InteractiveFly: GeneBrief

Axud1: Biological Overview | References

Gene name - Axud1

Synonyms - CG4272

Cytological map position - 22E1-22E1

Function - presumed transcriptional regulator

Keywords - tumour suppressor, imaginal discs, restriction of mitotic progression

Symbol - Axud1

FlyBase ID: FBgn0261647

Genetic map position - 2L:2,365,335..2,370,114 [+]

Classification - phospho-acceptor site for acidophilic serine/threonine kinases

Cellular location - nuclear

NCBI links: Precomputed BLAST | EntrezGene

Cell division rates and apoptosis sculpt the growing organs, and its regulation implements the developmental programmes that define organ size and shape. The balance between oncogenes and tumour suppressors modulates the cell cycle and the apoptotic machinery to achieve this goal, promoting and restricting proliferation or, in certain conditions, inducing the apoptotic program. Analysis of human cancer cells with mutation in AXIN gene has uncovered the potential function of AXUD1 as a tumour suppressor. Human AXUD1 has been described as a nuclear protein. This study found that a DAxud1-GFP fusion protein is localised to the nucleus during interphase, where it accumulates associated to the nuclear envelope, but becomes distributed in a diffused pattern in the nucleus of mitotic cells. The function of the Drosophila AXUD1 homolog was analyzed, and it was found that DAxud1 behaves as a tumour suppressor that regulates the proliferation rhythm of imaginal cells. Knocking down the activity of DAxud1 enhances the proliferation of these cells, causing in addition a reduction in cell size. Conversely, the increase in DAxud1 expression impedes cell cycle progression at mitosis through disturbance of Cdk1 activity, and induces the apoptosis of these cells in a JNK-dependent manner (Glavic, 2009).

Proliferation and apoptosis are the principal contributors to the establishment of organs and organisms size and shape. Both processes are genetically controlled, being proliferation rhythms and events of morphogenetic apoptosis a key component of animal development. Besides the basic cyclin-based machinery, an increasing number of signalling pathways and regulatory proteins, like proto-oncogenes and tumour suppressors, have been described to modify the cell cycle, often operating at the G1/S or G2/M transitions. In Drosophila, the Cdk2/Cyclin E complex is one important element controlling the G1/S transition. Other proteins such as the E2F/Dp heterodimer, Myc, Ras and several other signalling pathways modify the levels of Cyclin E, consequently changing the rate of the G1/S transition. The Cdk1/Cyclin B complex ultimately modulates the G2/M transition. Cdk1 is regulated by an elaborated arrangement of phosphorylation and dephosphorylation events, and in turn, controls the activity of Cdk1/Cyclin B complexes. The Cdc25/String phosphatase is one of the best-characterised activators of Cdk1 activity, and Tribbles, Wee1 and Myt1 kinases regulate protein levels and its phosphatase activity. In addition to G1/S and G2/M restrains, a third control point exists in the M/G1 transition. Spindle checkpoint proteins, anaphase promoting complex and Cdk1 are some of the elements that limit the rate of mitosis progression and subsequent re-enter into G1 phase. Tumour suppressors participate at the control points by imposing rate-limiting inputs, and mutations in these proteins in pathological conditions or in experimental situations where their levels are augmented, promote unrestrained proliferation or lead to cell death (Glavic, 2009).

The uncontrolled activation of canonical Wnt pathway has been described in a variety of cancers and tumour cell lines and thus elements that reduce Wnt signalling have been described to have tumour suppressor activity. Axin is one of the negative components of this pathway and mutations on it have been associated with cancer cells. Recently, analysing the transcription profile of human cancer cells with mutations in the AXIN gene, Ishiguro (2001) described a transcript that is under-represented in cancer cells compared with normal cells. AXIN expression was able to induce its transcription in a LEF/TCF independent mode, and as a result, this putative tumour suppressor gene was named AXUD1 (Axin Upregulated 1). Later, Gingras (2007) analysed the in vivo function of the three mouse paralogs (cysteine- serine-rich nuclear proteins CSRNP-1, -2 and -3) by generating the corresponding knockout animals, but no evidence was found of a role for these genes as tumour suppressors (Gingras, 2007). There is a single orthologue of the AXUD1/CSRNP family in Drosophila that maintains the structural features described for AXUD1 and CSRNPs proteins, which has been named DAxud1. This study investigated the function of DAxud1 using the genetic and molecular advantages of the fly model, aiming to analyse its potential activity as a tumour suppressor. The results indicate that DAxud1 is a nuclear protein that antagonises proliferation during mitosis in a Cdk1-dependent manner. DAxud1 reduction confers cells with a proliferation advantage that leads in the adult wing to higher cell densities accompanied by a reduction in cell size. Conversely, over-expression of DAxud1 blocks mitosis and promotes apoptosis through the activation of the JNK pathway (Glavic, 2009).

The CG4272 gene is located in the left arm of chromosome 2 at cytological position 22E1. CG4272 produces 2 transcripts of 4126 (CG4272-RA) and 3782 (CG4272-RB) nucleotides encoding a single polypeptide of 852 amino acids. Protein blast using the blastp algorithm identifies a number of related sequences in both vertebrate and invertebrate organisms. The vertebrate homologues belong to the Axin-upregulated family (Axud), also named TGFβ-induced apoptotic proteins (TAIP). Conservation of Drosophila CG4272 and related invertebrate members range between 38% identity (Aedes aegypti) to 60% identity (Anopheles ganbiae). Identity between CG4272 sequence and its vertebrate counterparts is between 43% (Danio rerio) to 45% identity (Homo sapiens AXUD1). To analyse the conservation of the Drosophila CG4272 protein a phylogenetic analysis was performed with sequences recovered using the BLAST-PSI programme. The criterion to select the putative orthologues was the highest level of similarity between the human AXUD1 protein and one predicted coding sequences in each genome. Thus, vertebrate paralogues were excluded to simplify the phylogenetic tree. Protein sequence comparison using CLUSTALW and consensus tree analysis using the PAUP* program in the neighbour joining algorithm configuration, shows that AXUD proteins in vertebrates are highly conserved (91%-98% identity among mammalian orthologues), and that the main conserved domain with Drosophila and vertebrates is a central acidic domain of 88 amino acids. This domain is predicted to be a presumptive phospho-acceptor site for acidophilic serine/threonine kinases (Glavic, 2009).

In situ hibridization analysis shows that DAxud1 is maternally contributed, and its transcripts localise homogenously in the early embryo and during the blastoderm stage. Later on, the expression is restricted to the cephalic furrow and mesodermal tissue along the anterior-posterior axis, as well as to the anterior and posterior gut precursors. Afterwards, DAxud1 expression decreases from tissues that have completed morphogenetic movements, and continues mostly in posterior and anterior gut precursors as well as in muscle precursors. The expression of DAxud1 is generalised in the wing blade of the wing disc and in the proliferative domains of the eye disc anterior to the morphogenetic furrow. Note that no expression is detected in differentiating ommatidia, and a clear reduction is observed posterior to the morphogenetic furrow, where the second mitotic wave occurs (Glavic, 2009).

Human and mouse AXUD proteins localise in the cell nucleus. In fact, mice AXUD/TAIP paralogues cluster within a family of proteins with transcription factors characteristics. Computational prediction suggests that Drosophila Axud1 is also a nuclear protein (99.9%). To study DAxud1 sub-cellular localisation, a DAxud1-GFP fusion protein was generated and expressed in the wing imaginal disc. Interestingly, the level of DAxud1-GFP is consistently lower in mitotic cells, those appearing in the apical side of the epithelium, than in cells in other phases of the cell cycle. Staining of sal-Gal4/UAS-DAxud1-GFP discs with the nucleic acid marker Topro shows that in interphase cells DAxud1-GFP protein accumulation is higher in a region not labelled with Topro that might correspond to the inner nuclear envelope, and that DAxud1-GFP is excluded from the nucleolus. The accumulation of DAxud1-GFP in the inner nuclear membrane is also observed in salivary gland cells, where DAxud1-GFP co-localises in part with the nuclear envelope protein Nup214 (Glavic, 2009).

This study investigated the function of Drosophila Axud1 during development, focusing on its effects on cell division and apoptosis in imaginal discs. Phylogenetic analysis demonstrate that AXUD proteins (CSRNP in mice) possess a stretch of Cysteine residues of variable length common to all species analysed. Although a functional analysis of this domain is still missing, the structural hallmarks of AXUD proteins suggest they constitute a novel conserved family of transcription factors (Glavic, 2009).

To address the function of DAxud1 an in vivo approach used both loss- and gain-of-function experiments. Over-expression analysis in imaginal discs revealed that excess of DAxud1 impairs organ development, causing a reduction in organ size. Cell death analysis in these tissues indicates that DAxud1 over-expression consistently activates apoptosis, indicating that size reduction can be attributed in part to a diminution in cell number. Several experimental situations that lead to apoptosis in imaginal discs, including morphogenetic apoptosis, implicate the inappropriate activation of the JNK pathway. This study found that the adult wing phenotype caused by DAxud1 over-expression is accompanied with the activation of the JNK pathway in the wing disc. Accordingly, this phenotype is partially reverted in heteroallelic combinations with several components of the JNK pathway or by expressing its negative regulator. Taken together, these results suggest that DAxud1 pro-apoptotic effect is caused by the activation of the JNK pathway in imaginal cells (Glavic, 2009).

Since activation of the Dpp and Wg pathways are necessary for normal wing disc development, the cell death elicited by DAxud1 could be a consequence of an inhibitory effect on these pathways. However, no effects were detected in the expression of Dpp and Wg target genes, indicating that DAxud1 functions independently of Dpp and Wg activity. Due to the proposed role as a tumour suppressor for AXUD1, the function of DAxud1 in cell cycle progression was studied. A blockage in proliferation was observed accompanied by the accumulation of cells in mitosis. The reduced levels of BrdU incorporation and the concomitant increase in the number of cells positive for mitotic marker PH3 could arise from an extremely fast progression through the cell cycle. Because a relatively short BrdU pulse (30 min) was performed and the tissue was immediately fixed afterwards, it is very unlikely that mitotic dilution of BrdU explains the observation. Since DAxud1 also promotes apoptosis, compensatory proliferation and cell competition could account for the higher amount of PH3 positive cells within DAxud1-expressing domain. However, reduced BrdU incorporation and high number of PH3 positive cells are observed in wing discs co-expressing DAxud1 and the apoptotic inhibitor p35, strongly arguing against both alternatives. It is suggested that increasing DAxud1 retards or blocks some stage during mitosis and consequently reduces proliferation. This blockage leads to a JNK-dependent apoptosis, a possibility supported by the following observations: decreasing JNK activity in heteroallelic combinations reduces the DAxud1 over-expression phenotype, a fraction of TUNEL positive cells express PH3 and the pro-apoptotic effects of DAxud1 are suppressed by Cdk1 expression. In addition, the expression of E2F/Dp or Cdk2, which accelerate the G1/S transition, cause stronger cell death and higher accumulation of PH3 positive cells. This is likely a consequence of more cells reaching mitosis, and them becoming affected by the blocking activity of DAxud1. In rescue experiments, expression of Cdk1 is unlike to cause changes in Cyclin B or Cdc25/String phosphatase, suggesting that a CycB/Cdk1-independent process is the target of DAxud1 activity. Analysis of Cdk1 function in Drosophila imaginal cells, using the hypomorph temperature-sensitive allele Cdc2E1-24 has shown that reduction in its activity leads to mitotic blockage. However cdc2E1-24 cells progress to DNA replication, endoreplicating their DNA and acquiring higher cell size. In the current results, DAxud1 over-expression likewise impairs mitotic progression but no endoreplication is detected. A possible explanation is that DAxud1 over-expression represses very effectively Cdk1 activity, reducing Cdk1 function below the levels reached by the Cdc2E1-24 allele. In this manner, these cells became blocked in mitosis without progressing to endoreplicative cycles. This reasoning also explains the inability of CycB to rescue DAxud1 over-expression and the exclusive role of Cdk1 to revert the DAxud1 over-expression phenotype. Finally, the effects of DAxud1 over-expression are restricted to mitotically active tissues, as no significant apoptosis or differentiation defects are detected when DAxud1 is expressed, for example, in cells posterior to the second mitotic wave domain in the eye disc (Glavic, 2009).

The observed restriction imposed by DAxud1 upon mitosis progression implies that cells where the activity of this gene is decreased should exhibit a proliferative advantage. This is indeed what was observed in cells after knocking down DAxud1, which increases proliferation as detected by BrdU incorporation. These mutant cells also display a reduction in cell size, which finally result in a smaller wing posterior compartment size. The reduction in cell size could be a consequence of reduced length of G1 and/or G2, or, alternatively, it might imply a function of the protein in the signalling pathways that regulate cell size. The mechanisms by which DAxud1 influences cell size were not addressed (Glavic, 2009).

It is remarkable that the reduction of DAxud1 expression confers cells a proliferation advantage, as observed in twin analysis. Thereby, cells with decreased levels of DAxud1 behave as cells with increased mitogenic potential. This behaviour, together with the consequences of increasing DAxud1 levels, lead to the hypothesis that DAxud1 might antagonize the effect of pro-proliferative signals in proliferating cells. In this manner, DAxud1 might act as a sensor of pro-proliferative inputs within the cell that imposes a restriction of progression through mitosis. When over-expression of DAxud1 breaks this balance cells triggers JNK-dependent apoptosis. Recently, Gingras (2007) described the transcriptional activator ability and individual knockouts for the orthologues of DAxud1 in mice (CSRNP-1, -2 and -3). That study did not detect any obvious effects on mouse development, hematopoiesis or T cell functions. Interestingly, CSRNP-1 was cloned from IL-2 treated T-cells, a general pro-mitogenic signal for these cells. The apparent discrepancy between the lack of phenotypes in knockout mice and the requirement of Drosophila DAxud1 might rely on redundancy among proliferation control mechanisms in mice. No tumoural behaviour was detected in cells with lower DAxud1 levels, as might be expected from the proposed function of human AXUD as a tumour suppressor (Ishiguro, 2001). The proliferation advantage of cells with lower levels of DAxud1 in genetic mosaics is compatible with a facilitating role in tumour progression, although to generate hyperplasic proliferation other priming pro-tumourigenic events must occur in the cell. The effects of DAxud1 in cell viability; cell competition and cell cycle progression identified in imaginal cells should open new avenues to understand the function of this family of related proteins during development and cancer progression (Glavic, 2009).


Search PubMed for articles about Drosophila Axud1

Gingras, S., Pelletier, S. Boyd, K. and Ihle, J. N. (2007). Characterization of a Family of Novel Cysteine- Serine-Rich Nuclear Proteins (CSRNP). PLoS ONE 2 (8) e808. PubMed ID: 17726538

Glavic, A., Molnar, C., Cotoras, D. and de Celis, J. F. (2009). Drosophila Axud1 is involved in the control of proliferation and displays pro-apoptotic activity. Mech. Dev. 126(3-4):184-97. PubMed ID: 19084594

Ishiguro, H., et al. (2001). Identification of AXUD1, a novel human gene induced by AXIN1 and its reduced expression in human carcinomas of the lung, liver, colon and kidney. Oncogene 20: 5062-5066. PubMed ID: 11526492

Biological Overview

date revised: 20 February 2010

Home page: The Interactive Fly © 2009 Thomas Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.