Page 125 - 82_02
P. 125
The new levels of redox regulation of S-adenosylmethionine synthesis
liver disease (4,13,79,80). Moreover, the Mat1a-/- mouse SAM is a small compound that can be transported
developed severe hepatic pathology with age, most of through the nuclear pore, but this structure limits the size
these mice showing hepatocellular carcinoma (81). In of the proteins that can be translocated from one side to
contrast, most patients with hypermethioninemia due to another of the nuclear membrane to near 50 kDa. This is
mutations in this same gene either do not develop
symptoms or those are neurological, including a few cases almost the size of MAT a-subunits and is far below that of
showing signs of demyelination by magnetic resonance MAT I (180 kDa). Thus, understanding how MATa1 is
imaging (MRI), as explained in a previous section distributed between the nucleus and the cytoplasm was our
(70,71,82). Therefore, in our opinion, there was a need to next objective. Many nuclear proteins present localization
reexamine the tissue expression profile of MAT1A. Using signals (NLSs) in their sequence (84), others contain
real-time RT-PCR this analysis was performed in rat conformational signals that mimic the classical NLSs (85),
tissues, where expression was detected in almost all of and in other cases binding to proteins including
those examined. Several expression levels were found: 1) appropriate signals is preferred (piggy-back)(86). Among
high, shown in liver; 2) medium, exhibited by pancreas NLSs, the best-known or classical signals are mono- and
and lung; and 3) low, found in the rest of the tissues bi-partite sequences rich in basic residues, for whose
examined (83). This same pattern was further confirmed identification several bioinformatic tools are available.
by western blot using an anti-MATa1 antibody developed Using this software we were unable to identify a classical
in the last years in our laboratory using highly purified
preparations of the protein (64,83). NLS in the rat MATa1 sequence, and hence the crystal
structure was utilized to determine the existence of
4.7. MATs are cytoplasmic and nuclear proteins exposed clusters of basic residues that could mimic such
NLSs. Our analysis revealed that the C-terminal of
Localization of MATs to the cytoplasm was a well-
established characteristic, based mainly in traditional MATa1 is abundant in basic residues, and those exposed
subcellular fractionation studies and activity to the surface were selected for site-directed mutagenesis.
measurements. However, lack of good antibodies The subcellular distribution of the mutants was analyzed
precluded confirmation of this statement for a long time. by confocal microscopy and subcellular fractionation and
Again, using the new available antibody and the combined results allowed the identification of two
immunohistochemistry we were able to detect MATa1 in overlapping areas in the monomer structure that were
both the cytoplasm and the nucleus, although nuclear involved either in nuclear localization or cytoplasmic
localization was preferred in extrahepatic rat tissues (83). retention (83).
This discovery led to new questions regarding whether
there was nuclear MAT activity and which were the Subcellular localization was independent of MAT
isoenzymes present in that subcellular compartment. The activity, since an inactive mutant prepared on a residue
answer to these points required the development of involved in methionine binding (F251D) showed the same
improved subcellular fractionation protocols and activity distribution than the wild type protein (83). Moreover,
measurements that allowed not only detection of the small deletions at the C-terminal end, similar to those found in
amount present in the nucleus, but also, and more some patients showing demyelination (lacking 45
important, to guarantee lack of contamination from the residues), favored nuclear localization of the truncated
cytoplasm. In order to detect nuclear MAT activity, a
highly concentrated nuclear extract from a whole rat liver MATa1 (83). Additionally, it was demonstrated that
was required, together with 4-fold increased levels of the nuclear localization of MATa1 correlated with increased
radioactive tracer and longer reaction times. In parallel, levels of the trimethylation of lysine 27 in histone 3
LDH activity was measured to ensure complete (me3K27H3) (83), a well-known epigenetic repression
elimination of the cytoplasm, and even more critical, the mark. In contrast, other histone methylations examined
activity measurements (LDH and MAT) were carried out
in the supernatant before nuclear protein extraction to were increased by overexpression of MATa1,
ensure that the detected activity was specific of the nuclear independently of the subcellular distribution observed for
compartment. Following this protocol, the specific MAT the mutants.
activity detected in the nucleus was approximately 250-
fold lower than in the cytoplasm (83). Regarding the In 2011, MATa2 was also found in the nucleus during
nuclear isoenzymes, AGFC followed by dot-blot detection a proteomic analysis in search for protein-protein
of MATa1 showed the protein as tetramers and interactions of the MafK transcription factor (87). This
monomers, the amount of MAT I being small (83).
Therefore, the low nuclear MAT activity measured study showed MATa2 acting as a transcriptional
correlated with the presence of equivalent levels of MAT I corepressor of MafK on the heme oxygenase 1 (HO-1)
in this compartment.
gene. Moreover, a role for MATß and MAT activity in this
4.8. Overlapping areas of the C-terminal end are involved repression system was described. This manuscript together
in the subcellular distribution of MATa1 with an Editorial in the same issue of Molecular Cell
(87,88), supported the hypothesis, previously postulated by
our group (83), suggesting that MATs are transported to
nuclear locations where SAM synthesis is needed, whereas
organelles such as the mitochondria use transport systems
to obtain the compound from the cytoplasm.
4.9. Nuclear accumulation of MATa1 is induced during
@Real Academia Nacional de Farmacia. Spain 239
