See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/346121585

The Infliximab-Based Self-Healing Hydrogel Composite Scaffold Enhances

Stem Cells Survival, Engraftment, and Function in Aiding Rheumatoid
Arthritis Management

Article in SSRN Electronic Journal · January 2020

DOI: 10.2139/ssrn.3687960

CITATIONS READS

5 155

10 authors, including:

Yue Zhao Hou Liu

Nanyang Technological University Sichuan University
44 PUBLICATIONS 2,148 CITATIONS 26 PUBLICATIONS 1,342 CITATIONS

SEE PROFILE SEE PROFILE

Hangrui Henry Liu Zuhao Li

UCSF University of California, San Francisco 39 PUBLICATIONS 2,192 CITATIONS
33 PUBLICATIONS 534 CITATIONS
SEE PROFILE
SEE PROFILE

All content following this page was uploaded by Hangrui Henry Liu on 07 December 2021.

The user has requested enhancement of the downloaded file.

JID: ACTBIO
ARTICLE IN PRESS [m5G;December 11, 2020;11:46]
Acta Biomaterialia xxx (xxxx) xxx

Contents lists available at ScienceDirect

Acta Biomaterialia
journal homepage: www.elsevier.com/locate/actbio

Infliximab-based self-healing hydrogel composite scaffold enhances

stem cell survival, engraftment, and function in rheumatoid arthritis
treatment
Yue Zhao a,c,1, Chaohua Gao b,1, Hou Liu a, Hangrui Liu c, Yubin Feng a, Zuhao Li b, He Liu b,∗,
Jincheng Wang b,∗, Bai Yang a, Quan Lin a,∗
a
State Key Lab of Supramolecular Structure and Materials, College of Chemistry, Jilin University, Changchun130012, China
b
Orthopaedic Medical Center, the Second Hospital of Jilin University, Changchun130041, China
c
RC Centre of Excellence for NanoscaleBioPhotonics, Department of Physics and Astronomy, Macquarie University, Sydney, NSW2109, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Rheumatoid arthritis (RA) is a severe inflammatory autoimmune disease, but its treatment has been very
Received 25 August 2020 difficult. Recently, stem cell-based therapies have opened up possibilities for the treatment of RA. How-
Revised 27 November 2020
ever, the hostile RA pathological conditions impede the survival and differentiation of transplanted cells,
Accepted 2 December 2020
and it remains challenging to fabricate a suitable biomaterial for the improvement of stem cells survival,
Available online xxx
engraftment, and function. Here we construct anoptional scaffold for RA management through the in-
Keywords: tegration of 3D printed porous metal scaffolds (3DPMS) and infliximab-based hydrogels. The presence
Inorganic-organic composite of rigid 3DPMS is appropriate for repairing large-scale bone defects caused by RA, while the designed
Self-healing hydrogels infliximab-based hydrogels are introduced because of their self-healable, anti-inflammatory, biocompat-
Biomimetic scaffolds ible, and biodegradable properties. We demonstrate that the bioengineered composite scaffolds support
Bone regeneration adipose-derived mesenchymal stem cells (ADSCs) proliferation, differentiation, and extracellular matrix
Stem cell-based therapies
production in vitro. The composite scaffolds, along with ADSCs, are then implanted into the critical-sized
bone defect in the RA rabbit model. In vivo results prove that the bioengineered composite scaffolds are
able to down-regulate inflammatory cytokines, rebuild damaged cartilage, as well as improve subchon-
dral bone repair. To the best of the authors’ knowledge, this is the first time that using the antirheumatic
drug to construct hydrogels for stem cell-based therapies, and this inorganic-organic hybrid system has
the potential to alter the landscape of RA study.
© 2020 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

Statement of significance
4. The composite scaffolds have abilities of enhancing ADSCs
survival, engraftment, and function in aiding rheumatoid
1. A new scaffold for rheumatoid arthritis management was
developed through the integration of infliximab-based hy- arthritis management.
drogels and 3D printing porous metal scaffolds.
2. The composite scaffold scan modulate the survival and
differentiation of ADSCs in vitro.
3. The composite scaffolds, along with ADSCs, are able to de- Introduction
press inflammatory cytokines, rebuild damaged cartilage,
as well as improve subchondral bone repair in vivo re- Rheumatoid arthritis (RA) is a severe inflammatory autoim-
sults. mune disease that affects about 25 million people among world
regions. Once RA occurs, the synovium becomes hyperplasic, vari-
ous cytokines, chemokines, as well as autoantibodies are produced
[1–3]. If insufficiently treated, RA can lead to systemic complica-
∗ tions, progressive destruction of joints, irreversible disability, and
Corresponding authors.
E-mail addresses: heliu@jlu.edu.cn (H. Liu), wangjinc@jlu.edu.cn (J. Wang), even a reduction in life expectancy. Although the exact etiology
linquan@jlu.edu.cn (Q. Lin). of RA is not fullyelucidated, research on this issue will be signifi-
1
These authors contributed equally to this work. cant for RA management [4,5]. Recently, stem cell-based therapies

https://doi.org/10.1016/j.actbio.2020.12.005
1742-7061/© 2020 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

Please cite this article as: Y. Zhao, C. Gao, H. Liu et al., Infliximab-based self-healing hydrogel composite scaffold enhances stem cell
survival, engraftment, and function in rheumatoid arthritis treatment, Acta Biomaterialia, https://doi.org/10.1016/j.actbio.2020.12.005
JID: ACTBIO
ARTICLE IN PRESS [m5G;December 11, 2020;11:46]

Y. Zhao, C. Gao, H. Liu et al. Acta Biomaterialia xxx (xxxx) xxx

Scheme 1. Schematic illustration showing the fabrication process of self-healing infliximab-based hydrogels combined with 3DPMS to deliver ADSCsin aiding RA manage-
ment.

open alternative options for the treatment of RA due to their im- a hydrogel via noncovalentsupramolecular interactions (e.g., hy-
munomodulation and tissue repair capabilities [6,7]. Among var- drophobic interactions, π -π stacking, Van der Waals forces, and
ious types of pluripotent stem cells, adipose-derived mesenchy- hydrogen bonds) [25,27–29]. As a result, this formulation hydrogel
mal stem cells (ADSCs) are an appealing cell source for regener- not only has advantages of traditional hydrogels, but also displays
ative medicine. Compared with other kinds of stem cells, ADSCs notable advantages in the treatment of various diseases. Nonethe-
can be easily procured from healthy donors by minimally inva- less,to the best of our knowledge,there has been no report on the
sive access. Moreover, they are promising candidates for clinical use of antirheumaticdrugs to construct hydrogels. Distinguished
applications, since such cell transplantation is regarded as auto- from the conventionaldrug delivery concept whereantirheumatic
transplantation and without ethical problems [8–10]. Although sig- drug molecules loaded in hydrogels andthe hydrogels just func-
nificant achievements have been achieved for stem cell-based RA tion as reservoir systems, this developed antirheumatic drug-based
management, widespread application of this method has been lim- hydrogel acts as topical medication and regulates the hostile mi-
ited. First of all, the hostile RA pathological conditions (e.g., over- croenvironment of transplant sites.
production of protease, intense inflammation, and high levels of Moreover, advanced treatments using biological therapeutic an-
reactive oxygen species) impede the survival and differentiation of tibodies, such as infliximab, have been proven advantageousin the
transplanted ADSCs [8,11]. Therefore, to avoid situations where the treatment of RA. This therapeutic antibody that works by depress-
inflammatory environment affectscell activity, desirable biomateri- ing the effects of tumor necrosis factor alpha (TNF-α ), has abilities
als for ADSCs delivery should work even under the chronic inflam- to reduce inflammation, depressing disease activity, andslowing
matory microenvironment. Concomitantly, in order to meet the cartilage/bone destruction [30]. In this work, we focused on the de-
complex structure of bone defect in RA, biomaterials must possess velopment of infliximab-based hydrogels for regulating the hostile
good processability and appropriate mechanical properties. Bio- microenvironment of the RA site. Previously, our group developed
compatibleis also another important criterion, which canovercome an inorganic-organiccomposite scaffold to aid RA managementby
stem cells poor migration and uncontrolled differentiation prob- integratingstiff 3D printed porous metal scaffolds (3DPMS) and soft
lems. Thus, there is an urgent need to explore a delivery system, multifunctional polysaccharide hydrogels [31]. In this work, we
which will enable ADSCs to more effectively promote RA manage- fabricated a self-healing hydrogel from dynamic reactions of mod-
ment. ified polysaccharides and infliximab (Scheme 1). The infliximab-
Hydrogels are generally fabricated by the cross-linking of hy- based hydrogels can modulate the survival and differentiation of
drophilic polymer chains within an aqueous microenvironment ADSCs in vitro, which was integrated with 3DPMS to enhance
[12–16]. These biomaterials are characterized by good biocompat- ADSCs survival, engraftment, and function in aiding RA manage-
ibility, water-rich nature, tunable mechanical property, and tissue- ment.In vivo experiments demonstrate that the composited scaf-
like elasticity [17–19]. With the above special physicochemical and fold can significantly inhibit inflammatory cytokines, rebuild dam-
biological properties, hydrogels provide important applications in aged cartilage, as well as improve subchondral bone repair in rab-
many fields, particularlyas vehicles for therapeutic cells, drugs, and bit RA animal model after 3 months. Our results demonstrate that
other bioactive molecules delivery [20,21]. Recently, there has been the concept of exploiting self-healing anti-inflammatory hydrogel
increasing interest in developing hydrogels which were made from can enhance ADSCs survival, engraftment, and function in aiding
therapeutic agents [22–26]. In this type of hydrogel, the thera- RA management, and broaden stem cell-based therapies on clini-
peutic agent itself is chemically crosslinked or self-assembled into cal applications.

2
JID: ACTBIO
ARTICLE IN PRESS [m5G;December 11, 2020;11:46]

Y. Zhao, C. Gao, H. Liu et al. Acta Biomaterialia xxx (xxxx) xxx

Experiment section Belgium). The percentage of bone volume out of ROI (BV/TV, Tb.
Th, Tb. Sp and BMD) wasreconstructed and calculated viasoftware
Materials NRecon version 1.6.6 (Bruker micro-CT N.V., Kontich, Belgium).

Sodium hyaluronate (Mw=90-100 KDa, Mw=10 0 0-20 0 0 KDa) Preparation of hydrogels

and adipic acid dihydrazide (ADH, ~98%) were from Yuanye
Biological Technology Co., Ltd (Shanghai, China). Sodium pe- (1) Preparation of Hyaluronic Acid-Hydrazide (HA-HYD): HA-
riodate (NaIO4 , ~99%), phosphate buffer solution (PBS), N-(3- HYD was prepared as previously described. Briefly, sodium
Dimethylaminopropyl)-N’-ethylcarbodiimide hydrochloride (EDC), hyaluronate (Mw=90-100 KDa, 1.0 g), adipic acid dihydrazide (9.0
ovalbumin (OVA, ~99%), p-nitrophenyl phosphate (pNPP) assay kit, g), and EDC (2.0 g) were dissolved into distilled water (100 mL),
bicinchoninic acid (BCA) assay kit, β -glycerol phosphate, ascor- and the pH of the mixed solution was adjusted to 4.5 with HCl
bic acid, alizarin red staining, ammonium hydroxide, RNA ex- (1 M). After that, the mixture solution was stirred for 24 hours
tract kit, PrimeScript RT reagent kit, Ketamine, bovine serum albu- under pH=4.5. After the process, the mixture was further dia-
min, DAPI, Masson staining, and Freund’s adjuvant were obtained lyzed (MWCO 3500) against deionized water for 3 d with repeated
from Sigma-Aldrich. Hematoxylin Eosin was from ThermoFisher change of water, followed by freezing and lyophilizing. Finally, the
Scientific Co., Ltd (Shanghai, China). Povidone iodine(KI, ~98.5%) pure HA-HYD powder was obtained and stored at 4°C for fur-
was obtained from Sinopharm Chemical Reagent Co., Ltd (Beijing, ther use [24]. The fabrication process of HA-HYD was illustrated in
China). Ethanol (C2 H5 OH, ~99.7 %) and ethylene glycol (~99%) were Scheme S1. (2) Preparation of Hyaluronic Acid-Aldehyde (HA-ALD):
given by Beijing Chemical Works (Beijing, China). Paraformalde- HA-ALD was synthesized as previously described. Briefly, sodium
hyde (~99%) and acrylic acid (~99%) were from Xilong Chemical hyaluronic acid (10 0 0-20 0 0 KDa, 1.0 g) and sodium periodate (1.0
Co., Ltd. Fetal bovine serum and streptomycin-penicillinwere pur- g) were dissolved in deionized water (100 mL) and magnetically
chased from Gibco Life Technologies (USA).Triton X-100 was pur- stirred at room temperature for 3 h in the dark. Subsequently, 10
chased from DingguoChangsheng Biotechnology Co. (Beijing, Peo- mL of ethylene glycol was added to the above mixture followed
ple’s Republic of China). Methyl methacrylate was purchased from by stirred for another 1 h to stop the reaction. After the process,
Alfa Aesar. Rhodamine-phalloidin was obtained from Molecular the mixture was further dialyzed (MWCO 3500) against deionized
Probes (Eugene, OR, USA). Verso cDNA synthesis kit was obtained water for 3 d with repeated change of water, followed by freezing
from Thermo Scientific (USA). Calcein-AM and cell counting kit-8 and lyophilizing. Finally, the pure HA-ALD powder was obtained
were obtained from Biobetimes Biotechnology Co., Ltd (Changsha, and stored at 4°C for further use [31,32]. The fabrication process of
China). Ti6 Al4 V (Sandvik, Sweden) powder was used for manufac- HA-ALD was illustrated in Scheme S2. (3) Hydrogels were formed
turing porous titanium alloy scaffolds by EBM (Q10, Arcam, Swe- immediately by the addition of HA-ALD (10 wt%, 500 μL) to the
den) as previously description. The porous scaffolds were based on mixed solution of HA-HYD(10 wt%, 500 μL) and different concen-
dodecahedron unit cells with the following design (nominal) di- tration of infliximab (0.5, 1.0, 2.0, 4.0, and 8.0 mg/mL). The gelation
mensions: strut size=300 mm, pore size=400 mm, porosity=70%. mechanism of hydrogels was illustrated in Scheme S3. When there
Disk-shaped scaffolds (diameter: 10 mm, height: 3 mm) were used are cells, 1 × 105 ADSCs were mixed with the solution of HA-HYD
for in vitro assays, and the columnar-shaped scaffolds (diameter: 5 (10 wt%, 160 μL) and infliximab (1 mg/mL), which were injected
mm, height: 10 mm) for in vivo study. After EBM processing, the into the 3DPMS. Then HA-ALD (10 wt%, 160 μL) was also injected
scaffolds were removed from the 3D printer and cleaned ultrason- into the 3DPMS. After a few seconds, hydrogels were formed im-
ically in acetone and ethyl alcohol for 20 min to remove powders mediately within the 3DPMS.
that were not melted during the EBM process. The ultrapure water
used throughout the experiments was purifiedwith a Milli-QA10
filtration system (Millipore, Billerica, MA, USA). In vitro infliximab release

Characterization The release of infliximab from hydrogels was determined. In

this experiment, infliximab labeled with FITC was used in detecting
SEM (JEOL JSM-6700F) was applied to observe the microstruc- protein concentration through the plate reader. Briefly, the sample
ture of hydrogels and other experimental scaffolds. All SEM sam- with 1 mg infliximab was immersed in 1 mL of phosphate buffer
ples were in dry state and sputtered with gold ions before obser- solution (0.01 M, pH 7.4) at 37°C for 28 days. The supernatant in
vation, and observed under anacceleration voltage of 3 kV. Rhe- each well was gathered and replaced with PBS at pre-determined
ological measurements were carried out on bulk gel samples us- time points (1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 14, 17, 21 and 28 days).
ing a rheometer (TA Instruments-waters LLC) at room temperature. Finally, the cumulative release amount of infliximab was evaluated
The universal testing machine (Landmark, MTS Inc. MN, U.S.A.) using Infliximab Elisa Kit.
was used to detect the push-out force of each scaffold in the
implant-bone interface. All stained samples were observed through Cell culture
a confocal microscope(TCS SP5, Leica, Germany). The absorbance
of samples was recorded by the spectrophotometric microplate Animal experiments were performed based on National Insti-
reader (Bio-Rad 680, USA). RT-PCR on Runx2, ALP, OCN and Osterix tutes of Health’s Guide for the Care and Use of Laboratory Ani-
were measuredby SYBR Premix Ex TaqII. For staining, the samples mals (NIH Publications No. 8023, revised 1978), and all animal ex-
were sliced through EXAKT systems (EXAKT Apparatebau, Norder- perimental protocols were approved by the Animal Care and Use
stedt, Germany). All photographs were taken by a digital cam- Ethics Committee of Jilin University.ADSCs were extracted from
era (Canon camera). The OD values were determined usingthe Mi- the subcutaneous fat pads of healthy New Zealand rabbits, and
croplate reader (ThermoFisher Scientific VariosbonFlsh).The confo- monocytes were isolated by Percoll discontinuous density gra-
cal microscopy images were observed and taken with an Olympus dient centrifugation. The isolated ADSCs were cultured in stan-
Fluoview FV10 0 0.Tissue slice images were measured using an op- dard α -MEM medium with 10% fetal bovine serum and 1% peni-
tical microscope (Olympus Optical Co., LTD). In order to assess the cillin/streptomycin at 37°C under 5% CO2 atmosphere. The medium
new bone formation, the samples were scanned using the micro- was changed every 3 days, and the 3 rd passage cells were used in
CT System (SkyScan 1076 scanner, Bruker micro-CT NV, Kontich, the following study [31,32].

3
JID: ACTBIO
ARTICLE IN PRESS [m5G;December 11, 2020;11:46]

Y. Zhao, C. Gao, H. Liu et al. Acta Biomaterialia xxx (xxxx) xxx

Cell viability Alizarin Red S (ARS) staining

To determine cell viability, a live/dead assay was carried out Alizarin Red S staining was applied to determine the miner-
based on the manufacturer’s instructions. In this assay, live cells alization of ADSCs. ADSCs (1 × 105 cells per well) were seeded
can be stained byCalcein-AM (green), while dead cells can be onto each group in 48-well culture plates. After 14 and 21
stained bypropidium iodine (PI)(red). In brief,ADSCs (1 × 105 cells days of osteogenic induction culture, ADSCs were fixed with 4%
per well) were incubated at 37°C under 5% CO2 atmosphere, and paraformaldehyde for 30 min, and then washed three times with
then seeded onto each experimental scaffold in 48-well plates, PBS. In order to remove salt residues, the fixed ADSCs were washed
respectively. At different time points, PBS solution (500 μL) con- with distillated water. Sequentially, osteoblasts in each group were
taining 4.0 mM PI and 2.0 mMCalcein-AM solution was added stained with40 mM ARS for 20 min (pH 4.1, 4°C), and rinsed three
into each experimental group. After cultured in dark for 30 min, times with distillated water. The ARS staining was imaged by an
the culture medium wastransferred out in eachsample, and ADSCs inverted microscope (CKX41). To quantify coloration of ARS, all
were rinsed with PBS to eliminate background interference.Then, samples were incubated with acetic acid (10%, v/v) solution for 30
the above samples were observed using a fluorescent microscope min (RT) under oscillation conditions of 50 rpm, and then trans-
[33]. ferred the solution containing osteoblasts (scraped from samples)
to a vial. Incubated (85°C, 10 min) these samples, cool on ice, and
then centrifuged. Then, the supernatant was neutralized with 10%
Cell proliferation ammonium hydroxide to neutralize excess acid. Finally, the ARS
concentrations were determined at 405 nm by a microplate reader
Cell proliferation could be expressed in terms of cell viability. (Bio-Rad 680, USA) [39,40].
The cell counting kit-8 (CCK-8) was done on days 1, 4, 7 and 14
days (n=6) according to the manufacturer‘s instructions. To sim-
ulate the RA microenvironment, the cultured medium was pre- Alkaline phosphatase (ALP) activity
pared by mixing 5mL complete medium (complete medium sup-
plemented with 10 mM β -glycerol phosphate and 50 mg/L ascor- To evaluate ALP activity of ADSCs in the early stage of dif-
bic acid) and 1mL blood serum of RA rabbits. ADSCs (1 × 105 cells ferentiation, an ALP colorimetric assay was carried out based on
per well) were incubated in the prepared cultured medium, and the manufacturer’s instructions. To simulate the RA microenviron-
then seeded onto each experimental scaffold in 48-well plates, re- ment, the cultured medium was prepared by mixing 5mL complete
spectively. Afterwards, CCK-8 solution (50 μL) was added into each medium (complete medium supplemented with 10 mM β -glycerol
group, and all groups were incubated at 37°C under 5% CO2 at- phosphate and 50 mg/L ascorbic acid) and 1mL blood serum of RA
mosphere. After incubation for 2 h, ADSCs were rinsed with PBS rabbits. ADSCs (1 × 105 cells per well) were incubated in the pre-
to eliminate background interference. Then the number of ADSCs pared cultured medium, and then seeded onto each experimental
seeded on each scaffold was counted by a microplate reader (Ab- scaffold in 48-well plates, respectively. The ALP assay of ADSCs was
sorbance at a wavelength of 450 nm) [33]. conducted after cultured for 4 and 7 days. In brief, at different time
points, cells were lysed by 1% Triton X-100 for 30 min (4°C), and
then centrifuged for 5 min (120 0 0 rpm) to collect the supernatant.
Cell morphology Subsequently, the ALP activity was measuredwiththe p-nitrophenyl
phosphate (pNPP)assay kit, and total intracellular protein wasmea-
To observe cell morphology, cultured ADSCs were fixed in 4% sured with thebicinchoninic acid (BCA) assay kit. The absorbance
paraformaldehyde (25°C, 10 min) and then permeabilized with of the yellow pnitrophenol(pNP) production and the total intracel-
0.15% Triton X-100 (25°C, 5 min). Blocking solution (1.0% bovine lular protein was detected at 490 nm and 570 nm through a spec-
serum albumin in PBS) was applied to block nonspecific bind- trophotometric microplate reader, respectively. The final ALP activ-
ing sites (25°C, 30 min). Afterwards, samples were washed with ity of ADSCs in each group can be calculated as pNPP per total
PBS for three times, and then rhodamine-phalloidin (200 nM, 200 intracellular protein content [41–43].
μL) was added to each sample (25°C, 30 min). Washing with
PBS 3 times, the samples were then stained with diamidino-2-
phenylindole (100 nM, 200 μL) for 30 min. The stained samples RA animal model establishment and in vivo bone regeneration study
were observed with a confocal microscope [34–37].
Forty 30-week-old female New Zealand white rabbits (about
2.5-3.5 kg) were used as in vivo experiments. Animal experiments
Quantitative real-time PCR (RT-PCR) for osteogenic markers were performed based on National Institutes of Health’s Guide for
the Care and Use of Laboratory Animals (NIH Publications No.8023,
The initial density of ADSCs in 48-well culture plates is revised 1978), and all animal experimental protocols were ap-
1 × 105 cells/well. To evaluate the osteogenic differentiation of AD- proved by the Animal Care and Use Ethics Committee of Jilin Uni-
SCs, the mRNA expression of osteogenic-related genes (Runx2, ALP, versity. Before implantation, the rabbit RA model was established
OPN, and OCN) was determined. The expression of Runx2, ALP and by weekly injection of ovalbumin/Freund’s complete adjuvant (20 g
Osterix was measured after incubation for 4 and 7 days, while OCN L−1 , 1 mL) mixed solution into the scapular area for 3 weeks. The
was measured after 7 and 14 days. At different time points, the to- fifth week after the first injection, the ovalbumin/Freund’s com-
tal RNA in ADSCs was isolated usingthe RNA extract kit, and RNA plete adjuvant mixed solution (20 g L−1 , 0.5 mL) was injected into
concentration wasdetectedby spectrophotometry. Using a Prime- the left knee joint cavity to induce RA.
Script RT reagent kit, the extracted RNA can be reverse-transcribed After successfully establishing the RA model, the RA rab-
into the first strand complementary DNA (cDNA). The Bio-Rad CFX bits were anesthetized by intravenous injection of Ketamine (50
Manager system was chosen to perform real-time PCR. The re- mg/kg), and all animal operations were performed under aseptic
lated primers of targeted gene are shown in Table S1 and the ex- conditions. The rabbit’s left knee was exposed as surgical site, and
pression of osteogenic related markers was normalized based on the skin was shaved and sterilized using 0.5% povidone iodine.
GAPDH. Relative expression level of each target gene was calcu- Skin, fascia, muscle, and drag patella ligament were separated until
lated through the formula 2−Ct [38,39]. its inward dislocation, and a 2 cm lateral incision through left knee

4
JID: ACTBIO
ARTICLE IN PRESS [m5G;December 11, 2020;11:46]

Y. Zhao, C. Gao, H. Liu et al. Acta Biomaterialia xxx (xxxx) xxx

was operated and exposed femoral condyle. Afterwards, a cylindri- the outer side was carefully moved away, and the collected sam-
cal defect (10 mm depth, 5 mm diameter) was produced using a ple was fixed with customized holder for the loading alignment.
drill, and then the experimental scaffold was implanted into the Then the push-out experiment was conducted with a constant rate
predrilled defect. To further prevent infection, penicillin (at a dose at 1.5 mm/min. The maximum push strength was recorded as the
of 0.1 g/kg) was given during surgery and postoperatively for 3 push-out force of each scaffold.
days. After surgery for 12 weeks, the RA rabbits were sacrificed un-
der anesthesia. Then femoral condyles with the experimental scaf-
Statistical analysis
fold were collected for further study [31].
All the results were expressed as the mean ± standard devia-
Histology and histomorphometry analysis
tion (SD). For multiple comparisons, a two-way analysis of variance
followed by Tukey’s post hoc was used for statistical comparisons.
All collected samples were dehydrated in serial concentrations
Differences were considered to be significant at a level of p< 0.05,
of ethanol (70, 85, 95, and 100%). After dehydration treatment,
0.01 and 0.001.
the above samples were embedded in methylmethacrylate (MMA)
solution that polymerized at 37°C within 1 week. Sections with
50 μm thickness were sliced through EXAKT systems. Then sliced 3. Results and discussion
samples were stained with hematoxylin and eosin (H&E) and Mas-
son staining. Finally, histological changes were recorded by a mi- Inspired by the inorganic-organic structure of the bone ma-
croscope, and the qualitative analysis of bone in-growth was per- trix, [45–47] we developed a composite scaffold from stiff3DPMS
formed with the system of digital image analysis (Image-Pro Plus and soft multifunctional hydrogels to aid RA management. To bet-
software, Silver Spring, MD) [38,44]. ter control cell behavior, a self-healing hydrogel was fabricated
from dynamic reactions of modified polysaccharides and inflix-
In vivo measurement of cytokines imab. Polysaccharides were selected as the template for hydrogel
synthesis because they are easy to chemically modify, biodegrad-
At 3 months, 2.0 mL of peripheral blood was collected from able, biocompatible, water-soluble, and abundant in nature [48–
the ear central artery using an EP tube containing 50 mL of hep- 50]. In our case, hyaluronic acid (HA)withwell-defined size ranges
arin sodium solution (10 0 0.0 IU/mL). The above blood was cen- was firstly chemically modified to contain either aldehyde (ALD)
trifuged, and cytokines (IL-1β , IL-6, TNF-α , and OVA-Ab) in serum or hydrazide (HYD) functional group, respectively. The above de-
were measured by ELISA Assay Kits. According to the manufac- signed polysaccharides were successfully synthesized as verified by
turer’s instructions, the standard solution was prepared in enzyme Fourier transform infrared spectroscopy (FT-IR) (Figure S1 and Fig-
coated plate. Then, the collected serum (10 μL) was added into the ure S2). The sol-to-gel phase transformation process was displayed
plate, and diluted into 50 μL. The above plate was incubated (37 in Fig. 1a. The mixture of prepared HYD-modified HA, (HA-HYD)
°C) for 30 min, and washed five times. Afterwards, the ELISA (50 and infliximab solution presented in the sol state at roomtempera-
μL) reagent was added into the above enzyme coated plate. The ture, and it quickly experienced sol-gel transitionafter mixing with
above plate was again incubated (37 °C) for 30 min, and washed ALD-modified HA (HA-ALD) solution. The gel formation mechanism
five times. Then 50 μL of reagent A and B were added into the is mainly due to a one-step condensation reaction, where theacyl-
above enzyme coated plate, and incubated in dark for 15 min. At hydrazone (hydrazine groups of HA-HYD reacted withHA-ALD) and
last, 50 μL of stop reagent was added into plate, which was mea- imine bonds (amino groups of infliximab reactedwith HA-ALD) are
sured at 450 nm through a spectrophotometric microplate reader. generated with water as the only byproduct [50]. Since the for-
Then the concentration of each cytokines was calculated through mation of the dynamic bonds requires a certain amount of time,
the standard curve [38,44]. the prepared hydrogel has injectable propertiesbefore solidifica-
tion. According to Fig. 1b, the mixed macromersolution can be in-
Microcomputed tomography (μCT) analysis jected through a 26-gauge (260 μm) needle, and then the hydro-
gel was rapidly formed within a few seconds at room tempera-
After 6 and 12 weeks of post-surgery, the rabbits were sac- ture. Rationally design of the injectable hydrogel could make it
rificed byintravenous injection of air under anesthesia and the serve as a suitabledeposit for therapeutic agent, and also provide
femoral condyles were collected. All specimens were fixed in 4% a biological tissue-like microenvironment [51,52]. Clinically, inflix-
paraformaldehyde in PBS for one week, and then scanned using imab is usually administered intravenously for the treatment of
the micro-CT System to appraise the new bone formation. The X- RA, while the traditional systemic injection is not quite effective in
ray source voltage was set at 102 kV and the beam current at bone tissue engineering [53]. In addition, repeated administration
100 mA. In addition, the scanning resolution is about 27.5 μm of infliximab has the potential of side effects such as increased in-
in the period of scanning (1.0 mm Al/0.25 mm Cu filter, rotation fection risks, unexpected inflammatory disorders, and systemically
step: 0.75°, scan time: 17 min). The area of an implant is chosen toxic to organs [54,55]. Therefore, the discovery of a more effective
as the region of interest (ROI), and bone phase is defined in the and safer delivery platform for RA treatment is still a challenge for
range from 10 0 0 to 2250 HU. The bone in-growth of ROI was re- researchers.In the pre-experiment, different content of infliximab
constructed and calculated through software NRecon version 1.6.6 (0.5, 1.0, 2.0, 4.0, 8.0 mg/mL) was used in the cell proliferation. Ac-
(Bruker micro-CT N.V., Kontich, Belgium), including parameters of cording to the CCK-8 assay (Figure S3), when the concentration of
bone volume/tissue volume (BV/TV) ratio,trabecular thickness (Tb. infliximab was 1.0 mg/mL, the relative ADSCs viability was highest.
Th, mm), trabecular separation (Tb. Sp, mm), and bone mineral In our system, the designedhydrogel with optimized concentration
density(BMD) [41–43]. of infliximab couldtotally degrade within 30 days, accompanied by
the sustained-release of infliximab. As illustrated in Fig. 1c, there
Push out test were mainly three stages of drug delivery including a faster re-
lease with 40% of drugs in the first stage (1-8 days), a stable re-
The push-out force of each scaffold in the implant-bone inter- lease with 20% of drugs in the second stage (8-17 days), and a sus-
face was detected using the universal testing machine. In order tained release with 10% of drugs in the third stage (17-28 days).
to disclose the inner side of each scaffold, the periosteal bone at Therefore,this injectable hydrogel could increase the release time

5
JID: ACTBIO
ARTICLE IN PRESS [m5G;December 11, 2020;11:46]

Y. Zhao, C. Gao, H. Liu et al. Acta Biomaterialia xxx (xxxx) xxx

Fig. 1. (a) The macroscopic observation of the gelation progress. (b) The injectable property of the hydrogel. (c) Infliximab release curve. (d) Time sweep measurements of
the hydrogel (temperature: 25°C, frequency: 10 rad/s, and strain: 0.1%). (e) The macroscopic self-healing procedures of the hydrogel. For the visualization observation, one
of hydrogel was dyed with malachite green (blue), and another hydrogel was dyed with sunset yellow (yellow). (f) Strain amplitude sweep measurements (γ =0.01-100%) of
the hydrogel with a fixed angular frequency of 10 rad/s. (g) Step strain measurements of the hydrogel with a fixed frequency of 10 rad/s. Each strain interval was kept as
200 s.

of infliximab in vitro, and maintain a therapeutic infliximabconcen- when the shear strain decreased to 1%, G’and G’’ almost com-
tration in the target site for a prolonged period. pletely returned to their original values within 200 s. This pro-
The mechanical properties of hydrogels could be easily ad- cess could be repeated multiple times, demonstrating the repeat-
justed by changing the concentrations of the mixed polysaccha- able mechanicalself-healing property, since the acylhydrazoneand
ride solution. In this study, the hydrogel with optimal properties imine groups in the hydrogel can dynamically associate and disso-
was applied for further self-healing properties study (Fig. 1d). The ciate at room temperature [56,58–61].
self-healing behavior of the developed hydrogel was displayed in In healthy adults, bone is a dynamic tissue that constantly re-
Fig. 1e. Two round-shaped gels were physically cleaved in four news through a series of stem cell behaviors [46,47,62]. How-
halves with a scissor, and then immediately brought into contact. ever, under RA conditions, the excessive infiltration of inflamma-
After 2 hours of repair, the four halves could be automatically re- tory cytokines could depress the bioactivity oftransplanted cells.
covered under ambient conditions, and the repaired hydrogel sam- As a result, the stem cell-based therapy efficiency for RA is seri-
ples also exhibited good mechanical properties and self-supporting ously compromised [3]. On the basis of this problem, the optimal
properties. This ability was endowed by the reversible nature of infliximab-based hydrogel was designed to overcomecell loss un-
the created dynamic bonds thatcould be recovered under mild con- der inflammatory microenvironment, as well as enhance cell sur-
ditions without the demand for any external stimulus [56,57]. In vival, proliferation, and function in the days and weeks that fol-
order to further study the self-healing properties of the hydro- lowtreatment.Prior to examining these effects of infliximab-based
gel, the rheological properties of the hydrogel were analyzed un- hydrogel in vitro, ADSCs were cultured under inflammatorymi-
der strain sweep mode (0.01-100%). As shown in Fig. 1f, the G’ croenvironment and cell counting kit-8 (CCK-8) assay was per-
and G’’ both maintained constant values in the low strain region. formed to value the cell metabolism. For comparison,ADSCs were
As the strain increased, the G’ and G’’ curves intersected at the seeded on HA hydrogels crosslinked by infliximab or not, respec-
strain of 5.72351%, which is the required critical strain value to de- tively. It was found that there was no significant difference be-
struct of the hydrogel network. When the strain further increased tweenthe two groups for 1 day, while the proliferation of AD-
to 100%, the values of G’ and G’’ decreased rapidly due to the col- SCsseeded on infliximab-based hydrogels was higherthan HA hy-
lapse of the hydrogel network structure. Based on the strain am- drogel group at 4, 7 and 14 days (Fig. 2a). Overall, infliximab-
plitude sweep results, we quantitatively analyzed the self-healing based hydrogels can construct a proper environment for ADSCs
properties of the hydrogel.As shown in Fig. 1g, we applied 100% proliferation under RA condition. Corresponding to the results of
and 1% continuously varying shear strains to the hydrogel, each CCK-8, the live/dead staining resultsacted as an evidence that
with the duration of 200 s. It can be observed that at higher shear ADSCs showed the better vitality on infliximab-based hydrogels
strains (100%), G’’ was always higher than G’, indicating that the (Fig. 2b). Therefore, under the inflammatory conditions of RA, the
network structure of the hydrogel was destroyed. Subsequently, designed infliximab-based hydrogelcan provide an environment

6
JID: ACTBIO
ARTICLE IN PRESS [m5G;December 11, 2020;11:46]

Y. Zhao, C. Gao, H. Liu et al. Acta Biomaterialia xxx (xxxx) xxx

Fig. 2. (a) The relative ADSCs viability was study by CCK-8 assay. (b) Cell viability of ADSCs on the surface ofhydrogels using the live/dead assay. Live cells appear green and
dead cells appear red.∗ P< 0.05, ∗ ∗ P< 0.01, and ∗ ∗ ∗ P< 0.001. All experiments were performed at least three times.

Fig. 3. (a)The SEM observation of the self-healing hydrogel. (b) The CAD model (i) andmacroscopical view (ii) of the 3D printed porous metal scaffold. (c) The SEM observa-
tion of the 3D printed porous metal scaffold. (d) The combination of the 3D printed porous metal scaffold and the hydrogel.

that is suitable and can relieve inflammation the proliferation of shaped Ti6 Al4 V-based 3DPMS consist of multiplanar hexagonal
ADSCs. unit cellstructures. The surfacemorphologies of the porous Ti6 Al4 V
The lyophilized infliximab-based hydrogels possessed a well- scaffold before and after infliximab-based hydrogelsmodification
defined porous network structure with pore diameters in the range are further observedby usingscanning electron microscopy (SEM).
of 20 0-30 0 nm (Fig. 3a). The macroporous structure was benefi- Fig. 3c showed that the original Ti6 Al4 V porous scaffold has a ho-
cial tocell migration and proliferation [63]. However, the weak me- mogeneous porous structure interconnected bymetal struts with
chanical strength and biodegradable property of hydrogelsrestrain average pore size of 1 mm, which is unsatisfactory for cell mi-
their use as long-term bone scaffolds for RA. Aiming for optimal gration, proliferation, and ECM deposition. After modification of
designand fabrication of scaffolds for ADSCsdelivery, we combined infliximab-based hydrogels, the large innerpores of Ti6 Al4 V porous
infliximab-based hydrogels with 3DPMS. Clinically, the emerging scaffoldwere completely filled with polymeric networks (Fig. 3d).
3D printing technique allows for tailoring Ti6 Al4 V-based metal Thus, the introduction of infliximab-based hydrogels shows thesu-
scaffolds for individualized treatment of osteochondral defects, periority of accommodating cells, inducing more pronounced cell
[31,35] especially large-scale bone defects caused by RA. According cellular growth, communication, and function for RA treatment.
to the CAD model and macroscopical view (Fig. 3b), the columnar- was not only more suitable for accommodating cells, but could also

7
JID: ACTBIO
ARTICLE IN PRESS [m5G;December 11, 2020;11:46]

Y. Zhao, C. Gao, H. Liu et al. Acta Biomaterialia xxx (xxxx) xxx

Fig. 4. (a) Statistical analysis of ALP activity of ADSCs cultured in samples at 4 and 7 days. (b) Alizarin red staining of ADSCs cultured in the MS, MS+I, MS+H and MS+HI at
14 and 21 days. (c) Runx2, (d) Col-I, and (e) OCN relative gene expression levels of ADSCs cultured on MS, MS+I, MS+H and MS+HI. Error bars represent standard deviation
of measurements within each experiment.∗ P< 0.05, ∗ ∗ P< 0.01, and ∗ ∗ ∗ P< 0.001. All experiments were performed at least three times.

lead to more pronounced cell growth, communication, and func- formation of themineralized matrix (Figure S5) [39,40]. At a gene
tion for RA treatment. level, the expressions of multiple osteogenic differentiation mark-
It is necessary to examine the relationships between cell mor- ers on different scaffolds were monitored by Real-time quantitative
phologyand 3D scaffold, because the cell morphology in matrix has Polymerase Chain Reaction (RT-qPCR). Under RA condition, ADSCs
a significantlyeffect on cellular behaviors [34]. For comparison, AD- were cultured in different scaffolds for 7 and 14 days, respectively.
SCs were seeded on the 3D printed porous metal scaffolds (MS), The typical osteogenic gene markers, such as runt-related gene 2
3D printed porous metal scaffolds loaded with infliximab (MS+I), (Runx-2), osteocalcin (OCN), and type I collagen (Col-I) were mea-
3D printed porous metal scaffolds modified with HA hydrogels sured [44]. According to Fig. 4c-e, the levels of all osteogenic dif-
(MS+H), and 3D printed porous metal scaffolds modified with ferentiation marker were significantly increased in MS+HI scaffold.
infliximab-based hydrogels (MS+HI), respectively.The cytoskeletal Thus, the present in vitro results demonstrated that the combi-
organization of ADSCs attached on the above implantswas evalu- nation of 3DPMS and infliximab-based hydrogels can accelerate
atedby staining actin filaments with phalloidin, and cell morphol- the differentiation of ADSCs toward the osteogenic lineage even
ogy was monitored by fluorescence microscopy.Figure S4 showed in RA conditions. This phenomenon was mainly attributed to the
that the attached ADSCs on MS and MS+I scaffolds possessed a sustained release of infliximab decreasing the level of inflamma-
spindle-shaped morphology similar to that of 2D culture system, tion and boostosteogenic differentiation.Moreover, the extracellular
whereas ADSCs on MS+Hand MS+HI scaffoldsdisplayed a predom- matrix-mimicking properties of hydrogel may favor the osteogenic
inately spherical shape with weaklydeveloped cytoskeleton. The differentiation of ADSCs.
later morphology tends to mimic the cell shape in vivo. Further- Based on the above in vitro results of the composite scaffolds
more, the osteogenic potential of scaffolds was monitored in the described, we next considered whether the composite scaffolds
following experiments, including the expressionof alkaline phos- might also exert osteogenesisand immunomodulatory in vivo ef-
phatase(ALP) activity, Alizarin Red staining, and the expression of ficacy in an experimental RA animal model. In this study, the fe-
multiple osteogenic differentiation markers. ALP acts as an early male rabbit was used as a RA animal model to imitate pathologic
osteogenic differentiation indicator that plays an essential role in change of RA. The hyperplastic synovium and invaded bone tis-
bone mineralization [41–43]. As shown in Fig. 4a, the ALP activity sue (Figure S6) indicated the successful establishment of the RA
of ADSCs in MS+HI scaffold was highest among all groups in in- model. Subsequently,a full-thickness cylindricalosteochondral de-
flammatory microenvironment, especially after cultured for 7 days. fect (diameter, 5 mm; length, 10 mm) was first created in the
Furthermore, the calcium nodules on scaffolds were measured by knee joint to mimic the large-scale bone erosion of a typical RA.
Alizarin Red staining (Fig. 4b). Based on gross appearances of scaf- To study the in vivo efficacy, the representative scaffold (MS+HI)
folds, the ADSCs cultured on MS+HI scaffold showed more homo- loaded with ADSCs (ADSCS@MS+HI) and four additional control
geneous distributed and much deeper color than other groups at scaffolds (MS, MS+H, ADSCS@MS+H, MS+HI) were constructed
14 and 21 days, implying the significantly upregulatedmineral nod- and then implanted into the defect. After 3 months of healing,
ules. The colorimetric quantitative analysis exhibited ADSCs cul- hematoxylin-eosin (H&E) staining of synovium was performed to
tured on MS+HI scaffold possessed the highest level on calcium evaluate the inflammatory response in local joint.As illustrated in
deposition, thereby demonstrated MS+HI scaffold can promote the Fig. 5a, when defects were treated with the four control scaf-

8
JID: ACTBIO
ARTICLE IN PRESS [m5G;December 11, 2020;11:46]

Y. Zhao, C. Gao, H. Liu et al. Acta Biomaterialia xxx (xxxx) xxx

Fig. 5. (a) H&E staining of the synovium in the five experimental groups(i: MS; ii: MS+H; iii: ADSCs@MS+H; iv: MS+HI; v: ADSCs@MS+HI) at 3 months of post-surgery to
characterize inflammatory response. Immunological correlated cytokines expressions of TNF-α (b), OVA-Ab (c), IL-1 β (d), and IL-6 (e) at 3 months in the five experimental
groups(i: MS; ii: MS+H; iii: ADSCs@MS+H; iv: MS+HI; v: ADSCs@MS+HI). Error bars represent standard deviation of measurements within each experiment. ∗ P < 0.05, ∗ ∗ P<
0.01, and ∗ ∗ ∗ P< 0.001. All experiments were performed at least three times.

folds, trachychromatic lymphocytes were densely distributed on ADSCS@MS+HI group exhibited good performance on decreasing
synovium. Intriguingly, after treating with the ADSCS@MS+HI scaf- the inflammation level of RA.
fold, there were not obvious inflammatory cells demonstrating Clinically, RA frequently destroys cartilage and causes perma-
the good anti-inflammatory.Histological ICRS score of the sur- nent joint damage, eventually resulting in irreversible disability,
rounding synovium further demonstrated the above result (Fig- systemic complications, and even early death [3,54]. So far, re-
ure S7). These results indicated that the ADSCs@MS+HI group pairingosteochondral defects in RA remain challenging due to the
could inhibit the infiltration of inflammatory cells around scaffold different components of cartilage and subchondral bone. Desirably,
[38,44]. our designed ADSCS@MS+HI scaffold can effectively alleviate cy-
As a kind of autoimmune disease, RA can produce a lot of im- tokine expressions, but also improve cartilage and bone reconstruc-
munological cytokines, such as tumor necrosis factor alpha (TNF- tion. According to Fig. 6a, in MS and MS+H groups, the scaffold
α ), Anti-OVA,interleukin 6 (IL-6),and interleukin 1 beta (IL-1β ). was clearly visible while a large cavity was existed on the sur-
These immunological cytokinescan take pleiotropic actions in the face of cartilage. Moreover, in ADSCS@MS+H and MS+HI groups,
pathophysiological processes of RA, and the abnormal expression the regenerated cartilage was also limited. It’s remarkable that in
of cytokines is also attributed to sustainable RA inflammation ADSCS@MS+HI group, the scaffold was completely covered by re-
and bone damage [30]. The levels of immune-related cytokines in paired cartilage and subchondral bone, and the articular surface
serum were further monitored after treatment with the experi- was smoothest among all groups. Meanwhile, microcomputed to-
mental scaffolds (Fig. 5b-e). It is worth noting that the expression mography reconstruction (Micro-CT) was performed on the artic-
of inflammatory factors (TNF-α , Anti-OVA, IL-6, and IL-1β ) in the ular joint samples to estimate subchondral bone regeneration. As
MS, MS+H, MS+HI and ADSCS@MS+H groups remained at high illustrated in Fig. 6b-d,ADSCS@MS+HI group elicited the most re-
level after 3 months. It is revealed that when the ADSCS@MS+HI markable new bone formation (yellow section) capacity in compar-
groupapplied to treat RA, proinflammatory cytokine levels were ison with other control scaffolds which corresponds tohistological
significantly reduced. The alleviation of inflammation is due to analysis. According to Masson staining in Fig. 6e, there was barely
the fact that hydrogel can provide a considerably suitable envi- new bone tissue formed in the defect for MS group, and limited
ronment for stem cell survival, engraftment, and function on the bone regeneration was observed in MS+H, ADSCS@MS+H, MS+HI
immunoregulatory of RA. On one other hand, the sustained re- groups.It should be noted that in ADSCS@MS+HI groups, relatively
lease of infliximab from hydrogels can inactivate TNF-α . It is well- thick bone tissue was formed and new bone formation around and
established that TNF-α can be regarded as a dominant cytokine within the scaffold implied the fine implant integration at the in-
in RA, since it appears to be upstream of other major cytokines terface of bone-implant. We believe the above regenerated bone
and affects the production of others. These results proved that structure is crucial for proper load-bearing function of regenerated

9
JID: ACTBIO
ARTICLE IN PRESS [m5G;December 11, 2020;11:46]

Y. Zhao, C. Gao, H. Liu et al. Acta Biomaterialia xxx (xxxx) xxx

Fig. 6. (a) Macrophotographs showed the defects in the five experimental groups(i: MS; ii: MS+H; iii: ADSCs@MS+H; iv: MS+HI; v: ADSCs@MS+HI)at 3 months of post-
surgery. (b) Micro-CT 2D images indicated the osseointegration of five groups at 3 months of postsurgery (dark gray section represents the 3DPMS scaffold). The front (c)
and vertical (d) views of 3D reconstruction images of the five experimental groups at 3 months of post-surgery (yellow section represents the new bone, and gray color
stands for the scaffold, respectively). (e) Representative histological sections stained with Masson staining at 3 months in the five experimental groups(i: MS; ii: MS+H; iii:
ADSCs@MS+H; iv: MS+HI; v: ADSCs@MS+HI). The quantitative micro-CT analysis of bone volume/ total volume (f), bone mineral density (g), trabecular thickness (h), and
trabecular spacing within the defect for the five experimental groups. Error bars represent standard deviation of measurements within each experiment. ∗ P < 0.05, ∗ ∗ P<
0.01, and ∗ ∗ ∗ P< 0.001. All experiments were performed at least three times.

bone tissue. Thus, the push-out test was executedto assess the bi- crostructure. As illustrated in Fig. 6h and i, the ADSCS@MS+HI
ological fixation strength at the interface of bone-implant (Figure scaffold could significantly enhanced Tb. Th, while upgraded Tb.
S8). Expectedly, the RA rabbit treated by ADSCS@MS+HI scaffold Sp. The successful bone regeneration capacity is attributed to the
exhibited a high ultimate push-out force of 350 N after 3-month fact that the natural extracellular matrix of hydrogels that can con-
implantation. Compared with other groups, the push-out force in struct an ideal microenvironment for ADSCs to differentiate into
defect after being treated with ADSCS@MS+HI scaffold was closer osteoblasts. On the other hand, sustained release of infliximab may
to the fixation strength of normal rabbit knee. Such restoration decrease the level of inflammation in RA condition, and minimize
of fixation strength in the fractured bone tissue demonstrated our the damage of inflammatory cytokines to osteogenic differenti-
ADSCS@MS+HI scaffold can locally enhance implant osseointegra- ation.Therefore, the composited ADSCS@MS+HI scaffold can im-
tion in RA condition. The fine integration between host bone and prove the bone volume, bone quality, and bone microstructure in
ADSCS@MS+HI scaffold can also avoid implant failure in RA pa- RA condition.
tients.
Furthermore, according to quantitative statistics, the bone vol- 4. Conclusion
ume/total volume (BV/TV) and bone mineral density (BMD) of
ADSCS@MS+HI group were also obviously highest compared with In this work, we have developed an advanced scaffold for
other groups after 3 months (Fig. 6f and g). Concomitantly, the RA management through the integration of infliximab-based hy-
trabecular thickness (Tb. Th)and trabecular separation (Tb. Sp)of drogels and 3DPMS. The presence of rigid 3DPMS was appro-
bone tissue are alsoimportant indicators to evaluate trabecular mi- priate for large-scale bone defects caused by RA. In particu-

10
JID: ACTBIO
ARTICLE IN PRESS [m5G;December 11, 2020;11:46]

Y. Zhao, C. Gao, H. Liu et al. Acta Biomaterialia xxx (xxxx) xxx

lar, the designed infliximab-based hydrogels were introduced be- References

cause of their self-healable, anti-inflammatory, biocompatible, and
biodegradable properties. Benefitting from thebenefits that inor- [1] Z. Chen, A. Bozec, A. Ramming, G. Schett, Anti-inflammatory and immune-reg-
ulatory cytokines in rheumatoid arthritis, Nat. Rev. Rheumatol. 15 (2019) 9–17.
ganic and organic materials provided, the ADSCs cultured in the [2] K.D. Deane, V.M. Holers, The natural history of rheumatoid arthritis, Clin. Ther.
composite scaffoldmaintained a high viability, enhanced prolifer- 41 (2019) 1256–1269.
ative capacity, and osteogenic differentiation even under RA con- [3] F. Zhang, K. Wei, K. Slowikowski, C.Y. Fonseka, D.A. Rao, S. Kelly, S.M. Good-
man, D. Tabechian, L.B. Hughes, K. Salomon-Escoto, G.F.M. Watts, A.H. Jons-
dition. Considering the in vitro results, the composite scaffold son, J. Rangel-Moreno, N. Meednu, C. Rozo, W. Apruzzese, T.M. Eisenhaure,
was implanted into the critical-sized bone defect in the RA rab- D.J. Lieb, D.L. Boyle, A.M. Mandelin, A.Accelerating Medicines Partnership
bit model.Notably, we found that the composite scaffolds also ex- Rheumatoid 2nd, C. Systemic Lupus Erythematosus, B.F. Boyce, E. DiCarlo,
E.M. Gravallese, P.K. Gregersen, L. Moreland, G.S. Firestein, N. Hacohen, C. Nus-
ertedosteogenesisand immunomodulatory in vivo efficacy, such as
baum, J.A. Lederer, H. Perlman, C. Pitzalis, A. Filer, V.M. Holers, V.P. Bykerk,
down-regulated inflammatory cytokines, rebuilt damaged cartilage, L.T. Donlin, J.H. Anolik, M.B. Brenner, S. Raychaudhuri, Defining inflammatory
as well as improvedsubchondral bone repair. This inorganic-organic cell states in rheumatoid arthritis joint synovial tissues by integrating sin-
gle-cell transcriptomics and mass cytometry, Nat. Immunol. 20 (2019) 928–
hybrid system provides anoptional strategy for stem cell-based
942.
therapies, and has the potential to alter the landscape for RA study. [4] I.B. McInnes, G. Schett, Pathogenetic insights from the treatment of rheumatoid
arthritis, The Lancet 389 (2017) 2328–2337.
Author contributions [5] J.S. Smolen, D. Aletaha, A. Barton, G.R. Burmester, P. Emery, G.S. Firestein,
A. Kavanaugh, I.B. McInnes, D.H. Solomon, V. Strand, K. Yamamoto, Rheuma-
toid arthritis, Nat. Rev. Dis. Primers 4 (2018) 18001.
The manuscript was written through contributions of all au- [6] S. Ansboro, A.J. Roelofs, C. De Bari, Mesenchymal stem cells for the manage-
thors. All authors have given approval to the final version of the- ment of rheumatoid arthritis: immune modulation, repair or both? Curr. Opin.
Rheumatol. 29 (2017) 201–207.
manuscript. Yue Zhao and ChaohuaGao contributed equally to this- [7] Z. Liu, M. Tang, J. Zhao, R. Chai, J. Kang, Looking into the Future: Toward ad-
work. Yue Zhao designed experiments, performed statistical anal- vanced 3d biomaterials for stem-cell-based regenerative medicine, Adv. Mater.
ysis, and manuscript editing; ChaohuaGao, Zuhao Li and He Li- 30 (2018) e1705388.
[8] M.A. Gonzalez, E. Gonzalez-Rey, L. Rico, D. Buscher, M. Delgado, Treat-
ucarried out animal and cell experiments; Hou Liu and Hangrui ment of experimental arthritis by inducing immune tolerance with human
Liu performed literature research; YubinFeng carried out cell ex- adipose-derived mesenchymal stem cells, Arthritis. Rheumatol. 60 (2009)
periment data analysis; Quan Lin,Jincheng Wang and Bai Yang did 1006–1019.
[9] L. Cai, R.E. Dewi, A.B. Goldstone, J.E. Cohen, A.N. Steele, Y.J. Woo, S.C. Heilshorn,
manuscript revision. regulating stem cell secretome using injectable hydrogels with in situ network
formation, Adv. Healthc. Mater. 5 (2016) 2758–2764.
Funding sources [10] P. Lavrador, V.M. Gaspar, J.F. Mano, Mechanochemicalpatternableecm-mimetic
hydrogels for programmed cell orientation, Adv. Healthc. Mater. 9 (2020)
e1901860.
This work was supported by the National Nature Science Foun- [11] J. Yeo, Y.M. Lee, J. Lee, D. Park, K. Kim, J. Kim, J. Park, W.J. Kim, Nitric ox-
dation of China (Grant Nos. 51861145311, 21174048, 81671804, and ide-scavenging nanogel for treating rheumatoid arthritis, Nano Lett. 19 (2019)
81772456). 6716–6724.
[12] D.R. Griffin, W.M. Weaver, P.O. Scumpia, D. Di Carlo, T. Segura, Accelerated
wound healing by injectable microporous gel scaffolds assembled from an-
Supporting information nealed building blocks, Nat. Mater. 14 (2015) 737–744.
[13] T. Matsuda, R. Kawakami, R. Namba, T. Nakajima, J. Gong, Mechanorespon-
sive self-growing hydrogels inspired by muscle training, Science 363 (2019)
FTIR analysis was applied to confirm the structure of HA-ALD 504–508.
(Fig. S1); FTIR analysis was applied to confirm the structure of HA- [14] D. Macaya, K. Ng, M. Spector, Injectable collagen-genipin gel for the treatment
HYD (Fig. S2); Analysis of ADSCs morphology on four experimen- of spinal cord injury: in vitro studies, Adv. Funct. Mater. 21 (2011) 4788–4797.
[15] A.C. Daly, L. Riley, T. Segura, J.A. Burdick, Hydrogel microparticles for biomed-
tal groups (Fig. S3); The semi-quantitative analysis of alizarin red ical applications, Nat. Rev. Mater. 5 (2019) 20–43.
staining at 14 and 21 days (Fig. S4); Macrophotographs of hyper- [16] T. Richardson, C. Wiegand, F. Adisa, K. Ravikumar, J. Candiello, P. Kumta,
plastic synovium and eroded bone tissue, and quantitative analysis I. Banerjee, Engineered peptide modified hydrogel platform for propagation of
human pluripotent stem cells, ActaBiomater. 113 (2020) 228–239.
of joint diameter hyperplastic synovium and eroded bone tissue, [17] Y. Zhao, Z. Li, Y. Jiang, H. Liu, Y. Feng, Z. Wang, H. Liu, J. Wang, B. Yang, Q. Lin,
and quantitative analysis of joint diameter (Fig. S5); The push-out Bioinspired mineral hydrogels as nanocomposite scaffolds for the promotion
test and the quantitative analysis of push-out force in defect after of osteogenic marker expression and the induction of bone regeneration in os-
teoporosis, ActaBiomater. 113 (2020) 614–626.
treated with the five experimental groups (Fig. S6).
[18] M. Konieczynska, J. Villa-Camacho, C. Ghobril, M. Perez-Viloria, K. Tevis,
W. Blessing, A. Nazarian, E. Rodriguez, M. Grinstaff, On-demanddissolution of
Declaration of Competing Interest adendritic hydrogel-based dressing for second-degree burn wounds through
thiol-thioester exchange reaction, Angew. Chem. Int. Ed. 55 (2016) 9984–9987.
[19] J. Weaver, D. Headen, J. Aquart, C. Johnson, L. Shea, H. Shirwan, A. García, Vas-
All authors declare no conflict of interest. culogenic hydrogel enhances islet survival, engraftment, and function in lead-
ing extrahepatic sites, Sci. Adv. 3 (2017) e1700184.
Acknowledgements [20] J. Basuki, F. Qie, X. Mulet, R. Suryadinata, A. Vashi, Y. Peng, L. Li, X. Hao, T. Tan,
T. Hughes, Photo-modulated therapeutic protein release from ahydrogel depot
using visible light, Angew. Chem. Int. Ed. 56 (2017) 966–971.
This work was supported by the National Nature Science Foun- [21] D. Lin, L. Lei, S. Shi, X. Li, Stimulus-responsive hydrogel for ophthalmic drug
dation of China (Grant Nos. 51861145311, 21174048, 81671804, delivery, Macromol. Biosci. 19 (2019) e190 0 0 01.
[22] Q. Liu, C. Zhan, A. Barhoumi, W. Wang, C. Santamaria, J.B. McAlvin, D.S. Ko-
and 81772456). Jincheng Wang and He Liu are thankful to the hane, A supramolecular shear-thinning anti-inflammatory steroid hydrogel,
Science and Technology Development Program of Jilin Province Adv. Mater. 28 (2016) 6680–6686.
(Grant Nos. 20190304123YY, 20180623050TC, 20180201041SF and [23] L. Mao, H. Wang, M. Tan, L. Ou, D. Kong, Z. Yang, Conjugation of two comple-
mentary anti-cancer drugs confers molecular hydrogels as a co-delivery sys-
20170204004GX), the Cultivation Program from the Second Hos- tem, Chem. Comm. 48 (2012) 395–397.
pital of Jilin University for National Natural Science Foundation [24] T. Wu, C. Cui, Y. Huang, Y. Liu, C. Fan, X. Han, Y. Yang, Z. Xu, B. Liu, G. Fan,
(KYPY2018-01), and the Youth Talents Promotion Project of Jilin W. Liu, Coadministration of an adhesive conductive hydrogel patch and an in-
jectable hydrogel to treat myocardial infarction, ACS Appl. Mater. Interfaces 12
Province (Grant No. 192004).
(2020) 2039–2048.
[25] F. Zhao, M.L. Ma, B. Xu, Molecular hydrogels of therapeutic agents, Chem. Soc.
Supplementary materials Rev. 38 (2009) 883–891.
[26] Y.H. An, J. Lee, D.U. Son, D.H. Kang, M.J. Park, K.W. Cho, S. Kim, S.H. Kim, J. Ko,
M.H. Jang, J.Y. Lee, D.H. Kim, N.S. Hwang, Facilitated transdermal drug delivery
Supplementary material associated with this article can be using nanocarriers-embedded electroconductive hydrogel coupled with reverse
found, in the online version, at doi:10.1016/j.actbio.2020.12.005. electrodialysis-driven iontophoresis, ACS Nano 14 (2020) 4523–4535.

11
JID: ACTBIO
ARTICLE IN PRESS [m5G;December 11, 2020;11:46]

Y. Zhao, C. Gao, H. Liu et al. Acta Biomaterialia xxx (xxxx) xxx

[27] E.A. Appel, M.W. Tibbitt, M.J. Webber, B.A. Mattix, O. Veiseh, R. Langer, Self- [45] S. Bai, X. Zhang, X. Lv, M. Zhang, X. Huang, Y. Shi, C. Lu, J. Song, H. Yang,
-assembled hydrogels utilizing polymer-nanoparticle interactions, Nat. Com- Bioinspired mineral-organic bone adhesives for stable fracture fixation and ac-
mun. 6 (2015) 6295. celerated bone regeneration, Adv. Funct. Mater. (2019) 1908381.
[28] N.M. Sangeetha, U. Maitra, Supramolecular gels: functions and uses, Chem. Soc. [46] C. Li, A.K. Born, T. Schweizer, M. Zenobi-Wong, M. Cerruti, R. Mezzenga,
Rev. 34 (2005) 821–836. Amyloid-hydroxyapatite bone biomimetic composites, Adv. Mater. 26 (2014)
[29] H. Wang, J. Wei, C. Yang, H. Zhao, D. Li, Z. Yin, Z. Yang, The inhibition of tumor 3207–3212.
growth and metastasis by self-assembled nanofibers of taxol, Biomaterials 33 [47] F. Nudelman, K. Pieterse, A. George, P.H. Bomans, H. Friedrich, L.J. Brylka,
(2012) 5848–5853. P.A. Hilbers, G. de With, N.A. Sommerdijk, The role of collagen in bone apatite
[30] P. Wojdasiewicz, L.A. Poniatowski, P. Nauman, T. Mandat, A. Paradowska-Go- formation in the presence of hydroxyapatite nucleation inhibitors, Nat. Mater.
rycka, K. Romanowska-Prochnicka, D. Szukiewicz, A. Kotela, L. Kubaszewski, 9 (2010) 10 04–10 09.
I. Kotela, I. Kurkowska-Jastrzebska, R. Gasik, Cytokines in the pathogenesis of [48] Z. Sun, F. Lv, L. Cao, L. Liu, Y. Zhang, Z. Lu, Multistimuli-responsive, moldable
hemophilic arthropathy, Cytokine Growth Factor Rev. 39 (2018) 71–91. supramolecular hydrogels cross-linked by ultrafast complexation of metal ions
[31] Y. Zhao, Z. Wang, Y. Jiang, H. Liu, S. Song, C. Wang, Z. Li, Z. Yang, H. Liu, and biopolymers, Angew. Chem. Int. Ed. Engl. 54 (2015) 7944–7948.
J. Wang, B. Yang, Q. Lin, Biomimetic composite scaffolds to manipulate stem [49] H. Wang, J. Di, Y. Sun, J. Fu, Z. Wei, H. Matsui, A.C. del Alonso, S. Zhou, Bio-
cells for aiding rheumatoid arthritis management, Adv. Funct. Mater. 29 (2019) compatible PEG-chitosan@carbon dots hybrid nanogels for two-photon fluores-
1807860. cence imaging, near-infrared light/pH dual-responsive drug carrier, and syner-
[32] B.P. Purcell, D. Lobb, M.B. Charati, S.M. Dorsey, R.J. Wade, K.N. Zellars, gistic therapy, Adv. Funct. Mater. 25 (2015) 5537–5547.
H. Doviak, S. Pettaway, C.B. Logdon, J.A. Shuman, P.D. Freels, J.H. Gorman, [50] Z. Wei, J.H. Yang, Z.Q. Liu, F. Xu, J.X. Zhou, M. Zrínyi, Y. Osada, Y.M. Chen, Novel
R.C.Gorman 3rd, F.G. Spinale, J.A. Burdick, Injectable and bioresponsive hydro- biocompatible polysaccharide-based self-healing hydrogel, Adv. Funct. Mater.
gels for on-demand matrix metalloproteinase inhibition, Nat. Mater. 13 (2014) 25 (2015) 1352–1359.
653–661. [51] J.D. Kretlow, S. Young, L. Klouda, M. Wong, A.G. Mikos, Injectable bioma-
[33] Y. Yao, P. Wang, X. Li, Y. Xu, G. Lu, Q. Jiang, Y. Sun, Y. Fan, X. Zhang, A di– terials for regenerating complex craniofacial tissues, Adv. Mater. 21 (2009)
self-crosslinking hyaluronan-based hydrogel combined with type I collagen to 3368–3393.
construct a biomimetic injectable cartilage-filling scaffold, ActaBiomater. 111 [52] Z. Wang, Y. Zhang, J. Zhang, L. Huang, J. Liu, Y. Li, G. Zhang, S.C. Kundu,
(2020) 197–207. L. Wang, Exploring natural silk protein sericin for regenerative medicine: an
[34] O. Chaudhuri, L. Gu, M. Darnell, D. Klumpers, S.A. Bencherif, J.C. Weaver, injectable, photoluminescent, cell-adhesive 3D hydrogel, Sci. Rep. 4 (2014)
N. Huebsch, D.J. Mooney, Substrate stress relaxation regulates cell spreading, 7064.
Nat. Commun. 6 (2015) 6364. [53] A.D.Medrano Acevedo, M.FajardoGaitan, Infliximab: pharmacology, uses and
[35] Y.F. Feng, L. Wang, Y. Zhang, X. Li, Z.S. Ma, J.W. Zou, W. Lei, Z.Y. Zhang, Effect limitations.
of reactive oxygen species overproduction on osteogenesis of porous titanium [54] F. Gao, Q. Yuan, P. Cai, L. Gao, L. Zhao, M. Liu, Y. Yao, Z. Chai, X. Gao, Au clusters
implant in the present of diabetes mellitus, Biomaterials 34 (2013) 2234–2243. treat rheumatoid arthritis with uniquely reversing cartilage/bone destruction,
[36] Y.W. Ge, Z.Q. Liu, Z.Y. Sun, D.G. Yu, K. Feng, Z.A. Zhu, Y.Q. Mao, Tita- Adv. Sci. 6 (2019) 1801671.
nium particlemediatedosteoclastogenesis may be attenuated via bidirectional [55] J. Seo, S.H. Park, M.J. Kim, H.J. Ju, X.Y. Yin, B.H. Min, M.S. Kim, Injectable click-
ephrinB2/ephB4 signaling in vitro, Int. J. Mol. Med. 42 (2018) 2031–2041. -crosslinked hyaluronic acid depot to prolong therapeutic activity in articular
[37] G. Jiang, A.H. Huang, Y. Cai, M. Tanase, M.P. Sheetz, Rigidity sensing at the lead- joints affected by rheumatoid arthritis, ACS Appl. Mater. Interfaces 11 (2019)
ing edge through alphavbeta3 integrins and RPTPalpha, Biophys. J. 90 (2006) 24984–24998.
1804–1809. [56] T.C. Tseng, L. Tao, F.Y. Hsieh, Y. Wei, I.M. Chiu, S.H. Hsu, An Injectable, Self-
[38] C. Deng, Q. Yao, C. Feng, J. Li, L. Wang, G. Cheng, M. Shi, L. Chen, J. Chang, -healing hydrogel to repair the central nervous system, Adv. Mater. 27 (2015)
C. Wu, Retracted: 3D printing of bilineage constructive biomaterials for bone 3518–3524.
and cartilage regeneration, Adv. Funct. Mater. 27 (2017) 1703117. [57] Y. Zhang, B. Yang, X. Zhang, L. Xu, L. Tao, S. Li, Y. Wei, A magnetic self-healing
[39] X. Zhang, Y. Zhu, L. Cao, X. Wang, A. Zheng, J. Chang, J. Wu, J. Wen, X. Jiang, hydrogel, Chem. Commun. 48 (2012) 9305–9307.
H. Li, Z. Zhang, Alginate-aker injectable composite hydrogels promoted irregu- [58] F. Ding, S. Wu, S. Wang, Y. Xiong, Y. Li, B. Li, H. Deng, Y. Du, L. Xiao, X. Shi, A
lar bone regeneration through stem cell recruitment and osteogenic differen- dynamic and self-crosslinked polysaccharide hydrogel with autonomous self-
tiation, J. Mater. Chem. B 6 (2018) 1951–1964. -healing ability, Soft Matter 11 (2015) 3971–3976.
[40] M.C. Embree, M. Chen, S. Pylawka, D. Kong, G.M. Iwaoka, I. Kalajzic, H. Yao, [59] Y. Gao, Q. Luo, S. Qiao, L. Wang, Z. Dong, J. Xu, J. Liu, Enzymetically regulat-
C. Shi, D. Sun, T.J. Sheu, D.A. Koslovsky, A. Koch, J.J. Mao, Exploiting endoge- ing the self-healing of protein hydrogels with high healing efficiency, Angew.
nous fibrocartilage stem cells to regenerate cartilage and repair joint injury, Chem. Int. Ed. Engl. 53 (2014) 9343–9346.
Nat. Commun. 7 (2016) 13073. [60] B. Yang, Y. Zhang, X. Zhang, L. Tao, S. Li, Y. Wei, Facilely prepared inexpensive
[41] C. Deng, R. Lin, M. Zhang, C. Qin, Q. Yao, L. Wang, J. Chang, C. Wu, Mi- and biocompatible self-healing hydrogel: a new injectable cell therapy carrier,
cro/nanometer-structured scaffolds for regeneration of both cartilage and sub- Polym. Chem. 3 (2012) 3235.
chondral bone, Adv. Funct. Mater. 29 (2019) 1806068. [61] Y. Zhang, L. Tao, S. Li, Y. Wei, Synthesis of multiresponsive and dynamic chi-
[42] L.B. Jiang, D.H. Su, S.L. Ding, Q.C. Zhang, Z.F. Li, F.C. Chen, W. Ding, S.T. Zhang, tosan-based hydrogels for controlled release of bioactive molecules, Biomacro-
J. Dong, Salt-assisted toughening of protein hydrogel with controlled degrada- molecules 12 (2011) 2894–2901.
tion for bone regeneration, Adv. Funct. Mater. 29 (2019) 1901314. [62] Y. Wang, T. Azais, M. Robin, A. Vallee, C. Catania, P. Legriel, G. Pehau-Arnaudet,
[43] H. Maleki, M.A. Shahbazi, S. Montes, S.H. Hosseini, M.R. Eskandari, S. Zaun- F. Babonneau, M.M. Giraud-Guille, N. Nassif, The predominant role of colla-
schirm, T. Verwanger, S. Mathur, B. Milow, B. Krammer, N. Husing, Mechani- gen in the nucleation, growth, structure and orientation of bone apatite, Nat.
cally Strong Silica-Silk Fibroin Bioaerogel, A hybrid scaffold with ordered hon- Mater. 11 (2012) 724–733.
eycomb micromorphology and multiscale porosity for bone regeneration, ACS [63] H. Yi, F. Ur Rehman, C. Zhao, B. Liu, N. He, Recent advances in nano scaffolds
Appl. Mater. Interfaces 11 (2019) 17256–17269. for bone repair, Bone Res. 4 (2016) 16050.
[44] W. Shi, M. Sun, X. Hu, B. Ren, J. Cheng, C. Li, X. Duan, X. Fu, J. Zhang, H. Chen,
Y. Ao, Structurally and functionally optimized silk-fibroin-gelatin scaffold using
3d printing to repair cartilage injury in vitro and in vivo, Adv. Mater. (2017) 29.

12

View publication stats