To cope with harsh environments, the Artemia shrimp produces gastrula embryos in diapause, a state of obligate dormancy, having cellular quiescence and suppressed metabolism. The mechanism behind these cellular events remains largely unknown. Here, we study the regulation of cell quiescence using diapause embryos of Artemia. We found that Artemia DEK (Ar-DEK), a nuclear factor protein, was down-regulated in the quiescent cells of diapause embryos and enriched in the activated cells of post-diapause embryos. Knockdown of Ar-DEK induced the production of diapause embryos whereas the control Artemia released free-swimming nuaplii. Our results indicate that Ar-DEK correlated with the termination of cellular quiescence via the increase in euchromatin and decrease in heterochromatin. The phenomena of quiescence have many implications beyond shrimp ecology. In cancer cells, for example, knockdown of DEK also induced a short period of cellular quiescence and increased resistance to environmental stress in MCF-7 and MKN45 cancer cell lines. Analysis of RNA sequences in Artemia and in MCF-7 revealed that the Wnt and AURKA signaling pathways were all down-regulated and the p53 signaling pathway was up-regulated upon inhibition of DEK expression. Our results provide insight into the functions of Ar-DEK in the activation of cellular quiescence during diapause formation in Artemia.
Diapause is a period in an organism's life cycle when growth, development and physical activity are slowed or even reversibly halted. The initiation and termination of such a state seem often to be triggered by specific environmental conditions . Cellular quiescence is a mechanism where some species can produce embryos in a state of diapause or dormancy, which often enables the offspring to survive remarkably harsh environmental conditions [2,3]. Development, reproduction and metabolic activities are highly suppressed in this state, meaning that embryos are protected until environmental conditions become favorable, at which point the cell cycle resumes [4–6]. The balance between quiescence and activation is crucial for tissue homeostasis, repair and regeneration and plays a key role in the growth, development and health in higher multicellular organisms [7,8]. In general, quiescent cells possess a transcriptional profile different from activated cells. This is achieved by the down-regulation of genes relating to proliferation and metabolism, and the up-regulation of genes not only related to cell cycle inhibition but also those that give the cells new properties [9,10].
DEK is highly conserved, being present in almost all higher eukaryotes, but is not found in lower eukaryotes such as yeast or C. elegans [11,12]. DEK has two functional DNA binding domains, one located in the center of the protein corresponding to the SAP/SAF box, and the other in the carboxyterminal region of DEK, overlapping a di- or multimerization domain [13,14]. Previous studies on DEK have suggested that DEK binds chromatin and functions as a genome organizer which is related to multiple critical cellular processes including DNA replication, RNA transcriptional regulation and mRNA splicing. Elevated DEK levels have been shown to promote proliferation, motility, invasion and tumorigenesis [15–18]. In melanoma cell lines, short hairpin RNA (shRNA)-mediated DEK depletion resulted in cell cycle arrest and enhanced cellular senescence, as well as increased doxorubicin (Dox)-induced cellular apoptosis . Elevated DEK levels in breast cancer cell lines have been reported to correlate with disease recurrence and metastasis [20,21].
Epigenetic studies have shown that the chromatin structure is involved in maintaining the reversibility of quiescence and that histone methylation contributes to the control of gene expression associated with quiescence regulation [22,23]. A study of the activation and differentiation of skeletal muscle stem cells has shown the important role of epigenetic processes controlling heterochromatin formation in the regulation of muscle stem cell quiescence . Heterochromatin exists in two varieties, constitutive heterochromatin and facultative heterochromatin. They act to silence the expression of related genes by their highly condensed structure. Heterochromatin is characterized by distinct epigenetic signatures, including high levels of trimethylation of H3K9 (H3K9me3) and H3K27 (H3K27me3) for facultative heterochromatin, and of H4K20 (H4K20me3) for constitutive heterochromatin. This stands in contrast with euchromatin, which has a high level of acetylation of H3K9 (H3K9ac) [25–28]. Analyses of histone modifications have revealed that H4K20me3 increases during cell quiescence, displaying tight chromatin compaction [29,30].
Artemia, also known as the brine shrimp, is found in severely hypersaline environments such as salt lakes, which are among the most hostile environments on earth [31,32]. In order to cope with environmental stresses under unfavorable conditions, mature females produce and release encysted embryos that enter diapause, a state of obligate dormancy. Under more favorable conditions, they release free-swimming nauplius larvae . For these cyst-encased diapause embryos, if suitable environmental conditions are resumed, their diapause is terminated and they hatch and continue to develop into nauplii and on to complete their normal life cycle. In our previous Artemia study, we found that an La-related protein controls cell cycle arrest by the nuclear retrograde transport of tRNAs during diapause formation . We also reported that SETD4, a member of SETD family, regulates embryonic cell quiescence and catalyzes H4K20me3 during diapause formation in Artemia . However, the molecular mechanism of the termination of diapause has not yet been reported.
In this Artemia study, we found that Ar-DEK was down-regulated in diapause embryos that existed under cellular quiescence but was expressed abundantly in the active cellular state of post-diapause embryos. Meanwhile, H4K20me3, a marker of constitutive heterochromatin was enriched in diapause, but subsequently reduced in post-diapause embryos. Knockdown of Ar-DEK in adult females induced the production of diapause embryos, even when kept in optimal environmental conditions, whereas the control Artemia continued to produce free-swimming nuaplii. These results indicate that Ar-DEK correlated with the termination of diapause in Artemia by reducing constitutive heterochromatin. Analysis of RNA sequences in Artemia and MCF-7 cells revealed that the Wnt and AURKA signaling pathways were all down-regulated and the p53 signaling pathway was up-regulated upon the inhibition of DEK expression. Our results provide insights into the functions of DEK in Artemia that can also be used to characterize the function of cellular quiescence regulation in other contexts.
Materials and methods
Parthenogenetic Artemia were collected from Gahai Lake, China. Ovoviviparous and oviparous Artemia were classified by the morphology of Artemia shell glands. For oviparous Artemia, swimming nauplii were reared in 8% (wt/vol) artificial seawater (Blue Starfish, Zhejiang, China) under a 4 h light/20 h dark cycle to release diapause cysts. For ovoviviparous Artemia, the swimming nauplii were reared in 4% (wt/vol) artificial seawater under a 16 h light/8 h dark cycle to produce free-swimming nauplius larvae directly. The water was maintained at 25°C and was supplemented with Chlorella powder (Fuqing King Dnarmsa Spirulina Co. Ltd.) as brine shrimp food every 2 days. For both reproductive pathways, early oocytes are formed in the paired ovaries and mature in the oviducts to form late oocytes. These pass into the ovisac (uterus), where they become early embryos. The late embryo stage then gives rise to either nauplius larvae or diapause cysts, which are then released into the environment.
Real-time quantitative PCR
Total RNA from pre-diapause embryos, diapause embryos, post-diapause embryos (12 h after activation of diapause embryos at −30°C) and nauplii were isolated using TRIzol reagent and treated with RNase-free DNase I (New England Biolabs, MA, U.S.A.) to remove any contaminating genomic DNA. Specific primers of Ar-DEK, AURKA and internal control actin (RT-Ar-DEK F [GCACAATCTTCAAAACAAAAA] and RT-Ar-DEK R [CGTTACACAAACAGGTACATACA] for Ar-DEK; AURKA F [TCGCAGTGATAGACCTGAGCC] and AURKA R [TTTGCCAGCCAGTATTTCGTT] for AURKA; actin F [GCCCTTGACTTCGAACAGGA] and actin R [GGAAGGTGGACAGAGAAGCC] for actin) were designed according to cDNA. Real-time PCR reactions were performed using the Bio-Rad MiniOpticon real-time PCR system with SYBR Premix Ex Taq (TaKaRa Bio, Shiga, Japan). All data are given as means ± SD of independent experiments from three separate RNA pools. Statistical analyses were performed using one-way ANOVA followed by Tukey's all pair comparison test using the STATISTICA software, and differences were considered signiﬁcant at P < 0.01.
Cell culture and treatment of doxorubicin and palmitic acid
Breast cancer (MCF-7) cells and Human gastric cancer cells (MKN45) cells were cultured at 37°C in 5% CO2 in EMEM (Genom, Hangzhou, China) and RPMI-1640 (Corning, New York, U.S.A.) supplemented with 10% fetal bovine serum (FBS) (Gibco), 100 IU/ml penicillin and 100 µg/ml streptomycin (Sigma–Aldrich). Cells were cultured in a 24-well plate (1 × 104 cells/well) in 10% FBS in EMEM or RPMI-1640 for 24 h and then was treated with 0.3 mM Dox and 0.3 mM palmitic acid (PA), respectively. In order to promote autophagosome generation within PA treatment, cells were treated with 10 µM chloroquine (CQ) for 4 h before collection.
Western blotting analysis
Proteins extracted from the tissues of Artemia of different developmental stages were prepared using TRIzol reagent (Invitrogen Life Technologies, Burlington, ON) according to the manufacturer's instructions. Total cell lysates were prepared from MCF-7 and MKN45 cells using a RIPA Lysis Buffer (Beyotime, China) and were heated in a boiling water bath for 15 min and then quantified using the Bradford method. Each total protein sample (4 µg) was separated on SDS–PAGE gel and then transferred to PVDF membranes (Bio-Rad, CA, U.S.A.). After incubation of the membranes with primary antibodies at 4°C overnight, specific proteins were detected using a BM Chemiluminescence Western Blotting Kit (Roche, Mannheim, Germany). The densitometry analysis for all data of Western blotting using ImageJ (National Institute of Health), as previously described .
Artemia samples were fixed with 4% paraformaldehyde and embedded with paraffin. Cell samples were fixed in 4% formaldehyde overnight and then washed with PBS. Tissue slides (5 µm thick) and cell samples were treated with 0.25% Triton X-100 for 10 min. The samples were blocked in an antiserum dilution buffer containing 1% BSA and 0.1% Triton X-100 for 1 h and then incubated with appropriate antibodies at 4°C overnight. The slides were incubated in a secondary Alexa Fluor 647-conjugated (Invitrogen) for 2 h at room temperature and the nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI) (Sangon Biotech, Shanghai, China). All confocal images were collected by using a Zeiss LSM 710 (Carl Zeiss, Zeiss, Germany) confocal microscope equipped with a 40×, 1.4-numerical-aperture (NA) objective lens.
siRNA synthesis and microinjection
For the preparation of the double-stranded RNA (dsRNA), a PET-T7 plasmid that contained two inverted T7 polymerase sites flanking the cloning region was used as the expression vector. To obtain the reconstructed plasmid expressing Ar-DEK dsRNA, two fragments in the coding region of the Ar-DEK gene were amplified with specific primers (ds-Ar-DEK-1 F [GCTCTAGAAGAGCACAATCTTCAAAACAA] and ds-Ar-DEK-1 R [CGGAATTCAAATAGCACAAATGGAAACCC], Xba1 and EcoR1 sites are underlined; ds-Ar-DEK-2 F [TGCTCTAGAGGAAGAGGCTGAGAAAACA] and ds-Ar-DEK-2 R [CGGGATCCGCTGCAGACTTAGGCGT], Xba1 and BamH1 sites are underlined). These amplified fragments were digested using XbaI, EcoRI and BamH1 (TaKaRa) and subcloned into PET-T7. The plasmid expressing the dsRNA targeting green fluorescent protein (GFP) was constructed as described previously and used as a negative control. The dsRNA was produced and purified as described previously by Yodmuang et al. . Approximately 1 µg of Ar-DEK or GFP-specific dsRNA was injected into the body cavity of Artemia using an Ultra-MicroPump II equipped with a Micro4TM MicroSyringe pump controller (World Precision Instruments, Micro4).
Lentiviral transduction of MCF-7 and MKN45
An shRNA targeting nucleotides 1165–1185 of DEK (shDEK 1) (GGATAGTTCAGATGATGAAC) and 860–880 of DEK (shDEK 2) (GTGATGAAGATGAAAAGAAA) were achieved by Hanbio Inc., Shanghai, China . MCF-7 and MKN45 were transduced with LV-RFP lentivirus (Hanbio Inc., Shanghai, China) as previously described. Briefly, cells were cultured in a 24-well plate (1 × 104 cells/well) in 10% FBS in EMEM or RPMI-1640 for 24 h. The medium was then replaced with 1 ml of fresh medium and 8 µg/ml polybrene. Then, 10 µl of lentivirus solution (108 IU/ml) was added to each well, followed by incubation at 37°C and 5% CO2 for 24 h. After incubation, the treatment medium was replaced with fresh medium containing 10% FBS, and the cells were cultured at 37°C and 5% CO2 until >50% confluence was achieved. Transduced cells were selected using 3 µg/ml puromycin and then used for in the following experiments or harvested for Western blot analysis.
Recombinant of histidine-tagged DEK
On the basis of the sequence of the human DEK gene in GenBank, the His-DEK-IRES-EGFP expression plasmid was synthesized with specific primers (Bac-DEK-IRES-EGFP F [CGGAATTCAAATGTCCGCCTCGGCCCCT] and Bac-DEK-IRES-EGFP R [GCTCTAGATTACTTGTACAGCTCGTC]; these amplified fragments were digested using EcoRI and XbaI (TaKaRa) and subcloned into a viral packaging plasmid (pFastBac™). Then recombinant packaging plasmid transforms into DH10Bac™ cells to obtain the recombinant Bacmid DNA, and then transfers the insect cells SF9, the viral supernatant of recombinant Baculovirus was collected 48 h later. After filtration through a 0.22 µm filter, the recombinant Baculovirus was used to infect SF9 to obtain the recombinant protein. His-tagged protein was purified using a His-tag Protein Purification Kit (Beyotime, China). The recombinant full-length histidine (His)-tagged DEK (His-DEK) was directly added into the cell culture medium of DEK knockdown (DEK-KD) cells. Ten micrograms of recombinant full-length His-DEK was added into the medium of DEK-KD cells (1 × 106), as previously described .
Reagents and antibodies
CQ and PA (Sigma), Dox (Sangon, Shanghai, China) and polybrene (Hanbio Inc., Shanghai, China) were used. The primary antibodies used in this study were as follows: anti-Ar-DEK, anti-p26, anti-Artemin, anti-Ar-SETD4 and anti-Ar-Larp were raised in rabbits (Hangzhou HuaAn Biotechnology Company, Hangzhou, China); anti-H3 (Cell Signaling, Boston, MA, U.S.A.); anti-p (Ser10) H3 (Epitomics), anti-Rb (Epitomics); anti-p (Thr356) Rb (Abcam, London, U.K.); and anti-tubulin-α (Sigma–Aldrich, Milwaukee, WI, U.S.A.), polyclonal anti-LC3B (Sigma), anti-H3K9me1 (catalog number ab9045; Abcam), anti-H3K9me2 (catalog number 1349-1; Epitomics), anti-H3K9me3 (catalog number ab1773; Abcam), anti-H3K27me1 (catalog number 7693s; Cell Signaling Technology), anti-H3K27me2 (catalog number 9728s; Cell Signaling Technology), anti-H3K27me3 (catalog number ab6174; Abcam), anti-H4K20me1 (catalog number sc-134221; Santa Cruz Biotechnology), anti-H4K20me2 (catalog number GTX630545; GeneTeX), anti-H4K20me3 (catalog number 5737s; Cell Signaling Technology), anti-Ki67 (catalog number ab16667; Abcam), and anti-PCNA (catalog number ab92552; Abcam), anti-γH2AX (catalog number ab81299; Abcam), anti-DEK (catalog number 66194-1-Ig; Proteintech), anti-His-tag (catalog number 66005-1-Ig; Proteintech), anti-p27 (catalog number 2552; Cell Signaling Technology), anti-active-caspase3 (catalog number ab32042; Abcam). Detection was performed using BM Chemiluminescence Western Blotting Kits (Roche, Basel, Switzerland) according to the manufacturer's instructions.
The mRNA-Seq experiments were performed by Novogene (Beijing, China). The mRNA-seq library was prepared for sequencing using standard Illumina protocols. Briefly, total RNA from MCF-7 infected with short hairpin RNA interference (RNAi)-expressing lentiviruses of DEK (shDEK) and control lentiviruses (shRFP) was isolated using TRIzol reagent and treated with RNase-free DNase I (New England Biolabs, MA, U.S.A.) to remove any contaminating genomic DNA. mRNA extraction was performed using Dynabeads oligo(dT) (Invitrogen). Double-stranded complementary DNAs were synthesized using Superscript II reverse transcriptase (Invitrogen) and random hexamer primers. The cDNAs were then fragmented by nebulization and the standard Illumina protocol was followed thereafter to create the mRNA-seq library.
For the data analysis, basecalls were performed using CASAVA. Reads were aligned to the genome using the split read aligner TopHat (v2.0.7) and Bowtie2, with default parameters. HTSeq was used for estimating their abundances. Differential expression analysis of two conditions/groups (two biological replicates per condition) was performed using the DESeq R package (1.18.0). Gene Ontology (GO) enrichment analysis of differentially expressed genes (DEGs) was implemented using the GOseq R package, in which gene length bias was corrected. GO terms with corrected P values of less than 0.05 were considered significantly enriched by DEGs.
The Artemia mRNA-Seq experiments were performed by Sangon (Shanghai, China). Post-diapause embryos for RNA-seq were collected over 3 months after activation of diapause embryos (−30°C). Total RNA from pre-diapause embryos, diapause embryos and post-diapause embryos was extracted as described above. RNA integrity was evaluated using a 1.0% agarose gel. Thereafter, the quality and quantity of RNA were assessed using a Nano Photometer® spectrophotometer (IMPLEN, CA, U.S.A.) and an Agilent 2100 Bioanalyzer (Agilent Technologies, CA, U.S.A.). Sequencing libraries were generated using VAHTSTM mRNA-seq V2 Library Prep Kit for Illumina® following the manufacturer's recommendations and index codes were added to attribute sequences to each sample. Paired-end sequencing of the library was performed on the HiSeq XTen sequencers (Illumina, San Diego, CA). FastQC (version 0.11.2) was used for evaluating the quality of sequenced data. Unigenes were blasted against NCBI Nr (NCBI non-redundant protein database), SwissProt, TrEMBL, CDD (Conserved Domain Database), Pfam and KOG (eukaryotic Orthologous Groups) databases (E-value < 1 × 10−5). At the same time, TransDecoder (version 3.0.1) was used to predict the CDS sequences of the un-aligned Unigenes.
Statistics and quantification
The sample size for each experiment is indicated in the figure legends. Data are presented as the mean ± SD. For analysis of two groups of data, a two-tail Student's t-test was used. For three or more groups, data were analyzed using a one-way ANOVA. P values less than 0.05 were considered significant and P values less than 0.01 were considered to be highly significant. The sequenced cDNA and the deduced amino acid (aa) sequence of Ar-DEK were analyzed by using EditSeq v5.00 (DNAStar). The homologies were analyzed by using BLAST (NCBI).
Molecular signatures of diapause embryos in Artemia
Two independent reproductive pathways are available for Artemia which enable adaptation to widely fluctuating environments. Under favorable conditions, mature Artemia females release free-swimming nauplius larvae via the ovoviviparous pathway. Alternatively, under unfavorable conditions, encysted embryos are released in a protective shell and in a state of diapause via the oviparous pathway (Figure 1A).
Expression of Ar-DEK at each developmental stage during diapause formation in Artemia.
To characterize the molecular signatures of quiescent cells in diapause embryos of Artemia, RNA sequencing of embryos in pre-diapause, diapause and post-diapause stages was performed. The DEGs within the embryos of these different stages are presented as a heatmap (Supplementary Figure S1A). In total, 9458 genes were differentially expressed between pre-diapause embryos and diapause embryos, and 6356 between diapause and post-diapause embryos (Supplementary Figure S1B). Among them, the expression of 3421 genes, including notable markers of diapause (p26 and Artemin) and of cell division (Ki67, PCNA), was markedly higher or lower in diapause embryos than in either pre-diapause or post-diapause embryos (Supplementary Figure S1C).
These results reveal the key molecular characteristics of quiescent cells in diapause embryos. GO enrichment analysis was performed on these genes that were differentially expressed between diapause and post-diapause embryos (Supplementary Figure S2). This revealed that the expression of genes associated with cellular activity, including proliferation and differentiation, autophagy and apoptosis, was suppressed in diapause embryos. Results of this analysis revealed highly inhibited metabolic activity during states of quiescence, in which the genes associated with oxidative phosphorylation, cholesterol and the xenobiotic metabolism were all down-regulated in diapause embryos. Differentiation target genes were more highly enriched in post-diapause than in diapause embryos. Similarly, apoptosis targets genes were more associated with post-diapause than diapause embryos.
Expression patterns of DEK highly correlate with diapause and termination of diapause in Artemia
Differentially expressed genes between embryos of different stages, as identified from the above RNA sequencing, revealed a particular gene of interest that encoded a DEK protein. The expression of Ar-DEK was enriched in ovoviviparous Artemia and in pre-diapause and post-diapause embryos but occurred only at very low levels in oviparous diapause embryos (Figure 1B). This expression pattern was also validated by the analysis of real-time quantitative PCR (RT-qPCR) and Western blot at each developmental stage. The mRNA level of Ar-DEK was six-fold lower in diapause embryos than that in pre-diapause embryos (Figure 1C). Similarly, the protein expression of Ar-DEK was six-fold lower in diapause embryos than in pre-diapause embryos, as revealed by analysis using ImageJ (Figure 1D). Furthermore, immunofluorescence analysis showed that Ar-DEK was barely detected in diapause embryos but was observed at the nucleus in pre-diapause embryos, post-diapause embryos and nauplii (Figure 1E). Thus, down-regulation of Ar-DEK may be important for cellular quiescence during diapause formation in Artemia.
To investigate the role of DEK during the diapause formation of Artemia, the cDNA of Ar-DEK was cloned from Artemia. The full-length of Ar-DEK was shown to be 1567 bp, with a 1092-bp open reading frame encoding a 363-aa protein (Supplementary Figure S3A). The deduced amino acid sequence of the protein was noted as similar to those of other DEK family members, and phylogenetic analysis suggested that Ar-DEK could be grouped with DEK in the Chordata Danio rerio (Supplementary Figure S3B). Ar-DEK is a basic protein (pI = 9.47), yet it harbors a copy of the SAP domain (aa 149–183), a putative nuclear localization sequence (aa 287–305) and it also has a high sequence identity with the DEK-C superfamily in the C-terminal region (aa 308–361) (Supplementary Figure S4A). These domains are found in the DEK proteins of many other species, ranging from plants to humans. Although the amino acid sequence of the DNA binding domain of Ar-DEK shows 27–36% sequence similarity across different species, the amino acids predicted to interact with DNA in Artemia remain entirely conserved (Supplementary Figure S4B). These results reveal that DEK is conserved both structurally and functionally over the course of evolution.
DEK terminates diapause by activation of quiescent cells in Artemia
A gene knockdown system using RNAi was used to study the involvement of Ar-DEK in the formation and maintenance of embryonic diapause. Based on full-length Ar-DEK, two dsRNA were designed and injected into ovoviviparous Artemia at early oocyte stage. After the injection of ovoviviparous Artemia adults with 1 µg of Ar-DEK dsRNA 1 or Ar-DEK dsRNA 2, the Ar-DEK mRNA level decreased to 30 and 35% of the level in control Artemia adults which had been injected with a GFP-specific dsRNA (GFP dsRNA), respectively (Figure 2A). The Ar-DEK protein expression level had also decreased significantly (Figure 2B). The results showed that an average of 70 and 65% of ovoviviparous Artemia adults produced diapause embryos after injection of Ar-DEK dsRNA 1 and Ar-DEK dsRNA 2, respectively. However, as the control, ovoviviparous Artemia produced 1% of diapause embryos after injection of GFP dsRNA (Figure 2C). Furthermore, the diapause-specific proteins p26 and Artemin were expressed abundantly in Ar-DEK knockdown embryos but were absent from control embryos (Figure 2D). These results indicate that the knockdown of Ar-DEK had enhanced the formation of Artemia diapause embryos.
Effect of knockdown Ar-DEK in ovoviviparous by using double-stranded RNA.
Our previous Artemia study showed that the cells of diapause embryos are in a quiescent state [34,35]. Here, immunofluorescence analysis revealed that the proliferation marker Ki67, which marks all phases of active cells, was completely absent from diapause embryos but was expressed in pre-diapause and post-diapause embryos and in hatched nauplii (Figure 2E). The expression pattern of PCNA, as another marker of active cells, was also expressed in exactly the same way as Ki67 (Figure 2E). In addition, the levels of H3S10ph and RbT356ph, proliferation markers for mitotically dividing cells during mitosis and the G1/S phase, were not detected in the diapause and post-diapause stages but were abundant in the pre-diapause and hatched nauplii (Figure 2F and Supplementary Figure S5). These results indicated that the cells in diapause embryos were in a quiescent state and were subsequently activated post-diapause, in which they were arrested at the G0/G1 phase.
The cell cycle state was then examined after knockdown of the Ar-DEK gene in ovoviviparous Artemia. Upon Ar-DEK knockdown, the expression level of Ki67 and PCNA were lower in the knockdown embryos than in the controls (Figure 2G) as were the levels of pH3S10 and pRbT356 (Figure 2D). The expression of diapause-related protein Ar-SETD4 and Ar-Larp, which were abundant in diapause embryos, were significantly increased after Ar-DEK knockdown (Figure 2D). These results indicated that DEK-KD caused cellular quiescence in the diapause embryos, and Ar-DEK correlated with the termination of diapause by repression of Ar-SETD4 and Ar-Larp.
Knockdown of DEK induced the cell quiescence via the increase in constitutive heterochromatin and the decrease in constitutive euchromatin
To characterize the chromatin structure of quiescent cells in the diapause embryos, we determined the chromatin structures by histone modifications in nuclei of their quiescent cells using a combination of different antibodies. Firstly, we found that the level of H3K9ac, a marker for euchromatin, was notably low whereas H4K20me3, a marker for constitutive heterochromatin, was considerably enriched in the cells of diapause embryos in contrast with those of pre-diapause and post-diapause (Figure 3A,B). However, we did not observe any major differences in the facultative heterochromatin markers of H3K9me3 and H3K27me3 and HP1α in the cells of pre-diapause, diapause and post-diapause (Supplementary Figures S6 and S8). These results indicated that the quiescent cells of diapause embryos contain high levels of constitutive heterochromatins but far lower levels of euchromatin.
Levels of methylation of H3 and H4 at various lysine residues during diapause formation.
To elucidate the function of Ar-DEK in the regulation of chromatin formation, we analyzed the changes of chromatin structures in ovoviviparous Artemia. We found that the level of H3K9ac had decreased and the level of H4K20me3 had increased after Ar-DEK knockdown, in contrast with the controls (Figure 3C,D). There was also no significant difference in the level of the facultative heterochromatin markers, H3K9me3 and H3K27me3, after the Ar-DEK knockdown (Supplementary Figure S7). These results indicated that the composition of euchromatin had decreased and constitutive heterochromatin had increased after Ar-DEK knockdown. Therefore, we suggest that DEK terminates diapause by the activation of quiescent cells via the facilitation of euchromatin formation and by the reduction in constitutive heterochromatin structures.
DEK regulated pathways in Artemia
To investigate the pathways downstream of DEK-terminated diapause that is active in activating cell quiescence, RNA sequencing of Ar-DEK knockdown in ovoviviparous Artemia was performed. The transcriptome profiles of three replicates of Ar-DEK knockdown and those of controls were compared. We performed a cluster analysis of DEGs in the entire transcriptome and observed the overall gene expression pattern in Ar-DEK knockdown Artemia. The gene expression pattern was markedly different between Ar-DEK knockdown and controls (Figure 4A). A total number of 23 970 genes had been differentially expressed (Figure 4B). The GO enrichment analysis of DEGs is shown in Figure 4C. Genes involved in the p53 signaling pathway and in the negative regulation of cell proliferation had been up-regulated in the cells of Ar-DEK knockdown Artemia, while genes involved in cell division, the MAPK and Wnt/β-catenin signaling pathways and many other metabolic pathways had been down-regulated. Our results demonstrate that Ar-DEK may regulate cellular quiescence in Artemia.
RNA sequencing analysis of Ar-DEK knockdown in ovoviviparous Artemia.
We previously reported that the PLK1–MEK–ERK–RSK1 signaling pathway regulates cellular quiescence in diapause embryos [38,39]. Here, transcriptome analysis showed that the expression of AURKA, the upstream kinase of PLK1–MEK–ERK–RSK1, was very low in diapause embryos. This indicated that the AURKA signaling pathway had been suppressed (Figure 4D). We performed the analysis of RT-qPCR to confirm the transcript levels of AURKA, as shown in Figure 4E, the results were consistent to those based on RNA sequencing. The expression pattern of this kinase pathway was confirmed by Western blot, and the phosphorylation of PLK1, MEK, ERK and RSK1 was highly reduced after Ar-DEK knockdown (Figure 4F). These results indicated that PLK1–MEK–ERK–RSK1 had occurred downstream of Ar-DEK regulation and in response to Ar-DEK during diapause formation in Artemia. We analysed the transcriptome and the results showed that the expression of Ar-SETD4 and Ar-Larp were up-regulated, indicating that Ar-DEK transcriptionally regulates Ar-SETD4 and Ar-Larp expression in Artemia.
Our study on the regulation of cellular quiescence during Artemia diapause formation is summarized in Figure 4G. Analysis of the transcriptomes of diapause embryos released from the Artemia without Ar-DEK knockdown (Supplementary Figures S1A and S2) and with Ar-DEK knockdown (and Figure 4A,C) showed that the expression of genes associated with cell activity, including cell proliferation and differentiation had been similarly suppressed after Ar-DEK knockdown. The results of this analysis revealed that Ar-DEK is a regulator occurring highly upstream in the process of diapause termination in Artemia and that the genes associated with the oxidative phosphorylation, cholesterol, the xenobiotic metabolism and apoptosis were downstream of Ar-DEK.
DEK induced increase in constitutive heterochromatin in response to stress in human cancer cell lines
To identify the function of DEK in the regulation of resistance to environmental stress, two cancer cell lines, MCF-7 and MKN45, were used and treated with 0.3 mM Dox and 0.3 mM PA that act to induce DNA damage and autophagy, respectively. Here, we found that DEK was down-regulated during Dox treatment. Meanwhile, the expressions of γH2AX, a DNA damage marker, were significantly increased during the early, 0–8 h, treatment in MCF-7 and 0–4 h treatment in MKN45 (Figure 5A,B). However, the increase was inhibited after 8 h in MCF-7 and after 4 h in MKN45, in which the expression of DEK was then decreased in both cell lines. These results indicate that the decrease in DEK correlative with DNA damage during Dox treatment. In addition, DEK was also down-regulated after PA treatment. LC3B, an autophagy marker, was also significantly increased during the early 0–6 h period in MCF-7 and MKN45 (Figure 5C,D). However, the increase was not observed after 6 h in MCF-7 and MKN45, in which DEK was then dramatically decreased in both cell lines. These results indicate that the decrease in DEK also correlative with autophagy during PA treatment. Thus, we suggest that the decrease in DEK expression is as a response to Dox and PA stress.
Expression of DEK responses to the stresses in human cancer cell lines.
To further confirm the response of DEK in cells resistant to such stresses, DEK was depleted by infecting the cells with short hairpin RNAi-expressing lentiviruses (shDEK 1 and shDEK 2). The cells were then rescued by adding recombinant His-DEK-GFP in MCF-7 and MKN45. Recombinant His-DEK was overexpressed and purified (Supplementary Figure S10A–C). We performed Western blot and immunofluorescence analysis of His-tag-DEK using anti-His-tag antibody, the results have shown that His-tag-DEK was in the inside of the cells (Supplementary Figure S10D,E). Western blot and immunofluorescence analyses revealed that after knockdown of DEK, the protein levels of DEK had decreased compared with those of the controls and had subsequently increased after His-DEK rescue (Figure 6A and Supplementary Figure S11A). We performed an analysis of Western blot to detect the yH2AX and LC3B levels in untreated control and DEK shRNA cells. The results showed that the LC3B levels had reduced, but no significant change of yH2AX expression was found in DEK-KD MCF-7 and MKN45 cells (Figure 6A and Supplementary Figure S11A). In addition, we found that the expression level of γH2AX and LC3B had decreased after DEK-KD, as compared with the control groups, but high levels of γH2AX and LC3B were observed after the addition of His-DEK to both cell lines under the conditions of Dox and PA treatments (Figure 6B,C and Supplementary Figure S11B,C). These results confirmed that the down-regulation of DEK could prevent DNA damage and lipotoxicity.
DEK increase resistance to environmental stress by an increase in constitutive heterochromatin.
Western blot was performed to characterize the chromatin structure after DEK-KD and then rescued by adding the recombinant His-DEK to the MCF-7 and MKN45 cells. Here, we found that H3K9ac, a euchromatin marker, H3K27me3 and H3K9me3, the facultative heterochromatin markers, were decreased, but H4K20me3, a marker for constitutive heterochromatin was enriched in MCF-7 and MKN45 after DEK-KD (Figure 6D and Supplementary Figure S11D). Furthermore, differences in HP1α expression were noted after DEK-KD. 24 h after adding His-DEK into DEK-KD cells, the levels of H3K27me3 and H3K9me3 were promoted, but the levels of H4K20me3 were decreased. However, the level of H3K9ac significantly increased at 48 h than that at 24 h after adding the recombinant His-DEK in MCF-7 and MKN45 DEK-KD cells (Supplementary Figure S13). This indicates that DEK plays a function in the regulation of chromatin structure via facilitating constitutive heterochromatin formation.
Pathways upstream and downstream of DEK regulation in MCF-7 cells
To elucidate the effect of DEK on proliferation and apoptosis of MCF-7 and MKN45, we performed analysis of Western blotting of p27, Ki67, PCNA, pH3S10 and active-caspase3 to verify the proliferation and apoptosis of MCF-7 and MKN45 after DEK-KD. We found that cell proliferation was inhibited by DEK RNAi during the period of 7 days. A high level of p27, a cyclin-dependent kinase inhibitor, was detected since the 4 days of DEK RNAi, indicating that they were in a quiescent state during the period. However, the significant apoptosis was observed after 6 days (Supplementary Figure S14).
To identify the signaling pathways by which DEK regulates cellular quiescence, RNA sequencing of DEK-KD in MCF-7 cell line was performed. We performed a cluster analysis of DEGs in the entire transcriptome and observed the overall gene pattern expression after DEK-KD in MCF-7. The gene expression pattern differed markedly after DEK-KD in MCF-7 cells (Supplementary Figure S15A). In total, 9565 genes were differentially expressed between MCF-7 DEK-KD cells and control cells (Supplementary Figure S15B). Gene set enrichment analysis revealed a down-regulated expression of genes involved in cell activation, proliferation and signaling pathways such as Wnt and MAPK (MEK–ERK–RSK1), and an up-regulated expression of p53, nuclear factor κB (NF-κB) in MCF-7 DEK knockdown cells (Supplementary Figure S15C). GO analysis showed that the expression of specific genes up- or down-regulated in MCF-7 DEK-KD cells were significantly enriched for GO terms linked to the regulation of chromatin stability, proliferation, differentiation, metabolism and related signaling pathways (Supplementary Figure S15D). Genes involved in the p53 signaling pathway and negative regulation of cell proliferation were up-regulated in cells after DEK-KD, while genes involved in cell division, the Wnt and AURKA signaling pathways, and many other metabolic pathways, were down-regulated. Our results demonstrate that DEK activates cellular quiescence in both Artemia and MCF-7 cells.
Cellular quiescence is a reversible state of arrest that relates to growth, proliferation and other cellular activities. The escape of such quiescent cells to reenter the cell cycle occurs in response to physiological cell stimuli and is thought to represent a homogenous state induced by diverse anti-mitogenic signals [9,23]. The appearance of cellular quiescence in the course of evolution involves central aspects of tissue function and homeostasis. Unlike proliferation, little is known about the molecular biology of cellular quiescence. Artemia is an ideal model system to study the regulation of cellular quiescence because Artemia embryos can remain in diapause for unusually prolonged periods where the cell cycle only resumes upon reactivation, leading to the ongoing development of post-diapause embryos. The present study demonstrates that Ar-DEK acts to terminate the diapause of Artemia embryos. More specifically, Ar-DEK is required for the activation of cellular quiescence during the termination of diapause in Artemia. Our results provide insights into the functions of DEK in Artemia that can be used to characterize the regulation of cellular quiescence in other contexts.
To study the function of DEK, DEK-KD has been performed in many previous studies. Depletion of DEK by short hairpin RNA reduces cellular proliferation, whereas its overexpression promotes proliferation and prevents differentiation of both keratinocytes and multiple breast cancer cell lines [21,40]. DEK has also been identified as a senescence inhibitor as DEK expression is reduced during replicative senescence, whilst the overexpression of DEK prolongs the lifespan of both primary and transformed keratinocytes . Much like its complex role in transcriptional regulation, DEK is involved in multiple cellular functions that have implications for cancer biology, including proliferation, differentiation, senescence and apoptosis. In this study, we found that the knockdown of Ar-DEK could induce embryonic cellular quiescence and diapause formation in Artemia. However, we did not find this situation in MCF-7 and MKN45 human cancer cell lines, in which the knockdown of DEK reduced proliferation during an initial period, but caused apoptosis beyond that point. Our results indicate that DEK is required for the proliferation involved in embryonic development and cancer growth. However, whilst the lack of DEK induces a semi-perpetual cellular quiescence during diapause formation in Artemia, after a brief period of quiescence, it ultimately causes apoptosis in cancer cell lines such as MCF-7 and MKN45. A previous study has reported that murine DEK is necessary for proper hematopoietic stem cell homeostasis and highly involved in multipotent progenitor cellular responses to radiation-induced damage, likely via regulation of the cellular decision to maintain quiescence versus enter a proliferative state . In addition, Cheung et al. had described the role of DEK in regulating quiescence in muscle stem cells . Based on these results, we suggest that down-regulated DEK may induce prolonged quiescence in stem cells, such as embryonic cells and hematopoietic stem cells, can only trigger a brief period of quiescence in differentiated cells such as cancer cells.
As an abundant and ubiquitous chromatin protein, DEK has two functional DNA binding domains, SAP/SAF and C-terminal and is distributed in the chromatin. [44,45]. A study of DEK in Arabidopsis thaliana showed that DEK contributes to the modulation of chromatin structure and function . A previous report has shown that DEK is a key factor in maintaining the balance between heterochromatin and euchromatin . Immunoprecipitation studies have also confirmed that DEK associates with activating histone modifications such as H3K4me2/3, rather than with repressive modifications such as H3K9me3 [14,47]. Consequently, the knockdown of DEK markedly reduces the distribution of facultative heterochromatin . In the current study of Artemia and MCF-7 and MKN45 cells, we found that the depletion of DEK induces an increase in constitutive heterochromatin and a decrease in constitutive euchromatin. These results indicate that DEK plays an important role in the maintenance of the balance between heterochromatin and euchromatin and also epigenetically regulates the relative gene expressions of diapause formation in Artemia.
Based on the analysis of transcriptomes in Artemia and MCF-7, we found that the p53 signaling pathway was up-regulated after DEK depletion. p53 is downstream of DEK and plays a role in the quiescence of various cell types . Consistently, we found that p53 target genes were up-regulated in diapause embryos but were down-regulated in pre-diapause and post-diapause embryos. The Hippo signaling pathway, involved in the regulation of cellular proliferation, apoptosis and stem cell self-renewal, was up-regulated after DEK depletion. This indicates that the Hippo signaling pathway is evolutionarily conserved and is an important downstream pathway for DEK regulation of cellular quiescence [49–51]. With the exception of the AURKA–PLK–MEK–ERK–RSK1 signaling pathway, which regulates cellular quiescence in diapause embryos and which was also suppressed (as mentioned above [52,53]). Wnt signaling pathway was down-regulated after DEK depletion in both Artemia and MCF-7 cells. Further support for DEK as a regulator of the Wnt/β-catenin signaling was published by Shibata et al., where DEK was shown to up-regulate the expression of Wnt10b, a canonical Wnt ligand which can activate the Wnt/β-catenin pathway, in neuroendocrine carcinomas of the lung . Wnt signaling helps to maintain the quiescence of hematopoietic stem cells. This may underlie why Wnt is required to preserve the self-renewal capability of these cells . A recent report by Liu et al. also demonstrated that DEK depletion in Caski cervical cancer cells resulted in phosphorylation of IκBα and the nuclear translocation of p65 with increased DNA target sequence binding . NF-κB signaling pathway also plays critical roles in the maintenance of human embryonic stem cell pluripotency . This indicates the far-reaching impact that NF-κB has on development and homeostasis. Taken together, we suggest that the AURKA–PLK–MEK–ERK–RSK1, p53, Wnt, NF-κB signaling pathways are simultaneously required for the regulation of the cellular quiescence primarily regulated by DEK.
Furthermore, genes related to aspects of metabolism, including cellular oxidative phosphorylation, cholesterol, xenobiotic and carbohydrate metabolic process, autophagy, as well as genes related to cell proliferation, including those associated with the mitotic cell cycle and translation regulator activity, were all down-regulated after DEK depletion. In contrast, genes associated with histone dephosphorylation and the negative regulation of cellular metabolic processes, cellular biosynthesis and transcription by RNA polymerase II transcription factor activity were up-regulated. Our results demonstrate that DEK may regulate cellular quiescence in both Artemia and MCF-7 cells.
differentially expressed genes
mitogen-activated protein kinase 3
fetal bovine serum
green fluorescent protein
trimethylation of H3K27
acetylation of H3K9
trimethylation of H3K9
phosphorylation of histone H3 at Ser10
trimethylation of H4K20
histidine (His)-tagged DEK-IRES-GFP
ERK activator kinase 1
nuclear factor κB
phosphorylation of Plk1 at Thr210
phosphorylation of retinoblastoma (Rb) at Thr356
- RSK1 (Ser221) ph
phosphorylation of RSK1 p90 at Ser221
ribosomal protein S6 kinase alpha-1
real-time quantitative PCR
short hairpin RNA
W.Y. and W.J. designed the experiments and wrote the manuscript. W.J. performed the experiments and collected and analyzed the data. A.L., J.F., Y.D., S.Y. and J.Y. performed the experiments of cell culture, Western blot and analysis of RNA sequencing. All authors discussed the results and contributed to editing the manuscript.
This work was supported by the National Natural Science Foundation of China (Project No. 31730084) and the National Major Research and Development Project (2016YFA0101201).
The Authors declare that there are no competing interests associated with the manuscript.