Abstract |
Mammalian glutamate dehydrogenase (GDH), an enzyme central to cellular
metabolism, is present in all living organisms. During evolution, GDH acquired
mitochondrial targeting and complex allosteric regulation. In mammals, GDH is
among the most abundant mitochondrial proteins (constituting up to 10% of matrix
proteins), although there are also suggestions for the extramitochondrial presence of
this protein. In the human, GDH exists in two isoforms, hGDH1 and hGDH2,
encoded by the GLUD1 and GLUD2 genes respectively, with distinct tissue and
cell-type expression profile. Both isoproteins are predicted to have an unusually
long N- terminal mitochondrial targeting signal (MTS), containing 53 amino acids
(N53). This MTS is proteolytically removed upon translocation to mitochondria.
Even though the two human GDHs share, in their mature form, high homology,
their MTS are more divergent.
In the first part of this PhD thesis, we studied the subcellular targeting of the two
hGDH isoforms in cell lines. To this end, we constructed expression vectors, in
which hGDH1 or hGDH2 was fused with the enhanced green fluorescent protein
(EGFP) and used them to transfect human and mammalian cell lines. Cotransfection
experiments using organelle-specific markers, followed by confocal
microscopy, revealed that hGDH1 or hGDH2 co-localized with the mitochondrial
marker DsRed2-Mito and to a lesser extent with the endoplasmic reticulum (ER)
marker DsRed2-ER. Subcellular fractionation and immunoblotting detected two
GDH-EGFP specific bands: 1) a ~90 kDa band associated with the mitochondrial
fraction, corresponding to the mature hGDH-EGFP following removal of the N53,
and 2) a ~95 kDa band mainly associated with the ER-containing cytosolic fraction,
corresponding to the precursor hGDH-EGFP. Deletion of the N53 abolished
mitochondrial targeting, thus confirming the role of N53 as a mitochondrial
targeting signal.
In the second part of the study, we sought to characterize the N53 sequences of
hGDHs, to understand their structure and the properties that are important for
efficient mitochondrial targeting. The main characteristics of the N53 of hGDHs are
the positive charge and the tendency to form two α-helices (α1: N1-10 and α2:N16-
32), α1 being more amphiphilic. The first helix, α1, is absolutely necessary for
mitochondrial targeting, since selective deletion of the α1 helix abolished the mitochondrial import of hGDH2. Moreover, α1 but not α2, displayed autonomous
mitochondrial-targeting capacity, being able to direct non-mitochondrial proteins
into mitochondria. However, the efficient targeting of hGDH2 required the
synergistic interaction of both helices α1 and α2, while α1 has the leading role.
Those experimental findings in human cell lines were confirmed by import
experiments in isolated yeast mitochondria, carried out by our collaborators in
IMBB. These experiments also revealed that hGDH2 is imported and/or processed
faster than hGDH1 in yeast mitochondria, supporting the concept of an enhanced
mitochondrial targeting capacity of hGDH2. Moreover, the proteolytic removal of
the N53 in the mitochondrial matrix was not found to be essential for the import of
hGDHs in mitochondria.
The last part of the study sheds light into the evolution of the GDH presequence
from lower eukaryotic organisms to mammals and human. With the use of
bioinformatic tools and in silico predictions, we suggest that the first evidence for
the presence of a GDH cleavable mitochondrial presequence is found in the
kingdom of ciliophora. In mammals, GDH presequence evolved into a larger, more
complex and highly efficient targeting signal. Additionally, we propose that the α1
helix in mammals is more conserved than the α2 helix. In order to test some of the
in silico predictions, we performed experimental studies looking for the
mitochondrial targeting capacity of GDH presequences from different eukaryotic
model organisms. Specifically we showed that the GDH presequences from T.
thermophila (ciliate), C. elegans (roundworm), D. melanogaster (fly) and X. laevis
(frog) could target the non mitochondrial proteins EGFP and DHFR into
mitochondria. However, the GDH presequence of T. thermophila was not sufficient
to target efficiently hGDH2 into mitochondria.
In conclusion, the present study provides important insights into the subcellular
targeting and the mechanisms that govern the mitochondrial transport of a major
mitochondrial matrix protein and contributes into the understanding of the function
and evolution of the mitochondrial targeting signals.
|