ARTICLE
Introduction
Apoptosis is a genetically controlled, cell death process, which is
important at various developmental stages, as well as for the maintenance
of tissue homeostasis [1]. During the past few years, many of the key
players in this process have been identified, including receptors, adapter
proteins, proteases and other positive and negative regulators. One of
the positive mediators of apoptosis, which was cloned in our laboratory,
is DAP-kinase [2]. DAP-kinase (DAPk) was discovered using a functional
approach to gene cloning, based on transfection of mammalian cells with
an anti-sense cDNA library, and subsequent isolation of death protective
cDNA fragments [2, 3]. Anti-sense DAPk RNA expression protected HeLa cells
from interferon-g-induced cell death. DAPk is a pro-apoptotic Ser/Thr
kinase that participates in a wide array of apoptotic signals including
IFN-g, TNF-a, Fas, activated c-Myc, TGF-b and detachment from extracellular
matrix [2, 4-6]. This microfilament-associated Ca2+/calmodulin
(Ca2+/CaM)-dependent kinase has a unique multidomain structure
that mediates protein-protein interactions such as ankyrin repeats, and
the death domain. The death promoting effects of DAPk depend on its catalytic
activity, the correct intracellular localization, and on the presence
of the death domain [2].
The apoptotic function of DAPk must be under tight control, to ensure
on one hand its silence under normal growth conditions, and to allow,
on the other hand, rapid activation in response to the appropriate apoptotic
signal. While extensive work was undertaken in characterizing the apoptotic
systems where DAPk functions, with respect to the structure/function features
of the protein, and to the impact of its loss in cancer development [2,
4, 5, 7-9], the mechanism of DAPk activation has only recently been deciphered
in detail (Figure 1).
Catalytic properties of
DAPk
The identification of the myosin light chain (MLC) as a substrate of
DAPk facilitated the performance of in vitro kinase assays enabling
the analysis of different aspects of its catalytic activity and its mode
of regulation. Structure/function dissection of the protein revealed two
molecular layers of regulation, which control the catalytic activity and
hence the pro-apoptotic effect of DAPk. The first comprises the CaM-regulatory
segment, which has an auto-inhibitory effect on the catalytic activity
and is relieved by binding to Ca2+-activated CaM [4]. Consistently,
the mere deletion of this segment from DAPk (DCaM mutant) generated a
constitutively active kinase, which displayed CaM-independent substrate
phosphorylation in vitro and promoted the apoptotic activity in
vivo [4].
In addition to the basic regulatory mechanism of CaM-mediated relief
from auto-inhibition, a second layer of regulation that is mediated by
an autophosphorylation mechanism was recently revealed as detailed below
[10]. It emerged from previous work, which showed that DAPk undergoes
autophosphorylation [4]. Since the pro-apoptotic effect of DAPk depends
on its catalytic activity, we wondered whether any of these autophosphorylation
sites may reside in the kinase domain, and whether they provide an additional
layer of regulation, modulating the catalytic activity. We mapped a functionally
relevant autophosphorylation site to Ser308 which resides within
the CaM regulatory segment of the enzyme. In vitro kinase assays
revealed that the autophosphorylation of Ser308 is Ca2+/CaM-independent
and is strongly inhibited by the addition of Ca2+/CaM [10].
The biochemical implication and functional relevance of the phosphorylation
status of Ser308 was established in this study by two, single
point mutations mimicking the different phosphorylation forms of this
specific site. DAPk in which Ser308 was converted to alanine
served to simulate a dephophosphorylated state of Ser308, while
the substitution to aspartic acid mimicked the phosphorylated form of
Ser308 [10].
Biochemical implications of autophosphorylation
on Ser308
Biochemical analysis of these mutants revealed two distinct properties
which are influenced by the phosphorylation status of serine 308. The
first comprised an in vitro elevation in the basal Ca2+/CaM-independent
catalytic activity upon dephosphorylation (Ser to Ala substitution). This
Ca2+/CaM-independent activity was minimized by the Ser to Asp
substitution [10]. It is suggested that Ser308 autophosphorylation
strengthens a "locking device", probably generated by the interaction
of the CaM-autoinhibitory domain with the catalytic cleft. Absence of
the phosphate group, conversely, weakens this "lock" and therefore partially
relieves the autoinhibition, resulting in Ca2+/CaM-independent
activity. In fact, a model structure of the catalytic domain of DAPk,
which was constructed based on the X-ray structure of phosphorylase kinase
in complex with a substrate peptide (PDB code 2phk), predicts that the
peptide derived from the CaM-regulatory region binds in the active site
with Ser308-PO3 positioned next to the ATP binding
P-loop [10]. According to this model, the phosphate moiety of Ser308interacts
favorably with this loop and with the positively charged Lys141.
Thus, the ionic interactions between these residues may stabilize the
phosphorylated CaM regulatory domain in a locked position in the catalytic
cleft.
The second biochemical property, which is governed by the Ser308
phosphorylation status is the CaM binding activity. The prevention of
Ser308 phosphorylation resulted in a marked elevation in CaM
binding activity as assessed by two different CaM binding assays in
vitro [10]. Conversely, the opposing substitution to aspartic acid
very strongly reduced the binding to CaM. The elevation could result from
enhanced accessibility of the dephosphorylated form to interaction with
CaM due to the "weakened lock", and/or from an intrinsic property of the
dephosphorylated segment which permits a better interaction with CaM.
Thus, we concluded that phosphorylation of Ser308 inhibits
different aspects of the catalytic activity of DAPK in vitro.
DAPk in apoptosis
Next it became of interest to test whether DAPk is subjected to dephosphorylation
in response to an apoptotic stimulus in cells. We verifyed the Ser308
autophosphorylation in vivo by using anti-phospho-serine antibodies,
and showed that DAPk indeed undergoes autophosphorylation at Ser308
in growing cells. Of importance was the finding that Ser308
undergoes dephosphorylation in response to ceramide, which utilizes active
DAPk as part of its mode of action in some cells [10].
The cellular relevance of the two phosphorylation forms of DAPk in
vivo, i.e. being autophosphorylated in growing cells and dephosphorylated
upon apoptotic stimuli, was studied by analyzing the cell death-promoting
capacity of the single mutants in transfection-based assays. The S308D
substitution, mimicking phosphorylation and simulating the state
of DAPk in growing cells, gave rise to a Ôloss of function' mutant which
possesses very minor killing activity. In contrast, S308A representing
imposed dephosphorylation, and simulating the state of DAPk in response
to an apoptotic stimulus, generated a Ôgain of function' super-killer
mutant. Thus, phosphorylation on Ser308 silences the pro-apoptotic
activity, whereas dephosphorylation of Ser308is part of the
activation process of DAPk [10].
CONCLUSION We
propose a mechanism in which autophosphorylation at the basal state serves
as a locking device ensuring the activation of the kinase only upon the
appropriate apoptotic signal. The localization of Ser308 within
the catalytic cleft promotes its phosphorylation, thereby simultaneously
reinforcing the existing "locked" conformation and decreasing the possibility
that the enzyme will be activated by random fluctuations of cellular Ca2+
levels. Upon triggering of death signals by the addition of ceramide, the
restraints imposed by the negative autophosphorylation of Ser308
are removed. This Ser308 dephosphorylation may be achieved by
a specific phosphatase, the identity of which is as yet unknown. The dephosphorylated
state results in a highly active enzyme with enhanced cell killing capacity.
We propose now that this super-killing activity emerges from increased CaM-independent
activity of the kinase as a result of "weakened lock" and increased binding
activity to CaM which facilitates its full activation even at low Ca2+/CaM
levels. This enhanced sensitivity to suboptimal levels of Ca2+/CaM
could potentiate the response of DAPk to the Ca2+ spike that
takes place during cell death and could also serve to widen the window of
DAPk activation beyond the time where intracellular Ca2+levels
are maximal.REFERENCES
1. Jacobson MD, Weil M, Raff MC. 1997. Programmed cell death
in animal development. (Review) Cell 88: 347.
2. Deiss LP, Feinstein E, Berissi H, Cohen O, Kimchi A. 1995.
Identification of a novel serine/threonine kinase and a novel 15-kD protein
as potential mediators of the gamma interferon-induced cell death. Genes
Dev 9: 15.
3. Deiss LP, Kimchi A. 1991. A genetic tool used to identify
thioredoxin as a mediator of a growth inhibitory signal. Science 252:
117.
4. Cohen O, Feinstein E, Kimchi A. 1997. DAP-kinase is a
Ca2+/calmodulin-dependent, cytoskeletal-associated protein
kinase, with cell death-inducing functions that depend on its catalytic
activity. EMBO J 16: 998.
5. Raveh T, Droguett G, Horwitz MS, DePinho RA, Kimchi A.
2001. DAP-kinase activates a p19ARF/p53-mediated apoptotic checkpoint
to suppress oncogenic transformation. Nat Cell Biol 3: 1.
6. Jang CW, Chen CH, Chen CC, Chen JY, Su YH, Chen RH. 2002.
TGF-b induces apoptosis through Smad-mediated expression of DAP-kinase.
Nat Cell Biol 4: 51.
7. Cohen O, Inbal B, Kissil JL, Raveh T, Berissi H, Spivak-Kroizaman
T, Feinstein E, Kimchi A. 1999. DAP-kinase participates in TNF-a- and
Fas-induced apoptosis and its function requires the death domain. J
Cell Biol 146: 141.
8. Inbal B, Cohen O, Polak-Charcon S, Kopolovic J, Vadai
E, Eisenbach L, Kimchi A. 1997. DAP-kinase links the control of apoptosis
to metastasis. Nature 390: 180.
9. Raveh T, Berissi H, Eisenstein M, Spivak T, Kimchi A.
2000. A functional genetic screen identifies regions at the C-terminal
tail and death-domain of death-associated protein kinase that are critical
for its proapoptotic activity. Proc Natl Acad Sci USA 97: 1572.
10. Shohat G, Spivak-Kroizman T, Cohen O, Bialik S, Shani
G, Berrisi H, Eisenstein M, Kimchi A. 2001. The pro-apoptotic function
of death-associated protein kinase is controlled by a unique inhibitory
autophosphorylation-based mechanism. J Biol Chem 276: 47460.
|
Version imprimable
