The triple-helical conformation of collagen (see Figure 1) has long been recognized for its
role in structural stabilization of connective tissue. The dissolution of the
collagen triple-helix has thus been implicated in a variety of diseases that
effect the structural integrity of various components of the body.
Figure 1: The triple-helical peptide-amphiphile
construct
Collagen also
provides a barrier to “compartmentalize” cellular activities; destruction of
this barrier plays a role in tumor cell invasion and metastasis. The matrix
metalloproteinase (MMP) family has been recognized for their collagenolytic
activities, and has thus been the subject of intense research efforts, in order
to elucidate their mechanisms of action and allow for rational design of
inhibitors. We have previously developed novel methodology for constructing
triple-helical peptides (THPs) and have applied these synthetic proteins for the
study of enzyme interactions with collagens. In the course of our early studies,
we have discovered that “triple-helical peptidase” activity may be more
widespread amongst proteases than previously believed. It appears that the
unique aspect of collagenolytic activity may not be the ability to cleave a
triple-helix, but rather the ability to bind and orient the native collagen
molecule properly. This paradigm shift could have significant effects on the
design of inhibitors against collagenolytic activity. To further explore the
nature of triple-helical peptidase activity, a series of THP substrates will be
assembled, incorporating known sites of collagen hydrolysis and varying over a
range of Tm values.
Substrate thermal stability will be correlated to enzyme activity. In tandem, 2D
NMR experiments using 15N-labeled amino acids will examine the
mobilities of the peptide backbone in these substrates. Individual kinetic
parameters and activation energies will be evaluated for MMP, trypsin,
thermolysin, cathepsin K, and aggrecanase hydrolysis of THPs. MMP-1, MMP-2,
MMP-8, and cathepsin K mutants will be utilized to determine the regions within
these enzymes critical for triple-helical peptidase activity. Finally, selective
substrates will be designed and improved methods for the synthesis of THP
substrates will be developed. Overall, triple-helical peptidase activity will be
systematically evaluated for a variety of proteases as a function of substrate
sequence and conformational flexibility, and the mechanism of collagenolytic
activity will better understood.
Investigation of collagen structural modulation of
melanoma cellular behaviors. We have shown that melanoma cell adhesion,
spreading, and motility is promoted by the a1(IV)531-543
region of type IV collagen. We will compare the promotion of cell adhesion,
spreading, and motility by triple-helical (homo- and heterotrimeric) and linear
versions of the ligand. We will synthesize single-stranded and triple-helical
peptides with Lys substituted by Hyl and glycosylated (Od) Hyl for Lys543 and
Lys540 from the human a1(IV)531-543
gene sequence. Protein sequence analysis has indicated that both of these sites
are Hyl in the native protein and that at least one is glycosylated. Both mono-
and disaccharides will be incorporated, as is found in native collagen.
Investigation of collagen structural modulation of signal
transduction pathways. We have shown a correlation between melanoma cell
binding to a1(IV)531-543
via the a3b1
integrin and induction of p125FAK and paxillin phosphorylation. We will compare
cell binding to triple-helical (homo- and heterotrimeric) and linear (glycosylated
and non-glycosylated) versions of the ligand, examine time courses of induction,
levels of induction, and look at the effects of clustered versus non-clustered
ligand. We will expand our initial signaling studies to look at downstream
events, i.e. which pathways are effected past pp125FAK, such as Tyr
phosphorylation and binding of paxillin or Grb2. We will use immunoprecipitation
analysis to detect paxillin and Grb2, and "peptide insertion" technology to
inhibit potential SH2 domain binding of Grb2 or pp60src to pp125FAK
intracellularly. We will also use an all-D peptide variant of
a1(IV)531-543 and determine if the
same signal transduction pathways are induced as for the all-L peptide. We have
shown that virtually identical levels of binding of the
a3b1 integrin to the
a1(IV)531-543 sequence are
achieved regardless of ligand stereochemistry. For therapeutic purposes, the
all-D peptide represents a potential in vivo inhibitor of metastasis, as it is
resistant to proteolysis.
Investigation of collagen structural modulation of
cellular production of basement membrane degrading metalloproteinases. We
have shown a correlation between cell binding to a1(IV)531-543
via the melanoma cell a3b1
integrin and induction of certain metalloproteinases. We will compare binding to
triple-helical and linear versions of the ligand, examine time courses of
induction, and look at clustered versus non-clustered ligand. The enzyme
products to be studied are several members of the matrix metalloproteinase (MMP)
family as well as a novel ~50 kDa metalloenzyme that we have discovered.
Metalloproteinases will be characterized via gel zymography and assaying with
fluorogenic and type IV collagen substrates. The novel metalloenzyme will be
further purified and characterized.
Introduction
Numerous
methods have been described for improving biomaterial biocompatibility. Our
approach involves coating biomaterial surfaces with cell active ligands that
maintain clearly defined structures at physiological temperatures.
"Peptide-amphiphiles," whereby hydrocarbon chains are covalently
linked to peptide sequences, have been shown previously to (a) form
a-helical
(see Figure 2) or triple-helical (see Figure 1) molecular architectures and (b) induce melanoma cell
adhesion and spreading. In the present study, a series of peptide-amphiphiles
incorporating cell adhesion sites from type I collagen were screened for
endothelial cell adhesion. These adhesion promoting molecules were then mixed
with the endothelial cell proliferating region of SPARC (Secreted Protein
Acidic and Rich in Cysteine). The greatest adhesion and spreading activity was
achieved by using a mixture of two peptide-amphiphiles and decanoic acid
(C10). Decanoic acid was required for proper packing of the peptide-amphiphiles
to allow for optimal cellular interactions. This mixture will now be deposited
onto the surface of a biomaterial, cardiovascular
stents. These stents will then be tested in a porcine angioplasty/restenosis
model.
Figure
2: The
a-helical peptide-amphiphile
construct
Samples
used for preliminary studies
-
Multiple
triple-helical peptide-amphiphiles from type I collagen were synthesized
and purified to be used for screening purposes.
-
Three
sequences were determined to have cell adhesive properties; A1, B1, and
F1.
-
These
samples were then mixed (1:1) with an a -helical peptide-amphiphile
containing a sequence from SPARC known to cause endothelial cell
proliferation.
-
Samples
were also mixed with varying amounts of decanoic acid to allow for uniform
spacing of peptide-amphiphiles on the surface.
Future
Studies
-
Additional
adhesion assays will be performed to indicate which receptors the cells
are using to adhere to the F1 peptide-amphiphile.
-
Affinity
chromatography will be performed to confirm receptor identity.
-
Further
assays will be performed to measure the extent of endothelial cell
activation (nitric oxide production, presence of E-selectin, VCAM-1, or
ICAM-1).
-
Cardiovascular
stents will be coated with peptide-amphiphiles. These stents will be used
in an in vivo porcine model of angioplasty/restenosis.
In recent years, the
United States has seen an alarming 1 in 4 deaths resulting from cancer-related
illnesses, accounting for the deaths of more then half a million Americans
per year (1)
. Mortality from cancer is
frequently due to metastasis, given that surgical excision of the primary tumor
considerably enhances a patient’s prognosis and prolongs survival
(2)
. The metastatic process is a complex series of events involving 1)
primary tumor intravasation (migration into the circulatory system), 2)
traveling through the blood stream to a secondary site where 3) adhesion and
migration on the extracellular matrix (ECM) components occurs, leading to 4)
tumor cell activation and 5) invasion of the basement membrane to ultimately
result in the 6) establishment of a secondary site of tumor growth. It is the goal of our laboratory to further elucidate the
processes that lead to tumor metastasis, concentrating on the interactions
between the tumor cell and specific components of the ECM, as well as the
differential response of the tumor cells to these elements.
Tumor cells interact with basement membrane collagen (type
IV) at the site of extravasation through distinct cellular receptors, including
the
a1b1,
a2b1,
and
a3b1
integrins, and cell
surface proteoglycans, such as CD44.
These receptors are well characterized as being differentially expressed
in metastatic tumors, relative to the normal cells, depending on tumor type and
stage of progression (3)
. Immediately following integrin-ligand
binding, activation of the tumor cell occurs which modulates a tyrosine-kinase
signaling cascade within the tumor cell.
Down-stream, induction of the mitogen-activated protein kinase (MAPK)
pathway occurs, resulting in the expression of specific ‘factors’ that
facilitate the invasion of the secondary tissues.
Among these factors are proteolytic enzymes, such as the matrix
metalloproteinases (MMPs), which are responsible (here) for compromising the
basement membrane, as well as the production of specific cytokines that recruit
the endothelium into the extravasation sequence.
The
objectives for this project focus on further elucidating this tumor-cell
receptor/collagen interaction and the subsequent biochemical consequences, and
are as follows:
Objective 1.
Preparation and characterization of peptide ligands specific
to CD44 and
a2b1
integrins. Using solid-phase methodology, we will prepare the following peptides
sequences derived from type IV collagen, in both single stranded and triple
helical forms: the CD44 specific IV-H1 (a1(IV)1263-1277),
in both glycosylated and non-glycosylated forms, and the recently solved
a2b1
integrin
specific sequence (a1(IV)405-410).
We shall verify peptide integrity by Edman-degradation sequence analysis,
amino acid analysis, and MALDI-TOF mass spectrometry, followed by circular
dichoism and 2D NMR studies to demonstrate triple helicity.
Objective 2. Examination
of the specific signaling-cascades induced by melanoma cell receptor-ligand
binding.
The binding of melanoma cell integrins to the basement membrane collagen (type
IV) is known to induce a tyrosine-kinase signaling cascade, resulting in the
phosphorylation of a 125 kDa focal adhesion kinase (p125FAK), as well
as the down-stream activation of the MAPK pathway.
Through western blot analysis we will reveal a time course and relative
induction levels for the known signaling components, in highly metastatic
melanoma cells stimulated with the peptides prepared in Objective 1.
Furthermore, signaling studies will also be performed with melanocytes
and poorly metastatic melanoma cell lines in an effort to create an ‘induction
profile’ for the various stages of tumor progression.
Objective 3. Identification of the specific
protease, TIMP, and cytokine response by the melanoma cells to ligand binding.
Previously, we investigated tumor cell induction with the Hep III sequence of
type IV collagen (a1(IV)531-543)
and documented production of MMPs 1, 2 and 9, as well as a novel 50 kDa
species, which may be a member of the ADAM family of proteases. Along these
lines, we shall examine post-induction media from highly metastatic melanoma
cells stimulated with the peptides from Specific Aim 1 for cytokines, tissue
inhibitors of metalloproteinases (TIMPs), MMPs, and members of the
ADAM/ADAM-TS family of proteases. We will use substrate zymography and western
blot techniques, as before, but also introduce the techniques of quantitative
RT-PCR and gene micro-array technology to reveal modulation of selected gene
products. These studies will be reiterated on melanocytes and poorly
metastatic melanoma cell lines in an attempt to evaluate the differential
expression of these gene products at various stages of tumor progression.
Synthesis of Glycosylated hydroxylysine
derivatives
Hydroxylysine (Hyl) is the major glycosylation site
within collagens. The d-hydroxyl group may be posttranslationally modified by
the monosaccharide galactose (b-D-galactopyranosyl) or the disaccharide glucose-galactose
[a-D-glucopyranosyl-(1®2)-b-D-galactopyranosyl]. Interest in collagen glycosylation stems from the recent reports of T-cell recognition of a
glycosylated sequence within type II collagen, the identification of melanoma
and breast carcinoma binding sites within type IV collagen that contain
glycosylated Hyl residues, and the activation of specific tyrosine receptor
kinases by glycosylated type I collagen. We have developed a strategy that allows for coupling of a properly
protected monosaccharide or disaccharide derivative
to an amino acid that contains protected carboxyl and amino groups and a free
hydroxyl group.
A Fmoc-glycosylated L-hydroxylysine
{Fmoc-L-Hyl(e-Boc,O-2,3,4,6-tetra-O-acetyl-b-D-galactopyranosyl)} [see Diagram 1]
was prepared in our Lab.
This
derivative was then used to prepare a model glycopeptide from tumor cell recognition
site within type IV collagen and was recently being examined to see the effect of
ligand glycosyaltion on tumor cell signaling to study the mechanism tumor cell
interaction with or invasion of basement membrane.
Diagram 1: Fmoc-glycosylated L-hydroxylysine
Currently we are making an attempt to prepare Fmoc-Hyl[(e-Boc,
O-d-(2-O-a-D-glucopyranosyl)-b-D-galactopyranosyl] using a similar approach used
for the preparation of monoglycosylated hydroxylysine.
Biophysical Characterization of
Peptide-Amphiphile Mini-Proteins
2D NMR Analysis of Type II Collagen Substrate/Enzyme Dynamics
To examine
the site-specific conformational dynamics of collagen in response to matrix
metalloproteinase cleavage using multidimensional NMR techniques, including
relaxation studies utilizing appropriately 15N-labeled collagen
mimics. This research is initially focused on human collagen type II (769-783
region) and the specific collagenolytic (vs. triple helicase) properties of
MMP-1. The results are anticipated to facilitate the drug design of specific
MMP inhibitors with therapeutic applications in such pathologies as
osteoarthritis and cardiovascular diseases.
