FUSION PROTEINS WITH A TARGETING FUNCTION - Fusion Protein Technologies for Biopharmaceuticals: Applications and Challenges - page 377

Intravenous administration of EPO-TAT resulted in a sixfold
higher cerebrospinal fluid (CSF) concentration than EPO
without the TAT peptide, indicating an improved BBB
crossing [56]. A more universal concept is the combination
of CPP with antibodies to reach intracellular targets. In a first
attempt, TAT was fused to the B domain of staphylococcal
protein A, which can bind the Fc fragment of IgGs. The
noncovalent complex of the fusion protein and a fluores-
cently labeled, attached antibody could penetrate mouse
fibroblast cells as detected by confocal microscopy [57].
Many other CPPs have been identified and characterized
since the first observation that TAT could translocate to the
nucleus [58]. But all share common features; CPPs are
typically less than 30 amino acids long and have a high
positive net charge. The simplest form is polyarginine
(Arg 8 ). Besides polycationic peptides, amphipatic variants
can also facilitate cell penetration. The transmembrane
transport works with both: covalent linkage to the payload
and noncovalent complexes. One general downside of CPP
is the lack of specificity for cell types that results from the
receptor independent uptake. Instead of receptors, the inter-
action between CPPs and the cell takes place via cell surface
proteoglycans. Clustering of these surface molecules trig-
gers the remodeling of the actin network and activates small
GTPases. After this event, the CPPs can enter the cell either
through changed membrane fluidity or through perturbation,
clathrin, or caveolin-dependent endocytosis or macropino-
cytosis. The major limitation after passage into the cell is the
escape from the endosome [59].
The use of CPP which have also been called protein
transduction domains (PTD) has of course included antibody
fragments as well. Antibodies fused to PTD can be termed
transbody because they are able to translocate across cell
membranes. An alternative to transbodies are intrabodies
that rely on direct expression in a cell after transfection with
recombinant DNA. However, there a safety concerns that
restrict this form of gene therapy. Cell permeable antibodies
however would certainly be safer, but suffer from a limited
half-life [60].
CPPs are the carrier of choice to transfer proteins into the
cell to correct an aberrant cellular status caused by the
absence or malfunction of a protein. A good example is
the introduction of the tumor suppressor p53 into cancer
cells to restore normal function. The TAT-p53 fusion protein
increased life span of mice suffering from terminal perito-
neal carcinomatosis sixfold and even created disease free
animals [61]. Further applications and examples of CPPs can
be found in Chapter 26.

that certain autoantibodies can enter cells. One of these
antibodies, called 3E10 from mice with lupus nephritis, is
directed against DNA but not cytotoxic. Initially, the full
murine antibody was conjugated to catalase. After incubation
with the conjugate, the enzyme reached the cytoplasm and
nuclei of cells and remained active. To further improve
cellular uptake, one aspartate residue in the V H region was
exchanged to asparagine. Additionally, the antibody was
miniaturized to a single chain fragment that maintained the
full cell penetration ability [62]. The intracellular route of
3E10 was unclear at the beginning. But then it was found out
that the nuclear transport of this autoantibody is dependent on
the presence of the equilibrative nucleoside transporter ENT2.
This finding confirmed that here the nucleoside salvage path-
way is used for protein transport [63].
Owing to the unique nucleus targeting, the scFv variant of
3E10 was also tested in fusion to a p53-derived peptide. Since
the elimination of cancer cells with a carboxy-terminal p53
fragment was not satisfactory, in later studies, full-length p53
was used. In the majority of tested cancer cells, the killing
efficiency reached 90% [64]. The same fusion protein signifi-
cantly reduced liver metastases in a colon cancer model in
mice in vivo [65]. A further example is the 3E10-mediated
transfer of nuclear transcription factor, FOXP3, into cancer
cells. The fusion protein induced apoptosis and diminished
tumor burden in a Balb/c mice colon cancer model [66].
Besides its applications in cancer therapy, the 3E10 scFv
was also used to transfer microdystrophin to myoblasts to
treat Duchenne muscular dystrophy [67] or heat shock
protein 70 (Hsp70) to protect neurons from hydrogen per-
oxide toxicity [68].

25.3.4 Endosome Escape and Cytosol Translocation

The most frequently used strategy for internalization relies
on receptor-mediated uptake by endocytosis. However, the
bound ligands end up in the endosome that later fuse with the
lysosome with the risk of degradation by low pH or prote-
ases. Therefore, mechanisms are needed that overcome
endosomal entrapment of internalized molecules. Natural
proteins that are able to reach the cytosol after internaliza-
tion are the different bacterial toxins. Figure 25.4 gives a
schematic overview of the best-studied toxins containing a
translocation domain.
The Pseudomonas exotoxin A (PE) consists as many
other toxins from microorganisms of three domains with
different functions. Before entering the cell, the C-terminal
lysine is removed by a protease to expose the sequence
REDL at the C-terminus. Domain I binds to the a 2-macro-
globulin receptor/low-density LRP which is internalized via
coated pits. Proteolytic cleavage by furin and the reduction
of disulfides separates the N-terminal binding domain I from
the C-terminal rest. Passage to the cytosol is enabled by
retrograde endoplasmic transport through the trans-Golgi

25.3.3 Antibody-Mediated Transduction

Cancer therapy that utilizes the introduction of tumor sup-
pressor proteins can also be initiated by antibody-mediated
transduction (AMT). This concept is based on the observation