Diacylglycerol kinase theta and zeta isoforms: regulation of activity, protein binding partners and physiological functions

(1)

Diacylglycerol kinase theta and zeta isoforms: regulation of activity,

protein binding partners and physiological functions

Los, Alrik Pieter

Citation

Los, A. P. (2007, January 25). Diacylglycerol kinase theta and zeta isoforms: regulation of

activity, protein binding partners and physiological functions. Retrieved from

https://hdl.handle.net/1887/9451

Version: Corrected Publisher’s Version

License: Licence agreement concerning inclusion of doctoral thesis in the

Institutional Repository of the University of Leiden

Downloaded from: https://hdl.handle.net/1887/9451

Note: To cite this publication please use the final published version (if applicable).

(2)

S t r u c t u r e – a c t iv it y

r e la t io n s h ip o f

d ia c y lg ly c e r o l k in a s e - e

(BBA. 2004 Mar 22;1636(2-3):169-174)

Alrik P. Los

J ü rg e n v a n B a a l

J oh n d e W id t

N u llin D iv e c h a

W im J . v a n B lit t e rsw ijk

Chapter 2

(3)

46

(4)

Ab stract

D iacylglycerol kinase (

D G K

) phosphorylates the second m es-

senger diacylglycerol to phosphatidic acid. Am ong the nine

m am m alian isotypes identified,

D G K

e is the only one w ith

three cysteine-rich dom ains ( instead of tw o) in its

N

-term inal

regulatory region. W e previously reported that

D G K

e b inds to

and is negatively regulated b y active

R

ho

A

. W e now report that

R

ho

A

strongly b inds to the

C

-term inal catalytic dom ain, w hich

w ould ex plain its inhib ition of

D G K

activity. T o help finding

a physiological function of

D G K

e , w e further determ ined its

activity in vitro as a function of 1 5 different truncations and

point m utations in the prim ary structure. M ost of these alter-

ations, located throughout the protein, inactivated the enz ym e,

suggesting that catalytic activity depends on all of its con-

served dom ains. T he m ost

C

-term inal cysteine-rich dom ain

is elongated w ith a stretch of 1 5 am ino acids that is highly

conserved am ong

D G K

isotypes. M utation analysis revealed

a num b er of residues in this region that w ere essential for

enz ym e activity. W e suggest that this cysteine-rich dom ain

ex tension plays an essential role in the correct folding of the

protein and /or in sub strate presentation to the catalytic

region of the protein.

Structure–activity relationship of diacylglycerol kinase-e

(5)

48

1. I ntroduction

Diacylglycerol kinase (DGK) p hosp horylates the second messenger diacylglycerol (DAG) to p hosp hatidic acid. Nine mammalian isotyp es have b een identif ied. T hey are group ed among f ive distinct classes, b ased on the comp osition of conserved domains in their p rimary structure. T he p hysiological f unctions of these isoz ymes and their conserved domains are rather p oorly def ined (reviewed in (T op ham and Prescott, 1 9 9 9 ; van Blitterswijk and H oussa, 1 9 9 9 ; van Blitterswijk and H oussa, 2 0 0 0 ; K anoh et al., 2 0 0 2 ; T op ham and Prescott, 2 0 0 2 )). T he DGKe isotyp e (H oussa et al., 1 9 9 7 ) is the only memb er of the f if th class. I ts p rimary structure contains, f rom N- to C-terminus resp ectively, a p roline-rich region, three cysteine-rich domains (instead of two f or all other isotyp es), a so-called ‘Ras-associating’ (RA) domain p artly overlap p ing a p leckstrin homology (PH ) domain, and a catalytic domain. Ap art f rom this latter domain, no regulatory f unction of the other domains has b een def initely estab lished. We rep orted p reviously that DGK e is p redomi- nantly ex p res sed in b rain, and that DGKe uniq uely b inds to and is negatively regulated b y the small G p rotein RhoA (H oussa et al., 1 9 9 9 ). H owever, it remains unclear how RhoA inhib its DGK e and very little is known ab out the f unction of this lip id kinase in vivo.

I n the p resent study, we determined what structural f eatures of DGKe are essential f or activity. An ex tensive structure-activity analysis was carried out in vit r o, in which we rationally designed p oint mutations in conserved regions and truncations f rom the N- and the C-terminus. F rom this ap p roach we learned three new things. F irst, the enz yme is ex tremely sensitive to (inactivated b y) most of these mutations, indicating that an intact catalytic domain b asically needs the p resence of all other (conserved) regions of the p rotein f or enz ymatic activity. S econd, we identif ied an ex tension of 1 5 residues in the cysteine-rich domains that is sp ecif ic f or DGK and is essential f or its activity. T hird, we f ound that RhoA b inds to the catalytic domain, which would ex p lain its inhib itory action on this lip id kinase.

2 . Materials and methods

2 . 1 . G e n e r a t i o n o f D G K

e

p o i n t - m u t a n t s

Point mutations in the CRD3 ex tended region were generated b y using DGKe mutated p rimers (T ab le 1 ), the Q uikChange^{T M} S ite-Directed M utagenesis K it f rom S tratagene and human DGKe-V S V cDNA (H oussa et al., 1 9 9 9 ) as a temp late (V S V , encodes a p ep tide-tag f rom vesicular stomatitis virus). M utants were sub cloned into the E coR1 -site of p cDNA3 vector (I nvitrogen) and conf irmed b y DNA seq uencing.

A kinase-death (kd; G6 4 8A) construct of DGKe was made b y a p olymerase chain reaction (PCR)-b ased ap p roach. Brief ly, a PCR p roduct was generated with the f orward p rimer 5 ’-CCGAAT T CAT GGCGGCGGCGGCCGAGCCC-3 ’ and

(6)

the reverse primer: 5’-CCAGTCGGTACCGTGTCTCCTCCAGGGCGCCAAGCAC- CCAGCCCACAGTGTCATCGCC-3’ using wtDGKe-VSV as a template, digested with Ec oRI and K p nI and then substituted with the EcoRI-K p nI fragment of wtDGKe in pcDNA3-FLAG vector (see below).

2 . 2 . Tr u n c a t i o n m u t a n t s

DGKe truncation cDNAs were made by PCR, using appropriate primers and human DGKe-VSV as a template. The resulting cDNAs were subcloned in frame into the Ec oRI and X h oI sites of pcDNA3-FLAG or EcoRI site of pEGFP-N3 vector (Clontech).

DGKe-61-575 and DGKe-6575-941 cDNAs were VSV-tagged at the 3’-end and ligated into EcoRI-digested pcDNA3 vector. The vector pcDNA3-FLAG (in which the DNA encoding the FLAG peptide was placed in the X h oI-X b a I site) was kindly provided by Dr. Peter ten Dijke (at our Department). All constructs were verified by DNA sequencing.

2 . 3 . E x p r e s s i o n o f D G K c o n s t r u c t s i n c e l l s a n d i n v i t r o e n z y m a t i c a c t i v i t y a s s a y

DGK constructs were transfected into CO S-7 cells using the DEAE-dextran method, and after two days the expression of the (tagged) constructs was monitored by Western blotting using anti-FLAG monoclonal antibody M2 (Sigma), anti-VSV monoclonal antibody P5D4 (Roche) or anti-GFP polyclonal antibody (BD Biosciences).

DGK activity assays were performed essentially as described previously (Houssa et al., 1997). Briefly, assays were performed in the presence of 1 mM deoxycholate, 1 mM phosphatidylserine (Sigma), [ a-³²P]ATP and 1 mM dioleoyl-s n-glycerol (Sigma) as a substrate. The radioactive phosphatidic acid (PA) generated was separated by TLC, using the solvent system chloroform/ methanol/ acetic acid, 65:15:5 (v:v:v). 32P-PA formation was quantified by a PhosphoImager and normal- ized by the amount of expressed protein (arbitrary units).

Mutant Forward primer Reverse primer

G236R CCCGAGTGTGGCTTCAGGCGTCTGCGCTCCC GGGAGCGCAGACGCCTGAAGCCACACTCGGG

L241V CGTCTGCGCTCCGTGGTCCTGCCTC GAGGCAGGACCACGGAGCGCAGACG

P244A CCCTGGTCCTGGCTCCCGCGTGC GCACGCGGGAGCCAGGACCAGGG

P244L CTCCCTGGTCCTGCTGCCCGCGTGCGTGCG CGCACGCACGCGGGCAGCAGGACCAGGGAG

P245L CCTGGTCCTGCCTCTCGCGTGCGTGCGCC GGCGCACGCACGCGAGAGGCAGGACCAGG

S240T CGGGCGTCTGCGCACCCTGGTCCTGCC GGCAGGACCAGGGTGCGCAGACGCCCG

Primers are shown in the 5’A3’ direction. Mutated codons are underlined, and mutated nucleotides are in bold.

T ab l e 1 . Primers used in site-directed mutagenesis of DGKe-extCRD.

(7)

50

2 . 4 . R h o A b i n d i n g a s s a y

FLAG-tagged or VSV-tagged DGKe constructs or VSV-tagged DGK_ were expressed together with Myc-tagged V14-RhoA in COS-7 cells, as described (Houssa et al., 1999). The RhoA construct was immunoprecipitated with monoclonal antibody 9E10. Coprecipitated DGKe constructs were detected by Western blotting using antibodies against the FLAG-tag or the VSV-tag (P5D4).SDS-PAGE and Western blotting were performed as described previously (Houssa et al., 1997).

3. Results and Discussion

For the structure-function analysis of DGKe, we made a number of truncations and point mutations in the primary sequence of the protein. Figure 1 gives a schematic overview of the DGKe primary structure with its conserved domains, and the various truncations that were made. Point mutations were made at conserved residues in the catalytic domain and in a conserved region extending the third CRD of the protein (extCRD).

Fig. 1.

DGKe constructs used: wild-type (wt), kinase-dead (kd; G648A, position marked by asterisk) and various truncation mutants (denoted by residues deleted).

TheC-termini of all constructs were fused to GF P or were F L AG-tagged.

The position of conserved domains in DGKe are indicated. P ro, proline -rich region;CRD, cysteine -rich domain (numbered 1 to 3); ext, extended CRD; P H , pleckstrin homology domain.

(8)

Fig. 2.

Sequence alignment of 15 amino acids

‘extension’ of the most C-terminal cysteine-rich domains (extCRD) of DGK isotypes. Data base accession numbers are indicated.

The alignment starts from the last (C-terminal) conserved Cys (in italics and bold-face; residue number indicated) of the canonical CRD (CRD3 in

DGKe). Fully conserved residues are in bold-face and partly conserved (similar) residues are shaded. DGKe residues mutated in this study are indicated with asterisks. Abbreviations: at, Arabidop - s is th al iana; ce, C ae norh abditis e l e g ans ; dd, D ic ty os te l iu m dis c oide u m ; dm, D ros oph il a m e l anog as te r; hs, H om o s apie ns .

3 . 1 . C o n s e r v e d r e s i d u e s i n t h e e x t e n d e d C R D a r e e s s e n t i a l f o r D G K a c t i v i t y

We previously discovered that the second CRD in DGK isotypes (the third CRD in DGKe) contains a C-terminal 15 amino acids conserved ‘extension’ that is specific and characteristic for the entire family of DGK isozymes (Houssa and van Blitters- wijk, 1998). Figure 2 shows the sequence alignment of this extension, denoted as extCRD, among DGK isotypes in five different species. We now investigated for human DGKe^CRD3 if the conserved residues in this extension are important for enzymatic activity. We mutated three conserved residues, Gly-236, Pro-244 and Pro-245 (in bold in Fig. 2) and two non-conserved residues, Ser-240 and Leu-241, within the extCRD of DGKe (marked by asterisks). The generated mutants were cloned into pcDNA3 vector, expressed in COS-7 cells and immunoprecipitated from cell lysates. Half of the precipitates was used for Western blotting (expression controls) and the other half for DGK activity measurement. Figure 3 shows that all three substitutions of the conserved residues lead to inhibition of DGK activity, while the mutations of the non-conserved residues have no effect on enzyme acti vity. Not all mutations of the conserved residues showed the same extent of inhibition: Gly-236 and Pro-245 mutations resulted in a loss of 93% of DGK activity, while substitution of Pro-244 either with an alanine or a leucine led to only 60%

and 70% reduction of activity, respectively. This may not be surprising since

(9)

52

mammalian DGKa and ArabidopsisDGK1 contain a leucine and a cysteine, respectively, at this position.

In conclusion, our data demonstrate that the conserved residues in the ext- CRD are important for DGK activity, possibly because they are essential for the correct folding of the protein or because they participate directly in the catalytic reaction.

3 . 2 . Tr u n c a t i o n s a t t h e N- o r t h e C- t e r m i n u s i n a c t i v a t e D G K

e

We made a number of truncations both at the C-terminal and the N-terminal side of the DGKe primary sequence, as shown in Figure 1. Each ^DGKe construct was made as a GFP fusion protein (GFP linked to the C-terminus) as well as FLAG- tagged (also at the C-terminus). These constructs were transfected into COS cells, and protein expression was quantified by Western blotting and densitometry of the bands. DGK activities were determined in the COS cell lysates and were corrected for the (small) variations in expression.

Figure 4 shows that truncations at the catalytic (C-terminal) side of the mole- cule entirely inactivated the enzyme. Even the smallest truncation of 33 amino acids (DGKe-6908-941) notably outside the catalytic domain, surprisingly, rendered the enzyme fully inactive. We initially found this with the GFP-fusion protein and sus- pected that the bulky GFP group might sterically hinder the catalytic domain if the connecting peptide is too short (Fig. 1 and 4). However, the same 33 residues- truncation mutant extended by a small FLAG-tag was likewise inactive, while the

Fig. 3.

Conserved residues in the C-terminal extension of CRD3 (extCRD3; marked by asterisks in Fig. 2) are necessary for DGK activity.

CO S-7 cells were transfected with V SV- taggedDGKe constructs containing the indicated mutations. Cells were lysed andDGKe was immunoprecipitated using monoclonal antibody P5D4 directed against the V SV-tag. Half of the immunoprecipitate was used for West- ern blotting (upper panel). The other half was used for DGK activity assay. PA, the product of DGK was separated by TLC (middle panel). Radioactivity (³²P) in the PA spots was q uantified by Phos- phoImager (lower panel). Data are expressed as percentage of activity of the wild-type DGKe and are represent- ative of two experiments with the same results. Similar percentages were obtained when the CO S-7 cell lysates were assayed directly, without immu- noprecipitation (data not shown).

(10)

GFP-fused full-length DGKe was fully active. We therefore conclude that even the very C-terminus of DGKe is essential for its activity.

The N-terminal portion of DGKe contains a number of regulatory domains such as the proline-rich region, the three cysteine-rich domains and the PH domain.

Truncations at this side of the protein primary structure, like at the C-terminus, dramatically reduced enzymatic activity (Fig. 4). Only one of these constructs, DGKe-61-116, lacking the N-terminal proline-rich region and the first of the three CRDs, showed some statistically significant residual activity of 10-20% of the wild-type enzyme. This was true for both the GFP- and the FLAG-tagged construct, corroborating the genuineness of this observation. This result is unexpected in view of the finding that DGKe containing all three CRDs but without the proline- rich domain it totally inactive. A possible explanation is that the N-terminal proline-rich region (perhaps including the first CRD), although essential for DGKe activity, exerts at the same time some moderate auto-inhibitory activity on the enzyme. DGKe is not unique in this regard: In class I DGK isotypes (_, `, a), the N-terminal RVH (recoverin homology) (C1) domain together with the Ca²⁺-binding EF-hands have been shown to exert an auto-inhibitory action (Sakane et al., 1991;

Kanoh et al., 2002).

Fig. 4.

Enzymatic activity of wild-type (wt), kinase-dead (kd, G648A) and various truncation mutants of DGKe.

Truncations from the N-terminus (6N- term) and C-terminus (6C-term) are numbered as summarized in Figure 1.

GFP-fused (black bars) or FLAG-tagged (open bars) DGKe constructs were transfected into COS-7 cells. Cell lysates were used to measure DGKe expression (Western blotting) and enzymatic activity in vitro. The amount PA formation was divided by the relative quantity of protein expressed (band density scanned in arbitrary units). Value of wild-type DGKe was set at 100% . Data are means of 3 independent experiments ±S.D. AnANOVA statistical test revealed that the activity of the two 61-116 truncations were significantly (P< 0.05) higher than the other constructs (except wild- type). Expressions of FLAG-tagged 61-575 and 61-388 were too low to allow reliable plotting in the figure. Ctrl, empty vector control.

(11)

54

3 . 3 . C a t a l y t i c d o m a i n

It has been debated whether or not the isolated catalytic domain of DGK by itself would show enzymatic activity. For DGK_, the C-terminal catalytic domain lacking the whole N-terminal region including the CRDs was reported to possess enzymatic activity (Sakane et al., 1996), although our lab could not confirm this. As regards DGKe, Figure 4 shows that the catalytic domain without the entire N-terminal regulatory region is completely inactive. Likewise, it has been reported that intact CRDs are required for activity of DGKc (Santos et al., 2002).

The catalytic domain contains an ATP binding site that differs from that in protein kinases in the sense that the conserved Gly-X-Gly-X-X-Gly motif (van Blit- terswijk and Houssa, 2000) does not need a downstream Lys residue for enzymatic activity (Schaap et al., 1994). A point mutation (G648A) made at the second Gly residue of this conserved motif rendered DGKe fully inactive (Fig. 4). Overexpres- sion of this kinase-dead mutant may have a dominant-negative effect in vivo, as was reported for the DGKc and ^DGK_ isotypes (van Blitterswijk and Houssa, 2000;

Topham et al., 1998; Sanjuan et al., 2001).

3 . 4 . B i n d i n g o f Rh oA

We previously reported that DGKe specifically binds to and is negatively regulated by active RhoA (Houssa et al., 1999). This property appeared relevant in vivo, since constitutively active V14-RhoA inhibited _-thrombin-stimulated DGKe activity in the nucleus of IIC9 embryonic fibroblasts (Bregoli et al., 2001). To find out how this small G protein blocks DGK activity, we individually expressed the regulatory and the catalytic domains of DGKe together with active ^V14-RhoA into COS-7 cells and examined their mutual binding. The Western blot in Figure 5 shows that immunoprecipitated (myc-tagged) RhoA co-precipitates both (FLAG-tagged) DGKe domains. The expression of the individual domains was consistently much lower

Fig. 5.

RhoA binding to the catalytic and the regulatory domain of DGKe.

FLAG-tagged full-length DGKe (wild- type; wt), DGKe-61-575 (catalytic domain; cat), DGKe-6575-941 (entire regulatory domain; reg) or empty vector were transfected into COS-7 cells. Myc- taggedV14-RhoA was co - expressed (lower left panel) and immunoprecipitated (IP) with monoclonal antibody 9E10. The amount of FLAG-DGKe construct protein in the IPs and the total cell lysates was determined by immu- noblotting (IB) using anti-FLAG antibodies. The amount of myc-V14-RhoA in the immunoprecipitate is shown in the lower right-hand control panel.

Mol. weight markers are indicated.

(12)

than the wild-type DGKe. Considering the different expression levels, it seems that the FLAG-tagged catalytic domain binds RhoA somewhat stronger than does the FLAG-tagged regulatory domain. Similar results were obtained with the respective VSV-tagged DGKe constructs, while VSV-DGK_ (serving as a negative control) did not co-precipitate with myc-RhoA (data not shown). The binding of RhoA to the DGKe catalytic domain may explain why RhoA inhibits the activity of this lipid kinase.

We tried to test the idea that RhoA might bind to the so-called Ras-associating (RA) domain located within the PH domain (Houssa et al., 1997). U nfortunately, a protein corresponding to the isolated PH domain construct failed to express in cells, so that we are unable to pin-point RhoA binding to this particular region of the regulatory domain.

3 . 5 . C o n c l u d i n g r e m a r k s

This study reveals that the DGKe isozyme is extremely sensitive to mutations.

Even the smallest truncations in the DGKe primary structure dramatically affect (abolishes) enzymatic activity in vitro. This surprising result is not an assay- related artifact, since similar results were obtained in a different, detergent-free assay system (data not shown), and when using two different tags (GFP or FLAG).

It therefore seems that catalytic activity is extremely sensitive to correct folding of the protein, and requires the presence of all of its conserved domains. Other DGK isozymes are not so sensitive to small truncations. For example, DGKc maintains activity when its C-terminal ankyrin repeats and PDZ-binding motif are deleted (Santos et al., 2002). A DGK` splice variant with a deletion of the C-terminal 35 amino acids also maintains activity (Caricasole et al., 2002).

DGKe is the only isozyme that binds to and is negatively regulated by active RhoA. This would suggest, and we actually confirmed here, that RhoA binds to the regulatory domain of DGKe. However, we found even stronger binding to the catalytic domain, which may provide a direct explanation for the inhibitory effect on the catalytic activity. It may be speculated that in a three-dimensional setting, active RhoA acts as a wedge between the regulatory and catalytic domain, thus preventing DAG access and/or its phosphorylation (see also below for a specu- lation how DAG might be presented to the catalytic domain).

More insight has been gained, but some uncertainty still remains about the function of the CRDs in DGKs. They were found to be essential for DGKe (this study) and DGKc activity (Santos et al., 2002), but not for DGK_ activity (Sakane et al., 1996). Intact CRDs were also required for plasma membrane translocation, as described for DGKc (Santos et al., 2002). Based on the homology with CRDs in protein kinases C that bind phorbol esters and DAG (Z hang et al., 1995; van Blitterswijk, 1998; Shindo et al., 2001), the CRDs in DGKs have been proposed to interact simi- larly with DAG (Houssa and van Blitterswijk, 1998). Recently, Shindo et al. have shown that the N-terminal CRD (CRD1) of DGKa and -` isotypes binds phorbol ester and that DAG competes with this binding, indirectly suggesting that DAG also binds to this domain (Shindo et al., 2001; Shindo et al., 2003). However, CRDs of the other DGKs including DGKe do not bind phorbol ester. We previously noted

(13)

56

that the C-terminal CRD2 in all DGK isotypes (CRD3 in DGKe) bears less homology to CRDs in protein kinase C but is highly specific for DGK and, in addition, contains a conserved extension (extCRD) (Houssa et al., 1997; Houssa and van Blitterswijk, 1998). It is, therefore, conceivable that this domain, unlike CRDs in protein kinase C, does not serve as a particularly strong DAG binding site. Here we showed that conserved residues in the ‘extension’ of this extCRD are essential for catalytic activity of at least DGKe and DGKc (A.P. Los, unpublished data). It may be speculated that the extCRD, by transient (relatively weak) binding of DAG, participates in catalysis

in conjunction with the catalytic domain. In other words, the extCRD may somehow

‘present’ DAG to the catalytic domain. The other CRD(s) could likely be sensing DAGgeneration at the plasma membrane or at other subcellular sites, thus facili- tating DGK translocation to that location. For DGKe, membrane translocation and sub sequent activation was found in noradrenalin-stimulated small arteries (Walker et al., 2001), while nuclear translocation and activation occurred in _-thrombin- stimulated embryonic fibroblasts (Bregoli et al., 2001). The structure-activity relationship presented here may eventually contribute to our further understand- ing of the function of this isozyme.

(14)

Bregoli, L., Baldassare, J.J., and Raben, D.M. (2001).

Nuclear diacylglycerol kinase-theta is activated in response to alpha-thrombin. J. Biol. Chem. 276, 23288-23295.

Caricasole, A., Bettini, E., Sala, C., Roncarati, R., Kobayashi, N., Caldara, F., Goto, K., and Terstappen, G.C. (2002). Molecular cloning and characteriz ation of the human diacylglycerol kinase beta (D G K beta) gene: alternative splicing generates D G Kbeta isotypes with dif f erent proper ties. J. Biol. Chem. 277, 4790-4796.

Houssa, B., Schaap, D., van der, W.J., Goto, K., Kondo, H., Y amakawa, A., Shibata, M., Takenawa, T., and van Blitterswijk, W.J. (1997). Cloning of a novel human diacylglycerol kinase (D G Ktheta) containing three cysteine-rich domains, a proline-rich region, and a pleckstrin homology domain with an overlapping Ras-associating domain. J Biol. Chem. 272, 10422-10428.

Houssa, B. and van Blitterswijk, W.J. (1998). S pecif i- city of cysteine-rich domains in diacylglycerol kinases and protein kinases C. Biochem. J 3 3 1 , 677-679.

Houssa, B., de Widt, J., Kranenburg, O., Moolenaar, W.

H., and van Blitterswijk, W.J. (1999). D iacylglycerol kinase theta binds to and is negatively regulated by active RhoA.J. Biol. Chem. 274 , 6820-6822.

Kanoh, H., Y amada, K., and Sakane, F. (2002). D iacylglycerol kinases: emerging downstream regulators in cell signaling systems. J Biochem. (Tokyo) 1 3 1 , 629-633.

Sakane, F., Imai, S., Y amada, K., and Kanoh, H. (1991).

The regulatory role of E F-hand motif s of pig 8 0 K diacylglycerol kinase as assessed using truncation and deletion mutants. Biochem. Biophys. Res.

Commun. 1 8 1 , 1015-1021.

Sakane, F., Kai, M., Wada, I., Imai, S., and Kanoh, H.

(1996). The C-terminal par t of diacylglycerol kinase alpha lacking z inc f ingers serves as a catalytic domain. Biochem. J. 3 1 8 , 583-590.

Sanjuan, M.A., Jones, D.R., Izquierdo, M., and Merida, I. (2001). Role of diacylglycerol kinase alpha in the at tenuation of receptor signaling. J. Cell Biol.

1 5 3 , 207-220.

Santos, T., Carrasco, S., Jones, D.R., Merida, I., and Eguinoa, A. (2002). D ynamics of diacylglycerol kinase z eta translocation in living T-cells.

S tudy of the structural domain req uirements f or trans location and activity. J. Biol. Chem. 277, 30300-30309.

Schaap, D., van der, W.J., and van Blitterswijk, W.J.

(1994). Consensus seq uences f or ATP-binding sites in protein kinases do not apply to diacylglycerol kinases. Biochem. J. 3 0 4 , 661-662.

Shindo, M., Irie, K., Masuda, A., Ohigashi, H., Shirai, Y ., Miyasaka, K., and Saito, N. (2003). S ynthesis and phorbol ester binding of the cysteine-rich domains of diacylglycerol kinase (D G K) isoz ymes.

D G K gamma and D G Kbeta are new targets of tumor-promoting phorbol esters. J. Biol. Chem. 278 , 18448-18454.

Shindo, M., Irie, K., Ohigashi, H., Kuriyama, M., and Saito, N. (2001). D iacylglycerol kinase gamma is one of the specif ic receptors of tumor-promoting phorbol esters. Biochem. Biophys. Res. Commun.

28 9 , 451-456.

Topham, M.K., Bunting, M., Zimmerman, G.A., McIntyre, T.M., Blackshear, P.J., and Prescott, S.M.

(1998). P rotein kinase C regulates the nuclear localiz ation of diacylglycerol kinase-z eta.

Nature 3 9 4 , 697-700.

Topham, M.K. and Prescott, S.M. (1999). Mammalian diacylglycerol kinases, a f amily of lipid kinases with signaling f unctions. J Biol. Chem. 274 , 11447-11450.

Topham, M.K. and Prescott, S.M. (2002). D iacyl - glycerol kinases: regulation and signaling roles.

Thromb. Haemost. 8 8 , 912-918.

van Blitterswijk, W.J. (1998). H ypothesis: ceramide conditionally activates atypical protein kinases C, Ra f -1 and K S R through binding to their cysteine- rich domains. Biochem. J. 3 3 1 , 679-680.

van Blitterswijk, W.J. and Houssa, B. (1999). D iacylglycerol kinases in signal transduction. Chem.

Phys. Lipids 9 8 , 95-108.

van Blitterswijk, W.J. and Houssa, B. (2000). P roper ties and f unctions of diacylglycerol kinases. Cellular Signalling 1 2, 595-605.

Walker, A.J., Draeger, A., Houssa, B., van Blitterswijk, W.J., Ohanian, V., and Ohanian, J. (2001). D iacylglycerol kinase theta is translocated and phos- phoinositide 3-kinase-dependently activated by noradrenaline but not angiotensin I I in intact small ar teries. Biochem. J 3 5 3 , 129-137.

Zhang, G., Kazanietz, M.G., Blumberg, P.M., and Hurley, J.H. (1995). Crystal structure of the cys2 activator-binding domain of protein kinase C delta in complex with phorbol ester. Cell 8 1 , 917-924.

References

(15)