Identification of novel genes involved in the commitment of endodermal cells to the thymic epithelial cell fate

The thymus provides the microenvironment for the maturation and selection of the majority of
peripheral T cells. Endodermal cells of the ventral aspect of the third pharyngeal pouch (3rdpp) at
10.5 days of mouse gestation (E10.5) adopt a thymic epithelial cell fate while cells of the dorsal
part of the 3rdpp give rise to the parathyroid glands. To identify novel genes potentially involved
in the commitment of endodermal cells to the thymic epithelial cell fate, the transcriptome of the
3rdpp was compared to that of the 2ndpp and to that of the 4thpp with the help of laser
microdissection and gene expression profiling (microarrays). Similarly, the transcriptome of the
ventral circumference of the 3rdpp was in addition compared to its dorsal counterpart. Taken
together, fifty genes were identified by microarray and confirmed by quantitative RT-PCR as
being differentially expressed between the ventral and the dorsal aspects of the 3rdpp while 12
genes were revealed as being upregulated if not exclusively expressed in the entire 3rdpp when
compared to the 2ndpp. Among the genes revealed to be differentially expressed within the 3rdpp,
two expressed sequence targets (ESTs) were found as being expressed in the ventral aspect of the
3rdpp but not in the dorsal side of the pharyngeal pouch while 5 genes (Tbx1, FoxA1, FoxA2,
Sfrp2 and CXCL12) demonstrated an upregulation in the dorsal aspect of the 3rdpp. Analysis of
fetal thymic tissue at different stages of development (E10.5, E12.5 , E16.5, E18. 5) demonstrated
that 8 of the candidate genes (Nrxn1, WIF1, Bmp4, Fst, c-Myc, Phlda2 and Flrt3) further
examined were expressed throughout development. Analysis of embryos at E10.5 by
immunohistochemistry for the protein expression of CCL21, Meox2, CD44, WIF1, Fst, Phlda2
confirmed an upregulation if not an exclusive expression in the 3rdpp. Moreover, an analysis of
the thymic expression revealed that two of the candidate genes examined (WIF1 and Flrt3) are
expressed in adult primary thymic epithelial cells but not in thymocytes in contrast to other
candidate genes analysed (e.g. Nrxn1, Bmp4, Fst, c-Myc and Phlda2). Other candidate genes like
Sp8 and Phlda2, for which deficient embryos were available to us, respectively, were analysed
for their thymic architecture by immunohistochemistry for several markers (e.g. K5, K8 and
CD45). However, any significative difference in comparison to wild type littermates could be
noticed for these two genes. A functional analysis by Fetal thymic organ cultures (FTOCs) of
E13.5s in the presence of human WIF1 recombinant proteins revealed that WIF1 can positively
influence the overall cellularity of thymocytes if not expressed at too high levels. However, in
contrast to FTOCs in presence of Bmp4 proteins, the in vitro overexpression expression of WIF1
did not inhibit the normal development of thymocytes in these FTOCs.
In conclusion, this project allowed to identify several candidate genes using microdissected
tissues to not only provide global information on gene expression during early development of
the thymus but it also provides novel targets to study the inductive signalling pathways that direct
the patterning and the differentiation of endodermal cells to the thymic epithelial cell fate. In that
regard, several of the candidate genes are known to be involved in Wnt, Tgfβ2 signaling
pathways or other singaling pathways, predicting that several pathways seem to play a role in
early thymus organogenesis.