Leslie Tetteh3SBS671W1298035Molecular dynamics simulations of
the unfolding of mutant prion variants at elevated
temperaturesAbstractThe goal of this investigation was to analyse
the unfolding pattern carried out by mutant variants of the WT
prion protein as well as the WT itself at 300k and 500k. The mutant
variants used were T188A/K and E219K, all of which are associated
with Creutzfeldt-Jakob disease. The T188 mutants are known to
increase the risk of pathogenesis, whilst E219K is known to guard
against it, prolonging conversion time.It was found that several
different types of partially unfolded energy states seemed to be
occupied, and that these were relatively stable, causing long
periods of fluctuation about a point in terms of RMSD. The most
unstable mutant was found to be T188K, and there was a sharp
contrast between the unfolding pattern of E219K and the T188
mutants. From the literature search there was also found to be a
possible mode to formation of an initial scrapie seed under
partially denaturing conditions, which could be determined by
experiment.IntroductionMalconformed versions of the wild-type prion
protein (PrPc), commonly known as the scrapie form (PrPSc) act as
infectious agents and are the cause of a number of diseases in
mammalian species such as scrapie in sheep and cattle, whilst in
humans they cause Gerstman Straussler-Scheinker syndrome (GSS) and
Creutzfeldt-Jakob disease (CJD). An initial scrapie seed causes a
change in the conformation of the WT-peptide such that
fibrillogenesis occurs; these fibrils cause tissue damage and cell
death and are toxic to humans and other large mammals, eventually
leading in the long-term to widespread brain damage, ataxia and
death (Chiti 2009, Otvos 2002). Pathogenesis arises sporadically in
approximately 85% of human disease cases; genetic inheritance
accounts for around 15% of all cases; and less than 1% can be
attributed to exposure to infected tissue (Lawson et al 2005).
The scrapie form of the prion protein PrPSc has been shown to
contain a secondary structure of about 43% beta structure and 30%
alpha helix, which is relatively insoluble to detergents whilst
being only partially prone to digestion by proteinase K. In
contrast the wild-type prion protein PrPC is a largely -helical
structure consisting of 42% alpha-helices and 3% beta-sheet
content, which is soluble in detergents and susceptible to complete
digestion by Proteinase K. Digestion of PrPSc by Proteinase K
leaves a 27-30kDa protease resistant core (Stahl 1993, Lawson
1995).
The WT-peptide is a sialoglycoprotein usually found bound to the
cell membrane by means of a GPI anchor (Pan 1993). It is composed
of two distinct sections defined as the C-terminal and the
N-terminal domains. The C-terminus has been crystallised by use of
NMR spectroscopy and has been found to be a globular structure
composed of three alpha helices, H1 H2 and H3, and an anti-parallel
beta sheet formed by beta-structures S1 and S2 which flank the H1
helix. The N-terminal region has been found to be unstructured
under crystallising conditions, whilst most of the SNPs known to
affect protein properties with regard to amyloid formation have
been found in the C-terminal globular domain (Riek 1996, Riek 1997,
Riek 1998).
PrPC is formed at an initial translated length of 253 amino
acids. The GPI anchor is attached to the globular part of the
protein following cleavage of a carboxy-terminal signal sequence,
whilst a 23 amino-acid signal sequence in the N-terminal region of
PrPC targets it to the endoplasmic reticulum for processing. Post
translational modification results in a mature length of 208 amino
acids for huPrPC (Otvos 2002). Following glycosylation, PrP is
localized to detergent resistant microdomains of the cell membrane
(Vey et al 1996). The process of fibril elongation leading to the
formation of amyloid arises from a conformational change of the
secondary structure within PrPC. This changes its biochemical
properties and leads to oligomerization from an initial soluble
PrPSc seed (Ma 2002, Baumketner 2008). The fibril formed via this
process eventually elongates and becomes an insoluble fibrillar
growth (Caughey 2003). The susceptibility of this fibrillar
formation to pressure is thought to allow for breakage and the
formation of new fibrillar seeds (Cordeiro 2004).
The sequence structure used for dynamic simulations was the
C-terminal domain of the protein residues 125-228. It has been
previously determined that the N-terminal region of the protein is
not found within most protease resistant amyloid aggregates
(Prusiner 1982), and NMR performed on HaPrP amyloid aggregates has
shown that it is the C-terminal domains which associate to form the
fibrils structured core (Tycko 2010). Although the N-terminal
region does not affect the structure of the protein, it appears to
carry out biological functions and does affect the biochemical
properties of PrPC.
The N-terminal region within hPrPC contains within it four
tandem copper-binding repeats with sequence (PHGGGWGQ)4 which
allows it to bind Cu2+ ions with high specificity, and also to
interact with other cations such as calcium within the body (Pushie
2007, Campioni 2010). This Copper-binding sequence, termed an
octarepeat, is located between residues 51-91 in humans and is
highly conserved amongst all mammalian species implying that it is
essential to the proper function of PrPC (Rheede 2003, Mastriani
2000). X-ray crystallographic data has been obtained showing the
copper bound form of the OR repeat sequence (Burns 2002).
Studies on murine deletion mutants rPrP51-90 and rPrP32-121 have
shown that whilst this N-terminal region is not needed for
structure, deletion does affect the stability of the protein making
for faster fibrillar formation with a shorter lag phase, a higher
susceptibility to pressure, and a lower unfolding temperature
(Cordeiro 2005). The effect of pressure may explain why fibrillar
formations are so prone to breakage, and the shorter lag phase
offers support for the theory that fibril break-up leads to the
initiation of new seeds (Cordeiro 2004). The amount of beta-sheet
structure formed upon subjection to fibrilising conditions has also
been shown to increase with the amount of the N-terminal region
deleted;19% in rPrP23-231; 25% in rPrP51-90 and 27% rPrP32-121, a
trait which was also expressed in the Syrian hamster recombinant
rPrP90-231 (Torrent 2004). The evidence thus demonstrates that the
N-terminal region of PrP protects against fibrillogenesis by
preventing beta-sheet formation.
An explanation for this difference in stability after the loss
of the N-terminal region can be offered by the study of cation-pi
interactions within the Prion protein. The strength of cation-pi
bonding within the Prion Protein has been shown to be on a par with
the energies demonstrated by other non-covalent bonds such as Van
Der Waal interactions, hydrogen bonding and salt bridges, measuring
at energies of between 10 and 150kJ/mol (Priyakumar 2004). These
electrostatic bindings are created between the electron-rich pi
bonds in aromatic and other pi-bonded residues and the positive
charge present on cations. On average three cation-pi bonds were
found to exist within the helix regions and two within beta sheet
structures (George 2013). Given that the amino-terminal region is
shown to have a high valence for binding cations this may be an
important finding with regard to the loss of the N-terminal regions
in most fibrillar formations.
The effect of hydration and packing on the tendency of PrPc to
form beta-sheets and on its natural folding stability was carried
out by Cordeiro et al in 2004 and a molecular dynamics study on the
same subject was carried out by Simone et al in 2005. This study
demonstrated that a less hydrated PrP structure is inherently more
susceptible to pressure, which also offers some explanation as to
the difference in stability of the rPrP deletion mutants, which
have a less hydrated structure. However the chemical treatment of
the deletion mutants with denaturing agents such as urea showed no
significant differences in fibrillogenesis, indicating that the
chemical mode of action leading to amyloidosis still centred around
the C-terminal region of the protein (Cordeiro 2005).
The C-terminal region of PrPC is very well conserved amongst
mammalian species and it has previously been found that the vast
majority of single nucleotide polymorphisms which lead to genetic
inheritance/resistance against prion diseases are located within
the C-terminus (Billeter 1997). This carries within it the inherent
notion that initiation sites for conversion to PrPSc are likely to
be found within this region, whilst the high level of conservation
across mammalian species implies that these initiation sites are
likely to be similar in different forms of PrPC.
Point mutations that are known to lead to increased likelihood
of pathogenesis are
GSSP102L, P105L, A117V, G131V, Y145X, F198S, D202N, Q217R,
M232T
CJDD178N, V180N, T183A, T188A/K/R, E200K, R208H, V210I
Whilst E219K and M129V are examples of mutations that lead
increased resistance to sCJD.
As can be seen in the case of GSS and CJD, within the field of
Prion diseases or transmissible TSEs there are noticeable strain
types, which can be identified by the incubation period; time
between first exposure and development of symptoms, and the pattern
of lesions which develops within the brain (Bruce 2003).
Polymorphisms can account for some of the difference, but it has
been shown in the studies of different clinical phenotypes of human
TSEs, sCJD and sporadic Fatal Familial insomnia, that other
variances are also present. Western blotting of the two
aforementioned PrPSc types exposed differences that suggested
glycan heterogeneity (Pan et al 2001).
The human prion protein has two asparagine linked glycosylation
sites at 181N and 197N, which allow mannose (a C2 epimer of
D-glucose) based oligosaccharide chains to be attached to the
protein (Vey 1996). The WT peptide can be either un-, mono-, or
diglycosylated when found in nature and experiments carried out by
Rudd et al in 1999 on Syrian hamster PrPc and PrPSc identified more
than 50 different polysaccharide chains which can be attached to
the two sites. The unique glycoform ratio encountered in hPrPSc
first provided the link between variant CJD and Bovine Spongiform
Encephalopathy (Lawson 2005).
The differences encountered in the conformations of the Prion
protein based on the source of the PrPSc offers some support for
the protein-only propagation theory first expounded upon by Alper
in 1967, however as evidenced by laboratory testing the type of
strain produced also appears to be dependent on the available pool
of PrPc. It was shown by Priola and Lawson in 2001 that the
introduction of a species specific residue into muPrPC induced a
conformational change that led to proteinase K resistance only with
unglycosylated muPrPC. Korth et al also demonstrated in 2000 that
mutations in the glycosylation consensus sequence resulted in a
change in hPrPC distribution pattern within cells containing ScN2a.
This alteration affects the pattern of hPrPSc formation.
It has also been demonstrated via treatment with tunicamycin, an
inhibitor of N-linked glycosylation in eukaryotic organisms, and
the alteration of the 181N and 197N consensus sites that
unglycosylated PrPC molecules have been found that reproduce some
of the characteristics of PrPSc such as detergent insolubility and
partial proteinase K resistance (Esko 2009, Lehmann 1997). These
studies provides some evidence that glycosylation confers some
resistance to the formation of the scrapie form, which further
supports the model of templated assembly put forth in the
protein-only model (Prusiner 1982).
The post-translational modifications which are carried out on
the C-terminal region of the protein such as glycosylation,
cleavage and attachment of a GPI anchor have been demonstrated to
not significantly affect the structural dynamics of PrPC in silico.
The lack of impact of the N-terminal region on the NMR derived
crystalline structure of the protein, and the lack of an observable
effect that these modifications have further enables the sole use
of the C-terminal globular region for molecular dynamics
simulations. It should however be noted that oligomerization of the
Prion Protein arises from different pathways, which allows for
distinct amyloid formations from the same initial structure.
Kinetic partitioning has been put forward as a way of explaining
amyloid formation via different transition pathways (Harrison et al
2001). This process was modelled dynamically by Dima and Thirumalai
in 2002 via the use of a three-dimensional beaded lattice model
containing a limited number of distinct monomer conformations.
Aspects of this model showed the progression to aggregated states
through the occupation of a state other than U (representing
complete unfolding) or N (the native structure of the lattice)
indicating that partially structured intermediates were able to
form aggregated states, without necessarily having to undergo full
unfolding. This model also highlighted that aspects of the
aggregation process were dependent on several physiological factors
such as temperature, pressure, concentration and the isoelectric
point (pI) of salts in solution.
The effect of physiological factors on the unfolding
characteristics of the prion protein have been greatly exploited by
computer modellers. Under normal conditions for the WT-peptide,
unfolding is very hard to simulate, given the tendency of the
protein to occupy a minimised energy state which corresponds to its
native structure under biological conditions (Daggett 2002, Rathore
2004). Altering physiological factors such as temperature, pH and
hydrodynamics has been shown in some cases to greatly accelerate
the unfolding speed of the prion protein and put it within a time
scale that can be effectively simulated (Gu 2003).
In the present simulation the use of higher temperatures were
used to induce unfolding of the prion protein in a way that
highlighted potential partially structured intermediates, which can
arise from protein unfolding and misfolding. The mutations chosen
were T188A, T188K, and E219K, as examples of mutations known to
have a varied effect on the pathogenesis of the inherited prion
diseases Creutzeldt-Jakob Syndrome. The hypothesis was that the
mutants would show differences in unfolding which related to their
tendency to form aggregated disease states. This is highest in T188
variants and lowest in E219K.
The protein structures used for simulations were based on the
NMR derived structure of the major prion protein 1QLX.
Methods
The model for the C- terminal globular domain of the WT peptide
1QLX was used. The model was downloaded from the protein database
and the structures for the C-terminal fragment 125-228 were
constructed using Modeller9.13. Initially modelled were the
structures for naturally occurring mutant variants T188A, T188K,
E219K, G142S, and N171S, which were constructed by single amino
acid substitution their structures. The FASTA sequences for the
C-terminal fragments are shown below. WT
Peptide(125-228)LGGYMLGSAMSRPIIHFGSDYEDRYYRENMHRYPNQVYYRPMDEYSNQNNFVHDCVNITIKQHTVTTTTKGENFTETDVKMMERVVEQMCITQYERESQAYYQR
T188A(125-228)LGGYMLGSAMSRPIIHFGSDYEDRYYRENMHRYPNQVYYRPMDEYSNQNNFVHDCVNITIKQHAVTTTTKGENFTETDVKMMERVVEQMCITQYERESQAYYQR
T188K(125-228)LGGYMLGSAMSRPIIHFGSDYEDRYYRENMHRYPNQVYYRPMDEYSNQNNFVHDCVNITIKQHKVTTTTKGENFTETDVKMMERVVEQMCITQYERESQAYYQR
E219K(125-228)LGGYMLGSAMSRPIIHFGSDYEDRYYRENMHRYPNQVYYRPMDEYSNQNNFVHDCVNITIKQHTVTTTTKGENFTETDVKMMERVVEQMCITQYKRESQAYYQR
MD simulations were carried out using the OPLS-AA/L all-atom
force field and the GROMACS-4.6.3 package. Two hundred steps of
steepest descent method was used to minimize the structure.
Solvation was carried out with 3 Cl- ions. At temperatures of 300 K
and 500 K the modelling was carried out using the NVT ensemble.
Equilibration for 200 ps of solute position was carried out using
RMD. All mutations were subject to UMD for 2 ns using the LINCS
algorithm and a pH of 7 was used for the simulations. The time step
used for each system was 2 femtoseconds. Three runs were carried
out for each variant. RMSDs were calculated for each variant on
each runstructure were calculated. Determination of the
distribution of secondary structure was carried out using DSSP.
Results
Fig 1: (a-c) C RMSD values for E219K runs 1-3, (d-f) DSSP
generated secondary structures for E219k runs 1-3)
Fig 2: (a-c) C RMSD values for T188A runs 1-3, (d-f) DSSP
generated secondary structures for T188A runs 1-3
Fig 3: (a-c) C RMSD values for T188K runs 1-3, (d-f) DSSP
generated secondary structures for T188K runs 1-3
Fig 4: (a-c) C RMSD values for WT-Peptide runs 1-3, (d-f) DSSP
generated secondary structures for WT-Peptide runs 1-3
The data gained from plotting the C root mean square derivative
(RMSD) values over time suggests a progressive unfolding pattern,
whereby several relatively stable partially unfolded states appear
to occupy several local energy minima. The unfolding of the native
protein at 300k was carried out over the period of 1 nanosecond and
shows the local minimum occupied by the normal structure of the
wild type protein. This state appears to be between 0.25 and 0.35nm
in terms of the RMSD, and all three repeats show consistency in the
time taken to reach this state.Once restraints are lifted at t=0,
most RMSD values of proteins at 520k rise steeply until they reach
an RMSD of roughly 0.3nm, which corresponds to the structure of the
prion protein in its native form. From this point there seem to be
several different types of unfolding characteristics which
correspond to several different stable unfolded intermediates. The
highest mutant variants A local minimum energy state representing a
partially unfolded intermediate is evidenced by a very slow rise in
RMSD value along with a long fluctuation period between 0.4-0.5nm
which is characterised by very little loss of secondary structure,
instead consisting of a partially disordered state where most of
the helical structure of helices H1, H2 and H3 is retained, with
only slight fraying around the edges. This local minimum state can
be observed in practically all of the prion variants but is
particularly visible with a longer residence time in E219K runs 1+2
(fig.1a + b), T188A run 3 (fig 2c.) and T188K runs 2+3 (fig. 3b +
c).Interestingly the native run at 520k seems to occupy a lower
energy minimum for a longer period of time, corresponding to the
native structure of the protein, maintaining a RMSD value of below
0.4nm for around 1ns in 2 out of 3 of its runs (fig 4a + c),
perhaps indicating a greater stability, or the slight difference in
structure already caused by the polymorphisms of the mutant
variants.At an RMSD value of around 0.6nm there seems to be another
local energy minimum in which resides a stable partially unfolded
intermediate, characterised by a greater loss of helical character
in helices H2+H3. This is particularly evidenced by E219K runs 1+3
(fig 1a + c), T188A runs 1+2 (fig 2a + b), T188K run 2 (fig. 3b)
and the native run 2 (fig 4b.). Interestingly T188A runs 1+2 seem
to largely bypass or have a very short residence time in the first
local minimum state progressing very quickly to the greater loss of
helical structure associated with the local minimum at around
0.6nm. This is particularly evident with the almost immediate
disintegration of helix H2 in T188A run 2.Although the run times
for unfolding were cut short at 2ns it appears as though there may
also be another local energy minimum between 0.70.8nm, evidenced by
T188A run 1 (fig 2a) and T188K run 1 (fig 3a), which is
characterised by a loss of helical structure for helices H1+H2+H3,
and appears to be differentiated from the other unfolding
structures of the same variants (see figs 2b, 2c, 3b, + 3c) by the
greater loss of helical content in helices H1 and H3. All prion
simulations conducted at 520K showed some level of unfolding
compared to the WT-peptide at 300k with all helices H1 H2+ H3
showing some level of structural change in one or more of the
simulations and no obvious elongation of the beta sheets S1 and S2
or formation of beta-sheet character.DiscussionMost of the
mutations associated with prion disease can be found in helices H2
and H3. This is the case of the T188A and T188K mutations found
within the H2 helix, and with the E219K polymorphism found in helix
H3. These two mutations are known to have contrasting effects on
the mechanism of disease state formation, with T188A/K mutations
known to be pathogenic and E219K having been demonstrated to
protect against disease states.
The RMSD values of E219K and the T188K can be compared directly
to contrast the differences in their stability. Whilst the highest
RMSD values in the T188 variants can be seen to top 0.9nm and 0.8nm
for T188K and T188A (fig 2a, fig 3a) respectively the highest RMSD
value in the E219k is only at around 0.7nm (see fig 1.). This value
obtained for the E219K variant also falls below the highest RMSD
values obtained in the WT peptide which also obtains peak values at
above 0.8nm.
Interestingly the E219K variant runs showed significant
unfolding of the H2 helix with very little helical character
seeming to remain in runs 1 and 3 (fig. 1d + f) of the unfolding
process. This showed significant congruence with the possible
energy minimum exhibited at RMSD of 0.6 nm, with the loss of
helical character seeming to coincide with the reaching of this
point.
The mechanism of inhibition of E219K is a currently under
discussion. A recent 2012 study carried out by Biljan et al found
that, whilst the Lysine polymorphism did not interfere with the
stabilizing influences around it, it did introduce new structural
features into the protein, and the replacement residue showed
evidence of reduced backbone flexibility. As Lysine in solution
exhibits considerably different character to that of Glutamic acid,
being a basic residue as opposed to one that is positively charged,
this is not an unexpected result.
It is also known that it is the heterozygous polymorphism of
E219K which results in the most resistance to amyloid formation
with the murine E219K variant only being shown to exhibit
resistance to scrapie formation when the gene expressed is
heterozygous. The reason for this is not currently known but it was
also speculated by Biljan that this may be due to the formation of
a dimer complex between Glutamate and Lysine 219 variants.
Interestingly another polymorphism associate with resistance to
prion disease the M129 mutation was shown in the cysteine variant
to increase stability via an extension of the natural
beta-sheet.
It is also notable that the T188K variant achieves the highest
RMSD value amongst any of the simulations undertaken (fig. 3a).
This again is consistent with the pathogenic nature of this
mutation, which would imply a loss of stability within the protein
structure. This mutation results from the replacement of threonine
which in nature is a polar amino acid with Lysine which is a
charged residue. This change in character is known to create new
electrostatic interactions within the protein, which would lead to
an alteration in stability.
It has been shown by prior studies that the T188 mutation is
known to trigger flexibility and displacement of H1, but it was
shown in our study, that the relatively stable fluctuation at 0.6nm
is evidenced across the board, and a similar breakdown primarily in
the H2 and H3 helices was observed. At an RMSD value of 0.7-0.8nm
there is however shown to be a considerable breakdown of Helix H1
in both T188A and T188K variants (fig. 2d + 3d). Although there are
other H1 shifts within the mutants variants, this is the only one
that appears to correspond to a particular energy minimum and
correspond to a stable unfolded intermediate. This would have to be
corroborated by repeat testing
With regards to the secondary structure of the protein it is can
also be seen that upon reaching an RMSD level of 0.6nm there is a
congruent level of unfolding in the H2 helix. This is evidenced by
T188K runs 2+3 (fig 3e + 3f). Similar levels of correspondence
between the reaching of an RMSD value of around 0.6nm and loss of
helical character in H2 can be seen as previously mentioned in
T188A run 2 (fig 2e).
The understanding of the unfolding process is crucial to the
understanding of the mechanism of the formation of oligomers from
PrPc and proteins in general. This is because investigative work
done beforehand shows that fibril formation occurs fastest at
denaturing conditions, indicating that a denatured or partially
denatured protein is essential to the formation of the amyloid
complex (Lawson 2005).
It is often common under constraints similar to those provided
here for proteins to become trapped in local minimum energy states
(Tang 2012). This is a natural tendency of proteins and can be
exploited to find potential characteristics of partially unfolded
states. Whilst it has become generally accepted that partially
structured intermediates are precursors in fibril formation, there
is a lot of contrasting evidence suggesting one pathway or the
other to amyloid formation (Abedini 2009).
The main point of contention in the literature around the
mechanics of prion formation appears to be which structure is most
unstable and prone to misfolding within the prion protein thus
leading to scrapie formation. The vast range of physiological
factors that can be altered in the modelling process has
exacerbated this point of contention, with some studies favouring
helix H1 as the most unstable structure within the prion protein,
and others generally favouring a combination of helices H2 and H3
(Langella 2004, Guo 2011).
In the data provided by the DSSP view of the secondary structure
it is apparent that loss of helical character in H1 is almost
non-existent in most of the simulations undertaken, and in the case
of run 1 for T188A and T188K does not appear to occur until the
RMSD value is approximately around 0.8nm. This is in agreement with
some studies which posit that a loss of helical character in H1 is
unlikely to occur in the partially structured intermediate before
in amyloid formation, whilst it contradicts other studies which
posit that the unfolding of the H1 helix is the most populated
state at melting point (Hosszu 2010, Tang 2012).
An explanation for this incongruity is offered by Garrec et al.
in a recent paper published in 2013. Within this it is demonstrated
by use of molecular dynamics simulations that at a pH of around 4.5
the slightly buried residue H187 assumes a protonated state which
can cause one of two changes to take place in the prion protein via
interaction with the linked residue R136. The repulsion caused
between the protonated imidazole ring and the guanidium group on
R136 appears to cause either a conformation where R136 or H187 is
pushed out of place, disrupting the native conformation of the WT
peptide.
The residues H187 and R136 correspond to 2 different parts of
the prion protein; H187 being part of the H2 helix and R136 being
located in the join between S1 and H1. Displacement of R136 from
its natural cavity was shown to cause the relaxation of the S1-H1
backbone constraint and the consequent close association of regions
S1-H1 and S2-H2 resulting in an elongation of the beta sheet in
this region. The displacement of H187 from its cavity conversely
causes a loss of intra-helical stability by repulsion with residue
T183. Though both of these actions result in the loss of helical
character from H2, the R136(out) populated state results in H1
destabilization as well as in the H2 helix, whilst the H187(out)
populated state results in a loss of character from only the H2
helix
This result may hold implications for the possible minima
observed in the T188A/K variants with T188 being very closely
related within the structure, which similarly seems to display 2
states, one in which there is only a loss of helical character from
Helices H2 + H3, and a second at 0.8nm where a loss of helical
character is found in the H1 helix as well.
This finding may also be of importance in relation to the
general process of prion formation. It is noted by Garrec et al.
that the ph level of 4.5 falls within the range of physiological pH
within the endosome. The transmembrane from of PrPc is found in two
topological isoforms where either the C-terminal end or the
N-terminal portion of the protein is directed towards the inside of
the cell and the endoplasmic reticulum known respectively as CtmPrP
and NtmPrP. Only CtmPrP has been implicated with any form of prion
disease.
PrPC is known to be subjected to endoproteolytic cleavage which
results in the production of an N-terminal and C-terminal fragment,
which is then either subjected to endocytosis or enzymatically
removed from the cell membrane. Marijanovic et al in 2009
demonstrated that the conversion of PrPc appears to take place
within endosomal compartments, and it has been demonstrated by use
of computer modelling that the H2H3 region alone can be converted
to a rich form, as was remarked upon earlier in this paper with the
example of the N-terminally truncated mutants rPrP51-90 and
rPrP32-121 fibrils produced by Cordeiro et al. The unglycosylated
form of protein as reported earlier also lowers resistance to
scrapie formation, and was shown to reproduce detergent
insolubility and protease resistance by Lehmann et al 1997. This
makes it a greater possibility that an initial scrapie seed would
be formed from an unglycosylated form of this terminal fragment,
via the loss of stability in the structure and initial unfolding
generated from the effect of the protonated residue H187.
Given the information provided by in this paper it appears
likely to this writer that these effects, or a combination thereof,
may form a possible pathway for formation of the an initial scrapie
seed from PrPc. This would need to be tested by use of all the
factors elucidated here which are an unglycosylated, C-terminal
fragment produced by cleavage at residues 110-111, subjected to
endosomal pH levels of around pH 4.5-5. From the evidence contained
within this paper this writer theorizes that this would likely lead
to an amyloid formation with a very similar character to PrPSc
including protease resistance and detergent insolubility.
This process could be accelerated by use of
phosphatidylethanolamine or PE, a common membrane protein shown to
act in concert with an initial scrapie seed to increase the rate of
amyloid formation by a factor of 105 (Deleault 2012). As scrapie
formation is characterised by a long incubation time, this may
increase the rate of reaction enough to be viewed in a suitable
frame of time. Pathogenic mutations associated with disease could
also be incorporated into testing to establish this.
Conclusion
From the effects seen within this paper possible supporting
evidence was used to show a difference in stability between E219K
and the T188 mutants. It was also shown that the proteins seem to
occupy partially structured intermediates which correspond to local
minimum energy states at denaturing conditions. It was also
concluded from a search of the literature that it may be possible
to reproduce an amyloid with all the characteristics of an initial
scrapie seed under physiological conditions. This has the potential
to be laboratory tested with different polymorphisms.