FUSION PROTEINS WITH TOXIC ACTIVITY - Fusion Protein Technologies for Biopharmaceuticals: Applications and Challenges - page 251

GM-CSF, or tumor necrosis factor a (TNF- a ) at the carboxy
terminal end of the heavy chain. Surprisingly, all variants
had a much shorter half-life than the parental antibody,
which is useful since this reduces unwanted effects through
rapid clearance, but retains the fusion protein in the tumor
because of the high specificity of the antibody. All protein
variants achieved only transient tumor regression because no
memory cells were generated. A promising candidate to
design immunocytokines is the human chemokine LEC
(liver-expression chemokine) that strongly attracts a number
of different lymphocytes and monozytes. Chemokines
require a free amino terminus; therefore, fusions to anti-
bodies have to be arranged at the variable domain. Similarly,
to other immunocytokines, fusions with LEC result in a high
tumor to organ ratio, minimizing unwanted side effects. An
alternative to the classical cytokines to attract and stimulate
immune cells is the utilization of co-stimulatory molecules
such as B7 to induce T-cell proliferation. B7 and LEC fusion
proteins were able to achieve complete regression in a
number of tumor models.
The next class of molecules suitable to induce indirect
killing of malignant cells through immune cells are members
of tumor necrosis factor superfamily (TNFSF) such as
OX40L, CD137L, and GITRL. These proteins can be found
on the surface of many activated immune cells and trigger a
sustained T-cell response after initial activation, thus creat-
ing an immunological memory. Tumor-targeted fusion
proteins containing these ligands are highly efficient to
eradicate tumors and establish long-term protection [5].
Overall, three effects of immunocytokines have to be
orchestrated to obtain a long lasting cure: first, tumor
infiltration of immune cells must be achieved by chemokine
attraction and antibody-mediated targeting; second, immune
cells must be activated, ideally in a co-stimulatory way to
obtain a memory effect; third, tumor-mediated immune
tolerance must be abolished. All these can be achieved by
a careful design of immune modulating fusion proteins [42].
Further details on targeted immunotherapy can be found in
Chapter 19.

targets are the multiple signal transduction cascades that are
regulated through phosphorylation. Aberrant phosphoryl-
ation often causes uncontrolled proliferation. The obvious
countermeasure to trigger self-destruction or to halt prolif-
eration would be to introduce kinases. Enzymes not requir-
ing cellular uptake for their toxic activity have been
elegantly designed as antibody-dependent enzyme prodrug
therapy (ADEPT). Novel therapeutic molecules to fight
cancer are the various antibody-enzyme fusion proteins [43].

17.4.1 Kinases

The utilization of kinases to force tumors into cell death is a
relatively new strategy. In order to achieve specificity and
receptor-mediated uptake, the kinase has to be fused to a
targeting molecule that identifies a marker or surface antigen
that can be found only on malignant cells. The kinase of
choice should restore cellular protection mechanisms such
as tumor suppressors. A good example is the inactivation of
the tumor suppressor death associated protein kinase
(DAPK2) in Hodgkin lymphoma. DAPK2 itself is a kinase.
Introducing a constitutively active version of DAPK2 fused
to a ligand of a surface molecule such as CD30 induces
apoptosis [44]. Since both molecules are of human origin, no
immunogenicity should be expected. A further advantage is
the lack of a toxic effect as long as DAPK2 is not introduced
into a cancer cell. The kinase itself is not harmful when
present in the bloodstream. This example is illustrated in
more detail in Chapter 21.

17.4.2 RNAses

The process to transcribe a gene and translate it into a protein
is a vital step for all organisms. The central molecule linking
transcription with translation is the ribonucleic acid (RNA)
in form of the messenger RNA (mRNA) or as amino acid
carrying transfer RNA (tRNA). Interrupting translation or
transcription ultimately kills cells. Therefore, RNA was
selected as an important cytotoxic target. RNA is a labile
molecule that can be destroyed enzymatically by RNAses.
One complication limiting the therapeutic access could be
the presence of RNAse inhibitor (RI) in the cytosol. How-
ever, it was demonstrated that an RNAse fusion protein
neutralizes RI and causes apoptosis [45]. Alternatively,
RNAses of nonhuman origin being insensitive to RI could
be utilized. Most frequently, human pancreatic RNAse,
angiogenin (ANG) or amphibian RNAses were used. All
RNAse fusion proteins are notoriously difficult to produce,
forming inclusion bodies or killing the production host.
Here, the same production strategies can be applied as
mentioned in Chapter 1. Of course, the enzymes can be
linked to any targeting molecule, either in the form of
antibody fragments or the cancer receptor specific ligands
such as growth factors for instance. However, RNAses

17.4 HUMAN ENZYMES

Enzymes as proteins with a wide range of functions can also
be utilized to kill aberrant cells. As before, it can be
distinguished between internal and external activity. Internal
targets for enzymes in principle are all functions that are
either essential for survival or trigger cell death through
apoptosis signals. Many of the toxins address the protein
generating process, blocking the ribosome or transcription
factors. The essential element for the translation is ribonu-
cleic acid (RNA). The sensitive nature of this molecule
makes it an ideal target for enzymatic damage through
RNA degrading enzymes, the RNAses. Other intracellular