Plant Molecular Biology 2:199-206 (1983). 9 Martinus NijhoJf/Dr W. Junk Publishers, The Hague. Printed in the Netherlands.
Developmental biochemistry of cottonseed embryogenesis and germination XVll. Developmental expression o f genes f o r the principal storage proteins Leon Dure III*, Jana B. Pyle, Caryl A. Chlan, Jean C. Baker & Glenn A. Galau l Department o f Biochemistry, University o f Georgia, Athens, GA 30602, U.S.A. l Present address: Department o f Botany, University o f Georgia, U.S.A.
Keywords: cotton storage proteins, cloned cDNA, RNA dot hybridization, in vitro translation
Abstract The developmental time period and the magnitude of expression of the genes for the principal cottonseed storage proteins have been measured by several means. R N A was extracted from cotton cotyledons at stages during embryogenesis and the relative amounts of the m R N A S for these proteins were determined by cell-free translation in the wheat germ system and by dot and northern hybridization of the RNA with cloned c D N A probes representing the three subfamilies of the major storage protein genes. The rates of reassociation in solution of some of the RNAs with one of the c D N A clones were also determined. Data from all four procedures show that the storage protein m R N A s are demonstrable in very small embryo cotyledons, rapidly reach a high abundance level that is maintained during most of embryo growth, and then fall precipitously in amount in the last days of embryogenesis. The expression of all three gene subfamilies appears coordinate. Further, c D N A reverse transcribed from the poly(A) + m R N A from a stage of m a x i m u m storage protein synthesis was hybridized to saturation with c D N A clones representing each of the subfamilies. These data indicate that the m R N A s for two of the families reach the same relative level in the total m R N A population which is about 15% of the total m R N A mass. The m R N A of the third subfamily comprises only 5% of the total m R N A mass at this stage. This apparent 3:3:1 ratio of m R N A s does not change during the period of storage protein synthesis. Based on the amounts of the storage protein species in the mature seed, the m R N A s of each subfamily appear to be translated to the same extent during embryogenesis.
Abbreviations
Introduction
SDS, EDTA, poly(A) +, poly(A)-, Cot,
The principal storage proteins of the cotton seed are produced by the processing of preproproteins of about 69 and 60 kD in size (4). We have reported that the 69 kD protein set emanates from a subfamily of m R N A s all of whose members can hybridize at moderate criteria with a single cloned c D N A sequence (9). In contrast, we have reported that the 60 kD protein set emanates from two sequence subfamilies of m R N A s whose c D N A representatives do not cross-hybridize except at a low criterion (9). Thus, the principal storage proteins are encoded in three subfamilies of genes that share enough homology to allow for immunological cross-reactivity a m o n g all of the principal storage proteins and their
kD, kb,
sodium dodecyl sulfate ethylenediamine tetraacetic acid polyadenylated nonpolyadenylated concentration nucleic acid in moles nucleotide 1 i • time in seconds kiloDalton kilobase
*To whom reprint requests should be addressed.
200 precursors (4, 9), but that have diverged to the extent that they can be separated into three subfamilies by nucleic acid hybridization techniques. The construction of cloned c D N A representing the m R N A s of each of the three major storage protein subfamilies (9) made it possible to follow the expression of storage protein genes during embryogenesis. We present here the time period and magnitude of expression of each of the gene subfamilies during embryogenesis as determined by in vitro translation of m R N A , RNA hybridization and by the kinetics of solution hybridization.
solution at 70 o C in 0.38 M NaC1, 0.001 M EDTA, 0.02 M PIPES-HC1, pH 6.9. The total c D N A was made radioactive during its synthesis and its extent of hybridization determined as the percent of total radioactivity made nuclease S-1 resistant (8). The values of percent of c D N A hybridized with the cloned cDNAs were normalized for the percent of the total c D N A which could hybridize with total cotton DNA. Hybrid formation in solution between a cloned c D N A insert probe and total high molecular weight RNA was performed as described (8). The data were computer-fitted by the method of Pearson et al. (10).
Methods and materials General
Results
The species of cotton used was G o s s y p i u m hirsutum, variety Coker 201. Embryos were obtained from greenhouse plants. The extraction of cotyledon proteins (3), the purification of R N A and poly(A) + R N A (7), in vitro translation in the wheat germ system (4, 7), the protocol for gel electrophoresis of proteins and the fluorography of dried gels (5), northern blot analysis, synthesis of cDNA, and hybridization of filter bound RNA with radioactive recombinant plasmid D N A (9) were all carried out as previously described. For dot blot analysis of m R N A concentrations during development, total high molecular weight RNA was bound to nitrocellulose filters and hybridized to radioactive plasmid D N A containing a c D N A insert representing one of the storage protein m R N A subfamilies (9). The processing of the nitrocellulose filters before, during and after hybridization was essentially as described (9). The construction and cloning of c D N A sequences corresponding to the subfamilies of storage protein genes are described in the preceding paper (9) as is the nick-translation of D N A to produce radioactive molecules. The isolation of inserted sequences after plasmid isolation involved sequence liberation by Pst I digestion, separation of fragments on acrylamide gels, and electroelution of the inserted sequence as described by Southern (12) and Yang, et al. (14).
The proteins of developing cotton cotyledons can be separated into a readily soluble fraction that can be solubilized in neutral dilute salt solutions and an insoluble residue that requires higher salt concentrations or 8M urea or 2% SDS at neutral pH to be dissolved (3). This latter fraction is composed chiefly of the proteins of the protein bodies of the tissue. Figure 1A shows the relative concentration in cotyledons of the proteins of the insoluble fraction during the development of the tissue as given by Coomassie staining of a one-dimensional SDS polyacrylamide gel. The youngest stage of embryo development used was embryos of 5 mg in wet weight. Just prior to the desiccation of the embryo to form the mature seed, embryos typically weighed 125 mg. Thus the developmental time span covered encompasses a 25-fold increase in embryo size and corresponds to the last 3/5ths of the total embryogenic period. It can be seen from this figure that the 52 and 48 kD proteins become the principal species by the dry seed stage. Their precursors (70 and 67 kD species, respectively), which are prominent in the early phase of storage protein deposition, have all been cleaved to form the mature species by the dry seed stage. The accumulation of the 52 kD mature species appears to lag relative to the 48 kD species during the early phase of deposition, but this is likely due to the very slow cleavage of the 70 kD precursors. In contrast, the 67 kD species does not accumulate due apparently to its more rapid processing. Both the 70 and 67 kD species occur in protein bodies, indicating that their cleavage to
Solution hybridization
Hybridization to saturation of total c D N A with individual cloned c D N A probes was carried out in
201
Fig. 1. A: One dimensional gel electrophoresis in SDS of protein from the insoluble fraction of cotyledons at various stages of development visualized by Coomassie staining. The developmental stages are given along the top of the figure in terms of embryo wet weight in mg. The approximate size of some of the proteins is given in kD on the right margin. Approximately 20 #g of protein were loaded in each well, except for the 5 and 10 mg stages which contained 15 ~g each. B: One dimensional electrophoresis in SDS of the products of in vitro translation in the wheat germ from RNA extracted from cotyledons at the same developmental stages as in A visualized by fluorography. An equal amount of radioactivity was loaded in each well. C: Dot hybridization of RNA from the same developmental stages as in A and B with radioactive cloned probes listed on the left margin. Two/.tg of RNA was immobilized in each dot and hybridized with plasmid DNA of a specific radioactivity of I 2 X 10~ dpm ~g l at a concentration of 2 • 106 dpm ml I as described in (9). Perpendicular lines connecting A-B-C are intended to line up equivalent developmental stages.
202 form the mature proteins takes place in situ within these bodies. The 70 and 52 kD proteins are glycosylated (3) and the true mass of their protein component is unknown, but is likely to be tess than that of their 67 and 48 kD counterparts since they originate from a shorter set of m R N A s and from smaller initial translation products (9). All of these species (70, 67, 52 and 48 kD) are comprised of isolectric variants (3, 4). The other products of the final cleavage of the 70 and 67 kD species which should have apparent molecular weights between 18 and 20 kD are a m o n g the lower molecular weight proteins of this fraction but have not yet been identified. Figure 1A also shows an apparent storage protein precursor of about 38 kD that accumulates and disappears coordinately with the 70 kD species. This species probably is cleaved in situ to smaller fragments seen in the figure, but they have not been identified. Figure 1B shows the profile of proteins synthesized in the wheat germ system from RNA prepared from cotyledons of the same developmental stages as in Fig. 1A. An additional electrophoretic lane shows the proteins synthesized from R N A of 12-h germinated cotyledons as well. The data show that m R N A s giving rise to proteins of 69 and 60 kD increase from a very low level in 5 mg embryos to become the predominant species by the 30 mg stage. We have previously equated the 69 kD translation products with the 67 and 48 kD proteins and the 60 K D translation products with the 70 and 52 kD proteins (4, 9) and have shown that these initial translation products are comprised of isoelectric variants as well (4). The abundancy of these m R N A s changes little throughout the remainder of embryogenesis until the embryos reach about 1 l0 rag. After this stage these m R N A s become undetectable by this technique, nor are they discernible in germinating cotyledons. Thus, m R N A s for both the 69 and 60 kD proteins appear to rise and fall in concentration coordinately. Notable in Fig. 1B is the fact that there is no m R N A species detectable by in vitro translation of a size and abundancy requisite for the synthesis of the proteins about 38 kD (Fig. 1A). Total RNA from the same stage cotyledons as in Fig. 1B was dot blotted and hybridized with each of the three cloned cDNAs (9) representing the three storage protein m R N A subsets. The relative extent of hybridization of each clone with a constant
amount of total R N A from cotyledons of each developmental stage is shown in Fig. 1C. Clone C-72 has been shown to hybridize with all the m R N A which synthesizes the 69 kD proteins, whereas clones C-94 and C-134 each hybridize with a specific subfamily of the m R N A s that synthesize the 60 kD proteins (9). The developmental picture observed with this technique is similar to that observed in in vitro translation and shows, in addition, that the two m R N A subfamilies for the 60 kD proteins change in concentration coordinately. (The intensities of the radioactive signals observed with each of the three probes are not necessarily a comparison of m R N A concentrations between the three subfamilies, since the probes differed somewhat in specific radioactivity and exposure times, for the filters were not normalized.) The changes in abundance with develo?mental time seen in Fig. 1C were corroborated by northern blots of the same total RNAs separated by molecular weight on agarose gels (9) (data not shown). In these experiments hybridization of the cloned cDNAs to each stage R N A was exclusively to the m R N A s of the expected size, i.e. 2.26 kb m R N A with C-72 and 1.96kb m R N A with C-94 and C-134 (9). The foregoing approaches to measuring the storage protein m R N A ahundancy present the same pattern: that of a rapid increase in young embryo cotyledons, a subsequent high steady-state level that is maintained until the embryos are nearly full size, and finally a precipitous drop to a level barely detectable by the hybridization techniques. The validity of this pattern was tested by measuring the rate of hybridization in solution of one of the cloned c D N A inserts with total high molecular weight R N A prepared from cotyledons of several developmental stages. However, in order to use these rates of hybridization to measure the concentrations of storage protein mRNAs, we first determined the concentrations of the m R N A for each of the three subfamilies of proteins in a single developmental stage m R N A population. This was accomplished by hybridizing to saturation c D N A constructed from young embryo cotyledon poly( A ) + m R N A with each of the plasmids carrying one of the three c D N A inserts. The final percent of the c D N A that hybridized indicates the percent of the poly(A)+mRNA mass that is comprised of the m R N A sequences for each storage subfamily. This assumes, as has been shown in other studies (I), that
203
A
A
.
.
.
.
o o~o~ o~c~176
.
.
j .o
.
~ ~176176
.//
~ ~ . /%,.
/
D ~
7,,'
-5
.~
I
-4
~'" -
o
-3
-2
-I
60
o
B
4C
9
o
9
o (~
9
.
50
9
~
~L
66.
A
9
o
40
30
o
9
"
o
20
10
o
o o
o
9
o
o o
o
~
-~
o
1
Q
2
3
log Coi Fig. 2. A: Saturation hybridization of radioactive cDNA ( ~ 400 nucleotides) prepared from poly(A)+ mRNA of 40 mg stage cotyledons with each of the cloned probes for the principal storage protein subfamilies. Plot is of percent hybridization versus log plasmid insert Cot. zx, hybridization with plasmid C-72; o, with plasmid C-94; 121,with plasmid C-134, Plasmid insert DNA was present in at least 250-fold mass excess to cDNA. The data were computer fit with second order kinetics (10). Co values were not corrected for salt concentration. The figure insert is a plot of percent hybridization versus plasmid insert Cot. B: Kinetics of hybridization of total high molecular weight RNA with the cDNA-derived insert of plasmid C-72. The plot is of percent hybridization versus log RNA Cot. The nick-translated insert was approximately 150 nucleotides. RNA was prepared from: 9 20 mg embryo cotyledons; O 40 mg embryo cotyledons; n 110 mg embryo cotyledons; 9 120 mg embryo cotyledons; 9 dry seed; [] cotyledons of 24-h germinated seedlings. RNA was present in at least 2.2 • 103-fold mass excess of the insert DNA. Reactions were normalized to 50% from observed values that ranged from 30 to 43%. Cot values were not corrected for salt concentration. The data were computer fit with pseudo-first order kinetics (I0).
204 reverse transcriptase transcribes all m R N A s equally without respect to sequence. The plasmids were hybridized with c D N A representing 40 mg stage poly(A) + m R N A , and the data, which show considerable scatter, are presented in Fig. 2A. The lines in this figure are the result of computer-fits of the data. Although these fits indicate a small difference in the saturation values between plasmid C-72 and C-94, we do not feel this difference is significant in view of the imprecision of the measurements and have rounded offthe saturation value for these two plasmids at 15% of the cDNA. Surprisingly, C-134, whose insert represents one of the 60 kD proteins, hybridizes only to about 5% of the cDNA. Inserted into Fig. 2A is a plot of percent c D N A hybridized vs. plasmid insert Cot which shows that apparent saturation is reached at the Cot values used. These values of course only refer to the concentration of these m R N A s in poly(A) + m R N A , but in cotton cotyledons both poly (A) + and poly(A) m R N A appear to contain the same m R N A sequences (including storage protein mRNAs) in similar proportion (5 and unpublished data) and, consequently, we believe the data presented in Fig. 2A reflect the concentration of these m R N A s in total m R N A as well as poly(A) + m R N A . Other data confirm the results of Fig. 2A for this extrapolation. I n v i t r o translation of total, poly(A) + and poly(A) RNA gives the same relative amounts of radioactive proteins as seen in Fig. lB. Further, hybridization of R N A with c D N A prepared from 40 mg stage cotyledons indicates the existence of R N A components of this high abundance in both total and poly(A) + RNA. The in v i t r o protein synthesis pattern shown in Fig. 1B indicates that m R N A s for the 69 and 60 kd proteins not detectably change relative one to another during embryogenesis. Further, the RNA dot blot hybridization pattern (Fig. 1C) and northern hybridization substantiate this. This pattern also shows that there is no change in the m R N A s for the two 60 kD protein subfamilies relative one to another. Thus the relative abundance of the three subfamilies seen in Fig. 2A does not change in embryogenesis. Knowing the concentration of these m R N A s at the 40 mg stage, it was possible to measure the absolute abundance during this period of all three m R N A subfamilies by measuring the rate of hybridization of a single cloned insert (C-72) with
RNA from several developmental stages relative to the rate of hybridization with this reference 40 mg stage RNA. Hybrid formation (Fig. 2B) was measured as the percent of the cloned c D N A probe that became resistant to S-1 hydrolysis. Since only a single strand of the probe is complementary to the m R N A sequences of the subfamily, only 50% of the probe can hybridize. In actuality a small portion of the probe becomes fragmented in the nick-translation step into pieces too small to form a stable hybrid. For this reason the computer derived hybridization in Fig. 2B has been normalized such that the degree of hybridization ranges from 0 to 50%. The concentration of the probe used in these experiments was so low that probe: probe reassociation should contribute negligibly to the S-1 resistant radioactivity. The probe was reassociated with total RNA from 20, 40, 110 and 120 mg embryo cotyledons, from dry seed cotyledons and 24-h germinated seedling cotyledons. Since the amount of m R N A in these RNA preparations has been determined previously (7), it was possible to calculate the concentration of the m R N A for 69 kD storage protein in terms of the fraction of total m R N A as well as the fraction of total RNA. The number of m R N A transcripts present at each developmental stage were also calculated per embryo and per haploid genome equivalent. These calculations are presented in Table 1. It is clear from Table 1 and from a visual comparison of Fig. 2A, B that a considerable amount of m R N A from 20, 40, and 110 mg embryo cotyledons is comprised of sequences for the 69 kD proteins, and by extension from the previous data (Fig. 1C, Fig. 2A), of sequences for the 60 kD proteins. The data indicate that there is an abrupt drop in the concentration of m R N A for these proteins from the 110 mg stage to the 120 mg stage. Even more of these sequences are lost as the seed desiccates. However, a trace remains in cotyledons after 24 h of germination. Northern blot analysis (data not shown) detected mature sized m R N A transcripts in all of these later developmental stages, in approximately the same concentration as determined in Fig. 2B and Table 1. Clearly, the picture presented by the kinetics of clone C-72: RNA reassociation confirms that suggested by in v i t r o translation (Fig. I B) and hybridization of the cloned probes with RNA dots (Fig. IC) and northern blots. Quantitation of m R N A levels by all these techniques agrees remarkably well.
205 Table 1. Concentration of 2.26 kb storage protein mRNA during cotton cotyledon development measured by the kinetics of hybridization of C-72 with total RNA.
Developmental s t a g e
Post-anthesis ~ (days)
Ksfob (M t s 1)
Concentration Fraction of RNA
Embryo (wt weight) 20rag 40 mg ll0mg 120rag Dry seed 24-H germination
24 26 44 51 60+
l.l 1.6 9.0x 10-I 3.9x 10-2 6.8X 10-3 1.9 X 10-3
Molecules
Total RNAc
mRNAd
Per embryoe Per haploid genomef
3.0x 10 3 4.4X 10 3 2.5• 3 I.IX 10 4 1.9• 5 5.2 X l0 6
1.0x 10 r 1.5 X 10 1 8.6>(10 2 3.8x 10 3 I.IXI0 3 1.6 X l0 4
5.3X 10~~ 2.9 x 10~ 3.8X l011 1.7x 10j~ 3.0X 109 8.3 X l0s
6.2x 103 1.6 • 104 1.1 x 104 4.9x I02 8.6X 10l 2.4 X 10~
a Walbot & Dure (13). b Least squares best estimate of the pseudo first order rate constant of the hybridization of C-72 with total RNA, as presented in Fig. 2B. Fractional uncertainties are approximately • 0.09 (one standard error). All other calculations in the table thus have associated uncertainties at least this large. c Calculated assuming the 2.26 kb mRNA comprises 15% of the 40 mg stage poly(A)+ mRNA and total mRNA (Fig. 2A and discussion in Results) and a total mRNA content of 2.9% (8). a Calculated assuming 2.9% of embryo and 1.7% of dry seed (8) and 3.3% of 24-h seedling cotyledon RNA is total mRNA (7). e Based on the total high molecular weight RNA contents of the cotyledons (unpublished data). f Based on the DNA content of the cotyledons during development (13) and a haploid genome size of 1.8 x 109 nucleotides. G. hirsutum is an allotetraploid.
Discussion The levels of the m R N A s c o d i n g for the storage proteins in c o t t o n c o t y l e d o n s during e m b r y o g e n e sis have been measured by f o u r i n d e p e n d e n t means, three of which m a d e use of the cloned probes for the three m R N A subfamilies. D if f e r e n t pieces of inf o r m a t i o n can be ascertained f r o m the techniques, 1. All three m R N A subsets increase rapidly in m i d - e m b r y o g e n e s i s , but decline d r a m a t i c a l l y in a short p e r i o d late in e m b r y o g e n e s i s as s h o w n by all techniques. These studies c o n f i r m in most details the q u a n t i t a t i v e c o n c l u s i o n s r e g a r d i n g storage protein m R N A s c o n c e n t r a t i o n s in c o t t o n c o t y l e d o n s made by Gal au & D u r e (8), which were based on h y b r i d i z a t i o n of different d e v e l o p m e n t a l stage m R N A p o p u l a t i o n s with h o m o l o g o u s and heterologous stage c D N A s . 2. All three gene subfamilies are expressed as m R N A in a c o o r d i n a t e f a s h i o n as s h o w n by the h y b r i d i z a t i o n of cloned c D N A probes to R N A dots or n o r t h e r n blots and (to a lesser extent) by in v i t r o translation. H o w e v e r , the stained protein gel (Fig. 1A) shows that the 52 k D m a t u r e p r o t e i n set does not increase as ra p id ly in early e m b r y o g e n e s i s as does the 48 k D set. This a p p a r e n t discrepancy is
caused by the fact t h a t a substantial a m o u n t of the 52 k D proteins are in the 70 k D p r e c u r s o r species (which are very ev i d en t on the stained gel) until Synthesis and processing is c o m p l e t e in late e m b r y o genesis. 3. Th e m R N A s for two of the subfamilies a p p e a r to exist in the same a m o u n t s , whereas one of the m R N A subfamilies a p p e a r s to be expressed at only 1 / 3 the level of the o t h er t w o subfamilies as indicated by the c D N A s a t u r a t i o n hybridization. These m e a s u r e m e n t s are of m R N A levels and say n o t h i n g a b o u t the n u m b e r of p r o t ei n copies translated f r o m those m R N A s in v i v o , n o r a b o u t the n u m b e r of genes c o m p r i s i n g each subfamily. We have r e p o r t e d (3) that in the m a t u r e seed the ratio of 52 k D p r o t e i n s to 48 k D proteins is a b o u t 4:3 based on several m e t h o d s o f p r o t e i n d e t e r m i n a t i o n . In actuality, the 52 k D proteins are glycosylated and the p r o t e i n f r a c t i o n of these molecules m ay represent only 40 k D of p r o t ei n mass. ( D e d u c e d f r o m the fact that the increase in a p p a r e n t m o l e c u l a r weight o b s e r v e d on gels u p o n g l y c o s y l a t i o n of the p u t a t i v e 58 k D p r o p r o t e i n f o r m of the 60 k D p r e p r o p r o t e i n to the 70 k D p r e c u r s o r is 12 kD.), If the final 52 and 48 k D proteins in fact c o n t a i n 40 and 48 k D of protein, respectively, and each of the m R N A sub-
206 sets were translated equally in embryogenesis, a r o u g h l y 8:7 mass ratio of 52 to 48 k D proteins would be expected. This is because the 52 k D proteins, a l t h o u g h c o n t a i n i n g less protein mass per molecule t h a n do the 48 kD proteins, are derived f r o m two m R N A subsets which in total a m o u n t comprise a b o u t 20% of the m R N A d u r i n g the period of m a x i m u m expression, whereas the 48 kD m R N A s constitute a b o u t 15%. F r o m this reasoning it seems likely that the three subsets of m R N A s are translated to a n equal extent d u r i n g embryogenesis. Cell division in these cotyledons ceases at a b o u t the 40 mg stage and D N A synthesis ceases at a b o u t the 80 mg stage, while the m R N A s for these storage proteins are at their high level. Since no change in their c o n c e n t r a t i o n is observed immediately before or after these transitions, the r e g u l a t i o n of their t r a n s c r i p t i o n must be i n d e p e n d e n t of the regulation of the genes responsible for cell division or D N A replication. The fact that the m R N A s for the storage proteins are expressed c o o r d i n a t e l y and are m a i n t a i n e d at the a p p a r e n t 3:3:1 ratio t h r o u g h o u t their period of expression may indicate that the same ratio exists for the n u m b e r of genes for these subfamilies a n d that their expression is not individually modulated. The drastic decrease in these m R N A s at the end of embryogenesis (between the 110 a n d 120 mg stages) represents a b o u t a 25-fold drop in concent r a t i o n that occurs over a 3-to-4-day period. If this is solely the result of a sudden cessation in the t r a n s c r i p t i o n of these genes, the half life of the m R N A s would be between 15 a n d 20 h. This is a realistic value for m R N A half life, a n d there is no reason to evoke a special m e c h a n i s m for the disapp e a r a n c e of these m R N A s . Both the data in Fig. 2B a n d n o r t h e r n blot analysis (data not shown) indicate that storage protein m R N A s decline in concent r a t i o n (as a fraction of total R N A ) even more rapidly in the first 24 h of g e r m i n a t i o n . The species of c o t t o n used is a n a t u r a l allotetraploid ( a m p h i d i p l o i d ) . It carries what can be recognized as a n ancestral A and D g e n o m e (6). It is t e m p t i n g to conjecture that the 69 k D protein genes derive from one of the ancestral genomes and the 60 k D proteins from the other. However, p r e l i m i n a r y results in which proteins from a diploid species carrying only the A g e n o m e were electrophoresed in one and two d i m e n s i o n s showed the existence of both molecular weight sets of proteins. Storage protein synthesis has been followed on
the m R N A level in other plant species, principally legumes a n d cereal of agricultural i m p o r t a n c e . Much of this work has been collated recently by Spencer & Higgins (11) a n d in a N A T O A.S.I. p u b l i c a t i o n (2).
References I. Axel, R, Feigelson, P & Schutz, G, 1976. Analysis of the complexity and diversity of mRNA from chicken liver and oviduct. Cell 7:247 254. 2. Ciferri, O & Dure III, LS, 1983.The Structure and Function of Plant Genomes. Plenum Press, New York and London. 3. Dure 111,LS & Chlan, CA, 1981. Developmental biochemistry of cottonseed embryogenesisand germination XII Purification and properties of the principal storage proteins. Plant Physiol. 68:187 194. 4. Dure Ill, LS & Galau, GA, 1981. lbid XII1 Regulation of the biosynthesis of the principal storage proteins. Plant Physiol. 68:187 194. 5. Dure III, LS, Greenway, SC & Galau, GA, 1981. ibid XIV Changing messenger ribonucleic acid populations as shown by in vitro and in vivo protein synthesis. Biochem. 20: 4162-4168. 6. Fryxell, PA, The Natural History of the Cotton Tribe, Texas A & M University Press, College Station and London. 7. Galau, GA, Legocki, AB, Greenway, SC & Dure Ill, LS, 1981. Cotton messenger RNA sequences exist in both polyadenylated and nonpolyadenylated forms. J. Biol. Chem. 256:2551-2560. 8. Galau, GA & Dure lI1, LS, 1981. Developmentalbiochemistry of cottonseed embryogenesis and germination XV Changing messenger rihonucleic acid populations as shown by reciprocal heterologous complementary deoxyribonucleic acid hybridization. Biochem. 20:4169-4178. 9. Galau, GA, Chlan, CA & Dure lII, LS (companion paper). Ibid XVI Analysis of the principal cotton storage protein gene family with cloned cDNA probes. (Submitted to Plant Mol. Biol.). 10. Pearson, WR, Davidson, EH & Britten, RJ, 1977. A program for least squares analysis of reassociation and hybridization data. Nucleic Acids Res. 4:1727 1735. ll. Spencer, D & Higgins, TJV, 1982. Seed maturation and deposition of storage proteins. In: A Smith & D Grierson (eds.) The Molecular Biology of Plant Development. University of California Press, Berkeley and Los Angeles. 12. Southern, E. M., 1979. Gel electrophoresis of restriction fragments. In: A. Wu (ed.) Methods in Enzymology,68: pp. 152 176. Academic Press, New York, NY. lY Walbot, V. & Dure Ill, LS, 1976. Developmentalbiochemistry of cottonseed embryogenesisand germination VII Characterization of the cotton genome. J. Mol. Biol. 101: 503 536. 14. Yang, R. C. A., Lis, J & Wu, R, 1979. Elution of DNA from agarose gels after electrophoresis. In: A Wu, (ed.) Methods in Enzymology,68: pp 176 182.Academic Press, New York, NY. Received 24 May 1983; in revised form and accepted 2 August 1983.
Developmental biochemistry of cottonseed embryogenesis and germination XVll. Developmental expression o f genes f o r the principal storage proteins Leon Dure III*, Jana B. Pyle, Caryl A. Chlan, Jean C. Baker & Glenn A. Galau l Department o f Biochemistry, University o f Georgia, Athens, GA 30602, U.S.A. l Present address: Department o f Botany, University o f Georgia, U.S.A.
Keywords: cotton storage proteins, cloned cDNA, RNA dot hybridization, in vitro translation
Abstract The developmental time period and the magnitude of expression of the genes for the principal cottonseed storage proteins have been measured by several means. R N A was extracted from cotton cotyledons at stages during embryogenesis and the relative amounts of the m R N A S for these proteins were determined by cell-free translation in the wheat germ system and by dot and northern hybridization of the RNA with cloned c D N A probes representing the three subfamilies of the major storage protein genes. The rates of reassociation in solution of some of the RNAs with one of the c D N A clones were also determined. Data from all four procedures show that the storage protein m R N A s are demonstrable in very small embryo cotyledons, rapidly reach a high abundance level that is maintained during most of embryo growth, and then fall precipitously in amount in the last days of embryogenesis. The expression of all three gene subfamilies appears coordinate. Further, c D N A reverse transcribed from the poly(A) + m R N A from a stage of m a x i m u m storage protein synthesis was hybridized to saturation with c D N A clones representing each of the subfamilies. These data indicate that the m R N A s for two of the families reach the same relative level in the total m R N A population which is about 15% of the total m R N A mass. The m R N A of the third subfamily comprises only 5% of the total m R N A mass at this stage. This apparent 3:3:1 ratio of m R N A s does not change during the period of storage protein synthesis. Based on the amounts of the storage protein species in the mature seed, the m R N A s of each subfamily appear to be translated to the same extent during embryogenesis.
Abbreviations
Introduction
SDS, EDTA, poly(A) +, poly(A)-, Cot,
The principal storage proteins of the cotton seed are produced by the processing of preproproteins of about 69 and 60 kD in size (4). We have reported that the 69 kD protein set emanates from a subfamily of m R N A s all of whose members can hybridize at moderate criteria with a single cloned c D N A sequence (9). In contrast, we have reported that the 60 kD protein set emanates from two sequence subfamilies of m R N A s whose c D N A representatives do not cross-hybridize except at a low criterion (9). Thus, the principal storage proteins are encoded in three subfamilies of genes that share enough homology to allow for immunological cross-reactivity a m o n g all of the principal storage proteins and their
kD, kb,
sodium dodecyl sulfate ethylenediamine tetraacetic acid polyadenylated nonpolyadenylated concentration nucleic acid in moles nucleotide 1 i • time in seconds kiloDalton kilobase
*To whom reprint requests should be addressed.
200 precursors (4, 9), but that have diverged to the extent that they can be separated into three subfamilies by nucleic acid hybridization techniques. The construction of cloned c D N A representing the m R N A s of each of the three major storage protein subfamilies (9) made it possible to follow the expression of storage protein genes during embryogenesis. We present here the time period and magnitude of expression of each of the gene subfamilies during embryogenesis as determined by in vitro translation of m R N A , RNA hybridization and by the kinetics of solution hybridization.
solution at 70 o C in 0.38 M NaC1, 0.001 M EDTA, 0.02 M PIPES-HC1, pH 6.9. The total c D N A was made radioactive during its synthesis and its extent of hybridization determined as the percent of total radioactivity made nuclease S-1 resistant (8). The values of percent of c D N A hybridized with the cloned cDNAs were normalized for the percent of the total c D N A which could hybridize with total cotton DNA. Hybrid formation in solution between a cloned c D N A insert probe and total high molecular weight RNA was performed as described (8). The data were computer-fitted by the method of Pearson et al. (10).
Methods and materials General
Results
The species of cotton used was G o s s y p i u m hirsutum, variety Coker 201. Embryos were obtained from greenhouse plants. The extraction of cotyledon proteins (3), the purification of R N A and poly(A) + R N A (7), in vitro translation in the wheat germ system (4, 7), the protocol for gel electrophoresis of proteins and the fluorography of dried gels (5), northern blot analysis, synthesis of cDNA, and hybridization of filter bound RNA with radioactive recombinant plasmid D N A (9) were all carried out as previously described. For dot blot analysis of m R N A concentrations during development, total high molecular weight RNA was bound to nitrocellulose filters and hybridized to radioactive plasmid D N A containing a c D N A insert representing one of the storage protein m R N A subfamilies (9). The processing of the nitrocellulose filters before, during and after hybridization was essentially as described (9). The construction and cloning of c D N A sequences corresponding to the subfamilies of storage protein genes are described in the preceding paper (9) as is the nick-translation of D N A to produce radioactive molecules. The isolation of inserted sequences after plasmid isolation involved sequence liberation by Pst I digestion, separation of fragments on acrylamide gels, and electroelution of the inserted sequence as described by Southern (12) and Yang, et al. (14).
The proteins of developing cotton cotyledons can be separated into a readily soluble fraction that can be solubilized in neutral dilute salt solutions and an insoluble residue that requires higher salt concentrations or 8M urea or 2% SDS at neutral pH to be dissolved (3). This latter fraction is composed chiefly of the proteins of the protein bodies of the tissue. Figure 1A shows the relative concentration in cotyledons of the proteins of the insoluble fraction during the development of the tissue as given by Coomassie staining of a one-dimensional SDS polyacrylamide gel. The youngest stage of embryo development used was embryos of 5 mg in wet weight. Just prior to the desiccation of the embryo to form the mature seed, embryos typically weighed 125 mg. Thus the developmental time span covered encompasses a 25-fold increase in embryo size and corresponds to the last 3/5ths of the total embryogenic period. It can be seen from this figure that the 52 and 48 kD proteins become the principal species by the dry seed stage. Their precursors (70 and 67 kD species, respectively), which are prominent in the early phase of storage protein deposition, have all been cleaved to form the mature species by the dry seed stage. The accumulation of the 52 kD mature species appears to lag relative to the 48 kD species during the early phase of deposition, but this is likely due to the very slow cleavage of the 70 kD precursors. In contrast, the 67 kD species does not accumulate due apparently to its more rapid processing. Both the 70 and 67 kD species occur in protein bodies, indicating that their cleavage to
Solution hybridization
Hybridization to saturation of total c D N A with individual cloned c D N A probes was carried out in
201
Fig. 1. A: One dimensional gel electrophoresis in SDS of protein from the insoluble fraction of cotyledons at various stages of development visualized by Coomassie staining. The developmental stages are given along the top of the figure in terms of embryo wet weight in mg. The approximate size of some of the proteins is given in kD on the right margin. Approximately 20 #g of protein were loaded in each well, except for the 5 and 10 mg stages which contained 15 ~g each. B: One dimensional electrophoresis in SDS of the products of in vitro translation in the wheat germ from RNA extracted from cotyledons at the same developmental stages as in A visualized by fluorography. An equal amount of radioactivity was loaded in each well. C: Dot hybridization of RNA from the same developmental stages as in A and B with radioactive cloned probes listed on the left margin. Two/.tg of RNA was immobilized in each dot and hybridized with plasmid DNA of a specific radioactivity of I 2 X 10~ dpm ~g l at a concentration of 2 • 106 dpm ml I as described in (9). Perpendicular lines connecting A-B-C are intended to line up equivalent developmental stages.
202 form the mature proteins takes place in situ within these bodies. The 70 and 52 kD proteins are glycosylated (3) and the true mass of their protein component is unknown, but is likely to be tess than that of their 67 and 48 kD counterparts since they originate from a shorter set of m R N A s and from smaller initial translation products (9). All of these species (70, 67, 52 and 48 kD) are comprised of isolectric variants (3, 4). The other products of the final cleavage of the 70 and 67 kD species which should have apparent molecular weights between 18 and 20 kD are a m o n g the lower molecular weight proteins of this fraction but have not yet been identified. Figure 1A also shows an apparent storage protein precursor of about 38 kD that accumulates and disappears coordinately with the 70 kD species. This species probably is cleaved in situ to smaller fragments seen in the figure, but they have not been identified. Figure 1B shows the profile of proteins synthesized in the wheat germ system from RNA prepared from cotyledons of the same developmental stages as in Fig. 1A. An additional electrophoretic lane shows the proteins synthesized from R N A of 12-h germinated cotyledons as well. The data show that m R N A s giving rise to proteins of 69 and 60 kD increase from a very low level in 5 mg embryos to become the predominant species by the 30 mg stage. We have previously equated the 69 kD translation products with the 67 and 48 kD proteins and the 60 K D translation products with the 70 and 52 kD proteins (4, 9) and have shown that these initial translation products are comprised of isoelectric variants as well (4). The abundancy of these m R N A s changes little throughout the remainder of embryogenesis until the embryos reach about 1 l0 rag. After this stage these m R N A s become undetectable by this technique, nor are they discernible in germinating cotyledons. Thus, m R N A s for both the 69 and 60 kD proteins appear to rise and fall in concentration coordinately. Notable in Fig. 1B is the fact that there is no m R N A species detectable by in vitro translation of a size and abundancy requisite for the synthesis of the proteins about 38 kD (Fig. 1A). Total RNA from the same stage cotyledons as in Fig. 1B was dot blotted and hybridized with each of the three cloned cDNAs (9) representing the three storage protein m R N A subsets. The relative extent of hybridization of each clone with a constant
amount of total R N A from cotyledons of each developmental stage is shown in Fig. 1C. Clone C-72 has been shown to hybridize with all the m R N A which synthesizes the 69 kD proteins, whereas clones C-94 and C-134 each hybridize with a specific subfamily of the m R N A s that synthesize the 60 kD proteins (9). The developmental picture observed with this technique is similar to that observed in in vitro translation and shows, in addition, that the two m R N A subfamilies for the 60 kD proteins change in concentration coordinately. (The intensities of the radioactive signals observed with each of the three probes are not necessarily a comparison of m R N A concentrations between the three subfamilies, since the probes differed somewhat in specific radioactivity and exposure times, for the filters were not normalized.) The changes in abundance with develo?mental time seen in Fig. 1C were corroborated by northern blots of the same total RNAs separated by molecular weight on agarose gels (9) (data not shown). In these experiments hybridization of the cloned cDNAs to each stage R N A was exclusively to the m R N A s of the expected size, i.e. 2.26 kb m R N A with C-72 and 1.96kb m R N A with C-94 and C-134 (9). The foregoing approaches to measuring the storage protein m R N A ahundancy present the same pattern: that of a rapid increase in young embryo cotyledons, a subsequent high steady-state level that is maintained until the embryos are nearly full size, and finally a precipitous drop to a level barely detectable by the hybridization techniques. The validity of this pattern was tested by measuring the rate of hybridization in solution of one of the cloned c D N A inserts with total high molecular weight R N A prepared from cotyledons of several developmental stages. However, in order to use these rates of hybridization to measure the concentrations of storage protein mRNAs, we first determined the concentrations of the m R N A for each of the three subfamilies of proteins in a single developmental stage m R N A population. This was accomplished by hybridizing to saturation c D N A constructed from young embryo cotyledon poly( A ) + m R N A with each of the plasmids carrying one of the three c D N A inserts. The final percent of the c D N A that hybridized indicates the percent of the poly(A)+mRNA mass that is comprised of the m R N A sequences for each storage subfamily. This assumes, as has been shown in other studies (I), that
203
A
A
.
.
.
.
o o~o~ o~c~176
.
.
j .o
.
~ ~176176
.//
~ ~ . /%,.
/
D ~
7,,'
-5
.~
I
-4
~'" -
o
-3
-2
-I
60
o
B
4C
9
o
9
o (~
9
.
50
9
~
~L
66.
A
9
o
40
30
o
9
"
o
20
10
o
o o
o
9
o
o o
o
~
-~
o
1
Q
2
3
log Coi Fig. 2. A: Saturation hybridization of radioactive cDNA ( ~ 400 nucleotides) prepared from poly(A)+ mRNA of 40 mg stage cotyledons with each of the cloned probes for the principal storage protein subfamilies. Plot is of percent hybridization versus log plasmid insert Cot. zx, hybridization with plasmid C-72; o, with plasmid C-94; 121,with plasmid C-134, Plasmid insert DNA was present in at least 250-fold mass excess to cDNA. The data were computer fit with second order kinetics (10). Co values were not corrected for salt concentration. The figure insert is a plot of percent hybridization versus plasmid insert Cot. B: Kinetics of hybridization of total high molecular weight RNA with the cDNA-derived insert of plasmid C-72. The plot is of percent hybridization versus log RNA Cot. The nick-translated insert was approximately 150 nucleotides. RNA was prepared from: 9 20 mg embryo cotyledons; O 40 mg embryo cotyledons; n 110 mg embryo cotyledons; 9 120 mg embryo cotyledons; 9 dry seed; [] cotyledons of 24-h germinated seedlings. RNA was present in at least 2.2 • 103-fold mass excess of the insert DNA. Reactions were normalized to 50% from observed values that ranged from 30 to 43%. Cot values were not corrected for salt concentration. The data were computer fit with pseudo-first order kinetics (I0).
204 reverse transcriptase transcribes all m R N A s equally without respect to sequence. The plasmids were hybridized with c D N A representing 40 mg stage poly(A) + m R N A , and the data, which show considerable scatter, are presented in Fig. 2A. The lines in this figure are the result of computer-fits of the data. Although these fits indicate a small difference in the saturation values between plasmid C-72 and C-94, we do not feel this difference is significant in view of the imprecision of the measurements and have rounded offthe saturation value for these two plasmids at 15% of the cDNA. Surprisingly, C-134, whose insert represents one of the 60 kD proteins, hybridizes only to about 5% of the cDNA. Inserted into Fig. 2A is a plot of percent c D N A hybridized vs. plasmid insert Cot which shows that apparent saturation is reached at the Cot values used. These values of course only refer to the concentration of these m R N A s in poly(A) + m R N A , but in cotton cotyledons both poly (A) + and poly(A) m R N A appear to contain the same m R N A sequences (including storage protein mRNAs) in similar proportion (5 and unpublished data) and, consequently, we believe the data presented in Fig. 2A reflect the concentration of these m R N A s in total m R N A as well as poly(A) + m R N A . Other data confirm the results of Fig. 2A for this extrapolation. I n v i t r o translation of total, poly(A) + and poly(A) RNA gives the same relative amounts of radioactive proteins as seen in Fig. lB. Further, hybridization of R N A with c D N A prepared from 40 mg stage cotyledons indicates the existence of R N A components of this high abundance in both total and poly(A) + RNA. The in v i t r o protein synthesis pattern shown in Fig. 1B indicates that m R N A s for the 69 and 60 kd proteins not detectably change relative one to another during embryogenesis. Further, the RNA dot blot hybridization pattern (Fig. 1C) and northern hybridization substantiate this. This pattern also shows that there is no change in the m R N A s for the two 60 kD protein subfamilies relative one to another. Thus the relative abundance of the three subfamilies seen in Fig. 2A does not change in embryogenesis. Knowing the concentration of these m R N A s at the 40 mg stage, it was possible to measure the absolute abundance during this period of all three m R N A subfamilies by measuring the rate of hybridization of a single cloned insert (C-72) with
RNA from several developmental stages relative to the rate of hybridization with this reference 40 mg stage RNA. Hybrid formation (Fig. 2B) was measured as the percent of the cloned c D N A probe that became resistant to S-1 hydrolysis. Since only a single strand of the probe is complementary to the m R N A sequences of the subfamily, only 50% of the probe can hybridize. In actuality a small portion of the probe becomes fragmented in the nick-translation step into pieces too small to form a stable hybrid. For this reason the computer derived hybridization in Fig. 2B has been normalized such that the degree of hybridization ranges from 0 to 50%. The concentration of the probe used in these experiments was so low that probe: probe reassociation should contribute negligibly to the S-1 resistant radioactivity. The probe was reassociated with total RNA from 20, 40, 110 and 120 mg embryo cotyledons, from dry seed cotyledons and 24-h germinated seedling cotyledons. Since the amount of m R N A in these RNA preparations has been determined previously (7), it was possible to calculate the concentration of the m R N A for 69 kD storage protein in terms of the fraction of total m R N A as well as the fraction of total RNA. The number of m R N A transcripts present at each developmental stage were also calculated per embryo and per haploid genome equivalent. These calculations are presented in Table 1. It is clear from Table 1 and from a visual comparison of Fig. 2A, B that a considerable amount of m R N A from 20, 40, and 110 mg embryo cotyledons is comprised of sequences for the 69 kD proteins, and by extension from the previous data (Fig. 1C, Fig. 2A), of sequences for the 60 kD proteins. The data indicate that there is an abrupt drop in the concentration of m R N A for these proteins from the 110 mg stage to the 120 mg stage. Even more of these sequences are lost as the seed desiccates. However, a trace remains in cotyledons after 24 h of germination. Northern blot analysis (data not shown) detected mature sized m R N A transcripts in all of these later developmental stages, in approximately the same concentration as determined in Fig. 2B and Table 1. Clearly, the picture presented by the kinetics of clone C-72: RNA reassociation confirms that suggested by in v i t r o translation (Fig. I B) and hybridization of the cloned probes with RNA dots (Fig. IC) and northern blots. Quantitation of m R N A levels by all these techniques agrees remarkably well.
205 Table 1. Concentration of 2.26 kb storage protein mRNA during cotton cotyledon development measured by the kinetics of hybridization of C-72 with total RNA.
Developmental s t a g e
Post-anthesis ~ (days)
Ksfob (M t s 1)
Concentration Fraction of RNA
Embryo (wt weight) 20rag 40 mg ll0mg 120rag Dry seed 24-H germination
24 26 44 51 60+
l.l 1.6 9.0x 10-I 3.9x 10-2 6.8X 10-3 1.9 X 10-3
Molecules
Total RNAc
mRNAd
Per embryoe Per haploid genomef
3.0x 10 3 4.4X 10 3 2.5• 3 I.IX 10 4 1.9• 5 5.2 X l0 6
1.0x 10 r 1.5 X 10 1 8.6>(10 2 3.8x 10 3 I.IXI0 3 1.6 X l0 4
5.3X 10~~ 2.9 x 10~ 3.8X l011 1.7x 10j~ 3.0X 109 8.3 X l0s
6.2x 103 1.6 • 104 1.1 x 104 4.9x I02 8.6X 10l 2.4 X 10~
a Walbot & Dure (13). b Least squares best estimate of the pseudo first order rate constant of the hybridization of C-72 with total RNA, as presented in Fig. 2B. Fractional uncertainties are approximately • 0.09 (one standard error). All other calculations in the table thus have associated uncertainties at least this large. c Calculated assuming the 2.26 kb mRNA comprises 15% of the 40 mg stage poly(A)+ mRNA and total mRNA (Fig. 2A and discussion in Results) and a total mRNA content of 2.9% (8). a Calculated assuming 2.9% of embryo and 1.7% of dry seed (8) and 3.3% of 24-h seedling cotyledon RNA is total mRNA (7). e Based on the total high molecular weight RNA contents of the cotyledons (unpublished data). f Based on the DNA content of the cotyledons during development (13) and a haploid genome size of 1.8 x 109 nucleotides. G. hirsutum is an allotetraploid.
Discussion The levels of the m R N A s c o d i n g for the storage proteins in c o t t o n c o t y l e d o n s during e m b r y o g e n e sis have been measured by f o u r i n d e p e n d e n t means, three of which m a d e use of the cloned probes for the three m R N A subfamilies. D if f e r e n t pieces of inf o r m a t i o n can be ascertained f r o m the techniques, 1. All three m R N A subsets increase rapidly in m i d - e m b r y o g e n e s i s , but decline d r a m a t i c a l l y in a short p e r i o d late in e m b r y o g e n e s i s as s h o w n by all techniques. These studies c o n f i r m in most details the q u a n t i t a t i v e c o n c l u s i o n s r e g a r d i n g storage protein m R N A s c o n c e n t r a t i o n s in c o t t o n c o t y l e d o n s made by Gal au & D u r e (8), which were based on h y b r i d i z a t i o n of different d e v e l o p m e n t a l stage m R N A p o p u l a t i o n s with h o m o l o g o u s and heterologous stage c D N A s . 2. All three gene subfamilies are expressed as m R N A in a c o o r d i n a t e f a s h i o n as s h o w n by the h y b r i d i z a t i o n of cloned c D N A probes to R N A dots or n o r t h e r n blots and (to a lesser extent) by in v i t r o translation. H o w e v e r , the stained protein gel (Fig. 1A) shows that the 52 k D m a t u r e p r o t e i n set does not increase as ra p id ly in early e m b r y o g e n e s i s as does the 48 k D set. This a p p a r e n t discrepancy is
caused by the fact t h a t a substantial a m o u n t of the 52 k D proteins are in the 70 k D p r e c u r s o r species (which are very ev i d en t on the stained gel) until Synthesis and processing is c o m p l e t e in late e m b r y o genesis. 3. Th e m R N A s for two of the subfamilies a p p e a r to exist in the same a m o u n t s , whereas one of the m R N A subfamilies a p p e a r s to be expressed at only 1 / 3 the level of the o t h er t w o subfamilies as indicated by the c D N A s a t u r a t i o n hybridization. These m e a s u r e m e n t s are of m R N A levels and say n o t h i n g a b o u t the n u m b e r of p r o t ei n copies translated f r o m those m R N A s in v i v o , n o r a b o u t the n u m b e r of genes c o m p r i s i n g each subfamily. We have r e p o r t e d (3) that in the m a t u r e seed the ratio of 52 k D p r o t e i n s to 48 k D proteins is a b o u t 4:3 based on several m e t h o d s o f p r o t e i n d e t e r m i n a t i o n . In actuality, the 52 k D proteins are glycosylated and the p r o t e i n f r a c t i o n of these molecules m ay represent only 40 k D of p r o t ei n mass. ( D e d u c e d f r o m the fact that the increase in a p p a r e n t m o l e c u l a r weight o b s e r v e d on gels u p o n g l y c o s y l a t i o n of the p u t a t i v e 58 k D p r o p r o t e i n f o r m of the 60 k D p r e p r o p r o t e i n to the 70 k D p r e c u r s o r is 12 kD.), If the final 52 and 48 k D proteins in fact c o n t a i n 40 and 48 k D of protein, respectively, and each of the m R N A sub-
206 sets were translated equally in embryogenesis, a r o u g h l y 8:7 mass ratio of 52 to 48 k D proteins would be expected. This is because the 52 k D proteins, a l t h o u g h c o n t a i n i n g less protein mass per molecule t h a n do the 48 kD proteins, are derived f r o m two m R N A subsets which in total a m o u n t comprise a b o u t 20% of the m R N A d u r i n g the period of m a x i m u m expression, whereas the 48 kD m R N A s constitute a b o u t 15%. F r o m this reasoning it seems likely that the three subsets of m R N A s are translated to a n equal extent d u r i n g embryogenesis. Cell division in these cotyledons ceases at a b o u t the 40 mg stage and D N A synthesis ceases at a b o u t the 80 mg stage, while the m R N A s for these storage proteins are at their high level. Since no change in their c o n c e n t r a t i o n is observed immediately before or after these transitions, the r e g u l a t i o n of their t r a n s c r i p t i o n must be i n d e p e n d e n t of the regulation of the genes responsible for cell division or D N A replication. The fact that the m R N A s for the storage proteins are expressed c o o r d i n a t e l y and are m a i n t a i n e d at the a p p a r e n t 3:3:1 ratio t h r o u g h o u t their period of expression may indicate that the same ratio exists for the n u m b e r of genes for these subfamilies a n d that their expression is not individually modulated. The drastic decrease in these m R N A s at the end of embryogenesis (between the 110 a n d 120 mg stages) represents a b o u t a 25-fold drop in concent r a t i o n that occurs over a 3-to-4-day period. If this is solely the result of a sudden cessation in the t r a n s c r i p t i o n of these genes, the half life of the m R N A s would be between 15 a n d 20 h. This is a realistic value for m R N A half life, a n d there is no reason to evoke a special m e c h a n i s m for the disapp e a r a n c e of these m R N A s . Both the data in Fig. 2B a n d n o r t h e r n blot analysis (data not shown) indicate that storage protein m R N A s decline in concent r a t i o n (as a fraction of total R N A ) even more rapidly in the first 24 h of g e r m i n a t i o n . The species of c o t t o n used is a n a t u r a l allotetraploid ( a m p h i d i p l o i d ) . It carries what can be recognized as a n ancestral A and D g e n o m e (6). It is t e m p t i n g to conjecture that the 69 k D protein genes derive from one of the ancestral genomes and the 60 k D proteins from the other. However, p r e l i m i n a r y results in which proteins from a diploid species carrying only the A g e n o m e were electrophoresed in one and two d i m e n s i o n s showed the existence of both molecular weight sets of proteins. Storage protein synthesis has been followed on
the m R N A level in other plant species, principally legumes a n d cereal of agricultural i m p o r t a n c e . Much of this work has been collated recently by Spencer & Higgins (11) a n d in a N A T O A.S.I. p u b l i c a t i o n (2).
References I. Axel, R, Feigelson, P & Schutz, G, 1976. Analysis of the complexity and diversity of mRNA from chicken liver and oviduct. Cell 7:247 254. 2. Ciferri, O & Dure III, LS, 1983.The Structure and Function of Plant Genomes. Plenum Press, New York and London. 3. Dure 111,LS & Chlan, CA, 1981. Developmental biochemistry of cottonseed embryogenesisand germination XII Purification and properties of the principal storage proteins. Plant Physiol. 68:187 194. 4. Dure Ill, LS & Galau, GA, 1981. lbid XII1 Regulation of the biosynthesis of the principal storage proteins. Plant Physiol. 68:187 194. 5. Dure III, LS, Greenway, SC & Galau, GA, 1981. ibid XIV Changing messenger ribonucleic acid populations as shown by in vitro and in vivo protein synthesis. Biochem. 20: 4162-4168. 6. Fryxell, PA, The Natural History of the Cotton Tribe, Texas A & M University Press, College Station and London. 7. Galau, GA, Legocki, AB, Greenway, SC & Dure Ill, LS, 1981. Cotton messenger RNA sequences exist in both polyadenylated and nonpolyadenylated forms. J. Biol. Chem. 256:2551-2560. 8. Galau, GA & Dure lI1, LS, 1981. Developmentalbiochemistry of cottonseed embryogenesis and germination XV Changing messenger rihonucleic acid populations as shown by reciprocal heterologous complementary deoxyribonucleic acid hybridization. Biochem. 20:4169-4178. 9. Galau, GA, Chlan, CA & Dure lII, LS (companion paper). Ibid XVI Analysis of the principal cotton storage protein gene family with cloned cDNA probes. (Submitted to Plant Mol. Biol.). 10. Pearson, WR, Davidson, EH & Britten, RJ, 1977. A program for least squares analysis of reassociation and hybridization data. Nucleic Acids Res. 4:1727 1735. ll. Spencer, D & Higgins, TJV, 1982. Seed maturation and deposition of storage proteins. In: A Smith & D Grierson (eds.) The Molecular Biology of Plant Development. University of California Press, Berkeley and Los Angeles. 12. Southern, E. M., 1979. Gel electrophoresis of restriction fragments. In: A. Wu (ed.) Methods in Enzymology,68: pp. 152 176. Academic Press, New York, NY. lY Walbot, V. & Dure Ill, LS, 1976. Developmentalbiochemistry of cottonseed embryogenesisand germination VII Characterization of the cotton genome. J. Mol. Biol. 101: 503 536. 14. Yang, R. C. A., Lis, J & Wu, R, 1979. Elution of DNA from agarose gels after electrophoresis. In: A Wu, (ed.) Methods in Enzymology,68: pp 176 182.Academic Press, New York, NY. Received 24 May 1983; in revised form and accepted 2 August 1983.
