Biomedical Engineering Reference
In-Depth Information
the success of monoclonal antibodies, yet
increase the
alternatives to antibodies with a focus on DARPins and
fusion proteins.
therapeutic efficacy.
In principle, an antibody is a natural example of a simple
but innovative fusion protein: Binding moieties (Fv or Fab
fragments) are combined with an effector and half-life
extension function (Fc fragment). This, of course, inspired
many antibody derivative approaches. Specificities were
trimmed, valency was adapted, fragments were removed,
and effector functions were reduced, extended, or adapted.
In other cases, additional effector functions such as small
molecule toxins or interleukins were added to antibodies.
Many of these successful applications are presented in other
chapters of this topic.
With engineering antibodies, a new era of therapeutic
protein engineering was started with a great prospect for
novel therapeutics, ultimately to the benefit of patients.
Considering that this research has been started about three
decades ago, surprisingly few of these approaches have
reached the market or marketability. This can be attributed
to several limitations imposed by antibodies.
One of the limitations concerns the composition of
immunoglobulin domains and their production. Antibodies
are complex molecules that rely on disulfide bonds and
glycosylation, requiring production in mammalian cells. To
be able to better work with antibodies, for example, in
selection procedures or for protein production, strong efforts
were put in dismantling the antibody into single pieces, such
as scFv or Fab fragments [2]. While very respectable
progress was made, the main limitations of the immuno-
globulin domain typically remain and a streamlined research
process is only possible with large investments.
Another limitation is the conceptual limitation imposed
by antibodies to our view on current drug development. The
success of monoclonal antibodies has limited our thinking of
what is possible with biologics. Limited tissue penetration
and i.v. injection are just two of the aspects that often are not
touched upon when developing new drugs.
To overcome such limitations and enable completely
novel therapeutic approaches, two things must happen: (i)
The research process must be fast-forward and as simple as
possible allowing for rapid generation of novel proof-of-
concept biologics. This can be achieved using antibody
fragments if a considerable investment is made, or using
novel approaches, that is alternatives to antibodies, that
allow simple and straightforward generation of novel thera-
peutic candidates. (ii) We must think of novel therapeutic
concepts, not limited by the limitations of antibodies.
Biologics can not only be administered by i.v./s.c. but
also by other routes. Biologics can be made to penetrate
tissue efficiently. Any fusion protein is possible. Any com-
bination of binding specificities and functionalities is feasi-
ble. If we think along these lines, we will come up with novel
therapeutic approaches with clear benefit for the patients.
This chapter highlights recent activities in the field of
34.2.2 Natural Binding Proteins
A look at natural human binding proteins reveals interesting
insights. Next to immunoglobulin domains, nature has
evolved a diverse set of other folds to be powerful pro-
tein-protein interaction mediators such as ankyrin repeat
proteins [3,4], leucine-rich repeat proteins [5,6], fibronectins
[7], as well as the SH2, SH3, and PDZ domains [8]. The
prominence of repeat proteins is underlined by the fact that
jawless vertebrates [9] and plants use this protein class as
binding proteins in their adaptive immune systems, just like
higher vertebrates use antibodies. The affinities of natural
binding proteins to their target proteins cover a broad range.
For example, the placental ribonuclease inhibitor quasi-
covalently inhibits ribonuclease (K
I
ΒΌ
10
14
M) [10].
Ankyrin repeat proteins interact with their target proteins
with high affinities in the picomolar range [11]. Other
proteins show interactions that are more in the nanomolar
to micromolar range, depending on the requirements to the
binders [8,12]. While many binding proteins show specific
binding, nature's strength is also in generating nonspecific
binding molecules that can fulfill several tasks [13].
Many binding proteins are typically specified for binding
certain types of target molecules. For example, SH2, SH3,
and PDZ domain proteins are specialized in binding peptides
[8]. Lipocalins are specialized in binding small chemicals
[14]. Fibronectins are typically involved in extracellular
matrix protein-protein interactions to connect cells [7].
Immunoglobulins and repeat proteins are different with
respect to their target selectivity. They show more diverse
target types, ranging from nonpeptides to peptides and
proteins. For example, ankyrin repeat proteins were shown
to be able to bind RNA [15], peptides [16], and proteins [11].
The chemical compositions and molecular structure of
natural binding proteins are as diverse as one could expect
from such a broad set of molecules. Some proteins such as
the fibronectins and ankyrin repeat proteins do not contain
disulphide bonds or free cysteines [4,7]. Other binding
proteins contain disulfide bonds for stabilization similar
as seen in immunoglobulin domains. Fibronectin has an
immunoglobulin-like fold with loops similar to CDRs of
antibodies [7]. Lipocalins have a beta-barrel structure and
also displays loops [14]. Ankyrin repeat protein use a
conceptually different approach to binding targets in that
they consist of repeated units of a turn-helix-loop-helix-loop
motif, forming an elongated protein domain with a large
target-interaction surface exposed on rigid secondary struc-
ture elements (Figure 34.1) [4]. Size-wise, the ankyrin repeat
domains of natural binding molecules are very flexible,
ranging from two to more than thirty repeats [17], while
typical ankyrin repeat proteins consist of four to six repeats,
5.9
Search WWH ::
Custom Search