Biomedical Engineering Reference
In-Depth Information
It is also important to test the SEED scaffold in therapeutic
formats with fusion to other protein domains. To accomplish
this we have compared the clinically validated antibody C225
(Fc-based) with C225-SEED fusion proteins in both bivalent
andmonovalent formats (formats illustrated in Figure 37.3A).
For example, an accelerated thermal stability study was
conducted by holding C225-Fc, C225-SEED (bivalent), or
C225-AG/GA (monovalent) proteins at 50
C over a period of
7 days, then analytical SEC profiles of the postincubation
samples were compared to a control held at 4
C to assess
changes in aggregation or loss of protein due to precipitation.
The amount of C225-Fc did not change over 7 days at 50
C,
while a small fraction of C225-SEED and C225-AG/GA
proteins were lost (Figure 37.4B). Differential scanning cal-
orimetery (DSC) was also used as an alternate method to
measure the stability of the proteins. DSC analysis showed
that the SEED CH3 domain has a lower melting temperature
than the CH3 domain of Fc. However, themelting temperature
of the SEED CH3 domain is
C225-Fc and C225-SEED (bivalent proteins) and mono-
valent C225-AG/GA and C225-GA/AG all had similar
potency to induce ADCC against A431 target cells (Figure
37.5B).
CDC effector function was tested using a different anti-
body, hu14.18 Fab, which is known to induce CDC against
M21 cells [20]. Bivalent hu14.18-Fc and hu14.18-SEED
induced CDC against M21 target cells with similar efficacy
(Figure 37.5C). The results from ADCC and CDC assays
thereby demonstrated that SEED retains antibody-like effec-
tor functions.
Mouse PK studies (Figure 37.5d) were performed to test
if SEED-based fusion proteins retain long in vivo half-life
similar to antibodies. By comparing Fc and SEED scaffolds,
and C225-Fc, C225-SEED, and C225-AG/GA proteins, we
found that SEED-based proteins and Fc-based proteins had
similar PK properties in mice, suggesting that engineering
the SEED platform did not
impair its in vivo half-life
68
C, which is still compatible
for use as a therapeutic protein, as neither protein productivity
nor stability (tested by incubation in different pH or at 50
C)
was greatly affected. An example comparing theDSC profiles
of C225-Fc, C225-SEED (bivalent), and C225-AG/GA
(monovalent) is shown in Figure 37.4C.
As a final example of our physical characterization of the
SEED design, we show in Figure 37.4D the crystal structure
of SEED, determined to 2.5 A
. This crystal structure con-
firms that the SEED design, as predicted, produces AG and
GA SEED domains that each fold in a similar way to Fc CH3
domains. Furthermore, the crystal structure physically con-
firmed the specific interaction interface between AG and GA
SEED domains.
In summary, our biophysical assays and structure deter-
mination confirm that SEED has physical properties com-
patible with therapeutic applications.
compared to IgG.
All together, our experiments (exemplified in Figure
37.5) showed that SEED retains the functional and biologi-
cal properties of therapeutic antibodies while efficiently
forming heterodimeric proteins.
37.3 POTENTIAL THERAPEUTIC APPLICATIONS
The evolution of biotherapeutics has been driven by both
technological advances as well as by improved understand-
ing of disease mechanisms. While the first effective protein-
based drugs consisted of naturally derived molecules such as
insulin and growth hormone purified from native sources,
more recent developments in molecular biology have
allowed the generation of engineered proteins suited to
specific needs and purposes. In general, biotherapeutics
have proven to be a very important drug class based on
safety and the efficacy provided by their exquisite specific-
ity. Antibodies in particular were identified early as having
great potential in targeting disease-associated processes
because of this characteristic. However, early antibodies
generated in nonhuman hosts proved to be highly immuno-
genic, which greatly hindered their clinical utility. More
recently, technologies allowing the humanization of non-
human antibodies, together with the generation of new
display technologies to generate fully human antibodies,
have largely overcome this hurdle and, as a result, antibodies
have rapidly become a major category of disease modifying
therapeutics. While the list of clinically validated antibody
targets continues to grow, it is also clear that in many cases
the clinical efficacy observed with conventional antibodies
is often much poorer than originally predicted. The reasons
for this are complex and disease specific but may include the
following: (1) multiple and often redundant signaling path-
ways involved in driving complex diseases are inadequately
37.2.4 Biological Properties
In addition to biophysical properties, we have tested the
biological properties of SEED-based proteins. An important
part of the SEED design is that it should retain desirable
biological properties of therapeutic antibodies, such as target
binding, effector functions, and a long half-life in vivo
(Figure 37.5). We again used the clinically validated anti-
body C225 to assess if SEED retains therapeutic antibody
properties. The C225 antibody has been shown to bind to the
EGF receptor extracellular domain and to inhibit EGF
binding to its receptor. Figure 37.5A shows in a ligand
competition assay that bivalent C225-SEED binds to
EGFR and inhibits EGF binding in a similar way as
C225-Fc, while monvalent C225-AG/GA and C225-GA/AG
bind to EGFR but have a higher K
i
in this assay.
An important feature of antibodies is their effector func-
tions, specifically antibody-dependent ADCC and CDC.
Search WWH ::
Custom Search