In vivo imaging of graftversus-host disease and graftversus-leukemia - Immune Biology of Allogeneic Hematopoietic Stem Cell Transplantation - page 61

and in vivo [18,19] . Alternatively, one can simply isolate a cell type of
interest from a transgenic mouse expressing luciferase. One such mouse,
developed jointly by the Contag and Negrin laboratories, places expres-
sion of firefly luciferase under the control of a chicken β-actin promoter
and the cytomegalovirus enhancer resulting in expression of luciferase in
all hematopoietic cells of the mice, named L2G85, which are on an FVB/N
(H-2 q ) background [20] . Since then, these transgenic mice have been back-
crossed onto other genetic backgrounds (C57BL/6, BALB/c,) as well as var-
ious knockout mouse strains, by our group and others. Using cells isolated
from these mice, a minimum of 100-1000 cells can be detected at specific
anatomical sites [21] . However, as with all optical imaging methodologies,
the signal intensity can be affected by tissue properties [22] , and other fac-
tors such as the source of bioluminescence in the animal, absorption of
light by hemoglobin, and the efficiency and sensitivity of the image collec-
tion device.

Exploring GVHD with bioluminescence

Bioluminescence imaging has revealed many important aspects about the
biology of GVHD induction and progression. In a major-MHC mismatch
model, alloreactive T cells were observed to infiltrate lymph nodes and
spleen within 24-48 hours after transplantation, proliferate, and migrate
to GVHD targets such as gastrointestinal tissue (GIT) and skin by day 6
( Figure 4.1 A) [23] . We observed CD4 + T cells to be the first to infiltrate all
lymphoid organs, followed (several days later) by CD8 + T cells. As previous
studies indicated that Peyer's patches are important for GVHD induction
[24] , we analyzed the proliferation of these cells by both histology and BLI.
Although these structures can be very difficult to visualize in irradiated mice,
they were clearly seen by BLI after HCT [23] . This study was important in
that it identified the kinetics of migration and proliferation of T cells during
GVHD induction, and further analysis could be conducted, for example,
by re-isolating T cells from the appropriate tissues at different time points
post transplant. We found that donor-derived T cells infiltrated secondary
lymphoid organs, became activated within the lymphoid organs during
these early time points and then began to proliferate, which was associated
with a number of phenotypic changes including for example the downregu-
lation of CD62L, and upregulation of α4β7, a gut-homing receptor [23] .

62

One important hypothesis in the field was whether the site of T-cell priming
imprinted the T cells to infiltrate and attack specific tissues. This was sup-
ported by a study by Mora et al. which found that APCs from particular sites,
such as Peyer's patches, resulted in T-cell activation and migration to the
GIT, and not the skin [25] . We directly addressed this question with several
experimental models where animals were prepared which lacked Peyer's
patches, lymph nodes or spleens (or a combination of all of these), as well
as using antibodies to block entry into secondary lymphoid organs [26] .
Contrary to a report that GVHD was prevented in Peyer's patch knockout
recipients in the first 30 days after HCT [24] , we could, via BLI, visualize that
remaining secondary lymphoid organs could compensate for those tissues
that were lacking. These findings indicated that there are multiple redun-
dant priming sites throughout the body and that only by blocking access to
all of these sites can one prevent the onset of GVHD.