Line 1. Role played by diaglycerol kinase isoforms during cell
growth and differentiation of myoblasts (C2C12 cells) and neural
(PC12) cells.
Line 2. PI3K/Akt/mTOR signaling as a new therapeutic target for
innovative strategies for treatment of patients with acute myeloid
and lymphoblastic leukemias.
Line 1. Several DGK isoforms,
including -alpha, -gamma, -delta, -zeta, -iota, and -theta have
been reported to be present within the nucleus. DGK isoforms can be
resident in the nucleus (for example DGK-theta) or translocate
there in response to agonists (DGK-alpha). A functional NLS has
been identified only in DGK-zeta and -iota. This motif of DGK-zeta
is similar to the phosphorylation-site domain of the MARCKS
protein. In COS-7 cells overexpressing a cDNA lacking the sequence
of DGK-zeta homologous to the phosphorylation-site domain of
MARCKS, the enzyme was almost entirely excluded from the nucleus.
This motif contains a serine residue (Ser 265) which is
phosphorylated by PKC-alpha and -gamma. In A172 cells
overexpressing these PKC isozymes, the amount of DGK-zeta present
in the nucleus was reduced, whereas treatment with phorbol esters
(that downregulate the amount of the two PKC isoforms) resulted in
enhanced intranuclear localization of DGK-zeta as well as in an
increase in nuclear DGK activity. Similar results have been
reported for DGK-iota. As to the subnuclear localization of DGK
isoforms, it has been shown that at least some isoforms (-alpha,
-theta) display a discrete intranuclear localization. It has been
demonstrated that DGK-theta mainly localizes to nuclear speckles, a
domain which is enriched in factors involved in mRNA splicing. The
association of DGK-theta with nuclear speckles appears extremely
interesting, because several elements of the phosphoinositide cycle
are present within this subnuclear compartment. These elements
include: phosphoinositide-specific phospholipase C-beta1,
PIPKIalpha, phosphatidylinositol 4,5 bisphosphate, phosphoinositide
3-kinase C2alpha, and SHIP-2.Concerning the function(s) played by
nuclear DGK, there is evidence that they are involved in the
control of nuclear DG levels. The group of Prescott compared
nuclear DG levels in control A172 cells with levels in cells with a
high amount of nuclear DGK-zeta (i.e. those treated with phorbol
esters, see above). When A172 cells were exposed to epidermal
growth factor for 10 min, nuclear DG mass in cells not pre-treated
with phorbol esters increased about 2.5-fold above the basal
levels, whereas nuclear DG levels in cells pre-treated with phorbol
esters increased only 1.3-fold. Furthermore, in cells
overexpressing an inducible DGK-zeta, the doubling time increased
about two-fold over controls. Exposure to insulin-like growth
factor-1 (IGF-1) of quiescent Swiss 3T3 cells resulted in the
stimulation of a nuclear DGK activity, but not of the DGK activity
present in whole cell homogenate. Two pharmacological inhibitors of
DGK markedly potentiated the mitogenic effect of IGF-1. Thus, these
findings confirmed that nuclear DGK plays a key role in regulating
the levels of DG present in the nucleus and that DG is a key
molecule for the mitogenic effect which IGF-1 exerts on Swiss 3T3
cells. Also the group of Raben has shown that DGK-theta was
responsible for the increased nuclear DGK activity which followed
stimulation of quiescent IIC9 cells with alpha-thrombin. The
function of DGK-theta in IIC9 cell nuclei would be to increase PA
production early (3-5 min) after challenge with alpha-thrombin.
Afterwards (from 10 min to 1 h of stimulation) activated
phospholipase D would ensure PA production. The reason for such a
switch is unclear, even though we may imagine that, depending on
the source, PA is generated in different subnuclear domains and/or
possesses distinct molecular species of fatty acids. Nevertheless,
an increased activity of nuclear DGK-theta has been recently
demonstrated to occur in nerve growth factor (NGF)-stimulated PC12
cells, i.e. a classical model for neural differentiation. This
might indicate that the functions of nuclear DGKs are not only
related with cell proliferation.
The function of PA (if any) in the nucleus is completely unknown.
Since the nuclear matrix is considered by some investigators to be
the equivalent of the cytoskeleton, it might be that nuclear PA is
involved in regulating some aspects of the structure of this
nuclear scaffold, perhaps through actin polymerization.
Interestingly, upon NGF treatment DGK-theta associates with the
nuclear matrix.
As far as the regulation of nuclear DGK isozymes is concerned, our
knowledge is very limited. Nuclear DAG-theta activity could be
blocked by active RhoA , but the regulation mechanisms of other
nuclear isoforms still await elucidation. Most likely,
phoshorylation and compartmentalization of DGKs are of fundamental
importance at the nuclear level to ensure correct termination of DG
signaling, especially if one considers the fact that in intact
nuclei exogenously added DGK phosphorylates only the
detergent-resistant, nuclear matrix-associated DG species derived
from phosphoinositide hydrolysis but not DG from PC (which
constitutes the bulk of nuclear DG and is located in the nuclear
envelope). We have recently demonstrated the important role played
by nuclear DGK-zeta in regulating myogenic differentiation and cell
cycle progression in C2C12 mouse myoblasts (Evangelisti et al., J Cell Physiol. 2006
Nov;209(2):370-8; Evangelisti et al., FASEB J. 2007
Oct;21(12):3297-307). Experiments are currently underway to
identify the targets of nuclear DGK-zeta which are important for
its roles (see also Hasegawa et al.,
J Cell Biochem.
2008 Oct 15;105(3):756-65).
Line 2. Since 2003
several published papers have addressed the issue of constitutive
PI3K/Akt/mTOR signaling activation in AML cells. There is general
consensus over the fact that this pathway is important for the
survival of AML blasts, including leukemic stem cells. However,
there are several unresolved issues. Indeed, it is not clear
whether activation of this pathway represents a positive or
negative prognostic factors for de novo AML patients (Tamburini et
al, Blood, 110, 1025-1028, 2007). Moreover, the downstream targets
of this signal transduction network are poorly understood in AML
cells and there exist contradictory reports regards this issue. For
example, glycogen synthase kinase 3beta (GSK3beta) was found
phosphorylated in AML cells with upregulated Akt function (Cheong
et al, Br J Haematol 122, 454-456, 2003). However, others found
that downregulation of PI3K/Akt signaling in AML primary cells did
not result in GSK3beta dephosphorylation (Grandage et al, Leukemia
19, 586-594, 2005). The same holds true for p70S6K which was found
to be phosphorylated in a PI3K/Akt-dependent fashion in AML primary
cells (Xu et al, Blood 102, 972-980, 2003).
Nevertheless, others failed to detect any relationship between
PI3K/Akt signaling upregulation and p70S6K phosphorylation in AML
blasts (Grandage et al, Leukemia 19, 586-594, 2005).
As to clinical studies, rapamycin gave some promising results when
used alone in a small study where it was able to induce a
significant and rapid clinical response in 4 of 9 patients with
either refractory/relapsed de novo or secondary AML (Recher C et
al, Blood 105, 2527-2534, 2005). Nevertheless, we do not know why
the drug was not efficacious in the other patients.
Other conflicting results regard the possibility that mTOR
inhibitors would either downregulate (Zeng et al, Blood 109,
3509-3512, 2007) or upregulate p-Akt levels in AML cells [Tamburini
et al, Blood. Sep 18; (Epub ahead of print) PMID: 17878402,
2007].
Given its aims, we anticipate that the results from this project
could further in a considerable manner our general knowledge about
PI3K/Akt/mTOR signaling in AML and could clarify some of the above
mentioned issues.
The phase II clinical trial results which is part of this project,
will indicate whether a combined treatment which includes a
rapamycin analogue and clofarabine is efficacious and safe for
treating older AML patients with relapsed/refractory AML. For the
first time we will try to correlate the clinical response with the
levels of proteins/phosphoproteins (Akt and 4E-BP1) which could be
important for determining cell sensitivity to mTOR inhibitors
(Kurmasheva et al, Br J Cancer 95,955-960,2006). Moreover, we will
investigate the expression of P-glycoprotein which could negatively
affect the response to therapy, as well as the state of p53
(wild-type vs. mutated) which could also have a detrimental effect
on clinical response (Kurmasheva et al, Br J Cancer 95, 955-960,
2006). It should be reminded here that in B-chronic lymphocytic
leukemia cells, the efficacy of fludarabine (a deoxynucleoside
analogue similar to clofarabine) was potentiated by nutlin-3a, an
inhibitor of MDM2-p53 interactions which caused an increase in the
levels of p53 (Coll-Mulet et al, Blood 107, 4109-4114, 2006). In
this connection, it is known that Akt-mediated phosphorylation of
MDM2 enhances its interactions with p300, allowing ubiquitination
and degradation of p53 (Zhou et al, Nat Cell Biol 3, 973-981,
2001).
Furthermore, SNP analysis, as well as gene expression profiles,
could reveal some specific patterns which are linked with
resistance/response to therapy, as indicated for other forms of
cancer and related therapies (Hedge et al, Mol Cancer Ther 6,
1629-1640, 2007; Xu et al, Oncogene 26, 2925-2938, 2007). The same
holds true for the proteomic/phosphoproteomic studies which will be
carried out in these patients (Posadas et al, Cancer 109,
1323-1330, 2007; Posadas et al, Cancer 110, 309-317). Taken
together, the results coming from these studies could bring
important information with regards to the reasons of the outcome of
the therapeutic treatment of older patients with AML.
It should be emphasized that no translational studies have been
carried out so far regarding the effects of a
Temsirolimus+clofarabine combined treatment using AML cells.
Actually, there is a paucity of studies performed with clofarabine
alone in AML cells (Lindemalm et al, Haematologica 88, 324-332,
2003). Thus, results from this set of experiments could help to
understand what is the best sequence for the combined treatment, if
the two dugs induce apoptosis, if the proapoptotic ASK1/JNK/c-Jun
is activated, and whether or not the combined treatment results in
hyperactivation or downregulation of p-Akt levels. All of these
findings could then be advantageously used in the future to design
better and more effective therapeutic protocols.
The gene expression profile studies carried out in AML cell lines
and AML blasts with constitutively active PI3K/Akt/mTOR signaling
could bring valuable information about the genes which are
downstream targets of this signaling network. This kind of studies
have never been performed in AML cells, however in other cell types
they have provided important information as to the genes which are
under the control of this pathway (Sivertsen et al, Br J Haematol
135, 117-128).We expect that also the proteome/phosphoproteome
analysis studies will shed some light about what targets are truly
downstream of PI3K/Akt/mTOR signaling in AML cells, as they have
done in other cell lines (Barani et al, J Proteosome Res 5,
1636-1646, 2006; Vandermoere et al, Moll Cell Proteomics 6,
114-124, 2007).
If one considers how severe is the prognosis of patients with AML
(both young and old), it is possible to understand why at present a
growing interest surrounds the development of new targeted
therapies, which could be less toxic and more effective. This
growing interest is testified by the ever increasing number of
manuscripts, published in top rated international journals, which
aim to highlight aberrantly regulated signal transduction pathways
in AML cells, and how they could be targeted for the development of
the therapeutic strategies.
All in all, we expect to provide a better picture of PI3K/Akt/mTOR
signaling in AML cells, and also to gain valuable insight into the
genes/proteins which could be involved in sensitivity/resistance to
mTOR inhibitors such as CCI-779 (Temsirolimus). The data could pave
the way for future functional studies (downregulation,
overexpression of a given protein) aimed to better understanding
the effective role(s) played by some molecules in determining the
outcome of therapeutic protocols. This information could then be
translated into the clinic, to improve future therapeutic
strategies which will include pharmacological inhibitors of this
signaling pathway.