Developmental changes in poly(A) polymerase activity in Artemia

Eur. J. Biochem. 135, 69-74 (1983) 0FEBS 1983

Developmental changes in poly(A) polymerase activity in Artemia Leandro SASTRE and Jesus SEBASTIAN Instituto de Enzimologia y Patologia Molecular del Consejo Superior de Investigaciones Cientificas, Facultad de Medicina de la Universidd Autonoma de Madrid

(Received February 9, 1983) - EJB 83 0110

The levels of poly(A) polymerase activity have been determined during Artemiu early development. Poly(A) polymerase activity increases steadily during postgastrular embryonic development reaching a maximum shortly after hatching. The rise of poly(A) polymerase is concomitant with an increase in poly(A) content and with a change in the subcellular distribution of the enzyme activity, the major increase corresponding to the nuclear fraction. Only one isoenzyme of poly(A) polymerase has been identified in Artemiu embryos and nauplii despite changes in enzyme levels and subcellular changes during early development. Poly(A) polymerase is not associated with the cytoplasmic poly(A)-containing ribonucleoprotein particles stored in Artemiu dormant embryos.

Poly(A) polymerase is the enzyme implicated in the synthesis of the poly(A) chain present at the 3‘ end of most mRNA molecules. This enzymatic activity has been characterized in several biological systems [I], including Artemiu dormant embryos [2]. Polyadenylation is a general step in the nuclear processing of mRNA sequences [3]. Besides nuclear polyadenylation, elongation of the poly(A) chains can also take place in the cytoplasm [4]. Cytoplasmic polyadenylation can play a role in the stability of mRNA and in the activation of maternal mRNA in developing systems. Dormant embryonic systems and mature oocytes have a population of polyadenylated mRNA stored as mRNP particles [5,6], which are translated after fertilization dr resumption of development [7- 101. Activation of stored mRNA in embryonic systems is an important example of posttranscriptional regulation of gene expression. However, the mechanisms involved in this process are unknown. Activation of stored inRNA is accompanied by an increase in the poly(A) chain length of mRNA in several systems [I 1 - 131, suggesting that polyadenylation of stored mRNA could be involved in their activation. Little is known about the developmental function of cytoplasmic poly(A) polymerase besides its possible role in the elongation of the poly(A) chain of stored mRNA. Poly(A) polymerase has been described in sea-urchin embryos [14], where the activity is high in the oocyte, decreasing progressively during embryonic development. The high levels of enzymatic activity in the oocyte could account for the increase in poly(A) after fertilization [15]. We have approached the study of poly(A) polymerase activity during Artemia early development. The oviparous developmental pathway of Artemiu involves a cryptobiotic period at the gastrula stage. The encysted dormant embryos are viable for years and can resume development giving rise to free swimming nauplii in about 16- 18 h at 30°C under suitable conditions of temperature and oxygenation. Artemiu cysts contain stored polyadenylated mRNP particles, which Abbreviation. mRNP, messenger ribonucleoprotein Enzyme. Poly(A) polymerase (EC 2.7.7.19).

are membrane-bound [16] or free in the cytoplasm [17]. Amaldi et al. [18] have shown that protein synthesis starts as soon as dormant embryos resume development. During postgastrular embryonic development there is a net increase in the levels of poly(A) and a mobilization of stored mRNA to polysomes [19--211. Therefore, Artemiu is a useful model system to study the mechanisms involved in the activation of stored mRNA after dormancy and the role of polyadenylation in this process. In this paper we report the characterization and the changes in enzyme levels and subcellular distribution of poly(A) polymerase during early development of Artemiu. The partial purification and properties of a poly(A) polymerase from Artemiu dormant embryos has been previously described

PI. MATERIALS AND METHODS Chemicals and buffers

Nucleotides and homopolymers were obtained from Sigma Chemical Co., Bio-Rex 70 was purchased from Bio-Rad and DEAE-cellulose from Serva. Ficoll 400 was obtained from Pharmacia. Pancreatic RNase and Torulu RNA were purchased from Calbiochem and 3H-labelled nucleotides and [3H]poly(U) from Amersham International. Glass-fiber filters were obtained from Whatman. All other chemicals were of analytical grade. Buffer A contains 50mM Tris/HCl, 0.2 mM EDTA, 5 mM 2-mercaptoethanol, 0.6 M ( N H ~ ) Z S O20 ~ ,% glycerol, pH 8.5. Buffer B contains 50 mM Tris/HCl, 0.2 mM EDTA, 5 mM 2-mercaptoethanol, 20 % glycerol, pH 8.5. Buffer C contains 25 mM Hepes, 0.3 M sucrose, 15 ”/, Ficoll 400, 5 mM MgC12, 0.5 mM CaC12, pH 7.5. Orgunisms and growth conditions Artemiu cysts were obtained from San Francisco Bay Brand, Division of Metaframe Co. (Menlo Park, CA, USA).

70

Treatment of dry cysts and growth conditions were as described elsewhere [22]. Isolation of newborn nauplii, synchronous cultures of developing nauplii and feeding conditions of larvae were as reported [23]. 5 and 10-day-old nauplii were obtained from the same batches of cysts. They were grown at the Instituto de Acuicultura del CSIC (Castel16n) by Dr F. Amat.

Table 1 . Subcrllular distribution Xenesis Developmental stage, with development time (h)

of p o l y ( A )

during Artemia embryo-

Poly(A) content nuclear fraction

post-

total

nuclear fraction

pg/I Os emhryos

Poly(A) polymerase assay

_.___

The poly(A) polymerase assay was carried out according Sastre and Sebastian [2] using Torula RNA as primer. 1 unit of poly(A) polymerase activity was defined as the amount of enzyme that catalyzes the incorporation of 1 nmol AMP into trichloroacetic-insoluble material in 60 min under assay conditions.

Dormant embryos (To) Developing embryos (T5) Developing embryos (Tlo) New hatched nauplii (TI*)

0.7 1.3 3.1 3.3

~~

2.2 3.3 6.7 7.9

2.9 4.6 9.8 11.2

RESULTS Subcellular,fractionation

Dormant, developing embryos and nauplii were homogenized in 2.5~01.buffer C with a glass/glass manual homogenizer. The homogenate was centrifuged at 6000 x g for 5 min to obtain a particulate fraction containing nuclei and yolk granules [24]. The supernatant was the postnuclear fraction. The supernatant was centrifuged at 15000 x g for 15 min to sediment a membrane-rich fraction containing mitochondria [25].The supernatant was centrifuged again at 105000 x g f o r 120 min to obtain the microsomal and cytosolic fractions.

Nucleic acid extraction and polyadenylic acid determination Aliquots of subcellular fractions or fractions from the isopycnic sucrose gradient were supplemented with 250 pg heparin/ml and sodium dodecyl sulfate to 0.5 o/, final concentration and extracted three times with phenol/chloroform/ isoamyl alcohol (25:24:1). The aqueous phase was made 0.3 M in NaCH3COO and precipitated with 2 vol. cold ethanol at - 20 "C overnight. After centrifugation at 1 5 000 x g for 10 min the pellet was resuspended in distilled water. Buffers used for R N A preparation were treated with 1 o/, diethyl pyrocarbonate and sterilized. Polyadenylic acid content was determined by hybridization with [3H]poly(U)using poly(A) as standard. The hybridization mixture contained 300 ng [3H]poly(U) (50- 200 counts min-' ng-') in 0.5 ml NaCl/Cit. After addition of the RNA, the hybridization mixture was incubated at 45°C during 18 h. Non-hybridized R N A and H]poly(U) were hydrolyzed by incubation of the hybridization mixture with 15 pg pancreatic ribonuclease at 30°C for 15 min. The reaction was stopped by addition of 150 pg Torula RNA and 3 ml ice-cold 5 "/, trichloroacetic acid/O.l uranyl acetate. After 15 min at 4 "C, trichloroacetic-acid-insoluble material was collected on Whatman glass filters. Filters were washed, dried and counted in a toluene-based scintillation fluid in a liquid scintillation counter.

r3

Protein determinulion Proteins were determined by the method of Lowry et al [26] using bovine serum albumin as standard.

Levels ofpoly ( A ) during the postgastrular embryonic development of Artemia Levels of poly(A) were determined in embryos at different stages of development and in newly hatched nauplii. Nuclear and postnuclear fractions were obtained, nucleic acids extracted and poly(A) content determined as described in Materials and Methods. Table 1 shows the changes in poly(A) content in these fractions. Poly(A) increases continuously from the time of resumption of development. The amount of poly(A) increases almost 5 times in the nuclear fraction and 3.5 times in the postnuclear fractions. The increase of poly(A) in the nuclear fraction could be correlated with the activation of messenger R N A transcription and the increase in the postnuclear fraction with the adenylation of stored cytoplasmic mRNA and transport of nuclear polyndenylated mRNA, concomitant with the activation of protein synthesis. Levels and suhcellular distribution of poly ( A ) polymerase activity during Artemia early development We have studied the levels of poly(A) polymerase activity during Artemia early development to look for a possible role of this enzyme in the regulation of the reported developmental changes in poly(A) concentration. Synchronous populations of Artemiii cysts and nauplii were homogenized in 2.5 vol. buffer A in order to solubilize the total poly(A) polymerase activity. Buffer A was supplemented with 50 pg soybean trypsin inhibitor/ml to prevent proteolysis from Artemiu larval proteases [27]. Homogenates were centrifuged at 10500Oxg for 2 h and the resulting supernatants were dialyzed against buffer B. Poly(A) polymerase activity was determined using Tomla R N A as primer. Fig. 1 shows total poly(A) polymerase activity present at diflerent times of Artemiu development. Poly(A) polymerase activity increases twofold during embryonic development. The enzymatic activity decreases thereafter and levels off about 24 h after hatching and remains almost constant through late larval development. The use of poly(A) as R N A primer and the substitution of manganese by magnesium in the enzymatic assay do not alter the pattern of poly(A) polymerase activity during Artemia development. The results reported in this paper d o not allow us to make conclusions about the mechanisms involved in the changes of levels in enzymatic activity. Mixing experiments show that enzyme inhibitors are not involved in the regulation of poly(A) polymerase levels. Poly(A) polymerase activity,

71 hatching

i

200

x

.-

L

> .c _ 0 0 ' " 0

m o

.Ec

60

t

a

150

b 9 E, 100 -

owl -a .= C

Ix < 2 50

-0

!-A-lP

a

0 I---I--L---LI--I*tL 0 10 20 30 LO 5 10 Time of development lhl

ldoysl

Fig. 1. Totalpoly ( A ) polymerase activity during Artemia development. 2 g embryos and nauplii at different stages of development were homogenized with 2.5 vol. buffer A. The homogenates were ccntrifuged at 105000 x g to obtain the soluble extracts. They were dialyzed against buffer B and assayed for poly(A) polymerase activity using Torula RNA as primer and manganese as divalent cation

Table 2. Suhcellular distribution ofpoly ( A ) polymerase activity during Artemia cwly developmmt Developmental stage, with development time (h)

Poly(A) polymerase activity -

~.

~~

+

nuclei yolk granules

-~

.-

cytosol

units/I05 animals Dormant embryos (To) Developing embryos (T5) Developing embryos (Tlo) Newly hatched nauplii (T, 8) Developing nauplii (T24) Developing nauplii (T30)

20 25 35 80 72 50

50 47 46 85 70 50

determined in mixed extracts from late and newly hatched nauplii, is that expected from the addition of the activities from the individual extracts. The subcellular distribution of poly(A) polymerase activity has been determined in dormant, developing embryos and nauplii. All the activity present in the cellular homogenates was found in the nuclear and cytosolic fractions, whereas the mitochondria1 and microsomal fractions contain almost undetectable levels. Table 2 shows the distribution of poly(A) polymerase activity in nuclear and cytosolic fractions from embryos and nauplii at different times of development. The increase in poly(A) polymerase activity takes place mainly in the nuclear fraction, where there is a fourfold increase of the enzymatic activity during postgastrular embryonic development. The increase of poly(A) polymerase in cytoplasm is less than twofold during the same period of development. Therefore, the increase of poly(A) polymerase activity during embryonic development is mostly due to the enzyme located in nuclei.

Churacterization of poly ( A ) polymerase activity .from Artemia nauplii Multiple poly(A) polymerase with different subcellular location and chromatographic properties have been reported in several eukaryotic systems [1,28]. The increase in poly(A) polymerase activity during Artemiu development as well as

10

20 30 LO Fraction number

50

Fig. 2. Column electrofbcusing of p o l y ( A j polymerase .from Artemia nauplii. 2 mi soluble extract from Arfernia nauplii were subjected to electrofocusing in a 105-ml column loaded with a 5 - 50 sucrose gradient containing 2.5';/, ampholytes pH 3- 10. The electrofocusing was run at 1200 V for 60 h at 0°C. Fractions of 1.5 ml were collected from the bottom of the column and the pH and poly(A) polymerase activity were determined

its changes in subcellular distribution makes worthwhile the investigation of the existence of different poly(A) polymerase isoenzymes in Artemia nauplii. Newborn nauplii were homogenized in 2.5 vol. buffer A in a Kontes-Dual grinder at 400 rev./min for 5 min. The homogenate was centrifuged at 205000xg for 2 h and the resulting supernatant was the soluble extract, which contained all the poly(A) polymerase activity present in homogenates. Several approaches have been used to investigate the existence of isoenzymes of poly(A) polymerase. Poly(A) polymerase activity from nauplii soluble extracts shows a monophasic curve of thermal inactivation and a unique and symmetrical curve of dependence on pH, with a maximum at p H 8.5. Fig. 2 shows the column electrofocusing of a nauplii soluble extract. Poly(A) polymerase is recovered in a single peak with an isoelectric point of 6. Fig.3A shows the chromatography of nauplii extracts on DEAE-cellulose. Only one peak of poly(A) polymerase is eluted at 80 m M KCI. N o other peaks of enzymatic activity were eluted from the column at concentrations of KCI up to 1 M. Fig. 3 B shows the rechromatography on Bio-Rex 70 of the fraction containing poly(A) polymerase activity from the DEAE-cellulose column. Again only one peak of enzymatic activity is eluted at 0.2 M KCI. The recovery of both columns was close to 90 %,. These results strongly support the existence of only one form of poly(A) polymerase in Artemia nauplii. In order to support this conclusion further, the poly(A) polymerase activities from nuclear and cytosolic fractions have been partially purified and the enzymatic properties of both enzyme preparations have been compared. 50 g newborn nauplii were homogenized in 2.5 vol. buffer C and the cytosolic and nuclear fractions were obtained. The nuclear fraction was resuspended in 50ml buffer A and homogenized in order to solubilize the poly(A) polymerase activity. Poly(A) polymerase from both fractions was partially purified following the method described for the enzyme from dormant embryos [2]. The enzyme purified from the nuclear fraction had a specific activity of 110 units/mg protein and the enzyme purified from the cytosolic fraction of 302 units/mg protein. The activities of the nuclear and cytosolic enzymes are fully dependent on the presence of a divalent cation in the

72

80

1

lA

0.5

., .t

100

1

-"

c

a ' 0

e

20

10

60 Y

D

0

LO

-

x Q 0

0

0

20

LO 60 Fraction number

Fig. 3. DEAE-cellulose(A) and Bio-Rex 7 0 ( B ) chromatography of poly(A) polynzerase from Artemia nauplii. 3 g newly hatched nauplii were homogenized with 2.5 vol. buffer A containing 50 pg soybean trypsin inhibitor/ml. The homogenate was centrifuged at 10000 x g for 30 min and the supernatant was centrifuged again at 105000xg for 120 min to obtain the soluble extract. The soluble extract was dialyzed against buffer B containing 10 mM KCI and was subjected to chromatography on a 20-ml column of DEAE-cellulose, equilibrated with buffer B containing 1 0 m M KCI. After loading the sample the column was washed with 20 ml buffer B plus 10 mM KCl and eluted with a linear gradient of 50 ml x 2 buffer B containing 10 mM and 0.5 M KCI. Fractions with poly(A) polymerase activity from the DEAE-cellulose column were pooled, diluted with 1 vol. buffer B and loaded on a 20-ml Bio-Rex 70 column, equilibrated with buffer B containing 50 mM KCI. The column was washed with 20 ml of the same buffer eluted with a linear gradient of50 ml x 2 buffer B containing 50 mM and 0.5 M KCI

assay. Both enzymatic preparations have maximal activities with manganese and about 10% and 6 % respectively with magnesium. Both enzymes are highly specific for ATP, although they can also use dATP, as is the case for the poly(A) polymerase purified from dormant embryos [2]. Michaelis constants for ATP are 0.06 mM and 0.04 mM for the nuclear and soluble enzyme, respectively. Both enzymes were inhibited by ATP at concentrations higher than 0.5 mM. The inhibition con$ant for ATP using Torula RNA as primer was 1 mM for both enzyme preparations. The study of the RNA primer specificity showed that Torula RNA was the best RNA primer for both enzymes, followed by poly(A). The comparison of the enzymatic properties of poly(A) polymerase from nuclear and cytosolic fractions from Artemia nauplii shows very few differences. Moreover, these properties are similar to those found for the enzyme from dormant embryos [2]. Therefore, it can be concluded that embryos and nauplii contain only one isoenzyme of poly(A) polymerase which is present both in nuclei and cytosol. L a c k of association of poly(Aj polymerase with cytoplasmic polyadenylated m R N P particles

One feature that can be relevant to the question of the physiological role of cytoplasmic poly(A) polymerase in the

0

5

10 15 20 25 Fraction number

30

Fig. 4. Isopycnic sucrose gradient centrqugatiorr of poly(A) f RNP particles and p o l ~ , ( Apolymeruse ) activity from the postmitochondrial supernatant of Artemia embryos. 2 g dormant embryos were homogenized with buffer C and the postmitochondrial supernatant was obtained. 3 ml of this fraction were loaded in a sucrose gradient made with three layers of 94%, 62% and 34"/, ( w k ) sucrose in buffer containing 30 mM Tris/HC1 pH 7.6, 0.1 M KCI and 5 mM MgC12. The gradient was centrifuged in a SW40Ti rotor at 35 000 rev./min for 90 h at 4°C. Fractions of 0.4 ml were collected. Poly(A) polymerase activity was assayed and nucleic acids extracted for poly(A) determination by hybridization with [3H]poly(U) as described in Materials and Methods

activation of stored mRNA is its possible association with cytoplasmic mRNP particles in dormant embryos. The postmitochondria1 fraction was obtained after homogenization of the embryos and it was subjected to isopycnic centrifugation in a sucrose gradient. Free cytoplasmic polyadenylated mRNP particles were identified in the fractions from the gradient using poly(A) tracts as marker. The fractions were also assayed for poly(A) polymerase activity. Fig. 4. shows the results of this experiment. There is a peak of poly(A) sedimenting at a density of 1.29 g/cm3, corresponding to the sedimentation of Artemia free cytoplasmic mRNP particles [17]. The bulk of poly(A) polymerase is in fractions sedimenting at lower density. Therefore, cytoplasmic poly(A) polymerase is not associated with cytoplasmic polyadenylated mRNP particles in the embryos. The enzyme does not seem to be a protein component of the particles.

DISCUSSION The results reported in this paper show the existence of an increase of poly(A) during early Artemirr development. This increase starts soon after resumption of development, since the levels of poly(A) double during the first 5 h of postgastrular development. Several processes can be involved in the increase of poly(A) during Artemiu embryogenesis, including polyadenylation of non-polyadenylated stored mRNA. Sierra et al. [29] have reported that half of the mRNA stored in dormant embryos lack poly(A) tails. Elongation of

73

the poly(A) chain of polyadenylated m R N A with short poly(A) tails can also account for the increase in total poly(A). In this respect the length of the poly(A) chain of mRNA from free cytoplasmic m R N P particles is very heterogeneous in Artemiu cysts, ranging from 12 to 138 nucleotides with a mean value of 75 nucleotides [16,30]. The third process, which can contribute to an increase of total poly(A), is the activation of transcription, which takes place early after rehydration of dormant embryos [31]. James et al. [I91 have demonstrated an increase in the complexity of Artenziu m R N A populations during postgastrular embryogenesis as a result of the synthesis of new mRNA sequences. The impermeability of the encysted embryos to radioactive precursors or inhibitors of RNA synthesis [32] makes it difficult to study the contribution of each of these processes to the increase of poly(A) in Artemiu. The results reported in this paper show the existence of an increase of poly(A) polymerase activity during postgastrular embryonic development. This increase of enzymatic activity can be correlated with the poly(A) increase that occurs in the nuclear and postnuclear fractions in Artemiu during the same developmental period (Table 1). An increase of poly(A) levels during early development has also been reported in several eukaryotic systems [12,13,15,33-381. The changes in total poly(A) polymerase are higher in the nuclear fraction, where the enzyme activity increases fourfold during embryonic development while the activity in cytosol increases less than twofold. Therefore, besides a net increase in activity there is a change in the subcellular distribution of poly(A) polymerase. The majority of the activity is cytosolic in dormant embryos, whereas the enzyme activity is equally distributed between cytosol and nuclear fractions in newly hatched nauplii. A change in the subcellular distribution of poly(A) polymerase also occurs in sea-urchin eggs, where the enzyme is primarily in the cytosol of the egg and is progressively transferred to the nucleus during embryogenesis [14,39]. After hatching there is a decrease of poly( A) polymerase activity reaching lower levels during late larval development than those found in dormant embryos. This decrease is not due to experimental conditions, since the enzymatic activities were determined in fed nauplii and the extracts were prepared in the presence of soybean trypsin inhibitor to prevent proteolysis by Artrmiu larval protease [27]. Moreover. extracts prepared with a mixture of newly hatched and late nauplii yield the expected levels of poly(A) polymerase activity. The development changes of total poly(A) polymerase activity in Artemiu are similar to those reported for RNA polymerase I, which also have an increase during embryogenesis followed by a decrease after hatching [22,23]. The presence of poly(A) polymerase activity in the nuclear and the cytoplasmic fractions of embryos and nauplii raises the question of the existence of different isoenzymes in Ar/clmia. The existence of poly(A) polymerase activities both in nuclei and cytosol has also been described in other eukaryotic systems [I ,281. However, the direrent nature of cytosolic and nuclear poly(A) polymerases has recently been challenged. Some of the multiple forms of poly(A) polymerase, obtained by ion-exchange chromatography, could be due to different stages of phosphorylation of the enzyme [40]. In the case of Artemiu, only one peak of enzymatic activity is obtained after chromatography of nauplii extracts on DEAEcellulose and Bio-Rex 70 and after eleclrofocusing. Studies on the pH dependence and thermal stability of poly(A) polymerases in homogenates also support the existence of one form of the enzyme. Besides, the properties of poly(A) polymerase, partially purified from nuclear and cytosolic

fractions, were the same for both enzyme preparations and similar to those of the enzyme previously characterized from the soluble fraction of dormant embryos [2]. Therefore, the results presented in this paper support the fact that Arterniu nauplii contain only one form of poly(A) polymerase, which is located both in the nuclear and the cytosolic fractions. The subcellular distribution of poly(A) polymerase suggests a different biological function for the enzyme in the nucleus and the cytoplasm of cells. Nuclear poly(A) polymerase can account for the polyadenylation of newly transcribed mRNA sequences. In this respect the increase of poly(A) polymerase in the nuclear fraction correlates with the fourfold to fivefold increase of poly(A) levels that occurs in nucleii after resumption of Artemiu development. A possible function for cytoplasmic poly(A) polymerase is the polyadenylation of non-polyadenylated mRNA or the elongation of the poly(A) chain of polyadenylated mRNA stored in Artemia. Artemiu embryos contain stored niRNA associated with proteins in ribonucleoprotein particles, which are in the cytoplasm [16,17,41]. Artemiu m R N P particles are substrates for poly(A) polymerase in vitro and the affinity of the enzyme for the particles is two orders of magnitude higher than for naked m R N A [42]. The fact that m R N P particles are good substrates for poly(A) polymerase and that a net increase in the amount of poly(A) has been observed concomitant with the activation of translation in many developing systems [ 12,15,33 - 381 suggest that cytoplasmic polyadenylation can be involved in the activation of stored mRNA. However, our results show that the bulk of poly(A) polymerase activity is not associated with m R N P particles after isopycnic sucrose gradient centrifugation of the postmitochondrial fraction. Therefore, poly(A) polymerase is not an intrinsic protein component of cytoplasmic m R N P particles from Artemiu dormant embryos. These results are in agreement with the lack of poly(A) polymerase activity assayed in m R N P particles purified by oligo(dT)-cellulose chromatography [30,42]. The fact that poly(A) polymcrase is not associated with cytoplasmic mRNP particles does not disprove the hypothesis of the role of cytoplasmic polyadenylation in the activation of stored mRNA, since a transient association between poly(A) polymerase and m R N P particles could take place only during the enzymatic polyadenylation of the particles. We are grateful to D r J. Renart for critical reading the manuscript and to Elvira Dominguez and Amalia Montes for technical assistance. This work was supported by a grant from Fondu Ncicional para el Des~~rrollo de la Investigacidn CientifTfi'c.a. L. S. has a fellowship from Cuju dr Ahorros dr Mudrid.

REFERENCES 1 . Ednionds, M. (1982) i n Tlie Eniymi>s(Boyer, P. D., ed.) vol. XV part B, pp. 21 7 - 244, Academic Press, New Y ork. 2. Sastre, L. & Sebastian, J. (1981) Biochirn. BiopA>,s. Actn. 661, 54-62. 3. Darnell, .I.F. (1982) Nuture (Land.) 297, 365-371. 4. Brawerman, G. & Dicz, J . (1975) Crll, 5, 271 -280. 5 . Ajtkozhin, M. A , , Doschanov, Kh. J . & Akhanov, A. V. (1976) FEBS Lett. 6 6 , 124- 126. 6. Jenkins, N. A., Kaumeyer, J . F., Young, E. M. & Raff. R. A. (1978) Drv. Biol.6 3 , 279-298. 7. Humphreys, T. (1971) Dev. Biol.26, 201-208. 8. Woodland, H. R . (1974) Dcv. Biol. 40, 90-102. 9. Ilan. J. & Ilan, J . (1978) Drv.B i d . 6 6 , 375-3385.

14 10. Mermod, J. J., Schatz, G. & Crippa, M. (1980) Dev. Biol. 75, 177 - 186. 11. Wilt, F. H. (1977) Cell, 11, 673-681. 12. Sagata, N., Shiokawa, K. & Yamana, K. (1980) Dev. Biol. 77, 431 -448. 13. Clegg, K. B. & Pik6, L. (1982) Nature (Lord.) 295, 342-345. 14. Slater, D. W., Slater, I. & Bollum, F. J. (1978) Dev. Biol. 63, 94-110. 15. Wilt, F. H. (1973) Proc. Natl Acad. Sci. U S A , 70, 2345-2349. 16. Nilsson, M. 0. & Hultin, T. (1975) FEBS Lett. 52, 269-272. 17. Slegers, H. & Kondo, M. (1977) Nucleic Acid Res. 4, 625-639. 18. Amaldi, P. P., Felicetti, L. & Campioni, N. (1977) Dev. Biol. 59, 49-61. 19. James, T. C., Hentschel, C. C. & Tata, J. R. (1979) in BiochemistryofArtemia Development (Bagshaw, J. C . &Warner, A. H., eds) pp. 167 - 189, University Microfilms International, Ann Harbor. 20. Felicetti, L., Pierandrei, P., Moretti, S., Campioni, N . & Urbani, C. (1975) Cell Differ. 4, 339 -354. 21. Sastre, L. & Sebastian, J . (1982) Biol. Cell. 45, 160. 22. Renart, J. & Sebastihn, J. (1976) Cell. Differ. 5, 97-107. 23. Osuna, C. & Sebastian, J. (1980) Eur. J . Biochem. 109,383 -389. 24. Vallejo, C. G., Perona, R., Garesse, R. & Marco, R. (1981) Cell. Differ. 10, 343 356. 25. Nilsson, M. 0. & Hultin, T. (1982) Biochim. Biophys. Acta, 696, 253 - 259. 26. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. J . Biol. Chem. 193, 265 -275. -

27. Osuna, C., Olalla, A,, Sillero, A., Gunther-Sillero, M. A. & Sebastian, J. (1977) Dev. Biol. 61, 94- 103. 28. Edmonds, M. & Winters, M. A. (1976) Prog. Nucleic Acid Res. Mol. Biol. 17, 149-179. 29. Sierra, J. M., Filipowicz, W. & Ochoa, S. (1976) Biochem. Biophys. Res. Commun. 69, 181 - 189. 30. De Herdt, E., Piot, E., Kondo, M. & Slegers, H. (1982) Eur. J . Biochem. 122,453-460. 31. McClean, D. K. &Warner, A. H. (1971) Dev. Biol. 24, 88-105. 32. Finamore, F. J. & Clegg, J. S. (1969) in Thp Cell C y d e (Padilla, G. M., Whitson, G. L. & Cameron, I. L., eds) pp. 249-278. Academic Press, New York. 33. Slater, D. W., Slater, I. & Gillespie, D. (1972) Nature (Lond.) 240, 333 - 337. 34. Lamb, M. M. & Laird, C. D. (1976) Dev. Biol. 52, 31 -42. 35. Pawlowski, P. J. & Rodriguez, V. (1978) Dev. Biol. 40, 71 -73. 36. Harris, B. & Dure, L. (1978) Biochemi.ytr.v, 17, 3250-3256. 37. Hecht, R. M., Gossett, L. A. & Jeffery, W R. (1981) Dev. Biol. 83, 374- 379. 38. Piko, L. & Clegg, K. B. (1982) Dev. B i d . 89, 362-378. 39. Egrie, J. & Wilt, F. (1979) Biochemistry, 18, 269-274. 40. Rose, K. M. &Jacob, S. T. (1979) J . Biol Chem. 254, 1025610262. 41. De Herdt, E., Slegers, H. & Kondo, M. (1979) Eur. J . Biochem. 96,423 430. 42. Sastre, L. & Sebastian, J. (1982) Biochem. Biophys. Res. Commun. 108, 1609-1613. -

J. Sebastian, Instituto de Enzimologia y Patologia Molecular del Consejo Superior de Investigaciones Cientificas, Facultad de Medicina de la Universidad Autonoma de Madrid, Arzobispo Morcillo sjn, Madrid-34, Spain L. Sastre, Sidney Farber Cancer Institute, 44 Binney Street, Boston, Massachusetts, USA 021 15