The Effects of Hypoxia-Reoxygenation in Mouse Digital Flexor Tendon-Derived Cells

Objective Ischemia-reperfusion injury refers to the exacerbated and irreversible tissue damage caused by blood flow restoration after a period of ischemia. The hypoxia-reoxygenation (H/R) model in vitro is ideal for studying ischemia-reperfusion injury at the cellular level. We employed this model and investigated the effects of cobalt chloride- (CoCl2-) induced H/R in cells derived from mouse digital flexor tendons. Materials and Methods Various H/R conditions were simulated via treatment of tendon-derived cells with different concentrations of CoCl2 for 24 h, followed by removal of CoCl2 to restore a normal oxygen state for up to 96 h. Cell viability was measured using the Cell Counting Kit-8 (CCK-8) assay. Cell growth was determined via observation of cell morphology and proliferation. Oxidative stress markers and mitochondrial activity were detected. The expression levels of hypoxia-inducible factor- (HIF-) 1α, vascular endothelial growth factor-A (VEGF-A), collagen I, and collagen III were determined using Western blot (WB), real-time PCR, and immunofluorescence staining. Cellular apoptosis was analyzed via flow cytometry, and the expression of apoptosis-related proteins Bax and bcl-2 was examined using WB. Results The cells treated with low concentrations of CoCl2 showed significantly increased cell viability after reoxygenation. The increase in cell viability was even more pronounced in cells that had been treated with high concentrations of CoCl2. Under H/R conditions, cell morphology and growth were unchanged, while oxidative stress reaction was induced and mitochondrial activity was increased. H/R exerted opposite effects on the expression of HIF-1α mRNA and protein. Meanwhile, the expression of VEGF-A was upregulated, whereas collagen type I and type III were significantly downregulated. The level of cellular apoptosis did not show significant changes during H/R, despite the significantly increased Bax protein and reduced bcl-2 protein levels that led to an increase in the Bax/bcl-2 ratio during reoxygenation. Conclusions Tendon-derived cells were highly tolerant to the hypoxic environments induced by CoCl2. Reoxygenation after hypoxia preconditioning promoted cell viability, especially in cells treated with high concentrations of CoCl2. H/R conditions caused oxidative stress responses but did not affect cell growth. The H/R process had a notable impact on collagen production and expression of apoptosis-related proteins by tendon-derived cells, while the level of cellular apoptosis remained unchanged.


Introduction
Tendon injury is frequently accompanied by a lack of blood supply, exposing the tendon to a hypoxic condition. Oxygen tension, which is influenced by blood supply, is vital to cell survival and proliferation. In vitro studies have shown that low oxygen tension enhances the expansion capacity of newborn pig tenocytes without affecting the cellular phenotype and functions [1], and hypoxic conditions significantly improve the cell proliferation and self-renewal capacity of human tendon stem cells [2,3]. Other studies, however, have shown that hypoxia inhibits cell proliferation and alters the synthesis of matrix components in synovial tissue [4]. Cobalt chloride-(CoCl 2 -) induced hypoxia has been shown to alter the expression of the bcl-2 family proteins and trigger caspase cascade apoptosis [5,6]. Collectively, the conflicting results in these studies underscore the complicated effects of hypoxia on cell growth.
Most tissues subjected to oxygen deprivation after injury undergo reperfusion [7][8][9][10]. Reperfusion injury, which is manifested by blood flow-deprived and oxygen-starved organs following blow flow restoration and tissue reoxygenation, is a tissue response to stimulate oxidative metabolism [11][12][13][14][15]. Tissues with reperfusion injury share some of the characteristic features of injury responses to oxygen deprivation and subsequent reoxygenation, such as necrosis, apoptosis, impaired microvascular function, and edema. It has been shown that considerable vascular remodeling takes place during tendon healing to enhance perfusion [16,17]. Stange et al. reported a peak in changes of relative blood volume 7 days after in situ freezing-induced tendinopathy in a rat patellar tendon [18]. A significant uniform hypervascularization occurred during the healing process of chronic Achilles tendinopathy after operative tenolysis in rabbits [17]. The increased vascularization facilitates reoxygenation during tendon repair. Therefore, the process of ischemia/reperfusion after tendon injury, which causes hypoxia-reoxygenation (H/R) changes at the cellular level, presents a major impediment for tendon recovery [19]. However, the sensitivity of tendon-derived cells to hypoxia and the effect of H/R on cell growth and characteristics have been rarely addressed.
In this study, we aimed to establish an in vitro model of cobalt chloride-(CoCl 2 -) induced hypoxia-reoxygenation (H/R) in mouse digital flexor tendon-derived cells and investigate the effects of H/R on these cells [20]. Hypoxic conditions were simulated by treating cultured cells with CoCl 2 , whereas reoxygenation conditions were simulated by replacing the culture media following CoCl 2 treatment. The principle behind CoCl 2 -induced hypoxia is the replacement of Fe 2+ by Co 2+ in heme-based oxygen sensors, which prevents the oxygen sensor from combining with oxygen [21]. Compared with culturing cells in a low-oxygen incubator, the mimic of the hypoxic niche chemically exhibits several advantages, including simple preparation of CoCl 2 , flexible dosage adjustment, treatment suspension at any time, and no impact on oxygen levels for cells that require normal oxygen tension [22]. To have better control over the degree of hypoxia, we decided to employ different CoCl 2 concentrations. Following H/R, cell viability, cell growth, oxidative stress markers, mitochondrial activity, and expression of HIF-1α, vascular endothelial growth factor-A (VEGF-A), collagen I, and collagen III were assessed. Cellular apoptosis and expression of apoptosis-related protein markers Bax and bcl-2 were analyzed.

Materials and Methods
2.1. Experiment Design. We first confirmed the success of the CoCl 2 -induced hypoxia model by examining the expression of HIF-1α protein in tendon-derived cells treated with low-to-high concentrations of CoCl 2 (0.05, 0.1, 0.2, and 0.4 mM), according to previous studies [20,23,24]. Next, the cells were divided into the following groups: control (cells cultured in regular medium), H (cells treated with a concentration gradient of CoCl 2 for 24 h), and R (cells treated with CoCl 2 for 24 h, followed by culturing in regular culture medium for 24 h, 48 h, 72 h, and 96 h (R24h, R48h, R72h, and R96h)). Cell viability was compared across the control, H, and R groups, as well as among different concentrations of CoCl 2 . Then, we determined an appropriate concentration of CoCl 2 for the hypoxia model to perform the following experiments: cell growth, oxidative stress markers, mitochondrial activity, mRNA and protein expression of HIF-1α, VEGF-A, collagen I, and collagen III, cellular apoptosis, and expression of the apoptosis-related proteins Bax and bcl-2.

Isolation of Mouse Flexor Tendon Cells and H/R Model.
The animal procedures were approved by the Administration Committee of Experimental Animals of our university. All animal experiments were carried out in accordance with the Experimental Animal Management Ordinance (National Science and Technology Committee of China) and the Guide for the Care and Use of Laboratory Animals (National Institutes of Health (NIH), Bethesda, MD, USA). Tendon-derived cells were isolated from mouse flexor digitorum profundus (FDP) tendons (C57BL/6, female, 4 weeks old). The mice were sacrificed by cervical dislocation, and the FDP tendons of the index, middle, and ring fingers of the hind paws were harvested. The tendon samples were washed with sterile phosphate-buffered saline (PBS) and then cut into small pieces. The cut samples were digested in a mixture of 3 mg/ml type I collagenase (Gibco, Thermo Fisher Scientific, Grand Island, NY) and 4 mg/ml dispase (Gibco, Thermo Fisher Scientific) for 2 h at 37°C. Undigested tissues and debris were filtered through a 70 μm cell strainer (Merck Millipore, Cork, Ireland). The released cells in the filtrate were centrifuged at 600 g for 10 min and resuspended in low-glucose Dulbecco's modified Eagle's medium (LG-DMEM; Gibco, Thermo Fisher Scientific) supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 100 μg/ml streptomycin. The isolated nucleated cells were plated as passage 0 (P0) at a seeding density of 500 cells/cm 2 , as determined in a previous study [25]. The cells were cultured at 37°C with 5% CO 2 . After 2 days, the cells were washed twice with PBS to remove nonadherent cells. After 10 days, the cells at 95% confluence were collected via local trypsin digestion and split 1 : 3 for subsequent passages. The medium was changed twice a week. Cells at P3 were used for all the experiments.  (20 μl) were mixed with the WST-8/enzyme solution (160 μl) and the reaction starting solution (20 μl) and incubated at 37°C for 30 minutes. The absorbance at 450 nm was measured using a microplate reader (Bio-Rad).   Table S1. Next, the filters were washed 3 times, 8 min each, with TBS-T and then incubated with a conjugated affinity-purified secondary antibody labeled with IRDye 800v for 1 h at room temperature. Afterwards, the membranes were washed again, and the protein bands were detected with an Odyssey imager (LI-COR, Inc., Lincoln, NE). The intensity of each band was quantified with ImageJ software (NIH, Bethesda, MD, USA). The expression levels of proteins were normalized to β-actin.

Cell
2.8. Real-Time PCR. Total RNA of cells was isolated using TRIzol ® Reagent (Ambion, Life Technologies). First-strand cDNA were reverse-transcribed with HiScript II qRT Super-Mix plus gDNA wiper (R223-01, Vazyme, China). QRT-PCR was carried out using 1xAceQ qPCR SYBR Green Master Mix (Q111-02, Vazyme, China) according to the manufacturer's instructions. Specific primers were synthesized for the HIF-1α, VEGF-A, collagen I, and collagen III mRNAs. GAPDH was used as an internal reference, and the geometric mean of its expression was used for normalization. The sequences of qPCR primers are listed in Table S2. Relative quantification of the target genes was performed in triplicate and analyzed using the 2 −ΔΔCt method.
2.9. Immunocytochemistry. The cells were plated at a density of 1 × 10 4 cells/well on coverglasses in 24-well plates for 2 days. The expression of 4 protein markers, including HIF-1α, VEGF-A, collagen I, and collagen III, was measured using immunocytochemistry. Briefly, cells were fixed in 4% paraformaldehyde for 30 min at room temperature. Following washing, the cells were permeabilized with 0.25% Triton X-100 in PBS and blocked with 1% bovine serum albumin (BSA) for 1 h at room temperature. The cells were then incubated with antibodies against HIF-1α, VEGF-A, 3 Oxidative Medicine and Cellular Longevity collagen I, and collagen III overnight at 4°C. After 3 washes with PBS, the cells were incubated with Alexa Fluor488 Goat anti-rabbit IgG (H+L) for 2 h at room temperature. Finally, cells were counterstained with Hoechst. The stained cells were then examined under a fluorescence microscope (Olympus IX71).
2.10. Flow Cytometry Analysis. Annexin V labeling was used in conjunction with flow cytometry to detect phosphatidylserine on the outer membrane of apoptotic cells. First, the binding buffer was diluted to 1: 4 with deionized water (4 ml binding buffer+12 ml deionized water). Then, the cells were harvested using 0.05% trypsin with no EDTA solution    2.11. Statistical Analysis. All experiments were repeated at least 3 times for replication. The data were expressed as mean ± SD. The results were analyzed by one-way ANOVA with post hoc Tukey testing or by the t-test for pairwise comparisons. All data were analyzed using Prism 5.0b (GraphPad Software, La Jolla, CA, USA). P < 0:05 was considered statistically significant.  (Figure 2(b)). At R48h, cell viability significantly increased in cells treated with 0.2-0.4 mM CoCl 2 (P < 0:05 or P < 0:005) (Figure 2(c)). At R72h, a significant increase in cell viability was only observed in cells treated with 0.25 mM of CoCl 2 (P < 0:05) (Figure 2(d)). The cell viability of all CoCl 2 -treated cells was similar to that in the control group at R96h (Figure 2(e)).

CoCl 2 -Induced
At each concentration, cell viability was compared across the control, hypoxia, and reoxygenation groups. Compared with the control group, the cells treated with low concentrations of CoCl 2 showed significant increases in cell viability under hypoxia, with a 14% increase at 0.05 mM and 15% at 0.1 mM (P < 0:05) (Figures 3(a) and 3(b)). The cells treated with higher concentrations of CoCl 2 at 0.15, 0.2, 0.25, and 0.3 mM exhibited significantly enhanced cell viability after reoxygenation for 24 h-72 h (P < 0:05 or P < 0:005) (Figures 3(c)-3(f)). At the 2 highest CoCl 2 concentrations, cell viability significantly increased at R48h compared with those in the control and H groups (P < 0:005) (Figures 3(g) and 3(h)).

H/R Does Not Inhibit Cell Growth.
During the experiment, we observed relatively healthy growth of cells treated with 0.3 mM CoCl 2 after reoxygenation. The cells were shrunk slightly when under hypoxic conditions, but the morphology returned to normal after reoxygenation (Figure 4(a)). Therefore, we selected 0.3 mM CoCl 2 to simulate hypoxia for subsequent experiments. Cell growth curves revealed a slightly increased cell proliferation under H/R compared with that of the control group (Figure 4(b)).

H/R Mediates Oxidative Stress and Mitochondrial
Activity. To determine the impact of H/R on oxidative stress, the intracellular levels of MDA and SOD were measured. The MDA levels significantly increased during reoxygenation compared with those in the control and H groups (P < 0:005) (Figure 5(a)). The SOD levels significantly decreased under H/R compared with that in the control group  Oxidative Medicine and Cellular Longevity (P < 0:005) ( Figure 5(b)). The MitoTracker Red staining showed that the intensity of bioactive mitochondria was higher in the H and R groups than in the control group ( Figure 5(c)).

H/R Enhances Protein Expression of HIF-1α and VEGF-A and Decreases Expression of Collagens I and III.
The protein expression of HIF-1α significantly increased when the cells were under hypoxia and returned to normal levels after reoxygenation (P < 0:05) (Figure 6(a)). By comparison, VEGF-A levels showed a similar but not significant increase under hypoxia and at R24h and returned to normal from R48h and after (Figure 6(b)). Both collagen I and collagen III protein levels significantly decreased at R48h, R72h, and R96h compared with those of the control group and H group (P < 0:05 or P < 0:005) (Figures 6(c) and 6(d)).

H/R Regulates mRNA Expression of HIF-1α, VEGF-A,
Collagen I, and Collagen III. Under H/R conditions, the mRNA expression of HIF-1α was significantly suppressed (Figure 7(a)). The VEGF-A mRNA level significantly increased when the cells were under hypoxia and at R24h compared with that of the control group (P < 0:05) but significantly decreased when the cells were at R72h and R96h compared with that of the H group (P < 0:05) (Figure 7(b)). The mRNA expression of collagen I significantly increased under hypoxia, returned to normal levels at R24h, and significantly decreased from R48h to R96h compared with those of the control and H groups (P < 0:05 or P < 0:005) (Figure 7(c)). The collagen III mRNA levels also significantly increased under hypoxia compared with that of the control group but remained at normal levels with different periods of reoxygenation (P < 0:05) (Figure 7(d)).   7 Oxidative Medicine and Cellular Longevity at R24h, and at R48h (P < 0.05) (Figure 8(CID="C002" value="B")). The percentages of collagen I-and collagen IIIpositive cells were significantly lower at R48h, R72h, and R96h compared with those in the control and H groups (P < 0:05 or P < 0:005) (Figures 8(CID="C004" value="C") and 8(CID="C006" value="D")).

H/R Does Not Affect Cellular Apoptosis but Alters Protein
Expression of Bax and bcl-2. value="Flo"value="w" cytometry was conducted to investigate the apoptotic levels of cells under H/R conditions. The flow cytometric analysis showed that no significant differences in the number of apoptotic cells were found across the control, hypoxia, and reoxygenation groups (Figures 9(a) and 9(b)). Meanwhile, the ratio of Bax/bcl-2 protein expression was significantly upregulated after reoxygenation compared with those of the control and hypoxia groups (P < 0:05), with a peak at R48h (Figure 9(c)).

Discussion
The current study uncovered the cellular responses of tendon-derived cells to hypoxia and subsequent reoxygenation. Our results showed that different concentrations of H/R conditions can cause oxidative stress, which is a major mechanism involved in the pathogenesis of H/R injury [26][27][28][29]. Therefore, we examined the levels of two oxidative stress markers, MDA and SOD, and the activity of bioactive mitochondria in cells under H/R. MDA, a byproduct of lipid oxidation, is a biomarker of oxidative stress of cells [30]. SOD is the main antioxidant enzyme in cells that plays important roles in scavenging oxygen free radicals and resisting the damage of oxygen free radicals [31]. Mitochondria, accounting for the majority of oxygen consumption, can help to tune cellular and organismal hypoxia responses [32]. Our results showed that the MDA level increased, the SOD level decreased, and the activity of bioactive mitochondria was enhanced under H/R, indicating that H/R induced notable oxidative stress in the tendonderived cells. However, the oxidative stress did not affect the cell growth in this study. We speculate that the increased mitochondrial activity may retain the metabolic demand of the cells and induce an adaptive response to oxidative stress [33][34][35][36]. 9 Oxidative Medicine and Cellular Longevity HIF-1α is a key transcription factor in response to hypoxic stress and is widely expressed in mammals, including humans, under hypoxic conditions [37,38]. In the present study, the HIF-1α protein level increased notably in the cells under hypoxia and then returned to the same level in the control cells immediately after reoxygenation. We noted, however, that the changes in HIF-1α mRNA and protein expression were inconsistent. Specifically, HIF-1α mRNA expression decreased significantly during hypoxia and reoxygenation, which is inconsistent with previous studies [39,40]. A possible explanation is that the expression of HIF-1α mRNA changes prior to the expression of protein, and the peak of mRNA expression occurred in less than 24 hours. Then, the inhibited HIF-1α mRNA expression caused the recovery of the HIF-1α protein level in the cells under reoxygenation. These results indicate that tendon-derived cells can regulate the expression of HIF-1α rapidly under H/R conditions. It may be beneficial to the adaptation of tendon-derived cells to oxygen deprivation, leading to the high tolerance of this type of cells to H/R. VEGF-A is an essential growth factor for most tissues in their response to traumatic injuries involving disrupted blood supply. Previous studies demonstrated that the expres-sion of the VEGF gene significantly increased during the early period of tendon healing and that intraoperative delivery of VEGF notably enhanced tendon healing in a chicken model, indicating the importance of VEGF in tendon healing [41,42]. In our study, we found that VEGF-A was upregulated under hypoxia when HIF-1α was activated, and its levels returned to normal during reoxygenation along with the passage of time. Our results were consistent with previous studies investigating the role of the HIF-1α/VEGF pathway in hypoxia. Liang et al. showed that hypoxia markedly upregulated VEGF-A and that appropriate vascular response might be essential for normal repair and remodeling. VEGF-A elevation is a rapid and strong response to hypoxic insult typically seen in most tissues [43]. Therefore, hypoxiapromoted expression of VEGF may be a self-protective mechanism after tendon injury.
Collagen is the major substance secreted by tendon cells. It is the main component of tendon ECM, and thus, synthesis of collagen is essential for maintaining tendon structure [44,45]. A normal tendon comprises 65%-85% of collagen I, which is the most abundant collagen, followed by collagen III [46,47]. Collagen III is highly upregulated during tendon healing [48]. Collagen III primarily forms an abundant but

10
Oxidative Medicine and Cellular Longevity disorganized collagen matrix in the proliferative phase of tendon healing. Webster et al. showed that when tendon-derived cells were cultured at low oxygen tensions, cellular metabolism was depressed and both total protein and collagen production were reduced [49]. It was hypothesized that the hypoxic environment might not have satisfied the physiological oxygen requirements of the cells, and it may have deleterious effects on collagen production [50].
A study also reported that there was an abrupt reduction of collagens I and III in the early stages of Achilles tendinitis; the tendon subsequently developed substantial impairment and Achilles tendinitis eventually occurred [51]. Our results showed that the expression of collagens I and III was downregulated during reoxygenation. Therefore, the in vitro reoxygenation process may correspond to the in vivo repair process after ischemia. Interestingly, the decline in the expression of collagen appears to contradict the increased cell viability. We speculate that although cell growth is not affected, other cellular functions are inhibited under hypoxic stress. Apoptosis plays a critical role in the homeostasis of normal tissue [52]. In human rotator cuff tendinopathy, HIF-1α accumulation was associated with cellular apoptosis, which provided an early support for the role of hypoxia-induced damage to cell loss via apoptosis [53]. However, Sasabe et al. reported that forced expression of HIF-1α suppressed hypoxia-induced apoptosis of human oral squamous cell carcinoma cell lines [54]. Based on flow cytometry analysis in the present study, CoCl 2 -induced H/R had a minimal effect on the cellular apoptosis. However, the protein expression of Bax was upregulated while that of bcl-2 was downregulated, resulting in a significant increase in the Bax/bcl-2 ratio during reoxygenation. The Bax/bcl-2 ratio reached its peak at 48 h after reoxygenation and began to decrease thereafter. This suggested that although the apoptotic pathway was activated, the cells did not actually undergo apoptosis within the observed period. We speculate that tendon-derived cells are highly resistant to H/R-induced apoptosis, and their inherent cell characteristics may prevent or even reverse the occurrence of apoptosis. Another possibility is that the initiation of the apoptotic mechanism precedes the occurrence of apoptosis. To this regard, extending the observation time may allow us to obtain a more complete picture of apoptotic events after H/R.

Conclusions
Collectively, the present study successfully established a CoCl 2 -induced H/R model in tendon-derived cells, which provided a framework for future studies to understand the tendon-specific features of this widely observed stress response mechanism. The data in this study showed that hypoxia followed by reoxygenation for a certain period of time promoted cell viability in a concentration-dependent manner. Tendon-derived cells exhibited considerable tolerance to hypoxia. H/R caused oxidative stress responses but did not affect cell growth. H/R altered the expression of HIF-1α, VEGF-A, collagen I, and collagen III. Cellular apo-ptosis was not affected by H/R, but the Bax/bcl-2 ratio increased during reoxygenation.

Data Availability
All data generated or analyzed during this study are included in this published article.

Conflicts of Interest
All authors have read the journal's policy on disclosure of potential conflicts of interest and declare no conflict of interest.