48x36 Poster Template

1
Immunotherapy by Chimeric Antigen Receptor
Modified Hematopoietic Stem Cell Transfer Michael
Badowski, Tong Zhang Ph.D., Brian Pipes, Tom C.
Tsang Ph.D., David T. Harris Ph.D. The University
of Arizona, Tucson, Arizona
Outline of work to do
Specificity
Introduction
Signaling
The process of malignant transformation
results in a fundamentally changed cell.
Mutations to DNA and epigenetic changes
accumulate that eventually result in abnormal
protein expression patterns. Some of these
protein products are immunogenic and can be
recognized by the host immune system. Although
cancer reactive T-cells are often part of the
immune repertoire, most cancer patients dont
mount an effective anti-tumor T-cell response.
While some studies have focused on expansion of
the native repertoire, the development of an
entire new population of anti-tumor immune cells
may hold more promise. We hypothesize that a
chimeric antigen receptor (CAR) can provide
anti-tumor specificity, and that adoptive
transfer of CAR-modified stem cells offers us a
method to transfer and expand tumor specific
T-cells. This adoptive immunotherapy can
eliminate malignant cells or supplement
traditional therapies. Currently, in vitro
engineered primary T-cells are the most commonly
used source of cells for both laboratory and
clinical work. Although adoptive transfer of
engineered tumor reactive T-cells offers some
anti-cancer protection, the introduction of a
full length T-cell receptor (TCR) into primary
T-cells raises questions. Poor efficacy as well
as potential autoimmunity may loom with such an
approach. The use of a single chain TCR (scTCR)
in which the TCR Va and Vb segments are joined by
a flexible linker provides a solution to unwanted
specificities. In addition to our scTCR we are
developing a single chain antibody molecule
(scFv) to increase avidity to the tumor antigen
and avoid the potential limitation of MHC
restriction. Our lab has previously developed a
signaling cassette based on the CD28 and p56Lck
proteins which are prominent in the T-cell
signaling pathway. The single chain specificities
are linked to the signaling cassette that we have
shown to function in T-cells. With specificity
and signaling coupled, the chimeric antigen
receptor can be transduced into hematopoietic
stem cells (HSC) via a lentivirus vector.
Engraftment of the transduced HSC gives us a
method of in vivo expansion of tumor specific
immune cells. Active immunization may also
enhance the in vivo development of anti-tumor
T-cells with our engineered single chain
molecule. We are in the process of testing that
the combination of adoptive transfer of chimeric
antigen receptor modified hematopoietic stem
cells followed by active immunization could be an
integrated and effective approach for cancer
immunotherapy.
Antibody heavy and light chains were derived
from a her2/neu specific hybridoma obtained from
ATCC. Message RNA was isolated and reverse
transcribed followed by PCR. Sufficient
degeneracy in the primers allowed PCR expansion
of all possible vheavy and vkappa or vlambda
chains. Sequences were compared to known
antibodies in BLAST and IMGT. After single chain
molecules were designed, the 3D structure and
folding was modeled and verified by SWISS-MODEL
and SwissPDB viewer. The resulting scFv should
have identical specificity and affinity as the
original antibody for her2/neu antigen.
As the first element in the signal
transduction chain, the protein p56Lck is
critical in the signal transduction of
lymphocytes. The Lck is a tyrosine kinase of the
src family. The SH2 kinase domain will
phosphorylate ZAP-70 which then acts on other
cellular substrates. Incorporation of the Lck
protein allows our chimeric receptor to activate
T-cells. Conversely, cells without this signaling
cascade, such as macrophage or neutraphils, will
not be activated. Transduction of stem cells will
potentially allow development of the single chain
chimeric receptor in all immune cells. However,
only cells that utilize p56Lck in the signal
chain will be activated by the cognate antigen.
After engraftment, the presence of
transgenic HSCs will be verified and monitored by
PCR and flow cytometry. The proportion of
transgenic cells will be monitored in bone
marrow, blood, spleen, thymus and lymph nodes.
Animals will then be challenged with tumor and
monitored for tumor growth and survival. Since
the single chain receptors will be expressed on
both CD4 and CD8 T cells, we should be able to
avoid the limitation of lack of T cell help as
was observed with use of full-length TCR.
It is likely that single chain receptors will be
expressed on a percentage of non-T immune cells
as well as our target lymphocytes. These
receptors would then serve as adhesion molecules.
Although these cells would not have a functional
single chain receptor due to lack of Lck
signaling machinery, it is possible that
additional co-localization of NK cells,
monocytes, dendritic cells, etc. would result.
This may enhance the overall anti-tumor effect.
An important factor to consider is the
inherent differences between antibody and TCR. We
expect the higher innate affinity of antibodies
to provide an scFv with much higher affinity than
the scTCR. Additionally, the scTCR will be MHC
restricted. The higher affinity and unrestricted
nature of the scFv should provide overall more
effective anti-tumor therapy than the scTCR.
We will also be testing a different method of
increasing T cell numbers and anti-tumor
immunity. Since many common antigens are adept at
eliciting a robust T cell expansion, we will
attempt to utilize this to enhance therapy. After
transplant of single chain modified HSCs, the
resulting T cells should have two receptors a
normal full length TCR for cognate antigen and a
single chain for tumor antigen. Antigen from
common viruses, such as influenza, or alloantigen
can expand the T cell pool 20 or more with a
blast of specific clones. This method of
expansion holds the potential to be much more
effective than others that depend on weakly
immunogenic tumor antigens. The choice of
immunizing antigen and route of immunization will
help determine the degree of expansion and
location of the T cell pool.
Figure 1. Single chain molecule derived from
antibody (scFv). Homology modeling by Phyre using
SWISS-Prot database allows us to determine a
likely structure for the scFv.
Figure 2. In conjunction with an earlier scTCR,
various signaling modules were tested. The
highest signal as measured by IL-2 production in
transduced cells is highlighted. The
transmembrane portion derived from CD3z and
cytoplasmic element of CD28 allow for proper
co-localization of the receptor complex and
signal elements. The cytoplasmic portion of
p56Lck is needed for signaling. Consequently the
CD3z/CD28/p56Lck signaling module is being used
with current single chain molecules.
Light and heavy chains along with linker
areas were connected by splice overlap extension
(SOE-PCR). Similarly, the scTCR was developed
from a her2 reactive T cell clone provided by
Elizabeth Jaffe (Johns Hopkins Medical
Institutes). Single chain molecules were then
SOE-PCR attached to the signaling domain.
Production of Lentiviral vector
Hypotheses
Lentiviral vectors were chosen because of
their ability to transfect both dividing and
non-dividing cells. Since most hematopoietic stem
cells are non-dividing in vivo, this quality
critical for use with stem cells. The lentivirus
was originally derived from HIV-1 but has been
heavily modified. To reduce the possibility of
competent virus being produced, the genes for
tat, env as well as accessory proteins have been
removed. Additionally, three separate plasmids
encode the portions of the lentivirus making
correct recombination exceedingly unlikely.
Replacing the env protein, the virus is
pseudotyped with VSV-G protein. Studies have
shown VSV-G pseudotyped HIV-1 vectors can infect
hematopoietic cells.

• The transplantation of scFv modified HSC is a
  more effective anti-tumor therapy than scTCR
  modified HSC due to innate higher affinity for
  tumor antigen.
• Immunization with a strong commonly available
  antigen, such as influenza, is more effective at
  increasing anti-tumor T cell numbers than
  immunization with cognate tumor antigen
• Intrathymic transfer of single chain modified HSC
  is equally effective as transfer via bone marrow
  transplantation in generating a tumor specific
  immune system, and offers the advantage of T cell
  specific expression and limited effector cell
  lifespan

References
Zhang T, He X, Tsang TC, Harris DT Transgenic TCR
expression comparison of single chain with
full-length receptor constructs for T-cell
function. Cancer Gene Therapy. 2004
Jul11(7)487-96. IMGT, the international
ImMunoGeneTics information system
http//imgt.cines.fr (founder and director
Marie-Paule Lefranc, Montpellier, France).
Miyoshi H, Smith K, Mosier D, Verma I, Torbett
B. Transduction of human CD34 cells that mediate
long-term engraftment of NOD/SCID mice by HIV
vectors. Science,1999 283 682-8. Guex, N. and
Peitsch, M.C. SwissModel and the
Swiss-PdbViewer An environment for comparative
protein modeling. Electrophoresis. 1997 (18),
2714-2723.
Figure 3a (above). Plasmids encoding the single
chain transgene are co-transfected along with the
accessory packaging proteins. The 293T producer
cells then manufacture all the components of the
lentiviral vector. The single chain chimeric
receptor is labeled with a separately transcribed
GFP tag. This allows us to easily track
transduction efficiency of the purified stem
cells. Figure 3b (right). Transduced cells were
cultured for 10 days to evaluate stem cell
potency by colony forming unit (CFU) assay. Here
we show several examples of colonies under
brightfield and fluorescent microscopy. We can
see that not individual cells but entire colonies
are either GFP or GFP-. This is strong evidence
that stem cells have been virally transduced.