Introduction
A progressive neurological disease, which mainly affects the senile
population, Alzheimer’s disease (AD) is one of the most prominent form
of dementia [1]. The advances in molecular biology, structural
biology and neurobiology in the recent past have enhanced our
understanding of the pathological process which underlies this
progressive neurodegenerative disease, but still we do not have a
complete understanding of the molecular mechanisms associated with the
disease. An estimated 55 million people across the globe suffer from
Alzheimer’s and other related dementia, with no disease modifying drugs
or therapy still available [2]. The current strategy of treatments
revolves around providing symptomatic relief with cognition enhancement
drugs and early diagnosis with biomarkers [3]. Central to the
neuropathology of AD is the loss of synapse and synaptic plasticity,
which reduce the ability of individuals to make and store new memories
[4]. The two most prominent hypotheses which help explain this
cognitive decline are the amyloid cascade hypothesis and the tau
hyperphosphorylation hypothesis [5, 6].
Tau protein also called microtubule associated protein tau, is an
intrinsically distorted protein (IDP), which does not have a complete
secondary or tertiary structure when it’s in the unbound form. Encoded
by the MAPT gene located on chromosome 17, with an mRNA of 16 exons, tau
protein occurs in different isoforms in the human brain [7, 8].
Alternate splicing of the exons 2, 3 & 10 leads to the formation of six
different isoforms with chain lengths ranging from 352-441 amino acids
[9, 10]. The major role of tau protein in neurons is that they bind
and stabilize the cytoskeletal fragments called microtubules. This in
turn helps in the proper intracellular trafficking [11, 12]. Once
bound to the microtubules, the tau protein will have residue specific
secondary structure which stabilizes the binding [13, 14].
Hyperphosphorylation of tau protein have been attributed as one of the
main causes of AD because of its ability to destabilize the
tau-microtubule binding and induce the formation of neurofibrillary
tangles [15, 16].
The structure of the longest tau isomer with 441 amino acids can be
divided into three different domains, the projection domain, proline
rich domain and microtubule binding domain. The projection domain is the
region that is close to the N terminal and has two sub regions called
the N1 & N2 (1-165), which are not attached but projected away from the
microtubules. The proline rich domain extends from amino acid 166 to 242
having a high density of proline residues. The microtubule binding
region consists of four repeat regions (R1, R2, R3 & R4) extending from
residues 243 to 367, which binds to the microtubules and have certain
residue specific secondary structures, followed by the C terminal region
[17, 18].
Post translational modifications (PTMs) have been associated with the
pathophysiology of many intrinsically distorted proteins like tau
[19]. Alquezer and coworkers have summarized various PTMs on tau
protein where they have given a detailed review of all modifications
along with the probable modification sites [20]. The most widely
studied PTM associated with AD is tau hyperphosphorylation.
Glycosylation is another important PTM on tau, which has been shown to
be correlated with AD [21, 22]. Tau undergoes O-GlcNAcylation, which
involves an attachment of the β-N-acetylglucosamine (GlcNAc)
monosaccharide to either Serine or Threonine residues, and it has been
found that the O-GlcNAcylation levels have been reduced in AD brains.
Since both phosphorylation and O-GlcNAcylation takes place on
serine/threonine residues, the presence of one is thought to hinder the
other [23, 24]. Another interesting PTM is N-glycosylation of the
tau protein. It was found that tau protein obtained from AD brains have
been N-glycosylated but not those from the healthy brains [25].
N-glycosylation of tau was an unusual observation as the machinery
needed for tau N-glycosylation, i.e., endoplasmic reticulum and Golgi
are located on pathways for secreted proteins and usually not accessible
to tau protein, which is an intracellular protein [26-28]. It was
also observed that N-glycosylation takes place on asparagine residues
having sequence N–X–S/T (where X can be any amino acid except proline)
[29] and the three potential N-glycosylation sites of human tau
protein are Asn167, Asn359, and Asn410; numbering based on longest tau
isoform [26].
The fact that tau is N-glycosylated in AD brains and not in healthy
brains, suggested the possible involvement of this PTM in AD pathology.
It was also found that paired helical filament (PHF) tau when
deglycosylated, changed the structure from the folded state to straight
filaments [30]. A recent in vitro study conducted by Losev
and coworkers using site specific mutation of putative N-glycosylation
sites in tau protein expressed in Drosophila and SH-SY5Y cells,
suggested the involvement of the three potential N-glycosylation sites
in tau aggregation. SH-SY5Y cells and transgenic Drosophila expressing
tau protein with asparagine to glutamine mutants at the putative
N-glycosylation sites were generated, and it was found that
N-glycosylation at Asn167 & Asn359 enhances the tau aggregation, while
Asn410 reduces tau aggregation. They also found that expression of the
N359Q hTau mutant leads to amelioration of AD-like symptoms, reduced
neurodegeneration and importantly, N359Q is the only mutant that
improved the lifespan of the flies [31].
Even though these in vitro results suggest that N glycosylation
of tau protein at specific residues (Asn359) leads to tau aggregation,
the exact mechanism by which it occurs is still not fully understood.
Understanding the N-glycosylation induced effects on protein dynamics of
tau can help us gain some insights into misfolding patterns and some of
the important residue interactions. This in turn can help the scientific
community to further explore the pathogenic role played by
N-glycosylation in AD and identify new targets in the pathogenic
N-glycosylation pathway. Here, in this study we used molecular dynamics
(MD) simulations coupled with essential dynamics, free energy landscape
analysis (FEL) & residue interaction network analysis to evaluate the
effects caused by the N-glycosylation on Asn-359 residue, which forms
part of the microtubule binding region of the tau protein. Through this
study, we aimed to study whether N-glycosylation can induce tau protein
folding and what are the important residue specific interactions which
helps in the process. We hope that the information that we gain from
this study can be used in future research to build new anti-tau
aggregation compounds as disease modifying therapeutics against AD and
other related tauopathies.