest, which represents the amount of bone formation, in the six groups
showed the same trend (Fig. [205]8c, d). These results confirmed the
synergistic effects of PEDOT-PSF and H[2]S in promoting bone
regeneration in periodontitis under diabetic environment.
Fig. 8. Periodontium regeneration in periodontitis under diabetic condition.
[206]Fig. 8
[207]Open in a new tab
a–d Micro-CT images of the rat maxillary first molar at the site of
implantation. Quantification analysis of BV/TV (b) and RF-ABC distance
(d) in the six groups (mean ± SEM, n = 4 independent biological
samples, one-way ANOVA). BV bone volume, TV total volume, RF root
furcation, ABC alveolar bone crest. e HE staining of the defect area at
4 weeks after implantation. AB alveolar bone. FT fiber tissue. R root.
f, g Masson’s trichrome staining of the defect area in various groups
and quantification analysis (mean ± SEM, n = 6 independent biological
samples, one-way ANOVA). h HE staining images of newly formed
periodontal ligament. in periodontal defects at 4 weeks post-operation.
The black angles indicated the angulation of newly formed periodontal
ligament. PDL periodontal ligament. i Analysis of the angular values of
newly formed ligaments for five groups, and dotted line represented
average angle of native mature ligament fibers (mean ± SEM, n = 5
independent biological samples, one-way ANOVA). j Schematic
illustration of the mechanism of the BNP-PEDOT-PSF-AG hydrogel in
promoting integrate periodontium regeneration under diabetic
periodontic condition. MSC mesenchymal stem cell, ROS reactive oxygen
species, MSC mesenchymal stem cell.
Histological analysis revealed that 4 weeks after hydrogel
implantation, soft tissue occupied the space of alveolar bone loss
under the root furcation in the AG and PSF-AG groups (Fig. [208]8e),
whereas the BNP-PSF-AG, PEDOT-PSF-AG and BNP-PEDOT-PSF-AG groups
exhibited more bone formation under the root furcation (Fig. [209]8e).
Furthermore, the maturation of new bone was evaluated by Masson
staining. The proportions of mature bone in both the PEDOT-PSF-AG
(1.00 ± 0.14) and BNP-PEDOT-PSF-AG (1.94 ± 0.21) groups were
significantly higher than in the AG (0.44 ± 0.07) group (Fig. [210]8f,
g). Specifically, the maturation of new bone in the BNP-PEDOT-PSF-AG
group was two times of that in the PEDOT-PSF-AG group. The superior
alveolar bone formation in the BNP-PEDOT-PSF-AG group revealed the
synergistic effects of H[2]S and PEDOT-PSF on hard tissue regeneration
in diabetic periodontitis.
The quality of periodontal fiber arrangement was reflected by the
angles of periodontal ligament (PDL) fiber bundles relative to the root
surface. Compared to the AG group (17.84 ± 3.31°), the PDL angles in
the PEDOT-PSF-AG (30.20 ± 1.65°) and BNP-PEDOT-PSF-AG (42.65 ± 3.65°)
groups were closer to the native mature angulation (48.21°)^[211]62,
which we attribute to the conductivity of PEDOT-PSF (Fig. [212]8h, i).
Furthermore, the mean fiber angulation of the PDL in the
BNP-PEDOT-PSF-AG group was higher than in the PEDOT-PSF-AG group
(Fig. [213]8h, i). These findings demonstrated that PEDOT-PSF and H[2]S
not only promote bone formation, but also guide orderly arrangement of
the PDL fibers, which is essential for functional PDL regeneration.
The excellent performance of the BNP-PEDOT-PSF-AG hydrogel in
integrated periodontal regeneration in ta diabetic setting can be
attributed to the following mechanisms (Fig. [214]8j): (1) Local
inflammation and oxidative stress are alleviated by the hydrogels
through excellent ROS-scavenging of PEDOT-PSF and H[2]S. Subsequently,
osteoclastogenesis is significantly reduced in the periodontal tissue.
(2) The synergistic effect of PEDOT-PSF and H[2]S modulates the local
macrophage polarization through regulation of lipid metabolism. (3)
MSCs are recruited to the periodontal tissue in a timely manner by the
sustained release of H[2]S from BNPs and then attach to the hydrogel
thanks to the synergistic effect of PSF. Sufficient stem cells in the
region ensure the subsequent angiogenesis. (4) Activated bone formation
indued by synergetic effect of PEDOT-PSF and H[2]S via promoting
autophagy in PDLSCs. (5) PEDOT-PSF endows the hydrogel with good
conductivity. The conductive hydrogels transmit endogenous
bioelectricity and open Ca^2+ ion channels under electric stimulation.
The inflow of Ca^2+ derived from AG hydrogels potentiate osteogenesis
in situ. (6) The conductive hydrogels improve cell arrangement,
enabling functional PDL regeneration. Together, these factors
synergistically contribute to improving the periodontal environment and
potentiating integrated periodontal regeneration under the conditions
of diabetic periodontitis.
Discussion
In this study, the hydrogel was engineered by incorporating conductive
PEDOT-PSF and a BNP-based H[2]S sustained-release system into a
physiochemical dual-crosslinked AG network. Compared with traditional
biomaterials and biological modulation techniques, this study targeting
the pathological microenvironment in diabetic periodontitis, creates a
polyphenol-mediated redox-active hydrogel that couple endogenous
bioelectricity and H[2]S gas to reverse hyperglycemic inflammatory
microenvironment. The hydrogel exhibits a number of advantages for
periodontitis treatment in diabetes patients. First, the ultralong,
uniformly dispersed PEDOT-PSF microfibers enhance the mechanical
properties of the hydrogel, which meet the desired mechanical criteria
for a periodontal local implant material. Second, PEDOT-PSF confers the
hydrogel with excellent antioxidation and anti-inflammatory properties.
With the synergistic effect of H[2]S, the hydrogel exhibits
immunomodulatory activity, which in turn converts inflammatory
periodontium into a microenvironment favorable for regeneration. In
addition, sustained release of H[2]S from BNPs recruits MSCs to the
periodontal tissue, promoting angiogenesis and osteogenesis. More
importantly, with the PEDOT-PSF-facilitated conductive pathways,
endogenous bioelectricity is transmitted to cells, which enhances Ca^2+
influx and improves cell arrangement. Subsequently, the inflow of Ca^2+
derived from the AG hydrogel potentiates osteogenesis in situ.
Accordingly, the hydrogel achieves integrated periodontal bone healing
through the synergistic effects of electrical conductivity, alleviation
of oxidative stress, reduction of inflammation, immunomodulation, MSC
recruitment, and osteogenesis. Therefore, this strategy for the
coupling of H[2]S gas and electrobiological activity could be a
prospective strategy for the treatment of periodontitis in diabetes.
Methods
This study was approved by the Research Ethics Committee of West China
Hospital of Stomatology (WCHSIRB-D-2021-163). The study design and
conduct complied with the Department of Health and Human Services, the
Declaration of Helsinki, and all relevant ethical regulations. Written,
informed consent was provided by all patients before enrollment. All
animal experiments were performed according to protocols approved by
the Research Ethics Committee of West China Hospital of Stomatology
(WCHSIRB-D-2020-485).
Fabrication of PEDOT-PSF
First, PSF was fabricated through a PDA-Assisted Protection-Extraction
Process. Silk cocoons was boiled in a Na[2]CO[3] solution (pH = 9) for
30 min, and stirred (room temperature) for 60 min. Then, the dried and
degummed mSF (0.6 g) were immersed and stirred in dopamine (DA)
tris-HCl buffer (0.05 M, 180 mL, pH = 8) for 10 h. Afterward, the
solution was rested with half of the solution replaced to DI water.
Subsequently, NaOH (0.42 g) was added in the solution (70 °C) for 3 h
and sonicated for 1 h. Finally, the PSF was obtained after
centrifugation for 3 times^[215]63.
Then, PEDOT-PSF was prepared through Assisted PEDOT Assembly. PSF
(0.05 g), EDOT (200 μL), and FeCl[3]·6H[2]O (1 g) were dispersed in
50 mL EtOH/H2O (3:1) solution. After 3 h, the PEDOT-PSF was obtained
after centrifugation, washing, and drying^[216]63.
Syntheses of NaHS@BNP
BNPs were fabricated by the desolvation method^[217]64. BSA (0.6 g)
were added to distilled water (30 mL). NaHS (0.03 g) were dissolved in
Ethanol (30 mL) to form the NaHS/Ethanol complex. Then the NaHS/Ethanol
complex was added to BSA solution at rate of 0.5 mL/min. 18.75 mg
N-(3-dimethylaminopropyl)- N’-ethylcarbodiimide hydrochloride (EDC) was
added to the mixture for BNPs coupling. After stirring for 12 h, the
NaHS-loaded BNPs were collected by centrifugation (19,760 × g, 15 min,
4 °C).
Preparation of the BNP-PEDOT-PSF-AG hydrogel
For synthesis of the BNP-PEDOT-PSF-AG hydrogel, 0.8 g of gelatin were
dissolved in 10 ml 2 wt. % sodium alginate. The solution was stirred
for 1 h at 65 °C until the gelatin powders were completely dissolved.
0.02 g PEDOT-PSF and 0.01 g BNPs were then added to the mixture, 600 μL
PEGDE was added to the mixed solution to cross-link the gelatin. After
24 h cross-link, the hydrogel was soaked in 5% CaCl[2] solution for 1 h
to further cross-link of the sodium alginate. For comparison, we
prepared the PEDOT-PSF-AG hydrogels, BNP-PSF-AG hydrogels, PSF-AG
hydrogels and AG hydrogels using the same assembly processes. The
detailed contents are listed in Supplementary Table [218]1.
Characterizations of hydrogels
The morphology characteristics and structure of mSF, PSF, PEDOT-PSF,
NaHS@BNP, and the BNP-PEDOT-PSF-AG hydrogel were observed by scanning
electron microscopy (SEM, JSM 6390, JEOL). The H[2]S release profiles
were conducted by immersing the BNP-PEDOT-PSF-AG hydrogels in PBS at
37 °C. Then, H[2]S releasing profile was measured by endogenous
hydrogen sulfide (H[2]S) assay kit (Catalog number: A146-1-1, Nanjing
JianCheng Bioengineering Institute).
XPS analysis for recharging reduction of PSF, PEDOT-PSF, and H[2]S/PEDOT-PSF
X-ray photoelectron spectroscopy (XPS; Kratos, Axis Ultra DLD, DK) was
used to investigate the reduction of quinone groups into phenolic
hydroxyl groups. PSF, PEDOT-PSF, and H[2]S treated PEDOT-PSF were
tested. The H[2]S treated PEDOT-PSF were prepared by soaking PEDOT-PSF
in NaHS solution for 30 min. A monochromatic Al Kα X-ray source
(hν = 1486.6 eV) running at 15 kV and 150 W was used. Regarding the
PSF, the O 1 s spectrum was fitted to 2 components: the hydroxyl group
C-O at ~532.8 eV and the carbonyl group C = O at ~531.9 eV.
Mechanical property test
The compressive and tensile properties of hydrogel were measured by the
universal testing machine (Instron 5567, Boston, MA, USA). The
compressive strength, tensile strength strain at fracture were
subsequently statistically analyzed. Tensile tests of the samples
(width: 15 mm, length: 30 mm, thickness: 2 mm) were performed at an
extension speed of 10 mm/min. The cyclic stress-strain test of
BNP-PEDOT-PSF-AG hydrogels (10 mm (height) × 15 mm (diameter)) were
performed at a speed of 5 mm/min; the maximum strain was 80%.
Conductivity measurement
The conductivities of the BNP-PEDOT-PSF-AG hydrogels (10 mm
(height) × 15 mm (diameter)) were measured by a two-probe method using
electrochemical system (CHI 660, Chenghua, China), according to our
previous study^[219]63. The conductivity (κ, S/m) was calculated by
[MATH: κ=IVLA, :MATH]
where A and L are the cross-sectional area and height of the hydrogel,
respectively. Conductivity of hydrogels with different contents of
PEDOT-PSF and compression also exhibited by light-emitting diode (LED)
test.
Cell culture
After administration of local anesthesia, the premolars were loosed and
removed from its socket. 20 extracted premolars from 10 patients were
obtained and immediately immersed in PBS containing 5%
penicillin/streptomycin (Catalog number: P1400, Solarbio) at 4 °C. Then
periodontal ligament was isolated from medium 1/3 of root^[220]65.
After digested with collagenase I (Catalog number: 8140, Solarbio) at
37 °C for 1 h, tissues and cells were cultured with α‐MEM medium
containing 10% fetal bovine serum (Catalog number: 100–700, Gemini) and
1% penicillin/streptomycin. PDLSCs were used at passage 2–4. All tests
performed under high glucose and inflammatory conditions (33 mM
glucose,100 μg/L LPS).
Transcriptome sequencing and data analysis
PDLSCs were cultured with the BNP-PEDOT-PSF-AG, PEDOT-PSF-AG, and
PSF-AG hydrogels under high glucose and inflammatory conditions (33 mM
glucose,100 μg/L LPS) for 3 days with electrical stimulation (600mV).
THP-1 cells were cultured with the BNP-PEDOT-PSF-AG, PEDOT-PSF-AG, and
AG hydrogels under high glucose and inflammatory conditions (33 mM
glucose,100 μg/L LPS) for 3 days. The PDLSCs and THP-1 cells were then
treated with Trizol reagent (Invitrogen) and stored at −80 °C before
sequencing. RNA sequencing was performed using Illumina HiSeq X10
(Illumina). Bioinformatic analysis was performed using the OmicStudio
tools ([221]https://www.omicstudio.cn/tool).
High-throughput electrical stimulation of PDLSCs
Different ES levels (0, 300 and 600 mV) were applied to PDLSCs on
hydrogels for 30 min per day through a home-made high-throughput ES
device^[222]63,[223]66. After 3 days application of ES, cell
proliferation and cell morphologies were measured. The cells were
stained with cyanine dye 3 (Cy3)-conjugated phalloidin diluted in PBS
with 1% BSA (F-actin; 1:500, Catalog number: 23111, ATT Bioquest). The
cell spreading area of PDLSCs on the different hydrogels was analyzed
using the ImageJ software. The cell viability was quantified using a
3-(4,5-dimethylthiazol-2-yl)-2,5- diphenyltetrazolium bromide (MTT)
assay.
RNA extraction and quantitative real-time polymerase chain reaction (PCR)
After application of ES, total RNA was extracted from cell samples
using RNA Extraction Kit (Tsingke). Then synthesized complementary DNA
(cDNA) were gained using HiScript III RT SuperMix for RT-PCR (Vazyme).
The primers used for target genes purchased from Tsingke are listed in
Supplementary Table [224]2. All samples were performed in triplicate.
The relative mRNA expression was normalized to GAPDH and calculated
using the 2 − ΔΔCt method.
Intracellular calcium measurement
PDLSCs were seeded on the BNP-PEDOT-PSF-AG hydrogel and cultured at
37 °C for three days, followed by treatment with 30 min ES (600 mV) or
without ES per day. Then cells were incubated with Fluo-8 AM (Catalog
number: [225]B21080, AAT Bioquest) for 30 min at 37 °C and collected
for flow cytometry test (FACSAriaIII, BD).
Cellular ROS scavenging activity
A reactive oxygen species assay kit (2′,7′-dichloro fluorescein diac-
etate, DCFH-DA; Catalog number: S0033S, Beyotime) was used to test the
cellular ROS scavenging activity. THP-1 cells were cultured with AG,
PSF-AG, BNP-PSF-AG, PEDOT-PSF-AG, or BNP-PEDOT-PSF-AG hydrogels in
12-well plates for 12 h. After incubation with a DCFH-DA probe for
20 min, the ROS levels were measured using CLSM (NI-E, Nikon) and flow
analyzer (FACSAriaIII, BD).
Polarization of macrophages
THP-1 cells were cultured with the BNP-PEDOT-PSF-AG, PEDOT-PSF-AG,
BNP-PSF-AG, PSF-AG, and AG hydrogels under high glucose and
inflammatory conditions (33 mM glucose,100 μg/L LPS) for 3 days. The
cells were incubated with BV421 Mouse Anti-Human CD86 diluted in PBS
(1:20; Catalog number: 562432; BD Biosciences) and FITC Mouse
Anti-Human CD206 diluted in PBS (1:6; Catalog number: 551135; BD
Biosciences) for 1 h at room temperature and analyzed using a flow
analyzer (FACSAriaIII, BD, USA).
Establishment of diabetic periodontic rat model and hydrogel implantation
145 SD rats (male; 200–220 g; 8 weeks) were intraperitoneally injected
with 55 mg/kg streptozocin (STZ; Sigma) to establish a diabetic model.
In order to avoid experimental differences caused by sex, rats of the
same sex were used. After 1 week, the maxillary first molars of the rat
were ligated with a 3-0 silk suture for 4 weeks to establish
periodontitis. After palatal incision of the maxillary first molars and
flap operation, AG, PSF-AG, BNP-PSF-AG, PEDOT-PSF-AG, or
BNP-PEDOT-PSF-AG hydrogels (Φ 3 × 2 × 0.5 mm^3) were implanted into
operation area to evaluate the osteoinductivity in vivo. Then the
gingivae were sutured to fix the hydrogels. The rats were euthanized 7
and 28 days later. The samples were fixed for histology,
immunofluorescence, and micro-CT assays.
Micro-computed tomography (μCT) and histology analysis of the periodontal
repair
The harvested maxillaries were scanned with μCT (μCT50, SCANCO Medical)
at 80 kV, 500 µA, and with a spatial resolution of 15 μm. The built-in
software of the microCT imaging system was applied for the analysis of
the volume of newly formed bone. After μCT analysis, the fixed
maxillaries were assigned to histological analysis. The tissue sections
were prepared into 4 μm thickness parallel to the long axis of the
skull. Hematoxylin/eosin (HE) and Masson trichrome staining were
conducted to show the periodontal regeneration situation. The
quantified analysis of the bone formation in Masson stain was
illustrated by the ratio of mature bone to the osteoid matrix by the
ImageJ 1.48 v software. Additionally, tartrate‐resistant acid
phosphatase (TRAP) staining labeled activated osteoclasts.
For immunofluorescence staining, the fixed samples were incubated in
30% sucrose, embedded in Tissue-Tek, and then cut into 8-μm-thick
sections parallel to the long axis of the skull. The slides were
incubated overnight (4 °C) with the following primary antibodies
diluted in universal antibody diluent (Catalog number: WB500D; NCM
Biotech): rat anti-CD68 (1:100; Catalog number: Ab53444; Abcam), rabbit
anti-iNOS (1:100; Catalog number: ab15323; Abcam), mouse anti-CD163
(1:100; Catalog number: sc-58965; Santa Cruz), rat anti-CD90 (1:100;
Catalog number: ab3105; Abcam), rabbit anti-CD31(1:500; Catalog number:
ab182981; Abcam), rabbit anti-α-SMA (1:100; Catalog number: ab5694;
Abcam), mouse anti-RUNX2(1:100; Catalog number: sc-101145; Santa Cruz),
and rabbit anti-OCN(1:100; Catalog number: DF12303；Affinity). After
three washes with PBS, the sections were incubated with secondary
antibodies of donkey anti-rat Alexa Fluor 488 (1:800; Catalog number:
A21208; Thermo Fisher Scientific), donkey anti-rabbit Alexa Fluor
555(1:800; Catalog number: A31572; Thermo Fisher Scientific), goat
anti-mouse Alexa Fluor 555 (1:800; Catalog number: A21423; Thermo
Fisher Scientific), donkey anti-mouse Alexa Fluor 488 (1:800; Catalog
number: A32766; Thermo Fisher Scientific), and donkey anti- rabbit
Alexa Fluor 488 (1:800; Catalog number: A32790; Thermo Fisher
Scientific) for 1 h at RT in dark. Cellular nuclei were labeled with
DAPI (Beyotime, Shanghai, China). Images were examined with confocal
(N-STORM & A1, NIKON).
Measurement of GSH/GSSG ratio and Gpx in periodontium tissue
Periodontal tissue was extracted from of maxillary first molars 4 weeks
after hydrogel implantation. Each piece was further ground with
sufficient liquid nitrogen. Then GSH/GSSG ratio and Gpx were measured
by GSH and GSSG Assay Kit (Catalog number: S0053, Beyotime) and Total
Glutathione Peroxidase Assay Kit with NADPH (Catalog number: S0058,
Beyotime).
Statistics and reproducibility
The sample size in each experiment is indicated in the figure or
corresponding figure legends. The SEM micrographs were repeated three
times independently with similar results. Comparisons for two groups
were conducted by a two-tailed Student’s t test. Comparisons among
groups were assessed by one-way analysis of variance (ANOVA) (for
multiple groups) followed by the Tukey-Kramer test. The results were
expressed as means ± SEM, P < 0.05 was considered statistically.
Reporting summary
Further information on research design is available in the [226]Nature
Portfolio Reporting Summary linked to this article.
Supplementary information
[227]Supplementary Information^ (1MB, pdf)
[228]Peer Review File^ (3.4MB, pdf)
[229]Reporting Summary^ (114.5KB, pdf)
Source data
[230]Source data^ (32.1MB, xlsx)
Acknowledgements