- Email: [email protected]
Phosphorylation and guanine nucleotide exchange on polypeptide chain initiation factor-2 from Artemia embryos
Biochimica et Biophysica Acta. 1007 (1989) 55-60 Elsevier
55
BBA 91913
Phosphorylation and guanine nucleofide exchange on polypepfide chain in/fiat/on factor-2 from Artemia embryos Mauricio G. Mateu, Federico G. Maroto, Oscar Vicente and Jos6 M. Sierra Centro de Biologla Molecular, Consejo Superior de Investigaciones Cientificas and Umversidad Aut6noma de Madric~ Madrid (Spare)
(Received 29 April 1988) (Revised manuscript received ? September 1988)
Key words: Protein synthesis; Initiation factor 2: Guanine nucleotide exchange: Protein phosphory|ation; (Artemia)
Et~taryotic initiation factor-2 (e|F-2) |tom Artemia embryos is able to exchange guanine nuclcotides at the s~t~e r~te in the presence or absence of Mg 2+ when the reaction is carried out with either purified e|F-2 at 30°C or less ~ r i [ i ~ preparations at any temperature (|0-~30°C). No exchange factor appears to catalyze this reaction. However, with purified e|F-2 at lower temperatures (|0 o C) the exchange is clearly impa|red by Mg 2. and this intpai|'ment is overcome by the guanine nuclcotide exchange factor (GEF) of rabbit reticu|ocytes. Thus, A~emia e|F-2 is able to exchange guanine nuc|eotides by two alternative mechanisms ~hat may reflect two states of the protein. Pbospbov~|afion of the e|F-2a submit by the heine-controlled inhibitor (HC|) of rabbit retieulucytes abolishes the GEF-dependent reaction, but has no effect on the factor-independent one. The search for e|F-2a kinases in A rtemia e.nbryo |ed to the detection of only one such enzyme, which was identified as a casein kinase type ||. None of the exchange reactions is affected by the pbusphory|ation of the elF-2a sublmit by this kinase, suggesting that, irrespective of the kind of mechanism for guanine nucleotide exchange that is actuary operating in Artemig, it might not be a target for regulation by e|F-2a phospboryiation.
Introduction
GEF is involved in elF-2 recycling during polypeptide chain initiation in mammalian cells (for reviews, see Refs. 1-3). Upon completion of a round of initiation an elF-2. GDP complex is formed and because at physiological concentrations of Mg 2+ the affinity of elF-2 for GDP is much higher than for GTP, the exchange of bound GDP for free GTP, which is required to form the ternary complex elF-2. GTP. MettRNA~, is impaired. This impairment, however, is overcome by the catalytic action of GEF, which promotes the G D P - G T P exchange. The GEF-dependent reaction is a target for regulation of polypeptide chain initiation, at least in rabbit reticulocytes. Phosphorylation of elF-
Abbreviations: GEF, guanine nucleotide exchange factor; elF-2, eukaryotic protein synthesis initiation factor-2; HCI, heme-controlled translational inhibitor (an elF-2a kinase); Met-tRNA i, initiator methionyl-tRNA, SDS, sodium dodecyi sulfate; PAGE, polyacrylam|de gel electrophoresis. Correspondence: J.M. Sierra, Centro de Biologla Molecular. Universidad Autbnoma de Madrid, Cantoblanco, 28049 Madrid° Spain.
2a subunit by HCI blocks the exchange reaction be, cause all the available GEF gets tied up as an inactive elF-2(aP). GEF complex [4-7]. The above mechanism of translational regulation is being investigated, among other lower eukaryotes, in Artemia. In developing embryos of Artemia it was reported the presence of a protein that stimulated the activity of elF-2 in ternary complex formation [8,9], This effect was blocked when elF-2 was preincubated with rabbit HCI in the presence of ATP [8]. However, Wahba and collaborators have been unable to find a requirement for a G D P / G T P exchange factor with Artemia elF-2 [10-12]. They have shown that guanine nucleotide exchange on this factor can proceed readily in the presence of Mg 2+ in a reaction which is not stimulated by GEF from rabbit reticulocytes [11,12]. On the other hand, it is also known that the a-subunit of Artemia elF-2 is a substrate for rabbit reticulocyte HCI [10,12,13] and casein kinase type II isolated from Artemia or mammalian cells [14,15]. However, neither of these kinases abolished the above.mentioned GEFindependent exchange [12,15]. We have reinvestigated the guanine nucleotide exchange on Artemia elF-2 and have found that this
0167-4781/89/$03.50 © 1989 Elsevier Science Pubfishers B.V. (Biomedical Division)
56 factor, as previously shown for Drosophila embryo elF-2 [16], is able ~o exchange guanine nucleotides by two alternative mechanisms that may reflect two states of the protein. We confirm the occurrence of a GEF-independent exchange but we also find conditions where nucleotide exchange on Artemia elF-2 is impaired by Mg 2+ and this impairment is overcome by rabbit GEF. The ability of Artemia elF-2 to respond to rabbit GEF has allowed us to study also the effect of eIF-2a phosphorylation, by either rabbit HCI or Artemia casein kinase type II, on this reaction. While, as expected, the GEF-dependent exchange is abolished by rabbit HCI, it is not abolished in the presence of Artemia casein kinase type 11. This establishes an important difference between the two kinases with respect to the regulation of guanine nucleotide exchange by elF-2 phosphorylation. Materials and Methods Assays To assay elF-2a kinase activity, samples (36 ~tl)
containing 20 mM Hepes (pH 7.6), 60-200 mM KCI, 4 mM Mg(CH~COO)2, 1 mM dithiothreitol, ¢0/tM [y~2P]ATP (22200 cpm/pmol), 2.8 ttg of purified Artemia elF-2 and protein fractions to be ~¢,~sted,are incubated for 7 min at 30 o C. Labelled polypeptides were analyzed by SDS-PAGE [17] and autoradiography using Kodak X.Omat S film and !lford intensifying screen. The area corresponding to the phosph~4ated a-subunit of Artemia ¢iF-2 (,*2 kDa) is scot,ned at 626 n~~ Jn a Chromoscan 3 (Joyce-l.oebl) densi~ometer. Casein kinase activity was assayed essenti:d~ as described [18]. Guanine nucleo~ide exchange assays ~.verecan:ed out as in Ref. 16 but 1.8 ItM of [3HIGDP (8900 cpm/pmol, Amercham) and Anemia elF-2, in the amounts indicated in each experiment, were used. The time and temperature of incubation are indicated in the legends to the figures. Conditions for ternary and 40 S initiation complex formation have been reported [19]. Preporations
Batches of 125 g (dry weight) of encysted Artemia sp. embryos (San Francisco Bay Brand, Newark, CA) wer~ deve!~ped in ~.~fici~l ~ea water [20] at 27 ° C for 14-15 h and the postribosomal supernatant and polysomal fractions prepared as described [21]. Polysomes were rinsed with extraction buffer [21l, gently suspended in 25 ml of high.salt buffer (buffer C in ReL 21) and stirred for 4 h at 0 e C. The suspension was centrifuged for 165 rain at 219000 x g and the clear supernatant was collected and frozen. This fraction will be referred to as the ribosomal wash. A protein kinase capable of phosphorylating tee
a-subunit (42 kDa in SDS-PAGE) of Artemia elF-2 was detected in developing embryos of Artemia [10,13]. This activity is present in both the high-salt wash of polysomes and the postribosomal supernatant and has been further purified from the latter fraction by precipitation with ammonium sulfate. The 30-50~ saturation cut, in buffer A (20 mM Tris-HC! (pH 7.6)/1 mM dithiothreitol/0.1 mM EDTA/10~ glycerol) containing 130 mM KCI, was applied to a DEAE-cellulose column and the bulk of elF-2a kinase activity eluted with buffer A containing 350 mM KCI. This fraction was dialyzed against buffer B (50 mM Tris-HCl (pH 7.5)/1 mM dithiothreitol/0.1 mM EDTA/10~ glycerol)containing 50 mM KCI, applied to phosphocellulose and the retained pt~otein was eluted stepwise with buffer B conraining 200 raM, 350 mM or 1200 mM KCI. The elF-2a kinase, which eluted at 1200 mM KCI, was rechromatographed through phosphocellulose and purified further by glycerol density gradient centrifugation. The functional and structural properties of Artemia elF-2a kinase were found to be very similar to those reported [14] for a casein kinase type 11 isolated from a ribosomal high-salt wash of encysted Artemia embryos. That both activities were indeed associated with the same protein was also supported because they copurifled together. Artemia elF-2 was purified from the high-salt wash of developing embryos according essentially to the proce6ure of Macrae et ~1. [22], but the hydroxyapatite chromatography was omitted and the phosphocellulose step was substituted by CM.Sephadex chromatography as follows, elF-2 from DEAE-cellulose was applied to a CM-Sephadex column (10 mg protein/ml gel) equilibrated in buffer A containing 180 mM KCI. After washing the column with this buffer, elF-2 was eluted with buffer A containing 400 mM KCI and 0.5 mM phenylmethylsulphonyl fluoride. Occasionally, this fraction was further purified by sedimentation through glycerol density gradients. To obtain a partially purified fraction of Artemia elF-2, the high-salt wash (45 ml, 28 mg of protein) obtained from developing embryos was dialyzed against buffer B containing 80 mM KCI and applied to a CM-Sephadex column (1.6 x 12 cm) equilibrated in the same buffer. Retained protein was eluted stepwise with buffer B containing 180 mM and 360 mM KCI. elF-2 (20 mg of protein), which was eluted at 360 mM KCI, was dialyzed against buffer B containing 80 mM KCI and further fractionated through heparin-Sepharose (Pharmacia, 0.6 x 2 cm) with 6 ml of a 80-500 mM KCi gradient. Fractions with higher elF-2 activity were pooled and kept frozen until use. HC! and GEF, both from rabbit reticulocytes, Artemia 40 S ribosomal subunits, [y-32p]ATP, [3H]MettRNA i, and determination of protein concentration were as in previous work [16,19].
57 1
A
23
~
S
0.6
-2 O
0
E O. "O C
E
0.4
\ \
C1.
,,Hcx,G F
Q,
+Mg 2÷,
"O C
+GDP
O
O .a
{3. 1 C~
¢:L
e-,--.,t
0.2
\M0
; ATP, K,, GEF "
:x:::
,6-.-
2
A.
6
Time (rain}
8
12
3{56
7891Gll
-÷
÷ ÷ ÷ ÷
÷" ÷ - ÷
÷ .
.
.
.
÷
- ÷ ÷ - ---÷÷
12 ÷ ATP HCI
_
~"
KA
FiB. I. GEF-dependent nucleotide exchange on Artemia elF-2 and the effect on this reaction of the elF-2a phosphorylation by Iabbit HC| or drlemia casein kinas¢ type !! (KA). elF-2.[3H]GDP complexes were preformed with 1.4 pg CA) or 4 pg (B) of purified Artemia c1|:-2 (CM-Sephadex fraction) and incubated in the presence of I mM Mg(CH~COO) 2 (@, [3, v, A, in (A): ! - ! 2 in (B)) or water (o in CA)), 40/~M ATP ([3° % in (A); 2-7, 9, 11, 12 in (B)), and 0.2/ag of HC! (13, in CA); $o 8, 9 in (B)) or 22/ag of Artemza elF-2a kinas¢ (KA) (DEAE-cellulose fraction) (st in (A); 1, 6, 10, 11, in (B)) for 5 rain at 30°C. Then, nucleotide exchange reactions in the presence of 2 mM EDTA (sample 12B), 20/aM GDP (all samples except IB, 2B) and 1/ag of rabbit reticulocyte elF-2. GEF (ra. st, A in (A); 7-11 in (B)) were carried out at l0 o C (A). 30 o C ( 1, 2, 4-12 in B) or 37 ° C (31}) for 4 min (B) or as indicated in (A). Background value without elF-2 (0.08 pmol) has been substracted. (C) Phosphorylation of Artemia elF-2a under reaction conditions of the experiment in panel B. [y-32p]ATP (15/aM, 72000 cpm/pmol) instead of unlabellcd ATP was used. Samples in tracks 1, 3 and 4 correspond to samples 7, 11 and 9 of panel B, respectively. Control samples with Artemia kinase (track 2) or HC! (track 5) were as samples 6B and 5B, respectively, bw without elF-2. Samples were subjected to SDS-PAGE and the gel was dried and autoradiographed. Arrow shows the position of the a-subunit of Artemia elF-2. The level of ©iF-2a phosphorylation with Artemia kinase is higher compared to that obtained with HCI just because the relative amounts of each kinase used.
Results Guanine nucleotide exchange on Artemia elF-2 can proceed by two different mechanisms
We have used the reaction: e~F-2. [3HIGDP + GDP ~ elF-2. GDP + [3HIGDP
as a model [1,4-7,16] to study guanine nucleotide exchange on Artemia elF-2. Mehta et al. [11] have pointed out that, in contrast to rabbit reticulocyte elF-2, the Artemia factor was able to readily exchange bound GDP for free GTP in the presence of Mg 2+ in a reaction which was not stimulated by added rabbit GEF. However, we have found that with relatively highly purified preparations of Artemia elF-2, guanine nucleotide exchange may or may not be impaired by Mg 2+, depending on the reaction temperature. Moreover, this impairment can be overcome by rabbit reticulocyte G E F (Fig. 1). At 10 ° C the exchange of elF-2 bound [3H]GDP with free unlabelled GDP proceeds readily in the absence of Mg 2+ (Fig. 1A, open circles) but it is clearly impaired in the presence of this cation (Fig. 1A, filled circles). The impairment by Mg 2+ can also be seen at 30°C (Fig. 1B, compare bars 4 and 12,
which correspond to samples containing or lacking free Mg 2+, respectively), but to a lesser extent because a significant percentage of exchange has already taken place at this temperature in the presence of Mg 2+ (Fig. I B, compare bars 2 and 4). However, guanine nucleotide e~change is not affected at all by the presence ¢1" Mg 2+ when the reaction is carried out at 37°C (Fig. ~.B, bar 3). These results prompted us to investigate the effect of rabbit reticulocyte GEF on the impairment by Mg 2+ of guanine nucleotide exchange on Artemiq~ elF-2. As shown in Fig. IA, the impairment by Mg 2÷ observed when the reaction is carried out at 10 °C is completely abofished by the addition of rabbit reticulccyte GEF (Fig. 1A, compare filled circles and filled triangles). The same result is obtained when the exchange reaction is carried out at 30°C in the presence of GEF (Fig. 1B, compare bars 4 and 7, which correspon~ to samples lacking or containing GEF, respectively). These data show that Artemia elF-2 can respond to rabbit GEF. In other words, a GEF-mediated mechanism for guanine nucleotide exchange can function with Art'emia elF-2. We have further found, confirming a~d extending previous observations [11,12], that when less purified preparations of Artemia elF-2 are used, guanine
58
-GDP 1.2
A
_-)
~0.8 ,ID
÷ t,
flOP
.
1
2
3
z,
Time (rain)
10
20
30
Temperature (eC)
u
1 345 678910111213 -÷,÷, - ÷ - ÷ - ÷ - ÷ - - ÷ . . . . ÷ ÷ . . . . .
11, - ATP HCI
"''÷ . . . .
" KAil) - KA(2)
. . . . . ++'" + . . . . . . + ÷
Fig. 2. ~ ~ a s nucleotideexcbat~e on Anemia elF.2 whichis not impaired by Mg=* is unaffected by the phosphorylation of elF-2a. (A) and (B) elF-2.[SH]GDP compl~es werepreformed wi~h9 tt$ of partially purified Artemia elF.2 and incubated as for nu¢leotide exchange reaction in the ~ of I tam M~CHsCOO)a (o), water (o) and 20 MMGDP (0, o). (C) elF-2.[3H]GDP complexes preformed with 12 itS of elF-2 were incubated in the presenceof ! mM Mg(CH3COO)=, 35 ~tM ATP (2-$, 7, 9, 11.13). and 0.4 tt8 of HCI (3. 8. 9) or Artemia kinase (K^) (?it 8 of the 8 1 ~ 1 ~-adieat fraction in 4, 10. 11; 8.8 p$ of the DEAE-cdlulosefraction in S. 12. 13) for 5 rain at 30 o C. Then. ,ucleotide exchange reactionsin the presenceof 2 tam EDTA(sample 14) and 20 ptM GDP (6-14) were carried out. Backgxoundvalues without elF.2 (0.~ pmol in (A). 0.03 pmol in (B) and (C)) have been substracted, nucleotide exchange can proceed at a high rate, whether Mg 2+ is added or not (Fig. 2A) and independently of the temperature of incubation (Fig. 2B). A straightforTABLB I A~ o/,~i~t~,4e e~cloa,Re /actor m p,m~ally purified Anemia
¢1F~2 B~aarycomplexeswere performed with Anoma elF-2 ( $ 1 ~ wadieat ffaoti~) in the amount ~own and further incubated as for n ~ ~c~ ~ t i o n in the abtea~ or presenceof partially purified A ~ d elF..2 and the other ~pone.ats indicated. Back,, 8gouad value8with elF-2 (0`10and 0.25 pmol without and with ?.$ tt8 of partially purified elF-2 respectively)haw been substracted, n.d., not detmnined. IExp, Temp. Nudeotide exchange No, (eC) ~ t i o n
10
[ ~HIGDP bound (pn~)
elF.2, I3H) GDP
partially GDP +Ms=* p~fied OS ~M) (t raM) elF.2
- M s 2÷
0,9
0 0 6 12
+ + +
0,39 0,39 0,39 0.43
n.d, 0`10 0`03 n,d.
0 0
+
1.37 0.32
n.~. . 9~8
3 "1.5 12
+ + +
0`51 0.30 0.28
n.d. 0,02 n,d.
3.0
ward explanation for these results is the efistence, in partially purified Anemia elF-2, of a guanine nucleotide exchange factor which would be responsible for the endogenous reaction. However, this possibility seems to be ruled out on the basis of the experiments shown in Table I. The impairment by Mg 2+ of the [ ~ H ] G D P / G D P exchange on purified elF-2 at 1 0 ° C or 3 0 ° C is not overcome by tile addition to the reaction mixture of an excess of the less purified Artemia elF-2 as a source of the putative exchange factor, Therefore, it appears that in certain experimental conditions (by using less purified preparations of elF-2 or increasing the temper° ature at which the exchange reaction is performed) Artemia elF.2 is able to exchange guanine nucleotides readily, irrespective of the absence or presence of Mg 2+ and without the requirement of an exchange factor.
Effect of phosphorylation of the elF-2a subunit by rabbit reticulocyte HC! or Artemia casein kinase H on the guanine nucleotide exchange reactions Phosphorylation of the a-subunit of Artemia eIFo2 by rabbit reticulocyte HCI abolishes the GEF-dependent guanine nucleotide exchange reaction (Fig. 1A, compare filled triangles and open squares; Fig. 1B bar 9). The level of phospho~lation of eIF-2a by this kinase, under the experimental conditions with which the exchange reaction was carded out, is shown in Fig. 1C (track 4). In the light of this result we have searched
59 for eIF-2a kinases in developing Artemia embryos and found only one such enzyme that turned out to be very similar, if not identical, to a previously reported [|4] casein kinase type II isolated from encysted embryos of Artemia. In contrast to rabbit HCL drtemia casein kinase type II has no effect at all on the GEF-dependent nucleotide exchange reaction (Fig. IA, compare open triangles and open squares: Fig. 1B, bar 11) in spite of its being able to phosphorylate the ¢-subunit of elF-2 (Fig. 1C, track 3). The above data suggest that the GEF-dependent guanine nuc|eotide exchange on Artemia elF-2 observed in vitro may be capable of regulation by elF-2a kinases of the rabbit reticulocyte HCI type. In addition, they establish an important difference between Artemia casein kinase type 11 (with elF-2a kinase activity) and rabbit reticulocyte HCi with respect to the regulation of guanine nucleotide exchange by elF-2 pbospborylation. With respect to the factor-independent nuclcotide exchange (Fig. 2, A and B), we have confirmed [12,15] that it is not affected by the phosphorylation of the a-subunit of Artemia eIF-,'~ by either rabbit HCI (Fig. 2C, bar 9) or Artemia casein kinase type II (Fig. 2C, bars 11 and 13). It should be noted that the lack of inhibition by HCI of this nucleotide exchange is a further support for the notion that it is not mediated by a GEF-like protein. Also, any regulatory role for Artemia casein kinase type 11 in either of the two guanine nucleotide exchange mechanisms observed in vitro with Artemia elF-2 seems to be ruled out. Discussion
From the studies on guanine nucleotide exchange with Drosophila elF-2 [16] we have suggested the existence of two alternative mechanisms for this reaction: a GEF.dependent exchange, similar to that described in rabbit reticulocytes and other mammalian cells, and a factor-independent exchange. This hypothesis can be now extended to Artemia elF-2 on the basis of the results shown in this paper. In addition to confirm and extend the observation [11,12] that Artemia elF-2 can exchange bound GDP for free. nucleotide at the same rate in the presence or absence of Mg 2+, ih a reaction in which no .~xchange factor appears to be involved, we have found that with purified elF-2 guanine nucleotide exchange is clearly impaired by Mg 2+ provided the reaction is carried out at low temperatures (10°C). Moreover, this impairment is overcome by rabbit GEF. The ability of Artemia elF-2 to respond to rabbit G E F proves that it has all the structural requirements for the interaction w/th a guanine nucleotide exchange factor and therefore opens the possibility that a GEF-mediated mechanism for guanine nucleotide exchange may be operative in Artemia embryos. However, this possibility should be reinforced by the isolation from this organism
of the putative GEF-Iike protein that has remained elusive until now. Our data are also compalible with the notion that the response of Artemia elF-2 to rabbit GEF is just an evolutionafily conserved property acquired before GEF and/o~ HCI appeared in mammalian cells. In this respect, it would be interesting Io ascertain whether this property, which is also observed in Drosophila elF-2 [16], can be extended to other lower eukaryotes, such as yeast or wheat, in which no evidence for the occurrence of a GEFolike factor has so far been reported [23,24]. On the other hand, the inhibition by HCI of the GEF-dependent nucleotide exchange shown in th/s paper supports the notion that if a GEF-mediated mechanism for guanine nucleotide exchange e~dsted in Artemia it should be sensitive to regulation by the phosphorylation state of the eIF-2¢ subunit. We and others [14,15] have shown that the a-subunit of Artemia elF-2 is phosphorylated by a casein kinase type I1 isolated from this organism. In fact, th/s is the only elF-2a kinase we have found till now in developing embryos of Artemia. The availability of a GEF-dependent reaction for nucleotide exchange on Arten~a elF-2 gave us the opportunity to test the effect of th~ kinase under the experimental conditions in which an HC|-like regulatory mechanism is operative. As shown in this paper, the Artemia casein kinase type i|, unlike rabbit reticulocyte HCI, has no effect on either the GEF-dependent nucleotide exchange or the factor-independent one. So, we can speculate that, irrespective of the kind of mechanism for guanine nucleotide exchant,e that is actually operating in viva, Artemia elF-2 might not be a target for regulation via elF-2a phosphoryla~ion. Aeknowledgement~ This investigation was supported by grants ham the Comisi6n Asesora de Investigaci6n and the Fondo de |nvestigaciones Sanitarias (Spain). We thank Dr. S. Ochoa for helpful discussions and Drs. M. Plana and E. Itarte for their help in the casein kinase type |I assays of Artemia fractions. F.G.M. and O.V. were recipients of fellowships from the Fondo de Investigaciones Sanitaria~ and Ministerio de Educacibn y Ciencia (Spain), respectively. The expert technical assistance of F. Ocaha is acknowledged. Relerences 10ehoa, S. (1983) Arch. Biochem,Biophys.223. 325-349. 2 Moldave, K. (1985) Annu. Rev. Biochem. 54, 1109~|149. 3 Pain, V.M. (1986) Bioehem.$. 235, 625-637.
4 Malls, R.L., Levin, D.H. and London, I.M. (1983) Pw¢. Nail. Acad. Sei. USA 80, 2559-2563. 5 Siekierka,J., Manne, V. and Ochoa, S. (1984) Proc. Natl. Acad. Sci. USA 81, 352-356.
60 6 Thomas, N.S.B., Matts, R.L., Petryshyn, R. and London, I.M. (1984) Pro¢. Natl. Acad. Sci. USA 81, 6998-7002. 7 Marts, R.L, Levin, D.H. and London, I.M. (1986) Pro¢. Natl. Acad. Sci. USA 83~ 1217-1221. 8 De Haro, C., Datta, A. and Ochoa, S. (1978.) Pro¢. Natl. Acad. Sci. USA 75, 243-247. 9 Malathi, V.G. and Mazumder, IL (1978) FEBS Lett. 86, 155-159. 10 Woodley, C.L., Roychowdhury, M., Macrae, T.H., Oisen, K.W. and Wahba" AJ. (I981) Fur. J. Biochem. 117, 543-$$1. 11 Mehta, ll.B., Woodley, C.L. and Wahba, A.J. (1983) J. Biol. Chem. 258, 3438-3441. 12 Wahba, AJ. and Woodley, C.L. (1984) Prog. Nucleic Acid Res. Mol. BioL 31, 221-265. 13 Mateu, M.G., Vicente, O., CarvaUo, P., Allende, J., Sierra, J.M. and Oehoa" S. (1983) 1501 FEILq Meetinll, p. 239 (abstr.). 14 Tho~, C. and Slegers, H. (1985) Biochim. Biophys. Aeta 825, 268~279, 15 Mehta, H.B., Dholakia" J,N,,Roth, W.W., Parekh, B.S., Montelaro,
16 17 18 19 20 21 22 23 24
R.C., Woodley, C.L. and Wahba, A.J. (1986) J. Biol. Chem. 261, 6705-6711. Msteu, M.G. and Sierra, J.M. (1987) Eur~ J. B]ochem. 165, 507-513. Laemmi, U.K. (1970) Nature 227, 680-685. Martos, C., Plan& M., Guasch, M.D. and itarte, E. (1985) Biochem. J. 225, 321-326. Mateu, M.G., Vieente, O. and Sierra, J.M. (1987) Eur. J. Biochem. 162, 221-229. Warner, A.H., Macrae, T.H. and Wahba, A.J. (1979) Methods Enzymol. 60, 298-311. Sierra, J.M., Meier, D. and Ochoa" S. (1974) Pro¢. Natl. Acad. Sei. USA 71, 2693-2697. Macfae, T.H., Royehowdhury, M., Houston, K.J., Woodley, C.L. and Wahba" A J. (1979) Fur. J. Biochem. 100, 67-76. Abroad, M.F., Nasrin, N., Bagehi, M.K., Chakravarty, !. and Gupta" N.K. (1985) J. Biol. Chem. 260, 6960-6965. Ostethout, J.J., Lax, S.R. and Ravel, J.M. (1983) J. Biol. Chem. 258, 8285-8289.
55
BBA 91913
Phosphorylation and guanine nucleofide exchange on polypepfide chain in/fiat/on factor-2 from Artemia embryos Mauricio G. Mateu, Federico G. Maroto, Oscar Vicente and Jos6 M. Sierra Centro de Biologla Molecular, Consejo Superior de Investigaciones Cientificas and Umversidad Aut6noma de Madric~ Madrid (Spare)
(Received 29 April 1988) (Revised manuscript received ? September 1988)
Key words: Protein synthesis; Initiation factor 2: Guanine nucleotide exchange: Protein phosphory|ation; (Artemia)
Et~taryotic initiation factor-2 (e|F-2) |tom Artemia embryos is able to exchange guanine nuclcotides at the s~t~e r~te in the presence or absence of Mg 2+ when the reaction is carried out with either purified e|F-2 at 30°C or less ~ r i [ i ~ preparations at any temperature (|0-~30°C). No exchange factor appears to catalyze this reaction. However, with purified e|F-2 at lower temperatures (|0 o C) the exchange is clearly impa|red by Mg 2. and this intpai|'ment is overcome by the guanine nuclcotide exchange factor (GEF) of rabbit reticu|ocytes. Thus, A~emia e|F-2 is able to exchange guanine nuc|eotides by two alternative mechanisms ~hat may reflect two states of the protein. Pbospbov~|afion of the e|F-2a submit by the heine-controlled inhibitor (HC|) of rabbit retieulucytes abolishes the GEF-dependent reaction, but has no effect on the factor-independent one. The search for e|F-2a kinases in A rtemia e.nbryo |ed to the detection of only one such enzyme, which was identified as a casein kinase type ||. None of the exchange reactions is affected by the pbusphory|ation of the elF-2a sublmit by this kinase, suggesting that, irrespective of the kind of mechanism for guanine nucleotide exchange that is actuary operating in Artemig, it might not be a target for regulation by e|F-2a phospboryiation.
Introduction
GEF is involved in elF-2 recycling during polypeptide chain initiation in mammalian cells (for reviews, see Refs. 1-3). Upon completion of a round of initiation an elF-2. GDP complex is formed and because at physiological concentrations of Mg 2+ the affinity of elF-2 for GDP is much higher than for GTP, the exchange of bound GDP for free GTP, which is required to form the ternary complex elF-2. GTP. MettRNA~, is impaired. This impairment, however, is overcome by the catalytic action of GEF, which promotes the G D P - G T P exchange. The GEF-dependent reaction is a target for regulation of polypeptide chain initiation, at least in rabbit reticulocytes. Phosphorylation of elF-
Abbreviations: GEF, guanine nucleotide exchange factor; elF-2, eukaryotic protein synthesis initiation factor-2; HCI, heme-controlled translational inhibitor (an elF-2a kinase); Met-tRNA i, initiator methionyl-tRNA, SDS, sodium dodecyi sulfate; PAGE, polyacrylam|de gel electrophoresis. Correspondence: J.M. Sierra, Centro de Biologla Molecular. Universidad Autbnoma de Madrid, Cantoblanco, 28049 Madrid° Spain.
2a subunit by HCI blocks the exchange reaction be, cause all the available GEF gets tied up as an inactive elF-2(aP). GEF complex [4-7]. The above mechanism of translational regulation is being investigated, among other lower eukaryotes, in Artemia. In developing embryos of Artemia it was reported the presence of a protein that stimulated the activity of elF-2 in ternary complex formation [8,9], This effect was blocked when elF-2 was preincubated with rabbit HCI in the presence of ATP [8]. However, Wahba and collaborators have been unable to find a requirement for a G D P / G T P exchange factor with Artemia elF-2 [10-12]. They have shown that guanine nucleotide exchange on this factor can proceed readily in the presence of Mg 2+ in a reaction which is not stimulated by GEF from rabbit reticulocytes [11,12]. On the other hand, it is also known that the a-subunit of Artemia elF-2 is a substrate for rabbit reticulocyte HCI [10,12,13] and casein kinase type II isolated from Artemia or mammalian cells [14,15]. However, neither of these kinases abolished the above.mentioned GEFindependent exchange [12,15]. We have reinvestigated the guanine nucleotide exchange on Artemia elF-2 and have found that this
0167-4781/89/$03.50 © 1989 Elsevier Science Pubfishers B.V. (Biomedical Division)
56 factor, as previously shown for Drosophila embryo elF-2 [16], is able ~o exchange guanine nucleotides by two alternative mechanisms that may reflect two states of the protein. We confirm the occurrence of a GEF-independent exchange but we also find conditions where nucleotide exchange on Artemia elF-2 is impaired by Mg 2+ and this impairment is overcome by rabbit GEF. The ability of Artemia elF-2 to respond to rabbit GEF has allowed us to study also the effect of eIF-2a phosphorylation, by either rabbit HCI or Artemia casein kinase type II, on this reaction. While, as expected, the GEF-dependent exchange is abolished by rabbit HCI, it is not abolished in the presence of Artemia casein kinase type 11. This establishes an important difference between the two kinases with respect to the regulation of guanine nucleotide exchange by elF-2 phosphorylation. Materials and Methods Assays To assay elF-2a kinase activity, samples (36 ~tl)
containing 20 mM Hepes (pH 7.6), 60-200 mM KCI, 4 mM Mg(CH~COO)2, 1 mM dithiothreitol, ¢0/tM [y~2P]ATP (22200 cpm/pmol), 2.8 ttg of purified Artemia elF-2 and protein fractions to be ~¢,~sted,are incubated for 7 min at 30 o C. Labelled polypeptides were analyzed by SDS-PAGE [17] and autoradiography using Kodak X.Omat S film and !lford intensifying screen. The area corresponding to the phosph~4ated a-subunit of Artemia ¢iF-2 (,*2 kDa) is scot,ned at 626 n~~ Jn a Chromoscan 3 (Joyce-l.oebl) densi~ometer. Casein kinase activity was assayed essenti:d~ as described [18]. Guanine nucleo~ide exchange assays ~.verecan:ed out as in Ref. 16 but 1.8 ItM of [3HIGDP (8900 cpm/pmol, Amercham) and Anemia elF-2, in the amounts indicated in each experiment, were used. The time and temperature of incubation are indicated in the legends to the figures. Conditions for ternary and 40 S initiation complex formation have been reported [19]. Preporations
Batches of 125 g (dry weight) of encysted Artemia sp. embryos (San Francisco Bay Brand, Newark, CA) wer~ deve!~ped in ~.~fici~l ~ea water [20] at 27 ° C for 14-15 h and the postribosomal supernatant and polysomal fractions prepared as described [21]. Polysomes were rinsed with extraction buffer [21l, gently suspended in 25 ml of high.salt buffer (buffer C in ReL 21) and stirred for 4 h at 0 e C. The suspension was centrifuged for 165 rain at 219000 x g and the clear supernatant was collected and frozen. This fraction will be referred to as the ribosomal wash. A protein kinase capable of phosphorylating tee
a-subunit (42 kDa in SDS-PAGE) of Artemia elF-2 was detected in developing embryos of Artemia [10,13]. This activity is present in both the high-salt wash of polysomes and the postribosomal supernatant and has been further purified from the latter fraction by precipitation with ammonium sulfate. The 30-50~ saturation cut, in buffer A (20 mM Tris-HC! (pH 7.6)/1 mM dithiothreitol/0.1 mM EDTA/10~ glycerol) containing 130 mM KCI, was applied to a DEAE-cellulose column and the bulk of elF-2a kinase activity eluted with buffer A containing 350 mM KCI. This fraction was dialyzed against buffer B (50 mM Tris-HCl (pH 7.5)/1 mM dithiothreitol/0.1 mM EDTA/10~ glycerol)containing 50 mM KCI, applied to phosphocellulose and the retained pt~otein was eluted stepwise with buffer B conraining 200 raM, 350 mM or 1200 mM KCI. The elF-2a kinase, which eluted at 1200 mM KCI, was rechromatographed through phosphocellulose and purified further by glycerol density gradient centrifugation. The functional and structural properties of Artemia elF-2a kinase were found to be very similar to those reported [14] for a casein kinase type 11 isolated from a ribosomal high-salt wash of encysted Artemia embryos. That both activities were indeed associated with the same protein was also supported because they copurifled together. Artemia elF-2 was purified from the high-salt wash of developing embryos according essentially to the proce6ure of Macrae et ~1. [22], but the hydroxyapatite chromatography was omitted and the phosphocellulose step was substituted by CM.Sephadex chromatography as follows, elF-2 from DEAE-cellulose was applied to a CM-Sephadex column (10 mg protein/ml gel) equilibrated in buffer A containing 180 mM KCI. After washing the column with this buffer, elF-2 was eluted with buffer A containing 400 mM KCI and 0.5 mM phenylmethylsulphonyl fluoride. Occasionally, this fraction was further purified by sedimentation through glycerol density gradients. To obtain a partially purified fraction of Artemia elF-2, the high-salt wash (45 ml, 28 mg of protein) obtained from developing embryos was dialyzed against buffer B containing 80 mM KCI and applied to a CM-Sephadex column (1.6 x 12 cm) equilibrated in the same buffer. Retained protein was eluted stepwise with buffer B containing 180 mM and 360 mM KCI. elF-2 (20 mg of protein), which was eluted at 360 mM KCI, was dialyzed against buffer B containing 80 mM KCI and further fractionated through heparin-Sepharose (Pharmacia, 0.6 x 2 cm) with 6 ml of a 80-500 mM KCi gradient. Fractions with higher elF-2 activity were pooled and kept frozen until use. HC! and GEF, both from rabbit reticulocytes, Artemia 40 S ribosomal subunits, [y-32p]ATP, [3H]MettRNA i, and determination of protein concentration were as in previous work [16,19].
57 1
A
23
~
S
0.6
-2 O
0
E O. "O C
E
0.4
\ \
C1.
,,Hcx,G F
Q,
+Mg 2÷,
"O C
+GDP
O
O .a
{3. 1 C~
¢:L
e-,--.,t
0.2
\M0
; ATP, K,, GEF "
:x:::
,6-.-
2
A.
6
Time (rain}
8
12
3{56
7891Gll
-÷
÷ ÷ ÷ ÷
÷" ÷ - ÷
÷ .
.
.
.
÷
- ÷ ÷ - ---÷÷
12 ÷ ATP HCI
_
~"
KA
FiB. I. GEF-dependent nucleotide exchange on Artemia elF-2 and the effect on this reaction of the elF-2a phosphorylation by Iabbit HC| or drlemia casein kinas¢ type !! (KA). elF-2.[3H]GDP complexes were preformed with 1.4 pg CA) or 4 pg (B) of purified Artemia c1|:-2 (CM-Sephadex fraction) and incubated in the presence of I mM Mg(CH~COO) 2 (@, [3, v, A, in (A): ! - ! 2 in (B)) or water (o in CA)), 40/~M ATP ([3° % in (A); 2-7, 9, 11, 12 in (B)), and 0.2/ag of HC! (13, in CA); $o 8, 9 in (B)) or 22/ag of Artemza elF-2a kinas¢ (KA) (DEAE-cellulose fraction) (st in (A); 1, 6, 10, 11, in (B)) for 5 rain at 30°C. Then, nucleotide exchange reactions in the presence of 2 mM EDTA (sample 12B), 20/aM GDP (all samples except IB, 2B) and 1/ag of rabbit reticulocyte elF-2. GEF (ra. st, A in (A); 7-11 in (B)) were carried out at l0 o C (A). 30 o C ( 1, 2, 4-12 in B) or 37 ° C (31}) for 4 min (B) or as indicated in (A). Background value without elF-2 (0.08 pmol) has been substracted. (C) Phosphorylation of Artemia elF-2a under reaction conditions of the experiment in panel B. [y-32p]ATP (15/aM, 72000 cpm/pmol) instead of unlabellcd ATP was used. Samples in tracks 1, 3 and 4 correspond to samples 7, 11 and 9 of panel B, respectively. Control samples with Artemia kinase (track 2) or HC! (track 5) were as samples 6B and 5B, respectively, bw without elF-2. Samples were subjected to SDS-PAGE and the gel was dried and autoradiographed. Arrow shows the position of the a-subunit of Artemia elF-2. The level of ©iF-2a phosphorylation with Artemia kinase is higher compared to that obtained with HCI just because the relative amounts of each kinase used.
Results Guanine nucleotide exchange on Artemia elF-2 can proceed by two different mechanisms
We have used the reaction: e~F-2. [3HIGDP + GDP ~ elF-2. GDP + [3HIGDP
as a model [1,4-7,16] to study guanine nucleotide exchange on Artemia elF-2. Mehta et al. [11] have pointed out that, in contrast to rabbit reticulocyte elF-2, the Artemia factor was able to readily exchange bound GDP for free GTP in the presence of Mg 2+ in a reaction which was not stimulated by added rabbit GEF. However, we have found that with relatively highly purified preparations of Artemia elF-2, guanine nucleotide exchange may or may not be impaired by Mg 2+, depending on the reaction temperature. Moreover, this impairment can be overcome by rabbit reticulocyte G E F (Fig. 1). At 10 ° C the exchange of elF-2 bound [3H]GDP with free unlabelled GDP proceeds readily in the absence of Mg 2+ (Fig. 1A, open circles) but it is clearly impaired in the presence of this cation (Fig. 1A, filled circles). The impairment by Mg 2+ can also be seen at 30°C (Fig. 1B, compare bars 4 and 12,
which correspond to samples containing or lacking free Mg 2+, respectively), but to a lesser extent because a significant percentage of exchange has already taken place at this temperature in the presence of Mg 2+ (Fig. I B, compare bars 2 and 4). However, guanine nucleotide e~change is not affected at all by the presence ¢1" Mg 2+ when the reaction is carried out at 37°C (Fig. ~.B, bar 3). These results prompted us to investigate the effect of rabbit reticulocyte GEF on the impairment by Mg 2+ of guanine nucleotide exchange on Artemiq~ elF-2. As shown in Fig. IA, the impairment by Mg 2÷ observed when the reaction is carried out at 10 °C is completely abofished by the addition of rabbit reticulccyte GEF (Fig. 1A, compare filled circles and filled triangles). The same result is obtained when the exchange reaction is carried out at 30°C in the presence of GEF (Fig. 1B, compare bars 4 and 7, which correspon~ to samples lacking or containing GEF, respectively). These data show that Artemia elF-2 can respond to rabbit GEF. In other words, a GEF-mediated mechanism for guanine nucleotide exchange can function with Art'emia elF-2. We have further found, confirming a~d extending previous observations [11,12], that when less purified preparations of Artemia elF-2 are used, guanine
58
-GDP 1.2
A
_-)
~0.8 ,ID
÷ t,
flOP
.
1
2
3
z,
Time (rain)
10
20
30
Temperature (eC)
u
1 345 678910111213 -÷,÷, - ÷ - ÷ - ÷ - ÷ - - ÷ . . . . ÷ ÷ . . . . .
11, - ATP HCI
"''÷ . . . .
" KAil) - KA(2)
. . . . . ++'" + . . . . . . + ÷
Fig. 2. ~ ~ a s nucleotideexcbat~e on Anemia elF.2 whichis not impaired by Mg=* is unaffected by the phosphorylation of elF-2a. (A) and (B) elF-2.[SH]GDP compl~es werepreformed wi~h9 tt$ of partially purified Artemia elF.2 and incubated as for nu¢leotide exchange reaction in the ~ of I tam M~CHsCOO)a (o), water (o) and 20 MMGDP (0, o). (C) elF-2.[3H]GDP complexes preformed with 12 itS of elF-2 were incubated in the presenceof ! mM Mg(CH3COO)=, 35 ~tM ATP (2-$, 7, 9, 11.13). and 0.4 tt8 of HCI (3. 8. 9) or Artemia kinase (K^) (?it 8 of the 8 1 ~ 1 ~-adieat fraction in 4, 10. 11; 8.8 p$ of the DEAE-cdlulosefraction in S. 12. 13) for 5 rain at 30 o C. Then. ,ucleotide exchange reactionsin the presenceof 2 tam EDTA(sample 14) and 20 ptM GDP (6-14) were carried out. Backgxoundvalues without elF.2 (0.~ pmol in (A). 0.03 pmol in (B) and (C)) have been substracted, nucleotide exchange can proceed at a high rate, whether Mg 2+ is added or not (Fig. 2A) and independently of the temperature of incubation (Fig. 2B). A straightforTABLB I A~ o/,~i~t~,4e e~cloa,Re /actor m p,m~ally purified Anemia
¢1F~2 B~aarycomplexeswere performed with Anoma elF-2 ( $ 1 ~ wadieat ffaoti~) in the amount ~own and further incubated as for n ~ ~c~ ~ t i o n in the abtea~ or presenceof partially purified A ~ d elF..2 and the other ~pone.ats indicated. Back,, 8gouad value8with elF-2 (0`10and 0.25 pmol without and with ?.$ tt8 of partially purified elF-2 respectively)haw been substracted, n.d., not detmnined. IExp, Temp. Nudeotide exchange No, (eC) ~ t i o n
10
[ ~HIGDP bound (pn~)
elF.2, I3H) GDP
partially GDP +Ms=* p~fied OS ~M) (t raM) elF.2
- M s 2÷
0,9
0 0 6 12
+ + +
0,39 0,39 0,39 0.43
n.d, 0`10 0`03 n,d.
0 0
+
1.37 0.32
n.~. . 9~8
3 "1.5 12
+ + +
0`51 0.30 0.28
n.d. 0,02 n,d.
3.0
ward explanation for these results is the efistence, in partially purified Anemia elF-2, of a guanine nucleotide exchange factor which would be responsible for the endogenous reaction. However, this possibility seems to be ruled out on the basis of the experiments shown in Table I. The impairment by Mg 2+ of the [ ~ H ] G D P / G D P exchange on purified elF-2 at 1 0 ° C or 3 0 ° C is not overcome by tile addition to the reaction mixture of an excess of the less purified Artemia elF-2 as a source of the putative exchange factor, Therefore, it appears that in certain experimental conditions (by using less purified preparations of elF-2 or increasing the temper° ature at which the exchange reaction is performed) Artemia elF.2 is able to exchange guanine nucleotides readily, irrespective of the absence or presence of Mg 2+ and without the requirement of an exchange factor.
Effect of phosphorylation of the elF-2a subunit by rabbit reticulocyte HC! or Artemia casein kinase H on the guanine nucleotide exchange reactions Phosphorylation of the a-subunit of Artemia eIFo2 by rabbit reticulocyte HCI abolishes the GEF-dependent guanine nucleotide exchange reaction (Fig. 1A, compare filled triangles and open squares; Fig. 1B bar 9). The level of phospho~lation of eIF-2a by this kinase, under the experimental conditions with which the exchange reaction was carded out, is shown in Fig. 1C (track 4). In the light of this result we have searched
59 for eIF-2a kinases in developing Artemia embryos and found only one such enzyme that turned out to be very similar, if not identical, to a previously reported [|4] casein kinase type II isolated from encysted embryos of Artemia. In contrast to rabbit HCL drtemia casein kinase type II has no effect at all on the GEF-dependent nucleotide exchange reaction (Fig. IA, compare open triangles and open squares: Fig. 1B, bar 11) in spite of its being able to phosphorylate the ¢-subunit of elF-2 (Fig. 1C, track 3). The above data suggest that the GEF-dependent guanine nuc|eotide exchange on Artemia elF-2 observed in vitro may be capable of regulation by elF-2a kinases of the rabbit reticulocyte HCI type. In addition, they establish an important difference between Artemia casein kinase type 11 (with elF-2a kinase activity) and rabbit reticulocyte HCi with respect to the regulation of guanine nucleotide exchange by elF-2 pbospborylation. With respect to the factor-independent nuclcotide exchange (Fig. 2, A and B), we have confirmed [12,15] that it is not affected by the phosphorylation of the a-subunit of Artemia eIF-,'~ by either rabbit HCI (Fig. 2C, bar 9) or Artemia casein kinase type II (Fig. 2C, bars 11 and 13). It should be noted that the lack of inhibition by HCI of this nucleotide exchange is a further support for the notion that it is not mediated by a GEF-like protein. Also, any regulatory role for Artemia casein kinase type 11 in either of the two guanine nucleotide exchange mechanisms observed in vitro with Artemia elF-2 seems to be ruled out. Discussion
From the studies on guanine nucleotide exchange with Drosophila elF-2 [16] we have suggested the existence of two alternative mechanisms for this reaction: a GEF.dependent exchange, similar to that described in rabbit reticulocytes and other mammalian cells, and a factor-independent exchange. This hypothesis can be now extended to Artemia elF-2 on the basis of the results shown in this paper. In addition to confirm and extend the observation [11,12] that Artemia elF-2 can exchange bound GDP for free. nucleotide at the same rate in the presence or absence of Mg 2+, ih a reaction in which no .~xchange factor appears to be involved, we have found that with purified elF-2 guanine nucleotide exchange is clearly impaired by Mg 2+ provided the reaction is carried out at low temperatures (10°C). Moreover, this impairment is overcome by rabbit GEF. The ability of Artemia elF-2 to respond to rabbit G E F proves that it has all the structural requirements for the interaction w/th a guanine nucleotide exchange factor and therefore opens the possibility that a GEF-mediated mechanism for guanine nucleotide exchange may be operative in Artemia embryos. However, this possibility should be reinforced by the isolation from this organism
of the putative GEF-Iike protein that has remained elusive until now. Our data are also compalible with the notion that the response of Artemia elF-2 to rabbit GEF is just an evolutionafily conserved property acquired before GEF and/o~ HCI appeared in mammalian cells. In this respect, it would be interesting Io ascertain whether this property, which is also observed in Drosophila elF-2 [16], can be extended to other lower eukaryotes, such as yeast or wheat, in which no evidence for the occurrence of a GEFolike factor has so far been reported [23,24]. On the other hand, the inhibition by HCI of the GEF-dependent nucleotide exchange shown in th/s paper supports the notion that if a GEF-mediated mechanism for guanine nucleotide exchange e~dsted in Artemia it should be sensitive to regulation by the phosphorylation state of the eIF-2¢ subunit. We and others [14,15] have shown that the a-subunit of Artemia elF-2 is phosphorylated by a casein kinase type I1 isolated from this organism. In fact, th/s is the only elF-2a kinase we have found till now in developing embryos of Artemia. The availability of a GEF-dependent reaction for nucleotide exchange on Arten~a elF-2 gave us the opportunity to test the effect of th~ kinase under the experimental conditions in which an HC|-like regulatory mechanism is operative. As shown in this paper, the Artemia casein kinase type i|, unlike rabbit reticulocyte HCI, has no effect on either the GEF-dependent nucleotide exchange or the factor-independent one. So, we can speculate that, irrespective of the kind of mechanism for guanine nucleotide exchant,e that is actually operating in viva, Artemia elF-2 might not be a target for regulation via elF-2a phosphoryla~ion. Aeknowledgement~ This investigation was supported by grants ham the Comisi6n Asesora de Investigaci6n and the Fondo de |nvestigaciones Sanitarias (Spain). We thank Dr. S. Ochoa for helpful discussions and Drs. M. Plana and E. Itarte for their help in the casein kinase type |I assays of Artemia fractions. F.G.M. and O.V. were recipients of fellowships from the Fondo de Investigaciones Sanitaria~ and Ministerio de Educacibn y Ciencia (Spain), respectively. The expert technical assistance of F. Ocaha is acknowledged. Relerences 10ehoa, S. (1983) Arch. Biochem,Biophys.223. 325-349. 2 Moldave, K. (1985) Annu. Rev. Biochem. 54, 1109~|149. 3 Pain, V.M. (1986) Bioehem.$. 235, 625-637.
4 Malls, R.L., Levin, D.H. and London, I.M. (1983) Pw¢. Nail. Acad. Sei. USA 80, 2559-2563. 5 Siekierka,J., Manne, V. and Ochoa, S. (1984) Proc. Natl. Acad. Sci. USA 81, 352-356.
60 6 Thomas, N.S.B., Matts, R.L., Petryshyn, R. and London, I.M. (1984) Pro¢. Natl. Acad. Sci. USA 81, 6998-7002. 7 Marts, R.L, Levin, D.H. and London, I.M. (1986) Pro¢. Natl. Acad. Sci. USA 83~ 1217-1221. 8 De Haro, C., Datta, A. and Ochoa, S. (1978.) Pro¢. Natl. Acad. Sci. USA 75, 243-247. 9 Malathi, V.G. and Mazumder, IL (1978) FEBS Lett. 86, 155-159. 10 Woodley, C.L., Roychowdhury, M., Macrae, T.H., Oisen, K.W. and Wahba" AJ. (I981) Fur. J. Biochem. 117, 543-$$1. 11 Mehta, ll.B., Woodley, C.L. and Wahba, A.J. (1983) J. Biol. Chem. 258, 3438-3441. 12 Wahba, AJ. and Woodley, C.L. (1984) Prog. Nucleic Acid Res. Mol. BioL 31, 221-265. 13 Mateu, M.G., Vicente, O., CarvaUo, P., Allende, J., Sierra, J.M. and Oehoa" S. (1983) 1501 FEILq Meetinll, p. 239 (abstr.). 14 Tho~, C. and Slegers, H. (1985) Biochim. Biophys. Aeta 825, 268~279, 15 Mehta, H.B., Dholakia" J,N,,Roth, W.W., Parekh, B.S., Montelaro,
16 17 18 19 20 21 22 23 24
R.C., Woodley, C.L. and Wahba, A.J. (1986) J. Biol. Chem. 261, 6705-6711. Msteu, M.G. and Sierra, J.M. (1987) Eur~ J. B]ochem. 165, 507-513. Laemmi, U.K. (1970) Nature 227, 680-685. Martos, C., Plan& M., Guasch, M.D. and itarte, E. (1985) Biochem. J. 225, 321-326. Mateu, M.G., Vieente, O. and Sierra, J.M. (1987) Eur. J. Biochem. 162, 221-229. Warner, A.H., Macrae, T.H. and Wahba, A.J. (1979) Methods Enzymol. 60, 298-311. Sierra, J.M., Meier, D. and Ochoa" S. (1974) Pro¢. Natl. Acad. Sei. USA 71, 2693-2697. Macfae, T.H., Royehowdhury, M., Houston, K.J., Woodley, C.L. and Wahba" A J. (1979) Fur. J. Biochem. 100, 67-76. Abroad, M.F., Nasrin, N., Bagehi, M.K., Chakravarty, !. and Gupta" N.K. (1985) J. Biol. Chem. 260, 6960-6965. Ostethout, J.J., Lax, S.R. and Ravel, J.M. (1983) J. Biol. Chem. 258, 8285-8289.