Biomedical Engineering Reference
In-Depth Information
presence of LAP monomers, that is, LAP-cytokine fusions
that have not dimerized to form the LAP shell around the
cytokine and are therefore biologically active. As discussed
previously, correct dimerization of the LAP molecules is an
absolute requirement to achieve full latency.
The appearance of LAP monomers is likely the result of
the overexpression of the recombinant protein which over-
taxes the cell glycosylation and folding machinery, so that
only a proportion of the expressed protein undergoes dimer-
ization, and the rest is secreted as monomers. Attempts to
supplement the relevant cellular enzymes (e.g., protein
disulfide isomerase) to increase the rate of formation of
the disulfide bonds required for dimerization of the LAP
molecules have not, thus far, been successful. These are not
insurmountable problems in terms of purification because
chromatography using a gel filtration column effectively
separates the three different species: aggregates, dimers, and
monomers (unpublished data). However, these issues do
present a problem when considering the total amount of
LAP protein that is “lost” in aggregates and monomers. In
practical terms, these species do in fact represent lost
material as these forms of the protein are not suitable for
therapeutic use.
The nature and extent of the glycosylation present on the
recombinant proteins produced is a critical factor for ensur-
ing an extended half-life in vivo, but is also important for
expression and purification. There are three N-glycosylation
sites in the LAP molecule, the sialylation of which are
necessary to prevent clearance of the protein from the
circulation [43]. Expression of the protein in CHO cells
appears to result in the production of LAP molecules that
have a range of levels of glycosylation. This is evident from
lectin chromatography using immobilized wheat germ
agglutinin (WGA). The elution profile of LAP fusion pro-
teins from WGA is indicative of LAP molecules with
differing levels of sugar moieties, as evidenced by the higher
concentration of N-acetyl-glucosamine required to elute
some of the LAP protein (unpublished data). However,
although this makes the use of lectin chromatography
unsuitable for the purification of latent cytokines, the gly-
cosylation levels achieved in CHO cells appear to be satis-
factory in terms of the half-life and tissue distribution of the
latent cytokine achieved in vivo as demonstrated by Adams
et al. [7].
Once the latent cytokine is cleaved by MMP, there is
release of the LAP part of the molecule that could conceiv-
ably have biological effects independent of those of the
cytokine. In fact, LAP has been shown to associate with
mature TGF-
b
in mice, thereby inhibiting the cellular effects
of the cytokine [100,101]. However, the expectation is that
latent cytokine cleavage, and subsequent release of LAP will
occur at sites of inflammation where there are high levels of
free radicals also abound. The capacity of LAP to bind to
TGF-
b
is abrogated by nitrosylation [102] and thus it is
highly unlikely that the LAP released by MMP cleavage will
antagonize TGF-
b
function. Despite early reports that LAP
does not have biological activity independent of TGF-
b
[103] a recent study has suggested that LAP could indeed
have immune functions such as the promotion of monocyte
chemotaxis and the blocking of inflammation in vivo [104].
However, in our experience, we have not seen any anti-
inflammatory effects by expressing free LAP by gene deliv-
ery in several models of inflammatory disease. No adverse
effects of the LAP released as part of treatment with latent
cytokines has ever been observed, but it is perhaps worth re-
investigating this issue in particular experimental settings to
show conclusively that LAP has no biological effects when
used as part of a latent molecule to treat disease.
16.8 CONCLUSIONS AND FUTURE
PERSPECTIVES
The latent cytokine technology is undergoing continual
development and refinement. Current work is focused on
repeating the therapeutic effects of latent cytokines on
disease pathology as demonstrated by Adams et al. [7],
this time using recombinant protein, rather than the gene
therapy approach used previously. This work also aims to
demonstrate cleavage of the latent fusion proteins at the sites
of disease in vivo. Release of active cytokine from the latent
complex by MMPs present in the biological fluids of patients
with inflammatory disease has already been demonstrated
by Adams et al. [7], and to show that this process also occurs
efficiently in vivo is an important step in the evolution of this
technology.
The latent cytokine technology described here was devel-
oped as a targeting mechanism for use with therapeutic
cytokines. The biological potency and pleiotropy of these
molecules dictate that the risk of unwanted side-effects are
extremely high and in many cases presented insurmountable
barriers to the use of cytokines as therapeutics. This tech-
nology has obvious advantages when applied to the use of
cytokines as therapeutics. However, the use of this technol-
ogy is by no means limited to use for the delivery of
cytokines to sites of disease. In fact, the LAP technology
could be used to “package” and deliver any therapeutic
molecule to sites of disease where MMP activity is up-
regulated.
There are already some applications of the latent cytokine
technology to other therapeutic molecules. Holle et al.
[105,106] have used this technology to develop a fusion
protein to deliver the pore-forming toxin from honey bees,
melittin, to tumor cells. As lytic peptides are fast-acting and
have short systemic half-lives [107] as well as demonstrating
high potency, the advantages of producing melittin as a
LAP-fusion protein are very similar to those for therapeutic
cytokines. The latent melittin produced is different from the
Search WWH ::
Custom Search