Biomedical Engineering Reference
In-Depth Information
showed that a group of basic amino acids contribute to
heparin binding and that heparin binding influences the
infectivity [40]. Importantly, the capsid of the AAV-2 vector,
which is being used for human gene therapy of congenital
muscle disorders, has been shown to be taken up into human
dendritic cells producing activation of capsid-specific T
cells. This uptake is thought to be due to heparin binding
and has the potential to produce toxicity in human clinical
trials [41,42]. Consistently, engineered or natural variants of
AAV lacking heparin-binding capability were less likely to
activate T cells directed against the capsid and did not seem
to compromise gene transfer.
In summary, the use of heparin-binding protein motifs as
a source of tissue-specific targeting is widely used in nature
from growth and differentiation factors to viruses. The
"barcode" provided by HSPGs in the ECM is often used
as low-affinity targeting system that often requires a second,
more highly specific receptor-ligand interaction to exert its
biological effects. As will be described later, we are cur-
rently developing such a system of tissue-specific targeting
for biopharmaceutical development to enhance tissue-spe-
cific delivery and produce sustained efficacy using the HBD
from the neuregulin 1 growth factor.
identified that alters the binding of FGF9 to HSPGs through
preventing homodimerization of the FGF9 protein. Since
monomeric FGF9 binds to heparin with lower affinity than
dimeric FGF9, the lack of dimerization results in increased
diffusion of the altered FGF9 protein through developing
tissues leading to ectopic FGF9 signaling, and repression of
joint and suture development.
An alternative method used to define HBDs is to generate
synthetic peptides replicating putative HBDs from a given
protein and study these peptides for their heparin-binding
properties. This has been done for the FGF receptor [48]. A
specific sequence in one of the immunoglobulin-like loops
in the extracellular domain of the FGF receptor tyrosine
kinase was identified to interact with heparin, independent of
FGF. In fact, a synthetic peptide of this sequence inhibited
both heparin as well as FGF binding to the FGF receptor.
Consistently, lysine residues within this sequence were
required for this interaction and an antibody against this
domain was also an antagonist of FGF-stimulated cell
growth. Another example where peptides have been used
to map heparin-binding portions is from the cell fusion
glycoprotein of human respiratory syncytial virus (RSV-
F) [49]. Synthetic overlapping peptides derived from the F-
protein sequence were assayed for heparin-agarose affinity
resulting in the identification of 15 peptides representing
eight linear HBDs.
Although many studies on HS-protein interactions have
been reported, a common consensus sequence for protein-
HS binding has yet been established [50]. However, there is
a general principle that ionic interactions between positively
charged amino acids and negatively charged sulfate groups
from HS are critical for protein-HS binding. There is also
some data from crystal structural studies of heparin-protein
complexes to suggest that arginine residues contribute more
than lysine residues, despite their identical charges [51].
While the affinity and specificity of the binding is mainly
determined by the secondary and/or tertiary spacing of these
basic residues, in some cases binding between multiple
proteins to HSPGs might require the formation of multi-
protein complexes, where HS stabilizes the interactions.
27.6 DISSECTING HEPARIN-BINDING PROTEIN
DOMAINS FOR TISSUE-SPECIFIC TARGETING
Given the tremendous diversity, yet specificity, of HS struc-
tures in different tissues that are critical for normal devel-
opment and function [43-45], it is equally important that
there be a similar diversity in specificity of heparin-binding
proteins. This is in fact the case with a wide variety of
peptide structures that show selectivity toward specific HS
structures. Understanding how HS specificity occurs in these
naturally occurring heparin-binding proteins will not only
provide clues to study the biological roles of HSPGs, but can
also generate novel methods for the development of tissue-
specific targeting of biopharmaceuticals.
One method that has been used to identify the HBDs
within proteins that bind HSPGs is site-directed mutagenesis
of specific amino acids. Site-directed mutagenesis was used
to identify key positively charged amino acid residues
required for the vaccinia viral envelope protein A27 to
interact with both heparin and HSPGs for cell-surface
attachment [46]. Using site-directed mutagenesis and solu-
tion NMR, a “KKPE” amino acid segment with a turn-like
conformation was identified as a key structural region that
mediates specific heparin binding. In addition, spontaneous
mutants have been found reveal specific structural features
required for heparin binding. The elbow knee synostosis
(Eks) mouse shows elbow and knee joint synosotsis, and
premature fusion of cranial sutures [47]. A missense muta-
tion in the FGF9 gene was identified in the Eks mouse was
27.7 FUSION PROTEINS INCORPORATING HBDs
Given that proteins that bind heparin are targeted to selective
tissues, have increased activity, and are resistant to proteol-
ysis, incorporating these features into biopharmaceuticals
could be an effective way to improve therapeutics. In addi-
tion to our work that will be described below using the
neuregulin 1's HBD, two other efforts to generate fusion
proteins that bind heparin have been undertaken. Vascular
endothelial growth factor A (VEGF-A) is a vascular perme-
ability factor and a potent inducer of vascular leakage that
can interact with HSPGs [52]. VEGF-E from the genome of
Search WWH ::
Custom Search