Elsevier

Gene

Volume 295, Issue 1, 24 July 2002, Pages 79-88
Gene

Genomic organization of the human mi-er1 gene and characterization of alternatively spliced isoforms: regulated use of a facultative intron determines subcellular localization

https://doi.org/10.1016/S0378-1119(02)00823-5Get rights and content

Abstract

mi-er1 (previously called er1) is a fibroblast growth factor-inducible early response gene activated during mesoderm induction in Xenopus embryos and encoding a nuclear protein that functions as a transcriptional activator. The human orthologue of mi-er1 was shown to be upregulated in breast carcinoma cell lines and breast tumours when compared to normal breast cells. In this report, we investigate the structure of the human mi-er1 (hmi-er1) gene and characterize the alternatively spliced transcripts and protein isoforms. hmi-er1 is a single copy gene located at 1p31.2 and spanning 63 kb. It contains 17 exons and includes one skipped exon, a facultative intron and three polyadenylation signals to produce 12 transcripts encoding six distinct proteins. hmi-er1 transcripts were expressed at very low levels in most human adult tissues and the mRNA isoform pattern varied with the tissue. The 12 transcripts encode proteins containing a common internal sequence with variable N- and C-termini. Three distinct N- and two distinct C-termini were identified, giving rise to six protein isoforms. The two C-termini differ significantly in size and sequence and arise from alternate use of a facultative intron to produce hMI-ER1α and hMI-ER1β. In all tissues except testis, transcripts encoding the β isoform were predominant. hMI-ER1α lacks the predicted nuclear localization signal and transfection assays revealed that, unlike hMI-ER1β, it is not a nuclear protein, but remains in the cytoplasm. Our results demonstrate that alternate use of a facultative intron regulates the subcellular localization of hMI-ER1 proteins and this may have important implications for hMI-ER1 function.

Introduction

We previously described the isolation and characterization of cDNAs for the mesoderm induction early response 1 (mi-er1) gene from Xenopus (Paterno et al., 1997) and from human sources (Paterno et al., 1998). Xmi-er1 was isolated as a novel fibroblast growth factor (FGF)-regulated immediate-early gene from Xenopus embryonic cells induced to form mesoderm (Paterno et al., 1997). Characterization of the cDNA and protein product revealed that XMI-ER1 is a nuclear protein with the ability to function as a transcriptional regulator. In addition, it possesses a signature SANT domain characteristic of a number of proteins involved in both transcriptional activation and transcriptional repression, including N-CoR, TFIIIB and the MYB oncoprotein (Aasland et al., 1996). Subsequent sequence analysis also revealed the presence of an ELM-2 domain; this domain was first described in egl-27, a gene that has a critical role in embryonic patterning in Caenorhabditis elegans (Solari et al., 1999).

XMI-ER1 protein is expressed ubiquitously in the Xenopus embryo, however its subcellular localization is developmentally regulated (Luchman et al., 1999). XMI-ER1 protein is retained in the cytoplasm of the early Xenopus embryo, even though it possesses a strong nuclear localization signal (NLS) (Post et al., 2001). At mid-blastula, it first begins to appear in the nuclei of presumptive mesodermal cells, and by early gastrula, several hours later, XMI-ER1 can be detected in all nuclei (Luchman et al., 1999). The mechanism involved in regulating nuclear translocation of XMI-ER1 is currently unknown.

The human orthologue of mi-er1 was isolated from a testes cDNA library (Paterno et al., 1998) and the encoded protein displayed 91% overall similarity to XMI-ER1, with 100% identity in the SANT domain and a completely divergent carboxy (C)-terminus (Paterno et al., 1998). Dot blot analysis of polyA+ RNA revealed ubiquitous but extremely low level expression in normal human tissues. Breast carcinoma cell lines and breast tumours, on the other hand, showed elevated levels when compared to normal cell lines and tissue (Paterno et al., 1998).

Given the potential role of mi-er1 as a novel transcription factor in the regulation of embryonic development and its association with the neoplastic state, we have further characterized the genomic structure, expression pattern and protein isoforms of the human mi-er1 gene. In this report, we show that hmi-er1 is a complex gene with multiple transcripts arising from use of alternate promoters, exon skipping, facultative intron usage and differential polyadenylation. We also describe a novel hMI-ER1 isoform, hMI-ER1β, containing a C-terminus distinct from that previously reported (Paterno et al., 1998), that arises from use of a facultative intron and that targets this isoform to the nucleus.

Section snippets

PCR cloning of hmi-er1 cDNAs

Overlapping segments of hmi-er1 cDNAs were amplified by 5′ and 3′ RACE as described (Paterno et al., 1998), using Marathon-ready testis, mammary, ovary or mammary carcinoma cDNAs (Clontech, Inc.) as templates and adaptor-specific forward or reverse primers in combination with the appropriate hmi-er1-specific forward or reverse primers. PCR products were cloned into pCR2.1, using the TOPO TA cloning kit (Invitrogen Life Technologies) and over 50 clones were sequenced on both strands with the

A single copy human mi-er1 gene gives rise to multiple transcripts

We previously reported the cloning of a human orthologue of Xmi-er1 (Paterno et al., 1998), an FGF-inducible immediate-early gene (Paterno et al., 1997). In the present study, we have employed a PCR-based strategy to identify, clone and characterize variant forms of human mi-er1 cDNAs. Using forward and reverse primers designed against various regions of our original published hmi-er1 sequence (Paterno et al., 1998) in combination with cDNAs from a number of human tissues and cell lines, we

Discussion

Our determination of the human mi-er1 gene structure revealed that, like most other vertebrates genes (Hawkins, 1988), the exon size was small, with almost all being less than 160 bp, while the intron size is considerably larger, averaging several kb. The one exception to this is the last exon, exon 16, which is 1.7 kb in size. While the last exon in a gene is usually the longest (Berget, 1995), the average size is 600 bp (Hawkins, 1988). In fact, in the 4.8 kb transcripts which include the

Acknowledgements

This work was supported by a grant from the Canadian Institutes of Health Research to L.L.G. and G.D.P. The human genomic sequence data (EMBL/GenBank Accession number: AL139216 and AL500525) used for determination of the exon/intron boundaries and gene structure were produced by the Sanger Centre Chromosome 1 Mapping Group Sequencing Group at the Sanger Centre and can be obtained from ftp://ftp.sanger.ac.uk/pub/ftp://ftp.sanger.ac.uk/pub/human/chr1

References (36)

Cited by (0)

Sequence data from this article have been deposited in the GenBank Database under the accession numbers AF515446, AY124191, AY124194 (encoding hmi-er1 N3β); AF515447, AY124190, AY124193 (encoding hmi-er1 N2β); AF515448, AY124189, AY124192 (encoding hmi-er1 N1β); AY124186 (hmi-er1 N1α), AY124187 (hmi-er1 N2α), AY124188 (hmi-er1 N3α).

View full text