Mouse MC3T3-E1 pre-osteoblast cells were obtained from the Chinese Academy of Sciences Cell Bank (Shanghai, China) and cultured in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS) (both from Hyclone; GE Healthcare Life Sciences, Logan, UT, USA), 100 U/ml penicillin G and 100 mg/ml streptomycin, at 37°C in a humidified atmosphere of 5% CO₂ in air. To induce osteoblast differentiation, the DMEM was supplemented with 10% FBS, 0.1 mM dexamethasone, 10 mM β-glycerophosphate, and 50 µg/ml ascorbic acid, and was changed every 2–3 days.

Total mRNA was extracted using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA), according to the manufacturer's instructions. The RNA concentration was quantified using a NanoDrop 1000 Spectrophotometer (Thermo Fisher Scientific, Inc.). RNA quality was tested using an Agilent 2100 bioanalyzer (Agilent Technologies, Inc., Santa Clara, CA, USA) and measured using the RNA integrity number (RIN). RNA samples were submitted for further analysis if the RIN score was >5.0.

For RT, first-strand cDNA was synthesized using the PrimeScript^® RT reagent kit (cat. no. DRR037; Takara Bio, Inc., Otsu, Japan). The expression levels of target genes were determined quantitatively using a CFX96 qPCR system (Bio-Rad Laboratories, Inc., Hercules, CA, USA) using SYBR^® Premix Ex Taq^™ II (cat. no. DRR820A; Takara Bio, Inc.), according to the manufacturer's instructions. The primer pairs were as follows: Runt-related transcription factor 2 (Runx2; GenBank accession no. NM_053470) forward, 5′-CCATAACGGTCTTCACAAATCC-3′ and reverse, 5′-GCGGGACACCTACTCTCATACT-3′; transcription factor Sp7 (Osx; accession no. NM_001037632) forward, 5′-CAGTAATCTTCGTGCCAGACC-3′ and reverse, 5′-CTTCTTTGTGCCTCCTTTTCC-3′; polypyrimidine tract-binding protein 2 (Ptbp2; accession no. NM_001310711) forward, 5′-AAAGTCGCTCTGAGTTGTTAT-3′ and reverse, 5′-GCGAAGAGTTTGTCCTCAACC-3′; transportin-1 (Tnpo1; accession no. NM_001048267) forward, 5′-AATTCGCGGTGACTCAGTCTGG-3′ and reverse, 5′-TCCATCTTGGTTTGCGAGGC-3′; exostosin-1 (Ext1; accession no. NM_010162) forward, 5′-GAAGAGCACAGTGGTCGGAA-3′ and reverse, 5′-CTCGATGGCCGCTAGAATGT-3′; NONMMUT044983 (NONCODE gene ID NONMMUG027774.1) forward, 5′-CGGCAGGCCTAGTCTTGTAT-3′ and reverse, 5′-ACAGCAGAGAGAGCCAAGGA-3′; NONMMUT023474 (gene ID NONMMUG014520.1) forward, 5′-TCTCGAACCCTAGGAGAGCA-3′ and reverse, 5′-GGGACAAGGTAAATGGCTCA-3′; NONMMUT018832 (gene ID NONMMUG011720.1) forward, 5′-AACATCTGAGGCTTGGCACT-3′ and reverse, 5′-TCATGGTACTGGCATCTCCA-3′; and GAPDH (accession no. NM_008084) forward, 5′-CAGTGCCAGCCTCGTCTCAT-3′ and reverse, 5′-AGGGGCCATCCACAGTCTTC-3′. GAPDH was used as an internal control. The qPCR thermal cycling was performed as follows: Initial incubation for 15 sec at 95°C, followed by denaturing for 40 cycles at 95°C for 5 sec and annealing for 31 sec at 60°C. The relative expression of target genes was calculated using the 2^−ΔΔCq method (15).

Cells were seeded at 1×10⁶ cells/well in 6-well plates (Corning Inc., Corning, NY, USA) and cultured for 24 h. When ALP activity was determined, confluent cell layers were washed with PBS, lysed with 0.1 mol/l M-PER mammalian protein extraction reagent (Pierce; Thermo Fisher Scientific, Inc.) for 15–30 min, and centrifuged at 12,000 × g for 15 min at room temperature, according to the manufacturer's instructions. The supernatants were collected for determining ALP activity using an ALP assay kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). Protein concentrations were measured using a Bicinchoninic Acid Protein Assay kit (Pierce; Thermo Fisher Scientific, Inc.). ALP activity (IU/l) was defined as the release of 1 nmol p-nitrophenol/min/µg total cellular protein.

The Rotating Wall Vessel Bioreactor (RWVB) clinostat is an effective, ground-based device which is used to simulate microgravity (7). Microgravity is achieved by keeping cells rotating uniformly around a horizontal axis. Therefore, there is a vector-averaged reduction in the apparent gravity acting on the cells when rotated by 360°. A 2D-RWVB (China Astronaut Research and Training Center, Beijing, China) was used in the present study, as previously described (16). MC3T3-E1 cells were seeded on coverslips and incubated until cell confluence reached 70%. The coverslips were subsequently fixed in the bioreactor and placed 12.5 mm away from the rotational axis. The bioreactor was completely filled with culture medium. Gentle aspiration was performed to clear away air bubbles in order to avoid shear stress during rotation. The bioreactor was fixed onto the clinostat and rotated around a horizontal axis at 24 rpm. The group rotating around a vertical axis was regarded as the control. The entire system was placed in a humidified incubator at 37°C under 5% CO₂.

An Affymetrix GeneChip^® Mouse Transcriptome assay (version 1.0; Affymetrix Inc., Santa Clara, CA, USA) provided global profiling of transcripts in the mouse genome, which were selected from the most authoritative databases, including NONCODE (www.noncode.org/index.php) and ENSEMBL (www.ensembl.org/index.html). In the microarray, every transcript was detected with 10 specific probes in order to improve the confidence of the statistical results. Protein-coding and non-coding genes were represented on a separate array to supply coincident hybridization. RNA labeling and array hybridization were performed according to the manufacturer's protocol. The microarray work was performed by Shanghai Ming Information Technology Co., Ltd. (Shanghai, China).

GO analysis was applied to analyze the primary function of the differentially-expressed genes according to the principles of Gene Ontology, which is able to organize genes into hierarchical categories and reveal the gene regulatory network on the basis of biological process and molecular function (17). Pathway analysis was additionally performed to examine the significant pathways of the differentially-expressed genes according to the Kyoto Encyclopedia of Genes and Genomes, Biocarta and Reatome databases (18). An online Bioinformatics enrichment tool (DAVID; david.ncifcrf.gov) was used to perform the GO and pathway analysis in the present study (19). Biological processes of GO terms were illustrated in the GO analysis. The P-value indicated the significance of GO term and pathway term enrichment in the differentially-expressed mRNA list (P<0.05 was considered to be statistically significant).

A co-expression regulatory network is an undirected graph, where each node corresponds to coding or ncRNAs, and a pair of nodes is connected by an edge if there is a significant co-expression association between them (20). lncRNA-mRNA networks were built to identify the interactions between the differentially-expressed genes and lncRNAs, according to the normalized signal intensity of specific expression in the gene and lncRNA. For lncRNA-gene pair, the Pearson correlation was calculated and the significant correlation pairs with which to construct the network were selected (21). In network analysis, degree centrality, which is defined as the number of connections between one node and another, is the simplest and most important measure of the centrality of a gene or lncRNA within a network which determines its relative importance (22).

Experiments were repeated a minimum of three times. Data were statistically analyzed using SPSS 19.0 software and expressed as the mean ± standard deviation. A one-way repeated measures analysis of variance followed by Dunnett's post hoc test was used to compare the time course-dependent variables. Two group comparisons were performed using a two-tailed t-test. P<0.05 was considered to indicate a statistically significant difference.

MC3T3-E1 cells were cultured in an RWVB clinostat and tested following rotation for 72 h. The effect of simulated microgravity on osteoblasts was assessed by observing the gene expression of Runx2 and Osx, in addition to the activity of ALP, which were considered to be important transcription factors for osteoblast differentiation. The data demonstrated that the gene expression of Runx2 and Osx and the protein activity of ALP were significantly decreased, of which, Osx and ALP reached lowest levels at 48 h and Runx2 decreased slightly further at 72 h (Fig. 1A). The results suggested that osteo-differentiation process of MC3T3-E1 cells was inhibited by simulated microgravity.

In order to detect the differentially-expressed lncRNAs and mRNAs in MC3T3-E1 when exposed to simulated microgravity, a transcriptome assay was performed to compare expression profiles between the normal gravity and simulated microgravity groups. The expression levels of 1,481 ncRNAs were observed to be significantly altered, which is presented in the hierarchical clustering analysis heat map (Fig. 1B). Among them, 857 lncRNAs whose length exceeded 200 nucleotides, including 168 upregulated and 689 downregulated, were screened out and are exhibited on a scatter plot (Fig. 1C). In addition, 2,264 mRNAs were demonstrated to be differentially-expressed, including 459 upregulated mRNAs and 1,805 downregulated mRNAs, presented on a clustering heat map (Fig. 1D).

Locational distributions of the aberrantly-expressed lncRNAs and mRNAs were analyzed synchronously. Significantly altered lncRNAs and mRNAs were spread across the chromosomes, with the largest amounts on chromosomes 1, 2 and 11. A relatively consistent variation pattern was observed between aberrant lncRNA and mRNA species, except for on chromosome 7 (Fig. 2A). In order to elucidate the detailed association, a large and complex lncRNA-mRNA co-expression network was generated according to the normalized signal intensity of differentially-expressed lncRNAs and mRNAs. This co-expression network consisted of 916 nodes and 4,813 connections between 354 lncRNAs and 562 mRNAs. Within this network, there were 3,798 pairs presenting as positive regulatory associations and 1,015 pairs as negative associations. Subsequently, mRNAs in the network were submitted for GO enrichment analysis. The results demonstrated that the aberrant mRNAs were frequently enriched in such biological process as phosphorylation, response to DNA damage, the cell cycle, regulation of gene expression and cell proliferation (Fig. 2B), and were involved in the cell signaling pathways of the cell cycle, focal adhesion, cancer and viral infection (Fig. 2C).

In order to further elicit information from the co-expression network and examine the possible functional lncRNAs, the nearby mRNAs located within 100 kb up or downstream of aberrantly-expressed lncRNAs were identified and matched with the aberrantly-expressed mRNAs in the microarray data. Among them, 132 differentially-expressed lncRNAs and nearby mRNAs pairs were identified (data not shown). GO analysis indicated that the nearby genes were involved in the significant biological processes, including the cell cycle, nervous system development, response to growth factors, regulation of RNA splicing, circulatory system development, enzyme-linked receptor protein signaling pathway and positive regulation of fibroblast proliferation (Fig. 3A). Pathway analysis demonstrated that the most significant pathways were chronic myeloid leukemia, protein digestion and absorption, and control of gene expression by vitamin D receptor (Fig. 3B). A literature review further demonstrated that ≥15 mRNAs included in 17 lncRNA-mRNA pairs have been reported to be involved in the regulation of osteoblast proliferation, differentiation and mineralization (Table I).

In the co-expression network, degree centrality was considered to be the simplest and most important measure by which to estimate the core regulatory factors. In order to obtain more accurate candidate lncRNAs, the degree centrality of the 17 lncRNAs whose nearby genes had been reported to be associated with osteoblast function was calculated. The lncRNAs NONMMUT044983, NONMMUT023474 and NONMMUT018832 were screened out via the degree centrality analysis (degree >15). Their associations with deregulated mRNAs were obtained from the overall co-expression network and rebuilt (Fig. 4A). The locations of the lncRNAs compared with mRNAs on the chromosome are exhibited in Fig. 4B. In addition, RT-qPCR analysis was performed to determine the expression of the three lncRNAs and their nearby genes in MC3T3-E1 cells exposed to the RWVB clinostat for 48 h. The results of the present study demonstrated that the expression of each lncRNA and mRNA was significantly decreased under simulated microgravity (Fig. 4C), consistent with the microarray data.