Regulation of mRNA stability via BRF1 and other AU-binding proteins

Steady state levels of mRNAs are determined by the rate of synthesis and degradation. A well-known cis-element conferring instability to mRNA is the so-called AU-rich element (ARE), which is present in the 3’ untranslated region (3’UTR) of many cytokines, chemokines, growth factors or proto-oncogenes. The ARE is recognized by a variety of ARE-binding proteins (AUBPs), which decide about the fate of the RNA. Multiple signaling cascades regulate the activity of the AUBPs. Butyrate response factor(BRF1), a Tis11 protein family member, was functionally cloned in our lab, as an ARE-mRNA destabilizing protein. However, not much is known about the mode of action of this protein and its physiological role. This thesis deals in a first part with the regulation of BRF1. Analysis of BRF1 protein sequence revealed multiple putative phosphorylation sites, where BRF1 activity could be regulated. Serine 92 (S92) was identified by coworkers as a protein kinase B (PKB/Akt) phosphorylation site. To confirm this finding in vivo a phospho-specific antibody was raised in rabbits. Using this antibody, phosphorylation of S92 could be corroborated in vivo. Further, it could be shown that ARE-containing mRNA is stabilized under conditions of phosphorylated BRF1. Inhibitor experiments indicated that S92 is not only phosphorylated by PKB, but also by at least one other kinase, probably from the ERK1/2 MAPK pathway. Apart from S92, also other sites seem to be phosphorylated, as arsenite-treatment of mouse fibroblasts (NIH3T3) or insulin-treatment of rat fibroblasts, which overexpress the human insulin receptor (HIRc-B), result in a protein phosphatase sensitive shift of BRF1 on SDS-PAGE. These additional phosphorylation events were analyzed in a second part. Comparison of the sequence of all three Tis11 protein family members revealed that S203 in BRF1 is homologous to the MK2 target site S178 in TTP. Indeed, S203 seems to be phosphorylated. Inhibitor experiments suggested that the ERK pathway might be involved. In addition, mutation of S203 renders BRF1 insensitive to PKB and
MK2 mediated inhibition in cotransfection ActD-chase experiments
and, therefore, seems to play a role in regulation of BRF1 activity.
Another way to control the activity of AUBPs is the regulation of
their localization in the cell. Therefore, in a third part, the question
was addressed, where the AUBPs HuR, AUF1p37, BRF1 and TTP are
localized in the cell and whether their localization might play a role
in regulating the activity of these proteins. The effects of activated
signaling cascades, known to stabilize ARE-mRNA, on AUBP
localization were investigated. In the case of p38 MAPK and PI3-K
stabilization of ARE-mRNA is, indeed, accompanied by translocation
of the stabilizing AUBPs HuR and AUF1p37 to the cytoplasm.
Surprisingly, PKB, a downstream kinase of PI3-K, does not affect
HuR and AUF1p37 localization, indicating that PI3-K regulates their
localization via another pathway. Probably, the effect is mediated via
PKC, as stimulation by TPA, a potent activator of PKC, does also lead
to cytoplasmic accumulation of HuR and AUF1p37. In addition, HuR
localization alters early during mitosis. At the onset of prophase, HuR
is exported to the cytoplasm, indicating, that HuR might play a role in
mitosis.
In the case of BRF1 and TTP no changes in localization could be
observed in response to external stimuli. These two proteins are
equally distributed in the cell, indicating that nuclear storage of these
proteins cannot account for their inactivation.
To shed light on the physiological role of BRF1, a cell line was
constructed with doxycycline-repressible BRF1 expression. This line
was used in the last part of this thesis, to investigate, whether BRF1
plays a role in cell cycle control. Overexpression of BRF1 did accelerate
progression of serum starved mouse fi broblast cells from G0/G1 phase
to S phase. Downregulation of BRF1 by siRNA, on the other hand, did
have the opposite effect. However, the effect was reproducible but only
marginal. Therefore, the project was discontinued at this stage.
Taken together, the data presented gives insight into the mechanisms,
linking extracellular signaling to mRNA stability. For BRF1 two sites
of regulation, namely S92 and S203 could be identifi ed. Further,
export of the two stabilizing AUBPs HuR and AUF1p37 seems to play
an important role in ARE-mRNA stability control. A dual role could be
assigned to the PI3-K signaling pathway: PI3-K inactivates the AREmRNA
destabilizing protein BRF1 by PKB mediated phosphorylation
at S92 and, via a PKB independent pathway, PI3-K exports HuR from
the nucleus, further stabilizing ARE-containing transcripts.