Bioglass promotes wound healing by affecting gap junction connexin 43 mediated endothelial cell behavior

Bioglass promotes wound healing by affecting gap junction connexin 43 mediated endothelial cell behavior

Accepted Manuscript Bioglass promotes wound healing by affecting gap junction connexin 43 mediated endothelial cell behavior Haiyan Li, Jin He, Hongfe...

2MB Sizes 0 Downloads 136 Views

Accepted Manuscript Bioglass promotes wound healing by affecting gap junction connexin 43 mediated endothelial cell behavior Haiyan Li, Jin He, Hongfei Yu, Colin R. Green, Jiang Chang PII:

S0142-9612(16)00047-8

DOI:

10.1016/j.biomaterials.2016.01.033

Reference:

JBMT 17309

To appear in:

Biomaterials

Received Date: 19 October 2015 Revised Date:

12 January 2016

Accepted Date: 15 January 2016

Please cite this article as: Li H, He J, Yu H, Green CR, Chang J, Bioglass promotes wound healing by affecting gap junction connexin 43 mediated endothelial cell behavior, Biomaterials (2016), doi: 10.1016/ j.biomaterials.2016.01.033. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Bioglass promotes wound healing by affecting gap junction connexin 43 mediated endothelial cell behavior Haiyan Li1a, Jin He2a, Hongfei Yu1, Colin R Green3*, Jiang Chang1, 4*

1954 Huashan Road, Shanghai 200030, China 2

Department of Pediatric Orthopaedics, Xin Hua Hospital Affiliated to Shanghai Jiao Tong

University, 1665 Kongjiang Road, Shanghai, 200092, China. 3

Department of Ophthalmology, Faculty of Medical and Health Sciences, The University of

Auckland, Private Bag 92019, New Zealand 1142.

Shanghai Institute of Ceramics, Chinese Academy of Sciences, 1295 Dingxi Road, Shanghai

M AN U

4

200050, China a

RI PT

Med-X Research Institute, School of Biomedical Engineering, Shanghai Jiao Tong University,

SC

1

The two authors contributed to the work equally.

Abstract

It is well known that gap junctions play an important role in wound healing, and bioactive

TE D

glass (BG) has been shown to help healing when applied as a wound dressing. However, the effects of BG on gap junctional communication between cells involved in wound healing is not well understood. We hypothesized that BG may be able to affect gap junction mediated cell

EP

behavior to enhance wound healing. Therefore, we set out to investigate the effects of BG on gap junction related behavior of endothelial cells in order to elucidate the mechanisms through which

AC C

BG is operating. In in vitro studies, BG ion extracts prevented death of human umbilical vein endothelial cells (HUVEC) following hypoxia in a dose dependent manner, possibly through connexin hemichannel modulation. In addition, BG showed stimulatory effects on gap junction communication between HUVECs and upregulated connexin43 (Cx43) expression. Furthermore, BG prompted expression of vascular endothelial growth factor and basic fibroblast growth factor as well as their receptors, and vascular endothelial cadherin in HUVECs, all of which are beneficial for vascularization. In vivo wound healing results showed that the wound closure of full-thickness excisional wounds of rats was accelerated by BG with reduced inflammation during initial stages of healing and stimulated angiogenesis during the proliferation stage. Therefore, BG can stimulate wound healing through affecting gap junctions and gap junction related endothelial

ACCEPTED MANUSCRIPT cell behaviors, including prevention of endothelial cell death following hypoxia, stimulation of gap junction communication and upregulation of critical vascular growth factors, which contributes to the enhancement of angiogenesis in the wound bed and finally to accelerate wound healing. Although many studies have reported that BG stimulates angiogenesis and wound healing,

RI PT

this work reveals the relationship between BG and gap junction connexin 43 mediated endothelial cell behavior and elucidates one of the possible mechanisms through which BG stimulates wound healing. Key words: bioglass, gap junction, vascularization, wound healing

SC

1. Introduction

Bioglass (BG) has been well known to be able to regenerate or repair bone defects due to its

M AN U

excellent osteogenic bioactivity in stimulating proliferation and osteogenic differentiation of osteoblastic cells or stem cells, and in promoting mineralization of the extracellular matrix [1-6]. In addition, some studies have indicated that BG not only stimulate bone repair or regeneration but also enhance wound healing of soft tissue [5, 7-15]. Although the antibacterial ability of BG may contribute to this role [5, 7-13], the underlying mechanism through which BG stimulates wound

TE D

healing remains unknown. Recently, BG has been found to stimulate vascularization [3, 4, 16] and blood vessel ingrowth into a BG scaffold [17-20]. It has been reported that angiogenic indicators, including vascular endothelial growth factor (VEGF) and basic fibroblast growth factor VEGF

EP

(bFGF) , could be increased through both direct and indirect contact of relevant cells with 45S5 Bioglass® particles or with its dissolution products [1, 4, 6]. However, these studies focused

AC C

primarily on the stimulatory effects of growth factors but neglected the fact that, during the proliferation stage of wound healing, endothelial cells go through many steps towards vessel formation. These steps involve cellular behavior changes in response to their fluctuating microenvironment, including cell survival, migration and vascularization [21-23]. Interestingly, all these endothelial cell behaviors are related to gap junctional cell-cell communications as connexin 43 (Cx43) plays an important role in determining endothelial cell fate, with cell-cell communication necessary for vascularization and angiogenesis in wound healing [24-26]. Intercellular communication mediated by gap junctions is considered to play an important role during the growth and development of tissues and organs, and in the maintenance of normal

ACCEPTED MANUSCRIPT metabolism and tissue homeostasis [27-31]. A variety of connexin types are reported in the skin and play an important role in wound healing, among which Cx43 is the most ubiquitous of the connexins in skin and is localized in the cutaneous vasculature, fibroblasts, dermal appendages, as well as the basal and lower spinous cell layers of the epidermis [32, 33]. As a result, most studies

RI PT

on the role of gap junctions in wound healing have been focusing on channels formed by Cx43 [32, 34-37]. However, its precise role in wound healing is complicated. On one hand, Cx43 gap junctional communication has been shown to spread cell death signals after injury [38, 39], in part owing to pathological hemichannel opening where cells are compromised if these large, relatively

SC

non-specific membrane channels open for sustained periods [40, 41]. On the other hand, Cx43 cell-cell gap junction channels play a positive role during other stages of the wound healing

during late stage of wound healing.

M AN U

process [24-26] as they are necessary for vascularization of endothelial cells and angiogenesis

With the critical role of Cx43 hemichannels and gap junction channels in wound healing, Cx43 mimetic peptides or Cx43 antisense have both been used to improve wound healing. These are small peptide sequences designed against the extracellular regions of the connexin molecule

TE D

and used to block connexin hemichannels, or antisense oligodeoxynucleotides used to reduce connexin protein expression at the wound site. Both reduce swelling and cell death as well as inflammation during the wound healing process [35, 42]. Other types of biomaterials that have

EP

been widely used to help wound healing are wound dressing materials, which mainly includes various types of natural-derived or synthetic biopolymers [43-47].

AC C

However, little attention has been paid not only to the influence of the biomaterials on cell-cell communications via gap junctions, but also to the effects of biomaterials on gap junctions during wound healing. Since gap junctions plays an important role in wound healing, it is critical to investigate the effect of biomaterials on gap junctions, especially Cx43 gap junctional communication, during wound healing process, and answering this question may help to understand biomaterial mediated wound healing and optimize design of wound healing biomaterials. We hypothesize that biomaterials, such as BG, may affect gap junctions and an underlying mechanism of BG enhanced wound healing may be related to the effects of BG on Cx43 gap junctional communication in endothelial cells and gap junction related endothelial cell behaviors. We therefore set out to elucidate the mechanism by which BG stimulates wound

ACCEPTED MANUSCRIPT healing, focusing on the effects of BG ion products on gap junction formation, connexin channel function and gap junction related endothelial cell behaviors in vitro, including the survival of endothelial cells under hypoxic conditions. In parallel, the effect of BG on endothelial cell expression of critical vascular growth factors, including VEGF and bFGF, their receptors and

RI PT

vascular endothelial cadherin (VE-cad), was investigated. Finally, we applied BG to full-thickness excisional wounds in rats to investigate its effect on wound healing, and to determine tissue responses to BG, including angiogenesis, Cx43 expression and inflammation at the wound site.

2. Materials and methods

SC

2.1 BG ion extracts

Bioglass 45S5 (BG) powder (Over 400 mesh) was kindly provided by the Shanghai Institute

M AN U

of Ceramics, Chinese Academy of Science. The sieved BG particles were characterized with scanning electronic microscopy (SEM). The ion extracts of BG were prepared according to the procedures previously reported [48, 49], and which were adapted from the standard procedure in ISO10993-1. One gram of BG powder was placed in a 10 cm diameter culture dish and soaked in 5 ml serum-free endothelial culture medium (Sciencell, USA). Afterwards, the mixture was

TE D

incubated in a cell culture incubator containing 5% CO2 at 37℃. After 24 h, the mixture was centrifuged and the supernatant was collected. The supernatant was then sterilized by filtering (Millipore, 0.22 µm) to obtain BG ion extract.

EP

For further use, BG ion extracts were diluted in endothelial cell medium, consisting of endothelial cell basal medium + 5% FBS (fetal bovine serum) + 1% endothelial cell growth

AC C

supplement + 1% P/S (penicillin/streptomycin) (Sciencell, USA). The BG ion extract was diluted at ratios of 1/8, 1/16, 1/32, 1/64, 1/128, 1/256 and 1/512 and the concentrations of Ca, Si and P in these extracts were detected by inductively coupled plasma atomic emission spectroscopy (ICP-AES; Optima 3000DV, Perkin Elmer, USA).

2.2 Cell isolation and culture Human umbilical vein endothelial cells (HUVECs) were isolated from human umbilical cord veins according to the method of Jaffe et al. with the culture procedures similar to those used in our previous studies [50, 51]. An Institutional Review Committee of Shanghai Jiao Tong University, School of Biomedical Engineering approved all these protocols. The isolated cells

ACCEPTED MANUSCRIPT were briefly cultured with endothelial cell medium (ECM, ScienCell, USA) in a humidified incubator containing 5% CO2 at 37 ℃. The culture medium was refreshed every 3 days until the primary HUVECs became confluent. Afterwards, cells were trypsinized and passaged. Cells were immunofluorescently stained with von Willebrand Factor and VE-cad to confirm their endothelial

RI PT

cell properties. Only early passages (passages 2–7) of the HUVECs were used in this study.

2.3 Cell proliferation assay

The diluted BG ion extracts at ratios from 1/8 to 1/512 (recorded as BG 1/8, BG 1/16, BG 1/32, BG 1/64, BG 1/128, BG 1/256 and BG 1/512, respectively) were used to determine the

SC

cytocompatibility of the BG ion extracts at different ion concentrations. Endothelial cell medium without BG ion extracts addition was used as the non-treatment control medium.

M AN U

HUVECs were seeded in 96-well plates at a density of 5 ×103 cells per well for culture. After 12 h, the medium in each well was removed and a Cell Counting Kit (CCK)-8 assay (Beyotime) was used according to manufacturer’s instructions to determine cell numbers based upon reaction product OD (optical density) value. In order to do this, 100 µl fresh medium containing 10 µl CCK-8 was added into each well and the cells were cultured for a further 2.5 h during which time

TE D

the CCK-8 reagent reacted with cellular metabolic products. The absorbance of the reaction products was then measured with a spectrophotometer at a wavelength of 450 nm using an enzyme-linked immunoadsorbent assay microplate reader (Synergy 2, Bio-TEK). The readings

EP

(OD value) represent the metabolic activity of cells, which reflects the number of cells present. After detecting the day 0 cell OD value, cell culture media was changed with the prepared BG ion

AC C

extracts at different ratios from 1/8 to 1/512. The cells were then continuously cultured for 7 days, and the CCK-8 assay was reapplied to detect cell OD values at days 1, 3 and 7.

2.4 HUVEC viability following hypoxia with and without BG solution In order to evaluate the protective effect of BG ion extracts on HUVECs that had experienced

hypoxia conditions, an in vitro cell viability assay adapted from Zhou et al. and Danesh-Meyer et al. was performed [41, 52]. HUVECs were seeded into 24 well plates (1 x 105 cells/well) containing endothelial cell medium and allowed to settle for 16 hours. The medium was replaced with DMEM/F12 containing 0.5% FBS and 1% glutamine. The whole plate seeded with HUVECs was then placed into a pre-warmed Billups-Rothenburg Modular Incubator Chamber. The chamber was flushed with 95% N2 and 5% CO2 for 5 minutes (20 L/min) to simulate hypoxia conditions.

ACCEPTED MANUSCRIPT After the chamber was filled with 95% N2 and 5% CO2, it was placed into a 37oC incubator for 3 hours. As there might be oxygen trapped in the plasticware, the chamber was regassed with 95% N2 and 5% CO2 after the first hour to ensure hypoxia was maintained. At the end of hypoxia period the culture media was replaced with fresh medium, or medium containing diluted BG ion extracts

RI PT

at the different ratios. Cells were further incubated at 37oC, 5% CO2 for 6 hours. The cells were then directly stained with a Live-dead staining assay kit, where the live cell is stained with Calcein-AM and dead cell is stained with Ethidium Homodimer (EH) (Invitrogen) and observed using an inverted microscope and images were taken. As the dead cells are stained in red and live

SC

cells are stained in green, this assay can confirm the percentage of viable cells. Cells that had not been exposed to low serum and hypoxia were counted as controls with viable cell counts

M AN U

expressed as a percentage of the control. Each experiment contained three duplicate wells for each culture medium with data from the three wells averaged. Six independent experiments were performed.

In order to confirm that HUVEC death was potentially due to hemichannel opening, a propidium iodide (PI) dye uptake assay was carried out to according to the procedures described

TE D

by Braet et al. [53]. HUVECs were plated into a 24-well plate at a density of 1 x 105 cells/well and subjected to hypoxia-reperfusion conditions described above. Following just one hour of reperfusion, PI was applied to HUVECs at a final concentration of 20 µg/ml. After 30 minutes, the cells were washed thoroughly with Phosphate Buffered Saline (PBS) to remove all PI from the

EP

media and fixed with 10% (wt./vol.) paraformaldehyde at 4oC for 5 min. Cells were then imaged

AC C

with an inverted microscope and CCD camera. Four images were taken for each well at a 10X magnification and the number of PI positive cells was counted. In order to exclude low levels of PI uptake by the background, normal cells were treated with PI in a same way and their images were used to set a threshold so that only cells with a PI uptake above this level were counted as PI positive. A bright field image corresponding to each PI image was taken to determine the total number of cells. The level of PI uptake was recorded as percentage of intact cells containing PI divided by the total number of cells.

2.5 Assessment of gap junction communication in HUVECs To determine the effect of BG ion extract on gap junctional communication between HUVECs, a cell settlement assay adapted from the method described by O’Carroll et al. was

ACCEPTED MANUSCRIPT applied [42]. HUVECs were seeded in two 24-well plates at a density of 6 x 104 cells/well and incubated overnight. The following day culture medium in the respective wells was changed to either endothelial cell medium or medium containing various BG ion extract concentrations and the cells were cultured for another day. One plate of cells was then incubated with medium

RI PT

containing 4 µM Calcein-AM (Molecular Probes) for 30 min to label the cells green (the HUVECs in the other plate were left unlabeled). The stained cells were washed 3 times for 5 min with PBS, trypsinized with 0.05% trypsin-EDTA and cell pellets were collected by centrifuge at 1000 rpm for 5 mins. The pellets were resuspended in 1 ml medium and 50 µl of each cell suspension (cultured

SC

in different media) was added to a well containing a confluent monolayer of unlabeled cells cultured in the equivalent media. The cells were incubated for another hour before they were

M AN U

observed under an inverted microscope and images taken with a CCD camera. For each well, six images were taken. The number of cells to which Calcein-AM dye had been transferred from labelled feeder cells were then counted.

2.6 Quantitative real-time polymerase chain reaction (Q-RT-PCR) To detect the effects of BG on gene expression of VEGF, VEGF receptor 2 (KDR), basic

TE D

fibroblast growth factor (bFGF) and bFGF receptor (bFGFR) from HUVECs, cells were plated in 6-well plates at a density of 1×105 cells/well and allowed to settle in endothelial cell medium overnight. The medium was then was removed and various culture media, including endothelial

EP

cell medium with and without BG 1/32, BG 1/64, BG 1/128, BG 256 and BG 1/512, were applied to the HUVECs. The cells were further cultured for 3 or 7 days. At the determined time points,

AC C

cells were washed twice with cold PBS and RNA was extracted using E. Z. N. A Total RNA kit I (OMEGA, Bio-tek) according to the manufacture’s guidelines. RNA concentration was measured using a nanodrop 1000 reader (Thermo SCIENTIFIC). cDNA was then synthesized using a ReverTra Ace-a kit (Toyobo Co., Ltd, Japan). To prepare a RT-PCR running solution, 1 µl of cDNA diluted at 1:10 in sterilized Mill-Q water was mixed with 9 µl SYBR-Green Master Mix containing primers (ToYOBO Co., Ltd) and was loaded in a 384-well plate. Primers of VEGF, KDR, bFGF, bFGFR, Cx43 and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (all from Sangon Biotech (Shanghai) Co., Ltd) were used at a final concentration of 400 nM. GAPDH was used as a housekeeping gene. The sequences for primers are listed in Table S1. A 7900 Real-time PCR system (Applied Biosystems) was used to run the real-time PCR analysis according to the

ACCEPTED MANUSCRIPT following program: denaturation was first performed for 1 min at 95 oC followed by 40 cycles (95 o

C for 15 s, 60 oC for 15 s, 72 oC for 45 s) PCR. Data was analyzed and compared by the ∆∆Ct

method, and each Q-RT-PCR was performed in triplicate. Data was normalized to the GAPDH mRNA expression of each condition and was quantified relative to the corresponding gene

RI PT

expressions from control samples.

2.7 Immunofluorescence

The expression and location of KDR and VE-cad were assessed using immunohistochemistry. After being cultured on 1 cm diameter glass coverslips for determined time points, cell layers were

SC

washed twice with preheated Hanks’ buffered salt solution (HBSS) and fixed with 4% (wt/vol) paraformaldehyde at room temperature for 15 min. Afterwards, the cells were permeabilized with

M AN U

cold methanol for 5 min at room temperature and blocked with PBS + 1% (wt/vol) bovine serum albumin (BSA) for 1 h at 37 oC. Then, cells were incubated in primary antibody solutions containing rabbit anti-human KDR (Abcam, diluted with HBSS-0.5% BSA at 1/100) or mouse anti-human VE-cad (Abcam, diluted in HBSS-0.5% BSA at 1/100) at 37oC for 2 h. After incubation with the primary antibody, the cells were washed twice with HBSS before Alexa 488

TE D

goat anti-rabbit IgG secondary antibody (Invitrogen) solutions (diluted with PBS + 0.5% BSA at 1/1000) or Alexa 488 goat anti-mouse IgG secondary antibody (Invitrogen) solutions (diluted with PBS + 0.5% BSA at 1/1000) were added to the cells. The cells were then incubated with

EP

secondary antibody solution at 37oC for another hour. The glass coverslips with cell layers were then removed from the culture plate wells and inverted onto glass slides using ProLong® Gold

AC C

antifade regent containing DAPI as a mounting medium(Invitrogen). The DAPI is a nuclear stain. Slides were examined using confocal immunofluorescence microscopy (Leica TCS-SP5 microscope).

2.8 Application of BG in full-thickness excisional wound model in rat An Institutional Review Committee of Shanghai Jiao Tong University, School of Biomedical

Engineering approved all animal study protocols. Four-week old male rats were purchased from Shanghai Laboratory Animal Center, CAS (SLACCAS). Five rats were used for each time point (2 day, 6 day and 12 day post-wounding). Full-thickness excisional wounds were created according to previously described methods [54]. Each rat was first anesthetized with an intraperitoneal injection of Chloral hydrate (350 mg/kg) and the whole back of the rat was sterilized with gauze

ACCEPTED MANUSCRIPT soaked with 75% ethanol. Once fully anaesthetized, the rats’ backs were shaved and four x 1 cm diameter full-thickness excisions were created on each rat. Two excisions were for control wounds without BG treatment and the other two excisions were for BG treated groups. There were, therefore, 60 excisions in total. The BG powders were sterilized by autoclaving prior to

RI PT

application and 0.015g BG powder was placed into each excision and spread evenly. After applying the BG powders, wounds were covered with Hydrofilm transparent dressing and bandaging in order to avoid dislocation of the dressing. Control wounds were left untreated but dressed similarly.

SC

2.9Analysis of wound healing 2.9.1 Wound closure

M AN U

The percentage of wound closure was then calculated as: % of wound closure= (area of original wound-area of actual wound)/(area of original wound)×100. A wound was considered completely closed when re-epithelialisation was complete.

2.9.2 Tissue specimen preparation and histological staining

Tissue samples were retrieved at various time points and prepared for histological and

TE D

immunohistochemical examination. The specimens were fixed overnight in 3.7% (v/v) buffered formalin (Sigma, USA), embedded in paraffin and then cross-sectioned longitudinally into 5 µm sections using a Leica RM2245microtome for histological staining and analysis. The sections

EP

were collected on glass slides and sections were stained with Gill’s 3 hematoxylin (Sigma-Aldrich) and aqueous eosin Y solution (Sigma-Aldrich) (H&E) to visualize the overall tissue morphology.

AC C

All samples were analyzed using an upright microscope (Leica DM4000, German) and images were acquired using a CCD camera (Leica, Germany). 2.9.3 Immunohistochemistry staining Rabbit anti-rat CD31 (Abcam), rabbit anti-rat Cx43 (Abcam) and rabbit anti-rat neutrophil

(Abcam) antibodies were used to perform immunohistochemical labelling to detect blood vessel formation, Cx43 expression and neutrophils respectively. Briefly, tissue sections were deparafinized in xylene and rehydrated in descending ethanol solution sections. The samples were then immersed in 0.3% H2O2 and then quenched in methanol to terminate peroxidase activity. After being immersed in a 99oC water bath for 20 min to improve antigen exposure, sections were blocked with 5% BSA in PBS for 30 min at room temperature. Then, sections were incubated in a

ACCEPTED MANUSCRIPT primary antibody solution diluted with 1% BSA in PBS with CD31 and Cx43 antibody concentrations of 1:300, and the neutrophil antibody at 1:100 dilution for 12 h at 4oC. Subsequently, the sections were rinsed with PBS and then incubated and reacted with the solutions in a DAB kit (Gene Tech, Shanghai) according to the manufacture’s instruction. Finally, the slides

RI PT

were counter-stained with hematoxylin, dehydrated and mounted using Antifade Mounting Medium (Beyotime). Normal tissue specimens were also prepared as positive controls. Five samples from each condition were analyzed manually. An upright microscope (Leica DM4000, German) was used to observe the specimen sections and a CCD camera used to take images.

SC

Blood vessels were identified in H&E-stained tissues and CD 31 stained samples at 200× magnification by defining lumens, identifying the presence of red blood cells within their

M AN U

boundaries and locating brown lines that showed up with CD 31 positive labelling. Cx43 expression were visible in images via immunolabelling (brown structures).

2.10 Statistical analysis

Data was expressed as means ± standard deviation (SD). Three independent experiments were carried out and at least 5 samples per test were taken for statistical analysis. A one-way

TE D

analysis of variance (ANOVA) with Tukey’s post hoc test was used for statistical analysis of multiple comparisons. A difference was considered when p < 0.05 (*) and p < 0.01 (**).

3. Results

EP

3.1 BG ionic products

SEM images show that the sieved BG particles have irregular shapes with particle size

AC C

ranging from 10-50 µm (Fig. S1). The concentrations of ions in BG ion extracts were detected and are recorded in Table S2. There is little difference between the concentrations of Ca in endothelial cell medium alone and in those media with BG ion extracts. However, in endothelial cell medium, very little Silicon (Si) was detected while there were much higher Si concentrations in BG 1/8-1/256. Even in BG 1/512, Si ions were detectable. It can be concluded that most favorable concentration of Si ions is in the range of 0.6-1.7 µg/ml. The concentration of P is increased in BG extracts diluted at 1/8, 1/16 and 1/32 as compared to that in the ECM. However, in the BG extracts diluted at 1/64, 1/128 and 1/256, the concentration of P is almost seem to that in the ECM.

3.2 Effects of BG on proliferation of HUVECs

ACCEPTED MANUSCRIPT BG ion extracts were diluted in endothelial cell medium at ratios from 1/8-1/512 and were applied to endothelial cell cultures in order to assess the effects of BG ion products on HUVEC proliferation. It can be seen from Fig.1 that culture media with high ion concentrations (BG 1/8 and BG 1/16) suppressed proliferation of HUVECs when compared to control medium alone.

RI PT

However, when the proliferation of HUVECs cultured in BG ion extracts diluted at lower ratios from 1/32-1/512 was compared to that of control medium, no obvious difference in proliferation was noted, indicating that the BG ion extracts diluted at ratios from 1/32-1/512 did not impact upon HUVEC growth and proliferation.

SC

3.3 Protective effects of BG on HUVECs exposed to hypoxia

Fig.2 and Fig.S2 show the protective effects of BG on cells after being exposed to hypoxia

M AN U

conditions. After being exposed to hypoxic conditions for 3 h, HUVECs were cultured in the different media (with and without BG) at normal oxygen levels for a further 6 h with significant differences apparent in subsequent cell viability. As shown in Fig.2A and B, the Calecin-AM live cell assay (green label) shows that over 80% of HUVECs exposed to hypoxia conditions but then cultured in BG ion extracts diluted at 1/128 remained viable, with their survival similar to that of

TE D

non-hypoxic controls. However, when re-cultured in endothelial cell medium alone or BG diluted at 1/512, only about 65% cells remained alive. This was confirmed by the EH dead cell assay (see Fig 2A, ECM image, showing red EH stained nuclei overlaid onto the Calcein-AM green label

EP

cell image). This indicates that BG ion extracts at appropriate concentrations are able to protect HUVECs that have been exposed to a period of hypoxia.

AC C

To investigate the mechanism by which BG might be sparing cells, a shorter post-hypoxia incubation period was used with PI uptake assessed as indicator of connexin hemichannel opening. Fig.S2 shows PI dye uptake in HUVECs exposed to hypoxic conditions and then cultured in different media for just one hour post-hypoxia. The assumption is that the cells remained viable at the one hour time point and nuclear staining indicates dye uptake through open hemichannels. Results indicate that about

25-30% of HUVECs have taken up PI during the one hour

post-ischemia culture in endothelial cell medium or BG at the lower 1/512 concentration. However, less than 10% have taken up the dye when cultured post-ischemia in BG ion extracts diluted at ratios from 1/32-1/256.

3.4 Effects of BG on gap junction communication between HUVECs

ACCEPTED MANUSCRIPT To determine the effect of BG on functional gap junction communication of HUVECs in different media, a cell settlement assay was used. After the cells had been cultured in the different media for 1 day and loaded with dye, they were parachuted onto a monolayer of unloaded cells in the equivalent respective media and imaged 1 h later. The spread of Calcein-AM dye into the

RI PT

unlabelled cells is seen in Fig. 3. Compared to the dye spread of HUVECs cultured in normal medium, BG 1/32 and BG 1/512, the spread of dye was significantly enhanced in HUVECs cultured in BG 1/64 and BG 1/128. BG 1/32 and BG 1/256 had no effect on Calcein dye spread, being similar to medium alone.

SC

3.5 Effects of BG on the expression of vascular growth factors and Cx43 expression in HUVECs

M AN U

VEGF, KDR, bFGF and bFGFR have been reported as critical growth factors and growth factor receptors for vascularization by HUVECs. At 1 day in culture and when compared to endothelial cell medium, culture media containing BG 1/64, BG 1/128 and BG 1/256 significantly upregulated bFGF, bFGFR, VEGF and KDR gene expression in HUVECs with the high concentration BG 1/32 and lowest concentration BG 1/512 having no effect on the expression of

TE D

these growth factors and receptors. After the cells were cultured for 3 days, the upregulation effects by BG extracts continued and higher gene expression were detected in the HUVECs cultured with the media containing BG 1/64, BG 1/128 and BG 1/256 BG extracts as compared to

EP

those detected in HUVECs cultured in same media for 24 hours. However, the high concentration BG 1/32 and lowest concentration BG 1/512 still had no effect on the expression of these growth

AC C

factors and receptors (Fig.4). At day 7, the stimulatory effects of BG ion extracts on these growth factors and receptors was still observed although their expression, especially that of VEGF and KDR, had decreased again. The expression of KDR in HUVECs cultured for 3 days was further analyzed using fluorescent immunohistochemical labelling and observed using confocal microscopy. More KDR positive immunofluorescence staining was observed in HUVECs cultured in BG 1/64, BG 1/128, and BG 1/256 than in cells cultured in medium alone or with BG 1/32 and BG 1/512 dilutions (Fig.5). In addition, the KDR expression in HUVECs cultured with different media is biphasic. The KDR expression in HUVECs cultured with BG 1/32, 1/512 and ECM control are similar, which are lower than that in HUVECs cultured with BG 1/64, 1/128 and 1/256. ECM control medium, BG 1/32, and BG 1 /512 have Si ion concentration of 0.18 µg/ml, 2.51

ACCEPTED MANUSCRIPT µg/ml and 0.37 µg/ml, respectively. However, the BG 1/64, BG 1/128 and BG 1/256 have Si ion concentration of 1.69 µg/ml, 0.84 µg/ml and 0.55 µg/ml, respectively. These results indicate that there is an effective Si ion concentration range for activating KDR in HUVECs. Higher or lower Si ion concentration cannot have the similar beneficial effects.

RI PT

Cell-cell communication between endothelial cells themselves and with other cell types is important for angiogenesis. Cx43 is a major connexin in blood vessel gap junction communication and VE-cad plays an important role in cell adhesion. Quantification of gene expression (Fig.6A) shows that the expression of Cx43 in HUVECs cultured in BG 1/64 and BG 1/128 for 7 days was

SC

much higher than in cells cultured in endothelial cell medium alone, BG 1/32 or BG 512. Fig.6B shows that after the cells had been culture for 3 and 7 days, BG 1/32, 1/64 and 1/128 had

M AN U

significantly upregulated VE-cad expression in HUVECs when compared to cells grown in medium alone, BG 1/256 or BG 1/512. In addition, the immunofluorescence images taken at the seven day time point indicate that most of the VE-cad was concentrated at cell-cell interfaces in those cells cultured in BG 1/32, 1/64 and 1/128 while VE-cad was dispersed in HUVECs cultured in endothelial cell medium alone, BG 1/256 or BG 1/512 (Fig.6C).

TE D

3.6 Effects of BG on wound healing 3.6.1 Effects of BG on wound closure

To test the effects of BG on wound healing dry powder was applied to excision lesions on

EP

rats. Closure measurements (Fig.S3) show that at day 2 post injury the percentage of wound closure did not vary between control and treated wounds. At day 6 and day12 postoperative,

AC C

however, BG had enhanced wound closure when compared to controls, the highest percentage of wound closure (85% ± 8) being significantly different (P < 0.05) to the control (65% ± 3) at day 12.

Wound healing was further analyzed after H&E staining (Fig.7). At day 6, initial formation of

a neoepidermis, which could be found underneath the eschar, was observed in the BG group wounds (Fig.7 D). Under the eschar was granulation tissue formation, acting as a template for the formation of a vascularized neodermis which began from day 6 onwards. The granulation tissue appears to be more organized in the BG group wounds at day 12 post wounding (Fig.7F). However, in the wound tissue without BG (untreated control), no neoepidermis could be found at this time point despite noticeable granulation tissue being present and an eschar starting to form

ACCEPTED MANUSCRIPT (Fig.7 C). 3.6.2 Effects of BG on neovascularization The impact of BG on neovascularization was analyzed by observing blood vessels within the wound bed following H&S staining of sections. At day 12 postoperative, significantly more blood

RI PT

vessels were present in the bed of BG treated wounds as compared to the control (Fig.S4F). This was confirmed with CD 31 labelling of endothelial cell adhesions (Fig.S4B and E). Those blood vessels appeared in the neodermis underneath the neoepidermis. In contrast, there were much

of normal (unwounded) animals (Fig.S4C and F). 3.6.3 Effects of BG on Cx43 expression in the wound bed

SC

fewer blood vessels in the wound areas of the control group (Fig.S4A and D) or beneath the skin

M AN U

Cx43 in wound sites was immunolabelled at days 2 and day 12 postoperative (Fig. 8). At day 2 postoperative, Cx43 expression was evident in the wound bed of control wounds (Fig. 8A) and BG treated wounds, including between adipocytes (Fig. 8B) adjacent to the wound bed, if anything at slightly reduced levels in both those groups compared to Cx43 expression in normal (unwounded) tissue (Fig.8C). The key finding, however, was that by day 12 postoperative, higher

TE D

expression of Cx43 was found in the BG treated wound beds (Fig. 8E ) than in untreated wounds (Fig.8D) or normal tissues (Fig. 8F). It is of note that in wound sites treated with BG, most of Cx43 positive staining was found in the cells around newly formed blood vessels, as indicated by

EP

arrow heads.

4. Discussion

AC C

Several publications indicate that BG can help wound healing [14, 15] but the mechanism remains unclear. Wound healing is a complex process and it is well accepted that vascularization/angiogenesis is a critical stage in the proliferative phase of wound healing where newly formed blood vessels supply oxygen and nutrients to cells to the wound area and regenerating tissues. This is especially important for the chronic wound healing process when blood vessels have been damaged within the wound site [55-58]. Since BG has been reported to be able to stimulate vascular growth factor secretion from endothelial cells or from the neighboring cells and improve vascularization or angiogenesis of endothelial cells [3, 4, 59], it can be reasoned that BG may enhance wound healing through stimulating blood vessel formation in wound bed.

ACCEPTED MANUSCRIPT However, to date, only vascular growth factors have been studied in attempts to understand BG’s vascularization stimulatory effects but the effects of BG on gap junction between endothelial cells have not been reported. It is well known that vascularization is a complicated process, which involves endothelial cell-cell communication indirectly via various growth factors, but also

RI PT

directly via cell-cell junctions, such as gap junctions [21, 23, 24, 60-62]. As gap junctional communication plays an important role in mediating the behavior of cells involved in wound healing, it is critical to know if BG can affect gap junction and the gap junction mediating cell behaviors. In this study, we have demonstrated that, in addition to stimulating expression of

SC

critical vascular growth factors from endothelial cells, BG ion extracts appear to affect gap junction channels and gap junction related endothelial cell behaviors, in particular, resistance to

during in vivo vascularization.

M AN U

cell loss following hypoxia and increased Cx43 expression and gap junction communication

Cx43 gap junctional communication has been demonstrated to play a crucial role in mediating endothelial cell behavior that is important in wound healing [24, 32, 35, 37, 41, 62, 63]. Although various biomaterials have been applied as wound dressings to assist wound healing, the

TE D

effects of biomaterials on gap junctional cell-cell communication still remain unclear. Faucheux et al. reported that cellulose substrate could modulate Cx43 phosphorylation and induce cell-cell communication between Swiss 3T3 cells, which subsequently affected the biological processes

EP

such as cell migration, proliferation, differentiation or apoptosis [64]. However, this study was limited to the effects of the biomaterial on cell behavior and was not related to wound healing.

AC C

In a wound with damaged blood vessels, the oxygen supply is limited to the cells in/around the wound site. Thus, the cells will experience hypoxia condition, which will affect behaviors of cells, especially the viability of cells. Furthermore, vascularization or angiogenesis is triggered by hypoxia [56, 65] and endothelial cells first need to survive under those conditions. Therefore, the ability for endothelial cells to resist apoptosis is important for angiogenesis in wound healing. In addition, some studies have shown that, when cells or tissues are under stress, hemichannel opening allows the release of signaling molecules into the extracellular space [35, 40, 42]. However, excessive hemichannel opening is not favorable for cell survival [38, 40-42, 69-72]. Following ischemia, Cx43 hemichannel opening contributes to lesion spread, inflammation and direct loss of cells’ ability to osmoregulate, leading to endothelial cell death [32, 41, 69, 72-74],

ACCEPTED MANUSCRIPT while Cx43 mimetic peptides that inhibit hemichannel opening reduced the death of endothelial cells both in vitro and in vivo. The question is whether BG is able to protect HUVECs under hypoxia condition. To answer this question, the hypoxia chamber study was designed according to literatures, where viability of cells following various ischemia conditions was evaluated. Our

RI PT

results demonstrated that, BG indeed enhanced survival rate of HUVECs under hypoxia condition. Interestingly, we also found that reduced PI dye uptake of the cells in the first hour post-hypoxia suggests that BG ion extracts at appropriate concentrations may prevent cell death by preventing hemichannel opening. However, more evidence are required to confirm the assumption of this

SC

mechanism.

Of particular note is that BG ion extracts at 1/64 and 1/128 stimulated [41] functional gap

M AN U

junction communication between HUVECs. This effect was dose dependent with the higher BG ion extract concentration 1/32 and lower extract dilutions 1/256 and 1/512 having no effect on gap junction communication. It is normally assumed that both connexon opening and docking of gap junctions will be perturbed when using extracellular acting connexin mimetic peptides to modulate channel function, but this is also dose dependent with low peptide doses used to block

TE D

hemichannel channel opening without impact on cell-cell coupling, at least in the short term [42, 65, 66]. BG, however, appears to reduce cell death post-injury, but also leads to increased cell-cell communication over a similar concentration range. These findings are consistent with studies

EP

using an intracellular acting mimetic peptide, Peptide9, which mimics the cytoplasmic tail of Cx43 to perturb its interactions with the cytoplasmic loop. Peptide9 prevents hemichannel opening

AC C

whilst retaining gap junction coupling [67]. Since there are only bioactive ions in BG ionic products, and we have no evidence of direct interaction between BG ions and hemichannels, the mechanism through which BG ion extracts affects gap junctional communication and modulates hemichannels appears to be different from that for at least extracellular acting connexin mimetic peptides. We propose that BG ion extract prevents hemichannel opening and affects gap junction communication through an indirect way, possibly by changing the microenvironment (for example pH or ionic composition) around endothelial cells to mediate their behavior, rather than direct binding of BG ions to the hemichannel itself. These mechanisms remain to be elucidated. In addition, we found that the expression of Cx43, VEGF, bFGF and VE-cad in HUVECs could be significantly upregulated after being cultured with BG ion extracts 1/64, 1/128 and 1/256

ACCEPTED MANUSCRIPT for 7 days, all of which are beneficial for angiogenesis and vascularization [24, 68]. In early studies, recombinant VEGF or FGF protein has been used to enhance vascularization [69-71]. However, this method is limited due to several reasons [4]. Recent studies have suggested that stimulation of VEGF and bFGF expression by silicate bioceramic materials provides an alternative effective

way

to

supply

those

angiogenic

factors

to

endothelial

cells

for

RI PT

and

angiogenesis/vascularization [4, 17, 48, 50, 72]. In addition, simultaneous delivery of multiple angiogenic factors is more effective than the delivery of a single angiogenic factor for enhancing vessel density and maturity [73-76]. In this study, BG ion extracts upregulated VEGF and bFGF

SC

expression simultaneously.

The upregulation of VEGF and KDR by BG ion extracts may further trigger KDR association

M AN U

with the VE-cad–Src complex which enhances the VE-cad expression [77]. VE-cad also plays a critical role in different stages of capillary tube formation [77, 78]. Lack of VE-cad may result in disorganized adherens junctions between endothelial cells and as a result the cells might not be able to adhere to each other or communicate sufficiently to form a network structure [68, 79, 80]. In our study, BG ion extracts not only stimulated the expression of VE-cad from HUVEC, but also

TE D

enhanced the concentration of VE-cad at the junction of cells after 7 days in culture, suggesting that BG ion extracts might be able to stabilize newly formed capillaries during later stages of vascularization.

EP

Unlike Cx43 antisense treatments [35], however, we did not observe faster wound closure (epithelialization) during the first two days, but from that point onwards did see enhanced healing

AC C

rates and significantly higher levels of vascularization in the wound bed. The lack of early effect may be owing to time required to get BG ions from the delivered powder form into the wound bed. Interestingly, at day 12, Cx43 expression was markedly increased in the BG treated wound bed and most of that Cx43 appeared to be expressed by endothelial cells forming new blood vessels. This result is inconsistent with the in vitro results where the Cx43 expression in HUVECs was upregulated by BG after the cells were cultured for 7 days. Day 7 to day 12 is the proliferation and angiogenesis stage of wound healing and the upregulation of Cx43 correlated with the enhanced angiogenesis observed during wound healing. Therefore, we propose a possible mechanism for BG to enhance wound healing as follows: In the early stage of wound healing when hypoxia results from the injury, ion products released from

ACCEPTED MANUSCRIPT BG can protect endothelial cells potentially by reducing the open probability of hemichannels. In the migration and proliferation stage, BG stimulates endothelial cells to migrate into wound bed by up-regulating vascular growth factor expression from existing endothelial cells or other types of cells. After the endothelial cells arrive at the wound bed, gap junctional communication

RI PT

between endothelial cells, and possibly between endothelial cell and other types of cells, is increased, stimulating vascularization.

Based upon our in vitro results we can conclude that Si ions contribute significantly to the effect BG has on wound healing. This is because the concentrations of calcium and phosphate ions

SC

in BG extracts are similar to those in normal control medium. In addition, the most effective concentrations of Si ion are in the range of 0.6-1.7 µg/ml, which is consistent with the

M AN U

concentrations (0.6-1.8 µg/ml) obtained in our previous studies on the effects of ionic products from calcium silicates on HUVECs, interactions between HUVECs and fibroblasts as well as interactions between HUVECs and human bone marrow stromal cells [48-50, 81]. However, a higher Si ion concentration (2.5 µg/ml) sometimes also exhibited positive effects on certain growth parameters. We also observed the increased concentration of phosphate ions in BG extracts

TE D

1/8, 1/16 and 1/32. Endothelial cells are very sensitive to the concentration of phosphate ions as the phosphate variations in the plasma will be first sensed by endothelial cells, which may result in modulation of the cells’ function. It has been reported that high concentration of phosphate ions

EP

will result in apoptosis of HUVECs through disrupting the mitochondrial function of endothelial cells [82]. In our results, the BG extracts 1/8, 1/16 and 1/32 did not possess the effective ion

AC C

concentration on proliferation and gap junction communication of endothelial cells. Nonetheless we found that direct application of BG powder to the wound site enhanced

wound healing rate with BG ion delivery moderated by the in vivo environment. It is indeed very difficult to determine the concentration of bioactive ions in the wound area. We also know that the ionic concentration in vivo must be different than in vitro situation, since the in vivo system is a dynamic, buffering and complicated system. Using the in vitro model, we identified the optimal concentration of ions for their angiogenic bioactivity, which indicates that the materials have the bioactivity. However, we need to keep in mind that the optimal in vitro concentration may not be the optimal concentration for in vivo situation. When we directly applied the BG powders on the wounds, the solid/liquid ratio must be different as that of in vitro situation, and considering the

ACCEPTED MANUSCRIPT dynamic condition of the body fluids, the ion concentration must be changing within certain concentration range. On the other side, the reaction of the cells to ions in vivo must also be different as the in vitro condition, since they are surrounded by tissues. Based on the results obtained we can only speculate that, due to the slow release of ions from BG and the dynamic

RI PT

regulation of the body fluid, effective ion concentration will be reached, which facilitated wound healing. Further experimental technique needs to be developed to precisely measure the ion concentration in order to determine the correlation between the in vitro and in vivo ion concentration. In addition, it is worth to indicate that, although we have demonstrated the

SC

stimulatory effect of the bioactive ions on angiogenesis in vitro, a direct addition of bioactive ions to wound may not exert same beneficial effects in vivo as the BG materials did, since the bioactive

M AN U

ions will be easily diluted by the dynamic in vivo system. A long term sustained release of bioactive ions by BG is critical for keeping the wound in a microenvironment with bioactive ions during the wound healing process.

5. Conclusion

In this study, we set out to investigate the effects of BG on gap junctions and connexin

TE D

channel mediated cell behavior as well as on wound healing. We demonstrated that that BG has strong effects on Cx43 gap junction channels and gap junction related behavior of endothelial cells, which contributes to the wound healing stimulatory effects of BG. In vitro studies indicated that

EP

BG ion extracts prevented death of HUVECs following hypoxia, appearing to block hemichannel opening. In addition, BG had stimulatory effects on HUVEC gap junction mediated intercellular

AC C

communication. After cells were cultured for 7 days, Cx43 expression as well as the expression of critical vascular growth factors (VEGF, bFGF and their receptors as well as VE-cad) was upregulated by BG ionic products. In vivo results showed that BG treatment on skin wounds could reduce inflammation and Cx43 expression in wound site at day 2 but increase Cx43 and stimulate vascularization at day 12, resulting in enhanced wound healing. We conclude that BG can enhance wound healing by protecting endothelial cells whilst by enhancing cell-cell communication. BG may have significant potential for wound healing because of its stimulatory effects on vascularization and wound healing through Cx43 mediated endothelial cell behavioral changes.

Acknowledgment

ACCEPTED MANUSCRIPT This work was supported by Natural Science Foundation of China (Grant No.: 81190132 and 31470918) and the Innovation Program of Shanghai Municipal Education Commission (Grant no. 14ZZ032).

References [1] Gerhardt L-C, Boccaccini AR. Bioactive glass and glass-ceramic scaffolds for bone tissue

RI PT

engineering. Materials. 2010;3:3867-910.

[2] Rahaman MN, Day DE, Bal BS, Fu Q, Jung SB, Bonewald LF, et al. Bioactive glass in tissue engineering. Acta Biomater. 2011;7:2355-73.

[3] Day RM. Bioactive glass stimulates the secretion of angiogenic growth factors and angiogenesis in vitro. Tissue Eng. 2005;11:768-77.

SC

[4] Gorustovich AA, Roether JA, Boccaccini AR. Effect of bioactive glasses on angiogenesis: a review of in vitro and in vivo evidences. Tissue Eng Part B Rev. 2010;16:199-207.

[5] Stoor P, Soderling E, Salonen JI. Antibacterial effects of a bioactive glass paste on oral

M AN U

microorganisms. Acta Odontol Scand. 1998;56:161-5.

[6] Hoppe A, Guldal NS, Boccaccini AR. A review of the biological response to ionic dissolution products from bioactive glasses and glass-ceramics. Biomaterials. 2011;32:2757-74. [7] Mortazavi V, Nahrkhalaji MM, Fathi MH, Mousavi SB, Esfahani BN. Antibacterial effects of sol-gel-derived bioactive glass nanoparticle on aerobic bacteria. Journal of biomedical materials research Part A. 2010;94:160-8.

[8] Kawashita M, Tsuneyama S, Miyaji F, Kokubo T, Kozuka H, Yamamoto K. Antibacterial silver-containing silica glass prepared by sol–gel method. Biomaterials. 2000;21:393-8.

TE D

[9] Munukka E, Lepparanta O, Korkeamaki M, Vaahtio M, Peltola T, Zhang D, et al. Bactericidal effects of bioactive glasses on clinically important aerobic bacteria. Journal of materials science Materials in medicine. 2008;19:27-32.

[10] Chen QZ, Blaker JJ, Boccaccini AR. Bioactive and mechanically strong Bioglass-poly(D,L-lactic acid) composite coatings on surgical sutures. Journal of biomedical materials research Part B, Applied

EP

biomaterials. 2006;76:354-63.

[11] Hu S, Chang J, Liu M, Ning C. Study on antibacterial effect of 45S5 Bioglass. Journal of materials science Materials in medicine. 2009;20:281-6.

AC C

[12] Allan I, Newman H, Wilson M. Antibacterial activity of particulate bioglass against supra- and subgingival bacteria. Biomaterials. 2001;22:1683-7. [13] Allan I, Newman H, Wilson M. Particulate Bioglass reduces the viability of bacterial biofilms formed on its surface in an in vitro model. Clin Oral Implants Res. 2002;13:53-8. [14] Gillette RL, Swaim SF, Sartin EA, Bradley DM, Coolman SL. Effects of a bioactive glass on healing of closed skin wounds in dogs. Am J Vet Res. 2001;62:1149-53. [15] Blaker JJ, Nazhat SN, Boccaccini AR. Development and characterisation of silver-doped bioactive glass-coated sutures for tissue engineering and wound healing applications. Biomaterials. 2004;25:1319-29. [16] Leu A, Leach JK. Proangiogenic potential of a collagen/bioactive glass substrate. Pharm Res. 2008;25:1222-9. [17] Leu A, Stieger SM, Dayton P, Ferrara KW, Leach JK. Angiogenic response to bioactive glass promotes bone healing in an irradiated calvarial defect. Tissue Eng Part A. 2009;15:877-85.

ACCEPTED MANUSCRIPT [18] Kyriakidou K, Lucarini G, Zizzi A, Salvolini E, Mattioli Belmonte M, Mollica F, et al. Dynamic co-seeding of osteoblast and endothelial cells on 3D polycaprolactone scaffolds for enhanced bone tissue engineering. J Bioact Compat Polym. 2008;23:227-43. [19] Kirkpatrick CJ, Fuchs S, Unger RE. Co-culture systems for vascularization--learning from nature. Adv Drug Deliv Rev. 2011;63:291-9. [20] Carano RA, Filvaroff EH. Angiogenesis and bone repair. Drug Discov Today. 2003;8:980-9. [21] Lamalice L, Le Boeuf F, Huot J. Endothelial cell migration during angiogenesis. Circ Res.

RI PT

2007;100:782-94.

[22] Munoz-Chapuli R, Quesada AR, Angel Medina M. Angiogenesis and signal transduction in endothelial cells. Cellular and molecular life sciences : CMLS. 2004;61:2224-43.

[23] Senger DR, Davis GE. Angiogenesis. Cold Spring Harb Perspect Biol. 2011;3:a005090.

[24] Wang HH, Su CH, Wu YJ, Li JY, Tseng YM, Lin YC, et al. Reduction of connexin43 in human endothelial progenitor cells impairs the angiogenic potential. Angiogenesis. 2013;16:553-60.

SC

[25] Gartner C, Ziegelhoffer B, Kostelka M, Stepan H, Mohr FW, Dhein S. Knock-down of endothelial connexins impairs angiogenesis. Pharmacological research : the official journal of the Italian Pharmacological Society. 2012;65:347-57.

M AN U

[26] Walker DL, Vacha SJ, Kirby ML, Lo CW. Connexin43 deficiency causes dysregulation of coronary vasculogenesis. Dev Biol. 2005;284:479-98.

[27] Goodenough DA, Goliger JA, Paul DL. Connexins, connexons, and intercellular communication. Annu Rev Biochem. 1996;65:475-502.

[28] Kumar NM, Gilula NB. The gap junction communication channel. Cell. 1996;84:381-8. [29] Yeager M, Nicholson BJ. Structure of gap junction intercellular channels. Curr Opin Struct Biol. 1996;6:183-92. 2004;62:228-32.

TE D

[30] Söhl G, Willecke K. Gap junctions and the connexin protein family. Cardiovascular research. [31] Mese G, Richard G, White TW. Gap Junctions: Basic Structure and Function. J Invest Dermatol. 0000;127:2516-24.

[32] Coutinho P, Qiu C, Frank S, Tamber K, Becker D. Dynamic changes in connexin expression correlate

EP

with key events in the wound healing process. Cell Biol Int. 2003;27:525-41. [33] Churko JM, Laird DW. Gap junction remodeling in skin repair following wounding and disease. Physiology (Bethesda). 2013;28:190-8.

AC C

[34] Mori R, Power KT, Wang CM, Martin P, Becker DL. Acute downregulation of connexin43 at wound sites leads to a reduced inflammatory response, enhanced keratinocyte proliferation and wound fibroblast migration. J Cell Sci. 2006;119:5193-203. [35] Qiu C, Coutinho P, Frank S, Franke S, Law LY, Martin P, et al. Targeting connexin43 expression accelerates the rate of wound repair. Current biology : CB. 2003;13:1697-703. [36] Becker DL, Thrasivoulou C, Phillips AR. Connexins in wound healing; perspectives in diabetic patients. Biochimica et biophysica acta. 2012;1818:2068-75. [37] Coutinho P, Qiu C, Frank S, Wang CM, Brown T, Green CR, et al. Limiting burn extension by transient inhibition of Connexin43 expression at the site of injury. Br J Plast Surg. 2005;58:658-67. [38] Cronin M, Anderson PN, Cook JE, Green CR, Becker DL. Blocking connexin43 expression reduces inflammation and improves functional recovery after spinal cord injury. Molecular and cellular neurosciences. 2008;39:152-60. [39] Rawanduzy A, Hansen A, Hansen TW, Nedergaard M. Effective reduction of infarct volume by gap

ACCEPTED MANUSCRIPT junction blockade in a rodent model of stroke. J Neurosurg 2009;87:916-20. [40] Froger N, Orellana JA, Calvo CF, Amigou E, Kozoriz MG, Naus CC, et al. Inhibition of cytokine-induced connexin43 hemichannel activity in astrocytes is neuroprotective. Molecular and cellular neurosciences. 2010;45:37-46. [41] Danesh-Meyer HV, Kerr NM, Zhang J, Eady EK, O'Carroll SJ, Nicholson LF, et al. Connexin43 mimetic peptide reduces vascular leak and retinal ganglion cell death following retinal ischaemia. Brain. 2012;135:506-20.

RI PT

[42] O'Carroll SJ, Alkadhi M, Nicholson LF, Green CR. Connexin 43 mimetic peptides reduce swelling, astrogliosis, and neuronal cell death after spinal cord injury. Cell Commun Adhes. 2008;15:27-42.

[43] Metcalfe AD, Ferguson MW. Tissue engineering of replacement skin: the crossroads of biomaterials, wound healing, embryonic development, stem cells and regeneration. Journal of the Royal Society, Interface / the Royal Society. 2007;4:413-37.

[44] Choi JS, Leong KW, Yoo HS. In vivo wound healing of diabetic ulcers using electrospun nanofibers

SC

immobilized with human epidermal growth factor (EGF). Biomaterials. 2008;29:587-96. [45] Ueno H, Yamada H, Tanaka I, Kaba N, Matsuura M, Okumura M, et al. Accelerating effects of 1999;20:1407-14.

M AN U

chitosan for healing at early phase of experimental open wound in dogs. Biomaterials. [46] Jayakumar R, Prabaharan M, Sudheesh Kumar PT, Nair SV, Tamura H. Biomaterials based on chitin and chitosan in wound dressing applications. Biotechnol Adv. 2011;29:322-37. [47] Doillon CJ, Silver FH. Collagen-based wound dressing: effects of hyaluronic acid and firponectin on wound healing. Biomaterials. 1986;7:3-8.

[48] Li H, Chang J. Bioactive silicate materials stimulate angiogenesis in fibroblast and endothelial cell co-culture system through paracrine effect. Acta Biomaterialia. 2013;9:6981-91.

TE D

[49] Li H, Xue K, Kong N, Liu K, Chang J. Silicate bioceramics enhanced vascularization and osteogenesis through stimulating interactions between endothelia cells and bone marrow stromal cells. Biomaterials. 2014;35:3803-18.

[50] Li H, Chang J. Stimulation of proangiogenesis by calcium silicate bioactive ceramic. Acta Biomater. 2013;9:5379-89.

EP

[51] Jaffe EA. Culture of human endothelial cells. Transplant Proc 1980;12:49-53. [52] Zhou Y SG, Zheng J, Huang Z, Gao F, Zhang Y, et al. . The protective effects of Egr-1 antisense oligodeoxyribonucleotide

on

cardiac

microvascular

endothelial

injury

induced

by

AC C

hypoxia-reoxygenation. Biochem Cell Biol 2010;88:687-95. [53] Braet K, Aspeslagh S, Vandamme W, Willecke K, Martin PE, Evans WH, et al. Pharmacological sensitivity of ATP release triggered by photoliberation of inositol-1,4,5-trisphosphate and zero extracellular calcium in brain endothelial cells. J Cell Physiol 2003;197:205-13. [54] Cerqueira MT, Pirraco RP, Martins AR, Santos TC, Reis RL, Marques AP. Cell sheet technology-driven re-epithelialization and neovascularization of skin wounds. Acta Biomaterialia. 2014;10:3145-55. [55] Kaully T, Kaufman-Francis K, Lesman A, Levenberg S. Vascularization--the conduit to viable engineered tissues. Tissue Eng Part B Rev 2009;15:159-69. [56] Rouwkema J, Rivron NC, van Blitterswijk CA. Vascularization in tissue engineering. Trends Biotechnol. 2008;26:434-41. [57] Hudon V, Berthod F, Black AF, Damour O, Germain L, Auger FA. A tissue-engineered endothelialized dermis to study the modulation of angiogenic and angiostatic molecules on

ACCEPTED MANUSCRIPT capillary-like tube formation in vitro. The British journal of dermatology. 2003;148:1094-104. [58] Laschke MW, HARDER Y, AMON M, Martin I, FARHADI J, RING A, et al. Angiogenesis in Tissue Engineering: Breathing Life into Constructed Tissue Substitutes. Tissue Eng Part A. 2006;12:2093-104. [59] Day RM, Boccaccini AR, Shurey S, Roether JA, Forbes A, Hench LL, et al. Assessment of polyglycolic acid mesh and bioactive glass for soft-tissue engineering scaffolds. Biomaterials. 2004;25:5857-66. [60] Risau W, Flamme I. Vasculogenesis. Annual review of cell and developmental biology. 1995;11:73-91.

RI PT

[61] Dejana E. Endothelial cell-cell junctions: happy together. Nat Rev Mol Cell Biol. 2004;5:261-70.

[62] Gartner C, Ziegelhoffer B, Kostelka M, Stepan H, Mohr FW, Dhein S. Knock-down of endothelial connexins impairs angiogenesis. Pharmacological Research. 2012;65:347-57.

[63] Greaves NS, Ashcroft KJ, Baguneid M, Bayat A. Current understanding of molecular and cellular mechanisms in fibroplasia and angiogenesis during acute wound healing. Journal of Dermatological Science. 2013;72:206-17.

SC

[64] Faucheux N, Zahm JM, Bonnet N, Legeay G, Nagel MD. Gap junction communication between cells aggregated on a cellulose-coated polystyrene: influence of connexin 43 phosphorylation. Biomaterials. 2004;25:2501-6.

M AN U

[65] O'Carroll SJ, Gorrie CA, Velamoor S, Green CR, Nicholson LFB. Connexin43 mimetic peptide is neuroprotective and improves function following spinal cord injury. Neuroscience Research. 2013;75:256-67.

[66] Davidson JO, Green CR, Nicholson LF, O'Carroll SJ, Fraser M, Bennet L, et al. Connexin hemichannel blockade improves outcomes in a model of fetal ischemia. Ann Neurol. 2012;71:121-32. [67] Wang N, De Vuyst E, Ponsaerts R, Boengler K, Palacios-Prado N, Wauman J, et al. Selective inhibition of Cx43 hemichannels by Gap19 and its impact on myocardial ischemia/reperfusion injury.

TE D

Basic Res Cardiol. 2013;108:309.

[68] Bai F, Zhang J, Wang Z, Lu J, Chang J, Liu J, et al. The effect of pore size on tissue ingrowth and neovascularization in porous bioceramics of controlled architecture in vivo. Biomed Mater. 2011;6:015007.

[69] Seghezzi G, Patel S, Ren CJ, Gualandris A, Pintucci G, Robbins ES, et al. Fibroblast growth factor-2

EP

(FGF-2) induces vascular endothelial growth factor (VEGF) expression in the endothelial cells of forming capillaries: an autocrine mechanism contributing to angiogenesis. J Cell Biol. 1998;141:1659-73.

AC C

[70] Kanczle JM, Ginty PJ, Barry JJ, Clarke NM, Howdle SM, Shakesheff KM. The effect of mesenchymal populations and vascular endothelial growth factor delivered from biodegradable polymer scaffolds on bone formation. Biomaterials 2008;29:1892–900. [71] Sohier J, Moroni L, van Blitterswijk C, de Groot K, Bezemer JM. Critical factors in the design of growth factor releasing scaffolds for cartilage tissue engineering. Expert Opin Drug Deliv. 2008;5:543-66.

[72] Keshaw H, Forbes A, Day RM. Release of angiogenic growth factors from cells encapsulated in alginate beads with bioactive glass. Biomaterials. 2005;26:4171-9. [73] Asahara T, Bauters C, Zheng LP, Takeshita S, Bunting S, Ferrara N, et al. Synergistic effect of vascular endothelial growth factor and basic fibroblast growth factor on angiogenesis in vivo. Circulation. 1995;92:II365-71. [74] Cao R, Brakenhielm E, Pawliuk R, Wariaro D, Post MJ, Wahlberg E, et al. Angiogenic synergism, vascular stability and improvement of hind-limb ischemia by a combination of PDGF-BB and FGF-2.

ACCEPTED MANUSCRIPT Nat Med. 2003;9:604-13. [75] Yoshiji H, Kuriyama S, Yoshii J, Ikenaka Y, Noguchi R, Hicklin DJ, et al. Synergistic effect of basic fibroblast growth factor and vascular endothelial growth factor in murine hepatocellular carcinoma. Hepatology. 2002;35:834-42. [76] Goto F, Goto K, Weindel K, Folkman J. Synergistic effects of vascular endothelial growth factor and basic fibroblast growth factor on the proliferation and cord formation of bovine capillary endothelial 1993;69:508-17.

RI PT

cells within collagen gels. Laboratory investigation; a journal of technical methods and pathology. [77] Carmeliet P, Collen D. Molecular basis of angiogenesis. Role of VEGF and VE-cadherin. Ann N Y Acad Sci. 2000;902:249-62; discussion 62-4.

[78] Wallez Y, Vilgrain I, Huber P. Angiogenesis: the VE-cadherin switch. Trends Cardiovasc Med. 2006;16:55-9.

[79] Ukropec JA, Hollinger MK, Woolkalis MJ. Regulation of VE-cadherin linkage to the cytoskeleton in

SC

endothelial cells exposed to fluid shear stress. Exp Cell Res. 2002;273:240-7.

[80] Wright TJ, Leach L, Shaw PE, Jones P. Dynamics of vascular endothelial cadherin and b-catenin Exp Cell Res 2002;280:159-68.

M AN U

localization by vascular endothelial growth factorinduced angiogenesis in human umbilical vein cells. [81] Zhai W, Lu H, Chen L, Lin X, Huang Y, Dai K, et al. Silicate bioceramics induce angiogenesis during bone regeneration. Acta Biomater. 2012;8:341-9.

[82] Di Marco GS, Hausberg M, Hillebrand U, Rustemeyer P, Wittkowski W, Lang D, et al. Increased inorganic phosphate induces human endothelial cell apoptosis in vitro. American journal of physiology

AC C

EP

TE D

Renal physiology. 2008;294:F1381-7.

M AN U

Figure S1. SEM images of as obtained BG particles.

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

(This figure is a single column fitting image.)

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure S2. BG ion extracts appear to prevent connexin hemichannel opening in HUVECs following exposure to hypoxic conditions. Following three hours hypoxia HUVECs were placed

TE D

back into normal medium and after just one hour PI uptake (assumed to be through open hemichannels) was assessed. The percentage of dye uptake into HUVECs cultured with different medium following exposure to hypoxic conditions is shown. Control means that the HUVECs

EP

were cultured under normal conditions. ECM means that the HUVECs were cultured with ECM after being exposed to hypoxic conditions. BG 1/32-1/512 means that HUVEC were cultured with

AC C

BG ion extracts diluted with ECM at different ratios after being exposed to hypoxic conditions. BG ion extracts diluted at 1/32 to 1/256 appears to reduce hemichannel opening in HUVECs following exposure to hypoxic conditions.

(This figure is a single column fitting image.)

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure S3. (A) Gross observation of rat excision wound healing at different time points. (B) Would closure at different time points, calculated from the size of the wound. BG powder treatment

TE D

accelerated wound closure as compared to that of control group. The wounds healed at similar rates during the first two days but BG treated wounds then heal faster, reaching statistical

EP

significance from day 6 onwards.

AC C

(This figure is a single column fitting image.)

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure S4. Blood vessels (arrow heads) immunolabelled for CD 31 and viewed in sections from tissue samples after rat excision wounds were treated with or without BG powder for 12 days.

TE D

More blood vessels were present in wound tissue treated with BG powders (B and E) than in those without BG powder (A and D). Images of (C) and (F) show the blood vessels in normal tissue.

AC C

EP

(This figure is a single column fitting image.)

ACCEPTED MANUSCRIPT Table S1 Gene sequence of primers used in this study. Gene bank

Primer sequences

Tm (oC)

VEGF165

AB_021221

F: 5’TGCGGATCAAACCTCACCA 3’ R: 5’CAGGGATTTTTCTTGTCTTGCT 3’

58

KDR

NM_002253

F: 5’ GTGATCGGAAATGACACTGGAG 3’ R: 5’ CATGTTGGTCACTAACAGAAGCA 3’

60

bFGF

NM_002006.4

F: 5’ CAATTCCCATGTGCTGTGAC 3’ R: 5’ ACCTTGACCTCTCAGCCTCA 3’

61

bFGFR

NM_021923

F: 5’GACGGCTCCTACCTCAA 3’ R: 5’GCTGTAGCCCATGGTGTTG 3’

60

VE-cad

NM_001795

F: 5’GGCTCAGACATCCACATAACC 3’ R: 5’CTTACCAGGGCGTTCAGGGAC 3’

63

Cx43

NM_000165

F: 5’ GGA GGG AAG GTG TGG CTG TC 3’ R: 5’ GGC AGG GCT CAG CGC ACC AC 3’

62

GAPDH

NM_002046

F: 5’GATTTGGTCGTATTGGGCG 3’ R: 5’ CTGGAAGATGGTGATGG 3’

SC

M AN U

TE D EP AC C

RI PT

Gene

60

ACCEPTED MANUSCRIPT

Table S2 Ion concentrations of BG extracts diluted at the different ratios. Si (µg/ml)

P (µg/ml)

ECM

57.23±0.25

0.18±0.12

19.23±0.34

BG 1/8

53.83±0.35

8.41±0.18

33.60±045

BG 1/16

54.72±0.38

5.09±0.12

27.92±0.25

BG 1/32

55.89±0.34

2.51±0.11

BG 1/64

56.67±0.42

1.69±0.12

BG 1/128

56.72±0.56

0.84±0.22

BG 1/256

56.87±0.32

0.55±0.12

BG 1/512

57.06±0.44

0.37±0.11

RI PT

Ca (µg/ml)

24.00±0.22 22.34±0.32 21.16±0.34 20.91±0.25

AC C

EP

TE D

M AN U

SC

21.00±019

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

Figures and captions

Figure 1. Effects of BG ion extracts on HUVEC proliferation. BG ion extracts diluted at ratios from 1/32 to 1/512 do not effect cell proliferation and no sign of toxicity.

AC C

EP

TE D

(This figure is a single column fitting image.)

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 2. BG ion extracts are protective for HUVECs exposed to hypoxic conditions. Following 3

M AN U

hours in hypoxic media HUVECs were placed back into normal medium with and without BG ions for a further 6 hours. They were then stained with a Live-dead assay kit. Representative images are shown in (A). The percentage of viable HUVECs cultured with different medium after being exposed to hypoxic conditions (relative to those cultured under control conditions) were counted and are shown in (B). Control means that the HUVECs were cultured under normal

TE D

conditions without hypoxia. ECM means that the HUVECs were cultured with ECM after being exposed to hypoxic conditions. BG 1/32-1/512 means that HUVEC were cultured with BG ion extracts diluted with ECM at different ratios and after being exposed to hypoxic conditions. Cell

EP

viability assays indicate that BG ion extracts diluted at 1/32 to 1/256 reduce the loss of HUVECs

AC C

following exposure to hypoxia.

(This figure is a 1.5-column fitting image.)

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 3. Calcein-AM cell settlement–dye spread assay used to investigate the level of gap junction communication between HUVECs cultured in media containing various concentrations of BG ions. Quantified dye spread shows that BG ion extracts diluted at 1/64 and 1/128 stimulate gap

TE D

junctional communication between HVUECs with accelerated Calcein-AM dye spread from loaded HUVECs to those in the underlying confluent monolayer.

AC C

EP

(This figure is a single column fitting image.)

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 4. The expression of critical genes associated with wound repair assessed in HUVECs after being cultured with different media for, 1, 3 and 7 days. (A) and (B): bFGF and VEGF expression

TE D

in HVUECs. (C) and (D): bFGFR and KDR expression in HVUECs. The Y axes indicates the gene expression in different samples relative to GAPDH and compared to the same gene expression in HUVEC cultured with ECM for 1 day. BG ion extracts diluted at 1/64-1/256

EP

stimulated bFGF and VEGF and the expression of their respective receptors in HUVECs.

AC C

(This figure is a 1.5-column fitting image.)

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 5. Immunofluorescence labelling showing KDR protein expression and localization in HUVECs cultured with different media for 3 days. BG ion extracts diluted at 1/64-1/256

TE D

stimulates KDR expression in HUVECs.

AC C

EP

(This figure is a single column fitting image.)

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 6. Cx43 gene expression in HUVECs cultured with different medium for 7 days, assessed using Q-RT-PCR (A). VE-cad expression in HUVECs cultured with different medium for 3 and 7

TE D

days, detected using Q-RT-PCR (B). The Y axes indicates the gene expression in different samples relative to GAPDH and compared to the same gene expression in HUVEC cultured with ECM for 1 day. VE-cad expression in HUVECs cultured with ECM and BG extracts diluted at 1/128 for 7 days detected using immunofluorescence labelling (C). Both images (protein localization and

AC C

HVUECs.

EP

levels) and graph (mRNA levels) show that BG ion extracts stimulate VE-cad expression in

(This figure is a 1.5-column fitting image.)

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 7. HE staining of sections from tissue samples after rat excision wounds were treated with or without BG powder for 2 days (A,B), 6 days (C,D) and 12 days (E,F). ES, eschar; NE,

EP

neoepidermis; ND, neodermis; GT, granulation tissue. Functional blood vessels have a clearly defined lumen containing red blood cells (arrows). At day 12, more blood vessels were found in

AC C

wounds treated with BG (F) than that in the control group (E).

(This figure is a single column fitting image.)

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 8. Cx43 (arrow heads) immunolabelled in sections from tissue samples after rat excision wounds treated with or without BG powder for 2 and 12 days. At day 2, less positive staining is

TE D

observed in deep dermal wound tissue treated with BG powder (B) than in wounds without BG powder (A). In contrast, by day 12, more positive Cx43 staining is observed in deep dermal wound tissue treated with BG powder (E) than in wounds without BG powder (D). Images of (C),

EP

and (F) show the Cx43 in normal tissue. The region of interest is shown from just below the

AC C

surface of the wound.

(This figure is a 1.5-column fitting image.)