Effects of Co-Culture of Graphene Oxide Scaffolds with Different Concentrations and Umbilical Mesenchymal Stem Cells on the Proliferation and Differentiation of Stem Cells

The of the is to enable cells to in the while promote their and differentiation ability. mesenchymal stem cell, the could provide an for them to their phenotype, and all necessary and to basic characteristics large rigidity and biodegradability. Thus, the two-dimensional may be mesenchymal stem cell


Methods
In this study, the effect of GO on the value-added differentiation activity of mesenchymal stem cell was systematically investigated.

Results
It was found that low concentration of GO and su cient concentration of umbilical cord mesenchymal stem cells are suitable for the second Co-culture. Furthermore, the addition of hyaluronic acid will make this culture more evenly distributed.

Conclusions
The adsorption of GO on umbilical cord mesenchymal stem cells can also make the two closely linked, which avoids the impact of animal joint activities on cells.

Background
Recently, the signi cant advantages of graphene oxides (GO) used as cell scaffold materials have attracted tremendous interests due to their large surface areas, friendly biological compatibility and good hydrophility. [1][2][3][4][5] As an ideal seed cell in tissue engineering, mesenchymal stem cell (MSC) has been widely used in the eld of regenerative medicine. [6][7][8][9][10][11] MSC has the effect of regulating local immunity and improving the in ammatory environment. At the same time, MSC is low in immunogenicity, which will not bring risks to transplantation. [12][13][14][15][16][17] Many early studies have con rmed that MSC transplanted in damaged structures can promote the repair effect of damaged areas. Up to now, the possibility of GO combined with MSC as the cell scaffold has also been initially explored. [18][19][20][21][22] However, there are few related studies and bone marrow stromal cells (BMSC) are mostly used as seed cells in the research.
At present, umbilical cord mesenchymal stem cells (UCMSC) has been employed to try to replace bone marrow stromal cells (BMSC). [23][24][25] On one hand, the human-derived BMSC needs to puncture the human bone marrow, which brings pain, safety risks and ethical problems to the human to some extent.
On the other hand, bone marrow-derived MSCs are actually very small in content. The extremely low proportion of MSCs in bone marrow as well as the low-activity have a great impact on the results of research. Although some studies have con rmed that in vitro cultured MSCs and GO are biocompatible, they are also limited to the observation of the effect of GO on MSC survival without further in-depth studies and speci c dose-effect relationships.
The dynamic knee joint is one of the most important structures that carries the human body's movements. [26][27][28] The joint cavity must maintain the stability of mechanical and biochemical microenvironment. MSC inhibits the in ammatory response through paracrine effects. Meanwhile, MSC releases growth factors such as platelet-derived growth factor (PDGF), transforming growth factor (TGFβ), vascular endothelial growth factor (VEGF) and insulin-like growth factor (IGF). These growth factors could improve the biochemical micro-environment of the knee joint and promote repair of cartilage tissue damage.
The problems of cell ow and low cell survival rate can be effectively solved by using cell scaffolds. The GO scaffold with favorable hydrophility, biocompatibility, ductility and damping properties can increase the electrical activity of cells. [29][30][31] In addition to being equipped with MSC, GO scaffold can also be combined with sodium hyaluronic acid (HA) to prepare materials with lubricating effect on the joint cavity. Furthermore, GO can also improve the joint as lubricants. Herein, in this experiment, the culture effect of UCMSC with GO was initially explored in order to provide a novel strategy for the treatment of osteoarthritis.

Materials and instruments
GO is purchased from C6G6Technology Co., Ltd, China. The basic characterization of GO is in the Supporting Information ( Figure S1 and Figure S2). The 3rd or 4th generation human umbilical cord mesenchymal stem cells (Boya stem cell technology co. LTD). DMEM/F12 medium and HA injection 2.5 ml/piece (Japan, H20050370). High-throughput multi-sample tissue grinding machine (Nanjing, Xianou Instrument Manufacturing Co., Ltd.). Electronic platform scale (Switzerland, Mettler Toledo). Constant temperature and humidity incubator (SHEL LAB in the United States). OLYMPUS photomicroscope (OLYMPUS in Japan). ADAM automatic cell counter (NanoEnTek, Korea).

Mixture of GO and HA
GO is placed in a small reagent tube and put it into a tissue grinder. The control frequency is set as 70 Hz and the stop time is 60 s. After the machine stops running, remove the reagent tube, put it into a centrifuge, and centrifuge Operation: According to the needs of the experiment, use an electronic balance to weigh the masses of graphene oxide required by 4 groups of GO solution, and put them into four small reagent tubes labeled with groups, put them on the ultraviolet operating table, and use 1 ml without The pipette tip of the bacteria tube transfers a part of the HA injection in the unsealed syringe containing 2.5 ml of HA injection into a small reagent tube lled with GO. The GO combined HA solution is con gured to two concentrations: 15ug/ml GO + 0.5% HA and 30 ug/ml GO + 0.25% HA. After mixing, use the pipette Transfer all the liquid guns to the large test tube cap, and then transfer the HA injection solution in the remaining syringes to the large test tube caps of the same group. Use the pipette head to mix them thoroughly and turn on the ultraviolet light., Perform UV sterilization for 35 min, and then use a pipette to transfer the mixture in the large test tube cap into the original syringe.

Isolation, culture and identi cation of human UCMSC
Take a healthy full-term fetal umbilical cord, rinse it thoroughly with PBS, remove the umbilical arteriovenous vein under sterile conditions, cut the remaining interstitial tissue (Walton's gel) into 1.0-2.0 mm size tissue blocks, and atten the tissue blocks on In the cell culture ask, add an appropriate amount of DMEM/F12 culture solution containing 10% fetal bovine serum in volume, and place in 37 °C, 5% CO 2 volume incubator for incubation. Replace the culture solution according to the cell growth rate.
When the cells have reached the bottom of the culture ask, remove the tissue blocks, pass them down according to the number of cells, observe with an inverted microscope and take a video. Take the 3rd or 4th generation human UCMSC, digest with 0.25% trypsin, centrifuge at 1200 r / min for 5 minutes, count the cells, and use about 2 × 10 cells per tube. Use 0.1% sodium azide and 0.5 Wash the cells twice with PBS, resuspend the cells in PBS, add mouse anti-rat CD34, CD45, CD90, and CD105 primary antibodies (1:50 dilution), leave them at 4 ℃ for 30 min, wash them twice with PBS, add isocyanate Rabbit antimouse IgG secondary antibody labeled with uorescein thiocyanate, placed at 4 ℃ for 30 min, washed twice with PBS, resuspended the cells with PBS without BSA, and detected cell surface markers CD34, CD45, CD90 and CD105 expression.

Experimental grouping and processing
The experiment was divided into 4 groups. GO was mixed with HA injection after being treated with a tissue grinder and a centrifuge, and a mixture of 10 ug/ml, 20 ug/ml, 30 ug/ml and 40 ug/ml was used as a culture medium. Mark the cells separately, divide the human UCMSC into four groups of approximately equal amounts, mark the serial numbers, and transfer them to the corresponding medium with a pipette for cultivation.

Effect of the mixture of GO and HA injection on the proliferation of human UCMSC
The four concentration mixtures were placed in 96-well culture plates, and 100 µL of medium was added to each well. The inoculated 96-well culture plate was placed in a constant temperature incubator (5% CO 2 , 100% humidity, 37 ℃ constant temperature) for 24 hours, and observed under continuous microscope.

In vitro culture of UCMSC and GO particles
UCMSC and GO were co-cultured for 24 hours. It was observed that UCMSC grew normally and GO did not cause a large number of deaths of UCMSC cells. Meanwhile, GO nanosheets began to aggregate while UCMSC cells tended to aggregate toward GO until they gradually wrapped GO nanosheets (Fig. 1).
In Fig. 2, UCMSC and GO were co-cultured for 72 hours. UCMSC grew normally. GO did not cause a large number of deaths of UCMSC cells. The tendency of UCMSC cells to further aggregate toward GO was observed and climbed toward GO particles to wrap them into clusters. UCMSC began to grow in three dimensions and stimulated proliferation.
UCMSC and GO were co-cultured for 7 days. UCMSC grew normally. GO did not cause a large number of deaths of UCMSC cells. It was observed that almost all UCMSC cells aggregated to GO. The GO particles were tightly wrapped to form a group structure of GO + UCMSC (Fig. 3) According to the above experimental results, following results can be concluded. Firstly, UCMSC can coexist with GO in vitro. Relatively large GO was chosen to co-culture with MSC. It can be observed that not only a large number of cell death did not occur after one week of culture, but the situation of cell proliferation appeared. This phenomenon indicates that GO at this concentration is not cytotoxic to UCMSC and it can promote the growth of UCMSC to some certain extent. Secondly, GO nanoparticles are attractive to UCMSC. Cells farther away from the GO particles can also be attracted and wrapped by them, forming a three-dimensional approximately spherical structure. This phenomenon indicates MSC and GO can form a good binding without resorting to other material structures, so as not to affect the treatment effect.

Culture results of UCMSC and GO with different concentrations
In Fig. 4, UCMSC was co-cultured with GO with different concentrations for 48 h. It was observed that the GO concentrations of 30 mg/ml, 10 mg/ml, 300 ug/ml, 200 ug/ml and 100 ug/ml were too high, which is not suitable for co-culture with UCMSC. GO was observed to aggregate into small clumps and was surrounded by cells after co-culturing of UCMSC and GO (15 ug/ml and 30 ug/ml) for 48 h (Fig. 5).
After Co-culturing of GO (30 ug/ml) and UCMSC with different concentrations for 48 h, it was observed that the cell concentration of 4.0 × 10 4 /ml and 2.0 × 10 4 /ml were lower in both groups (Fig. 6).
The above Co-culturing results of GO and UCMSCs with different concentrations con rmed that excessively high concentrations of GO are not conducive to the in vitro culture of cells. This may be ascribed to the high concentration of GO, which compressed the cell survival space. The research results show that concentrations of 30ug/ml and below are suitable for coexistence with UCMSC. However, if the concentration of GO is too low, it may reduce its lubrication and protection effect on articular cartilage surface. On the other hand, the low concentration of UCMSC cells under the low concentration of GO environment can not re ect the migration and proliferation effect. It is due to a certain distance limitation of GO on the cell forms a separate culture of GO and UCMSC, which cannot form a co-cultivation environment. However, it does not mean that the higher the density of UCMSC is better. Actually, the higher the density of inoculation will also cause cell death. Therefore, GO needs a low concentration environment to ensure its safety. On the contrary, UCMSC must at least ensure a su cient concentration to form a co-culture system.

Culture results of UCMSC and GO nanoparticle lubricants
UCMSC was mixed with a granular lubricant with a concentration of 15ug/ml GO + 0.5% HA and 30ug/ml GO + 0.25% HA for 7 days (Fig. 7). It was observed that GO showed a relatively uniform distribution while HA with a low concentration. Also, UCMSC@GO particles are evenly distributed and can promote the growth and proliferation of UCMSC. The 30 ug/ml GO + 0.25% HA group had a slightly stronger promotion effect on UCMSC than the 15ug/ml group (Fig. 8). Meanwhile, low concentration of HA was more conducive to the uniform distribution of UCMSC + GO structure. The 24 h culture results in UCMSC culture dishes reveals that the detectable cell concentration in the saline group was 4.45 × 10 6 /ml and the cell survival rate was 8%. The detectable cell concentration in the GO particle lubricant group was 3.41 × 10 6 /ml and the cell survival rate 13% (Fig. 9).
The above results indicate the relationship between the ratio of GO and HA mixed lubricants and the culture of UCMSC. Sodium hyaluronate injection is a macromolecular structure and has a high viscosity. Thus, it will have a certain effect on the migration of MSC and also hinder the aggregation to GO. Therefore, the concentration of HA should not be too high. However, HA also hinders the aggregation of GO, which is bene cial to the uniform distribution of GO. Small size of GO is more safe and less toxic than large size of GO. Therefore, the mixing of HA is conducive to the long-term uniform distribution of UCMSC + GO. Furthermore, low concentration of HA is more suitable for co-cultivation of the two.

Serum NO
The difference between GO + MSC group and blank group in group A was statistically signi cant (P < 0.01), and there was no signi cant difference between MSC group, GO group and blank group (P > 0.05). The difference was not statistically signi cant (P > 0.05). The difference between GO + MSC group and blank group in group B was statistically signi cant (P < 0.01), the difference between MSC group, GO group and blank group was statistically signi cant (P < 0.05). The difference between GO + MSC group and MSC group was It has statistical signi cance (P < 0.05). Compared between groups, the difference between GO + MSC group, MSC group and GO group was not statistically signi cant (P > 0.05), and the difference between blank group was statistically signi cant (P < 0.05) ( Table 1). Compared between groups, the difference between GO + MSC group and MSC group was statistically signi cant (P < 0.01), and the difference between GO group and blank group was not statistically signi cant (P > 0.05) ( Table 2).

Serum GAG
In group A, the difference between GO + MSC group and blank group was statistically signi cant (P < 0.01), the difference between MSC group and blank group was statistically signi cant (P < 0.05), and the difference between GO group and blank group was not statistically signi cant. Signi cance (P > 0.05).
The difference between GO + MSC group and MSC group was statistically signi cant (P < 0.01). In group B, the difference between GO + MSC group, MSC group and blank group was statistically signi cant (P < 0.01), the difference between GO group and blank group was statistically signi cant (P < 0.05); comparison between GO + MSC group and MSC group. The difference was statistically signi cant (P < 0.05).
Compared between groups, the difference between the blank group and the GO + MSC group was statistically signi cant (P < 0.01), and the difference between the MSC group and the GO group was not statistically signi cant (P > 0.05) ( Table 3). In group A, the difference between GO + MSC group, MSC group, GO group and blank group was statistically signi cant (P < 0.01); the difference between GO + MSC group and MSC group was statistically signi cant (P < 0.01). In group B, the difference between GO + MSC group, MSC group, GO group and blank group was statistically signi cant (P < 0.01); the difference between GO + MSC group and MSC group was statistically signi cant (P < 0.05).
Compared between groups, the difference between GO + MSC group was statistically signi cant (P < 0.01), the difference between blank group was statistically signi cant (P < 0.05), the difference between MSC group and GO group was not statistically signi cant (P > 0.05) ( Table 4).

Serum TNF-α
In group A, the difference between GO + MSC group, MSC group, GO group and blank group was statistically signi cant (P < 0.01); the difference between GO + MSC group and MSC group was statistically signi cant (P < 0.01). In group B, the difference between GO + MSC group, MSC group, GO group and blank group was statistically signi cant (P < 0.01); the difference between GO + MSC group and MSC group was statistically signi cant (P < 0.05).
Compared between groups, the difference between GO + MSC group was statistically signi cant (P < 0.01), the difference between blank group was statistically signi cant (P < 0.05), the difference between MSC group and GO group was not statistically signi cant (P > 0.05) ( Table 5).

Conclusions
In summary, we have con rmed that GO can coexist with UCMSC in vitro and play the role of adsorption and promotion. Cell survival and growth under different concentrations of environment were preliminary discussed. It was found that low concentration of GO and su cient concentration of UCMSC are suitable for the second Co-culture. Furthermore, the addition of HA will make this culture more evenly distributed. The adsorption of GO on UCMSC can also make the two closely linked, which avoids the impact of animal joint activities on cells. Meanwhile, two KOA animal models were selected in this experiment. The modi ed Hulth + cartilage defect model in group A is the main sports injury model. For the current multiple ligament tears and meniscus destruction, this injury will cause the intra-articular in ammation to gradually develop into KOA The results show that the role of the three-dimensional mesh scaffold of this kind of GO particles is more important, which can make UCMSC better adhere to the cartilage defect area, so as to grow and proliferate. Can make UCMSC better adhere to cartilage defect areas. The model of group B is a model of cartilage degeneration induced by chemical factors. It can be seen that cartilage has different degrees of necrosis, which is relatively close to the pathogenesis of degenerative osteoarthritis in clinic. GAG, IL-6, and TNF-α have a statistically signi cant difference compared with the blank group, which also shows that the use of GO particle lubricants to carry UCMSC has a better therapeutic effect. UCMSC loaded with graphene oxide can promote the chondrocyte secretion of two knee osteoarthritis animal models, reduce the level of in ammation in the joints, and play a role in cartilage repair. Our ndings may provide some help for the e cacy of subsequent knee osteoarthritis (KOA) animal models exploration.

Declarations
Ethics approval and consent to participate Not aplicable.

Consent for publication
Not aplicable.

Availability of data and materials
The data that supports the ndings of this study are available within the manuscript and its supplementary material.

Competing interests
All the authors declare no con icts on nancial interests of the manuscript. Authors' contributions AL came up the basic ideas and analyzed the data of this manuscript. SG and YW conducted this experiment. QW and YY were major contributors in writing the manuscript. All authors read and approved the nal manuscript.  UCMSC with different concentrations of GO and co-culture for 48h (40× magni cation).

Figure 5
UCMSC co-cultured with 15ug/ml GO, 30ug/ml GO for 48h (The picture above is magni ed 40 times and the picture below is magni ed 100 times).
Page 17/19 Figure 6 Different concentrations of UCMSCs co-cultured with 30 ug/ml GO for 48h (The picture on the left is magni ed 40 times, and the picture on the right is magni ed 100 times). UCMSC co-cultured with 15 ug/ml+0.5%HA and 30 ug/mlGO+0.25%HA for 24h (The picture above is magni ed 40 times and the picture below is magni ed 100 times).

Figure 8
UCMSC co-cultured with 15ug/mlGO+0.5%HA for 7d (The picture on the left is magni ed 40 times, and the picture on the right is magni ed 100 times).