HIF-1α/Beclin1-Mediated Autophagy Is Involved in Neuroprotection Induced by Hypoxic Preconditioning
Hypoxic preconditioning (HPC) exerts a protective effect against hypoxic/ischemic brain injury, and one mechanism explaining this effect may involve the upregulation of hypoxia-inducible factor-1 (HIF-1). Autophagy, an endogenous protective mechanism against hypoxic/ischemic injury, is correlated with the activation of the HIF-1α/Beclin1 signaling pathway. Based on previous studies, we hypothesize that the protective role of HPC may involve autophagy occurring via activation of the HIF-1α/Beclin1 signaling pathway. To test this hypothesis, we evaluated the effects of HPC on oxygen-glucose deprivation/reperfusion (OGD/R)-induced apoptosis and autophagy in SH-SY5Y cells. HPC significantly attenuated OGD/R-induced apoptosis, and this effect was suppressed by the autophagy inhibitor 3-methyladenine and mimicked by the autophagy agonist rapamycin. In control SH-SY5Y cells, HPC upregulated the expression of HIF-1α and downstream molecules such as BNIP3 and Beclin1. Additionally, HPC increased the LC3-II/LC3-I ratio and decreased p62 levels. The increase in the LC3-II/LC3-I ratio was inhibited by the HIF-1α inhibitor YC-1 or by Beclin1-short hairpin RNA (shRNA). In OGD/R-treated SH-SY5Y cells, HPC also upregulated the expression levels of HIF-1α, BNIP3, and Beclin1, as well as the LC3-II/LC3-I ratio. Furthermore, YC-1 or Beclin1-shRNA attenuated the HPC-mediated cell viability in OGD/R-treated cells. Taken together, our results demonstrate that HPC protects SH-SY5Y cells against OGD/R via HIF-1α/Beclin1-regulated autophagy.
KeywordsHypoxic preconditioning Oxygen-glucose deprivation/reperfusion Autophagy HIF-1α Beclin1 SH-SY5Y cells
Cerebrovascular diseases, such as stroke, have a high incidence of morbidity and mortality. Despite decades of extensive research and clinical practice, effective treatments for stroke are still limited; thus, current research continues to explore effective means of prevention (Jain et al. 2016). One possible therapeutic approach is the use of hypoxic preconditioning (HPC). HPC, also called hypoxia-induced tolerance, is defined as a phenomenon in which a brief period of hypoxia protects against subsequent, otherwise lethal insults (Stetler et al. 2014). HPC activates multiple endogenous protective mechanisms against severe or even fatal ischemic stroke and other acute attacks (Elena and Mikhail 2015). In vivo and in vitro studies have confirmed the protective effects of HPC against cerebrovascular stroke, but the mechanism behind this protection remains unclear (Lu et al. 2005; Yu et al. 2013). Studies have shown that HPC protection is eliminated in hypoxia-inducible factor-1α (HIF-1α) knockout mice subjected to neonatal hypoxia/ischemia (Sheldon et al. 2014). HPC induces tolerance to oxygen-glucose deprivation (OGD) in astrocytes, at least in part, via HIF-1α-linked upregulation of the P450 epoxygenase pathway (Liu and Alkayed 2005). HIF-1α is critically involved in ischemic preconditioning against stroke by modulating inflammatory responses in rats (Yang et al. 2018), suggesting that HIF-1α plays a key role in the regulation of HPC protection.
Autophagy is a lysosome-dependent degradation pathway that can clear away damaged organelles such as mitochondria and biological macromolecules including proteins, thus maintaining intracellular homeostasis (Denton et al. 2015). Rapamycin (Rapa) regulates autophagy through inhibition of the nutrient-sensing molecule known as mammalian target of rapamycin (mTOR) (Nair and Ren 2012). Phosphatidylinositol 3-kinase (PI3K) is a key regulator of autophagy, as it plays an important role in many biological processes, including controlling the activation of mTOR. 3-Methyladenine (3-MA) inhibits autophagy by blocking autophagosome formation via the inhibition of class III PI3K (Abdulrahman et al. 2011; Jin et al. 2015). The autophagy level is low under physiological conditions but can be increased under hypoxic conditions (Park et al. 2018); this increase can aggravate injury but can also play a protective role (Denton et al. 2015). The effective control of autophagy is important for the development of new treatment strategies for hypoxia-induced brain injury. Recent studies show that HPC plays a neuroprotective role by activating modest autophagy (Sheng et al. 2014; Wang et al. 2013b; Wang et al. 2014), but the mechanism involved is unclear.
The HIF-1α signaling pathway is closely related to hypoxia-induced autophagy (Wu et al. 2015). HIF-1 is a critical nuclear transcription factor in hypoxia adaptation and is composed of a labile HIF-1α subunit and a stable HIF-1β subunit (Loboda et al. 2012; Semenza 2012). HIF-1α is degraded by proteasomes under normoxic conditions. In response to hypoxia, however, HIF-1α is transferred into the nucleus and interacts with the HIF-1β subunit to form HIF-1 to initiate or enhance transcription of effector genes, including BNIP3/BNIP3L, insulin-like growth factor II, and vascular endothelial growth factor (Chu and Jones 2016; Xiong et al. 2015).
B cell leukemia/lymphoma 2 (Bcl-2)/adenovirus (E1B)-19KD-interacting protein 3 (BNIP3) is one of the Bcl-2 homology 3 (BH-3)-only proteins. BNIP3 is known to be involved in the regulation of both apoptosis and autophagy (Vasagiri and Kutala 2014). This protein activates autophagy by inhibiting the function of mTOR (Shi et al. 2014) and dissociates Beclin1 from Beclin1/Bcl-2 or Beclin1/Bcl-XL complexes by competing with Beclin1 during the formation of the autophagosome (Ney 2015). Accumulated evidence shows that the activation of autophagy is regulated by the HIF-1α/BNIP3 pathway (Lu et al. 2016; Wu et al. 2015; Zhao et al. 2012). Beclin1 is a well-known key regulator of autophagy; this protein controls the autophagic process by regulating class III PI3K-dependent generation of phosphatidylinositol 3-phosphate (PI3P) and the subsequent recruitment of additional autophagy-associated (ATG) proteins that promote autophagosome formation (Wirawan et al. 2012), which plays an essential role in autophagy activation (Maejima et al. 2016; Sinha and Levine 2008). Recent studies have confirmed that mimicking hypoxia with CoCl2 induces autophagy via activation of the HIF-1/BNIP3/Beclin1 pathway in skeletal myotubes (Chen et al. 2017b). In this study, we adapted an oxygen-glucose deprivation/reperfusion (OGD/R) model in SH-SY5Y cells and determined whether the HIF-1α/Beclin1 signaling pathway modulates autophagy and whether such regulation contributes to HPC-induced protection against OGD/R.
Materials and Methods
Human SH-SY5Y neuroblastoma cells were purchased from the cell bank of the Chinese Academy of Sciences (Shanghai, China) and maintained in Invitrogen Dulbecco’s modified Eagle’s medium (DMEM; Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum, 3.7 g/L sodium bicarbonate, 100 units/mL penicillin, and 100 g/mL streptomycin (Gibco-Thermo Fisher Scientific) at 37 °C in a tissue culture incubator with 5% CO2 and 95% relative humidity. We added 5 mM 3-MA (Sigma, St. Louis, MO, USA) or 25 mM Rapa (Cell Signaling Technology, Danvers, MA, USA) to inhibit or induce autophagy, respectively.
OGD/R and HPC
Ischemic injury was induced in vitro using OGD/R, as described previously (Darshit and Ramanathan 2016; Guo et al. 2014; Zhu et al. 2016). Briefly, 5 × 104/mL cells were maintained in glucose-free and serum-free DMEM in a hypoxia incubation chamber (Thermo, Forma 3427, USA) under hypoxic conditions (1% O2, 5% CO2, and 94% N2) for 10 h and then cultured under normoxic conditions (5% CO2, 95% air) in normal DMEM for a further 12 h. For HPC (Peterson et al. 2011; Tzeng et al. 2010; Wang et al. 2013b), cells were successively incubated (5% O2, 5% CO2, and 90% N2) for 9 h and under normoxic conditions for 12 h in normal DMEM. The control culture was always maintained in normal DMEM and put in the incubator under normoxic conditions.
Cell Viability Assay
Cell viability was determined using MTT (3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide) assays. Briefly, 2 × 104 cells per well were seeded in 96-well plates and incubated at 37 °C for 24 h. After the indicated treatment, 10 μL MTT (5 mg/mL; Sigma-Aldrich) was added to each well, followed by incubation for 4 h at 37 °C. Thereafter, the culture medium was replaced with 150 μL dimethyl sulfoxide (DMSO; Sigma-Aldrich) to dissolve any blue formazan crystals. After 10 min of incubation at room temperature, the optical density (OD) was measured at a wavelength of 570 nm using a microplate reader (BioTek Instruments, Inc., Winooski, VT, USA).
Treatment with YC-1
3-(5′-Hydroxymethyl-2′-furyl)-1-benzylindazole (YC-1) (Cayman Chemical Company, Ann Arbor, MI, USA) was dissolved in DMEM containing 1% DMSO (Bu et al. 2011; Tsai et al. 2013; Yan et al. 2011). SH-SY5Y cells were pretreated with 10 μM YC-1 for 1 h before HPC (Tsui et al. 2013).
In Situ One-step TUNEL
Cell apoptosis was determined using a terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) kit according to the manufacturer’s instructions (Nanjing KGI Biological Technology Development Co. Ltd., China). Briefly, cells were fixed with 4% paraformaldehyde in PBS for 1 h followed by permeabilization with 0.1% Triton X-100 in PBS for 10 min and then incubated with TUNEL detection solution at 37 °C for 1 h. Cell nuclei were counterstained with DAPI (Roche Molecular Systems, Inc., Pleasanton, CA, USA) for 20 min. TUNEL-positive cells had pyknotic nuclei with dark green fluorescent staining, indicative of apoptosis. Images were captured using a fluorescence microscope (Olympus IX71, Japan). Five random visual fields from each of five slides were observed at a magnification of × 200. TUNEL-positive cells were counted using Image-Pro Plus 5.0 (Media Cybernetics Inc., Rockville, MD, USA). The experiments were independently repeated three times.
Flow Cytometric Evaluation of Apoptosis
Quantification of apoptotic cells was also performed by flow cytometry using Annexin V-fluorescein isothiocyanate (FITC) and propidium iodide (PI). In brief, cells were seeded in 24-well plates at a density of 1 × 105 cells per well. Cells were harvested by trypsinization and gently washed twice with PBS after the various treatments, as indicated in the figures. The cells were stained with Annexin V-FITC using an Apoptosis Detection Kit (Beijing Biosea Biotechnology Co. Ltd., China) and PI for 15 min in the dark. Finally, apoptosis was quantified using a flow cytometer (FACSCalibur, BD Biosciences, San Jose, CA, USA). The data were collected from at least 10,000 cells derived from three independent experiments.
Western Blotting Analysis
Total protein extracts were prepared from SH-SY5Y cells using radioimmunoprecipitation assay (RIPA) buffer. Equal amounts of protein samples (20–40 μg) were resolved on SDS-PAGE gels and transferred onto Immobilon polyvinyl difluoride membranes (GE Healthcare, Little Chalfont, UK) using standard protocols. The membranes were blocked with 4% bovine serum albumin (BSA) for 1 h at room temperature and probed with primary antibody against HIF-1α (1:2000; Cell Signaling Technology), BNIP3 (1:1500; Cell Signaling Technology), Beclin1 (1:1000; Abcam, Cambridge, MA, USA), LC3 (1:1000; Cell Signaling Technology), p62 (1:1000; Abcam), cleaved caspase-3 (1:2000; Cell Signaling Technology), GAPDH (1:2000; Cell Signaling Technology), or β-actin (1:1000; Beyotime Biotechnology, Haimen, China) at 4 °C overnight. After being washed with TBST (a mixture of tris-buffered saline and polysorbate 20), the blots were incubated with secondary goat anti-mouse or anti-rabbit antibody conjugated with horseradish peroxidase at room temperature for 2 h. After being washed three times, the proteins were visualized using an enhanced chemiluminescence kit (GE Healthcare). The proteins were quantified with ImageJ software and normalized to β-actin in parallel experiments.
Knockdown of Beclin1 Expression in SH-SY5Y Cells
ShRNA sequence information
Total RNA was extracted from cells with TRIzol Reagent (Thermo Fisher Scientific) and transcribed into cDNA using a reverse transcription kit (Takara Biotechnology Co. Ltd., Dalian, China). The relative mRNA level of the target gene was detected using the Roche LightCycler® 480 Real-Time PCR system (Roche, Basel, Switzerland) according to the SYBR® Premix Ex Taq instructions (Takara Biotechnology Co. Ltd.). The primer sequences were as follows: forward 5′-CAGGAACTCACAGCTCCATT-3′ and reverse 5′-CATCAGATGCCTCCCCAATC-3′ for Beclin1 and forward 5′-ACCACACCTTCTACAATGA-3′ and reverse 5′-ATAGCACAGCCTGGATAG-3′ for β-actin. The results were quantified by the comparative CT (threshold cycle) method using β-actin as an internal control (Chen et al. 2017a).
Data were analyzed using SPSS 18.0 software (SPSS Inc., Chicago, IL, USA). All values are presented as the mean ± standard deviation. Statistical analyses for the comparison of two groups were performed using Student’s t test, and one-way analysis of variance followed by a post hoc least significant difference multiple comparison test was used for the comparison of more than two groups. P < 0.05 was considered statistically significant.
HPC Protects SH-SY5Y Cells Against OGD/R Injury
HPC Activates Autophagy in SH-SY5Y Cells
Protein p62, one of the selective substrates of autophagy, is localized at autophagosomes and constantly degraded by the autophagy-lysosomal pathway, which increases in response to a decrease in autophagic activity or a total loss of autophagic function (He et al. 2015). As shown in Fig. 2c, d, the p62 level was decreased in cells treated with HPC or OGD/R compared with that in cells kept under normal conditions. The p62 level in cells subjected to HPC followed by OGD/R was also decreased, reaching a value significantly lower than that in cells treated with OGD/R alone.
The results demonstrated that both HPC and OGD/R upregulated the autophagy level in SH-SY5Y cells. Moreover, HPC further increased the autophagy level, followed by OGD/R.
Autophagy Is Involved in the Neuroprotection of HPC Against OGD/R-Induced Injury
HPC-Induced Autophagy Is Associated with the HIF-1α/BNIP3/Beclin1 Signaling Pathway in Control Cells
HPC Enhanced Autophagy and the HIF-1α/BNIP3/Beclin1 Signaling Pathway in OGD/R-Injured Cells
HIF-1α/Beclin1 Signaling Is Involved in the Neuroprotective Effect of HPC Against OGD/R-Induced Injury
Ischemia reperfusion causes neuronal injury (Lin et al. 2016). SH-SY5Y human neuroblastoma cells are derived from the neural crest during neurodevelopment, exhibiting similar biochemical and functional properties to neurons (Shipley et al. 2016). In this study, a model of OGD/R-induced SH-SY5Y cell injury was established by using a hypoxic incubator, as in previous reports (Darshit and Ramanathan 2016; Guo et al. 2014; Zhu et al. 2016).
HPC refers to a nonfatal hypoxia exposure that confers a high degree of tolerance to subsequent severe hypoxia or other normally fatal stress, thereby offering a protective mechanism (Nikiforou et al. 2015). HPC has been confirmed to play a protective role not only in cardiovascular and renal systems but also in the central nervous system. However, the underlying mechanisms remain poorly understood. Recent study found that autophagy contributes to the neuroprotection of HPC against global cerebral ischemia and ethanol neurotoxicity in rats (Wang et al. 2013b; Zhan et al. 2017). Our study showed that HPC protects SH-SY5Y cells against OGD/R injury. A previous study reported that HPC did not cause neurotoxicity but induced autophagy (Shao and Lu 2012), and our data showed that HPC not only activates autophagy in normal SH-SY5Y cells but also further enhances autophagy in OGD/R-treated SH-SY5Y cells.
Autophagy, a mechanism of defense and stress regulation, plays an important role in maintaining cellular homeostasis and cell survival (Galluzzi et al. 2014). Autophagy has both pro-survival and pro-death functions, and recent studies have confirmed that enhancing autophagy has a protective effect against brain damage (Chen et al. 2012; Dai et al. 2017; Shen et al. 2016; Xie et al. 2016). Our results demonstrated that the autophagy agonist Rapa provided protection against OGD/R-induced injury. Moreover, 3-MA attenuated HPC-induced protection. Taken together, these results suggest that autophagy is critically involved in the protective effect of HPC, consistent with previous studies (Cui et al. 2015; Tzeng et al. 2010).
The HIF-1α signaling pathway can contribute to hypoxia-induced autophagy (Wu et al. 2015). HIF-1α increases the expression of BNIP3 under hypoxic conditions (Bellot et al. 2009; Li et al. 2013). In addition, BNIP3 is closely related to autophagy (Fu et al. 2018; Wang et al. 2013a); the activation of BNIP3 leads to the accumulation of LC3-II under hypoxia (Hanna et al. 2012). Beclin1 is an essential gene related to autophagy, and Beclin1-PI3KC3 complexes are key signaling complexes required for autophagosome formation (Wirth et al. 2013). Furthermore, mimicking hypoxic conditions with CoCl2 induces autophagy via activation of the HIF-1α/BNIP3/Beclin1 signaling pathway (Chen et al. 2017b).
Our results also showed that HPC activated autophagy while upregulating the HIF-1α/BNIP3/Beclin1 signaling pathway in SH-SY5Y cells, an effect that was counteracted by YC-1 treatment. Moreover, knockdown of Beclin1 expression suppressed HPC-induced autophagy without any effect on the expression of HIF-1α or BNIP3. Based on these findings and on previous studies (Bellot et al. 2009; Zhao et al. 2012), we conclude that HPC activates autophagy through the HIF-1α/BNIP3/Beclin1 signaling pathway in SH-SY5Y cells. In addition, HPC further activates autophagy and enhances HIF-1α/BNIP3/Beclin1 pathway in OGD/R-treated cells. Either treatment with YC-1 or knockdown of Beclin1 expression attenuates HPC-induced neuroprotection, suggesting that the HIF-1α/Beclin1 signaling pathway is involved in HPC-induced autophagy and neuroprotection.
In conclusion, our study reveals that HPC attenuates OGD/R-induced injury in SH-SY5Y cells by activating autophagy, and the HIF-1α/Beclin1 signaling pathway is involved in HPC-activated autophagy. Furthermore, HIF-1α and Beclin1 are also involved in HPC-induced neuroprotection.
This work was supported in part by two grants from the National Natural Science Foundation of China (NSFC; grant numbers: 81771517, 81571205) and by a Science and Technology Key Research Project Grant from the Department of Education of Henan Province (No. 14A310009).
Compliance with Ethical Standards
Conflict of Interest
The authors declare that they have no conflicts of interest.
- Abdulrahman BA, Khweek AA, Akhter A, Caution K, Kotrange S, Abdelaziz DH, Newland C, Rosales-Reyes R, Kopp B, McCoy K, Montione R, Schlesinger LS, Gavrilin MA, Wewers MD, Valvano MA, Amer AO (2011) Autophagy stimulation by rapamycin suppresses lung inflammation and infection by Burkholderia cenocepacia in a model of cystic fibrosis. Autophagy 7:1359–1370. https://doi.org/10.4161/auto.7.11.17660 CrossRefPubMedPubMedCentralGoogle Scholar
- Bellot G, Garcia-Medina R, Gounon P, Chiche J, Roux D, Pouysségur J, Mazure NM (2009) Hypoxia-induced autophagy is mediated through hypoxia-inducible factor induction of BNIP3 and BNIP3L via their BH3 domains. Mol Cell Biol 29:2570–2581. https://doi.org/10.1128/MCB.00166-09 CrossRefPubMedPubMedCentralGoogle Scholar
- Chen CX, Huang J, Tu GQ, Lu JT, Xie X, Zhao B, Wu M, Shi QJ, Fang SH, Wei EQ, Zhang WP, Lu YB (2017a) NAMPT inhibitor protects ischemic neuronal injury in rat brain via anti-neuroinflammation. Neuroscience 356:193–206. https://doi.org/10.1016/j.neuroscience.2017.05.022 CrossRefPubMedGoogle Scholar
- Dai R, Zhang S, Duan W, Wei R, Chen H, Cai W, Yang L, Wang Q (2017) Enhanced autophagy contributes to protective effects of GM1 ganglioside against abeta1-42-induced neurotoxicity and cognitive deficits. Neurochem Res 42:2417–2426. https://doi.org/10.1007/s11064-017-2266-0 CrossRefPubMedGoogle Scholar
- Fu R, Deng Q, Zhang H, Hu X, Li Y, Liu Y, Hu J, Luo Q, Zhang Y, Jiang X, Li L, Yang C, Gao N (2018) A novel autophagy inhibitor berbamine blocks SNARE-mediated autophagosome-lysosome fusion through upregulation of BNIP3. Cell Death Dis 9:243. https://doi.org/10.1038/s41419-018-0276-8 CrossRefPubMedPubMedCentralGoogle Scholar
- Hanna RA, Quinsay MN, Orogo AM, Giang K, Rikka S, Gustafsson AB (2012) Microtubule-associated protein 1 light chain 3 (LC3) interacts with Bnip3 protein to selectively remove endoplasmic reticulum and mitochondria via autophagy. J Biol Chem 287:19094–19104. https://doi.org/10.1074/jbc.M111.322933 CrossRefPubMedPubMedCentralGoogle Scholar
- Jain IH, Zazzeron L, Goli R, Alexa K, Schatzman-Bone S, Dhillon H, Goldberger O, Peng J, Shalem O, Sanjana NE, Zhang F, Goessling W, Zapol WM, Mootha VK (2016) Hypoxia as a therapy for mitochondrial disease. Science 352:54–61. https://doi.org/10.1126/science.aad9642 CrossRefPubMedPubMedCentralGoogle Scholar
- McFarland AJ, Grant GD, Perkins AV, Flegg C, Davey AK, Allsopp TJ, Renshaw G, Kavanagh J, McDermott CM, Anoopkumar-Dukie S (2013) Paradoxical role of 3-methyladenine in pyocyanin-induced toxicity in 1321N1 astrocytoma and SH-SY5Y neuroblastoma cells. Int J Toxicol 32:209–218. https://doi.org/10.1177/1091581813482146 CrossRefPubMedGoogle Scholar
- Tzeng YW, Lee LY, Chao PL, Lee HC, Wu RT, Lin AM (2010) Role of autophagy in protection afforded by hypoxic preconditioning against MPP+-induced neurotoxicity in SH-SY5Y cells. Free Radic Biol Med 49:839–846. https://doi.org/10.1016/j.freeradbiomed.2010.06.004 CrossRefPubMedGoogle Scholar
- Xie C, Ginet V, Sun Y, Koike M, Zhou K, Li T, Li H, Li Q, Wang X, Uchiyama Y, Truttmann AC, Kroemer G, Puyal J, Blomgren K, Zhu C (2016) Neuroprotection by selective neuronal deletion of Atg7 in neonatal brain injury. Autophagy 12:410–423. https://doi.org/10.1080/15548627.2015.1132134 CrossRefPubMedPubMedCentralGoogle Scholar
- Yang J, Liu C, Du X, Liu M, Ji X, Du H, Zhao H (2018) Hypoxia inducible factor 1α plays a key role in remote ischemic preconditioning against stroke by modulating inflammatory responses in rats. J Am Heart Assoc 7:e007589. https://doi.org/10.1161/JAHA.117.007589 CrossRefPubMedPubMedCentralGoogle Scholar
- Zhan L, Chen S, Li K, Liang D, Zhu X, Liu L, Lu Z, Sun W, Xu E (2017) Autophagosome maturation mediated by Rab7 contributes to neuroprotection of hypoxic preconditioning against global cerebral ischemia in rats. Cell Death Dis 8:e2949. https://doi.org/10.1038/cddis.2017.330 CrossRefPubMedPubMedCentralGoogle Scholar
- Zhu Q, Zhang Y, Liu Y, Cheng H, Wang J, Zhang Y, Rui Y, Li T (2016) MLIF alleviates SH-SY5Y neuroblastoma injury induced by oxygen-glucose deprivation by targeting eukaryotic translation elongation factor 1A2. PLoS One 11:e0149965. https://doi.org/10.1371/journal.pone.0149965 CrossRefPubMedPubMedCentralGoogle Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.