To characterize the physiological role of Ebp1, we generated Ebp1 deficient mice exhibiting a gene trap introduced into intron 2 of the Ebp1
gene. The gene trap leads to deletion of all but the first 20 amino acids of Ebp1 which have no biological function on their own. The most striking feature of the Ebp1-/-
phenotype was the small size of the mice. It is of interest to note that inhibition of expression of the potato homologue of Ebp1 also results in a dwarf phenotype [4
pups were on average 25% smaller than their wild type and heterozygous littermates until Day 30, the approximate onset of puberty. Mice were well proportioned with no skeletal abnormalities other than a kinky tail in approximately 1/3 of the mice. An unusual feature of our in vivo
analysis is the fact that the reduced weight was recovered after the prepubescent stage, while Ebp1 is still normally transcribed and translated and therefore likely to have a regulatory role. Although we are unable to explain the mechanism underlying this recovery, we report this observation as a possible example of temporal restricted competence in growth regulation of specific cell types by Ebp1. This phenotype is similar to that observed for the OTX 1 gene, a homeobox containing gene that in vitro
activates transcription of GH, FSH, and LH [28
It is of interest that the phenotype of Ebp1 knock out animals also resembles that of the IMP1
knock out mice [29
]. IMP1 belongs to a family of RNA binding proteins implicated in mRNA localization, turnover, and translational control. These mice also exhibit a dwarf phenotype with reduced proliferation of knock out MEFs. In these mice, growth retardation is related to a hypoplastic small intestine. In contrast, Ebp1 mice exhibit a transient dwarfism and their intestines are both grossly and histologically normal at 10 weeks. However, we were unable to assess intestinal anatomy and histology at earlier time points, and it remains possible that abnormalities occur when the weight differences are most marked. Ebp1 has also been shown to be an RNA binding protein with a sigma- 70 like motif. This motif has been identified as a new RNA binding domain unique to the superfamily of Imp4 or Brix proteins that are required for ribosome biogenesis [30
]. A role for Ebp1 in ribosomal biogenesis is supported by studies indicating its localization in the nucleolus, its association with 5 S ribosomal RNA, and its complexing with proteins such as B23 which are involved in ribosome processing[19
]. Whereas the significance of RNA binding proteins in spatial and temporal control of gene expression during development is evident in invertebrate organisms, the function of RNA binding proteins during mammalian development is less clear. With the exception of FMRP, Dazla, and Nova-1, which are associated with mental retardation, azospermia, and neuronal viability respectively [31
], the majority of the mammalian RNA binding proteins are constitutively expressed and exhibit a relative broad RNA-binding specificity. Further work is needed to uncover the role that Ebp1's RNA binding activity has in the development of this phenotype and the target RNAs that specifically interact with Ebp1.
The molecular pathways leading to dwarfism are likely to involve several mechanisms. As reduced growth was manifested in the MEFs from Day 12.5 embryos, we used these cells to measure changes in gene expression associated with loss of Ebp1. Global gene expression analysis indicated that expression of IGF-1 and the IGF binding proteins IGFBP3 and 4 was decreased. The in vivo
growth promoting role of the IGFs has been demonstrated conclusively from the dwarfing phenotypes observed after targeted mutagenesis of the mouse Igf1
]. IGF-1 functions both during embryonic and post natal development. IGF-1 mutants exhibit a birth weight that is 60% of normal and the difference between their size and that of wild type mice increases progressively with age [35
]. The Ebp1 phenotype is not as severe as that observed in IGF-1 deficient mice. The function of IGFBPs in cell growth is more complex. However, one role is to sequester and stabilize IGF-1 [36
]. Thus, decreases in IGFBPs may result in destabilization of IGF-1 and may be partially responsible for lower levels of serum IGF-1.
We found that MAPK was constitutively phosphorylated in MEFs derived from knock out mice. This constitutive phosphorylation was associated with the overexpression of SOS-1, an upstream activator of the Ras -MAPK signal transduction pathway. It is possible that overexpression of SOS-1 led to increased basal phosphorylation of MAPK. In contrast, both the basal and HRG-induced activation of the MAPK pathway is increased in EBP1
transfected breast cancer cell lines [11
]. The reasons for these apparently discordant findings may be related to the ligand used, the differential expression of ErbB receptors in breast cancer cells versus fibroblasts, and the cellular background.
We have previously shown that inhibition of Ebp1 expression results in increases in phospho AKT in human breast cancer cell lines, whereas ectopic expression of ebp1
inhibits HRG induced AKT activation [15
]. Here, we found that Ebp1 had no effect on AKT phosphorylation after EGF stimulation of MEFs. However, AKT was constitutively phosphorylated in mammary epithelial cells of knock-out mice. The mechanism of the increase in AKT phosphorylation is unclear. Paradoxically, Ahn et al [16
] have shown the binding of Ebp1 to nuclear AKT increases AKT phosphorylation [37
] in PC-12 neuronal cells. It is not known if Ebp1 binds AKT in normal mammary epithelial cells or affects its phosphorylation. These studies point up the importance of tissue context in the activation of specific signaling molecules. As activation of AKT is important in development of mammary adenocarcinoma [38
], we are currently monitoring breast tumor incidence with age in Ebp1 knock out mice. We have not yet seen increased tumor incidence in a small cohort of mice that have been carried until one year.
Finally, our laboratory has demonstrated that ectopic expression of Ebp1
results in decreased expression of AR and AR target genes and inhibition of growth of prostate cancer cells both in vivo
and in vitro
]. In the current study, deletion of the Ebp1
gene led to increased expression of AR in prostates of adult mice and upregulation of the metastasis associated gene AGR2
. AR plays a key role in the development and progression of prostate cancer, and we anticipate that the Ebp1
knockout mouse may have a hyperactivated AR signaling axis that may result in increased incidence of prostate cancer. Also with regard to prostate cancer, decreased expression of IGFBP-3 is correlated with increased risk of developing prostate cancer[39
]. Western blot analysis verified that Ebp1
knockout led to reduced protein levels of IGFBP-3 in liver and kidney. We were unable to detect IGFBP-3 in prostates of either wild type or knock out mice, but studies are ongoing to assess if there are changes in IGFBP-3 expression in the prostates of knock out mice. We are also monitoring the incidence of prostate cancer of the Ebp1-/-
mice as they age. Nevertheless, current studies, together with our previous reports, strongly suggest that Ebp1, as an endogenous regulator of ErbB-AR crosstalk, may regulate expression of a series of genes that are involved in aggressive prostate cancer growth.