ARID3A regulates autophagy related gene BECN1 expression and inhibits proliferation of osteosarcoma cells.

Osteosarcoma (OS) is the most common primary malignant bone tumor which has unclear pathobiology. Hence, enlightening the exact molecular mechanism underlying osteosarcoma progression is crucial for developing new treatment strategies. One member of the ARID family of DNA binding proteins is ARID3A that is implicated in osteosarcoma pathogenesis. ARID3A could bind E2F1 and regulate the transcription of E2F1 targets. At the same time, BECN1 is a well-characterized autophagy regulator gene that is a direct target of E2F1. The present study aimed to investigate the effect of ARID3A on the expression of BECN1 in osteosarcoma cells. First, we determined gene expression levels of ARID3A, BECN1, and E2F1 in U-2 OS by qPCR and confirmed with online datasets from GEO database. In addition, the prognostic value of these genes was also evaluated from Kaplan-Meier plotter database. Next, ARID3A was overexpressed and silenced in order to investigate the effect of ARID3A on BECN1 expression and proliferation of U-2 OS cells. Our results demonstrated that BECN1 was negatively correlated with E2F1 and positively correlated with ARID3A based on initial expression and prognostic effect in OS. Overexpression of ARID3A upregulated BECN1 while silenced ARID3A downregulated BECN1 expression in U-2 OS cells. Additionally, silencing of ARID3A promoted colony formation and proliferation, whereas overexpression of ARID3A suppressed colony formation and proliferation of U-2 OS cells. Taken together, these results indicate that ARID3A could function as tumor suppressor and affect the expression level of BECN1 in U-2 OS cells.

Long noncoding RNA ERICD interacts with ARID3A via E2F1 and regulates migration and proliferation of osteosarcoma cells.

Long noncoding RNA (lncRNA) dysregulation is known to be taking part in majority of cancers, including osteosarcoma. In one of our previous studies, we showed that lncRNA MEG3 is being regulated by microRNA-664a (miR-664a) suppresses the migratory potential of osteosarcoma cells (U-2OS). We now report a novel lncRNA, namely, ERICD, which is linked to the transcription factor AT-rich interaction domain 3A (ARID3A) in U-2OS cells. We show that ARID3A binds to ERICD and indirectly interacts with each other via the E2F transcription factor 1 (E2F1). Furthermore, small interfering RNA (siRNA)-mediated knockdown of ERICD inhibited cell migration, formation of colonies, and proliferation in U-2OS cells. Overexpression of ARID3A inhibited cell migration, colony formation, and proliferation, whereas siRNA-mediated knockdown of ARID3A promoted cell migration, colony formation, and proliferation. Our findings indicate that ARID3A and lncRNA ERICD have plausible tumor suppressive and oncogenic functions, respectively, in osteosarcoma. Our data demonstrate the converse interaction between ARID3A and lncRNA ERICD that target DNA-binding proteins and dysregulation of their expression through E2F1 augments osteosarcoma progression. The cell rescue experiment also indicated E2F1 to be involved in the regulation of ARID3A and ERICD.

Extended Culture Conditions for Multipotent Bone Marrow-Derived Mesenchymal Stem Cells.

Mesenchymal stem cells (MSCs) offer a promising source of cells for musculoskeletal regeneration because of their potential to differentiate into bone, cartilage and fat. However, their proliferation and multilineage differentiation potential decreases with aging or increased time in in vitro culture. To determine culture conditions capable of enabling maintenance of MSCs for extended periods of time, human bone marrow-derived MSCs (BM-MSCs) were cultured in growth medium containing various combinations of growth factors and small chemical compounds. Upon reaching confluence, MSCs were subcultured continuously and then tested for differentiation capacity. After screening various growth factors and small chemical compounds, we found a combination capable of maintaining the proliferation potential of BM-MSCs obtained from a 19-year-old donor (young MSCs) up to passage 13 (P13). In contrast, unsupplemented MSCs reached senescence at P10. Total population doublings of control (P10) and supplemented MSCs (P12) were estimated at 20.4 and 42, respectively. Young MSCs cultured with supplements maintained osteogenic, adipogenic and chondrogenic differentiation capacities at P12 as confirmed by expression of lineage-specific differentiation markers. Furthermore, the supplementation of to BM-MSCs obtained from 65- and 79-year-old donors (aged MSCs) also continued to proliferate until P12, and maintained osteogenic and adipogenic differentiation capacity until P7 and P8, respectively, whereas, unsupplemented aged MSCs stopped proliferating at P8. These results indicate that our extended culture conditions maintained the proliferative capacity of young MSCs while retaining their multipotent differentiation potential, and improved both proliferation and differentiation of aged MSCs.

Critical role of ARID3B in the expression of pro-apoptotic p53-target genes and apoptosis.

ARID3A and ARID3B are transcriptional targets of p53. Recently, it has been reported that ARID3A plays a critical role in the transcriptional activation of pro-arrest p21 in response to DNA damage. However, the role of ARID3B in the p53 regulatory pathway remains poorly understood. Here we show that ARID3A and ARID3B specifically bind to putative ARID3-binding sites in p53 target genes in vitro and in vivo. ARID3B and, to a lesser extent, ARID3A silencing blocked transcriptional activation of pro-apoptotic p53 target genes, such as PUMA, PIG3, and p53. Furthermore, ectopic ARID3B, to a lesser extent, ARID3A expression activated the pro-apoptotic gene expression, and only ARID3B induced apoptosis. Finally, ARID3B but not ARID3A silencing blocked apoptosis induction following DNA damage. These results indicated that, although ARID3B and ARID3A share overlapping functions, ARID3B play a key role in the expression of pro-apoptotic p53-target genes and apoptosis.

Integrated Bone Formation Through In Vivo Endochondral Ossification Using Mesenchymal Stem Cells.

Conventional bone regeneration therapy using mesenchymal stem cells (MSCs) is difficult to apply to bone defects larger than the critical size because it does not have a mechanism to induce angiogenesis. Implanting artificial cartilage tissue fabricated from MSCs induces angiogenesis and bone formation in vivo via endochondral ossification (ECO). Therefore, this ECO-mediated approach may be a promising bone regeneration therapy in the future. An important aspect of the clinical application of this ECO-mediated approach is establishing a protocol for preparing enough cartilage to be implanted to repair the bone defect. It is especially not practical to design a single mass of grafted cartilage of a size that conforms to the shape of the actual bone defect. Therefore, the cartilage to be transplanted must have the property of forming bone integrally when multiple pieces are implanted. Hydrogels may be an attractive tool for scaling up tissue-engineered grafts for endochondral ossification to meet clinical requirements. Although many naturally derived hydrogels support MSC cartilage formation in vitro and ECO in vivo, the optimal scaffold material to meet the needs of clinical applications has yet to be determined. Hyaluronic acid (HA) is a crucial component of the cartilage extracellular matrix and is a biodegradable and biocompatible polysaccharide. Here, we show that HA hydrogels have excellent properties to support in vitro differentiation of MSC-based cartilage tissue and promote endochondral bone formation in vivo.

Hyaluronic acid hydrogels support to generate integrated bone formation through endochondral ossification in vivo using mesenchymal stem cells.

Engineered cartilage tissue from differentiated mesenchymal stem cells (MSCs) can generate bone in vivo through endochondral ossification (ECO). This ECO-mediated approach has the potential to circumvent the severe problems associated with conventional MSC-based bone tissue engineering techniques that lack mechanisms to induce angiogenesis. Hyaluronic acid (HA) is a key component in the cartilage extracellular matrix. However, the ECO-supporting properties of HA remain largely unclear. This study aimed to compare the ability of HA and collagen hydrogels to support in vitro differentiation of MSC-based hypertrophic cartilage tissues and to promote endochondral bone formation in vivo. Following the chondrogenic and hypertrophic differentiation in vitro, both HA and collagen constructs accumulated sulfated glycosaminoglycan (sGAG) and type 1, type II, and type X collagen. However, HA hydrogels exhibited a more uniform distribution of sGAG, type 1 collagen, type X collagen, and osteocalcin proteins; in addition, the cells embedded in the hydrogels had more rounded cell morphologies than those in the collagen constructs. At week 5 of in vitro culture, two to three constructs were implanted into a subcutaneous pocket in nude mice and harvested after 4 and 8 weeks. Both HA and collagen constructs promoted endochondral bone formation with vascularization and bone marrow development; however, the HA constructs fused to form integrated bone tissues and the bone marrow developed along the space between the two adhered grafts in all implanted pockets (n = 5). In the collagen constructs, the integration was observed in 40% of the pockets (n = 5). Microcomputer CT analysis revealed that the bone volume of HA constructs was larger than that of collagen constructs. In conclusion, compared to collagen hydrogels, HA hydrogels had superior potential to generate integrated bone with vascularization and bone marrow development. This study provides valuable insights for applying ECO-mediated bone tissue engineering approaches for the repair of critical-sized bone defects.

ARID3A and ARID3B exert direct regulatory control over the long non-coding RNAs (lncRNAs) MALAT1 and NORAD within the context of non-small cell lung cancer (NSCLC).

Lung cancer, known for its high mortality rates and poor prognosis, remains one of the most prevalent cancer types. Early detection and effective treatment methods are crucial for improving survival rates. Non-small cell lung cancer (NSCLC) accounts for approximately 85 % of all lung cancer cases. Long non-coding RNAs (lncRNAs), which play vital roles in various biological processes, have been implicated in the development of cancer and can impact key therapeutic targets in different cancer types. In NSCLC, the dysregulation of specific lncRNAs, such as MALAT1 and NORAD, has been associated with neoplastic initiation, progression, metastasis, tumor angiogenesis, chemoresistance, and genomic instability. Both MALAT1 and NORAD directly regulate the expression of the transcription factor E2F1, thereby influencing cell cycle progression. Additionally, MALAT1 has been reported to affect the expression of p53 target genes, leading to cell cycle progression through the repression of p53 promoter activity. NORAD, on the other hand, is indirectly regulated by p53. The AT-rich interaction domain (ARID) family of DNA-binding proteins, particularly ARID3A and ARID3B, are involved in various biological processes such as cell proliferation, differentiation, and development. They also play significant roles in E2F-dependent transcription and are transcriptional targets of p53. The intricate balance between promoting cellular proliferation through the pRB-E2F pathway and inducing growth arrest through the p53 pathway underscores the crucial regulatory role of ARID3A, ARID3B, and their interaction with lncRNAs MALAT1 and NORAD. In this study, we aimed to investigate the potential interactive and functional connections among ARID3A, ARID3B, MALAT1, and NORAD in NSCLC, considering their involvement in the pRB-E2F and p53 pathways. Our findings strongly suggest that ARID3A and ARID3B play a regulatory role in controlling MALAT1 and NORAD in NSCLC. Specifically, our study demonstrates that the activities of MALAT1 and NORAD were markedly increased upon the overexpression of ARID3A and ARID3B. Therefore, we can conclude that ARID3A and ARID3B likely contribute significantly to the oncogenic functions of MALAT1 and NORAD in NSCLC. Consequently, targeting ARID3A and ARID3B could hold promise as a therapeutic approach in NSCLC, given their direct control over the expression of MALAT1 and NORAD.

Cooperation between ARID3A and p53 in the transcriptional activation of p21WAF1 in response to DNA damage.

ARID3A/DRIL1/Bright is a family member of the AT rich interaction domain (ARID) DNA-binding proteins that are involved in diverse biological processes. We have reported that p53 activates ARID3A transcription, and ARID3A overexpression induces G1 arrest. However, the role of ARID3A in the p53 pathway remains unclear. Here, we show that ARID3A cooperates with p53 to transcriptionally activate p21(WAF1), a p53-target gene important for cell-cycle arrest. ARID3A bound to its binding sites in the p21(WAF1) promoter in vivo and in vitro, and induced p21(WAF1) transcription in U2OS cells expressing wild-type p53 but not Saos-2 cells lacking p53. The co-expression of ARID3A with p53 cooperates to activate p21(WAF1) transcription and the stably transfected p21(WAF1) promoter. Mutation of the ARID3A binding sites reduced the p21(WAF1) promoter activity, and siRNA-based ARID3A knockdown suppressed the transcription of p21(WAF1), but not the proapoptotic NOXA and PUMA in response to DNA damage. Furthermore, p53 knockdown decreased ARID3A transcription, and, conversely, ARID3A overexpression and knockdown resulted in an increase or decrease in p53 stability, respectively. These results indicate both cooperative and interdependent roles for ARID3A and p53 in the transcriptional activation of p21(WAF1) in response to DNA damage.

Inability of p53-reactivating compounds Nutlin-3 and RITA to overcome p53 resistance in tumor cells deficient in p53Ser46 phosphorylation.

The p53 tumor suppressor protein plays key roles in protecting cells from tumorigenesis. Phosphorylation of p53 at Ser46 (p53Ser46) is considered to be a crucial modification regulating p53-mediated apoptosis. Because the activity of p53 is impaired in most human cancers, restoration of wild-type p53 (wt-p53) function by its gene transfer or by p53-reactivating small molecules has been extensively investigated. The p53-reactivating compounds Nutlin-3 and RITA activate p53 in the absence of genotoxic stress by antagonizing the action of its negative regulator Mdm2. Although controversial, Nutlin-3 was shown to induce p53-mediated apoptosis in a manner independent of p53 phosphorylation. Recently, RITA was shown to induce apoptosis by promoting p53Ser46 phosphorylation. Here we examined whether Nutlin-3 or RITA can overcome resistance to p53-mediated apoptosis in p53-resistant tumor cell lines lacking the ability to phosphorylate p53Ser46. We show that Nutlin-3 did not rescue the apoptotic defect of a Ser46 phosphorylation-defective p53 mutant in p53-sensitive tumor cells, and that RITA neither restored p53Ser46 phosphorylation nor induced apoptosis in p53Ser46 phosphorylation-deficient cells retaining wt-p53. Furthermore, treatment with Nutlin-3 or RITA together with adenoviral p53 gene transfer also failed to induce apoptosis in p53Ser46 phosphorylation-deficient cells either expressing or lacking wt-p53. These results indicate that neither Nutlin-3 nor RITA in able to induce p53-mediated apoptosis in the absence of p53Ser46 phosphorylation. Thus, the dysregulation of this phosphorylation in tumor cells may be a critical factor that limits the efficacy of these p53-based cancer therapies.

Essential role of caspase-8 in p53/p73-dependent apoptosis induced by etoposide in head and neck carcinoma cells.

BACKGROUND: Caspase-8 is a key upstream mediator in death receptor-mediated apoptosis and also participates in mitochondria-mediated apoptosis via cleavage of proapoptotic Bid. However, the role of caspase-8 in p53- and p73-dependent apoptosis induced by genotoxic drugs remains unclear. We recently reported that the reconstitution of procaspase-8 is sufficient for sensitizing cisplatin- but not etoposide-induced apoptosis, in chemoresistant and caspase-8 deficient HOC313 head and neck squamous cell carcinoma (HNSCC) cells. RESULTS: We show that p53/p73-dependent caspase-8 activation is required for sensitizing etoposide-induced apoptosis by utilizing HOC313 cells carrying a temperature-sensitive p53G285K mutant. Restoration of wild-type p53 function under the permissive conditions, together with etoposide treatment, led to substantial transcriptional activation of proapoptotic Noxa and PUMA, but failed to induce apoptosis. In addition to p53 restoration, caspase-8 reconstitution was needed for sensitization to etoposide-induced apoptosis, mitochondria depolarization, and cleavage of the procaspases-3, and -9. In etoposide-sensitive Ca9-22 cells carrying a temperature-insensitive mutant p53, siRNA-based p73 knockdown blocked etoposide-induced apoptosis and procaspase-8 cleavage. However, induction of p73 protein and up-regulation of Noxa and PUMA, although observed in Ca9-22 cells, were hardly detected in etoposide-treated HOC313 cells under non-permissive conditions, suggesting a contribution of p73 reduction to etoposide resistance in HOC313 cells. Finally, the caspase-9 inhibitor Ac-LEHD-CHO or caspase-9 siRNA blocked etoposide-induced caspase-8 activation, Bid cleavage, and apoptosis in both cell lines, indicating that p53/p73-dependent caspase-8 activation lies downstream of mitochondria. CONCLUSIONS: we conclude that p53 and p73 can act as upstream regulators of caspase-8, and that caspase-8 is an essential mediator of the p53/p73-dependent apoptosis induced by etoposide in HNSCC cells. Our data suggest the importance of caspase-8-mediated positive feedback amplification in the p53/p73-dependent apoptosis induced by etoposide in HNSCC cells.

Distinct and overlapping roles of ARID3A and ARID3B in regulating E2F­dependent transcription via direct binding to E2F target genes.

The AT­rich interacting domain (ARID) family of DNA­binding proteins is involved in various biological processes, including the regulation of gene expression during cell proliferation, differentiation and development. ARID3A and ARID3B are involved in chromatin remodeling and can bind to E2F1 and retinoblastoma tumor suppressor protein (RB), respectively. However, their role in regulating E2F target gene expression remains poorly understood. E2F transcription factors are critical regulators of cell cycle progression and are modulated by RB. Herein, putative ARID3­binding sites (BSs) in E2F target genes were identified, including Cdc2, cyclin E1 and p107, and it was found that ARID3A and ARID3B bound to these BSs in living cells. The mutation of ARID3 BSs reduced Cdc2 promoter activity, while ARID3A and ARID3B overexpression increased the promoter activity, depending on both ARID3 and E2F BSs. ARID3B knockdown blocked the transcription of Cdc2, cyclin E1 and p107 in normal human dermal fibroblasts (NHDFs), whereas the effects of ARID3A knockdown varied depending on the target genes. ARID3B overexpression, but not that of ARID3A, upregulated the transcription of E2F target genes, and activated cyclin E1 transcription and induced cell death with E2F1 assistance. Finally, ARID3A and ARID3B knockdown attenuated the cell cycle progression of NHDFs and T98G cells, and suppressed tumor cell growth. On the whole, these results indicate that ARID3A and ARID3B play distinct and overlapping roles in E2F­dependent transcription by directly binding to the E2F target genes. The present study provides novel insight into the mechanisms underlying the E2F dysregulation caused by ARID3A and ARID3B overexpression, which may have a significant influence on the progression of tumorigenesis.

E2F-like elements in p27(Kip1) promoter specifically sense deregulated E2F activity.

The transcription factor E2F, the main target of the RB tumor suppressor pathway, plays crucial roles not only in cell proliferation but also in tumor suppression. The cyclin-dependent kinase inhibitor p27(Kip1) gene, an upstream negative regulator of E2F, is induced by ectopically expressed E2F1 but not by normal growth stimulation that physiologically activates endogenous E2F. This suggests that the gene can discriminate between deregulated and physiological E2F activity. To address this issue, we examined regulation of the p27(Kip1) gene by E2F. Here we show that p27(Kip1) promoter specifically senses deregulated E2F activity through elements similar to typical E2F sites. This E2F-like elements were activated by deregulated E2F activity induced by forced inactivation of pRb but not by physiological E2F activity induced by serum stimulation, contrary to typical E2F sites activated by both E2F activity. The endogenous p27(Kip1) gene responded to deregulated and physiological E2F activity in the same manner to the E2F-like elements. Moreover, the E2F-like elements bound ectopically expressed E2F1 but not physiologically activated E2F1 or E2F4 in vivo. These results suggest that the p27(Kip1) gene specifically senses deregulated E2F activity through the E2F-like elements to suppress inappropriate cell cycle progression in response to loss of pRb function.

Association of caspase-8 mutation with chemoresistance to cisplatin in HOC313 head and neck squamous cell carcinoma cells.

Caspase-8 is a critical upstream mediator of apoptosis in the death receptor pathway. However, the relationship between caspase-8 mutation and chemosensitivity remain unclear in head and neck squamous cell carcinoma (HNSCC) carrying p53 mutation. In this study, we identified a caspase-8 nonsense mutation, accompanied by the loss of the second allele, in a drug-resistant HOC313 HNSCC cell line. The nonsense mutation (R68X) leads to truncation of all defined functional domains. Reconstitution of caspase-8 by stable transfection of wild-type caspase-8 sensitized the cells to cisplatin-, but not etoposide-induced apoptosis. Consistent with this, cisplatin, but not etoposide, induced TNF-alpha and TRAIL mRNA in caspase-8 reconstituted HOC313 cells, accompanied by activation of the reconstituted caspase-8 and its downstream caspase-3. These results indicate that the loss of caspase-8 plays an important role in acquisition of chemoresistance to cisplatin in HOC313 cells.

Inhibitory effect of CT domain of CCN3/NOV on proliferation and differentiation of osteogenic mesenchymal stem cells, Kusa-A1.

CCN3/NOV activates the Notch signal through the carboxyl terminal cysteine-rich (CT) domain. CCN3 transfection to Kusa-A1 inhibited osteogenic differentiation and cell proliferation, which is accompanied by upregulation of Hes/Hey, Notch downstream targets, and p21, a CDK inhibitor. Upregulation of Hes/Hey and p21 was abrogated by the deletion of CT domain. Anti-proliferative activity of CCN3 was also abrogated by CT domain deletion whereas anti-osteogenic activity was not completely abrogated. We found that CT domain-deleted CCN3 still possesses antagonistic effect on BMP-2. These results suggest that CCN3 employs Notch and BMP pathways in anti-osteogenic activity while it inhibits cell proliferation uniquely by Notch/p21 pathway.

Interaction of E2F-Rb family members with corepressors binding to the adjacent E2F site.

Cell cycle-dependent transcriptional repression of the E2F1 and B-myb promoters is mediated through E2F-binding sites and adjacent corepressor site (cell cycle gene homology region (CHR)/downstream repression site (DRS)). Here, we show that a factor binding to the B-myb CHR is co-purified with E2F DNA-binding activity, and coimmunoprecipitated with components of E2F/Rb-family repressor complexes, E2F4 and retinoblastoma (Rb) family proteins. In spite of structural and functional similarities, however, the E2F1 and B-myb CHRs exhibited distinct factor-binding specificities. Furthermore, substitution of E2F1 CHR with the B-myb CHR in the E2F1 promoter revealed that the B-myb CHR was unable to repress the E2F1 promoter completely in the G0 phase. These results suggest that transcriptional repression of the E2F1 and B-myb promoters is mediated by physical interaction of E2F/Rb-family repressor complexes with promoter-specific corepressors.

High-risk human papillomavirus type 16 E7 oncogene associates with Cdc25A over-expression in oral squamous cell carcinoma.

Cells expressing high-risk human papillomavirus (HPV) E7 protein display impaired checkpoint control after DNA damage and exhibit elevated rates of mutagenesis. Repression of HPV E7 expression results in the subsequent accumulation of hypophosphorylated retinoblastoma protein and repression of the Cdc25A genes. No study has been conducted to elucidate the role of Cdc25A in the development and progression of human oral carcinomas. To confirm Cdc25A protein expression together with HPV, immunohistochemistry, Western blotting, polymerase chain reaction (PCR), and reverse transcriptase (RT)-PCR were performed using various histological subtypes of oral carcinomas. Cdc25A protein was localized predominantly in the cell nuclei in carcinomas, and high expression was found in 54% of primary tumors. HPV-16 E7 was not found in non-neoplastic oral tissues, whereas it was observed in eight (36%) of 22 oral carcinomas. We found a significant correlation between Cdc25A over-expression and HPV-16 E7 positive carcinomas. There was a strong positive correlation between Cdc25A over-expression and tumor size and TNM stage. This study suggests that Cdc25A is likely to be an important mediator in the progression of oral tumors, and HPV-16 E7 may be a sensitive indicator of the involvement of viral oncogenes in oral carcinogenesis.

FUCA1 is induced by wild-type p53 and expressed at different levels in thyroid cancers depending on p53 status.

Fucose residues of cell surface glycans, which play important roles in growth, invasion and metastasis, are added by fucosyltransferases (FUTs) and removed by α-L-fucosidases (FUCAs). By the differential display method, we isolated a 3' non-coding region of α-L-fucosidase-1 (FUCA1) (a gene coding for the lysosomal fucosidase-1 enzyme) as a wild-type p53-inducible gene: 18S and 20S FUCA1 mRNA species were induced in Saos-2 cells transfected with a temperature-sensitive p53 mutant at the permissive temperature. By microarray analyses of thyroid cancer biopsy samples, FUCA1 RNA expression levels were found to be lower in anaplastic thyroid cancer samples (ATCs), while they were higher in papillary thyroid cancer samples (PTCs) and in normal thyroid tissues. Since most ATCs were reported to carry the mutated form of p53, while PTCs carry mostly the wild-type form of p53, it is likely that FUCA1 expression levels are regulated, at least in part, by the p53 status in thyroid cancers. In order to better understand the role played by FUCA genes in thyroid tumorigenesis, we examined the clonogenic potential in vitro of thyroid cell lines transfected with either FUCA1 or FUCA2 (the latter gene coding for a secreted, non-lysosomal enzyme). We found that α-L-fucosidases did not suppress grossly cell growth. Contrary to what we observed with the expression of FUCA1, the FUT8 expression levels were found high in ATCs but lower in PTCs and normal thyroid tissues. Taken together, these results suggest the possibility that the higher fucose levels on cell surface glycans of aggressive ATCs, compared to those of less aggressive PTCs, may be at least in part responsible for the more aggressive and metastatic phenotype of ATCs compared to PTCs, as the expression levels of FUCA1 and FUT8 were inversely related in these two types of cancers.

Critical role of cyclin D1 nuclear import in cardiomyocyte proliferation.

Mammalian cardiomyocytes irreversibly lose their capacity to proliferate soon after birth, yet the underlying mechanisms have been unclear. Cyclin D1 and its partner, cyclin-dependent kinase 4 (CDK4), are important for promoting the G1-to-S phase progression via phosphorylation of the retinoblastoma (Rb) protein. Mitogenic stimulation induces hypertrophic cell growth and upregulates expression of cyclin D1 in postmitotic cardiomyocytes. In the present study, we show that, in neonatal rat cardiomyocytes, D-type cyclins and CDK4 were predominantly cytoplasmic, whereas Rb remained in an underphosphorylated state. Ectopically expressed cyclin D1 localized in the nucleus of fetal but not neonatal cardiomyocytes. To target cyclin D1 to the nucleus efficiently, we constructed a variant of cyclin D1 (D1NLS), which directly linked to nuclear localization signals (NLSs). Coinfection of recombinant adenoviruses expressing D1NLS and CDK4 induced Rb phosphorylation and CDK2 kinase activity. Furthermore, D1NLS/CDK4 was sufficient to promote the reentry into the cell cycle, leading to cell division. The number of cardiomyocytes coinfected with these viruses increased 3-fold 5 days after infection. Finally, D1NLS/CDK4 promoted cell cycle reentry of cardiomyocytes in adult hearts injected with these viruses, evaluated by the expression of Ki-67, which is expressed in proliferating cells in all phases of the cell cycle, and BrdU incorporation. Thus, postmitotic cardiomyocytes have the potential to proliferate provided that cyclin D1/CDK4 accumulate in the nucleus, and the prevention of their nuclear import plays a critical role as a physical barrier to prevent cardiomyocyte proliferation. Our results provide new insights into the development of therapeutics strategies to induce regeneration of cardiomyocytes. The full text of this article is available at http://www.circresaha.org.

Distinct recruitment of E2F family members to specific E2F-binding sites mediates activation and repression of the E2F1 promoter.

The activity of E2F transcription factors plays a crucial role in mammalian cell-cycle progression and is controlled by physical association with the pocket proteins (pRb and its related p107 and p130). The E2F1 promoter, which contains two overlapping E2F-binding sites, is activated at the G1/S transition in an E2F-dependent manner. Mutational experiments have shown that the distal E2F-binding site on the E2F1 promoter is required for transcriptional repression in the G0 phase, whereas the proximal E2F-binding site contributes to transcriptional activation at the G1/S boundary. Consistent with these results, chromatin immunoprecipitation assays have revealed that the E2F4/p130 repressor complex specifically binds to the distal E2F-binding site, whereas E2F1 and E2F3 activators preferentially bind to the proximal E2F-binding site. The assays also showed that the specific binding of E2F4/p130 complex to the distal site was dramatically impaired by a mutation introduced into the contiguous repression site (cell Cycle gene Homology Region; CHR). Taken together, these findings indicate that the two E2F-binding sites play distinct roles in the regulation of E2F1 transcription by interacting with different sets of E2F members and cooperating with the contiguous repressor element.

Role of cyclin D1 cytoplasmic sequestration in the survival of postmitotic neurons.

Cyclin D-dependent kinases phosphorylate the retinoblastoma (Rb) protein and play a critical role in neuronal cell cycle control and apoptosis. Here we show that cyclin D1 became predominantly cytoplasmic as primary cortical progenitor cells underwent cell cycle withdrawal and terminal differentiation. Furthermore, ectopically expressed cyclin D1 sequestered in the cytoplasm of postmitotic neurons, whereas it efficiently entered the nucleus of proliferating progenitor cells. Cytoplasmic cyclin D1 were complexed with cyclin-dependent kinase 4 (CDK4), and also with CDK inhibitors, p27(Kip)(I) or p21(Cip)(I), which positively regulate assembly and nuclear accumulation of the cyclin D1-CDK4 complex. Although overexpression of p21(Cip)(I) promoted cyclin D1 nuclear localization, inhibition of either glycogen synthase kinase 3beta- or CRM1-mediated cyclin D1 nuclear export did not, suggesting that the inhibition of its nuclear import, rather than the acceleration of nuclear export, contributes to cytoplasmic sequestration of cyclin D1 in postmitotic neurons. In differentiated progenitor cells, nuclear localization of ectopic cyclin D1 induced apoptosis, and the DNA-damaging compound camptothecin caused nuclear accumulation of endogenous cyclin D1, accompanied by Rb phosphorylation. These results indicate that nuclear accumulation of cyclin D1 is inhibited in postmitotic neurons and suggest a role of its subcellular localization in neuronal death and survival.