BIOSCIENCE REPORTS

ACCEPTED MANUSCRIPT

RhoA inhibits the hypoxia-induced apoptosis and mitochondrial dysfunction in

chondrocytes via positively regulating the CREB phosphorylation

Kai Zhang, Dianming Jiang

Chondrocytes which are embedded within the growth-plate or the intervertebral discare
sensitive to environmental stresses, such as inflammation and hypoxia. However, little is
known about the molecular signaling pathways underlining the hypoxia-induced
mitochondrial dysfunction and apoptosis in chondrocytes. In this study, we firstly
examined the hypoxia-induced apoptosis, mitochondrial dysfunction and the activation of
cAMP response element-binding protein (CREB) signaling in human chondrocyte cell
line, C28/I2, and then investigated the regulatory role of RhoA, a well-recognized
apoptosis suppressor, in such process, with gain-of-function strategy. Our results indicated
that hypoxia induced apoptosis and inhibited CREB phosphprylation in chondrocytes,
meanwhile, the dysfunctional mitochondria with upregulated mitochondrial superoxide
and ROS levels, whereas with a reduced mitochondrial membrane potential (MMP) and
Complex IV activity were observed in the hypoxia-treated C28/I2 cells. However, the
overexpressed RhoA blocked the hypoxia-mediated reduction of CREB phosphprylation
and inhibited the apoptosis induction, along with an ameliorated mitochondrial function in
the hypoxia-treated C28/I2 cells. In conclusion, the present study confirmed the reduced
CREB phosphorylation, along with the apoptosis induction and mitochondrial dysfunction
in the hypoxia-treated chondrocyte cells. And the overexpression of RhoA ameliorated the
hypoxia-induced mitochondrial dysfunction and apoptosis via blocking the hypoxia-
mediated reduction of CREB phosphorylation.

Cite as Bioscience Reports (2017) DOI: 10.1042/BSR20160622

Copyright 2017 The Author(s).

This is an Accepted Manuscript; not the final Version of Record. You are encouraged to use the final Version of Record that, when
published, will replace this manuscript and be freely available under a Creative Commons licence.
All other rights reserved.
 1 RhoA inhibits the hypoxia-induced apoptosis and mitochondrial dysfunction in
2 chondrocytes via positively regulating the CREB phosphorylation
3
4 Running title:
5 RhoA regulates MD and CREB phosphorylation in chondrocytes
6
7 Kai Zhang, Dianming Jiang *
8
9 Department of Orthopaedics, First Affiliated Hospital of Chongqing Medical
10 University, 400016, Chongqing, China
11
12 * Dianming Jiang, dianm_jiang@126.com
13
14 No. 1, Medical Collage Road, Yuzhong District, 400016, Chongqing, China,
15 Tel: 0086-23-89011212; Fax: 0086-23-89011217
16

1
17 Synopsis
18 Chondrocytes which are embedded within the growth-plate or the intervertebral
19 discare sensitive to environmental stresses, such as inflammation and hypoxia.
20 However, little is known about the molecular signaling pathways underlining the
21 hypoxia-induced mitochondrial dysfunction and apoptosis in chondrocytes. In this
22 study, we firstly examined the hypoxia-induced apoptosis, mitochondrial dysfunction
23 and the activation of cAMP response element-binding protein (CREB) signaling in
24 human chondrocyte cell line, C28/I2, and then investigated the regulatory role of
25 RhoA, a well-recognized apoptosis suppressor, in such process, with gain-of-function
26 strategy. Our results indicated that hypoxia induced apoptosis and inhibited CREB
27 phosphprylation in chondrocytes, meanwhile, the dysfunctional mitochondria with
28 upregulated mitochondrial superoxide and ROS levels, whereas with a reduced
29 mitochondrial membrane potential (MMP) and Complex IV activity were observed in
30 the hypoxia-treated C28/I2 cells. However, the overexpressed RhoA blocked the
31 hypoxia-mediated reduction of CREB phosphprylation and inhibited the apoptosis
32 induction, along with an ameliorated mitochondrial function in the hypoxia-treated
33 C28/I2 cells. In conclusion, the present study confirmed the reduced CREB
34 phosphorylation, along with the apoptosis induction and mitochondrial dysfunction in
35 the hypoxia-treated chondrocyte cells. And the overexpression of RhoA ameliorated
36 the hypoxia-induced mitochondrial dysfunction and apoptosis via blocking the
37 hypoxia-mediated reduction of CREB phosphorylation.
38

2
39

40 Introduction
41 Chondrocytes which are embedded within the growth-plate or the intervertebral
42 disc survive in an almost avascular and hypoxic milieu, and are sensitive to
43 environmental and multiple cellular stresses, such as inflammation, endoplasmic
44 reticulum(ER) stress and hypoxia [1, 2]. Hypertrophic chondrocytes die through the
45 induction of programmed cell death (apoptosis) in the background of cervical
46 spondylosis or the joint disorder containing osteoarthritis (OA) characterized by
47 progressive breakdown of articular cartilage [3, 4]. Reactive oxygen species (ROS)
48 generated from the hypoxia/ischemia trigger apoptotic cell death [5].
49 Mitochondria are the key targets of ROS, which have been suggested to regulate
50 the processes involved in the mitochondrial dysfunction and in the promotion of
51 apoptosis. The elevated ROS production causes mitochondrial damage, such as
52 collapse of the mitochondrial membrane potential, complex IV inactivation, which
53 opens up the mitochondrial permeability transition pores, leading to the release of
54 cytochrome c into the cytoplasm, to the cleavage of proapoptotic proteins [6, 7]. Then
55 the released cytochrome c initiates the mitochondria-dependent caspase-3 pathway
56 which plays a pivotal role in the apoptotic cell death [6, 7]. ROS also have indicated
57 to cause structural and functional damage to mitochondria [10]. However, little is
58 known about the molecular signaling pathways underlining the
59 mitochondria-dependent caspase pathway.
60 Cyclic adenosine monophosphate (cAMP) response element-binding protein
61 (CREB) is necessary for the cell proliferation and apoptosis [11], via regulating the
62 expression of a repertoire of genes associated with cell survival, such as B cell
63 lymphoma 2 (Bcl-2), B-cell lymphoma-extra large (Bcl-xL) and c-fos [11],
64 particularly in the condition of hypoxia [14]. And the inhibition of CREB-Ser133
65 phosphorylation results in the suppression of anti-apoptotic genes [15, 16]. Therefore,
66 the CREB signaling might be implicated in the hypoxia-induced apoptosis in
67 chondrocytes.
68 In this study, we firstly examined the hypoxia-induced apoptosis and the activation
69 of CREB signaling in human chondrocyte cell line, C28/I2, and then investigated the
70 regulatory role of RhoA, a well-recognized apoptosis suppressor [17] in the
71 hypoxia-promoted apoptosis in C28/I2 cells with gain-of-function strategy. Our
72 results implied the involvement of CREB signaling and the protective role of RhoA in
73 the hypoxia-induced apoptosis in chondrocytes.
74

75 Materials and methods

76 Cell culture and treatments
77 Human chondrocyte cell line, C28/I2, was purchased from American Type Culture
78 Collection (ATCC) and were cultivated in Dulbeccos modified Eagles medium
79 (DMEM) (Gibco) containing 10% FBS (Invitrogen), 100 units/ml penicillin and 100
3
 80 g/ml streptomycin (Sigma-Aldrich) under 5% CO2 at 37 C. The cells were
81 sub-cultured after reaching about 90% confluence. For hypoxia treatment, cells were
82 placed in a hypoxia incubator with 5% CO2 and 3% oxygen. Oxygen concentration
83 was monitored continuously (Forma 3130, Thermo Scientific). To overexpress RhoA
84 in C28/I2 cells, the homo sapiens RhoA coding sequence was cloned into the
85 pcDNA3.1(+) vector (Invitrogen). And the RhoA-pcDNA3.1(+) or control
86 pcDNA3.1(+) plasmid was transfected into C28/I2 cells with Lipofectamine 2000
87 (Invitrogen). Lysophosphatidic acid (LPA) (Santa Cruz Biotechnology) was utilized a
88 RhoA activator.
89 MTT assay for cellular viability
90 Cellular viability of C28/I2 cells was examined by the MTT assay with 3-(4,
91 5-dimethylthiazol-2-yl)-2, 5-diphenyl tetrazolium bromide (MTT) assay kit
92 (Invitrogen). In brief, C28/I2 cells, post treatment, were added with MTT reagent was
93 for incubation at 37 C for 4 hours, and then were updated with dimethyl sulfoxide
94 (DMSO) to dissolve formazan crystals. Optical densities (OD) were measured at 450
95 nm by spectrophotometer (Bio-Rad). Cellular viability was presented as average
96 OD450 value.
97 Apoptosis assay and Caspase 3 assay
98 Apoptosis induction in C28/I2 cells was assayed with the annexin V/FITC
99 apoptosis detection kit (Abcam). In brief, 1 106 C28/I2cells post treatment were
100 stained with annexin V-FITC and propidium iodide and detected byaFACScan flow
101 cytometer (BD Biosciences). The apoptosis was evaluated by a percentage of annexin
102 V-positive cells to total cells.The caspase 3 activity in C28/I2 cells was examined
103 with an Caspase SenSolyte kit (for caspase 3) (AnaSpec). Cells were collected and
104 were washed for three times with ice-cold PBS, then were resuspended in 100
105 L(final volume) of a caspase buffer solution supplemented with the fluorogenic
106 peptide substrate Ac-DEVD-AMC for incubation for 30 min at 37 C. And the
107 caspase3 activity was determined by assessment of Asp-Glu-Val-Asp (DEVD)-AMC
108 cleavagein a fluorescence spectrophotometer (Hitachi) with an excitation wavelength
109 of 390 nm and an emission wavelength of 460 nm. And the activity was expressed as
110 percent value of fluorescence intensity of AMC to control.
111 Western blotting assay
112 The mitochondrial and cytosolic proteins in C28/I2 cells were isolated with the
113 Mitochondria/Cytosol Fractionation Kit (Abcam) and were supplemented with a
114 protease inhibitor cocktail (Roche). Then, protein samples were separated with 12%
115 SDS-PAGE gels and were electro-transferred onto nitrocellulose membranes
116 (Millipore). The membranes were incubated by rabbit antibody against cytochrome c
117 (Cyt c), caspase 3 (CASP 3), CREB, CREB with Ser133 phosphorylation, RhoA (all
118 Santa Cruz Biotechnology) or -actin (Sigma-Aldrich). Then, the specific binding
119 signal was acquired using the ECL detection systems with the incubation with
120 horseradish-peroxidase-conjugated secondary antibodies (Pierce).
4
121 Measurement of Mitochondrial Superoxide and Reactive Oxygen Species
122 The mitochondrial superoxide was examined with MitoSOXTM Red mitochondrial
123 superoxide indicator (Invitrogen) (Ex/Em=580/510 nm) which highly selective
124 detects the superoxide in the mitochondria. C28/I2 cells were incubated with 5M
125 MitoSOXTM Red at 37 C for 30 minutes and then were detected after the removal of
126 the reagent and three-time washing. Intracellular ROS level of was quantified by the
127 ROS-sensitive fluorophore 5-(and-6)-chloromethyl-2, 7-dichlorodi-hydrofluorescein
128 diacetate (Invitrogen) according to the products manual. Briefly, confluent C28/I2
129 cells were incubated with the 5-chloromethyl-2, 7-dichlorodihydrofluorescein
130 diacetate probe at 37 C for 30 minutes and washed with PBS for three times. Then
131 the cells were rinsed with HBSS and were measured using the fluorescence
132 spectrophotometer (Hitachi) with an excitation wavelength of 488 nm and an
133 emission wavelength of 530 nm. Results were presented as a percent level to control.
134 Measurement of mitochondrial membrane potential and complex IV activity
135 TMRE-Mitochondrial Membrane Potential Assay Kit (Abcam) was used to
136 determine the mitochondrial membrane potential (MMP) in C28/I2 cells. 3 x 105
137 C28/I2 cells were incubated with the MMP-sensitive fluorescent TMRE for 30
138 minutes at 37 C (1000 nM FCCP was added to the positive control cells 10 minutes
139 prior of TMRE). Cells were then trypsinized, centrifuged and resuspended in 0.4 mL
140 of DPBS with 0.2% BSA and were analyzed for TMRE fluorescence by the
141 fluorescence spectrophotometer (Hitachi) with an excitation wavelength of 549 nm
142 and an emission wavelength of 575 nm. Complex IV activity was measured by
143 Complex IV Rodent Enzyme Activity Microplate Assay Kit (Abcam) according to
144 manufacturer's instructions. In brief, C28/I2 cells were collected and lyzed with the
145 detergent extraction. Then protein samples were serially diluted and were incubated at
146 room temperature for 3 hours in the plate in each well of which the Enzyme-linked
147 monoclonal antibody against complex IV has immobilized. Then the binding was
148 assayed by the fluorescence spectrophotometer (Hitachi) with an excitation
149 wavelength of 549 nm and an emission wavelength of 575 nm. Results were presented
150 as a percent level to control.
151 Statistical analysis
152 Statistical analyses were performed using SPSS18.0 software (IBM SPSS). All data
153 were expressed as mean standard error of the mean (SEM). Student's t test was
154 performed for the difference between two groups. A p value <0.05 was considered
155 statistically significant.
156

157 Results
158 Hypoxia induces apoptosis and inhibits CREB phosphorylation in chondrocytes
159 In order to investigate the apoptosis induction by hypoxia in chondrocytes, we
160 examined the cellular viability with MTT assay, the apoptosis induction with flow
5
161 cytometric analysis and the caspase 3 activity with Caspase SenSolyte kit (for caspase
162 3) in the chondrocyte C28/I2 cells under hypoxia or normoxia. MTT assay results
163 (Figure 1A) demonstrated that the hypoxia treatment for 24 or 48 hours significantly
164 reduced the viability of C28/I2 cells (p<0.05 or p<0.01). The apoptosis level was also
165 significantly higher in the hypoxia-treated C28/I2 cells than in the C28/I2 cells under
166 normoxia (Figure 1B, p<0.01 or p<0.001). The activity of active caspase 3, revealing
167 by the fluorescence intensity AMC, in the hypoxia-treated C28/I2 cells was also
168 significantly higher in the hypoxia group (Figure 1C) at 12, 24 or 48 hour post
169 treatment (H.P.T.) (p<0.01 or p<0.001).
170 Western blotting assay was also performed to examine the apoptosis-associated
171 markers, such as released cytochrome c (Cyt c) and cleaved caspase 3 (Clv-CASP 3)
172 in the hypoxia- or normoxia-treated C28/I2 cells. Figure 1D indicated that there were
173 significantly high levels of Cyt c release and Clv-CASP 3 (Figure 1E) in the
174 hypoxia-treated C28/I2 cells at 12, 24 or 48 hour post treatment. CREB has been well
175 recognized to function as a pro-survival signal [18], and our study has also
176 investigated the activation of CREB signaling in the hypoxia-treated C28/I2 cells
177 (Figure 1F). The western blotting assay also indicated that the phosphorylated CREB
178 was also markedly downregulated in the hypoxia-treated C28/I2 cells than in the
179 normoxia-treated C28/I2 cells (Figure 1F), whereas the CREB without
180 phosphorylation was not markedly regulated by hypoxia in C28/I2 cells. Taken
181 together, we confirmed the apoptosis induction by hypoxia in chondrocyte C28/I2
182 cells with the downregulation of CREB phosphorylation.
183 Hypoxia induces mitochondrial dysfunction in chondrocyte C28/I2 cells
184 We then investigated whether hypoxia induced chondrocyte mitochondrial
185 dysfunction. Firstly, we examined the superoxide generation, using MitoSOX, a
186 live-cell-permeable and mitochondrial localizing superoxide indicator, in C28/I2 cells
187 under hypoxia or normoxia. As shown in Figure 2A, there was a significantly high
188 level of superoxide in the hypoxia-treated cells at both 24 and 48 hour post treatment
189 (p <0.01 or p <0.001). Secondly, we measured ROS production in the hypoxia-treated
190 C28/I2 cells. Figure 2B demonstrated that hypoxia enhanced the ROS production
191 from 12 to 48 hours post treatment (p <0.05, p <0.01 or p <0.001). On the other side,
192 MMP, as a well-confirmed indicator of mitochondrial function was significantly
193 reduced in the hypoxia-treated C28/I2 cells at 24 or 48 hour post treatment (p <0.05
194 for 24 H.P.T. or p <0.01 for 48 H.P.T., Figure 2C). And the activity of mitochondrial
195 respiratory chain complex IV was markedly downregulated in the hypoxia-treated
196 C28/I2 cells (Figure 2D), at 12, 24 or 48 hour post treatment (p <0.05, p <0.01 or p
197 <0.001). Therefore, hypoxia promotes mitochondrial dysfunction in chondrocyte
198 C28/I2 cells.
199 RhoA overexpression reduces the hypoxia-induced CREB phosphorylation and
200 apoptosis in C28/I2 cells
201 RhoA is a well-recognized apoptosis suppressor [18]. To further investigate the
202 molecular pathway underlining the hypoxia-induced, CREB-associated apoptosis in
6
203 chondrocytes, we then overexpress RhoA in the hypoxia-treated C28/I2 cells, and
204 then re-evaluated the apoptosis induction and CREB phosphorylation. Figure 3A
205 indicated that the RhoA was markedly promoted by the transfection with
206 RhoA-pcDNA3.1(+) plasmid at both 24 and 48 hour post transfection (p <0.001
207 respectively). Moreover, the phosphorylated CREB level was also significantly
208 promoted by the RhoA-pcDNA3.1(+) transfection (Figure 3B, p < 0.01 respectively
209 for 24 or 48 H.P.T.). Then we investigated the regulation of the overexpressed RhoA
210 on the hypoxia-induced apoptosis in C28/I2 cells. As shown in Figure 3C, the
211 hypoxia-mediated viability reduction was markedly ameliorated by the
212 RhoA-pcDNA3.1(+) transfection (p<0.05 for 24 H.P.T. or p<0.01 for 48 H.P.T.); and
213 the RhoA overexpression also reduced the hypoxia-mediated apoptosis of C28/I2 cells
214 (either p<0.05 for 24 or 48 hours). In addition, caspase 3 activity, AMC fluorescence
215 intensity in the hypoxia-treated C28/I2 cells was significantly inhibited in the RhoA
216 overexpressrion group, than in the control group (p<0.05 for 24 or p<0.01 for 48
217 hours post transfection). Therefore, RhoA overexpression attenuated the
218 hypoxia-induced apoptosis in chondrocyte C28/I2 cells.
219 RhoA overexpression reduces the hypoxia-induced mitochondrial dysfunction in
220 chondrocytes
221 We then investigated the regulation of RhoA overexpression on the
222 hypoxia-induced mitochondrial dysfunction. Firstly, we examined the mitochondrial
223 superoxide level and the ROS generation in the hypoxia-treated C28/I2 cells which
224 were transfected with RhoA-pcDNA3.1(+) or control pcDNA3.1(+) plasmid. Figure
225 4A indicated that the promoted mitochondrial superoxide production was significantly
226 reduced in the RhoA-pcDNA3.1(+)-transfected C28/I2 cells under hypoxia than in the
227 control pcDNA3.1(+) plasmid-transfected C28/I2 cells under hypoxia (Figure 4A,
228 p<0.05 for 24 or p<0.01 for 48 hours post transfection). And the ROS release was also
229 blocked by the RhoA overexpression (Figure 4B, p<0.01 for 24 or p<0.001 for 48
230 hours post transfection). Moreover, the hypoxia-mediated inhibition of MMP and
231 mitochondrial respiratory chain complex IV was markedly ameliorated by the RhoA
232 overexpression (Figure 4C and 4D, p<0.01or p<0.001). Taken together, our results
233 showed a mitochondrial protection of RhoA overexpression in the hypoxia-treated
234 C28/I2 cells.
235 RhoA activator lysophosphatidic acid (LPA) reduces the hypoxia-induced apoptosis
236 in C28/I2 cells via ameliorating mitochondrial dysfunction
237 To reconfirm the inhibitory role of RhoA on the hypoxia-induced mitochondrial
238 dysfunction and apoptosis in C28/I2 cells, we then pre-treated C28/I2 cells with 10
239 nM LPA, and then re-evaluated the hypoxia-induced apoptosis and mitochondrial
240 dysfunction. It was shown in Figure 5A that the LPA treatment did not regulate the
241 RhoA expression. However, the RhoA-induced CREB phosphorylation was
242 significantly promoted by the LPA treatment (p < 0.01 respectively at either 24 or 48
243 hour post treatment, Figure 5B). Moreover, the hypoxia-mediated apoptosis was also
244 markedly reduced in the LPA-treated C28/I2 cells at 48 hour post treatment (10 nM)
7
245 (p<0.05, Figure 5C). AMC fluorescence intensity in the hypoxia-treated C28/I2 cells
246 was significantly inhibited by the 10 nM LPA at 24 or 48 hour post treatment (p<0.05
247 respectively, Figure 5D). Therefore, RhoA activation by LPA also attenuated the
248 hypoxia-induced apoptosis in chondrocyte C28/I2 cells.
249 In addition, we also examined the regulation by LPA on the hypoxia-induced
250 mitochondrial dysfunction. Results demonstrated that both the hypoxia-induced
251 superoxide production (Figure 6A) and the ROS release (Figure 6B) were markedly
252 reduced in the LPA-treated (10 nM) C28/I2 cells than in the control C28/I2 cells
253 (p<0.05 or p<0.01). On the other side, the hypoxia-mediated inhibition of both MMP
254 and mitochondrial respiratory chain complex IV was significantly ameliorated by the
255 LPA treatment (p<0.01 or p<0.001, Figure 6C and 6D). Thus, our results reconfirmed
256 the involvement of mitochondrial dysfunction in the RhoA-mediated inhibition of
257 hypoxia-induced apoptosis in C28/I2 cells.
258

259 Discussion
260 Oxygen affects the activity of multiple types of cells and is involved in many
261 processes that are important for degenerative bone disease [21-23]. The current
262 studies found that hypoxia promoted apoptosis in the chondrocyte C28/I2 cells, via
263 upregulating caspase 3 activity and via upregulating inducing mitochondrial
264 dysfunction. Apoptosis can be initiated through membrane receptor-associated and
265 mitochondrial-initiated pathways that converge and mediate their downstream
266 apoptosis effects [24]. Therefore, the mitochondria dependence was confirmed in the
267 hypoxia-induced apoptosis in C28/I2 cells. Mitochondrial dysfunction such as the
268 overeproduction of mitochondrial superoxide and ROS has been identified to be one
269 of importance mechanisms in the hypoxia-mediated damage to chondrocytes [23, 25,
270 26]. On the other side, hypoxia inhibits complex I, III or IV [27] and MMP [28] in
271 chondrocytes. Our study confirmed the high promotion of mitochondrial superoxide
272 and ROS and the inhibition of MMP and complex IV in the hypoxia-treated C28/I2
273 cells.
274 Oxidative stress has been implicated in the pathogenesis of osteoporosis [29]. It is
275 conceivable that mitochondrial-mediated ROS generation under oxidative stress leads
276 to damage to ROS-produced cells or peripheral chondrocytes [30]. The present study
277 demonstrated that the hypoxia also significantly exerted mitochondrial dysfunction in
278 chondrocyteic C28/I2 cells. Hypoxia has been implicated in the pathogenesis of
279 osteoporosis [31, 32], at least partly via inducing the mitochondrial dysfunction [30,
280 33]. Mitochondrial-mediated ROS generation under oxidative stress leads to damage
281 to ROS-produced cells or peripheral chondrocytes [30]. The Allicin treatment
282 ameliorated the H2O2-mediated repression of CREB signaling [30]. Previously, PI3K
283 and CREB were shown to be implicated in chondrocyte-like cell proliferation and
284 differentiation [34, 35]. In the present study, RhoA overexpression in C28/I2 cells also
285 promoted the CREB signaling, and thus ameliorated the hypoxia-induced
286 mitochondrial dysfunction and apoptosis.
287 Multiple bone anabolic agents have been confirmed to prevent chondrocyte
8
288 apoptosis, such as parathyroid hormone (PTH) [36] and insulin-like growth factor-I
289 [37]. RhoA has been well recognized to promote cell survival and survival-related
290 signaling [38, 39], such as increased Bcl-2 expression [17] via binding and
291 hydrolyzing GTP. The signaling pathway through which RhoA prevents apoptosis
292 varies in different tissues. And our results confirmed that the hypoxia-induced
293 apoptosis was inhibited by the RhoA overexpression, implying the RhoA signaling is
294 implicated in the anti-apoptosis response in chondrocytes. RhoA is one number of
295 Rho small GTPase families [40]. Activated Rho GTPase directs a variety of cellular
296 responses [41, 42]. RhoA has been suggested to modulate the cell cycle and survival
297 [43, 44] of cancer cells. Moreover, RhoA has implicated several anti-apoptotic
298 pathways in the suppression of apoptosis [45] via activating ERK [45], via
299 up-regulating anti-apoptotic Bcl-2 expression [46, 47] or via regulating p53
300 pro-apoptotic pathway [48]. More recently, RhoA has been identified to regulate the
301 CREB activity [49]. And in this study, we found that RhoA also exerted a protective
302 role against the hypoxia-induced apoptosis in chondrocytes via regulating the CREB
303 phosphorylation.
304 In conclusion, the present study confirmed the reduced CREB phosphorylation,
305 along with the apoptosis induction and mitochondrial dysfunction in the
306 hypoxia-treated chondrocyte cells. And the overexpression of RhoA ameliorated the
307 hypoxia-induced mitochondrial dysfunction and apoptosis via blocking the reduction
308 of CREB phosphorylation.
309 Conflict of interest
310 Authors declare that there is no conflict of interests regarding the publication of this
311 article.
312 Acknowledgments
313 No supported grant.
314
315 References
316 1. Horton, W. J., Bennion, P. andYang, L. (2006). Cellular, molecular, and matrix
317 changes in cartilage during aging and osteoarthritis. J Musculoskelet Neuronal
318 Interact. 6, 379-381
319 2. Chang, Z., Huo, L., Wu, Y. andZhang, P. (2014). HIF-1 alpha had pivotal effects
320 on downregulation of miR-210 decreasing viability and inducing apoptosis in
321 hypoxic chondrocytes. ScientificWorldJournal. 2014, 876363
322 3. Sun, Y. Q., Zheng, S., Yu, J., Yan, K. andTian, W. (2013). Effect of total disc
323 replacement on atypical symptoms associated with cervical spondylosis. Eur.
324 Spine J. 22, 1553-1557
325 4. Triantafillou, K. M., Lauerman, W. andKalantar, S. B. (2012). Degenerative
326 disease of the cervical spine and its relationship to athletes. Clin Sports Med. 31,
327 509-520
328 5. Misra, M. K., Sarwat, M., Bhakuni, P., Tuteja, R. andTuteja, N. (2009). Oxidative
329 stress and ischemic myocardial syndromes. Med Sci Monit. 15, A209-A219
9
330 6. Balaban, R. S., Nemoto, S. andFinkel, T. (2005). Mitochondria, oxidants, and
331 aging. Cell. 120, 483-495
332 7. Choi, E. M. (2012). Magnolol protects osteoblastic MC3T3-E1 cells against
333 antimycin A-induced cytotoxicity through activation of mitochondrial function.
334 Inflammation. 35, 1204-1212
335 8. Reed, J. C. (1997). Cytochrome c: can't live with it--can't live without it. Cell. 91,
336 559-562
337 9. Kroemer, G., Zamzami, N. andSusin, S. A. (1997). Mitochondrial control of
338 apoptosis. Immunol Today. 18, 44-51
339 10. Chomyn, A. andAttardi, G. (2003). MtDNA mutations in aging and apoptosis.
340 Biochem Biophys Res Commun. 304, 519-529
341 11. Wang, P., Huang, S., Wang, F., Ren, Y., Hehir, M., Wang, X. and Cai, J. (2013).
342 Cyclic AMP-response element regulated cell cycle arrests in cancer cells. PLoS
343 One. 8, e65661
344 12. Ferron, L., Ruchon, Y., Renaud, J. F. andCapuano, V. (2011). T-type Ca(2)+
345 signalling regulates aldosterone-induced CREB activation and cell death through
346 PP2A activation in neonatal cardiomyocytes. Cardiovasc. Res. 90, 105-112
347 13. Han, X., Zhao, X., Lu, M., Liu, F., Guo, F., Zhang, J. and Huang, X. (2013).
348 Electroacupuncture Ameliorates Learning and Memory via Activation of the
349 CREB Signaling Pathway in the Hippocampus to Attenuate Apoptosis after
350 Cerebral Hypoperfusion. Evid Based Complement Alternat Med. 2013, 156489
351 14. Meng, F. J., Jiao, S. M. andYu, B. (2012). Picroside II protects cardiomyocytes
352 from hypoxia/reoxygenation-induced apoptosis by activating the PI3K/Akt and
353 CREB pathways. Int. J. Mol. Med. 30, 263-270
354 15. Aggarwal, S., Kim, S. W., Ryu, S. H., Chung, W. C. andKoo, J. S. (2008). Growth
355 suppression of lung cancer cells by targeting cyclic AMP response
356 element-binding protein. Cancer Res. 68, 981-988
357 16. Naqvi, S., Martin, K. J. andArthur, J. S. (2014). CREB phosphorylation at Ser133
358 regulates transcription via distinct mechanisms downstream of cAMP and MAPK
359 signalling. Biochem. J. 458, 469-479
360 17. Gomez, J., Martinez, C., Giry, M., Garcia, A. andRebollo, A. (1997). Rho
361 prevents apoptosis through Bcl-2 expression: implications for interleukin-2
362 receptor signal transduction. Eur. J. Immunol. 27, 2793-2799
363 18. Dworkin, S., Malaterre, J., Hollande, F., Darcy, P. K., Ramsay, R. G.
364 andMantamadiotis, T. (2009). cAMP response element binding protein is required
365 for mouse neural progenitor cell survival and expansion. Stem Cells. 27,
366 1347-1357
367 19. Zhang, J., Liu, X. B., Cheng, C., Xu, D. L., Lu, Q. H. andJi, X. P. (2014).
368 Rho-kinase inhibition is involved in the activation of PI3-kinase/Akt during
369 ischemic-preconditioning-induced cardiomyocyte apoptosis. Int J Clin Exp Med.
370 7, 4107-4114
371 20. Kumar, A., Al-Sammarraie, N., DiPette, D. J. andSingh, U. S. (2014). Metformin
372 impairs Rho GTPase signaling to induce apoptosis in neuroblastoma cells and
373 inhibits growth of tumors in the xenograft mouse model of neuroblastoma.
10
374 Oncotarget. 5, 11709-11722
375 21. Vanharanta, S. (2015). A hypoxic ticket to the bone metastatic niche. Breast
376 Cancer Res. 17, 122
377 22. Dalle, C. L., Matte', A., Valenti, M. T., Siciliano, A., Mori, A., Schweiger, V.,
378 Zampieri, G., Perbellini, L. and De Franceschi, L. (2015). Hypoxia-reperfusion
379 affects osteogenic lineage and promotes sickle cell bone disease. Blood.
380 23. White, R. andGibson, J. S. (2010). The effect of oxygen tension on calcium
381 homeostasis in bovine articular chondrocytes. J Orthop Surg Res. 5, 27
382 24. Hengartner, M. O. (2000). The biochemistry of apoptosis. Nature. 407, 770-776
383 25. Milner, P. I., Wilkins, R. J. andGibson, J. S. (2007). The role of mitochondrial
384 reactive oxygen species in pH regulation in articular chondrocytes. Osteoarthritis
385 Cartilage. 15, 735-742
386 26. Lane, R. S., Fu, Y., Matsuzaki, S., Kinter, M., Humphries, K. M. andGriffin, T. M.
387 (2015). Mitochondrial respiration and redox coupling in articular chondrocytes.
388 Arthritis Res Ther. 17, 54
389 27. White, R. andGibson, J. S. (2010). The effect of oxygen tension on calcium
390 homeostasis in bovine articular chondrocytes. J Orthop Surg Res. 5, 27
391 28. Clerigues, V., Murphy, C. L., Guillen, M. I. andAlcaraz, M. J. (2013). Haem
392 oxygenase-1 induction reverses the actions of interleukin-1beta on
393 hypoxia-inducible transcription factors and human chondrocyte metabolism in
394 hypoxia. Clin Sci (Lond). 125, 99-108
395 29. Wu, Q., Zhong, Z. M., Pan, Y., Zeng, J. H., Zheng, S., Zhu, S. Y. and Chen, J. T.
396 (2015). Advanced Oxidation Protein Products as a Novel Marker of Oxidative
397 Stress in Postmenopausal Osteoporosis. Med Sci Monit. 21, 2428-2432
398 30. Ding, G., Zhao, J. andJiang, D. (2016). Allicin inhibits oxidative stress-induced
399 mitochondrial dysfunction and apoptosis by promoting PI3K/AKT and
400 CREB/ERK signaling in osteoblast cells. Exp Ther Med. 11, 2553-2560
401 31. Ma, H. P., Ma, X. N., Ge, B. F., Zhen, P., Zhou, J., Gao, Y. H., Xian, C. J. and
402 Chen, K. M. (2014). Icariin attenuates hypoxia-induced oxidative stress and
403 apoptosis in osteoblasts and preserves their osteogenic differentiation potential
404 in vitro. Cell Prolif. 47, 527-539
405 32. Cao, B., Chai, C. andZhao, S. (2015). Protective effect of Edaravone against
406 hypoxia-induced cytotoxicity in osteoblasts MC3T3-E1 cells. Iubmb Life. 67,
407 928-933
408 33. Lane, R. K., Hilsabeck, T. andRea, S. L. (2015). The role of mitochondrial
409 dysfunction in age-related diseases. Biochim Biophys Acta. 1847, 1387-1400
410 34. Carpio, L., Gladu, J., Goltzman, D. andRabbani, S. A. (2001). Induction of
411 osteoblast differentiation indexes by PTHrP in MG-63 cells involves multiple
412 signaling pathways. Am J Physiol Endocrinol Metab. 281, E489-E499
413 35. Matsuo, N., Tanaka, S., Gordon, M. K., Koch, M., Yoshioka, H. andRamirez, F.
414 (2006). CREB-AP1 protein complexes regulate transcription of the collagen
415 XXIV gene (Col24a1) in osteoblasts. J. Biol. Chem. 281, 5445-5452
416 36. Jilka, R. L. (2007). Molecular and cellular mechanisms of the anabolic effect of
417 intermittent PTH. Bone. 40, 1434-1446
11
418 37. Grey, A., Chen, Q., Xu, X., Callon, K. andCornish, J. (2003). Parallel
419 phosphatidylinositol-3 kinase and p42/44 mitogen-activated protein kinase
420 signaling pathways subserve the mitogenic and antiapoptotic actions of
421 insulin-like growth factor I in osteoblastic cells. Endocrinology. 144, 4886-4893
422 38. Gallagher, E. D., Gutowski, S., Sternweis, P. C. andCobb, M. H. (2004). RhoA
423 binds to the amino terminus of MEKK1 and regulates its kinase activity. J. Biol.
424 Chem. 279, 1872-1877
425 39. Kobayashi, K., Takahashi, M., Matsushita, N., Miyazaki, J., Koike, M., Yaginuma,
426 H., Osumi, N., Kaibuchi, K. and Kobayashi, K. (2004). Survival of developing
427 motor neurons mediated by Rho GTPase signaling pathway through Rho-kinase. J.
428 Neurosci. 24, 3480-3488
429 40. Vega, F. M. andRidley, A. J. (2008). Rho GTPases in cancer cell biology. Febs
430 Lett. 582, 2093-2101
431 41. Jaffe, A. B. andHall, A. (2005). Rho GTPases: biochemistry and biology. Annu
432 Rev Cell Dev Biol. 21, 247-269
433 42. Bishop, A. L. andHall, A. (2000). Rho GTPases and their effector proteins.
434 Biochem. J. 348 Pt 2, 241-255
435 43. Wong, W. W., Dimitroulakos, J., Minden, M. D. andPenn, L. Z. (2002).
436 HMG-CoA reductase inhibitors and the malignant cell: the statin family of drugs
437 as triggers of tumor-specific apoptosis. Leukemia. 16, 508-519
438 44. Graaf, M. R., Richel, D. J., van Noorden, C. J. andGuchelaar, H. J. (2004). Effects
439 of statins and farnesyltransferase inhibitors on the development and progression
440 of cancer. Cancer Treat. Rev. 30, 609-641
441 45. Bijian, K., Takano, T., Papillon, J., Le Berre, L., Michaud, J. L., Kennedy, C. R.
442 and Cybulsky, A. V. (2005). Actin cytoskeleton regulates extracellular
443 matrix-dependent survival signals in glomerular epithelial cells. Am J Physiol
444 Renal Physiol. 289, F1313-F1323
445 46. Gomez, J., Martinez, C., Giry, M., Garcia, A. andRebollo, A. (1997). Rho
446 prevents apoptosis through Bcl-2 expression: implications for interleukin-2
447 receptor signal transduction. Eur. J. Immunol. 27, 2793-2799
448 47. Fromigue, O., Hay, E., Modrowski, D., Bouvet, S., Jacquel, A., Auberger, P. and
449 Marie, P. J. (2006). RhoA GTPase inactivation by statins induces osteosarcoma
450 cell apoptosis by inhibiting p42/p44-MAPKs-Bcl-2 signaling independently of
451 BMP-2 and cell differentiation. Cell Death Differ. 13, 1845-1856
452 48. Li, X., Liu, L., Tupper, J. C., Bannerman, D. D., Winn, R. K., Sebti, S. M.,
453 Hamilton, A. D. and Harlan, J. M. (2002). Inhibition of protein
454 geranylgeranylation and RhoA/RhoA kinase pathway induces apoptosis in human
455 endothelial cells. J. Biol. Chem. 277, 15309-15316
456 49. Lesiak, A., Pelz, C., Ando, H., Zhu, M., Davare, M., Lambert, T. J., Hansen, K. F.,
457 Obrietan, K., Appleyard, S. M., Impey, S. and Wayman, G. A. (2013). A
458 genome-wide screen of CREB occupancy identifies the RhoA inhibitors Par6C
459 and Rnd3 as regulators of BDNF-induced synaptogenesis. PLoS One. 8, e64658
460
461

12
462 Figure legends
463 Figure 1 Cellular viability, apoptosis and CREB phosphorylation in the
464 hypoxia-treated chondrocyte C28/I2 cells.
465 C28/I2 cells were inoculated under hypoxia or normoxia for 12, 24 or 48 hours, and
466 then were analyzed for the cellular viability with MTT assay, for the apoptosis with
467 annexin V/FITC apoptosis detection kit, for the caspase 3 activity with AMC
468 Caspase-3Assay Kit and for the CREB phosphorylation with western blotting assay.
469 MTT assay results (A), Percentage of apoptotic cells (B), percentage of caspase 3
470 activity (C) of the C28/I2 cells under hypoxia or normoxia were presented
471 respectively; D-F: Western blotting assay of released cytochrome c (Cyt c release)
472 (E), cleaved caspase 3 (Clv-CASP 3) (E), CREB with (p-CREB) or without
473 phosphorylation (CREB) (F) in the hypoxia-treated C28/I2 cells for 0, 12, 24 or 48
474 hours. Each result was averaged for triple independent experiments, Statistical
475 significance was shown as * p <0.05, ** p <0.01, or *** p <0.001, ns: no significance.
476

477 Figure 2 Mitochondrial dysfunction in the hypoxia-treated C28/I2 cells.

478 C28/I2 cells were inoculated under hypoxia or normoxia for 12, 24 or 48 hours, and
479 then the percentage of mitochondrial superoxide levels (A), of ROS level (B), of
480 mitochondrial membrane potential (MMP) (C) and of Complex IV activity (D) were
481 assayed respectively. All results were averaged for three independent experiments.
482 *p<0.05, **p<0.01, *** p < 0.001 or ns: no significance.
483

484 Figure 3 RhoA overexpression and its regulation on the hypoxia-induced

485 apoptosis in C28/I2 cells.
486 C28/I2 cells were transfected with RhoA-pcDNA3.1(+) (pcDNA-RhoA) or with
487 control pcDNA3.1(+) (pcDNA-Con) and were incubated for 24 or 48 hours under
488 hypoxia, then the RhoA in protein level (A), the CREB with or without
489 phosphorylation (B), the cellular viability (C), percentage of apoptotic cells (D) and
490 the caspase 3 activity (E) were assayed respectively. All experiments were performed
491 independently in triplicate. Statistical significance was shown as * p <0.05, ** p
492 <0.01 or *** p <0.001, ns: no significance.
493

494 Figure 4 Mitochondrial dysfunction in the hypoxia-treated C28/I2 cells, post the
495 RhoA overexpression.
496 C28/I2 cells were transfected with pcDNA-RhoA or with pcDNA-Con and were
497 inoculated under hypoxia for24 or 48 hours, and then the percentage of mitochondrial
498 superoxide levels (A), of ROS level (B), of mitochondrial membrane potential (MMP)
499 (C) and of Complex IV activity (D) were assayed respectively. All results were
500 averaged for three independent experiments. *p<0.05, **p<0.01, or *** p < 0.001.
501

13
502 Figure 5 Lysophosphatidic acid (LPA) reduces the hypoxia-induced apoptosis in
503 C28/I2 cells.
504 C28/I2 cells were treated with 0 or 10 nM Lysophosphatidic acid (LPA), and then
505 were incubated for 24 or 48 hours under hypoxia, then the RhoA in protein level (A),
506 the CREB with or without phosphorylation (B), the percentage of apoptotic cells (C)
507 and the caspase 3 activity (D) were assayed respectively. All experiments were
508 performed independently in triplicate. Statistical significance was shown as * p <0.05,
509 or ** p <0.01, ns: no significance.
510

511 Figure 6 LPA ameliorates the hypoxia-induced mitochondrial dysfunction in

512 C28/I2 cells.
513 C28/I2 cells were treated with 0 or 10 nM Lysophosphatidic acid (LPA), and then
514 were incubated for 24 or 48 hours under hypoxia, and then the percentage of
515 mitochondrial superoxide levels (A), of ROS level (B), of mitochondrial membrane
516 potential (MMP) (C) and of Complex IV activity (D) were assayed respectively. All
517 results were averaged for three independent experiments. *p<0.05, **p<0.01, or *** p
518 < 0.001.
519
520

14