Effect of transfer RNA from various sources on placental messenger RNA translation

Molecular and Cellular Endocrinology, 29 ( 1983) 18 1- 195 Elsevier Scientific Publishers Ireland, Ltd.

181

EFFECT OF TRANSFER RNA FROM VARIOUS PLACENTAL MESSENGER RNA TRANSLATION Susan J. KELLY Department Received

*, Jean LORIA,

of Reproductive

M.T. GYVES

Biology, Case Western Reserve

5 July 1982; accepted

12 October

SOURCES

ON

and Judith ILAN ** University School of Medicine.

1982

Poly(A+ )-containing mRNA from human term placenta was used to direct protein synthesis in a nuclease-treated rabbit reticulocyte lysate, which is dependent on mRNA and tRNA for maximal activity. The major protein product was human pre-placental lactogen (hPL). Addition of tRNA from rabbit liver, rabbit reticulocyte, human first trimester and term placenta, human liver and yeast resulted in 2-5-fold stimulation of [35S]methionine incorporation into total protein. Although all mammalian tRNA increased hPL synthesis, the relative synthesis as compared to endogenous globin was markedly different and most efficient with tRNA from term placenta. Addition of yeast tRNA increased total incorporation 3-fold but decreased incorporation of [35S]methionine into pre-hPL. These results suggest that the population of isoacceptor tRNAs may influence the expression of hPL in term placenta. Results are discussed by showing codon bias and usage of mRNA coding for hPL, LY-and P-hCG, rabbit globin and yeast alcohol dehydrogenase I. Keywords:

human

placental

lactogen;

cell-free translation;

codon usage

During placental development there is a major switch in gene expression. In first trimester placenta major gene products of the syncytiotrophoblast are (Y- and P-hCG while in term placenta a major gene product is hPL. It was reported that in a differentiating cell system such as calf lens (Virmaux et al., 1969) a direct correlation between the relative composition of aminoacyl-tRNA and the distribution of amino acids in

* Present address: Department of Zoology, University NC 27514 (U.S.A.). ** To whom correspondence should be addressed. 0303-7207/83/0000-0000/$03.00

0 Elsevier Scientific

of North

Publishers

Carolina,

Ireland,

Chapel

Ltd.

Hill,

182

S.J. Kelly et ul.

the crystallins exists. A similar correlation was found in the posterior part of the silk gland of Bombyx mori (i&Id et al., 1970, 1971; Chavancy et al., 1975). They found a linear correlation between the amino acid distribution of the fibroin and the corresponding tRNA acylated during the secretion phase. Therefore, it appears that there is a functional adaptation of the level of specific tRNA to the levels of amino acid incorporated into a major protein product. When chicks are treated with estrogen a major liver protein product is vite~logenin, one-third of which will split into phosvitin. The latter contains 50% serine residues. During estradiol induction there is a huge increase in serine tRNA in the liver, which is limited to two serine isoacceptors out of four. The amount of each is proportional to the frequency used in phosvitin mRNA translation (Bernfield and Maenpaa, 1971; Maenpaa and Bernfield, 197.5). In addition to specific synthesis and accumulation of message for a differentiated protein, the adjustment of major species of tRNA to codon utilization of major protein product may greatly influence the rate of specific mRNA translation. This will be manifested by the rate of appearance of the final gene product-the protein. Recently, large changes in the rate of polypeptide chain elongation, which might be affected by tRNA, have been reported in numerous eukaryotic systems influenced by developmental or hormonal stimuli. This was shown for the translation of specific proteins (Gehrke et al., 1981a; Roper and Wicks, 1978) and for the rate of elongation of total protein under hormonal and developmental stimuli (Brandis and Raff, 1978, 1979; Hille and Albers, 1979; Gehrke et al., 1981b). In order to study whether tRNA may influence the regulation of specific protein during placental development, we analyzed the effect of tRNA from various sources on hPL translation using a poly(A+) RNA template from term placenta. For this we used a nuclease-treated rabbit reticulocyte cell-free system. The above treatment resulted in a lysate which is partially dependent on mRNA as well as tRNA for translation. Our results indicate that added tRNA from various sources may affect the rate of synthesis of hPL and that this phenomenon is probably correlated to codon frequencies used in hPL and the corresponding major tRNA species in the source of tRNA used for the translational assay. It also points out that estimation of the amount of specific hPL mRNA as percentage of total mRNA by translational assay may be greatly influenced by the cell-free system used.

183

MATERIALS

AND METHODS

Chemicals were reagent grade. Glassware was baked overnight at 300°C. All solutions were treated with diethylpyrocarbonate, then heated to remove the excess reagent or were autoclaved prior to use (Bast et al., 1977). First trimester placentas were obtained immediately after therapeutic abortion. Normal term placentas were obtained within 10 min of delivery. Tissue was placed in cold saline. Transfer RNA was isolated from placentas according to the procedure of Roe (1975) using fresh tissue. Poly(A)-containing RNA was isolated from polysomes prepared according to Kelly et al. (1980) by chromatography on oligo(dT)cellulose (Collaborative Research, Inc., Waltham, MA) according to the procedure of Aviv and Leder (1972) as modified by Pemberton et al. (1975), substituting LiCl for NaCl. Translation of the poly(A)-containing RNA was carried out according to Patterson et al. (1977) in nuclease-treated reticulocyte lysate prepared according to Pelham and Jackson (1976). Standard assays of [35S]met~onine inco~oration into protein were carried out in a volume of 13 ~1 which included 30 pg/ml poly{A~ )mRNA and 29 FCi [35S]met~onine. Incubations were continued for 30 min or as otherwise indicated, but always within the linear range of incorporation (45 min), after which the samples were frozen at - 70°C. 1 ,ul samples were withdrawn at timed intervals during the incubation and spotted onto glass filter pads (Whatman GF/A, 2.4 cm) and dropped into a beaker containing 5% TCA in acetone for 10 min. The pads were then heated and washed 5 times in 5% TCA according to the procedure of Mans and Novelli (1961). Dried filters were counted in 5 ml Formula 963 scintillation fluid (New England Nuclear) in a Beckman LS-9000 liquid scintillation counter. For product analysis 30 ~1 of gel sample buffer (0.3 M Tris, 0.16 M H,PO,, 5% SDS, 50 mM DTT, pH 6.8) was added to incubation tubes containing 8 ~1 of translation mixture. These mixtures were heated in boiling H,O for 5 min. 10 ~1 ahquots were then applied to 12.5% SDS-slab gels and the proteins were separated by electrophoresis (Laemmli, 1970). The gels were fixed for at least 1 h at 45% (v/v) methanol, 9.2% (v/v) acetic acid and then stained for 1 h in the same solution containing 0.25% (w/v) Coomassie Brilliant Blue R (Sigma). Gels were destained in 5% (v/v) methanol, 7.5% acetic acid, then photographed and processed for fluorography (Bonner and Laskey, 1974).

S.J. Kelly et ai.

184

RESULTS The rabbit reticulocyte cell-free system incorporates amino acids into protein linearly for 45 min. Therefore all measurements were carried out for 30 mm and thus represent rates of incorporation. Treatment of the lysate with nuclease (Pelham and Jackson, 1976) partially destroyed the endogenous mRNA and tRNA, thus providing a model system to study the effect of tRNA on translation of pIacental message. Nuclease treatment did not alter the time period in which incorporation of amino acid into protein was linear. Poly(A’ ) placental mRNA stimulates [ ‘~S]met~onine inco~oration into protein with maximum stimulatio~l achieved at about 70 pg of RNA per ml (Fig. 1); since each point represents 30 min incubation, the curve in Fig. 1 represents rate differences. Therefore, in all subsequent incubations 30 pg/ml of poly(A’ ) RNA was used, which is about half of the concentration which gives linear response. Effects of added tRNA on total protein synthesis The effect of added yeast and rabbit liver tRNA on the incorporation of [3rS]methionine into protein is shown in Fig. 2. The rate of

f

,

I

I

loo

I

I

I

t

t

I

‘

I

200

,ug/ml of polyA - mRNA Fig. I. Stimulation of [35S]methionine incorporation into protein by poly(A’ ) RNA. Poly(A+ ) RNA was isolated from term placental polysomes by oligo(dT) cellulose chromatography and translated in the presence of [35S]metbionine at 30°C for 30 min. The samples were then analyzed for incorporation.

Placental mRNA

185

translation

50

100 [t-RNA]

150

(pg/ml)

Fig. 2. Effect of added tRNA on total incorporation of [‘sS]methionine into protein. Conditions for the cell-free system are the same as those described in Fig. 1 and in Materials and Methods, except for the concentration of term placental mRNA which was 30 pg/ml and for the concentration of tRNA as shown. 0, yeast tRNA; 0, rabbit liver tRNA; n , yeast tRNA added to non-nuclease-treated lysate. Similar results (no stimulation) were obtained with untreated lysate and rabbit liver tRNA.

[35S]methionine incorporation into protein was examined over a tRNA concentration range of 40- 160 yg/ml. Maximum rate of incorporation is about 3-fold higher than that of mRNA alone. Saturation occurred at about 100 pg/ml of tRNA. It is also shown that lysate which has not been nuclease-treated is not activated by addition of tRNA. Effects of added tRNA from various sources on the rate of p-e-hPL synthesis When term placental poly(A+) mRNA is added to nuclease-treated rabbit reticulocyte cell-free system and the [ 3sS]methionine-labeled translational product is analyzed by SDS-gel electrophoresis, two major translational products are visualized by fluorography. One is pre-hPL with an estimated molecular weight of 24600 and the second is globin (Kelly et al., 1980). This molecular weight is in agreement with the determination of pre-hPL by others who reported a range of 24 000-25 000 for the molecular weight of pre-hPL (Szczesna and Boime, 1976; Hubert et al., 1981; Shine et al., 1977; Sherwood et al., 1979). We identified the

S.J. Kelly

186

et al

Pre hPL

giobin

Fig. 3. Analysis of radioactive products of a nuclease-treated cell-free system directed by poly(A+ ) mRNA from term placental polysomes. Conditions for cell-free systems are as described in Materials and Methods and Fig. 2. Each reaction mixture contained 30 pg/ml of mRNA and rabbit liver tRNA as indicated. The labeled translation products were analyzed by 12.5% SDS-polyacrylamide slab gel electrophoresis and visualized by fluorography. It was underexposed to emphasize major translational products. The bands for pre-hPL and globin were cut out of the gel and counted as described in the text. A, control: B, 30 pg/ml term placenta mRNA. The migration of marker proteins was determined on the same gel and visualized by staining.

Placental mHNA

187

translation

Pre hPL

glabin

Fig. 4. Effect of term placenta tRNA on pre-hPL synthesis in a cell-free system directed by poly(A+ ) mRNA from term placenta. Experimental conditions are similar to those described in Fig. 3 and Materials and Methods except that term placenta tRNA was added as follows @g/ml): (1) 40 pg; (2) 80 pg; (3) 120 pg; (4) 160 pg. After radioautography the areas of the gel corresponding to pre-hPL and globin regions were cut, dissolved in H,O, and counted.

S.J. K&y et ul.

188

pre-hPL on the gel by immunoprecipitation. No other apparent labeled protein migrates to the same location on the gel. Moreover, when the gel is underexposed the fluorograph obtained shows mainly pre-hPL and globin bands. Such a fluorogram is shown in Fig. 3. In this experiment a nuclease-treated rabbit reticulocyte lysate was incubated for 30 min in the presence of 30 pg/ml of term placenta poly(A+) mRNA and increasing concentrations of rabbit liver tRNA. At the end of the incubation one aliquot was taken for gel electrophoresis and one for measuring total inco~oration of ~35S~methionine into protein. For an accurate estimate of counts incorporated into pre-hPL, the band was cut, dissolved in H20, and counted. This was compared to total incorporation. The effect of term placenta CRNA on pre-hPL synthesis is depicted in Fig. 4. It is apparent that the rate of pre-hPL synthesis is much higher than that of globin. The amount of pre-hPL and globin was quantitated

0 0

I

50

1

8

100

150

TRANSFERRNA (!Jg/d)

Fig. 5. Effects of yeast tRNA on [3SS]methionine incorporation into hPL and total protein in n&ease-treated lysate directed by term placental mRNA. Conditions for incorporation and analysis are similar to those described in Fig. 3. The results are expressed as relative incorporation. Without added yeast tRNA the relative value was considered as 1. Total incorporation (0) was determined from aliquots taken from the incubation medium at the end of the incubation. Pre-hPL (0) was determined by cutting and processing the hPL band from the gel. All incubations were carried out at 30°C for 30 min in the presence of 30 f.~g/ml term placental mRNA.

by cutting the gel bands as described above. The results revealed that even the smallest concentration of term placenta tRNA gave a huge increase in the rate of hPL synthesis as compared to globin when poIy(A*) mRNA from term placenta was added to the cell-free system. This is shown in Table 2. In contrast, even though increasing ~once~trations of yeast tRNA in nuclease-treated reticulocyte iysate stim~ate by 2-3-fold the inco~orat~o~ of [35~]met~oni~e into protein, the relative amount incorporated into pre-hPL is sharply decreased with increasing amount of yeast tRNA (Fig. 5, Tables 1 and 21, We analyzed the capacity of other sources of tRNA to stimulate [ 35S]methionine incorporation into protein directed by poly(A+ ) message. Under the same conditions reported above, 80 ,ug/ml tRNA from first trimester human placenta stimulates incorporation by 2.2-fold while term placenta was twice as active and stimulates the incorporation by 4.8-fold. Htunan liver tRNA stimulates incorporation by S-fold and rabbit reticulocyte tRNA by 3.1-fold (Table 1). We also analyzed the relative ~~co~orat~o~ of [3~~]met~on~e into pre-hPL as in~ue~ced by the same tRNAs as those described in Table 1. The results are summarized in Table 2 and show that even though rabbit ret~cu~ocyte tRNA was able to stimulate total incorporation by 3-fold, it has very little influence on the synthesis of hPL. Human liver tRNA also had little effect on the rate of pre-hPL synthesis and was comparable to that of reticulocyte. However, term placenta tRNA had the highest stimulatory Table I St~rnuiatioll of (3sS]methion~ne inco~orat~on into protein by tRNA from various sources Source of tRNA

tRNA concentration (ag/mi)

40

First trimester Term placenta Human liver Rabbit reticulocyte Yeast

80

(fold stimulations 2.5 1.5 4.8 4.8 3.0 2.1 3.1 1.9 I.5 1.3

120

I60

2.2 4.7 3.4 3.2 2.3

21 4.8

2.2

--

Incubation conditions for the n&ease-treated reticulocyte lysate were carried out at 30°C for 30 min as described in Materials and Methods. Each incubation reaction contains 30 pg/ml of ploy-containing mRNA from term placenta. Control value (without added tRNA) was 3.5X IO4 cpm, and without tRNA and no message resulted in 2.6X IO4 corn.

Table 2 Effect of added tRNA to globin tRNA

added

from various

sources on the rates of pre-hPL

synthesis

as compared

Source of tRNA

(p&m’) Rabbit reticulocyte

0 (control) 40 80 120 160

(relative 1.0 1.5 1.7 1.6 1.5

Human liver

1st trimester placenta

rates of pre-hPL/g~abin~ 1.0 1.0 2.5 1.5 3.0 1.7 3.1 1.8 2.8 1.5

Term placenta

Yeast

1.0 4.0 5.0 5.1 4.9

1.0 0.8 0.5 0.3 0.1

Incorporation of {3iS]methionine into protein directed by term placenta message was carried out in the nuclease-treated lysate as described in Fig. 1 and Materials and Methods. The translational products were analyzed by 12.5% SDS-polyacrylamide gel electrophoresis and visualized by fluorography. Pre-hPL and globin bands were cut from the gels and processed for counting as described in Figs. 3 and 4. The ratio of counts (pre-hPL/globin) of the control was normalized to 1.

effect on the relative rate of pre-hPL synthesis and this was apparent even at the lowest concentration of tRNA used. The results in Table 2 indicate that tRNA from term placenta may have a functional adaptation to the codon used in the translation of mRNA for the major protein product. Moreover, it indicates that estimation of the amount of message in a heterologous cell-free system by translational assay may be biased by the system used. The effect of tRNA on selective translation of pre-hPL mRNA could be valid only if the mRNA coding sequence shows a significant bias in the choice of which of several degenerate triplets are used to code for a particular amino acid. Therefore, we tabulated the codon frequencies in rabbit globin mRNA (Grantham et al., 1980) in hPL, OL-and &hCG (Fiddes and Goodman, 1980) and for yeast alcohol d~hydrogenase I which is a very abundant protein in yeast (Bennetzen and Hall, 1982). It can be seen in Table 3 that each of the mRNAs shown is highly biased in codon selection.

Placental mRNA

191

translation

Table 3 Codon Amino acid

frequencies Codon

in yeast, rabbit Yeast a ADH-I

globin and human

hPL ’

Rabbit globin b o[

placental

protein

whCG

’

P-hCG ’

P

0

0

0

0

0

6

0

0

0

0

0

0

8 0

7 7 0 7

21

24 12 0 0 18

17 17

47 IS 0 0 0

CUA cut CUG cuu UUA UUG

3 0 19 0 2 19

0 14 99 0 0 7

0 14 110 0 0 0

18 42 65 0 0 0

17 9 17 9 0 17

0 29 59 12 0 12

Ser

UCA ucc UCG ucu AGC AGU

0 0 0 14 0 0

0 28 0 21 21 7

0 21 0 21 0 27

6 30 24 12 24 6

17 35 9 17 0 9

6 35 6 6 29 0

Thr

ACA ACC ACG ACU

0 9 0 5

0 71 0 14

0 14 0 14

18 30 12 6

17 9 26 26

6 53 0 6

Pro

CCA ccc CCG ecu

10 10 0 2

0 35 7 7

7 0 0 21

6 12 6 0

35 9 9 9

24 59 29 18

Ala

GCA GCC GCG GCU

0 16 0 19

7 41 7 48

0 64 30 11

6 12 0 0

17 9 9 26

12 29 0 12

Gly

GGA GGC GGG GGU

0 3 0 41

0 41 7 27

4 61 19 11

0 24 18 0

0 9 9 17

0 29 29 6

Val

GUA GUC GUG GUU

0 17 0 19

0 14 82 27

0 8 30 4

0 6 24 0

17 35 26 9

0 18 53 0

Arg

CGA CGC CGG CGU AGA AGG

0

0 0

0 0

192 Table 3 (continued) Amino acid

Codon

Yeast a ADH-I

Rabbit globin b cy

hPL c

cy-hCG c

/3-hCG c

P

LYS

AAA AAG

4 20

21 62

4 68

6 48

35 26

0 24

Asn

AAC AAU

I1 0

27 27

23 4

36 6

35 0

1% 6

Gin

CAA CAG

9 cl

0 27

4 34

12 42

9 35

12 24

His

CAC CAU

10 1

34 21

11 4

12 12

26 17

0

GAA GAG

20 0

21 41

27 83

30 48

17 17

29

Asp

GAC GAU

14 2

21 7

30 8

77 6

0 26

29 12

Qr

UAC UAU

13 0

14 7

15 8

24 24

i? 43

IS 0

Cys

UGC UGU

a 8

0 7

15 4

18 6

70 17

47 24

Phe

UUC uuu

8 0

34 21

30 4

48 12

35 17

18 0

Xle

AUA AUC AUU

0 12 9

0 0 1

0 11 0

0 30 12

9 9 0

0 29 0

Met

AUG

7

7

19

30

26

24

Trp

UGG

5

14

15

6

0

6

Gill

-

0

Calculated per 1000 residues based on nucleic acid sequence. a Bennetzen et al. (1982). b Grantham et al. (198Oa. b). ’ Fiddes and Goodman (1980).

DISCUSSION Nuclease-treated endogenous tRNA

reticulocyte and mRNA

lysate which is partially depleted of was used to study the effect of tRNA

Plucentd

mRNA

translation

193

from various sources on the synthesis of pre-hPL directed by poly(A+) mRNA from term placenta. The concentration of mRNA utilized was that which achieved about half-maximal stimulation of amino acid incorporation in the cell-free system and the time of analysis was within the linear range of incorporation. Therefore the differences observed represent rate differences. It is obvious that tRNA from various sources stimulated the rate of total incorporation of [ “Slmethionine into protein in the presence of term placental message by 3-4-fold (Fig. 2 and Table 1). The rate of the stimulation of protein synthesis increased linearly for most tRNAs used up to a maximum of about 100 pg/ml, with the exception of term placental tRNA which was saturating at 40 pg/ml and stimulated the system by 4%fold. The effect of tRNA is specific for the expression of pre-hPL. This is illustrated in Fig. 4. Addition of yeast tRNA stimulates total rate of incorporation of [35S]methionine into protein, while at the same time the rate of incorporation into pre-hPL is decreased (Fig. 4). This is also illustrated in Table 2. Here we compared the relative rate of synthesis of pre-hPL to that of globin. Reticulocyte tRNA and first trimester tRNA were hardly effective while human liver tRNA and term placental tRNA greatly increased the rate of pre-hPL synthesis as compared to the rate of globin synthesis. These results may be explained by assuming functional adaptation of tRNA to the hPL which is a major protein product of term placenta. Such functional adaptation of tRNA to amino acid composition of a differentiated protein product was reported for lens crystallin (Virmaux, 1969), for silk gland protein (Garel et al., 1970, 1971; Chavancy et al., 1975) and for the induction of liver vitellogenin (Maenpaa and Bernfield, 1975). This interpretation is valid only if the mRNA coding triplets show a significant bias in the choice of which of several degenerate triplets are used exhibiting a different pattern of non-random codon usage. Thus a codon bias in the mature mRNA primary sequence may serve as a function for selective translation efficiency. For this reason we tabulated the codon selection in hPL, (Y- and ,&hCG, rabbit globin and yeast alcohol dehydrogenase I. We chose yeast alcohol dehydrogenase I since it is a very abundant protein in yeast. It can be seen that in yeast alcohol dehydrogenase I the usage of 61 possible codon triplets is highly biased, only 32 codons being used. In P-globin 41 are used, in hPL 47, in cx-hCG 50 and in P-hCG 43. Not only is the usage biased but there is an even greater bias of specific degenerate codons used. For instance, UCC (Ser) is not used by yeast alcohol

194

S.J. K.&v et uf

dehydrogenase while it is a major code word in rabbit glohin hPL and hCG (Y- and ,&mRNA. On the other hand, UUG (Leu) of yeast alcohol dehydrogenase is a major codon utilized but it is not utilized at all in hPL and ,&globin. Another example is valine GU& which is not utilized by yeast while GUG is a major codon used by rabbit globin hPL and cy- and &hCG. Upon careful examination of Table 3 it can be seen that for most code words there are significant differences in the usage of code words for a given amino acid. The effect of yeast tRNA on the differential translation of hPL message could be explained by assuming that there is a correlation between codon usage and tRNA abundance. In yeast it was shown that for each of the 16 amino acids whose tRNAs have been sequenced, the major isoaccepting species present in fact is that with an anticodon allowing it to translate the most frequently used codon for that amino acid (Bennetzen and Hall, 198213). Moreover, it was shown that major proteins in yeast such as alcohol dehydrogenase I and glyceraldehyde-3phosphate dehydrogenase shared the same usage of codon bias. In contrast, mRNA for minor proteins in yeast do not share this codon bias. The same phenomenon was reported for E. coli. Minor proteins expressed in E. coii exhibit a codon usage which is highly biased toward the same sets of preferred codons and there is a correlation between this codon usage pattern and the abundance distribution of 35 isoaccepting tRNA in E. cdi (Post et al., 1979; Nakamura et al., 1980; Post and Nomura, 1980). Because the concentration of charged cognate tRNA governs the step time required to add an amino acid opposite each codon, rapid translation is favored by the use of triplets for abundant tRNAs. Therefore, it seemed likely that in placenta there is a selection for a high rate of translation of highly expressed genes by a bias toward codon usage which matched the most abundant tRNA acceptors. In this tissue, therefore, the high output of protein is achieved by (1) increased rate of transcription of specific mRNA, (2) accumulation of the mRNA which indicates stability and (3) high rate of translation. All these genetic parameters might equally well have served to provide a high output of gene product. The increase in hPL mRNA during placental development has been reported (Hubert et al., 198 1; McWilliams et al., 1977). Our results indicate that in addition to mRNA accumulation, other factors may come into play to contribute to the translation of hPL mRNA. They also indicate that a translational assay in a heterologous cell-free system may only indicate relative abundance of hPL message and this may well be correlated to the amount found by hybridization techniques. However,

the amount of protein expressed as the final gene product may be of values much higher or lower than could be accounted for by the message in the tissue because of translational parameters, as was shown recently for the induction of vitellogenin in chick liver (Gehrke et al., 198 I b).

ACKNOWLEDGEMENTS We wish to thank Dr. Bruce Roe for his gifts of rabbit reticulocyte tRNA and human liver tRNA. This work was supported by NXH Grants HD-14072 and HD-I 1596.

REFERENCES Bast, R.E., Garfield, S.A., Gehrke, L. and Ilan, J. (1977) Proc. Natl. Acad. Sci, (U.S.A.) 74, 3133-3137. Bennetzen, J.L, and Hall, RD. (1982~~) J. Biol. Chem. 257, 30183025. Bennetzen, J.L. and Hall, B.D. (198%) 3. Biol. Chem. 257, 3026-3031. Bonaer, W.H. and Laskey, R.A. (1974) Eur. J. Biochem. 46, 83-88. Fiddes, J.C. and Goodman, H.M. (1980) Nature (London) 286,684-687. Grantham, R., Gautier, C., Gouy. M., Mercier, R. and Pave, A. (198Oa) Nucleic Acids Res. 8, r49-r62. Grantham, R., Gautier, C. and Gouy, M. (1980b) Nucleic Acids Res. 8, 1893-1912. Hubert, C., Mondon, F. and Cedard, L. (1981) Mol. Cell. Endocrinol. 24, 339-355. Kelly, S., Folman, R., Hochberg, A. and Ilan, J. (1980) Biochim. Biophys. Acta 609, 278-285. Laemmli, U.K. (1970) Nature (London) 227, 680-685. Mans, R.J. and Novelli, G.D. (1961) Arch. Biochem. Biophys. 94, 48-53. Nakamura, K., Pirtle, R.M., Pirtle, I.L., Takeishi, K. and Inouye, M. (1980) J. Biol. Chem. 255,210-216. Patterson, B.M., Roberts, B.E. and Kuff, E.L. (1977) Proc. Natl. Acad. Sci. (U.S.A.) 74, 4370-4374. Petham, R.R.B. and Jackson, R.J. (1976) Eur. J. B&hem. 67,247-256. Post, L.E. and Nomura, M. (1980) J. Biol. Chem. 255, 4660-4666. Post, LE., Strycharz, G-D., Nomura, M., Lewis, H. and Dennis, P.P. (1979) Proc. Natl. Acad. Sci. (U.S.A.) 76. 1697-1701. Roe, B.A. (1975) Nucleic Acids Res. 2, 21-42. Sherwood, L.M., Burstein, and Schecter, 1. (1979) Proc. Natl. Acad. Sci. (U.S.A.) 76, 3819-3823. Shine, J., Seeburg, P.H., Martial, J.A., Baxter, J.D. and Goodman, H.M. (1977) Nature (London) 270, 474-499. Szczesna, E. and Boime, I. (1976) Proc. Natl. Acad. Sci. (U.S.A.) 73, 1179-l 183.