Biomedical Engineering Reference
In-Depth Information
It has to be noted that this approach does not lead to cell
killing, but prevention of secretion [21].
Therapeutic variants of BT replace the binding domain by
a specific targeting domain, for example, a peptide ligand.
Furthermore, the light chain can be directly linked to the
catalytic domain by a protease-sensitive spacer to allow
release to the cytosol. These engineered proteins have a huge
therapeutic potential to treat diseases that involve secretion
processes. Currently, there is one fusion protein termed
AGN-214868 in Phase II trials [22].
Another bacterial toxin tested for therapeutic potential is
Shiga toxin from Shigella dysenteriae. It is composed of the
catalytic A subunit associated to a homopentamer of the
small B subunit. The B subunits bind to a cell surface
glycolipid, and the toxin complex enters the cell by endo-
cytosis. In the endosome, the A domain is proteolytically
processed into an A1 and A2 fragment that is still linked by a
disulfide bridge. After retrograde transport through Golgi
and endoplasmic reticulum, the bond is reduced and the
catalytically active N-glycosidase of A1 is released into the
cytosol, where it modifies an adenine of the 28S ribosomal
ribonucleic acid (rRNA) of the ribososme which disrupts the
interaction with EF2 and protein synthesis [23].
Fusion proteins combining a chemokine as targeting
moiety and the A1 chain of Shiga toxin are called leukocyte
population modulators since they can deplete certain
unwanted leukocyte subtypes. A molecule consisting of
human CCL2 (monocyte chemoattractant protein-1) chemo-
kine fused to a truncated form of the enzymatically active A1
domain of S. dysenteriae holotoxin (SA1) was prepared for
clinical trials of glomerulonephritis [24].
Cells infected by viruses often contain a high concentra-
tion of proteases. The presence of these proteases can be
utilized to increase the specificity of therapeutic toxins that
can eliminate infected cells. This concept was successfully
tested by designing the so-called Zymoxins consisting of a
targeting, translocation, and a toxin domain that was kept
inactive by an inhibitory protein. This inhibitor can only be
removed in cells expressing a certain virus protease. This
then activates the toxin, which kills the cell. In an in vitro
model, the viral NS3 protease in hepatitis C virus (HCV)
infected cells cleaved the linker between diphtheria toxin
and a protective defensin or the spacer connecting ricin
catalytic domain with a stalk peptide [25].
A special case utilizing toxins is the generation of tumor-
targeted superantigens (TTSA). Antibody linked superanti-
gens target cytotoxic T-cells to tumors. Besides the direct
killing, secondary effects are induced by the release of
cytokines. This secondary effect is very useful for eliminat-
ing cancer cells that do not present the specific antigen.
In general, superantigens secreted from bacteria help
pathogens evade the immune system. They bind to major
histocompatibility complex (MHC) class II molecules on
antigen-presenting cells
receptors activating the T cells in an antigen independent
manner, which leads to the elimination of APCs [26].
The first generation of TTSA consisted of staphylococcal
enterotoxin A (SEA) fused to a Fab fragment directed
against the 5T4 oncofetal antigen that can be found on
many solid tumors but not in normal adult tissues. However,
despite its reduced binding to APC, this molecule, called
anatumomab mafenatox, caused systemic T-cell activation
and required individual dosing in Phase II clinical trial.
Furthermore, neutralizing antibodies could be found in
patients [27]. Therefore, it was replaced by naptumomab
estafenatox consisting of an improved Fab against the same
target and a mutagenized SEA with a 100-fold reduced
systemic toxicity fused to the heavy chain. This molecule
has demonstrated a beneficial safety and efficacy profile in
the clinic [28]. Further details can be seen in Chapter 24.
17.2.2 Plant Toxins
Plant toxins are derived from the family of ribosome inac-
tivating proteins (RIP). They are smaller than bacterial
toxins and can be dissected into three groups. Type I RIP
contain only an enzymatic domain of
30 kDa and cannot
enter cells without conjugation to a targeting moiety nor
translocate to the cytosol easily. Members of this group are
saporin from Saponaria officinalis, bouganin from Bougain-
villea spectabilis, and gelonin from Gelonium multiforum.
Ricin from Ricinus communis belongs to the group of type II
RIP. They are heterodimers with a catalytic domain
(A chain) and a lectin-like domain (B chain). Both domains
have a size of
30 kDa and are connected by a disulfide
bridge. The last group is type III RIP that have no lectin
domain and require proteolytic processing to become
active [29].
One advantage of class I RIP is their low systemic
toxicity, which is a result of their lack of binding to cells.
This renders them particularly suitable for coupling to
specific targeting molecules. The small size also limits
the number of potential antigenic epitopes and minimizes
mutagenesis efforts to reduce immunogenicity. The least
toxic type I RIP is bouganin of B. spectabilis. The numerous
evaluations of plant-derived toxins have now yielded a
variant, which is in early stage clinical trials. This toxin,
called citatuzumab bogatox (VB6-845), is a de-immunized
variant of bouganin combined with a Fab fragment against
epithelial cell adhesion molecule (EpCAM). Between the
targeting Fab and the toxin is a furin-sensitive linker to
enable specific uncoupling of bouganin to execute its toxic
function. This immunotoxin demonstrated nanomolar
potency and high selectivity toward several cancer cell
lines [30].
Another member of the type I RIP is saporin (S6) from
S. officinalis. The toxin has primarily been used as conjugate
to antibodies or fragments [31]. Only recently attempts were
(APC)
and particular T-cell
Search WWH ::
Custom Search