Transgenic models to study TGF-[beta] function in hematopoiesis

Studies of hematopoietic pathologies involving the growth factor TGF-β have provided
important evidence of its keyrole in the regulation of human hematopoietic stem/progenitor
cell quiescence, proliferation, and differentiation. The inactivation of one of the various genes
involved in the TGF-β signal transduction pathway may represent a possible mechanism by
which some early hematopoietic progenitors, which are normally quiescent, escape from cellcycle
inhibition. Abnormalities in the expression of TGF-β receptors have been described in
proliferative syndromes including both early myeloid and lymphocytic leukemia 1,2. In these
cases the loss of the growth inhibitory TGF-β signal might provide a selective advantage to
the malignant cell. Additional autocrine TGF-β production and thereby inhibition of
neighboring cells leads to an overgrowth of the malignant clone. In patients with
myeloproliferative disorders, reduced mRNA levels of the TGF-β signaling components
Smad4 and type II TGF-β receptor were reported 3-5 further establishing a role of abolished
TGF-β signaling in the pathogenesis of hematopoietic malignancies. The role of TGF-β in the
regulation of hematopoiesis has also been analyzed in vivo using different mouse models. For
example, the administration of TGF-β in mice revealed an inhibition of thrombopoiesis and
erythropoiesis 6. A variety of knockout mice have been generated to study the effect of TGF-
β in vivo. The most of these approaches were hampered by the early lethality of the knockout
like in the case of the Smad proteins and the TGF-β receptors I and II 7,8. Homozygous TGF-
β1 knockout mice have a 50% intrauterine death rate because of severe developmental
retardation. The other 50% die within several weeks after birth due to a severe inflammatory
autoimmune disease 9. Nevertheless, TGF-β knockout mice display defective hematopoiesis
with elevated platelet counts and reduced numbers of erythroid cells 9. However, as most of
the knockout approaches for TGF-β signaling components resulted in early embryonic
lethality, the exact functions of the different elements of the TGF-β signaling cascade in
hematopoiesis are still controversial.
In this thesis work I describe different transgenic approaches to gain insight into the function
of TGF-β signaling components in hematopoiesis, with a focus on megakaryopoiesis. In the
first part I describe the generation of a transgenic mouse strain for the tissue-specific deletion
of target genes in megakaryocytes and platelets. Many of the genes potentially involved in
megakaryopoiesis are difficult to study by conventional knockout approaches, as they are
ubiquitiously expressed and therefore their germline deletion is embryonically lethal. One
way to circumvent the obstacles of early embryonic lethality is the use of the Cre/loxP system
for tissue-restricted target gene deletion 10. Hence, we generated a transgenic mouse for the
megakaryocyte-specific expression of the Cre recombinase. As short plasmid based
transgenes are often hampered by position variegation effects, like mosaic expression or
transgene silencing, we decided to modify a large genomic DNA fragment using ETrecombination
in E.coli 11. The coding sequence of the Cre recombinase was placed under the
control of the Pf4 gene embedded in a 100kB bacterial artificial chromosome (BAC). The
modified BAC-insert was used to generate PF4Cre transgenic lines. Analysis of the resulting
transgenic lines revealed differences in tissue-specific expression of the Cre recombinase,
dependent on copy numbers. Accordingly, strains with low copy numbers revealed very
specific Cre expression in megakaryocytes and platelets, while strains with higher copy
numbers displayed ectopic Cre expression. The evaluation of excision efficiency in
megakaryocytes of the different PF4Cre strains revealed that the strain with 5 integrations
excised with 90%, whereas the strains with 1 or 2 copies excised with 60-70% efficiency.
However, I used these strains to delete the TGF-β signaling components type II TGF-
β receptor (TBRII) and Smad4 in megakaryocytes by mating the PF4Cre strains with either
TBRIIlox/lox or Smad4lox/lox mice. Homozygous offspring was analyzed for peripheral
blood counts. Surprisingly, no change in the numbers of circulating platelets was detected in
any of these mice in comparison to control mice. I confirmed these results using the
transgenic Mx1Cre mouse for inducible deletion of target genes in hematopoietic stem cells.
Again, no changes in the numbers of circulating platelets were detected neither in
TBRIIlox/lox-Mx1Cre mice, nor in Smad4lox/lox-Mx1Cre mice. Together these results argue
against an involvement of TGF-β signaling components in the onset of myeloproliferative
disorders and additionally reveal that TGF-β signaling is dispensable for functional
megakaryopoiesis.
In a second mouse model we intended to disrupt Smad-mediated TGF-β signaling in
hematopoiesis by the induced deletion of the TGF-β signal transducer Smad4. We used the
Mx1Cre transgenic strain to induce Smad4 deletion in the bone marrow of Smad4lox/lox-
Mx1Cre mice. Smad4 deleted mice developed a severe haemolytic anemia 4-5 weeks after the
induction of Cre recombinase expression, accompanied by extramedullary hematopoiesis and
splenomegaly. Anemia in Smad4lox/lox-Mx1Cre mice was not autoimmune-mediated as
revealed by a negative direct antiglobulin test (DAT). The hyperplasia of the spleens in
Smad4lox/lox-Mx1Cre mice was due to a massive increase of immature myeloid cells. FACS
analysis revealed the myeloid cells in the spleen are TER119high/CD71high erythroblasts, which
argues for a maturation block in erythropoiesis as the cause for anemia in Smad4lox/lox-
Mx1Cre mice. Transplantation of Smad4lox/lox-Mx1Cre bone marrow into lethally irradiated
C57BL/6 recipients revealed that the anemia is not transplantable and thus can be
compensated by host-derived factors. Furthermore, Smad4lox/lox-Mx1Cre bone marrow
transplanted recipients did not develop a wasting syndrome. This is in complete contrast to the
previously described induced deletion of TBRII and TBRI in TBRIIlox/lox- and TBRIlox/lox-
Mx1Cre mice. In both of these mouse models deletion of the TGF-β signaling caused a severe
inflammatory phenotype, which is transplantable. Together, these results implicate that the
autoimmune phenotype in TGF-β receptor deleted mice is not Smad-mediated, as Smad4 is
the quintessential for signaling through activated Smads.
In the last part of my thesis I describe the generation of a new tool to study gene function in
human hematopoietic stem/progenitor cells. For this purpose I took advantage of the rapid
advances in the RNA-interference field and the demonstrated capability of lentiviruses to
infect non-cycling human hematopoietic stem cells. I modified a lentiviral vector by the
insertion of a expression cassette for short-interfering RNAs (siRNA), which drives siRNA
expression under the control of the H1 promotor. Originally thought to target TBRII in human
hematopoietic stem cells, the system was first established to target the human p53 mRNA, as
a functional siRNA sequence for this target was available at that time. Human cord blood
derived CD34+ cells were infected with the lentiviral construct pWPXLp53si and p53 mRNA
from infected cells was analyzed by quantitative real-time PCR. Infection efficiencies were
typically around 50% as revealed by the enhanced green fluorescent protein reporter gene
(EGFP). Infected CD34+ cells not only revealed p53 mRNA reduction to 3-10% of the control
levels, but also functional p53 silencing was demonstrated by the increased resistance to
apoptotic stimuli of pWPXLp53si-infected CD34+ cells. We also demonstrated that the
lentiviral system was able to silence p53 in early hematopoietic progenitors by growing
infected CD34+ cells under long-term culture initiating cell (LTC-IC) conditions. In
summary, we revealed that lentiviral delivery of siRNA can be used for efficient and stable
gene silencing in human hematopoietic progenitors. This system will be very valuable to
study the function of key regulatory genes in human hematopoiesis.