SIGNAL CONVERTER PROTEINS - Fusion Protein Technologies for Biopharmaceuticals: Applications and Challenges - page 440

coinhibitory) trans signals concurrently delivered by APC.
Specifically, costimulator (either B7-1 Fc g 1 or Fc g 1 4-
1BBL) and pro-apoptotic (Fc g 1

blockade of this costimulator pathway using soluble

CTLA-4 Ig increases cellular susceptibility to Fas-mediated

apoptosis [21,40]. As another CTLA-4 derivative, CTLA-
4

FasL) Fc fusion proteins
were quantitatively painted, in varying ratios, onto surrogate
APC precoated with palmitated protein A. By this approach,
we demonstrated that at a given level of TCR triggering,
the quantitative balance between costimulator (B7-1 or
4-1BBL) and FasL dictates the magnitude of the prolifera-
tive T-cell response. Furthermore, when the costimulator
density is kept constant, there is also a quantitative balance
between TCR-directed and FasL signals.
Thus, AVC engineering and coinhibitor painting nucle-
ated a compelling concept—fusion proteins can be used to
functionally engineer immunoregulatory cells. The protein
paints invoked in these earlier studies required bringing the
paint into direct contact with the cells to be painted, whether
ex vivo or in confined in vivo compartments (such as tumor
beds). This beckoned the development of next-generation
fusion protein paints that could be injected systemically and
would home to their cellular targets in vivo.

FasL shares this capacity to potentiate Fas-induced apo-
ptosis through costimulator interference, but with an added
twist—in this instance, both the costimulator blockade
function, and the FasL signaling function that is being
potentiated, are uniquely integrated into a single fusion
protein.
Other CTLA-4 FasL functional attributes stem from its

association with cell surfaces. This chimeric protein anchors
FasL to cell surfaces, thereby tapping into the higher efficacy
of surface-associated FasL [41,42]. Yet, as a two-faced,
Janus-like entity, CTLA-4

FasL is at the same time anchor-
ing CTLA-4 to activated T-cell surfaces. This in-and-
of-itself may have some interesting side benefits. CTLA-4
on T cells has been ascribed a cell-extrinsic, antigen-
independent mechanism of action whereby it can “catch”
and trans-endocytose its B7 (B7-1 and -2) ligands away from
antigen-presenting cell surfaces, thereby preempting CD28
costimulation [43,44]. The CTLA-4 anchored to Fas þ cells
via CTLA-4 FasL could well do the same. Furthermore, the

CTLA-4 end of CTLA-4 30.3 TRANS SIGNAL CONVERTER PROTEINS FasL has yet another potential, that

is, to reverse signal through its cognate B7 ligand and
thereby induce B7-bearing cells to produce suppressive
indoleamine 2,3-dioxygenase (IDO) [45].
In essence, CTLA-4

A novel fusion protein class, designated as trans signal
converter proteins (TSCP), was developed with in vivo
engineering of AVC in mind. The driving concept was to
combine within a single fusion protein homing and effector
elements, together geared toward converting an intercellular
signal between two cells. By judiciously choosing the
homing element, one can elicit additional functional effects
that reinforce the efficacy of the effector element. This was
in evidence in our paradigmatic AVC-generating TSCP,
CTLA-4

CoI
TSCP, can be thought of as in vivo superpaints, designed
not just for coinhibitor painting per se, but also for their
added complementary functions. An ability to generate AVC
in vivo, their core function, is piggy-backed onto their
inherent costimulator blocking capacity. Costimulator
blockade has been explored in great depth, and has made
it into the clinic. The B7-blocker CTLA-4

FasL, as well as other CoSR

FasL [37]. The CTLA-4 (CD152) component
within this fusion protein serves a dual purpose, targeting
the protein to APC surfaces bearing counter-receptors for
this homing domain [B7-1 (CD80) and B7-2 (CD86)], and at
the same time, passively blocking their costimulator func-
tion. The FasL (CD95L) component of this fusion protein
sends an active inhibitory, pro-apoptotic trans (intercellular)
signal to its cognate Fas receptor (CD95), which is upregu-
lated on activated T cells. Thus, this first TSCP, with a
costimulator

Ig has received
the most attention, with its well-documented capacities to
induce anergic hyporesponsiveness and prolong allograft
survival [46,47]. However, CTLA-4 Ig has its limitations,

including inability to reproducibly establish a permanent
state of peripheral tolerance in vivo [48] and to prevent graft
rejection in various transplantation settings [49]. By reach-
ing beyond passive blockade and melding it with active
inhibition, CTLA-4 FasL provides a route for overcoming

CoI) configura-
tion, functions as a signal converter, as it exchanges an
active inhibitory (apoptosis-inducing) signal for a T-cell-
activating costimulator one that is being passively blocked.
Interestingly, there are other functional advantages that
inhere in CTLA-4 FasL, arising from known functional
properties of the component proteins and their associated
signaling axes. B7 (CD80, CD86) proteins bind to CD28 and
CTLA-4 (CD152), thereby transmitting stimulatory and
inhibitory signals to T cells, respectively [38]. Increased
B7-triggering of the CD28 costimulator receptor interferes
with Fas-mediated apoptosis [39], and conversely, passive

receptor coinhibitor (CoSR CTLA-4 Ig's limitations. This is especially important where

there are multiple costimulator pathways in play.
CTLA-4

FasL exhibits striking potency in vitro, when
compared to CTLA-4 Ig or soluble FasL, alone or in com-

bination [37,50-52]. Interestingly, CTLA-4

FasL not only
deletes activated T cells, but also induces anergic prolifera-
tive hyporesponsiveness, as evidenced by its ability to
inhibit secondary responses in mixed lymphocyte reactions
[50]. This dual effect fits with other evidence in the literature
pointing to both Fas-mediated deletion [15,20,21,53]
and anergy [54]. CTLA-4 FasL-mediated inhibition of allo-

geneic responses was documented in the in vivo context as