http://orcid.org/0000-0003-2532-5218
Figures
Abstract
Tibetan pigs have survived at high altitude for millennia and they have a suite of adaptive features to tolerate the hypoxic environment. However, the molecular mechanisms underlying the regulation of hypoxia-adaptive phenotypes have not been completely elucidated. In this study, we analyzed differentially expressed genes (DEGs), biological pathways and constructed co-expression regulation networks using whole-transcriptome microarrays from lung tissues of Tibetan and Duroc pigs both at high and low altitude. A total of 3,066 DEGs were identified and this list was over-represented for the ontology terms including metabolic process, catalytic activity, and KEGG pathway including metabolic pathway and PI3K-Akt signaling pathway. The regulatory (RIF) and phenotypic (PIF) impact factor analysis identified several known and several potentially novel regulators of hypoxia adaption, including: IKBKG, KLF6 and RBPJ (RIF1), SF3B1, EFEMP1, HOXB6 and ATF6 (RIF2). These findings provide new details of the regulatory architecture of hypoxia-adaptive genes and also insight into which genes may undergo epigenetic modification for further study in the high-altitude adaptation.
Citation: Jia C, Kong X, Koltes JE, Gou X, Yang S, Yan D, et al. (2016) Gene Co-Expression Network Analysis Unraveling Transcriptional Regulation of High-Altitude Adaptation of Tibetan Pig. PLoS ONE 11(12): e0168161. https://doi.org/10.1371/journal.pone.0168161
Editor: Ramona Natacha PENA i SUBIRÀ, Universitat de Lleida, SPAIN
Received: August 30, 2016; Accepted: November 27, 2016; Published: December 9, 2016
Copyright: © 2016 Jia et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files. All rawdata files are available from Gene Expression Omnibus (GEO) database (Accession number: GSE84409).
Funding: This work was supported by the National High Technology Research and Development Program 863 (http://program.most.gov.cn/) Grant # 2013AA102503, to XG; National Natural Science Foundation of China (http://www.nsfc.gov.cn) Grant # U1036604, to ZHW and National Natural Science Foundation of China (http://www.nsfc.gov.cn) Grant # 31260542 to DWY. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The hypoxic environment at high altitude imposes extreme physiological challenges in mammals. Classical response to hypoxia is characterized by systemic changes in cardiovascular, respiratory, and hematopoietic functions that impact convective oxygen transport. Failure of these systems to adapt leads to altitude illness or even death [1, 2]. Native high-altitude species have been selected through evolutionary processes to have heritable genetic adaptations [3] in morphological, anatomical, physiological [4], biochemical, and behavioral traits to survive under low oxygen tension [5–7]. For example, Tibetan human populations showed lower hypoxic pulmonary vasoconstrictor response and higher blood oxygen saturation with similar hemoglobin levels to those expected at sea level [6–9]. Recently, a number of genome-wide analyses have been conducted to determine the genetic bases of adaptation to high altitude in species such as: highlander [10–15], yak [16, 17], Tibetan antelope [18], Tibetan mastiff [19], Tibetan pig [3, 20], Tibetan grey wolf [21], Tibetan Chicken [22], Tibetan cashmere goat [23] and Tibetan sheep [24]. Notable candidate genes related to high altitude adaptation included members of the HIF signaling pathway as well as genes associated with hemoglobin levels. However, the molecular mechanisms underlying adaptation to high altitude are yet to be completely elucidated.
The Diqing Tibetan pig (Sus scrofa) is a native breed that mainly inhabits Diqing Tibetan Autonomous Prefecture of the Yunnan province of China (average altitude: 3,700 m). Although their meat production, reproduction and growth rate is not as good as domesticated pigs living at sea level, adaptation to hypoxia, cold climate, disease and harsh environment makes the Tibetan pig the only adapted breed to these conditions. Tibetan pigs have adapted to high altitude with well-developed heart and lungs and a blunted erythropoietic response [20]. Since porcine physiology and anatomy are similar to humans [25], the Tibetan pig could be a suitable animal model for human high altitude diseases.
The objective of this study was to identify transcripts, pathways and potential regulators involved in high-altitude adaptation. Transcriptome profiling of lung tissues from Tibetan pigs and Duroc pigs was analyzed using Agilent microarrays, and potential pathway regulators were identified using two methods: partial correlation and information theory (PCIT) [26] and regulatory impact factor (RIF) algorithms [27–29]. The goal of this research is to identify pathways involved in high-altitude adaptation that may provide potential therapeutic solutions of hypoxia-induced diseases.
Materials and Methods
Ethics Statement
All animal experimental procedures were performed according to the Guide for the Care and Use of Laboratory Animals (Ministry of Science and Technology of China, 2006). The protocol was approved by the Experimental Animal Management Committee (EAMC) of Northwest A&F University ([2005]142).
Animal and sample preparation
Tibetan pigs were raised at Yunnan (Shangri-La Specialized Farmers Cooperatives, Tibetan Autonomous Prefecture of Diqing, 3,500 m) (Tibetan pigs at high altitude, herein mentioned as TH), and Duroc pigs were raised at Guangdong (Guangdong Yifa Breeding Co. Ltd, Huizhou, 40 m) (Duroc pigs at low altitude, herein mentioned as DL). In our experiment, we employed a migrant study design. Sixteen Tibetan female piglets from the TH population and sixteen Duroc female piglets from the DL population were selected with similar weight and non-genetic relationship. Among these, 8 piglets from each groups were migrated to low altitude (TL) or high altitude (DH) from their original rearing facility at 1 month of age, respectively (Fig 1). Animal were fed and watered ad libitum. At 6 month of age, 2 pigs were selected at random from each group for harvest (TH, TL, DH and DL). These animals were feed-restricted for 12 h and sacrificed by electric shock. Lung samples were harvested immediately and snap frozen in liquid nitrogen and then stored at -80°C until utilized for RNA extraction.
The native population included Tibetan pigs living at high altitude (TH) and Duroc pigs living at low altitude (DL). At 1 month of age, half of the animals were migrated to low altitude (TL) or high altitude (DH). Animals were harvested at 6 month of age.
Total RNA isolation and microarray hybridization
Total RNA was extracted from each lung sample using the Trizol RNA isolation method (Invitrogen, USA). RNA was purified and DNase treated with QIAGEN RNeasy® Mini kit (QIAGEN, Valencia, CA) according to the manufacturer’s instructions. The integrity and quality of the extracted RNA was evaluated by the Agilent 2100 bioanalyzer (Agilent Technologies, Inc., Santa Clara, CA, USA). The RNA Integrity Number (RIN) values of all samples were greater than 8.0. The RNA was amplified and labeled using Agilent Low Input Quik Amp Labeling Kit, One-Color (Agilent Technologies, USA). Labeled cRNA from 8 animals were purified by QIAGEN RNeasy mini kit and hybridized individually (1 sample per array) using commercially available Agilent Whole Porcine Genome Oligo (4 × 44 K) Microarrays (Agilent-020109, one-color platform) according to the instructions in the Agilent Expression Analysis Technical Manual. After the hybridization, the gene chips were washed using Agilent Gene Expression Wash Buffer Kit at room temperature.
Microarray data analysis and statistics
After drying, slides were scanned using an Agilent G2505B microarray scanner (Agilent Technologies, USA). Image files were analyzed with Agilent Feature Extraction software (version10.7.3.1) to obtain the raw expression data. All raw gene expression values have been deposited in the Gene Expression Omnibus (GEO) database (Accession number: GSE84409). If a transcript was flagged as “A” by the scanner, i.e. signal intensity was less than background level, it was considered to be “not expressed”. Similarly, the transcripts with “P” flags, expression greater than background levels, were considered to be “expressed”. In our study, expressed transcripts were defined as being present in at least one sample. In total, 34,359 expressed probes were used for subsequent analyses, among these, only 16,946 were annotated, corresponding to about 10,000 genes (including LOC). More probes were annotated using GenBank and Basic local Alignment Search tool program (BLAST, NCBI) (http://www.ncbi.nlm.nih.gov/).
Expression data were normalized using Robust Multichip Analysis (RMA). Statistical analysis was performed using two-way ANOVA in an R script developed by Pavlidis et al. [30] to identify differentially expressed genes (DEGs) related to high altitude tolerance. The criterion on detection of DEGs was set at FDR < 0.0001. Gene expression, y = log (expression), was modeled as follows:
Where yijk was dependent variable, μ was the mean expression level of the gene, Si was fixed effect of breed (i = 1 to 2), Rj was fixed effect of altitude (j = 1 to 2), (S•R)ij was interaction between breed and altitude, and Ɛ ijk represented the random error. Thus, the expression level y could be impacted by main effects due to S, R, and interactions between them (S•R), plus random error.
Pathway and functional analysis of DEGs
The probe set IDs for GenBank Accession Number based on the Agilent porcine array annotation file were first mapped to NCBI Entrez gene IDs generated by BioGps software [31]. Gene Ontology and pathway analysis was conducted by PANTHER version 11.1 [32] (http://www.pantherdb.org/) and the online STRING tools (http://string-db.org/) to identify significantly over-represented GO terms, PANTHER (Bonferroni correction P < 0.05) and KEGG pathways (FDR<0.01).
Gene co-expression network construction and RIF analysis
A modified version of PCIT [26] was used to compute the co-expression correlation in the TH (Tibetan at high altitude) or DH (Duroc at high altitude) group, respectively. Only transcripts with a significant partial correlation |r| ≥ 0.95 were selected to construct gene co-expression networks. Transcriptional regulators (TRs) include mainly transcription factors, signaling molecules and chromatin remodelers [28]. In this study, these molecules were mined from the literature in human [33] and bovine [28] and from the SWISS-PROT database for swine and mouse (http://www.uniprot.org/) [34]. The final non-redundant transcriptional regulator list of genes in our annotation array totaled 1,897 (S3 Table). According to the RIF and phenotype impact factor (PIF) algorithm [28, 29], the pivotal transcriptional regulators related to adaptation to high altitude can be identified based on the RIF1 or RIF2 rank. Co-expression networks were visualized in Cytoscape v3.2.1 software [35] for TH and DH by selecting the top five positive and negative TRs based on the RIF1 or RIF2 z-scores and their targets (DEGs and differential PIF) with significant partial correlation from PCIT results. The statistical analysis of differential PIF was performed using the same method described for DEGs (two-way ANOVA) and the criterion was set at FDR < 0.0001. All processes, experiments and analyses are shown in Fig 2.
Abbreviations: TH = Tibetan pigs at high altitude, DH = Duroc pigs at high altitude, TL = Tibetan pigs at low altitude, DL = Duroc pigs at low altitude, DEGs = Differential Expression Genes, PCIT = Partial Correlation and Information Theory, RIF = Regulatory Impact Factor, PIF = Phenotype Impact Factor.
Results
Differentially expressed genes
A total of 3,431 differential expression transcripts were identified by two-way ANOVA based on a threshold of FDR < 0.0001, which were represented 3,066 non-redundant genes. These DEGs reflected the difference between breeds (Tibetan vs. Duroc) affected by the interaction that may be responsible for the adaptation of Tibetan pigs to high altitude (S1 and S4 Tables). These genes were significant both due to the interaction effect and also the breed effect. These genes were selected to emphasize the objective to identify genes DE due to adaptation of the breed to differing altitudes.
Functional analysis of DEGs affected by hypoxia
The annotations for DEGs were tested for over-representation of biological pathways and processes. Significantly enriched terms included: 7 biological processes (BP), 4 molecular functions (MF), 3 PANTHER pathways, 7 cellular components, 3 PANTHER protein class (Bonferroni corrected P < 0.05) and 29 KEGG (FDR < 0.01) (S2 Table). The most significantly enriched terms were metabolic process (GO: 0008152), catalytic activity (GO: 0003824) and CCKR signaling map (P06959) in BP, MF and PANTHER pathway (Table 1), respectively. The most significantly enriched KEGG pathways were the metabolic pathway and PI3K-Akt signaling pathway (Fig 3).
Putative transcription regulators of DEGs identified by RIF and PIF
To identify which crucial TRs may affect gene expression difference and above-mentioned GO categories and pathways, we computed PIF, RIF1 and RIF2 scores. All probe’s PIF values are provided in S3 Table. Among these, 3,656 differentially PIF transcripts (FDR < 0.0001) were identified using two-way ANOVA (S4 Table). All RIF scores are provided in S5 Table. We identified 369 non-redundant TRs with RIF z-scores < -2 or > 2. Among them, 117 TRs had RIF1 z-scores > 2 and 101 had RIF1 z-scores < -2. In addition, 136 TRs had RIF2 z-scores > 2 and 125 had RIF2 z-scores < -2. Moreover, 91 TRs were common for RIF1 and RIF2 based on z-scores < -2 or > 2. Tables 2 and 3 listed the top 5 positive or negative extreme regulators, which were selected according to RIF1 or RIF2 z-scores.
Co-expression gene patterns at high altitude
Here we applied the correlation relationships from the PCIT results to construct co-expression correlation network (|r| ≥ 0.95) between 20 TRs (Tables 2 and 3) with significantly DE and PIF genes in the TH and DH breed groups, respectively. Networks were visualized with Cytoscape software and are presented in (Fig 4). We identified four TRs (TiPARP, SF3B1, HABP4, ERG) with no connection to DEGs and PIF genes in the TH group, and three TRs (EFEMP1, MAGED1, HIST1H3E) also with no connection in DH group.
(A) Tibetan group at high altitude (660 nodes and 1878 edges) (B) Duroc group at high altitude (512 nodes and 1641 edges). Node shape and color were mapped to the different gene types: transcription regulators (triangles, yellow), differential expression genes and differential PIF genes (circles, black).
Discussion
Tibetan pigs have lived at high altitude for thousands of years, which provided an excellent animal model for investigating mechanisms of hypoxia adaptation. Natives of high altitude have been documented to possess genetic variations that may be the results of compensatory genetic changes and natural selection. However, the high altitude environment also plays an important role in adaptation, because environmental effects modify gene transcription and/or translation that may irreversibly affect the phenotype [36]. If we take an integrative approach to analyze the mechanisms related to hypoxia adaptation, the effect of the gene must be evaluated with respect to the way that it interacts with the environment. Therefore, we employed a complete migrant design, where Tibetan pigs and Duroc pigs were raised in both their native altitude environments and also as migrants in a non-native environment. Changes in gene expression due to this migration indicated gene-by-environment interaction [37, 38]. In our study, a total of 3,066 DEGs were identified (FDR < 0.0001) as overlapping due to genetic variance (breed) and genetic×environment interaction effects. These DEGs were significantly over-represented for genes involved in metabolic process, catalytic activity, CCKR signaling and angiogenesis pathway by PANTHER classification categories (S2 Table) (Bonferroni corrected P < 0.05), and metabolic pathways and the PI3K-Akt signaling pathway by KEGG categories (Fig 3 and S2 Table) (FDR < 0.01).
Gene expression is regulated by TRs that tend not to be differentially expressed, because their activation is often modulated post-transcriptionally, i.e. via reversible phosphorylation, cellular localization, cofactor or ligand binding and so on [39]. To understand the genetic adaptation of Diqing Tibetan pigs to low oxygen, the RIF and PIF algorithms were used to identify potential TRs of DEGs. Correlation of gene expression levels within a network was determined using PCIT to construct regulation and co-expression networks (Fig 4). The PIF value weights the differential expression of a given gene by its overall abundance. The number of differentially PIF genes was greater than the number of DEGs (S4 Table), since differentially PIF genes could either have a large DEG or be a high abundant gene not hugely DEG [29]. RIF1 scores identify potential regulators [29] with large differential wring (i.e. change in correlation) to highly abundant, high differentially expressed genes between the two breeds at high altitude. RIF2 scores rank TRs as potential biomarkers to predict the abundance of DEGs [29]. In our study, we ranked 3,373 TRs (1,897 non-redundant) by RIF1 and RIF2 scores (S5 Table), and the top ten RIF1 or RIF2 TRs were listed in Tables 2 and 3.
In previous studies, the native highlanders were identified to have many adaptive features such as blunted hypoxic pulmonary vasoconstriction, increased erythropoiesis and metabolic reprogramming [5–7]. Moreover, mounting evidence suggests that the observed physiological adaptations are dependent on the HIF pathway [40–42]. In our study, 16 DEGs were also enriched as members of the HIF pathway (Fig 3 and S2 Table). Besides erythropoiesis, additional processes regulated by HIF are angiogenesis, glucose uptake and metabolism, which regulate oxygen delivery and consumption [11, 12, 14, 43]. In our study, most of the TRs identified as putative regulators are involved in response to hypoxia, angiogenesis and metabolic processes. In the top ten RIF1 gene list, inhibitor of kappa light polypeptide gene enhancer in B-cells, kinase gamma (IKBKG) gene, Kruppel-like factor 6 (KLF6) and recombination signal binding protein for immunoglobulin kappa J region (RBPJ) were identified (Table 2). IKBKG encodes the inhibitor of the κB kinase gamma (IKKγ) protein, also known as nuclear factor κB (NF-κB) essential modulator (NEMO). IKKγ is essential for the activation of the NF-κB pathway that plays a major role in immune response, inflammation, cell adhesion, cell survival and development [44]. IKKγ also protects HIF2α from proteasomal degradation and promotes its accumulation [45]. Under hypoxic conditions, HIFα forms a heterodimer that stabilizes HIFβ, which binds to hypoxia response elements in the promoter regions of hypoxia sensitive genes such as VEGF to induce gene transcription [46, 47]. HIF-dependent transcriptional changes lead to a broad range of cellular adaptations, including metabolic, proliferative, apoptotic, and angiogenic changes [48, 49]. KLF6 is a ubiquitously expressed member of the Kruppel-like family of transcription factors. KLF6 regulates the expression of multiple genes and is involved in tissue differentiation [50]. Under hypoxia condition, KLF6 is a potential transactivator of inducible nitric oxide synthase expression, which can promote the synthesis of Nitric oxide (NO) [51]. Elevated NO production helps populations living at high altitude avoid hypoxia by promoting vasodilation of pulmonary vasculature [52]. RBPJ is a transcription factor in the Notch signaling pathway [53]. Hypoxia activates the Notch intracellular domains associated with RBPJ recruitment of HIFα as transcriptional coactivators to target Notch downstream genes to regulate sprouting angiogenesis and vascular remodeling [54].
Splicing factor 3b subunit 1 (SF3B1), epidermal growth factor containing fibulin-like extracellular matrix protein 1 (EFEMP1), HOXB6 and activated transcription factor 6 (ATF6) were identified as the top-ten ranked RIF2 genes (Table 3). SF3B1 is considered to be a core factor of the spliceosome that contributes to the production of multiple distinct mRNAs [55]. It is identified as a HIFα target. Under hypoxia, the HIFα dependency of induction of a SF3B1 signal pathway enhances activation of fructose metabolism. Fructose metabolism is a central component of hypoxia-driven metabolic programs to maintain macromolecular such as lipids, proteins and nucleotide biosynthetic capacity to support the anabolic energy requirements [56]. EFEMP1 is a member of the fibulin family of extracellular glycoproteins. Fibulins have been shown to modulate cell morphology, growth, adhesion and motility [57]. EFEMP1 expression was positively correlated with microvascular density and the expression of VEGF [58]. HOXB6 is considered as a hallmark of erythropoiesis. Homeobox genes appear to be strong candidate genes to regulate hematopoiesis [59, 60]. The regulation of erythropoiesis is crucial for optimal oxygen delivery to the tissues under hypoxia [61]. ATF6 is an endoplasmic reticulum (ER) stress sensors. “ER stress” can be caused by a multitude of factors, including severe hypoxia [62]. ATF6 modulates ER function to protect cells against chronic stress [63]. Hypoxia induced changes in gene expression likely occur beyond the change of nucleotide sequence through genome-wide alterations in histone modification and DNA methylation [64]. Such modifications may result in global modulation of transcription levels [65]. In the top ten RIF1 or RIF2 genes, WD repeat domain 5 (WDR5), histone cluster 1, H3e (HIST1H3E) and chromobox protein homolog 5 (LOC100519426), were identified as putative regulators. These proteins are associated with histone modifications [66–68]. This result may indicate the potential for epigenetic regulation of adaptation to hypoxia; however, further investigation of offspring in hypoxic environments is needed to determine if this is plausible.
Conclusions
In the present study, we used a complete migrant design to study the interaction between genetic and environmental factors involved in the adaptation of Tibetan pigs to hypoxia. DEGs were identified due to breed by interaction effects that enriched for genes involved in metabolic processes and pathway, PI3K-Akt signaling pathway and HIF signal pathway. These biological processes could have been directly or indirectly regulated by the partial pressure of oxygen from the environment. In addition, putative pathway regulators of hypoxia adaptation were identified by RIF and co-expression methods. These regulators may affect gene expression by post-transcriptional or epigenetic modification. These findings contribute to a better understanding of the molecular mechanisms potentially underlying hypoxia adaptation, and may provide important information relevant to other species, especially humans.
Supporting Information
S1 Table. Results of microarray data analysis.
https://doi.org/10.1371/journal.pone.0168161.s001
(XLSX)
S2 Table. GO, PANTHER and KEGG pathways results of DEGs.
https://doi.org/10.1371/journal.pone.0168161.s002
(XLSX)
S4 Table. Differential expression genes and differential PIF.
https://doi.org/10.1371/journal.pone.0168161.s004
(XLSX)
S5 Table. The results of RIF1 and RIF2 z-scores.
https://doi.org/10.1371/journal.pone.0168161.s005
(XLSX)
Acknowledgments
We are grateful to Qiang Wang and Xiaofeng Zhang for their kind help in transport and management of our sampled pigs.
Author Contributions
- Conceptualization: CLJ ZHW.
- Formal analysis: CLJ ZHW JEK.
- Funding acquisition: ZHW.
- Investigation: XYK CLJ XG SLY SXL DWY.
- Project administration: SXL.
- Supervision: ZHW.
- Visualization: CLJ.
- Writing – original draft: CLJ XYK ZHW.
- Writing – review & editing: JEK ZHW SXL.
References
- 1. Hackett PH, Roach RC. High-altitude illness. The New England journal of medicine. 2001; 345(2):107–14. pmid:11450659
- 2. MacInnis MJ, Koehle MS, Rupert JL. Evidence for a genetic basis for altitude illness: 2010 update. High altitude medicine & biology. 2010; 11(4):349–68.
- 3. Li M, Tian S, Jin L, Zhou G, Li Y, Zhang Y, et al. Genomic analyses identify distinct patterns of selection in domesticated pigs and Tibetan wild boars. Nature genetics. 2013; 45(12):1431–8. pmid:24162736
- 4. Sakai A, Matsumoto T, Saitoh M, Matsuzaki T, Koizumi T, Ishizaki T, et al. Cardiopulmonary hemodynamics of blue-sheep, Pseudois nayaur, as high-altitude adapted mammals. The Japanese journal of physiology. 2003; 53(5):377–84. pmid:14975184
- 5. Beall CM. Tibetan and Andean patterns of adaptation to high-altitude hypoxia. Human biology. 2000; 72(1):201–28. pmid:10721618
- 6. Moore LG. Human genetic adaptation to high altitude. High altitude medicine & biology. 2001; 2(2):257–79.
- 7. Beall CM. Two routes to functional adaptation: Tibetan and Andean high-altitude natives. Proceedings of the National Academy of Sciences of the United States of America. 2007; 104 Suppl 1:8655–60.
- 8. Beall CM, Brittenham GM, Strohl KP, Blangero J, Williams-Blangero S, Goldstein MC, et al. Hemoglobin concentration of high-altitude Tibetans and Bolivian Aymara. American journal of physical anthropology. 1998; 106(3):385–400. pmid:9696153
- 9. Wu T, Wang X, Wei C, Cheng H, Wang X, Li Y, et al. Hemoglobin levels in Qinghai-Tibet: different effects of gender for Tibetans vs. Han. Journal of applied physiology (Bethesda, Md: 1985). 2005; 98(2):598–604.
- 10. Simonson TS, Yang Y, Huff CD, Yun H, Qin G, Witherspoon DJ, et al. Genetic evidence for high-altitude adaptation in Tibet. Science (New York, NY). 2010; 329(5987):72–5.
- 11. Beall CM, Cavalleri GL, Deng L, Elston RC, Gao Y, Knight J, et al. Natural selection on EPAS1 (HIF2alpha) associated with low hemoglobin concentration in Tibetan highlanders. Proceedings of the National Academy of Sciences of the United States of America. 2010; 107(25):11459–64. pmid:20534544
- 12. Yi X, Liang Y, Huerta-Sanchez E, Jin X, Cuo ZX, Pool JE, et al. Sequencing of 50 human exomes reveals adaptation to high altitude. Science (New York, NY). 2010; 329(5987):75–8.
- 13. Bigham A, Bauchet M, Pinto D, Mao X, Akey JM, Mei R, et al. Identifying signatures of natural selection in Tibetan and Andean populations using dense genome scan data. PLoS genetics. 2010; 6(9):e1001116. pmid:20838600
- 14. Xu S, Li S, Yang Y, Tan J, Lou H, Jin W, et al. A genome-wide search for signals of high-altitude adaptation in Tibetans. Molecular biology and evolution. 2011; 28(2):1003–11. pmid:20961960
- 15. Peng Y, Yang Z, Zhang H, Cui C, Qi X, Luo X, et al. Genetic variations in Tibetan populations and high-altitude adaptation at the Himalayas. Molecular biology and evolution. 2011; 28(2):1075–81. pmid:21030426
- 16. Qiu Q, Zhang G, Ma T, Qian W, Wang J, Ye Z, et al. The yak genome and adaptation to life at high altitude. Nature genetics. 2012; 44(8):946–9. pmid:22751099
- 17. Qiu Q, Wang L, Wang K, Yang Y, Ma T, Wang Z, et al. Yak whole-genome resequencing reveals domestication signatures and prehistoric population expansions. Nature communications. 2015; 6:10283. pmid:26691338
- 18. Ge RL, Cai Q, Shen YY, San A, Ma L, Zhang Y, et al. Draft genome sequence of the Tibetan antelope. Nature communications. 2013; 4:1858. pmid:23673643
- 19. Gou X, Wang Z, Li N, Qiu F, Xu Z, Yan D, et al. Whole-genome sequencing of six dog breeds from continuous altitudes reveals adaptation to high-altitude hypoxia. Genome research. 2014; 24(8):1308–15. pmid:24721644
- 20. Ai H, Yang B, Li J, Xie X, Chen H, Ren J. Population history and genomic signatures for high-altitude adaptation in Tibetan pigs. BMC genomics. 2014; 15:834. pmid:25270331
- 21. Zhang W, Fan Z, Han E, Hou R, Zhang L, Galaverni M, et al. Hypoxia adaptations in the grey wolf (Canis lupus chanco) from Qinghai-Tibet Plateau. PLoS genetics. 2014; 10(7):e1004466. pmid:25078401
- 22. Zhang Q, Gou W, Wang X, Zhang Y, Ma J, Zhang H, et al. Genome Resequencing Identifies Unique Adaptations of Tibetan Chickens to Hypoxia and High-Dose Ultraviolet Radiation in High-Altitude Environments. Genome biology and evolution. 2016; 8(3):765–76. pmid:26907498
- 23. Song S, Yao N, Yang M, Liu X, Dong K, Zhao Q, et al. Exome sequencing reveals genetic differentiation due to high-altitude adaptation in the Tibetan cashmere goat (Capra hircus). BMC genomics. 2016; 17:122. pmid:26892324
- 24. Wei C, Wang H, Liu G, Zhao F, Kijas JW, Ma Y, et al. Genome-wide analysis reveals adaptation to high altitudes in Tibetan sheep. Scientific reports. 2016; 6:26770. pmid:27230812
- 25. Stepanek J, Klocke D, Malvin G, Parisi J, Tazelaar H. The piglet as an animal model for hypobaric hypoxia. Wilderness & environmental medicine. 1998; 9(1):8–13.
- 26.
Koesterke L, Milfeld K, Vaughn MW, Stanzione D, Koltes JE, Weeks NT, et al., editors. Optimizing the PCIT algorithm on stampede's Xeon and Xeon Phi processors for faster discovery of biological networks. Proceedings of the Conference on Extreme Science and Engineering Discovery Environment: Gateway to Discovery; 2013: ACM.
- 27. Reverter A, Chan EK. Combining partial correlation and an information theory approach to the reversed engineering of gene co-expression networks. Bioinformatics (Oxford, England). 2008; 24(21):2491–7.
- 28. Hudson NJ, Reverter A, Dalrymple BP. A differential wiring analysis of expression data correctly identifies the gene containing the causal mutation. PLoS computational biology. 2009; 5(5):e1000382. pmid:19412532
- 29. Reverter A, Hudson NJ, Nagaraj SH, Perez-Enciso M, Dalrymple BP. Regulatory impact factors: unraveling the transcriptional regulation of complex traits from expression data. Bioinformatics (Oxford, England). 2010; 26(7):896–904.
- 30. Pavlidis P. Using ANOVA for gene selection from microarray studies of the nervous system. Methods (San Diego, Calif). 2003; 31(4):282–9.
- 31. Wu C, Orozco C, Boyer J, Leglise M, Goodale J, Batalov S, et al. BioGPS: an extensible and customizable portal for querying and organizing gene annotation resources. Genome biology. 2009; 10(11):R130. pmid:19919682
- 32. Mi H, Muruganujan A, Casagrande JT, Thomas PD. Large-scale gene function analysis with the PANTHER classification system. Nat Protoc. 2013; 8(8):1551–66. pmid:23868073
- 33. Vaquerizas JM, Kummerfeld SK, Teichmann SA, Luscombe NM. A census of human transcription factors: function, expression and evolution. Nature reviews Genetics. 2009; 10(4):252–63. pmid:19274049
- 34. Bastian FB, Chibucos MC, Gaudet P, Giglio M, Holliday GL, Huang H, et al. The Confidence Information Ontology: a step towards a standard for asserting confidence in annotations. Database: the journal of biological databases and curation. 2015; 2015:bav043. pmid:25957950
- 35. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome research. 2003; 13(11):2498–504. pmid:14597658
- 36. Singh RS. Darwin's legacy II: why biology is not physics, or why it has taken a century to see the dependence of genes on the environment. Genome / National Research Council Canada = Genome / Conseil national de recherches Canada. 2015; 58(1):55–62.
- 37. Cheviron ZA, Brumfield RT. Genomic insights into adaptation to high-altitude environments. Heredity. 2012; 108(4):354–61. pmid:21934702
- 38.
Roach RC, Wagner PD, Hackett PH. Hypoxia: from genes to the bedside: Springer Science & Business Media; 2013.
- 39. Hudson NJ, Reverter A, Wang Y, Greenwood PL, Dalrymple BP. Inferring the transcriptional landscape of bovine skeletal muscle by integrating co-expression networks. PloS one. 2009; 4(10):e7249. pmid:19794913
- 40. Petousi N, Robbins PA. Human adaptation to the hypoxia of high altitude: the Tibetan paradigm from the pregenomic to the postgenomic era. Journal of applied physiology (Bethesda, Md: 1985). 2014; 116(7):875–84.
- 41. Bigham AW, Lee FS. Human high-altitude adaptation: forward genetics meets the HIF pathway. Genes & development. 2014; 28(20):2189–204.
- 42. Beall CM. Andean, Tibetan, and Ethiopian patterns of adaptation to high-altitude hypoxia. Integrative and comparative biology. 2006; 46(1):18–24. pmid:21672719
- 43. Lorenzo FR, Huff C, Myllymaki M, Olenchock B, Swierczek S, Tashi T, et al. A genetic mechanism for Tibetan high-altitude adaptation. Nature genetics. 2014; 46(9):951–6. pmid:25129147
- 44. Shifera AS. The zinc finger domain of IKKgamma (NEMO) protein in health and disease. Journal of cellular and molecular medicine. 2010; 14(10):2404–14. pmid:20345847
- 45. Bracken CP, Whitelaw ML, Peet DJ. Activity of hypoxia-inducible factor 2alpha is regulated by association with the NF-kappaB essential modulator. The Journal of biological chemistry. 2005; 280(14):14240–51. pmid:15653678
- 46. Wang GL, Semenza GL. Purification and characterization of hypoxia-inducible factor 1. The Journal of biological chemistry. 1995; 270(3):1230–7. pmid:7836384
- 47. Forsythe JA, Jiang BH, Iyer NV, Agani F, Leung SW, Koos RD, et al. Activation of vascular endothelial growth factor gene transcription by hypoxia-inducible factor 1. Molecular and cellular biology. 1996; 16(9):4604–13. pmid:8756616
- 48. Wiesener MS, Turley H, Allen WE, Willam C, Eckardt KU, Talks KL, et al. Induction of endothelial PAS domain protein-1 by hypoxia: characterization and comparison with hypoxia-inducible factor-1alpha. Blood. 1998; 92(7):2260–8. pmid:9746763
- 49. Clerici C, Planes C. Gene regulation in the adaptive process to hypoxia in lung epithelial cells. American journal of physiology Lung cellular and molecular physiology. 2009; 296(3):L267–74. pmid:19118091
- 50. Blanchon L, Bocco JL, Gallot D, Gachon AM, Lemery D, Dechelotte P, et al. Co-localization of KLF6 and KLF4 with pregnancy-specific glycoproteins during human placenta development. Mechanisms of development. 2001; 105(1–2):185–9. pmid:11429296
- 51. Warke VG, Nambiar MP, Krishnan S, Tenbrock K, Geller DA, Koritschoner NP, et al. Transcriptional activation of the human inducible nitric-oxide synthase promoter by Kruppel-like factor 6. The Journal of biological chemistry. 2003; 278(17):14812–9. pmid:12590140
- 52. Zhang F, Wu W, Deng Z, Zheng X, Zhang J, Deng S, et al. High altitude increases the expression of hypoxia-inducible factor 1alpha and inducible nitric oxide synthase with intest-inal mucosal barrier failure in rats. International journal of clinical and experimental pathology. 2015; 8(5):5189–95. pmid:26191216
- 53. Wang L, Wang CM, Hou LH, Dou GR, Wang YC, Hu XB, et al. Disruption of the transcription factor recombination signal-binding protein-Jkappa (RBP-J) leads to veno-occlusive disease and interfered liver regeneration in mice. Hepatology (Baltimore, Md). 2009; 49(1):268–77.
- 54. Borggrefe T, Lauth M, Zwijsen A, Huylebroeck D, Oswald F, Giaimo BD. The Notch intracellular domain integrates signals from Wnt, Hedgehog, TGFbeta/BMP and hypoxia pathways. Biochimica et biophysica acta. 2016; 1863(2):303–13. pmid:26592459
- 55. Dolatshad H, Pellagatti A, Fernandez-Mercado M, Yip BH, Malcovati L, Attwood M, et al. Disruption of SF3B1 results in deregulated expression and splicing of key genes and pathways in myelodysplastic syndrome hematopoietic stem and progenitor cells. Leukemia. 2015; 29(8):1798. pmid:26242354
- 56. Mirtschink P, Krek W. Hypoxia-driven glycolytic and fructolytic metabolic programs: Pivotal to hypertrophic heart disease. Biochimica et biophysica acta. 2016; 1863(7 Pt B):1822–8.
- 57. Gallagher WM, Currid CA, Whelan LC. Fibulins and cancer: friend or foe? Trends in molecular medicine. 2005; 11(7):336–40. pmid:15961345
- 58. Chen J, Wei D, Zhao Y, Liu X, Zhang J. Overexpression of EFEMP1 correlates with tumor progression and poor prognosis in human ovarian carcinoma. PloS one. 2013; 8(11):e78783. pmid:24236050
- 59. Magli MC, Largman C, Lawrence HJ. Effects of HOX homeobox genes in blood cell differentiation. Journal of cellular physiology. 1997; 173(2):168–77. pmid:9365517
- 60. Zimmermann F, Rich IN. Mammalian homeobox B6 expression can be correlated with erythropoietin production sites and erythropoiesis during development, but not with hematopoietic or nonhematopoietic stem cell populations. Blood. 1997; 89(8):2723–35. pmid:9108390
- 61. Streeter RG, Pepelko WE, Cain SM. Tissue oxygenation and splenic erythropoiesis during chronic hypoxia and hypercapnia. Journal of applied physiology. 1975; 38(2):309–14. pmid:235506
- 62. Sit B. The Mammalian Unfolded Protein Response: A Potential Source of Novel Therapeutic Targets in the Hypoxic Tumour Fraction. University of Toronto Journal of Undergraduate Life Sciences. 2013; 7(1).
- 63. Liu L, Qi X, Chen Z, Shaw L, Cai J, Smith LH, et al. Targeting the IRE1alpha/XBP1 and ATF6 arms of the unfolded protein response enhances VEGF blockade to prevent retinal and choroidal neovascularization. The American journal of pathology. 2013; 182(4):1412–24. pmid:23395094
- 64. Perez-Perri JI, Acevedo JM, Wappner P. Epigenetics: new questions on the response to hypoxia. International journal of molecular sciences. 2011; 12(7):4705–21. pmid:21845106
- 65. Johnson AB, Denko N, Barton MC. Hypoxia induces a novel signature of chromatin modifications and global repression of transcription. Mutation research. 2008; 640(1–2):174–9. pmid:18294659
- 66. Wu MZ, Tsai YP, Yang MH, Huang CH, Chang SY, Chang CC, et al. Interplay between HDAC3 and WDR5 is essential for hypoxia-induced epithelial-mesenchymal transition. Molecular cell. 2011; 43(5):811–22. pmid:21884981
- 67. Watson JA, Watson CJ, McCann A, Baugh J. Epigenetics, the epicenter of the hypoxic response. Epigenetics. 2010; 5(4):293–6. pmid:20418669
- 68. Yu YH, Chiou GY, Huang PI, Lo WL, Wang CY, Lu KH, et al. Network biology of tumor stem-like cells identified a regulatory role of CBX5 in lung cancer. Scientific reports. 2012; 2:584. pmid:22900142