Developmental expression of Cu,Zn superoxide dismutase in Xenopus. Constant level of the enzyme in oogenesis and embryogenesis

Eur. J. Biochem. 186, 421 -426 (1989) FEBS 1Y89

(c;

Developmental expression of Cu,Zn superoxide dismutase in Xenopus Constant level of the enzyme in oogenesis and embryogenesis Luisa MONTbSANO, Maria Tercsd CARRI, Pdolo MARIOTTINI, Frdncesco AMALDI and Giuseppe ROTILIO Dipartimento di Biologia, 11 Universiti di Roma ‘Tor Vergdta’, Italy (Reccned February 9iJune 13, 3989) - EJB 89 0162

cDNA clones for Xenopus Iuevis Cu,Zn-superoxide dismutase were isolated, sequenced and used as probes lo study the expression of the corresponding gene during oogenesis and embryogenesis ; Cu,Zn-superoxide dismutase activity was also monitorcd throughout development. It has been observed that its mRNA is actively synthesized during early oogenesis, reaching a maximum level at stage 11, and is utilized through oogenesis. This results in an accumulation of enzyme activity during oocyte growth, paralleling the accumulation of the several other cellular components which are stored in the oocytc to bc utilized later on by the developing embryo. In fact, Cu,Znsuperoxide dismutase iictivity is present at an approximately constant level until late embryonic development, while its mRNA disappears soon after fertilization to be accumulated again only during the last part of embryogenesis. This developmental expression behaviour can be viewed as typical of an housekeeping function and suggcsts that Cu,Zn-superoxide dismutase activity is a constant need of the cell rather than being subject to regulation by oxygen metabolism. Superoxide dismutases (SOD ; superoxide : superoxide oxidoreductase) are enzymes that catalyze dismutation of superoxide radicals (0; j into oxygen and hydrogen peroxide in an extremely efficient fashion [I]. This reaction is considered to play a central role in the protection of aerobic cells against oxygen toxicity; in fact 0; - can initiate and propagate oxygen-dependent free-radical chain reactions which result in cell damage and death. With few exceptions, eukaryotic cells have a manganese-containing enzyme (Mn-SOD), which is mainly mitochondria-associated in most species and is resistant to inhibition by cyanide, and an ubiquitous predominant copper/ zinc-containing enzyme (Cu,Zn-SOD) which is CN- sensitive. While recent studies have reported on molecular aspects concerning the biosynthesis of Mn-SOD in eukaryotes [ 2 , 31, not much is known about the control ofCu,Zn-SOD biosynthesis. However, the housekeeping nature of Cu,Zn-SOD activity is suggested by the ubiquitous presence of the protein at reasonably comparable levels, irrespective of the species [4] or tissue [5] examined. Knowledge of enzyme expression in early embryonic stages would be extremely important in this respect. Therefore, it seemed interesting to start investigating Cu,ZnSOD in Xenopus luevis, an amphibian species which provides a useful model for the study of gene activity during development. The presence of Cu,Zn-SOD has been detected by gel electrophoresis in the cytosolic fraction of X.lurvis heart and kidney homogenates [6] but no further characterization has been reported. Moreover, although the amino acid sequence

has been determined for Cu,Zn-SOD of several species [l], no data for amphibians are available in the literature except for a very partial sequence from Runa catesbeiuna [7]. Here we describe the isolation of cDNA clones for X . laevis Cu,Zn-SOD. The nucleotide and deduced amino acid sequences are given for the cDNA corresponding to one of the two Cu,Zn-SOD gene copies present in the genome of X . laevis. A developmental analysis of Cu,Zn-SOD gene expression was carried out by following both the Cu,Zn-SOD mRNA level, using the cDNA clone as probe, and the enzyme activity during oogenesis and embryogenesis. The results obtained suggest that the mRNA level is regulated in order to maintain the specific activity of the enzyme constant.

Correspondence fo Ci. Kotilio. Dipartimento di Biologia, I1 Universiti di Roma ‘Tor Vergata’, via Carnevale. 1-00173 Roma, Italy Ahhveviarion. SOD. superoxide dismulase. Ennzyne. Superoxide dismutase (BC 1.15.1 .I). Note. The novel nuclcotidc sequence data published hcrc has been deposited with the EMBL sequence data banks and is available under accession number X16585. The novel amino acid sequence dala published here has been deposited with the EMBL sequence data bank.

Screening of a c D N A library und subcloning

~

MATERIALS AND METHODS Biological materials

X . luevis adults were purchased from the African Xenopus Facility, Noordhoek, South Africa. Ovaries were disaggregated with 0.5 mgiml collagenase and 1 mg/ml papain at room temperature for 1 h; oocytes were then extensively washed in modified Barth solution 181 and staged according to Duniont [9]. Embryos were obtained by in vitro fertilization as described by Brown and Littna [lo], grown at 22°C in dechlorinated water and staged according to Nieuwkoop and Faber [l 11.

A cDNA bank, constructed in lgtlO with poly(Aj-rich RNA from X.luevis oocytes by D. Melton and kindly made available to us, was screened by plaque hybridization. A human cDNA clone, kindly provided by Y. Groner, which contains a 620-bp PstI insert including 459 bp coding for hSOD-I [12] was used as probe. UNAs from positive recombinant phages were digested with EcoRI and run on agarose gels.

422 EcoRI inserts of the cxpectcd length were purified and subcloned in pSP65. D N A sequence analysis Determination of the cDNA sequences was performed according to the chemical modification method of Maxam and Gilbert [13] with the addition of a T-specific reaction [14]. Fragments were end-labelled with T4 polynuclcotide kinase and strand separated. The chemical reaction products were electrophoresed on urea/polyacrylamide (30 : 1) gels. Sequences were analyzed using the IBI/Pustell package software. Cellfractionation and RNA preparation All glassware and solutions were sterile. KNA was extracted essentially according to Probst et al. [15]: pools of 10 oocytes or 10 embryos for each stage were homogenized in 10 mM EDTA, 0.3 M NaCI, 2% SDS, 100 mM Tris/HCI (pH 7.3, and incubated for 30 min at room temperature in the same buffer plus 1 mg/ml proteinase K. Samples were subsequently extracted twice with phenol and twice with phenoljchloroform, then ethanol-precipitated. The amount of total RNAjoocyte or embryo was determined by spectrophotometric measurement at 260 nm. Northern blot and dot blot of R N A Dry RNA precipitates were dissolved in buffer A (20 mM Mops, 5 mM sodium acetate, 1 mM EDTA and 2.2 M formaldehyde, pH 7) plus 50% formamide, heated at 60 "C for 15 min, ice-cooled and mixed with 1/10 vol. 50% glycerol, 1 mM EDTA, 0.2% bromophenol blue, prior to loading on the gel. Gels were made 1.2% (massjvol.) agarosc in buffer A and run for about 4 h at 100 V. At the end of the run, gels were soaked for a few minutes in 20 x SSC (3 M NaC1,0.3 M sodium citrate) and RNA was transferred to nitrocellulose filters as described by Southern [16] overnight in 10 x SSC. Dry filters were baked for 2 h at 80°C. For dot blot analysis, RNA was spotted on nylon filters in a minifold device according to Chien and Dawid [17]. Preparation of probes and hybridization The purified inserts of recombinant plasmids were radioactively labeled by nick translation [18] or, alternatively, by the random oligonucleotide primer mcthod [I 91. Hybridization to phage DNA for plaque screening was carried out according to Maniatis et al. [18], while prehybridization and hybridization of Northern and dot blots were performed in 50% formamide as previously described [20]. Quantification of autoradiograms was carried out using a LKB Ultroscan XL laser densitometer. Polysomelm R N P distribution of m R N A Cytoplasmic extracts from 30-60 oocytes of each stage were prepared, loaded on 15 - 50% sucrose gradients, centrifuged in a Beckman ultracentrifuge with a SW41 rotor for 100 min at 37000 rpm as previously described [20]. A260was monitored while collecting gradient fractions; these were precipitated with 3 vol. ethanol overnight at -2O'C, and total RNA was extracted from precipitates with proteinase K/ phenol/chlorophorm [15]. Amounts of RNA equivalent to 10 oocytes were analyzed by Northern blot hybridization.

Protein preparation and Cu,Zn-SOD uctivitv dptemrination Pools of 10 oocytes or embryos for each stage were gently homogenized in 25 mM phosphatc buffer (pH 7.8) and centrifuged for 30 min at 12000 x g at 4'C. Total protcin content of the postmitochondrial supernatants were measured as dcscribed by Lowry et al. [21]. Samples were made 50 mM 'Tris/ HCI @H 6.8), 10% glycerol and 0.04% bromophenol blue, and aliquots equivalent to one oocyte or embryo were sub.jected to discontinuous polyacrylamide gel electrophoresis undcr non-denaturing conditions [22].The gels were stained for SOD activity by the method of Beauchamp and Fridovich [23] in the presence and absence of 1 mM KCN to discriminate Cu,Zn-SOD and Mn-SOD activities [24]. Densitometric quantifications were performed with a LKB Ultroscan X L laser densitometer.

RESULTS Cloning and sequencing XlSODcD N A A X . laevis AgtlO cDNA library was screened using as a probe a human SOD cDNA (see Materials and Methods). 64 out of 10000 recombinant phages gave a positive signal by plaque hybridiLation. Four different recombinant phages (AXISODcDNA.1 to iXlSODcDNA.4) were chosen for subcloning in pSP65 and further analysis. Three of them (l.XlSODcDNA.1 to iXlSODcDNA.3) have similar EcoRI inserts, about 640 bp long and have been sequenced. Fig. 1 shows the nucleotide and derived amino acid sequences for pXlSODcDNA.2. These three cDNAs contain the complete coding region for Cu,Zn-SOD except that none of them has the ATG start codon; they all begin with a GTG codon for a valine at position 2, according to the amino acid alignement with human SOD1. pXlSODcDNA.1 differs from the other two by a single nucleotide insertion after position 405, which puts the code reading out of frame. The deduced amino acid sequence is in agreement with the sequence of the Cu,ZnSOD purified from cytosolic extracts of liver and oocytes (unpublished results). The fourth recombinant clone, pXlSODcDNA.4, has a quite different EcoRI insert containing only 241 bp of SOD sequence. Analysis of this cDNA fragment (not shown) suggests that it contains an incomplete coding sequence originating from a different copy of the Cu,Zn-SOD gene. Cu,Zn-SOD activity and S O D - m R N A uccumulation puttern during X. laevis oogenesis The availability of a cDNA for X . laevis Cu,Zn-SOD. allowed a developmental study of the SOD gene expression. Since the purpose of the analysis was to obtain quantitative evaluations of the transcripts, we avoided purifying poly(A)rich RNA by oligo(dT)-cellulose columns which could cause uncontrolled losses of material. Therefore, total RNA, corresponding to a fixed number of oocytes of different stages, was analyzed by Northern or dot blot hybridization. Northern analysis (see, for example, insert of Fig. 2) shows a single band corresponding to an RNA of about 0.8 kb. The autoradiograms obtained in several experiments on different batches of oocytes were quantified by dcnsitoinetric scanning and average values are given in Fig. 2. As a control, we determined the amount of total RNA in each sample which rcflects the well-known accumulation of ribosomes during oogenesis. Moreover the same filters were subsequently rehybridized (not

423 , ,,

GTG

AAG GCG GTG TGT GTG CTG GCG GGA AGC GGC GAT GTA M G GGT GTG GTG CGC TTC

' i a l Lys 4 1 2 Val Cys Val Leu .&la G:y

GAG CAA CAG GAC S I T FGT GAC GTG ACA GTG GAP GGG M G

96

Ser G l y A s p Val iys ily Val Val Arz ? h e Glu Gln 'Glp ASD Asp Gly Asp 'Jal Thr Val G l u Gly LYS

ATC GAG GGT CTG ACT GAC GGA M T CAC GGC TTC CAT ATC CAC GTG T l l &G GAC AAC ACT MT GGC TGT CTC AGC GCC GGC CCT CGT TK PAC CCG CAG

195

I l r G l u Gly Leu T5r Asp G l y Asn 31s Gly ? h e H r s ILe H i s Val Phe Gly Asp Asn T h r Asn Gly Cys Leu Ser Ala G l y Pro 315 P h e Asn P r o G l n

MCM G MT CAC GGA TCC CCC PM GAT GCC GAC AGG CAC GTT GGG GAT CTT GGC M T GTG ACG GCT GM GGT GGT GTT GCT CAG TTC AnA TTC ACA GAC As"

L y s Asn His Gly S e r

Pro Ljs Asp Ala Ssp Arg His Val Gly

Asp

CCT CAG ATC TCC CTG APA GGC GAG CGC TCG ATC A n GGC CGT ACT GCT GTG GTC CAT GAG M G CAA GAT GAC TTG GGC PM GGT GGA GAT GAC GAG AGT P r o Gln I l e Se- Leu Lys Gly GLu Arg Ser Ile Ile n l y A r p Thr 41a Yal 'la1 H i s G ~ Lys J :In

CTG AAA

An GGC M C GCC GGG GGT CGC 7TG GCA TGT GGT GTG A

294

Leu 1 l y Asn Val T h r Ala Glu G l y Gly Vai A?a G l n P h e Lys P h e T h r Asp

593

Asp A s p Leu 3Ly Lys Glq G l y Asp Asp G111 Ser

T GGA TTC TGC CCG TGA CCATGTCCCGCACGTGATGCCCCCAAGCTCTTAGmATGTCMTTGCTC

503

Leu Ly5 Thr ily Asn Ala Gly Gly Arg Leu Ala Cys Gly 'Val I l e gly Phe Cys ? r o

T C C T T P A ~ M T T C T C A T C T G C A G ~ G C m C T C T C T C T C G T M C G A T C T C C A ~ C T T G n G T G ~ C A T C C T A C A ~ ~ G G A C A C C C C C ~ G C,C. .T T C ~

GO8

Fig. 1. Niiclrotide sequence qf X. laevis Cu,Zn-SODcDNA (XISODcDNA.2) and its deduced urnino ucid sequence. The nucleotide residues are numbered in the 5' to 3' direction starting from the first nucleotide of the cloned cDNA fragment which, on the basis of a comparison with human Cu,Zn-SOD, lacks the ATG initiation and the second codon.

Po 1ysomes

mRNP

1 VI L- /&

I

11

Ill

I

I

I

1v

v

VI

Stages

Fig. 2. Accumuiution pattern of Cu,Zn-SOD m R N A during oogenesis. Total RNA from a fixed number of oocytes at different stages has been analy7ed. in four independent experiments, by Northern or dot blot hybridization t o a X . luevis CuJn-SOD probe. The extent or hybridization was quantified by densitometric scanning of the autoradiograms and the values exprcsscd relative to the highest one (stage I1 in all experiments). The averages of the values obtained in four experimentsare given (O-0 )together with standard errors. The insert shows a Northern blol experiment as an example. ToVal RNAloocytc at different stages is also shown ( A- - - A)

shown) with probes for unrelated proteins, such as ribosomal proteins, whose accumulation patterns are well established [20]. It appears from Fig. 2 that the amount of SOD-mRNA/ oocyte increases very rapidly at the beginning of oocyte growth, reaches its maximum level at stage I1 and decreases thereafter. In order to determine the proportion of Cu,Zn-SOD mRNA actually translated at different stages of oogenesis, we fractionated postmitochondrial supernatants of oocytes on sucrose gradients. R N A was extracted from gradient fractions and analysed by Northern blot hybridization. As shown in Fig. 3 about 70"/0of SOD mKNA is recruited onto polysomes at stagc I; this m R N A utilization is increased at later stages. 'lhe developmental study has also been extended at the level of Cu,Zn-SOD enzymatic activity, which in other animal systems has been shown to parallel SOD protein level [5]

Fig. 3. Polysome/mRNP distribution of Cu,Zn-SOD mRNA in ooc:ytes. Cytoplasmicextracts from oocytes of stages I, 11and VI were fractionated on sucrose gradients. Five fractions were collected from each gradient; the RNA was extracted and analyzed by Northern blot hybridization to a X . laevis Cu,Zn-SOD probe (and referenccs therein). For this purpose, aliquots of postmitochondrial supernatants corresponding to a fixed number of oocytes at different stages were run on polyacrylamide gels and stained for SOD activity. An example is shown in the insert of Fig. 4, where SOL) activity appears as a main electrophoretic band and three fainter bands. Different electrophoretic patterns are generally obtained with different animals, suggesting genetic polymorphism. The fact that mitochondria had been previously eliminated by centrifugation of oocyte homogenates ensures that all bands are due to Cu,ZnSOD activity. Accidental contamination by Mn-SOD due to mitochondria breakage is ruled out by the absence of bands when gel activity staining was carried out in the presence of cyanide (not shown). Quantification of enzyme activity has been performed by densitometry of the activity gels [25]. However, since the gelstaining assay is quantitative only within a limited range of activity, in some experiments different numbers of oocytes were used for the various stages, so as to obtain comparable levels of activity. The results of dcnsitometric analysis of several gels are reported in Fig. 4, together with the measured amount of postmitochondrial supernatant proteins/oocyte. It appears that Cu,Zn-SOD activity increases in parallel with the progressive accumulation of proteins which occurs during oogenesis.

424 1

,.. e, +J >

e, Y

2 1

0

10

E

-. .z 1 A

c

.-

w W

8

5

g

+J

U O

6

.5

P

+. 0

v)

I

23 32 3 8 4 2 46

a,

5 . 3

‘0

16

0 z L n

V

\ 0

0 0

8

LII 0

c 0

I

I1

111

IV

v

VI 8

Fig. 4. Cu,Zn-SOU uctivitj, during oogenesis. Postmitochondrial supernatants from oocytes of different stages were electrophorcsed on polyacrylamidc gels and stained for SOD activity. The bands were quantified by densitometric scanning and the values normalized relative to the hghcst one (stage VI). The averages of the values obtained in scveral experiments are given (0-0). The insert shows. as an example, an expcriment in which the material equivalent of one oocyte was loaded for each stage. Total postmitochondrial protein contcnt/oocyte at different stages is also given ( A - - - A)

1

8

16

23

32

38

0 A

A’

.\.

*/ 1

8

16 73

32

38

16 23

32

3 38

4 42

Days 46 Stages

Fig. 6 . Cu,Zn-SOD activity during embryogenesis. Postniitochondrial supernatants from embryos of different stages were electrophoresed on polyacrylamide gels and stained for SOD activity. The bands were quantified by densilometric scanning and thc values normalized relative to the last point. The averages of values obtained in several The insert shows, as an example, experiments are givcn (0-0). an experiment in which the material equivalent of one embryo was loaded for each stage. Total postmitochondrial protciii contentienibryo at different stages is also given (A - - - A)

42

-

I

2

1

Stages

42

46 Stages

Fig. 5. Accurniilutioriputtern of Cu,Zn-SOD m K N A during embryogenehis. Total RNA from a fixed number of embryos at different stages has been analyLed, in iive experiments, by Northern or dot blot hybridization to a X . laevis Cu.Zn-SOD probe. The extent of hybridization was quantified by dcnsitomctric scanning of thc autoradiograms and the values expressed relative to the last point. The values obtaincd in the five experiments were averaged when referring to the same stage (0-0). The insert shows a Northern blot experiment as an example. Total RNA/embryo at dillerent stages is also given ( A - - - A )

Cii.Zn-SOD activity and SOD-mRNA accumulation pattcrns during X. laevis embryogenesis Total RNA extracted from embryos at different developmental stages was analyzed by Northern or dot blot hybridization. as an example, one of the Northern gels is shown in the insert of Fig. 5. A single hybridization band is visible (ofabout 0.8 kb as in the oocytes), its intensity being clearly stagedependent. The data obtained in several experiments per-

formed with different batches of embryos are combined in Fig. 5. It appears that the small amount of SOD mRNA of maternal origin and present in the zygote, disappears during the early stages of embryonic development and that accumulation of new SOD mRNA is resumed only later in the embryogenesis, after stage 32. In some experiments, as a control that no losses of RNA had occurred, the amount of total RNAIembryo was also determined. As expected, the amount of total RNA (about 4 pg/embryo, mainly due to the maternal store of ribosomes) remains constant during the first part of embryogenesis and begins to increase after stage 26 to double at the end of embryogenesis. As a further control, the Northern blots were also rehybridized with ribosomal protein probes as already mentioned (not shown). Cu,Zn-SOD enzymatic activity during embryogenesis was followed as described above. Fig. 6 summarizes the quantification of several experimcnts, one of which is shown as an example in the insert (in this particular case SOD activity appears as two major bands). It can be seen that SOD activity remains more or less constant at the beginning of embryogenesis, to slightly increase later on. For comparison the progressive increase of postmitochondrial supernatant proteins is also shown in Fig. 6.

DISCUSSION

A’. luevis, which represents an animal model system particularly suited to analysis of gene expression and rcgulation during development [8], has never been used, to our knowledge, for similar studies on Cu,Zn-SOD. This prompted us to start a project in this line, in view of its relevance to the understanding of patterns and mechanisms of regulation of an enzymatic activity, which has been agreed upon to represenl a major factor of cell survival against oxygen toxicity. The isolation and nucleotide sequencing of cDNAs ior the Cu,Zn-

425 SOD of this amphibian species was the first step in this perspective. Since we present here the first complete sequence of an amphibian Cu.Zn-SOD, some structural considerations seem appropriate. The X . laevis SOD cDNA sequence shows 66% similarity with human SOD cDNA [26, 271. When compared at the amino acid level, the similarity between these two species is also 66%, suggesting that several regions of the protein are not subjected to very strong structural constrains during evolution. Well conserved are the critical amino acid residues involved in the binding of copper and zinc, as well as the glycine residues known to be essential to the formation of the charactcristic P-strand barrel structure of this protein. It is therefore likely that the three-dimensional folding of the Xenopzrs enzyme is similar to that already established for the bovine protein [28]. It is well known that a whole genome duplication took place in X. lclevis about 30 million years ago [29]. In fact most genes studied ilp to now in this species are present in two copiesihaploid genome (for references see [30]). Three out of the four SOD-cDNAs analyzed clearly derive from the same gene copy. On the contrary the fourth incomplete cDNA has several nucleotide substitutions and therefore derives from a second copy of the Cu,Zn-SOD gene. Work is in progress in order to charactcrizc the protein products of the two genes. The level of Cu,Zn-SOD has been followed during development by measuring the enzyme activity; in fact it has been shown in several animal systems that this reflects well the Cu,Zn-SOD concentration measured by immunological techniques [5] (and references therein). The analysis of Cu,ZnSOD cnzymatic activity through oogenesis and embryogenesis has revealed that the level of this enzyme approximately follows the developmental pattern of total cytosol proteins, in linc with the pattern expected for a housekeeping gene product. In particular, it is accumulated progressively during oocyte growth, in parallel with the accumulation of many other components (yolk, mitochondria, ribosomes, mKNAs etc.) which are stored to be utilized after fertilization during the first part of embryogenesis. According to this pattern, Cu,Zn-SOD mRNA is synthesized and accumulated to its maximum level early in the oogenesis (stage 11). Its utilization during oocyte growth, demonstrated by polysome/mRNP distribution analysis, results in a progressive accumulation of the enzyme throughout oogenesis. After fertilization and during the first part of embryogenesis, Cu,Zn-SOD activity remains constant in spite of the drop of its mRNA to a minimum level. In the absence of data on the turnover of Cu,Zn-SOD, these results may be considered as strong evidence for the metabolic stability of this enzyme. They are in line with the well-known molecular stability of Cu,Zn-SOD, which is known to be insensitive to the action of protcases, and with the delayed and only partial decay of the enzyme activity in copper-depleted animals with respect to other copper-containing enzymes [31]. At later stages the enzymatic activity increases slightly when the synthesis of new Cu,Zn-SOD mRNA is resumed in the embryo around stage 32. It is interesting to notice that a constant levcl of Cu,Zn-SOD activity is also maintained during those embryogenesis stages when important changes in oxygen consumption are known to occur. As a matter of fact, amphibian embryos live a day without oxygen and develop dependence on it at later stages [32] with a peak of oxygen consumption during neurulation (for a review see [33]).In Xenopus, on the other hand, Cu,Zn-SOD apparently does not follow a pattern of enzyme activity that matches the oxidative metabolism pattern. This conclusion may seem in contrast

with previous reports suggesting post-natal increase of this enzyme activity in rat hepatocytes [34] or perinatal increases in rabbit lung [35] as responses to the relatively hyperoxic extrauterine environment. However, it should be recalled that a perinatal surge has not been found in rat [36] or human [37] lung. Furthermore the level of the enzyme in rat fetal hepatocytes [34] is of the same order of magnitude (one third) with respect to adult cells, in contrast with thc other antioxidant enzyme, glutathione peroxidase, which is extremely low in the fetal cells (one tenth of the adult value). It was also found that rat Cu,Zn-SOD activity was nearly constant from the earliest examined fetal age (15 days) to birth, while the functionally related catalase activity increased linearly from a vcry low value at 15 days pregnancy to the 15 times higher value measured at birth. No substantial changes of Cu,ZnSOD activity were seen during development of the fruit fly Ceratitis capitata in contrast with the large increase of MnSOD at the end of the pupal stage 1381. In this context it is important to mention that expression of Cu,Zn-SOD in rat has been found to be nearly equivalent in all tissues [3] and not to be related to tissue oxygen tension [5]. On the other hand, Mn-SOD levels, which are generally found to be responsive to oxygen in bacteria [39] and mammals [40], havc been found to be very low in rat lung, muscle and testis and high in heart and brain, where a predominantly respiratory metabolism requires high oxygen tensions [3]. It can therefore be concluded that reports documenting variation in Cu,Zn-SOD level as a function of varying oxygen tension [41, 421 appear to deal with clearly extreme pathological or stress conditions, rather than with physiological regulation. From the results here presented it appears also that the maintainance of a proper level of Cu,Zn-SOD activity in Xenopus oogenesis and embryogenesis is attained for a large part by the regulation of mRNA accumulation, in agreement with data obtained in adult rats [ 5 ] , which showed that the slight tissue-specific variations of the Cu,Zn-SOD activity are accompanied by corresponding quantitative changes of the mRNA. However, it has not been established if the observed variations in the SOD-mRNA levels are due to differential transcriptional rates or different mRNA stability throughout development. We wish to thank Dr Y. Groner for kindly making available a

human SOD cDNA clone and Dr D. Melton for his Xenopus laevis cDNA library. This work was partially supported by Consiglio Nazionale delle Ricerche Special Project Biotecnologia e Bioslrumenluzioni.

REFERENCES 1 . Bannister, J. V.: Bannister. W. H. & Rotilio, G. (1987) CRC Crit. Rev. Biochem. 22, 1 11 - 180. 2. Autor, A. P. (1Y82) J . Biol. Chem. 257, 2713 -271%. 3. Ishikawa. T., Hunaiti, A. R., Piechot, G. &Wolf, B. (1987) Eur. J . Biochem. 170, 317-323. 4. Maral, J., Puget, K. & Michelson, A. M. (1977) Biochem. Biophys. Res. Commun. 77,1525-1535. 5. Delabar, J. M., Nicole, A,, D’Auriol, L., Jacob, Y., MeunierRotival, M., Gdlibcrt,F., Sinet,P. M. & Jerome, H. (19x7) Eur. J . Biochem. 166, 181 - 187. 6. Nzingula, N. V., Shome, E. & Picard, I. J. (1980) Arch. Int. Physiol. Biochim. 88, 292 -293. 7. Abc, Y., Okazaki, T., Shukuya, R. & Furuta. H. (1984) Comp. Biochem. Physiol. 77B, 125 - 130. 8. Gurdon, J. B. (1974) The control of gene expression in unimul

development, Oxford University Press, Oxford. 9. Dumont, J. N. (1972) J . Morphol. 136, 153-280.

10. Brown. D. D. & Liltna, E. (1964) J . Mol. Biol. 8, 669-687. 11. Nieuwkoop, P. D. & Faber, J. (1956) Normal table ofXenopus laevis (Daudin), North-Holland, Amsterdam. 12. Liemari-Hurwik, J., Dafni, N., Lavie, V. & Groner, Y. (1982) Proc. Natl Acad. Sci. USA 79,2808 - 281 1 . 13. Maxam, A. M. &Gilbert, W. (1980) Melhods Enzymol. 65,499560. 14. Rubin, C. M. & Schmid, C. W. (1980) Nucleic Acids Res. 8, 4613 -4619. 15. Probst, E., Kressmann. A. & Birnstiel, M. L. (1979) J. Mol. Biol. 135,709 - 732. 16. Southern, E. M. (1975) J . Mol. Biol. 98, 503-517. 17. Chien, Y. H. & Dawid, I. B. (1984) Mol. Cell, Biol. 4,507-535. 18. Maniatis, T., Fritch, E. F. & Sambrook, J. (1982) Molecular cloning: a laboratory manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. 19. Feinberg, A. P. & Vogelstein, B. (1984) Anal. Biochem. 132, 613. 20. Pierandrei-Amaldi, P., Beccari, E., Bozzoni, I. & Amaldi, F. (1985) Cell 42, 317-323. 21. Lowry, 0. H., Roscbrough. N. J., Farr, A. L. & Randall, R. J. (1951) J . Biol. Chem. 193,265-275. 22. Davis: R. J. (1964) Ann. IV. Y. AcadSci. 121, 404-427. 23. Beauchamp, C. & Fridovich, I. (1971) Anal. Biochem. 44,276287. 24. Beauchamp, C. & Fridovich, I. (1973) Biochim. Biophys. Acta 317, 50-66. 25. Mavelli, I.. Ciriolo, M. R.. Rossi, L., Meloni, T., Forteleoni, G., De Flora, A , , Benatti, U., Morelli, A. & Rotilio, G. (1984) Eur. J . Biochem. 139, 13- 18. 26. Sherman, L., Dafni, N., Lieman-Hurwitz, J. & Groner, Y. (1983) Proc. Natl Acad. Sci. U S A 80; 5465 - 5469. 27. Hallewell, R. A., Masiarz, F. R., Najarian, K. C., Puma, J. P., Quiroga, M. R., Randolph, A,, Sanchez-Pescador, K.,

28. 29. 30. 31. 32. 33. 34.

Scandella, C. J., Smith, B.. Sleiner. K. S. & Mullcnbach, G. T. (1985) Nucleic Acids Res. 13, 2017-2034. Tainer, J. A., Getzoff, E. D., Beem, K. M.: Richardson, J. S. & Richardson, D. C. (1982) J . Mol. €301. 160, 181 -217. Besbee, C. A., Baker, M. A., Wilson. A. C., Hadji-Azimi, I. & Fishberg. M. (1977) Science 195, 785-787. Mariottini, P., Bagni, C., Anncsi, F. & Amaldi, F. (1988) Gene 67,69 - 74. Bohnenkamp, W. & Weser, U. (1977) in Supcroxide und superoxide dismutases (Michelson, A. M., McCord, J. M. & Fridovich, I., eds) pp. 387 - 394, Acadcmic Prcss. London. Adolph, E. F. (1979) Am. J . Physiol. 236, 282-291. Brachet, J. & Alexandre, H. (1986) Introduction to mokculur embryology, Springer-Verlag, Bcrlin. Mavelli, I., Autuori, F., Dini, L.. Spinedi. A,, Ciriolo, M. R. & Rotilio, G. (1981) Biochem. Biophys. Rex Commun. 302. 911 916. Frank, I. & Groseclose, E. E. (1984) Pediatr. Res. 18, 240-244. Gerdin, E., Tyden, 0. & Eriksson, U. J. (1985) Pediatr. Kes. 19, 687 - 691. Strange, R. C., Cotton. W., Frycr, A. A., Drcw, R., Bradwcll, A. R., Marshall, T., Collins, M. F., Bell; J. & Hume, R . (1988) Riorhim. Riophys. Artu 964, 260 - 265. Fernandez Sousa, J. M. & Michclson, A. M. (1976) Biochem. Biophys. Kes. Commun. 73, 21 7 - 222. Gregory, E. M. (1985) Arch. Riochem. I3ioph.y.~.238, 83--89. Stevens, J. B. & Autor, A. P. (1977) J . Biol. Chem. 252, 35093514. Crapo, J. D. & McCord, J. M. (1 976) Am. J . Physiol. 231, 11961203. Hoffman, M.; Stevens, J. B. & Autor, A. P. (1980) Toxicology 16, 23 5 -225. -

35. 36. 37. 38. 39. 40. 41. 42.