[proteamdavis] Fwd: helix-sheet intraconversion

  • From: Paul Limb <paulimb@xxxxxxxxx>
  • To: proteamdavis@xxxxxxxxxxxxx
  • Date: Sat, 3 Jul 2004 11:25:17 -0700

---------- Forwarded message ----------
From: Paul Limb <paulimb@xxxxxxxxx>
Date: Mon, 28 Jun 2004 17:25:55 -0700
Subject: Fwd: helix-sheet intraconversion
To: jtmorgan@xxxxxxxxxxx

---------- Forwarded message ----------
From: Paul Limb <paulimb@xxxxxxxxx>
Date: Mon, 28 Jun 2004 09:36:37 -0700
Subject: helix-sheet intraconversion
To: paulimb@xxxxxxxxx

A 16-amino acid oligopeptide forms a stable =CE=B2-sheet structure in
water. In physiological solutions it is able to self-assemble to form
a macroscopic matrix that stains with Congo red. On raising the
temperature of the aqueous solution above 70=C2=B0C, an abrupt structural
transition occurs in the CD spectra from a =CE=B2-sheet to a stable =CE=B1-=
without a detectable random-coil intermediate. With cooling, it
retained the =CE=B1-helical form and took several weeks at room temperature
to partially return to the =CE=B2-sheet form. Slow formation of the stable
=CE=B2-sheet structure thus shows kinetic irreversibility. Such a formation
of very stable =CE=B2-sheet structures is found in the amyloid of a number
of neurological diseases. This oligopeptide could be a model system
for studying the protein conformational changes that occurs in scrapie
or Alzheimer disease. The abrupt and direct conversion from a =CE=B2-sheet
to an =CE=B1-helix may also be found in other processes, such as protein
folding and protein=E2=80=93protein interaction. Furthermore, such drastic
structure changes may also be exploited in biomaterials designed as
sensors to detect environmental changes.

biomaterials | ionic residues | repeating sequences | scrapie related
diseases | temperature-induced transition


Here we report an unusual phenomenon: a peptide organized in a =CE=B2-sheet
is able to convert directly into an =CE=B1-helical conformation under the
stimulus of temperature or pH changes. No evidence is found for the
existence of a random-coil intermediate. The conversion is found in an
oligopeptide DAR16-IV (see Table 1), which was first found to form a
=CE=B2-sheet. This type of structural transformation phenomenon is similar
to that seen in a number of diseases, such as Alzheimer or scrapie.
These diseases are associated with the accumulation of similarly
stable =CE=B2-sheet structures that are resistant to degradation. These
substances, often called amyloid, stain with Congo red and represent a
focal point in trying to understand the pathogenesis of these

=CE=B1-Helices and =CE=B2-sheets are the major secondary structural motifs =
organize the three-dimensional geometry of proteins (1, 2). The amino
acids have been extensively studied for their ability to form either
helical or sheet structures (3). Experiments have demonstrated that
the sequence of amino acids plays a key role in determining the
propensity for protein secondary structure (4=E2=80=936). It is generally
considered that secondary structures are stable once they are formed.
These structures are believed to be of great importance in protein
folding pathways that may be quite complex (7=E2=80=9310). It is well known
that poly-lysine, poly-glutamate, and oligo-alanine can undergo
secondary structure changes in different solvent environments, such as
varying pH or salt concentration (11, 12). Furthermore, several block
copolymers with identical composition but with different
sequence=E2=80=94e.g., (Val=E2=80=93Lys)n, Ala20=E2=80=93Glu20=E2=80=93Phe8=
, Glu20=E2=80=93Ala20=E2=80=93Phe8=E2=80=94and
others form helical structures with quite different content in the
same environment (13, 14).

We have previously described a class of ionic self-complementary
oligopeptides. This class of oligopeptides forms =CE=B2-sheet structures
that are exceedingly stable in some extreme conditions including high
temperature, a wide pH range, high concentrations of denaturation
reagents, and a variety of proteases (15=E2=80=9317). The =CE=B2-sheet form=
ionic complementary oligopeptides have repeating sequences and are
classified by the repetitions of ionic charge groups. These =CE=B2-sheets
have a hydrophobic side and a charged hydrophilic side that has
alternating + and =E2=88=92 charged amino acid residues. For example,
molecules of modulus I have =E2=88=92 + =E2=88=92 + =E2=88=92 + =E2=88=92 +=
; modulus II, =E2=88=92 =E2=88=92 + + =E2=88=92 =E2=88=92 +
+; modulus IV, =E2=88=92 =E2=88=92 =E2=88=92 =E2=88=92 + + + +, etc. Severa=
l =CE=B2-sheet ionic
self-complementary oligopeptides including modulus I, II, and IV can
spontaneously self-assemble to form macroscopic matrices in the
presence of monovalent alkaline salts. These matrices are readily
stained by Congo red and are visible to the naked eye (15=E2=80=9317). In t=
scanning electron microscope, it was shown that the matrices are like
felt interwoven from individual fibers =E2=89=8810=E2=80=9320 nm in diamete=
r with 50-
to 100-nm pores. A variety of mammalian cells have been found to
utilize these peptide matrices as scaffolding for cell attachment


Materials. All oligopeptides used in this study (EAK16-IV, KAE16-IV,
DAR16-IV, and RAD16-IV; Table 1) were synthesized by t-Boc chemistry
on an Applied Biosystems automated model 430A peptide synthesizer and
purified by reverse phase HPLC at the Biopolymers Laboratory at the
Massachusetts Institute of Technology (16). The peptides were
dissolved in deionized water and stored at room temperature. The
concentrations were determined by amino acids hydrolysis as described

CD Analysis. CD spectra of serial dilution of either the =CE=B1- and =CE=B2=
of DAR16-IV peptide solutions showed that these structures are stable
even at 0.7 =CE=BCm in water. The value of mean residue ellipiticity at
[=CE=B8]218nm for =CE=B2 and [=CE=B8]222nm for =CE=B1 versus the concentrat=
ion remain
essentially unchanged in both cases. On dilution the =CE=B1-form has an
isosbestic point at 198 nm and the =CE=B2-form has an isosbestic point at
205 nm. Two pairs of modulus IV self-complementary oligopeptides with
identical compositions and length, EAK16-IV, KAE16-IV and DAR16-IV,
RAD16-IV have been systematically analyzed. To measure the effect of
various pH on the peptides, they were incubated overnight at room
temperature in solutions with pH range from 1=E2=80=9312 before measuring w=
the CD. For ionic strength changes, a peptide solution was divided
equally into six tubes and NaCl was adjusted to a final concentration
from 10 mM to 160 mM. These peptide solutions were then heated for 32
min at 75=C2=B0C and cooled to 25=C2=B0C before measurement.

Sedimentation Equilibrium Analysis. For determination of molecular
weight, the =CE=B1-helical form of DAR16-IV oligopeptide was heated in
water to 75=C2=B0C for 35 min and its =CE=B1-helical structure was confirme=
d by
CD. Three different concentrations of the oligopeptide were then
loaded into a hexa-cell chamber with three cells containing water as
the references. The samples were then centrifuged at 50,000 rpm at
25=C2=B0C for 24 h to sedimentation equilibrium. Each sample was then
measured several times independently. All points were collected and
analyzed. In all cases, the samples showed an average molecular mass
1870 which is close to the calculated monomeric molecular mass of 1670
Da. We therefore concluded the helical form of DAR16-IV is a monomer
in water.

Data Base Search. A search of the current protein data bases using
BLAST through the internet did not identify a sequence identical to
that of DAR16-IV. However, there are many segments containing clusters
of alternating charged repeating residues in a number of proteins with
charge distributions similar to DAR and EAK.

Other Analysis. Preliminary NMR examination of DAR16-IV using either
heated or unheated samples showed distinctive spectra suggesting that
the secondary structures of the molecules differ from each other. A
systematic analysis with NMR will be reported elsewhere.


Structure Properties. We have previously described this class of ionic
self-complementary oligopeptides which form remarkably stable =CE=B2-sheets
and can self-assemble to form macroscopic matrices that stained with
Congo red (15=E2=80=9317). Every other residue in the peptide is hydrophobi=
such as alanine, and the other residues are repeating clusters of
positively and negatively charged residues. If the ionic residues
alternate with two positive and two negative residues, they are
described as modulus II (15=E2=80=9317). One of the modulus II oligopeptide=
EAK16-II [AEAEAKAKAEAEAKAK], was originally found in a yeast protein,
zuotin, which was characterized as a putative left-handed Z-DNA
binding protein (18). Several modulus II and modulus I peptides have
been systematically analyzed (S.Z., unpublished data). Recently, we
have studied several modulus IV peptides containing 16 amino acids,
EAK16-IV, KAE16-IV, DAR16-IV, and RAD16-IV (Table 1). All four
oligopeptides displayed typical =CE=B2-sheet CD spectra at ambient
temperature. However, when the solutions were heated, they exhibited
different behavior. For example, both EAK16-IV and KAE16-IV, similar
to other members of the class, are extremely stable even at high
temperature. Their =CE=B2-sheet CD spectra were essentially unchanged from
20=C2=B0C to 90=C2=B0C (Fig. 1A). In the case of RAD16-IV, after heating at=
for 10 min, the room temperature CD spectrum showed that the =CE=B2-sheet
content remained essentially unchanged as indicated by the
ellipiticity at 218 nm. However the ellipiticity at the 195-nm region
was apparently reduced. This reduction probably reflects a change in
the right-handed twist of the =CE=B2-strands (Fig. 1B) because almost all
=CE=B2-strands in proteins are believed to possess some degrees of
right-handed twist (19, 20).

An Abrupt Structure Conversion. A more surprising result was found
with the peptide DAR16-IV. On incubation at elevated temperature, it
underwent a structural transition from a =CE=B2-sheet directly to an
=CE=B1-helix without an observable random-coil intermediate. Thus the
secondary structure of the oligopeptide had drastically changed as a
function of temperature (Fig. 2). The =CE=B2 to =CE=B1 transition was first
observed by measuring the CD spectra of the peptide solution at
different temperatures in the heating chamber of the instrument. At
elevated temperature, the =CE=B2-sheet spectrum was replaced by an =CE=B1-h=
spectrum. We also examined the DAR16-IV solution by heating it at 90=C2=B0C
for 10 min in a water bath and took its CD spectrum at 20=C2=B0C; a
distinctive structure transition was observed (Fig. 2A). To find out
more about the transition, several identical samples were heated at
75=C2=B0C for various time periods and their spectra were measured at 20=C2=
(Fig. 2B). No transition was observed on heating from 1 to 8 min.
However, changes were seen after a longer period of heating and the
transition was complete after 32 min of heating. Prolonged incubation
did not further promote the transition. Once the =CE=B1-helical structure
is formed, it is quite stable. When the DAR16-IV oligopeptide in
helical form was re-examined by reheating at different temperatures,
the helical content was gradually reduced as a function of incubation
temperature but it did not return to the =CE=B2-sheet form nor completely
denature into a random coil (Fig. 2C). The mean residue ellipiticity
[=CE=B8]222nm showed that =E2=89=8860% of peptide was in the =CE=B1-helical=
 form at 0=C2=B0C
and 30% remained at 90=C2=B0C. Furthermore, after the helical structure is
formed, it is very stable at room temperature. After several weeks
only a small fraction had returned to the =CE=B2-sheet form. This delayed
hysteresis suggests that the structural conversion is kinetically
irreversible. It should be noted that slow formation of a stable
=CE=B2-sheet is found in the pathogenesis of a number of protein
conformational diseases.

Structural Stability at Low Concentrations. Both the =CE=B1-helical and the
=CE=B2-sheet forms of DAR16-IV are stable at submicromolar concentrations.
The oligopeptide solutions were serially diluted with water and their
CD spectra were measured. The spectra showed a typical =CE=B1-helical
profile with an isosbestic point at 198 nm for the helical form (Fig.
3A). This result is not surprising because many short =CE=B1-helical
structures are monomeric and are not sensitive to concentration
changes. This finding is in agreement with other helical oligopeptides
of similar length (21, 22). The existence of DAR16-IV as a monomer in
the =CE=B1-helical form is consistent with the analytical equilibrium
sedimentation centrifugation as described in Materials and Methods.
Interestingly, the =CE=B2-sheet structure was likewise unaffected by
dilution of the oligopeptide and CD spectra had an isosbestic point at
205 nm (Fig. 3B). This is unusual because =CE=B2-sheet structures of
oligopeptides are in general affected by changes in concentration.
=CE=B2-Sheet structures consist of short sequences largely stabilized by
intermolecular interactions that usually disassociated upon dilution.
However, this result is in a good agreement with our previous analysis
of this class of =CE=B2-sheet ionic self-complementary oligopeptides in
which very stable =CE=B2-sheet structures are largely independent of
peptide concentration (16). This may be related to their architecture
in which repetitive + and =E2=88=92 charged residues can form ionic bonds o=
one side of the sheet while the alanines on the opposite hydrophobic
side form staggered van de Waals interactions; part of the structure
resembles interactions found in silk (23, 24).

Ionic Effect. Ionic strength can inhibit the structural transition.
Addition of NaCl to 10 mM or greater effectively inhibited the =CE=B2- to
=CE=B1-structure transition (Fig. 4). NaCl at 10, 20, and 40 mM inhibited
the structural conversion. However, increasing the NaCl concentration
to 80 and 160 mM resulted in reducing the =CE=B2-sheet structure CD spectra
in the 195-nm region. This change may reflect the slight structural
alteration of the backbone twists of the =CE=B2-sheet peptide. Salts such
as NaCl and KF have been shown to stabilize the =CE=B2-sheet structures,
and for some peptides they even induce formation of a =CE=B2-sheet from a
random coil or an =CE=B1-helix (13, 14). In this case, NaCl not only
inhibits the structure conversion, but also induces the peptide
self-assembly into a peptide matrix that can be stained Congo red
(data not shown) (15=E2=80=9317).

pH Effect. We also investigated secondary structural conformational
changes as a function of pH. Our previous study of EAK16-II and
several other ionic self-complementary oligopeptides at various pH
values showed that their =CE=B2-sheet conformation was quite stable over a
wide range (16). When oligopeptide DAR16-IV was incubated at a pH of 1
and 2, it showed a helical structure. When DAR16-IV was incubated at
pH 3 and 4, the CD spectra displayed a =CE=B2-sheet form but without a
clear isosbestic point, suggesting that there may be some structural
intermediates near pH 3 (Fig. 5A). However, when the peptide was
incubated in solutions from pH 5=E2=80=9311, the CD spectra exhibited a
typical =CE=B2-sheet form with an isosbestic point at 212 nm and a gradual
decrease of ellipiticity as a function of increasing pH (Fig. 5B).
These observations are consistent with previous reports that polymeric
peptides, such as polylysine can readily undergo secondary structural
transitions as a function of pH (11, 12).


Almost all =CE=B2-sheet ionic self-complementary oligopeptides studied to
date spontaneously assemble to form a macroscopic matrix in the
presence of monovalent salts (15=E2=80=9317). We also tested the property o=
matrix formation of DAR16-IV in the =CE=B2-sheet form. When DAR16-IV was
tested in phosphate-based saline at room temperature, it readily
formed a visible matrix that stained with Congo red; however, it
failed to form such a matrix after the same sample was heated at 75=C2=B0C
or higher temperature for 30 min (data not shown). These observations
are consistent with our previous observations that the =CE=B2-sheet
structure is an important structural component in facilitating its
self-assembly to a macroscopic structure. The unique structural
feature of this class of ionic =CE=B2-sheet oligopeptides is characterized
by hydrophobic residues entirely on one side of the =CE=B2-sheet (alanine,
in this case) and both positively and negatively charged residues on
the other side which vary in an alternating repeating pattern. In the
case of DAR16-IV there are four negatively charged aspartic acid
residues at the N terminus and four positively charged arginines at
the C terminus. It is interesting that this =CE=B2 to =CE=B1 conversion doe=
s not
occur when the polarity of the polypeptide is reversed, as in the case
of RAD16-IV. This suggests that the orientation of the =CE=B1-helix dipole
moment may contribute to stabilization of the =CE=B1-helix in one case but
not in the other. Previous reports have shown that the =CE=B1-helix has a
dipole moment with the positive pole at the N terminus and the
negative pole at the C terminus (25). It is interestingly that
negatively charged residues, Asp and Glu, are more frequently found
located at the N terminus of the helices in proteins, whereas the
positively charged residues, Arg, Lys, and His, tend to be found at
the C terminus of the helices (25, 26).

Structure Conversion. There appears to be a threshold effect involved
in the conversion as shown in Fig. 2B. Heating DAR16-IV at 75=C2=B0C
produces very little change in the CD spectra for the first 8 min, but
the change to an =CE=B1-helix is essentially completed after 32 min of
heating. Thus it takes some time to thermally disrupt the =CE=B2-sheet
lattice to free peptides to form the =CE=B1-helix. Likewise, when the
=CE=B2-sheet is destabilized by lowering the pH, the destabilization only
begins to become evident in the CD spectra when the pH is lowered
below 4, and is essentially complete when the pH is dropped to 2. It
is likely in this case that the aspartate residues are gradually
becoming protonated, and that they destroy the stability of the ionic,
self-complementary =CE=B2-sheet interactions.

Implications in Amyloid Formation. Once the =CE=B1-helix is formed, it
takes many weeks at room temperature to slowly reform the =CE=B2-sheet
structure. The cycle of =CE=B2-sheet formation disrupted by heat to form an
=CE=B1-helix, and then upon cooling the slow reformation of the =CE=B2-shee=
shows considerable hysteresis; the process is not kinetically
reversible. The rate of reforming the sheet at room temperature is
much slower than the rate at which the helix forms once the =CE=B2-sheet is
disrupted. This slow process of going from an =CE=B1-helix to =CE=B2-sheet =
one that is studied widely at present because it is believed to form
the molecular basis of a number of disorders, including amyloid
formation in neurological tissues in diseases such as scrapie, bovine
spongiform encephalitis, or Alzheimer disease (27=E2=80=9334). There it is
postulated that secondary structural conformation changes occur in
which a cleaved segment of a protein that is normally found in an
=CE=B1-helical form, is converted to a stable =CE=B2-sheet. Many investigat=
believe that these protein conformation changes are central to the
disease process. Thus the progression of the disease may be related to
changes in protein secondary structure leading to the formation of
insoluble =CE=B2-sheet plaques. In the infectious diseases such as scrapie,
a segment of the protein in the =CE=B2-sheet form (a prion) is believed to
catalyze the further conversion of segments from =CE=B1-helix to the
=CE=B2-sheet form (31, 32). By further studying DAR16-IV and its
derivatives we may perhaps gain some insight into factors that
influence these conformational changes and this might have relevance
to the pathogenesis of these diseases. It is interesting that these
diseases seem to be characterized by a slow conversion of protein into
an insoluble =CE=B2-sheet form, and DAR16-IV shows similarly slow kinetics
of conversion.

In contrast, the =CE=B2 to =CE=B1 conversion is fast, acting like a molecul=
switch with a dramatic conformational change. This type of
conformational switch may also occur in proteins. It could be an
element in protein folding, or it might even occur in some proteins
during the mechanism of their action. An example of a large
conformational change is found in the hemagglutinin molecule where
membrane fusion is brought about through a pH drop in endosomes
accompanied by conversion of an unstructured peptide segment into an
=CE=B1-helix (35, 36). The =CE=B2 to =CE=B1 conversion we are describing he=
re results
in a considerable geometrical change; DAR16-IV has an extended =CE=B2-sheet
length of =E2=89=885 nm compared with the helical length of =E2=89=882.5 nm=
. Thus,
changes of this type might occur in protein mechanisms with
significant shape changes.

Structural transitions also have been found in other systems. Perutz
and coworkers (37, 38) have reported that glutamine repeats found in
Huntington and several other neurological diseases form =CE=B2-sheets with
side chains forming complementary hydrogen bonds. One of these repeats
changes its conformation from a =CE=B2-sheet to a type I =CE=B2-turn at hig=
temperature (37, 38). These are some of the examples in which protein
conformation can be altered as a consequence of perturbations in the
environment of a particular sequence within a protein. This phenomenon
may be more widespread than previously realized, as nature is likely
to exploit these properties.

Applications in Biomedical Engineering. Such a drastic structural
change from a =CE=B2-sheet to an =CE=B1-helix found in DAR16-IV, with a 50%
change in distance, from 5 nm to 2.5 nm in a 16-residue peptide, may
also be exploited for biomedical engineering applications. For
example, one can design peptide biosensors so that they can rapidly
response to in vivo or in vitro environmental pH or temperature
changes. This type of biosensor may be further developed for a variety
of diagnostic devices.

Other related posts:

  • » [proteamdavis] Fwd: helix-sheet intraconversion