Photo courtesy of Nupur
Ghoshal
Page written by Aida Abraha, Ph.D.
INTRODUCTION:
Alzheimer's disease (AD) is, in part, characterized
by the polymerization of tau into paired helical and straight filaments
(PHF/SF) which together comprise the fibrillar pathology in degenerating
brain regions. Some of these tau filaments are known to undergo
phosphorylation and C-terminal truncation during the course of the disease.
Antibodies that recognize tau phosphorylated at Serine 396 and 404 (S396/404)
and truncated at Glutamate 391 (E391) do not stain normal control brains,
but do stain brain sections very early in the disease process. We modeled
these phosphorylation and truncation events by creating pseudo-phosphorylated
and deletion mutants derived from the longest isoform of the human tau
protein (ht40).
RESULTS:
Photo courtesy of Nupur
Ghoshal
Figure 1. Hippocampal sections containing the subiculum were taken
from the brain of a mild (MMSE = 26) AD case and reacted with monoclonal
antibodies (A) Tau-5, (B) AD2, and (C) MN 423. Hematoxylin counter
stain; Bar = 50µm.
Figure 2. Pseudo-phosphorylation enhances tau polymerization.
(A) Bar diagram of ht40 and mutant [396/404]S
to E, including the positions of the alternatively spliced
exons 2 and 3 (e2 and e3), the proline-rich region, and the four MTBRs
(m1-m4). The sites of pseudo-phosphorylation are approximated by
asterisks. The numbering of the molecule is based on the longest
human isoform of tau. (B) Representative polymerization reactions
using ht40 (o) and [396/404]S to
E (.) monitored
by changes in 90° laser light scattering (LLS). At the end of the
six hour polymerization reaction, filament formation was confirmed by electron
microscopy. Representative micrographs for (C) ht40 and (D) [396/404]S
to E are shown. Bar = 500nm.
Figure 3. C-terminal truncation at E391 enhances tau polymerization.
(A) Bar diagram of ht40 and deletion mutant D392-441
indicating the extent of the deletion. The results of the ELISA assay
confirming the immunoreactivity of D392-441
to MN 423 are also illustrated (+ vs -). (B) Representative
curves for polymerization reactions consisting of ht40 (o) and D392-441
(.) performed at room temperature and monitored by
laser light scattering for six hours. Filament formation at the end
of each polymerization reaction was confirmed by electron microscopy:
(C) ht40 and (D) D392-441. Bar = 200nm.
Discussion and Conclusion:
The formation of pathological tau filaments in neurofibrillary
tangles, neuropil threads, and tau positive plaques occurs early in the
neurodegenerative cascade and can be readily identified in AD brain through
the use of monoclonal antibodies such as Tau-5 (Fig 1A). In addition
to filament formation, other molecular markers can be used to identify
pathological changes in the tau molecule very early during the course of
AD. For example, tau protein that is abnormally phosphorylated at
Serine 396 and 404 can be identified by the phosphorylation-dependent monoclonal
antibody AD2. AD2 often reacts with pre-tangle neurons as well as neuropil
threads and NFTs in vulnerable brain regions prior to the onset of overt
dementia (Fig 1B).
Another molecular event purported to be involved
in the formation of the fibrillar pathology in AD brain is the abnormal
truncation of tau’s C-terminus. A specific monoclonal antibody, MN
423, recognizes tau truncated at E391 (Fig. 1C) in sections taken from
the same very mild AD case and brain region (subiculum) depicted in Fig.
1A & B. Like the abnormal phosphorylation illuminated by AD2,
truncation seems to appear early in the disease process and has also been
detected in pre-tangle neurons
Since it is nearly impossible to control the phosphorylation
of specific residues among the many potential phosphorylation sites of
tau, a pseudo-phosphorylation construct was created using site-directed
mutagenesis to substitute glutamic acid residues for Serine 396/404, thus
creating a [396/404]S to E double mutant
(Fig. 2A). Mutant [396/404]S to E
assembly, as assayed by LLS measurements, occurred at a greatly accelerated
rate when compared to that of wild type tau (Fig. 2B). The filaments assembled
by wild type ht40 and mutant [396/404]S to E
were seemingly identical to each other, appearing as flexible unpaired
SF-like polymeric structures ca. 15 nm in diameter by negative stain electron
microscopy (Figures 2C and D).
As indicated previously, the E391 truncation recognized
by the MN 423 monoclonal antibody is known to occur during the course of
AD pathogenesis, and our results (Fig. 1C) suggest that it is a relatively
early event. Accordingly, we created the D392-441
deletion mutant terminating at E391 for use in our in vitro studies (Fig.
3A). Using enzyme linked immunosorbent assay (ELISA), MN 423 only reacted
with D392-441 in vitro (Fig. 3A). When
assessed by LLS, the rate of polymer formation for this mutant was significantly
higher than wild type (Fig. 3B). Additionally, the average extent
of polymer formation by mutant D392-441, at
the end of the six-hour period, was also greater (Fig. 3B). Furthermore,
the filaments formed by wild type ht40 and D392-441
were morphologically similar (Fig. 3C and D).
In conclusion, the C-terminus of tau clearly inhibits
the assembly process; this inhibition can be partially reversed by site-specific
phosphorylation and completely removed by truncation events at various
sites from S320 to the end of the molecule (for more information see Abraha
et. al., 2000).